Cool Memories:

Notes on the Ligeti Viola Sonata

john stulz
(2012)
“Who are we, who is each one of us, if not a combinatoria of
experiences, information, books we have read, things imagined?
Each life is an encyclopedia, a library, an inventory of objects,
a series of styles, and everything can be constantly shuffled and
reordered in every way conceivable” -italo calvino

there is nothing
outside the text

the experience of hearing and interpreting a musical work cannot

be isolated from the tangled web of knowledge buried within each
listener. No matter how focused or

If musical compositions contain the aesthetic property of

multiplicity, if they are constructed as a network of
references, quotations, imitations or non-musical metaphors,
what are the ramifications for the performer/interpreter?

This essay is an attempt to explore the idea of multiplicity in

musical composition through Ligeti’s Sonata for Solo Viola

[The act of listening to a musical composition cannot occur in isolation from

the tangled web of knowledge present in each listener.] When confronted with
musical material on all levels, from phraslet to symphony to concert, we
cannot help but connect the dots between what is being heard and bits of
information located in our personal encyclopedia of experience. It is as if
our minds create an instantaneous structuralist analysis relating the notes
(^being) heard to whatever it can grasp on to. This explains how a
progression in Schumann can simultaneously remind us of Bach, Mahler, the
Beatles and a Giraffe.
conveying referential meaning through performance

-Polystylistic Playing (play in the style of...)

-Phrasal Emphasis (segment the horizontal stream to highlight certain symbols/signifiers)

-Education (use verbal/linguistic cues to trigger connections in the audience)

In the 1970s and 80s music took a radical hermeneutic turn away from the musical
absolutism of the avant-garde toward the referentialism and multiplicity
associated with post-modern aesthetics. The method of transmitting meaning
through a musical message expanded beyond the boundaries of the work in-itself to
include the work in its cultural and performative context. Composers began
looking outward, using quotation, poly-stylism, historical regression and the
free play of reference to impart meaning on their musical compositions. For
performers of contemporary music, this radical shift in the way meaning is
transmitted through music has profound consequences on how and with what aim we
approach the performative act.

In the late 1970s György Ligeti faced a hermeneutic crisis typical of composers
during that period. He had exhausted himself composing an opera about the end of
the world, the musical avant-garde was dead and there seemed no way to produce a
musically meaningful composition without resorting to decadence, musical
digression, or kitsch. As he told students in a lecture 15 years later “I am in a
prison: one wall is the avant-garde, the other wall is the past, and I want to
escape” (Ross 465). In the midst of this crisis he made hundreds of unfruitful
attempts at writing a piano concerto (Steinitz 245) only to finally break free
five years after the completion of Le Grande Macabre with his most expressive
work, the horn trio of 1982. The solution that finally freed him from his
aesthetic prison was the synthesis of a unique language of multiplicity
referencing an entire world of ideas ranging from chaos theory and french
spectral music to the musics of Africa, the carribean and Conlon Nancarrow. For
performers of Ligeti’s late music this multiplicity presents a unique set of
challenges: how can references be shown in performance? What is the
responsibility of the performer-interpreter when it comes to recognizing
references? How do you perform an extra-musical concept? and how can we ever
expect the audience to understand this language of reference?
 useful quotes

“A method of knowledge, and above all as a network of

connections between the events, the people, and the things
of the world” (Calvino 105)

“novel as a vast net” (Calvino 124)

“nostalgia for a homeland that no longer exists” (NY Times

obituary of Ligeti)

“I am in a prison: one wall is the avant-garde, the other

wall is the past, and I want to escape” (quoted, Ross 465)

d
o un
ar te
l t p le
i m y
bu ual co onl
r d s e
n ge ivi k a ted hos o
lo ind wor oca c w s t
no he l usi on
s c al is m ati
i f t si g o l
at o mu in t re
th acy he ean me, on
c m , t m fra d
si ri e e
mu e p ent hos me ess
h m w t i r
t ate e t
g its but
st ssa
e s ns
me sid g i sig
n
in ani al
e n
m ter
ex
 cool memories
-an open sign ambiguously interpretable as a direct referent to a pre-existing sign/
concept

