Vous êtes sur la page 1sur 350

«••••••••••••• !

•••• ••••••••• •
•••••••••••••
•••••••• • • •
I8S58S .asassssssssasssis
MflTHEMflTlCRLrg ..::::::

THEORY jjjjjjj ! ! ! ! ! ! ! ! ! ••I


OF MUSiC
!•••••••
S8SSSSSSS •••••••••Hi !•••••!

._••••••
Mass::""""8ma • „ Franck Jedrzejewski

••

••••i
••••••••••••••••I
!••••••••••• • • • • I
m m

!•••••!
!•••••••!
!•••••••!
!••••• !••• !••••••_
!•••••! !•••••! •••••••• _
!••. ••••••••••••
••••••
E CHE R l C fl L E S
•• M U S i Q U E / S C i E N C E S
•••••••••••••••••••••••••••

••••••••••I
••••••I
i
• • • • • • • • • • • • • • • I! » !
II
••••••••••••••••
••• !••••••

•••••• •••••••I
• •••••I •
•••••••I ••••••••!
•••• ••••I
H •••••••••••••••••:
:sssi :gf •••
__•!__
!•••«!
••••••••I ••••••••I
•••I
!••«••!
ircam ElELATOUR • •m •a • • • • •
EDITIONS

Centre F R A N C E
Pompidou www.editlons-delatour.conn

•••••••
MATHEMATICAL THEORY
OF MUSIC
Collection Musique/Sciences
directed by Jean-Michel Bardez & Moreno Andreatta

The Musique/'Sciences series contributes to our understanding of the relationship bet-


ween two activities that have shared intimate links since ancient times : musical and
scientific thought. The often-cited Quadrivium (music, astronomy, geometry, and arith-
metic) reminds us that, in an age imbued with the spirit of the Gods, it was not un-
common to think of these two modes of thought as twins. During the twentieth century,
music and science developed new links, establishing relationships with mathematics and
opening new lines of musical research using information technology. Modeling, in its theo-
retical, analytical and compositional aspects, is more than ever at the center of a rich
musicological debate whose philosophical implications enrich both musical and scientific
knowledge. The pleasure of listening is not diminished when it is more active, more aware
of certain generating ideas - au contraire.

Published works
Gerard Assayag, Prangois Nicolas, Guerino Mazzola (dir.), Penser la musique avec les
matMmatiques ? (2006)
Andre Riotte, Marcel Mesnage, Formalismes et modules musicaux, 2 vol. (2006)
Carlos Agon, Gerard Assayag, Jean Bresson (ed.), The OM Composer's Book (2006)

Forthcoming
Moreno Andreatta, Jean-Michel Bardez, John Rahn (dir.), Autour de la Set Theory
Guerino Mazzola, La virite du beau dans la musique
MATHEMATICAL THEORY
OF MUSIC

Pranck Jedrzejewski

Collection Musique/Sciences

ircam r r \w? n :i
Centre
•£ -J
bELATOuri
F R A N C E
Pompidou
Editorial Board
Carlos Agon, Ircam/CNRS, Paris
Gerard Assayag, Ircam/CNRS, Paris
Marc Chemillier, University of Caen
Ian Cross, University of Cambridge
Philippe Depalle, McGill University, Montreal
Xavier Hascher, University of Strasbourg
Alain Poirier, National Conservatory of Music and Dancing, Paris
Miller Puckette, University of California, San Diego
Hugues Vinet, Ircam/CNRS, Paris

Editorial Coordination
Claire Marquet

Cover Design
Belleville

Tous droits de traduction, d'adaptation et de reproduction par tous procedes reserves pour tous pays.

Le code de la propriete intellectuelle du ler juillet 1992 n'autorise, aux termes de Particle L. 122-5, 2e
et 3e a), d'une part, « que les copies ou reproductions strictement reservees a I'usage du copiste et non
destinees a une utilisation collective » et, d'autre part, « que les analyses et les courtes citations dans
un but d'exemple et d'illustration ». « Toute representation ou reproduction integrate ou partielle, faite
sans le consentement de I'auteur ou ayants cause, est illicite » (article L. 122-4). Cette representation
ou reproduction par quelque procede que ce soit constituerait done une contrefagon sanctionnee par les
articles L.335-2 et suivants du Code de la propriete intellectuelle.

ISBN 2-7521-0027-2 et 2-84426176-0

© 2006 by Editions DELATOUR FRANCE/Ircam-Centre Pompidou


w ww .editions-delatour .com
www.ircam.fr
Contents

Preface xi

1 Harmonic Models 1
1.1 Musical Sounds 1
1.2 Musical Intervals 2
1.3 Fourier Analysis 3
1.4 Physical Models 6
1.5 Sonance Models 15
1.6 Geometric Models 17
1.7 Affinity Models 21

2 Taxonomy of Chords and Scales 29


2.1 Group Actions 30
2.2 P61ya's Enumeration 33
2.3 Classification in Tempered Music 35
2.4 Motives and Mosaics 39
2.5 Dihedral Classification 40
2.6 Cyclic Chords 41
2.7 Enumerations of Pcsets Classes 43

3 Limited Transpositions Sets 57


3.1 General Theory 57
3.2 Messiaen Modes 60
3.3 Diatonic Scales and Microtonality 68
3.4 Quarter-Tone Limited Transpositions Sets 70
3.5 Enumerations 76

4 Neo-Riemannian Investigations 83
Contents

4.1 Set Theory 84


4.2 Generalized Interval Systems 88
4.3 Riemannian Transformations 91
4.4 Isographies and K-nets 92
4.5 Parsimonious Graphs 95
4.6 Permutational Transformations 97
4.7 Morris Groups 102

5 Knots and Braids 105


5.1 Classification of Knots 105
5.2 Dodecaphonic Knots 106
5.3 Gauss Diagrams 108
5.4 All-Interval Series 112
5.5 Serial Algebra 113
5.6 Combinatoriality 114
5.7 Markov Chains 116
5.8 Groups and Diagrams 119
5.9 Tuning Braids 121
5.10 Textual Analysis 124

6 Theory of Temperaments 127


6.1 Twelve Tone Temperaments 127
6.2 Pythagorean Tunings 129
6.3 Equal Temperaments 130
6.4 Zarlinean Temperaments 130
6.5 Meantone Temperaments 130
6.6 Historical Temperaments 131

7 Approximations of Temperaments 133


7.1 Continued Fractions 133
7.2 Meantone Approximations 138
7.3 Harmonic Approximations 140
7.4 Multiple Continued Fractions 141

8 Tuning Groups 145


8.1 Definitions 145
8.2 Formal Languages 149
8.3 Graphical Representations 151
8.4 Harmonic Metrics 152
8.5 Hellegouarch Commas 154
8.6 Consistency and Completeness 155

9 Contemporary Tunings 157


9.1 Euler-Fokker Genera 157
9.2 Farey Temperaments 160
9.3 Harry Partch Tonality Diamond 163
9.4 Ben Johnston Lattices 165
9.5 Ervin Wilson Tunings 166
9.6 Non-octaviant Tunings 167
9.7 Cyclic Tunings 169

10 Groups and Symmetries 173


10.1 Symmetry Groups 173
10.2 Frieze Groups 174
10.3 Polyhedral Groups 177
10.4 Coxeter Groups 178
10.5 Crystallographic Groups 179
10.6 Bravais Temperaments 188
10.7 Sporadic Groups 189
10.8 Constellations and Cacti 189

11 Tessellations and Canons 195


11.1 Musical Tilings 195
11.2 Perfect Tilings 200
11.3 Messiaen Chords 202
11.4 Reflexive Chords 206
11.5 Young Diagrams 207
11.6 Prime Canons 208
11.7 Multiple Canons 212
11.8 Regular Canons 214
11.9 Magic Squares 217

Appendix A: Dodecaphonics Knots 221

Appendix B: All-Interval Series 277

Bibliography 299

Index 331
Preface

The aim of this book is to provide an introduction to the mathematical theory of music,
with emphasis on the fascinating and original connections with musical analysis. Despite
the complexity of the subject, we have tried to make this introduction accessible to
a large audience. Many examples are given for a better understanding. Musicians and
mathematicians are invited to contemplate some nontrivial aspects of the mathematical
construction.
The book is divided into eleven chapters. In the first chapter, we introduce the language
of harmonic models. After a review of some physical models based on the wave equation,
which try to modelize the inharmonicity of the spectrum of certain frequencies, we look
briefly at sonance models. These models are built around the notion of dissonance curves,
which go back to the Hermann Helmholtz's book Sensation of Tones. The next section
is devoted to affinity models introduced by Edmond Costere. Costere's theory could be
generalized to other weight systems, but the crucial question is to know what is the
best weight system for a given context and to give an algebraic demonstration of the
classification obtained. Costere has just given the results for a particular weight system
obtained from a computational software. Today, we have to understand what is behind
these representations and to give the properties of the associated weight systems. In the
last section, we review some geometric methods which lead to the theory of modulation.
Chapter 2 on Taxonomy of chords and scales, is devoted to the classification of pitch-
class sets. We first introduced the notion of a group action, and review Polya's enumer-
ation theory. Four main group actions are considered : the cyclic classification proposed
by Edmond Costere (1954), the dihedral classification proposed by Hanson (1960), Za-
lewski and Forte (1972), the affine classification proposed by the Czech composer Pinos
(1971) and the symmetric classification proposed by the Mexican composer Julio Estrada
(1994). The chapter ends with the enumeration of sets classes.
Chapter 3 investigates the question of determining the limited transposition sets for
a given temperament, and to find a general formula to enumerate them. This problem
is envisaged in the n-tone equal temperament. The enumeration is given in particular
Preface

for the 96-tone equal temperament (1/16 of tones) that is also used by some young
composers.
Chapter 4 is devoted to the Neo-Riemannian investigations. The chapter begins with a
survey of the main results of the Set Theory developed by Allen Forte, briefly illustrated
with short examples. The next section deals with David Lewin's transformation theory
and the Generalized Interval Systems. The legacy of Hugo Riemann is evocated through
the Riemaniann transformations which are used in musical analysis. In this section, we
propose a new interpretation of these transformations using permutations. The group
generated by the new Riemannian transformations is the dihedral group. The permuta-
tions are well-adpated to the design of chords's lattices and isographies.
Chapter 5 on Knots and Braids is one of the most important. It shows how to clas-
sify the 12! twelve tone rows with 554 Gauss diagrams. Schoenberg series are mapped
to singular knots with six double points. From knots or chord diagrams, universal in-
variants are computed. We hope that they will be a precious help in the computational
musical analysis of serial scores. They could also be used as a kind of measurement of
the aesthetic positions when comparing two scores. Braids are introduced to construct
new temperaments. Lastly, in textual analysis, low dimensional topologies are found as
an expression of hidden knots, in the same way that the famous linguist Ferdinand de
Saussure tried to find anagrams under the words.
Chapter 6 on Theory of Temperaments introduces the problem of temperaments, from
Pythagoras to the modern systems. General formulas are established for the division of
the Pythagorean comma and for the division of the syntonic comma. For each choice of a
set of frequencies, the question is to know if one could build an efficient harmonic theory,
which could help us to understand the interactions between notes in chords and musical
progression.
Chapter 7 on Approximations of temperaments begins with a survey on continous
fractions. The approximations of irrational temperaments (i.e. temperaments whose fre-
quency ratios are irrational numbers) by rational temperaments and the converse problem
are investigated. The approximation of the meantone temperament is generally based on
the use of rational approximation of some numbers such as the square root of 5. The
chapter ends with a short description of the Jacobi-Perron algorithm.
Chapter 8 on Tuning groups introduces many new problems. Among them, one difficult
question is to determine the Hellegouarch commas of a given group. Another is to find a
metric well suited to the measurement of harmonic distance.
Chapter 9 on Contemporary Tunings describes modern tunings used by 20 th century
composers, as Harry Partch, Warren Burt, Ben Johnston, Ezra Sims and others. The
chapter ends with a study of the cyclic tunings which have a strange hierarchical organ-
ization.
Chapter 10 deals with Groups and symmetries. We review the mathematical inter-
pretation of symmetry in art and in musical scores. The frieze groups are well known
and have been found in many scores. By substituting the atoms of a crystal lattice by
the frequency ratios derived from the coordinates of the atom, we introduce a new in-
terpretation of crystallographic temperaments. The Bravais temperaments are treated
as an example. Sporadic groups have been used by Olivier Messiaen. Constellations and
cartographic groups end the chapter.
In Chapter 11, Tesselations and Canons, the relationships between tilings and canons
are introduced with many examples. Some musical structures such as chords and twelve
tone rows could be used to tile in two or three dimensions. The classification of canons

xii
Preface

is an important open mathematical problem connected with the decomposition of finite


abelian groups which were first investigated by De Bruijn, Haj6s and Vuza. The chapter
ends with a section on magic squares which have been used by many composers such as
Peter Maxwell Davies, Manfred Kelkel and Ivan Wyschnegradsky.
This book grew out of several talks I gave at the MaMuX (Mathematics, Music and
Relations to other fields) Seminar at IRCAM (Paris) during the years 2001-2004 and
at the AMS Joint Meeting in Phoenix (USA). It is a great pleasure to express my
thanks to all participants, to the members of the Society for Mathematics and Com-
putation in Music and in particular to Carlos Agon, Charles Alunni, Emmanuel Amiot,
Moreno Andreatta, Didier Aschour, Gerard Assayag, Alain Bancquart, David Benson,
Amine Beyom, Georges Bloch, Per Aage Brandt, Peer Bundgaard, Chantal Buteau, Vit-
torio Cafagna, Costin Cazaban, Marc Chemillier, Elaine Chew, Jean-Pierre Cholleton,
Jean-Marc Chouvel, Philippe Codognet, Celestin and Irene Deliege, Anja Fleischer, Al-
len Forte, Harald Fripertinger, Martha Grab6cz, Karim Haddad, Rachel Hall, Xavier
Hascher, Yves Hellegouarch, Richard Hermann, Tuukka Ilomaki, Peter Johnstone, Tom
Johnson, Oren Kolman, Olivier Lartillot, Fabien Levy, Michael Leyton, Alain Louvier,
Pierre Lusson, Mikhail Malt, Guerino Mazzola, Stefan McAdams, Dave Meredith, Mar-
cel Mesnage, Nicolas Meeus, Benott Meudic, Frangois Nicolas, Yann Orlarey, Thomas
Noll, Athanase Papadopoulos, Frederic Patras, Robert Peck, Jean Petitot, Iann Quinn,
John Rahn, Andre Riotte, Stephan Schaub, Makis Solomos, Marco Stroppa, Andranik
Tangian, Petri Toiviainen, Dmitri Tymoczko, Luigi Verdi, Domenico Vicinanza, and John
Wild for their involvement and, for some of them, for valuable discussions and comments.
I owe special thanks to all reviewers, and in particular to Moreno Andreatta, for their
remarks and their efforts to put this text in correct form. I have tried to make the bib-
liography reasonably complete, but some gaps are inevitable, and I apologize to those
authors whose relevant work has inadvertently been omitted.

xiii
1
Harmonic Models

The science of sound studies the communication of sound messages created by vibrations
in the propagation medium. It depends on several connex domains. A sound is a physical
phenomenon, but also a perceptual phenomenon. Physical attributes of a sound such as
frequency and amplitude are distinct from perceptual correlates as pitch and loudness.
Sounds are created by compression and expansion of gas molecules in the propagation
medium.

1.1 Musical Sounds


Prom a physical point of view, a pure sound is a sine wave and more complicated periodic
sounds are a collection of pure waves. This collection determines the spectrum. A pure
wave is characterized by its frequency, amplitude and phase. Frequency is the number of
times per second that object vibrates. It is measured in hertz (Hz) or cycle per second.
Low frequencies are perceived as bass notes, while high frequencies sound high. Human
ears perceive sound from about 16 to 20 000 Hz. Concert A is defined as 440 Hz. The
different frequencies present in a sound are called partials. The lowest frequency is called
the fundamental and the frequencies above the fundamental are the overtones. Overtones
are either harmonic or inharmonic. They are harmonic if their frequency is an integer
multiple of the fundamental frequency / , 2 / , 3 / , etc. otherwise they are inharmonic.
The harmonic overtones are also called the harmonic series, harmonic partials or just
harmonics. Pitch is a subjective quantity related to frequency and to the overtone series.
The Fechner law states that the perceived pitch is proportional to the logarithm of
the frequency. The amplitude of the sound wave evolves over time and determines the
envelope of a sine wave. The sound envelope is the collection of the envelopes of all
partials. It is characterized by four segments: attack, sustain, decay and release. A sine
wave emits at a point O of the space will be perceive by a man located at a point M with
a decay 6 = d/c where d is the distance OM and c the speed of the sound. The period
1. Harmonic Models

of the sound is defined as the inverse of the frequency, T = 1 / / . The wavelength X = cT


is the distance covered by the sound during T seconds at the speed c. The speed of the
sound depends on the medium characteristics. At the atmospheric pressure and in air,
the speed varies with temperature: at 0 C, c = 330 m/s, at 10 C, c = 337 m/s, at 15 C,
c = 343 m/s, at 20 C c = 334 m/s and at 30 C, c = 349 m/s.
The acoustic power I is the total energy due to the movement of air molecules. For a
sine wave of amplitude a, the acoustic power is given by

£L = 4TT W / 2
I = pc
where P is the acoustic pressure. The acoustic power is measured in bels or in decibels
(dB). The level is defined from a threshold of hearing I0 = 1012 Watt/m 2 and P0 = 2-10" 5
Pa.
d £ = 10-log-^ = 2 0 - l o g ^ -
h Po
The limit of audibility is 0 dB. A low voice is about 20 dB, a normal voice around 40 dB
and a tutti of a symphonic orchestra about 100 dB.
The sum of two sine waves of closed but distinct frequencies is perceived as a single
sound whose intensity slowly oscillates from large to small values. These vibrations are
called beats. When the amplitude is large, the interference is said to be constructive and
when the amplitude shrinks to zero the interference is destructive. If the frequencies differ
by an amount A / the resulting sound

sin(2?r/t) + sin(27r(/ + Af)t) = 2 cos(nAft) sin(27r(/ + A//2)*)

can be viewed as a sine wave of frequency / + A / / 2 with a slowly varying envelope


A = 2cos(irAft) of frequency A / . The number of beats per second is the difference of
the two frequencies. For example, two sounds of 440 Hz and 444 Hz create 4 beats per
second. Beats are also considered between partials. The third partial of A 440 Hz and
the second partial of E 659 Hz create

3/i - 2/ 2 = 3 x 440 - 2 x 659 = 2 beats/second

1.2 Musical Intervals


By the Fechner law, perceived pitches are approximately proportional to the logarithm
of the frequency. The ratios of the frequency are measured in cents. The musical interval
of two sounds of frequencies /1 and /o is given by
1200
1 // / ^ *
j—y M / 1 / / 0 ) cents

There are 100 cents in the tempered semitone and 1200 cents in the octave. In the n-tone
equal temperament (n-tet), the octave is divided in n tones. Two consecutive tones are
separated by the interval of 2*/ n or 1200/n cents. The ratios of the just intonation scale
are: unison (1), minor tone (10/9), major tone (9/8), minor third (6/5), major third
(5/4), fourth (4/3) and fifth (3/2). The following intervals will be used in the chapters
about tuning and temperaments.

2
Fourier Analysis

The Pythagorean comma is the difference between twelve just fifths and seven octaves

(3/2P = 3^ = 531441
p
27 219 524288
The syntonic comma is the difference between a major Pythagorean third (81/64) and
a just third (5/4)
_ 81/64 34 81 M
C. = —-.— = —:— = — = 22 cents
5/4 24-5 80
The Holderian comma is the semitone in the 53-tet

CH = 2 1 / 5 3 = 23 cents

The Fokker diesis is the 1/31 part of one octave

DF = 2 1 / 3 1 = 39 cents

The septimal comma is the difference between the Pythagorean minor seventh (16/9)
and the just minor seventh (7/4)

_ 16/9 26 64 _ t
C7 = = 27centS
W = 3^7 63"
The leimma is the difference between three octaves and five just fifths

23 28 256 m
5 = TTF
5 = TTTTT = 90 cents
(3/2) 3 243

The apotome is the difference between seven just fifths and four octaves

. (3/2) 7 37 2187 _ +
A = ' 4 = TTTT
11 = ^TTTTT = 114 cents
2 2 2048

1.3 Fourier Analysis


Jean-Baptiste Fourier has shown that the spectrum of periodic sounds, as any periodic
signal, can be decomposed into a sum of sine waves. The first sine tone with frequency
/ is called the fundamental, and the others 2 / , 3 / , etc. are called harmonic partials.
For more complex sound, spectrum is inharmonic that means it includes non-harmonic
sounds such as noise (a range of frequencies) or partials (a frequency such that its ratio
to the fundamental is not an integer). The spectrum is the collection of all partials or
overtones. The Fourier Theorem states that each periodical phenomenon is decomposed
into a sum of sine waves. For example, the sound

f(t) = 8sin 3 (27r/i) - 6sin2(27r/£) + 3

is written as a sum of three sine waves.

f(t) = 6sin(27r/£) + 3sin(47r/£ + TT/2) + 2sin(67r/£ - n)

3
1. Harmonic Models

The first sine wave is the fundamental of frequency / . The two others are the second
and third partials with frequencies 2 / and 3 / . More generally, a periodic function f(t) =
f(t + T) of period T and frequency f = 1/T can be written as a trigonometric series
oo
cos
/(*) = y + X > (27r/nt) + bn sm(2irfnt)
n=l

The coefficients are called the Fourier coefficients and are given by the following formulas

a
2 frT
n = ™ / f(x) COS(2TTfnx)dx
1
Jo

and
2 fT
bn= sin 27r nx dx
T f^ ( f )
For non periodic waveforms, we use the Fourier transform, which is a generalization of
the Fourier series. The Fourier transform represents the magnitude of the sound against
the frequency. For a real or complex value function of the Lebesgue space L1, the Fourier
transform is
H-oo
/(«) = / e-2iirvtf(t)dt
-oo
J—o
and the inverse Fourier transform is given by the following formula
r +oo

/(*)= / ezznvtf(v)dv
J—c

Recall that the total energy of a signal is equal to the total energy of its spectrum. So
there is an equality between the L 2 -norm of a function and the L 2 -norm of its Fourier
transform called the ParsevaVs Formula.
P+00 i^ ,2 Z^oo 0
/ \f(v)\ dv= / \f(t)\2dt

The convolution of two functions


+oo
f(x)g(t - x)dx
/ -oo
is preserved under Fourier transform

/ * 9(v) = f(v)d(v)

Real sound data are usually analyzed by Fast Fourier Transform (FFT). Another kind
of generalization of the Fourier transform is the wavelet transform. Formally, a wavelet
is a real function ^(t) such that the constant

l~ I2
-l-oo
c# = 2TT / -—, ' dx
/ -oo l»l
4
Fourier Analysis

belongs to ]0, +00[. For example, the Mexican hat defined by

for which the Fourier transform is

$(v) = v2e~v2l2

is a wavelet with cy = 1. The wavelet transform of a waveform / is a function of two


variables a ^ 0 and b defined by

a
dt
V |a| Cy J-00 \ J

The inverse transform of a function g(a, b) is the double integral


/>+oo /.+00 -j /t-h\
L%(h)(t) = / / _ z_ _ p ( a > b ) 9 ( —a ) dadb
J-00 J-00 a y/\a\cy \ /

For a square integral function / , we have L^L^(f) = f almost everywhere, and in


particular, for all points where / is continuous.

We discuss now a way of constructing spectral density of a random process X(t) and
we illustrate the use of the FFT on Olivier Messiaen's Mode de valeurs et d'intensites.
In the first voice, Messiaen uses a kind of series. Each note is encoded from 1 to 12. The
time series is composed of the pitch variations X(t) as a function of time, where X(t)
belongs to the set {1, 2, ..., 12}. The data has the following representation

100 150 200 260

This data is used to compute the normalized autocorrelation function defined by


T
Y,(X(T)-X)(X(t + T)-X)
T=0
Cx{t) =
J2(x(r)-xy
T=0

where X is the mean value of pitch variations. The graph of the correlation function is
shown in the next figure.

5
1. Harmonic Models

20 40 60 80 100 120

The corresponding power spectrum is obtained by Fast Fourier Transform (FFT) of the
autocorrelation function.

b 2do '4itoi 6oo"'" addi 1000 1260

Several peaks characterize elements of musical significance, which would be re-emphasized


by the study of the other voices.

1.4 Physical Models


Physical models are based on the resolution of the wave equation in different contexts.
The wave equation describes the propagation of vibrations on a geometric object, mod-
elizing a musical instrument. It is a partial differential equation

^ l2 = c2Au + F(t,x)
dt
where t is a positive real (the time) and x an element of R 3 . This equation is subject to
initial conditions
u(0,x) = f(x)
and
-j£(09x)=g(x)

It has been shown that the wave equation admits the solution

u(t, x) = -Hct * g(x) + Dt ( -Hct * f(x)) + / -Hc{t_s) * Fs(x)dx

6
Physical Models

where Fs(x) = F(s,x) and Ht(x) = H(t,x) verifies

<^r sin£|MI
\\v\\
and ^
A # t ( t O = cos(t||t/||)

Vibrations of Strings. The equations describing the vibrations produced by an


instrument admit periodic solutions whose frequencies are a model of partials of the
instrument. The first historical equation is the one dimensional wave equation solved by
d'Alembert in 1746. This equation describes the small displacements u(t, x) of a string

d2u _ 2d2u
dt2 " ° dx2
Suppose that the string of length £ is fixed at both ends. With the boundary conditions
u(0,t) = u(£,t) = 0, and initial conditions u(x,0) = f(x) and du(x,0)/dt = g(x),
D'Alembert shows that the general solution of this equation is given by the sum of
two arbitrary functions representing two waveforms, one traveling to the right and one
traveling to the left, each at the velocity c.

u{x, t) = fix + ct) + g(x — ct)

Prom the boundary conditions, we get g(x) = —f(—x). Thus, the solution becomes

u(x, t) = f(x + ct) - f(-x 4- ct)

where / satisfies
f(x) = f(x + 2l) VxeR
Since / is a periodic function of period 2£9 it admits a Fourier series expansion. If the
fundamental is a sine wave
/ ( z ) = a s i n ( — +tp)

then the n-th harmonic is


, v . fnirfx + ct) \ . (WK(—x-\-ct) \
un(x,t) = asm! + <p J - asm I + (p\

. (nirct\ . /nirxx

The general solution is the sum of all possible values


oo

u(x, t) = ^An cos (unt) sin \—fj


n=\
with
U-KC
27r
OJn = ~r = fn
and
2 / / TITTX \
An = j u(x,0)sin \—g-J dx
1. Harmonic Models

It is well-known that the frequency of the n-th partial is given by the formula

fn =
Ye]j^ « = i.2,3,...
where T is the tension of the string and p is its linear density (c = y/T/p).

1) For a string fixed at one end and free at the other, the mode frequencies are given
by

f - -

Inharmonicity models are yet far from reality. At the end of the 19 th century, Lord
Rayleigh proposed to describe the ideal string as a bar of length I

d2u 2 d4^ ,-, 2 # 4 u

where K2 = I/S is the sectional moment of the bar, p is the density and E the Young's
modulus. This equation can be approximated by the Euler-Bernoulli beam equation which
is a fourth-order differential equation
d2 u 2d
4
u n

with c 2 = EI/pS. If we look for separable solutions u(t,x) = f(x)g(t), it is easy to show
that solutions are given by

u(£, x) = (A sin kx + B cos kx + C sinh kx + D cosh kx) sm(wt + ip)

with

c
2) If the string is free at both ends, the boundary conditions at x = 0, d2u(t, 0)/dx2 = 0
and d3u(t,0)/dx3 = 0 imply

-k2(D - B) sm(wt + if) = 0

and
k3(C - A)sm(wt + <p) = 0
that is C = A and D = B.
At x = £, the boundary conditions d2u(t,t)/dx2 = 0 and d3u(t,t)/dxz = 0 yield

k2 [A(smh H - sin kt) + £(cosh kd - cos hi)] sin(wt + tp) = 0

and
ks [i4(cosh kt - cos kl) + B(sinh kt - sin kt)] sin(wt + ip) = 0
That is
^4(sinh kt - sin kt) = £(cosh kt - cos kt)
,4(cosh kt - cos kt) = £(sinh kt - sin kt)

8
Physical Models

Eliminating A and B by dividing the first equation by the second, and cross-multipling
yield to
cosh2 kt — 2 cosh kt cos kt + cos2 kt = sinh 2 kt — sin 2 kt

2-2coshAtfcosAtf = 0
or
cos A cosh A = 1 (A)
with A = kt. The allowed frequencies are given by

\lc \l I EI
fn 2 2
2TT£ 27rt ]j pS

where An are solutions of the previous equation (A). Using the identities

o 1 — cos 2x
tan x = - —
1 + cos 2x
and
.o cosh 2x — 1
tanh x = — —
cosh 2x + 1
the equation (A) can be rewritten as

tan
( Y ) = ±tanh(-rr)

Solutions are of the form


An = ^(2n + l ) 7 T - ( - l ) n a n

where the numbers an tends rapidly to zero when n grows. The interval ratio between
two partials is then

Asymptotic forms for the rn ratios are [Rayleigh, 1894]

2n + 3 x 2
,2n + l

or [Hueber, 1988]
n + 3 /(n + l ) 2 + l
rn
~ n y ( n + 2) 2 + l

The frequency ratios from the fundamental is the product of each interval ratio (or the
sum when the ratios are expressed in cents) modulo the octave.
n

9
1. Harmonic Models

Numerical computations give the following values

11n An
4.7300407448
rn (cents)
1755
vn (cents)
555
2 7.8532046240 1165 520
3 10.9956078380 870 190
4 14.1371654912 695 885
5 17.2787596573 578 263
6 20.4203522456 496 759
7 23.5619449020 433 1192
8 26.70353756 385 377
9 29.84513021 347 724
10 32.98672286 315 1039
11 36.12831552 289 128
12 39.26990817 267 395
13 42.41150082 247 642
14 45.55309348 231 873 |

3) If the string is clamped at x = 0 and free at x = £, the equation becomes

l + cosAcoshA = 0 (B)

tan(-) = ±coth(-)
The frequencies are

where An are solutions of the previous equation (B). These solutions are of the form
[Rayleigh, 1894]
An = i ( 2 n - l ) 7 r - ( - l ) ^ n n = 1,2,3,.-.
The numbers /3 n tend to zero as n tends to infinity,

The asymptotic ratios are the given by

/2n + l \ 2

Vibrations of Circular Drums. The governing partial differential equation of a


vibrating membrane is
d2u 2A
2
2/d u d2u\

For a circular drum, this equation is written in polar coordinates


d2u ldu 1 d2u 1 d2u
QTI + r Qr + ^2 dd2 ~ c2 dt2

10
Physical Models

Looking for separable solutions of the form u(t, r, 6) = f(r)g(6)h(t), we obtain


/"(r) l/'(r) lg"{6) 1 h"(t)
+ +
/(r) r /(r) r2 5 (0) c 2 /i(*)

Since 5 is a 2n—periodic function, we get


/i"(0 + w2/i(«) = o
{ 5 "(0) + rc25(0) = O
and

This Bessel equation has the solutions of the form

f(r) = Jn(w-c)
where J n is the n-th Bessel function. The solutions of the wave equation are
T
u(t} r, 0) = AJn(w-) sm(wt + (p) sin(n# + i>)
If the radius of the circular membrane is i?, the solution verifies the boundary condition
f(R) = 0. It follows that w could only takes the values
c
w
— Jnim~^

where j n % m is the m-zero of the n-th Bessel function. The frequency ratios between two
adjacent partials are
Jn,m
^n,m —
Jn—lyTi

For nominal modes (m = 1), let's denote


Jn,l
rn =

and
In vn
Vn = T\T2'"rn mod U vn = 1200
In 2
where the product is taken modulo 2. The numbers vn are the ratios vn are expressed in
cents. The following values show the inharmonicity of the first nominal modes
n rn (cents) vn (cents)
1 806 806
2 507 114
3 376 489
4 300 789
5 251 1040
6 216 56
7 190 246
8 169 415
9 153 568
10 140 708

11
1. Harmonic Models

Vibrations of a R e c t a n g u l a r M e m b r a n e . For a rectangular membrane of dimen-


sions tx and £y the separable solution

u(x,y,t) = f(x)g(y)h(t)

of the governing equation gives three equations

dx2
d29(y) +, k. 22g(y)
, = 0
dx2
If the membrane is fixed on the boundary u = 0 for x = 0, x = £x, y = 0 and y = (,y, the
solutions are given by

n(c, a;, y) = [A sin wt-\- B cos tu£J sin —— sin —-—


Zx ty

where m, n are integers ra, n = 1,2,3,... The modal frequencies are

- ^ I IV lib
/n m = +
' 2y£[ 72-

Vibrations of Circular P l a t e s . The vibrations of a stiff metal plate are described


by the biharmonic equation
d2u+cVu=0
2_4
w
where c is given by
2 Eh2
12p(l - a2)
E is the Young's modulus, h the thickness of the plate, p the density of the plate and a
the Poisson's ratio, and
4 _ d4u d4u d4u
U 1+ 2 2+
~dx dx dy W
Separable solutions of the form u(t, x, y) = f(x, y)g(t) verify

V 4 / = k4 f

and
g"(t) + w2g(t) = 0
The numbers k and w are related by

k2=w = w /I2p(l-a3)
c h V £
The biharmonic equation is factorized by

( V 2 - * 2 ) ( V 2 + * 2 )« = 0

12
Physical Models

So the solutions can be written uniquely as a sum u\ + ^2, where u\ satisfies V2u = k2u
and U2 satisfies V2u = —k2u. Introducing the hyperbolic Bessel functions

leads to the following solutions

u(t, r, 6) = [A Jn(kr) + BIn(kr)} sin(wt + y) sin(n0 +1>)

1) If the plate of radius a is clamped at r = a, the boundary conditions require u = 0


and du/dr = 0 leading to

f j4Jn(fca) + B I n ( f c a ) = 0
\ Aj;(fo) + fl/;(fca)=0
Allowed solutions of k depend on integers n, m. They are asymptotically given by
(n + 2ra)7r
£•
"'n,?! 2a
when n tends to infinity. Modal frequencies are

f -—k2

The fundamental mode is obtained for n = 0 and m = 1 with a/co,i « 3.189

- ch
/o,i « <*-«2
a'
with a « 1.618. The ratios a n>m of the modal frequencies

Jntm — ^n,m/0,1

are given in the following table

I n TO &&n,m ^n,7n a n ,m (cents)


1 1 4.612 2.08 1268
2 1 5.904 3.41 2124
0 2 6.306 3.89 2352
3 1 7.130 5.00 2786
1 2 7.801 5.95 3087
4 1 8.328 6.82 3324
2 2 9.400 8.28 3660
0 3 9.425 8.72 3749

2) If the plate has a free edge, the fundamental mode is now obtained forn = 2, m = 0
and

with ft « 0.836. The ratios / ? n m of the modal frequencies


=
/n,TO Pn,m/2,0

13
1. Harmonic Models

are given in the following table.

1 n m Pn,m Pn.m (cents)


0 1 1.73 949
0 2 7.34 3451
1 1 3.91 2361

i—l
2 11.40 4213
2 1 6.71 3296
2 2 15.97 4797
3 0 2.328 1463
3 1 11.07 4162
4 0 4.11 2447

3) If the plate has a simply supported edge, it can be shown that the fundamental
mode is obtained for n = 0, m = 1 and
j, ch

with 7 & 0.792. The ratios 7 n m of the modal frequencies

Jn>m '7n,77i/0,l

are given in the following table for the first values of n and m

n m / n,m 7n.m (cents)


0 2 5.98 3096
0 3 14.91 4678
1 1 2.80 1783
1 2 9.75 3942
1 3 20.66 5243
2 1 5.15 2837
2 2 14.09 4580
2 3 26.99 5705

Stochastic Differential Equations. Stochastic integral relatively to a Brownian


motion is the quadratic limit of the following quantity according to the subdivision
to = a < ti < ... < tn < tn+i = b of the interval [a, b]

f f{t,w) dW(t,w) = lim^fiU.wXWiU^) - W{ti))

This definition proposed by ltd leads to a new differential calculus. For example, it is
well known that in the integral

Ws dWa = \w? - I
Jo
it appears a supplementary term t/2. This equation could also be rewritten in the dif-
ferential form
dWf = 2WtdWt + dt

14
Sonance Models

More generally, for a stochastic process Xt under suitable conditions, we consider the
stochastic differential equation

dXt = a(t, Xt)dt + b(t, Xt)dWt

The ltd formula gives the drivative of /(£, Xt)

M(t Y\ (^IMA . n(t y Mt,xt) i.2(. d2f(t,xt)


df(t,Xt) = (—^—+o(t,X0 dx +^b(t,Xt) Qx2 )dt

+b(t,Xt)MhMdWt

In particular, for a Brownian motion Wt, we have

df(Wt) = f'(Wt)dWt + \f"{Wt)dt


The density p of the process Xt is governed by the Fokker Planck Equation

dp = ld2(b2(t,y)p) _ d(a(t,y)p)
dt 2 dy2 dy
We have seen in the first chapter, that the governing partial differential equation of a
vibrating membrane is
d2u 2.

If the vibrating membrane is disturbed by a random noise £(£), the equation becomes

and could be treated as a stochastic differential equation. The solutions of the stochastic
equation are new frequencies which tend to the original frequencies when noise goes to
zero. In some cases, inharmonicity could be interpreted as a random perturbation.

1.5 Sonance Models


Sonance is a general concept to term consonance or dissonance. There are several defin-
itions and several models and the phenomena is not well understood. It is not clear if
consonance is the opposite sign of dissonance. James Tenney [Tenney, 1988] discussed
at least five distinct ways of sonance: melodic, polyphonic, contrapuntal, functional and
sensory. Melodic consonance refers to successive notes and depends on their surround-
ing melodic context. Polyphonic consonance focuses on the vertical aspect of the music.
Contrapuntal consonance refers to its role in counterpoint, while functional consonance
tries to establish relationships between tones and fundamental root, such as basse fonda-
mentale [Rameau, 1722], real or virtual tonic, usually based on overtones series. Sensory
consonance is usually credited to Helmholtz and refers, at the origin, to the presence or
absence of beats.
Pythagoreans focus on the numerical properties of musical intervals, and wishes to
explain the laws of nature by the arithmetic of integers, presuming that human prefers

15
1. Harmonic Models

simple frequency ratios. The simplest ratios, such as octave (2:1), fifth (3:2), fourth (4:3),
major third (5:4) and minor third (6:5), are consonant intervals. A simple ratio is defined
as a ratio involving small integers or, for some theorists, small prime numbers. In the
Pythagorean meaning, harmonic laws govern also the motion of the planets. The whole
cosmos is embedded in numbers.
Some theorists explain consonance by tonal fusion [Stumpf, 1898] when tones fuse per-
ceptually: Two tones are consonant if they sound like a single tone. Others evoke amp-
litude fluctuations, rapid beating [Helmholtz, 1877], virtual pitches [Terhardt, 1974], ex-
pectation violations [Cazden, 1980] or tonotopic interferences [Greenwood, 1961], [Plomp,
1965], [Kameoka, 1969].
As described in Sensations of Tone, Hermann von Helmholtz [Helmholtz, 1877] in-
troduced the notion of sensory consonance and dissonance by comparing violin tones
sounded together. Helmholtz calculated the resulted dissonance by considering rough-
ness as a function of beats. Fast beats are considered to be rough with a maximum at 32
beats per second. Helmholtz drew dissonance curves by plotting roughness produced by
partials against intervals. Maximal consonances are obtained for the unison, the octave
and the fifth, followed by fourth, major sixth and major third. For each frequency in the
range of an octave, dissonance curves are formed by interactions between a given partial
of the note and the nearest partial of a referenced note labelled c' [see Helmholtz, p. 193].
Like Helmholtz, Harry Partch has drawn a graph of comparative consonance, the One-
Footed Bride [Partch, 1949]. It shows the relative consonance of each interval of his 43
just tone scale. The graph is symmetric around 600 cents and ratios (r, 2/r) are coupled.
"In this exposition 2:1, 3:2 and 4:3 (ratios of the 3-limit) are the Inter-
vals of Power. Those ratios that lie between 4:3 and 3:2 ("tritone" intervals)
-27:30, 11:8, 7:5, 10:7, 16:11, 40:27 are the Intervals of Suspense. The ra-
tios between 21:16 and 7:6, at the threshold of the descent to 1:1, and those
between 32:21 and 12:7, at the threshold of the ascent to 2:1 are the Emotional
Intervals. The ratios starting with 8:7 and descending to 1:1 and those start-
ing with 7:4 and ascending to 2:1, are the Intervals of Approach." [Partch,
1949, p. 156-157].
The tonotopic theory of sensory dissonance was first proposed by Greenwood [Green-
wood, 1961] and extended by Plomp and Levelt [Plomp, 1965], and Kameoka and Kuriy-
agawa [Kameoka, 1969]. In the mid 60's, Plomp and Levelt [Plompt, 1965] asked a large
number of listeners to judge the consonance of a variety of intervals, when sounded by
pairs of pure sine waves. They showed that the interval of maximum sensory dissonance
remains fairly constant relative to the critical band and they observed that in almost all
frequency ranges, the maximal point of roughness occurred at about 1/4 of the critical
bandwidth. The data of the experiment shows an averaged dissonance function of the
form
h(x) = e~^x - e~^x
with /?2 = 3.51 and P1 = 5.75. They also calculated the consonance for two tones, each
with six harmonics. Their results were confirmed in experiments carried out by Kameoka
and Kuriyagawa [Kameoka, 1969] who have proposed a model of calculating the absolute
dissonance of complex tones. A mathematical model is designed in the book of Sethares,
see [Sethares, 1998, p. 300].
Let S be a spectrum with partials at frequencies f\ < $2 < ... < fn and respective
amplitudes ai, <22, ..., an. Sethares defines the dissonance of the sound S by the sum of

16
Geometric Models

the dissonance of all pairs of partials. Let H(r, s) be the function


n
1
2
- i

with
„, *\rfj-sfi\
13
a2mm(rfj,sfi) + a1
where h(x) = e~^x — e~P^x is the previous function, the constants K = 0.24, a\ = 18.96,
a2 = 0.0207 and n is the number of partials. The dissonance of S at an interval r is given
by
D(r) = H(l, 1) + H (r, r) + H(l, r)
The dissonance of a chord of three notes at the intervals 1, r and s is defined by

D(r, s) = D(r) + D(s) + H(r, r) + # ( s , 5) + H(s, r)

Assuming that all amplitudes are equal (a* = 1), it is easy to show that the dissonance
curve D(r) has at most 2n 2 local minima. These minima tend to be located at intervals
r for which r = fa/fj for some partials S. Remark that we could define several models
by changing the function h(x).
Jean-Marc Chouvel [Chouvel, 1998, p. 156] defined the concordance of two partials by
introducing a Gaussian function. For the interval r, the concordance is

and is expressed in decibels by

CdB{r) = mogl0C(r)

For three sounds (1, r and 5), the concordance is defined by

C ( r , 5 ) = C ( r ) + C( 5 ) + C ( r - 5 )

For several sounds, Chouvel obtains dissonance graphs similar to Sethares plots.

1.6 Geometric Models


Geometric models have been studied by Guerino Mazzola, Daniel Muzzulini and Thomas
Noll. In this section, we follow [Mazzola, 1985, 2003].
Definition 1 Let R be a commutative ring. A local composition is a pair (L, M) where
L is a non-empty finite subset of M and M a R-module called the ambient space.
In Topos of Music [Mazzola, 2003], a local composition is a denotator whose form is
of power type. The ambient space is called the coordinator.
P r o p o s i t i o n 2 The collection of local compositions Loc# is a category whose objects
are local composition (L, M) and morphisms ^ : (L, M) —> (if, N) are homomorphisms
between ambient spaces from M to N, ty £ Hom(M, N).

17
1. Harmonic Models

Examples. Chords, scales, motives, onsets and most of the usual musical structures are
local compositions. The chords (^4, Z n ) modulo the octave form a local composition over
the n-tone equal temperament R = Zn. For n = 12, we recover the space of classical
harmony.

The following theorem has been established by [Mason, 1970] and generalized in
[Mazzola, 1985]. Pcsets are defined in the chapter 4.1 on Set Theory.

Theorem 3 Let n be a positive integer, and let A, B be two pcsets of same cardinality
m. Then there are two non-negative integers k (called sharps) and I (called flats) with
k-\-l = m — n and k>l>0 and a set of integers called alterations

JBT = {—£, —I + 1 , . . . , fc — 1, *:}

such that any pc-set A can be translated to any equipollent pcset B by some alterations.
In particular, in the classical temperament (n=12) any scale of seven notes is mapped to
any scale of the seven elements by 3 sharps and 2 flats.

Example. Remark that the definitions of sharps and flats can be exchanged. For example,
A = {0,1,2,3,4,5,6} and B = {0,1,7,8,9,10,11}, the subset {0,1,2,3} of A is mapped
to {10,11,0,1} by the translations of —2, and the subset {4,5,6} is mapped to {7,8,9}
by the translations of 3.

Definition 4 Let (S, M) be a local composition over the ring R. The left module R.S
is generated by the set of {s — x,s G S} for any element x of S. The module is free if it
admits a basis.

Proposition 5 The map F : S —> R.S is a functor from the category of local composition
LOCR to the category of modules ModR.

Mazzola defines also the category of ObLoc of objective local compositions whose
objects are the collection of all local compositions and morphisms are morphisms between
local compositions. He has shown that there is a fully faithful functor from ObLoc to Loc.

Definition 6 Let R be a commutative ring and (L, M) a local composition. A global


composition (7, (j)t) is defined by a finite non-empty covering I = (It)teT of L indexed
by the parameter t G T and an atlas of charts (cj)t): Lt —• It such that for each pair (s,t)
the maps (j)3jt = <pj ocj)t

Kt' {<i>i\hnis),Mt) -> (<t>:\itnislMs)


induce an isomorphism of local compositions (Lt^Mt) over the ring R.

Definition 7 Let (L, M) be a local composition. An interpretation L^ is the canonical


global composition defined by the trivial covering of L = \\{i} and the charts of the atlas
iei
are the identity on I. A sub-interpretation is a subset of an interpretation.

Example. In classical harmony, let C be the C-major scale

C= {0,2,4,5,7,9,11}

18
Geometric Models

The trichords interpretation of C is the set of degrees

C7(3) = {/c, He, IHc, IVc, Vc, VIC, VIIc]


where the degrees are defined as usual Ic = {0,4,7}, He = {2,5,9}, IIIc = {4,7,11},
IVc = {0,5,9}, VC = {2,7,11}, Vic = {0,4,9}, VIIC = {2,5,11}.
Definition 8 Let (I, </>t) and (J, ipt) be two global compositions with I a covering of K
and J a covering of L over the ambient space M. ^4 morphism F : (7,</>t) —• (J,ipt) of
global compositions is a pair F = (/, g) where f : I —> J and g : K —> L such that:
(l)Viel, g-.i^ m
(2) The map ij)s o g o </>t : Kt —> Lt is a morphism of local compositions
Definition 9 The category Glob of the global compositions is defined in the follow-
ing way. The objects of Glob are the collection of global compositions. The morphisms
between global compositions define the morphisms of the category of global compositions.
Definition 10 The geometric interpretation of a global composition is the simplicial
complex of the trichords interpretation.
Example. The simplicial complex S(K^) of the diatonic scale is the Moebius band.
II V I IV VII

VII HI VI II

Definition 11 Let (K, M) be a local composition and Kj a triadic sub-interpretation.


A subset J c I is called cadential if it has a non-ambigous sub-interpretation Kj on the
set of all interpretations of K, i. e. if it does not exist a translation Tn (n ^ 0) such that
Kj = (TnK)j. The set J is a minimal cadential set if it has no proper subset which is
also cadential.
Example 1. Let C be the C-major scale. C = {0,2,4,5,7,9,11}. The set J = {1,11} is
not cadential because its interpretation is ambigous

{Ic, He] = {0,2,4,5,7,9} = {VF,VIF}


Usually, minimal cadential sets are determined by computer. Daniel Muzzulini [1995] has
calculated all minimal cadential sets for different local compositions. Guerino Mazzola has
shown that for the well-tempered tuning M = Z ^ , if K is the major scale, there are only
five minimal cadential sets Jx = {IIJII}, J2 = {II, V}, J 3 = {IIIJV}, J4 = {IV, V},
J5 = {VII}. For the harmonic minor scale K = {0,2,3,5,7,8,11}, there are 21 minimal
cadential sets. Every pair of two different degrees defines a minimal cadential set.

Example 2. In the diatonic Pythagorean tuning, the frequency ratios are generated by
pure fifths {C (1), D (9/8), E (81/64), F (4/3), G (3/2), A (27/16), B (243/128)}. The

19
1. Harmonic Models

notes are represented by the powers of 3 in the previous ratios: F (-1), C (0), G (1),
D (2), A (3), E (4), B(5). For the major scales, the minimal cadential sets are the five
same sets of the 12-tone equal temperament. For the harmonic minor scale, there are
16 minimal cadential sets, the singleton {III} and all pairs of any two different degrees
different of III. For the melodic minor scale K = {0,2,3,5,7,9,11}, there are 10 minimal
cadential sets Jx = {III}, J2 = {I, II}, h = {I,V}, J 4 = {I,VII}, J5 = {II,IV},
Je = {II, VI}, J7 = {IV, V}, J 8 = {IV, VII}, J9 = {V, VI}, Jio = {VI, VII}.

Example 3. In the Euler Plane, the frequency ratios are given by {C (1), D (9/8), E (5/4),
F (4/3), G (3/2), A (5/3), B(15/8)}. In the basis <3,5>, the notes are represented by the
coordinates determined by the powers of 3 and 5 in the previous ratios, namely C (0,0),
D (2,0), E (0,1), F (-1,0), G (1,0), A (-1,1), B (1,1). The major scale has six minimal
cadential sets Jx = {II}, J2 = {VII}, J 3 = {III,VI}, J4 = {III,IV}, J 5 = {IV,V},
JQ = {V, VI}. The melodic minor scale has also six minimal cadential sets, J\ = {I},
J2 = {II}, J 3 = { / / / } , J 4 = {VI}, J 5 = {VII}, J 6 = {IV,V}. The harmonic minor
scale has 9 minimal cadential sets J\ = {II}, J2 = {HI}, ^3 = {VII}, J4 = {I,IV},
h = {I, V}, J6 = {I, VI}, J7 = {IV, V}, Js = {IV, VI}, J9 = {V, VI}.

Definition 12 A scale K is rigid if its symmetry group Sym(K) is trivial.

Definition 13 Let K be a scale, L = Tn(K) the translated scale and if3, L3 be two tri-
adic sub-interpretations. A modulator is a map defined by the composition of the transla-
tion Tn with an inner symmetry of K which induces an isomorphism between the triadic
sub-interpretation K3 ~ L3.

Example. Let K = {0,2,4,5,7,9,11} be the C-major scale, L the C() major scale L =
Ti(K). The map s(x) = 4 — x defines an inner symmetry of K. Let K3 = {He, Vc} =
{2,5,7,9,11} and L 3 = {IIc$, Vc$} = {0,3,6,8,10} be two sub-interpretations. The map
m(x) = T\ o I(x) = 5 — x defines an isomorphism between K% and L3.

Definition 14 Let K, L be two different triadic interpretations. A modulation from K


to L is a pair (m, J) consisting of a modulator m for (K, L) and a minimal cadential set
J.

Definition 15 A modulation (m, J) is quantized if it exits a set Q called the modulation


quantum such that
(1) The modulator m is an element of Sym(Q)
(2) All triads in J are subsets of Q.

Proposition 16 The setTnQ is covered by triads ofT^y The degrees of the interpret-
ation (T fl Q)(3) of the trace Q inT are called the pivots.

Example. The C-major scale is denoted K = {0,2,14,5,7,9,11} and the C(t major scale
L = {0,1,3,5,6,8,10}. Let J = {II, V} be a minimal cadential set of J in the 12-tone
equal temperament. Let

K3 = {IIC,VC} = {2,5,7,9,11}

and
£3 = {IIct,Vct} = {0,3,6,8,10}

20
Affinity Models

be two sub-interpretations. The set


Q = J UTi( J ) = {0,2,3,5,6,7,8,9,10,11}
is a modulation quantum. The set
T H Q = {0,3,5,6,8,10} = {IICh Illcfr VCh VIIc*}
is rigid and the pivots are the degrees
{II, III, V, VII}
The last theorem of this section has been demonstrated by computer. It has been
stated by Daniel Muzzulini.
T h e o r e m 17 In the classical temperament, all the seven elements scales have at least a
modulation quantum except the scales M52 = {0,1,2,4,6,7,8} andM55 = {0,1,2,3,4,7,9}
for translation parameter n = 1,11, and the scales M%% = {0,1,2,3,4,5,6} and MQ2 =
{0,1,2,4,6,8,10} for translation parameter n = 5, 7.

1.7 Affinity Models


In the mid 50's, Edmond Costere [Costere, 1954] proposed a theory of harmony based on
a model of affinity of tones. Twenty years later, E. Terhardt [Terhardt, 1974] proposed
another interpretation of the affinity of tones. In his model, Costere affects a weight of
one unit to the ascending and descending minor second, fifth and octave or unison and
develops an elementary algebra on weighted sets. The elements of the well tempered
system are identified with X = Z/12Z = {0,1,2,..., 11}, namely (C = 0, Ctf = 1, D = 2,
..., B = 11). The cyclic group C12 acts on the pitch class sets F(X) = {0,1} X . The
quotient A = F(X)/C\2 defines the set of musical assemblies, also called class sets or
simply sets. A perfect chord is a major or a minor triad, that is to say, respectively, a
transposition of [0, 4, 7] or a transposition of [0, 3, 7]. The set of perfect chords is denoted
by P.
Definition 18 The affinity ofiGX relatively to an element j € X is
w(ij) =l 1 / ^ - 1 = 0,1,5,7,11 (mod 12)
and w(i,j) = 0 otherwise. The affinity of j G X relatively to a set L £ A is

The numbers R = {^(j)}jex define the affinity vector of the setL. A collection of affinity
vectors is called an affinity table.
Definition 19 Let R = {^s(j)}jex be the affinity vector of the set L, the cardinal density
(or tonal gravity,) of a set B G A is

d{B) = 5>(i)

The cardinal density of each perfect chords defines two tables, one for the major triads
and another for the minor triads.

21
1. Harmonic Models

Example. The pentatonic scale (called the Chinese pentatonic scale by Costere) L =(C,
Oji, F&, Gtt, Bb)={0, 3, 6, 8, 10}

'<£ „ J. II- ^ ^
has the following affinity table:
C c« D Dtt E F Ftt G Gtt A Bb B
c 1 I 0 0 0 1 0 1 0 0 0 1
DB 0 0 1 1 1 0 0 0 1 0 1 0
0 1 0 1 1 1 0 0 0 1
n
Gfl 0 1 0
0
1
0
0 0 0 1 1 1 0 0
Bb 0 0 0 1 0 1 0 0 0 1 1 1
R 1 (3 1) 3 (1 3) 1 (3) 2 (2) 2 (3)

The last row is the affinity vector. Parentheses indicate the notes that are not belonging
to the set L. For each major chord, we compute the tonal gravity. The set Mi = [0,4,7]
has the density d(M\) = 1 + 1 + 3 = 5, the set M 2 = [1,5,8] has the density d(M 2 ) =
3 + 3 + 2 = 8, etc. The same is done for the minor chords. The set mi = [5,8,0] has the
density d(rai) = 3 + 2 + 1 = 6, the set m 2 = [6,9,1] has the density d(m2) = 1 + 2 + 3 = 6,
etc. Results are presented in the following table.

[W 8 4 8 6 6 6 7) 6 (6 6 ~W]
\ (6 6 6 8 4 8 5 7 6 6) 6 (7) |

The first row is the density of the major triad from the first chord based on 0 to the
chord based on 11. The second row is the minor chord from the chord finishing by 0 to
the chord finishing by 11. Numbers outside parentheses indicate chords for which all the
notes of the chord belong to the set L. The major chord [8, 0, 3] belongs to L, as well as
the minor chord [3, 6, 10].

Remark that additional weights are given by Costere to reinforce some notes (for
example if a note is repeated or doubled). The affinity table is also applied to microtones
(quarter tones) in [Costere, 1962, p. 35] or to other musical objects [Costere, 1974, p. 184].
It is clear that there are several models according to the definition of the primary weights.
For example, Terhardt considers four relations: the descending fifth, the descending major
third, the ascending and descending major second. In this case, the affinity table of the
pentatonic scale would be

C ctt D Dtt E F Ftt G GJJ A Bb B


c 1 0 1 0 0 1 0 0 1 0 1 0
Dtt 0 1 0 1 0 1 0 0 1 0 0 1
Ftt 0 0 1 0 1 0 1 0 1 0 0 1
GB 0 1 0 0 1 0 1 0 1 0 1 0
Bb 1 0 0 1 0 0 1 0 1 0 1 0
R 2 (2 2) 2 (2 2) 3 (0) 5 (0) 3 (2)

Costere uses the tonal gravity to partition the musical set A in four distinct categories.
Let L be a set of A. Define the following maxima. The first maximum is taken over all

22
Affinity Models

perfect chords belonging to the set L

u = max d(t)
v
tcL '
and the second maximum is taken over all perfect chords that are not included in the set
L.
v = max d(t)

Definition 20 A set L is tonally stable or simply stable (S) if the maximum of the
tonal gravity of the constitutive perfect chords is greater than the maximum of the tonal
gravity of the non-constitutive perfect chords, in others words if u> v, i.e.

max d(t) > max d(t)


tcL t£L

Definition 21 A set L is balanced (B) if the maximum of the tonal gravity of the con-
stitutive perfect chords is equal to the maximum of the tonal gravity of the non-constitutive
perfect chords, in others words ifu = v, i.e.

max d(t) = max d(t)

Definition 22 A set L is cadential (C) if the maximum of the tonal gravity of the
constitutive perfect chords is less than the maximum of the tonal gravity of the non-
constitutive perfect chords, in others words ifu<v, i.e.

max d(t) < max d(t)


t(ZL t^L

and if the maximum v is obtained for only one triad, that is

3\teV, d(t) = max d(t)


t<£L

Definition 23 A set L is jarred (J) if the maximum of the tonal gravity of the con-
stitutive perfect chords is less than the maximum of the tonal gravity of the non-constitutive
perfect chords, in others words ifu<v, i.e.

max d(t) < max d(t)


tcL t<£L

and if the maximum v is obtained for several distinct triads, that is

3tlyt2eV, t1^t2} d(ti) = d(t2) = max d(t)


t<£L
Theorem 24 The set of the 351 musical sets A from 1 to 12 notes has a partition in
four distinct categories: 81 stable sets, 102 balanced sets, 84 cadential sets and 84 jarred
sets.
A =S U B U C U J

23
1. Harmonic Models

Costere also distinguishes between tonic (T) and non tonic sets (T c ). There are 213
tonic musical sets and 138 non-tonic sets, with the following division, see [Costere, 1954].

s B C J
T 51 65 48 49
rpc 30 37 36 35
T U Tc 81 102 84 84
In the vocabulary of Costere, some properties are structural (not depending on affin-
ity) and others are cardinal properties (depending on affinity). Atonality is a structural
property and polytonality a cardinal property. We investigate first some structural prop-
erties.
Definition 25 A set L has limited intervals if one interval does not belong to L.
For example, the set L = [0, 4, 8, 10] has no fifth. The set [0, 7, 9] has no major third.
Definition 26 A transposition is a map Tn from A to A, defined by
Tn(x) = x + n mod 12
Definition 27 A set L has limited transpositions if two transpositions give the same
set.
For example, the whole-tone scale L = [0, 2, 4, 6, 8, 10] has only two transpositions L
and Ti(L) = [1, 3, 5, 7, 9, 11]. We will explore this concept in details in chapter 3.
Definition 28 A set L has a median symmetry if placed on the unit circle (by the
application k £ L —> (cos(2A;7r/12),sin(2fc7r/12)) a diameter is a symmetry axis.
For example, the set L = [0, 1, 5, 6, 9] has a symmetry around the axis (3, 9).
Definition 29 The inversion of a set L is the map I from A to A, defined by
I(x) = —x mod 12
and the inversion of order n is
In{%) = —x + n mod 12
Definition 30 A set L is reversible if X\L or X\I(L) is a transposition of L. In other
words if it exists m^0 such that
Tm(X\L) =L or Tm(X\I(L)) = L
For example, the set L = [0, 2, 4, 7, 9, 11] is reversible because TQ(L) = X\L. The set
L = [0, 3, 4, 5, 7, 11] is reversible because Ti(X\J(L)) = L.
Theorem 31 There are 94 limited intervals sets, 15 limited transpositions sets (excluded
the chromatic scale), 50 sets with median symmetry and 32 reversible sets with the fol-
lowing division
S B C ~T\
Limited intervals 22 25 20 27
Limited transpositions 3 0 0 12
Median symmetry 16 6 8 20
Reversible 8 3 12 9
where S are the stable sets, B the balanced sets, C the cadential sets and J the jarred sets.

24
Affinity Models

In Costere's theory, cardinal properties are considered in order to investigate the stability
and tonality of sets.
Definition 32 A set Lofn notes is dense or cardinally stable if the density of the set
L is greater or equal to the maximum density of n notes not belonging to the set

d{L) > max d(N)


N<£L,\N\=n
otherwise L is transitive.
For example, the set L = [0, 6, 9] is transitive. The affinity vector of L is

1 (2 1 0 1 2) 1 (2 I 1) 1 (1 I 2)
The density of the set L is d(L) = 1 + 1 + 1 = 3. It is less than the density of [1, 5, 7].
The Tristan chord L = {F, B, Dtf, G(t} has the following affinity table

C Ctf D D|J E F Fit G G|l A Bb B


F 1 0 0 0 1 1 1 0 0 0 1 0
B 1 0 0 0 1 0 1 0 0 0 1 1
D|t 0 0 1 1 1 0 0 0 1 0 1 0
GJJ 0 1 0 1 0 0 0 1 1 1 0 0
R (2 1 1) 2 (3) 1 (2 1) 2 (1 3) 1

The density of the weights of the set is d(L) = 2 + 1 + 2 + 1 = 6. Let N be the set
N = {C, E, Gb, Bb}, then d(N) = 2 + 3 + 2 + 3 = 10. Thus the Tristan chord is cardinally
unstable or transitive.
Definition 33 A set L has a tonic pole if it exists i G L such that

9 ( 0 = max 9(j) > max 9(j)


j€L jex\L

A set L has an extrinsic pole if it exists i G X\L such that

^(i) = max ^(j) > max Q(j)


jex\L jeL
Definition 34 A set L is cardinally balanced if

Vi G L, 3(i) = max S(j)

For example, the set L = [0, 6, 9] of the previous example is cardinally balanced,
because each element of L has the same affinity equal to 1 (see the affinity vector above).
Definition 35 A set L is modulant if there is equality of the affinity for all notes of
at least one perfect chord. In other words, L is modulant if for all i G L, one of the
following relations is true
(1) 3(i) = 3(t + 3) = 9(i + 7)
(2) 3(i) = 9(i + 4) = 9(t - 7)
(3) 9 ( 0 = 3(» - 3) = 3(t + 7)
(4) 3(i) = 3(z - 4) = 9(i - 7)
where the sums are taken modulo 12.

25
1. Harmonic Models

For example, the set L = [0, 4, 7, 9] is modulant. The affinity vector of L is

2 (1 2 1) 2 (2 1) 2 (2) 2 (1 2)

One could verify that for each element of L the second relation of the previous definition
is true.

Theorem 36 There are 105 transitive sets, 152 sets with tonic poles, 33 cardinally bal-
anced sets and 159 modulant sets with the following division

S B C J\
Transitive 4 15 42 44
Tonic poles 43 42 39 28
Cardinally balanced 8 4 6 15
Modulant 55 51 31 27

where S are the stable sets, B the balanced sets, C the cadential sets and J the jarred sets.
Investigations around tonality, modality, polytonality and atonality combine structural
and cardinal properties.

Definition 37 A set L is atonal if it does not contain a fifth or a major or a minor


triad.

For example, the whole-tone scale L = [0, 2, 4, 6, 8, 10], the diminished seventh chord
[0, 3, 6, 9] the cluster [0,1, 2, 3, 4] and the set [0, 1, 2, 4, 10] are atonal sets.
Definition 38 A set L is neutral if it contains only neutral fifth, that is fifth without
associated minor or major thirds.

For example [0, 1, 2, 6, 7, 8], [0, 1, 2, 7, 8, 9] and [0, 1, 2, 3, 4, 5, 6] are neutral sets.
Definition 39 A set L is binary when it contains (at least) one couple formed by the
major and the relative minor triad.

A binary set could contain several couple of major and minor triad. For example
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] contains six couples. Recall that if [z, i + 4, i + 7] is a major
triad, the relative minor triad is the set [z, i + 3, i + 7] (mod 12). The set [0, 1, 2, 4, 8, 9]
is a binary set. It contains a major chord [9, 1, 4], the relative minor chord [9, 0, 4] and
a minor chord [1, 4, 8].

Definition 40 Let V+ be the set of major perfect triads and V be the set of minor
perfect triads. The inverse minor map opp from V^ to V~ associates for all major triad
t = [i, i + 4, i + 7], the opposite perfect triad opp(t) = [i — 7, i — 4, i\.

Definition 41 A set L is major if L is not binary and verifies

3 t e T+ fl L, d(t) > max d(u)


uev-nL
Definition 42 A set L is minor if L is not binary and verifies

3 t € V~ fl L, d(t) > max d(u)

26
Affinity Models

Definition 43 A set L is majinor (major+minor) if L is not binary, L has the same


number of major and minor perfect triads, and
VtGP+flL, 3! seVnL, d(t) = d(s)
Theorem 44 There are seven distinct classes of sets with the following division
S B C ~JH
Major 4 15 13 13
Minor 4 15 13 13
Majinor 13 5 6 12
Polytonal 19 28 24 9
Neutral 19 19 9 6
Binary 22 20 10 10
Atonal 0 0 9 21
The sum of each column gives the number of stable (S), balanced (B), cadential (C) and
jarred (J) sets.

Comparision of weight systems are not easy. For example, if we take the chord {G, B,
D, F} we could calculate the affinity table
Gfl Bb
G
C
1
CJ
0
D
1

0
E
0
F
0
ni G
1 1
A
0 0
~B1
0
B 1 0 0 0 1 0 i 0 0 0 1 I
D 0 1 1 1 0 0 0 1 0 1 0 0
F 1 0 0 0 1 1 1 0 0 0 1 0
E (3 1) 2 (1 2) 1 (3) 2 (1 1 2) 1

and deduce the densities of the major and minor perfect chords, namely
Gtf Bb
Maj
C
7

3
D
6

5
E
4
F
5
n6 G
5 5
A
4 5
B
5
Min 5 5 6 3 6 4 6 6 4 4 6 5

We see that the only major chord of highest density is {C, E, G} and that they are 5
minor chords of density 6. Two chords are related to fifth affinity: {C, Eb, G}, {A, C,
E} and three chords are related to chromatic affinity, {Eb, Gb, Bb}, {G, Bb, D} and
{B, D, F§}. We now look at the Terhardt weights, the affinity table is
C c» D Dtt E F Fit G G» A Bb ~B1
G 1 0 0 1 0 1 0 1 0 1 0 0
B 0 1 0 0 1 0 0 1 0 1 0 1
D 1 0 1 0 1 0 0 1 0 0 1 0
F 0 1 0 1 0 1 0 1 0 0 1 0
E (2 2) 1 (2 2) 2 (0) 4 (0 2 2) 1

and the densities of the perfect 3-chords are


C CB D Dtt E F Ftt G GJJ A Bb B
Maj 8 4 3 8 3 6 4 6 4 6 4 3
Min 4 4 7 3 6 6 2 8 4 5 4 7

27
1. Harmonic Models

Two major chords are dominant of weight 8 {C, E, G} and {Eb, G, Bb}. The only minor
chords of the same density is the chord {C, Eb, G}.

Consider for example the first prelude extract from the 24 Preludes of Maurice Ohana.
The first measure is a cluster of eight notes (Bb, C, A, B, D, Ab, Db Eb). The affinity
table calculated with the Costere weights shows an homogenious repartition of the affinity
between 3 and 4 for the pitches of the cluster:

c c» D DJ E F F» G GB A Bb B
Ab 0 1 0 1 0 0 0 1 1 1 0 0
A 0 0 1 0 1 0 0 0 1 1 1 0
Bb 0 0 0 1 0 1 0 0 0 1 1 1
B 1 0 0 0 1 0 1 0 0 0 1 1
C 1 1 0 0 0 1 0 1 0 0 0 1
Db 1 1 1 0 0 0 1 0 1 0 0 0
D 0 1 1 1 0 0 0 1 0 1 0 0
Eb 0 0 1 1 1 0 0 0 1 0 1 0
E 3 4 4 4 (3 2 2 3) 4 4 4 3

The Terhardt weights show a similar repartition, with no dominant density.

C Ctt D Dtt E F Ftl G Gtt A Bb ~B~1


Ab 0 1 0 0 1 0 1 0 1 0 1 0
A 0 0 1 0 0 1 0 1 0 1 0 1
Bb 1 0 0 1 0 0 1 0 1 0 1 0
B 0 1 0 0 1 0 0 1 0 1 0 1
C 1 0 1 0 0 1 0 0 1 0 1 0
Db 0 1 0 1 0 0 1 0 0 1 0 1
D 1 0 1 0 1 0 0 1 0 0 1 0
Eb 0 1 0 1 0 1 0 0 1 0 0 1
E 3 4 3 3 (3 3 3 3) 4 3 4 4

28
2
Taxonomy of Chords and Scales

The purpose of this chapter is to propose a general framework of chords classification us-
ing groups action and to rediscover the cyclic classification proposed by Edmond Costere
(1954), the dihedral classification proposed by Howard Hanson (1960), Zalewski and Al-
len Forte (1972), the affine classification proposed by the Czech composer Alois Pinos
(1971), and later by Daniel Starr (1978) and Guerino Mazzola (1985), and the symmetric
classification proposed by the Mexican composer Julio Estrada (1994) (see the review
[Verdi, 2003]). The pitch-class sets theory was extensively studied since the beginning of
the 20 th century, from a theoretical [Lewin, 1960], [Martino,1961], [Howe, 1965], as well as
compositional perspective [Babbitt, I960]. The theory presents the numbering of possible
combinations of the twelve pitches of the equal temperament and was widely studied by
Edmond Costere [Costere, 1954] and by Allen Forte in the 1970,s [Forte, 1973]. Forte's
theory classifies major and minor chords under the same structure and thus is limited
to non-tonal music. In Costere's theory, the inversion used in the dihedral classification
must be excluded. This then leads to a new hierarchy of sounds which no longer has
223 but 351 pitch class sets (corresponding to the cyclic or dihedral group action). We
will illustrate it by examples taken from musical score of the beginning of the century
or from works using enlarged tonality, modality or polytonality [Milhaud, 1923]. The
modes of limited transposition, which were largely used by Olivier Messiaen [Messiaen,
1944], take on a significant structural characteristic for the organization of these classes.
The classification of hexachords is particularly revealing as it shows the tropes of Josef
Matthias Hauer [Hauer, 1926] who put forth at the same time as Schoenberg a different
serial theory. Mazzola's classification in 157 classes and Estrada's classification are also
presented in this chapter. The cyclic classification which includes all the structures of the
dihedral classification, appears to be well-suited to the analysis of pieces in which the
tonal functions are more or less abandoned, like Franz Liszt's last works, Bela Bartok's
Sonates for violin and piano (the first sonata is based on C sharp minor and the second
on C major) or the polytonal works of Frank Bridge and Darius Milhaud, or the modal
works of Olivier Messiaen, Jacques Charpentier, Charles Koechlin, Henri Dutilleux, and
2. Taxonomy of Chords and Scales

even the atonal works. The complementation function is then its implicit definition. It
structures the hexachords and determines the tropes of Joseph Matthias Hauer. The
cyclic classification of Edmond Costere explains analytically and with extreme accuracy
the harmonic mutations of modern and contemporary music up to the advent of serial
techniques. Mathematical investigations and applications of group-theoretic methods in
music theory have been done by D. Halsey, E. Hewitt [Halsey, 1978], H. Pripertinger
[Pripertinger, 1992, 1993, 1997, 1999], M. Andreatta [1995, 2003, 2004] and by Th. Noll
[Noll, 2004].

2.1 Group Actions


Definition 45 A group action of a multiplicative group G on a set X is given by a
mapping G x X on X, (p, x) —> g.x which satisfies the following conditions
(1) 1.x = x, for all x e X and where 1 is the unit element of G
(2) (gh).x = g{h.x), for allg,h e G and x G X
The group action induces an equivalence relation x ~ y given by

3 g G G, y = g.x

The orbits are the equivalence classes of this relation. Two elements x, y of X are in the
same orbit if there is an element g of the group G such that y = g.x.
Definition 46 The orbits of G on X are the equivalence classes G(x) of the previous
equivalence relation
G{x) = {g.x \geG}
The set of all orbits is denoted by G\\X = {G(x) | x G X}. The action is transitive (or
G acts transitively,) if there is only one orbit
Definition 47 For each x G X, the stabilizer Gx of x is the set

Gx = {g G G I g.x = x}

The set of all fixed points of g e G is denoted by

Xg = {x G X | g.x = x}

Examples. (1) A subgroup H of G acts by multiplication from the right on G. The orbit
H(g) is the left coset gH. In other words, two elements g, g' G G are in the same left
coset of H if there is an element h G H such that g' = hg. The left coset is denoted by
G/H or gH
G/H = {gh\heH}
The set of the right cosets is denoted by H\G or sometimes Hg.
(2) A group acts on itself by conjugation G x G -» G, (#, h) \—• ghg~l. The orbit

G{h) = {ghg-1 \g<=G]

is the conjugacy class of h. The stabilizer of h is the set of its commutative elements

Gh = {g € G, gh = hg}

30
Group Actions

Lemma 48 Let H be a subgroup of G. There is a bisection between the right and the left
cosets. In particular, the sets have the same cardinality

\G/H\ = \H\G\

Proof Let g\ and #2 be two elements of G. We have

H9l = Hg2 (1)


(1) «=• gifar^H
*=> for1)-1.^)-1 e H

The map Hg —• g_1H is an embedding from G/H in H\G. It is onto because all classes
gH can be rewritten as (g~1)~1H. •

Definition 49 Le£ H be a subgroup of G. The index o / i J m G denoted by \{G : H)\ is


the cardinal of the left cosets (and of the right cosets)

\{G:H)\ = \G/H\ = \H\G\

The index of the trivial subgroup H = {1} is called the order of G

\(G:1)\ = \G\

Definition 50 A subgroup H ofGis normal if the right cosets are equal to the left cosets

V<?eG, gH = Hg

Theorem 51 (Lagrange) The order and the index of a subgroup H of a finite group
G are divisors of the order of group G

\(G:H)\ = \G\/\H\

Proof The map h —> gh is a bijection from H to gH. All classes gH forms a partition
of distinct subsets of the number of the left cosets. But the number of left cosets is the
index of H in G, so
\G\ = \(G:H)\.\H\
a
Theorem 52 The order of the orbit of x is equal to the index of the stabilizer of one of
its elements
\G(x)\ = \G:Gx\

Proof The map sending the left coset gH to the element g.x G X i s a bijection between
the left cosets of H in G and the orbit of x. •

Theorem 53 A group action of a finite group G on a set X induces a group homo-


morphism from G to the symmetric group Sx by g —• # where ~g is the mapping x —> g.x
which is called a permutation representation of G on X.

31
2. Taxonomy of Chords and Scales

Lemma 54 (Burnside) Let G be a multiplicative finite group acting on a finite set X.


The number of G-orbits is the average number of fixed points

\G\\X\ = -±-J2\X9\

where Xg is the set of all fixed points of g € G

Xg = {x € X, gx = x}

Proof

771 El G «
= 1IGI
x€X

- 1
wi^^1
' xex 9eGx

1
' g€G xeXg

1
*geG

D
In order to introduce P6lya's enumeration theorem, we need a generalized version of
this lemma. Let R be a commutative ring such that Q is a subring of R and G be a finite
group acting on a finite set X. A weight function is a function w : X —> R constant on
each G-orbit, i.e.

w{g.x) = w(x) Vg G G, Vx e X

The weight of an orbit G(x) is defined as the weight of any element of this orbit w(G(x)) =
w(x). We recover the original version of this lemma by setting w(x) = 1 for all x G X.

Lemma 55 (Burnside) The sum of weights of G-orbits is the average number of weighted
fixed points

ueG\\x ' igeGxexg

32
Pblya's Enumeration

Proof

u€G\\X u€G\\X x£u ' '


w(u)
£ £'|G(z)|

1
'x€JY

= ^ E 5>(*)
= i^iE Ew(*)
'^GG xexg
D

2.2 Polya's Enumeration


The results of this section have been established by D.L. Reiner [Reiner, 1985], R.C.
Read [Read, 1997] and H. Pripertinger [Pripertinger, 1999]. In the whole section, X^Y
are two finite sets, R is a commutative ring containing Q as a subring and G is a group
acting on X. The set of configurations Yx is the set of functions from X to Y. The group
action of G on X induces an action of G on the set Yx by G x Yx —> Yx,

(g,f)^fog-1
where p is the permutation representation of g acting on X and the weight function
h : Y —> i? induces a weight function on F x by CJ : Yx -* R

^(/)=ii/i(/(a;))
This function u is constant on the G-orbits on Yx. For each # G G,

«fo/)=n M/G" 1 *))=n ww)= w (/)


xex xex
Definition 56 Let G be a group acting on a finite set X. The cycle index of this action
is a polynomial P(G)X) o/Q[£i, ...,£|x|] defined by

1 l
geGk=l

where jfc(p) is the number of cycles of length k of the permutation g in its decomposition
as a product of independent cycles.

33
2. Taxonomy of Chords and Scales

Theorem 57 (P61ya) The sum of weights of G-orbits on Yx is given by

E «(«) = ]grEII E%)'


ueG\\Y* ' ' pGG k=l \yeY
where jk (p) is the number of cycles of length k of the permutation g in its decomposition
as a product of independent cycles.
Proof According to the generalization of Burnside's lemma, we have to compute over
the fixed point function / , the sum

£*(/) = £ n^/W)
fEY* fGY* x€X

But / is constant on the cycles of the permutation representation of g

£ w (/) = E n w(«))|ui
f&T* feY* u€<g>\\X

= n E%) Hu€<g>\\X y€Y

n E%)fc
k=l \yeY
D
Since conjugate elements have the same number of A;-cycles in their decomposition, it
is sufficient to compute the cycle index over the conjugacy classes

P(W*i,...,tlx,) = 1 ir£l c l fU*(5c)


1
' c€C k=l
where C is the set of all conjugacy classes. If for the weights of the elements of Y we take
power of an independent variable z, the power series in the P61ya's formula are called
the configuration counting series and the figures counting series. The coefficient of order
k of the first series is the number of classes in Yx of weight zk, and the coefficients of
the second series are the numbers of elements of Y of weight zk.

Example. Let G = S3 be the group of permutation of the set {1,2,3}. S3 has six
elements si=(123)=(l)(2)(3), s 2 =(132)=(l)(23), s 3 =(213)=(12)(3), s 4 =(321)=(13)(2),
s 5 =(231)=(123) and s 6 =(312)=(123). For each element, the number of cycle of length
k is ji(*i) = 3,.72(si) = 0,j 3 (si) = 0, ji(s 2 ) = 1,J2(>2) = 1,J3(«2) = 0, .71(53) =
1,^2(53) = 1,^3(52) = 1, ..., ji(se) = 0, j2(86) = 0,j3(«e) = 1. Thus the cycle index is
the polynomial
P(s3)x)(tut2,t3) = i(«? + 3ti« 2 + 2^3)
Let Y be the set Y = {0,1}, X = {1,2,3} and G = S3. The weight function is defined
on Y by w(0) = 1, w(l) = z. The sum of weights of G-orbits is by Pblya's theorem the
following counting series

34
Classification in Tempered Music

obtained from P(5 , 3,X) by substituting tk by 1 + zk. For the symmetric group Sn we
have the following result.

Proposition 58 The cycle index of the symmetric group Sn of a set X = {1,2,..., n} of


n elements is given by

where the sum is taken over all j = (j'1,.72,..., jn) verifying


n
^kjk = n
k=l

2.3 Classification in Tempered Music


In the chord counting problem of the n tone music, the set X is identified with Zn and
the group G is the cyclic C n , dihedral X>n, symmetric Sn or affine groups An- A pitch
class set corresponds to a characteristic function taking the pitches in the set to 1 and
the remainding notes to 0. There is a one-to-one correspondence between pitch class sets
and functions from X = Zn to Y = {0,1}. The set of all pcsets is denoted by Yx or
F(Zn). The group G acts on F(X) and induces an equivalence relation on fc-chords. A
pcset of length fc or a A;-chord is a subset of the Cartesian product Xk with cardinal
number k. The results of this section have been established by George Halsey and Edwin
Hewitt [Halsey, 1978] and by Harald Pripertinger [Pripertinger, 1993].

Definition 59 In the classical temperament, the twelve pitch classes (C, CJJ, D, ..., B)
are identified with Z\i = {0,1,.., 11} according to the code C = 0, CJt = 1, D = 2, ..., B
= 11. A pitch class set (pcsets) is a subset of Z12. A pitch class set of cardinality k is
called a pcset of length k or a k-chord.

This definition can be generalized to any temperament of n pitch classes.

Definition 60 Let G be a group acting on the set F(Zn) of pcsets . The elements of the
quotient set F(Zn)/G defines the pcsets classes or simply sets classes relatively to the
action of the group G A pcsets class of cardinality k is a is called a pcsets class of length
k. For the dihedral group Vn, the elements of the quotient set F(Zn)/Vn are called the
set classes under the action of the dihedral group or simply the d-classes. For the cyclic
group Cn, the elements of the quotient set F(Zn)/Cn are called the set classes under the
action of the cyclic group or simply the c-classes or musical assemblies.

Two pcsets are equivalent under the cyclic group Cn if they are reducible to each other
by transposition. A transposition is a mapping

Ta : Zn —> Zn Ta(x) = x + a mod n

Two pcsets are equivalent under the dihedral group Vn if they are reducible to the same
form by transposition or by inversion followed by transposition. The inversion is the
mapping
i": Zn —> Zn I(x) = —x mod n

35
2. Taxonomy of Chords and Scales

The composition Ta o I defines the inversion Ia of order a by

Ia : Z n —> Z n I(x) = —x + a mod n

Two pcsets are equivalent under the affine group An if they are reducible to the same
form by affine transformation, i.e. a mapping

Ma,b : Z n -> Z n Ma,b(x) = ax + b mod n

Two pcsets are equivalent under the symmetric group Sn if they are reducible to the
same partition of n. For a pcset A = [01, ...,afc], a\ < ... < aki the structure of A is the
vector [A] = [Aai,..., Aa^] where Aa* = a^\ — a; for z = 1,..., k — 1 and Aa& = ai — a/..
The partition of A is the increasing reordering of [A].

The number of A;-chord classes is the number of G-orbits which is the coefficient of zk in
the cycle index for variables
tk = 1 + zk

Proposition 61 The cycle index of the cyclic group Cn is the polynomial

d\n

where if is the Euler totient function.

Example. For n = 12, the cycle index is

^(c12lzia)(*i,.-, t12) = ^(t\2 + 4 + 2t\ + 2*2 + 2t\ + 4*12)

Substituting tk by 1 -f zk, we get the following counting series

P(c12,z12)(z) = l + 2 + 6*2 + 192 3 +43* 4 + 6625 + 80;z6


+66* 7 + 43* 8 + 19z9 + 6z10 + z11 + z12

Each coefficient of zk is the number of &-chord classes under the Cn group action. There
are, for example, 19 trichords.

Recall that the Euler's totient function ip for integer m is the number of positive
integers not greater than and coprime to m. For m = p prime, ip(p) = p — 1 since all
numbers less than p are relatively prime (= coprime) to p. If m = pa is a power of a
prime p, then <p(p) = pa(l — £). If rai and m2 are coprime, then totient function have
the multiplicative property

<£>(raira2) = <p(mi)ip(m2)

The first values are: p(l) = 1, tp(2) = 1, (p(3) = 2, (p(4) = 2, tp($) = 4, y>(6) = 2,
V (7) = 6, p(8) = 4, y>(9) = 6, ip(10) = 4, y>(ll) = 10, <p(12) = 4, etc.

36
Classification in Tempered Music

Proposition 62 The cycle index of the dihedral group Vn is given by

Pl , _ / 5P(C„,Zn) + 5 * i 4 n _ 1 ) / 2 if n is odd
{V Zn)
"' " J |P( Cn ,z„) + W r 2 ) l 2 + tT) Hn is even

Example. For n = 12, the cycle index is

fy>i2,Zi2)(*i. -,*i2) = ^ ( < i 2 + *2 + 2t43 + 2t\ + 2t\ + 4ti 2 ) + ±t?*§ + 1%

Substituting tk by 1 + zk, we get the following counting series


P
(D12M2)(Z) = l + z + Qz2 + 12z3 + 29z4 + 38z5 + 50z6
+S8z7 + 29z8 + 12z9 + 6z10 + z11 + z12

For the affine group *Ani the result is more complicated. Wan-Di Wei and Ju-Yong (see
[Wei, 1993]) have established the following results.

Proposition 63 The cycle index of the affine group An is for n = 2a

iU,z2)(*i,i2) = \(t2 + t2)

iWz4)(*i.*2>*3>*4) = l(tj + 2t2t2 + 34+2t4)

and for a > 3

•r(A2i,%2a) o2
1_ f22(a-l)-lt2a + J2 (2W-V + ^(2i"1)2a"1) #"' +
V t=i
2a~i
2 i11
5>(*) (vtr+2°-H tr - - ) (n t2>)
i=0 \fc=l

Proposition 64 For a prime number p different from 2, the cycle index P = P^ a,z a)
of the affine group An is for n = pa and a > 1

" - ; ^ ^ (!>*<<-•>(-i>r+

2 E^W^^CO *i*ri_,-i)/- (n V-I


i=0d|p-l \k=l /

where S(x) = 1 if x > 1 and 5(1) = 0.

Proposition 65 Le£ X, Y be two finite sets, G, H be two finite groups acting on X and
Y. The cycle index of the natural action of G x H on X xY is the product

P(GxHfXxY) = P(G,X) * P(HyY)

37
2. Taxonomy of Chords and Scales

The product of two polynomials


n m
fc
P(hM,...,tn) = 5> (i) r[4 , Q(*l,*2,..., *m) = X>tf)I[ #
fc=1
(t) fc=l (j)

where the sum is taken over n or m-tuples (j) = (ji,..., j m ) is defined by


n 771

P*Q = S E « 0 ) I I I l Q M ) ( M )
a 6
(i) (j) *=u=i

where icd(fc, i) is the smallest integer which is a multiple of k and I, and gcd(k, I) =
kl j lcd(k,l). In particular, for n = p\l...p^T, where pj are prime numbers, the cycle
index of Z n is the product

Example. For p = 3, the cycle index of ^ 3 is

J U A ) ( * i , * 2 , t s ) = l(t3!+Sht2 + 2t3)

The product

^ U fo) = P(A3M) * PlJUM = ^(t\2 + 3ti4 + 2t\ + 2*5*5 + 6*2*i +


4tlt6 + 12*| + 6*1 + 8*2 + 4*i2)
The substitution tfc = 1 + zk leads to the enumeration series

^Mn.Zi 9 )(*) = l + z + 5z 2 + 9 z 3 + 2 1 z 4 + 25z5 + 34z 6


+ 2 5 / + 21* 8 + 9z 9 + 5z 10 + z 11 + z 12

Summarizing the number of assemblies from each class, we get the following table.

k-chords Mazzola (An) Costere (Cn) Forte (Vn) Estrada (<Sn)


1 Unison 1 1 1 1
2 Intervals 5 6 6 6
3 Trichords 9 19 12 12
4 Tetrachords 21 43 29 15
5 Pentachords 25 66 38 13
6 Hexachords 34 80 50 11
7 Heptachords 25 66 38 7
8 Octachords 21 43 29 5
9 Enneachords 9 19 12 3
10 Decachords 5 6 6 2
11 Endecachords 1 1 1 1
12 Dodecachords 1 1 1 1
Total 157 351 223 77

Except for the symmetric group, the problem is symmetric: the unison has a com-
plementary 11-note assembly, the intervals have complementary 10-note structures, a

38
Motives and Mosaics

structure of k notes has a complementary structure of (12 — k) notes and for each group
(cyclic or dihedral) the complementary classes have the same number of elements. The
symmetry envelops the hexachords which are symmetrical with respect to themselves,
which explains for now that Matthias Hauer's 44-trope structures do not appear, as
it needs complementation within the hexachords. The structure of the dodecachords is
unique and is equivalent to the chromatic scale whose sounds can be used simultaneously.
These 12-sound chords appear in certain pieces, specially at the conclusion, such as in
Darius Milhaud's Chorfyhores, composed in 1915, or Ivan Wyschnegradsky's Journee de
['Existence for narrator, chorus and orchestra, composed in 1916. When the piece uses
serial techniques, a sub-classification of 12-sound series is necessary. It is possible to clas-
sify the series into 554 serial schemes by structuring the assemblies by tritones within
the series. The unison or single sound has a eleven-note complementary structure, as we
will see in the chapter on knots and braids.

2.4 Motives and Mosaics


The results of this section is due to H. Pripertinger [Pripertinger, 1999]. The notion of
motives has been established to describe simultaneously several musical parameters such
as rhythms and pitches. All parameters are coded by numbers and what we have to
consider is a couple of two parameters, one evolving in Z m and the other in Z n . Thus a
fc-motives is a fc-subset of the Cartesian product X = Zm x Zn. The affine group J?n of
all affine mappings from Z^ to Z^ is indicated by

Al = {vA,b\ AeGL2(Zn),beZ2n}
where GL,2(Zn) is the general linear group. The mapping o~A,b is defined by

Applying P61ya's enumeration theorem, the computation of cycle indices of the action of
the group J?n on Z\ leads to formulas such as for n = 2

P(Alz>) = 24 (*i + 6tft 2 + 8M3 + 3*| + 6t 4 )

and for n = 12, the enumeration series is given by

P{A22)Z22)(z) = 1 + z + 5z2 + 26z3 + 216*4 + 2024*5 + 27806*6 + 417209*7 + ....


that is to say, there are one 1-motive, five 2-motives, twenty six 3-motives, etc. The
Pripertinger cycle index formula shows that the number of motives is growing quickly.
There are more than 9 • 107 motives of 9 elements, more than 5 • 10 16 motives of 18
elements and more than 1036 motives of 69 elements.
H. Pripertinger has introduced the concept of G-mosaics as G-orbits of partitions. A
partition IT of a set X is a collection of non-empty subsets of X such that for each x in
X there is exactly one subset P G n containing x. The set of all partitions of X = Zn is
denoted by II n and the set of all partitions of size m is denoted by n n > m . The action of
a permutation group G on X = Zn induces the action on the set of all partitions:

G x nn - • nn, (<7,TT) H - • gn = {gP \ P € TT}

39
2. Taxonomy of Chords and Scales

The G-mosaics are the G-orbits on the set of all partitions II n and the GA:-mosaics or
G-mosaics of size k are the G-orbits on Hnik-

Theorem 66 Let Sm be the symmetric group on the set {0,1, 2, ..., m-1} and let Mk
be the number

1
' ' *W)€GxS fc i=l
where ji(g) and Ji(cr) are the numbers of i-cycles in the decomposition of g € G and
a e Sk The numbers of G-mosaics in Zn is Mn and the number of Gk-mosaics is
Mk - Mk-\ (with M0 = 0).

Example. For n = 12, the previous formula gives the following results [Fripertinger, 1998].

k A12 Ci2 v12


1 1 1 1
2 87 179 121
3 2155 7254 3838
4 13730 51075 26148
5 30121 115100 58400
6 28867 110462 56079
7 13835 52376 26696
8 3667 13299 6907
9 571 1873 1014
10 63 147 96
11 5 6 6
12 1 1 1
Total 93103 351773 179307

2.5 Dihedral Classification


The dihedral classification used by Forte's theory classifies in the same structure the sets
of notes which are different according to the following criteria: (1) by enumeration - for
example the major chord (C, E, G) is equivalent to the permutation (E, C, G), (2) by
transposition - for example, the major chord (C, E, G) has the same structure as the
chord (E, Gj|, B) which is the C major chord transposed a major third higher, and finally
(3) by inversion - the major chord (C, E, G) is equivalent to the minor chord (C, Eb, G).
Based on the major chord (C, E, G) the reverse chord is constructed by reversing each
interval (C, GJJ, F) which, reordered and transposed to the lower fourth, is equivalent
to the minor chord. Mathematically, Forte identifies the twelve sounds of the temperate
scale to the twelve first numbers such as C = 0, Cfl = 1, D = 2, etc., and determines
the normal order of a pcset by reducing the gap between the first and the last note of
the set in ascending order. Certain structures having the same ambitus, the set starts
with the integer 0 and such a structure is called a prime form. Two pcsets will be
equivalent if they can be reduced to the same primitive form through transposition or
through inversion followed by transposition. From a mathematical point of view as we
have seen in the previous chapter, the transposition is a sum over the integers modulo
12. Transposing, for example, the reduced seventh chord [0, 3, 6, 9] (C, Eb, F|j, A) to the

40
Cyclic Chords

higher fifth is equivalent to adding 7 (a fifth is worth 7 half-tones) to each number [7,
10, 13, 16] and to keep only the balance of the division by 12 [7, 10, 1, 4] i.e. [1, 4, 7,
10] (Cfl, E, G, Bb), which is a structure similar to the initial chord. Inversion, which is
one of the essential functions of the serial theory, and even of the fugues of J.-S. Bach,
cannot be integrated esily to the classification. To consider that a major and a minor
chord have the same structure is impossible in the analysis of a tonal piece and even in
the case of modal or atonal works, the distinction between major and minor structures
should be made in most cases [Eimert, 1924], [Perle, 1962]. In Alban Berg's Sonata for
piano, the first phrase finishes on the fourth beat by a B minor chord which colors the
whole analysis. The first of the Vingt regards sur Venfant Jesus by Olivier Messiaen,
which is a work written in Mode 2, is colored by the tonality of F sharp major ; limited
transposition sets are, according to Messiaen, "in the atmosphere of several tonalities at
once, without poly tonality" [Messiaen, 1954, p. 57]. Another problem linked to inversion
is the complementation function which is in the dihedral classification slightly modified
as compared to its traditional form with respect to conventional sets. For example, for
sets, the structure [0, 1, 2, 3, 4, 6 ] is complemented by the notes which are not in this
set, i.e., [5, 7, 8, 9, 10, 11] which, transposed to the lower sixth minor, is a new structure
of the initial form [0, 2, 3, 4, 5, 6]. On the other hand, if the inversion is integrated, the
initial set becomes the set [11, 9, 8, 7, 6, 5] which, according to dihedral classification,
is equivalent to itself, and therefore is its own complement, which is no longer patterned
after sets.

2.6 Cyclic Chords


Chains of chords have been widely used in the past. Jean-Philippe Rameau (Genera-
tion harmonique, 1737), Camille Durutte (Technie ou lois generates du syst&me harmo-
niquef 1855) have used chains of thirds. Chains of fourths have been studied by Ivan
Wyschnegradsky (Etude sur Vharmonie par quartes superpos&es, 1935 and Preface a un
Traits d'harmonie par quartes superposies, 1944) and by Guerino Mazzola [Mazzola,
1990, p. 125]. The following definition generalizes the notion of chains of chords.
Definition 67 A (p,q)-chain is a sequence (ai,a2,...,afc) of pitch classes such that the
difference
CLj+i — dj = p mod p + q
k is the length of the chain. An extended (p, q)-chain is a subsequence (aj) such that for
all j , one of two consecutive pitch classes (a^a^\) belongs to the sequence
Example. The (l,2)-chain is the chain of seconds. The (3,4) chain also called the 3-chain
is the chain of thirds. The (5,7)-chain, also called the 5-chain is the chain of fourths or
fifths. The (6,6)-chain is the tritonus chain.
Theorem 68 Under the group action of the affine group An, there are exactly k set
classes of third chains of length k.
Example. The third chains are the chords of intervals 3 or 4. There are only two chains of
length 2, namely (0, 3) and (0, 4). Three chains of length 3: the chain (3, 3) corresponds
to the chord (0, 3, 6) (class 15 in Mazzola's classification), the chain (3, 4) and the
chain (4, 3) correspond to the same class (0, 3, 7) or (0, 4, 7) (class 10.1 in Mazzola's
classification) and the chain (4, 4) is associated with the class 16 of (0, 4, 8).

41
2. Taxonomy of Chords and Scales

Definition 69 A cyclic chord is a pitch class set for which the free monoid generated
by the affine mapping r
< r > (x) = {x, r(x), r 2 (#),...}
is finite. The cycle index v is the period of the cycle r p + v = rp.
Definition 70 The transition matrix Taib of a cyclic chord is the matrix defined by the
generator r = Ma$ as follows

(Taib)ij = 1 if Mafb(i) = j (mod 12)

and 0 otherwise. The characteristic of the chord is the characteristic polynomial of the
transition matrix
P(x) = det(Ta>6 - xl)
Definition 71 Let fin be the cyclic group of prime roots £ such that £ n = 1. The cyclo-
tomic polynomial is defined by

<i>n(x)= n ^ - o
<e/i n

where the product is done over all generators of the cyclic group \xn.
Proposition 72 (1) The coefficients of the cyclotomic polynomials are integers
4>n{x) e Z[x]
(2) For all integer n > 1
xn -1=n^M
d\n

(3) For all integer n > 1


K M = n v - I)M(WA0
d\n
(4) Ifp is prime
p-i

M ) = Y<xk
x
k=0
where [i is the Moebius function defined by fi(l) = 1, /x(n) = 0ifnisa square factor and
fj,(n) = (—l) r if n is a product of r distinct prime numbers.
Theorem 73 A class set is a cyclic chord for the action of the affine group if and only
if its characteristic polynomial is different from </>|2, </>i</>2, </>?</>2, 0o^i> 01 $$^2 ^3 where
(j)j are the cyclotomic polynomials.

Recall that the first cyclotomic polynomials are </>Q(X) = a;, (/>i(x) = x—1, <p2(x) = x+1,
(j)s(x) = # 2 + x + l, 04(x) = z 2 + l, </>6(z) = x2-x + l, (j>i2(x) = x4-x2 + l. For different
values of a and 6, the characteristic polynomials are
Po,b(x) = P6ib(x) = xu(x - 1) = 0 l V i ( a ) for all b = 0,.., 11.
Pifi{x) = {x-l)™ = <ff{x)
P\fik+l{x) = P\$k+s{x) = X12 - 1 = <f>i<l>2<l>3<l>4</>e<f>i2(x)
4 2
PIA(X) = Plt8(x) = (x- l) (x + x + l ) 4 = </>i4(x)

42
Enumerations of Pcsets Classes

P2>b(x) = P8,b(x) = x9(x + l)(x - l ) 2 = <p90<j>l<f>2(x) for all 6 = 0, ...11.


P3,b{x) = 4>l<t>\~*<IJ2+1{x) with j = 0 if b even and j = liib odd.
P*MX) = PutM*) = x9(x ~ !) 3 = <fil<f>\{*)
P4,3k+i(x) = Pi,Zk+2{x) = Pio,3k+i(x) = Pio,3k+2(x) = (plfafoix)
P5,2k+l(x) = Ph3(x) = Ph9(X) = (X- 1)3(X + 1)3(X2 + l) 3 = <j>\<t>l4{x)
PsAk{x) = (x + \)\x - If = <t>\<t>\{x)
P5,2k+2(X) = Plfi(x) = tf4(X)
P7,6fe(x) = (a; + l ) 3 ( a ; - l ) 9 = 0?^(x)
P7,6k+l(x) = P7fik+b{x) = Pl,2(xl = Pl,10(x) = <^4>2<Pl4>l(x)
Plfik+2{x) = PTfik+i{x) = 01</>20306(X)
P9,4k(x) = x*(x ~ I ) 4 = <t>Wl(x)
P9,4fc+l(a;) = P9,4fc+3(z) = X8(X - 1)(X + 1)(X2 + 1) = t l ^ ^ x )
P9Ak+2(x) = P3,1(X) = X\X - \)2{X + l) 2 = tlfi&ix)
PnMx) = (x~ i) 7 (* +1) 5 = 4>14(X)
Pn,2k+i(x) = (x- l)6(x +1)6 = <t>\4>l{x)
In the following table, the first column gives the cyclic chords. The second column is
the generated monoid and the last column gives the cycle index.

