Harmony and Climax in the Late Works of Sergei Rachmaninoff

by

Blair Allen Johnston

A dissertation submitted in partial fulfillment

of the requirements for the degree of
Doctor of Philosophy
(Music Theory)
in The University of Michigan
2009

Doctoral Committee:

Associate Professor Ramon Satyendra, Chair

Professor James M. Borders
Professor Andrew W. Mead
Professor Herbert Graves Winful
Assistant Professor Karen Jeanne Fournier
© Blair Allen Johnston
2009
 Acknowledgements

I owe an immense debt of gratitude to Ramon Satyendra and Andrew Mead,

without whose support, criticism, and encouragement this study would have been quite
impossible. The fact that we are all Rachmaninoff fans makes a nice icing on the cake.
I want to thank James Borders and Karen Fournier for advice and feedback over a
period of several years; and Herbert Winful for stepping in on short notice. I also
recognize the contributions of many other teachers and graduate student colleagues over
the years, too numerous to mention individually but by no means forgotten.
My colleagues at Indiana University have been extremely supportive during the
final stages of this project, and for that I am very grateful.
A special place in these acknowledgements is reserved for two who are no longer
among us. First, Ellwood Derr, from whom I learned more about musical structure and
style than I can say. It is my sincere hope that the dissertation would have met with his
approval. And second, Lyman Bodman, who introduced me to the music of
Rachmaninoff when I was young and who first drilled me on matters of harmony and
form. He will be missed.
Finally, my dear Kaye, without whom I would never have survived the ordeal. I
cannot express the extent of her contribution and therefore will not try to do so. I will
instead only promise a return to something approaching normal life in the coming
months.

ii
 Preface

It has not been possible to include scores of the works analyzed in the
dissertation. In many cases the analytic figures and reductions provided contain sufficient
information to make scores unnecessary, but the reader is nevertheless encouraged to
have copies at hand if possible. Works composed before Rachmaninoff left Russia in
1917 (Opp. 1–39) are in the public domain and may be found in numerous editions.
References in the dissertation to the late concert works (Opp. 40, 41, 43, 44, and 45) and
to the revised version of the Sonata No. 2, Op. 36 follow the most recent Boosey &
Hawkes editions (listed in the bibliography), with the exception of Op. 41, which has no
North American publisher and is available only in old Soviet editions.
References to specific locations in scores are made using measure numbers or
rehearsal numbers as appropriate for a given work, and occasionally using both. In the
body of the dissertation, rehearsal numbers are printed in boldface (e.g. 22). A subscript
attached to a rehearsal number indicates a specific number of measures after the rehearsal
number. Thus, 227 = seven measures after 22. In the captions of figures, rehearsal
numbers are abbreviated “r.” while measure and measures are abbreviated “m.” and
“mm.” In the body of the dissertation and in captions to figures, works identified only by
title are Rachmaninoff’s. Works not by Rachmaninoff are identified by composer, title
and, where appropriate, opus number, etc.
Russian names are given in the transliterations most familiar to a general reader.
Thus, “Rachmaninoff,” “Prokofiev,” “Scriabin,” “Tchaikovsky,” and “Rimsky-
Korsakov,” not “Rakhmaninov,” “Prokofieff,” “Skryabin,” “Chaikovsky,” and “Rimsky-
Korsakoff.” (Rachmaninoff’s preference for the “-off” spelling of his name as opposed to
the “-ov” spelling is respected.) Terms in Russian are italicized in the dissertation (e.g.
nega, peremennost). Such terms are without exception drawn from published research on
Russian music, and transliterations follow those of the sources.

iii
 Table of Contents

Acknowledgments ............................................................................................................. ii
Preface............................................................................................................................... iii
List of Figures................................................................................................................... vi
Abstract............................................................................................................................. xi

Chapter 1. Introduction ....................................................................................................1

Overview of the Repertory and Preliminary Survey of Literature ................................4
The Compound Harmonic Syntax of Rachmaninoff’s Mature Works ..........................6
Characteristics of Rachmaninoff’s Style, 1909–1940 .................................................13
Toward an Interpretive Model: Chromaticism, Climax, and Culmination..................18
Organization of the Dissertation ..................................................................................25

Chapter 2. Conceptualizing Harmonic Tension in Rachmaninoff’s Mature Style....27

Postromantic Deformation and Structural Tension......................................................29
Hyperdissonance: Definition and Initial Analytic Applications..................................40
Formalizing the Model: Tension Arcs .........................................................................44
Exaggeration of Tonal Premises ..................................................................................48
Distortion of Tonal Premises .......................................................................................62
A Parenthesis: Neutralization of Tonal Premises ........................................................71
Conclusion ...................................................................................................................72

Chapter 3. Overview of Harmonic Structures and Their Rhetorical Associations ...79

Four Short Analyses.....................................................................................................84
Conclusion ...................................................................................................................98
Summary of Chapters 2 and 3......................................................................................99

Chapter 4. “Fantastic” Chromatic Structures ............................................................100

Octatonic Structures (Interval 3/6/9 Basis)................................................................104
Rachmaninoff’s and Scriabin’s Symmetrical Structures Compared .........................125
Hexatonic Structures (Interval 4/8 Basis) ..................................................................135
Whole-Tone Structures (Interval 2 Basis) and Hybrid Structures .............................154
Conclusion .................................................................................................................159

Chapter 5. Modal Structures ........................................................................................160

The Church Modes.....................................................................................................161
Peremennost, Diatonic Oscillation, and Diatonic Stacks ..........................................165
Nega ...........................................................................................................................189

iv
 Phrygian Organization ...............................................................................................208
Conclusion .................................................................................................................232
Summary of Chapters 4 and 5....................................................................................232

Chapter 6. Climax in the Last Three Concert Works ................................................234

The Dies irae in Opp. 43, 44, and 45.........................................................................235
D♭-Major Focal Points ...............................................................................................236
Organization of the Analyses.....................................................................................238
Rhapsody on a Theme by Paganini, Op. 43 (1934) ...................................................241
Symphony No. 3, Op. 44 (1936, rev. 1938)...............................................................253
Symphonic Dances, Op. 45 (1940) ............................................................................264

Concluding Remarks .....................................................................................................276

Bibliography ...................................................................................................................282

v
 List of Figures

1.1. Selected theoretic and analytic approaches to chromatic music.................................20

1.2. Interpretive model.......................................................................................................23

2.1. Richard Strauss, Elektra, r. 177–1784 .........................................................................32

2.2. Elektra, r. 177–1772, analysis .....................................................................................33
2.3. Elektra, r. 177–1782, analysis and tension diagram....................................................34
2.4. Moiré interference pattern and analogy with Elektra passage....................................35
2.5. Richard Strauss, An Alpine Symphony, “Elegy,” r. 100, analysis...............................36
2.6. An Alpine Symphony, “Calm Before the Storm,” mm. 1-6, analysis ..........................38
2.7. An Alpine Symphony, “Calm Before the Storm,” analytic overview..........................39
2.8. Symphony No. 3, Op. 44, i, analysis of climax at the end of the development..........42
2.9. Generalized tension arc...............................................................................................45
2.10. Symphony No. 3, i, deformed tension arc at r. 22–25 ..............................................46
2.11. Candace Brower’s schema for phrase structure........................................................47
2.12. Hyperdissonant exaggeration....................................................................................49
2.13. Prokofiev, Piano Concerto No. 3, Op. 26, ii, analysis of r. 87 .................................50
2.14. Hyperdissonant exaggeration in passages from Rhapsody on a Theme by
Paganini, Op. 43 .....................................................................................................51
2.15. Rhapsody on a Theme by Paganini, mm. 41–42, 300–303, 344–347, analysis........52
2.16. Schenker’s analysis of Paganini’s theme (Free Composition, Fig. 40, 9)................53
2.17. A Quasi-Schenkerian reading of Rhapsody, Variation IX, r. 26 ..............................54
2.18. Tonal functions mapped onto a prototypical tension arc ..........................................56
2.19. Prelude in D major, Op. 23, No. 4, analysis of A section (mm. 1–35).....................57
2.20. Prelude in D major, analysis of climax in B section (mm. 35–53)...........................59
2.21. Richard Strauss, Salome, Op. 54, 3605 - 361 (Kaplan’s Example 1-25) ..................60
2.22. Diagram of hyperdissonant exaggeration at the Salome climax...............................61
2.23. Mahler, Symphony No. 10, i, interpretation of climax (m. 208) following
Kaplan ......................................................................................................................62
2.24. Hyperdissonant distortion .........................................................................................63
2.25. “A-u!” Op. 38, No. 6, analysis of climax (mm. 21–28)............................................65
2.26. Analysis of OCT(1,2) structure at the “A-u!” climax .................................................66
2.27. “A-u!” postlude, analysis ..........................................................................................67
2.28. Schema of “A-u!” climax..........................................................................................68
2.29. Elegy in E♭ minor, Op. 3, No. 1, analysis of climax.................................................69
2.30. Elegy, analysis of distorted cadential figure at climax .............................................70
2.31. “Daisies,” Op. 38, No. 3, analysis of mm. 1–4 .........................................................74

vi
2.32. “Daisies,” analytic overview of climax ....................................................................76
2.33. Scales and pentachords in “Daisies”.........................................................................77
2.34. Relationship between the harmonic motive and the scales in “Daisies” ..................77

3.1. Components of Rachmaninoff’s mature harmonic language......................................80

3.2. Harmonic components and their rhetorical associations ............................................82
3.3. Probable locations of special chromatic and modal structures ...................................83
3.4. “From the Gospel of St. John,” WoO, analysis ..........................................................86
3.5. “From the Gospel of St. John,” overview ...................................................................87
3.6. “From the Gospel of St. John,” motivic analysis........................................................87
3.7. The relationship between A Lydian and OCT(0,1) .......................................................87
3.8. “From the Gospel of St. John,” diagram of climax.....................................................88
3.9. The Bells, Op. 35, i, analysis of climax (mm. 106–162).............................................89
3.10. The Bells, i, analysis of mm. 32–39 ..........................................................................91
3.11. Etude-Tableaux in E♭ minor, Op. 39, No. 5, mm. 1–12, analysis ............................93
3.12. Etude-Tableaux in E♭ minor, mm. 12–22, analysis ..................................................94
3.13. Rhapsody on a Theme by Paganini, Op. 43, Introduction (mm. 1–9), analysis .......97

4.1. Octatonic scales ........................................................................................................104

4.2. Sample chord cycles and oscillations in OCT(0,1) .....................................................106
4.3. Mussorgsky, Boris Godunov, “Coronation Scene,” OCT(2,3) oscillation..................107
4.4. Isle of the Dead, Op. 29, global structure in relation to OCT(0,1) ..............................108
4.5. Isle of the Dead, octatonic Dies irae canon after r. 22 .............................................108
4.6. Isle of the Dead, climax at r. 22 ................................................................................109
4.7. Concerto No. 3, Op. 30, i, analytic overview of mm. 1–203....................................110
4.8. Concerto No. 3, i, analytic overview of mm. 203–235.............................................111
4.9. Concerto No. 3, i, m. 235 through recapitulation in cadenza ...................................112
4.10. Concerto No. 3, i, main theme, introduction of E♭ .................................................112
4.11. Concerto No. 3, i, end of cadenza through coda, analytic overview ......................113
4.12. Symphony No. 3, Op. 44, iii, r. 79–807 ..................................................................115
4.13. Symphony No. 3, iii, r. 80, octatonic chord pair and melodic cell .........................116
4.14. Symphony No. 3, iii, analytic overview of r. 80 through coda...............................117
4.15. Octatonic diminished seventh chord techniques.....................................................118
4.16. Symphonic Dances, Op. 45, iii, analysis of local climax at r. 79 ...........................119
4.17. Rimsky-Korsakov, The Golden Cockerel, Prologue, analysis of mm. 1–38 ..........121
4.18. The Golden Cockerel, Prologue, mm. 1–38, analytic overview .............................122
4.19. Comparison of Symphonic Dances theme and Golden Cockerel theme ................123
4.20. Symphonic Dances, Op. 45, i, introduction (mm. 1–16), analytic overview..........124
4.21. Scriabin, Prelude, Op. 74, No. 3, analytic overview of mm. 1–13 .........................127
4.22. Prelude, Op. 74, No. 3, “entangled” diminished seventh chords............................128
4.23. Prelude, Op. 74, No. 3, analytic overview of the entire piece ................................128
4.24. The Bells, Op. 35, ii, annotated reduction of mm. 1–14 .........................................129
4.25. The Bells, ii, overview of thematic material ...........................................................130
4.26. The Bells, ii, OCT(2,3) in mm. 1–10.........................................................................132

vii
4.27. The Bells, ii, analytic overview...............................................................................134
4.28. The Bells, ii, octatonic oscillations in measures 1, 72, and 111..............................135
4.29. Hexatonic scales......................................................................................................136
4.30. Sample hexatonic rotations and oscillations, including extended versions ............136
4.31. Prokofiev, Romeo and Juliet, “The Girl Juliet,” analysis of mm. 1–2 ...................138
4.32. Rhapsody on a Theme by Paganini, Op. 43, Variation XIII, analysis....................140
4.33. Rhapsody on a Theme by Paganini, Variation XIII into Variation XIV ................141
4.34. Rhapsody on a Theme by Paganini, Variation XIV, HEX(0,1) analysis ..................142
4.35. Waltz in A major, Op. 10, No. 2, analytic reduction ..............................................144
4.36. Waltz, Op. 10, No. 2, HEX(0,1) cycle and auxiliary seventh chords........................145
4.37. Waltz, Op. 10, No. 2, auxiliary seventh chords and melodic details ......................145
4.38. Sonata No. 2 in B♭ minor, Op. 36, i, climax at recapitulation ................................147
4.39. Sonata No. 2, i, diatonic and hexatonic third relations in the exposition ...............148
4.40. Sonata No. 2, i, comparison of exposition and recapitulation ................................150
4.41. Sonata No. 2, i, schema of recapitulation climax and solution...............................151
4.42. Sonata No. 2, i, hexatonic and diatonic/modal third relations in coda ...................153
4.43. Etude-Tableaux in E♭ minor, Op. 39, No. 5, climax in middle section ..................155
4.44. Symphonic Dances, Op. 45, i, mm. 16–40, analytic reduction...............................157
4.45. Symphonic Dances, i, mm. 42–78, analytic reduction ...........................................158

5.1. Rhapsody on a Theme by Paganini, Op. 43, analysis of Variation VII....................162

5.2. Rhapsody on a Theme by Paganini, end of Variation VII into Variation VIII.........163
5.3. Symphony No. 3, Op. 44, i, analytic reduction of mm. 1–10 (introduction)............164
5.4. The Bells, Op. 35, iv, analytic overview of mm. 1–19 .............................................171
5.5. The Bells, iv, analytic overview of mm. 20–138 ......................................................172
5.6. The Bells, iv, analytic reduction of mm. 138–end. ...................................................174
5.7. Three Russian Songs, Op. 41, i, melodic highpoints in the five chorus phrases ......176
5.8. Three Russian Songs, i, analytic overview through r. 7 ...........................................177
5.9. Three Russian Songs, i, analytic overview of climax through end...........................178
5.10. All-Night Vigil, Op. 37, v , analytic reduction of phrases 1 through 4 ...................179
5.11. All-Night Vigil, v, analysis of phrases 5 and 6........................................................180
5.12. All-Night Vigil, v, analytic overview ......................................................................181
5.13. Sonata No. 2 in B♭ minor, Op. 36, ii, analysis of mm. 1–23 ..................................182
5.14. Sonata No. 2, ii, analysis of climax in mm. 53–62 .................................................183
5.15. Symphonic Dances, Op. 45, i, analysis of transition, mm. 91–98 ..........................184
5.16. “Vocalise,” Op. 34, No. 14, peremennost resolution in mm. 31–36 ......................185
5.17. Symphony No. 3, Op. 44, iii, analysis of peremennost in the second theme
(exposition) ............................................................................................................187
5.18. Symphony No. 3, iii, overview of exposition .........................................................188
5.19. Basic nega idiom in D♭ major.................................................................................189
5.20. Basic nega idiom in B♭ minor.................................................................................190
5.21. Rimsky-Korsakov, Scheherazade, iii, analysis of mm. 1–8 ...................................190
5.22. Elegy in E♭ minor, Op. 3, No. 1, analysis of middle section ..................................191
5.23. Musical Moment in B♭ minor, Op. 16, No. 1, analysis of mm. 38–41 ...................192
5.24. Piano Concerto No. 2, Op. 18, ii, analysis..............................................................194

viii
5.25. Piano Concerto No. 2, ii, connection between introduction and climax.................195
5.26. Piano Concerto No. 2, ii, analysis of climax...........................................................196
5.27. Piano Concerto No. 2, ii, analytic overview ...........................................................197
5.28. Prelude in E major, Op. 32, No. 3, analysis of mm. 1–40 ......................................198
5.29. Prelude in E major, analytic overview....................................................................199
5.30. Prelude in E major, mm. 50–55 ..............................................................................200
5.31. Prelude in E major, motive x in the coda (m. 55–57) .............................................200
5.32. Chromatic harmonization of the nega tone.............................................................201
5.33. “A-u!” Op. 38, No. 6, analysis of mm. 1–13 ..........................................................202
5.34. “In the Soul of Each of Us,” Op. 34, No. 2, analysis of mm. 1–14 ........................203
5.35. Complex resolution of a chromaticized nega chord ...............................................204
5.36. Gustav Holst, The Planets, vii (“Neptune”), m. 101 ..............................................205
5.37. Rhapsody on a Theme by Paganini, Op. 43, Variation XVI, mm. 575–577 ..........206
5.38. Rhapsody on a Theme by Paganini, Variation XVII, mm. 613–622......................207
5.39. Symphony No. 3, Phrygian motto theme in mm. 1–5 ............................................208
5.40. Aleko, Phrygian organization in the introduction ...................................................209
5.41. Symphony No. 1, Op. 13, i, Phrygian opening gesture ..........................................209
5.42. Borodin, Symphony No. 2, i, Phrygian organization in mm. 1–3 ..........................209
5.43. “To Her,” Op. 38, No. 2, Phrygian structure in mm. 1–4 .......................................211
5.44. “To Her,” Phrygian organization ............................................................................211
5.45. “To Her,” analytic overview ...................................................................................212
5.46. Rimsky-Korsakov, Capriccio Espagnol, iv, excerpts.............................................213
5.47. Capriccio Espagnol, iv, Phrygian organization......................................................214
5.48. Capriccio Espagnol, overview................................................................................214
5.49. Mussorgksy, Pictures at an Exhibition, No. 6, overview .......................................215
5.50. Mussorgksy, Pictures at an Exhibition, No. 6, interaction of scales ......................216
5.51. “Polichinelle,” Op. 3, No. 4, analytic reduction of mm. 1–56................................218
5.52. “Polichinelle,” analysis of climax, mm. 83–98.......................................................220
5.53. “Polichinelle,” mm. 126–130..................................................................................221
5.54. Etude-Tableaux in D major, Op. 39, No. 9, analytic reduction of mm. 1–8...........222
5.55. Etude-Tableaux in D major, Phrygian events in the main theme, mm. 6–14 .........223
5.56. Etude-Tableaux in D major, overview....................................................................224
5.57. Etude-Tableaux in D major, analysis of climax in section A2, mm. 78–83 ...........225
5.58. Etude-Tableaux in D major, mm. 89–92 ................................................................226
5.59. Symphonic Dances, Op. 45, iii, analytic reduction of mm. 1–30 ...........................228
5.60. Symphonic Dances, iii, analytic reduction of main theme at r. 58 .........................229
5.61. Symphony No. 3, Op. 44, ii, analysis of mm. 1–14................................................230
5.62. Symphony No. 3, ii, last three measures.................................................................231

6.1. Marked D♭ events in well-known Rachmaninoff works from all periods ................237
6.2. Synthesis of thematic material, form, and hyperdissonant climax events ................239
6.3. Rhapsody on a Theme by Paganini as a four-movement structure...........................242
6.4. Overview of main climax events in Rhapsody on a Theme by Paganini .................243
6.5. Comparison of (A) Paganini’s theme and (B) Rhapsody on a Theme by
Paganini...................................................................................................................244

ix
6.6. Rhapsody on a Theme by Paganini, Variation XI, nega figure at r. 31....................246
6.7. Derivation of nega figure from Paganini’s theme ....................................................247
6.8. Rhapsody on a Theme by Paganini, Variation XVIII, analytic overview
of climax #1 .............................................................................................................249
6.9. Climax in Variation XVIII, analytic details..............................................................250
6.10. Melodic and tonal “inversion” in Variations XVI – XVIII ....................................250
6.11. Rhapsody on a Theme by Paganini, Variation XXII, analytic overview
of climax #2 ...........................................................................................................252
6.12. Rhapsody on a Theme by Paganini, Variation XXIV, final statement of the
Dies irae.................................................................................................................253
6.13. Symphony No. 3, Op. 44, i, hyperdissonant climax at the end of the
development...........................................................................................................254
6.14. Symphony No. 3, i, form ........................................................................................256
6.15. Symphony No. 3, i, harmonic content of the first movement climax chord...........257
6.16. Piano Concerto No. 4, Op. 40, i and iii, climax events...........................................258
6.17. Symphony No. 3, i, analytic overview....................................................................260
6.18. Symphony No. 3, ii, analytic overview...................................................................262
6.19. Motivic chord group in Symphonic Dances, Op. 45 ..............................................265
6.20. Symphonic Dances, i, analytic overview................................................................267
6.21. Symphonic Dances, ii, analysis of the introduction................................................268
6.22. Symphonic Dances, ii, overview of section A1......................................................269
6.23. Symphonic Dances, iii, overview of section A1.....................................................270
6.24. Symphonic Dances, iii, theme I: Dies irae in short rhythmic values .....................270
6.25. Symphonic Dances, iii, theme II, beginning...........................................................271
6.26. Symphonic Dances, iii, octatonic–Phrygian hybrid at section A1 climax..............271
6.27. Symphonic Dances, iii, overview of section B .......................................................272
6.28. Symphonic Dances, iii, Dies irae in long rhythmic values.....................................272
6.29. Symphonic Dances, iii, climax in section B ...........................................................273
6.30. Symphonic Dances, iii, overview of section A2.....................................................274
6.31. Symphonic Dances, iii, octatonic–Phrygian hybrid at section A2 climax..............274
6.32. Symphonic Dances, iii, analytic overview..............................................................275

C.1. Arnold Schoenberg, Violin Concerto, Op. 36, i, reduction of mm. 1–4 ..................279

x
 Abstract

This dissertation develops a framework for interpreting the interaction of

functional tonal structures, equal-interval chromatic structures, and modal structures in
Sergei Rachmaninoff’s mature compositions (1909–1940). Three areas of research are
involved: 1) harmonic materials, compositional techniques, and expressive characteristics
in Rachmaninoff’s style; 2) chromatic and modal theories in general but with special
emphasis on Russian repertories; and 3) theories of tonal tension, expressive shape, and
climax as applied to Postromantic music.
The harmonic language of Rachmaninoff’s mature works may be understood as
an amalgam of well formed, differentiated components drawn from the Western common
practice and Russian musical traditions. I show that different harmonic components have
different rhetorical associations in the works studied; that different components are
generally associated with different locations in form; and that acknowledging the
interaction of different kinds of harmonic structures in a work contributes significantly to
an understanding of expressive trajectory and large-scale organization, and—especially—
to exegesis of climax events.
Previous studies of Rachmaninoff’s works have with rare exceptions downplayed
the significance of both Russian idioms and climax in his works. I argue that reevaluation
is warranted on both counts. Although scholars have generally treated climax events as
problems to be contained in tonal analysis, I treat them as core events around which to
organize an analysis using a strongly tension-oriented approach.
Chapters 1 and 2 address issues of form and harmony in Postromantic works in
general and Rachmaninoff’s works in particular. I develop a theory of hyperdissonance to
aid interpretation of extraordinary harmonic tensions and formal problems that resist
explanation in conventional tonal and Formenlehre terms. Chapter 3 outlines the
rhetorical associations that the variegated components of Rachmaninoff’s harmonic

xi
language have. Chapters 4 and 5 address equal-interval chromatic structures (octatonic,
hexatonic, whole-tone) and modal structures (church modes, peremennost, nega) in
Rachmaninoff’s mature works. In Chapter 6, the interpretive apparatus of Chapters 1
through 3 and the technical apparatus of Chapters 4 and 5 are applied to Rachmaninoff’s
last three compositions: Rhapsody on a Theme by Paganini, Op. 43 (1934), Symphony
No. 3, Op. 44 (1936), and Symphonic Dances, Op. 45 (1940).

xii
 Chapter 1

Introduction

Rachmaninoff does not retreat before extremely complex problems, having worked out already
his technical methods and the characteristics of his style. Of these, the greatest power in his
creative hands and his favorite is harmony—which is full of colour, lush, often bold and
sometimes even rather tough.1

Rejecting a cherished modus operandi of Rachmaninoff scholars, I will not begin

this study with a defense of the composer. If recent trends can be trusted, the scholarly
tide has turned and a fuller reckoning of his achievements may be forthcoming. The
brutal dismissal of the composer in the 5th edition of Grove’s Dictionary of Music and
Musicians has been supplanted by Geoffrey Norris’s far more sympathetic account in
recent print and online editions; and, as the preliminary review of literature in this chapter
testifies, several high-grade dissertations and books have appeared during the last quarter-
century.2 These advances provide traction for new research on Rachmaninoff’s works.
The topic of this dissertation developed from a desire to better understand certain
complex chromatic and modal structures in Rachmaninoff’s mature works. Three main
questions are addressed:

1. How do the special chromatic and modal structures identified in the

dissertation interact with the strong functional tonal bases of
Rachmaninoff’s works?

1
A.V. Ossovsky, writing in 1904; quoted in Stuart Campbell, Russians on Russian Music, 1880–1917
(Cambridge: Cambridge University Press, 2003), 176.
2
Rosa Newmarch, “Rakhmaninov, Sergey Vassilievich,” in Grove's Dictionary of Music and Musicians,
ed. Eric Blom (New York: St. Martin's Press, 1954); and Geoffrey Norris, “Rachmaninoff, Serge,” in
Grove Music Online, ed. Laura Macy, http://www.grovemusic.com (accessed March 16, 2009).

1
 2. Can an interpretive framework be developed that allows the analyst to
gather meaning from the identification of special chromatic and modal
structures, without denying the desire to hear Rachmaninoff’s works as
powerfully integrated and tonally unified?

3. What implications does awareness of special chromatic and modal

structures have for the interpretation of form, expressive/rhetorical
trajectories, and climax events?

The dissertation incorporates three distinct areas of research: 1) harmonic

materials, compositional techniques, and expressive characteristics in Rachmaninoff’s
mature style; 2) chromatic and modal theories in general but with special emphasis on
Russian repertories; and 3) theories of tonal tension, expressive shape, and climax in late
Romantic and Postromantic music. I have therefore distributed the requisite survey of
literature over multiple sections in the dissertation. Incorporated into this introductory
chapter is a survey of pertinent sources concerned primarily or wholly with
Rachmaninoff’s compositions and a preliminary review of chromatic theories. The topics
in Chapters 2 and 3 of the dissertation emerge from a more extensive review of literature
on chromatic theory, the analysis of Postromantic works, and tonal tension and climax.
Chapters 4 and 5 incorporate a survey of literature on chromatic and modal structures in
Russian music.
The selection of harmony as a main focus for study is not arbitrary. As
Ossovsky’s comments (quoted at the head of the chapter) indicate, Rachmaninoff’s
highly individual pitch language was noticed quite early in his career. Even in his early
compositions there is a “boldness”—a “toughness,” even—of harmony, which got
amplified in the jagged, intensely chromatic works of the composer’s later years. I
suggest that this boldness remains to be understood, and that it involves not just a high
level of local dissonance resulting from complex linear elaboration of functional tonal
syntax, but also, crucially, special chromatic and modal structures that have as yet been
largely unexamined in his music.
The boldness was recognized and valued by Rachmaninoff’s Russian
contemporaries and by later Soviet musicologists. K.A. Kuznetsov observed in 1945 that
Rachmaninoff’s “musical language is invariably progressive even if permanently
connected with sane foundations of Russian and world classicism,” and that the

2
“harmonic boldness” (Kuznetsov uses the same term as Ossovsky) of works composed in
the 1910s “exemplifies a certain tribute paid by him…to modernism.”3 Such remarks
seem mildly astonishing to us in an academic culture that has conventionally considered
Rachmaninoff to be, as Gerald Abraham put it, a “pale shadow” of his Russian musical
forebears.4 When, writing at various points in the 1940s, Kuznetsov, Soviet musicologist
Daniel Zhitomirski, and Russian-born writer Nicholas Slonimsky all point out, as
Geoffrey Norris has done more recently, similarities between Rachmaninoff’s and
Prokofiev’s works, conventional musicological wisdom about Rachmaninoff is turned on
its head.5 Norris is explicit about the new features of works composed by Rachmaninoff
after leaving Russia in 1917, noting their “biting chromaticism,” “curious, shifting
harmonies,” “rhythmic incisiveness,” and “almost Prokofiev-like grotesquery,” though he
provides no analytic support for these points.6
I will go somewhat further than Norris in the initial chapters of this study,
suggesting and demonstrating through analysis certain similarities between Rachmaninoff
and not just Prokofiev but a number of other composers with whom Rachmaninoff has
traditionally been contrasted, not compared—Rimsky-Korsakov, Scriabin, Mahler,
Richard Strauss, and even Shostakovich—before making a strategic retreat to ensure that
Rachmaninoff’s own style does not disappear under the weight of too many comparisons.
As Zhitomirski put it, Rachmaninoff “overcame the stylistic inertia of modernism by
selecting and transfiguring in his own way the most fresh and vital musical
agglomerations of the first quarter of the century.”7 Barrie Martyn has echoed this
observation more recently, noting that “the fundamental fact of Rachmaninoff’s place in
Russian musical history is that he stands Janus-like between the old Russia and the new,
looking back to the flowering of Russian nineteenth-century ‘classical’ music as also
ahead to the first generation of Soviet Composers.”8

3
Quoted in Joseph Yasser, “Progressive Tendencies in Rachmaninoff's Music,” Tempo 22 (1951-52): 21.
4
Gerald Abraham, A Hundred Years of Music, 4th ed. (London: Duckworth, 1974), 248.
5
Zhitomirski and Slonimsky are quoted in Yasser, “Progressive Tendencies,” 23.
6
Norris, “Rachmaninoff, Serge.” His full-length biography is not much more detailed. See Geoffrey Norris,
Rachmaninoff (New York: Schirmer Books, 1994).
7
Quoted in Yasser, “Progressive Tendencies,” 22.
8
Barrie Martyn, Rachmaninoff: Composer, Pianist, Conductor (Aldershot, England: Scolar Press, 1990), 3.

3
 Overview of the Repertory and Preliminary Survey of Literature

Detailed analyses of Rachmaninoff’s compositions are rare in the scholarly

literature. The well-known biographies of the composer written by Sergei Bertensson and
Jay Leyda, John Culshaw, Geoffrey Norris, and Barrie Martyn contain only superficial
analytic notes, though a wealth of background and contextual information.9 Most
dissertations on Rachmaninoff’s works have been addressed to the performer or general
reader.10
Among the relatively few serious studies of Rachmaninoff’s music that have
appeared, the late works have been generally neglected, despite a general agreement that
the mature style, represented by works composed from the 1910s through the 1930s, has
unique characteristics in relation to the repertory in general and Rachmaninoff’s earlier
works in particular. The most thorough study of Rachmaninoff’s harmonic language yet
produced, Robert Cunningham’s impressive dissertation of 1999, includes in-depth
analyses of several of the Op. 33 and Op. 39 Etudes-Tableaux (1911, 1917), but does not
include any music composed after 1931 and considers only one work composed after
1917, the “Corelli” Variations, Op. 42 (1931), which cannot be called Rachmaninoff’s
most analytically challenging late composition.11 Charles J. Smith’s short, unpublished
study of Rachmaninoff’s chromatic techniques deals with works Rachmaninoff
composed before leaving Russia.12 David Cannata analyzes the tonal design and form of
the Symphony No. 3, Op. 44 (1936) in his important and scholarly 1999 book (developed
from a dissertation of 1992), but he does this as much by examining manuscripts and

9
Sergei Bertensson and Jay Leyda, Sergei Rachmaninoff: A Lifetime in Music (New York:
New York University Press, 1956); John Culshaw, Rachmaninov (New York: Oxford University Press,
1950).
10
See for example Heejung Kang, “Rachmaninoff’s Rhapsody on a Theme by Paganini, Op.43: Analysis
and Discourse” (D.M.A. diss., University of North Texas, 2004); Pamela Wilder, “Sergei Rachmaninoff:
Understanding the Composer through the Etudes-Tableaux, Op. 33” (D.M.A. diss., University of Alabama,
1988); and Glenn Winters, “An Analysis of Sergei Rachmaninoff's Preludes, Opus 23 and Opus 32, and
Etudes-Tableaux, Opus 33 and Opus 39” (D.M. diss., Northwestern University, 1986).
11
Robert. E. Cunningham, Jr. “Harmonic Prolongation in Selected Works of Rachmaninoff, 1910-1931”
(Ph.D. diss., The Florida State University, 1999).
12
Charles J. Smith, “Is It Original, Or Is It Good? The Paradox of Rachmaninoff's Intra-
Tonal Chromatic Harmony” (Paper delivered at the Rhodes International Rachmaninoff Conference
October 23, 2005).

4
detailing compositional process as by analyzing the music itself.13 The Symphonic
Dances, Op. 45 (1940) have yet to be seriously examined, as do the Piano Concerto No.
4, Op. 40 (1926; revised several times) and the Three Russian Songs, Op. 41 (1926).
Scholarly treatment of works composed before 1917 is more erratic. The many
important vocal works composed by Rachmaninoff during the 1910s have been
neglected. The Op. 34 and Op. 38 songs (published in 1912 and 1916, respectively) and
the choral symphony The Bells, Op. 35 (1913) have yet to enter into the scholarly
discourse. Stephen H. Prussing’s dissertation on the choral Vespers, Op. 37 (1915) is the
only substantial, analysis-oriented document about the work.14 Not surprisingly, piano
works from all periods have received more attention; but here the tendency toward
performer-oriented documents is especially noted. Heejung Kang’s recent dissertation on
the Rhapsody on a Theme by Paganini, Op. 43 (1934) is representative of such
documents.15 This document is perhaps more analytic in orientation than most, but aims
to demonstrate that conventional tonal and stylistic principles apply relatively unchanged;
the work is treated, albeit lovingly, as outdated. A similar perspective is common even
among serious analysts of Rachmaninoff’s works. For example, in his dissertation on
Rachmaninoff’s symphonies, Dana Collins states his central point as follows: “the
harmonic analysis traces and helps evaluate [Rachmaninoff’s] progression from a daring
to an anachronistic composer.”16 I reject this view as too limited, and believe that
rigorous study of Rachmaninoff’s later works must be undertaken to overturn such
blanket evaluations of his late style.
In part to fill this vacuum, and in part because I believe Rachmaninoff’s later
works are richer and more complex than his earlier ones, I focus in the dissertation on
works composed after 1909, and especially on the relatively few works composed after
1926.17 With the exception of a few shorter passages meant to demonstrate core

13
David Butler Cannata, Rachmaninoff and the Symphony (Innsbruck: Studien-Verlag, 1999).
14
Stephen H. Prussing, “Compositional Techniques in Rachmaninoff’s ‘Vespers, Opus 37’” (Ph.D. diss.,
The Catholic University of America, 1980).
15
Heejung Kang, “Rachmaninoff’s Rhapsody on a Theme by Paganini, Op.43: Analysis and Discourse”
(D.M.A. diss., University of North Texas, 2004).
16
Dana Livingston Collins, “Form, Harmony, and Tonality in S. Rakhmaninov’s Three Symphonies”
(Ph.D. diss., The University of Arizona, 1988), abstract.
17
Rachmaninoff composed nothing of substance between 1917 and 1926.

5
techniques or to show precedents for complex structures in later works, all of the analyses
in this study are of works from 1909 or later.
Cannata has divided the composer’s output into four periods: 1890–1896 (Opp. 1–
16), 1900–1908 (Opp. 17–28), 1909–1917 (Opp. 30–39), and 1926–1940 (Opp. 40–45),
which for the sake of clarity I refer to as the “early Russian,” “middle Russian,” “late
Russian,” and “exile” periods, respectively.18 For Cannata, Isle of the Dead, Op. 29
(1909) sits between periods, and represents a landmark in Rachmaninoff’s development.
The seventeen opuses written between 1909 and 1940 (Isle of the Dead through the
Symphonic Dances, Op. 45, the last composition) show Rachmaninoff’s style in varying
degrees of its full maturity, and, while a number of earlier works are as or more famous
(for example the Concerto No. 2, Op. 18, the Symphony No. 2, Op. 27, several of the Op.
23 preludes, and of course the C♯ minor prelude, Op. 3, No. 2), it is to the later works
that one must turn if the composer’s development is to be charted. In this study, particular
attention will be paid Rachmaninoff’s last three compositions, all of which are large
concert works: the Rhapsody on a Theme by Paganini, the Symphony No. 3, and the
Symphonic Dances. This choice has been made partly because the characteristics and
techniques described in this study are particularly evident in Opp. 43–45, and partly
because these three works have received scant treatment in the scholarly literature.

The Compound Harmonic Syntax of Rachmaninoff’s Mature Works

Analysts of Rachmaninoff’s works have generally concentrated on demonstrating

the music’s tonal and/or motivic coherence. Comparatively little attention has been paid
to ways that sharp contrasts—harmonic, rhetorical, motivic/thematic, etc.—are set up and
exploited.19 Rachmaninoff had something of a Dionysian side as a composer, which
comes through especially clearly in works from the late Russian and exile periods.
Apollonian analytic approaches disguise the extent to which he was, like many Russian
composers before and after him, an eclectic composer in whose music the fusion of
different melodic-harmonic idioms sometimes seems as much a mad improvisation as a

18
Cannata, Rachmaninoff and the Symphony, 65.
19
As discussed below and in Chapter 6, Cannata’s analyses of Rachmaninoff’s large concert works are
exceptions; they are very “problem-oriented.”

6
conventionally-formed musical argument. This is not to say that a convincing argument is
not to be found, only that, as Gerald Abraham put it, “compositional superabundance”—
textural, melodic, and especially harmonic—is a recognizable characteristic of
Rachmaninoff’s style.20 Abraham means the term disparagingly, but it need not be taken
so. In my view, the superabundance is not just on the musical surface (“a lot of notes”),
but something deeper that emerges from Rachmaninoff's unusual position as an inheritor
of two traditions: the conventional European tonal practice and the less conventional
nineteenth-century Russian practice. No other composer absorbed both as fully as
Rachmaninoff did. In the works studied, I recognize a confluence of three musical
streams. One is generic; two have distinctly “Russian” overtones. These constitute three
components of a rich, compound harmonic syntax:

1. Functional tonal organization.

2. “Fantastic” equal-interval chromatic structures.
3. Special modal structures.

Each category is outlined in brief below, preliminary to more detailed discussion in later
chapters.

Functional Tonal Organization

The first and, in terms of structure if not necessarily expressive content, most
important component is diatonic-functional tonal organization, which forms the basis of
Rachmaninoff’s works even at their most tonally advanced. By functional, I mean goal-
oriented tonal organization in which root-relations by perfect fifth and tendency tone
resolutions are critical. In the analytic diagrams in the dissertation, functional tonal
patterns are often represented using an adaptation of the formula outlined by Marion
Guck and also used by Steven Laitz as the basis of his “phrase model.”21 Authentic tonal
progressions are represented by the formula T1 – (x) – PD – D – T2, where T1 an T2

20
Gerald Abraham, “Rachmaninow,’ in Friedrich Blume, ed. Die Music in Geshcichte und Gegenwart
(Kassel: Barenreiter, 1949), 10:1843.
21
Marion Guck, "The Functional Relations of Chords: A Theory of Musical Intuitions," In Theory Only 4
(1978): 29-41; and Steven G. Laitz, The Complete Musician, 2nd edition (Oxford: Oxford University Press,
2008), 291-96.

7
indicate initial and goal tonics, D indicates dominant function, PD indicates predominant
function and (x) indicates any number of potential contrapuntal and harmonic expansions
of the initial tonic.22 Recognizing that, as Anatole Leikin has put it, “in Rachmaninov,
plagality becomes quintessential,” subdominant-oriented tonal progressions (rarer than
authentic, but sometimes structurally significant in the works analyzed) are represented
by the formula T1 – (x) – SD – T2, where SD represents subdominant function (as distinct
from predominant function).23
As a young composer, Rachmaninoff was strongly influenced by Moscow-based
composers Tchaikovsky and Taneyev, and if, as Joseph Yasser argued a half century ago,
Rachmaninoff outgrew the Muscovite aesthetic and harmonic limitations fairly early in
his career, he nevertheless depended throughout his career on functional tonal patterns
and goal-oriented, arc-shaped phrase designs derived from common-practice models
more than St. Petersburg composers Mussorgsky or Rimsky-Korsakov did.24 Indeed, arc-
shaped melodic structures, and the clear departure-return strategies they suggest on
various scales (discussed more fully in Chapter 2), may be Rachmaninoff’s principle
inheritance from Tchaikovsky. The prominence of arc shapes in many musical
dimensions (melodic contour, harmonic organization, form) strongly differentiates
Rachmaninoff’s works from those of more progressive Russian composers working at the
same time (especially Scriabin and Stravinsky). Although Rachmaninoff’s harmonic
language can be considerably more adventurous and variegated than is generally
recognized, expressive trajectories retain a basis in nineteenth-century models.
This has encouraged many scholars to approach Rachmaninoff’s works entirely
through the door of the German common practice and its late-Romantic extensions.
Cunningham, whose detailed analyses are easily the best yet produced by a
Rachmaninoff scholar, speaks very much from within the Schenkerian tradition even as
he addresses the tonally complex compositions of the late Russian and exile periods.

22
When unambiguous, conventional Roman numerals and figures are also used.
23
Anatole Leikin, “From Paganism to Orhodoxy to Theosophy: Reflections of Other Worlds in the Piano
Music of Rachmaninov and Scriabin,” in Voicing the Ineffable: Musical Representations of Religious
Experience, ed. Siglind Bruhn (Hillsdale, NY: Pendragon Press, 2002), 37. Cunningham’s analyses also
recognize the possibility of structurally significant subdominants. On the other hand, the “plagality” of
Russian music is challenged in Marina Frolova-Walker, “On ‘Ruslan’ and Russianness,” Cambridge Opera
Journal (1997), 21-45.
24
Yasser, “Progressive Tendencies.”

8
Cunningham extends the Schenkerian perspective in moderate ways: by referring to
Daniel Harrison’s influential dualist theory, by acknowledging Deborah Stein’s
recognition that, in late tonal music, the “plagal axis” may occasionally match the tonic-
dominant polarity in structural significance, and by incorporating Howard Cinnamon’s
work (also undertaken from within the Schenkerian tradition) on equal division of the
octave in Liszt’s music.25 However, Cunningham’s stated goal is a fundamentally
conservative one: to show “the solid tonal foundations beneath Rachmaninoff’s
progressive harmonies.”26 For him, “symmetrical pitch structures such as the octatonic
collection and progressions by equal divisions of the octave are deployed in contexts
where they emphasize deeper-level harmonies and strong, tonality-affirming chordal
motion…These innovations arise within complex but generally unambiguous structures,
which enhance a listener’s grasp of the work’s tonal conception.”27 For David Cannata,
Rachmaninoff’s works represent a Russian culmination of post-Wagnerian, Post-Lisztian
syntax.28 But, as Rachmaninoff biographer Max Harrison put it, “whatever the music
looks like, it never really sounds like Wagner.”29
For all the insights Cunningham and Cannata’s studies have provided, they fail to
account for the crucial structural and expressive roles played by other kinds of melodic-
harmonic organization. The analyses in this study suggest that it may be more
interpretively useful to construe marked chromatic and modal structures in
Rachmaninoff’s works as having the potential to problematize or disrupt conventional
tonal patterns. This has significant implications for the analysis of large-scale design,
expressive shapes, and climax. The Russian accretions in Rachmaninoff’s idiom—the
second and third components of the compound syntax—are perhaps not so easily
dismissed.

25
Daniel Harrison, Harmonic Function in Chromatic Music: A Renewed Dualist Theory and an Account of
Its Precedents (Chicago: The University of Chicago Press, 1994); Deborah Stein, Hugo Wolf’s Lieder and
Extensions of Tonality, Studies in Musicology Series (Ann Arbor, MI: UMI Research Press, 1985); Howard
Cinnamon, “Third-Related Harmonies as Elements of Contrapuntal Prolongation in Some Works by Franz
Liszt,” In Theory Only 12 (1992): 1-30. See Cunningham, “Harmonic Prolongation,” 20-23 for an overview
of his incorporation of Harrison’s, Cinnamon’s and Stein’s ideas into his methodology.
26
Cunningham, “Harmonic Prolongation,” xvii (abstract).
27
Ibid., xvi (abstract).
28
Cannata, Rachmaninoff and the Symphony, 29-37.
29
Max Harrison, Rachmaninoff (London: Continuum, 2005), 351.

9
“Fantastic” Chromatic Structures

This category has gone largely unrecognized in Rachmaninoff’s works despite

being a subject of considerable interest in studies of other Russian composers’ music. It
involves Rachmaninoff’s use (and highly individual extensions) of equal-interval
chromatic structures—what Richard Taruskin has called Russian “fantastic” harmony,
common in Russian works from the last decades of the nineteenth century and the first
two of the twentieth.30 As discussed in Chapters 2 through 4, “fantastic” chromaticism
transcends raw structure and emerges as a kind of expressive topic in Rachmaninoff’s
mature works.31 Here, Rachmaninoff’s deep but little-known interest in the music of
progressive St. Petersburg composers one generation his senior (Mussorgsky and,
especially, Rimsky-Korsakov) is relevant. It may also be that Rachmaninoff’s
performances of Scriabin’s works after the latter composer’s death in 1915 helped spur
the intense chromatic developments of 1916–1917 (the six songs published as Op. 38 and
the Etudes-Tableaux, Op. 39 are among Rachmaninoff’s most harmonically complex
works), though the important concert pieces of 1913 (The Bells, Op. 35, and the Piano
Sonata No. 2, Op. 36) already show traces of the new chromatic procedures.32
Appreciation of this side of Rachmaninoff’s compositional persona has suffered
from the traditional differentiation, part fact and part musicological fiction, of the
Moscow and St. Petersburg musical traditions. However, the once pervasive idea that
Rachmaninoff was simply a conservative Muscovite—Abraham’s “pale shadow”—is
now generally rejected by scholars. Norris has suggested that Rimsky-Korsakov exerted
as powerful an influence on Rachmaninoff as Tchaikovsky; and Martyn has noted that
the influence of Rimsky-Korsakov’s chromatic experiments on Rachmaninoff’s harmonic
language actually increased after 1909—that is to say, during the late Russian and Exile

30
Richard Taruskin, Stravinsky and the Russian Traditions (Oxford: Oxford University Press, 1996); see
especially Chapter 4, “From Chernomor to Kashchey: Harmonic Sorcery” (255-306), Chapter 5, “Bells,
Bees, and Roman Candles” (307-368), and Chapter 10, “Punch into Pierrot (Petrushka)” (661-778).
Taruskin’s work is considered in more detail in Chapter 3.
31
As outlined in Chapters 2 and 3 and discussed more fully in Chapter 4 (pp.104ff.), fixed zero
labels are used for identification of equal-interval structures, e.g. OCT(0,1) for the octatonic
collection containing C and D♭, and HEX(1,2) for the hexatonic collection containing C♯ and D.
32
On Rachmaninoff’s performances of Scriabin works, see Martyn, Rachmaninoff, 261 and 435-436.

10
periods.33 Rachmaninoff himself described his indebtedness to the central figures of both
Moscow and St. Petersburg, while maintaining his individuality:
My music is the product of my temperament, and so it is Russian music; I never
consciously attempted to write Russian music, or any other kind of music. I have
been strongly influenced by Tchaikovksy and Rimsky-Korsakov; but I have
never, to the best of my knowledge, imitated anyone.34

Rachmaninoff’s words support Adolfo Salazar’s brief summary of the composer’s place
in Russian music history: “After Glazunov…it is no longer possible to differentiate
between the two schools [Moscow and St. Petersburg], which finally become firmly
united in Rachmaninoff.”35

Modal Structures

The third component of the compound harmonic syntax I describe in this study
comprises a variety of well-defined modal structures. As detailed in Chapter 5, some of
the modal structures in the works studied—for example, the basic church modes—are
familiar and need little theoretical description. However, others have origins in Russian
liturgical and folk traditions and may be unfamiliar even to readers with extensive
knowledge in music theory, and will therefore require significantly more description and
analytic demonstration in that chapter. Michel Dimitri Calvocoressi described the
importance of modal structures in Mussorgsky’s works as follows: “Exactly as
Mussorgsky’s syntax represents and adjustment between the tonal principle and the
modal (including particular treatment of modes exemplified in Russian folk-music), so do
his most interesting forms.”36 Calvocoressi’s words apply to Rachmaninoff, too, although
it must be recognized that Rachmaninoff’s works show considerably greater reliance on
conventional tonal structures than Mussorgsky’s.

33
Norris, “Rachmaninoff, Serge”; Martyn, Rachmaninoff, 31.
34
From a 1941 interview with David Ewen in The Etude, quoted in Bertensson and Leyda, Sergei
Rachmaninoff, 369.
35
Adolfo Salazar, Music in Our Time: Trends in Music since the Romantic Era, trans. Isabel Pope
(Wesport, CT: Greenwood Press, 1946), 124.
36
Michel Dimitri Calvocoressi, Mussorgsky (London: Rockliff, 1956): 290.

11
 Rachmaninoff’s “Russian-ness” has often been claimed, even if there has been no
agreement on what exactly this means or what effects it may have had on his composing.
Nikolai Medtner commented that “Rachmaninov is so profoundly Russian himself that he
has no need of folk music.”37 In the book Artists in Exile, Joseph Horowitz claims
somewhat amusingly that “amid the Russian musical floodtide sweeping the United
States in the early twentieth century … Rachmaninoff was the most complete musician—
and the most incurably Russian.”38 Alexander Goedicke recalled the diversity of
Rachmaninoff’s liturgical and folk music interests:

“[Rachmaninoff] loved church singing very much and quite often, even in winter,
would get up at seven o’clock in the morning and hail a cab in the darkness,
mostly to drive to the Taganka, to the Andronyev monastery, where he stood in
the half-darkness … listening to the austere ancient chants from the Oktoekhos,
sung by the monks in parallel fifths … It commonly happened that on the same
evening he would go to a symphony concert … and then, more often than not, go
on to have supper at the restaurant Yar or the Strelna, where he would stay late
into the night, listening with great enthusiasm to the singing of the gypsies.”39

While this would seem to indicate that close scrutiny of actual Russian liturgical
and folk music might yield significant insights into the nature of Rachmaninoff’s tonal
language, Alfred J. Swan pointed out that the “verisimilitude [of Rachmaninoff’s modal
structures] was still vastly handicapped by his own view of harmony. At best he arrived
at only a sort of semi-modal conception.”40 Swan recognized the crucial point: modal
structures in Rachmaninoff’s works are invariably combined with non-modal melodic
and harmonic structures. As a result, an attempt to understand modal structures in
Rachmaninoff’s music by rigorous comparison with actual Russian chant, actual Russian
folk music, or Russian modal theory will likely be as “handicapped” as the modal
structures themselves. To avoid compounding the handicap, the description of modal
structures in Chapter 5 is therefore limited to only four clearly defined, frequently
encountered, and rhetorically significant types.

37
Quoted in Alfred J. Swan, Russian Music and Its Sources in Chant and Folk-Song (New York: Norton,
1973), 172.
38
Joseph Horowitz, Artists in Exile (New York, Harper Collins, 2008), 199.
39
Alexander Goedicke, quoted in Martyn, Rachmaninoff, 30.
40
Swan, Russian Music, 176.

12
 One goal of this dissertation is to detail through analysis the specific special
chromatic and modal structures used in Rachmaninoff’s mature works. A difficulty
emerges, however, when moving from descriptions of melodic-harmonic components to
consideration of the compositional whole. The difficulty is not unique to Rachmaninoff’s
works, but rather affects a wide range of extended tonal works from the late nineteenth
and early twentieth centuries, in which traditional tonal methods and a variety of extreme
chromatic procedures are often applied simultaneously. Music analysts have in general
been concerned primarily with demonstrating unity in compositions, not with describing
in a convincing and meaningful way the interactions of different structural types in a
single composition. The crucial question here is how “abnormal” harmonic and melodic
structures in extended late tonal works, and particularly in Rachmaninoff’s mature
compositions, can be made essential to interpretation. This question is the central concern
of the dissertation as a whole and Chapters 2 and 3 in particular.

Characteristics of Rachmaninoff’s Style, 1909–1940

Analysis of a large number of Rachmaninoff’s works has suggested some

preliminary observations about the nature of his mature style in comparison to his earlier
works. The music composed after Isle of the Dead, Op. 29 and the Piano Concerto No. 3,
Op. 30, which, as discussed earlier, represent a landmark in the composer’s development,
especially with regard to tonal design and harmonic complexity, is characterized by the
following eight characteristics:

1. Greater concision of thematic material. Although long-spun melodies remain in

the later works (several are in fact quite famous), they are fewer than in earlier
works. Thematic material in works from the late Russian and exile periods tends
to be broken up into shorter units, and, overall, less musical time is spent in
exposition.

2. Increased transparency of texture and orchestration, as noted by Barrie Martyn.41

41
Martyn, Rachmaninoff, 31.

13
 3. Greater rhythmic and metric complexity. Changing meters are increasingly
common, as are syncopated patterns. In the case of certain figures that appear in
the Rhapsody on a Theme by Paganini, Rachmaninoff’s acknowledged interest in
jazz may have been a factor, as may his familiarity with Gershwin’s Rhapsody in
Blue in particular.42

4. Greater amounts of local dissonance corresponding to the use of more complex

vertical sonorities and more adventuresome linear treatment of the ordinary
materials of tonal syntax. (This is the central topic of Cunningham’s dissertation.)

5. Increasingly prominent use of the Dies irae chant, or its distinctive melodic
pattern, generalized. Rachmaninoff’s interest in—obsession with, perhaps—the
Dies irae is well-documented, but will probably never be adequately explained.43

6. Increased emphasis on idiosyncratic melodic and harmonic structures referable to

recognized Russian chromatic and modal idioms.

7. Emphasis on a special kind of structural melodic-harmonic tension that I call

Hyperdissonance. Marked hyperdissonance events are anomalous in
Rachmaninoff’s earlier works, but represent a regular expressive and structural
feature of the later works.

8. Increasing problematization of core formal strategies, especially the departure-

return principle and conventional tonal structures with which it is associated.

Nos. 1 through 4 on the list are not addressed with any rigor in the dissertation.
No. 5 is addressed on a case-by-case basis in the works studied. Nos. 6 through 8
constitute the core of the dissertation. These three characteristics are closely related, and
transport Rachmaninoff’s style past the nineteenth century in distinctive ways. No. 6 is
the subject of Chapters 4 and 5 of the dissertation. No. 7 is treated in Chapter 2 of the

42
Harrison, Rachmaninoff, 246.
43
But see discussion in Martyn, Rachmaninoff, 98-99. As described by Martyn, the Dies irae is used
thematically in the Symphony No. 1, Op. 13, where it is related to the biblical epigraph “Vengeance is
mine, I will repay.” Martyn states on p.99 that “after the traumatizing catastrophe of the [symphony’s]
premiere, it is hardly surprising that the symphony’s musical motto was to haunt the composer throughout
his life, appearing in his work with increasing insistence as he grew older and approached his own day of
judgment.” On the use of the Dies irae in concert music generally and in Rachmaninoff’s works
specifically, see Malcolm Boyd, “‘Dies Irae’: Some Recent Manifestations,” Music & Letters 49 (1968),
347-356; Robin Gregory, “Dies Irae,” Music & Letters 34 (1953), 133-139; and Susan Jeanne Woodard,
“The Dies Irae as Used by Sergei Rachmaninoff: Some Sources, Antecedents, and Applications” (D.M.A.
diss., The Ohio State University, 1984).

14
dissertation. No. 8 is treated in Chapters 2 and 3 of the dissertation, and forms the basis
for the longer analyses in Chapter 6.
Consideration of the style characteristics identified above reveals limitations in
existing Rachmaninoff scholarship. Cunningham’s analyses are penetrating and
technically excellent; but characteristics Nos. 5 through 7 of the list cannot be
incorporated into his approach, for the following reasons:

1. The Schenkerian perspective cannot easily account for expressive trajectories and
structural tensions that emerge from the interaction and layering of variegated
components of a compound syntax. As discussed in more detail in Chapters 2 and
3, I consider such interactions to be crucial to meaningful interpretation of the
works studied.