-this is in direct contrast to a hot memory which is a closed sign only interpretable
as a pre-existing sign/concept

the value of the later lies in the fact

cool hot that the pre-existing sign has been
recontextualized in some fashion giving it
sign? open closed new aesthetic life. For a cool memory
value lies in the openness or ambiguity of
value? sign+concept recontextualization the sign itself, the sign can be
interpreted in a given way as a reference
perceived? through knowledge through knowledge as memory to a pre-existing sign. In both
+ interpretation cases cultural knowledge of the referenced
sign is essential to their perception on
reference? vague direct the part of the listener but in the case
of a cool memory the reference also
requires interpretation to be uncovered

taxonomy of reference

Direct -restatement of sign

Translated -same information presented in a different medium
Similar -new information that bears resemblance to previous information
Allusion -new information that suggests previous information

unrecognizable

really a continuum
translation

allusion

esoteric
similar
exact

hot memory cool memory no relation

a question of the relationship between the information present in a given

referencing sign and the information present in the referenced sign.

metaphor
quotation, allusion, modeling
 cool memory -an open sign, (^ambiguously) interpretable as a direct referent
to a pre-existing signified concept

hot memory -a closed sign only interpretable as a direct referent to a pre-

existing signified concept

For most composers situated in the mainstream of cultural thinking, the

1960s marked the beginning of a new ear in musical thought. That old
fairytale of an inevitable modernist progression from Beethoven to Wagner
to Schoenberg to Webern to Boulez and beyond turned out to be just that, a
fairytale used to explain the necessity of unnecessary music. While the
hyper-modernity of works like “Structures IA” and “Gruppen” certainly
produced fascinating results for the specialist, to the ears of everyday
concertgoers they were incomprehensible

For Ligeti and many composers of his generation escape was found through
the free play of cool memories

Ligeti Sonata’s “Cool Memories”

-Romanian folk song/generic folk songs (Hora Lunga/Facsar)

-Chaos (Loop/Presto)
-Harmony (Facsar)
-Baroque Dance (Chaconne)
-Lamento Motif (Lamento/Chaconne)
-Nancarrow (Loop/Presto/Chaconne)
-Jazz (Loop/Chaconne)
-Machines (Loop/Presto/Chaconne)
-Hindemith (Presto)
-Spectralism (Hora Lunga)
 Study No. 11 (Nancarrow) as presto

web of reference: ligeti viola sonata

spectralism

L. horn trio

L. violin
concerto
jazz
Hora Lunga

folk
music
modalism
Loop

L. piano
machines
études
Facsar

chaos Presto

Nancarrow

tonality
Hindemith

pattern
mecanico

Lontano
generative
Chaconne
form

bach
baroque
lamento
dance
The Wikipedia Model of Multiplicity: a sign is presented, the listener
reacts by looking up a possible referential understanding (i.e. searching
for the sign within their personal database), the understanding is present
in a series of definitions, allusions, and related terms all hyperlinked to
their own referential understanding with further definitions, allusions,
and related terms. The searching is endless, a (^randomly) linked
contextual multiplicity bringing variegated referential understanding to
the whole. Could have resulted in any number of possible combinations of
referents and is by nature incomplete.

conceptual replication of signs

A Music is built of gestures, signs bent

through time, acoustic phenomena with
representable parametric values, sounds that
are temporally and vertically segmented
through perception on a variety of
contextual levels, ranging in size from the

C
quiet plucking of a ukelele to the entirety
of Mahler’s second symphony, with all the
motives, phrases, sections and movements in
between

B
Meaning can be transmitted through referencing on every level of the work and along a
continuous spectrum of vagueness. In its most clear form a composer such as John Adams in
Son of Chamber Symphony will directly restate a well known musical figuration, the opening
bars of Beethoven’s ninth, or as in the third movement of Berio’s Sinfonia, an entire
musical work. Rather than direct quotation, a composer may wish to impact the reference
with their own musical personality through a translation, bending the material to a new
musical lenguage as in Busoni or Stokowski’s transcriptions of Bach. If the composer wants
even more distance from the original referenced material they may allude to it through
incorperating only certain distinctive features of a figuration or language. The art of
allusion on a macrocosmic scale, that of an entire work or compositional style, was
perfected by both the neo-classicists of the first half of the 20th century and the neo-
romantics later in the century; however, composers as diverse as Kurtag Lachenmann and
Philip Glass have all found subtle applications of allusion on the micro-levels of a phrase
or sonority. [In general, the clarity of a reference is directly related to its similarity
with the original referent, the more loose that relationship is the less likely the
allusion is to be understood as a cogent reference.
The surprising consequence of loosening the relationship between the signifier (that
which references) and the signified (that which is referenced)]