Chords Generator Cycle | Chords Generator Cycle


{0,1,2 11} <Ti> (0) 12 | {0,3,9} < M 6 i 3 > (0) 1
{0,2,4,6,8,10} <Mi ) 2 > (0) 6 {0,1,7} <M 6 ,i> (0) 1
{0,1,8,9,4,5} <M 7 l i> (0) 6 {0,1} <Mn,i> (0) 2
{9,10,8,0,4} <M 1 0 , 4 > (0) 3 {0,2} <M 11>2 > (0) 2
{1,0,3,6,9} <M 9>3 > (1) 4 | {0,3} < M i i , 3 > (0) 2
{0,16,7} <M 5 ,i> (0) 4 | {0,4} < M n > 4 > (0) 2
{0,3,6,9} <T 3 > (0) 4 | {0,6} <M 11>6 > (0) 2
{0,2,10,6} <Af 4l2 > (0) 3 {0,1} <M 0 ,i> (0)
{0,4,8} <T 4 > (0) 3 {0,2} <M 0 ,2> (0)
{11,0,4,8} <M 4 ) 4 > (11) 3 {0,3} <M 0 ,3> (0)
{0,1,3,7} < M 2 i l > (0) 2 {0,4} <Mo, 4 > (0)
{0,2,8} <M 3 l 2 > (0) 2 {0,6} <M 0 i 6 > (0)
{0,1,4} <M 3 ,i> (0) 2 | 1 {0} <M 0 ,o> (0)

2.7 Enumerations of Pcsets Classes


Intervals and Decachords. The most compact interval is the half-tone or second
minor whose reversal is the seventh minor and whose complement a cluster of ten notes.
The intervals have twelve transpositions, except for the triton, which has only six. The
decaphonic scale

is the basic scale for Scriabin's Prelude opus 74 no.l. The final cadence is written in this
scale.

43
2. Taxonomy of Chords and Scales

Paul Hindemith [Hindemith, 1940] classified the intervals by order of harmonic affinity.
He obtained a series (C, G, F, A, E, E\>, A\>, D, Bb, D\>, B, Pji) which he used in the
piano piece Ludus Tonalis as a structural element. The series is even printed in spiral
form on the cover page. The intervals lead to a complete redefinition of tonalities. In
the following table, the first column is the representative of the sets of the dihedral or
cyclic classification, the second column is the number of the class set, the third gives the
prime form of this pcset, the fourth column is the interval vector and the last give the
complement of the set.

| Representative B Sets Iv Complement sets


**0000000000 C2-1 0,1 11-1 0,1,2,3,4,5,6,7,8,9
*0*000000000 C2-2 0,2 10-2 0,1,2,3,4,5,6,7,8,10
*00*00000000 C2-3 0,3 93 0,1,2,3,4,5,6,7,9,10
*000*0000000 C2-4 0,4 84 0,1,2,3,4,5,6,8,9,10
*0000*000000 C2-5 0,5 75 0,1,2,3,4,5,7,8,9,10
*00000*00000 C2-6 0,6 66 0,1,2,3,4,6,7,8,9,10

In the affine classification, there are also six classes, but two classes have the same
symmetry group, reducing the number to only five classes. In the following table, the
first column is the representative of the set, the second column gives the number of the
class set under the action of the cyclic group, the third gives the class names according
to Mazzola's classification table, the fourth column provides a representative and the last
column is the symmetry group of the class.

| Representative tt Maz. Sets Sym. Groups


**0000000000 C2-1 3 0,1 < Mn,ii >
*0000*000000 C2-5 3.1 0, 5 < Mil,ii >
*0*000000000 C2-2 4 0,2 {1,T7,MIO,II,MIO,II}
*00*00000000 C2-3 5 0,3 {l,T5,M9,7,M9>ii}
*000*0000000 C2-4 6 0,4 {1,T7,M8J5,M8)II}
*00000*00000 C2-6 7 0,6 Hi |

The last symmetry group is defined by

ifa = {l9T3,Ts,T7iT9tTluMa,6} with a = 0,1,3,5,7,9,11.

In the symmetric classification, there are six classes of 2-chords which are the same as
in the dihedral classification. But there are only two classes of 10-chords represented by
the following representatives.

Representatives
**********QQ
B Sets
S10-1 0,1,2,3,4,5,6,7,8,9
*********Q*Q S10-2 0,1,2,3,4,5,6,7,8,10

Trichords and E n n e a c h o r d s . By forming all the possible combinations of superim-


posing major thirds and minor thirds, the four classical chords are obtained: the perfect
major chord [0,4,7], the perfect minor chord [0,3,7], the diminished fifth chord [0,3,6] and
the augmented fifth chord [0,4,8], which is not used for tonal music: Scriabin's Sonata

44
Enumerations of Pcsets Classes

no. 10, opus 70 opens with an augmented fifth chord (Gb, Bb, D) and a diminished fifth
chord (E\>, G\>, B\>\>). The inflection of the fifth of classical chords results in inflected
chords, the most famous of which is the major chord with descending inflection of the
fifth (C, E, G\>). Scriabin used the transpostion of the enneaphonic scale

in the first measures of the fourth Prelude opus 74-

| Representative tl Sets Iv Forte Complements


| ***000000000 C3-1 0,1,2 1011 3-1/9-1 0,1,2,3,4,5,6,7,8
| **o*oooooooo C3-2 0,1,3 912 3-2/- 0,2,3,4,5,6,7,8,9
1 *o**oooooooo C3-3 0,2,3 921 -/9-2 0,1,2,3,4,5,6,7,9
**00*0000000 C3-4 0,1,4 813 3-3/- 0,1,3,4,5,6,7,8,9
1 *o*o*ooooooo C3-5 0,2,4 822 3-6/9-6 0,1,2,3,4,5,6,8,10
*00**0000000 C3-6 0,3,4 831 -/9-3 0,1,2,3,4,5,6,8,9
| **000*000000 C3-7 0,1,5 714 3-4/- 0,1,2,4,5,6,7,8,9
1 *o*oo*oooooo C3-8 0,2,5 723 3-7/- 0,1,2,3,4,5,7,9,10
*00*0*000000 C3-9 0,3,5 732 -/9-7 0,1,2,3,4,5,7,8,10
*000**000000 C3-10 0,4,5 741 -/9-4 0,1,2,3,4,5,7,8,9
**0000*00000 C3-11 0,1,6 615 3-5/- 0,1,2,3,6,7,8,9,10
*0*000*00000 C3-12 0,2,6 624 3-8/- 0,1,2,3,4,6,8,9,10
*00*00*00000 C3-13 0,3,6 633 3-10/9-10 0,1,2,3,4,6,7,9,10
*000*0*00000 C3-14 0,4,6 642 -/9-8 0,1,2,3,4,6,7,8,10
*0000**00000 C3-15 0,5,6 651 -/9-5 0,1,2,3,4,6,7,8,9
*0*0000*0000 C3-16 0,2,7 525 3-9/9-9 0,1,2,3,5,6,7,8,10
*00*000*0000 C3-17 0,3,7 534 3-11/- 0,1,2,3,5,6,8,9,10
*000*00*0000 C3-18 0,4,7 543 -/9-11 0,1,2,3,5,6,7,9,10
*000*000*000 C3-19 0,4,8 444 3-12/9-12 0,1,2,4,5,6,8,9,10 |

In the affine, classification, there are only 9 classes (12 classes if we include Mazzola's
extension ".1" see [Mazzola, 2002]). In the following table, the first column is a repres-
entative of the set. The second column is the number in the cyclic classification. The
prime form is given in the third column. The interval vector is in the fourth column.
The columns 5 and 6 give respectively the name of the chord in Forte's and in Mazzola's
classification. The last column is the symmetry group of the set.

| Representative tt Sets Iv Forte Maz. Sym. Groups


| ***000000000 C3-1 0,1,2 1011 3-1/9-1 8 < Mio,n >
| *0*0000*0000 C3-16 0,2,7 525 3-9/9-9 8.1 < Mio.ii >
| **0*00000000 C3-2 0,1,3 912 3-2/- 9 {1}
*0*00*000000 C3-8 0,2,5 723 3-7/- 9.1 {1}
**00*0000000 C3-4 0,1,4 813 3-3/- 10 {1}
*00*000*0000 C3-17 0,3,7 534 3-11/- 10.1 {1}
**000*000000 C3-7 0,1,5 714 3-4/- 11 <T5>
**0000*00000 C3-11 0,1,6 615 3-5/- 12 < M6j7 >
*0*0*0000000 C3-5 0,2,4 822 3-6/9-6 13 {l,T7,M8,ii,M8,5}
*0*000*00000 C3-12 0,2,6 624 3-8/- 14 <T7>
*00*00*00000 C3-13 0,3,6 633 3-10/9-10 15 {l,T5,M6,ii,M6j7}
*000*000*000 C3-19 0,4,8 444 3-12/9-12 16 H

45
2. Taxonomy of Chords and Scales

where H is the group

iJ={l,Mij0,MM,Mi>8 with z ^ 3 , 6 , 9 }

Harmony achieved through the superimposition of fourths appealed to many composers


of the twentieth century who saw in it an alternative to Rameau's theories: Jolivet, in
his pieces for the piano Mana (particularly in the first Beaujolais), Stravinsky in The
Rite of Spring, etc. By superimposing perfect and augmented fourths, three new chords
are obtained ([0,5,6], [0,1,6] and [0,2,7]). The [0,5,6] structure is not individualized in
dihedral classification. Lastly, the complementary structure of a set of three notes is a
set of nine notes. These sets have particularly been used in the form of scales. Alexander
Tcherepnin, for example, developed polyphonic music called "intrapunctus" and tried to
find a metatonal scale which would include major and minor chords. He found a scale of
nine sounds (C, D\>, E\>, E, F, G, A\>, A, B) which he used in his opus 10 Feuilles libres.

In the symmetric classification, there are 12 classes of trichords and 3 classes of ennea-
chords. The last column of the following table gives the 3-partitions of twelve.

| Representatives tt Sets Partitions


***000000000 S3-1 0,1,2 1,1,10
**0*00000000 S3-2 0,1,3 1,2,9
**00*0000000 S3-3 0,1,4 1,3,8
**000*000000 S3-4 0,1,5 1,4,7
**0000*00000 S3-5 0,1,6 1,5,6
*0*0*0000000 S3-6 0,2,4 2,2,8
*0*00*000000 S3-7 0,2,5 2,3,7
*0*000*00000 S3-8 0,2,6 2,4,6
*0*0000*0000 S3-9 0,2,7 2,5,5
*00*00*00000 S3-10 0,3,6 3,3,6
*00*000*0000 S3-11 0,3,7 3,4,5
*000*000*000 S3-12 0,4,8 4,4,4 |

The three classes of enneachords are given in the following table.

Representatives
****#****QQQ
t Sets Partitions
S9-1 0,1,2,3,4,5,6,7,8 1,1,1,1,1,1,1,1,4
********Q*QQ S9-2 0,1,2,3,4,5,6,7,9 1,1,1,1,1,1,1,2,3
*******Q*Q*Q S9-3 0,1,2,3,4,5,6,8,10 1,1,1,1,1,1,2,2,2 |

Tetrachords and Octachords. In the classical theory, chords of four sounds are
reduced to the seventh chords. In the harmony by superimposition of fourths, the chords
found in Wagner's Tristan, Liszt's Mephistovalses, and in works by Milhaud, Honegger
and Stravinsky, are important structural chords. By superimposing fourths, five chords
of main fourths are obtained ([0, 2, 5, 7], [0, 2, 6, 7], [0, 1, 5, 6], [0, 1, 5, 7] and
[0, 1, 6, 7]) which create harmony by steps of fourths. The structures of four sounds
are also very present in modal compositions. The chords theme in Olivier Messiaen's
Vingts regards sur Venfant Jesus is made up of three chords of four sounds (of structure
[0, 2, 5, 7], [0, 1, 5, 6], [0, 3, 4, 9], [0, 4, 5, 6]). The atonal pieces also use chords of four
sounds. Each of Aaron Copland's Piano Variations, written in 1930 explores a motif of

46
Enumerations of Pcsets Classes

four notes. Tetrachords are widely used in Serge Prokofiev's Visions fugitives, opus 22.
In his Second sonata, Nikolai' Roslavetz uses a scale of eight notes (E, F, F$, G |J, A$} B,
Q,D)

In the cyclic and dihedral classification, the following table shows the representative,
the number of the set in the cyclic classification, the prime form of the set (third column),
the interval vector (fourth column), the name of the set in Forte's classification and the
prime form of the complement set.
| Representative Sets Iv Forte | Complements
| ****00000000
1*C4-1 0,1,2,3 9111 4-1/8-1 0,1,2,3,4,5,6,7
1 ***o*ooooooo C4-2 0,1,2,4 8112 4-2/- 0,2,3,4,5,6,7,8
| **o**ooooooo C4-3 0,1,3,4 8121 4-3/8-3 0,1,2,3,4,5,6,9
| *0***0000000 C4-4 0,2,3,4 8211 78-2 0,1,2,3,4,5,6,8
1 ***oo*oooooo C4-5 0,1,2,5 7113 4-4/- 0,1,3,4,5,6,7,8
| **0*0*000000 C4-6 0,1,3,5 7122 4-11/- 0,2,3,4,5,6,7,8
1 **oo**oooooo C4-7 0,1,4,5 7131 4-7/8-7 0,1,2,3,4,5,8,9
*0**0*000000 C4-8 0,2,3,5 7212 4-10/8-10 0,2,3,4,5,6,7,9
1 *o*o**oooooo C4-9 0,2,4,5 7221 -/8-11 0,1,2,3,4,5,7,9
*00***000000 C4-10 0,3,4,5 7311 -/8-4 0,1,2,3,4,5,7,8
***000*00000 C4-11 0,1,2,6 6114 4-5/- 0,1,2,4,5,6,7,8
| **o*oo*ooooo C4-12 0,1,3,6 6123 4-13/- 0,2,3,5,6,7,8,9
1 **oo*o*ooooo C4-13 0,1,4,6 6132 4Z15/- 0,1,3,4,5,6,7,8
| **000**00000 C4-14 0,1,5,6 6141 4-8/8-8 0,1,2,3,4,7,8,9
1 *o**oo*ooooo C4-15 0,2,3,6 6213 4-12/8-12 0,1,3,4,5,6,7,9
1 *o*o*o*ooooo C4-16 0,2,4,6 6222 4-21/8-21 0,1,2,3,4,6,8,10
| *0*00**00000 C4-17 0,2,5,6 6231 -/8Z15 0,1,2,3,4,6,8,9
1 *oo**o*ooooo C4-18 0,3,4,6 6312 V- 0,2,3,4,5,6,8,9
1 *oo*o**ooooo C4-19 0,3,5,6 6321 -/8-13 0,1,2,3,4,6,7,9
| *000***00000 C4-20 0,4,5,6 6411 -/8-5 0,1,2,3,4,6,7,8
1 ***oooo*oooo C4-21 0,1,2,7 5115 4-6/8-6 0,1,2,3,5,6,7,8
1 **o*ooo*oooo C4-22 0,1,3,7 5124 4Z29/- 0,2,3,4,6,7,8,9
1 **oo*oo*oooo C4-23 0,1,4,7 5133 4-18/- 0,1,3,4,6,7,8,9
1 **ooo*o*oooo C4-24 0,1,5,7 5142 4-16/- 0,1,2,4,6,7,8,9
| **0000**0000 C4-25 0,1,6,7 5151 4-9/8-9 0,1,2,3,6,7,8,9
| *0**000*0000 C4-26 0,2,3,7 5214 4-14/8-14 0,1,2,4,5,6,7,9
| *o*o*oo*oooo C4-27 0,2,4,7 5223 4-22/- 0,1,2,3,5,7,9,10
| *o*oo*o*oooo C4-28 0,2,5,7 5232 4-23/8-23 0,1,2,3,5,7,8,10
| *0*000**0000 C4-29 0,2,6,7 5241 -/8-16 0,1,2,3,5,7,8,9
| *00**00*0000 C4-30 0,3,4,7 5313 4-17/8-17 0,1,3,4,5,6,8,9
1 *oo*o*o*oooo C4-31 0,3,5,7 5322 -/8-22 0,1,2,3,5,6,8,10
| *00*00**0000 C4-32 0,3,6,7 5331 -/8-18 0,1,2,3,5,6,8,9
*000**0*0000 C4-33 0,4,5,7 5412 7- 0,2,3,4,5,7,8,9
1 *ooo*o**oooo C4-34 0,4,6,7 5421 -/8Z29 0,1,2,3,5,6,7,9
*000*000*000 C4-35 0,1,4,8 4134 4-19/- 0,1,2,4,5,6,8,9,10
| **000*00*000 C4-36 0,1,5,8 4143 4-20/8-20 0,1,2,4,5,7,8,9
1 *o*o*ooo*ooo C4-37 0,2,4,8 4224 4-24/8-24 0,1,2,4,5,6,8,10
*o*oo*oo*ooo C4-38 0,2,5,8 4233 4-27/- 0,1,3,4,6,7,8,10
*0*000*0*000 C4-39 0,2,6,8 4242 4-25/8-25 0,1,2,4,6,7,8,10
*00**000*000 C4-40 0,3,4,8 4314 -/8-19 0,1,2,4,5,6,8,9
*00*0*00*000 C4-41 0,3,5,8 4323 4-26/8-26 0,1,2,4,5,7,9,10
*00*00*0*000 C4-42 0,3,6,8 4332 -/8-27 0,1,2,4,5,7,8,10
*00*00*00*00 C4-43 0,3,6,9 3333 4-28/8-28 0,1,3,4,6,7,9,10 |

47
2. Taxonomy of Chords and Scales

In the affine classification, there are only 21 classes (29 extended classes) of tetrachords
and octachords, obtained by complementation. Remark that there are also ten classes
that have a symmetry group reduced to {1}, that is ten class sets with no symmetry. Some
authors have seen in this fact a foundation or a contribution to the atonality problem.
The following table shows the representatives of the class, the number of the set in the
cyclic classification (second column), Mazzola's number in the affine classification. The
last column gives the symmetry groups of the set.

| Representative It Maz. Sym. Groups


****00000000 C4-1 17 < M 9 ,u >
1 *o*oo*o*oooo C4-28 17.1 < M 9 ,n >
| ***0*0000000 C4-2 18 {1}
1 *o*o*oo*oooo C4-27 18.1 {1}
***00*000000 C4-5 19 {1}
| **0*0000*000 C4-33 19.1 {1}
| ***000*00000 C4-11 20 {1}
| **ooo*o*oooo C4-24 20.1 {1}
1 ***oooo*oooo C4-21 21 {l,T7,Mi0>ii,Mio,5}
| **o**ooooooo C4-3 22 < M 8 ,n >
*0*00*000*00 C4-41 22.1 < M8,ii >
1 **o*o*oooooo C4-6 23 <T5 >
**0*00*00000 C4-12 24 < M6|7 >
| **0*000*0000 C4-22 25 {1}
1 **oo*o*ooooo C4-13 25.1 {1}
| **0*00000*00 C4-18 26 {1}
| *0*00*00*000 C4-38 26.1 {1}
1 **o*oooooo*o C4-8 27 {ljMg.r.Mio.s.Mii.n} 1
**00**000000 C4-7 28 < M7,u >
**000*00*000 C4-36 28.1 < M7,n >
**00*00*0000 C4-23 29 <T7>
**00*000*000 C4-40 30 < M4,5 >
**oo*oooo*oo C4-30 31 {1,MII,II,M8J5,M3>7}
**000**00000 C4-14 32 {l,T5,M6|ii,M6|7}
**oooo**oooo C4-25 33 Hl
*0*0*0*00000 C4-16 34 {l,T7,M6lii,M6,5}
*0*0*000*000 C4-37 35 {l,r7,M4|ii,M6,5}
*0*000*0*000 C4-39 36 H2
*00*00*00*00 C4-43 37 H3 |

The groups Hi, H2 and # 3 are defined by

# 1 = {l,T7,Mi 1 ,ii,M 1 i,5,M6,i,M6,7,M5,5,M 5 ,ii}

iJ2 = {l,T7,Mio,ll,Mio,5,M6,i,M 6j 7,M4,5,M4,ii}

H<s = { ^ M ^ M ^ M ^ M ^ , with i odd}

In the symmetric classification, there are only fifteen classes of tetrachords given in
the following table. The last column is the 4-partition of twelve.

48
Enumerations of Pcsets Classes

| Representatives tt Sets Partitions


| ****00000000 S4-1 0,1,2,3 1,1,1,9
***0*0000000 S4-2 0,1,2,4 1,1,2,8
***00*000000 S4-3 0,1,2,5 1,1,3,7
***000*00000 S4-4 0,1,2,6 1,1,4,6
***oooo*oooo S4-5 0,1,2,7 1,1,5,5
**o*o*oooooo S4-6 0,1,3,5 1,2,2,7
**o*oo*ooooo S4-7 0,1,3,6 1,2,3,6
**0*000*0000 S4-8 0,1,3,7 1,2,4,5
**oo*oo*oooo S4-9 0,1,4,7 1,3,3,5
**00*000*000 S4-10 0,1,4,8 1,3,4,4
*0*0*0*00000 S4-11 0,2,4,6 2,2,2,6
*o*o*oo*oooo S4-12 0,2,4,7 2,2,3,5
*0*0*000*000 S4-13 0,2,4,8 2,2,4,4
*0*00*00*000 S4-14 0,2,5,8 2,3,3,4
*00*00*00*00 S4-15 0,3,6,9 3,3,3,3 |

In the symmetric classification, there are five classes of octachords.

Representatives
*******QQQQQ
tJ Sets Partitions
S8-1 0,1,2,3,4,5,6,7 1,1,1,1,1,1,1,5
******Q*QQQQ S8-2 0,1,2,3,4,5,6,8 1,1,1,1,1,1,2,4
******QQ*QQQ S8-3 0,1,2,3,4,5,6,9 1,1,1,1,1,1,3,3
*****0*o*ooo S8-4 0,1,2,3,4,5,7,9 1,1,1,1,1,2,2,3
*****Q*Q*Q*Q S8-5 0,1,2,3,4,6,8,10 1,1,1,1,2,2,2,2 |

Pentachords and Heptachords. The first atonal piece of the 20th century is a
work from the composer Abel Decaux, who settled in the United States in 1923 where
he taught at the Eastman School of Music of Rochester. Called the French Schoenberg,
Decaux composed four pieces for piano called Clairs de lune between 1900 and 1907.
The first piece, Minuit passe, develops a cell of three notes in a mysterious atmosphere
and stops on a chord of five sounds (B\>, D, Gtf, B, F$). The second piece, La ruelle,
begins with a chord of structure [0, 2, 4, 6, 8] and ends with a chord of the first piece.
Franz Liszt's Nuage gris also ends on an atonal chord of five sounds (A, E\>, G, B, F$).
Elliott Carter's Concerto for orchestra is built on 38 different possibilities of five-sound
chords. The pentatonic modes and in particular the major pentatonic (C, D, E, F, A)
and the minor pentatonic (C, D, E\>, G, A\>) were exploited by A. Tcherepnin (Etudes
de piano sur la gamme pentatonique, opus 51, Cinq etudes de concert chinoises, opus 52,
Etudes techniques sur la gamme pentatonique, opus 53). As for structures of seven sounds,
they are found in transition works used as scales. For example, Bartok uses a scale from
Roman folklore (C, D, E, F, Fj{, A, B), Scriabine uses a heptaphonic scale made of a
synthetic chord completed by the fundamental fifth (D, D\>, E, JFJJ, G, A, B\>) and many
Hispanic compositions use the melodic Andalusian minor. This scale (A, B\>, CJf, D, E,
Fj{, G|J) gives Hispanic elements found in Albeniz, Debussy (Iberia, La puerta del vino),
Granados (Goyescas), Lalo (Symphonie espagnole), Ravel (Alborada del gracioso), etc.
Dutilleux's Metaboles on the chord (D, F$, D$, B\>, E). The low E played by the double
basses gives the tenor of the whole piece.

The 72 Carnatic modes of Indian music only offer 36 different structures, as differ-
ent modes may have the same structure. This is the case for the Dheragankarabhama

49
2. Taxonomy of Chords and Scales

which is only the scale in C major and which can be found in five transpositions: Hanu-
matodi (C Phrygian), Narabhairavi (C Hypodorian), Harikambogi (C Mixolydian) and
Matsyakaliani (C Lydian). The Indian modes have also been used in Western music.
For example, August Chapuis in his Suite for piano uses the Eastern scale (D, E, F,
GU, A, Bb, C)J) which is the Srimendra mode, also used by Albert Roussel in his opera
Padm&vati, in the transposed form Mayamalavagaula. But the systematic exploitation of
these modes is due to composer Jacques Charpentier (72 Etudes kamatiques) who wrote
a work for each mode between 1956 and 1961. The classification of five-sound structures
shows the classical scales within the cyclic classification: major scale, minor scale, as-
cending melodic minor scale and descending melodic minor scale but cannot distinguish
the tenors of the ecclesiastic modes which are structurally different only in the starting
note, whether it is tonic or final. Jolivet's Etude sur les modes antiques exploits the A,
F and E modes. In Charles Koechlin's Heures Persannes opus 65, pentachords are used
for example in the fourth piece, entitled Matin frais dans la haute vallee.

Representative B Sets Iv Forte Complements


*****ooooooo C5-1 0,1,2,3,4 81111 5-1/7-1 0,1,2,3,4,5,6
****o*oooooo C5-2 0,1,2,3,5 71112 5-2/- 0,2,3,4,5,6,7
***o**oooooo C5-3 0,1,2,4,5 71121 5-3/- 0,3,4,5,6,7,8
**o***oooooo C5-4 0,1,3,4,5 71211 -/7-3 0,1,2,3,4,5,8
*o****oooooo C5-5 0,2,3,4,5 72111 77-2 0,1,2,3,4,5,7
****oo*ooooo C5-6 0,1,2,3,6 61113 5-4/- 0,1,3,4,5,6,7
| ***o*0*00000 C5-7 0,1,2,4,6 61122 5-9/- 0,2,4,5,6,7,8
| ***oo**ooooo C5-8 0,1,2,5,6 61131 5-6/- 0,1,4,5,6,7,8
1 **o**o*ooooo C5-9 0,1,3,4,6 61212 5-10/- 0,2,3,4,5,6,9
| **o*0**00000 C5-10 0,1,3,5,6 61221 5Z12/7Z12 0,1,2,3,4,7,9
| **oo***ooooo C5-11 0,1,4,5,6 61311 -/7-6 0,1,2,3,4,7,8
1 *o***o*ooooo C5-12 0,2,3,4,6 62112 5-8/7-8 0,2,3,4,5,6,8
| *o**0**00000 C5-13 0,2,3,5,6 62121 -/7-10 0,1,2,3,4,6,9
| *o*0***00000 C5-14 0,2,4,5,6 62211 -/7-9 0,1,2,3,4,6,8
| *oo****ooooo C5-15 0,3,4,5,6 63111 77-4 0,1,2,3,4,6,7
1 ****ooo*oooo C5-16 0,1,2,3,7 51114 5-5/- 0,1,2,4,5,6,7
| ***o*oo*oooo C5-17 0,1,2,4,7 51123 5Z36/- 0,2,3,5,6,7,8
***00*0*0000 C5-18 0,1,2,5,7 51132 5-14/- 0,1,3,5,6,7,8
***000**0000 C5-19 0,1,2,6,7 51141 5-7/- 0,1,2,5,6,7,8
**o**oo*oooo C5-20 0,1,3,4,7 51213 5-16/- 0,1,3,4,5,6,9
**o*o*o*oooo C5-21 0,1,3,5,7 51222 5-24/- 0,2,4,6,7,8,9
**o*oo**oooo C5-22 0,1,3,6,7 51231 5-19/7-19 0,1,2,3,6,8,9
**00**0*0000 C5-23 0,1,4,5,7 51312 5Z18/- 0,2,3,4,5,8,9
**oo*o**oooo C5-24 0,1,4,6,7 51321 7- 0,1,2,3,6,7,9
**ooo***oooo C5-25 0,1,5,6,7 51411 -11-1 0,1,2,3,6,7,8
*0***00*0000 C5-26 0,2,3,4,7 52113 5-11/7-11 0,1,3,4,5,6,8
*0**0*0*0000 C5-27 0,2,3,5,7 52122 5-23/- 0,2,4,5,6,7,9
*o**oo**oooo C5-28 0,2,3,6,7 52131 -/7Z18 0,1,2,3,5,8,9
*o*o**o*oooo C5-29 0,2,4,5,7 52212 -/7-23 0,2,3,4,5,7,9
*0*0*0*0*000 C5-30 0,2,4,6,7 52221 -/7-24 0,1,2,3,5,7,9
*o*oo***oooo C5-31 0,2,5,6,7 52311 -/7-14 0,1,2,3,5,7,8
*00***0*0000 C5-32 0,3,4,5,7 53112 7- 0,2,3,4,5,7,8
*oo**o**oooo C5-33 0,3,4,6,7 53121 -/7-16 0,1,2,3,5,6,9
*oo*o***oooo C5-34 0,3,5,6,7 53211 -/7Z36 0,1,2,3,5,6,8
*ooo****oooo C5-35 0,4,5,6,7 54111 77-5 0,1,2,3,5,6,7 1

50
Enumerations of Pcsets Classes

| ***0*000*000 C5-36 1 0,1,2,4,8 1 41124 5-13/- 0,2,3,4,6,7,8


| ***00*00*000 C5-37 0,1,2,5,8 41133 5Z38/- 0,1,3,4,6,7,8
| ***000*0*000 C5-38 0,1,2,6,8 41141 5-15/7-15 0,1,2,4,6,7,8
1 **o**ooo*ooo C5-39 0,1,3,4,8 41214 5Z17/7Z17 0,1,2,4,5,6,9
1 **o*o*oo*ooo C5-40 0,1,3,5,8 41223 5-27/- 0,2,4,5,7,8,9
| **0*00*0*000 C5-41 0,1,3,6,8 41232 5-29/- 0,2,3,5,7,8,9
1 **o*ooo**ooo C5-42 0,1,3,7,8 41241 5-20/- 0,1,2,5,7,8,9
| **oo**oo*ooo C5-43 0,1,4,5,8 41313 5-21/7-21 0,1,4,5,7,8,9
| **00*0*0*000 C5-44 0,1,4,6,8 41322 5-30/- 0,1,3,5,7,8,9
| **00*00**000 C5-45 0,1,4,7,8 41331 5-22/7-22 0,1,2,5,6,7,9
1 **ooo*o**ooo C5-46 0,1,5,7,8 41421 -/7-20 0,1,2,4,7,8,9
| *0***000*000 C5-47 0,2,3,4,8 42114 -/7-13 0,1,2,4,5,6,8
| *0**0*00*000 C5-48 0,2,3,5,8 42123 5-25/- 0,2,3,5,6,7,9
1 *o**oo*o*ooo C5-49 0,2,3,6,8 42132 5-28/7-28 0,1,3,5,6,7,9
| *0*0**00*000 C5-50 0,2,4,5,8 42213 5-26/7-26 0,1,3,4,5,7,9
| *0*0*0*0*000 C5-51 0,2,4,6,8 42222 5-33/7-33 0,1,2,4,6,8,10
| *0*0*00**000 C5-52 0,2,4,7,8 42231 -/7-30 0,1,2,4,6,8,9
| *o*oo**o*ooo C5-53 0,2,5,6,8 42312 V- 0,2,3,4,6,8,9
| *o*oo*o**ooo C5-54 0,2,5,7,8 42321 -/7-29 0,1,2,4,6,7,9
| *00***00*000 C5-55 0,3,4,5,8 43113 5Z37/7Z37 0,1,3,4,5,7,8
| *00**0*0*000 C5-56 0,3,4,6,8 43122 V- 0,2,4,5,6,8,9
| *00**00**000 C5-57 0,3,4,7,8 43131 7- 0,1,4,5,6,8,9
| *oo*o**o*ooo C5-58 0,3,5,6,8 43212 -/7-25 0,2,3,4,6,7,9
1 *oo*o*o**ooo C5-59 0,3,5,7,8 43221 -/7-27 0,1,2,4,5,7,9
*oo*oo***ooo C5-60 0,3,6,7,8 43311 -/7Z38 0,1,2,4,5,7,8
**o*oo*oo*oo C5-61 0,1,3,6,9 31233 5-31/- 0,2,3,5,6,8,9
**00*0*00*00 C5-62 0,1,4,6,9 31313 5-32/- 0,1,3,5,6,8,9
**00*00*0*00 C5-63 0,1,4,7,9 31331 -/7-32 0,1,3,4,6,8,9
*0**00*00*00 C5-64 0,2,3,6,9 32133 -/7-31 0,1,3,4,6,7,9
*0*0*0*00*00 C5-65 0,2,4,6,9 32223 5-34/7-34 0,1,3,4,6,8,10
*0*0*00*0*00 | C5-66 0,2,4,7,9 32232 5-35/7-35 0,1,3,5,6,8,10 |

In the affine classification, there are 25 classes (36 extended classes) instead of the
38 classes of the dihedral classification. The following table shows the representative,
the number of the set in the cyclic classification, the prime form of the set, the interval
vector, the name in Mazolla's classification and the symmetry groups of the sets.

| Representative tJ Sets Iv Maz. Sym. Groups


| *****ooooooo C5-1 0,1,2,3,4 81111 38 < M8,ii >
| *0*0*00*0*00 C5-66 0,2,4,7,9 32232 38.1 < M8,ii >
****0*000000 C5-2 0,1,2,3,5 71112 39
**0*0*0000*0 C5-27 0,2,3,5,7 52122 39.1
****00*00000 C5-6 0,1,2,3,6 61113 40
**0*00*0*000 C5-41 0,1,3,6,8 41232 40.1
****ooo*oooo C5-16 0,1,2,3,7 51114 41
***00*0*0000 C5-18 0,1,2,5,7 51132 41.1
***0**000000 C5-3 0,1,2,4,5 71121 42
**0*0*00*000 C5-40 0,1,3,5,8 41223 42.1
***0*0*00000 C5-7 0,1,2,4,6 61122 43
**0*0*0*0000 C5-21 0,1,3,5,7 51222 43.1
***0*00*0000 C5-17 0,1,2,4,7 51123 44 <T7> \

51
2. Taxonomy of Chords and Scales

| ***0*000*000 C5-36 0,1,2,4,8 41124 45 {1}


1 ***o*oooo*oo C5-32 0,3,4,5,7 53112 46 < M8,5 >
| ***o*ooooo*o C5-12 0,2,3,4,6 62112 47 < Mio.ii >
| ***oo**ooooo C5-8 0,1,2,5,6 61131 48 {1}
| **o*ooo**ooo C5-42 0,1,3,7,8 41241 48.1 {1}
***oo*oo*ooo C5-37 0,1,2,5,8 41133 49 {1}
**0*0000**00 C5-23 0,1,4,5,7 51312 49.1 {1}
***oo*ooo*oo C5-55 0,3,4,5,8 43113 50 < Mio.ii >
**o**ooo*ooo C5-39 0,1,3,4,8 41214 50.1 < Mio,n >
***ooo**oooo C5-19 0,1,2,6,7 51141 51 <T7>
***000*0*000 C5-38 0,1,2,6,8 41141 52 {l,Mio,ll,M 6 > 7,M4,5}
**o**o*ooooo C5-9 0,1,3,4,6 61212 53 {1}
**o*oo*ooo*o C5-48 0,2,3,5,8 42123 53.1 {1}
**0**00*0000 C5-33 0,3,4,6,7 53121 54 {1}
**oo*o*oo*oo C5-62 0,1,4,6,9 31313 54.1 {1}
**0*0**00000 C5-10 0,1,3,5,6 61221 55 {l,T5,M6,ii,M6,7}
**00*0*0*000 C5-44 0,1,4,6,8 41322 56 <T7>
**o*oo**oooo C5-22 0,1,3,6,7 51231 57 < M6,7 >
**o*oo*oo*oo C5-61 0,1,3,6,9 31233 58 < M6>7 >
**o*ooo*o*oo C5-53 0,2,5,6,8 42312 59 <T7 >
**oo**oo*ooo C5-43 0,1,4,5,8 41313 60 <T5>
**oo*oo**ooo C5-45 0,1,4,7,8 41331 61 {l,T7,M4,ii,M4|5}
*o*o*o*o*ooo C5-51 0,2,4,6,8 42222 62 {l,T7,M4,ii,M4>7} |

In the symmetric classification, there are only 13 classes of pentachords given in the
following table.

|~ Representatives tt Sets Partitions


*****0000000 S5-1 0,1,2,3,4 1,1,1,1,8
****0*oooooo S5-2 0,1,2,3,5 1,1,1,2,7
****00*00000 S5-3 0,1,2,3,6 1,1,1,3,6
****ooo*oooo S5-4 0,1,2,3,7 1,1,1,4,5
***o*o*ooooo S5-5 0,1,2,4,6, 1,1,2,2,6
***0*oo*0000 S5-6 0,1,2,4,7 1,1,2,3,5
***0*000*000 S5-7 0,1,2,4,8 1,1,2,4,4
***00*00*000 S5-8 0,1,2,5,8 1,1,3,3,4
**0*0*0*0000 S5-9 0,1,3,5,7 1,2,2,2,5
**0*0*00*000 S5-10 0,1,3,5,8 1,2,2,3,4
**0*00*00*00 S5-11 0,1,3,6,9 1,2,3,3,3
*0*0*0*0*000 S5-12 0,2,4,6,8 2,2,2,2,4
*0*0*0*00*00 S5-13 0,2,4,6,9 2,2,2,3,3 |

and seven classes of heptachords.

Representatives tl Sets Partitions


*******00000 S7-1 0,1,2,3,4,5,6 1,1,1,1,1,1,6
******o*oooo S7-2 0,1,2,3,4,5,7 1,1,1,1,1,2,5
******oo*ooo S7-3 0,1,2,3,4,5,8 1,1,1,1,1,3,4
*****Q*Q*QQQ S7-4 0,1,2,3,4,6,8 1,1,1,1,2,2,4
*****Q*QQ*QQ S7-5 0,1,2,3,4,6,9 1,1,1,1,2,3,3
****0*o*o*00 S7-6 0,1,2,3,5,7,9 1,1,1,2,2,2,3
*** 0 *o*0*0*0 S7-7 0,1,2,4,6,8,10 1,1,2,2,2,2,2 |
Enumerations of Pcsets Classes

Hexachords and Hauer Tropes. In 20 th century music, hexachords can be found


in polytonal, modal and atonal pieces. The chord used by Prank Bridge in his Piano
Sonata written in 1924 uses the bitonal style: he superimposes a minor chord and a
major chord one tone higher (C, Eb, G, D, Ffr, A). The first of Maurice Ravel's Valses
nobles et sentimentales opens on a transformed chord of seventh (G, D, A, C, E$, B).
Charles Koechlin's Heures persannes which uses the modal style, even metamodal style,
uses several hexachords. Dutilleux's fourth Metaboles (Torpide) is built on a chord of six
notes (C, E, G, D, F$, A\>). The string quartet Ainsi la nuit is structured on a chord
that superimposes three fifths (Cft, G§, F, C, G, D). The central movement {lento) of
the Third Sonata by Michael Tippett varies four times a succession of 17 hexachords.
Elliott Carter's Sonate pour piano is built on the following chord {B, F$, C|J, G(j, D§, A§).
Scriabin's synthetic chord {C, F&, B\>, E, A, D) is the basis for Promethie. It can be noted
that the structure of six-note sets is unusual because the complement of an hexachord is
itself an hexachord. Certain complementary sets have the same structure. This is the case
for the cluster of six sounds. Due to this fact, there are only 44 different sets with 8 of
them being their own complementary, which makes a total of 2 x (44-8)+8=80 structures.
These 44 sets were established in 1926 by Josef Matthias Hauer [Hauer, 1926] who used
them in many Zwolftonspiele.

| Representative tl Sets Iv Forte Com pi.


| ******QOOOOO C6-1 0,1,2,3,4,5 711111 6-1/6-1 id.
| *****o*ooooo C6-2 0,1,2,3,4,6 611112 6-2/- 0,2,3,4,5,6
1 ****o**ooooo C6-3 0,1,2,3,5,6 611121 6Z3/- 0,3,4,5,6,7
| ***o***ooooo C6-4 0,1,2,4,5,6 611211 6Z4/6Z37 0,1,2,3,4,8
| **o****ooooo C6-5 0,1,3,4,5,6 612111 -/6Z36 0,1,2,3,4,7
| ****Q*O*0000 C6-6 0,1,2,3,5,7 511122 6-9/- 0,2,4,5,6,7
| ****oo**0000 C6-7 0,1,2,3,6,7 511131 6-5/- 0,1,4,5,6,7
1 ***o**o*oooo C6-8 0,1,2,4,5,7 511212 6Z11/- 0,3,5,6,7,8
| ***Q*O**0000 C6-9 0,1,2,4,6,7 511221 6Z12/- 0,2,5,6,7,8
| ***oo***0000 C6-10 0,1,2,5,6,7 511311 6Z6/6Z38 0,1,2,3,7,8
1 **o***o*oooo C6-11 0,1,3,4,5,7 512112 6Z10/Z39 0,2,3,4,5,8
| **o**0**0000 C6-12 0,1,3,4,6,7 512121 6Z13/Z42 0,1,2,3,6,9
| **o*0***0000 C6-13 0,1,3,5,6,7 512211 -/6-Z41 0,1,2,3,6,8
| **00****oooo C6-14 0,2,3,4,5,7 521112 6-8/6-8 id.
| *o***0**0000 C6-15 0,2,3,4,6,7 521121 V- 0,3,4,5,6,8
| *o**0***0000 C6-16 0,2,3,5,6,7 521211 -/6Z40 0,1,2,3,5,8
| ***o**00*000 C6-17 0,1,2,4,5,8 411213 6-15/- 0,3,4,6,7,8
| ***Q*0*0*000 C6-18 0,1,2,4,6,8 411222 6-22/- 0,2,4,6,7,8
1 ***o*oo**ooo C6-19 0,1,2,4,7,8 411231 6Z17/- 0,2,3,6,7,8
| ***oo**o*ooo C6-20 0,1,2,5,6,8 411311 6Z43/- 0,1,4,6,7,8
1 ***oo*o**ooo C6-21 0,1,2,5,7,8 411321 6-18/- 0,1,3,6,7,8
| ***ooo***ooo C6-22 0,1,2,6,7,8 411411 6-7/6-7 id.
| **0***00*000 C6-23 0,1,3,4,5,8 412113 6-14/6-14 id.
1 **o**o*o*ooo C6-24 0,1,3,4,6,8 412122 6Z24/- 0,2,4,5,6,9
**o**oo**ooo C6-25 0,1,3,4,7,8 412131 6Z19/- 0,1,2,5,8,9
**0*0**0*000 C6-26 0,1,3,5,6,8 412212 6Z25/- 0,2,3,4,7,9
**0*0*0**000 C6-27 0,1,3,5,7,8 412221 6Z26/Z48 0,1,2,5,7,9
**00***0*000 C6-28 0,1,4,5,6,8 413112 6-16/- 0,2,3,4,7,8
**oo**o**ooo C6-29 0,1,4,5,7,8 413121 -/6Z44 0,1,2,5,6,9
*o***o*o*ooo C6-30 0,2,3,4,6,8 421122 6-21/- 0,2,4,5,6,8
*o**o**o*ooo C6-31 0,2,3,5,6,8 421212 6Z23/Z45 0,2,3,4,6,9 1

53
2. Taxonomy of Chords and Scales

| *0**0*0**000 C6-32 0,2,3,5,7,8 421221 -/6Z47 0,1,2,4,7,9


1 *o*o**o**ooo C6-33 0,2,4,5,7,8 422121 -/6Z46 0,1,2,4,6,9
| *oo***0**000 C6-34 0,3,4,5,7,8 431121 -/- id.
| **o**0*00*00 C6-35 0,1,3,4,6,9 312123 6-27/- 0,2,3,5,6,9
1 **o**oo*o*oo C6-36 0,1,3,6,7,9 312132 6Z49/Z28 0,1,3,5,6,9
**o*o*o*o*oo C6-37 0,1,3,5,7,9 312222 6-34/- 0,2,4,6,8,9
**o*o*oo**oo C6-38 0,1,3,5,8,9 312231 6-31/- 0,1,4,6,8,9
**o*oo**o*oo C6-39 0,1,3,4,7,9 312312 6-30/6-30 0,2,3,6,8,9
**0*00*0**00 C6-40 0,1,3,6,8,9 312321 6Z29/Z50 0,1,4,6,7,9
**oo**oo**oo C6-41 0,1,4,5,8,9 313131 6-20/6-20 id.
*o**o*o*o*oo C6-42 0,2,3,5,7,9 321222 6-33/- 0,2,4,6,7,9
*o*o**o*o*oo C6-43 0,2,4,5,7,9 322122 6-32/6-32 id.
*o*o*o*o*o*o C6-44 0,2,4,6,8,10 222222 6-35/6-35 id.

In the affine classification, there are only 34 classes (36 extended classes) instead of 80
in the cyclic classification and 50 in the dihedral classification. The following table gives
the representative, the number in the cyclic classification, the prime form of the set, the
interval vector (fourth column), the number in the affine classification (fifth column) and
the symmetry groups of the sets.

| Representative tt Sets Iv Maz. Sym. Groups


| ******oooooo
1 C6-1 0,1,2,3,4,5 711111 63 < M7,n >
| **o*o*oo*o*o C6-43 0,2,4,5,7,9 322122 63.1 < M 7 ,ii >
1 *****o*ooooo C6-2 0,1,2,3,4,6 611112 64 {1}
| **o*0*0*00*0 C6-42 0,2,3,5,7,9 321222 64.1 {1}
| *****oo*0000 C6-3 0,1,2,3,5,6 611121 65 {1}
| ***o*00*0*00 C6-26 0,1,3,5,6,8 412212 65.1 {1}
| *****ooo*ooo C6-4 0,1,2,4,5,6 611211 66 < M 8 ,ii >
1 ***oo*o*o*oo C6-27 0,1,3,5,7,8 412221 66.1 < M8,n >
| ****o*o*0000 C6-6 0,1,2,3,5,7 511122 67 < Mio,5 >
| ****o*00*000 C6-8 0,1,2,4,5,7 511212 68 {1}
| ****o*000*00 C6-11 0,1,3,4,5,7 512112 69 {1}
| **o**0*0*000 C6-24 0,1,3,4,6,8 412122 69.1 {1}
| ****o*0000*0 C6-14 0,2,3,4,5,7 521112 70 {1,T 5 ,M 9 ,11,M 9 , 7 } |
| ****oo**oooo C6-7 0,1,2,3,6,7 511131 71 {1}
| ***oo*o**ooo C6-21 0,1,2,5,7,8 411321 71.1 {1}
| ****oo*o*ooo C6-13 0,1,3,5,6,7 512211 72 < M 6 ,7 >
| ****oo*oo*oo C6-12 0,1,3,4,6,7 512121 73 < M9,n >
1 **o*oo*o**oo C6-40 0,1,3,6,8,9 312321 73.1 < M9,n >
****ooo**ooo C6-10 0,1,2,5,6,7 511311 74 < M9,n > |
***o**oo*ooo C6-17 0,1,2,4,5,8 411213 75 {1}
**oo**oo*o*o C6-38 0,1,3,5,8,9 312231 75.1 {1}
***0**ooo*oo C6-23 0,1,3,4,5,8 412113 76 < M 7 ,5 >
***o*o*o*ooo C6-18 0,1,2,4,6,8 411222 77 < M 4 ,5 > |
***o*o*ooo*o C6-30 0,2,3,4,6,8 421122 78 {1}
** 0 *o*0*0*00 C6-37 0,1,3,5,7,9 312222 78.1 {1}
***0*00*00*0 C6-31 0,2,3,5,6,8 421212 79 {l,T7,Mio,5,Mio,n}
***0*000**00 C6-28 0,1,4,5,6,8 413112 80 < M 8 ,5 >
***00**0*000 C6-20 0,1,2,5,6,8 411311 81 < M6,7 >
***00**00*00 C6-25 0,1,3,4,7,8 412131 82 {1}
**0**00**000 C6-29 0,1,4,5,7,8 413121 82.1 {1} 1

54
Enumerations of Pcsets Classes

| ***ooo***000 C6-22 0,1,2,6,7,8 411411 83 Hi


**0**0*00*00 C6-35 0,1,3,4,6,9 312123 84 < M3,7 >
**o**oo*o*oo C6-36 0,1,3,6,7,9 312132 85 {1,77^8,11^8,5}
**0*00**0*00 C6-39 0,1,3,4,7,9 312312 86 H2
**oo**oo**oo C6-41 0,1,4,5,8,9 313131 87 H3
*0*0*0*0*0*0 C6-44 0,2,4,6,8,10 222222 88 H4 |

The groups ifi, #2, # 3 and if4 are defined by

#1 = {1,M7JO,M5,2,MI1J2,MI>6,M7,6,M5,8,MII)8}

iJ 2 = {l,M7,o,M5,4,Mii j4 }
H3 = {l,T5,M4 a ,M8,i,M4,5,M 8 j 5,Mii, 1 i,M3,ii,M7,ii,Mi 1 j,M3,7,M 7 ,7}
H4 = {l.Tj.Mij with i = 5,7,11 and j = 2,4,6,8,10}
In the symmetric classification, there are only 11 classes of hexachords. They are given
in the following table.
| Representatives B Sets Partitions
| ******oooooo S6-1 0,1,2,3,4,5 1,1,1,1,1,7
*****o*ooooo S6-2 0,1,2,3,4,6 1,1,1,1,2,6
*****oo*oooo S6-3 0,1,2,3,4,7 1,1,1,1,3,5
*****000*000 S6-4 0,1,2,3,4,8 1,1,1,1,4,4
****0*0*0000 S6-5 0,1,2,3,5,7 1,1,1,2,2,5
****0*00*000 S6-6 0,1,2,3,5,8 1,1,1,2,3,4
****oo*oo*oo S6-7 0,1,2,3,6,9 1,1,1,3,3,3
***o*o*o*ooo S6-8 0,1,2,4,6,8 1,1,2,2,2,4
***0*0*00*00 S6-9 0,1,2,4,6,9 1,1,2,2,3,3
**0*0*0*0*00 S6-10 0,1,3,5,7,9 1,2,2,2,2,3
*0*0*0*0*0*0 S6-11 0,2,4,6,8,10 2,2,2,2,2,2 |
3
Limited Transpositions Sets

Limited transpositions sets are of first importance in the combinatorial study of the n-
tone equal temperament (n-tet). They structure chords, modes and the whole tuning
assemblies by their singular properties. The problem of finding the subsets of the chro-
matic scale in the n-tet which are globally invariant by transposition has been studied
by Messiaen since 1929. Let us recall that in tempered music, a combination of notes has
limited transpositions when it gives back the same notes after less than twelve transpos-
itions. For example, the tritone (diabolus in musica) has only six transpositions, instead
of other intervals which have twelve transpositions. This chapter gives the mathematical
background for studying limited transpositions sets in n-tone equal temperament and
enumerates these sets for different values of n.

3.1 General Theory


The elements of the n-tet are identified with Zn= {0,1,2,3,..., n — 1}.
Definition 74 A pcset class of length k or a k-chord is a subset of Z n of cardinal k.
The translations T{ : x —> x + i (i G Z n ) act on the additive group Wjn and induce an
equivalent relation by
A ~B & BTi9 Ti{A) = B
Definition 75 A pcset A is the transposition of B if A is in relation with B. The number
of transpositions of a pcset is the cardinal of the orbit
TA = {T(A), T € T}
where T is the group of translations.
In general, each pcset has n different transposition sets (To = id), but in some cases,
the number of transpositions is less than n. This leads to the following definition.
3. Limited Transpositions Sets

Definition 76 A pcset A is a limited transpositions pcset if


3ieZn*=Zn\{0}> Ti(A) = A
In other words, if the isotropy group of Zn for the translations
HA = {ieZn, Ti(A) = A}
is not reduce to {0}. Let L be the set of the limited transpositions sets.
Definition 77 A "limited transpositions set class" or simply a "limited transpositions
set" is an element of the quotient set C = L/~ of the limited transpositions pcsets by the
cyclic group or equivalently by the translations. A generator set is a set of elements of
the quotient set C containing {0}.
Remark. This choice is well-defined because if A is a limited transpositions set (A G L)
then all its transpositions belong to L.

Vj Ti(Tj(A)) = Tj(Ti(A) = Tj(A)


The isotropy group HA is cyclic. So it is generated by an element i$.
Proposition 78 Let A € L such that Ti(A) = A for some i. If a e A, then
V m G Zn, Tmi(a) G A
In other words, if a is a note of the limited transpositions pcset A, then a-\-i, a-\-2i, ...,
a-\-mi are also in A.
Proof We have
Tmiip) G Tmi(A) = T^.^A) = ... = A

Theorem 79 The number of transpositions of the set A G L is the nonnegative integer
k such that
k x \HA\ = n
where n is the cardinal of the set Zn. In particular, the order of the isotropy group HA
is a divisor of n.
Proof As k is by definition the cardinal of the orbit T^, k is also the index of HA in the
group of transformations G isomorphic to Zn. This is in fact the Lagrange's Theorem.
The order of any finite group is divisible by the order of any subgroup of it. •
Corollary 80 Let A be a limited transpositions pcset containing {0} and HA its isotropy
group, each generator set is the union of A and sets of the form Tj(A) for j G HA-
Proof Let
B = AU
\JTi*M
U=i
be a such set. Since A is invariant under Ti and

Vji Ti(Tji(A))=Tji(Ti(A)) = Tji(A)


the set B is invariant under %. So B is in L. D

58
General Theory

Theorem 81 If n is a prime number, the set of limited transpositions pcsets is trivial,


i.e. reduces to one element, the chromatic scale in the n-tet.
proof It is well known that the subgroups of Z n are fcZn, where k is a divisor of n. If
n is prime, the subgroups are reduced to {0} and Z n itself. Thus the isotropy group is
trivial. •
Definition 82 Let x be a k-chord. The structure of x = {x\, ...,£&} where x\ < ... < xn
is the vector
[x] = [Axi,...,Axk]
where Axi = Xi+i — X{ for i = 1,..., k — 1 and Axk = x\ — Xk mod n.
Theorem 83 A pcset x is a limited transpositions set if and only if the structure fx] is
a partition of n which has a symmetry of order d, i.e. such that [x]=[A,...,A] where the
subset A is repeated d times and \x\ = d. \A\.
Examples. The set x = {0,1,3,6,7,9} has a structure [x] = [1,2,3,1,2,3]. This structure
has a 2-symmetry [x] = [A, A] with A = [1,2,3].

Algorithm. The last theorem leads to the following algorithm. Recall that a partition of
n is a way of writing the integer n as a sum of positive integers. For each partition of n,
we have to determine if a reordering of this partition has a d-symmetry. If it is the case,
we must add it to the set of solutions only if it (or a permutation of it) has not been
added before.

Examples. Let's discuss some examples for different values of n (2 < n < 10).

Lt-sets of Z2.The 2-tet case is trivial. It is reduced to one musical interval, the triton of
frequency ratio \f2 (600 cents).

Lt-sets of Z3. The 3-tet, identified with Z^ {0,1,2}, is based on the interval of frequency
ratio 2 1 / 3 (400 cents). Since 3 is prime, there is only one limited transpositions set
£ = {0,1,2}.

Lt-sets of Z4. The 4-tet, identified with Z4= {0,1,2,3}, is based on the interval of fre-
quency ratio 2 1 / 4 (300 cents). Prom the relation k x \HA\ = 4, we see that the order of
the isotropy group HA is 2 or 4. If \HA\ = 4, the isotropy group is Z 4 . There is only
one limited transpositions set which is Z4 itself. If \HA\ = 2, then the isotropy group is
HA = {0,2}. If 1 is in A, then, by the the previous theorem, A contains also 3. In this
case, there are two sets of the form Tj. (A), {0,2} and {1,3}. By union, we construct the
set {0,2} U {1,3} = {0,1,2,3} = Z 4 , but it is already obtained. Thus there is one limited
transpositions set, the set {0,2}. To sum up, there are two limited transpositions sets in
the 4-tet, E\ = {0,2} and E2 = {0,1,2,3}. The set E\ has two transpositions E\ and
Ti{E\) = {1,3} and the set E2 has only one transposition.

Lt-sets of Z 5 . The 5-tet has only one limited transpositions set because 5 is prime.

Lt-sets of ZQ. The 6-tet is the whole tone scale. There are 4 limited transpositions sets.
The set {0,3} has 3 transpositions. The set {0,2,4} has 2 transpositions. The set {0,1,3,4}
has 3 transpositions and the set {0, 1, 2, 3, 4, 5} has only one transposition.

59
3. Limited Transpositions Sets

Lt-sets of Z7. The 7-tet has only one limited transpositions set because 7 is prime.

Lt-sets of Zs- The 8-tet has five limited transpositions sets. The set {0,4} has 4 transpos-
itions. The set {0,2,4,6} has 2 transpositions. The set {0,1,4,5} has 4 transpositions. The
set of 6 notes {0,1,2,4,5,6} has 4 transpositions and the chromatic set {0,1,2,3,4,5,6,7}
has only one transposition.

Lt-sets of Zg. The 9-tet has three limited transpositions sets L = {£?!, E2, -£"3}. If \HA\ =
9, then the isotropy group is a limited transpositions set E\ = {0,1,2,3,4,5,6,7,8}. If
\HA\ = 3, then the isotropy group is HA = {0,3,6} = E2. The transpositions are
Ti{HA) = {1,4,7} and T2(HA) = {2,5,8}. The union HAVT1(HA)\JT2(HA) is equal
to Ei. The set E3 = HAUTi(HA) = {0,1,3,4,6,7} is a limited transpositions set. The
union HA U T2(HA) = {0,2,3,5,6,8} is the transposition by 2 of the set £3.

Lt-sets of Z10. The 10-tet identified with Z10 is based on the interval of frequency ratio
2i/io ^ 2 0 cents). It has 8 limited transpositions sets. The set {0,5} has 5 transposi-
tions. Its complementary set {0,1,2,3,5,6,7,8} has also 5 transpositions. The set {0,2,5,7}
and {0,1,5,6} has 5 transpositions as well as the complementary sets {0,1,3,5,6,8} and
{0,1,2,5,6,7}. The chromatic set {0,1,2,3,4,5,6,7,8,9} has only one transposition.

3.2 Messiaen Modes


In the first half of the twentieth century, Olivier Messiaen introduced the concept of
limited transpositions sets, but he only used some of them in his own compositions.
Hhe called these sets limited transpositions modes. Some sets were already known before
Messiaen presented his theory. The whole tone scale was used by many composers. Claude
Debussy wrote a prelude (Voiles) using only this mode. Messiaen's third mode is the
enneaphonic scale that Scriabin uses in his Piano Prelude, opus 7^. An analysis of the
combination reveals that there is in the system of half tones sixteen assemblies with
limited transpositions. The following table shows the limited transpositions sets. The
last column indicates the number of transpositions.
|~Names Limited Transpositions Sets Tr.
| Tritone (0,6) 6
Aug. Fifth (0, 4, 8) 4
| Dim. Seventh (0, 3, 6, 9) 3
| Alt. Seventh (0, 2, 6, 8) 6
Double Fifths (0, 1 , 6 , 7 ) 6
| Mode 1 (0, 2, 4, 6, 8, 10) 2
Mode 2 (0, 1, 3, 4, 6, 7, 9, 10) 3
Mode 3 (0, 1, 2, 4, 5, 6, 8, 9, 10) 4
| Mode 4 (0, 1, 2, 3, 6, 7, 8, 9) 6
Mode 5 (0, 1, 2, 6, 7, 8) 6
Mode 6 (0, 1, 2, 4, 6, 7, 8, 10) 6
Mode 7 (0, 1, 2, 3, 4, 6, 7, 8, 9, 10) 6
Mode A (0, 1, 4, 5, 8, 9) 4
Mode B (0, 1, 3, 6, 7, 9) 6
Mode C (0, 1, 4, 6, 7, 10) 6
Chromatic Scale ( 0 , 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10, 11) 1 |

60
Messiaen Modes

The five first sets are classified or altered chords, the seven following modes (mode 1
to mode 7) show Messiaen's modes and the last three (modes A, B and C) are defective
because they are included in Messiaen's modes: mode A (or mode 8) in mode 3, mode
B (or mode 9) in mode 2 and mode C (or mode 10) is included in mode 6. Remark that
the mode 6 is included in the mode 7 which is not really used by Messiaen. Lastly, the
chromatic total has only one transposition and contains all the other modes. The following
figure shows the distribution of the limited transpositions sets in tempered music. The
rc-axis represents the number of notes and the y-axis the number of transpositions. The
lefts side of the tree corresponds to the chords and the right side corresponds to the
Messiaen modes.

2 »o

I
(4 transp.)

(6 transp.)
(9 notes)

(8 notes)
Mode 3

Mode 4
J
(6 transp.)
(8 notes)
ModeS

61
3. Limited Transpositions Sets

Let's introduce Messiaean's modes in the classical notation and in the molecular
presentation. In the molecular presentation, the x-axis represents the major thirds (trans-
lation T4) and the y-axis represents the tritonus (translation TQ). In this frame {T^TQ),
the fifths (or fourths) are placed along the diagonal axis.
The whole tone scale is the first Messiaen's modes. It has only two transpositions.

$ . . .il'iM'

$ !»!>»'*

In a frame (T^^TQ), the molecular presentation is

The mode 2 (also called the Bertha's mode) has three transpositions.

i >^.M«'^
22
^ ^
$

23

I , ^ 1»fl»^ m »

The molecular presentation shows the limited transpositions subsets: the modes B
{0, 1, 3, 6, 7, 9} and C {0, 1, 4, 6, 7, 10}, the chord of diminished seventh {0, 3, 6,
9}, the chord of altered seventh {1, 3, 7, 9}, the double fifth {0, 1, 6, 7}, various trans-
positions and four tritonus represented by the vertical segments.

This mode is widely used in the first regard of Messiaen's Vingt regards sur Venfant
Jesus.

62
Messiaen Modes

The mode 3 has four transpositions.

» . I>» \* 0^ ^

The molecular presentation shows the limited transpositions subsets of mode 3 the
mode 1 {0, 2, 4, 6, 8, 10}, the mode A {0, 1, 4, 5, 8, 9}, the chord of altered seventh
{0, 2, 6, 8} and the chord of augmented fifth {0, 4, 8}.

The mode 4 has six transpositions.

I ^ i » 'i*«» ^ ^
It has several limited transpositions subsets: the mode 5 {0, 1, 2, 6, 7, 8}, the mode B
{0, 1, 3, 6, 7, 9}, the mode C {0, 2, 3, 6, 8, 9}, the chords of diminished seventh {0, 3,
6, 9}, of altered seventh {0, 2, 6, 8}, the double fifth {0, 1, 6, 7}.

The mode 5 has six transpositions. It is included in the mode 4.

ipppp^s^
It includes the chord of altered seventh {0, 2, 6, 8} and the double fifth {0, 1, 6, 7}.
1,

The mode 6 has six transpositions.

$ » . « » ll» **

63
3. Limited Transpositions Sets

The molecular presentation shows the limited transpositions subsets of mode 6: the mode
1 {0, 2, 4, 6, 8, 10}, the mode 5 {0, 1, 2, 6, 7, 8}, the mode B {1, 2, 4, 7, 8, 10}, the mode
C {0, 1, 4, 6, 7, 10}, the chord of diminished seventh {1, 4, 7, 10}, the chord of altered
seventh {0, 2, 6, 8}, the double fifth {0, 1, 6, 7} and the chord of augmented fifth {0, 4,
8}.

The mode 7 has six transpositions.

i» ii» i«
$ • i * ii«

The mode 7 includes all the sets of limited transpositions except the modes 3 and A. It
is a good complement of these modes.

The other modes are included in the previous modes.


The mode A is included in the mode 3 2 (second transposition of mode 3).

The mode B is included in the mode 2. This mode is cited by Messiaen in his Technique
de mon langage musical (example 342).

The mode C is included in mode 6 3 (third transposition of mode 6). This mode is also
cited in Messiaen's Technique de mon langage musical (example 341).

64
Messiaen Modes

The following theorem has been established by B. Parzysz [Parzysz, 1983].

Theorem 84 In the 12-tet, there are only 16 limited transpositions sets which can be
classified with the number of transpositions k.
1) If k = 1, there is only one limited transpositions set,
2) If k = 2, there is only one limited transpositions set,
3) If k = 3, there are two limited transpositions sets,
4) Ifk = 4, there are three limited transpositions sets,
5) If k = 6, there are nine limited transpositions sets.
Proof The 12-tet is identified with Z12. Since the order of the isotropy group is a divisor
of 12, we have to study five cases.
1) If l-ff^l = 12, then the chromatic scale

^ = {0,1,2,3,4,5,6,7,8,9,10,11}

is the only limited transpositions set.

2) If \HA\ = 6, then the isotropy group is the whole tone scale

HA= E2 = {0,2,4,6,8,10}

With Ti(i?2), we can make by union, the set

E2 U T\ (E2) = Ei

which is already obtained.

3) If \HA\ = 4, then the isotropy group is the diminished seventh chord

HA=E3 = {0,3,6,9}

By making the unions of E$ with T\{E3) and T2(E^)i three new sets are obtained. The
set
£3 ^ ( £ 3 ) = £4 = {0,1,3,4,6,7,9,10}
is Messiaen's mode 2. The set

£3 U T2(E3) = {0,2,3,5,6,8,9,11} = T 2 (£ 4 )

is the transposition of E4 by 2. The set E3 U Ti(E3) U T2(E3) is the chromatic scale


already obtained.

65
3. Limited Transpositions Sets

4) If \HA\ = 3, then the isotropy group is the augmented fifth

HA=E5 = {0,4,8}

Three sets are formed by unions with Tj(Es). The set

E 5 UT 1 ( J B 5 ) = £ 6 = {0,1,4,5,8,9}

is the mode A. But the set


EbUT3(Eb) = T3(E6)
is the transposition of the previous set. The combination

-E?5 U T2(E§) = E2

leads to the whole tone scale. The set

£5 U Tj(E5) U T2(E5) = E7 = {0,1,2,4,5,6,8,9,10}

is Messiaen's mode 3. The other combinations are two transpositions of this mode, namely

E5 U Ti(Es) U T3(E5) = T3{E7)

and
£5 U T2(E5) U T3(E5) = T2{E7)

5) If |.ff>i I = 2, then the isotropy group is the tritone

^ = E 8 = {0,6}

The set

E8UT1(ES)UT2(ES)UT3(ES) UT4(E8) = E9 = {0,1,2,3,4,6,7,8,9,10}

is Messiaen's mode 7. The other combinations are transpositions of this mode

tfgUTj^ura^UTaC^UTsCEs) = T5(E9)
T
£8UT1(£8)UT2(£'8)ur4(£:8)ur5(£8) = 4(£g)
EsUT1(Es)UT3(E8)UT4(E8)UT5(Es) = T3(E9)
E8l)T2(E8)UT3(E8)\JT4(Es)uT5(E8) = T2(E9)

The set
E8l)T1(E8)UT2(E8)L)T3(E8) = Ew = {0,1,2,3,6,7,8,9}
is Messiaen's mode 4. The following unions are transposition sets.

E8UT1(E8)l)T2(E8)UT5(E8) = T5(E10)
E 8 UT 1 (E 8 )UT 4 (£' 8 )UT5(£ , 8 ) = T4{E10)
E8l)T3(E8)UT4(E8)UT5(E8) = T3(E10)

The set
Es U T^Es) U T2(E8) U T4(E8) = Eu = {0,1,2,4,6,7,8,10}

66
Messiaen Modes

is a new limited transpositions set (Messiaen's mode 6). The other sets are transpositions
of mode 6.
ESUT1(ES)UT3(ES)UT5(ES) = T5(En)
E8UT2(E8)UT4(E8)l)T5(E8) = T4(En)
E8UT2(E8)liT3(E8)uT4(E8) = T2(En)
The sets
E8UT1(E8)UT3(E8)\JT4(E8) = E4
E8UT2(E8)UT3(E8)UT5(E8) = T2(E4)
are transpositions of the mode 2 already obtained. The study of the combinations of the
form E8 U Ti(E8) U Tj(E8) reveals three new limited transpositions sets. The set
E8 UT^Eg) UT2(E8) = E12 = {0,1,2,6,7,8}
is the mode 5. The following sets are transposition sets of mode 5.

EsUT1(Es)UT6(Es) = T5(E12)
E8UT4(E8)UT5(E8) = T4(E12)
The set
E8UT1(E8)L)T3(E8) = E13 = {0,1,3,6,7,9}
is the mode B. The following sets are transposition sets of mode B
E8UT2(E8)UT5(E8) = T5(E13)
E8UT3(E8)UT4(E8) = T3(E13)
The set
E8 U I\(£ 8 ) U T4(E8) = E14 = {0,1,4,6,7,10}
is the mode C. The following sets are transposition of the preceding set.
E8UT3(E8)UT5(E8) = T5(E14)
E8UT2(E8)UT3(E8) = T2(E14)
The set
E8\JT2(E8)l)T4(E8) = E2
is the whole tone scale already obtained. The next step is the study of the sets E8\JTj(E8).
This leads to two new limited transpositions sets
E8 U T^Es) = E8U T5(E8) = E15 = {0,1,6,7}
and
E8 U T2(E8) = E8U T4(E8) = E16 = {0,2,6,8}
The set
£$ U T3(E8) — E3
is already obtained. D

67
3. Limited Transpositions Sets

3.3 Diatonic Scales and Microtonality


Limited transpositions sets have nice relationships with major and minor scales. It is well-
known for the 12-tone classical temperament that the presence of limited transpositions
sets in the structure of major scales has strong implications in the modulation process.
Some authors have tried to extend these properties to the n-tone equal temperament.
Most of the results of this section are coming from [Balzano, 1980], [Gould, 2000] and
[Broue, 2001].
Proposition 85 Let u be an element ofZn. The transpositions T% with k = 0, ...,r (i.e.
the cyclic group) generate the n transpositions of Z n if and only if u is prime with n.
Example. For n = 12, the numbers ^ = 1,5,7,11 lead to the 12 transpositions (11 = —1,
7 = — 5). The first couple (1, 11) is the chromatic transposition at the lower or upper
semitone. Thus the powers Xi, T? = T2, ..., Tf = Tn generate the n transpositions. The
second couple (5, 7) is the transposition at the fifth or at the fourth. The powers of T 5
generate n transpositions.

The following proposition shows that the limited transpositions set L is a structural
element of A in the sense that L connects the transpositions of A and thus makes easier
the modulations.
Proposition 86 Let d be a divisor of n. Denote Ld the set
L d = {0,fc,2fc,...,(d-l)fc}
where n = kd. Let A be a pcset class with exactly n transpositions (A is not a limited
transpositions set) and containing a set L of pcset class in Ld. Then L is included in
exactly d pcset classes Tm(A), with m = 0,1,..., (d — 1).
Definition 87 A set A of m pitch classes is neighbor of a set B of the same cardinality
if A and B have m — 1 pitch classes in common
\AnB\=m-l
The following result shows that for n = 12, there is only one major scale. It also means
that there is only one dichotomy of the 12-tone in 7 white keys and 5 black keys.
Theorem 88 Let m,n,u be non-negative numbers with 0 < m <n, 1 <u < n, and u is
prime with n. There is only one pcset class A verifying
(1) The cardinality of A is m, \A\ = m
(2) A has exactly n transpositions, A £ C
(3) A is in the neighborhood ofTu(A), \AnTu(A)\ =m-l
Let x be a pitch class ofZn, a representative of A is given by the set,
A= {x,Tu{x),Tl{x),...,TZ-l{x)}
Example. For n = 12, m = u = 7 and x = 5, the set A is the C major scale
A = {5,0,7,2,9,4,11}, that is the white keys. For m = 5 and x = 6, A is the black
keys, A = {10,3,8,1,6}.

With the three criteria given in the previous theorem, one could define a diatonic scale.
Thus the theorem says that there is only one diatonic scale for each m. Balzano [1980]

68
Diatonic Scales and Microtonality

and Gould [2000] defined diatonic scales from the diagonal sets emerging from the tiling
of the plane. In the following figure, the z-axis represents the translations T4 and the
y-axis the translations T3.

2-
1 11
11-

8' 8
VL
IV
5
I
2
The limited transpositions sets are also structural elements of the minor scale.
Proposition 89 In the 12-tone equal temperament, there is only one pcset class such
that each transposition A has six elements and A is the union of an element of L3 and
an element of L4. A is the minor scale without its tonic.
Example. Let's consider A minor. Without its tonic (i.e. the pitch class 9), A is the set
A = {0,2,4,5,8,11}. This set is the union of an element of L3, {0, 4, 8} and an element
of L4 {2, 5, 8, 11}. Adding a pitch class to A leads to form a set of 7 notes. There are
six choices, but only two are in the neighborhood of the major scale.
Theorem 90 In the 12-tone equal temperament, there are only two pcset classes in the
neighborhood of the major scale: the minor scale A, and its transposition at the third of
the inversion of A, 14(A) = T4l(A).
Example. In the neighborhood of the C major scale {0, 2, 4, 5, 7, 9, 11}, there are only
two scales A ={9, 11, 0, 2, 3, 5, 8}, A minor and the transformation /4(A) = {11, 0, 2,
4, 5, 7, 8}.

Definition 91 Let m,n,u be non-negative numbers with 0 < m < n, 1 < u < n, and u,
n are coprimes. A pcset class A with cardinality m is a diatonic scale if
(1) A does not contain three consecutive pitch classes
Vx e Z n , {x, x + 1, x + 2} £ A
(2) A contains one pcset of Ld, with d divisor of n
3x € Z n , Ts({0,fc,2fc,..., (k - l)d}) e A
(3) A is in the neighborhood ofTu(A)
\AnTu(A)\=m-l
Theorem 92 In the n-tone equal temperament, for a given u as above, there is only one
diatonic scale A of cardinal m, given by

where x is a pitch class ofZn.

69
3. Limited Transpositions Sets

Examples. (1) For n = 18. The integer u is one of 1, 5, 7, 11, 13, 17 (11 = —1, 13 = —5
and 17 = —7). For u = 5, the set of divisor of n is {1, 2, 3, 6, 9}. The sequence 0, 5, 10,
15, 15, 2, 7, 13, 17, 4, 9, 14, ... contains two elements of L2 (0, 9) and (5, 14). Thus, in
the field of the thirds of tone (i.e. 18-tet), the diatonic scale of length m = 10 is the set
whose elements are located before 14 in the previous sequence. A = {0, 2, 4, 5, 7, 9, 10,
12, 15, 17}. For u = 7, we get A = {0, 2, 3, 6, 7, 9, 10, 13, 14, 17}.
(2) For n = 24. The cycle of the multiples of u = 7 ends with the second element of L2 (7,
19). Thus in the field of the quarter tones (i.e. 24-tet), the diatonic scale of length m = 13
is the set A = {0, 1, 4, 5, 7, 8, 11, 12, 14, 15, 18, 21, 22}. It is composed of G major scale
{0, 4, 8, 12, 14, 18, 22} = 2 x {0, 2, 4, 6, 7, 9, 11} and the transposition a quarter tone
up of B\> major scale without triton 2 x {0, 2, 3, 5, 7, 10}, to avoid ambiguity.
Remark. Our definition of diatonic scales are different from the definitions of G. Balzano
and M. Gould. However in some cases, the Bolzano-Gould scale is a subset of the unique
diatonic scale. For example, in the case Z ^ + i ) , with k = 4, the tiling of the plane with
the translations T4 vertically and T§ horizontally leads to the following figure.

9 14 19 4 9
1
1
5 10 15 0 -- 5
1
1
1 6 11 -- 16 1
1
1
17 2 -- 7 12 17
1
1
13 -- 18 3 8 13
1
1
9 14 19 4 9
The diagonal set U = {0, 2, 5, 7, 9, 11, 13, 18} (which is considered as a diatonic scale by
M. Gould) is included in the set A = U U {3,14} which is precisely the diatonic scale for
n = 18, u = 9, and m = 11. The cycle of the multiple of 9 is not stopped at 9, although
9 appears two times in the diagonal. Thus a limited transpositions set of L2 (3, 13) is
included in the diatonic scale A in order to characterize the transpositions of the set A.

3.4 Quarter-Tone Limited Transpositions Sets


In this section, we extend the results of B. Parzysz [Parzysz, 1983] to the 24-tet.

Theorem 93 In the 24-tet, there are only 381 limited transpositions sets which can be
classified by the number of transpositions k.
1) If k = 1, there is only one limited transpositions set,
2) If k = 2, there is only one limited transpositions set,
3) If k = 3, there are two limited transpositions sets,
4) If k = 4, there are three limited transpositions sets,
5) If k = 6, there are nine limited transpositions sets,
6) Ifk = 8, there are 30 limited transpositions sets,
7) If k = 12, there are 335 limited transpositions sets.

70
Quarter-Tone Limited Transpositions Sets

Proof Let H be the isotropy group. Since Card(H) is a divisor of 24, the number k
takes its value in the set {1,2,3,4,6,8,12}. It is then easy to construct in each case,
the isotropy group H containing the note {0}. By union of separate sets Ti(if), one
constructs the set of all limited transpositions sets. Since the subgroups of Z24 are not
separate, we could find, during the analysis, some sets already obtained. The proof can
be simplified by symmetry arguments, but we will not do that, because the complete
demonstration gives also the explicit forms of the limited transpositions sets, and not
only their enumeration. Seven cases have to be considered:

1) If k = 1, then Card(H) = 24 and H = Z24. There is only one limited transpositions


set. It is the chromatic scale of 24 notes.

2) If k = 2, Then H = 2Z 24 = B = {0,2,4,6,8,10,...,22}. There is only one limited


transpositions set, which has two transpositions. It is the chromatic scale with 12 notes.

3) If k = 3 then H = 3Z 24 = {0,3,6,9,12,15,18,21}. We construct all the possible


unions from the sets Ti{H) with i = 0,1,2 in order to make a set containing the note
{0}. There are two generator sets C\ = H of 8 notes and C2 = H U T\(H) of 16 notes
because
HUT2(H) = T2(HUT1(H))

4) If k = 4 then H = 4Z 24 = {0,4,8,12,16,20}. From the four sets T^H), i = 0,1,2,3


we find three generator sets. A set
D1=H
of six notes. Another set of twelve notes

D2 = HUT1(H)
and its transposition sets

H\JT2{H) =B s,ndHUT3(H) = T3(HUT1(H))

A set of 18 notes
£3= | J Tj{H)
j=l,2,3
and its transpositions

.7=1,2,3
T H
T2(Ds) = U ^ )
.7=0,2,3

T3(D3) = | J Tj(H)
j=0,l,3

5) If k = 6 then H = 6Z24 = {0,6,12,18}. There are nine generator sets,

a) A set of four notes: E\ = H.

71
3. Limited Transpositions Sets

b) Two sets of eight notes


E2 = HUT1(H)
and its transposition set
HUT3(H) = C1

which are already obtained. The two other sets are transpositions of E2 and E3

HUT4(H) = T4(E2)

and
HUTb(H) = n(Es)

c) Three sets of twelve notes. Prom the six sets Ti(H) , i = 0,1, ...5, one makes (g) = 10
sets of twelve notes containing {0}, but only three are generator sets. The set

E4= | J Tj(H)
i=o,i,2

and its transposition sets


T4(E4) = | J Tj(H)
J=0,4,5

and
T H
T5(£4) = (J ^ )
j=0,l,5

The set
E5= | J Tj(H)
j=0,l,3

and its transposition sets


T3(E5)= (J 21(10
i=0,3,4

and
T5(E5)= | J ?}(#)
i=0,2,5

The set
E 6= |J ^(ff)
i=0,l,4

and its transposition sets


T2(E6)= | J T^IT)
j=0,2,3

and
T5(E4)= (J ^(tf)
J=0,3,5

The tenth set H U T2(.ff) U T4(H) = B is already obtained.

72
Quarter-Tone Limited Transpositions Sets

d) Two sets of 16 notes. Prom the six sets Ti(H) with i = 0,1,2, ..,5, one constructs
(5) = 10 sets of 16 notes containing {0}, but two only are generator sets. The set

E7 = |J Tj(H)
J'=0,l,2,3

has four equivalent forms Tj(E7) with j = 0,3,4,5. The set

E8 = |J TjiH)
i=0,l,2,4

has also four equivalent forms Tj(Es) with j = 0,2,4,5. The set

C2= [J T,(ff)
j=0,l,3,4

and its equivalent form T2(C2) are already mentioned.

e) A set of 20 notes
E9= |J Tj(H)
j=0,l,2,3,4

6) If k = 8 then # = 8Z 24 = {0,8,16}. Prom the eight sets Ti(H) with i = 0,1,...., 7,


one can build 2 7 = 128 sets containing {0}, but only 30 are generator sets.

a) A set of three notes: F\ = H.

b) Three sets of six notes. Prom the seven sets Ti(H) with i = 1,2,..., 7, we get

F2 = HUT1(H) because H L\T7(H) = T7(F2)

F3 = HUT2(H) because HUT6(H) = T6(FS)


F4 = HUT3(H) because HUT6(H) = T 5 (F 4 )
The seventh set is already mentioned. It is

D2 = HUT4(H)

c) Seven sets of nine notes. From the eight sets Ti(H) with i = 0,1,.., 7, one can build
(2) = 21 sets of the form H U Ti(H) U Tj(H) with i ^ j . However, each of those sets
has three equivalent forms (including itself), so there are only 21/3 = 7 sets, denoted by

*6, ,Fn.

d) Eight sets of twelve notes. There are (3) = 35 sets of the following form.

[" 3 1
J' = 1

73
3. Limited Transpositions Sets

But two sets have been already mentioned. It is the set

D 2= |J Tj(H)
i=0,l,4,5

and the set


B= |J TjiH)
j=0,2,4,6

Consequently, there are three forms to subtract to the 35 sets, so 35 — 3 = 32. Each
of these 32 sets has four equivalent forms (including itself), so we have 32/4 = 8 sets,
denoted by F12,...., F\g

e) Seven sets of 15 notes, denoted by F20, ....,^26- There are Q = 35 sets of the form
X^k) with k = 5 and
ffc-i
X$k)=H\J

Each of these sets has five equivalent forms, so it remains only 35/5 = 7 sets of 5x \H\ = 15
notes.

f) Three sets of 18 notes. There are Q = 21 sets of the form X\ ' with k = 6. But the
set
D 2= |J Tj(H)
i=0,2,3,4,6,7

is already obtained. Since it has three equivalent forms, there are only 21 — 3 = 18 sets
which has, each, six equivalent forms. So, there are 18/6 = 3 sets of 18 notes.

g) A set of 21 notes
F30= |J Tj(H)
j=0,l,...,5

7) If k = 12 then H = 12Z 24 = {0,12}. One constructs all the unions from the sets Ti(H)
with i = 0,1,..., 11. We get

a) One set of 2 notes: G\ = H.

b) Five sets of 4 notes. There are 11 sets of the kind X\ ' with k = 2. Since the set

E1=HUT6(H)

is already mentioned, it remains 11 — 1 = 10 sets, each with two equivalent forms. So,
there are 10/2 = 5 sets of 4 notes.

c) 18 sets of 6 notes. There are ^) = 55 sets of the kind X\ ' with k = 3. Since the set

DX = (J Tj(H)
.7=0,4,8

74
Quarter-Tone Limited Transpositions Sets

is already mentioned, we have only (55 — l ) / 3 = 18 sets of 6 notes.

d) 40 sets of 8 notes. There are (^J = 165 sets of the form X\ ' with k = 4. Since the
sets
Ci= (J Tj{H) = Ti{C1)
j=0,3,6,9

fori = 0,3,6,9.
E2 = (J Tj(H)=T6(H)
j=0,l,6,7

and its transposition T^(E2) = Tn(E2)

E3= |J Tj(H)=T6(E3)
j=0,2,6,8

and its transposition T^Es) = Tio(^3), have been already mentioned, there are only
(165 - 5)/4 = 40 sets of 8 notes.

e) 66 sets of 10 notes. There are ^) = 330 sets of the kind X\ ' with k = 5. Consequently,
we have 330/5 = 66 sets of 10 notes.

f) 75 sets of 12 notes. There are (g1) = 462 sets of the form x\k) with k = 6. But the
sets

B= |J TiiJS)
j=0,2,4,6,8,10
and
D 2= (J Tj(H)
.7=0,1,4,5,8,9

and its equivalent Ts(D2), the set

E4= |J T^iJ)
j=0,l,2,6,7,8

and its equivalents T±{E±) and T^E^), the set

£5 = (J Tj(H)
j=0,l,3,6,7,9

and its equivalents T^Es) and T5(£ 5 ), the set

E6= |J T,(ff)
j=0,l,4,6,7,10

and its equivalents T2(EQ) and T5(i?6) are already mentioned. Amongst the sets of the
form X\ ' some sets can be identified with Fi for i = 12 to 19. We will prove by reductio
ad absurdum that it is impossible. If so, the sets Fi will verify the identity

T12(Fi) = Fi

75
3. Limited Transpositions Sets

But the notes {16,17} belongs to all the sets Fi, because the sets Fi are of the form

Fi= |J ^({0,8,16})
j=0,l,p,q

So, the notes {4,5} belongs to all the sets Fi , it means that

Fi= |J ^({0,8,16})
5=0,1,4,5

for all i. But this set is the set D2 which is already mentioned. The hypothesis Tyz^Fi) =
F{ is absurd. The seeking sets are not of the form Fi. Thus there are 12 sets already
mentioned, and consequently, there are (465 — 12)/6 = 75 sets of 12 notes.

g) 66 sets of 14 notes. There are ft) = 462 sets of the kind X\ ' for k = 7 and so
462/7 = 66 sets of 14 notes.

h) 40 sets of 16 notes. There are ft) = 330 sets of the kind x\k) with k = 8. Since the
sets C2 and its transposition, the sets E7 and E$ with their four equivalent sets, that is
a total of 10 sets, have been already mentioned, there are (330 — 10)/8 = 40 sets of 16
notes.

i) 18 sets of 18 notes. There are (g1) = 165 sets of the kind X\ ' with k = 9. Since the
three sets equivalent to D3 are already mentioned, there are (165 — 3)/9 = 18 sets of 18
notes.

j) 5 sets of 20 notes. There are ft) = 55 sets of the kind x\ ' with k = 10. Since the five
sets equivalent to Eg are already mentioned, there are (55 — 5)/10 = 5 sets of 20 notes.

k) A set of 24 notes. There are 11 sets of the kind X\ ' with k = 11. Consequently, there
are 11/11 = 1 set of 24 notes. •

3.5 Enumerations
The goal of this section is to find the number of limited transpositions sets for a given
equal temperament of n notes. P61ya theorem gives a generator polynomial in which the
coefficient of xk is exactly the number A;-chord in n-tone equal temperament. For the
action of the cyclic group, the polynomial is given by the formula

d\n
where ip is the Euler's function. For x = 0, this formula gives the number of pcsets classes

n d
l^
d\n
The following result has been established in [Ballon, 2004]. We denote Ln the number of
limited transpositions sets, Mn the number of modes, Kn the number of modes associated
with limited transpositions sets. A pcset class of k degrees is associated with at most k
modes.

76
Enumerations

Theorem 94 The number of limited transpositions sets is given by


2
Ln = P n + -(Kn-Mn)
n
where Pn is given by the Pdlya theorem, Mn is the number of modes Mn = 2 n _ 1 and Kn
is given by
1 r r—fc+1 r—fc+2 r
fc+1 2J
^ = i£(-i) E E •••• E ^^
k=l i i = l i2=ii + l ifc=z fc _i+l
where the decomposition in prime factors of n is n = p^p^-'Pr1*, with pi > 1, ki > 0,
(i = l,...,r) a n d r > 0.
Proof Let Pn,k be the number of pcsets classes of k notes, and Mnyk the number of
modes of k notes. We have
n n
Pn k aild
Pn = Yl > Mn = ^Mn,k
fc=l k=\

Let Ln>fc the number of limited transpositions set of k notes, and KU}k the number of
modes associated with the limited transpositions sets of k notes. A pcset class of k notes
which is not a limited transpositions set has exactly k modes, thus the numbers of the
respective quantities verify the relation

Pn,k — Ln,k = -j-(Mn,k ~ Kn,k)

As the number of pcsets classes of k notes and the number of pcsets classes oin — k notes
are equal, the sum over k leads to

Pn-Ln = -(Mn-Kn)
n
Prom this relation, we get the expression of Ln. On the other hand, the number of modes
of k notes is
(n-l)!
M, l k
' ~ [k - 1 j ~ (n • *)!(fc - 1 ) !
This expresison leads to the total number of modes:
n-l

k=i fc=ov K J

The relation Kn is more difficult to prove. Let Mn be the set of modes and /Cn the set
of modes associated with limited transpositions sets. Let n be n = Pilp22'~Prr a n d m be
a divisor of n such that m ^ 1. The elements of /Cn are constructed from the elements
of the sets Mn/p. with j = 1,..., r. For each divisor m of n, the injection

maps a mode M of n/m notes with a limited transpositions modes of /Cn by repeating
n/m times the structure of M. The set /Cn is the union of the sets
r
K,n = \JfPj {Mn/pj)
3. Limited Transpositions Sets

By applying the following result to a collection of r sets Ai,..., Ar and by denoting \A\
the cardinal of the set A,

r r r—k+1 r—fc+2 r
[JAj = ]T(-i)*+1 £ £ .. £ |^nA i2 n...n^ t |
j=l fc=l ii=l i2=ii+l i fc =ifc_i+l

we get

r r-A:+l r-fc+2
w
^ = i/t;i=B-i) E E -
k=l ii=l i2=ii+l
r
/ y |/pi1--Pifc(^n/pi1..pifc)|
*fc=ifc-i+l

For two divisors mi and m2 of n, the function / m verifies

/ m i (,«A/ln/mi

But / m is an injective mapping. For all divisor m of n, we have

\fm{Mn/m)\ = \Mn/m\ = Mn/m

and then

\Jpii~Pik \-Mn/pi1..pik )\ — \^n/pil..pi = 2PhPi2-Pik

Replacing this relation in the expression of Kn leads to the result. •


The same demonstration can be used for the enumeration of the modes of k notes
associated with transposition limited sets.

Theorem 95 Let n be the number of notes per octave in the equal temperament, and k
an integer less than n. LeVs denote m the number of common prime factors of n and k
and qi>q2>~">Qm these factors. The number of modes Mnj/- is given by

m m—fe+lm—/c+2 m
fc+1 Qilqi2-"(lik
Mn,k = £(-D E E •••• E k ;)

By convention, the sums with terms in reverse order are ignored.

The number of pcsets classes (Pcsc) in the n-tet and the number of limited transpos-
itions sets classes (Ltsc) are given in the following table for different values of n, in the
cyclic classification.

78
Enumerations

13 n Ltsc
1
Pcsc
3
114n Ltsc
20
Pcsc
1 181
4 2 5 15 9 2 191
5 1 7 16 35 4 115
6 4 13 17 1 7 711
7 1 19 18 69 14 601
8 5 35 19 1 27 595
9 3 59 20 110 52 487
10 8 107 21 21 99 879
11 1 187 22 188 190 745
12 16 351 23 1 364 723
13 1 631 | 24 381 699 251 |
For divisions of ra-th of tone (m > 5, m = n/6), we get large values

1/ra n Ltsc Pcsc


1/5 30 2 300 35 792 567
1/6 36 14 939 1 908 881 899
1/8 48 703 331 5 864 062 367 251
1/12 72 1 909 580 799 65 588 423 374 144 427 519
1/16 96 5 864 196 582 931 825 293 359 523 589 782 053 586 451

Another approach to get the enumeration is to solve diophantine equations.

T h e o r e m 96 Let H he the set of divisors of n strictly less than n and Lk he the set of all
values I G H such that Lk = 0 mod n. The numbers x^^ ofk-chords with £ transpositions
verify for each k the equation
(n-l)l ^f l\
k\(n-k)\ ^ \ n)

where Xkye are positive integers and Pn^ is the coefficient of xk in the Pdlya generator
polynomial of k-chords.

The previous formula gives the number of limited transpositions sets. But in some
cases, there are several solutions for the Xkyi and another argument to solve the system
is needed.

Examples. Lt-sets of Z30. For the 30-tone equal temperament (Fifths of tone), we have
to solve several equations.
For fc = 6, the equation

5rc6)5 + 4z 6j io + 3z 6) i5 = 111
has solutions
#6,5 = n\
#6,10 = ^1 + 3 n 2
#6,15 = 37 - 3ni - 4n 2

79
3. Limited Transpositions Sets

where n i , n^ are positive integers. To determine these integers, we need another argu-
ment. Let a be a limited transpositions set of 6 notes which has 5 transpositions. T h e
structure of a is of t h e form [a] = [ai,..., as] with a\ 4-... 4- CLQ = 30. Since a must have
5 transpositions, the only solution is a\ = ... = a^ = 5. Thus n\ = 1. To determine n^
we have to look at the limited transpositions sets with 6 notes and 10 transpositions.
Since they are 4 partitions of 10 of length 2, there are only 4 limited transpositions sets
and thus n^ = 1. T h e last equation gives t h e number of limited transpositions sets of 65
notes which have 15 transpositions, #6,15 = 30. For k = 10, the equation

9a;io,3 + 8zio,6 + 5xio,i5 = 1025


has several solutions
#io,3 = n i
#io,6 = 2ni + 5n 2
#10,15 = 205 - 5ni - 8n2
External arguments give ri\ 1, n2 = 0. For k = 12, the equation

5^12,5 + 4^12,10 + 3Xi2,15 = 1089


leads to
#12,5 = ^1
#12,10 = ni + 3n2
#12,15 = 363 - 3ni - 4n2
T h e examination of limited transpositions sets structures lead to ri\ = 2, n<i = 6. There
are 2 limited transpositions sets of 12 notes with 5 transpositions, 20 limited transposi-
tions sets with 10 transpositions and 333 limited transpositions sets with 15 transposi-
tions. For k = 15, t h e equation

7X15,2 + 6X15,6 + 5Xi5, 10 150

has solutions of t h e form


#15,2 = rti
#15,6 = 3ni + 5n2
#15,10 = 30 — 5ni — 6ri2
T h e solutions are obtained for n\ = 1 and n^ = 0. T h e numbers of limited transpositions
sets classes (Ltsc) and pcsets classes (Pcsc) for the 30-tet are given in the following table.

1 Notes Pcsc Ltsc |1 8,22 195 143 91


30 1 1 9,21 476 913 12
1,29 1 0 10, 20 1 001 603 203
2,28 15 1 11, 19 1 820 910 0
3,27 136 1 12, 18 2 883 289 355
4,26 917 7 13, 17 3 991 995 0
5,25 4 751 1 14, 16 4 847 637 429
6,24 19 811 35 15 5 170 604 29
7,23 67 860 o I | Total 35 792 567 2300

Examples. Lt-sets of ZyQ. For the 36-tone equal temperament (sixths of tone), we have
to solve several equations. For k = 4, the equation

3#4,9 + 2X4}i8 = 19

80
Enumerations

has solutions
f x 4)9 = 1 + 2ni
\ #4,18 = 8 - 3ni
As they are only one limited transpositions set of 4 notes with 9 transpositions, n\ = 0.
For k = 6, the equation
5x 6j6 + 4z 6) i2 + 3x 6) i 8 = 160
has several solutions
#6,6 = ^1
#6,12 = 1 + n i + 3 n 2
#6,18 = 52 - 3ni - 4n 2
Let a be a limited transpositions set of 6 notes which has 6 transpositions. The structure
of a is of the form [a] = [ai, ...,ae] with ai + ... + a$ = 36. Since a must have 6 trans-
positions, the only solution is a\ = ... = a^ = 6. Thus n\ = 1. In the same way, if a is a
limited transpositions set of 6 notes which has now 12 transpositions. The structure of
a must be of the form [A, A, A] with A = [ai, a2] and ai + a 2 = 12. The partition of 12
leads to 5 solutions [1, 11], [2, 10], [3, 9] [4, 8] and [5, 7]. Thus z 6 ,i 2 = 5 and n 2 = 1. To
sum up, we have one set with 6 transpositions, 5 sets with 12 transpositions and 45 sets
with 18 transpositions. For k = 8, the equation

3xS}9 + 2^8,18 = 348

has solutions of the form


f #8,9 = 2ni
1 #8,18 = 174 - 3ni
Let a be a limited transpositions set of 8 notes which has 9 transpositions. The structure
of a is of the form [a] = [ai, ...,as] with a\ + ... 4- &% = 36. The structure of a must be
of the form [A, A, A, A] with A = [ai,a 2 ] and ai + a 2 = 9. The partitions of 9 lead to 4
solutions. Thus n\ = 2. For k = 9, the equation

4#9,4 + 3x9 j i 2 = 58

has solution of the form


f xgi4 = 1 + 3ni
\ #9,12 = 18 - 4ni
where ni = 0. For k = 12, the equation
llxi 2 ,3 + 2zi 2)6 + 9zi 2 ) 9 + 4XI2JI2 + 6xi 2j i8 = 6588

has solutions

( #12,3 = ^1
#12,6 = 1*2
#12,9 = ni+ 2?23
#12,12 = 2ni + n 2 + 3n4
[ #12,18 = 1098 - 6ni - 3n 2 - 3n 3 - 4n 4
Let a be a limited transpositions set of 12 notes. The structure of a is of the form
[a] = [ai,...,ai 2 ] with a\ + ... + ai 2 = 36. If a has 3 transpositions, the only solution
is a,j = 3 for all j . Thus n\ = 1. If a has 6 transpositions, the structure of a is of the
form [A, A, A, A, A, A], where A = [ai,a 2 ] and a\ + a 2 = 6. There are two solutions

81
3. Limited Transpositions Sets

A = [1,5] and A = [2,4]. Thus n\ = 2. If a has 9 transpositions, the structure of a is of


the form [A, A, A, A], where A = [ai,a 2 ,a 3 ] and a\ + a 2 + a 3 = 9. The examination of
the partitions leads to 9 solutions. Thus n% = 4. If a has 12 transpositions, the structure
of a is of the form [A, Ay A], where A = [ai, a2, a3,04] and ai + <i2 + as + a4 = 12. The
examination of the partitions of 12 of length 3 leads to 40 solutions. The solutions are
m = 1, n<2 = 2, n% = 4 and n 4 = 12. There are 1 set of 3 transpositions, 2 sets of 6
transpositions, 9 sets of 9 transpositions, 40 sets of 12 transpositions and 1026 sets of 18
transpositions. For k = 16, the equation

3xi6,9 + 2z 16 ,i8 = 4890


has solutions
f zi 6 j 9 = 2ni
1 #16,18 = 2445 - 3ni
In this case, n\ = 7 and we get #16,18 = 2424 limited transpositions sets of 16 notes with
18 transpositions. For k = 18, the equation
17xi8j2 + 12xi8)i2 + 15xi8,6 + 16xi 8 , 4 + 9xi 8 ,i 8 = 25278
has several solutions
( Xi 8 ,2 = n i
#18,4 = m + 3n2
< #18,6 = ™3
#18,12 = 2 + ni + 2n 2 + n 3 + 3n 4
[ #18,18 = 2806 — 5ni - 8n2 - 3n3 — 4n 4
The solutions are obtained for m = 1, 712 = 0, 713 = 3 and n 4 = 23. Let a be a limited
transpositions set of 18 notes. The structure of a is of the form [a] = [a\,..., ais] with a\ +
... + ais = 36. If a has 2 transpositions, the only solution is a,j = 2 for all j . Thus ri\ = 1.
If a has 4 transpositions, the structure of a is of the form [A, A, A, A, A, A, A, A, A], where
A = [ai, a2] and a\ + a2 = 4. There is only one solution A = [1,3]. Thus ni = 0. If a has
6 transpositions, the structure of a is of the form [A, A, A, A, A, A], where A = [ai, a^ a^]
and ai + a 2 + a 3 = 6. There are 3 solutions A = [1,1,4], A = [1,4,1], and A = [1,2,3]..
Thus n3 = 3. If a has 12 transpositions, the structure of a is of the form [A, A], where
A = [ai,..., as] and ai + ... + CLQ = 12. The study of the partition of 12 shows that there
are 75 solutions. Thus n 4 = 23. In conclusion, the numbers of limited transpositions sets
classes (Ltsc) and pcsets classes (Pcsc) for the 36-tet are given in the following table.

