Analysis of 18th- and

19th-Century Musical Works

in the Classical Tradition
Analysis of 18th- and
19th-Century Musical Works
in the Classical Tradition

David Beach
University of Toronto

Ryan McClelland
University of Toronto
First published 2012
by Routledge
711 Third Avenue, New York, NY 10017
Simultaneously published in the UK
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2012 Taylor & Francis
The right of David Beach and Ryan McClelland to be identified as authors of
this work has been asserted by them in accordance with sections 77 and 78 of
the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks, and are used only for identification and explanation without intent
to infringe.
Library of Congress Cataloging in Publication Data
Beach, David, 1938– author.
Analysis of 18th- and 19th-century musical works in the classical tradition/
David Beach, Ryan McClelland.
p. cm.
Includes bibliographical references and index.
1. Musical analysis—Textbooks. 2. Harmony—Textbooks. 3. Musical form—
Textbooks. I. McClelland, Ryan C., author. II. Title.
MT6.B294 2012
780.9′033—dc23 2011042055

ISBN: 978-0-415-80665-7 (hbk)

ISBN: 978-0-415-80666-4 (pbk)
ISBN: 978-0-203-12183-2 (ebk)

Typeset in Bembo
by Florence Production Ltd, Stoodleigh, Devon
Contents

Preface ix

PART I
Principles of Harmony and Phrase Design 1

1 A Review of Diatonic Harmony 3

Designation of Pitch and Chords 3
Tonal Functions 7
Common Sequential Patterns 13
Passing and Neighboring Chords 18
Functions of Individual Chords 23
Summary of Concepts and Terminology 26
Suggested Assignments 27

2 Expanding the Diatonic Palette 32

Secondary Dominants and Diminished Seventh Chords 32
Chromaticized Sequences 38
Diatonic Modulation 46
Modulation from an Initial Major Key 46
A Summary Example: Bach, French Suite No. 5, Gavotte 50
Modulation from an Initial Minor Key 52
A Summary Example: Bach, French Suite No. 2, Menuet 57
Suggested Assignments 61

3 Phrase Design 65
Phrase Design 65
What is a phrase? 65
Phrase Construction and Combination—Part 1: The Period 66
Phrase Construction and Combination—Part 2: The Sentence 70
Phrase Construction and Combination—Part 3: Larger Period Design 73
vi Contents

Phrase Elision and Hypermeter 74

Phrase Elision 74
Hypermeter 78
Phrase Expansion 79
External Phrase Expansions 80
Internal Phrase Expansions 83
A Summary Analysis (Haydn) 93
Suggested Assignments 97

4 Further Expansion of the Harmonic Palette 105

Modal Mixture 105
Two Important Chromatic Chords: The Neapolitan Sixth and the Augmented Sixth 112
Analysis (Beethoven) 118
Modulation 120
Suggested Assignments 136

PART II
Analysis 149

5 Binary and Rounded Binary Forms 151

Binary Form 151
Rounded Binary Form 153
Haydn, Symphony No. 101 in D Major (“Clock”), III 156
Beethoven, Piano Sonata in A Major, op. 26, II 176
Brahms, Piano Sonata in F Minor, op. 5, III 182
Suggested Assignments 193

6 Classical Sonata Form 196

Mozart, Piano Sonata in F Major, K. 332, I 197
Beethoven, Piano Trio in C Minor, op. 1, no. 3, I 211
Suggested Assignments 232

7 Sonata Form in the Nineteenth Century 235

Beethoven, Piano Sonata in F Minor, op. 57, I 236
Brahms, Clarinet (or Viola) Sonata in F Minor, op. 120, no. 1, I 258
Suggested Assignments 277

8 Ternary Form 279

Beethoven, Piano Sonata in D Major, op. 10, no. 3, II 279
Schubert, Impromptu in G Major, op. 90, no. 3 289
Brahms, Intermezzo in B Minor, op. 117, no. 2 300
Suggested Assignments 312
vii Contents

9 Rondo Form 314

Rondo Form 314
Mozart, Piano Sonata in B  Major, K. 333, III 316
Brahms, Violin Sonata in D Minor, op. 108, IV 331
Suggested Assignments 349

Appendix: Notes on Musical Reductions 351

Notes 356
Index of Musical Works 363
Index of Names and Terms 366
Preface

This is a book about analyzing music. What does this mean? Analysis is a process of discover-
ing how a piece of music “works.” A fuller answer to this question depends both on the
methodological choices of the analyst and the type of music under consideration. This text
focuses on instrumental music composed in the eighteenth and nineteenth centuries by Bach,
Haydn, Mozart, Beethoven, Schubert, Schumann, Chopin, and Brahms. All of these
composers created music within a relatively consistent set of organizational principles
referred to as the common-practice tonal system. Central to this system is the domain of harmony—
of chords and their progressions. Of particular importance is the way successive harmonies
relate to one another, as these harmonic progressions define key areas and also create the
sense of a musical journey, with its rising and falling tension, across a composition. Like
much poetry, musical works of common-practice tonality generally fall into a specific type
of form, in other words a conventional layout of melodic themes and key relationships.
Analysis recognizes formal types and then identifies how individual works satisfy or challenge
formal expectations throughout their musical journeys. Recurring principles of musical form
operate on all levels of musical structure from that of an entire piece (e.g., a sonata form or
a rondo form) down to the individual phrase. Closely allied with form is the temporal
dimension of music, broadly referred to as rhythm. In the music we are studying, the relative
lengths of formal units, especially at the level of the musical phrase, shape musical experience
in significant ways. Thus, our approach to this repertoire emphasizes elements of harmony,
form, and rhythm.
This text is directed at upper-level undergraduate and graduate students from all musical
disciplines who are required or elect to take a course in music analysis. Although we assume
that musicians in such courses will have completed several courses in music theory earlier
in their studies, we do provide a review of foundational principles in Part I.1 Even those
who are secure in their understanding of tonal harmony should spend some time with these
introductory chapters to acquaint themselves with our approach to this material. Chapter 1
reviews chord construction and labeling and then discusses harmonic progression within a
single key (diatonic harmony). Chapters 2 and 4 consider change of key (modulation) and
chords created through accidental alterations (chromatic harmony). Our approach to
harmony emphasizes that in music, just as with words in language, not all harmonies are
equally significant. Recognizing the multi-leveled, or hierarchical, nature of tonal harmony
is essential to understanding the larger arcs of directed motion that animate tonal music.
Satisfying performances of the repertoire we are studying invariably balance local detail with
x Preface

long, sweeping gesture; a hierarchical approach to harmonic analysis nurtures sensitivity to

the latter. In addition, we probe how melodic lines similarly embed multiple levels of
structure, revealing correlations between simpler underlying melodic frameworks and the
harmonic pillars supporting them. Chapter 3 discusses the basic element of musical form,
the phrase, and we explain some concepts that might not be as frequently included in
introductory theory courses as are harmonic topics. In particular, we devote considerable
attention to techniques that expand phrases beyond their expected lengths and also to the
notion that, like beats within a measure, not all musical measures are equally strong. This
latter phenomenon, now known as hypermeter, was recognized by eighteenth-century writers
as the alternation of “strong” and “weak” measures, and hypermetric analysis again develops
awareness of longer musical motions. Our hierarchical approach to tonal and rhythmic
structures is influenced by, and fully consistent with, the theories of Heinrich Schenker, but
this book is not a text in Schenkerian analysis. That is, many of Schenker’s ideas are present
without attempting to convey all that is required to construct graphic analyses that are
consistent with his concept of multi-leveled musical organization.
Part II is primarily devoted to detailed analyses of twelve compositions. These chapters
are organized by formal types, proceeding from binary and rounded binary forms (Chapter
5) to sonata forms (Chapters 6 and 7), ternary form (Chapter 8), and finally rondo form
(Chapter 9). For each of the twelve pieces that we discuss thoroughly, we provide a complete
score along with some annotations. These annotations identify formal sections, offer an
interpretation of hypermeter, and label harmonic progressions. It is strongly recommended
that the student listen to each piece, preferably more than once, before reading the analytic
discussion. Listening to each piece while following the annotated score will greatly facilitate
reading the analytic discussion. Due to their greater length, the pieces in Chapters 6–9 are
also accompanied by outlines of formal-tonal organization, and these figures should be
scanned before reading our analytic commentary. Typically, our discussion begins with a
survey of the formal organization, usually with some mention of the piece’s main key areas.
Then we enter the composition more deeply, focusing principally on the hierarchical nature
of tonal and rhythmic structures and the manipulation of pitch motives. Often we visually
depict our analytic understanding through voice-leading simplifications or representations,
as explained in the Appendix. For instructors who find there are too many pieces included
in Part II or that our discussions are too challenging, two suggested remedies would be to
skip Chapter 7 or to omit the last piece analyzed in each of Chapters 7–9.
In choosing the twelve pieces for detailed analysis, we have been guided by two primary
considerations. First, we have selected substantial works that are studied by university-level
students in their applied lessons and chamber music coachings and that in the broader musical
community are frequently performed and recorded. As a result, we cannot pursue every phrase
of these pieces in as much detail as we might like, but we fervently believe that analysis courses
should tackle pieces that are as long and as intricate as the ones students are preparing for
juries or recitals. Second, we have included pieces for varying instrumentation, including two
duos (one for clarinet and piano, the other for violin and piano), a piano trio, and an orchestral
movement, to the extent possible given the limitations of space. The suggested assignments
at the end of each chapter similarly involve a significant number of chamber works.
Due to the detailed nature of analysis and our choice to include complete scores, we
have had to restrict ourselves to a particular subset of the musical works composed in the
xi Preface

eighteenth and nineteenth centuries. Although Bach’s music appears quite often in Chapters
1, 2, and 4 as we review tonal harmony, we do not examine any of his works in detail in
Part II. Such a study would require explication of types of pieces that were not much pursued
by composers of the late eighteenth and nineteenth centuries. Put differently, the formal
conventions applicable to Baroque music differ quite markedly from those applicable to
Classical and Romantic music. At the later end of our chronological spectrum, we have
focused on those composers most directly within the Classical tradition of Haydn, Mozart,
and Beethoven; composers such as Berlioz, Liszt, Wagner, Tchaikovsky, and Strauss are not
studied. These and other composers of the later nineteenth century began to extend both
formal and tonal conventions in fundamentally new directions, and any serious engagement
with their music would stretch our book to an impractical length. We have opted to provide
numerous illustrations of analysis within a repertoire that is relatively uniform in organizing
principles rather than deliver a more diffuse sampling of analyses of widely differing
compositions. Even within our core group of composers, we decided eventually not to
include a detailed study of any of their vocal music. This decision was a practical one of
space, but also it has been our experience that students who can find their way through a
ternary-form piano piece by Schubert or Brahms, for example, can easily navigate one of
their songs; the reverse is less often the case. In part, this results from the generally shorter
lengths of vocal works, but the presence of a text also makes the identification of phrases
and sections more straightforward and the text’s meaning provides an accessible avenue to
beginning analysis of a vocal work.2
We have also had to restrict the topics we address in our analytic commentary due
to our principal emphases on tonal, formal, rhythmic, and motivic organization. Although
we are aware of performance implications of our analytic observations, we only address them
occasionally. This is due, in part, to practical limitations of space, but it is also due to the
magical way that differing performance interpretations can realize a particular analytic
observation. We believe that in-class demonstrations by instructors or student performers
as well as comparative assessment of recordings will generate sounds and class discussions
that are much more evocative than words we might commit to paper. Closely related to
performance concerns is musical expressivity. Throughout the analyses, we suggest how
particular musical features have expressive effects, but we do not do so consistently, and we
leave the happy task of exploring this topic further to instructors and students. And it is
important to bear in mind that ours is a text on music analysis, not music history. While
we give a general sense of how tonal harmony and musical forms evolved during the
nineteenth century, we do not do so in any comprehensive way. And we certainly make
no attempt to address how familiarity with composers’ biographies, their social milieus, and
the intended functions of their musical works enrich understanding. The ability to analyze
music is among the tools musicians need, and it is an important one. It is the topic of the
present text, but it is by no means the only intellectual enterprise that can enhance the lifelong
development of the musically sensitive mind.
We want to thank Constance Ditzel, Senior Editor (Music), and Denny Tek, Senior
Editorial Assistant (Music), for their enthusiastic support of this book from its initial stages
through its publication. We also acknowledge Mhairi Bennett, production manager, for her
expert transformation of our manuscript into its present form.
PART I

Principles of Harmony and

Phrase Design
1 A Review of Diatonic Harmony

Designation of Pitch and Chords

A preliminary step in our review of diatonic harmony must be to establish how we will
designate pitches and chords throughout this text. When we want to specify a particular
pitch in a specific octave, we will follow the system adopted by the Acoustical Society of
America, where middle C is C4, as indicated in Example 1.1.
EXAMPLE 1.1 Pitch Designation

Triads are designated by Roman numerals specifying the degree of the scale on which
they are built, and in this text we will designate the quality of each triad by upper- and
lower-case numerals, upper-case for major and lower-case for minor. Diminished triads are
indicated by a lower-case numeral followed by a superscript zero, e.g., viio. Inversions of
triads are indicated by the addition of figured-bass symbols (Arabic numbers and/or
accidentals) to the right of the Roman numeral, 6 for first inversion and 64 for second
inversion.1 Accidentals are needed where the interval above the bass is altered in relation to
the key signature. An accidental alone indicates that the third above the bass is altered;
otherwise, the accidental precedes the Arabic numeral, indicating the interval above the bass
that is affected, e.g., 6.
Example 1.2 shows the triads and their inversions in both the major and minor
modes. Note, however, that we have listed only those 64 chords representing I, IV and V,
since these are the only ones commonly employed.2 In the minor mode we have included
the major triad built on the natural seventh degree of the scale as well as the diminished
triad on the leading tone. Recall that in the minor mode the leading tone requires an
accidental ( or ).
4 Principles of Harmony and Phrase Design

EXAMPLE 1.2 Triads and Inversions in Major and in Minor

5 A Review of Diatonic Harmony

Triads within a key are commonly referred to by the following names beginning with
I/i and ascending to viio: tonic, supertonic, mediant, subdominant, dominant, submediant,
and leading-tone triad. These names are easy to remember once you realize the origin of
the terminology, which implies a symmetry around the tonic. The dominant is the fifth
above the tonic, and the subdominant is a fifth below. The mediant is midway between
tonic and dominant, and the submediant is midway between tonic and subdominant. This
leaves the supertonic, meaning the triad above the tonic (super is the Latin word for “above”).3
During the eighteenth century, the gradual acceptance of the passing seventh (8–7) as
a legitimate chord tone gave rise to seventh chords built on each degree of the scale. In the
major mode, there are four different types of seventh chord: 1) the major triad with major
seventh on scale degrees 1 and 4; 2) the major triad with minor seventh on scale degree 5;
3) the minor triad with minor seventh on scale degrees 2, 3, and 6; and 4) the diminished
triad with minor seventh (known as the half-diminished seventh chord) built on the leading
tone. In the minor mode, the major triad with major seventh occurs on scale degrees 3 and
6; the major triad with minor seventh on scale degree 5 (with the raised leading tone), but
also on natural scale degree 7 (VII7= V7 of III); the half-diminished seventh chord on scale
degree 2; and the fully diminished seventh chord (diminished triad and diminished seventh)
on the leading tone. These seventh chords and their designations in both the major and
minor modes are listed in Example 1.3. Note the shorthand designation of the diminished
seventh chord, o7, and half-diminished seventh chord, ø7. The diminished seventh chord
often functions in place of the dominant seventh chord, and occasionally the two occur
together forming the dominant ninth chord, as shown at the end of Example 1.3.

EXAMPLE 1.3 Seventh Chords in Major and in Minor

6 Principles of Harmony and Phrase Design

We will not write out all the inversions of all the seventh chords in major and minor.
Instead we have shown the inversions of the most common ones, the dominant seventh
chord and the diminished seventh chord, in Example 1.4.

EXAMPLE 1.4 Inversions of the Dominant Seventh and Diminished Seventh Chords

The chordal seventh is a dissonance requiring resolution down by step to a note of the
following chord. When the resolution does not occur as expected, this is significant. In
Example 1.5 we have shown the normal resolution at (a), but at (b) the resolution is
transferred to the bass, allowing the top voice to continue its upward motion.4

EXAMPLE 1.5 Resolution of the Dissonant Seventh

7 A Review of Diatonic Harmony

Tonal Functions
A logical way to think of harmonic progression is in terms of three basic functions: tonic,
dominant, and preparation for the dominant. As we shall see, a considerable portion of the
tonal literature can be understood in terms of these three basic functions. The first two,
tonic and dominant, are self-explanatory, while the last requires some clarification. When
we speak of preparing the dominant, we mean progressing to the dominant, either by a
strong fifth progression, that is, from a supertonic chord in root position, or more frequently
by step in the bass and most normally from below, thus supporting the subdominant or the
supertonic in first inversion, as shown in Example 1.6 at (a) and (b). Frequently there is a
5–6 linear motion above scale degree 4 in the bass, seemingly creating a harmonic change
from IV to ii6, but where this change is the result of linear rather than harmonic
considerations, we will indicate this change as shown at (c). In this instance, the 5–6 motion
avoids parallel perfect fifths between the bass and alto parts. Finally, we also find the root
of the dominant harmony approached from a step above, supporting either the submediant
or, more often, the subdominant harmony in first inversion, as shown at (d).

EXAMPLE 1.6 Harmonic Functions and Cadences

Cadences articulate important points of arrival or division in music. For our present
discussion, we need to distinguish two types: the half cadence (HC), which ends on the
dominant, thus coming to partial rest, and the authentic cadence, which ends on the tonic.
Depending on the final position of the top part, the authentic cadence is called perfect
(ending on scale degree 1) or imperfect (ending on scale degrees 3 or 5), abbreviated as
PAC and IAC respectively. In Example 1.6, the progressions at (a), (b), and (c) all culminate
with perfect authentic cadences, while (d) ends with a half cadence. Not represented here
is the cadential 64, a common component in music of the eighteenth and early nineteenth
centuries. The cadential 64 is the result of linear motion—of the passing sixth and fourth on
their way to the fifth and third over scale degree 5 in the bass—and thus the resulting chord
is not considered as a separate harmony from its resolution. It, along with its resolution, is
labeled as V to show its dominant function. This illustrates a very important point, namely,
that the labeling of chords should not result from the mechanical application of chord
inversion. The collection of notes of the cadential 64 is the same as the tonic triad, but its
function is not tonic, but rather an accented passing motion delaying arrival at the dominant
8 Principles of Harmony and Phrase Design

triad. If the 64 chords in Example 1.7 were to be removed, a melodic gap would be created
in each case. At (a) the sixth and fourth above the bass, A4 and F4, resolve to the fifth and
third leading to a perfect authentic cadence in F major. The second example, in D minor,
concludes with a deceptive progression, where VI replaces the expected tonic. The
progression at (c) shows a common variant of the normal voice leading; here the fourth still
resolves to the third above the bass, but the sixth progresses to the seventh rather than to
the fifth. In other words, in (c) the cadential 64 resolves to V7 rather than to V.

EXAMPLE 1.7 The Cadential 64

Though brief, the preceding review is sufficient for us to examine three short excerpts
from Mozart’s piano sonatas. All three examples have the same formal organization. Each
consists of two phrases, the first—the antecedent phrase—ends on the dominant (half
cadence) and the second—the consequent phrase—concludes with a perfect authentic
cadence. The two phrases are said to have parallel construction (as opposed to contrasting),
meaning that they open with the same melodic and harmonic material.5 Together the two
phrases form a musical period.
The first Mozart excerpt is the opening eight measures of the last movement from his
Piano Sonata, K. 545 (Example 1.8). The opening harmony is incomplete, but in retrospect,
considering where the phrase leads, the only logical choice is tonic. Furthermore, these
eight measures start the final movement of a sonata in C major. The half cadence is prepared
by IV5–6, where the change to ii6 is emphasized by the chromatic passing tone C and
the subsequent change of bass note. Passing tones, which connect notes belonging to the
underlying harmony, have been placed in parentheses and labeled “P.” In the second and
third measures, these passing tones occur “offbeat,” here on the second and fourth sixteenths,
but the C in m. 4 falls on the beat, and is thus called an accented passing tone. The
situation in the consequent phrase is somewhat different. Here the A5 on the second beat
of m. 6 and the C5 on the following downbeat are not approached by step, a characteristic
of the passing tone. Both are incomplete neighbors (IN), the first diatonic, the second
chromatic.6
9 A Review of Diatonic Harmony

EXAMPLE 1.8 Mozart, Piano Sonata in C Major, K. 545, III, mm. 1–8

Our second example is the opening period from the third movement of the Sonata,
K. 333 (Example 1.9). As in the preceding example, the antecedent phrase leads to a half
cadence, in this case involving the cadential 64, and the consequent phrase completes the
motion to the tonic. The G4 (inner part, left hand) on the fourth quarter of m. 1 (and later
in m. 5) arises conceptually from an implied F4 and is thus interpreted below as resulting
from a linear motion (5–6) above the bass note B3. Although we have placed the Roman
numeral vi in parentheses below the 6 here, we will not do so later in the text, since the
resulting chord is the product of voice leading, not harmony. In m. 3, the melodic notes
corresponding to the harmony are displaced by incomplete neighbors (appoggiaturas). In
m. 4, D5 is embellished by its upper neighbor before resolving to C5, and the last three
eighth notes are passing, leading up to F5. Regarding the harmonic analysis, we have labeled
all chords, though the result can be misleading, since this implies that all chords in the
succession are equal. They are not. We will return to this example later in the chapter to
uncover levels of harmonic organization.
10 Principles of Harmony and Phrase Design

EXAMPLE 1.9 Mozart, Piano Sonata in B Major, K. 333, III, mm. 1–8

Our third example is the opening period from the third movement of the Sonata,
K. 280. An interesting feature of this theme is the exchange of the outer parts of the chords
V43 – I (mm. 2–3 and later 10–11) resulting in the progression V42 – I6 (mm. 4–5 and 12–13).
The function of the two dominant seventh chords in this progression is to extend tonic
harmony through m. 5, which then leads to a half cadence in m. 8. The dominant is prepared
by IV5–6, and the actual change to ii is then clarified by the change of bass in
m. 7. Once again we have labeled all chords, but here we have shown the progression
operating at two levels. The five chords in mm. 1–5 (and later mm. 9–13) prolong the tonic
harmony, which is then followed by the cadential progression IV – ii – V and in the conse-
6–5
quent phrase IV – V4–3 – I. Note how this level of organization is articulated by the dynamics:
the forte dynamic clarifies the connection between the initial I and the I6 in the fifth measure
of each phrase, and the forte then continues into the cadential progression.
In examining these three excerpts from Mozart’s piano sonatas, we have touched on
two new topics, phrase structure and non-chord tones. The first of these will be developed
in detail in Chapter 3. For now it is sufficient that you understand what is implied by the
terms “phrase” and “period.” A phrase is a unit of tonal progression separated from its
continuation by a cadence. A period consists of two or more phrases that end with an
authentic cadence. Here the periods all consist of two parallel phrases, an antecedent leading
to a half cadence and a consequent ending with a perfect authentic cadence.
11 A Review of Diatonic Harmony

EXAMPLE 1.10 Mozart, Piano Sonata in F Major, K. 280, III, mm. 1–16

The second topic concerns non-chord tones: passing tones (P), accented and unaccented,
chromatic and diatonic; neighbor notes (N); and incomplete neighbor notes (IN), also called
appoggiaturas. All these are notes of melodic embellishment, and it is useful to label them
in harmonic analyses until distinguishing between chord and non-chord tones becomes
automatic. A related phenomenon we have not yet touched on is the suspension, which
results from rhythmic displacement rather than melodic embellishment. A suspension arises
from the extension of a note belonging to one chord into the time-span occupied by a
12 Principles of Harmony and Phrase Design

EXAMPLE 1.11 Bach, “O Lamm Gottes, unschuldig” (Chorale No. 165)

13 A Review of Diatonic Harmony

subsequent chord, where it becomes a dissonance requiring resolution to the note it has
temporarily displaced. So the process involves two chords: 1) the chord preceding the
suspension, which contains the chord tone to be suspended; and 2) the chord carrying the
suspension and its resolution.
An excellent source for studying the use of melodic embellishment and suspensions, as
well as the imaginative use of harmony to articulate a text, are the chorales of J.S. Bach.
His setting of “O Lamm Gottes, unschuldig” (“O Innocent Lamb of God”) is reproduced as
Example 1.11. Note that all suspensions, which are marked on the score, resolve down
by step to the note they have displaced. The most common types of suspension are those
represented here: 9–8, 7–6, and 4–3; these numbers signify the intervals above the bass
(i.e., figured-bass numerals). A suspension may be sounded again, as with the first 4–3
suspension, or held over from the previous chord, as with the first 9–8 suspension, both
occurring in m. 1. But beware, not all ties signify a suspension, as, for example, in the bass
part directly after the repeat; here, the tied F3 is a chord tone and the following E3 is a
passing tone. Finally, the 4–3 suspension in m. 4 has an embellished resolution; that is, the
resolution of F4 is the E4 on the second quarter note, not the submetrical E4 on the second
half of the first quarter.

Common Sequential Patterns

In this section we will examine sequential patterns commonly encountered in the music of
the eighteenth and nineteenth centuries. A sequence consists of a repeated voice-leading
pattern that connects two important points. From a harmonic perspective, some sequential
patterns are circular, that is, they return to the point of origin, while others are progressive,
connecting one harmony to another. Most important always are the boundaries, the point
of departure and the point of arrival.
The first type of sequence we will examine, the descending fifth sequence, is very
common. It utilizes chords built on all degrees of the scale, and, when complete, returns
harmonically to the point of origin, the tonic. In major keys: I – IV – viio – iii – vi – ii –
V – I. See Example 1.12a, mm. 18–26 from the first movement of Mozart’s Piano Sonata,
K. 545. This excerpt is in the key of the dominant, G major. In this example, the sequential
pattern alternates between first-inversion and root-position triads: I6 – IV – viio6 – iii – vi6
– ii – V6 – I.7 The sequence extends the initiating tonic harmony and is followed by a
confirming perfect authentic cadence. A simplification of the voice leading of this passage
is provided in Example 1.12b. Observe that the note values employed in the first four
measures of this analytic representation do not correspond to Mozart’s written durations.
Why have we done this? We perceive a stepwise descending line in the top part resulting
from notes struck on successive downbeats, and likewise one in the inner part resulting
from those struck on the second half of each measure. The note values, then, represent
the durations between successive notes in these lines.8 As indicated above the top part in
Example 1.12b, the sequence supports a descending fifth D6–G5, which is then followed
by the cadential pattern ii6 – V4–36–7
– I. Careful examination of these measures reveals that
they support a more encompassing descent of the fifth D6–G5 occurring over the span of
the entire phrase. The notes of the descent at this level are circled in Example 1.12b. This
example represents a major step in the analytic process, one we do not expect you to
14 Principles of Harmony and Phrase Design

EXAMPLE 1.12a Mozart, Piano Sonata in C Major, K. 545, I, mm. 18–26

15 A Review of Diatonic Harmony

EXAMPLE 1.12b Voice-leading simplification of Example 1.12a

reproduce at this early stage, but one you can follow. We have gone beyond the stage of
describing the sequence only in harmonic terms to show its function within the passage as
a whole. That is one of the goals of music analysis: to reveal the function(s) of the parts
within a larger context.
Our second example is a transitional passage from the first movement of Mozart’s Piano
Sonata, K. 332 (Example 1.13a). In this excerpt, Mozart changes from the major to the
minor mode before initiating a descending fifth sequence, this time utilizing a series of seventh
chords in root position.9 The result is a pattern of alternating 10–7 intervals above the bass,
as shown in the simplification of the voice leading of these measures (Example 1.13b). In
this instance the goal of the progression is the dominant, and the function of the extension
of the dominant is to allow the top part to complete its descent to G4. An important rhythmic
feature of this passage is the hemiola pattern indicated by brackets above the score (Example
1.13a, mm. 200–201).10
Our next example is the opening section of the Trio from Haydn’s String Quartet, op.
20, no. 2 (Example 1.14). Here the descending fifth progression involves alternating root-
position triads and seventh chords in first inversion. The sequence connects i (m. 57) to i6
(m. 64), followed by a passage in octaves leading to V.
The second type of pattern we will examine is referred to in the literature both as a
descending 5–6 and a descending third sequence. The first refers to the pattern of
intervals above the bass, and the second to the interval between the roots of alternating triads.
An example is the opening phrase from Beethoven’s Piano Sonata, op. 109 (Example 1.15).
The intervallic pattern 5–6 is shown directly below the music and the harmonic analysis
below that. The dotted brackets mark the repeated pattern. As in the previous example, the
sequence connects I with I6.
16 Principles of Harmony and Phrase Design

EXAMPLE 1.13a Mozart, Piano Sonata in F Major, K. 332, I, mm. 192–203

EXAMPLE 1.13b Voice-leading simplification of Example 1.13a

17 A Review of Diatonic Harmony

EXAMPLE 1.14 Haydn, String Quartet, op. 20, no. 2, III, mm. 57–68
18 Principles of Harmony and Phrase Design

EXAMPLE 1.15 Beethoven, Piano Sonata in E Major, op. 109, I, mm. 1–4

Our third type of pattern is the ascending 5–6 sequence, a voice-leading corrective
to avoid parallel fifths. We will offer a single example, the opening of Leporello’s famous
aria “Madamina” from Mozart’s Don Giovanni (Example 1.16). Here the 5–6 pattern of mm.
3–4 is repeated twice in the next four measures at successive higher steps connecting the
initial tonic harmony to its first inversion (m. 8), coinciding with Leporello’s arrival at D4.
This is shown on the score below the system in Example 1.16. Labeling the intervening
chords makes little sense; what is important are the boundaries of the sequence, I . . . I6.
This is followed by a cadential progression that is repeated in root position leading to a
perfect authentic cadence and strong arrival at scale degree 1 in the vocal part.

Passing and Neighboring Chords

In this text we will make a distinction between passing chords, those formed by passing
notes in either or both of the outer parts, and neighboring chords, those formed by
neighboring notes in one or both of these parts. While it is important for us to understand
chord-to-chord syntax, there is a danger inherent in providing a label for each chord, since
this implies that they all have equal status. Yet this is not true, and we have already seen
proof of this in our examination of at least two of our examples so far: the opening period
from the third movement of Mozart’s Piano Sonata, K. 280 (Example 1.10); and mm. 18–26
from the first movement of Mozart’s Piano Sonata, K. 545 (Example 1.12). The succession
of chords in the consequent phrase (mm. 9–16) from the first of these reads as follows:

I – V43 – I – V42 – I6 – IV – V6–5

4–3 – I

Examination of the opening measures of this phrase reveals that the chord in m. 10
harmonizes the passing note B4 between the previous C5 and the following A4, and that
the chord in m. 12 harmonizes the passing note G4 between the A4 and following F4. Thus
the melodic motion from the initial C5 to the F4 on the downbeat of the fifth measure of
19 A Review of Diatonic Harmony

EXAMPLE 1.16 Mozart, “Madamina” from Don Giovanni, mm. 1–16

20 Principles of Harmony and Phrase Design

the phrase outlines the tonic triad C5–A4–F4, an enlargement of the phrase’s opening
gesture. The intervening chords are passing. Thus a more accurate picture of the harmonic
organization of this phrase is:

I – V43 – I – V42 – I6
________________
I IV – V6–5
4–3 – I

In this text we will also make a distinction between chord and harmony. The famous
Austrian musician Heinrich Schenker (1868–1935) used the German word Stufe to designate
the latter, by which he meant harmonic scale-step, a harmony that may control several
chords in succession.11 In this example the first five chords are controlled by one harmony,
the tonic, and the harmonic succession (not the chord succession) is I – IV – V – I. Simi-
larly we have seen that the descending fifth sequence in the second excerpt cited above
(Example 1.12) extends a single harmony, the tonic, which is then followed by a standard
cadential progression leading to a perfect authentic cadence. We do not mean to dismiss
the sequential progression as insignificant, but rather to suggest that there are different
levels of harmonic organization in operation. Recognizing these levels is part of the analytic
process.
The notion of passing chords is clearly illustrated in the opening measures from the second
movement of Beethoven’s Piano Sonata, op. 10, no. 1 (Example 1.17). The first sixteen
measures of this movement form a musical period, divided into an antecedent phrase leading
to the dominant and a consequent phrase concluding with a perfect authentic cadence. The
first phrase is divided into two subphrases articulated by the change in melodic material in
m. 5. Careful examination of mm. 1–5 reveals that mm. 3–4 are a response to the first two
measures; that is, melodically they repeat the gesture of mm. 1–2 a step higher.12 Looking
at the melodic detail of m. 1, we see that the C5 on the second quarter is an incomplete
neighbor (appoggiatura) displacing B4, which is a neighbor to the A4 in mm. 1 and 2.
Already we have an interesting conflict between surface design and underlying structure.
The descending third C5–B4–A4 in mm. 1–2, answered by the third a step higher
in mm. 3–4, is aurally prominent. Yet this clearly articulated motivic gesture masks the
underlying neighbor-note pattern A4–B4–A4 harmonized by the progression I – V65 – I.
It is clear that the second of these chords is neighboring to the outer two. Likewise, the
underlying melodic structure of mm. 3–4 is B4–C5–B4 harmonized by V65 – I – V. In this
instance the tonic chord in the middle is neighboring, providing support to the neighbor
note C5. The melodic line then progresses via the chromatic passing tone B4 to C5 on
the downbeat of m. 5. The harmonic analysis below suggests three levels of organization.
In the first two measures the controlling harmony is the tonic supporting A4, and in the
next two measures the controlling harmony is the dominant supporting B4, which moves
up to C5 supported by tonic harmony in m. 5. The harmonic progression at level 2 is
then I – V – I. At the deepest level of organization (level 3), we can see (and hear!) that
the middle V chord is passing (it supports the passing tone B4 in the ascending third
A4–B4–C5). The tonic harmony spanning mm. 1–5 progresses to the subdominant and
eventually to the dominant in m. 8. But what about the tonic harmony at the end of m. 7?
We have placed the chords V65 – I that occur here in parentheses to show that they are
passing between IV and V. Certainly there is not a return to stable harmony here. It is
21 A Review of Diatonic Harmony

EXAMPLE 1.17 Beethoven, Piano Sonata in C Minor, op.10, no. 1, II, mm. 1–16
22 Principles of Harmony and Phrase Design

impossible to play this part of the phrase in such a way to make this chord the goal of motion;
rather the motion passes through this point to V.
Let us return briefly to the opening phrase from the third movement of Mozart’s
Piano Sonata, K. 333. When we examined this phrase initially, we labeled all the chords
(see Example 1.9). Is there something in the music to suggest different levels of harmonic
organization? First we might note that the melodic gesture of m. 2 is an answer to the previous
measure, though rhythmically altered and with G5 displacing the C5 that would be there
if Mozart had repeated the descending pattern of the opening arpeggio. We hear the second
measure as passing, leading to tonic harmony on the downbeat of m. 3. This suggests that
all the chords up to the downbeat of m. 3 fall under the tonic scale-step, to use Schenker’s
term. If we now consider the melodic motion F5 (m. 1) to E5 (m. 2), this predicts D5 on
the downbeat of m. 3, but this does not follow. Instead, the melodic line begins from B4
(actually from the incomplete neighbor temporarily displacing B4), the resolution of the
A4 (leading tone) from the previous measure, and the expected D5 is not reached until the
third quarter over tonic harmony. There is a specific relationship between the first and third
quarters of m.3, known as a voice exchange, where the notes of the two voices exchange
places, here B4 over D4 (first quarter) to D5 over B3 (third quarter). This is shown in
Example 1.18, a simplification of the motivic content and the voice leading of the phrase,
by the crossed lines.13 The harmonic analysis below indicates that all chords through the
third quarter of m. 3 fall under the tonic scale-step or harmony, which is followed by ii6
preparing the dominant at the half cadence. There is a subtle suggestion regarding
performance in this interpretation, namely that there could be the slightest separation between
the third and fourth beats of this measure, despite the continuous surface pattern. Mozart
has articulated this change in harmonic level with the addition of a third voice, starting with
the ii6 harmony.
There is a pattern emerging from our examination of the last two examples, namely
that it is the opening tonic harmony of the phrase that is extended by a progression of chords
up to the cadential progression: I . . . ii6 – V – I. Though any harmony can be prolonged
in this way, we will see that this is by far the most common pattern.

EXAMPLE 1.18 Voice-leading simplification of Example 1.9

23 A Review of Diatonic Harmony

Functions of Individual Chords

The following is a summary of the most common usages or functions of the diatonic chords
in both modes.

Tonic
1. The normal point of departure and the eventual goal of tonal motion.
2. The tonic chord, like all chords, can function in a passing or neighboring capacity. See
Example 1.17, m. 3 (neighboring) and m. 7 (passing).

Supertonic
1. Most frequently employed as preparation for the dominant. See Example 1.8, mm. 1,
5, and 7; Example 1.9, mm. 2, 3, 6, and 7; Example 1.10, m. 7; Example 1.12a, mm.
22–23; and Example 1.16, mm. 10 and 14.

Mediant
1. In the major mode, the mediant triad is not an independent chord. It does occur:
a. in a descending fifth sequence (see Example 1.12a, m. 19); and
b. as a substitute for I6 (see Example 1.19)
2. In the minor mode, III is a frequent goal of motion.

EXAMPLE 1.19 Substitution of iii for I6

24 Principles of Harmony and Phrase Design

Subdominant
1. Often used as preparation for the dominant. See Example 1.10, mm. 6 and 14; and
Example 1.17, mm. 6–7 and 14.
2. Provides support to the neighbor note in motions involving scale degrees 5–6–5 or
3–4–3. See Example 1.20: Haydn, Chorale St. Antoni, mm. 1–5.
3. The subdominant chord can also occur as a passing chord, as shown in Example 1.21,
where it offers temporary consonant support to the passing seventh of the dominant.

EXAMPLE 1.20 Haydn, Chorale St. Antoni, mm. 1–5

EXAMPLE 1.21 IV6 as Passing Chord

25 A Review of Diatonic Harmony

Dominant
1. Frequent goal of motion, e.g., half cadence, or proceeding to the tonic.
2. As a neighboring or passing chord. See Example 1.17, m. 1 (neighboring) and m. 7
(passing).

Submediant
1. The submediant is sometimes employed as a third divider in the descending fifth
progression from I to IV or ii6: I – vi – IV/ii6. See Example 1.8, m. 1.
2. It is also used as a substitute for the tonic after dominant harmony (the “deceptive” pro-
gression). See Example 1.22: Chopin, Prelude in B Minor, mm. 15–18. The harmonic
analysis of this excerpt is difficult. We have been guided by the fact that m. 16 is an
embellished repetition of m. 15. The B4 on the downbeat of m. 16 is thus a suspension
displacing the leading tone, and the G4 in the alto part displaces the F4, anticipating
the submediant chord in the next measure. Note the passage’s beautiful shadings of the
submediant with its welcome major-mode color: the submediant serves first an illusory
sonority at the end of m. 16, then briefly as a harmony in m. 17, and finally as the
phrase’s tonal goal.

EXAMPLE 1.22 Chopin, Prelude, op. 28, no. 6, mm. 15–18

Leading-Tone Triad
1. Most frequently found as a passing chord between I and I6: I – viio6 – I6.
2. Also occurs as part of a descending fifth sequence. See Example 1.12a, m. 19. VII in
the minor mode, the subtonic chord, leads to III as part of a sequence or as its dominant.
26 Principles of Harmony and Phrase Design

SUMMARY OF CONCEPTS AND TERMINOLOGY

We have covered a lot of territory in this chapter, some of it undoubtedly familiar, but other
ideas are most likely new. As a summary and an aid in checking your understanding of the
main concepts, we offer the following outline.

A. Review the following:

1. All chords (triads and seventh chords), their names, their inversions, and their designations
in major and minor keys.
2. Cadences
a. half
b. authentic, perfect and imperfect
c. deceptive
3. Notes of melodic/rhythmic embellishment
a. passing tone, accented and unaccented
b. neighbor note
c. incomplete neighbor (appoggiatura)
d. suspension
4. Form
a. phrase and period
b. antecedent and consequent

B. Make sure you understand these concepts:

1. Tonal functions: tonic, dominant, and dominant preparation
2. Passing and neighboring chords
3. Distinction between chord and harmonic scale-step
27 A Review of Diatonic Harmony

SUGGESTED ASSIGNMENTS

Sequences
In all of the following you should try your hand at simple reductions to show the underlying
voice leading. Also provide a harmonic analysis for each.

1. Bach, Partita No. 1, Allemande, mm. 24–27. (Note that the key signature does not
reflect the actual key of this excerpt.)

2. Bach, Keyboard Suite in A Minor, Giga, mm. 39–45

28 Principles of Harmony and Phrase Design

3. Schubert, Impromptu, op. 90, no. 2, mm. 25–32

4. Bach, French Suite No. 5, Gavotte, mm. 1–4

29 A Review of Diatonic Harmony

Periods
1. Mozart, Piano Sonata, K. 331, I, mm. 1–8. What is the chord on the downbeats of
mm. 3 and 7? Does it have harmonic meaning?

2. Mozart, Piano Sonata, K. 333, I, mm. 23–38. Note that each of the eight-measure phrases
is divided into subphrases of four measures. (Note that the key signature does not reflect
the actual key of this excerpt.)
30 Principles of Harmony and Phrase Design

3. Mozart, Piano Sonata, K. 284, II, mm. 1–16. Study levels of harmonic organization
and melodic structure. As a guide for your analysis, review our analysis of harmonic
levels in Beethoven’s Piano Sonata, op. 10, no. 1, II, mm. 1–16 (Example 1.17).
31 A Review of Diatonic Harmony
2 Expanding the Diatonic Palette

To contemporary Western ears, a diet of purely diatonic music would soon become boring.
It is the expansion of the diatonic palette to include chromatic elements and modulation
that provides interest and color to this basic sound world.

Secondary Dominants and Diminished Seventh Chords

The first step to expanding diatonic progression from a harmonic perspective is the inclusion
of secondary dominant and diminished seventh chords to the stable triads in the key other
than the tonic.1 In the major mode, V(7) of ii, of iii, of IV, of V, and of vi; and in the minor
mode, V(7) of III, of iv, of v, and of VI. The purpose of the secondary dominant is to place
emphasis on a chord within the diatonic progression. Consider the two progressions in
Example 2.1. The basic progression at (a) is I – ii – V. It is expanded diatonically by the
passing chord I6 on the fourth beat of the first measure, connecting ii and ii6. Note that
there is a voice exchange between the outer parts of these two chords shown by the crossed
lines. The progression is further elaborated harmonically by secondary dominant seventh
chords leading to ii and later to V. Rather than writing out V65 of ii followed by ii, we have
employed a shorthand system: [V65] ii. We will utilize this system where it is convenient
throughout this text. The basic progression at (b) is i – iv – V in A minor. This progression

is expanded initially by a passing viio 6 connecting i and i6 and later by a passing 64 connecting
iv and iv6. In each of these cases there is a voice exchange between the outer parts of the
two inversions of the same harmony. The progression is further expanded by the secondary
dominant seventh chord of iv.
In Chapter 1 we noted that the diminished seventh chord often substitutes for the
dominant, and this applies to secondary chords as well. The diminished seventh chord
occurs naturally in the minor mode built on the leading tone, but it occurs in both major
and minor as chromatic chords emphasizing stable triads in the key. At (a) and (b) in
Example 2.2, we have taken the progressions in Example 2.1 and rewritten them, substituting
diminished seventh chords for the secondary dominants. Note the shorthand notation: [o7].
We will notate the inversions of secondary diminished seventh chords in a similar fashion.
At (c) we have written out the dominant ninth chord in A minor, since this chord—formed
when the dominant seventh and diminished seventh chords occur together—also appears
as an applied chord. We will encounter the dominant ninth chord shortly in our examina-
tion of the opening section from the second movement of Beethoven’s Piano Sonata,
op. 31, no. 2.
33 Expanding the Diatonic Palette

EXAMPLE 2.1 Expanding Diatonic Progressions with Secondary Dominants

EXAMPLE 2.2 Expanding Diatonic Progressions with Secondary Diminished Sevenths

With the addition of secondary dominants and diminished seventh chords, our sound
world now contains all twelve pitches, though, of course, not twelve equal pitches. The
chromatic elements serve as elaborations of the notes of the diatonic scale, which have their
own hierarchy. This is demonstrated by the schemata in Example 2.3. At (a) the top line is
an ascending chromatic scale, missing only A, where the chromatic elements are harmonized
as secondary dominants. At (b) the chromatic scale is now in the lowest voice, and the
chromatic elements support secondary diminished seventh chords.
Our first example from music literature is from the first movement of Mozart’s Piano
Sonata, K. 333 (Example 2.4). This passage is in the key of the dominant, F major. The
tonic harmony is stated clearly in the first measure (m. 42), but then in the next two measures
we are faced with incomplete chords. Nevertheless, the progression is clear: IV – V – vi
with the motion to vi emphasized by its diminished seventh chord.2 The submediant then
leads to a perfect authentic cadence, but melodic closure is avoided at the last minute by
delaying the sounding of F5 in m. 46, thus creating forward momentum. The entrance of
34 Principles of Harmony and Phrase Design

EXAMPLE 2.3 Secondary Dominant and Diminished Sevenths in C Major

EXAMPLE 2.4 Mozart, Piano Sonata in B Major, K. 333, I, mm. 42–50

35 Expanding the Diatonic Palette

E5 in the second half of this measure suggests we might be headed once again to the
subdominant, but the chromatic change in the bass creates a secondary dominant leading
instead to ii. Measure 48 is an answer to m. 47 a step lower, completing the progression to
6–7
I, which is then followed by a secondary dominant leading to the dominant (V4–3 ) and
closure. This particular variant of this standard cadential pattern—where the sixth is delayed
in the top part until the chord has resolved to the true dominant, thus sounding like an
incomplete neighbor or appoggiatura—is a hallmark of Mozart’s style.
Our second example is a parallel period from the second movement of Beethoven’s
Piano Sonata, op. 14, no. 1 (Example 2.5). The third and fourth measures of the antecedent
phrase present an interesting problem in notating the harmonic progression. The chord in

EXAMPLE 2.5 Beethoven, Piano Sonata in E Major, op. 14, no. 1, II, mm. 33–51
36 Principles of Harmony and Phrase Design

the third measure is the diminished seventh of V: A–C–(E)–G. Instead of leading to V,

what follows is a substitute for V, the diminished seventh chord D–F–A–C. (The G
on the downbeat of the fourth measure is a 7–6 suspension.) So we have two diminished
seventh chords in a row.3 The dilemma is how to indicate most accurately what is occurring
in the music. The notation [o7] o43 makes little sense, so we decided to write out o7 of
V to o43 (substitutes for V), which is accurate and describes the functions of the two chords.
The equivalent spot in the consequent phrase is more straightforward. Here C is an
incomplete neighbor displacing B; the chord is o43 of iv leading to iv6. This is repeated twice,
followed by a voice exchange between the outer parts of the iv6 and iv chords, which
then leads to a perfect authentic cadence. The underlying progression of this consequent
phrase is i – VI – iv6 – V – i. In this instance it is not the initial harmony, the tonic, that is
prolonged by intervening chords, but the subdominant.
Our third example, the opening phrase from the second movement of Beethoven’s
Piano Sonata, op. 10, no. 3, also makes extensive use of diminished seventh chords (Example
2.6). The diminished seventh chord on the downbeat of m. 3 is extended by passing 64 chords
connecting o7 to o65 and back to o7 on the downbeat of m. 4. The expected resolution of
this chord is to the tonic, but instead it leads directly to V42 of iv. As a result, the leading
tone C does not resolve to D, but rather falls by semitone to C, the seventh of the secondary
dominant. The subdominant is then further emphasized by [V65] iv. Though not indicated
on the score, note the relationship between iv6 and iv: a voice exchange between the outer
voices (G4 over B2 to B4 over G2). As we have seen, this is a very common means of
prolonging a single harmony. The subdominant leads to a o43 chord. Once again this chord

EXAMPLE 2.6 Beethoven, Piano Sonata in D Major, op. 10, no. 3, II, mm. 1–9
37 Expanding the Diatonic Palette

EXAMPLE 2.7 Beethoven, Piano Sonata in D Minor, op. 31, no. 2, II, mm. 1–17

does not resolve to a tonic harmony (i6), but rather the bass progresses up this time by
semitone to G supporting o7 of V. It would seem, then, that the function of this o43 in m.
6 is not really harmonic, but linear, offering support for E5 as it reaches up to F5 in m. 7.
Our final example in this section, the opening period from the second movement of
Beethoven’s Piano Sonata, op. 31, no. 2, also makes extensive use of secondary dominant
and diminished seventh chords (Example 2.7). The first of these, o65 of ii6 occurs on the
third beat of m. 6. The following ii6 is delayed until the second beat of m. 7 by suspensions
in the top and “tenor” voices. This is followed by a passing 64 connecting ii6 to ii43, and the
introduction of the E transforms the latter into V43 of the following dominant at the half
38 Principles of Harmony and Phrase Design

cadence in m. 8, which is decorated by suspensions in the inner voices. In the consequent

phrase we have a very clear example of a dominant ninth chord, where the ninth, G, is
heard as the chromatic upper neighbor of F (note the motion G–F at the end of m. 12).
The dominant is subsequently prolonged by a motion to vi, where this G is now respelled
as F supporting the o7 of vi (mm. 14–15).

Chromaticized Sequences
Sequences frequently include chromatic elements, including secondary dominants. We will
begin by examining two examples of chromatic fifth sequences. The first (Example 2.8),
written in a Baroque imitative style, requires little explanation. It utilizes secondary dominant
seventh chords to elaborate a simple diatonic progression of descending fifths in C minor.
The second example, a transition passage from Chopin’s Mazurka, op. 68, no. 4, is far more
complex (Example 2.9a). Here the underlying progression is elaborated by chromatic
passing tones and suspensions, so we have enclosed the chords in boxes to make it easier to
follow the discussion. From the opening G major chord (V of V in F minor), the music
progresses by a series of secondary dominant seventh chords (D7 – G7 – C7 – F7 – B7 –
E7) until m. 37, where the pattern changes4 and the chords progress down by chromatic
steps in parallel motion leading to the cadential 64. A representation of the underlying voice
leading of this complex passage is provided in Example 2.9b.5 As shown below the system,
the intervening seventh chords connect V of V (m. 31) to V (m. 40), which is subsequently
extended until its resolution to the tonic at the conclusion of the following phrase.

EXAMPLE 2.8 Descending Fifth Sequence with Secondary Dominants

39 Expanding the Diatonic Palette

EXAMPLE 2.9a Chopin, Mazurka in F Minor, op. 68, no. 4, mm. 31– 40

EXAMPLE 2.9b Voice-leading representation of Example 2.9a

40 Principles of Harmony and Phrase Design

Next we will examine two examples of chromatic ascending 5–6 sequences. The
first of these is found in mm. 19–21 of the allemande from Bach’s Keyboard Partita No. 4
in D Major (Example 2.10a). This passage leads from the subdominant to the tonic in the
key of the dominant, A major. The 5–6 sequence and the chromatic notes creating
secondary dominant 65 chords are very clear in the left-hand part. However the right-hand
part is complex, and for this reason we have provided a simplification that removes all
suspensions and passing tones in the right-hand part (Example 2.10b). A two-level harmonic
analysis is provided below the simplified score. Though the chord-to-chord harmonic syntax
makes perfect sense, it is important to keep in mind that the purpose of a sequence is to
lead from the point of initiation (IV) to the goal (I), here through the dominant.