2. More specifically, within Cunningham’s framework, it is not possible to consider

the rhetorical associations that special chromatic and modal components may
have, or the general locations (in relation to the functional tonal basis and in
relation to form) at which they are most likely to be found. However, analysis of a
large number of works suggests that special modal idioms are generally
introductory, expository, or post-climactic, and are therefore found at the
beginnings and ends of sections, whereas idioms derived from “fantastic” equal-
interval chromaticism are generally associated with intensification,
destabilization, and climax.

I agree with Robert Hatten that, following Saussure, “musical meaning is

difference.”44 In fact, the special modal and chromatic idioms described in this
dissertation may be “topics” in the sense that Hatten and Ratner have used the term; the
idioms are certainly “marked,” and beyond question rhetorically differentiated.45 Unlike
Cunningham, I want unorthodox tonal features in Rachmaninoff’s works late works to
remain unorthodox in the interpretation, in order to capture the meaning that difference
can engender, while developing analytic contexts in which the features can be understood
as part of a coherent whole. Cunningham’s Schenkerian graphs achieve his stated goal of
providing “a frame of reference whereby the analyst, listener, or performer can
effectively grasp the structure of a work and recognize idiomatic features of the

44
Robert S. Hatten, Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation
(Bloomington: Indiana University Press, 1994), 276.
45
Ibid.; Leonard Ratner, Classic Music: Expression, Form, and Style (New York: Schirmer, 1980).

15
composer’s style.”46 However, the analytic agenda motivating the graphs—demonstration
that complex chromatic events invariably fold into and prolong basic tonal functions, and,
more controversially, perhaps, demonstration that the traditional scholarly view of
nineteenth-century chromatic expansion as a challenge to conventional tonal organization
and a seed in the destruction of that organization—prohibits incorporation of expressive
and structural features that in my view depend on differentiation.47
Cunningham renders nonessential those features of each work that arguably carry
the greatest expressive weight, and that most immediately identify the work as “a
Rachmaninoff.” In Cunningham’s approach, background structures are predetermined to
the extent that the individual elements in a Postromantic composition cannot easily be
used as a basis for interpreting the work’s unique expressive qualities. In the final
assessment, as demonstrations of analytic muscle, Cunningham’s graphs are impressive;
but they are interpretively limp, because he fails to recognize the rewards offered by a
somewhat more flexible tonal ontology. As Joseph Dubiel has put it, “the best
understanding clearly lies not in the simplest explanation of the data, but in the most
complex interpretation of them.”48
David Cannata’s interpretations, on the other hand, suppose no strict form of
predetermined background structure.49 Cannata’s analytic approach derives mainly from
Robert Bailey’s work on Wagner.50 Cannata emphasizes large-scale key relations, and
complex structures that emerge from the exploitation of Bailey’s “double-tonic
complexes.”51 As a result, he is able to propose individualized structures for each work
analyzed, several of which are very convincing. However, within the large structures

46
Cunningham, “Harmonic Prolongation,” 318.
47
See especially the explicit discussion in “Harmonic Prolongation,” 318-320. The conventional view
Cunningham rejects is perhaps stated most powerful in the writings of Ernst Kurth. See Lee A. Rothfarb,
Ernst Kurth as Theorist and Analyst (Philadephia: University of Pennsylvania Press, 1988); and Rothfarb,
Ernst Kurth: Selected Writings (Cambridge: Cambridge University Press, 1991).
48
Joseph Dubiel, “Contradictory Criteria in a Work of Brahms,” in Brahms Studies, ed. David Brodbeck,
vol. 1 (Lincoln: University of Nebraska Press, 1995), 82.
49
It should be noted that analysis is only one component of Cannata’s project; his excellent study of
sketches and compositional process is of great value.
50
Robert Bailey, “An Analytical Study of the Sketches and Drafts.” In Wagner: Prelude and
Transfiguration from “Tristan and Isolde,” ed. Robert Bailey, 113–46 (New York: Norton, 1985); Bailey,
“The Structure of the Ring and Its Evolution,” 19th Century Music 1 (1977-78), 48-61.
51
Bailey’s double-tonic complexes are developed theoretically in Patrick McCreless, Wagner’s
“Siegfried”: Its Drama, History, and Music (Ann Arbor, MI: UMI Research Press, 1982).

16
Cannata describes, details are generally lacking, and in the case of the Symphony No. 3,
his interpretation is significantly weakened by a failure to recognize important special
modal and chromatic structures. (See the analysis of the Symphony in Chapter 6 of the
dissertation.) This fault might have been remedied had Cannata studied more works by
other Russian composers while preparing the analyses. For, while Rachmaninoff was
certainly post-Wagnerian, he was also post-Mussorgskian, post-Rimsky-Korsakovian,
post-Borodinian. In fact, the only Russian composer other than Rachmaninoff treated in
Cannata’s book is Tchaikovsky, whose Fourth, Fifth, and Sixth Symphonies are
presented in brief as a prelude to analysis of Rachmaninoff’s works.52 But, as I have
already suggested, by 1909 (to say nothing of 1917, when Rachmaninoff composed his
last Russian-period works, or 1936, when the Symphony No. 3 was premiered), the
influence of Tchaikovsky on Rachmaninoff was significantly diminished.
Although her work does not approach Cunningham’s in analytic sophistication or
detail, nor Cannata’s in musicological sophistication, Patricia Brady’s dissertation on the
Etudes-Tableaux, Op. 33 and Op. 39 is a document of some substance.53 In it, she
suggests that diatonic-functional tonal methods are not unassailable pillars in
Rachmaninoff’s works, though she recognizes that functional tonal organization is
ultimately the paramount factor in each work’s structure. Proceeding from the very point
of view that Cunningham vehemently rejects, she observes that “many etudes contain
certain forces which establish or reinforce tonality and other forces which weaken it.”54
She notes more specifically that “the use of modal harmony is a characteristic feature of
Rachmaninoff’s writing. Modal structures appear in most of the seventeen etudes.”55 In
her view, chromatic and modal structures not only decorate but have an effect on the tonal
basis:

Insofar as the nature of tonality is concerned, the Etudes-Tableaux are typical

of many late nineteenth century compositions. Both chromaticism and
modality—two opposing forces which serve to weaken and obscure functional
major-minor tonality—are conspicuously present in the etudes… [The] “B”

52
Cannata, Rachmaninoff and the Symphony, 66-68.
53
Patricia Brady, “Rachmaninoff’s Etudes-Tableaux” (D.M. diss., Indiana University, 1986).
54
Ibid., 112.
55
Ibid., 124.

17
 sections of most etudes are characterized by increased chromaticism, often to
the degree that tonal center is completely obscured.”56

Brady’s remarks echo the observations made decades earlier by the Russian and Soviet
authors quoted at the beginning of the Introduction; and, although her analyses lack rigor
and therefore cannot be considered strong scholarly statements, she to some degree
anticipates some of my central points:

1) A compound melodic-harmonic syntax applies in Rachmaninoff’s mature works.

The compound syntax is based on diatonic-functional tonal structures, but
incorporates distinct chromatic and modal components that have as much
expressive and formal significance as the diatonic-functional framework, and in
some cases more.

2) Tension between components of the compound syntax—that is to say, between

special chromatic and/or modal structures and the diatonic-functional
framework—has implications for the interpretation of tonal structure, expressive
design and form.

3) Certain chromatic and modal components used with great frequency by

Rachmaninoff (especially after 1909) can be loosely associated with specific
rhetorical functions, and with particular locations in relation to musical form.
Brady notes that most “B” sections of the Etudes-tableaux are “characterized by
increased chromaticism”—this, combined with the observation that most of the
“B” sections are climactic, suggests the tantalizing possibility that clear
associations of pitch structure, location, and rhetorical function may be
developed.

Toward an Interpretive Model: Chromaticism, Climax, and Culmination

As the discussion above suggests, music analysts have differed greatly in their
approaches to the problem of interpreting challenging chromatic structures in nineteenth-
and early twentieth-century works in which functional tonal methods still apply. Figure
1.1 is a diagrammatic overview of a number of well known and influential analytic
approaches. The figure provides a context in which to outline the hermeneutic I adopt in
this dissertation.

56
Ibid., 106-7.

18
 In Figure 1.1, theories are identified by author and organized into four boxes
according to the degree to which extreme chromatic events in works that retain tonal
methods (in varying amounts) are or are not subsumed by diatonic-functional tonal
contexts. The theorists in box 1 (Howard Cinnamon, Robert Cunningham, and Heinrich
Schenker) present extreme chromatic structures as only superficially complex; chromatic
structures at the foreground and middleground levels do not disrupt but in fact support an
underlying diatonic-functional framework, and they are entirely dependent on voice
leading operations. These theorists present extreme chromatic structures as occurring in
conventional tonal methods.
The theorists in box 2 (Daniel Harrison, David Kopp, and Charles E. Smith) and
box 3 (Robert Bailey, David Cannata, Patrick McCreless, and Arnold Schoenberg) differ
greatly from one another in important ways; but they all present extreme chromatic
structures as at least partly independent of diatonic-functional ones.57 According to
theorists in box 2, in late tonal music, new, independent chromatic functions emerge.
These are legitimate on their own terms, and equal in significance to the diatonic-
functional functions, even if they ultimately derive in important ways from the diatonic-
functional ones. In box 2 theories, voice leading is only partly responsible for the
coherence of chromatic structures. According to theorists in box 3, tonality is expanded
by advanced chromatic procedures to the point that, as McCreless put it, “the background
is no longer given but chosen” and a true twelve-tone context exists.58 Theorists in boxes
2 and 3 present extreme chromatic structures as tonal methods.

57
See entries under these authors’ names in the bibliography.
58
Patrick McCreless, “An Evolutionary Perspective on Nineteenth-Century Semitonal Relations,” in The
Second Practice of Nineteenth-Century Tonality, ed. William Kinderman and Harald Krebs (Lincoln:
University of Nebraska Press, 1996), 103.

19
 Figure 1.1. Selected theoretic and analytic approaches to chromatic music

extreme chromatic structures as tonal methods • independent chromatic harmonic

functions emerge
2 • partly dependent on voice leading (tension not emphasized)

Harrison, Kopp, Smith

• tonal “background” freely chosen

in true twelve-tone chromatic context
• “expanded” tonality in dialogue with
3
conventional tonal methods

Bailey, Cannata, McCreless, Schoenberg,

20
atonality
extreme chromatic structures in tonal methods extreme chromatic structures and/against/around tonal methods

• chromatic elements are subsumed • traditional elements are subsumed and

by diatonic-functional contexts transformed by new contexts
1 • dependent on voice leading 4 • not always dependent on voice leading

Cinnamon, Cunningham, Schenker Baker, Samson, Straus, Wilson, Zimmerman

(tension between pitch structures sometimes emphasized)

harmonic boundaries in Rachmaninoff’s mature works

 Theorists in box 4 (James Baker, Jim Samson, Joseph Straus, Paul Wilson, and
Daniel Zimmerman) go a step further, developing interpretive contexts in which
chromatically-expanded tonal syntax leads to and then interacts with early atonal (or
post-tonal) structures.59 In their theories and analyses, extreme chromatic structures are
presented not in or as tonal methods, but and/against/around tonal methods, to the point
that tonal methods are themselves sometimes subsumed by the emergent contexts. Some
of the repertories treated in these theories are clearly more radical in matters of pitch
organization than Rachmaninoff’s works; but as examples of a particular way of thinking
about chromatic structures in relation to conventional tonal ones, they stand. Because
they emphasize the special tensions this sometimes creates in musical works, the findings
of Samson, Straus, and occasionally Zimmerman are consonant with those in the present
study. As Straus points out, “twentieth-century works often incorporate traditional
elements that are structurally distinct from the prevailing musical syntax...Our
understanding of them will be enriched if we can fully appreciate their clash of distinct
structures."60
Rachmaninoff’s mature works challenge in part because, as shown at the bottom
of Figure 1.1, tonal organization is variegated in such a way as to make each of the four
categories of theory potentially applicable in different ways for different works or
passages, or in different interpretive contexts. This variegation allows Cunningham’s
approach, with its rigid precompositional system, to seem plausible in some contexts,
while also encouraging something like Paul Wilson’s theory of “structural overlay” in the
music of Béla Bartók, in which different pitch structures are activated simultaneously in a
work, and precompositional systems are avoided as a matter of principle.61
In response to this challenge, Chapters 2 and 3 of the dissertation develop an
interpretive approach that synthesizes features from the different boxes in Figure 1.1.
Linear analysis techniques are used extensively, but I propose no Schenkerian Urlinien or
Ursätze. Following Bailey, Cannata and McCreless, the global structure of each work
studied is uniquely determined. The kind of compound syntaxes developed by Baker,

59
See entries under these authors’ names in the bibliography.
60
Joseph N. Straus, “The "Anxiety of Influence" in Twentieth-Century Music,” The Journal of Musicology
9 (1991), 437.
61
Paul Wilson, The Music of Béla Bartók (New Haven, CT: Yale University Press, 1992), especially 14,
51-52, 191.

21
Straus, Wilson and Zimmerman provide traction for the (more conservative) compound
syntax I find in Rachmaninoff’s mature works. Like Wilson, I develop an analytic
approach in which different structures are conceived as different layers in a compound,
stratified melodic-harmonic environment.
The interaction of different—and, as discussed in Chapters 2 and 3,
differentiated—harmonic structures in a work is a central factor in the interpretation of
the work. While Daniel Harrison’s elegant theory proposes a synthesis of function and
chromaticism, I prefer to savor the tension that can exist between functional tonal
“norms” and chromatic or modal “abnorms.”62 As explored more fully in the following
chapters, Rachmaninoff’s works are climax-centric. Form is organized around climax
events to a degree matched perhaps only in the works of Mahler; and climaxes are very
often characterized by dramatic changes in the nature of the interaction of layers in the
compound melodic-harmonic environment—changes, that is to say, in the relative status
of the norms and the abnorms. This has led me to the interpretive model outlined in
Figure 1.2.

62
The concept of “norms” and “abnorms” is developed in Joseph Dubiel, “Contradictory Criteria.”

22
 Figure 1.2. Interpretive model

interaction of components of compound syntax

(possibly hyperdissonance)

characterizes

climax & culmination

is organized
around

form

This loose hermeneutic is based in part on close study of many compositions from
the late Russian and exile periods and in part on Rachmaninoff’s own understanding of
musical form as a process that directs every musical work toward a “point” of
“culmination,” which “must seem a liberation from the last material obstacle, the last
barrier.”63 “The composition itself determines this culmination; the point may come at its
end or in the middle, it may be loud or soft...”64 I offer no strict interpretation of
Rachmaninoff’s theory. “Culmination” in Rachmaninoff’s terms resists precise
definition. Furthermore, “culmination” (as conceived by Rachmaninoff) and “climax” (as
generally understood) are not necessarily coextensive, though they are in many cases

63
From a letter to Marietta Shaginyan, quoted in Bertensson and Leyda, Sergei Rachmaninoff, 195.
64
Ibid.

23
closely associated. According to Rachmaninoff, a point of culmination may be quiet; but
a climax is virtually always a noisy affair, involving a simultaneous intensification of
many musical parameters—dynamics, melodic tessitura, textural density, rhythmic
activity, harmonic tension, chromatic activity—toward what V. Kofi Agawu calls a
“highpoint.”65 Agawu observes that “the phenomenon of climax is central to our musical
experience,” but that existing music-theoretic approaches tend to de-emphasize it in the
interest of greater “seriousness.”66 Indeed, attempts to organize analyses of
Rachmaninoff’s works around climax events are conspicuously absent from the literature.
An exception to this is Jason T. Stell’s master’s thesis, which, drawing from Agawu and
others, deals directly with climax events in three piano preludes (one each from the early
Russian, middle Russian, and late Russian periods).67 Stell’s “expressive curves” share
some features with the “tension arcs” I develop in Chapter 2 of the dissertation.68 For
Stell, a “highpoint” is a “crux” in Robert Hatten’s sense of the word—“the point of
expressive focus or greatest intensity in an entire piece.”69 This recalls Rachmainonff’s
point of culmination, which need not be a noisy affair (and is therefore unlike climax in a
narrow sense), but which need always be the point of expressive focus in a work.
This being said, in the large majority of cases climax and culmination are
coextensive—the climax is usually the “point.” The Oxford English Dictionary defines
“climax” in the proper rhetorical sense as “a figure in which a number of propositions or
ideas are set forth so as to form a series in which each rises above the preceding in force
or effectiveness of expression.”70 Many of the climax events analyzed in the dissertation
fit this definition—a series of stages of gradually increasing intensity. However, the same
dictionary defines “climax” in the general sense as “the highest point of anything reached

65
V. Kofi Agawu, “Highpoints in Schumann’s ‘Dichterliebe’,” Music Analysis 3 (1984), 159-180.
66
Ibid., 159-160. Agawu identifies Peter Bergquist’s Schenkerian analysis of the first movement of
Mahler’s Tenth Symphony, in which a pair of extraordinary climax events are reduced to “foreground
events,” as a particularly egregious case in point.
67
Jason T. Stell, “Rachmaninov’s Expressive Strategies in Selected Piano Preludes: Highpoints, Dramatic
Models, and Dynamic Curves” (M.A. thesis, The Pennsylvania State University, 1999).
68
Stell’s, Agawu’s, and my curves/arcs all share something of an origin in Leonard Ratner’s “dynamic
curves.” See Leonard Ratner, Music: The Listener’s Art (New York: McGraw-Hill, 1966), 2nd edition: 314.
69
Stell, “Rachmaninov’s Expressive Strategies,” 17. Stell magnifies Hatten’s original definition of crux—
“the point of expressive focus or greatest intensity in a phrase or gesture” (Hatten, Musical Meaning in
Beethoven, 289).
70
Oxford English Dictionary, 2nd ed. (1989), s.v. “climax, n.”

24
by gradual ascent; the culmination, height, acme, apex,” suggesting the degree to which
climax and culmination may overlap.71
Analyses in the dissertation are organized to show how climax events relate to
interactions of diatonic, chromatic, and modal structures in a work; how expressive focal
points (cruxes) emerge as different structures (with different rhetorical associations and
expressive implications) come together or culminate; and how variegated melodic and
harmonic materials may be integrated in a climax- and culmination-oriented conception
of musical form.

* * *

The dangers of over-interpretation and over-systematization threaten a project

such as this one. I have made every effort to avoid these dangers without sacrificing
music-theoretic rigor or analytic detail. A passage from an article written by Jay Reise on
the music of Scriabin is appropriate here:

When we teach the Bach chorales to beginning harmony students, we do not seek
to present an airtight system of composition but rather the elements of a style.
Similarly, I am not trying to reveal a rigid system in Skriabin’s work, but rather a
few components of a relatively flexible method of composition, which can
explain or at least describe certain characteristics of his style.72

Similarly, the goal of this dissertation is explanation (or at least description) of certain
characteristics of Rachmaninoff’s mature style, not the presentation of an “airtight
system.”

Organization of the Dissertation

The dissertation is in six chapters (including this introductory chapter), which fall
loosely into three parts. In Chapters 2 and 3, the interpretive apparatus sketched above is

71
Ibid.
72
Jay Reise, “Late Skriabin: Some Principles Behind the Style,” 19th Century Music 6 (1983), 226.

25
developed. These chapters may be of interest even to readers who are unfamiliar with
Rachmaninoff’s mature works. In Chapters 4 and 5, the special chromatic and modal
structures that appear most frequently in the works analyzed are described in detail. By
no means are Chapters 4 and 5 meant to be a comprehensive survey of Rachmaninoff’s
tonal language; additional research will surely expand the view presented here.
Throughout Chapters 1 through 5, numerous analytic vignettes are presented, showing
how theory and concepts in the chapters may be applied to the interpretation of climax
events in works from the mature period. Most of these shorter analyses are of partial
works, of single movements, or of complete short works. In Chapter 6, the full
interpretive apparatus of Chapters 2 and 3 and the full technical apparatus of Chapters 4
and 5 are applied to the three large concert works Rachmaninoff composed during his
final decade: Rhapsody on a Theme by Paganini, Op. 43, Symphony No. 3, Op. 44, and
Symphonic Dances, Op. 45.

26
 Chapter 2

Conceptualizing Harmonic Tension in Rachmaninoff’s Mature Style

The term “Postromantic” has recently fallen somewhat into disuse.1 The term was
accepted enough in the recent past to be used as a subchapter heading in the well-known
fourth edition of the ubiquitous undergraduate textbook A History of Western Music.2 But
it is entirely absent from the sixth and seventh editions of the same text, due perhaps to
increasing awareness that historical demarcations of this sort may be more fluid than
scholars sometimes imagine them to be.3 The standard music dictionaries currently have
no entries for the term. However, I believe “Postromantic” has value, and would reclaim
it for a repertory from the late nineteenth and early twentieth centuries that displays a
certain set of aesthetic and technical characteristics. A comprehensive description of
these characteristics is surely impossible, given the extraordinary diversity of tonal and
quasi-tonal music written during the period ca. 1890–1940 (which I take as reasonable
outer limits of Postromanticism, and which correspond very nearly to the boundaries of
Rachmaninoff’s working years). However, the Encyclopædia Britannica provides a
starting point:

[A] musical style typical of the last decades of the 19th century and first decades
of the 20th century and characterized by exaggeration of certain elements of the
musical Romanticism of the 19th century. Postromanticism exhibits extreme
largeness of scope and design, a mixture of various musical forms (e.g., opera
and symphony), and heightened contrapuntal complexity (i.e., a long or vast

1
An earlier draft of material from this chapter was read at the 2007 Annual Meeting of the Society for
Music Theory in Baltimore as “Maximally Rough and Loving It: Appreciating Expressive
Hyperdissonance in the Early Twentieth Century.”
2
Donald J. Grout and Claude Palisca, A History of Western Music, 4th edition (New York: Norton, 1988),
755-771. I make no distinction between various forms of the term—“Postromantic,” “postromantic,” or
“Post-Romantic.”
3
Donald J. Grout and Claude Palisca, A History of Western Music, 6th edition (New York: Norton, 2001); J.
Peter Burkholder, Donald J. Grout and Claude Palisca, A History of Western Music, 7th edition (New York:
Norton, 2006).

27
 array, or both, of simultaneous but independent musical lines or events). Often
Postromanticism also embraces vivid religious or mystical fervour, a sense of
longing, and a sense of the grim and the grotesque.4

“Exaggeration” and “vast array…of simultaneous but independent musical lines

or events” are particularly telling. From them, I offer a more specific observation: if the
Romantic is characterized by chromatic expansion and the development of striking
elaborations of linear tonal syntax, then the Postromantic is characterized by exaggeration
and ultimately fragmentation of tonal syntax, and the juxtaposition or superimposition of
conventional functional tonal structures and intense chromatic and/or modal structures
that challenge and even deform the functional tonal basis. In my view, complex
interaction of variegated melodic-harmonic components is one source of the continuing
fascination Postromantic music holds.
A basic claim in the present document is that Rachmaninoff was a Postromantic
composer, not an anachronistic Romantic composer. Echoing many of the authors quoted
in Chapter 1, who hear in Rachmaninoff’s music something progressive or at least
idiosyncratically “modern,” the analyses in this study demonstrate that Rachmaninoff
was not unaffected by musical developments in “Silver Age” Russia or in the early
twentieth century generally.5 Peter Burkholder has aptly suggested the difficulty of
categorizing composers of Rachmaninoff’s generation in the most recent revision of the
venerable history textbook cited above, asking the question, “Late Romantic or Modern?”
and then responding, “all the composers of this generation have aspects of both eras,
combining nineteenth-century elements with twentieth-century sensibilities.”6 The
characteristics of Rachmaninoff’s mature style identified in Chapter 1 of the
dissertation—especially Nos. 6–8 on the list, which involve the articulation of
conventional tonal and formal structures and unconventional structures that challenge and
disrupt them—correspond in clear ways to the above description of general Postromantic
characteristics.

4
Encyclopædia Britannica Online, s.v. “Postromantic music,”
http://www.britannica.com/EBchecked/topic/472266/Postromantic-music (accessed March 14, 2009).
5
“Silver Age” is the term generally preferred over fin de siècle by scholars of Russian arts and literature.
6
Burkholder, Grout, and Palisca, A History of Western Music, 6th ed., 799.

28
 To be sure, tonal configurations that simultaneously engage more than one kind of
melodic-harmonic structure have their origin in the music of nineteenth century
composers. In his dissertation on Liszt’s music, Ramon Satyendra writes that “chromatic
tonality is best seen not as an exclusively diatonic or exclusively chromatic system but as
an interaction between pitch-space systems” in which distinct pitch structures may exist
simultaneously in “stacked spaces.”7 In Postromantic and early modernist works,
however, stacked or layered configurations involving conventionally tonal and
unconventional structures are taken to new heights of explicitness and complexity.
Recognizing this, Joseph Straus has written that “twentieth-century composers use
traditional methods, but transform them.”8 Straus suggests that “what we need now is a
critical framework for understanding this sort of thing. The framework we need should,
above all, be sensitive to the tension in these works between the traditional elements and
the new musical context that transforms them.”9 The present chapter outlines one
possible framework, tailored for highly chromatic late tonal works in general and
Rachmaninoff’s mature compositions in particular.

Postromantic Deformations and Structural Tensions

Marked dialogue between different kinds of musical organization—that is to say,

between conventional structures and unconventional structures, or, to borrow again from
Joseph Dubiel, between tonal “norms” and “abnorms”10 —may be considered a defining
characteristic of Postromantic and nascent modernist styles. Such a dialogue might be
primarily one of melody and harmony, or one of phrase design and/or form; it might
occur on a large scale or a small one; but it will almost certainly have implications for the
interpretation of a work’s overall design and expressive trajectory. James Hepokoski has
observed that “a central feature of the modernist aesthetic game…was to implicitly or
fragmentarily refer to the generic formal conventions, perhaps as lost gestures or the
founding gestures of the game, but then to override them. By the last third of the

7
Ramon Satyendra, “Chromatic Tonality and Semitonal Relationships in Liszt's Late Style” (Ph.D. diss.,
The University of Chicago, 1992), vi.
8
Straus, “The Anxiety of Influence,” 431.
9
Ibid., 435.
10
Dubiel, “Contradictory Criteria.”

29
nineteenth century there had arisen a whole arsenal of … ‘deformations’ of the
Formenlehre (standard-textbook) structures.”11 That is to say, comprehension of form
and expressive trajectory in a Postromantic work can depend on recognizing that the
“game” involves an interaction between conventional bases and new structures that may
conflict with those bases. In Postromantic repertory, there is often a sense that
conventional tonal organization is somehow endangered, and that the danger is part of an
aesthetic stance. As Charles Wilson has put it, “hence, for instance, in the symphonies of
Sibelius and Nielsen, the long-range articulations of functional tonality prevail only after
a prolonged struggle, even then leaving a palpable sense of their impermanence and
vulnerability.”12
However it is taken aesthetically, this entanglement of functional syntax and
chromatic procedures in late tonal music has fascinated theorists for a long time. Gregory
Proctor has formulated a “double syntax” for certain repertories.13 Daniel Zimmerman,
James Baker, and Allen Forte have developed a variety of “compound analysis”
techniques.14 Daniel Harrison’s powerful theory offers a synthesis of function and
chromaticism, as does David Kopp’s recent book; but other scholars prefer to emphasize
the friction that can arise between tonal norms and chromatic abnormalities, variously
defined.15 Thinking in terms of Dubiel’s “norms” and “abnorms” suits many passages in

11
James Hepokoski, Sibelius: Symphony No.5 (Cambridge: Cambridge University Press, 1993), 5.
Hepokoski uses “modern” to refer to a number of composers, including Strauss, Sibelius, and Elgar, not
ordinarily associated with the term. See also Hepokoski, “Fiery-Pulsed Libertine or Domestic Hero?
Strauss’s Don Juan Reinvestigated,” in Richard Strauss: New Perspectives on the Composer and His Work,
ed. Bryan Gilliam, 135-176 (Durham, NC: Duke University Press, 1992).
12
Charles Wilson, “The Twentieth Century,” in Grove Music Online, ed. Laura Macy,
http://www.grovemusic.com (accessed March 16, 2009).
13
Gregory Proctor, “Technical Bases of Nineteenth-Century Chromaticism” (Ph.D. diss., Princeton
University, 1978).
14
James Baker, The Music of Alexander Skryabin (New Haven, CT: Yale University Press, 1986); Allen
Forte, “Schoenberg’s Creative Evolution: The Path to Atonality,” The Musical Quarterly 64 (1978): 133-
76; and Daniel J. Zimmerman, “Families Without Clusters in the Early Works of Sergei Prokofiev” (Ph.D.
diss., The University of Chicago, 2002).
15
Harrison, Harmonic Function in Chromatic Music; David Kopp, Chromatic Transformations in
Nineteenth-Century Music (Cambridge: Cambridge University Press, 2002). Approaches that emphasize
various kinds of friction include Edward T. Cone, “Sound and Syntax: An Introduction to Schoenberg's
Harmony,” in Music: A View From Delft, edited by Robert P. Morgan, 249-66 (Chicago: University of
Chicago Press, 1986); Richard A. Kaplan, “The Musical Language of Elektra: A Study in Chromatic
Harmony” (Ph.D. diss., The University of Michigan, 1985); Lawrence Kramer, “The Mirror of Tonality:
Transitional Features of Nineteenth-Century Harmony,” 19th Century Music 4 (1981): 191-208; and
Andrew W. Mead, “Listening to Reger,” The Musical Quarterly 87 (2004): 681-707. On the related matter
of conflict between tonal elements and post-tonal contexts, see Joseph N. Straus, Remaking the Past:

30
early twentieth century works very well—passages in which there is a clear diatonic-
functional basis, and intense, well-defined chromatic activity interacts with and
complicates essential premises of that basis.
In theories of the common practice, dissonance results from tension between
contrapuntal elements in a single well-formed syntax. In extended tonal works from the
early twentieth century, traditional dissonance of course remains; but a higher-order
dissonance is often suggested as well—a kind of hyperdissonance that results from
tension between different layers of a stratified compound harmonic environment. In such
a configuration, a chromatic structure in one harmonic layer may exaggerate, distort, or
even directly contradict the functional tonal premises stated in another layer. I suggest
that analysis of hyperdissonance events allows better appreciation of the harmonic
roughness and expressive tensions that characterize Postromantic music; and,
furthermore, that the construction of hyperdissonance as a category provides firmer
music-theoretic ground for the style observations made by Hepokoski and Straus.
It will be useful to examine instances of hyperdissonance in short passages from
the early twentieth-century repertory in general before turning specifically to the
interpretation of climax events in Rachmaninoff’s works.16
Figure 2.1 is a passage from Strauss’s Elektra (1908) David Murray writes that
Strauss’s Elektra “absolutely presupposes a secure tonal norm against which to measure
its harsh, disorienting dramatic effects for an audience with late Romantic ears.”17 In
Figure 2.1, there is a layer of functional, “normal” tonal activity, and there is a
“disorienting” chromatic layer.

Musical Modernism and the Influence of the Tonal Tradition (Cambridge, MA: Harvard University Press,
1990).
16
In this chapter, only diatonic-functional structures and chromatic structures are considered. Modal
structures are added in Chapter 3 and described more fully in Chapter 5.
17
David Murray, “Elektra,” in Grove Music Online, ed. Laura Macy, http://www.groveonline.com
(February 20, 2009).

31
 Figure 2.1. Richard Strauss, Elektra, r. 177–1784

In a well-known article on Schoenberg’s music, Edward T. Cone examines the

interaction of “sound and syntax” in highly chromatic music.18 “Sound” in this context
refers (primarily) to vertical sonority, while “syntax” refers to harmonic progression.
Analysis of sound and syntax in Figure 2.1 provides a way into the passage, and lays the
groundwork for interpreting longer passages and, ultimately, entire works. Figure 2.2 is
an analysis of the first two measures of the Elektra passage; in the figure, two harmonic
“layers” are identified. As shown in Figure 2.2, the harmonic syntax implied in layer x
and the vertical sounds produced by the x+y compound are not concordant. Layer x
implies a three-point design, stable-unstable-stable based on leading-tone activity, as
indicated on the figure; but the addition of layer y distorts this. The resolution to tonic in
layer x at timepoint 3 is contradicted: harmonic tension of the x+y compound is increased
at timepoint 3 rather than decreased. That is to say, a basic premise of basic tonal
18
Edward T. Cone, “Sound and Syntax: An Introduction to Schoenberg’s Harmony,” in Music: A View
from Delft, ed. Robert P. Morgan, 249–66 (Chicago: University of Chicago Press, 1986).

32
organization—resolution to the tonic—has been contradicted and a kind of tonal tension
not to be found anywhere in the common practice obtains.

Figure 2.2. Elektra, r. 177–1772, analysis

! !

The analysis is continued through rehearsal 178 in Figure 2.3a. As shown by

Roman numerals in the example, layer x implies a complete functional progression,
chromatically altered but coherent. Layer y is a string of chromatically-descending
diminished triads that does not engage that syntax. Layer z provides a constant B minor
! !" !
reference-point. Timepoints 3, 6, and 9 are all points of local leading-tone-type resolution
to functional pillars inside layer x. But intervallically, they are points of maximal tension
in the x+y compound, as shown by the tension diagram at the bottom of Figure 2.3a and
supported by examination of interval vectors in Figure 2.3b. The effect is perhaps related
to the textbook deceptive cadence, whose melodic leading-tone resolution is undermined
by a failure to resolve to tonic harmonically. However, harmonic roughness in the Elektra
passage is much greater than in that common-practice prototype, because the entire triads
of resolution are actually heard in layer x.

33
 Figure 2.3. Elektra, r. 177–1782, analysis and tension diagram

(a)

! !" !

(b) interval vectors of x + y compounds by timepoint

1: [001110] (triad) 4: [012111] (7th chord) 7: [004002] (7th chord)

2: [102111] 5: [102111] 8: [213211]
3: [225222] 6: [233241] 9: [233331]

The Elektra passage suggests an analogy from the visual domain: the
“interference pattern” created when two grid patterns are superimposed, as in Figure 2.4.
In the figure, the appearance of an ordinary grid is distorted by the superimposition of a
second grid. Both grids are well-defined. It is not a case of seeing either the one grid or
the other; the visual surface is both grids including the interference pattern created. The
analogy is offered informally; but it suggests a more general schema for the complicated

34
tonal energetics of the Elektra passage: layers x and y may be likened to the two grids.
The functional activity in layer x is distorted by superimposition of non-diatonic layer y.
A paradox of tonal energetics results: timepoints 3, 6, and 9 are recognizably stable and
undeniably unstable at the same time.

Figure 2.4. Moiré interference pattern and analogy with Elektra passage

The embedding of a familiar tonal idiom in an extraordinary chromatic

environment can have the effect of “defamiliarizing” the tonal idiom, creating new
expressive or rhetorical effects while maintaining a functional tonal basis. Figure 2.5,
analysis of the opening of the “Elegy” section in Richard Strauss’s An Alpine Symphony
(1915), demonstrates.

35
 Figure 2.5. Richard Strauss, An Alpine Symphony, “Elegy,” r. 100, analysis

chromatic substitution

100

### %$ & !# & &! & & ! & &! & & ! & &! & !%
" $ $ $
$
$ !!%%!! & #& & !& !!&& #& & !& ##$& &"' &
& !& & !#&% !& #%$
$ $ & & $
A
( ### $ $ ! %% !! !&& ! && !&& ! && ##!&&& $
$
$ $ $

^
7

## #& #& & !& #&

"# &
#& & &&
( ### ##$&
B
& &
& &
( ! II )
6
8 - 7

i V i

In the Elektra passage, the different harmonic layers are quite explicit; in the
“Elegy” passage, they are implicit but still clear. In Figure 2.5, layer A contains actual
musical material from the passage; layer B shows a simple functional i – V – i prototype
that provides a basis for understanding layer A: A is interpreted as a chromatic substitute
for B. In the third measure of the passage, leading tone E♯ is treated instead as
enharmonically-equivalent F♮, and harmonized with an F major triad rather than the
hypothetical C♯ major triad (or dominant-seventh chord, etc.) shown in B. It is not
difficult to hear A as fulfilling the same basic tonal and phrase functions as B, but in a
more energetic way. One striking effect of the chromatic inflection is that passing tone
F♯ (the tonic of the passage) in measures 3 and 4 of A is intensely dissonant with the F♮
root of the triad, whereas in prototype version B passing tone F♯ is only a mild
dissonance. Similarly, melodic tone C♯ in the second half of measure 4 is, because of the

36
inflection, dissonant with C♮ of the F major triad, whereas it is a chord member in B.
These are more than foreground details. The melody of the passage, which retains the
implications of diatonic-functional B, sounds highly charged—even grotesque, to recall
the Encyclopedia Britannica entry—in the chromatic context.; yet its notes have not been
changed, only the context in which they are heard. Specific melodic pitches (the F♯’s, the
C♯) which are not problematic in hypothetical layer B have quite different effects in layer
A. The leading tone has been exaggerated by enharmonic reinterpretation (E♯ → F♮) and
chromatic substitution to such a degree that the diatonic basis of the melody is
problematized.19 Because it has not been adapted to suit the chromatically inflected
context, the melody sounds like it does not belong; the result is a tension between the
melody and the harmonic setting.
The event in Figure 2.5 is quickly followed in the work by a more powerful
corruption of diatonic-functional premises involving the same melodic material. As
shown in Figure 2.6, at the beginning of the “Calm Before the Storm,” melodic material
from the “Elegy” is implanted at pitch into the key of B minor. In the B minor context,
primary melodic tone C♯ (marked x1 in Figure 2.6) is now extremely dissonant; and the
dissonance is never resolved. Instead, C♯, reiterated in the oboe part throughout the
section (as D♭), is gradually transformed into the upper third of B♭ minor—the global
tonic of the work, which appears as dissonant x2 in Figure 2.6—as the “Calm”
continues.20 The change in tonal orientation over the course of the passage is understood
as a shift in the status of the norms and abnorms. An overview of the entire section is
shown in Figure 2.7. At the beginning of Figure 2.7, elements in layer y are heard as the

19
Even more basically, the raw melodic structure can be heard as an arpeggiation of the tonic triad over the
course of five measures: C♯6 down to C♯5 at a rate of one chord tone per measure. This interpretation
makes the actual chromatic setting even more remarkable.
20
I reject Charles Youmans’s claim that the Alpine Symphony should be understood entirely in the key of
E♭, beginning and ending on the minor dominant. See Youmans, “The Twentieth-Century Symphonies of
Richard Strauss,” The Musical Quarterly 84 (2000): 247. B♭ is the first and final tonic in the Alpine
Symphony. Youmans, however, interprets the entire work as a dysfunctional sonata form in E♭ major. In his
scenario, the opening of the work articulates the dominant (minor, then major), and the recapitulation takes
place entirely in the minor dominant. I agree that there is conflict between B♭ and E♭ in the Alpine
Symphony, but a global E♭ interpretation actually lessens the impact of the large-scale tonal conflict. Better
to say that E♭ implications—and even a form in E♭—are embedded into the global B♭ minor context, and
work against that context. It then becomes clear that the work embodies a large-scale tonal deformation
similar in many ways to the smaller-scale ones analyzed in the “Elegy” and “Calm Before the Storm”
episodes.

37
local norm, while elements in layer x are dissonant abnorms. At the end of the figure, as
the next section of the work (“Thunder and Storm”) begins, precisely the reverse holds.
By the end of the passage in Figure 2.7, the torque applied to elements in layer x has
dissipated, but not as the result of any clear functional process. Rather, a chromatic
progression—c elements in Figure 2.6—leading to a diminished seventh chord at the
midpoint of the passage acts as connective tissue.

Figure 2.6. An Alpine Symphony, “Calm Before the Storm,” mm. 1-6, analysis

38
 Figure 2.7. An Alpine Symphony, “Calm Before the Storm,” analytic overview

39
 In the “Calm Before the Storm,” local tonal function in B minor is distorted by the
implantation of thematic material in another key. At the same time, a different tonic (B♭
minor) insinuates itself into the passage in an unconventional way—as a dissonance
ultimately made consonant as though by force of compositional will. The result is a
compound structure that simultaneously articulates diatonic-functional syntax, with its
precise implications of tension and resolution, and an intense chromatic structure that
undermines those implications.

Hyperdissonance: Definition and Initial Analytic Applications

Although the core concept is intuitively clear, hyperdissonance has proved

difficult to define with precision. It is perhaps as much an epistemological anchor in a
repertory that has sometimes seemed intractable as a bona fide technique. Nevertheless, I
offer the following:

hyperdissonance

In a Postromantic work, tension between a diatonic-functional tonal basis and

some explicit chromatic (or potentially modal) structure or structures, resulting in
exaggeration, distortion; neutralization of functional premises, fragmentation of
tonal patterns; and/or deformation on larger scales. Hyperdissonance is associated
with higher-order structural processes, as variegated components of a compound
melodic-harmonic environment are juxtaposed and worked out. In Rachmaninoff’s
mature style, hyperdissonance is often associated with climax.

Hyperdissonance events occur frequently in works Rachmaninoff composed

during the late Russian and exile periods, providing support for the claim that they are
fundamentally Postromantic, not Romantic, in expressive and rhetorical orientation.

40
Common practice models are invoked, but deformed. In the works studied,
hyperdissonance and climax are very often associated. Figure 2.8 is an analytic reduction
of the climax before recapitulation in the first movement of Rachmaninoff’s Symphony
No. 3 in A minor, Op. 44. The climax is interpreted as involving a tension between
conventional harmonic functions embedded in Dies irae-like layer A and
characteristically Russian “fantastic” chromaticism in layer B.21 The basis of layer A is
the establishment of dominant-type leading-tone energy and the functional resolution to
tonic, as shown. The “fantastic” chromatic setting distorts the functional basis, yet
preserves the pitch-class framework that defines the basis.

21
“Fantastic” chromatic structures involving equal division of the octave—octatonic, hexatonic, and whole-
tone—are treated more fully in Chapter 4. The convention of using Arabic numerals to represent pitch
classes to avoid issues of enharmonic notation is adopted in this dissertation. For example, following
Joseph Straus, OCT(0,1) refers to the octatonic collection containing pitch classes C and C♯ (or D♭). See
Straus, An Introduction to Post-Tonal Theory, 3rd ed. (Upper Saddle River, NJ: Pearson/Prentice Hall,
2005).

41
 2.8. Symphony No. 3, Op. 44, i, analysis of climax at the end of the development

42
 A fundamental premise of tonality is that achieving the tonic corresponds to a
lowering of harmonic tension. At the climax of the Rachmaninoff symphony passage, as
in the Elektra passage, that principle is turned on its head: arrival of the tonic triad after
rehearsal 21—made especially clear by a tonic statement of the symphony’s motto theme
in the brass at 215 and the rising melody in the violins at rehearsal 22—coincides with a
heightening of tension, because the intense chromatic activity (specifically, the long-held
G♯–F♯ seventh in the lower staff of layer B) does not accept the resolution.22
To say that this tonic triad is stable throughout the climax, that it simply takes
some time for the other elements to catch up to it (at rehearsal 25), is to miss the dramatic
point, especially as it takes a full 32 measures for the others to catch up. In Figure 2.7, the
tonic and the image of a progression that achieves that tonic are recognizable, but the
chromatic context has dramatically defamiliarized them. The tonic triad is under
considerable duress.23 When the melodic and dynamic apex of the passage is achieved, ff,
in the flutes and violins at rehearsal 23 the leading tone G♯ is finally resolved in register;
yet even here the long-held G♯ - F♯ minor seventh in the lower staff of layer B does not
support the resolution to tonic. Only at the actual moment of recapitulation (rehearsal 25)
is the tonic stabilized; the discharge of its unusual energy is invisible, but not inaudible—
that is to say, unlike a conventional dissonance, the dissonant tonic does not itself move
to resolve; rather, the context is adjusted around it, correcting the tonal error, as it were.
There are some partial precedents for such an event in the common practice. A
rhetorically emphasized cadential 6
4 may raise tonic awareness even in an unstable
context. This happens, for example, when Beethoven in the first movement of his
“Appassionata” Sonata, and Rachmaninoff in the first movement of his Second
€
Symphony, Op. 27 (rehearsal 1717), begin a movement’s recapitulation over a dominant
pedal. Another precedent may be heard in the first movement of Beethoven’s “Eroica”

22
The sense of A minor as tonic is inescapable. At the same time, the suggestion of possible dominant
function in the key of C♯ throughout the climax (V13, with B♯ spelled as C; see “Possible interpretations”
box at top right of Figure) is important. As discussed in the longer analysis of the Symphony in Chapter 6
the dissertation, C♯/D♭ has a special significance as a tonal center in the Symphony, and is an extremely
important key in Rachmaninoff’s works in general. To put it another way, Rachmaninoff’s adherence to the
conventions of monotonal sonata form demands that the climax chord be resolved to A minor, but the
chord itself has additional potential.
23
The defamiliarized tonic suggests what Richard Taruskin has called "the old Russian ploy of parading
Self as Other." See Taruskin, Defining Russia Musically (Princeton: Princeton University Press, 1997), 107.

43
Symphony. In the “Eroica,” at what Leonard Ratner identifies as the point of furthest
remove in the development (the location of greatest tonal stress in a conventional sonata
form), tonic note E♭ is re-imagined as highly-charged D♯, the leading-tone of distant E
minor.24 The unstable tonic chord in Figure 2.8 is different, however, in that it occurs not
at a point of remove, but at a point of return; tonal tension is measured not in terms of
distance from the tonic, but in terms of what has been done to the tonic triad to give it
special meaning. The climax in the “Eroica” movement is an extraordinary (early)
Romantic tension event; the climax in Figure 2.8, on the other hand, is an extraordinary
Postromantic tension event.
Such dramatic maltreatment of the tonic as a result of intense equal-interval
chromatic structures plays a special role at climaxes in many of Rachmaninoff’s works—
indeed, as discussed more fully in Chapters 3 and 4, intensification and climax are the
two rhetorical characteristics most clearly associated with special chromatic structures in
his mature works. Whereas composers of late tonal music often withhold the tonic to
heighten the sense of tension, Rachmaninoff is often keen to emphasize the tonic, which
in extremely chromatic contexts can create a very different kind of tension. “To postpone
the first clear presentation of a composition’s tonic is a characteristic Brahsmian gambit,”
Dubiel writes.25 One might say that to insist upon a composition’s tonic even in
chromatic contexts that apparently deny it is a characteristic Rachmaninoffian strategy.26

Formalizing the Model: Tension Arcs

Research on tonal tension, expressive shape, rhetorical design, and climax

suggests ways to develop a more precise conceptual and interpretive framework for
events like the one in the Symphony No. 3 passage analyzed above. Leonard Ratner has
described the tonal design of classic sonata form as a “two-stage action”: “centrifugal
motion (away from I)” begins during the exposition and continues until a “critical point”

24
Ratner, Classic Music, 227.
25
Dubiel, “Contradictory Criteria,” 81.
26
Such emphasis on the tonic even (especially) in highly chromatic contexts recalls Joseph Yasser’s
observation that Rachmaninoff’s chromatic language is characterized by strong “intra-tonal” organization,
as opposed to the “inter-tonal” organization of Wagner’s chromaticism. See Yasser, “Progressive
Tendencies,” 21.

44
(a “point of furthest remove”) is reached in the development, after which “centripetal
motion (toward I)” begins.27 Therefore, in his view, the “principal object of the
development … is to regain the tonic.”28 Ratner’s model suggests applications beyond
sonata form. A generalized version is shown in Figure 2.9.

Figure 2.9. Generalized tension arc

critical
point cen
tri
al pe
tal
tr ifug
cen maximal
tension
tonic tonal fluctuation tonic

(functional resolution)

Figure 2.9, which I refer to as a “tension arc,” recalls Ratner’s own “dynamic
curves,” which were briefly discussed at the end of Chapter 1. The arc applies in many
musical contexts, not just in sonata form movements, and it may be used to model tonal
and formal events on many scales. One premise of the Figure 2.9 model is that the crisis
at the apex of the tension arc (the point of greatest tonal instability and, most likely,
expressive focus—the “crux,” or “highpoint” in Hatten’s, Stell’s, and Agawu’s sense)
and the resolution to tonic are separate events in separate locations. As noted above, in
the Rachmaninoff Symphony No. 3 passage, this premise is radically undermined.
Arrival of the tonic after rehearsal 21 coincides with a heightening of tension, because the
intense chromatic context, involving symmetrical harmonic idioms of a characteristic
Russian sort, does not support the embedded resolution to tonic. The tension arc
suggested by the functional tonal framework and, more generally, the tenets of sonata
form, is deformed, as shown in Figure 2.10.

27
Ratner, Classic Music, 209, 225.
28
Ibid., 225.

45
 Figure 2.10. Symphony No. 3, i, deformed tension arc at r. 22–25

return to tonic contradicted

Chromatic layer

dissonance?

leading-tone,
Functional basis tonic dominant function tonic!

resolution

Tension arcs provide a way to represent the effects that hyperdissonance can have
on the expressive trajectory of a Postromantic passage or work. There is considerable
support for such a model. Departure-return metaphors, and various kinds of arch or arc
diagrams to show goal-orientation, expressive shape and/or the ebb and flow of tension,
are found throughout the literature. Leonard Ratner’s description of sonata form and his
“dynamic curves” have already been noted, as has Jason T. Stell’s discussion of
“expressive curves” in select piano preludes by Rachmaninoff. Similarly, in Candace
Brower’s cognitive theory of musical meaning, departure and return (one example of her
more general “source-path-goal” model) is a primary “music-metaphorical schema”—a
basic pattern to which musical events are matched and from which musical meaning is
gathered.29 Brower’s Figure 27 (reproduced here as Figure 2.11) provides a prototype for
“how the phrases of a musical work can be understood as a series of goal-directed
motions, with smaller arcs of motion nested within larger ones. The diagram “captures
the way that harmony, melody, and rhythm work together to articulate a series of
completed motions within an overall progression of departure and return. In its depiction
of a specific number of phrases and relatively specific tonal plan, it constitutes more of a
prototype than a schema.”30 Brower’s description echoes Ratner: “the overall trajectory
of harmonic motion shows the expected cycling of harmony away from the tonic and

29
Candace Brower, “A Cognitive Theory of Musical Meaning,” Journal of Music Theory 44 (2000): 323-6,
331.
30
Ibid., 350.

46
expansion of the tonic-dominant cycle...revealing a general tendency toward motion
leading away from tonic to a point of greatest tonal distance followed by a motion of
return...Each phrase is represented as having two distinct goals: the climax of the
phrase—the turning point between tension and relaxation, and the maximally stable event
at the end of the phrase.”31 For both Ratner and Brower, some kind of crisis—a limit, a
change of direction or conceptual reversal, a climax—happens at the point of furthest
harmonic remove. The apex of the arc in Brower’s diagram is, then, quite likely to be a
tensional highpoint in Agawu’s terms; and in Brower’s diagram, it is represented as an
actual melodic peak, too.

Figure 2.11. Candace Brower’s schema for phrase structure

Fred Lerdahl’s theory of tonal tension also generates arc-shaped diagrams for
tonal structures, showing melodic and harmonic fluctuation around a referential tonic (a
pitch in the case of melody, a chord in the case of harmony).32 Wallace Berry has gone so
far as to suggest that in chromatic contexts, “one is almost tempted to assert that tonal
structure is best characterized not in terms of specific tonics, but, rather, in terms of the
pattern (tripartite) stability-fluctuation-stability.”33 Berry recognizes that, even when a
passage resists description in functional or linear terms, a trajectory referable to basic

31
Ibid.
32
Fred Lerdahl, “Calculating Tonal Tension,” Music Perception 13 (1996): 319-82.
33
Wallace Berry, Structural Functions in Music (New York: Dover, 1987), 140.

47
tonal premises often remains. I take this to be a core feature of Postromantic works in
general.

Exaggeration of Tonal Premises

There are three basic ways that hyperdissonance can impact a passage’s
expressive shape, or, more formally, its tension arc: exaggeration, distortion, and
neutralization. I suggest that each of these is a kind of deformation in Hepokoski’s
generalized sense of the word. The first two are dealt with at length in the following
pages. The third (neutralization) represents a more radical tonal possibility that has little
practical application in the analysis of Rachmaninoff’s music, and is therefore presented
only in passing.34 Although I have attempted to define exaggeration and distortion as
precisely as possible to maximize their potential in analysis, in the final assessment they
represent a way of thinking about the expressive effects of unusual kinds of pitch
organization, not a strict system of classification.
If tension between diatonic-functional premises and a chromatic structure does
not explicitly undermine the resolution to tonic (or other strong local goal), but, rather,
amplifies the tension arc to an extent not possible in conventional tonal syntax alone, then
exaggeration occurs. This may be thought of as hyperdissonance at the point of remove,
as diagrammed in Figure 2.12. Essential tonal premises—departure/fluctuation generates
tension, functional resolution to a goal provides stability—are not undermined, but
greatly amplified. In such situations, the exaggeration itself can be the most significant
expressive, rhetorical, or style factor, overtaking the functional tonal basis in
significance.

34
But see the comparison of Rachmaninoff’s and Skryabin’s equal-interval structures in Chapter 4, where
the possibility of functional neutralization is considered in more detail.

48
 Figure 2.12. Hyperdissonant exaggeration

hyperdissonance

Chromatic Layer

Functional Basis tonic [fluctuation] tonic

(or other strong local goal)
resolution

Figure 2.13 gives an example of hyperdissonant exaggeration at rehearsal 87 in

the second movement of Prokofiev’s Piano Concerto No. 3, Op. 26 (1921). As shown in
the figure, a plagal gesture in the orchestra at 87 in the orchestra is converted into an
octatonic version in the solo piano part two measures later. Although the resolution to E
minor is not seriously disrupted, substitution of an OCT(1,2) structure for the plagal one
has a number of unusual effects. Tonic note E♮, part of the plagal sonority, is instead
momentarily dissonant in the octatonic version; and the subdominant “root” (A♮), must
be understood as a non-collection tone. These local effects are not powerful enough to
seriously disturb (or distort) the tonal basis, but the octatonic substitution turns the solo
piano version into something mildly grotesque.
A similar event in the Rhapsody on a Theme by Paganini is analyzed in Figure
2.14. The figure shows (1) Paganini’s theme as it is first heard at measure 33 in the
Rhapsody, (2) a passage from Variation VIII, and (3) a passage from Variation IX.
Paganini’s theme is in binary form; the first part is repeated, but the second is not.
(Rachmaninoff invariably writes the repeat out in full to allow double-variations
procedures; I have used a repeat sign in Figure 2.14 to save space.) Of interest at present
are the first four measures of the second part of the theme (mm. 37–40 of (1) in Figure
2.14). The measures are sequential, tonicizing iv and then III, and they contain the first
harmonies other than tonic and dominant in the theme. Note the appearance of B♭ in
measure 37, which strengthens the tonicization of iv. Measures 37–40 represent a clear

49
departure, and are followed by a clear return over the course of the rest of the theme and,
ultimately, a strong resolution to tonic.

Figure 2.13. Prokofiev, Piano Concerto No. 3, Op. 26, ii, analysis of r. 87

!
!"#$#"%&'()"*+,)- 01+#+)-,1%'2#$$'&#+,)-%
./
#$ ## $
$
#
" $ $
$ $ $
! "%"##$ %## $
$
$
" ' "' %$
!"#$%&'"( )*(+!