I Notes Pcsc Ltsc Notes Pcsc Ltsc


36 1 1 10,26 7060984 476
1,35 1 0 11,25 16689036 0
2,34 18 1 12,24 34769374 1078
3,33 199 1 13,23 64188600 0
4,32 1641 9 14,22 105453584 1768
5,31 10472 0 15,21 154664004 66
6,30 54132 51 16,20 202997670 2438
7,29 231880 0 17,19 238819350 0
8,28 840652 172 18 252088496 2780
9,27 2615104 19 Total 1 908 881 899 14939

82
4
Neo-Riemannian Investigations

This chapter starts with a review of the main results of American Set Theory (AST). It
is not very easy to know when the AST was really born. Perhaps the most appropriate
introduction is to examine the papers of Milton Babbitt, who gave fruitful intuitions
since his paper on Set Structure as a Compositional Determinant published in 1961, and
other works on the studies of musical sets and partitions [Martino, 1961], [Howe, 1965],
[Gamer, 1967]. Set Theory was set out by Allen Forte who published The Structure of
Atonal Music in 1973. The pcsets classes were the first expression of the dihedral group
action on the twelve notes. In the 80's, Georg Perle's Serial Composition and Atonality
(1981) devoted to the atonal music of Schoenberg, Webern and Berg, and John Rahn's
Basic Atonal Theory (1980) become standard references in the field of contemporary
music theory. The group-theoretic properties of musical systems were pointed out by M.
Babbitt (1965) and G. Balzano (1980).
The Generalized Interval System (GIS) introduced by D. Lewin in musical theory
provides a new abstract construction. Its book Generalized Musical Intervals and Trans-
formation (1987) redefines the transformations introduced in its previous papers leading
to a neo-Riemannian theory arising in response to analytical problems posed by chro-
matic and extended tonal music. J. Clough, G. Myerson (1985) and E. Agmon (1989)
explore diatonic systems. John Clough and Jack Douthett investigate abstracts properties
of musical groups. J. Kovachi, D. Clampitt, R. Cohn, R. Peck and Ian Quinn also use
general mathematical tools to understand the conceptual transformations emphasized
by neo-Riemannian theory. Henry Klumpenhouver proposed new analytical investiga-
tions based on isographies and networks to make clear harmonic and melodic transitions
through contextual transformations, see for example [Klumpenhouwer, 1991], [Lewin,
1990], [Hascher, 2005], [Lambert, 2002] and [Stoeker, 2002].
4. NeoRiemannian Investigations

4.1 Set Theory


We review some standard concepts of American Set Theory by following Allen Forte's
approach [Forte, 1973].

Definition 97 A pitch class (pc) is one of the twelve notes designated by the integers
{0, 1, ..., 11} identifying with Z12. C is labeled by 0, Oft is labeled by 1, ..., and B is labeled
by 11. Sometimes, the pitch classes are coded with the hexadecimal representation by the
integers {0, 1, 2, .., 8, 9, a, b). The letters a and b stands for 11 and 12 respectively.
Enharmonic equivalents (such as Oj[ and Db) are coded with the same symbol.

Definition 98 A pitch class set fpcset,) is a subset 0/Z12, i>£- a set of pitch classes.
A serie (or tone row,) is an ordered pcset. A fc-serie (or a k-row^ is an ordered pcset
of cardinality k. A pcset is said to be in normal order if arranged in ascending or-
der, the set is put as compactly as possible by a cyclic permutation. More precisely,
let A = [i4o,^4i,...,Ak-i] be a pcset arranged in ascending order Ao < ... < Ak-\.
For each cyclic permutation a, the index vector is defined from the permuted pcset
a(A) = [AT(0), Ar(i),---,47(fc-i)] by the vector (uu...,uk-\) with

Uk-j-i = Aa(j) — Aa(0) mod 12

The normal order is obtained for a given permutation ao if the number built on the
coordinates of the index vector, called the compact number

N(a(A)) = Ul • lO*" 2 + u2 • 10 fc " 3 + ... + uk.2 • 10 + uk-X

is as small as possible. The index number Aao(o) — A$ is the transposition number in


semitones of the normal order above the reference pcset.

Example. The set [0,3,6,8,9] is not in normal order. Its normal order is [3,6,8,9,0]. The
compact numbers of the five cyclic permutations are: iV([0, 3, 6, 8, 9]) = 9863, iV([3, 6, 8,
9, 0]) = 9653, iV([6, 8, 9, 0, 3]) = 9632, AT([8,9,0,3,6]) = 10741, AT([9,0,3,6,8]) = 11963.

Definition 99 The dihedral group D\2 acts on the set of all pcsets. Two pcsets are equi-
valent (or D\2 -equivalent) if "they are reducible to the same prime form by transposition
or by inversion followed by transposition". Pc-sets classes (also called simply set classes)
are identified with the dihedral classification. The set is in prime form (dihedral or Forte's
prime form) if it is transposed so that the first integer is 0 and it is the most compact
form among its inversion.

Example. The set A = [2,3,5,7,11] is not in prime form. The normal order of A is
[11,2,3,5,7], and transposed one semitone up is [0,3,4,6,8] and AT([0, 3, 4, 6, 8]) = 8643.
The inversion of A, 1(A) = [1,5,7,9,10] is not in normal order. The normal order is
[5,7,9,10,1] and transposed a fourth down [0,2,4,5,8]. As N([0, 2, 4, 5, 8]) = 8542 < 7V([0,
3, 4, 6, 8]) then the prime form of A is [0,2,4,5,8]. The determination of the prime form
can be done easily with the circle representation. In this representation, the pitches are
placed on a circle as drawn below. Transpositions are obtained by rotating the circle.
The inversion is just a mirroring of the figure.

84
Set Theory

Definition 100 The set matrix of the prime form A = [AQ, A\, ...,Ak-i] is the k x k
matrix U defined by
Uij = Ai-i + Aj-i mod 12

Proposition 101 Let A be a pcset of length k, A= [Ao, A\,..., -Afc-i]. If any number m
appears k times in the set matrix U, or in other words, if there exists a number m such
that
k
= J2 kUij=m)

where lx is the counting function (lx = 1 if X is true and lx = 0 otherwise) then


Im(A) = TmI(A) has the same pitch as the original set.

Im(A) = TmI(A) = A

Definition 102 The comparison matrix of the pcset

A=[Ao9Au..nAk-i\

is the k x k matrix C defined by

Cij = sgn(Aj-i — Ai-i)

Proposition 103 The comparison matrix is skew-symmetric, i.e. CT = —C. The com-
parison matrix of the inverse set 1(A) is the transposition matrix of C.

C(I(A)) = CT

The comparison matrix of the retrograde form R(A)

C(R(A)) = CR

is the n-rotation of the matrix elements around its center CR with


=
Cij C!fe+i-t,fc+i-j

The comparison matrix of the retrograde-inversion RI(A) is the codiagonal transposition

C{RI{A)) = CRT = CTR

defined by
C ij = Ck+l-j,k+l-i

85
4. Neo-Riemannian Investigations

Example. Let A be the 4-serie BACH, that is the pcset A = [10,9,0,11]. The comparison
matrix is
/ 0 -1 -1 1 \
C(A) = 1 0 - 1 1
1 1 0 1
V - l -1 -1 0/
The inverse set 1(A) = [1,2,11,0] has a comparison matrix CT. The retrograde set
R(A) = [11,0,9,10] has a comparison of the form

( 0 -1 -1 - 1 \
1 0 1 1
C(R{A))
1 -1 0 1
V l -1 -1

and the retrograde-inverse RI(A) = [0,11,2,1] has the following comparison matrix

1
C(RI(A)) =

Thus C(RI(A)) = C(R(A))T. But remark that C(IR(A)) is different from C(RI(A)).
Definition 104 The complement Ac of a pcset A is the set of all elements of Zu not
contained in A, i.e. Ac = Zyi\A. The set (of six pitch class sets) is self-complement if
the set A and its complement Ac are reductible to the same prime form.

Example. The set A = [0, 1, 2, 4, 5, 8] and its complement Ac = [3, 6, 7, 9, 10, 11] are
reducible to the same prime form A. A is self-complement. The set B = [0,2,4,5,7,9] is
also self-complement. But the set C = [0,2,3,5,6,8] has a complement reducible to the
prime form Cc = [0,2,3,4,6,9] different from the prime form C.
Definition 105 The interval class (\c) of two pitch classes x,y is the mapping d :
Z12 x Z12 - • {0,1,2,..., 6} defined by

mod 12 if \x - y\ < 6
d(xyy)
y\ mod 12 if \x — y\ > 6

There are only seven interval classes or types of intervals (ascending or descending):
unison (0), semitone (1), tone or dyad (2), minor third (3), major third (4) fourth or
fifth (5) and tritone (6).

Definition 106 The derivation of a serie S = [So,..., Sk] of length k + 1 with So < ... <
Sk is the pcset D(S) = [D0,...Dk] defined by Dj = d(Sj,Sj+i) = Sj+i — Sj mod 12
for all j = 0,..., k — 1, and Dk = Sk — So- The iterated derivation is defined recursively
Dm(S).
Definition 107 The interval vector (iv) of a pcset A is a 6-tuple that represents the
number of each interval classes included in A. The first coordinate is the number of the
smallest interval (a semi-tone in the twelve tone music), the second is the number of
seconds, etc. until the number of tritones.

86
Set Theory

Example. The whole tone scale A = [0,2,4,6,8,10] has an interval vector iv(A) =
0600603. The Prometheus hexachord B = [0,1,3,5,7,9] has iv(B) = 142422 which
means that B is composed of 1 semitone, 4 tones, 2 minor thirds etc. Remark that,
sometimes, the interval vector is defined as a 12-tuple for the twelve interval classes.

Definition 108 The interval class content vector (icv) of two pcsets A and B is the
7-tuple icv(A, B) = [vo, •••^6] defined by v0 = \A D B\ and for i = 1,.., 6, Vi = \AC\kB\ /2
where \X\ is the cardinality of the set X and Adk B is the set of all pairs of interval
class k :
A fife B = {(x,y) GAxB, d{x,y) = k}

Definition 109 A Z-related pair is a pair of sets A, B with the same interval vector
iv(A) = iv{B), but not reducible to the same prime form.

Example. The sets A = [0,1,2,4,5,7] and B = [0,1,2,3,5,8] have the same interval
vector iv(A) = iv(B) = 333231, but have different prime form.

Definition 110 Two sets A and B of same cardinality m are in similarity relation Rp
if two representatives A and B of the pcsets classes of A and B contain at least one
common subset of cardinality m—1.

A% B ^ 3C \C\ = m-l, CcADB

Definition 111 Two sets A and B of same cardinality m are in similarity relation i?o
if they have no corresponding interval vector coordinates in common

A% B «=• Vt, iv(A)i^iv(B)i

Definition 112 Two sets A and B of same cardinality m are in similarity relation Ri
if they have four out of their six interval vector digits equivalent and the two dissimilar
digits are switched in position.

Definition 113 Two sets A and B of same cardinality m are in similarity relation R2
if they have four out of their six interval vector digits equivalent and the two dissimilar
digits are not equivalent.

Example. The sets A = [0,1,2,4,6] and B = [0,1,3,5,7] are in Rp relation, because


they contain the set C = [0,2,4,6] in common as C is of cardinality 4 and C C A
and C C Tn(B). The sets A = [0,1,2,6,7], with iv(A) = 310132 and B = [0,2,3,4,6],
with iv(B) = 232201 have no common digits in their interval vector and thus are in
Ro relation. The sets A = [0,1,2,3,5,7] with iv(A) = 342231 and B = [0,1,4,5,6,8]
with iv(B) = 322431 are in R1 relation. The sets A = [01346], iv(A) = 223111 and
B = [01356], iv(B) = 222121 are in R2 relation as iv(A)i = iv(B)i except for i = 3 and
5.

Definition 114 Two sets classes A, B of distinct cardinalities (2 < \A\ < 10, 2 < \B\ <
10,) are in subset relation if one is included in the other, that is

ADCB «=> A C B or B C A

87
4. Neo-Riemannian Investigations

Definition 115 The set complex of a set class A, denoted K(A, Ac) or simply K(A) is
the set of all set classes B in subset relation with A or Ac, in other words, the sets B
such that B contains A or is contained in A or B contains Ac or is contained in Ac.
BeK(A) ^=> BDCA or B DC Ac

A set used as a reference of a complex set is called a nexus set.


Example. The set complex of the set A = [0,3,6,9] is the set K(A) = {i?i,i?2, # 3 , £4}
with B1 = [0,3,6], B2 = [0,1,3,6,9], B3 = [0,1,3,4,6,9] and B4 = [0,1,3,6,7,9].
Definition 116 The set subcomplex Kh(A, Ac) (or simply Kh(A)) of a set class A is
the set of all set classes B in subset relation with A and with its complement Ac, that is
B contains A or is contained in A and B contains Ac or is contained in A c .
B G Kh{A) <=> BDCA and B DC Ac

Example. The subcomplex of the set A = [0,2,4,6,8,10] is the set Kh{A) = {i?i, £ 2 ,
.., B7} with Bx = [0,2,4], B2 = [0,2,6], Bz = [0,4,8], B4 = [0,2,4,6], B5 = [0,2,4,8],
S 6 = [0,2,6,8],B 7 = [0,2,4,6,8].

4.2 Generalized Interval Systems


The notion of Generalized Interval System (GIS) was introduced and explored by David
Lewin in [Lewin, 1987].
Definition 117 Let X be a finite set, G the (multiplicative) group of intervals on X, a
generalized interval system (GIS) is a triple (X, G, / ) where f : XxX —• G is a function
called the intervallic function, also denoted by int(x> y) verifying:
(1) f(x,y)f(y,z) = f(x,z) for all x,y,z£X
(2) for all x in X, and for all g inG there is only one value y € X such that / ( # , y) = g.
Let X, Y be two (non-null) pcsets. We denote lx the characteristic function of X,
i.e. lx(u) = 1 if u G X and lx(u) = 0 otherwise. We define the adjoint function of the
characteristic function by l^-(^) = l x ( — u ) .
Definition 118 The interval function h(xfY) or IFUNC(X, Y) of two pcsets X and Y
is the convolution function of the characteristic functions \*x * l y (z),

tKxtY)(!) = E VxU) Mi ~ 3) = E Mk) Mi + *)


3 k

in other words, the interval function counts how many times the note k of X has its
i-transpose in Y. The interval vector (also denoted by \X,Y)) is the 12-tuple whose
coordinates are the interval functions h(x,Y)(i) f°r i = 0,..., 11.
Example. The interval function of the sets X = {4,10} and Y = {1,5,9} is

The interval vector is the 12-tuple vector


fyx,y) = (0,1,0,1,0,1,0,1,0,1,0,1)

88
Generalized Interval Systems

Definition 119 Let X, Y be two non-null pcsets. The injection function INJ(X, Y)(f)
relatively to a transformation f is the number of elements x of X such that f(x) £Y

INJ(X,Y)(f)=y£l{fix)eY)
xex
Theorem 120 Let T{ be the translation ofi, X and Y two non-null pcsets, then

INJ{X,Y){Ti) = IFUNC{X,Y){i)

Proof

INJ{X,Y){Ti) = Y, HTMSY) = E UWKY)


xex xex
= £ l x ( f c ) . l y ( t + fc) = IFUNC(X,Y)(i)
k

D
C
Theorem 121 Let A be an hexachord \A\ = \A \. For all bijectionf, we have

INJ(A,A)(f) = INJ(Ac,Ac)(f)

and in particular
IFUNC(A,A) = IFUNC(AC,AC)
Proof This theorem is a consequence of the properties of the characteristic functions.
11
INJ(A,A)(f) = £ l(Hx)€A) = J2 lA(k) lf-HA)(k)
xeA k=0

= £(i-i A .(fc))(i-i/-W*))
k

= ^-E^w-E1/-1^)^)
k

= 12-6-6+ ^ l(f(x)€Ae)
c
xeA
= INJ(Ac,Ac)(f)
a
Theorem 122 Let A be an hexachord with \A\ = \AC\ = 6. For all bisection f, the
injection function verifies

INJ(A,A)(f) + INJ(A,Ac)(f) = \A\

In particular,
IFUNC(A, Ac) = 6 - IFUNC{A, A)

89
4. Neo-Riemannian Investigations

Proof

\A\ = 22 l(xeA) = ^2 1
(®Gi4)-l(/(a;)€i4uA«)
=
S1(/(*)eA) + ^^meA*)
xEA XEA

= INJ(A,A)(f) + INJ(A,Ac)(f)

Applying the same result to Ac yields

INJ(AC, Ac)(f) + INJ(Ac,A)(f) = \AC\ = \A\ = 6

and
IFUNC(AC, A) + IFUNC(AC, Ac) = 6


If we define the Fourier transform of the characteristic function by

then it is well known that the Fourier transform of the convolution product is the point-
wise product of their Fourier transforms. So the Fourier transform of the interval function
is the pointwise multiplication of the complex conjugate of F(\x) with «F(ly) that is

J T ( X , r ) = ^ ( b F ) .F(ly)

David Lewin [Lewin, 2001] proved the following results.

Theorem 123 The interval function of two non-null pcsets X and Y is constant if and
only if its Fourier transform Th{X, Y)(i) vanishes for i = 1,..., 11.

Lewin has shown that it is the case if the set S = X or S = Y satisfies the following
properties:

(FP1) S is the disjoint union of tritone sets (#i+i — Xi = 6) and (or) augmented triad
sets (#i + i — Xi = 4).

(FP2) | 5 n T 3 ( i 4 ) | = \SnA\ for all set A of the class A = {0,1,6,7} i.e. the number
of notes in common between A and S is the same as the number of notes in common
between S and the 3-transpose of A.

(FP3) \S D T6(B)\ = \S fl B\ for all augmented triad sets B.

(FP4) | S f l C | = | S n T i ( C ) | = | S n r 3 ( C ) | for all diminished seventh chord set C =


{0,3,6,9}.

(FP5) \S fl D\ = \S fl Ti(D)| where D is the whole tone scale

D = {0,2,4,6,8,10}

90
Riemannian Transformations

Theorem 124 The interval function of two non-null pcsets X and Y is alternate h(X, Y)
= (p>#>P><7>P><Z>P><Z>P>tf>P><z) if and only if its Fourier transform T'h{X,Y){i) vanishes
for all i different ofi = 0 and i = 6. In other words, if and only if either X or Y satisfies
the first four previous properties (FP1) to (FP4).

Theorem 125 Let X, Y, Z be three non-null pcsets. The interval function /i(X, Y) is
equal to h(X, Z) if and only if the Fourier transform of lx or the Fourier transform
of ly — l z vanishes everywhere. In other words, if and only if Y and Z have the same
cardinal number \Y\ = \Z\ and either X verifies the following (FPi) properties or Y and
Z verify the (FR*) properties (for i = 1,..., 5).

(FR1) \Y H J\ - \Y H T 6 (J)| = \Z H J\ - \Z n T 6 (J)\, for any dyad J.

(FR2) \Y fl A\ - \Y fl T3(A)\ = \ZnA\-\Zn T3(A)\, for any set A of the class A =


{0,1,6,7}.

(FR3) \Y D B\ - \Y D T6(B)\ = \ZHB\-\Zn T6(B)\, for any augmented triad sets B.

(FR4) \Y fl C\ - \Y H T 4 (C)| = \Z H C\ - \Z n T 4 (C)|, for all diminished seventh chord


set of the class C = {0,3,6,9}.

(FR5) \Y H D\ - \Y flTi(D)| = \ZnD\-\ZD TX{D)\, where £> is the whole tone scale
D = {0,2,4,6,8,10}.

4.3 Riemannian Transformations


Since the seminal work of Hugo Riemann and Arthur von Oettingen at the end of the 19 th
century, new analytical methods have been developed using contextual transformations,
called Parallel, Relative and Leittonwechsel transformations. These transformations are
involutions. They act on triads and maintain two points fixed. The Parallel transforma-
tion P exchanges major triads and minor triads.

[a, a + 4, b] <—> [a, a + 3, b]

Examples. P([0,4,7]) = [0,3,7] and P([5,8,0]) = [5,9,0].


The Relative transformation R exchanges a major triad with the relative minor triad,
that is the minor triad located a minor third down.

[a,6,6 + 3] 3-> [ a - 3 , a , b ]

Examples. i?([0,4,7]) = [9,0,4] and #([5,8,0]) = [8,0,3].


The Leittonwechsel transformation L exhanges a major third triad with the minor
triad located a major third up.

[a - 4, a, b] <—> [a, 6, b + 4]

Examples. £([0,4,7]) = [4,7,11] and L([5,8,0]) = [1,5,8].

91
4. Neo-Riemannian Investigations

The progression J- VI-IV- V-I is interpreted with Riemannian Transformations as


I _R^ VI _^ IV (RLf V LR^ I

The composition of these transformations has been defined by Lewin. The suhdominant
transformation S is the composition
S = RoL
(denoted by S = RL). S transforms a major triad to a major triad transposed a fifth
up, and a minor triad to a minor triad transposed a fifth down.
Examples. S([0,4,7]) = [7,11,2] and S([0,3,7]) = [5,8,0]. Remark that S is not an
involution.

The dominant transformation D is the composition


D= LoR
(or for short D = LR). D transforms a minor triad to a minor triad transposed a fifth
up, and a major triad to major triad transposed a fifth down.
Examples. £([0,4,7]) = [5,9,0] and £>([0,3,7]) = [7,10,2].

4.4 Isographies and K-nets


The group generated by the three Riemannian operations P, L, R is isomorphic to the
dihedral group of order 24, i.e. the group generated by the translations
Tn : x —> x-\-n mod 12
and the inversions
In : x —> — x + n mod 12
In his thesis, Henry Klumpenhouwer [Klumpenhouwer, 1991] observes that every pitch
collection can be interpreted as a network of T and J operations. This perspective allows
an interpretation of each chord as a network in which each pitch classes are linked by Tn
or In arrows. For example, the pentachords
A =[0,1,4,6,7], B = [0,4,6,7,10] and C = [0,1,6,7,10]
are built on two tritones. They are defined through a five nodes network that contains
two TQ operations and two inversions. Moreover, if we consider the two transformations
as fixed points, the operation [T3] acts on the inversions as [T3](/n) = In+3- So, the
three pentachords evolves in a Klumpenhouwer network (K-net for short) showing the
structural morphology and the logical progressions of chords.

7 •1

k J5L iw I» J 5 L I,
•+6

[l.o

92
Isographies and K-nets

Xavier Hascher [Hascher, 2002] has shown how the 4-chords of Franz Liszt's Ladilaus
Teleki are organized in isographies. Robert Peck [Peck, 2003] analyzed the opening canon
of no. 8, Nacht from Arnold Schoenberg's Pierrot Lunaire, op. 21. The trichords of the
first three measures display a triangular network of the same pcset 3-3 [0,1,4]. Moreover,
theses chords are also the vertices of a supernetwork, in which edges represent hyper-
operators [Tn] and [In\.
Definition 126 Hyper-operators between two K-nets are positive isographies if
(1) The two K-nets have the same disposition of T and I relations,
(2) The values j of the translations Tj are the same in the two networks,
(3) The values j of the inversions Ij in the second network minus the values i of the
correspondant U arrows in the first network are equal to m (with m > 1).

j—i=m mod 12

If m = 0, the graphs are equivalent and the two networks are said to be in strong
isography.
Definition 127 Hyper-opertors between two K-nets are negative isographies if
(1) The two K-nets have the same disposition of T and I relations,
(2) The values j of the translations Tj in the second network are the inverse of the
values i of the correspondant T{ arrows in the first network,
(3) The values j of the inversions Ij in the second network minus the inverse values i
of the correspondant Ii arrows in the first network are equal to m, with m > 1.

j — I(i) = m mod 12

The following definition of (n, m)-isographies generalizes the previous ones. For n = 1,
we recover the definition of positive isographies and for n = 11, we recover the definition
of negative isographies.
Definition 128 Hyper-opertors between two K-nets are (n,m)-isographies if
(1) The two K-nets have the same disposition of T and I relations,
(2) The values j of the translations Tj in the second network are the translations Tn
(with n € {1,2,.., \\}) of the values i of the correspondant T{ arrows in the first network

3=Tn(i)
(3) The values j of the inversions Ij in the second network minus the values Tn(i) of
the correspondant Ii arrows in the first network are equal to m, with m > 1.

j — Tn(i) = m mod 12

In the two following examples, we show that some graphs could preserve the same
structure, even if there are no inversions in these graphs. To avoid confusion, we replace
hyper-operators by t-operators which act on the notes of the isographies. In the Second
Sonate for piano of Nikolai Roslavetz, the same structure is kept. The sets include a
minor third and a fifth transposed a semitone up. The set is completed by a tritone,
which is drawn on the bottom of the graph by the translation TQ. The first measures
evolve by simple translations, denoted by t-operators [T3] or [T7], at the minor third or
at the fifth. These t-operators are composed of the same translations as the horizontal
operators (T3 and Tj) of the initial structure. Here are the structure of the first measures:

93
4. Neo-Riemannian Investigations
M«,4

8 « - ^ - 1 - ^ 4 | U^i-4^7 I

1,1,1- JSlJ I , I , I- JMJ ITU*


I-
2 4 - ^ - 7 - ^ 1 0

4 | l- 11 |
2 |

MM.S Met 6-7 Mas. 8

1«JL.6_!i*9

J1L U'J" Uu I , I , I- J^U I, 2 4 - 1 2 - 7 — ! & • 10 5<-k-10-Xl


I, I''
7<HJl- o—if,3

l- I-
4
I- 7 J
9
I
MM. 9 Met. 10 MM. 11-13

1 6 4-3L11-JJU2 I ,JL,_S., I
m I , I , I-
O4J1-5-X.8
-M. I, I, I- Uu I, I, I'' 2 4JL 7-2*10
I"
2
l- I- 9 4
I I I I
The Prelude opus 74 n° 4 of Scriabin is difficult to analyze using isographies. The first
measures are analyzed in the following nets:
A • Mode 3 transp. 2 mes. 1 £-Mode 3 transp 2mes 2 A-Mode 3 transp. 2

04-i-9—^1 I [«•!, M\ oJ-.Jk., I


I-
4 |
H
5 |
I" 4 I
T10(C) - Mode 3 transp. 4 T6(C)-Mode 3 transp 4 T9(C)-Mode 3 transp 3

tttl 94JLe^Uio IT.J, 5 4 - 1 - 2—=*8


JH. , _ i . , j k . , i
3 11
I"
2 |

A- Mode 3 transp. 3 16(A) • Mode 3 transp 1 • me » 4 T 10(A) -Mode 3 transp. 1

1o*JL7.Ji>11 I
[4.tl IV, ja
I"

!'•
4 10
2 |

B - Mod* 3 transp. 2 1rH(B)-Mode3transp 2 - m ss 5 T10(B)-Mode3transp 4

[JiT,ol [T„l p lU

1
I''
0
1"
11 |
Td<B) • Mode 3 transp. 4 B - Mode 3 transp. 2 A-Mode 3 transp 2

I 8 4-Il-5J!l*o,
IT„] g | 5 «JL.2-iU6 [T,l, jJY^

l"
10 |_
1"
1 |
I" 4 |

94
Parsimonious Graphs

This Prelude is written in the Messiaen's mode 3. This mode has four transpositions.
The first measures begin with the second transposition of the mode. Three types of
chords are used: the chord A {0, 1, 4, 9}, the chord B {1, 5, 8, 9} and the chord C
{0, 5, 8, 11}. Each chord has a common horizontal structure composed of one minor
third and one major third (denoted by T3 and T 4 ).

The t-operators have two parts. A part in which the notes are simply transposed and
denoted by [Tn]. And a second part which modifies only the vertical components of the
graph, denoted by [Jn]. This part transforms the vertical arrows of translation T; by
=
Jn\J-i) J-i+n-

For example, the passage of the first graph to the second transforms a chord i to a
chord C. The action of the translation [Tn] translates the notes {0, 9, 1} of the chord
A to the notes {11, 8, 0}. The second part of the t-operator [J2] transforms the vertical
arrow T3 into the translation T3+2 = T5. Consequently, the note {4} of the chord A is
first translated by Tn to give the note {3}, and then modified by [J2] to give the note
3 + 2 = 5 of the chord C.

The t-operator [JmTn] acts on the notes of the horizontal arrows by the translation
[Tn] and on the notes of the vertical arrows by the translation [T n + m ].

4.5 Parsimonious Graphs


The concept of parsimony (the law of the shortest way) was first introduced by Richard
Cohn [Cohn, 1966]. It has been extended by Adrian Childs [Childs, 1998] and generalized
by Jack Douthett and Peter Steinbach [Douthett, 1998]. In this section, we follow this
last paper.

Definition 129 Let X and Y be two pcsets of same cardinality. The pcsets X and Y
are Pij related if, in the notes that are not common to both X and Y, i notes move by
interval class 1, and j notes move by interval class 2. More precisely, if the non-common
tones are denoted X\Y = {x\,X2> ...,£i+j}, there exists a bisection f : X —» V verifying

m
t( \ _ ( ^fc i 1 ° d 12 if 1 < k < i
^Xk'~{ xk±2 mod 12 if i+l<k<i+j

This definition covers the definition of Cohn (Pi,o and Po,i-stated) and the definition
of Childs (P2,o-related). The contextual Riemannian transformations P and L are P\}Q-
related and the R transformation is i^i-related. In the Chicken-Wire Torus represented
in the following figure, the edges represents the contextual transformations P, L, R. The
figure depicts the parsimony of triads in the sense of Richard Cohn.

95
4. Neo-Riemannian Investigations

The hexatonic mode of limited transposition [0,3,4,7,8,11] is organized in six triads (C,
Cm, E, Em, Ab and Abm) which defined the hexatonic cycle. These triads themselves
are P^o-related. They are placed at the vertices of a cube. The two remaining vertices
of the cube are occupied by two augmented triads, which have the greatest potential to
couple hexatonic cycles via P^o-relations. The graph obtained is called the Cube Dance.

Another method induced by P^o and Po,i-relations is depicted by a graph called Power
Towers. We invite the reader to refer to the paper of Douthett and Steinbach [Douthett,
1998].

96
Permutational Transformations

4.6 Permutational Transformations


The purpose of this section is to show the power of permutations for theoretical and
analytical purpose. First, we redefine the neo-Riemannian contextual transformations by
pointwise transformations that globally exchange the set of major triads into the set of
minor triads. Each transformation again denoted by P, L, R (Parallel, Leittonwechsel and
Relative) is an involution (P2 = L2 = R2 = 1). The main advantage of the permutational
method is that A;-chords can be used with permutations. We will see also that the P, L,
R transformations generate the dihedral group. See also Robert Morris [Morris, 1982]
and Ilflmaki Tuukka [Ilomaki, 2005].
The Parallel transformation P is defined by the permutation

/ 0 1 2 3 4 5 6 7 8 9 10 11 \
^ 7 6 5 4 3 2 1 0 11 10 9 8 j

or in cycle notation
P = (0,7)(l,6)(2 ) 5)(3 ) 4)(8 ) ll)(9,10)
It transforms a major triad based on the pitch n into a minor triad based on the pitch
(12 — n) mod 12. This transformation is the inversion h(x) = — x + 7 mod 12.
The Relative transformation R defined by

/ 0 1 2 3 4 5 6 7 8 9 10 11 \
^ " ^ 4 3 2 1 0 1110 9 8 7 6 b )

and in cycle notation

fi = (0,4)(l ) 3)(5,ll)(6 ( 10)(7 ) 9)(2)(8)

It exchanges a major triad based on the pitch n with a minor triad based on the pitch
(9 — n) mod 12. This transformation is the inversion U(x) = — x + 4 mod 12.
The Leittonwechsel transformation L is defined by

/0 1 2 3 4 5 6 7 8 9 10 11 \
\ 11 10 9 8 7 6 5 4 3 2 1 0 j
and in cycle notation

L = (0,ll)(l,10)(2 ) 9)(3 ) 8)(4,7)(5,6)

transforms a major triad based on the pitch n into a minor triad based on the pitch
(4 — n) mod 12. This transformation is the inversion I\\{x) = — x + 11 mod 12.
The new transformations lead to three important groups: the hexatonic, octatonic and
the PLR or dihedral group.
The Hexatonic group is the group generated by the transformations P and L. It has
the presentation:

Gi = (P,L | P2 = L2 = 1, PLP = LPL)


Let TQ be the identity of the group, the group G\ is composed of six elements

G\ = {To, 77,^4, ^3,^8, i n }

97
4. Neo-Riemannian Investigations

where In is the inversion of order n (x —> —x + n mod 12) and Tn is the transposition
of order n (x —> x + n mod 12). The group is represented by an hexagon. The vertices
are the elements of the group and the edges are alternatively the transformations P and
L. Let A be a pcset of k notes, the graph of G\ represents a pitch class, given by the
action of the group G\ on A. For example, the atonal triad A = [0,6,11] gives five others
triads P(A) = [7,8,1], LP(A) = [4,3,10], PLP(A) = [3,4,9], (LP)2(A) = [8,7,2] and
P(LP)2(A) = [11,0,5]. Remark that each triad is the same pitch class set (3-5 in the
Forte's classification).
The Octatonic group is generated by the two permutations P and R.

G2 = {P,R\P2 = R2 = 1, (RP)2 = (PR)2)

The group has eight elements. With the previous notation of transpositions and inver-
sions, we get
<?2 = {To,/7,T9,/io,T6,/i,r3,/4}
The graph is an octagon, whose edges are alternatively the transformations P and R.
As in the case of the hexatonic group, one vertex of the graph accepts one pcset, which
is propagated through other vertices via the P and R permutations.
The group PLR is generated by the three permutations. It has the presentation

r _ I P , L , R | P2 = L2 = R2 = (PL) 3 \
u
- \ = (PRL)2 = (PR)4 = 1 /

Substituting P, L, R by the inversions J7, 1^ 7n, one verifies that G is the dihedral
group. The elements of G are drawn on the following figure.
-x+6

The interior octogone is rotated by 37r/4 and identified with the exterior octogone. The
figure is then a torus on which three octagons are drawn in the meridian plane and four
hexagons, based on the succession PLPLPLP, lie on the transversal planes of the torus.

98
Permutational Transformations

Because of the planar representation, these hexagons are not easily seen. Starting from
a point situated on the interior octogone, the path PLPLPLP leads to a point located
on the exterior octogone which is exactly the same as the starting point.
As G is the dihedral group, the figure is a tesselation of the transpositions and inver-
sions of a pitch-class set. In the tonal music context, one chord, for example a major triad
[C, E, G] (but it could also be a chord of four notes or more), is placed on a vertex of the
figure and propagated with the P, L, R transformations. Prom the figure obtained, the
consonant triads are connected with the transformation LR, when the base pitch moves
a fifth up.

LR LR LR LR
D E B F#

Example. We understand why isographies could easily be drawn with this representation.
But this representation does not concern only consonant triads, and our scheme can be
filled with any pcset, for example, with the atonal triad [0, 1, 6]. In the piano piece of
George Crumb entitled Gargoyles (Makrocosmos vol. 77), the right hand moves

PR
[2,3,8] - ^ - + [5,6,11] - ^ - + [0,1,6]
PL
[3,4,9] [7,8,1] - ^ t [1,2,7]

and the left hand moves in a same way

LR PR
[1,0,7] - ^ - > [10,3,4] [11,10,5]
PL (RP)2
[1,2,8] [0,5,6] [6,11,0]
The same transformations are used in both hands, except in the last triads of our example,
in which the reverse transformation is used (PRPR on the right hand, and RPRP on the
left hand). But if we allow hands crossings, the P-transform brings the set [7, 8, 1] to the
set [6, 11, 0] and [0, 5, 6] to [1, 2, 7]. In the last chord, the triads are played together.

Example. Thomas Noll has studied afRne transformations Ma$ : x —» ax + fi mod 12. He
has shown that some of them generate monoids. The following graph depicts a monoid
based on two afRne mappings u and v. On this picture, a, b and c are constants, and the
functions u and v verify the relations u3 = u, v3 = v and u2v = vu2.

-• u

If u and v are of the form f(x) = ax + (3 mod 12 and satisfy f(A) C A for the major
chord A = {0,4,7}, then the functions u and v are 0, 4, 7, x, 3a; + 7, 4a:, 8a; + 4 or 9a; + 4.
If we choose for example ^(a;) = 3a; + 7 and v(x) = 8x + 4, then the monoid has three

99
4. Neo-Riemannian Investigations

constants a = 7, b = 0 and c = 4. In the same way, if we want to keep the minor triad
unchanged A = {0,3,7} the functions u and v are 0, 3, 7, a;, 3a; + 3, 4a; + 3, 8a; + 7 and
9a;. The choice u(x) = 3x + 3 and v(a;) = 8a; + 7 leads to the constants a = 0, b = 7 et
c = 3.
But others monoids are possible. Hexacycles can be found using the affine transform-
ations
fa (x) = 7x + 1 h2(x) = 7x + 4
/^(z) = 7a;+ 7 /i 4 (z) = 7a; + 10
In the following figure, the choice of three indices z, j , k G {1,2,34} leads to the hexacycle.

h,(x+4) h,(x)

k
*

x+4 x+8

K
h/x+8)

Each transformation is associated with a permutation. The transformation hi(x) is


associated with the permutation
2 3 4 5 6 7 8 9 10 11
hi
\1 8 3 10 5 0 7 2 9 4 11 6
If we choose i = 1, j = 2 and k = 4, the seventh chord [7,11, 2, 5, 8] is transformed in
five others chords, which are shown in the following figure.

7,11,2,5,8
7x+10^ .Jx+1

3,7,4,1,10 2,6,3,0,9

7x+10l 7x+1

11,3,6,9,0 3,7,10,1,4

7x+4 7x+4
1,5,2,11,8

100
Permutational Transformations

Example. The mode 2 of Messiaen (C, CJ|, E\>, E, F$, G, A, B\>). which is a limited
transpositions mode, has only three transpositions. Trying to find affine transformations,
we obtained two permutations closely related to mode 2. The following permutation

/ 0 1 2 3 4 5 6 7 8 9 10 11 \
a
\^16 1 1 4 9 2 7 0 5 10 3 8 J

or in the cycle notation

a =(0,1,6,7X2,11,8,5X3,4,9,10)

is the affine transformation x —» 5a; + 1 mod 12 and the permutation b

/ 0 1 2 3 4 5 6 7 8 9 10 11 \
1 ^ 3 8 1 6 11 4 9 2 7 0 5 10 )

given in the cycle notation by

a=(0,3,6,9)(l,8,7,2)(4,ll,10,5)

is the affine transformation x —• 5x + 3 mod 12. The group H is generated by the two
permutations and has the presentation

TT _ I a> b | a 4 = ( a _ 1 6 a - 2 ) 3 = 1, aab = baa, \


H
~ \ bab~la? = ab /

The group has only twelve elements and appears to be well adapted to the three trans-
positions of mode 2. Consider a characteristic chord of mode 2 (C, E, G, Fjj), that is
a chord that only belongs to one of the three transpositions, the group H acts on this
chord and the action produces twelve chords. Remark that the major triad is included
in this chord.

In the previous figure, there are only four chords of mode 2 1 (first transposition), four
chords of mode 2 2 (second transposition), and four chords of mode 2 3 (third transposi-
tion). The group facilitates the transpositions inside mode 2. It is clear that it could not

101
4. Neo-Riemannian Investigations

account for modulations towards limited transpositions modes, as a modulation from


mode 2 to mode 6. Another group must be found to get suitable isographies, but the
group H is well adpated for music written in mode 2.

Example. He de feu 2 is a piece for piano composed in 1949-50 with three other pieces in
Quatre etudes de rythme. It is based on various inversions of two series of twelve notes,
which is interpreted by two permutations. The first permutation is given below

0 1 2 3 4 5 6 7 8 9 10 11 \
6 5 7 4 8 3 9 2 10 1 11 0 J
or expressed in the cycle notation

a =(0,6,9,1,5,3,4,8,10,11X2,7)

The second permutation is

/ 0 1 2 3 4 5 6 7 8 9 10 1l\
^ 5 6 4 7 3 8 2 9 1 1 0 0 11/

or in the cycle notation

6 =(0,5,8,1,6,2,4,3,7,9,10)(11)

These two permutations a and b generate the Mathieu group Mi 2. It has the following
presentation.

njr _/ a,b\a10 = bu = {ab~1)2 = aba-^a^b"1


Ml2 2 l 2
- \ = bab- a- b ab-*a = {a?babf = 1

This group has 95040 elements and, of course, only some of them are used by Messiaen.
However, it induces a real potential of combinations, which maybe charges the music of
inner perspectives.

4.7 Morris Groups


The permutations which exchange notes in a given interval has been studied by Robert
Morris and classified in several types. We study in this section the groups generated by
the elements of the cyclic or dihedral group and the following multiplication

Ma : x —» a.x mod 12

Only the values a = 5, 7 and 11 correspond to some permutations. For a = 5, the


application M 5 defines the permutation

_ / 0 1 2 3 4 5 6 7 8 9 10 11 \
5
V 0 5 10 3 8 1 6 11 4 9 2 7 J

The notes {0, 3, 6, 9} are fixed points. The others are exchanged in four cycles

MB = (1,5)(2,10)(4,8)(7,11)

102
Morris Groups

In the same manner, the multiplication by 7 induces the permutation

_ / 0 1 2 3 4 5 6 7 8 9 10 11 \
7
" \ 0 7 2 M 11 6 1 8 3 10 5 J

which let the notes {0, 2, 4, 6, 8, 10} fixed and exchanges the others in three cycles

M 7 = (1,7X3,9X5,11)

The multiplication by 11 is the inversion, a well-known element of the dihedral group

/ = (1,11)(2,10)(3,9)(4,8)(5,7)

The permutation associated with the first transposition is

T1 = (0,1,2,3,4,5,6,7,8,9,10,11)
The group generated by the transpositions and the multiplication by 5 or 7 has 24
elements
\(TUM<>)\ = \(T1,M7)\ = 2A
If we add the inversion, the group has 48 elements

1 ( ^ , 7 ^ 5 ) 1 = 1 ( ^ , 7 ^ 7 ) 1 = 48

In the alpha familly, Morris considers the permutations that exchange the neighbor notes.
The prototype of this familly is the permutation

a = (0,1)(2,3)(4,5)(6,7)(8,9)(10,11)
The group generated by the permutations a, Ti, M5, M7 and J defines some new trans-
formations which can act on pcsets or on twelve-tone series. The smallest group is the
group
Ga = (Tua)
generated by the transpositions T\ and the permutation a of neighbor notes. It has 72
elements and has the presentation

Ga = (Tua\a2 = T^aT^a = Ta12 = (T^af = l)

The largest group is the group (T\, a, / , M) where M is the multiplication by 5 or by 7.


This group has 144 elements and corresponds to the action of transpositions, inversion,
retrogradations, exchanges of first neighbors and multiplication. It is a generalization of
the classical dodecaphonic transformations to a wide set of operators.

In the beta familly, Morris considers the permutations of 2-cycles which exchange the
notes distant of one tone. The prototype of this familly who has 8 permutations is the
beta permutation
/3 = (0,2)(1,3)(4,6)(5,7)(8,10)(9,11)
The smallest group generated by this permutation (Ti,/J) has 108 elements and the
largest (Ti,/3, J,M 5 ) has 432 elements.

103
4. Neo-Riemannian Investigations

In the gamma familly, Morris considers the permutations of 2-cycles which exchange
the notes having the distance equal to one minor third. The prototype of this famille
who has 8 permutations is the element

7 = (0 > 3)(l,4)(2,5)(6,9)(7,10)(8,ll)

The smallest group generated by this permutation (Ti, 7) has 96 elements and the biggest
(Ti,7,7,M 5 ) has 192 elements.

In the delta familly, Morris considers the permutations which exchange the notes dis-
tant of one major third. The prototype is

5 = (4,8)(5,9)(6,10)(7,11)

The group (Ti,<$,M5) has 648 elements and the group (Ti,<S,/,M) has 1296 elements.
The following table gives the order of the group indicated in the first column when x is
one of the permutations a, /?, 7 or S.

G a P 7 5 1
(Tx,x) 72 108 96 648
(Ti,I,x) 144 216 192 1296
(T!,M5,x) 144 216 192 648
(TuM7,x) 72 216 96 1296
(TuM5,I,x) 144 432 192 1296
{TltM7tI,x) 144 432 192 1296

The same study has been done with the permutations which exchange the notes of a
limited transpositions set and leave the others unchanged. One can show that the group
(Ti,M 5 ,a;) extends the possibility of schoenbergian transformations, since the biggest
group has 46080 elements. The following table gives the order of the group (Ti,Ms,x)
for each limited transpositions set.

| Names X \{TuM5,x)\
| Tritone (0,6) 768
Aug. Fifth. (0, 4, 8) 648
| Dim. Seventh (0, 3, 6, 9) 384
| Alt. Seventh (0, 2, 6, 8) 2304
Double Fifths (0, 1 , 6 , 7 ) 23040
| Mode 1 (0, 2, 4, 6, 8, 10) 144
Mode 2 (0, 1, 3, 4, 6, 7, 9, 10) 384
Mode 3 (0, 1, 2, 4, 5, 6, 8, 9, 10) 3888
| Mode 4 (0, 1, 2, 3, 6, 7, 8, 9) 23040
| Mode 5 (0, 1, 2, 6, 7, 8) 46080
Mode 6 (0, 1, 2, 4, 6, 7, 8, 10) 3840
Mode 7 (0, 1, 2, 3, 4, 6, 7, 8, 9, 10) 46080
Mode A (0, 1, 4, 5, 8, 9) 3888
Mode B (0, 1, 3, 6, 7, 9) 46080
Mode C (0, 1, 4, 6, 7, 10) 46080
Chromatic Scale (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11) 24 |

104
5
Knots and Braids

This chapter shows three applications of knots theory to the musical field. The first ap-
plication deals with dodecaphonic series, which can be classified with Gauss diagrams and
so leads to a better understanding of their structural meanings. The second application
concerns tuning and temperaments and shows how to build a new temperament from a
knot. The last application is more general as it could be used in any text. It suggests
how to highlight knots in literary or musical texts, and shows new topological structures
of low dimension.

5.1 Classification of Knots


The theory of knots has today several applications both in mathematics and in physics
but also in biology where proteins structures are modelized by knots. In mathematics,
the main research is to determine knots invariants which can describe knots and links
without any ambiguity. Applications of these invariants concern quantum groups as well
as manifolds. In Physics, these invariants could be useful for the description of Feynman
Diagrams or others diagrammatic methods used in Statistical Physics concerning the
Yang Baxter Equation. This chapter suggests a new application of the theory of knots
to the musical field.
Amongst the objects that you can make in tieing a cord or laces, there are knots
built with only one cord, and links that are made with several laces. A knot is a thin
tangled rope in space whose ends are glued together. From a mathematical point of view,
a knot is a smooth embedding of the circle S1 in E 3 . Historically, the knots appear in
the book of J.B. Listing, a student of C.F. Gauss, who publishes in 1848 the drawing of
a knot on the cover of his book Vorstudien zur Topologie. The first list of knots has been
published by P.G. Tait and C. Little in 1877, and the concept of equivalence of two knots
has been formalized by Reidmeister in 1932. Knots in space are represented by planar
5. Knots and Braids

diagrams. Two knots are equivalent (ambient isotonic) in space if their planar diagrams
are equivalent under the three Reidmeister moves.
One defines the number of crossings c(K) of a knot K as the minimal number of
crossing points in the planar representation of the knot K. A crossing of an oriented
knot is said to be positive or right (resp. negative or left) if somebody moving on the
higher branch sees the lower cord passing from right-hand side to the left (resp. from the
left-hand side to the right). The index of a right crossing is +1 and the index of a left
crossing is —1. For a knot K, the linking number is the sum of all the index crossings,
and for a link K with Kn components (strands), the linking number is the half of the
sum of the linking number of each component

lk{K) = ±5>0K„)
n

One way of classifying knots is to consider the minimal number of crossings. The sum of
two oriented knots is obtained by removing a small arc from each knot and to glue the
four endpoints respecting the orientation of each knot. The unknot O is the untangled
circle S1.

Definition 130 A knot K, different from the unknot, is called a prime knot if the sum
K = K\ + if2 implies that K\ or K<i is the unknot.

Knots are designated by their crossing number and an order number. The knot 3i
is the trefoil knot, the knot 4i the eight knot, etc. Prime knots have been tabulating
until 16 crossings. The first work was done by Thomas K. Kirkman in the ISSO's. The
classification was studied by C. N. Little and reviewed by Kurt Reidmeister up to 9
crossings in 1932. John Conway, Hugh Dowker, Jim Hoste and Morwen Thistlethwaite
contributed to the classification which is now known up to 16 crossings. The following
table gives for each number of crossings (first column) the number of prime knots (second
column).
Cross. Primes || Cross. Primes
3 1 10 165
4 1 11 552
5 2 12 2176
6 3 13 9988
7 7 14 46972
8 21 1516 253293
9 49 |1 1388705

To characterize knots, mathematicians try to find new invariants. These invariants are
numbers (like the linking number), polynomials or groups. For a general introduction,
see [Adams, 1994], [Lickorish, 1997], [Sossinsky, 1997], etc.

5.2 Dodecaphonic Knots


In this section, we will show that the dodecaphonic series are classified according to
Gauss diagrams (also called chord diagrams) of singular knots with six double points. In

106
Dodecaphonic Knots

the first paragraph, we show how to build the Gauss diagram of a twelve-tone row. Let
us choose a series, for example

C, A\>, Gb, G, D, E, A, B, D\>, F, E\>, B\>

According to the identification of Z12 with the twelve notes, this series could also be
written as
0,8,6,7,2,4,9,11,1,5,3,10
Put the notes of the series on a circle
G
Gb
" \ 8 Ab

\
0 C
\ /
*°Bb
Bit
V Eb

Db

And join the tritones together by a chord

Rub out notes. What is left is an abstract structure called a Gauss or a chord diagram.

It is easy to verify that the 48 derived forms of the 12-tone row are mapped on the same
Gauss diagram. That is why Gauss diagrams are used to classify dodecaphonic series.
All the transpositions of the series have the same chord diagram (you have eventually to
rotate the diagram). The retrogradation of the series is placed on the diagram which is
the mirror symmetry of the original diagram. The inversion of the series uses the same

107
5. Knots and Braids

chord diagram and the retrogradation of the inversion is mapped by the mirror diagram.
All the 48 derived serial forms use the same chord diagram (viewed in the space, possibly
mirroring or rotating).

5.3 Gauss Diagrams


Let us show now that Gauss diagrams of twelve-tone rows are related to singular knots
with six double points. A double point is a point where the two segments of the rope are
glued. As we have seen, the Gauss diagram of order six is the directed circle on which
one placed six couples of points connected by chords. To build a Gauss diagram starting
from a singular knot with six double points, it is enough to fix an orientation of the
knot, to number the intersections of double points and to copy these numbers on the
circle of the Gauss diagram. Moving around the knot, we number all the intersections of
double points that we meet. Each double point of the knot receives a couple of numbers
describing the intertwining of the knot. In the following example describing a knot with
four crossings and six double points, starting from the flag and walking through the knot,
we mark each double point encountered by the numbers 0, 1, 2, 3, etc.

The couples are (0, 2), (1, 4), (3, 8), (5, 11), (6, 10) and (7,9). Put the numbers on the
circle from 0 to 11 and join the numbers of the couples by a chord, we obtain the Gauss
diagram associated with the singular knot

108
Gauss Diagrams

Remark that in general several knots are associated with the same Gauss diagram
(change for example one of the undercrossings to an overcrossing). In the same way,
several twelve-tone rows are associated with the same Gauss diagram. By labelling each
chord by a letter and by reading the word starting from 0 one obtains the following
diagram

which represents the Gauss word


abacbdefcfed
The mathematical question that arises now is: how many chord diagrams of dodecaphonic
knots are there ? We know that there are 12! twelve-tone series. If we consider that the
48 serial derived forms are equivalent, the number of twelve-tone rows is less than 12!.
The following result is shown in [Reiner, 1985].
Theorem 131 Let S be the set of classes of tone rows equivalent under transposition,
retrograde, and inversion. Then
\S\ =9,985,920
Proof Let X be the set of tone rows. The cardinal of X is \X\ = 12!. Let R be the
retrograde transformation, and let T and J be the transposition and the inversion. The
problem of counting the tone row of S is equivalent to count the number of orbits of the
group G = Z>24 x %2 generated by the transformations T, R and J, on the set X. D24
denotes the dihedral group. The Burnside's lemma says that the number of orbits is the
average number of fixed points, namely

1 ]
geG

Since G has 48 elements, we have to determine the fixed points for each element. But
the elements Tn (1 < n < 11), TnR (n ^ 6), T n J , TnIR (for n even) have no fixed
points and the identity has 12! fixed points. It remains to look at seven elements of G,
the operation T6R and TnIR for n odd. The fixed tone rows by the operation T6R are
the tone rows in which the six last entries are the reverse of the six first transposed by
a tritone. So there are
£ = 1 2 x l 0 x 8 x 6 x 4 x 2 = 2 6 -6!
ways of doing this. The same can be done for the transformation TnIR: the first six
entries determine the number of fixed tone rows. So they are also 2 6 • 6! series, for each
n odd. Finally there are
12! + 7 • 2 6 • 6!

109
5. Knots and Braids

fixed tone rows. Thus we get | 5 | = (12! - t)/48 = 9985920 tone rows classes.
More generally, for the n-tet, we have the following result.

Theorem 132 In the n-tone equal temperament (n > 3), there are exactly n\ series of
notes. Under the equivalence of the derived forms (i. e. under the action of the dihedral
group), there are

f I(n-l)! + 2(^)/2I( f + l)! if n is even


\ J(n-l)! + 2<n-5>/2(2=i)! if n is odd
series ofn notes.
The proof is a direct application of the de Bruijn extension of the P6lya enumeration
therorem.
Theorem 133 Under the action of the cyclic group C2n, the number of equivalent dia-
grams is
Cn =
2^£p(*)"n(0
i\2n
f
where ip(i) is the Euler s function and un is defined for all the divisors of2n by

t n /*(2n/t - 1)!! ifi is odd


vn(i) = { LTJI / 2 n / t \
ik(2k-l)U ifi is even

Under the action of the dihedral group, the number of Gauss diagrams in the 2n-tet is
given by

with
L»J
n\
E k\(n-2k)\
k=0
where \x\ indicates the greatest integer less than or equal to x.
The proof of this theorem is given in the paper of Khruzin [Khruzin, 2000].
The following table gives the number of Gauss diagrams (dn) for the 2n tone equal
temperaments. For the usual equal temperament (n = 6), there are exactly 554 Gauss
diagrams which classify dodecaphonic series with respect to their structural properties.
n Cn dn Temp.
3 5 5 6-tet
4 18 17 8-tet
5 105 79 10-tet
6 902 554 12-tet
7 9749 5283 14-tet
8 127072 65346 16-tet
9 1915951 966156 18-tet
10 32743182 16411700 20-tet
11 625002933 312702217 22-tet

110
Gauss Diagrams

A complete catalogue of the 554 Gauss diagrams has been set up. The diagrams are
classified by the Gauss word order. Each of the six chords are labelled by the numbers 1,
2,.., 6 and the Gauss word is a permutation of 112233445566. For two Gauss diagrams D
and L, D < L if and only if w(D) < w(L) where w(D) is the Gauss word of D. We can
also write the Gauss word in a form slightly different by locating the chords of different
types: the chord a connects two consecutive points of the Gauss diagram, b connects
the two extreme points of three consecutive points of the circle, and so on, until the
type / which connects the two extreme points of seven consecutive points of the circle.
Moreover, one uses the opposite (e.g. e _ 1 is a chord linking the two extreme points of
eight consecutive points). For example, the series of Jean Barraque Au-deld du hasard
(0, 8, 7, 1, 4, 2, 10, 3, 11, 5, 6, 9) is represented by the word

X = aba1-1 dab

and corresponds to the diagram D m in the classification of the Gauss diagrams. The
structural vector of a Gauss diagram is made of six components which count the number
of chords of each type a, 6, c, d, e, / . For example, the structural vector of D m is 220200.
This means that this diagram includes two chords of the type a, two chords of the type
6, none of the type c, two chords of the type d and none of the types e and / . In the
classification, the first Gauss diagram D\ is used by B.A. Zimmermann in his opera Die
Soldaten.

Di X = a6
Gauss Word 112233445566
Structural Vector 600000
Permutation
(0 1) (2 3) (4 5) (6 7) (8 9) (10 11)
B.A. Zimmermann, Die Soldaten, Acte I

Another example is the following diagram D349 which is used by Karel Goeyvaerts in
his Sonata for two pianos.

^349 X = afd~xe2a
Gauss Word 112345662453
Structural vector 200121
Permutation
(0 1) (2 8) (3 11) (4 9) (5 10) (6 7)
Karel Goeyvaerts, Sonata for two pianos.

The symmetric diagram D358 has been used by Anton Webern in his Symphony, opus

m
21.

Z/35g X = ac~1e~1eca
Gauss Word 112345665432
Structural Vector 202020
Permutation
(0 1) (2 11) (3 10) (4 9) (5 8) (6 7)
A. Webern, Symphony, opus 21

in
5. Knots and Braids

The last diagram of the dodecaphonic classification is used by B.A. Zimmerman in the
Interludes of his opera Die Soldaten.

£>554 X= f
Gauss Word 123456123456
Structural Vector 000006
Permutation
(0 6) (1 7) (2 8) (3 9) (4 10) (5 11)
B.A. Zimmerman, Interludes.
Ecce Ancilla Domini of Jean Etienne Marie is built on a series of 24 notes in the
space of quarter-tones (F %+, Ftf, G, B, C\+, A, D+, C, F, G|t, A$+, Ctf, B+, E, G$+,
Aft, G+, D%, F+, E-h, A-h, D, D$+, C+). The chord diagrams in this space can be
represented by a Chinese diagram linking two Gauss diagrams.

5.4 All-Interval Series


All-interval series have been studied by Andre Riotte [Riotte, 1962] and Herbert Eimert
[Eimert, 1964] who have published a complete list of the 1928 all-interval series. These
series are classified under 63 Gauss diagrams. They have been used for example by Alban
Berg in Suite lyrique {Dz$%)

5,4,0,9,7,2,8,1,3,6,10,11
by Luigi Nono in Canto Sospeso (D^s)
9,10,8,11,7,0,6,1,5,2,4,3
by Ernst Krenek in the Studies in Counterpoint (Di3$)
3,6,1,7,0,2,11,10,8,4,5,9
by Andre Riotte in many pieces and in particular in Multiple for string quartet composed
in 1962 and in his Inventions for piano solo (D50)
0,1,5,2,7,6,8,11,9,3,10,4
and by Milton Babbitt in Three Compositions for piano (£353)
10,3,5,2,0,1,7,11,6,9,8,4

112
Serial Algebra

The Gauss diagram of these series includes two (and only two) chords of type a, because
the first and the last note of the series is a tritone. Remark that all the type of chords
are not necessarily present in an all-interval series. For example, in the Gauss diagram
D14 which is denoted by X = a2bc2b, there are no chords of type d, e and / .

5.5 Serial Algebra


The classification of the twelve-tone rows proposed by Edmond Costere [Costere, 1962]
and formalized by Bernard Parzysz [Parzysz, 1983], is a classification based on the divi-
sion of the series S in two parts. The set of the six first notes of the series is denoted A
and the set of the six last notes is denoted Ac.

Definition 134 A transpose series is a series in which there is a transposition from the
set A to the set Ac. If the set of the transpose series is denoted by T, we have

SeT&3i Ti(A) = Ac

Examples. The twelve-tone row of Arnold Schoenberg's Sirinade opus 24


(4,2,3,11,0,1,8,6,9,5,7,10) which has the Gauss diagram D^oi and the series of B.A.
Zimmermann's Die Soldaten (9, 5, 6, 1, 4, 2, 8, 10, 7, 0, 11, 3) which has the Gauss
diagram -D358 are transpose series, because in each case, the transposition of the set A
gives the set Ac (T6(A) = Ac).

Definition 135 A reverse series is a series for which it exists a transposition from the
set 1(A) = —A to the set Ac. If we note TZ the set of reverse series, we have

SeTl^3i Ti(-A) = AC

Examples. The twelve-tone row of Luigi Dallapicolla's II Prigioniero


(8, 11, 2, 7, 5, 10, 9, 3, 4, 0, 1, 6) which has the Gauss diagram D238 is a reverse
series because we have Tn(—A) = Ac. The series of the Structures for two pianos by
Pierre Boulez (3, 2, 9, 8, 7, 6, 4, 1, 0, 10, 5, 11) is also a reverse serie, because we have
T7(-A) = Ac.

Definition 136 A semi-transpose series is a series for which it exists a transposition


different from identity which sends the set A to itself Let T' be the set of semi-transpose
series, we have
SeT'&3i^0, Ti(A) = A

Examples. The series of Schoenberg's Ode to Napoleon (4, 5, 1, 0, 8, 9, 11, 10, 2, 3, 7, 6)


which has the Gauss diagram DQSS is a semi-transpose serie, because T±(A) = A

Definition 137 A semi-reverse series is a series for which it exists a non-identity trans-
position which sends the set —A to the set A. If TV denotes the set of semi-reverse series,
we have
S e TV <# 3 t ^ 0, Ti(-A) = A

Examples. The series of Luigi Nono's Canto Sospeso (9, 10, 8, 11, 7, 0, 6, 1, 5, 2, 4, 3)
which has a Gauss diagram D358 is a semi-reverse serie, because Tj(—A) = A. The series

113
5. Knots and Braids

A of Jean Barraque's Concerto (7, 4, 9, 0, 6, 1, 2, 10, 5, 11, 3, 8) which has the Gauss
diagram D271 is a semi-reverse serie.

Many series do not belong to the four previous types. For example, the series of Pierre
Boulez's Sonatine for flute and piano (0, 11, 7, 1, 8, 4, 3, 9, 2, 10, 5, 6), the Cycle de
Partisan furieux of Le Marteau sans Maitre (3, 1, 4, 5, 9, 6, 10, 7, 8, 11, 2, 0), Karel
Goeyvaerts's Opus 2 for 13 instruments (11, 6, 5, 4, 7, 8, 3, 2, 9,10,1, 0), Jacques Lenot's
Third Sonate for piano (5, 8, 7, 2, 6, 0, 3, 9, 10, 1, 4, 11), Peter Schat's Polonaise opus
2 (6, 8, 9, 11, 2, 10, 0, 1, 3, 4, 5, 7), Jan van Vlijmen's Costruzione per due pianoforti
(0, 1, 5, 4, 7, 3, 2, 9, 8, 10, 6, 11) not belong to the set T U V U Tl U W. The following
figure shows the different relationships between the serial sets.

R Reverse series 31 % (144)


Webern Op.28

Schoenberg Op. 31 D_244 F Semi-transpose "


Berg Suite lyrique y.innu muiiin Die Soldateii d1
Schoon. Op. 41 D538 I T Transpose series
\ T J\
%m Webern Op.24 / U
B A Zimmermann Die Soldaten,
1%(0) . Arte I sc. 1 D 358,
Schoen. Valse Op. 23 D_407 Acte I sc. 2, D_544
_ ^ - - " 4 % (18)
Schoen. Op. 37 D_193
Stockhousen Klav. IX
Schoen. Op, 26 D_414 3% (12)
Schoen. Op. 24 D„301

18 % (84)
A. Schoenberg Op. 30

J. Barraque Concorto J. Barraque, Le temps restltue


B. Ferneyhough. Superscript D_447
Berio Senqenza V D. 256
A. Webern Op 19.23.26
J. Barraque Sonate
J. Barraque. au-dela du hasard

K. Stockhausen. Licht A. Berg Concerto pour violon

R' Senii-revGfse Series

Remark that the classification of Costere-Parsysz works with non well-ordered sets
and does not preserve the structural properties of the series. This classification is fully
compatible with the Gauss diagrams classification.

5.6 Combinatoriality
The combinatoriality is a notion first introduced by Milton Babbitt. Since combinatori-
ality applies to any set, we only consider the combinatoriality of hexachords in order to
compare the two classifications of Costere and Babbitt. The basic transformations and
their associated rows are denoted by P (basic row), I (inversion), R (retrograde) and RI
(retrograde of the inversion).

Definition 138 A semi-combinatorial row is a twelve-tone row such that any of its
transformations other than the retrograde can be transposed so that the first six notes are
equivalent to the last six notes of the original set, not necessarily in the same order.

114
Combinatoriality

Example. The twelve-tone row


S= [6,4,8,2,0,3, 9,5,1,10,11,7]
is a semi-combinatorial set, because in the first transposition of the inversion
TiJ(S) = [7,9,5,11,1,10, 4,8,0,3,2,6]
the set of the last six notes is equivalent to the set of the first six notes of the original
set S. Remark that any set is combinatorial with its own retrograde.
The combinatoriality of rows has been first considered by Ernst Krenek. In [Krenek,
1940], Krenek studied the "symmetric row"
S= [9,1,11,2,10,0, 8,4,6,3,7,5]
composed of the first hexachord A and the transposition of the inversion
S = AuTn(I(A))
Proposition 139 The basic transformations denoted by P (basic row), I (inversion),
R (retrograde) and RI (retrograde of the inversion) is the Klein group.
The multiplicative table is given by
P I R RI
p P I R RI
I I P RI R
R R RI P I
RI RI R I P
Definition 140 The all-combinatorial row is a row whose first hexachord forms a twelve-
tone row with any of its basic transformations (P, I, R, RI) transposed.
Proposition 141 The second hexachord of all-combinatorial rows is always a literal
transposition (Ti with i^0) of the first hexachord.
Example. The chromatic row
S =[0,1,2,3,4,5,6,7,8,9,10,11]
is formed of the first hexachord A = [0,1,2,3,4,5] whose transposition a tritone higher
leads to the second hexachord.
S = AUT6(A)
T h e o r e m 142 Let A be the first hexachord of a twelve-tone row S. The all-combinatorial
rows S are classified in six types of unordered sets and gathered in four orders.
(1) First order combines A with only one transposition TQ
S = AUT6(A)

A is one of the three unordered sets


A = {0,1,2,3,4,5}
A = {0,2,3,4,5,7}
A = {0,2,4,5,7,9}

115
5. Knots and Braids

(2) Second order combines A with two transpositions T3 and TQ

S = AL\T3(A) = AuT9(A)

A is the unordered set


A = {0,1,2,6,7,8}
(3) Third order combines A with three transpositions T<i, T$, Tio

S = AUT2(A) = AUT6(A) = Al)T10(A)

A is the unordered set


A = {0,1,4,5,8,9}
(4) Fourth order combines A with six transpositions T\f T3, T 5 , T7, Tg, T\\

S = AL\T1(A) = AUT3(A) = AUT5(A)


= AL\T7(A) = AuT9(A) = AuTn(A)

A is the unordered set


A= {0,2,4,6,8,10}
Theorem 143 Any hexachord A and its complementary set Ac have the same interval
vector.
Example. In the twelve-tone row of the Schoenberg's Fourth String Quartet

S =[0,11,7,8,3,1,2,10,6,5,4,9]

the two hexachords have the same interval vector

Iv(S) = [3,2,2,4,3,1]

5.7 Markov Chains


In this section, we show that Markov chains can be used to define a distance between
series. Markov chains on finite or countable number of states E are models for determining
the unfolding of musical events through time. The movement of a Markov chain Xn is
governed by its transition probability matrix

Pij=¥(Xn = j\Xn.1=i)

The Markovian property is expressed by the equation

¥(Xn = k\X1= xi,..., X n _ ! = x„_i) = F(Xn = k I X n _ ! = a; n _i)

which means that the Markov chain doesn't take into account the whole history of the
process but depends only on the last state (at time n — 1). Under suitable conditions, it
exists a unique stationary distribution of the chain A satisfying the condition \P = A.
More precisely for all states, we have

VyeE, Y/X(x)P(x,y) = X(y)

116
Markov Chains

In this case, for all function f:E-+R A—integrable, we have

n J
k=o

and the entropy of the chain is given by

S = - J2XiPi* lo
&Pii

Markov chains have been used by Iannis Xenakis for example in his piece Analogique B.
The composer defines some Markovian random processes according to range of pitch
classes, intensity and density. He uses two transition matrices

/ 1/5
^ 4/5
4/5
1/5 ) <K'")
The entropies of the two matrices are quite similar: S(P) = 0.707 and S(Q) = 0.722.
In Archorripsis, Iannis Xenakis used a discrete distribution of independent events.
On the finite set (#i, #2, > •••> xn) the Poisson distribution depending on the parameter A
attributes the probability

Archorripsis is divided in 28 sets of measures and 7 sets of instruments providing a matrix


of 7 rows and 28 columns. Each element of this matrix supports from 0 to 4 independent
events. The expectation of the discrete distribution is given by

EX- Y2*i PC* = xi)


i=l

One shows that the mean and the variance of the Poisson distribution is equal to the
A parameter. In Archorripsis, the unfolding of the pitch classes, speeds and durations
are also organized according to the Poisson distribution. For each musical parameter,
random points are controlled by the parameter A. The matrix used by Xenakis is given
in the following figure. The number of events of an element of the matrix is proportional
to the darkness of each cell. Dark cells represent 4 events by measure, and white cells
represent no event in the measure.

Time

Markovian analysis could also be used to measure some gaps in serial compositions.
For example, consider the transitions matrices of Anton Webern's opus 17, n° 1 defined
as follows. The theoretical transition matrix Qij represents the transitions from the pitch

117
5. Knots and Braids

class j to the pitch class i in the 48 forms of the dodecaphonic serie. Write the original
series and its 11 transpositions. Do the same for the retrogradation series, the inverse
and the retrogradation of the inverse: we obtain 48 series of 12 tones. Now count the
transitions from j to i and divide by the total number of transitions, we obtain the
probability transition Qij which is a 12x12 matrix.

Qi Q* \
Q
= Y4 Q 2 Qi J

with
/0 0 1 \
7 7 2
0 0 2
Qi = 2 7 0
2 0 7
\1 2 0/
and
/0 2 1 0 7 \
1 1 0 2 0
2 0 1 2 2
Q% = 2 1 2 1 2
0 2
2 1 0 1
V7 0 2 2 1 0 /

The transition matrix Py represents in the score the empirical probability of transition
from the pitch class j to the pitch class i, that is the number of j followed by pitch class
i divided by the total number of transitions.

p= Pl Pi\
\ * Pj
with
°
I 3/283/17
0
7/34
2/7
3/34 0 1/17 \
1/14 0 1/14
0 0 0 1/26 1/13 2/13
Pi = 2/27 2/27 1/9 0 5/27 1/27
6/25 3/25 1/25 2/25 0 1/25
V 1/26 0 1/26 3/26 2/13 o J
The submatrix Pi is

( 1/17 1/17 3/34 1/34 2/17 2/17 \


1/14 1/14 1/14 1/28 3/28 3/28
5/26 1/26 1/26 0 3/13 /13
P2 = 0 1/9 4/27 5/27 2/27 0
0 1/25 6/25 1/5 0 0
V 5/26 4/13 1/26 0 1/26 1/13 /

118
Groups and Diagrams

The submatrix P3 is

/ 1/30 1/10 0 2/15 1/10 1/15 \


1/10 2/15 1/15 1/10 1/10 0
3/29 2/29 3/29 3/29 5/29 2/29
P3 5/33 4/33 5/33 5/33 2/33 2/33
1/25 2/25 1/25 1/25 1/25 6/25
\ 0 1/13 1/13 1/13 0 5/26 /

and the submatrix P4 is

°
( 1/10 1/6
0
7/30 1/30 1/10
1/6 1/6 1/30
1/30 \
1/30
1/29 2/29 0 6/29 0 2/29
P4 =
1/33 1/33 2/33 0 2/33 4/33
2/25 1/25 3/25 3/25 0 4/25
^ 7/26 1/13 0 0 3/13 0 /
Using the 2-norm of a matrix A, defined as the supremum of the eigenvalues of A.A1
we find for the Webern's piece
||P - Q | | 2 « 0.65
The 2-norm between P and Q defines the discrepancy of a serial composition, a kind of
distance between the musical piece and the raw materials. It is another method to map
the serial pieces.

5.8 Groups and Diagrams


In this section, we consider permutations associated with chord diagrams in order to
measure the generating power of these diagrams. Each diagram represents a set of series.
But the symmetries limit the number of series associated with one diagram. We define a
group whose order represents this generating power.
Each diagram is associated with one permutation. For each chord diagram, we number
the vertices of the diagram clockwise, and consider the pairs of numbers corresponding to
linked vertices. The set of the pairs defines the permutation of the diagram. For example,
the diagram D358 which is the diagram of the series of Anton Webern's Symphony opus
21, Luigi Nono's Canto Sospeso, Alban Berg's Suite lyrique, and Karlheinz Stockhausen's
Klavierstuck IX.

119
5. Knots and Braids

The labelling of the vertices yields to the pairs

(0,1)(2,11)(3,10)(4,9)(5,8)(6,7)

These pairs define the permutation P358 associated with the diagram Ds^s-

/ O l 2 3 4 5 6 7 8 9 10 11 \
358
^ 1 0 1110 9 8 7 6 5 4 3 2 )

We propose the following definition.


Definition 144 Let Dn be a Gauss diagram. We denote Pn the associated permutation,
T\ the transposition of a semitone up and I the inversion. The group of the Gauss diagram
Dn is the group generated by the three operators

Gn = (P n ,Ti,7)

The order of this group is called the generating power of the diagram Dn.
Examples. 1) The group associated with the diagram D358 is the group

G358 = CP358»?i,I)

The GAP software gives the presentation of this group in terms of generators and rela-
tions
G358 = (a, 6, c I a 2 = b12 = c2 = ab^c = l )
The group has
|G 3 5 8 | = 24
elements. This group and the dihedral group have the same number of elements.
2) The series D554 used by B.A. Zimmermann in his opera Die Soldaten is also used
by Pierre Boulez in the Cycle des bourreaux de solitude (Piece 6) of Le Marteau sans
Maitre.
2,3,4,5,6,7,8,9,10,11,0,1
The associated group

G554 = (a, 6, c I a 2 = c2 = abab~l = (ac)2 = (b^c)2 = b6a = l )

has 24 elements.

3) The series used by Anton Webern in his Concerto for nine instruments, opus 24

11,10,2,3,7,6,8,4,5,0,1,9

is associated with the diagram £549 whose group


n _/ a,b,c\a2 = c2 = {b~lc)2 = (acb)2
G
*49 - \ = (ab2c)2 = (afe"1)4 = (abab-1)3 = 1
is of order 192.
4) The series used in Schoenberg's Ode to Napoleon and Kees van Baaren's Variations
in Isometrical Series
5,10,6,9,7,8,2,1,3,0,4,11

120
Tuning Braids

is associated with the diagram -D538. The associated group generated by the three oper-
ators
G538 = (Pb38,Ti,I)
has 7680 elements.
5) The series used by Dallapiccola in Quademo Musicale di Annalibera
10,11,3,6,8,2,7,1,5,9,0,4
is associated with the diagram D$Q. The group generated by the three operators

has the presentation

n _/ a,b,c\a2 = c2 = (b-lc)2 = (ac)4 = acb~\abc)2 \


56
° - \ = (acb*)2 = (ab2ab-2)2 = b12 = (acab~2)4 /
It is a group of order 28800.
6) Many diagrams have a maximum of generating power. It is the case, for example,
of the series D244 used in Schoenberg's Variations for Orchestra, of the series of diagram
DIG used in Jacques Lenot's Third Sonate for piano, and of the series of diagram .D353
used in Milton Babbitt's Three Compositions for piano. In each case, the associated
group is isomorphic to the permutation group Si2 of order

1^121 = 4 7 9 001 600

5.9 Tuning Braids


A braid is a set of n strings, each string joining the n points A\, ...,An of the top bar
with Ai = (i, 1) to the points Bi,..Bn of the bottom bar Bi = (iy 0). The set is a tangle
if one string starting from a bar does not intersect the other bar but intersects the same
bar. The closure of the braid is obtained by gluing together the top and the bottom bar.
Braids are described by a word listing the crossings from top to the bottom. The nth
string crossing over the (n + 1 ) string is denoted by an and the nth string crossing under
the (n + 1) string is denoted by the inverse cr"1.
1 2 i i+1 n

I I X I
1 2 i i+1 n

I I X I
The set Bn of braids with n strands has a natural structure of group, called the Artin
group. The product of two braids a and b is the braid ab obtained by putting the two
braids end to end. The inverse of a braid b is a mirror image of b.

121
5. Knots and Braids

Theorem 145 (Artin) The braid group Bn is isomorphic to the group whose generators
are G\, ...,cr n _i and verifying the following relations

GiGj1 = O^Oi = 1
aiaj=o-jai if\j-i\>2 for ij = 1,2, ...,n - 1
OiGi+\Oi = <7i + i<7i<7i + i % = 1,2,..., n — 2

These relations can be used to verify that two braids are equivalent. For example, the
braid a^1 (c^o^c^cr^"1 is equivalent to cr2

Some braids are used to describe tuning and temperaments. Meantone temperaments
are based on two different elementary intervals u\ and cr2. The first interval G\ is the
distance between the first two notes (C-CJJ) and the second interval cr2 is the distance
between the second notes (C(t-D). In the meantone temperaments, two consecutive notes
are separated by either o\ or <J2. The first interval concerns the notes (C-Cfl), (Eb-E),
(F-F(t), (G-GjJ) and (Bb-B). The other intervals are of type cr2. Meantone temperaments
are described by the word

X = ((Ji<J 2 (j 2 cri(7 2 )cricr 2 (aicr 2 cr 2 (7icr 2 )

The first interval is


37 ^ 8 1 ^ - 7 "
2 11 V80;

and the second interval is


5n
<»-£(2\
3 vsoy s

where n is a parameter: n = 1/4 in the classical meantone of Pietro Aaron (1523),


n = 1/10 (Lanfranco, 1533), n = 2/7 (G. Zarlino, 1558), n = 1/3 (F. Salinas, 1577),
n = 2/9 (Lemme Rossi, 1666), n = 1/5 (J. Sauveur, 1710), n = 1/6 (G. Silbermann,
1748), n = 5/18 (R. Smith, 1749) and n = 3/10 (J.B. Romieu, 1758) see for example
[Jedrzejewski, 2002]. The closure of the meantone braid gives a link with two components.

Another example is the Marpurg H temperament built on two basic intervals denoted
a and b: a is equal to 96 cents and b to 102 cents. The structure of the temperament
repeats (abb) four times.

(abb)(abb)(abb)(abb)

This can be drawn with a braid, each crossing materializing an interval.

122
Tuning Braids

a
P"^^c 1 1
b
b

^^^^.J.. a
h.J
b

a
,, J G
b.
b

IsTS a
b ••'• Bb

1 1 b f'" B
L_ fi
In the same way used by Harry Partch, one can build new temperaments. Let's choose
a knot of twelve crossings and a law of crossing. For example, just multiply the interval
of the crossed branch. To have a consistent law of crossing, the intervals must verify the
two closure relations

c 2 = ab2

and

a2bcs = 1

The knot is represented by

By choosing different values for a, b and c verifying the closure relations, one can do the
computation of all intervals modulo the octave, possibly by multiplying or dividing by a
power of 2 in order to get each interval in the same octave. For example, for a = 16/9,
b = (20) 1 / 4 /a and c = 56/4 we get a = 773 cents, 6 = 524 cents and c = 789 cents. This
new temperament is represented in the following figure.

123
5. Knots and Braids

5.10 Textual Analysis


As it was done by Ferdinand De Saussure with his anagrams, we are looking for low
dimensional topological structures in a given text. For example, consider the following
contemporary text

[ [ black shirt ;
red noise
(( black towel ]
; red) shadow
red ; red

One can ask to find all the word white. Each occurrence will be denoted by A, the
occurrence of the inverse word black will be denoted by small a. The same can be done
for other colors: the occurrences of the word red are denoted by B, and the occurrences
of its inverse blue are denoted by small 6. The word obtained is

aBaBBB

One can check if this word has a corresponding braid or not. The word aBaBBB cor-
responds to the knot 62. In this way, we hope to find many hidden braids in a given text
and we suppose that this will show topological relationships. All parameters are involved.
If we look at the punctuation marks, and if we assign the bracket [ to the letter A, the
inverse bracket ] to the letter small a, the sign ; to the letter B and the parenthesis ( t o
C and its inverse ) to the letter c, we find the word

AABCCaBcB

which represents the knot 74. Other occurrences could form a knot such as lexical units
or classes, isotopic relations, semantic units, as well as contextual representations. In a
musical text, in the same way, one could choose a pattern A and its inverse a, for example
the retrograde notes or rhythms, or other parameters. One can choose also other patterns
B, C, etc. and try to find a word that corresponds to a braid. This method can be used
to find invariant knots. For example, in the evolution between languages in linguistics
or in musicology, we could find a knot or a set of knots based on structures of a given

124
Textual Analysis

language L\ and the same knots in a language Li which is more recent than Li, This
could be an argument to maintain that the language L2 has relationships with L\. In
that case, the relation is supposed to be relevant and through the analysis of many texts
leads to some new categorical classifications.

125
6
Theory of Temperaments

The tuning of musical instruments has been widely studied in the world by all civiliz-
ations. The disciples of Pythagoras, Aristoxenus, Ptolemy and others have set out the
Greek tuning theories. The Pythagorean system based on the natural fifth and on the
octave was used in Europe in the fifteenth century. Later, musicians tried to complete the
cycle of fifths. But it is well known that there is no integer solution to this problem. Many
approximation solutions appear, consisting of tempering the fifths, that is diminishing
the ratio of the natural fifth to adjust successive fifths in an entire number of octaves.
Other musicians and theorists gave up with the idea of diminishing the fifths and tried
with meantone temperaments to set a tuning system with natural thirds. Others with
just intonation tried to use small and whole numbered ratios, and others tried to establish
a coherent set of ratios using irregular divisions. As there are great implications from the
esthetical point of view, the problem has passed through the centuries and is still active.

6.1 Twelve Tone Temperaments


For the twelve tone systems, the ratios of the historical temperaments, which based the
reduction of the fifths on fractions of the syntonic comma or Pythagorean comma, can
be established as follows. Let k be the ratio of the syntonic comma (K)
81 _ 3 4
80 24 • 5
and s the ratio of the schisma (S)
_38-5
S
~ 2 15
The ratio of the Pythagorean comma (P) is denoted by
3l2
t
6. Theory of Temperaments

In the twelve tone temperament, every fifth is tempered by a fraction a^ of syntonic


comma and an amount bi of the schisma. If a* = 6^, then the fifths are tempered by a
fraction a* of the Pythagorean comma. The difference between the fifth of the tempera-
ment and the pure fifth (3/2) is denoted by
AFi K + hS
In the same way, the difference between the major third of the temperament and the
pure major third (5/4) is denoted by A T M and the difference between the minor third
of the temperament and the pure minor third (6/5) is denoted by AT m

1 N. Fr6q. AF ATM ATm


0 1 aiK + biS (l + s4)K + r4S s9K+(l+r9)S
1 S72-liks7sr7 a8K + b8S (l+s^K + r^S -(l + sl)K-rlS
2 322-3fcS2sr2 a3K + b3S (l + s«)K + r°S s?K+ ( 1 + ^ ) 5
3 392-14fcS9sr9 a±oK + bioS -s%K-{l+rl)S -(l + s^K-r^S
4 342-6fc34sr4 a5K + b5S (l + s*)K + r*S -(l + s*)K-r*S
5 3 1 1 2- 1 7 fc 5 i 1 s r i 1 0>V2.K + &12<S -s^K- (1+^)5 -(l+sii)tf-riiS
6 362-9fc56sHj a7K + b7S (l+s™)K + r™S -(l + s64)K-r%S
7 3.2-1kai8ri a2K + b2S (l + s*)K + ri>S sl°K+(l+r™)S
8 S82~12ks»sr» a9K + b9S -s8K-(l+r8)S -(l + sl)X-r«5
9 332"4Aj535r3 aAK + b4S (l + sl)K + rlS -(l + s3)K-r3S
10 3 2 A; 5 ios r io
10 -15
auK + buS -si°K-(l+rl<>)S -(l+sl°)K-r™S
11 3 5 2- 7 & s 5s r5 a6K + b6S (l + sl)K + r*S -(l+s*)K-4S \

The coefficients verify the relations (in what follows, all sums are considered modulo 12)
12 12

i=l i=l
The partial sums are defined by

Sa = ^CLi Sn = S7l = ^CLi


i=l

and

r!t = I > rn=rn1=Y)>i


i=\
To characterize a temperament by a number, James Murray Barbour [Barbour, 1951]
has proposed to consider the mean deviation

^ = 1^2 E, | i + ¥ ^
where a is defined by
1200 37
a= In
ln(2) V213/12
that is about 14 cents.

128
Pythagorean Tunings

6.2 Pythagorean Tunings


Pythagorean tunings are based on the cycle of fifths (which is in fact a spiral). To built
a Pythagorean system of n notes, we must write down all the natural fifths considered
(....jlbbb, £bbb, £bbb, F\>\>, Cbb; ...., Ebb, £bb, F\>, C\>, ..., B, F%, CH, ... , 4tf, £fl, B$,
Fx, Cx, ...) with their ratios (Fbb, 2 2 4 /3 1 5 , 271 cents) Starting from one note, say C of
ratio 1, we cover the spiral in the ascending way. The ratios are multiplied by 3/2 and,
eventually divided by a power of 2, to keep the note in the octave [1,2[: C (1), G (3/2,
702 cents), D (3 2 /2 3 , 204 cents), A (3 3 /2 4 , 906 cents), etc. In the descending way, the
ratios are divided by 3/2 and eventually multiplied by a power of 2: C (1), F (4/3, 498
cents), B\> (2 4 /3 2 , 996 cents), E\> (2 5 /3 3 , 294 cents), etc. We choose n consecutive fifths,
rewrite the ratios in an octave and put them in the increasing order leading to the n tone
Pythagorean system. For example, the twelve tones Pythagorean system is displayed in
the following table. All the fifths are pure, except the wolf fifth (B-F§). There are two
kind of intervals: the leimma £ of 90 cents (256/243 = 2 8 /3 5 ) and the apotome a of 114
cents.

| Notes Ratios Interv. Cents AF ATM ATm


C 1 £ 0 0 K -K
Db 256/243 a 90 0 K S
D 9/8 £ 204 0 -S -K
Eb 32/27 a 294 0 K -K
E 81/64 £ 408 0 -S -K
F 4/3 £ 498 0 K -K
Gb 1024/729 P 588 0 K S
Ffl 729/512 £ 612 0 -S -K
G 3/2 £ 702 0 K -K
Ab 128/81 a 792 0 K S
A 27/16 £ 906 0 -S -K
Bb 16/9 a 996 0 K -K
B 243/128 £ 1110 -P -S -K \

If all the fifths are tempered by a fraction ai of Pythagorean comma (i.e. if all 6^ = 0),
the table of the general temperament of twelve tone can be written as follows.

| Freq. AF ATM ATm


1 aiP (1 + S4)K + s4S S9K + (1 + S9)S
| 372-np57 a8P (l+s^K + s^S -(l + sl)K-slS
| 3 2 2~V 2 a3P (l + s*)K + s*S s?K + (l + s?)S
| 3 9 2 -1 V }9 aioP -s*K-(l + s°)S -(l+s97)K-s97S
3 4 2~V 4 a5P (l + s*)K + s*S - ( 1 + s%)K - s*S
3 ll2-7p5ii ai2P -s^K-il + si^S -(l+s^)K-s?S
3 6 2"V 6 a7P (l+s^K + s^S -(1 + S6)K_S65
3.2-V 1 a2P (l + sS)K + s*S sl0K + (l + slQ)S
3 8 2" 1 V 8 agP -s8K - (1 + s8)S -(1 + 4)K-4S
3 3 2"V 3 aiP (1 + s\)K + s\S -(l + s3)K-s3S
3 102-15 p sio auP - 4 ° i t - ( l + S i°)S -(l+sl°)K-slQS
3 5 2~V 5 a6P (l + s96)K + slS -(1+4)K-8*S \

129
6. Theory of Temperaments

6.3 Equal Temperaments


The twelve tone equal temperament (12-tet) is the system used in Western countries. It
appears in the work of Giovanni Maria Lanfranco (Scintille de musica, Brescia, 1533).
In Harmonie Universelle by Marin Mersenne published in 1636 there is one of the first
computation of the values of the frequencies. In the equal temperament, the octave is
divided in twelve equal parts. The ratio of two consecutive notes is 2 1 / 1 2 (100 cents).
The notes are as follows: C (1), Cfl (2 1 / 1 2 , 100 cents), D (2 1 / 6 , 200 cents), D(J (2 1 / 4 , 300
cents), E (2 1 / 6 , 400 cents), F (2 5 / 1 2 , 500 cents), Ftf (2 1 / 2 , 600 cents), G (2 2 / 3 , 800 cents),
A (2 3 / 4 , 900 cents), A tf (2 5 / 6 , 1000 cents), B (2 1 1 / 1 2 , 1100 cents), C (2, 1200 cents). This
tuning is generalized to a system of n notes: the n tone equal temperament (n-tet) in
which each interval is equal to 21/71 or 1200/n cents.

6.4 Zarlinean Temperaments


Zarlinean systems are based on the structure of a perfect chord (C, E, G) transposed
in three positions: F-A-C, C-E-G and G-B-D. Once the acoustic ratios of the three
fundamentals F (4/3), C (1) and G (3/2) is fixed, we compute the other ratios: A (4/3
x 5/4 = 5/3), B (15/8) and D (3/2 x 3/2 = 9/4 or 9/8 in the octave). This leads to
the zarlinean diatonic system: C (1), D (9/8), E (5/4), F (4/3), G (3/2), A (5/3), B
(15/8). The chromatic zarlinean system is established by considering the sharp (multiply
by 25/24) and the mol (divide by 25/24). This leads to the system of 21 degrees: C (1), Cfl
(25/24), Db (27/25), D (9/8), Dfi (75/64), E\> (6/5), E (5/4), F\> (32/25), E$ (125/96),
F (4/3), F$ (25/18), G\> (36/25), G (3/2), GJ (25/16), A\> (8/5), A (5/3), A$ (125/72),
B\> (9/5), B (15/8), C\> (48/25), B% (235/64).
The system proposed by Charles Edouard Joseph Delezenne (1776-1866) is built on the
diatonic system by lowering (multiply the previous note by 16/15) or raising (multiply
the next note by 15/16). This leads to the system of 17 notes: C (1), Q (135/128), Db
(16/15), D (9/8), m (75/64), Eb (6/5), E (5/4), F (4/3), JJ (45/32), Gb (64/45), G
(3/2), GJ (25/16), Ab (8/5), A (5/3), Aft (225/128), Bb (16/9), B (15/8).

6.5 Meantone Temperaments


In the meantone temperaments, each fifth is diminished by a fraction r of the syntonic
comma. The system presented here has been built on the fifths from Eb to GJ|. The
classical meantone temperament (r = 1/4) has been proposed by Pietro Aaron (ca.
1485-1545) in his book Toscanello in musica published in 1523. In the meantone tem-
peraments, all the fifths are tempered, that is are pure fifths diminished by a fraction
of the syntonic comma, except the wolf fifth (Ab-Eb). Many theorists and musicians
have proposed different meantone tunings with different fractions: r = 1/4 Aaron (1523),
r = 1/10 Lanfranco (1533), r = 2/7 G. Zarlino (1558), r = 1/3 F. Salinas (1577), r = 2/9
L. Rossi (1666), r = 1/5 J. Sauveur (1710), r = 1/6 G. Silbermann (1748), r = 5/18 R.
Smith (1749), r = 5/18 J.B. Romieu (1758). For r = 1/4 (the classical case), there are
eight pure major thirds. For r = 1/3, there are nine minor pure thirds. In the general

130
Historical Temperaments

case, there are two kinds of intervals

a
=2*k
and
2 ,5r
35
The general form of the meantone temperaments is given in the following table.

| Notes Ratios Inter v. AF ATM ATm


C 1 a -rK -(4r - l)K {Sr-1)K
3 72-llfc-7r b -rK SrK-S (3r-l)K
cj
D 322-3A.-2r b -rK -{Ar - l)K (3r-l)K
Eb 2 5 3" 3 fc 3r a -rK -{Ar - 1)K -9rK + S
E 3 4 2~ 6 A;- 4 r b -rK -(4r - \)K (3r-l)K
F 223~1.kr a -rK -{Ar-\)K -9rK + S
362-9k-6r b -rK 8rK-S (3r-l)K
nG 3-2-1fe-r a -rK -{Ar - l)K (3r-l)K
G|t 3 8 2" 1 2 A;- 8 r b (lln-1) K-S SrK-S (3r-l)K
A 332-4&"3r b -rK -{AT - 1)K (3r-l)K
Bb 2 4 3" 2 fc 2r a -rK -{Ar - l)K -9rK + S
B 3 5 2" 7 A;- 5 r b -rK SrK-S {3r-l)K

6.6 Historical Temperaments


We have seen that by multiplying the frequency of a note by the fifth ratio (3/2), we go
up in the fifths' spiral. Starting form the note Ci with acoustic ratio (1), we can build a
sequence of fifths d (3/2), D2 (3 2 /2 3 ), A2 (3 3 /2 4 ), E3 (3 4 /2 6 ), B3 (3 5 /2 7 ), etc. About
seven octaves (2 7 ), we expected to get a note C#. But we only get a Bfa of ratio (3 1 2 /2 1 9 ).
The difference between seven octaves and twelve fifths defines the Pythagorean comma.
It is impossible to close the cycle of fifths, because from a mathematical point of view,
the equation

©'
has no integer solutions (except the trivial case (0,0)). Pythgoras has proposed an approx-
imate solution (12, 7). Musicians and theorists have tried to get approximate solutions,
usually by diminishing the pure fifths with a fraction of comma. The system thus obtained
is called a temperament.

Divisions of the Pythagorean Comma, In the second column of the following


table, the vectors are defined with the following cycle of fifths (C, G, D, A, E, B, Ffr, C§,
GH, D$, i4((, F). The first coordinate of the vectors indicates the fraction of Pythagorean
comma needed for tempered the fifth C-<7, the second coordinate concerns the fifth G-D,
etc. Many other temperaments have been studied by Neidhardt and Marpurg (see for
example [Jedrzejewski, 2000]).

131
6. Theory of Temperaments

1 Zwolle (15th) ( 0 , 0 , 0 , 0 , 0 , - p , 0,0,0,0,0,0)


| Bermudo (1555) (0, - p / 6 , 0 , -p/6,0, - p / 6 , 0 , 0 , - p / 2 , 0 , 0 , 0 )
| Werckmeister III (-P/4, - p / 4 , -p/4,0,0, -p/4,0,0,0,0,0,0)
| Werckmeister IV ( - P / 3 , 0 , - p / 3 , 0 , - p / 3 , 0 , - p / 3 , 0 , +P/3, + p / 3 , - p / 3 , 0 )
| Werckmeister V (0,0, - p / 4 , - p / 4 , 0 , 0 , - p / 4 , - p / 4 , + p / 4 , 0 , 0 , - p / 4 )
Werckmeister VI (-p/7, -4p/7,p/7,0,0, -p/7, -2p/7,0,p/7,0, -p/7,0)
Bendeler I (1739) (-P/3, -p/3,0,0,0, -p/3,0,0,0,0,0,0)
Bendeler II (-p/3,0, -p/3,0,0,0, -p/3,0,0,0,0,0)
Bendeler III (-p/4, -p/4,0,0, -p/4,0,0,0, -p/4,0,0,0)
von Wiese I (0,0,0,0,0, - p / 2 , - p / 2 , 0 , 0 , 0 , 0 , 0 )
von Wiese III (0,0,0,0,0, - p / 2 , 0 , 0 , 0 , - p / 2 , 0 , 0 )
Young II (1800) ( - p / 6 , - p / 6 , -p/6, - p / 6 , -p/6, - p / 6 , 0 , 0 , 0 , 0 , 0 , 0 ) |

Divisions of the Syntonic Comma. Instead of building a temperament by the


division of the Pythagorean comma, many theorists have proposed to distribute the
syntonic comma. The following table shows the distribution of the syntonic comma K
and the schilisma S on the cycle of fifths (C, G, D, A, E, B, ify Cfl, Gtf, Dtf, A$, F). In
the Ramis temperament, the just second fifth G-D is diminished by a syntonic comma
(K) and the eigth fifth D\>-A\> is diminished by a schilisma (5). The others fifths remain
just (3/2).

B. Ramis (1482) (0, -K, 0,0,0,0,0, - 5 , 0 , 0 , 0 , 0 )


M. Agricola (1539) ( 0 , 0 , 0 , 0 , 0 , -K, 0,0,0, - 5 , 0 , 0 )
S. de Caus (1615) (0, -K, 0,0,0, -K, 0,0,0, K - S,0,0)
Kepler I (1619) (0,0,0, -K, 0,0,0,0, K - 5,0, -K, 0)
Kepler II (1619) (0,0,0, -K, 0,0,0, K - S, 0,0, -K, 0)
Mersenne I (1636) (0,0, -K, 0,0, -K, 0,0,0, K - 5,0,0)
Mersenne II (1636) (0, -Ky 0,0,0, K - 5 , 0 , 0 , 0 , - X , 0,0)
Mersenne III (1636) (0,0, -K, 0,0, K - 5 , 0 , 0 , 0 , 0 , -K, 0)
Malcolm (1721) (0,0, -Ky 0,0,0, K - 5,0,0, -K, 0,0)
Euler (1739) (0,0, -K, 0,0,0, -K, 0,0,0, K - 5,0)
Rousseau (1768) (0,0, -Kt 0,0, -K, 0, IK - 5,0,0, -K, 0)
7
Approximations of Temperaments

This chapter investigates the methods of approximations of temperaments by continued


fractions. After a review of the main results on the approximation of an irrational number
by a sequence of rational numbers, we show how to construct rational approximations
of meantone temperament. Conversely, we explore the problem of finding the best equal
temperament with n pitch classes in a neighborhood of a given rational temperament.

7.1 Continued Fractions


Continued fractions appear, apparently for the first time, in the work of the Hindu
mathematician Aryabhata in the late fifth century. A simple continued fraction is a
representation of a real number a given by

1
a = a0-\ =
ai +
1
a2 +
or in a compact abbreviated notation

a = [a 0 ;ai,a 2 ,...]

with a semicolon to separate the integer part. When the integer part is null (ao = 0), the
notation a = [a\, a 2 ,...] is also used. For typographic convenience, the continued fraction
is also written as
1 1 1
a0 + ; ; "...
It is well known that every real number a has a unique continued fraction expansion.
This expansion is finite if a is rational (a € Q) and infinite if a is irrational (a € '.
7. Approximations of Temperaments

Let [a] be the integer part of a, the largest integer less or equal a. Define the sequence
xn by the relations x$ = a and
1
Xn+l =

The numbers an are then the integer part of xn

Define inductively the numbers pn and qn by the relations


= fl a
PO = Go, Pi 0l + !> Pn+l = Pn^n+l + P n - 1 , U>\
QO = 1) 2l = Gl> tfn+1 = ^n^n+1 + 2 n - l > n > 1

The convergents are the rational numbers c n defined by

Pn r , 1
c n = — = [ao;ai,...,a n J = a 0 H =j
*n a2 +
1
+—
Example 1. The computation of the n-th convergent of

TT= [3; 7,15,1,292,1,1,...]

leads to c0 = 3, c1 = 22/7, c2 = 333/106, c 3 = 355/113, etc.

Example 2. The golden ratio

has the following convergents: co = 1, ci = 2, C2 = 3/2, C3 = 5/3, C4 = 8/5, C5 = 13/8,


c6 = 21/13, etc.

Proposition 146 Continued fractions of a real number a verifies the following proper-
ties:

I- (Qn) is a strictly increasing sequence and tends to +00

2. Vn G N, Pn+iQn -Pnqn+i = (~l)n


3- fak) is an increasing sequence, and fak+i) is a decreasing sequence.