In general, the form of a reference is exactly that of a Sausserian sign: the composer uses
a signifier (either a figuration, work or language) to refer to a signified concept (again,
a well known figure such as the opening of Beethoven’s ninth symphony, an entire piece such
as the third movement of Mahler’s second symphony or an entire musical style). In either
case, the clarity of a reference is a direct function of the signifier’s similarity to the
signified and the general cultural commonality of the signified.
The surprising consequence of loosening the relationship between the signifier and
the signified is that it allows for a multiplicity of references within a given signifier.
Through such ambiguity a single musical sign can take on many possible interpretations, a
listener hearing the double stop harmonics in Lamento may identify them as a restatement of
ocarinas in Ligeti’s own violin concerto, as a baroque or renaissance ground bass, or as a
reference to medieval parallel organum. In each case, correctly identifying a fresh
signified concept does not pit it in a dialectical argument with a previously identified
reference but rather adds a new level of depth to the overall understanding of the message.
Taken to its extreme, as in the novels of Carlo Emilio Gadda and James Joyce, the artwork
of Hieronymus Bosch, or the “Histoire(s) du Cinema” of Jean-luc Godard, the aesthetics of
multiplicity can transform an artwork into “a method of knowledge [using] a network of
connections” (Calvino 105) between a vast treasury of external objects, events and
concepts.

By no means must a composer limit themselves to referencing purely musical concepts. From
at least the time of Monteverdi, composers have used certain musical gestures to reference
non-musical sounds such as a cuckoo bird or thunder. Taking the idea further, composers
routinely directly mimic poetic language with musical figurations ranging from a rising
scale to represent an ascent into heaven or a tritone to represent the devil. Most
interestingly, and of particular importance to Ligeti, composers have used the form of a
piece itself to illustrate an extra-musical concept. Bach’s infinitely rising cannon from a
musical offering can represent the ever rising glory of God (or his patron Frederick the
Great) and the hypothetically hours long extension of color and talea of the first movement
of Messiaen’s Quatour Pour le Fin du Monde can represent the eternal expansion of heaven.
 hungarian rock

44 the number
violin concerto

Personal language and Reference

The idea that one work by a given composer can have loose
references to other previously or concurrently conceived
works within the oeuvre of that same composer is certainly
suspect. Given the long held beliefs that each composer
develops their (^his or her) own personal language, how could
an interpreter possibly differentiate between the
characteristics of that composer’s language and (^the
possibility of) somewhat vague references to other works
within their oeuvre. How could one identify without a written
or verbalized (^statement) by the composer, the legitimacy of
an interpretation that posits a given element as a reference
as opposed to a stylistic coincidence?
A clear example of reference within one’s own catalogue
occurs in Ligeti’s fourth piano étude “Fanfares” which is
built around an eight-note ostinato repeated 208 times. It’s
pitches are identical to those of “Hungarian Rock” (repeated
176 times) and the second movement of the horn trio (with a
3-3-2 accent grouping instead of the 3-2-3 of “Fanfares”).
Given the late Ligeti’s aesthetics of multiplicity and
reference, it is difficult to analyze the reusing of material
as the result of stylistic concerns as opposed to intentional
semiological networking. In contrast (^within the critical
framework of earlier musical hermeneutics), the use of
similar re-used ostinati (^by) within the work of earlier
composers such as Bach, Beethoven or even Bartók could rarely
be considered significant. In order to generate some kind of
method whereby an analyst can convincingly speak of (^a
dichotomy between) personal referencing and personal style,
the factors weighed must include both the way an individual
composer expresses meaning through music (in gangly technical
jargon the composer’s hermeneutical praxis) as well as the
nearness to exact repetition a given sign lies on our
continuum of reference.
Given this dichotomy, clear loose references exist
between the viola sonata and three other pieces from within
Ligeti’s work: The Violin Concerto, Piano Etudes, and Horn
Trio.
I[0]=Initial conditions
I[1] = F[1]{ I[0] }
I[2] = F[2]{ F[1]{ I[0] }}
I[3] = F[3]{ F[2]{ F[1]{ I[0] }}}
...
...
...
I[n] = F[n]{ F[n-1]{ ... F[1]{ I[0] } ...}}

