Elements of sonata theory: norms, types, and deformations in the late-eighteenth-century sonata 9780195146400, 9780199890231, 9780199773916


117 103 150MB

English Pages [691] Year 2006

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Frontmatter
Terms and Abbreviations (page xxv)
About the Authors (page xxix)
ONE Contexts (page 3)
TWO Sonata Form as a Whole: Foundational Considerations (page 14)
THREE The Medial Caesura and the Two-Part Exposition (page 23)
FOUR The Continuous Exposition (page 51)
FIVE The Primary Theme (P) (page 65)
SIX The Transition (TR) (page 93)
SEVEN The Secondary Theme (S) and Essential Expositional Closure (EEC): Initial Considerations (page 117)
EIGHT S-Complications: EEC Deferral and Apparent Double Medial Caesuras (TMB) (page 150)
NINE The Closing Zone (C) (page 180)
TEN The Development (Developmental Space) (page 195)
ELEVEN The Recapitulation (Recapitulatory Space; Recapitulatory Rotation) (page 231)
TWELVE Non-Normative Openings of the Recapitulatory Rotation: Alternatives and Deformations (page 255)
THIRTEEN Parageneric Spaces: Coda and Introduction (page 281)
FOURTEEN Sonata Form in Minor Keys (page 306)
FIFTEEN The Three- and Four-Movement Sonata Cycle (page 318)
SIXTEEN Sonata Types and the Type 1 Sonata (page 343)
SEVENTEEN The Type 2 Sonata (page 353)
EIGHTEEN Rondos and the Type 4 Sonata (page 388)
NINETEEN The Type 5 Sonata: Fundamentals (page 430)
TWENTY The Type 5 Sonata: Mozart's Concertos (R1: The Opening Ritornello) (page 469)
TWENTY-ONE The Type 5 Sonata: Mozart's Concertos (Solo and Larger Expositions: Solo 1 + Ritornello 2) (page 496)
TWENTY-TWO The Type 5 Sonata: Mozart's Concertos (Development and Recapitulation: From Solo 2 through Ritornello 4) (page 563)
Appendix 1. Some Grounding Principles of Sonata Theory (page 603)
Appendix 2. Terminology: "Rotation" and "Deformation" (page 611)
Bibliography (page 623)
Index of Names (page 633)
Index of Works (page 639)
Index of Concepts (page 649)
Recommend Papers

Elements of sonata theory: norms, types, and deformations in the late-eighteenth-century sonata
 9780195146400, 9780199890231, 9780199773916

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

ELEMENTS of SONATA THEORY

This page intentionally left blank

ELEMENTS of SONATA THEORY Norms, Types, and Deformations in the Late-Eighteenth-Century Sonata

James Hepokoski Warren Darcy

OXFORD

UNIVERSITY PRESS Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellence in research, scholarship, and education.

Oxford New York Auckland Cape Town Dares Salaam HongKong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in

Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Copyright © 2006 by Oxford University Press, Inc. Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Www.oup.com Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Hepokoski, James A. (James Arnold), 1946—

Elements of sonata theory : norms, types, and deformations in the late-eighteenth-century sonata / James Hepokoski, Warren Darcy. p. cm. Includes bibliographical references and index. ISBN-13 978-0-19-514640-0 ISBN 0-19-514640-9 1. Sonata form. 2. Instrumental music—18th century—Analysis, appreciation.

I. Darcy, Warren. II. Title MT62.H46 2006

784.18'3'09033—dce22 2005006674

246897531 Printed in the United States of America on acid-free paper

Preface

To book offers a fresh approach to one of the most familiar topics in the field of music: the study of sonata-form movements and the larger workings of multimovement sonatas, symphonies, and chamber music of the “early classical” and “clas-

sical” period. While remaining in dialogue with the several current approaches to this subject, it provides something different, and from time to time it challenges established views of the sonata. Both of the authors have been leading classes and seminars in this method over the past decade at Yale University, Oberlin College Conservatory, and the University of Minnesota. Large portions of Elements of Sonata Theory, both in earlier incarnations and in this much-expanded one, have been required reading in these courses. From one perspective the Elements is a research report, the product of our analyses of hundreds of individual movements by Haydn, Mozart, Beethoven, and many surrounding composers of the time (as well as later composers). In our work we have been looking for patterns within sonata-composition, for shared gestures, for ranges of options, for a sense of the typical. Our intention was to devise an explanation of how varying degrees of the normative can be altered, stretched, or occasionally overridden altogether to produce an individualized “deformation.” To be sure, the theoretical discussions of eighteenth- and early- nineteenth-century writers are relevant (Koch, Galeazzi, Reicha, and others), and these insights are integrated into the book. Our preference, though, was to let the composers themselves teach us how sonatas work. Our method of understanding sonatas (“Sonata Theory”) strikes a balance between inductively inferred norms and the unpredictability that one finds in these pieces.

vi Preface

Late-eighteenth-century sonatas are most productively heard within the context of a broad, flexible background-knowledge of what had come to be more or less standard compositional options at each point in the sonata. Any individual work within a genre (such as sonata form) interacts with the listener’s (or composer’s) expectations. Our book provides a detailed map of those expectations at that point in history. Not surprisingly, this turns out to be a complex matter. How can we know whether Haydn’s choice here or Mozart’s there was to be heard as normative, as strikingly unusual, or as something in between? And how can a composer’s early choices influence the range of continuation-options down the road? Understanding any compositionally selected gesture requires an awareness of the backdrop of typical choices against which it was written and within whose world of norms the piece was to be grasped in the first place. The Elements seeks to fill in many details of that backdrop. This perspective has the advantage of permitting one to pass beyond the confines of the acoustic surface alone (what one literally hears, what is actually notated) in order to notice, for instance, which normative things might be absent. It could be that such absences—generically expected events that the composer might keep from happening within an individual work—should be understood as essential constituents of the piece’s meaning. This book divides into two large parts. In the first of these, chapters 1-15, we lay out the basics of the essential system, work-

ing section by section, zone by zone, through the most often encountered type of sonata form (“Type 3,” with an exposition, development, and recapitulation), and considering also the differing implications within minor-mode sonata form and the multimovement sonata as a whole. In several of our analytical seminars earlier versions of chapters 1-15 alone served as the text. The discussions found in the second part, chapters 16-22, are more complex and extended, especially from chapter 17 onward. ‘These chapters provide elaborately detailed studies of the other sonata formats of the period (Types 1, 2, 4, and 5). The increased intensity of these chapters is no accident. Confronting these differing formats at all—the “sonata without development” (Type 1), the “binary” sonata (Type 2, without a full recapitulation), the sonata-rondo (Type 4), and concerto first-movement form (Type 5)—throws one directly into the midst of ongoing debates and passionately held, sharply diverging views. Given the existing state of the discussion, we were obliged to present these thorny issues with an enhanced rigor, constructing step-by-step solutions to these often misconstrued matters and providing evidence and justification for our decisions along the way. Nowhere 1s this situ-

ation more evident than in the case of the first movements of Mozart’s concertos. This Type 5 structure is the most difficult of the sonata types, and it is a topic concerning which even the most rudimentary features of terminology and sonata-form perception

Preface vil have been ardently contested over the past century. Covering this problem adequately required four extended chapters (19-—22)— virtually a separate monograph on the Mozartian Type 5 sonata, though one that is entirely dependent on one’s grasp of the book’s first eighteen chapters.

In addition to furnishing a new mode of analysis for the late-eighteenth-century instrumental repertory, the Elements also provides a foundation for considering works from the decades to come—late Beethoven, Schubert, Weber, Mendelssohn, Schumann, Liszt, Brahms, Bruckner, Strauss, Mahler, the “nationalist composers,” and so on. As we point out from time to time, most of these sonata norms remained in place as regulative ideas throughout the nineteenth century, even as the whole sonata-form genre, with its various options, was continuously updated, altered, and further personalized with unforeseen accretions, startling innovations, and more radical deformations. (The “three-key expositions” sometimes found in Schubert and Brahms, for example—though surfacing in some earlier composers as well—seem to have been encouraged by the eighteenthcentury expositional strategy of the “trimodular block” and its “apparent double medial caesuras.” Similarly, the “de-energizing transition” and occasional suppression of the medial caesura in, say, Schumann or Brahms, surely emerged from the precedents of the “blocked medial caesura” coupled with “expanded caesurafill” in Haydn and Mozart.) What follows is a blend of musicological and music-theoret-

ical thinking. What at first may seem to be a work of music theory turns out in the end to be a set of reflections on what sonata form is and how it can be understood to mean anything at all. In its drive to get to the bottom of things, Sonata Theory 1s informed by a not-always-tacit dialogue with current philosophy and literary criticism. While the book does not flaunt its intersections with certain strands of thought of the past decades—

genre theory, phenomenology-oriented hermeneutics, reader (listener)-generated artistic texts, the slippage and dispersion of meanings once supposed to be unitary, and so on—the importance to us of those modes of thinking should be evident to most readers. (The more generalized axioms grounding our conceptual system and modes of inquiry are laid out in appendix 1 at

the end of the book.) There are no tacit social agendas to our research of which we are aware, except that of seeking to understand what sonatas are and how they work. Still, Sonata Theory does have an interpretational, self-reflective, or philosophical tilt to it, an urge to explore a more fundamental questioning of this music’s methods and purposes. We are committed to understanding musical practice not only as a self-contained technical language but also as a metaphor for human action or communtication. We hope that our work will illuminate other perspectives and will open the investigation of sonata form and its diversified

viii Preface meanings to questions of serious concern to a new, younger generation of musicians and scholars. The musical examples in this book were created by Marcus Lofthouse, a recent graduate of the Oberlin Conservatory of Music, using Sibelius 3. Passages for solo piano are reproduced in full; string quartet excerpts are presented on two staves, but nothing has been omitted. Most orchestral passages are presented in twostave reductions that eliminate some octave doublings but retain the melody and bass lines, accompaniment, and any inner voices or counterpoint, all in their original registers. Space limitations necessitated reducing most concerto passages to two staves, one for the soloist (usually piano), the other for the orchestral accompaniment. The solo passages are presented as completely as possible, eliminating only a few low doublings as well as those measures where the piano functions as a thoroughbass instrument. The orchestral passages required a bit more in the way of compromise, but the one-staff reductions do show all the essentials of the textures. The examples were checked against the most authoritative editions available. Although fidelity to the score is balanced by practical considerations of legibility, our aim has been to make these examples as complete, as faithful to the original, and as helpful as possible. The figures and tables in the various chapters were reproduced by Zachariah Victor, using Adobe [lustrator 10. In referring to individual works in the text we normally use the full versions of the most widely known, easily recognizable titles and numberings (and even nicknames), even when those designations might be more popular—or customary—than scholarly. (Additionally, when we do not explicitly flag a key as “minor,” we mean that it is major: “in D” means “in D major.”) As all scholars of the period are aware, Mozart’s “Symphony No. 39 in E-flat, K. 543” 1s not at all his thirty-ninth composed symphony—nor did

the composer think of it in that way—although for a very long time it has been commonly referred to as that in standard discussions and shows no sign of even beginning to shake off this now “historically fixed” number. And merely to refer to the work, de haut en bas, as the Symphony in E-flat [Major] or only as K. 543, without any other identifying reference, could either oblige some readers to scurry off to Kéchel-number lists or discourage them from trying to remember which piece this actually is. These issues are particularly noticeable in references to Mozart’s piano concertos, which in the literature are often referred to only by Kéchel number, sometimes accompanied by the key (“K. 488” alone or “Piano Concerto in A, K. 488,” as opposed to our preferred—though not literally correct—‘Piano Concerto No. 23 in A, K. 488”). We recognize the historical inaccuracies embedded in “No. 39” and “No. 23,” but our intention, because we cite so many references to so many individual works, has been to simplify things for the reader. For similar reasons, within the text

Preface 1x

proper our references to Kéchel numbers are only to the familiar, “traditional” numbers, thereby avoiding the clutter and pedantic flavor of the double-descriptions that append the revised K. numbers as well, when such numbers exist. Thus instead of the scholarly precise “Piano Concerto in C, K. 415/387b” we prefer the more reader-friendly (albeit “incorrect” or not fully up-to-date with regard to the catalogue) “Piano Concerto No. 13 in CG, K. 415.” (No reader could possibly be confused by the absence of the much less familiar “updated number.”) Not all of the K. numbers

have these issues associated with them, but when they do, the dual number 1s provided in the index. Related issues and choices were made in citing the works of Haydn, Beethoven, Schubert, and others.

Any book of this scope is inevitably indebted to the many colleagues and students—too numerous to mention individually— with whom, along the way, we have shared information, proposed new ideas, developed concepts, and worked through analyses. We are grateful for all of these conversations and critiques, which have helped to shape our own thinking over the years. Apart from its use in academic classes and seminars, much of the first half of this book was “officially launched” at a workshop of the Mannes Institute for Advanced Studies in Music Theory (Institute on Musical Form) on June 24—27, 2004. This workshop was ably led by Daniel Harrison, and we thank him for doing so. We are also pleased to acknowledge the assistance and encouragement of Oxford University Press from the book’s initial inception back in the mid- and late-1990s onward. We are grateful to Maribeth Payne, then music editor at Oxford, and to the group of anonymous reviewers that read and commented on an early version of this text around seven years ago. And we are indebted to the team at Oxford, Kimberley Robinson, Eve Bachrach, Robert Milks, Norman Hirschy, and others, who have been crucial in suiding this book through the production process and into print. Still additional thanks are due to Thomas Hepokoski, who helped to sustain this project to the end in important ways. Finally, we thank our immediate families—and especially our wives, Barbara and Marsha—for having the willingness, love, and patience to persevere through the seemingly endless sessions of our research, writing, and revisions. There may be, finally, light at the end of the tunnel.

This page intentionally left blank

Contents

Terms and Abbreviations xxv About the Authors xxix ONE

Contexts 3 Differing Approaches to Sonata Form = 3

The War against the Textbooks 6 Sonata Theory: Introductory Remarks 9 TWO

Sonata Form as a Whole: Foundational

Considerations 14 Exposition 16 Development 18 Recapitulation 19 Repetition Schemes 20 THREE

The Medial Caesura and the Two-Part Exposition 23 The Two-Part Exposition 23 The Medial Caesura (MC): Definitions and Overview 24 The Medial Caesura: Harmonic Defaults 25 The Medial Caesura: Common Characteristics 30 The Deployment Sequence of Medial Caesura Options 36

Caesura-Fill (CF) 40

xu Contents Medial Caesura Declined 45 Deformation: The Blocked Medial Caesura 47 Troubleshooting MC Identifications 48 FOUR

The Continuous Exposition 51 Continuous Exposition Subtype 1 (“Expansion-Section”

Subtype) 52 Continuous Exposition Subtype 2: Early PAC in the New Key Followed by (Varied) Reiterations of the Cadence 60 Difficult Cases: Incipient or Not-Fully-Realized Medial

Caesuras 63 FIVE

The Primary Theme (P) 65 Structure 69 Thematic/Modular Designations: Numberings 71 Exceptions in Thematic Numberings 72 “Zero-Modules”: General Considerations 72 Tonal Under- and Overdetermination 73

Some Special P-Types 77 Pas Grand Antecedent 77 Mozartian “Loops”: A Specialized Variant of the Sentence 8&0

P°- and P!.°-Modules/Themes 86 The “Circular” 8—b7—6—47-8 Pattern in P-Space 91 Peer and Pt! Themes 92 STX

The Transition (TR) 93 When Does a Transition Begin? 94 Common Transition Strategies: The Independent and Developmental

Types 95 The Independent (Separately Thematized) Transition 95 The Developmental Transition 95 Common Transition Strategies: The “Dissolving” Types 101 The Dissolving Restatement 101 The Dissolving Consequent 101 Period with Dissolving-Consequent Restatement (or Sentence with

Dissolving-Continuation Restatement) 102 The Dissolving P-Codetta: Reiterated Cadences (Dissolving Cadential Reinforcement) 102

Contents xi The Dissolving P-Codetta: Tonic Prolongation via the “Circular”

8—b7-6-47-8 Model 103 Sentence with Dissolving Continuation Module 105 The Dissolving Hybrid 106 Larger, Rounded Structure (ABA', aa'ba") with Dissolving

Reprise 108 Special Minor-Mode and Other Mixed Cases 111 Premature or Delayed Tutti Affirmations 113 Other Problematic TR-Issues 114 Tutti-Affirmation Full Restatements: One or More I:PACs within

TR 114 TR Rhetoric Lacking? The Modulating Consequent 114 TR Rhetoric Lacking? Multiple Phrases Ending HC Eventually

LeadtoS 115 TR Rhetoric Lacking? P Ends with I:PAC or I:LAC and S Follows

Directly 116 TR as Energy-Loss? 116 SEVEN

The Secondary Theme (S) and Essential Expositional Closure (EEC): Initial Considerations 117 Historical Discussions of P- and S-Space 118 Tonal Choices forS 119 Essential Expositional Closure: The First-PAC Rule 120 Proportions of S- and C-Space 124

Structure 124 Theme Types within Allegro Movements 131 The Bustling, Staccato, Energetically Galant, or Jauntily

Self-Confident S 132 The Lyrically “Singing” or Gracefully Cantabile S 133

The P-Based S135 S as “Contrasting Derivation” from P 136

The Forte S136 S as Virtuosic Figuration 137 The “Learned-Style” or Fugal/Imitative S 137 The Multimodular S (MMS): Lengthy S-Themes (or S-Modular

Groups) 139 P- or TR-Material in the Interior of S-Zones 140 Minor-Mode Modules within S141

S? and S'° Themes 142 S? or S!9 Following a V:HC Medial Caesura (or III:HC in Minor-Mode Sonatas) 142

xiv Contents S® or S!-9 Following a I: HC Medial Caesura in Major-Mode

Sonatas 145 Other S° and S!" Types 145 Additional Issues within S-Space 145 Gendered S-Themes? (Masculine/Feminine) 145 Some Schenkerian Implications 147 EIGHT

S-Complications: EEC Deferral and Apparent Double Medial Caesuras (TMB) 150 Retrospective Reopenings of the First PAC with Following

Material 151 Repetition of the Immediately Preceding S-Melody or Its Concluding

Portion 151 Persistence of S-Material Past the Firsts PAC 151 Revitalization of a Portion of S- (or FS-) Material after Starting a

New Module 152 Production of an Additional MC-Effect or Nonelided Cadence

Shortly into Presumed C-Space 159 Problems with a Potential First PAC: Ineffective or Weakened

Cadences 163 The First PAC Arrives “Too Early”; The Implied S-Zone Is Contextually Too Underdeveloped; Thematic “Loops” 163 Substitution of an Imperfect Authentic Cadence (LAC) for the More

Usual PAC 167 The Evaded PAC 169 The Attenuated PAC 170 Apparent Double Medial Caesuras: The Trimodular Block

(TMB) 170 Deformation: Failed Expositions 177 No Secured EEC within the Exposition 177 Failed Exposition: EEC-Substitute in the Wrong Key 178 NINE

The Closing Zone (C) 180 The “Non-S-ness” of C181 Exceptions 182 C as S-Aftermath 182 C-Theme Types 183 Codetta-Module/s] 184 The P-Based C184

Contents xv

The TR-Based C185 New Themeas C 186 Crescendo-Module as Onset of C186 Concluding C-Module as Piano Afterthought 187

C? Themes 187 S© Themes: Apparent C-Zones in the Absence of an EEC 190

The Retransition 191 TEN

The Development (Developmental Space) 195 Tonal Layout 196 The Development asa Whole 196 Substitutes for V, at the End of the Development (Lower-Level

Defaults) 198 Rhetorical/Thematic Layout: Developmental Rotations 205 Developmental Rotations: First Principles 206 The Onset of the Development; P-Material as the Norm;

Fifth-Descents 207 Episodic Openings: “Writing Over” 212 C-Based Openings 215 S-Based Openings 216 Developmental Rotation Types 217 Half-Rotations: P-TR = 217

Half-Rotations: S-C? 217 Double Rotations 218 Double or Triple Half-Rotations 218 Episodic Interpolations or Substitutes in the Center of the

Developmental Space 218 Tonic-Centered Episodes 219 Introductory Material in Developments 219 “Slow-Movement” Episodes within Allegro-Tempo Movements (or Slow

Movement as Development) 220 The “False-Recapitulation Effect” 221 The Continuum of Haydn’s “False-Recapitulation Effect” 223 The Off-Tonic False-Recapitulation Effect 226 Interpreting Developments: Models for Analysis 228 Topical Dramas: The Ordering of Established Rhetorical

“Topics” 228 Sequence-Blocks: The Ratz-Caplin Model 228 Sonata Theory: Overview of Typical Pathways for the

Developmental Space 229

Xv1_ Contents

ELEVEN

The Recapitulation (Recapitulatory Space; Recapitulatory

Rotation) 231 What Qualifies as a Recapitulation? 231 ESC: Tonic Presence and the Precipitation of the Tonic as a

Crystallized Reality 232 Recompositions, Reorderings, Interpolations 233 The Recapitulatory TR = 235 Altered MC Treatments in the Recapitulation; “Wrong-Key” Starts

for S 237 The Crux 239 Correspondence Measures and the Crux 239 Correspondence Measures and Referential Measures 241 The “Sonata Principle”: A Problematic Concept 242 What Needs to Be Resolved in the Recapitulation? 242 The Fallacies of “Closer Relation” and a “Resolving” Fifth-Transposition 245 Nonresolving Recapitulations: S Does Not Attain the ESC 245 Recapitulations with an S That Fails 245 Truncated Recapitulations: Suppression of the S-C Block 247 The Larger Role of the Recapitulation within the Sonata 250 Tonal Potential, Tonic Presence 250 Narrative Implications: The Sonata as Metaphor for Human

Action 251 TWELVE

Non-Normative Openings of the Recapitulatory Rotation: Alternatives and Deformations 255 Recapitulations That Appear to Begin after P!! 256 The “Disjunct” Recapitulation? 256 Non-“Disjunct” Recapitulations That Appear to Begin with Tonic-

Pl.2, P2, or TR = 257 Rhetorical Recapitulations Beginning in the Parallel Mode 258 Rhetorical Recapitulations Beginning in a Nontonic Key 260

False Starts 260 Recapitulatory Rotations That Begin in IV 262 Recapitulatory Rotations That Begin in vi, VI, or BVI 268 Recapitulatory Rotations That Beginin V = 275 Recapitulatory Rotations That Begin in Other Keys 279 Double-Recapitulation Effects? 279

Contents xvi THIRTEEN

Parageneric Spaces: Coda and Introduction 281 The Coda 281 Definitions, Traditional Views 281 Thematic-Rhetorical Material in Codas: Rotational

Implications 283 Discursive Codas 284 Characteristic Functions of the Discursive Coda 286 Expanded C-Space and Coda-Rhetoric Interpolation

(CRI) 288 The Introduction 292 Slow Introductions and Genre or Movement Types 295 Characteristic Zones within a Slow Introduction 297 Variants and Later Deformations of the “Slow” Introduction 299

Expressive or Representational Functions 300 The Introduction-Coda Frame 304 POURTEEN

Sonata Form in Minor Keys 306 The Extra Burden of Minor-Mode Sonatas 306 Major and Minor as Binary Opposites (Positive and Negative) 307 Expositions in the Minor Mode: EEC in III 310 Recapitulations and Codas in Sonatas with Expositions of the i-III

Type 312 Expositions in the Minor Mode: EEC in v (the Minor

Dominant) 314 (Nineteenth-Century) Expositions in the Minor Mode:

EEC in VI 317 PIFTEEN

The Three- and Four-Movement Sonata Cycle 318 Number of Movements 319 First Movement 321 Slow Movement 322 Key Choice in Slow Movements: Major-Mode Sonatas 323 Key Choice in Slow Movements: Minor-Mode Sonatas 327

Minuet/Scherzo 329 Potential Correspondences with the First Movement 330 Expressive Connotations: Minuets and Scherzos; Major and Minor

Modes 331 The Nontonic Minuet/Scherzo 331

xvul Contents The Key and Character of the Trio 332 Deferral of the ESC in Compound Ternary (ABA) Forms 333

Finale 333 The Role of the Finale in the Trajectory of the Whole Work 334

Major- and Minor-Mode Finales in Minor-Mode Works 335 The Multimovement Cycle as a Complete Gesture 336 The Three-Movement Pattern 336

Multimovement ESC Deferral 337 The Standard Four-Movement Pattern 337 Alterations of the Normative Movement- or Key-Order

Scheme 337 The Role of the Listener 340 SIXTEEN

Sonata Types and the Type 1 Sonata 343 Five Sonata-Form Types 343 The Type 1 Sonata 345 Terminology 346 Historical Origin 348 The Type 1 Sonata with P-Based Discursive Coda 349 The Expanded Type 1 Sonata 349 The Type 3 Sonata with Expositional-Repeat Feint: A Related but Differing Structure 350 SEVENTEEN

The Type 2 Sonata 353 Inappropriateness of the Term “Recapitulation” 353 Historical Considerations 355 Earlier Musicological Treatments of Type 2 Sonatas: “Binary-Variant”

Structures and the Problem of “Reversed Recapitulations” 365

The Type 2 Sonata as a Constellation of Generic Options 369 The End of the Exposition and the Beginning of Rotation 2:

Two Options 369 P-Based Openings to the Developmental Space 372 Episodic Openings to the Developmental Space 373 S- or C-Based Openings to the Developmental Space:

Type 3 Type 2 Conversions 376 Developmental Activity Proper 378

The Crux 379 The Tonal Resolution (StC) 380 The “Recapitulation” Question and Tonal Resolutions in

Continuous Second Rotations 380

Contents X1x

Type 2 Sonata Forms without Codas 381 Type 2 Sonata Forms with P-Based Codas and the “Reversed

Recapitulation” Fallacy 382 The P-Based Coda: The Potential for a Secondary

“Completion-Effect”? 384

P-Compensatory Codas in Non-Type 2 Sonatas 356 Confronting Hard Cases: Flexibility in Sonata-Type Recognition 386 EIGHTEEN

Rondos and the Type 4 Sonata 388 Definitions 389 Rondeaux, Rondos, and Rotations: A Preliminary Note 390

The Rondeau 392 Traditional and Expanded Rondeaux 392 The Two-Couplet Rondeau 393 The Multicouplet Rondeau 393 The Symmetrical Three-Couplet Rondeau 394 The Expanded Rondeau 396

The Rondo 397 Influence of Earlier Dance-Suite Binary Formats? 397 The Retransition: A Crucial Marker of the Rondo 398 Refrains of “Rondo Character”; Occasional “False Predictions” and Composers’ “Mislabelings” 398 The Five-Part Rondo (AB—AC-A) 399 Variants of the Five-Part Rondo (AB—[...]C—A and

AB-AB-A) 400 The Seven-Part Rondo or “Chain Rondo”

(AB—AC-AD-A) 401 The Symmetrical Seven-Part Rondo (AB—AC-—AB'-A) 402 Refrain-Material in Nontonic Keys: The Rondos of C. P. E.

