Analyzing The Music of Living Composers

Analyzing the Music of Living Composers
(and Others)
Analyzing the Music of Living Composers
(and Others)
Edited by
Jack Boss, Brad Osborn,

Tim S. Pack and Stephen Rodgers
Analyzing the Music of Living Composers (and Others),
Edited by Jack Boss, Brad Osborn, Tim S. Pack and Stephen Rodgers
This book first published 2013
Cambridge Scholars Publishing
12 Back Chapman Street, Newcastle upon Tyne, NE6 2XX, UK
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Copyright © 2013 by Jack Boss, Brad Osborn, Tim S. Pack and Stephen Rodgers and contributors
All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system,
or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or
otherwise, without the prior permission of the copyright owner.
ISBN (10): 1-4438-4227-3, ISBN (13): 978-1-4438-4227-3

TABLE OF CONTENTS
Preface ........................................................................................................ ix
Jack Boss
PART I: PROCESS IN THE MUSIC OF LIVING COMPOSERS
Introduction to Part I ................................................................................... 3

Tim S. Pack
Chapter One................................................................................................. 5
Making the Most out of Minimalism: Multiple Simultaneous Processes
in Torke’s Telephone Book
Brent Yorgason
Chapter Two .............................................................................................. 27

Coherence and Comprehensibility in Joan Tower’s Purple Rhapsody
Andrew Gades
Chapter Three ............................................................................................ 51

Games, Simplicity, and Trees: An Analysis of Arvo Pärt’s Arbos
Erik Heine
Chapter Four.............................................................................................. 70
An Introduction to the Music of Willem Ceuleers
Tim S. Pack
PART II: THE TONAL TRADITION
Introduction to Part II .............................................................................. 105

Stephen Rodgers
Chapter Five ............................................................................................ 106

Harmony and the Myth of Narcissus in ‘Du liebst mich nicht’
Jessie Layne Thornton
vi Table of Contents
Chapter Six .............................................................................................. 115

Mahler’s Musical Idea: A Schenkerian-Schoenbergian Analysis
of the Adagio from Symphony No. 10
Jack Boss
Chapter Seven.......................................................................................... 132

Debussy’s Unterbrechung
Gary Don
PART III: POP MUSIC AND BEYOND
Introduction to Part III............................................................................. 161

Brad Osborn
Chapter Eight........................................................................................... 164

When Pop Stars Collide: Mashups as Musical Destiny
Christine Boone
Chapter Nine............................................................................................ 176

Groove Topics in Improvised Jazz
Garrett Michaelsen
Chapter Ten ............................................................................................. 192

The Music of Mario, Link, and Samus: Harmony, Form, and Meaning
in Early Compositions by Koji Kondo and Hirokazu Tanaka
Nathan Baker
Chapter Eleven ........................................................................................ 218

Hearing Heima: Ecological and Ecocritical Approaches to Meaning
in Three Icelandic Music Videos
Brad Osborn
PART IV: MUSICAL GEOMETRY AND TEMPORAL STRUCTURING
Introduction to Part IV............................................................................. 239

Jack Boss
Chapter Twelve ....................................................................................... 240

Serial N-Cubes
Paul Lombardi
Analyzing the Music of Living Composers (and Others) vii
Chapter Thirteen...................................................................................... 254

Temporal Structuring in Colin Matthews' Eleven Studies in Velocity
Aleksandra Vojcic
Contributors............................................................................................. 275
PREFACE
In March of 2010, a rather unique scholarly meeting was held in Eugene,

Oregon under the auspices of the West Coast Conference of Music Theory
and Analysis. Our stated purpose was to focus on applying traditional music-
analytic techniques, as well as new, innovative techniques, to describing the
music of composers of the late 20th and early 21st centuries. In any era,
recently-composed music is often the most difficult to understand—and that
difficulty increases greatly in our own era, when such a wide variety of
musical styles are practiced and even combined. Our goal was to take steps
toward making music of our time a bit less impenetrable for our colleagues,
students and other listeners by showing how it follows, varies, and sometimes
controverts the organizational schemes of older music. As is often the case
with our meetings, our call for papers was open to a broad range of topics—
so there are chapters analyzing music of older eras as well, which, as it turns
out, throw light on the analysis of recent music in unexpected ways.
Therefore, like our previous proceedings volume published with Cambridge
Scholars, Musical Currents from the Left Coast (2008), this book “contains
what may be understood as a snapshot, from one vantage point at least, of the
present state of North American research in music theory and analysis.” It
also will be useful to musicologists, music analysts, and performers of the
repertoire considered, as well as interested amateurs.
In the introduction to Musical Currents from the Left Coast, Bruce
Quaglia and I commented on how the studies contained in it “present a cross-
section of thought at a particular instant that then reflects a network of
trajectories, and suggests possible developments and outcomes.” This new
volume has its own trajectories, and we anticipate that it will give rise to its
own developments and outcomes, but, to a much larger degree than Musical
Currents, these very well could include responses from living composers in
the language of music. Thus the book’s contribution to our art could go well
beyond enhancing the fields of music theory and musicology.
—Jack Boss, September 2012

PART I:
PROCESS IN THE MUSIC

OF LIVING COMPOSERS
INTRODUCTION TO PART I
Part I explores “Process in the Music of Living Composers.” It opens

with Brent Yorgason’s exploration of compositional procedures in
Michael Torke’s Telephone Book, which range from techniques derived
from minimalism to cyclical patterns of Stravinsky, Brahmsian motivic
transformations, the harmonic and rhythmic language of jazz, and melodic
riffs derived from rock and pop. Yorgason explains how Torke speeds up
and combines musical processes to create momentum, and he asserts that
Torke’s profusion of as well as treatment of these processes reflects
postminimalism.
The recurring theme of motivic transformation, now understood
through the lens of Schoenberg’s theoretical work, also forms the basis for
Andrew Gades’s study of Joan Tower’s Purple Rhapsody. Gades not only
describes the motivic processes that give the work coherence, but accounts
for the unique roles they play within different formal sections of the work.
After identifying the motivic “seeds,” Gades examines how motivic
deployment articulates formal divisions throughout Purple Rhapsody. His
essay considers other form-delineating elements, such as pitch collection,
tempo, texture, and timbre. Gades also describes Tower’s gradual
transformation of rhythmic and melodic motives as a means of lending
coherence to Purple Rhapsody, and he explains how the main theme’s dual
character—both static and dynamic—is paralleled at multiple levels of the
piece to generate drama and long-range momentum.
The focus then turns to Arvo Pärt’s music, as Erik Heine investigates
the processes that generate the early tintinnabuli work Arbos. Relying on
the composer’s own comments put forth in Paul Hillier’s Arvo Pärt, Heine
defines and illustrates the tintinnabuli method as being based on a diatonic
melody voice (M-voice) and a tintinnabuli voice (T-voice), which is
always a note of the tonic triad; he points out that in relation to the melody
voice, the T-voice can assume different positions such as above or below
and first or second position. Heine then goes on to show how the narrative
portrayed by the triple mensuration canon in Arbos has continued to set
that work apart from Pärt’s other tintinnabuli compositions.
Part I concludes by introducing readers to the musical language of the
Belgian composer Willem Ceuleers, whose steadily growing output thus
far includes over 780 works. In addition to relaying the composer’s own
4 Introduction to Part I
remarks about decisions made prior to and throughout the compositional

process, Timothy Pack demonstrates Ceuleers’ skill and stylistic breadth
by examining four works: Stabat sancta Maria, op. 655; Requiem, op. 735;
Cantata voor zondag Trinitatis, op. 610; and Orgelmis voor het Heilig
Hart van Jezus, op. 688. Pack explains how Ceuleers masterfully
incorporates an extensive array of styles spanning the last eight centuries
and cutting across several genres, including vocal and purely instrumental.
Pack’s essay shows how Ceuleers uses each earlier style as one of many
compositional parameters for creating, in his own musical language and in
his own meta-style, a new work for today’s listeners to enjoy.
—Tim S. Pack
CHAPTER ONE
MAKING THE MOST OF MINIMALISM:

MULTIPLE SIMULTANEOUS PROCESSES
IN TORKE’S TELEPHONE BOOK
BRENT YORGASON
In the classic minimalist compositions of the 1970s, the gradual

playing-out of clearly audible musical processes was a central feature.
Minimalist composers such as Philip Glass, La Monte Young, and Steve
Reich developed a number of techniques for working with musical
processes such as phasing, the subtle shifting of metric accents, and
additive and subtractive patterns. Post-minimalist composers incorporated
many of these processive techniques into their own compositional styles,
but did not adopt the aesthetic or the style of minimalism.1 Unlike early
minimalism, which typically features hypnotic repetitions and slowly-
paced changes within a single ongoing process, post-minimalistic music
normally features more rapid changes in style, less repetition of musical
ideas, and the simultaneous use of multiple processes.
All of these characteristics can be seen in Michael Torke’s Telephone
Book, a three-movement work for chamber ensemble that includes “The
Yellow Pages,” “The Blue Pages,” and “The White Pages.”2 Torke is one
of those rare individuals who literally hears colors in music, a condition
known as synesthesia.3 In discussing this piece, Torke mentions that he has
always heard the key of G major as “a darkish, burnt yellow.”4 Therefore,
“The Yellow Pages” is written in G major. For similar reasons, he uses
two sharps for “The Blue Pages” and no sharps or flats for “The White
Pages.”
Telephone Book uses a number of musical processes derived from
minimalism, but employs them in post-minimalistic ways, by (1) speeding
up the processes used, (2) using processes in simultaneous combination,
and (3) combining minimalist processes with other compositional styles
and techniques, such as the cyclical patterns of Stravinsky, Brahmsian
6 Chapter One
motivic transformations, the harmonic and rhythmic language of jazz, and

melodic riffs that are derived from rock and pop music. The result is a
work with a constantly evolving, restlessly optimistic, dance-like musical
surface that is supported by layers of logical processes. Unlike classic
minimalism, our attention as listeners is not necessarily drawn to the
underlying processes, since there are many interesting melodic and
rhythmic ideas decorating the surface. However, upon reflection, these
processes are not very far from the surface, and listeners can follow along
with them fairly easily, even when multiple processes occur simultaneously.
“The Yellow Pages” begins by introducing a handful of repeating
patterns that will permeate the work. The first of these is a pattern of
ascending fifths in the violin (see Example 1-1). A one-measure ostinato
pattern (taken from the bass line of a song by Chaka Khan) then enters in
the cello in m. 5 (see Example 1-2). An important contour motive, which
can be represented as [0 1 0 / 0 1 2], is also introduced in m. 5, passed
back and forth between the overlapping piano and violin patterns (see
Example 1-3). One additional two-measure pattern, shared by the flute and
clarinet, begins in m. 7 (see Example 1-4). In its initial form it is only
active near the barline. After the introduction of these basic patterns, the
process of systematically altering the patterns begins.
Example 1-1, “Yellow Pages,” violin pattern, m. 4

Copyright © 1995 by Adjustable Music. Bill Holab Music, Sole Agent. All Rights
Reserved. Used by permission.
Example 1-2, “Yellow Pages,” ostinato pattern, m. 5

Making the Most of Minimalism 7
Example 1-3, “Yellow Pages,” [0 1 0 / 0 1 2] contour motive in violin and piano

patterns, m. 5
Example 1-4, “Yellow Pages,” initial woodwind pattern, m. 7
The process from which “The Yellow Pages” derives its name occurs
in the opening section of the work, and also reappears in each of the other
movements. This is a process that Torke calls “static transposition” (see
Table 1-1).5 For each two bars of music in this section, a sharp is added in
the order of the circle of fifths. However, none of the patterns are actually
transposed. Each pattern retains the same note names, adding accidentals
only when required by the process. For example, in mm. 11-12, the first
sharp, C#, is added. Here, the patterns continue as they did before, but all
of the Cs are replaced by C#s. The next accidental in the circle of fifths,
G#, is substituted for all of the Gs in mm. 13-14, and so on, continuing
around the circle. Example 1-5 provides the first few phases (mm. 9-16) of
this substitutional process for the ostinato part in the cello.
8 Chapter One
Table 1-1, Static transposition in “The Yellow Pages,” mm. 9-386
Example 1-5, “Yellow Pages,” ostinato pattern, mm. 9-16
As this slow transpositional process continues, the listener will

experience a gradual series of tonal transformations: G major is eventually
replaced by G# minor, which changes into G# major (enharmonically
respelled as Ab major), and so on. When the pitches of G major are
regained in m. 33 (after having gone through the entire circle of fifths),
Torke terminates the process by going directly back to G major, shifting
all of the parts down a major second. This return firmly reestablishes the
home key and has a framing effect on the section as a whole.7
The composer compares this slow transpositional process to the

process of thumbing through the yellow pages in a telephone book, with
the categories of phone numbers changing slowly from “Automobile
Wrecking” to “Aviation Consultants” to “Awnings and Canopies.” 8
Although Torke’s metaphor is evocative, it is not entirely apt, since in
thumbing through the yellow pages, only the initial letter or letters remain
the same—all of the other letters change rather unpredictably. This is not
very descriptive of the process in this piece, wherein only one pitch
changes at a time.
A more fitting metaphor, in my opinion, is the word game called a
“Laddergram” or a “Word Chain,” which involves changing one word into
another by altering only one letter at a time (see Figure 1-1). For example,
to change MILK into PAIL, we could use the series MILK - SILK - SILL -
PILL - PALL - PAIL. To change MICE into RATS, we can use the series
MICE - RICE - RACE - RATE - RATS. Although not completely
systematic, this type of transformation is much more representative of the
type of process that occurs in “The Yellow Pages.”
MILK to PAIL MICE to RATS

MILK MICE
SILK RICE
SILL RACE
PILL RATE
PALL RATS
PAIL
Figure 1-1, Sample “Laddergrams” or “Word Chains”
This process of “static transposition” is simultaneously combined with

two other processes in this passage (see Figure 1-2). The first of these is an
additive process in the woodwinds. At each repetition of the two-measure
pattern, a few additional notes are added, replacing rests in the previous
instance (see the last column in Table 1-1). By m. 29, the pattern has
reached its final state (as illustrated in Example 1-6), having been
transformed from a few syncopated phrases in the background to an
elaborately designed, flowing melody that is very much in the foreground.
The process occurring in the violin part is opposite that of the
woodwinds—for each two measures of music, a note is subtracted from its
pattern. This subtracted note always corresponds to the pitch being altered
in the other parts (see Table 1-1). For example, in m. 11, C# is the added
pitch; at this point, all Cs are eliminated from the violin pattern. In m. 13,
all Gs are removed, and so on. The previously prominent violin part thus
10 Chapter One
becomes progressively sketchier until it disappears entirely (see Example

1-7). The simultaneous combination of these three processes—static
transposition, the additive woodwind process, and the subtractive violin
process—gives the listener plenty to attend to. It may not be possible to
hear all of these changes taking place in a single hearing.
Figure 1-2, Simultaneous processes in “The Yellow Pages,” mm. 9-38
becomes
Example 1-6, “Yellow Pages,” initial and final woodwind patterns, mm. 9-10 / 29-
30
Example 1-7, “Yellow Pages,” initial and final violin patterns, mm. 9-10 / 25-26
The two-measure pattern that was developed by the woodwinds in this

first section (see Example 1-6 above) is taken directly by the piano at letter
B (see Example 1-8), effectively linking the sections and creating a sense
of continuity. The texture in this section is monophonic, with the complete
melodic line given only to the piano. The other parts pass pieces of the
melody back and forth between themselves. The effect is somewhat
pointillistic in the upper parts, although smoothly supported by the full
melodic line of the piano.
A thinning process takes place in this section, which allows an offbeat
accent to emerge strongly, leading to the perception of a shifted downbeat.
This gradual process of shifting begins with the addition of a series of
offbeat chords in m. 43 (see Example 1-9a). In the next two-measure unit,9
the upper voices drop the melodic line and align themselves with the
syncopated chords. It is at or near this point that the sense of downbeat
changes, with the perceived downbeat shifting one-half of a beat to the
right, coinciding with the beginning of the syncopated chordal pattern.
Because of the regularity and strength of the accented offbeats, the
sixteenth-sixteenth-eighth pattern in the piano is easily reinterpreted as
eighth-sixteenth-sixteenth (see Example 1-9b). This perceived shift of
downbeat is further strengthened by the complete disappearance of the
melody at m. 47 and confirmed by the entry of the ostinato pattern on the
offbeat rather than on the notated downbeat in m. 51. I typically find it
very hard not to shift my metric focus to the offbeats when I listen to this
passage.
12 Chapter One
Example 1-8, “Yellow Pages,” beginning of the second section, m. 39
a.
b.
Example 1-9, Emergence of a shifted downbeat in “The Yellow Pages,” m.43

Example 1-10, Alternating meters section in “The Yellow Pages,” m. 56
In the next section (at letter C) a process of alternation begins between

the real (notated) downbeat and the “perceived” downbeat (see Example 1-
10). The first such occurrence catches the listener off guard (in m. 56);
here, the downbeat of the familiar ostinato pattern seems to have arrived
“too soon.” Example 1-10 illustrates how certain notes have been omitted
and inserted in order to create the effect of shifting between a measure of
7/8 and a measure of 9/8. As quirky as this metric scheme may be, its
alternating pattern continues to underlie the work for the next thirty-six
measures.
Accompanying the ostinato in the right hand of the piano are the
syncopated chords derived from the previous section, clearly outlining the
contour pattern [0 1 0 / 0 1 2] (see Example 1-11).
Example 1-11, Successive levels of elaboration in “The Yellow Pages,” mm. 55-69
14 Chapter One
This pattern, which remained in the background in the first section, now
moves to the forefront. When the violin is added in m. 59, its line is a
simple embellishment of this pattern in the piano (see the violin part in
Example 1-11). The entry of the clarinet in m. 63 in turn embellishes the
violin line. And the flute enters in m. 67 with an elaboration of the clarinet
part. The process of creating successive levels of elaboration of a single
idea establishes a high degree of unity between the parts when they are all
combined at m. 67.
Even before these parts are combined, a new process has begun in the
cello and piano: a cyclic harmonic process similar to the one used in the
opening section. But instead of motion by fifths every two measures, this
section involves harmonic motion downward by a third every four
measures. It begins by moving from G major to E major in m. 63, to C
major in m. 67, then to A major, F Major, D major, and B major, arriving
at G major again in m. 87. This process differs from the previous one in
that the ostinato pattern does modulate to the keys in the cycle. But as the
cello and left-hand piano move through a series of third-related keys, the
other parts once again remain stationary, preserving their note names while
incorporating the required accidentals. Thus, their relationship to the tonal
center given by the ostinato is constantly changing. As the harmonic
sequence continues, a canon ensues at letter D (mm. 71-90 in the score),
with successive canonic entries in the piano, violin, clarinet, and flute.
The character of the music beginning at letter E (m. 91) is much more
developmental. Once again, the monophonic piano melody is taken from
the idea that was developed by the woodwinds in the previous section, and
the upper voices pass pieces of the melody back and forth. But there is also
another process that begins here. The four upper parts begin to move
within the narrow confines of repeating motivic cells. The range of each
cell is only three pitches. For example, in m. 91 (see Example 1-12), the
pitches A B C are used by the flute, D E F by the clarinet, and B C D by
the violin and cello. 10 These motivic patterns are all derived from the
monophonic melody in the piano and proceed at different speeds in each
voice. Each pattern consists of four notes in the contour pattern [1 0 1 2].
This contour can be understood as a contraction of the original [0 1 0 / 0 1
2] pattern.
At letter F (m. 103), repetitions of this motivic cell appear in
hierarchical augmentation levels, with the cello and flute patterns lasting
roughly one measure, the violin two measures, the clarinet four measures
and the left-hand piano in the low register lasting eight measures (see
Figure 1-3). At the same time, a shifting process is taking place, with each
repeating pattern dropping its final pitch.
Example 1-12, Repeating motivic cells, each with a [1 0 1 2] contour motive, in

“The Yellow Pages,” m. 91
16 Chapter One
Figure 1-3, Shifting hierarchical layers in “The Yellow Pages,” mm. 103-08
For example, the melodic idea in the flute omits its final sixteenth note,
shifting the beginning of each repetition to the left by one sixteenth (see
Example 1-13). Similarly, the cello part shifts one sixteenth to the left on
each repetition, the violin shifts one eighth to the left, and the clarinet
shifts one quarter to the left. As Figure 1-3 illustrates, all of the parts
(except for the steady piano part), slowly begin to drift away from the
barline. Although there is clearly some temporal coordination between the
upper shifting voices (which are aligned at their beginnings), the overall
aural effect is that things never quite line up the same. At the end of each
eight-measure unit, a fragment of the ostinato idea from the opening
section interjects and the process of shifting layers begins anew. Elements
from the opening section continue to appear throughout this developmental
section, hinting at an eventual return.11
Example 1-13, Shifting rhythmic patterns in “The Yellow Pages,” m. 106
Example 1-14, Embedded piano melody in the upper parts, “The Yellow Pages,”
m. 149
18 Chapter One
The expected return arrives in m. 143 with a fortissimo statement of

the woodwind pattern from the opening section (m. 35) in vigorous
rhythms and a full texture, accompanied by a decisive return to G as the
tonal center. This woodwind pattern is also embedded within the
contrapuntally conceived upper lines at letter I. This becomes especially
apparent in m. 149, where the full woodwind melody is played by the
piano, and each of its notes is doubled somewhere in the upper parts (see
Example 1-14).
A subtractive process begins to take place in m. 153. The melody here
is rhythmically sparser, comparable to the state of the woodwind melody
in m. 21. In m. 157, the melody is comparable to the version at m. 15, and
the melody in m. 161 is the same as the woodwind pattern in its initial
state in m. 7. Thus, instead of starting with an abbreviated pattern and
building up to a full melody, as in the opening section, this passage
pursues the reverse process, beginning with the full pattern and reducing it
to its initial abbreviated form. These successive reductions of the melody
continue to be embedded within the contrapuntal lines of the upper parts,
which become less and less active as the section progresses.
Example 1-15, Closing theme in “The Yellow Pages,” m. 159
Throughout this process of reduction, a new idea has been emerging—

a slow descending line that acts as a closing theme (see Example 1-15). It
began in the cello in m. 147 and reaches its final form in m. 159, where it
is reinforced by the violin. As the upper lines wither away, this closing
theme does not diminish in strength. By m. 163, only two ideas remain:
the new closing theme in the strings and the initial woodwind pattern in
the piano and winds. The transition to the final section at letter J (in m.
167) is quite smooth. The woodwind pattern is retained in the flute and
clarinet (see Example 1-4), the cello comes in with the ostinato (Example
1-2), and the piano returns to its initial accompaniment pattern (Example
1-3). The violin carries over the descending closing theme, which is
combined nicely with this opening material.
When Torke composed the second and third movements of Telephone
Book ten years after writing “The Yellow Pages,” he returned to the idea
of static transposition in combination with other processes. In the
composer’s note in the score, he states that “each movement explores a
slightly different application of this treatment.” 12 I would now like to
briefly examine this “slightly different application” of the process of static

transposition in these two latter movements.
Most of the second movement, “The Blue Pages,” (which is written in
a decisively bluesy style) involves a single set of simultaneous processes,
once again combining static transposition with additive and subtractive
rhythmic processes (See Figure 1-4). But here, only some of the parts
remain static (keeping the same letter names) while the others participate
in a harmonic sequence. And rather than adding and subtracting pitches
within a unit of unvarying length (such as the two-measure units in “The
Yellow Pages”), here Torke allows his basic units to expand and contract.
“The Blue Pages” constantly alternates between two contrasting ideas.
The first is the principal melody—a bluesy theme, made up of two
phrases, which at each repetition is one measure shorter in length. Torke
shortens it by removing one beat from the beginning and end of each of its
two phrases (see Example 1-16). The second idea is a freely-swinging
transitional theme, which at each repetition is one measure longer. He
lengthens it by adding two beats of music to the beginning and to the end
(see Example 1-17). Thus, these two themes expand or contract
progressively from their middles.
Figure 1-4, Simultaneous processes in “The Blue Pages,” mm. 7-73

20 Chapter One
Example 1-16, Initial transformations of the bluesy main theme in “The Blue
Pages,” mm. 7-22.
becomes
Example 1-17, Swinging transitional theme in “The Blue Pages,” mm. 15 and 23-
24
At each repetition of the principal theme, the accompanimental parts—

an oscillating piano vamp and bluesy bass line—move up by a major
second, shifting the tonal center from E Dorian to F# Dorian to G# Dorian,
and so on, until E Dorian is re-attained (in m. 66). Each of these shifts
adds two new sharps to the key signature. Thus, the static transposition
process is more accelerated in this movement (as shown in Table 1-2). For
example, in m. 16, with the shift to F# Dorian, all of the Gs and Ds in the
main theme (and in the transitional theme) are replaced with G#s and D#s
(see Example 1-16). Then in m. 25, with the shift to G# Dorian, all of the
Es and As are replaced with E#s and A#s. And so on. Table 1-2
summarizes all of the processes that occur in this passage.13
22 Chapter One
Table 1-2, Static transposition in “The Blue Pages,” mm. 7-73
In the third movement, entitled “The White Pages,” Torke again

combines static transposition with additive processes, but also incorporates
progressively increasing rhythmic shifts (see Figure 1-5).
Figure 1-5, Simultaneous processes in “The White Pages,” mm. 13-32

There are two principal ideas presented in the opening section: a quick
woodwind flourish played by the flute and the clarinet (see Example 1-18)
and a rock-style piano accompaniment (doubled in the strings—see
Example 1-19). Even before the process of static transposition begins, the
doubled parts begin to drift apart from each other rhythmically. For
example, in m. 14, the piano shifts an eighth note ahead of its doubled part
in the strings. This displacement process also occurs between the clarinet
and the flute. In the course of time, the displacement between matched
parts increases to a quarter duration, a dotted quarter, and a half (as shown
in Example 1-20).
Example 1-18, Woodwind flourish in “The White Pages,” mm. 4-6

Example 1-19, Rock piano accompaniment in “The White Pages,” mm. 5-6
Example 1-20, Rhythmic displacement between the woodwind flourishes in “The

White Pages,” mm. 16-31
24 Chapter One
The process of static transposition begins in m. 19, adding one sharp at

a time around the circle of fifths (this can be seen in Example 1-20,
although all of the steps in the process are not shown). In this case, all of
the parts retain the same note names. At the same time, an additive process
is taking place in the flute and string parts, which become rhythmically
more active, replacing rests with notes as the music progresses. For
example, the flute part in Example 1-20 grows by prepending two notes in
m. 22, four notes in m. 26 and six notes in m. 30. All of these processes
are summarized in Table 1-3.
Table 1-3, Static transposition in “The White Pages,” mm. 13-32
In the “Yellow Pages” and the “Blue Pages,” static transposition

always involved the ascending circle of fifths, adding one or two sharps at
a time. In “The White Pages,” Torke again uses the ascending circle of
fifths in the opening section, but also experiments with other approaches to
static transposition. For example, in mm. 113-122, he traverses the circle
of fifths in reverse order, adding one flat at a time, as illustrated in Table
1-4. Similarly, in mm. 32-54 he uses a key signature transformation that
adds three flats at a time, moving more quickly through this reverse-order
circle of fifths.
Table 1-4, A passage of static transposition involving descending fifths in “The

White Pages,” mm. 113-122
The appeal of Torke’s Telephone Book is due not only to its catchy
rhythms and bright, optimistic tone. It is also supported by layers of
interesting processes and is unified by musical materials that are
interrelated in many ways. In the liner notes to Overnight Mail, Torke
states:
It is genius to be able to put together the elements of music in such a way

that’s immediately satisfying, and it’s surprising how much consensus will
follow. But for me, that’s only half the battle. Once I find something that I
believe is arresting, I want to develop this material to create a structure that
on repeated listenings will have depth and weight. Presentation isn’t
enough for me.14
With “The Yellow Pages,” Torke has struck a very good balance: while
the piece is pleasing to the senses, it is also interesting to the mind, and at
the same time, is not challenging to the ear. The profusion of musical
26 Chapter One
processes in this work, as well as the way in which these processes are
sped up and simultaneously combined, creates an actively changing,
dynamic surface that presents the listener with something new upon each
hearing. In other words, “The Yellow Pages” clearly makes the most of the
minimalist processes on which it is based.
Notes
1
For an overview of the differences between minimalism as an aesthetic, as a style,
and as a technique, see Timothy A. Johnson, “Minimalism: Aesthetic, Style, or
Technique?” The Musical Quarterly 78, no. 4 (1994): 742-773.
2
Readers may want to follow along with the published score of Telephone Book
(New York: Boosey & Hawkes, 1995).
3
“Synesthesia,” <http://en.wikipedia.org/wiki/Synesthesia>, accessed February 24,
2010.
4
“Michael Torke,” <http://www.michaeltorke.com/compositions.php>, accessed
February 24, 2010.
5
Julian Hook has described this passage in terms of signature transformations. See
“Signature Transformations” in Music Theory and Mathematics: Chords,
Collections, and Transformations, ed. Jack Douthett, Martha M. Hyde, and Charles
J. Smith (University of Rochester Press, 2008).
6
The “Other Processes” shown in this table (an additive process and a subtractive
process) will be discussed below.
7
Note that the harmonic motion governing this section is represented in miniature
by the violin motive that initiates the section (see Example 1-1). Its structural
pitches (G D A E B) correspond to the first five key areas in the cycle of fifths.
8
Torke, liner notes in Overnight Mail (Argo 455 684), 7.
9
As in the initial section, events here continue to occur in two-measure units.
10
The tonic pitch G is the only one missing here, and correspondingly the sense of
tonal stability is weakened.
11
For example, the rhythms in the right-hand piano in this section are clearly
derived from the opening woodwind pattern.
12
Torke, “Note by the Composer,” in Telephone Book, Adjustable Music, 1997.
13
Note that the alternating pattern breaks off when a 2-measure theme is reached.
This is instead followed by a 1-measure version of the theme, a transition, an 8-
measure return of the full theme in the home key, and a full 8-measure transition.
14
Torke, liner notes in Overnight Mail (Argo 455 684), 7.
CHAPTER TWO
COHERENCE AND COMPREHENSIBILITY

IN JOAN TOWER’S PURPLE RHAPSODY
ANDREW GADES
Every instant of music has a past, a present and a future. The present, of
course, is what’s happening at a given moment. The past is everything
that’s come before, everything that’s led up to that moment. For me, it’s
very important that the present grow out of the past, that past and present
combined contain the seeds of the future. As a piece goes on, it develops
more and more past: it takes on more shape, and the more shape it has the
more you know about where it’s headed. It’s like a tree: When it first
sprouts you don’t know how it’s going to grow, but after it’s been growing
for a few years you have a pretty good idea of what it will grow into.1
Introduction
For a contemporary American composer, Joan Tower has received
considerable attention. In particular, Tower’s interest in organic metaphors
has been well documented in interviews, in articles by music critics, as
well as in Ellen Grolman’s Comprehensive Bio-Bibliography (2007) and
Denise Von Glahn’s Skillful Listeners: American Women Composing Nature
(2013). Indeed, Joan Tower often refers to her compositional process with
organic metaphors, and the titles of many of her compositions, including
Sequoia, Wings, and White Granite, reflect this interest in nature.
Tower did not give Purple Rhapsody (2005) a nature-oriented title,
however. As a viola concerto, Purple Rhapsody joins Tower’s other works
for viola, Wild Purple (2001) and Simply Purple (2008). The color purple
is a common theme for Tower’s pieces for viola—in her own words, “I’ve
always thought of the viola sound as being the color purple. Its deep,
resonant, and luscious timbre seems to embody all kinds of hues of
purple.”2 While not suggested by the title, Purple Rhapsody contains many
28 Chapter Two
of the organic elements Tower describes as part of her compositional

process.
The analytical literature on Joan Tower’s music, while sparse, also
focuses on aspects of organic unity. Important contributions in this area
include Judy Lochhead’s article, “Joan Tower’s Wings and Breakfast
Rhythms” (1992) and Ryan McClelland’s paper, “Melodic Process and
Parallelism in Joan Tower’s Silver Ladders” (2002). Although different in
analytical method, the resulting observations have some similarities.
Lochhead focuses on repetition and its role in creating and perceiving
formal types. Of particular relevance, Lochhead suggests that “repetition
not only acts as a temporal marker through its reference to a prior unit or
event but also retrospectively shapes the earlier occurrence as well as
itself.”3 Repetition has similar formal functions in Purple Rhapsody, but
my analysis compares the similarity of the repetitions in order to derive a
common idea and demonstrate the development of shared motivic
material.
McClelland utilized a salience-based method to chart large-scale linear
trajectories and unveil parallelism. The following analysis also uncovers
motivic parallelism, but does not rely on salience criteria. Rather, the
parallelism I observe in Purple Rhapsody is an outgrowth of investigating
the motivic material at the musical surface.
The goal of my analysis is to determine the motivic “seeds” in Purple
Rhapsody, and the transformations of musical material throughout the
course of the piece. This approach was inspired by Tower’s statements
about her compositional process:
You see, I don’t do sketches in advance. I do start out with a basic idea,
but I’m not very “pre-compositional” in my thinking. I used to be, but that
was because I felt insecure and needed some sort of map to get me through
the infinity of choices that were available. Now I’m more of an “organic”
composer [emphasis added].4
Tower’s description of her compositional process is suggestive of

developing variation, and my analysis draws upon Schoenberg’s concept
of the “musical idea.” By first considering a single theme and breaking it
down into its constituent motives, and demonstrating how the motive is
used and developed throughout the piece, I aim to identify the generative
basic idea in Purple Rhapsody. After determining the basic idea, it is then
possible to consider how it is used to shape the music as it progresses.
In working with Purple Rhapsody and attempting to identify a basic
idea from which the piece progresses, I draw heavily on Schoenberg’s
Coherence and Comprehensibility in Joan Tower’s Purple Rhapsody 29
discussions of musical idea, coherence, comprehensibility, Gestalt, and

Grundgestalt.
[Schoenberg] conceives of the whole as a balance of forces: the unrest

inherent in the material, the imbalance produced by such unrest, and the
restoration of balance. The idea is the contrast that challenges the state of
rest—and by means of which that state is restored.5
In this way, Schoenberg saw a musical work as “a totality because it is the

realization of a thought.”6
Overview of Purple Rhapsody

Like many of Tower’s compositions, Purple Rhapsody is a single-
movement work. It is one of Tower’s longer pieces, with a performance
time of about nineteen minutes. The work is sectional within its single
movement, and these divisions can be made on the basis of tempo and
other considerations.
Tower has included thirty-six specific metronome markings in the
score. Usually these tempo changes accompany changes in pitch
collection, pitch center, theme, and so forth; however, this is not always
the case. Sometimes tempo changes are almost imperceptible due to drawn
out ritardandi and accelerandi, or because frequent and changing tuplet
patterns disguise any consistent rhythmic pulse. In other instances,
discernible changes are created by shifts in texture, timbre, or pitch
collections without a change in tempo. Nevertheless, the tempo indications
provide a useful starting point for dividing the piece into smaller sections.
These initial divisions can then be further divided or grouped together
based on pitch content, thematic material, repetition, and cadence points.
Following thematic and tempo similarities, Purple Rhapsody divides
into four large sections as shown in Table 2-1. The A sections are slower
and more melodic than the B sections. Both the A and the B sections
contain two main thematic ideas, and within each theme there are two
motivic ideas that are used consistently throughout: a neighbor figure and
scalar runs. These two ideas unify the entire work, permeating the musical
surface and deeper structure. They form a connection based on common
musical ideas between the otherwise seemingly-contrasting sections.
30 Chapter Two
A B Ac Bc
1-99 100-246 247-318 319-452
Table 2-1, Formal diagram of Purple Rhapsody
In order to determine a basic idea that provides unity and cohesion, the
themes must be broken down into their constituent motivic ideas. In the
Fundamentals of Musical Composition, Schoenberg describes a motive by
saying:
Even the writing of simple phrases involves the invention and use of
motives, though perhaps unconsciously. Consciously used, the motive
should produce unity, relationship, coherence, logic, comprehensibility,
and fluency. The motive generally appears in a characteristic and
impressive manner at the beginning of a piece. The features of a motive are
intervals and rhythms combined to produce a memorable shape or contour
that usually implies an inherent harmony. Inasmuch as almost every figure
within a piece reveals some relationship to it, the basic motive is often
considered the “germ” of the idea. Since it includes elements, at least, of
every subsequent musical figure, one could consider it the “smallest
common multiple.” And since it is included in every subsequent figure, it
could be considered the “greatest common factor.”7
Melodic Motives
The A section
The very first motive played by the solo viola in Purple Rhapsody is a
prominent and distinctive feature of the slower A sections, a lower
neighbor motion of a whole-step which is used through the first four
measures as shown in Example 2-1. Tower deliberately turns the otherwise
unremarkable interval of a whole step into an important motivic idea by
singling it out in this way. The first change in this initial theme is the
introduction of a tritone leap to F#, from which another whole-step
neighbor motion spins off in mm. 5-7.
Example 2-1, First solo viola theme with melodic motives marked, mm. 1-11
At the same time, the D pedals played by the orchestra, together with
the consistent return to D by the viola in the first four measures, establish
pitch centricity on D. The scale or pitch collection remains ambiguous
until m. 8, as the only pitch classes to that point have been C, D, F#, and
G#. These pitch classes suggest a whole-tone collection, especially with
the emphasis placed on the interval of a major second.
The ambiguity disappears in m. 8 with the introduction of another
portion of the [0, 2] octatonic scale, linking F# and G# with another D-C-
D neighbor motion an octave above the initial D3-C3 pitches. In this way,
the first theme can be understood as motive a, a whole-step neighbor
motion on D and C, passing through a similar motion on F# and G# as it
ascends an octave, linked together by motive b, portions of an octatonic
scale. Example 2-1 shows the first theme with the neighbor and scalar
motives marked as motives a and b respectively.
This theme returns, albeit slightly altered, three additional times in the
A section, and then once more at the beginning of the Ac section. The first
three restatements of this theme are shortened and occur at different pitch
levels. In m. 57, shown in Example 2-2, only the first few measures
resemble the initial theme before breaking into different material.
Although significantly shorter, this recurrence of the initial theme contains
the necessary melodic motives that contribute to its comprehensibility. It
begins with whole-step lower neighbors and then moves on to an octatonic
scale in a lower voice as a pedal tone is maintained above it. Measures 77-
81 are similar to mm. 57-60, but the cello and bass play the theme, set
against the rest of the orchestra playing first chromatic and then whole-
tone scales—usually signifying a transitional section.
32 Chapter Two
Example 2-2, Solo viola, mm. 57-60
The last return of this theme in the A section, shown in Example 2-3, is
marked by further reduction of the theme statement. The distinctive whole-
step neighbor motion is still used to begin the theme, but it is followed by
leaps of a perfect fourth and an augmented fourth rather than stepwise
scalar motion through an octatonic collection.
Elsewhere, Tower uses inversion and transformation to change the

descending whole-step neighbor into an ascending half-step neighbor. The
first example of this variation occurs in m. 30, shown in Example 2-4, with
the variation of the motive marked as ac. The similarities between this
section and the whole-step theme are striking; both begin with repeated
neighbor motives and then expand with scalar motion following the
octatonic scale. This theme is the second theme presented in the piece, and
its proximity to the initial whole-step theme reinforces the perception of
the half-step theme as a variation. As with the whole-step theme, later
recurrences are fragmented and simplified.

This half-step theme is invoked again in mm. 48-55. At first glance,

this portion of music would seem to be merely an ascending octatonic
scale, however, the ascent includes half-step neighbor patterns.
Additionally, slurs connect half-step groupings together, strengthening the
perception of the half-step motions. Not only do these slurs reinforce the
neighbor motions as the scale ascends, but they emphasize all half-step
motion within the ascent of the octatonic scale. Most importantly, this
instance of the theme begins with a neighbor motive that connects it to
earlier variations.
Together, these themes dominate the melodic landscape of the A
section. All other areas of the A section are transitional or subordinate
areas consisting mainly of either chromatic or whole-tone scalar motion,
lacking a distinctive thematic idea. While the thematic areas often have a
stable pitch center, the transitional sections are unstable. Tower creates
additional contrast in texture; the solo viola presents the themes and the
transitions include more of the orchestra—reminiscent of traditional
concerto-ritornello form, but with the solo sections rather than tutti
returning to the theme. Even these transitional areas, however, refer back
to the main themes. For example, the whole-tone section in mm. 44-47
begins with the same whole-step pattern with dotted rhythm from C to D
of m. 1.
Table 2-2 shows the overall form for the A section. The section begins
with the whole-step theme and then transitions into the half-step theme,
which remains prominent until m. 57. At this point, the whole-step theme
attempts to return, but as mentioned earlier, these returns are incomplete,
fragmented, and without a clear octatonic pitch collection for support. It is
not until m. 94 that the section closes with a fairly recognizable whole-step
theme, although still incomplete. This incomplete and fragmented return
denies a sense of closure, especially since none of the themes return with
the original pitch-class center of D. The only return to D is the final
cadence at the end of the section, providing some closure.
34 Chapter Two
Measures Melodic content Pitch collection

1-11 whole-step theme [0,2] octatonic
12-20 transition chromatic
21-29 transition odd whole-tone
30-36 half-step theme [0, 1, 3, 4] and [7, 8, t, e]
37-41 half-step theme chromatic
41-47 transition even whole-tone
48-56 half-step theme [0,1] and [0,2] octatonic
chromatic with octatonic
57-76 whole-step theme -> transition
subsets
77-85 whole-step theme and transition chromatic and whole-tone
fragments of whole-, and half-

86-93 octatonic subsets
step themes
94-99 whole-step theme [1,2] octatonic
Table 2-2, Formal structure of the A section
The Ac Section
As shown in Example 2-5, the Ac section begins with the same whole-
step theme from the A section, but this time in a brief canon between the
solo viola and two cellos. The upper cello part bears the closest
resemblance to the first whole-step theme from mm. 1-11, alternating
between D and C before moving to F# and G# in m. 251 and eventually
moving up in m. 354 to the C an octave above, as in the original theme.
The half-step variant returns in mm. 265-272, shown in Example 2-6.
Here, Tower has transposed the theme up a tritone, which reverses the
order in which pairs of pitches appear when compared with m. 30. In m.
30, the theme started with G and Ab moving up a tritone to C# and D#,
ending with a whole-step motion. At m. 265, the theme begins with C#
and D, moving up a tritone to G and Ab. The scalar runs used in the
transitions become another variation of the b motive, although Tower uses
them to embellish the structural melody of the theme, rather than forming
an intrinsic part of the theme itself.
Example 2-5, Viola and cellos, mm. 247-54
The two themes are combined in another duet for the final thematic
statement in mm. 286-90, again embellished and quickly fragmented,
moving to a cadence in m. 316. As shown in Example 2-7, both parts start
with some resemblance to either the half-step or whole-step theme, but
quickly break apart into descending octatonic scales. Tower has combined
elements of the whole-step and half-step themes, the characteristic features
of the initial descending whole-step or ascending half-step neighbor
patterns giving way to a scalar motion. In this instance, the basic outline
has been embellished by short scale segments, as in the earlier themes of
the Ac section. This synthesis does not last, though, as the duet becomes
increasingly frantic, fragmented, and chromatic, leading to a relatively
consonant cadence in mm. 314-16.
36 Chapter Two
Example 2-7, Solo viola and violin, mm. 286-88
The B Sections
The B sections are characterized by a faster tempo and more driving,
frenetic passagework, which creates a marked contrast from the A
sections. The constant motion present in the B section removes some
thematic distinctiveness, and for that reason, subsections of the B section
are easier to identify by changes of texture and timbre rather than tracing
thematic elements. While the B sections are relatively athematic when
compared to the A sections, melodic motives are still important.
The most distinctive sound of the B sections is the solo viola’s sul
ponticello, which typifies mm. 108-30 and mm. 204-37. The sul ponticello,
playing on or near the bridge, introduces a new and different timbre that
sets these portions of music apart. The first sul pont. section begins with
the solo viola quickly alternating among C#, F#, and G, as in Example 2-8.
The beginning of the B section, mm. 100-07, introduces this instance of
set-class (016) with support from the rest of the orchestra playing groups
of either (016) or (0167). By m. 110, this pattern breaks into an ascending
[0, 1] octatonic scale. These two ideas, the fast alternation among pitch-
classes of a given set and the octatonic scale, form the basis for the
musical content of the B section and are shown in Example 2-8 as motives
x and y, respectively.

The second sul pont. section is similar to the first, but it begins with a
different pitch-class set before starting its ascending scale. Previously,
(016) or (0167) were featured, but in mm. 204ff Tower uses (01), a subset
common to both.
After the first sul pont. section, a solo clarinet follows by expanding on
the x motive. As shown in Example 2-10, it begins with the same set
classes, (016) and (0167), that were employed earlier. All melodic material
is limited to the specific transposition of [0, 1, 6, 7] for several measures,
sustaining this pc set. Gradually, the pitch classes are removed until a
dyad, pc-set [6, 7], remains. Set [0, 1, 6, 7] drops the C# and leaves [0, 6,
7] to continue until the C is removed as well. Tower uses this liquidation
process to move to different set classes or different transpositions of the
same set class. The clarinet moves to (014), using the common dyad (03)
to move to (0235).
Example 2-10, Clarinet, mm. 131-37
As seen in Example 2-11, the strong tension of the (0167) relaxes as it

moves through (014) to (0235). At the same time that this reduction of
tension takes place, the energy increases in tempo, dynamic, and texture.
The music reaches a high point in m. 150 as the (0235) reduces first to
(03) and then to (02). While the emphasized interval narrows, the texture
thickens as it approaches a full octatonic collection in m. 151. This
progression is shown in Example 2-11a, which gives a reduction showing
the pitch-class collections used in the B section to this point. Example 2-
38 Chapter Two
11b shows much of the same information, but as the texture thickens in
mm. 147-51, it shows only a subset of the entire collection in order to
illustrate the specific dyad set-class emphasized melodically. Both
Example 2-11a and 2-11b show the smooth voice-leading connections
Tower uses in the B section.
Throughout the B section, the most prominent set-classes are
(016)/(0167), (03), and (014)/(04). Besides liquidation and using a
common dyad to pivot, Tower also gradually expands or contracts the
interval inside a set-class. Tower uses these simple processes and a focus
on a small number of set-classes to create cohesion in the B section.
(a) Reduction showing pitch-classes in use
(b) Reduction showing set-classes given melodic emphasis
Example 2-11, Orchestral reductions, mm. 101-51
The Bc section is similar, although it progresses from an emphasis on

(0134) and (014) towards an emphasis on (016) and (0167), opposite to the
order of the B section. Also, Bc is interrupted by a cadenza in mm. 391-
412, although it maintains the emphasis on the (016) trichord.
Rhythmic Motives
Pitch relations are not the only musical criteria that create motivic
unity within a piece; rhythm plays an important role as well. Rhythm is
particularly important for Tower, as Grolman notes:
Rhythmic prominence has long been a hallmark of a Tower composition.