"%'$ "'
#$ # $ & $
$
! $
" '" '"
( (

!
" $ $
! %$ $
" '
! ' '' ""'' %''
' '
&" ' "' 3 ' "'
' "' '
,%-(.%,

3%)&,$,-#"%45%6&))+7%89% :
-)+%;';<'&%)=%1)""'1+,)-

50
 Figure 2.14. Hyperdissonant exaggeration in passages from Rhapsody on a Theme by Paganini, Op. 43
B♭
Theme
33  
 
                        
 
1                            
                        
 
p   

V iv V III

Var. VIII
300
                                  
                      
   
 
 
ff ff
2

51
 
         
                
                             
                   
                                                 

iv III

* *

Var. IX
344
    
                   
     
                  
                                
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
mf
3   
                
                          
 
   
     
 

 
f f
   
cresc.
 
p cresc. p

* ! * III

 At the corresponding location in Variation VIII (2), which Rachmaninoff marks
ff, a B♭ minor triad is substituted for V of iv, resulting in not a fifth root relation but a
chromatic major third root relation (marked with an asterisk in Figure 2.14). As a result,
the roles played by A♮ and B♭ in the melody are exchanged. This is shown by the
stemming in Figure 2.15, whose (1), (2), and (3) correspond to those in Figure 2.14. A♮,
primary in (1), is dissonant in (2); B♭, dissonant in (1), is a local chord root in (2).
Although the local resolution to iv and then to III is maintained in Variation VIII, the ff
chromatic outburst has exaggerated the departure stage of the form, and mildly
problematized certain basic tonal details of the functional tonal theme.35

Figure 2.15. Rhapsody on a Theme by Paganini, mm. 41–42, 300–303, 344–347, analysis

! " #

# "# # # # # " # # %# # "# # "# # %# #

%# # # "# $# " ###
" #
# ##
)
! # "" ### $ ## %## $# #
" # "#
)
# #
1)2)3
, */ % */ % *

In Variation IX, chromatic complication of basic tonal syntax is taken a step

!"#$%&$'()!"##$%&'(&)*+*&,()*+,-)./()0
further at the corresponding location (m. 344). The B♭ minor triad is retained (it is in fact
% rather than two); however, it resolves in measure 347 not to a D
held for three measures
$

minor triad but to an A minor triad—that is to say, to the tonic triad. This is shown in
simplified form in Figure 2.15 (3). This has the effect of more seriously deforming the
functional tonal basis of the theme. A clear resolution to the tonic triad in the departure
stage of syntax or form—even as a passing event—is usually of considerable interest. At
measure 347 in Variation IX (Figure 2.15 (3)), the tonic results from a gradual process of
chromatic substitution. Step 1: in Variation VIII, a chromatic chord substitutes for V of

35
As the longer analysis of the Rhapsody in Chapter 6 dissertation shows, the substitution of a chromatic
third relation for a diatonic relation in Variation VIII has implications for the large-scale structure of the
work, which can be understood as involving a cycle of major thirds.
"& ( ( ( ( (

"!
) 52 ' # " !

# " # # %# ## # # " # # %# $## # # "# # "#

-&.&
"#
"# %# # # "# $ # %# # # %#
 ! " #

# "Variation
# # # IX,# the chromatic chord
# " # is#retained, " # is#
%# # but" #a different #chord %# #
# # ""###
iv. Step 2: in
%#
) ! for its resolution, altering the "shape
## of the passage.
$## %## #
# ##
)
substituted
# # "" #
# $ # "# "# $#
#
#
Comparison of Figure 2.14 and Figure 2.15 (3) with Figure 2.16, Heinrich
1)2)3
Schenker’s , */ theme, suggests%the extent of*/the deformation, while
analysis of the % also *
revealing that core tonal premises nevertheless remain intact.36

Figure 2.16. Schenker’s analysis of Paganini’s theme (Free Composition, Fig. 40, 9)
!"#$%&$'()!"##$%&'(&)*+*&,()*+,-)./()0
%
$

As shown in Figure 2.15, the process of chromatic substitution process changes

"&
the structure of the top ( ( (
voice in Variations VIII and IX: in Figure 2.15 (2) and 2.15 (3), ( (

"!
) ' # " !
the descent to F♮ and D♮ is stunted as the melody remains focused on A♮, and Schenker’s
# " # 4# (see
%# ##Figure 2.16)
# " is# therefore # # "If
-&.&

# "# # %# # eliminated.
$#
# a# 5-line
" # %# #reading $is# to
# # "#
"#
"# ##
scale degree %# obtain,
# all%#
!the events #
# # # % ## # #
# " # $ # # %#
" # " # $# $ #
$# "#
" # "#
) which follow the chromatic substitution must be reevaluated. Such an $#
# ##
interpretation is possible, as shown in Figure 2.17, suggesting that tonal premises have
not truly been undermined in Variation IX. Nevertheless, the effect of the chromatic
intensification is pronounced. Indeed, the dynamic, textural, and registral characteristics
+,- * ...
&
" .. , *
of the passage, shown crudely in Figure 2.17 and more precisely in Figure 2.15, suggest
that Rachmaninoff intended the chromatic substitution in Variations VIII and IX to be
(#03$*,+#,)*+4
jarring. In Variation IX, the return of comparatively normal tonal functions later in the

36 /#."#0)*,1$+#,)*&,
Heinrich Schenker, Free Composition (Neue Musikalische Theorien und Phantasien, III: der Freie Satz),
ed. and "*)*,1$"#1*)+#"
trans. Ernst Oster (New York: Schirmer Books, 1979). 20--*,1$"#1*)+#"

53
 %

$
passage is, by contrast, rhetorically unmarked. The exaggeration of the tension arc is the
main issue in these two variations.

Figure 2.17. A Quasi-Schenkerian reading of Rhapsody, Variation IX, r. 26

"& ( ( ( ( (

"!
) ' # " !

# " # # %# ## # # " # # %# $## # # "# # "#

-&.&
"#
"# %# # # "# $ # %# # #
#
# #
# %# # % ## #
"""### " # $# "# $$ ###
! #
)
" # "# $# # # %# # $# #
#

+,- * ... " ..& , *

(#03$*,+#,)*+4

/#."#0)*,1$+#,)*&,
"*)*,1$"#1*)+#" 20--*,1$"#1*)+#"

Chromatic exaggerations such as those in the Prokofiev and Rachmaninoff

passages just analyzed are of course not entirely new in the Postromantic. Chromaticism
in more conventional contexts may generally be thought of as intensifying the
departure/fluctuation stage of tonal organization, sometimes to the point that the
instability that results seems the defining musical characteristic. For example, Geoffrey
Chew has examined the effects of intense chromaticism on local climaxes in the Abschied
from Act II of Wagner’s Tristan und Isolde along lines suggested by Ernst Kurth.37 Chew
observes that, while in Schenker’s theory the tonic and dominant are stable pillars to
which all other contrapuntal and harmonic events are at some point bound in analysis, in
a Kurthian interpretation of harmonic “instability” in chromatic contexts, events neither

37
Geoffrey Chew, “Ernst Kurth, Music as Psychic Motion and Tristan und Isolde: Towards
a Model for Analysing Musical Instability,” Music Analysis 10 (1991): 171-193.

54
the tonic nor the dominant may be considerably more crucial, and may in a sense become
the conceptual pillars. Chew notes that:

“...Dominant harmonies at the end of the refrains are points at which the emotion
(dependent on leading-note tension) has passed its peak and subsided...The
climaxes themselves – the points of greatest tension and instability – give the
Abschied its characteristic dramatic shape, and so they may have some claim to be
thought of as Grundpfeiler in a Kurthian sense, even though they cannot be
regarded as such in any Schenkerian sense.”38

In other words, intense chromaticism achieves its most powerful effects before the
dominant stage of syntax, and the phrase or passage is thereby restructured in some sense.
As Kurth put it:

“...One experiences the leading-note effect in these contexts the more strongly
where originally, in the blandest and most hackneyed form of the scale, no
semitones occur; for this reason, every chromatically altered note has an effect
even stronger than the chromatic tension of the leading note in its normal
position…”39

This relocation of harmonic tension from the dominant to the predominant stage in syntax
is, for Chew, a defining characteristic of the chromatic idiom.
It is possible to map functional progressions onto a basic tension arc in a way that
successfully models a majority of cases, reflecting Brower’s, Kurth’s, and Chew’s
observation that the predominant stage of a functionally-organized phrase or passage is,
in chromatic Romantic and Postromantic works, generally the most intense and the most
rhetorically accented. (See Figure 2.18.)

38
Ibid., 187.
39
Quoted in Geoffrey Chew, “The Spice of Music: Towards a Theory of the Leading Note,” Music
Analysis 2 (1983): 45. See also the discussion of Kurth’s theories in Lee A. Rothfarb, Ernst Kurth as
Theorist and Analyst (Philadephia: University of Pennsylvania Press, 1988); and Rothfarb, “Energetics,” in
The Cambridge History of Western Music Theory, ed. Thomas Christensen, 927-55 (Cambridge:
Cambridge University Press, 2002).

55
 Figure 2.18. Tonal functions mapped onto a prototypical tension arc

(PD)
(x)
apex (D)
(SD)

departure maximal return

tension
tonic expansion, progression, fluctuation tonic

(functional resolution)

Many passages in Rachmaninoff works from all periods follow this plan without
necessarily suggesting interpretation in layers or hyperdissonance as an analytic
framework. Figures 2.19 and 2.20 present an overview of the main climax in the well-
known middle-Russian-period Prelude in D major, Op. 23, No. 4 (1903).40 The chromatic
materials in the passage are of a very conventional kind, including modal mixture and the
Neapolitan, and neither hyperdissonance nor harmonic stratification is suggested.
In Figure 2.19(a), the motivic material of the prelude is shown: a neighbor figure
(N), and a rising figure (R). The piece is in ternary form. Figure 2.19(b) shows that the A
section is periodic; its defining perfect authentic cadence in measure 35 overlaps with the
beginning of the B section. Although the period is parallel, there is development within
the A section. Most obviously, a countermelody in triplets is added in the consequent
phrase. Less obviously is the emphasis on a new chromatic tone in the consequent phrase:
E♭ joins A♯/B♭, as shown in Figure 2.19(b).

40
This prelude is discussed in Stell, “Rachmaninov’s Expressive Strategies,” 77–92. Many of my findings
are consonant with his. One difference, however, is that I adhere to a ternary form interpretation of the
work, while he finds a ternary interpretation insufficient.

56
 Figure 2.19. Prelude in D major, Op. 23, No. 4, analysis of A section (mm. 1–35)

a) Motivic material

!
! " ##
"
! " # # # # # # # # # # # # # # "# #
" "

#
!

$ "" # #

#
!

$%&'(!)*+,-

b) Period design and introduction of chromatic tones

()*"+,$%-

!"#$%&'
78"#2)9 !"#$%$&$"# ()"*$+,$"#
! # !" !% #$
'( -!(

! "
.//0!///1"#2)&,%$& ! "
.//0!///&$3$4)5$&
!
6///1"#2)&,%$&

These basic materials are developed into the climax shown in Figure 2.19. The
neighbor figure is retained, and the rising figure is gradually extended. The climax that
arrives in measure 51 is unmistakable. It is the moment of peak chromatic intensity in the
piece, an arrival at registral extremes following an ascent in the upper voice and a descent
in the bass, the occasion of peak textural and dynamic intensity (ff; compare the pp at the
beginning of the section); and it immediately impels a reversal of melodic
directionality.41 The climax chord—♭II in measure 50—represents a synthesis of the two

41
Stell considers the expressive implications of the reversal in detail on pp. 87–92.

57
chromatic tones exposed in the two phrases of the A section (see again Figure 2.18(b)).
But the Neapolitan is not treated conventionally. As Stell has noted, the addition of C♯ in
an upper register suggests the German augmented sixth of G major/minor, correct
resolution of which would continue the wedge shape in the outer voices.42 Instead,
however, ♭II is directly replaced by ♮II in a fashion that Rachmaninoff would duplicate
four years later in the D major second theme of the Symphony No. 2’s finale.43 At the
moment of climax, B♭ in the right hand part becomes A♯ and pushes to B♮, while E♭ is
blocked from taking the next step (to D♮), and leads back to E♮.
Hyperdissonant exaggeration may be considered an extreme, historically specific
form of chromatic amplification. Analysts have previously recognized its significance in
Postromantic works, but have never to my knowledge named or formalized it. Richard
Kaplan’s analysis of the dense chord heard at the ff climax near the end of Richard
Strauss’s Salome is one example. His analysis is reproduced in Figure 2.21; the passage is
shown in reduction in his system “a” and parsed into two neighbor chords in his system
“b”.44 Kaplan interprets the climax chord as a compound of the two neighbor chords,
entangled yet clarified by register. The chord shown on the top staff of system “b” is
“generated principally by the diatonic melody” and as such is tightly bound to the
diatonic-functional basis, as its Roman numeral label suggests.45 Kaplan observes that the
chromatic neighbor chord in the bottom staff of system “b” “combines as dissonantly as
possible with this melody: each pitch class of the lower chord forms a semitonal
relationship with a pitch class of the upper chord.”46

42
Stell, “Rachmaninov’s Expressive Strategies,” 84–86.
43
See mm. 185–188 of the symphony movement. The prelude and the symphony theme have other features
in common, including a wedge formation of the outer voices and the gradual exposition of chromatic tones
B♭ and E♭. The similarities suggest that the prelude might have served as a model for the symphony theme.
44
Kaplan, “The Musical Language of Elektra,” 48
45
Ibid.
46
Ibid.

58
 Figure 2.20. Prelude in D major, analysis of climax in B section (mm. 35–53)

B section

PAC 35 37 39 41 43
 N R R  N R  N R 
N R
  N                  
     
   
    
                          
pp
    
      
vi 7 # § 6
5

V = iv

[
V I = ii V ii
departure


59
climax reprise
51 A section
ascent  
49 release
45 47 R         53
R         
N R R N
                                     
        
                  
 
  (N) p cresc. ff dim. mf
  
      
descent

§ !
7
 
IV 7
B T

V I
II ii
+6
(Ger of G)
chromatic amplification return
 Figure 2.21. Richard Strauss, Salome, Op. 54, 3605 - 361 (Kaplan’s Example 1-25)

%%%
! #### # ! % % % %% %% %% %% %%
3
%% %% % &
&
&
" # #! $ %% %% %% % %% %% %% %% &
&
% %
(( ((( &
#### # ! &
3
(( &
a " # # ! ' (( )()((( &
#&
#&
(( ))((( #&
* # ## # ! ' &
" # # #! ( )( &
( )( &

! #### #
(( ((( ((
(( (( ((
" ##
ø7
vii
b
(( ) (( # ((
* # ## # ( )( #(
" # ## ( (
+(
)(
6
I 5 I

In Figure 2.22, I contextualize Kaplan’s analysis using the current framework; the
three harmonies (tonic–compound neighbor chord–tonic) are mapped onto a three-point
tension arc. As the figure shows, intense friction between the two neighbor chords
exaggerates the basic shape suggested by the diatonic melody. Both neighbor chords
resolve to tonic; nevertheless, the friction at the point of remove—a moment of
hyperdissonance in a layered configuration—essentially defines the moment. That is to
say, the complex, unstable sonority, not the tonic resolution, is the conceptual anchor of
the event.

60
 Figure 2.22. Diagram of hyperdissonant exaggeration at the Salome climax

!"#$%&"'(
&)"*+,%-( !"#$%&'($)*
'+%-!(

!"#$%& /0-"'$"%&1( !"#$%&

'+-%.#$"'(
&)"*+,%-( !"#$%&'($)*
'+%-!(

"+,--"!,'($)
$.*()#',/(%('0

2$#,3)( 2$#,3)(

Similarly, the famous nine-note chord heard at the climax in the first movement of
Mahler’s Tenth Symphony (measure 208), has been interpreted by Kaplan, and Agawu
following him, as a radical, layered chromatic exaggeration of a functional basis.47 The
chord is considered a compound sonority referring to two keys at once—V♭9 of F♯ major
plus V9 of B♭ major, both of which key areas are significant in the work, but only one of
which (F♯ major) emerges from the climax event. Figure 2.23 contextualizes their
analysis according to the current theory. As the figure shows, the event demands tonal
comprehension, yet depends almost entirely on a unique set of structural circumstances
that reside outside ordinary tonal grammar.

47
Kaplan, “The Musical Language of Elektra,” 49; V. Kofi Agawu, “Tonal Strategy in the First Movement
of Mahler's Tenth Symphony,” 19th Century Music 9 (1986): 222-233.

61
 Figure 2.23. Mahler, Symphony No. 10, i, interpretation of climax (m. 208) following Kaplan

)*$+,-

!"#$%&( !
! %% $$$
!
!
$$
' # 2),3

& $$$
"
"$
"$
.
'
"$

"
" 24$*3
!"#$%&'

#$%&'($)*+,)-
./0,1%(+.'.0+*+*

"
/"0-!'&!"*1!$,*+***,-.$!
"
!"#$%&'($)*'$*+***,-.$!

Distortion of Tonal Premises

If a clear statement of tonic or resolution to some other strong local goal in one
harmonic layer is explicitly contradicted by a well-defined structure in another layer, as
happens in the Symphony No. 3 passage analyzed earlier in this chapter, then a kind of
distortion occurs. An essential tonal premise has been undermined, and an extremely
conflicted, powerfully marked tonal situation results. As demonstrated in the analysis of
the Symphony No. 3 climax in Figure 2.8, this may be thought of as hyperdissonance at
the point of return (or, occasionally, at the point of departure), and is diagrammed in

62
Figure 2.24. In such a case, the chromatic “abnorms” overwhelm and distort tonal
premises, which nevertheless remains in another layer. Although there are some partial
precedents for hyperdissonant distortion in the common practice, I believe it is for the
most part a Postromantic trait, and as yet little understood. The theory developed here is
only a starting point for the interpretation of such events; further work in the repertory
will be necessary before the extent to which this model applies in a wider repertory will
be determined.

Figure 2.24. Hyperdissonant distortion

return to tonic contradicted

Chromatic Layer

hyperdissonance

Functional Basis tonic [fluctuation] tonic!

(or other strong local goal)
resolution

Figure 2.25 presents an analytic overview of the climax in Rachmaninoff’s song

“A-u!” Op. 38, No. 6 (1916). The climax involves clear hyperdissonant distortion—that
is to say, severe conflict between some functional tonal premise and a strong, well-
defined chromatic structure at a point of return.48 Specifically, the climax is interpreted as
the result of conflict between OCT(1,2) and OCT(2,3) structures in layer 2 and D♭ major
tonic elements in layer 1. Layer 1 is organized around dyad F/A♭ (motivic in the song)
and includes pitch class D♭ (C♯) after measure 24. Functional D♭ major is established at
the start of the song (not shown in Figure 2.25) by passing from the tonic to the dominant
between measure 1 and measure 4; but it is largely abandoned (or at least radically de-
emphasized) as octatonic methods gradually intrude and come to prominence. As shown

48
In Figure 2.25, the poem by Konstantin Balmont is omitted from the vocal part except at the climax. The
song’s title comes from the shriek heard at the climax.

63
in Figure 2.25, the return of tonic—really more a tortured projection of the tonic,
prepared by scale degree 5 (A♭) throughout measure 23—at the box in measure 24 is
extremely tense. Restabilization is a gradual process, not complete until measure 28 (if
even then), and, like the restabilization of tonic in the Symphony No. 3 passage (see
again Figure 2.8), more an “invisible discharge” as layers are reoriented than a genuine
resolution of the unstable tone(s).
Figure 2.26 provides a more detailed analysis of layer 2 octatonic activity at the
“A-u!” climax. Figure 2.26(a) shows how an octatonic oscillation like that in the
“Coronation Scene” of Mussorgksy’s opera Boris Godunov is extended to the form heard
in measures 24–25.49 This is connected to measures 26–27 in Figure 2.26(b), showing
how the two diminished-seventh chords contained in Oct(1,2) (labeled #1 and #2) are
entangled, creating a dense OCT(1,2) frame against which the projection of D♭ major tonic
is highly unstable.

49
On octatonicism in the “Coronation Scene” of Mussorgsky’s Boris Godunov, see Chapter 4 of the
dissertation and Allen Forte, “Musorgsky as Modernist: The Phantasmic Episode in 'Boris Godunov',”
Music Analysis 9 (1990): 3-45.

64
 Figure 2.25. “A-u!” Op. 38, No. 6, analysis of climax (mm. 21–28)

65
 Figure 2.26. Analysis of OCT(1,2) structure at the “A-u!” climax

(a) Extension of “Coronation”-type octatonic oscillation

(b) Entanglement of diminished-seventh chords and OCT(1,2) melodic segment

The climax event has an effect on the harmonic material of the song’s postlude.
As Figures 2.27(a) and 2.27(b) show, layers 1 and 2 merge to create a sort of hybrid D♭
major collection that is partly diatonic and partly octatonic; and things are left dangling at
the end, in D♭, but with a quasi-octatonic binary that looks elsewhere as it were (Figure
2.27(c)).50 Figure 2.28 provides a schema for the entire passage: at the climax, “little” D♭
major on top is highly unstable in relation to the octatonic layer, which is, however,
highly charged in relation to “big” D♭ major underneath. It is a wrenching moment—the

50
The hybrid and the dangling E♭ recall measure 2 of the song, where they occur in passing. The hybrid is
in fact the so-called “acoustic collection,” whose significance in other Russian works from the period is
discussed in Clifton Callender, “Voice-Leading Parsimony in the Music of Alexander Skryabin,” Journal
of Music Theory 42 (1998): 219-33; and Zimmerman, “Families without Clusters.”

66
point of maximally roughness and expressive intensity in the song; the climax, and, as the
tension recedes into the postlude, probably the point of culmination, too.

Figure 2.27. “A-u!” postlude, analysis

!"#$%&'(

!"#$%&'(

67
 Figure 2.28. Schema of “A-u!” climax

In “A-u!” there is no real harmonic activity after the climax event; as though all
energy has been spent, the postlude takes place over what I call a “post-climactic pedal
point”—common in Rachmaninoff’s mature works, as subsequent analyses in the
dissertation will show.
In each of the examples presented so far, juxtaposition of a diatonic-syntactic
basis and recalcitrant chromaticism creates some sort of intense anxiety about the tonic.
In the Strauss and Mahler passages, the separation into harmonic layers, individual but
interactive, is texturally and registrally obvious. The Rachmaninoff examples are perhaps
not as obviously counter-traditional, nor are their harmonic layers quite as explicit, but
they are equally dramatic: insistence upon the tonic triad in unremitting chromatic
contexts creates moments of climactic roughness—quite the opposite of what one expects
from a statement of tonic. In the case of the Rachmaninoff Symphony No. 3 passage (see
again Figures 2.8 and 2.10), the result is a dramatic twist, a hyperdissonant accent, on
what was by 1936 utterly passé: the return of tonic at a sonata-form recapitulation.
While I believe that hyperdissonant distortion, in particular, is characteristic of
Rachmaninoff’s mature style (that is to say, of works composed in the late Russian and

68
exile periods), there is at least one clear precedent in the early Russian period: the climax
in the Elegy in E♭ minor, Op. 3, No. 1 (1892). As the analysis in Figure 2.29 shows, the
event shares important features with the “A-u!” climax, and might even be heard as a
prototype for it.

Figure 2.29. Elegy in E♭ minor, Op. 3, No. 1, analysis of climax

!"#$%&'() *"#$%&'()
$ # # !$" # # # # !$" # #
%&
#####
" ## #
!" #$
## # # #
! " """" # # $# # # #$"## # $##
+,)'-(
## #
'()*
## "#
# $### # ##

# ## # # # #
!! '()*
% "" " "
'()*
"" # ## #
# # $# " # ## # # # # #
# # "# # #
'()*

3 '''56
4 .',)'-(

5 777 3

& "
&B05.'C''D!
%+',-.
@')%AB*
$# $ # &# &# $# !"#$%&'()(*+(,
/0 &# $# $ # & # & # $# $ # &1 (
"" &# &# $# $# &# $# &#&# $#
! " " "" $# &# &# $# $# $# $# "# "# $# $#' "# "#
""$### ""$### & $## ""$### &$$### ""$### &$$### ""$###
% " " " &$$###
&$##
""" "# $## " # $#
$#
" # $# " # $# "#
$#
#
) )
""" 89:;<=>? $
&B05.'! "
E5+E$F
!"#$%&'() .E%*+(/

"# # "# # # $# #
&$ &!
" -.%/(01%2
! " """" # # # "## "##
# ## # ## # # #
## "## ## # # # # # # # $"#
# #
# #
$ ## ## # # # !! '()*
% """
""" # # # #
!

5 34'''56
&B05.'! "
23-43,356'78.3,)9+:;' ,&%G$(

69
 The piece is in ternary form (ABA). Its main melodic material can be understood
as emerging from a basic - - descent in measures 3–5. This provides the kernel for
the hyperdissonant distortion event at the climax. As shown in Figure 2.29, the B section
(a departure in Ratner’s sense) progresses through III to V, the latter of which hosts a
wedge progression that “should” prepare the return of tonic but goes to far, leading to the
climax event that begins in measure 80. At the climax, tonic E♭ is entangled with A
major, resulting in an octatonic-type binary reminiscent of the ones at the “A-u!” climax
in Figure 2.25. In this case, however, the melodic material does not follow suit: it is not
octatonic. E♭ emerges at measure 82, but as a major-minor seventh chord (again,
characteristic of octatonic organization)—not a minor triad. As shown in the boxes in the
A sections of Figure 2.28, this sonority functions as V7 of iv in the functionally normative
A sections of the piece. At the climax, hyperdissonance emancipates it from a functional
role.

Figure 2.30. Elegy, analysis of distorted cadential figure at climax

*:-/!@=693/>853= *.-/01234567789:9;/567;84;689/:;/<=6>:?
" " "
! # $ ""!% "
#
"
$
"" # # #
! " " "" $# $# %# %# "# $# "# "#
" ##
& """ $#
""" # # # $# " # $# " #
! " !
&
'()*+,$-

As Figure 2.30 shows, a basic cadential formula ( - - ) is contradicted by

extreme chromatic activity at the climax, applying a kind of expressive torque to the tonic
triad. As shown in Figure 2.29, the discharge of the hyperdissonance occurs during a one
measure cadenza-like passage that Rachmaninoff notated in small noteheads. Although it
is perhaps possible to hear an A♭ minor sonority in the cadenza, suggesting functional

70
resolution of the E♭7 chord, such an interpretation is weak. I prefer to hear the same kind
of “invisible” discharge as that featured in the “A-u!” and Symphony No. 3 passages
analyzed above. As the A section of the Elegy begins again in measure 87, the
extraordinary climax chord is restored to its original, ordinary function (V7 of iv).

A Parenthesis: Neutralization of Tonal Premises

As stated above, the third possible effect that hyperdissonance can have on a
functional tonal basis—neutralization of functional premises rather than exaggeration or
distortion of them—does not appear to have much analytic usefulness when it comes to
Rachmaninoff’s works. Tonal underpinnings are simply too strong. However,
recognizing that such neutralization is possible provides a way to differentiate
Rachmaninoff’s works from works that use similar or identical chromatic structures to
different ends. The music of Alexander Scriabin is a particularly good case in point.
As I demonstrate analytically in Chapter 4, the equal interval structures that many
authors have identified in Scriabin’s works are in and of themselves not unlike those in
Rachmaninoff’s mature works.51 However, the effect of the structures is very different
indeed. Simon Morrison describes octatonic and whole-tone materials (and the so-called
“mystic” chord that combines features of both) in Scriabin’s works as “inert acoustic
structures modeled on traditional harmonies but devoid of functionality.”52 Taruskin’s
view is similar: “Since it is harmonic progression that had always articulated the
structural rhythm of music, which is to say its sense of directed unfolding in time, a
music based on universal invariant harmonies becomes quite literally timeless, as well as
emotionally quiescent.”53 On the other hand, I have suggested, and will do so more
forcefully in Chapters 3 and 4, that equal interval structures in Rachmaninoff’s works are
associated with intensification and climax. According to the above authors, in Scriabin’s
late works, equal-interval structures, though modeled on “traditional harmonies”
(translation: they are tertian sonorities) do not engage tonal functions at all. Tonal

51
See again notes in Chapter 1 on Rachmaninoff’s performances of Skryabin’s works in the 1910s. The
two composers were classmates.
52
Simon Morrison, “Skryabin and the Impossible,” Journal of the American Musicological Society (1998):
315-316. See also Reise, “Late Skriabin: Some Principles Behind the Style.”
53
Taruskin, Defining Russia Musically, 348-349.

71
premises are therefore neutralized by symmetrical pitch structures in Scriabin’s works,
while they are exaggerated or distorted by the same in Rachmaninoff’s. (Rachmaninoff’s
and Scriabin’s equal-interval structures are compared in more detail in Chapter 4.)

Conclusion

“Practically anything in music can be labelled passing-note or appoggiatura,”

Calvocoressi warns us.54 The interpretive framework Calvocoressi develops for
Mussorgsky’s music to some degree resonates with the one I am proposing for
Rachmaninoff’s, insofar as both proceed from the premise that an interaction of
functional tonal structures and special chromatic and modal structures is significant.
Calvocoressi suggests that expansion of tonal premises is only part of that significance:

[Mussorgsky’s] music embodies the genre omnitonique foretold by Liszt, but in a

form depending upon the fundamental properties of Mussorgsky’s ideas, melodic and
harmonic, rather than upon the artifices by means of which his Western contemporaries
were extending the boundaries of the major-minor system.55

In a similar vein, Jim Samson has suggested that “there is…a distinction between
the expansion of classical tonality from within and its modification from without. In the
music of the Russian nationalist composers the modification of tonality was particularly
thoroughgoing…The remarkable flowering of Russian music in the nineteenth century
was characterized by a fascinating dialogue between indigenous traits–often the product
of a distinctive folksong heritage–and Western traditions which were alternately
embraced and rejected.”56 That is to say, the means by which Russian composers
developed new melodic and harmonic resources in the late nineteenth and early twentieth
centuries were only in part related to the “internal” expansion of syntax that characterizes
late Romantic music in the West. For example, numerous octatonic passages occur in the
works of Chopin and Liszt as a result of internal tonal expansion. In a great deal of
54
Michel Dimitri Calvocoressi, Mussorgsky (London: Rockliff, 1956), 238.
55
Ibid., 257.
56
Jim Samson, Music in Transition: A Study of Tonal Expansion and Atonality, 1900–1920 (New York:
Norton, 1977), 9–10.

72
Russian music from the late nineteenth and early twentieth centuries, on the other hand,
octatonicism (along with other “fantastic” chromatic structures) is treated as a source—a
topic, something conceptually marked, expressively accented.
While, as the analyses in this chapter all demonstrate, the insights afforded by
linear analysis are invaluable for the interpretation of tonal music from any era, such
methods are not necessarily well configured to capture certain kinds of melodic-harmonic
configurations—especially configurations that involve idioms which may have topical
associations or special kinds of motivic significance, or when stratification and
hyperdissonance may be involved.57 As melodic-harmonic configurations gain in
complexity and explicit or implicit deformational procedures become a central point in
the undertaking, analysis emerges from considerations of foreground, middleground, and
background levels into a grayer area that recalls Roland Barthes’s familiar claim: “to
interpret a text is not to apply meaning to it, but on the contrary to appreciate the plurality
that constitutes it.”58
Put simply, conventional linear analysis methods and functional analysis methods
require supplementation in the study of Postromantic music. It is my hope that the
interpretive framework sketched in this chapter has suggested the potential benefits of an
approach that incorporates the abnormal as abnormal, that allows a “surface-level”
deformation or anomaly to in fact be deformational or anomalous. Identification of a
linear or a functional prototype is therefore only a starting point for interpretation.
Figures 2.31 through 2.34, collectively an analysis of Rachmaninoff’s song “Daisies,”
Op. 38, No. 3 (1916), demonstrate.
Figure 2.31 contains the first four measures of the song. Three features stand out:
a plagal orientation, various statements of the melodic figure E♭-C or F- E♭-C (marked y
in Figure 2.31), and a striking harmonic relation that is extracted into a box on the figure.
It is tempting to interpret the harmonic event—a “harmonic motive”—in exclusively
linear terms. Motion from the unstable first chord into the second is characterized by
semitones in all voices. Resolution of the leading tone (E → F) is amplified by
simultaneous resolution of three additional, “artificial” leading tones (D♭ → C, B → C,

57
I have already suggested that by reducing octatonicism to a simple prolongational technique in
Rachmaninoff’s works, Cunningham eliminates an important rhetorical consideration.
58
Roland Barthes, S/Z, trans. Richard Miller (New York: Hill and Wang, 1974), 58.

73
and A♭ → A♮). The spelling of the chromatic chord, in fact, makes the linear basis
entirely clear: Rachmaninoff has chosen a non-tertian spelling of what sounds like a D♭
minor seventh chord, thereby emphasizing the quadruple semitonal voice-leading
resolution.

Figure 2.31. “Daisies,” Op. 38, No. 3, analysis of mm. 1–4

#$%&'()*+&',)-.

!
# %$ $

# % &&$$$ %$$$
!

]
&#'("

) $ $ $" $+ $$ & $ $ $ $ $
%(($" &(
!"#$%
$
!
# &" ' % $"
$$
%$
$
* * &*$ $* $ $ $
* *) *
! !
# & "! ,
" !"#$%
$ $ $
] $ $
$ $ &$ $ $ $ $ $ $ &$ $ $ $ $ $ &$ $ $ $ $ $
$
[
[

$ % $ $ &$ $ $

- &! $ $
"
&'() *

However, this harmonic formula, which if considered without regard to spelling

involves root motion by chromatic major third, has a deep structural and motivic
significance in the song. This is made clear at the climax (measures 1–18). Measures 17–
18 of the song are clearly climactic: the piano part achieves its highest and lowest
registers (the low D♭ represents a plunge more than two octaves below anything
previously heard in the song), and the voice achieves its highest dynamic indication, f,
and its highest note upon the word gotov’ (“prepare”). An overview of the climax in the
context of the entire song is shown in Figure 2.32. As shown in Figure 2.32, the climax
results from a gradual ascent, in stages (marked 1a, 2, 3, and 4 on the figure) in the vocal

74
line, which is rendered in simplified form. The climax occurs on D♭ major—previously
the unstable chord in the harmonic motive; and it is setup up by an F7 chord. Figure 2.32
reveals a kind reciprocal relationship between F major (the overall tonic of the song) and
D♭. In other words, the harmonic motive can be reversed: either chord can be the basis of
a seventh chord that resolves into the other chord. The chromatic major-third relationship
is motivic, and, because it is exploited on the largest scale of the song’s harmonic
structure, and provides the source of a climax event, it is structural. The potential
reciprocality of the chromatic major-third relation heard at the onset of the song is
actualized at the climax.
Figures 2.33 and 2.34, in conjunction with the annotations of measures 1–4 in
Figure 2.31, show how the melodic material in the song reflects the reciprocal nature of
the harmonic motive. The melodic figure labeled y in Figure 2.30 establishes pitch class
E♭ as a significant melodic feature; as shown in Figure 2.32(a), this combines with
frequently sounded B♮ (which is also contained in the harmonic motive) to suggest the F
acoustic scale, which contains an F Lydian pentachord. In the climactic D♭ region of the
song, melodic tone G♮ is emphasized. As shown in Figure 2.32(b), this creates a Lydian
pentachord. Figure 2.33 shows the relationship between the F and D♭ Lydian pentachords
and the two chords of the harmonic motive, and shows that, collectively, all of these
features suggest a symmetrical background source: WT1, or the whole-tone scale
containing D♭. WT1 and the chromatic major third relationship in the harmonic motive
both suggest equal division of the octave, which, as discussed in Chapters 3 and 4, is
strongly associated with intensification and climax in the works studied.
In view of all of these features, the F7 chord leading into the climax is in my view
not a passing event, but rather represents a kind of tonal involution. The main harmonic
motive of the piece is reversed in its directionality—its “charge”—and in being reversed
is made to serve not a composed-out “tonical” structure but a chromatic, non-tonical
background harmonic source. The large-scale design and the melodic material of the
work therefore reflect, or perhaps rely on, the harmonic motive.

75
 Figure 2.32. “Daisies,” analytic overview of climax

climax
vocal line, simplified
4
3
2 1B
1A

    

    
           
        
  postlude
      
   
post-climactic pedal
  
 
(SD) !5 §6 !5 §6 !7 8ba (SD) DT
4 4 3 Part 1 (28-31):

76
p authentic harmonic motive echoes
harmonic motive (D) (SD) harmonic motive cadence
Part 2 (32-37):
harmonic motive plagal-oriented closure
in climax registers
F major D F major

11 14 18 24 28

reversal of harmonic motive

tonic charged

departure return
F major
 Figure 2.33. Scales and pentachords in “Daisies”

!"#$%&'"(%"!)*

"#
( !" $#
!4
# # # #
( !" # # # #
1(2!3#0(%"!)*

)+,'!-(.*-&!"/#0,

( !" $#
54

( ! " "# "# # "# "# "# # "#

Figure 2.34. Relationship between the harmonic motive and the scales in “Daisies”

9
%

"# # !$"# " #

# $# # "#
$#
!
+7! !" # #

&"
9

'()*+,-+./012*.!34,2+.5/6)1
! &"
"# "# # $##
% "#
87! ! " # # # $## #

"#$

77
 The climax in “Daisies” is neither as gritty as the “A-u!” climax nor as grim as the
Symphony No. 3 climax; it is, on the contrary, a kind of expressive breakthrough at a
point of remove, not a collapse at the point of return. The projection of a charged,
unstable tonic at the point of remove, recalling the climax in the development of
Beethoven’s “Eroica” (see discussion of that work in relation to the Symphony No. 3 first
movement climax, above), brings the chromatic distance traveled—the dislocation, the
defamiliarization, or, to be precise, the exaggeration—into expressive focus. Certainly
there is no diatonic structure of comparable significance; and, as in the “A-u!” climax,
everything following the climax unfolds over a post-climactic pedal point, as though
emphasizing that the chromatic climax event is the event in the song.

* * *

It is perhaps typical of Rachmaninoff that “extraordinary” chromaticism is in the

end somehow subsumed under some “ordinary” tonal procedure—if this failed to happen,
he would be not a Postromantic composer but a modernist one. Yet the moments when
the “extraordinary” overwhelms the “ordinary” largely define the music’s expressive
content. In considering tonal design in the usual sense, the emphasis may be on harmonic
norms; but emphasis on the effects of the abnorms reveals more dynamic designs that are
reminiscent of Wallace Berry’s “intensity curves.”59 Lee A. Rothfarb has traced Berry’s
idea back to Ernst Kurth, who was working around the same time Rachmaninoff.60
Kurth’s theories, with their emphasis on disruptive, even destructive forces, perhaps
reflect the tonal music of his own day better than generally recognized. In the examples
presented in this chapter, tension between components of a variegated harmonic language
shapes larger structural and expressive processes, and, to borrow a phrase from Kurth,
“contradictions are thus transformed into an overpowering accord.”61

59
Berry, Structural Functions, 4.
60
Rothfarb, Ernst Kurth: Selected Writings, 33.
61
Ibid., 191.

78
 Chapter 3

Overview of Harmonic Structures and Their Rhetorical Associations

In Chapters 4 and 5 of the dissertation, the special chromatic and modal

components of Rachmaninoff’s mature harmonic language are described in detail, and
their technical characteristics are defined in order to prepare the longer analyses in
Chapter 6. The present chapter modulates, as it were, from the abstraction of Chapters 1
and 2 to the specificity of Chapters 4 and 5 by considering the rhetorical and expressive
characteristics that special chromatic and modal structures have in the context of the
interpretive model developed in Chapters 1 and 2.
Figure 3.1 provides an overview of the compound harmonic language described in
Chapters 4 and 5. In addition to listing the main melodic and harmonic components found
in the works studied and a number of the most important techniques through which they
are articulated, Figure 3.1 lists the salient characteristics of each (bulleted and italicized)
to make clear some of their essential differences. Some of the chromatic and modal types
listed in the figure, e.g. the ordinary church modes, will be familiar to a reader with even
limited knowledge of music theory and therefore require little discussion. Others, e.g.
peremennost, may be unfamiliar even to a readers conversant in the literature, and will
therefore require more extensive theoretical description in Chapters 4 and 5.
The present chapter details the rhetorical associations that special chromatic and
modal structures have in Rachmaninoff’s mature style. The goal is to demonstrate in
analysis that Rachmaninoff used different kinds of pitch organization for different
rhetorical purposes, and that interpretation of form and climax events in the works
analyzed benefits substantially from awareness of these rhetorical associations, which
depends in turn on not reducing the special chromatic and modal structures to diatonic
prototypes.

79
 Figure 3.1. Components of Rachmaninoff’s mature harmonic language

!"#$%&'#()*+'#()*!,(-./',0
*
!"#$%&'"'(%#)&'&#*+,#-.&/&#*
!"0&0#1")(/-#&$%,"234"5"4"5"6"7"84"5"69
!"#(%:(%'*"#$%(, !"#$%$&'(%()#*
!"'/(-)":(;-)#<)(5)(#<)%"=$:(/, !"#$%$&'$($&#)*&%+*&%,-.
!">$-/5$)&(%#(:

1!(#%(2%&$3*45,'-(%&$*6%,"$%",.2 >'?()*6%,"$%",.2
728--.%,&$()9.:"();&#%.,<()=

8,';()(=(%%$,# '9'2$1#")$&'"*&$//1#$")
8,'"&#1#")$& 7,'.(A1#")$&
7,'%(>-"$2$"!*
$+9B9C'#(5#$1)'&3&/(*9?1$5* $:9H'#(5#$1)'&3&/(*9?1$5* +,'"52$)153'&.65&.'!"2(*
DE'#(&.)$F6(* 5(&$?5"&1/'1551)<(!()#
:,';.53<$1)'"5<1)$=1#$")'>*?(&$1/@
G+9B9C'!(/"2$&'#(&.)$F6(* G:9H'!(/"2$&'#(&.)$F6(*

!"'-%";)$./(=-#&?("#$%&'"'(%#)&'&#*
+,'-."/(0#")('1)2'.345$2'*#56&#65(* !"/&=&#":&-#$%&'"#(%:(%'*"#$%(,

!";)$./(=-#&?("#$%&'"'(%#)&'&#*
!":(%*"0&0#1")(/-#&$%,@"'$=;/&'-#(">$-/5$)&(%#-#&$%
!"2=-*"-//$A",$=("#(%:(%'*"#$%(,9
!"<,<-//*".-,(:"$%")$#-#&$%"-%:"$,'&//-#&$%
!"'-%".("(B#(%:(:".(*$%:"C;<)(D"'$//('#&$%".-,&,

As suggested in Chapters 1 and 2, Rachmaninoff’s music is not unique in

combining diatonic-functional syntax and special chromatic and modal idioms. Research
on late Romantic, Postromantic, and early modernist works written by composers of
many nationalities has suggested various kinds of compound syntax. However, as Jim
Samson has pointed out, Russian music from the late nineteenth and early twentieth

80
centuries is characterized by such combinations to an especially large degree, and
scholars have not been reluctant to incorporate this into their analyses.1
James Baker’s work on Scriabin and Michel Dimitri Calvocoressi’s description of
functional, modal and chromatic structures in Mussorgsky’s works have been cited in
Chapters 1 and 2. Simon Morrison has written about Rimsky-Korsakov’s Sadko in
similar terms.2 Morrison recognizes a kind of triple syntax, noting "the equivalency of the
‘diatonic,’ ‘modal,’ and ‘octatonic’ passages. No single musical syntax dominates."3
Morrison suggests that the dialogue between special chromatic music (in this case,
octatonic) and modal music is not just structural in Sadko; it is meaningful. Octatonic
music is “supernatural,” and associated with the character Volkhova (she is a fantasy, a
projection of the imagination), while modal music is “natural,” and associated with the
title character, Sadko. “That her "supernatural" (octatonic) music derives from his
"natural" (modal) music signals that she is as much an aural as a visual object of
masculine conjuring, a product, in short, of synaesthesia.”4 Differentiation of pitch
structures is therefore crucial in Morrison’s interpretation of the opera.
I argue that differentiation is similarly useful in the interpretation of
Rachmaninoff’s mature works, though in less precise, less “plotted” ways than in
Morrison’s interpretation of Sadko. Analysis of a large number of works suggests that
different kinds of pitch structures in Rachmaninoff’s works are generally associated with
different basic rhetorical functions. Figure 3.2 lists the most important special chromatic
and modal components and their basic rhetorical associations. As the table shows,
recognition of rhetorical characteristics makes possible the identification of probable
locations of special chromatic and modal structures in relation to the underlying
functional tonal framework.

1
Samson, Music in Transition, 9-12.
2
Simon Morrison, “The Semiotics of Symmetry, or Rimsky-Korsakov’s Operatic History
Lesson,” Cambridge Opera Journal 13 (2001): 261-293.
3
Ibid., 291.
4
Ibid., 269.

81
 Figure 3.2. Harmonic components and their rhetorical associations

!"#$ %&&'()*+)',& -'&+./'00',.1'(*+)',&

87#/!.,'/-0%.'/-0%'$2-/,9-.,'/:
8*.$#!.#$-0%7$-&)@'$A: 84)$3-*,3):
82)/)$,!%;0,/)-$<%!"$'&-.,!,*&:

,/.$'(#!.'$1
!"#$!"%&'()*
,/,.,-.,/2
!"#"$"%%&'(%+%(,-.'/,!%'*!,00-.,'/ 6)2,//,/2%-/(%)/(%'7%)4,*'()+-$!
(,2$)**,3)
%")*%,(,'&*
4'*.5!0,&-!.,!

B"$12,-/ '!"+,*-.+*'"

,/.)/*,71,/2
=7-/.-*.,!>%!"$'&-.,!,*&
()*.-6,0,9,/2 ,/.'%",2"4',/.+!0,&-C
)?#-05,/.)$3-0%*.$#!.#$)*
!0,&-!.,!

As shown in Figure 3.2, functional tonal organization (almost invariably

decorated by generic linear chromatic events, which should not be confused with special,
“fantastic” chromatic structures) provides the structural framework for all
Rachmaninoff’s works, even in the most extreme chromatic and/or modal contexts. In
most cases, modal structures—including the familiar church modes, peremennost
techniques, and the related nega idiom, all of which are discussed in Chapter 5—appear
in introductory passages, serve to initiate thematic exposition or to close a section (after a
highpoint or climax), or, occasionally, appear as digressions inside longer sections. In
most cases, “fantastic” chromatic structures—that is to say, structures that can be
understood as involving equal division of the octave, possibly extended to include
additional tones—appear in passages that intensify, destabilize, and lead to climaxes on
various scales. (The Phrygian mode is a special case treated at length in Chapter 5.)
Figure 3.3 shows the probable locations of these structures in relation to the
prototypical, Ratnerian tension arc developed in Chapter 2. The figure also identifies a
few additional musical characteristics toward the goal of enhancing the associations
being made between specific pitch structures and expressive trajectories: processes of
intensification generally engage several musical parameters simultaneously, including

82
harmonic structure, dynamics, texture, rhythm, and register; and, as demonstrated in
several analyses in this and later chapters, pedal points following climaxes are a common
feature in Rachmaninoff’s works, and may coincide with post-climactic modal structures.

Figure 3.3. Probable locations of special chromatic and modal structures

!"#!$%"&'
!" 23
1*4
)5(
()"*+, ')
*()
!2+
#&
5!2
3(%
*

!"#$"%!&!'(#!)"
!!!!!"!#$%&'()
!!!!!"!*+,*-.+
!!!!!"!./$*/'
!!!!!"!.+0(1*+.

#!$%&'()"*+)*",( ,0*%&1-2),"3%&'()"*+)*",( #!$%&'()"*+)*",(

&-.,&/ &-.,&/ &-.,&/

Figures 3.2 and 3.3 are not meant to present an iron-clad formula. Rather, they are
based on the observation that, in the works analyzed, clear modal organization is more
likely toward the beginning or end of an episode or a section or a work, while equal-
interval chromatic structures are more likely in passages that intensify and at highpoints
and climaxes. This by no means implies that modal organization is found at the beginning
and at the end of every section of music in the works analyzed, nor that equal-interval
chromaticism is found at every intense moment or climax. Many passages and even entire
works are composed using primarily or only functional tonal methods. But when modal
and/or “fantastic” methods are emphasized to a substantial degree, the result is likely to
follow the broad outlines given in Figures 3.2 and 3.3.
Four short analyses demonstrate the point. In these analyses, I will of necessity
anticipate some of the technical points made more fully in Chapters 4 and 5, and I will
use terminology and labels from those chapters in the interest of consistency throughout

83
the document, even at the risk that some details in the present analyses may not yet be
fully understood. However, the importance of the main point these analyses are intended
to make—that clear rhetorical differentiation of special chromatic and modal structures
exists and is significant—supersedes any organizational drawback.

Four Short Analyses

“From the Gospel of St. John,” WoO (1915)

Figure 3.4 is an analysis of the posthumously published song “From the Gospel of
St. John,” composed in the late Russian period. The song is little more than a fragment. It
is only thirteen measures long (lasting about a minute and a half in performance), and
lacks any real resolution at the end; it ends as it begins—on a first-inversion A major
triad. The song has no key signature, but A major serves as the tonic, melodically and
harmonically (never, however, in root position). There is a limited amount of functional
tonal activity, which, together with the A major tonic, serves as a framework for
interpretation.
As Figure 3.4 shows, the piano prelude suggests the Lydian mode, with the
exception of F♮, which may be interpreted as a neighbor tone. With the entry of the voice
in measure 3, the mode is disrupted, and linear chromatic activity prepares the arrival of
V6 in measure 5. (This motion from I6 to V6 represents the extent of functional tonal
progression in the song.) As Figure 3.4 shows, the complex passage that follows (mm.6-
8) may be understood as an ornamented version of an octatonic chord cycle, as shown
more clearly in Figure 3.5. Octatonic structures are formally discussed in Chapter 4, and
the full details of Figures 3.4 and 3.5 may not be clear until then; however, some analytic
observations pertinent to this particular song are possible now.
As Figure 3.4 shows, the octatonic cycle in “From the Gospel of St. John”
involves minor third root relations. It emerges from the initial A major triad, proceeds
through seventh and ninth chords on roots F♯, E♭ (or D♯), and C♮, and is ornamented by
passing and neighbor tones that to some degree obfuscate the octatonic basis.
Simultaneously, two different diminished seventh chords are articulated in the passage

84
(labeled º7: [0,3,6,9] and º7: [1,4,7,10] in the figure); these diminished seventh chords
combine to form the complete OCT(0,1) collection. The vocal part presents a tonic triad
“frame,” which, with the addition of a few chromatic tones, represents a composing-
out—in inversion—of the motive labeled “x” in the piano prelude, as shown in Figure
3.6.
As Figure 3.4 shows, the song’s climax occurs as the octatonic cycle reaches its
limit (the Cb9 chord in measure 8) and then cycles back to I(6). The dynamic marked is ff,
and there is a clear acceleration into and through the climax event. The appearance of G♯
at the end of measure 8 (marked “*” in Figure 3.4) represents the dissolution of
octatonicism (G♯ does not belong to the OCT(0,1) collection) and a return of Lydian
organization, now in a post-climactic capacity and above a post-climactic pedal point. A
Lydian and OCT(0,1) are closely related. As Figure 3.7 shows, they share many tones—
most significantly for present purposes, A♮ and D♯/E♭. D♯ characterizes the Lydian mode
on A, and participates as a chord root in the climactic OCT(0,1) cycle; and both the Lydian
and octatonic collections contain the A major tonic triad. In the absence of genuine tonal
progression, the “modulation” from Lydian A to OCT(0,1) and back to Lydian A is
perhaps the most important factor in the song’s overall harmonic design.
In “From the Gospel of St. John,” octatonicism is associated quite explicitly with
intensification and climax, while Lydian organization fulfills initiating and post-climactic
rhetorical functions. Furthermore, the tension that exists between the large-scale tonic
frame in the voice and the octatonic structure in the piano has taken the place of
functional tonal progression as the main event of harmonic interest, and serves as grist for
the climax. Figure 3.8 summarizes these points in a diagram.

85
 Figure 3.4. “From the Gospel of St. John,” WoO, analysis

climax of song

3 4 5 6 7 8 9

voice tonic frame

     
            
 
p ff mf
(x’)
acceleration through climax
11 12 13

º7: [1,4,7,10]
*
piano   

86

  
  
 
  
 
                
 
  
     
 

   
 
    º7           
 
   
   

     
                  
 
  
    
            
x

octatonic scale
º7: [0,3,6,9]

pedal point

roots related by minor thirds

A Lydian OCT(0,1) cycle A Lydian

introductory intensifying, climactic post-climactic

 Figure 3.5. “From the Gospel of St. John,” overview

./01*)2345- ! " # $ %

! " #" " #" $" "%"#" " #" " #" " #" " #" #" "
#" "

#"" " # $# """ # # # """ & & $ """ & & & """ # $# """ # $"""
! " #"" $" #" " % # "" #" $$ " " &" &$ " " #"
#" #"
!"#$%&'()*+*,-

6)7+8140 6)7+8140

Figure 3.6. “From the Gospel of St. John,” motivic analysis

!"

" #" " # " "

!
! " #" #" !
! "

"#$%&'!()*!+,-.)/ 0-12.-21,!&3!4&.$5!5#%,'!(()678

Figure 3.7. The relationship between A Lydian and OCT(0,1)

!"#$%&'"(')$*

! " #" #" " #" #" " $ " %" #" #" " #" %" "
" "
!+,
#
!"#$%&'( )*+,-./0

87
 Figure 3.8. “From the Gospel of St. John,” diagram of climax

!"#$%&#'("!)""#'!&#%*'+,-."'-#/'012'*3*4"

!"#$%&'()*+&$#&,"$%+
45(7)8

!!"#"$" !!##"+$ !!"#"$"

!"%&'()* ,-./0123"4&456 !"%&'()*

The Bells, Op. 35 (1913), First Movement Climax

Similar associations between modal structures and initiating or post-climactic

rhetorical functions, and between “fantastic” chromatic structures and intensifying or
climactic rhetorical functions obtain even in works or passages whose overall tonal
designs are more conventional than “From the Gospel of St. John.” Figure 3.9 is an
analytic overview of the main climax in the first movement of The Bells, Op. 35 (1913).
In this movement, functional tonal organization is considerably more prominent than it is
in “From the Gospel of St. John”; but similar chromatic and modal structures may be
heard, and the same basic rhetorical associations exist.

88
 Figure 3.9 The Bells, Op. 35, i, analysis of climax (mm. 106–162)

89
 The movement is in ternary form, and its tonic is A♭ major. Figure 3.9 shows, in a
simplified form, the B section and then, in more detail, the A’ section, followed by a
portion of the coda. Rehearsal numbers and measure numbers are provided. Marked “n”
throughout Figure 3.9 is a neighbor figure of significance throughout the movement. In
measure 1 of the movement (not shown in the diagram), the upper neighbor of A♭ and the
upper neighbor of E♭ are activated simultaneously; in Figure 3.9, neighbor figures on A♭
and E♭ occur several times. Marked “x” throughout Figure 3.9 are various forms of a
separate motivic figure that plays a more direct role in the climax event. Motive x
saturates both the A section (not shown) and the climactic A’ section.
As Figure 3.9 shows, the climax event is framed by functional tonal activity in the
key of A♭ major, but itself has a strong octatonic basis. OCT(2,3) is articulated by seventh
chords on “nodes” D, B, and G♯. As shown in Figure 2.10, The seeds for the climactic
octatonic structure are planted in the first A section: the seventh chords on C♭ (B) and D
in measures 33, 35, and 38 are the first substantive chromatic events heard in the work. In
this early passage, the C♭ and D sonorities are combined with A♭ major tonic elements in
a complex layered structure, and an A♭ seventh chord is heard as well, heightening the
suggestion of octatonic organization. This early passage sets up the OCT(2,3) events
exploited in the climax.
As Figure 3.9 shows, several non-collection tones may be heard in the climactic
octatonic passage. These may be interpreted as passing events within the octatonic
structure, and do not significantly disrupt the overall octatonic structure of the measures.
Furthermore, the non-octatonic events all involve the same type of sonority as the
structural octatonic harmonies—major-minor seventh chord—and they occur when the
dynamic is reduced down to p before another buildup to ff (into measure 152) that is
associated once more with explicitly octatonic organization.

90
 Figure 3.10. The Bells, i, analysis of mm. 32–39

32

35

38

In Figure 3.9, as in “From the Gospel of St. John,” two different diminished
seventh chords are simultaneously articulated in different registers; these combine to
form OCT(2,3). One of the diminished seventh chords (º7: [0,3,6,9]) emerges from the
functionally significant E♭ in the melody—more specifically, it emerges from chromatic
inflection of motive x. Diatonic x in measure 139 spans a perfect fourth from primary
note E♭ down to B♭. With OCT(2,3), diatonic x is inflected to x’, which spans a tritone and
is therefore integrated into the octatonic structure and º7: [0,3,6,9]. After rehearsal 23, x’
is simplified to just the tritone—in this context, an especially raw signifier of
octatonicism.
In the A’ section, the octatonic structure is associated with a process of
intensification, and directly sets up the movement’s climax. By sharp contrast, the coda,
which commences in measure 155, begins with a passage in the Lydian mode. The D♮

91
from OCT(2,3) (which, as described above, is set up early in the movement) is thus
retained in the coda, but the rhetorical effect is different. Here, modal organization, as in
“From the Gospel of St. John,” is post-climactic; and again there is a post-climactic tonic
pedal point. In the coda, the falling melodic contours (heard most clearly in the sequential
figure marked “*” in Figure 3.9), supported by a long diminuendo, contrast sharply with
the rising tessitura and crescendo of the octatonic passage.