Iterative Processing

The basic compositional device used by Ligeti

throughout the viola sonata is that of the iterative
process or recursion. Each moment begins within the
presentation of a string of information, what I call
Iteration zero of I0. The material is then extended by
running the information contained in I0 through some
process of alteration in order to derive a new
iteration I1. This new (^string of) information
simultaneously contains both the initial material and
the new material by presenting I1 as a clear function
of I[0] (I1=Fx{I[0]}). This process is repeated again
(^and again) creating I[3], I[4], (^all the way) to
I[n] when the movement finishes itself.
Mathematically, we can represent this process of
nested functions in the figure below, although a
simpler explanation would be to say that each
iteration is created by altering the previous
iteration.
Hora Lunga

The first movement of Ligeti’s sonata, “Hora Lunga”, is written entirely on the C-
string and consists of a series of eight sections. These divide roughly into a
bipartite form with sections 1-5 being mirrored in sections 6-8. The primary
melodic material is a synthesis of two distinctive musical (^denotative) systems
(or paradigms), that of the “family of slow, stereotyped melodic formulas and
figures which are strung together in Romanian folk songs from the Maramues region
in the northern Carpathians” (Ligeti 17) and that of french spectralism
(particularly the approach of Claude Vivier). While both are unique languages with
fixed sets of signifiers and signifieds through which the composer/musician may
create a musical discourse, there exists points of convergience (mainly the co-
dependent use of the harmonic spectrum as the source of melodic inspiration) which
Ligeti explicitly highlights (both with his verbal commentary and with stylistic
technique) to create a semiological multiplicity.
AHCB: the dual nature of musical symbols
In the opening measure of “Facsar”, Ligeti gives us a prime example of a musical
symbol, the four-note Bach Motif re-arranged as A-B-C-Bflat (where “B” is labeled
“H” in the German system). This famous motif pops up all throughout the
(^literature) from the last notes of Kunst Der Fugue to the music of Webern and
Boulez. It is by all means safe to assume that there is a certain class of well
versed listener’s who, when hearing the opening of “Facsar”, will make this
connection and all of the subsequent (^connections) implied by its statement
(i.e. the direct links to Bach through implied harmony, four-part chordal playing
technique à la the solo sonatas and partitas, and even possibly the distinctive
broken double stop first found in measure 13 that harkens back directly to the
andante of Bach’s a-minor violin sonata). However, in order for this to be
understood without prompting two requirements must be met: the listener must have
a prior knowledge of the Bach motif and the listener must have enough auditory
awareness to recognize the notes A-B-C-Bflat as a re-ordered statement of the
Bach motif. The auditory awareness required will be different from listener to
listener, ranging anywhere from perfect pitch to a hunch that just happens to be
right. Either way, it is certainly safe to assume that just as there is a class
of well versed listener’s able to make the connection, the number of listeners in
that class is small.

différance and interpretation

Once recognized as a statement of the Bach motif, the four-note phraslet
described above is transformed from a purely musical signifier - whose non-
linguistic meaning can only be defined through the parametric relationships
present in the gesture - to a symbolic signifier with the entire network of
concurrent musical, linguistic, and subjective signifiers available to each
interpreter. What was once a conceptual unit “(^the parametric relations of) the
opening four-note phraslet of Ligeti’s Facsar” has become “a re-ordered statement
of the Bach motif”. Through the process of what Derrida has called Différance we
have replaced one signified with (^another) signifier. A purely musical element
whose meaning cannot lie outside of the parametric relations presented has become
an iteration of a symbol in our personal encyclopedia of extra-musical signs.
Without différance Shostakovich’s ironies, Mahlerian stream of consciousness,
Berlioz’s tone poems and Ars Nova symbolism would all be meaningless sonic
events. Even when you feel emotion or (^directly) describe the parametric
features of a musical event you cannot escape the process of différance.
once the process of referential linking begins, once the auditor applies the lens
of interpretation through reference, the identification of referents multiplies
exponentially from both directions of decoding and applying.