4. Vfc € N , C2k < OL < C2fc+i.

Proof (1) It is easy to see by induction that (qn) is a strictly increasing sequence, since
qo = 1 < qi = a\ and if qn-\ < qn we have
<?n+l n , Qn-1 . „ . -,

134
Continued Fractions

(2) Eliminating an between the equations

Pn+1 = Pn^n+1 + P n - 1
Qn+1 = qnQ>n+l + Qn-1

leads to
Pn+ltfn " P n t f n + l = - ( P n t f n - 1 ~ Pn-lQn-l)

and since
Piqo -poqi = l
we get
Pn+l?n -Pnqn+1 = (-1)*
(3) Dividing the previous equation by qnqn+\> we get

Pn+1 Pn _ (~l)n+1
tfn+1 qn qnqn+1

or
(-l)n+1
Cn+l ^n —
#n<7n+l
Thus, the distance between cn and c n _2 is

Pn Pn-2
Cn - C n - 2 =
<Zn qn-1
Pn-1 Pn-2 ("I)71"1
<Zn-l <7n-2 qn-iqn
n 2
(-l) ~ (-l)^1
qn-2qn-i qn-iqn

#n-l tfn tfn-2


(-l)nan
<7n-2<7n-l

T h e even convergents C2& form an increasing sequence and t h e odd convergents C2k+i
form a decreasing sequence. •

T h e o r e m 1 4 7 For n € N, we have

1 Pn 1
< a <
tfn(tfn+l +tfn) tfn qnqn+\

Proof Substituting in the expression [ a o , a i , . . . , a n + i ] , a n + i by the rest a n + i , the real


a is
Pn<*n+1 + P n - 1
a =
^n^n+1 + Qn-1
Since a n + i , p n , Qn are positive

Pn\ Pn+1 _ Pn
a— <
Qn\ | #n+l tfn | tfntfn+1

135
7. Approximations of Temperaments

On the other hand, the distance

|Pn+l+Pn Pn (Pn+1 +Pn)qn ~Pn(gn+l + qn)


Qn+1 + Qn Qn QniVn+l +Qn)
(~l)n
tfnfan+l + g n )
1

T h e induction relation induces

Qn-i > qn-2 and qn > 2 g n _ 2

thus
1 1 1
tfntfn+i
< ^
2gn_ign
<
4qn-iqn-2
<...
T h e number a is t h e sum of t h e alternate serie
oo
OO / \

^°°tfn qo ^ V ? n Qn-lJ
n
Po 1 (-l)
- ^
40 + tfotfl + ... + tfntfn+l
- — + ...

T h e latest property leads to t h e following theorem which gives the best approximation
of a real number.
T h e o r e m 1 4 8 Among the fractions p/q with q<qn, the closest to the real number a is
the convergent pn/qn.

Proof Suppose t h a t there exists a fraction p/q such t h a t

V Pn
a —q <^ a - qn
and suppose, in a first time, t h a t t h e fraction p/q is between cn = pn/qn and c n + i =
Pn+i/<?n+i- T h e sequence (..., Pn/qn,p/q,Pn+i/qn+i, ~) is increasing or decreasing. For a
non-negative integer N,

Pn _P N Pn+l 1
<
qn q qnq qn+i qn #n<Zn+l

t h a t is q > Nqn+\ > qn+i. If t h e sequence is increasing, using the property (2), we get

Pqn > qPn > qn+lPn+l > Pn+tfn

thus q > Pn+i and if t h e sequence is decreasing, we get

N n + l > qPn+1 > qn+lPn+\

Consequently p > p n + i . Now if p/q does not belong to the interval [c n , c n +i] since

1 1
qnqn+i < ^ 2qn-1qn < ..

136
Continued Fractions

the length of the interval [c n ,c n + i] is at most the half of the length of [c n _i,c n ]. Thus
p/q belongs to [c n _i, c n ]. Prom the first part of the proof, we see that

q>qn and p > pn

So for a fraction p/q such that

V Pn
a - < a - q
Q n

we have q > qn and p > pn. The approximation of the real a by a convergent is better
than the approximation by other fractions simpler than the convergent. •
If the rational p/q is a best approximation to a, then p/q is a convergent. But it is not
true that if the rational p/q satisfies

1
a
q2
then p/q is a convergent of a. However a theorem of Hurwitz states that if a rational
number satisfies
a <
q 2q2
then it is a convergent in the continued fraction expansion of a.

Examples. To find q just fifths (3/2) lying in p octaves, one has to solve the equation

This equation has no integer solutions. So, we have to find the best approximation

f-^|) = 0

The convergents are c\ = 1, c<i = 1/2, C3 = 3/5, C4 = 7/12, c 5 = 24/41, CQ = 31/53,


etc. The denominators of the convergents give the number of notes n of the best n-tet
approximations: 12, 41, 53, etc.

Another application of the theory of continued fractions is the computation of the


convergents of the irrational numbers
2*71200
&k

representing the frequency of the individual intervals of the 1200-tet, for each k =
1,..., 1200. In the sequence of the convergents c£\ which are expressed in cents by the
formula
1 (n)
» 1200 log Ck
log 2
we keep only the first convergent (denoted simply c&) such that the approximated ratio
is less than 0.5 cents
» k\ < l / 2

137
7. Approximations of Temperaments

Thus we build an approximate 1200-tet, a = 1731/1730, c2 = 866/865, c 3 = 577/576,


etc. The approximation reveals different behaviors of the individual intervals. The fol-
lowing figure shows the distribution of the inverse of the numerator of the convergents
Ck against the number k. It is a kind of Stern-Brocot tree (see chapter 8).

0.3

0.25

0.2

0.15:

0.1:

0.05 :

~ d ' " " '260' ' " 400 ' ' 6do" " ' 800' " ' iobd ' ' 1200

Superpartial ratios (i.e. ratios of the form (fc + l)/k) have an important property:
superpartial ratios are Hellegouarch commas. They are ratios of the first just intervals
(2/1, 3/2, 4/3, 5/4, 6/5, etc) and also they are the best approximations of microtones.
In the sequence of convergents of the approximation of a microtone of frequency 21/71 in
the n-tet, there is always a superpartial ratio. The following table shows the superpartial
ratios in the computation by continued fractions of the microtones (second column), the
number of degrees n (n-tet, first column) and the values in cents of the microtones (third
column).

110n Ratios
15/14
Cents
120
136n Ratios
52/51
Cents
33
12 17/16 105 | 53 77/76 23
17 25/24 71 54 79/78 22
18 26/25 68 60 87/86 20
19 28/27 63 66 95/94 18
24 35/34 50 72 104/103 17
31 45/44 39 | 96 138/137 13 |

7.2 Meantone Approximations


Let k be the syntonic comma. Applying the continued fractions of the n-comma meantone
temperaments leads to the best equal temperament approximations of m degrees. The
number m is determined from the equation

138
Meantone Approximations

by the computation of the convergents of

V = log(2)
q log(3/(2ft»))

The following table gives the approximations for the classical meantone temperaments.
The number of degrees m of the approximated m-tet is the denominator of the conver-
gents. The two first convergents (1 and 1/2) are omitted. The last column gives the usual
approximation.

1 0r Authors
Pythagoras
Convergents
3/5, 7/12, 24/41, 31/53, 179/306
Us.
53
1/9 Romieu 3/5, 4/7, 7/12, 130/223, 137/235 223
1/8 Romieu 3/5, 4/7, 7/12, 74/127, 81/139 139
1/7 Romieu 3/5, 4/7, 7/12, 46/79, 53/91, 364/625 91
1/6 Silbermann 3/5, 4/7, 7/12, 32/55, 39/67, 71/122 55
3/17 Riccati 3/5, 4/7, 7/12, 25/43, 32/55, 569/798 55
1/5 Sauveur 3/5, 4/7, 7/12, 18/31, 25/43, 768/1321 43
3/14 Riccati 3/5, 4/7, 7/12, 18/31, 25/43, 43/74, 74
5/23 Smith 3/5, 4/7, 7/12, 18/31, 43/74, 534/919 74
2/9 Rossi 3/5, 4/7, 7/12, 18/31, 43/74, 61/105 74
1/4 Aaron 3/5, 4/7, 7/12, 11/19, 18/31, 101/174 31
5/18 Smith 3/5, 4/7, 7/12, 11/19, 29/50, 707/1219 50
2/7 Zarlino 3/5, 4/7, 7/12, 11/19, 29/50, 69/119 50
3/10 Harrisson 3/5, 4/7, 7/12, 11/19, 40/69, 51/88 69
1/3 Salinas 3/5, 4/7, 7/12, 11/19, 733/1266 19 1

The classical meantone temperament has irrational ratios (C, 1, Db, 5 7 / 4 /16, D >/5/2,
Eb 4/5 3 / 4 , E, 5/4, F 2/5 1 / 4 , Ffl, 5>/5/8, G 5 1 / 4 , G(J, 25/16, A 5 3 / 4 /2, Bb, 4/y/E, B, 5 5 / 4 /4).
To get a rational system that approximates this tuning, one has to find fractions to be
used instead of powers of 5. If we carefully look at the system proposed by Verheijen, as
it has been stated by Simon Stevin in Van de Spiegeling der Singeonst (1884), one could
show that if we use the following approximations

^ m v^ieo 4^i95
559 107 128
all the ratios of the Verheijen system are described (C, 1, D\>, 40/39, D, 625/559, E\>,
125/107, E, 5/4, F, 107/80, Fj, 3125/2236, G, 160/107, Ab, 25/16, A, 107/64, B\>,
1000/559, B, 200/107).

Another example of rational approximations is given by the system of Otto Gibelius


(Propositiones mathematico-musicae, 1666). The approximations

/=• 3 6 0 4 / - 160 1 216


v 5 cz v 5 cz —= ~
161 107 4^/5 323
leads to the system (C, 1, D\>, 675/646, D, 180/161, Df, 125/107, E\>, 128/107, E, 5/4,
F, 432/323, ify 225/161, G, 160/107, Gfl, 25/16, A\>, 8/5, A, 540/323, B\>, 288/161, B
200/107).

139
7. Approximations of Temperaments

7.3 Harmonic Approximations


Donald E. Hall has introduced [Hall, 1985] a function of how well an equal temperament
can provide close approximations to harmonic intervals. Let { n , r 2 , . . . , r n } be a set of
harmonic intervals defined by their frequency ratios. Define their frequency values in
cents by
Cj = 12001n2(r,-)
The structural interval of the n-tone equal temperament is
1200

Let aij be the number of steps giving the closest approximation to Cj in the i-tone equal
temperament and the error e^ of this approximation

e =
ij \cj ~~ 0i s
ij j\

Suppose that the target interval size is a random variable, then the probability of getting
an error is

Hall's remarkability function is defined by


n
Ri(n) =-Y[\n(Pij)
3=1

Hall has done a systematic evaluation of equal temperaments through n = 612. It appears
that for the sequence of just intervals

F = {1,5/4,4/3,7/5,3/2,8/5,5/3,7/4}

the remarkability function exceeds 2.5 for 12,19, 31,41, 53, 62, 68, 72, 84, 87, 99,103,109,
etc. We have done the computation of the remarkability function for other temperaments.
Some results are given in the following table.

| Farey 12, 19, 31, 41, 53, 62, 68, 72, 84, 87, 99, 103, 109, 115, 118
| Partch 7-limit 19, 27, 31, 41, 53, 62, 68, 72, 84, 87, 99, 103, 109, 118
Partcli 11-limit 31, 41, 46, 58, 72, 80, 94, 118, 130, 152, 159, 183, 190, 198
Partch 13-limit 58, 66, 72, 87, 111, 130, 183, 190, 198, 224, 270, 296, 311
Zarlinian diatonic 12, 19, 22, 24, 31, 34, 41, 46, 53, 65, 75, 77, 84, 87, 94
Pythagoras 12, 24, 29, 41, 53, 65, 77, 82, 94, 106, 118, 130, 135, 147, 159
Kepler I 12, 41, 53, 65, 77, 94, 106, 118, 130, 159, 171, 183, 195, 205
al-Kindi 12, 24, 29, 41, 53, 65, 82, 94, 106, 118, 135, 147, 159, 171, 188
Mayumi Reinhardt 10, 66, 74, 87, 103, 113, 130, 140, 147, 177, 187, 198, 206, 207

The remarkability of the temperament of Mayumi Reinhardt {1, 14/13, 13/12, 16/13,
13/10, 18/13, 13/9, 20/13, 13/8, 22/13, 13/7, 208/105} based on the prime number

140
Multiple Continued Fractions

13 is very different from the other scales. Other functions such as the Paul Erlich's
harmonic entropy have tried to understand the relationship between non-equal and equal
temperaments, and to provide better approximations.

David C. Keenan has proposed the following definition of the harmonic error (limited
to three prime numbers 3, 5, 7) to measure the discrepancy of equal tunings in a way
very close to the remarkability.

Definition 149 Let R be the round function. For each prime number p > 3, we define
the p-th harmonic error of the n-tet by

1200
ep(n) log
nlog2 2#(nlogP/log2)

The harmonic error is defined by

M§#(5?)
The following plot shows the variations of the inverse of the harmonic error for n from
5 to 400.

2-\

100 200 300 400

7.4 Multiple Continued Fractions


A generalization of the one-dimensional continued fraction algorithm to two dimensions
was first considered by Jacobi in the 1830s. Later Perron [1907] performed what is now
called the Jacobi-Perron algorithm [Berstein, 1971]. The ergodic properties of this al-
gorithm was first studied by F. Schweiger [1973], and later by Gordin (1970), Mayer
(1984) and Ito and Yuri (1987). Since the publication of the Jacobi-Perron algorithm,

141
7. Approximations of Temperaments

many other multidimensional continued fraction algorithms have been proposed. We


briefly describe the algorithm proposed by Viggo Brun and the modified Jacobi-Perron
algorithm studied by D. Hardcastel and K. Khanin [2001] in two-dimensions. First remark
that it is well-known that if OJI, OL^ ...,a n , are real numbers with at least one irrational
number, then there are an infinite number of ways of approximating such numbers by
rationals numbers with the same denominator OL\ « pi/q, oti ~ P2/q, ...,«n « Pn/q such
that the errors are less than l / g 1 + 1 / n (see for example [Hardy-Wright, 1980]).

Brun Algorithm. The algorithm proposed by V. Brun in 1919 is a generalization of


the one-dimensional algorithm. Given three real numbers l > a > / 3 > 7 > 0 , find a
sequence of numbers xny yn and zn such that

XJI Oi Xn Oi

V. Brun starts with the following scheme


«0 = oe x0 = 1 yo = o *o = 0
Po = P 0 l 0
7o = 7 0 0 1
At the first step and on the first column, V. Brun places the difference between the
greatest and the next-greatest number (a —ft)and leaves unchanged the other numbers.
On the right-hand side, he copies the first and the third lines of the previous matrix and
replaces the second line by the sum of the numbers of the first and the second lines of
the previous step.
ai = a — /3 X\ = 1 yi = o *i = 0
\Pi = P 1+0 0+1 0+0
7i = 7 0 0 l
At the n-step, the first line gives the values of xn, yn and zn. For example, the Brun
algorithm applied to
a = log 2, /3 = log(3/2), 7 = log(4/3)
gives the following values
xn = 5, yn = 3, zn = 2 (n = 6)
xn = 12, yn = 7, zn = 5 (n = 8)
xn = 41, yn = 24, zn = 17 (n = 11)
xn = 53, y„ = 31, * n = 22 (n = 12)
xn = 306, y„ = 179, zn = 127 (n = 17)
For the numbers
a = log 2, P = log(3/2), 7 = log(5/4)
some other triplets (xn, yn, zn) are (12, 7, 4), (19, 11, 6), (31, 18, 10), (34, 20, 11), (53,
31, 17), (87, 51, 28), (118, 69, 38) etc.

Modified Jacobi-Perron Algorithm. The modified Jacobi-Perron algorithm for


finding the approximations of two numbers a and (3 such that
1 >a>/?>0

142
Multiple Continued Fractions

starts with the definition of the sequence (a n ,6 n ). The initial values are ao = a and
bo = p. The sequence is defined by the 2-dimensional Gauss transformation

T{x y)= m
' { ^(wVx)'mia(wl))
where {x} denotes the fractional part of x and

(a n +i, 6 n + i) = T(an, bn)

At each step, we compute the sequence


1
un = LT-i(a,/?)_

where [x] denotes the integer part of x, and the sequence

vn = v(Tn-1(a,/3))

where the function v{x^y) denotes the position of l/{x} in T(x,y), i.e. v(x)y) = i when
the z-th coordinate of T(x,y) is !/{#}. The matrices A(un, vn) are given by

i( un 1 0
A
(UnA) = 0 0
\° 0 1

and
j( un 1 0
A
(un,2) = \ 0 0 1
0 0
It has been shown that the product

Cn = A(univn)A(un-uvn-i)...A(uuVi)

is a matrix of the form


(&
r» si
Cn=\
Qi ri si
\ ll rl si
For each j = 0,1,2 and when n tends to infinity, the ratios

-^^a and % -»0


3 3
qh qh
are approximations of (a, /?). For example, for the numbers
log(3/2) log(5/4)
H
log(2) log(2)

we find the approximations (7/12, 4/7, 2/3) and (1/3, 2/7, 1/3), (11/19, 7/12, 2/3) and
(6/19, 1/3, 1/3), (18/31, 2/3, 11/19) and (10/31, 1/3, 6/19), (10/17, 11/19, 18/31) and
(11/34, 6/19, 10/31), etc. The modified Jacobi-Perron algorithm has been generalized to
d-dimensions.

143
8
Tuning Groups

The twelve tone equal temperament is generated by the repetition of the semitone inter-
val. Rational temperaments are also generated by the repetition of one or several intervals,
but some relations must be added, such as the approximation of the pythagorean comma
by the unit ratio, in order to close the cycle of fifths. The various methods to consider a
temperament are explained in this chapter.

8.1 Definitions
Equal temperaments are cyclic groups. They are generated by the repetition of a single
ratio { l , r , r 2 , . . . . , r n _ 1 } . This set is closed by the octavian relation rn = 2, which induces
an isomorphism of this set with the cyclic group. For the 12-tone equal temperament,
the ratio r is the semitone r = 2 1 / 1 2 . Unfortunately, most of temperaments are not cyclic
groups. As we would like to have a complete description of the frequency ratios by the
powers of r, we introduce the notion of a pseudo-cyclic group. Consider a temperament
T of n pitch-classes, i.e. an increasing sequence of frequency ratios xo = 1 and Xj G [1,^[
for j = 1,.., n — 1 (in general, 77 = 2 is the octave)

T = {l,xiyX2,...,xn-i}

The pseudo-cyclic group associated with T is given by a generator r = Xi and n — 1


regulators C2,...,c n such that Xk = rkCk, for k = 2, ...,n — 1 and a closure relation
rn
c n = 77. The set T is written in the pseudo-cyclic form as

T={l,r,r2c2,r3c3,...,rn-1cn_i}

The regulators define a set of units like

r r
8. Tuning Groups

Example. The Pythagorean temperament T = {1, 256/243, 9/8, 32/27, 81/64, 4/3,
1024/729, 3/2, 128/81, 27/16, 16/9, 243/128} is a pseudo-cyclic group whose gener-
ator is the leimma r = 256/243 and the regulators are c<i = cz = p, C4 = C5 = CQ = p 2 ,
c7 = c$ = eg = p 4 and c\\ = c\2 = pb. The ratio p = 531441/524288 is the Pythagorean
comma. The set

T = {1, r, r2p, r3p, r4p2, r5p2, r6p2, r V , r V , r V , r 1 0 p W }

describes each ratio and ends with the closure relation r12p5 = 2. The regulators define
a unique relation
P = i

which means that in order to close the cycle of fifths, the Pythagorean comma must be
a unit approximation or a unit like.
Definition 150 Two pseudo-cyclic groups T\ and T2 of same period 77 and same gener-
ator r are equivalent T\ 1Z T2 if they have the same units like.
Definition 151 Let r be a generator and ui, ..,u m be m units like. The tuning group G
generated by r modulo the relations Uj = 1, for j = 1, ..., m

G =< r, ui = 1,..., um = 1 >

is the class of all pseudo-cyclic groups for the equivalence relation 1Z.
Examples. (1) The set

H = | l , r , ra, r a^r a , r a , r a , r a , ,r a , r a ,r a }

is a system built with a generator r and only one regulator a. The closure relation is
r6a6 = 2. This system corresponds to the tuning group

G =< r, - = 1>
r
Meantone temperaments are pseudo-cyclic groups with one regulator.
(2) The set

H = {1, r, ra, ra6, r2a&, r 2 a 2 6, r 2 a 2 6 2 , r 3 a 2 6 2 , r 3 a 3 6 2 , r 3 a 3 6 3 , r«4a363, r 4 a 4 6 3 }

is a pseudo-cyclic group with two regulators and the closure relation r 4 a 4 6 4 = 2. This
pseudo-cyclic group is associated with the tuning group
a
^ •. b
G =< r, - = 1, - = 1>
r r
For example, the group

9 3 ? 3 4
n - i - 1
G-< = 1 = 1 >
8' 2^ ' 215
describes the diatonic Zarlinian temperament.

Conversely, a tuning group describes a set of temperaments. The canonical tempera-


ment associated with a given tuning group is the temperament for which each ratio has

146
Definitions

the smallest height. For example, let's consider the following presentation of the tuning
group
9 3?
-,
G = <
8 ' 2 * s l >
This group has one generator r = 9/8 and one relation u = 1, with u = 3 7 /2 1 1 . The
group G has elements of the form l , r , r 2 , r 3 , . . . . , r n modulo the relation given by u.
The ratios X{ = rluk (where i = 1,2, ...,n and k G Z) form an increasing sequence
#i < #2 < #3 < .... < xn limited by the octave. As there are several possibilities to
choose the number fc, there are several ratios. One choose the ratio Pijq% which have the
smallest height
ftmin = infmax(pi,gi)

The first note of our tuning system is C with ratio 1. If we multiply this ratio by the
generator r, we get x\ = r = 9/8. If we multiply again this value by the generator, we
have r2 = 81/64 or 408 cents. But this ratio has not the smallest height, because it can
be divided by u
_ r 2 _ 3 * 2 ^ _ 2* _ 32
X2
" "JT " ¥ ~W " 33 " 27
If we multiply this new ratio by the generator, we get the new value
2532 4
^3 = ^ = 3 3 2 3 = 3

which has the smallest height. Multiplying again the ratio #3 by the generator r leads to
X4 = x^r = 3/2. We build the temperament by the powers of the generator, modulo the
smallest height. We obtain: C (1), D (9/8), E (32/27), F (4/3), G (3/2), A (27/16), B
(16/9) and C (2).

Remark. One could also define a tuning group with several generators. Let R = {ri, ...r n }
be the set of generators and U = {^1, ...yum} the set of relations, the tuning group

G=<ri,...,rn, ui =u2 = . . . =um = 1>

describes the class of the pseudo-cyclic groups whose elements are of the form

r r
l 2 * ' * rn u
l '"um

For example, a model with two generators r and 5,

G=<10/9, 9/8, | g = l >

with r = 10/9 and s = 9/8 is composed of all elements of the form r V , that is r, 5, r 2 ,
rs, s 2 , r 3 , r2s, rs2, s3, etc. modulo the relation. One can verify that this group leads to
a temperament of 26 notes.

T h e o r e m 152 Letpi, p2, P3,..., pn be prime numbers in increasing order and let r\ be
the number
n =Pl Vl P'S -'Pn'

147
8. Tuning Groups

Let G be the group generated by the prime numbers pi, P2, P3,---> Pn and H be the group
generated by the ratios

n = Pi P2 -Pnn with i = 1 , . . . , n

The number r\ is a generator of the group G/H if and only if the determinant of the
matrix A = (a^) is ±1

&i,i ai,2 ai}3 • •• a\yU


^2,1 ^2,2 &2,3 * * * &2,n
detA = ^3,1 ^3,2 ^3,3 • •• a3jn
±1

Proof Let's denote


a
xi «i,2 i,3
^2 «2,2 a 2 ,3 &2,n
#3 ^3,2 a3>3 «3,n ^ l ^ l , n + l + #2^2,71+1) ••• + ^n<ln,n+l

•^n,™

According to the Bezout's theorem

gcd(ai)n+i,a2,n+l,---)^n,n+l) = 1

is equivalent to the existence of a vector (a^i, ...,a n> i) G Z n such that


n
/ J*i,\a>iin+l = =tl
i=l

Thus the n column vectors of the matrix A (ai,i,...,oi | n ), (a2,i>--->a2,n)> and


(ai >n , ..., a n j n ) form a basis of Z n . If we set
n
L= 0Z(ai>i,...,anji)
i=2

the quotient Z/L is a group isomorphic to Z generated by r i .



Examples. For the tuning group
9
37 „
G = <
8 ' 2 * s l >

the generator is r = 2 - 3 3 2 = 9/8, and the relation 2 - 1 1 3 7 = 1, leading to the determ-


inant
detA = - 3 - 1 1 -21 + 22 = 1
2 7
The Kirnberger II temperament is described by the tuning group

__ 28 31* 3*__
G
-<35> 2i9 = 1 > 24 = >

148
Formal Languages

The generator of the group r = 2 8 3~ 5 5° leads to a\^ = 8, a2,i = —5 and c^i = 0.


The first relation 2~ 19 3 1 2 5° = 1 gives ai ) 2 = —19, a^2 = 12 and a^^ = 0. The second
relation 2 - 4 3 4 5 - 1 = 1 gives a\$ = —4, a^3 = 4 and 0,3,3 = — 1. These values yield to
the determinant

8 -19 -4
8 -19
detA = -5 12 4 = — -5 = -(96-95) = - !
12
0 0 -1
If we note p the Pythagorean comma, k the syntonic comma, the usual tuning groups
are
G = < 21/71 >, the n-tone equal temperament
G = < r , p = 1 >, the Pythagorean temperaments
G = < r , p = 1, fc = 1 >, the syntonic temperaments
G =<r, k12n = p >, the meantone temperaments
G = < (3/2)*/ n >, the non-octaviant temperaments (n = 7, TEQJ of Serge Cordier,
n = 9, Wendy Carlos's alpha, etc.)
G =<p, 2 65 = 3 4 1 >, the Janko temperament of 41 notes
G = < p, 2 84 = 3 5 3 >, the Mercator temperament of 53 notes

8.2 Formal Languages


A tuning system can be described by its intervals. Denoting ai, a2, ...., an the different
intervals of the system, one can form a word to describe the period of the system (usually
one octave). For example, the word

X = (ab)(ab)(ab)(ab) = (ab)4

describes a system in which the first interval is a, the second 6, the third a, etc. With
this convention, the tempered system is denoted by a 12 . The problem underlying the
description of the tuning systems is to understand the relationships of the chords and
scales in non-tempered systems which are modified under transpositions. For example
the major scale have a structure of the form: tone, tone, semi-tone, tone, tone, tone,
semi-tone (ttsttts). In the meantone temperaments which are described by the word

X = (abbab)ab(abbab)

the C-major scale has the following structure

U = {ab)2{b){abfb

In this expression, parenthesis represent the intervals of the scale: the first interval is
(ab), the second (ab) the third 6 etc. One can see easily that the major scale D, F, G,
A, B\> has the same structure. But Cft major has a complete different structure

(b2)(ab)(a)(ab)(b2)(ab)(a)

The study of these relationships and their topography could explain the predominance
of some modulations.

149
8. Tuning Groups

Example 1. The tempered system is the simplest group. It is built by the repetition of
the semitone interval, which is the generator of the tuning group
G =< 2 1 / 1 2 >
It is described by the concatenation of the twelve repetition of the interval a of 100 cents.

X = a12
All transpositions have the same structure.

Example 2. The Pythagorean system is a tuning group in which the leimma (£) is the
generator and the Pythagorean comma (a/£) is a unit like

The Pythagorean system is described by the word

X = {iataU)p{iHataH)
in which p is the Pythagorean comma between G\> and F§, I is a leimma of 90 cents
(256/243 = 2 8 /3 5 ) and a denotes the apotome of 114 cents. Leimma, apotome, and
Pythagorean comma satisfy the relation
a = £ . p = 3 7 /2 1 1

Example 3. The meantone temperament of Pietro Aaron is a tuning group built with the
semitone generator and in which the enharmonic comma is a unit like
57/4 27 ,

In terms of formal language, it is described by the word

X = (abbab)ab(abbab)
where a is the semitone of 76 cents (5 7 / 4 /16) and 6 is a semitone of 117 cents (8/5 5 / 4 ).

Example 4- More generally, the meantone temperaments, defined by a fraction n of syn-


tonic comma, can be represented by the following tuning group

G=<k; *"» sp > fc-fi P=y


These systems are represented by the word

X = (abbab)ab(abbab)
7
3 28
7 n anc sanotner
where a is the semitone ( ^ j ^ ~ ) * ^* semitone (-^ fc5n). Let's denote c the
difference between a and 6, the meantone temperaments are described by a pseudo-cyclic
group of the form
H =< l , r , r 2 c , r 3 c 2 , r V , r V , r V , r 7 c 4 , r 8 c 4 , r V , r 1 V , r 1 1 c 6 >
The generator r is the semitone a and the regulator is c = b/a. The closure relation is
given by
a567 = 2

150
Graphical Representations

8.3 Graphical Representations


Many graphical representations of temperaments have been proposed. Jean Lattard in his
book Gammes et temperaments musicaux proposed to place horizontally the name of the
notes in the cycle of fifths order and to put vertically the difference in savarts between the
note in the considered temperament and the same note in the equal temperament. Marc
Champollion places vertically the difference in cents between the thirds and fifths with
respect to the equal temperament and to the just intonation. Rene Sandoz has proposed
in the Revue de Suisse Romande a presentation built from the difference in savarts with
respect to the equal temperament. Albrecht Wolf proposed a complete description of the
temperament with graph and tables for each interval.

,k Delta second thirds


(Syntonic comma)
+1
\yk/2

ih
D

l-lll-V
C-E-G +1
-1 viV *
Deltafirstthirds
A Yy (Syntonic comma)

XF

viA
B -1 X

Kimberge r II (C Major)

We proposed [Jedrzejewski, 2000] another graphical representation, aiming to visualize


the properties of perfect chords in a given scale (see the previous figure). In this graph,
each perfect triad is represented by a point. The horizontal axe represents the difference
of the first third of the perfect chord relatively to the pure third and the vertical axe
represents the difference of the second third relatively to the pure third. Units are usually
syntonic commas. For example, in the Kirnberger II temperament and in the C-major
scale, the first degree characterized by the perfect triad (C-E-G) is represented by the
point (0,0). The major third (C-E) is just, so the difference is null. The minor third
(E-G) is also just. Then the first degree is placed on the origin of the graph. The second
degree of the temperament is characterized by the perfect chord (D-F-A). The minor
third (D-F) is less than the syntonic comma from the pure, so the abscisse is —1. The
major third (F-A) is greater than an half syntonic comma from the pure. The coordinates
of the second degree is then (—1, 1/2). By placing the seven perfect triads of major scale,
we have a complete topography of the scales, locating "strong" degrees (as I, IV, V in
the tempered system), and other properties as well.

151
8. Tuning Groups

8.4 Harmonic Metrics


Many authors have proposed new definitions of harmonic distance. It seems that the
mathematical distance on rational numbers are not well adapted to musical theory. Most
of the definitions presented here have been established by Yves Hellegouarch [1980], John
Chalmers [1993] and Ervin Wilson [1979]. The p-adic norm is also well suited to musical
considerations.
Definition 153 The harmonic metric of a rational frequency ratio p/q (p, q are relat-
ively prime) in the basis b > 1 is

db(Z) = logbh(?)

where h(p/q) = max(p,q). For a k-chord of rational frequencies x = (1, fi, ..., fk-i) the
harmonic metric is defined by
db(x) = log6 max(A, A/i,.., Xfk-i)
where fi = pi/qi and A = gcdfa).
Remark that the map db is a metric on Q + \ { 0 } . When the frequencies are not rational,
the harmonic metric is extended using the property
h(xn) = [h(x))n VneN
Examples. In the Pythagorean tuning, the major chord x = (C\^E^G^C2) has the har-
monic distance equal to
34 3
d2(x) = log2max(l,26,-,2)
= log 2 max(2 6 ,3 4 ,2 5 .3,2 7 )
= log 2 (2 7 ) = 7.
In the 12-tone temperament, for the same chord x = (1, 2 1 / 3 , 2 7 / 1 2 , 2) we have
x12 = (1, 2 4 , 2 7 , 2 12 ) and
h(x12) = 2 12 = (h(x))12
So h(x) = 2 and
d2(x) = log2(2) = 1
Definition 154 The harmonic gap of a rational frequency p/q (p, q relatively prime and
P, Q>1) is the logarithm sum of its numerator and denominator
H{p/q) = logp + logg
The harmonic distance of two ratios r\ = p\jq\ and r2 = P2/Q2 is defined by
d(r1,r2) = \H(r2)-H(r1)\
The harmonic gap verifies the following properties.

(1) H(p/q)>0 farallp/0 6[l,2]


(2) H ((p/q)a) = aH (p/q) for all p/q G [1,2], a € R
(3) H (rir 2 ) = H (n) + H (r 2 ) n = -pi/qi and r 2 = £2/92-

152
Harmonic Metrics

Definition 155 The harmonic complexity of a frequency ratio p/q whose decomposition
in prime factors has the form p/q = 2a°p^1 ...p%n where pi,..., pn <we prime numbers
greater than 2 is the sum of the prime factors different of 2
C(p/q)=Pi + ...+pn
Definition 156 The indigestibility of a prime number p is
. , 2(p-l)2 . .
X\P) = ¥ P %s prime
V
The indigestibility of a product of two prime numbers is (by definition) the sum of the
indigestibility of its factors
x(pq) = x(p) + x(q)
The harmonicity of the ratio p/q is
sgn(x(p) - x(q))
hip/q) =
x(p) + x(q)
where sgn{x) = \x\ /x is the sign of x ^ 0.

Example. The indigestibility of the first numbers is x(2) = 1, x(3) = 8/3, x(5) = 32/5.
The harmonicity of the pure fifth 3/2 is h{3/2) = 3/11. The harmonicity of the syntonic
comma /i(81/80) = 52/5. The harmonicity of the Pythagorean comma is /i(3 12 /2 19 ) =
1/51.

Definition 157 The p-adic norm of the rational frequency ratio x = par/s where p is
a prime number, r and s are integers not divisible by p, is defined by

\x\D=p-a
with the convention that |0| = 0.

Example. For the syntonic comma, 81/80 = 3 4 /(2 4 • 5), the p-adic norms are
81 81 1 81
= 16,
80 2 80 3 "81' 80

The p-adic norm is a metric which verifies the following properties


v
l^lP = MpMp ^2/
and
\x + y\p<max(\x\p,\y\p) Vx,y
For two ratios r\ and r<i of the form r = p/q = 2 °p* ...p£n one can consider the function
a 1

d{r\,r2) = ||r*2 - n | | with

- + ....+ \q\
\q\ "Pi
or
- = sup
\q
l-l l-l
\q\ p\

153
8. Tuning Groups

8.5 Hellegouarch Commas


All the results of this section have been established by Yves Hellegouarch [Hellegou-
arch, 1983]. Let pi,P2,Pn be n prime numbers with p\ < P2 < ••• < Pn- We denote
G = < pi, ...,p n > the subgroup of Q+\{0} generated by the prime numbers pj.
Definition 158 A comma of the group G (or a best approximation of 1 in G) is a
number a ^ 1 verifying

6GG\{1} and |log6| < |loga| =* h(b) > h(a)

where h{p/q) = max(p)q).


Remark that if a is a comma of a subgroup if of G, a is not necessarily a comma of
G. For example, 2 8 / 3 5 is a comma of H =< 2,3 >, but not a comma of G =< 2,3,5 >.
If a = pXl ...Pn11 is a comma then gcd{x\,.., xn) = 1.
Proposition 159 Let a be a positive rational number, then (a + l ) / a is a comma ofG.
Proof Let p/q > 1 be a positive rational. One must show that if p/q > 1 and log (p/q) <
log((a + l)/a) then p > a + 1. But since the function x —> (x + \)/x is decreasing on
]0, +oo[ and (q-\-l)/q < p/q < ( a + l ) / a , we have q > a. The inequalitiesp > q+l > a + 1
lead to the result. •

Theorem 160 Let p, q be two prime numbers with p < q. The commas of the group
G =< p,q> are the convergents of the irrational number a = log pj log q.
In the general case G =< piy ...,p n >, commas are given by the Chemillier-Duchamp
algorithm, see [Chemillier, 1999].

Proof We must show that all comma is a convergent. Let a = pxqy be a comma. Then x
and y have opposite signs and \x/y\ is a best approximation of a. Thus a is a convergent.
Conversely, let's define the sequence (an) by
q rn-2
r _ i = p, 7*0 = — , rn = r—
p°o {rn-irn
Since p < q, the number a$ exists and r 0 is the first convergent of a. There exits some
numbers an > 1 such that
C-i < rn-2 < C-l 1
because r n _ i < r n _2 and (log(r n _i), log(r n _2)) are linearly independent on Q. The
number rn is a comma. Let b = pxqy such that

|log p (6)|<|log p (r n )|

that is
\x + V logp(g) | < \xn + yn log p (g)|
and
(x„,J/„) = ( ( - i r - V n , ( - l ) n 9 n )
Since pn/qn is a best approximation of log p (g), we have \y\ < qn and \x\ > pn. Con-
sequently, x and y have opposite signs and then h(b) > h(rn). D

154
Consistency and Completeness

Definition 161 Let G be the group G =< p,q > and rn a comma of G. The group
G/ <rn> is isomorphic to Z and defines natural scales. For p = 2 and q = 3, the scales
are called Pythagorean scales.

Examples. For the first convergents, we recover the classical tunings

Hn=<2,3> / <rn >

- n = 4 / 3 , # i = {1,2}
-r2 = 9/8, H2 = {1,3/2,2}
-r3 = 2 8 /3 5 , H3 = {1,9/8,4/3,3/2,16/9,2}
- r\ = 3 1 2 /2 1 9 , H4 is the twelve Pythagorean temperament
- 7"5 = 2 6 5 /3 4 1 , H& is the Janko temperament of 41-degrees
- TQ = 3 5 3 /2 8 4 , HQ is the Mercator temperament of 53 degrees.

8.6 Consistency and Completeness


The notions presented in this section have been made up by Paul Hahn and commented
by Paul Erlich, John Chalmers, Patrick Ozzard-Low and others. Our definitions are
slightly different. In the following, the n-tet is identified with Z n = {0,1,2, ...,n — 1}.

Definition 162 Let X be a temperament and n be a positive integer. The mapping


An : X —* Zn associates to each integer ratio r the number k = An(r) which is the index
of the nearest approximation defined by

\2Vn-r\< \2e'n-r\ W<EZ n

Definition 163 A temperament X is n-consistent if for each triad {a,b,c} of X such


that c = ab we have
An(ab) = An{a) + An(b)

Example. The Partch 3-limit H3 = {1, 4/3, 3/2, 2} is 12-consistent. Since A12(l) = 0,
^12(4/3) = 5, A12(3/2) = 7, A12(2) = 12. We have for example

^ 1 2 ( 5 X 5 ) = 5 + 7 = ^12(2)

The Partch 5-limit H5 = {1, 6/5, 5/4, 4/3, 3/2, 8/5, 5/3, 2} is not 11-consistent. The
approximated indices in the 11-tet are respectively (0, 3, 4, 5, 6, 7, 8, 11) and

A 1 1 (|) + X 1 1 (J)-» + 4 ^ 1 ( J x f ) - A l ( | ) . .

Definition 164 Let X be a n-consistent temperament. X is n-complete if each indi-


vidual interval of the n-tet can be generated by some combination of the nearest approx-
imations in n-tet to the intervals ofX.In other words, X is n-complete if for each j £ Z n

155
8. Tuning Groups

there exists an integer r smallest as possible such that the ratios a\, a2, ..., ar € X\{1}
verify
An(ar) - An{ar-i) + ... + ( - l ) r _ 1 j 4 n ( a i ) = j (modn)
For all degrees j , the maximum value taken by r is called the diameter of X and is
denoted by 5n(X).

Example. The Partch 5-limit is 12-complete, but it is 24-incomplete, because there is


no combination to reach all odd degrees, since the nearest approximations An(x) are
even. The Partch 3-limit H% is approximated by the indices (0, 2, 3, 5) in the 5-tet. The
diameter is 2 because the degrees 1 and 4 are reached by a combination of two steps:
Scale degree 1 is generated by 3 steps minus 2 and scale degree 4 by 2 steps minus 3
(—1 = 4 mod 5). The approximated degrees of H% in 7-tet are (0, 3, 4, 7). The diameter
is 3 (J7(H3) = 3), because the index 2 needs at least three terms

156
9
Contemporary Tunings

The rediscovery of microtones at the beginning of the 20 th century and some new in-
vestigations on just intonation have led to a set of new theories which have been applied
in many musical works. Whereas Julian Carrillo, Alois H&ba, Ivan Wyschnegradsky and
others have used the equal temperament based on the equal division of tone, Adriaan
Fokker has developed a theory of 31-tone equal temperament. Many new tunings have
been explored in the field open by Harry Partch, Lou Harrison, Ezra Sims, Ervin Wilson
and Ben Johnston.

9.1 Euler-Fokker Genera


In the 18 th century, the Swiss mathematician Leonhard Euler (1707-1783) had made up a
new procedure to carry out tuning systems called musical genus (genus musicum) from a
simple mathematical formula (Tentamen Novae Theoriae Musicae, 1739). Starting from
an original pitch level n, he appoints intervals by their rank in the partials decomposition
of a fundamental pitch: 3 for the fifth, 5 for the third, 7 for the seventh, etc. The formula,
for example
32-53n

describes the set of all products of the form 3a • 56 • n where a and b take all integer values
a = 0, 1, 2 and b = 0, 1, 2, 3. So, one gets the set of possibilities, namely: n, 3n, 5n,
9n, 15n, 25n, 45n, 75n, 125n, 225n, 375n, 1125n. If one gives the value of n, for example
n = 1/5, the sequence becomes 1/5, 3/5, 1, 9/5, 3, 5, 9, 15, 25, 45, 75, 225. Introducing
the power of 2 in order to put all values in the octave [1, 2[, we get a complete tuning
system: C (1), D (9/8, 204 cents), D+ (75/64, 275 cents), D|f (6/5, 316 cents), E (5/4,
386 cents), Pji (45/32, 590 cents), G (3/2, 702 cents), G+ (25/16, 773 cents), Gfl (8/5,
814 cents), A+ (225/128, 977 cents), A% (9/5, 1018 cents), B (15/8, 1088 cents). The
9. Contemporary Tunings

Euler genera are given by the formula

3 a • 56 • V • • • n

They could describe many well-known temperaments. The genus 3 6 -n is the Eratosthene
diatonic system described by Ptolemaus. The genus 3 1 1 • n is the Pythagorean tempera-
ment built on the fifths ascending (if n = 311) and descending (if n = 1/3 11 ). The genus
3 3 • 5 2 • n corresponds to the zarlinian temperament. In the 20 th century, this procedure
has been widely studied by Adriaan Fokker (1889-1972) in relation with lattices proper-
ties. From the Euler expression 3 P • 5 9 • 7 5 , Fokker draws in a multi-dimensional space, p
fifths in the x direction, q thirds in the y direction over the fifths and s sevenths in the z
direction. For example, the genus 3 4 • 5 in the Fokker representation is described by the
lattice
E - B - F I - Q - G j l
I I I I I
C — G — D — A — E
All notes separated by a horizontal line are fifths, and all notes separated by vertical
line are major thirds. As in a crystallographic description, we get a lattice representation
where atoms are replaced by notes. So many results of crystallography, and in particular
results based on symmetry, can be applied to Fokker representation.

Another great contribution of Adriaan Fokker concerns the 31-tone equal temperament
(31-tet), also called tricesimoprimal or trentunisonic system. This system had been stud-
ied for the first time by Christiaan Huygens (1629-1695) who published in 1691 Novus
cyclus harmonious. In this temperament, the octave is divided in 31 equal parts called
diesis. The notes are named by prefixation. Let re be a note, xi is the note x plus one
diesis, xis is the note x plus two diesis, xisi is the note plus three diesis and xisis is the
note plus four diesis. For example, ais is A augmented by two diesis. Similar prefixes are
used for the descending order, xeh is the note x minus one diesis, xes is x minus two
diesis, xeseh is x minus three diesis and xeses is x minus four diesis. The tone is divided
in five diesis, and semitone has three diesis. In the 31-tet, enharmonic notes are different
from the 12-tet. In the next figure, the genus 3 2 • 5 • 7 is represented.

Adriaan Fokker, The genus 3 2 • 5 • 7

158
Euler-Fokker Genera

The drawback of the Euler-Fokker genera is that some genera are equivalent. For
example, the genus 3 • 7 2 built on C has the same notes as the genus 7 5 built on G. Then,
it is natural to ask: how many different genera can be built ? We have the following
result.

Theorem 165 Let (p, q, r) be the description in diesis of three constituent intervals of
the Euler-Fokker genus 3 a • 5 6 • 7C.
(1) There are exactly 156 different genera with (p, q,r) = (18,10,25), i.e. with a fifth of
18 diesis, a third of 10 diesis and a seventh of 25 diesis.
(2) If we substitute the major third by a minor third of 8 diesis, there are only 136
different genera, with (p,#,r) = (18,8,25).
(3) There are 125 different genera with (p,q,r) = (20,10,30).

Adriaan Fokker composed some musical pieces with the Euler-Fokker genera under the
pseudonym of Arie de Klein, for the tricesimoprimal organ: Canzone (3 2 • 7), Preludium
chromaticum (3-5-7), Improvisando (5 2 • 7), Tenacita and Canzonetta (3 • 7 2 ), Inventio
(5 • 72) and Capriccio (7 3 ). He has also commanded works to other dutch composers
through the Huygens-Fokker Foundation (Stichting Nauwtuisterendheid).
Henk Badings (1907-1987) has composed many 31-tet pieces. The Suite van kleine
stukken was written in 1954 with the Euler genus 3 3 • 5 2 Reeks van kleine klankstukken
for organ used 3 3 • 5 2 in Preludio, 3 3 • 5 3 (Siciliano), 5 3 • 7 2 (Perpetuum mobile), 5 2 • 73,
3 3 • 7 2 , 3 2 • 5 • 7 (Passepied) and 3 • 5 2 • 7.
Hans Kox (1930) also used Euler genera. Drei Stukken for violin is written in the 3 2 • 7 3
genus. Vue des Anges on a poem of Rilke (1959) is built on the 3 3 • 72 genus. The serenade
(1968) for two violins in five parts is written in the genus 3 • 7 2 in different transpositions.
The canon of the first part used two transpositions 3 • 7 2 (di -f£h) and 3 • 7 2 (ges-a£h)
included in the genus 3 • 5 • 7 2 (di-feh).
Jan van Dijk (1918) used many different genera in Musica per organo trentunisono.
Each of the twelve pieces is written in different genera: Prelude and fugue for oboe and
organ uses 5 • 3 4 (g, dis) 7 • 3 3 • 5 (ai, gis) 7 • 3 4 (ai, 6), the second piece Fugue for violin,
alto and organ uses: 7 3 • 3 3 (ai, d£h), 7 3 • 3 2 (ai, ei), 7 4 • 3 (ai, g), 7 • 3 4 (di, e), the third
piece Canon for organ: 7 5 (ai, ais), the fourth piece Fugue for flute, oboe and organ:
7 • 3 2 • 5 2 (ai, feh), The fifth piece 5a, Toccata for organ solo: 7 2 • 3 2 • 5 (ai, b£h), and the
fifth piece 5b, Fugue for oboe, violin and organ: 7- 3- 5 3 (ai, dih), the sixth piece Prelude
and fugue for organ solo: 7 2 • 3 • 5 2 (ci, b£h), the seventh piece, Corrente alia fuga for
organ solo uses 7 3 • 3 • 5 (ai, dis), the eighth 8a, Capriccio for flute in G and organ uses
7 • 5 4 (ai, b£h), the eighth part 8b, Fugue for flute in G and organ 7 2 • 5 3 (ai, ei), the
ninth piece, Duettino e Ciaconna for two flutes and organ: 7 3 • 5 2 (ai, bis) 7 4 -5 (ai, c)
and the tenth Basso solo for pedalier uses 7 5 (ai, ais).
The british composer Alan Ridout (1934-1996) also used Euler genera in his String
Trio (1961). The first movement Chants uses the genus 3 3 • 7 2 , the second movement
Dances uses the genera 5 2 • 7 3 and 3 • 5 2 • 7, the third and final movement takes back the
elements of the first part.
Many other composers have been involved in the 31-tet history, such as Anton de Beer
(1924), Jon Catler (1957), Ivor Darreg (1917-1994), Harold Fortuin, Eugen Frischnecht,
Jaap Geraedts (1924), Joop van Goozen (1960), Alois Haba (1893-1973), Anthon van der
Horst (1899-1965), Martinus J. Lttrsen, Joel Mandelbaum (1932), Peter Schat (1935), Al-
phonse Stallaert (1920), Paul Christian van Westering (1911) and Ivan Wyschnegradsky
(1893-1979), who wrote a piece for organ with the technique of non-octaviant spaces.

159
9. Contemporary Tunings

9.2 Farey Temperaments


Farey temperaments are defined from Farey sequences.

Definition 166 The Farey sequence Fn of order n is the increasing sequence of irre-
ductible fractions p/q (i. e. gcd(p) q) = 1) between 0 and 1 whose denominators do not
exceed n. The Farey temperament TFn is the shifted Farey sequence. It is defined by
adding one to each ratio of the Farey sequence, namely TFn = Fn + 1.

The Farey sequences are built recursively from the set

Fi = {0,l}

by inserting mediants of each neighboring pairs of rational numbers. If p/q and p'/q' are
such pairs, the mediant is denned by

P + rf
q + q'
The first Farey sequences are

1 X
F -1° \

,r_r0 1 1 1 2 3 1
4
MU's^'sU'i'
Remark that the Farey temperaments TFn are not the set of all rational numbers p/q
such that p and q are less or equal n. For example, in the Farey temperament TF5 we
have the number 7/4 which does not verify this property. However, each number x of TFn
is the mediant of two consecutive numbers of TFn-\. The Partch n-limit temperament
Hn verifies this property, if p/q G Hn then q <p <n. The main properties of the Farey
sequences are

Proposition 167 (1) Ifpi/qi, Pv/qi are two consecutive ratios p\/q\ <Pi/q2 then

\piq2~qiP2\ = 1

(2) Ifpi/qi, P2/92, P3/93 are three consecutive ratios p\/q\ <p^/q2 <Pz/qz then

P2 _ Pi +P3
q2 q\ + qz
(3) The number of terms in the Farey sequence of order n (i.e. the cardinal of the set
Fn) is
n
|FB| = l + 5>(fc)
fe=l
where ip(k) is the Euler totient function.

160
Farey Temperaments

The fact that the sequence of the denominators of a Farey sequence is palindromic, is
not obvious. It is proved by the embedding of the Farey sequences into the Stern-Brocot
tree. The Stern-Brocot tree is a binary tree obtained by iteratively inserting the mediant
of two adjacent fractions and starting with points 0/1 and 1/0. The relations between the
Stern-Brocot Tree and Well-formed Scales are described in [Clampitt, 1997] and [Carey,
1989].
The ratios on the left side of the tree are less than 1 and the ratios on the right side are
greater than 1. For the ratio on the left side, the place of each ratio p/q can be encoded
by a word describing the way to get the number starting from 1/1. Denoting L the left
movement and R the right movement, the word LLRL or simply L2RL represents the
ratio 3/8.
1
0

4 5 5 4 3 3 2 1

More generally, movements are associated with continued fractions. The function given
by the exponents rj of the dj (= L or R) components

f(a\\o%. ..<") = [ 0 ; r i , r 2 , . . . , r n _ i , l + r n ]

describes the continued fraction of the point associated with the word

a a
l 2 ~'an

For example, the word L 2 RL is associated with the continued fraction

f(L2RL) = [0; 2,1,2] = ^ - = f

1 +
2
The sum of the exponents minus 1 (fc = ri +... + r n — 1) is the row number in which the
fraction is located. For example 3/8 is located in the fourth row (2 + 1 + 2 — 1 = 4) of
the Stern-Brocot tree.

The Farey sequences (and then the Farey temperaments) are related to Ford circles.
The Ford circle C(p,q) is the circle centered at (p/q)±l/(2p2)) and of radius l/(2q2).
If we put the Farey ratios Pi/qi on the real line, we could draw the circles C(pi,<fc). If

161
9. Contemporary Tunings

Pi/Qi) P2/92, P3/93 are three consecutive terms of the Farey sequence, then the circles
C(pi,gi) and C(p2>#2) are tangent at the point a\ given by

„ _ (V2 gi l__\

and the circles C(p2>#2) and C(ps,qs) are tangent at the point 0,2 given by

U2 92(92+9*)'92+9$)
The point a\ lies on the circumference of the circle built on the two points of the real
line (pi/#i,0) and (p2/</2»0). The point a2 belongs to the circle built on (p2/<72,0) and
(^3/^3,0). If we plot all the fractions on the z-axis and their corresponding Ford circles
generated from a Farey series of order n, they form a set of tangent circles [Ford, 1938].

Remark that Farey sequences have been used by Ivan Wyschnegradsky in his research
on rhythmical figures of ultrachromatism [Wyschnegradsky, ed. 1996]. Another important
point is that phase transitions occur in Farey temperaments.

Phase transitions are singularities of variables such as free energy. For a dynamical
system characterized by a Hamiltonian H, we define the Gibbs measure by the probability
of density
e-PH
/(*) = Z
where /? is linked to the temperature T of the system, namely /? = 1/kT. The partition
function is
Z = I e-^*) dx
The internal energy of the system is the expectation of the Hamiltonian, considered as a
random variable
B
"-•*-- -W
The free energy is defined by

F = -kTlogZ = U-TS
The entropy is given by the following equations
f)F f)F
s = = fc(log z + m = km2
~ ar W
162
Harry Partch Tonality Diamond

and the calorific capacity is

*-(SX-^(W
Usually, phase transitions are singularities of the calorific capacity. For the Farey tem-
peraments, we will see that the free energy has a discontinuity atft= 2. Recall that the
Farey sequences are the set Fm of all rational fractions of [0,1] whose denominators are
less or equal m. The set Fm-\ is obtained from the set Fm by inserting between two
consecutive fractions pi/qi and p2/q2 the number

Pi +P2
tfl+42
For each m, let Fm be the Farey set

=
I'm l^lj^r-j^mj

and for each Xi = Pi/qi and Xi+i = pi+i/c^+i define the number

v% = Pi+i + Qi

Definition 168 The partition function Zm of the Farey set Fm is

i=l

Examples. The first values are

Zl = 2-f Z2 = 2~p + 3^ Z3 = 2-f3 + 3- A-?

Definition 169 The free energy of the Farey temperaments is defined by

FW)= lim Z l l o g ^ )
m->oo p m

Theorem 170 The free energy of the Farey temperaments F(fi) exhibits a unique phase
transition atft= 2.

Proof See the paper of P. Kleban and A.E. Ozluk [Kleban, 1998]. •

9.3 Harry Partch Tonality Diamond


Harry Partch (1901-1974) has found a theory of just intonation and new temperaments
based on the tonality diamond. To build the diamond, one has to choose an odd number,
for example 5. The diamond will be set upon the three numbers < 1 , 3, 5>. In each box of
the diamond matrix, we place a fraction of the form p/q such that the fractions of each
line of the diamond matrix have the same numerator, and the fractions of the opposite
line have the same denominator. So we have

163
9. Contemporary Tunings

1/3
1/5 5/3
1/1 5/5 3/3
5/1 3/5
3/1

Each ratio of the tonality diamond is then multiplying or dividing by a power of 2


in order to map each ratio in the octave. We then have the tonality diamond of 5-
limit. Taking the ratios and putting them in the ascending order, we get an heptatonic
temperament {1, 6/5, 5/4, 4/3, 3/2, 8/5, 5/3} in which each ratio is rather simple and
built from the prime numbers 2, 3 and 5. For musical reasons, Harry Partch has added to
the heptaphonic system five ratios taken in the monophonic fabric to set up the 5-limit
temperament {1, 16/15, 10/9, 9/8, 6/5, 5/7, 4/3, 3/2, 8/5, 5/3, 16/9, 9/5, 15/8}. The
5-limit temperament is described by the word

X = A • d • A1 with A = (abc)(aba)

where a, b} c, and d denote the intervals of the temperament and A1 the palindome of
A, A1 = (aba)(cba). More generally, we have the following definition.
Definition 171 Let p be a prime number. A p-limit system is a temperament in which
each ratio is a rational number whose denominator and numerator factors are products
of prime numbers less or equal p.
The Partch 7-limit temperament is built in the same manner on the numbers <
3,5,7 > and leads to the system of 13 degrees: {1, 8/7, 7/6, 6/5, 5/4, 4/3, 7/5, 10/7,
3/2, 8/5, 5/3, 12/7, 7/4}. The system is described by the palindromic word

X = A-g-Al with A = (abcdef)

where a = 8/7, b = 49/48, c = 36/35, d = 25/24, e = 16/15, / = 21/20, and g = 50/49.

The Partch 9-limit is built on the numbers < 3,5,7,9 > and leads to the 19 degrees
temperament {1, 10/9, 9/8, 8/7, 7/6, 6/5, 5/4, 9/7, 4/3, 7/5, 10/7, 3/2, 14/9, 8/5, 5/3,
12/7, 7/4, 16/9, 9/5}. If we denote the constitutive intervals by a = 10/9, b = 81/80,
c = 64/63, d = 49/48, e = 36/35, / = 25/24, g = 28/27, h = 21/20, i = 50/49, the
9-limit is the palindromic word

X = A'i-Ai with A = (abed) (efe) (gh)


The Partch ll-limit is the temperament of 29 degrees {1, 12/11, 11/10, 10/9, 9/8,
8/7, 7/6, 6/5, 11/9, 5/4, 14/11, 9/7, 4/3, 11/8, 7/5, 10/7, 16/11, 3/2, 14/9, 11/7, 8/5,
18/11, 5/3, 12/7, 7/4, 16/9, 9/5, 20/11, 11/16}. If we denote the constituent intervals
by a = 12/11, b = 121/120, c = 100/99, d = 81/80, e = 64/63, / = 49/48, g = 36/35,
h = 55/54, i = 45/44, j = 56/55, k = 99/98, I = 28/27, m = 33/32, n = 50/49, the
ll-limit temperament is given by the palindromic word

X = A-n-Ai

with
A = abcdef ghijklm

164
Ben Johnston Lattices

The Partch 13-limit is the temperament of 41 degrees {1, 14/13, 13/12, 12/11, 11/10,
10/9, 9/8, 8/7, 7/6,13/11, 6/5,11/9,16/13, 5/4,14/11, 9/7,13/10,4/3,11/8,18/13, 7/5,
10/7, 16/11, 3/2, 20/13, 14/9, 11/7, 8/5, 18/11, 5/3, 22/13, 12/7, 7/4, 16/9, 9/5, 20/11,
11/16, 24/13, 13/7}. If we denote the constituent intervals a = 14/13, b = 169/168,
c = 144/143, d = 121/120, e = 100/99, / = 81/80, g = 64/63, h = 49/48, i = 78/77,
j = 66/65, k = 55/54, / = 65/64, m = 56/55, n = 99/98, o = 91/90, p = 40/39,
q = 33/32, r = 50/49, the 13-limit is the palindromic word
X = A • r • A1
with
A = (abcdefghijk)c(lmnopq)c(o)
The Partch monophonic system is the 43 tone temperament that has been used by
Harry Partch for tuning the chromelodeon an instrument built in 1941. The ratios are
{1, 81/80, 33/32, 21/20, 16/15, 12/11, 11/10, 10/9, 9/8, 8/7, 7/6, 32/27, 6/5, 11/9, 5/4,
14/11, 9/7, 21/16, 4/3, 27/20, 11/8, 7/5, 10/7, 16/11, 40/27, 3/2, 32/21, 14/9, 11/7,
8/5, 18/11, 5/3, 27/16, 12/7, 7/4, 16/9, 9/5, 20/11, 11/6, 15/8, 40/21, 64/33, 160/81}.
The temperament is symmetric around 7/5 and is described by the word
X = A-i>A*
with
A= (abc)(jdef)(ajgj)(abdc)(hgj)(abc)

9.4 Ben Johnston Lattices


The American composer Ben Johnston (1926) has developed in Rational Structure in
Music a theory of forming temperament by lattices. Let (pi, £>2> •••> Vn) be n prime
numbers greater than 2. Each prime number represents a dimension of the n-dimensional
Johnston lattice. In one direction, the points of the lattice are the powers of pi replaced
in the octave by multiplying or dividing by a power of 2. The center of the lattice is the
fraction 1/1. For example, the lattice (3-5) is a planar lattice. The points of the horizontal
axis are the points derived from the sequence (..., 1/3 2 , 1/3, 1/1, 3, 3 2 , 3 3 , 3 4 , ...) by
powers of 2 (...., 16/9, 4/3, 1/1, 3/2, 9/8, 27/16, ...). The points of the vertical axis are
the points derived from (..., 1/5 2 , 1/5, 1/1, 5, 5 2 , 5 3 , 5 4 , ...) by powers of 2 (...., 32/25,
8/5, 1/1, 5/4, 25/16, 125/64, ...). Outside the axis, the pairs (x,y) are formed by the
pointwise multiplication x-y-2k (where k is such that the fraction belongs to the octave
[1, 2[). Once the lattice is built, Ben Johnston states selected rules to extract a subdomain
of the lattice. The fractions of this subdomain define the set of the temperament. For
example, the selection of a subdomain in the (3.5) lattice gives the scale (C 1, C|j 25/24,
D 9/8, E\> 6/5, E 5/4, Gb 36/25, G 3/2, A\> 8/5, A 5/3, B\> 9/5, B 15/8, C 2). This
temperament corresponds to the tuning group

25 1 312 K 34
G = < r = - , 1= ^ , 5 E E - >

The lattice (3 • 7) leads to (C 1, D\> 49/48, D 9/8, Eb 7/6, E 9/7, F 4/3, Pfl 72/49, G
3/2, Ab 14/9, A 12/7, Bb 7/4, B 27/14), which is the tuning group
49 1 _ 312 ^ _ 26
G = < r = 1 = 7 = >
4 8 ' 2^' 3 ^

165
9. Contemporary Tunings

The lattice ( 3 - 5 - 7 ) leads to the 22 degrees temperament (C 1, D\>' 28/27, D\>" 16/15,
D> 10/9, Db"' 8/7, E\>' 7/6, £b 6/5, E 5/4, £ b " 9/7, F' 21/16, F " 27/20, i^|' 45/32,
F&" 81/56, G 3/2, A\>' 14/9, 4b 8/5, 4 5/3, AV 12/7, £ b ' 7/4, 5b 9/5, 5 15/8, Bb"
37/14, C 2), which corresponds to the tuning group

2 35
G
28
1 = 5
_3^_ 7
f_
= < r = -
27' 322' 212' - 32 >

This lattice is used in One Man (1967) for trombone solo, Rose (1971) for choir, Mass
(1972) for choir, ten trombones, double-basse and percussions, and in the Fourth String
Quartet (1973). Works such as The Song of Innocence (1975) for soprano and instruments
and Two Sonnets of Shakespeare (1979) use the bidimensional lattice (3 • 11). Diversion
(1979) for 11 instruments and the Sixth String Quartet is built on the lattice (3• 5-7-11).
Duo for two violins (1978), the Fifth String Quartet (1979) use the five prime numbers
(3 • 5 • 7 • 11 • 13). The Microtonal Suite for piano (1978) and Twelve Partials (1980) use
the lattice (3 • 5 • 7 • 11 • 13 • 17 • 19) and the Ninth String Quartet, written in 1987, uses
a 10-dimensionnal lattice (3 • 5 • 7 • 11 • 13 • 17 • 19 • 23 • 29 • 31).

9.5 Ervin Wilson Tunings


Ervin Wilson has proposed a method for building temperaments named Combination
Product Sets (CPS). This procedure has been commented by many authors such as
John Chalmers, Paul Erlich, Brian McLaren and Paul Rapoport. Starting from a set of
generators G = [#1, #2> •••>0n]> the method consists in forming all the possible factors by
pointwise multiplication of p elements. The ( £ J factors are reordered and doubles are
eliminated. This leads to a new set A. The elements (#, y) of the Cartesian product Ax A
are then divided by themselves x/y and adjusted by a power of 2 to be in the octave
[1, 2[. The reordered set or CPS is denoted by

V n [#i>£2,-»,0n]

For example, to construct the hexany

2 4 [1,3,5,7]

we have to form the six products 1 • 1, 1 • 3, 1 • 5, 1 • 7, 3 • 5, 5 • 7.

1.1 |1 1.3 1.5 1.7 3.5 5.7


L3 1
1.5
rw
3/5
5/3
1/1
7/3
7/5
5
3
35/3
7
1.7 3/7 5/7 1/1 15/7 5
3.5 1/5 1/3 7/15 1/1 7/3
3.7 1/7 5/21 1/3 5/7 5/3
5.7 |13/35 1/7 1/5 3/7 1/1

Normalizing the ratios by the powers of 2 yields to the following numbers

166
Non-octaviant Tunings

1/1 7/6 5/4 35/24 5/3 7/4


1/1 21/20 6/5 7/5 3/2 7/4
1/1 15/14 5/4 10/7 3/2 12/7
1/1 7/6 4/3 7/5 8/5 28/15
1/1 8/7 4/3 10/7 5/3 40/21
11/1 8/7 6/5 48/35 8/5 12/7

Eliminating the doubles, we get a 19 degrees temperament. Many CPS have a great
number of degrees. But the CPS

3 9 [1,2,..., 9]

leads to a system with 15 degrees, and the CPS

4 8 [1,2,..., 8]

gives a 12 degrees temperament.

9.6 Non-octaviant Tunings


Most of the temperaments use the periodicity of the octave. Some authors have tried
to construct systems based on a period which is not an integer multiple of the octave.
This is the case of the Equal Temperament with Just Fifths (ETJF) of Serge Cordier,
in which the octave is augmented ((3/2) 1 2 / 7 , 1203 cents) to include all just fifths. The
interval between two consecutive notes is 100.26 cents. The table gives the ratios and the
values in cents of all intervals.

(3/2) 6 / 7 602
Notes
C
Freq.
1
Cents |
0 \nG 3/2 702
Ctf (3/2)^7 100 G« (3/2) 8 / 7 802
D (3/2) 2 / 7 201 A (3/2) 9 / 7 903
Dfl (3/2) 3 / 7 301 AJJ (3/2) 10 / 7 1003
E (3/2)4/7 401 B (3/2)n/7 1103
F (3/2)5/7 501 | C (3/2) 12 / 7 1203

Wendy Carlos has proposed equal temperaments which are good approximations to
the just fifth (3/2) and the just major third (5/4). The fifth is divided in 9,11 or 20 equal
parts leading respectively to the alpha, beta and gamma temperaments. These numbers
come from the expansion in continued fraction of the ratio
log(3/2)
= [1; 1,4,2,6,1,10,135,...]
log(5/4)
whose numerators and denominators of convergents (1, 2, 9/5, 20/11, 129/71, ..) lead to
the sequence (1, 2, 5, 9, 11, 20, 71, ...).

The alpha temperament is the system of 18 degrees built on the repetition of the interval
of ratio (3/2) a / 9 , about 78 cents. In this system, nine intervals (702 cents) approximate

167
9. Contemporary Tunings

the just fifth (3/2, 702 cents), five intervals (390 cents) approximate the just major third
(5/4, 386 cents) and four intervals (312 cents) approximate the just minor third (6/5,
316 cents). The ratio x verifying this approximation is given by the formula

(372 j (vi J (575)= 1


This equation has one solution x = (3/2)*/ 9 .

The beta temperament is the system of 22 degrees built on the repetition of the interval
of ratio (3/2) 1 / 1 1 , about 64 cents. In this system, 11 intervals (702 cents) approximate
the just fifth (3/2, 702 cents), six intervals (383 cents) approximate the just major third
(5/4, 386 cents)) and five intervals (319 cents) approximate the just minor third (6/5,
316 cents). The ratio x verifying this approximation is given by the formula

(372) (5/4J (e/sJ = 2


This equation has one solution x = (3/2) 1 / 1 1 .

The gamma temperament is the system of 35 degrees built on the repetition of the
interval of ratio (3/2) 1 / 2 0 , about 35 cents. In this system, 20 intervals (702 cents) ap-
proximate the just fifth (3/2, 702 cents), 11 intervals (386 cents) approximate the just
major third (5/4, 386 cents)) and nine intervals (316 cents) approximate the just minor
third (6/5, 316 cents). The ratio x verifying this approximation is given by the formula

\3/2j U/47 U/V "


This equation has one solution x = (3/2) 1 / 2 0 .

Refinements of these tunings are obtained by dividing each interval into two new
intervals. This leads to alpha prime, beta prime and gamma prime presented in the
following table.

Systems Card. Intervals Cents


alpha 18 (3/2) 1 /* 78
alpha prime 36 (3/2) 1 /i8 39
beta 22 (3/2)1/!! 64
beta prime 44 (3/2)V22 32
gamma 35 (3/2) 1 /20 35
gamma prime 70 (3/2) V40 18

The Bohlen-Pierce temperament is a non-octaviant system of 14 degrees in which the


octave (2) is replaced by an octave plus a perfect fifth (3/1). It has been proposed by
Heinz Bohlen in 1972, and rediscovered by John Pierce and Max V. Mathews in 1984. It
is built on the triad 7:5:3 instead of the usual 6:5:4. The intervals (3/1, 5/3, 7/3, 7/5)
and their inverses (9/5, 7/5, 15/7) given by the product of the ratios of the Bohlen-Pierce
tuning are presented in the following table.

168
Cyclic Tunings

1 No. Ratios A Cents | 1 No. Ratios A Cents


1 1 a 0 8 9/5 1) 1018
2 27/25 d 133 9 49/25 c 1165
3 25/21 a 302 10 15/7 b 1319
4 9/7 b 435 11 7/3 a 1467
5 7/5 c 583 12 63/25 d 1600
6 75/49 b 737 13 25/9 a 1769
7 5/3 a 884 1 1 14 3/1 1902 |

The Bohlen-Pierce tuning is represented by the word

X = (adabcb)a(bcbada)

with a = 27/25 (133 cents), b = 49/45 (147 cents), c = 375/343 (154 cents) and
d = 625/567 (169 cents).

9.7 Cyclic Tunings


Many irrational numbers have been used in artistic forms for their harmonic proportion,
beauty and strange fascination. Cyclic tunings are made by the repetition of any number
w, with the following procedure. Write the sequence of powers of the number w

{...., 1/w2, 1/w, 1, w, w2, ...}


then choose a limit k of the number of degrees (for example, 2/c + 1)

{w~~k,w~k+l, ....w - 1 ,l,iu, ....wk}

and normalize the ratios by a power of 2 to put them in the same octave [1, 2[. This leads
to new temperaments which have many common properties. One of them is the existence
of prime scales generated by only two intervals. The most famous examples of the cyclic
tunings are the Pythagorean tunings based on the pure fifth (w = 3/2) and the golden
scale based on the golden fifth (a fifth which depends on the golden ratio, solution of the
equation <>/ =(/)-{-1). But here, we will consider new tunings given by irrational number
w such as 7T = 3.14159..., e = 2.71828..., the Euler constant 7 = 0.5772156649..., the
Catalan number C = 0.9159655942, and the Feigenbaum number F = 4.6692106...

Pythagorean Tunings. The simplest Pythagorean scale (k = 1) based on the just


fifth (co = 3/2 = 702 cents) is the system composed of three notes {F, C, G} and which
is described by the word

^ 3 = C1C2C1

where c\ = 4/3 = 498 cents and c<i = co/ci = 9/8 = 204 cents. The next system
(k = 2) is made from the fifths {B\>, F, C, G, D} and composed of two intervals c<i and
c
3 = C1/C2 = 32/27 = 294 cents. The scale well ordered is made of 5 notes {C, D, F, G,
B\>} and represented by the word

P 5 = C2C3C2C3C2 = (C2C3)2C2

169
9. Contemporary Tunings

For k = 3, we get the seven degrees Pythagorean scale {C, D, E\>, F, G, A, B\>} repres-
ented by the word

Pi = c2c4(c2fc4c2
in which the two constituent intervals are related to the previous one by the relation
c4 = C3/C2 = 256/243. This system has only two constituent intervals c2 and c 4 . If we
continue the algorithm with k = 4, we construct the new scale {C 1, D 9/8, E\> 32/27,
E 81/84, F 4/3, G 3/2, A\> 128/81, A 27/16, B\> 16/9} which is described by the word

P9 = a(bcb)a(bcb)a

where the intervals a, b and c are given by a = 9/8, b = 256/243 and c = 2187/2048. We
see that this scale has three constituent intervals instead of two. It is not a prime scale.
The first scale which contains all the previous scales and has only two intervals is the
scale P12, involving the two intervals c4 = c$/c2 = 256/243 and C5 = C2/C4 = 2187/2048.
The following table shows the next steps. The first column gives the number of degrees
in the scale, the second and third columns are the numbers of constituent intervals. The
fourth column gives the relation of the constituent interval and the last column is its
value in cents.

3 2ci 1C2 C2= co/cj= 9/8 204


5 3c2 2c3 C3= CI/CJJ= 32/27 294
7 5C2 2c4 c4= c3/c2= 256/243 90
12 7c4 5c5 c 5 = 02/^= 2187/2048 114
17 12c4 5C6 C6= c5/c4= 3 1 2 / 2 i a 23
29 17c6 12c7 c7=c4/cfi=227/31' 67
41 29c6 12c8 c 8 =c 7 /c f i = 2 4 6 /3^ a 43
53 41c6 12c9 09= cs/cR= 2 6 5 / 3 4 1 2
65 53cg 12cio d o = CQ/C^ 3 5 3 /2 8 4 3.6

Golden Tunings. Golden cyclic scales are based on the golden fifth of 696 cents
_ 15-N/5
w= 2^—

The constituents of the prime scales are presented in the following table whose columns
have the same meaning as the previous one. The first constituent interval is c\ = 2/w =
504 cents.

3 2ci lc 2 c2= w2/2 192


5 3c2 2c 3 cz= ci/c2= A/w6 311
7 5c2 2c 3 ^4= cz/c2= 8/wb 119
12 7c 4 5c5 C5= c2/c4= w 7 /16 74
19 12c5 7c6 cg= C4/c5= 128/u> 45
31 19c6 12c7 c7= c5/ce= w™/2iL 28
50 31c7 19c8 c8= c6/c7= 2ls/w*i 17

170
Cyclic Tunings

Exponential Tunings. Exponential cyclic scales are based on the powers of the num-
ber e, the basis of the Neperian logarithms e = 2.71828... The constituents of the prime
scales are given in the following table (same meaning as previous). The first constituent
interval is c\ = e/2 = 531 cents.

3 2ci lc 2 c2 = 8/e* 138


5 3c2 2c3 C3 = ci/c 2 = e 3 /16 394
7 5C2 2c4 c4 = c 3 /c 2 = e 5 /128 256
9 7c2 2c5 c5 = c 4 /c 2 = e 7 /1024 119
16 9c5 7c6 eg = CB/CB = 2 1 3 /ea 19
25 16c6 9c 7 C7 = c 5 / Cfi = e /2" 3
16
100
34 25c6 9c8 c 8 = ci/Cf. = e 25 / 2 a d 81
34
43 34c6 9c9 C9 = C8 /c fi = e /2 4M 62
52 43c6 9c 10 cio = c 9 /c 6 = e /2° a
43
43
61 52c6 9cii en = c 9 /c 1 0 = e 5 2 /2'° 24

P i Tunings. John Harrison (1693-1776) is a British watchmaker who has studied the
7r-systems based on the ratio 21/271" of 191 cents. Here we studied the tunings built with
the powers of the number w. The constituents of the prime scales are presented in the
following table. The first constituent interval is c\ = 4/n = 418 cents.

[3 2ci lc 2 C2 =7T2/8 364


5 3c 2 2c3 c 3 = C I / G , = 32/TT 3 55
8 5c3 3c4 C4 = C2/C3 =7T5 /256 309
11 8c 3 3c5 C5 = C4/C3 =7T8 /213 254
14 Ilea 3c6 c6 = C5/C3 =7r n /214 200
17 14c3 3c7 c7 = c 6 /c 3 =?r 14 /223 145
20 17c3 3c8 c 8 = c 7 /c 3 = T T 1 7 / 2 2 8 91
23 20c3 3c9 c9 = c 8 /c 3 =7r20 /233 36
43 23c9 20c10 cio = c 3 /c 9 = 238/TT 2 3 19
66 43c 10 23cn e n = c 9 /ci 0 = T T 4 3 / 2 7 1 17

Eulerian Tunings. Made with the Euler constant

7 = 0.5772156649...

the Eulerian cyclic tunings are presented in the following table. The first constituent
intervals are c\ = 2j = 249 cents and c<i = I/27 2 = 703 cents.

171
9. Contemporary Tunings

3 2ci lo? c2 = l / 2 7 2 703


4 3ci lC3 C3 = Cl/C 2 = 1/47 3
454
5 4ci IC4 c4 = c 3 /c 2 = I/87 4 206
9 5C4 4c5 c 5 = c 4 /c 2 = I67 5 43
14 9c 5 5c6 c6 = C 2 / c 5 = 1/12879 162
19 14c5 5c7 c 7 = c 5 /c fi = l / 2 " / y 1 4 119
24 19c5 5c8 c 8 = c 7 /c fi = 1/2 10 . 7 19 76
29 24c5 5c9 c 9 = c 8 /c fi = l / 2 i a . 7 2 4 33
53 29c9 24c10 c 10 = c 9 /c fi = 2 2 3 .7 2 9 10
82 53cio 29cn di=c9/c10 = l/24V3 23

C a t a l a n Tunings. Catalan tunings are made from the powers of the Catalan number
C = 0.9159655942... The constituents of the prime scales are presented in the following
table. The first constituent interval is c\ = 1/C = 152 cents.

\n~ 2ci lea c2 = 2C 2 896


4 3C! ICB c3 = c2/c1=2C3 744
5 4ci IC4 c4 = ca/c, = 2C 4 592
6 5ci lc 5 c5 = c 4 /cj = 2C 5 440
7 6ci lc 6 c6 = CB/C! = 2C* 288
8 7cx 1C7 c 7 = CB/<4 = 2C 7 136
15 8c 7 7c8 c 8 = C l / c 7 = 1/2C 8 16
23 15c8 8c 9 c9 = c 7 /c 8 = 4C 1 5 121
31 23c8 8c 10 C10 = c 9 /c 8 = 8C 105
39 31c8 8cn en = Cio/c8 = 16C 31 89
47 39c8 8c 12 C12 = c n / c 8 = 32C 73
55 47c8 8c 13 cis = ci 2 /c 8 = 64C 47 58
63 55c8 8c ] 4 en = cis/cg = 128C 42

Feigenbaum Tunings. Feigenbaum tunings which occurred in fractal music, are built
on the Feigenbaum constant F = 4.6692106.... The constituents of the prime scales are
presented in the following table. The first constituent interval is c\ = F/A = 268 cents.
3 2ci 1C2 c2 =25/^2 664
4 3ci 1C3 c3 =c2/c1 = 27/F3 397
5 4ci 1C4 C4 =C3/Cl =29/F4 129
9 5C4 4C5 c5 =ci/c4 = F 5 / 2 n 139
13 9c 4 4C6 ce =c5/c4 =F9/220 10
22 13ce 9c 7 c7 =c4/c6 =229/F13 118
31 22ce 9c 8 c8 =c7/c6 = 2 4 9 / ^ 2 2 108
40 31ce 9c 9 C9 = C 8 / C 6 =269/F31 98
49 40ce 9cio cio = c 9 / c 6 = 2 8 9 / F 4 0 87
58 49ce 9cn en =c10/c6 =2109/F49 77

172
10
Groups and Symmetries

The symmetries that appear in music are related to group theory. Symmetries of periodic
patterns and big lattices such as tunings and temperaments are well described by multi-
dimensional crystallographic groups. The symmetry of a geometric object (i.e. a subset
of a metric space) is the set of all transformations which brings the object into the same
figure as the original or its mirror image. The set of all transformations that leaves an
object unchanged is called the symmetry group (or point group) for that object. These
transformations are generated by elementary symmetry operations which are reflection,
rotation and translation. They can be applied to any musical object. This is why each
musical object has a symmetry group, possibly reduced to the identity element.

10.1 Symmetry Groups


The parameters of music (e.g. pitch, dynamics, texture, tempo, form, etc.) can be ordered
and substituted by symbolic letters taken in the usual alphabet to form patterns that
reveal the symmetry group. For example, the palindromic arch form ABCBA can be found
in many musical objects, and it is well known that the Phrygian scale (stttstt) where s
is the semitone and t the tone is a palindrome of the Ionian scale (ttsttts). Numerous
examples of symmetries occur in Bach's Musical Offering and Goldberg Variations. In
the first movement of Bartok's Music for Strings, Percussion and Celesta, counterpoint
is arranged about the A axis. It has been observed that the John Cage' Sonatas and
Interludes are distributed from the second and third Interludes (S1S2S3S4I1S5S6S7S8I2
and S16S15S14S13I4S12S11S10S9I3), expression in which we have used the notation S n for
the n-th sonata and I m for the ra-th Interludes. We recall some definitions.

Definition 172 Let X be a set, F(X) be the free group and R be a set of elements of
F(X). Each element of R can be written as a finite product X\ ...XJI where each Xj for
j = 1, ...,n is an element of X or an inverse of an element of X. Let N be the smallest
10. Groups and Symmetries

normal subgroup of F(X) containing R. Then the factor G = F(X)/N is called the
presentation of the group G =< X \ R = 1 > determined by the generators X and the
relations R.

Examples. The infinite cyclic group

Coo=<t>

is generated by one element without relation. It represents the repetition of the same
pattern
p p p p p p p p p

The cyclic group Cn is a group of order n.

Cn=<r\rn = l>

This group is generated by the rotation r about a fixed axis of angle 2kir/n where
fc = 0 , l , 2 , . . . ( n - l ) .

The infinite dihedral group DQO is defined by

Doo=<x,y \y2 = (xy)2 = l>

The dihedral group Dn of order 2n

Dn =< x,y \xn = y2 = (xy)2 = 1 >

is the group of (proper) rotation (det(r) = +1) which leaves a regular dihedral n-gon
invariant.

The derived group G' of a group G is the group generated by the set of commutators
[x, y] = x~1y~1xy in G. If G = < X \ R > and H = < Y \ S > are two groups generated
by respectively the sets X and Y", then the direct product has the presentation

GxH=<X,Y | R,S,[X,Y]>

where [XyY] is the set of commutators [X, Y) = {[x,y],x € X,y e Y}.

10.2 Frieze Groups


A frieze is an infinite plane figure whose translational symmetries are just the iterates of
some translation along an axis. In the plane R 2 , there are only four non-trivial isometries:
rotation, translation, reflection and glide reflection. A reflection is a mirror symmetry.
A glide reflection is a translation followed by a reflection. Rotations and translations
preserve orientation (but not reflections and glide reflections). Rotations and reflections
have fixed points (but not translations and glide reflections).

Definition 173 A frieze group is a discrete subgroup of the euclidean plane whose trans-
lation subgroup is infinite cyclic.

174
Frieze Groups

Frieze groups are symmetry groups of infinite strip patterns. More precisely, frieze
groups are infinite subgroups of isometric transformations of the euclidian plane M2,
which map a strip (i.e. Rx[—1, +1]) and are discrete. All translations are parallel.
T h e o r e m 174 In the euclidean plane, there are exactly seven frieze groups.
We now describe these groups. The crystallographers notation uses two dimensional
symbols. The first symbol represents the element of symmetry perpendicular to the direc-
tion of the translation. The second denotes the element of symmetry parallel or perpen-
dicular to the direction of the translation. A number n means a n-fold axis. Sometimes
the notation starts with p for periodic, but generally is omitted. Symbol m (mirror) de-
notes a reflection and g (glide) represents a glide reflection. We denote xy the conjugation
y~xxy.
1. The group F\\ (symbol p l l , 11 or pi) is generated by one translation denoted by t
Fn =< t >
It represents the repetition of the same pattern

P P P P P P
For example, the same rhythm

n n. n. n n n
2. The group F\m (symbol plm or lm, or pm) is generated by a translation t and a
reflection
Flm=<t,r \r2 = l,tr = t>
It represents the repetition of the same pattern with reflection through the hori-
zontal axis
P P P P P P
b b b b b b
For example,

nnnnnnnn.
wwu wwv wu
3. The group Fmi (symbol ml or p/m) is generated by a translation t and a mirror
reflection perpendicular to the horizontal axis
Flm=<t,r \r2 = l,tr = t~1 >
It represents the repetition of the same pattern vertically mirroring at each step
p q p q p q p q p q p
For example,

nnnn.Dn.nn.
175
10. Groups and Symmetries

4. The group F\g (symbol pg) is generated by a translation t of two units and a glide
reflection
Flg=<t,r \r2 = t,tr = t>
It represents the repetition of the same pattern with reflection through the hori-
zontal axis followed by a translation of one step

P P P P P
b b b b b

For example,

nj ni ni n\
r w r cj" r u r u
5. The group F\i (symbol 12) is generated by a translation t and a rotation about a
2-fold axis
F12=<t,s | s2 = t,t8 =t~1 >
It represents the repetition of the same pattern with a rotation about 2TT.

P P P P P P
d d d d d d

For example,

nn.nn.nnn.nn
uuuuuuuuu
6. The group Fmm (symbol pmm) is generated by a translation t, a reflection and a
glide reflection

Fmm =< t,s,r | s2 = l,ts = t~\r2 = l , t r = t,(sr)2 = 1 >

It represents a mirroring in the horizontal and the vertical directions.

p q p q p q p q p q
b d b d b d b d b d

For example,

n nn nn nn n
u uu uu wu w
176
Polyhedral Groups

7. The group Frag (symbol pmg) is generated by a translation t of two units, a rotation
about a 2-fold axis and a glide reflection

Fmg =< t,s,r\s2 = l,ts = t~\r2 = t,tr = t, (sr)2 = 1 >

It represents the repetition of the same pattern with a rotation about 2ir.

p q p q p q p q p q
d b d b d b d b d b

For example,

nnn.nn.nnn
u wcr wu uu w
10.3 Polyhedral Groups
Definition 175 A polytope in W1 is an intersection of finitely many closed half-spaces
that is bounded and has non-empty interior. Polytopes are named polygons {when n = 2)
and polyhedra (when n = 3).
Definition 176 A flag is the n-tuple (F^F\^.^Fn-i) where Fk is a closed face of di-
mension k such that Fk-i C Fk for k = 0,1, ...,n — 1.
Proposition 177 A polytope with no holes (of genus 0) f faces, v vertices and e edges
verifies the Euler's formula
v-e+f=2
Definition 178 A polytope P is regular if the symmetry group of P acts transitively on
the set of its flags.
It can be shown that P is regular if all its faces are isometric regular polytopes in
dimension n — 1 and all its dihedral angles are equal. In the space M3, convex regular
polyhedra are called platonic solids.
Definition 179 The polyhedral group of the polytope P o / R n is the symmetry group
of P.
Definition 180 Let P be a polytope ofRn. The sequence {pi,p2j--->Pn-i} where p^ is
the number of k-dimensional (or (k — l)-dimensional) faces of the polytope F^+i con-
taining in the face Fk-2 (with F-i = 0, and Fn = P) is called the Schlafli symbol.
It can be shown that the presentation of the polyhedral group as the following form.
Let P be a polytope of W1. li Pk denotes the fc-th Schlafli symbol, the presentation of
the polyhedral group is given by

G =< n , . . . , r n | (rm)2 = 1, for \k - l\ > 2


(rkrk+\)2 =Pk for k = l , . . . , n - l >

177
10. Groups and Symmetries

Polyhedral groups are Coexeter groups. In the space M3, the polyhedra are designed by
the Schl&fli symbol {p, g}, in which q > 3 is the number of regular p-gons (p > 3) that
meet at any one vertex of the polyhedron. The polyhedral group is presented by

G =< r, s | rp = sq = (sr)2 = 1 >

It has two generators: a rotation r about axes passing through a q-iold vertex and a
rotation s about axes passing through the mid-point of a p-fold face from the center of
the polyhedron.

Theorem 181 In the euclidean space R3, there are exactly five convex regular polyhedra,
named platonic solids. / / G is the polyhedral group of the polyhedron P given by the
Schlafli symbol {p, q} with \G\ /p faces, \G\ jq vertices and \G\ /2 edges, then we have the
Poincare- Wyle formula
1 1 ^ 2
p q \G\
The characteristics of the platonic solids are as follows

Names {P.?} \G\ G


Tetrahedron {3,3} 12 T =<r,s \r3 = s 3 = (sr)2 = 1 >
Octahedron {3,4} 24 0 = < r, s | r 3 = s4 = (sr)2 = 1 >
Cube {4,3} 24 0 =< r, s | r 3 = s4 = (sr)2 = 1 >
Icosahedron {3,5} 60 Y =< r, s | r 3 = s 5 = (sr)2 = 1 > |
Dodecaheron {5,3} 60 Y =< r, s | r 3 = s5 = (sr)2 = 1 >

10.4 Coxeter Groups


In 1934, H. Coxeter showed that all reflection groups in a n-dimensional euclidean space
are Coxeter groups and proved, the following year, that every finite Coxeter group is
isomorphic to some reflection group whose elements have a common fixed point. The
classification of the finite Coxeter groups has been achieved in 1935.

Definition 182 A Coxeter group G is defined by the following presentation

G=(ru...rn | (rirj)m* = l)

where mij is a matrix called the Coxeter matrix of integers verifying ma = 1 (so that
r\ = 1) and mij = rriji (for j ^ i) is an integer > 2 or equal to oo (in this case, there
is no relation between r^ and rj). The integer n is the rank of the group and mij are its
exponents.

The polyhedral groups are Coxeter groups. The symmetric group Sn is a Coxeter group
with ri = (i,i + 1) for i = 1,2,..., (n — 1). Every group generated by two elements is a
Coxeter group.

Coxeter group are represented by Dynkin diagrams (or Coxeter graphs).

178
Crystallographic Groups

Definition 183 A Dynkin diagram is a graph with n vertices, one for each generator
and edges defined by the following rules
(1) a thick edge (or an edge labelled with oo) if mij = 0,
(2) no edge between i and j if m^ = 2
(3) an edge between i and j if m^ = 3
(4) a (mtj — 2)-fold edge between i and j ifrnij > 4.

The classification of the finite Coxeter groups is presented in the following table
Groups Exponents | | Groups Exponents
An,n > 1 1 , 2 , . ..,71 #8 1,7,11,13,17,19,23,29
Bn,n>2 l,3,..,2n-l F4 1,5,7,11
Dn,n>4 l,3,..,2n- l , n - l Hz 1,5,9
-^6 1,4,5,7,8,11 HA 1,11,19,28
E7 1,5,7,9,11,13,17 | 1 GJ\m>5 1,771 — 1

10.5 Crystallographic Groups


The classification of crystallographic groups was achieved at the end of the 19 th century
by E.S. Fedorov, Pricke and Klein in 1897, a little bit later by A. Schoenflies and in 1924
by Polya and Niggli. In 1910, L. Bieberbach gave a solution to Hilbert's 18 th problem by
demonstrating that for any n, there are only finitely many n-dimensional crystallographic
groups (up to equivalence). There are 17 planar and 230 spatial crystallographic groups
(219 without orientation preserving affine transformations).
Definition 184 A fundamental domain of a topological space X for the action of a
topological group G on X is a subset D C X that is the closure of an open connected
subset and is such that the subsets gD for g £ G have pairwise no common interior points
(g{Int{D)) C\D = 0) and form a locally finite covering of X. If G is a discrete group of
transformations of a topological space X, a fundamental domain of G is a subset D C X
containing one representative for each orbit of G.
Definition 185 A crystallographic group r is a discrete group of an Euclidian space
E n of dimension n having a bounded fundamental domain.
Definition 186 Two crystallographic groups T and V of W1 are equivalent if they are
conjugate in the group of affine transformation A, i.e. there is an element g of A such
that
T'=9-lTg
Theorem 187 Two crystallographic groups V and V are equivalent if and only if they
are isomorphic.
Theorem 188 The group T is a crystallographic group iff the space En/T is compact.
Definition 189 The translations T in the group of affine transformations A of the n-
Euclidiean space En is a normal subgroup of A and the quotient A/T is the group of
linear transformations. Let T be a subgroup of A The translations in T form a normal
subgroup T = T n l \ The quotient G = Y/T is called the symmetry group or the point
group ofY.

179
10. Groups and Symmetries

T h e o r e m 190 Let n be a positive integer n € N*. There are finitely many isomorphism
classes of finite subgroups in GL(n,Z).
Definition 191 A T-subgroup of a group G is a normal abelian subgroup TofG without
torsion and such that G/T is finite and T is equal to its own commutant

T={geG,yt€T,gt = tg}

The definition of a crystallographic group can be generalized as follows.

Definition 192 A group T is a crystallographic group ifT has a free abelian T-subgroup
of finite type. The rank ofT is called the dimension ofT.
The group T is the set of all parallel translations in the crystallographic group T. It
is a normal subgroup of finite index isomorphic to Z n . T is the lattice. The symmetry
group G = T/T preserves the lattice T. Relative to a basis of T, the elements of G
are represented by matrices with integers entries. The description of the n-dimensional
crystallographic groups is then reduced to the description of all finite groups of squares
matrices with integer entries of M{n, Z). The problem is then to compute the cohomology
group H\G,RnIZn).
P r o p o s i t i o n 193 Two cohomology classes define equivalent crystallographic groups if
and only if they are transformed into one another by the normalizer of G in GL(n, Z).
T h e o r e m 194 Let G be a symmetry group of a crystallographic group T. The natural
homomorphism
H\G,Rn/Zn) - • iJ 2 (G,Z n )
is an isomorphism.

Definition 195 A Bravais group is a subgroup of GL(n, Z) that are lattice symmetry
groups.
The Bravais groups are the stabilizer subgroups for the natural action of GL(n, Z) on
the set of positive definite quadratic forms in n variables. The number of Bravais groups
are 1 (n = 1), 5 (n = 2), 14 (n = 3), 64 (n = 4), etc.

Planar Groups
Definition 196 A discrete subgroup G of the Euclidean plane with its translation group
isomorphic to Z 2 is called a plane crystallographic group or a planar group.
Definition 197 A tiling of the plane is a covering ofR2 with non-overlapping polygons.

Definition 198 A crystal system is a group of transformations that leave a lattice in-
variant. It is characterized by the repetition of a unit cell.

T h e o r e m 199 In the euclidean plane, there are 4 crystal systems and 5 Bravais groups.
The unit cell is defined by two axes of vectors (a, b) and an angle a between the axis
fa).

180
Crystallographic Groups

In the Euclidean plane, the four crystal systems are

1. Oblique (general parallelogram lattice) \a\ ^ |6| and a = (a, b) ^ 7r/2.

2. Rectangular for \a\ ^ |6| and a = n/2.

3. Square for \a\ = \b\ and a = 7r/2.

4. Hexagonal for |a| = |6| and a = 2TT/3.

By adding to the primitive unit cell of the four crystal systems an atom centered, we see
that only rectangular lattices can be centered. Thus there are five different symmetry
types of plane lattices, the five Bravais groups. This new class is called rhombic or centered
lattice. Another way to show that there are only five Bravais groups is to show that in the
euclidean plane, repeating patterns could not have 5-fold symmetry. Let 0 be a point of
the lattice. This point is an axis of n-fold symmetry. Any non-trivial rotation is a power
of the rotation r = r(0,27r/n) where O is the center of r. Let A, A' be two points in
the neighborhood of O. As the lattice is isomorphic to Z 2 , we could choose OA = 1,
OA! = — 1. Let a be the angle between A and B = r(A). Denote B' the point r~1(A).
B and B' belong to the same row, so BB' = 2cos(a) is an integer and can only take
on values -2, - 1 , 0, 1, 2. Thus cos{a) can only be equal to - 1 , -1/2, 0, 1/2, 1, that is
a = 27r/n with n = 1,2,3,4,6. So there are only five types of symmetry. Materials with
5-fold symmetry (which are not repeating patterns) are called quasicrystals.

Theorem 200 In the Euclidean plane, there are exactly 17 planar crystallographic groups.
Five of them are orientation preserving subgroups and tillings of the plane. The twelve
remaining subgroups are orientation reversing.

We now describe the 17 planar crystallographic groups. The first five groups are the
symmetry groups of tesselations of the plane. The others are the twelve orientation
reversing groups of the plane. We denote the conjugation ax = x~lax.

1. The first group (symbol pi) is generated by the translations a and b (of vectors a
and b) in the oblique crystal system

Pi = (a, b | ab = 6a)

ru r n
r7Ji
inWi J"i J\ ' " /

R
2. The second group (symbol p2) has a 2-fold symmetry in the oblique crystal system

p2 = (a, 6,5 | ab = ba, s2 = 1, as = a" 1 , bs = 6" 1 )

181
10. Groups and Symmetries

3. The third group (symbol p3) is a trigonal lattice with a 3-fold symmetry

p 3 = ( a , 6, s | ab = 6a, ss = 1, a s = a _ 1 6, 6s = a" 1 )

4. The fourth group (symbol p4) is a square lattice with a 4-fold symmetry

p 4 = ( a , 6, s | a6 = 6a, s 4 = 1, as = 6, 6s = a" 1 )

rT
i-L . "U
nn
•u
a 6 D

h A
0

f'
3
w 0

r0
a 0
J) r P

h1- °i—Di

, , — • «
*
^
4
f

r*
K 3
^0
h
MV*
p J!i p
r # of

ri ^L 1 Lh/ h V fp >J rp

5. The fifth group (symbol p6) is an hexagonal lattice with a 6-fold symmetry

P 6 = (a, b, s | ab = 6a, s 6 = 1, as = b, bs = a _ 1 6)

W
•0..
k\M N\/&\ /A "\/A ; /K\ /I \ n
••<>• 6 - ^^ • A H-.L5&

182
Crystallographic Groups

6. The first group (symbol pm) of the twelve orientation reversing groups is a rectan-
gular lattice with a mirror symmetry

G\ = {a,b,r \ ab = ba, r2 = 1, ar = a, br = 6" 1 )

n7 J> w J71
rp rp rp rp

JJ JJ> Ji> J;>

7. The group G 2 (symbol cm) is a centered lattice with parallel mirror and glide planes

G\ = (ayb,r \ ab = 6a, r2 = 1, ar = 6, br = a)
LiJiLiJjLil

JiLUILil

__.
JlLlJiL!

JJ> J>
rp rp rp
JJ
rt rp rp
—J —J JJ> J> J>
~""> "~" "-il rp rP rp

The group G\ (symbol pg) is a rectangular lattice with glide planes

G\ = (a,b,r \ ab = ba, r2 = a, a r = a, 6 r = 6 _ 1 )

u7 JJ>

rp rp

j> JJ>

9. The group G\ (symbol pmm) is a rectangular lattice with perpendicular mirrors


and a 2-fold symmetry

r19 — (/ a
A r > 52 I a& = &r a> fi2 =
1) a* = a_1
> &5 = fr-1>
° ~ r = 1, a = a, 6' = b~\ (sr)2 = 1

183
10. Groups and Symmetries

rp j ?r j rp jpr irp 1
'jjr>jT"'jJilJT"jJ!'
'"""T''pr'l'"rp"]lr''['rp''r
iS'EMffliSJ
J.. D r i rp i pr ! rp i

10. The group G\ (symbol pmg) is a rectangular lattice with mirror, glide reflection
and a 2-fold symmetry

n2 _ I a, 6, r, s | a6 = 6a, s2 = 1, a3 = a \ b3 = 6 *,

I LrmJ Lirxl iJ I Al I AM
I r^"] j —-y j~| ipr ! I
7JT lW UJH
I
L j. H .—j
i i
j.
\ I
-.—I
rp I i rp ! i rp I
L L ..........J..... . . . . . . ( . . . — . . . . : . . . . . . — i - J
fiji i'xii r
' ! PM i Pr i I
11. The group G | (symbol cmg) is a centered lattice with alternating mirrors and glide
planes in both directions and a 2-fold symmetry

r& _ / Q>,b>r,s\ ab = ba> s2 = 1, a3 = a *, bs =b *,


2
° ~ \ r 2 = 1, ar = 6, br = a, (sr) 2 = 1

i i —i Al !JJ> i
r — ! r \"\
. ".j..,. \ j ..,.