EXAMPLE 2.10a Bach, Partita No. 4 in D Major, Allemande, mm. 19–21

EXAMPLE 2.10b Voice-leading simplification of Example 2.10a

The second example of this type of sequence comes from the second movement of
Beethoven’s Piano Sonata, op. 7 (Example 2.11a). Again the 5–6 sequence is clear in the
left-hand part (mm. 17–18), whereas the right-hand part is a bit more complex. A detailed
harmonic analysis is provided below the score, showing the elaboration of an ascending
diatonic step progression by secondary dominants. Especially important in this regard is the
diminished seventh chord of vi () in m. 18, which supports F5, the upper neighbor of E5.
Looking ahead you will see that we have placed parentheses around mm. 20–23. This
41 Expanding the Diatonic Palette

parenthetical insertion, which delays arrival at the tonic harmony, expands the length of the
phrase from six to ten measures. This and other types of phrase expansion will be discussed
in detail in Chapter 3. A simplification of the voice leading is provided in Example 2.11b.
Here it is possible to see the sequence’s underlying voice leading and how it fits into the
phrase as a whole. The outer voices of the sequence proceed in ascending parallel tenths

EXAMPLE 2.11a Beethoven, Piano Sonata in E Major, op. 7, II, mm. 15–24

EXAMPLE 2.11b Voice-leading simplification of Example 2.11a

42 Principles of Harmony and Phrase Design

EXAMPLE 2.12a Beethoven, Piano Sonata in C Major, op. 53, I, mm. 1–9

EXAMPLE 2.12b Voice-leading representation of Example 2.12a

43 Expanding the Diatonic Palette

connecting E4 over C3 (I) to C5 over A3 (vi), while the inner line proceeds by 5–6 motions.6
Above this, E5 is introduced by its upper neighbor, and we have connected this to the point
of origin, E4. As shown by the arrow, the top line is then transferred down to the original
octave position, temporarily covered by A4–G4. Once again we have a situation where it
is possible to label all chords in the sequence, but most important are the point of initiation
(I) and the goal of the motion (vi). This level of harmonic organization is indicated below
Example 2.11b.
We turn now to the chromatic descending 5–6 sequence with an example from
the opening of Beethoven’s “Waldstein” Sonata, op. 53. The repeated 5–6 pattern is marked
beneath the score (Example 2.12a). A cursory examination of the bass line reveals that it is
a chromatic descending tetrachord leading from tonic to dominant (V7). A closer examination
reveals that these measures are organized in two groups of four, where the fourth measure
is a kind of response to the third in the higher octave; that is, mm. 5–8 are a slightly modified
restatement of mm. 1–4 a step lower. Harmonically the first group progresses from I to V6,
the second leads to IV6and then to V7. Thus the overall progression is I – V6 – IV6 – V7,
as shown at the lowest level below the representation of the voice leading in Example 2.12b.7
The issue here is how to represent the logic of the chord-to-chord progression. It might
occur to you to label the B chord in mm. 5–6 as VII, but that does not indicate how it
functions as leading to IV6 in m. 7. A better description is that it functions in relation to
the subdominant as IV of IV leading to V42 of IV. We represent this through expansion of
the square brackets: [IV – V42] IV6. Looking back at mm. 1–4, we can see that the opening
tonic also functions in retrospect as a subdominant in relation to V6.
A similar situation is found in mm. 15–21 from the opening movement of Schubert’s
String Quartet, D. 887 (Example 2.13). Once again the bass is a descending chromatic
tetrachord leading from tonic to dominant. And mm. 17–18 are an answer to mm. 15–16
a step lower, so V6 (m. 16) is answered by IV6 (m. 18). The third statement, beginning
from the E chord, is altered, leading through a passing 64 on the downbeat of m. 20 to IV,

EXAMPLE 2.13 Schubert, String Quartet in G Major, D. 887, I, mm. 15–21

44 Principles of Harmony and Phrase Design

EXAMPLE 2.14 Mozart, Piano Sonata in F Major, K. 280, I, mm. 14–26

thus extending the subdominant until it resolves to the dominant, decorated by a 4–3
suspension, in m. 21. In Beethoven’s op. 53 we were able to ascribe harmonic function to
each chord between the terminal points, but here the chords in mm. 17 and 19 do not have
harmonic function. They are solely the product of repetition of the 5–6 pattern at successively
lower steps.
This last example illustrates an important point, namely, that we cannot—nor should
we attempt to—ascribe harmonic function to chords that result purely from linear
considerations. In the preceding examples we have indicated chord-to-chord harmonic syntax
along with deeper levels of harmonic organization as long as doing so has provided further
insight into the logic of a passage. But there are clearly some instances where labeling every
chord not only can be meaningless, but can be misleading. This is illustrated by examining
the sequential passage in Example 2.14, the transition to the second theme from the first
movement of Mozart’s Piano Sonata, K. 280. The tonic harmony is established in mm. 14–17.
Beginning in m. 18 there is a repeated descending pattern of o43 chords leading to first-
45 Expanding the Diatonic Palette

inversion triads. If we were to label these triads, the result would be V6 – IV6 – III6 – II6
– I6, each preceded by its o43 chord. This is a meaningless string of symbols from the point
of view of harmonic logic. What is important here is that this repeated descending pattern
connects I (m. 17) to I6 (m. 22), indicated by the arrow in Example 2.14. This is a good
illustration of the distinction we have made before between chord and harmony (harmonic
scale-step). The harmony controlling this passage is the tonic. Providing the intervening
chords, those between I and I6, with Roman numerals would ascribe to them harmonic
status, where their real function is passing.
A more complex example of passing chords—in this case not a sequence—is provided
in Example 2.15. It is the opening phrase of the development section from the first
movement of Mozart’s Piano Sonata, K. 333. This phrase is in the key of the dominant, F
major. It opens with the tonic harmony and proceeds to the passing chord viio6 (or V43 if
you count the grace note C), which is repeated in the lower octave in the third measure.
The diatonic continuation would normally be to I6, but instead Mozart writes a secondary
diminished seventh chord directing the motion more strongly to the ii6 chord in the next
measure.8 So far, so good. It is the next chord that is potentially a problem. Its notes say it
is V7, but its function is not to lead to the tonic. Rather it is a passing chord connecting ii6
to the o65 of V, similar to the way viio6 had previously connected I to the o65 of ii6. The
“real” dominant comes in the seventh measure of the phrase. Labeling the chord in the
second half of the fifth measure as V7, at least without indicating its passing function, would
give a distorted picture of the harmonic logic of this phrase.

EXAMPLE 2.15 Mozart, Piano Sonata in B Major, K. 333, I, mm. 64–71

46 Principles of Harmony and Phrase Design

The purpose of these last two examples is to demonstrate that the interpretation of the
harmonic organization of a passage is not a mechanical, automatic process. You should always
consider the larger picture first, that is, to see and hear where a passage is leading before
deciding how to label the details. There is a very significant difference between chord and
harmony. All harmonies are chords, but not all chords are harmonies.

Diatonic Modulation
Modulation—change of tonal center—occurs in various ways and in varying strengths.
Regarding the latter, we will make a distinction in this text between a transient modulation,
where there is a temporary shift of tonal center, and a true modulation that is carefully
prepared and confirmed by a cadence. We will also refer to the new key in relation to the
tonic, for example, a modulation to the key of the dominant, G major. Such a statement
carries much more information than saying simply that the piece has modulated to G major.
We are concerned here primarily with the tonal style in which a piece or movement begins
and ends in the same key, though possibly with a change of mode, and it is important to
describe internal changes of key in relation to this tonic.
Though modulation can occur from any key to any other key, our concern in this chapter
is diatonic modulation, which is limited to those keys whose tonic triads are stable triads in
the original key. Mode—major or minor—depends on the quality of the triad in the original
key. Thus, from C major we modulate to D minor (ii), E minor (iii), F major (IV), G major
(V) and A minor (vi); and from A minor we modulate to C major (III), D minor (iv),
E minor (v), F major (VI) and G major (VII).
At a minimum, a tonicization or a passing modulation requires the dominant of the new
key and at least temporary continuation in that key. True change of key also requires
confirmation by a cadence in the new key. The smoothest modulations are often those where
there is a pivot, a chord that functions in both the original and the new key, most frequently
a chord preparing the new dominant. We identify pivot chords in retrospect, since we are
not aware that a modulation is about to take place until after this point, when the dominant
of the new key is sounded. A list of frequently encountered pivots is provided in Figure
2.1. We will provide examples of the most common modulations, and, where appropriate,
indicate the pivots.

Modulation from an Initial Major Key

I to V
1. Bach, Prelude in D Major (WTC I), mm. 1–6 (Example 2.16). Pivot: I5–6 becomes
ii6. This is an interesting passage to examine from the perspective of near, but not exact,
sequential repetition. For example, note that the right-hand pattern in the second half of
m. 1 is repeated with minor alteration in the first half of m. 2, and that the right-hand pattern
in m. 3 is repeated a third lower, again with minor alteration, in m. 5. In both instances,
the changes are the direct result of differences in the supporting harmonies. Note also that
the G5 on the downbeat of m. 4 is not resolved in that octave within this initial phrase,
a sure sign that Bach will return to that register shortly.
47 Expanding the Diatonic Palette

Major Mode Pivot

I to ii ii/i (read: ii becomes i) or IV/III

to iii vi/iv or I/VI

to IV ii/vi or IV/I

to V vi/ii or I/IV

to vi ii/iv or IV/VI or vi/i

Minor Mode

I to III iv/ii or VI/IV or i/vi

to iv iv/i or VI/III

to v i/iv or III/VI

to VI iv/vi or VI/I

to VII i/ii or III/IV

FIGURE 2.1 Common Pivots in Modulations

EXAMPLE 2.16 Bach, Prelude in D Major (WTC I), mm. 1–6

48 Principles of Harmony and Phrase Design

2. Beethoven, Piano Sonata, op. 31, no. 1, I, mm. 1–11 (Example 2.17). Pivot: vi
becomes ii. The opening phrase, with its staggered entrances between the hands and the
sudden dynamic outburst, sets the mood for this playful movement.

EXAMPLE 2.17 Beethoven, Piano Sonata in G Major, op. 31, no. 1, I, mm. 1–11

3. Haydn, String Quartet, op. 77, no. 2, Menuetto, mm. 1–12 (Example 2.18). As
in the preceding example, the pivot is vi/ii (actually vi6/ii6). However, vi6 is not an
independent harmony here, but arises from I5–6, in this instance elaborated by the intervening
chromatic passing tone C, which is emphasized by the marking .

I to IV
Schumann, Dichterliebe, no. 4, mm. 1–8 (Example 2.19). There are a couple of unusual
features of the harmonic syntax of these eight measures. First, the progression leads in a
normal way to the dominant in m. 4, but there, instead of returning to I, the accompaniment
insists on ii. This harmony, the supertonic, becomes the pivot in the next measure in the
modulation to IV: ii becomes vi. This is followed by a secondary dominant in 42 position
leading to IV6 in the new key. From there the progression continues through a passing
6 6 6
4 to ii5. The expected continuation is to the cadential 4, but instead Schumann interjects
6
I – IV (forte) before the perfect authentic cadence in the new key. We have placed
this I6 – IV in parentheses and below we have shown the normative progression. So why
might Schumann do this? It preserves the parallel motion between voice and bass, and at
the same time emphasizes the words “ganz und gar,” which mean totally or completely.
49 Expanding the Diatonic Palette

EXAMPLE 2.18 Haydn, String Quartet in F Major, op. 77, no. 2, III, mm. 1–12
50 Principles of Harmony and Phrase Design

EXAMPLE 2.19 Schumann, Dichterliebe, no. 4, mm. 1–8

A Summary Example: Bach, French Suite No. 5, Gavotte (Example 2.20)

The form of this engaging stately Gavotte is binary; it is divided into two parts, each of which
is repeated. The first part consists of a period divided into two parallel phrases, the second
of which modulates to the dominant. The second part comprises two periods, each divided
into two phrases. The first phrase ends on a half cadence in the key of the submediant, and
the second phrase completes the modulation to vi. The final period is organized similarly,
with the first phrase ending on the dominant and the final phrase completing the return to
the tonic. Each of the periods ends with a perfect authentic cadence.
51 Expanding the Diatonic Palette

EXAMPLE 2.20 Bach, French Suite No. 5 in G Major, Gavotte

52 Principles of Harmony and Phrase Design

A cursory examination of the Gavotte reveals the frequent recurrence of a three-note

motive, repetitions of which are marked on the score by brackets. The initial phrase opens
with two statements of this idea harmonized by the descending 5–6 sequence. The second
phrase opens with a modified statement of the motive leading to V42 in the new key of V.
We have indicated the previous tonic harmony as the pivot in the modulation to V: I/IV.
And we have circled the bass note G3 on the downbeats of mm. 5 and 6, the seventh of
the dominant in the new key. Our expectation is that this dissonance will resolve down by
step at the change of harmony, but instead the resolution is temporarily transferred to the
top part (F5) in mm. 6–7. Note, however, that our expectation is satisfied by the sounding
of the F in the bass on the second quarter note of m. 7 (I6 in the new key).
The second part opens with the inversion of the motive. Temporarily we have returned
to the tonic key (note the consistent use of C rather than C), and it is from this key that
Bach prepares the modulation to the submediant (E minor). The pivot chord is IV/VI

followed by iv5–6 V in the new key. The dominant is subsequently prolonged throughout
the next phrase until the final cadence in mm. 15–16 confirming the modulation. This
prolongation of the dominant consists of a modified statement of the opening right-hand
part now in the lower part, above which Bach has added counterpoint extending the eighth-
note motion of the preceding phrase, broken only at the cadence. The final period then
opens with another statement of the inverted motive in the right-hand part as the bass picks
up the eighth-note motion, which this time continues through the phrase division until the
final cadence.
Following our earlier practice, we have placed all non-chord tones in parentheses, but
because they are so numerous we have not labeled them. The one exception is the appear-
ance of a type not yet mentioned, the anticipation of the final G5 in the top part.

Modulation from an Initial Minor Key

i to III
1. Bach, Prelude in C Minor (WTC I), mm. 1–8 (Example 2.21). The first four
measures of this prelude establish the main key and the main motivic idea, which alternates
between the top part and an inner voice (left hand). The 64 chords on the last quarter note
of mm. 1 and 2 are neighboring, and the tonic chord on the last quarter of m. 3 is passing
between V65 and V. The continuation, beginning in m. 5, is a sequence by descending fifths
alternating between 42 and 65 chords. The pivot in the modulation to III (E major) is shown
as VI42/IV42 at the outset of the descending fifth sequence.9 Arrival at the new key is marked
by repetition of the main motive.

2. Mozart, Piano Sonata, K. 310, I, mm. 9–16 (Example 2.22). These measures are
the consequent phrase (modulating consequent) of a period. The goal of this phrase is the
dominant in the new key (III). The pivot in the modulation is shown as VI5–6/IV5–6, which
is followed by a secondary dominant leading to ii and then V in the new key. In this phrase
the tonic is initially embellished by o7 over a tonic pedal (mm. 9–11). An important motivic
53 Expanding the Diatonic Palette

EXAMPLE 2.21 Bach, Prelude in C Minor (WTC I), mm. 1–8

EXAMPLE 2.22 Mozart, Piano Sonata in A Minor, K. 310, I, mm. 9–16

54 Principles of Harmony and Phrase Design

component of these measures—as well as the antecedent phrase—is the decoration of E5

by its upper and lower neighbors (F5 and D5). If we skip ahead to the measure marked
calando, we see that Mozart has cleverly built reference to this motivic idea into the latter
part of the phrase leading to the dominant of III. By delaying the entrance of F5, D5, and
E5, we hear the last of these sounded one final time in reference to an A minor chord, now
heard as vi in the new key rather than tonic.

3. Beethoven, Piano Sonata in C Minor, op. 13, I, mm. 1–5 (Example 2.23). This
is an interesting passage to examine with respect to passing chords. First, the 64 in m. 1 is a
passing chord between i and i6, then in m. 2 the root-position dominant is passing between
o4 o4
2 and 3; and finally this same progression occurs twice in relation to V. In each of these
cases, there is a voice exchange between the outer voices, shown in Example 2.23 by the
crossed lines. The pivot in the modulation to III (E) is iv6/ii6, in this case emphasized by
its secondary dominant. The extensive use of diminished seventh chords in this introductory
passage contributes to the “Pathétique” quality of this movement.10

EXAMPLE 2.23 Beethoven, Piano Sonata in C Minor, op. 13, I, mm. 1–5
55 Expanding the Diatonic Palette

i to v
1. Beethoven, Piano Sonata in D Major, op. 10, no. 3, I, mm. 23–30 (Example
2.24). The pivot in the modulation to v (F minor) is i6/iv6.

EXAMPLE 2.24 Beethoven, Piano Sonata in D Major, op. 10, no. 3, I, mm. 23–30
56 Principles of Harmony and Phrase Design

2. Bach, Prelude in E Minor (WTC I), mm. 1–16 (Example 2.25). The first four
measures establish the main key by the progression I – iv64 – viio – i (an alternate to the
more common i – iiø42 – V65 – i). This is followed by the progression i – VI – iv – V – i,
where VI and iv are introduced by their secondary dominants. The pivot for the modulation
to v (B minor) is i/iv. The new key is subsequently confirmed by a strong cadence in mm.
15–16.

EXAMPLE 2.25 Bach, Prelude in E Minor (WTC I), mm. 1–16

57 Expanding the Diatonic Palette

A Summary Example: Bach, French Suite No. 2, Menuet (Example 2.26)

This is an excellent example with which to conclude this chapter, since it illustrates many
of the ideas expressed so far, but also points the way toward new territory we will be discussing
as we get deeper into music analysis. We begin by examining the movement’s formal
organization and overall tonal structure. This movement, like the gavotte discussed earlier
(Example 2.20), is in binary form, and we have outlined its main formal divisions and
subdivisions on the score (Example 2.26). The first part consists of an eight-measure period
that modulates to III (E). It is divided into two phrases, the first ending on the dominant
and the second modulating to III. The pivot in the modulation is i5–6/IV6.11 The two phrases
are marked “forephrase” and “afterphrase,” a correct if somewhat awkward translation of
the German terms Vorsatz and Nachsatz.12 The second part opens with a period that is similar
in certain respects to the opening one. The forephrase leads from tonic to dominant in the
new key, and the afterphrase modulates to another new key. In this case there is no pivot,
but rather the bass note E of m. 9 is led via the chromatic passing tone E in mm. 13–14
to F in m. 15, after which the modulation is confirmed by the perfect authentic cadence in
mm. 15–16. Heard in relation to E (III), this is a modulation to its supertonic, but, in relation
to the original key, it is a modulation to iv. The next eight measures are a descending fifth
sequence connecting the subdominant to the dominant. We might expect a return to the
tonic at this point, but instead the forephrase of the final eight measures extends the dominant,
the tonic coming only at the beginning of the afterphrase, which concludes with a perfect
authentic cadence on the tonic. The overall tonal organization, then, is i – III (m. 8) – iv
(m. 16) – V (m. 24) – i (m. 32).
The discussion above tells us a good deal about this short movement, but, in fact, it has
much more to tell us if we look more closely. To this end we have prepared a simplification
of the voice leading with additional analytical notations (Example 2.27). If we were to ask
ten different music analysts to prepare a simplified representation of the score, we would
probably get ten slightly different results depending on decisions about how much to include
or exclude, but they would all agree in principle. The point is this: there is no “right” solution,
but several “right” ones; however, there are clearly “wrong” ones, where there is a mis-
reading of the harmony or voice leading. We will not explain every measure, but rather
we will talk through the decisions made in the opening phrase. Later on, beginning in
m. 17, a conscious decision was made to include less detail than earlier.
In the first measure, the motivic gesture in the right-hand part masks an underlying
stepwise motion in quarter notes moving in contrary motion to the bass. The relationship
between the voices is a voice exchange, where the middle of the three notes is passing. In
the second measure, a modified statement of the motive in eighth notes is taken by the left-
hand part, masking a motion in parallel tenths with the top part leading back to tonic harmony
supporting E5 on the downbeat of m. 3. In the third measure E5 is decorated by a turn
figure before progressing to D5 and on down to B4 supported by the dominant in m. 4.
Once again the eighth-note motion in m. 4 covers an underlying quarter-note progression
in the top part, which forms a voice exchange with the lower part. We have indicated
the motion in parallel tenths in mm. 2–3 between the staves in Example 2.27, and above
mm. 2–4 we have indicated that these measures should be compared to mm. 17–24. The
latter engages a new topic, that of motivic parallelism, a topic we will touch on here and
58 Principles of Harmony and Phrase Design

EXAMPLE 2.26 Bach, French Suite No. 2 in C Minor, Menuet

59 Expanding the Diatonic Palette

EXAMPLE 2.27 Voice-leading simplification of Example 2.26

60 Principles of Harmony and Phrase Design

there, as appropriate, in later chapters. In this instance, the motion in parallel tenths in mm.
2–3, beginning from A5 over F4 and leading to E5 over C4, is expanded by the sequence
beginning in m. 17, where each 10 falls every other measure, the pattern breaking at E5
over C3 in m. 23. If this parallel is not convincing at first, compare the continuation (mm.
23–24) to mm. 3–4. The arrow from the E5 in m. 23 to the D3 two measures later indicates
that the line, clearly articulated by the eighth-note motive, continues in the lower part. Once
the parallel between mm. 2–4 and 17–24 is pointed out, this aspect of the composition’s
fabric becomes clear. If we were to scan this short movement for additional statements of
this motivic idea A–G–F–(E), we would find that it occurs numerous times in a variety of
contexts. See mm. 5–6 (inner voice), 15–16 (top voice), 25–26 (inner voice), 29–30 (inner
voice), and 30–31 (top voice), where the descent to closure occurs. Less obvious, perhaps,
are references to this idea in mm. 10–11 (right hand) and 12 (left hand). One goal of this
text is to make you aware of relationships of this sort so you can uncover them in your own
study of music.
In this chapter we have employed two types of reductions, which we have referred to
as simplifications and representations of the voice leading. In the first type, note values
represent relative durations once notes of melodic and rhythmic embellishment have been
removed. See Examples 2.10b, 2.11b, and 2.27. You should use the last of these as a model
in preparing the assignments suggested below. In the second type, we are interested only in
representing the underlying voice leading, as in Examples 2.9b and 2.12b, where note values
do not represent duration. A fuller explanation of these two types of reductions is provided
in the Appendix.
61 Expanding the Diatonic Palette

SUGGESTED ASSIGNMENTS

Secondary Dominants and Diminished Seventh Chords

1. Beethoven, Piano Sonata, op.14, no. 2, II, mm. 1–20. Note the unusual resolution of
the B4 in m. 7. What gives rise to this? Where is its resolution?
62 Principles of Harmony and Phrase Design

Modulation
We suggest you prepare analyses of one or both of the following movements by Bach:
Menuet from French Suite No. 3 (a) and the Sarabande from French Suite No. 1 (b). Follow
these steps:
1. On the score identify the major formal divisions (periods) and their subdivisions into
phrases, identifying the relationship between phrases (e.g., antecedent-consequent,
forephrase-afterphrase). Indicate the harmonic goals of phrases and, where appropriate,
indicate the pivot of a modulation.
2. Prepare a simplification of the voice leading by removing all notes of melodic embellish-
ment and rhythmic displacement. You may find this difficult at first, so the instructor
may want to begin this as an in-class project. After completing this step, consider specific
relationships between voices, like motion in parallel tenths or voice exchange, and
indicate below your reduced score levels of harmonic organization.

(a) Bach, Menuet from French Suite No. 3

63 Expanding the Diatonic Palette

(b) Bach, Sarabande from French Suite No. 1

64 Principles of Harmony and Phrase Design
3 Phrase Design

Up to this point our focus has been largely on understanding the principles that direct the
harmonic organization of the music of the eighteenth and nineteenth centuries. In this
chapter, we begin our study of formal organization. First, we consider what constitutes a
musical phrase and discuss the most common ways that phrases are constructed and
combined. Then we consider how phrases can be expanded. The chapter concludes with
an analysis of a complete piece by Haydn that comically plays with our expectations about
phrase design. Although the main purpose of this chapter is to establish formal principles in
advance of our study of longer pieces in Part II of this book, even within this chapter we
will see how composers both worked within a set of norms for phrase construction and also
stepped outside these norms, balancing the need for familiarity with the pleasure of surprise,
the goal of comprehensibility with the passion of expression.1

Phrase Design

What is a phrase?
The term “phrase” is one that all musicians use frequently, often without giving a second
thought to what constitutes a musical phrase. There is a shared sense among musicians that
“phrase” designates a segment of music that belongs together and that conveys motion—a
feeling of “beginning – middle – end.” Where musicians’ conceptions of “phrase” differ is
typically in the length of musical passage referred to. In the intensity of a rehearsal or coaching,
many things are often called “phrases”—from units of perhaps two measures to sixteen or
more measures. And, it is certainly true that skilled performers provide a sense of shape both
to short spans of music and to longer ones at the same time. The most inspiring performances
ensure that local shaping never interferes with the projection of larger-scale musical spans.
This is one of the reasons why music analysts have become increasingly precise in our usage
of the term phrase, reserving it for the larger spans that have a true capacity for “beginning
– middle – end.”
In the repertoire under consideration in this book, musical phrases come in a variety of
lengths, but four measures and eight measures are the most common. Length does not
determine what constitutes a musical phrase, but it is helpful to remember that a true musical
phrase hardly ever has a length shorter than four measures. Phrases are defined by cadences:
if there has been no tonal motion leading to a cadence, there has been no phrase. Thus, our
discussion of phrase design must begin with a review of cadences.
66 Principles of Harmony and Phrase Design

Although cadences are classified in terms of their harmonic content, they arise from
a combination of harmonic, melodic, and rhythmic factors. Cadences are formed only
by a very small number of harmonic progressions, but the mere presence of one of these
harmonic progressions does not automatically indicate the presence of a cadence. The
harmonic progression must be wedded to a degree of completion in the melody.
Often, there is also a reduction in rhythmic activity at the cadence, but this is not always
the case, especially in music of the early eighteenth century, such as much of Bach.
Remember, however, that a cadence rarely occurs less than four measures away from adjacent
cadences.
There are two principal types of cadences: those that end on tonic harmony and those
that end on dominant harmony. The former, referred to as authentic cadences, move
from a root-position V (or V7) to a root-position I. Both the V and I chords must be in
root position for an authentic cadence to occur. Cadences that end on the dominant are
referred to by a variety of interchangeable names: half, inauthentic, and semicadence; we
will refer to them as half cadences (HC). In a half cadence, there is only a restriction on
the ultimate chord, which must be V in root position.2 The preceding chord is often one
with predominant function (such as IV, ii6, or V65 of V), but this need not be the case. Some
writers on music also describe “deceptive” and “plagal” cadences, corresponding to the
harmonic motions V (or V7) – vi and IV – I respectively. As will become clear below, these
progressions hardly ever articulate the level of form corresponding to the musical phrase and
thus rarely constitute cadences. Most often, a V – vi progression delays arrival at the harmonic
goal, necessitating another run at closing the phrase—a motion that achieves an authentic
cadence. When IV – I occurs at the end of a melodic unit, its function is almost always
post-cadential: it comes after a V – I progression that actually brought closure to the phrase
with an authentic cadence.
It is useful to distinguish between an authentic cadence where the upper line ends on
1 and one where the upper line remains on 3^ or 5^.3 The former, referred to as a perfect
^
authentic cadence (PAC), has a greater degree of finality than the latter, the imperfect
authentic cadence (IAC). Thus, in degree of finality, there is an increase from half cadence
to imperfect authentic cadence to perfect authentic cadence. Certain types of cadences are
used in particular locations within pieces to provide the appropriate degree of closure.

Phrase Construction and Combination—Part 1: The Period

In Chapter 1, we already introduced an important type of phrase organization: the parallel
period, which consists of an antecedent phrase and a consequent phrase. We discussed
three examples from Mozart’s piano sonatas (Examples 1.8, 1.9, and 1.10), and we reproduce
the second of these here as Example 3.1. Two phrases constitute a musical period if the first
ends with a weaker cadence (HC or IAC) and the second ends with a stronger cadence
(IAC or PAC) and it makes sense to group the two phrases together into a larger formal
entity. By far, the most common cadential pairing is HC – PAC, as seen in Example 3.1.
The term “parallel” refers to the deployment of the same melodic material at the beginnings
of the antecedent and consequent phrases. The vast majority of periods are parallel. A period
where the consequent does not begin with the same material as the antecedent is termed a
contrasting period, and we will provide an example of this later (see Example 3.2).
67 Phrase Design

EXAMPLE 3.1 Mozart, Piano Sonata in B major, K. 333, III, mm. 1–8

Note that in Example 3.1 the two phrases are of equal length. This is what usually
happens in a period, and when this is not the case it is almost always because the consequent
is longer than the antecedent. We will explore the topic of phrase expansion later in this
chapter. The most common length for a period is eight measures (4 + 4), but other lengths
are found, especially sixteen measures (8 + 8). Within each of the phrases in a period there
is often a binary division, that is, a subdivision of the phrase into two-measure subphrases.
In Example 3.1 this is particularly clear given the rest near the end of the second measure
and the highly contrasting rhythm and melodic contour of mm. 3–4 compared to mm. 1–2.
If we list only the initial and final harmonies of each phrase in Example 3.1, our analysis
would be I – V for the antecedent and I – I for the consequent. Since it is possible to have
cadential pairings other than HC – PAC, other tonal plans are possible. The four tonal plans
below account for nearly all period designs:

Antecedent phrase Consequent phrase

I – V (HC) I – I (PAC) interrupted
I – V (HC) X – I (PAC) continuous
I – V (HC) I – X (PAC) progressive
I – I (IAC) I – I (PAC) sectional

A full description of the organization of Example 3.1, then, would be “parallel interrupted
period.” We use the term interrupted because of the stoppage of the harmonic motion at
the dominant, after which the consequent begins again with tonic harmony. Often, there
is an accompanying melodic interruption as well, whereby the melody descends to 2^ at the
end of the antecedent (the C in m. 4 of Example 3.1) and, after beginning again at its original
pitch, descends to 1^ at the end of the consequent (the B at the start of m. 8).
We now offer an example of each of the other tonal plans listed above. In continuous
and progressive periods, the antecedent phrase still ends on V, but either the beginning or
the end of the consequent does not present the initial tonic harmony. In a continuous
period, the consequent phrase begins with a non-tonic harmony, almost always the dominant
or the supertonic. In the initial eight measures of the minuet from Beethoven’s Piano Sonata,
op. 22, the consequent phrase picks up with the dominant harmony from the end of the
68 Principles of Harmony and Phrase Design

antecedent (Example 3.2). Unlike our other examples, the consequent phrase does not begin
with the same material as the antecedent; thus, this is a contrasting period. The turn figure
with which the consequent begins does derive from the antecedent, but not from the begin-
ning of the phrase. When we are making the distinction between parallel and contrasting
periods we are comparing the beginnings of phrases. In a contrasting period, the consequent
often has a motivic connection to some portion of the antecedent, just not to the begin-
ning. The minuet from Beethoven’s Piano Sonata, op. 10, no. 3 (Example 3.3) begins with
a parallel continuous period, but here the harmony at the start of the consequent is
supertonic (not dominant). Note that this period spans sixteen measures (8 + 8). Of course,
this period does not take twice as long to perform as the previous one given its faster tempo
(observe the absence of any subdivisions smaller than the quarter note).

EXAMPLE 3.2 Beethoven, Piano Sonata in B major, op. 22, III, mm. 1–8

A progressive period involves modulation during the consequent phrase. In theory,

the modulation could be to any key, but in practice the modulation is usually to the dom-
inant (in the case of major mode) or to the relative major (in the case of minor mode).
Instead of the term “progressive,” some authors refer to this organization as a “period with
modulating consequent.” An example is the opening eight measures of the finale of Haydn’s
Piano Sonata No. 34 in E minor (Example 3.4). The consequent preserves the first measure
of the antecedent before the pivot chord (m. 6: iv6/ii6) takes the period to a PAC in the
relative major.
69 Phrase Design

EXAMPLE 3.3 Beethoven, Piano Sonata in D Major, op. 10, no. 3, III, mm. 1–16

EXAMPLE 3.4 Haydn, Piano Sonata in E Minor, Hob. XVI/34, III, mm. 1–8
70 Principles of Harmony and Phrase Design

EXAMPLE 3.5 Mozart, Piano Sonata in B major, K. 281, I, mm. 1–8

The final tonal plan, the sectional period, differs from all of the others in that its first
phrase is closed. Mozart’s Piano Sonata, K. 281, opens with a parallel sectional period due
to the tonic harmony at the end of the antecedent phrase (Example 3.5). Note that the
cadence in m. 4 is an IAC since 3^ (D) occurs in the melody.

Phrase Construction and Combination—Part 2: The Sentence

A period is a design involving two phrases. We turn now to a common way in which an
individual phrase can be organized: sentence design.4 In a musical sentence, a basic idea is
stated and immediately repeated, and then the phrase continues to a cadence. We will refer
to the statement of the basic idea and its repetition as a presentation and the remainder of
the phrase as a continuation. The continuation is typically the same length as the
presentation; when this is not the case it is almost always because the continuation is longer
than the presentation. Most often, a musical sentence consists of eight measures: a two-
measure basic idea, its repetition, and a four-measure continuation. However, sentences of
sixteen measures (4 + 4 + 8) and, in slower tempos, of four measures (1 + 1 + 2) are by
no means unusual.
As with period design, a sentence can have various tonal plans. It is useful to identify
three different tonal plans for presentations. First, the repetition of the basic idea can be
exact, meaning that it expresses the same harmony as the initial statement. (This harmony
is almost always tonic harmony.) Second, the repetition of the basic idea can be a response
to the initial statement, meaning that it exchanges dominant harmony for the tonic harmony
71 Phrase Design

of the original statement. This can occur in two ways. If the basic idea expresses only tonic
harmony, then the response will express only dominant harmony; if the basic idea progresses
I – V, then the response goes V – I. Bear in mind that even when these harmonies occur
in root position they do not have a cadential function: a basic idea and its repetition do not
form a melodic unit that is sufficiently complete to constitute a musical phrase. Third, the
repetition of the basic idea can be sequential, meaning that it transposes the initial statement
of the basic idea to another harmony (other than the dominant); this other harmony is usually
supertonic (and thus this type of presentation is not characteristic of sentences in the minor
mode).
The continuation typically brings a feeling of intensification. This quality often arises
from a shortening of the melodic subphrases. In a sentence with a two-measure basic idea,
the continuation often begins with a pair of one-measure segments before a two-measure
unit leads to the cadence. This shortening of melodic units generally results from a literal
fragmentation of the presentation’s basic idea, but the continuation need not have a tight
thematic connection to the presentation. Other features commonly found in a continuation
include an increase in harmonic rhythm, a speeding up of surface rhythm, and the use of
sequence.
A quintessential example of a musical sentence is the opening of Beethoven’s Piano
Sonata, op. 2, no. 1 (Example 3.6). The presentation is of the response type, since the basic

EXAMPLE 3.6 Beethoven, Piano Sonata in F Minor, op. 2, no. 1, I, mm. 1–8
72 Principles of Harmony and Phrase Design

idea expresses tonic harmony, while the repetition expresses dominant harmony. The start
of the continuation brings a return to tonic harmony. In Chapters 1 and 2, we frequently
mentioned that harmony operates on different levels. When one looks at the overall tonal
structure of a sentence, it is almost always the case that the local chord changes within the
presentation participate in a prolongation of the initial tonic harmony that lasts at least until
the start of the continuation. In Beethoven’s sentence, note that the first two measures
of the continuation present a fragment—the second half—of the basic idea. There is thus
both a shortening of subphrase length and a clear motivic connection to the presentation.
In addition, the harmonic rhythm accelerates to one chord per measure in mm. 5–6 and
then to two chords per measure in m. 7. A very similar construction occurs near the start
of Beethoven’s Fifth Symphony, except that the sentence spans sixteen, rather than eight,
measures (Example 3.7). We will provide examples of presentations with exact repetition
and with sequential repetition later in this chapter (see Examples 3.11 and 3.15 respectively).

EXAMPLE 3.7 Beethoven, Symphony No. 5, op. 67, I, mm. 6–21

73 Phrase Design

Phrase Construction and Combination—Part 3: Larger Period Design

We now turn to a larger design based on the concept of the period, namely a period where
the antecedent and consequent phrases are each in sentence design. Since the most common
length for a sentence is eight measures, the most common length for a period comprised of
two sentences is sixteen measures. The opening of the slow movement from Beethoven’s
Piano Sonata, op. 10, no. 1, which was studied in Chapter 1 (Example 1.17), provides a
good illustration; we reproduce Example 1.17 as Example 3.8. As noted in Chapter 1, there

EXAMPLE 3.8 Beethoven, Piano Sonata in C Minor, op. 10, no. 1, II, mm. 1–16
74 Principles of Harmony and Phrase Design

is an eight-measure antecedent leading to the dominant and an eight-measure consequent

ending with a PAC. Looking more closely at the antecedent, we observe a basic idea in
mm. 1–2 that is repeated over dominant harmony in mm. 3–4. Thus, like our previous two
examples, the presentation is of the response type. At the start of the continuation (m. 5),
there is a sense of acceleration since the new double-dotted melodic figure is stated on three
successive beats in contrast to the slower repetition of mm. 1–2 in mm. 3–4. Recognizing
that mm. 1–8 deploy melodic material in sentence design helps us realize immediately that
the root-position dominant in m. 4 does not end the phrase by articulating a HC. The
melodic structure is not sufficiently complete for a cadence to occur at this point.
We have only discussed the most common types of phrase organization. Not every phrase
can be described as a sentence design or understood as part of a musical period. Some phrases
are simply phrases. Being aware of the most common ways of organizing the melodic content
of a phrase, however, expedites determining where phrases begin and end. As we will see,
it also enables us to recognize when a composer writes a phrase that is modeled on a common
phrase design but departs from it in some way. Such departures have structural and expressive
significance.

Phrase Elision and Hypermeter

Phrase Elision
In all of the examples up to this point, every phrase ends before the subsequent phrase
begins. This is not always the case. Sometimes the final event of a phrase coincides
with the start of the next phrase. The slow movement of Beethoven’s Piano Trio, op. 1,
no. 2 (Example 3.9) provides an example. The first phrase is in sentence design; mm. 1–4
constitute a statement-response presentation of the I – V, V – I type (rather than the simpler
I, V type seen thus far). The continuation does not achieve its cadence within four measures;
the tonal motion does not stop on the V7 chord at the end of m. 8. The V7 chord resolves
to the E major harmony in m. 9, which coincides precisely with the start of the melody in
the violin. We will refer to this as phrase elision; some writers use the equivalent term
phrase overlap. Phrase elision helps to maintain musical flow and prevent the music from
becoming excessively segmented.
Another instance of phrase elision occurs near the start of Haydn’s Symphony No. 104;
Example 3.10a provides the orchestral score, while Example 3.10b gives a piano reduc-
tion with analytic annotations.5 After the eight-measure antecedent, one expects an eight-
measure consequent leading to a PAC. Tonic harmony does arrive eight measures after the
start of the consequent, but the dynamics are suddenly fortissimo and winds, brass, and timpani
join the strings in a jolting orchestral tutti. The subsequent music continues in this new
dynamic and texture; in other words, the cadential arrival elides with the next phrase.
We will return to both the Haydn symphony and the Beethoven trio below and explore
a bit further why the Haydn feels quite different from the Beethoven even though both
exhibit the same phenomenon of phrase elision.
75 Phrase Design

EXAMPLE 3.9 Beethoven, Piano Trio in G Major, op. 1, no. 2, II, mm. 1–9
76 Principles of Harmony and Phrase Design

EXAMPLE 3.10a Haydn, Symphony No. 104, I, mm. 17–33

77 Phrase Design

EXAMPLE 3.10b Annotated piano reduction of Example 3.10a

78 Principles of Harmony and Phrase Design

Hypermeter
On the examples in this chapter, we have been including Arabic numerals: 1 2 3 4. It probably
appeared that they were simply an aid in counting measures, but they actually have a deeper
significance. Turn back to the excerpt we studied from Beethoven’s Fifth Symphony
(Example 3.7), and try to conduct it. Beethoven’s metronome marking is  = 108, meaning
that each measure lasts just over half a second. Clearly, it is not possible to conduct the
2
quarter notes in Beethoven’s notated 4 meter. Instead, the pulse that a conductor would
beat in this movement corresponds to the notated measure. A conductor would beat a
4
quadruple pattern (i.e., the conducting gestures for 4 meter) across four-measure units. Thus,
the “1 2 3 4” annotations on the score represent a type of meter—not just a counting of
numbers of measures. All of the measures marked “1” have a strength or metric accent in
4
the same way that the first beat in a 4 measure does. In the music of the eighteenth and
nineteenth centuries, meter—like harmony— operates on multiple levels. Meter that
operates above the level of the notated meter is referred to as hypermeter. In the case of
Beethoven’s Fifth the hypermeter actually corresponds to the spontaneously perceived meter
of the listener; often, the notated meter is the perceived meter and hypermeter is perceptible
only when one specifically focuses on its slower pulses. We will use the term hypermeter,
though, for any meter above the level of the notated measure, regardless of whether it is
the perceived meter or a perceptible meter.
In discussing hypermeter, analysts adapt terminology used in discussing meter. Just as
measure refers to a cyclical grouping of beats within a meter, the term hypermeasure
refers to the analogous grouping within a hypermeter. Thus, a four-measure unit labeled as
“1 2 3 4” would constitute a hypermeasure within a quadruple hypermeter. The start of
a hypermeasure is a hypermetric downbeat—just as the first beat of a measure is a down-
beat—or in abridged form, hyperdownbeat. The subsequent beats in a hypermeasure
can be designated as “second (or third, or fourth) hypermetric beat” or more succinctly
“hyperbeat 2 (or 3, or 4).”
Not all hypermeter is quadruple, but this is by far the most common scenario. Extended
passages with triple hypermeter are rare. A famous instance occurs in the scherzo of
Beethoven’s Ninth Symphony, and the composer specifically tells the conductor “ritmo di
tre battute.” As in the Fifth Symphony, this very fast movement is a piece where the perceived
meter (and the meter given by the conductor’s gestures) is actually the hypermeter. The
explicit indication of triple hypermeter in this passage is helpful, since elsewhere the music
proceeds in quadruple hypermeter (which Beethoven indicates as “ritmo di quattro battute”).
Some pieces have duple hypermeter—that is, a simple alternation of strong and weak
measures rather than the “strong–weak–medium–weak” feeling of quadruple hypermeter.
However, as we will see below, it is possible to have an underlying quadruple hypermeter
even when the music does not literally proceed in four-measure units throughout. Therefore,
it is important not to infer duple hypermeter without considering the possibility of quadruple
hypermeter.
Except for a very few pieces like the scherzo from Beethoven’s Ninth, hypermeter is not
explicitly notated in the score. This is different from pitch or rhythm or meter at the level
of the notated meter. What, then, creates our feeling of hypermeter? How can we discern
hypermeter when looking at a score? Hypermeter is closely related to phrase and subphrase
design. Phrases (and large subphrases) generally begin at or slightly before hypermetric
79 Phrase Design

downbeats. In addition, when musical material is repeated, it retains its original hypermetric
identity as much as possible. The onset of an event that is sustained for a relatively long time—
a pitch, a harmony, a texture—can also be indicative of a hypermetric downbeat. Thus, the
sense of “beginning” is associated with hypermetric downbeats. As we shall see, when events
with strongly initiatory quality occur unexpectedly close to, or far from, one another, there
is often a contraction, or expansion, within the hypermeter.
We return now to the examples of phrase elision from the Beethoven trio and the Haydn
symphony. In the Beethoven trio (Example 3.9), the elision keeps the hypermeter periodic.
Since the first phrase does not end until the downbeat of m. 9, only by launching the new
phrase at that downbeat can the quadruple hypermeter continue without disruption. In the
Haydn symphony (Example 3.10), on the other hand, the elision introduces an irregularity
in the hypermeter. A measure that we expected to be hypermetrically weak—hyperbeat
4—is made to function as a hypermetric downbeat for the following phrase. In Example
3.10b, notice that the hypermetric analysis of the measure containing the elision is 4 = 1.
Just as the tonic harmony at the start of this measure both concludes a phrase and launches
the next one, hypermetrically this downbeat both provides the last hyperbeat of the
preceding phrase and the hypermetric downbeat for the following one. When a measure
has two hypermetric functions, we refer to this as a hypermetric reinterpretation.6 The
hypermetric reinterpretation is another reason why this measure has such a jolting effect,
besides the shock of the fortissimo and entrance of winds, brass, and timpani. By far, the most
common hypermetric reinterpretation is 4 = 1, but it is certainly possible for a new
hypermetric downbeat to intrude even further into a preceding hypermeasure (e.g., 3 = 1
in a quadruple hypermeter). As we have seen, phrase elision may or may not involve
hypermetric reinterpretation. Hypermetric reinterpretation, though, almost always involves
elision, if not elision of two phrases then elision of two subphrases.
It is difficult to make generalizations about where elision tends to occur. One piece of
advice: If a pair of phrases that begin similarly can be placed in an antecedent-consequent
relationship (i.e., a HC at the end of the first phrase), this is almost always the best analysis
(rather than an authentic cadence at the end of the first phrase elided to the start of the
second phrase). Glance back at Examples 3.1 and 3.8. In both cases there are some short
notes that provide rhythmic continuity between the end of the antecedent and the start of
the consequent. The underlying structure, though, is the arrival on the V triad; the melodic
embellishment provides some momentum towards the next downbeat but does not continue
the phrase to the next downbeat.

Phrase Expansion
The phrases in all of our examples thus far have been either four, eight, or sixteen measures
in length. While these are the most common lengths, phrases of many other lengths occur.
When they do, it is usually possible to understand how they relate to a four-measure, eight-
measure, or sixteen-measure model. Generally, phrases of irregular length are longer, not
shorter, than the “normative” model to which they may be related. Thus, this section is
titled “Phrase Expansion,” although phrase contraction occasionally takes place. We will
discuss several techniques by which phrases can be expanded. In addition, we will consider
the hypermetric consequences of phrase expansion.
80 Principles of Harmony and Phrase Design

External Phrase Expansions

Phrase expansions can be divided into two types: those that are external to the phrase and
those that are internal. An external expansion is one that occurs before the real beginning
of a phrase or that extends a phrase after its cadence. In the analytic literature, one encounters
a variety of terms to denote an expansion before the beginning of a phrase, including intro-
duction, vamp, and prefix. Similarly, multiple terms exist for expansions after the cadence,
including post-cadential expansion, codetta (for an extension of an authentic cadence),
standing-on-the-dominant (for an extension of a half cadence), and suffix. We prefer the
terms introduction and post-cadential expansion (or extension) to refer to the two types of
external expansion.
The slow movement of Beethoven’s String Quartet, op. 135, demonstrates both types
of external expansion (Example 3.11). The underlying phrase is an eight-measure sentence
(mm. 3–10): mm. 3–6 constitute a presentation with exact repetition and mm. 7–10 a
continuation leading to a PAC on the second beat of m. 10. Note the shortening of melodic
subphrases from two measures to one measure when the continuation begins, as well as the
quickening of the harmonic rhythm and the use of sequence. Before this sentence begins,
there is a two-measure introduction that unfolds the tonic harmony; this is an external
expansion. After the PAC in m. 10, there is a two-measure external expansion. The cello
imitates the first violin’s cadential figure A–F–E–D, which returns to the first violin in the
next measure. Harmonically, mm. 11–12 prolong the tonic harmony from the second beat
of m. 10 until the downbeat of m. 13, whereupon an ornamented repetition of the eight-
measure sentence begins in a higher register. The first twelve measures are thus best
understood as an expanded eight-measure phrase.
What are the hypermetric consequences of these phrase expansions? When there is a
significant amount of expansion present, it is often useful to distinguish between a surface
hypermeter and an underlying hypermeter. In the Beethoven string quartet, there is a
continuous duple hypermeter throughout the entire excerpt; the odd-numbered measures
are hypermetrically stronger than the even-numbered ones. This constitutes the surface
hypermeter of the passage. If we take the phrase expansions into account, we can under-
stand an underlying quadruple hypermeter in mm. 3–10. Measures that do not participate
in the underlying hypermeter can be designated with “———” in the hypermetric analysis
to indicate that they expand another hyperbeat or by placing parentheses in the hypermetric
analysis. In this excerpt, the surface and underlying hypermeters have a straightforward
relationship; measures that are hypermetrically strong (or weak) in the underlying hyper-
meter possess the same relative hypermetric strength in the surface hypermeter. As we will
see below, the relationship between surface and underlying hypermeters can be more
complex.
We turn now to a phrase studied near the end of Chapter 2 (Example 2.22), the second
phrase of Mozart’s Piano Sonata, K. 310 (Example 3.12). As discussed in the last chapter,
the phrase modulates from A minor to C major, ending with a half cadence. The phrase is
an expanded eight-measure sentence; note the acceleration in harmonic rhythm in the fifth
measure. The shortening of melodic units is less obvious, but observe the handling of the
descending slurred two-note idea. It occurs in the second and fourth measures of the
presentation, and it occurs three times in close succession at the start of the continuation.
The goal dominant arrives at the phrase’s eighth measure (m. 16) and is extended for seven
81 Phrase Design

EXAMPLE 3.11 Beethoven, String Quartet in F Major, op. 135, II, mm. 1–14
82 Principles of Harmony and Phrase Design

measures (mm. 16–22). The relationship between this post-cadential expansion and the
underlying phrase differs from that in the Beethoven string quartet. The expansion does not
simply repeat the cadential gesture; rather it enters boldly with a new left-hand texture and
a return of the dotted-note motive from the start of the phrase. There is thus some initiatory
quality within the expansion, and the expansion itself proceeds in two-measure melodic
units with a change in melodic content and rhythmic texture at its fifth measure (m. 20).
In the underlying hypermeter, the final hyperbeat coincides with the arrival of the dominant
at m. 16, and the post-cadential expansion delays the next hyperbeat until the subsequent
phrase begins in m. 23. The expansion projects its own surface hypermeter, reinterpreting
hyperbeat 4 in the underlying hypermeter as hyperbeat 1 in its surface hypermeter. Thus,
the downbeat of m. 16, which is hypermetrically weak in the underlying hypermeter, is a
hypermetric downbeat in the expansion’s surface hypermeter. Distinguishing underlying and
surface hypermeters recognizes the two qualities of musical time operative in this passage:
the expansion is rhythmically energetic but ultimately does not advance the tonal story of
the music.

EXAMPLE 3.12 Mozart, Piano Sonata in A Minor, K. 310, I, mm. 9–23

83 Phrase Design

Internal Phrase Expansions

Phrases can be expanded internally through several techniques, including repetition, composed-
out deceleration, parenthetical insertion, and cadential evasion. We will demonstrate each of
these techniques, either acting individually or in combination.
Consider the opening of Beethoven’s Piano Sonata, op. 2, no. 3 (Example 3.13).
Measures 1–8 consist of an eight-measure sentence with statement-response presentation
and a continuation that begins with fragmentation of the presentation’s basic idea. The
cadence in m. 8 is an IAC. Beethoven repeats the entire continuation (with the melody
placed in the bass), leading to a PAC at the downbeat of m. 13. The cadence elides with
the beginning of the next phrase, an elision that maintains the quadruple hypermeter. Taken
as a whole, the true cadence of the phrase is the PAC—not the IAC—and thus we can
consider the expansion as internal to a phrase spanning mm. 1–13. Strictly, there is also a
composed-out deceleration involved. Notice that in m. 8 the dominant harmony lasts only
two beats, whereas in the repetition of the continuation dominant harmony governs mm.
11–12. This delay allows the goal tonic harmony to arrive on a downbeat, which in addition
to the melodic closure on 1^, gives the cadence greater impact than the third-beat cadence
of m. 8.