She attributes her fascination with rhythm to early exposure to the complex
meters of South American dance music and her personal interest in
percussion, as well as to the rhythmically intricate serial music she

performed as a pianist for so many years.8
Just as the melodic motives were formed from extremely basic ideas,
the rhythmic motives are also simple, almost trivial, ideas. However, the
emphasis Tower places on these basic ideas marks their importance—just
as with the simple neighbor motion. The basic rhythmic idea in Purple
Rhapsody is a simple long-short pattern that is connected with the opening
whole-step oscillation. The long-short motive first appears as a dotted half
note and a quarter note, but quickly transforms through retrograde,
augmentation, or diminution to different duration values.
The A Sections
The first two measures of the solo viola, shown in Example 2-1,
contain the long-short rhythmic idea, but in m. 3, a diminution of the long-
short motive is nested within the larger long-short pattern of the whole
measure. Similar diminutions embellish in mm. 9-10; these
embellishments create an element of rhythmic unity within the theme.
The repetitions of the whole-step theme include the same long-short
pattern, as seen in Examples 2-2 and 2-3, but each repetition simplifies the
rhythm and the embellishing diminutions are no longer present.
Additionally, these repetitions alter the long-short motive by beginning the
measure with a quarter-note rest.
This long-short pattern typifies the half-step theme as well. In fact, the
common usage of the long-short rhythmic motive in both themes
reinforces both the importance of the rhythmic motive and the connection
of the two themes. The similar use of the long-short motive can be
observed when comparing Examples 2-1 to 2-3 with Example 2-4.
The rhythmic motive returns with the whole-step theme in the Ac
section, but with additional alterations. As seen in Example 2-5, the initial
long-short pattern is the same, but the short note has been changed to equal
sixteenth notes interrupted by rests. This new pattern is similar to the
earlier one seen in Example 2-1, and is aurally perceived as similar, as
both have similar attack points.
The long-short motive also appears in the return of the half-step theme,
but the pattern reverses to short-long, as seen throughout Example 2-6.
The same altered diminution from Example 2-5 returns again here. When
the themes are paired in mm. 286-88, they retain the same long-short or
short-long motives as shown in Example 2-7; the whole-step theme
40 Chapter Two
contains the original long-short rhythm while the half-step theme keeps the
short-long pattern it was given in m. 265.
The transitional or subsidiary sections do not use the long-short motive
as prominently as the thematic areas—occasionally the long-short motive
does appear, but typically within the context of fragments of thematic
material. The transitions are instead marked by constant, fluid motion that
shifts between triple and duple divisions of the beat.
The B sections
Because of the constant, driving motion of the B sections, it is more
difficult to speak of a specific rhythmic idea in the B sections. They share
characteristics of the transitions in the A sections by alternating between
triple and duple divisions of the beat. However, the B sections are not
without distinctive rhythmic elements.
The motivic use of triplets can be seen in Example 2-10, but shortly
after this excerpt the triplet pattern changes from equal triplets to the long-
short pattern shown in Example 2-12. At other moments in the B sections,
duple divisions of the beat prevail. The duple divisions group into different
categories based on corresponding changes in pitch or articulation. As seen
in Example 2-8, while the rhythmic motion is constant sixteenth notes, the
perceived motion begins at the eighth-note level since the pitch changes at
that rate. The rate of change in pitch, then, is different from the rhythmic
motion. This is particularly evident in the pattern that begins in m. 113.
Example 2-13 shows this new pattern created by the repetition of pitches.
The repetition of the middle pitches creates an illusion of a short-long-
short rhythm, but this conflicts with the slurs connecting the sixteenth
notes into different pairs. The alteration to triplets removes this distinction
by eliminating the repeated pitch within each beat.
Example 2-12, Clarinet in Bb, mm. 141-42

Both triple and duple division patterns appear throughout the B

sections, and in particular, the slurred pattern seen in Example 2-13
appears in augmentation at m. 157. Strong emphasis is placed on it, since
it is played throughout the orchestra at a loud dynamic, indeed, one of the
loudest in the B section, while surrounded by softer dynamic levels on
either side. Other climactic points in the B section use the uneven triplet
pattern.
The emphasis placed on these rhythmic features helps to distinguish
them as important motives within the B sections. Both the triplet and
sixteenth-note motives reappear in the Bc section, but the slurred sixteenth-
note motive is considerably more prevalent in Bc than the long-short triplet
motive.
Coherence and Comprehensibility

Schoenberg’s interest in the motive and basic musical idea relies in
part on the question of how smaller parts relate to the whole work, how a
smaller motive can create the driving force for the larger work.
Through the connection of tones of different pitch, duration, and stress

(intensity), an unrest comes into being: a state of rest is placed in question
through a contrast. From this unrest a motion proceeds, which after the
attainment of a climax will again lead to a state of rest or to a new (new
kind of) consolidation that is equivalent to a state of rest.9
Tower shares this idea with what she calls “motivated architectural
thinking,” which she sees in the music of Beethoven. “The thing that really
interests me about Beethoven’s music is how the music is strongly
propelled forward by the inherent musical motivation of the actions and
reactions of each phrase within the long range direction of the music.”10
Comprehensibility
Comprehensibility is an important part of Schoenberg’s concept of

formal structure. According to Schoenberg, form is not needed for the
composer, but for listeners in order that they might better understand and
comprehend the content of a piece of music. In other words, Schoenberg
did not consider form to be an external frame imposed upon a piece, but
the overall shape of a piece that “articulates the musical organism so as to
make comprehensible the musical idea.”11 In this way, form is simply a
42 Chapter Two
natural and necessary outgrowth of the musical idea rather than an

arbitrary taxonomy. While Tower has said that she does not compose with
a specific form in mind, a listener can still hear a form within the piece
while attempting to comprehend what is perceived.
Comprehensibility is the property of a piece of music that allows it to
be understood—comprehended—by the listener.
A musical content is musically comprehensible when its small and smallest

parts (periods, sentences, phrases, motives) share such coherence among
one another and with the whole as would in general be required for
comprehensibility. In music, comprehensibility depends on coherence…
Coherence, however, does not always guarantee comprehensibility.12
Schoenberg provides an extensive list of laws of comprehensibility in

The Musical Idea, some of which will be discussed here in relation to
Purple Rhapsody.
Schoenberg begins his discussion of comprehensibility by talking
about main and subordinate ideas and the important differences in their
presentation. In Purple Rhapsody, the opening theme presents the main
idea, and specifically the oscillating motive a. Tower’s presentation of the
theme and its motives follows Schoenberg’s criteria because this theme is
frequently repeated, relatively stable, and the theme makes the repetitions
of the Grundgestalt apparent. These qualities primarily identify the
thematic areas in the A sections: the primary motive is repeated, apparent,
and accompanied by a stable pitch center.
The transitional material in Purple Rhapsody is, then, relegated to what
Schoenberg refers to as the subordinate idea. As Schoenberg describes,
Clearly, this motion of the subordinate idea stands out in contrast to the
state of rest of the main idea, and the abandonment of what is characteristic
has a purpose opposite to that of the repetitions of the main idea: the
secondary matters of the transition are not intended to be noticed as
something essential.13
The subordinate areas in Purple Rhapsody unsettle the music by

moving away from stable pitch centers, sliding up and down octatonic,
whole-tone, or chromatic scales without forming a specific pitch center.
They also lack the characteristic features of the main theme, consisting
primarily of scale segments.
While the presentation of material and formal structure can contribute
to comprehensibility, coherence is a necessary component. Two of
Schoenberg’s laws of comprehensibility rely heavily on coherence,
requiring both a clarity of the coherent parts themselves as well as how

they are connected to each other.14
Coherence
“Coherence in general refers to conditions that bind together an object,
bringing its components into a meaningful interaction.” 15 While the
various melodic and rhythmic components of Purple Rhapsody have been
described above, the question of how they relate to each other remains. Its
themes and motives do have a high level of coherence, which in turn
contributes to the comprehensibility of the work.
After listening and analysis, the neighbor figure is the obvious choice
for a central musical idea in Purple Rhapsody, particularly when paired
with the dotted rhythmic figure. Its several repetitions in the A sections
reinforces its primacy. Tower develops the motives gradually in the A
section as the themes return, but alters them with each new appearance.
While the A and B sections appear to be unrelated at first glance, the
changes Tower makes in the A section isolate specific parts of the motive
that are extensively used in the B sections.
Consider mm. 1-4 and mm. 30-33, shown in Examples 2-1 and 2-4
respectively; the neighbor figure emphasizes a single pitch. Even the
added leap of a third and the following triplet in mm. 32-33 merely expand
the neighbor motion and return to the original pitch. In both m. 5 and m.
34, the stasis is broken with leaps of a tritone, which are then sustained
with additional neighbor motion. The alterations made in mm. 30-36 allow
Tower to begin integrating the neighbor motion with the scalar motion.
The combination of the two ideas occurs again in the repetitions of the
theme, but one alteration that is consistent in later sections is the leaping
third following either the initial half- or whole-step neighbor figure.
Example 2-14 shows the appearance of this expanded neighbor motion.
Instead of oscillating with an interval of a second, it now uses a third. In
mm. 31-32, the familiar dotted rhythm reverses as the half-step expands to
the minor third. Measures 58-59 retain the interval of a minor third, but the
pattern is inverted—descending instead of ascending. The rhythm is
altered by adding a note in m. 59 to allow the return to E on the fourth
beat. The same applies to mm. 78-79 where another note is also added in
m. 79, repeating the C#. In this fashion, Tower gradually alters the
stepwise neighbor figure to span a third and changes the dotted rhythm to
even quarter notes.
44 Chapter Two
Example 2-14, Progression of the neighboring third
This gradual alteration allows the theme’s variations to have many

similar features. The similarities provide coherence by connecting the
more distant variation in m. 79 with the initial neighbor motive. The
gradual progression from one to the next reinforces the comprehensibility
of this connection. Tower uses the variation of mm. 78-79 frequently in
the B section, as mentioned in reference to Example 2-13. This new
variation first appears in the B section using the interval of a minor third;
the same interval used in the example above, and this is the interval most
often associated with it. This motivic relationship creates a strong
connection between the motivic ideas in the A and B sections. Tower
further adapts the motive in the B section by expanding it to other intervals
beyond the minor third, but the basic profile remains the same.
The gradual liquidation of the distinctive elements evidenced in the
final statement of the theme in the A section also sets up important
thematic characteristics seen in the B section. The last statement of the
whole-step theme in the A section, shown in Example 2-15, removes
entirely the stepwise scalar motion and replaces it with an outline of an
octatonic subset (016), a set class that is important in the B section.
Indeed, the opening line of the B section uses (016) in a manner
reminiscent of the ascending half-step motive, as shown in Example 2-15.
The use of (016)/(0167) at the beginning of the B section connects the
opening material of the B section with the closing material of the A
section. Additionally, this adaptation of the neighbor motive changes it
from the oscillation between two pitches to an alternation among three
pitches in a trichord.
Example 2-15, (016) outline
The first sul pont. area of the B section uses this new variation of the
neighbor motive. Instead of oscillating between members of the dyad (02),
the pitches now cycle among the pitches of a (016) trichord. The motivic
development employed through the A section and in the introduction of
the B section connects this material with the whole- and half-step themes
from the A section, as shown in Example 2-16. Example 2-16a shows one
of the themes from the A section, succinctly illustrating both the a and b
motives. Example 2-16b is taken from the beginning of the sul pont.
portion of the B section with motive x outlining (016) and motive y
moving up an octatonic scale. However, both themes use the basic musical
idea and have the same basic structure due to the development of motive a
that connects it to motive x.
(a) Theme from A, mm. 57-60
(b) Theme from B section, mm. 108-12
(c) Theme from B section, mm. 120-23
Example 2-16, Solo viola themes
This connection of thematic ideas is even more pronounced in Example

2-16c where a descending whole-step is used, combined with the rhythmic
idea from Example 2-14. This example appears after the excerpt from
Example 2-16b, perhaps because the earlier one was drawing from the
46 Chapter Two
material at the end of A to create a smoother transition between the two. In

any case, the theme in Example 2-16c reinforces the connection between
the material in the A and B sections, fusing together a number of different
motivic ideas. This example demonstrates Schoenberg’s idea of more
remote formulations derived from a single musical idea based on the
unifying element of one common factor. Sometimes that common factor is
a shared melodic unit that is held stable while rhythms change, or a
rhythmic idea that remains the same while the pitch content changes.
Tower uses these techniques to reach variations of the theme which do not,
at least on the musical surface, resemble its first statement. Coherence is
achieved through the gradual transformations of the rhythmic and melodic
motives’ characteristic features.
Motivated Architectural Thinking

How, then, do all of these developments fit into Tower’s idea of
“motivated architectural thinking?” Consider the main thematic idea
Tower used in various forms throughout Purple Rhapsody; it consists of a
neighbor motive and scalar motion. Within the context of Schoenberg’s
conception of a musical idea, these two motives can be considered as stasis
and motion.
The initial neighbor motive stands as a microcosm of this process; a
single static note is brought into unrest by the motion to a neighboring
note, but rest is regained with the return to the original pitch. The unrest
created by the neighbor motion manifests in the contrasting scalar idea
when the stepwise motion does not return to the stable pitch, but continues
forward with a scale. This duality and conflict set up an “action and
reaction” that creates drama and propels the music forward.
The stasis and motion exemplified in the theme is paralleled at
numerous levels in Purple Rhapsody. The establishment of pitch centricity
is one observable example, shown in Table 2-3. The first pitch center at
the beginning of the piece is D, created by a consistent drone on that pitch
class. The only deviation from the pedal D in the first eleven measures is
an echo of the descending whole-step neighbor in the low strings in mm.
5-6.
Measures Pitch Center

1-11 D, with whole-step down to C
12-20 Eb down to G
21-29 G down to C
30-41 C, with half-step up to Db
41-47 Ab down to F
48-56 F, becomes unstable
57-76 E, becomes unstable
77-85 unstable
86-93 unstable, emphasis on (0257)
94-99 cadence on D
Table 2-3, Progression of pitch centers in A
The transitional areas attempt to assert a stable pitch center, but they
are unsuccessful and devolve into scalar motion. The first transition area in
mm. 12-20 briefly moves up a half step to Eb before beginning an
octatonic descent to G. The next transitional area is similar to the first,
briefly centering on G before beginning its octatonic descent to C.
The next thematic area, beginning at m. 30, maintains the C-centricity
and supports it with a strong fifth relationship as both C and G are heard
together, as seen in Example 2-4. The only movement away from C is a
brief half-step motion to Db, similar to the whole-step in the earlier pedal.
The thematic sections, then, are the only areas where a strong pitch
center is established while the transitional areas are unstable. Near the end
of the A section, this pattern deteriorates as the themes are increasingly
fragmented and merge more smoothly with the transitions that follow. The
half-step theme beginning in m. 48, as well as the whole-step theme in m.
57, are examples of this dissolution of pitch centricity.
The final theme in the A section begins with uncertain centricity on
C#. The lack of an accompanying pedal tone, as well as the strong
emphasis on (0257) arranged in stacked fourths, contributes to the lack of
a definitive centric pitch. Eventually, however, the final thematic
statement comes to rest on D, returning to the original pitch class that
opened the piece.
In this way, each thematic area in the A section can be considered an
area of relative rest or stasis that is contrasted by the motion of the
transitional areas. Measures 1-29 represent an augmentation of the whole
48 Chapter Two
step theme: mm. 1-11 are the stasis of motive a, and mm. 12-29 are the
unstable motion of the scalar motive b. In fact, the entire A section has a
similar structure. The strongest pitch centricity is established at the initial
statements of the whole-step and half-step themes in mm. 1-11 and mm.
30-41, respectively. These two areas, with the eventual return to D at the
end of the A section, mirror the initial whole-step motion, D-C-D.
The Ac section follows a similar pattern with the thematic areas holding
stable pitch centers, but the large-scale pattern is not as clear. This lack of
clarity is not entirely unexpected, as melodic and rhythmic embellishments
have been added. Even so, a larger scale pattern can be observed in the
pitch centers. As with A, the initial theme establishes centricity on D, and
contains a notable motion down to C. No return to D is present, however,
leaving any closure with regards to a return to D to the Bc section.
Because the B sections generally lack a specific pitch center, similar
parallelism exists at a larger level between the A and B sections. The
slower A sections containing thematic areas that establish pitch centricity
are at relative rest compared to the constant motion of the B sections.
While the constant motion continues until the end of the piece once the Bc
section starts, it begins to approach a resolution by sustaining a single idea
through the fast rate of surface motion. At this level, the A sections
represent motive a and the B sections represent motive b.
In this way, the single originating idea of an oscillating pitch governs
the shape of the phrases and the larger sections, an example of what Réti
calls the inner form-building force at work.16 The parallelism also suggests
a high level of coherence, aiding the comprehensibility as well. Perhaps
this is what Tower means when she talks about the “inherent musical
motivation of a phrase within the long range direction of music;” the
fundamental musical idea from the opening phrase provides the larger
framework for the whole piece.
Notes
1
Joan Tower, quoted in James Wierzbicki, “Every Instant of Music Has Past,
Present and Future,” St Louis Dispatch, January 4, 1987.
2
Ellen. K. Grolman, Joan Tower: The Comprehensive Bio-Bibliography (Lanham,
MD: Scarecrow Press, 2007), p. 101.
3
Judy Lochhead, “Joan Tower’s Wings and Breakfast Rhythms I and II: Some
Thoughts on Form and Repetition,” Perspectives of New Music 30/1 (Winter,
1992): 136.
4
Ibid.
5
Arnold Schoenberg, The Musical Idea and the Logic, Technique, and Art of Its
Presentation, ed. and trans. Patricia Carpenter and Severine Neff (New York:
Columbia University Press, 1995), p. 21.
6
Ibid.
7
Arnold Schoenberg, Fundamentals of Musical Composition, ed. Gerald Strang,
with the collaboration of Leonard Stein (London: Faber and Faber, 1967), p. 8.
8
Grolman, Joan Tower, p. 36.
9
Schoenberg, The Musical Idea and the Logic, Technique, and Art of Its
Presentation, p. 103.
10
Joan Tower, email correspondence with the author.
11
Schoenberg, The Musical Idea, p. 46.
12
Ibid., pp. 23-24.
13
Ibid., p. 135.
14
Ibid., pp. 132-43.
15
Ibid., p. 24.
16
Rudolph Réti, The Thematic Process in Music (New York: The Macmillan
Company, 1951), pp. 109-14.
Bibliography
Bryden, Kristi. “Musical Conclusions: Exploring Closural Processes in
Five Late Twentieth-Century Chamber Works.” PhD diss., University
of Wisconsin-Madison, 2001.
Fletcher, John M. “Joan Tower’s Fascinating Rhythms For Band: Genesis
and Analysis.” DMA document, University of Oklahoma, 2002.
Gann, Kyle. “Uptown Dropout.” Village Voice, September 22, 1998, p.
132.
Grolman, Ellen K. Joan Tower: The Comprehensive Bio-Bibliography.
Lanham, MD: Scarecrow Press, 2007.
Lochhead, Judy. “Joan Tower’s Wings and Breakfast Rhythms I and II:
Some Thoughts on Form and Repetition.” Perspectives of New Music
30/1 (Winter, 1992): 132-56.
McClelland, Ryan. “Melodic Process and Parallelism in Joan Tower’s
Silver Ladders.” Conference paper, annual meeting of the Society for
Music Theory, Columbus, OH, November 3, 2002.
Naughtin, Matthew. Dawn of a New Era. Program notes. Omaha
Symphony Orchestra Library, 2005.
Omaha Symphony Orchestra. Joan Tower’s Purple Rhapsody. Audio CD.
Joann Falletta, dir., Paul Neubauer, viola. Omaha Symphony Orchestra
Library, November 5, 2005.
ProMusica Chamber Orchestra. Triumvirate. Audio CD. DCD573, Summit
Records, 2011.
50 Chapter Two
Reichardt, Sarah Jane. “An Analysis of Joan Tower’s Island Prelude.”

Master’s thesis, University of Texas at Austin, 1998.
Réti, Rudolph. The Thematic Process in Music. New York: The Macmillan
Company, 1951.
—. Thematic Patterns in Sonatas of Beethoven. Edited by Deryck Cooke.
New York: The Macmillan Company, 1967.
Rylands, Ann. “The Violin Concertos of Ellen Taafe Zwilich (1997) and
Joan Tower (1992): Evolution of an American Style.” PhD diss.,
University of South Carolina, 2002.
Schoenberg, Arnold. Fundamentals of Musical Composition. Edited by
Gerald Strang. With the collaboration of Leonard Stein. London: Faber
and Faber, 1967.
—. Style and Idea: Selected Writings of Arnold Schoenberg. Edited by
Leonard Stein. Translated by Leo Black. New York: St. Martin’s Press,
1975.
—. The Musical Idea and the Logic, Technique, and Art of Its
Presentation. Edited and translated by Patricia Carpenter and Severine
Neff. New York: Columbia University Press, 1995.
Shouha, Laura. “The Musical Language of Joan Tower: An Energy Line
Analysis of Island Prelude for Oboe and Wind Quintet.” PhD diss.,
University of North Texas, 2001.
Singleton, Glen R. “Contrast and Unity in Silver Ladders by Joan Tower.”
Master’s thesis, Baylor University, 1993.
Solie, Ruth A. “The Living Work: Organicism and Musical Analysis.”
19th-Century Music 4/2 (Autumn, 1980): 147-56.
Stoecker, Philip. “An Analysis of Joan Tower’s Sequoia.” Master’s Thesis,
University of Ottawa, 1997.
Tower, Joan. “On Breakfast Rhythms I and II.” DMA diss., Columbia
University, 1978.
—. Purple Rhapsody: Concerto for Viola and Orchestra. Score. G.
Schirmer Rental Library, 2005.
Von Glahn, Denise. Skillful Listeners: American Women Composing
Nature. Bloomington, IN: Indiana University Press, 2013.
CHAPTER THREE
GAMES, SIMPLICITY, AND TREES:

AN ANALYSIS OF ARVO PÄRT’S ARBOS
ERIK HEINE
Arvo Pärt’s music underwent a radical change in the 1970s, after

having tried serial compositions, collage, and extended tonal techniques
earlier in his career. After a lengthy break from composition beginning in
1968, Pärt withdrew into the music of Johann Sebastian Bach, on whose
music much of Credo is based, as well as early plainchant, the music of
the Medieval and Renaissance eras, and the music of the Russian
Orthodox Church. From this immersion in earlier styles, and after nearly
ten years with only one published work, Pärt developed the tintinnabuli
style, which means “in the style of tinkling bells,” and which uses specific
textural compositional techniques. His first works in this new style were
written in 1976 during which time he composed seven pieces. The M-
voice, or melodic line, uses diatonic pitches of a scale or mode while the
T-voice, or accompaniment, uses only members of the tonic triad. Arbos,
an early tintinnabuli work from 1977, displays many of the same
characteristics as other pieces composed from 1976 to 1978, including the
use of the Aeolian mode and voices composed in proportional durations.
Paul Hillier, in his book Arvo Pärt, mentions that Arbos is a mensuration
canon. However, unlike most mensuration canons, the canon in Arbos
actually begins with simultaneous attacks in all canonic voices.
Beyond providing a traditional analysis of Pärt’s Arbos, I also want to
situate Pärt, the composer, within his early serial and avant-garde career,
and his more familiar and popular compositional style, showing how his
early tintinnabuli works are each governed by a different “game.”
Additionally, following Pärt’s own drawing of the piece, two brief
readings will be proposed, one secular, and one sacred.
Arvo Pärt was born in 1935 in Paide, Estonia, which was then a Soviet
Republic. Although Pärt was born in Estonia, he does not draw on folk
music for styles or melodies, unlike Polish composer Henryk Górecki,
52 Chapter Three
another composer with whom Pärt is often grouped.1 While attending the
Tallinn Conservatory, Pärt was exposed to serialism, initially the music of
Anton Webern, Ernst Krenek, and Pierre Boulez. However, the serial and
atonal works of Arnold Schoenberg and Igor Stravinsky were deemed
unacceptable in the Soviet Union. As Lyn Henderson writes,
Although the political ‘thaw’ of the late 1950s and early 1960s enabled
composers of the Soviet block to hear previously banned music, including
12-note works, from the European mainstream, their own serialist
excursions in general ran the gauntlet of continuing official censure. The
Russian composer Andrey Volkonsky had led the way with his Musica
stricta (1956), though it was only three years since he had suffered
expulsion from the Moscow Conservatory for being found in possession of
scores by Schoenberg and Stravinsky.2
As a student at the conservatory, Pärt was very familiar with the music of
Shostakovich, Prokofiev, and Khachaturian, all three of whom were
mentioned in the infamous Zhdanov Decree,3 but began following his own
path composing in a serial fashion, as highlighted in Henderson’s article.
However, the allure of serialism began to wane for Pärt. As Peter Schmelz
states,
By the mid-1960s, [Pärt, Schnittke, Silvestrov, and Gubaidulina] had

grown dissatisfied with that they considered to be the abstractions and
constraints of serialism. They all began turning in their own fashions to
something that felt freer: both to actual freedom in the music, that is,
aleatory elements, and often to various invocations of tonality.4
Pärt’s serial style was becoming less musically interesting to him, and he
began looking for a new path. The premiere of his work Credo in 1968
proved to be the catalyst for change. Although the premiere received a
standing ovation from the audience, it was surprising that it made it past
the censors, due to its overtly religious text and its obvious musical
symbolism.5 Schmelz writes,
In Credo, Pärt was very obviously, perhaps too obviously, laying out a sort
of morality play. The religious text that he added to this ongoing musical
drama was the final straw and, according to Hillier, resulted in a ban on
this and his other works as well.6
The Credo was the piece where Pärt began to move toward a new style.
After Credo, Pärt would compose only one piece until 1976, and that was
his Symphony No. 3 in 1971. Once again, quoting Schmelz,
Games, Simplicity, and Trees: An Analysis of Arvo Pärt’s Arbos 53
Thus Credo marked the first, decisive step in Pärt’s long process of
musical and spiritual redefinition, a process of redefinition irrevocably
connected with an emerging and increasingly refined mimetic sensibility,
and tied to a gradual abandonment of serial structures.7
Henderson adds to the commentary on Credo by stating,
Just as, some seven years before, he had flouted the Soviet authorities in
his forthcoming serial explorations, so now, as a loyal member of the
Russian Orthodox Church, he would do so once again, for he became
determined to set prohibited religious texts…While being the first in his
long series of devotional works, it is also the final example of the 12-note
period and it marks the very summit of a decade of creative experiment. In
a return to the 1964 format [the works Quintettino and Collage on
B-A-C-H], the work is framed within a single key-centre.8
The last statement in Henderson’s quote is the significant one, that the
Credo is framed within a single tonal area. The non-modulatory nature of
Pärt’s early tintinnabuli works can be seen as derived from his early
collage/serial/tonal works. It is the adherence to a single tonic pitch,
without tonicization of any other pitches, that seems to evoke the greatest
reaction from scholars.
Because Pärt’s music has enjoyed a great deal of recognition since his
emigration from the Soviet Union to Germany, musical discussion of his
works have made their way into American textbooks, both theoretical and
musicological.9 In three typically used textbooks for a twentieth-century-
specific music theory course, Pärt’s music is only mentioned in two:
Stefan Kostka’s Music and Materials of the Twentieth Century and Miguel
Roig-Francoli’s Understanding Post-Tonal Music.10 Of these two books,
only Roig-Francoli’s makes any attempt to analyze Pärt’s music, and after
introducing the concept of tintinnabuli, presents an analysis of Cantus in
less than two full pages of prose.11 This is due, in large part, to the fact that
Kostka’s book is in its third edition, originally published in 1990, while
Roig-Francoli’s book is in its first edition, published in 2007. A similar
trend has occurred in music history textbooks. While Bryan Simms’s book
Music of the Twentieth Century only mentions Pärt for two pages, and
does not use or define the term “tintinnabuli,” Richard Taruskin’s Music in
the Late Twentieth-Century, the fifth of a five-volume collection, dedicates
nearly ten pages to Pärt’s music, covering pieces such as the St. John
Passion and Tabula Rasa.12
Despite the amount of newly available access to both the composer and
his music, both Pärt and his wife, Nora, are hesitant to discuss the pieces in
detail. In an interview conducted with both Arvo and Nora Pärt, Geoff
54 Chapter Three
Smith asked about titles of works acting as metaphors for Pärt’s career.
Nora responded by stating,
I would like to add a comment here. I feel we are moving on to dangerous

ground. We’ve been in the West for twenty years now. Since then there has
been a growing number of texts on Arvo’s music but very little of it is
musicologically founded. In effect, almost nothing. This deficit in
musicological methodology is always smoothed over by biographical or
personal information, which cannot necessarily be linked to Arvo’s music
directly. Naturally, you can always connect ideas—biographical or not—
with Arvo’s music. Yet the meaning of the music is purely musical. Arvo
is predominantly concerned with musical forms and structure.13
This quotation, and stance, seems intended to accomplish multiple goals.

First, it asserts that the writings about Pärt contain little musicological, or
theoretical, substance. Second, the quote seems to position Pärt in the role
of a “mystical composer,” one whose music defies analysis, and asks the
listener to become aware of the musical and/or spiritual content. Last, the
quote unequivocally states that the largest compositional concern of Pärt is
with form and structure, placing the composer back into the world of
standard analysis, but paradoxically allowing his music to defy analysis.
Pärt has referred to serialism as a “child’s game,” and that “religion
influences everything. Not just music, but everything.”14 When combined
with Josiah Fisk’s comment regarding the “New Simplicity” that “we have
a music that, for the listener who seeks more than the token artifacts of
classical music, offers no dialogue, no ambiguity, and no inner life. What
we are left with is a surface that speaks of depth, and a depth that speaks of
nothing,” 15 what is left seems invaluable or without substance. If the
musical content is vacuous, if musicology is supplanted by biography, and
if form is the primary compositional concern, then how does one analyze
Pärt’s music? Benjamin Skipp offers thoughts concerning Pärt’s works,
strongly disagreeing with Fisk’s comments, and referring to Pärt’s pieces
as “self-contained, that is, they do not engage in citation of music of the
past.”16 Skipp also discusses Pärt’s “non-dialectical forms” by writing,
While his tintinnabuli works do display moments of contrast in both

rhythmic organization and between certain lines of counterpoint, there is
rarely a sense of formal transformation at a deeper level…In a work such
as Passio, where there are no differentiated formal sections, the tension and
synthesis which is inherent to classical forms cannot function.17
Skipp is stating that traditional forms that employ traditional tonal,

harmonic, and developmental conflict, cannot be used to analyze Pärt’s
music. Instead, the analysis of the music must involve different criteria
beyond canonic procedures.
Before embarking on an analysis of Pärt’s Arbos (1977, rev. 1986),
Pärt’s early tintinnabuli works, composed from 1976-78, should be briefly
mentioned. As mentioned earlier, the word tintinnabuli means “in the style
of tinkling bells.” In this style, two voices exist: the M-voice, or the
melody, and the T-voice, or the tintinnabuli voice. 18 Essentially, no
harmonic progression is present; the home tonality of the work is where it
remains throughout. The M-voice presents members of the scale, most
often moving stepwise. The T-voice plays members of the tonic triad,
usually the closest member of the tonic triad to the melodic note. 19
Example 3-1 comes from Paul Hillier’s book and shows various
possibilities for the composing-out of the M- and T-voices.
Example 3-1, Positions of Melody and Tintinnabuli voices20
In 1976, Pärt composed seven works in this new tintinnabuli style, as

seen in Table 3-1.
56 Chapter Three
Year Title of Composition Tonality/Modality

Calix (withdrawn) —
Modus A Aeolian
Für Alina B Aeolian
Trivium D Aeolian
1976
Pari Intervallo C Aeolian
An den Wassern zu Babel sassen wir
A Aeolian
und weinten
Wenn Bach Bienen gezüchtet hätte Unclear: chromatic
Arbos D Aeolian
Cantate Domino B-flat Major (Ionian)
A Aeolian (with chromatic
Fratres
inflections)
D Aeolian (one mvt. in
Missa Sillabica
F Ionian)
1977 Variationen zur Gesundung von I-III. A Aeolian
Arinuschka IV-VI A Ionian
I. A Aeolian
Tabula Rasa
II. D Aeolian
Cantus in memoriam Benjamin
A Aeolian
Britten
Summa G Aeolian
1978 Spiegel im Spiegel F Major (Ionian)
Table 3-1, List of compositions by Arvo Pärt, 1976-1978
Of those seven works, the first two, Calix and Modus, were written before
he had fully achieved the tintinnabuli style, and Pärt ultimately rejected
Calix from his canon.21 From looking at the modality column in Table 3-1,
it is clear that Pärt most commonly uses the Aeolian mode, and seems to
favor D and A as tonic pitches. Hillier describes the sense of harmonic
stasis as part of a ritual. He writes,
The use of repetitive patterns and harmonic stasis suggests an awareness of

time quite different from the materiality and Western ‘clock’ time, though
just as real to the person who experiences it…With Pärt, we are taken one
step further, though we are obliged to accept, or at least trust, the

framework of Christianity within which he works.22
Pärt is also invoking a sense of spirituality, of Christianity. David Clarke

refers to Pärt’s music and states, “[I]ts religious aura clearly has a strong
allure for a secularized culture that no longer has any collective way of
articulating the spiritual.”23
One of the earliest pieces in this new style is 1976’s Für Alina. This
short piano piece is in B Aeolian and typifies Pärt’s tintinnabuli style of
soft dynamics, slow rates of change and stepwise motion or leaps within
the tonic triad. The manuscript of this work is reproduced on pages 88-89
of Hillier’s book. The top staff is the M-voice while the middle staff is the
T-voice.
Each work or movement of Pärt’s early tintinnabuli music is governed
by a different principle; each piece is a game, like the title of Paul
Hindemith’s collection of fugues and interludes, Ludus Tonalis, “The
Game of Tones.” I do not believe that it is by chance that the first
movement of Tabula Rasa is titled “Ludus.” Other famous works from this
period include Summa, Fratres, Tabula Rasa, and Spiegel im Spiegel. In
Summa, which is in G Aeolian, Pärt begins with two melodic lines, one in
the second violin, and one in the cello. The objective is to rotate through
the melodic line, while starting with the same first note each time, and
return to the original melody. Example 3-2 shows the ordering of pitches
in both parts, as well as a comparison of the original two melodic lines to
each other.
The two original statements are related almost entirely by inversion.
The scale degrees in bold highlight the moments where the inversion is not
present, and those scale degrees are either 1, 3, or 5.
Spiegel im Spiegel works in a similar fashion. The translation of the
title is “Mirror in the Mirror,” making it clear that inversion will play a
significant role in this piece as well. The point of rest, and goal, for the
melodic instrument is scale-degree 3.24 Example 3-3 shows the statements
of the melodic instrument in terms of scale degrees and solfège syllables;
the (+) symbol indicates an ascending gesture while the (-) symbol
indicates a descending gesture.
Spiegel im Spiegel is composed in F Ionian, one of Pärt’s few major
mode works. When the soloist lands on scale-degrees 1 or 3, the piano
arpeggiates an F major triad. When the soloist is on scale-degree 5, an A
minor triad is sounded, and when the soloist is playing scale-degree 7, a C
major triad is arpeggiated. When the soloist sounds scale-degrees 2, 4, or
6, the arpeggiation in the piano involves some diatonic dissonance, as a
triad is not clearly stated.
58 Chapter Three
Example 3-2, Ordering of pitches (in scale degrees) in Summa. Scale degree 8
indicates “descending;” scale degree 1 indicates “ascending”.
23-43
123-543
2 8 7 (+)3 - 4 5 6 (-)3
67123-76543
2 8 7 6 5 (+)3 - 4 5 6 7 1 (-)3
4567123-2876543
2 8 7 6 5 4 3 (+)3 - 4 5 6 7 1 2 3 (-)3
234567123-432876543
Example 3-3, Melodic statements (in scale degrees) in Spiegel im Spiegel
Fratres, one of Pärt’s best-known works, and the piece that is the most
widely arranged, is in A Aeolian, but contains chromatic inflections, with
the major third scale degree as well as the minor second scale degree.
Once again, this piece is based on a pattern, which is shown in Example 3-
4. The objective of the pattern is to move down a third each time, and the
goal is a pitch over two octaves lower than the beginning. Like Cantus and
Arbos, Fratres utilizes an additive process.
Example 3-4, Reduction of initial statement in Fratres
Tabula Rasa, or “clean slate,” has been referred to as the last piece in
Pärt’s new style. It is composed in two movements. The first movement,
“Ludus,” or “Games,” uses a similar additive process to the ones already
discussed and is in A Aeolian. Essentially the objective is to begin with
four notes, all As, and expand the motive out to ascend an octave, return to
the original A, descend an octave, and ascend to the original A. Example
3-5a shows the initial and second statements and Example 3-5b shows the
final statement. Here is the opening of the first movement of Tabula Rasa.
Example 3-5a, Reduction of first two statements in Tabula Rasa, (I)
Example 3-5b, Reduction of Final Statement in Tabula Rasa, (I)
The second movement, titled “Silentium,” or “silence,” is in D

Aeolian, an unusual move for Pärt because traditionally, even multi-
movement works remain in the same key. Once again, the additive process
is used as the initial statement presents both upper and lower neighbor
pitches to the tonic of D, and the goal is to ascend two octaves, return,
descend two octaves, and return. However, the final statement of this
60 Chapter Three
process fails to finish. The opening and final statements are shown in
Examples 3-6a and 3-6b.
Example 3-6a, Reduction of first two statements in Tabula Rasa, (II)
Example 3-6b, Reduction of final completed statement in Tabula Rasa, (II)
This movement contains the most similarities to Arbos as any of Pärt’s

pieces in this period of composition. Both this movement and Arbos are in
D Aeolian. Both pieces involve mensuration canons. Both pieces begin
with simultaneous beginnings in all three proportional voices. However,
the second movement of Tabula Rasa is soft, as indicated by the title
“Silentium.” The second movement of Tabula Rasa has no percussion, nor
are ends of statements articulated in any way. Most importantly, the
process for the mensuration canon is incomplete. In contrast, Arbos is loud
throughout the entire piece, is written for brass and percussion, ends of
statements are articulated, and the compositional process is complete at the
end of the work. While the second movement of Tabula Rasa is certainly
in the style of tinkling bells, the whole of Arbos is not.
Of the works listed in Table 3-1, the most familiar are likely, in no
particular order, Für Alina, Fratres, Cantus, Tabula Rasa, and Spiegel im
Spiegel. Arbos is not a member of this “most familiar” group of works.
The five works previously mentioned are familiar, largely because they
appear in texts and writings concerning Pärt, or are used in films,
particularly Cantus and Spiegel im Spiegel. The fact that Arbos is not as
well known as other early tintinnabuli works does not make it a lesser
piece; the way the piece is scored, and its nature as a mensuration canon,
when Cantus is also a mensuration canon, can explain why it does not
appear often in texts. Additionally, Hiller offers a brief analysis of the
work, but it is less than one page of prose.25 He also likens Cantus to
Arbos, indicating that even though Arbos was Pärt’s first mensuration
canon, it is not his most familiar work in that genre.26
Arbos is a piece originally composed in 1977 for seven or eight
recorders and three triangles in D Aeolian. The piece was revised in 1986
for a brass chamber ensemble and three percussionists, and this is the
version of the piece that I will be discussing. The piece is scored for four
trumpets, four trombones, and three percussionists. However, the
groupings of the instruments are atypical of score order. The first group
places Trumpets 1, 2, and 4 together, the second groups places Trumpet 3
with Trombones 1 and 2, and the third group contains Trombones 3 and 4.
The percussion, located beneath the brass, is separated into Tubular Bells,
Tam-tam, and Timpani. The piece is written in a meter of 12/4; the first
group is subdivided 3+3+3+3, the second group is 4+2+4+2, and the third
group is 8+4. The percussion, like the first group, is written in 3+3+3+3.
The indication “bell-like with anticipation” is printed above all brass parts,
and those parts are all marked at a dynamic of sempre forte. The
percussion do not have the same marking as the brass, and their dynamics
are all sempre mezzo-forte. Based on this scoring, Pärt is showing the
groups present in the mensuration canon. Even the subdivisions above the
groupings are different, indicating how the mensuration canon should
proceed. In stark contrast to the sound of Für Alina, Arbos is on the
opposite end of the spectrum: it is loud and completely contrasts with the
fragile sounds of Für Alina.
A mensuration canon is a canon where voices move at proportional
rates of speed. One of the earliest and most famous uses of the
mensuration canon is in Johannes Ockeghem’s Missa prolationum,
composed in the second half of the fifteenth century. Ockeghem’s Mass is
significant enough that Allan Atlas refers to it as a “technical tour de
force.”27 A typical mensuration canon features the beginning of the initial
statement in the shortest durations, then the next shortest durations, and so
on. In a sense, this is how Arbos operates. However, instead of beginning a
statement and having a dux and a comes, or dux and multiple comes
voices, all voices enter simultaneously with the first note of the piece.
As in many of his works, the highest M-voice, or M-voice 1 in Arbos,
moves primarily stepwise in a scalar fashion; its objective is to move from
scale-degree 5 down the interval of a perfect twelfth and end on scale-
degree 1. Example 3-7 shows a reduction of M- and T-voices 1.
62 Chapter Three
Example 3-7, Reduction of M-voice 1 and T-voice 1
T-voice 1 is in the Inferior Second Position. Of interest is where Trumpet

2 changes from T-voice 1 to M-voice 1. At this same point, Trumpet 4
continues T-voice 1 and Trumpet 1 rests. A second trait typical of Pärt’s
music is the use of an additive compositional process. The M-voice
initially states the opening pitch twice, followed by the opening pitch
twice and the third note, followed by the first four notes, and the pattern
continues. Example 3-8 shows the beginning of this additive process, and
by the sixteenth iteration, the process is complete. However, the other
voices are not yet complete, so M- and T-voices 1 play the sixteenth
iteration of the additive process ten more times so that the other
proportional parts can complete their respective patterns.
Example 3-8, Reduction of Additive Process in Arbos
Each note in the pattern alternates between long and short, half notes and
quarter notes. However, the beginning of each iteration is inconsistent; the
first note alternates between half notes and quarter notes. Because of the
pattern, after the initial statement, the following two statements begin with
the same duration; statements 2 and 3 begin with quarter notes, statements
4 and 5 begin with half notes, etc. Hillier states that,
The rhythm is the same throughout, but every second phrase reverses the
pattern from trochaic to iambic, so that the effect is of an alternating
rhythmic current passing through the ever-lengthening melodic contours.28
However, once the end of the process is reached, the opening notes for the
seventeenth through twenty-sixth statements alternate between beginning
with a half note and beginning with a quarter note.
M-voice 2 is located in the Trombone 1 part. The difference with this
second group is that M-voice 2 is surrounded by T-voices. T-voice 2a is
located in the Superior Second Position, while T-voice 2b is located in the
Inferior Second Position. This second grouping contains durations twice as

long as the first grouping. Example 3-9 provides a reduction of the M- and
T-voices 2.
Example 3-9, Reduction of M-voice and T-voice 2
Once again, the alternation of notes between long and short is present, as is
the alternation between trochaic and iambic. One final similarity between
M-voice 1 and M-voice 2 is that the objective in both parts is to move
down the interval of a perfect twelfth, and it takes sixteen iterations for the
process to be complete. Only one additional complete statement is sounded
for M-voice 2, for a total sounding of seventeen iterations, as opposed to
twenty-six for M-voice 1.
M-voice 3 is in the Trombone 4 part, while Trombone 3 plays T-voice
3 in the Superior Second Position. Once again, the durations of notes are
twice as long as the group immediately above, so the third group is in a 4:1
proportion to the first group, and a 2:1 proportion to the second group.
Because of the durations of the notes, it is not possible for M-voice 3 to
cover the same intervallic span as M-voices 1 and 2. Although M-voice 3
begins with an A, like M-voices 1 and 2, and it does cover the distance of
a perfect twelfth, M-voice 3 omits the final two pitches from M-voices 1
and 2. Only 11 iterations of M- and T-voices 3 occur. Example 3-10 shows
the reduction of M- and T-voices 3.
Example 3-10, Reduction of M-voice and T-voice 3
The percussion serves to articulate the conclusion of each iteration of

the three M- and T-voices. The tubular bells play the notes F and D, both
members of the tonic triad. The pitch F articulates the end of M- and T-
voice 1, and occurs the quarter note after the last note in the additive
process. Therefore, in a measure of 12/4, the pitch F can occur in eight
different places: beats 1, 3, 4, 6, 7, 9, 10, and 12. The pitch D articulates
64 Chapter Three
the end of M-and T-voice 2, and occurs the half note after the last note in
the additive process. In a measure of 12/4, the pitch D only falls on beats
1, 5, 7, and 11. The timpani, which rolls on the pitch D, and sounds an
accented D an octave lower, and tam-tam together articulate the end of M-
and T-voice 3, and they occur the whole note after the last note in the
additive process. The timpani and tam-tam only articulate on beats 1 or 9
in the 12/4 measure. The timpani states a final D to signal the end of the
piece.
The title of the piece comes from the Latin word “arbor” meaning tree.
In this particular instance, Pärt is not referring to a grove of trees, but to a
family tree. In her article titled “Toward a Theory of Minimalist Tropes,”
Rebecca Leydon states that, “In fact Pärt claims to be representing familial
ancestors to the generations propagating at the foreground.”29 In the liner
notes to the ECM recording, Pärt himself states,
This score seeks to create the image of a tree or family tree. The deeper the
line, the more slowly it moves, and vice versa. This results in three layers
of movement, a mensuration canon in which each of the three layers
presents the theme in a different tempo.30
Pärt’s reference to the family tree can easily be heard. Hillier states that
“The title of Arbos signifies the shape of a tree and the different tempos of
life cycle it contains (branches, trunk, roots).” 31 In 1995, Pärt made a
drawing of his own interpretation of his early tintinnabuli works: Tabula
Rasa, Cantus, Arbos, An den Wassern…, and Sarah… Pärt’s image of
Arbos appears below.
Figure 3-1, Pärt’s drawing of Arbos, reprinted by permission.

Arvo Pärt "Arbos" © 1981 by Universal Edition A.G., Wien/UE 17443
The younger members of the family, likely children, are much more active
and energetic, and are represented in Voice 1; they can move through the
canonic process much more quickly and still have energy to continue
restating the final iteration even after the process is complete. The parents
are represented in Voice 2 as the tree trunk. They also complete the
canonic process, but in a more slow and deliberate way, so much so that
only one extra iteration is needed. Voice 3 is the equivalent of
grandparents, the roots of the family, and although their process does not
quite complete itself, it does not need to do so because of the depth, both
in terms of pitch and in terms of duration, of Voice 3.
Within Pärt’s compositional output from 1976 to 1978, the form and
construction of Arbos is unique. Only one other complete work of the
fifteen that Pärt composed in this three-year range is a mensuration canon,
and that is Cantus, like Arbos, written in 1977. Hillier acknowledges the
differences between the pieces when he states, “[T]he differences, and not
only those of texture and tempo, are significant.”32 The most significant
difference between Arbos and Cantus is that the Cantus is strictly a canon,
meaning that a clear dux and comes are present. Additionally, the Cantus
also uses a single tubular bell, the tonic pitch of A. However, the tubular
bell in Cantus is part of a regularly repeating pattern that does not
correspond to endings of iterations as in Arbos. Miguel Roig-Francoli
provides an analysis of Cantus in his book Understanding Post-Tonal
Music and writes, “This is also a piece in which, to paraphrase [Steve]
Reich, the compositional process and the finished, sounding product are
identical.” 33 While the published score for Cantus displays a dynamic
growth from ppp to fff, the manuscript shows a different dynamic arc.34
Each entering voice is one dynamic louder than the previous entry. First
violins have a dynamic of ppp, second violins have pp, violas have p,
cellos enter with mp, and basses have mf. The overall dynamic grows to f,
but begins to diminuendo at rehearsal 16, and the ensemble ends at a
dynamic of ppp.
As Arvo Pärt’s music continues to grow in its popularity, more texts
will undoubtedly be written. In fact, The Cambridge Companion to Arvo
Pärt, edited by Andrew Shenton, was only recently published in May
2012. Based on the table of contents, only one of the ten chapters is strictly
devoted to analysis of Pärt’s music. Shenton continues to try to unlock
Pärt’s musical mystery. He and Michael Scott Cuthbert have used a
program called Music21 to try to discover the “nucleus,” as Pärt refers to
it, of the music.35 Skipp states that,
Theological considerations of Pärt’s works as religious objects or practices

therefore seem disingenuous, because of their popularity and appropriation
as film and television soundtracks…At the heart of what can be viewed as
the Enlightenment musical meta-narrative is the idea of a continually
66 Chapter Three
developing relationship, or tension, between form and content…It is this

meta-narrative, wrapped up with the issues of canon formation that
originated in the 18th century and particularly in the figure of Bach, which
played the formative role in Pärt’s primary grasp of musical creation.36
Using Nora Pärt’s comment that Arvo is mainly concerned with form, the
tension in Arbos can be seen to be the tension between familial
relationships, between the specific ways in a which a tree can grow,
bearing branches and leaves, and waiting for the contrast in dynamic, but
never receiving it. The lack of dynamic contrast is one that often occurs in
Pärt’s music. In the case of Arbos, it lacks an overtly religious nature and
text, so superimposing that “meta-narrative” onto the piece might not
work. However, three different lines are present, and could be seen to
represent the Holy Trinity of Father (tree roots), Son (trunk), and Holy
Spirit (branches). Continuing analysis and musicological discovery remain
necessary and relevant to comment on the music of this immensely
popular living composer.
Arvo Pärt’s 1977 work Arbos, later revised in 1986, stands out in his
compositional output of early tintinnabuli works for multiple reasons.
First, the instrumentation is unlike anything Pärt used elsewhere,
exclusively brass and percussion; generally, Pärt uses piano, voices, or
string ensembles. Second, the dynamic throughout the entire piece is forte,
with a crescendo at the conclusion. The vast majority of Pärt’s works are
at a piano dynamic. Third, Arbos is a mensuration canon that has all three
voices presenting simultaneous attacks at the beginning of the piece, and
the line moves in a largely descending fashion. Arbos is representative of a
family tree, with the greatest durational proportion in the lowest voice.
While the second movement of Tabula Rasa and Cantus come closest to
Arbos in terms of their composition, Arbos still prominently stands out in
Pärt’s output as a work that is different from the rest of his early
tintinnabuli pieces. Hillier states,
When we create music, we express life…Pärt uses the simplest of means—

a single note, a triad, words—and with them creates an intense, vibrant
music that stands apart from the world, and beckons us to an inner
quietness and inner exultation.37
Arbos expresses life through the image of a family tree, and despite its
loud dynamic, asks us to reflect on our own heritage.
Notes
1
Paul Hillier, Arvo Pärt (New York: Oxford University Press, 1997), p. 24.
2
Lyn Henderson, “A Solitary Genius: The Establishment of Pärt’s Technique
(1958-68),” The Musical Times, 149/1904 (Autumn 2008): 82-83.
3
Peter Schmelz, Such Freedom, If Only Musical: Unofficial Soviet Music During
the Thaw (New York: Oxford University Press, 2009), pp. 26-27.
4
Schmelz, p. 220.
5
Hillier provides a large section on this piece in his book on pp. 58-63.
6
Schmelz, p. 232.
7
Ibid., pp. 232-33.
8
Henderson, pp. 86-87.
9
I am not trying to provide a complete survey. I am trying to show how Pärt’s
music is discussed (or not discussed) in commonly used era-specific books.
10
Joseph Straus’s Introduction to Post-Tonal Theory does not mention or analyze
Pärt’s music. Straus’s book is currently in its third edition, originally published in
1990.
11
Miguel Roig-Francoli, Understanding Post-Tonal Music (Boston: McGraw-Hill,
2007), pp. 334-37. The specific analysis occurs on pages 336-37.
12
Richard Taruskin, Music in the Late Twentieth Century (New York: Oxford
University Press, 2010). Taruskin discusses Pärt’s music on pages 400-10.
13
Geoff Smith, “An Interview With Arvo Pärt: Sources of Invention,” The Musical
Times 148/1868 (August, 1999): 21.
14
Jamie McCarthy, “An Interview With Arvo Pärt,” The Musical Times 130/1753
(March, 1989): 132.
15
Josiah Fisk, “The New Simplicity: The Music of Górecki, Tavener and Pärt,”
The Hudson Review 47/3 (Autumn 1994): 411.
16
Benjamin Skipp, “Out of Place in the 20th Century: Thoughts on Arvo Pärt’s
Tintinnabuli Style,” Tempo 63/249 (2009): 7.
17
Ibid.
18
Hillier spends an entire chapter of his book explaining the style on pages 86-97.
19
Even though Roig-Francoli discusses the M- and T-voices in his textbook, he
does not provide musical examples. Instead, he provides a reference to Hillier’s
book.
20
Hillier, p. 94.
21
Hillier lists Calix as the first work composed in 1976 on p. 98, but in the “List of
Works by Arvo Pärt” on pp. 208-10, it is not listed. This list comes with an asterisk
that states, “This list only contains those works currently recognized by the
composer as part of his canon.”
22
Hillier, p. 17.
23
David Clarke, “David Clarke Reappraises the Music and Aesthetics of Arvo
Pärt,” The Musical Times 134 (1993): 680.
24
I am using the term “melodic instrument” since multiple arrangements of this
piece exist, most notably for violin and piano and cello and piano.
25
Hillier, p. 101.
26
Ibid., p. 102.
68 Chapter Three
27
Allan Atlas, Renaissance Music (New York: W. W. Norton & Co., 1998), p.
154.
28
Hillier, p. 101.
29
Rebecca Leydon, “Towards a Typology of Minimalist Tropes,” Music Theory
Online 8/4 (December 2002).
30
Arvo Pärt, liner notes to Arbos, ECM New Series 1325, ECM 422831 959-2,
1987.
31
Hillier, p. 101.
32
Hillier, p. 102.
33
Miguel Roig-Francoli, Understanding Post-Tonal Music (New York: McGraw-
Hill, 2008), p. 336. His analysis appears on pp. 336-37, preceded by a brief
discussion of Pärt’s compositional technique on pp. 334-35.
34
The manuscript is reproduced in the Deluxe Edition of the album Tabula Rasa,
released by ECM in 2010. The reproduction appears on pp. 95-104.
35
Shenton presented preliminary information and analysis at the Forum on Music
and Christian Scholarship, 16-18 February, 2012, held at Calvin College. The
computer program can be accessed at http://mit.edu/music21. Pärt refers to the
“nucleus” on pages 19-20 of Smith’s interview.
36
Skipp, p. 11.
37
Hillier, p. 1.
Bibliography
Atlas, Allan. Renaissance Music. New York: W. W. Norton & Co., 1998.
Clarke, David. “David Clarke Reappraises the Music and Aesthetics of
Arvo Pärt.” The Musical Times, 134 (1993): 680-84.
Fisk, Josiah. “The New Simplicity: The Music of Górecki, Tavener, and
Pärt.” The Hudson Review 47/3 (1994): 394-412.
Griffiths, Paul. Modern Music and After, 3d ed. New York: Oxford
University Press, 2010.
Henderson, Lyn. “A Solitary Genius: The Establishment of Pärt’s
Technique (1958-68).” The Musical Times 149/1904 (2008): 81-88.
Hillier, Paul. Arvo Pärt. New York: Oxford University Press, 1997.
Kostka, Stefan. Materials and Techniques of Twentieth-Century Music, 3d
ed. Upper Saddle River, New Jersey: Prentice Hall, 2006.
Leydon, Rebecca. “Towards a Typology of Minimalist Tropes.” Music
Theory Online 8/4 (December, 2002),
<http://www.societymusictheory.org/mto/issues/mto.02.8.4.leydon.html>
McCarthy, Jamie. “An Interview with Arvo Pärt.” The Musical Times
130/1753 (1989): 130-33.
Moody, Ivan. “Górecki: The Path to the ‘Miserere.’” The Musical Times
133/1792 (1992): 283-84.
Pärt, Arvo. Arbos. With the Hilliard Ensemble. ¤1987 by ECM New
Music Series, ECM 422831 959-2. Compact Disc.
Roig-Francoli, Miguel. Understanding Post-Tonal Music. New York:
McGraw-Hill, 2008.
Schmeltz, Peter. Such Freedom, If Only Musical: Unofficial Soviet Music
During the Thaw. New York: Oxford University Press, 2009.
Simms, Bryan. Music of the Twentieth Century, 2d ed. New York:
Schirmer Books, 1996.
Skipp, Benjamin. “Out of Place in the 20th Century: Thoughts on Arvo
Pärt’s Tintinnabuli Style.” Tempo 63/249 (2009): 2-11.
Smith, Geoff. “An Interview with Arvo Pärt: Sources of Invention.” The
Musical Times 140/1868 (1999): 19-25.
Straus, Joseph. Introduction to Post-Tonal Theory, 3d ed. Upper Saddle
River, New Jersey: Prentice Hall, 2004.
Taruskin, Richard. Music in the Late Twentieth Century. New York:
Oxford University Press, 2010.
CHAPTER FOUR
AN INTRODUCTION TO THE MUSIC