Bach 403 The Sonata-Rondo (The Type 4 Sonata) 404 What Qualifies as a Sonata-Rondo? 404 The Standard Type 4 Sonata (The Type 3 Sonata-Rondo

Mixture) 405 The Type 1 Sonata-Rondo Mixture (Type 4!) 407 The Expanded Type 1 Sonata-Rondo Mixture (Type 4'-°xr) 409 Haydn’s Treatments of Type 4 Finales 413 Type 4! Variants 415 Omission of RT and Any Tonic Reference to P* after the

Exposition 416 More Radically Recomposed Recapitulatory Spaces 416

No Return to P¥ as a Literal Refrain after the Recapitulation 417

xx Contents Type 4 Procedures in Mozart’s Concerto Finales 417

Type 4 Formats in Mozart’s Concerto Finales 418 Pf and the Possibility of Tutti-Extensions (“Ritornello 1”) 420 The Solo Re-entry in the Tonic (the Sujet-Libre TR!) 425 ESC Issues in Type 4 Sonatas: The Double Perspective 428 NINETEEN

The Type 5 Sonata: Fundamentals 430 Historical Overview and Initial Questions of Terminology 433 Vogler, Koch, and Others: Six Tutti-Solo Layouts 435 Vogler’s Description (1779) of the Concerto 435 Koch (1793) and the Two Four-Ritornello Layouts:

Subtypes A and B 436 The Problematic Status of the Recapitulatory Tutti,

R3 (R383) 440 Tutti-Solo Layouts, Subtypes C-F 441 Overview: Sonata Form, Mozart, and the Seven-Part (Four-Ritornello) Framework 443 More on Terminology: “Ritornello” or “Tutti”? 445 The Status of Ritornello 1(R1) 447 R1 and the Subsequent Exposition: The Question of Conceptual

Primacy 447 Conclusion: The Three Structural Functions of Ritornello 1 449 Sonata Theory’s Thematic Labeling for Type 5 Sonatas 451 Basic Concerto Principles: Mozart’s Early Pastiche Concertos,

K. 107 453 Adapting J. C. Bach into K. 107, No. 1/i: Opening Ritornello and Smaller and Larger Expositions 454 EEC Issues in Type 5 Sonatas: The “Double Perspective” 462 Developmental Spaces and Recapitulations in the K. 107

Concertos 466 A Summary of Structural Axioms Exemplified by K. 107 467 TWENTY

The Type 5 Sonata: Mozart’s Concertos (R1: The

Opening Ritornello) 469 The Paradox of Mozart’s Concertos 470 Ritornello 1: The Logic of Modular Succession 471 Modular Descriptions: Sonata Theory and Some Alternatives (Leeson-Levin, Kiister, Stevens) 471 Ritornello 1: The Individual Zones 475 The Primary Thematic Zone: R1:\P_— 475

Contents Xx “Motto” R1:\P as Idée Fixe or Later “Wild Card” 482 The Transition: R1\TR = 483 R1: Medial Caesura and Caesura-Fill 484 No R1:\S Produced: The Continuous R1 487 The Secondary-Thematic Zone: R1:\S (Tonality) 488 The Secondary-Thematic Zone: R1:\S (R1:\EEC Issues) 490 The Secondary-Thematic Zone: R1:\S (Implications for Later

Rotations) 492 The Closing Zone: R1\C 493 Recurrences of R1 Modules in Later Rotations 494

TWENTY-ONE

The Type 5 Sonata: Mozart’s Concertos (Solo and Larger

Expositions: Solo 1 + Ritornello 2) 496 The Soloist Enters: The Interaction Begins 496 Hermeneutic Issues: Individual and Group 496 Option 1: Solo Enters with R1:\P Material (Onset of Solo Exposition and Rotation 2) 498 Option 2: Solo Enters with New Material Preceding the Onset of R1:\P: Links, Bridges, and Prefaces (“Rotation 2”

Ambiguities) 498 Option 3: Solo Substitutes a New Theme for R1:\P (Onset of Rotation 2 with an S1 “Replacement Theme” 520

Implications: How Free Will Rotation 2 Be? 521 Post-P-Theme Options: Orchestral Flourish as Affirmational Tutti Interjection 1 (P-Codetta or Link to a Sujet-Libre

Transition) 521 The Normative Case and Its Hermeneutic Implications: The

Nonmodulatory P-Codetta 522 The Sujet-Libre Transition Type 525 Sujet-Libre TR-Variant 1: “Linkage Technique” 527 Sujet-Libre TR-Variant 2: S1\TIT! Omitted 528 S1:\TT! as Mediator between the P- and TR-Zones 528 The Modally Shifting or Modulatory S1:\TT! and Sujet-Libre

Transitions That Begin Off-Tonic 528 Two Other Solo 1 Transition Types 529 Alternative 1: Solo 1 Transition as Dissolving R1:\TR!!, Usually with Subsequent Expansions 532 Alternative 2: Solo 1 Transition as Dissolving P>TR

Merger 533 From the (First) MC to S1:\EEC (S- or TMB-Space) 533 The S1:;\EEC: A New Requirement in the Type 5 Sonata 534 The Musical Content of S1:\S-Space 535

Xxll_ Contents

Options in Which All or Some of R1:\S Appears in Solo 1 535 Options in Which R1:\S Does Not Appear in Solo 1 540 The Display Episode (S1:;\DE) and the S1\EEC 542 The Relationship of the Display Episode to the S1:\EEC 543 The Morphology of Display Episodes 544 Display-Episode Variants: Playing with Signs 545 Ritornello 2 and Its Role in the Larger Exposition 548

The R2\EEC 552 Option 1: R2 Begins with a Rotationally Inert Space before Shifting

into Rotationally Participatory Modules (R1:\T'R!! and Other

Openings) 552 Option 2: R2 Begins with an R1:\C Module Unsounded

inS1 557 Option 3: R2 Begins with an R1:\S Module Unsounded

in St 557 Option 4: Alternative or Quasi-Deformational R2 Spaces 558 Alternative Tracks for the End of R2: Modulations, Half

Cadences 559 TWENTY-TWO

The Type 5 Sonata: Mozart’s Concertos (Development and Recapitulation: From Solo 2 through

Ritornello 4) 563 The Developmental Space: S2 or S2 + “R3” 563 Modular Content: Frequency of “Episodic” Developments 563

Modular Content: Linkage Technique 564 Rotational Developments 570 Overall Shape: Event Zones 571 Tutti Interjections = 573 The Retransitional “Ritornello 3” Option Proceeding to a Solo-Led

Recapitulation 573 The Solo Recapitulation: S3, R3 + S3, or R3=9S3 577 Solo and Larger Recapitulations 577 The Recapitulation as Rotation 1—Rotation 2 Synthesis 577 Varying Relationships of the Solo Recapitulation to S1 = 579 Solo-Led (S3) Openings 581

Double-Start Openings (R3 + S3) 582 R3=>S3 Merger Openings 585 The Solo Recapitulatory Transition 586 Solo Recapitulatory S- and C-Space (S3:\ESC) 588 Option 1: No Restoration of R1:\Post-MC Material Suppressed in the Larger Exposition 588

Contents Xxill Option 2: Restoration of R1:\Post-MC Material Suppressed in the

Larger Exposition 590 Ritornello 4 and the Conclusion of the Larger Recapitulation 596

R4! and R4 Subdivisions 596 R2 Material within Parts of R4: Three Options 597

The Cadenza 600

Appendix 1. Some Grounding Principles of Sonata

Theory 603 The Intellectual Backdrop 603 “Texts Always Take Place on the Level of Their Reader’s

Abilities” 605

Appendix 2. Terminology: “Rotation” and “Deformation” 611 Rotation 611 Deformation 614 Connotations of “Deformation”: Paradoxes of the “Normative” and

“Non-Normative”; the Need for Nuance 614 Precedents in the Use of the Term 619

Bibliography 623 Index of Names 633 Index of Works 639 Index of Concepts 649

This page intentionally left blank

Terms and Abbreviations

PAC = perfect authentic cadence (a phrase-concluding formula featuring V-I root-position bass motion; the upper voice ends on scale-degree 1 above the tonic chord) IAC = imperfect authentic cadence (similar to PAC, but the upper voice ends on scale-degree 3or 5above the tonic chord) HC = half-cadence (a cadence ending on an active V chord; this dominant chord will also end a phrase) DC = deceptive cadence (V-vi, or V followed by any non-tonic chord containing 1, where an authentic cadence is expected) Vy =a V that is tonicized; the dominant sounded as a key (as in second themes of major-mode expositions)

Va, =a V that is an active chord, not a key; the A stands for “active,” and it indicates that the dominant is being sounded but not tonicized; instead, it implies a resolution to the existing or implied tonic. C = closing zone (within an exposition, musical material following the EEC. Its internal modules are designated as C!, C?, etc.; in this case the superscript integers should be advanced only after a PAC.) CF = caesura-fill (connective material, of variable length, bridging a caesura—either a medial caesura or a final caesura—to the next thematic module) Cpre-EEC = A “C”like theme that occurs before the EEC proper within a continuous exposition. (Within a two-part exposition, such a theme 1s designated as S©.)

CRI = coda-rhetoric interpolation (coda-like material interpolated shortly before the close of the recapitulatory rotation, which then resumes to complete the recapitulation proper) DE = display episode (in a Type 5 sonata [concerto movement] XXV

xxvi_ Terms and Abbreviations

the solo-virtuosic closing portion, ending with an emphatic trill cadence, of $1 and S3—the solo exposition and solo recapitulation. The location of the display episode is usually included in the label, as in S1:\DE. See S1:\.) EEC = essential expositional closure (within an exposition, usually the first satisfactory PAC that occurs within S and that proceeds onward to differing material. An immediate repetition of the melody or cadence—or certain other procedures, outlined in chapter 8—can defer this point to the next PAC.) ESC = essential structural closure (within a recapitulation, usually the first satisfactory PAC that occurs within S and that proceeds onward to differing material. Like the EEC, the ESC can also be deferred through certain procedures to the next PAC. The ESC 1s normally the recapitulation’s parallel point to the exposition’s EEC, although exceptions do exist.) FS = Fortspinnung modules (usually in the continuous-exposition context of TR=>FS) MC = medial caesura (within an exposition, I: HC MC represents a medial caesura built around the dominant of the orig-

inal tonic; V:HC MC represents an MC built around V/V; etc. The presence of an MC identifies the exposition-type as two-part—the most common type—and leads directly to an S theme. In nearly all cases, if there is no MC, there 1s no S. Cf. the alternative, TR=>FS.) MMS = multimodular S (an S that tracks through two or more different, often contrasting ideas—S!:!, S!-2, and so on—before driving to its first satisfactory PAC with a cadential module. The numbers after the decimal point—the “decimal designators’—provide a method of labeling and identifying these separate modules. A trimodular S is particularly common: see TMS.)

P = primary-theme zone (whose individual modules may be described as P!-!, P!-2, etc. A module that precedes or sets up what is taken to be the “P-theme proper” may be designated as P® or P!-°.)

PMC = postmedial caesura (any emphatic MC-effect that occurs in an exposition after the first MC; a “second” MC-production, sounded several measures past an initial, fully successful MC.)

Ptf= the specialized P-theme within a Type 4 sonata—sonata-rondo—that also functions as a recurring, refrain theme with “rondo character,” often also displaying a characteristic refrain-theme structure. R1 = the initial ritornello (Ritornello 1 or opening tutti) at the opening of a Type 5 sonata (concerto movement). Similarly, R2,R3, and R4 stand for the second, third, and fourth ritornellos (or tuttis), each of which also has a specialized function and role to play within a Type 5 sonata. R1:\ = prefix indicating material within R1 of a Type 5 sonata

(concerto movement). (Thus R1:\P, R1:\S, and R1:\EEC

Terms and Abbreviations xxvii

represent the modules functioning as the primary theme, the secondary theme, and the rhetorical EEC within the opening tutti of a Type 5 sonata.) RT = retransition (a connective passage of preparation, usually leading to the onset of a new rotation, that is, to the repeat of the exposition, to the onset of the recapitulation, or to the beginning of the coda) S = secondary-theme zone (follows an MC. This is built from precadential, pre-EEC thematic modules. Differing musical ideas within it, when they exist, are designated with superscripts as S!!, S!-2, and so on. [See MMS and TMS.] A module that precedes or sets up the S-theme proper may be designated as S° or S!-°, Not to be confused with S1.) S1 = the first solo section, Solo 1, of a Type 5 sonata (concerto movement), typically marked by the first entrance of the soloist following the orchestral R1 and ending with a trill cadence

precipitating the onset of the second ritornello or tutti, R2. S1 is also the “solo exposition,” even though, as discussed in Chapters 19 and 21, this is normally extended into a “larger exposition” —rotationally defined—with the addition of the immediately subsequent R2. Similarly, within concerto movements S2 and S3 stand for the second and third solo section. S2 is usually the developmental space of the Type 5 sonata. S3 (or sometimes R3=>9S3) is normally the “solo recapitulation,” also extendable into the “larger recapitulation” with the addi-

tion of R4 (chapter 22). (Notice that in the concerto-space designation, S1, the numeral 1s not superscripted. When it 1s, as in S!, S?, and so on, it refers not to Solo 1 but to a portion of secondary-theme space. In concerto movements the two may appear in the same description, as with S1:\S!-2, or “the second module of S-space within Solo 1 of a Type 5 sonata.” See S.) S1:\ = prefix indicating material within the S1 zone of a Type 5 sonata. (Thus S1:\P, S1:\S, and S1:\EEC represent the modules functioning as the primary theme, the secondary theme, and the EEC within the Solo 1 space of a Type 5 sonata. See S1.)

S© = a theme within S-space (and thus before any clear articulation of an EEC) that, for any number of reasons, seems to take on the features and style more characteristic of a closing theme (C). Cf. Cpre-EEC,

TI = tutti interjection (in a Type 5 sonata, any brief, interrupting tutti impulse within what is otherwise a solo section, such as S1, S2, or S3. The first of these to appear, S1:\TI', shortly into Solo 1, 1s often formulaic and stylized, as noted in chapter 21.)

TMB = trimodular block (an especially emphatic type of multimodular structure in an exposition or recapitulation, always associated with the phenomenon of apparent double medial caesu-

ras. Individual modules may be designated as TM!, TM2, and TM3. Of these, TM! and TM? are usually “thematic.” TM!

xxvil Terms and Abbreviations

follows the first apparent MC, TM? often reinvigorates the TR-style [often TM! merges into TM?, TM!=TM?] and helps to set up the second apparent MC, and TM> follows that second MC-effect. A TMB leads, at its end, to the EEC. Either TM! or TM? may give the impression of being the “real” S depending on the individual circumstances. Cf. TMS.) TMS = trimodular S (a common type of MMS with three Smodules. Within the sonata narrative the first proves “unable” to produce a PAC; the second often thematizes the threat or difficulty; the third is a decisive cadential module. It differs from the TMB in its lack of apparent double medial caesuras: there is no second “apparent” MC after the second S module.) TR = transition (following P, the energy-gaining modules driving toward the medial caesura)

TR=> FS = the broad middle section of a continuous exposition that begins as a transition (TR) but at a crucial “point of conversion” midway through 1s often better described as Fortspinnung (FS) or, in other cases, a chain of thematic modules. Either procedure avoids producing a clear MC and the resultant two-part exposition. The > (“becomes” or “merges into”) represents the conceptual point of conversion.

About the Authors

James Hepokoski, professor of music at Yale University, specializes

in formal structure and hermeneutic issues in sonata-form-based repertories, ca. 1750-1920. He is the author of four books and numerous articles in a variety of areas, including Italian opera (Verdi), early-modernist composers (Sibelius, Strauss, Elgar), American and Germanic music-historical methodologies, and current literary-critical/cultural approaches to music. He was a co-editor of the journal 19th-Century Music from 1992 to 2006.

Warren Darcy is professor of music theory and former director of the Division of Music Theory at the Oberlin College Conservatory. He has lectured and published widely on Wagnerian opera, and his book Wagner’s “Das Rheingold” (Oxford, 1993) won the Society for Music Theory’s 1995 Wallace Berry Award. He has also published on the music of Bruckner and Mahler and is currently engaged in a large-scale study of rotational form in Mahler’s symphonies.

XXIX

This page intentionally left blank

ELEMENTS of SONATA THEORY

This page intentionally left blank

CHAPTER ONE © "ISLE 0

Contexts

Differing Approaches to Sonata Form of the larger playing field. In influential Englishlanguage scholarship today one might recognize

There is no consensus regarding the manner in four general trends: two broad musicological which sonata form in the decades around 1800 lines and two broad music-theory lines. To be is to be grasped. On the contrary, analysts are sure, the categories overlap —they are anything confronted with a clutch of diverse approaches but airtight—and within each there are differwith differing emphases, interests, and termi- ences and varied accents in the way the gennologies. This is contested terrain, particularly eral method 1s formulated. Still, musicology and since the structure is basic to how we conceptu- music theory have often pursued distinct paths, alize the Austro-Germanic art-music enterprise generating different questions and answers. stemming from Haydn, Mozart, Beethoven, and The two broad musicological approaches,

Schubert. Our contribution, Sonata Theory, sometimes intersecting, are: (1) the style of provides a via media among these approaches, eclectic analytical writing favored by Donald remaining open to the positive insights that Francis Tovey and carried on (and varied) by each has to offer and for the most part remain- such differing writers as Joseph Kerman and ing methodologically compatible with them all. Charles Rosen and (2) the more strictly “hisAt the same time we propose new, genre-based torical-evidentiary-empirical” concerns of perspectives, along with useful ways of formu- such diverse figures as William S. Newman, lating analytical questions and moving on to _—_— Jan LaRue, Eugene K. Wolf, Leonard G. Rat-

productive hermeneutic endeavors —interpre- ner, and their successors. ‘The two broad mu-

tations of meaning. sic-theoretical approaches are: (3) Schenkerian Situating oneself within a conflicted field is and post-Schenkerian methodologies and (4) a risky, fallible enterprise, in part because one is lines of analysis emphasizing motivic growth obliged reductively to characterize the work of — from small musical cells, as well as the idenothers —and those others nearly always object tification of phrase-shapes and the patterns of (often rightly so) to such characterizations. And _ larger sectional blocks—a style of analysis as-

yet it may be helpful to sketch out some rough sociated with Arnold Schoenberg, Rudolph descriptions of viable approaches to the subject Réti, and Hans Keller, and including the work of sonata form, if only to suggest an impression of Erwin Ratz and, most recently, William E. 3

4 Elements of Sonata Theory

Caplin.! At the risk of oversimplification (and section”) that needs to be resolved in the recawith apologies to those unmentioned), we pitulation.® A central feature of Rosen’s writing might characterize the interests of these four (as well as that of Tovey and Kerman) was the categories by citing an example of an important description of individual compositional styles

text within each. and preferences, along with the pronouncement 1. Our first-category illustration is Charles of cleanly-divided aesthetic judgments of the Rosen’s Sonata Forms (1980, rev. 1988).2 Draw- works at hand—strong praise for the mastering on the analytical and prose style of Tovey works contrasted with tart dismissals of works and grounded in a vast knowledge of the rep- deemed not to make the grade.

ertory, Rosen’s magnum opus stressed the va- 2. The second category is best represented riety of procedures that one can encounter in by Leonard G. Ratner’s Classic Music (1980).° the “texture” or “process” that we now call so- Somewhat parallel to the scholarly-inventory nata form. (Hence his plural, “forms,” echoing work of William S. Newman and Jan LaRue, Tovey.)? Rather than elaborating an intricate Ratner sought to reconstruct the concept of the background plan for the form, Rosen preferred eighteenth-century style from the point of view to demonstrate how difficult— or futile —it is of the eighteenth century itself. The book was to provide a set of detailed expectations regard- to be ing it because of the unique things that occur in individual pieces by composers of genius. As a full-scale explication of the stylistic premises of 4 matter of principle Rosen shunned the idea classic music, a guide to the principles according of a “general practice” for the construction of to which this music was composed. . . . The exsonatas—except for a few tonal requirements position of 18th-century musical rhetoric is found and common textural choices— although there m theoretical and critical treatises. . . . [These writings] point to what was current then, illumiwere clearly better and more masterly solu- . . nating our present view of the music. Coordi-

tions to the general set of problems atmusic hand.*itself, . . _ the data - nated with analysis of the This somewhat intuitive approach, acute and gleaned from these writings make it possible to invariably musical, also emphasized the con- determine the basic criteria of expression, rhetocept of tonal “polarization” (usually tonic and ric, structure, performance, and style that govern dominant) in expositions and famously regarded classic music... . This book allows the student to the expositional shift to a non-tonic key as an approach the music and musical precepts of the “opposition[al]” move, a “large-scale disso- 18th century in much the same way a listener of nance” (“structural dissonance” or “dissonant that time would have done.’

1. But even these broad categories are too limiting. In- University Press, 1944) [reissued in 1956 under the title termixed throughout them all are the various traditions The Forms of Music|, pp. 208-32. passed on in the Formenlehre, the academic textbooks 4. Rosen, Sonata Forms, rev. ed., pp. 4—7. Cf. the differof form, which seem to have a separate reception-life ing impression conveyed in W. Dean Sutcliffe’s review, of their own. In addition, other influential European in Music & Letters 79 (1998), 601—4, of Rosen’s modest perspectives that sometimes escape from or provide al- revision of his earlier work The Classical Style: Haydn, ternative havens within the above four categories have Mozart, Beethoven, exp. ed. (New York: Norton, 1997 also proven provocative for current work— one thinks, [orig. ed., 1971]). This review, in part, calls attention to for example, of the work of Jens Peter Larsen and Carl the earlier book’s apparent “emphasis on the normative Dahlhaus. Moreover, in recent years differing scholars aspects of the style . . . stereotypes and formulas” —conhave begun to seek new ways to blend together formerly cerns that raise a host of questions in these more skep-

differing methodologies. tical times and ones that Rosen himself had sought to

2. Rosen, Sonata Forms, rev. ed. (New York: Norton, clarify in the later Sonata Forms.

1988 [first ed. 1980]). 5. Rosen, Sonata Forms, rev. ed., pp. 98-99, 229, 287.

3. Donald Francis Tovey, “Sonata Forms,” originally See also Rosen, The Classical Style, exp. ed., p. 33. two different entries for the 11th (1911) and 14th (1929) 6. Ratner, Classic Music: Expression, Form, and Style eds., the latter of which 1s reprinted in Tovey, Musical (New York: Schirmer, 1980). Articles from the Encyclopaedia Britannica (London: Oxford 7. Ratner, Preface to Classic Music, pp. xiv—xv1.