EXAMPLE 3.13 Beethoven, Piano Sonata in C Major, op. 2, no. 3, I, mm. 1–13
84 Principles of Harmony and Phrase Design

Extremely similar is the start of the second movement of the same sonata (Example 3.14).
This movement opens with a four-measure antecedent phrase leading to the dominant. The
antecedent is itself in sentence design, an instance of a four-measure sentence (note the Adagio
tempo). The consequent phrase, also in sentence design, begins on the supertonic—as often
happens in a continuous period—and reaches an IAC in the middle of m. 8. Beethoven
repeats the continuation (with the left and right hands exchanging material) to attain a PAC
at the downbeat of m. 11, which elides with the start of the next phrase. As in the first
movement of this sonata, the repeat of the continuation is slightly lengthened so that the

EXAMPLE 3.14 Beethoven, Piano Sonata in C Major, op. 2, no. 3, II, mm. 1–11
85 Phrase Design

EXAMPLE 3.15 Mozart, Piano Sonata in C Major, K. 309, III, mm. 1–19

tonic harmony arrives on a downbeat rather than in the middle of a measure. Our careful
study of phrase design in the initial themes of these two movements has revealed a striking
similarity in construction, despite their vastly different moods.
Expansion can result from repetition that occurs earlier in a phrase. Slightly earlier is
the repetition in the second phrase of the finale from Mozart’s Piano Sonata, K. 309 (Example
3.15). Based on eight-measure sentence design, this phrase is on the brink of a PAC when
the first three measures of the continuation are repeated with slight variation. Instead of
eighth notes expressing a rising third (as in mm. 13–14), sixteenth notes fill in falling sixths
(mm. 16–17).
Mozart’s Clarinet Quintet, K. 581, provides a fascinating instance of phrase expansion
in its minuet movement. The second trio of that movement begins with the parallel period
given in Example 3.16.7 This period is striking in several regards. First, the antecedent ends
with V7 – vi; the deceptive progression actually serves in a cadential capacity here. Second,
note that the consequent returns the clarinet melody and upper strings exactly but places
C rather than A in the cello, substituting first-inversion for root-position tonic harmony.
86 Principles of Harmony and Phrase Design

EXAMPLE 3.16 Mozart, Clarinet Quintet, K. 581, III, Trio II, mm. 1–12
87 Phrase Design

Third, and most importantly for our present discussion, the consequent phrase has four
measures of internal expansion (mm. 7–10). In the antecedent ii65 – V7 occurs in mm. 2–3;
in the consequent ii6 begins, as expected, in m. 6 but is reiterated for an additional four
measures until the dominant arrives at m. 11. Notice that on the last beat of m. 10 the
clarinet returns to the exact same pitches it had on the last beat of m. 2 (shown by the dotted
bracket on the score). This makes the return from the digressive measures particularly smooth,
as does the clarinet’s touching on F5 at the downbeat of m. 10, as this was the clarinet’s
last pitch before the expansion began. Since the expansion consists of four measures and the
underlying hypermeter is quadruple, the surface hypermeter continues without interruption
during the expansion (i.e., 3 4 1 2 in mm. 7–10).
It is perfectly clear that Mozart’s consequent phrase has been expanded and exactly where
this has occurred. Less obvious is whether the expansion arises through repetition or
parenthetical insertion. The first violin echoes the clarinet’s melody—giving the impression
of expansion through repetition—but the expansion has an abruptly new texture and
accompaniment pattern (or lack thereof), which suggests a parenthetical insertion. Agreeing
on the primary technique that generates this expansion is of much less importance than
recognizing the presence of expansion.8
A straightforward example of parenthetical insertion occurs in the minuet from Haydn’s
String Quartet, op. 74, no. 1 (Example 3.17). The phrase is headed for a PAC at the downbeat
of its eighth measure when three measures of highly contrasting material intrude. The
sforzandi, introduction of triplets, and the two-beat grouping of material (in contrast to the
prevailing triple meter) all distinguish the insertion from the surrounding music. This
expansion is quite similar to the one we studied in the C major piano sonata by Mozart
(Example 3.15), except that there the three extra measures repeated material that had occurred
earlier in the phrase. In the Haydn, note that there is also a brief external expansion as the
concluding tonic harmony is extended by a measure.
The only one of the four techniques of internal phrase expansion listed above that we
have not seen is cadential evasion. There are various ways that the resolution of a root-position
dominant can be evaded, but the most common are through the deceptive progression (to
the submediant) or a descending bass line that leads from V through V42 to first-inversion tonic
harmony. The former progression occurs in the second phrase of the finale of Beethoven’s
Violin Sonata, op. 24 (Example 3.18). The first two phrases constitute a parallel interrupted
period with each phrase in sentence design. In the seventh measure of the consequent
(m. 15), a PAC seems imminent but at the end of the measure the bass moves up to vi instead,
necessitating another run at the cadence, which is achieved two measures later.
The slow movement of Mozart’s String Quartet, K. 589, provides an explicit example
of cadential evasion with V42. Example 3.19 gives the movement’s initial four phrases. In
the first two phrases (mm. 1–4 and 5–8), the cello soars above the other strings, which results
in the viola supplying the bass line. These two phrases form a contrasting interrupted period.
Starting in m. 9, the first violin takes the melody and leads a repeat of the preceding two
phrases. The first phrase is repeated exactly in mm. 9–12, but the second phrase is expanded
by two measures. At the end of its third measure (m. 15), the cello steps down from B to
A, and two additional measures are required to attain a PAC (whose tonic harmony elides
with the start of the next phrase). Mozart’s  indication beautifully marks the expansion,
as does the first violin’s momentary touching of the high E6.
88 Principles of Harmony and Phrase Design

EXAMPLE 3.17 Haydn, String Quartet in C Major, op. 74, no. 1, III, mm.49–60
89 Phrase Design

EXAMPLE 3.18 Beethoven, Violin Sonata in F Major, op. 24, IV, mm. 1–18
90 Principles of Harmony and Phrase Design

EXAMPLE 3.19 Mozart, String Quartet in B Major, K. 589, II, mm. 1–18
91 Phrase Design

It is crucial to bear in mind that a minority of deceptive progressions or V42 – I6

progressions generate a phrase expansion. More often, these progressions partake in the
progression that makes up the underlying phrase. Context plays a critical role in determining
whether or not a phrase expansion has occurred. Sometimes the decision is easy, as in the
previous example, where a version of the phrase occurs first without expansion and then
with expansion. In period structures, an expansion of the consequent is readily detected as
well, since the antecedent sets up a definite expectation for the length of the consequent.
Similarly, with sentence design, the length of the presentation specifies the normative length
for the continuation. Sometimes, though, the music offers no such model to assist in our
analytic decision-making. In these cases, the lengths of surrounding phrases still play an
important role. If all of the surrounding phrases proceed in four-measure units (or multiples
thereof ), then a phrase with a non-quadruple-based length is much more likely to be an
expanded phrase than if there are other nearby phrases with unusual lengths. A five-measure
phrase without any expansion rarely occurs, but when it does it would almost surely be
close to (likely adjacent to) another phrase of five measures (or at the very least, a phrase
with a non-duple length).
We close our discussion of phrase expansion with one example of phrase contraction.
The second movement of Schubert’s Piano Sonata, D. 664, opens with a parallel sectional
period where the first phrase consists of seven measures and the second consists of eight
(Example 3.20). A close look at the first phrase reveals that it begins with a four-measure
statement-response presentation. The continuation starts in typical fashion by repeating a
one-measure fragment of the basic idea. The cadence, however, is achieved in a single
measure. The conclusion of the phrase feels early not only because of the contracted phrase
length but also due to the suddenly faster harmonic rhythm in m. 7 that arises from the
passing bass motion inserted between 5^ and 1^. Note that our hypermetric analysis indi-
cates “1 2 3” for the continuation; the hypermeasure simply ends after hyperbeat 3.
The next measure functions only as a hypermetric downbeat for the following phrase; it in
no way continues the preceding hypermeasure (i.e., the hypermetric analysis 4 = 1 would
be inappropriate for m. 8). The consequent phrase is a normative eight-measure sentence.
The movement’s first sonority is an interesting one. Given the slow tempo, we do hear a
first-inversion B minor triad for some duration, but ultimately we understand the B as an
embellishment of the ensuing A (a 6–5 motion above the bass). This motive completely
dominates the melodic structure of these phrases, but its reach extends into the harmony
too. The passing bass motion in mm. 6–7, which was noted above, enables a B minor sonority
to occur on the downbeat of m. 7. In the consequent the presentation is sequential rather
than statement-response, and the sequence is to the submediant (mm. 10–11). The B minor
harmony extends into the first measure of the continuation, which results in mm. 8–13
proceeding from I to I6 via a three-measure vi chord!
92 Principles of Harmony and Phrase Design

EXAMPLE 3.20 Schubert, Piano Sonata in A Major, D. 664, II, mm. 1–15
93 Phrase Design

A SUMMARY ANALYSIS (HAYDN)

As a conclusion to this chapter, we will examine the minuet from Haydn’s String Quartet,
op. 64, no. 6 (Example 3.21). Not only does this analysis allow us to review several of the
chapter’s main concepts, but it also demonstrates some of the challenges that can arise in
studying phrase expansion and hypermeter. The analysis of phrase expansion and hypermeter
sometimes admits the possibility of multiple good interpretations.
The minuet consists of two repeated parts with a return of the opening material mid-
way through the second part (mm. 21ff.), a form known as rounded binary form and that
will be presented in detail in Chapter 5. The first part of this minuet consists of two phrases
that form a parallel progressive period modulating to the dominant. Each of these phrases
is six measures long, and our hypermetric analysis treats them as real six-measure phrases—
not as expanded four-measure phrases (or as contracted eight-measure phrases). It would,
of course, be possible to recompose mm. 1–12 as two four-measure phrases, a possibility
demonstrated in Example 3.22. However, these phrases seem quite short, and the
recomposition removes their most distinctive feature, namely the accented dissonant chord
in the third measure of each phrase. A further argument in support of our hypermetric reading
is the fact that there are two adjacent phrases of the same length. As noted earlier, when an
irregular phrase length recurs, it is much easier to hear the seeming irregularity as normative.
Finally, bear in mind that the hypermetric analysis 1 2 3 4 5 6 implies a regular alternation
of strong and weak measures; that is, hyperbeats 1, 3, and 5 are stronger than hyperbeats 2,
4, and 6. It is notable that in these six-measure phrases, there is a slowing of rhythmic activity
in the fourth measure. In each case, though, the harmony cannot serve as a cadential goal:
IV in the first phrase and a V65 of B in the second phrase. While it is harmonically clear that
the fourth measures are not cadential arrivals, the melody’s rhythmic slowing draws attention
to the lack of cadential articulation and the unusual length of the phrases.
At the start of the minuet’s second part, Haydn works with the eighth-note figure played
by violin I at the approach to the preceding cadence (mm. 11–12). This is a good example
of linkage technique, which entails beginning a new phrase with the melodic figure that
closed the preceding phrase. After two repetitions that continue the dominant harmony,
the figure moves through a descending fifth sequence that returns the music to the tonic
key of E major. This eight-measure phrase leads to a half cadence in the tonic at m. 20.
In m. 21, the opening music returns but reworked to avoid modulation. The goal of
this material is now the PAC in E major at m. 32. Thus, we again have a twelve-measure
span (mm. 21–32), but it is treated very differently than the first twelve measures of the
minuet. Instead of two phrases that form a period, there is a single phrase. Further, the
hypermeter, which is strongly articulated through rests and the change in dynamics, is now
quadruple. On the score, we have shown mm. 21–32 as a twelve-measure phrase that con-
sists of three hypermeasures. Another possibility, indicated above the score, would be to
consider the piano measures (mm. 25–28) as a parenthetical expansion within an underlying
eight-measure phrase. A parenthetical interpretation would be motivated by the sudden
changes in dynamics and the rests that separate these measures from their surroundings.
In addition, there is a clear harmonic connection between the preceding IV harmony
(m. 24) and the ii6 harmony (m. 30) that is the first consonant sonority in the subsequent
forte music. On the other hand, the piano measures trace an almost chromatic stepwise ascent
from the C5 above the IV harmony (m. 24) to the F5 above the ii6 harmony (m. 30). This
94 Principles of Harmony and Phrase Design

EXAMPLE 3.21 Haydn, String Quartet in E major, op. 64, no. 6, III, mm. 1–36
95 Phrase Design

EXAMPLE 3.21 continued

EXAMPLE 3.22 Eight-measure recomposition of mm. 1–12 of Example 3.21

96 Principles of Harmony and Phrase Design

melodic connection cuts across the rests and sharp dynamic changes to support hearing the
piano measures as integral to the phrase. While either of these interpretations is defensible,
a third interpretation would not be. It would not be satisfactory to consider the vi chord
in m. 28 as the goal of a deceptive cadence; mm. 21–32 must be considered as a single
phrase and not as two phrases. Neither mm. 21–28 nor mm. 29–32 is at the same formal
level as mm. 1–6 and mm. 7–12. Regardless of whether one interprets mm. 21–32 as
an expanded eight-measure phrase or as a true twelve-measure phrase, one notes the
presence of quadruple hypermeter and the momentum of a relatively long phrase leading
to the PAC. Compared to mm. 1–12, the reworking of the thematic material resolves the
six-measure lengths and provides a phrase that seems more balanced and satisfying. A beautiful
detail is Haydn’s incorporation of the same chromatic chord in m. 29 as in m. 9; it has a
slightly different function—in m. 9 as part of the modulation to the dominant, in m. 29
as an applied diminished seventh of ii6—but the reuse of this striking sonority provides a
compelling connection amidst the very significant amount of thematic reworking that has
occurred.
The minuet ends with a short four-measure unit that consists solely of a cadential gesture.
When we study complete pieces in Part II of this text, we will see that this type of passage
frequently occurs at the end of a piece after the “big” PAC, and we will refer to such passages
as codettas. Note that this codetta is based on the cadential figure from the minuet’s first
phrase (m. 5). This was an element of the opening phrase that had never returned
subsequently. Thus, although the codetta is not needed to bring tonal closure to the minuet
(the PAC at m. 32 does that), it provides the valuable service of ensuring that every thematic
element of the opening phrase recurs somewhere else in the minuet.
Many of the concepts outlined in this chapter may be entirely new, as introductory
theory courses and textbooks necessarily focus largely, if not entirely, on chord construction
and usage. Therefore, it might be interesting to know that all of the concepts discussed here
were written about—using different terminology—in the eighteenth century. Material that
is especially similar to our discussion can be found in the composition handbooks authored
by Heinrich Christoph Koch between 1782 and 1793. While the music of composers such
as Haydn, Mozart, and Beethoven provides the best evidence of the utility of intensive study
of phrase design and hypermeter, there is historical justification to do so as well. In this
chapter, we have restricted ourselves almost entirely to the music of these three composers
because phrase design is more crisply articulated in their music than in the music of earlier
composers (especially Bach) and later ones. In Part II of this book, we will analyze phrase
design and hypermeter across a broader range of composers, especially later ones.
97 Phrase Design

SUGGESTED ASSIGNMENTS
In all of the following, you should identify where phrases begin and end and what types of
cadences occur. Then, consider if the phrases can be described in terms of sentence and/or
period structures. Finally, look for phrase elision and phrase expansion, and consider their
implications for hypermeter.

1. Mozart, Piano Sonata, K. 279, III, mm. 1–10

2. Mozart, Piano Sonata, K. 576, I, mm. 1–8

98 Principles of Harmony and Phrase Design

3. Mozart, Piano Sonata, K. 283, I, mm. 1–16

4. Beethoven, Piano Sonata, op. 31, no. 3, II, mm. 1–9

99 Phrase Design

5. Mozart, Piano Sonata, K. 457, I, mm. 23–63

100 Principles of Harmony and Phrase Design
101 Phrase Design

6. Mozart, Piano Trio, K. 502, I, mm. 1–12

102 Principles of Harmony and Phrase Design

7. Haydn, String Quartet, op. 33, no. 2, II, mm. 1–26

103 Phrase Design

8. Mozart, Piano Sonata, K. 311, I, mm. 1–16

104 Principles of Harmony and Phrase Design
4 Further Expansion of the
Harmonic Palette

In Chapter 2 we explored extension of the diatonic realm by means of secondary dominant

and diminished seventh chords and by modulation to what are generally referred to as “closely
related” keys, namely, those keys whose tonic chords are stable triads within the original
key. Here we will consider additional means of expanding the harmonic realm: 1) modal
mixture; 2) chromatic chords; and 3) modulation to non-diatonic keys.

Modal Mixture
Modal mixture involves the borrowing of elements from the parallel mode, for example,
the use of notes or chords from C minor within the context of C major, or the opposite,
thus altering the quality but not the function of the chord in question. Most often the term
refers to the former, that is, to borrowing from the parallel minor. In its simplest form, modal
mixture involves the borrowing of a single note, most often 3^ or 6^, from the minor mode.
The first of these is expressed frequently as a change in the tonic harmony. Of the many
examples in the literature, we offer two, the first from Schubert’s String Quartet, D. 887.
The quartet opens with a clear statement of I/i leading to V in the fourth measure. See
Example 4.1a. As shown by the Arabic numerals between the viola and cello parts, the fifth
measure, which has the character of a distant echo, extends the initial four-measure gesture
and is thus considered as an extension of the hypermetric unit. These five measures are then
repeated a fifth higher. Not shown here is the opening of the recapitulation, beginning in
m. 278, where Schubert reverses the mixture: i/I. At the close of the movement (Example
4.1b), Schubert makes final reference to this modal duality. Here the duple/quadruple
hypermeter is clearly expressed.
A slightly more complex example of mixture occurs in the opening of Brahms’s Third
Symphony, a piano reduction of which is provided in Example 4.2. Here the opening tonic
harmony (F) is initially embellished by a diminished seventh chord utilizing 3^ (A). The
major tonic chord is then restated in m. 3 coinciding with the initial statement of the motive
marked with a bracket,1 but in the fourth measure 3^ is once again introduced, this time as
a coloration of the tonic harmony. The process of borrowing from the minor mode continues
in the fifth measure with the introduction of 6^ (D), which is cancelled by the D in the
next measure as part of the embellishing diminished seventh chord, the same one we heard
in the second measure, which now functions as o43 of V leading to V6. Within these initial
measures we have two modal inflections, 3^/3^ (A/A) and 6^/6^ (D/D).
106 Principles of Harmony and Phrase Design

EXAMPLE 4.1a Schubert, String Quartet in G Major, D. 887, I, mm. 1–10

EXAMPLE 4.1b Schubert, String Quartet in G Major, D. 887, I, mm. 437–444

107 Further Expansion of the Harmonic Palette

EXAMPLE 4.2 Brahms, Symphony No. 3, op. 90, I, mm. 1–7

The use of 6^/6^ is clearly expressed in mm. 49–50 of our next example, Schubert’s
famous song, “Du bist die Ruh” (Example 4.3a). Schubert’s setting of the final verse of this
Rückert poem expands this modal inflection into the harmonic realm. The simultaneous
introduction of 3^ (G) and 6^ (C) in m. 55 creates the VI chord in first inversion, which
is subsequently prolonged by its dominant. The VI chord then leads to V in m. 58, which

in turn resolves to the tonic in m. 59, but instead of a stable tonic, it is I 7 leading on to the
subdominant at the climax of the phrase. The simplification of the score in Example 4.3b
shows the function of VI within the tonic scale-step. Corresponding to the prolongation
of the tonic harmony in mm. 54–59 is the unfolding of the melodic sixth B4–G5, indicated
by the slur in Example 4.3b. This sixth is answered by the octave skip A5–A4 over
subdominant harmony, after which the melodic line continues its descent to closure,
supported by dominant leading to tonic. At the deepest level of organization the harmonic
8–7
progression is I – IV – V6–5 – I supporting the descending fifth B4–E4, the notes of which
4–3
are circled in Example 4.3b. Modal mixture is employed here within the tonic scale-step,
which is transformed from stable harmony into the dominant seventh leading to IV.
The topic of modal mixture is extensive if we consider all possibilities, so we will limit
our review to mixture of elements from the parallel minor mode in a major key. The chords
involved are indicated in Figure 4.1. The first column lists the diatonic chords in the major
mode, and the middle column shows the chords (and their constituent scale degrees)
borrowed from the parallel minor. We have used boldface for III and VI, which involve
double mixture (two borrowed tones), since among the chords borrowed from the parallel
minor they serve most frequently as goals of modulation. The right column lists chords
resulting from compound mixture, that is, from two stages of mixture. The first stage involves
change from the diatonic chord (e.g., iii) to the one borrowed from the minor mode (III),
and the second involves modal change (III to iii).
108 Principles of Harmony and Phrase Design

EXAMPLE 4.3a Schubert, “Du bist die Ruh,” mm. 48–65

109 Further Expansion of the Harmonic Palette

EXAMPLE 4.3b Voice-leading simplification of Example 4.3a

Diatonic Simple Compound

Chords Mixture Mixture

I i

ii iio 6 (^
6)

iii III (^

3,^
7) iii
IV iv (^
6)

V v (^
7)

vi VI (^
6,^
3) vi
viio VII (^
7) vii

FIGURE 4.1 Modal Mixture (major key)

110 Principles of Harmony and Phrase Design

We will conclude this section with two passages from Schubert’s Moment musicale,
op. 94, no. 2. The first is the consequent phrase from the opening section of this work and
the first measure of the following section (Example 4.4). The first borrowing from the minor
mode occurs in m. 12, where the minor tonic chord appears as a neighboring chord to the
dominant. The introduction of C prepares the following diminished seventh chord of V,
which leads to the half cadence in m. 14. This is followed by [o7] IV and a confirmation
of the modulation to the subdominant. In the process, the chromatic passing tone B in the
bass (iv6 in the key of IV) anticipates the key of the B section, vii, enharmonically written
as F minor rather than G minor. This is an example of compound mixture: A major to
G minor, which is a modal alteration of G major. The subdominant, the goal of the initial
modulation, becomes V of the new key.

EXAMPLE 4.4 Schubert, Moment musicale, op. 94, no. 2, mm. 9–18

The first six measures of the second passage, reproduced as Example 4.5, are the same
as the opening of the previous example. Here, however, we have a parenthetical excursion
to III (C) rather than a modulation to IV.2 This passage is difficult to sort out, so we will
talk through the harmony. In part the difficulty results from deciding how best to indicate
what is taking place musically. For instance, the chord on the downbeat of m. 42 appears
111 Further Expansion of the Harmonic Palette

EXAMPLE 4.5 Schubert, Moment musicale, op. 94, no. 2, mm. 36–47

to be V43 of V in C, but it never really functions in that capacity. Instead it has been labeled
as an altered supertonic with raised third, which, following the chromatic voice exchange
indicated on the score, becomes IV in the local key.3 Note the exchange of F and F, 6^ and
6^ in the original key, within this measure. A final reference to this exchange occurs in the
alto voice in m. 46 as the phrase heads toward closure in A. Meanwhile the return to A
has been accomplished by the sudden return to the dominant, now in six-five position, which
had been abandoned in m. 41. Note the registral connection in the bass between the two
dominant chords.4 This return to the original key is made smoother by the overlap between
the digression to C and the re-establishment of the original key, which is not shown in the
harmonic analysis. But you should be aware that the progression V65 – I in mm. 44–45 is
also heard as [V65] VI in C (III).
112 Principles of Harmony and Phrase Design

Two Important Chromatic Chords: The Neapolitan Sixth

and the Augmented Sixth
The two chords under consideration here, the Neapolitan Sixth and the Augmented Sixth,
are derived from the minor mode, though both occur in the major as well as minor mode.
The first of these is the major triad built on the lowered second degree of the scale with
the third of the chord, scale degree 4, in the bass: F–A–D in C minor. It is used in place
of the subdominant or supertonic as preparation for the dominant. In its voice leading, the
lowered second scale degree descends to the leading tone (supported by dominant harmony),
passing almost always through scale degree 1. We offer two examples, the first from
the nineteenth song in Schubert’s famous cycle, Die schöne Müllerin (Example 4.6). Note
that 2^, the A4 in m. 8, passes through G4 to F4, harmonized by the dominant, in the next
measure. The harmonic progression in mm. 8–10 is II6 – V – i.

EXAMPLE 4.6 Schubert, “Der Müller und der Bach,” mm. 1–10

The voice leading in our second example, taken from Mozart’s D minor Fantasy, is a
bit more complex (Example 4.7). The Neapolitan sixth chord is stated forcefully in the second
half of m. 52 and subsequently elaborated by the ascending arpeggiation to B5. The lowered
second scale degree, E5, descends first to D5 as part of the arpeggiated [o7] V and then as
the fourth above the bass in the cadential 64 before progressing to C4 via octave transfer as
113 Further Expansion of the Harmonic Palette

part of the V7. The notes involved in the underlying melodic progression E5–D5–C4 are
circled on the score.
The second of our two chords being considered here is the augmented sixth chord.
Actually we are talking about three slightly different versions of the same basic chord. These
three, known as the Italian, French, and German augmented sixth chords, are shown in
Example 4.8 in the key of A minor. The basic chord, the Italian augmented sixth, is derived
from iv6 with the sixth raised, creating the interval of an augmented sixth with the bass (6^
in the minor mode or 6^ in the major mode). The interval of an augmented sixth expands
in contrary motion to the octave of V. The French version of this basic chord adds the
fourth above the bass, and the German version adds the fifth, normally requiring the
intervening six-four over the dominant to avoid parallel fifths. As shown below the system,
two of these chords, the Italian and the German, are derived from the subdominant, while
the root of the French is ii. Technically this is correct, though in fact they are all slightly
different versions (with slightly different sounds) of the same basic harmony. For this reason
we will not designate them by Roman numeral in this text, but by the labels It. 6, Fr. 6,
and Ger. 6.

EXAMPLE 4.7 Mozart, Fantasy in D Minor, K. 397, mm. 45–54

114 Principles of Harmony and Phrase Design

EXAMPLE 4.8 Augmented Sixth Chords

Below we offer examples of each type with minimal commentary. In several of these
examples, the augmented sixth chord is preceded by a diatonic chord or secondary dominant
having the same function, namely, to lead to the dominant.

EXAMPLE 4.9 Italian Augmented Sixth Chords

(a) Beethoven, Piano Sonata in C Minor, op. 10, no. 1, III, mm. 1–4

(a) Beethoven, Piano Sonata in C Minor, op. 10, no. 1, III, mm. 1–4. A very clear example
of an Italian augmented sixth chord that is melodically embellished before its resolution.

(b) Beethoven, Piano Sonata in F Major, op. 78, II, mm. 1–4

(b) Beethoven, Piano Sonata in F Major, op. 78, II, mm. 1–4. This movement opens in
an unusual way with an augmented sixth chord resolving to V. Note the local
prolongation of the supertonic harmony in m. 3 by voice exchange between the outer
parts. The chord in the last eighth of that measure is the half-diminished seventh chord
of the following dominant.5
115 Further Expansion of the Harmonic Palette

EXAMPLE 4.10 French Augmented Sixth Chords

(a) Schubert, “Der Wegweiser,” mm. 1–5

(a) Schubert, “Der Wegweiser,” mm. 1–5. In this instance the French augmented sixth
chord is introduced by o42. Note the elaborated resolution of the 4–3 suspension in
m. 4.

(b) Beethoven, Piano Trio in B Major, op. 97, III, mm. 1–8

(b) Beethoven, Piano Trio in B Major, op. 97, III, mm. 1–8. Here the opening tonic
harmony is prolonged by a voice exchange between the outer parts, which is subse-
quently reversed leading to ii6 – V, the latter introduced by its half-diminished seventh
chord. The second half of the phrase opens with I6, which is the pivot for the
modulation to V. This IV6 chord in the new key leads to the French sixth on the third
beat of m. 5, which strengthens the motion to the dominant and establishment of the
new key. The following cadential 64 is not a requirement of the voice leading, but of a
metric consideration—to create a balance with the initial four-measure unit.
116 Principles of Harmony and Phrase Design

EXAMPLE 4.11 German Augmented Sixth Chords

(a) Beethoven, Piano Sonata in C Minor, op. 27, no. 2, III, mm. 1–9

(a) Beethoven, Piano Sonata in C Minor, op. 27, no. 2, III, mm. 1–9. An example of the
German sixth introduced by a descending chromatic bass line supporting the progression
I – V6 – [V42] – iv6 – Ger. 6 – V. Here one can clearly see and hear the derivation of
the augmented sixth chord from iv6. Note that the German sixth chord does not lead
to an intervening 64, a choice made possible by the omission of the fifth above the bass
just before the resolution to V.
117 Further Expansion of the Harmonic Palette

EXAMPLE 4.11 continued

(b) Beethoven, Piano Sonata in F Minor, op. 2, no. 1, I, mm. 140–152

(b) Beethoven, Piano Sonata in F Minor, op. 2, no. 1, I, mm. 140–152. Here the Ger. 6
resolves in textbook fashion through a cadential 64. The latter part of the phrase is an
extended sequence by descending fifth that alternates 65 chords and triads and leads to a
perfect authentic cadence.

(c) Beethoven, Piano Sonata in F Minor, op. 57, II, mm. 1–8

(c) Beethoven, Piano Sonata in F Minor, op. 57, II, mm. 1–8. As indicated by the analysis
below the music, the I – IV – I progression that opens both the forephrase and afterphrase
prolongs I. The subdominant is a neighboring chord. The German augmented sixth chord
introduced on the last eighth of m. 6 looks odd because Beethoven has written the F as
E.6 As a result the following progression to the dominant without an intervening 64 may
look fine, but in fact this does not correct the sounding parallel fifths.
118 Principles of Harmony and Phrase Design

Analysis (Beethoven)
A very clear example of modal mixture—specifically of the interaction of 6^/6^ as well as
3^/3^—occurs in the opening movement of Beethoven’s Lebewohl [Farewell] Sonata, op. 81a.
We will examine its first twenty-one measures in some detail. The score with analytic
additions is provided in Example 4.12, and a representation of voice leading is provided in
Example 4.13.
When listening to this passage, one is struck immediately by the avoidance of the tonic
harmony, except in passing, until m. 21. The first statement of the Lebewohl motive,
G4–F4–E4, is harmonized by vi, and the repeat of this idea in mm. 7–8 by VI.7 This
establishes a polarity between C and C that becomes a primary motivic component of this
movement. Notice how Beethoven avoids tonic on the downbeat of m. 7 by harmonizing
G4 with a diminished seventh chord, which forces this important point of articulation to
become passing within a dominant seventh chord that finds its resolution instead to VI.
Observe also that the Lebewohl motive descends from G4 and that the melodic goal in m.
21, where we get our first stable tonic harmony, is G5. Furthermore, the return to G4 in
m. 7 is through its upper neighbor A4, and the establishment of G5 in m. 21 is likewise
accomplished through its upper neighbor A5, first stated in mm. 15–17 harmonized by
iv6/IV6 and later as seventh of the dominant in m. 20. To summarize, Beethoven has
established G4/G5 as the focal pitch, and to highlight its importance we have notated it in
Example 4.13 (our representation of the structure of this passage) with a half note. The
primary motive of this passage is G4–F4–E4, the statements of which are highlighted in
Example 4.13 by brackets. Especially important is the interaction between C, expressed
initially as the bass of vi and later as the bass of IV6 (m. 16), and C, expressed as VI in mm.
8–9 and later as the bass of iv6 (m. 15). As noted above, the interaction of C/C becomes
a primary motivic idea later in the movement. Finally, from the perspective of voice leading,
A, the upper neighbor, is important as the means by which G is established as the focal
pitch associated with tonic harmony.
Let’s now look at this passage in even greater detail. We hear the harmonization of the
opening statement of the Lebewohl motive as I – V – vi and the following material in mm.
3–4 in relation to vi (C minor). The local harmonic progression is shown between the staves
in Example 4.12. This material forms a basic four-measure phrase, and Beethoven could
have returned directly to the restatement of the Lebewohl motive via a dominant seventh
chord on the last eighth of m. 4. Instead he repeats m. 4 an octave higher with the added
change of B5 to B5 on the last eighth in anticipation of the change of harmony on the
next downbeat. This addition to the basic phrase prepares the return to that register and
eventual establishment of G5 in m. 21. As shown in the box at the end of the top system
of Example 4.13, the upper neighbor note A4 that returns to G4 is approached via a
descending chromatic third from C5, which is temporarily broken by the repetition of the
initial part of this motion in the upper register.
The progression in C (VI) beginning in m. 8 is a bit more complex. First VI is prolonged
by its V43 until E5 is established as a tenth above the bass note C on the third eighth note
of m. 9. This is the initial point of a sequential descending progression by step (based on a
descending fifth pattern) connecting VI and iv. The underlying motion in parallel tenths is
shown by the curved lines on the score (Example 4.12), and the underlying voice leading
is shown in Example 4.13. This is followed immediately by a passing i6 chord leading to
119 Further Expansion of the Harmonic Palette

EXAMPLE 4.12 Beethoven, Piano Sonata in E Major, op. 81a, I, mm. 1–21
120 Principles of Harmony and Phrase Design

EXAMPLE 4.13 Voice-leading representation of Example 4.12

V43 on the downbeat of m. 12. Considered in the larger context, this dominant does not
find its resolution until m. 21. But first Beethoven must progress from the inner voice tone
D5 up to the seventh of the chord, A5. He does this, but redirects the harmony to provide
temporary stable support for the A5, first by iv6, then IV6, allowing him to make reference
once again to the important interchange between C and C. Once this is accomplished, the
top voice descends via a series of parallel 63 chords back to V43 and then ascends once again
to A5. This time A5 is harmonized as seventh of the dominant, resolving to G5/I in the
next measure.

Modulation
In this section we will examine modulations to keys borrowed from the parallel minor,
notably III and VI, and two examples in the major mode of modulation to iii/III. We will
then examine the role of the German augmented sixth chord in enharmonic modulation
by half-step, a favorite device of nineteenth-century composers, as well as the use of chains
of descending thirds. These are a sampling of the range of chromatic and enharmonic
modulations composers devised.
Modulation to keys borrowed from the minor mode are easily achieved by changing
first to the parallel mode, for example, from A major to A minor, and then modulating to
a key associated with the minor key. There are countless examples of this procedure in the
Classical literature, one of which occurs in the minuet of Haydn’s Symphony No. 103
(Example 4.14).
121 Further Expansion of the Harmonic Palette

EXAMPLE 4.14 Haydn, Symphony No. 103, III, mm. 1–27

122 Principles of Harmony and Phrase Design

EXAMPLE 4.14 continued

123 Further Expansion of the Harmonic Palette

EXAMPLE 4.14 continued

124 Principles of Harmony and Phrase Design

EXAMPLE 4.14 continued

125 Further Expansion of the Harmonic Palette

The first part of this movement consists of two contrasting four-measure phrases, the
first ending on the tonic and the second ending with a half cadence. The second phrase is
extended for two measures by echoes in two registers of the falling fifth from the phrase’s
cadence. Immediately following the repeat, the mode changes to minor, and in the next
eight-measure phrase Haydn modulates to III of E minor (III of E major). As shown below
the score, the pivot in the modulation is VI/IV. The music then returns to E minor for an
imitative passage based on the minuet’s opening motive, which leads to a Ger. 6 resolving
to V (mm. 24–25), which is then repeated. (Note: The German sixth does not lead to an
intervening 64, since the fifth above the bass, stated by the clarinets,8 is no longer present as
the chord resolves to V.) The dominant is subsequently extended for another four measures
before the return to the opening material.
Two very different passages involving modulation to VI can be found in Beethoven’s
Piano Sonata, op. 7. The first occurs in mm. 43–86 of the third movement (Example 4.15).
This section opens with an eight-measure antecedent phrase that ends with a half cadence.
The consequent phrase, which is greatly expanded, switches to the minor mode, progressing
deceptively to VI (VI in E minor) in the fourth measure (m. 54). As in the Haydn example,
the motion to VI is prepared by the change to the parallel minor. What follows is an extended
excursion—actually a parenthetical insertion—into the realm of VI (mm. 54–69).
Confirmation of C as the new tonic is accomplished by the progression beginning in m.
58, which is repeated and expanded immediately following, leading to the It. 6 in m. 68.
The derivation of this chord from the minor subdominant (m. 66) is emphasized by the
chromatic voice exchange between the two chords: C5/A3 to A4/C4. Arrival at the
augmented sixth chord returns us to the realm of the tonic, thus ending the excursion to
VI. Before leaving this passage you should examine carefully the harmony and voice leading,
an analysis of which is provided below the music.
Initially one might expect the repetition of B in mm. 70–71 to represent a cadential 64
following the augmented sixth chord, but it soon becomes apparent that this is really a tonic
harmony, which is extended, first as a 64 and then later as a root-position chord until the
climax of this passage is reached in m. 79 following the long crescendo. Here in m. 79, D is
added to the chord, transforming tonic into [V65] IV. This whole passage beginning in m.
70 is a complex extension of a single passing chord, the tonic, connecting VI and IV. The
means by which this extension is accomplished is shown as the exchange of parts—
represented by two different types of brackets in Example 4.15—between the right-hand
and left-hand parts, first over a B pedal and then over E. In the latter case these exchanges
involve chromatic alteration in the embellishing pitches. So much time and energy has elapsed
in connecting VI and its extension to IV that Beethoven does not allow the subdominant
to complete its motion through the dominant to a perfect authentic cadence immediately,
but instead extends this passage further by twice avoiding closure to the tonic by means of
deceptive progressions (mm. 82 and 84).
If we take a final survey of this passage, we see an eight-measure antecedent and a greatly
expanded consequent that encompasses all of mm. 51–86! This phrase is extended first by
the excursion to VI, second by the extension of the passing tonic harmony leading to IV,
and finally by twice avoiding closure to the tonic through deceptive progressions. The
function of VI in this extraordinary passage is to participate as middle member of the
descending arpeggiation from tonic to subdominant, I – VI – IV, which then leads through
V to I.
126 Principles of Harmony and Phrase Design

EXAMPLE 4.15 Beethoven, Piano Sonata in E Major, op. 7, III, mm. 43–86
127 Further Expansion of the Harmonic Palette

A more direct approach to VI occurs in the second movement of this sonata. The last
few measures of its opening section, ending with a perfect authentic cadence in the tonic
(C major), is provided at the beginning of Example 4.16. It is followed immediately by the
introduction of VI. A change of key of this type, where the tonics of the old and new keys
have a single note in common, in this case C, and where the change is direct, is referred to
as a common-tone modulation. So, here we are suddenly in A major, which is then
confirmed as the new key by the progression in mm. 25–28. This is followed by a
progression leading to iv (F minor). In the process of confirming F as the new tonic, we
encounter an interesting chord with B as the lowest note on the second eighth note of m.
31. If this B had occurred in an inner voice above D, the chord would be a French
augmented sixth in “normal” position, that is, built above scale degree 6 of F minor. Instead
it is inverted, so that the interval of an augmented sixth has become a diminished third. This
change does not alter its function. It leads to a perfect authentic cadence confirming the
modulation to F minor.
So far the larger progression is similar to what we discovered in this sonata’s third
movement: I – VI – iv, except here the subdominant is minor, not major, and it does not
progress directly to V. Instead the progression by descending thirds continues to D, where
we have a varied restatement of the earlier four-measure idea in A. That is, the progression
from F to D imitates the earlier progression from C to A. In relation to F minor, this is a
progression to VI, but in relation to our original tonic C, D is II. Overall, then, Beethoven
has extended the descending third pattern to the chord on II: I – VI – iv – II, and we
might expect him to treat this II as a Neapolitan. Instead, Beethoven begins to repeat the
four-measure idea from mm. 25–28 a fourth higher. The change in pattern comes in the
third measure, where the harmony continues sequentially to [o65] iii6 instead of V7 – I in
D. This F minor chord (iii6) becomes the pivot (iv6), which is subsequently altered
chromatically to become the German augmented sixth chord leading to the dominant in
the original key. This progression leading to V of the original key is abrupt, no doubt one
reason why an extended retransition follows this tonal arrival, giving the listener time to
readjust his/her orientation.
The next two examples involve modulation to the mediant in the major mode. The
first of these, Schubert’s “Nachtgesang,” is a diatonic modulation from E major to G minor
(Example 4.17). What places discussion of this song here is the pivot: IV6 in E, which is
emphasized by its dominant, is then treated as II6 in G minor (iii). The second example,
taken from the slow movement of Beethoven’s Fifth Symphony, is the famous passage leading
from A major (I) to C major (III). See Example 4.18. The passage begins with an ascending
figure of a third leading to C5, then to E5. The third statement reaches up to G5 as the
bass progresses chromatically from A to A. Our expectation is that this diminished seventh
chord will lead to ii (with G5 resolving to F5), but in the seventh measure of this phrase
the bass returns once again to A in the lower octave. In terms of sound, this chord is I 7


(V7 of IV), and we might expect it to lead to the subdominant, but Beethoven has rewritten
the G as F, changing its spelling to the German sixth in C. The chord does not lead us to
8–7
IV, but resolves to the V6–5 to I in C (III) for the triumphant statement of this theme, now
4–3
complete with brass.9
128 Principles of Harmony and Phrase Design

EXAMPLE 4.16 Beethoven, Piano Sonata in E Major, op. 7, II, mm. 22–37
129 Further Expansion of the Harmonic Palette

EXAMPLE 4.17 Schubert, “Nachtgesang,” mm. 1–8

EXAMPLE 4.18 Beethoven, Symphony No. 5, II, mm. 23–31

130 Principles of Harmony and Phrase Design

This last example highlights an important feature of the German augmented sixth chord,
namely that it has the same sound as a dominant seventh chord and thus can be used as a
source of enharmonic modulation. This interchange between pre-dominant and dominant
function became part of the harmonic language of the nineteenth century and thus deserves
our careful attention. Example 4.19 shows three scenarios, the last a variant of the second
one. At (a) the second chord is a Ger. 6 in A minor, a pre-dominant chord that would
normally lead to V in that key. Instead the D has been treated as an E, allowing the chord
to function as V7 in B major, a half-step above the original key. At (b) the major tonic
chord with added seventh, potentially V7 of IV, is treated as a Ger. 6 leading to V – I in
C (III in the original key). This is analogous to the situation encountered in the Beethoven
example above. The situation at (c) is a less common variant of (b). Here V7 in A minor is
treated as a Ger. 6 in A, a half-step lower. Examples of these three scenarios are provided
below.
Example 4.20, taken from Beethoven’s Rondo a capriccio, op. 129, provides a very clear
example of the scenario given at (a) above. Here Beethoven carefully prepares the
introduction of the German sixth chord in m. 165, first by changing to the parallel minor
mode (mm. 162–163) and then by introducing the C as part of the [o65] V in m. 164. The
initial appearance of the augmented sixth chord resolves as expected to the dominant, but
in its next appearance (m. 168), the C is rewritten as D, and this same chord (now respelled)
functions as a dominant seventh leading us to A, a half-step above the point of origin.

An example of the scenario given at (b) above, where I 7 functions both as a German
7
sixth and as V of IV, occurs in the first movement of Beethoven’s Horn Sonata, op. 17
(Example 4.21). The first appearance of the chord in question occurs in mm. 136–137, where
the chord is spelled as an augmented sixth chord resolving to the dominant in A minor.
This modulation from F major (I) to A minor (iii) is then repeated. Then, in m. 146, we
hear this same chord again (ff), this time spelled with E rather than D, now functioning
as [V7] IV.
Finally, an example of the third scenario occurs at the end of the A fugal section from
the final movement of Beethoven’s Piano Sonata, op. 110 (Example 4.22). The fugue ends
with the arpeggiation of V7, the seventh of which (D) is subsequently treated as if it were
C, the augmented sixth above E. This D/C leads up to D5 as part of the cadential 64 in
G minor. The following treatment of this 64 is unusual. Instead of resolving to 53 , the 64 leads
directly though arpeggiation to the new tonic in root position.
131 Further Expansion of the Harmonic Palette

EXAMPLE 4.19 Enharmonic Modulations with German Augmented Sixths

EXAMPLE 4.20 Beethoven, Rondo a capriccio, op. 129, mm. 158–171

132 Principles of Harmony and Phrase Design

EXAMPLE 4.21 Beethoven, Horn Sonata in F Major, op. 17, I, mm. 134–149
133 Further Expansion of the Harmonic Palette

EXAMPLE 4.22 Beethoven, Piano Sonata in A Major, op. 110, III, mm. 111–115

Schubert’s Moment musicale, op. 94, no. 6, provides an interesting study of the augmented
sixth chord. The excerpt provided in Example 4.23 features the return of the opening A
major material (designated as a′ on Example 4.23 at the upbeat to m. 54) as well as the
preceding retransitional passage (mm. 40–53). The F major chord at the beginning of the
excerpt is VI, and the return to A is accomplished via the It. 6 leading to V,10 which is
extended to the end of m. 45. The tonic is then decorated by a 5–6 motion introducing
F5, which is suspended over the following dominant with G3 in the bass. This F5 then
descends chromatically to D5, the seventh of the dominant, in preparation for the return.
The following eight-measure phrase is a repetition of the antecedent phrase from the opening
of the piece. Examine the harmony of this phrase and the provided analysis of it carefully.
The chord in m. 54 is not iv7, as it may at first appear, but rather the C5 is a suspension
displacing B4 as root of the following ii65. Likewise the chord in m. 56 is not vi64, but rather
F5 is a suspension prepared by the preceding half-diminished seventh chord momentarily
displacing the E5 as fifth of I6. In the latter half of the phrase the controlling harmony is
the dominant. As shown by the slurs added below the harmonic analysis of this passage, the
chords on the third beats of mm. 58 and 59 are passing.
The consequent phrase, beginning with the upbeat to m. 62, opens with notes borrowed
from the parallel minor, F (6^) and also the chromatic passing tone C (3^). This gesture is
then repeated. What follows is an E major-minor 42 chord, enharmonically the notes of the
German augmented sixth chord (F–A–C–D), but with D in the bass. It is not going to
function here as a German sixth, but rather as V42 of II, rewritten enharmonically by Schubert
as an A major rather than a B major chord. This passage looks complicated, but, in fact,

the sound is as follows: [V42] – II6 . . . V 6–5
4–3 – i. Here Schubert has realized the potential
of this versatile augmented sixth chord, which earlier had resolved to the dominant in A,
to function now as V7 of the Neapolitan (II).
As a final example in this section on modulation we will examine a passage taken from
the second movement of Schubert’s “Unfinished” Symphony in B minor (D. 759). The
main key of the movement is E major, but the section we have selected is in C minor (vi).
See Example 4.24, a reduction of the orchestral score of mm. 66–83. A feature of the
accompaniment in the opening measures of this passage is the decoration of G4 by its upper
134 Principles of Harmony and Phrase Design

EXAMPLE 4.23 Schubert, Moment musicale, op. 94, no. 6, mm. 40–77
135 Further Expansion of the Harmonic Palette

neighbor A4 as the clarinet melody ascends to A5. In m. 72, with a prominent change of
dynamic (), the harmony changes to V65 of II, to which it resolves in the next measure
while the clarinet sustains A5. In the next few measures, as the clarinet begins its descent,
the harmony moves toward F major, III of D minor (ii) or III of D major (II), but at the
last moment the progression pulls back to a D minor chord (ii) via a deceptive progression.
As shown below the lower staff, the controlling harmony in these measures is II/ii, and the
suggestion of F major is embellishing. (However, if you examine the entire movement you
will discover that this brief suggestion of F major foreshadows the dramatic emphasis on
F major, II in E major, in mm. 252–253.) The clarinet’s return to the passage’s opening
third (C5–E5) in m. 77 is harmonized by V7 of II (D). Here, however, the seventh of the
chord, G4, is changed to F4, transforming its spelling as well as its function into a German
augmented sixth, which leads us back to C minor, which at the last minute changes to the
parallel major. Here, in this passage, we have a very clear example of the dual function of
the German augmented sixth chord as dominant of II and as leading to the dominant in
the local tonic.

EXAMPLE 4.24 Schubert, Symphony in B Minor (“Unfinished”), D. 759, II, mm. 66–83
136 Principles of Harmony and Phrase Design

SUGGESTED ASSIGNMENTS

Chromatic Harmony

1. Schumann, Waltz, op. 124, no. 4

2. Mozart, Piano Sonata, K. 284, I, mm. 11–17

137 Further Expansion of the Harmonic Palette

3. Mozart, Piano Quartet, K. 478, III, mm. 330–351

138 Principles of Harmony and Phrase Design
139 Further Expansion of the Harmonic Palette

4. Beethoven, Piano Sonata, op. 57, I, mm. 1–16

140 Principles of Harmony and Phrase Design

5. Beethoven, Piano Sonata, op. 13, III, mm. 193–210

141 Further Expansion of the Harmonic Palette

Chromatic Modulation
1. Schubert, “Unfinished” Symphony, D. 759, II, mm. 96–111. How does Schubert modulate from C minor
to D major?
142 Principles of Harmony and Phrase Design

2. Schubert, Piano Sonata, D. 845, I, mm. 1–26. Determine the harmonic progression, paying particular
attention to mm. 20–23.
143 Further Expansion of the Harmonic Palette

3. Haydn, String Quartet, op. 76, no. 6, Fantasia, mm. 9–16 and 31–39. This movement is a study in
modulation. The excerpt at (a) modulates from B major (I) to C minor (ii). How does Haydn accomplish
this? What is the chord on the third beat of m. 14? The excerpt at (b) returns to B major from B major.
Note the mixture of sharps and flats in mm. 35–36, where you must use your ear, not your eyes, to determine
the progression.
144 Principles of Harmony and Phrase Design

4. Beethoven, Piano Sonata, op. 90, II, mm. 114–126. Determine the harmonic progression, paying particular
attention to mm. 120–121. What is the function of the 64 chord on the downbeat of m. 121? What is the
relationship between the chords on the second halves of mm. 120 and 121? (Hint: To answer this question
you will have to respell the chord on the second half of m. 121.)

5. Chopin, Nocturne, op. 27, no. 1, mm. 41–52. Determine the harmonic progression, paying particular
attention to mm. 45–52. What are the harmonies in m. 45 and on the last beat of m. 48?
145 Further Expansion of the Harmonic Palette

6. Schubert, Piano Sonata, D. 960, I, mm. 1–48. An important feature of this movement is the emphasis given
to VI, first hinted at in m. 8 in the bass. Provide a detailed harmonic analysis of each section; then consider
the large-scale harmonic plan. An interesting issue to consider is the role of the section beginning in m.
20 in relation to the eventual goal of the passage given (m. 48). Stated somewhat differently, what is the
role of the return to B major prior to the modulation to F minor?
146 Principles of Harmony and Phrase Design
147 Further Expansion of the Harmonic Palette
PART II

Analysis
5 Binary and Rounded Binary
Forms

Beginning with this chapter, we shift exclusively to the analysis of complete pieces. This
chapter focuses on movements in binary, or two-part, form. There are two types of binary
forms: those without a return of the opening music in the tonic key in the second part
(binary or simple binary) and those with such a return (rounded binary). The former
type of binary form is commonly found in dance-inspired pieces from the early eighteenth
century. It is the usual form of movements in, for example, Bach’s cello and keyboard suites
and his violin partitas. Although examples of binary form can be found throughout the
eighteenth century, rounded binary became the predominant type by the time of Haydn,
Mozart, and Beethoven.1 Rounded binary form is the standard form for Classical minuets
(and their trios) and, in the nineteenth century, for scherzos (and their trios). We will briefly
review two examples of binary forms from Bach’s keyboard suites studied in Chapter 2 and
one example of rounded binary form from a Haydn string quartet studied in Chapter 3. We
will devote most of this chapter to the handling of rounded binary form in a minuet–trio
movement from a symphony by Haydn and in two scherzo–trio movements from piano
sonatas by Beethoven and Brahms.