The basic building block of meaning in solyptic networks is the

reference, a form of sign relating its articulation to an external
concept. Following the general conception of a sign as “the union
of a signifier and a signified” (Barthes 38), a reference, when
identified as such, is a total combine of its expression and its
content, the musical event and the idea to which it is linked. It
is interpretively indivisible functioning exactly as a hyperlink
containing both its expression as physical fact and the code
linking it to a related site. As such, references are
simultaneously interpretable through either decoding its
expression (“x references y”) or applying its content (“y seems
related to x”), both of which are reliant on the auditor’s
capacity to make the connection between expression and content.
The act of interpretation becomes integral to the meaning within
the message, it is an opening of the hermeneutical process whereby
the traditional dichotomy of that which the artist is trying to
say and that which the audience thinks he is saying is replaced by
an endless network of connections between the physical events and
externalized referents that are identified by the interpreter.

by definition, reference does not resort to a meta-language of interpretation,

while we use the linguistic formulations “x references y” or “y is referenced by
x” there is conceptually no need for language, what is heard immediately becomes
a psychological unit that links directly to other psychological units (memories)
that are not part of a synthetic framework of understanding.

the connection between material expression and referenced concept is in Barthes

the “Signification”

It is “motivated” in that a clear relation exists whereby the reference models

the referent in its physical materialization
“Loop” consists of a
three bar introduction
followed by nine
complete iterations of
a string of 45 dyads
and one virtual
iteration in which only
the first three dyads
are stated. each
iteration is an exact
restatement of the 45
dyads, neither the
order nor the pitch
material changes. In
order to generate an
over-arching dramatic
form out of the
constant looping of
fixed information,
ligeti subjects the
rhythm of each point to
irrational alteration
and weights these
alterations negatively
in the direction of
reduction. Figure 1
shows the rhythm of the
first three dyads as
they change with each
iteration. Notice how
with each iteration the
rhythmic value of a
given pitch could be
lengthened or shortened
without any
compositionally logical
reasoning; however, the
overall trend is
(^weighted) towards
reduction.
Simulating Ligeti
An Algorithmic Recreation of Loop in OpenMusic

I Introduction

Throughout the twentieth century there was a dramatic increase in musical works created
through compositional process as opposed to compositional whim. This interest clearly
starts with the combinatorial methods of the second Viennese school and runs through
Boulez, Cage, Xenakis, Reich, and beyond. In order for compositional theory to be of any
use we must move away from analyzing the results toward analyzing the process through
which those results are achieved. By studying a musical work through statistical,
algorithmic, and logical analysis we can better understand both the processes through
which the work itself is created and the meaning buried beneath the notes. This paper is
an attempt to do just that by studying the rhythmic parameter of Loop from György
Ligeti’s Sonata for Solo Viola.
After a description of the general construction of Loop, exploring its generative
form and its relationship to chaotic systems, I will take the reader through various steps to
creating a method of generating algorithmic simulations of Loop. This method will
involve a combination of Markov Analysis and programming logic using a LISP
Graphical User Interface for composition called OpenMusic (OM). By properly
understanding the conditions which lead to successful simulations of Loop we are able to
clearly distinguish the limits of its compositional process and the implementation of that
process by an artistically sensitive composer such as Ligeti.
II General Construction