Contexts 5 Not surprisingly, Ratner paid close attention torically defensible musical “topics” (standardto the early theorists’ descriptions of what came ized musical gestures or types within phrases) to be called (c. 1824-1845) “sonata form.” The and eighteenth-century conceptions of “rheto-

Newman-LaRue-Ratner projects (however ric’ in this repertory. they might differ in other respects) were ones 3. Moving to the music-theory side of things, of data-gathering and recovery. One of their the touchstone of the third category is Heinfeatures was to urge analysts to sideline nine- rich Schenker’s Der freie Satz (1935, translated teenth- or twentieth-century views of sonata as Free Composition).'!! For many music theorists form in order to gain a more period-conscious interested in sonata form, no text 1s more cenconception of the form.® (In this regard these tral than this one. Opposed to traditional ways interests are not without parallel to the perfor- of discussing musical structure, Schenker was mance-practice movement and its quest for “au- convinced that he had discovered a new thethenticity.”) To varying degrees scholars within ory of form, “a new concept, one inherent in this circle seek to describe sonata form (and the works of the great masters; indeed, it 1s the other forms) from the perspective of late-eigh- very secret and source of their being: the conteenth-century theorists — favoring their term1- cept of organic coherence.”!? This theory was nology and concerns and being cautious about to be grounded not in phrase- or section-repsoing beyond them.? Writers influenced by this etitions or in thematic manipulation but rather point of view call upon the authority of late- in linear-contrapuntal views of the sonata as the eighteenth-century or early nineteenth-cen- unfolding of a “fundamental structure” (Ursatz) tury writers on the form (such as the important by means of more elaborate middleground and statements of Heinrich Christoph Koch, Fran- foreground structures. Middlegrounds and forecesco Galeazzi, Augustus Kollmann, and Anton erounds are understood as florid “diminutions” Reicha). Several of them have also tended to of more simple, elemental background gestures view harmony (modulations, key-areas visited, elaborated over the course of an entire moveand so on) as the primary feature of sonata form ment. The method 1s highly sensitive to conin the years from roughly 1750 to 1820— giv- trapuntal, linear voice-leading, long-range proing it the upper hand over thematic arrange- longations or descents of important individual ment. In the mid-twentieth century Ratner fa- pitches, and the like. Here sonata form is unmously contested the earlier, thematic view of — derstood as divided into two parts (expositionthe sonata, which he regarded as discredited, an development || recapitulation) with a crucial anachronistic, nineteenth-century (mis-)under- harmonic “interruption” (||) at the end of the standing of the form as it had been originally development and a subsequent rebeginning at grasped in Beethovenian and pre-Beethovenian the onset of the recapitulation, which restates decades.!° Some writers influenced by Ratner’s and finally completes the fundamental structure work are also concerned with identifying his- interrupted at the end of the first part.!9 8. See, e.g., Eugene K. Wolf, “Sonata Form,” in The 11. Schenker, Free Composition (German original, Der New Harvard Dictionary of Music, ed. Don Michael Ran- freie Satz, 1935), trans. and ed. (with additional com-

del (Cambridge, Mass.: Harvard University Press, mentary) Ernst Oster (New York: Longman, 1979). Es1986), pp. 764-67. This essay outlines the rhetorical- pecially relevant is part 3, ch. 5 (“Form”), pp. 128-145. tonal structure at hand and provides a historical over- “Section 3,” on “Sonata Form” (including Oster’s faview of the origins and transformations of the form. mous footnote), is found on pp. 133-41. 9. In other respects Ratner-related styles of analysis 12. Schenker, Free Composition, p. xxi. seem to be musicological variants of the well-estab- 13. Also to be noted in terms of Schenkerian and postlished sector of music theory, “history of music theory.” Schenkerian analysis is the summary of sonata form in A more purely music-theoretical analogue is Joel Les- Allen Cadwallader and David Gagné, Analysis of Tonal ter, Compositional Theory in the Eighteenth-Century (Cam- Music: A Schenkerian Approach (New York: Oxford Uni-

bridge, Mass.: Harvard University Press, 1992). versity Press, 1998), esp. ch. 11, “Sonata Principle,” 10. The locus classicus of this position is Ratner, “Har- pp. 303-59. Similarly, one should mention William monic Aspects of Classic Form,” Journal of the American Rothstein, Phrase Rhythm in Tonal Music (New York:

Musicological Society 11 (1949), 159-68. Schirmer, 1989), particularly ch. 4, “Phrase Rhythm

6 Elements of Sonata Theory

4. Our example of the fourth category 1s wake of Tovey’s similar assertions!©— has been William E. Caplin’s Classical Form (1998).'4 Its the repeated declaration that the “textbook” opening paragraph proclaimed the need for “a — view of sonata form is inadequate to deal with new theory of classical form,” one that avoids the actual musical structures at hand. At best, “ill-defined concepts and ambiguous terminol- such a scheme represents a conformist trap that ogy derived from theories that have long fallen master-composers avoid falling into. In addiinto disrepute.” Following the work of Schoen- tion, the implication has sometimes been that berg and Ratz,!° Caplin viewed form as a group- to undertake any such “textbook” description ing structure, and he set out to identify and classify of norms, however nuanced or sophisticated, is

the “formal functions” of smaller thematic/for- a mistaken enterprise. It is not difficult to find

mal units. In practice, this entailed close at- conventionalized avowals on these matters. tention to the structures and subparts of three Here is a strong version of the credo from Claufundamental theme types: the sentence (consist- dio Spies, excerpted from an essay in a book of ing, for Caplin, of presentation, continuation, Brahms Studies (1991): and cadential functions; or basic idea [usually repeated, perhaps with variation] + fragmenta- There is nothing new about “forms” with whose tion + cadence); the period (antecedent + con- aid pieces of music are easily and lazily categosequent); and the small ternary (A—B-—A'). Much rized or typified, tagged, pigeon-holed, and conattention was also given to the anatomy of nu- veniently stored away without further—or even merous “hybrids” that mix aspects of the more prior — hearing, and without further thought. standard theme types (as defined by the author). We were all initiated KO the NON-MYSEETIONS SEO

As the musical parts are assembled, they can lidities of form, particularly the most fictitious one of all, “Sonata Form.” Nor is there, I hastake on “framing functions,” “interthematic ten to add, anything new in the notion that such functions,” “harmonic functions,” “initiating “forms” —and especially “Sonata Form” —are functions,” “continuation functions,” and so fictions to whose specifications and proclaimed on, often at more large-scale levels. One aim norms very few pieces of music worth any further of analysis is to be able to recognize the theme thought actually conform in any appreciable way.

types (and hybrids) and to place them into a ...Itis almost as if Brahms had decided to comlarger functional system of interrelated parts. pose [the Tragic Overture] as a potent rebuttal of In the end, what was provided was an elabo- notions propounded by the tenets of Formenlehre, rate taxonomy of different kinds of phrase-and- although [it] is by no means unique among his

section juxtapositions. works in this respect.

The same point, put more gently —and after an

The War against the Textbooks admirably detailed study of Brahms—may be

found from James Webster in the same volume:

One prominent feature of the study of sonata form in recent decades—very much in the From examples like these it is clear that norms of formal procedure, whether the bad old textbook

and Form: Some Preliminaries,” pp. 102—20. This is an menlehre: Uber Formprinzipien in den Inventionen und Fugen analytically sophisticated discussion of forms in general J. S. Bachs und ihre Bedeutung fiir die Kompositionstechnik

(including sonata form) and, in part, it seeks to blend Beethovens, 3rd ed., enl. (Vienna: Universal, 1973 [1st some of the concerns of Schenkerians with the more ed., 1951]). musicological (and often emphatically non-Schenker- 16. See, e.g., Donald Francis Tovey, “Some Aspects

ian) studies by Rosen, Ratner, and others. of Beethoven’s Art Forms” and “Musical Form and 14. Caplin, Classical Form: A Theory of Formal Functions Matter,” in The Main Stream of Music and Other Essays for the Instrumental Music of Haydn, Mozart, and Beethoven (Oxford: Oxford University Press, 1949), pp. 272-73,

(New York: Oxford University Press, 1998). 160-62; and Tovey, “Sonata Forms,” pp. 210-12 15. Schoenberg, Fundamentals of Musical Composition, (“There are no rules whatever for the number or distried. Gerald Strang and Leonard Stein (London: Faber bution of themes in sonata form’’). & Faber, 1967); Ratz, Einfiihrung in die musikalische For-

Contexts 7 models or the numerical averages developed ear-_ = was more complex than the reflex denunciations lier in this paper, can never satisfactorily account suggested. The reiterated conviction that there

when Brahms’s technique seems most paradox1- eighteenth century, true enough in its narrow,

cal—as in the timeless, themeless, tonic retransi- ; , ,

; have Ie literal sense,artistic rises to result the level tion welesed—_th have just—_ analysed—the is of i. .ani. error when it

often the oe poetic ” is naively taken either to dismiss the presence of

) substantially more complex systems of standard

Remarks along these lines could hardly be practices or to discourage inquiry into those more familiar. Even earlier, by midcentury, it practices. Is there a more effective way of examhad become a scholarly point of honor to de- ining conventional musical gestures (or calling clare war on the textbooks and, for some (again forth that which was conventional within indi-

in varying degrees), on the often wooden vidualized musical gestures) without producing limitations of classifying schemes in general. ideas that were reductive, stiff, mechanical, preWhether uttered in stronger or gentler versions scriptive? Is an aesthetically sensitive openness such declarations advanced unswervingly or- to the study of convention within composition

. . ossible? they were caught up in the traditional philo- roneey ES 8 .

thodox late-twentieth-century convictions, and P Th ly F lated , . _. . e most stron ormulated arguments sophical dilemma of universals and particulars. against generalized principles of sonata practice For the most part—again, much as Tovey had concealed a substantial weakness: in their intendone—they took partisan positions on behalf sity they tempted one to overstate the degree to of the particulars, or at least on behalf of the which such classifications were ever intended to

. a. be equivalent to scientific laws. Within the hu-

ultimate noncapturability of the greatargumasters. 1.4 . 4. manities generic options, Apart from .assessing this norms, neonominalist 5 Pand more-orment on its own terms. it would also be valu less standard procedures are not laws at all. And

, , since they are not, there was no need to supable to investigate the assumptions , ih ¢ - P . . . ose that themodernist existence of numerousPexceptions that made such views possible: the mystification —_ . P of genius; the belief in the compulsion of the or deviations invalidated the norm. Perhaps . . the many deviations were purposeful dialogues

. with theor background norm. applied rules systems; the precept thatBut whatthis 5 —_wou

true artist to escape from confining, externally “ch .. back 4 P - hy: 8 id we most revere in music must not only be be mean, paradoxically, that the deviations helped yond the grasp of academic minds and rational to reinforce the socially shared norm that was

. being temporarily overridden. (Otherwise classification but must always be declared to be _ how so: and so on could they be perceived as deviations at all?) But oe . .and . what is meant by a norm? And Studying teaching musicology and music y .how could one

theory in the 1960s, 1970s, and 1980s, the au come to an understanding of what such norms

, might have Weinto beganthe to seek way out of thors of this book absorbed suchbeen? views on 8a omey marrow of their bones. We also agreed—and the dilemma. The most profitable guidelines for we continue to agree —that prior textbooks had our solution lay within the domains of current

. . . enre theory and hermeneutics.

invited a too rigid understanding of sonata form. sents y _.

. . architecture of later-eighteenth-century literal point is correct and has the added benefit - arecomy So far as the gravamen of the charge goes, the Given the flexibility found in the large-scale

of bringing caution to any new analytical in- position, the main descriptive problem was the quiry. Still, the problem of determining the role difficulty of positing convincing categories of

, ‘ a: _ ays typical As scholars ofyP eighteenthof convention within procedures. this “classical” repertory P8 17. Spies, “‘Form’ and the Tragic Overture: An Adjura- Analytical and Historical Perspectives, ed. George S. Bo-

tion,” and James Webster, “The General and the Partic- zarth (Oxford: Clarendon, 1990), pp. 391 (Spies) and ular in Brahms’s Later Sonata Forms,” in Brahms Studies: 75 (Webster).

8 Elements of Sonata Theory

9 y 9 19

century music perennially point out, surprising ible one. . . . What is needed, then, is a general occurrences and variants abound —all the more theory of form that can account for conventional so when one’s investigation takes a panoramic patterns and at the same time do justice to the imview, extending beyond Haydn, Mozart, and mense diversity that exists within the framework early Beethoven to include the works of less- of these patterns.

explored composers. It 1s for this reason that at- hus the chall culate the implied

.. .ices. .m11 1_--

tempts to describe normative sonata procedures us the chabenge: to articu ate Ee umpire tend to bog down in trying to account for a host pattern-types that appear in some of the clearof seemingly unusual cases (of which there 1s an est OF most notable exemplars and to do this

; with as much supply detail and specificity as_the maespecially abundant in Haydn’s euvre).

So much is evident, but the only alternative terial encourages. These heuristic norms need to throwing up one’s hands in the face of such not be considered as literally existing “things. diversity (rallying around the cry, “Anything can nates Mine may oe ee eestood aS what Pane

happen!,” which 1s obviously untrue) was to find alts, TO Share op eer, a. ees t 1.

a reasonable middle ground between confiningly tyP ©s OF W at we PFener to CONSTCCE as Teg tia

_ tive guides for interpretation. Moreover, these

rigid schemata and the claim of a near-total free- dh he defined neither b

dom. It was necessary to retrieve a workable her- MOF , wou he fo be “ - A t CP DY _ meneutic space between the reductive textbook sta ond “a, hos ne Rathe her coud te

models of the nineteenth and early twentieth MOFE standard Choices at cr, they wou + Oe centuries and the unhelpful (though still fash- rive from the standard choices themselves, insoionable) “lowest-common-denominator” har- far as the frequency of those choices (not their monic models, whose claims to adequacy have inevitability) permit one, inductively, to infer been challenged on both historical and concep- ‘ background set of sure shared by Com tual grounds in an important essay from 1991 by posers ane a community OF “asteners at a given Mark Evan Bonds.!8 In that essay Bonds distin- historical time and place. As we constructed

; « ; 55between « these models, then, we were concerned to idencuished “conformational” and “genera-

tive’ concepts of sonata form, traced the fortunes tity types or tendencies that (in retrospect) were of these concepts historically, and submitted the influential generic participants in the eventual mid-twentieth-century ascendancy of the gen- crystallization or early reification phase of the

erative models to a critique. Among his conclu- sonata in the mid-eighteenth century, when

sions: the preferred optionssomewhat became both clearer and more consistent.?9 The result was the

m th ll Th ,

Few analyses [today] openly acknowledge the syste t at we ca Sonata Theory oo extent to which composers worked within the Our intention 1s not to lay down binding laws context of formal conventions... . But it would or invariant rules concerning either the parts of be ludicrous to argue that sonata form was not a sonata or the sonata as a whole. Instead, we at least in part an a priori schema available to the are trying to sketch the outlines of a complex composer. .. . Sonata form, for Haydn, was in set of common options or generic defaults. It fact a point of departure, a mold, albeit a flex- _is not that any attempt to recover standard pat-

18. Bonds,” The Paradox of Musical Form,” ch. 1 of related genres, works for chamber ensemble, and solo Wordless Rhetoric: Musical Form and the Metaphor of the and accompanied sonatas in all but a few major cenOration (Cambridge, Mass.: Harvard University Press, ters.” Similar observations regarding the increasing

1991), pp. 13-52. normativity of certain kinds of sonata procedures— es19. Bonds, “The Paradox of Musical Form,” p. 29. pecially those identified with the Viennese Classicism 20. E.g., as articulated in Wolf, “Sonata Form,” of Haydn, Mozart, and early Beethoven in the period The New Harvard Dictionary of Music, p. 766: “By circa 1770—1800— may be found in the writings of virabout 1765, however, full sonata form [1.e., with full tually every author who has investigated such things. recapitulation] —though never the rigid textbook va- See, e.g., the many similar remarks in Charles Rosen, riety — was rapidly becoming the norm in fast move- Sonata Forms, rev. ed., pp. 145, 153, 156-58, 161, and ments and many slow movements of symphonies and 286-87.

Contexts 9 terns is a flawed enterprise; rather, it is that prior lowed: symphony movement? overture? sonata? attempts have been inadequately conceived. We chamber music? how long or “grand” a moveoffer Sonata Theory as a heuristic construct that ment? how complex? how “original”? how “‘incan help the task of analysis and hermeneutics. tense” or “challenging” to listeners? what is the At any point, the method outlined here can be expected audience? for connoisseurs or amateurs expanded or modified through criticism, cor- (Kenner or Liebhaber)? how “unusual” in its inrection, or nuance. Indeed, we invite this. The ternal language and manner of presentation? in proposed construct 1s intended only as a begin- competition with whom? whom am I trying to ning, as a work-in-progress — not as a fixed set impress? for what occasion? and so on.

of finalized dicta. As an assemblage of separate Once these gateways had been determined subparts, each of which should be subjected to and work begun in earnest—the task of creconstant testing and refinement, the utility of | ating an engaging musical pathway through Sonata Theory as a whole does not rest on the pre-established, generically obligatory staunexceptionable validity of any correctible sub- tions—the composer faced practical issues of

part. musical continuation from one idea to its successor. (A succeeding phrase, even an utterly contrasting phrase, would typically be heard

Sonata Theory: Introductory Remarks as “reacting to” what had been established up to that point— moving outward to another What follows lays the groundwork of a method branch of the musical ramification.) A sonata of approaching analytically any sonata-form — form required that certain audible goals be sucmovement from the period of Haydn, Mozart, cessively articulated and secured, even though and Beethoven. A central premise of this method the individual details of each sonata journey is the conviction that we must seek to under- could differ remarkably. A composer’s choices stand the backdrop of normative procedures involved not only varying senses of the propriwithin the different zones or action-spaces of ety of “what sorts of things could reasonably be the late-ecighteenth-century sonata. Much of this expected next” within the style but also how book sketches out key technical features of those delectable surprises, even varying degrees of

norms as we currently understand them. seeming transgressions, might be folded into At any given point in the construction ofa —_ the expanding network of ideas. Within each sonata form, a composer was faced with an ar- compositional zone (action-space) or subsecray of common types of continuation-choices tion these “internalized” features included such established by the limits of “expected” architec- things as generically appropriate types of themes ture found in (and generalized from) numerous and textures; reasonable lengths of individual generic precedents. (To produce a keyboard-so- passages (which depended on the anticipated nata or symphonic movement was to place one’s length and complexity of the whole composiindividual achievement into a dialogue with a tion); dynamics; degrees of anticipated contrast; community-shared pool of preexisting works, standard “topics” or thematic formulas; propprobably including some well-known ones, that erly placed cadences and/or cadential delay or formed the new work’s context of understand- frustration; the handling of major- and minoring.) This is not to say that any skilled composer mode coloration; boundaries of taste; and the soberly pondered these choices, one by one, in limits of eccentricity. the act of composing. Surely the most common The options available from compositional decisions were made efficiently, expertly, and zone to zone existed conceptually within the tacitly on the basis of norms that had been in- —— knowledgeable musical community as something

ternalized (rendered automatic) through expe- on the order of tasteful generic advice — enrience and familiarity with the style. Still, even abling and constraining guidelines (not inviobefore a sonata form was begun, a composer __lable rules) within the “sonata-game” — given might, consciously or not, confront an array of — by ashared knowledge of precedents. Moreover, initial questions acting as a filter for all that fol- the available suldelines for each moment (pri-

10 Elements of Sonata Theory

mary theme, transition, medial caesura, second- implication is that not to choose the first-level ary theme, and so on) were not accessible 1n an default would in most cases lead one to consider arbitrary, non-weighted fashion. Some choices what the second-level default was—the next were virtually obligatory; others less so, some- most obvious choice. If that, too, were rejected, times in discernible degrees. (For novice-com- then one was next invited to consider the thirdposers, one might wittily fantasize—provided _level default (if it existed), and so on. Or perhaps that the image 1s not taken too literally —some- at some point in this process a composer might thing on the order of an aggressively complex decide to do something unusual by rejecting all “wizard” help feature within a late-eighteenth- of the default choices altogether, in pursuit of a century musical computer application, prompt- deformation of that compositional moment.

ing the still-puzzled apprentice with a welter of As might be imagined, the whole system numerous, successive dialog boxes of general in- was highly complex, typically involving at any formation, tips, pre-selected weighted options, compositional point more than two default levand strong, generically normative suggestions as els of options. This is why it requires so much the act of composition proceeded. What would time—and space—to reconstruct the backhave been urged here were such things as the- sround system. But it is only through an unmatic-modular shape, style, effect, and format derstanding of what the main options were that appropriate to the relevant action-space mo- we can come to grips with the implications of a ment— not literal content, the burden of which composer’s choices from moment to moment.

was still placed on the composer.) In confronting any individual composition we Within the late-e1ghteenth-century style seek to determine which gestures in it were norsome of the options were much more frequently mative within the style, which were elaborate, elechosen: To suggest the strength and pre-estab- gant, or strained treatments of the culturally availlished hierarchical ordering of these options we able norms, and which were not normative at all. call the more normative procedures first- and sec- Sonata Theory starts from the premise that an inond-level defaults within the various zones.2! Most dividual composition is a musical utterance that is simply put, composers selected (or adapted) set (by the composer) into a dialogue with implied first-level options more frequently than second- norms. ‘This is an understanding of formal procelevel ones, and so on. (Writers of minor-mode dures as dynamic, dialogic. Our conception of the sosonatas, for instance, more often modulated to nata as an instance of dialogic form 1s not accurately the major mediant, III, in the exposition, than described as seeking to reinstate a bluntly “conforto the minor dominant, v—a less common op- mational” view of that structure (in Bonds’s origition.) As we use it, however, the term default nal sense of that category). Viewed more subtly, it

connotes more than a merely preferred option is not the obligation of a sonata to “conform” to for otherwise detached consideration. First-level a fixed background pattern, which then, in turn, defaults were almost reflexive choices—the might be construed as an “ideal” or “well formed” things that most composers might do as a mat- shape from which deviations might be regarded as ter of course, the first option that would nor- compositional errors or aesthetically undesirable

mally occur to them. More than that: not to distortions. Rather, the composer generates a soactivate a first-level-default option (for example, nata—which we regard as a process, a linear series to provide an expositional move to v instead of | of compositional choices—to enter into a dialogue to III) would require a more fully conscious de- with an intricate web of interrelated norms as an cision—the striving for an effect different from ongoing action in time. The acoustic surface of that provided by the usual choice. An additional any sonata form (what we literally hear) sets forth

21. At some level the literal, computer-definition con- sense of ongoing, strongly weighted advice, standard cept of default—an assumption prebuilt into the large- choices, and normatively arrayed options). As menscale automatic (but alterable) decisions of a software tioned earlier, the metaphorical implication, if appliprogram at the moment of its initialization—1s not fully cable at all, is to be worn loosely. congruent with our free adaptation of it here (in the

Contexts 11 the sonic traces of this individualized, processual of expectations, presumably in order to generate dialogue, one that, from the standpoint of recep- an enhanced or astonishing poetic effect.??

tion, it is the task of the analyst to reinvigorate. Deformations— unusual or strongly characThe backdrop of norms against which a sonata or terized, ad hoc moments—are common within any of its successive zones 1s placed into dialogue the works of many different late-eighteenthis no monodimensional, reified “thing.” On the century composers. Indeed, they are rampant contrary, that backdrop comprises complex sets (or = in Haydn, who delighted in producing surprisconstellations) of flexible action-options, devised ing effects. Such occurrences, in dialogue with a to facilitate the dialogue. Understanding form as norm, should not be regarded as redefining that dialogue also helps us to realize that in some cases norm unless the composer continued to employ standard procedures may be locally overridden for that idiosyncratic feature in other works (thus certain expressive effects. These eftects differ from customizing the norm for his own use) or uncomposition to composition: each needs to be 1n- less later composers picked up the deformation

terpreted individually. The more piece-specific as one of their more or less standard options. one’s readings can be along these lines, the better. When this later occurrence happens, the origiIn any analysis merely to assert that something 1s nal exception 1s no longer to be regarded as a dedone “for expressive reasons” or “for reasons of | formation per se but becomes one of the lower-

variety’ 1s obviously inadequate. level defaults within the Sonata-Theory system. Background norms and standard options are W hat was a deformation in Beethoven could beclassifiable into common and less common se- come a lower-level default in Schumann, Liszt, lections at different times and different places. or Wagner— part of a larger network of nine-

Within an individual composition, a mark- teenth-century sonata-deformation families. edly exceptional procedure here or there 1s just The essence of Sonata Theory lies in uncovthat— exceptional. We call such an occurrence a ering and interpreting the dialogue of an indigeneric deformation: a stretching or distortion of | vidual piece with the background set of norms. a norm beyond its understood limits; a pointed This style of analysis considers every aspect of overriding of a standard option. The term “de- the individual work: themes, harmonic and conformation,” in this specific context, is a narrow- trapuntal motion, large- and small-scale shapes, definitional, technical one, grounded in prece- textures, dynamics, instrumentation, tempos, dents in literary theory and other research areas. repeat conventions, and so on. The main reIn its strictly limited, analytical usage within So- quirement for the application of the method 1s nata Theory, “deformation” carries no negative to grasp the controlled flexibility of the implicit charge, no negative assessment. On the contrary, underlying system of conventions. Elaborating such deformations are typically engaging, aes- that system is the goal of the Elements.

thetically positive occurrences that contribute At every turn, our aim has been to focus on to the appeal and interest of a piece. As we use the most basic features of the sonata and never the term, it signifies only a purposely strained or to forget why we perform and listen to this munon-normative realization of a musical action- sic in the first place. To overlook fundamental space, a surprising or innovative departure from things leads one’s analyses astray or renders them the constellation of habitual practices, an imagi- sterile, bookish, or irrelevant. The best analytinative teasing or thwarting, sometimes playful, cal system is the one that seeks to reawaken or 22. It would be a mistake, therefore, to read into this “deformation” —to be able to perceive in it a genre-entechnical usage any residual connotations of the evalu- abled, positive sense of strain, a deliberately manufactured

atively negative, such as the “deformed” (in its more tension set apart in this aesthetic-analytical, “artificial” typical meaning), the “disfigured,” the “misshapen,” the context from any implication of criticism or (much less) “abnormal,” the “poorly formed,” or the “ugly.” Those censure. These connotational points are revisited and are not our connotations, and within the framework of amplified in the “Deformation” section of appendix 2, Sonata Theory terminology we distance ourselves from which also offers further reflections on the concepts of them as strongly as we can. The central thing is to be dialogic form and sonata-form action-spaces. able to grasp the intended nuance of the technical term

12 Elements of Sonata Theory

re-energize the latent drama, power, wit, and rently “orthodox wisdom” regarding sonatas to wonder within individual compositions. When- radical questioning— comfortable trenches of ever an analytical system diverts attention from thought that had long been part of our own rethe impact of the music as real experience — or, flexive modes of approaching this music. From even more, when it fails to heighten our own the beginning we sought to listen carefully to experience of the music—then that analytical this repertory, trying to remain open to what it system is in need of correction. We hope that seemed to want to tell us on its own terms, inSonata Theory, in its practical application, will sofar as we could apprehend those terms in our lead beyond the academic explanations and in- own, very different times. Before long we came terpretations of the self-enclosed work into a to understand that everything that we had conlarger reflection on the changing meanings of sidered to be established about sonata-analysis

this music within society. had to be rethought. If only for this reason, we In part, we do this by redirecting analyti- realize how curious Sonata Theory might at first cal attention to those portions of the sonata that appear, especially to scholars habituated within have been taken for granted or passed over in other modes of analysis and accustomed to other relative silence in most preceding discussions. kinds of theoretical questions. The value of any These include the composer’s treatment of cae- analytical system, however, lies in the robust-

suras (medial and final), the textural drive to- ness of its interpretive power. It is that interward important cadences (including especially pretive adequacy that we have been seeking. the moments of what we call essential exposi- Whenever existing terminology was adequate, tional closure [EEC] and essential structural closure we have retained it; whenever it was misleading [ESC]), the rotational aspect of the sonata move- or connotatively unhelpful, we have decided to ment as a whole (its tendency to cycle repeat- change it; whenever it lacked a term for a cruedly through large, thematically differentiated cial concept, we have been obliged to devise a

blocks), and many other considerations. Al- new one. though this was by no means clear to us when Readers might initially find that the basic we began this project, one result of our work concerns of Sonata Theory are learned relatively has been to defamiliarize the sonatas of Haydn, quickly —like the moves of chess. These conMozart, and Beethoven — permitting us to hear cerns may seem simple precisely because they them in what we have found to be more re- are simple. At all points in the analysis of a sowarding ways. To some extent, we discovered nata, we have tried to emphasize the most esearly on that we often had to overcome our own sential features and dramatized musical goals. patterns of habituation in analysis and under- Beyond the elementary principles of Sonata standing “in order [to adapt the words of Viktor Theory, though, les an elaborate network of Shklovsky] to return sensation to our limbs, in possibility, nuance, flexibility, sophistication, order to make us feel objects, to make a stone and detail that takes patience to master. As with feel stony.”?5 The idiosyncratic concerns — even chess, again, one may learn the moves rapidly, the idiosyncratic terms— of Sonata Theory can but to play the game at a fully proficient level 1s

help in this regard. more difficult. Notwithstanding its many pos-

For the authors, one of the most challenging tulates and axioms, Sonata Theory is no meburdens in devising Sonata Theory has been to chanical system. Rather, in proper application remain willing to submit all components of cur- it is an art that requires training, musical sensi-