OF WILLEM CEULEERS
TIM S. PACK1
Over the last five or six years, I have had the distinct pleasure of
corresponding with numerous renowned composers from all over the
world; all of them have generously shared with me their time and music,
and most have even been so kind as to discuss their work directly with my
students. Among these outstanding musicians, one of the most exceptional,
most prolific, and yet most underrated composers is Willem Ceuleers. The
skill and breadth of styles displayed across his huge output as well as his
contribution to music of our time are unquestionably deserving of detailed
study and appreciation. This paper endeavors to initiate such a task by
providing biographical information as well as a survey of the four
following works, which demonstrate the composer’s skill and stylistic
depth:
1) Stabat sancta Maria, op. 655 (2005)

2) Requiem, op. 735 (2010)
3) Cantata voor zondag Trinitatis, op. 610 (2005)
4) Orgelmis voor het Heilig Hart van Jezus, op. 688 (2007)
Willem Ceuleers was born on June 21, 1962 in Watermaal-Bosvoorde,

Belgium. When he was eight years old, he began his formal training in
music at the Muziekacademie van Deurne, which is just outside Antwerp.
From 1978 to 1980, he continued his musical training at the Rijksinstitute
in Antwerp. It was here in 1978 that he completed his first work, the
chorale prelude Vater unser im Himmelreich, op. 1. At age eighteen, he
entered the Koninklijk Muziekconservatorium in Antwerp, where for five
years he studied organ, recorder, harmony, and counterpoint. During this
same period, he also studied voice and carillon at the Stedelijk
Conservatorium in Mechelen. From 1985 to 1988, Ceuleers furthered his
An Introduction to the Music of Willem Ceuleers 71
studies in organ, voice, and harmony at the Koninklijk Muziekconservatorium

in Brussels, and from 1988 to 1992, he returned to the Koninklijk
Muziekconservatorium in Antwerp to continue studying voice and
harpsichord. In the mid 1990s, he studied cello at the Muziekschool Ward
de Beer in Antwerp.
Since the late 1970s, Ceuleers has been a bass singer in several
prominent ensembles while holding concurrent positions as organist at
various churches and cathedrals in and around Antwerp. In 1976, when he
was only thirteen, he was a bass singer in the ensemble Pratum Musicum,
led by Marcel Onsia. From 1981 to 1987, he sang in Currende, directed by
Paul van Nevel. Since 1990, Ceuleers has been a bass singer in the
Huelgas Ensemble, and in 2004, he was commissioned to write the thirty-
five-voice Nomen mortis infame to commemorate the ensemble’s thirty-
fifth anniversary. Nomen mortis is the work for which Ceuleers is perhaps
most widely known, especially outside of Belgium. In addition to his
involvement with the Huelgas Ensemble, Ceuleers has been a bass singer
in the Josquin Capella since 1998. He has also sung with renowned
ensembles such as the Capilla Flamenca (Belgium), Cappella Pratensis
(Netherlands), Egidius Kwartet (Netherlands), and Grande Chapelle (Spain).
As organist, Ceuleers served at various churches in and around
Antwerp between 1978 and 1992. From 1992 to 2006, he was titular organist
at Sint-Catharinakerk in Sinaai, and from 2002 to 2004, he concurrently
served as the choirmaster at Sint-Goedelekathedraal in Brussels. 2 Since
2006, Ceuleers has served as the regional organist over three churches in
Laken; in addition to his duties at Sint Clara and Onze-Lieve-Vrouwekerk,
he is the resident organist at Sint Lambertuskerk.
In 2001, Ceuleers and theologian Dr. Dick Wursten founded the
Collegium Musicum in Antwerp. The group includes a choir as well as a
Renaissance orchestra and Baroque orchestra comprising sixty to seventy
members, who play and sing as necessitated by each piece. The Collegium
has a vast repertory ranging from the music of Du Fay to works by
Ceuleers himself. The various church jobs of the members offer excellent
venues for the group. Nearly all of Ceuleers’ compositions are performed
either by the Collegium or in recitals, though he is receiving more and
more commissions from other ensembles and organizations.
Ceuleers is a highly prolific composer whose output thus far includes
780 compositions comprising a broad array of styles and genres (see Table
4-1). Stylistic influences range from Pérotin (fl. c. 1200) to Charles
Tournemire (1870-1939) and Marcel Dupré (1886-1971).
72 Chapter Four
Quantity Genre
289 chorale preludes
215 motets
55 works for instrumental ensemble
36 organ verses
35 free organ works
33 cantatas
26 masses
16 polyphonic Lieder
15 Lieder
13 chorale variations
12 organ symphonies
10 concertos
10 works for cello
6 chorale fantasies
6 song variations
2 works for harpsichord
1 opera
1 symphony
Table 4-1, Ceuleers’ output arranged by genre
Although his compositions incorporate a diverse range of styles, Ceuleers

rarely combines different styles within the same piece. When asked about
his use of various styles of previous great composers, he replied “I often
put myself in the place of a certain master’s student and develop from this
point my own musical language.”3 In previous discussions on this matter,
he commented, “in this respect, my compositions are certainly modern,
because they evoke my personal choices with a living audience in mind.”4
In our most recent discussion, I asked him which styles he prefers most.
He responded first by pointing out that he does not dislike any style, but
that he does have one set of stylistic preferences for vocal music and
another for instrumental music; each is listed in Table 4-2.5
Vocal music
1. Flemish polyphony (especially the styles of Josquin and Gombert)
2. Baroque anthems (especially the styles of Monteverdi and Schütz)
3. cantatas (especially the styles of Buxtehude and Bach)
Instrumental music
1. organ and harpsichord music (especially the styles of Byrd and Sweelinck)
2. organ music
chorale variations (influenced especially by Scheidemann and Weckmann,
who himself was influenced was the Italian style of Gabrieli and Monteverdi)
3. symphonic organ works of Franck, Vierne, Tournemire, and Dupré
Table 4-2, Ceuleers’ preferred styles for vocal and instrumental music
In this conversation, Ceuleers reiterated that he hardly ever fuses styles; he

stressed that he uses the instruments and styles for what they were
originally intended.
His works are not mere reproductions of works written in earlier styles;
rather, he compares his approach to that of a builder today using old
materials—for example, stones, mortar, and timber from a Romanesque
structure—to construct a new building. Such a builder (à la Ceuleers)
would not use these materials in a way contrary to the way they were
originally used: the timber would never be used for the foundation of the
new building, nor would the mortar ever be used for the roof. If the builder
wanted to evoke the Romanesque style most clearly, s/he would construct
shapes, such as semi-circular arches and barrel vaults, which characterize
the style. The end result would give the initial impression that the structure
is unquestionably new but could have been built during the Romanesque
period; closer inspection of the structure would reveal subtle features
distinguishing it from the earlier period. Despite incorporating materials,
designs, and shapes characteristic of Romanesque architecture, the new
structure would be original because of having been constructed by a
contemporary builder who made his/her own decisions throughout the
creative process.
This analogy is of fundamental importance for a good understanding of
Ceuleers’ compositional approach. For him, each style is a different means
of giving coherence and unity to his music. Once he decides which style to
incorporate for a specific piece, he works within the scope of that style to
74 Chapter Four
create in his own musical language a new piece for living audiences to
appreciate.
Regardless of the style he draws upon, an important feature of his
overall compositional approach is his predilection for working with one
main theme, which may be either borrowed or newly composed. He may
treat the theme or subject as a cantus firmus, or he may use it as a model
for creating variations or altogether new music; however, he tends to avoid
working with multiple themes, unless they are related. This explains why,
as illustrated in Table 4-1, he has composed numerous chorale preludes,
motets, organ works, cantatas, and masses, yet has written only one
symphony and only one opera.
Having provided information about Ceuleers and his compositional
approach, I would now like to direct our attention to a few works that
demonstrate his skill and stylistic breadth. The earliest style Ceuleers
incorporates is that of Pérotin, and the work that best illustrates his
adaptation of this style is Stabat sancta Maria, op. 655 (2005). Though a
very serious work with a solemn text about the Virgin Mary full of
sorrows standing by the cross of Jesus, the piece was written as a musical
joke for a Lenten concert performed by the Collegium. Composers on the
program ranged from Du Fay to Josquin, so the audience was quite
astonished when they heard this piece after a full concert of Renaissance
music for Lent. Moreover, both the score and the program listed the piece
as anonymous; until after the concert, the audience and even the singers
assumed the piece had been written by an anonymous Parisian composer
from the early thirteenth century.
Stabat sancta Maria perhaps most resembles Pérotin’s Viderunt omnes,
though Ceuleers did not have the Notre Dame composer’s work in front of
him when writing his own piece. Both works are examples of organum
quadruplum with clausulae. Examples 4-1 and 4-2 respectively show the
four-voice organum in mm. 1-16 and clausula in mm. 25-32 of Stabat
sancta Maria. For the sake of comparison, Examples 4-3 and 4-4 illustrate
the organum in mm. 1-5 and clausula in mm. 157-62 of Pérotin’s Viderunt
omnes.
Example 4-1, Ceuleers, Stabat sancta Maria, op.655 (2005), mm. 1-16
76 Chapter Four
Example 4-2, Ceuleers, Stabat sancta Maria, op. 655 (2005), mm. 25-32
Example 4-3, Pérotin, Viderunt omnes, mm. 1-5
Example 4-4, Pérotin, Viderunt omnes, mm. 157-62

With respect to thematic material, rhythm, counterpoint, and

dissonance treatment, Ceuleers’ piece is quite comparable to that of
Pérotin. There are, nonetheless, subtle differences concerning timbre,
length, liturgical function, and formal design. The lowest voice in
Pérotin’s organa quadrupla is a tenor; in Viderunt omnes, the tenor’s range
is F3-F4, and in Sederunt principes, it is D3-F4. Ceuleers, on the other
hand, prefers a darker color for the cantus firmus voice, so he situates it in
a baritone range (A2-Bb3). Each of the works by Pérotin is much longer
than the one by Ceuleers; this difference most likely stems from the
intended purpose of the pieces. Pérotin’s two organa quadrupla are
graduals derived from the Psalms. Viderunt omnes, which comes from
Psalm 97:3-4, is for Christmas Day and therefore concerns rejoicing over
God’s salvation. Sederunt principes, which derives from Psalm 118:23 and
86, is for the Feast of St. Stephen;6 although this text is also sung during
Christmastide (December 26), it is a plea for salvation from unjust
persecution. Unlike Pérotin, Ceuleers chose a Tract for Lent; this non-
biblical text portrays the grief of the Virgin Mary as she stands by the
cross of Jesus.
Table 4-3 compares the formal design of all three works. In both of his
works, the Notre Dame composer proceeds from organum directly to
plainchant; he then alternates between organum and clausula sections and
concludes with the plainchant. Ceuleers’ setting begins with organum but
proceeds directly to a clausula on the last syllable of the first word; rather
than alternating between organum and clausula passages, Ceuleers
subsequently returns to chant and repeats the procedure over again. His
design is therefore organum—clausula—chant, organum—clausula—
chant. As shown in Table 4-3, Pérotin uses the pattern organum—chant
before alternating between organum and clausula sections and ending with
chant.
In Pérotin’s time, most of the music throughout the church year was
monophonic; even when polyphony was used, two- or three-voice texture
was more common than four-voice texture. Viderunt omnes and Sederunt
principes were reserved for very special times of the year (e.g., Christmas
or perhaps Easter) and were considered exceptionally grand. These works
would have formed a very distinct part of the service.
Ceuleers wrote his piece to achieve a different kind of striking effect.
He was not endeavoring to create a shock by introducing polyphony to an
audience primarily accustomed to hearing monophonic plainchant, nor was
he aiming to astound his listeners with a grandiose work of epic
proportions for High Mass during Christmas time. Instead, he composed
the piece so that his singers and audience could enjoy a clever surprise at a
78 Chapter Four
Table 4-3, Formal design in Pérotin and Ceuleers organa quadrupla
concert of sacred music for Lent. A large-scale work in the style of Pérotin
would have undermined the effect he wanted to create. Likewise, Pérotin
undoubtedly could have written a work following Ceuleers’ plan, but in
early-thirteenth-century Paris, such a plan would have almost certainly
undermined the effect he desired to create.
After the Notre Dame style of Pérotin, the next style that Ceuleers likes
to incorporate is Franco-Flemish polyphony, especially that of Gombert
and Josquin. The Requiem, op. 735 (2010) is just one of several of
Ceuleers’ works that draws on this early-sixteenth-century style. Ceuleers
composed the piece shortly after our first meeting (August 25, 2010). At
the time, I had just given a lecture at a conference in Brugge, and Ceuleers
had very kindly agreed to meet with me in Antwerp to discuss his work. I
had already been planning for spring 2011 a graduate seminar on the
polyphonic requiem from its inception to the present, so I asked Ceuleers
what type of requiem he would choose to write; without hesitation, he told
me that he would write a six-voice requiem modeled after Gombert’s
Media vita in morte sumus.7 Before our conversation had ended, he agreed
to compose such a work, and within forty-eight hours, he had completed

both the Introit and Kyrie.
Prior to composing the large multi-movement work, Ceuleers made
some important decisions and considered the resulting challenges and
various solutions. After choosing the style, he decided to score the piece
for six voices, because doing so would ensure sonorous harmonies and a
polyphony that is complex yet not too dense. To add to the richness of the
polyphony, he chose to limit the choir range so that the inner voices would
cross regularly and would never be more than a fourth apart from each
other.
Ceuleers then decided that the entire requiem would be set on the
plainchant. This posed several challenges. Because the requiem chants are
in different modes, there are potential problems with maintaining the
coherence, balance, and unity of the piece as a whole. As illustrated in
Table 4-4, much of the plainchant is in the so-called “high” modes: modes
6, 7, or 8.
movements text mode finalis

Introit Requiem 6 F
Kyrie Kyrie 6 F
Gradual Requiem 2 d
Tract Absolve 8 G
Requiem Sequence Dies irae 1 d
Offertory Domine Jesu Christe 2 d
Sanctus Sanctus 8 G
Agnus Dei Agnus Dei 8 G
Communion Lux aeterna 8 G
Responsory Libera me 1 d
Absolution
Antiphon In paradisum 7 G
Table 4-4, Modes of the requiem plainchants
To counteract the bright, major-like, open character of these chants,

Ceuleers situates the entire choir in a low tessitura (ATTTBB) with
compact spacing between the voices; the result gives the tonal palette a
dark quality.8
Another challenge stemming from Ceuleers’ decision to set the whole
piece on the plainchant is that the chant tunes are often not of great
melodic interest, especially parts of the Introit, Tract, Offertory, and
Communion; this is most likely because in the chant, expressing the
80 Chapter Four
meaning of the text took priority over melodic fecundity. In order to address
this issue and create contrast in these movements, Ceuleers decided to have
the plainchant migrate throughout the voices and have the voices without the
chant develop their own polyphony. To create coherence in these
movements and throughout the piece, he chose to use the incipits of the
chants as motives to be imitated “more or less intensively”.9
Ceuleers also had to confront the potential dilemma caused by the
varying lengths of the texts of each movement. The Sequence is very long,
and he had already set the “Dies irae” in his op. 657 (2006); he therefore
omitted the “Dies irae” from his Requiem, op. 735. Nonetheless, the
Offertory and Responsory “Libera me” have very lengthy texts; working
out these texts extensively would make the Requiem too long for use in a
regular funeral service. Ceuleers wanted his Requiem to be liturgically
functional, and he knew that a lengthy setting might never be sung.
Example 4-5 shows the opening fourteen measures of the Introit. In
this first section, the choir has a dense stretto-like figure, while the third
voice paraphrases the cantus firmus; in measure 4, the chant migrates to
the top voice, as the other voices continue the opening motive. The thick
imitative texture in these opening measures and elsewhere throughout
Ceuleers’ Requiem is very similar to that of Gombert’s polyphony; a
superficial impression of this work might well suggest that the piece was
written at the time of Gombert.
Example 4-5, Ceuleers, Requiem, Introit, mm.1-14 (continued on next page)

Example 4-5, continued

82 Chapter Four
However, closer inspection reveals several decisions that Gombert and his
contemporaries most likely would not have made. Gombert would
probably never have written the diminished fourth between F# and Bb at
the end of m. 9 (see Example 4-5). Ceuleers’ counterpoint is also
occasionally more progressive than that of Gombert and his
contemporaries; the Bb2 that enters just after the A3 in the phrase
beginning in m. 27 (shown in Example 4-6) is not very common in the
sixteenth-century Flemish style.
Example 4-6, Ceuleers, Requiem, Introit, mm. 27-30
Another subtle difference between the settings by Ceuleers and his

predecessors has to do with the texts set to polyphony; in all requiems by
Renaissance composers, the “Te decet” section of the Introit is chanted,
and either the “Et tibi” or “Exaudi me” is set polyphonically; Ceuleers, on
the other hand, has all of these texts chanted.
In the Kyrie, Tract, and Offertory, Ceuleers abandons cantus firmus
and paraphrase techniques and instead opts to work with the motives from
the plainchants. In addition to treating the motives in imitative
counterpoint throughout the voices, he sometimes draws on the six-voice
texture to state motives in two-voice canon(s) with free counterpoint in the
other voices; mm. 13-23 of the Tract (Example 4-7) shows one of many
such passages.
Example 4-7, Ceuleers, Requiem, Absolve Domine, mm. 13-23

84 Chapter Four
At this point during the compositional process, Ceuleers decided that

he would use inversion during the Offertory and Sanctus and combine the
two directions in the Agnus Dei.10 However, once he began writing the
Offertory, he abandoned this idea for two reasons: 1) the plainchant for the
Offertory did not offer melodies of adequate interest for motivic inversion
and 2) postponing the use of inversion until the Sanctus reinforces the
large-scale binary structure of the Ordinarium.11 He decided to structure
the Offertory in a way similar to that of the Introit by polyphonically
setting only the first section (the Responsories “Domine Jesu Christe” and
“quam olim Abrahae”) and by having the second section (the verse
“Hostias et preces tibi”) sung in chant. Out of twelve different requiem
settings by Renaissance composers, only one has the “Hostias” sung
entirely in chant; this is yet another subtle example in which Ceuleers
deviates from the conventional practice of his Renaissance predecessors.12
To emphasize the bipartite design of the entire Requiem Mass,
Ceuleers begins the Sanctus (see Example 4-8) with a return to migratory
cantus firmus technique for the first time since the Introit. For contrast, he
begins the Sanctus with the same motives from the Introit but inverts them.
His use of imitative texture, cantus firmus technique, and his counterpoint
in general are all very similar to what one would expect in Franco-Flemish
polyphony of the early sixteenth century. However, from the standpoint of
harmony, Ceuleers’ Sanctus represents a striking departure from this style.
Without question, very few if any Renaissance composers would ever
write a B-major sonority, especially in a Requiem Mass; yet, as illustrated
in Example 4-8, this passage prominently features numerous recurrences
of the B-major sonority in different positions.
To lend balance and unity to the work as a whole, Ceuleers gives the
Agnus Dei the same overall form as the Kyrie; this allows the outer
movements of the Ordinarium to form a closed circle. To create
momentum in the Agnus Dei, he combines the work’s principal motives
with their inversions. In the first section of the Agnus Dei, he uses only the
ascending motive, which occurs in the Kyrie and Sanctus; in the second
section, he employs only the descending motive, which occurs in the
Introit and Sanctus; and in the final section, he combines the ascending
and descending motives. He also returns to canonic treatment of the
plainchant during the threefold statement of “dona eis requiem,” but he
situates each statement in a different pair of voices.
Example 4-8, Ceuleers, Requiem, Sanctus, mm. 1-18 (continued on next page)
86 Chapter Four

The remaining three movements are similar in construction to the

previous six. Imitation of chant motives in dense polyphony and
occasional cantus firmus technique are prevalent. On the last word of the
entire Mass—“requiem”—Ceuleers brings back the opening theme of the
Introit, which also has the same word. This attention to large-scale
symmetry is prominent in many sixteenth-century settings of the Requiem.
After the sixteenth-century Flemish style, the next style Ceuleers
favors is that of the cantatas by Buxtehude and Bach. The works that
incorporate the eighteenth-century cantata style form a very important part
of Ceuleers’ entire output; in many ways, they are his compositional
fingerprint and reflect his greatest preferences. One piece that very clearly
illustrates this predilection is his Cantata voor zondag Trinitatis, op. 610
(2005).
Cantata voor zondag Trinitatis was commissioned for the inauguration
of the new organ at Christuskerk13 in Antwerp on Trinity Sunday, May 22,
2005. This new organ is a copy of the organ Bach designed at the church
in Störmthal. On November 2, 1723, Bach inaugurated the organ at
Störmthal with his cantata Hochsterwünschtes Freudenfest, BWV 194, and
he revived the work for Trinity Sunday in 1724, 1726, and 1731.14 The
church in Antwerp initially hoped to dedicate their new organ by
performing Bach’s cantata; however, the church did not have enough room
to seat all the musicians required for the work, 15 so Ceuleers was
commissioned to write a cantata specifically for the Antwerp church and
its new organ. 16 Also for the occasion, the church commissioned the
renowned Dutch poet, writer, and preacher Jaap Zijlstra (b. 1933) to write
a text that could easily be fitted as a cantata.
Once Ceuleers had the text, he immediately set to work on the cantata.
Like BWV 194, Ceuleers’ cantata comprises two large parts. Table 4-5
shows a comparison of the movements in each cantata.
88 Chapter Four
Table 4-5, Scoring and Movements in BWV 194 and op. 610
The opening chorale of op. 610 is in the style of Bach, but it is Ceuleers’
own melody; after an orchestral ritornello, the choir enters, and the
soprano sings the newly composed chorale tune in long note values. For
coherence, Ceuleers has the orchestra continue stating motives from the
ritornello even after the choir’s entrance; subsequent ritornelli are varied,
but they develop the same motives.
In the subsequent sections, Ceuleers raises the tension and momentum
from one recitative to the next and from one aria to the next. He does so
not only by increasing texture and melodic interest but also by employing
greater stylistic contrast, richer text painting, and more vivid expression of
the musical narrative. The first aria, for example, is a French gavotte, but
the next one is Italianate and more reminiscent of Vivaldi.
In his BWV 194, Bach also incorporates a variety of styles. Bach
begins with a chorale based on the structure of a French overture. The
following recitativo secco, in which the bass is supported only by basso
continuo, is Italianate in style; the subsequent aria is a gigue, and the
soprano recitative after that is a gavotte.
Like Bach, Ceuleers makes clever use of text painting. In mm. 17-18 of
the first recitative, the soprano sings a relatively lengthy and fluid melisma
on the words “as water flows” (see Example 4-9).
Example 4-9, Ceuleers, Cantata voor zondag Trinitatis (2), mm. 14-18
However, in the second aria, the text painting becomes much more vivid,
and the musical narrative even more eloquent.
This da capo aria for bass opens with the following words:
Wij vragen om de goede Geest, We ask for the good Spirit,

het orgel zij U toegewijd, the organ dedicated to you,
de gratie van de boventoon, the grace of the overtone,
de warme gloed, the warm glow,
de onderstroom. the undercurrent.
Both text and music convey two narratives. The surface narrative, which is
for Trinity Sunday, asks that the Holy Spirit would come during our time
of adversity.17 It asks that the organ be dedicated to the Holy Spirit and
entreats the Holy Spirit for His grace, warmth, and motion (undercurrent)
through the congregation. Together “Boventoon” and “onderstroom”
(from the heights to the depths) depict the omnipresence of the Holy Spirit.
The deeper narrative relates the Holy Spirit to the “good wind” flowing
through the organ pipes. Here, the congregation asks that the organ (also
referring to the Holy Spirit) would have graceful overtones, a warm glow
or timbre, and an undercurrent. This undercurrent is the Holy Spirit, but it
is also the wind that flows invisibly through the pipes of the organ.
Example 4-10 shows numerous instances in which Ceuleers musically
conveys the narrative in the aria.
90 Chapter Four
Example 4-10a, Ceuleers, Cantata voor zondag Trinitatis (5), mm. 25-29
On the word “orgel,” shown in Example 4-10a, Ceuleers gives the

organ a prominent, virtuosic solo, while the other voices sustain their
pitches. According to Laurence Dreyfus, Bach composed six cantatas that
include a “substantial concerted solo part for the organ.”18 None of his
cantatas for Trinity Sunday—not even BWV 194, which had been used for
the dedication of the organ at Störmthal in 1723—includes scoring for the
organ.19 The abridged version of BWV 194 used for the performance in
1726, however, does rescore two arias to replace one of the oboes with
obbligato organ.20 The text in all surviving versions of BWV 194 focuses
more on the consecration of the church than on that of the organ; unlike
Ceuleers’ cantata text, Bach’s text makes no mention of the organ but
instead celebrates the newly erected sanctuary. 21 From this standpoint,
Ceuleers’ cantata is even better suited for the Antwerp Church than Bach’s
cantata.
On the word “boventoon,” shown in Example 4-10b, Ceuleers has the
organ gracefully ascend to and hover around its highest note C6. In
Example 4-10c, on the word “onderstroom,” Ceuleers has the organ
descend to the bottom of its range. Throughout the passage in mm. 25-36,
the organ serves a dual role as both protagonist and narrator. It is the
protagonist of the textual and musical narrative depicting the instrument’s
properties; at the same time, it serves as a narrator musically portraying the
qualities of the Holy Spirit.
Example 4-10b, Ceuleers, Cantata voor zondag Trinitatis (5), mm. 30-32
92 Chapter Four
Example 4-10c, Ceuleers, Cantata voor zondag Trinitatis (5), mm. 33-36
In addition to using the work to consecrate the organ at the Antwerp

church, the cantata serves as a personal dedication of a “sixth flute” made
for Ceuleers by his recorder professor Baldrick Deerenberg.22 Today, this
instrument would be called a sopranino recorder in D, but in the baroque
era, it was considered a high treble recorder or “sixth flute,” since it is
transposed a sixth up from the conventional recorder in F. Before the alto’s
entrance in the cantata’s penultimate movement, Ceuleers gives this little
flute a prominent, virtuosic solo, just as he had done for the organ earlier
(see Example 4-11).
94 Chapter Four
As in Bach’s BWV 194, the prima pars of Ceuleers’ cantata ends with
a chorale consisting of a two-verse prayer to the Holy Spirit. Referring
back to Table 4-5 shows that both composers alternate between recitatives
and arias in the secunda pars of their cantatas; unlike Bach, Ceuleers
proceeds from the last aria to the closing chorale and therefore has only
eleven numbers instead of twelve, as Bach has. Both composers reserve
duets for the secunda pars; both frequently employ recitativo secco; and
both incorporate similar styles and techniques such as gavotte style and
fugal writing. Bach ends his cantata with a two-verse chorale making
supplication to God in general; the tune he uses is based on “Nun laßt uns
Gott dem Herren” by Nikolaus Selnecker (1528-92). 23 However, as
illustrated in Example 4-12, Ceuleers bases his closing chorale on the late-
nineteenth-century English tune Lauda anima by Sir John Goss (1800-80).
Unlike Bach, Ceuleers ends his cantata with a three-verse chorale in which
each verse expresses thanks to each Person of the Trinity. Ceuleers’
chorale thus seems more specifically suited to Trinity Sunday than Bach’s.
The most modern style that Ceuleers likes to incorporate is that of
Charles Tournemire (1870–1939) and Marcel Dupré (1886–1971), and one
of his best examples drawing from these composers is his Orgelmis voor
het Heilig Hart van Jezus, op. 688 (2007). Tournemire was a pupil of
Franck and went on to teach Langlais, Messiaen, and Duruflé. Along with
other French contemporaries such as Debussy, Tournemire contributed
new ideas about harmony. Dissonances no longer had to resolve, but
became part of the independent harmony; they produced colors that could
stand on their own. This led to new ideas about tone color and harmonic
stability. Tournemire’s style, though often modal, plays with the ambiguity
between major and minor modes. For Tournemire, an important source of
musical momentum is the tension between juxtaposed tonal systems, such
as the eight church modes, major/minor modes, added chords, and

pentatonic as well as other pitch collections. Like his pupil Duruflé,
Tournemire often supports modal chant with extended tertian harmonies.
Dupré, who described Tournemire as a formidable rival,24 studied with
Vierne and Widor. He taught two generations of organists and composers
including Messiaen, Langlais, and Alain. Described by Messiaen as “the
Liszt of the organ and perhaps the greatest of all the virtuosi that have ever
existed,”25 Dupré developed standards of the French Symphonic tradition
more than any other organist of his generation, and he achieved an
unprecedented degree of fame as a composer, recitalist, liturgical organist,
author and editor. 26 Features of his style include extreme virtuosity,
meticulous attention to color, high chromaticism, and thick chords comprising
as many as nine different pitches. 27 Like Tournemire, Dupré relies on
Gregorian themes and religious imagery. 28 According to Daniel Dries,
Dupré’s organ works represent the pinnacle of the French Symphonic
Organ tradition.29
The Introit of Orgelmis voor het Heilig Hart van Jezus, op. 688 begins
with a repeating subject; as the subject repeats, the harmony becomes
increasingly thicker. With each statement of the subject, Ceuleers adds a
new voice in fourths, sevenths, ninths, and elevenths (see Example 4-13).
In this passage and elsewhere throughout the piece, dissonances often arise
from linear processes. Each pitch in each chord plays a crucial role in
producing the desired color. Like Tournemire, Ceuleers always employs
the plainchant, and like Dupré, he usually situates the chant in a sound
world of dense, highly chromatic counterpoint producing complex and
colorful harmonies.
96 Chapter Four
Example 4-13, Ceuleers, Orgelmis, op. 688 (I), mm. 1-9
In the Gradual, Ceuleers seamlessly blends the styles of Tournemire

and Dupré by alternating between a thin, modal texture and a thick,
dissonant texture (see Example 4-14a and 4-14b). After contrasting
between these textures, Ceuleers concludes the movement with a measure
that is reminiscent of Duruflé. As illustrated in Example 4-14c, the free
meter, colorful harmonies, and prominent chant melody are all qualities
found in the works of both Tournemire and Duruflé.
Example 4-14a, Ceuleers, Orgelmis, op. 688 (II), mm. 1-6
Example 4-14b, Ceuleers, Orgelmis, op. 688 (II), mm. 11-14
Example 4-14c, Ceuleers, Orgelmis, op. 688 (II), m. 21

98 Chapter Four
The Offertory (fourth movement) shows a rare instance in which

Ceuleers fuses different styles within the same work. Inspired by Josquin,
the movement opens with paired duets followed by a tutti passage. By
measure 16, the consecutive seconds in a three-voice texture are inspired
by Dupré. The chromatic motion in mm. 50-59 and bitonality especially in
mm. 70-77 are also features of Dupré’s works such as his Symphony for
Organ and Orchestra, op. 25 and Organ Concerto, op. 31.
The sixth movement, entitled “Sortie: toccata,” evokes the toccatas of
Dupré’s teacher Vierne. In a recent interview, Ceuleers specifically
mentions Vierne’s Symphony No. 5 as an inspiration for much of this
movement.30 However, I would also like to suggest the last movement of
Vierne’s Symphony No. 1 as a potential inspiration for Ceuleers; Example
4-15 offers a comparison of excerpts from the two works.
Example 4-15, Comparison of excerpts by Vierne and Ceuleers
Measure 58 of this movement shows an excellent example of the kind

of tonal and harmonic ambiguity inspired by Dupré’s music. The sonority
on the downbeat initially appears to be a German augmented-sixth chord31
in the key of Bb major or Bb minor; with this interpretation, the E5
resolves to a metrically weak F4 in the same measure, and the Gb never
resolves. Another possibility is to interpret the sonority as a dominant-

seventh chord in B minor; with this interpretation, the Gb becomes the root
F#, and the chordal seventh E, does not resolve. As illustrated in Example
4-16, the sonority does eventually resolve to B minor in m. 60.
Example 4-16, Ceuleers, Orgelmis, op. 688 (VI), mm. 57-60
The influence of Franck, Vierne, and especially Tournemire and Dupré is

evident throughout Ceuleers’ piece. An initial impression would reasonably
suggest the work was composed in the 1920s. As we have already seen,
however, this and all of the works examined reveal Ceuleers’ own musical
language. Moreover, all of the pieces are not mere copies of earlier works
and styles; rather, they are entirely the result of the composer’s own
decisions, which were made before, during, and throughout the
compositional process.
As we look back at the music presented over the course of this essay,
imagining that one and the same person made the compositional decisions
for all of these works may seem rather difficult. We see an exceptionally
broad variety of styles spanning the last eight centuries, and we notice an
equally diverse range of vocal and purely instrumental genres. Also quite
astonishing is the fact that all four of the works discussed in this article
were written within only five years of each other (2005-10); moreover,
referring back to Table 4-1, we notice that these works constitute only a
very small fraction (0.5%) of the composer’s current output.
In addition to being highly prolific across an extensive spectrum of
styles, the composer displays remarkable preciseness and skill. From his
technical prowess with respect to counterpoint, harmony, and thematic
development, to his vivid expressivity achieved through creative
manipulation of musical elements in text painting and portraying narrative,
Ceuleers has an eloquent and multi-faceted musical language that
continues to be shaped and polished by his experiences as a performer,
student, and composer of music. He writes for our ears to hear: I hope that
100 Chapter Four
this introductory study will serve as a catalyst for further research, greater
exposure, and deeper appreciation of the music of Willem Ceuleers.
Willem Ceuleers (2008)

Notes
1
The author wishes to thank Willem Ceuleers for kindly and very generously
sharing his time, insight, and materials.
2
N.B. this is the same church where Benedictus Appenzeller (c.1480/88-after
1558) worked as choirmaster from 1556 to 1558.
3
Tim S. Pack, personal interview with Willem Ceuleers, Wijnegem, Belgium,
August 29, 2012.
4
Tim S. Pack, personal interview with Willem Ceuleers, Antwerp, Belgium,
August 25, 2010; Willem Ceuleers, e-mail message to author, April 8, 2011.
5
Pack, interview, August 29, 2012.
6
I am using Vulgate numbering for the Psalms.
7
Ceuleers’ Requiem does not incorporate thematic or melodic material from
Gombert’s work; rather, he uses the style of Gombert’s work as a springboard for
writing his own piece.
8
Comments furnished by the composer on September 20, 2010.
9
10
11
The Mass consists of two large sections: the Liturgy of the Word and the Liturgy
of the Eucharist. The Sanctus bifurcates the Mass in that it is the first movement of
the Ordinary occurring in the Liturgy of the Eucharist. Most composers who set the
Mass to music reinforce this bipartite structure by creating some type of special
contrast in the Sanctus.
12
Settings compared include those of Anerio, Appenzeller, La Rue, Magalhães,
Moulinie, Ockeghem, Palestrina, Prioris, Pujol, Richafort, Sermisy, and Vásquez.
Vásquez is the only composer to avoid polyphony in the Offertory verse.
13
Christuskerk is a Protestant church at 13 Bexstraat in Antwerp.
14
Werner Neumann, Handbuch der Kantaten Johann Sebastian Bachs
(Wiesbaden: Breitkopf & Hartel, 1974), p. 154.
15
Bach’s cantata calls for full choir, STB soloists, 3 oboes, bassoon, strings, and
basso continuo; Ceuleers’ cantata reduces this number to SATB soloists, recorder,
oboe, violin, organ solo, and basso continuo.
16
Willem Ceuleers, e-mail message to author, September 11, 2012.
17
The previous recitative ends with the words, “In a time of harsh words and hard
facts, we pray for tenderness of language, warmth of voice, and a quiet place to
be.” The subsequent aria then calls on the Holy Spirit for grace and warmth.
18
Laurence Dreyfus, “The Metaphorical Soloist: Concerted Organ Parts in Bach’s
Cantatas,” Early Music 13/2 (1985): 237.
19
Alfred Durr, The Cantatas of J.S. Bach: With Their Librettos in German-English
Parallel Text (Oxford: Oxford University Press, 2005), p. 719.
20
Thomas Braatz, “Cantata BWV 194 Discussions: Part 2,” <http://www.bach-
cantatas.com/BWV194-D2.htm> (accessed September 13, 2012).
21
Durr, p. 719.
22
Willem Ceuleers, e-mail message to author, September 13, 2012.
23
The text of the closing chorale in BWV 194 is by Paul Gerhardt (1607-76).
102 Chapter Four
24
Marcel Dupré, Recollections, trans. Ralph Kneeream (Melville: Belwin-Mills,
1975), p. 96.
25
Claude Samuel, Conversations with Olivier Messiaen (London: Stainer & Bell,
1976), p. 27.
26
Daniel Dries, “Marcel Dupré: The Culmination of the French Symphonic Organ
Tradition (Doctor of Creative Arts thesis, University of Wollongong, 2005), pp. xi
and xiv.
27
Ibid., p. 148.
28
Ibid., p. 176.
29
Ibid., p. 176.
30
Tim S. Pack, personal interview with Willem Ceuleers, Wijnegem, Belgium,
August 29, 2012.
31
Ceuleers himself describes this sonority as a modified c chord with a flat ninth,
lowered fifth, and omitted root in the key of B major/b minor. He agrees that the
chord functions both as a predominant in B /b and as a dominant (F ) in B/b.
PART II:
THE TONAL TRADITION

INTRODUCTION TO PART II
Part II, “The Tonal Tradition,” considers a small portion of the vast
repertoire of compositional techniques that living composers have
inherited, or could inherit, from previous eras. Jessie Thornton’s chapter
on Schubert’s “Du liebst mich nicht” describes the song’s progression
from spiraling modulations to harmonic clarity, and interprets this in terms
of the myth of Narcissus, the song’s principal subtext. Jack Boss’s article
explores the ways Mahler’s sense of musical coherence may have
influenced his friend and successor, Arnold Schoenberg, using a blend of
Schenkerian and Schoenbergian approaches to analyze the Adagio from
the Tenth Symphony. The significance of Schenkerian analysis for early
twentieth-century music is explored in a different way by Gary Don, who
explores “interruption” structures that involve root movements by third in
Debussy’s Sarabande from Pour le piano.
The three essays in this section are also linked by an analytical
approach that treats these pieces not as abstract artworks isolated from the
currents of history and culture, but rather as products of their particular
time, which are best understood only if we remain aware of what
influenced them and how they were received. One sees evidence of this
historically-informed approach in Jessie Thornton’s equal interest in the
harmonic peculiarities of Schubert’s song as well as in the larger ideas that
likely inspired them, in Jack Boss’s attention to the musical consequences
of a close mentorship, and in Gary Don’s careful consideration of how
Debussy’s music has been interpreted by his contemporaries and latter-day
listeners alike. In this sense as well, these works are truly living, even if
their composers are no longer with us.
—Stephen Rodgers
CHAPTER FIVE
HARMONY AND THE MYTH

OF NARCISSUS IN SCHUBERT’S
“DU LIEBST MICH NICHT”
JESSIE THORNTON
Schubert’s setting of August von Platen’s poem “Mein Herz ist

zerissen, du liebst mich nicht” is remarkable, even when compared with
his other songs, for its extravagant harmonies and unconventional
modulations that seem to thwart tonal logic. This fact has not gone
unnoticed by commentators on the song; some have even gone so far as to
take these extravagances as an indication that the song is of questionable
quality. Kristina Muxfeldt notes in her article, “Schubert, Platen, and the
Myth of Narcissus,” that “assessments of the song’s abnormal harmonic
language have generally concluded that it is, in the end, an experimental,
not entirely persuasive work.”1 Kofi Agawu writes that “the overall effect
of the song transcends what we would expect from normative tonal-
functionality,” adding that the song “may be seen to aspire towards (but
perhaps to miss?) a higher status.”2 Analyses of “Du liebst mich nicht”
have tended to focus on the spiraling modulations in the first half of the
piece, paying less attention to the sudden turn to harmonic clarity at the
repeat of the last four lines of the poem.3 The “problem” this piece poses,
and part of the reason that it has not always been accorded the status it
deserves, is that these two halves do not seem to fit together. There is an
apparent lack of coherence among the harmonic languages of the song’s
various sections: the tonal audacity of the first half, the harmonic
normalcy of the second half, and the seemingly inappropriate harmonic
departures of the final section, which place an inordinate emphasis on the
(frankly) unremarkable final lines of the poem.
My paper explores how these seemingly disparate sections relate, in an
effort to better account for how Schubert uses harmony and tonality to
Harmony and the Myth of Narcissus in Schubert’s 107
“Du liebst mich nicht”
present a deeply sensitive—and altogether successful—musical

interpretation of Platen’s poem. Specifically, I contend that the
disorientation of the first half of the song draws attention, via the sudden
emergence of tonal logic, to the revelatory climaxes of the second half.
Further, I propose that knowledge of the Narcissus myth is essential if we
are to fully appreciate the deeper meaning of the song’s harmonic design.
The following discussion is organized into three sections. The first section
presents a detailed analysis of the first half of the song and aims to show
that its modulations are not in fact haphazard, as Agawu suggests,4 but
rather part of a strategic pattern of ultimately denied tonal expectations.
The second section analyzes the double climaxes on the words
“vermissen” (to miss) and “Narzissen” (a bulbous flower), and proposes
that the crux of the second half of the song actually depends on the aural
disorientation of the first half. The final portion of my paper draws on
Muxfeldt’s work to connect the threads that link the composer, the poet,
and the myth to this piece, and then demonstrates how an
acknowledgement of these connections can help us to understand the
piece’s overall harmonic organization.
* * *
Platen’s poem is a five-couplet Persian-style ghazal that repeats its title

phrase as an obsessive refrain, reiterating the poet’s agony in unrequited
love. The original poem and an English translation appear below.5
Mein Herz ist zerrissen, du liebst mich My heart is torn apart, you love me
nicht! not!
Du ließest mich’s wissen, du liebst mich You have let me know it, you love me
nicht! not!
Wiewohl ich dir flehend und werbend Though I have pleaded with you,
erschien, wooed you,
Und liebebeflissen, du liebst mich nicht! Appeared in ardent pursuit, you love
Du hast es gesprochen, mit Worten me not!
gesagt, You have spoken it, cast it in words,
Mit allzugewissen, du liebst mich nicht! In all too certain ones, you love me
So soll ich die Sterne, so soll ich den not!
Mond, So shall I the stars, so shall I the
Die Sonne vermissen? du liebst mich moon,
nicht! The sun relinquish? you love me not!
Was blüht mir die Rose, was blüht der What to me are the blooming rose, the
Jasmin? jasmine,
Was blühn die Narzissen? du liebst mich The Narcissi in bloom? you love me
nicht! not!
Example 5-1, “Mein Herz ist zerissen, du liebst mich nicht,” August von Platen
108 Chapter Five
The reason for the lack of reciprocity is never addressed, as only the
poet’s own voice bemoans the events that lead to this emotional
tribulation. The essential storyline, revealed through the poet’s
impassioned outburst, is that the beloved was not swayed by the poet’s
advances; at the beloved’s cold rejection, the poet is thrown into hysterical
despair, rejecting all of nature—the sun, the moon, the stars, and the
flowers—because his love will never lead to anything beyond the fleeting
hope that created it.
The first half of the song (mm. 1-40) projects a spiral of tonal centers,
organized around the prolonged expectation and ultimate denial of F
tonality. (Readers will want to consult a score as they read the following
analysis. The score is readily available in a Dover edition, as well as from
other publishers. It can also be found online at the International Music
Score Library Project [IMSLP].)6 The four-bar introduction, with the
resolution to F major in m. 4, sets up the expectation of further
development in the F realm, but at the same time pales in comparison with
the distant modulations that follow. The abrupt harmonic change from E to
C7 in m. 13 leads us to expect further development of F major, an
expectation heightened by the prolongation of a C dominant over the next
three measures. But this dominant does not resolve to a tonic; instead, it is
completely undercut by a swerve toward G minor in mm. 17-18 (at “und
liebebeflissen”). The dominant of G minor likewise fails to lead to a G
minor tonic—this key area is never solidified with an authentic cadence,
as we tumble to a sudden and discordant arrival on Ab major (m. 20).
From here onward, we see perfect authentic cadences in Ab major (m.
20), G major (m. 24), and Gb major (m. 28). Agawu suggests that this first
section lacks any specific tonal orientation—that, in regard to these
spiraling modulations, “we simply end up where we end up, not because
we set off from that point.”7 I would argue that the descending half-step
modulations suggest not haphazard points of arrival, but rather a strategic
pattern of tonal expectations that substantially increase the longed-for
resolution to F. First, notice that these arrival points occur every four bars;
if they were indiscriminately designed, why the temporal consistency?
Second, if the sequence of chromatically descending modulations were to
continue, the next authentic cadence would undoubtedly be in F major,
resolving the harmonic expectation that had been established ever since
the unresolved dominant in mm. 13-16. Third, if the pattern of keys
established with authentic cadences (from the beginning of the piece until
m. 28) were reconstructed as individual pitches, they would spell out an
A-Ab-G-Gb chromatically descending tetrachord, or a truncated lamento
bass line. This subtle harmonic reference to a lament is consistent with the
grief-laden subject matter of the poem.
Schubert’s initial setting of the last four lines of the poem (mm. 29-40)
is marked by increased harmonic and hypermetrical instability—a
reflection of the poet’s increasing hysteria—and also by the culmination of
F expectancy, along with its ultimate and conclusive denial. The unstable
modulations that characterized the first section of the song were
nevertheless paired with normative four-bar phrase structures. Therefore,
when even this rudimentary stabilizer is removed and the harmonic chaos
persists, the piece seems to spin out of control: starting in m. 29, the
phrases expand from four to five bars in length and remain as such for the
duration of the song. Example 1 shows the first of these five-bar phrases,
in mm. 29-33.
Example 5-2, Five-bar phrase in mm. 29-33
In addition to phrase expansion, the poet’s cries climb staggeringly

higher and higher in register, propelled forward by the melodic doubling
in the piano accompaniment on each new utterance of the line “du liebst
mich nicht,” in accordance with the poet’s hysterical renunciation of
nature (mm. 32-33, 36-37, and 38-39). In the midst of these destabilizing
agents, the harmonic insistence on F remains constant. Recall that
Schubert’s setting of the first six lines ended in Gb major, the last key
center in the string of chromatically-descending authentic cadences (m.
28). In the next line (“So soll ich die Sterne…”), we expect F major to be
the next cadence; Schubert increases this feeling of anticipation with a
string of E diminished seventh chords in root position (mm. 29-31),
prolonged for three measures before being reinterpreted as a common-tone
diminished seventh chord that resolves not to F, but to the dominant of the
key of A minor (m. 33). Against all expectation, this section does not lead
to an authentic cadence in F major, but rather to a half cadence in A major,
the parallel of the home key. Example 5-3 shows how the E diminished
seventh chord might have resolved, and then how it actually resolves.8
Example 5-3, Resolutions of the E diminished seventh (possible and actual)