Binary Form
Binary forms consist of two parts, which are nearly always repeated (with repeat signs rather
than varied repetition).2 Tonally, we can make a basic distinction between binary forms that
close on the initial tonic harmony at the end of the first part (sectional binary) and those
that do not (continuous binary). Continuous binary is somewhat more common than
sectional binary.3 In the major mode, the first part of a continuous binary piece invariably
moves to the dominant. In the minor mode, the modulation is usually to the relative major,
but the minor dominant is occasionally found instead; either way, the modulation is
confirmed with a perfect authentic cadence. The tonal structure of the start of the second
part is much more variable. Most often, there is a tonicization (with cadential confirmation)
of a closely related key, usually the subdominant or the submediant. This portion of the
form can instead feature very brief digressions to one or more closely related keys without
cadential confirmation, or it can even be in the tonic key (perhaps with emphasis on
dominant, rather than tonic, harmony). Regardless of the tonal structure of the start of the
second part, at some point the tonic key returns and is cadentially confirmed. Due to the
relative brevity of most binary-form pieces, it is not infrequent for there to be only one
152 Analysis

Continuous binary (major mode) ||: A :||: B :||

I V      X I
(PAC) (PAC)

Continuous binary (minor mode) ||: A :||: B :||

i III*    X i
(PAC) (PAC)
* sometimes v

Sectional binary (major mode) ||: A :||: B :||

I I      X I
(PAC) (PAC)

FIGURE 5.1 Basic Models of Binary Form

cadence in the tonic key during the second part—the perfect authentic cadence at the very
end. Figure 5.1 summarizes the basic tonal plans for continuous and sectional binary forms.
In Chapter 2 we studied two binary-form pieces by Bach, one in G major and one in
C minor. Both involved modulation in the first part: G major to D major and C minor to
E major respectively. We will briefly review each of these pieces in light of our general
comments on binary form. The piece in G major was the gavotte from the French Suite
No. 5 (Example 2.20). The first part consists of two phrases that combine into a parallel
progressive period ending with a PAC in the dominant in m. 8. The second part consists
of four phrases that combine into two periods. The first period begins back in the tonic key
but ends with a PAC in the submediant (m. 16), a key often emphasized at this point in
binary form. Note that the start of the second part begins with a melodic inversion of the
gavotte’s initial three notes. In Bach’s binary forms, there is usually a direct thematic
connection between the start of the first part and the start of the second part; sometimes,
there is melodic inversion involved, as here, but in other cases the initial melodic material
reappears in its original version but transposed to a new key. Observe also the shifting of
the main melodic material into the left-hand part in mm. 12–14; this, too, is commonly
found in the second parts of Bach’s binary forms for keyboard. The gavotte’s concluding
period modulates back to G major, ending with a PAC. The start of the last phrase (mm.
20–21) bears a strong melodic similarity to the gavotte’s opening, but notice that the harmonic
support here is dominant, not tonic. There is some recall of the opening material after the
return to the tonic key—more exposed recall than there is in many binary pieces—but it
does not provide the feeling of “beginning again” that defines rounded binary form.
The binary piece in C minor from Chapter 2 was the minuet from the French Suite
No. 2 (Example 2.26). Again, the first part consists of an eight-measure progressive period,
153 Binary and Rounded Binary Forms

although its two phrases are thematically contrasting rather than parallel (and the modulation
is to the relative major, not the dominant). Unlike the gavotte, the start of the second part
remains in the key from the end of the first part. Notice the direct thematic correspondence
between the first measure of the first and second parts; these measures are related by exact
transposition (rather than melodic inversion as in the gavotte). Like the gavotte, the first
two phrases of the second part, lead to a PAC in a new key, this time the subdominant.
The minuet’s second part is eight measures longer than the gavotte’s, principally as a result
of the descending fifth sequence in mm. 17–24 that leads back to the tonic key. Very
occasionally, the two parts of a binary form are of equal length, but nearly always the second
part is longer, sometimes much longer. The minuet’s final phrase bears some similarity to
the first part (compare especially m. 29 with m. 5), but again there is no moment of
“beginning again” as one finds in rounded binary form. As in the gavotte, the only PAC
in the tonic key occurs in the final measure of the minuet.
There is one thematic relationship often found in Bach’s binary forms that doesn’t happen
to occur in either the gavotte or the minuet. In many continuous binary pieces there is a
direct relationship between the final two (or so) measures of the first part and the
corresponding ones in the second part. Giving these two cadences the same melodic content
(but transposed) contributes a further sense of unity to the piece. Some authors refer to binary
forms with this “cadential rhyme” as balanced binary.
We examined the motivic design of these two pieces, especially the minuet, in some
detail in Chapter 2. It was clear that both pieces redeploy their initial materials in multiple
guises, sometimes through repetition, transposition, or simple embellishment, but other times
through enlargements that create hidden motivic parallelisms. Binary forms are predicated
upon thematic unity much more than thematic contrast (compared to a form like ternary
or rondo). A central component of any analysis of a binary piece must be a thorough
assessment of its handling of motivic materials at various levels of design.

Rounded Binary Form

In rounded binary form, there is a return of the initial thematic material in the tonic key in
the middle of the second part. Usually, this return is preceded by an arrival on the dominant
of the home key—in other words by a half cadence. Occasionally, this arrival on the dominant
is extended for a few measures as a brief retransition before the thematic return. The amount
of the initial thematic material that returns varies. Since there is no modulation in the first
part of sectional rounded binary, it is possible for the entire first part to return without—
or with minimal—modification. In continuous rounded binary, the modulation in the first
part necessitates recomposition, at least from the point corresponding to the initial
modulation, though often the exact thematic restatement breaks off even earlier. The
important point to remember is that it is the return of the opening measures of the first part
in the tonic key that creates rounded binary form; the second part of a simple binary form
may incorporate a tonic return of another portion of the first part or a non-tonic return of
the opening measures. Figure 5.2 summarizes the basic outlines of continuous and sectional
rounded binary forms.
The diagrams in Figure 5.2 might give the impression that rounded binary form is, in
some way, a three-part form due to the A B A′ thematic labels. Keep in mind, first of all,
154 Analysis

Continuous rounded binary

(major mode) ||: A :||: B A′ :||
I V (X) V I
(PAC) (HC) (PAC)
|___________________| |________|

Continuous rounded binary

(minor mode) ||: A :||: B A′ :||
i III X V i
(PAC) (HC) (PAC)
|___________________| |________|

Sectional rounded binary

(major mode) ||: A :||: B A′ :||
I I X V I
(PAC) (HC) (PAC)
|___________________| |________|

FIGURE 5.2 Basic Models of Rounded Binary Form

that in the eighteenth century the form nearly always consists of two repeated sections; only
in the nineteenth century do composers sometimes omit the repeat of the second part, but
by then the usage of rounded binary form for certain types of movement had become an
established convention. Second, although it is traditional to use the letter B to designate the
start of the second part in rounded binary form (just as it is traditional to use B for the entire
second part of binary form), there is no strong thematic contrast here. The outset of the
second part of rounded binary form nearly always elaborates upon the thematic material
from A. When the thematic connection between B and A is less clear, it is usually because
B is very short. It happens not infrequently in major-mode continuous rounded binary that
B consists only of a few measures based entirely around the dominant, transforming it from
its own key area back into V of the tonic key. Thus, when B is less thematically dependent
on A, it is generally very evident that it does not constitute an independent formal unit.
After we examine ternary form in Chapter 8, it will be perfectly clear why rounded binary
form is not a ternary form, despite the traditional usage of the labels A B A′.
To this point, we have not particularly distinguished between the typical tonal structures
of binary forms and those of rounded binary forms. This is completely appropriate for the
first part and the start of the second part of these forms. However, because of the return of
155 Binary and Rounded Binary Forms

the opening material in the tonic in the second part, rounded binary form projects a two-
part tonal structure. The first branch leads from the initial tonic to the dominant arrival,
midway through the second part, that prepares the thematic return. The tonal structure then
starts over with tonic harmony, which is subsequently confirmed with a perfect authentic
cadence at the end. These tonal divisions are shown by the brackets beneath the diagrams
in Figure 5.2; note that this two-part tonal structure does not align with the two-part formal
design. Occasionally, binary form has similarly clear rearticulation of tonic harmony midway
through the second part that balances a prior I – V motion, but in general binary form gives
the impression of a single tonal arch. Recall, for example, the i – III – iv – V – i motion
that spanned the C minor minuet. We will consider the interaction of tonal structure and
formal design more closely in our discussion of sonata form in Chapter 6, as this issue becomes
more significant in that larger form, which is an outgrowth of rounded binary form.
At the end of Chapter 3, we studied the minuet (but not the trio) from Haydn’s String
Quartet in E Major, op. 64, no. 6. The score is available in Example 3.21, and we provide
a voice-leading representation here as Example 5.1. As we would expect in a Classical minuet,
the form is rounded binary; the tonal type is the more common continuous structure with
modulation to the dominant during the first part. The first part consists of a parallel progressive
period, which is an exceedingly common design for this portion of a binary or rounded
binary form. The second part continues the dominant for a couple of measures before
launching a descending fifth sequence that returns to the tonic key. Note, however, that
the goal of this phrase is a half cadence in m. 20. Thus, even though there is a return to the
tonic key around mm. 16–17, there is no rearticulation of the tonic harmony at a high level
of structure. The E major harmony on the third beat of m. 18 is a passing harmony within
a broader span of predominant harmony (mm. 17–19). This brief, metrically unaccented
chord harmonizes the melodic passing tone G5 within the descending fifth C6–F5 that frames
the predominant harmonic function within the phrase. The local nature of the harmonic
motions within these measures is further projected by the dropping out of the cello; the
cello sounds B2 at the perfect authentic cadence in m. 12 and again in m. 15, and the true
bass register of the cello comes back only with the E3 at the thematic return (m. 21). In
longer pieces, the connection between the dominant harmony at the end of the first part
and the one that sets up the thematic return will seem more abstract. Even with greater
separation between these arrivals and more elaborate intervening material, however, this
deeper harmonic connection is still understood. In the Haydn, there is a clear corresponding
melodic connection; the F5 in the first violin can be heard extending across the PAC in m.
12 as well as the HC in m. 20. (And, in fact, one can trace a stepwise octave descent between
these two points in the first violin line—with a registral adjustment in m. 16 that will permit
the concluding F to sound in its original register.)
The thematic return during the second part was discussed at length in Chapter 3 due
to the reworking of phrase length and hypermeter. It is in the fourth measure of the thematic
return where modifications begin to occur, and tonal closure is provided with the PAC in
m. 32. The minuet concludes with a four-measure codetta, a brief passage consisting of
repeated cadential gestures that takes place after tonal closure has been attained. A codetta
is not uncommon at the end of a rounded binary form. (It is uncommon to find a codetta
at the end of a binary form.)
156 Analysis

EXAMPLE 5.1 Voice-leading representation of Haydn, String Quartet, op. 64, no. 6, III, mm. 1–32

With our survey of the basic principles of binary and rounded binary form complete,
we turn now to the analyses that are this chapter’s main focus. We will first examine the
minuet–trio movement from Haydn’s Symphony No. 101 in D Major, another Haydn
movement where the treatment of phrase design and hypermeter is especially notable. We
will then proceed to the scherzo–trio movements from Beethoven’s Piano Sonata in A
Major, op. 26, and Brahms’s Piano Sonata in F Minor, op. 5. Both of these depart somewhat
from the norms for the tonal structure of rounded binary form.

Haydn, Symphony No. 101 in D Major (“Clock”), III

Haydn’s Symphony No. 101 is a large-scale work written for the composer’s second trip to
London in 1794–95. An annotated score of its minuet appears in Example 5.2. The minuet’s
first part begins with a period (specifically, a parallel progressive period), but it spans twenty
measures and is followed by additional material confirming the key of the dominant. The
period’s eight-measure antecedent is in sentence design. In the repetition of the basic idea
(mm. 3–4) notice that the upper-voice pitches from mm. 1–2 occur in the bass line (bracketed
157 Binary and Rounded Binary Forms

on the score). This is not particularly salient, since throughout mm. 2–7 the bass simply
moves in parallel tenths with the melody, but Haydn develops this feature later. The
chromatic descent at the end of the phrase (marked by a dotted bracket) seems like an
innocuous way to descend to scale degree 2 at the half cadence, but it too provides significant
material for subsequent development.
The twelve-measure consequent is expanded internally. The modulation to the dominant
occurs during the repetition of the basic idea (mm. 11–12). Instead of peaking around A5,
as in the first phrase, the melodic motion continues up to E6. Since this is already the sixth
measure of the phrase, it is not hard to imagine why Haydn might have decided to expand
the phrase in order to descend convincingly for the melodic cadence. The expansion results
from a four-measure stasis on I6 of the new key (mm. 14–17), an expansion that is quite
apparent because it is based on repetition. The I6 alternates with its neighboring V42, except
in the fourth measure of the expansion where the I6 persists throughout the entire measure.
This permits the bass motion to D to fall on the subsequent downbeat, where it supports
ii6 as the cadential progression begins to take shape. Notice also that in the last measure of
the expansion the final melodic tone consonant with the I6 harmony is E5; in other words,
the expansion transfers E6 down to E5. From this point there is an embellished descent
through the fifth E5–A4 to melodic closure in m. 20; these pitches are circled in the first
violin part. We show two levels of hypermeter on the score. Most important is the
underlying hypermeter, which corresponds to the basic phrase, but we also note that the
expansion, given its length, could produce a momentary surface hypermeter of its own.
The remainder of the first part confirms the modulation to the dominant and in a sense
could be considered an expansion. It does, however, present a complete phrase, which is
repeated, and has a distinct melody and interesting dialogue of sforzandi between upper and
lower instruments. This is why we have labeled mm. 20–23 (and mm. 23–26) as a closing
phrase. The junctures between these phrases require comment. In our hypermetric analysis,
we show no reinterpretation at the start of the closing phrase, but we do show a
reinterpretation (4 = 1) at the start of its repetition. Yet, the two locations are very similar.
In both cases, we understand the presence of phrase overlap; the syncopated and accented
pitch that launches the closing phrase “should have” occurred on the downbeat—it has been
displaced by a beat. (Note that the clarinets, which double the melody in the closing phrase,
do enter on the downbeat of m. 20.) Recall from our discussion in Chapter 3 that phrase
overlap may or may not generate a hypermetric reinterpretation; sometimes overlap is
necessary to preserve periodic hypermeter. The latter is the case at m. 20, where the conse-
quent phrase’s concluding harmony spills over onto the downbeat of m. 20 rather than
entering during the eighth measure of the basic phrase. The closing theme is precisely four
measures long, which means that when its repetition begins in m. 23 the overlap does induce
a hypermetric reinterpretation. The repetition of the closing phrase is extended with a couple
of measures that feature rapid chord alternations. After the expanded consequent, there is a
feeling of acceleration throughout the remainder of the first part. The closing phrases are
short (four measures) and further energized by the syncopations and sforzandi, and this is
followed by the dizzying repetitions within the final extension (mm. 26–28). These rhythmic
features give the first part a powerful sweep.
158 Analysis

EXAMPLE 5.2 Haydn, Symphony No. 101 in D Major (“Clock”), III, mm. 1–80
159 Binary and Rounded Binary Forms

EXAMPLE 5.2 continued

160 Analysis

EXAMPLE 5.2 continued

161 Binary and Rounded Binary Forms

EXAMPLE 5.2 continued

162 Analysis

EXAMPLE 5.2 continued

163 Binary and Rounded Binary Forms

EXAMPLE 5.2 continued

164 Analysis

EXAMPLE 5.2 continued

165 Binary and Rounded Binary Forms

EXAMPLE 5.2 continued

166 Analysis

EXAMPLE 5.2 continued

167 Binary and Rounded Binary Forms

EXAMPLE 5.2 continued

168 Analysis

The second part consists of its usual two segments: a portion leading to the home
dominant (mm. 29–48) and the thematic return (mm. 49–80). The former includes an
extension of the home dominant for several measures preceding the thematic return; thus,
mm. 29–42 alone provide the underlying content of the B section. These fourteen measures
can be viewed as a contraction of sixteen. To understand this, we need to examine the
sequence that launches the second part of the minuet. The rapid descending fifth sequence
beginning in m. 35 is introduced by four-measure sequential units beginning in m. 29.
Example 5.3 recomposes this passage with an exact sequence of mm. 29–32 in the following
four measures. Comparing this recomposition against Haydn’s version, we observe that in
the third measure of the second unit the sequence breaks off, avoiding the expected arrival
on G major harmony. Instead a new “pseudo-sequence” begins; the harmonic basis of mm.
35–38 is descending fifths (B, E, A, D), but the melodic content is not exactly repeated
between mm. 35–36 and 37–38. There is enough of a sequence, though, to convey that
mm. 35–38 constitutes a four-measure unit; combined with its internal repetition and the
subsequent hastening of surface chord changes in mm. 39–40, mm. 35–42 give the impression
of a sentence design. This breaking off of the initial sequence in m. 35 in favor of this eight-
measure sentence motivates our hypermetric reading of m. 35 as 3 = 1. Haydn could have
written the eight-measure recomposition provided in Example 5.3 and followed this with
his eight-measure sentence; this is why we mentioned earlier that mm. 29–42 can be viewed
as a contraction of sixteen measures into fourteen. This interpretation rests on understanding
pitch structure, but Haydn’s sudden forte in m. 35 vividly and directly projects the unexpected
change in direction that occurs here. As is often the case with performance indications, this
change of dynamics does not merely provide an increase in volume but is intimately
connected to the thematic content.

EXAMPLE 5.3 Recomposition of mm. 29–34 as eight-measure unit

Before leaving the B section, we will point out three details. First, notice the deployment
of the chromatic descending figure from the end of the initial antecedent phrase at the
beginning of B. In the Haydn string quartet movement studied above, we also noted the
re-use of a cadential figure as an initiatory melodic gesture in the analogous location. Second,
there is an imitative relationship between the outer parts in the pseudo-sequence of mm.
35–38. Even though the imitation is inexact, it is unmistakable since this passage consists of
only two lines, each doubled at the octave. This might be viewed as related to the hidden
imitative relationship mentioned in mm. 1–4. Third, this B section is one that does not
tonicize a new key; instead, it heads directly back to the home key and transforms V from
169 Binary and Rounded Binary Forms

a key area into the dominant of the home key. Note that the controlling harmony for a
sizable expanse of the B section—mm. 32–40—is D major in root position. However, this
D major harmony is circumscribed by A major chords at the boundaries of B. Thus, there
is a prolongation of the dominant from the end of the first part of the minuet until the
thematic return, but internal to that prolongation is, in fact, a strong presence of root-position
D major. Once we are aware of the multiple levels on which harmony operates, these D
major sonorities do not diminish the tension of the extended dominant and the desire for
tonal return coordinated with the thematic return.
In A′, the antecedent phrase is the same as before, and the closing phrases are simply
transposed into the tonic key. The only element that is substantively modified is the
consequent phrase, which is further expanded from twelve to sixteen measures (mm. 57–72).
The additional four measures result from a repetition of the presentation (mm. 61–64 =
mm. 57–60). The upper instruments have a four-measure presentation, and then the lower
instruments have their own. Thus, the imitative relationship between upper and lower
instruments that was submerged in mm. 1–4 is explicit here. The expansion within the
continuation—the stasis on the I6 harmony—is the same harmonically as before, but the
melodic content is altered. The expansion in mm. 66–69 unfolds the descending sixth D6–F5
in response to the quickly ascending sixth E5–C6 in the measure preceding the expansion.
The F5 at the end of the expansion promptly leads to E5 over ii6 and the ensuing dominant
and to melodic closure at m. 72.
Two other aspects of the consequent phrase in A′ bear noting. First, the bass line at the
start of the consequent does not solidly articulate tonic harmony. Instead, the lower strings
(and two measures later the lower winds) emphasize the dominant. It is only when the melody
goes into the bass for the presentation by the lower instruments that D occurs in the bass
register. Note that the timpani is silent during the first four measures of the phrase, and re-
enters only to reinforce the return of D. The second aspect is perhaps less evident but is a
wonderful detail. In m. 64, the descending chromatic figure from m. 7 is incorporated,
though altered. Harmonically, the progression I6 – [V7] V7 – I – ii6 replaces a chain of
parallel thirds leading between I6 and V. Hypermetrically, the figure no longer has a
strong–weak placement, as it falls on hyperbeats 4 and 5 rather than 7 and 8. This conflict
between the memory of the original hypermetric placement of this melodic figure and the
hypermeter established by the immediate context has the effect of diluting the strength of
hyperbeats in the latter, thereby increasing the phrase’s desire for both tonal closure and a
strongly projected hypermetric downbeat. In writing this thematic return, Haydn makes
sure to “outdo” the expanded consequent from the minuet’s first part!
We offer only a few brief comments on the trio; the score, with hypermetric annotations,
is provided as Example 5.4. Trios are usually “simpler” in one or more ways compared to
their framing minuets (or scherzos). In orchestral works, there is almost always a reduction
in instrumentation, as is the case here. Often, the dynamics are softer, the harmonic rhythm
slower, and the lengths of sections shorter. All of these occur in this trio. Note that the trio’s
first part is only sixteen measures in length; mm. 97–112 are a written-out repeat of mm.
81–96 to allow for a very slight harmonic alteration (in mm. 102–103 the upper strings
briefly depart from their repeated D major chord).4 Due to the lack of tonal motion in the
initial twelve measures, the first part constitutes only a single phrase, leading from the tonic
to a PAC in the key of the dominant.
170 Analysis

EXAMPLE 5.4 Haydn, Symphony No. 101 in D Major (“Clock”), III, mm. 81–160
171 Binary and Rounded Binary Forms

EXAMPLE 5.4 continued

172 Analysis

EXAMPLE 5.4 continued

173 Binary and Rounded Binary Forms

EXAMPLE 5.4 continued

174 Analysis

EXAMPLE 5.4 continued

175 Binary and Rounded Binary Forms

As in the minuet, the B section returns quickly to the tonic key and does not tonicize
any other keys. The chord at the start of B—F major in first inversion—is a shock,
reharmonizing the F4 from the start of A with this chromatic chord. We can understand
this chord as part of a motion from the dominant at the end of A to the D major chord
reached early in the B section. Example 5.5 provides a literal chord reduction of this passage
at (a) and shows how it can be understood as a contraction of an ascending 5–6 sequence
at (b). Example 5.5 also points out that the goal of B is the home dominant, which is reasserted
at m. 127 and extended until m. 136. Thus, as we saw in the minuet, there is a salient root-
position D major chord subsumed within the larger prolongation of the dominant from the
end of A to the end of B. We have suggested one hypermetric interpretation of B; we view
the repetition in mm. 121–122 and in mm. 130–135 as outside of the underlying hypermeter.5

EXAMPLE 5.5 Sequential basis of B

The most fascinating component of the trio is its extensively rewritten thematic return
(mm. 137–160). As at the start of B, the introductory vamp lasts only two measures (and
here it is also placed in a higher register), but it is now much more vigorously cut off
by the full ensemble—not a solo flute. (Thus, we have shown the third measure of A′ as a
hypermetric downbeat.) When the opening melody re-enters, it does so in the bass, and
this motivates a recomposition that includes strong harmonic changes. In place of the placid
twelve-measure tonic pedal, there is an overall progression of I – vi – IV – ii6 – V4–3
6–5
spanning
mm. 137–148. This phrase seems destined for closure; in addition to the harmonic
progression, there is a clear arrival on scale degree 2 in the melody over the ii6 harmony
(m. 145). Instead, the phrase breaks off, and after some silence the pianissimo D major chords
return. The cellos and basses provide an independent bass line that introduces C to create
[V42] IV, a conventional signal of post-cadential function. Thus, although we have not yet
had a PAC in D major during the trio, there is a feeling that the trio is already past the point
of such a cadence. There are further melodic closes at m. 156 and m. 160—and the former
is reinforced by an A–D leap in the cellos and basses—but look at the horns. Already at m.
154 they have begun to sustain a tonic pedal in a low register (D2/D3). The trio ends on
its tonic harmony, but there is never a true PAC. This qualification of closure is possible
because the trio is only the middle section in the larger minuet–trio–minuet design.
176 Analysis

Finally, the hypermeter in mm. 150–160 requires comment. Throughout the trio, we
have understood the onset of the main flute melody to correspond with a hypermetric
downbeat. We do so as well here, but there are two complications. First, the flute enters
after three measures of vamp (rather than four or two). Second, and more significantly, the
added bass line suggests a hypermeter non-aligned with the flute’s melody. Since the bass
line remains on C during the first measure of the flute melody, the bass line suggests that
this measure is hyperbeat 4 in a quadruple hypermeter. This reading is further reinforced
by the onset of the long tonic pedal in the horns in the next measure. There is a conflict
here between the hypermetric cues of the accompaniment and those of the melody; when
this is the case, the larger melodic design generally predominates. Here, the eighth-note
melody is presented twice—mm. 153–156 and 157–160—and its implied quadruple
hypermeter gradually subsumes the accompaniment’s competing hypermeter.
This movement has a very straightforward harmonic language—none of the chords
introduced in Chapter 4, for example, are found. Primarily we have focused on the
movement’s more distinctive treatment of phrase design and hypermeter. While minuets
and scherzos often play with our expectations concerning phrase lengths and manipulate
hypermeter in interesting ways, many also have remarkable tonal features. Although tonal
features may be highlighted somewhat less here than in other chapters, as we proceed
to movements by Beethoven and Brahms there will be increasing discussion of tonal
elements.

Beethoven, Piano Sonata in A Major, op. 26, II

In the second movement of Beethoven’s Piano Sonata in A Major, op. 26, we find a very
lively scherzo, marked Allegro molto, which encloses a serene trio. Both scherzo and trio
sections have remarkable formal and tonal features. We will discuss the scherzo first, then
the trio, and finally the connection between scherzo and trio.
Example 5.6 provides an annotated score of the scherzo, and a cursory glance reveals it
to be a rounded binary form. The repeat of the first part is written out (mm. 9–16 = mm.
1–8) to permit the addition of some passing tones (compare, for example, m. 1 with m. 9,
m. 2 with m. 10). Wanting to ensure that the first part is not played four times (!), Beethoven
writes “La prima parte senza repetizione” in the score. The second part (mm. 17–67) receives
its customary exact repeat, and the thematic return occurs at the pickup to m. 45 with the
melody placed into the left hand. It is subsequently played by the right hand eight measures
later. We turn now to a closer look at tonal organization.
The scherzo, as is expected in a multi-movement work of Beethoven’s time, is in the
key of the overall work, here A major. But this is not the key established in the scherzo’s
opening measures. Instead, we hear a cadential progression in E major in mm. 1–4. The
motion to A major in mm. 5–8 results simply from transposing mm. 1–4 up a fourth. In
addition to the off-key beginning, the scherzo’s initial harmony is not a tonic or a dominant
of E major but its supertonic; in terms of the eventual A major, the initial chord is
submediant. This non-tonic beginning, and specifically its F minor chord, has implications
for the design of the second part. Finally, note the continuous ascent of the upper line in
the first part from C5 to A5: (A–B)–C5–D–E–[F]–E, (D–E)–F–G–A–[B]–A5. This,
too, influences the subsequent course of the music.
177 Binary and Rounded Binary Forms

EXAMPLE 5.6 Beethoven, Piano Sonata in A Major, op. 26, II, mm. 1–67
178 Analysis

EXAMPLE 5.6 continued

179 Binary and Rounded Binary Forms

The second part opens with an ascending sequence that connects the A harmony from
the end of the first part to a C harmony that is vastly extended (mm. 24–44). As shown in
Example 5.7, the sequence is a 5–o7 variant of the ascending 5–6 sequence. Although the
melodic material is directly based on the opening music, the placement of the melody with
respect to the harmony has changed. In the initial F minor chord, the melody moves from
third to fifth; at the start of part 2, the melody moves from chordal fifth to seventh. As a
result, the melodic line needs to descend, at least temporarily, and after a two-measure delay
the G moves to F (m. 20). The larger melodic framework is still an ascent, but a much
slower one, progressing from E (m. 17) to F (m. 20) to G (m. 24), where each of the latter
two tones is approached by a local descent. Supporting the G is the extended C major
harmony, which is destabilized through addition of its seventh and later its ninth, clarifying
its function as dominant of F minor. Thus, we see the logic of Beethoven’s pause on the C
major harmony: he is preparing the thematic return in a perfectly appropriate way—with
the dominant of its initial harmony. An expansion of the home dominant would be less
effective here, as it would not lead as strongly to the initial harmony and it would undercut
the emphasis on the key of the dominant in the first few measures of the thematic return.
In the upper part, there is a clear connection as well: the motion from E to F to G ultimately
leads to A at the thematic return. Just as with the earlier motions to F and G, the approach
to A comes from above, the seventh—B—having been introduced already at m. 26.6

EXAMPLE 5.7 Voice-leading representation of mm. 16–45

We have remarked already that the principal melody occurs in the left hand at the
thematic return. Initially, it is not the lowest voice; the bass motion is the same as at the
beginning, simply transposed down an octave. In the third and fourth measures of the return,
the left-hand melody does function as the bass voice (with the addition of a syncopated
chromatic passing tone—E—that recalls, albeit briefly, the extended dominant of F minor
from earlier). What is less immediately apparent in the thematic return is that the eighth
notes in the upper register are an embellished version of the middle voice from the opening.
The pitches corresponding to the middle voice in mm. 1–8 have been circled on the score
in mm. 45–52.
180 Analysis

The repetition of the melody by the right hand has a few significant effects. By emerging
after the left-hand statement, it powerfully conveys the ascending sweep characteristic of
this material. Although the left-hand statement momentarily breaks this sweep in mm. 49–52
by shifting down a fifth rather than up a fourth, one can nonetheless trace a motion from
the C4 at the start of the left-hand statement all the way to the A5 at the end of the right-
hand statement (m. 60). Yet the phrase is not over at m. 60 since the running eighths of
the left hand have arrived only on scale degree 3. The I6 internally expands the phrase, and
the PAC arrives four measures later (m. 64). Both during the phrase expansion and the four-
measure codetta that begins at m. 64, the upper part lands on C6—first supported by [V]
ii, then I6, and finally I. This focus on C6 is a beautiful outcome of the ascending melody,
taking the focal pitch of m. 1 and giving it a climactic registral treatment and supporting it
eventually with root-position tonic harmony.
While the expansions at the end of the scherzo’s final phrase attract the most attention,
there is a subtle recomposition at the very beginning of this phrase. Consider the eighth
notes in the left hand in m. 53. The note F does not arrive until the fourth eighth note; on
the downbeat, only A and C sound. As we pointed out in Chapter 1, two pitches that form
the interval of a third usually suggest the root and third of a chord, and this is clearly the
case here since the preceding measure contains root-position A major. Thus, this final
statement of the principal melody begins from tonic harmony with only a vague suggestion
of its characteristic F minor chord at the last instant before the next change of chord. We
might not even hear this brief suggestion of F minor at all, but it is worth noting that
Beethoven alters the eighth notes in m. 53 from the ones in the right hand eight measures
earlier in a way that is consistent with a nod towards the F minor chord. In m. 45, the
fourth eighth note was C (not F), as the lowest part in the left hand provided the F for the
F minor chord.
The trio certainly fulfils our expectation for simplification: the surface rhythm and
harmonic rhythm are slow and consistent, the texture is exclusively chordal, and the outer
voices proceed almost entirely in parallel tenths. The score appears in Example 5.8 and a
voice-leading representation in Example 5.9. The form is binary (not rounded binary), and
the dominant at the end of the first part is handled as V of the principal key (not as a key
area of its own). The dominant harmony extends from the end of the first part all the way
to the cadential dominant at the end of the second part; note the bass motion through the
octave A2–A3 across mm. 75–89.
There is one striking parallelism between the first and second parts of the trio. As shown
by the brackets in Example 5.9, the bass line’s initial approach to the dominant and its
approach to the subsequent return of the dominant are the same: scale degree 4 (G) prolonged
first by a lower neighbor (F) and then a chromatic passing tone (G). This might not
seem a particularly unique or salient melodic motion, but the sonorities that accompany
the bass F are striking, albeit different. In the first part, the upper voices move to the notes
of ii65; usually with a bass embellishment the upper voices either shift to be somewhat
more consonant with the bass, or they simply stay on their original notes. Moving to a
different sonority that is not consonant with the simultaneous bass embellishment stands out.
In the second part, the F initially supports [V65] IV but in the ensuing measure the fifth of
the chord is raised, temporarily producing an augmented triad [V6 ] IV. In addition, the
motions of the melody are related in these two locations. As shown by the dotted brackets
181 Binary and Rounded Binary Forms

EXAMPLE 5.8 Beethoven, Piano Sonata in A Major, op. 26, II, mm. 68–95
182 Analysis

EXAMPLE 5.9 Voice-leading representation of the trio

in Example 5.9, in both instances there is a leap downward from B, although it does not
occur in the same alignment with the bass line. Further, the leap is not to the same pitch—
E the first time, F the second. However, the F serves to delay E and does resolve there
once the dominant harmony arrives. The strongest argument for this melodic relationship,
though, is the minimal melodic motion in this trio; the leaps B4–E4 and B5–F5 are the
only leaps that span an interval larger than a third.
A remarkable formal feature of this movement is the presence of a retransition between
the trio and the da capo return of the scherzo. This type of retransition was rare in the
eighteenth century but became quite common during the nineteenth century. Here, the
thematic material derives entirely from the scherzo, reversing the tonal motion of mm. 1–8
as a way of modulating smoothly back from the trio’s D major. We will encounter a similar
retransition in the scherzo movement by Brahms to which we now turn.

Brahms, Piano Sonata in F Minor, op. 5, III

The scherzo movement from Brahms’s Piano Sonata in F Minor, op. 5, shares many
similarities with the Beethoven movement just studied. In addition to a retransition between
the trio and da capo return of the scherzo, Brahms’s scherzo also begins with a non-tonic
harmony, which again has implications for the tonal structure of the scherzo’s second part.
But Brahms’s scherzo is much more experimental. There are no exact repeats, and the
written-out repeat of the first part is considerably more varied than in the Beethoven. The
tonal goal of the first part is an unexpected one—the subdominant—and the beginning of
the second part is thematically independent from the rest of the scherzo. We will examine
these and other features of the scherzo first, and subsequently turn to the lengthy trio.
An annotated score of the scherzo is provided as Example 5.10. With its left-hand pattern
of a low bass octave followed by two higher chords, the scherzo’s opening resembles a waltz,
183 Binary and Rounded Binary Forms

EXAMPLE 5.10 Brahms, Piano Sonata in F Minor, op. 5, III, mm. 1–100
184 Analysis

EXAMPLE 5.10 continued

185 Binary and Rounded Binary Forms

EXAMPLE 5.10 continued

186 Analysis

EXAMPLE 5.10 continued

and it possesses the strongly projected four-measure hypermeter characteristic of a waltz.

Decidedly less waltz-like are the tonal elements. The first part of the scherzo consists of
sixteen measures, leading to the subdominant and not to the usual relative major. Moreover,
the way Brahms approaches this arrival is unusual: a stepwise, and almost fully chromatic,
bass line traces the diminished twelfth from E up to B. There is a sense of two parallel
phrases—mm. 1–8 and 9–16—due to the similarity in the melodic material at corresponding
locations in these two passages. However, the chord in m. 8 is iv6, which is not suitable as
a goal harmony. Thus, the feeling of two phrases is illusory, as a single tonal motion connects
the beginning with eventual arrival at the key of B minor. But, tonally speaking, what exactly
constitutes the beginning of this scherzo? The initial chord—o7—is not only non-tonic but,
187 Binary and Rounded Binary Forms

in fact, dissonant. At first, we think it will resolve to tonic harmony in m. 2, but note that
the fifth above the bass note F is displaced by the sixth (D); in harmonic terms, a VI6 sonority
replaces the expected i. In m. 3, we encounter o65, an inversion of the opening chord, and
only at m. 4 does an F minor chord arrive, but it is in first inversion. And, in m. 5, it is
transformed into an applied chord to the subdominant! Combined with the stepwise motion
of the bass throughout mm. 1–16, the de-emphasis of tonic harmony contributes to the
unsettled, turbulent quality of this opening.
In the preceding Beethoven scherzo, the repeat of the first reprise was written out to
allow for the addition of a few passing tones. In the Brahms, the recomposition of mm.
1–16 in mm. 17–36 is much more extensive. Overall, the pianist’s hands exchange material.
This occurs most literally in mm. 17–20 (= mm. 1–4) and mm. 25–28 (= mm. 9–12), where
the left hand states the former right-hand melody in melodic inversion. Notice that,
although the harmonies are quite similar, there is a slight stabilization of F minor in the
initial four measures of the varied repetition: the second measure now presents i6 (rather
than VI6) and the fourth measure lands on i rather than i6. The segments corresponding to
mm. 5–8 and 13–16 are more freely varied; Example 5.11 provides simplifications of mm.
5–8 and its varied repetition in mm. 21–24. The simplification reveals the transfer of
F–G–A–B from right hand into the left, and the partial transfer of the A–B–C–D from
left hand into the right. The transfer of the latter is, in fact, somewhat more complete as in
m. 22 the B is delayed by an embellishing A. Harmonically, the most significant change is
in the first measure (m. 21), where an applied chord of the Neapolitan replaces [V65] iv.7
The final segment of the first part is expanded by four measures in its varied repetition. Due
to the repetitions, we can easily discern that mm. 29–36 represent an expanded and varied
repetition of mm. 13–16. Notably, it is the two measures that depart furthest from their
model (mm. 13–14) that are stated multiple times—an expansion Brahms highlights through
his sostenuto indication.8 The varied repeat ends with a clear restatement of its final measure
(m. 36 = m. 16); notice the recomposition of the penultimate measure that enables a root-
position dominant chord and thus a perfect authentic cadence in mm. 35–36, in contrast to
the stepwise bass motion in mm. 15–16.

EXAMPLE 5.11 Voice-leading simplifications of mm. 5–8 and 21–24

188 Analysis

We noted above the stepwise motion through the diminished twelfth from E up to B
during the scherzo’s first part. The second part continues the ascent from B (m. 37) up to
G (mm. 50–69), dividing the span at D (m. 42) and E (m. 46) and thereby projecting the
opening diminished seventh harmony linearly. The unusual bass design of B leads smoothly
to the return of the diminished seventh at the onset of A′ (m. 70), the non-dominant
preparation responding to an idiosyncrasy of the thematic material. As in the first part, the
bass ascent during B is stepwise, although here the stepwise motion happens more slowly;
the bass tones participating in the ascent occur every other measure: B (m. 36), C (m. 38),
C (m. 40), D (m. 42), E (m. 44), E (m. 46), F (m. 48), G (mm. 50–69). After the arrival
on D (m. 42), the upper line moves in parallel tenths with the bass; in mm. 36–42, the
upper line descends from B6 to F6. These observations are summarized graphically in
Example 5.12. Note that within mm. 42–50 the chords in odd-numbered measures are
subsidiary to those that participate in the pattern of parallel tenths.9 This entire passage is
principally organized around the large linear motion in the bass rather than standard
harmonic progressions; this is why we have not provided a detailed harmonic analysis on
the score.

EXAMPLE 5.12 Voice-leading representation of mm. 36–70

We have addressed the tonal design of the B material, but we have not commented on
two unusual aspects of its formal design. First, due to the discontinuous gestures in mm.
38–41, the B section does not seem to begin until the arrival on D major and the
continuous right-hand material at m. 42. For this reason, on the score we have labeled mm.
38–41 as a transition to the B section. There is, in fact, some blurring of the boundary
between the end of A and the start of this transition. The descending arpeggio that marks
the cadence on B minor at the end of A (m. 36) is answered by the bass in the next measure;
the latter seems simply like a one-measure post-cadential extension. However, these two
measures provide the thematic material for both of the two-measure units within the
transition in mm. 38–41. From a tonal perspective, mm. 36–37 (beat 2) provide a close to
the A section, but thematically they prepare the transition to the B section. Blurring of
189 Binary and Rounded Binary Forms

boundaries between sections is typical of Brahms’s style, and we will encounter more
elaborate overlaps in the Brahms works studied in Chapters 7 and 8. The second unusual
aspect of the B section is its thematic independence from the A section. As we have seen
in the Haydn and Beethoven examples, B is normally a development of material from A.
In the Brahms, however, the rhythmic pattern from the transition (mm. 38–41) serves as
the basis for every measure in mm. 42–62; the rhythm from the scherzo’s first measure
disappears completely until m. 63. There is thus a stark contrast in thematic material that
begins at m. 42—a contrast more typically found between a scherzo and a trio than between
the A and B sections of a scherzo (or of a trio). In A′ only the rhythmic pattern from the
scherzo’s opening returns, which means that the scherzo’s contrasting rhythmic motives are
not brought into direct conflict or reconciliation. It is notable, though, that in the approach
to the final cadence (mm. 87–95), the rhythmic content becomes dramatically simpler: each
measure consists of a half-note octave followed by a quarter-note chord. This may be viewed
as a liquidation of the rhythmic content to its underlying basis; this long-short succession is
suggested by the tonal content of the two rhythmic figures. In m. 1, the brief C is a passing
tone between D and B; similarly, in m. 42, the F is prolonged through arpeggiation and
then moves to D on beat 3. (Of course, the half-note to quarter-note motive is present
throughout mm. 42–54 in the left hand.)
In A′ we encounter yet another variation on the opening material. Both right and left
hands give the opening melody (mm. 70–73), combining the modes of presentations seen
in the first part and its written-out varied repeat. The first eight measures of A′ closely
resemble the corresponding portion of A, but the presence of F minor is further stabilized.
Both in the second and fourth measures of A′ is there a root-position F minor chord. It is
only at the eleventh measure of A′ (m. 80) where significant divergence from A occurs in
order to avoid modulation. After an expansion of the phrase through the sequential
repetition of the diminished seventh arpeggio figure, a cadential segment (mm. 87–94) leads
to closure at m. 95. A short codetta further extends this already massive phrase of twenty-
six measures (mm. 70–95).
Particularly beautiful is the trio’s placement in D major, since it thereby seems to
germinate from the trio-like D major passage that began the scherzo’s second part.10 With
its soft, slow-moving, homorhythmic texture, the trio provides necessary and obvious
expressive contrast. The trio’s phrase design and hypermetric structure also contribute to
this effect. Throughout most of the scherzo, the bass line announces a forceful change of
harmony at each downbeat. At the trio’s third measure, this fails to occur; instead there is
a syncopation at the level of the measure, and this becomes the trio’s most distinctive rhythmic
feature. We provide a score with hypermetric annotations as Example 5.13.
As the annotations on Example 5.13 suggest, the syncopated figure maintains a consistent
placement on the second hyperbeat. This hypermetric identity is strongly established by the
beginnings of the trio’s first two phrases. When the gesture returns at the end of the third
phrase (mm. 129–132), it is easily heard with the same hypermetric placement due to the
end of the two-measure melodic parallelisms in mm. 123–128. After the symmetrical eight-
measure phrases in mm. 101–116, the asymmetry within mm. 117–132 is striking; the
sixteen-measure phrase does balance the previous two phrases, but is subdivided as 6 + 6 +
4. Brahms returns to this thematic material at the end of the second reprise and fashions a
version that preserves four-measure hypermeter (mm. 192–203).
190 Analysis

EXAMPLE 5.13 Brahms, Piano Sonata in F Minor, op. 5, III, mm. 101–212
191 Binary and Rounded Binary Forms

EXAMPLE 5.13 continued

192 Analysis

The second part of this trio is noteworthy for two bold phrase expansions. The first
occurs in mm. 139–153 where an eight-measure unit is expanded to fifteen measures. This
phrase is especially similar to the trio’s second phrase (mm. 109–116), which makes the
expansion of the [ø7] V particularly obvious. Recognizing the expansion explains a seeming
anomaly: the phrase consists of an odd number of measures and the concluding dominant
chord (m. 153) seems a measure too short. When the expansion is taken into account,
however, it is clear that the final chord would only last one measure in the underlying
prototype. Many phrase expansions maintain relatively periodic surface hypermeter, but this
one decidedly does not. In fact, without awareness of the deeper level of hypermeter, the
conclusion of this phrase—which provides the dominant arrival that sets up the thematic
return—feels awkward. Within the expansion, a nice touch is the inclusion of the syncopated
gesture in the second measure (m. 146). Somewhat puzzling is the chromatic slide to G in
the bass that momentarily produces a dominant seventh of C major. Brahms appropriately
revisits this harmony in A′.
The remaining phrase expansion is much longer than the previous one, and it is used
to prepare the trio’s climax. In A′, the second phrase begins as a minor-mode variant on
the trio’s second phrase. At the end of its fourth measure, the phrase goes off track tonally,
rhythmically, and registrally. The phrase at first seems headed for F major (mm. 166–168)
but ultimately goes to C major (mm. 169–172), a destination that seems related to the
tonal content of the expansion just before A′. Rhythmically, the characteristic half-note to
quarter-note succession is augmented from three beats into three measures (mm. 169–171);
the six-measure unit in mm. 166–171 expands a four-measure model through this written-
out deceleration. As the goal C major triad (VII) unfolds into the home dominant, the
rhythmic content further hints at the passage’s digressiveness through the bass line’s
reintroduction of the scherzo’s primary rhythmic idea. The entire passage from mm. 166–179
occurs in a very high register. The impression that these measures constitute a large
parenthetical insertion is confirmed by the arrival of I6 at m. 180; I6 is the chord that should
have occurred at the phrase’s fifth measure. Brahms does not allow the phrase to end in a
half cadence at this point, but keeps reiterating these long-expected measures (with the
glorious melodic arpeggiation D, F, A, D connecting consecutive hyperdownbeats). These
repetitions culminate with a restatement of the latter portion of the final phrase from the
first part, in invertible counterpoint (mm. 192ff.). Previously this phrase was not consonant
with four-measure hypermeter, but this recomposed version is. Especially significant is the
arrival of the trio’s structural dominant on a hyperdownbeat (m. 200) and its interaction
with the right-hand melody. The right-hand chord is tied over from the previous measure
and thus is not a syncopation—as everywhere else in the trio—but a suspension. It is
normative for a suspension to enter in a metrically weak position and sustain through a
stronger beat. Thus, the stylistically marked syncopation is transformed into a stylistically
unmarked suspension. This final phrase, then, not only brings hypermetric resolution to the
material originally presented at the end of the trio’s first reprise but also provides a climactic
working out of the trio’s characteristic rhythmic motive. After this powerful buildup—there
is essentially a single tonal motion from tonic to dominant spanning mm. 162–200—Brahms
provides a tonic harmony only at the start of an eight-measure retransition that leads to the
da capo repeat of the scherzo.
193 Binary and Rounded Binary Forms

SUGGESTED ASSIGNMENTS
The ultimate goal of analysis is to identify what is special or unique about a particular work.
To that end, we suggest you proceed through the following steps:

1. Identify the main formal divisions and subdivisions (A, B, A′, retransition, codetta, etc.).
2. Identify important motives that are repeated/developed in the course of the movement.
3. Examine features of the phrase rhythm, that is, the interaction of phrase design and
hypermeter.
4. Identify important features of the overall tonal organization and details of the harmony,
particularly in more complex passages. You may want to prepare a simplification of the
voice leading of certain passages, identifying harmony, hypermeter, important motives,
etc.

For each of the five pieces below, we have provided a few questions to point towards some
of the movement’s unique aspects. We encourage you to study the music on your own and
to complete the analytic steps listed above before considering our questions.

Haydn, String Quartet in B Major, op. 76, no. 4, III

1. Identify several techniques by which Haydn develops the opening eighth-note motive.
2. Compare the harmonic progressions and outer-voice motions of mm. 1–8 and 29–36.
3. What are the chords in mm. 17–18?
4. How does Haydn’s deployment of the cello in mm. 9–29 clarify the formal-tonal
organization?
5. What is the formal function of mm. 37–50?
6. In what way(s) is the thematic material of the trio similar to that of the minuet?

Haydn, String Quartet in C Major, op. 74, no. 1, III

1. We studied the end of the minuet in Chapter 3 (see Example 3.17) as an example of a
three-measure parenthetical insertion (mm. 56–58). This thematic material originates
in mm. 11–12. What is the formal function of mm. 11–12 within mm. 1–14?
2. Upon what model is the tonal structure of mm. 15–23 based? (In what other work studied
in this chapter was a similar progression used at the same formal location?)
3. What is the formal function of mm. 27–31?
4. This is a minuet in which one might say that the thematic rounding occurs twice. Explain
and speculate on why Haydn might have included the additional statement of the opening
material.
194 Analysis

5. Comment on the D in m. 40. What pitch did we expect here? What does the D
“trigger”?
6. Compare the first violin line in the first five measures of the minuet to that in the first
five measures of the trio.
7. In addition to the melodic similarity noted in the preceding question, what other
similarities are there between the thematic materials of the minuet and those of the trio?
8. What is the formal-tonal role of mm. 97–112? How is the modulation from m. 112
back to the da capo restatement of the minuet achieved?

Haydn, String Quartet in F Major, op. 74, no. 2, III

1. Why do mm. 1–4 not constitute a complete phrase? How could one rewrite m. 4 so
that mm. 1–4 would be a complete phrase? If one undertook such a rewriting, to what
phrase design would mm. 1–8 correspond?
2. Describe the phrase design of mm. 18–37. Comment on the relationship of this passage
to mm. 1–8 (and the hypothetical recomposition of mm. 1–8 mentioned in the
preceding question).
3. What is the formal function of mm. 38–41?
4. What melodic relationship exists between the end of the minuet and the start of the
trio?
5. What is the formal-tonal role of mm. 68–78? On what earlier measures are mm. 68–69
based, and what is the effect of this on the listener?

Beethoven, Violin Sonata in C Minor, op. 30, no. 2, III

1. Why is there no repeat sign in m. 18?
2. Comment on the significance of the piano left hand in m. 16 (beats 1 and 2).
3. With what harmony does Beethoven prepare the scherzo’s thematic return? Does this
provide a smooth harmonic return, or is another type of connection present?
4. Comment on the relationship between the piano and violin parts in the trio.
5. With what harmony does Beethoven prepare the trio’s thematic return? How many
measures in total are occupied by this harmony?