23. Shklovsky, Theory of Prose [from second edition, of art makes perception long and ‘laborious’... . Art is a 1929], transl. Benjamin Sher (Elmwood Park, II1.: means of experiencing the process of creativity. The artifact itself

Dalkey Archive Press, 1990), p. 6. In order to ac- is quite unimportant.” Sher defends his translation, “encomplish these things, declared the Russian Formalist stranging” (as opposed to the more traditional choices, Shklovksy, “[we have] been given the tool of art... . By “defamilarizing” and “estranging’’), on p. xix. ‘enstranging’ objects and complicating form, the device

Contexts 13

tivity, and much experience with the repertory The goal and the course to the goal are pri-

in question. mary. Content comes afterward: without a goal At the heart of the theory is the recognition there can be no content, .

and interpretation of expressive/dramatic trajecto- In the art of music, as in life, motion toward

,; ,;toward the goal encounters obstacles, reverses, ries generically obligatory cadences. For the . disappointments, and involves great distances, detours,

present, we might only register the degree to . . . hich th; ‘th Heinrich expansions, interpolations, and, in short, retardawhich ¢ i CONCCEN resonates wi CEN TTC tions of all kinds. Therein lies the source of all Schenker’s much-quoted description of musical artistic delaying, from which the creative mind motion and dramatized process in Free Composi- can derive content that is ever new.24 tion (Der freie Satz, 1935):

24. Schenker, Free Composition, p. 5.

CHAPTER TWO © "ISLE 0

Sonata Form as a Whole Foundational Considerations

Sn"ture form the most movements important large of the symphony [or sonata]” disposed of is individual fromstructhe inallegro “two sections” [zwey Theile| and three “main “common-practice” tonal era. It sets forth and periods” [Hauptperioden| (Koch 1793); within resolves its musical discourse within a large-scale “larger pieces of music” a “well-conducted mel-

binary format. The term “sonata form” was al- ody [!] . . . divided into two parts, either conmost surely unknown to Haydn, Mozart, early nected, or separated in the middle by a repeat Beethoven, and their contemporaries: it seems sign” (Galeazzi 1796); “an elaborate movement to have surfaced only in the 1820s and 1830s. In [or] along movement . . . generally divided into the late-cighteenth and early-nineteenth century two sections’ (Kollmann 1799); “grand binary this structure would have been grasped primar- form” [grande coupe binaire] (Reicha 1826); and ily as the customary design of first movements so on.! Still, “sonata form” (Sonatenform) seems within sonatas, chamber music, and symphonies, to have been a familiar term by the mid-1820s, although it was by no means confined only to at least in A. B. Marx’s Berliner allgemeine musikafirst movements (nor only to rapid-tempo move- lische Zeitung circle, where it referred both to the ments). The varying descriptions from contem- multimovement cycle as a whole and, occasionporary theorists were more convoluted. There ally, to the form of an individual movement.? It the form was variously described as: “the first was only in 1838 and 1845, though, 1n technical

1. Heinrich Christoph Koch, Versuch einer Anleitung zur on Practical Musical Composition (London, 1799; rpt. New

Composition (Leipzig: Adam Friedrich Béhme, 1793; York: Da Capo Press, 1973), p. 4 [ch. 1, section 10]; rpt., Hildesheim: Georg Olms, 1969), pp. 304—5 (from Anton Reicha, Traité de haute composition musicale (Paris,

section 101), trans. Nancy Kovaleff Baker in Koch, 1826), discussed, e.g., in Ian Bent and William Drabkin, Introductory Essay on Composition: The Mechanical Rules Analysis (New York: Norton, 1987), pp. 18—20, and es-

of Melody, Sections 3 and 4 (New Haven, Conn.: Yale pecially Peter A. Hoyt, “The Concept of développement University Press, 1983), p. 199; Francesco Galeazzi, El- in the Early Nineteenth Century,” in Music Theory in ementi teorico-pratici di musica, vol. 2 (Rome: Puccinelli, the Age of Romanticism, ed. lan Bent (Cambridge: Cam-

1796), the relevant extracts of which were excerpted bridge University Press, 1996), pp. 141-62. and translated in Bathia Churgin, “Francesco Galeazzi’s 2. In the journal’s first year of publication (1824) the Description (1796) of Sonata Form,” Journal of the Ameri- term ‘sonata form” appeared in both senses. The first, can Musicological Society 21 (1968), 181—99 (above quota- apparently initially the more common, was a descrip-

tions from pp. 189-90); A. F. C. Kollmann, An Essay tion of the entire multimovement cycle (used by Marx, 14

Sonata Form asa Whole 15 discussions of the form’s particulars, that Marx ized action involving differing types of idealized put the stamp of approval on the term “Sonaten- mud- and late-eighteenth-century personalities. form” with regard to the individual-movement (Its potential for “extramusical” connotations structure. Throughout this book we use that and analogues is discussed in the final section of term as a familiar quick-reference, even as we chapter 11, “Narrative Implications: The Sonata realize that that designation was not current in as Metaphor for Human Action.”) Sonata form

the eighteenth century. emphasized short-range topical flexibility, grace, Sonata form is neither a set of “textbook” and forward-driving dynamism combined—1in rules nor a fixed scheme. Rather, it is a con- both the short and long range—with balance, stellation of normative and optional procedures symmetry, closure, and the rational resolution of that are flexible in their realization—a field of — tensions. By the mid-eighteenth century it had enabling and constraining guidelines applied in become obligatory for the first movement of a the production and interpretation of a famil- standard multimovement instrumental work; it iar compositional shape. Existing at any given had also become a common, if optional, choice moment, synchronically, as a mappable con- for the slow movement and the finale. Slow stellation (although displaying variants from = movements and finales sometimes also displayed one location to another, from one composer to different adaptations of the form. Although the another), the genre was subjected to ongoing suidelines in most of this book were written diachronic transformation in history, changing predominantly with first and last movements via incremental nuances from decade to decade. and single-movement overtures in mind (all enHaydn’s conception of what was customary ergetic “Allegro movements”’), they are also apwithin sonata form in 1770 differed somewhat plicable, occasionally with some modifications, from Beethoven’s conception in 1805. However to slow movements. such models might be said to have differed, they From the compositional point of view sonata also shared certain crucial, genre-defining fea- form was an ordered system of generically availtures that make them all recognizable as sonata able options permitting the spanning of ever form. Here we are dealing primarily with the larger expanses of time. A sonata-form project model that crystallized during the second halfof = was a feat of engineering, like the constructhe eighteenth century and that reached a peak tion of a bridge “thrown out” into space. In in the mature works of Haydn and Mozart and the eighteenth-century style this temporal span

the early works of Beethoven. was to be built from rather simple materials: W hat we now call sonata form was developed trim, elementary musical modules whose brevas a response to aspects of the world view of the ity and small-scale balances seemed best suited Enlightenment and the concomitantly emerg- to short-winded compositions. In the hands of ing modernism. Considered generally, it could most composers, constructing a sonata-form be understood as an abstract metaphor for disci- movement was a task of modular assembly: the plined, balanced action in the world, a general- forging of a succession of short, section-specific

Heinrich Joseph Birnbach, and others), a usage that per- kurzes Allegro (A-moll) in der Sonatenform,” Bam Z, sisted throughout much of the nineteenth century, es- 1824, 410b). See the discussion of terminology and quopecially in German-speaking regions. The other use of tation of sources in the entry by Hans-Joachim Hinrich“Sonatenform” referred to the structure of an individ- sen, “Sonatenform, Sonatenhauptsatzform” [1996], in ual movement. It first appeared in a casual, unexplained Hans Heinrich Eggebrecht, ed., Handwérterbuch der musiway —as if it were already a common label—in Marx’s kalischen Terminologie (Stuttgart: Steiner, n.d.), pp. 1-7. 1824 essay on the E-minor second movement (Prestis- 3. A. B. Marx, Die Lehre von der musikalischen Komposisimo) of Beethoven’s Piano Sonata in E, op. 109 (“Es tion, praktisch-theoretisch, vols. 2 and 3, 1st eds. (Leipzig,

bildet mit dem letzen Satze die eigentliche Sonate und 1838 and 1845), 2:482, 497; 3:195; cited in Hinrichsen, ist auch in der Sonaten-Form hingeworfen,” BamZ, I, “Sonatenform, Sonatenhauptsatzform” [1996], pp. 6—7. 1824, 37b) and in Carl Loewe’s discussion of the first See also Marx, Musical Form in the Age of Beethoven: Selected

movement of Beethoven’s Cello Sonata, op. 102 no.1 Writings on Theory and Method, ed. and trans. Scott Burn-

(“Hart und rauh, im minnlichen Zorne, beginnt ein ham (Cambridge: Cambridge University Press, 1997).

16 Elements of Sonata Theory

musical units (spaces of action) linked together priate for their specified location. These three into an ongoing linear chain— pressing down spaces can be viewed as expansions of the three and connecting one appropriately stylized mu- phases of the continuous rounded binary form sical tile after another.* One of the challenges (the rounded binary structure in which the first

facing the mid- and late-eighteenth-century part ends in a secondary key). We shall take composer was to use a seemingly unassuming, up these spaces individually. (In figure 2.1 we galant language, grounded in structural punc- have provided two diagrams of Sonata Theory’s tuation and periodicity, to produce ever more conception of the most common type of sonata spectacular spans for occasions of enhanced dig- form: 2.1a refers to the exposition; 2.1b to the nity, prestige, or social importance. Ever-larger, whole sonata-form movement.) thematically differentiated binary structures (sonata forms, often with built-in repetitions of

individual sections), eventual accretions to the Exposition structure (slow introductions and longer codas), and multimovement conventions all had their — As with all of the action-spaces the exposition roles to play in this process of generic enlarge- is assigned a double-task, one harmonic and the

ment. And ultimately they led to the grandly other thematic-textural (“rhetorical”). Its harmonumental, personalized structures of Haydn, monic task 1s to propose the initial tonic and then,

Mozart, and Beethoven. following any number of normative (and drama-

The most typical sonata forms (what we call tized) textural paths, to move to and cadence in Type 3 sonatas) articulate an overall rounded =a secondary key. In major-mode sonatas— the binary structure. The two parts of this larger | most common in the eighteenth century— this structure are, in modern terminology: (1) the | was the key of the dominant (which may be inexposition and (2) the development and recapit- dicated as V;, meaning “a V that is tonicized”), ulation. As will be elaborated at the end of this _ thereby generating tonal tension. In minor-mode chapter, both parts may be marked for repeat, or sonatas this was usually the key of the major methe composer may eliminate the repeat of part diant (IID), although a less-often-selected choice 2 or, under some circumstances, both repeats. (second-level default) was the minor dominant Notwithstanding its binary origins, the norma- (v). The differing psychological and structural tive, Type 3 sonata consists of three musical ac-_ | world of minor-mode sonatas is dealt with in tion-spaces (again, the exposition, development, chapter 14. Here, for the most part, we shall foand recapitulation), laid out ina large AI/BA' — cus on major-mode practice. format. Hence the common observation that The exposition’s rhetorical task, no less 1mporthe form consists of an originally binary struc- tant, is to provide a referential arrangement or ture often arrayed in a ternary plan. Each of the layout of specialized themes and textures against three spaces 1s usually subjected to thematic and which the events of the two subsequent spactextural differentiation. Each is marked by sev- es—development and recapitulation—are to eral successive themes and textures, all of which be measured and understood. We refer to this are normally recognizable as generically appro- layout as Rotation 1 or the expositional rotation.° 4. To be sure—and particularly in the hands of the mas- 11, subsection “Recompositions, Reorderings, Interpoter composers of the period—certain passages within lations” (especially n. 2 and the text to which it refers), individual sonata forms may from time to time give the and ch. 18, subsection “Haydn’s Treatments of Type 4 impression of a broader continuity of internal ramifi- Finales” (especially n. 49 and related text). cation. This is especially the case with the startlingly 5. Sonata-form structures are centrally concerned with original musical language of Haydn, who, even within the formal principle that we call rotational form or the a generally modular and “sectionalized” concept of for- rotational process: two or more (varied) cyclings— rotamal practice, often favored passages of ongoing Fortspin- tions —through a modular pattern or succession laid nung (a moment-to-moment “spinning-out” of modular down at the outset of the structure. Appendix 2 progrowth and elaboration). For brief characterizations of vides a broader introduction to this principle, which Haydn’s often-“vitalistic” compositional style, see ch. pervades the discussion of sonata form in this book.

a PAC final

a. Exposition only: the Essential Expositional Trajectory (to the EEC)

a MC EEC cadence

o Continuation modules wn my

Pd a i (series of energy-gaining modules) | a

f Energy-gain + A Relaunch Postcadential Appendix or

Acceptance of P set of “accessory ideas” fAi /Often New key May be multisectional (C!, C2, etc.) forte Usually piano and of varying lengths. Usually forte

Launch ~

P Either modulatory or nonmodulato

; or gaining in rhetorical force.

Often lyrical, etc.

Proposes the main idea for the sonata

in V (or, if P was in minor, in III or in v)

Tonic key Nontonic key

Exposition, Part 1 Exposition, Part 2

b. The entire structure: the Essential Sonata Trajectory (to the ESC)

ee me final Ms

o MC 4 cadence interruption s

:i Often “ “a fDevelopment TR S C o MC cadence P- or TR-dominated ra 5

i Perhaps rotational i per |

ee | I V or Il as chord | I I Exposition Development Recapitulation Restart Tonal resolution Often recomposed

Vv (emph. IV?)

One central mission: laying out the S, as agent, carries out the central strategy for the eventual attamment of generic task of the sonata—securing the ESC: a structure of promise. the ESC: a structure of accomplishment. FiGurRE 2.1. The Generic Layout of Sonata Form

18 Elements of Sonata Theory

Because the exposition’s succession of events closure at the EEC. In performing or listening serves, especially in its second half, to pre- to any sonata-form exposition one should sense dict the plan and purpose of the entire third the broad drive of these generic vectors. Whenspace —the recapitulation, which finally re- ever one hears the onset of S-space within any solves the work —its layout may be understood exposition, one should listen with an alert sense as articulating a structure of promise (indicating of anticipation for any subsequent PAC — how how it proposes that “things work out” in the it might be approached, secured, delayed, recapitulatory rotation-to-come). Because the thwarted, or deferred. One should experience arrangement of rhetorical modules in Rotation any sonata form with a strongly “directed” pre1 provides the ordered set of events that articu- paratory set, pressing forward conceptually and lates the uniqueness and specific personality of — anticipating genre-defining events-to-come. that piece, it should be kept in mind when as- Following the EEC one or more additional sessing all of the later events in the movement. cadences (PACs) may follow within the closing Within the expositional rotation the tonal zone or Closing space (C). (Not all expositions conand rhetorical tasks unfold simultaneously, in- tain C-modules; it is possible for the S-concludtertwined with each other in mutually reinforc- ing EEC to be delayed until the end of the exing ways. The exposition begins with a primary position, in which case there is no closing zone.) theme or primary idea (P) in the tonic that sets the Whether or not C-modules are present, the final

emotional tone of the whole work. The most cadence of the exposition will generally be a common layout for the remainder of the exposi- perfect authentic cadence in the secondary key tion continues with an energy-gaining zone of —_ (again, V:PAC, III:PAC, or v:PAC). This final transition (TR) that leads to a mid-expositional cadence might not occur directly at the double break or medial caesura (MC). This 1s typically bar. Frequently the final cadence is followed by a followed by the onset of a specialized, second- C-module that prolongs the newly reinforced ary-theme zone (S) in the new key. The generi- tonality by means of a pedal-point or some other cally essential tonal purpose of the exposition device. Additionally, the final cadence is someis to drive to and produce a secure perfect au- times followed by a reactivation of V in prepathentic cadence (PAC) in the new key (notated ration for a repeat of the entire exposition: if so, as V:PAC in major-mode sonatas, III:PAC or this reactivating passage 1s the retransition (RT). v:PAC in minor-mode ones). We refer to the first satisfactory PAC within the secondary key

that goes on to differing material as the point Development of essential expositional closure (the EEC): this 1s

one of the central concepts of Sonata Theory This action-space renders the established tonal and one that is dealt with at length in other tension more fluid and complex. While the exchapters.° Producing the EEC is the generically position had split its tonal assertions into two assigned task of the S-idea(s). The large dot- broad blocks or contrasting planes (I and V in ted-line arrow in figure 2.1a suggests a broadly |= major-mode sonatas), the development typically vectored trajectory from the start of the exposi- initiates more active, restless, or frequent tonal tion to the EEC; the smaller dotted-line arrow — shifts—a sense of comparative tonal instabilbelow it suggests a subordinate trajectory from ity. Here one gets the impression of a series of the beginning of S to its own point of PAC- changing, coloristic moods or tonal adventures,

6. For the moment, we might emphasize that the first ahead. Additionally, there are other ways of deferring satisfactory PAC in the new key 1s often but not always the sense of a clear EEC (ch. 8). The clearest way of sugthe first PAC in that key. A first PAC, for instance, might gesting all of this in brief is to define the EEC as the first be followed by a thematic repetition of all or part of the new-key PAC that proceeds onward to differing or contrasting S-idea that we have just heard— which would automat- material—or, of course, that closes the exposition itself, ically defer the EEC to the next satisfactory PAC further if there are no closing modules that follow that PAC.

Sonata Form asa Whole 19

often led (in major-mode works) through the ordered thematic pattern established in the exsubmediant key, vi, or other minor-mode keys position). Developments often refer back to (or with shadowed, melancholy, or anxious con- take up as topics) one or more of the ideas from notations. Any authentic-cadence attainment the exposition, most commonly selected, as in a non-tonic key is to be understood as an it happens, from Rotation 1’s first half (P and important developmental event—a cadential TR). More often than not, the modules taken ratification of an attained tonal station. (A v1: up and worked through in the development are PAC 1s especially common in major-mode so- presented in the order that they had originally natas.) Ultimately, the standard development appeared in the exposition (even though sevculminates on an active dominant (V,, mean- eral expositional modules are normally left out ing “a V that is an active chord, not a key”). entirely). Thus the modular succession encounAt this point the dominant from the end of the tered in the development— not only the exposi(major-mode) exposition is usually recaptured, tional events referred to, but also the possibility

detonicized, and reactivated. of an episode or largely new theme —is never to This last point needs underscoring. In the de- be considered arbitrary. On the contrary, even velopment the final cadence 1s usually a half-ca- within this more unpredictable, developmendence in the tonic (I: HC), although a cadence 1n tal texture the thematic choice and arrangea related minor key, normally followed by a brief ment is of paramount importance and derives reactivation of V, 1s also a possibility. In addition, its significance through a comparison with what a I:HC is frequently followed by a prolonga- had happened in the exposition. The develoption of dominant harmony, a “dominant-lock” ment 1s variable in length, although in the peor “dominant preparation.” The typical I: HC riod 1760—90 one would normally expect it to conclusion of the development — just before the occupy a smaller space than that established by onset of the recapitulation— brings us to a har- the exposition. Longer, more elaborate developmonic interruption. (This crucial interruption is a ments in the 1780s, 1790s, and later decades are defining feature of the Schenkerian conception monumentalized statements that invite special of sonata form.) The V, at the end of the devel- attention. opment is not resolved to the I that usually begins the recapitulation. Rather, the phrase —and

the development section as a whole —is normally Recapitulation “interrupted” on V, (notwithstanding any foregrounded or local, connective “fill” that might This action-space resolves the tonal tension origbridge the end of the development to the reca- inally generated in the exposition by rebeginpitulation), and the next cycle of events 1s newly ning on the tonic (with the initial theme in the launched with the opening of the recapitulation. | most common Sonata Types, 1, 3, 4, and 5) and True, this more fundamental interruption on usually by restating all of the non-tonic modules the dominant may sometimes be masked on the from part 2 of the exposition (S and C mateforeground with an apparent V—-I cadence (with rial) in the tonic key. For this reason—its largely the I triggering the recapitulation). But the more referential retracing of the rhetorical materials fundamental or background concept is that of laid out in the exposition (Rotation 1)—we also harmonic interruption on V,. (Those unfamiliar call the recapitulation the recapitulatory rotation. with the Schenkerian, linear-contrapuntal view (Exceptions and reorderings of thematic mateof things might notice that this interruption di- _—_ rial may be found in some sonatas.) Because of vides the entire sonata form at the end of the de- its function in bringing tonal closure to the envelopment. This contrasts with the eighteenth- tire form, we refer to the S/C complex in the century “binary” division of sonata form at the recapitulation as the tonal resolution. Its shape and

end of the exposition.) manner of unfolding had been established by the In terms of their rhetorical strategies, devel- —_ exposition’s structure of promise. Correspondopments may or may not be fully or partially ingly, we consider the recapitulation to articulate rotational (that is, guided in large part by the a structure of accomplishment. Minor-mode sonatas

20 Elements of Sonata Theory

that had sounded S and C in the major mediant exposition :||: development—recapitulation : 11). (III) in the exposition have the additional option This is the most formal and earliest norm. Many

of sounding them in either the major or minor late-century first movements, especially those

mode in the recapitulation. after about 1760, repeat only the first part (the The recapitulation’s S, launching the tonal exposition), although in works prior to 1790 one resolution following a recapitulatory MC, leads need not be surprised to see the second part also to the production of a satisfactory I:PAC that repeated. After that date, repeating the second goes on to differing, non-S material. This is the part is an uncommon gesture that invites analytmoment of essential structural closure (the ESC), ical interpretation. It is also possible to find both most often a point parallel to the exposition’s parts unrepeated. This occurs in lighter works, EEC. The ESC represents the tonal goal of the in some midcentury symphonies (some Stamitz entire sonata form, the tonal and cadential point symphonies from the 1750s; some early Mozart toward which the trajectory of the whole move- symphonies; and so on) and in some slow movement had been driving: this is suggested by the ments (especially those in the format of the less longest dotted-line arrow in figure 2—1b. From expansive, Type 1 sonata, lacking a developthe perspective of Sonata Theory, it is only here ment). The nonrepeated exposition is also a gewhere the movement’s tonic is fully called forth, neric feature of the overture or sinfonia. (In other stabilized as a reality as opposed to a mere po- words, expositional repeats will not appear in e1-

tential. As in the exposition, C-material will ther operatic or concert overtures; this is also follow, now in the tonic. The recapitulation’s true of the overture’s mid-nineteenth-century final cadence 1s generally a I:PAC (or, in minor, offspring, the symphonic poem).’ In this aspect

sometimes a1:PAC), although this too may be the lighter overture is to be distinguished gefollowed by a prolongation of tonic harmony or nerically from the more formal first movement by a transition leading either back to a repeat of a sonata or grand symphony, which at least (of the entire development and recapitulation) had available the common option of expositional

or forward into the coda. repetition. Nonrepeated expositions within first A coda (outside of sonata space) may or may movements do sometimes occur in more broadly not follow the recapitulation. More information scaled and ambitious works after 1780, but when about codas, along with a discussion of the other they do—as in Mozart’s Symphony No. 35 in optional or parageneric feature of some sonatas, D, K. 385, “Haffner,”’® or in Beethoven’s Vio-

the introduction, may be found in chapter 13. lin Sonata in C Minor, op. 30 no. 2, his Piano Sonata in F Minor, op. 57, “Appassionata,” and his String Quartet in F, op. 59 no. 1— they are

Repetition Schemes exceptional and need to be considered as consciously expressive choices.’

Within eighteenth-century sonatas and sym- One curious (and rare) possibility is that phonies one may find both parts repeated (Il: of literally writing out an expositional repeat, 7. Thus the rule. Exceptions are extremely rare and dis- those of the Serenades in D, K. 320, “Posthorn,” and in concertingly puzzling, such as the repeat of the exposi- E-flat, K. 375. Such examples—perhaps related to eartion in young Mozart’s Overture to Apollo et Hyacinthus, K. lier or existing concepts of repeat-convention options 38 (1767), labeled as the “Prologus/Intrada” to the opera. in overture-symphonies, in smaller-scale symphonies, This piece 1s a Type 2 sonata (Chapter 17) whose first rota- or in some serenades—require individual attention. tion (exposition) is provided with a repeat sign. Much later, Within the larger symphony it may be that during the the odd “expositional” (?) written-out and slightly varied 1770s (though not, it seems, in the 1780s) Mozart was repetition in Berlioz’s Overture, Le carnaval romain 1s also exploring the possibility of the omission of the exposicurlous, suggesting that the form of this unusual piece 1s tional repeat as a lower-level default. more purely rotational (or perhaps instrumental-strophic 9. The solution of Beethoven’s op. 59 no. 1/1, which with fortissimo refrain) than a sonata per se, although it 1s initially suggests an expositional repeat only to abort it also manifestly in dialogue with certain sonata norms. almost immediately in favor of development, 1s antici8. Other examples within Mozart’s major works in- pated in the first movements of Mozart’s Serenade in Eclude the first movements of his Symphonies Nos. 31 flat for Eight Winds, K. 375, and Haydn’s Piano Sonata

in D, K. 297, “Paris,” and 34 in C, K. 338, along with in D, Hob. XVI:51.

Sonata Form asa Whole 21

normally including variants the second time Control, balance, generic identification, and foraround. This occurred most famously in C. P. E. mal architectural splendor: these would appear Bach’s unusual set of six keyboard Sonaten mit to be the central reasons why literal repetition verdnderten Reprisen, H. 136—39, 126, 140 (W. played such a prominent role in the style.