110 Chapter Five
It is at this point that the ultimate enervation of F begins. The entire

harmonic progression of mm. 29-31, which begged to resolve to F major,
is repeated verbatim in mm. 34ff., though enharmonically respelled and
transposed up a minor second. This rising-semitone sequence finally
provides us with F at the end of m. 3—but, as David Gramit notes, “in a
form that undermines any sense of resolution.”9 The fact that a new
diminished seventh chord appears above F in m. 34 changes our aural
perception; it forces us to revise our harmonic expectations, such that
when we finally arrive on an F major triad, it is heard not as a tonic, but as
the dominant of Bb minor. This is yet another key that is suggested but
ultimately never attained, as the music makes a sudden dive toward A
minor in mm. 37-38. The cadence in mm. 37-38 is repeated in mm. 39-40
(thus extending this five-bar phrase by two additional measures). This
hypermetrical expansion repeats what we might call the “mockery of F,”
before leading to a cadence in A major on the downbeat of m. 40.
* * *
The song could have reasonably ended here: poetically and musically,
there is no need to repeat the last four lines of the poem—so the fact that
Schubert repeats them is a sign of how important they are to his musical
interpretation of Platen’s text. The repetition of these lines marks the
official return of harmonic coherence, followed by two bold harmonic
departures to C# minor on the words “vermissen” and “Narzissen.” After
the cadence in A major (in m. 40), the harmonic language finally
stabilizes—paradoxically enough, the more standard harmonies sound
jarring when juxtaposed with the modulatory chaos that came before them.
Though a normal harmonic syntax is reestablished, it does not stall the
dynamic momentum that began with the denial of F only measures before.
On the contrary, the voice continues to climb higher in register, becoming
ever more frantic. The irregular phrase lengths continue as well, brought
about by an augmentation of the declamatory rhythm at the climactic high
points and subsequent text repetitions: if the half notes on the second
syllables of “vermissen” and “Narzissen” had been quarter notes, and if
the words “du liebst” had not been repeated in the following measures, the
two phrases of the final section could easily be rewritten to span four bars
rather than five and seven, respectively. Finally, Schubert maintains a
degree of harmonic dissonance through the emphatic repetition of a tonic
pedal tone in mm. 40-42 and again in mm. 45-47. Each musical
component coalesces to form one catalyst, propelling the music forward to
the double climaxes on the words “vermissen” and “Narzissen.” These
climaxes represent the only instance of genuine departure from A major.
Harmonically, registrally, dynamically, and rhythmically, they are marked

as the most important moments in the song.
To some extent, the emphasis on C# minor makes musical sense, since
the chord lies a third above the tonic A, just as F major lies a third
below—indeed, the strong presence of C# makes the absence of F all the
more apparent. But the sudden mode change to a bright A major and the
more straightforward harmonic idiom of the final section are rather harder
to explain. David Gramit writes, “The turn to A major for the repeat of the
final lines seems wholly out of touch with the devastating reality that the
drastic rhythmic displacements and dynamic climaxes of the final
repetitions proclaim, just as the text’s extravagant farewell to the sun,
moon, and stars is followed by a disconcertingly modest enumeration of
flowers.”10 After the vortex of (organized) harmonic chaos, the
unexpected change to a key that is not only harmonically stable but major
seems entirely “out of synch” with the rest of the piece. Furthermore, one
might argue that the two climaxes place too much emphasis on
“vermissen” and “Narzissen.” How are we to account for these oddities?
* * *
Though the musical climax on “vermissen” can be reasonably justified,

the presence of an equally accentuated climax on a word that bears so little
significance to the poetic narrative seems altogether peculiar, without
knowledge of the Narcissus myth and how it relates to Schubert, Platen,
and “Du liebst mich nicht.”
In her article on Schubert’s song, Kristina Muxfeldt recounts the
Greek myth of a young man named Narcissus, whose exceptional beauty
caused everyone to fall in love with him.11 His indifference toward his
admirers often drove them into deep despair, or even compelled them to
commit suicide. One admirer, whose attachment to Narcissus is of direct
importance to this song, was Echo, a nymph who had been robbed of her
ability to initiate speech, so that the only way she could communicate was
by “echoing” another’s words and changing their inflection to reflect her
meaning. Communication between Echo and Narcissus met with
disastrous results: Echo tried desperately to express her love in
Narcissus’s own words, but he did not understand and became
increasingly annoyed by her “mocking” repetition. When Echo tried to
embrace Narcissus, he cried out, “Hands off! Embrace me not! May I die
before I give you power over me!” In a last attempt to communicate her
love, Echo repeated the words “I give you power over me!” Overcome
with anger, Narcissus cruelly rejected Echo, and she melted away until
nothing but her own voice remained behind in the hills. After numerous
112 Chapter Five
other encounters that met with similar heartbreaking results, a young man,
whom Narcissus had disdained, prayed that Narcissus would futilely fall
in love the way others had fallen in love with him. That day, tired and
thirsty from hunting, Narcissus came to a pool of water. Upon looking into
the pool, he became mesmerized by his own reflection. He tried to
embrace his beloved, becoming ever more desperate to be united with the
image. It was finally the lack of an echo that caused him to realize his
folly, and upon realizing that he could not be united with his reflection he
dissolved into a pool of tears and a Narzissen flower grew up in his
place.12
To appreciate how this myth relates to the music, we must first
understand how this myth relates to Platen and Schubert. Muxfeldt claims
that the poem “Mein Herz ist zerrissen, du liebst mich nicht” could
“almost be carved out of [Platen’s] diary entries so closely do their words
correspond.”13 Furthermore, she argues that the character of Narcissus was
an “object of reflection,” or an “idée fixe,” that came up repeatedly both in
Platen’s poems and in his diary entries.14 Muxfeldt suggests that Narcissus
was such a prevalent image in the writings of Platen because of Platen’s
own homosexual tendencies, and because of the frustrating social
restrictions that forced him to abstain from pursuing these desires.
Specifically, while attending school in Würzburg, Platen had fallen in love
with a young man who, in his mind, closely resembled the Narcissus
protagonist. Though Platen was indeed extremely attracted to this younger
student, he feared making contact and resorted instead to recording every
interaction they had in his diary, safely entertaining the notion of a
romantic relationship from a distance. On December 10, 1818, the
following statement appeared in Platen’s diary: “Wohl sah ich, dass du
liebst aber du liebst nicht mich!” (Surely you love, but it is not me you
love!). The records we have of Platen’s diary reveal that he was familiar
with and had a great interest in the myth of Narcissus. Platen even goes so
far as to compare his “beloved” with Narcissus, addressing him by that
name in his diary.
The salient connections between Platen’s diary entries, the poem, and
the Narcissus myth aside, the question that still needs to be answered is
how much Schubert knew (if anything) about Platen’s diaries, his
homosexuality, and the myth of Narcissus. According to Muxfeldt,
Schubert and Platen had an invested mutual friend, Franz von Bruchmann.
Bruchmann and Platen went to school together at the University of
Erlangen where they, ironically enough, shared a class on mythology.
Bruchmann took great interest in Platen’s poetry, often taking multiple
copies of his work back to the community of scholars in Vienna, where he
“frequently headed the reading circle Schubert and his friends had
formed.”15 Though we have no direct documentary evidence that Platen

and Schubert ever met, the music of Schubert’s song offers its own sort of
evidence. The peculiarity of Schubert’s setting of “Du liebst mich
nicht”—with its striking emphasis on the word “Narzissen,” its
compositional incorporation of an echo, and its repetition of the last four
lines—suggests that he was quite familiar with the mythical innuendos
hidden in Platen’s poem.
Knowledge of the Narcissus myth then becomes essential for us to
understand the climatic departures in the second half of the song.
Muxfeldt argues that these two climaxes “convey an indirect meaning by
the same strategy Echo used to communicate with Narcissus: [Schubert]
inflects the meaning of a deceptively bland text—‘Was blühn die
Narzissen’—by means of a well-placed echo.”16 This explains the
somewhat confusing presence of two climatic moments rather than one. In
addition, it explains why both climaxes are almost identical, the only
difference being that in m. 47 the piano leaps down to the A1 one beat
earlier than it had in m. 42. Schubert thereby offers a musical imitation of
the “echo” from the poem. Furthermore, Schubert’s decision to set the
repetition of the final lines in a stable A major, after the competing
tonalities of the previous measures, makes the harmonic departures on
“vermissen” and “Narzissen” sound like true harmonic departures—that
is, a stable tonal backdrop is required for these departures to be fully
registered. The chromatic moves to C# minor are not flaws, inordinate
inflections of a “deceptively bland text”; they are signs of Schubert’s deep
understanding of that text, and his willingness to let the submerged
intertextual meanings of the poem affect the harmonic design of the music.
In conclusion, “Du liebst mich nicht” is about more than bizarre
chromatic modulations that are undermined by a turn to tonal conformity;
rather, the myth supplies revelatory meaning to the poet’s unreciprocated
love story. On the sole basis of a more or less literal poetic reading,
without knowledge of the Narcissus myth, the entire second half of the
song is all too easily misunderstood and therefore dismissed. However, the
second half of the piece cannot be understood without the first—and vice
versa. The forced normality of the end of the piece, falling on the heels of
the extraordinary modulations of the first half, draws attention to the
poetic revelation of the final section of the song: chaos, in short, cedes to
logic. If the entire piece were to uniformly disregard normative tonal
syntax, the harmonic departures on the words “vermissen” and
“Narzissen” would be imperceptible; on the other hand, if this work
followed a more conventional tonal organization throughout, the
heartbreak of the poet would not be so keenly felt. It is the stark contrast
that brings the piece and poem to life.
114 Chapter Five
Notes
1
Kristina Muxfeldt, “Schubert, Platen and the Myth of Narcissus,” Journal of the
American Musicological Society 79/3 (1996): 487. An updated version of the
article appears in Muxfeldt’s recent book, Vanishing Sensibilities: Schubert,
Beethoven, Schumann (Oxford: Oxford University Press, 2012), pp. 160-96.
2
V. Kofi Agawu, “Schubert’s Harmony Revisited: The Songs ‘Du liebst mich
nicht’ and ‘Dass Sie hier gewesen,’” Journal of Musicological Research 9/1
(1989): 32.
3
See, for example, Agawu and also Susan Youens, “Schubert and the Poetry
of Graf August von Platen-Hallermünde,” The Music Review 45/1: 19-34.
4
Agawu, p. 31.
5
The English translation is taken from Muxfeldt, p. 488.
6
“IMSLP / Petrucci Music Library,” http://imslp.org/wiki/Main_Page (accessed
August 13, 2012).
7
Agawu, p. 31.
8
The E diminished seventh chord actually resolves to an E dominant ninth, but I
have simplified the voice leading and used a dominant seventh in my example.
9
David Gramit, “Orientalism and the Lied: Schubert’s ‘Du Liebst mich nicht,’”
19th-Century Music 27/2 (2003): 104.
10
Ibid., p. 106.
11
Muxfeldt, pp. 497-98.
12
My retelling of the myth has benefited from my reading of Muxfeldt’s article.
13
Muxfeldt, p. 500.
14
Ibid., p. 497.
15
Ibid., p. 503.
16
Ibid., pp. 506-07.
CHAPTER SIX
MAHLER’S MUSICAL IDEA:

A SCHENKERIAN-SCHOENBERGIAN
ANALYSIS OF THE ADAGIO
FROM SYMPHONY NO. 10
JACK BOSS
Gustav Mahler’s multiple roles as friend, inspiration, and benefactor to

the emerging composer Arnold Schoenberg are amply documented in
Schoenberg’s letters to Mahler, in Alma Mahler’s memoirs, and in
Schoenberg’s essay on Mahler written as a tribute after the older
composer’s death. Alma Mahler writes about “youthful” Schoenberg and
his brother-in-law, Alexander von Zemlinsky, making trips to the Mahler
household for dinner that did not end well in some cases:1
Schoenberg, on the other hand, was inspired by a youthful rebelliousness

against his elder, whom at the same time he revered. They [Schoenberg
and Zemlinsky] used to come in in the evening. After one of our
devastatingly simple meals, all three went to the piano and talked shop—at
first in all amity. Then Schoenberg let fall a word in youthful arrogance
and Mahler corrected him with a shade of condescension—and the room
was in an uproar. Schoenberg delighted in paradox of the most violent
description. At least we thought so then; today I should listen with
different ears. Mahler replied professorially. Schoenberg leapt to his feet
and vanished with a curt good night. Zemlinsky followed, shaking his
head.
As soon as the door had shut behind them, Mahler said: “Take good care
you never invite that conceited puppy to the house again.” On the stairs
Schoenberg spluttered: “I shall never again cross that threshold.” But after
a week or two Mahler said: “By the way, what’s become of those two?” I
did not, of course, say: “But you told me not to ask them again,” but lost
116 Chapter Six
no time in sending them an invitation; and they, who had only been
waiting for it, lost no time in coming.
Schoenberg’s ongoing financial dependence on the older composer

(who was at that time gainfully employed as Music Director at the Vienna
Court Opera) is portrayed in heart-rending detail by the following excerpt
from a letter to Mahler dated August 2, 1910.2
I am without money and I have the rent to pay. It was, I know, very
shortsighted of me to take a larger flat when I was earning less. But there
are many excuses—the disappointment of hopes so near fulfillment that
any one might have counted on them, not to mention me. I am compelled
therefore to beg of you the loan of 300-400 crowns. I shall certainly be able
to repay them next year when I am at the Conservatoire.
I cannot tell you how unhappy it makes me to have to cloud my relations

with you in this way.
A letter to Mahler the following day, August 3, indicates a favorable

result: “I have today received 800 crowns in your name from Miethke.”3
And Mahler continued to consider himself responsible for Schoenberg’s
welfare up until his death, as Alma reports: “During his last days and
while his mind was still unclouded, his thoughts often went anxiously to
Schoenberg. ‘If I go, he will have nobody left.’ I promised him to do
everything in my power.”4
Finally, in several letters from Schoenberg to Mahler, often written
right after a performance of one of Mahler’s symphonies that Schoenberg
attended, the younger composer enthusiastically explains his veneration
for the elder in terms that anticipate the language he would use later on to
describe his own notion of coherence and meaning in music. Schoenberg
typically speaks of Mahler’s direct expression of the “truth,” or of the
inner depths of Mahler’s soul (which for Schoenberg seems to amount to
the same thing), completely unobscured by considerations of musical style.
An even better example of this kind of language can be found in
Schoenberg’s essay written as a tribute and rejoinder to Mahler’s critics
after the latter’s death.5
In reality, there is only one greatest goal towards which the artist strives: to
express himself. If that succeeds, then the artist has achieved the greatest
possible success; next to that, everything else is unimportant, for everything
else is included in it: death, resurrection, Faust, fate—but also the lesser
and yet not less important moments, the emotions of the soul and spirit
which make a man creative. Mahler, too, tried only to express himself….
He expressed only that which, independent of style and flourish, portrays
Mahler’s Musical Idea 117
himself and himself alone, and which therefore must remain inaccessible to
anyone else who tried to achieve it merely by imitating the style. But this
style itself seems, in an enigmatic and heretofore unfamiliar way, to
exclude imitation. Perhaps this is because here, for the first time, a mode of
expression is so inseparably bound up with the subject to which it applies
that what usually appears as a symptom of the outward form is here,
simultaneously, material and construction as well….
The genius lights the way [to the future], and we strive to follow. Do we
really strive enough? Are we not bound too much to the present?
We shall follow, for we must. Whether we want to or not. It draws us

upward.
We must follow.
This, it seems to me, is what Gustav Mahler’s work, like the work of every
great man, was allowed to tell us.
Considering the close personal, artistic, and philosophical relationship

between the two men, it is only natural that music scholars would
investigate the influence of Mahler’s music and ideas on Schoenberg’s
music and ideas, in an attempt to grasp better what remains hard to
interpret in the music of both composers. And, in fact, a great deal has
been written (primarily in German) on connections between the two
composers’ music and thought.6 But hardly any of these works explores
whether Mahler’s music can be explained in ways that Schoenberg would
later associate with the concepts of “Grundgestalt” and “musikalische
Gedanke”—according to which conflicts or oppositions inherent in or
around the work’s first phrase undergo various elaborations and
extensions through the course of the work, leading to further and deeper
conflicts, all of which are resolved at or near the end (in a tonal work, by
showing clearly the relationship of the opposing elements to the home
key).7 Furthermore, the literature by Patricia Carpenter and her students
devoted to illustrating musical idea as an analytic tool tends to deal with
more traditional nineteenth-century German composers such as
Beethoven, Schubert, or Brahms, or with Schoenberg’s own tonal music.8
This seems curious, because Mahler himself used to speak of the “musical
idea” (Idee, not Gedanke) that his music represented. In a letter of
February 1897 to Arthur Seidl about the Second Symphony, he admits that
he hadn’t yet found a way to express his “idea” sheerly musically, like
Beethoven in his Ninth Symphony: “Whenever I plan a large musical
structure, I always come to a point where I have to resort to the ‘word’ as a
vehicle for my musical idea.”9
118 Chapter Six
To be sure, Mahler’s conception of “idea” as expressed in his letter is

much more vague in outline than Schoenberg’s—we do not know to what
extent it involved conflict, elaboration and resolution. Still, it seems
worthwhile to attempt an analysis of one of his symphonic movements in
Schoenbergian terms to see what it might tell us about how certain unusual
features of its tonal organization fit into a coherent dramatic narrative. I
have chosen the first movement, the Adagio, of the Tenth Symphony, a
work that seems (to me) to express clearly and exquisitely the notion of
resolution after conflict, especially at its final cadence.
The method I will use is a hybrid of Schenkerian and Schoenbergian
techniques that I introduced in a Music Theory Online article in 1999:
essentially, it consists of using the Schenkerian analysis to define
segments in the music that I then identify as opposing, elaborating, or
resolving within a Schoenbergian master process.10 Such a hybrid will
describe the piece more exhaustively than a typical Schoenbergian
analysis (in other words, there will not be any segments whose connection
to the structure as specific pitches is not made explicit). On the other hand,
the Schoenbergian component will enable this hybrid to characterize the
foreign quality of certain elements in a way that Schenkerian analysis
alone cannot, while giving me increased power to describe relations across
the surface of the piece. The combination of an analytic method that
describes a synchronic hierarchy of diminutions with a method that traces
the diachronic development of the Grundgestalt’s tonal contexts will
reflect more accurately, I hope, the listener's perception of coherence both
“back into” and “across” the music.11
* * *
The Grundgestalt of the Adagio is mm. 1-17, the viola introduction,

analyzed in Example 6-1. It presents us with two oppositions to the main
F# major tonality, a G triad and an A# triad. Example 6-1a uses part of a
Schoenbergian “chart of regions” to illustrate how foreign these chords are
to F#. Neither is within the group of “direct and close” keys, designated by
the central cross. G major, according to Schoenberg, is “indirect and
remote”: it can only be reached through the minor subdominant, B minor,
as its submediant triad. A# major is “indirect but close,” because it can be
attained by changing the mode of F#'s mediant triad.12 The problem of
how these foreign, conflicting chords fit into the whole is what is worked
out through the Adagio. Let us focus first on how the piece works out the
opposition between G and F#—by elaborating it and then by explaining
G's role in the home key. Further on, the opposition between A# and F#
will be documented similarly.
Example 6-1, Mahler, Tenth Symphony, Adagio, mm. 1-17
A# a# C# c# E e
D# d# F# f# A a
G# g# B b D d
G
Example 6-1a, chart of regions in F# major (major keys are represented by capital
letters, minor by small)
One principal way in which Mahler elaborates the opposition between

G and F# is using the dominant seventh chord of G as the culminating
chord of progressions that move away from F# into remote areas. There
are several examples in the Adagio; one of the more dramatic ones is mm.
178-83, illustrated by Example 6-2. Here we have a middleground 3-line,
the first two notes of which are counterpointed with an ascending fifth-
progression. But instead of a tonic triad, Mahler harmonizes the first scale
degree with the altered dominant seventh of G, an intensified deceptive
cadence. The D7/b5 chord has a disruptive influence in this context;
through it, Mahler is making the F#-G opposition “penetrate to the
profoundest depths,” as Schoenberg would put it.13
Some earlier passages in which the D7 chord had had a similar
deceptive, disruptive function are mm. 162-67 and mm. 140-41. In the
first, a prolonged dominant seventh (in F#) is followed by a sforzando D7
on the last half of 167 before resolving to a I in m. 168. In the second,
which occurs at the juncture between the “elaboration” and “restatement”
sections according to Kofi Agawu,14 the V7 (in F#) ending the
“elaboration” goes to a D7 before progressing to the I that supports the
120 Chapter Six
“restatement” of m. 16’s first theme. In this latter instance, the deceptive

D7 is disrupting the characteristic dominant-tonic progression marking the
joint between sections equivalent to development and recapitulation in a
variant of sonata form.
Which reduces to:
Mahler begins to suggest solutions to the F#-G problem as early as m.

141, so we do not have a simple chronological sequence of problem
followed by solution in this piece. It may be significant that the solutions
begin to appear at the onset of Agawu’s “restatement” section (the
equivalent of a recapitulation in a standard sonata form), but it must also
be remembered that the dramatic elaboration of the problem at mm. 178-
83 is also part of the “restatement.” Perhaps all we can say is that
elaborations of the problem and solutions to it seem to alternate in the
“restatement” section. One solution, in mm. 141-42, is illustrated by
Example 6-3. Here, Mahler explains the G triad as a neighbor to the F#
tonic, its simplest role in the key of F# major. Not long after, in mm. 143-
53, illustrated by Example 6-4, he shows us that G can serve as an altered
supertonic in a circular progression in the home key. The natural II chord
in m. 145 progresses to its primary triad, IV, which in turn acts as a
dominant preparation. (An alternative view of this passage, which would
follow more closely Schoenberg’s comments in Structural Functions of
Harmony about the natural II chord, would claim that G major progresses
to B major as altered submediant to tonic, then B in turn progresses as
subdominant to C# and eventually F#. Either way, the passage explains
G's role.)

122 Chapter Six
Since Mahler has used the dominant seventh of G repeatedly as a

disruption within the F# major context, it is necessary for him to explain
its role as well within that key. An explanation occurs at mm. 221-25, part
of the “closure” section (again, Agawu's term). Here, a sustained D7
chord, which had been introduced as natural VI, becomes a German sixth
in F# and leads in the customary way to I 6/4. See Example 6-5. The fact
that Mahler sustains and arpeggiates this explanatory D7 makes it easier to
hear a connection between it and the disruptive D7 back at m. 183; the
earlier chord had been sustained through arpeggiation in a similar way.
The latter D7 seems to solve the problem that the earlier one had posed.
One further passage conclusively suggests a role for G within the F#

major context—mm. 204-08, the onset of the background dominant triad.
See Example 6-6. Here G natural completes a large group of stacked thirds
above the V chord, the first four of which are simultaneous and the last
three successive.
Mahler elaborates the problem of A# (how can it relate to F# major?)
more thoroughly and methodically than the G problem. Both Steven Bruns
and Richard Kaplan have provided detailed accounts of the appearances of
A#/Bb as chord and key area in this movement, so it is necessary only to
cite a few examples of problems and solutions pertaining to A# here.15
In mm. 16-18, an area of relative F# major stability, there is an unusual
emphasis in the first violin melody on the third and fifth of the tonic, A#
and C#. See Example 6-7. These, of course, are the two notes F# has in
common with A# minor, and their emphasis can be interpreted as an initial
suggestion of the dual role they will be asked to play in the Adagio, as
scale degrees 3 and 5 of F# and as 1 and 3 of A#.
A# and C#’s dual role becomes more overt when Mahler juxtaposes F#
and A#/Bb triads, as he does frequently in the Adagio. The earliest
occurrence is at m. 26, where iii follows I6 with A# and C# held as
common tones (see Example 6-8). Measures 71 and 141 contain similar
progressions.16
124 Chapter Six
At two places in the movement, A# minor becomes prominent enough

to warrant its own key signature (enharmonically spelled as Bb minor).
These are mm. 91-105 and 172-77. Examples 6-9a and b provide graphs
of both sections. In mm. 91-94, the Bb minor tonic is prolonged by
neighbor chords which include the F# minor seventh chord, producing still
another juxtaposition of F# and A#/Bb harmonies, but this time with Bb as
primary.17 In mm. 94-98, a middleground 3-line establishes Bb minor and
major, though it is obscured somewhat by a sonority in m. 95 which
superimposes an Ab major triad over the root of the Bb minor tonic. The
progression becomes clearer after m. 96. Measures 98-99 balance the
descending line with an ascending one, still establishing Bb major.
Finally, mm. 100-02 bring further neighboring progressions, followed by a
move to a Gb 6/5 chord in m. 102, another reference to the key area Bb
minor/major is trying to supplant.
Example 6-9a, Mahler, Tenth Symphony, Adagio, mm. 91-107

126 Chapter Six
Example 6-9b, Mahler, Tenth Symphony, Adagio, mm. 172-78
The bass support for prolongations of Bb minor in mm. 172-77 is much

weaker than in the previous passage. Nevertheless, one can hear scale
degrees 3, 2, and 1 in m. 172, counterpointed in parallel sixths, followed
by 5, 4, and 3 in m. 173, counterpointed in parallel thirds. The first of
these lines has further significance in that its scale degree 1 is harmonized
by a Gb triad, another reference to the piece's primary key in a Bb context.
By this point in the piece, the listener sensitive to specific pitch classes
should have the impression that Bb/A# and Gb/F# are vying for
supremacy, since chords based on the two roots have been alternating
within both keys. These passages (mm. 91-105 and 172-77) represent the
furthest extent to which A#/Bb challenges F# as primary key, the ultimate
elaboration of the A# problem.
At the same time, mm. 91-105 and 172-77 provide the solution to the
A# problem—both sections clearly demonstrate how A# fits into an F#
context, at their ends. Measures 104-05 summarize the first passage with
scale degrees 3 - 1 in Bb minor; but this key-defining third is soon
followed by Gb, and the subsequent music (mm. 105-07) clearly indicates
that Db and Bb should be heard as scale degrees 5 and 3 in Gb major, F#’s
enharmonic equivalent.
The end of the second passage indicates even more conclusively that
Bb belongs to Gb/F# major as its scale degree 3. Measures 177-78 offer a
middleground 3-line in Bb minor, the first two pitches of which are
harmonized by iv7 and viio7 (the iv7 is obscured by a dissonant bass note,
but can be recognized as holding over from the preceding measures). But
the scale degree 1, rather than bringing I in Bb minor, is spelled
enharmonically as A# and harmonized by I in F# major. A key-defining
middleground progression in Bb leads to a scale degree 1 that is
reinterpreted as 3 in F#: this is a clear, conclusive answer to the A#/Bb
question. Since both mm. 91-105 and 172-77 present A# minor, obtained
through a mode change from the original A# problem in the Grundgestalt,
and then treat the tonic of A# minor as third scale degree in the home key,
Mahler's solution to the A# problem is essentially the same as the answer
Schoenberg gives for major III in Structural Functions of Harmony (refer
again to my pp. 118-19).
Associated with the elaboration and solution of the A# problem is a
motivic subplot, involving the search for a missing, key-defining tonic
note, that Mahler resolves only in the final measures of the piece. For me,
this subplot is the most salient expression of the musical idea in Mahler's
Adagio, and its solution accounts, I believe, for the overwhelming sense of
resolution I sense in the final measures. We have already discussed the
motivic subplot’s first stage: it begins in m. 16, where Mahler emphasizes
A# and C# within an F# major prolongation. (See Example 6-7 again.) It
is important to notice that A# and C# appear in the first violin without F#;
this omission, as well as A# and C#’s emphasis, helps us to hear these
pitches as ambiguous with respect to tonal context. When m. 16’s theme
recurs elsewhere in F# major, the opening melodic gesture invariably
emphasizes A# and C# and leaves out F#. See mm. 49, 58, 69 (second
violin/viola), 141 (first and second violins), 178 (Ex. 6-9b graphs this
measure), and 213. By the time we hear the bassoon's statements of the
motive in mm. 235-37, there is an almost painful sense of incompleteness.
See Example 6-10. A# and C# long for a third pitch to make their tonal
128 Chapter Six
context definite; the active tone B is not it, and Mahler repeats the motive
several times, as if groping for the right completion. A subsequent passage
having a similar effect is mm. 253-55, where A# and C# lead to Gx (A
natural), part of a linear half-diminished seventh chord.
The true completion for A# and C#, which fixes them securely as scale
degrees 3 and 5 in F# major, comes only at mm. 272-73, the final cadence.
See Example 6-11. The first flute and second violin/viola combine to form
A#-C#-F#. It is difficult not to hear these three notes as a solution to the
problem elaborated earlier in the movement—whatever uncertainty we

had had about the melodic continuation of A#-C# or their proper tonal
context is dispelled here.
This chapter has analyzed only a few segments of the Adagio.
However, there are other elaborations of and solutions to the F#-G and F#-
A# problems in this movement, enough for me to claim that something
parallel to Schoenberg's musikalische Gedanke, operating on segments
isolated and brought to the listener's attention by a Schenkerian analysis,
contributes to structuring the movement and accounts for some of the
harder-to-explain tonal progressions within it. And it seems to me that
characterizing aspects of the Adagio as problem and solution best accounts
for the overwhelming resolution I sense at the end of it. But analyzing the
piece this way also suggests a way that Mahler influenced Schoenberg’s
ideas about musical coherence. Though Schoenberg’s “musical idea” has
many roots—in rhetorical theory of the eighteenth century, German
idealist philosophy, and music theorists such as Adolph Bernhard Marx—
it may well be that one of the strongest influences on the younger
composer’s thought was the direct, unclouded musical communication of
the “idea” by his benefactor and friend.18 In that way, Mahler truly did
light the way to the future, just as Schoenberg claimed.
Notes
1
Alma Mahler, Gustav Mahler: Memories and Letters, 3rd ed., ed. Donald
Mitchell, trans. Basil Creighton (London: John Murray, 1973), p. 78.
2
Ibid., p. 341.
3
Ibid., p. 341.
4
Ibid., pp. 197-98.
5
Arnold Schoenberg, “Gustav Mahler” (1912, 1948), in Style and Idea: Selected
Writings of Arnold Schoenberg, 2nd rev. ed., ed. Leonard Stein, trans. Leo Black
(Berkeley and Los Angeles: University of California Press, 1984), pp. 454 and
471.
6
There is one source in English that covers the personal history between the two
composers in much more detail than I have done here, and also discusses
Schoenberg’s efforts to preserve Mahler’s memory and promote interest in
Mahler’s music after the older composer’s death. Then it goes on to associate the
formation of Schoenberg’s notion of the sheerly musical idea with the responses
both composers made to Schopenhauer’s, Hanslick’s, Wagner’s and Liszt’s
proclamations about meaning in absolute music, “music drama,” and program
music. This is Julia Bess Hubbert, “Mahler and Schoenberg: Levels of Influence”
(Ph.D. dissertation, Yale University, 1996). When Hubbert begins to discuss
musical connections between the two in a later chapter, however, she does not
focus on a process of conflict, elaboration and resolution (as I am about to do).
Instead, she asserts that Mahler after the Fifth Symphony wrote in a polyphonic
130 Chapter Six
style characterized by an increased independence of individual lines, which sets

the stage for Schoenberg’s more dissonant polyphony in the First Chamber
Symphony, and eventually his introduction of twelve-tone music.
7
An interesting exception to the general trend is an article by Michael Mathis on
the “Urlicht” movement of Mahler’s Second Symphony. Mathis defines the
Grundgestalt of the movement as Ab/G# “elaborated by corresponding modal
inflections of the third and sixth degrees” in the keys of Db major and minor (p.
247), and then shows how the projection of these elements into the work’s
middleground gives it coherence, in the absence of a traditional dominant
supporting the second scale degree. See Mathis, “Arnold Schoenberg’s
Grundgestalt and Gustav Mahler’s Urlicht,” International Journal of Musicology 5
(1996): 239-60.
8
A few among many examples: Patricia Carpenter, “Grundgestalt as Tonal
Function,” Music Theory Spectrum 5 (1983): 15-38, which uses the first movement
of Beethoven’s op. 57 Piano Sonata; idem, “Schoenberg’s Tonal Body,” Theory
and Practice 30 (2005): 35-68, which uses Brahms’s Piano Quartet op. 60;
Severine Neff, “Aspects of Grundgestalt in Schoenberg’s First String Quartet, op.
7,” Theory and Practice 9 (1984): 7-56; and Murray Dineen, “Tonal Problem,
Carpenter Narrative, and Carpenter Motive in Schubert’s Impromptu, op. 90, no.
3,” Theory and Practice 30 (2005): 97-120.
9
Alma Mahler, Selected Letters of Gustav Mahler, ed. Knud Martner, trans. Eithne
Wilkins, Ernst Kaiser, and Bill Hopkins (New York: Farrar, Straus, Giroux, 1979),
p. 212.
10
Jack Boss, “’Schenkerian-Schoenbergian Analysis’ and Hidden Repetition in
the Opening Movement of Beethoven's Piano Sonata Op. 10, No. 1,” Music Theory
Online 5/1 (January 1999).
11
The terms synchronic and diachronic, borrowed from linguistics, have been used
by a number of writers on music for various purposes. Allan Keiler uses them to
describe how music and music history unfold organically both ways in Schenker’s
thinking: Keiler, “On the Origins of Schenker's Thought: How Man is Musical,”
Journal of Music Theory 33/2 (Fall 1989): 273-98. For Leonard Meyer, they
represent two perspectives on motivic relations, the second of which (diachronic
relations) has been underexplored. See Meyer, “A Pride of Prejudices; Or, Delight
in Diversity,” Music Theory Spectrum 13/2 (Fall 1991): 241-51.
12
See Schoenberg, Structural Functions of Harmony, rev. ed., ed. Leonard Stein
(New York: Norton, 1969), pp. 19-21 for an introduction to the chart of regions.
Pages 68-75 present Schoenberg’s classification of key relationships to the tonic.
13
This quotation comes from the chapter on “Melody and Theme” in Schoenberg’s
Fundamentals of Musical Composition, ed. Gerald Strang and Leonard Stein
(London: Faber and Faber, 1967), p. 102: “The unrest in a melody need not reach
below the surface, while the problem of a theme may penetrate to the profoundest
depths.”
14
V. Kofi Agawu, “Tonal Strategy in the First Movement of Mahler's Tenth
Symphony,” 19th-Century Music 9/3 (Spring 1986): 224.
15
Steven M. Bruns, “Mahler's Motivically Expanded Tonality: An Analytical
Study of the Adagio of the Tenth Symphony” (Ph.D. dissertation, University of
Wisconsin, 1989), pp. 198-226; Richard Kaplan, “The Interaction of Diatonic
Collections in the Adagio of Mahler's Tenth Symphony,” In Theory Only 6/1

(November 1981-82): 29-39.
16
Bruns discusses the suggestions and statements of A# covered in the last two
paragraphs in greater detail. See Bruns, “Mahler's Motivically Expanded
Tonality,” pp. 200-13.
17
Bruns also comments on the alternation of Bb and F# triads in this section. See
pp. 213-18.
18
I describe in detail these influences on the formation of Schoenberg’s “musical
idea,” and others, in the first chapter of my forthcoming book, Symmetry and the
“Musical Idea” in Schoenberg’s Twelve-Tone Music.
CHAPTER SEVEN
DEBUSSY’S UNTERBRECHUNG
GARY DON
The earnest debate about whether Debussy’s music is essentially tonal

or post-tonal has produced an impressive number of diverse and thought-
provoking books and articles in the decades since his death. The
contemporaneous view of Debussy’s emerging compositional style tended
to focus on his radical departures from the traditional tonal framework,
often from a critical point of view, including Saint-Saëns’s sour comment
that Debussy “cultivated the absence of style as well as the absence of
logic and common sense.”1 Debussy’s conversations with Ernest Guiraud,
well known for the “pleasure is the law” quote, also yield his enigmatic
comment “Il faut noyer le ton.” Does noyer mean “to drown” or “to blur”
in this instance? Does le ton refer to the individual tone or to the overall
tonality? Subsequent comments suggest the latter meaning of both words.
When Guiraud played a French augmented sixth chord and said “But
when I play this it has to resolve,” Debussy replied, “I don’t see that it
should. Why?” He added,
Music is neither major nor minor. Minor thirds and major thirds should be
combined, modulation thus becoming more flexible. The mode is that
which one happens to choose at the moment. It is inconstant.2
Debussy’s blurred tonality raises issues common to those confronted in

the works of post-tonal composers, including a perceived lack of
monotonality or primary key, a lack of large-scale linear progressions, and
a related lack of prolongation. Joseph Straus and Steve Larson debate the
question of whether clearly-defined concepts of consonance and stability
must precede the application of prolongation in an analysis in order for it
to be meaningful, or whether the reverse is true.3 Straus concludes with a
preference for
Debussy’s Unterbrechung 133
a more balanced view that sees consonance or stability and prolongation

acting in reciprocal, mutually reinforcing ways… This notion of mutual
reinforcement poses a theoretical conundrum—you can’t tell what is
consonant until you know what is being prolonged, but you can’t tell what
is prolonged until you know what is consonant—but one with little
practical consequence. You simply pull yourself up by your bootstraps,
moving flexibly back and forth between the relevant categories.4
Carl Schachter reaches a similar conclusion:
[I]f one needs to understand the background to make sense of the

foreground, one also needs to understand the foreground to make sense of
the background—a seemingly hopeless impasse. Actually, it’s a heuristic
problem that confronts people all the time and in areas far removed from
musical analysis: one can grasp neither the part without the whole nor the
whole without the part. But one copes, somehow.5
These issues come into play with every attempt to apply Schenkerian
analysis to the music of Debussy. Adele Katz argues eloquently that
Debussy’s style “demonstrates that the art of prolongation is not static,
confining the composer to a prescribed set of rules. On the contrary, its
elasticity is its greatest artistic asset so long as the effects it achieves are
not gained at the expense of tonal stability.”6 However, it becomes clear
that each analysis must define “tonal stability” for that particular piece,
even for early and seemingly unproblematic works. Katz concedes that “it
is probable that a new system of analysis is needed to understand the new
concepts defined by the whole-tone, polytonal, and twelve-tone systems
and the new and different techniques they disclose.”7 Richard Parks
concurs. “The absence of Urlinie descent has a parallel in the absence of
structural harmonic fifth-relationships,” he writes. “After the early works
[of Debussy] especially, one rarely finds dominant-tonic closure in a
structural sense… [L]inear progressions tend to be of the simplest and
only of local significance.”8
The concept of prolongation, specifically dominant prolongation, is at
the heart of Schenker’s concept of interruption (Unterbrechung). He
identifies a distinction between the term “half cadence” and
the prolongational significance of this dominant at the first level… The

interruption not only creates more content; it also has the effect of a delay,
or retardation, on the way to the ultimate goal, scale degree 1 over I. The
interruption is able to produce this effect only because it carries within it
the fundamental structure, which must achieve its fulfillment despite all
detours.9
134 Chapter Seven
He proposes the term “dividing dominant” or simply “divider” to clarify

the distinction. The dividing dominant is prolonged through the
interruption, despite the fact that it is followed by a tonic chord, because
scale degree 2 moves to scale degree 3 instead of scale degree 1 at this
point. Schenker avoids the term “imperfect authentic cadence,” asserting
that the movement to scale degree 3 is “not a cadence.”10 The fundamental
structure is not complete until the dominant resolves to the tonic and scale
degree 2 resolves down by step to scale degree 1, forming a perfect
authentic cadence. Stability, rather than consonance, is the driving force in
the case of interruption, since all of the scale degrees of the interrupted
Ursatz are consonant in relation to their harmonizing chords, but scale
degree 1 is more stable than scale degrees 2 or 3.
Identifying an Ursatz in Debussy’s compositions—interrupted or
not—can be problematic. Nevertheless, a number of authors in recent
publications have revitalized Katz’s notion of the “elasticity” of
prolongation, and concluded that they can stretch it to encompass a sizable
number of Debussy’s compositions previously viewed as resistant to this
approach. Boyd Pomeroy writes, “While many aspects of Debussy’s tonal
practice do indeed represent a radical break with tradition, the larger-scale
musical architecture is nevertheless often underlain by creative
transformations of traditional tonal paradigms.”11 Matthew Brown
identifies an Ursatz that encompasses the entirety of L’Isle joyeuse, and
prolongational middleground layers (Example 7-1). Brown notes the
similarities between the scale degree 2-lowered scale degree 2-scale
degree 1 progression in his Ursatz and Schenker’s analysis of Chopin’s
Mazurka in E minor, Op. 41, No. 2, “where the chromatic alteration F# to
F creates the impression of the Phrygian mode.”12
Brown’s Ursatz and middleground layers for his analysis of L’Isle
Joyeuse do in fact bear a strong resemblance to Schenkerian tonal
prototypes, and they demonstrate a consistent set of transformations from
one layer to the next. One issue that arises in Brown’s analysis pertains not
to the prototypes themselves, but to the relationship between the
prototypes and the tonality that governs them. Brown refers to L’Isle
joyeuse as a “monotonal composition” that “is firmly rooted in the key of
A major.”13 This assertion brings to mind Debussy’s comment that he “has
no faith in the supremacy of the C major scale,” and that “[t]he tonal scale
must be enriched by other scales.”14 “Enrichment” can have different
shades of meaning, and analysts must make crucial decisions: are these
“other scales” ornamental, or are they of equal importance? For example,
there is a world of difference between analyzing the D# as a melodic
lower neighbor and as a chord tone of V of V, as Brown does in levels b
and c of his graph (Example 7-1), and analyzing the D# and the A-D#
tritone as integral parts of the A Lydian and the A whole-tone scales.
Example 7-1, Matthew Brown’s graph of L’Isle joyeuse
Kip Wile assigns equal weight to the other scales in his analysis of
three Preludes. For example, his analysis of the descending voice leading
in Des pas sur la neige takes into account the interactions between modes
and major, minor, octatonic, and whole-tone scales.15 Wile posits self-
defined middleground levels, based on three categories of recurrence:
recurrent linear progressions, recurrent objects, and recurrent centricity.
He notes that his recurrent linear progression “bears comparison to
Schenkerian usage, yet does not correspond in every respect,” and that
136 Chapter Seven
[t]he term “middleground” is used [in Wile’s analysis] not in the strict
Schenkerian sense of an underlying structure that relates to contrasting
levels through traditional contrapuntal embellishments. Rather, it is used
analogously to that meaning in that: 1) it describes medium-range voice-
leading events; 2) the events produce structure that underlies and joins to
the compositional surface; and 3) the events elaborate still deeper
organization.16
Thus, Wile’s “middleground” is not prolongational, according to Larson’s

assertion that “prolongation is embellishment.”17 Rather, Wile separates
recurrence and prolongation, noting that Debussy’s music demonstrates
an expanded conception of recurrence, representing “solutions that reflect
the musical issues of his time.”18
Boyd Pomeroy’s analysis of Gigues from Images combines
Schenkerian voice-leading graphs with “directional tonality,” a term
applied to pieces that begin in one key and end in another, the tonic keys
usually related by major or minor third.19 Examples of directional tonality
from the works of Chopin include the Scherzo No. 2, Op. 31 (Bb minor to
Db major); Sonata No. 2, Op. 35, second movement (Eb minor to Gb
major); Ballade No. 2, Op. 38 (F major to A minor); and the Fantasy, Op.
49 (F minor to Ab major). In the case of Gigues, Pomeroy interprets C5 as
a prolonged scale degree 5 in F minor, then as a prolonged scale degree 3
when the tonal center shifts to Ab major (Example 7-2). The prolonged
notes simply change scale-degree function when the key changes.20
Example 7-2, Boyd Pomeroy’s graph of Gigues from Images, mm. 1-105
It is clear that Brown, Wile, and Pomeroy focus on the connections

between Debussy’s compositional practices and those of his predecessors,
Chopin and Wagner in particular, by applying Schenkerian principles and
the concept of directional tonality to the analysis of Debussy’s music, and
that there is a delicate balancing act between perils and rewards inherent in
doing this kind of analysis. Monotonal analyses are consistent with
Schenkerian practice, but they run the risk of reducing important features
to the level of subordinate details. Using analogies to Schenkerian
principles in an analysis or applying them in multi-tonal contexts runs the
risk of reducing their explanatory power even while expanding their range
of application. The rewards include a deeper level of insight into the ways
that traditional principles of recurrence and large-scale voice leading
continue to manifest themselves in different contexts.
This article undertakes a similar perilous journey through the study of
interruption structures, directional tonality, and tonic complexes in the
Sarabande from Pour le piano and Reflets dans l’eau from Images Book
I, and the “veiling” and “submersion” of these structures in ways similar
to those observed by John Crotty in his analysis of Prélude à l’après-midi
d’un faune. 21 Crotty constructs a musical analogy to Symbolist poetry and
prose that “depends on the displacement of a tonal object with something
that can sufficiently veil the identity of that object without either
destroying its identity or disrupting the tonal syntax.” 22 In the case of the
Sarabande and Reflets dans l’eau, quartal and extended tertiary chords
veil the tonal objects and the syntax, including the interruption structures,
in ways that not only do not disrupt or destroy them, but enhance their
expressive power.
There are two versions of the Sarabande, one in an autograph
manuscript dated 1894 and titled Images and the other in the autograph
manuscript for Pour le piano, dated 1901. Both versions incorporate the
spirit and the structure of the traditional sarabande through the use of a
slow triple meter and second-beat accents. Example 7-3 is the complete
score of the Pour le piano version, with the major motives and their
alterations labeled. Both versions feature a wedge-shaped progression
(that is, an ascending upper line paired with a descending lower line) in
beats 2 and 3 of m. 6, followed by a repetition and extension of this
pattern to an implied dominant in the key of E major by means of a
continued ascent to D#5 in the upper line and a continued stepwise descent
to B1 in the lower line in mm. 7 and 8 (Motive C in Example 7-3). The
middle voices alternate between A major and G# minor triads for beats 2
and 3 of m. 6, then repeat and extend this pattern through m. 7 before
continuing up to C# minor and B major triads at the end of m. 7 and the
beginning of m. 8. These triads, combined with the contrary motion in the
138 Chapter Seven
outer voices, produce an added-fourth chord and seventh chords that veil
the motion to the dominant in m. 8 without disrupting it, due to the strong,
directed stepwise motion to scale degree 5 in the bass line. This implied
dominant does not resolve to I in E major: rather, the D#5 leading tone
resolves to E5 in m. 9, but this E is the fifth of an A major triad (IV in E
major), followed by an implied dominant in G# minor in m. 14.
Example 7-3, Sarabande from Pour le piano, complete score with primary motives

140 Chapter Seven
Measures 6-8 contribute more than localized coherence to the

Sarabande. They are part of a large-scale interruption, in which the
material of mm. 1-4 (Motive A) is repeated exactly beginning in m. 15, the
top line of mm. 6-8 (Motive C) reappears an octave higher in mm. 47-49,
and the dominant of G# minor in m. 14 reappears in m. 59 (Examples 7-3
and 7-4). After the large-scale interruption at m. 14, Debussy places the
major motives of the Sarabande in direct succession, albeit with octave
transfers and altered melodies and harmonies, from m. 42 through the end
of the piece. For example, the top line of Motive A in m. 1 is transposed
down an octave in m. 42, and the top line of Motive D in m. 31 is
transposed up an octave in m. 56. The D#6 in m. 49 does not resolve in the
same register to E as did the D#5 of m. 8, however: it reappears as the
highest pitch in m. 59, then it resolves as scale degree 2 to the tonic C#6 in
the final chord. Example 7-4 lists the harmonic functions of the chords at
the important structural points. It demonstrates that there are three tonic
keys prominent at different times: G# minor, E major, and C# minor.
Thus, Debussy’s interruption occurs at the compositional surface and at an
abstract middleground level that is prolonged by foreground
embellishments and articulated by “dividing dominants.”