Brahms, Cello Sonata in E Minor, op. 38, II

1. This “quasi Menuetto” is unusual in the length of its upbeat: the piano part has more
than a complete measure before the principal melody begins. What implications does
the lengthy upbeat have for hypermeter throughout the minuet?
195 Binary and Rounded Binary Forms

2. The initial pair of notes F–E and the motive of a descending semitone (especially 6^–5^
in minor) permeates both minuet and trio. Explain.
3. Compare the phrase design of mm. 2–14 and 60–76.
4. Is the trio an example of binary or rounded binary form? Justify your answer.
6 Classical Sonata Form

Sonata form in the Classical period—as expressed in the music of Haydn, Mozart, and early
Beethoven—is an expansion of rounded binary form, where the initial part is the exposition
and the second part contains both the development and the recapitulation. Both parts are
repeated, as specified in the score through repeat signs, though the repeat of the second part
is rarely taken in modern performances. As the dimension of sonata-form movements grew
in the nineteenth century, repeats disappeared, leaving a three-part design: exposition–
development–recapitulation. Although sonata form does appear in various types of move-
ments, it is almost always found in the first movement of instrumental works such as sonatas,
trios, quartets, and symphonies. All of our examples are first movements of such works.
There is always a danger in defining a model of Classical sonata form, since there are
many exceptions to the “rule.” Still it is helpful to have in mind the norm for sonata form
in both the major and minor modes.1 Thematically, the exposition typically contains an
initial theme, a transitional passage leading to a second theme, and a closing section that
may involve a full-fledged theme and/or shorter phrases. The content of the development
is harder to define; it may include new material, but typically involves development of
motives and themes from the exposition leading to a retransition. The recapitulation
normally follows the exposition’s ordering of material. In many sonata forms, additional
closing material—referred to as a coda—follows the recapitulation. Unlike thematic content,
tonal structure differs in major- and minor-mode sonata forms. In the major mode, the initial
theme is in the tonic key, and the following transition will typically lead to V of V.2 The
second theme and closing section will then appear in the key of the dominant. The
development digresses from the dominant but eventually returns to it for a retransition leading
to the opening theme in the tonic. In the recapitulation, it is normally the transition to the
second theme that is rewritten to remain in the tonic, ending on the dominant of the home
key. The remainder of the recapitulation presents a transposed version of the corresponding
exposition material. In a movement in the minor mode, the transition to the second theme
in the exposition typically leads to V of III, and the second theme and following material
are written in the key of the relative major (III).3 The development section leads eventually
to V and the retransition, and the transition in the recapitulation is usually rewritten to remain
in the tonic key, ending on its dominant. The remainder of the recapitulation presents a
transposed version of the corresponding exposition material, though often some—or even
all—occurs in the parallel major. This preserves the modal quality of this material, which
thereby retains more of its original expressive import. Again, bear in mind that this is a general
description of the norm, subject to a wide variety of modifications.
197 Classical Sonata Form

Exposition Development Recapitulation

||: th.1 – trans. – th.2 – cl. mat.:||: ( ) th.1 – trans. – th. 2 – cl. mat.:||

Major: I [V] V ----V ------ -V 7, I I

Minor: i [V] III --- III ----- -V 7, i i/I

|__________________________________________| |____________________|

FIGURE 6.1 Basic Models of Classical Sonata Form

A chart of the basic thematic/tonal models outlined above is provided in Figure 6.1.
As indicated by the repeat signs, the formal design is binary, though once the repeat signs
are removed, the form is ternary. From a tonal perspective, sonata form falls into two large
parts, as indicated by the brackets at the bottom of Figure 6.1. Thus we must be careful to
define exactly what it is we are describing about sonata form. The formal design is binary
as long as the repeats remain intact, but its tonal design is also binary—but a different binary:
the first part contains the exposition and development and the second part the recapitulation.
It is useful to distinguish the two by terminology: formal design vs. tonal structure.
In this chapter we will examine two opening movements in some detail: Mozart’s
Piano Sonata in F Major, K. 332, and Beethoven’s Piano Trio in C Minor, op. 1, no. 3.
The Mozart is straightforward in its formal-tonal organization, while the Beethoven is more
complex in its formal design.

Mozart, Piano Sonata in F Major, K. 332, I

In almost all respects, the initial movement from Mozart’s Piano Sonata in F Major, K. 332,
reflects the standard practice of the Classical period for a sonata movement in the major
mode. The score of this movement with analytic additions—to be discussed below—is
provided in Example 6.1, and Figure 6.2 provides a chart of the movement’s formal-tonal
organization. The left column of Figure 6.2 lists the three major sections (Exposition,
Development, and Recapitulation) and their subdivisions. The center column provides
corresponding measure numbers, and the right column indicates key areas and important
harmonic motions within these keys. As a first step in our analysis, we will talk through the
formal divisions, then go back and discuss some important features of the melody, the
harmony, and the metric organization.
The opening theme of this sonata movement is in two parts, both ending with perfect
authentic cadences in the tonic. It is this strong tonal closure as well as the introduction of
new melodic material that creates a two-part initial theme. This theme is followed by a
transition (T1) that leads to V of V (mm. 37–40), preparing the statement of the second
198 Analysis

EXAMPLE 6.1 Mozart, Piano Sonata in F Major, K. 332, I

199 Classical Sonata Form

EXAMPLE 6.1 continued

200 Analysis

EXAMPLE 6.1 continued

201 Classical Sonata Form

EXAMPLE 6.1 continued

202 Analysis

EXAMPLE 6.1 continued

203 Classical Sonata Form

EXAMPLE 6.1 continued

204 Analysis

EXAMPLE 6.1 continued

205 Classical Sonata Form

EXPOSITION
Theme 1
part 1 1–12 I
part 2 13–22 I
Transition (T1) 23–40 leads to V of V
Theme 2
antecedent 41–48 C(V): I – V
consequent 49–56 I
Transition (T2) 56–70 leads to V
Closing section
closing theme 71–76 IV – I
expanded repetition 77–86
closing phrase 86–93 I

DEVELOPMENT
initial phrase 94–101 I
repetition (octave lower) 102–109 I
extension (based on T2) 109–123, 123–132 leads to F(I): III

RECAPITULATION
Theme 1
part 1 133–144 I
part 2 145–154 I
Transition (T1) 155–176 leads to V
Theme 2
antecedent 177–184 I–V
consequent 185–192 I
Transition (T2) 192–206 leads to V
Closing section
closing theme 207–212 IV – I
expanded repetition 213–222 I
closing phrase 222–229 I

FIGURE 6.2 Outline of Mozart, Piano Sonata, K. 332, I

206 Analysis

theme in the key of the dominant, C major. Theme 2 consists of a sixteen-measure

antecedent-consequent construction, ending with a perfect authentic cadence (mm. 55–56)
in C major. Rather than proceeding directly to closing material, theme 2 is followed by a
second transition passage (T2), which contains a descending fifth sequence in the parallel
minor mode and leads once again to V of V. Now the closing theme enters, although it
does not open as we might expect with the local tonic (C major), but with the subdominant
harmony. This six-measure phrase is then repeated in varied form an octave higher and
expanded, leading to a perfect authentic cadence (m. 86) and a final closing phrase.
The development opens with an eight-measure phrase in the key of the dominant, which
is subsequently repeated an octave lower. Although by no means the only possibility,
continuing the key from the end of the exposition into the start of the development—
however briefly—is the most common tonal strategy at this formal juncture. This tonally
closed unit (mm. 94–109) is followed by a transitional passage based on the earlier transition
leading from the end of theme 2 to the closing theme, that is, T2, only here the sequence
leads us from the lower register to the one an octave higher. The goal of this motion is the
A major chord, reached in m. 123. This A major chord has the potential to be V of vi, but
it never functions in that capacity. It has been marked III on the score, and eventually we
must consider why Mozart would make an A major chord the goal of harmonic motion in
the development section. For now we note that the C is subsequently replaced by C
preparing the entry of V43 and then V7 to reintroduce the opening theme in the tonic key.
The recapitulation follows standard procedure. The only change from the exposition
comes in the transition to theme 2 (T1), which is rewritten to lead to V rather than V of
V. Thus the entire recapitulation remains in the tonic key.
Having taken this brief descriptive walk though the movement, we will now go back
to the beginning to examine various parts in greater detail. Our discussion will increasingly
focus on the musical elements that make this movement unique and memorable, although
we will continue to make some general comments on Classical sonata form. A solid
understanding of standard sonata-form procedures, in fact, helps to reveal the truly
distinguishing features of a particular movement.

Exposition
The first theme, part 1, consists of three four-measure hypermeasures, as indicated by the
Arabic numbers between the staves in Example 6.1. Initially, these hypermeasures seem
unusually distinct due to the clear changes in rhythm and texture. A close look at mm. 1–12
reveals elements that bind the twelve measures together and that also have significance for
the movement as a whole. The melodic line contains two important motivic components:
the arpeggiation of the tonic triad (marked with brackets in mm. 1–2, 6–7, and 10), and
the upper neighbor of scale degree 5, that is, D5 (mm. 5 and 9). The melody’s lower range
between the two iterations of D5 highlights this pitch, as does its function in m. 9 of
introducing the decorated return of the arpeggiation of the tonic triad. The varied return
of the opening arpeggiation coincides with the melody’s most active rhythm and provides
momentum towards the ensuing cadence. A third, less obvious, component of mm. 1–12
is the gradual descent of a fifth from C5 to F4: C5 (m.2)–B4 (m.3)–A4 (m.5)–A4 (m.9)–G4
(m.11)–F4 (m.12). These notes are circled on the score. Note that although the A4 in
207 Classical Sonata Form

m. 5 continues on down to F4 in m. 7, the latter motion has been interpreted as extending

A4 until it reappears at m. 9.4 This delay on A4 thus articulates a clear division of the fifth
into two thirds: C5–B4–A4 (mm. 1–5) and A4–G4–F4 (mm. 9–12). Before moving on to
the second part of theme 1, there is one final feature of the first part worthy of note, namely
the weak hemiola pattern implied by the interaction between melodic contour (duple) and
meter (triple) in mm. 5–6 (right hand) and 7–8 (left hand), marked by dotted brackets on
the score. Mozart’s legato slurs provide further clarification that the hemiola is not strongly
projected. Like the other features we have noted in the opening twelve measures, however,
this implicit hemiola pattern is developed later.
The second part of theme 1, stated in the upper octave, is clearly based on the primary
motivic component of the movement, the descending arpeggiation of the tonic triad
(bracketed on the score). Despite this motivic connection, the higher register, new dotted
rhythms, and simpler texture make the second part of theme 1 rather contrasting from its
first part. Particularly satisfying, then, is the exact return of the melody at the final cadence
(compare mm. 19–20 with 11–12). In the cadential melody, note that B is consistently picked
up and resolved an octave lower in an inner voice rather than proceeding by step to A in
its own register. At the juncture between the two parts of theme 1, B quickly returns in
the upper voice—on the last beat of m. 12—and moves by step down to A. At the end of
the second part of theme 1, B does not return immediately in the upper voice, but there
is a longer-range connection to the A5 in m. 24, a connection that is sealed with the
reiteration of B5–A5 across mm. 25–28. This extends a connective thread across the
otherwise sharp juncture between theme 1 and the transition.
Theme 1 ends with a two-measure extension that reiterates the cadential harmonies and
the melody’s central C–F fifth in the two registers explored during the theme. These
repetitions give theme 1 an unusually strong close, as does its separation from the ensuing
transition by a beat of rest. These features set the stage for quite a shock when the transition
begins: the gentle floating down from C6 to C5 proceeds not to C4 but to C4. The dynamics
shift to forte, and the simple harmonies from the end of theme 1 give way to an onslaught
of diminished seventh sonorities (mm. 25–26 and 29–30). We have already observed the
presence of one connection across this musical chasm—the resolution of the B5 “hanging”
in the upper line to the A5 in m. 24. Note that this A5 is preceded by a rising D minor
arpeggio (bracketed on the score). Thus, the first two measures of the transition are a variant
on the first two measures of the initial theme! Many transitions begin with a relatively literal
reuse of the start of the initial theme; Mozart’s disguised variant in this movement is more
exceptional. Example 6.2 provides a simplification of the transition’s tonal content, along
with some analytic notations. If one removes the sixteenth-note elaborations from the first
two measures of the transition, it becomes apparent that the arpeggiation of the D minor
chord in mm. 23–24 builds upon another element of theme 1: the conflict of three half
notes against two measures of triple meter (i.e., hemiola).5 This pattern is repeated in mm.
27–28 and then three successive times beginning in mm. 31–32. As shown between the
staves, the two-measure pattern is initially grouped into two metric units of four measures
each, then into three units of two measures each. This fragmentation of four-measure into
two-measure units increases the transition’s drive towards its tonal goal—the V of V at
m. 37. As is typical for the end of a transition in a sonata form, the goal harmony is reiterated
across several measures (mm. 37–40). Also common is the brief rest after the transition.
208 Analysis

EXAMPLE 6.2 Voice-leading simplification of mm. 22–40

The shift into the minor mode during the extension of V of V is not uncommon either;
this modal shift permits a brightening in color when the second theme begins.
The exposition’s overall tonal motion thus far is I – V of V, preparing the second theme
in the key of V. Note the use of a German augmented sixth chord (mm. 35–36) immediately
before the arrival on V of V. Augmented sixth chords frequently occur at the ends of
modulatory passages, where they clarify the new key by pointing out its dominant. In the
case of a modulation to the key of the dominant, the augmented sixth chord is closely related
to the initial tonic harmony. In this piece, the initial tonic triad is F–A–C, and the German
augmented sixth in the key of the dominant is F–A–C–E . To put it differently, the initial
tonic triad can be construed as a subdominant harmony in the key of the dominant, and
the German augmented sixth represents a chromatic intensification of that subdominant.6
Like the “hanging” B5, this tonal connection spans the sharply articulated boundary
between first theme and transition.
The second theme consists of two eight-measure phrases, grouped into 4 + 4, an
antecedent leading to V in the local key and a consequent leading to a perfect authentic
cadence. The theme begins from the inner-voice tone E5, but in the next measure G5, so
carefully prepared by the preceding transition, is stated above it, and from there descends
by step to D5 supported by V. The notes in this descent are indicated by upward-directed
stems added to the music. The consequent phrase traverses the same path, this time
completing the descending fifth G5–C5. The last three notes of the descent, E5–D5–C5
(mm. 55–56), have been circled to highlight the linkage to the following transition, which
elides with the final measure of the second theme. The linkage is subtle: E5–D5–C5 (right
hand, mm. 55–56) answered by E4–D4–C3 (left hand, mm. 56–58).7
The following transition (in its tonic-key version from the recapitulation) was discussed
in Chapter 1 in the section on Diatonic Sequences; see Examples 1.13a and 1.13b and the
accompanying discussion. It is by no means common to have a transition between the second
theme and the closing material; we have designated mm. 56–70 as a transition because of
its reduced melodic content and the presence of sequence. The one feature of this sequence
important to note here is that the hemiola pattern, while implicit earlier, now becomes
explicit in mm. 64–65 through the harmonic rhythm, use of bass register, and the dynamic
209 Classical Sonata Form

changes. As noted above, this transition leads us once again to V of V in preparation for the
closing material, which opens with subdominant harmony. Note that because of the formal
boundary between the end of the transition (m. 70) and the start of the closing material (m.
71) V does not proceed to IV; the subdominant harmony represents the beginning of a new
phrase. Launching the closing material from subdominant harmony fits well with its slower
rhythms and chordal texture, bringing a calmer mood after the minor mode and hemiola
conflicts of the preceding transition.
In Figure 6.2 we have shown a “closing theme,” which is repeated in expanded form,
and a “closing phrase.” The amount of material within the closing section of an exposition
can vary considerably, as the perfect authentic cadence at the end of the second theme has
already performed the exposition’s necessary tonal “work.” As a result of this variability,
we will use a variety of terms in discussing closing sections: closing theme, closing phrase,
and codetta. It is not always obvious which term best suits a given portion of the closing
section, and some analysts simply demarcate the subdivisions—marked by authentic
cadences—within the closing section without giving them further labels. We feel, however,
that even if there is ambiguity about the distinction between, say, a closing theme and a
closing phrase, it is preferable to reflect carefully on the construction of the closing section
and attempt to reveal something about how its organization is similar to, or different from,
the closing sections of other expositions. Thus, “closing theme” signifies a phrase or group
of phrases that possesses a length and melodic interest relatively commensurate with theme
1 or theme 2. A “closing phrase” is a single phrase that does not approach the scale of a
theme. Finally, codetta refers to a short passage of repeated cadential gestures. Bear in mind,
though, that the most important formal boundary is the one between the second theme
and the closing section; this moment is always articulated by a perfect authentic cadence in
the secondary key. This is usually the first such cadence, except in cases where the melodic
material of theme 2 is revisited immediately afterwards. In that much less common situation,
the next perfect authentic cadence in the secondary key would mark the move from second
theme to closing section.
Returning to the Mozart, we see that the closing theme consists of two phrases (mm.
71–76 and 77–86). The initial phrase, which is the first six-measure phrase in the exposition,
ends with an imperfect authentic cadence. The phrase is repeated an octave higher and is
heading towards a perfect authentic cadence at m. 82, but it is evaded through the
substitution of first-inversion C major harmony and the melodic leap up to G5. Only after
an additional four measures does the perfect authentic cadence occur. These intervening
measures (mm. 82–85) therefore represent an internal phrase expansion that arises through
parenthentical insertion (due to the sharply contrasting musical content).

Development
The development opens with an eight-measure phrase that might first appear to be a new
idea. Compared to Haydn and Beethoven, Mozart was fond of including a new theme in
his developments. In some respects this material is new, but careful examination reveals
similarities with earlier material, particularly the opening of the second theme. The fleeting
sixth G4–E5 of this earlier theme is now expressed as the broader arpeggiation G4–C5–E5,
and the top voice G5, which was introduced in the second measure of theme 2, is here
210 Analysis

delayed until the third measure. If we consider longer-range melodic continuities within
the development’s first phrase, we observe a neighbor motion G5–A5–G5 that soars above
the third E5–D5–C5, which moves towards closure in the phrase’s second and fourth
measures but only attains C5 in its final measure. The prolongation of scale degree 5 by its
upper neighbor while a lower line moves from scale degree 3 to scale degree 1 is exactly
analogous to the last four measures of the first part of theme 1 (mm. 9–12). So this new
idea incorporates important features of earlier ideas. This phrase is then repeated an octave
lower (mm. 102–109), the final measure eliding with the following sequential passage. Since
this sequence leads back to the upper octave, there is a registral connection between the G5
prolonged throughout the development’s initial phrase and the G5 as part of the Italian
augmented sixth chord leading to the A major chord (III) at the end of the sequential passage
(mm. 122–123).
This A major chord is clearly articulated as a tonal goal, and we should ask ourselves
why Mozart has made III rather than V the harmonic goal of the development section.
There is no definite answer, but we can note that this intermediate goal divides the space
between V (end of exposition) and I (beginning of the recapitulation), thus dividing the
large-scale bass motion of a fifth into two thirds. Recall that the opening theme, part 1,
descends a fifth, clearly divided into the thirds C5–B4–A4 and A4–G4–F4. We should not
overstate this connection as there are a few other instances where Mozart ends a development
on III , but imagining this very long-range connection stimulates a tantalizingly rich hearing
of the tonal motion across this development. The A major chord is sustained over four
measures, then changed into an A minor chord in preparation for the connecting V43 chord
that leads back to the tonic and the opening theme. Mozart does not do the expected here,
but rather states V7 in the upper register and piano before beginning the restatement of the
opening idea in the lower octave, which was carefully prepared by the V43.8 The result is
that B5, the dissonant seventh, is not resolved in that register until the second part of theme
1. This delayed resolution of B5 to A5 recalls the connection we heard earlier between the
end of theme 1, part 2, and the following transition.

Recapitulation
There is no need to discuss the recapitulation at great length, since the primary difference
is that the tonic key prevails throughout. As is usually the case, the necessary modification
occurs during the transition between the first and second themes, and it is achieved in the
most common manner. The transition begins exactly as in the exposition (mm. 155–163 =
23–31), several measures are recomposed (mm. 164–172 only roughly correspond to mm.
32–36), and the final measures of the transition exactly transpose the corresponding ones
from the exposition (mm. 173–176 transpose mm. 37–40). Although the second theme and
closing material are unmodified except for transposition, their placement into the tonic key
reveals a connection to the first theme that was previously obscure. The striking subdominant
harmony at the beginning of the closing theme (m. 207) now harmonizes D5, the upper
neighbor of C5 and a central element in the first part of theme 1. Further, in the approach
to its perfect authentic cadence, the closing theme now outlines the fifth C5–F4, the same
fifth outlined across the first part of theme 1 (and also, given the tonic transposition, across
theme 2 in the recapitulation).
211 Classical Sonata Form

At first glance, this movement seems to be a study in contrast. There are a wide variety
of melodic and rhythmic figures, sharp dynamic changes, and oppositions between different
registers. Many formal boundaries feel abrupt due to these contrasts and the frequent use of
rests at major junctures. One of the contributions of analysis is to reveal connections among
seemingly disparate materials and to bridge formal divisions. Analysis can provide a balance
between local detail and overall organization, enriching our experience of this movement
whether as a listener or as a performer.

Beethoven, Piano Trio in C Minor, op. 1, no. 3, I

In comparison to the Mozart, the initial movement of Beethoven’s Third Piano Trio is larger
in dimension and bolder in its dramatic expression. In most, but not all respects, it conforms
to the norm for a Classical sonata form in the minor mode. The exposition consists of two
distinct thematic groups, the first in the tonic and the second in the relative major (III). This
is followed by a closing section that confirms the new key. The development begins from
III and progresses to V with a lengthy expanse of iv in between. Thus the exposition and
development chart an overall tonal course found in many Classical sonata-form movements
in the minor mode: i – III – iv – V. With one notable exception, the recapitulation follows
the same order of events as in the exposition, except that the transition to theme 2 is rewritten
to stay in the tonic key. So far, very normal. However, as we shall see, the divisions of these
larger formal sections and their tonal contents are not so predictable.
An outline of the movement’s formal-tonal organization is provided in Figure 6.3. Like
Figure 6.2, the left column lists the major formal sections and their subdivisions, the middle
column lists the corresponding measure numbers, and the right column lists the basic
harmonic motion of each formal unit. A copy of the score with analytic additions—
identification of formal units, important motives, hypermeter, and, where appropriate,
harmony—is provided in Example 6.3. We will refer to both Figure 6.3 and Example 6.3
throughout the following discussion.

Exposition
The first theme consists of two phrases with distinct melodic content (labeled a and b); both
phrases end on the dominant.9 As a whole, these two phrases create the impression of a very
large antecedent, thereby giving rise to the expectation that a consequent will follow. Indeed,
the opening idea does return (m. 31), but at a different pitch level (A), which leads to the
dominant of the secondary key (III) at m. 39. The expected consequent is instead the
transition to the secondary key! It is common in Classical sonata form for the transition to
begin with the same melodic content as the start of the first theme (though usually at its
original pitch level). Since the expected consequent does not materialize, this means that
theme 1 ends with a half cadence. Unlike second themes and closing themes, which invariably
end with perfect authentic cadences, a first theme may end with either type of authentic
cadence or with a half cadence. Compared to Haydn and Mozart, Beethoven was fond of
ending first themes with a half cadence, a choice that often contributes to a feeling of agitation
or breathlessness, as it does here.
212 Analysis

EXPOSITION
Theme 1
phrase 1 (a) 1–10 i–V
phrase 2 (b) presentation 11–18 i–V
continuation 19–30 V
Transition
phrase 1 (a′) 31–39 E (III): IV – V
cadential expansion
(based on b) 39–58 V
Theme 2
presentation by piano 59–67 I
presentation by strings 67–75 IV
continuation 76–98 IV – V – I
Closing section
closing theme (initially
based on a) 98–(110)–124 I – I6 – II6 – V– I
codetta (based on b) 124–137 I

DEVELOPMENT
development of a 138–176 I – VI – ii/iv
development of b 176–198 c(i): iv – V
retransition 198–213 V

RECAPITULATION
Theme 1 – Transition
phrase 1 (a) 214–223 i–V
phrase 2 (a″)/transition 224–242 I – II6 – V
cadential expansion
(based on b) 242–261 V
Theme 2
presentation by piano 262–270 i
presentation by strings/piano 270–278 i – VI – iv
continuation 279–301 iv – V – i
Closing section
closing theme (initially based 301–(313)–327 i – i6 – II6 – V – i
on a)

CODA (based on b) 327–360 i

FIGURE 6.3 Outline of Beethoven, Piano Trio, op. 1, no. 3, I

213 Classical Sonata Form

EXAMPLE 6.3 Beethoven, Piano Trio in C Minor, op. 1, no. 3, I

214 Analysis

EXAMPLE 6.3 continued

215 Classical Sonata Form

EXAMPLE 6.3 continued

216 Analysis

EXAMPLE 6.3 continued

217 Classical Sonata Form

EXAMPLE 6.3 continued

218 Analysis

EXAMPLE 6.3 continued

219 Classical Sonata Form

EXAMPLE 6.3 continued

220 Analysis

EXAMPLE 6.3 continued

221 Classical Sonata Form

EXAMPLE 6.3 continued

222 Analysis

EXAMPLE 6.3 continued

223 Classical Sonata Form

EXAMPLE 6.3 continued

224 Analysis

EXAMPLE 6.3 continued

225 Classical Sonata Form

EXAMPLE 6.3 continued

226 Analysis

We will point out two features of the initial phrase (mm. 1–10). On first hearing it
sounds like an introduction, given its slow rhythms and ending on an embellished dom-
inant with fermata. In other words, the second phrase (mm. 11ff.) sounds like the start
of theme 1. However, motive a returns at the beginning of both the development
and recapitulation (as well as to initiate the transition and the closing section), and thus
mm. 1–10 really are an integral part of theme 1. Second, note that the opening phrase’s
length results from the varied repetition of mm. 3–4 a step higher, thus expanding an
eight-measure phrase to ten. An underlying quadruple hypermeter, then, is present, and it
will persist as a surface hypermeter throughout much of the movement.
The second phrase (mm. 11–30) is a sixteen-measure sentence whose continuation is
expanded by four measures through the extensive repetitions of the goal dominant harmony.
The phrase opens with a five-note descending figure (motive b), which is employed
extensively at various points throughout the movement. Here its statements articulate the
prolongation of E5 by its upper neighbor note, shown by the large slur in Example 6.3. In
mm. 17–18, the melody progresses to the covering third above, G5, thus preparing the
continuation, which prolongs the dominant. The continuation’s lack of harmonic motion
and its progressive rhythmic intensification resemble the end of a transition more than the
end of a first theme. This impression is furthered by the rests at the end of the phrase. The
keen listener knows that the prolonged G major harmony cannot prepare a second theme
in E major, but it seems that Beethoven is trying to unsettle the listener’s evolving sense of
the exposition’s formal design.
As noted above, the transition to theme 2 opens with a statement of motive a, beginning
with an A chord, which is the pivot in the modulation to E (III): VI/IV. This prominent
placement of A recalls the expansion in mm. 5–6 of theme 1. Since the transition begins
on A, its fifth and sixth measures introduce the Neapolitan sixth chord of E major, a striking
sound that is exploited later. After a passing secondary diminished seventh chord, the
dominant of the new key arrives and is expanded across mm. 39–58. Beethoven thus
“outdoes” the dominant expansion from the end of theme 1 when he has arrived at the
proper dominant to prepare theme 2. This post-cadential expansion divides into two parts,
the first (mm. 39–46) based on multiple statements of motive b. As in the transition from
the Mozart movement, modal mixture temporarily darkens the color until theme 2 arrives.
Unlike the Mozart, the transition ends softly and is not followed by any rests. Note that
Beethoven easily could have introduced rests immediately before theme 2 by using a gesture
like the last two measures of theme 1. (Imagine transposing mm. 29–30 to B and substituting
that in place of Beethoven’s mm. 57–58.) Instead, he fills in this juncture with two measures
that prepare the mode and mood of theme 2, possibly because he had deployed such a decisive
break after theme 1 and wanted to ensure that the exposition did not become too
sectionalized. The hypermeter further contributes to the recessive quality at the end of the
transition. Ever since the second phrase of theme 1, quadruple hypermeter has been strongly
projected, but this is not the case in the second part of the transition’s post-cadential
expansion. Measures 47–50 and 51–54 are hypermeasures, but the beginning gesture of mm.
47 and 51 does not recur at m. 55. Instead, there is a drawing out of the hypermeasure (to
eight measures), coordinated with the reduction in dynamics and rhythmic activity.
The second theme (mm. 59–98) is a massively expanded sixteen-measure musical
sentence, clearly divisible into various components. It opens with a four-measure basic idea
227 Classical Sonata Form

in the piano that is then repeated. This is followed by a further presentation of the same
idea by the violin a fifth lower, concluding with the subdominant harmony in m. 75.
Sentences usually begin with two statements of the basic idea, but this one has four, although
the dialogue between piano and violin gives the overall impression of binary grouping. In
works such as duo sonatas and trios, melodic ideas (and even complete themes) are frequently
repeated more than one would find in a solo sonata (or in a symphonic work) in order to
give more members of the chamber ensemble the opportunity to shine. The sentence’s
continuation spans twenty-three measures (mm. 76–98), a length even greater than the sixteen
measures of the expanded presentation. The vast size of this phrase builds up tension for the
perfect authentic cadence that will close theme 2 and confirm the secondary key. Beethoven’s
calando marking in the middle of the continuation adds to the suspenseful effect. The
continuation prolongs the subdominant, first IV, then its modal equivalent, iv, until the
dominant is reached in m. 92. Overall the harmonic motion of this sentence is I(6) – IV(6)
– V – I.
The hypermetric organization of Beethoven’s second theme is interesting. Due to the
arrival of E major in the first measure of the second theme (m. 59) and the hypermeter of
the preceding transition, m. 59 seems to be hypermetrically strong. Yet the melody enters
after the downbeat with an upbeat figure that leads to the next downbeat, whose strength
is further projected by the trill. This suggests that m. 60 initiates the theme’s hypermeasures,
and this is why we have designated both mm. 59 and 60 as hypermetric downbeats. The
most common scenario for consecutive hypermetric downbeats is, in fact, exactly this one—
an accompaniment pattern launches a hypermeasure and the melody enters later and
reorients the hypermeter. This one-measure hypermetric shift in mm. 59–60 is confirmed
by the start of the continuation at m. 76 (not at m. 75).
A representation of theme 2 is provided in Example 6.4. It is most significant, then,
that the piano enters fortissimo on a weak measure (m. 91) as if to shift back to the original
hypermeter, but without success. As far as tonal contents are concerned, the initial four-
measure idea elaborates a descending third B5–A4–G4 progressing in parallel tenths with
the bass. This idea is subsequently repeated a fifth lower (violin), E5–D4–C4, supported
by subdominant harmony. As shown in Example 6.4, there is a voice exchange between
the outer parts leading to m. 76, the beginning of the continuation. Not shown here, but
clearly marked on the score, are the subsequent voice exchanges prolonging the subdominant
harmony. In the final approach to the dominant, the E5 is picked up once again, which
then leads to D5 supported by V in the local key at m. 92. Measures 92–97 provide an
interesting study in levels of harmonic activity. It is possible to label every chord in this
succession, but at the largest level they all fall within the dominant scale-step. Looking at
the overall picture, there is no descent of a fifth from the initial B5, as one might expect,
given that themes ending with a perfect authentic cadence often feature an embellished
stepwise descent to that melodic goal. This fifth will occur later in conjunction with the
closing theme.
A simplification of the closing theme (mm. 98–124) is provided in Example 6.5. The
theme’s initial portion is a variant of the 5–6 sequence. After the fifth measure, this line,
which enters on the third beat of the measure, disappears, leaving the inner voice, which
proceeds in an 8–10 pattern with the bass. The end of the sequence and the change in
rhythmic texture articulate the vi chord in m. 106 as a tonal goal, but it initiates a linking
228 Analysis

EXAMPLE 6.4 Voice-leading representation of mm. 59–98

EXAMPLE 6.5 Voice-leading simplification of mm. 98–124

229 Classical Sonata Form

passage to I6, the real goal, in m. 110. As shown in Example 6.5, this passage prolongs E5
while the inner line traverses an entire octave from E4 to E5. The final portion of this
theme then opens with a four-measure idea (mm. 110–113) that decorates E5 by its
chromatic upper neighbor F5, harmonized by the Neapolitan sixth chord, recalling its use
near the start of the transition to prepare V of the secondary key. In the varied repetition
of this idea (mm. 114–117), the violin line reaches up to B5 and from there descends by
step to local closure, finally articulating the descending fifth that had not occurred earlier
within theme 2. At the last minute, closure is thwarted by the deceptive harmonization of
E5 by VI, which initiates a six-measure parenthetical insertion utilizing once again the
characteristic sound of the Neapolitan sixth chord. As indicated in Example 6.5, this
parenthetical insertion is itself an expanded four-measure unit (based directly on mm.
110–113), resulting from the extension of the melody’s second and third pitches. Like the
second theme, then, the closing theme ultimately consists of a single tonal motion—in our
terminology, a single musical phrase.
The codetta is based on varied statements of motive b by all three instruments (indicated
on the score by brackets). The return of both motive a and motive b during the closing
section provides an immediate feeling of unity for the exposition as a whole and also highlights
the change in expressive character experienced during theme 2.

Development
The development opens with an abbreviated statement of the opening phrase from theme
1 that leads from the parallel minor of III (E) to initiate a second statement of this idea in
B major, enharmonically C major or VI in relation to E. It soon becomes apparent that
this is only an intermediate goal between E and the C7 chord—V7 of ii—which is reached
in m. 152 (ff). This harmony is then extended over the next twenty-four measures until
it resolves to the F minor chord in m. 176. This is the pivot in the return to C minor: ii in
E = iv in C minor. The next section of the development deploys motive b in all three
instruments. The first presentation is in F minor (iv), the second in A (VI), to which the
lowered seventh is added in m. 191. This harmony has the potential to become V7 of II,
but instead Beethoven rewrites the G as F, transforming the chord built on A into the
German augmented sixth chord, which resolves to V in m. 198. Overall mm. 176–197 extend
the subdominant harmony, changing it from stable triad to a chromatically altered harmony,
the augmented sixth chord, requiring resolution. The extended A – A7 section beginning
in m. 184 functions within this larger prolongation of the subdominant as a composing-
out of the triad’s third. In characteristically Beethovenian fashion, the goal dominant from
m. 198 launches a sizable retransition (mm. 198–213). Recognizing the development’s
main harmonic arrivals reveals its contribution to the movement’s tonal motion thus far:
i – III – iv – V.10

Recapitulation
Our investigation of the recapitulation will be relatively brief, focusing on areas where it
differs from the exposition. The first change comes immediately after the first phrase, where,
instead of motive b, we get a lengthy phrase based on a. This phrase opens with a statement
230 Analysis

of the first two measures of a by the cello in C major. The bass line descends almost
chromatically to A, which supports varied statements of this motive by the violin in D (II),
another manifestation of this movement’s fascination with the Neapolitan. We might
expect the A to continue its descent to G; the A7 harmony at m. 236 could have been
enharmonically reinterpreted as a German augmented sixth, as it was immediately
before the retransition. Instead the bass ascends chromatically to C followed by F–F–G.
A representation of the voice leading of this passage is provided in Example 6.6. At the
deepest level it shows the extension of tonic harmony and scale degree 3, first E4, then E5,
by its upper neighbor F5. At the cadence in mm. 240–242, E5 descends to D5 and on to
the third below. Scale degree 2 is then transferred to the upper octave (D6) for the initial
statement of motive b by the piano at the outset of the post-cadential expansion.

EXAMPLE 6.6 Voice-leading representation of mm. 224–242

There are two important consequences of Beethoven’s extensive recomposition of the

early stages of the recapitulation. First, the quick departure from literal recapitulation and
the tonal remoteness of the Neapolitan provide a developmental quality. It is as if the impulse
for exploration has carried over from the development proper into the recapitulation. There
are many instances in Classical sonata-form movements where something “developmental”
occurs early on in the recapitulation. The second result of Beethoven’s recomposition is a
fusion of theme 1 with the transition. We are waiting for the second phrase of theme 1, but
it never comes. Instead, the new developmental phrase leads to the dominant, which receives
the same post-cadential expansion as did the dominant of the secondary key at the end of
the exposition’s transition. Fusion of first theme and transition into a single section in the
recapitulation occurs quite often in the repertoire. This tactic avoids repeating some of the
material that had already occurred in the tonic key in the exposition. Moreover, in
movements that close the first theme with a perfect authentic cadence, it eliminates that
cadence, which would usually mean that the close of the second theme provides the
recapitulation’s first perfect authentic cadence in the tonic key.
The second theme and closing theme are much the same as in the exposition, except,
of course, they are in the tonic key rather than III. Only in the final measures of the closing
theme—after the thwarting of closure by the deceptive progression—is there substantive
231 Classical Sonata Form

change. Compare mm. 118–124 of the exposition with mm. 321–327 in the recapitulation.
In the exposition, recall that the melody’s second and third pitches were extended by a
measure each. In the recapitulation, the third pitch is further elongated and when it does
move it incorporates a seemingly innocuous embellishment in the violin part (mm. 326–327).
The new embellishment E5–D5–C5 provides a link to the ensuing statement of motive b
at the outset of the coda. The coda’s prominent use of motive b over tonic harmony is
particularly satisfying since this was precisely the thematic element largely absent from the
recapitulation. As noted above, the recapitulation omits the second phrase of theme 1, and
the codetta is discarded as well. Motive b is only present in the post-cadential expansion
before theme 2, and there it expresses dominant rather than tonic harmony. Notably, the
coda ends with the same material that closed the exposition’s codetta. As a result, the coda
is tightly integrated with the recapitulation, seeming almost like a wonderfully inventive
expansion of the codetta rather than a thoroughly new section.
Many commentators have written about Beethoven’s “C minor style,” especially in
regards to his “Pathétique” Piano Sonata, Third Piano Concerto, and fate-knocking Fifth
Symphony. These and other of Beethoven’s C minor works have a particularly urgent
dramatic quality, and writers throughout the eighteenth and early nineteenth centuries opined
extensively about the expressive qualities of the different keys (although, by no means, did
they agree!). Our piano trio movement is Beethoven’s first published work in C minor, and
it certainly does not contradict the notion of a dramatic C minor style. Many features discussed
in our analysis—such as the sometimes obsessive reiterations of motive b, the premature
transition rhetoric at the end of theme 1 in the exposition, the vast single-phrase designs of
the second and closing themes, and the recomposition of theme 1 in the recapitulation—
play important roles in creating the movement’s bold expression.
232 Analysis

SUGGESTED ASSIGNMENTS
The ultimate goal of analysis is to identify what is special or unique about a particular work.
To that end, we suggest you proceed through the following steps:

1. Identify the main formal divisions and subdivisions (themes, transitions, closing sections,
etc.).

2. Identify important motives that are repeated/developed in the course of the movement.

3. Examine features of the phrase rhythm, that is, the interaction of phrase design and
hypermeter.

4. Identify important features of the overall tonal organization and details of the harmony,
particularly in more complex passages. You may want to prepare a simplification of the
voice leading of certain passages, identifying harmony, hypermeter, important motives,
etc.

For each of the four pieces below, we have provided a few questions to point towards some
of the movement’s unique aspects. We encourage you to study the music on your own and
to complete the analytic steps listed above before considering our questions.

I. Examine one or both of the following movements in the minor mode.

Mozart, Piano Sonata in A minor, K. 310, I

1. The first theme is divided into an antecedent phrase and a modulating consequent/
transition.
(a) What is Mozart bringing out in the calando measures (mm. 14–15)?
(b) How is the consequent/transition expanded?

2. Note that theme 2 begins with a four-measure idea that is subsequently repeated in
varied form and expanded. Examine this phrase carefully.

3. What is the relationship between mm. 35–40 and 40–45?

4. Development section: Identify the various subdivisions, noting harmonic progressions

and larger goals. Specific questions:
(a) Why does Mozart change the spelling of the C7 chord of m. 56 in the next measure?
(b) Speculate as to why Mozart might have changed the spelling of the embellishing
diminished seventh chord in m. 55, replacing the D with C, especially considering
that the change makes no sense in the local context. The answer might have to do
with the harmonic goal of the phrase.
233 Classical Sonata Form

5. Consider changes made in the recapitulation compared to the exposition.

(a) How are the phrases beginning in m. 88 (the transition) and later in m. 121
expanded?
(b) What is/are the structural and/or expressive effect(s) of the latter phrase expansion?

Beethoven, Piano Sonata in C Minor, op. 10, no. 1, I

1. On what phrase design is theme 1 based? How has it been expanded?

2. Take a careful look at the transition to theme 2. How is the modulation to III
accomplished?

3. Theme 2, although lengthy, consists of a single phrase. Explain how Beethoven manages
to do this.

4. What is unusual about the development section? How do you explain B minor/D major
within the context of C minor? Stated differently, what is the controlling key of the
development section, and how does that relate to the overall tonal progression of the
movement up to the recapitulation?

5. What is unusual about the tonal progression of the recapitulation?

6. In our discussion of Beethoven’s Piano Trio, op. 1, no. 3, we mentioned that his C
minor sonata forms tend to be highly dramatic. Do you feel this is the case in this piano
sonata? By referring to specific elements in the music and your analysis of it, defend
your choice.

II. Examine one or both of the following movements in the major mode.

Beethoven, Piano Sonata in G Major, op. 14, no. 2, I

1. Comment on the treatment of the
24 meter in the opening measures.

2. Comment on the relationship of the outer lines in mm. 47–58. What implications might
this have for the hypermeter?

3. Trace the thematic origins of the development and identify its main arrivals (about four).
Then consider the following:
(a) Launching the development in the parallel mode of the tonic key is not uncommon.
Explain why it is so effective here. (What element of theme 1 might hint at G minor?)
(b) What trick is Beethoven playing on the listener in mm. 99–103?
(c) The retransition occurs in two segments. Identify these segments and explain how
the second one prepares for the recapitulation (in ways other than providing
dominant harmony).
234 Analysis

4. The coda includes a restatement of theme 1 in the tonic key. Explain how this
restatement is more “stable” than the original version of theme 1.

Beethoven, Violin Sonata in F Major, op. 24 (“Spring”), I

1. What is the formal organization of the first theme and transition? (It may be interesting
to compare your analysis to our discussion of the corresponding portion of Beethoven’s
Piano Trio, op. 1, no. 3.)
2. Upon what pattern is the harmonic progression of mm. 26–33 based? How does it
compare to the corresponding portion of the recapitulation?
3. On what phrase design is the second theme based? What melodic interval unfolds over
the span of this theme?
4. What is the function of the A major chord that frames the development section? Can
you explain the role of the harmony between these poles?
5. The coda, mm. 210–247, opens like the development section, except a fifth lower. What
is the function of the D major chord in mm. 210–213?
6. Explain the harmonic basis of mm. 222–227. What interesting relationship do you notice
between mm. 222–223 and 224–225?
7. In the second part of the coda (mm. 232–247), there is a shocking moment. Where is
this? Why is it so shocking, and how does Beethoven handle its harmonic resolution?
(And how is that harmonic resolution different than elsewhere in the movement?)
7 Sonata Form in the Nineteenth
Century

The Classical model of sonata form—as described in the previous chapter—persists well into
the nineteenth century, but at the same time certain changes also start to appear, beginning
with the middle-period works of Beethoven. Here we find all of the repeat signs disappearing,
leaving a clear ternary design, and an expanded harmonic language. We will encounter both
in the first work examined in this chapter, the first movement of Beethoven’s Piano Sonata
in F Minor, op. 57. Schubert’s music brings two significant innovations. The first, the three-
key exposition, expands the path from tonic to dominant in the exposition to include an
intervening key with its own theme. For example, the exposition from the first movement
of the String Quintet in C Major, D. 956, consists of three distinct tonal areas (C–E–G),
each with its own theme. This type of expansion of the dimension and tonal language of
first-movement sonata form influenced subsequent generations of composers.1
A more radical change, one that is peculiar to Schubert, is the replication of the
key scheme of the exposition in the recapitulation but transposed to end on the tonic. For
example, in the first movement of the Piano Quintet in A Major (“Trout”), D. 667, the
recapitulation begins in the subdominant; that is, I – V of the exposition is answered in
the recapitulation by IV – I. In the fifth movement of the same work, I – IV (exposition)
is answered by V – I (recapitulation). Most radical in this regard is the key succession in
the work’s second movement. An initial statement of three themes follows the key scheme
F major – F minor – D major, which, following a transition to G, is answered by A major
– A minor – F major! Schubert has been criticized for this procedure, which potentially
avoids rewriting the transition in the recapitulation to remain in the tonic key, though, in
fact, he rarely transposes the exposition without some rewriting. In his late works, Schubert
reverts to the standard practice of the tonic return at the outset of the recapitulation, yet
often this tonic becomes V of IV; the tonal structure remains innovative but basically within
the framework of the Classical model. This hybrid procedure occurs in the first movements
of works such as the Piano Trio in E Major, D. 929, and in the C major string quintet
mentioned above.
This brief survey is not intended to provide a comprehensive overview of innovations
in nineteenth-century sonata form—a huge topic—but rather to point toward two important
areas of change: the expansion of content to compensate for elimination of repeats and the
expansion of the tonal language. Above it was noted that we will begin our investigation
with the first movement of Beethoven’s Piano Sonata in F Minor, op. 57. We will then
examine the opening movement of another work in F minor, Brahms’s Clarinet (or Viola)
Sonata, op. 120, no.1.
236 Analysis

Beethoven, Piano Sonata in F Minor, op. 57, I

Beethoven’s Piano Sonata in F Minor, op. 57, is one of his compositions that has acquired
a nickname: “Appassionata.” It is a stormy work where—unlike in his Symphony No. 5 in
C Minor—the minor mode persists even throughout the last movement. Like the first
movement of the Fifth Symphony, the opening movement of the “Appassionata” is
dominated by the rhythmic motive of three eighth notes leading to a longer note, and its
obsessive use contributes greatly to the music’s power. In its initial utterance, this rhythm
accompanies the pitches C–D–C. Our analysis will reveal that the neighbor-note motive
C–D–C (and its major-mode counterpart C–D–C) appears in multiple other guises as
well, pervading the tonal landscape. We will focus on this germinal idea in our commentary.2
In terms of overall formal design, the repeat of the exposition has been eliminated, and
there is a substantial coda that significantly extends the movement. Before examining
this movement in detail, we will talk though its formal-tonal organization, a chart of
which is provided in Figure 7.1. The score with a few added notations is provided in
Example 7.1.
In its overall plan, the exposition follows a traditional path with theme 1 in the tonic
key and theme 2 in the relative major. The first theme is divided into two phrases, an
antecedent that leads to the dominant and a modulating consequent that takes us to the new
key (III). The second theme consists of a four-measure idea followed by its varied and
extended restatement that changes to the parallel minor of III. The closing theme extends
this modal change.3 The development is divided into two parts, the first based on theme 1
and the second on theme 2. Part 1 opens with a statement of theme 1 in E major,
enharmonically F major or VI in A (III), which is followed by a sequence that begins in
E minor and leads through C minor back to A, now A7 or V7 of D (VI in the tonic key).
This passage looks complicated, but once we examine what Beethoven has written, we realize
he has prolonged the key of the relative major by a chain of descending major thirds, A –
E/e – c – A, and the return to A becomes V of D. This leads to part 2 of the development,
which opens with a sequence based on theme 2 that continues the pattern of descending
thirds leading from VI through iv to II in the tonic key. This leads us to the dominant and
preparation for the return.
The recapitulation follows the order of events as they occurred in the exposition with
the transition to theme 2 rewritten to remain in the tonic key. Theme 2 begins in the parallel
major, but then reverts in the varied restatement to the minor mode. The closing theme
now emphasizes II (rather than II in A), an important feature of this movement from the
very beginning. This leads to a weak close in the tonic in m. 203. More is needed to bring
the movement to a satisfactory conclusion. What follows is not just a closing phrase or codetta,
but an extended coda, really a second development section. It opens with a phrase based
on theme 2 that leads us to II6 in m. 218. This coincides with a change of surface pattern
(arpeggios) which progresses from II6 to iv6, which is extended, and leads eventually to the
6–7
cadential 4–3 and a perfect authentic cadence (m. 240). This is followed by a closing phrase,
initially based on theme 2, which not only confirms closure, but in the end also allows the
built-up energy to dissipate.
237 Sonata Form in the Nineteenth Century

EXPOSITION
Theme 1
antecedent 1–16 f: i – V6
modulating consequent/ 17–34 i – [V7] III
transition
Theme 2
statement 35–39 A (III): I
varied statement 40–51 I/i
Closing section

closing theme 51–65 i5– 6 – II – V7 – i
bridge to development 65–66

DEVELOPMENT
part 1 (based on theme 1)
theme 1 67–78 VI (E)
sequence 79–93 vi – iii – I

extension 94–108 I 7 (=V7 of D)
part 2 (based on theme 2)
sequence 109–123 f: VI – iv – II
o
retransition 123–135 7 – V7

RECAPITULATION
Theme 1
antecedent 136–151 f: i – V6
consequent/transition 152–173 I – V7
Theme 2
statement 174–178 I
varied statement 179–190 I/i
Closing section
closing theme 190–203 i5–6 – II – V7 – i

CODA [=second development]

phrase 1 (based on theme 1) 203–210 f: I – VI
phrase 2 (based on theme 2) 210–217 VI – i6
cadential buildup 218–239 II6 iv6 [o65] – V – i
closing phrase 239–262 i

FIGURE 7.1 Outline of Beethoven, Piano Sonata, op. 57, I

238 Analysis

EXAMPLE 7.1 Beethoven, Piano Sonata in F Minor, op. 57, I

239 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

240 Analysis

EXAMPLE 7.1 continued

241 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

242 Analysis

EXAMPLE 7.1 continued

243 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

244 Analysis

EXAMPLE 7.1 continued

245 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

246 Analysis

EXAMPLE 7.1 continued

247 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

248 Analysis

EXAMPLE 7.1 continued

249 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

250 Analysis

EXAMPLE 7.1 continued

251 Sonata Form in the Nineteenth Century

EXAMPLE 7.1 continued

252 Analysis

EXAMPLE 7.1 continued

253 Sonata Form in the Nineteenth Century

Exposition
A graphic representation of theme 1 and the opening statement of theme 2 is provided in
Example 7.2. The antecedent phrase, a sixteen-measure sentence, opens with a four-measure
idea that progresses from i to V6, which supports the neighbor-note figure C5–D5–C5.
The repetition of this idea occurs a half-step higher, establishing from the very beginning
an emphasis on II and on D, the upper neighbor of scale degree 5. This is followed by an
eight-measure continuation that begins with a restatement of the C5–D5–C5 figure, thus
articulating a larger pattern of C–D–C over mm. 1–10, shown in Example 7.2 by the larger
slur. Motion from D–C is thereby established as a primary motivic component of the
movement, as is the interchange between D and D. Immediately after restating C5–D5–C5
in mm. 9–10, we hear three statements of D2–C2 in the bass in the pervasive rhythmic
motive mentioned above. Finally, at the end of the phrase D4–C4 is harmonized by VI6 –
V6, a non-cadential progression. This is one of the rare instances where the overall
organization of melodic material—and the long chord and fermata in m. 16—clearly
indicate the end of a phrase despite the absence of a normative cadential progression. The
pause on this tense first-inversion, rather than root-position, dominant harmony adds to the
suspenseful atmosphere.
The consequent opens with a varied restatement of the initial four-measure idea now
extended to six measures. The C5–D5–(C5) figure of mm. 21–22 is answered in m. 23 by

EXAMPLE 7.2 Voice-leading representation of mm. 1–39

254 Analysis

C5–D5, the latter now harmonized as the seventh of the dominant in the key of the relative
major. The resolution of this dissonant seventh and completion of the neighbor-note motive
will not occur until the start of theme 2 (m. 36). Meanwhile we have an eleven-measure
extension of the new dominant harmony, which is shown in Example 7.2 as a parenthetical
insertion. This passage emphasizes the minor mode, thus anticipating the closing theme in
A minor.4 As indicated by the upper-most line of letters above the staff in Example 7.2,
the top part of this insertion articulates the neighbor-note figure F6–E6 in mm. 26–27,
which is then repeated in mm. 30–31 and, following transfer to the lower octave, continues
its descent to C5 as part of theme 2 in m. 36. Not only is the resolution of the dissonant
D withheld until theme 2, but there is also linkage between the insertion and theme 2, as
this idea reappears as F4–E4–D4–C4 in mm. 37–39. Once again we have the interchange
between 6^ and 6^, now F and F in the key of A. And, in the restatement of the first four
measures of theme 2, the mode shifts back to minor precisely at the melody’s arrival on
scale degree 6 (m. 42)!5
A graphic representation of the closing theme is provided in Example 7.3. The initial
melodic gesture is the descending third C–B–A, which is then restated a sixth higher
(A–G–F in m. 52). This leads to D5, the upper neighbor note of C, first supported by
II in the local key, then, following transfer to the lower octave, as seventh of the dominant
resolving to C and tonic harmony in m. 55. As shown in Example 7.3, mm. 51–55 articulate
a modally altered statement of the neighbor-note motive from theme 1, as C–D–C in the
context of the new key. The expanded repetition of this four-measure idea is an octave
higher (mm. 55–61). The expansion comes at the end, where the upper neighbor, now D6,
is first extended (m. 59), and then led through C6 (m. 60) supported by [o7] V, to B5
supported by V, and finally to A5 for local closure in m. 61. This melodic descent is shown
in Example 7.1 by the addition of stems to the notes D6 (m. 59), C6 and B5 (m. 60), and
A5 (m. 61). The registrally expanded descending third occurring over the span of the closing
theme, shown in Example 7.3 by the large slur, is an expansion of the motivic gesture that

EXAMPLE 7.3 Voice-leading representation of mm. 51–61

255 Sonata Form in the Nineteenth Century

opens the phrase. The remainder of the exposition (mm. 61–65) extends the local A minor
tonic harmony, leading to an open octave on A.