.=!. - ji> j «w N-M


• r - | —i
i
rp ! pr | rp
i
J JUI JJ> <
j! T 1
1 prj r.P J_J

i i i I

12. The group G\ (symbol pgg) is a rectangular lattice with perpendicular glide planes
and a 2-fold symmetry

_ I a,6,r,s | ab = 6a, s 2 = 1, a3 = a *, 6s = 6 *,
Gi =
r 2 = a, ar = a, 6 r = 6" 1 , (sr) 2 = 6

184
Crystallographic Groups

1 1 4 1 i -1
00
Ill ^ j [L \ fp \

r-.j.
;"ZJ'1
l.cr„...j
p^j-j |
....{...CTJ K1 i*lj K... ^1 VI
=i.j |.,.m] L"n..l r_pj - . j :P...•^ 1 rp I
J
ir-.Lt^...i..nr...i |....r-".j P f l J > l Pr ^ i p r l
LciL^J IL •*J 1 rp 1
1 i J>l 11] !|

13. The group G\ (symbol p3ml) is a triangular lattice with an horizontal mirror plane
and a 3-fold symmetry

i _ / a,b,r,s \ ab = ba, s6 = 1, as = b, b" = a 1b,


G\ „2 =_ 1, ar = a, br = ab'1, (sr)2 = 1

\ A: /i\ >x / r ';A: rjr A ; rt

1
/•>A ! / ^ « \ , {#?
9 g
, ! , .£A? , '
^ K-f i rWr

14. The group G 2 (symbol p31m) is a triangular lattice with bisector mirror planes
and a 3-fold symmetry

r23 — / °"> ^ , r2 ' s I °^ =r ^ a ' s6 =r •'•' a * = ^2 ^ = a_1


^
° " \ r = 1, a = 6, 6 = a, (sr) = 1

r\/£Kr/fl \/Ami/il
V^wA^Aw/^ ^ \ j / \ A ; \ I /\y
( ^-.=^f ^~~ L ^-M- m*
L\///\\7&\//A\/4
)f'-A..J ^•^•••1 j^....V.j -V--;

A^A^I A 4 A
'-=& *-=&

15. The group G\ (symbol p4m) is a square lattice with horizontal mirror planes and
a 4-fold symmetry

_ I a,b,r,s | ab = ba, s4 = 1, as = b, b3 = a 1
,
G\ =
r2 = 1, ar = a, V = b'1, (sr)2 = 1

185
10. Groups and Symmetries

... ?•—o--E - -6- - \- o.- T - 6 - i—,a- -1 - •

^4?.r^^f:4,iP.i?.f\i
» v #1 * V * I * \ * i * •' * i P \ * i * s * I

I v ? \ > j 1 > , ' T \ ; i i : } / T V il

16. The group G\ (symbol p4g) is a square lattice with glide planes and a 4-fold
symmetry
_ / a, 6, r, 5 I aft = 6a, s 4 = 1, as = b, bs = a ,
ci = r 2 = a, a r = a, 6 r = 6 _ 1 , (sr) 2 = 1

f\~7^~ : ^.>^.j ; '\^r^,>


LXX >PTxiM J^i/*4IJ^I/^L/SILAM
\K X'X : K XN frxN: J??TM:MS /% M
[y^iji\\/^\j\\r^\j\\
fr i<-X •*• -v)k tjK*;* ••••!•&" KMrV'"tl
X X •XX X I X
J. .\\ .) \ ./•! V • i \ y~! •«? < 1 \ V*1 V /(
/Sii AM>Mi^!^SiiAI»|
k X x?x'XjX
XrHiXIXiXX [K /^i/^J/>\j/SU AM
b-7 ..^...^...^...^...^
X i ; ;. .^. ^! ./.\^U x. ^^. /. \^
t ^. ; R

17. The group Gj (symbol p6m) is an hexagonal lattice with a 6-fold symmetry

1 _ / o, 6, r, a I 06 = 6a, s 6 = 1, a* = 6, 6s = i
Gl r 2 = 1, ar = b, br = a, (sr)2 = 1

K- K'
^

^ ^ j ^ ^
•nr-' •&•• '-' A....' hr.:...' r> ?.....*

Space Groups

The main result of the classification of crystallographic groups in the space (n = 3) is


the following
Theorem 201 In the Euclidean space, there are 7 crystal systems, 11 Laue classes, 14
Bravais lattices, 32 crystal classes, 73 finite space subgroups, 219 orientation reversing
groups and 230 space groups.

186
Crystallographic Groups

In the international notation of Hermann-Mauguin, the n-fold rotation about an axis


of vector u counterclockwise is denoted by n. The inversion J is denoted by x and the
rotation inversion IR of angle 27r/n about a vector u is denoted by n. The Schoenflies
notation is based on the words: C {Cyclic), S {Symmetry axis), D {Dihedron), T {Tet-
rahedron), 0 {Octahedron), h {horizontal), v {vertical), Y and Yft are the symmetry
groups of the icosahedron. There are only five finite subgroups of SO (R 3 ) of order n > 1
of presentation
G=(a,b\an = bm = {ab)k)

The axis order {n, m, k} is characteristic of the group G: Cn = {n, 0,0}, Dn = {n, 2,2},
T = {3,3,2}, O = {4,3,2} and Y = {5,3,2}. Combining with the inversion, we get nine
types of symmetry groups of finite order denoted by Klim notation Cni, Cnp = {2n, 0,0},
Cnv = {n,2,2}, Dni, Dnp = {2^,2,2}, Tu Tp = {4,3,2}, Ou Yi and three groups of
infinite order Cooi, Ax>v> A»i- The fullerene discovered in 1985 by Kroto and Smalley is a
molecule of 60 carbon atoms which belongs to the icosahedral inversion group Yi = YxCi.

The seven crystal systems are characterized by the coordinates of the unit cell of axes
{a, b, c) and of interaxial angles {a, /?, 7).

1. Triclinic. This crystal has no rational symmetry other than 1 or 1. Its symmetry
group is the group C%.

2. Monoclinic. The symmetry elements are the 2-fold rotation (2) and/or the re-
flection (2 or m). In this case, the interaxial angles verify a = j3 = n/2 and
a = 7 = 7r/2. The symmetry group is C<n.

3. Orthorhombic. Orthorhombic systems have three perpendicular 2-fold axes or mir-


ror planes, or two mirror planes and one 2-fold axis. The interaxial angles verify
a = ft = 7 = 7r/2. The symmetry group is D^i-

4. Trigonal Trigonal systems have a single 3 or 3 symmetry element. In this case,


\a\ = \b\, a = P = 7r/2 and 7 = 27r/3. Special cases are the rhombohedral systems
defined by \a\ = \b\ = \c\, and a = ft = 7. The symmetry group is D^.

5. Tetragonal. Tetragonal systems have one 4-fold or one 4 symmetry element. The
axes \a\ = \b\ and the interaxial angles verify a = fi = 7 = 7r/2. The symmetry
group is D4i.

6. Hexagonal. Hexagonal systems have a single 6 or 6 symmetry element. In this case,


\a\ = \b\, a = f3 = 7r/2 and 7 = 27r/3. The symmetry group is D^.

7. Cubic. Cubic (or isometric) systems have four 3-fold or 3 axes. In this case, \a\ = \b\,
and a = ft = 7 = 7r/2. The symmetry group is 0{.

By adding to the primitive cell of the seven crystal systems (P) an atom body centered
(I), one face centered (A, B or C) or to each faces, all face centered (F), special centered
(R) we define the Bravais lattices. There are exactly 14 Bravais lattices. Associated to
each Bravais lattice, there are 32 crystal classes or symmetry groups presented in the

187
10. Groups and Symmetries

following table.

| Crystal Class Bravais Symmetry Groups


| Triclinic P 1,1
Monoclinic P,C 2, m, 2/m
Orthorhombic P, C, F, I 222, mm2, 2/m2/m2/m
Trigonal P, R 3, 3, 32, 3m,32/m
Tetragonal P,I 4, 4, 4/m, 422, 4mm, 42m, 4/m2/m2/m
Hexagonal P 6, 6, 6/m, 622, 6mm, 6m2, 6/m2/m2/m
[ Cubic P,F,I 23, 2/m3432, 43m, 4/m32/m

10.6 Bravais Temperaments


Bravais temperaments are built from Bravais lattices. Let (pi, £>2> P3) be three prime
numbers greater or equal than 3 (pj > 3). At the place of each atom of the Bravais
lattice of coordinates (x, y, z), we associate the frequency ratio r given by

where the number p is chosen such that the ratio r belongs to the octave [1,2[. The
temperament thus obtained has the symmetry group of the Bravais lattice. For instance,
we consider the orthorhombic lattices. The orthorhombic P cell is a rectangular box with
atoms at each vertex. Suppose that the unit vectors have lengths a = 1, b = 2 and c = 3,
and are associated with the three prime numbers p\ = 3, P2 = 5, ps = 7. Each point
of the eight points of the unit cell is associated to a ratio 3X • 5y • 7*, as in the Euler or
Johnson lattices. The computation of all ratios leads to the following table

Coordinates Ratios Cents | | Coordinates Ratios Cents


(0,0,0) 1 0 (0, 3, 2) 6125/4096 697
(1.3,2) 3 • 53 • 7 7 2 1 4 199 (1, 0, 0) 3/2 702
(1.0,2) 3 • 77128 240 (0, 0, 2) 49/32 738
(1, 3, 0) 3 • 57256 661 J 1 (0, 3, 0) 125/64 1159

In the orthorombic C, we have to add two ratios placed at the center of the horizontal
faces. In the (a,b,c) basis, these points have coordinates (1/2, 3/2, 0) and (1/2, 3/2, 2).

188
Sporadic Groups

They give two ratios 3 1 / 2 • 5 3 / 2 /16 of 330 cents and 3 1 / 2 • 5 3 / 2 • 7 2 /512 of 1068 cents. In
the orthorhombic I lattice, we have to add a point in the center of the box of coordinates
(1/2, 3/2, 1) of 99 cents. In the all face centered orthorhombic F, the points (1/2, 3/2,
0) of 330 cents, (1/2, 3/2, 2) of 1068 cents, (1/2, 0, 1) of 720 cents, (1/2, 3, 1) of 679
cents, (0, 3/2, 1) of 348 cents and (1, 3/2, 1) of 1050 cents are added.

10.7 Sporadic Groups


The classification of the finite simple groups has led to the conclusion that every non-
abelian finite simple group is isomorphic to an alternating group of at least five letters, a
Lie group or one of the 26 sporadic groups. Five of these 26 groups were discovered by E.
Mathieu in the 19 th century. The five Mathieu groups are representable as permutation
groups. They are also representable by automorphisms of Steiner systems. A Steiner
system S(£,m,n) is a set of n elements in which there is a system of (™)/(7) subsets
called blocks, consisting of m elements of the set and such that every set of £ elements
is contained in one and only one block. A Steiner automorphism is a permutation of
S(£, m, n) which brings blocks into blocks. The Mathieu groups are the groups of the
Steiner automorphism.

Mathieu groups Steiner System Order of M


Mil S(4,5,ll) 7920
M12 S(5,6,12) 95 040
M22 S(3,6,22) 443 520
M23 S(4,7,23) 10 200 960
M24 S(5,8,24) 244 823 040
It has been noticed by W. Berry [Berry, 1976] that the Mathieu group M12 has been used
by Olivier Messiaen in lie de feu 2 for piano. The group is generated by two permutations
called the Mongean shuffles of twelve tone rows

{6, 5, 7, 4, 8, 3, 9, 2, 10, 1, 11,0}

and
{5, 6, 4, 7, 3, 8, 2, 9, 1, 10, 0, 11}
These two rows are described by the knot -D358 which is also used by Anton Webern in
his Symphony} opus 21.

10.8 Constellations and Cacti


In his famous Esquisse d'un programme, A. Grothendieck introduced the concept of dess-
ins d'enfants (children's drawings) as a special case of a map on a topological oriented
surface. The theory of dessins d'enfants is an attempt to link the theory of Riemann sur-
faces and Galois theory. Grothendieck studied the action of the Galois group Gal(Q | Q)
on the set of dessins d'enfants. It originates in the Belyi theorem which relates Riemann
surfaces defined over Q with meromorphic functions having three critical values {0,1,00}.
In the planar case, the Belyi functions are rational functions. So for each combinatoric

189
10. Groups and Symmetries

planar map, there is one associated rational function. The action of the Galois group on
maps has various invariants. One of them plays an important role: it is the cartographic
group. This group is directly readable on maps (that is graphs drawn on two-dimensional
manifolds). A map is a very concise presentation of permutations groups. For more de-
tails, see [Zvonkin, 2004]. Some musical works are based on the user of permutations:
Olivier Messiaen's Chronochromie, He de feu #, etc. Iannis Xenakis' Metastasis, etc.
To understand the power of permutations and to give a classification of the cartographic
groups, we review the elementary combinatoric theory of maps. This leads to the map of
the Mathieu groups used by Olivier Messiaen.
Definition 202 Let Sn be the symmetric group that acts on n objects. A constellation
of length k and degree n or a k-constellation is a sequence [<JI,<J2, •••0'fc] where Gj G Sn
verifying
(1) the group G = < cri,<72, ...,<7fc > acts transitively on the set of n points
(2) the product G\G<i—<Jk is the identity
GiG2~-crk = id

The group G is called the cartographic group.


Definition 203 Two constellations [c"i, cr2,...cr*.] and [ c i , ^ ) •••crfc] are isomorphic if
they have same length k, same degree n and if the elements Gj and G'J are conjugate
Gj = h~lGjh for all j = 1,2,..., k.
Definition 204 Let C = [cri,<72,...<Tfc] be a constellation The passport of C is the se-
quence [pi,...,Pfc] of partitions of n, where eachpj for j = 1,2,..., k is the cycle structure
of the permutation Gj.
Example. The permutation (1, 2, 3, 5) (4,6) (7,8) (9) (10) contains one cycle of length 4,
two cycles of length 2 and two cycles of length 1. The corresponding cycle structure is
(4, 2, 2, 1, 1). It is a partition of n = 10 and is denoted by 42 2 1 2 .
Definition 205 A combinatorial map is a ^-constellation [cri,cr2,cr3] where G<I is an
involution, without fixed points.
Example. Each map is represented by a drawing. The edges are labelled twice at each
end. It is possible to read the three permutations on the graph.

The permutation <72 associated with the four edges is made of the cycles of each labeled
edge.
a 2 = (l,2)(3,4)(5,6)(7,8)
The permutation G\ is given by the four vertices. It is made of the cycles composed by
the number around each vertex
a 1 = (2,5,3)(l,7,6)(4)(8)

190
Constellations and Cacti

The permutation a^ is associated with faces. Faces are the resulting areas after a cutout
along the graph.
a 3 = a r 1 ^ 1 = (l,3,4,5,7,8)(2,6)

Definition 206 Let V be the number of vertices, E the number of edges and F be the
number of faces. The Euler formula of a map verifies

V-E +F = 2-2g

The number g is called the genus of the map. A map of genus g = 0 is called plane.

Definition 207 An hypermap is a ^-constellation.

Another equivalent definition of an hypermap is a map whose vertices are colored in


black and white in such a way that each edge connects two vertices of distinct colors. A
bicolored plane tree is a hypermap.

The Mathieu group M\2 is associated with the icosaedron. Let ri be the permutation
of the neighbor of i in the rotation about the vertex i. The Mathieu group M\i is the
group generated by urj1 for z,j = 1,2,..., 12. Many maps are associated with M12. For
example, the map of the following figure has 4 vertices, 6 edges and 4 faces.

The permutations associated, with the edges

<72 = (1,2)(3,4)(5,6)(7,8)(9 > 10)(11 > 12)

with the vertices


CTl = (1,4,5X2,3X6,7,8,9,10,11,12)
and with the faces

a3 = o^o? = (1,3)(2,5,12,9,10,8,6,4)(7)(11)

generate the group M\i- The passport of this map is [732,2 6 ,821 2 ]. Another representa-
tion is given by the following figure

191
10. Groups and Symmetries

The permutation associated with the edges

a2 = (1,2)(3,4)(5,6)(7,8)(9,10)(11,12)

with the vertices


a 1 = (l,2,3,4,5,7)(6,9,8)(10,ll)(12)
and with the faces

a3 = a^a^1 = (1)(3)(2,7,9,11,12,10,6,4)(5,8)

generate the same group M12. The passport of this map is [6321,26,8212]. One can show
that the group M\i has 50 maps and 1437 hypermaps. The first 50 hypermaps are given
in the table.
Passport Maps Hypermaps
3 e
\ 3 ,2 ,(ll,l) 4 12
4 ^ , 2 ° , (11,1) 4 12
5 ^ , 2 ° , 6321 12 36
5*1^ 28,821 s 6 18
6321,2°, 6321 8 24
6321,2 e , 821* 12 36
82P,2 B ,821 : J 4 12
Total 50 150
Some passports have no maps. They are associated with 1287 hypermaps. For example,
the passport [3 4 ,2 4 1 4 , (10,2)] has no map and 12 hypermaps.

Definition 208 A cactus is a plane constellation [<7i,<72, ...0"/c,c] where c is the cyclic
permutation c= (1,2,.., n). Thus

CTi<72...<Tfc = C

Definition 209 Two cacti are isomorphic if they are isomorphic as constellations.

Cacti are represented by polygons. A rooted cactus is a cactus in which one polygon is
considered as the root of the cactus and marked 1.

Example. Major triads having one common pitch-class are represented by a cactus. Three
permutations are associated with these chords and verify

<7l<72<73 = C

In the following figure, the cactus represents the major triads

192
Constellations and Cacti

The bicolored tree of the cactus gives the three permutation.

The permutation 0-3 is obtained from black vertices around the white vertices labelled 1.

a3 = (1,2,3X4,5,7X6)
The second permutation represents the cycles made from the white vertices labelled 2.

a 2 = (l,4)(2)(3)(5)(6)(7)

The third permutation corresponds to the white vertices labelled 3.

<n = (l)(2)(3)(4)(5)(6,7)

The product of the three permutations verifies

<7i<72<73 = (1,2,3,4,5,6,7)

Harmonic progressions of triads with one common pitch-class can be classified by their
passports. In our example, the passport of the cactus is

[21 5 ,21 5 ,13 2 ,7]

One can show that the number of cactus having this passport is 49 (see [Zvonkin, 2004]).

193
11
Tessellations and Canons

Tilings have many applications in music theory. They have been used for building tem-
peraments on a lattice and in canons theory, and could also be very useful for covering
the plane with twelve tone rows, chords and scales. The first mathematical introduction
on Tilings and Patterns has been written by Branko Griinbaum and Geoffrey Shepard
in 1989. Canons are special case of tilings. Moreno Andreatta has shown that Regular
Canons are related to the classification of abelian finite groups.

11.1 Musical Tilings


In this section, we review some classes of planar tilings. Each face or each vertex is
associated with a musical object. The periodicity of the tiling induces a repetition of
musical objects. In chords tesselations, a chord is mapped to a tile. This tile is sufficient
to generate a pattern. According to the French theorist Jean-Philippe Rameau, harmonic
chords are built from major and minor thirds. The chain of major thirds is generated
by the translation T4 and the chain of minor thirds is generated by the translation T3.
The cartesian product Z3 x Z4 defines a tiling of the plane in which the major thirds
are represented by the vertical axis and the minor thirds by the horizontal axis. Musical
tilings are periodic, and this periodicity defines the original tile. Since the bottom of the
tile {0, 3, 6, 9} is identified with the top line, and the left side {0, 4, 8} with the right
side, the edges of the tiling are identified in the direction of the arrows to form a torus.
This tiling is a representation of the direct sum

Z12 = 4Z 3 0 3Z 4

A triangulation of each square leads to major and minor triads. This triangulation cor-
responds to the triangulation of a manifold M, and verifies the Euler-Poincare formula.
Let F be the number of faces, V the number of the vertices and E the number of edges.
11. Tessellations and Canons

We have for any triangulation that

F + V - E = x(M)

where x is the characteristic of Euler-Poincare of the manifold M. The Euler character-


istic of a sphere with g handles is x = 2 — 2g. In particular, a torus is a sphere with
one handle, thus x = 0. In the tiling of major and minor triads, F = 24, V = 12, and
E = 36, so F + V - E = 0.

C Eb F# A C

But the torus is not the only possibility of chords tesselations. If we consider the chain
of fifths represented by the translation Tj and the chains of tritone, represented by the
translations TQ, we get a tiling in which the gluing of the borders in the direction of the
arrows leads to the Klein bottle. A triangulation of this tilings verifies the Euler-Poincare
formula F + V - E = 0.

G 0 ~" A E B F*|

Db Ab Eb Bb F
6 1 8 3 10 5

7 * _ ^» 9 i Jj I
C G D A^ E B F#

Another instructive example is given by the tiling of the projective plane PR 2 with
Steiner systems. Let X be the set {1, 2, ..., n}.

Definition 210 A Steiner system S((,,m,n) of order n is the set of m-subsets A in X


called blocks such that every l-subset from X is contained in exactly one block of A.

Example. The Steiner triplet S(2,3,7) is the set of 7 triplets: {1,2,4}, {2,3,5}, {3,4,6},
{4,5,7}, {1,5,6}, {2,6,7} and {1,3,7}. The unique Steiner triple system of order 7 is
known as the Fano plane.

The Steiner triple system 5(2, s + 1, s2 + s + 1) is, for each integer 5, a finite project-
ive plane. Suppose now that we have a set of six musical objects, denoted by the first

196
Musical Tilings

positive numbers X = {1,2,..., 6}. The set of triplets such that all pair of numbers in X
occurs exactly in two triplets is not a Steiner triple system, but is related to 5(2,3,7).
It is a set of 10 triplets: Tx = {1,2,3}, T2 = {1,3,5}, T 3 = {1,5,6}, T4 = {1,2,4},
T 5 = {1,4,6}, T6 = {2,3,6}, T7 = {2,4,5}, T8 = {2,5,6}, T9 = {3,4,5}, Tl0 = {3,4,6}.
The triangulation of the projective plane is shown in the following figure.

The Euler characteristic of the projective plane (x = 1) verifies the relation F+V—E = x,
with F = 10, V = 6 and E = 15.

Archimedean tilings are uniform tilings of the plane by regular polygons, n-gons,
as squares (denoted 4), triangles (3) and hexagons (6), octogons (8) and dodecagons (12
or C). Each vertex has the same adjacent tiles and is named a\...an by these tiles. For
example 3463 means, that at each vertex we find a triangle (3), a square (4), an hexagon
(6) and another triangle (3). Each number of the tiling represents a sector of a 2n angle:
4 means a n/2 angle, 3 means a 7r/3 angle, 6 corresponds to a 27r/3, 8 is a 37r/4 angle
and 12 means a 57r/6 angle. This determines a function </> verifying

(j)(ai...an) = </>(ai) + ... + (/>(an) = 2n

The sum of each components of the tiling name is always 2ir. For example for the tiling
x given by C3C (where C is the hexadecimal notation of 12) the function

4>{x) = 20(12) + 0(3) = 2.5?r/6 + TT/3 = 2TT.

It has been shown that there are only eleven archimedean tilings: 333333, 4444, 666,
33336, 33344, 33434, 3463, 3636, 3CC, 46C and 488. Each tiling has a group of sym-
metry. The tiling 43433 has a right and left handed version. Its symmetry group is p4girn.
Archimedean tilings have been generalized to k-uniform tilings, that is tilings with k
kinds of vertices. Archimedean tilings have only one type of vertices with regular poly-
gons, fc-uniform tilings have been enumerated by Otto Krotenheerdt. They are 20 tilings
with two kinds of vertices, 39 tilings with 3 kinds of vertices, 33 tilings with 4 kinds of
vertices, 15 tilings with k = 5, 10 tilings with k = 6, 7 tilings with k = 7 and no tilings
for k greater or equal 8.

197
11. Tessellations and Canons

The following picture shows a tiling 333333 filled by twelve tone rows. In each triangle,
the twelve notes are shared with the neighboring triangles. The period of this tilling is
equal to 12.

A
1
•/ A

A J
0 3 7 11 8 3 7 11 4 3 7 11 0

The next picture shows a 6434 tiling filled by twelve tone rows. The wheels share the
notes of each row. The period is equal to 12.

Twelve \
Tono
G 7 0 Row I S l ^ ^ H ^
Tiling /
3
11
11
v
\<{o] 5 ]8>^ > 10
1 9 \ 1 / 3
4 5 4
' iQ ) 'A

0 7 6

Polygons tilings are figures assembled from other polygons as in the Tetris game.
For figure made from squares Solomon W. Golomb coined the term polyominoes in 1954.
Tilings made with figures from triangles are called polyamonds and tilings made with fig-
ures from hexagons are called polyhexes. There are only one monomino (a single square),
one domino (a double square), two triominoes (assemblies of three squares), five tet-
rominoes, twelve pentominoes, 35 hexominoes, 108 heptominoes, etc. Only four hep-
tominoes do not tile the plane.

198
Musical Tilings

For a tiling of polyominoes, one can associate a rhythm with each square. The polyominoe
and its rhythms are repeated through the whole tiling. In the following figure, the tile is
represented by five elementary squares (see for example the black pentomino). The tiling
of the whole rectangle is made of ten tiles. The distribution of the rhythmic figues is the
same in each tile.

Pentilings are tesselations made with plane pentagons. Figures with five-fold sym-
metry have been associated with magic and mysticism since ancient times, as it appears
to be a non natural symmetry. They are sometimes considered as an anticipation of
Penrose's aperiodic tilings. From the musical point of view, the vertices of a pentagon
are associated with a musical object such as a rhythm figure or a pentachord. The first
arrangements of pentagons were described by Diirer in a book published in 1525. In one
of them, the tile consists of two pentagons and one lozenge gap. Six pentagons are linked
by an hexagon as shown in the following left figure.

10 6

The right figure is built by applying a five-fold rotation symmetry about a central
pentagon, associated with the chord [0, 2, 5, 7, 10]. Five hexagons are distributed around
the central point of the figure. Six pentagons are included in a bigger pentagon. Two
adjacent pentachords have two common pitch classes.

199
11. Tessellations and Canons

Aperiodic tilings are tilings in which no subset of polygons tiles the plane period-
ically. Robert Berger has found an aperiodic tiling in 1964. Ten years later, Raphael
Robinson simplified the number of the tiles to six. Robert Ammann discovered a number
of aperiodic tilings and in 1974, Robert Penrose found an aperiodic tiling that uses only
two tiles. The first tile is a rhombus containing an angle of 36° and the second tile is a
rhombus containing an angle of 72°. If we locate the origin at one of the vertices of the
tiling and define the x and y axes such that the angles between edges of tiles and the x
axis are multiple of 7r/5, then we could represent the Penrose tilings by integers points
of R 4 . Using the rational expressions of angles

C O S(*V5)
v../v/ = —- A — sin(7r/5) = V2
4
and
to /K\ -! + ^ • /« ,K\ /zVs + VS
v
cos(27r/5) = 4 sin(27r/5)
' ' = v2- 4
we can represent the coordinates (x,y) of each vertex by four integers (#1,0:2,2/1,2/2)
defined by
J x = x\ cos(7r/5) + #2 cos(27r/5)
\ V = 2/1 sin(7r/5) + 2/2 sin(27r/5)
Wang's conjecture (1961) states that all tilings could always be arranged to tile the plane
periodically.

Reptiles. A Reptile (Replicating tile) is a shape that can be tiled with smaller congru-
ent copies of itself. The rep number, denoted by rep-n, is the number n of tiles required
to build a larger version. All triangles and parallelograms are rep-n2 (for some integer
n > 1). The 45-degree right triangle is rep-2. It is easy to show that if a figure is a rep-n
tile then it is also a rep-(2n-l) tile.

The reptile is said to be regular if the copies used to build the larger version have the
same size (self-similarity). If a figure is a regular rep-n tile then it is also a regular rep-n2
tile. It has been shown by Sze-Man Ngai, Victor Sirvent, J.J.P. Veerman and Yang Mang
that there are only six regular rep-2: the 45-degree right isocele triangle, the rectangle
of sides 1 and y/2, and four fractal reptiles: the Levy dragon, the Heighway dragon, the
twindragon and the tame twindragon. These tiles are used in fractal music. Fractal music
is the use of fractals computation to create music by a mapping of numerical values issued
by the computation to musical parameters.

11.2 Perfect Tilings


Perfect Rhythmic Tilings have been studied by the minimalist composer Tom Johnson
[Johnson, 2004] and used in his composition Tilework. If linear perfect tilings have been
used only in a rhythmic perspective, it is also interesting to notice that they have many
other applications. They are both reptiles and augmented canons. Let Mayi be the affine
transformation of Z n
Maj) : x —• ax + b mod n

200
Perfect Tilings

Definition 211 Let m, k be two non negative integers. A linear perfect tiling of length
n = mk and of order k is a partition of Zn in m sets of k elements Aij = Mij(Zk)

(hj)es

such that all coefficients i are distinct. The mutiplicative coefficients i are called the
augmentations and S the set of affine coefficients of the tiling.

Examples. For n = 15, the set of affine coefficients is

S = {(1,2), (2,4), (4,5), (5,1), (7,0)}

and the tiling is


Zis= |J Mitj(Z3)
(hj)es
The following table shows the set Z15 tiled by five augmentations of Z3.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14]
A+2 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0
2A + 4 0 0 0 0 0 0 0 0 1 0 1 0 1 0 0
4A + 5 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0
5A + 1 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0
7A 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1
It can be written in a compact form, keeping only the multiplicative coefficient of the
affine transformation,
(7,5,1,1,1,4,5,7,2,4,2,5,2,4,7)
For n = 21, there are only 9 perfect linear tilings (if we do not count the mirror tilings).
One of them is given by the set

S = {(1,3), (2,15), (3,7), (4,12), (5,1), (6,2), (9,0)}

or in the compact form

(9,5,6,1,1,1,5,3,6,9,3,5,4,3,6,2,4,2,9,2,4)

A computer program gives the following result.

Proposition 212 The number of the linear perfect tilings of order k = 3 are given in
the table
m n mn m n m
5 15 2 11 33 4 138
6 18 0 12 36 15 324
7 21 18 13 39 61 644
8 24 66 14 42 328 800
9 27 382 15 45 2 151878
10 30 1 104 L*6 48 17 660 110

201
11. Tessellations and Canons

Perfect tilings have been defined for any shape. Perfect square tilings are the tiling of a
square with smaller squares.

Definition 213 A perfect square tiling is a tiling of a square with finitely many similar
but non-congruent copies of the same square. The order of the tiling is the smallest
number of copies needed in a perfect tiling.

Example. Duijvestijn [Duijvestijn, 1978] has proved that the smallest order for a perfect
square tiling is 21.

WXm
50
1
15 17
9
I 18 24
29 : 25 16
1

33
37 g m
Proposition 214 The number of perfect square tilings of order k is given in the following
table

\ k \P\\ k W\
21 1 24 26
22 8 25 160
23 12 26 441 \

11.3 Messiaen Chords


In order to show the importance of topology in the musical field, we introduce in this
section the molecular presentation of chords. To build isographies, we need a graphical
presentation of chords. We choose a planar representation, in which the horizontal axis is
tiled by the major thirds (translation T4) and the vertical axis by the tritones (translation
TQ). The diagonal axis shows the relations of fifths (or fourths) in the following net:

202
Messiaen Chords

1 5 9 1

In his Traite de rythme, de couleur et d'ornithologie, Olivier Messiaen introduces sev-


eral invented chords that stand distinctly apart from his modes of limited transpositions.
However, these chords include some limited transpositions sets. We would like to show
that the molecular presentation is well suited to follow the included kernel of limited
transpositions sets.
Let's examine Messiaen's Chords. The first chord studied by Messiaen in his Traite*
(Tome 7, p. 135) is the chord of transposed inversion on the same bass note (accord a
renversements transposes sur une mime note de basse). This chord is a dominant ninth
chord in which the seventh degree is replaced by the first (C, F, G, Bb, D) and its color
is changed by the introduction of two added notes (ify B). The chord A {0, 2, 5, 6,
7, 10, 11} is transposed and forms the transposed chords B = Tio(A), C = Tf(A) and
D = T2(A). The transpositions have always the same bass note (C).

2
o

In the molecular presentation, the chord shows two limited transpositions sets, the
double fifth {0, 5, 6, 11} and the augmented fifth {2, 6, 10}.

The second type of chords used by Olivier Messiaen is called the chord of contracted
resonance (accord a resonance contract&e, Tome 7, p. 150). These chords have been used
for the first time in the Quatuor pour la fin du temps. The molecular presentation of
CCR1A {1, 2, 3, 4, 6, 7, 9}

203
11. Tessellations and Canons

shows the limited transpositions subsets, the altered seventh {1, 3, 7, 9} and the two
tritones.

The molecular presentation of CCR1B {1, 2, 3, 4, 5, 8, 10} shows the same subsets,
the altered seventh {2, 4, 8, 10} and the two tritones.

These two chords have the same number of fifths (notes located on the diagonal axis)
and the same limited transpositions subsets.

The second chords of contracted resonance (CCR2) A and B have the following present-
ation. The molecular presentation of CCR2A {0, 1, 5, 6, 9, 10}

shows the limited transpositions subsets, the chord of augmented fifth {1, 5, 9} and the
tritones {0, 6}.

The molecular presentation of CCR2B {0, 3, 4, 6, 10, 11} shows two different subsets,
the altered seventh {0, 4, 6, 10} and two tritones. We can easily see that this chord is
included in CCR1A.

6 10

204
Messiaen Chords

The third type of chords studied by Messiaen is the revolving chord (accord tournant,
Tome 7, p. 165). The molecular presentation of the revolving chord A {0, 1, 2, 3, 4, 6,
7, 8} looks like a long troncated triangle. The chord includes the mode 5 {2, 3, 4, 8, 9,
10}, the chord of altered seventh {2, 4, 8, 10}, the double fifth {2, 3, 8, 9}, the chord of
augmented fifth {0, 4, 8} and three tritones.

The molecular presentation of the revolving chord B {0, 1, 2, 3, 7, 8, 10, 11} has also a
triangular shape. The chord includes the double fifth {1, 2, 7, 8}, the chord of augmented
fifth {3, 7, 11} and two tritones.

The molecular presentation of the revolving chord C {0, 1, 3, 4, 5, 8, 10, 11} is built
on the axis of fifths. The chord includes the limited transpositions sets, the double fifth
{4, 5, 10, 11}, the chord of augmented fifth {0, 4, 8} and two tritones.

In the harmonic progression, the molecular representation helps us to follow the evol-
ution of the kernel of included limited transpositions sets. In the final movement of Et
exspecto resurrectionem mortuorum, the melodies uses the previous chords. The kernel
is a chord of altered seventh that we can follow as a rectangle shape (CCR1 mes. 9: 1A,
mes 10: IB, mes 11: 10A, mes. 12: 10B). Messiaen uses the revolving chord 12A which
includes the altered seventh and the double fifth. The double fifth becomes the new ker-
nel. This produces a kind of modulation. The double fifth is the only subset of limited
transpositions in the following chords (revolving chords 12B et 12C, mes 12-13).

205
11. Tessellations and Canons

11.4 Reflexive Chords


In a given tesselation, reflexive chords are the points of a polytope surrounding a unique
central pitch class. In the above figure, the tiling is the usual tiling of major and minor
thirds. The polytope P defines a reflexive chord [0, 3, 4, 7, 8, 10, 11]. The points of P
are placed around the central pitch class G (7).

Definition 215 The volume of a reflexive chord associated with a polytope P is the
volume of the polytope P. Let dP be the border of P and \P\ be the cardinal of P. The
volume of P is given by the Pick's formula.

vol{P) = \P\-\\dP\-l

Example. In the previous example, the chord [0, 3, 4, 7, 8, 10, 11] has the volume equal
to vol(P) = 7 - 6 / 2 - 1 = 3.
In the plane, there are only 16 reflexive chords. They are given in the following figure.

206
Young Diagrams

11.5 Young Diagrams


Young diagrams or Ferrers boards represent graphically integer partitions. The parts of
the partition are the rows of squares and the sum is done vertically. For each integer n,
we associate the integer 1 with a rhythmic element. The Young diagram gives all the
combination of building rhythms from the initial object.
In the following example, the quaver is the unit. The rhythms are classified by the
integer partitions of n = 6, given in the lexicographic order (6) > (5,1) > (4,2) etc.

J. (6) I I I I I I I

*• J- (3,3)

i J.J> F"
(3,2,"lT

(3,1,1,1)

JJi)i)(2,2,1,1)P

J J> -h J> J> (2,1,1,1,1,

J>J>J>J>J>J> (1.1.1.1.1.1)0

207
11. Tessellations and Canons

11.6 Prime Canons


The first canons appear in the 12 th century. The french chace and the italian caccia are
the primitive forms of canons. Despite the fact that canon is an old procedure, it has
been used at any time, from Renaissance composers (G. Ockeghem, M. Forestier, etc.)
to 20 th century composers. Famous canons have been composed by Jean-Sebastien Bach
(Musical Offering, Goldberg variations, Canonical Variations Vom Himmel hoch BWV
769), Gabriel Faure (Romances sans paroles, Thames et variations, Prelude #), Arnold
Schoenberg (Opus 25, 28, 40, 41), Alban Berg (Kammerkonzert), Anton Webern (Opus
2, 16, 28, 30, 31), Igor Stravinsky (Octuor, Sextuor, In memoriam Dylan Thomas), Paul
Hindemith (Fuga undecima in B in Ludus Tonalis) and many others. Today, musicians
distinguish at least five types of canons.
1. The canon perpetuus in which the different voices return identically to their begin-
ning, like for example in Bach's Musical Offering 2.
2. The canon per tonos in which the different voices return to their beginning, but at
different pitches (Bach's Musical Offering 3).
3. The canon cancrizans or retrograde. The upper voices are in retrograded invertible
counterpoint. One of the first canon cancrizans is Guillaume de Machaut's Ma fin
est mon commencement.
4. The canon per mensurationem in which voices are the same from the melodic but
different in rhythm (e.g. Josquin des Pres's Missa Uhomme arme).
5. The canon per augmentation in which voices are transformed by repetition of some
elements or addition of new elements.
For example, Olivier Messiaen uses a rhythmic canon in the seventh song of Har-
awi called Adieu, measures 59-66. The rhythms are counted relatively to the double
quaver (e.g. 3 means three double quavers, etc.). The upper voice juxtaposes three non-
retrogradable rhythms [3, 5, 8, 5, 3], [4, 3, 5, 3, 4] and [2, 2, 3, 5, 3, 2, 2]. The same
rhythms appear in the two other voices shifted in time.

From a mathematical point of view, canons are linked to the decomposition of finite
abelian groups.
Definition 216 Let G be an abelian group and R, S, X be three subsets of G. A set R
tiles the set X with S if X is a direct sum of R and S
X = R®S
i.e. if every element of x G X has a unique decomposition x = r + s with r G R and
s G S. R is called the tile and S is the translation set. When X = G, the couple (iJ, S)
is called a factorization of the group G.
Examples. The tile R = {0,2} tiles the set, Z4 = {0,1,2,3} with two translations S =
{0,1}, i.e. the sets R and R + 1.

0 1 2 ~3~1
R 1 0 1 0
R+ l 0 1 0 1

208
Prime Canons

Remark that the number 1 appears only one time by column. The motive R is also
written 101. The tiling corresponds to the decomposition of Z4 in the direct sum

Z4 = # 0 S ={0,2} 0(0,1}
= 2Z 2 0 Z 2

In the musical field, a factorization (i?, S) is called a canon or a tiling canon. The set R
represents the set of events of the ground voice (dux) and the set S represents the time
attacks: R + s with s G S are the different voices (comes) of the canon.

Definition 217 The characteristic polynomial of a set A C Z n is

A(x) = £>»
aeA

When working with polynomials, the direct sum of sets is interpreted as a product
of polynomials and the union of sets as a sum of the corresponding polynomials. In the
previous example, R = (0,2} is associated with R(x) = x2 + x° = x2 + 1 and the set
S = {0,1} corresponds to the polynomial S(x) = x + 1. The product

R(x)S(x) = (z 2 + l)(a; + l)
= x3 + x2 + x + 1

is the characteristic polynomial of Z4 = 2Z 2 0 Z 2 .

Proposition 218 The characteristic polynomials have the following properties.


(1) The cardinal of the set A is the value of the characteristic polynomial at the point 1.

\A\ = A{\)

(2) Let Pm(x) be the characteristic polynomial ofZm. Then the characteristic polynomial
of the set kZm is Pm(xk).
(3) If E = i? 0 S, then the characteristic polynomial of E is

E(x) = R(x)S(x)

and the cardinal of E is the product

\E\ = \B\x\S\ = R(lW)

(4) If F = R\J S, then the characteristic polynomial of F is the sum

F(x) = R(x) + S(x)

Remark that the canons are related to the underlying group (which defines the sum).
For example, the tile 10110100 associated with R = {0,2,3,5} tiles the integers Z, but
not the nonnegative integers N (there is always a hole in position 1). The tile R tiles
the group Zs = {0,1,2, ..,7} equipped with the sum modulo 8, but not the same set
Ns = {0,1,2, ...,7} equipped with the usual sum. The first example is a tiling of the
circle, whereas the second is a tiling of the line.

209
11. Tessellations and Canons

Definition 219 Two canons E = R® S and E' = Rl 0 Sl of Z n are equivalent if


\R\ = \R,\> | 5 | = \S'\, and if there exists a translation Tj and a permutation a in the
symmetric group S\s\ such that

Tj{R+si)=R' + 8aW

for i = 1,.., | 5 | . In other words, E and E' are equivalent, if the voices of the canon E
taken at any time is a permutation of the voices of the canon E\

Example. The canon E = {0,1} 0 {0,2} is equivalent to the canon E' = {0,3} 0 {0,2},
because T^(R) = R' and T^(R + 2) = R' + 2. The table of the second canon is obtained
from the table of the first by a translation of each element by T 3 (or by a circular
permutation of the columns)

h
0 1 2 ~3~] 0 2 3
R 1 1 0 0 R' 1 0 0 1
R+2 0 0 1 1 R' + 2 0 1 1 0
Definition 220 Two canons E and E* of Zn are called dual if E = i? 0 5 and E* =
S®R.

Example. The canon E = {0,1} 0 {0,2} is dual of the canon E* = {0,2} 0 {0,1}.
Remark that the tables can be very different, by exchanging the role of R and S.

0 1 2 31 0 1 2 3
-

R 1 1 0 0 s 0 0
R+2 0 0 1 1 5+1 0 1 0 1
Definition 221 Two canons E and E' ofZn are isomorphic if E' is equivalent to E or
to its dual E*. Non-isomorphic canons are called prime canons.

An important problem is to build and to determine the number of prime canons. The
problem seems to be easier on the line than on the circle. The first way of building
prime canons on the line (i.e. we work on Z n with the usual sum) is to consider the
decomposition of n in powers of primes. For n = qi...qr, we have

Z n = Zqi 0 q{Lq2 0 ... 0 (qi...qr)Zqr

Example. Writing all the decomposition products for n = 12, we get eight canons

n R
1x12 Zi 0 Z12
2x6 Z2 0 2Z6
3x4 Z 3 0 3Z4
4x3 Z4 0 4Z3
6x2 Z6 0 6Z2
2x2x3 Z 2 0 2Z 2 0 4Z3
2x3x2 Z 2 0 2Z 3 0 6Z2
3x2x2 Z 3 0 3Z 2 0 6Z2

210
Prime Canons

Another way to do the same thing is to use tensorial products. Define the horizontal
elements

Hrj i i •• i

and vertical elements by


1 0 01
Vg = 0 '•. 0
0 0 1

These elementary generators have the obvious properties

Hp®Hq = Hq®Hp = Hpq


vp®vq = vq®vp = vpq
(Hp®vqy Vp®Hq

For example, the product T = H2 ® V3 ® V2

1 0 "ol 1 0
T = H2 ® V3 ® V2 1 1 ® 0 1 0 ®
0 1
0 0 1

is calculated in the usual way

pTfl | 0 0
1 0
T = 0 111 1 1 0
0 1
0 0 111 1 1
which leads to the following canon

\1 1 0 0 0 0 0 0 0 0 0 0
0 0 1 1 0 0 0 0 0 0 0 0
0 0 0 0 1 1 0 0 0 0 0 0
T = 0 0 0 0 0 0 1 1 0 0 0 0
0 0 0 0 0 0 0 0 1 1 0 0
0 0 0 0 0 0 0 0 0 0 1 1I
which can be written

T = {0,1} 0 {0,2,4,6,8,10} = Z 2 0 2Z 6

But these generators could not be used for tiling the circle (when we use the sum
modulo n). The first prime canons are given in the following table. The column N n
gives the number of canons for tiling the line, and the column Z n gives the number of

211
11. Tessellations and Canons

new prime canons in tiling the circle.

n Nn Zn Total nr 2 2 4
2 1 0 1 10 3 3 6
3 1 0 1 11 1 0 1
4 2 0 2 12 8 15 23
5 1 0 1 13 1 0 1
6 3 0 3 14 3 10 13
7 1 0 1 15 3 22 25
8 4 2 6 16 8 41 49
We see in the table that for n = 8, two new prime canons are given. These canons are

E1 = 3Z 4 0 4Z 2 = {0,3,6,9} 0 {0,4}

and
E2 = 3Z 2 0 2Z 4 = {0,3} 0 {0,2,4,6}
which are not obtained from tensorial product of a combination of the elements
Hp and Vq.

11.7 Multiple Canons


Musicians are usually interested in tiling the line with more than one tile. For a given
integer n, by starting with p tiles Ai, ;4 2 , .., Ap in Z n , one constructs an embedding of
these canons in Zpn and writes each tile in pZ n + fc, for k = 0, ..,p — 1. For example,
consider two tiles A = {0,1} and B = {0,2} in Z$

0 1 2 3 4 ~5l
0 1 2 3 4 5
A 1 1 0 0 0 0
B 1 0 1 0 1 0
A+2 0 0 1 1 0 0
B+ l 0 1 0 1 0 1
A+4 0 0 0 0 1 1
These two tiles are embedded in Zi 2 , and then become A' = {0,2} and B' = {0,4,8}.
The new canon is A! 0 {0,4,8} U 5 ' 0 {1,3}. In the space Zi 2 , the tile A is written
according to the space {0, 2, 4, 8, 10} and the tile B is written according to the space
{1, 3, 5, 7, 9, 11}.

0 1 2 3 4 5 6 7 8 9 10
A' 1 0 1 0 0 0 0 0 0 0 0
in0
A' + 4 0 0 0 0 1 0 1 0 0 0 0 0
\ A' + 8 0 0 0 0 0 0 0 0 1 0 1 0
\ B' + l 0 1 0 0 0 1 0 0 0 1 0 0
\ B' + 3 0 0 0 1 0 0 0 1 0 0 0 1
Prime canons are used to tile the line with chords. If we call the notes and the voices 1,
2, 3, etc., the problem is to write a tiling in which each column is a complete chord with
no repeated note and is a different circular permutation compared to the other columns.

212
Multiple Canons

For a chord of three notes, the set {1, 2, 3} and its permutation {2, 3, 1} and {3, 1, 2}
form three columns of different transpositions of the same chord. The motive 100 tiles a
chord of 3 notes on 3 voices. Each prime canon is associated with a word. The tiling V3
in Z3 is associated with the word 123 composed of the number of the line in which the
1 appears, when we read the canon from the column 0 to the column 2.

0 1 ~2~|
Voice 1 1 0 0
Voice 2 0 1 0
Voice 3 0 0 1

Prom the word w, one constructs the different permutations an(w) with no equal number
in the same column
Hi 2 3 w
2 3 1 oi(w)
3 1 2 a2(w)
Thus the tiling of a three notes chord is derived from the prime canon V3.

Tiling chords could also be constructed starting from canons with different tiles. For
example, from the previous two tiles canon

A' 0 { 0 , 4 , 8 1 0 5 ' 0 ( 1 , 3 }

we get a word
w = 114522453345
This word and the permutation <72 (w) lead to a tiling chord with two notes on five voices

pr 1 4 5 2 2 4 5 3 3 4 5 w
4 5 2 2 4 5 3 3 4 5 1 1 a2{w)

The expanded expression of each voice is

f °"1 1
1
2
2
3
2
4
0
5
0
6
0
7
0
8
0
9
0
10 " 111
0 0
0 0 0 0 1 1 2 2 0 0 0 0
0 0 0 0 0 0 0 0 1 1 2 2
2 0 1 0 2 0 1 0 2 0 1 0
0 2 0 1 0 2 0 1 0 2 0 1

Remark that the permutation GQ(W) and aio(w) are other solutions of tiling on two
voices. But there are also five solutions on five voices. For example, one solution is given
in the following table

FT 1 4 5 2 2 4 5 3 3 4 5 w
3 4 5 1 1 4 5 2 2 4 5 3 a3(w)
5 3 3 4 5 1 1 4 5 2 2 4 a5(w)
4 5 2 2 4 5 3 3 4 5 1 1 o-ioM

213
11. Tessellations and Canons

The motive of the canon tiles a chord of four notes (1, 2, 3, 4) on five voices.

n 1 ?, 3 4 5 6 7 8 9 10 11
l 1 0 2 2 3 3 0 0 0 4 4
0 0 4 4 1 1 0 2 2 3 3 0
2 3 3 0 0 0 4 4 1 1 0 2
4 2 1 3 4 2 1 3 4 2 1 3
3 4 2 1 3 4 2 1 3 4 2 1

11.8 Regular Canons


In this section, the notion of Regular Complementary Canons of Maximal Category in-
troduced by Vuza [Vuza, 1991] is investigated. We review some results of regular canons
and recall that the determination of these canons is related to the classification of abelian
finite groups. The links with Haj6s groups was pointed out for the first time by Moreno
Andreatta [Andreatta, 1995]. See also Emmanuel Amiot for a review [Amiot, 2004, 2005].

Definition 222 Let G be an abelian group. A subset A of G is r-periodic if it exists


r G G\{0} such that A = r + A. The set A is periodic if A is r-periodic for some r.
Otherwise the set A is called aperiodic.

Example. Subsets of Zn are aperiodic if they are acyclic.

Definition 223 Let G be an abelian group. G is called an Hajds group if for each fac-
torization of G = i? © S at least one factor is periodic.

Remark that if G is a finite abelian group, the factors are not necessary subgroups of G.
For example, if R = {0, 1, 4, 5} and S = {0, 2}, R®S = Z 8 The set R is 4-periodic, but
none of R and S are subgroups of Zg.

In 1962, A.D. Sands gave the classification of all finite abelian groups which are Haj6s
groups.

Theorem 224 (Sands) The group Zn is an Hajds group if and only ifn is of the form:
pk for k > 0, pkq for k>l, p2q2, pqr, p2qr or pqrs for distinct primes p, q, r, s.

Theorem 225 (De Bruijn) The group Zn is a non-Hajds group if and only if n can
be expressed in the form piP2^i^2^3 where p\, P2 are primes, piUi > 2 for i = 1, 2, 3
and gcd(n\p\,nvp2) = 1.

Proof We will show that the De Bruijn's theorem is equivalent to the Sands's theorem.
Define the following sets

No = {pk, p prime, k > 0}


Ni = {phq, p, q distinct primes, k > 0}
^2 = {p2q2. Pi q distinct primes}
^3 = {pkQr, p, q distinct primes, k = 1,2}
•W4 = {pqrs, p, g, r, s distinct primes}

214
Regular Canons

and let N be the set


4
N=\jNi
i=0
Consider an integer n and its decomposition in prime factors. We have five cases.
(1) If n has only one prime factor, then n € No.
(2) If n has two prime factors, n is of the form n = paiqa2. Then n £ Nc if and only if
ot\ > 3 and a2 > 2 One could choose p\ = p, p2 = q, n\ = p a i ~ 2 , n2 = qa2 and 713 = p.
Thus n = piP2^i^2^3 and gcd(nipi,n2p2) = 1.
(3) If n has three prime factors, n is of the form n = paiqa^ra3. Then n € Nc if and
only if a i > 2, «2 > 2 and 0:3 > 1 With p\ = py p2 = g, ni = p ~ ) n2 = qa2~l and
ocl l

n 3 = r*"3, we have n = piP2n\n2n% and gcd(nipi,n2p2) = 1-


(4) If n has four prime factors, n is of the form n = pa^qa^rassai. Then n G N if and
only if n = pgrs. Thus n is of the form n = pi^2^i^2^3 with pi = p, P2 = tf, ^1 = 7"a3,
n2 = 5 a 4 and 713 = pai~lqa2~l. And we have gcd(n\pi,n2p2) = 1.
(5) If n has five prime factors or more, n = r^lr22r^3r^Ar^5...r^m, then n e Nc with
3 4 l l 2 l b n
Pi = n , P2 = 7*2, ni = rg , n 2 = r£ and n 3 = r^ ~ r2 ~ r^ ...r^i .
Thus

N c = {v\V2n\n2nz, pup2 primes, p ^ > 2, gcd(n1p1,n2p2) = 1}


The smallest values of n for which Z n is a non-Haj6s group are: 72, 108, 120, 144, 168,
180, 200, 216, 240, 252, 264, 270, 280, 288, 300, 312, 324, 336, 360, 378, 392, 396, 400,
408, 432, 440, 450, 456, 468, 480, 500, 504, 520, 528, 540, 552, 560, 576, 588, 594, 600,
612, 616, 624, 648, 672, 675, 680, 684, 696, 700, 702, 720, 728, 744, 750, 756, 760, 784,
792, 800, 810, 816, 828, 864, 880, 882, 888, 900, 912, 918, 920, 936, 945, 952, 960, 968,
972, 980, 984, 1000.
Definition 226 A regular canon (R, S) is a factorization of a non-Hajds group Z n into
two aperiodic subsets
Zn = R®S
Theorem 227 Let Z n be a non-Hajds group with n = p\P2n\n2n^. Let A and B be the
sets

A = n 2 n 3 (Zp 2 0 p 2 ^ i Z P l )
B = nin 3 (Z P l 0 p i n 2 Z P 2 )

and let Tj(B) = {j} 0 B be the translation of length j . Then the canon (R, S) defined by

S = n3(p2n2Zni 0piniZn2)
R = AUTi(JB)U...UV1(5)
is a regular canon.
Proof Using the following properties

Za0aZ6 = Za6
cLa 0 aZbc = %abc mod abc gcd(a, c) = 1

215
11. Tessellations and Canons

We compute the sum

S®A = p2n2n3Zni 0 p \ n i n 3 Z n 2 0 n 2 n 3 Z P 2 0p 2 n x n 2 n 3 Z P x
= 72 2 72 3 (p 2 £i n i ® Zp2) 0 77-172.3 ( p i Z n 2 0 p 2 n 2 Z P l )
= ra2n3(ZniP2) 0 n i n 3 ( p i Z n 2 0 p 2 n 2 Z P l )
= 7 i 2 n 3 ( ^ » n i P 2 0 p2n{ZPx) 0 nin 3 piZ n 2
= n2n3ZPlP2ni 0nin3piZn2
= ^3(^iPi^n 2 0 n2ZPlP2ni)
PlP2^1^2

and in the same way,

5 05 = p2n2n3Zni 0 P i n ! n 3 Z n 2 0 nin 3 Z P l 0 pi7iin 2 ra 3 Z P2


= P\n\n3{Zn2 0 n 2 Z P 2 ) 0 n 3 (p 2 n 2 Z n i 0 niZ P l )
= nin 3 (Z P l 0 p i Z n 2 P 2 ) 0 n2n3p2Zni
= n3{n{ZPlP2ri2 0 p 2 n 2 Z n i )
= 723^piP2nin2

Thus, we get

S0 R = n 3 Z P l P 2 n i n 2 U {1} 0 n3ZPlP2nin2 U .... U {n3 - 1} 0 n 3 Z P l P 2 n i n 2


— ^7
PlP2^1^2^3

Example. For n = 72, we have p1=n\=n3 = 2 and p 2 = n 2 = 3

,4 = 6 Z 3 0 36Z 2 = {O,6,12,36,42,48}
B = 4 Z 2 0 24Z 3 = {O,4,24,28,48,52}

The canon (R, S)

S = 1 8 Z 2 0 8Z 3 = {O,8,16,18,26,34}
R = A UTi (B) = {0,1,5,6,12,25,29,36,42,48,49,53}

is a regular canon.

Conjecture 228 All regular canons are generated by transformations of the sets A, B
and S. In other words, it exists some applications Uj, Vj (non necessary linear) such that

S = (p 2 n 2 n 3 ) u0(Zni) 0 (pi7ii n n 3 ) ui{Zn2)


R = v0(A)Uv1(B)U...Uvn3-1(B)

Example. For n = 72, three sets S

S = 18Z2 0 2 fc+3 Z 3 with £ = 0,1,2

216
Magic Squares

and three sets R


R = A\JT2(>+1{B) with £ = 0,1,2
form regular canons.

For n = 108, we have p\ = n\ = 2 and p2 = n2 = n% = 3. The sets

A = 9 Z 3 0 54Z2 = {O,9,18,54,63,72}
B = 6 Z 2 0 36Z 3 = {O,6,36,42,72,78}

leads to three sets S

S = 27Z2 © 3.2 fc+3 Z 3 avec k = 0,1,2

isomorphic to 5,55,7S. The sets

Rij = A\JTi(B)\JTj(B)
for different values (i,j) are regular canons.

For n = 120, we have p\ = n\ = n 3 = 2, p2 = 5 and n2 = 3. The sets

A = 6 Z 5 e 6 0 Z 2 = {0,6,12,18,24,60,66,72,78,84}
B = 4Z 2 0 24Z5 = {0,4,24,28,48,52,72,76,96,100}

lead to the six sets S


S = 30Z 2 0 2 fc+3 /i(Z 3 )
where ft(Z3) = {0,1,2} for fc = 0, 1, 2, 3 and h(Z3) = {0,1,5} for fc = 0,1, and nine sets
R
R = h(A)U({2e + 3}®llB)
where h(A) = A for t = 0,.., 5 and /i(A) = 6{0,2,3,4,6} 0 60Z 2 for I = 0,1,2.
Regular Canons are fascinating musical objects. This model has been implemented in
OpenMusic (a visual programming language for computer-aided composition developped
by Carlos Agon) and it has inspired many composers. The French composer Georges
Bloch wrote several pieces based on this model and on some compositional extensions of it
(rhythmical modulations between canons, the reduction of a regular canon to a collection
of self-similar canons, etc.). For an analytical account of Georges Bloch's compositional
techniques and a description of the implementational aspects see [Andreatta, 2003].

11.9 Magic Squares


A magic square consists of non-negative integers such that, in each line, in each column
or in each diagonal square, the sum of these numbers is the same constant. This sum is
called the magic constant. The most earliest known magic square is located in the 7" King
of Confucius, also called the Book of Permutations. This is the lo-shu.

[4 9 2
3 5 7
8 1 6

217
11. Tessellations and Canons

In the representation of the i" King, numbers are represented by points. According to
the legend, the emperor (ca. —2200) saw it on the shell of a tortoise near the Yellow
river.

It is easy to show that the set of magic squares (a^) formed by the n2 non-negative
integers is a vector space Mn of dimension

dimM n = n{n- 2)

The group A4 of the isometries of the square acts on the space Mn. This action is faithful.
A4 is composed of the eight transformations

A4 = {l,si,s 2 ,s 3 ,s 4 ,7*1,^2,7*3}

where 1 is the identity and the others are defined by

si(aid) = a
i,n—j+\
s2(aij) = a
n—i+lJ
s a
s( hj) = Q"n—j+l,n- - i + 1
S4(aii:j) = a
3,i
r
i(aij) = a
n—j+l,i
r a
2( ij) = Q"n—i+l,n- -3 + 1
r a
s( ij) = Q>j,n—i+1

The Roman knew an alphabetical magic square that can be read just as well horizont-
ally as vertically:

rsA A T 0 R\
R E P 0
T E N E T
0 P E R A
R 0 T A S
In the 13 th century, Manual Moschopoulos, raises of Maxime Planude stated a series
of rules to build a magic square. In the 16 th century, Albrecht Diirer (1471-1528) quoted
the magic square below in its famous engraving the Melancholy.

ri6~ 3 2 "131
5 10 11 8
9 6 7 12
4 15 14 1 I
The date of the engraving (1514) is formed by the two numbers in the center of the
last line. Each line, each column or diagonal gives a linear sum of 34. If this square is
divided in four smaller squares and a central square, each subsquare being made of four
elements, one notes that the sum of these four elements is always equal to the linear sum
(34). This property is not true for all the subsquares. For example, the square (3, 2, 10,
11) is a subsquare of sum 3 + 2 + 10 + 11 = 26, and not 34. On the other hand, the
Jupiter square admits a perfect subdivision in subsquare: the sum of the four elements
of an unspecified subsquare is equal to the constant of the Jupiter square.

218
Magic Squares

15 10 3 ~wi
4 5 16 9
14 11 2 7
1 8 13 12j
Dtirer's square and the Jupiter square are constructed by using the first 16 positive
integers (from 1 to 16) but are laid out in a particular order, to preserve the magic
constant. The Mars square uses the first 25 integers.

2 23 25 7 8
4 16 9 14 22
21 11 13 15 5
20 12 17 10 6
18 3 1 19 24
In the center, the subsquare of nine elements is also a magic square. The Venus square
uses the first 49 integers.

[4 29 12 37 20 45 281
35 11 36 19 44 27 3
10 42 18 43 26 2 34
41 17 49 25 1 33 9
16 48 24 7 32 8 40
47 23 6 31 14 39 15
22 5 30 13 38 21 46
The Mercury square is constructed by using the first 64 positive integers.

H*
49
58
15
59
14
5
52
4
53
62
11
63
10
1
56
41 23 22 44 45 19 18 48
32 34 35 29 28 38 39 25
40 26 27 37 36 30 31 33
17 47 49 20 21 43 42 24
9 55 54 12 13 51 50 16
64 2 3 61 60 6 7 57

In music, magic squares have been used in the 20 th century. Peter Maxwell Davies' piece
Ave Maris Stella composed in 1975 is based on the following square.

ri 6 2 7 3 8 4 9 5
6 2 7 3 8 4 9 5 1
2 7 3 8 4 9 5 1 6
7 3 8 4 9 5 1 6 2
3 8 4 9 5 1 6 2 7
8 4 9 5 1 6 2 7 3
4 9 5 1 6 2 7 3 8
8 5 1 6 2 7 3 8 4
5 1 6 2 7 3 8 4 9|

219
11. Tessellations and Canons

In Saturnalia Op 30, Manfred Kelkel uses the Saturn square.

nr 1 ~8~|
7 5 3
2 9 4
In Tabula Smaragdina, he uses the Jupiter square in which the integers are transformed
into rhythmic figures according to pre-established rules of transposition. In his Etude sur
le carre* magique sonore for piano, Ivan Wyschnegradsky uses a cyclic melody structure of
six measures, founded on non-octaviant spaces. The period is a seventh major of volume
11, i.e. consisted of 11 semitones. It is divided into four parts (E\>, F|J, C, D). This period
generates a complete cycle by repetition of the initial structure (3 semitones, 3 semitones,
3 semitones and 2 semitones) extended to the whole audible space.

220
Appendix A
Dodecaphonic Knots

The diagrams give the 554 chord diagrams of the dodecaphonic knots and for each,
the following properties: the Gauss word, the structural vector and the associated per-
mutations. The knots are classified by increasing Gauss words.

Description of knot 1 Description of knot 3


Di x = a6 Ds X = o4ca

Gauss word 112233445566 Gauss word 112233445665

Structural vector 600000 Structural vector 501000

(0 1) (2 3) (4 5) (6 7) (8 9) (10 11) (0 1) (2 3) (4 5) (6 7) (8 11) (9 10)

Description of knot 2 Description of knot 4


D2 x = a4b2 D4 X = a3bcb

Gauss word 112233445656 Gauss word 112233454656

Structural vector 420000 Structural vector 321000

(0 1) (2 3) (4 5) (6 7) (8 10) (9 11) (0 1) (2 3) (4 5) (6 8) (7 10) (9 11)


A. Dodecaphonic Knots

Description of knot 5 Description of knot 10


D$ X = a3bda Dio X = a3d2a

Gauss word 112233454665 Gauss word 112233456645

Structural vector 410100 Structural vector 400200

(0 1) (2 3) (4 5) (6 8) (7 11) (9 10) (0 1) (2 3) (4 5) (6 10) (7 11) (8 9)

Description of knot 6 Description of knot 11


DQ X = a3ea2 D\\ X = a3eca

Q Gauss word 112233455664


Structural vector 500010

(0 1) (2 3) (4 5) (6 11) (7 8) (9 10)
Gauss word 112233456654
Structural vector 401010

(0 1) (2 3) (4 5) (6 11) (7 10) (8 9)

Description of knot 7 Description of knot 12


D7 X = a3c3 D12 x = a2b4
Gauss word 112233456456 Gauss word 112234345656
Structural vector 303000 Structural vector 240000

(0 1) (2 3) (4 5) (6 9) (7 10) (8 11) (0 1) (2 3) (4 6) (5 7) (8 10) (9 11)

Description of knot 8 Description of knot 13


Ds X = a3cdb Dis X = a2b2ca
Gauss word 112233456465 Gauss word 112234345665
Structural vector 311100 Structural vector 321000

(0 1) (2 3) (4 5) (6 9) (7 11) (8 10) (0 1) (2 3) (4 6) (5 7) (8 11) (9 10)

Description of knot 9 Description of knot 14


Dg X = a3eb2 Du X = a2bc2b
Gauss word 112233456564 Gauss word 112234354656
Structural vector 320010 Structural vector 222000

(0 1) (2 3) (4 5) (6 11) (7 9) (8 10) (0 1) (2 3) (4 6) (5 8) (7 10) (9 11)


222
A. Dodecaphonic Knots

Description of knot 15 Description of knot 20


D\§ X = a2bcda D20 X = a2bebc
Gauss word 112234354665 Gauss word 112234356546
Structural vector 311100 Structural vector 221010

(0 1) (2 3) (4 6) (5 8) (7 11) (9 10) (0 1) (2 3) (4 6) (5 10) (7 9) (8 11)

Description of knot 16 Description of knot 21


D\§ X = a2beab D 21 X = a2bfb2
Gauss word 112234355646 Gauss word 112234356564
Structural vector 320010 Structural vector 230001

(0 1) (2 3) (4 6) (5 10) (7 8) (9 11) (0 1) (2 3) (4 6) (5 11) (7 9) (8 10)

Description of knot 17 Description of knot 22


D17 X = a2bfa2 D22 x
= a2beda
Gauss word 112234355664 Gauss word 112234356645
Structural vector 410001 Structural vector 310110

(0 1) (2 3) (4 6) (5 11) (7 8) (9 10) (0 1) (2 3) (4 6) (5 10) (7 11) (8 9)

Description of knot 18 Description of knot 23


Dig X = a2bdc2 D2 3 X = a2bfca
Gauss word 112234356456 Gauss word 112234356654
Structural vector 212100 Structural vector 311001

(0 1) (2 3) (4 6) (5 9) (7 10) (8 11) (0 1) (2 3) (4 6) (5 11) (7 10) (8 9)

Description of knot 19 Description of knot 24


D\Q X = a2bd2b -L/24 X = a2caca
Gauss word 112234356465 Gauss word 112234435665
Structural vector 220200 Structural vector 402000

(0 1) (2 3) (4 6) (5 9) (7 11) (8 10) (0 1) (2 3) (4 7) (5 6) (8 11) (9 10)


223
A. Dodecaphonic Knots

Description of knot 25 Description of knot 30


Z/25 X = a2dada Z/30 X = a2fada
Gauss word 112234453665 Gauss word 112234456635
Structural vector 400200 Structural vector 400101

(0 1) (2 3) (4 8) (5 6) (7 11) (9 10) (0 1) (2 3) (4 10) (5 6) (7 11) (8 9)

Description of knot 26 Description of knot 31


Z^26 x
= ^2eac2 D%1 X = a2e~1aca

Gauss word 112234456356 Gauss word 112234456653


Structural vector 302010 Structural vector 401010

(0 1) (2 3) (4 9) (5 6) (7 10) (8 11) (0 1) (2 3) (4 11) (5 6) (7 10) (8 9)

Description of knot 27 Description of knot 32


D27 X = a2eadb JD32 X = a2cd2c

Gauss word 112234456365 Gauss word 112234536456


Structural vector 310110 Structural vector 202200

(0 1) (2 3) (4 9) (5 6) (7 11) (8 10) (0 1) (2 3) (4 7) (5 9) (6 10) (8 11)

Description of knot 28 Description of knot 33


D28 X = a2fabc D33 X = a2cdeb

Gauss word 112234456536 Gauss word 112234536465


Structural vector 311001 Structural vector 211110

(0 1) (2 3) (4 10) (5 6) (7 9) (8 11) (0 1) (2 3) (4 7) (5 9) (6 11) (8 10)

Description of knot 29 Description of knot 34


D29 X = a2e~lab2 X = a2cec2
Gauss word 112234456563 Gauss word 112234536546
Structural vector 320010 Structural vector 203010

(0 1) (2 3) (4 11) (5 6) (7 9) (8 10) (0 1) (2 3) (4 7) (5 10) (6 9) (8 11)


224
A. Dodecaphonic Knots

Description of knot 35 Description of knot 40


D35 X = a2cfcb D40 X = a2fbea

Gauss word 112234536564 Gauss word 112234546635


Structural vector 212001 Structural vector 310011

(0 1) (2 3) (4 7) (5 11) (6 9) (8 10) (0 1) (2 3) (4 10) (5 7) (6 11) (8 9)

Description of knot 36 Description of knot 41


2^36 X = a2ce2a D41 X = a2e~xbda
Gauss word 112234536645 Gauss word 112234546653
Structural vector 301020 Structural vector 310110

(0 1) (2 3) (4 7) (5 10) (6 11) (8 9) (0 1) (2 3) (4 11) (5 7) (6 10) (8 9)

Description of knot 37 Description of knot 42


D37 X = a2cfda £> 42 X = a2 fa2
Gauss word 112234536654 Gauss word 112234556634

Structural vector 301101 Structural vector 400002

(0 1) (2 3) (4 7) (5 11) (6 10) (8 9) (0 1) (2 3) (4 10) (5 11) (6 7) (8 9)

Description of knot 38 Description of knot 43


D38 X = a2ebeb L>43 X= aVW
Gauss word 112234546365 Gauss word 112234556643
Structural vector 220020 Structural vector 400020

(0 1) (2 3) (4 9) (5 7) (6 11) (8 10) (0 1) (2 3) (4 11) (5 10) (6 7) (8 9)

Description of knot 39 Description of knot 44


D39 X = a2e~lbcb D44 X = a 2d 4

Gauss word 112234546563 Gauss word 112234563456

Structural vector 221010 Structural vector 200400

(0 1) (2 3) (4 11) (5 7) (6 9) (8 10) (0 1) (2 3) (4 8) (5 9) (6 10) (7 11)


225
A. Dodecaphonic Knots

Description of knot 45 Description of knot 50


Z^45 X = a2d2ec Z/5O X = a2ecec
Gauss word 112234563465 Gauss word 112234564365
Structural vector 201210 Structural vector 202020

(0 1) (2 3) (4 8) (5 9) (6 11) (7 10) (0 1) (2 3) (4 9) (5 8) (6 11) (7 10)

Description of knot 46 Description of knot 51


Z/46 X = a2decd Z} 5 1 X= a W
Gauss word 112234563546 Gauss word 112234564563
Structural vector 201210 Structural vector 203010

(0 1) (2 3) (4 8) (5 10) (6 9) (7 11) (0 1) (2 3) (4 11) (5 8) (6 9) (7 10)

Description of knot 47 Description of knot 52


D47 X = a2dfc2 D$2 X = a2fceb
Gauss word 112234563564 Gauss word 112234564635
Structural vector 202101 Structural vector 211011

(0 1) (2 3) (4 8) (5 11) (6 9) (7 10) (0 1) (2 3) (4 10) (5 8) (6 11) (7 9)

Description of knot 48 Description of knot 53


£>48 X = a2de2b D53 X = a2e-xcdb
Gauss word 112234563645 Gauss word 112234564653
Structural vector 210120 Structural vector 211110

(0 1) (2 3) (4 8) (5 10) (6 11) (7 9) (0 1) (2 3) (4 11) (5 8) (6 10) (7 9)

Description of knot 49 Description of knot 54


D49 X = a2dfdb D54 X = a2f2b2
Gauss word 112234563654 Gauss word 112234565634
Structural vector 210201 Structural vector 220002

(0 1) (2 3) (4 8) (5 11) (6 10) (7 9) (0 1) (2 3) (4 10) (5 11) (6 8) (7 9)


226
A. Dodecaphonic Knots

Description of knot 55 Description of knot 60


D55 X = a2e~leb2 UQQ X = a2e~1eca
Gauss word 112234565643 Gauss word 112234566543
Structural vector 220020 Structural vector 301020

(0 1) (2 3) (4 11) (5 10) (6 8) (7 9) (0 1) (2 3) (4 11) (5 10) (6 9) (7 8)

Description of knot 56 Description of knot 61


D56 X = a2e3a DQI X = ab2ab2
Gauss word 112234566345 Gauss word 112323445656
Structural vector 300030 Structural vector 240000

(0 1) (2 3) (4 9) (5 10) (6 11) (7 8) (0 1) (2 4) (3 5) (6 7) (8 10) (9 11)

Description of knot 57 Description of knot 62


D§j X = a2efda •^62 X = ab2aca
Gauss word 112234566354 Gauss word 112323445665
Structural vector 300111 Structural vector 321000

(0 1) (2 3) (4 9) (5 11) (6 10) (7 8) (0 1) (2 4) (3 5) (6 7) (8 11) (9 10)

Description of knot 58 Description of knot 63


L>58 X = a2e~ld2a ^63 X = ab3cb
Gauss word 112234566453 Gauss word 112323454656
Structural vector 300210 Structural vector 141000

(0 1) (2 3) (4 11) (5 9) (6 10) (7 8) (0 1) (2 4) (3 5) (6 8) (7 10) (9 11)

Description of knot 59 Description of knot 64


D59 X = a2f2ca DQ/± X = ab3da

Gauss word 112234566534 Gauss word 112323454665

Structural vector 301002 Structural vector 230100

(0 1) (2 3) (4 10) (5 11) (6 9) (7 8) (0 1) (2 4) (3 5) (6 8) (7 11) (9 10)


227
A. Dodecaphonic Knots

Description of knot 65 Description of knot 70


Z)(55 X = ab2dab D70 X = ab2d2a

Gauss word 112323455646 Gauss word 112323456645

Structural vector 230100 Structural vector 220200

(0 1) (2 4) (3 5) (6 10) (7 8) (9 11) (0 1) (2 4) (3 5) (6 10) (7 11) (8 9)

Description of knot 66 Description of knot 71


Dffi X = ab2c3 Djl X = ab2eca

Gauss word 112323456456 Gauss word 112323456654

Structural vector 123000 Structural vector 221010

(0 1) (2 4) (3 5) (6 9) (7 10) (8 11) (0 1) (2 4) (3 5) (6 11) (7 10) (8 9)

Description of knot 67 Description of knot 72


Dtf X = ab2cdb Df2 X = abcbca

Gauss word 112323456465 Gauss word 112324345665

Structural vector 131100 Structural vector 222000

(0 1) (2 4) (3 5) (6 9) (7 11) (8 10) (0 1) (2 4) (3 6) (5 7) (8 11) (9 10)

Description of knot 68 Description of knot 73


DQ$ X = ab2dbc .D73 X = abc3b

Gauss word 112323456546 Gauss word 112324354656

Structural vector 131100 Structural vector 123000

(0 1) (2 4) (3 5) (6 10) (7 9) (8 11) (0 1) (2 4) (3 6) (5 8) (7 10) (9 11)

Description of knot 69 Description of knot 74


DQ$ X = ab2eb2 jL/74 X = abc2da

Gauss word 112323456564 Gauss word 112324354665

Structural vector 140010 Structural vector 212100

(0 1) (2 4) (3 5) (6 11) (7 9) (8 10) (0 1) (2 4) (3 6) (5 8) (7 11) (9 10)


228
A. Dodecaphonic Knots

Description of knot 75 Description of knot 80


Z/75 X = abceab JL/go X = abceda

Gauss word 112324355646 Gauss word 112324356645

Structural vector 221010 Structural vector 211110

(0 1) (2 4) (3 6) (5 10) (7 8) (9 11) (0 1) (2 4) (3 6) (5 10) (7 11) (8 9)

Description of knot 76 Description of knot 81


DJQ X = abcdc2 Z/gi X = abcfca

Gauss word 112324356456 Gauss word 112324356654

Structural vector 113100 Structural vector 212001

(0 1) (2 4) (3 6) (5 9) (7 10) (8 11) (0 1) (2 4) (3 6) (5 11) (7 10) (8 9)

Description of knot 77 Description of knot 82


D77 X = abcd2b -^82 X =
abdaca

Gauss word 112324356465 Gauss word 112324435665

Structural vector 121200 Structural vector 311100

(0 1) (2 4) (3 6) (5 9) (7 11) (8 10) (0 1) (2 4) (3 7) (5 6) (8 11) (9 10)

Description of knot 78 Description of knot 83


D^g X = abcebc -L/g3 X = abeada

Gauss word 112324356546 Gauss word 112324453665

Structural vector 122010 Structural vector 310110

(0 1) (2 4) (3 6) (5 10) (7 9) (8 11) (0 1) (2 4) (3 8) (5 6) (7 11) (9 10)

Description of knot 79 Description of knot 84


Djg X = abcfb2 Z}g4 X = abfac2

Gauss word 112324356564 Gauss word 112324456356

Structural vector 131001 Structural vector 212001

(0 1) (2 4) (3 6) (5 11) (7 9) (8 10) (0 1) (2 4) (3 9) (5 6) (7 10) (8 11)


229
A. Dodecaphonic Knots

Description of knot 85 Description of knot 90


Z/35 X = abfadb DQO X = abd3c

Gauss word 112324456365 Gauss word 112324536456

Structural vector 220101 Structural vector 111300

(0 1) (2 4) (3 9) (5 6) (7 11) (8 10) (0 1) (2 4) (3 7) (5 9) (6 10) (8 11)

Description of knot 86 Description of knot 91


D%§ X = abdcdb Dgi X = abd2eb

Gauss word 112324534656 Gauss word 112324536465

Structural vector 121200 Structural vector 120210

(0 1) (2 4) (3 7) (5 8) (6 10) (9 11) (0 1) (2 4) (3 7) (5 9) (6 11) (8 10)

Description of knot 87 Description of knot 92


Us7 X = abdcea Z/92 X = abdec2

Gauss word 112324534665 Gauss word 112324536546

Structural vector 211110 Structural vector 112110

(0 1) (2 4) (3 7) (5 8) (6 11) (9 10) (0 1) (2 4) (3 7) (5 10) (6 9) (8 11)

Description of knot 88 Description of knot 93


D$$ X = abdeb2 D93 X = abdfcb

Gauss word 112324535646 Gauss word 112324536564

Structural vector 130110 Structural vector 121101

(0 1) (2 4) (3 7) (5 10) (6 8) (9 11) (0 1) (2 4) (3 7) (5 11) (6 9) (8 10)

Description of knot 89 Description of knot 94


Z/g9 X = abdfba Z/94 X = abde2a

Gauss word 112324535664 Gauss word 112324536645

Structural vector 220101 Structural vector 210120

(0 1) (2 4) (3 7) (5 11) (6 8) (9 10) (0 1) (2 4) (3 7) (5 10) (6 11) (8 9)


230
A. Dodecaphonic Knots

Description of knot 95 Description of knot 100


x
Z/95 X = abdfda ^100 = abe~1bc2
Gauss word 112324536654 Gauss word 112324546536
Structural vector 210201 Structural vector 122010

(0 1) (2 4) (3 7) (5 11) (6 10) (8 9) (0 1) (2 4) (3 10) (5 7) (6 9) (8 11)

Description of knot 96 Description of knot 101


U96 X = abebea DlOl X = abd^bcb
Gauss word 112324543665 Gauss word 112324546563
Structural vector 220020 Structural vector 131100

(0 1) (2 4) (3 8) (5 7) (6 11) (9 10) (0 1) (2 4) (3 11) (5 7) (6 9) (8 10)

Description of knot 97 Description of knot 102


D97 X = abe~lb3 •^102 x
= abe~1bea
Gauss word 112324545636 Gauss word 112324546635
Structural vector 140010 Structural vector 220020

(0 1) (2 4) (3 10) (5 7) (6 8) (9 11) (0 1) (2 4) (3 10) (5 7) (6 11) (8 9)

Description of knot 98 Description of knot 103


x
Z/93 X = abfbdc -^103 = abd~1bda
Gauss word 112324546356 Gauss word 112324546653
Structural vector 121101 Structural vector 220200

(0 1) (2 4) (3 9) (5 7) (6 10) (8 11) (0 1) (2 4) (3 11) (5 7) (6 10) (8 9)

Description of knot 99 Description of knot 104


Z/99 X = abfbeb D\04 X = abe2ab
Gauss word 112324546365 Gauss word 112324553646

Structural vector 130011 Structural vector 220020

(0 1) (2 4) (3 9) (5 7) (6 11) (8 10) (0 1) (2 4) (3 8) (5 10) (6 7) (9 11)


231
A. Dodecaphonic Knots

Description of knot 105 Description of knot 110


£>105 X = abefa2 X-^ilO X = abe~1dac

Gauss word 112324553664 Gauss word 112324556436

Structural vector 310011 Structural vector 211110

(0 1) (2 4) (3 8) (5 11) (6 7) (9 10) (0 1) (2 4) (3 10) (5 9) (6 7) (8 11)

Description of knot 106 Description of knot 111


^106 x
= obe~1cab ^111 X = abd~1dab

Gauss word 112324554636 Gauss word 112324556463

Structural vector 221010 Structural vector 220200

(0 1) (2 4) (3 10) (5 8) (6 7) (9 11) (0 1) (2 4) (3 11) (5 9) (6 7) (8 10)

Description of knot 107 Description of knot 112


Z)i07 X = abd~1ca2 Du2 X = abed3

Gauss word 112324554663 Gauss word 112324563456

Structural vector 311100 Structural vector 110310

(0 1) (2 4) (3 11) (5 8) (6 7) (9 10) (0 1) (2 4) (3 8) (5 9) (6 10) (7 11)

Description of knot 108 Description of knot 113


•^108 X = abfeac Z/113 X = abedec

Gauss word 112324556346 Gauss word 112324563465

Structural vector 211011 Structural vector 111120

(0 1) (2 4) (3 9) (5 10) (6 7) (8 11) (0 1) (2 4) (3 8) (5 9) (6 11) (7 10)

Description of knot 109 Description of knot 114


Diog X = abfab D\14 X = abe2cd

Gauss word 112324556364 Gauss word 112324563546

Structural vector 220002 Structural vector 111120

(0 1) (2 4) (3 9) (5 11) (6 7) (8 10) (0 1) (2 4) (3 8) (5 10) (6 9) (7 11)


232
A. Dodecaphonic Knots

Description of knot 115 Description of knot 120


D115 X = abefc2 ^120 x
= abe~1c2d
Gauss word 112324563564 Gauss word 112324564536
Structural vector 112011 Structural vector 112110

(0 1) (2 4) (3 8) (5 11) (6 9) (7 10) (0 1) (2 4) (3 10) (5 8) (6 9) (7 11)

Description of knot 116 Description of knot 121


DUQ X = abe3b D121 X = abd^c3
Gauss word 112324563645 Gauss word 112324564563
Structural vector 120030 Structural vector 113100

(0 1) (2 4) (3 8) (5 10) (6 11) (7 9) (0 1) (2 4) (3 11) (5 8) (6 9) (7 10)

Description of knot 117 Description of knot 122


x
DllJ X = abefdb D\22 = abe~1ceb
Gauss word 112324563654 Gauss word 112324564635
Structural vector 120111 Structural vector 121020

(0 1) (2 4) (3 8) (5 11) (6 10) (7 9) (0 1) (2 4) (3 10) (5 8) (6 11) (7 9)

Description of knot 118 Description of knot 123


JDH8 X = abfcd2 D\23 X = abd^cdb
Gauss word 112324564356 Gauss word 112324564653
Structural vector 111201 Structural vector 121200

(0 1) (2 4) (3 9) (5 8) (6 10) (7 11) (0 1) (2 4) (3 11) (5 8) (6 10) (7 9)

Description of knot 119 Description of knot 124


x
DllQ X = abfcec D\24 = abfebd
Gauss word 112324564365 Gauss word 112324565346
Structural vector 112011 Structural vector 120111

(0 1) (2 4) (3 9) (5 8) (6 11) (7 10) (0 1) (2 4) (3 9) (5 10) (6 8) (7 11)


233
A. Dodecaphonic Knots

Description of knot 125 Description of knot 130


x
£>125 X = abfbc ^130 = abfe2a

Gauss word 112324565364 Gauss word 112324566345

Structural vector 121002 Structural vector 210021

(0 1) (2 4) (3 9) (5 11) (6 8) (7 10) (0 1) (2 4) (3 9) (5 10) (6 11) (7 8)

Description of knot 126 Description of knot 131


x
-^126 = obe~ldbd D\Sl X= abfda

Gauss word 112324565436 Gauss word 112324566354

Structural vector 120210 Structural vector 210102

(0 1) (2 4) (3 10) (5 9) (6 8) (7 11) (0 1) (2 4) (3 9) (5 11) (6 10) (7 8)

Description of knot 127 Description of knot 132


D\27 X = abdrldbc •^132 X = abe~ldea

Gauss word 112324565463 Gauss word 112324566435

Structural vector 121200 Structural vector 210120

(0 1) (2 4) (3 11) (5 9) (6 8) (7 10) (0 1) (2 4) (3 10) (5 9) (6 11) (7 8)

Description of knot 128 Description of knot 133


£>128 X = abe-'fb2 J D133 X = abd~xd2a

Gauss word 112324565634 Gauss word 112324566453

Structural vector 130011 Structural vector 210300

(0 1) (2 4) (3 10) (5 11) (6 8) (7 9) (0 1) (2 4) (3 11) (5 9) (6 10) (7 8)

Description of knot 129 Description of knot 134


D129 X = abd~xeb2 ^134 X= abe~1fca

Gauss word 112324565643 Gauss word 112324566534

Structural vector 130110 Structural vector 211011

(0 1) (2 4) (3 11) (5 10) (6 8) (7 9) (0 1) (2 4) (3 10) (5 11) (6 9) (7 8)


234
A. Dodecaphonic Knots

Description of knot 135 Description of knot 140


-^135 x
= Q>bd~1eca Z/140 X = acaeb2

Gauss word 112324566543 Gauss word 112332456564

Structural vector 211110 Structural vector 221010

(0 1) (2 4) (3 11) (5 10) (6 9) (7 8) (0 1) (2 5) (3 4) (6 11) (7 9) (8 10)

Description of knot 136 Description of knot 141


-^136 X = aca2ca Z/141 X = acad2a

Gauss word 112332445665 Gauss word 112332456645

Structural vector 402000 Structural vector 301200

(0 1) (2 5) (3 4) (6 7) (8 11) (9 10) (0 1) (2 5) (3 4) (6 10) (7 11) (8 9)

Description of knot 137 Description of knot 142


DiSf X = acac3 -L/142 X = acaeca

Gauss word 112332456456 Gauss word 112332456654

Structural vector 204000 Structural vector 302010

(0 1) (2 5) (3 4) (6 9) (7 10) (8 11) (0 1) (2 5) (3 4) (6 11) (7 10) (8 9)

Description of knot 138 Description of knot 143


•^138 X = acacdb -^143 X = adacda

Gauss word 112332456465 Gauss word 112334254665

Structural vector 212100 Structural vector 301200

(0 1) (2 5) (3 4) (6 9) (7 11) (8 10) (0 1) (2 6) (3 4) (5 8) (7 11) (9 10)

Description of knot 139 Description of knot 144


-^139 X = acadbc i-^144 X = adadc2

Gauss word 112332456546 Gauss word 112334256456

Structural vector 212100 Structural vector 202200

(0 1) (2 5) (3 4) (6 10) (7 9) (8 11) (0 1) (2 6) (3 4) (5 9) (7 10) (8 11)


235
A. Dodecaphonic Knots

Description of knot 145 Description of knot 150


Zyi45 X = adad2b -^150 X =
cieacea

Gauss word 112334256465 Gauss word 112334524665

Structural vector 210300 Structural vector 301020

(0 1) (2 6) (3 4) (5 9) (7 11) (8 10) (0 1) (2 7) (3 4) (5 8) (6 11) (9 10)

Description of knot 146 Description of knot 151


-^146 X = adaebc Z/151 X = aead2c

Gauss word 112334256546 Gauss word 112334526456

Structural vector 211110 Structural vector 201210

(0 1) (2 6) (3 4) (5 10) (7 9) (8 11) (0 1) (2 7) (3 4) (5 9) (6 10) (8 11)

Description of knot 147 Description of knot 152


Z)i47 X = adafb2 D\§2 x
= aeadeb

Gauss word 112334256564 Gauss word 112334526465

Structural vector 220101 Structural vector 210120

(0 1) (2 6) (3 4) (5 11) (7 9) (8 10) (0 1) (2 7) (3 4) (5 9) (6 11) (8 10)

Description of knot 148 Description of knot 153


-^148 X = adaeda 2^153 X = aeaec2

Gauss word 112334256645 Gauss word 112334526546

Structural vector 300210 Structural vector 202020

(0 1) (2 6) (3 4) (5 10) (7 11) (8 9) (0 1) (2 7) (3 4) (5 10) (6 9) (8 11)

Description of knot 149 Description of knot 154


-^149 X = adafca •^154 X = aeafcb

Gauss word 112334256654 Gauss word 112334526564

Structural vector 301101 Structural vector 211011

(0 1) (2 6) (3 4) (5 11) (7 10) (8 9) (0 1) (2 7) (3 4) (5 11) (6 9) (8 10)


236
A. Dodecaphonic Knots

Description of knot 155 Description of knot 160


Z/155 X = aeae2a £>160 x
= a
Mc2
Gauss word 112334526645 Gauss word 112334562564
Structural vector 300030 Structural vector 202002

(0 1) (2 7) (3 4) (5 10) (6 11) (8 9) (0 1) (2 8) (3 4) (5 11) (6 9) (7 10)

Description of knot 156 Description of knot 161


-^156 X = aeafda DiQi X = afae2b

Gauss word 112334526654 Gauss word 112334562645


Structural vector 300111 Structural vector 210021

(0 1) (2 7) (3 4) (5 11) (6 10) (8 9) (0 1) (2 8) (3 4) (5 10) (6 11) (7 9)

Description of knot 157 Description of knot 162


D157 X = afafa2 Di62 X = afafdb
Gauss word 112334552664 Gauss word 112334562654
Structural vector 400002 Structural vector 210102

(0 1) (2 8) (3 4) (5 11) (6 7) (9 10) (0 1) (2 8) (3 4) (5 11) (6 10) (7 9)

Description of knot 158 Description of knot 163


£>158 X = afad3 Z?163 X = ac2ec2

Gauss word 112334562456 Gauss word 112342356456


Structural vector 200301 Structural vector 104010

(0 1) (2 8) (3 4) (5 9) (6 10) (7 11) (0 1) (2 5) (3 6) (4 9) (7 10) (8 11)

Description of knot 159 Description of knot 164


D159 X = afadec Z/164 X = ac2edb
Gauss word 112334562465 Gauss word 112342356465
Structural vector 201111 Structural vector 112110

(0 1) (2 8) (3 4) (5 9) (6 11) (7 10) (0 1) (2 5) (3 6) (4 9) (7 11) (8 10)


237
A. Dodecaphonic Knots

Description of knot 165 Description of knot 170


D\§§ X = ac2fbc D\Y0 x
= o,ce~1b2c

Gauss word 112342356546 Gauss word 112342456536

Structural vector 113001 Structural vector 122010

(0 1) (2 5) (3 6) (4 10) (7 9) (8 11) (0 1) (2 5) (3 10) (4 6) (7 9) (8 11)

Description of knot 166 Description of knot 171


Z?i66 X = afe^b2 Dtfl X = acd^b3

Gauss word 112342356564 Gauss word 112342456563

Structural vector 122010 Structural vector 131100

(0 1) (2 5) (3 6) (4 11) (7 9) (8 10) (0 1) (2 5) (3 11) (4 6) (7 9) (8 10)

Description of knot 167 Description of knot 172


D\§j X = ac2fda D\72 x
= Q>ce~lbda

Gauss word 112342356645 Gauss word 112342456635

Structural vector 202101 Structural vector 211110

(0 1) (2 5) (3 6) (4 10) (7 11) (8 9) (0 1) (2 5) (3 10) (4 6) (7 11) (8 9)

Description of knot 168 Description of knot 173


x
-^168 = o,c2e~1ca ^173 x
= acd~1bca

Gauss word 112342356654 Gauss word 112342456653

Structural vector 203010 Structural vector 212100

(0 1) (2 5) (3 6) (4 11) (7 10) (8 9) (0 1) (2 5) (3 11) (4 6) (7 10) (8 9)

Description of knot 169 Description of knot 174


x x =
-^169 = acfbdb -^174 acdedc

Gauss word 112342456365 Gauss word 112342536456

Structural vector 121101 Structural vector 102210

(0 1) (2 5) (3 9) (4 6) (7 11) (8 10) (0 1) (2 5) (3 7) (4 9) (6 10) (8 11)


238
A. Dodecaphonic Knots

Description of knot 175 Description of knot 180


Dyj§ X = acde2b ^180 X = acfceb

Gauss word 112342536465 Gauss word 112342546365

Structural vector 111120 Structural vector 112011

(0 1) (2 5) (3 7) (4 9) (6 11) (8 10) (0 1) (2 5) (3 9) (4 7) (6 11) (8 10)

Description of knot 176 Description of knot 181


DIJQ X = acdfc2 .Digi X = ace~x<?

Gauss word 112342536546 Gauss word 112342546536

Structural vector 103101 Structural vector 104010

(0 1) (2 5) (3 7) (4 10) (6 9) (8 11) (0 1) (2 5) (3 10) (4 7) (6 9) (8 11)

Description of knot 177 Description of knot 182


Z/177 X = acde _ 1 c6 £>182 X = acd~lc2b

Gauss word 112342536564 Gauss word 112342546563

Structural vector 112110 Structural vector 113100

(0 1) (2 5) (3 7) (4 11) (6 9) (8 10) (0 1) (2 5) (3 11) (4 7) (6 9) (8 10)

Description of knot 178 Description of knot 183


x =
-Dl78 x
= acdfea -^183 ace~1cea

Gauss word 112342536645 Gauss word 112342546635

Structural vector 201111 Structural vector 202020

(0 1) (2 5) (3 7) (4 10) (6 11) (8 9) (0 1) (2 5) (3 10) (4 7) (6 11) (8 9)

Description of knot 179 Description of knot 184


Z/179 X = acde~lda •^184 x
= a>cd~1cda

Gauss word 112342536654 Gauss word 112342546653

Structural vector 201210 Structural vector 202200

(0 1) (2 5) (3 7) (4 11) (6 10) (8 9) (0 1) (2 5) (3 11) (4 7) (6 10) (8 9)


239
A. Dodecaphonic Knots

Description of knot 185 Description of knot 190


-^185 x
= acfac D 190 X = acee l
c?