142 Chapter Seven

Example 7-4, Graph of Sarabande from Pour le piano, mm. 1-14

144 Chapter Seven
Example 7-4 continued, mm. 15-30

Example 7-4 continued, mm. 31-41

146 Chapter Seven
Example 7-4, continued, mm. 42-55

Example 7-4 continued, mm. 56-72 (end)
The recurrence of Motive C, the interruption motive, at mm. 47-49

requires a more detailed analysis. Although the top line of mm. 47-49
consists of mm. 6-8 transposed up an octave, the bass line has been altered
to create a continuous descending line from mm. 47-48, as opposed to the
repetition-extension of the bass line in mm. 6 and 7 (see Example 7-3).
There are also different harmonic implications, with a tonic chord in C#
minor in place of the tonic seventh chord in E major (compare the last
eighth-note chord of m. 6 with the last beat of m. 47). The altered bass line
of mm. 47 and 48 requires a leap in order to arrive on B1 at the downbeat
of m. 49, as opposed to the stepwise motion from m. 7 to m. 8. However,
the harmonic implication, dominant in the key of E major, is the same for
mm. 8 and 49.
148 Chapter Seven
Katz analyzes the Sarabande as a monotonal C# minor composition

with no interruptions. She concedes that the dominant of E major in m. 8
“would seem to be of structural importance,” but she concludes that “its
actual function is that of a passing chord between the C# minor and A
major chords,” and that “[i]t is part of a prolonging motion that expands
the space between the tonic and supertonic chords” in C# minor. Her
graph is reproduced as Example 7-5.23 Although Katz is not concerned
with the lack of a major dominant (a G# major chord), noting Debussy’s
penchant for modal (minor) dominants in the minor mode,24 her analysis is
in fact problematic, in view of Schenker’s analyses of minor-mode
compositions that clearly indicate that the final scale degree 2 of the
fundamental line is harmonized by a major dominant.25 The lack of scale
degree 3 (E) harmonized by a tonic C# minor chord at a structural level is
also problematic.26
Although Katz’s monotonal analysis of the Sarabande is viable,
conceding the issues noted above, and although it is also viable to analyze
the piece as an example of directional tonality, beginning in E major and
ending in the relative minor, C#, the piece is more complex and subtle
than that. The D#s of the dominant chords in E major in mm. 8 and 49 are
of structural importance, due to the fact that they are prolonged from m. 8
to m. 14 and m. 49 to m. 59, respectively, changing function from scale
degree seven to scale degree five in G# minor at these points. The piece
moves freely between the three third-related tonal centers with apparent
ease, bringing to mind Debussy’s reference to “flexible” modulations and
“inconstant” modes. It also brings to mind Robert Bailey’s concepts of
“tonal pairing” and the “double-tonic complex,” in which
[t]he two elements are linked together in such a way that either triad can
serve as the local representative of the tonic complex. Within that complex
itself, however, one of the two elements is at any moment in the primary
position while the other remains subordinate to it.27
Thus, the piece is not “bitonal,” in the sense of the simultaneous presence
of tonic keys of equal weight, but it is not “monotonal” either, since no
one key is in the primary position for the entire work.28
Example 7-5, Adele Katz’s graph of the Sarabande from Pour le piano.
From CHALLENGE TO MUSICAL TRADITION: A NEW CONCEPT OF
TONALITY by Adele T. Katz, ©1945 by Alfred A. Knopf, Inc., copyright
renewed 1972 by Adele T. Katz. Used by permission of Alfred A. Knopf, a
division of Random House, Inc. Any third party use of this material, outside of this
publication, is prohibited. Interested parties must apply directly to Random House,
Inc. for permission.
150 Chapter Seven
Example 7-6 reproduces Bailey’s complex of thirds, which goes

beyond the typical relative major-minor relationship, such as A minor and
C major, to incorporate parallel major and minor relationships and
chromatic mediant-related triads, such as A minor and C minor.29 He
represents ascending major thirds with diagonals to the right and
ascending minor thirds with diagonals to the left. Bailey applies this
concept in his analysis of Wagner’s Tristan and Isolde, stating that A
major and minor and C major and minor form an A/C complex in the
Prelude and Act I.
Example 7-6, Robert Bailey’s complex of thirds
In the Sarabande, the three tonal centers and the triads built from them
are incorporated into a progression of quartal chords (Motive E), first
introduced in m. 23, then extended in mm. 67-70 (Example 7-7). The roots
(that is, the bottom pitches of the stacks of fifths) ascend by alternating
minor and major thirds to the final tertian C# minor triad in m. 71 (B-D-
F#-A-C#), at the same time that the prolonged D#6 of m. 59 resolves
downward by step to C#6 in m. 71. D-natural, introduced in m. 42 as the
root of a D major triad when Motive A recurs, is incorporated into this
structure. The result is a vertical extension of Bailey’s complex, due to the
quartal chords, with ascending perfect fifths indicated by vertical lines.
Thus, the end of the Sarabande combines directional tonality, evident in
the music of Chopin, with a triple-tonic complex, an elaboration of the
double-tonic complex evident in the music of Wagner.
John Crotty has noted this same connection, applying Bailey’s concept
of the double-tonic complex to Prélude à l’après-midi d’un faune.30 He
identifies C# minor and E major as the double-tonic complex, two of the
three keys that form the triple-tonic complex of the Sarabande.31 Crotty
maintains Bailey’s distinction between a double-tonic complex and
polytonality:
However, an important point should be made: polytonality is not an issue

since the two triads are not set against one another. Structural tension
resides primarily in shifting tonal primacy from one triad to the other
within the tonic complex.32
Example 7-7, Sarabande from Pour le piano, complex of thirds
In his analysis of veiling in Prélude à l’après-midi d’un faune, Crotty

asserts that the “primary system is ‘diatonic’ tonality,” the “secondary
system is the whole-tone scale,” and “whole-tone tetrachords displace
either tetrachord of the diatonic E major scale to combine a diatonic fifth
with the whole-tone tritone in various scalar configurations.”33 Example
7-8 reproduces Crotty’s Examples 1a and b, which provide examples of
the displacement of both the first and the second tetrachords of the E
major scale with whole-tone tetrachords. At the end of the Sarabande, it is
152 Chapter Seven
the quartal chords that veil the identity of the three tonic triads by
displacing them, and it is the simultaneous completion of the chain of
ascending alternating major and minor thirds (B-D-F#-A-C#) and
resolution of D#6 to C#6 that clarifies the tonal syntax.
Example 7-8, John Crotty’s Examples 1a and 1b
Reflets dans l’eau is another example of a motivic interruption that

contributes to large-scale coherence and a veiled ending. There is only one
version, in the autograph manuscript for Images Book I, dated 1905. It
features a motivic interruption similar to that of the Sarabande in mm. 9-
11, as shown in Example 7-9. The interruption is framed by minor ninth
chords (F minor-Ab minor-Bb minor) that are the result of chromatic
ascent in all lines except for the top line. These ninth chords veil the
chromatic ascent in the bass line from scale degree 5 in Db major at the
beginning of m. 9 to scale degree 1 at the end of m. 10 without obscuring
it completely. A quartal chord, reminiscent of those of the Sarabande,
follows on the downbeat of m. 13, with Bb at the top of a five-note stack
of four ths.
Debussy’s dynamic markings emphasize the interruption by a
crescendo from pianissimo in m. 9, followed by a drop back to pianissimo,
followed by an extended crescendo in m. 10, followed by another drop
back to pianissimo. An examination of recordings by pianists Friedrich
Gulda, Arturo Michelangeli, and Vlado Perlemuter reveals that none of the
three follow Debussy’s markings. On the contrary, all of them insert a
diminuendo and a ritardando at the end of m. 10. Although the “Tempo
rubato” indication grants performers interpretive freedom, Debussy was
meticulous about the placement of expressive markings. In a letter to his
publisher regarding Images Book II, he urged the engraver to “respect the
exact placing of nuances—[they are] of extreme and pianistic
importance.”34 Debussy advised pianist Maurice Dumesnil “to observe
exact dynamics, to pedal according to one’s ears, to keep the textures
clear, and to play in a straightforward manner without romantic
affections” when performing Hommage à Rameau, Claire de lune, Reflets
dans l’eau, Children’s Corner, and Pour le piano.35 Based on these first-
hand accounts, it is likely that he wanted the dynamics of mm. 9 and 10 of
Reflets dans l’eau to be performed exactly as indicated.
Example 7-9, Reflets dans l’eau, mm. 7-15
Unlike both versions of the Sarabande, there is no recurrence of the

interruption motive in Reflets dans l’eau, and no double- or triple-tonic
complex. Although Db major emerges as the primary key, there are hints
of Db major/Bb minor bitonality throughout the piece. The form is more
traditional, with an ornamented repetition of mm. 1-8 as a recapitulation
and as a coda. The quartal chords at the end of the piece in mm. 90 and 91,
with Db at the bottom of the stacks of fifths (Example 7-10), veil the Db
major/Bb minor tonic chords in a manner reminiscent of the end of the
Sarabande. The Bb4 of mm. 91 and 92 ultimately resolves down by step
to Ab4 in the last tonic chord in m. 93 (scale degree 6-scale degree 5),
clarifying the tonal syntax. The resulting complex of thirds is broader than
that of the Sarabande, as a result of the minor ninth chords (Example 7-
11).
154 Chapter Seven
Example 7-10, Reflets dans l’eau, mm. 90-94 (end)
Example 7-11, Reflets dans l’eau, complex of thirds
Roy Howat has established that Debussy enhanced his Blüthner

boudoir grand piano “with the Aliquot system of added resonating strings”
in 1904 or 1905, at the same time that he was composing Reflets dans
l’eau.36 These strings were part of a frame that was laid over the strings of
the piano. They were not struck by the hammers: rather, they were
designed to vibrate in sympathy to the struck piano strings. It is probable,
then, that he played this piece on this enhanced instrument, thus adding a
physical and acoustical veil to the musical one that he was composing at
this time.
It is worthwhile to examine the differences between the two versions

of the Sarabande and Reflets dans l’eau, especially in regards to dynamic
markings. Measures 1-8 of the earlier version of the Sarabande from
Images are shown in Example 7-12.
There are C-natural and D-natural accidentals in m. 1 in the earlier
Images version, and the bass line of m. 7 adds octaves to the bass line in
m. 7 of the later Pour le piano version (compare Example 7-12 with mm.
1-8 of Example 7-3). For the purposes of this study, however, the
differences in dynamic markings for mm. 6-8 are of greater interest. These
markings are present in Debussy’s hand in the facsimiles of the autographs
of both works.37 There is a crescendo marked over beats 2 and 3 of m. 6
that is not present in the Pour le piano version, and the crescendo of m. 7
extends over the complete measure, instead of only the second half,
emphasizing the interruption structure. Debussy does not comment on the
revision in Pour le piano in the letters to his publisher. There is no
difference between versions for the recurrence of the interruption motive
in mm. 47-48, other than the placement of the crescendo above the staff in
the Images version. In both cases, the markings reflect the fact that the
bass line now follows a continuous descending line in mm. 47-49, rather
than a repetition-extension of mm. 6-8. The dynamic markings of the
Images version for mm. 6 and 7 are closer to the corresponding motive in
Reflets dans l’eau than those of the Pour le piano version.
Example 7-12, Sarabande from Images, mm. 1-8

156 Chapter Seven
While Reflets dans l’eau offers the clearest expression of the

interruption motive itself, from the standpoint of the detailed dynamic
markings, both versions of the Sarabande offer the most highly developed
composing-out of this structure, from the standpoint of the integration of
recurrent Schenkerian prototypes with the triple-tonic complex. Although
the conflicting implications of middleground and background pitches such
as the D# of the Sarabande and the repeated lack of a resolution to scale
degree 1 in E major frustrate localized tonal expectations, the Sarabande
fulfills expectations in the long term if one accepts Debussy’s creative use
of the triple-tonic complex on his own terms. It is not necessary to accept
Debussy’s admonition, “There is no theory. You have merely to listen,” a
statement certain to induce trauma in a music theorist.38 In fact, this
statement is often taken out of context. When Guiraud challenged
Debussy’s assertion that “music cannot be learned” by pointing out that he
(Debussy) spent ten years at the Conservatoire, Debussy conceded the
point: “True enough, I feel free because I have been through the mill.”39
The Sarabande demonstrates that Debussy learned his lessons well,
and that he was able to combine traditional tonal paradigms with the
innovative techniques of Chopin and Wagner in his own unique way.
Stefan Jarocinski notes,
As soon as [a musical] system collapses, the musical formulae which were

based on it are no longer valid; eventually they cease to exist, or else are
incorporated in some other new system in which they may sometimes be
invested with expressive functions completely different from those for
which they had originally been created.40
Recent scholarship and the examples presented in this article suggest that
the “or else” clause applies to Debussy’s music, however much he might
have objected to the choice of the word “system.” In the Sarabande,
Debussy invests Schenker’s monotonal “dividing dominant” with
additional expressive power by creating multiple dividing dominants with
multiple tonal implications. The result of this ambiguity is a “dynamic
system of signs activating the intelligence and stimulating one’s
sensibility”; that is, a musical analogue to Symbolist literature.41 Crotty’s
“veil” and Bailey’s “submersion” are also highly suggestive, the former
evoking Voiles and the latter La cathedrale engloutie. The act of analyzing
the dynamic interactions between the tonal syntax and the post-tonal veil
becomes a heuristic process. It is one of constantly comparing perceptions
of the processes and structures present in the pieces with the assumptions
inherent in the analytical process, and vice-versa. When the process
becomes time-consuming and frustrating, it is worthwhile to remember
Schachter’s reassuring conviction that “one copes, somehow.”
Notes
1
Edward Lockspeiser, Debussy: His Life and Mind, vol. 1 (London: Cassell and
Company, 1966), p. 28.
2
Ibid., p. 206.
3
Joseph N. Straus, “The Problem of Prolongation in Post-Tonal Music,” Journal
of Music Theory 31/1 (Spring 1987): 1-21; Steve Larson, “The Problem of
Prolongation in Tonal Music: Terminology, Perception, and Expressive Meaning,”
Journal of Music Theory 41/1 (Spring 1997): 101-36; Joseph Straus, “Response to
Larson,” Journal of Music Theory 41/1 (Spring 1997): 137-39.
4
Straus, “Response to Larson,” p. 138.
5
Carl Schachter, “A Commentary on Schenker’s Free Composition,” Journal of
Music Theory 25/1 (1981): 132.
6
Adele T. Katz, Challenge to Musical Tradition: A New Concept of Tonality
(London: Putnam and Co., 1947), p. 266.
7
Ibid., p. 293.
8
Richard S. Parks, The Music of Claude Debussy (New Haven and London: Yale
University Press), p. 4.
9
Heinrich Schenker, Free Composition (Der freie Satz), ed. and trans. Ernst Oster
(New York: Longman, 1979), p. 37.
10
Ibid., p. 36.
11
Boyd Pomeroy, “Tales of Two Tonics: Directional Tonality in Debussy’s
Orchestral Music,” Music Theory Spectrum 26/1 (2004): 87-88.
12
Matthew Brown, “Composing with Prototypes: Charting Debussy’s L’Isle
joyeuse,” Intégral 19 (2005): 151-88. For Schenker’s analysis of Chopin’s
Mazurka, see Schenker, Free Composition, p. 71, and Figure 75.
13
Ibid., pp. 164 and 167.
14
Lockspeiser, p. 206.
15
Kip Wile, “Recurrence, Level Organization, and Collection Interaction in Three
Piano Preludes by Debussy,” Indiana Theory Review 22/2 (2001): 59-67.
16
Ibid., pp. 53 and 57.
17
Larson, “Problem of Prolongation,” p. 130.
18
Wile, p. 82.
19
Pomeroy, pp. 88-90. Jim Samson traces the origin of this practice in Chopin’s
music to the influence of the “brilliant” style, a practice with roots in
improvisation (“Chopin’s Alternatives to Monotonality: A Historical Perspective,”
in The Second Practice of Nineteenth-Century Tonality, ed. William Kinderman
and Harald Krebs [Lincoln: University of Nebraska Press, 1996], pp. 34-44).
20
This analysis stands in contrast to Pomeroy’s earlier statement that “Debussy’s
music always remained rooted in triadic consonance and the principle of
monotonality” (“Debussy’s Tonality: A Formal Perspective,” in The Cambridge
Companion to Debussy, ed. Simon Trezise [Cambridge: Cambridge University
Press, 2003], p. 155).
21
John E. Crotty, “Symbolist Influences in Debussy’s Prelude to ‘The Afternoon
of a Faun’,” In Theory Only 6/2 (February 1982): 17-30. Crotty attributes the term
“submerged” to Robert Bailey (p. 18, n. 7).
158 Chapter Seven
22
Ibid., p. 19.
23
Katz, pp. 275-76.
24
Ibid., p. 252.
25
See Schenker, Free Composition, Figure 12.
26
Similarly, Harald Krebs cites the lack of a single Kopfton as an impediment to a
monotonal analysis of Schubert Lieder in “Alternatives to Monotonality in Early
Nineteenth-Century Music,” Journal of Music Theory 25/1 (Spring 1981): 6.
27
Robert Bailey, “An Analytical Study of the Sketches and Drafts,” in Richard
Wagner, Prelude and Transfiguration from Tristan and Isolde, ed. Robert Bailey
(New York: W.W. Norton, 1985), p. 122.
28
Harald Krebs applies the double-tonic complex to Schubert’s “Meeres Stille”
and “Der Wanderer,” in which a monotonal Schenkerian interpretation “is
‘correct’ in the abstract, but it does not accurately reflect my experience of the
song, particularly of its conclusion” (“Some Early Examples of Tonal Pairing:
Schubert’s ‘Meeres Stille’ and ‘Der Wanderer,’” in The Second Practice of
Nineteenth-Century Tonality, ed. William Kinderman and Harald Krebs [Lincoln:
University of Nebraska Press, 1996], p. 28).
29
Bailey, p. 120.
30
Crotty.
31
Krebs also identifies the C# minor/E major double-tonic complex in Schubert’s
“Der Wanderer” (“Tonal Pairing,” pp. 18-23).
32
Crotty, pp. 19–20.
33
Ibid., p. 19.
34
Claude Debussy, Oeuvres Complètes, Série 1, vol. 3 (Paris: Durand Editions,
1991), foreword, p. xviii.
35
Charles Timbrell, “Debussy in Performance,” in The Cambridge Companion to
Debussy, ed. Simon Trezise (Cambridge: Cambridge University Press, 2003), p.
264.
36
Roy Howat, “Debussy’s Piano Music: Sources and Performance,” in Debussy
Studies, ed. Richard Langham Smith (Cambridge: Cambridge University Press,
1997), pp. 99-100.
37
Claude Debussy, Oeuvres Complètes, Série 1, vol. 2, pp. 115-16.
38
Lockspeiser, p. 207.
39
Ibid.
40
Stefan Jarocinski, Debussy: Impressionism and Symbolism, trans. Rolo Myers
(London: Eulenburg Books, 1976), p. 43.
41
Ibid., p. 25.
PART III:
POP MUSIC AND BEYOND

INTRODUCTION TO PART III
Having considered music of the present and its precursors in the tonal
traditions of the past, the book proceeds to contemplate the processes and
techniques of popular music belonging to the recent past, present, and
future, in Part III, “Pop Music and Beyond.” Expanding the scope of pop
music scholarship from its common strongholds of rock and Top-40, the
contributions in this section examine seldom-studied styles that lie on the
fringes of what is normally considered “popular” music.
The discussion begins with Christine Boone’s investigation of the
twenty-first century mashup. Although groundbreaking in terms of genre,
Boone fruitfully reads this cutting-edge style as a (post)modern
manifestation of the timeless quotation principle in music. She shows how
several factors converged—historical, technological, economic, societal,
and aesthetic—to lead to the mashup’s emergence. The rhythmic patterns
of modern jazz, traced back to dance bands of the early 1920s and 30s, are
explored using “topic theory” by Garrett Michaelsen, and applied to an
analysis of Miles Davis’s 1964 solo on “My Funny Valentine.” In
addition to showcasing how techniques typically used to analyze Classic-
era music can provide new insight into jazz improvisation, his article also
serves as a useful compendium of common jazz grooves.
The next two chapters then turn toward the realm of two very different
types of musical multimedia. Nathan Baker illuminates a fascinating
repertoire that has been scarcely analyzed, considering unique formal and
harmonic characteristics of the early Nintendo music of Koji Kondo and
Hirokazu Tanaka. With his experiential analysis of how this music
interacts with the mindset of the video game player, Baker demonstrates
the integral and interactive link between player and soundtrack. Finally,
my own investigation of music videos by Icelandic artists Björk, Sigur
Rós, and Múm demonstrates how understanding concomitantly both
cinematic and musical elements of this medium can add rigor to an
analysis of meaning. Using traditional music-theoretical tools, as well as
more recent ones from the fields of ecological perception and ecocriticism,
the analyses reveal the indelible signature Iceland’s ecological features
have left on the compositional styles of these artists, which can be seen on
screen and also heard in the sounds of the recordings themselves.
When juxtaposed with the rest of the collection, the chapters in this
third section might also invite the reader to reflect upon the way that
162 Introduction to Part III
collaboration may destabilize the authorial sovereignty we typically afford

“the composer.” The mashup usually involves compositions by two or
more extant musicians, and then relies on yet another musician to
skillfully and artfully re-present and alter those musicians’ music. The
standard jazz tunes analyzed by Michaelsen have an original composer
(e.g., Herbie Hancock), but the distinctive musical attributes of the
recording, which constitutes Michalsen’s object of inquiry, reside in the
improvised performances of several musicians (or, perhaps more
accurately, in the interactive space between those musicians, what
Michaelsen calls the “groove”). Tanaka and Nintendo’s other early video
game programmers may come the closest to our traditional view of
composer, though, like the early electronic music pioneers in Bell labs,
composers such as these need to have at least as much skill in
programming and engineering as traditional composition, if not more. And
in their palpable sense of collaboration, the three Icelandic music videos in
my chapter should give us pause to reconsider authorship in popular
music. Aside from the interactions between musicians and
cinematographers, Björk’s video includes amateur vocal collaborators,
while the process-based, monothematic songs of Sigur Rós and Múm
resist the traditional conception of song-writing, and may better be
described as structured participation among a group of songwriters.
Perhaps, in the end, this section should be entitled “Analyzing the Music
of Living Composers? Or Engineers? Or Improvisers? Or Programmers?
Or Songwriters?”
Yet the last word in that alternate title also gives me pause. For as
someone who writes music that is also on the fringes of popular music—
music for drums, electric guitar, bass, and voice that uses odd-cardinality
meters, enharmonic modulations, and through-composed forms—I find it
frustrating that I am considered to be a “songwriter,” while a great deal of
music written for voice and piano using diatonic harmony, common time,
and da capo forms is written by “composers.” Too easily we associate the
former with simplicity and convention, and the latter with genius and
innovation. Last year I was commissioned to compose a piece for classical
guitar and baritone voice. At the premiere, I was flown in because the host
institution wanted a pre-concert lecture delivered by the composer. But,
had I added further layers to the music—recorded bass and drum parts in a
studio, added post-production effects on a computer, collaborated on a
music video with a producer—I would have been the songwriter, and the
work would certainly not have had a commission, a premiere at an
academic institution, and a pre-concert lecture.
Rather than read this closing personal anecdote as yet another plea to
bridge the divide between popular and classical music, or worse still, as
Analyzing the Music of Living Composers (and Others) 163
another sour grapes diatribe from a composer bemoaning the academy’s

reluctance to accept anything but traditional concert music, I invite the
reader to, over the course of these next four chapters, reflect on how this
mashed-up, improvised, collaborative music makes us reconstruct entirely
the variegated identities of the composer in contemporary music-making
practices.
—Brad Osborn
CHAPTER EIGHT
WHEN POP STARS COLLIDE:

MASHUPS AS MUSICAL DESTINY
CHRISTINE BOONE
In 2007, Robert Everett-Green wrote a provocative, short column for a

Canadian newspaper, entitled “The Rise of the Song.”1 In this column, he
argued that few people buy CDs at physical record stores anymore.
Instead, most people now purchase individual songs from online retailers
like Apple’s iTunes Store. This consumption practice, he says, has
reshaped the way that people think about and relate to their music.
Previously, a song was part of a larger whole—the album. In the age of the
MP3 download, however, this is no longer the case; the song stands more
or less alone. Of course, the single had been the dominant commodity of
the music industry until the late 1960s, so this situation is hardly new.2
Consumers today, however, exert a much larger degree of control over
their playlists, especially in the way they can quickly exchange playlists
with one another. Through the rapid exchange of playlists, songs become
embedded within contexts of other songs. But, unlike the similar situation
of radio playlists, these custom-made playlists represent personalized
relationships between songs that often unlock unexpected and idiosyncratic
significance in the songs themselves.
Everett-Green seems unaware of the long history of the 78 and 45
when he claims that “[e]ven without considering how they’re
disseminated, songs over the past decade have become much more
obviously related to other songs than to albums.”3 Even so, he raises an
important point regarding the way technologies of mixing and recording
seem to be undermining the stability of the recorded single. Everett-Green
notes, “[t]he ever-widening use of samples and loops, and the
normalization of remixes and mashups, promotes an idea of the song as a
porous, dynamic entity in a maelstrom of other songs.”4 Consumers today
think of songs as active components ready to be mixed up in a new recipe,
When Pop Stars Collide: Mashups as Musical Destiny 165
not as static artworks meant to be heard again and again in some inviolable
original setting. The rest of this chapter examines the ways that history and
technology converged at the beginning of the twenty-first century to create
a ground fertile for mashup production.
The mashup is, as Kembrew McLeod notes, a “pop music Frankenstein,”5
which challenges accepted legal standards for musical production and
destabilizes the cultural identity of recorded music. In making a mashup,
artists take two (or more) songs, take them apart using computer software,
mix those pieces together, and present the result as a kind of commentary
on the original recordings. The process of mashing thus changes
recordings into these “dynamic entities,” rejecting the more traditional
view of a song as a finished product and instead conceiving of it as raw
material for another iteration of artistic production. However, the concept
is still not quite as radical as it may seem. The roots of this idea—taking
an extant work and using it as the basis for a new work—go back more
than a millennium.6
Borrowing is as intrinsic to human nature as is creation. Every creator,
from architects to choreographers to composers, “must inevitably build
upon the foundations provided by his [or her] predecessors.”7 Although
mashups are a recent phenomenon, they have been influenced by a rich
history of musical borrowing and reuse. Not all of these historical trends
can be said to have directly influenced mashup artists, of course, but there
has been a continuous use of appropriation in musical composition since at
least as far back as the Middle Ages.8 Many accounts of digital sampling
and mashups begin with a brief history of borrowing in Western art music,
focusing on experimental composers of the mid-twentieth century.9 It
seems that some of these authors are citing art music composers in order to
validate the sampling aesthetic of popular contemporary artists. The brunt
of their collective arguments, therefore, is that because this type of
borrowing happened in art music, it is somehow more defensible, both
artistically and legally, when it happens in popular music. Such connections,
however, are quite artificial. In the following historical examination of
musical repurposing, I do not seek to create a synthetic lineage of
influence, nor to establish a hierarchy of classical and popular musics.
Instead, I briefly retrace this history to remind us that it seems to be in the
nature of composers to use and repurpose whatever they have on hand
when creating, regardless of the time period and genre. Mashups are the
result of centuries of musical borrowing, whether or not each historical
tradition can be said to have directly influenced the genre. The “musical
destiny” of this point in time is, as it has been for a long time, musical
repurposing.
166 Chapter Eight
This is not intended to be a complete history of musical borrowing. For

the sake of brevity, I will focus on examples which are the closest to
mashups either in their construction or intention. In the middle ages, tropes
were composed as additions to extant liturgical chant melodies. The
composers of these tropes used a chant melody as a starting point and
added their own embellishments. There were different kinds of tropes:
some involved melismas inserted into the middle of a chant melody, some
were simply the composition of additional words for a chant, and some
involved both music and text. These musical tropes were added into the
original melody in a strictly horizontal manner—there was no vertical
overlap between the chant melody and the trope. Tropes emerged at
roughly the same time as polyphony, another musical practice based on
repurposing existing musical materials. A principal voice sang a pre-
composed chant melody to which an organal voice was later added. While
tropes interacted with borrowed material on a purely horizontal plane,
organal voices added a vertical dimension, placing the chant melody in a
new contrapuntal context.
One of the most popular genres that used borrowed material during the
Baroque period was the quodlibet. Although developed in the fourteenth
century, the form’s popularity increased dramatically during the
seventeenth and eighteenth centuries.10 A quodlibet is a humorous piece
that combines various well-known melodies in succession, like a medley.
The point of the quodlibet was to show off the composer’s cleverness and
technical skill at combining different, unrelated melodies. For example,
the last movement of Bach’s Goldberg Variations is a quodlibet based on
two popular German folk songs.11 In terms of intentionality, the quodlibet,
in which a composer takes two or more pieces and juxtaposes them in a
humorous or skillful way, is the genre that most closely matches the
modern-day mashup. In both types of composition, humor often results
from songs of a differing style and lyrical topic being pushed up against
one another. Two things that should not work well together somehow do,
provided that the composer can bring out their hidden musical similarities.
The eighteenth and nineteenth centuries also saw the emergence of a
type of composition similar to the mashup called the potpourri. The
potpourri was a medley of melodies from popular operas, similar to
Hooked on Classics (but without the incessant disco beat). Czerny wrote a
potpourri based on arias from Louis Spohr’s Faust,12 for example, and
Sarasate’s various fantasies for violin and piano also fall into this
category. The idea behind this type of composition is to give the audience
many different recognizable themes in a short amount of time. This time
period also saw an increased number of insertion arias in opera. Both
Rossini and Wagner, for example, helped start their careers by writing
arias for particular singers to be inserted into other composers’ operas.13
Haydn and Mozart also composed insertion arias for specific singers in
their own operas. This practice, taken to an extreme, became what is
known as a pasticcio, or an opera made of pieces pasted together by
different composers. Like insertion arias, the numbers in the pasticcio
were chosen to suit particular singers’ voices. For this reason, the operas
had different numbers each time they were performed.
American popular song of the early twentieth century also frequently
featured musical borrowing. George M. Cohan, for example, used what J.
Peter Burkholder calls the “patchwork” technique to sew together
melodies from several traditional and patriotic American songs.14 Cohan
managed to quote “Yankee Doodle,” “Dixie,” “The Girl I Left Behind
Me,” “The Star Spangled Banner,” and “The Battle Cry of Freedom” all
within thirty-four measures of his 1904 song “Yankee Doodle Boy.”15
Classical composers continued to use borrowed music into the
twentieth century as well. Perhaps most famously, Charles Ives made
extensive use of borrowed musical material, quoting and reworking
various American folk songs and patriotic tunes.16 Stravinsky’s ballet,
Pulcinella (1920), was also based on preexisting compositions, namely
those of Pergolesi.17 Luciano Berio took borrowing to an extreme when he
wrote his most famous work, Sinfonia, in 1968. In the third movement, he
creates a collage using a large portion of the scherzo from Mahler’s
second symphony. Berio rearranges Mahler’s music and also quotes
Beethoven, Ravel, Debussy, Strauss, Schoenberg, Berg, Stockhausen, and
others.18 In addition to these musical quotes, Berio also instructs the
singers to speak from various extant texts, including sizable portions of the
Wozzeck libretto and Beckett’s L’Innomable.
In addition to our proclivity towards musical reuse, technological
developments of the twentieth century have also helped lead to the
inception of mashups. I have identified four streams of twentieth-century
music that use new technologies in order to repurpose musical material:
experimental music, popular music, live DJ sets, and sample-based hip
hop. In contrast with the older musical styles discussed above, these four
contemporary genres bear a more direct lineage to mashups of the twenty-
first century.
Recording technology greatly changed the way that composers were
able to borrow previously existing music. It had always been possible to
reuse various parts of a composition; the instrumentation, melody, chord
progression, lyrics, and form were easily imitated. Even direct quotations
from other composers could be used. But the advent of recording made it
168 Chapter Eight
possible not just to imitate, but to actually reuse a specific performance of

a work by literally copying it. Recording technology had not been around
long before composers started to experiment with it. As early as the 1920s,
Stefan Wolpe and Ottorino Respighi were incorporating recorded sounds
into their compositions.19 John Cage’s Imaginary Landscape pieces were
among the first to use record players and radios as instruments. Imaginary
Landscape No. 1, written in 1939, was a quartet for a piano, a cymbal, and
two turntables played at varying speeds.20 In 1951, Cage wrote Imaginary
Landscape No. 4, in which performers are instructed to tune twelve radios
to specific frequencies during the course of the piece.21 The piece, then, is
different during each performance. Cage uses other people’s sound
(whatever happens to be on the given frequencies: music, newscasts,
commercials, even static) to create his soundscape, but the result ends up
being entirely random.
The earliest example of sampling to appear on the Billboard charts was
a 1956 musical skit called “The Flying Saucer.”22 Bill Buchanan and
Dickie Goodman told the story of an alien invasion using popular songs of
the day. A news announcer called John Cameron Cameron (played by
Goodman) interviews people on the street about their reactions to the
invasion, and the people’s reactions are the samples. For example, after
the announcer asks, “Pardon me, madam, would you tell our audience
what would you do if the saucer were to land?” Little Richard sings his
response: “Duck back in the alley!” from the song “Long Tall Sally.” In
1961, American composer James Tenney combined popular music with
avant-garde techniques when he wrote Collage #1 (Blue Suede). Using
Elvis Presley records as his medium, Tenney chopped them up and
reassembled them, speeding up and slowing down their tempi.23 Elvis’s
characteristic voice is recognizable during the middle of the piece, but he
is only ever allowed to sing one or two words before Tenney cuts him off
with white noise or an extreme tempo shift.
Probably the earliest example of two pop songs being heard at the
same time, overlaid on top of one another, is Alan Copeland’s 1968
arrangement of the theme from Mission: Impossible and the Beatles’
“Norwegian Wood.”24 This is not technically a mashup, because no
previously recorded music is used; all the parts were recorded specifically
for this track.25 The songs are arranged so that the Mission: Impossible
theme is in G minor and “Norwegian Wood” is in G major. As a result of
this, the song constantly alternates between parallel major and minor
modes as they are juxtaposed horizontally. When the songs do overlap
vertically, a conflict between the B-flat and B-natural never occurs—
Copeland alters the melody of “Norwegian Wood” in order to make it sit
over the other song better. He also changes the meter of “Norwegian
Wood.” Originally in triple meter, Copeland speeds up certain rhythms in
order to condense two bars of 3/4 time into one bar of 5/4, the time
signature of Mission: Impossible. A short excerpt of the resulting piece is
shown in Example 8-1.
Example 8-1, Alan Copeland, “Mission: Impossible Theme/Norwegian Wood,”

(0:06-0:17), transcription by author
In the 1960s, popular music began to be influenced by contemporaneous

experimental music. Composers like Schaeffer, Varèse, and Stockhausen
made a great impression on several groups in the 1960s, particularly the
Beatles.26 The influence of their tape-collage style can be heard in
“Revolution 9” on the Beatles’ self-titled album from 1968, and in Frank
Zappa’s “The Chrome-Plated Megaphone of Destiny” on his album from
the same year, We’re Only in it for the Money.27
A different type of musical repurposing can be found in dance music
played and created by DJs. Literally a disc jockey, the DJ started out as a
mere employee of radio stations, charged with announcing each song and
its artist, and playing the record.28 However, the role of the DJ today,
especially the club DJ, has become something different. In the 1970s,
disco DJs began to work with two copies of the same record played on two
different turntables in order to make each song longer.29 Eventually,
170 Chapter Eight
twelve-inch disco medleys began to be released. These records featured

seamless transitions between disco hits; the beat was continuous, so that
people could keep dancing the entire time.30 In order to make these kinds
of seamless transitions in a live club setting, the DJ had to be skilled at
beat matching. First, the two songs chosen should have relatively similar
tempi. While one song is playing on one turntable, the DJ is listening to
the other song on the other turntable through headphones. The second
song’s tempo can be shifted, if need be, to ensure that the beats of the two
songs are lined up with one another. Finally, using a fader, the first song
may be gradually shifted into the second song as the audience keeps
dancing.31
Today, DJs and mashup artists concentrate on more than just
connectivity between songs. “Out has gone the idea of introducing records
and in has come the notion of performing them,”32 say Bill Brewster and
Frank Broughton. “DJs track down greatness in music and squeeze it
together. Like a master chef who picks just one perfect cherry from each
tree to make his pie, a DJ condenses the work and talent of hundreds of
musicians into a single concentrated performance.”33 DJs today take only
the best, most danceable portions of songs and extend them, interpolating
them with beats, chords, and vocal exclamations from other songs. In the
words of Daniel Hadley, “the musical text is continually being
transformed...The music contained on any piece of vinyl is not considered
to be a fixed and immutable creation, but rather raw material which must
be recast through its insertion into a flow of texts which interact both with
each other and with the bodies of the dancers.”34 What was once a series
of unrelated songs becomes a connected “set” of songs that interact and
converse with one another to form a narrative.35 The DJ creates a new
composition, in effect, out of other people’s songs.36
The beginnings of rap and hip hop music are very similar to the
beginnings of the DJ-as-composer/performer in the disco and club world.
In the 1970s, mobile DJs in the Bronx would play old funk and soul
records. Like the disco DJs, these “street DJs” would also have two
turntables, each with a copy of the same record. The DJ would extend the
break section of the song by playing it on one turntable,37 then playing it
on the other, and then back to the first, etc., using the same beat-matching
techniques as the club DJ. Although the music was different, the idea was
the same: keep people dancing. DJ Kool Herc, one of the first mobile DJs
to extend a break in this way, said that “[o]n most records, people have to
wait through a lot of strings and singing to get to the good part of the
record. But I give it to them all up front.”38 The rapping started when the
DJs began shouting “party phrases, such as ‘let’s jam, y’all,’”39 into a
microphone while mixing their records. Eventually the rapping and mixing
became too complicated for one person to handle, and DJs hired MCs to
make the rhymes.40
Hip hop DJs began to seriously alter (or “flip”) the material that they
sampled. They chopped up sounds into smaller pieces and rearranged
them, and they looped musical phrases.41 As with dance club DJs, hip hop
DJs, too, became involved in a creative act, carefully selecting source
materials for composing their own tracks. Referencing soul or funk songs
added a certain amount of power to a track; after all, a sample retains some
of its original meaning, even when it is placed in a new context. Other
combinations of songs were picked simply because of their humorous
juxtaposition.42 The skills to mix together sounds from different sources
were very important to a DJ’s reputation. Not only did one have to be able
to match beats and find records that complemented or conversed with each
other, but they had to do so at a moment’s notice (until the 1980s, all hip
hop was mixed live, not in a studio). Early hip hop DJs like Afrika
Bambaataa, Kool Herc, and Grandmaster Flash engaged in DJ battles to
see who could come up with the most innovative cuts and mixes.43 The
audience judged these competitions with their reactions to the DJs’ work.
The stranger, more esoteric songs mixed, the better.44 Prince Be Softly, of
the group PM Dawn, says that “[s]ampling artistry is a very misunderstood
form of music. A lot of people still think sampling is thievery but it can
take more time to find the right sample than to make up a riff. I’m a
songwriter just like Tracy Chapman or Eric B. and Rakim.”45
Until fairly recently, the only people with access to technology that
could potentially create a mashup were DJs and record producers, but now
almost anyone with a computer can do it. Computer programs like Acid
Pro, Logic, and Pro Tools are relatively cheap and easy to use (mashups
can even be produced using Audacity, which is completely free). These
programs allow users to change the tempo or pitch of a song independent
of one another, and to layer and juxtapose different tracks. One of the most
common types of mashup is one in which the vocals from a rap song are
played over the instrumental background of a pop song. Because rap is
spoken/chanted rather than sung, rap vocals eliminate one of the most
challenging parameters of making an effective mashup—the creator
doesn’t have to worry about transposing the songs to the same key, or
seeing that the notes in a melody fit with the specific harmonies of the
other song. The primary musical issues become rhythm and meter,
ensuring that the two songs align metrically. This task is easily
accomplished in any of the previously mentioned software programs by a
process known as “beatmatching” or “beatmapping.”
172 Chapter Eight
Another common type of mashup involves sampling the sung vocal

line from a hip hop or pop song, and playing it over the musical
background of another song. This type of mashup seems to be more highly
regarded by critics, and is certainly harder to create than one using rapped
vocals. As Sasha Frere-Jones puts it, “[t]he most celebrated mashups are
melodically tuned, positing a harmonic relationship between, say,
Madonna’s voice and the Sex Pistols’ guitars.”46 This is what happens in
DJ Freelance Hellraiser’s so-called “shotgun wedding”47 of Christina
Aguilera’s “Genie in a Bottle” and the Strokes’ “Hard to Explain.” The
Strokes’ song is in G major, and stays in G major for the mashup. “Genie
in a Bottle” is originally in F minor, but is transposed down to E minor.
After the transposition, the songs are in relative keys, and therefore use the
same diatonic scale. The result, called “A Stroke of Genie-us,” reveals a
new relationship between the vocal and instrumental parts. Chord tones
become non-chord tones and vice-versa. Frere-Jones observes that
Aguilera’s vocal line “lays over the Strokes’ chord changes so deliciously
you can’t imagine why the song didn’t always do that.”48 Examples 8-2
and 8-3 show Christina Aguilera’s vocal line both in its original context
and in its new harmonic context in Hellraiser’s mashup. Non-chord tones,
with respect to each harmonic context, are shown in parentheses.
Example 8-2, Christina Aguilera, “Genie in a Bottle,” (0:41-0:52), transcription by

author
Example 8-3, DJ Freelance Hellraiser, “A Stroke of Genie-us,” (1:06-1:18),

transcription by author
Roland Barthes notes that “[a]ny text is woven entirely with citations,
references, echoes, cultural languages, which cut across it through and
through in a vast stereophony.”49 It is truly impossible to produce an
entirely original idea. Artists are bombarded with influences from birth
that necessarily make their mark on that which they create. Joseph Straus
writes about Harold Bloom’s “anxiety of influence” as it applies to the

music of the twentieth century. Composers, he says, grab onto material
from the past, and transform it to clear their own creative space. This
transformation is what he calls “remaking the past.”50 This anxiety of
influence and the renovation of previous material has always existed in
poetry (the subject of Bloom’s original theory), as well as in music. Mark
Twain said, “[t]he kernel, the soul—let’s go further and say the substance,
the bulk, the actual and valuable material of all human utterances—is
plagiarism. For substantially all ideas are second-hand, consciously or
unconsciously drawn from a million outside sources.”51
Straus says that this concept applies more in the twentieth century than
it did before, that we are surrounded with these anxiety-inducing voices of
the past now more than ever. A variety of factors have led to our over-
saturation in music of the past: historical preservation and study,
globalization, the introduction of fixed media on which to preserve
recordings, and the distribution power of the internet, to name a few. We
hear music almost everywhere we go and, furthermore, can summon any
piece on demand with our personal computers or mobile devices. Due to
the conjunction of technology and influence at the beginning of this
century, mashups can perhaps be said to have come about quite inevitably.
“People are users. They are producers, storytellers, consumers, interactors
—complex, varied beings, not just people who go to the store, buy a
packaged good off the shelf and consume.”52 Mashups, like all of music
history, are a product of influences, transformed artistically into a new
creation.
Notes
1
Robert Everett-Green, “The Rise of the Song,” The Globe and Mail (Canada),
January 27, 2007, Weekend Review section.
2
Richard Harker, “It’s 1960 All Over Again,” FMQB Online, April 20, 2007,
<http://www.fmqb.com/article.asp?id=389486>.
3
Everett-Green, “The Rise of the Song,” p. R1.
4
Ibid.
5
Kembrew McLeod, Freedom of Expression ®: Overzealous Copyright Bozos and
Other Enemies of Creativity (New York: Doubleday, 2005), p. 79.
6
Mark Katz, Capturing Sound: How Technology Has Changed Music (Berkeley:
University of California Press, 2004), p. 139.
7
Hugh Arthur Scott, “Indebtedness in Music,” The Musical Quarterly 13/4 (1927):
497.
8
J. Peter Burkholder, “Borrowing,” Grove Music Online [accessed 23 September,
2012].
174 Chapter Eight
9
Mark Katz, Capturing Sound: How Technology Has Changed Music; Joanna
Teresa Demers, Steal This Music: How Intellectual Property Law Affects Musical
Creativity (Athens, GA: University of Georgia Press, 2006); McLeod, Freedom of
Expression®; Christoph Cox and Daniel Warner, Audio Culture: Readings in
Modern Music (New York: Continuum, 2004); David Toop, Ocean of Sound:
Aether Talk, Ambient Sound and Imaginary Worlds (London: Serpent's Tail,
1995); Hugh Davies, “A History of Sampling,” in unfiled: Music Under New
Technology, ed. Chris Cutler (London: ReR, 1994), pp. 5-12.
10
Maria Rika Maniates et al. “Quodlibet,” Grove Music Online.
11
Davies, “A History of Sampling,” p. 11.
12
Andrew Lamb, “Potpourri,” Grove Music Online.
13
Philip Gossett, “Gioachino Rossini: Early Years,” Grove Music Online.
14
J. Peter Burkholder, All Made of Tunes: Charles Ives and the Uses of Musical
Borrowing (New Haven: Yale University Press, 2005), p. 322.
15
Ibid., pp. 322-24.
16
Ibid.
17
Stephen Walsh, “Igor Stravinsky: Exile in Switzerland 1914-20,” Grove Music
Online.
18
David Osmond-Smith, Playing on Words: A Guide to Luciano Berio’s Sinfonia
(London: Royal Music Association, 1985).
19
Kembrew McLeod, Owning Culture: Authorship, Ownership, and Intellectual
Property Law (New York: Peter Lang Publishing, 2001), p. 110.
20
Cox and Warner, Audio Culture, p. 25.
21
Demers, Steal This Music, p. 75.
22
Kembrew McLeod, “Confessions of an Intellectual (Property): Danger Mouse,
Mickey Mouse, Sonny Bono, and My Long and Winding Path as a Copyright
Activist- Academic,” Popular Music and Society 28/1 (February 2005): 81.
23
Ibid.
24
McLeod, Freedom of Expression®, p. 162.
25
Copeland’s “Mission: Impossible Theme/Norwegian Wood” is actually an
example of what I call a “cover mashup.” See Christine Emily Boone, “Mashups:
History, Legality, and Aesthetics” (Ph.D. Dissertation, University of Texas at
Austin, 2011).
26
Chris Cutler, “Plunderphonia,” in Audio Culture: Readings in Modern Music,
ed. Christoph Cox and Daniel Warner (New York: Continuum, 2004), p. 147.
27
Ibid., 148.
28
I consciously use the pronoun “he” to refer to the DJ throughout this section.
There are exceptions, of course, but most DJs were historically, and remain
contemporaneously, men.
29
30
Ibid., 161.
31
Brian Todd Austin, “The Construction and Transformation of the American Disc
Jockey Occupation, 1950-1993” (Ph.D. Dissertation, University of Texas at
Austin, 1994), p. 156.
32
Bill Brewster and Frank Broughton, Last Night a DJ Saved My Life: The History
of the Disc Jockey (New York: Grove Press, 2000), p. 8.
33
Frank Broughton and Bill Brewster, How to DJ Right: The Art and Science of
Playing Records (New York: Grove Press, 2003), p. 12.
34
Daniel J. Hadley, “‘Ride the Rhythm’: Two Approaches to DJ Practice,” Journal
of Popular Music Studies 5/1 (1993): 58.
35
Brewster and Broughton, Last Night a DJ Saved My Life, p. 8.
36
Mark J. Butler, Unlocking the Groove: Rhythm, Meter, and Musical Design in
Electronic Dance Music (Bloomington: Indiana University Press, 2006), p. 33.
37
A “break,” sometimes called a “drum break,” is an interlude in a song where all
the parts except the drums drop out of the mix.
38
Quoted by Robert Ford, “B-Beats Bombarding Bronx: Mobile DJ Starts
Something With Oldie R&B Disks,” Billboard, July 1, 1978, p. 65.
39
Cheryl L. Keyes, Rap Music and Street Consciousness (Champaign: University
of Illinois Press, 2004), p. 1.
40
Ibid.
41
Joseph G. Schloss, Making Beats: The Art of Sample-Based Hip-Hop
(Middletown, CT: Wesleyan University Press, 2004), p. 106.
42
Siva Vaidhyanathan, Copyrights and Copywrongs: The Rise of Intellectual
Property and How it Threatens Creativity (New York: NYU Press, 2003), p. 135.
43
Steve Hager, Hip Hop: The Illustrated History of Breakdancing, Rap Music, and
Graffiti (New York: St. Martin's Press, 1984), p. 34.
44
Demers, “Sampling as lineage in hip-hop,” p. 28.
45
Quoted by Tricia Rose, Black Noise: Rap Music and Black Culture in
Contemporary America (Middletown, CT: Wesleyan University Press, 1994), p.
79.
46
Sasha Frere-Jones, “1 + 1 + 1 = 1: The New Math of Mashups,” The New Yorker
80/42 (January 10, 2005): 85-86.
47
48
Frere-Jones, “1 + 1 + 1 = 1: The New Math of Mashups.”
49
Roland Barthes, quoted by Jonathan Lethem, “The Ecstasy of Influence: A
Plagiarism Mosaic,” in Sound Unbound: Sampling Digital Music and Culture, ed.
Paul D. Miller (Cambridge, MA: The MIT Press, 2008), p. 43.
50
Joseph N. Straus, “The ‘Anxiety of Influence’ in Twentieth-Century Music,”
The Journal of Musicology 9/4 (Fall 1991): 477.
51
Mark Twain, quoted by Lethem, “The Ecstasy of Influence: A Plagiarism
Mosaic,” p. 43.
52
Yochai Benkler, quoted by Robert S. Boynton, “The Tyranny of Copyright?,”
New York Times Magazine, January 25, 2004, p. 43.
CHAPTER NINE
GROOVE TOPICS IN IMPROVISED JAZZ
GARRETT MICHAELSEN
In her book Saying Something, Ingrid Monson asserts that “musicians’

discussions of the higher levels of improvisational achievement frequently
emphasize time and ensemble responsiveness as the relevant framework
rather than, for example, large-scale tonal organization.”1 Analysis of jazz
improvisation, however, has traditionally emphasized parameters relating
to “large-scale tonal organization,” to a far greater extent than “time and
ensemble responsiveness.”2 Monson does not mean to devalue these
parameters, but rather to point out that elite performers “take for granted
the harmonic and melodic competence of the player.”3 Jazz analysts have
therefore often ignored an aspect of improvised performance that
musicians themselves prize most highly. These two interrelated concepts
of time and ensemble responsiveness combine to produce the “groove.” In
order to re-center jazz analysis toward the groove, I propose the use of a
finely honed tool in the analyst’s kit: topic theory.
As first set forth by Leonard Ratner, musical topics are “subjects for
musical discourse … [that may] appear as fully worked-out pieces, i.e.,
types, or as figures and progressions within a piece, i.e., styles.”4 In Classic
music, Ratner identifies a variety of dance types that composers import
into other works (such as the use of sarabande rhythms in a string
quartet). Ratner states that “[d]ances, by virtue of their rhythm and pace,
represented feeling.”5 Wye Allanbrook expands on this notion with her
idea of “rhythmic gesture”: “rhythm—the number, order, and weight of
accents and, consequently, tempo—is a primary agent in the projecting of
human postures and thereby of human character.”6 Allanbrook creates a
spectrum of dance meters ranging from ecclesiastical or exalted passions
to gallant or terrestrial passions, and subsequently identifies these dance
meters in Mozart’s operas. In doing so, Allanbrook models a way in which
rhythmic gestures influence our perception and understanding of Mozart’s
Groove Topics in Improvised Jazz 177
characters, plots, and potentially other musical parameters. In Classic

music, the dance topic allows for vague and ephemeral ideas of rhythmic
feel to be pinned down with greater accuracy, allowing for richer
interpretive meanings to emerge from analysis.
Analogously to the dance topic in Classic music, many of the rhythmic
patterns of modern jazz stem from those of the swing and dance bands of
the 1920s and ’30s. The small-group jazz of the bebop modernists in the
mid-1940s and onwards deviated from swing in that it was intended for
listening, not dancing.7 The faster tempos, greater rhythmic complexity,
and focus on improvised solos over notated ensemble passages all point to
this shift from the dancehall to the nightclub. Numerous continuities
between modern jazz and swing remained, however. Both styles
emphasized a core group of musicians—the rhythm section—who were
largely responsible for the creation of the groove. Rhythm sections
continued to consist primarily of piano, bass, and drums, and they
remained an interrelated unit defined in contrast to the role of the
improvising soloist.8
Dance bands performed a wide variety of dance types. Don DeMichael
and Alan Dawson’s 1962 drum manual presents the various possibilities:
two-beat, 3/4, lame duck, shuffle, ethnic, Latin, tango, beguine, bolero,
conga, rhumba, samba, calypso, mambo, cha cha cha, merengue, and
nanigo.9 These dances are rather strictly defined, with exact rhythmic
patterns a drummer would repeat and intersperse with notated ensemble
figures. In addition to drum parts, these dances would also dictate specific
styles of accompaniment for the bass and chordal instruments.
In modern jazz, elements of these dances are distilled into the grooves
I will discuss. As the musicians no longer need to articulate a stable and
repeating rhythmic pattern for dancing, the expression of a groove is far
more complex and variegated than that of a dance. Furthermore, Monson
notes that the term groove has two complementary meanings, one of
which uses the word as a noun and the other as a verb. Grooves as nouns
are “particular sets of rhythm-section parts that combine to produce
particular rhythmic patterns.”10 Larry Zbikowski expands on this
definition, writing that a groove is “a large-scale multi-layered pattern that
involves both rhythmic and pitch materials,” and can include contributions
from any member of an ensemble, not just the rhythm section.11 In my
formulation, every member of a jazz ensemble, not just the rhythm
section, participates in the creation of a groove.
The use of groove as a verb relates to Charles Keil’s concept of
“participatory discrepancies.”12 Participatory discrepancies are slight
inconsistencies occurring between players in performance that cause
178 Chapter Nine
music to “be personally involving and socially valuable.”13 Keil defines

two types of participatory discrepancies: (1) processual, and (2) textural,
or timbral. Processual discrepancies are the slightly unsynchronized or
unique note placements occurring between players in an ensemble that
give a groove its particular feeling. Textural discrepancies are the
differences in timbre or intonation between players that give an ensemble
its own unique sound. Monson sums up the concept of groove-as-verb by
noting that most musicians she interviewed “described grooving as a
rhythmic relation or feeling existing between two or more musical parts
and/or individuals,”14 a perspective that corresponds with Keil’s original
formulation.
According to Keil, participatory discrepancies are musicians’ default
ways of relating to their fellow performers. Certain performers might play
slightly ahead of, or behind, the beat. Keil does not consider the possibility
that musicians might alter their participatory discrepancies throughout a
performance, however. In fact, experimental research conducted by J. A.
Prögler suggests that many musicians adopt a shifting and inconsistent
approach towards these discrepancies.15 I will demonstrate through an
analytical example that musicians often alter their performance of a
groove for expressive effect throughout an improvisation.
My concept of groove relies on both meanings of the term—groove as
noun and verb. Grooves are therefore both the common rhythmic patterns
played by the rhythm section and soloists, as well as the particular ways of
playing those patterns (i.e., Keil’s participatory discrepancies). Matthew
Butterfield, in dialogue with Keil, writes that “participatory discrepancies
interact with aspects of syntactical pattern in systematic ways in the
production of engendered feeling in jazz and other groove-based
musics.”16 To study groove, one must therefore consider both the noun and
the verb.
As the foregoing discussion demonstrates, grooves have been given
fairly comprehensive treatment by previous commentators.
Reconceptualizing grooves as topics, however, adds to the conversation
the idea that the expressive correlations of various grooves can be
imported into specific improvisations. These expressive correlations are
elements of style that listeners and performers understand implicitly.
Correlations, according to Robert Hatten, “typically involve general
cultural units … or expressive states defined by basic semantic
oppositions in a culture.”17 The basic oppositions expressed by groove
correlations in jazz are hot and cool, excitement and relaxation, forward
motion and stasis. Grooves express these values through a combination of
rhythmic patterns and participatory discrepancies. As Butterfield has
shown, an identically notated drum part for a swing groove might be

perceived as “relaxed” or “laid back” when beats 2 and 4 are delayed,
whereas anticipating beats 2 and 4 slightly will result in a heightened
feeling of anacrusis and increased energy.18 Similarly, grooves with
different rhythmic patterns situate themselves on different regions of the
scale from hot to cool.
The most fundamental characteristic of a jazz groove is its manner of
rendering eighth notes. Almost all jazz is notated (when it is notated) and
conceptualized with the eighth note as the primary rhythmic subdivision,
but it is assumed that a player will perform eighth notes with varying
degrees of inequality.19 A rhythmic spectrum therefore exists with equal,
evenly spaced eighth notes on one extreme, and very unequal, dotted
rhythms on the other. Example 9-1 illustrates this spectrum of eighth-note
inequality. It is important to note that the example represents continuous
rather than discrete points along the spectrum. Musicians regularly inhabit
the space between the evenly spaced “straight eighths” and the triplet-
based eighths, for instance. Each musician tends to have a default way of
playing swing eighths that plays a large role in that performer’s unique
style.20 Improvisers may also choose to alter their swing eighths depending
on specific expressive situations.
Example 9-1, Spectrum of Eighth-Note Inequality and Expressive States
In general, the spectrum of equal to unequal eighths correlates with a

range from relaxation to excitement. Grooves with even eighths tend to
correlate with more relaxed and static expressive states while grooves with
unequal eighths correlate with more excited and energetic expressive
states. While each groove allows some degree of leeway in the specific
expression of eighth-note inequality, tempo imposes certain limits on this
flexibility. At faster tempos, eighth notes tend to be played closer to the
even side of the spectrum, mostly due to the extremely short amounts of
180 Chapter Nine
time such minute differences would involve. Slower tempos,

consequently, allow for a greater range of eighth-note expressions.
Example 9-2 illustrates the greater range of eighth-note inequality at
slower tempos compared to faster tempos.
Example 9-2, Relationship of Tempo and Eighth-Note Inequality
The specific jazz grooves to be discussed throughout this chapter fall

somewhere on the spectrum of eighth-note inequality given in Example
9-1. Example 9-3 summarizes the fundamental features of these grooves.
Common rhythmic patterns for the drums will be notated for each of the
grooves, and accompanimental roles for the bass and piano will be
discussed as well. To begin, the swing groove is the standard groove of
most jazz music. It permits a wide variety of eighth-note inequalities and
expressive states. Example 9-4 gives a common drum pattern for swing.
The essential features for the drummer are the hi-hat accents on beats 2
and 4, known as the backbeat, and the swing pattern in the ride cymbal.
This pattern is inflected in various individual ways, with certain drummers
playing the eighth notes closer to the even side and others closer to the
triplet or dotted-eighth side. The drum patterns notated for each of the
grooves would not be repeated unchangingly throughout a performance,
but would serve as the basis from which elaboration and communication
with the rest of the ensemble would take place. Bassists improvise what is
known as a “walking” bass line. This line consists of quarter notes on each
beat that, in addition to emphasizing chord tones, include elaborations
such as passing tones, neighboring tones, and appoggiaturas. Pianists add
to the groove by “comping,” a technique whose name derives either from
the words “accompanying” or “complementing.”21 Comping consists of
adding chordal accompaniment to the ensemble that may perhaps lock up

with a repeated rhythmic figure or fill in gaps in the musical texture.
Pianists serve this role in all of the grooves, and so pianists’ roles in
expressing a specific groove rely more on their choice of eighth-note
inequality and activity level.
Groove Tempo Eighth-Note Drums Bass Piano

Inequality
Swing Slow- Even Ride pattern, Walking Free
Medium- to dotted hi-hat on comping
Fast 2 and 4
Shuffle Medium Triplet Shuffle beat, Repeating Riff-based
to dotted repeated patterns comping
triplet
eighths
Waltz Medium- Even 3/4 pattern in Walking Comping
3/4 Fast to dotted ride
Ballad Slow Even Ballad “stir” Sustain, Legato
to dotted 1 and 3 comping,
sustain
Latin Medium- Even Constant Repeating Rhythmic
Fast eighths/ patterns, comping
sixteenths, emphasis on
clave roots and fifths
patterns
Example 9-3, Table of Modern Jazz Grooves
The swing groove gradually transforms into a few other grooves

depending on various methods of inflection. If an ensemble consistently
performs triplet eighth notes, effectively producing the sound of 12/8
meter, the groove shifts into a shuffle. This groove is the standard groove
of blues, and of rock styles based on blues. Example 9-5 gives a common
drum pattern from a recent drum manual.22 While the specific
combinations of bass, snare, and cymbal articulations vary, the
quintessential shuffle elements are the repeated triplet-eighths in the ride
cymbal, heavy emphasis on the 2-and-4 backbeat, and incorporation of the
bass drum into the pattern producing a darker, more grounded feeling in
contrast with the lightness of swing. Some musicians will take the triplet
eighths even further, producing an even more unequal dotted rhythm.
Rather than walking, bassists in a shuffle groove generally play repeated
motives, often called “riffs,” that are then altered to fit the chord
progression. Pianists comp in a similarly riff-based manner, often
choosing to lock up with the drums and bass rather than interacting with
the soloist. The shuffle groove thus represents an intensification of swing.
182 Chapter Nine
It produces greater levels of energy and is therefore often used at high

points in a solo.
Example 9-4, Drum Pattern for Swing Groove
Example 9-5, Drum Pattern for Shuffle Groove
Another groove related to swing is the 3/4 jazz waltz. This less
common groove in the jazz style places swing into a triple-meter context.
Its primary elements are summarized in Examples 9-3 and 9-6. As many
of its features are similar to those of swing, a detailed discussion of the
jazz waltz will not be given here.
Example 9-6, Drum Pattern for 3/4 Jazz Waltz
A number of grooves inhabit the region closer to the even end of the
eighth-note spectrum. The ballad groove is the one used in performing
both newly composed ballads and standard ballads from the American
songbook tradition. Ballads have slow tempos and generally employ even
eighth notes. A shift to more unequal eighths often results in increased
energy levels and potentially a shift into a slow shuffle. Being slow and
even, ballads tend to project the most relaxed and static expressive state of
all the grooves. The drummer commonly plays with wire brushes rather
than sticks and usually performs a “stir” on the snare drum. To produce
this quintessential timbre of the ballad groove, drummers press brushes on
the snare drum and circle them around the drumhead slowly and evenly,
producing a constant “shhh” sound. Additionally, the drummer often snaps
the hi-hat on 2 and 4 to maintain the backbeat. Example 9-7 summarizes
this pattern. Rather than walking, bassists tend to play in a “two-feel,”
with half notes on beats 1 and 3. This increases the feeling of relaxation in
the ballad groove and gives the bassist the option of walking to increase
the energy level of a performance. Pianists similarly comp in a more
subdued manner, often sustaining chords for longer periods of time than
they might in a swing groove. As with all of the grooves, greater eighth-
note inequality will inject activity into the ballad’s more static background
texture.
Example 9-7, Drum Pattern for Ballad Groove
One common shift of groove that occurs in ballads is to double time.