Development
The development opens with a statement of theme 1 in E major, which we hear as VI (F)
in relation to A minor, the exposition’s closing key. Following the modal change to E minor
(m. 79), a sequence leads through C minor back to A (mm. 87–108), which, with the
addition of its seventh, G, leads to D major in m. 109 for the second part of the
development. Example 7.4 offers a graphic representation of the underlying structure of the
entire development. The purpose of this type of graph—as opposed to ones like Examples
7.2 and 7.3—is to show underlying connections; this necessitates registral simplification and
omission of detail. From a melodic perspective, C within the exposition’s closing A minor
becomes B4, scale degree 5, in E major, which is prolonged throughout the development’s
initial phrase and retained as top note at the modal change to E minor (m. 79). As we progress
from m. 79, the beginning of the sequence, the main melodic line is no longer emphasized,
but rather is buried within the arpeggios. In fact, by the time Beethoven has reached the
A7 chord, the emphasis has clearly shifted to the seventh, G (see esp. mm. 91–93), which
is represented in Example 7.4 as coming from an inner voice. The entire passage from mm.
94–108, represented in our abstract sketch by parentheses, extends this dissonant harmony
until its resolution in m. 109. This parenthetical passage is directly based on the extension
of V of III that prepared theme 2 in the exposition, except transposed down by fifth since
the imminent recall of theme 2 will be in the key of D (VI).6

EXAMPLE 7.4 Voice-leading representation of development

The second part of the development opens with sequential statements of the opening
of theme 2 that progress from VI (m. 110) through iv (m. 114) to II (m. 118), the latter
two arrivals approached through deceptive progressions (i.e., as vi of the preceding key).
Up to this point, the hypermeasures are all quadruple, but the connection between II (which
is subsequently respelled enharmonically in m. 119) and the diminished seventh chord with
256 Analysis

D4 as the lowest note (m. 123) requires an additional measure, as the motion is achieved
through an ascending 5–6 sequence. Note that this dramatic arrival on the diminished
seventh, which launches the retransition, completes a two-octave ascent in the bass from
D2 (m. 109) to D4 (m. 123). The following arpeggios leading to m. 130 are based entirely
on this one diminished seventh chord. Beginning in m. 130 the right-hand part holds its
position, while the left hand crosses over, first stating D in the same octave position as in
m. 109 (D2), then D5, then D2–C2 and D5–C5 in the primary rhythmic motive. In
Example 7.4, we suggest large-scale connections involving these D2–C2 motions in both
of the structural outer voices.

Recapitulation
Beethoven follows Classical procedure by rewriting the transition to theme 2 to remain in
the tonic key. One result is that the neighbor-note motive of this passage, which in the
exposition was F–E, is now at the original pitch level: D–C. Statements of this idea in
mm. 165–166 and 169–170 are marked in the score by brackets. This passage is followed
by theme 2, the initial statement in F major and the extended restatement returning to the
minor mode in preparation for the closing theme.
A graphic representation of the closing theme is provided in Figure 7.5. In the initial
statement, scale degree 3 (here A) is prolonged by its upper neighbor (B), first harmonized
by II, then as seventh of the dominant. As in the transition, the transposition maps some
former F–E motions onto D–C; note especially the bass line in m. 193 (bracketed in
Example 7.5). Like at the end of the exposition, the repetition of this four-measure idea is
an octave higher and extended as the melodic descent to closure unfolds. The details are
slightly varied, as the continuation of B5 through A5 to G5 occurs in the left-hand part,
temporarily covered by the figuration above. The recapitulation, though, ends as did the
exposition with a few measures of tonic harmony (mm. 200–204 = mm. 61–65)—except
instead of the open octave sustained throughout m. 65, the momentary F octave at the start
of m. 204 immediately launches the coda.

EXAMPLE 7.5 Voice-leading representation of mm. 190–200

257 Sonata Form in the Nineteenth Century

Coda
A graphic representation of the voice leading of the coda is provided in Example 7.6. The
coda opens with an eight-measure phrase based on the ascending arpeggiation portion of
theme 1. The phrase leads from i to VI, above which the top part articulates the neighbor-
note motive F6–G6–F6. This is followed by an extended statement of theme 2 that leads
us to II6 in m. 218, whose arrival is marked by a change in surface pattern (sixteenth-note
broken chords). From this point the harmony progresses through a passing 63 chord in m.
220 to the iv6 chord in m. 222, whereupon parallel 63 chords traverse two octaves to the iv6
in m. 227, an arrival also marked by a change in surface pattern. The top voice soars from
B4 to B6 over this extended subdominant harmony with D, the pervasive upper neighbor
of C, as the lowest note. As in the development, the bass specifically traces the stepwise
ascent from D2 to D4. This D4 (m. 227) progresses to D4 (m. 229) on its way to C (m.
231), a reference to the interchange between D and D as neighbors to C from the very
beginning of the movement.7 The harmonic progression supporting this motion is iv6 – [o65]
– V. Meanwhile the top voice descends by step B6–A6–G6, but this can be interpreted as
a motion to an inner voice. B, now the dissonant seventh of the dominant harmony and
an octave lower (B5), is still very much in evidence. It is extended over the next four
measures, while the bass re-articulates the D–C motive four more times (mm. 235–238).
This dissonant seventh then resolves to A5 in m. 239 covered by F6.8

EXAMPLE 7.6 Voice-leading representation of mm. 203–249

258 Analysis

Though the cadence at the più allegro ends conclusively with F6 in the top-sounding
part, we must be aware of the importance of A5 as the resolution of the preceding dissonant
seventh. Indeed A5 emerges as primary in the following altered statement of theme 2, which
leads to B5 supported by subdominant harmony in m. 244 (= m. 247), and from there the
top voice descends by step through A5 to G5, supported by V6–5 4–3, and then to an implied
F5 over tonic harmony in m. 249. This implied F5, which completes the descent to closure,
is immediately covered by scale degree 5. In the following measures (not shown in Example
7.6) there are several references to the D–C motive, and, as the movement comes to an
end, C4, then C5, persists until it fades away entirely.
Despite its length, this movement displays a characteristically Beethovenian economy
of material. We have especially focused on the deployment of neighbor-note motives,
particularly the use of scale degree 6 (usually 6^) as an upper neighbor to scale degree 5,
either in relation to the tonic F or to local tonics. In the development, we also noted the
prevalence of modulation by falling third. These two ideas are not unrelated: in F minor, a
5–6 motion exchanges F–A–C with F–A–D, or in harmonic terms an F chord (or key)
with the chord (or key) a third lower. Sometimes this interrelationship of linear and harmonic
entities is explicit, as in the closing theme. In the exposition, a 5–6 motion in the first two
measures of the closing theme generates the pitches of F major, which (notated as E major)
becomes the first key area of the development. In the recapitulation, the same 5–6 move—
but now in relation to F—brings forth the pitches of D major, which turns out to be the
first key area of the coda. Elsewhere the linear 5–6 motion is less explicit on the musical
surface, but with falling-third modulations an element of 5–6 motion remains embedded in
the underlying voice-leading structure.9

Brahms, Clarinet (or Viola) Sonata in F Minor, op. 120,

no. 1, I
With this Brahms movement we are entering a new sound world. In contrast to the expansive
dramatic works of Beethoven, Schubert, and even earlier Brahms, this late work is compact,
devoid of extensive repetitions. There is also a clear difference in Brahms’s harmonic language,
which is manifest at times in chord-to-chord syntax as well as the means by which changes
of tonal center are achieved. Brahms’s harmony is sometimes difficult to understand,
particularly when viewed in light of the clarity of the Classical style. It is at times Classical,
at times decidedly non-Classical, but always logical. Once we unravel what is taking place
in this movement, we will discover that it is a musical essay on the linear motion 5–6, which
pervades many levels of the structure.
An annotated score is provided in Example 7.7 and an outline of formal-tonal
organization in Figure 7.2. Following a precedent established by Schubert, the exposition
consists of three themes in three different keys. Here, however, the second key is very
dependent on the first key and is not confirmed with a cadence. To varying degrees,
downplaying of the second key is Brahms’s standard practice in his three-key expositions.
Why then do analysts refer to such structures as three-key expositions? The answer lies in
a subtle shift in the conception of sonata form during the nineteenth century. In the
eighteenth century, sonata form was defined by tonal structure, namely the polarity between
the two keys presented in the exposition. Haydn occasionally reused his first theme as the
259 Sonata Form in the Nineteenth Century

EXPOSITION
“introduction” (pno.) 1–4
Theme 1 (cl.) 5–12 f: i – V
continuation 12–24 V (VI6 – iv6 – II6)
Theme 1 restatement (pno.)
→ Transition 25–37 i – V – [V7]
Theme 2 38–52 D (VI): I6 = c: II6
Theme 3
statement 1 (pno.) 53–56 c (v): V – i – [V] V
statement 2 (cl.) 57–60 V – i – iiø65
augmentation (pno.) 61–68 V – i – [V] V – V
continuation 68–76 V
Closing section
closing phrase 77–81 i – iiø65 – V7 – i
expanded repetition 81–89 i – iiø65 – V7 – i

DEVELOPMENT
phrase 1 (based on theme 2) 90–100 A (VI of c): I6 – ii65 – V – i
phrase 2 (based on theme 2) 100–115 E (VI of A): I6 – [iiø7 – V7] vi
phrase 3 (based on theme 3) 116–130 c (vi of E): i – to f (ii of f)
retransition (based on theme 1) 130–137 f (ii): i – V6 = f: i6–5

RECAPITULATION
Theme 1 138–145 f: i – V
dominant extension 145–152 V
Theme 2 153–167 I6 – i6
Theme 3
statement 1 (pno.) 168–171 V – i – [V] V
statement 2 (cl.) 172–175 V – i – iiø65
augmentation (pno.) 176–183 V – i – [V] V – V
continuation 183–191 V
Closing section
closing phrase 192–196 i – iiø65 – V7 – i
expanded repetition 196–205 i – iiø65 – V7 – i
closing phrase
(based on theme 1) 206–213 i – iiø65

CODA 214–236 V–i

FIGURE 7.2 Outline of Brahms, Clarinet Sonata, op. 120, no. 1, I

260 Analysis

EXAMPLE 7.7 Brahms, Clarinet Sonata in F Minor, op. 120, no. 1, I

261 Sonata Form in the Nineteenth Century

EXAMPLE 7.7 continued

262 Analysis

EXAMPLE 7.7 continued

263 Sonata Form in the Nineteenth Century

EXAMPLE 7.7 continued

264 Analysis

EXAMPLE 7.7 continued

265 Sonata Form in the Nineteenth Century

EXAMPLE 7.7 continued

266 Analysis

EXAMPLE 7.7 continued

267 Sonata Form in the Nineteenth Century

EXAMPLE 7.7 continued

268 Analysis

EXAMPLE 7.7 continued

269 Sonata Form in the Nineteenth Century

second theme, simply transposing it (the so-called monothematic exposition). During the
nineteenth century, thematic contrast came to be as important as tonal structure in defining
the thematic sections of an exposition. Thus, strong thematic contrast allied with a relatively
weak key change can suggest a three-key, three-theme model. In the clarinet sonata, Brahms’s
third key (C minor) is cadentially confirmed, providing the local tonal closure necessary at
the end of an exposition. Like the Classical model, there is a two-key polarity underlying
Brahms’s exposition, but there is also a new melodic idea in a different key wedged between
these tonal poles. Brahms’s handling of the three-key exposition can be viewed as a
consolidation of Schubert’s three-key innovation with the Classical two-key tradition. In
reading our analysis of this sonata, bear in mind that “theme 2” signifies something different
than what “theme 2” denotes in an eighteenth-century sonata exposition.
Brahms’s development section consists of four phrases, the first two based on theme 2.
The first phrase is in the key of A major, VI in relation to the dominant from the end of
the exposition. The second phrase begins in the key of E major, enharmonically VI of the
preceding A major, and it leads to C minor for the third phrase, which is an elaborate
combination of all three themes. This third phrase leads to F minor for the final phrase,
which is based on theme 1. These last two keys, C and F minor, enharmonically D and
G minor, vi and ii in relation to F minor, play important roles in this movement. This last
phrase is marked as a retransition, but this is not a retransition in the Classical sense of leading
back to the tonic through the dominant. There is no dominant, but rather, as we shall see,
a 6–5 linear motion, the reverse, in a sense, of the 5–6 motion that introduces the
development.
The recapitulation follows the sequence of events of the exposition, the main change
occurring in theme 2, which is now in the parallel major (F major). The next significant
change comes with the addition of a reminiscence of theme 1 after the closing phrase, which
leads to a beautiful coda, where the main melodic note of the movement, C5, persists,
growing ever fainter, but never relinquishing its position.

Exposition
A detailed representation of the voice leading of mm. 1–38, that is, up to theme 2, is provided
in Example 7.8. The sonata opens with a four-measure “introduction” stated by the piano
in open octaves, the main features of which are the decoration of C5 by its upper neighbor,
D5, and the subsequent descent by step to F4 via G4—not G4—immediately creating a
dark modal quality that characterizes much of the movement. The G and the extended
neighbor, D, become important features of this movement. We have placed “introduction”
in quotation marks because, although the piano’s hollow octaves of mm. 1–4 seem like a
preparation for the clarinet’s ensuing melody, they present thematic material that is integral
to theme 1.10 In fact, towards the end of theme 1 (mm. 19–24) the clarinet restates mm.
1–4 in their entirety, with only slight rhythmic variation. This tension over the formal
connectedness of mm. 1–4 with the subsequent music highlights the chromatic G4 of
m. 4. Although we understand this pitch to resolve to F4 in the piano accompaniment
in m. 5 (and have therefore drawn a large slur in Example 7.8 between C5 in m. 1 and
F4 in m. 5), there is a certain hesitancy to this resolution. This tentativeness is enhanced by
Brahms’s decrescendo in m. 4 and the delay of the piano chord in m. 5 until the second beat.
270 Analysis

EXAMPLE 7.8 Voice-leading representation of mm. 1–38

Not only do these musical details draw attention to the chromatic G4, signaling its
importance to the movement, but they also sow the seed for a recomposition of this passage
near the end of the coda.
The main feature of theme 1 is the descent of a fourth from C5, supported by tonic
harmony, to G4, supported by the dominant. It has already been noted that 5–6 linear motion
operates throughout the movement at multiple levels, almost as an Ur-motive. Looking now
to the details of the harmony and voice leading in theme 1, we see that its initial tonic
harmony is extended by a 5–6 linear motion, as is the following subdominant (5–6). As a
result, Brahms has immediately reintroduced the two important decorative pitches of the
introduction, D and G. Note that the following two chords—the C7 chord of m. 9 and
D chord of m. 10—have not been labeled. The problem with simply labeling them as V7
– VI is that this would suggest they function at the same level as the preceding subdominant
and following cadential pattern. Instead, they extend the subdominant until its chromatic
manifestation, V65 of V, is introduced on the downbeat of m. 11. This is an excellent example
of the distinction made in earlier chapters between chord and harmonic scale-step, which
may encompass many chords. The harmonic underpinning of this phrase is i – iv – V (mm.
5, 7, 12), which is expanded by 5–6 motions and passing chords.
The following four measures (mm. 13–16) extend the dominant, the clarinet melody
making clear reference to the neighbor-note pattern from the piano “introduction.” A varied
repetition of these four measures begins, but in the third measure (m. 19) Brahms introduces
271 Sonata Form in the Nineteenth Century

a D chord above F, which becomes the point of departure for the clarinet’s complete
restatement of the piano “introduction,” which now functions to close theme 1. What are
we to make of this D6 sonority? This is not D introduced by its dominant, but F with D
above, displacing C5. As indicated by the lowest level of Roman numerals in Example 7.8,
this sonority participates in a 5–6 motion that connects mm. 5 and 19. Looking ahead in
Example 7.8, we see that this motion is reversed between mm. 19 and 25, where we are
firmly back on tonic harmony for the seeming repetition of theme 1 that turns out to be a
transition. Note that in these final measures of theme 1 the harmonic progression outlines
descending thirds, “VI6” – iv6 – II6; thus, a direct progression from II6 to i—without an
intervening V—accompanies the tonal and thematic return at m. 25.11
The first eight measures of the transition basically follow the same path as mm. 5–12 of
theme 1, and thus their interpretation has not been written out in Example 7.8. This is
followed by a brief passage leading us back once again to a D chord over F, now with A
rather than D as the top note (m. 38). The initial melodic descent of the fourth C5–G4
(mm. 25–32) is shown to be embedded within the third C5–B4–A4 (mm. 25–38), where
B4 is supported by II6, by now a familiar sound, which is the pivot for the local modula-
tion to D (VI). As shown in the bass (stems up), there is an ascending line leading from
this bass B of this pivot through C, supporting V65 of VI, to D, but as an inner voice
above the bass F. Though the motion to D is stronger than before, Brahms leads us back
to the same bass note as before, F, and once again via G. For the first time in this movement,
the quadruple hypermeter is expanded by one measure (m. 36) to accommodate the
chromatic passing tone A4. This brief delay during the extension of V7 of D lends some
gentle support to the arrival on the D6 sonority of m. 38.
Example 7.9 is an abstract representation of the structure of this movement up to theme
3. We have already discussed mm. 1–25, which overall present the voice-leading motion
5–6–5 over F. This motion is embedded within a larger version of the same progression
leading to theme 2. This is what was meant earlier by the statement that this 5–6 idea pervades
deeper and deeper levels of musical organization. Though we can speak of a modulation to
D in a local sense, D is never stable or cadentially confirmed. As shown below Example
7.9, this D6 chord ultimately becomes the pivot in the modulation to the dominant, the
key of theme 3.

EXAMPLE 7.9 Tonal summary of exposition

272 Analysis

We have not provided a representation of themes 2 and 3; instead refer to the annotated
score (Example 7.7). The rhythmic drive of theme 3, with its repeated rhythmic pattern
, provides a contrast to the legato-like and registrally expansive first theme.
The initial statement of theme 3 is shared by clarinet and piano, but then the clarinet
predominates. This is followed by an augmentation of the rhythmic pattern in the piano12
above which the clarinet begins an elaborated descent from the sustained G5 in m. 61 directed
toward closure in the local key. Though this line concludes on C5 in m. 67, the harmony
is directed beyond this point to the dominant in the next measure, preceded by the descent
in octaves from A5 to D5. The following elaboration of the dominant begins with an
augmentation of this descent in the piano part, imitated two beats later by the clarinet (mm.
68–69). This gesture is then stated a step lower beginning from G5 (mm. 70–71), and the
imitating clarinet line once again reaches C5 on the downbeat of m. 72. Here, too, closure
is avoided by transforming the tonic into V7 of iv, where we now hear an elaborated descent
from this A5, the clarinet leading and the piano following (mm. 72–73). The descent once
again outlines the diminished fifth A5–D5, at which point the line skips to the lower octave
for the answering descent from G4 (mm. 74–76). After the first two steps in the piano descent,
Brahms separates the notes by rests, creating a written-out ritardando. If we expect finally to
hear melodic closure to C4 on the downbeat of m. 77, the beginning of the closing phrase,
we are once again disappointed.
The closing phrase occurs twice: in mm. 77–81 which elides with an expanded
repetition in mm. 81–89. A representation of the expanded version (mm. 81–89) and the
first measure of the development is provided in Example 7.10. Here we can see the melodic
emphasis placed on scale degree 2 in the local key, D5, first harmonized as V of V, then
later by its diatonic counterpart, iiø65. The long-awaited resolution to local scale degree 1
occurs in the clarinet part, as C4, at the end of the phrase. Note the connection to the
beginning of the development section via a 5–6 motion above the bass note C.

EXAMPLE 7.10 Voice-leading representation of mm. 81–90

273 Sonata Form in the Nineteenth Century

Development
A representation of the voice leading of the entire development is provided in Example
7.11. Before examining each of the four phrases in some detail, take note of the voice-
leading connection at both ends. We have just observed that the connection between
exposition and development is a 5–6 motion above the bass note C. The connection between
development and recapitulation reverses this motion, that is, 6–5, now over the bass
note F (mm. 136–138). In this instance the 6–5 motion is elaborated by an intervening
embellishing German sixth chord. The D6 chord (m. 136) is heard initially as V6 in the key
of F (G) minor.
The first phrase of the development is based on theme 2, and our representation of the
voice leading once again eliminates rhythmic displacements. The main feature of this phrase
is the descent of a fifth within the local key of A from the opening E4 via transfer to the

EXAMPLE 7.11 Voice-leading representation of development

274 Analysis

upper octave to G4 [A4] on the downbeat of m. 100. The transfer to the upper octave is
accomplished by repetition of the voice exchange of the third and fourth measures (the piano
part is an octave higher in mm. 94–95 compared to mm. 92–93). The repetition of these
measures is modally altered, incorporating F and thereby anticipating of the change of mode
at the end of the phrase (m. 100). This two-measure phrase expansion has been indicated
by the (3 4) between the staves on Example 7.11.
The second phrase presents a further expansion of the same idea (theme 2) a third lower.13
Once again Brahms has used the bass note as pivot to change the key. That is, the tonic
note of the previous key, A/G, is retained as the third of the new key, E major. This phrase
can be divided into two parts, the first leading from m. 100 (I6) back to the point of origin
(m. 107), although with a seventh added in the clarinet (D5). Note that this seventh chord
resolves to another seventh chord, part of a descending fifth sequence alternating between
4 6 
2 and 5 chords that breaks in m. 111. There is a registral connection between the bass G 2
at the outset of the sequence (m. 107) and the D2 (m. 112), which supports ii in the key
ø7

that is the goal of this phrase, C minor. Between these two landmarks, the bass had been
displaced upwards during the sequence, but in Example 7.11 we provide bass notes in the
lower octave parenthetically to show the implied stepwise connection in the bass as the
modulation from E major to C minor unfolds. Arrival at the C minor goal in m. 116
coincides with the completion of the melodic descent of a sixth E5–G4. Note that G4
(A4) was also the melodic goal of the development’s first phrase (m. 100).
The fact that this new key for phrase three (C minor) is stable—i.e., begins with a root-
position local tonic harmony—places it in association with the stable point of origin of this
motion, namely the C minor at the end of the exposition. The intervening keys of A major
and E major result from successive 5–6 motions; they are subsidiary tonalities between C
minor and C minor (=D minor), whose relationship may be considered a manifestation of
5–6 at a very deep level of structure. The third phrase opens with a bold four-measure idea
(mm. 116–119), marked forte and marcato, that is a thoroughly transformed version of theme
2—the dolce theme that has dominated the development thus far. As before, the melody is
split between the two instruments, but now the clarinet leads (E5–D5–C5) and the piano
provides the ascending chord (B4–F5–A5–G5). The passage sounds entirely different not
only due to the dynamics but also the plunge into minor mode and the altered scale-degree
placement of these melodic motives. The left hand of the piano maintains its earlier role,
but instead of moving in half notes it proceeds in quarter notes, thereby revealing a hidden
connection between themes 1 and 2. The bass motion in the first two measures of theme
2 was nothing other than an augmentation of mm. 1–2! Starting in m. 120, theme 3 provides
the thematic content for this C minor phrase, strengthening the connection between C
minor and C minor suggested above. The latter portion of the phrase focuses on the
descending fifth motive from the end of theme 3, and—as in the exposition—convincing
local melodic closure is hard to achieve. A stretto-like series of imitations of this motive
along a succession of secondary dominants leads to a momentary F minor chord at m. 126
with full stability arriving only at m. 130. We leave further details of the voice leading of
this phrase for you to explore with the aid of Example 7.11.
The development’s last phrase, beginning in m. 130, anticipates the thematic return.
The top part descends a fifth, C6–F5, then continues its descent, completing the octave.
This overlaps with an inner part that continues one step beyond the octave to resolve the
275 Sonata Form in the Nineteenth Century

D to C (6–5) over the bass note F. Despite the thick piano texture, these two lines emerge
due to the quasi-canonic relationship between them in mm. 130–131, an intricacy made
plain by Brahms’s careful upward stemming of the inner part. It was noted previously that
the D6 chord in m. 136 is, enharmonically, V6 in the key of F minor. We would also point
out that the bass motion leading into this chord—G–F–F in mm. 134–136—is
enharmonically equivalent to A–G–F, the pitch motion that led to the exposition’s
statements of theme 1 (i.e., mm. 4–5 and 23–25).
Example 7.12 is an abstract representation of key relationships in the movement so far,
eliminating transient tonal centers, such as D in the exposition and both A and E in the
development. The main keys of the exposition are F and C minor, and those of the
development are a half-step higher, D [C] and G [F] minor. These are the pitches that
were prominent in mm. 1–4!

EXAMPLE 7.12 Tonal summary of movement

Recapitulation and Coda

The changes in the recapitulation, beyond the use of more elaborate figuration in both the
clarinet and piano parts, are the following:

1. The extension of the first phrase of theme 1 is rewritten to lead directly to theme 2,
which is now in the parallel major. (There is no second statement of theme 1 by the
piano to launch a transition between themes 1 and 2.)
2. Themes 2 and 3 are in the tonic (F major, F minor).
3. There is a new closing phrase, based on theme 1, that begins in m. 206 and leads to the
coda.

A representation of this new closing phrase plus the opening measures of the coda is provided
in Example 7.13. The first four measures, which unfold over a tonic pedal, are perfectly
straightforward, presenting no analytic challenges, but the voice leading and particularly the
harmonic syntax of the next four measures are unusual. The succession of chords in mm.
210–213 is e6 – Cø7 – b6 – (Gmaj7 ) – Gø7, which then occurs in first inversion and resolves
as iiø65.This seems to make more sense if we understand this motion as occurring in relation
276 Analysis

to an implied bass B, as shown in Example 7.13, which is finally stated at the end of the
fourth measure (iiø65). In other words, this appears to be a variant of the opening statement
of theme 1 (compare mm. 8–11 and especially the triplet-embellished restatement in mm.
28–31). The recapitulation’s new closing phrase leads to the extended dominant harmony
at the beginning of the coda, above which the clarinet line descends a fifth once again to
closure, though immediately covered by the ever-present C5.

EXAMPLE 7.13 Voice-leading representation of mm. 205–219

The coda concludes with two sotto voce reminiscences of the piano “introduction.” We
observe something is absent—the chromatic G. In fact, neither G nor F makes a single
appearance, even as an embellishing tone, during the entire coda. This contributes to the
greater stability of mm. 227–236 compared to mm. 1–4, as do the presence of harmony
(rather than open octaves) and the conclusion with Picardy third. Although the end of the
coda strongly references mm. 1–4, thereby framing the movement thematically, it also alludes
to the final phrase of the development (esp. mm. 130–132). At the start of the F minor
phrase of the development, there was a quasi-canonic effect within the piano part, as the
F5–C6 fifth was answered by the C5–F5 fourth. This same answering of a fifth with a
fourth occurs at the end of the coda (m. 227, m. 231), suggesting again the central
importance of the F minor phrase to the development.
277 Sonata Form in the Nineteenth Century

SUGGESTED ASSIGNMENTS

Schubert, Piano Trio in B Major, D. 898, I

1. In Classical sonata form, the recapitulation is marked by a double return, that is, a
coordinated return of the opening theme and the tonic key. In this movement the
thematic return and tonal return do not coincide. What, then, are the possible
explanations for the thematic return in VI?
2. Consider these aspects of the exposition:
(a) How does the D major chord, mm. 18ff., function within the context of theme 1?
Likewise, what is the function of the A major chord, mm. 44ff., in the transition
to theme 2?
(b) In theme 1, note the use of A in m. 9. How does it function in this context? How
is this detail developed later in the movement? Hint: check out the coda.
(c) In theme 2, how does Schubert modulate to the dominant? What is the structure
of theme 2?
(d) What is the tonal organization of the closing theme?
3. The development section is divided into two parts followed by a retransition
(prolongation of V), then the thematic return in G (VI). Part 1, mm. 112ff., is based
on theme 1; part 2, mm. 139ff., is based on theme 2. Consider the following:
(a) What is the tonal organization of each part of the development?
(b) What is the combined tonal organization of these parts? Relate it to the coda.
(c) What is the function of the thematic return in VI in this larger scheme?

Brahms, Clarinet Sonata in E Major, op. 120, no. 2, I

This movement is as complex as the one we already examined in this chapter. Understanding
of its organization comes in many stages.

1. Step 1 is always to understand the formal organization at many levels. Where does the
development section begin? Is there a coda? If so, where (and how) does it begin? Then
determine the divisions of the major sections—themes, transitions, etc.—noting major
tonal connections.
2. Step 2 is to examine various sections in some detail, paying attention to phrase rhythm,
harmony, and important features. Here you may want to be selective, choosing specific
passages for careful scrutiny. Here are several suggestions:
(a) Consider mm. 1–10.
(i) Explain the harmony in mm. 5–8.
(ii) What is the function of mm. 9–10?
(iii) Note the use of G in m. 8 and again in m. 9. How is this detail developed
later in the movement?
278 Analysis

(b) The phrase in mm. 10–21 stops abruptly. What is Brahms doing here? How does
he accomplish the modulation to V? What happens at the corresponding spot in
the recapitulation?
(c) What is the relationship between the clarinet and piano left hand in mm. 22ff.?
Does this type of relationship occur elsewhere?
(d) What seems to be the function of mm. 26–27? In the next phrase as well, what is
going on with the phrase design?
(e) Explain the harmony and voice leading of mm. 44–52. This might require a simple
reduction.
3. More difficult to explain are the following and how they function within a broader
context:
(a) mm. 113–126
(b) mm. 150–166
(c) mm. 73–102
8 Ternary Form

Ternary form—normally designated A B A or A B A′—is based on the principles of

contrast and repetition. The portion designated B provides a contrast to A, and the second
A is either an exact repetition (written out or da capo) or a varied repetition (A′) of the first
part. In the Baroque period, ternary structures were frequently formed by pairing two
movements, e.g., Menuet I and Menuet II, where the individual movements were invariably
in binary form, each complete in itself. The larger ternary design resulted from repetition
of the first movement: Menuet I (A) – Menuet II (B) – Menuet I (A).1 The equivalent in
vocal music of the period was the da capo aria. This type of composite ternary form was
expressed in the Classical period as the Minuet – Trio – Minuet and later as the Scherzo –
Trio – Scherzo. As we saw in Chapter 5, the individual components were most always in
rounded binary form.
Though one can find examples of single movements in ternary form in the Classical
period (especially in slow movements), this is a form most often associated with character
pieces, particularly piano pieces, of the nineteenth century, no doubt because it is well suited
to the expression of strongly contrasting, often lyrical, ideas. There are many examples in
the piano music of Schubert, Schumann, Chopin, Mendelssohn and Brahms, to name a few.
Here we frequently find the ternary form expanded internally by a retransition leading back
to a varied repetition (A′) and externally by a coda, often combining elements from both
the A and B sections. In this chapter we will examine three representative works: the second
movement of Beethoven’s Piano Sonata in D Major, op. 10, no. 3, Schubert’s Impromptu
in G Major, op. 90, no. 3, and Brahms’s Intermezzo in B Minor, op. 117, no. 2.

Beethoven, Piano Sonata in D Major, op. 10, no. 3, II

We begin our investigation with one of Beethoven’s most tragic slow movements. This is
a highly dramatic work that expresses a wide range of emotional states from dark and
foreboding to lyrical but somber, from intense to tranquil. Overall it has a very serious quality,
only briefly providing a glimpse of a brighter world. A copy of the score with analytic
additions—indications of larger and smaller formal divisions, harmony, hypermetric
organization, and motivic repetitions—is provided as Example 8.1, and an outline of formal-
tonal organization is provided in Figure 8.1. We will begin by talking through the movement
in a general way, focusing on aspects of formal design and rhetoric. We will then examine
the movement from a more technical perspective, focusing on voice leading and longer-
range connections.
280 Analysis

A
phrase 1 (motive x) 1–9 d: i
phrase 2 9–17 i → VII (C)
phrase 3
initial statement 17–21 → v (a)
varied repetition 21–26 v
closing phrase 26–29 v

B
phrase 30–38 III (F) → V
retransition 38–43 V

A′
phrases 1 and 2 combined 44–56 i → VI (B)
phrase 3
initial statement 56–60 Ger. 6 – V – i
varied repetition 60–65

Coda
phrase (motive x) 65–76 i
closing phrase 76–87 i

FIGURE 8.1 Outline of Beethoven, Piano Sonata in D Major, op. 10, no. 3, II

The A section, mm. 1–29, is divided into three phrases and a fourth closing phrase. The
first phrase, which has a dark, foreboding, quality is closed in the tonic key of D minor.
The melodic content of the first measure is an important motive that is developed over the
course of the movement; it is bracketed and labeled x on the score. This is followed by
what comes closest to a real melody in this movement. This second phrase begins as a lyrical,
rather somber, idea that leads eventually to a perfect authentic cadence in the unexpected
key of C major (VII). As we approach the cadence, the music expands registrally and
dynamically, offering a brief glimpse of sunlight in an otherwise dark environment. This is
followed by a four-measure phrase leading to a cadence in A minor (v), internal to which
the initial gesture, the German augmented sixth resolving to the dominant, is repeated in
embellished form, followed by the dramatic gesture extending the diminished seventh chord
(forte) leading to the cadential pattern in mm. 20–21. The melodic gesture at the cadence is
a descending fifth (E5–A4), though not a complete descent by step. This four-measure idea
is then repeated in varied form with increased intensity, once again leading to a cadence on
A4.2 The closing phrase supplies the fully stepwise descent of a fifth missing in the previous
two cadences. The notes of the descent are circled on the score (Example 8.1).
281 Ternary Form

EXAMPLE 8.1 Beethoven, Piano Sonata in D Major, op. 10, no. 3, II

282 Analysis

EXAMPLE 8.1 continued

283 Ternary Form

EXAMPLE 8.1 continued

284 Analysis

EXAMPLE 8.1 continued

285 Ternary Form

EXAMPLE 8.1 continued

286 Analysis

The B section initially offers a strong contrast to the preceding material. The opening
measures, in F major (III), present a calm exterior, but this relative tranquility is short-lived,
giving way to the intensity of the repeated gesture of m. 35, which takes us to the G minor
harmony of m. 36, the pivot in the modulation back to the tonic. Here the intensity of m.
35, which is repeated a step higher in m. 37, is interrupted by the fleeting repeated three-
note gestures in the right-hand part over the G minor harmony. This pattern is then repeated
and extended for six measures over the dominant harmony beginning in m. 38. This leads
us back to the varied repetition of A.
The A′ section begins as expected, but instead of a complete statement of the opening
phrase, Beethoven presents us with a thirteen-measure phrase that combines features of the
first two phrases. He begins with the opening five measures of the first phrase (the passage
leading to the subdominant), though with more activity in the accompaniment than before,
but this then leads to a new connective passage taking us to a higher octave and the latter
part of the second phrase leading to a cadence in B major (VI). This connective passage—
featuring three successive statements of motive x—proceeds from the subdominant harmony
in m. 48. As shown on the score, the following 5–6 motion creates a II6 chord, which is
subsequently heard as IV6 in the key of B (VI). This E harmony is then altered to its minor
form before leading to the dominant seventh chord of the new key in m. 51. This last chord
is loaded with potential meaning. Enharmonically it is the German augmented sixth chord
that originally led us to A minor, but here it will eventually lead us to an arrival on B. But
Beethoven isn’t quite ready to get us there yet, and instead he pulls back once again to the
G minor harmony via a chromatic change in the bass that emphasizes this deceptive resolution
of the F7 chord. The reason for Beethoven’s choice of B as the eventual goal of this extended
phrase becomes immediately clear in the following phrase. B becomes the bass note of the
following German augmented sixth chord leading us to the dominant and eventually to the
tonic in the original key of D minor. As before, this four-measure idea is repeated with
greater intensity, but here leading to a coda rather than to a brief closing phrase.
Though the coda opens softly with statements of the x motive in the bass on i, then
VI, it is clear from the very beginning, especially with the repeated figure in the right-hand
6–7
part, that this is going to build in intensity. And, indeed, it does build to the V4–3 in mm.
72–73, repeated an octave lower in the next two measures. This leads to a closing phrase
beginning in m. 76, the first part of which is based on statements of the second half of motive
x. This is followed by the gesture G5–A5 stated three times like the tolling of a bell as the
melodic line descends the final third to closure. The music settles ominously in the dark
lower register before two fleeting statements of C–D and the final two beats of the bass
drum. A most effective ending to this dark, but beautifully expressive, work!
We now return to the beginning of this movement to examine certain passages in greater
detail. An interpretation of the initial phrase is provided in Example 8.2. The metric
organization of this nine-measure phrase is interesting. Something is unusual, at least for an
opening phrase, but just how it is unusual may not be immediately apparent. The hypermeter
is clearly duple (or quadruple) with m. 5 falling on a hypermetric downbeat, and if
we continue in this pattern m. 9 does as well. Measure 10, the start of the second phrase,
is also a hypermetric downbeat. Successive downbeat measures are not uncommon, but there
is one problem with this interpretation, namely that m. 6, where we hear the repetition
of the opening motive a step higher, is also heard as a downbeat, that is, as a beginning.
287 Ternary Form

EXAMPLE 8.2 Voice-leading representation of mm. 1–9

In fact, this motivic association is crucial to our interpretation of long-range pitch connections.
So a better interpretation of the hypermetric organization is that the successive downbeat
measures occur within the phrase, in mm. 5 and 6.
The x motive of m. 1 articulates the third D4–F4, and in the next measure F4 moves
temporarily to its upper neighbor, G4. This is followed by a diminished seventh chord
prolonged by a double voice exchange between the outer parts, after which [V42] iv leads
the top line back to the upper neighbor G4 via its leading tone. However, this time the
motion continues beyond the G4 to B4 as part of another voice exchange. This concludes
the first half of the phrase. The second part of the phrase opens with a statement of motive
x a step higher than in m. 1, resulting in the transfer of the upper neighbor, here the upper
third of the motive, to G5. This note moves down by step to F5, first harmonized by [o7]
V and then as part of the cadential 64. The continuation does not occur in this register, but
rather an octave lower, where E4, the resolution of F5/F4, is implied by the context (over
the sounding leading tone). This brief excursion into the upper octave will be immediately
exploited by Beethoven, as will the distribution of notes of a descending line in different
registers.
If we direct our attention back to the score, we see that the next phrase begins in the
upper octave with a decorated descending third leading to F5 supported by tonic harmony
in m. 10. A stem has been added to this note in Example 8.1 to show its importance. A
graphic representation of the material immediately following would show this F5 progressing
down to D5 through the passing E5, which is decorated by its own third. The D5 then
moves down to C5 over dominant harmony, and this is followed by D5–F5 supported by
a voice exchange with the bass. This is followed by a repetition of the upbeat figure that
leads once again to F5 and tonic harmony in m. 12, then F5 leading to the upper neighbor
G5, recalling the equivalent motion in the first phrase. The gesture here leading to m. 13
is a replica of the opening of this second phrase, but a step lower; it leads to E5 supported
by VII (C major). In the continuation of the phrase we expect the line to continue its descent
to C5, but Beethoven retains E5 on the downbeat of m. 15. In the repetition of the cadential
pattern that follows we do get the expected descent to local closure, but only after transfer
288 Analysis

to yet another octave higher and resolution two octaves below that. The notes of the descent
(E5–D4–C4) are indicated in Example 8.1 by added stems.
As it turns out, the descent to local closure in C major prolongs the tone that initiated
the descent, E5, which is reinstated on the downbeat of m. 20 via its upper neighbor.
We are now in the key of the dominant, and the following cadential pattern outlines a
descending fifth (E5–A4), though not a fully stepwise one. The expanded repetition of
this phrase also articulates this fifth similarly, and it is left to the closing phrase to complete
the descent by step to local closure. Though the preceding phrases are articulated from
one another by cadences, they are united by a single long-range pitch connection, the
descending fifth.
A representation of the voice leading of the B section is provided in Example 8.3. The
purpose of this example is to show the connections to the opening material, despite the very
different character of the music. Here we are once again in the lower octave with F4 as the
main melodic note, which is subsequently prolonged by its upper neighbor G4, introduced
here by the unharmonized chromatic passing tone F4. This is followed by a descent through
E4 to D4, first supported by the secondary dominant leading to the G minor chord in m.
36, the pivot in the modulation back to D minor. Not shown in this reduction are the
three-note figures that elaborate this harmony, which on the one hand interrupt the flow
of the music from m. 35 (ff) to m. 37 (ff), but on the other hand prepare the figuration
that extends the dominant in mm. 38–43. The longer-range connection across these
measures leading to the A′ section is F4/III – E4/V – D4/i.
The final portion of this movement to be examined is the coda leading to the final
phrase; a representation of the voice leading of this passage is provided in Example 8.4. The
first six measures extend the tonic, the top voice traversing the entire octave from D4 to
D5 chromatically with the harmony at the terminal points being i and i6. This is followed
by the standard cadential pattern iiø65 leading to the cadential 64 supporting the continued
ascent through E5 to F5. The V6–5 4–3 chords are elaborated by the three-note embellishing
pattern introduced in the B section, functioning primarily here, as there, to extend the
dominant. The voice leading shown in Example 8.4 represents the norm, that is, the regular

EXAMPLE 8.3 Voice-leading representation of mm. 30–44

289 Ternary Form

EXAMPLE 8.4 Voice-leading representation of mm. 65–76

resolution of 64 to 35 in two octaves, without trying to represent all aspects of the elaboration.
Again we find a descending third distributed over several octaves. We must wait to hear
this descending third in the same octave in mm. 82–84 as the bell tolls, and even here the
final note of resolution, D4, is delayed.

Schubert, Impromptu in G Major, op. 90, no. 3

One could hardly find a greater contrast to the Beethoven movement than this
impromptu by Schubert. Both are highly expressive works, but, whereas the Beethoven is
a somber, intense work, the Schubert is mostly tranquil, almost like a lullaby. The annotated
score is provided in Example 8.5, and an outline of formal-tonal organization is provided
in Figure 8.2.
The A section consists of an a b b′ design, each eight measures in length. The first eight
measures are further divided into an antecedent leading to the dominant and a consequent
ending with a perfect authentic cadence. The b phrase, which emphasizes ii, then IV, also
concludes with a perfect authentic cadence, as does the following varied repetition of the
phrase. This frequent coming to closure contributes to the music’s sense of peace and well-
being. The B section provides contrast. Its controlling mode is minor, E minor (vi), with
an internal excursion to C major (VI in E minor). More important, perhaps, is that the
phrase lengths are no longer predictable. The B section consists of two phrases, both
expanded. Though the phrases are not parallel, that is, they do not consist of the same
material, they are related by the characteristic rhythm of the left-hand part in their opening
measures, which helps in identifying where phrases begin and end. The basis for the phrase
expansion is indicated on the score. The first phrase is expanded internally by a slightly varied
repetition of the first three measures, which gives the initial impression that the hypermeter
has now shifted from quadruple to triple, and externally by extension of the goal harmony,
C, by an additional four measures (mm. 36–39). The second phrase is also expanded
internally, in this case by varied repetition of its first four measures. A particularly beautiful
moment occurs in m. 49, where Schubert leaps up to G5, momentarily breaking away from
the restricted range of the melody. This phrase is followed by a brief retransition leading
back to the opening material.
290 Analysis

A
a antecedent 1–4 G: I – V
consequent 5–8 I–V–I
b 9–16 [V] ii – [V] IV – V – I
b′ 17–24

B
phrase 1 25–39 e (vi): i → VI
o
phrase 2 40–51 7 – i – ii6 – V – I
retransition 51–54 I – iv/G: ii – V65

A′
a antecedent 55–58 G: I – V
consequent 59–62 I–V–I
b' expanded 63–74 [V] ii – [V] IV – V – I

Coda 74–86 I

FIGURE 8.2 Outline of Schubert, Impromptu, op. 90, no. 3

The A′ section consists of an exact repetition of the a period and an expanded version
of the b′ phrase. This four-measure internal expansion (mm. 70–73) begins with the
substitution of IV6 for the expected tonic (the expected fourth hyperbeat), which is followed
by a varied repetition of mm. 67–69. The final measure of this phrase, m. 74, overlaps with
the harmonically interesting coda. It will be discussed further below, but here we note another
one of the special expressive moments in this piece. In the coda the repeated underlying
melodic progression is G4–F4–G4. Three times Schubert has approached the leading tone
by the descending arpeggiation C5–A4–F4 (mm. 77, 81, and 82), but the fourth time the
music leaps up to E5, the ninth of the dominant, an extremely effective moment in this
very beautiful work.
A representation of the voice leading of the a and b phrases of the A section is provided
in Example 8.6. The antecedent of the a period proceeds to what Schenkerian analysts refer
to as an interruption. In an interruption, the melodic line descends as far as scale degree 2
supported by V, but instead of completing the descent to closure, this motion is interrupted
and returned to its point of departure, in this case scale degree 3. The consequent, then,
completes the motion. This is a very common paradigm in tonal music. The b phrase counters
this strong pull toward closure with an ascending fifth A4–E5 divided into two overlapping
thirds, the first progressing to ii and the second to IV at the climax of the phrase (m. 12).
This ascending fifth is subsequently answered by a descending fifth a step lower (D5–G4),
291 Ternary Form

EXAMPLE 8.5 Schubert, Impromptu in G Major, op. 90, no. 3

292 Analysis

EXAMPLE 8.5 continued

293 Ternary Form

EXAMPLE 8.5 continued

294 Analysis

EXAMPLE 8.5 continued

295 Ternary Form

EXAMPLE 8.5 continued

296 Analysis

EXAMPLE 8.5 continued

297 Ternary Form

EXAMPLE 8.5 continued

298 Analysis

EXAMPLE 8.6 Voice-leading representation of mm. 1–16

concluding with a perfect authentic cadence. As shown in Example 8.6, the descent through
B4 and A4 to G4 is delayed until the final moment.
A representation of the voice leading of both phrases of the B section is provided in
Example 8.7. The initial part of the first phrase, including the varied repetition of its first
three measures, prolongs the local tonic, E minor. As shown in Example 8.7, the melodic
motion in these measures occurs at two levels, a descending fifth from B4 to E4, a motion
to an inner voice represented by downward-directed stems, and the slower-moving upper
line that descends a third from B4 to G4. This takes us to m. 31, which becomes the point
of departure for the modulation to C major (VI). There is an immediate descent of a fifth
to the downbeat of m. 32, above which G4 is retained, and from that point there is a clear
descent of a fifth to the perfect authentic cadence on C in m. 35. Overall, then, there is a
descent of a third B4 to G4 followed by a descent of a fifth from G4 to C4. Together
these form a descending seventh, replacing an ascending step from B4 to its upper neighbor,
C5. Looking ahead, we see that the music indeed does arpeggiate up to this C5 in m. 39.
So, if we were to represent only the deepest level of connection across this span of fifteen
measures (mm. 25–39), it would show B4, supported by an E minor chord (i) progressing
to its upper neighbor C5, supported by a C major chord (VI).
Measures 35–39, the extension of the goal harmony, is worthy of some scrutiny. The
local tonic is initially prolonged by a neighboring motion involving the minor subdominant
chord, followed by a progression involving the German augmented sixth chord. When
dealing with so many flats, it takes a moment to determine the notes of this chord. Begin
with the lowered sixth degree of the scale, which is A, then above it add major third (C),
perfect fifth (E) and augmented sixth (F): A–C–E–F. The repetition of these two
measures of extension involves two subtle changes. First, the neighboring motion involves
the regular subdominant (IV6) and the German sixth chord is replaced by o65 of V, A having
replaced the earlier A. These very subtle changes in color are typical of Schubert’s
repetitions.
The first four measures of the second phrase, which are repeated, return us to E minor.
C 5, the upper neighbor, is now harmonized as seventh of o7 leading to B4, harmonized

299 Ternary Form

EXAMPLE 8.7 Voice-leading representation of mm. 25–54

by the dominant (end of m. 43). The varied repetition ends with V42, which leads now to
G in the bass (I6) passing on down to E. We have already noted the melodic leap up to
G5 at this point, a simple motion that seems to open up an entirely new sound world. This
motion covers a descent from B4 to E4 at the cadence in m. 51. This is followed by a brief
connection back to the opening material. Here B4 is reintroduced by its upper neighbor,
C5, this time harmonized as seventh of the dominant.
The final section we will examine in some detail is the coda, which contains a wonderful
example of enharmonic use of the augmented sixth chord. A representation of mm. 74–82
is provided in Example 8.8. In the first four measures the harmony progresses to the minor
subdominant, which prepares for its chromatic alteration as the German augmented sixth
chord. The second time Schubert respells the augmented sixth chord enharmonically as a
dominant seventh chord (D–F–A–C as opposed to E–G–B –C), which leads to the minor
Neapolitan (written as G–B–D rather than A–C–E). But he then pulls us back once
again to G major by re-employing this dual-function chord as an augmented sixth chord
6–7
resolving to V4–3 – I.
Though this work seems to float by almost effortlessly, it is actually very complex. As
always with Schubert, there are some interesting harmonic twists, and, as the work progresses,
we encounter some sophisticated manipulation of the lengths of musical phrases.
300 Analysis

EXAMPLE 8.8 Voice-leading representation of mm. 74–82

Brahms, Intermezzo in B Minor, op. 117, no. 2

Brahms indicates to the performer to play this intermezzo very expressively (con molto
espressione). It is, in fact, a deeply expressive work, much of it gentle and flowing, yet at
other points somewhat agitated. And though the dynamic level remains predominantly soft,
there are places where the music builds to forte, most notably at the climax in mm. 67ff.
Technically, this is an extremely sophisticated work, in which phrases are often expanded
and overlapped, the end of one almost indistinguishable from the beginning of the next.
The harmony is at times crystal clear, at other times momentarily obscured by surface
embellishment. A copy of the score is provided in Example 8.9, and an outline of its formal-
tonal organization is provided in Figure 8.3.
This intermezzo is in a ternary form that is expanded internally by a retransition leading
back to A′ and externally by a coda. Both the A and B sections consist of parallel phrases
in antecedent-consequent relationship, and, as we shall see, the two ideas are motivically
related. The goal harmony of the intermezzo’s first phrase, the dominant, is extended, melting
seamlessly into the beginning of the second phrase, which is expanded internally by a
digression to the key of the dominant. But the real goal is the relative major, D, the key
of the B section. The antecedent phrase of the B section ends on an F major chord (III), a
fleeting reference to the dominant of the main key, and, for the first time in this piece, the
consequent phrase ends with a strong cadence, an imperfect authentic cadence in the local
key. The approach to this cadence is the first time the dynamic level reaches forte. The B
section is followed by a retransition that leads us back, again almost seamlessly, to the opening
material, this time with D rather than D in the bass. Otherwise the antecedent is largely
the same as in the beginning. However, the consequent phrase is substantially rewritten,
leading to the climax of the piece, the tritone-related progression II7 – V43 harmonizing
scale degree 4, first E6, then dropping down to E4 in preparation for the coda. The coda
further prolongs the dominant until the first perfect authentic cadence of the piece and the
first tonic triad in root position three measures before the end of the intermezzo! Even here,
the arrival at scale degree 1 is delayed by a suspension.
A representation of the voice-leading structure of both phrases of the A section is
provided in Example 8.10, and an account of the chord-to-chord harmonic syntax is indicated
301 Ternary Form

A
antecedent 1–9 b: i6 – V7
modulating consequent 9–22 i6 – vi/D: iv

B
antecedent 22–30 D (III): V7 – I → III
consequent 30–38 V7 – I – IV – V – I
retransition 38–51 b: [Ger. 6 – V7] V7

A′
antecedent 51–60 I6 – V7
consequent 60–72 i6 – II7 – V7
 6 – 5
Coda 72–85 V6– –i

FIGURE 8.3 Outline of Brahms, Intermezzo in B Minor, op. 117, no. 2

on the score. The meaning of the opening progression is not immediately clear; it leads to
a B minor chord in first inversion, first with B4 as the top note, then F5. What follows is
a diatonic sequence of root-position seventh chords progressing by descending fifths from
i7 (third beat of m. 2) to iiø7 (m. 5). Melodically the top voice reaches up to A5, the seventh
of the tonic chord, and then the top voice descends in parallel tenths with the bass
(alternating with sevenths in a 10–7 pattern) until m. 5, where there is a voice exchange
with the bass prolonging ii and E5. The descending third from this E5 is then repeated in
the next measure supported by dominant harmony. At this point we expect that this E5,
now the dissonant seventh of the dominant, will resolve to D5 supported by tonic harmony,
but this expectation is never realized. Instead Brahms extends V7, melodically approaching
the seventh once again from a third above. This leads us to the expected D5 (m. 8),
harmonized not by the tonic but presented as an accented passing tone over an extended
dominant harmony. Following the cascading arpeggio down to the low E, this dominant
harmony is transformed into o43.3 Once the dissonant D resolves to C5 over E on the third
beat of m. 9 and this leads to B4 over D in the bass (i6) on the downbeat of m. 10, we
have already begun the consequent phrase. This is what was meant earlier by our comment
that the two phrases overlap almost seamlessly. The antecedent phrase has been expanded
by the extension of the dominant harmony and the composing-out of the third E5–D5–C5
over a span of four measures (mm. 6–9). This third is part of a larger motion of a descending
fifth F5–B4 crossing the phrase boundaries.4
The consequent phrase begins as before, but this time the top voice reaches up from
F5 to B5 initiating a series of secondary dominant seventh chords leading back to F5 over
D. This is not a seventh chord, but an augmented sixth chord initiating a four-measure
parenthetical excursion to F minor that is abruptly cut short by the reintroduction of this
302 Analysis

EXAMPLE 8.9 Brahms, Intermezzo in B Minor, op. 117, no. 2

303 Ternary Form

EXAMPLE 8.9 continued

304 Analysis

EXAMPLE 8.9 continued

305 Ternary Form

EXAMPLE 8.9 continued

306 Analysis

EXAMPLE 8.9 continued

307 Ternary Form

EXAMPLE 8.10 Voice-leading representation of mm. 1–23

same chord on the downbeat of m. 18, now spelled with a C rather than a B in the inner
part. These two chords are marked with asterisks in Example 8.10.5 This chord then
leads to the G minor chord in m. 20, the pivot in the modulation to D major (III): vi/iv.
This harmony is extended over three measures (mm. 20–22), further expanding the phrase.
This is a tricky passage to decipher, but in essence this G harmony is colored by a 5–6
motion (D–E), momentarily transforming the minor subdominant into a Neapolitan chord.
The voice leading once again continues across the phrase boundary, which is indicated by
the vertical dotted line in Example 8.10. The gesture initiating the new phrase is F4–E4–D4,
a motion to an inner voice. In the immediate context this F4 is a passing tone, and thus its
notation in Example 8.10 without stem. But, as we shall see, this note retains its important
function in the B section as third of the local tonic.
Before proceeding to the B section, we want to point to an important feature of the
opening phrase, since this feature is extensively exploited later in the intermezzo. This feature
is the reaching up from the F5 in m. 2 to A5 and the return by step to F5 two measures
later. Though the motion continues beyond this second F5 to E5 (m. 5), Schenker’s
unpublished analysis of this intermezzo identifies F5–A5–G5–F5 as an important motivic
idea exploited later on.6 This idea is marked on the score with a bracket and on our
representation of the voice leading of the two phrases of the B section, which is provided
as Example 8.11.
308 Analysis

EXAMPLE 8.11 Voice-leading representation of mm. 23–38

The initial melodic gesture of the first phrase of the B section, the descending third
F4–E4–D4 is answered by the ascending third D5–E5–F5, the last note covered by A5.
So the F5–A5 of m. 2 is now stated almost simultaneously and then prolonged by its upper
neighbors before resolving to G4 over E4 supported by supertonic harmony in m. 26,
displaced momentarily by their upper appoggiaturas. The following three measures, marked
legato espress. e sostenuto, are, at first glance, a bit difficult to sort out due to the notes of
melodic embellishment, but once we sort this all out, the harmony of m. 27 is ii5–6 initially
decorated by its secondary dominant, which passes through a tonic chord in first inversion
to ii65 in m. 29. The relationship across mm. 27–29 is a voice exchange between the outer
parts prolonging ii and G4. The chromatic inflection of this supertonic harmony deflects
away from the expected dominant at the end of the phrase to III. The result of this surprise
309 Ternary Form

is twofold. It not only creates a harmonic reference to B minor, but it also articulates
F4 as the melodic goal of the phrase, the completion of the motive F–A–G–F, while
preparing the repetition of the phrase’s opening gesture at the outset of the consequent.
Once again the answering third D5–E5–F5 is immediately covered by A5 in the
following measure (m. 32). And once again the third F5–A5 is decorated by its upper
neighbors, first in this octave, then an octave lower in m. 34, where it is supported by IV.
The following three measures, which prolong IV, are analogous to mm. 27–29. They are
marked espress. e sostenuto, but here Brahms allows the dynamic level to rise to forte. As before
the underlying harmony is temporarily obscured by the numerous notes of melodic
embellishment, but once this is sorted out we see (and hear!) the subdominant harmony
prolonged by its neighboring V65 followed by the chromatic voice exchange between the
6–7
outer parts, as shown in Example 8.11. This leads to the V4–3 – I9–8, the first authentic
cadence of the piece. The top voice leads through A4 to G4 (8–7 of the dominant) to
F4/I, completing another clear statement of our identified motive.
In some sources the form of this intermezzo has been incorrectly identified as sonata
form. The idea that this intermezzo is written in sonata form may derive in part from the
fact that the music to this point consists of two ideas that are linked to form a larger unit
that comes to a cadence, a temporary resting point, only in m. 38. Furthermore the following
retransition has the character of a development. It opens with a repetition of the initial
descending third in octaves: D–C–B. As shown in Example 8.12, a representation of the
voice leading of the retransition, this gesture, beginning from the D5, generates an ascending
line leading to G5, supported by the C7 chord, V7 of V, in m. 45. Schenkerian analysts
refer to the process by which this ascending line is created as “reaching over,” a term which
here describes the reaching up to a third above each step in the ascending line to approach
the next one from above, similar to the way F5 reaches up to A5 in m. 2 to lead to G5
on the next downbeat. The last step in this process involves just these pitches, except this
time the goal is G5, not G5. Note how the third above F5, that is, the A5, is extended
for three measures (mm. 42–44), while the supporting harmony is transformed from a B7
into an enharmonically spelled augmented sixth chord by means of a chromatic voice
exchange. This is the same chord that was given double meaning in the consequent phrase
of the A section (highlighted by the asterisks in Example 8.10).