Gauss word 112342556346 Gauss word 112342563564

Structural vector 202002 Structural vector 103020

(0 1) (2 5) (3 9) (4 10) (6 7) (8 11) (0 1) (2 5) (3 8) (4 11) (6 9) (7 10)

Description of knot 186 Description of knot 191


-^186 X = ace~1eac •^191 X = acefeb

Gauss word 112342556436 Gauss word 112342563645

Structural vector 202020 Structural vector 111021

(0 1) (2 5) (3 10) (4 9) (6 7) (8 11) (0 1) (2 5) (3 8) (4 10) (6 11) (7 9)

Description of knot 187 Description of knot 192


DiS7 X = ace2d2 D\g2 X= acee~ldb

Gauss word 112342563456 Gauss word 112342563654

Structural vector 101220 Structural vector 111120

(0 1) (2 5) (3 8) (4 9) (6 10) (7 11) (0 1) (2 5) (3 8) (4 11) (6 10) (7 9)

Description of knot 188 Description of knot 193


Z/188 X = acezc £>193 X = acfd3

Gauss word 112342563465 Gauss word 112342564356

Structural vector 102030 Structural vector 101301

(0 1) (2 5) (3 8) (4 9) (6 11) (7 10) (0 1) (2 5) (3 9) (4 8) (6 10) (7 11)

Description of knot 189 Description of knot 194


Z/189 X = acefcd -^194 X= acfdec

Gauss word 112342563546 Gauss word 112342564365

Structural vector 102111 Structural vector 102111

(0 1) (2 5) (3 8) (4 10) (6 9) (7 11) (0 1) (2 5) (3 9) (4 8) (6 11) (7 10)


240
A. Dodecaphonic Knots

Description of knot 195 Description of knot 200


x
-^195 = ace~1dcd ^200 x
= ac e
f lbc

Gauss word 112342564536 Gauss word 112342565364

Structural vector 102210 Structural vector 112011

(0 1) (2 5) (3 10) (4 8) (6 9) (7 11) (0 1) (2 5) (3 9) (4 11) (6 8) (7 10)

Description of knot 196 Description of knot 201


Dl96 X = acd~1dc2 •^201 x =
ace~1ebd

Gauss word 112342564563 Gauss word 112342565436

Structural vector 103200 Structural vector 111120

(0 1) (2 5) (3 11) (4 8) (6 9) (7 10) (0 1) (2 5) (3 10) (4 9) (6 8) (7 11)

Description of knot 197 Description of knot 202


Z/197 X = ace~1deb -^202 x
= acd~1ebc

Gauss word 112342564635 Gauss word 112342565463

Structural vector 111120 Structural vector 112110

(0 1) (2 5) (3 10) (4 8) (6 11) (7 9) (0 1) (2 5) (3 11) (4 9) (6 8) (7 10)

Description of knot 198 Description of knot 203


Digs x
= acd~ld2b D203 X = ace~2b2

Gauss word 112342564653 Gauss word 112342565634

Structural vector 111300 Structural vector 121020

(0 1) (2 5) (3 11) (4 8) (6 10) (7 9) (0 1) (2 5) (3 10) (4 11) (6 8) (7 9)

Description of knot 199 Description of knot 204


Z)i99 X = acfbd D204 X = aed-ifb2

Gauss word 112342565346 Gauss word 112342565643

Structural vector 111102 Structural vector 121101

(0 1) (2 5) (3 9) (4 10) (6 8) (7 11) (0 1) (2 5) (3 11) (4 10) (6 8) (7 9)


241
A. Dodecaphonic Knots

Description of knot 205 Description of knot 210


x
^205 x
= o,cf2ea •^210 = acd~1fca

Gauss word 112342566345 Gauss word 112342566543

Structural vector 201012 Structural vector 202101

(0 1) (2 5) (3 9) (4 10) (6 11) (7 8) (0 1) (2 5) (3 11) (4 10) (6 9) (7 8)

Description of knot 206 Description of knot 211


x
-^206 = acfe~1da D2II x
= adbedb

Gauss word 112342566354 Gauss word 112343256465

Structural vector 201111 Structural vector 120210

(0 1) (2 5) (3 9) (4 11) (6 10) (7 8) (0 1) (2 6) (3 5) (4 9) (7 11) (8 10)

Description of knot 207 Description of knot 212


D 207 x
= ace 1 2
ea D212 x
= adbe-W

Gauss word 112342566435 Gauss word 112343256564

Structural vector 201030 Structural vector 130110

(0 1) (2 5) (3 10) (4 9) (6 11) (7 8) (0 1) (2 6) (3 5) (4 11) (7 9) (8 10)

Description of knot 208 Description of knot 213


x
-^208 = acd~1eda -^213 x
= adbfda

Gauss word 112342566453 Gauss word 112343256645

Structural vector 201210 Structural vector 210201

(0 1) (2 5) (3 11) (4 9) (6 10) (7 8) (0 1) (2 6) (3 5) (4 10) (7 11) (8 9)

Description of knot 209 Description of knot 214


x
-^209 x =
<*>ce~2ca Z/214 = adbe~1ca

Gauss word 112342566534 Gauss word 112343256654

Structural vector 202020 Structural vector 211110

(0 1) (2 5) (3 10) (4 11) (6 9) (7 8) (0 1) (2 6) (3 5) (4 11) (7 10) (8 9)


242
A. Dodecaphonic Knots

Description of knot 215 Description of knot 220


Z>215 X = ac^b4 £*220 x
= aebfea
Gauss word 112343456562 Gauss word 112343526645
Structural vector 141000 Structural vector 210021

(0 1) (2 11) (3 5) (4 6) (7 9) (8 10) (0 1) (2 7) (3 5) (4 10) (6 11) (8 9)

Description of knot 216 Description of knot 221


D2IQ X = ad~lb2da L/221 X = aebe~1da
Gauss word 112343456625 Gauss word 112343526654
Structural vector 220200 Structural vector 210120

(0 1) (2 10) (3 5) (4 6) (7 11) (8 9) (0 1) (2 7) (3 5) (4 11) (6 10) (8 9)

Description of knot 217 Description of knot 222


D2YI X = ac~1b2ca D222 X= ac^b^b
Gauss word 112343456652 Gauss word 112343546562
Structural vector 222000 Structural vector 123000

(0 1) (2 11) (3 5) (4 6) (7 10) (8 9) (0 1) (2 11) (3 5) (4 7) (6 9) (8 10)

Description of knot 218 Description of knot 223


-D2I8 x
= ae&e26 D223 X = afbed2
Gauss word 112343526465 Gauss word 112343562456
Structural vector 120030 Structural vector 110211

(0 1) (2 7) (3 5) (4 9) (6 11) (8 10) (0 1) (2 8) (3 5) (4 9) (6 10) (7 11)

Description of knot 219 Description of knot 224


-^219 X = aebe~lcb L>224 X = afbe2c
Gauss word 112343526564 Gauss word 112343562465
Structural vector 121020 Structural vector 111021

(0 1) (2 7) (3 5) (4 11) (6 9) (8 10) (0 1) (2 8) (3 5) (4 9) (6 11) (7 10)


243
A. Dodecaphonic Knots

Description of knot 225 Description of knot 230


£>225 X = afbfcd ^230 X = ae~1bdec

Gauss word 112343562546 Gauss word 112343564265

Structural vector 111102 Structural vector 111120

(0 1) (2 8) (3 5) (4 10) (6 9) (7 11) (0 1) (2 9) (3 5) (4 8) (6 11) (7 10)

Description of knot 226 Description of knot 231


£>226 X = afbe-lc2 -^231 X = ad~lbdcd

Gauss word 112343562564 Gauss word 112343564526

Structural vector 112011 Structural vector 111300

(0 1) (2 8) (3 5) (4 11) (6 9) (7 10) (0 1) (2 10) (3 5) (4 8) (6 9) (7 11)

Description of knot 227 Description of knot 232


D227 X = afbfeb £>232 X = ac~lbdc2

Gauss word 112343562645 Gauss word 112343564562

Structural vector 120012 Structural vector 113100

(0 1) (2 8) (3 5) (4 10) (6 11) (7 9) (0 1) (2 11) (3 5) (4 8) (6 9) (7 10)

Description of knot 228 Description of knot 233


D22S X = afbe~ldb £>233 X = ad^bdeb

Gauss word 112343562654 Gauss word 112343564625

Structural vector 120111 Structural vector 120210

(0 1) (2 8) (3 5) (4 11) (6 10) (7 9) (0 1) (2 10) (3 5) (4 8) (6 11) (7 9)

Description of knot 229 Description of knot 234


D229 X = ae^bd3 £)234 X = ac~xbd2b

Gauss word 112343564256 Gauss word 112343564652

Structural vector 110310 Structural vector 121200

(0 1) (2 9) (3 5) (4 8) (6 10) (7 11) (0 1) (2 11) (3 5) (4 8) (6 10) (7 9)


244
A. Dodecaphonic Knots

Description of knot 235 Description of knot 240


x x
£>235 = ae-'bfbd D240 = ac-'bfb2
Gauss word 112343565246 Gauss word 112343565642
Structural vector 120111 Structural vector 131001

(0 1) (2 9) (3 5) (4 10) (6 8) (7 11) (0 1) (2 11) (3 5) (4 10) (6 8) (7 9)

Description of knot 236 Description of knot 241


x =
^236 o,e~lbe~1bc -D241 x
= ae~1bfea
Gauss word 112343565264 Gauss word 112343566245
Structural vector 121020 Structural vector 210021

(0 1) (2 9) (3 5) (4 11) (6 8) (7 10) (0 1) (2 9) (3 5) (4 10) (6 11) (7 8)

Description of knot 237 Description of knot 242


D237 X = ad~lbebd D242 x
= ae~1be~1da
Gauss word 112343565426 Gauss word 112343566254
Structural vector 120210 Structural vector 210120

(0 1) (2 10) (3 5) (4 9) (6 8) (7 11) (0 1) (2 9) (3 5) (4 11) (6 10) (7 8)

Description of knot 238 Description of knot 243


x
•^238 x
= ac_16e6c 243 = ac~1beda
Gauss word 112343565462 Gauss word 112343566452
Structural vector 122010 Structural vector 211110

(0 1) (2 11) (3 5) (4 9) (6 8) (7 10) (0 1) (2 11) (3 5) (4 9) (6 10) (7 8)

Description of knot 239 Description of knot 244


D239 X = ad~lbe~lb2 2^244 X = ac~1bfca
Gauss word 112343565624 Gauss word 112343566542

Structural vector 130110 Structural vector 212001

(0 1) (2 10) (3 5) (4 11) (6 8) (7 9) (0 1) (2 11) (3 5) (4 10) (6 9) (7 8)


245
A. Dodecaphonic Knots

Description of knot 245 Description of knot 250


-^245 X = ade~1ada Z?250 x
= afaec

Gauss word 112344256635 Gauss word 112344562365

Structural vector 300210 Structural vector 201012

(0 1) (2 6) (3 10) (4 5) (7 11) (8 9) (0 1) (2 8) (3 9) (4 5) (6 11) (7 10)

Description of knot 246 Description of knot 251


-^246 X = add~1aca -^251 x
= ae~1ead2

Gauss word 112344256653 Gauss word 112344563256

Structural vector 301200 Structural vector 200220

(0 1) (2 6) (3 11) (4 5) (7 10) (8 9) (0 1) (2 9) (3 8) (4 5) (6 10) (7 11)

Description of knot 247 Description of knot 252


J-^247 X = ac~1caca •^252 X = ae~1eaec

Gauss word 112344356652 Gauss word 112344563265

Structural vector 303000 Structural vector 201030

(0 1) (2 11) (3 6) (4 5) (7 10) (8 9) (0 1) (2 9) (3 8) (4 5) (6 11) (7 10)

Description of knot 248 Description of knot 253


-^248 X =
aee~1aea Z?253 X = ade3d
Gauss word 112344526635 Gauss word 112345263456

Structural vector 300030 Structural vector 100230

(0 1) (2 7) (3 10) (4 5) (6 11) (8 9) (0 1) (2 6) (3 8) (4 9) (5 10) (7 11)

Description of knot 249 Description of knot 254


£>249 X = a fad2 £>254 X = ade2fc

Gauss word 112344562356 Gauss word 112345263465

Structural vector 200202 Structural vector 101121

(0 1) (2 8) (3 9) (4 5) (6 10) (7 11) (0 1) (2 6) (3 8) (4 9) (5 11) (7 10)


246
A. Dodecaphonic Knots

Description of knot 255 Description of knot 260


x
£>255 = adefd2 ^260 x
= ade~1d3
Gauss word 112345263546 Gauss word 112345264536
Structural vector 100311 Structural vector 100410

(0 1) (2 6) (3 8) (4 10) (5 9) (7 11) (0 1) (2 6) (3 10) (4 8) (5 9) (7 11)

Description of knot 256 Description of knot 261


•^256 x =
a>dee~ldc D26I X = add~1d2c

Gauss word 112345263564 Gauss word 112345264563


Structural vector 101220 Structural vector 101400

(0 1) (2 6) (3 8) (4 11) (5 9) (7 10) (0 1) (2 6) (3 11) (4 8) (5 9) (7 10)

Description of knot 257 Description of knot 262

D257 x
= adef2b D262 X = ade-'dfb
Gauss word 112345263645 Gauss word 112345264635

Structural vector 110112 Structural vector 110211

(0 1) (2 6) (3 8) (4 10) (5 11) (7 9) (0 1) (2 6) (3 10) (4 8) (5 11) (7 9)

Description of knot 258 Description of knot 263


x
^258 = Q>dee~1eb -^263 x
= add~1deb
Gauss word 112345263654 Gauss word 112345264653
Structural vector 110130 Structural vector 110310

(0 1) (2 6) (3 8) (4 11) (5 10) (7 9) (0 1) (2 6) (3 11) (4 8) (5 10) (7 9)

Description of knot 259 Description of knot 264


x
£^259 = adfdfc 1^264 X = adfcd

Gauss word 112345264365 Gauss word 112345265346


Structural vector 101202 Structural vector 101202

(0 1) (2 6) (3 9) (4 8) (5 11) (7 10) (0 1) (2 6) (3 9) (4 10) (5 8) (7 11)


247
A. Dodecaphonic Knots

Description of knot 265 Description of knot 270


£)265 X = adfe-'c2 L>270 x
= adfa

Gauss word 112345265364 Gauss word 112345266345

Structural vector 102111 Structural vector 200103

(0 1) (2 6) (3 9) (4 11) (5 8) (7 10) (0 1) (2 6) (3 9) (4 10) (5 11) (7 8)

Description of knot 266 Description of knot 271


-^266 x =
ade~1ecd Z/271 X = adfe~1ea

Gauss word 112345265436 Gauss word 112345266354

Structural vector 101220 Structural vector 200121

(0 1) (2 6) (3 10) (4 9) (5 8) (7 11) (0 1) (2 6) (3 9) (4 11) (5 10) (7 8)

Description of knot 267 Description of knot 272


^267 x
= add~1ec2 D272 X = add~1e2a

Gauss word 112345265463 Gauss word 112345266453

Structural vector 102210 Structural vector 200220

(0 1) (2 6) (3 11) (4 9) (5 8) (7 10) (0 1) (2 6) (3 11) (4 9) (5 10) (7 8)

Description of knot 268 Description of knot 273


-^268 x =
ade~2cb -^273 X = add'1 f da

Gauss word 112345265634 Gauss word 112345266543

Structural vector 111120 Structural vector 200301

(0 1) (2 6) (3 10) (4 11) (5 8) (7 9) (0 1) (2 6) (3 11) (4 10) (5 9) (7 8)

Description of knot 269 Description of knot 274


=
-^269 X a>dd~lfcb ^274 X = afcefc

Gauss word 112345265643 Gauss word 112345362465

Structural vector 111201 Structural vector 102012

(0 1) (2 6) (3 11) (4 10) (5 8) (7 9) (0 1) (2 8) (3 6) (4 9) (5 11) (7 10)


248
A. Dodecaphonic Knots

Description of knot 275 Description of knot 280


Z/275 X = afce~1dc ^280 x
= ac~1cdeb

Gauss word 112345362564 Gauss word 112345364652

Structural vector 102111 Structural vector 112110

(0 1) (2 8) (3 6) (4 11) (5 9) (7 10) (0 1) (2 11) (3 6) (4 8) (5 10) (7 9)

Description of knot 276 Description of knot 281


£>276 X = afcfb D 281 X = ae 1
ce 1 2
c

Gauss word 112345362645 Gauss word 112345365264

Structural vector 111003 Structural vector 103020

(0 1) (2 8) (3 6) (4 10) (5 11) (7 9) (0 1) (2 9) (3 6) (4 11) (5 8) (7 10)

Description of knot 277 Description of knot 282


Z/277 X = afce~1eb -^282 x
= o,c~lce<?

Gauss word 112345362654 Gauss word 112345365462

Structural vector 111021 Structural vector 104010

(0 1) (2 8) (3 6) (4 11) (5 10) (7 9) (0 1) (2 11) (3 6) (4 9) (5 8) (7 10)

Description of knot 278 Description of knot 283


^278 x
= ac~1cd2c -^283 X = ad~1ce~1cb

Gauss word 112345364562 Gauss word 112345365624

Structural vector 103200 Structural vector 112110

(0 1) (2 11) (3 6) (4 8) (5 9) (7 10) (0 1) (2 10) (3 6) (4 11) (5 8) (7 9)

Description of knot 279 Description of knot 284


D279 X = adTlcdfb -^284 X = ac~1cfcb

Gauss word 112345364625 Gauss word 112345365642

Structural vector 111201 Structural vector 113001

(0 1) (2 10) (3 6) (4 8) (5 11) (7 9) (0 1) (2 11) (3 6) (4 10) (5 8) (7 9)


249
A. Dodecaphonic Knots

Description of knot 285 Description of knot 290


£>285 X = afe~1bfb £>290 x
= ae5
Gauss word 112345462635 Gauss word 112345623456
Structural vector 120012 Structural vector 100050

(0 1) (2 8) (3 10) (4 6) (5 11) (7 9) (0 1) (2 7) (3 8) (4 9) (5 10) (6 11)

Description of knot 286 Description of knot 291


£>286 X = afd~1beb £>291 X = ae3fd
Gauss word 112345462653 Gauss word 112345623465
Structural vector 120111 Structural vector 100131

(0 1) (2 8) (3 11) (4 6) (5 10) (7 9) (0 1) (2 7) (3 8) (4 9) (5 11) (6 10)

Description of knot 287 Description of knot 292


x
^287 = ac~1ebeb £>292 X = ae2fde
Gauss word 112345463652 Gauss word 112345623546
Structural vector 121020 Structural vector 100131

(0 1) (2 11) (3 8) (4 6) (5 10) (7 9) (0 1) (2 7) (3 8) (4 10) (5 9) (6 11)

Description of knot 288 Description of knot 293


D288 x
= adT2bcb D293 X = ae2e~1d2
Gauss word 112345465623 Gauss word 112345623564
Structural vector 121200 Structural vector 100230

(0 1) (2 10) (3 11) (4 6) (5 8) (7 9) (0 1) (2 7) (3 8) (4 11) (5 9) (6 10)

Description of knot 289 Description of knot 294


x
^289 = ac~1e~1bcb £>294 X = ae2f2c
Gauss word 112345465632 Gauss word 112345623645
Structural vector 122010 Structural vector 101022

(0 1) (2 11) (3 10) (4 6) (5 8) (7 9) (0 1) (2 7) (3 8) (4 10) (5 11) (6 9)


250
A. Dodecaphonic Knots

Description of knot 295 Description of knot 300


x =
^295 x =
ae2e-1ec -^300 o,ed~1dec

Gauss word 112345623654 Gauss word 112345624653

Structural vector 101040 Structural vector 101220

(0 1) (2 7) (3 8) (4 11) (5 10) (6 9) (0 1) (2 7) (3 11) (4 8) (5 10) (6 9)

Description of knot 296 Description of knot 301


x x
-^296 = o,efdfd -^301 = aefce

Gauss word 112345624365 Gauss word 112345625346

Structural vector 100212 Structural vector 101022

(0 1) (2 7) (3 9) (4 8) (5 11) (6 10) (0 1) (2 7) (3 9) (4 10) (5 8) (6 11)

Description of knot 297 Description of knot 302


x
Z/297 = aee~1d2e ^302 x
= aefe~1cd

Gauss word 112345624536 Gauss word 112345625364

Structural vector 100230 Structural vector 101121

(0 1) (2 7) (3 10) (4 8) (5 9) (6 11) (0 1) (2 7) (3 9) (4 11) (5 8) (6 10)

Description of knot 298 Description of knot 303


x
£>298 = aed^d3 -^303 x =
aee - 1 ece

Gauss word 112345624563 Gauss word 112345625436

Structural vector 100410 Structural vector 101040

(0 1) (2 7) (3 11) (4 8) (5 9) (6 10) (0 1) (2 7) (3 10) (4 9) (5 8) (6 11)

Description of knot 299 Description of knot 304


x
-^299 = aee~1dfc -^304 x =
aed~1ecd

Gauss word 112345624635 Gauss word 112345625463

Structural vector 101121 Structural vector 101220

(0 1) (2 7) (3 10) (4 8) (5 11) (6 9) (0 1) (2 7) (3 11) (4 9) (5 8) (6 10)


251
A. Dodecaphonic Knots

Description of knot 305 Description of knot 310


£>305 x
= aee-2c2 D310 X = aed~1e2b

Gauss word 112345625634 Gauss word 112345626453

Structural vector 102030 Structural vector 110130

(0 1) (2 7) (3 10) (4 11) (5 8) (6 9) (0 1) (2 7) (3 11) (4 9) (5 10) (6 8)

Description of knot 306 Description of knot 311


£>306 X = aedTlfc2 -^311 X = aee~2db

Gauss word 112345625643 Gauss word 112345626534

Structural vector 102111 Structural vector 110130

(0 1) (2 7) (3 11) (4 10) (5 8) (6 9) (0 1) (2 7) (3 10) (4 11) (5 9) (6 8)

Description of knot 307 Description of knot 312


Dm X = aefb J D312 X = aed^fdb

Gauss word 112345626345 Gauss word 112345626543

Structural vector 110013 Structural vector 110211

(0 1) (2 7) (3 9) (4 10) (5 11) (6 8) (0 1) (2 7) (3 11) (4 10) (5 9) (6 8)

Description of knot 308 Description of knot 313


D3O8 X = aefe~1eb £>313 X = afdefd

Gauss word 112345626354 Gauss word 112345632465

Structural vector 110031 Structural vector 100212

(0 1) (2 7) (3 9) (4 11) (5 10) (6 8) (0 1) (2 8) (3 7) (4 9) (5 11) (6 10)

Description of knot 309 Description of knot 314


x
^309 = aee~1efb ^314 X = afdefd2

Gauss word 112345626435 Gauss word 112345632564

Structural vector 110031 Structural vector 100311

(0 1) (2 7) (3 10) (4 9) (5 11) (6 8) (0 1) (2 8) (3 7) (4 11) (5 9) (6 10)


252
A. Dodecaphonic Knots

Description of knot 315 Description of knot 320


£>315 X= afdfc ^320 X = ae~1de~1cd
Gauss word 112345632645 Gauss word 112345635264
Structural vector 101103 Structural vector 101220

(0 1) (2 8) (3 7) (4 10) (5 11) (6 9) (0 1) (2 9) (3 7) (4 11) (5 8) (6 10)

Description of knot 316 Description of knot 321


D%IQ X = afde~1ec ^321 X = ac~1decd
Gauss word 112345632654 Gauss word 112345635462
Structural vector 101121 Structural vector 102210

(0 1) (2 8) (3 7) (4 11) (5 10) (6 9) (0 1) (2 11) (3 7) (4 9) (5 8) (6 10)

Description of knot 317 Description of knot 322


^317 X = ac^d4 £)322 X= ad^de^c2
Gauss word 112345634562 Gauss word 112345635624
Structural vector 101400 Structural vector 102210

(0 1) (2 11) (3 7) (4 8) (5 9) (6 10) (0 1) (2 10) (3 7) (4 11) (5 8) (6 9)

Description of knot 318 Description of knot 323


DsiS X = ad~1d2fc D323 X = ac-'dfc2
Gauss word 112345634625 Gauss word 112345635642
Structural vector 101301 Structural vector 103101

(0 1) (2 10) (3 7) (4 8) (5 11) (6 9) (0 1) (2 11) (3 7) (4 10) (5 8) (6 9)

Description of knot 319 Description of knot 324


-D319 X = ac~1d2ec D324 X= ae-'dfb
Gauss word 112345634652 Gauss word 112345636245
Structural vector 102210 Structural vector 110112

(0 1) (2 11) (3 7) (4 8) (5 10) (6 9) (0 1) (2 9) (3 7) (4 10) (5 11) (6 8)


253
A. Dodecaphonic Knots

Description of knot 325 Description of knot 330


.D325 X = ae~1de~1eb £>330 X = afe-'cfc

Gauss word 112345636254 Gauss word 112345642635

Structural vector 110130 Structural vector 102012

(0 1) (2 9) (3 7) (4 11) (5 10) (6 8) (0 1) (2 8) (3 10) (4 7) (5 11) (6 9)

Description of knot 326 Description of knot 331


x
£>326 X = ad-'defb ^331 = afd~1cec

Gauss word 112345636425 Gauss word 112345642653

Structural vector 110211 Structural vector 102111

(0 1) (2 10) (3 7) (4 9) (5 11) (6 8) (0 1) (2 8) (3 11) (4 7) (5 10) (6 9)

Description of knot 327 Description of knot 332


D327 X = ac^de^ ^332 X = ac~1ecec

Gauss word 112345636452 Gauss word 112345643652

Structural vector 111120 Structural vector 103020

(0 1) (2 11) (3 7) (4 9) (5 10) (6 8) (0 1) (2 11) (3 8) (4 7) (5 10) (6 9)

Description of knot 328 Description of knot 333


D328 x
= ad-lde-xdb D333 X = ad~2c3

Gauss word 112345636524 Gauss word 112345645623

Structural vector 110310 Structural vector 103200

(0 1) (2 10) (3 7) (4 11) (5 9) (6 8) (0 1) (2 10) (3 11) (4 7) (5 8) (6 9)

Description of knot 329 Description of knot 334


£>329 X = ac-'dfdb D334 X = ac^e^c3

Gauss word 112345636542 Gauss word 112345645632

Structural vector 111201 Structural vector 104010

(0 1) (2 11) (3 7) (4 10) (5 9) (6 8) (0 1) (2 11) (3 10) (4 7) (5 8) (6 9)


254
A. Dodecaphonic Knots

Description of knot 335 Description of knot 340


£>335 X = ae~2cfb D340 X = ac~1e~1cdb
Gauss word 112345646235 Gauss word 112345646532
Structural vector 111021 Structural vector 112110

(0 1) (2 9) (3 10) (4 7) (5 11) (6 8) (0 1) (2 11) (3 10) (4 7) (5 9) (6 8)

Description of knot 336 Description of knot 341


Z/336 X = ae~ld~1ceb D34I X = ae~3b2
Gauss word 112345646253 Gauss word 112345656234
Structural vector 111120 Structural vector 120030

(0 1) (2 9) (3 11) (4 7) (5 10) (6 8) (0 1) (2 9) (3 10) (4 11) (5 7) (6 8)

Description of knot 337 Description of knot 342


D337 X= ad-'fcfb £>342 X= ae-xd-lfb2
Gauss word 112345646325 Gauss word 112345656243
Structural vector 111102 Structural vector 120111

(0 1) (2 10) (3 9) (4 7) (5 11) (6 8) (0 1) (2 9) (3 11) (4 10) (5 7) (6 8)

Description of knot 338 Description of knot 343


-^338 x
= ac~1fceb £>343 X= ac-'fb2
Gauss word 112345646352 Gauss word 112345656342
Structural vector 112011 Structural vector 121002

(0 1) (2 11) (3 9) (4 7) (5 10) (6 8) (0 1) (2 11) (3 9) (4 10) (5 7) (6 8)

Description of knot 339 Description of knot 344


Z?339 X = ad~2cdb D344 X = ad~2eb2

Gauss word 112345646523 Gauss word 112345656423

Structural vector 111300 Structural vector 120210

(0 1) (2 10) (3 11) (4 7) (5 9) (6 8) (0 1) (2 10) (3 11) (4 9) (5 7) (6 8)


255
A. Dodecaphonic Knots

Description of knot 345 Description of knot 350


D345 X= ac-le~leb2 £>350 x
= afd-'fda
Gauss word 112345656432 Gauss word 112345662543
Structural vector 121020 Structural vector 200202

(0 1) (2 11) (3 10) (4 9) (5 7) (6 8) (0 1) (2 8) (3 11) (4 10) (5 9) (6 7)

Description of knot 346 Description of knot 351


x
£>346 x
= a f 4* •£\351 = ae~1ee~1ea

Gauss word 112345662345 Gauss word 112345663254


Structural vector 200004 Structural vector 200040

(0 1) (2 8) (3 9) (4 10) (5 11) (6 7) (0 1) (2 9) (3 8) (4 11) (5 10) (6 7)

Description of knot 347 Description of knot 352


x
D347 X= afe^ea ^352 = ac~1e3a
Gauss word 112345662354 Gauss word 112345663452
Structural vector 200022 Structural vector 201030

(0 1) (2 8) (3 9) (4 11) (5 10) (6 7) (0 1) (2 11) (3 8) (4 9) (5 10) (6 7)

Description of knot 348 Description of knot 353


x
^348 x
= afe~1efa 1)353 = ad~1ee~1da

Gauss word 112345662435 Gauss word 112345663524


Structural vector 200022 Structural vector 200220

(0 1) (2 8) (3 10) (4 9) (5 11) (6 7) (0 1) (2 10) (3 8) (4 11) (5 9) (6 7)

Description of knot 349 Description of knot 354


-0349 X = afd~1e2a ^354 x
= ac~1efda
Gauss word 112345662453 Gauss word 112345663542
Structural vector 200121 Structural vector 201111

(0 1) (2 8) (3 11) (4 9) (5 10) (6 7) (0 1) (2 11) (3 8) (4 10) (5 9) (6 7)


256
A. Dodecaphonic Knots

Description of knot 355 Description of knot 360


D355 X = ad~2d2a jD360 X = b 3c2b

Gauss word 112345664523 Gauss word 121234354656

Structural vector 200400 Structural vector 042000

(0 1) (2 10) (3 11) (4 8) (5 9) (6 7) (0 2) (1 3) (4 6) (5 8) (7 10) (9 11)

Description of knot 356 Description of knot 361


Z?35g X = ac~1e~1d2a ^361 x
= *>3dc2

Gauss word 112345664532 Gauss word 121234356456

Structural vector 201210 Structural vector 032100

(0 1) (2 11) (3 10) (4 8) (5 9) (6 7) (0 2) (1 3) (4 6) (5 9) (7 10) (8 11)

Description of knot 357 Description of knot 362


D357 X = ac^fca £>362 X = b 3d 2b

Gauss word 112345665342 Gauss word 121234356465

Structural vector 202002 Structural vector 040200

(0 1) (2 11) (3 9) (4 10) (5 8) (6 7) (0 2) (1 3) (4 6) (5 9) (7 11) (8 10)

Description of knot 358 Description of knot 363


Z/358 X = ac~1e~1eca D363 X = b3ebc

Gauss word 112345665432 Gauss word 121234356546

Structural vector 202020 Structural vector 041010

(0 1) (2 11) (3 10) (4 9) (5 8) (6 7) (0 2) (1 3) (4 6) (5 10) (7 9) (8 11)

Description of knot 359 Description of knot 364


£>359 X = 66 £>364 X = b3fb2

Gauss word 121234345656 Gauss word 121234356564

Structural vector 060000 Structural vector 050001

(0 2) (1 3) (4 6) (5 7) (8 10) (9 11) (0 2) (1 3) (4 6) (5 11) (7 9) (8 10)


257
A. Dodecaphonic Knots

Description of knot 365 Description of knot 370


£>365 X = b2cd2c DtfQ X = b2e-xbcb

Gauss word 121234536456 Gauss word 121234546563

Structural vector 022200 Structural vector 041010

(0 2) (1 3) (4 7) (5 9) (6 10) (8 11) (0 2) (1 3) (4 11) (5 7) (6 9) (8 10)

Description of knot 366 Description of knot 371


DffiQ X = b2cdeb £>371 X = b2d4
Gauss word 121234536465 Gauss word 121234563456
Structural vector 031110 Structural vector 020400

(0 2) (1 3) (4 7) (5 9) (6 11) (8 10) (0 2) (1 3) (4 8) (5 9) (6 10) (7 11)

Description of knot 367 Description of knot 372


DsQ7 X = b2cec2 L>372 X = b2d2ec
Gauss word 121234536546 Gauss word 121234563465
Structural vector 023010 Structural vector 021210

(0 2) (1 3) (4 7) (5 10) (6 9) (8 11) (0 2) (1 3) (4 8) (5 9) (6 11) (7 10)

Description of knot 368 Description of knot 373


D368 X = b2cfcb D373 X = b2decd
Gauss word 121234536564 Gauss word 121234563546
Structural vector 032001 Structural vector 021210

(0 2) (1 3) (4 7) (5 11) (6 9) (8 10) (0 2) (1 3) (4 8) (5 10) (6 9) (7 11)

Description of knot 369 Description of knot 374


DsQ9 X = b2ebeb ,0374 X = b2dfc2

Gauss word 121234546365 Gauss word 121234563564


Structural vector 040020 Structural vector 022101

(0 2) (1 3) (4 9) (5 7) (6 11) (8 10) (0 2) (1 3) (4 8) (5 11) (6 9) (7 10)


258
A. Dodecaphonic Knots

Description of knot 375 Description of knot 380


JD375 X = b2de2b DsS0 X = b2e~lcdb
Gauss word 121234563645 Gauss word 121234564653
Structural vector 030120 Structural vector 031110

(0 2) (1 3) (4 8) (5 10) (6 11) (7 9) (0 2) (1 3) (4 11) (5 8) (6 10) (7 9)

Description of knot 376 Description of knot 381


^376 x = b2dfdb £>381 X = b2f2b2
Gauss word 121234563654 Gauss word 121234565634
Structural vector 030201 Structural vector 040002

(0 2) (1 3) (4 8) (5 11) (6 10) (7 9) (0 2) (1 3) (4 10) (5 11) (6 8) (7 9)

Description of knot 377 Description of knot 382


D377 X = b2ecec £>382 X = b2e-^eb2
Gauss word 121234564365 Gauss word 121234565643
Structural vector 022020 Structural vector 040020

(0 2) (1 3) (4 9) (5 8) (6 11) (7 10) (0 2) (1 3) (4 11) (5 10) (6 8) (7 9)

Description of knot 378 Description of knot 383


D378 x = fcV1^ 1^333 x
= bcb2cb

Gauss word 121234564563 Gauss word 121323454656

Structural vector 023010 Structural vector 042000

(0 2) (1 3) (4 11) (5 8) (6 9) (7 10) (0 2) (1 4) (3 5) (6 8) (7 10) (9 11)

Description of knot 379 Description of knot 384


D379 X = b2fceb ^384 x
= 6c6c3
Gauss word 121234564635 Gauss word 121323456456
Structural vector 031011 Structural vector 024000

(0 2) (1 3) (4 10) (5 8) (6 11) (7 9) (0 2) (1 4) (3 5) (6 9) (7 10) (8 11)


259
A. Dodecaphonic Knots

Description of knot 385 Description of knot 390


Z/335 X = bcbcdb Z/390 X = bcdcdb
Gauss word 121323456465 Gauss word 121324534656
Structural vector 032100 Structural vector 022200

(0 2) (1 4) (3 5) (6 9) (7 11) (8 10) (0 2) (1 4) (3 7) (5 8) (6 10) (9 11)

Description of knot 386 Description of knot 391


£)386 x = bc4b Ds9l X = bcdeb2

Gauss word 121324354656 Gauss word 121324535646

Structural vector 024000 Structural vector 031110

(0 2) (1 4) (3 6) (5 8) (7 10) (9 11) (0 2) (1 4) (3 7) (5 10) (6 8) (9 11)

Description of knot 387 Description of knot 392


D387 X = bc2dc2 D392 X = bcd3c
Gauss word 121324356456 Gauss word 121324536456
Structural vector 014100 Structural vector 012300

(0 2) (1 4) (3 6) (5 9) (7 10) (8 11) (0 2) (1 4) (3 7) (5 9) (6 10) (8 11)

Description of knot 388 Description of knot 393


,0388 X = bc2d2b 1^393 X = bcd2eb

Gauss word 121324356465 Gauss word 121324536465

Structural vector 022200 Structural vector 021210

(0 2) (1 4) (3 6) (5 9) (7 11) (8 10) (0 2) (1 4) (3 7) (5 9) (6 11) (8 10)

Description of knot 389 Description of knot 394


Z?389 X = bc2ebc -D394 X = bcdec2
Gauss word 121324356546 Gauss word 121324536546
Structural vector 023010 Structural vector 013110

(0 2) (1 4) (3 6) (5 10) (7 9) (8 11) (0 2) (1 4) (3 7) (5 10) (6 9) (8 11)


260
A. Dodecaphonic Knots

Description of knot 395 Description of knot 400


x x
^395 = bcdfcb -^400 = bcedec
Gauss word 121324536564 Gauss word 121324563465
Structural vector 022101 Structural vector 012120

(0 2) (1 4) (3 7) (5 11) (6 9) (8 10) (0 2) (1 4) (3 8) (5 9) (6 11) (7 10)

Description of knot 396 Description of knot 401


Z/396 X = bcebdb -D4OI X = bce2cd
Gauss word 121324543656 Gauss word 121324563546
Structural vector 031110 Structural vector 012120

(0 2) (1 4) (3 8) (5 7) (6 10) (9 11) (0 2) (1 4) (3 8) (5 10) (6 9) (7 11)

Description of knot 397 Description of knot 402


Z/397 X = bcfbdc D402 X = 6ce/c2
Gauss word 121324546356 Gauss word 121324563564
Structural vector 022101 Structural vector 013011

(0 2) (1 4) (3 9) (5 7) (6 10) (8 11) (0 2) (1 4) (3 8) (5 11) (6 9) (7 10)

Description of knot 398 Description of knot 403


-^398 x
= bcfbeb D403 X = bce3b

Gauss word 121324546365 Gauss word 121324563645


Structural vector 031011 Structural vector 021030

(0 2) (1 4) (3 9) (5 7) (6 11) (8 10) (0 2) (1 4) (3 8) (5 10) (6 11) (7 9)

Description of knot 399 Description of knot 404


D399 X = bced3 D404 X = bcefdb
Gauss word 121324563456 Gauss word 121324563654

Structural vector 011310 Structural vector 021111

(0 2) (1 4) (3 8) (5 9) (6 10) (7 11) (0 2) (1 4) (3 8) (5 11) (6 10) (7 9)


261
A. Dodecaphonic Knots

Description of knot 405 Description of knot 410


D405 X = bcfcd2 L>410 X = bcd^cdb

Gauss word 121324564356 Gauss word 121324564653

Structural vector 012201 Structural vector 022200

(0 2) (1 4) (3 9) (5 8) (6 10) (7 11) (0 2) (1 4) (3 11) (5 8) (6 10) (7 9)

Description of knot 406 Description of knot 411


x
Z/406 = bcfcec Z/411 X = bcfebd

Gauss word 121324564365 Gauss word 121324565346

Structural vector 013011 Structural vector 021111

(0 2) (1 4) (3 9) (5 8) (6 11) (7 10) (0 2) (1 4) (3 9) (5 10) (6 8) (7 11)

Description of knot 407 Description of knot 412


Z/4O7 X = bce~1c2d £>412 X = bcfbc

Gauss word 121324564536 Gauss word 121324565364

Structural vector 013110 Structural vector 022002

(0 2) (1 4) (3 10) (5 8) (6 9) (7 11) (0 2) (1 4) (3 9) (5 11) (6 8) (7 10)

Description of knot 408 Description of knot 413


x
D40S = bcd^c3 £*413 X= bce^dbd

Gauss word 121324564563 Gauss word 121324565436

Structural vector 014100 Structural vector 021210

(0 2) (1 4) (3 11) (5 8) (6 9) (7 10) (0 2) (1 4) (3 10) (5 9) (6 8) (7 11)

Description of knot 409 Description of knot 414


-^409 x =
bce~1ceb Z/414 X = bcd~1dbc

Gauss word 121324564635 Gauss word 121324565463

Structural vector 022020 Structural vector 022200

(0 2) (1 4) (3 10) (5 8) (6 11) (7 9) (0 2) (1 4) (3 11) (5 9) (6 8) (7 10)


262
A. Dodecaphonic Knots

Description of knot 415 Description of knot 420


Z/415 X = bdcec2 £>420 X = bd2fb2
Gauss word 121342356456 Gauss word 121342535646
Structural vector 013110 Structural vector 030201

(0 2) (1 5) (3 6) (4 9) (7 10) (8 11) (0 2) (1 5) (3 7) (4 10) (6 8) (9 11)

Description of knot 416 Description of knot 421


-^416 X = bdcedb D42I X = bd2edc
Gauss word 121342356465 Gauss word 121342536456
Structural vector 021210 Structural vector 011310

(0 2) (1 5) (3 6) (4 9) (7 11) (8 10) (0 2) (1 5) (3 7) (4 9) (6 10) (8 11)

Description of knot 417 Description of knot 422


£>417 X = bdfbc2 Z>422 X = bd2e2b
Gauss word 121342456356 Gauss word 121342536465
Structural vector 022101 Structural vector 020220

(0 2) (1 5) (3 9) (4 6) (7 10) (8 11) (0 2) (1 5) (3 7) (4 9) (6 11) (8 10)

Description of knot 418 Description of knot 423


D41S X = bdfbdb JD423 X = bd2fc2

Gauss word 121342456365 Gauss word 121342536546


Structural vector 030201 Structural vector 012201

(0 2) (1 5) (3 9) (4 6) (7 11) (8 10) (0 2) (1 5) (3 7) (4 10) (6 9) (8 11)

Description of knot 419 Description of knot 424


D419 X = bd4b -D424 X = bdecdb
Gauss word 121342534656 Gauss word 121342543656

Structural vector 020400 Structural vector 021210

(0 2) (1 5) (3 7) (4 8) (6 10) (9 11) (0 2) (1 5) (3 8) (4 7) (6 10) (9 11)


263
A. Dodecaphonic Knots

Description of knot 425 Description of knot 430


Z/425 X = bdfcdc ^430 ^ = bdefcd
Gauss word 121342546356 Gauss word 121342563546
Structural vector 012201 Structural vector 011211

(0 2) (1 5) (3 9) (4 7) (6 10) (8 11) (0 2) (1 5) (3 8) (4 10) (6 9) (7 11)

Description of knot 426 Description of knot 431


Z/426 X = bdfceb D43I X = bdee~lc2
Gauss word 121342546365 Gauss word 121342563564
Structural vector 021111 Structural vector 012120

(0 2) (1 5) (3 9) (4 7) (6 11) (8 10) (0 2) (1 5) (3 8) (4 11) (6 9) (7 10)

Description of knot 427 Description of knot 432


D427 X = bde~xcz ^432 X = bdefeb
Gauss word 121342546536 Gauss word 121342563645
Structural vector 013110 Structural vector 020121

(0 2) (1 5) (3 10) (4 7) (6 9) (8 11) (0 2) (1 5) (3 8) (4 10) (6 11) (7 9)

Description of knot 428 Description of knot 433


D428 X = bde2d2 D433 X = bdee^db
Gauss word 121342563456 Gauss word 121342563654
Structural vector 010320 Structural vector 020220

(0 2) (1 5) (3 8) (4 9) (6 10) (7 11) (0 2) (1 5) (3 8) (4 11) (6 10) (7 9)

Description of knot 429 Description of knot 434


Z?429 X = bde3c D434 X = bdfd3
Gauss word 121342563465 Gauss word 121342564356
Structural vector 011130 Structural vector 010401

(0 2) (1 5) (3 8) (4 9) (6 11) (7 10) (0 2) (1 5) (3 9) (4 8) (6 10) (7 11)


264
A. Dodecaphonic Knots

Description of knot 435 Description of knot 440


Z/435 X = bdfdec D440 X = bdfbd
Gauss word 121342564365 Gauss word 121342565346
Structural vector 011211 Structural vector 020202

(0 2) (1 5) (3 9) (4 8) (6 11) (7 10) (0 2) (1 5) (3 9) (4 10) (6 8) (7 11)

Description of knot 436 Description of knot 441


^436 x =
bde~1dcd D441 X = bde~1ebd

Gauss word 121342564536 Gauss word 121342565436


Structural vector 011310 Structural vector 020220

(0 2) (1 5) (3 10) (4 8) (6 9) (7 11) (0 2) (1 5) (3 10) (4 9) (6 8) (7 11)

Description of knot 437 Description of knot 442


D437 x = bddrxd<? -D442 X = bebec2
Gauss word 121342564563 Gauss word 121343256456
Structural vector 012300 Structural vector 022020

(0 2) (1 5) (3 11) (4 8) (6 9) (7 10) (0 2) (1 6) (3 5) (4 9) (7 10) (8 11)

Description of knot 438 Description of knot 443


^438 X = bde~1deb Z?443 X = bebedb

Gauss word 121342564635 Gauss word 121343256465


Structural vector 020220 Structural vector 030120

(0 2) (1 5) (3 10) (4 8) (6 11) (7 9) (0 2) (1 6) (3 5) (4 9) (7 11) (8 10)

Description of knot 439 Description of knot 444


D439 x = bdd~H2b £>444 X = bfbfb2
Gauss word 121342564653 Gauss word 121343525646
Structural vector 020400 Structural vector 040002

(0 2) (1 5) (3 11) (4 8) (6 10) (7 9) (0 2) (1 7) (3 5) (4 10) (6 8) (9 11)


265
A. Dodecaphonic Knots

Description of knot 445 Description of knot 450


Z/445 X = bfbedc -D450 X = bede2c
Gauss word 121343526456 Gauss word 121345236456
Structural vector 021111 Structural vector 011130

(0 2) (1 7) (3 5) (4 9) (6 10) (8 11) (0 2) (1 6) (3 7) (4 9) (5 10) (8 11)

Description of knot 446 Description of knot 451


i?446 X = bfbe2b D45I X = bedefb

Gauss word 121343526465 Gauss word 121345236465


Structural vector 030021 Structural vector 020121

(0 2) (1 7) (3 5) (4 9) (6 11) (8 10) (0 2) (1 6) (3 7) (4 9) (5 11) (8 10)

Description of knot 447 Description of knot 452


DU7 X = bfbfc2 ^452 x
= bedfdc
Gauss word 121343526546 Gauss word 121345236546
Structural vector 022002 Structural vector 011211

(0 2) (1 7) (3 5) (4 10) (6 9) (8 11) (0 2) (1 6) (3 7) (4 10) (5 9) (8 11)

Description of knot 448 Description of knot 453


£>448 X = be^bed2 ^453 x
= befcec
Gauss word 121343562456 Gauss word 121345246356
Structural vector 020220 Structural vector 012021

(0 2) (1 8) (3 5) (4 9) (6 10) (7 11) (0 2) (1 6) (3 9) (4 7) (5 10) (8 11)

Description of knot 449 Description of knot 454


D449 X = be~1be2c D454 X = befcfb

Gauss word 121343562465 Gauss word 121345246365

Structural vector 021030 Structural vector 021012

(0 2) (1 8) (3 5) (4 9) (6 11) (7 10) (0 2) (1 6) (3 9) (4 7) (5 11) (8 10)


266
A. Dodecaphonic Knots

Description of knot 455 Description of knot 460


Z/455 X = bee~lcdc £>460 X = be2fd2
Gauss word 121345246536 Gauss word 121345263546
Structural vector 012120 Structural vector 010221

(0 2) (1 6) (3 10) (4 7) (5 9) (8 11) (0 2) (1 6) (3 8) (4 10) (5 9) (7 11)

Description of knot 456 Description of knot 461


x
At56 X = befbc Al61 = be^dc
Gauss word 121345256346 Gauss word 121345263564
Structural vector 021012 Structural vector 011130

(0 2) (1 6) (3 9) (4 10) (5 7) (8 11) (0 2) (1 6) (3 8) (4 11) (5 9) (7 10)

Description of knot 457 Description of knot 462


-D457 X = bee~1ebc D462 X = be2f2b
Gauss word 121345256436 Gauss word 121345263645
Structural vector 021030 Structural vector 020022

(0 2) (1 6) (3 10) (4 9) (5 7) (8 11) (0 2) (1 6) (3 8) (4 10) (5 11) (7 9)

Description of knot 458 Description of knot 463


jD458 X = be4d £>463 x
= be2e~leb
Gauss word 121345263456 Gauss word 121345263654
Structural vector 010140 Structural vector 020040

(0 2) (1 6) (3 8) (4 9) (5 10) (7 11) (0 2) (1 6) (3 8) (4 11) (5 10) (7 9)

Description of knot 459 Description of knot 464


D459 X = be3fc Z/464 X = befded

Gauss word 121345263465 Gauss word 121345264356

Structural vector 011031 Structural vector 010221

(0 2) (1 6) (3 8) (4 9) (5 11) (7 10) (0 2) (1 6) (3 9) (4 8) (5 10) (7 11)


267
A. Dodecaphonic Knots

Description of knot 465 Description of knot 470


£>465 X = befdfc D4J0 X = bee~1ecd

Gauss word 121345264365 Gauss word 121345265436


Structural vector 011112 Structural vector 011130

(0 2) (1 6) (3 9) (4 8) (5 11) (7 10) (0 2) (1 6) (3 10) (4 9) (5 8) (7 11)

Description of knot 466 Description of knot 471


£>466 x
= bee~ldz Dtfi X = bed~lec2

Gauss word 121345264536 Gauss word 121345265463


Structural vector 010320 Structural vector 012120

(0 2) (1 6) (3 10) (4 8) (5 9) (7 11) (0 2) (1 6) (3 11) (4 9) (5 8) (7 10)

Description of knot 467 Description of knot 472


£)467 X = bed~1d2c D472 X = bfce2c
Gauss word 121345264563 Gauss word 121345326456
Structural vector 011310 Structural vector 012021

(0 2) (1 6) (3 11) (4 8) (5 9) (7 10) (0 2) (1 7) (3 6) (4 9) (5 10) (8 11)

Description of knot 468 Description of knot 473


At68 X = be fed £>473 X = bfcfdc
Gauss word 121345265346 Gauss word 121345326546
Structural vector 011112 Structural vector 012102

(0 2) (1 6) (3 9) (4 10) (5 8) (7 11) (0 2) (1 7) (3 6) (4 10) (5 9) (8 11)

Description of knot 469 Description of knot 474


At69 X= befe-'c2 D474 X= bfbfb

Gauss word 121345265364 Gauss word 121345426365


Structural vector 012021 Structural vector 030003

(0 2) (1 6) (3 9) (4 11) (5 8) (7 10) (0 2) (1 7) (3 9) (4 6) (5 11) (8 10)


268
A. Dodecaphonic Knots

Description of knot 475 Description of knot 480


£>475 X = bfe4 ^480 X = bfee~1ec
Gauss word 121345623456 Gauss word 121345623654
Structural vector 010041 Structural vector 011031

(0 2) (1 7) (3 8) (4 9) (5 10) (6 11) (0 2) (1 7) (3 8) (4 11) (5 10) (6 9)

Description of knot 476 Description of knot 481


At76 X = bfe2fd D4SI X = bfdfd
Gauss word 121345623465 Gauss word 121345624365
Structural vector 010122 Structural vector 010203

(0 2) (1 7) (3 8) (4 9) (5 11) (6 10) (0 2) (1 7) (3 9) (4 8) (5 11) (6 10)

Description of knot 477 Description of knot 482


At77 X = bfefde £>482 X = bfd-'d3
Gauss word 121345623546 Gauss word 121345624563
Structural vector 010122 Structural vector 010401

(0 2) (1 7) (3 8) (4 10) (5 9) (6 11) (0 2) (1 7) (3 11) (4 8) (5 9) (6 10)

Description of knot 478 Description of knot 483


£>478 X = bfee~ld2 At83 X= bfe-'dfc
Gauss word 121345623564 Gauss word 121345624635
Structural vector 010221 Structural vector 011112

(0 2) (1 7) (3 8) (4 11) (5 9) (6 10) (0 2) (1 7) (3 10) (4 8) (5 11) (6 9)

Description of knot 479 Description of knot 484


x
D479 X= bfefc ^484 = bfd^dec
Gauss word 121345623645 Gauss word 121345624653
Structural vector 011013 Structural vector 011211

(0 2) (1 7) (3 8) (4 10) (5 11) (6 9) (0 2) (1 7) (3 11) (4 8) (5 10) (6 9)


269
A. Dodecaphonic Knots

Description of knot 485 Description of knot 490


£>485 X = 6/e-V £>490 X= bfd-'fdb
Gauss word 121345625634 Gauss word 121345626543
Structural vector 012021 Structural vector 020202

(0 2) (1 7) (3 10) (4 11) (5 8) (6 9) (0 2) (1 7) (3 11) (4 10) (5 9) (6 8)

Description of knot 486 Description of knot 491


£>486 X= bfd-'fc2 Dm x = c6
Gauss word 121345625643 Gauss word 123123456456
Structural vector 012102 Structural vector 006000

(0 2) (1 7) (3 11) (4 10) (5 8) (6 9) (0 3) (1 4) (2 5) (6 9) (7 10) (8 11)

Description of knot 487 Description of knot 492


£>487 X = bfb D492 X = c2d2c2

Gauss word 121345626345 Gauss word 123124356456


Structural vector 020004 Structural vector 004200

(0 2) (1 7) (3 9) (4 10) (5 11) (6 8) (0 3) (1 4) (2 6) (5 9) (7 10) (8 11)

Description of knot 488 Description of knot 493


At88 X = bfe-'eb £>493 X = c2ed2c
Gauss word 121345626354 Gauss word 123124536456
Structural vector 020022 Structural vector 003210

(0 2) (1 7) (3 9) (4 11) (5 10) (6 8) (0 3) (1 4) (2 7) (5 9) (6 10) (8 11)

Description of knot 489 Description of knot 494


At89 X = bfd~1e2b D494 X = c2e2c2

Gauss word 121345626453 Gauss word 123124536546


Structural vector 020121 Structural vector 004020

(0 2) (1 7) (3 11) (4 9) (5 10) (6 8) (0 3) (1 4) (2 7) (5 10) (6 9) (8 11)


270
A. Dodecaphonic Knots

Description of knot 495 Description of knot 500


£>495 X = c2fd3 1^500 X = cde~1cdc
Gauss word 123124563456 Gauss word 123142546356
Structural vector 002301 Structural vector 003210

(0 3) (1 4) (2 8) (5 9) (6 10) (7 11) (0 3) (1 5) (2 9) (4 7) (6 10) (8 11)

Description of knot 496 Description of knot 501


-D496 X = c2fdec £>501 x
= cdd^c3
Gauss word 123124563465 Gauss word 123142546536
Structural vector 003111 Structural vector 004200

(0 3) (1 4) (2 8) (5 9) (6 11) (7 10) (0 3) (1 5) (2 10) (4 7) (6 9) (8 11)

Description of knot 497 Description of knot 502


Dm X = c2f2c2 D502 X = cdfed2
Gauss word 123124563564 Gauss word 123142563456
Structural vector 004002 Structural vector 001311

(0 3) (1 4) (2 8) (5 11) (6 9) (7 10) (0 3) (1 5) (2 8) (4 9) (6 10) (7 11)

Description of knot 498 Description of knot 503


-D498 X = cde2dc L>503 X = cdfe2c
Gauss word 123142536456 Gauss word 123142563465
Structural vector 002220 Structural vector 002121

(0 3) (1 5) (2 7) (4 9) (6 10) (8 11) (0 3) (1 5) (2 8) (4 9) (6 11) (7 10)

Description of knot 499 Description of knot 504


x
Z/499 = cdefc2 D504 X = cdfcd
Gauss word 123142536546 Gauss word 123142563546
Structural vector 003111 Structural vector 002202

(0 3) (1 5) (2 7) (4 10) (6 9) (8 11) (0 3) (1 5) (2 8) (4 10) (6 9) (7 11)


271
A. Dodecaphonic Knots

Description of knot 505 Description of knot 510


£>505 x
= cde^d3 -^510 X = ce~1ced2

Gauss word 123142564356 Gauss word 123143562456

Structural vector 001410 Structural vector 002220

(0 3) (1 5) (2 9) (4 8) (6 10) (7 11) (0 3) (1 8) (2 5) (4 9) (6 10) (7 11)

Description of knot 506 Description of knot 511


-D5O6 x
= cde~1dec Z^5H X = ce~1ce2c

Gauss word 123142564365 Gauss word 123143562465

Structural vector 002220 Structural vector 003030

(0 3) (1 5) (2 9) (4 8) (6 11) (7 10) (0 3) (1 8) (2 5) (4 9) (6 11) (7 10)

Description of knot 507 Description of knot 512


^507 X = cdd~xdcd £>512 X = cd^cd3

Gauss word 123142564536 Gauss word 123143564256

Structural vector 002400 Structural vector 002400

(0 3) (1 5) (2 10) (4 8) (6 9) (7 11) (0 3) (1 9) (2 5) (4 8) (6 10) (7 11)

Description of knot 508 Description of knot 513


£>508 X= cfcfc2 -^513 X = cd~1cdec

Gauss word 123143526546 Gauss word 123143564265

Structural vector 004002 Structural vector 003210

(0 3) (1 7) (2 5) (4 10) (6 9) (8 11) (0 3) (1 9) (2 5) (4 8) (6 11) (7 10)

Description of knot 509 Description of knot 514


D509 X = cc^c4 1^514 X = ce4c

Gauss word 123143546526 Gauss word 123145236456

Structural vector 006000 Structural vector 002040

(0 3) (1 10) (2 5) (4 7) (6 9) (8 11) (0 3) (1 6) (2 7) (4 9) (5 10) (8 11)


272
A. Dodecaphonic Knots

Description of knot 515 Description of knot 520


£>515 X = ce2fdc •^520 x
= cee~1ded

Gauss word 123145236546 Gauss word 123145264356

Structural vector 002121 Structural vector 001230

(0 3) (1 6) (2 7) (4 10) (5 9) (8 11) (0 3) (1 6) (2 9) (4 8) (5 10) (7 11)

Description of knot 516 Description of knot 521


^516 x
= cee~1cec £>521 X = cee~ldfc

Gauss word 123145246356 Gauss word 123145264365

Structural vector 003030 Structural vector 002121

(0 3) (1 6) (2 9) (4 7) (5 10) (8 11) (0 3) (1 6) (2 9) (4 8) (5 11) (7 10)

Description of knot 517 Description of knot 522


L>517 X = cefe2d £>522 X = cfdfdc

Gauss word 123145263456 Gauss word 123145326546

Structural vector 001131 Structural vector 002202

(0 3) (1 6) (2 8) (4 9) (5 10) (7 11) (0 3) (1 7) (2 6) (4 10) (5 9) (8 11)

Description of knot 518 Description of knot 523


^518 x
= cefefc ^523 X = ce~1de2d

Gauss word 123145263465 Gauss word 123145362456

Structural vector 002022 Structural vector 001230

(0 3) (1 6) (2 8) (4 9) (5 11) (7 10) (0 3) (1 8) (2 6) (4 9) (5 10) (7 11)

Description of knot 519 Description of knot 524


L> 5 19 X = cefd2 ^524 X = ce^dfd2

Gauss word 123145263546 Gauss word 123145362546

Structural vector 001212 Structural vector 001311

(0 3) (1 6) (2 8) (4 10) (5 9) (7 11) (0 3) (1 8) (2 6) (4 10) (5 9) (7 11)


273
A. Dodecaphonic Knots

Description of knot 525 Description of knot 530


#525 x
= cdTld2ed #530 X = cf4c

Gauss word 123145364256 Gauss word 123145623645

Structural vector 001410 Structural vector 002004

(0 3) (1 9) (2 6) (4 8) (5 10) (7 11) (0 3) (1 7) (2 8) (4 10) (5 11) (6 9)

Description of knot 526 Description of knot 531


#526 X = cd-'tffc #531 X = cfe-'ec

Gauss word 123145364265 Gauss word 123145623654

Structural vector 002301 Structural vector 002022

(0 3) (1 9) (2 6) (4 8) (5 11) (7 10) (0 3) (1 7) (2 8) (4 11) (5 10) (6 9)

Description of knot 527 Description of knot 532


#527 X = c/V #532 X = ce-V

Gauss word 123145623456 Gauss word 123145632456

Structural vector 001032 Structural vector 001050

(0 3) (1 7) (2 8) (4 9) (5 10) (6 11) (0 3) (1 8) (2 7) (4 9) (5 10) (6 11)

Description of knot 528 Description of knot 533


#528 X = cfefd #533 X= ce-1e2fd

Gauss word 123145623465 Gauss word 123145632465

Structural vector 001113 Structural vector 001131

(0 3) (1 7) (2 8) (4 9) (5 11) (6 10) (0 3) (1 8) (2 7) (4 9) (5 11) (6 10)

Description of knot 529 Description of knot 534


# 5 2 9 X = cf2e~1d2 jD534 X = ce^ee^d2

Gauss word 123145623564 Gauss word 123145632564

Structural vector 001212 Structural vector 001230

(0 3) (1 7) (2 8) (4 11) (5 9) (6 10) (0 3) (1 8) (2 7) (4 11) (5 9) (6 10)


274
A. Dodecaphonic Knots

Description of knot 535 Description of knot 540


£>535 X= ce-'efc D540 X= defend2
Gauss word 123145632645 Gauss word 123415263546
Structural vector 002022 Structural vector 000321

(0 3) (1 8) (2 7) (4 10) (5 11) (6 9) (0 4) (1 6) (2 8) (3 10) (5 9) (7 11)

Description of knot 536 Description of knot 541


Z/535 X = ce~1ee~1ec •^541 x
= dee~1e2d
Gauss word 123145632654 Gauss word 123415264356
Structural vector 002040 Structural vector 000240

(0 3) (1 8) (2 7) (4 11) (5 10) (6 9) (0 4) (1 6) (2 9) (3 8) (5 10) (7 11)

Description of knot 537 Description of knot 542


A>37 X = d2f2d2 £>542 X = ded^ed2
Gauss word 123412563456 Gauss word 123415264536
Structural vector 000402 Structural vector 000420

(0 4) (1 5) (2 8) (3 9) (6 10) (7 11) (0 4) (1 6) (2 10) (3 8) (5 9) (7 11)

Description of knot 538 Description of knot 543


D538 X = d2e~led2 A>43 X = dfe2
Gauss word 123412564356 Gauss word 123415623456
Structural vector 000420 Structural vector 000123

(0 4) (1 5) (2 9) (3 8) (6 10) (7 11) (0 4) (1 7) (2 8) (3 9) (5 10) (6 11)

Description of knot 539 Description of knot 544


£>539 X = defed £>544 X = dfd
Gauss word 123415263456 Gauss word 123415623465
Structural vector 000222 Structural vector 000204

(0 4) (1 6) (2 8) (3 9) (5 10) (7 11) (0 4) (1 7) (2 8) (3 9) (5 11) (6 10)


275
A. Dodecaphonic Knots

Description of knot 545 Description of knot 550


#545 X = dfe-'e3 £>550 X = e/ 4 e

Gauss word 123415624356 Gauss word 123451623456

Structural vector 000141 Structural vector 000024

(0 4) (1 7) (2 9) (3 8) (5 10) (6 11) (0 5) (1 7) (2 8) (3 9) (4 10) (6 11)

Description of knot 546 Description of knot 551


£>546 X = dfe-'efd A>51 X = e/2e-1e2
Gauss word 123415624365 Gauss word 123451623546

Structural vector 000222 Structural vector 000042

(0 4) (1 7) (2 9) (3 8) (5 11) (6 10) (0 5) (1 7) (2 8) (3 10) (4 9) (6 11)

Description of knot 547 Description of knot 552


D 5 4 7 X = dd~xeA #552 X = efe-'efe

Gauss word 123415634256 Gauss word 123451624356

Structural vector 000240 Structural vector 000042

(0 4) (1 9) (2 7) (3 8) (5 10) (6 11) (0 5) (1 7) (2 9) (3 8) (4 10) (6 11)

Description of knot 548 Description of knot 553


£>548 x
= dd~1e2fd #553 X = ee-'ee-'e2

Gauss word 123415634265 Gauss word 123451632546

Structural vector 000321 Structural vector 000060

(0 4) (1 9) (2 7) (3 8) (5 11) (6 10) (0 5) (1 8) (2 7) (3 10) (4 9) (6 11)

Description of knot 549 Description of knot 554


£>549 X = dd^fdfd A>54 X= f
Gauss word 123415643265 Gauss word 123456123456

Structural vector 000402 Structural vector 000006

(0 4) (1 9) (2 8) (3 7) (5 11) (6 10) (0 6) (1 7) (2 8) (3 9) (4 10) (5 11)

276
Appendix B
All-Interval Series

The tables give a description of all-interval series classified by their chord diagram. The
series are denoted by their structural vector, whose components are the numbers of
semitones between two adjacent pitches.

Knot 14 1 n. 20
n. 21
3 6 7 10 81 9 5 4 2 11
3 6 8 11 7 2 9 4 5 1 10
1 n. 1 1 4 2 7 9 11 10 8 5 6 3 n. 22 3 6 10 1 5 4 9 2 7 11 8
n. 2 1 6 3 4 2 5 11 9 10 8 7 n. 23 3 6 11 2 4 5 9 1 8 10 7
n. 3 1 6 7 8 10 3 5 11 2 4 9 n. 24 3 8 10 1 7 9 2 4 5 6 11
n. 4 1 6 7 8 10 9 11 5 2 4 3 n. 25 38 10 1 7 9 2 4 11 65
n. 5 1 6 9 4 2 11 53 10 87 n. 26 415 10 3 8 7 11 269
n. 6 18 10 352 11 7964 n. 27 4 6 1 8 10 352 11 79
n. 7 18 10 35 11 2 4 9 6 7 n. 28 4 6 3 1 5 2 11 9 10 87
n. 8 18 10 9 11 5 2 4 3 6 7 n. 29 4 6 7 8 10 9 11 2 5 1 3
n. 9 1 10 873 11 4 2 5 6 9 n. 30 4 6 9 7 11 253 10 81
n. 10 2 5 1 8 9 10 11 7463 n. 31 47 11 10 9 8 1 5 2 6 3
n. 11 2 11 783 10 5 1 4 6 9 n. 32 5 24 11378 10 169
n. 12 3 1 5 2 11 9 10 8 7 6 4 n. 33 5 4 2 3 1 7 10 896 11
n. 13 3 4 2 5 11 9 10 8 1 6 7 n. 34 5 4 2 3 1 10 7 11 968
n. 14 3 4 2 5 11 9 10 8761 n. 35 5 4 2 9 7 1 10 836 11
n. 15 351 10 7 9 2 4 11 68 n. 36 5 6 3 8 10 1 7 9 2 4 11
n. 16 3 6 1 4 2 7 9 11 10 85 n. 37 5 6 9 8 10 7 1 3 2 4 11
n. 17 3 6 2 5 1 8 9 10 11 74 n. 38 56 11 4 2 3 1 7 10 89
n. 18 3 6 4 7 11 10 9 8 1 5 2 n. 39 56 11 4 2 9 7 1 10 83
n. 19 3 6 5 8 10 11 9 7 2 4 1 | n. 40 58 10 11 9 7 2 4 1 6 3 |
B. All-Interval Series

Knot 19
1 n. 1 1 2 4 10 9 7 11 3 5 6 8 1 n. 33 3 6 2 11 7 8 1 5 9 4 10
n. 2 1 3 7 11 9 2 8 10 5 6 4 1 n. 34 3 6 4 1 5 10 11 7 9 2 8
n. 3 1 3 7 11 9 8 2 4 5 6 10 n. 35 3 6 4 10 9 11 7 2 1 5 8
n. 4 1 8 10 4 9 7 11 3 5 6 2 n. 36 3 6 5 2 4 11 10 8 9 1 7
n. 5 2 6 1 8 10 4 9 7 11 3 5 n. 37 3 6 5 11 9 10 8 1 2 4 7
n. 6 2 6 5 3 11 7 9 4 10 8 1 n. 38 3 6 7 1 9 8 10 11 4 2 5
n. 7 2 6 7 8 10 4 3 1 5 9 11 n. 39 3 6 7 4 2 1 8 10 9 11 5
n. 8 2 6 11 9 5 1 3 4 10 8 7 n. 40 3 6 8 2 9 7 11 10 5 1 4
n. 9 4 6 1 3 7 11 9 2 8 10 5 n. 41 3 6 8 5 1 2 7 11 9 10 4
n. 10 4 6 5 10 8 2 9 11 7 3 1 n. 42 3 6 10 4 9 5 1 8 7 11 2
n. 11 4 6 7 9 1 5 3 2 8 10 11 n. 43 3 6 10 7 11 4 5 1 9 8 2
n. 12 4 6 11 10 8 2 3 5 1 9 7 n. 44 3 6 11 5 9 4 2 7 8 10 1
n. 13 5 3 11 7 9 4 10 8 1 6 2 n. 45 3 6 1 1 8 10 5 4 2 9 7 1
n. 14 5 3 11 7 9 10 4 2 1 6 8 n. 46 3 7 1 10 8 9 2 4 5 6 11
n. 15 5 4 2 8 9 11 7 3 1 6 10 n. 47 3 7 1 10 8 9 2 4 11 6 5
n. 16 5 10 8 2 9 11 7 3 1 6 4 | n. 48 3 7 11 8 2 4 5 9 1 6 10
n. 49 3 10 1 7 11 9 2 4 5 6 8
n. 50 3 10 8 11 2 4 5 9 1 6 7
n. 51 3 10 8 11 2 4 5 9 7 6 1
n. 52 3 11 5 2 4 9 10 8 1 6 7
n. 53 3 11 5 2 4 9 10 8 7 6 1
n. 54 3 11 7 4 10 8 1 9 5 6 2
1 n. 1 1 3 11 4 2 5 8 10 9 6 7
n. 55 4 1 5 10 11 7 9 2 8 6 3
n. 2 1 4 2 7 8 10 3 5 11 6 9
n. 56 4 6 1 8 10 9 7 11 5 2 3
n. 3 1 6 3 10 8 11 2 4 5 9 7
n. 57 4 6 3 2 5 11 7 9 10 8 1
n. 4 1 6 3 11 5 2 4 9 10 8 7
n. 58 4 6 7 8 10 3 1 5 11 2 9
n. 5 1 6 7 3 11 4 2 5 8 10 9
n. 59 4 6 9 2 11 5 1 3 10 8 7
n. 6 1 6 7 8 10 3 4 2 11 5 9
n. 60 4 7 11 10 5 1 3 2 8 6 9
n. 7 1 6 7 8 10 9 4 2 5 11 3
n. 61 4 10 3 5 1 2 7 11 8 6 9
n. 8 1 6 7 9 5 4 2 11 8 10 3
n. 62 4 10 9 11 7 2 1 5 8 6 3
n. 9 1 6 9 5 11 2 4 3 10 8 7
n. 63 5 2 4 11 10 8 9 1 7 6 3
n. 10 1 6 9 10 8 5 2 4 11 3 7
n. 64 5 3 7 8 10 1 4 2 9 6 11
n. 11 1 7 3 8 10 5 4 2 11 6 9
n. 65 5 4 2 3 8 10 7 1 9 6 11
n. 12 1 7 9 2 4 5 10 8 11 6 3
n. 66 5 4 2 9 8 10 1 7 3 6 11
n. 13 1 8 10 3 4 2 1 1 5 9 6 7
n. 67 5 4 2 9 11 7 1 10 3 6 8
n. 14 1 8 10 9 4 2 5 1 1 3 6 7
n. 68 5 6 3 2 4 7 10 8 1 9 11
n. 15 1 8 10 9 7 11 5 2 3 6 4
n. 69 5 6 3 7 1 10 8 9 2 4 11
n. 16 1 9 5 4 2 8 11 7 3 6 10
n. 70 5 6 9 1 7 10 8 3 2 4 11
n. 17 1 9 5 4 2 11 8 10 3 6 7
n. 71 5 6 9 2 4 1 10 8 7 3 11
n. 18 1 10 8 7 2 4 9 5 11 6 3
n. 72 5 6 11 3 7 8 10 1 4 2 9
n. 19 2 5 1 8 7 11-3 4 10 6 9
n. 73 5 6 11 4 2 3 8 10 7 1 9
n. 20 2 6 3 11 7 4 10 8 1 9 5
n. 74 5 6 11 4 2 9 8 10 1 7 3
n. 21 2 6 5 9 1 8 10 4 7 11 3
n. 75 5 6 11 9 1 8 10 7 4 2 3
n. 22 2 6 9 5 1 4 10 8 7 3 11
n. 76 5 8 10 11 4 2 3 1 7 6 9
n. 23 2 6 11 3 7 8 10 4 1 5 9
n. 77 5 9 1 8 10 4 7 11 3 6 2
n. 24 2 8 3 7 11 4 5 1 10 6 9
n. 78 5 9 1 8 10 7 4 2 3 6 11
n. 25 2 8 9 1 5 4 11 7 10 6 3
n. 79 5 11 3 4 2 1 8 10 7 6 9
n. 26 2 11 7 8 1 5 9 4 10 6 3
n. 80 5 11 9 10 8 1 2 4 7 6 3 |
n. 27 3 2 4 7 10 8 1 9 5 6 11
n. 28 3 2 4 7 10 8 1 9 11 6 5
n. 29 3 2 5 11 7 9 10 8 1 6 4
n. 30 3 6 1 7 9 2 4 5 10 8 11
n. 31 3 6 1 10 8 7 2 4 9 5 11
n. 32 3 6 2 8 9 1 5 4 11 7 10 |

278
B. All-Interval Series

Knot 33 Knot 35
1 n. 1 1 6 4 3 8 10 7 2 11 5 9 1 n. 1 1 4 5 9 2 7 11 8 3 6 10
n. 2 1 6 4 11 5 2 3 8 10 79 n. 2 1 8 7 11 2 5 9 4 3 6 10
n. 3 1 6 9 5 2 4 11 8 7 3 10 n. 3 2 6 3 4 7 11 10 9 1 8 5
n. 4 1 6 9 5 11 2 7 10 8 3 4 n. 4 2 6 3 8 9 1 10 7 11 4 5
n. 5 1 6 9 7 3 8 5 2 4 11 10 n. 5 2 6 5 4 11 7 10 1 9 8 3
n. 6 1 6 9 7 10 8 3 2 5 11 4 n. 6 2 6 5 8 1 9 10 1 1 7 4 3
n. 7 16 10 3 7 8 11 4 2 5 9 n. 7 2 6 9 4 1 5 10 3 7 8 11
n. 8 16 10 11 4 2 5 8 3 7 9 n. 8 2 6 9 8 3 7 10 1 5 4 11
n. 9 2 1 8 10 7 4 9 5 3 6 11 n. 9 26 11 4 5 1 10 7 3 8 9
n. 10 23 11 4 7 8 10 1 9 6 5 n. 10 26 11 8 7 3 10 5 1 4 9
n. 11 2 7 8 10 1 4 3 11 9 6 5 n. 11 3 4 7 11 10 9 1 8 5 6 2
n. 12 2 9 5 4 1 8 10 7 3 6 11 n. 12 3 4 9 5 2 11 7 8 1 6 10
n. 13 3 1 9 8 11 2 4 5 10 6 7 n. 13 3 8 9 1 10 7 11 4 5 6 2
n. 14 31 10 8 9 2 11 5 4 6 7 n. 14 3 8 11 7 2 9 5 4 1 6 10
n. 15 3 5 2 4 9 10 7 1 8 6 11 n. 15 5 4 11 7 10 1 9 8 3 6 2
n. 16 3 5 9 4 7 10 8 1 2 6 11 n. 16 5 8 1 9 10 11 7 4 3 6 2 |
n. 17 3 7 1 10 5 2 4 9 8 6 11
n. 18 3 7 10 8 1 4 5 9 2 6 11
n. 19 3 11 2 4 5 8 1 9 10 6 7
n. 20 3 11 5 2 1 10 8 9 4 6 7
n. 21 4 3 8 10 72 11 5 9 6 1
n. 22 4 5 11 2 9 8 10 1 3 6 7
n. 23 4 9 8 10 1 2 5 11 3 6 7
n. 24 4 11 5 2 3 8 10 7 9 6 1 Knot 38
n. 25 5 6 2 3 11 4 7 8 10 19
1 n. 1 1 3 7 11 5 4 2 8 9 6 10
n. 26 5 6 2 7 8 10 1 4 3 11 9
n. 2 1 3 10 8 5 4 2 11 9 6 7
n. 27 5 6 8 3 4 2 11 10 7 1 9
n. 3 1 6 3 5 2 4 11 8 10 9 7
n. 28 5 6 8 7 1 1 0 3 4 2 119
n. 4 1 6 7 3 10 8 5 4 2 11 9
n. 29 5 6 9 1 7 10 11 2 4 3 8
n. 5 1 6 7 9 10 8 11 4 2 5 3
n. 30 5 6 9 1 10 8 7 4 11 3 2
n. 6 1 6 9 11 2 4 5 8 10 3 7
n. 31 5 6 9 11 2 4 3 10 1 7 8
n. 7 1 7 2 4 9 8 10 11 5 6 3
n. 32 5 6 9 11 3 4 1 10 8 7 2 |
n. 8 1 7 8 10 3 2 4 1 1 5 6 9
n. 9 19 10 8 11 4 2 5 3 6 7
n. 10 2 6 3 4 10 8 7 1 5 9 11
n. 11 2 6 5 3 11 7 1 8 10 4 9
n. 12 2 6 9 4 10 8 1 7 11 35
n. 13 26 11 9 5 1 7 8 10 4 3
Knot 34 n. 14 2 8 1 5 9 7 11 10 4 6 3
n. 15 2 8 7 11 3 1 5 10 4 6 9
n. 1 1 8 7 3 4 11 5 2 9 6 10 n. 16 31 10 8 7 4 2 9 5 6 11
n. 2 2 6 3 10 7 1 8 9 5 4 11 n. 17 31 10 8 7 4 2 9 11 6 5
n. 3 2 6 5 4 11 3 8 7 1 10 9 n. 18 3 4 10 8 7 1 5 9 11 6 2
n. 4 2 6 9 10 1 7 8 3 11 4 5 n. 19 3 5 2 4 11 8 10 9 1 6 7
n. 5 2 6 11 4 5 9 8 1 7 10 3 n. 20 3 5 2 4 11 8 10 9 7 6 1
n. 6 3 2 11 5 4 9 1 8 7 6 10 n. 21 3 6 1 7 2 4 9 8 10 11 5
n. 7 3 10 7 1 8 9 5 4 11 6 2 n. 22 3 6 2 8 1 5 9 7 11 10 4
n. 8 54 11387110 9 6 2 n. 23 3 6 4 10 11 7 9 5 1 8 2
n. 24 3 6 5 11 10 8 9 4 2 7 1
n. 25 3 6 7 1 8 10 9 2 4 5 11
n. 26 3 6 8 2 7 11 9 1 5 4 10
n. 27 36 10 4 5 1 9 11 7 2 8
n. 28 36 11 5 4 2 9 10 8 1 7
n. 29 3 8 2 4 11 5 1 9 7 6 10 |

279
B. All-Interval Series

1 n. 30
n. 31
4 10 5 1 3 11 7 8 2 6 9
4 10 11 7 9 5 1 8 2 6 3
Knot 49
n. 32 5 3 2 4 1 8 10 7 9 6 11 1 n. 1 1 4 9 7 10 8 5 3 11 6 2
n. 33 5 3 11 7 1 8 10 4 9 6 2 n. 2 1 9 7 4 2 5 3 8 11 6 10
n. 34 5 6 3 1 10 8 7 4 2 9 11 n. 3 1 9 7 10 8 5 3 2 11 6 4
n. 35 5 6 9 7 10 8 1 4 2 3 11 n. 4 1 10 9 7 4 2 5 3 11 6 8
n. 36 5 6 11 3 2 4 1 8 10 7 9 n. 5 2 6 1 4 9 7 10 8 5 3 11
n.37 5 6 11 9 2 4 7 8 10 1 3 n. 6 2 6 5 9 11 8 10 1 3 4 7
n. 38 5 9 2 4 7 8 10 1 3 6 11 n. 7 2 6 7 4 3 1 10 8 11 9 5
n. 39 5 11 4 2 3 10 8 7 1 6 9 n. 8 26 11 3 5 8 10 7 9 4 1
n. 40 5 11 10 8 9 4 2 7 1 6 3 | n. 9 4 6 1 9 7 10 8 5 3 2 11
n. 10 4 6 5 2 9 11 8 10 1 3 7
n. 11 4 6 7 3 1 10 8 11 9 2 5
n. 12 4 6 11 2 3 5 8 10 7 9 1
n. 13 5 2 9 11 8 10 1 3 7 6 4
Knot 39 n. 14 5 8 9 11 2 4 1 3 7 6 10
n. 15 59 11 2 4 1 3 10 7 6 8
1 n. 1 1 6 2 5 8 10 9 11 7 4 3 n. 16 59 11 8 10 1 3 4 7 6 2 |
n. 2 1 6 3 4 7 11 9 10 8 5 2
n. 3 1 6 8 5 2 4 3 11 7 10 9
n. 4 1 6 9 10 7 11 3 4 2 5 8
n. 5 2 5 8 10 9 11 7 4 3 6 1
n. 6 2 11 8 10 3 5 1 4 9 6 7
Knot 50
n. 7 3 2 5 1 9 8 10 7 4 6 11 1 n. 1 1 4 5 10 3 8 7 2 11 6 9
n. 8 3 4 7 11 9 10 8 5 2 6 1 n. 2 1 6 3 10 9 4 5 2 11 8 7
n. 9 3 8 11 7 9 2 4 1 10 6 5
n. 3 1 6 4 3 2 5 11 9 10 7 8
n. 10 3 10 1 5 9 4 2 11 8 6 7 n. 4 1 6 78 5 2 11 4 3 10 9
n. 11 41 10 8 3 7 11 2 9 6 5
n. 5 1 6 7 8 11 2 5 4 9 103
n. 12 4 7 10 8 9 1 5 2 3 6 11
n. 6 1 6 8 7 4 3 11 5 2 9 10
n. 13 5 6 3 8 11 7 9 2 4 1 10
n. 7 1 6 8 7 10 9 11 5 2 3 4
n. 14 5 6 4 1 10 8 3 7 11 2 9 n. 8 1 6 9 10 3 4 11 2 5 8 7
n. 15 5 6 9 2 11 7 3 8 10 14
n. 9 16 10 9 2 5 11 3 4 7 8
n. 16 56 10 1 4 2 9 7 11 8 3 |
n. 10 1 8 2 5 11 4 3 7 9 6 10
n. 11 1 8 5 2 11 4 3 10 9 6 7
n. 12 18 11 2 5 4 9 10 3 6 7
n. 13 1 10 5 4 9 2 7 8 11 6 3
n. 14 23 10 7 1 9 8 5 4 6 11
Knot 46 n. 15 2 5 4 11 3 7 8 1 10 6 9
n. 16 2 6 3 5 9 8 1 7 10 4 11
1 n. 1 1 6 3 7 4 5 2 11 8 9 10 n. 17 2 6 5 4 10 1 7 8 3 11 9
n. 2 1 6 3 11 8 9 2 7 4 5 10 n. 18 2 6 9 11 3 8 7 1 10 4 5
n. 3 1 6 10 5 4 7 2 9 8 11 3 n. 19 26 11 4 10 7 1 8 9 5 3
n. 4 1 6 10 9 8 1 1 2 5 4 7 3 n. 20 29 10 1 7 3 8 11 4 6 5
n. 5 2 1 8 11 10 9 4 7 3 6 5 n. 21 2 11 4 5 9 1 8 7 10 6 3
n. 6 2 3 4 1 10 7 8 5 9 6 11 n. 22 3 1 9 4 5 11 2 8 7 6 10
n. 7 2 7 8 5 10 3 4 1 9 6 11 n. 23 3 2 9 8 1 10 7 4 5 6 11
n. 8 2 9 4 7 10 1 8 11 3 6 5 n. 24 3 2 9 8 1 10 7 4 11 65
n. 9 3 7 4 5 2 11 8 9 10 6 1 n. 25 3 5 9 8 1 7 10 4 11 6 2
n. 10 3 7 4 9 10 11 8 1 2 6 5 n. 26 3 6 1 10 5 4 9 2 7 8 11
n. 11 3 11 8 1 10 7 4 9 2 6 5 n. 27 3 6 2 11 4 5 9 1 8 7 10
n. 12 3 11 8 9 2 7 4 5 10 6 1 n. 28 3 6 4 1 2 7 9 11 10 58
n. 13 5 6 2 1 8 11 10 9 4 7 3 n. 29 3 6 5 2 1 8 9 10 11 4 7
n. 14 5 6 2 9 4 7 10 1 8 11 3 n. 30 3 6 7 4 11 10 9 8 1 2 5
n. 15 5 6 3 7 4 9 10 11 8 1 2 n. 31 3 6 8 5 10 11 9 7 2 1 4
n. 16 5 6 3 11 8 1 10 7 4 9 2 | n. 32 36 10 7 8 1 9 5 4 11 2 |

280
B. All-Interval Series

1 n. 33 3 6 11 8 7 2 9 4 5 10 1 1 n. 8 1 1 4 2 9 7 3 8 11 6 5 10
n. 34 3 10 9 4 5 2 11 8 1 6 7 n. 9 1 4 3 11 9 10 8 5 6 2 7
n. 35 3 11) 9 4 5 2 11 8 7 6 1 n. 10 1 4 6 9 7 11 2 5 10 3 8
n. 36 4 1 2 7 9 11 10 5 8 6 3 n. 11 1 4 9 5 3 10 8 11 6 2 7
n. 37 4 3 2 5 11 9 10 7 8 6 1 n. 12 1 7 6 8 9 10 11 5 2 4 3
n. 38 4 5 2 3 1 7 10 9 8 6 11 n. 13 1 7 6 9 4 2 11 5 10 3 8
n. 39 4 5 8 9 1 7 10 3 2 6 11 n. 14 1 8 10 3 5 2 11 4 6 9 7
n. 40 4 7 2 1 3 5 10 11 8 6 9 n. 15 1 8 10 9 11 2 5 4 6 3 7
n. 41 4 9 2 11 5 3 10 1 8 6 7 n. 16 1 8 10 9 11 5 2 7 6 3 4
n. 42 4 i:L 2 9 7 1 10 3 8 6 5 n. 17 1 9 6 4 11 2 5 3 10 8 7
n. 43 4 11 8 3 7 1 10 9 2 6 5 n. 18 2 1 6 7 4 3 11 9 10 8 5
n. 44 5 2 1 8 9 10 11 4 7 6 3 n. 19 2 1 6 7 4 9 5 3 10 8 11
n. 45 5 4 1 10 7 8 3 2 9 6 11 n. 20 2 7 6 1 4 3 11 9 10 8 5
n. 46 5 4 7 10 1 8 9 2 3 6 11 n. 21 2 7 6 1 4 9 5 3 10 8 11
n. 47 5 4 10 1 7 8 3 11 9 6 2 n. 22 3 1 5 2 11 10 9 8 6 7 4
n. 48 5 6 2 9 10 1 7 3 8 11 4 n. 23 3 1 6 4 7 2 9 11 10 8 5
n. 49 5 6 3 2 9 8 1 10 7 4 11 n. 24 3 1 6 7 8 10 9 11 2 5 4
n. 50 5 6 4 11 2 9 7 1 10 3 8 n. 25 3 4 2 5 11 10 9 8 6 7 1
n. 51 5 6 4 11 8 3 7 1 10 9 2 n. 26 3 4 6 1 8 10 9 11 5 2 7
n. 52 5 6 8 3 10 1 7 9 2 1 1 4 n. 27 3 5 1 10 7 2 9 4 6 11 8
n. 53 5 6 9 2 3 8 7 10 1 4 11 n. 28 3 5 6 8 11 10 9 7 2 4 1
n. 54 5 6 11 4 1 10 7 8 3 2 9 n. 29 3 5 6 11 4 2 9 7 10 1 8
n. 55 5 6 11 4 7 10 1 8 9 2 3 n. 30 3 7 6 1 8 10 9 11 2 5 4
n. 56 5 8 1 2 3 4 11 10 7 6 9 | n. 31 3 8 6 5 4 2 9 7 1 10 11
n. 32 3 8 10 1 7 2 9 4 6 11 5
n. 33 3 11 6 5 4 2 9 7 10 1 8
n. 34 4 1 2 3 5 10 8 11 6 9 7
Knot 52 n. 35 4 1 6 8 3 10 5 2 11 7 9
n. 1 1 8 2 5 11 7 9 10 3 6 4 n. 36 4 3 2 1 7 10 8 9 6 11 5
n. 2 1 10 5 11 7 9 2 8 3 6 4 n. 37 4 3 2 1 10 7 11 9 6 8 5
n. 3 3 2 9 11 7 1 10 4 5 6 8 n. 38 4 3 6 7 2 5 11 9 10 8 1
n. 4 3 4 10 9 11 7 1 2 5 6 8 n. 39 4 5 2 11 9 10 8 1 6 7 3
n. 5 3 8 2 9 7 11 5 10 1 6 4 n. 40 4 5 2 11 9 10 8 7 6 1 3
n. 6 3 10 9 7 11 5 2 8 1 6 4 n. 41 4 7 2 9 11 10 8 5 6 3 1
n. 7 n. 42 4 7 6 8 9 10 11 2 5 1 3
4 6 1 8 2 5 11 7 9 10 3
n. 8 4 6 1 10 5 11 7 9 2 8 3 n. 43 4 9 2 7 1 10 8 3 6 5 11
n. 9 n. 44 4 9 2 7 10 1 5 3 6 8 11
4 6 3 8 2 9 7 11 5 10 1
n. 10 4 6 3 10 9 7 11 5 2 8 1 n. 45 4 9 6 1 2 11 5 3 10 8 7
n. 11 n. 46 4 11 2 5 3 10 8 1 6 7 9
4 6 7 8 2 11 5 1 3 10 9
n. 12 4 6 7 10 11 5 1 3 2 8 9 n. 47 4 11 2 5 3 10 8 7 6 1 9
n. 13 4 6 9 8 2 3 1 5 11 10 7 n. 48 5 3 6 1 4 2 7 9 10 11 8
n. 14 n. 49 5 3 6 8 1 10 7 9 2 4 11
4 6 9 10 3 1 5 11 2 8 7
n. 15 5 2 1 7 11 9 10 4 3 6 8 n. 50 5 4 2 3 1 10 7 8 6 9 11
|_n. 16 n. 51 5 4 2 9 7 1 10 11 6 3 8
5 4 10 1 7 11 9 2 3 6 8 |
n. 52 5 4 2 9 7 10 1 8 6 3 11
n. 53 5 8 3 7 9 2 4 1 6 10 11
n. 54 5 8 6 9 11 7 10 1 2 3 4
Knot 74 n. 55 5 8 9 1 3 2 4 7 6 10 11
n. 56 5 8 10 9 11 3 4 1 6 7 2
n.~T~ 1 2 6 5 8 10 9 11 3 4 7 n. 57 5 8 10 9 11 3 4 7 6 1 2
n. 2 1 2 6 11 8 10 3 5 9 4 7 n. 58 5 8 10 11 9 2 7 4 6 1 3
n. 3 1 2 11 5 3 10 8 7 6 4 9 n. 59 5 9 6 8 7 10 1 3 2 4 11
n. 4 1 3 6 4 5 2 11 9 10 8 7 n. 60 5 10 6 1 4 2 9 7 3 8 11
[ n. 5 1 3 6 5 8 10 11 9 2 7 4 n. 61 5 10 6 7 4 2 3 1 9 8 11
n. 6 1 4 2 7 9 10 11 8 6 5 3 n. 62 5 10 7 1 3 2 4 11 6 8 9
n. 7 1 4 2 9 7 3 8 5 6 11 10 n. 63 5 11 6 4 9 2 7 1 10 8 3
n. 64 5 11 6 9 8 10 7 1 2 3 4 |

281
B. All-Interval Series

Knot 75 Knot 80
1 n. 1 1 1 4 2 9 6 5 8 3 7 11 10 1 n. 1 1 2 5 3 10 8 11 6 9 7 4
n. 2 1 4 3 11 7 2 6 5 8 10 9 n. 2 1 2 5 6 8 7 11 3 4 10 9
n. 3 1 7 11 10 9 8 6 5 2 4 3 n. 3 1 4 2 9 7 10 11 6 8 5 3
n. 4 1 1 0 8 3 6 5 11 7 2 9 4 n. 4 1 10 5 6 9 2 8 3 7 11 4
n. 5 2 1 5 9 4 7 6 3 10 8 11 n. 5 2 1 4 6 10 11 7 3 8 5 9
n. 6 2 1 5 9 4 10 6 3 7 11 8 n. 6 2 1 5 9 4 7 3 6 10 11 8
n. 7 2 3 10 5 1 4 6 9 7 11 8 n. 7 2 1 10 6 4 11 7 3 8 5 9
n. 8 2 5 1 3 6 4 7 11 10 9 8 n. 8 2 5 10 6 9 1 4 3 11 7 8
n. 9 2 5 1 9 6 10 4 3 11 7 8 n. 9 2 7 4 6 10 5 1 9 8 11 3
n. 10 2 7 11 3 4 1 6 9 10 8 5 n. 10 2 7 10 6 4 5 1 9 8 11 3
n. 11 2 7 11 3 4 10 6 9 1 5 8 n. 11 2 7 11 3 4 1 9 6 10 5 8
n. 12 2 8 3 7 11 4 6 10 1 5 9 n. 12 2 11 10 6 3 7 4 9 5 1 8
n. 13 2 8 3 7 11 10 6 4 1 5 9 n. 13 3 1 4 6 7 2 11 9 10 8 5
n. 14 2 8 9 1 5 4 6 10 7 11 3 n. 14 3 2 8 9 1 5 4 6 7 10 11
n. 15 2 8 9 1 5 10 6 4 7 11 3 n. 15 3 5 8 6 11 10 7 9 2 4 1
n. 16 2 9 10 11 7 4 6 3 1 5 8 n. 16 3 7 4 9 5 1 2 6 8 11 10
n. 17 2 11 7 3 6 10 4 9 5 1 8 n. 17 3 7 4 9 5 1 8 6 2 11 10
n. 18 2 11 7 9 6 4 1 5 10 3 8 n. 18 3 10 4 9 5 1 8 6 11 2 7
n. 19 3 1 5 2 6 8 9 10 11 7 4 n. 19 3 11 8 9 1 5 4 6 10 7 2
n. 20 3 1 5 8 6 2 9 10 11 7 4 n. 20 3 11 8 9 1 5 10 6 4 7 2
n. 21 3 2 4 7 6 10 5 1 9 8 11 n. 21 4 1 2 6 9 5 8 3 7 11 10
n. 22 3 4 2 5 6 8 9 10 11 7 1 n. 22 4 1 3 6 5 8 10 9 11 2 7
n. 23 3 5 1 4 6 10 9 2 7 11 8 n. 23 4 5 1 9 8 2 3 6 11 10 7
n. 24 3 5 1 10 6 4 9 2 7 11 8 n. 24 4 5 1 9 8 11 3 6 2 7 10
n. 25 3 7 11 2 6 8 1 5 9 4 10 n. 25 4 7 2 6 3 11 8 9 1 5 10
n. 26 3 7 11 8 6 2 1 5 9 4 10 n. 26 4 7 9 6 11 8 10 3 5 2 1
n. 27 3 8 10 1 6 4 9 2 7 11 5 n. 27 4 11 7 3 8 2 9 6 5 10 1
n. 28 3 10 8 11 6 2 1 5 9 4 7 n. 28 4 11 7 3 8 5 9 6 2 1 10
n. 29 3 11 7 4 6 10 5 1 9 8 2 n. 29 5 2 1 6 9 10 4 3 11 7 8
n. 30 3 11 7 10 6 4 5 1 9 8 2 n. 30 5 8 10 9 11 2 7 6 4 1 3
n. 31 4 1 5 3 6 8 11 7 2 9 10 n. 31 5 10 1 3 2 4 7 6 9 11 8
n. 32 4 1 5 9 6 2 8 3 7 11 10 n. 32 5 10 1 6 4 11 7 3 8 2 9 |
n. 33 4 1 5 10 3 2 6 8 11 7 9
n. 34 4 1 5 10 3 8 6 2 11 7 9
n. 35 4 3 2 1 5 8 6 9 11 7 10
n. 36 4 3 2 1 5 11 6 9 8 10 7
n. 37 4 5 1 9 8 2 6 3 11 7 10 Knot 81
n. 38 4 7 11 3 6 2 8 9 1 5 10
1 n. 1 2 3 8 6 4 11 10 7 1 5 9
n. 39 4 7 11 9 6 8 5 1 2 3 10
n. 2 2 5 4 6 8 9 10 1 7 11 3
n. 40 4 7 11 10 9 2 6 8 5 1 3
n. 3 2 9 8 6 4 5 10 1 7 11 3
n. 41 4 7 11 10 9 8 6 2 5 1 3
n. 4 2 11 4 6 8 3 10 7 1 5 9
n. 42 4 9 2 7 11 5 6 3 8 10 1
n. 5 3 7 11 5 2 1 8 6 4 9 10
n. 43 4 9 2 7 11 8 6 3 5 1 10
n. 6 3 7 11 5 2 9 4 6 8 1 10
n. 44 4 11 7 3 8 2 6 9 5 1 10
n. 7 3 11 7 1 10 5 4 6 8 9 2
n. 45 5 2 4 3 6 1 7 11 10 9 8
n. 8 3 11 7 1 10 9 8 6 4 5 2
n. 46 5 8 3 7 11 10 6 1 4 2 9
n. 9 4 3 2 11 5 1 9 6 10 7 8
n. 47 5 8 10 9 6 1 4 3 11 7 2
n. 10 4 3 10 6 9 1 5 1 1 2 7 8
n. 48 5 11 7 2 9 4 6 1 10 8 3 |
n. 11 4 5 2 6 3 11 7 1 10 9 8
n. 12 4 5 10 1 7 11 3 6 2 9 8
n. 13 4 9 2 5 11 7 3 6 10 1 8
n. 14 4 9 10 6 3 7 11 5 2 1 8
n. 15 4 11 2 6 9 5 1 7 10 3 8
n. 16 4 11 10 7 1 5 9 6 2 3 8

282
B. All-Interval Series

Knot 87 Knot 89
1 n. 1 1 3 6 7 10 89 11 254 n. 1 1 2 4 9 10 7 11 3 6 5 8
n. 2 1 9 4 5 2 7 11 863 10 n. 2 1 4 6 5 10 892 1173
n. 3 110 8 3 5 2 11 4 6 7 9 n. 3 37 11 298 10 5 6 4 1
n. 4 2 3 6 4 7 11 10 1895 n. 4 3 11 7 10 9 4 2 1 6 8 5
n. 5 2 9 6 4 1 5 10 783 11 n. 5 4 1 6 3 7 11 298 10 5
n. 6 3 1 6 4 5 2 11 98 10 7 n. 6 4 7 6 9 1 5 2 3 8 10 11
n. 7 3 2 6 5 9 8 1 10 11 74 n. 7 5 8 6 1 2 4 9 10 7 11 3
n. 8 3 5 6 8 1 10 7 9 4 2 11 n. 8 5 10 8 9 2 11 7 3 6 1 4
n. 9 3 10 6 1 9 4 5 2 7 11 8
n. 10 415 10 783 11 629
n. 11 452 11 98 10 7 6 3 1
n. 12 47 11 10 1 8 9 5 6 2 3
n. 13 4 11 2 5 3 8 10 1 6 9 7
n. 14 5 2 4 3 1 10 786 11 9
n. 15 536 11 2 4 9 7 10 18 Knot 96
n. 16 5 9 8 1 10 11 7 4 6 3 2 |
1 n. 1 1 3 6 5 11 7 8 2 4 9 10
n. 2 1 7 11 4 10 8 9 2 6 5 3
n. 3 1 9 6 5 2 8 10 3 7 11 4
n. 4 1 10 4 2 3 11 7 8 6 5 9
Knot 95 n. 5 2 1 5 9 11 7 10 4 6 3 8
n. 6 2 3 6 4 10 7 11 9 5 1 8
1 n. 1 1 2 4 9 7 10 11 6 3 5 8 n. 7 236 10 7 11 4 1 5 9 8
n. 2 1 2 5 3 8 10 11 6 4 7 9 n. 8 238 10 4 5 1 7 6 9 11
n. 3 1 2 5 6 3 4 10 97 11 8 n. 9 23 11 7 4 5 1 10 698
n. 4 1 4 2 8 5 3 11 6 10 79 n. 10 2 5 1 8 11 7 3 4 6 10 9
n. 5 1456 10 982 11 73 n. 11 2 5 1 8 11 73 10 649
n. 6 1 8 5 6 3 11 7 10 492 n. 12 27 11 351 10 4 6 9 8
n. 7 1 9 7 4 10 8 11 6 3 5 2 n. 13 2 9 5 1 4 11 7 10 638
n. 8 1 10 5 6 4 7 11 9 2 8 3 n. 14 2 9 6 4 10 153 11 78
n. 9 234 10 1596 11 87 n. 15 296 10 1 5 4 7 11 38
n. 10 2 5 3 6 11 8 10 4 7 9 1 n. 16 298 10 4 11 7 1 6 3 5
n. 11 294 10 7 11 3 6 5 8 1 n. 17 2 11 7 8 5 1 9 4 6 10 3
n. 12 2 11 9 6 5 8 10 4 1 3 7 n. 18 2 11 7 8 5 1 9 10 643
n. 13 3 1 4 6 5 10 89 11 27 n. 19 316 10 9 4 2 8 7 11 5
n. 14 31 10 659 11 8 2 4 7 n. 20 3 2 6 8 1 5 9 11 7 10 4
n. 15 34 10 97 11 8 6 1 2 5 n. 21 3 2 6 8 9 5 1 4 11 7 10
n. 16 3 5 2 6 1 9 7 4 10 8 11 n. 22 346 10 5 1 9 7 11 28
n. 17 3 5 8 6 1 2 4 9 7 10 11 n. 23 346 10 9 1 5 8 7 11 2
n. 18 37 11 289 10 6 5 4 1 n. 24 3 5 6 2 9 8 10 4 11 71
n. 19 3 8 2 9 11 7 4 6 5 10 1 n. 25 3 7 6 4 5 1 9 10 82 11
n. 20 3 11 7 10 4 9 2 6 1 8 5 n. 26 3 8 6 2 1 5 9 11 7 10 4
n. 21 4 1 3 6 7 2 11 98 10 5 n. 27 3 8 6 2 9 5 1 4 11 7 10
n. 22 4 1 5 3 2 8 9 6 7 10 11 n. 28 3 10 6 4 5 1 9 7 11 28
n. 23 4 7 9 6 1 2 5 3 8 10 11 n. 29 3 10 6 4 9 1 5 8 7 11 2
n. 24 47 11 9 2 8 3 6 1 10 5 n. 30 3 11 6 8 1 5 9 2 4 10 7
n. 25 5 2 1 6 8 11 79 10 43 n. 31 4 3 6 2 11 7 8 5 1 9 10
n. 26 5 4 1 6 3 7 11 289 10 n. 32 4 3 6 8 2 11 7 9 1 5 10
n. 27 5 8 1 6 2 9 4 10 7 11 3 n. 33 4 3 7 11 8 1 5 2 6 9 10
n. 28 58 10 4 1 3 7 6 2 11 9 n. 34 4 5 1 9 7 11 2 8 6 3 10
n. 29 59 11 8 2 4 7 6 3 1 10 n. 35 4 5 1 9 10 82 11 637
n. 30 5 10 1 3 4 2 7 6 8 11 9 n. 36 4 9 1 5 8 7 11 263 10
n. 31 5 10 1 6 3 8 2 9 11 74 n. 37 4 9 6 2 5 1 8 11 73 10
L_n. 32 5 10 89 11 2 7 6 3 1 4 | n. 38 4 9 6 8 2 5 1 3 7 11 10 |

283
B. All-Interval Series

1 n. 39 4 11) 1 5 3 11 7 2 6 8 9 n. 41 5 8 10 3 2 4 7 6 1 9 11
n. 40 4 11) 1 5 3 11 7 8 6 2 9 n. 42 5 8 10 9 2 4 1 6 7 3 11
n. 41 4 10 7 11 9 5 1 2 6 8 3 n. 43 5 8 10 9 2 4 7 6 1 3 11
n. 42 4 10 7 11 9 5 1 8 6 2 3 n. 44 5 9 1 6 7 4 2 3 10 8 11
n. 43 4 11 7 3 1 5 2 8 6 9 10 n. 45 5 9 7 6 1 4 2 3 10 8 11
n. 44 4 11 7 3 10 8 2 5 6 9 1 n. 46 5 10 8 11 4 2 3 6 7 1 9
n. 45 5 2 8 10 3 7 11 4 6 1 9 n. 47 5 11 3 6 7 2 4 1 8 10 9
n. 46 5 3 6 1 7 11 4 10 8 9 2 n. 48 5 11 9 6 7 8 10 1 2 4 3 |
n. 47 5 9 6 1 10 4 2 3 11 7 8
n. 48 5 11 7 8 2 4 9 10 6 1 3 |