50/1-—6, Sonatas with Varied Repetitions), Consequently, repeat signs should not be composed in 1758—59 and published in Berlin taken for granted, passed over lightly in analythe following year. In Haydn’s works the proce- sis, or omitted in performance. Repeat signs are dure surfaces only (and wisely, in ‘Tovey’s view) never insignificant.!! Block-repetitions are an in a few “purely lyric slow movements,” such as integral component of the style, and composthe Adagios of the Quartet in C, op. 33 no. 3, ers can work with this defining convention in a “Bird,” and the Symphony No. 102 in B-flat.!9 variety of ways. When previously obligatory (or (Both slow movements are in F major; in the exceptionally strong first-level default) exposiquartet the Adagio is the third movement; in tional repeats began gradually to disappear— es-

the symphony it is the second.) pecially in the early nineteenth century, with What are the purposes of large-scale repeats certain works of Beethoven (op. 30 no. 2, op. within sonata form? Central to the concept of — 57, op. 59 no. 1, and so on, and later with Menthe grand sonata or symphony is a system of — delssohn, Schumann, and others)—the genre

schematic repeat-conventions, balances, sym- itself was undergoing a major rethinking.!? metries and proportions that call attention to The familiar, current views—Schenkerian and and help to define the genre. The emphatically otherwise —that propose that some repeats are architectural construction calls attention to the structurally insignificant while others are more genre’s ordered formality —and 1n the case of the important (because of the unfolding of certain erand symphony, also to its grandeur and public structural tones or other significant events, persplendor. Repeats were an important feature of — haps under a first-ending sign) miss the larger

a sumptuous, high-prestige display of grand ar- point of repeat signs as generic identifiers.!5 chitecture, one to which large-scale repetitions Even when the structural-tone aspects might were essential —especially that of the exposi- be convincing (but, perhaps paradoxically, only tional repeat in the first movement. The styl- as local details), the gist of these claims seems ized form thus celebrated the “Enlightenment” to be based on later-nineteenth-century prem(or “modern”) culture that makes such an im- ises, Which came to look on all unaltered repetipressive, moving, or powerful art possible. One tion as an aesthetic error. Such a conviction also of the structure’s implications would have been came to affect performance in the omission of that this culture had devised a rational, balanced repeats or in the insistence on an altered intermeans to shape and contain the fluid, raw, el- pretation in the repeat. It may be, though, that emental power of music. By extension, the pro- saying the same thing twice was what the comcess probably also represented the controlling or poser had in mind. harnessing of those impulsive, instinctive, libidi- It is easy to object to our general argument nal, or “uncivilized” elements within ourselves. here. One could strive to minimize the impor10. Tovey, “Sonata Forms,” Musical Articles, p. 214: 12. Curiously, in 1826 Reicha suggested —in passing “Haydn saw that the only place for C. P. E. Bach’s de- and without explanation (Traité de haute composition muvice was in purely lyric slow movements. Even there he sicale, p. 300)—that finales may lack an explicit repeat: never had the patience to plod and pose (as C. P. E. Bach “When the first part 1s not repeated, as in overtures and did to the bitter end) through a repetition [recapitula- finales...” (“Quand la premi¢re partie n’a pas de reprise, tion] of both parts. When his second part comes to reca- comme dans les ouvertures ou dans les finales .. .”). pitulate the second group it combines both versions.” It may be that Reicha had sonata-rondos, Type 4 sona-

11. For the quintessential statement of that which the tas, in mind (ch. 18 ). present argument opposes, see Douglass M. Green, Form 13. Cf., e.g., Jonathan Dunsby, “The Formal Repeat,” in Tonal Music: An Introduction to Analysis, 2nd ed. (New Journal of the Royal Musical Association 112/2 (1987),

York: Holt, Rinehart, and Winston, 1979), p. 82: “HIS- 196-207. TORICAL NOTE [sic:] Ordinarily the repetition of a part is of little significance in formal analysis.”

22 Elements of Sonata Theory

tance of the usual repetition schemes by an ap- However we decide this matter, we should peal to history: deriving them step-by-step from note three things. First, the issue of notationally the earlier binary forms, then asserting that the indicating a repeat of section 2 was still part of persistent lingering of the repeat conventions the historical concept of “grand binary” form

into the 1780-1820 period of the grand sym- (within a symphonic first movement) around phony was an outdated survival, vestigial, un- 1800, even when that repeat was notationally necessary to the perception of the genre. The elided. Its conceptual presence remained there, larger question, though, is why the conven- counterpointed against the given, simpler struction remained available into the later phases ture. It persisted as historical-generic memory, of 1780-1820 period and beyond (particularly even when it was not made physically present on after Beethoven’s occasional removals of the the acoustic surface of the music. Second, any expositional repeat had occurred). The expo- retention of the second repeat toward the end of sitional repeat must have persisted, however _ the eighteenth century should be regarded as exsporadically, because it was not merely vestigial. pressively significant, especially since its stronIt continued to be genre-defining, a sign of spe- gest composers— Haydn and Mozart— were

cial grandeur and formality —with an ear at- apparently coming to believe that repeat 2 was tuned also to the grand tradition and historical not as obligatory as that of repeat 1. When the lineage that had led to the mid- and later-nine- repeat was called for, it must have been placed

teenth-century sonata and symphony. there for a reason, as in the slow movement and Of the two standard large-scale repeats, the finale of Mozart’s Symphony No. 41 in C, K. second, longer one (development-recapitula- 551 (“Jupiter”), where formal processes and tion) was the one more vulnerable to suppres- monumentalized grandeur are principal topsion. This second repetition was increasingly ics throughout the whole work. Third, given a reduced to the status of an easily discardable nineteenth-century work lacking an indication option in the 1780-1800 period.!* In some of that second block-repetition, any reworked cases, concerns of absolute length or a sense of | referencing back to this increasingly atavistic reredundancy in closing particularly dramatic so- peat 2 within a longer, discursive coda, as in the natas twice might have overridden the genre- first movement of Beethoven’s Symphony No. 3 defining principle of long-range architectural in E-flat, op. 55, “Eroica,” should be viewed as repetition and balance, at least with regard to such, not as an innovative addition or accretion this development—recapitulation section. Per- to a previously postulated, differing symphonic haps the logic of the situation suggested that the practice. obligatory repeat of section 1 alone (the expositional repeat) was to be viewed as sufficient as a genre definer.

14. See, e.g, Michael Broyles, “Organic Form and the Binary Repeat,” The Musical Quarterly 66 (1980),

339-60.

CHAPTER THREE © "ESRI 0

-P

The Medial Caesura and the Two-Part Exposition ‘

A exposition has both a tonal and a rhetori- The Two-Part Exposition cal function. Its basic tonal plot— moving from an initial tonic to a secondary key, then This is the format most frequently employed by securing that new key with one or more ca- most composers of the second half of the eighdences— constitutes the exposition’s tonal form. teenth century. Hence when one confronts any This tonal form, generally the same in all so- sonata form from this period, the most reasonnatas, is worked out in different sonatas in in- able initial expectation would be that one 1s dividualized ways, according to localized rhe- about to encounter a two-part exposition. (As torical plots. Tonal form is to be distinguished will be seen in chapter 4, a continuous exposifrom rhetorical form, which includes personalized tion often plays upon, then overrides, this exfactors of design and ad hoc expression: modular __ pectation.) The cardinal feature identifying this and textural layout, selection and arrangement exposition type 1s the presence of a sufficiently of musical topics, varieties of structural punc- deployed medial caesura and (often contrasting)

tuation, and so on. The compositional ordering second theme.

of these processes produces a distinct, singular Part 1 comprises the establishment of the musical shape. This layout serves as the referential | tonic and the energized drive to the medial caerotation (or expositional rotation) that also guides sura. It contains two action-spaces, the primaryour understanding of the ordering of modular theme zone (P) and the transitional zone (TR), and

events in the subsequent action-spaces of the culminates in the medial caesura (MC), which sonata— development, recapitulation, and coda. we indicate by an apostrophe (’). Part 2 comAn exposition may be disposed in either of two prises the post-MC material and lasts until the rhetorical formats: the two-part exposition (con- end of the exposition. This section is concerned taining a medial caesura) or the continuous expo- with the cadential affirmation of the new key sition (lacking a successfully articulated medial (V in major-key sonatas, III or v in minor-key caesura). In this chapter we are concerned with ones). Part 2 subdivides into the secondary-theme

the former. zone (S)—which normally concludes with the sounding of the first satisfactory perfect authentic cadence (PAC) in the new key that proceeds onward to differing material, the event that we

23

24 Elements of Sonata Theory

Part 1 Part 2 284, occupying all of m. 21, with the literal gap on beat 4.) In rapid-tempo compositions a medial caesura is usually built around a strong half

MC EEC cadence that has been rhythmically, harmoni-

, cally, or texturally reinforced. The half cadence P TR | 5 / C | proper—the moment of cadential arrival on an active dominant— often occurs before the MC

FicurE 3.1. The Two-Part Exposition itself. Very commonly, this active V (V,) 1s then prolonged, kept alive, for several more measures

as an actual or implied dominant pedal-point, call the moment of essential expositional closure a dominant-lock, driving aggressively toward the (EEC), indicated here by a slash (/)—and the MCE articulation. ‘Thus while the moment of the closing zone (C). The two-part exposition may | MC proper—the articulation of the gap—is

be represented as in figure 3.1. frequently not literally identical with the moWhen beginning the analysis of any exposi- ment of the half-cadence arrival (the HC, which tion, we recommend that the first task be to lo- | could have happened several bars earlier), the cate and identify the treatment of the MC —to larger drive to and execution of most MCs are determine, first, if one exists at all and, if so, to nonetheless “built around” a half-cadence effect investigate what kind it is, where it falls within or “domunant-arrival effect” in either the tonic the exposition, what complications might sur- — or the dominant key.?

round it, and whether the moment identified Viewed broadly, the entire process from actually leads to an acceptable secondary theme the half-cadence arrival proper through the (S). The second task should be to examine the literal execution of the terminal MC-break, strategy surrounding the EEC. Productive anal- = which might occur several measures further yses often start in the middle of the exposition ahead, expresses a purposefully activated and and work outward to the beginning and the prolonged half-cadence effect. In referring to

end. medial caesuras as often being “built around” half cadences, we of course distinguish between the point of initial half-cadence arrival and the

The Medial Caesura (MC): MC moment itself, in those cases where these Definitions and Overview two events differ. Nevertheless we also use the shorthand symbols I: HC MC (a medial caesura The medial caesura 1s the brief, rhetorically rein- that often terminates the sustaining of an acforced break or gap that serves to divide an ex- tive V in the tonic) or V:;:HC MC (one that ofposition into two parts, tonic and dominant (or ten terminates the sustaining of an active V in tonic and mediant in most minor-key sonatas).! the dominant) to suggest this whole complex of (A touchstone occurrence of this familiar break musical activity, one in which the literal MC may be consulted in example 3.1 below: the first moment is to be interpreted referentially to any movement of Mozart’s Piano Sonata in D, K. preceding moment of half-cadence arrival.

1. Much of what follows is adapted and updated from dential arrival,” for instance —our view of the activity Hepokoski and Darcy, “The Medial Caesura and Its surrounding the MC-event differs from his in several Role in the Eighteenth-Century Sonata Exposition,” respects, as will emerge. Music Theory Spectrum 19 (1997), 115-54. Some of the 2. Defining precisely what consitutes a half cadence is adaptations seek to clarify and make more precise is- no easy matter. For one version of the concept of domi-

sues raised regarding that article by William E. Caplin nant arrival, as opposed to a half cadence proper, see in “The Classical Cadence: Conceptions and Miscon- Caplin, Classical Form, pp. 79-81. Cf. nn. 6, 11, and 14 ceptions,” Journal of the American Musicological Society 57 below, on the claim that the dominant-lock is best re-

(2004), 51-117. While we do adopt some of Caplin’s garded as “postcadential.” cadential terminology —identifying a moment of “ca-

The Medial Caesura and the Two-Part Exposition 25

The medial caesura has two functions: it exceptions to the proposed idea— encountering marks the end of the first part of the exposition either the one MC or the other can be a signal (hence our adjective “medial’’), and it is simul- both of the level of complexity at hand and of taneously the highlighted gesture that makes the probable proportions of what is to follow. A available the second part. ‘The MC is the device I:HC MC helps to predict a work on a relatively that forcibly opens up S-space and defines the modest scale; a VHC MC—suggesting a more exposition type. Somewhat whimsically, it may harmonically complex option— proposes an be thought of as metaphorically analogous to exposition with “grander” proportions. (Such the moment of the opening of elevator doors considerations are related to the concept of the onto a higher floor—making S-space possible “deployment sequence of medial caesura opor opening to the second part of the exposi- tions,” discussed separately below: a composer tion. The medial caesura provides a firmly es- sometimes seems to pass up a I: HC MC postablished platform from which the secondary sibility in order to pursue a V:HC MC down

theme, launching part 2, may emerge. In or- the road.) der for the MC to do its job most effectively The later eighteenth century saw a general within rapid-tempo compositions, energy must increase in the expansiveness and ambition of be applied. This energy is furnished by TR, the individual movements. As a result, in most matransitional zone. Asarule of thumb, once TR —_—sjor-mode cases the MC 1s constructed around

has begun, the forte energy should be kept con- a half cadence or active-dominant arrival in stant or on the increase all the way to the medial the dominant key: the familiar V:HC option. caesura proper. Any flagging of energy or vigor — Because of this statistical frequency, we refer within TR—any diminuendo or faltering drop to it as a first-level default for this expositional to plano—1is countergeneric and constitutes an moment when we are dealing with works of at event that invites interpretation. It may suggest least moderate length.? (For more unassuming the production of something unusual: a medial pieces, on the other hand, one might argue that caesura deformation or the presence of a trou- the I:HC MC could be the first-level default bled expressive problem being unfolded in the or “most obvious” choice.) In the case of the

musical narrative. V:HC MC the transition will have modulated from the original tonic to the dominant. In many transitions the preparation for the major-

The Medial Caesura: Harmonic Defaults mode S (in V) is accomplished through a darkened or stressful pathway in the parallel minor As indicated above, the MC 1s most commonly of that new-tonic-to-come. In such situations, the final gesture—the “break” or “gap” at the therefore, the TR-drive to the V:HC MC is end—ofa more complex musical passage con- _— produced with a concomitant shift to the mistructed around and often sustaining a half ca- nor mode. This means that the moment of the dence (HC) or dominant arrival, in either the — MC is locally sounded as the terminal gesture tonic key (I: HC) or the dominant key (V:HC). of a prolonged half-cadence-effect in the minor As will become clearer as we proceed, I:HC dominant, v:HC, whereupon S follows, more MCs are generally more appropriate for shorter, brightly, in the major mode. (A particularly lighter pieces. On the other hand, V;IHC MCs charged instance of this dramatic chiaroscuro may tend to be more frequent within ambitious be found in the first movement of Beethoven’s works of moderate length and larger, especially Symphony No. 2 in D, op. 36, with mm. 61-71 toward the end of the eighteenth century. As delivered aggressively on A minor [MC at m. such —and even though one can come across 71] and S emerging in A major at m. 73, after 3. An array of statistical evidence regarding the fre- lution of the Viennese Classical Style,” Journal of the quency of what we call V:HC and I:HC medial cae- American Musicological Society 42 (1989), 275-337. (We, suras in Haydn, Mozart, and others has been compiled however, do not find the term “bifocal close” for the I: in Robert S. Winter, “The Bifocal Close and the Evo- HC MC to be helpful.)

26 Elements of Sonata Theory

two bars of major-inflected caesura-fill.) Such scending caesura-fill in the first violins lead to occurrences participate in a dialogue with the the E-flat-major start of Sin m. 52. In this case normative, major-mode MC expectation. The the rhetorical effect produced 1s that of the halfv:HC MC option is a commonly elected nega- cadence MC —the passage 1s clearly in dialogue tive overlay onto the conceptual first-level de- with that norm—but the music 1s dramatifault, V:HC. We still have a first-level default cally staged as being “unable” to produce the MC, but one subjected to the additional surface more normative HC at this point. Somewhat feature of temporary minor-mode mixture—a similarly, in Mozart’s C-major Overture to La momentary “lights-out” feature. (Chapter 14 clemenza di Tito, K. 621, a suddenly introduced considers the expressive implications of such a Ve /V (m. 28) seems to startle the music into

mixture.) nothing less than a fermata-stop (m. 29) that

Additionally, within the first-level default in serves as the exaggerated GP-gap of a V8/V MC either major-mode or minor-mode expositions deformation. S begins in G major in m. 30. one occasionally finds a seventh included in the At least in works of substantial length the V chord at the MC point. This seventh 1s best second most common major-mode option, the regarded as a passing tone. In the first movement second-level default, is to build an MC around a of Haydn’s Symphony No. 100 in G (“Mili- half cadence or dominant arrival in the original tary’), for example, the structural V/V, the half tonic, a J: HC. (As mentioned above, in shorter cadence proper, is articulated at the downbeat works, a case might be made that the I: HC MC of m. 62, after which it is prolonged. During is somewhat more appropriate: it might be rethis prolongation the seventh is added (entering garded as a first-level default 1n certain situafirst in mm. 64—65, though most prominently tions. For the present, the discussion is framed in mm. 69—73), suggesting a V®~’ figure. The around works with grander proportions.) In seventh (4 of the new key) resolves to an i1n- this second-level default, I: HC, the transition ner-voice 3 at the onset of S (m. 75). This ad- will not have modulated: it will have begun and dition of the seventh during the drive to the ended in the tonic, and it will be up to the ensuMC proper is not uncommon but is normally ing S-space to establish the new key, usually by limited to the first-level default (V:HC MC). beginning directly in it. Because second-level If the seventh were added to the dominant of | MC choices are not infrequent, TR-space cana second-level default (I: HC), the tendency of not be defined in terms of modulation. Once the resultant V’ to resolve to the tonic would again, first- and second-level MC defaults are preclude the requisite tonal shift to the key of — not expressively equivalent. The first, V:HC the dominant. Nevertheless, this can occur at MC, is a more decisive gesture: it announces the MC point of the initial tutti rotation of a the intention to open part 2 more solidly, with Type 5 sonata (concerto movement), where a its new key already in hand. The second, I:HC, modulation 1s not required, as in Mozart’s Piano is weaker, usually occurs early on, predicts a Concerto No. 21 in C, K. 467, mm. 20-26. briefer or less ambitious sonata, and sometimes On rare occasions one encounters the substi- purposefully generates problems in what foltution of an inversion for the V or V’ chord at lows. the MC point. Regardless of whether the domi- The minor-mode, derivative analogues to nant has previously appeared in root position, the above are III:HC or v:HC as strong firstthis situation should be understood as a medial- level defaults, depending on the key to which caesura deformation, which might well impact on one is modulating (moving to the minor domthe subsequent S. In Beethoven’s C-minor Co- inant occurs much less frequently), and 1:HC riolan Overture a strenuous TR manages to lock as the second-level default. While examples of onto V° of E-flat (here entering as E-flat minor) the former are frequent—an MC built around in m. 46. (Obviously, this cannot be construed a half cadence in the new key 1s the most comas a half cadence proper.) After four measures of | mon choice—examples of the latter are relaconvulsive upheavals around that chord, a V°/ tively rare. The reason why is obvious. While 11 MC-effect occurs in m. 50. Two bars of de- major-mode statements of the I:HC MC (sec-

The Medial Caesura and the Two-Part Exposition 27

ond-level default) may easily become the tonic effect in m. 39, which 1s returned to and recapchord of the new key (V) and the S-to-come, tured in m. 43) and assign the modulation to III this is not the case in minor-mode works. In to the “poetic” caesura-fill bridging the MC to other words, 1: HC MC (say, a G-major chord the onset of the cantabile S in D major (m. 47).

sounded as an active V in C minor) is obliged Most sonata forms display either a first- or to yield at once —as a quasi non sequitur—to the second-level default MC, one built around a tonic of the mediant major with the onset of — half cadence that may or may not have been S (the key of E-flat in a C-minor exposition, prolonged by means of a dominant-lock. Much whose appearance also produces a cross-rela- less frequently, one may find an MC-function tion between the original dominant chord’s B produced by a perfect authentic cadence in the natural and the new tonic’s B flat). When this new key (PAC). In major-mode sonatas this does occur, the effect can be striking: a sudden third-level default is V;PAC MC, which occurs, pull out of the ominous tonic minor into the for instance, in the first movement of Mozart’s brighter, more “hopeful” mediant major.* The Quartets in D, K. 155, m. 28; in E-flat, K. 428, touchstone examples may be found in the Pres- m. 40; and in B-flat, K. 589, m. 45. This procetissimo finale of Beethoven’s Piano Sonata in C dure 1s also found with some frequency in earMinor, op. 10 no. 1 G: HC MC at m. 16, stalled lier and briefer works. (The V:PAC MC in the with a quizzical fermata, followed directly by first movement of Haydn’s Symphony No. 10 the forthright S in E-flat at m. 17) and in both in D, m. 23—one of several examples in early the opening ritornello and solo exposition of | Haydn—could hardly be clearer.) The minorMozart’s Piano Concerto No. 20 in D Minor, mode-sonata analogue is normally III:PAC, as K. 466 (a literal 1: HC at m. 28, proceeding into in (though with subsequent complications) the a dominant-lock and 1: HC MC at m. 32 anda first movement of Mozart’s Quartet in D Minor, sudden shift to F major, III, for the opening of — K. 421, m. 24 (see example 4.3 and the accomthe ritornello’s secondary-theme space at m. 33; panying discussion in chapter 4). Alternatively, see also the solo exposition, with its 1: HC MC in works that shift to the minor dominant key, at m. 114). From time to time the production of | one might find a v:PAC MC.°

ai: HC MC can be presented as a compositional PAC MCs are stronger tonal and rhetorical problem that needs immediate emendation. In gestures than are HC MCs. Because they are the first movement of Beethoven’s Quartet in C heard as signs of closure, not of expectancy, and Minor, op. 18 no. 4, a1: HC MC-effect in m. 25 because they sound the same perfect authentic leads not to S proper but to a classic situation of | cadence that will define the EEC concluding “medial caesura declined” and the initiation of | the secondary theme, they present problems of

one type of trimodular block (TMB)—both understanding. From time to time they emerge of which strategies are discussed separately be- after a composer has already dallied with the low. Decades later, Mendelssohn would provide V:HC option— perhaps already producing an the B-minor Hebrides Overture with a 1:HC HCE arrival in that key, or very nearly so. When MC (m. 43; notice also the earlier 1: HC MC- this happens, it is as though the music at first 4. In effect the juxtaposition 1s produced by a chro- chord of the recapitulation, where the more typical V, matic 5—6 shift, in which both the third and the fifth sonority 1s sometimes replaced by V/vi (V of A minor of the active V (the G-major chord within a C-minor within a C-major movement, for example, giving way exposition) are altered. Recent neo-Riemannian theory almost immediately to the C-major recapitulation). might also describe this as a PL shift: a simultaneous 5. An extremely rare—and clearly deformational—al-

application of a color-shift to the parallel (P) mode (a ternative is found in Beethoven’s D-minor Largo e G-major chord, V of C minor, thus inflects to G minor) mesto slow movement of the Piano Sonata in D, op. 10 followed by a “leading-tone exchange” or Leittonwechsel no. 3, which features a VII:PAC MC (a PAC on V of III, (L) (the resultant G-minor sonority inflects its fifth, D, C major, in m. 17) —as if seeking to “close” early, albeit

to E-flat, thus producing an E-flat chord). This familiar in the “wrong” major mode. The S and C that follow juxtaposition is discussed in somewhat more detail in are in A minor (v). ch. 10 in conjunction with the final active-dominant

28 Elements of Sonata Theory

“decides” to drive toward the normative V:HC preempt the sense of tonal closure that we asMC, only suddenly to “change its mind,” aban- sociate with the EEC—the V:PAC that must doning the normative implications of the dom1- be re-produced down the road at the end of S. nant-lock (if that lock had indeed been initi- But how certain can we be that such a V:PAC ated), and pushing instead, impulsively, toward should not be taken for the EEC? This decision the stronger PAC in the new key. Thus a V:PAC is a crucial one. It concerns the structural imMC is sometimes produced in a context that has portance of that first V:PAC at what might well suggested, then overridden, a more normative be the MC: is it a secondary, local effect (perhalf-cadence-effect MC option.® (In K. 155/1, haps describable as the strongest possible “tonifor instance, the half cadence and dominant- cization” of V of I at the MC point)? Or is it to lock, V:HC, occurs at m. 20. The lock proper be taken as a decisive structural event within sustains the V, through mm. 20—23 but is aban- the genre—nothing less than the EEC? Or can doned at m. 24 in order to plunge into the em- there even be other options for interpretation? phatic V:PAC MC at m. 28. The subsequent S Deciding this matter in individual cases detheme, m. 29, as it happens, appears to register pends on three factors, each of which involves its surprise by beginning off-tonic, on the su- matters of interpretation and experience. The pertonic chord.) In some of these instances V: first is the question of how far into the exposiPAC MCs are elaborate, more decisive versions tion the V:PAC is sounded. Once we have proof caesura-fill of the 5-1 -descent type, and ceeded past about 65 or 70 percent through—in distinguishing between the two can be difficult other words, once we have experienced a proor very much a matter of individual interpreta- portionally overlong transition— V:PACs betion. (This last feature is discussed in more de- come less convincing as MCs, since a medial tail below, in the discussion of caesura-fill.) caesura normally occurs earlier in the exposiSuch observations lead to larger speculations: tion. The later the V:PAC is produced, the more the very concept of a V:PAC MC is potentially likely it will be taken as the EEC. (Put another problematic. What leads us to think that what way, normally the only way that a composer can we call a V:PAC MC is not the EEC? (Such an have a V:PAC serve as an MC 1s to expand the early EEC would define that exposition as con- proportions of what follows to the point where tinuous, not as two-part, since there would have it can be regarded as a convincing part 2: S and been no prior MC.) One might reason thatif = C. For this reason the V:PAC MC option typithe generic goal of an exposition is to produce a cally suggests an exposition and subsequent sosatisfactory PAC in the secondary key, any such nata of notable proportions.) The second factor V:PAC-effect at this point might initially lead concerns how the V:PAC was prepared and prous to suspect that that aim has been achieved. duced (idiosyncrasies in the preceding TR). The One soon learns that it is part of the expressive third is the character of the module that follows. character of a local V:PAC MC to threaten to If it is a clear, contrasting theme, 1s it S-like or

6. In response to one case of this as illustrated in our that a dominant-lock might be abandoned en route—in article, “The Medial Caesura,” pp. 129-30 (Beethoven, other words, that it might be staged as “changing its Piano Trio in G, op. 1 no. 2, first movement, mm. mind”’—1in order to proceed to a PAC. Such a proce98-99), Caplin (“The Classical Cadence,” pp. 108-12) dure would “unfreeze” the locked dominant (still an insisted that the PAC-effect at m. 99, while displaying active dominant, a V,, after all) and treat it as more of “cadential content” (the two-chord combination V7—]I), a “normal” V, that can proceed onward toward resocould not be considered a PAC proper because it lacked lution. In any event, we agree that Beethoven’s op. 1 “cadential function,” at least according to his “highly no. 2/1 provides a problematic case along these lines, constrained” (p. 56), much-restricted definition of that since the lurch to the V:PAC happens SO rapidly and so function. In Caplin’s view, once an HC-arrival has been closely resembles caesura-fill of the 5—1 -descent type. attained in TR, all that follows in the dominant-lock The following example mentioned in the text, Mozart’s must ipso facto be considered “postcadential” and by that K. 155/1, provides a clearer illustration of the process. definition incapable of producing a PAC at its end. That Some of these issues are revisited in n. 14. Cf. also n. 2 argument, however, fails to consider the possibility and nn. 11 and 14.