Loop is composed of a series of nine iterations of a 45-unit long string of dyads, preceded
by a two bar introduction. The overall rhythmic scheme throughout Loop is divisible by a
common denominator of one sixteenth note, with values ranging from 1-beat units to 8-
beat units. Beginning with iteration one (heretofore referred to as I1) starting at bar 4, we
have a rhythmically diverse presentation of the 45 dyads in a somewhat moderate rate. I1
uses almost the entire gamut of rhythmic units with 2- through 8-beat units all making an
appearance. As the work moves from I1 to I9 the rhythmic pattern morphs from moderate
and varied to fast and regular, with an almost constant stream of 1-beat units occurring in
I9. Table II.1 shows the entire sequence of 16th note units as they move from I1 to I9 with
their index (dyad #) in the top column.
What the data in Table II.1 show upon first inspection is the overall motion from
rhythmic volatility in I1 to rhythmic standardization in I9. This motion is shown even
more clearly when graphed as in Figure II.1, whereby each rhythmic value is plotted
along the y-axis against its index in the x-axis. Graphed as such, the actual values are less
interesting than the overall motion, we are able to see the trend from large variability and
rhythmic value to flat, low value and almost completely even.
However, not only can we observe the data horizontally in iterative units, we can
also see the transformation vertically of each dyad in the string. Thus a dyad such as
number 14 occurs nine-times throughout Loop with the rhythmic sequence 3-3-2-2-2-1-1-
1-1. When mapped individually, as in Figure II.2 for dyads 1 and 2, the generally
rhythmic decay is still quite clear while the path with which that decay happens is more
clearly represented. By comparing all 45 dyads mapped as such, two properties of the
process become quite clear: 1) each dyad follows a random path of decay, and 2) each
path of decay is unique to that dyad.
In general terms, we can say that Loop follows a generative process of decay
whereby each atom (in our case any one of the 45 dyads) takes a random and unique path
weighted towards 1-beat rhythmic units. The entire system, i.e. each iteration, moves
from a chaotic state of high variability, entropy, and complexity to an almost totally
ordered state of low variability, uniformity and simplicity. In nature this is the same
general process of boulders becoming sand, mountains becoming plains, or snowflakes
melting into water droplets.
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 4 3 3 5 3 2 2 3 3 8 2 2 3 3 2
2 3 3 5 8 3 3 2 2 2 2 2 2 3 3 3
3 4 3 3 5 3 2 2 3 3 5 3 2 2 2 2
4 2 2 3 3 2 2 2 2 2 2 3 3 2 2 1
5 1 2 3 3 2 3 2 2 1 2 2 1 2 2 1
6 1 2 1 1 2 1 1 2 2 2 1 1 2 1 1
7 1 1 2 2 1 2 2 1 2 1 1 1 2 1 1
8 2 1 1 1 2 1 1 1 2 1 1 1 1 1 1
9 1 1 1 1 2 1 1 1 2 1 1 1 1 1 1

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
1 2 2 2 4 6 3 2 3 4 6 2 2 2 2 2
2 2 3 2 5 3 2 2 2 2 4 2 2 2 3 2
3 2 3 2 3 2 2 2 2 2 3 2 2 1 2 2
4 1 2 2 2 3 1 2 2 1 2 3 2 2 1 2
5 1 2 2 2 3 2 1 2 3 3 2 2 1 2 2
6 2 2 1 3 2 1 1 1 2 1 2 1 1 1 2
7 1 2 1 2 2 1 2 1 2 2 1 2 2 2 1
8 1 1 1 2 1 1 1 1 1 2 1 1 1 1 1
9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
1 2 3 3 5 7 6 5 3 2 2 3 3 8 2 2
2 3 2 2 3 3 5 3 2 2 3 3 2 2 2 2
3 3 2 1 3 2 3 4 2 2 2 2 2 1 2 1
4 3 2 1 2 3 4 3 1 1 2 2 2 2 2 2
5 1 3 2 2 1 2 2 1 2 2 1 2 1 2 2
6 2 1 3 2 3 2 2 1 2 2 1 2 1 2 1
7 1 1 1 2 1 2 2 1 1 1 1 1 2 1 1
8 1 1 1 2 1 1 1 1 1 1 2 1 1 1 1
9 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1
Table II.1: Rhythmic Pattern for Loop
Figure II.1 General Rhythmic Transformation From I1 to I9

Figure II.2 A: Rhythmic Decay of Dyad 1

Figure II.2 B: Rhythmic Decay of Dyad 2

III General Algorithmic Process

In order to simulate the rhythmic decomposition of Loop we need to mechanize

the derivation of each iteration. Exactly how this is done is a matter of both careful
planning and taste; however, a general value that each simulation should be both varied
from and similar to the original will be kept. This means that we would like to create an
algorithm, or mechanized composition process, that can give us a variety of different
solutions which all appear to be quite similar to the solution given by Ligeti. The general
process of this algorithm is given in Figure III.1.

seed rhythm

rhythmic generator
no: repeat with
new values

is piece over?

yes
collect result

print all results

Figure III.1 General Algorithmic Process for Loop Simulations

In the schemata above, the seed rhythm is represented as a diamond which indicates a
“fed value” – a value which the user inputs, the functions are given in rectangles and the
logical operator “is piece over?” is represented in an oval. These three types of actions
represent the basic building blocks of all algorithmic processes. “Fed values” can be
either manually placed into the system or randomly derived; however, all algorithmic
processes need some kind of data to work. Functions provide the basic transformations
and/or actions taken on the data received and can range in complexity. Logical operators
act by asking questions and can produce various results based on the answers to those
questions. In the general algorithm presented in Figure III.1 one could set any kind of
value to answer the question “is piece over?”; however, for our purposes we will follow
the model of Ligeti by stating that the work is finished after eight repetitions of “rhythmic
generator”, taken with the seed value produces nine total iterations.