Etude-Tableaux in E♭ minor, Op. 39, No. 5 (1917), First Section

The previous two analyses demonstrate the utility of Figures 3.2 and 3.3 in the
analysis of large-scale climax events. A third analysis demonstrates that similar structures
and associations can apply even in passages whose climaxes are of local rather than
global significance. Figures 3.11 and 3.12 present in two parts an analytic overview of
most of the first section (section A of a ternary form) of the well-known Etude-Tableaux
in E♭ minor, Op. 39, No. 5.

92
 3.11. Etude-Tableaux in E♭ minor, Op. 39, No. 5, mm. 1–12, analysis

93
 3.12. Etude-Tableaux in E♭ minor, mm. 12–22, analysis

94
 As shown in Figure 3.11, phrase 1 of the etude is organized around a clear T–PD–
D–T progression over a tonic pedal point. (The pedal point also includes the upper fifth,
B♭, enriching the texture and adding dissonance.) Pitch classes of the tonic triad serve as
a frame for the melody; this partly explains the V13 in measure 11.5 Phrase 2 (mm. 12–
22), shown in Figure 3.12, begins with a reharmonization of phrase 1 melodic material. In
the reharmonization, the melodic tones are supported with different diatonically-related
triads in a manner suggesting the peremennost idiom described in Chapter 5: G♭ major,
III, and D♭ major, VII or III of B♭ minor, the goal of the passage. The modal (or, in this
particular case, perhaps only quasi-modal) structure initiates the phrase, supporting the
general associations outlined in Figure 3.3. The new diatonic chords in phrase 2 are
presented using a strategy of suggestion followed by confirmation. G♭ major is suggested
in measure 13, but under a dissonant melodic tone (because the melodic structure is the
same as that in phrase 1), and confirmed as a local goal in measure 16 (marked “√” in
Figure 3.12). Because the melody in measures 1–4 and 6–7 is set sequentially, a similar
strategy of suggestion-confirmation is suggested for the D♭ major triad, as shown in
Figure 3.12, but on a larger scale owing to the interference of an intensifying octatonic
passage that begins in measure 18.
As shown in Figure 3.12, the octatonic passage interrupts the (reharmonized)
tonic frame and functional tonal progression, leading to a local climax event. The
octatonic passage shares many superficial characteristics with the climax-inducing
octatonic passages in “From the Gospel of St. John” and the first movement of The Bells:
heightening melodic tessitura, increasing dynamic level, faster rhythms, and so on. It also
features the same basic octatonic structure: two different diminished seventh chords
articulated in different registers, and emphasis on seventh chords whose roots are related
by minor third. To these the passage adds a third, even more explicitly octatonic
technique: transposition of a melodic segment, along with its harmonic support, up by
minor third (T3).6 A number of non-collection tones can be understood as passing and
neighbor events within the octatonic structure.

5
Cunningham recognizes blurring of tonal functions as a significant feature in Rachmaninoff’s
style, calling it “hybrid function.” See Cunningham, “Harmonic Prolongation,” 99-112.
6
Octatonic structures are treated in more detail in Chapter 4.

95
Rhapsody on a Theme by Paganini, Op. 43, Measures 1–9

Even very brief passages on occasion display the rhetorical functions identified in
Figure 3.2 and Figure 3.3. Figure 3.13 is an analytic reduction of the first nine measures
of the Rhapsody on a Theme by Paganini. These measures are labeled “Introduction.” In
the passage, a tonic frame (referable to the initial motive of Paganini’s theme) is
decorated by a neighbor figure (E♮-F♮-E♮, which, as discussed in Chapter 6, comes to
climactic prominence later in the work). The tonic frame is sustained throughout the
passage. At the same time, an ascending OCT(0,1) scale in the bass supports a clear
octatonic seventh chord cycle through roots F♯, A♮, C♮, and E♭, which are boxed in
Figure 3.13; the chord “nodes” are connected by non-OCT(0,1) sonorities that provide
harmonic support for the motivic neighbor figure while maintaining smooth voice-
leading throughout the passage.
The passage begins fairly loudly (f); nevertheless, a crescendo is indicated before
the sf in measure 7, so that the octatonic structure is once again associated with a process
of intensification. The rhetorical association is maintained even in this brief introductory
flourish. In a manner very similar to that of “From the Gospel of St. John,” which in
some ways anticipates the structure of the Rhapsody passage, the octatonic structure
coexists with a tonic frame. The moment of maximal tension between the two
structures—one “structuring,” referable to the tonic triad, and proceeding into a
functional tonal progression (V7 – i), the other intensifying, involving equal division of
the octave, and referable to Russian “fantastic” chromatic techniques—is the occasion of
a climax event—or, as seems more appropriate for this very brief passage, a “highpoint.”
Although the passage differs from the other three passages analyzed in this chapter in that
no modal structures occur in the passage, even in this very limited context, the basic
rhetorical premises suggested in Figures 3.2 and 3.3 may be heard.

96
 Figure 3.13. Rhapsody on a Theme by Paganini, Op. 43, Introduction (mm. 1–9), analysis

hyperdissonant highpoint

sf
p

" " !#" "

# " " "" " " "" " "
f tonic frame
" !#"" " "
" " "
! " $" $" "
"" " "

!
motivic neighbor figure
"
!
" "

""" $ """ & """ " $ "" "

& """ #$ ""
#" " $ " & "" # "" $ ""
% $ "" $& "" "
$
"" " $" " &"
" &" " $" " "$ " " #" '
$" "

OCT(0,1)
OCT(0,1) scale
chord cycle

(i) PD V7 i

The analysis in Figure 3.13 suggests three additional analytic points:

1. In the passage, an octatonic structure is itself the subject of contrapuntal

elaboration. This demonstrates that octatonic idioms are well-defined in
Rachmaninoff’s works. Even if equal-interval structures in the tonal repertory
have their origins in elaboration of common-practice tonal structures, they
demonstrate a degree of self-sufficiency in works from the late nineteenth and
early twentieth centuries. This observation conflicts with Cunningham’s
interpretation of equal-interval structures in Rachmaninoff’s works.7

2. There is a clear change in functional orientation over the course of the octatonic
cycle— from tonic function to predominant function. The passage is therefore not
strictly prolongational.8

7
See again the discussion of Cunningham’s methodology in Chapter 1.
8
This assumes that “predominant” is a recognized tonal function. As explained in Chapter 2, I
regard intensification of the predominant stage of syntax to be a significant component of late
Romantic and Postromantic style.

97
 3. Neither the OCT(0,1) structure nor the tonal interpretation alone can account for the
passage’s effect; it is necessary to recognize the conflict that results from the
superimposition of the tonic frame and the octatonic structure.

As discussed in more detail in the longer analysis of the Rhapsody in Chapter 6 of

the dissertation, the passage in Figure 3.13 has a larger significance in the composition.
The rhetorical procedure exposed in the passage—statement of the tonic, simultaneous
statement of a clear octatonic structure, and exploitation of a tension that exists between
them—is the subject of a much larger climax in Variation XXII, involving precisely the
same pitch structures as Figure 3.13.9

Conclusion

Passages that contradict the rhetorical associations and probable locations

identified in Figures 3.2 and 3.3 of course exist: the theory does not prescribe. Exceptions
to the general rhetorical associations/locations, are, however, usually strongly marked.
Climaxes at which modal structures are emphasized rather than chromatic structures, do
occur—I take them to be special events.10 Other passages contradict the general rhetorical
associations because some modal or special chromatic idiom has especially strong
motivic significance in a work and is therefore found in a greater variety of contexts than
suggested in Figures 3.2 and 3.3. Some passages contradict the basic rhetorical
associations and locations described in this chapter on account of special expressive
circumstances—particularly potent is the effect of a passage that begins intensely or even

9
See the analysis of the Rhapsody in Chapter 6 of the dissertation, with special focus on rehearsal
66–68 in the work.
10
Examples of climaxes featuring strongly articulated modal structures may be found in the
Prelude in B minor, Op. 32 No. 10 (1910); the second movement of the Piano Sonata No. 2, Op.
36 (1913), which is analyzed in Chapter 5; the Etude-tableaux in C minor, Op. 39 No. 7 (1917).
All of these climaxes involve peremennost-derived “diatonic stacks,” which are discussed in
Chapter 5.

98
hyperdissonantly and therefore has an unusually deformed tension arc. This may be heard
in several works analyzed later in the dissertation.11
However, the simple view outlined in Figures 3.2 and 3.3 provides a useable
starting-point for interpreting Rachmaninoff’s compound syntax, and, moreover,
encourages appreciation of the special expressive effects that exceptional passages such
can produce.

Summary of Chapters 2 and 3

In Chapter 2, a framework for interpreting the interactions of variegated

components in a compound harmonic environment was sketched, suitable for the analysis
of Postromantic works in general and Rachmaninoff’s late Russian and exile works in
particular. Tension between different components was characterized as a kind of
hyperdissonance, which can have the effect of exaggerating, distorting, or (exceptionally)
neutralizing conventional tonal and formal premises. Following the work of several
scholars, it was suggested that the Postromantic style may be understood as involving a
variety of “deformations” and structural tensions, to which hyperdissonance events were
added as a category.
In Chapter 3, the basic rhetorical associations that “fantastic” chromatic and
modal structures have in Rachmaninoff’s mature works were identified. Close reading of
several works suggested that a trend toward increasing symmetry of pitch organization is
likely to correspond to processes of intensification and/or climax in a given passage or
work, while explicitly modal structures are likely to be associated with introductory,
expository, and/or post-climactic functions.

11
See form example the analysis of the second movement of The Bells in Chapter 4 and the
analysis of the Etude-Tableaux in D major, Op. 39, No. 9 in Chapter 5.

99
 Chapter 4

“Fantastic” Chromatic Structures

William Benjamin has suggested that chromatic structures in late tonal works may
do more than “fill in the cracks” between parts of a composed-out conventional tonal
structure.1 Although many chromatic events in Rachmaninoff’s works can be understood
as embellishments of functional syntax, analyses in Chapters 2 and 3 have shown that
some chromatic structures resist such description—and that the resistant structures are
often rhetorically marked. Most of these have involved equal division of the octave—
primarily chromatic minor-third relations (referable to the octatonic collection) and
chromatic major-third relations (referable to the hexatonic collection or to a background
whole-tone collection or augmented triad). Chromatic third relations in general have been
treated extensively by Gregory Proctor, Richard Cohn, Matthew Bribitzer-Stull, Howard
Cinnamon, David Kopp, and others.2
Richard Taruskin has traced the history of chromatic third relations in nineteenth-
and early twentieth-century Russian music from Glinka through Stravinsky; drawing
from his work, I refer to such idioms collectively as “fantastic” chromatic structures.3
Although Taruskin is concerned primarily with octatonic organization, he also recognizes
whole-tone organization, and he cites a number of passages that suggest hexatonic
organization without, however, using that term. In Taruskin’s outline, equal-interval
devices are traced from Schubert to Glinka and from Liszt to Rimsky-Korsakov.4 The
Russians absorbed equal-interval devices into the national idiom, and, in the process,

1
Benjamin, “Interlocking Diatonic Collections as a Source of Chromaticism in Late Nineteenth-Century
Music,” In Theory Only 1 (1976): 33.
2
See entries under these authors’ names in the bibliography, and discussion later in the present chapter.
3
Taruskin, Stravinsky and the Russian Traditions, in particular Chapter 4, “Chernomor to Kashchey:
Harmonic Sorcery” (255-306) and Chapter 10, “Punch into Pierrot (Petrushka)” (661-778, with special
focus on 737-59).
4
Ibid., 255-72.

100
certain technical features and the general rhetorical associations of chromatic third-
relations took on a uniquely Russian character, which Taruskin summarizes as follows:

1. “In Russian music...there is a notable tendency to make the symmetry of the third-
relations explicit in a literal way that composers to the west normally did not
exploit.”5

2. Octatonic and whole-tone structures are “equivalents: both were outgrowths of

mediant interval cycles, both were originally used as modulatory devices, both
first appeared as descending basses; and both, for Russian composers, were
evocative of evil magic.”6 This observation may be extended to hexatonic
structures.

In other words, Russian composers converted equal-interval operations from

something occurring “inside” syntax and therefore comparatively generic into something
“fantastic”— more explicitly symmetrical than similar chromatic devices in German
works, and transmitting a kind of expressive code. Taruskin’s view has been influential;
take for example the following comments by Anatole Leikin:

Rimsky-Korsakov applied the octatonic scale…to portray fantastic creatures in

his orchestral fantasy Sadko… All this constitutes the beginning of the long-
standing tradition of representing the supernatural in music…The whole-tone
scale, the augmented triad, and the octatonic scale have not acquired similar
semantic connotations in Western music.7

When Taruskin says that chromatic third relations are “modulatory,” he means
that they are unsettling or disruptive in either a structural or a rhetorical sense (or both),
even in cases where the passage in question begins and ends in the same key or on the
same chord (as many of Taruskin’s examples do). Taruskin’s approach to the analysis of
equal-interval chromatic structures is thus consonant in important ways with the ideas
presented in the first three chapters of this dissertation. It differs from the approach
adopted by theorists such as Cinnamon and Cunningham, who re-genericize chromatic

5
Ibid., 261-62.
6
Ibid., 267.
7
Leikin, “From Paganism to Orthodoxy to Theosophy,” 31.

101
third relations, treat them invariably as prolongations of tonic or dominant, and therefore
consider them fundamentally supportive of syntax, not disruptive.8
Although in Rachmaninoff’s works the specific semantic connotations of equal-
interval chromaticism (evil magic, the supernatural, etc.) are largely abandoned, its
markedness remains. In his use of octatonic structures (and extended structures derived
from octatonicism), especially, Rachmaninoff reveals a kinship with “progressive”
Russian composers with whom he is not often associated, including Rimsky-Korsakov,
whose extensive development of octatonic devices has been covered at length by
Taruskin; Rimsky-Korsakov’s student Stravinsky, as covered by Arthur Berger, Pieter
van den Toorn, Taruskin, and others;9 and Mussorgsky, Scriabin and Prokofiev, as
covered by Allen Forte, James Baker and Daniel Zimmerman, respectively.10 As
discussed in Chapter 1, the influence of Rimsky-Korsakov on Rachmaninoff seems to
have been especially significant. When Rachmaninoff fled Russia in late 1917, he took
with him only a single score by another composer: Rimsky-Korsakov’s Golden
Cockerel.11 Of the harmonic materials in this opera Rachmaninoff is reported to have
exclaimed: “And then the chromaticism. This is where the source of all the wretched
modernism lies hidden. But with Rimsky it is in the hands of a genius.”12 Still later, when
preparing for a summer of composing in 1934 (which resulted in the Rhapsody on a
Theme by Paganini), Rachmaninoff studied his own All-Night Vigil, Op. 37 and Rimsky-
Korsakov’s The Golden Cockerel and Kitezh.13 “The true greatness of Rimsky-Korsakoff
dawned on me gradually,” he said, “and I was very sorry that I never got to be his
pupil.”14
One notes a significant increase in passages based on equal-interval devices in

8
See again the discussion in Chapter 1.
9
Taruskin, Stravinsky and the Russian Traditions, 255-306
10
Arthur Berger, “Problems of Pitch Organization in Stravinsky,” Perspectives of New Music 2 (1963): 11-
42; Pieter van den Toorn, The Music of Igor Stravinsky (New Haven: Yale University Press, 1983); Forte,
“Musorgsky as Modernist”; Baker, The Music of Alexander Scriabin; and Zimmerman, “Families Without
Clusters in the Early Works of Sergei Prokofiev.” See also Steven Baur, “Ravel's ‘Russian’ Period:
Octatonicism in His Early Works, 1893-1908,” Journal of the American Musicological Society (1999):
531-92.
11
Martyn, Rachmaninoff, 287.
12
Quoted in Alfred J. and Katherine Swan, “Rachmaninoff: Personal Reminiscences,” Musical Quarterly
30 (1944): 178.
13
Harrison, Rachmaninoff, 301.
14
Swan, “Rachmaninoff: Personal Reminiscences,” 177.

102
Rachmaninoff’s late Russian and exile periods—that is to say, after Isle of the Dead
(1909). Vladimir Ashkenazy has suggested that Rachmaninoff’s late works, in
contradistinction to the works of earlier Russian composers and Rachmaninoff’s own
earlier compositions, are “no longer outgoing”—harmonies are “closing in on
themselves” in ways that “Tchaikovsky would never have dreamed of.”15 Although
Ashkenazy’s observation resists easy definition in music-theoretic terms, I take it to be in
part a response to the increased emphasis—especially at moments of structural or
rhetorical importance—on symmetrical chromatic structures.
The following pages describe in detail the special chromatic structures introduced
informally in analyses in Chapters 2 and 3. The structures described may be understood
as ultimately deriving from various kinds of equal-interval oscillation or rotation
operations. Rachmaninoff developed several idiosyncratic extensions and combinations
of “fantastic” structures, without, however, altering the basic technical procedures or
rhetorical associations. As described in Chapters 1 and 2, the appearance of equal-interval
pitch devices in a functional tonal environment represents in Rachmaninoff’s music a
kind of stress upon the functional system, which, as discussed in Chapter 3, is often
exploited expressively. In many of the analyses presented thus far, intensification, tonal
instability, and climax coincide with a trend towards symmetry in pitch organization.
These associations obtain in a large majority of works analyzed in the dissertation. In this
one regard, at least, Rachmaninoff is perhaps not so different from the early Stravinsky
after all. In Taruskin’s view, octatonic harmony in Stravinsky’s Petrushka "is animistic;
the Petrushka chord is conceived, nay motivated, by a sense of struggle, and antagonism
of order and chaos reflecting the roles of pianist versus orchestra...We are meant to hear
C and F-sharp in terms of an active, not a static, polarity—as competing centers, not
merely as docile constituents of a single, static, octatonically referable "hyperharmony,"
to borrow an apt term from Rimsky-Korsakov's own vocabulary.”16
“Fantastic” chromaticism differs from functional tonal organization not so much in
the specific scale or collection involved (the eight-note octatonic or six-note hexatonic
and whole-tone collections as opposed to the seven-note diatonic collection) as in the

15
Geoffrey Norris, “Vladimir Ashkenazy on Sergei Rachmaninoff,” Andante Online (December 2001),
http://www.andante.com/article/article.cfm?id=15463 (accessed March 10, 2009).
16
Taruskin, Stravinsky and the Russian Traditions, 756-57.

103
non-goal-directedness of the harmonic structures that result from the “fantastic”
collections’ inherent symmetries. Whereas functional tonality may be regarded as
basically goal-oriented (see Chapters 1 and 2), equal-interval chromatic structures are
primarily based on oscillation and rotation. See again Figure 3.1; the techniques outlined
in that figure are described more formally below.

Octatonic Structures (Interval 3/6/9 Basis)

“Octatonic” refers to a symmetrical eight-note scale or collection in which

semitones and whole-tones alternate. Assuming enharmonic equivalence, only three
transpositions of the scale are possible, as shown in Figure 4.1. In this dissertation, Pieter
van den Toorn’s “model A” (semitone + whole-tone) is used exclusively.17 As stated in
Chapter 2 of the dissertation, to deal with issues of enharmonic equivalence (e.g. C♯ =
D♭), I follow Joseph Straus’s fixed-zero labels: OCT(0,1), OCT(1,2), and OCT(2,3), as shown
in Figure 4.1. (There is no OCT(3,4), as it would duplicate the pitch-class content of
OCT(0,1).)

Figure 4.1. Octatonic scales

!"#$%&'( !"#$'&)( !"#$)&*(

! # " $" $" #" %" #" #" $" $" $" #" #" #" #"
$" #" #" #" #" #" $" #" %" %"

In Rachmaninoff’s works, octatonic structures are characterized by four closely-

related melodic-harmonic techniques, often in some combination:

1. Melodic presentation of segments referable to an octatonic scale

2. Rotation or oscillation of triads/seventh chords whose roots are related by minor
third or tritone
3. Special techniques involving diminished seventh chords
4. Transposition of melodic segments (purely octatonic or not) by T3, T6, or T9

17
Van den Toorn’s technical system is described in The Music of Igor Stravinsky, 31-72. His “Model B”
octatonic scale has a whole-tone + semitone configuration.

104
 Although technique 1 above might be considered sufficient for defining
“octatonicism,” recognition of techniques 2 through 4 allows identification of octatonic
structures in contexts that are not exclusively based on an octatonic scale—i.e. contexts
in which other harmonic structures are simultaneously articulated, or in which an
octatonic framework is itself chromatically decorated and the underlying scale is
therefore disguised. The word “octatonic” therefore means a modus operandi as much as
a referential collection, insofar as it indicates not only a source for melodic and harmonic
material, but a set of particular ways that that material is used in actual musical contexts,
and insofar as a structure might be clearly octatonic even if more than eight unique pitch
classes are used.18 As Vincent Persichetti observed more generally, a procedural melodic
or harmonic conception in fact often precedes a scalar conception: “It is advisable that
scales be allowed to form as a result of the impetus of melodic or harmonic patterns; the
material generated by thematic ideas may then be gathered up and placed into scale
formation.”19 This has already been shown informally in several short analyses. (See
especially the analysis of the first nine measures of the Rhapsody on a Theme by
Paganini and the analysis of “From the Gospel of St. John” in Chapter 3.)

Cycles and oscillations

Russian octatonicism through the era of Scriabin and Rachmaninoff is

fundamentally tertian, distinguishing it from the Bartók’s scale-oriented melodic
octatonicism. The octatonic is unique among symmetrical collections in the variety and
abundance of triads and seventh chords that may be obtained from it: major, minor, and
diminished triads, and major-minor, minor-minor, half-diminished, and fully-diminished
seventh chords. Rotations (or cycles; the terms are interchangeable in the present context)
of triads and/or seventh chords built upon the first, third, fifth, and seventh notes of the
scale, which are related by minor thirds) are a common manifestation of octatonicism in
Russian music from the late nineteenth century on. Oscillation between two chords is
equally common. Figure 4.2 gives sample cycles and oscillations in OCT(0,1).

18
A similar understanding of noncollection tones in octatonic and whole-tone structures in Scriabin’s
works is suggested in Jay Reise, “Late Skriabin: Some Principles Behind the Style.”
19
Vincent Persichetti, Twentieth-Century Harmony, 2nd ed. (New York: Norton, 1961), 43.

105
 Figure 4.2. Sample chord cycles and oscillations in OCT(0,1)

!
a) ascending minor triads b) descending minor triads c) ascending major triads
#" #" #" $#$""" $$$""" ##$""" %%%""" $$%""" $$$"""
! $$"" ## "" $%%""" $$$""" $#$""" $#$""" $$$""" $%%""" ## ""
$" $ " $" $"
& $" #" %" $" %" #" $" $" #" %" $"

OCT(0,1)

!
d) dominant 7th cycle e) dominant 7th cycle f) half-diminished 7th cycle
$ " # " $" $" $ " #$$$"""" $%$"$""" %%%$"""" ###"$""" #$$$"""" #$#"#""" $%%%"""" $$$""" $$$"#""" #$#"#"""
! #$$ """ # #"$"" %%%""" $%$""" #$$ """ %"
$" $ " $" $"
& $" #" %" $" %" #" $" $" %" %" $"

!
g) dominant 7th pair h) dominant 7th pair i) half-dim 7th pair j) minor-minor 7th pair
$ " $" $ " $%$"$""" ###"$""" $%$"$""" $%%"$""" $$$$"""" $%%"$"""
! #$$ """ %%%""" #$$ """ $$#$"""" $%%%"""" $$#$""""
& $" %" $" $" $" %" $" %" $" %"

A well-known example of octatonic oscillation is the tritone-related pair of major-

minor seventh chords used extensively in the “Coronation Scene” from the prologue of
Mussorgsky’s Boris Godunov (see Figure 4.3). The octatonic basis of the chord pair is
clear despite the fact that the collection is incomplete (as shown in the figure, only six
pitch classes are used).20 In more complex contexts, tertian sonorities may be
superimposed, as for example with the F♯ and C major triads of Stravinsky’s “Petrushka
chord,” or at the climax of Rachmaninoff’s “A-u!” in the Chapter 2 analysis.

20
See also Forte, “Musorgsky as Modernist.”

106
 Figure 4.3. Mussorgsky, Boris Godunov, “Coronation Scene,” OCT(2,3) oscillation

!
!"#$%&'(

! ###"""" %$"""" #" $" $" "

" #" " #"
& "
)*+,-./01,-23454167287,-93
:,,632-4;*6492<=26-06,14
#>,28,//,126,1432?234/06,142/,60,1

As the analysis of the Elegy in E♭ minor, Op. 3, No. 1 in Chapter 2 has shown,
passages that suggest equal-interval chromaticism appear in some early Rachmaninoff
works. However, extensive use of equal-interval chromatic structures and—more
tellingly—emphasis on such structures at important formal or expressive junctures
constitute in my view an important characteristic of the late Russian and exile periods.
This observation is consonant with David Cannata’s view, discussed in Chapter 1, that
Isle of the Dead (1909) represents Rachmaninoff’s real coming of age as a composer. The
tonal scheme used in Isle of the Dead “testifies to a stylistic sophistication hitherto
unknown in his works.”21 The structure Cannata describes is a complete rotation of keys
related by minor third: A minor – C minor – E♭ major – F♯ minor – A minor, as shown in
Figure 4.4.22 Cannata observes that the tonal structure is summarized in a remarkable
passage (beginning eleven measures after rehearsal 22) in which the Dies irae is treated
canonically, with entries beginning on C, E♭, A and F♯—that is to say, with entries on the
key notes of the work.23 The canonic passage is shown in Figure 4.5.

21
David Butler Cannata, “Rachmaninoff's Concept of Genre,” Studies in Music from the University of
Western Ontario 15 (1995): 72.
22
Cannata’s study of manuscripts has revealed that the E♭ and F♯ areas were added after the initial drafting
of the work. See Cannata, Rachmaninoff and the Symphony, 78-83.
23
Ibid., 81. Cannata suggests that the F♯ entry of the Dies irae is left incomplete to reflect the instability of
F♯ as a key center in the work.

107
 Figure 4.4. Isle of the Dead, Op. 29, global structure in relation to OCT(0,1)

!
!"#$%&'(
"" #""
# """
# $" %"

$. . .(
! " " $ "" % ""

& #" $" "

" "

!"#$%&'( )*+,-./0*-12.34356

Figure 4.5. Isle of the Dead, octatonic Dies irae canon after r. 22

(Only dies irae statements shown; octave doublings omitted)

!
" ! ## ## ## ## ### # ## # ## # # # # ## # # # # ##
Largo A

#$ #$ #$ # #$##$##$#### #$# #$##$#### #$# #$##$#### #$# #$##$#### #$# #$##$####

#$# #$# #$# ### #$# #$# #$# ###
$ $ $ $ $ $ $ $
p
## # #
# ### ### # #
# # #
# ### ####
% (# (# # # (# (# # )#
" !
& & & & & & & &
C

'# # '# # '# # '# # & '# & # & '# & #
OCT(2,3) '
E
(
!
F

" (* )* (* *
## ## ## ## ## ## ## ## # ### ### ### ### ### ## # ### ### ### ### ### ##
# # # # # # # # # # # # # # # #
# #
### ##### ######## # # # # # # # # ## (## ## (## ## ## ## ##
3 3 3 3 3 3 3 3

% #(# # (# # # # # #(# # )# # # # # # (# # (# # # # #
" & '# & # & '# & & '# & # & '# &
# #

What Cannata does not discuss is that the chromatic minor thirds key scheme
suggests a large-scale octatonic cycle, and that the canonic summary is explicitly
octatonic, as shown in Figure 4.5. Moreover, as Figure 4.6 shows, the summary canon
emerges from a climax event in which the four principal tones involved (A, C, E♭, and
F♯) are fused into a diminished seventh chord. The diminished seventh chord (º7:
[0,3,6,9] on the figure) is shared by OCT(0,1) (global key scheme) and OCT(2,3) (canonic
passage).

108
 Figure 4.6. Isle of the Dead, climax at r. 22

22 climax

!
" to Dies irae canon

! "

#""" $""" " " "

"" " ""
! "
#""" #" %" #" " $"
"" "" "" " "
&
" $" %" $" %" $" $" %" $" %" # " "
fff
intensification º7: [0,3,6,9]

OCT(2,3)

The first movement of the next work composed by Rachmaninoff, the Piano
Concerto No. 3 in D minor, Op. 30 (1909), is similar in certain ways to the Isle of the
Dead. Although Op. 29 and Op. 30 are in different keys (A minor, D minor), the climax
in the first movement of Op. 30 uses almost exactly the same structure—at pitch—as that
shown in Figures 4.4 through 4.6. The movement follows a conventional concerto
movement plan, and the music through measure 203 is characterized by strong functional
tonal organization. As shown in Figure 4.7, the first theme is in D minor, the second
theme (measure 93 and following) is in B♭ major, and the middle-section development
(starting with the false repeat of the first theme at measure 167) begins in fairly
unproblematic fashion. At measure 203, the minor dominant is attained. At this point, as
shown in Figure 4.8, an octatonic structure similar to those described in Isle of the Dead
replaces—or, perhaps, displaces—the conventional tonal organization displayed in the
movement up to that point. (The melodic transposition techniques indicated in Figure 4.8

109
are not discussed until later in this chapter; the reader is encouraged to return to Figure
4.8 at that point.)

Figure 4.7. Concerto No. 3, Op. 30, i, analytic overview of mm. 1–203

!
& '( &)* () (. +,(

"" ##
!"#$%&' *+,&#%--% *+,&/+/%
!"
! !"#$!% "#
$" % % %
""% % # # % % "% % % % % % % %
% "% #% "% % % #% % %

+ 0' !" !# " $ #

!$ %& !&
'# # '$ !# ) ) )
$ ( $ " " "
%
& '& '& '& 0

As shown in Figure 4.8, the climax at measure 235 is prepared by an extensive

octatonic structure, and involves diminished seventh chord [0,3,6,9]. (The octatonic
intensification into the climax is reflected by a crescendo—mf to ff to fff—and an
accelerando—Più vivo at measure 203, then Allegro at measure 223.) As shown in Figure
4.9, the climax chord and the OCT(2,3) melodic cell heard several times at the climax and
in the measures following it are not unlike the climax chord and Dies irae motive at the
Isle of the Dead climax (see again Figures 4.5 and 4.6).
However, the event in the concerto has additional, opus-specific significance. If
D♯ is respelled as E♭, it becomes clear that the climax event may be interpreted as a
development of the E♭ introduced into the movement’s main theme in measure 12 (Figure
4.10). This early E♭ is the first chromatic tone heard in the composition (other than the
ordinary leading tone, C♯). In measures 12 and 14 it tonicizes G minor (iv). At measure
235, at the climactic apex of an octatonic structure, the E♭ is radically exaggerated. As
shown in Figures 4.9 and 4.11, the climactic D♯/E♭ carries through the cadenza (which
contains a recapitulation of the first theme) into the flute solo which follows at rehearsal
19 and, ultimately, into the recapitulation of the second theme, which occurs in the key of
E♭ major.

110
 Figure 4.8. Concerto No. 3, i, analytic overview of mm. 203–235

OCT(2,3)
melodic basis: OCT(2,3) melodic cell
T3 T3 T3 14
Più vivo non-OCT: chromatic ascent 219

203   
                      
   
                           
          
mf cresc. ff
chord cycle: OCT(0,1)
WT ascent
     
           
                  

V7 of

V7 of

º7: [0,3,6,9]


111
T3 (aborted) º7
T3
Allegro
235
223
      
  
                                              
         
                         
            
             
fff
                    
      


º7: [0,3,6,9]

climax
 Figure 4.9. Concerto No. 3, i, m. 235 through recapitulation in cadenza

;96*+<

&$
7-)=1/-20)7>1*1

!"# #" $" :

" $"
!
!"#$%&'()*+,-./0)0+,,

# """% # " # " # " (# " + " # " ( # " + " # " (
!!!

! ##""& # " " *) #" #"#" &#

! !"
' ) )
,
!$% "&"
' ' ' !!!
" & $& & $& +& " " " " "$ " "
!!
!
" # $%
$!
"

'( #$ $
%

$)*+,-.

/+0123+4.1/+56789+76-2

Figure 4.10. Concerto No. 3, i, main theme, introduction of E♭

% % % % &% % % % ! % % % % % % (% %&) *% % % % % % % &% % % )

!""#$%&'()'*&*'+)*+&
!
( (
" # ! $ '
! !

!#

$ #% % ! % % % % *% #) % ! % % % % *%
% % % % ) % %&% &% % )! % %%
"

"# *%
' '
"# ! "#

112
 Figure 4.11. Concerto No. 3, i, end of cadenza through coda, analytic overview

.#

. # 89:

!
!" . # 89: !"#$% % #% % % #% % % &
" #! $
% #% % !
% % % %%
% % '% '% % % %%% '% % %
" # ! %"" %% %% %%
% %""
% % % % %""
%
% #% %""
% % %"" &%$'()

"" !"#$ !

-*'24)5

8;&<2(63=1236)7*)0*,&<)7>*%?&'&:
) *
% # % % '% % ( % (&) %" % % % % )% " % % % % % %
*%"+,-'.,%/'%0$.$12+#31.4.*%5/#2%/
89:
"# # # % $%
#
+
#&) % % % #& &
!
###
,
#% #%

,- %

./0123*'24)5
#$

!
67 89:
) * ) ) )
%" % %" &%% "" '%% %& "
%&'()*+
%* * ( %*
, # % " '% % "
- %%
% & %
## # + 8@)>2:
)
# %"
!!
, ## - & (& *
#- & (& %
(% $ %
. .
%
# *
A***++* * *******B******************6

-*'67)5

In Figure 4.11, pitch class E♭ is associated with a neighbor figure (marked “N” on
the figure), which is finally absorbed back into D minor at the coda. (However, even here
the story is not over. See rehearsal 48 in the third movement, where the first movement’s

113
second theme material is again treated in E♭ major; and see also the apotheosis E♭ major
chord, fff, now in the context of D major, after rehearsal 77 near the end of the concerto.)
To summarize: lowered scale degree 2, introduced conventionally early in the first
movement, is exploited octatonically at the movement’s climax, creating a structural
dissonance that is only resolved at the end of the first movement (and which has
implications for a point of culmination at the end of the third movement).
The Isle of the Dead and Piano Concerto No. 3 passages feature cyclical (or
rotational) articulation of octatonic structures. Similar cycles were featured in several of
the analytic vignettes in Chapters 1 and 2, including “From the Gospel of St. John,” and
the E♭ minor Etude-Tableaux, Op. 39, No. 5. Oscillations between two octatonically-
related chords are also common in Rachmaninoff’s mature works (and, exceptionally,
earlier works such as the Elegy in E♭ minor, Op. 3, No. 1), usually appearing at strongly
marked moments. (See for example analysis of the “A-u!” climax in Chapter 2.) Figure
4.12 shows an octatonic oscillation used in conjunction with a motivic melodic cell in the
finale of the Symphony No. 3, Op. 44. The passage, from rehearsal 79 through rehearsal
80, contains the end of the movement’s exposition, and forms a bridge to the fugal
episode that substitutes for a proper development in the movement. Fuller analysis of the
passage must wait until Chapter 6, where it is interpreted in the context of the entire
symphony; but some initial observations are possible here.

114
 Figure 4.12. Symphony No. 3, Op. 44, iii, r. 79–807

! $$$ !
!"
( * () $ +, + # $+ ++ #
( ( *( #
-(*( )( ) ,
!"#$%
# " % % & & '
! &'&%
' !"#$
,
$ $!
(")*+,%$-

# $ " *( & & % -(

-$++( ( $ ((
(( $$++ -(
+ (( - ++
( * ++# (
*- ((
*+#
!"#$
) ) - ,+ (
')--&&,
+#
. $$$ "! *'( *(-(-( ( (*( ( ( *(* (
"
-$*+,.-
** (( & & & & # ' $(
$(
!! "

#$

#
$$$ $*#/0%$-12+,1&($)3%-4
+# + () () () ,( # () () () ,+ 0+ # +# (' & &
/ /
""
, , , ,
! $$$ -( ( -( ( -( (-( -(
# &
5+,6-
& ' -(( (( -(( (( (( (( (( (( %$(7

$$ -( -( -( -( ( -( -( ( -( -(
""
-(( -
#$ % ( -( ( ( ( ( ( ( ( ' & $((( ' & -(( ' & & ' -( ( ()-() $()
/ / /
$$$ ' . 1 1 1 !
-( -( ' &$$(( ' & --((' &
-$*+,.-
-( ( -( ( (
" 1 & &
# ( -( ( -( ( -( $ ( %
(0 ( ( ( -(
-( -( - (/
"" %&'()*+!' ""
""

The second theme group ends in E♭ major (rehearsal 79), quietly. At rehearsal 80,
intensification of dynamics, texture, and instrumentation coincides with the onset of an
octatonic oscillation and a statement of the symphony’s motto theme in the trumpets on
C♯.24 In Figure 4.13, the events of Figure 4.12 are summarized. The motto is not
explicitly octatonic, being capable of numerous harmonizations; but it is, in this context,
an OCT(1,2) melodic cell. The chord oscillation is of the “Coronation” tritone type.

24
As discussed in Chapter 6, the motto theme is heard at a variety of pitch levels in the symphony: most
significantly on A in the first movement, and on C♯ in the second and third movements.

115
 Figure 4.13. Symphony No. 3, iii, r. 80, octatonic chord pair and melodic cell

!"##"

! #" " #" " #"

$%&'()*+
! ###$"""" $$"" #" $"#"
$"
$" # " $" $" $" $" #"

$%&'()*+,"-./001#/"2

Dynamics and instrumentation make clear the passage’s rhetorical significance.

The passage is also structurally significant, as shown in the analytic overview in Figure
4.14. (The events in Figure 4.14 are described more fully in Chapter 6.) The OCT(1,2)
oscillation at the end of the exposition (Figure 4.12) initiates a much larger cycle that
ultimately leads into the recapitulation by way of E♮, supporting V7 of the global tonic A
major. The fugue which substitutes for a proper development in the movement begins in
D major and climaxes on B major at rehearsal 93, as shown. The extraordinary pp
passage after rehearsal 94 —an interruption, raw in its “orientalism”—occupies the B♭
OCT(1,2) node, preparing the large-scale dominant which follows. As shown in Figure
4.14, the primary melodic tones and highpoints from rehearsal 80 through the
recapitulation involve notes from the motto theme on C♯ heard in the trumpets at 80. This
large-scale articulation of the motto carries over into the movement’s coda (after
rehearsal 110), at which point the motto is again stated on C♯, but this time in A major—
and note how the C♯ major-minor seventh chord from rehearsal 80 is incorporated into
the motto statement.

116
 Figure 4.14. Symphony No. 3, iii, analytic overview of r. 80 through coda

637$"8
695":2#2"3 /010& *23#&4405#2"3. ;&<=52#0>=#2"3 ("7=
!" !"$?+- #$ #%$?@ #&$?A ##$?B ''"$?B
!"#"$!%$&

'' '' '' (( '' %' ((

!"##"$#%&!&

! #"#"#" " " #" " ###

!"##"$#%&!& " #" !"##"$#%&!&
" " #""
! #"" #" $" " " #" " " " "
" ""

% ##""" $$""" &"" #$""" ### """ " #"" $"" "" ""
" #"
$" " " $"#" " " " "" " " " "
&"
!" !#
() (( !" &((
'()*+,-. !" ) (

Diminished seventh techniques

Because octatonic structures are characterized by minor thirds and tritones,

diminished seventh chords are particularly easy to come by. (See again the Isle of the
Dead climax chord in Figure 4.5, the analysis of the Concerto No. 3 first movement
climax in Figure 4.10, and numerous analyses in Chapters 2 and 3.) Any given octatonic
scale contains two different diminished seventh chords, which leads directly to several
notable techniques in the works studied. The first technique is shown in Figure 4.15a:
octatonic decoration of a (functional) diminished seventh chord with passing tones.25
However, Taruskin dismisses such decorated diminished seventh chords as only
superficially octatonic: “true octatonicism preempts functions normally exercised by the
circles of fifths, whether by a rotation of thirds or, more radically, by a tonally stable
diminished harmony.”26 More idiomatically octatonic is the technique shown in Figure
4.15b: two diminished seventh chords, which combine to form a complete octatonic
25
The Dies irae canon in Isle of the Dead may be interpreted as articulating an ornamented diminished
seventh chord.
26
Taruskin, Stravinsky and the Russian Traditions, 269.

117
collection as shown in Figure 4.15d, are entangled. Entangled diminished seventh chords
in octatonic contexts were featured but not discussed in detail in several analyses in
Chapters 2 and 3.

Figure 4.15. Octatonic diminished seventh chord techniques

!
a) decorated º7 b) entangled º7’s c) related dom7 cycle

#" #" #" %%%#"""" ###"$""" ####"""" %#%"#"""

! $" #" #" %"
& " " " %" %"""
"
P P P
$" " %" %"
%" #" #" %" #" "
functional resolution

d) º7’s combine to octatonic

! ##%""" %" %" #"

#" %#%#"""" %" #" " " "
º7 + º7 = OCT(1,2)

Although the entangled diminished seventh chords of Figure 4.15b are clearly
related to the chord cycle shown in Figure 4.15c, melodic emphasis in a given passage
may bring out the entanglement, greatly increasing the intensity of dissonance in the
passage. Figure 4.16 is such a case. (The passage has a D♭ major key signature in the
score but is notated without a key signature in Figure 4.16 to make plainer the interval
structure.) In this passage, tones of [2,5,8,11] are emphasized in the melodic material,
while the roots of the OCT(1,2) chord rotation emphasize [1,4,7,10].

118
 Figure 4.16. Symphonic Dances, Op. 45, iii, analysis of local climax at r. 79

!
+9:$743.#;'<=>?,#,'<<4#-&(@*A=*'>

!"
!"
6" 78
" " #" %"
! " $"
$" %"
" %" #" #" %"
%" %"
/01#2345464778
!
#"
!
! $"&& #$"" %" "" %"
&
"!"#$%&'()*#+,-.
#" ## #""" ## %""" ## $""" %# %""" %%%""" %%""
' ## %""" #" #" $" #" #" "
%" #" #"
%" %" %" %" %"
/01#274G4047H8

+9:$743.#,B=&>#&=-(-?=@#C#(DE?<<?(&F#-=@'*

#$ ! 0!/,&/%*,(#+1/!0"

!
%&'()*+,,-./.0123+4/5
2&)+'($)/%
!"#$%&'($)*'$*'$)(+*,(#'$!'",*-.*$+'/'$)(+*#'!&+'&!"

!! "

119
Transposition of melodic segments by T3, T6, or T9

Because the octatonic collection (0134679t) features an interval pattern that recurs
at the distance of a minor third, any octatonic material (melodic, harmonic or both)
transposed by minor third up or down or by tritone will stay inside the given octatonic
collection. Particular clear examples of this will be found in passages analyzed in the
section above, and in several analyses in Chapters 1 and 2 of the dissertation. The
T3/T6/T9 technique can be extended to involve transposition of material that is not
exclusively octatonic: when several T3, T6, or T9 operations are used in succession, the
effect may be “octatonic” even if the segment so transposed is not itself exclusively
octatonic. Figure 4.17, an annotated reduction of the opening of Rimsky-Korsakov’s
Golden Cockerel, demonstrates. (The passage features all of the octatonic techniques
identified above.)
In Figure 4.17, the pair of triads outlined in the first six measures—D♭ major and
F♭ (or E) major—establish OCT(1,2) as a potential basis for the passage; D♭ and E are two
of the four OCT(1,2) nodes. In measures 7 – 12 of the passage, diminished seventh chord
[2,5,8,11] is outlined by chain transposition of a melodic theme at T9. (The structure is
similar to that in Figure 4.15, but on a larger scale. Other similarities between The Golden
Cockerel and Rachmaninoff’s Symphonic Dances are noted below.) The transposition
process, and the diminished seventh chord it generates are disguised but not undone by
the presence of many non-octatonic auxiliary tones in the melodic motive itself.
After measure 11, diminished seventh chord [2,5,8,11] is sustained above bass
note G♮ (a third OCT(1,2) node). Octatonic organization gradually dissolves after measure
13, and a non-octatonic passage follows at measure 19. However, octatonic structures
again characterize the music at measure 27: OCT(1,2) node E♮ in the bass, diminished
seventh [1,4,7,10] (the other OCT(1,2) diminished seventh chord), and T9 transpositions of
the melodic theme. The fourth OCT(1,2) node, B♭, arrives at measure 38 (marked by an
asterisk); but by this point, OCT(1,2) has been left behind in favor of OCT(0,1), which is the
setting for similar T9 and diminished seventh techniques leading into measure 38.

120
 Figure 4.17. Rimsky-Korsakov, The Golden Cockerel, Prologue, analysis of mm. 1–38

T9
1 7 º7: [2,5,8,11] 11
                       
      
  

     
  
                
  
      

Basis: OCT(1,2) T9 º7: [2,5,8,11] over G

T9

D E G

27 T9 T T9
13         19                 9
      
                           
               
      

121
not 2x with change of octave
dissolution of OCT(1,2)    octatonic º7: [1,4,7,10]
   
  
     

     
  
         
   
6 N #7 H
D E 5 4
E

31
T9 38
       
               
            
      
(sim.)
      
º7: [0,3,6,9] over E    *
 
 
                   
   
 
6
  
º7: [2,5,8,11] over E Basis: OCT(0,1) !4
B
 Figure 4.18 is an analytic overview of the entire passage, showing OCT(1,2) nodes
D♭, E, G, and B♭ and diminished seventh chords that result from T9 treatment of the
melodic theme. Triads belonging to OCT(1,2) or, later in the passage, OCT(2,3) are boxed.

Figure 4.18. The Golden Cockerel, Prologue, mm. 1–38, analytic overview

,-./*'&0 ,-./1'*0

! $$#""" ##%"""
!"#$%&'(')'**+ !"#$%*'2'"'*1+ !"#$%&'(')'**+ !"#$%1'3'4'5+
$ " #" #" " %" $" #" #" #" %" %" #" "
! $" #" #" $" #" $$""

& $" #" #"

"
#" $"
!"#$%*'2'"'*1+

%*'2'"'*1+$6$%&'(')'**+$7$,-./*'&0

%*'2'"'*1+$6$%1'3'4'5+$7$,-./1'*0$

As discussed at the beginning of this chapter, Rachmaninoff showed great interest

in The Golden Cockerel over a period of many years. Barrie Martyn has noted a specific
melodic similarity between a theme in the Prologue of The Golden Cockerel and a theme
from the first movement of Rachmaninoff’s Symphonic Dances without, however,
developing the observation analytically. Recognition of octatonic structures allows a
more detailed comparison of the themes. Figure 4.19 shows the two themes; an analytic
reduction is shown beneath each theme. (Figure 4.19b is the melodic theme from Figure
4.17.)

122
 Figure 4.19. Comparison of Symphonic Dances theme and Golden Cockerel theme

(a) Symphonic Dances, Op. 45, i, analysis of measures 10–11

Allegro
)
10
# %
" # # ! $ %& ' (&% ' & ' (% ' #% % %& ' *%) '
ff

# % % %
"# # (% *%
º7

(b) The Golden Cockerel, analysis of measures 7–8

Allegro

+ # ! % % %*%(%(% % #% % % %(%#% %#% % % (%

7

##
pp

+ ## % % %
# (%
º7

The similarity of the themes’ general contours is obvious. Less obvious is the fact
that both themes outline a diminished seventh chord, and that the chord is decorated by a
descending chromatic line, which creates the distinctive contour. The similarity does not
end here. Figure 4.20 shows that the entire opening passage of Rachmaninoff’s dance
resembles the opening of Golden Cockerel.27

27
Bertensson and Leyda report Rachmaninoff’s recollection, many years after the fact, of an occasion when
he, Rimsky-Korsakov, and Scriabin discussed The Golden Cockerel (still a work in progress) at a café.

123
 Figure 4.20. Symphonic Dances, Op. 45, i, introduction (mm. 1–16), analytic overview

12%&34,-,5,67

!!

! "" #
!

! " $# "# $## "$## "##

%$### $"## $##
#
"%## %## $## "## $ ##
#
"## ## #
*&&.

! ;
" "# "# "#
"! " " $# %# # & $# # $# "#
"# # " #

!"#$#%&'()*+,-.

"
!"
!%

! "" #
!$ !& !#
# ""### ##
! " ## """### $## ""## ## % # $# & $$## # #
$ ## """#
$## $## ## ## " # ## $ ## $" ## ## $ #
"$ ## ## ##

&"
" "" #
"# # "# # $# "#
# " # $# # # #
# "# "# #
# $#
"# #

'()*+,-. '()*/,+. 3+,9,:,//7

0
34,-,5,67
8&3+,9,:7 8&32,/4,/7

!" " #

As shown in Figure 4.20, many non-octatonic tones are present in the sixteen
measure introduction to dance. It might be correct to say that the OCT(2,3) material shown
in Figure 4.20 is only a framework upon which a complex, highly individual structure is
built. However, certain features are common to the Rimsky-Korsakov and Rachmaninoff
passages: the diminished-seventh oriented melodic material at measure 10, and the
gradually filled-in diminished seventh chord of the opening measures. The Rachmaninoff
passage is perhaps richer than the Rimsky-Korsakov passage, however, because the
OCT(2,3) material is only one component of the structure. Of the three triads arpeggiated

Rachmaninoff comments, “What untold riches there are in the Coq d’Or! The beginning alone—how
novel…I don’t know what impression this conversation made on Scriabin. But I was deeply stirred.”
(Bertensson and Leyda, Sergei Rachmaninoff, 138-39).

124
at the beginning of the passage (bracketed and marked with an asterisk), only two belong
to the OCT(2,3) collection—the D major and A♭ major triads are part of the collection, but
the G♭ major triad does not belong (though the sustained G♭ does). Similarly, the A minor
triad at measure 7 and measure 14 is treated as a kind of non-octatonic alternative to the
A♭ major triad (a neighbor chord, perhaps); it enriches the harmonic content of the
passage, without, however, necessarily undoing the octatonic framework. The octatonic
structure of the introduction is “dirty”; but the dirtiness is significant, because the three
bracketed triads (plus the A minor triad) comprise a chord group that is used motivically
in all three movements. (See the longer analysis of Op. 45 in Chapter 6.) The octatonic
framework in this late Rachmaninoff passage is itself extensively elaborated and
decorated—it is given the same kind of flexible treatment that familiar tonal idioms are
given in the mature common practice.
I have been at pains to suggest throughout this dissertation that functional tonal
syntax remains central in Rachmaninoff’s idiom even when special chromatic and modal
structures are emphasized. Figure 4.20 is a good case in point. As marked on the figure,
tonal functions in C minor are engaged as the introduction ends, and a lengthy span of
music in C minor (actually C Aeolian) begins in measure 16. (That section is analyzed
later in the present chapter). No similarly clear functional engagement occurs within the
octatonic structure in the Golden Cockerel opening.

Rachmaninoff’s and Scriabin’s Symmetrical Structures Compared

By now, the exaggeration and distortion effected by equal-interval chromatic

structures in many passages from Rachmaninoff’s mature works are clear. These are, I
believe, a hallmark of the composer’s mature style. It is interesting to compare such
passages as the Symphony No. 3 first movement climax, the “A-u!” climax, and the
“Daisies” climax to the same kinds of symmetrical pitch structures as they occur in
Scriabin’s late music.
Figure 4.21 is an analytic overview of the first half of Scriabin’s Prelude, Op. 74,
No. 3 (1914). The prelude is almost entirely octatonic. It falls into two parallel halves,
followed by a short “tag” (measures 25–26). As Figure 4.21a and 4.21b show, the

125
octatonic techniques in the piece are similar to those used by Rachmaninoff and Rimsky-
Korsakov: a melodic motive (x) is transposed at T6 and T9; major-minor seventh chords
related by tritone appear; two different diminished seventh chords are entangled,
summing to OCT(0,1) (as shown in Figure 4.22); and in measures 9–12, an extended
segment of octatonic melodic patterns is transposed up by minor third to close the
section.
As shown in Figure 4.23, the second half of Scriabin’s prelude is a transposition
by tritone of the first half.28 The entire prelude remains inside one octatonic collection.
There is motion within the collection, but otherwise no motion at all, recalling Taruskin’s
observation (quoted in Chapter 2) that “music based on universal invariant harmonies
becomes quite literally timeless, as well as emotionally quiescent.”29 How different from
Rachmaninoff’s boiling octatonic climaxes! Although the equal-interval pitch structures
in Scriabin’s prelude and Rachmaninoff’s “A-u!” (see Figures 2.25, 2.26, and 2.27) are
nearly identical, all being derived from tertian models, and all involving the same kinds
of root relations, transposition operations, and diminished seventh techniques, the
rhetorical effects of the structures in the two passages could hardly be more different—
distortion and climax in the Rachmaninoff, invariant, changeless quiescence (motion
without motion) in the Scriabin. In Scriabin’s late style, functional premises are
effectively neutralized. In Rachmaninoff’s late style, a tension between conventional
functions and symmetrical chromaticism remains.

28
Commonly available editions have a misprint in measure 21, which has unfortunately been followed in
many performances and recordings. On the lower staff, the ♯ before G should instead be before D. This
restores the chord to OCT(0,1), and makes it an exact analog (at T6) to the chord in measure 9.
29
Taruskin, Defining Russia Musically, 348-49.

126
 Figure 4.21. Skryabin, Prelude, Op. 74, No. 3, analytic overview of mm. 1–13
(a) OCT(0,1) materials in measures 1 - 5: motive x, melodic segments and seventh chords at T6

T6
T6

1 3 5
P P
         
     P   
        
   
           

x x
motive x
  
    
T6 enharmonic
major-minor seventh chords
(“Coronation” type)

127
(b) Analysis of measures 1 - 13

emphasized melodic tones (motive x)

1 3 5 7 8 9 11 13
T6 T9 T9 (T9)
T6 T3
           

                         
                     
"tenor"    octatonic patterns
    
   

º7: [1,4,7,10] 
º7: [0,3,6,9]
 Figure 4.22. Prelude, Op. 74, No. 3, “entangled” diminished seventh chords

89:';45<5=5>? % 89:';65@59564? ! 01234567

! #" $" #" " "

#" # " $" #"
# " #" # " # " #" " "
!"#$%&'(')*++ ,-".$/,

Figure 4.23. Prelude, Op. 74, No. 3, analytic overview of the entire piece

'(

!
!"#$%& !"#$%)
!"#

! " !" #$
#" #" #" #" #" #"
! " #"
" #" " #" " #"# " # " #"
" # " $ " ## ""
% $" #" # " " #" $$""
#" #" #" #" "
#" " #" " #" #"
#" #"

$%&$'"()*+(+&'$'%*),*%-)$.&+,#.+,$/)0,$)1,(2$'+(*)(%,$&"3'4%-

Consider, as a contrast to the Scriabin prelude, the second movement of The Bells,
Op. 35 (1913), an intensely chromatic work almost certainly composed with Scriabin’s
and Rimsky-Korsakov’s harmonic experiments in mind. The movement is in D major.
The previous movement is in A♭ major (see again the discussion of the first movement in
Chapter 2).30 The tritone relation between the first movement’s A♭ major and the second
movement’s D major suggests octatonic possibilities. As suggested in Figure 4.24 (boxes
1 and 2) and in Figures 2.26 and 2.27, an octatonic structure is articulated when the two
keys—or, rather, triads representing the two keys—are superimposed at the beginning of
the second movement. The movement therefore begins, unusually for Rachmaninoff,
with a marked octatonic structure.

30
The Bells does not begin and end in the same key. David Cannata has suggested that the entire
composition is organized around the fourth movement’s C♯ minor/D♭ major. See Cannata, Rachmaninoff
and the Symphony, 83-87.

128
 Figure 4.24. The Bells, Op. 35, ii, annotated reduction of mm. 1–14

Lento
30
formation of theme d d d
1
   violas (con sordini )
  ( )            
p
                                
mf
poco a poco cresc.

strings pizz., clarinets, and harp f

horns and trumpets 
        
                   
sf sf   
f 
     
                
 
cellos and basses
     
sforz.
1 2
OCT(2,3) oscillation

129
( “Coronation”-type)

 3 3 3 3 3
                
dim.
       

  31 violins I and II
         
           
         
 
dim. p
 cellos


            
        

dim.

p
             
f
sforz. sforz. pp
 Figure 4.25. The Bells, ii, overview of thematic material

!(#"3&,2(0%&+"&4"01/2/"+"-5,%+6"01/"178/,-%))&+(+0"&8/+%+6

!"#$% 01/2/"+
$! %
!