EXAMPLE 8.12 Voice-leading representation of mm. 38–52

310 Analysis

Arrival at the C7 chord in m. 45 is followed immediately by the statement of the two-

note figure D–C three times in three successively lower octaves, the last harmonized by
V7 in m. 48. These three statements stand out not only because of the change in the mode
of presentation (as chords rather than arpeggios) but also because of their duple rhythmic
articulation, which creates a hemiola against the triple meter. It is just at this point that the
metric organization of this passage becomes less clear. If we look at the score, we see that
the changes of surface design suggest a quadruple hypermeter through m. 46, as indicated
between the staves. However, the continuation is not clear, which might very well have
been Brahms’s intention, since he takes great effort to disguise the return of the opening
material. We will return to this thought in a moment.
The melodic gesture that generates the retransition is the descending third D5–C5–B4
in mm. 38–39, and this third then becomes the overall gesture leading from m. 38 to the
B4 on the downbeat of m. 52, after the beginning of the A′ section. That is, the voice
leading in both melody and bass cross the formal boundary between the retransition and the
A′ section, which is indicated in Example 8.12 by the dotted vertical line. The long-range
third spanning these fifteen measures is shown by the large slur and extended stems in our
example. Brahms takes great effort to help us hear this connection. Though the overall
melodic gesture from the pick-up to m. 39 through m. 45 is D5 to G5, the connection of
D to C is established by the insistent repetition of that figure, and the following connection
of C to B is then filled in by this third at two levels, as shown by the notation in Example
8.12. Most fascinating is the hesitation after the first two notes of this third in mm. 51–52.
At this last moment Brahms rejects D–C–B for D–C–B as the bass leads through E to
D, not D. This creates a dominant ninth chord (B–D–F–A–C), which directs the
harmony beyond the beginning of the phrase to the subdominant in m. 54, further blurring
the formal boundary between retransition and return.
A representation of the voice leading of the consequent phrase of A′ and the coda, which
are linked to form a continuous unit, is provided in Example 8.13. Before discussing this
final portion of the intermezzo, we should review the structure of the initial phrase (Example
8.10), since this final section can be understood as an enlargement of that structure. After
F5 is established in the top part (m. 2), this voice reaches up to A5, then descends by step
to E5, first harmonized by the supertonic, then the dominant. This is followed by an
extension of the dominant, above which E5 progresses through the passing tone D5 to C5.
In other words, D5 is never given its own harmonic support in the descent to B4. The
final portion of the intermezzo, beginning in m. 61, is an expanded version of this opening
phrase. The initial four measures involve a motion to A5 progressing in parallel tenths with
the bass, from which point there is a chromatic descent to E5, spelled as D5, the third of
a B major-minor seventh chord, which enharmonically is II7, the Neapolitan seventh chord.
This prepares the reharmonization of scale degree 4, now transferred to E6 in m. 69 as
seventh of the dominant. This is the climax of the piece from which point E is transferred
down two octaves and the dynamic level gradually decreases in preparation for the coda
(Più Adagio). This final portion of the phrase extends the dominant for another eleven
measures (mm. 72–82), above which scale degree 3, first D, then D is extended until finally
resolving to C4 and then B3, supported for the first time by tonic harmony in root position.7
As in the opening phrase, scale degree 4 (E) is emphasized and extended, and then scale
311 Ternary Form

EXAMPLE 8.13 Voice-leading representation of mm. 61–85

degrees 4–3–2 in the descent to closure all occur over a prolonged dominant. The difference,
of course, is one of scale.
The coda is an integral part of the tonal plan, not just a section added at the end.
Nevertheless, it does provide in many ways a kind of summary in that it makes clear reference
to material from previous sections. As shown in Example 8.13, the 64 with D is prolonged
by a long-range voice exchange between the inner parts from mm. 73 to 80, after which
D4 is introduced and resolved immediately to C4. The final approach to scale degree 1
involves a diminution of the long-range melodic line, the descending fifth. Internal to the
voice exchange, there is first a reference to the key of the B section, D major (III), then
within the context of B minor a passing reference to F minor, the key suggested but never
confirmed in the parenthetical digression in the second phrase of the piece. This occurs in
mm. 77–80 within the context of the descending third D5–C5–B5 indicated by the beam
in Example 8.13.
This intermezzo features many of the most characteristic elements of Brahms’s style:
elaborate motivic development; complex use of embellishing tones; blurring of boundaries
between phrases and larger sections; de-emphasis of root-position tonic harmony and delay
of closure; and intricate textures, both contrapuntally and rhythmically. As such, his music
presents special challenges to performers, analysts, and listeners.
312 Analysis

SUGGESTED ASSIGNMENTS

Beethoven, Piano Sonata in E Major, op. 7, II

Though portions of this movement have been examined earlier in this book, it is worth
studying this movement as a whole. The following are some areas to focus on:

1. The form of the A section is a b a′. What are the tonal connections between the a and
b sections? How is the a′ section expanded?
2. Analyze the form and harmonic progression of the B section. How is it related to A?
How does Beethoven get back to the dominant (m. 37)? You can check your
interpretation against the one provided in Example 4.16.
3. What are the motivic and rhythmic bases for the retransition?
4. Analyze carefully the final section (coda) beginning in m. 74.

Schubert, Moment musical, op. 94, no. 2

Portions of this work were discussed in Chapter 4 (see Examples 4.4 and 4.5) in the context
of modal mixture. Still there is much to be learned from this relatively short piece. It is
written in extended ternary form, by which is meant that it is extended to a five-part form
by varied repetition of the A and B sections: A B A′ B′ A′′ (coda). Consider the following:

1. Harmony
2. Phrase rhythm, including phrase expansion
3. Schubert’s development of the opening neighbor-note motive

Schubert, Piano Sonata in B Major, D. 960, II

1. The A section consists of two phrases, mm. 1–17 and 18–42, where the second is a
reharmonization of the melody of the first. Both are expanded. How?
2. The A′ section involves some interesting harmonic digressions. How do you explain
the brief passage in C major, mm. 113ff.? The final phrase in C major can be explained
as an expansion of eight measures. How so?
3. The B section is more complex. What is its internal form? Can you explain the harmonic
progression of the phrases beginning in m. 59 and later in m. 76? These two phrases
involve an odd number of measures. Can these be explained as resulting from some
kind of phrase expansion? What is the overall harmonic logic of this section? How is it
related to the outer sections? (There are probably several correct answers to that
question!)
313 Ternary Form

Chopin, Etude in E major, op. 10, no. 3

1. The A section consists of an antecedent phrase and an expanded consequent. How is
the antecedent divided? How is the consequent expanded? What is unusual about the
resolution of the cadential 64 at the end of the second phrase?
2. The B section is difficult. First examine the phrases, noting repetitions, and indicate
harmonic goals. Do not try to analyze harmonically the extended chromatic passages,
mm. 38–41 and 46–53.
3. Examine carefully the retransition. How does Chopin lead back to the beginning?

Brahms, Intermezzo, op. 117, no. 1

1. The A sections of this work are relatively simple and repetitive, appropriate for a setting
of a simple Scottish lullaby. What rhythmic effect occurs in mm. 13ff.? How do mm.
17–20 relate to mm. 1–4, and how do they prepare the B section?
2. The B section is contrasting in several respects. How so, and how also is it related to
the outer sections? Here you might prepare a simplification of the voice leading and
then identify the important harmonies. What is the harmonic progression in mm. 25–26,
which is repeated in varied form in mm. 27–28?
9 Rondo Form

In our final chapter, we turn to the rondo, which is based on alternations between a recurring
theme, known as the rondo theme or refrain, and contrasting sections referred to as
episodes. Occasionally, composers wrote rondos as independent pieces, generally for solo
piano or for solo instrument with piano or orchestral accompaniment. More commonly,
we encounter the rondo as an alternative to sonata form for the last movement of multi-
movement instrumental works such as sonatas, string quartets, and symphonies. In concertos,
the final movement is much more often a rondo than a sonata form. Rondo forms
occasionally occur as slow movements, though less frequently since other formal types, such
as sonata form (sometimes without a development section), ternary form, and theme-and-
variations form, are found in slow movements in the Classical tradition. After a general
discussion of rondo form, this chapter will examine two rondos that serve as finales. The
first comes from a piano sonata by Mozart composed in the late 1770s, and the second is
from a violin sonata by Brahms written in the mid 1880s. Compared to other rondos by
each composer, both movements are expansive and formally complex; in relation to one
another, the Brahms movement aptly demonstrates the increasingly elaborate handling of
tonal and formal conventions in the late nineteenth century.

Rondo Form
In the previous chapter we studied ternary form, which arises from the restatement of an
entire section (A) after a contrasting section (B). Rondo form can be viewed as an extension
of ternary form. In a rondo, the initial section (A or refrain) is restated at least two additional
times. Its first appearance establishes the tonic key and almost always concludes with a perfect
authentic cadence in that key. Nearly all refrains have a length between sixteen and thirty-
two measures, and they generally feature symmetrical phrases with clear cadences. Sometimes,
a rondo refrain contains repeated sections, creating a complete binary or rounded binary
form.
The subsequent statements of the refrain can be unaltered, or they can have surface
embellishment or be shortened somewhat; the final refrain is often lengthened, especially
when not followed by a separate coda. Usually, all statements of the refrain occur in the
tonic key, but it is possible for a later refrain to start in the “wrong” key. On rare occasions
a refrain statement in the middle of the form will occur in a non-tonic key, or it will occur
in the tonic key but will proceed into the subsequent section without the expected perfect
authentic cadence.
315 Rondo Form

5-part rondo
A B A C* A (Coda)
major: I V, IV, or vi I V, IV, vi, or i I I

minor: i III, v, or VI i III, v, or VI i i (or I)

* The C section is often unstable tonally and features more than one key.

7-part rondo
A B A C* A B A (Coda)
major: I V I I, IV, vi, or i I I I I

minor: i III or v i i, iv, or VI i i (or I) i i (or I)

* The C section is often unstable tonally and features more than one key.

FIGURE 9.1 5-part and 7-part Rondo Forms

In between the statements of the refrain there are contrasting sections which we will
refer to as episodes and label with the letters B and C. These sections contrast the refrain
both in thematic content but also in key. As with refrains, episodes often have considerable
internal repetition.
The most common types of rondo are 5-part and 7-part rondos with the layout of
thematic materials shown in Figure 9.1. Except in the case of B episodes in 7-part rondos,
there is considerable variety in the key areas found in episodes; Figure 9.1 lists only the most
frequent choices. Often the C section will feature more than one key area, although when
this occurs it is unlikely that each key would be confirmed with a perfect authentic cadence.
Most rondos have connective material between some of the thematic sections. Nearly
all rondos have passages that prepare returns of the refrain. These passages typically sit on
the home dominant (usually V7) and are referred to as retransitions. Some rondos also
have passages after refrain statements that prepare the key of the following episode; we refer
to these as transitions. When a rondo contains multiple retransitions, these passages often
share similar material; the same is true of the transitions in a rondo with multiple transitions.
Since we have chosen to focus on two complex rondos in this chapter, we offer now
brief descriptions of two 5-part rondos by Haydn and Beethoven. The third movement of
Haydn’s Piano Sonata in D Major, Hob. XVI/37 provides a good illustration of a highly
sectionalized rondo form. Its refrain (mm. 1–20) consists of a complete rounded binary form
with both halves repeated. The B episode (mm. 21–40), in the parallel minor, is also in
rounded binary form with exact repetitions. After an unaltered refrain restatement, the C
episode (mm. 61–80) provides yet another rounded binary form, this time in the subdominant
316 Analysis

(G major). After the C section, a retransition (mm. 81–93) leads back to the home dominant
and playfully sits there for several measures. This is the only connective passage between
this rondo’s principal sections. The final refrain has its repetitions written out to allow for
a livelier sixteenth-note texture in the left hand. The movement ends without a coda; two
tonic chords are added at the conclusion of the final refrain.
More elaborate is the 5-part rondo found as the slow movement of Beethoven’s Piano
Sonata in C Minor, op. 13 (“Pathétique”). The refrain establishes the tonic key of A major
through an eight-measure phrase that is immediately repeated an octave higher. The B
episode spans a mere seven measures (mm. 17–23) but deceives the listener by tonicizing F
minor (vi) before achieving a perfect authentic cadence in E major (V). After a brief
retransition (mm. 24–28), the refrain returns (mm. 29–36) but is stated only once. Like the
B episode, the C episode (mm. 37–44) touches on two keys; following a tonicization of A
minor (i), the episode culminates with a perfect authentic cadence in E major (a convenient
spelling of VI). After a short retransition (mm. 45–50), the refrain returns in its original
sixteen-measure length (mm. 51–66) but with a faster inner-voice accompaniment (triplet
sixteenths rather than sixteenth notes). Unlike the Haydn, this rondo concludes with a coda
(mm. 67–73).

Mozart, Piano Sonata in B Major, K. 333, III

Mozart’s Piano Sonata in B Major, K. 333, is one of his most popular piano works. In large
part, this is due to its masterful coordination of Mozart’s characteristically tuneful and
charming melodic writing with a number of surprises. In the finale, these surprises include
clever manipulations of thematic materials, phrase expansions, and virtuosic passages
reminiscent of concerto movements. Our commentary will highlight these important
features. Figure 9.2 provides a summary of formal-tonal organization, and Example 9.1 gives
an annotated score of the complete movement.
Central to this movement’s effectiveness is its innocent beginning. The refrain consists
of a single parallel period (mm. 1–8), which is then repeated with louder dynamics and faster
rhythms in the left hand (mm. 9–16).1 As indicated on the score by circles, the melody in
mm. 1–8 outlines two descents from F5, the first ending on C5 at the end of the antecedent
and the second ending on B4 at the close of the consequent. When the antecedent is
repeated, its melody remains the same, but when the consequent is repeated its melody is
altered and achieves a more emphatic conclusion. This emphasis arises through the higher
register and the faster rhythms (compare m. 15 to m. 7). Despite the displacement of
embellishing pitches an octave higher in m. 15, the underlying melodic content remains the
same. The brief dabbling with this higher register suggests that the music will soon explore
it more fully.
The transition is an eight-measure sentence that achieves the most usual tonal goal, V
of V (m. 24). Again, structurally important melodic pitches have been circled on the score.
The transition’s melody comes to a clear termination on G4 on the second quarter note of
m. 24, although the left-hand accompaniment already begins the texture that supports the
initial measures of the B episode. We will return to the significance of this formal boundary
later. A particularly beautiful aspect of the transition is the prominence of the chromatic
pitch that signals the modulation, namely E (m. 20). The anacrustic motive x (bracketed
317 Rondo Form

REFRAIN (A)
antecedent 1–4 I–V
consequent 5–8 I
antecedent repeated 9–12 I–V
consequent repeated 13–16 I

Transition 17–24 leads to V of V

EPISODE 1 (B) 24–36 F(V): V–I

Retransition 36–40 leads to B(I): V7

REFRAIN (A) 41–56 I

Transition 57–64 leads to g(vi): V

EPISODE 2 (C → Development)

sentence 65–72 g(vi): i–V

link 73–76 leads to E(IV): I
development of transition 76–90 E(IV): I–V
development of refrain 91–103 leads to B(I): V

Retransition 103–111 B(I): V – V7

REFRAIN (A) 112–127 I

Transition 128–148 I–V

EPISODE 1 transposed (B′) 148–164 V–I

Transition to cadenza 164–171 I–V

Cadenza 171–199 V

REFRAIN (A′)
antecedent 200–203 I–V
consequent expanded 204–214 I

CODA 215–225 I

FIGURE 9.2 Outline of Mozart, Piano Sonata, K. 333, III

318 Analysis

EXAMPLE 9.1 Mozart, Piano Sonata in B Major, K. 333, III

319 Rondo Form

EXAMPLE 9.1 continued

320 Analysis

EXAMPLE 9.1 continued

321 Rondo Form

EXAMPLE 9.1 continued

322 Analysis

EXAMPLE 9.1 continued

323 Rondo Form

EXAMPLE 9.1 continued

324 Analysis

EXAMPLE 9.1 continued

325 Rondo Form

EXAMPLE 9.1 continued

326 Analysis

EXAMPLE 9.1 continued

327 Rondo Form

on the score) begins both statements of the sentence’s basic idea as well as the continuation,
but transposed a step higher each time. Due to this process of transposition, we expect E
on the last eighth of m. 20 and our attention is grabbed when we hear E instead. Similar
prominent handling of a pivotal chromatic pitch will occur in the subsequent transitions.
Like the transition, the B episode begins as an eight-measure sentence, but here the
continuation is expanded. Instead of the expected perfect authentic cadence at m. 32, an
imperfect authentic cadence occurs, deferring local closure in F major until m. 36. This
expanded continuation provides the most exciting music thus far: in addition to its nearly
continuous sixteenth-note motion, it touches on a new registral highpoint, D6 (m. 30).
Note also that the repeated continuation provides one measure more than necessary.
Whereas V6–5 4–3 span a single measure in that initial statement (m. 31), on the second cadential
approach they are stretched across two measures (mm. 34–35). Not only does this extra
measure increase the anticipation for cadential closure but it also permits the episode to
conclude on a hyperdownbeat rather than a weaker fourth hyperbeat. The end of B elides
with a short retransition that transforms the local tonic triad back into V7 of the home key.
Although mm. 36–40 have a connective formal function, they do introduce yet another
significant melodic-rhythmic motive, the repeated eighth notes in the right hand.
After an exact restatement of the entire refrain (mm. 41–56), the same transition material
returns but leads to a different tonal goal: V of G minor (vi). What pitch most clearly
distinguishes B major from G minor? These keys share the same key signature, but in
G minor the expression of dominant harmony requires F. Notice in m. 62 how prominently
the melody moves from F5 to F5. This is analogous to the highlighting of E mentioned
above in the first transition’s modulation to F major. The modulation to G minor is clarified
by the augmented sixth chord at the end of m. 63. The choice of vi for the second episode
is fairly common in major-mode rondos; almost always, the second episode of a major-mode
rondo has a considerable emphasis on the minor mode (if not vi, then usually i). Since major-
mode rondo refrains typically have a light-hearted character, the shift to minor during the
second episode, or at least a significant portion thereof, brings dramatic contrast.
The second episode starts out as a bold contrast. Besides the shift to a minor key, the
large ascending melodic leaps provide a surge in intensity. Despite the rhythmic similarity
between the melody in m. 65 and the one in m. 5, the overall effect is contrast, and it seems
best to label this new section with the letter C. Mozart, however, gradually returns to thematic
materials heard earlier. With the modulation to E major at m. 76, a humorously disguised
version of the transition occurs. Although the dotted rhythms are new, the pitch content
of the melody in mm. 76–79 is a transposed version of mm. 16–19 (and mm. 56–59). In
m. 76, the onset of the harmonic arrival on E major and the long melodic pitch suggest
the presence of hypermetric reinterpretation (4 = 1). As the phrase unfolds, it gradually
becomes apparent that the anacrustic beginning characteristic of the transition has been
preserved. This explains the dual hypermetric reading of mm. 77–81 shown in the score
annotations. The music does not attain cadential closure in E major; instead, after a half
cadence (m. 88), which is immediately repeated (m. 90), the music passes through C minor
and B minor. In these latter key areas, the development of earlier thematic material becomes
explicit, as rhythmically unaltered versions of the refrain’s beginning occur. The modula-
tion to B minor facilitates the end of this section, since the arrival on its dominant (m. 103)
can launch a retransition to prepare the reassertion of B major at the next refrain statement
328 Analysis

(mm. 112ff.). Due to the extensive manipulation of thematic material from earlier in the
movement, this central section of the rondo comes to function more as a development than
as a contrasting C section. Since the opening measures of the central section are highly
contrasting, the outline in Figure 9.2 suggests a transformation “C→development,” but if
one were to choose a single label “development” would be more appropriate than “C.”
Whether the thematic content of the central section of a 7-part rondo is new or develops
earlier thematic material is an important distinction. When this section consists entirely or
principally of development, as in this piece, the rondo takes on an attribute of sonata
form. For this reason, a 7-part rondo with a development as its central section is often referred
to as a sonata-rondo. The sonata component is greatly enhanced if there is a transition
between the initial refrain and the B episode, as there is in Mozart’s rondo. Thus, the initial
refrain, transition, and B episode resemble a sonata exposition; the central section provides
a development; the return of the refrain after the development, the ensuing transition,
and the home-key restatement of the B episode resembles a recapitulation. Of course,
compared to sonata form there are two extra refrain statements, one between the “exposition”
and “development” and one after the “recapitulation.” In addition, the themes themselves
usually have somewhat different internal construction than one encounters in a sonata form.
As noted above, rondo refrains and B episodes generally have more symmetrical phrase
structures and a greater amount of exact repetition than in the first and second themes
of a sonata form. Nonetheless, the substitution of a development for a contrasting C episode
increases the “weight” or “depth” of a rondo. A sonata-rondo typically provides a more
substantial conclusion to a multi-movement work than does a 7-part rondo with a C section.
Returning to Mozart’s rondo, we note that after the development section the refrain
returns in its entirety and without variation (mm. 112–127). Since the upcoming B episode
will occur in the home key, the transition that begins at m. 127 must be modified, as its
two previous statements modulated to the dominant and the submediant respectively. Like
the earlier transitions, the change is signaled prominently by the introduction of a chromatic
pitch, here A (m. 132). Initially, this A is harmonized by V65 of IV, but through a chromatic
voice exchange this chord evolves into V65 of ii and sets up a brief descending fifth sequence
that leads back to the tonic harmony (m. 137). Compared to the previous transitions, this
one is longer and has a thicker, quasi-orchestral texture; as we will see, Mozart revisits these
qualities later. The transition attains its expected goal, the dominant harmony, at m. 144
and expands it until m. 148.2 Where exactly does the transition end? If one compares
m. 148 with the corresponding m. 24, it is clear that the melody of B begins in the middle
of m. 148 with the C5. But if one listens to several recordings of this rondo, one will discover
that many performers connect the preceding F4 quarter note with that C5. This feels quite
natural since the F4 is registrally close to C5 and its single-note texture might seem more
similar to the lyrical B theme than to the thick F major chord on the measure’s downbeat.
However natural this feels under the hands, analytically it makes little sense. In the Classical
style, when a B episode (or sonata second theme) returns it does so beginning on the same
beat. Furthermore, transitions often end with two or three triads or open octaves (less often
a single note, as occurs here); these are punctuating articulations that define a formal ending,
not a beginning. This is a good example of how thinking precisely about formal boundaries
can shape a performance interpretation.
329 Rondo Form

In our earlier discussion of B, we noted that it provides intensification through phrase

expansion and registral expansion. Its transposed restatement does so to an even greater extent.
Like the initial B episode, the continuation is expanded due to the imperfect authentic
cadence at m. 156 (analogous to m. 32). Now, however, the perfect authentic cadence is
achieved only after an additional eight measures (at m. 164) rather than four measures (at
m. 36). This further expansion arises through an impressive drawing out of the cadential
6
4 across mm. 158–162, that is, across five measures rather than the single measure (m. 34)
encountered earlier. Since these eight measures are clearly based on a four-measure model,
we have shown two levels of hypermeter. There is a periodic surface hypermeter, but one
can easily infer an underlying hypermeter wherein a third hyperbeat is powerfully stretched
by this stasis on the cadential 64. In addition to delaying the cadence, this passage includes
the highest pitch and introduces the lowest pitch heard thus far—in fact, the highest and
lowest pitches available on the five-octave compass of Mozart’s instrument. This climactic
passage is one of the most virtuosic in all of Mozart’s solo piano music.
At this point, we expect a final statement of the refrain, possibly followed by a coda.
Mozart does something unexpected: he inserts a quasi-improvisatory section, which he labels
as a cadenza, to be played in tempo except for its concluding runs (marked ad libitum). In
Mozart’s numerous concertos with rondo-form finales, a cadenza does precede the final
refrain statement. Thus, he is mixing genres by incorporating a feature from the concerto
into the piano sonata. In his concertos, cadenzas are heralded by a loud orchestral passage
that stops on a cadential 64. This type of preparation occurs in this piano sonata; the cadential
6
4 arrives at m. 171 and is preceded by several forte measures with a relatively unusual texture
for Mozart’s piano music. The left-hand melody in octaves and the violinistic figuration in
the right hand conjure up, as best one can on a keyboard instrument, the impression of an
orchestra surging towards the cadential 64 that is the sign for the concerto soloist to improvise
a cadenza.
In a concerto, the cadential 64 preceding the cadenza resolves only at the V7 near the
cadenza’s conclusion. Melodically, there is usually a corresponding connection between
3^ above the cadential 64 and 2^ above the V7. For many of Mozart’s concertos there exist
cadenzas by the composer, and from these it is apparent that while the intervening material
explores harmonies other than the dominant—sometimes remote ones, often involving
mixture with the minor mode—the organization is such that the underlying sense of the
unresolved cadential 64 is not entirely attenuated. (As the nineteenth century progressed, it
became standard for composers to provide written-out cadenzas in the score, and these
sometimes have a more innovative tonal function.) In order to clarify what is going on in
this piano piece, we provide a voice-leading representation of the cadenza along with an
inventory of its thematic content at (a) in Example 9.2.
The cadenza begins with the refrain melody, but its tonal function is changed due to
its placement above a dominant pedal. The bass moves deceptively in m. 177, but this bass
G serves as a temporary upper neighbor to the dominant. The dominant persists throughout
mm. 179–185 as material drawn from the earlier retransitions occurs. Starting at m. 186,
Mozart develops the eighth-note rhythmic motive characteristic of the retransition, first with
a tonic pedal in the bass (mm. 186–188) and then in the right hand during a descending
fifth sequence (mm. 189–192). Although mm. 186ff. are based on a rhythmic motive from
330 Analysis

EXAMPLE 9.2 Voice-leading representations of cadenza and transition to restatement of B

the earlier retransitions, their pitch content is derived from the transition that leads to the
restatement of the B episode. Example 9.2 at (b) provides a voice-leading representation of
this transition (mm. 128–148) for easy comparison with mm. 186–199 at (a), also in Example
9.2. In both passages, the melody introduces A harmonized with an applied dominant of
IV, but this chord resolves deceptively to ii (recall that ii is equivalent to vi in the key of
IV). In the earlier sequential passage (mm. 133–137), the harmonic rhythm is slow and the
sequence stops at the tonic harmony; in the cadenza’s sequential passage (mm. 189–192),
the harmonic rhythm is faster, driving the sequence past tonic harmony towards vi which
is itself headed to the strong arrival of ii65 at m. 193. Thus, mm. 191–192 do not actually
correspond to anything in the earlier passage, but at m. 193 the correspondence becomes
more exact than ever. Except for the doubling of the harmonic rhythm in m. 195, mm.
193–196 (beat 1) recall mm. 138–142 both in harmony and voice leading. After the downbeat
of m. 196, the close relationship breaks down; note, however, that the overall harmonic
motion remains the same. The bass E prolonged throughout mm. 193–196 passes through
E on to F in m. 198, just as happened more directly in mm. 143–144. As shown in Example
9.2, the upper voice extends 2^ from the onset of ii65 until the arrival of the dominant, although
in the cadenza its elaboration is considerably more extensive. The cadenza comes to the
expected trill on 2^, but here again Mozart throws in a little surprise. In a concerto, the trill
on 2^ generally resolves emphatically to 1^, and this perfect authentic cadence coordinates
331 Rondo Form

precisely with the triumphant re-entry of the orchestra for its final section. In this sonata,
the trill on 2^ dissipates into further figuration that decelerates and leads smoothly back to
the final refrain statement, still marked piano.
Unlike the other refrain restatements, the final one (mm. 200–214) is modified. It consists
of only a single period, but one where the consequent phrase is expanded from four measures
to eleven. The antecedent remains very close to the refrain’s initial antecedent, but the
consequent jumps to forte, a dynamic previously reserved for the second half of each refrain
statement. This dynamic shift signals the elimination of the two-period, soft-loud design in
favor of a single period. The consequent is expanded through the use of I6 (m. 207, m. 210)
to delay the cadence. Within the phrase expansion note how the final push to the cadence
(mm. 210–213) is itself expanded compared to the preceding unsuccessful attempt (mm.
207–209) through the spreading of the cadential 64 and its resolution across two measures
(mm. 212–213). We observed the same technique in the B section (see mm. 34–35), and
as in the B section, this extra measure places the end of the final refrain statement on a
hyperdownbeat (m. 214) and thereby gives a bit of extra weight to this moment of structural
close.
Since m. 214 corresponds to the end of the refrain material and features a perfect authentic
cadence, we have designated the remainder of the movement as a coda (mm. 215–225).
The “after-the-end” quality is enhanced through the immediate introduction of 7^ as part
of V7 of IV. Unlike the iterations of this harmony in the cadenza and in the transition leading
to the restatement of B, this one resolves to IV (not deceptively to ii). The persistence of
the tonic pedal in the bass and the shorter lengths of melodic units contribute to a contented
conclusion for this bold and expansive movement.

Brahms, Violin Sonata in D Minor, op. 108, IV

Our second rondo form poses numerous analytic challenges, many of which relate to defining
aspects of Brahms’s language. Foremost among these is the lack of strong arrivals on tonic
harmony throughout most of the movement. The deferral of tonal closure occurs on several
levels of structure. Also of great importance is Brahms’s elaborate development of thematic
materials, both within individual sections and between them. These features make the
identification and labeling of sections more difficult than in Mozart’s rondo. Expressively,
these elements contribute to the movement’s dramatic sweep towards its tragic conclusion.
As we undertake a detailed discussion of Brahms’s movement, bear in mind that reinterpreting
Classical forms and rendering formal boundaries less clear are typical of much of Brahms’s
music. Figure 9.3 provides a formal-tonal outline of the movement, and Example 9.3 offers
an annotated score.
The movement’s first formal quirk occurs at the end of m. 4. After four measures the
music suddenly stops and then begins again in m. 5 with the violin spinning a true melody
from A5–E5–F5, the pitches in the top voice of the piano chords in mm. 1–2 (labeled x in
Example 9.3).3 The first four measures seem like an anticipation of the movement’s true
beginning—a “false start”; however, this introductory function is not clear because mm.
1–4 are in the movement’s main tempo, not in a slower one. This ambiguity about the
formal function of mm. 1–4 is why we have placed their introductory label in quotation
marks in Figure 9.3. Since rondo forms involve multiple returns of their opening materials,
332 Analysis

“Intro” 1–4 V – iv

REFRAIN (A) 5–16 V

Transition (Tx) 17–38 i – V/v

EPISODE 1 (1B)
antecedent 39–54 C (VII): I – V7[iii]
consequent expanded 55–72 I – a (v): VI
Transition (Ty) 73–76 a (v): V

EPISODE 1 (2B)
antecedent 77–84 V
consequent expanded 85–107 V–i
Transition (Ty) 108–113 V

“Intro” 114–117 V – d (i): iv

REFRAIN (A) 118–129 d (i): V

EPISODE 2 (Development)
part I 130–176 g (iv): i – f (iii): i
part II 176–194 f (iii): i – d (i): V
Transition (Tx) 194–217 d (i): V

EPISODE 1 transposed (1B)

antecedent 218–233 F (III): I – V7[iii]
consequent expanded 234–251 I – d (i): VI
Transition (Ty) 252–255 d (i): V

EPISODE 1 transposed (2B)

antecedent 256–263 V
consequent expanded 264–286 V–i
Transition (Ty) 287–292 V

“Intro” 293–296 V – iv

REFRAIN (A) 297–311 V–i

Coda 311–337 i

FIGURE 9.3 Outline of Brahms, Violin Sonata, op. 108, IV

333 Rondo Form

EXAMPLE 9.3 Brahms, Violin Sonata in D Minor, op. 108, IV

334 Analysis

EXAMPLE 9.3 continued

335 Rondo Form

EXAMPLE 9.3 continued

336 Analysis

EXAMPLE 9.3 continued

337 Rondo Form

EXAMPLE 9.3 continued

338 Analysis

EXAMPLE 9.3 continued

339 Rondo Form

EXAMPLE 9.3 continued

340 Analysis

EXAMPLE 9.3 continued

341 Rondo Form

EXAMPLE 9.3 continued

342 Analysis

EXAMPLE 9.3 continued

343 Rondo Form

EXAMPLE 9.3 continued

344 Analysis

the ambiguity about whether mm. 1–4 constitute an introduction to the refrain or the first
segment within the refrain is a rich compositional seed for Brahms to sow.
Starting in m. 5, the music proceeds with superficial regularity as the x motive launches
not only m. 5 but also m. 9 and m. 13. Closer inspection reveals that this strongly projected
hypermeter does not correlate with solid tonal pillars. In the initial statement of the x motive
(mm. 1–2), the terminal F5 was harmonized by root-position tonic harmony, but it occurred
in the midst of an ongoing melodic sequence. The larger harmonic connection is between
the dominant of m. 1 and that of m. 5, with the subdominant harmony of m. 4 providing
neighboring bass motion in between. When the violin takes up the x motive in mm. 5–6,
the F is a dissonance against dominant harmony and ultimately resolves with a 7–6 motion
against the bass of ii65 (m. 7). In the next statement, the F is replaced by F, which is
harmonized with an applied dominant to iv. At m. 13, the x motive is transposed to begin
from 1^ instead of from 5^. Here again tonal stability is somewhat undermined by the inclusion
of F in the harmonic support of 1^, allowing mm. 13–16 to begin with an exact transposition
of mm. 1–3 into G minor (iv). Thus, the ambiguity about the formal function of mm. 1–4
is reengaged as its transposed version now serves as the culmination of the rondo refrain.
This culmination seems headed for a powerful perfect authentic cadence, but at m. 17 the
violin drops out, the dynamics suddenly are soft, and the chordal texture vanishes. These
changes affect the degree of tonal resolution between m. 16 and m. 17. Certainly, there is
no melodic resolution, but the harmonic resolution is also reduced. While we would not
rule out the possibility of hearing an elided imperfect authentic cadence at m. 17, the passage
is closer to a half cadence in m. 16 with the root-position tonic harmony in m. 17 serving
as a launch to the transition but not also as a conclusion to the refrain. Either way, the degree
of tonal closure is significantly less than typically found in this formal location in rondos by
Haydn, Mozart, Beethoven, or Schubert.
As expected, the transition (mm. 17–38) modulates and ends on the dominant of a new
key, here V of A minor (v).4 A modulation to the relative major most frequently prepares
for the B episode in a 7-part minor-mode rondo, but the minor dominant is the next most
common choice. Based on a sixteen-measure sentence (with the goal V harmony expanded),
the transition is quite unremarkable except for its deft handling of motivic materials. The
transition opens with a new melodic idea moving in eighth notes, which we will refer to
as motive y, and it is shown by dotted brackets on Example 9.3.5 In the initial two statements
of motive y, one is drawn to the interplay between the descending third F4–E4–D4 (m. 17)
and the ascending third C4–D4–E4 (m. 18), circled in Example 9.3. In mm. 17–20 and
mm. 21–24, piano iterations of motive y are contrasted by forte intrusions of motive x. In
the sentence’s continuation, the statements of motive y in the violin preserve their ascending
thirds (though beginning not with a consonant pitch but with a dissonant D4); those in the
piano forgo their descending motion in favor of large ascending leaps that prepare the return
of motive x. Motive x, however, is varied (x′) through the addition of an appoggiatura (A5
in m. 30; D5 in m. 31). This appoggiatura allows one to hear stepwise motion within the
B5–G5 and E5–C5 thirds, thereby incorporating into motive x an aspect of motive y. When
the goal V arrives in m. 33, stepwise motion through a third is all that remains of the thematic
material (y′); simplification of melodic content often accompanies dominant pedals at the
ends of transitions, but Brahms’s distillation of an element embedded within y and
subsequently incorporated within x is particularly notable.
345 Rondo Form

The juncture between the end of the transition and the start of B involves both continuity
and surprise. Melodically, B grows directly from the transition; as shown in Example 9.3,
an augmentation of y′ in the bass connects the end of the transition with the start of B,
which has a melody that begins with another statement of y′. This is a good example of
linkage technique, whereby a motive near the end of a formal unit is immediately deployed
to initiate the next one. The surprising aspect of B is its tonal design; it touches on several
key areas before eventually confirming A minor at m. 104 with a perfect authentic cadence,
the strongest cadence in the movement thus far. We will now examine the tonal structure
and phrase organization of B in more detail.
The B episode contains two periods with contrasting thematic material and keys. Given
these strong contrasts, we have chosen to differentiate these periods as 1B and 2B. As noted
above 1B has a smooth thematic evolution from the transition, but the expected key of
A minor is displaced by its relative major, C major. This initial C major harmony is elaborated
through plagal motions, which raise the spectre of F major. The sixteen-measure antecedent
phrase features an ascending fifth sequence (mm. 43–48) that leads to the dominant of
E minor (mm. 48–54). Notice the treatment of the melody at the highpoint of the phrase.
The pitches B3–E4–G4–F4–A4–G4–F4 are first given by the piano (mm. 47–50) and are
then imitated by the violin. The violin’s imitation is partial as the melody reverts to the
piano at the end of m. 52; observe here the careful upward stemming of the F3 to highlight
the piano’s completion of the violin’s imitation. Due to the modulation within the antecedent
phrase, the beginning of the consequent back on C major sounds like a deceptive progression
within E minor. In fact, C major is only briefly asserted in the consequent, which quickly
shifts into F major and then emerges from its ascending-fifth sequence (mm. 61–65) in
A minor. The violin, which assumes a leading role during the consequent, provides the
melody at the phrase’s highpoint: E5–A5–C6–B5–D6–C6–B5–A5 (mm. 65–68). This is a
transposition of the end of the consequent from E minor into A minor, but with a terminal
1^ (here A5) included within the four-measure unit. This final note is harmonized by an
F major chord, VI of A minor; thus, the deceptive motion heard between the end of the
antecedent and the start of the consequent is now heard within the consequent’s final
measure. And, just as at the very start of 1B, F major—the most usual key for B in a 7-part
rondo in D minor—seems to be trying to surface. This impression is enhanced through the
phrase’s striking penultimate chord. While the upper pitches articulate V7 of A minor, the
pianist’s left hand plays 5^ of F major, C, rather than the expected E. The result is a hybrid
harmony that conflates elements from the dominant sevenths of A minor and F major! From
a larger perspective, it is clear that the terminal F major functions as VI in A minor, but the
preceding bass motion renders it somewhat more stable, as does its placement as goal of an
entire musical period.6
Following a repetition of the cadence, a brief transition leads to 2B and back to V of
A minor. Unlike 1B, 2B begins in A minor, but it still does not assert root-position tonic
harmony. Like 1B, 2B consists of a period with expanded consequent but differs in three
important respects. First, the roles of the violin and the piano reverse: in 2B the melody is
led by the violin in the antecedent and by the piano in the consequent. This is typical in
duo sonatas in the Classical tradition. Second, 2B possesses a feeling of acceleration owing
to the faster surface rhythms, the shorter phrase length of the antecedent (eight measures
compared to sixteen in 1B), and the obvious sentential design of the phrases with
346 Analysis

fragmentation at the outset of their continuations. Third, the expansion of the consequent
phrase in 2B is longer and more elaborate. In 1B, the expansions arise internally through
one-measure elongations of two harmonies (m. 57 and m. 60), but in 2B there are multiple
expansions. Initially, the consequent’s continuation is expanded through thematic fragments
that span not one measure—as one expects at the outset of the continuation of an eight-
measure sentence—but 1.5 measures (mm. 89–91). These unusual lengths support reading
two hyperbeats across three notated measures, an interpretation that responds to the melodic
development and preserves quadruple hypermeter. These sequential measures lead from
A minor to D minor, a shift emphasized through the deployment of augmented sixth chords
(m. 93 and m. 95). An additional passage (mm. 96–104) reattains A minor through a
chromatic octave ascent in the bass that connects the A major chord, functioning as V of
D minor, with an A minor sonority. The rising bass line, syncopated rhythms, and crescendo
imbue this phrase expansion with powerful energy that culminates with the melody found
at the highpoint of 1B (cf. mm. 104–107 with mm. 65–68), but this time achieving a perfect
authentic cadence in A minor. Thus, the drive to this delayed cadence is capped off with a
figure that had been left unresolved during 1B, and this thematic recycling helps draw
together 1B and 2B into a span of nearly eighty measures all headed towards this perfect
authentic cadence in A minor.
The open-octave transitional material that occurred between 1B and 2B returns after
2B (mm. 108–113), but it now leads to [o7] V in A minor—an odd goal given that a refrain
statement in the home key of D minor is expected. Instead, the quasi-introductory material
from mm. 1–4 comes back in A minor. More precisely, its first three measures occur in
A minor (mm. 114–116) and its final measure is in D minor (m. 117). This key change is
extremely jarring because it creates the progression (in D minor) of V – iv between m. 117
and m. 118. Given its “wrong-key” beginning, this four-measure unit seems even more
introductory than in mm. 1–4. The rondo refrain then occurs in its entirety and without
variation (mm. 118–129). Since the refrain is not followed by a section beginning with a
D minor chord (or even in the key of D minor!), its ending on V is highlighted. We are
reminded of the melody’s unresolved C due to the linkage technique wherein the D5–C5
from the refrain’s terminal measure is taken up in the chordal accompaniment of the following
section.
At the center of this rondo lies a development section based almost entirely on the initial
measures of the refrain (mm. 5–8). The development proceeds in two parts (mm. 130–176
and 176–194), both driven by sequences. In the first part, the thematic derivation is
obscured by the slower rhythms and softer dynamics, but a comparison of the melody in
mm. 5–8 with that in mm. 134–137 reveals that the latter is a transposed (and reharmonized)
simplification of the former: A–E–F–E–(G–B)–D–(C–B–A–G)–G becomes G–D–E–D–
C–F. In the second part, the thematic connection is more explicit, as the eighth notes in
mm. 176–177 clearly arise from those in mm. 7–8. The sequences in the second part are
similarly straightforward, proceeding via ascending fifths from F minor (m. 176) and C minor
(m. 181) through the key of G minor (m. 187) to the ultimate arrival on the dominant of
D minor (m. 194). In the first part, the sequences are more unusual, and we provide a
simplification in Example 9.4. Initially, the sequence outlines eight-measure units in rising
minor thirds: G minor (m. 134), B minor (m. 142), and C minor (m. 150). This passage
has a strong variation component, as the melody always sounds in a different part of the
347 Rondo Form

texture and the rhythmic activity continually increases. Thus, the simplification in Example
9.4 shows the “normative” voice leading for the chords involved in the sequence, but this
constitutes the actual voice leading only in the B minor unit. Elsewhere, several of the chords
occur in inversions different from those shown in Example 9.4, including the all-important
augmented sixth chords. At m. 158, a fourth sequential unit in E minor could have begun,
but instead the music continues V of E minor from the preceding measure. The eight-
measure sequential unit is fragmented to its second half, where the modulation to the next
key actually occurs. Beginning with the bass D in m. 158, this shorter sequential module
leads to F minor (m. 161) and A major (m. 163). Notice the acceleration of the harmonic
rhythm in m. 163 that contracts the latter sequential unit. After m. 163, the sequence
fragments further such that the German augmented sixth follows directly after the A major
chord; as a result, this unit moves up by a major, rather than minor, third to C major, which
Brahms immediately respells as D major. At this point, the goal of F minor could have been
reached through one further sequential unit with bass line D–C–F; however, the C that
occurs in m. 167 supports a passing 64 chord that leads to a new sequence with an urgently
brief two-chord sequential unit. The C that supports the cadential 64 arrives only at the

EXAMPLE 9.4 Simplification of sequences in part I of development (mm. 130–176)

348 Analysis

fortissimo in m. 171, which resolves to F minor at m. 176. The progressive reduction in the
number of chords in sequential units combines with the acceleration in harmonic rhythm,
syncopations, and crescendo to produce a powerful—but tonally remote—climax in the middle
of this movement.
As noted above, the second part of the development leads to V of D minor (m. 194),
and this is the expected tonal goal. Since this movement has been proceeding along the lines
of a 7-part rondo, we anticipate a retransition will lead to a restatement of the refrain.
However, the material in mm. 194ff. is based on the transition that led from A to 1B, and
Brahms does indeed omit the expected return of A. This rondo, then, is a variant of the 7-
part rondo that can be summarized as ABACBA, a 6-part rondo not typical of Classical
practice but found in a small number of rondos by Mozart. Moreover, Brahms makes a
critical change in the transition material: it is entirely set over a dominant pedal (except for
the double-neighbor motion in mm. 210–211). Brahms thereby not only avoids thematic
return but also tonal return. This is a wonderful illustration of the importance of remaining
aware of large-scale tonal structure in addition to chord-to-chord harmonic progression when
examining music in the Classical tradition.
The B material returns without alteration, except for transposition down a perfect fifth.
Due to its meandering tonal design, this means that a perfect authentic cadence in D minor
is not reached until m. 283. Recall that the transition after the initial statement of 2B led
to [o7] V in A minor; thus, upon transposition down a fifth, [o7] V in D minor is attained
(m. 292), and this leads smoothly into a restatement of mm. 1–4 (in mm. 293–296) to prepare
the long-awaited refrain restatement (mm. 297ff.). Except for a coloristic reharmonization
of the F in m. 302 and a two-measure phrase expansion (mm. 305–306), the refrain parallels
those heard earlier. The enormous change occurs in the measure directly after the refrain
where the pianist provides a thunderous D minor chord (m. 311). Unlike the piano D minor
chord of m. 17 or the abrupt move to G minor after the second refrain statement, this event
provides solid tonal closure to the refrain. Experiencing this moment is the payoff of the
multiple denials of tonic arrivals earlier in the movement.
The coda returns to materials from the transition that immediately followed the initial
refrain. The lilting figure of the transition’s opening (mm. 17–19) is transformed from a
gesture that leads away from D minor into one that confirms the tonic key (mm. 313–315).
The coda then turns to refrain material (mm. 325ff.), specifically its final four measures (mm.
13–16). Neither the violinist nor the pianist can complete the phrase; both get stuck on F
with a harmony that includes E, a pitch always present in the refrain’s penultimate measure
(m. 15, m. 128, m. 309) and through which the melody always descended. After the poco
sostenuto, the players regain their momentum and achieve not only harmonic closure but
also melodic closure that involves a descent through diatonic 2^ (see circled pitches on the
score in mm. 331–335). The coda is still manipulating thematic materials and rendering them
more tonally stable. As such, it provides further evidence of Brahms’s penchant for large-
scale thematic development and for giving all sections of the form dense and piece-specific
thematic content.7 Where composers of the late eighteenth century often allowed closing
sections to contain relatively generic melodic materials, Brahms and many other composers
of the later nineteenth century maximized the potential of thematic content within the
boundaries of the common-practice tonal system, and in this way looked forward to the
steps beyond that system in the twentieth century.
349 Rondo Form

SUGGESTED ASSIGNMENTS

Beethoven, Violin Sonata in A Major, op. 12, no. 2, III

1. Consider mm. 53 and 54. Which of these measures should be considered as the first
measure of a phrase? Why?
2. There’s a very unexpected rest in m. 79, but what does the “breaking down” of the
music in these measures also get rid of? (Tonally, what is typically going on at this spot
in rondo form?)
3. In m. 120, the piano left-hand note is marked . What connection is Beethoven
highlighting here, and what term describes this type of thematic reworking?
4. What is the form of mm. 120–183? Is this section more like a C section or a
development?
5. In mm. 202–205, there are some rather unusual accidentals (for a piece in A major).
(a) Based on these accidentals, what key is expected in m. 206?
(b) How does the key identified in (a) relate to the two keys that are tonicized in the
two phrases that immediately precede mm. 202–205? (Is there any pattern?)
6. As is often the case, the final statement of the refrain is altered.
(a) Explain precisely what is altered (measures omitted? added?).
(b) Explain how Beethoven prepares the final refrain, and explain why this preparation
might have suggested one of the alterations identified in (a).
(c) How do the alterations affect the strength of the refrain’s final cadence?
7. Quite near the end a very interesting rhythmic effect occurs. Where is this, and what
is it called?