When an improviser cues such a shift, the tempo doubles, producing a
medium-to-fast swing groove. The repeating harmonic framework of the
tune usually does not double, however, so that each harmonic change now
lasts twice as many measures (though this results in approximately the
same duration) as it did in the slower ballad groove. While the tempo shift
brings about a significant increase in energy, the now leisurely harmonic
pace continues to temper the energy level of the groove. Common aspects
of the swing groove discussed above apply equally to the double-time feel.
The only groove that rarely permits any sort of eighth-note inequality
is the Latin groove. Actually a category of specific grooves, Latin includes
a wide variety of rhythmic styles imported into jazz from Latin America.
While many musicians carefully study authentic forms of specific grooves
such as the Brazilian samba, the catchall term “Latin” is often used to
describe grooves that emphasize even eighth notes over swing eighths.23
Due to their location on the even side of the spectrum, Latin grooves
correlate with a more relaxed expressive state. That is not to say that the
rhythmic complexities of certain authentic Latin grooves cannot express
extremely energetic and excited states, but that in the generic form usually
184 Chapter Nine
referenced by modern jazz, Latin grooves evoke a calmer, more serene

manner than swing.24
Example 9-8, Standard Patterns for Latin Claves
While there is no single drum pattern that applies to all of the various
expressions of a Latin groove, many Latin grooves often include either the
son or rhumba clave, a rhythmic pattern that may be articulated
throughout the drum set. These two standard patterns are notated in
Example 9-8.25 The two are very similar to one another, with the rhumba
adding an additional syncopation on the third note of the “3” grouping.
Both claves are heard in 3+2 and 2+3 versions, the difference being
whether the group of three articulations falls in a hypermetrically stronger
or weaker position than the group of two. Drummers can play rhythmic
figures that either do or do not stress these clave patterns, depending on
the situation. Merely shifting to even eighths will often be enough for a
drummer to imply a Latin groove. Bassists usually play a repeating
rhythmic pattern that emphasizes roots and fifths of chords. Additionally,
the bass commonly accents the first and third beats in a measure. Pianists
comp freely in even eighths, either repeating rhythmic patterns or
responding to a soloist.
Groove topics may be used to provide unique analytical insights into
collectively improvised performances. As an example, I will investigate a
recording of a live performance given by the Miles Davis Quintet in 1964
of “My Funny Valentine.”26 This recording consists of almost all of the
members of his well-known second quintet: Herbie Hancock on piano,
Ron Carter on bass, and Tony Williams on drums, though with George
Coleman on tenor sax in place of Wayne Shorter. The tune itself plays an
important role in the expression of groove topics, so before examining the
performance I will turn to the tune.
Example 9-9, Lead Sheet of “My Funny Valentine”27
“My Funny Valentine” was composed by Richard Rodgers with words

by Lorenz Hart for their musical Babes in Arms (1937). Example 9-9
provides a “lead sheet” version of the tune, which imparts the melody, as
well as the harmonic framework in the form of chord symbols.28 The lead
sheet also delineates formal divisions. In the first two A sections, the tune
stays in its tonic key, C minor. The B section shifts to the relative major,
E-flat. The final A´ section returns to C minor but with an interesting
twist. Rather than continuing with eight-measure section lengths, four
extra bars are added to produce a concluding twelve-bar section. These
added four bars function tonally to effect a shift back to E-flat major, the
key of the B section. The tune’s form thus concludes away from tonic, in
E-flat major, a highly unusual quality in a jazz standard. There is yet
another interesting feature having to do with linear structure. In the B
section, an ascent from Bb4 to Eb5 is implied by the stepwise connections
between the first notes in mm. 17, 19, and 21. As the melody reaches
closer to its E5 goal, the appearance of C5 in m. 23 suddenly interrupts
this motion. As the A´ section returns in the key of C minor, a series of
new linear ascents begins. These ascents culminate with the achievement
of the E5 desired earlier, though undercut by C minor harmony. In light
of this expressive crux of the tune, the added four measures may be
regarded as a conciliatory gesture, bringing back Eb major, though notably
without the high E5. Davis’s quintet interacts with these unusual tonal,
186 Chapter Nine
formal, and linear aspects of the tune as they collectively improvise their
performance.
This analysis will investigate Davis’s solo that opens the recording.
Throughout this investigation of the various grooves used in the recording,
refer to Example 9-10. This chart maps out the groove topics each
musician projects, and indexes their appearances to time points in the
recording as well as locations in the tune’s form. In the example, each
groove is labeled by an abbreviation given in the key to symbols. Three
additional markings are important to note. First, grooves followed by a
question mark are suggested by a performer but do not emerge fully
throughout the whole ensemble. Second, dotted lines indicate the span of
time in which each performer projects each groove. Third, snippets of text
appear above these lines to highlight a noteworthy element of a player’s
expression of a groove.
The performance begins with an “out-of-time” introduction played by
Hancock. Jazz musicians call these sections “rubato,” a sense of the term
that conflicts somewhat with the term’s meaning in the Classical
repertoire. Rather than a temporary relaxation of tempo, rubato in jazz
often refers to extended passages that have no metric pulse. The musicians
interactively cue chord shifts in these passages. While rubato is not a
“groove” in the sense defined above, it often contains qualities of the
ballad groove due to its feeling of slower unfolding.
Davis enters at 0:30 and initiates the tune proper. Hancock and Davis
perform the A section together, merely hinting at aspects of the tune’s
harmony and melody. After stating the first four measures of the melody,
Davis departs from the melody, never to return to it in his five-minute
solo. Carter’s entrance at 0:59 cues the first real groove of the
performance. His articulation of beats 1 and 3 along with an emphasis on
the chordal root and fifth suggests a Latin groove, though with a triplet
feel. Williams, however, snaps the hi-hat on beats 2 and 4 setting up a
slow ballad. Due to the slow tempo and lack of additional support in the
drums, ballad takes precedence over Latin, but Carter’s utterances do give
the music a Latin inflection. Williams strengthens the ballad feel at 1:15
by beginning the ballad “stir” on the snare. At the same time, Carter halts
the root-fifth motion and takes up the usual ballad style. When the
musicians reach the B section at 1:31, Hancock begins to inflect the slow
ballad groove with swung double-time rhythms. The tune’s active
harmonies at 1:50 inspire further double-time hints from Hancock, with
his active rhythms, and Carter, with his walking quarter notes. Double
time does not immediately emerge, however, as the two immediately
sustain their notes following this outburst.
Example 9-10, Map of Groove Topics in “My Funny Valentine”

188 Chapter Nine
At 2:02 the musicians return to the Latin-inflected ballad groove from

the earlier passage at 0:59. Hancock’s sudden utterance at 2:21 seems to
inspire Davis to move unexpectedly into the high range. The following
passage sounds like it is just on the cusp of a change in groove, with all of
the musicians listening very intently to one another to see who will step in
and provide a definitive cue for a shift. Davis’s shriek at 2:35 and
subsequent swing figure offers the expected cue, and the musicians
gradually work their way into a double-time swing groove. An interesting
aspect of this shift is that it occurs during the “extra” four bars at the end
of the tune’s final A´ section. The musicians thus use these unorthodox
additional measures as a kind of pivot point, allowing a shift of groove to
take place between the first and second choruses of Davis’s solo.
When the second chorus begins at 2:52, the musicians are “in full
swing,” with all of its attendant increase in energy. Williams performs the
standard ride cymbal pattern, Carter walks, and Hancock comps
rhythmically. At 3:18 Williams gives a brief hint at straight eighth notes, a
hint of Latin that, for the time being, goes unanswered. At 3:26, however,
a subtle shift in groove does occur. Davis’s solo line becomes soft, yet
swings intensely. His eighth notes move toward the right side of the
spectrum provided in Example 9-1. Williams follows suit and adds a
similarly hard-swinging rhythm on his tom-toms. Hancock’s comping
suggests more of a riff-based style rather than the rhythmically freer stance
he had earlier adopted. All of these elements combine to create an
intensification of the swing groove, despite the lowered dynamics of this
passage. Indeed, the quietness of their utterances serves to make the
groove even more powerful; the very act of producing such a powerfully
swinging groove at such a soft dynamic level adds a whole new dimension
to the passage. The musicians depart from this soft and intense swing at
3:42, returning to the more typical swing groove they previously
employed.
Nearing 3:58, a succession of straight eighths in Williams’s part
suggests a shift to Latin. Following his utterance, the group shifts almost
simultaneously to a Latin groove, a shift that coincides with the beginning
of the tune’s B section. The use of the more relaxed Latin feel here
complements many aspects of this musical moment. Tonally, the tune
shifts from the minor mode to its relative major. Combined with the shift
in groove, the music takes on an otherworldly character, as if suddenly
imported from somewhere else. Davis supports this feeling with sustained
pitches, although interrupted by strange harmonic divergences. Carter
sustains a repeated pedal throughout much of this passage, and Hancock
holds sustained chords. Between 4:17 and 4:28, swing gradually creeps
back in as the tune shifts away from E-flat major. Swing reemerges with C
minor at 4:28.
The swing groove that appears at 4:28 is again inflected with the
various intensifying devices used in the passage at 3:26. Whereas the shift
from this soft-and-intense swing passage to Latin was mediated by a
passage of standard swing at 3:42, now the two are directly juxtaposed,
and their conflicting expressive correlations meet abruptly. Following
4:58, Williams begins to hint at straight eighths in the same manner he did
previously. Latin reemerges at 5:06, and nicely supports Davis’s more
easygoing sustained utterances along with the brief return of major-mode
harmony. Formally, the musicians again find themselves in the extra four
measures added to the A´ section, the strange point at which E-flat major
returns to conclude the tune. This last vestige of relaxation and stasis in
Latin-major does not get the final word, however, as an abrupt ascent in
Davis’s part cues a shift back to swing-minor. As a result of this return to
swing, Davis’s solo actually extends by four (double-time) measures into
George Coleman’s, which follows directly. Thus the topical needs of the
groove supplant standard jazz practice here, that of concluding one’s solo
before the start of the next player’s chorus.
To summarize, Davis’s solo is shaped primarily by the shift from the
beginning ballad groove into double-time swing. This shift brings about a
concomitant increase in energy and excitement. Once this primary shift
occurs, however, Davis inflects the prevalent swing groove into an
unusual and intense soft swing. Additionally, the musicians contrast
minor-key swing in Davis’s second chorus with major-key Latin. The
contrasting expressive correlations of these grooves create a great deal of
variety and pose a problem: which one will ultimately win out? The
answer to this question is not quite as simple as it might appear, due to the
fact that E-flat major returns at the end of the tune. Despite the brief return
of Latin for this final major-key swerve, minor-swing has the last word,
notably resulting in a bit of spill over into Coleman’s solo. The ensemble’s
use of the competing swing and Latin grooves also parallels the linear
tension of the tune itself. Latin groove, the goal of the second chorus’s B
section, does not emerge as a permanent change, but rather succumbs to
swing as the minor-key A´ section is reached. Latin reappears briefly with
the final E-flat major phrase, but this achievement is undercut by a fall
back into swing. Thus, just as the linear ascent to E5 is undercut by C
minor harmony, so is the Latin groove’s attempt to conclude the second
chorus.
The groove topic allows us to refocus jazz analysis on a set of musical
parameters often overlooked by jazz analysts. This emphasis on groove
190 Chapter Nine
and the ways in which it is interactively cued coincides closely with the
musical value placed on these parameters by jazz musicians. The
compendium of grooves discussed here is by no means exhaustive. There
are undoubtedly more expressive correlations to be discovered in these
grooves as well. In its relatively short history, topic theory has revealed
aspects of Classical style to which music theorists were formerly less
attuned. Similarly, it has the power to reveal expressive aspects of
improvised jazz that analysts have neglected to highlight. In this way,
topic theory shows that, whether through dance or groove, musicians
separated by hundreds of years have always been interested in making
music move.
Notes
1
Ingrid Monson, Saying Something: Jazz Improvisation and Interaction (Chicago:
University of Chicago Press, 1996), p. 29.
2
See, for instance, Henry Martin, Charlie Parker and Thematic Improvisation
(Lanham, MD: Scarecrow Press, 1996) and Steve Larson, Analyzing Jazz: A
Schenkerian Approach (Hillsdale, NY: Pendragon Press, 2009). I do not mean to
denigrate the Schenkerian approach adopted by both Martin and Larson, but rather
I intend to emphasize that such an approach often implicitly values tonal coherence
over other parameters that may be foremost in the minds of musicians, particularly
during the moment of a performance act.
3
Monson, p. 29.
4
Leonard Ratner, Classic Music: Expression, Form, and Style (New York:
Schirmer Books, 1980), p. 9.
5
Ibid., p. 9.
6
Wye Jamison Allanbrook, Rhythmic Gesture in Mozart (Chicago: University of
Chicago Press, 1983), p. 8.
7
Owens describes the bebop style as “the lingua franca of jazz;” see Thomas
Owens, Bebop: The Music and Its Players (New York: Oxford University Press,
1995), p. 4. While numerous small-group styles have emerged since the 1940s,
bebop serves as the basis from which these other styles depart or against which
they are defined. To a significant extent the term “jazz” refers to this “lingua
franca,” while the term “swing” is used when discussing the earlier style.
8
Guitarists, while quite common in swing rhythm sections, became somewhat less
common in modern jazz. That is not to say that the guitar was an aberration in
bebop, but rather that the typical rhythm section usually included only one chordal
instrument, and more often than not that instrument was the piano.
9
Don DeMichael and Alan Dawson, A Manual for the Modern Drummer (Boston:
Berklee Press Publications, 1962).
10
Monson, p. 67.
11
Lawrence Zbikowski, “Modeling the Groove: Conceptual Structure and Popular
Music,” Journal of the Royal Music Association 129/2 (2004): 282.
12
Charles Keil, “Participatory Discrepancies and the Power of Music,” Cultural
Anthropology 2/3 (1987): 275–83.
13
Ibid., p. 275.
14
Monson, p. 68.
15
J. A. Prögler, “Searching for Swing: Participatory Discrepancies in the Jazz
Rhythm Section,” Ethnomusicology 19/1 (1995): 21–54.
16
Matthew W. Butterfield, “The Power of Anacrusis: Engendered Feeling in
Groove-Based Musics,” Music Theory Online 12/4 (2006).
17
Robert Hatten, Musical Meaning in Beethoven: Markedness, Correlation, and
Interpretation (Bloomington, IN: Indiana University Press, 1994), p. 30.
18
Butterfield 2006.
19
Benadon uses the term “beat-upbeat ratio,” or BUR, to describe this eighth-note
inequality; see Fernando Benadon, “Slicing the Beat: Jazz Eighth-Notes as
Expressive Microrhythm,” Ethnomusicology 50/1 (2006): 73–98.
20
Benadon 2006 offers empirical evidence in support of this assertion.
21
As described in Paul Berliner, Thinking in Jazz: The Infinite Art of Improvisation
(Chicago: University of Chicago Press, 1994), p. 315.
22
See Larry Finn, Beyond the Backbeat: From Rock and Funk to Jazz and Latin
(Boston: Berklee Press Publications, 2000), p. 11.
23
Of course, rock grooves also feature even eighth notes, and they begin to be used
in jazz performances starting later in the 1960s. Rock and Latin grooves may be
distinguished from one another based on their unique rhythmic and
accompanimental patterns used by the rhythm section.
24
The association between Latin grooves and a relaxed expressive state stems
mainly from the prevalence of the bossa nova in jazz. Bossa nova arose as a cross
pollination between Brazilian samba rhythms and jazz harmonies during jazz’s
turn toward the “cool sound” in the 1950s, and thus reflects this subdued character.
For more on the bossa nova, see Chris McGowan and Ricardo Pessanha, The
Brazilian Sound: Samba, Bossa Nova, and the Popular Music of Brazil
(Philadelphia: Temple University Press, 2009).
25
As shown in Finn, pp. 32–33.
26
Miles Davis, The Complete Concert: 1964 (My Funny Valentine + Four &
More) (Sony 4712462, CD, [1964] 1993).
27
In this lead sheet, “-” indicates minor triads, “‫ ”ټ‬major, and “ø” half-
diminished. Superscript numbers add upper extensions that may be altered by
accidentals.
28
This version of the lead sheet is adapted from Howard Brofsky, “‘My Funny
Valentine’: The Evolution of a Solo,” Black Music Research Journal 3 (1983): 37.
CHAPTER TEN
THE MUSIC OF MARIO, LINK, AND SAMUS:

HARMONY, FORM, AND MEANING
IN EARLY COMPOSITIONS
BY KOJI KONDO AND HIROKAZU TANAKA
NATHAN BAKER
The main theme from Super Mario Bros., written by Japanese video
game composer Koji Kondo, is one of the best-known pieces of music
composed during the past three decades. The video game industry is the
fastest-growing source of jobs for composers, according to ASCAP,1 and
leading video game composers, such as Kondo, Nobuo Uematsu, and
Yasunori Mitsuda, are acclaimed for their music across the globe. Given
the prominence of video games (and their music) in today’s culture, it is
somewhat surprising that limited academic attention has been paid to
video game music.2 A study of this music, which on first listening sounds
so familiar to western ears, reveals a number of harmonic and formal
characteristics that distinguish it from traditional western music in
intriguing ways. In this chapter, I will analyze pieces of music from three
early video games for the eight-bit Nintendo Entertainment System—
Super Mario Bros. and The Legend of Zelda, composed by Koji Kondo,
and Metroid, composed by Hirokazu Tanaka—and discuss their prominent
harmonic and formal characteristics. I will also examine how the harmony
and form of these pieces contribute to their meaning within the context of
the games in which they are featured.3
Koji Kondo was the lead composer for Nintendo, and is best known for
his work in Nintendo’s flagship game lines Super Mario and Zelda. We
shall start with Kondo’s aforementioned main theme from Super Mario
Bros. (“Overworld/Main Theme”). Over the years, I have asked several
colleagues if they can quickly identify the form of this theme. Almost
The Music of Mario, Link, and Samus 193
inevitably, they hum through a few phrases, reach a repeat of the A

section, and answer that it is a looping binary form. When I then have
them listen to a recording of the piece, many are surprised by the subtle
complexity of its actual form. Examining the harmonic characteristics of
this piece phrase by phrase will help to arrive at a view of its overall form.
As shown in Example 10-1, the theme begins with a brightly
syncopated introduction that ends on a distinctly inconclusive cadence.
Since the initial melodic notes clearly reveal the key to be C major
(confirmed by the first phrase beginning in m. 3), we are able to recognize
the opening chords as V9/V followed by V in C. Kondo, a composer who
has stated that jazz, fusion, and Latin styles are major influences on his
idiolect,4 often uses the extended chords native to those styles, particularly
9th and added 6th chords.
Example 10-1, Super Mario Bros., “Overworld / Main Theme,” Introduction, mm.
1–3
The first phrase, four measures in length, follows a relatively standard

tonal progression, with the V in m. 4 behaving more like a neighbor
prolongation of the IV chords than a functional dominant. This phrase then
repeats, yielding an A section ending on a half cadence (see Example
10-2).
194 Chapter Ten
Example 10-2, Super Mario Bros., “Overworld / Main Theme,” Section A, mm.
3–6
The B section (Example 10-3) begins in m. 7 with a simple four-

measure progression. The answering four-measure phrase that concludes
the B section, however, offers what seems from a western classical
perspective to be an unusual progression involving flattened submediant
and subtonic harmonies. This particular modal borrowing commonly
occurs in Japanese video game music,5 with the major or major-minor
seventh quality subtonic harmony serving as a functional dominant. I
suspect that the common appearance of this harmony in video game music
stems partly from the strong influence of jazz and rock music on Japanese
video game music composers.6 At any rate, we are well served by stating
that the B section ends with the video game music analogue to an
authentic cadence.7 The theme then repeats both phrases of the B section.
Example 10-3, Super Mario Bros., “Overworld / Main Theme,” Section B, mm.
7–14
The C section (Example 10-4) begins in m. 15. The progression heard

in the first four-measure phrase helps to confirm the borrowed cadential
figure ending the previous phrase. (Colleagues who initially assess the
form of this piece as a looping binary form tend to describe this C section
either as a coda or linking transition to set up the loop back to the
beginning, so extending and confirming the “authentic” cadence makes
sense). The second phrase of the C section begins with a similar borrowed
cadential progression, but the last two measures are identical to the two-
measure introduction that started the theme. This traditional half cadence
(V9/V moving to V) strongly sets up the expectation that the A section will
return as we loop back to the beginning, particularly since it is identical to
the original introduction of the piece. The following music clearly signals
the return of the A section, thus confirming listener expectations.
196 Chapter Ten
Example 10-4, Super Mario Bros., “Overworld / Main Theme,” Section C, mm.
15–22
If we continue listening to the music past this “return to the

beginning,” however, we come to the feature that inevitably surprises my
colleagues when I ask them to identify the piece’s form: whereas before
we heard the B section following the A section, this time Kondo provides
entirely new musical material! Compare the cadence ending this D section
(Example 10-5) to the borrowed cadence ending the B section. Like the B
section, the D section repeats itself, and is followed by the C section.
Recovering from the initial surprise of hearing this D section, one could be
very tempted to return to the looping binary form theory, with the
provision that every other loop alternates between the B section and D
section. Kondo, however, has one more surprise in store: this latest C
section does not return to the A section as expected. Instead, Kondo inserts
another reprise of the D section before returning to the A section, finally
looping the entire piece.
Example 10-5, Super Mario Bros., “Overworld / Main Theme,” Section D, mm.
27–34
The overall form of the “Overworld” theme, then, is not simple and
predictable at all: [intro AA BB C AA DD C D]. The overall familiarity of
the music helps it stay in the background as a non-distraction as the player
plays the game, while the surprising formal elements provide enough
variation to keep it from becoming too monotonous and repetitive (which,
considering that this musical theme is constantly looped during 19 of the
game’s 32 levels, is an important factor indeed).
From the analysis presented here, it seems that early Nintendo
composers took advantage of the different psychological effects created by
combining subtle variation with sheer repetition to provoke a desired
mood in the player matching the character of the level. Compare the long
and formally complex loop of the main “Overworld” theme to the
“Underground” theme (Example 10-6). Notice the sparse musical
elements of the latter theme: two one measure re-ti-do statements in B-flat
major (barely enough to establish the tonic), two measures of the same
statement transposed to the subdominant, a highly dissonant liquidation
that ends with a chromatic linearization of VII (see Example 10-7), an
odd measure of silence at the end disrupting the expected hypermetrical
198 Chapter Ten
pattern, and frequent looping due to the shortness of the musical line
(which induces the ostinato effect). Kondo often uses these stylistic
techniques (fragmentation, chromaticism, liquidation, hypermetrical
distortion, and ostinato) when he composes music for underground
environments. The result has a distinct psychological effect on the player
of the game, creating a sense of tension and claustrophobia that matches
the game’s current environment.
Example 10-6, Super Mario Bros., “Underground”
Example 10-7, Super Mario Bros., “Underground,” Reduction of mm. 5–6
We can also see this stylistic effect at play in the “King Koopa’s
Castle” theme (Example 10-8), which is even shorter and more fragmented
than the “Underworld” theme. Notice the rapid chromatic tremolo, the
lack of a sense of tonality, and the way the form of the piece elides the
expected end of the phrase with the beginning of each loop. It is

noteworthy that levels featuring these two themes appear in the game
much less frequently than those featuring the main “Overworld” theme. Of
the 32 game levels, eight take place in King Koopa’s castle, and only two
levels feature the “Underworld” theme for the entire level (although the
theme does make temporary appearances in brief underground shortcuts or
bonus coin rooms contained within other levels).
Example 10-8, Super Mario Bros., “King Koopa’s Castle”
The three main themes from The Legend of Zelda, also composed by
Koji Kondo, feature many similarities to the three previously analyzed
themes from Super Mario Bros. Compared to the main “Overworld”
theme from Super Mario Bros., the main “Overworld” theme from The
Legend of Zelda contains even more appearances of the borrowed
submediant and subtonic chords. Example 10-9 is representative of the
piece as a whole, which uses the melodic minor scale (with le and te) as
much as it does the Ionian major scale.
200 Chapter Ten
Example 10-9, The Legend of Zelda, “Overworld,” Introduction, mm. 1–4
The “Overworld” theme from The Legend of Zelda has a much simpler
form than the “Overworld” theme from Super Mario Bros., but it features
a similarly surprising formal alteration after the listener thinks the piece
has first looped. The last measure of the first section (see Example 10-10)
is essentially the same as the last measure of the introduction. Thus, when
the second section (Example 10-11) begins with the same musical material
as the first, we are led to believe that we have already experienced the
loop. Beginning in the third measure of the second section, however, the
theme suddenly changes, and by the fifth measure an unexpected German
augmented sixth chord has started a four-measure internal expansion that
distorts the hypermeter, before the second section finally presents the
expected last measure that prepares the actual loop back to the beginning
of the first section. The resulting form is: [intro–A–A´], which is shorter
and simpler than the corresponding theme from Super Mario Bros., but
here the listener is perhaps more distracted because The Legend of Zelda
focuses more on exploration and free traversal across the game world
when compared to Super Mario Bros., which is a limited side-scrolling
game. The increased freedom and complexity of the gameplay in The
Legend of Zelda lets Kondo get away with less musical complexity. I also
suspect that the game itself takes up more of the limited cartridge memory
in The Legend of Zelda than it does in Super Mario Bros., leaving less
memory for musical themes.
Example 10-10, The Legend of Zelda, “Overworld,” Section A, mm. 5–12

202 Chapter Ten
Example 10-11, The Legend of Zelda, “Overworld,” Section A´, mm. 13–24
Turning now to the “Underworld” theme (Example 10-12), notice the

chromatically descending tetrachord from G to D in the middleground of
both soprano and bass parts. This dissonant ostinato created by short notes,
as well as the liquidation in the metrically unexpected 5/4 measure at the
end of the theme, is common in Kondo’s claustrophobic underground
style.
Example 10-12, The Legend of Zelda, “Underworld”
The “Death Mountain” theme plays with unexpected metrical elements

even more than the previously analyzed examples.8 At the start of the
piece, the relative melodic durations create the sense of a slightly
elongated compound meter (see the renotation in Example 10-15). When
the subsequent chromatic triplets start, however, the listener suddenly
loses all sense of the meter previously perceived and is set adrift until the
original melody resumes. The prominence of the tritone (both in the
parallel tritone lines of the melodies and in the tritone transposition
between the two sections) only further contributes to the psychological
discomfort experienced by the player, ratcheting up the tension for this
climactic point in the most dangerous dungeon level of the game.
204 Chapter Ten
Example 10-13, The Legend of Zelda, “Death Mountain,” Section A, mm. 1–4
Example 10-14, The Legend of Zelda, “Death Mountain,” Section A´, mm. 5–8
Example 10-15, The Legend of Zelda, “Death Mountain,” Metrical Renotation of

m. 1
Kondo musically highlights what is happening in these games in a

couple of other ways. The “Star” theme from Super Mario Bros. is a short
piece of music consisting entirely of the two-measure vamp shown in
Example 10-16. The extreme repetition and consonant jazzy harmonies
underscore the fact that the star power-up makes Mario invincible,
essentially suspending the normal rules of the game world and allowing
the player to run through enemies until the effect wears off.
206 Chapter Ten
Example 10-16, Super Mario Bros., “Star”
Another musical effect paralleling the gameplay takes place in the

underwater levels. The player must continually press the jump button to
bob up a short way through the water, while some force (gravity or
suction, the game does not explain which) constantly pulls Mario back
down toward the bottom of the screen. Portions of the “Underwater”
theme reflect Mario’s struggle through the water, with repeated notes in
the melody that “bob up” a step before being pulled back down by the
constantly descending lines notated in Example 10-17.
Example 10-17, Super Mario Bros., “Underwater,” mm. 20–29
Having examined Koji Kondo’s two most important early games for
the Nintendo Entertainment System, let us proceed to another video game
composer, Hirokazu Tanaka, and look at some of the pieces that he
composed for another important early Nintendo game, Metroid (which is
also one of Nintendo’s flagship series, although Tanaka was not involved
with any of the sequels). Hirokazu “Hip” Tanaka was the main sound
engineer at Nintendo, mostly working with programming and sound

effects, although he also composed the music for several games. Metroid
is a game set in space. The player controls a bounty hunter sent out to an
isolated planet to fight space pirates and some mysterious life form that
threatens all other life in the galaxy. The music reflects this sense of
isolation with its heavy use of sound effects (much more so than Kondo’s
soundtracks, which mostly stick to musical notes and percussion) and
dissonance to lend a more modern, futuristic feel to the soundtrack.
Tanaka, like many video game composers who entered the field from the
electronic side of the industry, works a lot with developing variation,
adding new layers to an initial theme in a continuous process similar to a
passacaglia.
The first of Tanaka’s pieces examined here is the introduction to the
“Title Screen” theme from Metroid (Example 10-18). What the
transcription cannot impart is the addition of some distinctive sound
effects to the music: the timbre of the low bass pedal creates an ominous
deep buzzing sound, and the highest voice features a crystalline timbre
with an added pulsating echo effect. To me, the overall sound is something
like cold starlight twinkling over the background radiation of deep space
(which matches the image of a desolate planetscape with twinkling stars in
the sky shown on the screen).
Example 10-18, Metroid, “Title Screen,” Introduction, mm. 1–8
After the introduction repeats, Tanaka begins layering voices together.

The first eight measure section (Example 10-19) features the echoing
crystal voice descending in octaves, another voice presenting a slow-
moving theme in whole notes, and a simulated snare drum in the white
noise channel. Note as well the shift from the Bb of D Aeolian to a
brighter-sounding B-natural. Coupled with the more active rhythmic
208 Chapter Ten
layers, the effect is much more hopeful than the stark loneliness of the
introduction.
Example 10-19, Metroid, “Title Screen,” Section A, mm. 9–16
The next variation of this theme adds a new melody in a type of

counterpoint to the whole-note line (Example 10-20). The melody is then
altered in the subsequent eight-measure iteration of the theme (Example
10-21).
Example 10-20, Metroid, “Title Screen,” Section B, mm. 17–24

210 Chapter Ten
Example 10-21, Metroid, “Title Screen,” Section C, mm. 25–32
The final development of the theme (Example 10-22) retains the

essence of the descending line with octave leaps in the top voice, but the
meter suddenly accelerates into 6/8 with a distinct increase in overall
rhythmic activity, culminating in a liquidation of the descending line over
a 6/4–5/3 resolution. The buzzing low pedal D then returns for four
measures before the piece loops back to the introduction.
Example 10-22, Metroid, “Title Screen,” Section D, mm. 33–42
Another clear example of this developing variation technique can be

heard in the theme from “Brinstar,” the first major area of the game
(Example 10-23). As with the central variation section from the “Title
Theme,” Tanaka begins with three voices: a simulated snare drum, a
whole-note line that will later become the countermelody, and a line to
establish the harmony (this time a rhythmically active bass line instead of
the high descending line of the previous piece). Compare this first two-
measure section (which is repeated once) to the following two-measure
section (which is repeated three times before continuing to a two-measure
transition into the next development). The underlying structure of the bass
line and countermelody remain the same, with embellishments added onto
this structure, and a melody voice has also been introduced to the texture.
212 Chapter Ten
Example 10-23, Metroid, “Brinstar,” Section A, mm. 1–6
The next variation changes the pitch center from G to C, and develops
the previous iteration by adding more melodic embellishments to the
melody and countermelody, with the countermelody having been
transposed up a fourth to match the new key (Example 10-24).
Example 10-24, Metroid, “Brinstar,” Section B, mm. 7–10
The next variation shifts key again, this time to B-flat, and greatly
transforms both the melody and especially the countermelody (Example
10-25). The triplet figure now saturates the melody line, and the
countermelody moves on nearly every beat.
Example 10-25, Metroid, “Brinstar,” Section C, mm. 11–14
The last part of the “Brinstar” theme (Example 10-26) rapidly cycles
from G major through Ab Lydian to a Badd6 chord, and finally to a D
major area (with a pedal A in the bass) that ultimately serves as the
dominant to loop back around to the original G tonality (foreshadowed by
the arpeggiated G chord in the countermelody). The liquidating effect of
this section is also enhanced by the increased melodic activity in the bass
line (which has been fairly static in previous sections), as well as the
abruptly static nature of all three lines in the final two measures before the
loop.
Example 10-26, Metroid, “Brinstar,” Section D, mm. 15–20

214 Chapter Ten
Some of Tanaka’s developing variation music is quite a bit sparser

than the previous examples, as is the case with the theme from “Kraid’s
Lair,” a section of the game leading up to a boss fight with Kraid, an alien
space pirate (Example 10-27). Here there are only two voices, a melody
and a bass line. The static E minor pedal of the opening two measures is
developed in the second section by transforming the leaping melodic
motives over a VI–VII progression. The next development again
increases the rate of change of the bass notes (now changing every beat
instead of every two beats), and transforms the leaping motive of the
melody into alternating sixteenth notes. The bass line changes twice per
beat in the following variation, while the sixteenth note figure is further
developed into a swinging arc of scales and arpeggios. The final section
provides the liquidating effect through a sudden cessation of motion in the
bass line (with the deep buzz sound effect from the introduction returning)
as the melody is constrained to a single four-note scale bounded by the
same pitches that enjoyed metric prominence in the first measure of the
melody: E and A. Notice as well that the swing feel so prevalent
throughout the rest of the piece has been completely eliminated by the
regularity of the eighth notes over the static bass pedal.
Example 10-27, Metroid, “Kraid’s Lair”

The final piece to be analyzed in this chapter is an example of Tanaka

using the underworld style previously discussed in Kondo’s music. This
theme, “Norfair,” plays in the background as the player explores a
dangerous lava-filled area deep beneath the surface of the planet (Example
10-28). Notice the short, fragmented nature of the melodic ideas, the
dissonance of the parallel major seventh chords in third inversion that
begin this piece, the metrical distortion created by the extra eighth note
rest added to create the 7/8 measures, and the subtle effect in the last
section produced by the simulated snare drum continuing to play its 6/8 +
7/8 pattern while the other voices eliminate the 7/8 grouping. The player,
who has by now gotten used to the hiccup of the 7/8 measure,
subconsciously perceives the unexpected metrical change, but is unlikely
to realize exactly what has happened (particularly since the bulk of the
player’s attention is undoubtedly focused on not getting killed!). The
resulting cognitive dissonance produces a distinct feeling of discomfort in
the player that heightens the sense of danger present in this challenging
part of the game.
Example 10-28, Metroid, “Norfair”

216 Chapter Ten
Some general characteristics that have emerged from the analysis of

these pieces by Koji Kondo and Hirokazu Tanaka include: a unique
harmonic language with a distinct use of modal characteristics,
particularly as represented by the stepwise VI–VII–I borrowed
progression; the use of formal variation to help minimize the potential for
tedium stemming from the forced reliance on relatively short loops of
music due to the memory limitations of the cartridge; and compositional
techniques, such as the underworld style, used to musically portray what is
happening in the game, thereby enhancing the experience for the player.
This chapter has only scratched the surface of harmony, form, and
meaning in video game music, focusing as it did on a limited selection of
three games, two composers, and the span of a single year of video game
music history.9 Hopefully this brief examination has demonstrated the
wealth of interesting musical features to be heard in the popular and
increasingly influential genre of video game music.
Notes
1
Todd Brabec and Jeff Brabec, “Licensing Songs for Video Games,” The ASCAP
Corner, accessed August 31, 2012. <http://www.ascap.com/Home/Music-
Career/articles-advice/ascapcorner/corner16.aspx>
2
Karen Collins is responsible for much of the extant research on video game
sound, having written one book on the subject and having edited another. See
Karen Collins, Game Sound: An Introduction to the History, Theory, and Practice
of Video Game Music and Sound Design (Cambridge, MA: MIT Press, 2008); and
Karen Collins, editor, From Pac-Man to Pop Music: Interactive Audio in Games
and New Media (London: Ashgate Publishing, 2008).
3
I highly recommend that the reader listen to recordings of these pieces as they are
discussed (particularly if the reader has not played the games), each of which is
easily accessible via YouTube.
4
The best reference to Kondo’s love of Latin and jazz is probably his 2005
interview with Mark MacDonald of Electronic Gaming Monthly. See Mark
MacDonald, “Interview with Koji Kondo,” Electronic Gaming Monthly, May
2005.
5
I am grateful to Peter Shultz for pointing out, in a personal correspondence, what
he calls the “victory progression,” which is prominently displayed in the fanfare
played at the end of each successfully completed level in Super Mario Bros. It also
makes frequent appearances in music by a wide variety of prominent Japanese
video game composers such as Hirokazu Tanaka, Yasunori Mitsuda, and Nobuo
Uematsu. I first noticed it as the last three chords of the “Prelude” from the Final
Fantasy series, composed by Uematsu.
6
Several attendees at the 2010 West Coast Conference of Music Theory and
Analysis shared some very useful comments regarding the use of this chord
progression in rock music of the 1960s and 1970s, for which I am greatly
appreciative.
7
Christopher Doll has named this particular subtonic function in rock music the
“rogue dominant.” See Christopher Doll, “Listening to Rock Harmony” (Ph.D.
Dissertation, Columbia University, 2007). I will discuss the theoretical
underpinnings and role of the bVII–I cadence more extensively in the doctoral
dissertation on which I am currently working.
8
I highly encourage the reader to conduct along with a recording of this piece in
order to experience the same surprise (and frustration!) that I did when I first
transcribed it (hint: listen to the melody, then count the pulses in the bass per
melodic note).
9
Super Mario Bros. was originally released on September 13, 1985; The Legend of
Zelda on February 21, 1986; and Metroid on August 6, 1986.
CHAPTER ELEVEN
HEARING HEIMA:
ECOLOGICAL AND ECOCRITICAL
APPROACHES TO MEANING
IN THREE ICELANDIC MUSIC VIDEOS
BRAD OSBORN
Múm’s “Green Grass of Tunnel” (2002), Björk’s “Triumph of a Heart”

(2004), and Sigur Rós’s “Glósóli” (2005) are music videos composed by
Icelandic recording artists in the last decade. Each of these pieces, in very
different ways, offers a commentary on the Icelandic ecosystem.1 By
examining concomitantly the cinematic and musical elements of these
three Icelandic music videos as they relate specifically to a place-centered,
ecological view of Iceland, this essay hopes to demonstrate one way in
which analysis of music videos reveals far more than can be gleaned from
recorded music alone.
Although the analysis of music videos has not received much attention
in the music-theoretical community, Nicholas Cook, following Goodwin
and perhaps more surprisingly, Schoenberg, has argued successfully that
music videos should, in their essence, be interpreted as musical entities:
A ‘musicology of the image’ [Goodwin’s term] would seek to interpret the

music video as, before anything else, a ‘musical entity’ [Schoenberg’s
term]...it would understand it as making music with the media of the
video...it might be possible to work from fairly basic music-theoretical
concepts toward an understanding of the relationship of music to words
2
and pictures.
My approach, like Cook’s, will be fundamentally driven by analysis of the

music, but with an additional toolset outside the bounds of what he would
Hearing Heima 219
consider “basic” music theory. Drawing on recent work in the fields of

ecomusicology and ecological perception will uncover fundamentally
Icelandic elements in the music videos.3 Prior to the analysis, a brief
introduction of ecologically inspired music criticism will help to situate
this work within what appears to be a rapidly expanding field of
scholarship,4 and a terse history of the modern Icelandic popular music
scene and its relationship to Iceland’s physical and socio-economic
landscape will provide context for these works.
Ecological approaches to music and music analysis have only been
emerging in the last two decades, and could be placed in two broad
categories.5 The first, often called “ecomusicology” or “ecocriticism,”
tends to concern itself with approaches to analyzing music that situate said
music within a particular place, as well as that place’s attendant cultural
milieu. Though one might argue that this has been a concern of music
historians for a long time, ecocriticism received significant attention in the
wake of the New Musicology movement, and Denise Von Glahn’s 2003
book has certainly stood as a benchmark for the field.6
The second involves a relatively new field of music psychology known
as “ecological perception.” Pioneered by J.J. Gibson, who was concerned
with visual perception, several articles on auditory perception that adapted
his work to sound appeared in journals in the 1990s, and Eric Clarke’s
2005 book forms the locus classicus on the topic.7 Allan Moore, in his
adaptation of Clarke’s theory to popular recorded song, sums up the
approach nicely: “invariants afford through specification.”8 Invariants are
the inherent properties of a sound, both physical (i.e., by nature of material
construction) and cultural (i.e., the use of drum in ritual) that specify either
the sound’s source or its reference to a specific group of listeners. In
Clarke’s view:
...just as sounds specify the invariants of the natural environment, so too do

they specify the constancies or invariants of the cultural environment. The
sounds of a muffled drum being struck with wooden sticks specify the
materials (wood, skin) and physical characteristics (hollowness, damped
vibration) of the material source—the drum; and they also specify the
9
social event (for instance, a military funeral) of which they are a part.
By analyzing invariants and the sources they specify, ecological

perception aims to account for what interpretations of meaning these
sounds afford and, perhaps just as importantly, which interpretations they
do not afford. Put together, these three concepts help us to solve a
perennial problem involving the meaning of a given piece of music: how
220 Chapter Eleven
do we account for the fact that, though each of us has our own unique
interpretation of a piece, there seems to be a commonly accepted range of
meanings shared between many listeners? Put differently, how do we find
a middle space between, on one extreme, pure, unbounded subjectivity
among individual subjects, and, on the other extreme, a single, inter-
subjective “encoded” meaning waiting to be “discovered” for each piece
of music?
Though one could theoretically apply these modes of analysis to any
body of music, Icelandic popular music—specifically its inextricable link
to modern conceptions of Icelandic culture and the Icelandic natural
landscape10—suggests the need for a more active link between ecology
and analysis. A landmark 2005 documentary Screaming Masterpiece
documents this integral link between the country’s musicians and the
natural features of the land.11 Three recurring themes mentioned by
musicians interviewed throughout the film seem particularly poignant in
this regard: (1) the geographical isolation of Iceland from neighboring
continents is isomorphic to the cultural isolation of the nation’s popular
music when compared to mainstream US or European styles; (2) the
geological wonders of the country, including geysers, glaciers, volcanoes,
mountains, geothermal hotpots, and vast lava fields marked by deep
fissures and deposited volcanic rock, are a continual aesthetic inspiration
for Icelandic artists of all disciplines; and (3) the relatively high amounts
of cold and darkness the country receives most of the year contribute to
the highly practiced and contemplative nature of the experimental art its
residents produce as they spend significant time indoors with only
artificial light. This sense of linking Iceland’s musical identity to place is
further reinforced by a 2007 film made by Sigur Rós themselves. The film,
entitled Heima (literally “home”), follows the band around the country as
they perform a series of free concerts, not only in cities, towns, and small
villages, but also in natural settings such as caves and open fields, most of
which were recorded live to imbue the film with the acoustic signatures of
those places.
Though Icelandic popular music was initially influenced a great deal
by British rock music in the 1960s, it has, especially in the last 15 years,
gained a distinctive and influential voice.12 The success of the Icelandic
popular music scene is undoubtedly bound with the immense international
success of Björk and Sigur Rós in the 1990s, which carved a space for
newer acts such as Múm, Mammut, Apparat Organ Quartet, and others in
the 2000s. Iceland’s impact on the modern experimental rock scene can be
gleaned best from the import of its yearly festival, Iceland Airwaves,
Hearing Heima 221
which, though it draws acts from around the world, highlights emerging
Icelandic rock artists. As evidenced by the festival’s 2012 lineup, many of
these artists are in fact so new that they have yet to release a full-length
record, and some do not even have record contracts. Seen internationally
as a hotbed of new, groundbreaking, experimental artists, major media
outlets flock to Reykjavík each year to broadcast from the festival.13 This
overwhelming international recognition has not only changed the face of
Icelandic music, but the music has in fact profoundly changed the current
social, political, and cultural climate of the country. As Dibben notes:
The success of Icelandic popular music abroad has a number of

consequences for national identity. First, the internationalism of Icelandic
popular music works against the idea of Iceland as a ‘peripheral’ nation
within a world context and demonstrates that it has a distinctive
contribution to make. Some Icelanders directly attribute their pride in the
Icelandic nation to its increased international profile within the popular
music industry... Second, the success of popular music has had direct
benefits for Iceland’s export economy, and indirect benefits for the tourist
industry... The cultural industries in Iceland are now a significant part of
the Icelandic economy, responsible for 4% of GDP, of which music
accounts for a quarter...As a consequence of this, state support for the
music industry increased with a reduction of tax charged on recorded
music (from 24.5% to 7%) in March 2007, and the creation in the same
year of the government-sponsored International Music Export...
With a view toward exposing how they contribute to this perceptible

link between Iceland’s popular music and the Icelandic ecosystem (which,
again, should be taken to stand for: the interactions of all organisms—both
human and non-human animals—and materials both natural and human-
made), I shall now undertake an analysis of the musical and visual
elements in three music videos by Björk, Sigur Rós, and Múm. The first of
these videos comments more directly on the artist’s interactions with the
Icelandic socio-cultural landscape, while the other two artists eschew self-
inclusion in their videos altogether to comment exclusively on the
Icelandic physical/geological landscape. Even between these last two, we
will observe a profound dichotomy between, in one video, the Icelandic
ecosystem as pastoral/lush, and in the other video, the Icelandic
ecosystem as imagined/harsh. Dichotomies such as these illustrate, and are
borne out of, the volatile and extreme geological contrasts that define and
shape the island.14
222 Chapter Eleven
Time Musical Form Time Cinematic Form

(video introduction not
0:01–0:27 heard on Medúlla) 0:01–0:38 SCENE A1 (home)
0:28–0:38 INTRODUCTION 0:39–0:53 TRANS (driving)
0:39–1:10 VERSE 1 0:54–1:22 SCENE B1 (bar, bored)
1:11–1:42 CHORUS 1 1:23–1:42 SCENE B2 (bar, happy)
1:43–2:24 REHEARSAL 1:43–2:46 SCENE B3 (bar, rehearsal)
2:25–2:57 VERSE 2 (live) 2:47–3:05 SCENE B4 (bar, party)
2:58–3:24 CHORUS 2 (live) 3:06–3:29 TRANS (walking)
3:25–4:02 BRIDGE 3:30–4:24 SCENE C (road)
4:03–5:25 CHORUS 3 4:25–5:25 SCENE A2 (home)
Example 11-1, Musical and Cinematic Form of Björk, “Triumph of a Heart”