& (' ! ' ') ' ! ' * ') (+ ! ' ') (' * ') ' ! ' ( ') (' ! ' ') (' * ') ' ( ') ' )
$%&'()"!*&+")&,-%+%#
# $ " !""#
! ('
"#$#%&%"#$#%$'()$*

01/2/"+

# $ ('" '" '" " '" " '" " '" '" '" " ('" " '" " '" '" '" " '" " '" "
$
*' (' (' ' (' (' *' ' '
"#

!9#":1/2/"+"%+"0&+%*";"2(<&,

01/2/"+

# $$ %
&' $%&'%+)"."(+-"..
" # & , ' ' ' '
' ' ' '
!

) ' ') ' ' $' *' ' ') ' ' ')
$
- $ % ) ' ')
*/''&)
$ & ' $' ' '
+! +! +!
"

(c) Chromatic theme

!"#$"%&'
!" & & & ( & & ) & & $& (&
&(& & )& *& &(&$*& (& *& ! & &$& & )& & & &$& & !
#(#)
$$ ! ' '
# "% % ' ' &
!!"#$%& "'()

130
 As shown in Figure 4.24, the opening measures of the second movement feature a
gradual spinning-out of thematic material (theme d, so labeled because of the contour
similarity with the Dies irae) to fill in the interval between pitch class E♭ and pitch class
A♭. This interval may be interpreted as a sort of holdover from the first movement’s
tonic. (Note that the D♮ gives this material an A♭ Lydian character, which is the modal
form heard at the end of the first movement. See again the analysis of that movement in
Chapter 3.) Theme d is important throughout the second movement, appearing in a
number of forms: see Figure 4.25a and 4.25b. The only other significant thematic
material, the “chromatic theme” shown in Figure 4.25c, is probably derived from the
cello countermelody to theme d shown in Figure 4.25b. Throughout the movement, the
sixteenth-note version of theme d shown in Figures 4.24 and 4.25a is associated with
octatonic structures.
The OCT(2,3) oscillation at the beginning of the movement (Figure 4.24) is one of
the more remarkable “fantastic” passages in Rachmaninoff’s works. It is similar in
several ways to Mussorgsky’s “Coronation Scene” prototype: it features two sonorities
related by tritone (here rooted on D and A♭) and it features the same kind of antiphonal
blagovest bell texture wherein the harmonic material is separated into two sonic layers,
one in the upper register and one in the lower.31 The blagovest texture in Mussorgsky’s
“Coronation” is explicitly associated with bells; in the Rachmaninoff movement, no
actual bells sound, but the title of the composition and the poem make the association
plain. The octatonic structure of the opening measures is detailed in Figure 4.26. Because
the D major tonic of the movement is entangled in a noisy octatonic structure at the
opening of the movement, this may be interpreted as hyperdissonance at not the point of
remove nor the point of return, but at the point of departure (or perhaps, if the first

31
See Edward V. Williams, “The Blagovest Theme in Russian Music,” Kennan Institute for Advanced
Russian Studies Colloquium (Washington, D.C.: Wilson Center, 1987). Williams defines the idiom as a
two-part sonic texture or “sound complex” referable to bells (Williams, “The Blagovest Theme in Russian
Music,” 40), and he identifies the opening of Concerto No. 2, Op. 18 as the paradigmatic example of the
idiom in Rachmaninoff’s music. Jason T. Stell identifies the blagovest idiom in some of Rachmaninoff’s
solo piano works, drawing from Williams’s research (“Rachmaninov’s Expressive Strategies,” 23-25). Stell
notes that “blagovest permeates Rachmaninov’s music to such an extent that it becomes a recurrent topic”
(“Rachmaninov’s Expressive Strategies,” 24).

131
movement is taken into account, the point of continuation). The unusual tension of the
event—an unstable tonic at the opening of a movement, not treated as an applied chord
(as for example the first chord in Beethoven’s Symphony No. 1)—is reflected in a
dynamic and textural crescendo across the opening measures. As shown in Figure 4.26c,
the clarification of D major as tonic at rehearsal 31 is entirely non-functional. Up to that
point, thematic presentation has strongly suggested that A♭ major will be the movement’s
tonic. When D major emerges as the legitimate tonic, at 31, the violins take up theme d
and give it an entirely new rhythmic and textural character.

Figure 4.26. The Bells, ii, OCT(2,3) in mm. 1–10

!
$.( $9( $+( !"
" $# %$$### $
! " %$ ## ## $%$#### "%##
#
% ## %$"### #
& "" # # "%###
0,0/64+6)78
"# $# %# %# %# #
!"#$%&'( ! # $# %#

)*+,--./,)0

1)23130/455
/)0,+
!"!#$%!&'("!)*
+,-"*%'("!

Figure 4.27 is an analytic overview of the entire movement. (Some material is

omitted to save space.) The figure shows the return of octatonic structures at several
points, and shows how the climactic E♭ major at measure 97 may be understood as
emerging from the octatonically-induced tension between E♭/A♭ and D major in the first
ten measures.32 I am tempted to say that the hyperdissonant opening—an unusual and

32
In David Cannata’s short analysis of the movement, he emphasizes the E♭ major/G♭ major material from
rehearsal 41 through 44 (including the climax event at measure 97) as a “contrasting tonal plateau,”

132
striking thing in Rachmaninoff’s works, to be sure—carries over into the climax.
Significantly, the climax event features theme d at the same pitch level heard in the
opening measures of the movement; and theme d is prominent in all of the octatonic
oscillations and at the climax event. As shown in Figure 4.28, the third octatonic passage
in the movement (at measure 111) connects back to the first, returning to OCT(2,3); it is
post-climactic, and marked p, unlike the two octatonic passages that came before. As
shown in Figure 4.27, a final OCT(2,3) event at measure 151 confirms D major as tonic.
All of this is to say that, while the “fantastic” structures used by Rachmaninoff
and Scriabin are similar in many superficial regards (and sometimes virtually identical),
the active, tensive hyperdissonance exploited by Rachmaninoff is expressively far
different from the static, “timeless” condition engendered by Scriabin’s symmetrical
structures. In the former case, insistence upon functional tonal premises as an underlying
basis leads to a structural tug of war (to exaggeration, to distortion, and ultimately to
climax events that exploit the tension); in the latter case, insistence that those premises
shall not—must not—apply results in their neutralization.

without, however, associating the event with any earlier events in the movement or work. See Cannata
Rachmaninoff and the Symphony, 86-87.

133
 Figure 4.27. The Bells, ii, analytic overview

(full release)
1st 2nd
[ movement ][ movement ] structural hyperdissonance climax
with tonic D major
E  elements
1 13 68 72 76 97 105 122 151 153


                                

134
              
 
   
        
theme d etc.
OCT(2,3)
material omitted

               

              
 
    
  
OCT(2,3)          OCT    
OCT(0,1) (2,3)

material omitted
sf f ff p p f pp

7
D major elements
II 7 V b5 I IV I
 Figure 4.28. The Bells, ii, octatonic oscillations in measures 1, 72, and 111

!" !# $%
! ""
"#$%$!! "#$%$!! "#$%$!!

!
$# # #%# # # %# # # # # #
! $# $# $# # # # #%# # # %#
$#
" $# %$$###
& " $%$### %%$#%###
$ # $ #
$%$#### "# $%$####
%# %# $%## # $%## # $ ##
' "" %$"### %$"### $%## $%## "## %%##
"##
%# %# %#
" #$ """

&'()*+,- &'().+/- &'()*+,-

Hexatonic Structures (Interval 4/8 Basis)

“Hexatonic” refers to a symmetrical six-note collection in which semitones and

minor thirds alternate, of which there are four distinct transpositions as shown in Figure
4.29 (transposition of any hexatonic scale by major third up or down duplicates pitch-
class content).33 Again, fixed-zero labels are used.34 If octatonicism is characterized by
consistent interval 3/6/9 and/or T3/T6/T9 operations and diminished seventh chord
techniques in a tertian context, hexatonicism is characterized by interval 4/8 patterns and
T4/T8 transposition activity in a tertian context, even if—as already suggested for
octatonicism—tones foreign to the hexatonic collection are present.

33
The term “hexatonic” was introduced in Vincent Persichetti, Twentieth-Century Harmony, 2nd ed. (New
York: W.W. Norton, 1961), 53; see especially his Example 2-37, where the term covers a variety of
different six-note collections. The more restricted definition adopted in the dissertation has been
popularized by Richard Cohn, “Maximally Smooth Cycles, Hexatonic Systems, and the Analysis of Late-
Romantic Triadic Progressions,” Music Analysis 15 (1990): 9-40.
34
These labels are derived from Straus’s octatonic labels, and are also suggested in Miguel A. Roig-
Francoli, Understanding Post-Tonal Music (New York: McGraw-Hill, 2008), 57-58.

135
 Figure 4.29. Hexatonic scales

!"#$%&'( !"#$'&)( !"#$)&*( !"#$*&+(

! # " $" #" #" $" #" #" $" $" #" #" #"
% " #" #" %" #" $" $" #" $" #" #" %"

As with octatonicism, structures based on chord rotation and oscillation are

common in the works analyzed. Samples are shown in simple form in Figure 4.30. In
Figure 4.30, note the inclusion of some “extended” hexatonic structures—that is to say,
structures involving seventh chords with tones foreign to the hexatonic collection. A pure
hexatonic collection does not allow major-minor seventh chords; but they are commonly
used in hexatonic-type progressions. In such a case, the structure may be said to have
HEX(x,x) as a basis, without necessarily committing to HEX(x,x) for its full content. (A
hexatonic-type structure involving a seventh chord not purely of the collection was
analyzed, though not labeled as such, in the song “Daisies,” Op. 38, No. 3 in Chapter 2;
the basic structure is reproduced in Figure 4.30h.)

Figure 4.30. Sample hexatonic rotations and oscillations, including extended versions

!
!"#$%&'!()*# !0#$%&'!()*# !1#$%&'!*)2# !3#$%&'!,)(# !4#$%&'!,)(#

" $" #" %""

! #""" %$"" $$ "" !$$$$$#
" $" #" %""
%"" ##""" $%"" !$$$$$# $$""" #%#"""" %%"""" $$$"""" ""
#" " " $$""" #" %"" ""
" "
"
# "" $"
"
% ""
$5"<78$98/"3: 5/678$98/"3: ; ;
& " $" $" " $" %" " $" "
$" " " #" %" $ " " " "
%&!'##()'%*
!"!#$ !"!#$ !"!#$ %&!'##()'%*

!
!+#$%&'!,)(# !-#$%&'!,)(# !.#$%&'!,)(# !/#$%&'!,)(#

! """ ##""" " """ $$$"""" %" """ # #"""" " """
% "" % "" % "" # """" % """
#" $"""
$" %" % "
& " #" " " $" " " #" "
" " "
; ;
;
+,('&'$&-

136
 Unlike octatonic rotations and oscillations, hexatonic rotations and oscillations
may simulate V – I or iv – I resolutions, including melodic resolution of the leading tone
or melodic resolution of scale degree 6 (lowered) to scale degree 5, or both. In other
words, although hexatonic structures deny root relations by perfect fifth just as octatonic
structures do, hexatonic structures do not completely eliminate tendency-tone activity.35
(In Figure 4.30, some tendency tones, are shown by small noteheads in parentheses.) The
availability of functional tendency tones allows hexatonic structures to substitute for
diatonic functional ones in some cases, making explicit the interaction of functional
syntax and symmetrical chromaticism and simplifying the introduction of equal-interval
chromaticism as a structuring mechanism.36 (See again the analysis of the Rhapsody on a
Theme by Paganini, Theme and Variations VIII and IX in Chapter 2.)
Figure 4.31 shows a HEX(3,4) cycle substituting for functional syntax in a work by
Prokofiev. Figure 4.31a is an analytic reduction of the first two measures of “The Girl
Juliet,” from the ballet Romeo and Juliet (composed 1935–36; later revised). Figures
4.31b through 4.31e detail the relationship of the hexatonic structure to an underlying
syntactical model—the hexatonic structure may be taken as chromatic exaggeration of a
kind not unlike that discussed in Chapter 1. Figure 4.31b shows a basic I – V – I
prototype. Figures 4.31c and 4.31d show the hypothetical inflection that yields the
version used in the passage. Figure 4.31e details the tendency tones that remain. As
Kevin J. Swinden has observed in his analysis of Wagner’s “Tarnhelm” motive,
chromatic major-third relations may be thought of as functional hybrids, combining
aspects of authentic and plagal harmonic motion.37 This is shown in Figure 4.31e by the
resolution of scale degree 7 (associated with dominant function) to scale degree 1 and the
resolution of lowered scale degree 6 (associated with subdominant function) to scale
degree 5.

35
On potential dominant function of chromatic major-third relations, especially when major-minor seventh
chords are involved, see Charles J. Smith, “The Functional Extravagance of Chromatic Chords,” Music
Theory Spectrum 8 (1986), 126-27.
36
However, an important difference should be noted: as discussed in relation to the “Daisies” climax in
Chapter 1, chromatic major third relations, quite unlike the arrangement of proper dominant and tonic in
functional syntax, are reciprocal—a chord may be constructed on any “root” in a hexatonic collection so as
to contain the leading tone of any other major or minor triad in the same collection, which means that a
chord “tonicized” hexatonically might well turn around and tonicize the chord that tonicized it.
37
Kevin J. Swinden, “When Functions Collide: Aspects of Plural Function in Chromatic
Music,” Music Theory Spectrum 27 (2005): 249-82.

137
 Figure 4.31. Prokofiev, Romeo and Juliet, “The Girl Juliet,” analysis of mm. 1–2

!"#$%%&'()*$"+",-./0$12340./5+
!6# !0# !3# !2# "
!
" " " # "
!
" " " "" "" "" " " %"" " "" #"" $"" %"" "
"
! "" " " " " ##"" %$"" %""
"
" "" "" #"" " % "" "#"%" "

#" %" " " " #" " " #"

&
" " " " %" " ! $" #" %"

789!:;<# # 789!:;<# #"$ "

%
=$$$$>$$$= =$$$$>=$$>?
$$=

! # ! ! "# # !

Perhaps because of the relative ease with which chromatic major-third relations
and tonal functions can be fused, hexatonic structures do not have the strong extra-
musical associations (evil magic, the supernatural, etc.) that octatonic structures do in
Russian music. However, in his article on “Fantastic” chromaticism, Taruskin identifies a
number of passages based on chromatic major-third relations. His statement, quoted
earlier in the chapter, that octatonic and whole-tone structures are functional
“equivalents” might be revised to read as follows: “octatonic, hexatonic and whole-tone
organization are (to some degree) functional equivalents.”
Important recent contributions to scholarship on chromatic major-third relations
in general have been made by Richard Cohn, Matthew Bribitzer-Stull, and David Kopp.
These scholars differ greatly in their basic conceptions, however.38 Cohn emphasizes the
“smoothness” of his chromatic major-third cycles.39 Bribitzer-Stull’s chromatic major-
third relations, on the other hand, are full of tension, reminiscent of the disruptive third-
relations Taruskin describes. And yet Cohn’s theory has value for the present study,
because he allows chromatic major-third relations to exist separately functional syntax—
indeed, he insists that they exist outside functional syntax, whereas Bribitzer-Stull, like

38
Matthew Bribitzer-Stull, ““The A-Flat-C-E Complex: The Origin and Function of Chromatic Major
Third Collections in Nineteenth-Century Music,” Music Theory Spectrum 28 (2006): 167-90; Cohn,
“Maximally Smooth Cycles”; and Kopp, Chromatic Transformations in Nineteenth-Century Music.
39
He means “smooth” from the standpoint of underlying voice-leading; but I find it difficult to separate the
technical smoothness from a rhetorical smoothness.

138
Cinnamon and Cunningham, is concerned primarily with the occurrence of chromatic
third relations inside functional organization. David Kopp’s theory of chromatic third
relations (major and minor) to some degree resembles Cohn’s in that it is based on a
belief that chromatic third relations “possess an identity and a quality which are
independent of the fifth relations and diatonic third relations of the tonal system,
displaying an independent functional identity.”40 Kopp, however, presents a larger, “well-
ordered harmonic system” of “common-tone tonality” that, like Daniel Harrison’s theory,
effectively eliminates a basis for any friction between diatonic and chromatic structures.41
This I regard as deeply problematic for Rachmaninoff’s music, for reasons that I hope are
clear by this point.
A sort of theoretical compromise is offered: chromatic major-third relations, like
chromatic minor-third relations, are presented as conceptually independent of functional
syntax in the present description of Rachmaninoff’s harmonic language; yet, unlike
Cohn’s hexatonic cycles, they do not always occur separately from syntax—in
Rachmaninoff’s dense, layered harmonic environments, they may be quite simultaneous.
They are not smooth (in any expressive or rhetorical sense, at least), and “component
[pitch-classes] are certainly” not “equally weighted,” as they are in Cohn’s cycles.42 For
the sake of simplicity, Cohn’s term “hexatonic” is retained to refer to chromatic major-
third relations in general; but his terms “Northern,” “Southern,” “Eastern,” and
“Western” for the four possible systems are not used.43 At the same time, Bribitzer-
Stull’s point is not lost, and his observation that consecutive chromatic major-third
relations can seriously disrupt a tonal context is taken seriously.44 Bribitzer-Stull’s
research has been primarily on works from the common practice. By the Postromantic
era, the structures he describes represent a common way of going about harmonic
business—yet the disruptive effects of chromatic major-third relations are never fully lost
in Rachmaninoff’s works. (See for example the analysis of “Daisies” in Chapter 2.)
As discussed in more detail in Chapter 6, hexatonic structures are especially
important in the Rhapsody on a Theme by Paganini, participating on the small scale and

40
Kopp, Chromatic Transformations in Nineteenth-Century Music, 3.
41
Ibid., 263.
42
Cohn, “Maximally Smooth Cycles,” 13.
43
Ibid., 17.
44
Bribitzer-Stull, “The A-flat–C–E Complex,” 176-77.

139
the large. The introduction of hexatonic-type chromatic major-third relations as a kind of
hyperdissonant exaggeration in Variations VIII and IX of the work has already been
discussed (see the analysis of the passage in Chapter 2). Figure 4.32 shows a similarly
intense hexatonic structure in Variation XIII. At the beginning of the variation, a diatonic
tonic-dominant (i – V) alternation originating in the arpeggio motive of Paganini’s theme
(labeled x on the figure) and a strong HEX(1,2) cycle are combined. In fact, the material
labeled “1” in Figure 4.32 and the hexatonic material labeled “2” are in a state of friction
with one another throughout the variation. The friction is intensified by upper and lower
pedal points: A♮ in the upper register (marked “3” on the figure) and D♮ in the lower,
which combine to maintain a sense of the local tonic amidst the intense chromaticism.
Note that, following the precedent established in Variations VIII and IX, the hexatonic
structure is marked ff, supporting the rhetorical associations developed in Chapter 3 of the
dissertation. In Variation XIII, hexatonic organization is associated with a state of
heightened dynamic, textural, and expressive intensity.

Figure 4.32. Rhapsody on a Theme by Paganini, Op. 43, Variation XIII, analysis

78
!"#$%&!'()*+&,& 79

%
) !
!"#$%&'( *
# "&$### &### "###
( !" # #
&## & $##
# $# "&&### $&##'' # $### ###
' ## $ ### ##
# $$### #
! %
# " # # # ## ## # # # # # #
!! !
# &# #
%( "
(
" # ## # # ##&$# # # # # ### #
# $# # # # ##$# "# #
# # "# #
%
!"#$%&"'("()(*(+,%-($./-/(
+ ,- - +
!"#$%&!'()*+&-&

"&',/#0"&1($/&0"%&(2/$3//&(45(65(#&!(7

140
 In the second half of Variation XIII (rehearsal 34, repeated with slight alterations
at rehearsal 35), the hexatonic structure of the opening measures is mitigated—but, as
shown by the dashed beam on the upper staff of Figure 4.32, a remnant remains. (The
chord marked “*” on the figure is variable in the Variation, and Rachmaninoff’s full
score and two-piano reductions do not completely agree on its pitch-class content: in the
two-piano reduction, it contains A♮ and B♮ after rehearsal 34 and A♮ and B♭ after
rehearsal 35, while in the full score it contains A♮ and B♮ after rehearsal 34 and just A♮
after rehearsal 35. B♮ goes against a HEX(1,2) reading, but the overall content of the chord
suggests a connection with the hexatonic framework established earlier in the variation.)
Engaged dramatically in Variation XIII, hexatonic organization carries over into
the following variation, as shown in Figure 4.33. A HEX(0,1) relation substitutes for
dominant-tonic resolution to F major (the new local tonic) at the beginning of Variation
XIV, and the HEX(0,1) chord pair, C♯ minor – F major, is stated many times over the
course of the variation. As shown on Figure 4.33, arpeggio motive x (see again Figure 3-
32) is inverted in Variation XIV, preliminary to the more famous inversion of the entire
theme in Variation XVIII. At the same time, a new version of the original (non-inverted)
motive, marked y on Figure 4.33, is introduced. (The derivation of y from arpeggio
motive x is not obvious in Figure 4.33, but is instantly audible, especially when rhythm is
considered.)

Figure 4.33. Rhapsody on a Theme by Paganini, Variation XIII into Variation XIV

1234#5556$7/8( 1234#516$9-3),69.*36:72)*37)(

! # $## %##
! "

! " ## $ # # # # #
# "# # % # #

& # ## ## #
## $$## % ####
" " # $#
#
#
0

!"#$%&'( !"#$%&'(
)*+),-,*,-./ )*+),-,*,-./

141
 As detailed in Figure 4.34, melodic idea y, the HEX(0,1) structure with which it is
associated, and the underlying functional syntax model bear an interesting three-way
relationship. In the figure, (a) through (d) show the derivation of the HEX(0,1) relation
used in Variation XIV from a diatonic prototype. (Note that Figure 4.34d invokes the
dominant-subdominant hybrid function proposed by Swinden.) As shown in the figure,
pitch class D♭ in melodic idea y is treated by Rachmaninoff as lowered scale degree 6,
moving down to C♮ (scale degree 5). Hexatonic chord root C♯, however, is not treated as
lowered scale degree 6, and does not move down to C♮. In Figure 4.34e, melodic idea y is
split into two voices to make plain its content in relation to the HEX(0,1) relation
underneath it. Figure 4.34 reveals a momentary clash between hexatonic values and
diatonic values.

Figure 4.34. Rhapsody on a Theme by Paganini, Variation XIV, HEX(0,1) analysis

!
"#$ "%$ "&$ "'$ "($!)#*+#,+-.!/0)
! !" #
"# ## $$ "%$# # $$ "%$# # $$ "%$##%$ # $$ "%$##%$ # $$

$
#%$'$$
& $ $$ $$ $$ "$ %$ '$ $$
!
" "# # $ " $$ $$ '$
"4!5!6 "
!!1
23 7 '4!5!6
3 !"#$#" !
++!!!!!!)!!!!0!!!!!!!!!!!!!)!!!!!!!!!!!!0!!!!!!!!!!!!!!!)!!!!!!!!!!!!0 %&'"()*%+&

"# # $ # $ # $ %#&# $

Unlike octatonic structures, which are rare in Rachmaninoff’s works before the
late Russian period, hexatonic structures—including cycles of chords (or, on a larger
scale, keys) related by chromatic major third and substitution of hexatonic oscillation for
conventional dominant-tonic or subdominant-tonic progressions—appear fairly

142
frequently in works from the early and middle Russian periods.45 This is perhaps because
chromatic major-third relations are more generic than octatonic ones, and because
hexatonic structures and conventional tonal functions may be synthesized more easily
than octatonic structures and tonal functions. Figure 4.35 shows the use of HEX(0,1)
organization in the early Waltz, Op. 10, No. 2 (1893-94).
In the waltz, the secondary key area of D♭ major bears a hexatonic relationship
with the home key of A major, suggesting the HEX(0,1) cycle shown in Figure 3-36a. The
A-D♭ key relationship in an of itself is probably not enough to suggest that the work is
“hexatonic” in orientation; but the chords marked “*” on Figure 4.35 support a hexatonic
reading. The “*” chords are all seventh chords built on triads from the HEX(0,1) cycle,
including F major (which does not appear as a key area in the piece)—auxiliary seventh
chords that create local hexatonic relations that mirror the large-scale chromatic major-
thirds structure of the work. The “*” seventh chords are extracted in Figure 4.36b.; when
the seventh chords are considered, all three hexatonic “roots” (A, F, and C♯/D♭) are
accounted for.

45
Bribitzer-Stull includes the second and third movements of Rachmaninoff’s Piano Concerto No. 2, Op.
18 in his list of works involving chromatic major-third relations (“The A-flat–C–E Complex,” 186).

143
 Figure 4.35. Waltz in A major, Op. 10, No. 2, analytic reduction

97
124
78 81
139
69 195
33 57
1
1 238
 
          
   94
     
    
                    
     
          

 
         
*
2

144
*           *      
     
    


 
 

mf ff p * sfff

A major D  major A major

HEX(0,1) basis
 Figure 4.36. Waltz, Op. 10, No. 2, HEX(0,1) cycle and auxiliary seventh chords

!"#$%&'!()*#$$+,+-. !/#$01.213.4$05$6"-78)$9:;*()$<0;=

! #" $" %" #" #"" %" %" $" #""

! "" $ "" % "" "" %" " " %" $$"" % ""
# "" ## "" #% "" # "" " " # ""
& %" " "
" $ " $" "
" " #"

#
" !$ # "
#

Figure 4.37. Waltz, Op. 10, No. 2, auxiliary seventh chords and melodic details

!
!""#$%&'())(* !
$$
- ./
%
# $ "! & %%%% %" ' % % % %" $% % % %" $% % % %$% % % %$% % % % % % %$% %
% % $% % %% % ( (

%%+ %%+ %+ %+ %%+ %%+ -

!" !""#$%&#&"'#("% ")*&+),)
) $$$ ! * *
" %+
% %
% # * $ % %% * $ % %% , , &"
%+ + %% %+ %% %+ * %% %%
+ +
+ +

!
!"

-0/ -/.
. $$$
&"
# . ... %&"%% " % /%&" "
%
& " ( / & "%% " %( " "
&"%% / &"
[
%& " ( / %& " ( % %
% & %% % %% %% % % %% %
% % % % % % % % % %% %% %% %% %% $ % % $%%
# - - - -
% && " % & % && " % %+ %+
# .&& "" /&& ""
!""#$% !"
$
&&"" &"" "
) .. , &" ) $$ % %%
. . . &&" &" &" - & & +

Examples of C♯ and F seventh chords in the waltz are boxed in Figure 4.37. (The
C♯ seventh chord at measure 53 is superimposed above tonic A♮.) Figure 4.37 also shows
how the reciprocal nature of the underlying hexatonic structure is manifest in certain
surface melodic details of the work. Melodic cell x in measures 53 – 56 connects a C♯

145
auxiliary seventh chord to a restatement of thematic material in A major at measure 57;
note the connection of measure 57 to measure 1. The same melodic cell is adapted in
measure 135 and following to connect an F auxiliary seventh chord (following the D♭
major section) to restatement of thematic material in A major. An enharmonic pun
bridges the change of key signature, and again suggests the symmetrical, reciprocal
nature of the underlying harmonic basis: C♮-D♭-E♭ = B♯-C♯-D♯.46
The Hexatonic junctures of the waltz, though colorful, are not rhetorically marked
to any great extent. The chromatic major-thirds structure of the work seems on the
contrary a kind alternative tonal plan, not a disruption or intensification—very much as
Kopp would have it. By the late Russian period, however, the rhetorical associations
presented in Chapter 3 crystallized—equal-interval and diatonic structures are
increasingly differentiated, tension between them is emphasized, and explicitly layered
formations come to the fore.
In the first movement of Rachmaninoff’s Sonata No. 2 in B♭ minor, Op. 36
(1913), for example, hyperdissonance resulting from an interaction of hexatonic and
diatonic structures at the beginning of the recapitulation results in a climax event and in a
substantial modification of traditional sonata form tonal design and rhetorical strategy.47
The passage shown in Figure 4.38 contains a climax in the proper rhetorical sense: a
series of events arranged in stages of increasing intensity—in this case, increasing
chromatic intensity. The stages are marked 1 through 3 in the figure.

46
In connection with the present discussion, see also the Mazurka in D♭ major, Op. 10, No. 7. In that work,
composed around the same time and published at the same time as the waltz, the same hexatonic cycle
(HEX(0,1)) is used; but D♭ major is the global tonic, F major is the key of the middle-section digression, and
A major triads are used as auxiliary chords. See especially the last 25 measures of the Mazurka, in which
the A major auxiliary chord is tonicized, ffff, and an explicit HEX(0,1) cycle is presented as a kind of
summary.
47
The analysis is based on the revised version of 1931.

146
 Figure 4.38. Sonata No. 2, Op. 36, i, climax at recapitulation

/0 121 123

!
!"#$%&#'(")*%#$

" !"#$%&'(

+%*!&#%,-)&+*.

In Figure 4.38, the music is separated into layers. Layer C is a harmonic

framework; layer B contains the three highly unstable chords at the highpoints of the
three stages of climax; and layer A contains, simply, tonic elements B♭ and D♭.48 The
blagovest texture of the passage (alternating material in the high register and in the lower

48
It is worth noting that the three unstable “highpoint” chords in Layer B represent the three key areas in
which the second theme is heard in the Sonata: D♭ major in the first movement exposition, F♯ (G♭) major in
the first movement recapitulation, and F♯ (E) major at the end of the second movement.

147
register, creating a bell-like effect; see again the discussion of blagovest in the second
movement of The Bells earlier in this chapter) accentuates the highpoint chords in layer
B. The climax reaches its peak with the F♯ major triad inside stage 3. Maximal expressive
intensity is thus attached to a highly-charged chord that contains, enharmonically, the
defining third of the just re-established global tonic, B♭ minor. At the climax, a powerful
chromatic torque is applied to tonic elements.

Figure 4.39. Sonata No. 2, i, diatonic and hexatonic third relations in the exposition

!"#$%&"'()&*$!+(,"-#$'.&/,$/0-"'&()$&)$012(3&'&()

!4#$501"'()&*$/0-"'&()$&)$012(3&'&()$*(,"

!"#$%&'(

148
 Stages 1 and 3 of the climax are defined by different kinds of third relations: stage
1 contains a diatonic (or modal) minor-third relation, B♭ minor – D♭ major, and stage 3
contains a chromatic major-third relation, D major – F♯ major (that is to say, a HEX(1,2)
relation). As Figure 4.39a and 4.39b show, both kinds of third relation are established
earlier in the piece: the minor-third relation in the exposition, the hexatonic relation in the
exposition coda. (The diatonic minor-third relation is related to the modal peremennost
idiom discussed in Chapter 5; it is sufficient for now to simply regard it as differentiated
from the hexatonic relation.)
Figure 4.40 compares the beginning of the recapitulation to the exposition. As the
dotted lines show, stages 1 and 2 of the recapitulation and the music following stage 3 all
correspond to parts of the exposition. But climactic stage 3 is new. Stage 3 represents a
chromatic insertion that distorts what was in the exposition a plainly functional large
progression, tonic-predominant-dominant-tonic. The recapitulation moves from B♭ minor
to G♭ major for the second theme; this, too, could have been a straightforward, basically
diatonic course. The insertion of stage 3, however, means that an extraordinary chromatic
path is traveled instead. As a result, a larger chain of major thirds is suggested for marked
events in the recapitulation as a whole, as beaming in Figure 4.40 indicates.
As shown in Figure 4.41, the start of the recapitulation at measure 98, the climax
at measure 104, the return of the second theme at measure 112, and the coda at measure
124 all occupy nodes in a large chain of HEX(1,2) major thirds. Climax occurs inside the
D major node—the chromatically rough stage 3 climax, which contains the highly-
charged F♯ major triad. In his recent article on chromatic major-third relations, Matthew
Bribitzer-Stull notes that successive chromatic major-third relations can seriously disrupt
an ordinary tonal context.49 Stage 3 of the Rachmaninoff climax is such an event. There
is first a chromatic major-third move from B♭ to D♮ between stage 2 and stage 3, and then
a chromatic major-third relation inside stage 3, which distorts tonic elements B♭ and D♭.50
The F♯ major triad represents an unstable chromatic complication inside a larger unstable
chromatic complication.

49
Bribitzer-Stull, “The A-flat–C–E Complex,” 176.
50
In Bribitzer-Stull’s article, disruptive successive major-third relations occur at the same level of tonal
hierarchy. In the Rachmaninoff passage, they do not; but the principle—disruption crucial premises of
ordinary tonal design, in this case tonic elements—is similar.

149
 Figure 4.40. Sonata No. 2, i, comparison of exposition and recapitulation

highpoint chords (layer B)

chromatic M3
(dominant function)

150
chromatic M3
(dominant function)
 Figure 4.41. Sonata No. 2, i, schema of recapitulation climax and solution

151
HEX(1,2)
 Between measures 106 and 112, a “proper” pathway is found, to stabilize “F♯-
ness” into more legitimate G♭ major. The second theme is thus set in G♭ major in the
recapitulation, not in B♭. The need to discharge the lingering tension from stage 3 of the
climax is more compelling than the urge to flatten the recapitulation into a single key.
The large-scale tonal crisis inherent in sonata form is as a result extended past the
development into the recapitulation—it is in fact greatly amplified in the recapitulation,
creating a context for the unconventionally located tension climax and delaying full
solution of the tonal problem.51
At the bottom of Figure 4.41 is a diagram of the entire recapitulation, showing
how equal-interval chromatic pressure is associated with a climax event, and how the
tension is then released in stages over the rest of the movement. In connection with this, it
is worth noting that the disruptive hexatonic progression of the climax is essentially
reversed at the start of the coda in measure 124. As Figure 4.42a shows, hexatonic and
diatonic/modal relations from stages 1, 2, and 3 are engaged in reverse; and for good
measure, as Figure 4.42b shows, the diatonic/modal minor-third relation is heard one
final time at the very end of the movement. Significantly, neither the direct reversal at
measure 124 nor the final minor-third relation at the end of the movement happens in the
original 1913 version of the sonata; they happen only in the revised version of 1931,
which suggests that Rachmaninoff’s controversial revisions go somewhat deeper than
generally recognized.

51
In the first movement of Rachmaninoff’s Symphony No. 2, Op. 27 (1907), the tonal crisis in the
development is similarly extended into the recapitulation. In the symphony movement the mechanism is
considerably simpler than in the sonata movement: reprise of exposition material over a dominant pedal-
point.

152
 Figure 4.42. Sonata No. 2, i, hexatonic and diatonic/modal third relations in coda

(a) mm. 112 – 126

0)4*.7)-:3.04;
!"#$%&'
3)7./<*+)/0,/6;4*).7,$=1*456,%9
$()*+,-+./01,2/.3,1*4561,',470,89

(b) mm. 136 – end

!"#$%&"'()%!#*
)"&%+,$-"+!.+/*#$"%&

153
 Whole-Tone Structures (Interval 2 Basis) and Hybrid Structures

Strict whole-tone organization in Rachmaninoff’s works is comparatively rare,

perhaps because the whole-tone collection contains no major or minor triads, only
augmented ones, and because there are only two distinct transpositions of the
collection—WT0 and WT1, starting on C and D♭, respectively. The idiom therefore has a
certain stagnant quality. Only a few whole-tone passages have been identified thus far in
the dissertation. A brief WT0 passage was analyzed in first movement of the Symphony
No. 3 as part of a general trend toward equal-interval organization at an important
structural and expressive moment (see Figure 2.7); and a whole-tone ascent in the bass
was identified in the first movement of the Concerto No. 3 (see Figure 4.8), again as part
of a general trend toward increased symmetry of pitch organization associated with an
intensification leading to a climax event.
Figure 4.43 shows whole-tone (and octatonic) organization at the climax in the
middle section of the Etude-Tableaux in E♭ minor, Op. 39, No. 5. (See again the analysis
of the etude’s first section in Chapter 3.) The structures shown in the figure are not
entirely whole-tone: whole-tone scales are involved, and T2 operations, but triads outside
the whole-tone collection are used.52
The climax in Figure 4.43 is associated with both octatonic and whole-tone
structures, and with gradual intensification of register, texture and dynamics. Roman
numerals shown on the figure are in relation to the global tonic, E♭ minor. As suggested
on the figure, each stage of functional syntax is exaggerated by equal-interval structures.
(Not shown on the figure is the resolution of V7 to E♭ minor for the reprise at measure 53;
note that the entire reprise occurs over a post-climactic pedal point.)

52
Taruskin has identified a similar structure—a whole tone scale connecting triads related by chromatic
major third—in the overture to Glinka’s Ruslan and Lyudmila. See Stravinsky and the Russian Traditions,
261-62.

154
 Figure 4.43. Etude-Tableaux in E♭ minor, Op. 39, No. 5, climax in middle section

26 29 33 35 37 T3 38 39 T3

       
                                               
   
     
          
p
     poco a poco cresc.     
      
           
       
   
(OCT broken) 6
T2 T3 T3
5
V V
OCT(1,2)
V tonicizes B major

155
climax
41 42 43 44 45 46

T2 T2
chromatic ascent   
                    
     
     
 
ff WT1 WT0
                 
 
      
             

          
 

T2 T2
T2 II7
VI V7 V7
 In Figure 4.43, different kinds of equal-interval structure are used in close
proximity at a climax event. In the late Russian and exile periods, the strong association
between equal-interval structures and processes of intensification and climax results in
many situations where more than one kind of “fantastic” structure is articulated at the
same time. Such a case is shown in Figures 4.44 and 4.45, an analysis of measures 16–78
of the first movement of the Symphonic Dances, Op. 45. The dance is in ternary form
(A1–B–A2), with an introduction and a coda. (The introduction was analyzed in Figures
4.19 and 4.20. Note that the “arpeggio motive” from the introduction is retained in
Figures 4.44 and 4.45.) Figure 4.44 is an analytic reduction of the thematic exposition
(measures 16–40) in the first A section. As shown in the figure, thematic exposition is
associated with modal organization (as suggested more generally in Chapter 3)—
specifically, an Aeolian structure in which the tonic and (minor) dominant are entangled
until the end of the first phrase. Following this, between measures 29 and 32, a chromatic
structure leads to a local highpoint; as indicated on the figure, the highpoint suggests
hexatonic organization without committing to it.
Figure 4.45 is an analytic reduction of the measures which follow—that is to say,
the rest of the A section, including a local climax event leading into measure 79 (note the
crescendo to ff). The passage may be understood as involving a large-scale HEX(3,4)
structure, important nodes of which are connected by marked whole-tone ascents (such
that WT0 is filled out in the structural bass over the course of the section), and above
which is superimposed a cycle of diminished seventh chords. To put it more plainly, three
different kinds of equal-interval structure are combined in this section. None of them by
itself is sufficient to account for the structure or expressive content of the passage; but
collectively, as a kind of general trend toward harmonic symmetry leading to the A
section climax—all the more potent after the modal exposition in Figure 4.44. As a last
piece of evidence that equal-interval structures in Rachmaninoff’s works may lead to
unorthodox kinds of tonal tension, observe that the point of furthest tonal remove in
Figure 4.45 (measure 62, marked with an exclamation point on the figure) is a hexatonic
node that involves the tonic note, C♮, arrived at by whole-tone ascent in the bass, and
entangled with diminished seventh chord [1,4,7,10], creating a nonfunctional,
nonresolving ninth chord on the tonic root.

156
 Figure 4.44. Symphonic Dances, Op. 45, i, mm. 16–40, analytic reduction

^ 2 ^
1 2
18
(from Intro.) 25
       
    
               
        
16
Aeolian
arp. motive
             
 
             
      
    
           


T PD D

157
local highpoint
fication
intensi
^ ^
3 2 1 4
29 31 32 33
27          40
              
        
 
                          
 
cresc. etc.
ff dim. pp

            

            
     
             
    

PD D T
? HEX(0,1) basis
 Figure 4.45. Symphonic Dances, i, mm. 42–78, analytic reduction
^ º7: [0,3,6,9]
3 º7 overlays: º7: [0,3,6,9] º7: [2,5,8,11]
C minor tonic frame:
48 52 56
42 45
   
 
 
         

          
      
               
                  

WT ascent
HEX(3,4) basis: Ab C E

º7: [1,4,7,10] º7: [2,5,8,11] º7: [0,3,6,9]

58 60 62 68 71
         
        
 
                                     
      

158
              
                 
          
WT ascent
Ab C ! E

^ ^ ^
3 2 1
º7: [0,3,6,9] 78
75 76 79

71
 
                            
  
                   
        
cresc. sff

              

      
             
WT ascent
Ab C
PD D T
 Conclusion

The analyses in this chapter have suggested that identification of equal-interval

chromatic structures in Rachmaninoff’s works allows more meaningful interpretation of
large- and small-scale organization, and contributes to better understanding of expressive
trajectory and climax events. Idioms featuring symmetrical pitch organization are thereby
rehabilitated: genericized in much recent music theory, they become “fantastic” once
again when it is recognized that they serve specific rhetorical functions in the works
analyzed—intensification, climax, disruption—and are therefore strongly differentiated
from the underlying functional tonal basis and from modal structures. The analyses have
also shown Rachmaninoff’s connection with the post-Wagnerian chromatic tradition in
general and—perhaps more tellingly—with progressive Russian composers of the late
nineteenth and early twentieth centuries in particular.

159
 Chapter 5

Modal Structures

This chapter is devoted to detailed technical description of specific modal

structures that appear with frequency in Rachmaninoff’s mature works. Many of the
modal structures described may be considered extensions of recognized Russian idioms
with origins in liturgical or folk music; in some cases, the modal structures are more
generic. To avoid avoid the risk of suggesting ad hoc structures, this chapter is limited to
only four kinds of modal organization. I consider the following to be clearly defined,
particularly common, and structurally significant in the works analyzed:

1. Use of the traditional church modes (Dorian, Aeolian, Lydian, etc.) as

straightforward substitutes for conventional major/minor tonality.

2. Russian peremennost idioms and extended diatonic tertian structures that can
result from their application.

3. A distinctive melodic-harmonic idiom Taruskin has associated with the Russian

word “nega.” Nega is related to peremennost but has its own expressive and
structural qualities.

4. Phrygian organization, which has particularly complex structural implications and

which in Rachmaninoff’s oeuvre seems to be associated with “gypsy” music.

Previous treatments of modality in studies of Rachmaninoff’s music have been

limited to category 1 above. (Anatole Leikin’s brief comments on peremennost, discussed
below, are an exception.1) It is hoped, therefore, that the present chapter may provide a
starting point for more extensive work on the topic. Throughout the analyses in this
chapter, the general rhetorical associations established in Chapter 3 apply: modal

1
Leikin, “From Paganism to Orthodoxy to Theosophy,” 36-37.

160
structures in Rachmaninoff’s mature works are generally associated with introductory,
initiating, digressive, and/or post-climactic rhetorical functions in the context of the
section in which they are heard or possibly in the context of the entire work. Phrygian
organization is a special case, as described later in this chapter. Phrygian structures are
especially important in the Symphony No. 3 and the Symphonic Dances, excerpts from
both of which are presented at the end of the chapter.

The Church Modes

The familiar church modes require no special treatment in the present context
(again, with the exception of the Phrygian mode). Passages referable to church modes
were identified in several works earlier in the dissertation—e.g. Lydian organization in
the opening and closing measures of the song “From the Gospel of St. John” in Chapter
3. In many passages in the works analyzed, a structure in a church mode substitutes in a
clear way for a conventional functional tonal structure.
Figure 5.1b shows Aeolian substitutions for conventional tonic-dominant relations
in Variation VII of the Rhapsody on a Theme by Paganini. Variation VII contains the
first explicit statement of the Dies irae theme in the composition (the Dies irae is
indicated on the figure), and the modal structure can therefore be considered expository.2
The syntax of Paganini’s theme is shown in Figure 5.1a. (The repetition of the first four
measures of the theme is written out in Figure 5.1a to simplify comparison with Figure
5.1a; recall that in the Rhapsody Rachmaninoff invariably writes out the repetition to
multiply the opportunities for variation.) In Variation VII, modal inflection results in a
“neutral”-sounding treatment of the theme: the leading tone (scale degree 7, G♯) is
eliminated in the Aeolian mode.

2
The Dies irae chant is in the Dorian mode. However, Rachmaninoff’s setting of its first seven notes in
Variation VII includes F♮ at prominent points, suggesting A Aeolian rather than A Dorian.

161
 Figure 5.1. Rhapsody on a Theme by Paganini, Op. 43, analysis of Variation VII

(a) Paired i – V gestures in Paganini’s theme

!"#$#%&%&'()*"&%%#()'+%,

$ ($ $ ($
# !" $% & $% $ $ $ $ & $ $'$'$ $ $ & $ $ $ $ $ $) * $% & $% $ $ $ $ & $$'$'$ $ $ & $$ $ $ $ $) *

! " ! " ! " ! "

$-&"#.)/#0%+"#0

(b) Exposition of Dies irae in Variation VII and Aeolian substitutions for i – V

!"#$%&'%((%)&*&+#',%&'%-#.*+%/

01 !"#$#%&%&'()*+',&-&#,.

! "
/&#0*&"1#
! "" """ """ #"" "" """ $"
""
""
" """
" "
"" #$$"
""
" "" """ """
" " "

%
"
" """" " """" "
"""" " "" "" "" " "" " " " " " "" " "
" " """
""
"
"" "
2 7 2 !#+34/ 666 56
!$1&"#,*2#0%3"#0*4#5'+#*0#63#(%&17. 666 2
!"#$%&'()*+),%,*,%#')
&8%.&5

Figure 5.2 shows a similar modal substitution at the end of Variation VII, this
time as an explicit reharmonization of the motive from Paganini’s theme; again, the
leading tone is eliminated. Recall that hexatonic structures are noisily articulated in
following two variations. (See again the analysis of Variations VIII and IX in Chapter 2.)
Variation VII (modal) is expository, initiating, neutral; Variations VIII and IX
(hexatonic) are intensifying. The dynamics marked throughout these variations support
this rhetorical framework—modal structures are generally p or pp, while hexatonic
structures are louder.

162
 Figure 5.2. Rhapsody on a Theme by Paganini, end of Variation VII into Variation VIII

'()%*+%!"#$%!&& !"#$%!&&&

,- %&'&("("

! """" "" " " " "

"" " "
! " " " " ""
"
! )*+,-./0
!"#$
!! "#

# " "
"
" " " " " "
" " " "

/ &&&. /
!"#$%&'()*+),%,*,%#'
%+*#%!

A similar but more rapid juxtaposition of modal and symmetrical chromatic

structures within a similar rhetorical framework may be heard in the introduction to the
first movement of the Symphony No. 3, as shown in Figure 5.3. The symphony’s
Phrygian motto theme is heard, unharmonized, on A♮ in the opening measures.3 Modal
organization gives way to chromatic organization as the passage intensifies—tones from
the Phrygian “cell,” A, B♭, and G act as pitch-class pivots between the modal opening
and the chromatic structure in measures 6 – 8. As shown on the figure, in measures 9 and
10 a HEX(0,1) relation substitutes for V – i; then, an Aeolian echo initiates a decrescendo
preliminary to the (modal) exposition of the primary theme in the following measures. As
in Variation VII of the Rhapsody, the Aeolian substitution for conventional V – i has a
de-intensifying effect, especially in proximity to the intense HEX(0,1) version.

3
The Phrygian motto is discussed again at the end of the chapter, and more fully in Chapter 6.

163
 Figure 5.3. Symphony No. 3, Op. 44, i, analytic reduction of mm. 1–10 (introduction)

de-i
nten
s ifica
romatic tion
ecomes ch : ch
tion: modal b rom
intensifica ati c be
com
es m
odal
1 6 10

tonic frame ^ ^
motto   ^6  ^7 1 ^6 ^7 1
 
  
            
                                  
pp 7 sff p
f
      


164
    
        
 
          
  
V i HEX(0,1) Aeolian

(Phrygian cell) PD D T D T D T

coalesces into
functional harmony

modal exposition of primary theme follows

 Peremennost, Diatonic Oscillation, and Diatonic Stacks

Certain unconventional harmonic structures in Rachmaninoff’s mature works may

be understood as based on his idiosyncratic extensions of the so-called ladovaya
peremennost (or simply peremennost) in traditional Russian music. The term is translated
as “modal mutability” by Anatole Leikin and as “tonal mutability” by Richard Taruskin.4
Leikin, citing Russian musicologist Andrey Myasoyedov, defines peremennost as a
shifting of harmony “between at least two equal tonics.”5 Taruskin’s definition is similar:
“the quality…whereby a tune seemed to oscillate between two equally stable points of
rest, as it were two ‘tonics’.”6 Leikin, again citing Myasoyedov, suggests that
peremennost developed from a kind of “protoharmony” found in “older liturgical
chants”—a system of melodic organization in which each of four pitches “can and does
carry the function of a temporary ‘tonic’ in a melody” so that there is “no single unifying
center, since each member of the protoharmony tends to be equal and independent.”7
The quotation marks placed around “tonic” by both Taruskin and Leikin are
significant; for “tonic” is in general too strong a word when peremennost idioms are
applied in art music contexts (as opposed to genuine liturgical or folk contexts). Rather,
peremennost idioms involve some kind of oscillation between or superimposition of
diatonically but non-functionally related chords, one of which is tonally more important
than the others, but which together form a kind of harmonic network that is distinctly less
center-specific than conventional tonal syntax and in some cases even approaches a
limited form of pan-diatonicism. In the present context, “peremennost” refers to a family
of related, non-functional diatonic structures. Peremennost idioms in Rachmaninoff’s
usage usually involve chords related by diatonic third.

4
Leikin, “From Paganism to Orthodoxy to Theosophy,” 37; Taruskin, Defining Russian Musically, 133.
The term is used with or without an apostrophe—peremennost or peremennost’.
5
Leikin, “From Paganism to Orthodoxy to Theosophy,” 37, citing Andrey Myasoyedov, O garmonii
russkoy muzyki (Moscow: Prest, 1998), 33-34, 49.
6
Taruskin, Defining Russia Musically, 133.
7
Leikin, “From Paganism to Orthodoxy to Theosophy,” 36, citing Myesoyedov, O garmonii russkoy
muzyki, 18-21. Leikin notes that Myasoyedov has identified protoharmony in several Rachmaninoff’s
middle-Russian period works, but I have not had access to this research. Leikin also suggests that
protoharmony may be the basis of the emphasis on “plagality” in Russian music in general and
Rachmaninoff’s music in particular (“From Paganism to Orthodoxy to Theosophy,” 37), which agrees with
the my finding that peremennost idioms intensify plagal action in several passages analyzed.

165
 Although peremennost idioms are like “fantastic” chromatic idioms in that they
are based on oscillation and superimposition of tertian sonorities as opposed to
functional, goal-oriented syntax patterns, they differ from “fantastic” structures in two
important ways:

1) Interval content. Peremennost idioms do not involve equal-interval structures, but

rather diatonically related triads (and seventh chords).

2) Rhetorical associations. Peremennost idioms, with few exceptions, generally have

introductory, expository, or post-climactic functions whereas equal-interval
structures (octatonic in particular) tend to be intensifying and climactic.

Central to peremennost is the articulation of a kind of melodic-harmonic structure

in which a specific pitch center is de-emphasized while a diatonic basis remains clear.
The concept of center-less (or multi-center) tonal systems figures prominently in a wide
range of scholarship on Russian liturgical, traditional, and concert music. In his
introduction to the Musica Russica edition of Rachmaninoff’s complete sacred choral
works, Vladimir Morosan notes that “equal emphasis between a key and its relative major
(or minor) is frequently found in both Russian Orthodox liturgical music and Russian
music in general, to the point that it may be deemed a stylistic trait.”8 The late nineteenth-
and early twentieth-century Russian musicologist Stepan Smolensky developed a pattern-
oriented theory of mode for Russian liturgical music. In Smolensky’s view, a mode is
determined not by a “final” note and a scale built around that final, but rather by a set of
characteristic melodic patterns, which might be organized around one or more of many
different finals.9 Rachmaninoff knew Smolensky well and dedicated the All-Night Vigil,
Op. 37 (1915) to his memory. (Yekovlev, writing in 1911—before the composition of the

8
Morosan, ed., Sergei Rachmaninoff: The Complete Sacred Choral Works, The Monuments of Russian
Sacred Music, vol. 9 (Madison: Musica Russica, 1994), lxxiii (footnote 111).
9
Smolensky’s theories of Znamenny chant were introduced to English readers in Alfred J. Swan, “The
Znamenny Chant of the Russian Church,” The Musical Quarterly 26 (1940): 232-43, 365-80. Smolensky’s
pattern or motive-oriented rather than scalar view is similar to to Gustav Reese’s statement that a mode “is
composed of a number of MOTIVES (i.e. short music figures or groups of tones) within a certain scale”
(Gustav Reese, Music in the Middle Ages (New York: W. W. Norton, 1940), 10).

166
All-Night Vigil—offers the tantalizing observation that “Rachmaninoff is under the
influence of the theories of S.V. Smolensky,” but adds nothing more.10)
Similar multi-tonic or multi-center interpretations of tonal structures in
nineteenth-century works by non-Russian composers have been presented by a number of
scholars. The well-known book The Second Practice of Nineteenth-Century Tonality
takes the “double-tonic complex” as one theoretical and analytical starting-point.11
Charles Rosen has suggested that tonal structures in nineteenth-century European music
generally may be understood as involving a conceptual fusion of relative major and
minor, greatly enlarging the number of possible tonal structures while at the same time
reducing the traditional tonal polarity between relative keys.12 However, peremennost
idioms as described above differ from the generic tonal relationships described by Rosen
et al. in three important ways:

1. Fluctuation between members of the peremennost pair or group is considerably

more immediate and explicit in Russian music than it is in mainstream European
music of the nineteenth century. This recalls Taruskin's observation that the
oscillations and rotations of chromatic third relations are more explicit in Russian
music than in Western European music. In some cases, members of the
peremennost pair or group are superimposed, resulting in extended diatonic
“stacks” that to my knowledge have no real counterpart in Western music of the
nineteenth and early twentieth centuries. As mentioned above and shown in
several analyses below, a limited pan-diatonicism can even result.

2. Specific melodic and harmonic idioms—in the works studied, often ostinato
patterns—are associated with peremennost, and these can retain their identity
even when used in complex, compound harmonic environments.

3. A significant differentiation between peremennost-type structures, the chromatic

structures described in Chapter 4, and the underlying functional tonal basis is
maintained in a majority of cases in the works studied, whereas in Rosen’s
generalized formulation, no rhetorical differentiation is suggested or perhaps even
possible.

10
Quoted in Stuart Campbell, Russians on Russian Music, 1880–1917 (Cambridge: Cambridge University
Press, 2003), 184. Yekovlev’s comments are not about the All-Night Vigil but about the Liturgy for St. John
Chrysostom, Op. 31, then recently composed.
11
William Kinderman and Harald Krebs, eds., The Second Practice of Nineteenth-Century Tonality
(Lincoln: University of Nebraska Press, 1996); Bailey, “An Analytical Study of the Sketches and Drafts”;
Bailey, “The Structure of the Ring and Its Evolution”; McCreless, Wagner’s “Siegfried”: Its Drama,
History, and Music.
12
Charles Rosen, Sonata Forms, Rev. ed. (New York: W.W. Norton, 1988), 368-69.

167
 David Cannata, drawing more on Robert Bailey’s theory of double tonic
complexes than on anything in Russian music theory, has identified structures involving
multiple tonal centers on a large scale in several of Rachmaninoff’s large concert works,
suggested that Rachmaninoff used a number double-tonic complexes involving relative
keys in his concert works and that he “equated relative keys to an advanced degree.”13
Although I suggest that in the strong “intra-tonal” contexts of Rachmaninoff’s
compositions there will rarely be any real ambiguity about a work’s governing or global
tonic, Cannata argues that the global tonic is in fact uncertain in some works until the
advent of some clarifying event (generally late in the work). I believe he overstates the
case here. I prefer the view, explained in Chapters 1 through 4, that problematization of
the tonic is not ambiguity, per se, but, rather, a kind of hyperdissonance resulting from
tension between different components of a compound melodic-harmonic environment—a
Postromantic structural and aesthetic idiosyncrasy that occurs as known tonal premises
are disrupted by unconventional tonal events. Whichever view of large-scale tonal design
is accepted, however, “de-centered” harmonic structures referable to peremennost are
fairly common in Rachmaninoff’s mature works. Such structures may involve any or all
of three specific techniques:

• Oscillation between and/or superimposition of diatonically related triads,

often with a melodic ostinato.14 Such oscillations often appear in the opening
and closing measures of a composition.

• Extended tertian structures—“diatonic stacks”—resulting from the

superimposition of two or more diatonically related triads. Often, these may
be interpreted as elaborations of the subdominant in structural plagal
structures.

• Modal reharmonization: statement of a melodic segment in one diatonic

harmonization, followed by restatement of the segment at its original pitch
level but with a different diatonic harmonization that emphasizes a different
local pitch center, such that a larger-scale oscillation between pitch centers
occurs around an unchanging melody.