Beethoven, Piano Sonata in D Major, op. 28, IV

1. Compare the initial statement of the rondo refrain with subsequent statements of the
refrain.
2. Are there any similarities between the rondo refrain and the thematic materials of the
episodes? If yes, describe.
3. Comment on the thematic content of the transitions and/or retransitions (if present).
Do any of them re-use the same thematic material? If yes, describe.
4. Some rondo forms are referred to as “sonata-rondos.” What feature(s) of this movement
would be found in a “sonata-rondo” form, and what feature(s) of this movement are
not typical of a “sonata-rondo” form?
350 Analysis

Beethoven, Piano Sonata in G Major, op. 31, no. 1, III

1. Compare the initial statement of the rondo refrain with subsequent statements of the
refrain. What are the main types of changes introduced in later refrain statements? Which
statement of the refrain is most similar to the initial one?
2. Are there any similarities between the rondo refrain and the thematic materials of other
sections of the form?
3. Compare mm. 200–205 with the opening refrain. What thematic connection can be
observed, and how is the motive transformed from its initial appearance in the refrain?
4. Consider the chord in m. 236. It is marked «ƒ.
(a) What term does «ƒ stand for, and what does it mean?
(b) This is the only time Beethoven uses this marking in the movement. What is so
special or unexpected about the chord in m. 236?

Mozart, Piano Sonata in C Major, K. 309, III

1. At what point does phrase expansion occur within the refrain? How is this moment
manipulated later in the movement?
2. How does Mozart delay local closure within the B episode? What happens to the moment
of closure in the home key restatement of the B episode?
3. What formal similarity exists between this rondo and the one by Brahms studied in this
chapter?
4. Comment on the thematic contents of this movement’s extensive coda. How does
Mozart signify a closing quality in the reminiscence of the refrain in mm. 245ff.?
5. Trace the use of the perfect fifth in the thematic material of each section of this rondo.

Schumann, Piano Trio in F Major, op. 80, II

1. Compared to “Classical” models, what is unusual about the tonal design of the refrain?
How is this feature handled (or not handled) in the final statement of the refrain?
2. The chord on the last beat of m. 13 serves as a pivot chord. How would it be analyzed?
3. What is unusual about the key choice for the second refrain statement?
4. Consider the B episode and its transposed restatement.
(a) By what interval is it transposed when it is restated?
(b) Other than transposition, in what way is the B material altered during its restatement?
Appendix: Notes On Musical
Reductions

In this book we have employed two types of reductions, referred to as simplifications and
representations of the voice leading, respectively. The first of these is really a metric
simplification of the score, where note values represent relative duration. The purpose of
these simplifications is to reveal underlying voice leading and often, as a result, to clarify the
harmony. The process of preparing a reduction of this type involves removal of notes of
melodic embellishment and normalization or realignment of rhythmic displacements,
including suspensions. The process often involves registral normalization as well, particularly
in the bass, to show voice-leading connections more clearly. The result should be a
recognizable representation of the musical passage, where an assigned note value no longer
represents the written value in the score but the total duration of a note and its melodic
and/or rhythmic elaborations. Reductions of this sort may vary from person to person
depending on how much an individual decides to eliminate or retain in the reduction. In
other words, there are different possible “correct” solutions depending on what one seeks
to demonstrate through this process.
Consider the opening phrase of the B major Prelude from Book I of the Well-Tempered
Clavier, which is provided at (a) in Example A1.1 An important feature of this phrase is the
repetition of a melodic-rhythmic figure that occurs in varied form, moving from one voice
to another in a three-voice contrapuntal setting. With two exceptions, the pattern elaborates
the first note, whether occurring on the beat, either struck or held over from the previous
beat, or sounded on the second sixteenth following a sixteenth-note rest. The two exceptions
occur in the second measure, where the motion to F5 on the second beat (top voice)
prolongs E5 and the motion to B4 on the fourth beat (inner voice) temporarily displaces
the underlying 7–6 suspension. Removal of these melodic embellishments is the first stage
in preparing the reduction at (b). In this instance the 7–6 suspension just noted and its later
repetition with the two parts inverted have been retained because of their perceived
importance in the construction and flow of the phrase. Note the registral normalization in
the bass in the second half of m. 4 and the rhythmic normalization of the inner voice in the
second half of the following measure, which helps to clarify the underlying voice leading
and harmony. The result of these steps—removal of melodic embellishments, the noted
rhythmic and registral normalizations, but the retention of the initial 7–6 suspension and its
varied repetition—is the metric simplification at (b).
Our second type of reduction is an interpretation of the voice leading, very much like
a Schenkerian graph without invoking all the terminology and notational procedures
352 Appendix: Notes on Musical Reductions

associated with that approach. Here note values do not represent relative duration, but rather
are used to represent hierarchical levels of the voice leading. Such an interpretative
representation of the Bach phrase is provided at (c) in Example A1. In this graph the opening
melodic gesture is shown to introduce D5 in the second measure via its upper neighbor
note, marked N and given the value of an eighth note to show this neighboring relationship
to the note to which it resolves, which has been assigned the value of a quarter note. The
diagonal line connecting the D5 to the bass note B3 indicates their association. Following
the establishment of D5, the top voice is shown to progress by step down to F4 at the
cadence with members of the descent all represented by quarter notes with stems of the
same length. This descending sixth is further delineated by a large slur. Extension of G4 in
this encompassing sixth by the descending third G4–F4–E4 is shown to occur at two levels,
first in the immediate context, represented by unstemmed note heads grouped by a slur,
which is embedded within the harmonized descent represented by quarter notes with short
stems and once again grouped by a slur. In this more encompassing version the passing note
F4 is provided temporary harmonic support by vi in the key of the dominant. This deceptive
progression is shown to exist within the prolongation of the supertonic harmony leading to
the cadence. Slurs in the inner voice and bass indicate extensions of the underlying step
progressions, and the Arabic numerals between the staves indicate the intervallic relationship
between the outer parts. Though the principles of notation employed here are simple enough,

EXAMPLE A1 Bach, Prelude in B Major (WTC I), mm. 1–6

353 Appendix: Notes on Musical Reductions

EXAMPLE A2 Mozart, Piano Sonata in F Major, K. 280, I, mm. 17–26

it should be clear from the description above that the decisions involved in this type of
interpretative reduction are far more complex than those leading to a metric simplification
of the voice leading. This is especially true of the cadential pattern in mm. 4–6.
Two additional examples are provided below. The first of these, the transition passage
leading to the second theme in the first movement of Mozart’s Piano Sonata in F Major,
K. 280, was discussed briefly in Chapter 2 (Example 2.14), where the futility of labeling the
chord progression connecting I (m. 17) and I6 (m. 22) was noted. This passage is reproduced
at (a) in Example A2, and a metric simplification of the voice leading is provided at (b). As
can be seen in the reduction, the progression of mm. 18–22 is a series of descending 63 chords,
each introduced by an applied o43, thus preserving the parallel tenths between the outer parts.
As shown below the system, the entire passage is controlled by the extended progression
I – ii6 – V6–5
4–3.
2

An interpretation of this passage is provided at (c) in Example A2. Here the passing
motion connecting I and I6 is shown to occur within a large-scale voice exchange prolonging
the initiating note A5 and the tonic harmony; both the A5 and the supporting bass note
have been assigned quarter-note values with long stems. The end points of the voice exchange
are also given stems, but short ones to indicate their role in the prolongation of A5/I. The
intervening notes are not stemmed. This extension of tonic harmony is followed by the
elaboration of both G4 and G5 by arpeggiation of the members of the supporting supertonic
354 Appendix: Notes on Musical Reductions

harmony. G5 is provided a stem equal to the A5 some seven measures earlier, and, though
the cadence follows in the lower octave, this G5 prepares the initial melodic note of the
second theme in this upper register (not shown in the example). Here length of stems is
used to indicate visually the long-range connections in the outer parts.
Our final example is the opening four measures of the sarabande from Bach’s first Cello
Suite, which is provided at (a) in Example A3. Though we are focusing primarily on later
repertoire in this book, this example was chosen specifically to illustrate how to deal in
reductions with music written for non-keyboard solo instruments, where lines often
disappear temporarily due to practical limitations of the instrument. A metric reduction of
these measures is provided at (b), which retains the second-beat emphasis characteristic of
the sarabande rhythm. Though the bass line temporarily drops out, we understand the material
up to the second quarter of m. 3, where the subdominant harmony enters, to be controlled
by G (I), so that bass note G has been added twice in parentheses. More important to our
understanding of this passage is the C4 over subdominant harmony that is implied by the
context; it too has been added in parentheses and connected to the C4 that is introduced
as a dissonance in the next measure. Finally A3 has been added above the D major chord
(V) at the cadence, clearly implied as the local goal of the preceding C4–B3 motion in the
top-sounding line.
Our interpretation of the voice leading is provided at level (c). The content of mm.
1–2 is shown to prolong tonic harmony and B3, the latter by its upper neighbor note while
the inner voice rises from D3 to G3. The leap in the top voice to D4 while the inner voice

EXAMPLE A3 Bach, Cello Suite in G Major, Sarabande, mm. 1–8

355 Appendix: Notes on Musical Reductions

descends to F3 predicts the following C4 over subdominant harmony. We might expect
this implied C4 to return to B3 as it had before, but this time the motion passes through
B3 on its way to the implied A3 at the cadence. The chords during the first beat of m. 4
serve as an elaboration of the underlying harmonic connection between the preceding
subdominant and the goal of the phrase, the dominant. As before, our use of note values—
quarter, eighth and unstemmed notes—indicates our interpretation of the levels of voice
leading, and in this case our understanding of this hierarchy is dependent in part on our
awareness of notes missing but clearly implied by the context.
Notes

Preface
1 Taken by itself, Part I is appropriate for a graduate review of tonal theory where the focus is on
understanding musical structure rather than part writing.
2 For those who want to pursue this topic, we refer you to Chapter 10 (“Text and Music”) in
David Beach’s Advanced Schenkerian Analysis (Routledge, 2012), which contains detailed studies
of vocal works by Mozart, Schubert, Schumann, and Brahms.

1 A Review of Diatonic Harmony

1 Figured bass is a shorthand system widely used in the seventeenth and eighteenth centuries to
indicate to the keyboard (or lute) player the intervals required above the given bass line. The
figure 6 for first-inversion triads is actually a short form for 63.
2 Theoretically, all inversions of all triads are possible, but such a listing gives the false impression
that all have equal status, which, as we shall see, they do not.
3 Subtonic is sometimes used in the minor mode to designate the triad below the tonic, that is, the
triad built on the natural seventh degree of the scale.
4 In this opening section we have covered a lot of ground, and in doing so we have assumed a
secure knowledge of fundamentals: scales and intervals. If this is not the case, security in these
fundamentals, as well as chord construction, must be attained before continuing into the study
of harmony.
5 This relationship is sometimes referred to as a question (antecedent) and answer (consequent).
We will explore phrase design in more detail in Chapter 3.
6 This type of non-chord tone, which is approached by leap, is sometimes referred to as an
appoggiatura.
7 Note that the diminished triad (viio) is among the first-inversion chords. While diminished triads
can occur in root position during sequences, composers tend to avoid them even in these
situations.
8 The different types of reductions and their criteria for construction are outlined in the Appendix.
9 The complete descending fifth sequence occurs more frequently in the minor mode because it
employs the natural seventh degree of the scale (except with the dominant), thus avoiding the
diminished triad viio.
10 Hemiola involves a conflict between two and three, here created by the repeated duple pattern
against the triple meter.
11 Heinrich Schenker, Harmony, English translation by Elizabeth Mann Borgese (Chicago, 1954),
133–153.
357 Notes

12 This relationship will be discussed further in Chapter 3, where it is shown to be an important

component in the formal design known as a “musical sentence.”
13 A voice exchange occurs only between the same two parts/voices.

2 Expanding the Diatonic Palette

1 Secondary chords are often referred to as “applied” chords.
2 One might argue that a reading of ii6 in m. 43 is equally as logical, but after having heard F so
strongly in the previous measure, IV seems the more logical choice. In general, the harmonic
interval of a third is interpreted as a root and third of a chord, not the third and fifth, unless
there is evidence to suggest the contrary.
3 The effect on the voice leading of substituting a diminished seventh chord for the expected
dominant is that the leading tone of the first chord (here A) does not lead as expected to B, but
instead falls by semitone to A .
4 It is significant that the pattern changes as the descending melodic line reaches D5 (seventh of
the E chord), since an important motivic component of this work is the decoration of C5 by
its upper neighbor D5.
5 Note values in Example 2.9b do not represent durations.
6 The C5 above the vi chord is implied by the context, but not stated in that octave and is thus
placed in parentheses in Example 2.11b.
7 As with Example 2.9b, this graphic representation of the voice leading does not represent relative
durations. More will be said about this near the conclusion of this chapter and in the Appendix.
8 The secondary o65 maintains the expected bass pitch A (i.e., as in I6).
9 Having the pivot occur on a seventh chord is relatively rare, resulting here in the retention of
the bass note G3 to become the seventh of the following chord.
10 The title Grande sonate pathétique was given to this work by Beethoven’s publisher, who was
impressed by its tragic qualities.
11 The astute reader might have noted that we have not been entirely consistent in how we have
notated pivots involving a 5–6 motion. Under normal circumstances we have taken the actual
pivot to be the chord represented by the 6 in 5–6. For example, I5–6/ii6 in Examples 2.16 and
2.18, or i5–6/iv6 in Example 2.26. In one instance we have shown the pivot differently, because
of the way the harmony is introduced (by its secondary dominant) and its continuation. See
Example 2.22. A certain amount of flexibility is needed to adjust to different circumstances. That
is, we must be sensitive to the context in which a musical event occurs.
12 This terminology follows that used by William Rothstein in Phrase Rhythm in Tonal Music
(Schirmer, 1989). Though employed rarely in our text, this terminology is useful in circumstances
like this to describe two phrases based on contrasting, as opposed to similar, melodic material
that forms a musical period. Phrase design will be discussed further in Chapter 3.

3 Phrase Design
1 We acknowledge the particular influence of two books on our presentation of phrase design and
expansion: William Rothstein, Phrase Rhythm in Tonal Music (Schirmer, 1989) and William E.
Caplin, Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart,
and Beethoven (Oxford, 1998).
2 When followed by a rest (or extended by a fermata), a V7 chord—or in very rare instances V6
or an inversion of V7—can conclude a phrase. Such instances are best viewed as exceptional
cases where a phrase concludes without a normative cadence.
358 Notes

3 A caret placed above an Arabic numeral is a shorthand notation for “scale degree.”
4 Our discussion of sentence design follows closely the terminology of William Caplin, as presented
in his book cited at the outset of this chapter. Since Caplin’s definition of “phrase” differs
considerably from ours, however, he describes the sentence as a two-phrase construction. It might
be of interest to know that Caplin’s theory in turn amplifies observations made by one of the
leading composers of the twentieth century, Arnold Schoenberg.
5 Since this is our first orchestral score, we will comment briefly on reading orchestral scores. First,
note the larger brackets at the left edge of the score to designate the instrumental families: winds,
brass, percussion, and strings. Within the winds—flutes, oboes, clarinets, and bassoons—the
clarinets are special. The notated pitch in a clarinet part is not the same as its sounding pitch.
Here, the clarinets are “in A,” which means that they sound A when C is notated; thus, the
sounding pitch is a minor third lower than the notated pitch. This is why the clarinet lines use
the key signature of F major—the key of F major is a minor third higher than the key of D
major. (The other type of clarinet in widespread use is “in B,” which sounds a major second
lower than the notated pitch. We will encounter this clarinet transposition in the next chapter,
Example 4.14.) The designation “a2” in flutes and bassoons indicates that both players are active,
playing in unison. In the brass, Haydn only deploys horns and trumpets in this symphony. In
this work, both are “in D,” which means that the notated C sounds as D (the horns sound a
minor seventh lower, the trumpets a major second higher). As is usual in the Classical repertoire,
the only percussion instruments are the two timpani tuned to tonic and dominant pitches; in
this excerpt only the timpani tuned to D is used. Finally, the strings—first violins, second violins,
violas, cellos, double basses—occupy the lowest four staves. The cellos and double basses are
notated together, but the double basses sound an octave lower than notated.
6 Some authors refer to hypermetric reinterpretation as metric reinterpretation. Since it is a
hyperbeat that is reinterpreted (not a beat), we prefer the more explicit term hypermetric
reinterpretation.
7 As in the Haydn symphony, the clarinet is in A, which means that it sounds a minor third lower
than notated. In the text, we always refer to the sounding pitch of the clarinet.
8 Incidentally, at the end of the trio, this period is further expanded. The consequent is expanded
by eight, rather than four, measures; the first three measures of the eight-measure expansion are
basically the same as in the earlier expansion.

4 Further Expansion of the Harmonic Palette

1 The brackets in Example 4.2 indicate two levels of motivic activity, the shorter idea beginning
with the dotted quarter and the larger gesture initiated on the downbeats of mm. 3 and 5.
2 Though the scale is much larger here, note that the excursion to III is brought about by a 5–6
linear motion, similar to the approach to VI in Schubert’s “Du bist die Ruh.”
3 A common pattern involving a voice exchange is IV6 – passing 64 – ii65, with or without chromatic
6
alteration. Here this common pattern appears as II 4 [ ] IV.
3
4 Note also that the space between E2 (V) in m. 41 and G2 (V65) in m. 44 is filled in by the passing
tones F – G (supporting IV – V in C), though an octave higher.
5 In Chapter 2, we did not explore the use of half-diminished seventh chords as secondary chords,
since they are used in this way much less frequently than are diminished seventh chords. Half-
diminished sevenths can only serve as secondary chords to major triads. For another example, see
Example 4.10b.
6 The substitution of E for F would normally occur if the line were leading up to F, so this is
particularly unusual here, since the line is descending to E.
359 Notes

7 The Lebewohl motive is the well-known horn call with open fifth. The harmonization of the
third interval, the sixth, first by vi and later by VI distorts the pastoral expression normally
associated with this idea.
8 The clarinet parts sound a major second lower than written.
9 Another example of modulation to the major mediant is found in the first movement of
Beethoven’s “Waldstein” Sonata, op. 53, which is in C major. The second theme is in E major,
but in this instance the pivot is diatonic as the modulation proceeds first from C major to E
minor with pivot vi6/iv6 (m. 21).
10 This is the quickest way to return from VI to I, namely, by changing the chord above  6^ into
an augmented sixth chord leading back to V in the original key.

5 Binary and Rounded Binary Forms

1 When simple binary occurs in the late eighteenth century, it does so most often in themes for
sets of variations.
2 For this reason, some authors refer to binary form as “two-reprise form.” We prefer the
terminology “part” rather than “reprise” since many writers employ the term “reprise” to mean
“return” or “recapitulation.”
3 Note that sectional binary is almost never found in a minor-mode piece.
4 To conserve space, Example 5.4 employs repeat signs and indicates the changes in mm. 102–103
with annotations above the violin and viola lines.
5 Another defensible interpretation of the first ten measures of B would be 12 1234 1234, where
the hypermeter is reoriented around the entry of the flute melody in the third measure of B.
(This type of reading will occur at the start of A′.)
6 The motion to the ninth (D) does not particularly disrupt the B to A connection. The ninth
resolves back to C within the span of the C harmony, and this C becomes the focal tone for the
left-hand melody at the thematic return. Although the eighth notes in the first measure of the
thematic return touch on C6, the principal pitch for the upper voice in m. 45 is A5.
7 We can understand the necessity of this alteration. In mm. 5–6, the right-hand motion from F
to G is delayed, creating a subdominant chord before the Neapolitan. When this melody is placed
into the bass, extending the F would result in a second-inversion subdominant. A beautiful way
of extending the F without generating this 64 chord is the [V65] II.
8 Observe that after the sostenuto indication Brahms writes a tempo at m. 33. For many composers,
sostenuto indicates only a shading of dynamics and articulation, but for Brahms sostenuto also
includes a slowing of tempo.
9 If one wanted to give harmonic labels to these intervening chords, they are best related to the
immediately preceding chord. For example, in m. 43 the A major chord is V of the preceding
I chord and does not progress to the ii in m. 44. Similarly, in m. 45 the A minor chord is iv of
the preceding ii chord (rather than a modally altered dominant chord).
10 The key of D major has a special role throughout the entire sonata. In the first movement, it
is the exposition’s tonal goal (rather than the normative A major). The second movement
(Andante) begins in A major but ends in D major—a rare instance of directional tonality in
Brahms’s oeuvre. As in the first movement, D major is the primary tonal contrast in the finale.

6 Classical Sonata Form

1 Although we do not adopt their terminology, our presentation of the norms of sonata form is
informed by the comprehensive discussion in James Hepokoski and Warren Darcy, Elements of
Sonata Theory: Norms, Types, and Deformations in the Late-Eighteenth-Century Sonata (Oxford, 2006).
360 Notes

2 A second, but less common, possibility for major-mode expositions is for the transition to end
on V of the tonic key. Unlike most of the features of Classical sonata form, the non-modulating
transition is not generally found in sonata forms composed after 1800.
3 In a significant number of minor-mode expositions, the secondary key is the minor dominant.
This tonal alternative is especially suited to movements with a persistently tragic or foreboding
nature.
4 Hearing the F4 at m. 7 as subsidiary to A4 is facilitated by the literal presence of F4 as an
inner voice beneath the A4 of m. 9. The inner voice in mm. 8–9 emerges directly from the F4
of m. 7.
5 This underlying pattern is discussed by Maury Yeston in The Stratification of Musical Rhythm
(Yale, 1976), 103–108.
6 In this broader tonal context, the prominent D minor at the outset of the transition can be viewed
as a subsidiary 5–6 motion arising within F major harmony.
7 This linkage was pointed out by Oswald Jonas in Introduction to the Theory of Heinrich Schenker,
trans. and ed. John Rothgeb (Longman, 1982), Example 199.
8 The hypermetric organization of mm. 113–132 is more complex than the duple grouping
indicated between the staves in Example 6.1. At the macro level these twenty measures are divided
into two groups of 10, which may explain why Mozart instinctively added the two measures of
V7 at the end, namely, to create balance in addition to providing a registral connection to the
A major chord of m. 123.
9 We do not refer to these phrases as “part 1” and “part 2” of theme 1, as we did in the Mozart
sonata just studied, because they are not separated by a perfect authentic cadence.
10 It is worth noting a long-range pitch connection that goes along with this tonal motion. At the
onset of III, the start of theme 2, the dolce melody sits on B5 (m. 59), and it is from this same
pitch that the descent to closure occurs later in the exposition. In the development, the upper
line regains B5 at the V7 of iv (m. 152) and then presents A5 at the arrival of iv (m. 176) and
G5 at the retransition (m. 198). Thus, a stepwise descent through a third associates these key
moments melodically.

7 Sonata Form in the Nineteenth Century

1 See, for example, the first movements of Schumann’s Symphony No. 2 in C major, op. 61 (1846)
and Brahms’s Symphony No. 2 in D major, op. 73 (1877).
2 Our observations about this movement have been informed by Heinrich Schenker’s discussion
and analytic graphs in Tonwille 7 (1924) and Der freie Satz (1935), Figures 54/8, 114/8, and 154/4.
3 A similar situation occurs in the opening movement of Beethoven’s “Waldstein” Sonata, op. 53,
where the second half of the exposition features a modal shift (theme 2 in E major, closing section
in E minor). Keep in mind that an exposition with a modal shift in the secondary key is not a
three-key exposition (where there are three distinct local tonics). Although the exposition with
modal shift might be viewed as a precursor of the three-key exposition, it does not undermine
the polarity between two keys that defines the eighteenth-century sonata exposition. In fact, there
are numerous eighteenth-century expositions with a modal shift (e.g., the first movement of
Beethoven’s Piano Sonata in C Major, op. 2, no. 3).
4 A similar anticipation of the minor mode in the dominant extension before the second theme
occurs in the “Waldstein” Sonata (mm. 23–30). It should be noted, however, that even in
expositions without a modal shift minor-mode inflections of the dominant extension at the end
of the transition are not infrequent; this feature also occurred in the Mozart sonata and Beethoven
trio studied in the previous chapter.
361 Notes

5 One comment on our hypermetric analysis of theme 2 is necessary. In Example 7.2, we show
successive hypermetric downbeats at the start of theme 2. With the arrival of A harmony,
m. 35 provides a hypermetric downbeat, but it ultimately functions as an accompanimental pattern
preparing the melody’s hypermetric downbeat in m. 36. (Successive hypermetric downbeats
occurred for the same reason at the analogous location, mm. 59–60, in the Beethoven trio studied
in the previous chapter.)
6 There is an interesting difference between the beginnings of the parenthetical passages in the
exposition (mm. 24–34) and development (mm. 94–108). In the exposition, the tonal goal
(V7of III) coincides with the beginning of the melodic material of the parenthesis; there is thus
an elision, and it results in a hypermetric reinterpretation. In the development, the tonal
goal (V7of VI) is articulated already at m. 93, one measure before the melodic material of the
parenthesis begins, meaning there is no elision this time. (Perhaps the very loud dynamics in the
development section motivated the extra measure before the piano parenthesis begins.)
7 In Example 7.6, we show an implied C4 in m. 231 to complete this pitch motion in a single
register, but the actual pitch at the downbeat of m. 231 is C2. C2 has a stepwise connection to
the low bass note supporting the II6 harmony at m. 218 and thereby further clarifies the overall
harmonic connection between II6 and the cadential dominant of mm. 231–238.
8 As shown in Example 7.6, the arrival at m. 239 falls on a hypermetric downbeat, but a statement
of theme 2 begins in the next measure. Thus, as with previous theme 2 statements, this one is
launched with successive hypermetric downbeats.
9 The close connection between 5–6 motions and modulations by descending third will emerge
again in the following Brahms analysis.
10 Beginning a movement with a few quasi-introductory measures that turn out to be thematically
integral occurs quite often in Brahms’s music. We will encounter this again in the finale of his
Violin Sonata in D Minor, op. 108, addressed in our study of rondo form in Chapter 9.
11 Note that in Example 7.8 we have removed rhythmic displacements in order to show the voice
leading more clearly. In the bass line in mm. 21–24, for instance, Brahms actually delays the first
pitch in each measure by a beat through the use of suspensions. We will continue to remove
such displacements in subsequent reductions.
12 The augmentation is not exact. To be exact, the fourth chord would be an eighth note, not a
sixteenth. This modification preserves an essential characteristic of the original pattern.
13 Reading of the clarinet line for the remainder of the development section is made more difficult
by its notation a diminished third rather than a major second higher than it sounds.

8 Ternary Form
1 There are numerous examples of alternating movement pairs in Bach’s suites and partitas.
2 Here, in both mm. 21 and 22, the augmented sixth chord is introduced by a strongly articulated
F major chord (VI), anticipating the key of the B section.
3 It might be tempting, though incorrect, to interpret the A in m. 8 as B, that is, as a G minor
chord (vi). The continuation makes it clear that Brahms means what he writes, which indicates
an extension of dominant function.
4 It is interesting that scale degree 3 is never supported by tonic harmony in this piece. Even in
the coda it is a passing tone—indeed, an extended one—over dominant harmony.
5 We will encounter this chord later in the retransition (m. 44), where it is spelled as a D seventh
chord but functions as an augmented sixth chord leading to V7 of V.
6 See Allen Cadwallader, “Schenker’s Unpublished Graphic Analysis of Brahms’s Intermezzo
Op. 117, No. 2: Tonal Structure and Concealed Motivic Repetition,” Music Theory Spectrum 6
(1984): 1–13.
362 Notes

7 The employment of both forms of scale degree 3 here reflects their use earlier in the bass at the
beginning of the antecedent and consequent phrases, D in m. 52 and D in m. 61.

9 Rondo Form
1 For detailed discussion of mm. 1–8 refer back to Examples 1.9, 1.18, and 3.1.
2 In Example 9.1, the hypermetric reading of this expansion is one of a few possible interpretations
of this passage.
3 The emphasis on F, A, and E in this sonata is thought to have programmatic significance, a
reference to Brahms’s lifelong friend, violinist Joseph Joachim, with whom Brahms had a
temporary falling out in the mid 1880s. (Brahms had supported Joachim’s wife’s position during
the couple’s divorce negotiations.) Joachim’s personal motto from his bachelor years was “Frei
aber Einsam” (“Free but lonely”), or F–A–E, and Brahms had encoded this musical cell in a few
of his early pieces. Encoding names or words in musical pitches was a favored device in the
works of Robert Schumann, a composer who influenced the young Brahms and the young
Joachim.
4 Since this movement contains transitions with two types of thematic content, we show this
distinction by labeling each transition as either Tx or Ty in Figure 9.3.
5 Motive y begins with A–F–E, a permutation of the three pitches in the initial statement of motive
x. We do not emphasize this connection since the lower register of the A places more attention
on the F and E and their continuation rather than on the initial three-note cell.
6 Given the parallelism of antecedent and consequent phrases and the following change in thematic
material, this arrival on VI is one of the rare instances where a deceptive progression truly has a
cadential function.
7 For a treatment of Brahms’s large-scale thematic development see Ryan McClelland, “Brahms
and the Principle of Destabilised Beginnings,” Music Analysis 28, no. 1 (2009): 3–61. This article
also discusses the distinction between rondo form and sonata form in relation to the finale of the
Violin Sonata, op. 108.

Appendix: Notes on Musical Reductions

1 An analysis by David Beach of the entire prelude is contained in “The Submediant as Third
Divider: Its Representation at Different Structural Levels,” Music Theory in Concept and Practice,
ed. Baker, Beach, and Bernard (Rochester, 1997).
2 In this instance Mozart has progressed to the dominant in the original key rather than the dominant
in the new key, leaving the process of modulation to occur at the start of the second theme.
Index of Musical Works

Page numbers that appear in bold indicate the presence of an annotated score and/or reduction

Bach, Johann Sebastian Piano Sonata in E Major, op. 7, II, 40–1,

Cello Suite No. 1 in G Major, Sarabande, 43, 127–8, 312
354–5 Piano Sonata in E Major, op. 7, III,
French Suite No. 1 in D Minor, 125–6
Sarabande, 63–4 Piano Sonata in C Minor, op. 10, no. 1, I,
French Suite No. 2 in C Minor, Menuet, 233
57–60, 152–3 Piano Sonata in C Minor, op. 10, no. 1,
French Suite No. 3 in B Minor, Menuet, II, 20–2, 30, 73–4, 79
62–3 Piano Sonata in C Minor, op. 10, no. 1,
French Suite No. 5 in G Major, Gavotte, III, 114
28, 50–2, 152 Piano Sonata in D Major, op. 10, no. 3, I,
Keyboard Suite in A Minor, Giga, 27 55
“O Lamm Gottes, unschuldig” (Chorale No. Piano Sonata in D Major, op. 10, no. 3,
165), 12–13 II, 36–7, 279–89
Partita No. 1 in B major, Allemande, Piano Sonata in D Major, op. 10, no. 3,
27 III, 68–9
Partita No. 4 in D Major, Allemande, 40 Piano Sonata in C Minor, op. 13
Prelude in C Minor (WTC I), 52–3 (“Pathétique”), I, 54, 231
Prelude in D Major (WTC I), 46–7 Piano Sonata in C Minor, op. 13
Prelude in E Minor (WTC I), 56 (“Pathétique”), II, 316
Prelude in B Major (WTC I), 351–3 Piano Sonata in C Minor, op. 13
(“Pathétique”), III, 140
Beethoven, Ludwig van Piano Sonata in E Major, op. 14, no. 1, II,
Horn Sonata in F Major, op. 17, I, 130, 35–6
132 Piano Sonata in G Major, op. 14, no. 2, I,
Piano Concerto No. 3, op. 37, 231 233–4
Piano Sonata in F Minor, op. 2, no. 1, I, Piano Sonata in G Major, op. 14, no. 2,
71–2, 117 II, 61–2
Piano Sonata in C Major, op. 2, no. 3, I, Piano Sonata in B major, op. 22, III,
83, 360 67–8
Piano Sonata in C Major, op. 2, no. 3, II, Piano Sonata in A Major, op. 26, II,
84–5 176–82, 187
364 Index of Musical Works

Piano Sonata in C Minor, op. 27, no. 2, Brahms, Johannes

III, 116 Cello Sonata in E Minor, op. 38, II,
Piano Sonata in D Major, op. 28, IV, 194–5
349 Clarinet Sonata in F Minor, op. 120,
Piano Sonata in G Major, op. 31, no. 1, no. 1, I, 258–76
I, 48 Clarinet Sonata in E Major, op. 120,
Piano Sonata in G Major, op. 31, no. 1, no. 2, I, 277–8
III, 350 Intermezzo in E Major, op. 117, no. 1,
Piano Sonata in D Minor, op. 31, no. 2, 313
II, 37–8 Intermezzo in B Minor, op. 117, no. 2,
Piano Sonata in E Major, op. 31, no. 3, 300–11
II, 98 Piano Sonata in F Minor, op. 5, III,
Piano Sonata in C Major, op. 53 182–92
(“Waldstein”), I, 42–3, 359, 360 Symphony No. 2, op. 73, I, 360
Piano Sonata in F Minor, op. 57 Symphony No. 3, op. 90, I, 105, 107
(“Appassionata”), I, 139, 235–58 Violin Sonata in D Minor, op. 108, IV,
Piano Sonata in F Minor, op. 57 331–48, 350, 361
(“Appassionata”), II, 117
Piano Sonata in F Major, op. 78, II, Chopin, Frédéric
114–15 Etude in E Major, op. 10, no. 3, 313
Piano Sonata in E Major, op. 81a Mazurka in F Minor, op. 68, no. 4,
(Lebewohl), I, 118–20, 359 38–9
Piano Sonata in E Minor, op. 90, II, 144 Nocturne in C Minor, op. 27, no. 1,
Piano Sonata in E Major, op. 109, I, 15, 144–5
18 Prelude in B Minor, op. 28, no. 6, 25
Piano Sonata in A Major, op. 110, III,
130, 133 Haydn, Joseph
Piano Trio in G Major, op. 1, no. 2, II, Chorale St. Antoni, 24
74–5, 79 Piano Sonata in E Minor, Hob. XVI/34,
Piano Trio in C Minor, op. 1, no. 3, I, III, 68–9
211–31, 233–4 Piano Sonata in D Major, Hob. XVI/37,
Piano Trio in B Major, op. 97, III, III, 315–16
115 String Quartet in C Major, op. 20, no. 2,
Rondo a capriccio, op. 129, 130–1 III, 15, 17
String Quartet in F Major, op. 135, II, String Quartet in E Major, op. 33, no. 2,
80–1 II, 102–3
Symphony No. 5, op. 67, I, 72, 78, 231, String Quartet in E major, op. 64, no. 6,
236 III, 93–6, 155–6
Symphony No. 5, op. 67, II, 127, 129 String Quartet in C Major, op. 74, no. 1,
Symphony No. 9, op. 125, II, 78 III, 87–8, 193
Violin Sonata in A Major, op. 12, no. 2, String Quartet in F Major, op. 74, no. 2,
III, 349 III, 194
Violin Sonata in F Major, op. 24, I, String Quartet in B Major, op. 76, no. 4,
234 III, 193
Violin Sonata in F Major, op. 24, IV, Haydn, String Quartet in E major,
87, 89 op. 76, no. 6, II, 143
Violin Sonata in C Minor, op. 30, no. 2, String Quartet in F Major, op. 77, no. 2,
III, 194 III, 48–9
365 Index of Musical Works

Symphony No. 101, III, 156–76 Schubert, Franz

Symphony No. 103, III, 120–5 “Du bist die Ruh,” 107–9, 358
Symphony No. 104, I, 74, 76–7, 79 Impromptu in E Major, op. 90, no. 2,
28
Mozart, Wolfgang Amadeus Impromptu in G Major, op. 90, no. 3,
Clarinet Quintet, K. 581, III, 85–7 289–300
Fantasy in D Minor, K. 397, 112–13 Moment musicale in A Major, op. 94,
“Madamina” from Don Giovanni, 18–19 no. 2, 110–11, 312
Piano Quartet in G Major, K. 478, III, Moment musicale in A Major, op. 94,
137–8 no. 6, 133–4
Piano Sonata in C Major, K. 279, III, 97 “Der Müller und der Bach,” 112
Piano Sonata in F Major, K. 280, I, 44–5, “Nachtgesang,” 127, 129
353–4 Piano Quintet in A Major (“Trout”),
Piano Sonata in F Major, K. 280, III, D. 667, I, 235
10–11, 18, 20 Piano Quintet in A Major (“Trout”),
Piano Sonata in B major, K. 281, I, 70 D. 667, II, 235
Piano Sonata in G Major, K. 283, I, 98 Piano Quintet in A Major (“Trout”),
Piano Sonata in D Major, K. 284, I, 136 D. 667, V, 235
Piano Sonata in D Major, K. 284, II, Piano Sonata in A Major, D. 664, II,
30–1 91–2
Piano Sonata in C Major, K. 309, III, 85, Piano Sonata in A Minor, D. 845, I,
350 142
Piano Sonata in A Minor, K. 310, I, 52–4, Piano Sonata in B Major, D. 960, I,
80, 82, 232–3 145–7
Piano Sonata in D Major, K. 311, I, Piano Sonata in B Major, D. 960, II,
103–4 312
Piano Sonata in A Major, K. 331, I, 29 Piano Trio in B Major, D. 898, I,
Piano Sonata in F Major, K. 332, I, 277
15–16, 197–211, 226 Piano Trio in E Major, D. 929, I,
Piano Sonata in B Major, K. 333, I, 235
29–30, 33–5, 45 String Quartet in G Major, D. 887, I,
Piano Sonata in B Major, K. 333, III, 43–4, 105–6
9–10, 22, 66–7, 79, 316–31 String Quintet in C Major, D. 956, I,
Piano Sonata in C Minor, K. 457, I, 235
99–100 Symphony in B Minor (“Unfinished”),
Piano Sonata in C Major, K. 545, I, D. 759, II, 133, 135, 141
13–15, 18, 20 “Der Wegweiser,” 115
Piano Sonata in C Major, K. 545, III,
8–9 Schumann, Robert
Piano Sonata in D Major, K. 576, I, 97 Dichterliebe, no. 4, 48, 50
Piano Trio in B Major, K. 502, I, 101 Piano Trio in F Major, op. 80, II, 350
String Quartet in B Major, K. 589, II, 87, Symphony No. 2, op. 61, I, 360
90 Waltz, op. 124, no. 4, 136
Index of Names and Terms

applied chords see secondary chords chain of thirds 120, 127, 236, 255, 258
augmentation 272, 274, 345, 361 chords (in general): designation 3–6; names
augmented sixth chords 112–17, 359; French 5; passing and neighboring 18–22, 32,
113–15, 127, 327; German 113–14, 45, 54, 110, 117, 270; tonal functions 7,
116–17, 120, 125, 127, 130–5, 208, 23–5
229–30, 280, 286, 298–301, 309, 346–7, coda 196, 231, 236, 257–8, 269, 275–6, 279,
361; Italian 113–14, 125, 210 286, 288–90, 299–300, 310–11, 314–16,
avoidance of the cadence see phrase expansion 331, 348, 361
codetta 80, 96, 155, 180, 189, 209, 229,
basic idea see sentence 231
Beach, David 356, 362
binary form 50, 57, 151–92, 314; balanced Darcy, Warren 359
153; continuous 151–5; rounded 93, deceptive motion/progression 8, 25, 66, 85–7,
151–5, 196, 314–15; sectional 151–3, 359; 91, 96, 125, 135, 229–30, 255, 286,
simple 151–3, 359 329–31, 345, 352, 362
borrowed chords see modal mixture development see sonata form
diminished third chord 127
cadence 7, 10, 66 dissonant seventh: resolution 6, 52, 120, 179,
cadence types: deceptive 66, 85, 362; half 210, 254, 257–8, 274, 301
(HC) 7–10, 22, 25, 37–8, 50, 66–7, 74,
80, 84, 93, 110, 125, 154–5, 157, 211, episode see rondo form
327, 344; imperfect authentic (IAC) 7, evasion of the cadence see phrase expansion
66–7, 70, 209, 300, 309, 327, 329, 344; exposition see sonata form
perfect authentic (PAC) 7–8, 13, 18, 20, expression 65, 74, 189, 196, 211, 229, 231,
33, 36, 48, 50, 57, 66–8, 74, 80, 83–5, 87, 279, 286, 289–90, 300, 331, 359; see also
93, 96, 117, 127, 151–5, 169, 180, 187, rhetoric
206–11, 227, 230, 236, 280, 289, 298, extended upbeat see phrase expansion
300, 314–16, 329–31, 345–6, 348, 360
cadential avoidance/evasion see phrase figured bass 3, 13, 356
expansion formal design vs. tonal structure 153–5, 196–7
cadential six-four 7–10, 13, 35, 38, 48, French augmented sixth chord see augmented
112–13, 115, 117, 125, 130, 236, 287–8, sixth chords
327, 329–31, 347
Cadwallader, Allen 361 German augmented sixth chord see augmented
Caplin, William 357–8 sixth chords
367 Index of Names and Terms

harmonic progression: chord vs. scale-step notes of melodic/rhythmic embellishment:

(Stufe) 20–2, 45–6, 107, 227, 270; anticipation 52; incomplete neighbor
hierarchical nature/levels 10, 20–2, 40, (appoggiatura) 8–11, 20–2, 26, 35–6, 308,
43–4, 72, 169, 230 344, 356; neighbor note 9–11, 26, 54,
hemiola 15, 207–9, 310, 356 110; passing tone 8, 11, 13, 20, 26, 40, 48,
Hepokoski, James 359 133, 176, 189, 288, 301, 310; suspension
hypermeasure 78 11–13, 25–6, 36–40, 44, 115, 133, 192,
hypermeter 78–9; surface vs. underlying 300, 351, 361
80–2
hypermetric beat (hyperbeat) 78 parallel motion: fifths 7, 18, 113, 117; first-
hypermetric downbeat 78–9 inversion triads 120, 257; tenths 41,
hypermetric reinterpretation 79, 358; 57–60, 118–19, 157, 180, 188, 227, 301,
examples 82, 157, 168, 327, 361 310, 353
parenthetical insertion see phrase expansion
imitation 38, 80, 125, 168–9, 272, 274–6, passing/neighboring chords see chords
345 performance 22, 65, 168, 196, 328
interruption 67, 290 period 8–10, 26, 29–31, 62, 66–70, 73–4;
Italian augmented sixth chord see augmented continuous 67–9, 84; contrasting 57, 66,
sixth chords 68, 87, 153, 357; interrupted 67, 87, 290;
parallel 8–10, 20–1, 35–8, 50–4, 66–70,
Jonas, Oswald 360 84–7, 89–91, 93, 125, 133–4, 152, 155–6,
169, 206, 208, 289, 300–1, 316, 331, 345;
Koch, Heinrich Christoph 96 progressive (modulating) 52–4, 67–9, 93,
152, 155–6, 236, 253; sectional 67, 70,
linear 5–6 motion see techniques of voice 91–2
leading phrase: 10, 65; antecedent and consequent
linkage technique 93, 208, 231, 254, 345–6, 8–10, 20–1, 35–8, 50–4, 66–70, 73–4, 79,
360 84–7, 89–91, 125, 133–4, 156–7, 169,
206, 208, 236, 253, 289, 300–1, 316, 331,
McClelland, Ryan 362 345; forephrase and afterphrase (Vorsatz
minuet 67–8, 85–7, 93–6, 120–5, 151, and Nachsatz) 57, 117
155–76, 279 phrase elision/overlap 74–7, 79, 83–4, 87, 157,
modal mixture 105–11; examples 118–20, 125, 208, 210, 272, 290, 301, 327, 361
226, 329; simple vs. compound 107–9 presentation see sentence
modal shift 125, 130, 208, 360 phrase contraction 79, 91–2, 168
modulation: chromatic 110, 120–9; common- phrase expansion: external by cadential
tone 127; diatonic 46–57, 68–9, 93–6, extension 80–2, 87, 125, 157, 189, 226,
110, 127; enharmonic 130–5, 231, 299 230, 289, 300–1, 328, 344; external by
motivic enlargement 20, 153, 210, 253–4, extended upbeat/vamp 80–1; internal by
310–11 avoidance (evasion) of the cadence 87,
motivic parallelism 60, 153, 180, 207 89–91, 125–6, 180, 290, 327, 329, 331;
internal by parenthetical insertion 41,
Neapolitan sixth chord 112–13, 133–4, 187, 85–7, 93–6, 125–6, 192, 206, 209, 229,
226, 229–30, 236–7, 257, 271, 286, 307; 300–1, 358; internal by repetition 83–5,
minor quality 299; as pivot chord 127, 105–6, 157, 169, 187, 189, 226–7, 254,
271; root position 135, 187, 253–6 256, 272, 274, 288–90, 311–12, 346, 348;
ninth chord (dominant) 5, 32, 38, 179, 290, internal by written-out deceleration 192,
310, 359 229, 271, 307, 327, 329, 346; internal to a
368 Index of Names and Terms

consequent phrase 85–7, 89–91, 125–6, ascending fifth progression 345–6; by

157, 169, 300–1, 331, 345–6, 358; descending 5–6 motion 15, 18, 42–4, 52;
internal to a continuation 83–5, 227, 327, by descending fifth progression 13–15,
329, 345–6 38–9, 44–5, 52, 57, 80–1, 93, 117, 153,
Picardy third 276 155, 168, 206, 208, 274, 301, 328–9;
pitch: designation 3 chromaticized 38–45, 80–1, 93, 155, 175,
pivot chord 46–58, 68, 115, 125–7, 226, 229, 179, 227–8, 236, 255–6, 328–9; diatonic
271, 286, 288, 307, 357, 359 13–19, 52, 57, 117, 153, 168, 206, 208,
274, 301
reaching over see techniques of voice leading sonata form 155, 196–231, 235–76, 309, 314,
recapitulation see sonata form 328; development 196–7; 206, 209–11,
refrain see rondo form 229, 236, 255–8, 269, 273–5, 328, 346–7;
retransition 153, 182, 192, 196, 229, 256, 269, exposition 196–7, 206–9, 211, 226–9,
279, 289, 300, 309–10, 315–16, 327, 237–8, 253–5, 258, 269–72, 275, 328;
329–30, 348 recapitulation 196–7, 206, 210–11,
rhetoric 231, 279; see also expression 229–31, 235–6, 256, 258, 269, 275, 328
rhythm: change in harmonic 71–2, 80, 91, sonata-rondo 328, 362
169, 180, 208, 330, 347–8; change in Stufe see harmonic progression
surface 66, 71, 79, 93, 157, 180, 206, 209, successive hypermetric downbeats 227, 286–7,
226–7, 316, 345–7; motivic 82, 189, 192, 361
236, 253, 256, 272, 289, 327, 329
rondo form 314–48, 362; episode 314–15; techniques of voice leading: linear 5–6 motion
refrain 314 7–10, 18–19, 40–3, 46–8, 52–3, 57, 133,
Rothstein, William 357 175, 256, 258, 269–74, 308, 357, 360;
reaching over 309; voice exchange 22, 32,
Schenker, Heinrich 20–2, 290, 307, 309, 36, 54, 57, 111, 115, 125, 227, 274, 287,
351–2, 356, 360–1 301, 308–9, 311, 328, 353, 357–8
scherzo 78, 151, 176–92, 279 ternary form 153–4, 197, 235, 279–311, 314
Schoenberg, Arnold 358 three-key exposition 235, 258, 269, 360
secondary chords 32–46, 114, 358; examples tonicization 46, 151, 316
48, 52–4, 56, 114, 274, 288, 301, 308 trio section 151, 169–76, 180–2, 189–92, 279
sentence (musical) 70–5, 91, 358; basic idea
70–1; continuation 70–1; examples 80–5, vamp see phrase expansion, external
87, 89, 91–2, 156, 168, 226–7, 253, 316, voice exchange see techniques of voice leading
327, 344–6; presentation 70–1 voice-leading reductions 60, 351–5
sequence 13; by ascending 5–6 motion 18–19,
40–1, 43, 175, 179, 227–8, 256; by Yeston, Maury 360