(2004)15
Example 11-1 depicts the basic formal outline of Björk’s music video
for “Triumph of the Heart” from her 2004 album Medúlla. Note that, in
this type of representation, I have provided formal cues for both the
musical and cinematic elements of the song. Accounts of musical form
here are informed by recent theories of rock form, including Summach’s
work on conventional forms, and my own work on more recent post-
millennial formal designs.16 In order to relate the cinematic form of the
movie more closely to ecological theory, I emphasize place as the visual
parameter most responsible for delineating form. For example, scenes A,
B, and C in “Triumph of the Heart” occur in three different spaces (a
home, a bar, and a road, respectively), while variations on those cinematic
units can be further defined using numbers. These numbers may be
applied for two different reasons, as demonstrated in Example 11-1. In
scenes B1 through B4, the physicality of place is continuously present, but
the ecological interaction between humans and said physical space differs
from scenes one through four. In the case of scenes A1 and A2, the
presentation of a single physical space (the protagonist’s home, first
shown in A1) is separated in time by two sequential intervening spaces
(the bar, then the road), thus the arrival of A2 can be viewed as a
cinematic recapitulation.
A brief plot synopsis will help to situate the analysis. At the beginning
of “Triumph of a Heart” we find the protagonist, played by the
actress/musician Björk, at home with her cat. The protagonist (hereafter
“Björk,” though a discernable complication will arise from this) is
Hearing Heima 223
noticeably bored, and perhaps irritated by her companion, as evidenced by

the ennui imparted by her facial expressions. She leaves the home in
frustration, the cat’s human wardrobe and judging glare while standing in
the doorway now suggesting a more domestic relationship between the
two. At the end of Scene A, Björk drives from her small pink home in the
middle of a vast, uninhabited grassland toward the city lights of
Reykjavík.17 Scene B1 shows her arriving at a bar in the early evening,
sitting down to drinks by herself, and remaining noticeably bored until
being joined by others in Scene B2. A complex “behind the scenes” shift
ensues from scene B3 to scene B4 involving the rehearsal of “Triumph of
a Heart,” which will be discussed in detail later. Björk becomes
increasingly happy and intoxicated throughout the night into scene B4,
climaxing in her sprint from the bar out into the streets, where she
promptly falls onto her face. Bleeding from the head, she initially finds
this amusing, but when her mood turns downtrodden, she walks along the
dimly lit streets from the city toward her home, only to faint in the grass
before she reaches her destination.
Scene C opens with Björk awakening to sunshine, having spent the
evening outside. She begins to continue her walk home, pink hearts now
inexplicably emanating from her mouth as she sings. The cat-partner, still
at home, sees the floating pink hearts from the window, and, seemingly
understanding that they must be coming from Björk, drives a car toward
their source. The cat pulls up next to her, still walking the road that has
now changed to dirt (apart from the main “ring road” that circumnavigates
the island, a very small number of roads in the county are paved), picks
her up, and drives her home. Upon returning home in scene A2, Björk
remembers her fond feelings for the cat, who, after receiving a kiss, grows
into a human-sized housecat who nonetheless is wearing a two-piece
suit.18 Via computer animation and live-action, the couple dances for
nearly a minute in what has to be the campiest ending even compared to
the artist’s notoriously campy videos (e.g., “It’s Oh so Quiet” and “Human
Behaviour”).
A closer reading of meaning in this music unearths two particularly
salient commentaries on the Icelandic ecosystem. The first, and most
obvious, may be the composing-out of the pastoral–urban–pastoral motive
in the video. Formally speaking, “Triumph of a Heart” is cast in a
modified compound AABA form,19 meaning that it features two
verse/chorus pairs, a contrasting bridge, then a recapitulatory chorus. As
analyzed in Example 11-2, the motivic contrast between the verse and
chorus, as well as between the verse/chorus pair and the contrasting
224 Chapter Eleven
bridge, reveals a structural similarity to sonata forms. Both initially pit two
themes against one another in a dominant/tonic relationship (e.g. the
Bb/Eb axis between Verse 1 and Chorus 1),20 contrast those two themes
with a developmental section, then return to one or both of the original
themes. Hermeneutic interpretations of this macro-formal structure,
germane to many western art forms, involve the narrative of [home-
journey-return].
Example 11-2, Compound AABA Motivic Structure in Björk, “Triumph of a

Heart” (2004)
Rather than interpret this with unbounded hermeneutics, drawing

visual evidence from the Icelandic ecosystem in the video grounds this
interpretation within social practice. The music/video pair both compose
out a dichotomy between Iceland as timeless [unspoiled/natural/pastoral],
and modern-day Iceland with its urban development and attendant cultural
scene. Part of this newly developed urban cultural practice revolves
around a Dionysian nightlife in Reykjavík that promotes massive amounts
of alcohol consumption.21 The opening frame of the video is not of Björk
or the cat but of the home itself. Nestled in the rolling green hills, butted
up against the rocky, mountainous interior (if this is believable walking
distance from Reykjavík—a long walk indeed—the camera must be
pointing either east or north toward the uninhabited interior), we barely
see any other human settlements, revealing the extraordinarily low
population density of the country.22
Hearing Heima 225
Just as the hard urban dance-beat enters at 0:27, Björk darts out the
door toward her car. Recall the concept of invariants and specifications
from ecological perception. Specification is to ecological perception as
signification is to semiotics. Sounds do not “signify” meaning by complex
semiotic processes, they instantly specify the invariant physical and
cultural meanings afforded to competent perceivers. For listeners familiar
with popular music, a four-on-the-floor dance beat instantly specifies a
host of cultural meanings, all of which involve [urbanism] in some way.
Further specifications may involve subjective experiences with these
sounds, including [sweat, alcohol, dance, sex, etc.]. Analysis of timbre
here reveals an immediate contrast to the first 26 seconds, in which the
perceiver only hears chanting by human voices. Specification is both
physical and cultural here. Our instant perception of the physical
invariants specify its source [human voice], and our awareness of the
cultural invariants of chanting specify a host of meanings including [old,
timeless], afforded more strongly by the depictions of the geologically
unspoiled, undeveloped Icelandic landscape.
From 0:27 to 3:24, Björk interacts with the urban ecosystem.
Undertaking wild adventures, she becomes liberated from her oppressive
and unhappy domestic situation (with the cat), but then longs for the
countryside. Returning home, she is happy to awaken in the lush volcanic
grasslands, and even happier to be reunited with her partner. As can be
seen from the Example 11-1 formal chart, this narrative structure in which
a protagonist leaves the home, departs on an adventure, then returns home,
aligns with the exposition/development/recap scheme that compound
AABA forms share with a host of other western art forms, including
sonata forms. The motivic sketches in Example 11-2 reinforce this sense
of journey in a manner quite uncommon to sonata forms. Björk’s voice
constantly makes an upward semitonal journey rising from the
Bb/Eb axis in the first verse/chorus pair, up to the B/E axis in the second
verse/chorus pair, and finally reaching up to the F major ending prepared
by the bridge heading into the final chorus.
Another quintessentially Icelandic element in the music stems from its
notable exception to the standard narrative flow in this medium. Music
videos typically present the album version of a piece (or a slightly
different mix of that recording) uninterrupted from beginning to end,
accompanied by a moving image of some sort to accompany the music.
While cinematic interpolations of various sorts are not unheard of in this
genre (the most famous example being Michael Jackson’s “Thriller”), the
interruption of Medúlla’s recorded version, spanning roughly 1:43 to 3:25,
226 Chapter Eleven
is especially self-referential. Instead of hearing the recorded versions of

verse 2 and chorus 2, the viewer hears newly recorded versions of this
material in which Björk is accompanied a capella by the mostly amateur
performers seen on the screen. In Scene B2, while the heard music is still
that of Medúlla, these humans are initially seen interacting with the
protagonist (played by Björk) as drunken bar compatriots. But in scene
B3, those same humans are now heard as performers rehearsing the parts
for the live recording (Björk’s management put out an open call for the
audition). More complex still, some of the on-screen performers are
professional a capella musicians who actually do appear on the Medúlla
version of “Triumph.” Furthermore, the viewer can hear, at different
times, both the Medúlla version of the studio-recorded a capella
performance by Japanese beat-boxer Dokaka, as well as the live version he
re-recorded at the bar.
Along with these “extras,” the protagonist’s persona shifts greatly
when this fourth wall is broken. Though characters may never escape the
associations of their actors, we can view the human actor, Björk
Guðmundsdóttir, as portraying a fictional character through Scene B2.
However, when Scene B3 arrives and we see the actress/musician leading
a rehearsal for a performance of her own composition, we are forced to
confront the idea that the actress Björk and the protagonist of the story are
one and the same (or at least that the sovereignty of those two personas
has been greatly compromised).
An ecological interpretation of this performer/participant conflation
emphasizes its commentary on the Icelandic socio-cultural landscape.
Having much to do with the fierce sense of national pride Icelanders have
cultivated since their liberation from Denmark in 1944, music—especially
group singing—is a highly celebratory and participatory event. Björk
interacts with the urban Icelandic ecosystem by leaving the isolation of the
artist, inviting all to participate in musical creation at the bar. But this is no
ordinary bar. Sirkus, which closed permanently in 2007, was the
downtown hotbed for local music—truly, a musicians’ bar.23 Steinunn
Jakobsdóttir’s memorial of the fabled bar for The Reykjavík Grapevine
read:
But although its walls, covered with music posters and artwork, might
collapse any minute, they’ve witnessed an essential part in the city’s
culture, as for years, Sirkus has been a hotbed of everything related to any
grassroots genre in art, music, fashion and filmmaking. Here, local bands
have taken their first steps and new talents have been discovered. In
Hearing Heima 227
between touring around the world, groups such as GusGus, Sigur Rós,
Múm have gone to Sirkus to relax.24
When the densest city in Iceland is under 200,000 people, it is nearly

impossible for even the country’s most celebrated international celebrity to
hide. (Björk “sightings” at bars in downtown Reykjavík are remarkably
common). Adaptation depends on interacting with an ecosystem in the
most efficient way. Thus, the radical participation seen and heard in
“Triumph of a Heart”—both between performer/participant and the
“behind the scenes” rehearsal at B3 to which we as viewers are privy —
demonstrates an ecological connection to the (urban)25 Icelandic socio-
cultural landscape.
Moving on now to the remaining two videos by Sigur Rós and Múm,
they contrast sharply in their depiction of the Icelandic ecosystem in that
they depict no urbanism whatsoever, and are devoid of adult human
interactions. Almost singularly focused on the geological landscape,
“Green Grass of Tunnel” includes a flock of birds, and “Glósólí” depicts a
small troupe of human children with a palpable degree of animism—itself
a commentary on the “human” interaction with the otherwise mineral and
flora-focused environment of the video. Despite these cinematic
similarities, I hope to show in the following music-driven comparative
analysis that they depict entirely different attitudes toward the Icelandic
geological landscape. Since neither song is structured using sections, per
se (they both instead develop a single theme), formal graphs such as
Example 11-1 will be of little musical merit here. Instead, the following
paragraphs of musical and cinematic description will serve to frame the
analysis to follow.
Unlike “Triumph of a Heart,” which features a standard rock formal
structure (despite its unique pitch ascent), Sigur Rós’s “Glósóli” is
structured more through its development of a single motive. It is also
perhaps the most explicit example of a rock form structured solely by a
dynamic process. The electric bass progression [G–D–E–C], which
accompanies singer Jón “Jónsi” Þór Birgisson as he steps through a [G4–
A4–B4] trichord occasionally neighbored on either side, unifies the song.26
The climax is the result of a crescendo spanning the entire track,
announced by Jónsi’s heroic ascent to D4 at 3:49.27 In addition to the
volume crescendo, the piece also utilizes an overarching rhythmic
crescendo. The bass and acoustic percussion begin by playing march-like
quarter-notes until 3:49, where they accelerate to eighth-notes en route to
the volume climax at 4:40. This type of form, which I have elsewhere
deemed a “Monothematic Form,”28 owes its shape to these two
228 Chapter Eleven
simultaneous processes applied to a single theme. The cinematic and

narrative content of the “Glósóli” video reveals many connections with the
musical form I identify. Note especially the striking visual parallel
between the track’s upward dynamic and spectral curve (Example 11-3)
and the slope of the cliff that coincides with the growth of these processes
at the video’s climax (Example 11-4).
As the video begins, we see a small boy with a military side drum
sitting alone on the beach, staring across a vast Icelandic sea. Marching to
the song’s quarter-note pulse, he wanders through the pastoral grassy,
rolling landscape, summoning other small children with the pulse of his
drum (which, as one might imagine, matches the pulse of the song). More
children are added to the group in a linear accumulation paralleling the
gain in amplitude throughout the song. Just before the shift to eighth-note
pulses at 3:49, the children lie down on a rock to sleep for the night. They
wake in the morning to bright sunlight (Glósóli means “glowing sun” in
Icelandic), and as they all gaze up a large hill, the drummer boy begins to
tap his new eighth-note pulse. The volume rises, the children’s facial
expressions grow more intent, and at the onset of the musical climax, the
boy points his mallet like a sword as the children race up the hill. As they
approach the hill’s crest, the camera pans out to reveal that the assumed
hill is actually a sheer cliff face, the children running toward it like
lemmings. Throughout the pounding climax, the children fly off—literally,
as birds—the cliff with bright smiles on their faces, as if it were a carnival
ride of sorts. Just as the last chord is struck, one last sheepish child
attempts to fly, but instead cannonballs off the precipice down into the
water. The fate of the fallen child is left to the viewer’s imagination as the
track fades out.
Nowhere in their entire discography does the comparatively relaxed
sound of Múm bear any resemblance to the climactic gestures at the end of
“Glósóli.” “Green Grass of Tunnel,” the second track from their 2002
record Finally We Are No One, exhibits a fairly uniform timbral and
dynamic profile. Except for the sparse valleys where the electronic
percussion momentarily drops out, the overall volume profile of the track
is exceptionally level. While this is not a notable feature in and of itself
(the extraordinarily high level of compression in modern pop recording
processes ensures this), the quietness, inactivity, and general sense of
“mellow” maintained over the course of the track is notable in its lack of
anything resembling a climax. Timbral analysis of the recording supports
this reading as well. Almost entirely electric/electronic, it is devoid of any
acoustic sounds besides the occasional appearance of singer Kristín Anna
Hearing Heima 229
Valtýsdóttir’s childlike and airy soprano, which does not enter until the
track is halfway over.29 Only the mineral-esque “tinkering” noises (to be
discussed in closing) interfere acoustically with the keyboard pads and soft
synth leads in the opening two minutes.
Example 11-3, Spectrographic Image of Sigur Rós, “Glósóli” (2005)
Example 11-4, Video Still from Climax of Sigur Rós, “Glósóli” (2005, 4:36)
230 Chapter Eleven
Múm’s video, entirely computer-generated, also contrasts sharply with

the sharp images of the Icelandic countryside in “Glósóli.” (Although the
analogy is certainly imperfect, it seems fitting that electronic instruments
pair with computer animation in the former, while acoustic instruments
pair with live action in the latter). Depicting the iceberg-filled lagoons in
winter, bordered by the sea on one side and the mountainous interior on
the other, the lush green color one might expect in “Green Grass of
Tunnel” nonetheless only appears once: crayoned underneath a rocket ship
in a child’s drawing hanging on the wall of an abandoned shed in the
whitewashed coastline (2:46). The camera pans upward from the ground
shortly afterward to reveal the actual rocket ship—the one depicted in the
drawing—slowly shooting skyward. Puffins and other birds, who have
been circling the icy cliffs since the video’s opening, are now drawn to the
homemade dirigible. Retroactively, we realize that the child’s drawing was
in fact set in this location—one can now see the shed in which the drawing
was hung, as well as the nearby lighthouse, underneath the purview of the
hovering rocket, and this entire scene equivocates the drawing in the
shed.30 A reasonable interpretation might be [escape], as it is presumably
the author of the drawing who is now piloting the rocket away from this
deserted place, or perhaps the pilot simply wishes to commune with the
birds and mountain peaks. The video ends with the pilot leaving the island
flying toward the iconic aurora borealis viewable throughout the island.
Though seen from just about any remote location in Iceland, the northern
lights do indeed originate from the north, and based on the angle of the
coastline relative to the aurora, we can deduce that the “scene” of the
video is now the northwestern fjords near Ísafjörður.
Comparing the two songs, both ostensibly representing the Icelandic
geological landscape, one can sense the contrasts, contradictions, and
dichotomies so germane to the island. Particularly salient are two
contraposed pairs, each of which maps onto the Sigur Rós and Múm
pieces, respectively: growth/stasis, and flora/mineral. An ecological
analysis of the musical elements in these two pieces will illuminate the
very different ways they depict the Icelandic geological landscape. I will
first begin with the compositional elements of the music, and then close
with a brief discussion of the recorded sonic landscape itself.
Example 11-5 shows the opening vocal statement in “Green Grass of
Tunnel.” This selected passage is representative of the entire track’s vocal-
melodic profile in its arpeggiation of the C major tonic triad and its
dedication to the pentatonic collection. Never articulating either member
of the B/F tritone necessary for truly tonicizing C major, the melody only
Hearing Heima 231
reinforces this tonal center by its aforementioned arpeggios and alternating

phrase endings on E4 and C4 over the tonic bass note. The commonly
occurring bass progression derives from a “gapped fifths cycle,”31 which
of course makes it a subset of the complete fifths cycle in the pentatonic
voice collection. Compare Múm’s pentatonicism with the clear tonality
expressed by three repeated vocal patterns in “Glósóli,” transcribed in
Example 11-6. Lower (F͓4) and upper (C5) semitonal neighbors around
the G/B dyad more actively reinforce a sense of tonality than the
pentatonic collection in Example 11-5. Furthermore, the presence of a
tonic triad unfolds gradually. We first hear F͓4 as a lower neighbor to G4
(6a), C5 as an upper neighbor to B4 (6b), and finally, C5 as a passing tone
to the climactic D5 at the arrival of the title lyric (6c).
Example 11-5, C Pentatonic melody in Múm, “Green Grass of Tunnel” video

(2002, 1:54)
By recognizing the lyrical narrative, which personifies the sun and the
process of its growth over the course of the track (see translation in
Example 11-6), we can see how this gradual unfolding of the tonic triad
works concomitantly in our search for meaning. Though there are many
ways to arpeggiate a G major triad at the keyboard, most of us would feel
in our fingers a sense of growth as we rise from G to B, and then a sense
of completion as we reach up to D. True as it may be that “Green Grass of
Tunnel” also thrives on arpeggiated tonic triads, it lacks the process of
growth that characterizes “Glósóli.” This is undoubtedly due to the
absence of tonal neighbors, with their attendant push and pull toward
members of the triad, as well as the manner in which the Múm melody
seems to be treating the members of said triad as undifferentiated scale
steps in a pentatonic collection.
232 Chapter Eleven
Example 11-6, G Major melodies in Sigur Rós, “Glósóli” video

(2005, 0:48, 2:58, 4:25)
In the videos, this dichotomy between tonal growth and pentatonic

stasis plays out as two very different commentaries on the Icelandic
landscape. Rich with lush greens and vibrant flora, Sigur Rós depicts
growth in perhaps its most recognizable form—plant life. Iceland owes
much of its greenery to the interaction between rich volcanic soils and
comparatively mild, marine-stabilized temperatures around its fertile ring.
Organicism is literally present in the visual elements, just as it is
metaphorically present in the gradual unfurling and eventual blossoming
of the tonic triad. The sonic and spectral growth shown in Example 11-3
couples with the cliff scene in Example 11-4 to further enhance this sense
of growth at the video’s climax.
“Green Grass of Tunnel” is not merely defined by its lack of this
growth. Rather, the stasis it projects through its pentatonic collection is a
direct commentary on the comparatively timeless nature of Iceland’s
glaciers. We may choose to hear the rising and falling vocal melody as
either stepping through a pentatonic scale or as skipping around the tonic
triad, but in either case, the rapidity with which this gesture happens seems
to suggest a bounding over the jagged, craggy mountains of the interior, as
well as those in the iceberg-filled fjords (the sense of playfulness and
whimsy in Valtýsdóttir’s delivery would seem to bolster this observation).
The pentatonic bass collection and parallel voice leading also deters any
Hearing Heima 233
sense of “progression” and is better heard as an immobile ostinato. Of

course, glaciers and icebergs do move, but, as opposed to the living flora
and fauna depicted in “Glósóli,” they do not do so under their own
volition.
Yet another way to express these two dichotomous depictions of
Iceland involves hearing Múm as mineral/cold, and Sigur Rós as
organic/warm. The preceding compositional comparison has relied on
metaphors commonly associated with tonal and pentatonic collections, but
the physical materials used to make the recordings leave their invariant
traces as well. The bass lines transcribed in Examples 11-5 and 11-6 are
produced using circuit boards and trees, respectively. “Green Grass of
Tunnel” bears the distinct signatures of computer/electronically processed
drum, bass, and synth sounds, while Glósóli leaves the listener with
distinct traces of human and plant activity in its rhythm section. We can
hear the sounds of bass strings resonating against a wooden fretboard, as
well as the preceding pick-attack that generates this vibration. One might
even go so far as to call the video titles ironic. “Green Grass of Tunnel”
depicts no greenery in its video and utilizes no organic materials in its
rhythm section production. “Glósóli” depicts very little sunlight on-screen
and, though the earth’s sun contains almost no carbon, the track relies
entirely on once-living organic materials for its instrument sounds.
Elsewhere, I have discussed the role of acoustic “tinkering” noises
created in the studio to add resonance to otherwise electronically produced
tracks like “Green Grass of Tunnel.”32 In closing, I would like to point out
yet another irony borne out by this observation. Sounds such as these,
heard clearly in the opening of the track, yet present throughout, seem to
be commonplace in post-millennial electronic rock music. It seems to be a
clever solution to a common problem. As anyone who has created
electronic recordings on a computer can attest, the end result is often flat,
lifeless, dull, and cold when compared to comparatively warm and
dynamic acoustic recordings. This is due to the dynamic compression
applied to most synthesized instruments and samples. Thus, the sounds of
small physical objects being manipulated in a live room are often added to
tracks such as these to add the desired ambience. What is ironic about
“Green Grass of Tunnel” is how the addition of these noises seems to have
the opposite effect. By attending to the physical invariants specified by the
sounds themselves, we hear mineral/metallic. That is to say, we hear yet
another iteration of the cold, mineral nature of Iceland’s glaciers and
icebergs in the sounds themselves. The video titles for “Glósóli” and
“Green Grass of Tunnel” may be ironic, but the invariant physical
234 Chapter Eleven
properties of the materials used to craft their sonic depictions of two very
different Icelandic landscapes—organic in the former, mineral in the
latter—instantly specify the same images as those depicted cinematically.
My hope is that the ecological methods espoused here and throughout this
chapter help us to produce a “musicology of the image” that directly
addresses invariant connections between music and ecosystem.
Notes
1
Throughout the essay, I prefer use of the term “ecosystem” rather than
“environment” or similar nouns. The former emphasizes an interactive space
between living organisms and non-living elements, both natural and human-made.
The latter tends to create distinctions between humans, non-human animals, plants,
and natural features.
2
Nicholas Cook, Analyzing Musical Multimedia (New York: Oxford University
Press, 2000), p. 150.
3
In addition to drawing from these two scholarly fields, my views on the Icelandic
landscape, especially its geo-physical landscape, are greatly informed by the two
trips I undertook to the island in 2010 and 2011. On the first of these trips, I spent
a great deal of time around the live music scene in Reykjavík, and on the second, I
experienced the wilderness by hiking and camping around the more sparsely-
populated southern and eastern coastlines.
4
As evidence of this current growth, one might note the special joint
Ecomusicologies “pre-conference” at the 2012 joint national meeting of the
Society for Music Theory, the American Musicological Society, and the Society
for Ethnomusicology, as well as the edited collection of essays to be released from
this meeting.
5
My intent in this introduction is not to provide a complete literature review of
these two fields, which is superfluous for the current application. No prior
understanding of these two fields will be necessary to perceiving the links between
music, film, and place in the three analyses to come. Instead, relevant details from
key sources in these fields will be presented throughout the analyses to come in
order to frame the ecological details of the music and cinema.
6
Denise Von Glahn, The Sounds of Place: Music and the American Cultural
Landscape (Lebanon, NH: Northeastern University Press, 2009).
7
Ways of Listening: an Ecological Approach to the Perception of Musical
Meaning (New York: Oxford University Press, 2005).
8
Allan Moore, Song Means: Analyzing and Interpreting Recorded Popular Song
(London: Ashgate Publication Company, 2012).
9
Eric Clarke, “Subject-Position and the Specification of Invariants in Music by
Frank Zappa and P.J.Harvey,” Music Analysis 18/3: 347–374.
10
Nicola Dibben, through a survey of over 45 Icelandic music videos, as well as
extensive field work conducted in 2006, has conclusively validated this aspect of
Hearing Heima 235
the country’s musical culture. See Dibben, “Nature and Nation: National Identity
and Environmentalism in Icelandic Popular Music Video and Music
Documentary,” Ethnomusicology Forum 18/1 (June 2009): 131–151.
11
Ari Alexander Ergis Magnusson, Screaming Masterpiece (Soda Pictures, 2005),
DVD.
12
Dibben 2009 provides a closer reading of Iceland’s popular music history, as
well as its briefer history of music video production.
13
The highly respected and influential indie-rock radio station KEXP (Seattle)
broadcasts live from Iceland Airwaves each year, taking up temporary residency in
a local hipster hostel coincidentally named KEX.
14
For example: fire/ice (active volcanoes and perpetual glaciers), farmland/tundra
(the uninhabited interior of the island and the fertile agrarian outer ring), and
dark/light (perpetual darkness in the winter, midnight sun in the summer).
15
It is hoped that the reader will take advantage of streaming video sites such as
YouTube in order to experience the music videos analyzed here.
16
See Jason Summach, “Form in Top-20 Rock Music, 1955–89” (Ph.D. Dissertation,
Yale University, 2012); and Brad Osborn, “Subverting the Verse/Chorus
Paradigm: Experimental Formal Structures in Post-Millennial Rock Music” (Ph.D.
Dissertation, University of Washington, 2010).
17
Of the country’s 320,000 total human population, just over half of those humans
live in the capital city of Reykjavík, located in the southeast corner of the country.
The rest of the population is either clustered into small towns or spaced out into
smaller-still villages, all of which reside only on the country’s outer ring (the
interior is a vast, mountainous tundra covered by snow and ice most of the year,
and is all but uninhabitable).
18
A gender-based analysis of this video that highlights the ambiguous identity of
the cat-partner, as well as the inter-species romantic overtones that accompany the
kiss, while outside the bounds of the current ecologically focused interpretation,
seems especially fruitful to me.
19
For more on the compound AABA form as a conventional formal structure see
John Covach, “Form in Rock Music: A Primer,” in Engaging Music: Essays in
Musical Analysis, edited by Deborah Stein (New York: Oxford University Press,
2005), pp. 65–76. The final A section of these compound AABA forms can either
be a verse/chorus pair, but is just as likely to be either one or the other. Usually, if
only one section serves as the final recapitulation, it will be the chorus (as here).
20
Of course, this relationship is mirrored relative to a traditional first and second
tonal area. Rather than think of this as some sort of dualistic relationship, or
highlighting the role of plagalism in rock music, I hear the composed-out Bb triad
of the first verse as a structural dominant anticipating the arrival of the Eb major
triad in the more memorable chorus.
21
So intense is this level of celebration that the U.S. State Department recently
issued the following warning to American tourists: “be aware that downtown
Reykjavik [sic] can become disorderly in the early morning hours on weekends.”
<http://travel.state.gov/travel/cis_pa_tw/cis/cis_1138.html#crime>, accessed July
22, 2012.
236 Chapter Eleven
22
The countryside is quite untouched indeed—93% of the country’s population
lives in some sort of urban environment.
<https://www.cia.gov/library/publications/the-world-factbook/geos/ic.html>,
accessed July 22, 2012.
23
My sincere thanks go to Kimberly Cannady for identifying this specific bar for
me. While I had passed by the graffiti-clad ruins of the bar on several occasions, it
was no longer open to the public by the time I had started visiting the country.
24
Steinunn Jakobsdóttir, “Last Call: Sirkus is Closing,” The Reykjavík Grapevine,
2007. <http://www.grapevine.is/Home/ReadArticle/Last-Call-Sirkus-is-Closing>
25
I emphasize the role of the urban in “Triumph of a Heart” inasmuch as it
contrasts sharply with depictions of the natural/pastoral in the two other videos
analyzed in this essay, as well as other Björk videos such as “Jóga” (1997).
26
The only modifications to this bass line happen through rhythmic and metric
alteration. Twice during the build-up to the climax, the rhythm is normalized to
equal values, beginning on the D instead of G [D–E–C–G].
27
The valleys at the end of the track’s spectrum and waveform graphics represent
the last chord being held out over 40 seconds until it gradually decays.
28
See Brad Osborn, “Understanding Through-Composition in Post-Rock, Math-
Metal, and other Post-Millennial Rock Genres,” Music Theory Online 17/3 (2011).
29
The cut chosen for the video is, in fact, a radio edit that omits about 45
seconds—her voice does not appear in the album version until 2:35.
30
The video’s lighthouse may have been directly inspired by the lighthouse-
keeper’s home in which Múm recorded this and one other album. See “Múm: The
Good Life” [interview] in The Milk Factory
<http://www.themilkfactory.co.uk/interviews/mumiw.htm>
31
For more on gapped fifths cycles, see Guy Capuzzo, “Sectional Tonality and
Sectional Centricity in Rock Music,” Music Theory Spectrum 26/2: 177–199.
“Glósóli” features the same gapped fifths collection in its bassline, and, although
that collection is also a subset of the larger diatonic collection heard in the voice
part, I hear the link between bass and voice collections as weaker in “Glósóli” due
to the three-pitch-class difference in cardinalities, as opposed to only a one pitch-
class difference between bass and voice in “Tunnel.”
32
See Osborn, “Subverting the Verse/Chorus Paradigm.”
PART IV:
MUSICAL GEOMETRY
AND TEMPORAL STRUCTURING
INTRODUCTION TO PART IV
In its concluding part, “Musical Geometry and Temporal Structuring,”

our collection of essays concerns itself with theoretical innovations that help
the analyst to more effectively describe the work of late twentieth-century
composers (and could also find application in the music of tomorrow’s
composers). It then returns to an account of how a living composer makes
use of rhythmic processes that can be understood theoretically and
represented graphically, to round off the volume with the same sort of
analytical study with which it began.
Paul Lombardi’s chapter on “Serial N-Cubes” extends some of his
previous published work. It considers the extension of the concept of a
twelve-tone matrix to multiple dimensions, which could incorporate twelve-
tone transformations other than the usual P, I, R and RI such as
multiplication, rotation and permutation, or account for multiple row forms
such as Stravinsky uses in the Requiem Canticles. To bring the part and
volume to a close, Aleksandra Vojcic then investigates some of the rhythmic
processes that structure the music of Colin Matthews’ Eleven Studies in
Velocity. These include structural markers in pulse streams, tempo
modulation, and pulse-stream polyphony.
—Jack Boss
CHAPTER TWELVE
SERIAL N-CUBES
PAUL LOMBARDI
Serialism can be understood through multiple dimensions, enhancing

our conception of music. Twelve-tone series have one dimension, such as
the one that follows.1
3, 2, 9,8,7,6, 4,1,0, A,5, B
Example 12-1, a twelve-tone series
Twelve-tone arrays have two dimensions. The series is in the top-most

horizontal row, while some form of the series extends downward in the
left-most column. The rows are transpositions of each other and the
columns are transpositions of each other. The top-most row and the left-
most column are axes.
3 2 9 8 7 6 4 1 0 A 5 B
4 3 A 9 8 7 5 2 1 B 6 0
9 8 3 2 1 0 A 7 6 4 B 5
A 9 4 3 2 1 B 8 7 5 0 6
B A 5 4 3 2 0 9 8 6 1 7
0 B 6 5 4 3 1 A 9 7 2 8
2 1 8 7 6 5 3 0 B 9 4 A
5 4 B A 9 8 6 3 2 0 7 1
6 5 0 B A 9 7 4 3 1 8 2
8 7 2 1 0 B 9 6 5 3 A 4
1 0 7 6 5 4 2 B A 8 3 9
7 6 1 0 B A 8 5 4 2 9 3
Example 12-2, a twelve-tone array

Serial N-Cubes 241
Twelve-tone cubes have three dimensions and are made by extending a

third axis, perpendicular to the other two axes. The axes meet at the origin.
Transpositions of the array are populated along the third axis. The axes are
arranged according to a right-handed axis system such that the right-
pointing horizontal axis is for dimension one, the down-pointing vertical
axis is for dimension two, and the axis pointing out the back of the page
(away from the reader) is for dimension three.2
Example 12-3, a twelve-tone cube
The specific coordinates of the pitch classes within series, arrays, and
cubes are referenced with indices; each dimension requires one index.
Order positions in series are indexed i, and coordinates in arrays and cubes
are indexed ij and ijk respectively. Indices begin with 0 instead of 1 to
facilitate mathematical operations.
Three dimensions can be comprehended intuitively, while objects with
more than three dimensions are intuitively abstract. A fourth axis
extending from the origin of a cube perpendicular to the other axes makes
a 4D (four-dimensional) cube. Transpositions of the 3D cube occur along
242 Chapter Twelve
the fourth axis, which is indexed l such that the coordinates in the 4D cube
are indexed ijkl.
Example 12-4, a 4-dimensional twelve-tone cube
Serial n-cubes can have any number of dimensions: 3D cubes are called 3-
cubes, 4D cubes are called 4-cubes, 5D cubes are called 5-cubes, etc.
Cubes with more than three dimensions are called hypercubes, and 4-cubes
are specifically called tesseracts.
In an earlier article, Michael Wester and I employed a tesseract to
capture the symmetric/antisymmetric design of Boulez’s Structures 1a.3 A
brief review of this design is given below to show a relevant application of
serial n-cubes. Then, this essay examines n-cubes in more detail and in a
more general context than was done in the Structures 1a article to provide
a basis for further applications that involve them. N-cubes generally
include twelve-tone transformations, but an example of an n-cube
conceived through permutation is examined as well. Finally, this essay
shows how rotational arrays can be elaborated into n-rotational cubes.
Boulez, Structures 1a
Boulez’s Structures 1a is a multi-serial composition for two pianos that
employs a single twelve-tone series (shown in Example 12-1) to govern
pitch class, duration, dynamics, and articulation. The serial organization
involves one T-array and two I-arrays. The T-array has axes made from
the prime and inverted forms of the series (P3 or just P, and I3 or just I),
while one I-array has axes made from just P, and the other has axes made
from just I. The dimensions of the three arrays are referred to as PxI, PxP,
and IxI respectively so that they correspond to the series forms of the axes.
Serial N-Cubes 243
PxI T-array
I

P3 2 9 8 7 6 4 1 0 A 5 B
4 3A9 8 7 5 2 1B6 0
9 8 3 2 1 0A7 6 4B5
A9 4 3 2 1B8 7 5 0 6
BA 5 4 3 2 0 9 8 6 1 7
0B6 5 4 3 1A9 7 2 8
2 1 8 7 6 5 3 0B9 4A
5 4 BA 9 8 6 3 2 0 7 1
6 5 0 BA 9 7 4 3 1 8 2
8 7 2 1 0B9 6 5 3A4
1 0 7 6 5 4 2 BA 8 3 9
7 6 1 0 BA8 5 4 2 9 3
PxP I-array IxI I-array

P I

P3 2 9 8 7 6 4 1 0 A 5 B I 3 4 9 AB 0 2 5 6 8 1 7
2 1 8 7 6 5 3 0B9 4A 4 5 AB 0 1 3 6 7 9 2 8
9 8 3 2 1 0A7 6 4B5 9A3 4 5 6 8B0 2 7 1
8 7 2 1 0B9 6 5 3A4 AB 4 5 6 7 9 0 1 3 8 2
7 6 1 0 BA 8 5 4 2 9 3 B0 5 6 7 8A1 2 4 9 3
6 5 0 BA 9 7 4 3 1 8 2 0 1 6 7 8 9B2 3 5A4
4 3A9 8 7 5 2 1B6 0 2 3 8 9AB 1 4 5 7 0 6
1 0 7 6 5 4 2 BA 8 3 9 5 6B0 1 2 4 7 8A3 9
0B6 5 4 3 1A9 7 2 8 6 7 0 1 2 3 5 8 9B4A
A9 4 3 2 1B8 7 5 0 6 8 9 2 3 4 5 7AB 1 6 0
5 4 BA 9 8 6 3 2 0 7 1 1 2 7 8 9A0 3 4 6B5
BA 5 4 3 2 0 9 8 6 1 7 7 8 1 2 3 4 6 9A0 5B
Example 12-5, arrays in Boulez’s Structures 1a
For the entire composition, every note corresponds to a serialized pitch

class and a different serialized duration. (Duration series are measured in
the number of thirty-second notes directly proportional to pitch classes.)
For example, the Piano 1 pitch classes for the first half of the composition
come from the twelve transpositions of the prime series in the PxI array in
the order P3, P4, P9, …, P7. In other words, the forms of the series are
presented methodically from the top horizontal row down to the bottom.
This is called the P forms of the series in the order of I because the specific
transpositions are ordered according to the pitch classes in I. The result is
that not only are the pitch classes serialized, but also the series forms. At
the same time, the durations of those notes correspond methodically to the
244 Chapter Twelve
twelve RI series forms in the IxI array from the top horizontal row to the
bottom. That is, the specific transpositions of the RI series forms are
ordered according to the pitch classes in RI (RI series forms in the order of
RI).
Meanwhile, the Piano 2 pitch classes come from the I series forms in
PxI, and the durations come from the R series forms in PxP. The second
half of the composition is constructed similarly, as shown in the table
below. This table shows which series forms and arrays are used for the
pitch classes and durations for each piano and for each half of the
composition.
Part A Part B
P series in the order of I RI series in the order of RI
PC PxI array IxI array
Antisymmetry Symmetry
Piano 1
RI series in the order of RI I series in the order of R
Duration IxI array PxI array
Symmetry Antisymmetry
I series in the order of P R series in the order of R
PC PxI array PxP array
Antisymmetry Symmetry
Piano 2
R series in the order of R P series in the order of RI
Duration PxP array PxI array
Example 12-6, series and arrays used in Boulez, Structures 1a
This intricate design contains an elaborate pattern of symmetries, and these

symmetries extend from the middleground to the background (n.b., the
retrograde series forms shown in the table above have a trivial effect on
the symmetries.) In the middleground, the symmetries are found in the I-
and T-arrays. I-arrays are symmetric around the main upper-left lower-
right diagonal, in that the pitch classes in the upper right are mirrored in
the lower left. T-arrays are antisymmetric around this diagonal in that the
negative of the pitch classes modulo 12 are mirrored, but the antisymmetry
is transposed by two times the value of the pitch class in the origin. (The
symmetry and antisymmetry is shown in the example below; the solid
arrow indicates symmetry and dashed arrow indicates antisymmetry.)
Because the transpositions of the series forms are serialized in the order of
the pitch classes in the series, the middleground corresponds to the
symmetry and antisymmetry in the I- and T-arrays. In other words, the
symmetries occur in the composition because the series forms are ordered
according to themselves.
Serial N-Cubes 245
I-array T-array
Example 12-7, symmetry and antisymmetry in the arrays of Boulez, Structures Ia
In addition to the symmetries in the arrays, some pairs of arrays are

also symmetric or antisymmetric, which can be seen with help from
another T-array: IxP. The PxI array is symmetric with the IxP array, and
the PxP array is antisymmetric with the IxI array.4 The symmetries
between the arrays can be balanced into a 4-cube. The axes of this 4-cube
are in order P, I, P, and I, thus the 4-cube has the dimensions PxIxPxI as
shown below where the dimensions are indexed and in the order ijkl.
Although the series forms and arrays are duplicated many times
throughout this 4-cube, significance is found in the ordering and
symmetries between the series forms and arrays.
Example 12-8, 4-cube with symmetries between arrays in Boulez, Structures Ia
The specific transpositions of the PxI, IxP, PxP, and IxI arrays are on
the surface of the 4-cube, which can be understood through the concept of
slices. Imagine a serial 3-cube and slice through it with a knife so that you
cut out a single array. This array is a 2D slice of the cube. In the same
246 Chapter Twelve
fashion, a series can be a 1D slice of an array, 3-cube, or 4-cube; an array

can be a 2D slice of a 3-cube or 4-cube; and a 3-cube can be a 3D slice of
a 4-cube. So, the PxI, IxP, PxP, and IxI arrays are 2D slices of the PxIxPxI
4-cube. Therefore, some 2D parts of the 4-cube are mirrored with other 2D
parts symmetrically or antisymmetrically, and the symmetries in the 4-
cube reflect the background design of the composition. Example 12-9
below shows the same arrays as Example 12-6 above, but PxI is replaced
by IxP as necessary. The lines and dashed lines show the symmetric and
antisymmetric relationships respectively between pairs of arrays.
Part A Part B
P series in the order of I RI series in the order of RI
PC
PxI IxI
Piano 1
RI series in the order of RI I series in the order of R
Duration
IxI IxP
I series in the order of P R series in the order of R

PC
IxP PxP
Piano 2
R series in the order of R P series in the order of RI
Duration
PxP PxI
Example 12-9, series and arrays used in Boulez, Structures Ia, including IxP to
show certain symmetries between arrays
Moving further into the background reveals one more antisymmetric

relationship. All of the arrays in Piano 1 occur on the surface of a 3-cube
with the dimensions IxPxI, and all of the arrays in Piano 2 occur on the
surface of a 3-cube with the dimensions PxIxP. These two 3-cubes have an
antisymmetric relationship between them, and each is a 3D slice of the
PxIxPxI 4-cube. Therefore, the antisymmetry between the two pianos is
the same as it is between some 3D slices of the 4-cube.
The 4-cube succinctly captures the design of Structures 1a at multiple
levels. The specific transpositions of the T- and I-arrays are on the surface
of the 4-cube. The middleground structure corresponds to symmetries
within certain 2D slices of the 4-cube; the background structure
corresponds to symmetries between certain 2D slices; and the structure
between the two pianos corresponds to symmetries between certain 3D
slices. Structures 1a was the inspiration for this investigation, as shown in
this brief overview. The next part of my essay delves into deeper details of
n-cubes.
Serial N-Cubes 247
Axes to Propagate Dimensions

The specific transpositions of the P and I axes used in the 4-cube above
are P3 and I3. For any n-cube, its axes must begin with the same pitch class
because the origin contains only a single pitch class. The equation below,
as a function of S, gives the inverted series form beginning on the same
pitch class. The pitch classes in I at order position i are determined by
substituting the pitch-class values from S at order positions s0 and si.
Si s0 , s1 , , sB
I i Si 2 s0 si mod12
The Structures 1a 4-cube uses only P and I series forms for axes, however
the appropriate transposition of any series form can serve as the basis of an
axis. The equations that maintain a common pitch class at order position 0
for the standard twelve-tone transformations and their retrogrades are
given as follows.
Prime (identity) Pi Si si R i Si
sBi s0 sB mod12
Inversion Ii Si 2 s0 si mod12 RIi S i
sBi s0 sB mod12
Multiplication by 5 M i Si 5si 4 s0 mod12 RM i S i 5sB i s0 5sB mod12
Multiplication by 7 MIi Si 7 si 6 s0 mod12 RMIi S i 7 sB i s0 7 sB mod12
Furthermore, other types of transformations can serve as the basis for

axes as long as they maintain the common pitch class at the beginning.
Some examples (shown below) are rotation, pitch-class/order-number
isomorphism,5 the Lewin re-ordering,6 and any permutation. (In the
equations below, r is the number of order positions that the series is
rotated to the left, and can have a value of 0–11; m multiplies the order
positions to get every mth order position, and can have a value of 1–12.)
Some of these transformations will be revisited later in this essay.
Rotation Oir Si s i r mod12

sr s0 mod12
Isomorphism H i Si is i0 s0 mod12
Re-ordering m
L Si
i s m i 1 1 mod13
sm 1 s0 mod12
Permutation X i Si xi
The equation for the traditional twelve-tone T-array is below. It gives the
pitch class at coordinate ij when the appropriate pitch-class values from S
248 Chapter Twelve
are substituted. The PxI in the superscript shows the axes on which the
dimensions are based.
Si s0 , s1 , , sB
A PxI
ij s s
i j s0 mod12
N-cubes are difficult to display graphically. Thus, it is valuable to

know the equations that enumerate them. The equations facilitate
performing mathematical operations on n-cubes. Also, an equation
succinctly contains the characteristics of its corresponding n-cube.
Equations for many examples of n-cubes follow with the next few
paragraphs. The PxI array above is only one of many possible 2D arrays.
To serve as the basis for constructing more n-cubes, the equations are
given below for all 2D arrays based on the standard twelve-tone
transformations and their retrogrades with P for dimension one. Equations
that begin with something other than P are redundant because they can be
made by simply substituting various series forms for S.
A ijPxP s s s mod12
i j 0 A ijPxR s s
i B j sB mod12
A PxI
ij s s s mod12
i j 0 A PxRI
ij s s
i B j sB mod12
A PxM
ij s 5s s mod12
i j 0 A PxRM
ij s 5s
i B j sB mod12
A PxMI
ij s 7s s mod12
i j 0 A PxRMI
ij s 7s
i B j sB mod12
If dimension one is limited just to P, there are 64 possible 3-cubes that

use only the same 8 twelve-tone series forms used above. Three examples
are below.
PxIxP
Cijk s s
i j sk mod12
PxIxM
Cijk s s
i j 5sk 4 s0 mod12
C PxIxR
ijk s s
i j sB k sB mod12
Similarly, if dimension one is limited just to P again, there are 512

possible 4-cubes (three examples follow).
Serial N-Cubes 249
PxIxPxI
Cijk " s s
i j sk s" s0 mod12
PxIxMxMI
Cijk " s s
i j 5sk 7 s" s0 mod12
PxIxRxRI
Cijk " s s
i j sB k sB " s0 mod12
N-cubes with any number of dimensions are possible. The axes can be
duplicated any number of times, and just a few of these possibilities follow
below.
PxIxMxMIxPxIxMxMI
§ si s j 5sk 7 s" ·
Cijk "mnop ¨¨ ¸¸ mod12
© sm sn 5so 7 s p s0 ¹
PxIxRxRIxMxMIxRMxRMI
§ si s j sB k sB " ·
Cijk "mnop ¨¨ ¸¸ mod12
© 5 sm 7 sn 5 s B o 7 s B p ¹
§ si s j sB k sB " ·
¨ ¸
PxIxRxRIxMxMIxRMxRMIxPxIxRxRIxMxMIxRMxRMI
Cijk ¨ 5sm 7 sn 5sBo 7 sB p ¸ mod12
"mnopqrstuvwx ¨s s s s ¸
¨ q r B s B t ¸
¨ 5s 7 s 5s 7 s ¸
© u v B w B x ¹
The 16-cube (the very long equation) contains two of each of the 8 twelve-
tone forms of the series discussed above, and thus contains many
symmetric and antisymmetric relationships between various sub-
dimensional slices. For example, the first 8 dimensions are symmetric with
the last 8, and the 3D IxRxMI slice is antisymmetric with the 3D PxRIxM
slice.
Permutation
As discussed in the previous section, any series form can serve as the
basis for an axis including a permutation that transforms one series into
another—whatever permutation that may be. Stravinsky’s Requiem
Canticles makes use of two twelve-tone series, and those series are used in
conjunction in the Interlude and Postlude movements. In this section of the
essay, these two series are used to build a 4-cube. The two series S1 and S2
(see below), are identified by their superscripts 1 and 2.
250 Chapter Twelve
Si1 5,7,3, 4,6,1, B,0, 2,9,8, A

S 2j 5,0, B,9, A, 2,1,3,8,6, 4,7
Example 12-10, two series in Stravinsky’s Requiem Canticles
The two series conveniently begin with the same pitch class so that neither
of them needs to be transposed to combine them into an n-cube. The
following array has them as axes. The transpositions of P1 are in the
horizontal rows in the order of P2, and the transpositions of P2 are in the
vertical columns in the order of P1. Because the two series begin with the
same pitch class, s0 can be taken from either of them.
1 2
A ijP xP = si1 s 2j s0 mod12
P2
Ļ
1
Pĺ 5 7 3 4 6 1 B 0 2 9 8 A
0 2 A B 1 8 6 7 9 4 3 5
B 1 9 A 0 7 5 6 8 3 2 4
9 B 7 8 A 5 3 4 6 1 0 2
A 0 8 9 B 6 4 5 7 2 1 3
2 4 0 1 3 A 8 9 B 6 5 7
1 3 B 0 2 9 7 8 A 5 4 6
3 5 1 2 4 B 9 A 0 7 6 8
8 A 6 7 9 4 2 3 5 0 B 1
6 8 4 5 7 2 0 1 3 A 9 B
4 6 2 3 5 0 A B 1 8 7 9
7 9 5 6 8 3 1 2 4 B A 0
Example 12-11, array using Stravinsky’s Requiem Canticles series as axes
Incorporating the prime and inverted forms of the two series with this
array yields a 4-cube. This 4-cube has the two prime forms for dimensions
one and three, and their two inverted forms for dimensions two and four.
Its equation is as follows.
1 1 2 2
P xI xP xI
Cijk " s 1
i s1j sk2 s"2 s0 mod12
This 4-cube is a complicated object, and it can be examined from a variety

of directions. For example, transpositions of an S2 P2xI2 array occur in the
Serial N-Cubes 251
order I1 along dimension two. The specific handful of series forms that
occur in Requiem Canticles are distributed throughout the 4-cube. The
combination of the two different series allows for the construction of an n-
cube with axes based on some permutation that transforms one of the
series into the other.
N-Rotational Cubes
Separate from combining two different series, a compositional device
commonly associated with Stravinsky is the rotational array. It is used in
several of Stravinsky’s final works, including the Requiem Canticles, and
one rotational array from this piece is shown below. The hexachord series
is in the top horizontal row, and its five rotations occur below it
horizontally. All of the rotations are transposed to begin with the same
pitch class. The equation for rotational arrays is also shown below, where
n + 1 is the cardinality of the series, which is 6 in this case.
5 7 3 4 6 1
5 1 2 4 B 3
Si = 5, 7, 3, 4, 6,1 5 6 8 3 7 9
5 7 2 6 8 4
Oij s i j mod n 1
si s0 mod12 5 0 4 6 2 3
5 9 B 7 8 A
Example 12-12, a rotational array from Stravinsky’s Requiem Canticles and the
equation that generates it
Because all the horizontal rows of a rotational array begin with the same
pitch class, the rows can be conveniently used for axes of an n-cube.
Rotational arrays have the same number of horizontal rows as the
cardinality of the series that generate them; therefore, an n-cube that uses
all of the rows as axes would have the same number of dimensions as the
cardinality of the series. These objects, coined here, are called n-rotational
cubes. The rotational arrays common to Stravinsky are mostly
hexachordal, but just as with the n-cubes presented earlier in this essay, the
series can be of any cardinality with or without pitch-class duplications.
Thus n-rotational cubes can have any number of dimensions. The
generalized equation for n-rotational cubes is shown below. The indices
supply the coordinates for the n dimensions. The superscript shows that
the axes are constructed from the series rotations as a function of S (i.e.,
Oir Si ).
252 Chapter Twelve
O ijS
1
xO ( S )xO 2
( S ) xxO n ( S ) ¨