13
Cannata, “Rachmaninoff's Concept of Genre,” 72.
14
Taruskin has suggested that Stravinsky’s well-known penchant for ostinatos was derived from Russian
folk music models (Stravinsky and the Russian Traditions, 957, 961). My analysis of a peremennost
ostinato in the first of Rachmaninoff’s Three Russian Songs, Op. 41 (see Figures 5.8 and 5.9) suggests a
similar connection between peremennost ostinatos and folk music.

168
 Note that in this conception peremennost-derived structures do not necessarily
involve “tonic(s),” specifically, but rather a conceptual fusion of tertian sonorities in a
variety of melodic-harmonic contexts. In the works studied, a peremennost idiom may
occur inside any stage of functional syntax, or it may resist a syntactical interpretation
altogether. In Rachmaninoff’s works, superimposition of tertian sonorities, as opposed to
oscillation between them (that is to say, vertical rather than horizontal peremennost), may
be considered a development in the late Russian and exile periods, as may the limited
pan-diatonicism that results from the peremennost-based extended tertian stacks and
reharmonization techniques described below.15
A number of structures involving peremennost may be found in passages
analyzed earlier in the dissertation. The diatonic/modal minor-third relations discussed in
the analysis of the first movement of the Sonata No. 2, Op. 36 in Chapter 4 may be
interpreted as a simple peremennost-based oscillation/superimposition. (See again
Figures 4.x through 4.x; B♭ minor and D♭ major are the triads involved; they are clarified
by a two-layer, bell-like blagovest texture.) Following closely the rhetorical associations
laid out in Chapter 3, the peremennost idiom in the sonata analysis initiates the
recapitulation and concludes the coda, while the more harmonically intense hexatonic
relation is associated with climax. The analysis of the first section of the Etude-Tableaux
in E♭ minor, Op. 39, No. 5 in Chapter 2 suggests peremennost-based diatonic
reharmonization of a melodic segment to emphasize different local centers without,
however, changing the melodic segment itself. (See again Figures 2.x and 2.x.) In the
analysis of the A section of the first Symphonic Dance in Chapter 4, the C minor tonic
and the minor dominant are superimposed in an Aeolian passage. (See Figure 4.x.) A
number of more striking peremennost structures are described below.
The last movement of The Bells features a particularly clear peremennost
oscillation as an ostinato in the opening measures. Figure 5.4 is an analytic overview of
measures 1–19. Although a voice-leading reduction of the oscillation might suggest its
origin in a neighbor figure (G♯–A–G♯), examination of the full score shows that
Rachmaninoff took pains to emphasize the chords of the oscillation (a rocking back and

15
Joseph Straus has suggested that overlapping or superimposed triads are an important component of
Stravinsky’s harmonic vocabulary, even at the level of deep structure. See Straus, “Stravinsky’s ‘Tonal
Axis’,” Journal of Music Theory 26 (1982): 261–290.

169
forth between C♯ minor and A major triads), not the abstract neighbor figure. The 5-6
contrapuntal motion has been hypostasized in a repeating chord pair. The oscillation itself
has several components: a drone in the harp, a layer in the upper strings, and a layer in
the lower strings. On Figure 5.4, the peremennost pattern (marked pp), which supports
statements of thematic material in the english horn, is thrice interrupted by triads in the
winds: A minor – F minor – D minor, marked forte. (A fourth disruption, involving a B♭
minor triad in measure 20, is shown in Figure 5.5.) The disruption triads bear various
chromatic relationships with the C♯ minor tonic, and chromatic third relations with each
other. Note that the triads on F and D are foreshadowed underneath the A minor triad in
measure 6, drawing the chromatic disruptions into an especially close association that
will bear climactic fruit later in the movement. As indicated on Figure 5.4, falling
contours characterize the opening measures of the movement. This may be heard most
clearly in the three english horn phrases shown in the figure (the descending melodic line
in each phrase is beamed); but the diatonic oscillation in the strings and harp is also
downward-oriented, as is the trajectory of the three chromatic disruption chords on a
somewhat larger scale.

170
 Figure 5.4. The Bells, Op. 35, iv, analytic overview of mm. 1–19

connect to Climax 2

chromatic disruptions

1 2 5 7 12 15 17 19

1 2 3
  
  
f f
f

171
(strings + harp) (english horn)
             
  

     
                              
        
  
         
    
pp pp
          
pp pp
diatonic oscillation diatonic oscillation diatonic oscillation

C  minor

falling contours
 Figure 5.5. The Bells, iv, analytic overview of mm. 20–138

gr
ad
ua
ld
ec
lin
from chromatic disruptions ing
ac
tio
nt
oC
climax 1 climax 2 #m
ino
r
113 117

4 Poco più mosso

theme d theme d 138
Allegro
20 24 50 54 70 93
       
ed ed ed
itt itt itt
               
m

om om
 
l l lo
ria ria ia
e e er
  
at at at

m m m

172
theme d theme d ff ff

120-121
       
    

     
          
       
 
     res. to
      D minor
    
 
ff
                     
triad

sf p
WT0
diatonic oscillation (expanded) diatonic stack
diatonic stack
iv (ext.) V
non-functional
SD
C  minor i
rising contours maximal chromatic pressure
in relation to tonic
Several elements from the opening passage are developed climactically later in the
movement, as shown in Figure 5.5. (As indicated on the figure, several passages are
omitted to save space. Changes of key signature are not shown, to make plainer the
relationships of all events to the global tonic, C♯ minor.) At measure 24, the diatonic
oscillation from the opening measures is expanded. This expanded peremennost structure
leads to the first climax event at measure 54. From measure 50 onward, contours
generally rise as dynamic levels generally increase. Dies irae-related theme d (see again
the analysis of the second movement of the work in Chapter 4) returns, leading at
measure 54 to a climactic extended diatonic stack—vertical peremennost as a
culmination of the horizontal peremennost that has characterized the movement thus far:
C♯ minor and A major triads (the two members of the oscillation at the beginning of the
movement) above F♯, suggesting a diatonic elaboration of the subdominant.
After this, an increase in tempo leads to a second, more powerful climax event
beginning at measure 113. As indicated on Figure 5.5, the second climax event may be
interpreted as involving greatly enlarged, intensified versions of the first three chromatic
disruptions heard at the beginning of the movement (A minor, F minor and D minor). The
F minor sonority is heard only in passing; but the A minor sonority is intensified by its
own peremennost diatonic stack, and the D minor sonority is intensified by a WT0
structure that acts as a kind of large equal-interval appoggiatura. The second climax
event, then, may be interpreted as a synthesis of the peremennost structures from the
movement’s exposition and intense chromatic structures more typical of Rachmaninoff’s
approach to climax. As shown in Figure 5.6, the movement closes in D♭ major—and the
final event in the work synthesizes the fourth chromatic disruption triad (B♭ minor; see
again Figure 5.5) and a peremennost-derived diatonic stack in a culminating plagal
gesture.

173
 Figure 5.6. The Bells, iv, analytic reduction of mm. 138–end

! """" # # # #
!"# !$"
)*+,

! ## ## $$$$$
## ## /011&()*+,(-,.%

!"#$%&"'(%)'"++#$"%&
% " "" # # # # # # # # # # # # #"#&# # &# #
" " #
# # # # # &# $$
# # # #
# # # # # # # & # &# # # # # # $ # $ $ $

$
"
,(((-"&%.

!$% %&'&()*+,(-,.% !&(

! $ $ # # # # # # # # # ##$$ # #$$##$$ # &# &# #$$"## $# "# &# "# &# &# # $# # # # & #
&# &# #
!$ $ $ "# & # "# & # &# " #
!"#$%&"'(%)'"++#$"%&
# # &# "#
/011&()*+,(-,.%

%$ ## # # #
" $ $$$
# &# $#
# # "# &# $# '

#"
"
$ $
$
/(((-#0%.

/&(01*232(-*&(*-42*5,%6
!&' !&%

! $$ $ # # $# # # # # # # # ###
# # # ####
! $$ #

# # ##
% $$ # # ## #
" $$$ # #

!"#$%&"'()$#'*

!" $

174
 Peremennost structures are quite common in the late Russian and exile periods.
Although in Rachmaninoff’s oeuvre peremennost is removed from its folk and liturgical
associations to a large degree, the appearance of particularly explicit peremennost idioms
in several liturgical and folk-based compositions suggest that the tether is not completely
severed. Figures 5.7 through 5.9 show that peremennost is a main structural component in
the first of the Three Russian Songs, Op. 41, No. 1 (1926), “Across the River.” The
composition is a setting of the folk song “Cherez rechku,” which, according to Barrie
Martyn, tells “the pathetic tale of a drake escorting a duck over a bridge; the duck
becomes frightened and flies away, leaving the drake forlorn and weeping.”16
The five phrases sung by the chorus of men’s voices are shown in Figure 5.7.
Note the gradual increase in tessitura, culminating in pitch E4 at rehearsal 8. Figure 5.8
contains an analytic reduction of the setting of the first three phrases. A peremennost
oscillation between E minor and C major is established as an ostinato in the opening
measures. Although there is little doubt that E minor (or E Aeolian) is the tonic of the
song, the C major component of the peremennost oscillation is solidified to a large degree
between rehearsal 1 and rehearsal 3, as shown on Figure 5.8. Between rehearsal 6 and
rehearsal 8, a limited pan-diatonicism emerges from the peremennost structure.
The climax of the song occurs with the choral highpoint at rehearsal 8. As shown
in Figure 5.9, the ff arrival of E4 in the chorus is distorted by an OCT(0,1) structure. The
OCT(0,1) structure, which incorporates a C major-minor seventh chord (first heard at
rehearsal 5, then more powerfully after rehearsal 8) may be interpreted as an outgrowth
of the C major member of the peremennost oscillation. The hyperdissonant clash at
rehearsal 8 between the arrival of tonic highpoint E4 in the chorus and the OCT(0,1)
structure is therefore a climactic compounding of equal-interval chromaticism and
peremennost. The diatonic oscillation returns with choral phrase 5 at rehearsal 10, as the
two diatonic chords (E minor and C major) are superimposed. E minor emerges cleanly
only at the very end of the song. Following closely the general rhetorical associations laid
out in this study, peremennost is associated with the opening (introductory, expository)
and closing (post-climactic) portions of “Across the River,” while an equal-interval
structure characterizes the climax event.

16
Martyn, Rachmaninoff, 309.

175
 Figure 5.7. Three Russian Songs, Op. 41, i, melodic highpoints in the five chorus phrases

B3

2
  
1
                            
pp

C4

4
        
                  
2   
p

D4

176
        
3
                     
mf

E4 hyperdissonant
8 climax of song
7   
   
4
                              

mf cresc. f mf cresc. f dim. p

10
  
5
                         
p
 Figure 5.8. Three Russian Songs, i, analytic overview through r. 7

peremennost
ostinato 1
2 “C major” 3 4 
      
        
 
     
            
      
    


  
      

etc.
[chorus] [chorus] etc.
     
             

                  
    
pp
      
pp E minor (Aeolian)

177
limited modal pan-diatonicism

5 6 7
                     
  
                  
                

[chorus] [chorus]
 etc.      etc.
                        
  
[no real bass] mf


(to OCT climax)

 Figure 5.9. Three Russian Songs, i, analytic overview of climax through end

hyperdissonant climax
Arrival on E4 in chorus phrase 4 distorted by intense chromatic structure

8  9 10 11
               

 
E and C
      
 
            

    
  
 superimposed
    

ff [chorus]
dim. pp
  
 

178
     
 
 
  
 
  
     
   
ff dim.
pp peremennost
ostinato
OCT(0,1)
(OCT broken)

equal-interval structure

modal structure
 Figure 5.10. All-Night Vigil, Op. 37, v , analytic reduction of phrases 1 through 4
2 6 10 11 12 14 16 17

 
  
     
     
                             
phrase 1 phrase 2 phrase 3
      
       

      
         
 
ppp p
i iv
peremennost oscillation
phrase 4 (concluded)
22
21

179
Climax
20
17 19
              
        
t
l ascen
gradua

           
  
      
       
phrase 4 (canonic) 
                
                        
     
f ff
 
mf

III
iv i V
(chromatic activity)
HC
 The best-known number in Rachmaninoff’s All-Night Vigil, Op. 37 (1915)—No.
5, “Nyne otpushaeshi,” a setting of the Nunc dimittis (Luke 2:29-32)—features extensive
use of a similar peremennost oscillation in the opening and closing measures, contrasted
with a chromatic climax event. Figure 5.10 provides an analytic reduction of the first four
phrases of the work. As shown in the figure, a peremennost oscillation—caught midway
between B♭ minor (i), G♭ major (VI), and D♭ major (III), as it were—is established as a
wordless ostinato in the opening measures. In phrase 4, the text “which Thou hast
prepared before the face of all people” is set canonically, and increasingly chromatically,
leading to the climax at measure 22, where the D♭ major component (III) of the
peremennost oscillation emerges, ff, as a highpoint on the way to a straightforward half
cadence. As shown in Figure 5.11, peremennost returns in phrases 5 and 6. Figure 5.12
gives an overview of the entire composition.

Figure 5.11. All-Night Vigil, v, analysis of phrases 5 and 6

./-"+$)0 ./-"+$)1

! "" " ## ## # # "## #

)* )+ ), *- ** *. */

! " " # # ## # ## "## " ## ## # #$$# # # #

#
# ###### # ## ### # # # #
# $ ## ## " # ## ## $ ##
% ""
" """
# # # # ##
# #
#
!
# # # # # #
# #
#
!!
!!!
! "# !
!"#"$"%%&'(
*+!'(("&'*% !"#$%&'"()!(*+,-$

180
 Figure 5.12. All-Night Vigil, v, analytic overview

66

!"#$%&

!"#$%&'()&*)+$&,-
%.,/!&'(+"/0
'/,$,&'(-1"/-!",-
2/3-"1()/1!$,&'(1).-)!
!/4&!/#(')1$!/-&'(&,-",3&-.
+"1"!",,$3-*$+,&%%'-&$( +"1"!",,$3-*$+,&%%'-&$(
!"#$%&'() !"#$%&'()

+)1/3"5 ' ( ) * + ,
...
# - # /%012/1
!!! ! "# # ## ! !! !!!

In Rachmaninoff’s late Russian and exile compositions, peremennost techniques

reach a point where the diatonic members are freely superimposed and even used in place
of one another. Figure 5.13 is an analytic overview of the opening measures of the second
movement of the Sonata in B♭ minor, Op. 36. Measures 1–6 of the movement establish D
major as a dominant-function sonority.17 As shown in Figure 5.13, this resolves not to G
major but to E minor for the start of “phrase A” (the start of the movement proper)—that
is to say, the dominant of G major is used directly as the dominant of peremennost-
related E minor. E and G chords are then interchanged and superimposed throughout
phrase A and phrase B. Note in particular the structure of phrase B—G major at measure
16, then G major and E minor superimposed at measure 18, then E minor and A minor
superimposed before the arrival of the dominant at measure 20. As shown in Figure 5.14,
the climax later in the movement involves a similar but more intense peremennost-type
superimposition as an elaboration of the subdominant, leading to the major tonic (E
major) at measure 64. (This is one of comparatively few climax events featuring modal
rather than chromatic structures in Rachmaninoff’s mature works.)

17
Note that chords on F♯ and D bookend the introduction, recalling sonorities from the first movement
climax event. See again the analysis of that movement in Chapter 4.

181
 Figure 5.13. Sonata No. 2 in B♭ minor, Op. 36, ii, analysis of mm. 1–23

1 6 phrase A (repeated, ornamented) 11

introduction
       
                                    
              
    
p
     
         
 

    
          
          
   

V V

of G major i of E minor

phrase B

182
15 16 18 20 22 23

         



              

                    
                             

                                      
                          
p p pp
 
        
   
    


V
of E minor peremennost (G major/E minor) V
of G major

V i / III
of G major
followed by
phrase A (ormamented)
m.24-
 Figure 5.14. Sonata No. 2, ii, analysis of climax in mm. 53–62

climax

53 54 55 58 61 62 63 64
     

     
      
etc.
E minor
,
etc.

183
            
           
    
     
C major
pp peremennost oscilation poco a poco cresc. ff dim. pp
  
   

material omitted
     
                   
  

peremennost stacked triads A minor

I

iv (ext.) (major tonic)
 A peremennost-based superimposition in the transition from the first A section to
the B section in the first movement of the Symphonic Dances, Op. 45 is shown in Figure
5.15. As in the Sonata No. 2 passage above, the peremennost event may be interpreted as
an intensification of the subdominant.

Figure 5.15. Symphonic Dances, Op. 45, i, analysis of transition, mm. 91–98

$%&'() +-/0-'()
*'+$,-.(% 2-0/%+.-.(%3 1'+$,-.(%

!"#"$"%%&'()!"#$%&'#(!&)&(*
! "" # # ## ;> # # # # %# # # %# %# # %# %# ;<&#
$$
" %#%##&# %#%##&# ;=
! ## $ $ # # %# %#
(3($

""
"! "
&#& #%# %# %#&#
# # %#
450%&*$)

" &%%### %&&%####

!09(#-(*$
!" " &# &#
%# & # %#
+,-%./&0*1
+&21
&
6777'&*(%8 !" #

*+#,-./%)#"'&01(.&%'
/%2)34)$.%&#)5.67
"$!+/'.8"2)(+#&1-+&1()9)'":(.&%

Peremennost provides a basis for understanding certain unconventional harmonic

events in works otherwise not strongly characterized by modal structures. Figure 5.16
shows an unconventional resolution to tonic in the well-known “Vocalise,” Op. 34, No.
14. As indicated on the figure, in measures 34–35 of the song, a tonicization of E major
(III) is strongly implied; but the expected resolution to E is denied, as the dominant of E
major instead resolves directly to the tonic C♯ minor at measure 36 in a manner
reminiscent of the resolution of D major directly to E minor in Figure 5.13. It is a

184
moment of considerable expressive weight in the song: an expected blossoming into the
major mode is undercut by a peremennost-derived substitution. The perfect authentic
cadence that follows in measure 37 occurs entirely in the shadow of the peremennost
event.

Figure 5.16. “Vocalise,” Op. 34, No. 14, peremennost resolution in mm. 31–36

% %% !" ( ( ( ( ( # ( "
!" !
(
$% ' ( ( ( (((

!
&

%%
$ % % (( ) (( ( (" (( ( !" (( (( ( (( (( ( # &( ( ( ( ( ((( ((
"#
( ( (
( * ( ( ( ( ( ( ( ( ( ( ( ( (( (( (( ((
!
, ( ( (
+ %%%% (( ( ( ( (& ( ( (
!" (& ( ( ( ( ( ( (
#& (( ( (( (
*
$

% % (" ( & ( ( ( ( ( ( ( ( ( (" ( ( (

$%%

!
* *
#$%&#'

%%%% & ( ( ( " " ( ( ( ( (( (( (" -( ( %.(

$ (( (( (( (( (( (( % (( ( ( - ((( (( % (((( ((( - ((( *
( !
(( ( ( ( ( ' ( (
( ( ( ( ( ( ( & %( (
!
%
+ %% % &( ( ( ( (
( ( (
-( ( ( (
& (( (
######

3
#$$" ! !"#
(
)
*
*+,$-./0$1&/$&#2&(&0
%&' #" !!" #"
{
]

$%&'&(&)))

A final example of peremennost demonstrates the large-scale implications that the

idiom may acquire in the complex contexts of Rachmaninoff’s mature works. Figure 5.17
shows the exposition of the second theme area in the third movement of the Symphony
No. 3, Op. 44. (The symphony is discussed more fully in Chapter 6; several passages
have already been analyzed in the dissertation.) As shown in Figure 5.17, the material is

185
layered: the melody strongly implies C♯ minor, and even introduces the leading tone of
that key (B♯). Underneath this material, A major and, at rehearsal 77, E major are
superimposed—that is to say, the notes of the movement’s overall tonic (A, C♯, and E)
provide a basis for the expanded peremennost domain in which the theme is heard. Figure
5.18 puts this theme in the context of the exposition. The figure shows the opening of the
movement in A major, a transition that moves to the gamut of D♭/C♯, the peremennost-
inflected second theme area (Figure 5.17), the ff “chromaticization” of C♯ before around
78, the closure of the exposition—unexpectedly—in E♭ major, and the octatonic
statement of the symphony’s motto theme on C♯ at rehearsal 80 that prepares the large-
scale statement of the motto over the course of the fugue that follows. (See again the
analysis of rehearsal 80 and following in Chapter 4; Figures 4.x through 4.x.) In Figure
5.18, arrows indicate points where C♯ in some form is strongly emphasized. Recall that
the hyperdissonant climax in the first movement (analyzed in Chapter 1; see again Figure
1.x) strongly suggested potential resolution to D♭/C♯. All of this material suggests that
the incorporation of D♭/C♯ into the symphony’s global A minor/major is a central
concern in the work. (Further evidence for this view is provided in the analysis of the
second movement later in the present chapter.) In this interpretive context, the thematic
statement in Figure 5.17 represents a synthesis of the respective tonal gamuts—A major
and its dominant, E major, and thematic material in C♯—via peremennost.

186
 Figure 5.17. Symphony No. 3, Op. 44, iii, analysis of peremennost in the second theme (exposition)

C  minor material
Meno mosso 77
      
       
             
      
      
f
        
mf cantabile
*

        
    
          
        
           

187
               
               
        
       
    
      
     
E major + (nega)
A major

peremennost
superimposition
 Figure 5.18. Symphony No. 3, iii, overview of exposition

motto peremennost
motto
augmented triad     
 
            
             
           
transition f ff pp ff
mf

        
 
         
  
         
6 6 4 !4
!4 3 §2 !3 !7 !5 #5 §9 !9

188
!5 §3 §3 7 !7
§3 #3 §5
octatonic
fusion of C  minor §3 6 7
4
V III iii
I V A major !
E major V
of F 
Theme area I Theme area II

emphasis on C 

71 73 74 75 76 77 78 79 80
 Nega

An interesting structure results when a peremennost-type oscillation between

mediant-related triads is filled in chromatically, as in Figure 5.19. (A similar device may
be clearly heard in Figure 5.17 after rehearsal 77—the E major and C♯ minor
components of the peremennost compound are connected the B♯.) Figure 5.19 resembles
the expressively-packed nega idiom that Taruskin has discussed at length.18 As he tells it,
the word nega “is usually translated as ‘sweet bliss,’ but it really connotes gratified
desire, a tender lassitude…In opera and song, nega often simply denotes S-E-X à la
russe, desired or achieved.”19 In its original musical form, nega is part of the standard late
nineteenth-century Russian “orientalist” package. Although certain rhythms and textures
are associated with nega, it is a melodic figure that really defines the idiom: “the
reversible chromatic pass between the fifth and sixth [scale] degrees is in fact the
essential nega undulation,” as Taruskin presents it.20 See again Figures 5.19; a corollary
in the minor mode is shown in Figure 5.20. Because a chromatic tone is involved, nega is
not strictly modal; but I view the idiom as an outgrowth of peremennost, and, as several
analyses below show, its general rhetorical associations (expository or digressive—in
middle sections—as opposed to climactic) support inclusion in the modal category.

Figure 5.19. Basic nega idiom in D♭ major

"
! &!" "
# "#" "
"
!
! " """ ### $# ##
## $# %# "###

!"""""""""""""""""""""""""#$""""""""""""""""""""""""""""!

18
Taruskin, Defining Russia Musically, 165-185; and “Russian Musical Orientalism: A Postscript,”
Cambridge Opera Journal 6 (1994): 81-84.
19
Taruskin, Defining Russian Musically, 165. Taruskin discusses the “orientalist” implications of nega at
length.
20
Ibid., 168. Taruskin has pointed out (ibid., 135-136) that Gerald Abraham identified the same kind of
“chromatic pass” as a characteristic of Russian music decades earlier, though Abraham did not associate
the technique with any particular expressive topic. See Gerald Abraham, “The Elements of Russian Music,”
Music and Letters 9 (1928): 51-58.

189
 Figure 5.20. Basic nega idiom in B♭ minor

"#"
" "
! !
"
! " """ ### ""# ### $# ##
##
#
! """ !

Taruskin identifies the nega idiom in works by Glinka, Borodin, Tchaikovsky,

and even the young Rachmaninoff (“Ne poy, kravitsa,” Op. 4, No. 4 of 1892, which does
not fare well in Taruskin’s hands).21 A particularly clear example of the idiom (clear both
musically and in terms of its strong exotic associations) not mentioned by Taruskin is in
the third movement of Rimsky-Korsakov’s Scheherazade (1888), “The Young Prince and
The Young Princess.” An analysis of the passage is given in Figure 5.21.22

Figure 5.21. Rimsky-Korsakov, Scheherazade, iii, analysis of mm. 1–8

(a) annotated reduction

!
% $ $
&
! %$ &
# " $ $ $ $ $% $ $ $% $ $ $ $ $ $ $ $ $ $ $ $ $% $& $ $ $% $ $ $% $ $ $% $ $& $ $ $ $ $ $ $%% $% $% $
!
))%% &
)% )))%% $% * $% *))$$%%%% +* $$%% ))%% +))$)%%%% , $% $%
$$%% $$$
' !" ( ( ))%% % ))%% )% )%
-
!!

!"#$%&'(!"#$%!$)"(*+%#,("-."/+%"-+-(#,("-."/+0)12#

(b) analysis

21
Taruskin, Defining Russian Musically, 165-185; Taruskin, “Russian Musical Orientalism: A Postscript,”
81-84.
22
In later editions of the work, Rimsky-Korsakov removed the movements’ programmatic titles. The third
movement came to be known simply as Andantino quasi allegretto.

190
 Several compositions from Rachmaninoff’s early Russian period feature the nega
idiom in a straightforward form—that is to say, emphasis on the “reversible chromatic
pass” between scale degrees 5 and 6 (in the major mode), effecting a rocking back and
forth between I and vi (as in Figure 5.19), often above a pedal tone. The nega idiom often
occurs at the beginnings of sections (and is in such contexts associated with thematic
exposition); and usually the idiom occurs in lyrical middle episodes or movements.
Figure 5.22, from middle section of the Elegy in E♭ minor, Op. 3, No. 1, contains a
structure quite like the one in the Scheherazade. (The hyperdissonant climax at the end of
the middle section of the Elegy was analyzed in Chapter 2; see again Figures 2.29 and
2.30.)

Figure 5.22. Elegy in E♭ minor, Op. 3, No. 1, analysis of middle section

(a) annotated score excerpt

!
# # # ! $$ $ $$ $ $$ $$ $$ $$ $$ $ $$ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ &$ $ $ $ $ $$$
!"

#
" ## $ $ $ $ $ $ $ $ $ $ $ $%$ $ %$ $ $ $ $ $ $ $$ $ $$ $$ $ $$ $ %$$ $ $$ $ $$ $ $$ $$ $#$$ $ $$ $ $ $ $
$ $ $ $ $ $ $ $ $ $ $
!! "#
' # # #! ( $ * * $ $ * * * $ * %* $ #$ $ $
( $ *
### $ ) $ ) $ $ ( )
$ ( $ ( $

!"#$%&'(#$%&'#$)"(*+%#,("-."/+%"-+-(#,("-."/+0)12#

(b) analysis

191
 Figure 5.23 shows Rachmaninoff’s use of the nega melodic idiom in a somewhat
more complex harmonic environment; the reversible chromatic pass is clear, however.
Like the excerpt from the Elegy above, the excerpt in Figure 5.23 is from the beginning
of a lyrical middle section. Note that the expressive focal point of the phrase is the area of
maximal nega activity. The straightforward cadential progression that follows is, by
comparison, unremarkable—and unmarked.

Figure 5.23. Musical Moment in B♭ minor, Op. 16, No. 1, analysis of mm. 38–41

$./0$((+1$%2")'3%/"+#!%"2%/40'($

!
!"#$%"&"
) ) )+ ) )$ ) ) ) ) $ '$ ' ' '$
$ $

) ) ) ) )*) '
!"
% + $ ) ) )$
$ % %%%% "! ( ) ) ) $ )+ )) ) *) ) %) ) ) "#
$
)
& " *)
) ) ) )
$

% )& ) * )& '

) ,) %) %)
!" ' ' ' ' ' ' ' ' '
+ && *)) ))# &
$
- %% % % ! &
"#
#$%&#' " ()*'

% % " &" ) *) &" ,) ) ) )

. )
-%./0!"#$%&'()$#*+#,- -%./0!"#$%&*$()$#*+#,-
-%./0!"#$%&'()$#*+#,- -%./0!"#$%&*$()$#*+#,-

!
+, '
% #
% %
$ % % % " )& ) ) ) ) ) "! ( )
$

* %) & " *) )
' ' '!"
- %% % % "# && ) "! && "
%% & )

'( '( )

192
 Analysis of works from the middle Russian, late Russian, and exile periods
suggests that Rachmaninoff maintained an interest in the nega idiom throughout his
career, developing it, however, in distinctive ways that transport it beyond the basic
nineteenth-century forms described above. In all the cases I have identified, an
association with romance (in song), with middle-section lyrical episodes (in shorter
instrumental works, or inside individual movements of longer instrumental works), or
with slow movements is retained. To take an especially well-known work as an example,
nega provides a starting point for interpreting the second movement of the Piano
Concerto No. 2, Op. 18. Figure 5.24a shows a suggestion of nega underneath the flute
solo in measures 9–11. Figure 5.24b shows nega more fully developed in the
harmonization of the main theme in measures 13–19: note the characteristic move from I
to vi and back, and the characteristic interplay of scale degrees 5 and 6 (here in an inner
voice), as indicated by arrows on the figure. As discussed below, in the context of the
entire movement, nega, combined with hexatonic organization, provides a basis for
interpretation of climax.

193
 Figure 5.24. Piano Concerto No. 2, Op. 18, ii, analysis

(a) nega in mm. 9–11

!"
!
"""" !"#$%&'("( $ "$ "% $ $ &$
# $ %

!
!
!" "#$%"##&

"""" $ $ $
!
$ $ $ $ $ $ &$$"$ $ $ "$ $ $ $ $ $&$$ $ $ $ $% $
$$ $$$
$
$% %
' """" (
#
)
% $ &$ $
$
% $
( ( (
'"()

(b) nega in main theme, mm. 13–19

!
%&
%%%% ! & & & & & & & & & & # ' & (& (& (& '" & ! & & &,& & &,& &
! $ " )* "+ & #

'&' " , '''' """ , '

!'()*+"'"',"-./"'",./",,0

&/ '' %''' ,'& " &/ #

. .
. ,'
- %%%% ! .
! " .
. '
#'
' ' '" "! .
. '
! !

!
!"#$

%1
%% ' & %(& (& (& '" & * !+ & & & & & & & & &
! $ % %# ) "
&' "
( (
'
- %%%% # '' ,''' %'''' "" & ,& '. & & &/
!
'
'
' '" "! ..
.
'
'
!
"#
!"#$

194
 As shown in Figure 5.25, the movement opens with a short HEX(3,4) passage that
effects a transition from the first movement’s C minor to the second movement’s E
major.23 As detailed above, nega in the movement involves E major (I) and C♯ minor
(vi), bridged by the unstable nega tone C♮/B♯. The nega tone, then, is established in the
HEX(3,4) structure in the movement’s opening measures. While the G♯ major triad in the
HEX(3,4) structure resolves directly to E major in measure 5, it may also be interpreted as
diatonic V of the nega alternate, C♯ minor—that is to say, V of vi. (Note that, in the
HEX(3,4) opening, a forte dynamic strongly emphasizes the G♯ major triad in a generally
pp context.) At measure 105 in the movement (rehearsal 23), a large-scale, ff resolution to
VI—here C♯ major rather than C♯ minor—occurs. Figure 5.26 puts the climax event in a
larger context.

Figure 5.25. Piano Concerto No. 2, ii, connection between introduction and climax
45678(

!
:0
!$#
&
$
% #% %
# ## #% %
" # ! #% % $
!"#$"!%
%
#$
$
$
' #### !

!
%
$
$ %
####
! " #
# %
" ! #% #% & &
(% )% (% (%
!%
(%
%

%% # #%% ## %% # %% # % %%
# % #
' #### ! (% )% )% (% (% (%% ###% #%
!!
) % #% ( % #% #%
#% (%
(%
%
"" !
&'()"*'+) "" !"#$%&'()*#+,&
,-+"-.,+/
%
0
,-./0123
-"789#*

23
As noted in Chapter 4, chromatic major-third relations are identified in the movement in Bribitzer-Stull,
“The A-flat–C–E Complex,” 186.

195
 Figure 5.26. Piano Concerto No. 2, ii, analysis of climax
Climax, part 1 Climax, part 2
23 25
Piu animato ascent (WT) cadenza

 
     
    (HEX)   
              
    
   
  
                 
1
   
         
§7

7 ff nega sffz
#3 descent
5 §6 7 6 #6
3 §3 5
5 #3


[
VI
SD 24
Tempo I

196
  
    

       
 
pp
 
2 p 
    
   

§5 6 §6 7
§3 4 §3
[

II
IV V V I
D
apex (cont.)

(E major) Return
 As detailed in Figure 5.26, the C♯ major triad (VI) is the first part of a two-part
climax that synthesizes the nega idiom and hexatonicism. A large articulation of the nega
figure (C♯-C♮-B) connects VI to IV at rehearsal 24 in the wake of the first stage of
climax. A second intensification at rehearsal 25 leads to a cadenza on the Neapolitan (F
major), which is hexatonically related to the C♯ major event. This may be interpreted as
large-scale harmonization of the nega tone. Figure 5.27 shows the two-stage climax in
the context of the entire movement.
In such a rich harmonic structure, peremennost, the “chromatic pass,” Taruskin’s
nega as an expressive topic, and hexatonic organization bleed together into a complex
multilayered environment. Certainly, any specific folk or liturgical implications are long
since erased. The harmonic materials involved are nevertheless strongly marked in
relation to the general functional tonal context, and retain the basic rhetorical associations
identified in Chapter 2: nega/peremennost in a straightforward form characterizes
thematic exposition in the movement, while emphasis on hexatonic structure
characterizes the climax events. The second movement of the Piano Concerto No. 2
demonstrates a kind of structural interaction taken to new heights in the late Russian and
exile periods.

Figure 5.27. Piano Concerto No. 2, ii, analytic overview

)7 )8
56 )5
-*./&'0
%$
# #$ % $ % $$$ $$$ $$
!"#$%&'(&()*+,)&
" #### ! $ $$ $$$ $$ $ $$$$ $$ % $$ $$
$ $ $ %$ #$$ $
$
$ $ #$ %$ $ %%$$
$ $$ $ $$

$ &
! "! % "$
"# "# "#

0 10 %2 1 %00 1 0

!"#$ !"#$%&'#(')'*+%,-* !"#$

!"#$'2'3-&%+4(#!
.-/%0+10- 0-+10(

197
 Figure 5.28. Prelude in E major, Op. 32, No. 3, analysis of mm. 1–40

Allegro vivace
peremennost reharmonization of x (I - vi)
(octave doublings freely omitted)
 
3 11 13 22
     
 
        
   
     
     

[
     

[
ff

[
ff x
     
ff

x x’
   
 
  
    
     
6

6
     
I #3
vi
(III)

no bass support
V

198
x
[

24 26 28 31 40
  
         
   


    
   
                   
 
f ff p cresc. ff
dim.
 

 
     
     

   
  
    

6
   

6 7

§6

#3

V bass support

! I
 A similar structure may be heard in the much shorter E major Prelude, Op. 32,
No. 3 (1910). Again, an intersection of nega, peremennost and hexatonic structures is
involved. Figure 5.28 provides an analytic overview of the prelude’s first forty measures.
As shown in the figure, a simple motive (x) is set in a number of contexts. Measure 22 is
a significant point of arrival in the work, establishing the G♯ major triad as a double-
function chord: it is hexatonically related to the E major tonic, and it is V of vi (the
peremennost partner of the E major tonic). Figure 5.29 shows this more plainly. As
shown in Figure 5.30, later in the prelude, pitch-class B♯ from the G♯ major triad is
treated as a nega tone underneath continued treatment of motive x, culminating in a more
explicit HEX(0,1) harmonization. As shown in Figure 5.31, motive x is harmonized with a
more straightforward E major tonic in the prelude’s coda.

Figure 5.29. Prelude in E major, analytic overview

!
** ++ +, +- +. 2* ,3

# # # # #
""
# # # # #
!"" # ##
# $## ## "# #
## ## # $## ##
# # " ##
## # " ##
!! /0$1 " !!
% """" # # #
# # # #
! !
#
! ! !
"
" #
!

!"#$%&'($)
*$+()/)'02&(%*&3
!"#"$"%%&'()(*($+
,
-./()0$10'.(!*#

199
 Figure 5.30. Prelude in E major, mm. 50–55

!
! $4256(
""
!"
& & & & & &
!"" #" ## # " ## "" ## ## "" ### """ ### """

!
#"$ ##
## ## ## ##
#$ " #% #$ " #% #$ " " ## # "
% $ % $ " ##
% $ ##
%
$ % % % $ % % % $ % % % $ % % % $ % % %
#$%$&'&#$%$&()*+
$ % % %
# # ##)$# # #
' """" # # # # # # # # ## )# # # # ## # ##
# " # # ##*
"# #
( ( + # # #

!
$ %"#& %$'&

!"#$%&'()*+,-./01+20./).3)!
""""
!, !!
###
,+
#)$ +++
! +++ )), +++ + + #
"" +++
!
% % % %
' """" + )) #+ # ,# )# "# )# &
# # )# # )# # , #
) # "" ## ) # # "# )# # # ) # #
$ % % % % % #% " #% ) #% #% # ) # #
!
% %
"
"#

Figure 5.31. Prelude in E major, motive x in the coda (m. 55–57 )

"""" # $ %'& (% ! )
%& % $ ' % ! %& %
%& ( )
!!

!
( (

Figure 5.32 shows in generalized form the “hexatonicization” of the nega idiom
suggested in the analyses of the Concerto No. 2 movement and the E major prelude. The
figure shows chromatic harmonization of the nega tone such that the same major-minor
seventh chord (on root F♮ in Figure 5.32) is used as the “dominant” of both peremennost-
related diatonic chords (I and vi). In Figure 5.32, the first resolution of the nega tone is
“diatonic”—in the sense that the major-minor seventh chord acts as a conventional
applied chord (V7 of vi)—while the second resolution of the nega tone is hexatonic

200
(specifically, HEX(0,1); recall from Chapter 4 that relations based on hexatonicism may in
the works analyzed contain extended tones foreign to the collection—in this case, the
seventh, E♭, is not a member of HEX(0,1)).

Figure 5.32. Chromatic harmonization of the nega tone

"
" # "
! !
"
! " """ ### $# ##
## $# %# "###

## #
& "" " ## $#### ##
## $#### "###
""
)* +)*,
! '( !
"#$%&

Figure 5.33 shows the structure from Figure 5.32 as it appears in the opening of
the the song, “A-u!” Op. 38, No. 6. (The octatonic, hyperdissonant climax of the song
was analyzed in Chapter 2; see again Figures 2.25 through 2.28.) In the beginning of the
song, a chromatic descent in the accompaniment provides a framework for the music
from measure 1 through the arrival on V7 in measure 4. In measure 5, the “other”
dominant seventh chord from Figure 5.32, V7 of vi, replaces diatonic V7, resolving to a
fused D♭/B♭ (I/vi) sonority and ultimately progressing to B♭ minor (vi)—above the tonic
note D♭, however—to end the first section of the song. After a fermata, a new section of
music begins (measure 12). In measures 12 and 13, the nega melodic figure implicit in
the structure of the first 11 measures of the song is made explicit (circling around the
unstable nega tone, A♮), providing the basis for a complex, quasi-octatonic oscillation
that will later develop into the true octatonic structure heard at the climax. (See again
Figure 2.26.)

201
 Figure 5.33. “A-u!” Op. 38, No. 6, analysis of mm. 1–13

1 4 5 6 7 8 9 10 12
voice
Andante tempo piu vivo. Appassionato
 
       
       
               
 p
pp ff play on nega tone A 
cresc. dim.
piano 
                   

        
 
  
   

202
  
           
      
        
peremennost
fusion: quasi-OCT
7 7 (develops into climax)
I V V (of vi) D /B  vi6
(I/vi)
 Figure 5.34. “In the Soul of Each of Us,” Op. 34, No. 2, analysis of mm. 1–14

4 5 6 9 10 12 13

                        

voice
     
      

nega      

(HEX) (HEX)

203
       
      

   
          

piano
     
   
       
        
    

7
6 !5
4 !3

I vi iv V I

T PD D T
 Figure 5.34 shows an intersection of nega and hexatonic structures in the first
fourteen measures of the late-Russian period song “In the Soul of Each of Us,” Op. 34,
No. 2. Although the nega tones are disguised by the chromatic context, the essence of
nega remains: the reversible chromatic pass and the exchange of I and vi (with iv added
in this context) are embedded in the passage.
As discussed earlier in the chapter, Rachmaninoff extended peremennost
techniques to include the possibility of vertical as well as horizontal presentation. The
nega idiom is similarly extended in several of the works analyzed: the “reversible
chromatic pass” may be resolved both up and down at the same time—i.e. the nega tone,
and any chord with which it is associated, may be resolved to two different diatonically
related chords simultaneously, as shown in Figures 5.35a and 5.35b. In the figure, the
nega tone has two enharmonic meanings at once. It is simultaneously scale degree ♯5
rising to scale degree 6 as the root of vi, and scale degree ♭6 (spelled enharmonically in
the figure) resolving to scale degree 5 as the fifth of the tonic triad, resulting in
conventional and chromatic resolutions at the same time. (In the figure, the superimposed
sonorities are labeled simply I+6.) The principle outlined in Figure 5.35 suggests a basis
for understanding certain complex structures in the Postromantic repertory more
generally. Figure 5.36, from the last movement of Gustav Holst’s The Planets, shows the
simultaneous resolution of a tendency tone in two directions at once.

Figure 5.35. Complex resolution of a chromaticized nega chord

! "" "
!"
%$% #" %$%
#"$##
! " " ### $### ## ## $$%$#### #
# # #
"&% "&%
& "" # $## "## ## $# #"$"##
" """ $#
!"#$ %!&'$ %!&'$ !"#$ %!&'$

204
 Figure 5.36. Gustav Holst, The Planets, vii (“Neptune”), m. 101

" """" ! $## !!

!"!

!
## !!
#$%&%'(%)*')*+,-,(./*'0*01.2)*3')%0*'4'56

! !

"""" ! !!
(!#)*&'

# ! %#!
% #!
! !
## %& &&
+&#,%-.!/0&'1.,.

&
%!" "
#

" """" ! ! !!
""#
! "#
"
# &&
!
! ! "& &
"""" ! !!
! #!
"
#! ##
"!!!#$%&'
!
#!

As explored more fully in Chapter 6, nega/peremennost is elevated in the

Rhapsody on a Theme by Paganini to the point that a hyperdissonant climax event in the
famous eighteenth variation may be interpreted as an outgrowth of nega on the global
scale. Although a detailed description of that moment must wait, a few preliminary
observations are possible here. Figure 5.37 shows an excerpt from Variation XVI in the
key of B♭ minor. The figure shows a momentary peremennost-derived superimposition of
B♭ minor and D♭ major, the latter of which will emerge as the key of Variation XVIII,
which is the centerpiece of the Rhapsody. B♭ minor is retained throughout the excerpt;
but melodic and harmonic resolutions to D♭ major as “tonic” are articulated. In the next
variation (XVII), more pronounced superimposition of B♭ minor and D♭ major is
enhanced by articulation of the nega idiom, as shown in Figure 5.38a. Throughout
Variation XVII, A♮ is treated as an unstable nega tone, resolving up to B♭ as the root of
B♭ minor and down to A♭, as the fifth of D♭ major, and, at measure 621, to both
simultaneously.
The climactic culmination of A♮ nega in Variation XVIII (where it is
compounded hexatonically) is taken up in Chapter 6. That the doubly-unstable nega tone

205
A♮ used throughout Variations XVII and XVIII is in fact the global tonic of the Rhapsody
constitutes one of the most intricate expressions of hyperdissonant exaggeration in any of
the works analyzed.

Figure 5.37. Rhapsody on a Theme by Paganini, Op. 43, Variation XVI, mm. 575–577

!"#$%&'()"*$#+,&$-(,$(. "
! "" " # # # #
!"
#$ &%
! "" # # # #
!

$) $)
"" # ## ' ## ## '
! " " " # (# ## #
!

' ##
# ' # "#### # ' ### ### ###
$* $* $ $ $ #$%$&$!'$( "
#
$ &
# ##$ ## # ## #
+ "" " # # # # # #
, ## ##
" "" * ' !
' , # #
-
& "
!
!"#"$"%%&'($)*+,-!.+/)!0!/'1$!$%$&&&

206
 Figure 5.38. Rhapsody on a Theme by Paganini, Variation XVII, mm. 613–622

mf mf
peremennost superimposition
        
      
 


p p

(a) 
 
see Rachmaninoff's meticulous stemming
       
p cresc.
cresc.
                                           

       

         
                

         
                 

207
continued play
nega figure on the nega tone
V - I in D 
i

XVI and XVII XVIII

(b) 
 

D  major
B  minor
 Phrygian Organization

Several analyses earlier in the dissertation featured flat scale degree 2 at strongly
marked moments. (See again the analyses of the second and fourth movements of The
Bells in Chapter 4 and the present chapter, respectively; and the analysis of the first
movement of the Piano Concerto No. 3 in Chapter 4.) While many such cases may be
interpreted as articulating the Neapolitan as a conventional chromatic predominant
harmony, in some works the lowered second scale degree is emphasized to such a degree
that a genuine Phrygian organization results.

Figure 5.39. Symphony No. 3, Phrygian motto theme in mm. 1–5

!"#$%

# # # #!
& & *
"! # $ # # %# # # # # ' # ( )
!! !"#$

Figure 5.39 shows the motto theme of Rachmaninoff’s Symphony No. 3 as it is

first heard in measures 1–5 of the work. Its most distinctive characteristic is the Phrygian
tone B♭. The motto may be understood as an upper and lower neighbor figure—B♭ and
G♮ orbiting tonic A♮.24 The motto is one in a distinguished line of opening Phrygian
gestures in Rachmaninoff’s concert works. Figure 5.40 shows a Phrygian figure used
extensively in the opening of Rachmaninoff’s opera Aleko (1892; the opera was his
graduation work). A similar figure at the opening of the Symphony No. 1, Op. 13 (1896)
is shown in Figure 5.41. Like the Symphony No. 3 motto, the figure circled in Figure
5.41 serves as a kind of connective tissue across the movements of the symphony. These
are probably all descendents of the essentially Phrygian theme heard at the opening of
Alexander Borodin’s Symphony No. 2 (1876), shown in Figure 5.42.

24
David Cannata’s interpretation of the Symphony No. 3 also centers on the B♭ (Cannata, Rachmaninoff
and the Symphony, 125-30). However, his conclusions are quite different from my own, as explored more
fully in Chapter 6. For Cannata, the B♭ suggests a large-scale D minor/major implication, which I regard as
insufficient, particularly as there is only one extended passage in the key of D in all three movements. My
reading of the work suggests a more intricate structure that emerges in part from the Phrygian nature of the
motto theme.

208
 Figure 5.40. Aleko, Phrygian organization in the introduction

!"#$"%&'($"%$)*+&

! %' #% % $ %' #%% % % $ %'

$ $ $

% #% % % % % % % #%$ % % % % % % % % )% % $
$ $

% ( (# *
$
$

! " #!
$ %( % $ %( $ %%% % % $
& & & ( & ( & %
! ! !

%% ( + ( # %,
"# ##
' ' ' '
$ $ $

""
$ $
#! $ %% ( #%% $$ %% ( % %#%% $$ %% ( #%% %% %% % ( % % % % % % % % % %#)%%% $$
$ % %#
!
& &

,-./0*$"'"&*0-)1.'2*03.& "#$%&'

!
!

! "#
' ( ' (
$

#$ %. %. % # % ##%%% / %%%
- . . %.
%## %##
' '
+ #-$ %
%. %. % #% % % ( )% (
" #
$
!
. . . )%

Figure 5.41. Symphony No. 1, Op. 13, i, Phrygian opening gesture

&''%(")*+#*,),*-")..)

! ! "
" # % $ $#$ & $' ( $ & $ $' )% $ $#$ & $' ( $ & $ $' )% $ $#$ & $' ( ( ## *$$, $$ $$'
+$ +
!"#$%
*.
.& #&
" "
!
- $ $
+
+ +
!
0 0 00 0 0 00 00 0 + + *$ $$ $$'
& $ $ ## $
###
' 0 ' *& #&
/ ! )
" #
& $ ( $ $# $ $ $ & $ ( $ $# $1 $ $ $ $ .
!
& $ 1 + . $ . $
### 21
* #
/0"1(2#, 34 2

Figure 5.42. Borodin, Symphony No. 2, i, Phrygian organization in mm. 1–3

!
# % %
" #!
$ ' ' ( ' ' #' ' (' $
!
& &
#
" #!
('
)*)+&,(,!
' #'
' (' '
!"#$%&'()*

209
 Both Aleko and the Symphony No. 1 explicitly invoke a non-Western context.
The libretto of Aleko is an adaptation of Pushkin’s poem “The Gypsies.” Barrie Martyn
has observed that the Symphony No. 1 represents a synthesis of liturgical music and
gypsy music in a symphonic context.25 In both cases, Phrygian material is associated with
a strongly marked musical and cultural content. Martyn tells how Rachmaninoff’s teacher
Nikolai Zverev introduced the composer to gypsy performers as a young man:

Like many Russian musicians Zverev himself was greatly attracted by gypsy
music, and in the course of preparing his ‘cubs’ for life he used to take them to the
fashionable Moscow restaurants at which gypsy musicians played and stirred the
Russian soul.26

Of perhaps more personal significance was Rachmaninoff’s close association (and

entirely unrequited infatuation) with the gypsy singer Anna Lodïzhensky in the early
1890s.27 Memories of the association might partly explain the gypsy melodic “sobs” and
strong Phrygian elements in the late Russian period song, “To Her,” Op. 38, No. 2
(1916). The poem of the song, by Andrei Bely (pseudonym for Boris Nikolayevich
Bugayev), “tells of a lover who hears, or imagines he hears, his beloved call to him but
waits for her in vain.”28 Figure 5.43 details the Phrygian structure of the opening
measures: a melodic ostinato based on an upper and lower neighbor figure not unlike the
Symphony No. 3 motto theme, and a recurring Phrygian resolution to F major/minor.
Figure 5.44 shows these features more plainly; the arrow shows the essential
three-note Phrygian cell. An analytic overview of the entire song is given in Figure 5.45.
Two different kinds of music are involved. “A” sections are based on the ostinato shown
in Figure 5.43. “B” sections, which do not contain the ostinato, involve local highpoints
and, in measures 33–35, the song’s climax. As shown be beams on the top staff of Figure
5.45, the first half of the song (measures 1–26, sections A1–B2) can be understood as a
large-scale articulation of a Phrygian melodic structure derived from the ostinato: the
essential tones are the highpoints of the vocal line, F-E♭-G♭-F-A♭-F.

25
Martyn, Rachmaninoff, 98-100.
26
Ibid., 56.
27
Ibid.
28
Ibid., 265.

210
 Figure 5.43. “To Her,” Op. 38, No. 2, Phrygian structure in mm. 1–4

' ( )

!"#$%&'

!"#$%&#!

Figure 5.44. “To Her,” Phrygian organization

!"#$%&'()*"+),-%.+(*-%

Phrygian organization largely disappears in the second half of the song (sections
A3, B3 and A4). However, the climax in section B3, which continues the trajectory of
highpoints in the vocal line begun in the first half of the song, involves a larger-scale
version of the F minor-major alternation that characterized the Phrygian opening
measures (see again Figure 5.43): F minor, with pitch class E♭ at the climax, resolves to F
major for the postlude. The resolution to F major involves the same Phrygian chord heard
at the opening of the song; both are marked “*” in Figure 5.45. At measure 39 in the
postlude, the ostinato figure is set in conventional F major rather than F Phrygian.

211
 Figure 5.45. “To Her,” analytic overview

A1 B1 A2 B2
voice

2 3 11  16 22
     

                  
                                    
              

f p p p
p mf f
                                             

    
  
* 

“III” V
I/i vi

212
highpoint of
vocal part end of
A3 B3 A4 (postlude)
vocal part
41
27 31 33 39
   
  35  

   
                  
         
      
p
   
p p pp
ff
                   
   

 
   
    
  
*
climax
V [i] i I
 The above examples suggest that Phrygian structures in Rachmaninoff’s works
have expressive associations and structural implications quite unlike those of the other
church modes. The Phrygian is for Rachmaninoff not a pseudo-religious mode, but a
pseudo-exotic one. Here, Rachmaninoff as an “Eastern” composer comes to the fore,
even if such a label may be problematic.29 The Phrygian mode is, like nega, a harmonic
sign for something not at all of the Western common practice. It is therefore strongly
differentiated from conventional tonal structures in Rachmaninoff’s music.

Figure 5.46. Rimsky-Korsakov, Capriccio Espagnol, iv, excerpts

(a) mm. 7–11

('
!""#$%#&&'
'' '' '( ' ' ' ' '( ' ''' '( ' '''
& & & &)& & &*) & & *
&* && &$&
(
! ! &
& & +& & & & &# && & & &
!
& & )& &
& & & &&&
*
& && &&
# $ " % &&) & % &&& % &) & % && % &) & % && % &&) & % &&& % &&) &
!' ' ' ' ' ' ' " ' '' ' ' '' ' ' '' '
' " ! ! " ! " !
, !
" $" & %& % & %& % & %& % & %& % & %

(b) mm. 22–25

! !"#$"%&
&&
!"#$%"
!" # # $ $$$$ $$$$$ # # $ $$$$ $$$$$ #
$$$$ $ $$$$ $$$$ $ $$$$
)' !! ' )$'
'()&& '
# # # &$$$ $$$ # # * #
% &$$$ $$ $
" " $( $$ # # *
(
# *
( (
!

Rimsky-Korsakov used Phrygian organization extensively in the fourth movement

of his Capriccio Espagnol (1887), “Scena e canto Gitano”—again, to portray a non-
mainstream European culture. Figure 5.46 shows two representative passages from the
movement; their tonal substance is extracted into Figure 5.47. As shown in Figure 5.47,

29
Taruskin considers the problem of locating Russian music between East and West in detail, historically,
hermeneutically, and aesthetically in Defining Russia Musically.

213
A♮ Phrygian major is involved. The one-flat key signature of the fourth movement must
not be construed as suggesting simply D minor, though that key may be suggested in due
course. Rimsky-Korsakov’s music does more than hover around the dominant of D
minor—A♮ clearly emerges as the legitimate pitch center of the movement.30 As shown in
Figure 5.48, the B♭ necessary for the mode may be interpreted as a holdover from the flat
regions in the second and third movements. A♮ Phrygian major is replaced by
conventional A major in the fifth movement.

Figure 5.47. Capriccio Espagnol, iv, Phrygian organization

-$.&*',)!"#$%&'()/'01#
!"#$%&'()*+,,
#+21,3-&1(
"" ""
! " "#"$" " #""
""
"
$ """

Figure 5.48. Capriccio Espagnol, overview

!" !!" !!!" !*" *"

#$%&'()( *('+(,+&-+ #$%&'()( ./0-(101/(-2&13+2(-& 4(-)(-3&1(526'+(-&

! """ """
# $ # $$ # $ # #
41:(;&' $
<11:(;&' #178'93+(-1:(;&'

#1:(;&' #1:(;&'

Mussorgsky used Phrygian organization in conjunction with other modal

structures in the sixth number of Pictures at an Exhibition (1874), “Samuel Goldenberg
und Schmuÿle.” Figures 5.49 and 5.50 provide analytic highlights. The piece is in ternary
form (A1-B-A2). As shown in Figure 5.49a and in Figure 5.50, the A sections are based
on the so-called “gypsy” minor scale, which is closely associated with OCT(1,2) (see the

30
The Prelude to Act IV of Bizet’s opera Carmen involves the same pitch structure, again with a one-flat
key signature. The opera predates Rimsky-Korsakov’s work by more than a decade.