The Medial Caesura and the Two-Part Exposition 29

C-like? Deciding this 1s not always easy. This or obstinate?) TR that, still in the grip of the general situation arises with some frequency erounding tonal principle of the P-zone, dwells in Haydn, who had a fondness for planting a on an unusually static tonic. This emphasis, in decisive V:PAC in the 55—70 percent range of — turn, demands analytical and hermeneutic inthe exposition.’ Alternatively, what follows the terpretation. first V:PAC might be less of a “theme” than one The classic example occurs in the first moveor more short modules that recapture or restate ment of Mozart’s String Quintet in G Minor, K. that mid-expositional cadence. When that oc- 516. In this extraordinary exposition the negacurs, we are dealing with what we regard as the tive pull of G minor is apparently so strong that second type of continuous exposition, a possi- TR (beginning in m. 9asa TR of the dissolving-

bility dealt with in chapter 4. consequent type) finds itself unable to escape its The even rarer option, a I:PAC MC, may be control. The result is one of the bleakest MCs in considered a fourth-level default. In eighteenth- the repertory, the 1:PAC at m. 29. The precedcentury works a I:PAC or [AC-substitute (im- ing, forte 1:PAC at m. 24, Neapolitan-enhanced perfect authentic cadence) leading to an obvi- (m. 23) and brusquely closing the door on the ous Sin the new key may occasionally be found fatalistic G minor, foreshadows this MC-effect. in light, small-scale works, in some telescoped What intervenes in mm. 25-29 is a “timid,” or abbreviated expositions, and in some slow failed attempt to wrest free of the clutches of G movements. Generally the PAC or IAC closes minor through a momentary glance at VI. Beoff a brief, straightforward P, and the resulting ing drawn back once again to G minor and to impression is that of omitting the TR-zone al- the 1:PAC in mm. 27—29 1s chilling—a second together. Because of the effective ellipsis of TR, confirmation of the countergeneric inability to the I:PAC or IAC at the end of P is asked to do escape from the gravitational negativity of the double duty as the rhetorical MC. This occurs tonic. The S that follows in m. 30 (the rhetorical in the first movement of Mozart’s Symphony signals make it clear that this is S) begins in the “No. 7” in D, K. 45, m. 16 (I:LAC MC, with S same, inescapable G minor and finally manages in V at m. 17) and his String Quartet in A, K. to hoist itself up to the proper mediant major in 169, m. 11 (I:PAC MC, with S in V at m. 12). mim. 36—37 (although further damage to S 1s

In larger, more ambitious pieces the ex- also apparent in subsequent measures). tremely infrequent I:PAC or IAC MC can carry It may also happen that a longer stretch of a different implication. Here, following P, one caesura-fill, branching out from the tonic auenters what seems rhetorically to begin as a nor- thentic cadence at the end of TR, 1s called mative TR. That TR, however, proves unable upon to accomplish the modulation to the new (or unwilling) to produce any of the three more key. This is uncommon in eighteenth-century standard MC defaults: V:HC, I:HC, or V:PAC. works but turns up occasionally in works in the

In some expositions it may “try” to produce nineteenth century, as in the first movement one of those—or to initiate a motionin one of | of Schubert’s Symphony No. 8 in B Minor, those directions —before being drawn back to “Unfinished,” D. 759, first movement—1:[AC the original tonic. In other cases it may simply MC at m. 38 (perhaps interpretable as a1:PAC), bask in an ultra-stable tonic without any ges- with a modulatory caesura-fill leading to an Sture toward a typical MC. In either situation space in G major that begins in m. 42.8 one confronts a “failed” (or gesturally weak? 7. Haydn often gives us an emphatic V:PAC at a point (Similar situations crop up also in midcentury sonatas where it is difficult to decide what its intended function of less well-known composers) However one regards it, might be: EEC? MC? witty or purposefully “difficult” this V:PAC option (not too far past the midpoint of the gesture? Is its very point to place us in an ambiguous exposition) is one of the important features of Haydn’s interpretive position? Or might it be the articulation of conception of sonata form. a different (third?) type of exposition altogether— per- 8. Related instances would include modulatory CF pas-

haps one customized by Haydn for individual use or sages following a purposefully “wrong-key” MC, such perhaps one known to him from more local traditions? as that in first movement of Schubert’s Piano Sonata

30 Elements of Sonata Theory

The Medial Caesura: Common Characteristics HC that is immediately followed by an “acceptable” new theme in the proper subordinate key, Within Allegro compositions (first movements that HC may be interpreted as a medial caesura. and finales, most overtures) the medial caesura (Put another way, the situation is staged as if the is often the final moment of articulation fol- apparent S-theme has “understood” that HC to

lowing one or more measures of preparation have been one.) An example may be found in on a prolonged structural dominant (dramati- the first movement of Mozart’s Piano Sonata in cally sustaining the earlier arrival of a I: HC or C, K. 309, in which TR begins in m. 21 and V:HC). A common sequence of events 1s: (1) the structural dominant is reached only with the initial stages of TR, by and large consis- the arrival of the V;:HC MC in m. 32. In adtently gaining in energy; (2) the attaining of the dition, many slow movements— generally lyristructural dominant by means of a half-cadence cal movements— omit the dominant-lock. The arrival (usually either a V:HC ora l:HC—or McC-effect in these gentler movements is often the minor-mode-sonata equivalents), which 1s produced by a mere half cadence without much frequently then locked onto as a literal or 1m- additional rhetorical emphasis. (The situation plied pedal-point (structural-dominant lock); (3) in the C-major Andante movement of Mozart’s the prolongation of this still-active V (V,) and Piano Sonata in G, K. 283, 1s typical: V:HC MC

the rhetorical drive to the medial caesura—a in m. 8.) It may be that any prolonged domidrive that sustains or even increases the energy nant-lock in slow movements was intended as accumulated thus far; (4) the articulation of the an unusually strong or expansive gesture.

MC proper, the terminal gesture of the entire Normally, however, in Allegro movements, process. Example 3.1 shows the opening of the in order to function as a normative medial caefirst movement of Mozart’s Piano Sonata in D, sura, the forte half-cadence arrival within TR K. 284: TR begins in m. 9; the half-cadence must be additionally reinforced. The whole arrival and the dominant-lock (I:HC), holding process often proceeds as follows: fast onto that V,, occur in m. 17; the second- 1. The structural dominant (the half-cadence level default MC (I: HC MC —the concluding arrival, which typically precedes the MC, somegesture of a prolonged half-cadence process) 1s times by several measures) is often approached

articulated in m. 21; S ensues in m. 22. through a chromatically altered predominant Not all Allegro compositions articulate all harmony that contains #4. (This scale-step is four of these events. It 1s possible—for a less reckoned in the key within which the half carhetorical effect—to sound the MC at the mo- dence is to be sounded. In the case of directed ment of the arrival of the half cadence, thus motion into a V:HC, 4 of the new key would be omitting the structural-dominant lock alto- 1 of the original tonic.) This altered predomigether. This would be an instance of a nonelab- nant is frequently an applied chord (V/V, V’/V, orate, straightforward articulation of the MC. vul?/V, or vil°’/V in root position or inversion) Generally considered—and if not overridden or an augmented sixth chord.’ The chromatic

by other evidence —if TR produces a notable line 4-#4-5 or 3-#4-5 often appears in one in C, D. 279. Here we have an unequivocal arrival on ceded by its own applied dominant (V of V, V®° of V, the “wrong dominant,” 111: HC (V of E minor!) at m. and so on), “we might refer to this goal as a tonicized 37, which V, is immediately frozen as a dominant-lock half cadence.” While accurate, this terminology might ending with a 11:HC MC, mm. 37-41. Four bars of be potentially confusing. This surely means only that an expanded fill, mm. 41—44, accomplish the modulation unequivocal half cadence 1s locally supported by its own to the generically proper key, G major (V), in which S dominant. One continues to experience that dominant

then begins, m. 45. arrival as an active V, as V,, not as a tonic (V,,), particu9. According to Allen Cadwallader and David Gagné, larly in “lighter” cases of a mere 4—#4—5 or 3-#4-5

Analysis of Tonal Music: A Schenkerian Approach (New motion in one of the outer voices— where any sense of

York: Oxford University Press, 1998), p. 409, n. “tonicization”’ in the normal sense of the word is virtu23 (referring to a situation in the first movement of ally nonexistent. Thus there can be no claim that anyBeethoven’s Piano Sonata in G, op. 49 no. 2, discussed thing like a “full” tonicization of that V, has occurred: on pp. 311—29), when the half cadence arrival 1s pre- the arrival on V 1s still a half cadence, not a concretized

The Medial Caesura and the Two-Part Exposition 31

of the outer voices (4—#4—5 as the bass-line (dominant-lock). This may involve alternating V approach to V, is especially characteristic); if an with a neighboring ¢, producing 3-373 neighbor augmented sixth chord is employed, the typical motion. Sometimes the neighboring $ is supbass line is b6—5 and the #4—5 move occurs in ported by 7? in the bass, creating an apparent the upper voice.!° The texture at this moment is V-I-V alternation. The larger point, though, vigorous, highly active; the dynamics, usually a is that the V, of the half-cadence arrival is vigstrong forte, will persist or even gain in intensity orously seized onto, frozen in place, kept alive in the subsequent drive to the medial caesura. by means of a specialized pedal-point effect that

As mentioned earlier, within major-mode announces that TR is ending with a continued sonatas it is not uncommon to encounter inter- push toward the MC. The sense of an “HCmixtures with the minor mode, perhaps even moment” is not released and left behind as a a shift to the minor mode — usually that of the mere past event—as happens with most other new key—in the vicinity of this half-cadence kinds of HC phrase endings— but rather is held arrival or dominant-lock point, although, if this onto, brandished as an achievement, sustained occurs at all, it may also take place earlier or as a continuing function with a specific role to later in TR, perhaps even persisting through the play at this point in the form. MC itself. An appearance of the negative mi- 3. The normative, unflagging drive in the nor mode participates in the generic expectation space between the locking onto the structural of the intensification process, either enhancing dominant and the actual articulation of the it or engaging in some other kind of dialogue MC is of paramount importance. Any attenuwith it. The mixture with minor may suggest ation of dynamics here should be viewed as the introduction of uncertainty, doubt, or peril countergeneric, or perhaps—especially by the into the narrative thread—or (as sometimes 1n later eighteenth century—as a less common, Beethoven) the onset of a grim struggle in the second-level default that calls attention to itself production of the MC and S. Following such a and challenges the prevailing norm of energymodal shift, the ensuing major-mode S emerges gain. Depending on the circumstances a dy-

with a sense of brightness and relief. namic collapse in this space might represent the 2. Once attained, the structural dominant staging of a momentary crisis of confidence in is frequently prolonged, perhaps by neighbor one’s decision to enter S-space. The S that folmotion, as part of the drive to the MC proper — lows the dynamically underprepared MC might tonicization of that V. The difficulties of interpretation theme, over that V. Such S-theme beginnings are disincrease, however, as the sense of that “tonicization”’ cussed in ch. 7. becomes increasingly intense with differing strengths of 11. Thus our view differs from that of Caplin (“The applied dominants to the V,. (It is possible to imagine Classical Cadence,” passim, but see especially pp. a continuum of differing applied-chord strengths, for 89-91, 98-100, 108-12), who, as mentioned above instance, that ultimately lead one to cases that appear to (n. 6), regards this dominant-lock as “postcadential” in

articulate a V:PAC as the MC.) function. We find it preferable to think of this stretch of 10. Fora stronger sense of rhetorical emphasis it is pos- music not so much as existing “after” the half-cadence sible to approach the structural dominant— or half-ca- arrival (which in a literal sense, it does) as keeping that dence arrival—more than once in fairly rapid succes- arrival alive, refusing to let it go, animatedly spreading sion. Thus once the structural dominant is sounded, it its sense of being still active over several more meacan be immediately re-sounded through energetic reit- sures. Put another way, the dominant-lock may be conerations of the half cadence. The music can go through sidered a special prolongational technique that extends the cadence several times, reapproaching and rearticu- and “holds in place” the HC-arrival effect for a specific lating it, helping to produce the rhetorical drive toward rhetorical purpose. See also n. 14. Caplin’s English-lanthe MC proper. The touchstone example occurs in the guage term for such harmonic locks is “standing on the first movement of Beethoven’s Piano Sonata in F Mi- dominant,” a translation of Erwin Ratz’s “stehen auf der nor, op. 2 no. 1, where the predominant 4—#4—5 bass Dominante,” which had appeared in the latter’s Einfiihmotion preceding the half-cadence arrival is first stated rung in die musikalische Formenlehre (3rd ed., enl., 1973).

in mm. 15-16, then restated twice more, mm. 17-18, (See Caplin, “The Classical Cadence,” especially p. 90, 19-20. (M. 20 is the MC proper, but, unusually, it also n. 101; and Caplin, Classical Form, pp. 16 (“postcadential initiates a dominant-lock and the onset of an S° (or S!-°) standing on the dominant’), 75, 77—78, 144-45.)

h S a PT SK 4 to. EXAMPLE 3.1 Mozart, Piano Sonata in D, K. 284, 1, mm. 1-24 Allegro

iery A4) 0 Owe elTm eeeee eeOe eeeeere es ee > Dk SC TY Af UT TT6ee e—= eeeee es eee e—=Ee ea BA =e eeeee = eee eee

Piano |esanceey ee peGnanane aa Ae, eel ccc Ee PD nel cos ODee §=

|COP Se aSc ee POT he fF. ee eee eee eee Oe ee 2 eee ee eee eee, eee ee eee eee eee

|i ee ||—a-_eeI— —o —a.es' esne oo 2 4oe a” a,XN —a |a— —no aEcf) aaA 0+ eed eed a cee ee a cred a a a aceed Oe ae ee eswee ye =” "ee «=, ee PN Sa ema Ta=_—a i KE ™® CCU . . ea ’ 4,

ane eeaeeee0eeeeeee eseeeeeQO CO eee Oe eee eeeee eee eseee ee eee eeOC ee |Oe | | eee Ff ft

| abe a AeS el ee ee ee A ee ee A ned A ne eee PsA QO GORA OO QQ

7a

P St P St Pp , aw a ————— —

fe @oe@rl eo. Pf,Leer |eeFF 6h, Nt OF AWA a ee ee eee ee ee.) eee Osee 2Phe ere ee |eeee eennn eeeBm ee eee eee2eee eee ira him —........j]_ 050 ——— See ee ee eee Aa Pe A eee

0 ee eee p | | | | CO ee ae 2 oe

be Ce Es | QO a ne ee” =

i)ae ana —GO = a Iee a2Fc ee rs ss GS SO CO CC QD A 1+a|een ae PSesCS SC ee esCS es ee eeCS , $= 9

ee Ee a QO CO i ApT1| TT el ael ff = ed o_o ee eee re ee Pp St p a ee el OF ee ee ee ee eee eee eee eee eee eee ee eee ee eee eee eee ee ee ee ee eee eee ee ee

|igTc =| Yr lee

a” ,eeee fv sl A OC OOOH tt eC Ct BA’ 4 ee wn)ee=eTee ee —__ ee gg Rswee Pn eaa Ee eca Re ee A es eeee

0 = ey es * A es ee ee ee ee ee ee ee eee eee eee

ee fe—s l4

Ft ae a ae a ae a a ee ee es es es Ss Qs GD GY Qn RRS RS GR Gr RS GS es ee

pp dlOO re(pee ee aA 8Pl eenee (Se | ee SO 32?

Pe EXAMPLE 3.1 (continued)

Ec ee es QO pm Pf Otme Ot aE Po gs

LU ~_tc ~~ Li(>: A. *QO«=e gM meee me~~ as$§$lrCU afce" ee cee__ jee ceemG cee nee nee pe | yp — —___/ ________® _____», |___Fan [Stee oe eee

EP A ee ee ee ee £. 0 «= = = = =) = =|

a a rr en re | | en eee

f ) 7A*lebdse’ aesSeS, | . - >” ba ne + PF

ir. TEIgg pM a 18

| wo li, ic Se ae Tc ke§$lmP UCU ..LUD Oe le ll GT ee sel —e e—“—m®_

rail AY 2 Ae Oe eee eee eee Sees 2 Oe Oe Oe ee ee eee 2

ns Pa a(C

20 f a ee ‘ *s .ee—“ . By eh ee a GT ew

Ec a a a a ee er ee ee SO ee ee am tlm esee CD OO QA, cu a Bh, ee esnc ee es | A, es ”>,. © S

Cw J fT ee egy eh le | | | (oem) ae ee A SSa=O A es00= ml GS (ee / » A Ay ee ee em i ace gm A se —_ ReesFt i SS ne ce 0aeses a ee es es 0es (US

iCo ee arem cc Ieepce oe QS Qeea =|ee aoeT ee ee a —=Ee OO Om eee es Ff ee i _»»pgp|)p|)|p|EI

ed aed ll

ft o«—_ eee FFF ee ee a2aeee ee leee Beue ee 2ee2eeeedeeee ee ee eee2eeeee ee

a La. Pe > I cee rec eet 2, Os Ot 13

mem |ee... =e ee“, tmeee me CE =m if. i gr2 gf fF «= {T— | =~ J] | gfala FrnS fF POee—=e, &® iam eeeFtee eelm eeFf —..._ see eee EG eee

ne an is

—) #£+S+ | |,”[| lisa 2— fi i a # | gr>| «| |] | -Ff__, g | gr iT|{ | | YT go ffstesssss ge ge yg ws » # ~~ ”|

|Rel loan UU ee ee” ee ~~ | | | ee) ee a” = i” $=” =< |6—— ee EE Se es aaa es ee ee cr eel

ele nnn” nn” nn A AGQ ee ree ee aeeAOP a ee ee SO QO RD QD QO es eseeISS QO CG A

— pp ee eee |nO re a a nc rs _| QO —_ eQO Q |iaes fF | get ge f ra| aa.a ga a es er.

PON gil Tt fe

A eeee a ee ee cr eeeel | aEE eS esl ee Deee eeaeea ee es ee Rel CS nn nn A a a ne ho =) |

aoTAe—PFw ce nce GQ Be

| oye ee OFT CN’. oT. §*§$>FS

to both procedures as FS. However it might be can pass through the zone of conversion in a disposed, this caesura-free succession typically number of ways. We may imagine the manifold occupies the large center portion of the exposi- possibilities as arranged on a continuum repretion (perhaps its middle 60—80 percent), now senting the various degrees to which we sense understood as a continuous, not a two-part, that a potential MC has been suggested. For exposition. We schematize our experience of heuristic purposes we might identify three situthis large, central section as TR=>FS, in which ations within this sliding scale (each case pushes the symbol => stands for the word “becomes.”* the sense of an MC toward a clearer articulaWhat begins (we think) as TR—it may even tion): (1) the TR=FS can move past the last provide indications that a predicted MC 1s in possible S-point with no caesura signals whatpreparation— shifts conceptually to the FS ever; (2) it can reach and perhaps prolong the modules (or, alternatively, to the unbroken the- structural dominant— even initiate a clear, gematic chain) characteristic of the continuous ex- neric drive to the MC — but fail to crystallize a position (an FS that will drive, without an S, to- medial caesura; or (3) it can actually articulate ward the EEC). Thus the symbol of “becomes,” a seeming (or potential) MC and perhaps even =>, also represents the process of conversion. All enter a process of caesura-fill but then both de-

of this may be represented as 1n figure 4.1. cline to furnish an immediate, subsequent S and At the basis of this understanding 1s the as- refuse to drive toward a more acceptable MC sumption that a listener adequate to the de- in the ensuing bars. ‘This last case is sometimes mands of the piece actually can experience such difficult to distinguish from extreme examples a process of conversion. Sensing it depends both of the second. It belongs generally to the cateon an experience with the style— having a large gory of medial caesura declined, but the psychology

inventory of normative exemplars at hand—and of its production 1s perhaps best understood in on grasping the proportions that a composer its relation to the first two cases. What is needed seems to promise at or near a piece’s outset. One at this point 1s a closer look at each possibility.° function of the opening ideas of each exposition 1. FS may move past the S-point without our is to predict the rhetorical scale that will follow. noticing it. In other words, we eventually come

4. We have adopted this symbol for “becomes” from 5. Closer discussions, with printed musical examples, Caplin, Classical Form, p. 47, where => also “denotes a are found in Hepokoski and Darcy, “The Medial Cae-

retrospective reinterpretation of formal function.” sura,’ pp. 133-38.

54 Elements of Sonata Theory

to realize that we are beyond any conceivable 2. The composer may create the expectation of S-point. By all reasonable standards, it is now an imminent MC only to veer away from it for more

too late for an S-theme, although we did not Fortspinnung or other elaboration. How close we register our having passed by its potential mo- get to the implied caesura-point varies from ment: we heard neither a medial caesura nor any case to case. The MC-point proper, of course, compelling generic signals of an approach to results from the laying-down of the structural one. To be sure, such pure instances of the con- dominant, the harmony that could potentially tinuous exposition are rare among celebrated articulate aI: HC, V:HC, or III: HC medial caeworks of the later-eighteenth-century compos- sura. The structural dominant may be touched ers—the Presto finale of Haydn’s Quartet in B lightly and immediately rejected (as if hot) with Minor, op. 33 no. 1 is a locus classicus: TR=>FS a new burst of Fortspinnung that overrides (or extends from m. 13 to the III:PAC EEC at m. writes over) the normal tendency of the exposi51, followed by C-material in mm. 52—63. An- tion to divide into two parts at this mid-expoother is the first-movement exposition of his sitional point. In other cases one might produce Symphony No. 13 in D—notable also because a half-cadence dominant arrival, lock onto the the TR=>FS portion of the exposition (m. 14 structural dominant to prolong the still-active V, beat 3 to m. 34) pushes all the way to its end: in and perhaps even furnish some additional signals other words, this exposition lacks a C-zone alto- to suggest the production of an MC —the music gether. This is a case, therefore, where the third begins to “fall into” pre-MC behavior— only of Larsen’s supposed “three parts” does not exist. to draw away from it before that MC turns into Still another, much later, is the exposition of the a reality. finale of his Piano Trio in C, Hob: XV: 27. Here An example of the latter situation is prothe relentlessly churning TR=>FS (mm. 43-81, vided in the first movement of Haydn’s String with C occupying mm. 81-93) follows its con- Quartet in E-flat, op. 33 no. 2 (“Joke,” example ceptual opposite, a square-cut, rounded-binary 4.1). Here TR sets out in the tonic in m. 13 P, mm. 1—43 (a characteristic finale theme of | and moves almost immediately to the dominant “rondo character,” even though what follows 1s arrival, V/V on the third beat of m. 14. This not a sonata-rondo: see chapter 18). This type newly locked structural dominant now underof continuous exposition also appears in pieces pins a generic drive to what we presume will be from the earlier part of the century. Elementary a standard V:HC MC, a drive beginning in earexamples may be found in some of the Sammar- nest with the reiterated figures in m. 15. The retini symphonies from around the early 1740s iterations and hypermetrical implications clearly and in several of the first movements of C. P. E. suggest the production of a normative medial Bach’s keyboard sonatas from the same time, caesura in m. 19. It would be easy to imagine such as the “Prussian” (1740—43) and “Wiirt- a differing m. 19 that consists (assuming the

temberg” (1742—44) Sonatas.°® most generic of choices) of three hammer-blow

6. In the first movement of Sammartini’s Symphony In C. P. E. Bach one often finds a similar format: “No. 3” in D Major (J-C 15, before c. 1742) the first an initial P-gesture; a modulatory TR=>FS (typically half of the binary (proto-sonata) structure may be con- sequential—and rarely very long) that proceeds to a strued as: P (mm. 1—8); a short-winded FS (mm. 9-19) PAC (the EEC); and a (brief) “appendix” theme (C) that never suggests anything caesura-like but does lead at the end to solidify the new key. Because C. P. E. to the EEC (V:PAC) at m. 19; a brief, cadential close Bach’s textures often feature breaks and discontinuities, (C, mm. 20—28). The score 1s available in The Sympho- the caesura situation is sometimes difficult to assess. For nies of G.B. Sammartini.: Vol. 1: The Early Symphonies, a general discussion of C. P. E. Bach and the frequent ed. Bathia Churgin (Cambridge, Mass.: Harvard Uni- inappropriateness of the concept of the “second theme,” versity Press, 1968), 76—77. “No. 3” is renumbered as see David Schulenberg, The Instrumental Music of Carl “No 14” in Newell Jenkins and Bathia Churgin, eds., Philipp Emanuel Bach (Ann Arbor: UMI Research Press, Thematic Catalogue of the Works of Giovanni Battista Sam- 1984), e.g., pp. 100-105; and William S. Newman, The martini: Orchestral and Vocal Music (Cambridge, Mass.: Sonata in the Classic Era, 2nd ed., pp. 420-21. Harvard University Press, 1976), p. 54.