§ si s ·
j 1 mod n 1 s1 s k 2 mod n 1 s2 ¸
mod12
n
¨¨ s ¸¸

© n n mod n
1
sn
¹
The quantities of the pitch classes in rotational arrays are symmetrical

around the pitch class in the origin.7 Similarly, the quantities of the pitch
classes in n-rotational cubes are symmetrical also. However, the quantity
of each pitch class in an n-rotational cube is n 1n1 times more than in a
rotational array made from the same series.
Summary
This chapter revisited the application of a 4-cube to Boulez’s
Structures 1a—an advantageous approach in that the dimensions in the 4-
cube correspond to the symmetric/antisymmetric design of the
composition.8 Then, characteristics of n-cubes were examined more fully,
adding details to the original tesseract analysis of Structures 1a, so that n-
cubes with any number of dimensions could be propagated and
represented by their algebraic equations. This additional information
allowed for the two series from Stravinsky’s Requiem Canticles to be built
into another 4-cube. The two series and their inverted forms were the axes
of the 4-cube, showing that axes derived from any type of transformation
can be the basis for n-cubes—even ones that are not the usual twelve-tone
operations. Since any type of transformation can derive axes, the
horizontal rows of rotational arrays were then taken for axes of serial
objects called n-rotational cubes, where the number of dimensions is equal
to the number of notes in the series. N-rotational cubes and the other n-
cubes presented in this essay are just a few of many types of higher-
dimensional serial objects that can enhance our conception of music.
Serial N-Cubes 253
Notes
1
This paper uses the term series instead of row so that row can be reserved for
another usage, which is horizontal row in conjunction with vertical column. The
term array is used instead of matrix to correspond to correct mathematical usage.
Furthermore, the series in this paper are twelve-tone and of cardinality 12, but
series of different lengths with or without pitch-class duplications can be applied to
the concepts in this essay.
2
A right-handed system can be determined as follows: point the fingers of one’s
right hand in the direction of the first axis, and then curl the fingers around to point
in the direction of the second axis; the thumb will then extend in the direction the
third axis should point.
3
Paul Lombardi and Michael Wester, “A Tesseract in Boulez’s Structures 1a,”
Music Theory Spectrum 30/2 (Fall 2008): 339–59.
4
These generalized symmetries, as well as the 3D ones that follow later in this
essay, are verified in Lombardi and Wester, pp. 348 and 350.
5
Andrew Mead, “Some Implications of the Pitch Class/Order Number
Isomorphism Inherent in the Twelve-Tone System: Part One,” Perspectives of New
Music 26/2 (Summer 1988): 96–163.
6
David Lewin, “On Certain Techniques of Re-Ordering in Serial Music,” Journal
of Music Theory 10/2 (Winter 1966): 276–87.
7
Paul Lombardi, “A Symmetrical Property of Rotational Arrays in Stravinsky’s
Late Music,” Indiana Theory Review 25 (2007): 81–82.
8
See Lombardi and Wester.
CHAPTER THIRTEEN
TEMPORAL STRUCTURING
IN COLIN MATTHEWS’S ELEVEN STUDIES
[INVENTIONS] IN VELOCITY
ALEKSANDRA VOJCIC
Straightforward [auffrichtig] instruction, in which amateurs of the

keyboard, and especially the eager ones, are shown a clear way not only
(1) of learning to play cleanly in two voices, but also, after further
progress, (2) of dealing correctly and satisfactorily with three obbligato
parts; at the same time not only getting good inventiones, but developing
the same satisfactorily, and above all arriving at a cantabile manner in
playing, all the while acquiring a strong foretaste of composition. (J. S.
Bach)
A young pianist playing Bach’s two-part inventions and three-part

sinfonias may wonder about their intended purpose—unlike many of the
preludes from the Well-Tempered Clavier, most of the inventions are
fundamentally not pianistic and feel awkward at the keyboard. This young
student may then become convinced, such as the current author once was,
that their piano teacher’s choice of repertoire was precisely due to this
awkwardness. Having later studied these works in analyses classes helped
with the understanding of succinct compositional ideas unique to each
short work, but fundamentally altered my understanding of their purpose: I
now believed they were instructions in composition, rather than keyboard
playing. However, in the preface to the score, Bach alluded to both types
of instruction: his intent was to groom the student to become a complete
musician, one capable of understanding the compositional plan and
executing it with utmost clarity of intent.
An initial encounter with Colin Matthews’s Eleven Studies in Velocity
for piano (1987) can engender a similar effect.1 Each of the eleven brief
works is based on a clearly defined compositional plan that is explored in
Temporal Structuring in Eleven Studies [Inventions] in Velocity 255
minute detail. Polyphony is featured throughout, but is largely non-

imitative and temporally derived. As a result, structural ideas in the
domain of rhythm frequently take precedence over the exact choice of
pitches even when those choices are far from random: the composer
admits to relying on “elements of serial technique and methods of note
generation” that he appropriated and developed in a manner particular to
him (Bruce 2006). However, the idea that there is a structured pre-
compositional plan, so that “to a certain extent every new piece has a bit
of reinventing the wheel about it,” transcends the domain of pitch and is
perhaps more immediately intuited from the manner by which Matthews
shapes his pieces from a temporal perspective (Bruce 2006). Matthews
prefaces his work in a manner similar to Bach—these studies are not about
explicit pianistic virtuosity as much as they are about the virtuosity
implicit in pianistically realizing a virtuosic compositional aim: 2
Although, as the title implies, these studies are all fast, they are not in the
first place studies in pianistic virtuosity – I am no pianist myself – so much
as studies in various ways of composing fast music. All are short and
concentrated, with few lasting more than a minute. Since the order of
performance and the number of studies played (there is not obligation to
play them all) is largely up to the pianist, there is no point in attempting to
describe them, except to say that in a complete version one (which should
be placed near the end) is a nocturne, with the only slow textures of the
eleven, and the final piece is in the nature of a toccata.
Currently, there is little interpretive attention devoted to Colin

Matthews’s work, and none on this impressive collection of piano studies.
In a brief review of Matthews’s music from 1975–90, David Wright
discusses style evolution and composition preferences in no fewer than
fourteen works. Eleven Studies in Velocity appear only in the appendix
comprising twenty-three selected works from that period (Wright 1990).
Similarly, a review of the sole commercially available recording of this
collection by Nicholas Unwin dwells on the other work from the same
album, Tippett’s Sonata, for several pages and has the following succinct
evaluation of Matthews’s work and the recording, here quoted in its
entirety: “…Studies in Velocity demand a consistently high technical
ability; each study is dynamically rhythmic, often very percussive, and—
almost always—fast” (Thomas 1996, 51).
Among the short few vignettes on these pieces, Richard Drakeford’s
informative paragraph expresses a clear admiration for the work
(Drakeford 1991, 448):
256 Chapter Thirteen
There is a hint of the aleatoric in the composer's willingness for these

pieces to be played in a variety of orders, creating through different
performances rather the effect of a 'mobile', but each study is sharply and
vividly characterized, always rhythmically vital, and with a wide variety of
textures, from the predominantly single line of the third (left hand) study to
the poetic harmonic writing and luminous sonorities of the tenth. In short,
this is a superb set of pieces, again requiring virtuosity from the pianist, and
from the listener a quick ear and an open mind.
The rhythmic vitality and vivid characterization Drakeford points to are

partly the motivation for my analyses that follow. The three studies from
the collection chosen for this chapter represent a small, self-contained
compendium of rhythmic processes that include recurring composite time-
signature patterns (that I identify as meta-measures), pulse streams,
structural markers in pulse streams, tempo modulation, and pulse-stream
polyphony. Additional analytical focus is devoted to the limited scope of
generating material. Two of the studies (Nos. 4 and 5) are arguably based
on the same short figure of <Long, Short> durations, and the overall form
is generated by temporal changes systematically applied to this basic unit.
Addressing his long-standing flirtation with American Minimalism,
Matthews, apparently harshly, declares little interest for styles whose pre-
compositional plan fits on the back of a postage stamp (Galloway 2001).3
The fascinating aspect of this statement is that the pre-compositional plans
of many of Matthews’s studies could arguably also fit on the back of a
(albeit larger) postage stamp, but the simultaneous sparseness of means
and density of expression in the resulting narrative often resemble Anton
Webern’s approach more than that of Reich.4 There is also somewhat of a
parallel in Matthews’ statement to that made by Ligeti, who also fruitfully
engaged with American Minimalists: “…Why a certain form is ready or
not ready I cannot say…with me, the plan and the piece develop at the
same time. I don’t believe in making plans” (Service 2003).5 Yet, as
numerous analyses have demonstrated, Ligeti’s works do indeed originate
from a well-defined plan, but the composer frequently interferes with
those plans, particularly when they involve rhythmic processes.6
Temporal Structuring
The primary structural backdrop for temporal events more generally,
and for these works more specifically, is here defined through the concept
of rhythmic hierarchy, comprising three temporal strata: (1) the formal
foreground encompasses pulses (subdivision of tactus), the tactus or the
counting beat, and tactus groupings; (2) the middleground consists of
measure groups and other phrase-level groupings whose boundaries are
structurally defined; and (3) the formal background, largely equivalent to

rhythmic scaffolding, is delineated by structural markers that indicate
pacing or formal segmentation.7 While structural markers primarily
emerge as a series of prominent rhythmic events that parse the time flow
and/or delineate formal segments, other elements of musical fabric, such
as harmonic rhythm, can represent a primary vehicle of series definition, a
point I will demonstrate via Matthews’s Study 1.8
In the context of defining structural markers I will also utilize an
analytical representation called the “pulse-stream graph.” Pulse streams
themselves are indicative of textural strata that are diverse in terms of
rhythm, or meter, or both. In generalizing textural strata, the following
two categories emerge: a) a single pulse stream can represent a sequence
of events unified by some parameter in the domain of rhythm, register, or
formal function; and b) pulse streams in combination form rhythmic
polyphony without reliance on accents or evenly spaced beats.9 The
second definition of pulse streams appears more vividly illustrated in the
analyses of formal and rhythmic middleground entities in the presence of
rhythmic polyphony. However, in addition to its role in rhythmic
polyphony, a single pulse stream can also stand for a series of structural
markers representing a unary phenomenon. The first definition of pulse
streams is demonstrated via Study 1, whereas the second definition
features prominently in the temporal structuring of Study 4 and Study 5.
Musical examples from Study 1 examine structural markers in order to
demonstrate how a pulse stream with two similar durational contours
articulates two sections of the form. Here, the structural markers are not
defined exclusively as rhythmic events, as they represent interaction
between harmonic events and their background rhythmic component.10 In
addition to the two entwined and swiftly moving upper lines, Study 1 also
features a prominent, slowly paced, bass line consisting of four different
trichords. The work begins with a metrically ambiguous gesture under a
fermata, and this gesture is repeated midway through the piece after a
general pause—an eighth-note rest in all parts, following a comma
indicated by Matthews; see Example 13-1. This rest in bar 14, boxed in
Example 13-1, is unique in all of twenty-six bars and represents a metric
hiatus.11 The addition of this eighth-note rest before the second section is
also the reason why bar 14 carries a time signature of 3/8 in contrast to the
corresponding moment in bar 1, which is notated in 2/8 time.

Example 13-1, Matthews, Study 1: comparable thematic events at the onset of two
sections— bb. 1–3 <fermata, trichord 1> and bb. 14–16 <rest, trichord 1>
The left-hand bass line is based on a chord sequence that groups

trichords into two sets of four. The summary of the repeated progression
of four trichords appears in Example 13-2. The two sets of four trichords,
representing repeated pitch class sets, are separated by the rest in bar 14
and divide the study into two equal halves. In Example 13-1, the
corresponding trichords at the onset of each section are indicated with
boxes. The chord changes occur every three to four notated bars, and each
change of harmony corresponds to a significant textural and formal marker
in Study 1. Entries of the chordal bass line represent a background pulse
stream that is roughly comparable, in terms of its structural importance, to
a pulse stream based on cadential markers, since the bass line is treated
conventionally here, as a “foundation, a guide for upper voices in a

polyphonic texture” (Rupprecht 2005, 28).
Bars: 1Ǧ13 15Ǧ26
Example 13-2, Matthews, Study 1: repeated progression of four trichords
A pulse-stream graph of the chordal bass line is shown in Example 13-

3.12 Durations of the four sustained trichords in the first section
correspond to a hastening pattern of eighth notes <8, 8, 7, 6>, while the
second section starts more broadly <10, 9, 7, 6> but ends rhyming with
section one. The graph appears to have two sloping curvatures,
corresponding to the bass-line pattern of the two sections. The downward
slope of the curved lines in the graph embodies progressively shorter
durations in the trichord sequence; these curves make the hastening of the
“harmonic” rhythm towards the cadence visually clear. Hence, the two
curves correspond to the two sections in Study 1 and depict its two-part
form. The steeper curvature of the second series of trichords may echo
Matthews’s concern with the concept of intensification, which he
customarily explores through the concepts of speed and dynamism: apart
from instinctive modifications of temporal pacing, Matthews also relies on
harmony as well as motives to create momentum (Bruce 2006). These are
precisely the two parameters featured in the current examples.

Example 13-3, Matthews, Study 1: pulse stream graph of the bass line, delineating
a two-part form
While Study 1 defines the overall form primarily through harmonic

rhythm, the three sections of Study 5 are related by a common foreground
motive, <Long, Short>, and the process of tempo modulation. Each
section is based on a different combination of duple and triple time
signatures. The proportional 3:4:5 tempo relation between the sections is
a result of the arithmetic expansion of the pervasive <Long, Short>
motive, in each section by an additional sixteenth note. The sixteenth note
is an agent of “common currency” in the tempo modulation between the
sections, but it represents the chronos protos and the sub-pulse level, not
the pulse (the pulse representing a subdivision of tactus or the counting
beat).13 Study 5 is therefore based on an isochronous tactus within each of
the three sections and a common chronos protos unit between the
sections.14
The main motivic element is a persistent rhythmic figure <L, S>; see
Example 13-4a. The time signatures in all three sections are representative
of composite meters, whose periodicity is defined by the recursion of
<triple, duple> meter.15 These recurring composite metric patterns do not
correlate with traditional hypermetric periodicities, and I term them meta-
measures. In short, meta-measures are recurring composite metric patterns
that organize tactus beats and tactus groups into middleground
periodicities. Simple meta-measure prototypes, like the ones featured in
Study 5, share the denominator, while complex prototypes do not.16

Categories are identical for both prototypes, and include layering of meta-
measures with rhythmic polyphony, tempo modulation, and flexible
cycles.
Example 13-4a, Matthews, Study 5: meta-measures in the first section (bb. 1–11
only)
In the opening section of Study 5, two bars of 9/16 are followed by

three bars of 12/16 in a fast tempo of MM=ca. 216 for a dotted eighth
note. The simple meta-measure <2 x 9/16, 3 x 12/16> initially carries
dynamic accents that bring out the perceived tactus—9/16 is played “in
one” and 12/16 “in two” and the effect is that of two longer beats
(MM=72) followed by two shorter beats (MM=108); see the annotations
in Example 13-4a.17 If this format were to persist, we would be hearing a
back-and-forth between two kinds of beats moving at different speeds—

there would be a tempo modulation built into the meta-measure pattern.
However, the notated accents in the end serve to enliven the continuum of
<Long, Short> figures in the rhythmic foreground and eventually draw our
attention to the level of the dotted eighth-long beats whose speed is
proportionally related to the proportionally slower tactus of the latter
sections. While the middleground pattern remains unchanged during the
course of the four statements of the meta-measure pattern in this section,
with the change in accent placement Matthews challenges our perception
of metric accent and, consequently, which level of beat constitutes tactus.
In this manner, the first type of variation in the work is the subtle interplay
between the two levels of beat structure.
In the <3/4, 2/4> flexible cycles of the second section (bb. 21–43 in
Example 13-4b), the exact number of bars in the <triple meter, duple
meter> pattern keeps varying. In general, the flexible cycle represents a
principle of pattern variation applied to a core pattern, and is exemplified
by a generalized ordered series of two or more values; here the value is a
metric unit. The metric pattern of the first section changes in the second
section in response to the lengthening of the long note (here a dotted
eighth note) in the persistent <L, S> foreground motive, and the tactus
speed slows down to MM=ca.160 for a quarter note. This speed would
correspond to a quarter note of the previous section, were there to be one.
Instead, the common denominator between the sections is the sixteenth
note, too short to be heard as tactus or pulse, yet an integral part of the
main motive as the only unchanging currency in the <L, S> figures that
persist throughout the sections.18 Rhythmic variety is provided by the
varying beat-group patterns in the flexible cycles <3/4, 2/4>, each marked
with a bracket in Example 13-4b.
In addition to the varying number of bars within each meta-measure
(<3/4, 2/4> flexible cycles), the rhythmic pattern is further enlivened by an
expanding motivic pattern delineated by sforzando dyads (each boxed in
Example 13-4b), at first appearing only in the left hand. These dyads (or
chords when dyads occur one in each hand) outline the following,
expanding, intervals of quarter-note multiples (* = less one sixteenth
note): <[4 6] [*4 8] [*5 11] *10 *18>. This series of chords is shown in
the pulse-stream graph in Example 13-5 and this graph shows a single
pulse stream receding from the foreground deeper into the
middleground.19 There are three dyads in the first meta-measure of the
second section, but none in the second meta-measure that, incidentally,
represents the core metric pattern without any duplicated units (simply
<3/4, 2/4>). The third meta-measure contains two dyads, whereas the
fourth and the fifth iteration have one in each hand and only at the end of
the two-beat unit (and not in relation to the notated bar line!). In the
process of expansion of their inter-onset interval, these sforzando chords
gradually assume structural function as they begin to delineate the last
three meta-measure patterns. The last meta-measure (bb. 38–43) helps
form a transition to the third section by combining the triple meter of the
second section (3/4) with a different duple meter of the third section
(21/2/4). The last sforzando chord appears at the end of the duple term in
this transitional meta-measure as well.
Example 13-4b, Matthews, Study 5: flexible cycle meta-measures and expanding

chord series in section II, with transitional meta-measure section III, and tempo
modulation in a 4:5 ratio (MM=160oMM=128)
Example 13-5, Matthews, Study 5: pulse-stream graph of the sforzando series in

bb. 21–43
The final section of Study 5, with interesting time signatures of 2½ and

3¾ quarter notes, is also based on the <L, S> motive, this time lengthening
the “long” note to a quarter (thus continuing the “add-the-sixteenth note”
progression). Unlike the transition between the first two sections, there is
now a transition between the second and third sections in form of a mixed
meta-measure (last line in Example 13-4b). The triple element is in 3/4,
while the duple element is in 21/2/4 time.20 This union is sealed with the
last sforzando dyad of the expanding series that governed flexible cycles
in the second section, appearing at the end of the duple element (after a
two-beat group, regardless of the position of the barline) of the meta-
measure pattern. This is the only sforzando dyad within the confines of
the concluding section. Additionally, while the meta-measure pattern of
the third section fundamentally has two triple-meter bars followed by four
duple-meter bars, the transitional meta-measure is different in that it
contains three bars of each meter.
It is the process of expanding the foreground motive <L, S> that brings
about time signature changes in the three sections. The audible result is
multiple tempi (in the simplest sense, they fall into a 3:4:5 ratio) due to the
arithmetic expansion of the <Long, Short> motive by one unit of the
“common currency” agent—the sixteenth note. The marked tempo
schema <216, 160, 128>, as indicated by the metronome markings, does
not exactly correlate to Matthews’s indications at the double bar lines,

those indicating the sixteenth-note equivalence between the three
sections.21 The process of tempo modulation (and the sixteenth-note
equivalence) determines the exact tempo proportion between the
sections—the actual metronome markings are approximate and for
courtesy only.
Despite some ambiguity in the opening segment, the tactus level in
Study 5 is mostly well defined, and the work relies on a steady tactus in
order to underscore the process of tempo modulation taking place between
the three sections. The pulse level is harder to find, particularly in the
third section, where each tactus is five sixteenths long and the eighth-note
level is bypassed altogether.
In contrast to the multiple composite meters of Study 5, Study 4 (see
Example 13-6) features three time signatures banded into one,
<6+7+8/16>, thus encompassing a composite metric pattern within one
actually notated bar. Another point of distinction between the two studies
is that the composite time signature in Study 4 remains unchanged
throughout, but the interior expression of tactus varies within most bars
and among the sections. If the speed of MM = 96 per dotted quarter note,
as indicated by Matthews, is observed, the metric pattern is likely heard as
a series of three beats rather than beat groups, making it a foreground
rather than a middleground event.
The first six bars explore different ways of illustrating the arithmetic
expansion embedded within the time signature by focusing on the
combination of a longer note value with the pulse subdivision, the
sixteenth note. Bars 1–5 form a kind of palindrome: there is a gradual
expansion of the longer note value within each bar, from an eighth to a
quarter note in bars 1–3, and the process becomes reversed in bars 3–5.22
Since the hands are in rhythmic unison, the texture represents a single
pulse stream, marked “stream 1” in Example 13-6. Despite all the
different figurations in stream 1, we never lose sight of the three (uneven)
tactus beats within each notated bar. The closing bar of stream 1 (bar 6) is
entirely in sixteenth notes and it is the only one of the kind in the whole
study.
Example 13-6, Matthews, Study 4: Homophonic presentation of two pulse-streams

in bb. 1-10
The following section (stream 2 in bars 7–10) highlights the

arithmetically elongated tactus by eliminating most of the sixteenths from
the foreground, except when needed to distinguish between the three long
beats of the metric pattern. The process of defining the long(er) beat in
stream 2 is less forthright than that of stream 1. The breakdown of this
process of differentiation is summarized in Example 13-7, where all of the
numbers (1–6) refer to sixteenth-note durations. While “6” indicates a
dotted quarter equivalent, with only one sound articulation, “6+1”
indicates a beat with two sound articulations. The process of defining the
three progressively longer units of the meta-measure is not mechanical,
even if it is patterned—note the recurring divisions such as 4+3 or 5+3.23
Example 13-7, Matthews, Study 4: defining the tactus in stream 2 (bb.7–10)
Each of the two pulse steams is first presented homophonically, (the

hands are in rhythmic unison) so that they can more easily be heard, and
possibly combined. Indeed, starting in bar 11, the two streams engage in
invertible counterpoint; see Example 13-8. At first the left hand (S-2 in
Example 13-8) is delayed by a dotted eighth note resulting in a metric
canon (the streams are out of phase by a “half beat”). After three
measures, the left-hand stream undergoes an arithmetical collapse as its
tacti get progressively shorter by a sixteenth note. A moment of
convergence (end of bar 15 through bar 17) is followed by new polyphony
as the two hands reverse and S-2, now in the right hand, goes through the
same process of speeding up.
Example 13-8, Matthews, Study 4: Pulse-stream polyphony and arithmetic

contraction in stream 2
The pulse stream graph in Example 13-9 serves to illustrate the process
just surveyed. The stable stream maintains the metric identity of the meta-
measure and is represented by diamond-shaped markers on the graph—it
progresses along a three-step cyclical path of <6, 7, 8> sixteenth-note
durations, which is visually apparent in the graph. Stream 2, however,
draws an analogous, but offset, pattern, and then gets progressively shorter
until each attack reaches the length of only one sixteenth note. In bb.16–
17 the two streams are again synchronized, as evidenced by the last two
pulses in the pulse-stream graph, where S-2 “leaps” from the one-sixteenth
duration back to seven to coincide with S-1, i.e., the square markers and
the diamond markers are overlaid. The interaction of the stable stream
with the hastening one in bars 18–20 is similar to, but dramatically
compressed from, the initial presentation. Exploring the temporal
structuring in Study 4 demonstrates how a basic meta-measure pattern can
become a kind of an abstract referent, even when it is not extant in the
formal middleground. The asymmetrical meter of an arithmetically
expanding series (<6, 7, 8/16>) can indeed serve as an organizing force in
different temporal processes falling under the broad term of rhythmic
polyphony.
ͻ
ͺ
͹
͸
Duration
ͷ ͳ
Ͷ ʹ
͵
ʹ
ͳ
Ͳ
ͳͲ ʹͲ ͵Ͳ ͶͲ ͷͲ ͸Ͳ ͹Ͳ ͺͲ ͻͲ
ClockTimein1/16ths
Example 13-9, Matthews, Study 4: interaction of two pulse streams (bb.11–17?)
While Study 1 demonstrates the simple, yet effective, organizational

power of a series of structural markers that are both harmonically and
temporally defined, the two analyses of Study 5 and Study 4 show how a
metrical pattern can be a source of metric stability, even when several
rhythmic processes both challenge and rely on the stability of these
asymmetrical metrical structures to engender formal segmentation and
variation. In addition, the process of expansion embedded in the time
signature(s), is driven by the same motive <Long, Short>, illustrated in
different ways in the foreground. The resulting tempo modulation and
composite time signatures are governed by processes of temporal
expansion in Study 5, while the phasing of pulse streams in Study 4, with
one stream contracting, rather than expanding, shapes the formal

middleground through the process of foreground temporal contraction.
In conclusion, the three selected studies are representative of a
compositional manner in which form can be defined through temporal
processes—the modified strophic form of Study 1 is driven by recurring
harmonies with modified harmonic rhythm, whereas the three sections of
Study 5 largely follow the arithmetic expansion of a single duration in a
<Long, Short> motive. The four-and-a-half sections of Study 4 (including
the codetta, bb. 21–23) rely on the same <Long, Short> motive, while
basing the homophonic sections on arithmetic expansions of the tactus
beat, and layering polyphonic sections with an arithmetically contracting
pulse-stream. The resulting vitality and freshness of each of these brief
studies is even more remarkable if one considers the limited conceptual
means by which they are created.
Although Matthews rejects an idea of belonging to any school of
composition, his affinity for compositional craft that is timeless in
conception and fresh in delivery seems very reminiscent of J. S. Bach.24
In this manner, Matthews’s studies resemble Bach’s process of composing
inventions, more than the works that Matthews titled “inventions.”
Notes
1
Aside from Matthews’s interests and connections with Debussy, Britten, Holst,
and Mahler, his connection to Bach can be superficially witnessed through the
sheer number of works that evoke baroque themes, even those specifically
reminiscent of Bach. In addition to a number of toccatas, Matthews’s list of works
written around the same time as the Studies in Velocity includes: (a) several types
of chaconne including Three Part Chaconne for string trio and piano left hand
(1989);1 and (b) multiple references to inventions, such as Triptych (tri-partite
inventions) for piano quintet (1984), Two Part Invention for 19 players (1988,
dedicated to Elliott Carter), or a later work, Three Preludes (2003), the third of
which Matthews describes as a “two-part invention in form of an ostinato with an
ever-growing crescendo.”
2
These program notes are available from the publisher’s website,
http://www.fabermusic.com/Repertoire-Details.aspx?ID=1184.
3
Matthews admits that his work Hidden Variables is “a rather vicious attack on
minimalism, which, although useful to me at an early stage, is something that I
think any composer should outgrow” (Galloway 2001, 16).
4
Reich himself notes kinship between his phase pieces and Webern’s music as
musical structures that conceptually originate in the same way (from a canon or
round), but sound dramatically different as completed works (Reich 2002, 71).
5
Also of note is Matthews’s admission of admiration for Birtwistle and Ligeti;
others, among his contemporaries, represented on this short list include Knussen
and Lindberg (Galloway 2001).
6
Many of Ligeti’s etudes spring from a succinct compositional idea, which is
pursued throughout an etude with rare instances of episodic contrast or
introduction of new material.
7
Structural markers, as prominent rhythmic events, delineate formal segments and
represent a type of rhythmic scaffolding. Structural markers include convergence
points in long-range polyrhythms, sectional shifts in phasing processes, and events
that represent a point of departure from one type of rhythmic process to another
(Vojcic 2007).
8
Matthews indicates that studies are individually numbered without determining
their order in performance, though he does recommend certain guidelines for a
possible sequence. The published order, which will be used in this study, is the
order chosen by the pianist for the first performance. The first of the eleven
studies in this published order will be called Study 1, for convenience’s sake.
9
John Roeder also prominently uses the concept of pulse streams, but his
analytical approach represents “rhythmic polyphony as two or more concurrent
‘pulse streams’ created by regularly recurring accents.” This often results in an
“irregular surface as the sum of several concurrent regular continuities” (Roeder
1994, 232).
10
As previously indicated, structural markers do not have to exclusively represent
rhythmic events such as patterns of duration or convergence points between pulse
streams. For example the rate of incidence of cadence points in common-practice
tonal music, as an agent of structural demarcation, can be envisioned as a series of
structural pulsations that can exist as a fairly background phenomenon. Mapping
cadences throughout a work or a movement can tell us much about its formal
proportions and the resulting view can analytically be similarly represented by a
pulse stream graph.
11
Metric hiatus is a term used by Christopher Hasty to denote "a break between
the realization of projected potential and a new beginning," meaning it takes place
when a projection is interrupted and has the effect of a restart—counting begins
anew (Hasty 1997, 88).
12
The pulse stream graph of the trichord sequence in Study 1 is a two-dimensional
representation, indicating the importance of the inter-onset interval between
structural markers as signposts for the temporal organization of this piece. Inter-
onset interval is a term used with a similar meaning by Dowling and Harwood
1986.
13
Chronos protos (Greek for a unit of time) denotes the smallest common
denominator between higher-level metric units.
14
Due to sheer speed, one possible exception to the “isochronous tactus” template
is the first section; see the discussion below.
15
As previously stated, subsequent arithmetic expansion of the long note, in each
section by an additional sixteenth note, brings about tempo modulation between
the sections, as well as the new time signatures.
16
Examples of complex prototypes include patterns such as <9/8, 3/4, 2/4>, <5/8,
2/4>, etc… For more detailed discussion of these phenomena, see Vojcic 2007.
17
The notated metronomic marking of eighth=216 is rather fast, and falls outside
the 60–190 speed range for tactus, which I adopted as perceptual boundaries for
tactus speed (after 50–200 boundaries established by Paul Fraisse). Nicolas
Unwin’s Metier recording adopts a markedly slower speed of ca.180 for the initial
dotted eighth-note, which would bring out a dotted-eighth-note tactus. In this
instance the performance speed is of paramount importance.
18
In this section, the <L, S> figure comprises a dotted eighth and a sixteenth, and
the tactus consistently corresponds to a quarter note.
19
Other textural elements, which remain steady, articulating most of the quarter-
note beats in the foreground, are not present on the pulse-stream graph in order to
avoid clutter.
20
In the last three bars of Example 13-4b (beginning of the third section), the long
note is a quarter note, which is one-sixteenth note longer that the dotted eighth
note from the previous section. Correspondingly, the overall tactus speed is
slower, MM=ca.128, rather than MM=160.
21
For instance: a group of four sixteenths in a tempo of MM=160 proportionally
modulates into a group of three sixteenths at MM=129.6, rather than MM=128 as
indicated. However, the difference between these speeds does not meet the
perceptual minimum of “just noticeable difference,” or the Weber Fraction, which,
in the psychological present (0.5–2 sec) appears to correspond to ca. 5% of
minimal requisite difference from the established (temporal) periodicity. Epstein
1995 summarizes the existing research on this topic at length. It is noteworthy that
Matthews chooses to “round-off” his metronome markings, as well as approximate
the duple/triple meter with curious time signatures in the third section of Study 5.
Other composers (like Elliott Carter) might have opted for the more precise, even
if unnoticeably different, notation of local tactus speeds.
22
In keeping with my previous distinction between meta-measures and notated
measures, notational units are referred to as bars (separated by barlines), reserving
the term measure for meta-measures that may include two or more notated bars. In
this instance, the composite metric pattern comprises only three tactus beats,
spanning only one notated bar and does not qualify as a meta-measure, a
middleground entity usually exceeding five tactus beats.
23
Each number in the chart represents duration in sixteenth notes with either one
or two sound articulations per tactus beat. The arrows point to the rhythmic
patterns retained from one bar to the next.
24
Matthews makes no excuses about his strong opinions on contemporary schools
of composition: “…(like Groucho, I wouldn’t want to belong one that would have
me as a member)—particularly cliques like the complexity people, or the French
spectralists, who have to form a mutual support group to reinforce their “us against
the world” outlook (Galloway 2001, 16).
References
Bach, J. S. 1978. Inventions and Sinfonias BWV 722–801, ed. Rudolph
Steglich. Duisburg: G. Henle Verlag.
Benadon, Fernando. 2004. “Towards a Theory of Tempo Modulation.”
Paper presented at the 8th International Conference on Music
Perception and Cognition, Evanston, IL.
Bruce, David. 2012. “Colin Matthews Interview.” Composition: Today
website.
http://www.compositiontoday.com/interviews/colin_matthews.asp
(accessed 15 March).
Drakeford, Richard. 1991. Review: “Keyboard Diversity.” The Musical
Times 132/1783: 448.
Dowling, Jay and Harwood, Dane. 1986. “Rhythm and the Organization
of Time.” In Music Cognition. Orlando: Academic Press,
pp. 178–201.
Epstein, David. 1995. Shaping Time: Music, the Brain, and Performance.
New York: Schirmer.
Faber Music website. 2012. Program notes for Eleven Studies in Velocity.
http://www.fabermusic.com/Repertoire-Details.aspx?ID=1184
(accessed 5 May).
Fraisse, Paul. 1978. “Time and Rhythm Perception.” In Handbook for
Perception volume 8: Perceptual Coding, ed. E. C. Carterette and M.
P. Friedmans. New York: Academic Press, pp. 203–54.
Galloway, Michael and Matthews, Colin. 2001. “Composer in Interview:
Colin Matthews.” Tempo (New Series) 215: 15–16.
Hasty, Christopher. 1997. Meter as Rhythm. New York: Oxford
University Press.
—. 1981. “Rhythm in Post-Tonal Music: Preliminary Questions of
Duration and Motion.” Journal of Music Theory 25/2: 183–216.
—. 1984. “Phrase Formation in Post-Tonal Music.” Journal of Music
Theory 28: 186–90.
London, Justin. 2002. "Temporal Asymmetries as Period Markers in
Isochronous and Non-Isochronous Meters." Paper presented at the 7th
International Conference on Music Perception and Cognition, Sydney,
Australia.
Matthews, Colin. 1987. Eleven Studies in Velocity. London: Faber
Music.
Mead, Andrew. 2007. “On Tempo Relations.” Perspectives of New
Music 45/1: 64–108.
Reich, Steve. 2002. “Postscript to a Brief Study of Balinese and African

Music.” In Writings on Music 1965–2000, ed. Paul Hillier. New York:
Oxford University Press, pp. 69–71.
Robison, Brian. 1999. “Carmen Arcadiae Mechanicae Perpetuum:
Toward a Methodology for Analyzing Harrison Birtwistle’s Music
Since 1977.” DMA dissertation, Cornell University.
Roeder, John. 1994. “Interacting Pulse Streams in Schoenberg’s Atonal
Polyphony.” Music Theory Spectrum 16/2: 231–49.
Rupprecht, Philip. 2005. “Above and Beyond the Bass: Harmony and
Texture in George Benjamin’s Viola, Viola.” Tempo: A Quarterly
Review of Modern Music 59: 28–38.
Service, Tom. 2003. “Prelude for Pygmies.” In The Guardian, 16
October.
Thomas, Philip. 1996. Review of Metier Recordings. Tempo (New Series)
195: 47–71.
Unwin, Nicholas. 1995. 20th-century British Piano Music. Preston:
Metier Sound and Vision.
Vojcic, Aleksandra. 2007. “Rhythm as Form: Rhythmic Hierarchy in
Later Twentieth-Century Piano Music.” Ph.D dissertation, City
University of New York, Graduate Center.
Wright, David C. H. 2001. “Matthews, Colin.” The New Grove
Dictionary of Music and Musicians, edited by Stanley Sadie and John
Tyrrell. London: Macmillan.
—. 1990. “Colin Matthews: An Introduction to His Recent Music.” The
Musical Times 131/1770: 418–21.

CONTRIBUTORS
Nathan Baker has been the Music Theory Coordinator at Casper College
(Wyoming) since fall of 2009. Originally from Missoula, MT, Baker
received a B. Mus. in Music Education and Music Composition from Utah
State University in 2002, an M.A. in Music Theory from the University of
Oregon in 2006, and is nearing completion of the Ph.D. in Music Theory
(with a supporting area in Music History) from the University of Oregon.
In 2004 he was the recipient of the University of Oregon School of
Music’s Excellence in Teaching award.
Baker’s research has ranged from neo-Riemannian theory and atonality
to the study of harmony and form in video game music; other topics of
interest include world music theory and music theory pedagogy. His
master’s thesis, “Neo-Riemannian Perspectives on the Early Music of
Arnold Schoenberg,” discovered a link between the harmonic progressions
found in late Romantic composers and the progressions used in
Schoenberg’s early atonal period. Baker’s proposed doctoral dissertation
will focus on “Form, Style, and Meaning in Japanese Video Game Music.”
A member of the Society for Music Theory, Baker has presented his
findings at the West Coast Conference for Music Theory and Analysis.
Baker’s other musical activities include playing trombone, composing,
and arranging music.
Christine Boone is Assistant Professor of Music Theory at Indiana State

University. She received her B.M. in vocal performance at Indiana
University and both her M.M. and Ph.D. in music theory at the University
of Texas. Her research interests are centered around popular music, and
she has presented papers on the Beatles in both the United States and the
United Kingdom. Her dissertation was entitled “Mashups: History,
Legality, and Aesthetics,” and it is on this subject that her current research
is focused.
Boone, a soprano, is also an active performer in both choral ensembles
and solo work.

276 Contributors
Jack Boss holds the Ph.D. in music from Yale University, where he
studied with Allen Forte, Claude Palisca and David Lewin. He is
presently Associate Professor of Music Theory at the University of Oregon.
He has also taught at Brigham Young University, Ball State University
and Yale.
He has published a number of influential articles on Schoenberg’s,
Beethoven’s and Bernard Rands’s music in Music Theory Spectrum,
Journal of Music Theory, Perspectives of New Music, Intégral, Music
Theory Online and Gamut. His book reviews can be found in the Journal
of Music Theory, Music Theory Online and Notes. He edited the essay
collection Musical Currents from the Left Coast (Cambridge Scholars,
2008) together with Bruce Quaglia, and has recently completed a book-
length survey of Schoenberg’s twelve-tone music that will be published by
Cambridge University Press.
Boss has also given numerous conference presentations on Schoenberg’s
music and related topics in the United States (AMS/SMT and CMS, as
well as many of the regional societies), Canada, England (SMA) and
Ireland (Dublin International Conference on Music Analysis). He served
from 1989-91 as Reviews Editor, Associate Editor and Editor in Chief of
the Journal of Music Theory, from 2000-05 as Reviews Editor for Music
Theory Online, and from 2005-10 on the Editorial Review Board of the
Journal of Music Theory Pedagogy. He was elected President of the West
Coast Conference of Music Theory and Analysis in 2003.
Boss’s other musical interests include singing, conducting, playing
clarinet (accompanied by his wife, pianist SunHwa Lee Boss) and
arranging music for the largest Baptist congregation in Eugene, Oregon.
Gary W. Don is Associate Professor of Music at University of Wisconsin-

Eau Claire. He taught theory and aural skills at the University of
Washington and Skidmore College before joining the UW-Eau Claire
faculty. His research interests include Goethe’s influence on music
theorists of the nineteenth and twentieth centuries, overtone structures in
the music of Debussy, modality in the music of Prokofiev, and theory
pedagogy. He has presented papers on these topics at the West Coast
Conference of Theory and Analysis, Music Theory Midwest, the Great
Lakes Chapter and national conferences of the College Music Society, and
at national meetings of the Society for Music Theory. He has published
articles in Computer Music Journal, In Theory Only, Perspectives of New
Music, and Music Theory Spectrum.

Andrew Gades is a PhD student in music theory at Florida State

University. His research focuses on critical theory, music/text relations,
and the analysis of twentieth and twenty-first century American music. He
has published in Mosaic: A Journal of Music Research, and was
designated a College Teaching Fellow for 2009-2010.
Erik Heine is an Associate Professor of Music Theory at the Wanda L.

Bass School of Music, Oklahoma City University, where he has taught
since 2005. His dissertation, The Music of Dmitri Shostakovich in “The
Gadfly,” “Hamlet,” and “King Lear,” analyzes the cues composed for the
three films, both independently and as they contribute to each film's
narrative structure.
He has presented papers at international conferences concerning film
music, both of Shostakovich and other composers, including multiple
international conferences during the centenary celebration of Shostakovich’s
birth in 2006, as well as papers concerning music theory and aural skills
pedagogy. He has also been published in the DSCH Journal and the
journal Music and Letters. His current research concerns music from the
films Signs and Solaris, the tintinnabuli works of Arvo Pärt, and the music
of Edgard Varèse.
Heine earned a Ph.D. in Music Theory from the University of Texas at
Austin, an M.M. in Music Theory from the University of Arizona, and a
B.M. in Percussion Performance from Illinois Wesleyan University. His
teachers have included Jim Buhler, Tim Kolosick, David Neumeyer, and
Ed Pearsall.
Paul Lombardi holds a Ph.D. in music composition from the University

of Oregon, and has studied composition with David Crumb, Robert Kyr,
Stephen Blumberg, and Leo Eylar. He is presently Assistant Professor of
Music Theory and Composition at the University of South Dakota. His
music has been performed in more than 20 states across the US, as well as
in other areas in North America, South America and Europe. Recordings
of his music are available from Capstone Records, Zerx Records, and
ERMMedia. Many groups have played his music, notably the Kiev
Philharmonic, the East Coast Composers Ensemble, Third Angle, and
Hundredth Monkey. He is the winner of the 2011 Renée B. Fisher Piano
Composition Competition, and he has received numerous commissions
including one by Oregon Bach Festival Composers Symposium in honor
of George Crumb on the occasion of his 75th birthday. Some of his scores
are published in the Anthology of Contemporary Concert Music (currently
in press).

278 Contributors
Lombardi’s theoretical work focuses on mathematics and music, and is

published in Music Theory Spectrum, Indiana Theory Review, Mathematics
and Music, and Mathematics and Computers in Simulation. He has
presented his research at numerous theory conferences, both national and
regional. He was the pianist for the Hundredth Monkey Ensemble from
2000 to 2003, and was a soloist for the Siskiyou Community Orchestra in
1994.
Garrett Michaelsen is earning his Ph.D. in music theory from Indiana

University. His research interests focus on twentieth-century music,
particularly jazz and avant-garde improvisation, and he has given
presentations at numerous national and regional conferences. He was the
2011 recipient of the Mary Wennerstrom Associate Instructor Fellowship,
and has chaired the Indiana University Graduate Theory Association’s
annual conference. He also has had an illustrious career as a jazz
trumpeter, and was a founding member of the Respect Sextet.
Brad Osborn is Visiting Assistant Professor of Music Theory at Ohio

University. His primary area of research concerns experimental
compositional procedures in recently conceived rock genres, or what he
deems “post-millennial experimental rock.” Osborn’s dissertation
specifically addresses the new formal structures that emerge from this
post-millennial corpus, and, along the way, theorizes emergent harmonic,
melodic, timbral, and rhythmic tendencies in the music as well. His
publications in this area include articles in Music Theory Spectrum, Music
Theory Online, and Gamut. Osborn is currently focusing his efforts on a
book-length study that theorizes how these expressive compositional
moments may intersect with a listener’s acquisition of meaning,
specifically, in the music of the English experimental rock band Radiohead.
Tim S. Pack has been an instructor of music theory at the University of

Oregon since fall, 2005. He received a B.A. in music from Huntingdon
College in 1993, an M.M. in composition from Westminster Choir College
of Rider University in 1998, and a Ph.D. in music theory from Indiana
University in 2005. His dissertation, titled "Axial-Tenor Composition in
the Renaissance," examines formal and temporal organization, cantus-
firmus placement and text usage, motivic development, harmony, voice
ranges and functions, and cleffing in five-voice tenor motets and masses of
the fifteenth and sixteenth centuries.
Although specializing in music of the fifteenth and sixteenth centuries,
he has also done research on music of other eras and has given several

presentations at scholarly conferences in Belgium, the Netherlands, Italy,

England, Wales, Australia, and the United States. His publications range
from articles on the motets of Obrecht and Orto to the liturgical works of
Distler and Messiaen; his most recent publications are an essay on
ostinato-tenor composition around 1500 (Brepols, 2012) and an article on
the choral-symphonic works of Boccherini (Ut Orpheus Edizioni, 2013).
Pack continues to expand his interests by examining music of living
composers including Willem Ceuleers, Colin Mawby, George Walker,
Karl Jenkins, Michael Torke, Arvo Pärt, Einojuhani Rautavaara, and
several others.
Stephen Rodgers is associate professor of music theory and musicianship

at the University of Oregon. Prior to coming to Oregon, he taught aural
skills at his alma mater, Lawrence University, where he received his B.A.
in Music and English (magna cum laude). He also taught music theory at
Yale University, where he completed his Ph.D. in 2005, with the help of a
Mellon Fellowship in Humanistic Studies.
Rodgers’s research reflects his interest in placing music analysis in a
broader context by bringing it into contact with musicology, hermeneutics,
literary theory, and other forms of humanistic inquiry. His publications
and presentations have dealt primarily with text-music relations, program
music, rhythm and meter, and musical form. His work has appeared in
Music Theory Online, the Journal of Musicological Research, Current
Musicology, Theoria, Indiana Theory Review, Chamber Music magazine,
and the collection Sounding Off: Theorizing Disability in Music
(Routledge, 2006). Forthcoming articles include a study of Berlioz’s
romances, to be published in Nineteenth-Century Music Review, and an
exploration of sentence forms in Die schöne Müllerin, to be published in
Music Theory Spectrum. Rodgers’s book Form, Program, and Metaphor
in the Music of Berlioz (Cambridge University Press, 2009), the first
devoted to Berlioz’s handling of musical form, considers how the shapes
of Berlioz’s pieces are related to the poetic and dramatic sentiments that
were his very reason for being. Currently he is working on a set of articles
about the music of Fanny Hensel (née Mendelssohn). Rodgers also serves
on the editorial board of Music Theory Online.
A composer as well as a theorist, Rodgers has received numerous
commissions from the Yale Collegium Musicum and the Madison
Children’s Choir, among other ensembles. He is also active as a tenor. He
has performed several lecture-recitals throughout the United States, and is
a member of Vox Resonat, a professional early music ensemble directed
by Eric Mentzel.

280 Contributors
Jessie Layne Thornton is an undergraduate student at the University of

Oregon, majoring in pre-med. She is an accomplished guitarist, and has a
profound interest in analysis and theory, evidenced by her successful
presentations on Schubert’s “Du liebst mich nicht” and Sibelius’s
Pohjola’s Daughter at the 2010 and 2011 meetings of the West Coast
Conference of Music Theory and Analysis.
Aleksandra Vojcic earned bachelors and masters degrees in Piano from

The Juilliard School, and the Ph.D. in Music Theory from the Graduate
Center, CUNY. Her doctoral dissertation is entitled, “Rhythm as Form:
Rhythmic Hierarchy in Later Twentieth-Century Piano Repertoire.” She is
a former faculty member at Juilliard and the University of Massachusetts
at Amherst.
Ms. Vojcic has been a piano soloist with the Belgrade Philharmonic,
the National Repertory Orchestra, New Juilliard Ensemble, Colby
Symphony Orchestra, Juilliard Chamber Orchestra, and the Josip
Slavenski String Orchestra. New York venues have included Steinway
Hall, Kosciuszko Foundation, Alice Tully Hall, and MoMa Summergarden.
She has presented lectures and papers in the UK, Lithuania, Austria, and
the United States and chaired a session at the Music Theory Society of
New York State in 2005. Her recordings include Heavenly Lullabies with
the trio D'Divaz. Vojcic was featured in an award-winning Swiss
documentary Yugodivas. She has appeared in broadcasts on WNYC,
KAJX, and PGP RTB.
Brent Yorgason is an Assistant Professor at Marietta College, where he

teaches music theory. He has previously taught at Indiana University and
the University of Texas at San Antonio. He recently finished his Ph.D. in
music theory at Indiana University; his dissertation is entitled "Dispersal,
Downbeat Space, and Metric Drift: Aspects of Expressive Timing and
Meter." In addition to studies of meter and performance, Brent's research
interests include such diverse topics as Schenkerian analysis, machine
metaphors in music, minimalism and postminimalism, music technology,
and hymnology.
Brent is the Managing Editor of Music Theory Online and serves on the
Networking committee of the Society for Music Theory. He has also
worked extensively as a computer programmer, developing music
pedagogical software such as ETDrill (ear training software), Music
Fundamentals Online, and various timeliner and score annotation tools for
the Variations2 digital music library project at Indiana University.

Analyzing The Music of Living Composers

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Analyzing The Music of Living Composers

Uploaded by

Copyright:

Available Formats

Analyzing the Music of Living Composers

Jack Boss, Brad Osborn,

This book first published 2013

Cambridge Scholars Publishing

12 Back Chapman Street, Newcastle upon Tyne, NE6 2XX, UK

British Library Cataloguing in Publication Data

ISBN (10): 1-4438-4227-3, ISBN (13): 978-1-4438-4227-3

PART I: PROCESS IN THE MUSIC OF LIVING COMPOSERS

Introduction to Part I ................................................................................... 3

Chapter Two .............................................................................................. 27

Chapter Three ............................................................................................ 51

PART II: THE TONAL TRADITION

Introduction to Part II .............................................................................. 105

Chapter Five ............................................................................................ 106

Chapter Six .............................................................................................. 115

Chapter Seven.......................................................................................... 132

PART III: POP MUSIC AND BEYOND

Introduction to Part III............................................................................. 161

Chapter Eight........................................................................................... 164

Chapter Nine............................................................................................ 176

Chapter Ten ............................................................................................. 192

Chapter Eleven ........................................................................................ 218

PART IV: MUSICAL GEOMETRY AND TEMPORAL STRUCTURING

Introduction to Part IV............................................................................. 239

Chapter Twelve ....................................................................................... 240

Chapter Thirteen...................................................................................... 254

In March of 2010, a rather unique scholarly meeting was held in Eugene,

—Jack Boss, September 2012

PROCESS IN THE MUSIC

Part I explores “Process in the Music of Living Composers.” It opens

remarks about decisions made prior to and throughout the compositional

MAKING THE MOST OF MINIMALISM:

In the classic minimalist compositions of the 1970s, the gradual

motivic transformations, the harmonic and rhythmic language of jazz, and

Example 1-1, “Yellow Pages,” violin pattern, m. 4

Example 1-2, “Yellow Pages,” ostinato pattern, m. 5

Example 1-3, “Yellow Pages,” [0 1 0 / 0 1 2] contour motive in violin and piano

Example 1-4, “Yellow Pages,” initial woodwind pattern, m. 7

Table 1-1, Static transposition in “The Yellow Pages,” mm. 9-386

Example 1-5, “Yellow Pages,” ostinato pattern, mm. 9-16

As this slow transpositional process continues, the listener will

The composer compares this slow transpositional process to the

MILK to PAIL MICE to RATS

Figure 1-1, Sample “Laddergrams” or “Word Chains”

This process of “static transposition” is simultaneously combined with

becomes progressively sketchier until it disappears entirely (see Example

Figure 1-2, Simultaneous processes in “The Yellow Pages,” mm. 9-38

The two-measure pattern that was developed by the woodwinds in this

Example 1-8, “Yellow Pages,” beginning of the second section, m. 39

Example 1-9, Emergence of a shifted downbeat in “The Yellow Pages,” m.43

Example 1-10, Alternating meters section in “The Yellow Pages,” m. 56

In the next section (at letter C) a process of alternation begins between

Example 1-12, Repeating motivic cells, each with a [1 0 1 2] contour motive, in

Example 1-13, Shifting rhythmic patterns in “The Yellow Pages,” m. 106

The expected return arrives in m. 143 with a fortissimo statement of

Example 1-15, Closing theme in “The Yellow Pages,” m. 159

Throughout this process of reduction, a new idea has been emerging—

briefly examine this “slightly different application” of the process of static

Figure 1-4, Simultaneous processes in “The Blue Pages,” mm. 7-73

At each repetition of the principal theme, the accompanimental parts—

Table 1-2, Static transposition in “The Blue Pages,” mm. 7-73

In the third movement, entitled “The White Pages,” Torke again

Figure 1-5, Simultaneous processes in “The White Pages,” mm. 13-32