214
line labeled “1” on the figure). The B section is based on D♭ Phrygian, which is similarly
closely associated with OCT(1,2) (see the line labeled “2” on the figure). The minor third
relation between the D♭ tonic of the B section and the B♭ tonic of the A sections may be
interpreted as a manifestation of this quasi-octatonic background association.31

Figure 5.49. Mussorgksy, Pictures at an Exhibition, No. 6, overview

(a) mm. 1–2

" # # # ! %& $ ( !)$ $ $ $' *

## ' (!
3
'
! $ $ $ $'
3

$ $ $ $ $ $ )$
) $+ + + +
f sf sf

(b) mm. 9–19 (portions omitted)

!
B A2

%%%% "
$ % &% % &
(2x)
%
(2x)
13 (2x)
9
""
17
'&% %"%
19
[$ % % % %] %
! " " " $## "%%$%% "%% ## $## "%%"$%% $%%$ ##
'$"%%% %% sf
mf
% f sf sf
( "" % "%% "%%% ) '% %'% % # "%
""" % *

D " Phrygian minor B " “gypsy” minor

31
Joel Lester has identified a similar intersection of OCT(1,2) and E Phrygian organization in the opening
measures of Part I in Stravinsky’s Symphony of Psalms (Joel Lester, Analytic Approaches to Twentieth-
Century Music (New York: Norton, 1989), 166–167).

215
 Figure 5.50. Mussorgksy, Pictures at an Exhibition, No. 6, interaction of scales

$
B “gypsy” minor [A1 & A2 sections]
$"
!
$ " # " $" $" #" #" $" #"

(1)
OCT(1,2) T3

! #" #" $" $" $$#""" $$$"""

$ " $ " $"" #" #"

(2)

! $" $" %" $" $"

$ " $" %" $"
$
D Phrygian minor [B section]

The foregoing discussion of non-mainstream European cultures and musical

representations is not meant to advance any specific hermeneutic agenda, nor to suggest
any specific extramusical content to the works analyzed, but, rather, to establish the
specialness of Phrygian organization in Rachmaninoff’s works. This may be understood
as an outgrowth of work undertaken by his Russian predecessors, and to some degree
continued by his Russian successors. See for example the Phrygian-type modes described
in Ellon D. Carpenter’s study of modality in Shostakovich’s music.32 In Rachmaninoff’s
works, Phrygian organization is in some ways a thing apart. It is unlike the functional
tonal basis and, as discussed below, problematic for that basis in many regards; and it is

32
Ellon D. Carpenter, “Russian Theorists on Modality in Shostakovich's Music,” in Shostakovich Studies,
ed. David Fanning, 76-112 (Cambridge: Cambridge University Press, 1995).

216
also unlike other modal structures and “fantastic” structures, although it may interface
with them in interesting ways.
Lori Burns considers Phrygian structures in conventional tonal contexts in detail
in Bach’s Modal Chorales.33 A number of her observations are relevant here. She
recognizes a Phrygian upper and lower neighbor formula similar to the neighbor-tone
formulas used by Rachmaninoff in the Symphony No. 3 motto, in the song “To Her,” and
in other Phrygian contexts discussed below and in Chapter 6.34 Burns suggests that the
Phrygian mode poses special problems in tonal contexts, because a Phrygian tonic may
be understood as a dominant in a conventional tonal context, and, conversely, what seems
like a tonic to tonal ears may be understood as the subdominant of a bona fide Phrygian
tonic.35 Burns concludes that interpretation of a larger context is generally necessary to
determine whether the modal final has a tonic function, or whether it should be
interpreted as an articulation of the dominant.36 A Phrygian final or “tonic,” then, may
have a number of potential tonal implications—and harmonic function and tonal
stability/instability may be very much in flux.
In Rachmaninoff’s Phrygian settings, it is sometimes neither possible nor
desirable to decide which of the above implications is in effect. In a number of the works
analyzed, a Phrygian tonic’s different implications are explored without complete
resolution of the issue. A tonic established in a Phrygian context may be unstable,
wanting, as it were, to become a dominant; yet “dominant” is in many cases too simple.
That, in Burns’s theory, two apparently identical structures might be interpreted quite
differently in different cases (tonic vs. dominant) points to the familiar premise that in
music analysis context is everything; but it also suggests something more directly
relevant to the study of Rachmaninoff’s (and other Postromantic) works: centricity and
tonal function may be quite distinct, and a tussle between the two may be quite salient.

33
Lori Burns, Bach’s Modal Chorales (Stuyvesant, NY: Pendragon Press, 1995), 25-30, 39-60.
34
Ibid., 53-54.
35
Ibid., 41-46.
36
See for example her analysis of “Aus tiefer Not schrei ich zu dir,” 61-84.

217
 Figure 5.51. “Polichinelle,” Op. 3, No. 4, analytic reduction of mm. 1–56

mm.36 - 55 = mm.16 - 35

to B section
peremennost oscillations peremennost oscillations peremennost oscillations
(WT1)


m.56

      

1
   
11
               16
       21
    26 28 30

     
    
       
ff p
         
 ff
sfff p p ff
        
   
ff

      
      
sfff

    

         
218

basis: HEX(1,2) *

F  Phrygian (major) F  Phrygian

climax of A section

primarily increasingly
modal
modal “fantastic”
 The early piano work “Polichinelle,” Op. 3, No. 4 (1892) is a compelling case in
point. The piece is in ternary form. As shown in Figure 5.41, F♯ Phrygian major
organization is articulated in the opening and closing measures of the A section, along
with peremennost oscillations and equal-interval patterns at the local climax.37 To
conventional tonal ears, the F♯ Phrygian major tonic sounds very tentative. The bulk of
thematic statement in measures 11 through 26 involves not F♯ but peremennost-related D
major and B minor, which to some degree act as surrogate tonics.38 F♯ major, by
comparison with D and B, sounds quite charged, even unstable. Yet F♯ cannot be
interpreted simply as an unresolved dominant—the three-sharp key signature chosen by
Rachmaninoff prohibits this. The only sensible interpretation of the key signature is F♯,
suggesting that Rachmaninoff considered that to be the effective tonic of the work.39
As shown in Figure 5.52, the return of Phrygian F♯ at the reprise of the A section
follows a climactic passage at the end of the B section in which the hexatonicism
suggested at the local climax in the A section is developed into a stronger and more
explicit hexatonic relation between B minor, one of the surrogate tonics in Figure 5.51,
and G minor, which emerges as the link back to Phrygian organization. As shown in
Figure 5.53, the piece ends in F♯ Phrygian major.
“Polichinelle” establishes a fairly straightforward precedent for the interpretation
of complex Phrygian structures in Rachmaninoff’s later works. Rachmaninoff’s most
elaborate essays involving Phrygian organization may be found in his last two
compositions—the Symphony No. 3, Op. 44 and the Symphonic Dances, Op. 45—in
which, as discussed more fully in Chapter 6, Phrygian organization emerges as a central
structural concern in large-scale, hyperdissonance-oriented Postromantic organization. In
the case of the symphony, key signature is again a significant clue to Rachmaninoff’s
structural conception.

37
There is some uncertainty about the exact bass pitches in measure 30 and following (at the location
marked “*” in Figure 5.51). In published scores, both E♯ and E♮ appear as neighbor tones to F♯. However,
in his Ampico piano roll recording of the piece (date?), Rachmaninoff plays exclusively E♮, bringing the
passage into even closer association with the Phrygian figure used in measures 1–10.
38
Note that F♯ is related to B minor as a conventional dominant and to D major as a “V of VI” dominant
(D major acting as “I”).
39
The use of an F♯ minor key signature rather than an F♯ major key signature, which might seem more
appropriate, may have been expediency on Rachmaninoff’s part, or, perhaps, an attempt to more closely
approximate the pitch-class content of the Phrygian major mode.

219
 
Figure 5.52. “Polichinelle,” analysis of climax, mm. 83–98

       
83 3

        

3

    
 
5

   
3 3

           
 

         
f   
       
 
      
  
  
 sfff sfff
(hexatonic)


             
3
 
3

 
5
  
3

             
3

           
87

     
      
          

 
220

 
       
     
  
 
 sfff sfff


(hexatonic)

        5    
   
91 95 98

           
  

   
  
material omitted

material omitted

   
   
      

 

     
      
   


sfff
sfff sfff sfff sfff    
return of F  Phrygian
 Figure 5.53. “Polichinelle,” mm. 126–130

!
#" #"
## %$# # # %$# $# $# $# $# $#
"" "# # # %# # %$# $# *
# #' )
!"#
& & (
!" & ' ' ' %# ' #( ' ( ' ( ' ' # # ' '
& ' &'
$ $ %# # $$
( !!! $ $
# # %$# $ $# $#
"" # #' # + %# # $ $# *
!" & %# # # # %# # ( # ' '! ' ' +# ' )
$ $% # # $ $ # $ # # ##
$ %# # $
$ $

"
!"#$%&"%'%%%()*+,&-"%.-/0*

A Phrygian link between the early Russian-period “Polichinelle” and Opp. 44 and
45—a gap of almost 50 years to fill—may be found in the last published work of the late
Russian period, the Etude-Tableaux in D major, Op. 39, No. 9 (1917). As is standard in
Rachmaninoff’s etudes-tableaux (and his character pieces in general), the work is in
ternary form. As shown in Figure 5.54, initial statement of the D major tonic is entangled
with a quasi-hexatonic structure in the introduction to the etude. Phrygian organization
emerges with the start of the section A1 proper in measure 5, and is maintained
throughout the exposition of thematic material in measures 6–14. Figure 5.55 details
Phrygian events (emphasizing E♭ in relation to D♮) in these measures.
An overview of the entire etude is given in Figure 5.56. As discussed above,
Phrygian organization in general may be construed as hybridizing tonic and dominant
functions. As shown in Figure 5.56, section B in G major may be interpreted as a
resolution of this hybrid function, and the arrival on C major at the end of section A1 may
be interpreted a large-scale harmonization of C♮, which, as suggested throughout Figure
5.55, is involved in a majority of Phrygian events in the opening measures.40

40
In a functional tonal D major context, C♮ has no clear role; but, in a D Phrygian major context, it is not at
all out of place.

221
 Figure 5.54. Etude-Tableaux in D major, Op. 39, No. 9, analytic reduction of mm. 1–8

Introduction A section

1 3 5 6 8

quasi - HEX(1,2) [main theme]

Phrygian *
   
         

 
         
    
     
   
       
         
ff mf

222
            
   
          
SD
        

T Phrygian Phrygian

D major

hyperdissonant
opening
thematic exposition
 Figure 5.55. Etude-Tableaux in D major, Phrygian events in the main theme, mm. 6–14

6    
                    
                       
              

mf

p cresc.
  
        
           
      


Phrygian events ^
2

223
10             
 
                                
                    
        
ff f ff
  
 
     
        
    

    
        
 

 Figure 5.56. Etude-Tableaux in D major, overview

Section A1 Section B Section A2

Introduction Transition

1-4 5 14 - 17 29 34 - 39 40 41 - 62 63 - 69 70 79 92 - 93 94 - 97
quasi quasi quasi quasi
HEX(1,2) HEX(0,1) HEX(3,4) HEX(1,2)

Diatonic

224
D major C major G major D major
Phrygian Phrygian D major

T IV V
I
D
Phrygian
climax event
V
 Figure 5.57. Etude-Tableaux in D major, analysis of climax in section A2, mm. 78–83

CLIMAX
^2

78
                           
                
                          


ff

dim.
  
           
     
           
            

   
  
 II

225

81
      
                
     
 
 
             
            
mf    cresc. ff
        
       
         
  
 

I

SD T
 Figure 5.58. Etude-Tableaux in D major, mm. 89–92

ygian ascent quasi - HEX(1,2)

complex harmonization of Phr

89
               
                     
        

               
 
          
  
cresc.
     

226
       
 
                   
             
       
                
         
         
V7 I V7 I V7 I 

D T D T D T

 The climax of the etude at measure 79 in section A2 is shown in both Figure 5.56
and Figure 5.57, and can be interpreted as a culmination of Phrygian organization. The
climax event—E♭ major in a larger D major context—is not unlike those analyzed earlier
in the dissertation in the first movement of the Piano Concerto No. 3 (see Figures 4.7
through 4.11) and second movement of The Bells (see Figures 4.25 through 4.28). As
shown in Figure 5.58, the final measures of the etude emerge from a reprise of the quasi-
hexatonic structure following a complex passage in which a clear dominant-tonic pattern
in one layer is set against a striking chromatic harmonization of a Phrygian ascent in
another layer.
It is interesting to note that the last movement of the last work of Rachmaninoff’s
exile period (the third movement of the Symphonic Dances, Op. 45)—and therefore the
last product of his career as a composer—and the last work in the last opus of the late
Russian period (the D major Etude-Tableaux just analyzed) are similar in several ways.
As shown in Figure 5.59, the third movement of the Symphonic Dances begins in D
Phrygian major with material derived from the Dies irae. As in several examples above,
the Phrygian tonic has a dominant function embedded in it. The exposition of the main
theme (also derived from the Dies irae) at rehearsal 58 is plainly derived from the
opening measures of the introduction. As discussed in Chapter 6, Phrygian organization
reaches a zenith in the central episode of the dance, interfacing with octatonic structure in
a climactic moment of extreme hyperdissonance that integrates the Dies irae, the
Phrygian mode, octatonic organization, and Rachmaninoff’s favorite “marked” key area,
D♭ major.

227
 Figure 5.59. Symphonic Dances, Op. 45, iii, analytic reduction of mm. 1–30

Introduction
Allegro vivace
Lento assai
Dies irae Dies irae
         56     
                  
 
                 

p p
        
sff sff
f

   (iv) (I)
      
(V)
 
               
(iv)
I
Phrygian

228
      
   
57
          
 
           
sff p sff
ff pesante      
     
              
                   
ff

(HEX)      

Phrygian
 Figure 5.60. Symphonic Dances, iii, analytic reduction of main theme at r. 58

-./0*1,$23

!" !"#$%"&'#
! "" # # # # $ # # # %# # $# #
# !"#$%"&'#

! # # # $# %# # $# %# # #

#$%&'()*+,
& "" #
" #

A final brief analysis will pave the way to the last chapter of the dissertation and
bring the exposition of technical material to a close. Figure 5.61 contains an analytic
reduction of the opening measures of the second movement of the Symphony No. 3, Op.
44. As indicated on the figure, Phrygian C♯ organization accompanies a statement of the
symphony’s Phrygian-inflected motto theme in C♯. (Note that Rachmaninoff provides no
key signature for the movement, suggesting that its complex harmonic structure is to be
understood in the larger context of the symphony’s A minor/major, and suggesting a
certain amount of tonal flux.) As may be expected in a strong Phrygian passage, C♯ has a
double meaning: it is Phrygian tonic, and it is the dominant of F♯, the focus on which
however is to some undermined by a preponderance of extended tertian sonorities—in
fact, only with the resolution to C♯ major in measure 10 is a clear triad articulated.41
Figure 5.62 contains a reduction of the final measures in the movement, showing a
statement of the motto theme on C♯ as a closing gesture, following some forty-one
measures in which C♯ is essentially never absent. As discussed earlier in the present
chapter and in Chapter 4, the motto theme is stated once more on C♯ at rehearsal 80 in
the third movement. The C♯ “problem,” so to speak, is ultimately solved only in the coda
of the third movement.

41
Cannata treats the movement as simply in the key of F♯ (Rachmaninoff and the Symphony, 127-28),
though he states that it is F♯ minor (which is defensible) on one page (127) and F♯ major (which is
indefensible) on another page (128).

229
 Figure 5.61. Symphony No. 3, Op. 44, ii, analysis of mm. 1–14

local highpoint: tones of Phrygian motto verticalized

36
Phrygian motto
6 10 14

       
      
horn
p f dim p
*
violin solo

    
           
            
      

230
  
winds

harp
         

      
  
   
  
      

  


SD T
F  minor

i9 V

iv9 I
C  Phrygian major
 Figure 5.62. Symphony No. 3, ii, last three measures

'()*%#+,-&..&,&+,/ $

!"#$%&

! !"#$%&!'($))* '
! " % % &
# # $# # # # # # $# # # # # # $#

* *
( $$))
+$%,!
)) )) $
$ )) '
+
+
$ $) ) ) $ $) #
) ) ) ) #, +
!
* *
# $#
-.#(
% # $# # # '
(
" " # # $# # # # # # $# # # # # %
# $# % &
!"#$%&!'($))*
! !"#$ !!

The most straightforward tonal explanation of the material in Figure 5.61 is that it
establishes the dominant of F♯ minor. But this explanation seems quite insufficient in the
last third of the movement, as C♯ becomes increasingly the focus, and especially in the
closing measures of the movement, as the Phrygian motto theme is stated on C♯ in a
manner analogous to its statement on A♮ at the end of the first movement (where it is
undeniably a tonic). A better explanation of the material in Figure 5.61 is that it strongly
establishes C♯ as a pitch center, and that the Phrygian context allows an interaction of
tonic and dominant functions within and around that center. It is too simple to say, as
Cannata does, that C♯ in the second movement of the symphony is simply the dominant
of F♯.42 Such a view misses the significance of Phrygian organization in Rachmaninoff’s
oeuvre generally, the central role C♯ plays in the symphony more specifically, and the
absence of a key signature in the movement most specifically.
A tension seems to be inherent in Phrygian organization—especially Phrygian
organization around a major tonic, as for example in the fourth movement of Capriccio
Espagnol, in “Polichinelle,” in the Etude-Tableaux in D major, in the third movement of

42
Cannata, Rachmaninoff and the Symphony, 127-28.

231
the Symphonic Dances, and in the second movement of the Symphony No. 3. As already
shown in the analyses of the second movement of The Bells and the first movement of the
Concerto No. 3, a tug of war between the tonic note and the lowered second scale
degree—the Phrygian tone—is not infrequently associated with climax even when
Phrygian modal organization per se is not otherwise strongly indicated in a passage or a
movement. The rhetorical and expressive associations of Phrygian structures are
unusually complex: clearly modal, and frequently expository or initiating; yet also a
stimulus for climax, which makes it unlike other modal structures in Rachmaninoff’s
works.

Conclusion

By recognizing types of modal organization whose significance is not

acknowledged in existing Rachmaninoff scholarship (peremennost, nega, and the
Phrygian mode), it has been possible in the present chapter to amplify the framework of
rhetorical associations laid out in Chapter 3. Although peremennost and nega are similar
in some regards to certain tonal formations used in mainstream European music of the
late nineteenth century, they remain at least implicitly “Russian” in Rachmaninoff’s
mature works. Modality emerges as more than an adornment of ordinary diatonic-
functional tonal syntax. Modal structures are marked in the works studied, and their
contributions to form, in large-scale tonal design, and in expressive trajectory are
generally different from the contributions of the functional basis and “fantastic”
chromaticism.

Summary of Chapters 4 and 5

Whereas Chapters 2 and 3 of the dissertation constitute an interpretive apparatus

suitable for Rachmaninoff’s late works, Chapters 4 and 5 constitute a technical apparatus.
In Chapter 4, equal-interval structures originating in traditional Russian representations
of the “fantastic” in music were described. Analysis of many works demonstrated

232
Rachmaninoff’s extensive use of octatonic, hexatonic, whole-tone, and hybrid structures
throughout the late Russian and exile periods. In Chapter 5, Rachmaninoff’s use of the
church modes, of peremennost-based diatonic oscillation and superimposition techniques,
of the expressively-packed nega idiom, and of complex Phrygian structures was detailed.
The analyses in Chapter 5 also featured increasingly complex combinations of functional
tonal, equal-interval chromatic, and modal structures, laying groundwork for the more
comprehensive analyses in Chapter 6.

233
 Chapter 6

Climax in the Last Three Concert Works

In this final chapter, the focus shifts from microscopic descriptions of individual
passages such as those in Chapters 2 through 5 to a macroscopic consideration of
Rachmaninoff’s last three compositions—Rhapsody on a Theme by Paganini, Op. 43,
Symphony No. 3 in A minor, Op. 44, and Symphonic Dances, Op. 45—as exemplars of a
hyperdissonance-oriented approach to large-scale Postromantic form. This chapter is
therefore an application of the interpretive and technical apparatuses developed in the
dissertation to large works in their entireties. Several passages from Opp. 43, 44, and 45
were analyzed in Chapters 2 – 5. This material is reviewed and expanded in the following
pages as it is incorporated into more comprehensive analyses. Rachmaninoff’s last three
opuses, composed in bursts of activity between 1934 and 1940, represent in many ways a
culmination—the composer’s own word seems appropriate—of his entire oeuvre. In
these works can be heard a powerful synthesis of four threads that had occupied
Rachmaninoff increasingly after the watershed works of 1909:

1. Complex combinations of functional tonal structures, equal-interval structures,

and modal structures within the general framework of rhetorical and expressive
associations described in Chapters 2 through 5.

2. Emphasis on points of intense hyperdissonant exaggeration and hyperdissonant

distortion as a way to articulate large form in a Postromantic—that is to say, in a
deformation-oriented—melodic and harmonic context.

3. Extensive use of the Dies irae as thematic material.

4. Emphasis on the region of D♭ major as a structural linchpin and expressive crux,

regardless of the global tonic of the work.

234
 Items 1 and 2 above involve techniques and tendencies described at length in the
preceding five chapters. Items 3 and 4 need special discussion here.

The Dies irae in Opp. 43, 44, and 45

Rachmaninoff’s use of the Dies irae melodic incipit—four notes, which might be
extended as a generic sequential pattern—was shown in several analyses in earlier
chapters.1 (See again the analysis of the second and fourth movements of The Bells in
Chapters 4 and 5, and the analytic overview of the fugal episode in the finale of the
Symphony No. 3 in Chapter 4.) The incipit is used prominently in many compositions
from the Symphony No. 1, Op. 13 on; yet Rachmaninoff apparently had no substantial
direct knowledge of the chant until after he completed the Corelli Variations, Op. 42 in
1931:

Shortly after the composition of the Corelli Variations Rachmaninoff had at last
tried to find out more about the theme which had never ceased to haunt him since the
disaster of the First Symphony and about which, paradoxically, he was still ignorant,
asking the musicologist Joseph Yasser about its origins, its full form (Rachmaninoff
invariably quotes only its opening phrase) and its meaning, without giving him any
clues as to why.2

Rachmaninoff’s familiarity with the Dies irae before his communication with
Yasser seems to have been gathered mainly from the general concert repertory, in which
the chant incipit had long been used as a motivic signal for death, judgment, and so on. In
the Rhapsody on a Theme by Paganini, which was the first work composed after the
Corelli Variations, the appearance of the Dies irae in something closer to its actual chant
form—a distinct opening phrase of seven notes rather than a generic set of four notes that
might be extended sequentially—is probably a result of the composer’s correspondence
with Yasser.

1
Additional research on appearances of the Dies irae in Rachmaninoff’s and other composers’ works was
cited in Chapter 1, and is listed in the bibliography.
2
Martyn, Rachmaninoff, 329.

235
 In the finale of the Symphony No. 3 and the third movement of the Symphonic
Dances, the Dies irae effectively displaces all other thematic material, as discussed in
more detail below. However, as stated in Chapter 1, no rigorous basis for understanding
Rachmaninoff’s frequent recourse to the Dies irae in the late works suggests itself.
Rachmaninoff claimed a quasi-programmatic meaning for the Dies irae in the Rhapsody
in a letter written to choreographer Michael Fokine when a ballet version of the work was
being planned: “all variations on the Dies irae would be for the evil spirit.”3 But the
spirited, even celebratory nature of the chant’s treatment in the third movement of the
Symphonic Dances and at the end of the final of the symphony suggest that a view in
which the chant is a straightforward sign for evil and/or death is too limited. In the end,
despite the consistency with which it appears, the Dies irae remains something of an
enigma in Rachmaninoff’s works.

D♭-Major Focal Points

Crucial to the analyses in this chapter is “D♭ major”—as a key area, as a concept.
The key of D♭ major has a special significance in Rachmaninoff’s oeuvre as a whole,
shared to a lesser extent by the enharmonic parallel minor, C♯ minor. It is the global tonic
of several important works. The most famous early composition is the Prelude in C♯
minor, Op. 3, No. 2 (1892). Eighteen years later, Rachmaninoff concluded his cycle of
twenty-four preludes in the key of D♭ major (Op. 32, No. 13), borrowing motivic material
from the early C♯ minor piece for the last prelude and thereby making a pair of bookends
for the cycle. The fourth movement of the The Bells begins in C♯ minor and ends in D♭
major, and Cannata has suggested that the entire four movement composition can be
heard in the gamut of D♭.4 (See again the analysis of the fourth movement in Chapter 5.)

3
The letter dates from 1937 and is published in Bertensson and Leyda, Sergei Rachmaninoff, 333.
4
Cannata, Rachmaninoff and the Symphony, 83-87.

236
 Figure 6.1. Marked D♭ events in well-known Rachmaninoff works from all periods
Work Movement Key Event or Significance

Prelude No.1 in C♯ minor, Op.3, No.2 C♯ minor First published prelude

Trio élégaique in D minor, Op.9 ii F major Lyric episode in D♭ major
Symphony No.1 in D minor, Op.13 ii F major D♭ major as a hexatonic partner to F major throughout lyric movement
Concerto No.2 in C minor, Op.18 ii E major Climax on C♯ major
iii C minor Lyric second theme in D♭ major in recapitulation
Chopin Variations, Op.22 C minor Lyric penultimate variation (XXI) in D♭ major (derived from theme)
Sonata No.1 in D minor, Op.28 i D minor Climax on D♭ major directly before recapitulation
Concerto No.3 in D minor, Op.30 ii * Lyric movement involves an interaction of D♭ major and F♯ minor
* probably a prototype for the second movement of Op.44
Prelude No.24 in D♭ major, Op.32, No.13 D♭ major Last published prelude
“Vocalise,” Op.34, No.14 C♯ minor Lyric final song of the set is in C♯ minor
The Bells, Op.35 D♭ major Entire composition organized around final culmination in D♭ major

237
Sonata No.2 in B♭ minor, Op.36 i B♭ minor Lyric second theme in D♭ major in exposition
All-Night Vigil, Op.37, v B♭ minor Climax on D♭ major
“Daisies,” Op.38, No.3 F major Climax on D♭ major
“Dreams,” Op.38, No.5 D♭ major
“A-u!” Op.38, No.6 D♭ major Last song cycle closes in lyric D♭ major
Concerto No.4 in G minor, Op.40 iii G minor Lyric second theme in D♭ major in exposition
Corelli Variations, Op.42 D minor Lyric variations (XIV and XV) in D♭ major
Rhapsody on a Theme by Paganini, Op.43 A minor Lyric variation XVIII in D♭ major; just one variation in the key
Symphony No.3 in A minor, Op.44 i A minor Central climax event suggests D♭; climax on D♭ major in recapitulation
ii * * Lyric movement involves an interaction of C♯ Phrygian major and F♯ minor
iii A major Lyric second theme in exposition involves C♯ minor
Fugue/development emerges from C♯ octatonic event
Symphonic Dances, Op.45 i C minor Lyric middle section in C♯ minor; climax on D♭ major
ii G minor (Middle section skirts D♭ major, but deflects)
iii D major Lyric middle section in D♭ major
 Even more significant are the many interior climax events and expressively-
packed lyric episodes on or in D♭ major in works from all four periods. Figure 6.1 is a list
of important events involving D♭/C♯ major/minor in a number of well-known works. The
list suggests the special significance this key area had for the composer. Of principal
interest at present are occasions in the large concert works composed after 1926 when
marked D♭ major moments (climaxes or lyric episodes/movements) emerge in the
contexts of D minor, A minor, G minor, and so on—that is to say, in keys with which D♭
major is only distantly related.5 As shown in Figure 6.1, a majority of Rachmaninoff’s
large instrumental works are in natural keys, in which contexts D♭ major sounds very
striking.
D♭ major looms throughout the analyses in this chapter. It is an expressive and
structural focal point in Rachmaninoff’s last works to such a degree that conventional,
relativistic key relationships (expressible in generic terms by Roman numerals, e.g. “I –
V” or “i – III”) seem replaced in part by a kind of “absolute” tonal organization in which
D♭ major emerges as a setting for important events regardless of what overall key is in
use. None of the last concert works is set in the key of D♭ major. Yet the Rhapsody, the
Symphony No. 3, and the Symphonic Dances all revolve in large part around core events
in or about D♭ major. It becomes not a question of if D♭ major will appear, but a question
of in what larger context it will appear, and through what technical means its role in
structure will be articulated. In Opp. 43, 44, and 45, D♭ major represents a realm of the
interior—distant, often lyrical, usually introspective by comparison with the more active
music on either side, and expressively packed.

Organization of the Analyses

Limitations of space make complete analytic reductions of Opp. 43, 44, and 45
impossible. (The three works take up approximately 480 pages in full score.) Myriad

5
For comments on a general trend in the nineteenth and early twentieth century music toward extreme flat
and sharp keys, see Hugh MacDonald, “[G-Flat Major Key Signature],” 19th-Century Music 11 (1988):
221-37. For related but more abstract comments, see Bertold Hoeckner, “Schumann and Romantic
Distance,” Journal of the American Musicological Society 50 (1997): 55-132.

238
features of interest must go without comment in the interest of holistic treatment. Instead,
in this chapter, synopses of form, thematic material, and tonal design on the large scale
provide frameworks in which analytic snapshots of structurally significant climax events
may be meaningful. Here, Rachmaninoff’s theory of “culminating points,” discussed in
Chapter 1, comes to fruition.

Figure 6.2. Synthesis of thematic material, form, and hyperdissonant climax events

'01.&'%().&'1+%&-

'1'0-2(.*$-'!-

!"!#$%&'"!%()*'+,(',+&-)%./-%(&'%"!*

!"#$%&'()!*+$%,-)*#.)#*!- /01$%,-)*#.)#*!-

05/1+$%**"!&!')(-%.&6)171!'*

2-"4&-)'"!&-)*'+,(',+1
-&+21)3"+.

*'+,(',+&-)+1*"-,'%"!
3%!&-)(,-.%!&'%"!

239
 In each of the following analyses, I show a clear correlation between large-scale
climax events (at most two or three in an entire composition) and large-scale
hyperdissonance events resulting from strong articulation of specific “fantastic”
chromatic and special modal structures at formal and expressive junctures. It is not
necessary to account for every note in each work to demonstrate how such climaxes are
developed. It is necessary only to show how Rachmaninoff unifies these large
compositions by drawing thematic material, harmonic structure, large-scale form and
tonal design, and hyperdissonance climax events into close associations. One way to
conceptualize such a synthesis is shown in Figure 6.2.
As suggested in Figure 6.2, the thematic materials used in Opp. 43, 44, and 45
have idiosyncracies that suggest various non-diatonic structures. The modal and
chromatic structures that result engender hyperdissonant climax events by complicating
or deforming a work’s large-scale tension arc (that is to say, its tonally- and formally-
derived trajectory of departure and return, as explored in Chapter 1). Structural resolution
is achieved at some late stage in a work as the melodic and harmonic components
involved are brought into a more harmonious arrangement, solving problems and
conflicts established earlier in the work. Rachmaninoff’s Postromantic aesthetic position,
as opposed to a Romantic position or to a modernist position, is clear: conflict,
fragmentation, distortion, and exaggeration beyond the boundaries of the Romantic, yes;
but also, in the end, unity of design and structural resolutions tied to conventional roots
that true modernists tried to sever.
With regard to the last point in the preceding paragraph, Olin Downes’s early
review of the Symphony No. 3 is compelling: “There is the impression of frustrated
strength, which gathers, to crash helplessly against some obstacle… His idiom is more
his own than ever before, and free of the indebtedness it once had to Tchaikovsky.”6
Jason T. Stell has noted in his study of Rachmaninoff’s piano works that the composer’s
practice setting up of musical obstacles to be overcome climactically later in a work
resembles processes in Bruckner’s music.7 Stell notes Warren Darcy’s study of “blocked
tendencies” and sonata deformations in Bruckner’s symphonies, specifically noting

6
Quoted in Bertensson and Leyda, Sergei Rachmaninoff, 325. Downes’s review, however, is generally
negative.
7
Stell, “Rachmaninov’s Expressive Strategies,” 15.

240
Darcy’s comment that “Bruckner causes all these achievements to converge and resonate
sympathetically in a climactic moment of splendor.”8 Yet, as the following analyses
show, the results here are uniquely Rachmaninoffian, insofar as they involve alloys of
functional tonal, “fantastic” chromatic and special modal structures that are unique to his
mature style.

Rhapsody on a Theme by Paganini, Op. 43 (1934)

The Rhapsody has a double-function form. The twenty-four variations (plus

Introduction and Tema) are played essentially without a break, forming a single large
movement. This large movement, however, may be understood as simultaneously
suggesting a multi-movement plan. Although Martyn claims that the Rhapsody “divides
naturally into three sections, corresponding to the form of a sonata or concerto,” several
factors suggest the four-movement plan shown in Figure 6.3.9
Martyn’s analysis treats Variations XII through XVIII as a single movement.
Indeed, Rachmaninoff had made a single hybrid movement from slow movement and
scherzo in the Piano Concerto No. 3, and he would do so again in the second movement
of the Symphony No. 3. However, in the Rhapsody, tempos and key structures (D minor
and F major as a pair of relatives, B♭ minor and D♭ major as a second pair of relatives)
suggest that Variations XII through XVIII represent two distinct stages in the
composition. Further support for a four-movement interpretation is provided by the codas
appended to the ends of Variation XV (at rehearsal 40) and Variation XVIII (at rehearsal
51), which add material not suggested by the structure of Paganini’s theme and not
included in any other variations. The codas strongly suggest that Variations XV and
XVIII should be heard as concluding utterances in separate internal movements.

8
Ibid., citing Warren Darcy, “Bruckner’s Sonata Deformations,” in Bruckner Studies, ed. Timothy L.
Jackson and Paul Hawkshaw (New York: Cambridge University Press, 1997), 77.
9
Martyn, Rachmaninoff, 328.

241
 Figure 6.3. Rhapsody on a Theme by Paganini as a four-movement structure

First movement Introduction A minor

Variation I (precedente)
Tema
Variations II - X

transition Variation XI cadenza #1

Minuet / Scherzo Variations XII - XV D minor

F major

Slow movement Variations XVI - XVIII B ! minor

D ! major Climax #1

Finale Variations XIX - XXIV A minor

Climax #2
cadenza #2
cadenza #3

As shown in Figure 6.3, Variation XI may be interpreted as a transitional,

modulating episode between the first and second movements. Variation XI is also the
first of three cadenzas. The second cadenza is at the end of Variation XXII, and the third
cadenza is at the end of Variation XXIII. The cadenzas punctuate stages in a large
trajectory of tonal departure through the slow movement (motion to regions increasingly
distant from A minor throughout Variations XII and XVIII) and return (with Variation
XIX and following) that forms a basis for interpretation of two main climax events in the
work. As shown in Figure 6.4, the first climax event occurs in Variation XVIII—the
point of furthest remove in the work, and the only variation in D♭ major. The second,
arguably more powerful climax event occurs in Variation XXII—around the point of
return. Climax #1 is associated with a hexatonic structure and with nega; Climax #2 is
associated with an octatonic structure.
In addition to the double-function form outlined above, I propose another
interpretation of the Rhapsody’s form: the large-scale structure of the work, including the

242
two climax events and the overall trajectory of departure and return, closely follows the
structure of Paganini’s theme, to such an extent that the work as a whole may be
considered an enormous variation on the theme. This is shown in more detail in Figure
6.5.10

Figure 6.4. Overview of main climax events in Rhapsody on a Theme by Paganini

()*+",%-./ ()*+",%-:/
01*23%14%45#36783%#7+197 01*23%14%#735#2

!"#$%&!''' !"#$%&&''
2
#" )%37
28*1 !
>%%%+"?1#
<35
");83#5
312

=%+*21# !"#$%&'()*+",'-.+'.-$ !"(+'&'()*+",'-.+'.-$

!"/0(1&0"'()*+")('$"'-$&'$2"&,".),'&10$"!"#$%3127 !"2$-*3$2"4-(5"6)'-(2.+'*()

The figure reveals strong correspondences between the theme (A) and the
Rhapsody as a whole (B). The first stage of activity in both establishes A minor as tonic.
Tonal departure begins with motion to the subdominant as part of a sequence. In both,
there is a provisional return to A minor (measure 10 in the theme; Variation XIX in the
Rhapsody), followed by a more powerful functional affirmation of A minor as tonic. The
three cadenzas in the Rhapsody (see again Figure 6.3) are indicated by fermatas on Figure
6.5. As dotted lines on Figure 6.5 show, the cadenzas are associated with the moments of
chromatic activity in Paganini’s theme. As explained more fully below, cadenza #1
establishes nega in the work by bringing pitch class B♭ into play; and cadenzas #2 and #3
are based on pitch classes E♭ and F, which may be interpreted as large-scale
manifestations of the two tones articulating the Italian augmented sixth chord in
Paganini’s theme, at the corresponding location in the large form of the Rhapsody. In
other words, specific chromatic moments in Paganini’s theme provide a basis for specific
large-scale structural features in the work.

10
For purposes of comparison, see again Heinrich Schenker’s analysis of the theme (Figure 2.16).

243
 Figure 6.5. Comparison of (A) Paganini’s theme and (B) Rhapsody on a Theme by Paganini

244
Climax #1

However, Figure 6.5 also reveals a crucial difference between the structure of
Paganini’s theme and the structure of the Rhapsody. Whereas the sequential passage in
Paganini’s theme is a conventional circle of fifths motion, the sequential structure at the
corresponding location in the Rhapsody is a HEX(0,1) structure. Recall from Chapter 2 that
hexatonicism is introduced as an intensifying device in Variations VIII and IX.11 In
Figure 6.5 (B), the basically diatonic framework of Paganini’s theme is deformed
hexatonically; the theme’s pattern of departure and return is thereby greatly exaggerated.
As a result, D♭ major emerges as the point of greatest difference between the Rhapsody’s
structure and Paganini’s theme, and the point of furthest remove in the Rhapsody. As
observed above, Rachmaninoff includes only a single variation in the key, thereby
bringing its special status into focus.
An earlier, fleeting suggestion of D♭ major may be heard at rehearsal 31 in
transitional Variation XI (the first cadenza), as shown in Figure 6.6. This tonal
foreshadowing is concomitant with a statement of a nega figure that will be featured
throughout the flat-key slow movement variations (XVI – XVIII). Figure 6.6 shows an
entanglement of A minor/major (the tonic of the preceding variations, which emerges as
the dominant of D minor), D minor (the tonic of Variations XII and XIII), and D♭ major
(something yet to come, later the setting of a major climax event), underneath the nega-
type melodic figure.
As suggested in Figure 6.7, the nega figure is derived from the B♭ in the fifth
measure of Paganini’s theme. The nega configuration of the melodic gesture—A♮ as an
unstable tone oscillating up to B♭ and down to G♮/ B♭—is actually first heard in Variation
VIII, where it is clearly associated with the introduction of hexatonicism. Recall from
Chapter 5 that peremennost and nega techniques strongly characterize the B♭ minor
portion of the Rhapsody (Variations XVI and XVII), bringing these variations into close
association with the D♭ major that is to come. (See again Figures 5.37 and 5.38, recalling
that nega idiom is generally associated with lyric movements, or lyric episodes and
digressions.

11
See again Figures 2.14–2.17.

245
Figure 6.6. Rhapsody on a Theme by Paganini, Variation XI, nega figure at r. 31

!"#$%0*1%'"2,3".4)*

!" )*!*#+*'
'#
!
(
!"#$%
#
&"$'(
# " &" "
""""" ",

/
" "&" "$" %""" &*" " + *
'
%" "
,
! " &" "
$ """ )
!" !"
#$%&'%(%)*+&,
- ,,
! '' ""
' "
'# ,
& '' ##
.
" %%%''
' "" ,
&
"
&
&"$'(-,(+)"$.(
5,,,0#6%)

7,0"$%)
0#6%) 5,0"$%)

Figure 6.7. Derivation of nega figure from Paganini’s theme

(a) Paganini’s theme, measures 5 - 6

" " " $" " " "

! # #
[

(b) nega figure + HEX in Variation VIII (at rehearsal 31)

! " $" " %"

" "
&""
$$ ""

(c) nega figure in Variation XI (after rehearsal 21)

'
! " $" " %" &"" " $" " "

246
 As shown in Figure 6.5, the emphasis on nega in Variations XVI and XVII occurs
within a framework of large-scale hexatonic exaggeration resulting from chromatic
deformation of the sequential episode in Paganini’s theme. To put it another way: as the
Rhapsody approaches its hexatonically-defined point of furthest remove, the tonic note
A♮ is treated as an unstable nega tone, suggesting a structural hyperdissonance involving
the tonic. Throughout Variation XVII, not just pitch class A♮ but whole triads on root A♮
are repeatedly articulated. Rachmaninoff keeps the global tonic triad in circulation even
in the distant key of B♭ minor, calling attention to the chromatic distance traveled.
Nega and HEX(0,1) come together powerfully at Climax #1 in Variation XVIII.
Figure 6.8 is an analytic reduction of the variation. As is well known, the famous melody
in Variation XVIII results from the inversion of Paganini’s theme.12 (Recall from Chapter
4 that the F major variations, XIV and XV featured partial inversion of the theme along
with hexatonic structures.13 Variation XVIII represents a more complete inversion.) The
inversion of the opening decorated arpeggio figure (x) in Variation XVIII is indicated on
Figure 6.8.14 Inversion procedures are continued throughout the variation. Whereas the
theme features a descending sequence leading to cadence, Variation XVIII features an
ascending sequence. The theme is treated doubly in the variation, resulting in a two-
phrase periodic structure. Phrase 1 cadences on the dominant. Climactic phrase 2
cadences on the tonic. A coda follows, featuring continued play on the nega tones (A♭ -
A♮ - B♭) above a post-climactic D♭ pedal point (not shown on Figure 6.8).
As shown in Figure 6.8, climax #1 may be understood an integration of a HEX(0,1)
cycle (A minor – F minor – D♭ major), which reflects the large-scale tonal design of the
12
Cannata’s study of draft materials has revealed that a sketch of the inversion of the theme was among the
first things Rachmaninoff produced when beginning work on the Rhapsody, and that the sketch dates from
the 1920s (Cannata, Rachmaninoff and the Symphony, 55–57). Paul Vining and Rollo Piaf have suggested
that the melody of Variation XVIII is also derived from the Credo of William Byrd’s Mass for three voices,
a performance of which they claim Rachmaninoff directed at Foulis Castle in Scotland in 1934 (before
composing the Rhapsody). See Paul Vining and Rollo Piaf, “Byrd Plagiarized,” Musical Times 118 (1977):
300. I have been unable to confirm Rachmaninoff’s participation in the Byrd performance, and Cannata’s
research suggests that the inversion of Paganini’s theme had been sketched before the 1934 Byrd
performance. If Vining and Piaf’s report is accurate, however, it might add another layer of meaning to
Rachmaninoff’s famous melody.
13
See again Figures 4.33 and 4.34.
14
The subdominant / tonic oscillations and subdominant-dominant hybrid / tonic oscillations in the opening
measures of the variation may also be interpreted as a kind of inversion—that is to say, a harmonic
inversion of the dominant / tonic oscillation featured in the opening measures of Paganini’s theme.

247
entire opus, and the nega figure from Variations XI and XVII in the key of D♭ major. The
harmonic material of the climax is clearly stratified: the HEX(0,1) cycle is articulated in
the upper register, while a functional D♭ major progression is propelled by the bass
motion. The most intensely dissonant element at the climax is the A minor triad. The
climax is therefore a hyperdissonant exaggeration in which the global tonic is entangled
in a hexatonic structure at a point of furthest remove at the end of a hexatonic structure
on a much larger scale. The event recalls the climax of the song “Daisies,” Op. 38, No. 3,
which also features an unstable tonic sonority at a hexatonically produced point of
remove.15
At the climax in Variation XVIII, the global tonic is treated as an unstable
element in both the nega idiom and the hexatonic structure that resolves to D♭ major.
Figure 6.8 also shows how the melodic structure of the sequential/climactic portion in
both phrases may be interpreted as articulating a large neighbor tone that seems to belong
more to A minor than to the local tonic D♭ major: C♮ – B♮ – C♮ in phrase 1, F♮ – E♮ – F♮
in phrase 2. The F♮ – E♮ – F♮ figure specifically recalls the opening measures of the
Rhapsody, which featured a recurring E♮ – F♮ – E♮ melodic figure. The connection
between the opening measures of the Introduction and Variation XVIII is shown more
clearly in Figure 6.9. The first chord heard in the Rhapsody—an A minor triad with
F♯/G♭ underneath—returns at the Variation XVIII climax, but in entirely different tonal
circumstances. The relative values of E♮ and F♮ are reversed: F♮ resolves to E♮ in the
Introduction, but E♮ (associated with tonic A minor) resolves to F♮ in Variation XVIII.
Figure 6.10 shows how this is part of a more general tonal “inversion” that characterizes
Variations XVI through XVIII as a whole. Pitch classes which are stable in the key of A
minor are, on the contrary, highly charged in B♭ minor and D♭ major, and vice versa—
including, because of the nega idiom in used throughout the flat-key variations, the tonic
A♮.

15
See again Figure 2.32.

248
 Figure 6.8. Rhapsody on a Theme by Paganini, Variation XVIII, analytic overview of climax #1

Phrase 1 sequence begins on V C-B-C

x (inverted)
              
   
         
 
                           
     
cresc. f mf

       
        




  

    
  
ascending fifths sequence PD D T

T D


249
HEX(0,1)
Climax
Phrase 2
octave doublings omitted sequence begins on I F-E-F 51
*
50

            
      
                   
   
         
 
       
 
             

mf cresc. nega ff

            

      
 
 
  

     

ascending sequence (altered) PD D

coda above post-climactic pedal point follows

T T

 Figure 6.9. Climax in Variation XVIII, analytic details

!"#$%&'()*+*,-
6,5789)5/):8158052/)#:...
&
&' & && & & ***)))) . - % && & & & ))))
%%%%% #"
!"#$%&!
(

#
$
./01234*052/ ./ .
- & *) . &
- & *& % && &&& &&&& &&&
)) &&&
-
%%%%% #" && & &&'' & && && **))) && &&& &&
. '$(%)
!
$" & , !*"& , && %&& && && )) && & &
& & & &&&( + &&( , ,
** &&& &
& && & , && & & , & & ,
,
/& & & & &
! !!
! !"#$!
&
0 !" 1 / , &/ , ) %%%%% #" ,,
)
&
& , , %& *&& $ *& 0 , %% && && , &
% & &&& &&& $ &
&& &&& &&&
0
%&
** && * &&& & % &
+& +) & ) && &
)
!"#$%&!*(%+*,)#%! .) +) )
'$(%)
;2/5*)<4/*052/ 5/0-/=-,+)35==2/8/0
>8172/5*8,,+)*>81?-3

Figure 6.10. Melodic and tonal “inversion” in Variations XVI – XVIII

!"#$%&'&

&" $ #" " #" $ " $%&()

! #" $ %" !""""""# #%"" &$

'" %" &" $

! %$ &" %$ %" &" %$ &$ %%""
!"#$"$%&'
("#)!*+!,+-./#0)1#2#0)111
3/!4&5#6-*#78+'!9#:"

Climax #2

The Introduction to the work also provides a basis for interpreting the second
climax event, which occurs in Variation XXII. In Chapter 3, an OCT(0,1) cycle in the

250
Introduction was analyzed.16 A tension between the A minor tonic and the octatonic cycle
characterizes these measures, building to a direct clash between A minor and E♭ in
measure 7. As shown in Figure 6.11, a large version of the same conflict is involved at
the Variation XXII climax.
The Dies irae returns in Variation XXII before the climax. It is set chromatically,
building in intensity until A minor is “broken” after rehearsal 64. The A major-F♯ minor
compound at rehearsal 67 (again recalling the very first chord of the composition) is
extremely unstable in the context of the E♭ major-minor seventh chord that is the basis of
those measures. As in climax #1 (Variation XVIII), a projection of the global tonic is
highly charged; but in Variation XXII, the climax occurs at the point of return in the
large-scale departure-return arc, not the point of remove. Climax #2 is therefore
interpreted as an large-scale hyperdissonant distortion according to the criteria
established in Chapter 2. The second cadenza in the work follows, expanding E♭. The
beginning of Variation XXIII capitalizes on the tonal dislocation effected in climax #2
and in the cadenza, as the piano and orchestra momentarily disagree about what key to
play in: the orchestra returns to A minor, but the piano resolves the E♭ seventh chord to
A♭ minor.
As described above, the two main climaxes in the Rhapsody may be understood as
involving hexatonic structures, octatonic structures, and nega ultimately developed from
marked chromatic tones in Paganini’s theme: B♭ and D♯/E♭. As shown in Figure 6.12, a
final statement of the Dies irae in Variation XXIV synthesizes these two pitch classes
firmly into the gamut of A minor (they are marked by arrows in the figure), and a
peremennost flourish recalling the now-familiar A/F♯ compound brings the work to an
end. The Dies irae statement may be interpreted as a final moment of culmination. The
“problems” caused by B♭/E♭ are fully resolved, and the hexatonic and octatonic climaxes
are contextualized.

16
See again Figure 3.13.

251
 Figure 6.11. Rhapsody on a Theme by Paganini, Variation XXII, analytic overview of climax #2

61 62 63 64
Dies irae

is m            

atic
  
om
  
r

ch
    

ng
        
 
easi
incr OCT(2,3)
pp ff sff
(Phrygian)

 
       
    

T
A minor tonic “broken” OCT(0,1)

252
hyperdissonant distortion (octatonic): tonic highly dissonant
Climax #2

67 A major 68

65 66 Paganini theme        
      
          
                   

                
 peremennost 
F  minor 

       
        
etc.
 
sff

   
Cadenza #2
Figure 6.12. Rhapsody on a Theme by Paganini, Variation XXIV, final statement of the Dies irae

!" !# "$

!
!"#$%"&'#
! " " " #" " $" " " " $" #" #" %" "" # ""%" %" " "
%%"" % """ $$""
"
% " "
!! ()#&#*#++,$-.
" $ "" "" $ ""
& " "" ""
"
$"" %"" #"" ""
" " " $" #"
" "
(' ' %
% &' %

Symphony No. 3, Op. 44 (1936, rev. 1938)

The Symphony No. 3 is in the same key as the Rhapsody, and echoes of the
earlier work may be heard throughout the symphony. Particularly salient here are the
incorporation of pitch class B♭ into the gamut of A minor/major, the use of HEX(0,1), and
an intervallically complex climax sonority that resembles the A minor/major + F♯
compound used prominently in the Rhapsody (most notably at its climaxes). Several
passages from the symphony were analyzed in other chapters of the dissertation. The
climax at the end of the development in the first movement was discussed in Chapter 2.17
The Phrygian motto theme stated in the opening and closing measures of the first
movement was discussed in Chapter 5.18 The C♯ Phrygian major statement of the motto
theme in the second movement was also described in Chapter 5.19 Modal structures in the
third movement’s second theme area, the octatonic statement of the motto theme at
rehearsal 80 in the third movement, and the large-scale articulation of the motto theme

17
See again Figure 2.8.
18
See again Figures 5.3 and 5.41.
19
See again Figures 5.61 and 5.62.

253
over the course of the central, developmental fugal episode in the movement were
described at various points in Chapters 4 and 5.20
These features can be synthesized into a coherent view of the entire work. In the
following analyses, two main components are highlighted:

1. Phrygian structures referable to the motto theme.

2. A large-scale hexatonic structure suggested by and emerging from the climax
event at the end of the development in the first movement.

The shattering climax event at the end of the development section in the sonata-
form first movement was presented in Chapter 2 as an example of hyperdissonant
distortion at a point of structural return. At the climax, intense equal-interval structures,
punctuated by the appearance of Dies irae-like material at rehearsal 20, undermine a
clearly-articulated return of the tonic (A minor) before rehearsal 22. A snapshot of the
beginning of the event is shown in Figure 6.13.

Figure 6.13. Symphony No. 3, Op. 44, i, hyperdissonant climax at the end of the development

)78*/1-66&+0+'(1-6'&/'-&+(.9-%0:

'&+-.('/-01(2,(%-+&/3
"#$

%&''&(')*%*(&+(, !!
! )
&%%&'%%&%% %% %) * + % % % + %),, #
%% %% %% %%, ## "! ,#
,, ## ,, ###
%, % , ,, ##
,# &%,, # &% , # %-
, ##
$
( ( %% ,# ,# -
!
-

&% * + % % % + &%, # # ,# ,# ,#
!! !!
)
!
"
. +
&% % % % &%/ "! , &% &% + # &-
-
+ + + "
!!

1&%-+0+'(45+.'-&+('&+*6

20
See again Figures 4.12–4.14 and 5.18.

254
 The passage, which lasts over 40 measures (from before rehearsal 22 to rehearsal
25), is of such intensity and of such dimensions that it calls for interpretation well outside
the conventional tonal box. As explored more fully below, it is the central event in the
first movement, on which the movement’s entire structure hinges; and the other two
movements emerge from its shadow. To put it more bluntly, the hyperdissonance of the
climax is not completely resolved until the coda of the third movement.
In the Rhapsody, pitch class B♭, derived from the first chromatic tone in
Paganini’s theme, plays an important structural role in several ways. However,
suggestions of bona fide Phrygian organization are limited to the beginning of Variation
XXII (see again Figure 6.11) and a handful of other locations. In the Symphony No. 3,
the Phrygian potential of the B♭ in the opening statement of the motto theme is much
greater, taking three movements to work out in full.21 Throughout the symphony, the
melodic motive A♮ - B♭ (with or without G♮, which is the other tone in the motto and the
other neighbor tone in the basic Phrygian cell), sometimes accompanied by auxiliary
tones from further on the flat side (especially E♭—again, the Rhapsody seems to be a
precedent), appears with such frequency and in such a variety of contexts that it would be
virtually impossible to list them all. As the following analysis suggests, this may be taken
as surface evidence of a deeper structural concern for integrating pitch class B♭ and the
Phrygian motto theme into the gamut of A minor/major.

The climax at the end of the development in the first movement, revisited

The first movement of the symphony has an outwardly conventional sonata form,
as shown in Figure 6.14. The opening and closing measures of the movement are
characterized by clear statements of the Phrygian motto theme. The proportions of the
movement are fairly balanced, as suggested by the fact that the midpoint of the

21
As cited earlier, the B♭ plays and important role in Cannata’s interpretation of the Symphony, too. For
Cannata, it implies a large-scale, subdominant-oriented double-tonic complex oriented, A minor - D minor,
which is resolved in favor of A in the finale following the D major fugal episode after rehearsal 80
(Cannata, Rachmaninoff and the Symphony, 125-30). Although the fugal episode is undoubtedly
significant, the present analysis suggests a more complex structure for the work as a whole than the one
suggested by Cannata. Cannata struggles to incorporate the second movement into his interpretation; and
largely fails to account for the general absence of significant passages in D minor or D major throughout
the first and second movements.

255
development section (measures 162) and the midpoint of the movement as a whole
(measure 159) are almost the same.

Figure 6.14. Symphony No. 3, i, form

1 – 11 Introduction motto theme: A Phrygian minor

12 – 97a Exposition

96b – 229 Development midpoint of movement: 159

midpoint of development: 162
230 – 310 Recapitulation

311 – 318 Coda motto theme: A Phrygian major

The core event of the first movement is the hyperdissonant climax at the end of
the development section (rehearsal 21 through 24, resolving at rehearsal 25). The climax,
which occurs at a point of expected tonal and formal return, seriously disrupts the sense
of regularity and balance that the movement’s proportions engender. The climax event is
of such length, and is so strongly emphasized, that the hyperdissonant tail threatens to
wag the tonal dog, so to speak.
Analysis of the harmonic material at the climax provides important clues to
interpreting the symphony’s overall organization. The material is condensed in Figure
6.15a. The similarity between Figure 6.15a and Figures 6.8 and 6.9 (climax #1 in the
Rhapsody) is striking. In both, the global tonic of the work (A minor in both cases) is a
highly charged body, entangled in a complex structure involving G♯/A♭ and F♯/G♭. In the
Rhapsody, the structure occurs in the nega-inflected context of D♭ major. In the
symphony, embattled A minor emerges as tonic (at rehearsal 25); but the gamut of D♭/C♯
is nevertheless strongly implied, as shown in hypothetical Figure 6.15b. In Figure 6.15b,
as in the climax in Variation XVIII of the Rhapsody, a conventional tonic – dominant
resolution and a hexatonic resolution (A minor – C♯ minor/major) are combined.

256
 Figure 6.15. Symphony No. 3, i, harmonic content of the first movement climax chord

! !
'()*+*,&-./ $
'0)*1***,&-./2,(3./

$$"$# """
! """ """ !
$ """
"
" " !