The Continuous Exposition 55

F-major chords (V of B-flat), followed by a to rest or anchor itself with an S-theme on its rest, a drop to piano, and (since this is Haydn) a other side. Instead, the fill 1s reinvigorated into monothematic S theme — or perhaps a contrast- an expanded Fortspinnung or “thematic” moduing one—beginning with the upbeat to m. 20. lar chain that takes on a life of its own. In such But instead, at the last moment, in m. 19, Haydn a situation the decisive change in the character slips out of the caesura-loop by sustaining the and function of the caesura-fill cancels the effect first violin and cello, unsettling the prolonged of the preceding MC. It represents a change of

V, dominant (thus inaugurating a new har- mind after the fact. monic progression at the precise moment when Such a situation occurs in the first movement we had expected everything to stop) and glid- of Haydn’s Symphony No. 96 in D (“Miracle,” ing forward into a reinvigorated melodic figure example 4.2). Setting aside the delicious comin the outer voices. This new figure (grounded plications that bring us to the V:HC MC point in much that has preceded it) 1s immediately (including a typically Haydnesque attempt to imitated in the second violin, and then, in m. reopen the I: HC MC possibility in mm. 48—51, 20, in the viola. In short, a renewed thematic aborted in m. 52, perhaps because the I: HC opidea emerges and pushes through the expected tion had already been used up earlier in m. 31), MC-moment (“writes over it”), canceling the we may note that mm. 54—55 drive to the new local MC implications with a new burst of Fort- structural dominant, V/V, which is attained in spinnung. Mm. 19-20 represent the point (or m. 56. This leads to the manufacture of a nearly zone) of conversion, the point at which a two- immediate V:HC MC with upward Nachschlag part exposition is renounced, and the Fortspin- on the first beat of m. 57, followed at once by nung continues by merging smoothly into a ca- an eighth rest. (As a result of the earlier complidential module beginning on the new tonic in cations, this is an exceptionally late first-levelm. 21 and expanding outward until the EEC is default MC, occurring, as we eventually learn, attained on the third beat of m. 28. The expo- some 61 percent of the way through the exposition itself ends four bars later, in m. 32. The sition, if we consider the expositional space as weak V:PAC-effect at m. 21 should not be con- continuing through m. 83.) The upbeat to m. 58 sidered the EEC: m. 21 1s a direct and relatively in the strings, with its characteristic energy-loss uninterrupted continuation of the figuration of | drop to piano, begins a recognizable expanded the preceding bars. This PAC is probably bet- caesura-fill in octaves. Its upward motion, howter understood not as concluding anything (it ever, Is non-normative, gaining rather than loslacks a truly cadential function) but as marking ing registral energy. Consequently, the caesurathe tonic-chord onset of a thematically profiled fill is made to overshoot its tonic-pitch goal in cadential module, a common feature of the con- m. 60, then to draw itself up questioningly on

clusion of Haydn’s expansion sections. 4 of A major (m. 61), and finally to abandon the As a whole, this passage from op. 33 no. “fill” function altogether with the incongru2 illustrates the procedure that we call the ous intercutting of a sforzando G’—C progresbait-and-switch tactic: Haydn baits us into antici- sion (momentarily calling our attention to 4III pating an imminent medial caesura, the hall- of the anticipated A major) in mm. 62-63. The mark of the two-part exposition, then swerves top voice in the strings of this C-major chord away from the caesura-point and switches to a (m. 63) recaptures the E, of the MC Nachschlag continuous exposition of the expansion-section (m. 57), whereupon a descending fifth progressubtype —all for the sake, one supposes, of high sion (from 5 in m. 63 through 4-3-2 in mm. generic play and the splendid exhilaration found 64-65 to T in m. 71) leads to a V: PAC in m. 71.

in sophisticated musical humor. The cadential ? of m. 67 recovers the dominant 3. In extreme cases of the bait-and-switch tactic we of A major, now understood as having been find the MC fully articulated before the plug is pulled prolonged from m. 56. The V: PAC of m. 71 on the two-part exposition. Such situations involve is no “late” medial caesura. Appearing some 82

undermining the caesura-fill that follows the percent of the way through the exposition and MC, thus refusing to permit the caesura-fill eliding with a clearly “codetta-like” C theme,

o4Vj ¥ = = i a aTFabfa llne” PR EXAMPLE 4.1. Haydn, String Quartet in E-flat, op. 33 no. 2, 1, mm.

13-28

pe ep

p Wed o \ —“ « Py SP A ae es ne SS Os eee TF TN De el™ _,sEE I PNT —spe TE “= |aa Tt ¥ wi aa F} Nene’ Pe mee Te e—y “eg — es _ [Allegro moderato]

fi bf | [f*l 2 [5 UUUUUUUUUUUUUUUULULULULULUmMLllUL CU eS, lll CU a HI

YTe Y fenuto 15 Tt tee Te = NS an 2 A — TS me Mm Te Ll ete | Dr iD AT) eee ee eee Oe Oe eehUTULULDhUmeL eee, eee ee. eee,OT

en F eS en a a QO pe eT ROY ag eteee esse J eee ineee iFT if. yh | Jf * ~*~ J FF er Um LL TLLULULULULULULULULULULULULULULULUDLULU OG, TT eH ae HT eH

ea IEF ———ee—e—e—eE—EeeeeeeeeoeoS OeeeeeeeeeHAEHF.SPF$e——— yl ert =te#HNNe GUE et

18 — mo oS me = 1 Mr } = =... = ~ — a= "Ae Fe i el TA A 0 es | si 5 | Fy. | * lm UUm™ ET o—=—e Ud CU—“e BUD e—“pm | ema’ lee gemld st Oe 7OU l—*e"F ee lL PO yeti ti— an = h 4 —_ lew eT —“=s= TT _OS—“e eT Po eee OE —e—m®e M@ FP A 5 ee ED(GO GOOeGO A A Pe DDDonn QS GO A

— ee

ne

2>” yoy TRL Aelryeee dda wd 7 i”Ssa A) es eelle ee ee nel= nel PD leeleee OA CP a a a Oa” l—Fe lel E fo Ue Ee el 2 ee ed ed el 2 ee | | ee Ny~ ee 7 2 a | «¢ po Lyfe ye me

ss i P pw KF it SarcotSA ae P CFESC.

2ee Fedoa SS ee ee GG ee a ee ceCO ee et se-|. GG do dod do tf —' ~" 6G YY , rr? ~~ nn TY yyy oy ' ~ im 6 ee ————___ 6 ery’ ot ig ee Eg

Pp pert t—“CsCsSSCSCSCSCSC“‘“(RE ee at~—FS zone still unclosed, a completely new

a straightfoward, periodic P (mm. 1—8), TR begins at codetta-like, sentential theme with C-rhetoric emerges m. 9, arriving with the HI:HC at the downbeat of m. 19 at m. 32 and is finally able to produce the needed III: into the suggestion of a light III: HC MC. Instead of act- PAC (the EEC) at m. 37.

ing directly upon that potential MC, the immediately 11. In ch. 9 we designate the equivalent situation ensuing caesura-fill expands outward (mm. 19-26), in within a two-part exposition as SC. This could involve, the process seeming to “lose” the earlier MC-effect, or for instance, the staging of a “breakdown” of S proper, perhaps seeking in vain to stabilize the whole passage followed by the subsequent articulation of a theme with into a convincing MC event. The pause with fermata at “C-character,” even though no EEC has yet been at-

m. 26, with its pronounced ninth, C,, held wide-eyed tained. over the sustained dominant seventh of E-flat major, 1s a

60 Elements of Sonata Theory

the continuous exposition, one in which the circumstance 1s, following the first PAC, the “C-like” theme can also be understood as the presence of an unusually brief succeeding modconcluding module of a broad TR=> FS space ule —one that is too brief to be considered a that had attempted, late in the game, to produce satisfactory S and one whose main function, in

a restored-caesura effect.!? such close proximity to the just-heard PAC, 1s to confirm that cadence with another one. Such cadential reiterations continue throughout much

Continuous Exposition Subtype 2: Early (sometimes all) of the remainder of the exposiPAC in the New Key Followed by tion. The result is a differing sort of mid-expo(Varied) Reiterations of the Cadence sitional expansion section, one that keeps reopening seemingly closed authentic cadences through

In the second continuous-exposition subtype varied modular repetitions (see figure 4.2). an early structural perfect authentic cadence One function of the cadential repetitions is (PAC) in the second key area (typically occur- to extend expositional space to a point at which ring around 50 to 70 percent of the way through an acceptable EEC may be sounded. In other the exposition) 1s followed not by a genuine sec- words, if the first V:PAC is not to be used as an ondary theme but by multiple, perhaps varied = MC (third-level default), then the next option

Oe

or expanded restatements of the immediately would be to have it function as the EEC. But preceding cadential module. In some instances in these situations, we are to suppose that this the “restatement” aspect is obvious; in other, _ is too early in the exposition to sound an EEC. more varied cases this may be less evident. In Thus more expositional space must be crafted most cases, though, a defining hallmark of this —by means of repetition and/or expansions of (“premature” PAC/EEC-effects, each reopened by cadential repetition)

“early” “real” EEC?! rep. _ rep.

| | | [etc. | io

(V:HC MC (multiple cadences)

may be “predicted” )

Figure 4.2. The Continuous Exposition (Subtype with Early PAC and Cadential Reiterations)

12. In this case the listener seems invited to conclude caesura-eftfect is to prepare the onset of another theme, that the “too-late’”” MC-effect tries to compensate for major section, or zone. The only thematic possibility its earlier absence, as if pretending (or hoping?) that the that remains at this relatively late expositional point is normative MC option were still legitimately open. But therefore the “C-rhetoric” option, which does arrive once the MC-restoration effect has been sounded—long “on schedule,” even though it has not been properly past the point where it normatively belongs—the S-op- prepared by a preceding EEC. The witty effect is that tion that such a gesture invites is shown to be futile. of C stepping onto the stage, blissfully “unaware” of any A “genuine” S 1s no longer available, since that option past difficulties (“All right! Here I am!”)—as if it had had been cast overboard by earlier, extended FS activ- been looking only at its “expositional pocket-watch”’ ity, which has now tracked us “too far” into the expo- and waiting for its pre-assigned moment of arrival, the sition for any convincing S. And yet the nature of any one allotted to it through an earlier generic agreement.

The Continuous Exposition 61

the cadence in question—until a sufficiently lows is not S (and the V:PAC is consequently convincing, proportionally apt EEC-moment no third-level default MC), because the music is attained. Nonetheless, the obsessive caden- consists entirely of varied repetitions of the catial repetitions themselves soon take on their dence that we have just heard. This produces own self-propelling, reiterative momentum and multiple, quasi-stuttering PACs in mm. 60, 66, sometimes have to be stopped through the ap- and 69—each overriding its predecessor as an

plication of outside compositional force. MC or EEC candidate — before the cadentialFor the analyst the initial indications that this repetition process breaks down with the three procedure is being employed involve the coor- successive diminished-seventh chords in mm. dinated appearance of four factors at the poten- 71-73. The EEC follows in m. 77, and the subtial MC and S-point. The first is an early and sequent C begins in m. 78.

emphatic PAC in the new key. (Initially one A similar procedure may be found in the might suppose this to be a third-level default first movement of Haydn’s Symphony No. 88 MC, to be followed by an S.) The second—as in G, in which the cadential repetitions of the mentioned above, one of the most important early V:PAC in m. 61 (approximately 54 percent clues—1s that the subsequent, potential S-mod- of the way through the exposition) are wittily ule is unusually brief (often four measures or overextended into a prolonged series of mulfewer): a concise, single phrase only, and one tiple, chattering reappearances—a recurring whose aims are emphatically cadential, leading cadence-idea that one cannot shake off. The at once to 1ts own PAC in the new key. The first movement of Beethoven’s Symphony No. third is that, upon examination, this cadential 7 in A, op. 92, also plays on the memory of this module may suggest a recapturing or restating continuous-exposition subtype, although in this of the preceding cadence, perhaps through a ref- case the first cadential repetition is extremely erence (however veiled) to the module that pro- brief (mm. 130—34) and what begins as its imduced the preceding PAC. The fourth is that the mediate restatement (m. 134) is deformationally cadential module (which is being assessed as a expanded, bringing us through an unexpected potential S) is often subjected to immediate rep- set of harmonic adventures before driving again etition— one that may be literal, made briefer, to an authentic cadence in m. 164. Regardless or, conversely, decorated through considerable of the subtype encountered, attempts to analyze expansion. When all of these factors are in play continuous expositions as if they were two-part

(especially 1, 2, and 4), one should normally expositions (by undertaking a fruitless search conclude that this is a continuous exposition for a second theme) can lead only to a misun(subtype 2) rather than a two-part exposition. derstanding of their internal processes. At this point one realizes that the exposition has The second subtype also affords the opportuno second theme and no medial caesura— only nity for creating delicious ambiguities between a new-key PAC with (varied) cadential reitera- the two-part and continuous expositions. One

tions. of the most artful may be found in the first

This procedure underpins the psychology of | movement of Mozart’s Quartet in D Minor, K. the first movement of Mozart’s String Quartet 421 (example 4.3). On the face of it, this would in B-flat, K. 458 (“Hunt”). Within TR mm. seem to be a clear two-part exposition, with a 42ff provide a characteristic half cadence and III:PAC MC on the third beat of m. 24 (after structural-dominant lock (V:HC), the apparent an initial dominant-lock feint in mm. 18—20) beginning of a drive toward a V:HC MC. The and a modestly contrasting, sentential S beginpossibility of that MC erodes away with mo- ning in F major at m. 25. (We had mentioned tivic repetition (and the piano dynamic, refus- this as a possible example of the third-level deing to energize), and the music gives the 1m- fault MC in chapter 3.) A closer examination, pression of changing its mind, unfreezing the though, notices that the proposed S begins with

dominant-lock, and plunging (mostly forte) an uncommonly short S-module (four bars) toward an early V:PAC in m. 54. What fol- that moves efficiently to another III:PAC on

EXAMPLE 4.3 Mozart, String Quartet in D Minor, K. 421,1, mm. 14-28

a P ae ' _ —se _—\ — — k lee. PY

l ‘, [Allegro moderato |A? Wey SS YoY A y. ee A | So 2 A 0eggAee08esnO be sas 2 =__e o_?——_83 2 = p NL y _ CFESC. oo ~~ ad | EE - £ oC s ae . he a, oO Oe TEE A—_ Ys—_ “Yes Sp —— Ff

So Ge Sen oo ee se ee ee ee OO 5$:§nNTN0N0—@0— — FE 0: 7-NT7[_—?70]

CKrESC. IP fr 20 o— f —~__

/) } (~ | JO SP ic a ~ of ey To xxkKe ee”, OR SD Ee

a # |fee ¢ Se he ee ge Aoo>frMl A > C—O _— —— A _/ ry ¥ ~ Y My Joe ae. F —_— f To ~~oe / [Se a_i — o™ Ne ee Oe =e OS 5SS eS 8 eee oe er er a, SE a ie a a aan a : ma vat oo om +Y P ee i Tae fs ete == F *D> , Jos JiJJe eT 2 tr? eye OE eS ->7->02.29"8€@a]8s0.- oOO0,--N7NTNOTcTcT*T+TY.Y"w-'Y’”!]

fey? _ _ “F>w™@-__-'-",—. _ —— TF? spe eS es

——p ee A 29 a — fF? — Yh TF? ee 2 yt) eee EE eee)

Cy TFr-72>->-"7"@#9 OR ee oeaD—_"_# go Tt FB Ot OO OO 8 oy. Oem” vaoo_NwWwWNWWWcONWMOc1/AWWY PF __— 5 DeeeOF eF>EFT_—TVvwvNwnNn ._ ... > OO 0 OT T—™*estéwv] Hey? #9—ssYen _ _voCO Ee ot 2? oF eee CS) — ___—_ oe _ 2 282__ 2 2... wvawiiz "2 ee wiesww (efee 2 ee ee

P an ‘y an ~ “y ae Sdedy [ddaced » des ey Tg eT gg OOOO Or jo 9 i>? i —?— si _}

The Continuous Exposition 63

A} 8 ri 7? ot le ee LL ~ }me ~ rom ‘crese EXAMPLE 4.3 _— (continued) —_~

Oe ee

) ———~ 9-9 —— | o Oe we

~ mY ~ 1 np fesssosey sasaas 2 Tid ttecesios the third beat of m. 28 —reclosing that cadence Difficult Cases: Incipient or Not-Fully-

“too soon” for any normative S. Moreover, the Realized Medial Caesuras bass-line of that S-module is merely a caden-

tial formula—that of the “expanded cadential We have already suggested some posprogression” leading off from I® in m. 25—and sibilities for MC deformation, including over its four bars it essentially expands the bass first-level-default MC-effects on chordal inverline of the cadence of the presumed MC (im. 24, sions and the blocked-MC effect. In addition,

beats 1—3, also a motion from I). somewhat common in early sonatas (from the From this perspective our supposed S may 1740s, 1750s, and 1760s) is the expositional sitube understood as an expanded reiteration of — ation of an apparently continuous exposition that the preceding III: PAC —in other words, as the “almost” produces an MC and second theme initial defining gesture of a continuous expo- that “almost” manages to divide the structure sition, subtype 2. The decorated reiteration of | into a two-part exposition. (In some instances, this four-bar cadential module in mm. 29-32 for example, this might signal the presence of a reinforces this view, and what follows may be light HC-effect but no unequivocally convincheard as compressed reiterations of the same ing “S” identifier in the next module). In such PAC —thus continuing to defer the real EEC cases feeling obliged to decide unequivocally further down the road. (Opinions may differ —_1n favor of the one or the other seems counabout which of these PACs serves as that EEC.) ter-intuitive, Procrustean. (One thinks of cell Is this, then, nothing but a continuous expo- division — mitosis: in metaphase and anaphase sition, subtype 2—foreshadowing that of the the two cells have begun to divide but have not “Hunt” Quartet to come? Merely to claim this fully succeeded in doing so.) In such cases it can also seems unsatisfactory, since the S-character be difficult to decide whether we should conof the first-violin theme in mm. 25-28 (varied sider the “new module” (?) in the middle to be in mm. 29—32) also seems strong. We have here an S or not, since that “almost viable” S-possia ravishingly clever ambiguity. It is as if in mm. bility may not be a fully characteristic S or may

25-29 the upper voice wishes to proceed with not be prepared by all aspects of the rhetoric S (and a two-part exposition) while the lower commonly associated with the MC — especially voice wishes to reiterate the preceding cadence. the clear MC-gap, one of the principal signals

This 1s a brilliantly crafted moment of struc- of an MC. In such cases—and with all simiture, a passage of the highest subtlety, poised lar problems of ambiguity within Sonata Thebetween the two exposition types—and par- ory—one should not force a decision into one

taking of both. rigid binary category or the other. This 1s pre-

64 Elements of Sonata Theory

cisely the kind of inflexible categorization that characteristics, and so on) help to provide the we seek to discourage. Instead, one explicates vocabulary to describe the nuances of the situwhat 1s actually there. Toward this end the more ation 1n question. By no means does every case normative categories of Sonata Theory (two- have to be slotted bluntly into either the one part and continuous expositions, medial-caesura category or the other.!5

13. Such situations turn up with some frequency in to a dominant lock (V/V) at m. 25—thus seeming to C. P. E. Bach’s early keyboard sonatas—the “Prussian” promise an immanent V:HC MC. But the music never sonatas and others. One also finds them in early sym- stops to articulate that MC, even though a new thematic phonic practice. They are not uncommon, for instance, module takes off, with an energetic forte, in m. 35. Was in the Allegro movements of Georg Christoph Wagen- m. 34 in some way an implied MC? The evidence for seil’s symphonies in and around the 1750s. In general, one is weak: the MC was never articulated as such, and Wagenseil favored the continuous exposition (subtype the Trommelbass figure keeps moving through TR and 1), but some of the rapid-tempo movements seem to into the new idea at m. 35. Here we are probably con“want” to divide into two parts. In the first movement fronting a continuous exposition that has almost subdiof the Symphony in E, WV 393, a TR leads eventually vided into a two-part format.

CHAPTER FIVE © "ISLE 0

The Primary Theme (P)

‘ie primary theme process.! (P) 1s the ideathis that be- the cisive, prepared to P orthea onset recognizable gins the sonata From point variant thereof return may indicate of a new large-scale trajectories toward the EEC and ESC conceptual rotation, which may be either full begin to take flight. With the initialimpulse of or partial (references to P only, or to P and TR P, we have taken the first step that will trigger only). This 1s self-evidently the case with the sequentially the other sonata stages; we have recapitulatory rotation, which normally recycles entered into a generic contract to carry out the all or most of the expositional materials. It may trajectory to the ESC—and sometimes beyond. also be the case in developments and codas, alThis Hauptgedanke (principal idea) may take ona though local circumstances and implications number of expressive roles: that of the emotional differ from piece to piece.

stance or referential character around which P may be launched in various ways. Many the subsequent sonata will be built; that of the characteristic topics (patterned styles) are approstructural decision to act decisively (launching priate for P-theme Hauptgedanken, and P-ideas the sonata with determination); or, especially in frequently visit a volatile succession of conminor-mode works, that of establishing the pre- trasting topics as they proceed. As Ratner has

vailing situation of the sonata-drama. noted, P-themes within symphonies were often At the same time P establishes its rhetorical styled as marches, befitting their roles as opening function as the initiator of rotations. Its first ap- pieces of eighteenth-century concerts. But many pearance signals the beginning of the exposi- other selections were also common. Ratner’s list tional rotation, the referential layout that serves of standard topics for what he called “key area as the rule for interpretation of much that fol- I” includes: hunt, polonaise, passepied, singing lows. Once the exposition 1s completed, a de- style, alla breve, brilliant style, and contredanse.?

1. Here and elsewhere, what we for convenience —and with characteristic textures rather than melodies in the

in part out of tradition—refer to as a “theme” indi- narrow sense of the term. cates only the leading musical idea (usually the initial 2. Ratner, Classic Music, pp. 222—23, which also include idea or initial-idea complex) of an expositional zone. examples of the topics mentioned. Ratner also includes “Theme” should not be understood exclusively to con- “recitative obligé” (Beethoven, Quartet in F Minor, op. note a melody, much less a self-contained and closed 95) and “waltz” (for the opening of the Eroica!). one. Many P-zones (and TR-, S-, and C-zones) begin 65

66 Elements of Sonata Theory

One formulaic device to initiate a P-zone the onset of some sort of bustling crescendo efis that of a triple-hammer-blow—three bold fect. The weak-launch option may stand on its chords, as if to awaken sonata-space with a vig- own (K. 201), but, as has been widely remarked, orous gesture — usually followed by a contrast- the sounding of a piano theme 1s especially ating topic of differing material. (As discussed in tractive after a slow introduction— something

chapter 3, triple-blows are also characteristic that the mature Haydn used to great effect.4 MC markers. They are additionally often used The weak-launch, lyrical-theme opening inas concluding signs of expositions and recapitu- vites a TR of the forte affirmation or restatelations. In other words the gesture can serve as ment type (chapter 6 ), a common strategy in an indicator of the major points of articulation § Allegro movements in symphonies (Mozart, K. within sonata-space.) Within P these chords 201; standard practice in many of later Haydn’s are sometimes largely preparatory, P® or P!° symphonies). modules, which are discussed below. The three From time to time one also encounters the chords may articulate the tonic alone, they may seemingly paradoxical use of closing formulas outline a 1—V—I gesture (perhaps with an acti- to begin a piece, although in some cases the vating 1 — 3 — 3 in the upper voice, as in the two-chord cadential idea, as a kind of preface, opening of Haydn’s Quartet in G, op. 76 no. is most strictly classified as preceding P-space 1), perhaps merely with 8— 7- 8, and so on), proper. This general idea was apparently a or they may be sounded in some other con- fairly widely recognized procedure in the eighfiguration. Other conventional possibilities for teenth century. Such an obvious displacement openings—sometimes intermixed with other of typical function must have had witty or other topics— called on fanfare-like gestures (sturdy clever resonances that were especially appealoutlines of triads) or the sounding of a musical ing to connoisseurs. The archetypal cadential idea all’unisono (in unison) or 1n octaves. formula, typically reduced to its most essential The P-theme may begin aggressively, ata progression of dominant (or dominant seventh) strong dynamic level. This is the strong-launch to tonic, was wrenched free of its normative option, and, as mentioned above, it often features function—that of concluding a larger phrase chordal or fanfare-like gestures, flashy coups or zone —in order to serve, incongruously, as a d’archet, dotted rhythms, octave drops or leaps, compact initiating gesture. triadic articulations, an emphatic, forte theme, As might be expected, the technique is chara bold, P® or P!° motto, or something similar. acteristic of Haydn, one strand of whose Witz Sometimes an initial forte basic idea is riddled featured modular dislocations —1deas in “wrong with back-and-forth interpolated piano respons- places” —and surprises of different kinds. One es—hesitant, gentler, contrasting, or question- obvious example is the purposely banal V’7—I ing replies, as in J. C. Bach, much of Mozart, (an apparent PAC-effect) at the Vivace asand so on. The strong-launch option 1s particu- sai opening of his Quartet in G, op. 33 no. 5, larly appropriate in large-scale, public, or ambi- mm. 1—2 (example 5.1). To be sure, the issue tious works that lack a slow introduction. of whether such an isolated two-chord gesture Alternatively, the sonata might set forth with may still be considered a “cadence” proper, the weak-launch option— beginning piano, either strictly considered, has been raised recently, but with an unassuming, lyrical melody (first move- au fond the dispute about such matters 1s idle ments of Mozart, Symphony No. 29 in A, K. and rests on an adamantly tenacious adherence 201; Piano Sonata in B-flat, K. 333) or with to overly exclusionary definitions.> There can 3. For a discussion of the latter see Wolfgang Gratzer, 4. As mentioned also, e.g., in Rosen, Sonata Forms, rev.