IV Markov Analysis

Given the rhythmic data provided by an iteration In of Loop it is possible to derive a very
rough imitation by creating a simple probability distribution of all possible rhythmic
events occurring. This is done by counting up the occurrences of each rhythmic unit and
dividing by the total. For instance, I2 has 24 occurrences of 2-beat units, 16 occurrences
of 3-beat units, 1 occurrence of a 4-beat unit, 3 occurrences of 5-beat units, and one
occurrence of an 8-beat unit. Using this material we can create the probability distribution
of I2 found in Table II.1.

2 3 4 5 8
P(x) 24/45 16/45 1/45 3/45 1/45
Table II.1: Probability Distribution for Rhythmic States in I2

Using the “ChoixMultiple” function in OMAlea, we can randomly select any number of
events following a given probability distribution (Figure II.1). We present the function (c)
with a probability vector (a) representing the distribution from Table II.1 and a list of
states (b) indexed to our probability vector. The function then randomly draws a state
based on the probabilities of (a).

Figure II.1: Using “choixmultiple” To Generate States

By repeating this method 45 times we can create a somewhat satisfactory imitation of I2

(Figure II.2, compared with the original). However successful this method is in
synthesizing an imitation of a given In it fails miserably to produce a satisfactory
imitation of the entire Loop process. This is because each state in a given iteration is not
randomly drawn from a probability distribution but is generated from the equivalent state
in the previous iteration. Each rhythmic state after I1 is imbued with a memory of its
previous state in earlier iterations, the entire transformation from I1 to I9 is thus a
Generative Process as opposed to a Random Process.
 In order to deal with generative processes such as Loop we need a different
approach to probability theory, one that takes into account the formulation “given State A
what is the probability that State B will occur”. The method through which generative
probability can be explored is known as “Markov Analysis”. As an analytical tool
Markov analysis is similar to the method of probability distribution above; however, it

Imitation

Ligeti

Imitation (continued)

Ligeti (continued)

Figure II.2: I2 imitated with a simple probability distribution compared to original

takes the set of probabilities of all initial states becoming resultant states. For Loop this
means measuring not the occurrences of states in an iteration In but the transformations of
states across two iterations (i.e. the probability of a X-beat unit from In-1 becoming a Y-
beat unit in In).
The most common method for presenting the resultant data is in what is known as
a “Transition Table” whereby the initial states are listed vertically in the left-most column
and the new states are listed horizontally on the top-most row. The probability of State A
becoming State B is then filled in for each cell of the table. Table II.2 presents the
transition table of a 1st Order Markov Analysis of I2. By looking at an initial state from I1
we can look up the probability that a given state will occur at the same dyad in I2.

1 2 3 4 5 6 7 8
1 - - - - - - - -
2 - 13/19 6/19 - - - - -
3 - 4/7 5/14 - 1/14 - - -
4 - 1/3 1/3 - 1/3 - - -
5 - - 2/3 - - - - 1/3
 6 - - 1/3 1/3 1/3 - - -
7 - - 1/1 - - - - -
8 - 1/1 - - - - - -
st
Table II.2: Transition Table I1 to I2 (1 order Markov Analysis)

This method can be expanded for all transitions I1 through I9 ddd

Another way to present the results of a Markov analysis is in a map known commonly as
a Markov Chain which shows the probability that any given state will become a resultant
state. This method better suits an alternative approach that focuses on the transformation
of each individual dyad as opposed to the total transformation from iteration to iteration
of all dyads as in the 1st order markov analysis above. Figure II.3 shows the Markov
Chain of dyad 25 from I1 to I9 in Loop. The usefulness of this diagram lies in exposing
the probability of possible paths of rhythmic decay given the information at hand and
allowing the synthesizer to recreate an alternative path based on those data. Starting with
an initial rhythmic value of 6 for I1, every simulation of dyad 25 will jump down to a
rhythmic value of 4 by I2 and 3 by I3 due to the 100% probability of the transitions
represented in the Markov Chain. From there the possibilities using 1st order Markov
Analysis double, the rhythmic value of I4 could be either 1 or 2 with equal probability. In
this way, the original rhythmic value of 6 in I1 can randomly shifts until it reaches the
end state (naturally, after passing through a rhythmic value of 1).