Knot 103 Knot 104


1 n. 1 1 2 4 7 8 10 3 6 11 5 9 1 n. 1 2 3 11 7 4 6 10 1 5 9 8
n. 2 1 3 5 6 11 8 10 9 2 4 7 n. 2 2 3 11 7 10 6 4 1 5 9 8
n. 3 1 3 11 6 5 8 10 9 2 4 7 n. 3 2 5 1 9 4 6 10 3 7 11 8
n. 4 1 4 2 3 10 8 5 6 1 1 9 7 n. 4 2 5 1 9 10 6 4 3 7 11 8
n. 5 1 4 2 3 10 8 11 6 5 9 7 n. 5 2 9 5 1 4 6 10 7 11 3 8
n. 6 1 4 2 9 10 8 5 6 11 3 7 n. 6 2 9 5 1 10 6 4 7 11 3 8
n. 7 1 4 2 9 10 8 11 6 5 3 7 n. 7 2 11 7 3 4 6 10 9 1 5 8
n. 8 1 7 3 6 11 10 8 5 4 2 9 n. 8 2 11 7 3 10 6 4 9 1 5 8
n. 9 1 7 9 6 11 4 2 5 10 8 3 n. 9 4 1 5 9 2 6 8 3 11 7 10
n. 10 1 8 10 7 2 4 9 6 11 5 3 n. 10 4 1 5 9 8 6 2 3 11 7 10
n. 11 1 9 5 6 11 8 10 3 2 4 7 n. 11 4 3 7 11 2 6 8 5 1 9 10
n. 12 1 9 11 6 5 8 10 3 2 4 7 n. 12 4 3 7 11 8 6 2 5 1 9 10
n. 13 2 1 5 8 7 11 3 6 10 4 9 n. 13 4 7 11 3 2 6 8 9 5 1 10
n. 14 2 7 11 8 1 5 9 6 10 4 3 n. 14 4 7 11 3 8 6 2 9 5 1 10
n. 15 2 8 3 6 10 11 7 4 5 1 9 n. 15 4 9 1 5 2 6 8 11 7 3 10
n. 16 2 8 9 6 10 5 1 4 11 7 3 n. 16 4 9 1 5 8 6 2 11 7 3 10 |
n. 17 3 1 5 10 7 11 4 6 9 2 8
n. 18 3 1 7 6 9 8 10 11 2 4 5
n. 19 3 2 4 11 8 10 5 6 9 1 7
n. 20 3 2 8 6 9 7 11 10 1 5 4
n. 21 3 4 2 1 10 8 7 6 9 11 5
n. 22 3 4 10 6 9 5 1 8 11 7 2
n. 23 3 5 1 2 11 7 8 6 9 10 4 Knot 108
n. 24 3 5 11 6 9 4 2 7 10 8 1
n. 25 3 7 1 6 9 2 4 5 8 10 11 1 n. 1 1 2 5 11 3 6 7 10 8 9 4
n. 26 3 7 11 4 1 5 10 6 9 8 2 n. 2 1 2 7 11 8 6 3 5 9 4 10
n. 27 3 8 2 6 9 1 5 4 7 11 10 n. 3 1 10 8 3 4 6 7 2 11 5 9
n. 28 3 8 10 5 2 4 11 6 9 7 1 n. 4 2 8 3 7 9 6 4 1 5 10 11
n. 29 3 10 4 6 9 11 7 2 5 1 8 n. 5 2 8 9 1 3 6 4 7 11 10 5
n. 30 3 10 8 7 4 2 1 6 9 5 11 n. 6 3 1 9 8 2 6 5 10 11 7 4
n. 31 3 11 5 6 9 10 8 1 4 2 7 n. 7 3 5 9 4 10 6 1 2 7 11 8
n. 32 3 11 7 8 5 1 2 6 9 4 10 n. 8 3 7 1 10 5 6 8 9 4 2 11
n. 33 4 5 1 10 11 7 9 6 8 2 3 n. 9 3 11 5 2 1 6 4 9 8 10 7
n. 34 4 10 3 6 8 1 5 2 7 11 9 n. 10 4 1 5 10 11 6 2 8 3 7 9
n. 35 4 10 9 6 8 7 11 2 1 5 3 n. 11 4 3 8 10 1 6 9 5 1 1 2 7
n. 36 4 11 7 10 5 1 3 6 8 2 9 n. 12 4 7 11 10 5 6 2 8 9 1 3
n. 37 5 3 1 6 7 4 2 9 10 8 11 n. 13 4 9 8 10 7 6 3 11 5 2 1
n. 38 5 3 7 6 1 4 2 9 10 8 11 n. 14 5 2 4 3 8 6 11 10 7 1 9
n. 39 5 4 2 11 10 8 9 6 7 1 3 n. 15 5 10 1 7 3 6 11 2 4 9 8
n. 40 5 8 10 3 2 4 1 6 7 9 11 | n. 16 5 10 11 7 4 6 3 1 9 8 2

284
B. All-Interval Series

Knot 109 Knot 111


n. 1 1 3 7 11 10 6 5 8 2 4 9 1 n. 1 1 2 4 7 3 6 11 10 8 5 9
n. 2 1 4 10 8 9 6 5 3 1 1 7 2 n. 2 1 3 10 8 5 6 11 9 2 4 7
n. 3 2 1 5 9 11 6 3 8 10 4 7 n. 3 1 3 10 8 11 6 5 9 2 4 7
n. 4 2 7 11 3 5 6 9 8 10 4 1 n. 4 1 4 2 3 5 6 11 8 10 9 7
n. 5 3 4 2 8 11 6 10 5 1 9 7 n. 5 1 4 2 3 11 6 5 8 10 9 7
n. 6 3 8 10 4 7 6 2 1 5 9 11 n. 6 1 4 2 9 5 6 11 8 10 3 7
n. 7 5 3 11 7 2 6 1 4 10 8 9 n. 7 1 4 2 9 11 6 5 8 10 3 7
n. 8 5 8 2 4 9 6 1 3 7 11 10 n. 8 1 7 2 4 9 6 11 5 10 8 3
n. 9 1 7 8 10 3 6 11 5 4 2 9
n. 10 1 8 10 7 9 6 1 1 4 2 5 3
n. 11 1 9 10 8 5 6 1 1 3 2 4 7
n. 12 1 9 10 8 11 6 5 3 2 4 7
n. 13 2 1 5 8 3 6 10 11 7 4 9
n. 14 2 7 11 8 9 6 10 5 1 4 3
Knot 110 n. 15 2 8 1 5 9 6 10 4 11 7 3
n. 16 2 8 7 11 3 6 10 4 5 1 9
1 n. 1 1 3 10 8 7 6 4 5 11 2 9 n. 17 3 1 5 10 4 6 9 11 7 2 8
n. 2 1 4 2 7 8 6 11 5 3 10 9 n. 18 3 1 10 8 7 6 9 11 2 4 5
n. 3 1 4 11 3 7 6 2 5 8 10 9 n. 19 3 2 4 11 5 6 9 10 8 1 7
n. 4 1 7 9 2 3 6 11 8 10 5 4 n. 20 3 2 11 7 8 6 9 10 1 5 4
n. 5 1 8 10 9 7 6 3 2 5 11 4 n. 21 3 4 1 5 10 6 9 8 11 7 2
n. 6 1 9 5 4 7 6 3 10 8 11 2 n. 22 3 4 2 1 7 6 9 8 10 11 5
n. 7 1 9 5 4 10 6 3 7 11 8 2 3 5 1 2 8 6 9 7 11 10 4
n. 23
n. 8 2 5 8 10 9 6 1 4 11 3 7 n. 24 3 5 2 4 11 6 9 7 10 8 1
n. 9 2 8 1 9 5 6 10 4 7 11 3 3 7 4 2 1 6 9 5 8 10 11
n. 25
n. 10 2 8 5 1 9 6 10 4 11 3 7 n. 26 3 7 11 4 10 6 9 5 1 8 2
n. 11 2 8 7 3 11 6 10 4 1 5 9 n. 27 3 8 5 1 2 6 9 4 7 11 10
n. 12 2 8 11 7 3 6 10 4 5 9 1 3 8 10 5 11 6 9 4 2 7 1
n. 28
n. 13 2 11 8 10 3 6 7 4 5 9 1 n. 29 3 10 7 11 4 6 9 2 5 1 8
n. 14 3 2 4 7 10 6 5 9 1 8 11 n. 30 3 10 8 7 1 6 9 2 4 5 11
n. 15 3 2 5 11 4 6 1 8 10 9 7 n. 31 3 11 7 8 2 6 9 1 5 4 10
n. 16 3 2 9 7 1 6 4 5 10 8 11 n. 32 3 11 8 10 5 6 9 1 4 2 7
n. 17 3 7 11 8 2 6 1 9 5 4 10 n. 33 4 5 1 10 9 6 8 7 11 2 3
n. 18 3 10 1 7 8 6 5 4 2 9 11 n. 34 4 10 5 1 3 6 8 2 7 11 9
n. 19 3 10 8 11 2 6 1 9 5 4 7 4 10 11 7 9 6 8 2 1 5 3
n. 35
n. 20 3 10 9 11 5 6 8 1 2 4 7 n. 36 4 11 7 10 3 6 8 1 5 2 9
n. 21 3 11 7 4 10 6 5 9 1 8 2 n. 37 5 3 2 4 1 6 7 9 10 8 11
n. 22 4 5 10 8 11 6 3 2 9 7 1 5 3 2 4 7 6 1 9 10 8 11
n. 38
n. 23 4 5 11 2 9 6 1 3 10 8 7 n. 39 5 4 2 11 9 6 7 8 10 1 3
n. 24 4 11 5 2 3 6 7 9 10 8 1 n. 40 5 8 10 3 1 6 7 4 2 9 11
n. 25 4 11 10 8 5 6 9 2 3 1 7 n. 41 5 8 10 3 7 6 1 4 2 9 11
n. 26 5 3 2 4 11 6 8 1 7 10 9 n. 42 5 8 10 9 1 6 7 4 2 3 11
n. 27 5 4 2 9 11 6 3 10 1 7 8 n. 43 5 8 10 9 7 6 1 4 2 3 11
n. 28 5 8 7 3 11 6 10 1 4 2 9 n. 44 5 9 2 4 1 6 7 3 10 8 11
n. 29 5 8 10 11 4 6 7 1 3 2 9 n. 45 5 9 2 4 7 6 1 3 10 8 11
n. 30 5 9 1 8 2 6 3 11 7 4 10 5 10 8 11 3 6 7 2 4 1 9
n. 46
n. 31 5 9 1 8 11 6 3 2 4 7 10 n. 47 5 11 4 2 3 6 7 1 8 10 9
n. 32 5 11 9 10 3 6 7 4 2 1 8 | n. 48 5 11 10 8 9 6 7 1 2 4 3 |

285
B. All-Interval Series

Knot 131
1 n. 1 1 1 2 4 7 9 5 6 11 8 3 10 1 n. 55 3 8 2 1 5 9 6 7 4 11 10
n. 2 1 2 4 7 9 11 6 5 8 3 10 n. 56 3 8 5 10 6 11 2 4 9 7 1
n. 3 1 2 4 10 3 8 6 11 5 9 7 n. 57 3 8 9 2 6 1 4 10 7 11 5
n. 4 1 2 4 10 9 8 6 5 11 3 7 n. 58 3 8 9 5 1 2 6 7 4 10 11
n. 5 1 2 7 4 6 3 8 10 11 5 9 n. 59 3 8 10 4 7 2 6 1 9 5 11
n. 6 1 2 8 5 6 10 11 7 3 4 9 n. 60 3 8 10 11 5 9 6 1 2 7 4
n. 7 1 3 5 8 10 11 6 2 9 4 7 n. 61 3 10 4 11 7 9 6 5 2 1 8
n. 8 1 3 11 5 6 8 9 10 4 2 7 n. 62 3 10 8 1 7 9 6 11 4 5 2
n. 9 1 4 3 2 6 5 10 8 11 9 7 n. 63 3 11 5 10 8 9 6 4 1 2 7
n. 10 1 4 5 10 6 9 8 2 7 1 1 3 n. 64 3 11 7 2 8 9 6 10 5 4 1
n. 11 1 4 9 2 6 11 10 8 5 3 7 n. 65 4 1 2 7 6 3 11 5 10 8 9
n. 12 1 4 10 5 6 9 8 3 11 7 2 n. 66 4 3 2 8 10 5 6 11 9 1 7
n. 13 1 4 10 7 11 5 6 3 8 9 2 n. 67 4 3 2 8 10 11 6 5 9 1 7
n. 14 1 7 3 5 6 11 10 8 2 9 4 n. 68 4 5 10 1 6 9 11 7 8 2 3
n. 15 1 7 3 11 6 5 10 8 2 9 4 n. 69 4 7 2 1 6 9 5 11 10 8 3
n. 16 1 7 3 11 6 10 5 8 2 4 9 n. 70 4 9 2 8 10 5 6 11 3 7 1
n. 17 1 7 9 4 2 11 6 10 5 8 3 n. 71 4 9 2 8 10 11 6 5 3 7 1
n. 18 1 7 11 2 8 5 6 10 9 4 3 n. 72 4 11 10 7 6 3 5 1 8 2 9
n. 19 1 8 7 10 6 9 2 4 11 5 3 n. 73 5 2 1 8 6 3 10 4 11 7 9
n. 20 1 9 5 11 6 3 8 10 4 7 2 n. 74 5 2 4 3 1 7 6 9 8 11 10
n. 21 1 9 5 11 6 8 3 10 4 2 7 n. 75 5 3 1 4 2 7 6 10 9 8 11
n. 22 1 9 11 8 10 5 6 2 3 4 7 n. 76 5 3 7 1 6 4 9 2 8 10 11
n. 23 1 10 5 4 6 3 2 8 7 11 9 n. 77 5 4 11 2 6 9 10 8 7 1 3
n. 24 1 10 8 3 5 11 6 9 4 7 2 n. 78 5 8 1 2 6 9 4 10 11 7 3
n. 25 2 1 4 3 6 5 11 9 8 10 7 n. 79 5 8 2 1 6 9 4 3 7 11 10
n. 26 2 1 4 10 8 9 6 5 11 3 7 n. 80 5 8 2 11 7 1 6 3 4 9 10
n. 27 2 1 5 9 8 3 6 11 10 4 7 n. 81 5 8 3 10 6 1 2 4 7 9 11
n. 28 2 1 8 5 6 3 7 11 10 4 9 n. 82 5 8 9 10 6 7 2 4 1 3 11
n. 29 2 3 4 1 6 7 9 11 8 10 5 n. 83 5 9 1 7 6 3 4 2 8 11 10
n. 30 2 3 4 7 6 1 9 11 8 10 5 n. 84 5 9 1 7 6 4 3 2 8 10 11
n. 31 2 3 8 9 6 11 5 1 10 4 7 n. 85 5 9 7 4 2 1 6 10 3 8 11
n. 32 2 5 4 11 6 9 7 1 8 10 3 n. 86 5 10 4 1 6 2 7 11 3 8 9
n. 33 2 7 4 9 6 11 5 3 8 10 1 n. 87 5 10 8 2 3 4 6 7 1 9 11
n. 34 2 7 4 10 8 3 6 11 5 9 1 n. 88 5 10 8 2 9 4 6 1 7 3 11
n. 35 2 7 8 11 6 9 1 5 10 4 3 n. 89 5 10 8 11 9 1 6 7 4 3 2
n. 36 2 7 11 3 8 9 6 5 10 4 1 n. 90 5 10 8 11 9 7 6 1 4 3 2
n. 37 2 9 4 1 6 7 3 5 8 10 11 n. 91 5 10 11 8 6 3 4 2 7 1 9
n. 38 2 9 4 7 6 1 3 5 8 10 11 n. 92 5 11 3 1 6 7 2 4 10 9 8
n. 39 2 9 8 3 6 5 11 7 10 4 1 n. 93 5 11 3 7 6 1 2 4 10 9 8
n. 40 2 11 4 5 6 3 1 7 8 10 9 n. 94 5 11 3 7 6 2 1 4 10 8 9
n. 41 3 1 5 4 10 9 6 8 7 2 11 n. 95 5 11 7 10 4 1 6 2 9 8 3
n. 42 3 1 7 8 10 9 6 2 11 4 5 n. 96 5 11 9 8 10 7 6 2 1 4 3 |
n. 43 3 2 4 5 11 9 6 7 8 1 10
n. 44 3 2 8 7 11 9 6 1 10 5 4
n. 45 3 4 1 2 6 7 10 8 9 11 5
n. 46 3 4 2 7 1 9 6 5 10 11 8
n. 47 3 4 2 8 11 10 6 5 9 1 7
n. 48 3 4 9 1 5 10 6 11 8 2 7
n. 49 3 4 9 10 6 5 8 2 1 1 7 1
n. 50 3 4 10 5 1 9 6 11 8 7 2
n. 51 3 5 1 8 2 9 6 4 11 10 7
n. 52 3 5 11 4 2 9 6 10 7 8 1
n. 53 3 7 1 2 4 9 6 8 5 10 11
n. 54 | 3 7 11 10 4 9 6 2 1 8 5 |
B. All-Interval Series

Knot 132 Knot 138


1 n. 1 1 4 2 3 5 11 6 8 7 10 9 1 n. 1 1 3 6 10 7 2 4 5 9 8 11
n. 2 1 7 9 10 8 11 6 3 2 5 4 n. 2 1 4 3 7 8 10 5 2 6 9 11
n. 3 1 8 2 5 6 9 10 3 7 11 4 n. 3 1 9 6 4 7 8 10 5 3 2 11
n. 4 110 4 5 6 8 7 11 3 2 9 n. 4 1 10 9 7 2 4 5 8 6 3 11
n. 5 3 2 5 4 6 1 7 9 10 8 11 n. 5 2 5 10 8 7 3 4 1 6 11 9
n. 6 3 2 9 5 1 8 6 11 4 10 7 n. 6 2 11 10 8 1 9 4 7 6 5 3
n. 7 3 10 1 8 6 5 11 9 2 4 7 n. 7 3 1 6 11 8 9 5 4 2 7 10
n. 8 3 10 9 1 5 4 6 7 8 2 11 n. 8 3 5 6 7 4 9 1 8 10 11 2
n. 9 4 5 1 9 10 3 6 11 2 8 7 n. 9 3 7 6 5 2 9 11 10 8 1 4
n. 10 4 5 2 3 6 11 8 10 9 7 1 n. 10 3 11 6 1 10 9 7 2 4 5 8
n. 11 4 11 2 9 6 5 8 10 3 1 7 n. 11 4 1 8 10 11 9 2 5 6 7 3
n. 12 4 11 7 3 10 9 6 5 2 8 1 n. 12 4 7 8 10 5 3 2 11 6 1 9
n. 13 5 2 8 1 6 4 11 7 3 10 9 n. 13 5 2 9 11 10 8 1 4 6 3 7
n. 14 54 10 1 6 9 2 3 11 78 n. 14 5 3 6 2 11 10 8 1 9 4 7
n. 15 58 10 3 1 7 6 4 11 29 n. 15 5 8 3 11 4 2 1 10 6 9 7
n. 16 5 11 9 2 4 7 6 3 10 18 | n. 16 5 9 6 8 11 4 2 1 3 10 7 |

Knot 134 Knot 139


1 n. 1 1 3 5 2 6 9 8 10 7 4 11 1 n. 1 1 3 4 11 7 8 5 2 6 10 9
n. 2 1 3 7 4 6 5 8 10 9 2 11 n. 2 1 4 6 8 3 10 7 11 2 5 9
n. 3 1 3 7 10 8 9 6 2 5 4 11 n. 3 1 4 6 8 9 10 7 11 2 5 3
n. 4 1 4 2 9 10 7 6 5 3 1 1 8 n. 4 1 9 10 11 7 2 5 8 6 4 3
n. 5 1 4 9 8 10 7 6 2 5 3 11 n. 5 2 3 6 11 9 8 1 5 4 7 10
n. 6 1 8 5 2 4 3 6 10 7 9 11 n. 6 2 5 8 7 11 4 3 1 6 9 10
n. 7 1 8 7 10 6 3 4 2 5 9 11 n. 7 2 9 6 5 3 8 7 11 4 1 10
n. 8 197 10 6 5 2 4 3 8 11 n. 8 2 11 8 1 5 4 9 7 6 3 10
n. 9 1 1 0 3 2 4 7 6 8 5 9 11 n. 9 3 1 10 7 11 2 9 4 6 8 5
n. 10 1 10 8 3 4 7 6 5 9 11 2 n. 10 3 2 6 10 7 4 5 1 8 9 11
n. 11 2 5 3 1 6 11 47 10 89 n. 11 3 4 6 8 5 2 7 11 10 9 1
n. 12 253 11 6 1 4 9 8 10 7 n. 12 3 5 2 11 7 10 9 8 6 4 1
n. 13 254 11 6 1 3 7 10 89 n. 13 3 7 10 1 5 2 9 4 6 8 11
n. 14 2 11 4 5 6 7 9 1 10 83 n. 14 3 8 6 4 1 10 11 7 2 9 5
n. 15 2 11 9 5 6 7 4 3 8 10 1 n. 15 3 10 6 2 11 8 1 5 4 9 7
n. 16 2 11 9 7 6 5 4 1 10 83 n. 16 3 11 251 10 9 8 6 4 7
n. 17 3 4 2 5 8 1 6 11 97 10 n. 17 4 1 6 3 5 2 11 7 10 98
n. 18 3 4 2 5 9 11 6 1 8 7 10 n. 18 4 1 6 9 5 2 11 7 10 38
n. 19 38 10 1 4 5 6 7 9 11 2 n. 19 41 10 11 7 2 9 5 6 3 8
n. 20 38 10 1 9 7 6 5 4 11 2 n. 20 4 3 2 5 1 10 796 11 8
n. 21 4 1 9 7 6 5 2 3 10 8 11 n. 21 432 11 7 10 1 9 6 5 8
n. 22 4 7 3 1 6 11 29 10 85 n. 22 4 3 6 1 9 10 11 7 2 5 8
n. 23 523 10 8 11 6 4 1 9 7 n. 23 4 7 6 3 11 251 10 98
n. 24 5 2 4 3 8 11 6 1 9 7 10 n. 24 4 7 6 9 11 251 10 38
n. 25 531 10 6 9 4 2 11 87 n. 25 47 10 5 1 2 3 11 698
n. 26 53 11 2496 10 187 n. 26 4 9 2 5 1 10 736 11 8
n. 27 53 11 8 6 1 4 2 9 10 7 n. 27 492 11 7 10 1 3 6 5 8
n. 28 541 10 8 3 6 2 11 97 n. 28 4 9 6 7 3 10 512 11 8
n. 29 54 11 2 6 3 8 10 197 n. 29 5 3 8 7 11 41 10 629
n. 30 5 8 9 4 2 11 6 10 137 n. 30 5 8 6 4 3 2 11 7 10 19
n. 31 58 10 92 11 6 1 3 7 4 n. 31 5 8 6 4 9 2 11 7 10 13
n. 32 59 11 261 10 8 3 4 7 | n. 32 5 9 2 7 11 10 1 4 6 8 3 |

287
B. All-Interval Series

Knot 145 Knot 146


1 n. 1 1 1 2 11 7 10 9 4 3 6 5 8 1 n. 1 1 1 3 6 10 11 8 2 4 5 9 7
n. 2 1 4 6 5 10 7 11 2 9 8 3 n. 2 1 3 11 4 2 5 8 7 6 9 10
n. 3 1 4 6 9 2 11 7 10 5 8 3 n. 3 1 3 11 4 2 8 5 10 6 9 7
n. 4 1 4 6 9 11 2 7 10 8 5 3 n. 4 1 7 6 8 9 10 4 2 5 11 3
n. 5 1 9 4 7 10 8 5 3 6 11 2 n. 5 1 7 6 9 5 11 2 4 10 3 8
n. 6 1 10 6 3 8 11 7 4 5 2 9 n. 6 1 7 9 2 4 10 5 8 6 11 3
n. 7 1 1 0 6 3 11 8 7 4 2 5 9 n. 7 1 8 11 2 4 5 9 7 6 10 3
n. 8 1 10 6 11 3 8 5 2 4 7 9 n. 8 1 9 5 4 2 8 11 10 6 3 7
n. 9 2 5 6 9 1 10 8 11 4 7 3 n. 9 1 9 6 10 5 8 2 4 11 3 7
n. 10 2 5 6 9 10 1 8 11 7 4 3 n. 10 1 9 6 11 5 3 10 8 2 7 4
n. 11 2 5 6 9 11 8 10 1 4 3 7 n. 11 1 10 6 9 2 5 11 7 4 3 8
n. 12 2 11 6 3 5 8 10 7 4 9 1 n. 12 2 1 4 10 8 7 3 5 6 11 9
n. 13 2 11 6 3 7 10 8 5 4 1 9 n. 13 2 1 4 10 8 7 3 11 6 5 9
n. 14 2 11 6 3 10 7 8 5 1 4 9 n. 14 2 3 6 5 4 7 10 8 1 9 11
n. 15 3 1 4 2 11 8 9 5 6 10 7 n. 15 2 5 6 4 3 8 11 7 1 10 9
n. 16 3 1 4 2 11 10 7 9 6 8 5 n. 16 2 7 4 10 8 1 9 5 6 11 3
n. 17 3 1 10 5 2 4 7 9 6 11 8 n. 17 2 7 4 10 8 1 9 11 6 5 3
n. 18 3 2 5 1 10 11 8 9 6 7 4 n. 18 2 9 6 11 4 1 10 8 7 3 5
n. 19 3 2 11 4 1 5 8 9 6 10 7 n. 19 2 11 6 4 9 8 5 1 7 10 3
n. 20 3 4 1 2 11 7 10 9 6 8 5 n. 20 3 1 6 7 9 5 4 2 8 11 10
n. 21 3 4 7 11 8 1 10 9 6 5 2 n. 21 3 2 6 11 9 1 8 10 7 4 5
n. 22 3 4 9 10 7 11 2 1 6 8 5 n. 22 3 2 11 5 1 4 9 8 6 7 10
n. 23 3 5 2 1 10 8 11 9 6 7 4 n. 23 3 5 6 11 9 1 8 10 4 7 2
n. 24 3 5 8 10 7 2 11 9 6 4 1 n. 24 3 7 1 10 8 2 9 4 6 11 5
n. 25 3 5 8 10 7 4 9 1 6 2 11 n. 25 3 7 6 1 9 5 4 2 8 11 10
n. 26 3 7 4 11 8 10 1 9 6 5 2 n. 26 3 7 6 4 1 2 8 10 9 11 5
n. 27 3 7 10 8 5 4 1 9 6 2 11 n. 27 3 10 6 7 9 5 4 2 11 8 1
n. 28 3 8 5 10 7 11 2 9 6 4 1 n. 28 3 10 7 1 5 8 9 4 6 11 2
n. 29 3 8 9 2 11 7 10 5 6 4 1 n. 29 3 11 5 2 4 10 9 8 6 7 1
n. 30 3 8 11 7 4 5 2 9 6 1 10 n. 30 3 11 6 5 9 1 8 10 4 7 2
n. 31 3 10 1 5 2 7 4 9 6 11 8 n. 31 3 11 6 8 5 1 0 4 2 9 7 1
n. 32 3 10 7 8 5 1 4 9 6 2 11 n. 32 4 1 2 8 10 9 11 5 6 3 7
n. 33 3 11 2 4 1 8 5 9 6 10 7 n. 33 4 3 2 8 10 7 1 9 6 11 5
n. 34 3 11 8 7 4 2 5 9 6 1 10 n. 34 4 3 8 11 7 1 10 9 6 2 5
n. 35 4 1 6 3 5 8 10 7 2 11 9 n. 35 4 7 2 8 10 3 5 11 6 9 1
n. 36 4 1 6 3 8 5 10 7 11 2 9 n. 36 4 9 2 8 10 1 7 3 6 5 11
n. 37 4 1 6 3 8 9 2 11 7 10 5 n. 37 4 9 8 5 1 7 10 3 6 2 11
n. 38 4 7 6 9 8 3 2 5 1 10 11 n. 38 5 2 6 9 10 1 7 11 8 3 4
n. 39 4 7 6 9 8 11 10 1 5 2 3 n. 39 5 3 6 2 7 4 10 8 1 9 11
n. 40 4 7 6 9 11 8 10 1 2 5 3 n. 40 5 3 7 8 10 1 4 11 6 9 2
n. 41 5 2 6 3 4 7 11 8 1 10 9 n. 41 5 3 7 8 10 4 1 2 6 9 11
n. 42 5 2 6 3 7 4 11 8 10 1 9 n. 42 5 4 7 10 8 1 9 11 6 2 3
n. 43 5 2 6 7 3 4 1 10 8 1 1 9 n. 43 5 9 1 8 10 4 7 2 6 3 11
n. 44 5 8 6 1 2 11 7 10 9 4 3 n. 44 5 9 6 2 1 4 10 8 7 3 11
n. 45 5 8 6 9 7 10 11 2 4 1 3 n. 45 5 9 6 7 1 3 2 4 10 11 8
n. 46 5 8 6 9 10 7 11 2 1 4 3 n. 46 5 11 6 4 9 2 8 10 1 7 3
n. 47 5 9 8 11 2 4 1 3 6 7 10 | n. 47 5 11 6 9 1 7 10 8 2 3 4
n. 48 | 5 10 7 11 2 9 8 3 6 1 4 | n. 48 5 11 9 10 8 2 1 4 6 7 3 |

288
B. All-Interval Series

Knot 152
1 n. 1 1 3 11 7 4 5 2 8 6 9 10 1 n. 33 4 7 10 1 9 8 2 3 6 1 1 5
n. 2 1 9 6 7 3 5 8 2 4 11 10 n. 34 4 10 1 7 9 2 11 3 6 8 5
n. 3 1 9 11 8 2 4 5 10 6 7 3 n. 35 4 10 7 1 3 2 5 9 6 8 11
n. 4 2 1 8 10 4 7 9 5 6 3 11 n. 36 5 3 6 7 4 10 11 9 8 1 2
n. 5 2 3 6 4 10 7 8 5 1 9 11 n. 37 5 3 8 7 2 9 4 1 6 10 11
n. 6 2 7 8 10 4 1 3 1 1 6 9 5 n. 38 541 10 783 11 692
n. 7 2 9 6 4 10 1 8 11 7 3 5 n. 39 5 8 2 1 3 4 11 10 679
n. 8 3 2 6 11 9 1 5 8 7 10 4 n. 40 5 8 6 3 11 2971 10 4
n. 9 3 7 6 10 5 4 2 8 11 9 1 n. 41 589 10 11 4 3 1 6 7 2
n. 10 3 10 6 7 9 5 1 4 11 2 8 n. 42 589 10 11 4 3 7 6 1 2
n. 11 3 11 6 2 1 8 10 4 7 9 5 n. 43 5 9 8 1 2 3 4 7 6 10 11
n. 12 4 10 1 8 11 7 3 5 6 2 9 n. 44 5981 10 74 11 623
n. 13 4 10 7 8 5 1 9 11 6 2 3 n. 45 5 10 6 1 4 9 2 7 8 3 11
n. 14 5 3 7 11 8 1 10 4 6 9 2 n. 46 5 10 6 7 4 3 2 1 8 9 11
n. 15 5 9 6 11 3 1 4 10 8 7 2 n. 47 5 11 6 3 2 8 9 1 10 74
n. 16 5 9 7 4 10 8 1 2 6 11 3 | n. 48 5 11 641 10 7 3 8 2 9 |

Knot 153 Knot 154


1 n. 1 1 2 6 5 8 9 10 11 4 3 7 1 n. 1 1 2 6 5 8 7 11 3 4 9 10
n. 2 1 2 6 11 8 3 10 5 4 9 7 n. 2 1 2 6 5 9 8 3 7 11 4 10
n. 3 1 3 4 11 10 9 8 5 6 2 7 n. 3 1 3 7 11 10 9 2 8 6 5 4
n. 4 1 3 6 11 8 2 7 9 4 5 10 n. 4 1 4 6 3 8 9 2 7 11 5 10
n. 5 1 4 6 3 7 10 9 11 5 2 8 n. 5 1 4 11 7 3 8 9 2 6 5 10
n. 6 1 4 9 2 7 8 3 5 6 11 10 n. 6 1 8 5 2 11 7 9 4 6 3 10
n. 7 1 4 9 2 7 8 3 11 6 5 10 n. 7 1 8 6 5 2 9 10 11 7 3 4
n. 8 1 4 10 5 3 8 7 2 6 11 9 n. 8 1 8 6 5 3 11 7 2 9 10 4
n. 9 1 7 6 3 10 4 9 5 2 11 8 n. 9 1 8 7 11 5 2 9 10 6 3 4
n. 10 1 7 6 8 5 2 11 3 4 10 9 n. 10 1 9 4 3 11 7 8 2 6 5 10
n. 11 1 8 5 2 11 4 3 7 6 9 10 n. 11 1 10 6 3 8 11 7 9 2 5 4
n. 12 1 9 4 5 2 11 8 7 6 10 3 n. 12 1 10 9 2 7 11 3 8 6 5 4
n. 13 1 9 4 5 10 3 8 11 6 2 7 n. 13 2 1 6 10 4 11 7 3 8 9 5
n. 14 2 1 6 7 3 4 11 10 9 8 5 n. 14 2 1 6 10 9 4 3 11 7 8 5
n. 15 2 1 6 7 9 4 5 10 3 8 11 n. 15 2 1 7 11 10 9 4 3 6 8 5
n. 16 2 1 8 9 11 10 4 7 6 3 5 n. 16 2 3 6 8 9 11 7 10 1 4 5
n. 17 2 3 6 5 9 8 1 10 7 4 11 n. 17 2 3 8 9 1 5 4 7 6 10 11
n. 18 2 7 6 1 3 4 11 10 9 8 5 n. 18 2 3 10 7 1 5 4 11 6 8 9
n. 19 2 7 6 1 9 4 5 10 3 8 11 n. 19 2 5 6 8 1 10 9 11 7 4 3
n. 20 2 7 8 3 5 10 4 1 6 9 11 n. 20 2 7 1 5 10 3 4 9 6 8 11
n. 21 2 9 6 11 3 8 7 10 1 4 5 n. 21 2 7 6 10 3 4 9 5 1 8 11
n. 22 3 1 6 10 5 4 9 7 2 8 11 n. 22 2 7 6 10 4 5 1 9 8 3 11
n. 23 3 2 6 11 4 7 10 1 8 9 5 n. 23 2 8 1 5 9 4 3 7 6 10 11
n. 24 3 2 8 9 1 10 7 4 6 5 11 n. 24 2 8 7 11 3 4 9 1 6 10 5
n. 25 3 5 6 2 1 8 9 11 10 4 7 n. 25 2 9 6 8 3 5 1 10 7 4 11
n. 26 3 7 10 9 11 5 2 8 6 1 4 n. 26 2 9 8 3 7 11 4 1 6 10 5
n. 27 3 10 4 9 5 2 11 8 6 1 7 n. 27 2 9 10 1 7 11 4 5 6 8 3
n. 28 3 10 6 7 8 11 2 5 4 9 1 n. 28 2 11 6 8 7 10 3 5 1 4 9
n. 29 3 11 2 9 7 1 10 4 6 5 8 n. 29 3 2 6 5 4 1 10 7 11 9 8
n. 30 4 1 6 8 2 5 11 9 10 7 3 n. 30 3 4 6 1 8 7 11 5 2 9 10
n. 31 4 1 10 7 3 8 2 9 6 5 11 n. 31 3 4 7 11 9 10 1 8 6 5 2
n. 32 4 7 6 8 2 11 53 10 19 | n. 32 3 4 9 10 11 7 1 2 6 5 8 |

289
B. All-Interval Series

1 n. 33 3 8 6 5 4 11 7 1 10 9 2 1 n. 20 4 3 7 9 11 10 5 8 6 2 1
n. 34 3 8 9 2 7 11 5 10 6 1 4 n. 21 4 5 6 8 3 11 9 7 2 1 10
n. 35 3 8 11 7 9 2 5 4 6 1 10 n. 22 4 7 2 1 3 5 9 8 6 11 10
n. 36 3 10 6 1 8 5 2 11 7 9 4 n. 23 4 7 3 1 2 5 10 11 6 8 9
n. 37 4 1 6 10 5 11 7 2 9 8 3 n. 24 4 9 1 3 5 10 11 2 6 8 7
n. 38 4 3 1 5 2 11 8 7 6 10 9 n. 25 4 9 1 3 5 10 11 8 6 2 7
n. 39 4 3 6 10 9 2 5 11 7 8 1 n. 26 4 9 6 8 11 3 5 10 1 2 7
n. 40 4 3 7 11 10 9 2 5 6 8 1 n. 27 4 11 6 8 9 5 3 1 2 7 10
n. 41 4 5 2 9 7 11 8 3 6 10 1 n. 28 5 4 6 10 1 2 7 9 11 3 8
n. 42 4 5 6 8 2 9 10 11 7 3 1 n. 29 5 8 1 7 10 4 9 2 6 11 3
n. 43 4 5 6 8 3 11 7 2 9 10 1 n. 30 5 9 6 11 8 7 1 10 4 3 2
n. 44 4 7 6 10 11 5 1 2 3 8 9 n. 31 5 10 6 4 1 2 7 9 11 3 8
n. 45 4 9 1 5 10 3 2 11 6 8 7 n. 32 5 10 11 2 7 9 1 4 6 3 8 |
n. 46 4 9 6 10 3 2 11 5 1 8 7
n. 47 4 9 7 11 2 5 8 1 6 10 3
n. 48 4 10 3 2 1 5 9 11 6 8 7
n. 49 4 10 9 2 7 11 3 5 6 8 1
n. 50
n. 51
4 11 2 3 1 5 8 9 6 10 7
4 11 6 8 2 3 10 5 1 9 7
Knot 172
n. 52 4 11 6 8 9 5 1 2 3 10 7 1 n. 1 1 2 8 10 11 9 5 6 3 7 4
n. 53 5 2 6 3 4 7 11 9 10 1 8 n. 2 1 4 5 6 3 8 11 7 2 9 10
n. 54 5 2 9 10 11 7 3 4 6 1 8 n. 3 1 8 5 6 2 9 10 11 7 4 3
n. 55 5 3 11 7 2 9 10 4 6 1 8 n. 4 1 9 7 2 4 10 5 6 8 11 3
n. 56 5 4 1 10 7 11 9 8 6 3 2 n. 5 2 3 10 5 1 4 9 6 7 8 11
n. 57 5 4 6 1 3 7 11 10 9 2 8 n. 6 2 3 10 5 1 7 9 6 4 11 8
n. 58 5 4 6 1 10 9 2 7 11 3 8 n. 7 2 5 4 6 3 1 7 11 10 9 8
n. 59 5 4 11 7 1 10 9 2 6 3 8 n. 8 2 9 10 11 7 1 3 6 4 5 8
n. 60 5 8 6 3 4 9 10 11 7 1 2 n. 9 2 9 10 11 7 4 3 6 1 8 5
n. 61 5 8 7 11 3 4 9 10 6 1 2 n. 10 2 11 4 6 9 7 1 5 10 3 8
n. 62 5 9 8 3 7 11 4 10 6 1 2 n. 11 3 1 7 11 10 9 2 6 8 5 4
n. 63 5 10 6 1 4 11 7 3 8 9 2 n. 12 3 1 7 11 10 9 8 6 2 5 4
n. 64 5 10 6 1 9 4 3 11 7 8 2 | n. 13 3 4 7 11 10 9 2 6 5 8 1
n. 14 3 5 11 7 2 9 4 6 10 1 8
n. 15 3 5 11 7 2 9 10 6 4 1 8
n. 16 3 7 4 6 1 2 8 10 11 9 5
Knot 159 n. 17 3 8 11 7 2 9 10 6 1 4 5
n. 18 3 11 8 6 5 10 4 2 7 9 1
1 n. 1 1 2 6 8 5 10 11 9 7 3 4
n. 19 4 1 8 6 3 5 11 7 2 9 10
n. 2 1 2 7 10 11 9 5 8 6 3 4
n. 20 4 1 9 6 11 3 5 10 8 2 7
n. 3 1 4 5 11 2 8 9 10 6 7 3
n. 21 4 3 2 1 5 11 9 6 8 7 10
n. 4 1 8 6 2 5 10 11 9 7 3 4
n. 22 4 5 2 6 8 9 10 11 7 1 3
n. 5 1 9 6 7 4 11 5 2 8 3 10
n. 23 4 5 8 6 2 9 10 11 7 1 3
n. 6 2 1 6 4 3 7 9 11 10 5 8
n. 24 4 7 3 6 5 9 11 10 8 2 1
n. 7 2 3 4 10 1 7 8 11 6 9 5
n. 25 4 7 8 6 9 11 5 1 2 3 10
n. 8 2 5 10 11 9 7 3 4 6 1 8
n. 26 4 9 2 7 11 5 3 6 8 1 10
n. 9 2 7 6 4 9 1 3 5 10 11 8
n. 27 4 11 2 6 8 3 10 5 1 7 9
n. 10 2 9 4 10 7 1 8 5 6 3 11
n. 28 4 11 8 6 2 3 10 5 1 7 9
n. 11 2 11 10 5 3 1 9 4 6 7 8
n. 29 5 4 1 6 10 9 2 7 11 8 3
n. 12 3 4 6 1 2 7 10 11 9 5 8
n. 30 5 8 1 6 3 4 7 11 10 9 2
n. 13 3 7 6 10 9 8 2 11 5 4 1
n. 31 5 9 11 10 8 2 1 6 4 7 3
n. 14 3 8 6 5 10 11 2 7 9 1 4
n. 32 5 10 4 2 7 9 1 6 3 11 8 |
n. 15 3 11 6 2 9 4 10 7 1 8 5
n. 16 4 1 2 7 9 11 3 8 6 5 10
n. 17 4 1 9 7 2 11 10 5 6 8 3
n. 18 4 3 6 8 5 9 11 10 7 2 1
n. 19 4 3 7 9 11 10 5 2 6 8 1 |

290
B. All-Interval Series

Knot 173 1 n. 15 2 11 4 3 1 10 9 6 5 8 7
n. 16 2 11 10 9 7 4 3 6 5 8 1
1 n. 1 1 3 4 11 7 9 10 6 2 5 8 n. 17 32 11 981 10 6 5 4 7
n. 2 1 9 10 11 7 3 4 6 8 5 2 n. 18 3 4 7 9 10 11 2 6 1 8 5
n. 3 2 3 5 1 8 9 11 6 4 7 10 n. 19 38 11 927 10 6541
n. 4 2 3 8 6 4 5 11 7 10 1 9 n. 20 3 10 7 9 4 5 2 6 1 8 11
n. 5 2 5 4 6 8 3 11 7 10 1 9 n. 21 4 3 7 9 11 10 1 6 2 8 5
n. 6 2 5 8 6 1 3 4 11 7 9 10 n. 22 4 9 1 3 5 10 7 6 2 8 11
n. 7 2 5 8 6 4 3 7 11 10 9 1 n. 23 4 10 1629 11 3785
n. 8 2 9 8 6 4 11 5 1 10 7 3 n. 24 4 10 7 6 2 3 5 9 1 8 11
n. 9 2 9 11 7 8 3 5 6 4 1 10 n. 25 5 2 7 9 11 3 8 6 1 4 10
n. 10 2 11 4 6 8 9 5 1 10 7 3 n. 26 5 2 8 6 1 4 11 379 10
n. 11 2 11 8 6 4 9 1 5 10 3 7 n. 27 5 4 1 6 3 8 11 927 10
n. 12 2 11 8 6 7 9 4 5 1 3 10 n. 28 5 4 7 6 3 2 11 981 10
n. 13 3 7 10 1 5 9 8 6 4 11 2 n. 29 5 8 1 6 2 11 10 9 7 4 3
n. 14 3 7 10 1 5 11 4 6 8 9 2 n. 30 5 8 2 6 1 10 11 9 7 3 4
n. 15 3 11 2 5 1 7 8 6 4 9 10 n. 31 5 8 7 3 11 9261 10 4
n. 16 3 11 2 5 1 9 4 6 8 7 10 n. 32 5 8 7 6 2 11 431 10 9 |
n. 17 4 1 10 6 2 9 11 7 8 3 5
n. 18 4 1 10 6 5 9 2 7 11 3 8
n. 19 4 3 7 11 2 5 9 6 10 1 8
n. 20
n. 21
4 3 7 11 10 9 1 6 2 5 8
4 3 10 6 9 5 2 1 1 7 1 8
Knot 183
n. 22 4 5 2 6 9 1 10 7 11 3 8 n. 1 1 4 11 3 7 8 5 6 2 9 10
n. 23 4 5 11 7 10 1 9 6 2 3 8 n. 2 1 9 5 6 2 8 7 3 11 4 10
n. 24 4 7 10 6 2 3 5 1 8 9 11 n. 3 2 3 10 6 7 4 5 9 1 8 11
n. 25 4 7 10 6 11 3 2 1 5 9 8 n. 4 2 8 1 9 5 4 10 6 7 3 11
n. 26 4 9 1 5 2 11 3 6 10 7 8 n. 5 2 8 7 3 11 4 10 6 1 9 5
n. 27 4 9 1 5 10 3 7 6 2 11 8 n. 6 2 9 10 6 1 4 11 3 7 8 5
n. 28 4 9 10 6 3 11 2 5 1 7 8 n. 7 5 8 7 3 11 4 1 6 10 9 2
n. 29 4 11 2 6 3 7 10 1 5 9 8 n. 8 5 9 1 6 10 4 11 3 7 8 2
n. 30 4 11 5 1 10 7 3 6 2 9 8
n. 31 5 3 8 7 11 9 2 6 10 1 4
n. 32 5 9 2 7 11 3 8 6 4 1 10 |

Knot 184
1 n. 1 134 11 2 5 9 6 10 78
n. 2 1 3 5 6 8 11 10 9 2 7 4
n. 3 13 10 52 11 9 6 4 7 8
Knot 179 n. 4 1729 10 11 5 6 8 3 4
n. 5 1 7 9 6 4 11 25 10 38
1 n. 1456 10 729 11 83
1 n. 6 1796 10 52 11 438
n. 14 10 6 5 2 7 9 1138
2 n. 7 21 10 7 8 3 4 6 5 11 9
n. 14 11 379 10 6 5 2 8
3 n. 8 2 5 4 6 11 981 10 73
n. 14 11 6 10 7 8 3 5 2 9
4 n. 9 27 10 1 8 9 4 6 11 53
n. 1 8 5 6 3 4 7 9 10 11 2
5 n. 10 2 11 4 6 5 3 8 7 10 19
n. 18 11 63 10 7 9 4 5 2
6 n. 11 3 1 7 6 8 9 4 5 2 11 10
n. 1 10 4 6 5 8 7 3 11 92
7 n. 12 3 1 7 6 8 9 10 11 254
n. 1 10 11 9 7 3 4 6 5 8 2
8 n. 13 31 10 729 11 6 4 5 8
n. 2 3 5 9 1 8 11 64 10 7
9 n. 14 352 11 10 9 7 6 8 1 4
n. 2 5 4 9 7 10 36 11 81
10 n. 15 35 11 6 4 9 2 7 10 18
n. 25 10 3 1 4 9 6 11 87
11 n. 16 35 11 6 4 9 8 1 10 72
n. 2 8 5 6 4 3 7 9 11 10 1
12 n. 17 37 10 189 11 6 4 5 2
n. 28 11 6 4 9 1 3 5 10 7
13 n. 18 3 11 2 5 4 9 7 6 8 1 10
[n. 14 29 11 3 7 8 5 6 4 10 1 n. 19 4 1 2 3 10 586 11 97 |

291
B. All-Interval Series

1 n. 20 4 1 8 6 7 9 10 11 2 5 3 1 n. 11 2 3 5 11 6 4 1 8 9 10 7
n. 21 4 3 2 1 10 7 8 6 9 11 5 n. 12 2 5 103 8 11 6 4 9 1 7
n. 22 4 3 8 6 5 11 10 9 2 7 1 n. 13 2 7 4 3 6 5 8 9 11 10 1
n. 23 4 3 8 7 10 1 2 6 9 11 5 n. 14 2 7 10 3 6 11 8 9 5 4 1
n. 24 4 5 2 6 3 7 10 1 8 9 11 n. 15 2 9 11 5 6 4 7 8 3 10 1
n. 25 4 5 2 11 10 9 8 6 7 1 3 n. 16 2 11 10 9 8 5 6 4 3 7 1
n. 26 4 5 8 6 3 1 10 7 2 9 11 n. 17 3 2 11 10 6 5 8 1 9 4 7
n. 27 4 7 2 9 10 11 8 6 5 3 1 n. 18 3 4 7 2 6 1 10 11 9 8 5
n. 28 4 7 8 6 1 3 10 5 2 11 9 n. 19 3 8 11 10 6 5 2 7 9 4 1
n. 29 4 9 2 7 10 1 8 6 3 5 11 n. 20 3 10 7 2 6 1 4 5 9 8 11
n. 30 4 9 8 1 10 7 2 6 3 5 11 n. 21 4 1 8 9 10 7 6 2 3 5 11
n. 31 4 9 8 6 11 5 10 3 2 1 7 n. 22 4 3 7 1 6 2 11 10 9 8 5
n. 32 4 11 2 5 10 3 8 6 1 7 9 n. 23 4 7 8 3 10 1 6 2 9 11 5
n. 33 4 11 2 6 9 1 10 7 8 3 5 n. 24 4 9 1 7 6 2 5 10 3 8 11
n. 34 4 11 8 6 9 7 10 1 2 3 5 n. 25 5 2 3 4 11 8 6 1 7 9 10
n. 35 5 3 1 6 4 7 2 9 10 11 8 n. 26 5 2 7 9 4 1 6 3 8 11 10
n. 36 5 3 2 1 10 7 9 6 8 11 4 n. 27 5 8 1 9 4 7 6 3 2 11 10
n. 37 5 3 8 7 10 1 9 6 2 11 4 n. 28 5 8 3 11 4 7 6 2 1 10 9
n. 38 5 1 1 9 6 2 1 10 7 8 3 4 n. 29 5 8 9 10 11 2 6 1 7 3 4
n. 39 5 11 9 6 8 7 10 1 2 3 4 n. 30 5 8 9 11 10 1 6 2 7 4 3
n. 40 5 11 10 9 2 7 1 6 4 3 8 | n. 31 5 11 3 8 6 1 4 9 2 7 10
n. 32 5 11 9 2 6 1 10 3 8 7 4 |

Knot 186
1 n. 1 1 8 7 3 10 6 9 2 5 11 4 Knot 207
n. 2 1 9 7 2 8 6 11 3 5 10 4
1 n. 1 1 2 8 5 6 9 7 3 4 11 10
n. 3 1 1 0 5 3 8 6 11 2 7 9 4
n. 2 1 4 9 7 2 11 6 3 8 5 10
n. 4 2 3 11 4 5 6 8 7 1 10 9
n. 3 1 4 10 5 6 2 7 8 3 11 9
n. 5 2 9 5 4 11 6 8 1 7 10 3
n. 4 1 10 5 3 8 11 6 2 7 4 9
n. 6 3 2 11 5 4 6 7 8 1 9 10
n. 5 2 1 4 3 6 7 10 11 9 8 5
n. 7 3 10 7 1 8 6 11 4 5 9 2
n. 6 2 1 8 9 5 3 6 7 4 10 11
n. 8 4 3 1 2 5 6 8 9 11 10 7
n. 7 2 7 4 9 6 1 10 5 3 8 11
n. 9 4 5 11 2 3 6 10 9 1 8 7
n. 8 2 7 8 3 11 9 6 1 4 10 5
n. 10 4 9 7 2 11 6 8 3 5 10 1
n. 9 3 1 9 4 5 10 6 7 2 8 11
n. 11 4 10 5 3 11 6 8 2 7 9 1
n. 10 3 4 1 2 6 5 8 9 11 10 7
n. 12 4 10 11 9 5 6 8 2 1 3 7
n. 11 3 5 9 8 1 2 6 11 10 4 7
n. 13 4 11 5 2 9 6 10 3 7 8 1
n. 12 3 8 5 10 6 1 4 9 7 2 11
n. 14 5 2 1 3 4 6 7 10 11 9 8
n. 13 5 2 1 3 4 7 6 10 11 8 9
n. 15 54 11 3 2 6 9 10 178
n. 14 5 8 2 1 6 10 11 4 3 7 9
n. 16 59 11 10 4 6 7 3 1 2 8 |
n. 15 5 8 9 11 10 7 6 3 4 1 2
n. 16 5 10 4 1 6 9 11 3 8 7 2 |

Knot 206
n. 1 1 4 3 7 8 11 6 10 5 2 9 Knot 208
n. 2 1 4 5 9 8 11 63 10 72
n. 3 1 4 9 2 7 10 65 11 38 n. 1 172 11 6 8 5 3 10 94
n. 4 1 4 9 7 2 5 6 10 11 83 n. 2 1 7 9 2 3 11 685 10 4
n. 5 1 7 3 4 6 5 8 9 10 11 2 n. 3 1 8 5 3 6 4 7 10 11 92
n. 6 179 10 6 5 2 3 4 11 8 n. 4 18 11 36 10 7 4 5 9 2
n. 7 1 10 3 8 7 4 6 5 11 92 n. 5 1 9 5 4 7 3 6 10 11 82
n. 8 1 10 11 9 8 5 6 3 4 7 2 n. 6 19 10 35 1168274
n. 9 2 1 4 9 6 11 835 10 7 n. 7 19 11 10 7 3 6 4 5 8 2
n. 10 21 10 9 6 5 8 3 11 47 n. 8 110 5 11 6 8 3 2 9 7 4

292
B. All-Interval Series

1 n. 9 2 1 4 10 6 11 3 7 8 5 9 1 n. 11 2 7 4 3 11 8 6 59 1 10
n. 10 2 3 5 10 1 4 6 9 11 8 7 n. 12 2 7 4 9 5 8 6 11 3 1 10
n. 11 2 3 11 41 10 6 9 5 8 7 n. 13 2 11 3 7 6 4 1 9 8 5 10
n. 12 2 5 8 7 3 10 6 11 4 1 9 n. 14 2 11 9 1 6 4 7 3 8 5 10
n. 13 2 7 4 1 0 6 5 9 1 8 11 3 n. 15 3 7 1 8 6 4 9 2 5 11 10
n. 14 2 8 5 4 6 3 7 10 11 9 1 n. 16 3 7 9 4 6 8 1 2 5 11 10
n. 15 2 8 5 1069 14 11 3 7 n. 17 3 11 5 4 6 8 9 10 1 7 2
n. 16 28 11 4 6 9 1 10 5 3 7 n. 18 3 11 9 8 6 4 5 10 1 7 2
n. 17 2 8 11 10 6 3 7 4 5 9 1 n. 19 4 1 9 8 5 10 6 2 11 3 7
n. 18 2 9 5 4 7 10 6 3 11 81 n. 20 4 1 9 8 11 10 6 2 5 3 7
n. 19 29 11 10 7 4 6 3 5 8 1 n. 21 4 3 1 9 6 10 5 11 2 7 8
n. 20 2 11 8 1 9 10 6 5 4 7 3 n. 22 4 3 2 11 5 10 6 9 1 7 8
n. 21 3 1 4 5 6 10 9 7 2 11 8 n. 23 4 5 10 1 7 2 6 3 11 9 8
n. 22 3 5 8 1 6 2 9 11 10 7 4 n. 24 4 5 11 3 6 2 7 1 10 9 8
n. 23 3 7 4 5 6 10 9 1 8 11 2 n. 25 4 7 3 8 5 10 6 2 11 91
n. 24 3 7 4 5 9 1 6 2 8 11 10 n. 26 4 7 3 8 11 10 6 2 5 9 1
n. 25 3 7 10 11 9 1 6 2 8 5 4 n. 27 4 9 2 5 11 10 6 3 7 1 8
n. 26 3 11 2 7 9 5 6 10 4 1 8 n. 28 4 9 7 3 6 10 11 5 2 1 8
n. 27 3 11 8 1 6 2 9 5 4 7 10 n. 29 4 11 5 9 6 2 1 7 10 3 8
n. 28 3 11 8 1 9 5 6 10 4 7 2 n. 30 4 11 10 7 1 2 6 9 5 3 8
n. 29 4 1 2 8 6 5 11 9 10 3 7 n. 31 5 9 1 10 6 2 7 4 3 11 8
n. 30 4 1 3 2 9 8 6 5 11 10 7 n. 32 5 9 7 10 6 2 1 4 3 11 8 |
n. 31 41 10 5 3 2 6 7 8 11 9
n. 32 4 3 10 9 11 8 6 5 2 1 7
n. 33 4 5 8 2 6 1 9 11 10 73
n. 34 4 7 2 8 6 11 53 10 91 Knot 213
n. 35 4 7 9 2 3 8 6 11 5 10 1
n. 36 4 7 10 11 9 2 6 1 8 5 3 1 n. 1 1 4 9 6 11 3 5 8 10 7 2
n. 37 4 9 10 3 5 8 6 11 2 7 1 n. 2 1 9 7 4 2 5 10 6 11 8 3
n. 38 4 10 5 8 6 1 1 3 2 9 7 1 n. 3 1 9 10 6 5 2 8 1 1 7 4 3
n. 39 4 10 11 8 6 5 9 2 3 1 7 n. 4 1 10 4 7 11 8 3 6 5 9 2
n. 40 4 11 8 2 6 7 3 5 10 19 n. 5 2 1 10 8 11 9 5 6 3 4 7
n. 41 5 2 1 7 6 4 3 10 9 11 8 n. 6 2 3 11 6 9 8 5 1 4 10 7
n. 42 5 4 1 3 6 8 11 2 7 9 10 n. 7 2 7 10 8 5 3 11 6 9 4 1
n. 43 5 4 7 3 6 2 11 8 1 9 10 n. 8 2 9 5 6 3 8 11 7 4 10 1
n. 44 5 9 1 8 11 3 6 2 7 4 10 n. 9 3 4 7 6 2 1 10 8 11 9 5
n. 45 5 9 2 3 1 7 6 4 10 11 8 n. 10 3 4 7 11 8 2 5 6 10 9 1
n. 46 5 9 7 2 11 3 6 8 1 4 10 n. 11 3 8 11 6 10 5 2 4 7 9 1
n. 47 5 11 9 10 3 7 6 4 1 2 8 n. 12 3 8 11 7 4 10 1 6 2 9 5
n. 48 5 11 10 7 6 4 1 3 2 9 8 | n. 13 5 2 8 11 7 4 3 6 1 9 10
n. 14 5 8 9 6 7 3 1 4 2 11 10
n. 15 5 9 2 6 1 10 4 7 11 8 3
n. 16 59 11 8 10 1 2 6 7 4 3 |

Knot 210
n. 1 1 9 5 2 6 10 11 8 3 7 4 Knot 214
n. 2 1 9 11 2 6 10 5 8 3 7 4
n. 3 2 1 4 3 11 8 6 5 9 7 10 n. 1 1 2 7 6 4 3 10 8 9 5 11
n. 4 2 1 4 9 5 8 6 11 3 7 10 n. 2 1 2 7 6 5 11 3 8 10 9 4
n. 5 2 1 7 10 3 8 6 4 11 5 9 n. 3 1 4 9 6 2 5 8 10 7 3 11
n. 6 2 1 7 10 11 4 6 8 3 5 9 n. 4 1 7 3 4 2 9 8 6 5 10 11
n. 7 2 5 3 7 6 4 1 9 8 11 10 n. 5 1 7 3 10 8 9 2 6 5 4 11
n. 8 2 5 9 1 6 4 7 3 8 11 10 n. 6 1 8 7 6 5 11 3 2 4 9 10
n. 9 2 7 1 10 5 4 6 8 9 11 3 n. 7 1 8 7 6 10 3 4 2 9 5 11
n. 10 2 7 1 10 9 8 6 4 5 1 1 3 n. 8 1 9 5 2 4 7 10 6 3 8 11

293
B. All-Interval Series

1 n. 9 1 9 5 8 10 7 4 6 3 2 11 1 n. 8 173 10 8 2 9 6 5 11 4
n. 10 1 109 68 5 24 73 11 n. 9 1 7 8 6 5 11 324 10 9
n. 11 238 10 9 1 7 6 5 4 11 n. 10 195 11 7 10 4 6 3 8 2
n. 12 258 10 73 11 6 1 4 9 n. 11 1976 10 524 11 83
n. 13 298 10 3 7 1 6 1145 n. 12 1 9 11 6 2 3 8 10 5 4 7
n. 14 2 1L 8 10 1 9 5 6 7 4 3 n. 13 1 9 11 8 10 4 7 6 3 5 2
n. 15 32 11 6 1 9 5 8 10 74 n. 14 21 10 8 7 4 3 6 5 9 11
n. 16 3 4 7 6 5 9 1 10 8 11 2 n. 15 238 10 5 4 1 6 7 9 11
n. 17 38 11 6 1 9 5 2 4 7 10 n. 16 238 10 5 4 7 6 1 9 11
n. 18 3 107659 142 11 8 n. 17 2 5 3 6 7 4 10 8 11 91
n. 19 41 10 8 11 3 7 6 5 2 9 n. 18 27 10 8 1 4 9 6 11 35
n. 20 43 10 895 11 6 1 2 7 n. 19 2 8 3 6 4 10 7 11 591
n. 21 47 10 8 5 9 1 6 11 23 n. 20 2 8 9 6 4 10 15 1137
n. 22 49 10 83 11 5 6 7 2 1 n. 21 298 10 11 4 1 6 7 3 5
n. 23 5 2 9 6 4 1 10 8 11 37 n. 22 298 10 11 4 7 6 1 3 5
n. 24 54 11 6 1 7 3 10 892 n. 23 2 11 9 6 1 4 10 8 5 3 7
n. 25 54 11 6 2 3 8 10 9 1 7 n. 24 31 10 6 5 9 7 4 2 8 11
n. 26 5 8 9 6 10 142 11 37 n. 25 328 10 9 1 7 6 4 5 11
n. 27 5 9 1 4 2 11 863 10 7 n. 26 3 4 7 8 10 126 11 95
n. 28 591 10 8 11 2 6 3 4 7 n. 27 34 10 6 5 9 1 7 11 28
n. 29 5 10 11 6 1 7 3 4 2 9 8 n. 28 3 5 2 6 1 9 11 8 10 47
n. 30 5 10 11 6 8 3 2 4 9 1 7 n. 29 3 8 2 6 1 9 5 11 7 10 4
n. 31 5 11 3 2 4 9 10 6 1 8 7 n. 30 38 11 425 10 6 7 9 1
n. 32 5 11 38 10 9 4 6 1 2 7 | n. 31 3 10 4 2 9 5 11 6 8 1 7
n. 32 45 11 6 3 2 8 10 9 1 7
n. 33 4 10 15 11 3 7 6 2 8 9
Knot 220 n. 34
n. 35
4 10 7 11 5 9 1 6 2 8 3
4 11 5 6 9 2 8 10 371
1 n. 1 1 2 8 6 9 5 3 11 7 4 10 n. 36 5 3 1 6 7 4 11 10 8 9 2
n. 2 1 8 2 4 11 5 9 6 7 3 10 n. 37 5 3 1 6 10 9 4 2 7 8 11
n. 3 19 10 6 7 8 2 4 5 11 3 n. 38 5 3 7 6 1 4 11 10 892
n. 4 23 11 6 9 1 7 8 10 45 n. 39 53 11 6 9 4 1 8 10 72
n. 5 2 8 5 1 9 7 3 6 4 10 11 n. 40 5 8 1 2 4 3 10 679 11
n. 6 28 11 7 3 1 9 6 4 10 5 n. 41 5 8 2 4 1 3 11 6 10 79
n. 7 2 9 5 6 3 7 1 8 10 4 11 n. 42 5 8 7 2 4 9 10 613 11
n.8 3 7 1 8 10 4 11 6 2 9 5 n. 43 5 9 1 7 11 2 8 6 3 4 10
n. 9 3 7 9 1 5 8 2 6 11 10 4 n. 44 5 9 7 4 2 8 11631 10
n. 10 3 11 5 4 2 8 7 6 10 91 n. 45 5976 10 3 4 2 1 8 11
n. 11 3 11 9 5 1 4 10 6 7 2 8 n. 46 59 11621 10 8 7 4 3
n. 12 4 10 5 6 2 8 11 7 3 1 9 n. 47 5 11 324 10 9 6 1 7 8
n. 13 4 10 11 6 2 8 5 1 9 7 3 n. 48 5 11 4 6 1 7 3 10 829 |
n. 14 54 10 8 7 1 9 6 1132
n. 15 5 9 2 6 11 4 10 8 1 7 3
n. 16 5 10 4 6 9 1 3 7 11 82 |
Knot 241
Knot 221 n. 1 1 2 5 9 6 8 3 11 7 10 4
n. 2 127 11 5 3 6 8 9 4 10
n. 1 1 3 5 6 2 9 8 10 11 47 n. 3 13 11 7 10 9 6 8 2 5 4
n. 2 1 3 5 6 11 8 7 2 4 9 10 n. 4 1 9 4 3 6 8 2 7 11 5 10
n. 3 1 3 7 6 9 8 5 4 2 11 10 n. 5 217 11 10 4 6 3 8 9 5
n. 4 13 11 6 5 8 7 2 4 9 10 n. 6 2 7 1 5 10 4 6 9 8 3 11
n. 5 145 10 8 3 2 6 1197 n. 7 2 8 3 4 6 9 7 1 5 10 11
n. 6 14 10 8 5 3 7 6 2 11 9 n. 8 2 8 9 4 6 3 1 7 11 10 5
n. 7 14 11 10 8 9 2 6 5 3 7 n. 9 317 11 10 5 6 2 8 9 4

294
B. All-Interval Series

1 n. 10 3 2 5 1 9 11 6 8 7 10 4 1 n. 9 1 9 10 8 7 4 6 5 3 2 11
n. 11 3 4 9 1 6 10 5 11 7 2 8 n. 10 1 10 9 7 6 8 5 4 2 3 11
n. 12 3 5 11 7 2 1 6 10 4 9 8 n. 11 2 5 10 8 9 11 6 1 4 3 7
n. 13 3 7 10 11 6 8 2 5 1 9 4 n. 12 2 11 10 8 3 5 6 7 4 9 1
n. 14 3 8 9 5 6 2 1 7 11 10 4 n. 13 3 1 9 7 6 5 10 4 2 11 8
n. 15 3 10 1 5 9 7 6 4 11 2 8 n. 14 3 1 10 5 6 7 9 4 2 11 8
n. 16 3 H 2 7 6 4 10 1 5 9 8 n. 15 3 5 2 1 6 11 9 8 10 7 4
n. 17 4 3 7 11 2 8 6 5 10 1 9 n. 16 3 5 9 11 6 1 2 8 10 7 4
n. 18 4 3 8 2 6 11 10 5 1 7 9 n. 17 4 1 8 10 3 7 6 5 2 9 11
n. 19 4 5 2 8 6 9 10 7 11 3 1 n. 18 4 1 10 8 2 7 6 5 3 11 9
n. 20 4 9 1 5 2 8 6 11 10 7 3 n. 19 4 1 10 8 3 5 6 7 2 11 9
n. 21 4 9 8 2 6 5 10 11 7 1 3 n. 20 4 7 8 10 9 1 6 11 2 3 5
n. 22 4 10 1 5 9 8 6 3 11 2 7 n. 21 4 7 10 8 2 1 6 11 9 5 3
n. 23 4 10 1 8 6 5 3 7 11 2 9 n. 22 4 7 10 8 9 11 6 1 2 5 3
n. 24 4 10 5 1 7 2 6 11 3 8 9 n. 23 5 2 9 11 6 4 1 8 10 3 7
n. 25 4 10 7 8 6 11 9 1 5 2 3 n. 24 5 3 2 11 6 1 9 10 8 7 4
n. 26 4 10 7 11 3 8 6 9 5 2 1 n. 25 5 3 8 10 1 4 6 9 11 2 7
n. 27 4 10 11 7 1 2 6 5 9 8 3 n. 26 5 3 8 10 11 2 6 1 9 4 7
n. 28 4 11 2 8 6 3 10 1 5 9 7 n. 27 5 3 11 9 6 4 1 10 8 2 7
n. 29 5 3 7 11 2 9 6 4 10 1 8 n. 28 5 8 3 11 6 10 1 2 4 9 7
n. 30 5 9 8 3 6 4 10 11 7 1 2 n. 29 5 9 2 4 11 8 6 1 3 10 7
n. 31 5 10 1 9 6 4 3 7 11 2 8 n. 30 5 9 8 11 6 1 3 4 2 7 10
n. 32 5 10 11 7 1 3 6 4 9 8 2 | n. 31 5 10 1 3 6 8 11 2 4 9 7
n. 32 5 10 4 2 11 8 6 3 1 9 7 |

Knot 242
1 n. 1 1 2 11 7 8 5 6 3 4 9 10
n. 2 1 3 11 5 6 8 7 2 4 10 9
n. 3 1 4 11 7 10 5 6 2 9 8 3
n. 4 1 7 3 5 6 9 2 8 10 11 4
Knot 252
n. 5 2 3 8 9 6 7 4 5 1 10 11 1 n. 1 1 4 9 7 2 8 3 6 11 5 10
n. 6 2 9 8 3 6 1 4 11 7 10 5 n. 2 1 7 2 6 11 8 3 5 10 4 9
n. 7 3 2 8 10 5 4 6 7 1 9 11 n. 3 2 1 4 6 3 7 10 11 9 5 8
n. 8 3 4 9 10 6 1 2 11 7 8 5 n. 4 2 1 7 6 3 4 10 11 9 8 5
n. 9 3 8 9 2 6 5 10 7 11 4 1 n. 5 2 5 9 11 10 7 3 6 4 1 8
n. 10 3 10 4 2 1 8 6 11 5 9 7 n. 6 2 7 1 6 9 4 10 5 3 8 11
n. 11 4 5 10 8 2 3 6 11 9 1 7 n. 7 2 7 4 6 9 1 10 5 3 11 8
n. 12 4 11 10 8 2 9 6 5 3 7 1 n. 8 2 11 3 5 10 1 9 6 4 7 8
n. 13 5 3 7 1 6 4 11 10 8 2 9 n. 9 3 4 10 11 9 8 5 6 2 1 7
n. 14 5 8 7 1 1 2 1 6 10 9 4 3 3 7 10 11 9 5 2 6 8 1 4
n. 10
n. 15 5 10 7 11 4 1 6 3 8 9 2 n. 11 3 7 10 11 9 5 8 6 2 1 4
n. 16 5 11 3 1 6 9 10 4 2 7 8 | n. 12 3 8 2 7 9 4 1 6 10 5 11
n. 13 3 11 2 7 9 1 4 6 10 5 8
n. 14 3 11 2 7 9 1 10 6 4 5 8
Knot 243 n. 15 4 1 2 6 8 5 9 11 10 7 3
n. 16 4 1 8 6 2 5 9 11 10 7 3
n. 1 1 2 5 3 6 4 7 10 8 9 11 n. 17 4 1 9 7 2 11 3 6 8 5 10
n. 2 1 2 8 10 7 4 6 3 5 9 11 n. 18 4 5 8 6 3 11 2 7 9 1 10
n. 3 1 3 4 2 5 8 6 9 7 10 11 n. 19 4 7 2 6 8 11 3 5 10 1 9
n. 4 1 3 4 2 7 10 6 5 9 8 11 n. 20 4 7 3 1 2 5 9 6 8 11 10
n. 5 1 3 7 9 6 8 5 2 4 10 11 n. 21 4 7 8 6 2 11 3 5 10 1 9
n. 6 1 3 10 7 6 5 9 2 4 11 8 n. 22 4 11 8 6 9 5 2 1 3 7 10
n. 7 1 4 3 7 6 2 5 10 8 9 11 n. 23 5 8 9 11 10 4 3 6 7 1 2
n. 8 1 9 4 7 6 5 3 8 10 11 2 n. 24 5 11 10 6 7 4 3 1 2 8 9 |

295
B. All-Interval Series

Knot 271 Knot 273


1 n. 1 1 2 8 3 5 9 6 7 4 11 10 n. 1 1 2 5 8 3 10 6 11 4 7 9
n. 2 1 4 5 10 6 7 2 8 9 11 3 n. 2 1 8 5 3 6 11 10 7 4 9 2
n. 3 1 9 5 8 6 3 11 2 7 10 4 n. 3 2 3 4 1 10 5 6 9 11 8 7
n. 4 1 9 73 11 2 68 5 10 4 n. 4 2 9 4 7 10 11 6 3 5 8 1
n. 5 1 9 7 3 11 86 2 5 10 4 n. 5 3 1 4 5 6 10 9 8 11 2 7
n. 6 2 1 8 5 6 3 7 9 4 10 11 n. 6 3 5 8 1 6 2 9 4 7 10 11
n. 7 2 5 9 1 3 7 6 4 10 11 8 n. 7 5 4 1 3 6 7 2 11 8 9 10
n. 8 25 10 4 6 1 9 7 3 11 8 n. 8 5 10 1 4 3 2 6 7 8 11 9
n. 9 2 7 8 11 6 9 1 3 4 10 5
n. 10 2 11 3 7 9 1 6 4 10 5 8
n. 11 2 11 10 4 6 7 3 1 9 5 8
n. 12 3 7 9 4 10 11 6 2 1 8 5 Knot 350
n. 13 3 7 10 11 2 8 6 5 9 1 4
n. 1 1 2 8 5 10 6 11 3 7 4 9
n. 14 3 11 2 7 10 4 6 1 9 5 8
n. 2 1 9 5 8 3 6 11 10 4 7 2
n. 15 3 11 9 8 2 7 6 10 5 4 1
n. 3 2 1 4 10 5 6 9 8 11 3 7
n. 16 4 1 2 8 6 5 9 11 3 7 10
n. 4 2 7 4 10 11 6 3 8 5 9 1
n. 17 4 1 9 5 3 11 6 8 2 7 10
n. 5 3 4 1 9 5 6 10 11 8 2 7
n. 18 4 1 9 5 6 8 2 11 10 7 3
n. 6 3 8 5 9 1 6 2 7 4 10 11
n. 19 4 7 2 8 6 11 3 5 9 1 10
n. 7 5 9 1 4 3 6 7 2 8 11 10
n. 20 4 7 3 11 6 8 2 5 10 19
n. 8 5 10 4 1 2 6 7 3 11 8 9
n. 21 4 7 3 11 9 5 6 8 2 1 10
n. 22 4 10 1 2 5 9 6 8 11 3 7
n. 23 4 10 5 2 6 8 11 3 7 9 1
n. 24 4 10 5 8 6 2 11 3 7 9 1 Knot 351
n. 25 4 10 7 2 11 3 6 8 5 9 1
n. 26 4 10 11 2 6 8 5 9 1 3 7 1 n. 1 1 9 4 7 2 6 11 3 8 5 10
n. 27 4 10 11 8 6 2 5 9 1 3 7 n. 2 1 9 10 7 8 6 11 3 2 5 4
n. 28 5 8 1 2 6 11 10 4 9 7 3 n. 3 2 1 4 3 7 6 10 11 8 9 5
n. 29 5 9 1 4 6 3 7 10 11 2 8 n. 4 2 5 9 1 4 6 10 7 3 11 8
n. 30 5 9 11 3 7 4 6 10 1 2 8 n. 5 2 5 9 1 10 6 4 7 3 11 8
n. 31 59 11 3 7 10 6 4 1 2 8 n. 6 2 7 4 9 1 6 10 5 8 3 11
n. 32 5 10 4 3 1 9 6 11 8 7 2 | n. 7 2 11 3 7 4 6 10 1 9 5 8
n. 8 2 11 3 7 10 6 4 1 9 5 8
n. 9 4 1 9 5 2 6 8 11 3 7 10
n. 10 4 1 9 5 8 6 2 11 3 7 10
n. 11 4 5 2 3 11 6 8 7 10 9 1
Knot 348 n. 12 4 7 3 11 2 6 8 5 9 1 10
n. 13 4 7 3 11 8 6 2 5 9 1 10
1 n. 1 1 2 7 4 3 6 8 9 5 11 10
n. 14 4 11 2 9 5 6 8 1 10 3 7
n. 2 1 9 5 11 3 6 8 2 7 4 10
n. 15 5 9 2 1 1 4 6 7 3 10 1 8
n.3 2 1 7 3 4 6 9 8 5 10 11
n. 16 5 9 8 11 10 6 7 3 4 1 2 |
n. 4 2 7 1 9 4 6 3 8 11 10 5
n. 5 2 8 5 10 4 6 9 1 7 3 11
n. 6 2 8 11 10 4 6 3 7 1 9 5
n. 7 3 4 7 2 1 6 10 11 5 9 8 Knot 353
n. 8 3 7 1 9 5 6 2 8 11 10 4
n. 9 3 8 11 10 5 6 2 7 1 9 4 n. 1 1 7 3 11 5 6 8 2 9 10 4
n. 10 3 11 5 9 1 6 10 4 7 2 8 n. 2 1 10 9 2 5 6 8 11 3 7 4
n. 11 4 3 7 1 2 6 11 10 5 8 9 n. 3 4 1 9 5 8 6 11 2 3 10 7
n. 12 4 9 1 7 2 6 5 10 11 8 3 n. 4 4 7 3 11 8 6 5 2 9 10 1
n. 13 4 10 5 8 2 6 11 3 7 1 9 n. 5 4 10 3 2 8 6 11 5 9 1 7
n. 14 4 10 11 8 2 6 5 9 1 7 3 n. 6 4 10 9 2 8 6 5 11 3 7 1
n. 15 5 9 1 7 3 6 4 10 11 8 2 n. 7 5 2 9 10 1 6 4 7 3 11 8
n. 16 5 10 11 8 3 6 4 9 1 7 2 | n. 8 5 11 3 7 1 6 4 10 9 2 8

296
B. All-Interval Series

Knot 354
1 n. 1 1 3 5 2 9 6 8 7 10 4 11 1 n. 35 1 2 11 9 5 8 6 4 7 3 1 10
n. 2 1 3 10 7 8 6 9 2 5 4 11 n. 36 2 11 9 7 4 6 8 5 3 1 10
n. 3 1 8 2 5 4 6 3 10 7 9 11 n. 37 3 1 4 5 10 6 2 7 8 11 9
n. 4 1 8 7 10 3 6 4 5 2 9 11 n. 38 3 1 10 5 4 6 8 7 2 11 9
n. 5 3 2 11 4 5 6 7 9 10 1 8 n. 39 3 2 5 4 11 6 1 8 7 10 9
n. 6 3 2 11 9 7 6 5 4 10 1 8 n. 40 3 2 11 4 5 6 7 8 1 10 9
n. 7 3 10 7 8 1 6 11 9 2 5 4 n. 41 3 4 1 2 7 6 5 10 11 8 9
n. 8 3 10 7 9 11 6 1 8 2 5 4 n. 42 3 4 7 2 1 6 11 10 5 8 9
n. 9 4 5 2 8 1 6 11 9 7 10 3 n. 43 3 5 2 1 8 6 4 11 10 7 9
n. 10 4 5 2 9 11 6 1 8 7 10 3 n. 44 3 5 8 1 2 6 10 11 4 7 9
n. 11 4 11 2 3 5 6 7 8 1 10 9 n. 45 3 7 4 11 10 6 2 1 8 5 9
n. 12 4 11 2 8 7 6 5 3 1 10 9 n. 46 3 7 10 11 4 6 8 1 2 5 9
n. 13 5 3 1 10 9 6 4 11 2 8 7 n. 47 3 8 5 10 11 6 1 2 7 4 9
n. 14 5 3 2 11 4 6 9 10 1 8 7 n. 48 3 8 11 10 5 6 7 2 1 4 9
n. 15 5 4 10 1 8 6 3 2 11 9 7 n. 49 3 10 1 8 7 6 5 4 11 2 9
n. 16 5 4 11 2 3 6 8 1 10 9 7 | n. 50 3 10 7 8 1 6 11 4 5 2 9
n. 51 3 11 2 7 8 6 4 5 10 1 9
n. 52 3 11 8 7 2 6 10 5 4 1 9
n. 53 4 1 3 5 2 6 10 7 9 11 8
n. 54 4 1 3 7 10 6 2 5 9 11 8
1 n. 1 1 2 7 4 3 6 9 8 5 10 11 n. 55 4 1 9 5 2 6 10 7 3 11 8
n. 2 1 2 7 4 9 6 3 8 5 10 11 n. 56 4 1 9 7 10 6 2 5 3 11 8
n. 3 1 3 4 7 2 6 10 5 8 9 11 n. 57 4 3 1 7 2 6 10 5 11 9 8
n. 4 1 3 10 7 8 6 4 5 2 9 11 n. 58 4 3 7 1 2 6 10 11 5 9 8
n. 5 1 4 3 2 5 6 7 10 9 8 11 n. 59 4 5 2 3 11 6 1 9 10 7 8
n. 6 1 4 9 2 5 6 7 10 3 8 11 n. 60 4 5 2 9 11 6 1 3 10 7 8
n. 7 1 8 7 10 3 6 9 2 5 4 11 n. 61 4 5 10 1 3 6 9 11 2 7 8
n. 8 1 8 7 10 9 6 3 2 5 4 11 n. 62 4 5 10 1 9 6 3 11 2 7 8
n. 9 1 9 4 7 2 6 10 5 8 3 11 n. 63 4 5 11 3 2 6 10 9 1 7 8
n. 10 1 9 10 7 8 6 4 5 2 3 11 n. 64 4 5 11 9 2 6 10 3 1 7 8
n. 11 1 10 3 8 5 6 7 4 9 2 11 n. 65 4 7 3 1 10 6 2 11 9 5 8
n. 12 1 10 9 8 5 6 7 4 3 2 11 n. 66 4 7 3 11 2 6 10 1 9 5 8
n. 13 2 1 4 3 7 6 5 9 8 11 10 n. 67 4 7 9 1 10 6 2 11 3 5 8
n. 14 2 1 4 9 7 6 5 3 8 11 10 n. 68 4 7 9 11 2 6 10 1 3 5 8
n. 15 2 1 7 3 4 6 8 9 5 11 10 n. 69 4 9 1 7 2 6 10 5 11 3 8
n. 16 2 1 7 9 4 6 8 3 5 11 10 n. 70 4 9 7 1 2 6 10 11 5 3 8
n. 17 2 1 8 5 3 6 9 7 4 11 10 n. 71 4 11 2 3 5 6 7 9 10 1 8
n. 18 2 1 8 5 9 6 3 7 4 11 10 n. 72 4 11 2 9 5 6 7 3 10 1 8
n. 19 2 3 5 11 4 6 8 1 7 9 10 n. 73 4 11 5 3 2 6 10 9 7 1 8
n. 20 2 3 11 5 4 6 8 7 1 9 10 n. 74 4 11 5 9 2 6 10 3 7 1 8
n. 21 2 5 3 1 4 6 8 11 9 7 10 n. 75 4 11 10 7 3 6 9 5 2 1 8
n. 22 2 5 3 11 8 6 4 1 9 7 10 n. 76 4 11 10 7 9 6 3 5 2 1 8
n. 23 2 5 9 1 4 6 8 11 3 7 10 n. 77 5 2 3 4 1 6 11 8 9 10 7
n. 24 2 5 9 11 8 6 4 1 3 7 10 n. 78 5 2 9 4 1 6 11 8 3 10 7
n. 25 2 7 1 3 4 6 8 9 11 5 10 n. 79 5 3 2 11 4 6 8 1 10 9 7
n. 26 2 7 1 9 4 6 8 3 11 5 10 n. 80 5 3 8 11 10 6 2 1 4 9 7
n. 27 2 7 4 3 1 6 11 9 8 5 10 n. 81 5 4 11 2 3 6 9 10 1 8 7
n. 28 2 7 4 9 1 6 11 3 8 5 10 n. 82 5 4 11 2 9 6 3 10 1 8 7
n. 29 2 7 8 11 3 6 9 1 4 5 10 n.83 5 8 3 10 1 6 11 2 9 4 7
n. 30 2 7 8 11 9 6 3 1 4 5 10 n. 84 5 8 9 10 1 6 11 2 3 4 7
n. 31 2 9 5 11 4 6 8 1 7 3 10 n. 85 5 9 2 11 4 6 8 1 10 3 7
n. 32 2 9 11 5 4 6 8 7 1 3 10 n. 86 5 9 8 11 10 6 2 1 4 3 7
n. 33 2 11 3 5 8 6 4 7 9 1 10 n. 87 5 10 11 8 3 6 9 4 1 2 7
n. 34 2 11 3 7 4 6 8 5 9 1 10 | n. 88 5 10 11 8 9 6 3 4 1 2 7 |

297
Bibliography

[1] AGMON (Etyan), "A Mathematical Model of the Diatonic System", Journal of
Music Theory, 33, 1989, pp. 1-25.
[2] AGMON (Etyan), "Coherent Tone-Systems: a study in the theory of diatonicism",
Journal of Music Theory, 40(1), 1996, pp. 39-59.
[3] AGMON (Etyan), "The Multiplicative Norm and its Implications for Set-Class
Theory", Perspectives of New Music, 40(1), 2002, pp. 216-234.
[4] AGON (Carlos), OpenMusic: Un langage visuel pour la composition musicale assis-
tte par ordinateur, Th&se, Paris VIII, 1998.
[5] AGON (Carlos), "Mixing Visual Programs and Music Notation in OpenMusic" in
Mazzola, Noll and Lluis-Puebla, Perspectives in Mathematical and Computational
Music Theory, Epos, 2004, pp. 233-244.
[6] ALEMBERT (Jean Henri d'), Elements de Musique tMorique et pratique selon les
principes de M. Rameau, Paris, 1752. Reprint Slatkine, Geneve.
[7] ALESSI (Rosario), Sonometro. Misura matematica del suono musicale. Metodo teo-
rico per la giusta divisione della scala cromatica (...) per accordare i pianoforti.
A. Morano, Naples 1881.
[8] A M I O T (Emmanuel), "Pour en finir avec le desir: la notion de symetrie en analyse
musicale", Analyse musicale, 22, Paris, 1991.
[9] AMIOT (Emmanuel), "Mathematique et analyse musicale: une fecondation reci-
proque", Analyse musicale, 28, Paris, 1992.
[10] AMIOT (Emmanuel), "La serie dodecaphonique et ses symetries", Quadrature, 19,
Paris, editions du choix, 1994, pp. 15-32.
[11] AMIOT (Emmanuel), "Des Vuza canons aux canons rythmiques 6conomiques", Ma-
MuX Seminar (Mathematiques, Musique et relations avec d'autres disciplines),
Ircam, 2002.
Bibliography

[12] AMIOT (Emmanuel), "A solution to Johnson-Tangian conjecture", MaMuX Semi-


nar (Mathematiques, Musique et relations avec d'autres disciplines), Ircam, 2002.
[13] AMIOT (Emmanuel), "Why Rhythmic Canons are Interesting" in Mazzola, Noll and
Lluis-Puebla, eds. Perspectives in Mathematical and Computational Music Theory,
Epos, 2004, pp. 194-213.
[14] AMIOT (Emmanuel), "Rhythmic Canons and Galois Theory", Colloquium on Ma-
thematical Music Theory, H. Pripertinger, L. Reich (Eds), Grazer Mathematische
Berichte, 347, 2005, pp. 1-21.
[15] AMIOT (Emmanuel), "A propos des canons rythmiques", Gazette des mathemati-
ciens, 106, Societe Mathematique de Prance, 2005, pp. 43-67.
[16] AMIOT (E.), ASSAYAG (G.), MALHERBE ( C ) , R I O T T E (A.), "Duration structure
generation and recognition in musical writing", Proc. of the ICMC 1986, La Haye,
1986.
[17] A M I O T (Emmanuel), ANDREATTA (Moreno), AGON (Carlos), "Tiling The (Mu-
sical) Line with Polynomials: Some Theoretical and Implementational Aspects",
Barcelona, Proceedings of the International Computer Music Conference, Barce-
lona, 2005, pp. 227-230.
[18] ANDLER (Daniel), "Ou resident les structures? Questions de niveau dans Petude
de la cognition", Mathematique et Musique: Logiques mathematiques, logiques mu-
sicales au XXe siicle, Quatrieme Forum Diderot, Ircam, 1999.
[19] ANDREATTA (Moreno), Gruppi di Hajos, Canoni e Composizioni, Tesi di Laurea
in Matematica, Universita di Pavia, 1995.
[20] ANDREATTA (Moreno), La Theorie mathematique de la musique de Guerino Maz-
zola et les canons rythmiques, Memoire de D.E.A. pour le doctorat en musique et
musicologie du XXe siecle, Ehess/Ircam, 1999.
[21] ANDREATTA (Moreno), Methodes algebriques dans la musique et musicologie du
XX6 siecle: aspects th&oriques, analytiques et compositionnels, these en musicologie
computationnelle, Ehess/Ircam, 2003.
[22] ANDREATTA (Moreno), "On Group-Theoretical Methods Applied to Music: Some
Compositional and Implementational Aspects" in Mazzola, Noll and Lluis-Puebla,
eds. Perspectives in Mathematical and Computational Music Theory, Epos, 2004,
pp. 169-193.
[23] ANDREATTA (Moreno), AGON (Carlos), CHEMILLIER (Marc), "OpenMusic et le
probleme de la construction de canons musicaux rythmiques", Actes des Sixidmes
Journees d'Informatique Musicale, CNET-CEMAMu, 1999, pp. 179-185.
[24] ANDREATTA (Moreno), VUZA (Dan Tudor), "On some properties of periodic se-
quences in Anatol Vieru's Modal Theory", Tatra Mountains Mathematical Publi-
cations, 23, 2001, pp. 1-15.
[25] ANDREATTA (Moreno), "Quelques aspects theoriques d'une approche algebrique en
musique", UOuvert, 112, 2005, pp. 1-18.
[26] ASSAYAG (Gerard), AGON ( C ) , LAURSON (M.), RUEDA ( C ) , "Computer Assis-
ted Composition at Ircam: Patchwork and OpenMusic", Computer Music Journal,
23(3), 1999, pp. 59-72.
[27] ASSAYAG, (Gerard), FEICHTINGER (H.G.), RODRIGUES (J.F.), eds. Mathematics
and Music. A Diderot Mathematical Forum, Springer Verlag, 2002.

300
Bibliography

[28] ASSELIN (Pierre-Yves), Musique et temperaments, Paris, Costallat, 1985.


[29] AWRAAMOFF (A.M), "Jenseits von Temperierung und Tonalit&t", Melos, 1, 1920,
pp. 160-166, 184-188.
[30] BABBITT (Milton), Words about Music, S. Dembski, J.N. Strauss (eds.), Madison,
University of Wisconsin Press, 1987.
[31] BABBITT (Milton), "Some Aspects of Twelve-Tone Composition", The Score and
IMA Magazine, 12, pp. 53-61, 1955; reprinted in "Twentieth Century Views of
Music History", New-York, Schribner, 1972, pp. 364-371.
[32] BABBITT (Milton), "Twelve-Tone Invariants as Compositional Determinants", Mu-
sical Quaterly, 46, 1960, pp. 245-259.
[33] BABBITT (Milton), "Set Structure as a Compositional Determinant", Journal of
Music Theory, 5(1), 1961, pp. 72-94.
[34] BABBITT (Milton), "Twelve-Tone Rhythmic Structure and the Electronic Medium",
Perspectives of New Music, 1(1), 1962, pp. 49-79.
[35] BACKUS (John), "Pseudo-science in Music", Journal of Music Theory, 4(2), 1960,
pp. 221-232.
[36] BACKUS (John), The Acoustic Foundations of Music, New York, Norton and Com-
pany, 1969.
[37] BAILHACHE (Patrice), "Temperaments musicaux et mathematiques", Sciences et
Techniques en Perspective, Universite de Nantes, 16, 1989, pp. 83-114.
[38] BAILHACHE (Patrice), Leibniz et la th&orie de la musique, Paris, Klincksieck, 1992.
[39] BAILHACHE (Patrice), Une histoire de Vacoustique musicale, Paris, Editions du
CNRS, 2001. .
[40] BAKER (James), BEACH (David), BERNARD (Jonathan), Music Theory in Concept
and Practice, Eastman Studies in Music, University of Rochester Press, 1997.
[41] BALLON (Fabien), "Denombrement des echelles et des modes dans les systemes
microtemperes", 2004, unpublished.
[42] BALZANO (Gerald J.), "The group theoretic description of 12-fold and microtonal
pitch systems", Computer Music Journal, 4(4), 1980, pp. 66-84.
[43] BALZANO (Gerald J.), "The Pitch Set As A Level of Description for Studying Mu-
sical Pitch Perception," in Manfred Clynes, Music, Mind and Brain, 1992, pp. 321-
351.
[44] BALZANO (Gerald J.), "What Are Musical Pitch and Timbre?", Music Perception,
3(3), 1986, pp. 297-314.
[45] BANCQUART (Alain), Musique: habiter le temps, Lyon, Symetrie, 2003.
[46] BARBAUD (Pierre), Introduction a la composition musicale automatique, Paris, Du-
nod, 1965.
[47] BARBAUD (Pierre), La musique, discipline scientifique, Paris, Dunod, 1968.
[48] BARBAUD (Pierre), Vademecum de Uinge'nieur en musique, Springer Verlag, 1993.
[49] BARBOUR (James Murray), Equal Temperament: its History from Ramis (1482) to
Rameau (1737), Dissertation, Cornell Institute, 1932.
[50] BARCA (Alessandro), Nuovi teoremi per la divisione delle ragioni, Bergamo, 17817

301
Bibliography

[51] BARKECHLI (Mehdi), "Mesure des intervalles harmoniques de la gamme a partir


de la sensation subjective de consonance", Acustica, 2, 1952, pp. 242.
[52] BARTOLUS (Abraham), Musica Mathematical Altenburg, 1614.
[53] BAZELOW (Alexander), BRICKLE (Prank), "A Partition Problem posed by Milton
Babbitt", Perspectives of New Music, 14(2), 15(1), 1976, pp. 280-293.
[54] BASS (Richard), "Models of Octatonic and Whole-tone Interaction: George Crumb
and his Predecessors", Journal of Music Theory, 38, 1994, pp. 155-186.
[55] BEAMENT (James), The violin explained: components, mechanism and sound, Ox-
ford University Press, 1997.
[56] BEAN (Calvert), HILLER (Lejaren), "Information theory analyses of four sonatas
expositions", Journal of Music Theory, 10(1), 1966, pp. 96-137.
[57] BEKESY (Georg von), Experiments in Hearing, Mc Graw Hill, 1960.
[58] BELLERMANN (J.G.H.), Die Grosse der Musikalischen Intervalle als Grundlage der
Harmonie, Berlin, 1783.
[59] BELIS (Annie), Aristoxdne de Tarente et Aristote: Le Traite d'harmonique, Paris,
Klincksieck, 1986.
[60] BENADE(Arthur H.), Fundamentals of Musical Acoustic, Oxford University Press,
1976.
[61] BENSON (David), Mathematics and Music, Cambridge University Press, 2005,
(to appear).
[62] BERAN (Jan), MAZZOLA (Guerino), "Analyzing Musical Structure and Perfor-
mance - A Statistical Approach", Statistical Science, 14(1), 1999, pp. 47-79.
[63] BERAN (Jan), MAZZOLA (Guerino), "Timing Microstructure in Schumann's TrSiu-
merei as an Expression of Harmony, Rhythm and Motivic Structure in Music Per-
formance", Computers and Mathematics with Applications, 39(5-6), 2000, pp. 99-
130.
[64] BERAN (Jan), Statistics in musicology, London, Chapman & Hall, CRC, 2003.
[65] BERNARD (Jonathan), "Problems of Pitch Structure in Elliot Carter's First and
Second String Quartets", Journal of Music Theory, 37, 1993, pp. 231-266.
[66] BERNARD (Jonathan), "Chord, Collection and Set in Twentieth Century", in B A -
KER J. et al. Music Theory in Concept and Practice, 1997, pp. 11-52.
[67] BERNSTEIN (Leon), The Jacobi-Perron algorithm. Its theory and application, Lec-
tures Notes in Mathematics, 207, Springer, 1971.
[68] BERRY (Wallace), Structural function in music, Prentice Hall, 1976 (Reprinted by
Dover, 1987).
[69] BICKIS (Mik), BREMMER (Murray), Statistical Aspects of John Cage's Etudes Aus-
trales, Talk given at the Joint Mathematics Meetings AMS-MAA, Phoenix, 2004.
[70] BLACKWOOD (Easley), The Structure of Recognizable Diatonic Tuning, Princeton
University Press, 1986.
[71] BOON (Jean-Pierre), DECROLY (Olivier), "Dynamical Systems Theory for Music
Dynamics", Chaos, 5(3), 1995, pp. 501-508.
[72] BLOCK (Steven), D O U T H E T T (Jack), "Vector Products and Intervallic Weighting",
Journal of Music Theory, 38(1), 1994, pp. 21-41.

302
Bibliography

[73] BORETZ (B.), CONE (E.T.), Perspectives on Contemporary Music Theory,


W.W. Norton and Company, New York, 1972.
[74] BOSANQUET (Robert Holford Macdowall), "On the mathematical theory of Mr.
Baillie Hamilton's string-organ", Philosophical Magazine, Fourth Series, 49(323),
1874, article no. 12, pp. 98-104.
[75] BOSANQUET (Robert Holford Macdowall), "On some points in the harmony of
perfect consonances", Proceedings of the Musical Association, 3, 1877, pp. 145-153.
[76] BOSSEUR (Dominique and Jean-Yves), Les revolutions musicales, Paris, Sycomore,
1979.
[77] BOSSEUR (Jean-Yves), Vocabulaire de la musique contemporaine, Paris, Minerve,
1992.
[78] BOULEZ (Pierre), Penser la musique aujourd'hui, Paris, Gonthier, 1964.
[79] BREDICEANU (Mihai), Topology of sound forms and music, Romanian Academy
Publishing House, 2002.
[80] B R O U £ (Michel), "Les tonalites musicales vues par un mathematicien", in D. Rous-
seau & M. Morvan eds., "Le temps des savoirs", Revue de VInstitut Universitaire
de France, Paris, 2001.
[81] BRUIJN (N. G. de), "On the factorization of cyclic groups", Indag. Math, 17, 1955,
pp. 370-377.
[82] BRUIJN (N. G. de), "On number systems", Nieuw Arch Wish. 4(3), 1956, pp. 15-17.
[83] BRUIJN (N. G. de), "P6lya's Theory of Counting ", in Beckenbach (ed.), Applied
Combinatorial Mathematics, New York, Wiley, 1964.
[84] BRUIJN (N. G. de), "On the number of partition patterns of a set", Nederl. Akad.
Wetensch. Proc. Ser., Indag. Math., 41, 1979, pp. 229-234.
[85] BRUN (Viggo), "Algorithmes euclidiens pour trois ou quatre nombres", XIII
Congres de MatMmatiques Scandinaves, Helsinki, 1957, pp. 45-64.
[86] BRUN (Viggo), "Mehrdimensionale Algorithmen, welche die Eulersche Kettenbru-
chentwicklung der Zahle verallgemeinern", Leonhard Euler zum 250. Geburtstag,
Berlin, Akademie-Verlag, 1959.
[87] BRUN (Viggo), "Euclidean Algorithms and Musical Theory", L'ensignement ma-
th&matique, Geneve, X, 1964, pp. 125-137.
[88] BuCHLER (Michael), "Relative saturation of interval and set classes: A new model
for understanding pcset complementation and ressemblance", Journal of Music
Theory, 45(2), 2001, pp. 263-343.
[89] BURGUBURA (H.), "Relations des mathematiques et de la musique dans Phistoire
des gammes", Revue des anciens dives de VEcole Centrale des Arts et Manufac-
tures de Paris, Paris, 1973, May pp. 9-15 and June pp. 5-14.
[90] BURN (David), "Further observations on stacked canon and Renaissance compo-
sitional procedure: Gascone's Ista est speciosa and Forestier's Missa L'Homme
Arme", Journal of Music Theory, 45(1), 2001, pp. 73-118.
[91] BURNHAM (Scott), "Method and Motivation in Hugo Riemann's History of
Theory", Music Theory Spectrum, 14(1), 1992, pp. 1-14.
[92] BUSCH (Hermann R.), Leonhard Eulers Beitrage zur Musiktheorie, Bosse, Regens-
burg, 1970.