"$#"
$ %
# $" %" "" # $"
$" $" $"
" "
! #
#6
7
!
& 5 &*2*4

8.,&-(-9*/:;.<=9&.-
>
?:@(9.-&A*/:;.<=9&.-*',(3./*9?&/8)

The chord shown in figure 6.15b is in fact an extended dominant sonority familiar
in Rachmaninoff’s works. Figures 6.16a and 6.16b show the appearance of the same type
of chord at climax events in the first and third movements of the Piano Concerto No. 4 in
G minor, Op. 40 (1926; rev. 1941). Figure 6.16a occurs at the hyperdissonant climax at
rehearsal 21 in the first movement of the concerto. The movement is in the key of G
minor. A long dominant pedal point (D♮) precedes the climax shown in Figure 6.16a,
which means that the climax represents a powerful harmonic and expressive
breakthrough to a new tonal level at a point where harmonic and formal processes have
led the listener to expect a return to G minor. At the climax, the movement’s tonic, G♮,
and pitch class A♭, which has figured prominently throughout the movement, are
entangled, resulting in an extended dominant chord that resolves to C major.22 In the third
movement, the same kind of extended dominant chord is used, climactically but not
hyperdissonantly, along with the same thematic material, between rehearsal 79 and 80,
this time resolving to G major. The climaxes of the first and third movements are

22
C major is the tonic of the concerto’s second movement, which has its own hyperdissonant climax at
rehearsal 36.

257
therefore closely associated, and it might be said that the event in the third movement
corrects the hyperdissonant “error” in the first.

Figure 6.16. Piano Concerto No. 4, Op. 40, i and iii, climax events

!
!"#$$%&'()$*+,-*-.)/$012-'3&((+.".)$45&*"6$")$-.3$+7$3-,-5+2*-.)

8#
! !!$$ !( %& %((( ((( ! '& & & %'& '& )( % & %(((
" ! % $ !& !& %& !!&& !( %& ( ( (
# !& & %& ' "
' "

% ((( ( %%(((( %)))(((( % (((

!!

* !! !%$$$ $$
$ %(($ %! ((( %(
$ $ $
$ $ $ $
+
%
,''')'3405.4'60+*7

!& % &&
* !! !%%&&& %&&&
&
! #$
!%
&
" !''()'!"'*+','-./012

!
!8#$$90&'3$*+,-*-.)/$45&*"6$")$4+3"

% ((( """" % ((( """"

9:
! & ! && ) && %! && %) (( %&
) !!&&&& - & !& &&&& &
!&&& %!&&&& !!)&&&& %!!&&&&& )&&&& & ( !& )&& )&& %&& )(( )%&& (
" , )!&&& !&& %& ) &
" "

("" ) (( % & % (( ""

#!! !! !! "
& && ! && ! & ! &( ) & % &
"

* ) &. -)!&&&!&&& ! %&& !)&& "

%&&&! )&&& &&& (( "
! & * %( !( & (( ""
& !& ! & &, - # ( &, - #

* ) )!!&&&& %&&
%&&&
&
!!#$
%
&
" !

258
 The evidence in Figure 6.15b and Figures 6.16a and 6.16b supports the claim that
the climactic sonority at the end of the development in the first movement of the
Symphony No. 3 (Figure 6.15a) may be heard as having strong implications in the key of
D♭ major/minor in addition to its more obvious (and realized) function as a gateway to
recapitulation in the key of A minor. As shown in Figure 6.17, the structure of the first
movement hinges on this implication: D♭ major is explicitly articulated at a climax event
at rehearsal 32 in the recapitulation, conceptually resolving the “other” side of the
hyperdissonance at rehearsal 22 and following. As also shown in Figure 6.17, the
exposition climaxes on F major. The climax events in the movement therefore articulate a
HEX(0,1) structure within the global A minor/major tonic context: A minor/major itself, F
major at the end of the exposition, D♭ major in the recapitulation, and the complex
sonority at the end of the development that simultaneously suggests both A minor/major
and D♭. Figure 6.17 shows how the tonal settings of the second theme material in both the
exposition and the recapitulation are “adjusted” in mid-stride to create the HEX(0,1)
climax events: from E major up to F major between rehearsal 7 and rehearsal 9 in the
exposition, and from C major/minor through A♭ major to D♭ major and, ultimately, to
global tonic A major between rehearsal 27 and rehearsal 33 in the recapitulation. In other
words, the key in which the second theme material is first heard—E major, the
dominant—turns out to be the “wrong” key for the HEX(0,1) structure; it is replaced by F
major as indicated on Figure 6.17. A more complex tonal structure characterizes the
second theme material in the recapitulation; but again manipulations are undertaken to
ensure a climax on D♭ major.
As shown in Figure 6.17, the hyperdissonant climax at the end of the development
(indicated by an asterisk on the figure) is entangled in both the large-scale functional
tonal structure and the HEX(0,1) structure of the climax events; and the global tonic, A
minor, is entangled in the climax chord.

259
 Figure 6.17. Symphony No. 3, i, analytic overview

HEX(0,1) climax events

extreme hyperdissonant distortion
A minor F major D  major A major

exposition climax central climax recapitulation climax

*
Dies irae
T es: circle of fifths
emphasized bass ton
Phrygian peremennost D Phrygian
“fantastic” PD D “fantastic” “fantastic”
PD D
                   
  
     
                  
(digression)
       
PD D ff ff T T ff
m.161 D
midpoint of DEV
T
midpoint of movement

260
T
feinted return to A minor & Theme I *
E major A  major
A minor T T A major

rehearsal # 1 2 3 4 5 7 8 9 11 13 14 15 17 18 19 20 21 25 26 28 29 32 33
6 10 12 16 22 27 30 34
23 31 35
24
transition

Intro Theme I Theme II Theme I Theme II Coda

Exposition Development Recapitulation

HD

tension arc at point of return

Hexatonic and Phrygian structures synthesized in the second and third movements

The structures just described, and the Phrygian motto theme, carry over into the
second and third movements of the symphony. The second movement has a hybrid form:
slow movement plus scherzo, as suggested in Figure 6.18.23 The figure provides an
overview of the movement, showing the F♯ minor / C♯ Phrygian major pitch centers of
the slow movement frame and the F minor tonic of the interior scherzo portion. As
suggested in the brief analysis of the movement in Chapter 5, tonic and dominant
functions are entangled in the C♯ Phrygian major setting of the motto theme in the outer
sections, and the fact that Rachmaninoff provided no key signature for the movement
suggests that the movement’s complex harmonic structure is to be interpreted in the
larger context of A minor/major.24 As suggested on Figure 6.18, the movement may be
interpreted as a synthesis of a HEX(0,1) structure from the first movement and Phrygian
organization, bringing it into close association with the events in the first movement.
The association between first and second movements is made explicit at the
climax in the second movement. As shown on Figure 6.18, the climax occurs shortly
between rehearsal 57, which is the midpoint of the scherzo portion of the movement
(Allegro vivace) and therefore conceptually of the symphony as a whole, and rehearsal
58. As the figure shows, the complex climax sonority from the end of the development in
the first movement reappears at the midpoint of the F minor scherzo. The three
participants in the HEX(0,1) structure—A minor/major, C♯ minor/major, and F
minor/major—are thereby bound across two movements by common association with the
climax chord.

23
As suggested in Figure 6.1, the prototype for this movement seems to have been the second movement of
the Piano Concerto No. 3, which is also a slow movement – scherzo hybrid, and which involves the same
two pitch centers—F♯ and D♭.
24
See again Figures 5.61 and 5.62.

261
 Figure 6.18. Symphony No. 3, ii, analytic overview

central climax chord

        
       



motto motto
ff

            
             
D
F  minor D D
D/T
T T

262
F minor
C  Phrygian major *
T/D
midpoint of Scherzo

36 39 40 41 43 44 45 47 51 53 56 57 58 59 60 66 67 68 69 70
rehearsal # 37 42 46 48 52 54 61
38 49 55 62
50 63
64
65

Slow movement Scherzo Slow movement

(Adagio ma non troppo) (Allegro vivace)
 As suggested in Chapter 5, the last forty-one measures of the second movement,
including a final clear statement of the motto theme on C♯, strongly favor C♯ Phrygian
major as a center over F♯ minor. As a result, the overall structure of the symphony takes
on the shape of a large-scale HEX(0,1) structure organized around climax events, and in
which the Phrygian motto theme acts as a kind of periodic structural articulation across
movements. The third movement, in A major, features C♯ prominently in several ways,
as discussed earlier in relation to Figures 4.12–4.14:

• The peremennost-inflected second theme area superimposes triads on C♯, A, and

E around rehearsal 77 in the exposition.

• The central fugue, in which the motto theme and the Dies irae are brought
together, emerges from an OCT(1,2) statement of the motto theme on C♯ (rehearsal
80) that explicitly recalls the C♯ statements of the motto theme in the second
movement. (The Dies irae pervades the rest of the movement.) The fugue
articulates a large-scale version of the motto theme on C♯, creating a large
OCT(1,2) cycle that achieves the home dominant.

• Most significantly, in the coda, the motto theme on C♯ is set in the key of A
major, with C♯ major triads explicitly incorporated into the gamut of the home
tonic (the Allegretto after rehearsal 110).

With these events, the structural tensions ultimately referable to the Phrygian motto in the
opening measures of the first movement and to the hyperdissonant climax at the end of
the development in the first movement are resolved. Recalling the end of the Rhapsody, a
flourish on the Dies irae in the last two measures of the symphony brings the B♭ back
into play (along with modal associate G♮ and auxiliary tone E♭)—a reminder, perhaps, of
the Phrygian starting point.

263
 Symphonic Dances, Op. 45 (1940)

The Symphonic Dances differ from the Rhapsody and the Symphony No. 3 in that
there is no single global tonic.25 The three movements are in three different keys: C
minor, G minor, and D major. But they are unified around a common group of chromatic
chords, which is shown in Figure 6.19.26 As briefly discussed in Chapter 4, these four
chords are first heard in measures 1–8 of the first movement.27 They appear in prominent
locations elsewhere in the first movement and in the other two movements, and I
therefore take them to be motivic material. Their distribution across the three movements
is of great analytic interest. I suggest that in Op. 45, unity is provided not by a large-scale
composing-out of a tonic nor even by shared thematic material, but by inter-movement
manipulation of a highly chromatic, distinctly non-tonical motivic chord group. Because
the motivic chords are stated plainly at the start of the opus, in a C minor context that
scarcely accommodates them tonally, one might say that a certain amount of
hyperdissonance is loaded into the work from the very start. The following analysis traces
the roles played by the motivic chords (singly or collectively) in the Symphonic Dances,
especially at climax events. I describe a gradual unfolding through the first two
movements leading to an acme in the third movement, where octatonic, hexatonic, and
Phrygian structures—all suggested by the motivic chord group—are brought together.
The motivic chord group resists easy description in functional tonal terms,
especially in the keys of C minor, G minor, and D major. Rachmaninoff took care to
present them at the beginning of the first movement in a way that obscures clear voice-
leading. (See again example Figure 4.20, where the triads are arpeggiated in different
instruments and in different registers.) In other words, Rachmaninoff establishes them as
chords, not as results of linear activity. The chords provide raw chromatic material that

25
In the present context, there is no compelling reason to make a distinction between the orchestral and
two-piano versions of the Symphonic Dances, which Rachmaninoff worked on simultaneously.
26
Another commonality between the three movements are the clear quotations from or references to
Rachmaninoff’s own earlier compositions. Several of these are indicated on the figures in this section of the
chapter, and are discussed at appropriate points in the analyis. At the end of the first dance, Rachmaninoff
appears to quote the primary theme from the first movement of the Symphony No. 1, Op. 13. In the third
dance, he refers to a passage (beginning three measures after rehearsal 10) in Isle of the Dead, Op. 29, and
he uses material from the ninth number (Blagosloven esi, Gospodi) of the All-Night Vigil, Op. 37.
27
See again Figure 4.20.

264
Rachmaninoff works into a variety of contexts. In the group, root relations by tritone,
major third, and minor third suggest the possibility of various “fantastic” structures, as
shown on Figure 6.19. The Phrygian structures that figure prominently in the following
analysis may similarly be understood as suggested by the motivic chord group. The chord
group therefore presents in a concentrated form several of the harmonic structures
preferred by Rachmaninoff in the late Russian and exile periods. In the analytic figures
that follow, the chords in the group are identified by circled Arabic numerals from 1 to 4,
corresponding to the order in which they are shown in Figure 6.19.

Figure 6.19. Motivic chord group in Symphonic Dances, Op. 45

! ###""" %$$"""
!

$##""" $$$ """

+,-./012

! " # $

%&' ()*

()*

Overview of movements i and ii

Each of the three movements in Op. 45 is in some kind or large ternary form.
Figure 6.20 is an analytic overview of the first movement. The introduction to the first
movement was analyzed in Chapter 4, as was section A1.28 Recall that the motivic chord
group is first heard in a “fantastic” context in the introduction, but that the thematic
28
See again figures 4.20, 4.44, and 4.45.

265
exposition beginning at rehearsal 2 is modal. As section A1 intensifies, equal-interval
structures come to the fore, confirming the general rhetorical associations outlined in
Chapter 3—but also calling attention to the tonally unsettled nature of the motivic chord
group as it is stated at the beginning of the movement.
As shown in Figure 6.20, the first movement incorporates motivic chord 1 (F♯ or
G♭ major) at an internal climax event, inside a large Phrygian structure (sections A1 and
A2 in C♮, section B1 and B2 a half-step higher in C♯/D♭). As in the Rhapsody and the
Symphony No. 3, C♯/D♭ emerges as the centerpiece in the overall design. The trajectory
of the B music mirrors the larger trajectory of the A music: motion from the minor mode
to the major mode. The key of C major at the end of the movement (starting four
measures before rehearsal 27) is the setting for what appears to be a loose quotation of
the primary theme from the composer’s Symphony No. 1, Op. 13.29
Figure 6.21 shows how motivic chords 2, 3, and 4 (on D, A♭, and A) are
incorporated into the key of G minor in the introductory measures of the second
movement. As suggested in Figure 6.17, the tritone root relation between chords 2 and 3
suggests an octatonicism that is realized at the beginning of the second movement. The
“slide” between A minor and A♭ major after rehearsal 31 recalls the similar slide between
the same two triads before rehearsal 2 in the first dance; but the tonal context is of course
different. An overview of section A1 of the large ternary in the second movement is
given in Figure 6.22. Section A1 itself describes a smaller ternary structure. As shown on
Figure 6.22, the section modulates from G minor to A♭ minor, which may be understood
as a larger articulation of motivic chord 3. Note also that motivic chord 1, on root F♯, is
suggested briefly at a local highpoint at rehearsal 36.

29
But note that the symphony theme is itself related to the incipit of the Dies irae. Martyn has suggested
that the melody which appears at the end of the first dance may be a reference more to the chant than to the
symphony theme (Martyn, Rachmaninoff, 350).

266
 Figure 6.20. Symphonic Dances, i, analytic overview

Intro A1 B1 B2 A2 Coda

1 2 10 11 17 18 20 21 22 27
19
chord group
chord quotation from
“fantastic” modal chord
1 Symphony No.1
“fantastic” 1 “fantastic”
modal (Phrygian)


              

 
 
 
peremennost

267
PD D
SD
II V
C# minor Db major T
(Aeolian) T

T
C minor C minor C major

internal climax
 Figure 6.21. Symphonic Dances, ii, analysis of the introduction

Introduction section A1

30 31 32

       
                   
           
        
violin solo

268
                     
                          

repeated, slightly altered

          
       

OCT(2,3)

II - V G minor

Tonic
chords
3 2 4 3
 Figure 6.22. Symphonic Dances, ii, overview of section A1

*. +

!"#$%&'(#)%" * + *,

:"#8(8&8"# (%"789'8"# (%"789'8"#

A<BC/D-E ;< =*< =*<

! !!??F??G >?@)"%$ !
*???@)"%$
*??! H?" *?" *??! I?#

-0 -. -/ -- -1 -2 -3 -4 -5 -6

Climax and culmination in movement iii

The events in the first and second movements lay groundwork for the D major
third movement, which represents a high-water mark in Rachmaninoff’s structural
thinking. Two passages from the movement were analyzed earlier in the dissertation.
Phrygian organization and the use of the Dies irae in the opening measures of the
movement were shown in Chapter 5.30 The internal climax of the central B section was
briefly presented in Chapter 4 as an example of hyperdissonant distortion involving an
octatonic structure at a point where resolution to the (local) tonic is expected.31 These
observations provide a framework for more comprehensive analysis.
Figure 6.23 provides an overview of section A1 in the third movement. The
introduction establishes Phrygian organization as structurally significant (involving pitch
class E♭ in a D major context, as discussed in Chapter 5), and it establishes the Dies irae
as thematic material. The Dies irae is taken up as theme I at measure 30, as shown in
Figure 6.24. The Dies irae appears in two main forms in the movement: the form stated at

30
See again Figures 5.60 and 5.61. It was suggested that the movement resembles the Etude-Tableaux in D
major, Op. 39, No. 9.
31
See again Figure 4.16.

269
measure 30 involves short rhythmic values (shown in Figure 6.24, and marked “Dies
irae: short” in Figure 6.23), while a form introduced later in the movement involves
longer rhythmic values (shown in Figure 6.28).

Figure 6.23. Symphonic Dances, iii, overview of section A1

23

!"#$% &'()(*! &'()(*!!

)*+,&*-.+/&,01-2 34.51,416+7&+,*8&91,:1;*

! "# $" !!%&'&!("

+%,-.
:+-+<+771,2 /.0)-1
4!5
+"#)*&' ("#)*&' !"#$%&' !"#)*&'
,-'./$)% ,-'./$)%"0"123)3&%$2

Figure 6.24. Symphonic Dances, iii, theme I: Dies irae in short rhythmic values

& & & & & & &

$$ ! % % % % ' % %& % (%& %&'%& % %& %& %& '%& (%& %& & %& & %&
!"#$"%&
!"

# " '% % )
!!

As shown in Figure 6.23, the goal of the first portion of section A1 is G major,
which reflects the smaller motion from D to G in measure 6 of the movement—a
resolution of the dominant side of the Phrygian tonic. Peremennost-type fusion of G
major and E minor leads to theme II, which is an extensive reworking of “Blagosloven
esi, Gospodi,” No. 9 from the All-Night Vigil, Op. 37. The beginning of this material is
shown in Figure 6.25. The E major climax of section A1 follows in measures 114–123,
and involves a hybridization of Phrygian and OCT(1,2) idioms, shown in Figure 6.26.

270
Pitch class C♮ (measure 118 and following) does not belong to OCT(1,2), but may be
understood as associated with the clearly articulated Phrygian upper and lower neighbor
figure cell. Pitch class D♯ in the bass is more problematic from a harmonic analysis
standpoint, being strictly associated with neither OCT(1,2) nor Phrygian E; but it does not
disrupts the overall sense of Phrygian and octatonic structures at the section A1 climax.

Figure 6.25. Symphonic Dances, iii, theme II, beginning

( *!
!"

# $$ !" % % % %
!"#$%&
(
% % % %
& & ' & & %& & & ' ' % %& %& %& %& %' % %& %& )%& )%
% %
!

Figure 6.26. Symphonic Dances, iii, octatonic–Phrygian hybrid at section A1 climax

!"#$%&'()*+,,

! "" ## ##
! "# #$$#
$# # $# # # #
!!" !!#

## # ##
# # # "## $ ## ##

% "" #
" &# # &# # # "# # &# #

-./01234

An overview of section B in the third movement is given in Figure 6.27. The

figure shows the emergence of D♭ major as an octatonic associate of E major in what
amounts to an extension of the OCT(1,2) structure from the climax at the end of section
A1. At measure 133, the Dies irae is stated in longer rhythmic values (marked “Dies
irae: long” in Figure 6.27), as shown more clearly in Figure 6.28. Both the form of the
Dies irae and the harmonic material in this passage refer to Isle of the Dead: compare

271
measure 133 and following in the dance to the passage beginning three measures after
rehearsal 10 in the earlier work.

Figure 6.27. Symphonic Dances, iii, overview of section B

Intro Lyrical Core: Db major

Dies irae: long
(Isle of the Dead)

124 133 152 162 200 208 - 214 215 - 234

Gb
SD
Local post-climactic
Phrygian
OCT(1,2) Climax pedal point
+ octatonic
idioms
OCT(1,2)
E major Db major Db major
chord

hyperdissonant
distortion

Figure 6.28. Symphonic Dances, iii, Dies irae in long rhythmic values

$$ ! !"##$% ( ) ) *!
!""

# " &% ! ' ! &% ! $' ! % ! &' ! $% ! % %! ' ! %! & ' ! $ % ! '! &% ! %
(
!"

Figure 6.27 provides a context for the local climax event in measures 208 – 214.
As shown more clearly in Figure 6.29, an accumulation of Phrygian and octatonic idioms
in the D♭ major music precedes a strong move to the subdominant—G♭ major, or motivic
chord 1—at measure 200. The incorporation of motivic chord 1 into D♭ major recalls
central section of the first movement.32 Resolution of the subdominant to local tonic D♭
major is powerfully distorted by the entangled diminished seventh chords of the OCT(1,2)
32
See again Figure 6.20.

272
structure at measure 208. A post-climactic pedal point follows, above which echoes of
octatonic and Phrygian idioms are heard.

Figure 6.29. Symphonic Dances, iii, climax in section B

$%&'()*+
$%&'()*+
,-.*.,+)- !!
,-.*.,+)-

! "" " # $# # #
!"# $### $ # $# # # # $" ## " # $ ## $" ## " # $ ## $#%
! " " ## $$ ## "## ## $ ## "## $ ## ## $ ## " ## $ ## "# $# "#
#
$" ## $$$ ### $" ### $ %### $""#### "#
"" ###
$#

# $ # $# $ # " # " # #
6789:43>3?322=
&" %# #
" " """ ##
# $# "# $# # $# "# $#
$$## " ## # $# "# $# # $# "# $# # $# # $ # "# #
! "# 6789:23;3732<=
!

/0!12345

%'@A&B)CC,+*+.9B)C.,&.),+8
&AC,DE.),+9.,9D,-*D9.,+)-9B)C&[email protected]

Section A2 returns to the tempo and material of section A1; but it is organized
very differently, as shown in Figure 6.30. The reprise of theme I in D major is delayed
until measure 334, making measure 235 and following more developmental than
recapitulatory. In section A2, the short and long forms of the Dies irae are brought
together, synthesizing material from section A1 and section B. Two strongly marked
statements of the long form in section A2 frame the main climax event of the movement:
a statement on A♭ major, preceding a long dominant pedal beginning in measure 287, and
a statement on D major at measure 328, as shown in Figure 6.30.
Figure 6.30 shows that the main climax of the movement may be interpreted as a
culmination involving the entire motivic chord group. The statement of the Dies irae on
A♭ major (motivic chord 3) is the goal of the first portion of section A1. The long
dominant pedal on A♮ (motivic chord 4) resolves to D major (motivic chord 2, and the
tonic of the movement) at measure 318. But the resolution is a passing event, not a
structural one, as the process of intensification begun at the start of section A2 continues,

273
pushing through D major to E major at measure 322 (thus recalling the climax at the end
of section A1) and then, climactically, to F♯ major (motivic chord 1—the first marked
event heard in the first movement of the opus, and associated with several earlier
climaxes) at measure 326.

Figure 6.30. Symphonic Dances, iii, overview of section A2

;<
!"#!$%"&
E3$1 '(")"*+ '(")"*++
C/$)3D
,-"."/0#)"123/4 :531)5;.3<)1=0;> :531)5;.3<)1=0;> ,-./01-0234)3156)7018095
? ?
:531)5;.3<)-04/ :531)5;.3<)-04/

!"# !$$ !%& "+%)*)""" ""' "'( "&#)*)"(!

*+',-$(%"$5$&#6
"""""""7"89: B053/
*F*50-3
""""""""""""7"20; :531)5;.3<)-04/ :531)5;.3<)-04/ C/$)3D
"
0"""#()&'
*+',-$(% !
."""#()&'
!"#()&' !"#()&'
. !"#$%&'
H =>?,@A<4 *+',-$(% *+',-$(%"1"234(4&%$3
I
"
/"""#()&' @
!"#()&' <

)02$.$5*5(0!-*6!07#

5/$)352$5*57/)$132$01*08*0#7%&*!"%0/72$01*08*6/093/*(:#"!-$%%01315"

Figure 6.31. Symphonic Dances, iii, octatonic–Phrygian hybrid at section A2 climax

!"#$%&'()*+,,

!
!"#
$ "$ $
#$$$ "##$$$$ %#$$$
$ $
! " #$$ %$$

#%$$
&" $
" $ "$ $

-./01234

274
 The hexatonic resolution of F♯ major to D major in measure 328 coincides with a
ff statement of the long-form Dies irae, prefacing the radically truncated reprise of theme
I at measure 334. The All-Night Vigil material returns as theme II at measure 349, leading
to a final climax event that, like the climax event at the end of section A1, may be
understood as involving a hybridization of Phrygian and octatonic structures (Figure
6.31).
These points are contextualized in Figure 6.32, an analytic overview of the entire
movement. The figure shows how a large-scale ascent across all three sections culminates
in a main climax event that integrates the motivic chord group fully into the gamut of D
major. The conventional dominant – tonic resolution at measure 318 is entirely
subordinate to the “fantastic” F♯ major–D major event in measures 326–328, which may
be understood the goal of the entire opus—the resolution of a global, opus-long
hyperdissonance between the symmetrical chromatic and Phrygian implications of the
motivic chord group and conventional tonal structures. The interaction of these
variegated components in such a powerfully climax-centric context makes the movement
a fitting culmination of the late Russian and exile periods as a whole.

Figure 6.32. Symphonic Dances, iii, analytic overview

-.'$/.0/123%41/.5678/#6$&%2'3%.2/%29.$9%2-/&.3%9%#/#:.4;/-4.65

417.$63%.2

! "# !!$ !%& &#' &/& &/0 #!/ #&% #&/ #$" #0'
#$%&'(
()*+,)-.*
$'4-1D7#'$1/'7#123

! "" # ## # # % % "# "# #

$$$$$ # %%% $# %# # #
$ # #
?@!A*B,C
()*+,)-.* " !!

! </&%2.4/=/&'>.4 ! "/&%2.4/=/&'>.4

)* + ),

275
 Concluding Remarks: Rachmaninoff in Context

Rachmaninoff’s openly stated disdain for the “modern music” of his day—“about
modern music I feel as about interviews before breakfast,” he told the San Francisco
press in 1937—has perhaps clouded scholarly judgment of his music’s aesthetic and
stylistic characteristics.1 It has been too easy to consider him an anachronism. A Webern
he was not; nor a Stravinsky; but neither was he a Tchaikovsky, or even a Glazunov. To
dismiss him as such is to accuse him of a kind of musical parochialism. A number of
photographs held by the Glinka Museum in Moscow show Rachmaninoff as a young man
in the early 1890s at the rural estate at Ivanovka (now a museum), rake in hand, his
relatives the Satins with him and (in at least one) Father Nikolay, their dour-looking
priest, at the rear.2 The photographs were perhaps intentionally stylized, but, still, to
twenty-first-century eyes Rachmaninoff appears very old-fashioned—a hay-bale, horse-
and-buggy figure. Forty years later he was a lover of motorboats and fast cars, a man who
enjoyed jazz, a cosmopolitan, globe-trotting figure.3 The cultural collision recalls the
musical collision described by Peter Burkholder: “All the composers of this generation
have aspects of both eras, combining nineteenth-century elements with twentieth-century
sensibilities.”4
By the same token, I have shown through analysis of many works that
Rachmaninoff’s mature music resists characterization exclusively in conventional tonal
terms. My analytic lens has suggested connections between Rachmaninoff’s mature
works and progressive European music of the early twentieth century, and suggested that
his Russian heritage is neither superficial nor dismissible. Particular modal and chromatic

1
Quoted in Bertensson and Leyda, Sergei Rachmaninoff, 327.
2
These photographs are in the public domain and may be viewed online in many locations. See for
example http://www.tstu.ru/en/tambov/kultur/composer/rahm/s1.htm (accessed April 23, 2009).
3
On Rachmaninoff and motorboats, see Bertensson and Leyda, Sergei Rachmaninoff, 319-20.
4
As discussed in Chapter 2 (Burkholder, Grout, and Palisca, A History of Western Music, 6th ed., 799).

276
structures have identities in the works studied. They are marked, and have clear rhetorical
associations: intensification, climax, and disruption in the case of “fantastic” equal-
interval structures; introduction, exposition, digression, and post-climax in the case of
modal structures (although Phrygian organization, as shown in several cases, has more
complex associations).
The challenge of developing an analytic strategy rigorous enough yet flexible
enough for such variegated harmonic environments has led me to reject existing
approaches that treat chromatic events as invariably adornments of functional structures.
Rachmaninoff was a Postromantic composer, and expressive trajectories in his music
involve processes of deformation, exaggeration, and distortion that result in part from
frictions between and integrations of differentiated components in a complex, compound
harmonic environment. Generalized, such processes may be taken as representative of the
Postromantic repertory in general. Analyses of several works by other composers in
Chapter 2 suggest potential applications of the approach developed in that chapter to a
larger repertory.
Rachmaninoff’s approach to form is at once Procrustean and plastic. This
apparent paradox, too, may be taken as a Postromantic trait. In his music we hear a fusion
of clear, conventional plans (ternary form and sonata form are preferred) and sui generis,
hyperdissonance-oriented shapes. The former supply frameworks; the latter supply
energy, and inform the interpretation of climax in the works analyzed. To return to a
metaphor suggested early in the dissertation, Dionysius is bound by Apollo; but he
nevertheless impels the action. In the works studied, interpretation of hyperdissonance
and climax has shown that a kind of expressive form is imposed on conventional form
and tonal design in flexible ways.

* * *

The analyses in the dissertation represent a departure from existing Rachmaninoff

scholarship in methodology, in depth of analysis, and in the variety of musical genres
considered. Though Robert Cunningham’s dissertation equals the present work in
analytic detail, his approach emphasizes a kind of unity and integration very different

277
from mine, because he fails to incorporate what I deem to be essential Russian chromatic
and modal idioms into his analyses. As analyses throughout the present dissertation have
shown, important rhetorical information is packed into these idioms.
If Barrie Martyn is correct and Rachmaninoff does indeed stand “Janus-like
between the old Russia and the new, looking back to the flowering of Russian nineteenth-
century ‘classical’ music as also ahead to the first generation of Soviet Composers,” then
it may be possible to hear in the works of later composers some of the structures and
techniques identified in the dissertation.5 In Chapters 2 and 4, analysis of octatonic and
hexatonic exaggeration in passages from Prokofiev’s works validated Geoffrey Norris’s
claim that similarities between the two composers may be more extensive than earlier
generations of musicians and scholars realized.6 In Chapter 5, Rachmaninoff’s extensive
use of Phrygian organization was shown to be a continuation of a practice established by
his Russian predecessors, and it was briefly suggested that Shostakovich continued the
practice (in, however, a radically adapted form).
But perhaps more revealing are ways that Rachmaninoff’s hyperdissonance-
oriented approach to musical form may resemble approaches in later composers’ works.
Yuriy Kholopov comments on Shostakovich:

Shostakovich’s new solution as a twentieth-century composer consists of finding

new effective means of contrast, an even higher order of dissonance. In the
develop section he now starts to place contrasted sound-layers one on top of
another. The unity of the harmony in the vertical dimension is broken. The layers
of polyharmony dissonantly contradict one another, as if the voices somehow are
not listening to one another; in some places they even try to out-shout one another
to see who can make the most noise. In places it becomes impossible to sense any
tonality whatsoever. Supercharging the discordant mass of sound leads to a huge
‘proclamation’ at the beginning of the recapitulation, where uncoordinated
shouting lines suddenly merge into a mighty unison. This type of solution imparts
new life to sonata form and other symphonised forms…7

5
Martyn, Rachmaninoff, 3.
6
See again discussion of Norris’s view in Chapter 1.
7
Yuriy Kholopov, “Form in Shostakovich’s Instrumental Works,” in Shostakovich Studies, ed. David
Fanning (Cambridge: Cambridge University Press, 1995),

278
 Kholopov’s comments recall with surprising clarity observations made in this
dissertation, suggesting that the characteristics he describes are not new in
Shostakovich’s music, only new in the extent to which they are featured. A “higher order
of dissonance” results from harmonic stratification. Conflict and contradiction generate
new kinds of expressive trajectories and rejuvenated formal processes. The “shouting
lines” finally come to some agreement, and the higher order of dissonance—the
hyperdissonance—is solved at a moment of climactic culmination.
It is true that harmonic materials in Rachmaninoff’s music are not as explicitly
stratified as they sometimes are in Shostakovich’s (or Richard Strauss’s)8. But a passage
like the one in Figure 5.58 is not far off: it features an unyielding layer of tonic triads on
the very bottom, functional resolutions to those tonics in the middle, and, on top, an
increasingly tense chromatic harmonization of a Phrygian ascent.
It is interesting to consider how hyperdissonant exaggeration, distortion, and—
perhaps more significantly—neutralization (as demonstrated in Skryabin’s Prelude, Op.
74, No. 3) may be manifest in radical but still recognizable ways in modernist works
from the first half of the twentieth century.9 At present, this is more speculation than
theory. The ground here is not at all firm underneath the analyst’s feet; but consider, for
example, the opening of Arnold Schoenberg’s twelve-tone Violin Concerto, Op. 36
(1936), shown in Figure C.1.10

Figure C.1. Arnold Schoenberg, Violin Concerto, Op. 36, i, reduction of mm. 1–4

!"#$%&'('")*+,")

$ # ! & # & ( $#
" ! $# ! &# &( $)## ! $#
' ' #
$## !! # $$##
!"#$"%
#
! )#
! $# , $
% )# % , &# #*
* +
&'("%)&

8
See again the analyses of Elektra and the Alpine Symphony in Chapter 2.
9
On the Skryabin prelude, see again Figures 4.21–4.23.
10
I am grateful to Andrew Mead for calling my attention to this passage.

279
 Taken on its own, the solo violin part seems to suggest a pair of tonal gestures in
B♭ minor (shown by arrows on the figure). A conventional goal-oriented rhythmic figure
leads to the downbeat as “leading-tone” resolves to “tonic” in measure 1; the second
gesture behaves similarly. The accompaniment in the strings of course completely denies
such an interpretation; but are the implications of the melody entirely lost, or just
embedded so deeply that it takes a kind of tonal archaeology to reveal them? I am
tempted to suggest in this very limited context that hyperdissonance—melodic
implications distorted or neutralized by the “chromatic” (really twelve-tone) context—is
so fully incorporated into the language of the piece that it is insoluble. The tension is
frozen in place, so to speak, and new harmonic processes are required to give shape to the
music.
Of course, nothing of this sort occurs anywhere in Rachmaninoff’s works.
Hyperdissonance is prepared; or, in the rare cases where a work begins hyperdissonantly
(e.g. the Etude-Tableaux in D major, Op. 39, No. 9 and the second movement of The
Bells, both analyzed elsewhere in the dissertation), it is at least resolved. Even the
dangling quasi-tonic at the end of “A-u!” comes nowhere near atonality.11
Rachmaninoff’s is a Postromantic ethos. But seeds planted in the Postromantic grew into
modernist plants. It is a quirk of musicology that modernism is perhaps better understood
than late Romanticism or Postromanticism. Continued work along lines suggested in the
dissertation, undertaken without preconceptions or prejudices, might fill in some of the
gaps.

* * *

Much remains to be done in the study of Rachmaninoff’s works. This dissertation

has of necessity been limited in scope and subject. Rhythm, texture, and orchestration
have not been considered in any detail, and have in fact never been taken up with any
rigor in the Rachmaninoff literature.
Also needed is a comprehensive comparison of Rachmaninoff and his schoolmate
Skryabin. Steps were taken in Chapter 4 of the dissertation, but many more will be

11
See again Figure 2.27.

280
necessary before century-old assumptions are replaced by solid conclusions. Skryabin
was perhaps not as utterly radical as James Baker has suggested; and Rachmaninoff was
surely not as utterly conservative as the literature has generally suggested.12 When
Rachmaninoff died in 1943, Skryabin had been dead nearly thirty years. Had
Rachmaninoff also died in 1915 (just after the Op. 34 songs, The Bells, the Sonata No. 2,
and the first set of Etudes-Tableaux, and with the Op. 38 songs and second set of Etudes-
Tableaux on the horizon), he would likely be remembered quite differently. His style did
not change radically after 1917; but this should be taken as refinement, not regression.13
Rachmaninoff’s adjacency to the German Postromantic tradition was suggested
by comparisons to Strauss and Mahler in Chapter 2. Further investigation will likely
reveal more parallels, and confirm Rachmaninoff’s position as a central figure in
Postromantic music. It would be satisfying to see his perennial popularity in the concert
hall reflected in a more widespread scholarly appreciation of his place in the repertory.
As Rachmaninoff put it, “taken individually the people in an audience may be poor critics
of music, but as a complete body, the audience never errs.”14

12
Baker, The Music of Alexander Scriabin.
13
For discussion of similar “refinement” in the apparently conservative late works of Richard Strauss, see
Kaplan, “The Musical Language of Elektra,” 176.
14
Bertensson and Leyda, Sergei Rachmaninoff, 362.

281
Bibliography

282
Abraham, Gerald. “The Elements of Russian Music.” Music and Letters 9 (1928): 51-58.

________. A Hundred Years of Music. 4th ed. London: Duckworth, 1974.

________. “Rachmaninow.” In Die Music in Geschichte und Gegenwart, edited by

Friedrich Blume, 10:1839-44. Kassel: Barenreiter, 1949.

Agawu, V. Kofi. “Structural 'Highpoints' in Schumann's 'Dichterliebe'.” Music Analysis 3

(1984): 159-80.

________. “Tonal Strategy in the First Movement of Mahler's Tenth Symphony.”

19th Century Music 9 (1986): 222-33.

Bailey, Robert. “An Analytical Study of the Sketches and Drafts.” In Wagner: Prelude
and Transfiguration from “Tristan and Isolde,” edited by Robert Bailey, 113–46.
New York: W. W. Norton, 1985.

________. “The Structure of the Ring and Its Evolution.” 19th Century Music 1 (1977–
78): 48–61.

Baker, James M. The Music of Alexander Scriabin. New Haven: Yale University Press,
1986.

Barthes, Roland. S/Z. Translated by Richard Miller. New York: Hill and Wang, 1974.

Baur, Steven. “Ravel's ‘Russian’ Period: Octatonicism in His Early Works, 1893–1908.”
Journal of the American Musicological Society (1999): 531-92.

Benjamin, William. “Interlocking Diatonic Collections as a Source of Chromaticism in

Late Nineteenth-Century Music.” In Theory Only 1 (1976): 31-51.

Berger, Arthur. “Problems of Pitch Organization in Stravinsky.” Perspectives of New

Music 2 (1963): 11-42.

Berry, Wallace. Structural Functions in Music. New York: Dover, 1987.

Bertensson, Sergei and Jay Leyda. Sergei Rachmaninoff: A Lifetime in Music. New York:
New York University Press, 1956.

Boyd, Malcolm. “‘Dies Irae’: Some Recent Manifestations.” Music & Letters 49 (1968):
347-56.

Brady, Patricia. “Rachmaninoff’s Etudes-Tableaux.“ D.M. diss., Indiana University,

1986.

283
Bribitzer-Stull, Matthew. “The A-Flat-C-E Complex: The Origin and Function of
Chromatic Major Third Collections in Nineteenth-Century Music.” Music Theory
Spectrum 28 (2006): 167-90.

Brower, Candace. “A Cognitive Theory of Musical Meaning.” Journal of Music Theory

44 (2000): 323-79.

Burkholder, J. Peter, Donald J. Grout and Claude Palisca. A History of Western Music. 7th
ed. New York: W. W. Norton, 2006.

Burns, Lori. Bach’s Modal Chorales. Stuyvesant, NY: Pendragon Press, 1995.

Callender, Clifton. “Voice-Leading Parsimony in the Music of Alexander Skryabin.”

Journal of Music Theory 42 (1998): 219-33.

Calvocoressi, Michel Dimitri. Mussorgsky. London: Rockliff, 1956.

Campbell, Stuart. Russians on Russian Music, 1880–1917. Cambridge: Cambridge

University Press, 2003.

Cannata, David Butler. Rachmaninoff and the Symphony. Innsbruck: Studien-Verlag,

1999.

________. “Rachmaninoff's Changing View of Symphonic Structure.” Ph.D. diss., New

York University, 1992.

________. “Rachmaninoff's Concept of Genre.” Studies in Music from the University of

Western Ontario 15 (1995): 59-73.

Carpenter, Ellon D. “Russian Theorists on Modality in Shostakovich's Music.” In

Shostakovich Studies, edited by David Fanning, 76-112. Cambridge: Cambridge
University Press, 1995.

Challis, Natalia. The Singer's Rachmaninoff. New York: Pelion Press, 1989.

Chew, Geoffrey. “Ernst Kurth, Music as Psychic Motion and Tristan und Isolde: Towards
a Model for Analysing Musical Instability.” Music Analysis 10 (1991): 171-93.

________. “The Spice of Music: Towards a Theory of the Leading Note.” Music Analysis
2 (1983): 35-53.

Cinnamon, Howard. “Third-Related Harmonies as Elements of Contrapuntal

Prolongation in Some Works By Franz Liszt.” In Theory Only 12 (1992): 1-30.

Cohn, Richard A. “Maximally Smooth Cycles, Hexatonic Systems, and the Analysis of
Late-Romantic Triadic Progressions.” Music Analysis 15 (1990): 9-40.

284
Cohn, Richard A. and Douglas Dempster. “Hierarchical Unity, Plural Unities: Toward a
Reconciliation.” In Disciplining Music, edited by Katherine Bergeron and Philip
V. Bohlman, 156-81. Chicago: The University of Chicago Press, 1992.

Collins, Dana Livingston. “Form, Harmony, and Tonality in S. Rakhmaninov’s Three

Symphonies.” Ph.D. diss., The University of Arizona, 1988.

Cone, Edward T. “Sound and Syntax: An Introduction to Schoenberg's Harmony.” In

Music: A View From Delft, edited by Robert P. Morgan, 249-66. Chicago:
University of Chicago Press, 1986.

Culshaw, John. Rachmaninov. New York: Oxford University Press, 1950.

Cunningham, Robert. E. Jr. “Harmonic Prolongation in Selected Works of Rachmaninoff,

1910–1931.” Ph.D. diss., The Florida State University, 1999.

________. Sergei Rachmaninoff: A Bio-Bibliography. Westport, CT: Greenwood Press,

2001.

Dahlhaus. Between Romanticism and Modernism. Translated by Mary Whittall. Berkeley:

University of California Press, 1980.

Darcy, Warren. “Bruckner’s Sonata Deformations.” In Bruckner Studies, edited by

Timothy L. Jackson and Paul Hawkshaw, 256-77. Cambridge: Cambridge
University Press, 1997.

Dubiel, Joseph. “Contradictory Criteria in a Work of Brahms.” In Brahms Studies, edited

by David Brodbeck, 1: 81-110. Lincoln: University of Nebraska Press, 1995.

Flanagan, William. “Sergei Rachmaninoff: A Twentieth-Century Composer.” Tempo

(1952): 4-8.

Forte, Allen. “Musorgsky as Modernist: The Phantasmic Episode in 'Boris Godunov'.”

Music Analysis 9 (1990): 3-45.

________. “Schoenberg’s Creative Evolution: The Path to Atonality.” The Musical

Quarterly 64 (1978): 133-76.

Frolova-Walker, Marina. “On ‘Ruslan’ and Russianness.” Cambridge Opera Journal

(1997): 21-45.

Gregory, Robin. “Dies Irae.” Music & Letters 34 (1953): 133-39.

Grout, Donald J. and Claude Palisca. A History of Western Music. 4th ed. New York: W.
W. Norton, 1988.

285
________. A History of Western Music. 6th ed.. New York: W. W. Norton, 2001.

Guck, Marion. "The Functional Relations of Chords: A Theory of Musical Intuitions." In

Theory Only 4 (1978): 29-41.

Harrison, Daniel. Harmonic Function in Chromatic Music: A Renewed Dualist Theory

and an Account of Its Precedents. Chicago: The University of Chicago Press,
1994.

Harrison, Max. Rachmaninoff. London: Continuum, 2005.

Hatten, Robert. Musical Meaning in Beethoven: Markedness, Correlation, and

Interpretation. Bloomington: Indiana University Press, 1994.

Hepokoski, James. “Fiery-Pulsed Libertine or Domestic Hero? Strauss’s Don Juan

Reinvestigated.” In Richard Strauss: New Perspectives on the Composer and His
Work, edited by Bryan Gilliam, 135-76. Durham, NC: Duke University Press,
1992.

________. Sibelius: Symphony No. 5. Cambridge: Cambridge University Press, 1993.

Hoeckner, Berthold. “Schumann and Romantic Distance.” Journal of the American

Musicological Society 50 (1997): 55-132.

Horowitz, Joseph. Artists in Exile. New York: Harper Collins, 2008.

Kaminsky, Peter. “Ravel’s Late Music and the Problem of “Polytonality.” Music Theory
Spectrum 26 (2004): 237-64.

Kang, Heejung. “Rachmaninoff’s Rhapsody on a Theme by Paganini, Op. 43: Analysis

and Discourse.” D.M.A. diss., University of North Texas, 2004.

Kaplan, Richard A. “The Musical Language of Elektra: A Study in Chromatic

Harmony.” Ph.D. diss., University of Michigan, 1985.

________. “Temporal Fusion and Climax in the Symphonies of Mahler.” Journal of

Musicology (1996): 213-32.

Karl, Gregory. “Cyclic Structure in Two Works of Rakhmaninov.” Music Research

Forum 3 (1988): 5-22.

Kholopov, Yuriy. “Form in Shostakovich’s Instrumental Works.” In Shostakovich

Studies, edited by David Fanning, 57-75. Cambridge: Cambridge University
Press, 1995.

286
Kinderman, William and Harald Krebs, eds. The Second Practice of Nineteenth-Century
Tonality. Lincoln: University of Nebraska Press, 1996.

Kopp, David. Chromatic Transformations in Nineteenth-Century Music. Cambridge:

Cambridge University Press, 2002.

Kramer, Lawrence. “The Mirror of Tonality: Transitional Features of Nineteenth-Century

Harmony.” 19th Century Music 4 (1981): 191-208.

Laitz, Stephen G. The Complete Musician. 2nd ed. Oxford: Oxford University Press,
2008.

Lasarenko, Kim Andrei. “A Style Change in Rachmaninoff's Piano Music as Seen in the
Second Piano Sonata in B-Flat Minor, Op. 36 (1913 and 1931 Versions).” D.M.A.
diss., The Ohio State University, 1988.

Leikin, Anatole. “From Paganism to Orthodoxy to Theosophy: Reflections of Other

Worlds in the Piano Music of Rachmaninov and Scriabin.” In Voicing the
Ineffable: Musical Representations of Religious Experience, edited by Siglind
Bruhn, 25-44. Hillsdale, NY: Pendragon Press, 2002.

Lerdahl, Fred. “Calculating Tonal Tension.” Music Perception 13 (1996): 319-82.

Lewis, Christopher. “The Mind’s Chronology: Narrative Times and Harmonic Disruption
in Postromantic Music.” In The Second Practice of Nineteenth-Century Tonality,
edited by William Kinderman and Harald Krebs, 114–49. Lincoln: University of
Nebraska Press, 1996.

MacDonald, Hugh. “[G-Flat Major Key Signature].” 19th Century Music 11 (1988): 221-
37.

Martyn, Barrie. Rachmaninoff: Composer, Pianist, Conductor. Aldershot, England:

Scolar Press, 1990.

McCreless, Patrick. “An Evolutionary Perspective on Nineteenth-Century Semitonal

Relations.” In The Second Practice of Nineteenth-Century Tonality, edited by
William Kinderman and Harald Krebs, 87-113. Lincoln, University of Nebraska
Press, 1996.

________. “Syntagmatics and Paradigmatics: Some Implications for the Analysis of

Chromaticism in Tonal Music.” Music Theory Spectrum 13 (1991): 174-78.

________. Wagner’s “Siegfried”: Its Drama, History, and Music. Studies in Musicology
no. 59. Ann Arbor, MI: UMI Research Press, 1982.

287
Mead, Andrew W. “Listening to Reger.” The Musical Quarterly 87 (2004): 681-707.

Meyer, Leonard B. Emotion and Meaning in Music. Chicago: The University of Chicago
Press, 1956.

Mitchell, William. “The Study of Chromaticism.” Journal of Music Theory (1962): 2-31.

Morosan, Vladimir, ed. Sergei Rachmaninoff: The Complete Sacred Choral Works. The
Monuments of Russian Sacred Music, vol. 9. Madison: Musica Russica, 1994.

Morrison, Simon. “The Semiotics of Symmetry, Or Rimsky-Korsakov's Operatic History

Lesson.” Cambridge Opera Journal (2001): 261-93.

________. “Skryabin and the Impossible.” Journal of the American Musicological

Society (1998): 283-330.

Murrary, David. “Elektra.” In Grove Music Online, edited by Laura Macy.

http://www.grovemusic.com (accessed October 5, 2007).

Newmarch, Rosa. “Rakhmaninov, Sergey Vassilievich.” In Grove's Dictionary of Music

and Musicians, edited by Eric Blom, 7: 27-29. New York: St. Martin's Press,
1954.

Norris, Geoffrey. Rachmaninoff. The Master Musician Series. New York: Schirmer
Books, 1994.

________. “Rachmaninoff, Serge.” In Grove Music Online, edited by Laura Macy.

http://www.grovemusic.com (accessed March 16, 2009).

________. “Vladimir Ashkenazy on Sergei Rachmaninoff.” Andante Online (December,

2001). http://www.andante.com/article/article.cfm?id=15463 (accessed March 10,
2009).

Persichetti, Vincent. Twentieth-Century Harmony, 2nd ed. New York: W. W. Norton,

1961.

Proctor, Gregory. “Technical Bases of Nineteenth-Century Chromaticism.” Ph.D. diss.,

Princeton University, 1978.

Prussing, Stephen H. “Compositional Techniques in Rachmaninoff’s ‘Vespers, Opus

37’.” Ph.D. diss., The Catholic University of America, 1980.

Rachmaninoff, Sergei. Piano Concertos Nos. 3 and 4. London: Boosey & Hawkes, 2004.

________. Rhapsody on a Theme by Paganini, Op. 43. London: Boosey & Hawkes,
2003.

288
________. Sonata No. 2, Op. 36 (original and revised editions). London: Boosey &
Hawkes, 1993.

________. Symphonic Dances, Op. 45. London: Boosey & Hawkes, 2005.

________. Symphony No. 3, Op. 44. London: Boosey & Hawkes, 2003.

Ratner, Leonard. Classic Music: Expression, Form, and Style. New York: Schirmer,
1980.

________. Music: The Listener’s Art. 2nd ed. New York: McGraw-Hill, 1966.

Reese, Gustav. Music in the Middle Ages. New York: W. W. Norton, 1940.

Reise, Jay. “Late Skriabin: Some Principles Behind the Style.” 19th Century Music 6
(1983): 220-31.

Riesemann, Oskar von. Rachmaninoff's Recollections Told to Oskar von Riesemann. New
York: MacMillan, 1934.

Roig-Francoli, Miguel. Understanding Post-Tonal Music. New York: McGraw-Hill,

2008.

Rosen, Charles. Sonata Forms. Rev. ed. New York: W.W. Norton, 1988.

Rothfarb, Lee A. “Energetics.” In The Cambridge History of Western Music Theory,

edited by Thomas Christensen, 927-55. Cambridge: Cambridge University Press,
2002.

________. Ernst Kurth as Theorist and Analyst. Philadelphia: University of Pennsylvania

Press, 1988.

________. Ernst Kurth: Selected Writings. Cambridge: Cambridge University Press,

1991.

Salazar, Adolfo. Music in Our Time: Trends in Music since the Romantic Era. Translated
by Isabel Pope. Westport, CT: Greenwood Press, 1946.

Samson, Jim. Music in Transition: A Study of Tonal Expansion and Atonality, 1900–
1920. New York: W. W. Norton, 1977.

Satyendra, Ramon. “Chromatic Tonality and Semitonal Relationships in Liszt's Late

Style.” Ph.D. diss., The University of Chicago, 1992.

289
________. “Conceptualizing Expressive Chromaticism in Liszt's Music.” Music
Analysis 16 (1997): 219-52.

Schenker, Heinrich. Free Composition. Edited and translated by Ernst Oster. 2 vols. New
York: Schirmer Books, 1979. First published in 1935.

Schoenberg, Arnold. Concerto for Violin and Orchestra, Op. 36. New York: G. Schirmer,
1939.

________. Structural Functions of Harmony. Rev. ed. Edited by Leonard Stein. New
York: W. W. Norton, 1969. First published in 1954.

________. Theory of Harmony. Edited and translated by Roy E. Carter, based on the
third edition (1922). Berkeley and Los Angeles: University of California Press,
1978. First published in 1911.

Smith, Charles J. “The Functional Extravagance of Chromatic Chords.” Music Theory

Spectrum 8 (1986): 94-129.

________. “Is It Original, Or Is It Good? The Paradox of Rachmaninoff's Intra-

Tonal Chromatic Harmony.” Paper delivered at the Rhodes International
Rachmaninoff Conference, October 23, 2005.

Stein, Deborah. Hugo Wolf’s Lieder and Extensions of Tonality. Studies in Musicology
Series. Ann Arbor, MI: UMI Research Press, 1985.

Stell, Jason T. “Rachmaninov's Expressive Strategies in Selected Piano Preludes:

Highpoints, Dramatic Models, and Dynamic Curves.” M.A. thesis, The
Pennsylvania State University, 1999.

Straus, Joseph N. “The "Anxiety of Influence" in Twentieth-Century Music.” The

Journal of Musicology 9 (1991): 430-47.

________. An Introduction to Post-Tonal Theory. 3rd ed. Upper Saddle River, NJ:
Pearson/Prentice Hall, 2005.

________. Remaking the Past: Musical Modernism and the Influence of the Tonal
Tradition. Cambridge, MA: Harvard University Press, 1990.

________. “Stravinsky’s ‘Tonal Axis’.” Journal of Music Theory 26 (1982): 261–290.

Swan, Alfred J. Russian Music and Its Sources in Chant and Folk-Song. New York: W.
W. Norton, 1973.

________. “The Znamenny Chant of the Russian Church.” Parts 1, 2. The Musical
Quarterly 26 (1940): 232-43, 365-80.

290
Swan, Alfred J. and Katherine. “Rachmaninoff: Personal Reminiscences.” Parts 1, 2. The
Musical Quarterly 30 (1944): 1-19, 174-91.

Swinden, Kevin J. “When Functions Collide: Aspects of Plural Function in Chromatic

Music.” Music Theory Spectrum 27 (2005): 249-82.

Threlfall, Robert, and Geoffrey Norris. A Catalogue of the Compositions of Sergei

Rachmaninoff. London: Scholar Press, 1982.

Taruskin, Richard. Defining Russian Musically. Princeton: Princeton University Press,

1997.

________. “Russian Musical Orientalism: A Postscript.” Cambridge Opera Journal 6

(1994): 81-84

________. Stravinsky and the Russian Traditions: A Biography of the Works through
Mavra. Oxford: Oxford University Press, 1996.

van den Toorn, Pieter C. The Music of Igor Stravinsky. New Haven: Yale University
Press, 1983.

Vining, Paul and Rollo Piaf. “Byrd Plagiarized.” The Musical Times 118:1160 (1977):
300.

Wilder, Pamela Wright. “Sergei Rachmaninoff: Understanding the Composer through the
Etudes-Tableaux, Op. 33.” D.M.A. diss., University of Alabama, 1988.

Williams, Edward V. “The Blagovest Theme in Russian Music.” Kennan Institute for
Advanced Russian Studies Colloquium (December, 1985). Washington, D.C.:
Wilson Center Press, 1987.

Winters, Glenn R. “An Analysis of Sergei Rachmaninoff's Preludes, Opus 23 and Opus
32, and Etudes-Tableaux, Opus 33 and Opus 39.” D.M. diss., Northwestern
University, 1986.

Woodard, Susan Jeanne. “The Dies Irae as Used by Sergei Rachmaninoff: Some Sources,
Antecedents, and Applications.” D.M.A. diss., The Ohio State University, 1984.

Wilson, Paul. The Music of Béla Bartók. New Haven: Yale University Press, 1992.

Yasser, Joseph. “Progressive Tendencies in Rachmaninoff's Music.” Tempo 22 (1951-

52): 11-25.

________. “The Opening Theme of Rachmaninoff's Third Piano Concerto and Its
Liturgical Prototype.” The Musical Quarterly 55 (1969): 313-28.

291
Youmans, Charles. “The Twentieth-Century Symphonies of Richard Strauss.” The
Musical Quarterly 84 (2000): 238-58.

Zimmerman, Daniel J. “Families Without Clusters in the Early Works of Sergei

Prokofiev.” Ph.D. diss., The University of Chicago, 2002.

292