Figure II.3
 1 2 3 4 5 8 Total
1 - - - - - - 0
2 1/6 2/3 1/8 - 1/24 - 24
3 - 9/16 5/16 1/8 - - 16
4 - - 1/1 - - - 1
5 - - 1/1 - - - 3
8 - - - - 1/1 - 1
st
Table 2: Transition Table I2 to I3 (1 order)

1 2 3 4 5 Total
1 1/4 3/4 - - - 4
2 6/25 3/5 4/25 - - 25
3 - 2/3 1/4 1/12 - 12
4 - 1/2 1/2 - - 2
5 - 1/2 1/2 - - 2
st
Table 3: Transition Table I3 to I4 (1 order)

1 2 3 4 Total
1 2/7 4/7 1/7 - 7
2 7/28 18/28 3/28 - 28
3 1/3 1/3 1/3 - 9
4 - 1/1 - - 1
st
Table 4: Transition Table I4 to I5 (1 order)

1 2 3 Total
1 8/11 2/11 1/11 11
2 10/27 15/27 2/27 27
3 5/7 2/7 - 7
st
Table 5: Transition Table I5 to I6 (1 order)

1 2 Total
1 6/11 5/11 22
2 3/5 2/5 20
3 2/3 1/3 3
Table 6: Transition Table I6 to I7 (1st order)

1 2 Total
1 23/26 3/26 26
2 15/19 4/19 19
Table 7: Transition Table I7 to I8 (1st order)
 1 2 Total
1 37/38 1/38 38
2 5/7 2/7 7
Table 8: Transition Table I8 to I9 (1st order)

((4800 5900) (5700 6200) (5500 6600) (6400 6900) (6200 7300) (6200 7200) (5500
6800) (5500 6600) (6200 7500) (5500 7100) (4800 7000) (4800 6600) (4800 6400) (6200
8000) (6200 7800) (6200 7700) (5500 7300) (5500 7100) (6200 7600) (4800 7000) (5500
6900) (5500 6500) (6300 6900) (6200 7200) (5500 6800) (6600 6900) (6500 6900) (6100
6200) (4800 5900) (5600 6200) (5500 6600) (6400 6900) (6200 7300) (5500 7700) (6200
7000) (6200 7800) (5500 7300) (6900 7100) (6900 7000) (6600 6900) (6200 6400) (6800
6900) (5500 6600) (6500 6900) (6300 6900))

(4 3 3 5 3 2 2 3 3 8 2 2 3 3 2 2 2 2 4 6 3 2 3 4 6 2 2 2 2 2 2 3 3 5 7 6 5 3 2 2 3 3 8 2 2)
(3 2 3 3 2 2 2 2 2 2 2 2 2 2 3 2 2 2 5 3 2 3 2 2 4 2 3 2 3 2 2 3 3 3 3 5 8 3 2 2 2 3 2 2 2)
(2 5 3 2 2 3 2 2 2 2 2 2 2 2 2 2 1 2 3 4 2 3 2 5 3 2 2 2 2 1 2 3 3 3 2 3 5 2 5 3 3 2 3 1 2)
(2 3 2 2 2 2 2 1 3 1 3 2 2 3 1 3 2 2 3 2 2 2 2 2 2 2 2 2 2 2 2 3 2 3 1 2 3 1 3 3 3 1 3 1 1)
(3 1 2 2 2 2 2 3 1 1 1 1 2 1 1 3 2 2 3 3 2 2 3 1 1 2 1 1 2 2 2 3 1 3 3 1 2 2 3 2 2 2 2 2 2)
(1 1 3 2 2 2 2 1 1 1 1 1 1 1 3 1 2 2 1 1 1 1 1 2 1 1 1 2 2 3 3 1 3 2 1 1 2 2 1 2 2 1 2 1 2)
(2 1 1 1 1 1 2 1 1 2 1 2 1 2 1 2 2 2 1 1 1 1 2 2 1 2 1 2 1 2 1 2 1 1 1 2 1 2 2 2 2 1 1 1 1)
(1 1 1 1 1 1 1 2 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 2 1 1 1)
(1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 2 1 1 1 1 2 1 1 1 1 2 1 2 1 2 1 1 1 2 1 2 1 1 1 1 1 1 1)