303
Bibliography

[93] BUTEAU (Chantal), Motivic Topologies and their Signification in Musical Motivic
Analysis, Masters Thesis, Universite de Laval, Quebec, 1998.
[94] BUTEAU (Chantal), A topological model of motivic structure and analysis of music-
Theory and Operationalization, PhD, University of Zurich, 2003.
[95] BUTEAU (Chantal), "Automatic Motivic Analysis Including Melodic Similarity For
Different Contour Cardinalities: Applications to Schumann's Of Foreign Lands and
People", Barcelona, Proceedings of the International Computer Music Conference,
2005, pp. 239-242.
[96] CAREY (Norman), "Well-Formed Scales and the Stern Brocot Tree", American
Mathematical Society Conference, Phoenix, 2004.
[97] CAREY (Norman), CLAMPITT (David), "Aspects of well-formed scales", Music
Theory Spectrum, 11, 1989, pp. 187-206.
[98] CAREY (Norman), CLAMPITT (David), "Regions: A Theory of Tonal Spaces in
Early Medieval Treatises", Journal of Music Theory, 40(1), 1996, pp. 113-147.
[99] CAREY (Norman), CLAMPITT (David), "Self-similar Pitch Structures, Their Duals,
and Rhythmic Analogues", Perspectives of New Music, 34(2), 1996, pp. 62-87.
[100] CARLOS (Wendy), "Tuning: At the Crossroads", Computer Music Journal, 11(1),
1987, pp. 29-43.
[101] CARLOS (Wendy), "Three Asymmetrical Divisions of the Octave", Pitch, 4(1),
1990, pp. 2-3.
[102] CASTELLENGO (Michele), FABRE (Benolt), G U Y O T (Frederique), "Etude de la ca-
tegorisation d'un corpus de bruits domestiques", in Daniele Dubois, Categorisation
et cognition, Paris, Kime, 1997, pp. 41-58.
[103] CAZDEN (Norman), "Musical consonance and dissonance: a cultural criterion",
Journal of Aesthetics, 4, 1945, pp. 3-11.
[104] CAZDEN (Norman), "Sensory theories of musical consonance", Journal of Aesthe-
tics, 20, 1962, pp. 310-319.
[105] CAZDEN (Norman), "The definition of consonance and dissonance", International
Review of Aesthetics and Sociology of Music, 11(2), 1980, pp. 123-168.
[106] CHALMERS (John H, Jr.), "The Application of Rothenberg's Pattern Recognition
Model to the Structure of Tetrachords and Tetrachordal Scales", Xenharmonikdn,
3, 1975.
[107] CHALMERS (John H, Jr.), "Microtonal Mathematics", Interval, 1(2), 1978.
[108] CHALMERS (John H, Jr.), "Construction and Harmonization of Microtonal Scales in
Non-12-Tone Equal Temperaments", Proceedings of the 1982, International Com-
puter Music Conference, ICMA, 1982, pp. 534-555.
[109] CHALMERS (John H, Jr.), "Polychordal Arrays of MOS Scales", Xenharmonikdn,
7-8, 1979.
[110] CHALMERS (John H, Jr.), "Some Additional With Nineteen-tone Scales", Xenhar-
monikdn, 7-8, 1979.
[Ill] CHALMERS (John H, Jr.), "Tritriadic Scales With Seven Tones", Xenharmonikdn,
9, 1986.
[112] CHALMERS (John H, Jr.), "Tritriadic Transformations", One/one, 3(1), 1987.

304
Bibliography

[113] CHALMERS (John H, Jr.), "Tritriadic Scales with Seven Tones, Part Two: Derived
Forms and Structural Properties", Xenharmonikdn, 10, 1988.
[114] CHALMERS (John H, Jr.), "Third Coast New Music Project", One/one, 4(3), 1988,
pp. 15.
[115] CHALMERS (John H, Jr.), "Tritriadic Scales with Seven Tones. Part Three: The
M -> T and D - • M Matrices", Xenharmonikdn, 11, 1989, pp. 40-68.
[116] CHALMERS (John H, Jr.), "Tetrachordal Scales and Complexes", Xenharmonikdn,
11, 1989, pp. 1-20.
[117] CHALMERS (John H, Jr.), "Three Approaches to Harmony in 13-TET", Xenhar-
monikdn, 13, 1991, pp. 53-64.
[118] CHALMERS (John H, Jr.), Divisions of the Tetrachord, Hanover NH, Prog Peak
Music, 1993.
[119] CHEMILLIER (Marc), "MonoYde libre et musique", RAIRO, Informatique tMo-
rique et applications, Paris, Gauthier Villars, 21(3), 1987, pp. 341-371, and 21(4),
pp. 379-418.
[120] CHEMILLIER (Marc), Structure et m&thodes algGbriques en Informatique musicale,
These Paris, L.I.T.P., Institut Blaise Pascal, 1990.
[121] CHEMILLIER (Marc), "Mathematiques et musiques de tradition orale", Musique et
mathematiques, Paris, Alea, 1997, pp. 133-143.
[122] CHEMILLIER (Marc), "Generateurs musicaux et singularites", Actes des Sixidmes
Journe'es d'Informatique Musicale, CNET-CEMAMu, 1999, pp. 167-177.
[123] CHEMILLIER (Marc), "Ethnomusicology, Ethnomathematics. The Logic Underlying
Orally Transmitted Artistic Practices", in ASSAYAG et al. 2002, pp. 161-183.
[124] CHEMILLIER (Marc), PACHET (F.), Recherches et applications en informatique mu-
sicale, Paris, Hermes, 1998.
[125] CHEMILLIER (Marc), DUCHAMP (Gerard), "Algorithme de reduction des degres
dans une gamme musicale (d'apres la theorie d'Yves Hellegouarch)", Gazette des
Mathematiques, Societe Mathematique de Prance, 82, 1999, pp. 26-30.
[126] CHEMILLIER (Marc), TRUCHET (Charlotte), "Computation of words satisfying the
rhythmic oddity property (after Simha Arom's works)", Information Processing
Letters, 86, 2003, pp. 255-261.
[127] CHILDS (Adrian), "Moving Beyond Neo-Riemannian Triads: Exploring a Trans-
formational Model for Seventh Chords", Journal of Music Theory, 42(2), 1998,
pp. 181-193.
[128] CHOUVEL (Jean-Marc), "La physique et Pesthetique. Une analyse epistemologique
des modalites de connaissance du phenomene harmonique", Paris, Editions Eska,
Musurgia, 2(4), 1995, pp. 88-102.
[129] CHOUVEL (Jean-Marc), Esquisses pour une pensie musicale. Les metamorphoses
d'Orphee, Paris, L'Harmattan, 1998.
[130] CHOUVEL (Jean-Marc), DOVAL (Boris), "La concordance harmonique: vers une
nouvelle approche de la consonance", Communication a la troisieme conference
internationale pour la perception et la cognition musicale, Liege, 23-27 juillet 1994.
[131] CHOWNING (John), "The Synthesis of Complex audio Spectra by Means of Fre-
quency Modulation", Journal of the Audio Engineering Society, 21(7), 1985,
pp. 526-534.

305
Bibliography

[132] CLAMPITT (David), "Some Refinements on the Three Gap Theorem, with Appli-
cations to Music", Muzica, 6(2), 1995, pp. 12-22.
[133] CLAMPITT (David), Pairwise Well-formed Scales: Structural and Transformational
Properties, Ph.D. Dissertation, State University of New York at Buffalo, 1997.
[134] CLOUGH (John), "Diatonic Interval Sets and Transformational Structures", Music
Theory Spectrum, 18, 1979, pp. 461-482.
[135] CLOUGH (John), "Diatonic Interval Cycles and Hierarchical Structures", Perspec-
tives of New Music, 32(1), 1994, pp. 228-253.
[136] CLOUGH (John), "A rudimentary geometric model for contextual transposition and
inversion", Journal of Music Theory, 42(2), 1998, pp. 297-341.
[137] CLOUGH (John), MYERSON (Gerald), "Variety and Multiplicity in Diatonic Sys-
tems", Journal of Music Theory, 29, 1985, pp. 249-270.
[138] CLOUGH (John), MYERSON (Gerald), "Musical Scales and the Generalized Circle
of Fifths", American Mathematical Monthly, 93(9), 1986, pp. 695-701.
[139] CLOUGH (John) and D O U T H E T T (Jack), "Maximally Even Sets", Journal of Music
Theory, 35,1991, pp. 93-173. Errata: Journal of Music Theory, 36(2), 1992, pp. 401-
402.
[140] CLOUGH (John), CUCIUREAN (John), and D O U T H E T T (Jack), "Hyperscales and
the Generalized Tetrachord", Journal of Music Theory, 41, 1997, pp. 67-100.
[141] CLOUGH (John), ENGEBRETSEN (Nora), and KOCHAVI (Jonathan), "Scales, Sets
and Interval Cycles: A Taxonomy", Music Theory Spectrum, 21(1), 1999, pp.74-104.
[142] COXETER (Harold Scott Macdonald), Introduction to geometry, John Wiley, 1959.
[143] COXETER (Harold Scott Macdonald), Regular Polytopes, Macmillan, 1963.
[144] COHN (Richard), "Maximally Smooth Cycles, Hexatonic Systems, and the Analysis
of Late-Romantic Triadic Progressions", Music Analysis 15(1), 1996, pp. 9-40.
[145] COHN (Richard), "Neo-Riemannian Operations, Parsimonious Trichords, and their
Tonnetz Representations", Journal of Music Theory, 41(1), 1997, pp. 1-66.
[146] COHN (Richard), "Introduction to Neo-Riemannian Theory: A Survey and A His-
torical Perspective", Journal of Music Theory, 42(2), 1998, pp. 167-180.
[147] COHN (Richard), "Square Dances with Cubes", Journal of Music Theory, 42(2),
1998, pp. 283-296.
[148] COLBOURN (Charles), ROSA (Alexander), Triple Systems, Oxford University Press,
1999.
[149] COSTERE (Edmond), Lois et styles des harmonies musicales, Paris, Presses Uni-
versitaires de Prance, 1954.
[150] COSTERE (Edmond), Mort ou transfiguration de Vharmonie, Paris, Presses Uni-
versitaires de Prance, 1962.
[151] COOKE (Deryck), The Language of Music, Oxford University Press, 1959.
[152] COONS (Edgar), KRAEHENBUEHL (David), "Information as a measure of structure
in Music", Journal of Music Theory, 2(2), 1958, pp. 127-161.
[153] COVACH (John R), "The Zwolftonspiel of Josef Matthias Hauer", Journal of Music
Theory, 36(1), 1992, pp. 149-184.

306
Bibliography

[154] COVEN (Ethan), MEYEROWITZ (Aaron), "Tiling the integers with translate of one
finite set", Journal of Algebra, 212, 1999, pp. 161-174.
[155] CROCKER (Malcolm), ed. Handbook of Acoustics, Wiley Interscience, 1998.
[156] DABBY (Diana S.), "Musical Variations form a Chaotic Mapping", Chaos, 6(2),
1996, pp. 95-107.
[157] DAHLHAUS (Carl), Untersuchung uber die Entstehung der harmonischen Tonalitat,
Barenreiter, Kassel, 1967.
[158] DALL'OLIO (Giambattista), "Sull'applicazione della matematica alia musica", Me-
morie delle Societa Italiana delle Scienze, IX, 1802, pp. 609-625.
[159] DAMMAN (Ralf), "Die Musica mathematica von Bartotus", Archiv fur Musikwis-
senchaft, XXVI, 1969, pp. 140-162.
[160] DANIELOU (Alain), Simantique musicale, Essai de psycho-physiologie auditive, Pa-
ris, Hermann, 1967.
[161] DELIEGE (Celestin), Les fondements de la musique tonale, Paris, Lattes, 1984.
[162] DELIEGE (Celestin), Cinquante ans de moderniti musicale: de Darmstadt a VIrcam.
Constribution historiographique a une musicologie critique, Bruxelles, Mardaga,
2003.
[163] DELIEGE (Irene), PADDISON (M.), Musique contemporaine. Perspectives th&oriques
et philosophiques, Bruxelles, Mardaga, 2000.
[164] D E L O N E (Richard), et al. eds. Aspects of 20th Century Music, Englewood Cliffs,
Prentice-Hall, 1975.
[165] DEVIE (Dominique), Le temperament musical, Societe de musicologie du Langue-
doc, 1990.
[166] D O U T H E T T (Jack), ENTRINGER (Roger), MULLHAUPOT (Andrew), "Musical Scale
Construction: The Continued Fraction Compromise", Utilitas Mathematica, 42,
1992, pp. 97-113.
[167] DOUTHETT (Jack), STEINBACH (Peter), "Parsimonious graphs: A study in parsi-
mony, contextual transformations, and modes of limited transpositions", Journal
of Music Theory, 42(2), 1998, pp. 241-263.
[168] DRIEBERG (Priedrich von), Die mathematische Intervallenlehre der Griechen, Ber-
lin, 1818.
[169] DUCHON (M.), "Survey of some topics of harmonic analysis", Tatra Mountains
Mathematical Publications, 23, 2001, pp. 165-169.
[170] DUDON (Jacques), "La gamme doree: Un temperament inegal issu du nombre
d'or", in Jean-Bernard Condat, Nombre d'or et Musique, Verlag Peter Lang, 1988,
pp. 147-157.
[171] DUDON (Jacques), "Musiques Harmoniques", One/one, 5(3), 1989, pp. 10-14.
[172] DuiJVESTUN (Arie), "A Simple Perfect Square of Lowest Order", J. Comb. Th.
Ser. B, 25, 1978, pp. 240-243
[173] DUNSBY (Jonathan), Music Analysis in Theory and Practice, New Haven: Yale
University Press, 1988.
[174] DURUTTE (Camille), Technie, ou lois g6n£rales du systeme harmonique, Mallet-
Bachelier, Paris, 1855.

307
Bibliography

[175] EBERLE (Gottfried), "Absolute Harmonie und Ultrachromatik", in 0 . Kolleritsch


(ed.), Alexandre Skrjabin, Graz, Universal Edition, 1980.
[176] ElTZ (Carl A.), Das mathematisch reine Tonsystem, Leipzig, 1891.
[177] ElMERT (Heimert), Atonale Musiklehre, Leipzig, 1924.
[178] ElMERT (Heimert), Grundlagen der musikalischen Reihentechnik, Wien, Universal
Edition, 1964.
[179] ENGEL (Gustav), Das mathematische Harmonium: Ein Hilfsmittel zur Veranschau-
lichung der reinen Tonverhaltnisse, Berlin, 1881.
[180] ESBROECK (Van), "Le probleme des gammes", Bulletin du GAM, 1, Universite
Paris VII, 1963.
[181] ESTRADA (Julio), The'orie de la composition: discontinuum-continuum, These, Uni-
versite de Strasbourg, 1994.
[182] EULER (Leonard), Tentamen novae theoriae musicae ex certissimis harmoniae
principiis dilucide expositae, St. Petersburg, 1729.
[183] EULER (Leonard), Exposition de quelques nouvelles vues matMmatiques dans la
theorie de la musique, Amsterdam, 1760.
[184] EULER (Leonard), Conjecture sur la raison de quelques dissonances generalement
recues dans la musique, Memoires de PAcademie des sciences de Berlin, 1764,
pp. 165-173.
[185] EULER (Leonard), Du veritable caract&re de la musique, Memoires de PAcademie
des sciences de Berlin, 1766, pp. 174-199.
[186] EULER (Leonard), De harmoniae veris pincipiis perspeculum musicum repraesenta-
tis, Novi commentarii Academiae scientiarum Petropolitanae, XVIII, 1774, pp. 330-
353.
[187] EULER (Leonard), Lettres a une princesse dfAllemagne sur divers sujets de physique
et de philosophic, I, Paris 1812, pp. 28-33 (Lettre VII: uDes douze tons du clavecin",
Berlin 3 mai 1760).
[188] EULER (Leonard), "Essai dfune nouvelle tMorie de la musique", in Musique ma-
th&matique, Librairie Scientifique et Philosophique, Paris, 1865.
[189] FAREY (John), "On Music", The Philosophical Magazine, XXVI, London, 1806,
pp. 171-176.
[190] FAREY (John), "On the Stanhope temperament of the musical scale", The Philo-
sophical Magazine, XXVIII, London, 1806, pp. 140-152.
[191] FAREY (John), "On different temperaments of the Musical Scale", The Philosophi-
cal Magazine, XXVII, London, 1807, pp. 313-322.
[192] FAREY (John), "On a new Mode of Equally Tempering the Musical Scale", The
Philosophical Magazine, XXVIII, London, 1807, pp. 65-66.
[193] FAREY (John), "Six theorems, containing the chief properties of all Regular Dou-
zeave Systems of music, with twelve corollaries thence deduced, containg the tem-
peraments of as many systems", The Philosophical Magazine, London, 1810, pp. 39-
53.
[194] FAREY (John), "On the recent Alterations said to be made by some tuners of
Musical Instrument, The Place of Wolves, or largely temperated Concords", The
Philosophical Magazine, LVI, London, 1820, pp. 341-342.

308
Bibliography

[195] FAREY (John), "On different modes of expressing the magnitudes and relations
of musical intervals, with some remarks, in commendation of Professor Fischer's
proportionally-tempered douzeave", The Silliman American Journal of Science and
Arts, II, London, 1820, pp. 65-81.
[196] FAUVEL (John), FLOOD (Raymond), WILSON (Robert), Music and Mathematics,
From Pythagoras to Fractals, Oxford University Press, 2003.
[197] FEICHTINGER (Hans G.), DORFLER (M.), eds. Diderot Forum on Mathematics
and Music: Computational and Mathematical Methods in Music, Vienna, Osterrei-
chische Computergesellschaft, 1999.
[198] FERNANDEZ (Claude), PERCHERANCIER (Jean-Michel), Les oeuvres pour piano et
orchestre, Geneve, Champion-Slatkine, 1988.
[199] FERRETTI (R.), MAZZOLA (G.), "Algebraic Varieties of Musical Performances",
Tatra Mountains Mathematical Publications, 23, 2001, pp. 59-69.
[200] FlCHET (Laurent), Les theories scientifiques de la musique aux XIXe et XXe si£cle,
Vrin, Paris, 1996.
[201] FISCHER (Myriam), "Leonhard Euler et la musique", Strasbourg, L'Ouvert, 112,
2005, pp. 19-31.
[202] FLETCHER (Neville H.), and ROSSING (Thomas D.), The Physics of Musical Ins-
truments, Springer Verlag, 1991.
[203] FOKKER (Adriaan D.), "Het muzikale toonstelsel van Christiaan Huygens, de nor-
male di&zenstemming", 1942.
[204] FOKKER (Adriaan D.), "Rekenkundige bespiegeling der muziek", (Arithmetical re-
flection of music), Noorduijn, Gorinchem, 1945, review by E.J. Dijksterhuis, in
Mens en Melodie, 1, 1946, pp. 59-62.
[205] FOKKER (Adriaan D.), "Les mathematiques et la musique", in Recherches musi-
cales. Th&oriques et pratiques, Den Haag, From Teyler Museum Archives, 10, 1947,
pp. 1-31.
[206] FOKKER (Adriaan D.), "De behoefte aan grotere nauwkeurigheid in de muzikale
notatie der toonhoogte", 1953.
[207] FOKKER (Adriaan D.), "Equal Temperament and the Thirty-one-keyed Organ",
1955.
[208] FOKKER (Adriaan D.), "Optelakkoorden", 1960.
[209] FOKKER (Adriaan D.), "Met verjongde oren", 1964.
[210] FOKKER (Adriaan D.), "Simon Stevin's views on music", Introduction to Vande
spiegheling der singconst ("On the theory of the art of singing"), in Principal Works
of Simon Stevin 5, A.D. Fokker (ed.), Amsterdam, 1955-1966, pp. 413-464.
[211] FOKKER (Adriaan D.), "On the Expansion of the Musician's Realm of Harmony",
This paper is a preparatory study for a report on the problem of the notation of
microtones, which will be presented to the general assembly of the International
Musicological Society in 1967.
[212] FOKKER (Adriaan D.), "Unison vectors and periodicity blocks in the three-
dimensional (3-5-7), harmonic lattice of notes", From Koninklijke Nederlandse Aka-
demie van Wetenschappen - Amsterdam, Proceedings, Series B 72(3), 1969 (Com-
municated at the meeting of February 22, 1969).

309
Bibliography

[213] FOKKER (Adriaan D.), New Music With 31 Notes, translated by Leigh Gerdine,
Orpheus - Series of Monographs on Basic Questions in Music, vol. 5, Verlag fur
Systematische Musikwissenschaft, Bonn, 1975.
[214] FOKKER (Adriaan D.), VAN DlJK, (Jan), "Experiences musicales avec les genres
musicaux de Leonhard Euler contenant la septieme harmonique", Recherches mu-
sicales. TMoriques et pratiques, Den Haag, Martinus Nijhoff, Teyler Museum Ar-
chives, 10, 1951.
[215] FORD (L.R.), "Fractions", American Mathematical Monthly, 45(9), pp. 586-601.
[216] FORTE (Allen), "A Theory of Set-Complexes for Music", Journal of Music Theory,
8(2), 1964, pp. 136-184.
[217] FORTE (Allen), "Sets and Nonsets in Schoenberg's Atonal Music", Perspectives of
New Music, 11(1), 1972, pp. 43-64.
[218] FORTE (Allen), The Structure of Atonal Music, New Haven and London, Yale
University Press, 1973.
[219] FORTE (Allen), Introduction to Schenkerian Analysis, New York, Norton, 1982.
[220] FRANCES (Robert), La perception de la musique, Paris, Vrin, 1984.
[221] FRIPERTINGER (Harald), "Enumeration in Musical Theory", S6minaire Lotharin-
gien de Combinatoire, 476(42), 1992, pp. 26-29.
[222] FRIPERTINGER (Harald), "Endliche Gruppenaktionen auf Funktionenmengen. Das
Lemma von Burnside - Repr&sentantenkonstruktionen - Anwendungen in der Mu-
siktheorie", Bayreuther Mathematische Schriften, 45, 1993, pp. 19-135.
[223] FRIPERTINGER (Harald), "Enumeration and construction in music theory", in H.G.
FEICHTINGER and M. DORFLER eds. Computational and Mathematical Methods in
Music Theory, Vienna, Austria, 1999, pp. 179-204.
[224] FRIPERTINGER (Harald), "Enumeration of Mosaics", Discrete Mathematics, 199,
1999, pp. 49-60.
[225] FRIPERTINGER (Harald), "Enumeration of Non-Isomorphic Canons", Tatra Mt.
Math. Publ. 23, 2001, pp. 47-57.
[226] FRIPERTINGER (Harald), "Enumeration of Mosaics, Enumeration of Canons", Ma-
MuX Seminar (Mathematiques, Musique et relations avec d'autres disciplines),
Ircam, 2002.
[227] FRIPERTINGER (Harald), "Tiling Problems in Music Theory" in Mazzola, Noll and
Lluis-Puebla, eds. Perspectives in Mathematical and Computational Music Theory,
Epos, 2004, pp. 153-168.
[228] FULLER (Ramon), "A Study of Interval and Trichord Progressions", Journal of
Music Theory, 16(1-2), 1972, pp. 102-141.
[229] GALLIMARD (Jean Edme), L'arithme'tique des musiciens, Paris, 1754.
[230] GALLIMARD (Jean Edme), La tMorie des sons ou Von demontre dans une exacte
precision les rapports de tous les intervalles de la gamme, Paris, Ballard, 1754.
[231] GAMER (Carlton), LANSKY (Paul), "Fanfare for the Common Tone", Perspectives
of New Music, 1976, pp. 229-235.
[232] GAMER (Carlton), "Some combinatorial resources of equal-tempered systems",
Journal of Music Theory, 11, 1967, pp. 32-59.

310
Bibliography

[233] GENEVOIS (Hugues), ORLAREY (Yann), eds. Musiques et Mathematiques, Paris,


Aleas, 1997.
[234] GIBELIUS (Otto), Propositiones mathematicae-musicae, Miinden, 1666.
[235] GILBERT (Steven), The Trichord: An Analytical Outlook for Twentieth Century
Music, Ph.D. dissertation, Yale University, 1970.
[236] GILBERT (Steven), "An Introduction to Trichordal Analysis", Journal of Music
Theory, 18(2), 1974, pp. 338-362.
[237] GLASBERG (Brian) and MOORE (Brian), "Derivation of auditory filter shapes from
notched noise data", Hearing Research, 47, 1990, pp. 103-108.
[238] GOLLIN (Edward H.), "Some Aspects of Three-dimensional Tonnetze", Journal of
Music Theory, 42(2), 1998, pp. 195-206.
[239] GOLLIN (Edward H.), Representations of Space and Conceptions of Distance in
Transformational Music Theories, Ph. D. Thesis, Harvard University, 2000.
[240] GOULD (Mark), "Balzano and Zweifel: Another look at generalized diatonic scales",
Perspectives of New Music, 38(2), 2000, pp. 88-105.
[241] GOTZE (Heinz), WlLLE (Rudolf), eds. Musik und Mathematik, Salzburger Musik-
gesprach 1984, Berlin, Springer, 1985.
[242] GREENWOOD (Donald), "Auditory masking and the critical band", Journal of
Acoustical Society of America, 33(4), 1961, pp. 484-502.
[243] GREENWOOD (Donald), "Critical bandwith and the frequency coordinates of the
basilar membrane", Journal of Acoustical Society of America, 33(4), 1961, pp. 1344-
1356.
[244] GREENWOOD (Donald), "A cochlear frequency-position function for several species
- 29 years later", Journal of Acoustical Society of America, 87(6), 1990, pp. 2592-
2605.
[245] GREENWOOD (Donald), "Critical Bandwith and consonance in relation to cochlear
frequency-position coordinates", Hearing Research, 54(2), 1991, pp. 164-208.
[246] GROSS (Dorothy), A Set of Computer Programs to Aid in Music Analysis, Ph. Diss
Indiana University, 1975.
[247] GRONBAUM (Branko), SHEPHARD (Geoffrey C ) , Tilings and patterns, Freeman,
1986.
[248] GuiTART (Rene), La pulsation matMmatique. Rigueur et ambiguite, la nature de
Vactivity matMmatique, ce dont il s'agit d'instruire, Paris, L'Harmattan, 1999.
[249] G U T (Serge), La tierce harmonique dans la musique occidentale, Paris, Heugel,
1967.
[250] H A J 6 S (Gy5rgy), "Sur la factorisation des groupes abeliens", Casopis Pest. Mat.
Fys. 4(3), 1950, pp. 157-162.
[251] HAJOS (Gy5rgy), "Sur le probleme de la factorisation des groupes cycliques", Acta
Math. Acad. Sci. Hungar., 1, 1950, pp. 189-195.
[252] HALL (Donald E.), "The Objective Measurement of Goodness-of-Fit for Tunings
and Temperaments", Journal of Music Theory, 17, 1973, pp. 274-290.
[253] HALL (Donald E.), "Quantitative Evaluation of Musical Scale Tunings", American
Journal of Physics, 42, 1974, pp. 543-552.

311
Bibliography

[254] HALL (Donald E.), Musical Acoustics, Wadsworth Publishing Compagny, Belmont,
California, 1980.
[255] HALL (Donald E.), "A Systematic Evaluation of Equal Temperaments Through
n = 612", Interface, 14, 1985, pp. 61-73.
[256] HALSEY (D.), H E W I T T (E.), "Eine gruppentheoretische Methode in der Musiktheo-
rie", Jahresber. der Dt. Math. Vereinigung, 80, 1978, pp. 151-207.
[257] HALUSKA (Jan), "Uncertainty and tuning in music", Tatra Mountains Mathema-
tical Publications, 12, 1997, pp. 113-129.
[258] HALUSKA (Jan), "Searching the frontier of the Pythagorean system", Tatra Moun-
tains Mathematical Publications, 16, 1999, pp. 273-281.
[259] HALUSKA (Jan), ed. Harmonic Analysis and Tone Systems, Bratislava, Tatra
Mountains Mathematical Publications, 23, 2001.
[260] HALUSKA (Jan), "Equal temperament and Pythagorean Tuning: a geometrical in-
terpretation in the plane", Fuzzy Sets and Systems, 114, 2000, pp. 261-269.
[261] HALUSKA (Jan), The Mathematical Theory of Tone Systems, Marcel Dekker, New-
York, 2003.
[262] HALUSKA (Jan), RIECAN (B.), "On a lattice of tetrachords", Tatra Mountains
Mathematical Publications, 16, 1999, pp. 283-294.
[263] HANSON (Howard), Harmonic Materials of Modern Music: Resources of the Tem-
pered Scale, New York, Appleton-Century-Crofts Inc., 1946.
[264] HARDCASTLE (David), KHANIN (Kostya), "Continued Fractions and the d-
dimensional Gauss Transformation", Communications in Mathematical Physics,
215, 2001, pp. 487-515.
[265] HARDY (Geoffrey), W R I G H T (Edward), An introduction to the theory of numbers,
Oxford University Press, 1980.
[266] HARRISON (Daniel), Harmonic function in chromatic music: A renewed dualist
theory and an account of its precedents, University of Chicago Press, 1994.
[267] HARRISON (Daniel), "Some Group Properties of Triple Counterpoint and Their
Influence on Compositions by J.S. Bach", Journal of Music Theory, 32(1), 1988,
pp. 23-49.
[268] HARTMANN (George C ) , "A numerical exercise in musical scales", American Jour-
nal of Physics, 55(3), 1987, pp. 223-226.
[269] HASCHER (Xavier), "Liszt et les sources de la notion d'agregat", Analyse Musicale,
43, 2002, pp. 48-57.
[270] HASCHER (Xavier), "Une analyse transformationnelle de l'opus 19 n° 4 de Schoen-
berg au moyen des K-reseaux", paper delivered at the International Symposium
"Around Set Theory", Ircam, 2003 (to appear).
[271] HASCHER (Xavier), "Famille cordique, relation de similarity et uniformite maxi-
male", LVuvert, 112, 2005, pp. 33-46.
[272] HAUER (Josef Matthias), Zwolftontechnik, Die Lehre von den Tropen, Vienna, Uni-
versal Edition UE 8438, 1926.
[273] HELLEGOUARCH (Yves), "Scales", Comptes Rendus de la Societe Royale du Ca-
nada, IV(5), 1982 and V(2), 1983.

312
Bibliography

[274] HELLEGOUARCH (Yves), "L'Essai d'une nouvelle theorie de la musique de Leonhard


Euler", in Destin de VArt, desseins de la Science, Actes du colloque de Caen, 1986,
IREM Basse Normandie, Caen, 1992.
[275] HELLEGOUARCH (Yves), "A la recherche de Parithmetique qui se cache dans la
musique", Gazette des Mathematiques, Societe Mathematique de Prance, 33, 1987,
pp. 71-80.
[276] HELLEGOUARCH (Yves), "Gammes naturelles", in Musique et mathematique, Publi-
cation de PA.P.M.E.P. 53, Paris, 1984. Other version in Gazette des Mathematiques,
Societe mathematique de Prance, 81, 1999, pp. 25-39, and 82, pp. 13-26.
[277] HELLEGOUARCH (Yves), "Outils diophantiens pour la definition d'une distance har-
monique", MaMuX Seminar (Mathematiques, Musique et relations avec d'autres
disciplines), Ircam, 2002.
[278] HELMHOLTZ (Hermann L.F, von), Die Lehre von der Tonempfindungen, als Physio-
logische Grundlage fur die Theorie der Musik, Heidelberg 1862. English Translation
by Ellis in "On the Sensations of Tone".
[279] HELMHOLTZ (Hermann L.F, von), Theorie physiologique de la musique, Paris, 1868,
Translation by M.G. Geroult.
[280] HELMHOLTZ (Hermann L.F, von), On the sensation of tone as a physiological basis
for the theory of music, Londres, 1885. Fac-sim. New York, Dover, transl. Alexander
J. Ellis, 1954.
[281] HERMANN (Richard), "Metonymy, Synecdoche, and Metaphor: Reflections of and
on Neo-Riemannian Analytical Techniques", MaMuX Seminar, Special Session On
Transformational Theory and Neo-Riemannian Analysis, Ircam, 2005.
[282] HERMANN (Richard), "Parsimonious Equivalence Classes for Voice-leading Bet-
ween Maximally Even and Near Maximaly Even Set-Classes", MaMuX Seminar,
Special Session On Transformational Theory and Neo-Riemannian Analysis, Ircam,
2005.
[283] HESSE (Horst-Peter), Grundlagen der Harmonik in Mikrotonalitat, Innsbruck, Hel-
bling, 1989.
[284] HESSE (Horst-Peter), "Die Sonanz als Basis der Harmonik in mikrotonaler Musik",
Innsbruck, Mikrotone 3, 1990.
[285] HlNDEMlTH (Paul), Unterweisung im Tonsatz, Mayence, Schott, 1940.
[286] HOERNER (Sebastian von), "The Definition of Major Scales, for Chromatic Scales
of 12, 19 and 31 Divisions per Octave", Psychology of Music, 4(1), 1976, pp. 12-23.
[287] HOFSTADTER (Douglas R.), Godel, Escher, Bach, Harvester Press, 1979.
[288] HOOK (Julian), "Uniform Triadic Transformations", Journal of Music Theory,
46(2), 2002, pp. 57-126.
[289] HOSTINSKY (Ottokar), Die Lehre von den musikalischen Klangen: ein Beitrag zur
aesthetischen Begriindung der Harmonielehre, Prague, H. Domenicus, 1879.
[290] HOWE (Hubert), "Some Combinational Properties of Pitch Structures", Perspec-
tives of New Music, 4(1), 1965, pp. 45-61.
[291] HUEBER (Kurt Anton), "Mathematisch-physikalische Grundlagen einer ekmeli-
schen Intervallehre", in Franz Richter Herf (eds), Mikrotone 2, 1988, pp. 115-124.

313
Bibliography

[292] HURON (David), "Tone and Voice: A Derivation of the Rules of Voice-Leading from
Perceptual Principles", Music Perception, 19(1), 2001, pp. 1-64.
[293] HuSMANN (Heinrich), Einfwrung in die Musikwissenschaft, Heinrichschofen, Wil-
helmshaven, 1975.
[294] HYER (Brian), Tonal Intuitions in Tristan und Isolde, Ph.D. diss., Yale University,
1989.
[295] HYER (Brian), "Reimag(in)ing Riemann", Journal of Music Theory 39(1), 1995,
pp. 101-138.
[296] ISAACSON (Eric), "Similarity of Interval-Class Content Between Pitch-Class Sets:
The IcVSIM Relation", Journal of Music Theory, 34(1), 1990, pp. 1-28.
[297] ILOMAKI (Tuukka), "Group Structures and Equivalence Classes in Extended
Twelve-Tone Operations", Proceedings of the International Computer Music Confe-
rence, Barcelona, 2005, pp. 487-490.
[298] JACKENDOFF (Ray), LERDAHL (Fred), A Generative Theory of Tonal Music, Cam-
bridge Mass., MIT Press, 1983.
[299] JEDRZEJEWSKI (Pranck), ed. La loi de la Pansonorite Ivan Wyschnegradsky (ver-
sion de 1953), Geneve, Contrechamps, 1996.
[300] JEDRZEJEWSKI (Pranck), MatMmatiques des systdmes acoustiques, Temperaments
et modeles contemporains, Paris, L'Harmattan, 2002.
[301] JEDRZEJEWSKI (Pranck), Dictionnaire des musiques microtonales, Paris, L'Har-
mattan, 2003.
[302] JEDRZEJEWSKI (Pranck), "Nceuds polychromes et entrelacs sonores. Vers de nou-
velles categories musicales", Musicae Scientiae, Special issue 2003-2004. Award
papers. Special 10th anniversary conference issue, pp. 73-84.
[303] JEDRZEJEWSKI (Pranck), "Alois Haba et Pexperimentation micro-intervallique",
Congres international de Musique Tcheque, Paris Sorbonne, 15 novembre 2002, in
L'attraction et la necessite. Musique tchdque et culture francaise au XXe sidcle.
Textes reunis par Xavier Galmiche et Lenka Stranka, Prague, Edition Barenreiter
Praha, 2004, pp. 169-176.
[304] JEDRZEJEWSKI (Pranck), ed. Une philosophie dialectique de Vart musical, Ivan
Wyschnegradsky, Paris, L'Harmattan, 2005.
[305] JEDRZEJEWSKI (Pranck), "Permutation Groups and Chord Tessellations", Procee-
dings of the International Computer Music Conference, Barcelona, 2005, pp. 231-
234.
[306] JONES (Evan), "Residue-Class Sets in the Music of Iannis Xenakis: An Analytical
Algorithm and a General Intervallic Expression", Perspectives of New Music, 39(2),
2001, pp. 229-261.
[307] JOHNSON (Tom), Self-Similar Melodies, Paris, Editions 75, 1996.
[308] JOHNSON (Tom), "Tiling the line. Self-Replicating Melodies, Rhythmic Canons,
and an Open Problem", Actes des Huiti&mes Journe'es d'Informatique Musicale,
Bourges, 2001, pp. 147-152.
[309] JOHNSON (Tom), Tiling the line in Theory and in Practice, MaMuX Seminar (Ma-
thematiques, Musique et relations avec d'autres disciplines), Ircam, February 2002.

314
Bibliography

[310] JOHNSON (Tom), Some Observations on Tiling Problems, MaMuX Seminar (Ma-
thematiques, Musique et relations avec d'autres disciplines), Ircam, Januay 2003.
[311] JOHNSON (Tom), "Tiling Melodies, Tiling Chords", Computer Modeling and Re-
trieval, 2003, LIRMM, CNRS, University of Montpellier.
[312] JOHNSON (Tom), Perfect Rhythmic Tilings, MaMuX Seminar (Mathematiques,
Musique et relations avec d'autres disciplines), Ircam, January 2004.
[313] JOHNSON (Timothy A.), "Harmonic Vocabulary in the Music of John Adams: A
Hierarchical Approach", Journal of Music Theory, 37(1), 1993, pp. 117-156.
[314] JOHNSTON (Ben), "Proportionality and Expanded Pitch Resources", Perspectives
of New Music, 5(1), 1966, pp. 112-120.
[315] JOHNSTON (Ben), "Rational Structure In Music", American Society of University
Composers Proceedings, 1-2, 1976-1977, pp. 102-118.
[316] JORGENSEN (Owen H.), Tuning, Michigan State University Press, 1991.
[317] KAMEOKA (Akio), K N O P O F F (Leon), "Consonance Theory I: Consonance of
dyads", Journal of the Acoustical Society of America, 45(6), 1969, pp. 1452-1459.
[318] KAMEOKA (Akio), K N O P O F F (Leon), "Consonance Theory II: Consonance of com-
plex tones and its calculation method", Journal of the Acoustical Society of Ame-
rica, 45(6), 1969, pp. 1460-1469.
[319] KARKOSCHKA (Erhard), Das Schriftbild der neuen Musik, Celle, Moeck, 1966.
[320] KARKOSCHKA (Erhard), "Adriaan D. F O K K E R ' Neue Musik mit 31 Tonen", Die
Musikforschung, XXII, 1969, pp. 115-116.
[321] KHRUZIN (Andrei), "Enumeration of chord diagrams", preprint ArXiv,
math.CO/0008209, 2000.
[322] KLEBAN (Peter), OZLUK (Ali), "A Farey Fraction Spin Chain", Communication in
Mathematical Physics, 203, 1999, pp. 635-647.
[323] KLEMM (Michael), Symmetrien von Ornamenten und Kristallen, Berlin, Springer,
1982.
[324] KLUMPENHOUWER (Henry), A Generalized Model of Voice Leading for Atonal Mu-
sic, Ph.D. diss., Harvard University, 1991.
[325] KLUMPENHOUWER (Henry), "The Cartesian Choir", Music Theory Spectrum,
14(1), 1992, pp. 15-37.
[326] KLUMPENHOUWER (Henry), "Some Remarks on the Use of Riemann Transforma-
tions", Music Theory Online, 0(9), 1994.
[327] KLUMPENHOUWER (Henry), "The Inner and Outer Automorphisms of Pitch-Class
Inversion and Transposition: Some Implications for Analysis with Klumpenhouwer
Networks", Integral, 12, 1998, pp. 81-93.
[328] KLUMPENHOUWER (Henry), "Dualist Tonal Space and Transformation in
Nineteenth-Century Musical Thought", in Christensen, The Cambridge History
of Western Music Theory, Cambridge University Press, 2002, pp. 456-476.
[329] KOLMAN (Oren), "Transfert principles for generalized interval systems", Perspec-
tives of New Music, 41(2), 2003.
[330] KORNERUP (Thorwald), Das Goldene Tonsystem, Aschehoug Forlag, Kopenhagen,
1935.

315
Bibliography

[331] KORNERUP (Thorwald), Die Vorldufer der gleichschwebende Temperaturen mit 19


oder 31 T'onen in der Oktave, translated from Danish to German by P. Priedrich
Paulsen and J. Joergensen, Kopenhagen, 1930.
[332] KRAEHENBUEHL (David), CONS (Edgar), "Information as a measure of structure
in music", Journal of Music Theory, 2(2), 1958, pp. 127-161.
[333] KRANTZ (Richard J.), D O U T H E T T (Jack), "A measure of the reasonableness of
equal-tempered musical scales", Journal of the Acoustical Society of America,
95(6), 1994, pp. 3642-3650.
[334] KRENEK (Ernst), Uber Neue Musik. Sechs Vorlesungen zur Einfuhrung in die theo-
retischen Grundlagen, Universal Edition, Wien, 1937. English Translation: Music
here and now, Norton, New York, 1939.
[335] KRENEK (Ernst), Studies in Couterpoint based on the Twelve-Tone Technique,
Schirmer, 1940.
[336] KRENEK (Ernst), "New Developments of the Twelve-Tone Technique", The Music
Review, IV(2), 1943, pp. 81-97.
[337] KRENEK (Ernst), "Extents and limits of serial techniques", The Musical Quarterly,
46(2), 1960, pp. 210-232.
[338] KRENEK (Ernst), Horizons Circles. Reflections on my Music, University of Cali-
fornia Press, Berkeley, 1974.
[339] KRUMHANSL (Carol), Cognitive foundations of musical pitch, New-York, Oxford
University Press, 1990.
[340] KRUMHANSL (Carol), "Perceived Triad Distance: Evidence Supporting the Rea-
lity of Neo-Riemannian Transformations", Journal of Music Theory, 42(2), 1998,
pp. 265-281.
[341] LAMBERT (Philip), "Isographies and Some Klumpenhouwer Networks They In-
volve", Music Theory Spectrum, 24(2), 2002, pp. 165-195.
[342] LATTARD (Jean), Gammes et temperaments musicaux, Paris, Masson, 1988.
[343] LAWVERE (F.W.), "An elementary theory of the category of sets", Proc. Natl.
Acad. Sci., 52, 1964, pp. 1506-1511.
[344] LEIBOWITZ (Rene), Introduction a la musique de douze sons, Paris, L'Arche, 1949.
[345] LEMAN (Marc), Music, Gestalt, and Computing: Studies in Cognitive and Syste-
matic Musicology, Springer Verlag Lecture Notes in Artificial Intelligence, 1317,
1997.
[346] LENDVAI (Erno), Symmetries in Music, Kodaly Institute, Kecskemet, 1993.
[347] LERDAHL (Fred), "Timbral Hierarchies", Contemporary Music Review, 2(1), 1987,
pp. 135-160.
[348] LERDAHL (Fred), JACKENDOFF (Ray), A Generative Theory of Tonal Music, MIT,
1983.
[349] LEVY (Fabien), "Plaidoyer pour une oreille subjective et partisane. Une approche
pythagoricienne de la perception culturelle des intervalles", L'Harmattan, Cahiers
de philosophic du langage, 3, 1998, pp. 43-71.
[350] LEWIN (David), "IntervaUic Relations between Two Collections of Notes", Journal
of Music Theory, 3(2), 1959, pp. 298-301.

316
Bibliography

[351] LEWIN (David), "The Intervallic Content of a Collection of Notes", Journal of


Music Theory, 4(1), 1960, pp. 98-101.
[352] LEWIN (David), "Forte's Interval Vector, My Interval Function and Regener's Com-
mon Note Function", Journal of Music Theory, 21(2), 1977, pp. 194-237.
[353] LEWIN (David), "A Formal Theory of Generalized Tonal Functions", Journal of
Music Theory, 26(1), 1982, pp. 23-60.
[354] LEWIN (David), Generalized Musical Intervals and Transformations, New Haven,
Yale University Press, 1987.
[355] LEWIN (David), "Klumpenhouwer Networks and Some Isographies That Involve
Them", Music Theory Spectrum, 12(1), 1990, pp. 83-121.
[356] LEWIN (David), Musical Form and Transformation: Four Analytical Essays, New
Haven, Yale University Press, 1993.
[357] LEWIN (David), "A Tutorial on Klumpenhouwer Networks, Using the Chorale in
Schoenberg's Op. 11 No. 2", Journal of Music Theory, 38(1), 1994, pp. 79-101.
[358] LEWIN (David), "Cohn Functions", Journal of Music Theory, 40(2), 1996, pp. 181-
216.
[359] LEWIN (David), "Special cases of the interval function between pitch-class sets X
and Y", Journal of Music Theory, 45(1), 2001, pp. 1-30.
[360] LINDLEY (Mark), TURNER-SMITH (Ronald), Mathematical Models of Musical
Scales: A New Approach, Orpheus-Schriftenreihe zu Grundfragen der Musik 66,
Verlag fur systematische Musikwissenschaft, Bonn, Germany, 1993.
[361] LUSSON (Pierre), Musique et mathematiques, Paris, Alea, 1997.
[362] M A C LANE (Saunders), Categories for the Working Mathematician, New York,
Springer, 1971.
[363] M A C LANE (Saunders), MOERDIJK (Ieke), Sheaves in Geometry and Logic. A First
Introduction to Topos Theory, New York, Springer, 1994.
[364] MAEDEL (Rolf), Mikrotone, Aufbau, Klangwert, Beziehungen. Innsbruck, Helbling,
1983.
[365] MAEDEL (Rolf), "Transformation", Mikrotone II, Innsbruck, Helbling, 1988,
pp. 105-114.
[366] MAEGAARD (Jan), "The Nomenclature of Pitch-Class Sets and the Teaching of
Atonal Theory", Journal of Music Theory, 29(2), 1985, pp. 299-314.
[367] MALT (Mikhail), Les Mathematiques et la Composition assists par Ordinateur
(Concepts, Outils, Modules), Paris, PhD thesis, Ehess, 2001.
[368] MANCINI (David), "Twelve-Tone Polarity in Late Works of Luigi Dallapiccola",
Journal of Music Theory, 30, 1986, pp. 203-224.
[369] MANDELBAUM (Joel), Multiple Division of the Octave and the Tonal Resources of
the 19-Tone Equal Temperament, PhD thesis, University of Indiana, 1961.
[370] MANTUROV (Vassily), Knot Theory, Chapman & Hall, CRC Press Company, 2004.
[371] MARTINO (Donald), "The Source Set and Its Aggregate Formations", Journal of
Music Theory, 5(2), 1961, pp. 224-273.
[372] MASON (Robert), "Enumeration of Synthetic Musical Scales", Journal of Music
Theory 14(1), 1970, pp. 92-126.

317
Bibliography

[373] MAZZOLA (Guerino), Gruppen und Kategorien in der Musik. Entwurf einer ma-
thematischen Musiktheorie, Berlin, Heldermann Verlag, 1985.
[374] MAZZOLA (Guerino), "ObertSne oder Symmetrie: Was ist Harmonie?", Mi-
krotone 1, Innsbruck, Helbling, 1986, pp. 191-210.
[375] MAZZOLA (Guerino), "Konsonanz-Dissonanz und verborgene Symmetrien", Mi-
krotone 2, Innsbruck, Helbling, 1988, pp. 95-104.
[376] MAZZOLA (Guerino), Geometrie der Tone, Basel, Birkhauser, 1990.
[377] MAZZOLA (Guerino), Mathematical Music Theory: An Informal Survey, Locarno,
Edizioni Cerfim, 1993.
[378] MAZZOLA (Guerino), "Semiotic Aspects of Musicology: Semiotics of Music" in
R. Posner et al. A Handbook of the Sign-Theoretic Foundations of Nature and
Culture, Berlin, Walter de Gruyter, 1998.
[379] MAZZOLA (Guerino), The Topos of Music, Basel, Birkhauser, 2002.
[380] MAZZOLA (Guerino), "Mathematical Music Theory - Status Quo 2000", in Maz-
zola, Noll and Lluis-Puebla, eds. Perspectives in Mathematical and Computational
Music Theory, Epos, 2004, pp. 42-77.
[381] MAZZOLA (Guerino), "Towards a Galois Theory of Concepts", in Mazzola, Noll and
Lluis-Puebla, eds. Perspectives in Mathematical and Computational Music Theory,
Epos, 2004, pp. 78-88.
[382] MAZZOLA (Guerino), NOLL (Thomas), LLUIS-PEBLA (Emilio), eds. Perspectives in
Mathematical and Computational Music Theory, Electronic Publishing Osnabrtick,
2004.
[383] MCLAREN (Brian), "General Methods of Generating Musical Scales", Xenharmo-
nikdn, 13, 1990, pp. 20-44.
[384] MCLAREN (Brian), "Those Difficult Teens (and Other Microtonal Scales)", Xen-
harmonikdn, 14, 1991, pp. 3-6.
[385] MCLAREN (Brian), "A Nostalgic Look At the 20s - and Other Musical Scales",
Xenharmonikdn, 14, 1991, pp. 6-8.
[386] MCLAREN (Brian), "The Low Profile of Those Higher Divisions of the Octave",
Xenharmonikdn, 14, 1991, pp. 8-11.
[387] MCLAREN (Brian), "The Uses and Characteristics of Non-Octave Scales", Xenhar-
monikdn, 14, 1993, pp. 12-22.
[388] M C L A R E N (Brian), "The Uses and Characteristics of Non-Just Non-Equal-
Tempered Scales", Xenharmonikdn, 15, 1993, pp. 27-41.
[389] MEAD (Andrew W.), "Some Implications of the Pitch-Class Order-Number Iso-
morphism Inherent in the Twelve-Tone System", Perspectives of New Music, 26(2),
1988, pp. 96-163.
[390] MEAD (Andrew W.), An Introduction to the Music of Milton Babbitt, Princeton
University Press, 1994.
[391] MEEOS (Nicolas), Heinrich Schenker. Une introduction, Bruxelles, Mardaga, 1993.
[392] MEEtJS (Nicolas), "Vecteurs harmoniques", Musurgia , 10(3-4), 2003, pp. 7-34.
[393] MENAGE (Marcel), "La Set Complex Theory: de quels enjeux s'agit-il?", Analyse
musicale, 17, 1989, pp. 87-90.

318
Bibliography

[394] MENAGE (Marcel), "Sur la modalisation des partitions musicales", Analyse musi-
cale, 22, 1991, pp. 31-46.
[395] MENAGE (Marcel), R I O T T E (Andre). "Modelisation informatique des partitions,
analyse et composition assistee", Les Cahiers de VIRCAM, 3, 1993, pp. 43-59.
[396] MENDEL (Arthur), "Pitch in the 16 th and early 17 th Centuries", Musical Quarterly,
34, 1948, pp. 28-45, pp. 199-221, pp. 336-357, pp. 575-593.
[397] MENDEL (Arthur), "On the Pitches in use in Bach's Time", Musical Quarterly, 41,
1955, pp. 332-354, pp. 466-480.
[398] MENDEL (Arthur), "Pitch in Western Music since 1500: a re-examination", Acta
Musicologica, 50, 1978, pp. 1-93.
[399] MERSENNE (Marin), Correspondances, Paris, CNRS, Edited by C. de Waard, 1932-
1945.
[400] MERSENNE (Marin), Harmonie Universelle, Sebastien Cramoisy, Paris 1636-37.
Reprinted by CNRS 1975.
[401] MERSENNE (Marin), Cogitata physico-mathematica, Paris, Bertier, 1644.
[402] MERSENNE (Marin), Harmonicorum Libri XII, Paris, Bauldry, 1648.
[403] MESSIAEN (Olivier), Techniques de mon langage musical, Paris, Alphonse Leduc,
1944.
[404] MESSIAEN (Olivier), Traite de Rythme, de couleur et d'omitologie, Paris, Alphonse
Leduc, 7 vols, 1992.
[405] MlLHAUD (Darius), "Polytonalite et atonalite", Revue Musicale, 4, 1923, reprinted
in Notes sur la musique: essais et chroniques, textes reunis et presentes par Jeremy
Drake, Paris, Flammarion, 1982, pp. 173-189.
[406] MORENO (Enrique), Expanded Tunings in Contemporary Music: Theoretical Inno-
vations and Practical Applications, Edwin Mellen Press, 1992.
[407] MORRIS (Robert), "Set Groups, Complementation, and Mappings Among Pitch-
Class Sets", Journal of Music Theory, 26(1), 1982, pp. 101-144.
[408] MORRIS (Robert), Composition with pitch-classes: a theory of compositional design,
New Haven, Yale University Press, 1987.
[409] MORRIS (Robert), "Pitch-Class Complementation and its Generalizations", Jour-
nal of Music Theory, 34(2), 1990, pp. 175-245.
[410] MORRIS (Robert), "Equivalence and Similarity in Pitch and their Interaction with
Pcset Theory", Journal of Music Theory, 39(2), 1995, pp. 207-243.
[411] MORRIS (Robert), "K, Kh and Beyond", in J. Baker et al, Music Theory in Concept
and Practice, 1997.
[412] MORRIS (Robert), "Voice-Leading Spaces", Music Theory Spectrum, 20(2), 1998,
pp. 176-208.
[413] MORRIS (Robert), Class Notes for Advanced Atonal Music Theory, 2 vol., Prog
Peak Music, 2001.
[414] MORRISON (Gary), "Decimal Music", Xenharmonikdn, 7-8, 1979, pp. 1-4.
[415] MORRISON (Gary), "Ten Equal: The Other Side of the Coin", Interval, 2(1), 1979,
pp. 7-8.

319
Bibliography

[416] MORSE (Philip), INGARD (K. Uno), Theoretical Acoustics, Mc Graw Hill, 1968,
reprinted by Princeton University Press, 1986.
[417] MUZZULINI (Daniel), "Leonhard Eulers Konsonanztheorie", Musiktheorie 9(2),
Laaber Verlag, Laaber, Germany, 1994.
[418] MUZZULINI (Daniel), "Musical Modulation by Symmetries", Journal of Music
Theory, 39(2), 1995, pp. 311-325.
[419] NAUERT (Paul), "Field Notes: A Theory of Fixed-Pitch Formations", Perspectives
of New Music, 41(1), 2003, pp. 180-239.
[420] NEDERVEEN (Cornelius Johannes), Acoustical Aspects of Woodwind Instruments,
Northern Illinois Press, 1998.
[421] NEIDHARDT (Johann Georg), Gdntzlich erschopfte mathematische Abtheilungen
des diatonisch-chromatischen, temperirten Canonis Monochordi, Kftnisgberg, 1732.
[422] NEIDHARDT (Johann Georg), Systema generis diatonici chromatid exnumeris serie
naturali procedentibus, Ktoigsberg, 1734.
[423] NETSKE (Andreas), NOLL (Thomas), "Inner metric analysis", Tatra Mountains
Mathematical Publications, 23, 2001, pp. 91-111.
[424] NEUWIRTH (Erich), Musical Temperaments, Springer Verlag, 1997.
[425] NEWMAN (D. J.), "Tesselation of integers", Journal of Number Theory, 9, 1977,
pp. 107-111.
[426] NICOLAS (Frangois), "Le feuillete du tempo. Essai sur les modulations metriques",
Paris, Entretemps, 9, 1990, pp. 51-77.
[427] NICOLAS (Frangois), "Quand Palgebre mathematique aide a penser (et pas seule-
ment & calculer) la combinatoire musicale", MaMuX Seminar (Mathematiques,
Musique et relations avec d'autres disciplines), Ircam, 2003.
[428] NOLAN (Catherine), "Structural Levels and Twelve-Tone Music: A Revisionist
Analysis of the Second Movement of Webern's Piano Variations Opus 27", Journal
of Music Theory, 39(1), 1995, pp. 47-76.
[429] NOLL (Thomas), Morphologische Grundlagen der abendldndischen Harmonik, Doc-
toral Thesis, TU Berlin, 1995, publ. in Musikometrika, 7, 1997.
[430] NOLL (Thomas), "Harmonische Morpheme", Musikometrika 8, 1997, pp. 7-32.
[431] NOLL (Thomas), "Geometry of Chords", in Puebla, Emilio Lluis, ed. Electronic
Bulletin of the Sociedad Matematica Mexicana, 1, 2001, to appear.
[432] NOLL (Thomas), "Tone Apperception and Weber-Fechner's Law", MaMuX Semi-
nar (Mathematiques, Musique et relations avec d'autres disciplines), Ircam, 2002.
[433] NOLL (Thomas), "Vade Mecum to Mathematical Music Theory", in Mazzola, Noll
and Lluis-Puebla, eds. Perspectives in Mathematical and Computational Music
Theory, Epos, 2004, pp. 14-28.
[434] NOLL (Thomas), BRAND (Monica), "Morphology of Chords", in Mazzola, Noll and
Lluis-Puebla, eds. Perspectives in Mathematical and Computational Music Theory,
Epos, 2004, pp. 366-398.
[435] NOLL (Thomas), GARBERS (J5rg), "Harmonic Path Analysis", in Mazzola, Noll
and Lluis-Puebla, eds. Perspectives in Mathematical and Computational Music
Theory, Epos, 2004, pp. 399-431.

320
Bibliography

[436] NORDEN (N. Lindsey), "A New Theory of Untempered Music", The Musical Quar-
terly 22(2), 1936, pp. 217-233.
[437] NORDEN (N. Lindsey), "A New Study of Intervals", Organ Institute Quarterly,
5(1), 1955, p. 31.
[438] NULL (Edwin von der), Moderne Harmonik, Leipzig, Pr. Kistner und C.F.W. Siegel,
1932.
[439] OBOUHOV (Nicolas), "Le livre de vie", Paris, La Revue Musicale, 290-291, 1972,
pp. 1-73.
[440] OETTINGEN (Arthur von), Harmoniesystem in Dualer Entwicklung: Studien zur
Theorie der Musik, Leipzig, W. Glaser, 1866.
[441] OTTERSTROM (Thorvald), A Theory of modulation, University of Chicago, 1935.
[442] OzONAM (Jacques), MONTUCLA (Jean-Etienne), Recreations mathematiques et
physiques, Paris, 1790.
[443] PAPADOPOULOS (Athanase), "Mathematics Music and J.S. Bach", Rend. Sem. Fac.
Sci. Univ. Caligari, 70(2), 2000, pp. 1-20.
[444] PAPADOPOULOS (Athanase), "Mathematics and music theory: from Pythagoras to
Rameau", Math. Intelligencer, 24(1), 2002, pp. 65-73.
[445] PAPADOPOULOS (Athanase), "The Mathematics of the Music of Olivier Messiaen",
Lett. Mat. Pristem., 47, 2003, pp. 27-41.
[446] PAPADOPOULOS (Athanase), "Consonance musicale et complexity mathematique",
LVuvert, 112, 2005, pp. 47-68.
[447] PARTCH (Harry), Genesis of a Music: An Account of a Creative Work, its Roots
and its Fulfillment, 2nd enl. edition, Da Capo Press, New York, 1974. Reprint,
1979. First edition 1949.
[448] PARZYSZ (Bernard), Musique et mathematique, Publication de PA.P.M.E.P., 53,
Paris, 1984.
[449] PECK (Robert W.), "Toward an Interpretation of Xenakis's Nomos Alpha", Pers-
pectives of New Music, 41(1), 2003, pp. 106-157.
[450] PECK (Robert W.), "Klein-Bottle Tonnetze", Music Theory Online, 9(3), 2003.
[451] PERLE (Georg), Serial Composition and Atonality, Los Angeles, University of Ca-
lifornia Press, 1962.
[452] PERLE (Georg), "Pitch Class Set Analysis: An Evaluation", Journal of Musicology,
8(2), 1990, pp. 151-172.
[453] PERRON (Oskar), "Grundlagen fur eine Theorie des Jacobischen Kettenbruchalgo-
rithmus", Math. Ann., 64, 1907, pp. 1-76.
[454] P I E R C E (John R.), The Science of Musical Sound, New York, Scientific American
Books, 1983.
[455] P I E R C E (John R.), MATHEWS (Max), "Control of Consonance and Dissonance with
Nonharmonic Overtones", Music by Computers, von FOrster and Beauchamp eds,
New York, 1969, pp. 129-132.
[456] PlKLER (Andrew G.), "History of Experiments on the Musical Interval Sense",
Journal of Music Theory, 10(1), 1966, pp. 54-95.
[457] PlRRO (Andr6), Descartes et la musique, Paris, 1907. New edition, 1974.

321
Bibliography

[458] PLOMB (R.), LEVELT (W.), "Tonal Consonance and Critical Bandwith", Journal
Acoustical Society of America, 38, 1965, pp. 548-560.
[459] POLANSKY (Larry), "Notes of a Lousy Carpenter", One/one, 1(1), 1985.
[460] POLANSKY (Larry), "The HMSL Intonation Environment", One/one, 3(1), 1987,
pp. 4-12.
[461] POLANSKY (Larry), "Paratactical Tuning: An Agenda for the Use of Computers
in Experimental Intonation", Computer Music Journal, 11(1), 1987, pp. 61-68.
[462] POLE (William), The Philosophy of Music, London, TrUbner and Co, 1879.
[463] P6LYA (George), "Kombinatorische Anzahlbestimmungen fur Gruppen, Graphen
und chemische Verbindungen", Acta Mathematica, 68, 1937, pp. 145-254.
[464] P6LYA (George), READ (R. C), Combinatorial Enumeration of Groups, Graphs,
and Chemical Compounds, Springer Verlag, 1987.
[465] POSNER (Roland) et al, Semiotics: A Handbook of the Sign-Theoretic Foundations
of Nature and Culture, Berlin, Walter de Gruyter, 1998.
[466] QuiNN (Ian), "Listening to Similarity Relations", Perspectives of New Music, 39,
2001, pp. 108-158.
[467] RAHN (John), Basic Atonal Theory, New York, Longman, 1980.
[468] RAHN (John), "On Some Computational Models of Music Theory", Computer Mu-
sic Journal, 4(2), 1980, pp. 66-72.
[469] RAHN (John), "Notes on Methodology in Music Theory", Journal of Music Theory,
33(1), 1989, pp. 143-154.
[470] RAHN (John), "New Research Paradigms", Music Theory Spectrum, 2(1), 1989,
pp. 84-94.
[471] RAHN (John), Music Inside Out Going too far in musical essays, Overseas Publi-
sher Association, 2001.
[472] RAICHEL (Daniel R.), The Science and Applications of Acoustics, Amer. Inst, of
Physics, 2000.
[473] RAMEAU (Jean Philippe), Traits de Vharmonie rMuite a ses principes naturels,
Paris, 1722.
[474] RAMEAU (Jean Philippe), Nouveau systeme de musique tMorique et pratique, pour
servir d'introduction au traite de VHarmonie, Paris, Ballard, 1726.
[475] RAMEAU (Jean Philippe), Generation Harmonique ou traite de musique tMorique
et pratique, Paris, Prault, 1737.
[476] RAMEAU (Jean Philippe), Demonstration duprincipe de Vharmonie, Paris, Durand
et Pissot, 1750.
[477] RAPOPORT (Paul), "The Structural Relationships of Fifths and Thirds in Equal
Temperaments", Journal of Music Theory, 37(2), 1993, pp. 351-389.
[478] RAPOPORT (Paul), "Just Shape, Nothing Central", MusicWorks, 60, 1994,
pp. 42-49.
[479] RAPOPORT (Paul), "Gimme Fives", MusicWorks 61, 1995, pp. 40-45.
[480] RASCH (Rudolph), "Wohltemperit en Gelijkzwevend", Mens and Melodie, 36,
Utrecht, 1981, pp. 264-273

322
Bibliography

[481] RASCH (Rudolph), "Gelijkzwevend stemmen", Mens and Melodie, 37, Utrecht,
1982, pp. 14-30.
[482] RASCH (Rudolph), "Description of Regular Twelve-Tone Musical Tunings", Jour-
nal of the Acoustical Society of America, 73(3), 1983, pp.1023-1035.
[483] RASCH (Rudolph), "Priedrich Suppig and His Labyrinthus musicus of 1722", The
Organ Yearbook, XV, 1984, pp. 33-53.
[484] RASCH (Rudolph), "Does Well-Tempered Mean Equal-Tempered ?" in Bach, Han-
del, Scarlatti - Tricentenary Essays, Peter Williams, Cambridge University Press,
1985, pp. 293-310.
[485] RASCH (Rudolph), "Relations between Multiple Divisions of the Octave and the
Traditional Tonal System", Interface, 14, 1985, pp. 75-108.
[486] RASCH (Rudolph), "Integer Approximation of Musical Scale Frequencies", Inter-
face, 15, 1986, pp. 25-34.
[487] RASCH (Rudolph), ed. Corpus Microtonale, vol. 1: Adriaan Daniel Fokker - Selected
Compositions, with introduction by Rudolf Rasch, Utrecht, 1987.
[488] RASCH (Rudolph), "The Theories of Adriaan Fokker, Part One", One/one, 4(1),
1988.
[489] RASCH (Rudolph), "The Theories of Adriaan Fokker, Part Two", One/one, 4(2),
1988.
[490] RAYLEIGH (John William), The Theory of Sound, London, 2 vols., Macmillan,
Second Edition, 1894.
[491] READ (Ronald C ) , "Combinatorial problems in the theory of music", Discrete
Mathematics, 167-168(1-3), 1997, pp. 543-551.
[492] REGENER (Eric), "The Number Seven in the Theory of Intonation", Journal of
Music Theory, 19(1), 1975, pp. 140-135.
[493] REINER (David), "Enumeration in Music Theory", Amererican Mathematical
Monthly, 92, 1985, pp. 51-54.
[494] R E T I (Rudolph), Tonality, Atonality, Pantonality: a study of some trends in Twen-
tieth Century music, New York, Macmillan, 1958.
[495] RlCHTER-HERF (Franz), ed. Mikrotone: Bericht ilber das 1. Internationale Sym-
posium Mikrotonforschung, Musik mit Mikrotonen, Innsbruck, Helbling, 1986.
[496] R I C H T E R - H E R F (Franz), ed. Mikrotone II: Bericht iiber das 2. Internationale Sym-
posium Mikrotonforschung, Musik mit Mikrotonen, Innsbruck, Helbling, 1988.
[497] R I C H T E R - H E R F (Franz), ed. Mikrotone III: Bericht iiber das 3. Internationale Sym-
posium Mikrotonforschung, Musik mit Mikrotonen, Innsbruck, Helbling, 1990.
[498] R I C H T E R - H E R F (Franz), ed. Mikrotone IV: Bericht iiber das 4- Internationale Sym-
posium Mikrotonforschung, Musik mit Mikrotonen (2-5 May 1991 in Salzburg),
Musikedition Nymphenburg, 1993.
[499] RlEMANN (Hugo), Musikalische Logik: Hauptziige der physiologischen und psycho-
logischen Begriindung unseres Musiksystems, Leipzig, C. Kahnt, 1873.
[500] RlEMANN (Hugo), Musikalische Syntaxis: Grundrifi einer harmonischen Satzbil-
dungslehre, Leipzig, Breitkopf und Hartel, 1877.
[501] RlEMANN (Hugo), Skizze einer Neuen Methode der Harmonielehre, Leipzig, Breit-
kopf und Hartel, 1880.

323
Bibliography

[502] RlEMANN (Hugo), Grosse Kompositionslehre, vol 1: Der homophone Satz (Melo-
dielehre und Harmonielehre), Berlin, Max Hesse, 1902.
[503] RlEMANN (Hugo), Handbuch der Harmonielehre, Leipzig, Breitkopf und Hartel,
5/1912, 6/1918.
[504] RlOTTE (Andre), Generation des cycles equilibres, Ispra, Euratom, Rapport interne
353, 1963.
[505] RlOTTE (Andre), Formalisation des structures musicales, PhD Thesis, Universite
Paris VIII, Departement d'informatique, Paris, 1979.
[506] RlOTTE (Andre), Formalisation des structures musicales, Lectures, Universite Pa-
ris VIII, Departement d'informatique, Paris, 1978-1990.
[507] RlOTTE (Andre), "Formalisation des echelles de hauteurs en analyse et en compo-
sition", Musique et assistance informatique, MIM, Marseille, 1990.
[508] RlOTTE (Andre), "L'utilisation des modeles mathematiques en analyse et en com-
position musicale, Quadrivium, Musique et Sciences, IPMC, Paris, 1992.
[509] RlOTTE (Andre), "Organisations du temps autour de la serie, de Webern a Boulez",
Les cahiers de philosophic, 20, La loi musicale, Lille, 1996.
[510] RlOTTE (Andre), "Mathematique du son, musique du nombre", in Genevois and
Orlarey, Musique et mathematiques, Aleas, 1997, pp. 11-25.
[511] RlOTTE (Andre), MESNAGE (Marcel), Formalismes et modules, Collection Mu-
sique/Sciences, Editions Delatour/Ircam, 2006.
[512] RlSSET (Jean-Claude), "Paradoxes de hauteurs", Proceedings of the 1th Interna-
tional Congress of Acoustics, Budapest, 1971.
[513] RlSSET (Jean-Claude), "Hauteur et timbre", Revue d'Acoustique, 42,1977, pp. 263-
268.
[514] RlSSET (Jean-Claude), "Musique, recherche, theorie, espace, chaos", Paris, Ircam,
Inharmoniques, 8-9, 1991, pp. 272-316.
[515] ROSCH (Eleanor), "Family resemblances: studies in internal structure of catego-
ries", Cognitive Psychology, 7, 1975, pp. 573-605.
[516] ROEDER (John), A Theory of Voice Leading for Atonal Music, Ph.D. diss., Yale
University, 1984.
[517] ROEDER (John), "A Geometric Representation of Pitch-Class Series", Perspectives
of New Music, 25, 1987, pp. 362-409.
[518] ROEDER (John), "A MaMuTh Achievement", Perspectives of New Music, 31(2),
1993, pp. 294-312.
[519] ROGERS (David W.), "A Geometric Approach to Pcset Similarity", Perspectives
of New Music, 37(1), 1999, pp. 77-90.
[520] ROMANOWSKA (Anna), "Algebraic Structure of the Tone System", Demonstratio
Mathematica, 7(4), 1974, pp. 525-542.
[521] ROMANOWSKA (Anna), "Algebraic Language in the Theory of Harmony", Tatra
Mt. Math. PubL, 23, 2001, pp. 113-123.
[522] ROSSING (Thomas D.), The Science of Sound, Addison-Wesley, Second Edition,
1990.
[523] ROSSING (Thomas D.), Science of Percussion Instruments, World Scientific, 2000.

324
Bibliography

[524] ROTHENBERG (David), "A Mathematical Model for Perception Applied to the
Perception of Pitch", in G. Goos and J. Harmanis, Lecture Notes in Computer
Science: Formal Aspects of the Cognitive Process, 1975, pp. 127-141.
[525] ROUSSEAU (Jean-Jacques), Dictionnaire de musique, Paris, 1767.
[526] SALMON (Thomas), A Proposal to Perform Musick in Perfect and Mathematical
Proportions, London, 1688.
[527] SALMON (Thomas), "The Theory of Music reduced to aritmetical and geometrical
proportions", Philosophical Transactions of the Royal Society of London, 24(302),
1705, pp. 2072-2100.
[528] SAMSON (Jim), Music in Transition: a study of tonal expansion and atonality,
1900-1920, New York, W.W. Norton, 1977.
[529] SANDS (Arthur D.), "On the factorization of finite abelian groups", Acta Math.
Acad. Sci. Hungar., 8, 1957, pp. 65-86.
[530] SANDS (Arthur D.), "The Factorisation of Abelian Groups", Quart. J. Math. Ox-
ford, 2(10), 1959, pp. 81-91.
[531] SANDS (Arthur D.), "On the Factorisation of Abelian Groups II", Acta Math. Acad.
Sci. Hungar., 13, 1962, pp. 153-159.
[532] SANDS (Arthur D.), "On Keller's conjecture for certain cyclic groups", Proc. Edin-
burgh Math. Soc., 22(2), 1977, pp. 17-21.
[533] SAUVEUR (Joseph), TraitS de la theorie de la musique, Paris, 1697.
[534] SAUVEUR (Joseph), "Sur la determination d'un son fixe", Histoire de VAcademic
Royale des Sciences, Paris 1700. Reprinted by R. Rasch, The Diapason Press,
Utrecht, 1988.
[535] SAUVEUR (Joseph), "Rapport des sons des cordes des instruments de musique, aux
fleches des cordes et nouvelle determination des sons fixes", Histoire de VAcad&mie
Royale des Sciences, Annee 1713, Paris 1716. Reprinted by R. Rasch, The Diapason
Press, 1988, pp. 324-349.
[536] SAUVEUR (Joseph), "Sur Papplication des sons harmoniques aux jeux de POrgue",
Histoire de VAcademie Royale des Sciences, Paris, Hachereau, 1720. Reprinted by
R. Rasch, The Diapason Press 1988.
[537] SAUVEUR (Joseph), "Methode generate pour former des systemes temperes de mu-
sique", Histoire de VAcademie Royale des Sciences, Annee 1707, Paris, Martin,
1730, pp. 203-222. Reprinted by R. Rasch, The Diapason Press 1988.
[538] SAUVEUR (Joseph), "Table generate des systemes temperes de musique", Histoire
de VAcademie Royale des Sciences", Annee 1711, Paris 1730, pp. 307-316. Reprin-
ted by R. Rasch, The Diapason Press, Utrecht, 1988.
[539] SAUVEUR (Joseph), "Systeme general des intervalles et des sons et son applicalion
a tous les systemes et a tous les instruments de musique", Histoire de VAcad&mie
Royale des Sciences, Annee 1701, Paris 1743, pp. 299-966. Reprinted by R. Rasch,
The Diapason Press, 1988.
[540] SCABOROUGH (Don L.), "Connectionist Models for Tonal Analysis", Computer
Music Journal, 13, 1989, pp. 49-55.
[541] SCHAUB (Stephan), Algebraic Structures and Symmetry in Western Music, Master
Thesis, Department of Mathematics, University of Arizona, 1996.

325
Bibliography

[542] SCHAUB (Stephan), L'hypothese Mathematique. Musique symbolique et composi-


tion musicale dans Herma de Iannis Xenakis, Memoire de D.E.A., Universite de
Paris IV, 2001.
[543] SCHENKER (Heinrich), Theorien und Phantasien III: Derfreie Satz, Wien, Univer-
sal Edition, 1935.
[544] SCHILLINGER (Joseph), The Schilling er System of Musical Composition, New York,
Carl Fischer, 2 vol., 1941.
[545] SCHILLINGER (Joseph), The Mathematical Basis of the Art, The Philosophical Li-
brary, 1948.
[546] SCHNEIDER (Sigrun), Mikrotone in der Musik des 20. Jahrunderts, Band 15 der Or-
pheus, Verlag fur systematische Musikwissenschaft GmbH, Bonn, Bad Godesberg,
1975.
[547] SCHNITZLER (Giinther), Musik und Zahl, Verlag fur systematische Musikwissen-
schaft, Bonn-Bad Godesberg, 1976.
[548] SCHOENBERG (Arnold), Style and Idea: Selected Writings of Arnold Schoenberg,
Edited by Leonard Stein, New York, St. Martins Press, 1975.
[549] SCHOENBERG (Arnold), Harmonielehre (1911), Vienna, Universal Edition, 1966.
[550] SCHREIBER (Heinrich), Algoritmus proportionum una cum monochordi generis dya-
tonici compositione, Cracovie 1514.
[551] SCHWEIGER (Fritz), The metrical theory of the Jacobi-Perron Algorithm, Berlin,
Springer, Lecture Notes in Mathematics, 334, 1973.
[552] SCOTT (Damon), ISAACSON (Eric J.), "The Interval Angle: A similarity Measure
for Pitch-Class Sets", Perspective of New Music, 36(2), 1998, pp. 107-142.
[553] SETHARES (William A.), Tuning, Timbre, Spectrum, Scale, Springer Verlag, 1998.
[554] SETHARES (William A.), "Repetition and pseudo-periodicity", Tatra Mountains
Mathematical Publications, 23, 2001, pp. 125-140
[555] SIMS (Ezra), "Yet Another 72-Noter", Computer Music Journal, 12(4), 1988,
pp. 28-45.
[556] SlMS (Ezra), "Reflections on This and That: Perhaps A Polemic", Perspectives of
New Music, 39(1), 1992, pp. 236-257.
[557] SlMONTON (Theodore E), "A New integral Ratio Chromatic Scale", Journal of the
Acoustical Society of America, 25, 1953, pp. 1167-1175.
[558] SINN (Christoph Albert), Die aus mathematischen Griinden richtig gestellete mu-
sikalische Temperatura practica, Wernigeroda, 1717.
[559] SLAYMAKER (Prank H.), "Chords from tones having strched partials", Journal of
the Acoustical Society of America, 47, 1970, pp. 1569-1571.
[560] SLOBODA (John), The Musical Mind: An Introduction to the Cognitive Psychology
of Music, Oxford, Calderon Press, 1985.
[561] SLONIMSKY (Nicolas), "The Schillinger system of composition", The Musical Quar-
terly, 32, 1946, pp. 465-470.
[562] STARR (Daniel), "Sets, Invariance and Partitions", Journal of Music Theory, 22(1),
1978, pp. 1-42.

326
Bibliography

[563] STARR (Daniel), MORRIS (Robert), "A General Theory of Combinatoriality and
the Aggregate", Perspectives of New Music, 1977-78, 16(1), pp. 364-389 and 16(2),
pp. 50-84.
[564] STEIN (Sherman K.), SZAB6 (Sandor), Algebra and Tiling, The Carus Mathema-
tical Monographs, 25, 1994.
[565] STOEKER (Philip), "Klumpenhouwer Networks, Trichords and Axial Isography",
Music Theory Spectrum, 24(2), 2002, pp. 231-245.
[566] STRAUB (H.), Beitrage zur modultheoretischen Klassifikation musikalischer Motive,
Diplomarbeit ETH Zurich, 1989.
[567] STRAUS (Joseph Nathan), Introduction to post-tonal theory, Englewood Cliffs,
Prentice Hall, 1990.
[568] STUMPF (Carl), "Konsonanz und Dissonanz", Beitr. Akust. Musikwissenshaft, 1,
1898, pp. 1-108.
[569] SWIFT (Richard), "Some Aspects of Aggregate Composition", Perspectives of New
Music, 14(2) and 15(1), 1976, pp. 236-248.
[570] SZABO (Sandor), "A type of factorization of finite abelian groups", Discrete Math,
54, 1985, pp. 121-124.
[571] TAMBA (Akira), La th6orie et Vesthetique musicale japonaises, Paris, Publications
Orientalistes de Prance, 1988.
[572] TANAKA (Shohe), "Studien im Gebiete der reinen Stimmung", Vierteljahrsschrift
fur Musikwissenschaft, 6, 1890, pp. 1-90.
[573] TANGIAN (Andranik), "The Sieve of Erathosthene and Johnson's Problem on Ryth-
mic Canons", Discussion paper No. 309, Fern University of Hagen, 2001.
[574] TANNER (Robert), "Le r6le de la fusion dans Pappreciation de la justesse des ac-
cords", in Colloque Acoustique musicale, Paris, CNRS, 1959, pp. 49-59.
[575] TANNER (Robert), La musique antique grecque, expliquie par une consequence de
la tMorie psycharithmetique, Paris, Richard Masse, 1961.
[576] TANNER (Robert), "Critique de la theorie de la resonance", in La resonance dans
les €chelles musicales, Colloque du CNRS, Paris, 1963, pp. 63-100.
[577] TANNER (Robert), "Le probleme des deux tierces (Pythagore et Zarlino). Sa solu-
tion psycharithmetique et experimentale", Acustica, 27(6), 1972, pp. 335-352.
[578] TANNER (Robert), "La fonction et la justesse melodique des intervalles", Acustica,
34(5), 1976, pp. 259-268.
[579] TEITELBAUM (Richard), "Intervallic Relations in Atonal Music", Journal of Music
Theory, 9(1), 1965, pp. 72-127.
[580] TENNEY (James), A History of Consonance and Dissonance, New York, Excelsior
Music Pub., 1988.
[581] TERHARDT (Ernst), "Pitch, Consonance and Harmony", Journal of the Acoustical
Society of America, 55, 1974, pp. 1061-1069.
[582] TlJDEMAN (Robert), "Decomposition of the integers as a direct sum of two sub-
sets", Number Theory (Paris 1992-1993), London Math. Soc. Lecture Note Ser.
215, Cambridge University Press, 1995, pp. 261-276.
[583] TREU (Abdias), Directorium mathematicum ad cuius ductum et informationem
tota mathesis Harmonica, Altdorf, 1657.

327
Bibliography

[584] VERDI (Luigi), "Collections symetriques par transposition: Quelques observations",


MaMuX Seminar (Mathematiques, Musique et relations avec d'autres disciplines),
Ircam, 2002.
[585] V E R D I (Luigi), "History of Set Theory from an European point of view", paper
delivered at the International Symposium "Around Set Theory", Ircam, 2003 (to
appear).
[586] VERDI (Luigi), Caleidocicli musicali. Simmetrie infrante dei suoni, Milano, Rug-
ginenti, 2005.
[587] ViClNANZA (Domenico), "Paths on Stern-Brocot Tree and Winding Numbers of
Modes", Proceedings of the International Computer Music Conference, Barcelona,
2005, pp. 235-239.
[588] ViERU (Anatol), Cartea modurilor, Bucarest, Ed. muzicala, 1980. English transla-
tion The Book of Modes, 1993.
[589] ViERU (Anatol), "Modalism - A Third World", Perspectives of New Music, 24,
1985, pp. 62-71.
[590] ViERU (Anatol), "Generating Modal Sequences (A Remote Approach to Minimal
Music)", Perspectives of New Music, 30(2), 1992, pp. 178-200.
[591] VOGEL (Martin), Die Lehre von dem Tonbeziehungen, Verlag fur systematische
Musikwissenschaft, 1975.
[592] VOLK (Anja), "Modeling Processive Perspective on Meter in Music", Barcelona,
Proceedings of the International Computer Music Conference, 2005, pp. 251-254.
[593] VUZA (Dan Tudor), "Aspects mathematiques dans la theorie modale d'Anatol
Vieru", Parts 1-4, Revue Roumaine de Mathematiques Pures et Appliqu^es, 27(2)
and (10), 1982, and 28(7-8), 1983.
[594] VUZA (Dan Tudor), "Some Mathematical Aspects of David Lewin's Book Gene-
ralized Musical Intervals and Transformations", Perspectives of New Music, 26(1),
1988, pp. 258-287.
[595] VUZA (Dan Tudor), "Supplementary Sets and Regular Complementary Unending
Canons", Perspectives of New Music, Part 1, 29(2), Part 2, 30(1), Part 3, 30(2),
Part 4, 31(1), 1991-1993.
[596] VUZA (Dan Tudor), "Supplementary Sets - Theory and Algorithms", Muzica, 1,
1995, pp. 75-99.
[597] W E I (Wan-Di), Xu (Ju-Yong), "Cycle index of direct product of permutation
groups and number of equivalence classes of subsets of Z„", Discrete Mathema-
tics, 123, 1993, pp. 179-188.
[598] W E Y L (Hermann), Symmetry, Princeton University Press, 1952.
[599] WERCKMEISTER (Andreas), Musicae mathematicae hodegus curiosus, oder Richti-
ger musicalischer Weg-weiser, Quedlinburg 1686-1687, reprint, fac-sim. Olms, 1972.
[600] WERCKMEISTER (Andreas), Musicalische Temperatur, Quedlinburg 1691. Reprin-
ted by R. Rasch, Utrecht, The Diapason Press, 1983.
[601] WlBBERLEY (Roger), "Three Essays on Syntonic Tuning: Quid non Ebrietas Dis-
signat? Willaert's Didactic Demonstration of Syntonic Tuning; Syntonic Tuning:
A Sixteenth-Century Composer's Soundscape; Syntonic Tuning: Creating a Model
for Accurate Electronic Playback", Music Theory Online, 10(1), 2004.

328
Bibliography

[602] WILCOX (Howard J.), "Group tables and the generalized hexachord theorem",
Perspectives of New Music, 21(1-2), 1983, pp. 535-539.
[603] WILD (John), "Tessellating the chromatic", MaMuX Seminar (Mathematiques,
Musique et relations avec d'autres disciplines), 2002.
[604] WlLLE (Rudolf), "Mathematische Sprache in der Musiktheorie", Feedback Papers
24, 1981, pp. 6-10.
[605] WlLLE (Rudolf), "Mathematik und Musiktheorie", in Musik und Zahl, Verlag fur
systematische Musikwissenschaft GmbH, Bonn, 1976, pp. 233-264.
[606] WILSON (Ervin M.), "The Bosanquetian 7-rank Keyboard After Poole and Brown",
Xenharmonikdn, 1, 1972.
[607] WILSON (Ervin M.), "On the Development of Intonational Systems by Extended
Linear Mapping", Xenharmonikdn, 3, 1975.
[608] WILSON (Ervin M.), "Pythagorean Intervals", Interval, 1(3), 1979.
[609] WILSON (Ervin M.), "The Purvi Modulations", Xenharmonikdn, 10, 1982.
[610] WILSON (Ervin M.), "The Marwa Permutations", Xenharmonikdn, 9, 1986.
[611] WILSON (Ervin M.), "The Generalized Marimba Keyboard", Interval, 1986-1987,
pp. 8-9.
[612] WILSON (Ervin M.), "D'Alessandro: Like A Hurricane", Xenharmonikdn, 12, 1990,
pp. 1-39.
[613] WILSON (Ervin M.), "Partch's Scale on the Bosanquet Generalized Keyboard",
One/one, 1(4), 1985.
[614] WiTTSTEiN (Theodor), Grundzuge der mathematisch-physikalischen Theorie der
Musik, Hannover, Hahn'sche Buchhandlung, 1888.
[615] W O L F (Daniel), "Four Tunings for Sound Environments", One/one, 6 (3), 1990.
[616] WURSCHMIDT (Joseph), "Logarithmische und graphische Darstellung der musika-
lischen Intervalle", Zeitschrift filr Physik 3, 1920, pp. 89.
[617] WURSCHMIDT (Joseph), "Viertel und Sechsteltonmusik, eine kritische Studie",
Neue Musik-Zeitung, 42, 1921, pp. 183.
[618] WURSCHMIDT (Joseph), "Die 19-stufige Skala", Neue Musik-Zeitung, 42, 1921,
pp. 215.
[619] WURSCHMIDT (Joseph), "Die rationellen Tonsysteme in Quinten-Terzengewebe",
Zeitschrift filr Physik, 41, 1928, p. 527.
[620] WURSCHMIDT (Joseph), "Tonleitern, Tonarten und Tonsysteme", Sitzungsberichte
der Physikalisch-medizinischen Sozietdt zu Erlangen, Band 63, 1931.
[621] WYSCHNEGRADSKY (Ivan), "Etude sur Pharmonie des quartes superposees", Le
MSnestrel, Paris, 12 and 19 april 1935.
[622] WYSCHNEGRADSKY (Ivan), "Preface a un Traite d'harmonie par quartes superpo-
sees", Polyphonie, Paris, 1944, pp. 56-62.
[623] WYSCHNEGRADSKY (Ivan), "L'ultrachromatisme et les espaces non-octaviants",
Paris, La Revue Musicale, 290-291, 1972, pp. 73-130.
[624] WYSCHNEGRADSKY (Ivan), La loi de la Pansonorite', Edited by Pranck Jedrze-
jewski with the collaboration of Pascale Criton, Contrechamps, Genova, 1996.

329
Bibliography

[625] WYSCHNEGRADSKY (Ivan), Une philosophie dialectique de Vart musical, Edited by


Pranck Jedrzejewski, Paris, L'Harmattan, 2005.
[626] WOURINEN (Charles), Simple Composition, New York, Longman, 1979.
[627] XENAKIS (Iannis), Musiques formelles, Paris, Stock Musique, 1963, English Trans-
lation Formalized Music, Indiana University Press, 1971. Revised edition Pendra-
gon Press, 1992.
[628] XENAKIS (Iannis), Musique, Architecture, Paris, Casterman, 1971.
[629] YASSER (Joseph), A Theory of Evolving Tonality, American Library of Musicology
Inc., 1932.
[630] ZARLINO (Gioseffo), Istitutioni harmoniche, Venise, 1558.
[631] ZVONKIN (Alexander), LANDO (S.K), Graphs on Surfaces and Their Applications,
Springer, 2004.
[632] ZWEIFEL (Paul F.), "Generalized Diatonic and Pentatonic Scales. A Group Theo-
retic Approach", Perspectives of New Music, 34(1), 1996, pp. 140-161.

330
Index

Acoustic Power, 2 Braids, 105, 121


Affine Groups, 37 Bravais Temperaments, 188
Affinity, 21 Bridge Prank, 53
Agmon Etyan, 83 Brownian Motion, 14
Agon Carlos, 217
All-combinatorial, 115 Cactus, 192
Ambiant Isotopic, 106 Cadential Sets, 23
Amiot Emmanuel, 214 Cage John, 173
Andreatta Moreno, 30, 195, 214 Canons
Aperiodic Tilings, 200 Canons cancrizans, 208
Apotome, 3 Multiple canons, 212
Archimedean Tilings, 197 Prime canons, 210
Artin Groups, 122 Regular canons, 215
Assemblies, 35 Vuza canons, 214
Atonal Sets, 26 Cardinal Density, 21
Autocorrelation Function, 5 Cardinaly Stable, 25
Carlos Wendy, 167
Babbitt Milton, 83, 112 Carnatic Modes, 49
Bach J.S., 173, 208 Carter Elliott, 49, 53
Balanced Sets, 23 Cartographic Group, 190
Balzano Gerald, 68, 83 Category of Local Compositions, 18
Barraque Jean, 111 Cents, 2
Bartok Bela, 29, 49, 173 Chalmers John, 152, 155, 166
Berg Alban, 41, 112, 208 Chapuis August, 50
Biharmonic equation, 12 Charpentier Jacques, 29, 50
Binary Sets, 26 Chemillier Marc, 154
Bloch Georges, 217 Chicken Wire Torus, 95
Bohlen Heinz, 168 Childs Adrian, 95
Boulez Pierre, 114 Chord Diagrams, 106
Index

Chouvel Jean-Marc, 17 Euler Leonhard, 157


Clampitt David, 83 Euler's Formula, 177
Clough John, 83 Euler's Totient Function, 36, 160
Cohn Richard, 83, 95 Euler-Fokker Genera, 157
Combination Product Sets, 166 Extrinsic Pole, 25
Combinatoriality, 114
Comma Farey Sequences, 160
Hellgouarch comma, 154 Farey Temperaments, 160, 163
Holderian comma, 3 Fast Fourier Transform, 4, 6
Pythagorean comma, 3, 127, 131 Faure Gabriel, 208
Septimal comma, 3 Fechner Law, 1
Syntonic comma, 3, 127 Ferrers Boards, 207
Comparison Matrix, 85 Fixed Points, 30
Complete Temperaments, 155 Fokker Adriaan, 159
Concordance, 17 Fokker Planck Equation, 15
Conjugation, 30 Ford Circles, 161
Consistency, 155 Forte Allen, 29, 83
Constellations, 190 Fourier Transform, 4
Continued Fractions, 133 Frequency, 1
Convergents, 134 Frieze Groups, 174
Copland Aaron, 46 Pripertinger Harald, 30, 33, 35, 39
Cordier Serge, 167
Costere Edmond, 21, 29, 113 Gauss C.F., 105
Coxeter Groups, 178 Gauss Diagrams, 108
Crystal System, 180 Gauss Words, 109
Crystallographic Groups, 179 Generalized Interval System, 83, 88
Cube Dance, 96 Goeyvaerts Karel, 111, 114
Cycle of fifths, 129 Golden Temperaments, 170
Cyclic Chords, 42 Gould Mark, 69
Cyclic Group, 36, 174 Group Actions, 30
Cyclic Temperaments, 169
Hahn Paul, 155
Davies Peter Maxwell, 219 Hajos Groups, 214
Debussy Claude, 60 Harmonic Complexity, 153
Decaux Abel, 49 Harmonic Metrics, 152
Diatonic Scales, 69 Hascher Xavier, 83, 93
Diesis, 3 Hauer Joseph Matthias, 29, 53
Dihedral Classification, 40 Hellegouarch Yves, 152, 154
Dihedral Group, 37, 84, 174 Helmholtz Hermann von, 15
Dissonance Curves, 16 Hindemith Paul, 44, 208
Dodecaphonic Knots, 106 Honegger Arthur, 46
Douthett Jack, 83, 95
Durutte Camille, 41 IlOmaki Tuukka, 97
Dutilleux Henri, 29, 49, 53 Indigestibility, 153
Injection Function, 89
Eimert Herbert, 112 Interpretation, 18
Equal Temperaments, 130 Interval Class, 86
Erlich Paul, 155, 166 Interval Function, 88
Estrada Julio, 29 Interval Vector, 86

332
Index

Inversion, 35 Muzzulini Daniel, 17


Isographies, 93
ltd Formula, 15 Negative Isographies, 93
Neutral Sets, 26
Jacobi-Perron Algorithm, 141 Noll Thomas, 17, 30, 99
Jarred Sets, 23 Nono Luigi, 112
Johnson Tom, 200
Johnston Ben, 165 Oettingen Arthur von, 91
Jolivet Andre, 46 Ohana Maurice, 28
Just Intonation, 163 Overtones, 1
Ozzard-Low Patrick, 155
k-chords, 35, 57
K-nets, 92 p-limit, 164
Kelkel Manfred, 220 P6lya's Theorem, 34, 76
Klumpenhouwer Henry, 83, 92 Parsimony, 95
Klumpenhouwer Networks, 92 Partch Harry, 16, 123, 163
Knots, 105 Partials, 1
Koechlin Charles, 29, 53 Partition Function, 163
Krenek Ernst, 112 Parzysz Bernard, 113
Pcsets, 35
Lambert Philip, 83 Peck Robert, 83, 93
Leimma, 3 Pentilings, 199
Lenot Jacques, 114 Perfect Tilings, 200
Lewin David, 83 Perle Georg, 83
Limited Transposition Sets, 58 Phase Transitions, 162
Liszt Prans, 29, 46, 49, 93 Pierce John Robinson, 168
Local Composition, 17 Pinos Alois, 29
Pitch Class Sets, 35
Magic Squares, 217 Planar Groups, 180
Majinor Sets, 27 Plates, 12
Marie Jean-Etienne, 112 Platonic Solids, 177
Markov Chains, 116 Polyhedra Group, 177
Mathews Max, 168 Polyominoes, 198
Mathieu Groups, 189 Polytope, 177
Mazzola Guerino, 17, 29, 41 Positive Isographies, 93
McLaren Brian, 166 Power Towers, 96
Mean Deviation, 128 Prime Knots, 106
Meantone Approximations, 138 Pseudo-Cyclic Groups, 145
Meantone Temperaments, 130 Pythagorean Temperaments, 129, 169
Membranes, 10
Messiaen Modes, 60 Quantized Modulations, 20
Messiaen Olivier, 5, 29, 41, 46, 57, 102 Quarter Tones, 70
Milhaud Darius, 46 Quasicrystals, 181
Modulant Set, 25 Quinn Ian, 83
Modulation, 20
Modulator, 20 Rahn John, 83
Morris Robert, 97 Rameau Jean-Philippe, 41
Mosaics, 40 Random Noise, 15
Motives, 39 Rapoport Paul, 166

333
Index

Ravel Maurice, 53 Transitive Actions, 30


Reflexive Chords, 206 Transitive Sets, 25
Remarkability Function, 140 Transpositions, 35
Reptiles, 200 Tristan Chord, 25
Riemann Hugo, 91 Tuning Groups, 146
Riemann Transformations, 91
Rigid Scales, 20 Vlijmen Jan van, 114
Riotte Andre, 112 Vuza Dan Tudor, 214
Rooted Cactus, 192
Roslavetz Nikolai, 47, 93 Wagner Richard, 46
Roussel Albert, 50 Wave Equation, 6
Wavelength, 2
Saussure Ferdinand de, 124 Wavelet, 5
Schat Peter, 114 Webern Anton, 111, 117, 208
Schisma, 127 Wilson Ervin, 152, 166
Schoenberg Arnold, 93, 208 Wyschnegradsky Ivan, 41, 220
Scriabin Alexander, 43, 49, 53, 94
Semi-combinatorial, 114 Xenakis Iannis, 117
Sensory Consonance, 15
Yang Baxter Equation, 105
Series, 113
Young Diagrams, 207
Sethares William, 16
Young's Modulus, 12
Sets Classes, 35
Similarity Relations, 87 Z-related Pairs, 87
Sonance, 15 Zarlinean Temperaments, 130
Space Groups, 186 Zimmermann Bernd Alois, 111
Spectrum, 1
Sporadic Groups, 189
Stabilizer, 30
Stable Sets, 23
Starr Daniel, 29
Steinbach Peter, 95
Steiner Systems, 189, 196
Stern-Brocot Tree, 161
Stochastic Integral, 14
Stoeker Philip, 83
Stravinsky Igor, 46, 208
Strings (Equation of), 7
Strong Isographies, 93
Subcomplex, 88
Superpartial Ratios, 138
Symmetry Group, 173

Tangles, 121
Tcherepnin Alexander, 46, 49
Tilings, 180, 195
Tippett Michael, 53
Tonal Gravity, 21
Tonality Diamond, 163
Tonic Pole, 25

334
Dep6t legal: 2 e trimestre 2006
Imprime en Prance
•••••••••••••

ill SBifii
£1
:ss jjj[:jjjjjjjj:jj[j![[[:! II

J •••••••••••••••••IIIIIIIIIIilillHIl
JML
!•••••••••••••• ••••••••••••••••••••
(•••••Ml
i?• • • • • • • • • • • • • • • • • • a ••«••••••••
(•••••••I

• ••••••••••••••I ••••••••••I ••••••••


:::::;
.__•••••••••••
I ••••••
!•••••••••••••
!•••••

•••• •••••

SI!!!!!![[ !•••
••••••••

••••••••• •••I !•••••••••••!


!••••! IM •••••
••
M U S i Q U E / S C i E N C E S

Franck Jedrzejewski

MflTHEMRTiCflL THEORY OF MUSiC


This book provides a unified account of mathematical tools
for music theory and analysis, as alternatives to empirical methods. It constitutes a unique reference.
The first chapters of the book take as its subject the staple (basic, essential, age-old...) topics in syste-
matic musicology: harmonic and sonance models, Helmholtz research, Costere affinity models as well
as chords and scales taxonomy using group action method. Messiaen's limited transposition modes
are generalized to n-tone equal temperament.
Methods reviewed by the author briefly include the popular techniques of Forte set theory and Lewin
generalized interval systems. The legacy of Hugo Riemann is evoked via the new definition of Riemannian
transformations with permutations, which appear in mathematical music theory as carriers for models
of some musical objects, as well as for the semantics of other types of musical logic. Neo-Riemannian
investigations also provide new interpretations in the design of chords lattices and isographies.
Among the novel features of this book are diagrammatic representations of knots and braids. Schoenberg
series are mapped to singular knots. The appendix gives the complete classification of the twelve tone
rows in chord diagrams.
In the chapters devoted to tuning and temperaments, the hierarchical organization of contemporary
tunings and historical temperaments is studied under various refinements of fundamental concepts:
continued fractions, harmonic distance and tuning groups. In the last chapter, the relationships between
tilings and canons are related to groups and topology.
This book will be of value to graduate students and research workers in systematic musicology.

With the participation of Ircam and the support of CNRS and SFAM

:s
•••••••••i
!iB
CENTRE NATIONAL
DE LA RECHERCHK
SCIENTIFIOUE

• •••••••!
w
Miiiiii m
.. •••••*••••*••••••
Nxpu^'W

•••••••••••••••••.
IIISHI:
Dep
!••••

Vous aimerez peut-être aussi