“Mozart, oder? Der Unisono-Beginn in Streichquar- ed., p. 243. tetten der Wiener Klassik: Fragment zu einer Poetik 5. The argument contra the term “cadence” in such situdes musikalischen Anfangs,” Mozart-Jahrbuch 1991 [1991 ations—and several others besides—has been made by

Salzburg Conference Report] (Kassel: Barenreiter, Caplin, “The Classical Cadence,” especially pp. 83-85

1992), pp. 641-49. (dealing with the famous, only partially parallel case at

The Primary Theme (P) 67

Dy , FT X La ee ee eee ss Tw le wT ee ea ee wl t Ak Gy a YOY SN o? 1 1 @ aad : i@er op a 00 aoe eT ea aPa iEgy EE OO pOi|yrSS ee —— td 6|7 yor 7 ————— Sf [) 4 ee ee — gm ee PF Ca A A A ray 2 @ go | UTae Rm ee Pr —, EXAMPLE 5.1 Haydn, String Quartet in G, op. 33 no. 5, 1, mm. 1-10

Vivace assal ror oy {y_% | 1 +) oe ge +g “*@ |e @ © @ |g “*s¢e -—~ _eae EO4ee eee

be little doubt that— “cadence” or not— Haydn at the end of the movement, where it is made to expected his listeners to understand the open- assume its most natural function.® A more exaging here as a witty “closing formula” that has gerated version of this cadential-ending-as-be-

been transferred to the apparently “wrong” ginning may be found in the Quartet in C, op. spot of the piece. As the movement proceeds 74 no. 1, mm. 1—2 (sounded virtually outside Haydn resituates that cadence formula in a more of any tempo or pulse, the dominant seventh is proper location—both at the end of the initial — provided with a fermata, with a second fermata

thematic phrase, mm. 9-10, for example, and separating the tonic chord from P-proper, the the opening of the trio in the third movement of Mo- op. 33 no. 5 would be one such situation—for hearing zart’s Symphony No. 41 in C, K. 551). In Caplin’s view this moment as a separate, cadential displacement (as an such a two-chord succession may have “cadential con- isolated two-chord cadential formula presented starkly tent” but not a “cadential function,” because “it cannot asa “bare fact,” shorn of its normative functional role of be construed to end a formal unit,” which is a sine qua larger-phrase closure) is the precise point of this initial non in Caplin’s definitional apparatus. (No chordal suc- gesture. (The situation in the K. 551/111 trio, we agree, cession rises to the level of the term “cadence” unless it is more problematic.) Another instance —similarly disdisplays both cadential content and phrase-ending func- allowed as “cadences” by Caplin (pp. 90-95) — would tion.) Needless to say, most cadences do indeed serve as be the succession of emphatic V—I chords (often withthe “last event” of a “larger formal unit” or “phrase” of out preceding tonic and predominant chords) somea handful of measures (pp. 57—58) that often display a times found at the ends of expositions. We suspect that tonic—predominant—dominant-tonic motion. But what most listeners intuitively hear such chords as recrafted, is true of most cadences within this style need not be emphatic reiterations of the final portion (the “cadence” elevated into a rigid criterion to apply to all of them. element proper) of a more complete cadential progresWithin the flexible sonata style, it is possible that in sion heard just before them. To strip them of legitimate special adaptations and under special, quasi-formulaic cadential status (in favor only of the “postcadential” role circumstances, some emphatically staged, two-chord of “codettas”’) is counter-intuitive. See also the discusprogressions might be heard as invocations of the “ca- sion in ch. 9. dence” proper and hence, pari passu, might be reason- 6. Inthe case of op. 33 no. 5 the opening two bars comably regarded as “cadences” (which in this case, perhaps bine features of the “brief in-tempo introduction” and more precisely, could mean “cadence-alluding formu- the P® or P!° concepts (discussed separately below in the las’). “Cadential” openings to such works as Haydn’s subsection on “P® and P!-° modules/themes’’).

68 Elements of Sonata Theory

Allegro moderato, in m. 3), while that of the of Schumann’s Symphony No. 4 in D Minor, Quartet in D, op. 50 no. 6, mm. 1-4, 1s slightly Franck’s later symphony in the same key, or more thematically elaborate. Unusual variants Mahler’s Symphony No. 1 in D—even though and expansions of the technique abound: recall, those introductions might not exemplify all the for instance, the (post-introduction) P-theme of features of the source-models. the first movement of the Symphony No. 94 in P-theme characteristics also vary according G, “Surprise.” Similar effects, filtered through to the prescribed performing forces of the work. different musical sensibilities, can be found here Certain kinds of P-themes are characteristic of

and there in Mozart and Beethoven. keyboard sonatas; others of chamber music; othWhen a slow introduction precedes the faster ers of orchestral pieces. In 1979, for example, P-theme, one normally expects to finda clear — Laszl6 Somfai submitted Haydn’s keyboard soseparation between them—a fermata or some natas to an extensive stylistic survey. Haydn’s other unmistakable conclusion to the introduc- first-movement opening ideas, he found, distion. This renders the onset of P independent played a wide variety of shapes and topical styles, and distinct. In intensely dramatic orchestral changing throughout his career. At least in this works from the 1790s and thereafter, however, repertory P often begins with a strongly emwe occasionally come across either a “run-on”’ phasized dotted-rhythm idea. This idea may be introduction without a clear pause or an intro- immediately contrasted with a cantabile theme duction ending with an accelerando or other cu- (Sonatas in D and E-flat, Hob. X VI:19 and 45); mulative effect that gains energy and merges the dotted-rhythm itself may be carried out in directly with P and that may also cross over the extenso (“monochrome” rhythm) or subjected to new tempo marking in the process.’ In the latter characteristic keyboard embellishments (Sonacase P springs forth from the wind-up accumu- tas in B-flat, C, F, C, and B-flat, Hob. X VI:18,

lation of the introduction instead of initiating 21, 23, 35, and 41); it may be in the French its own spark of momentum, and the exposi- Overture style, implicitly subject to overdottional rotation can begin several measures into ting (Sonata in D: Hob. XV1I:14); or it may be the Allegro tempo. One instance had occurred more complex, march-like, or wide-ranging in in Cherubini’s Overture to Les deux journées compass (Sonata in E-flat, Hob. X VI:52). (1800), but the procedure is most familiar from In Somfai’s view, sometimes Haydn’s keysome of Beethoven’s orchestral works: the first board P-theme, especially in minor-mode sonamovement of the Symphony No. 4 in B-flat, tas (C Minor, Hob. XVI:20), 1s more “abstract”

op. 60 (wind-up beginning in m. 35; Allegro (less purely keyboard-oriented), closed, and vivace in place at m. 39; exposition proper, P, at of an “instrumental character.” Other movem. 43 with expositional repetition marked at m. ments begin with a “striking instrumental ef45); the Leonore Overture No. 1, op. 138 (wind- fect” manifestly idiomatic for the keyboard: an up beginning in m. 37, which A. B. Marx be- initial rolled, broken, or asserted single chord lieved in 1859 to be the gathering of Leonora’s propelling the sonata into action (two sonatas courage and steely resolve, der Entschluss zu ihrer in E major, Hob: XVI: 13 and 22; Sonata in F, That;® Allegro con brio in place, m. 42; expo- Hob. XVI:29); or “bare-bones” staccato themes sition, P, at m. 58); and the Egmont Overture, that Somfai judged to be reminiscent of Scarop. 84 (Allegro in place, m. 25, itself initiat- latti (C major, Hob. XVI:50). Most of Haydn’s ing the wind-up; P proper shot forth at m. 29). opening movements were in duple time (4/4 or These early examples are among the sources for 2/4). Those in 3/4 that “are not stylized or acthe later-nineteenth-century generative intro- celerated minuets are relatively rare” (Sonata in ductions—to the first movements, for example, D and two sonatas in E-flat, Hob. X VI:24, 28,

7. The adjective “run-on” in this context comes from 8. Adolph Bernhard Marx, Ludwig van Beethoven: Leben James Webster, Haydn’s “Farewell” Symphony and the Idea und Schaffen (Berlin: Janke, 1859), 1:337. of Classical Style, pp. 162—65.

The Primary Theme (P) 69

and 49). Somfai also assessed P-themes in terms also be unfolded as a multimodular succession of symmetry and asymmetry: “The asymmetric comprising several ideas. (See the section on model is earlier and richer in variants. Perhaps it “Thematic-Modular Designations.’) is more characteristic of Haydn; at least 1t occurs Some P-ideas display the characteristics of

somewhat more frequently.””? periods and sentences at different structural levels. We might, for example, come across a period whose antecedent (and whose consequent as

Structure well, if the period 1s stated completely) 1s articulated as a smaller-scale sentence (a sentential an-

The structural type that the composer selects for tecedent). The unusually broad, antecedent-like P is no neutral choice: it is an important factor —_ phrase (mm. 26—39) that opens the first-move-

in the personality and drama of each individ- ment exposition of Mozart’s Symphony No. ual work. P may be shaped as a simple period, 39 in E-flat, K. 543, can serve as an illustraa sentence, a single phrase, or a more complex tion (example 5.2). Here the large antecedent 1s structure. For the sake of convenience in de- _ structured as a compound sentence ending with

scribing these formats, we use, though some- a half cadence: the presentation (mm. 26-29; times in adapted ways, most of the terminol- 30—33) is followed by a continuation module ogy advocated in Caplin’s Classical Form for the (mm. 34—39). In this case the antecedent is folstructure of periods, sentences (including the lowed by a sentential consequent (mm. 40-54), terms “presentation” “continuation,” and “con- whose I:PAC 1s elided with the onset of an in-

tinuation=cadential”’), and their hybrids.!° In dependent, forte transition. In other instances general, sentences are more active, more rest- the parallel consequent may dissolve into TR less, more forward-driving than periods, which _ rhetoric before attaining its expected cadence: tend to be more static and symmetrical. P might 9. Laszl6 Somfai, The Keyboard Sonatas of Joseph Haydn: p. 260, n. 5); a phrase is “minimally, a four-measure Instruments and Performance Practice, Genres and Styles unit, often, but not necessarily, containing two ideas” [orig. in Hungarian, 1979], trans. Somfai in collabo- (p. 256; cf. p. 263, nn. 4, 11). What Caplin calls a phrase ration with Charlotte Greenspan (Chicago: University we would often call a subphrase or module—although

of Chicago Press, 1995). This data, and other infor- “module” is intended to be a flexible term covering any mation not mentioned above, may be found in ch. 15, of a number of small building-blocks within a work, “The Primary Theme,” pp. 237-61 (“monochrome,” ranging from each of Caplin’s two smaller ideas, to any pp. 243-44; “abstract” and “instrumental character,” slightly larger unit without strong inner contrasts, to, at p. 247; “striking instrumental effect” and “idiomatic,” times, a consistent “phrase” itself. pp. 251-52; “bare bones” and Scarlatti, pp. 253-54; The differing understandings of the term “phrase” “are not stylized or accelerated minuets,” p. 254; “The (with or without the necessity of a cadence) are rooted in

asymmetric model,” p. 241). differing analytical traditions that we need not explicate 10. N. 14 below suggests one way in which our view here. We might note only that the varying conceptions of a sentence is more flexible that that of Caplin. (We of the phrase invite differently nuanced understandings also find the concept of “hybrids” to be problematic, of larger musical motion. One of the clearest advocates of although for the sake of convenience we have provi- the necessity of the terminal cadence is William Rothsionally employed that term in this book.) Still another stein, as in his book Phrase Rhythm in Tonal Music (New difference between our descriptions is in our diverg- York: Schirmer Books, 1989), p. 5: “A phrase should be ing conceptions of what constitutes a “phrase.” We re- understood as, among other things, a directed motion gard the normative “phrase” as a more or less complete in time from one tonal entity to another; these entimusical thought involving motion to a cadence. The ties may be harmonies, melodic tones (in any voice or presence of a cadence at its end—except, perhaps, in voices), or some combination of the two. If there is no deformational or other rare and extraordinary instances tonal motion, there is no phrase.” Oversimplifying the mat-

(which do sometimes occur)—is central to our pre- ter considerably (to confine ourselves for practical purferred view of the term. Caplin’s definition is cast in a poses only to discussions carried on in the past fifteen way that does not require a cadence. In his system the years), we may acknowledge a Caplin-Rothstein split word “phrase” 1s “a functionally neutral term of group- regarding this matter—one in which our preferences ing structure [that] refers, in general, to a discrete group lie with Rothstein. See Darcy, rev. of Caplin, Classical approximately four measures in length” (Classical Form, Form, in Music Theory Spectrum 22 (2000), 122-25.

70 Elements of Sonata Theory EXAMPLE 5.2 Mozart, Symphony No. 39 in E-flat, K. 543, 1, mm. 26—57 (melodic line only)

ST

Tn a po ”wo (@iF“oe tl SP p y§ 9 7 #0 i Allegro

26

P

Vin. 1 fey Pope ep pp 33

Nee

ey: _—_—— | yy 6h i st -f-+—_— 40) oFPoy ; ——_. " __—

eS (4 — (4 — ewe ty Ba Ne A PL LS YD (SGD GO (DG (O P

46

tt

nC PE sel, DS , § = ee A 9 A DS SE ON ee OE ON Oc

52

St

in this case we consider the TR to begin with tended to suggest an untexted song. Quite often

the consequent (chapter 6). such a song implies an “absent” text of around Another complex structure (a rarer one) 1s eight poetic lines in length. (A typical scheme that of a rounded binary form: ABA', with a is2+2+2+ 2 lines, perhaps responding more harmonic interruption at the end of the B sec- comfortably to the schema, aa'ba''.) We identify tion. (The A section may itself be structured as such a cantabile “song without words” as a lyric a period, sentence, or hybrid.) The presence of — binary, a subset of the binary form proper.

such an elaborate shape can produce not only When these more complex forms are closed a breadth or vastness to the P-idea, but also an (when their final modules are brought to a peruncommon roundedness or completeness for fect authentic cadence), one might even find loit. In some cases, it may happen that the onset cal repeat signs within the theme. Repeat signs of the last limb (the A' reprise) soon dissolves within an opening melody are more often siginto transitional material (TR of the dissolving- nals of a rondo theme and hence predictive of final-element type; we regard TR as beginning either a pure rondo or a Type 4 sonata (sonadirectly with that A', even before the moment ta-rondo). Still, things are not always so clear. of literal dissolution). This occurs in the mas- In the sparkling finale of Mozart’s String Quinsive opening F-minor paragraphs of Cherubi- tet in D, K. 593, one finds a standard sonata ni’s Overture to Médée (initial A ending with a form (a Type 3 sonata) in which P is a rounded half cadence, m. 20), in the first movement of | binary theme with local, interior repeats. The Schubert’s Quintet in C, D. 956 (B in m. 26, A' editors of the relevant volume of the Neue Moin m. 33), and in a few other pieces. For herme- zart Ausgabe misconstrued this and introduced neutic purposes it is sometimes helpful to notice without comment a nonexistent, additional rethat the ABA' structure, when more vocal, lyri- peat sign after P proper (at m. 37), thus providcal, or cantabile in character, might have been in- ing directions for an impossible, hopelessly mis-

The Primary Theme (P) 71

construed form, one whose expositional repeat on to P? only after the first PAC has been atreturns only to TR.!! The nineteenth-century tained. (Obviously, if what follows merely reMozart edition had provided the correct repeats. iterates or slightly varies the music leading into A similar interior-repeat situation, thankfully the cadence, that would be a full or partial repwithout the editorial error, may be found in the etition of P!, not a P? module in its own right. Allegro vivace finale of Schubert’s Symphony The same is true of modules within TR, S, or No. 5 in B-flat, and in several other works. On C.) Any P-module labeled P? should have been rare occasions one might find rounded-binary preceded by two differing thematic spans, each P themes within Type 3 finales in which the ending in a PAC. Thus the integer 1 means: internal repeats, instead of being notated with “belonging to the first perfect-authentic-carepeat signs, are fully written-out, with perhaps dential span” or “preceding the first PAC.” The only minor variants, if any. Examples occur in integer 2 means: “belonging to the second perthe last movements of Haydn’s Symphonies No. fect-authentic-cadential span” or “following the 76 in E-flat and No. 77 in B-flat and of Mozart’s first PAC.” Note that within P it is possible to Symphony No. 40 1n G Minor, K. 550. (Repeat have no PACs at all (tonal underdetermination sions within sonata-rondo P themes are consid- of P): the music can dissolve away (via a TR)

ered in chapter 18.) from that potential before an authentic cadence is reached. In that case the mere letter P can suffice to designate pre-IT'R space; or, if it 1s useful

Thematic/ Modular Designations: Numberings to subdivide the modules of the P-span further, one could designate it as P! with some added We have already established our letter-designa- indicator of the modules’ locations or functions, tions for the four potential zones (action-spaces) perhaps by means of the decimal designators deof a standard expositional layout, or first rota- scribed below. Yet another possibility, in each of tion: P (primary); TR (transition); S (second- the zones, is a preliminary or otherwise prepaary); C (closing). For many purposes it is suf- ratory “zero-module” (P° or P!-°, for example). ficient to indicate these zones without reference This is a more complex issue treated separately to any of their inner subdivisions, in which case further below.

the letters alone are appropriate. It sometimes Many P-zones (and other zones) are mulhappens, though, that one wishes to single out —_ timodular: they consist of smaller, sometimes individual modules or small subsections within contrasting units that we might want to single

these spaces. In order to indicate constituent el- out for any number of purposes. We may be ements within P-, TR-, S-, and C-space we use interested in parsing the theme itself. We may exponential numbers—superscripted numer- _— wish to suggest a functional or hermeneutic role als—such as P!, C2, and so on. These numbers for one or more of the P-modules. Or we may

are assigned according to certain principles. simply wish to be able to refer to that isolated Our first principle 1s that in none of the zones module later in the piece— perhaps only one or do we notch a superscript integer upward from two modules of P are actually used in the de1 to 2, from 2 to 3, and so on, unless a perfect velopment. How might P or P! be subdivided? authentic cadence has been sounded ina key __ If Pisa period, one might wish to designate its appropriate to that musical space. This is advis- two parts as Pt and Peons, A complementary able because so much of Sonata Theory is con- _ labeling plan could be used for the sentence. Or cerned with the attainment or nonattainment one might wish to adapt some of Caplin’s term1iof cadences. Music designated as P! will move nology from Classical Form, “basic idea” (b.i.),

11. See this measure of K. 593/iv in Mozart, Neue Aus- score (the second manuscript page of the finale) — with gabe simtliche Werk, Serie VIII, Werkgruppe 19, Ab- no indication of a repeat at this point—may be conteilung I: Streichquintette, ed. Ernst Hess and Ernst sulted in Ernst Hess, “Die ‘Varianten’ im Finale des Fritz Schmid (Kassel: Barenreiter, 1967), p. 134. A Streichquentettes KV 593,” Mozart-Jahrbuch 1960/61 photograph of the relevant page of Mozart’s autograph (Salzburg, 1961), pp. 68—77, facing p. 77.

72 Elements of Sonata Theory

“contrasting idea” (c.1.), and other descriptions, with apparent double medial caesuras) in the as part of one’s superscript designation of the middle of an exposition. Here the convention is modules. We advise flexibility along these lines. to identify the differing modules as TM!, TM?, One should use the subdesignations that seem and TM, even though in most cases the whole most appropriate to the analytical concerns TMB covers only a single cadential span (see

brought to bear on the themes themselves. chapter 8). The other exception concerns the In general, we refer to modules within a per- special case of “zero-modules” (especially P®, fect-authentic-cadential span by means of deci- S°, and C®), a topic dealt with directly below. mal designators: 1.1, 1.2, 1.3, and so on. Such Proceeding from a zero-module to the ensuing designators can furnish much information at “1” theme implies nothing with regard to the a glance. P!-?, for example, may be explicated existence or nonexistence of a PAC between the as “within the first perfect-authentic-cadential two modules. span of the P-zone (prior to the first sounding

of a I:PAC, if a CxIsts at all), the module that “Zero-Modules”: General Considerations follows the immediately opening idea.” P?:! means: “still in P-space, the next P-module af- It is not uncommon for individual zones— espeter the first PAC.” It should be underscored that cially P and S—to begin with music that, even the practice of decimal designators is no rigid while opening that zone, seems preparatory to system but merely a conceptual tool to be used a more decisive (or more fully launched) modby the individual analyst as he or she sees fit. ule that follows. This aspect can take on differMost flexibly understood, the numerals follow- ent realizations, some of which are “thematic,” ing the decimal need imply no more than that some of which are not. One might find: an i1nthey tally modules that we wish to notice for — troductory vamp or accompaniment figure; an one reason or another. Often, however, they initial group of “set-apart,’ emphatic chords; a will correspond to certain thematic features of quasi-fanfare motto, sometimes all’unisono, that the period or sentence, including “basic ideas,” “clears the way” and then proceeds onward “contrasting ideas,” the beginning of the con- to contrasting material; an obvious anacrusis tinuation of a sentence, and so on. (Whenever module or other preliminary module; a thepossible, we prefer to identify the beginning of | matic module that has not yet fully stabilized the continuation of a sentence with a 1.2 des- over a root position tonic (especially within the ignation, with the presentation modules being S-zone); and the like. A zero designation— P°, labeled as 1.1. But even here, given the details S°, and (the rarer) C°—1ndicates the results of of individual cases, absolute consistency must an interpretive decision that proposes either that sometimes be sacrificed.) Differing analysts and the module at hand displays an overt preparaanalytical purposes might arrive at differing tory function (often in the sense of “get ready!’’) decimal designations — which is generally fine. or that the initial module conveys the sense In most cases, however, the predecimal integers, of something “destabilized” or not yet fully registering perfect authentic cadences, should moored to tonic root-support. If the analyst deremain constant from one analysis to another. cides that such an introductory module is not as fully separate from what follows to merit the

. . . . “zero” label per se, a lighter alternative is the use

Exceptions in ‘Thematic Numberings of a 1.0 label: P19, S19, ClO, the next module of As indicated above, within the P zone and in which, still more decisive, would be understood all other zones it is normally only when one as 1.1. At issue here are only degrees of strength moves past a perfect authentic cadence onto dif- and analytical nuance: often either the “zero” fering material still in that zone that the integer or the 1.0 label will be workable. In either case, 1 gives way to 2, 2 to 3, and so on. There are the zero-module will lead directly into sometwo exceptions to this principle. One concerns thing more secured and normative for that zone. the convention of labeling the constituent parts Zero-modules rarely last longer than a few bars of a trimodular block (the TMB, associated (although a few broader exceptions do exist) be-

The Primary Theme (P) 73

fore calling forth, moving into, merging into, it is also true that according to the generic conor otherwise precipitating the relevant P!, S!, tract in force this P-key is under obligation to or C!. In highly exceptional cases, one might be lost or abandoned (for the keys of S / C; for confront a situation in which the “zero” passage the keys of the modulatory development) before is itself multimodular. Within the P-zone, for being reattained more securely in the recapituexample, this could suggest a P?.!, P9.? sequence lation. The tonic key at the opening exists as a

of numberings before P! is reached. proposition to be undermined (or unfolded) on Zero-modules are not musical ideas that stand the way to reaching a higher level of closure. To outside of the zone proper. A P®- or P!-°-module be sure, the vigorous postulating of a diatonic launches the P-zone and therefore belongs to collection within P leaves no doubt with regard

P-space (and hence to expositional space); to the local tonic being asserted. Nevertheless, an S°- or S!-° module launches the S-zone; erasping the still-provisional nature of the tonic on those rare occasions when it can be found, at P is central to the hermeneutic aspect of Soa C°- or C!.°-module launches the C-zone. In nata Theory. It is an essential component of its most cases (though not all) the zero-module will understanding of the raison d’étre of the trajecnot be separated from its “integer-one” succes- tory of the sonata as a whole toward the ESC. sor by a PAC. Nonetheless—as an exception (See chapter 11 for the concept of tonic pres-

to the general principle of numbering— one ence.) notches up any P® number from zero to one at On this understanding, it becomes important this point, in large part to indicate that the pre- to notice whether P 1s tonally open or closed paratory or “not fully opened” character of the before it proceeds into TR. Open, closed, and music has now taken on a more decisive and multiply closed P-themes cannot be regarded as

normative aspect. expressively neutral choices. Obviously, a P that The zero concept covers a wide range of pre- is brought to a single I:PAC (or IAC) has fully paratory or otherwise quasi-“‘tentative” initiat- declared its tonic and stabilized it, at least loing functions within these zones. Since the zero cally, with a cadence. Here the tonic proposiconcept indicates a function, not a thematic or tion is fully carried out and may be considered modular type, zero-modules can differ widely adequately determined (or, more simply, norfrom each other in character and format. Some mal). It often happens that the I:PAC moment standard types of P® modules are not character- is elided with TR —frequently a forte or aggresistic of S° modules, and vice versa. One should sive TR, as occurs in Mozart’s Symphony No. not expect the ones to “sound like” the others. 39 in E-flat, K. 543, m. 54 (example 5.2). In Apart from these generalized remarks intend- such cases P 1s articulated as a completed, locally ing to introduce the zero concept broadly, one centered entity before embarking on the process should consult the separate discussions of P°- of destabilization.

or P!°-modules (below), S°- and S!°-modules In other situations P proper never attains (chapter 7), and C°- and C!.°-modules (chapter a I:PAC at all: it is destabilized with TR ma9) in order to deal with specific instances and terial before reaching its own tonic closure.

typical cases. This is the case in many instances of TRs of the “dissolving” type: [Rs that begin as a closing element of the P-idea (a consequent, a sen-

Tonal Under- and Overdetermination tence-continuation, a concluding final A’ element, and so on) but that redirect the expected As elsewhere in sonata-space, cadence attain- —_ thematic closure into TR-Fortspinnung or other ment or nonattainment within P isacrucial as- —_— recognizable devices. Such P-themes should be pect of its character. One structural function of regarded as tonally underdetermined: their tonics P is to set forth an unambiguous tonic that will are clearly understood but not secured with an nevertheless be attained as a fuller, more sta- authentic cadence. Celebrated examples include bilized reality at the point of the ESC. While _ the large-antecedent openings (with dissolvingP-themes are rarely if ever tonally ambiguous, consequent TRs) of Beethoven’s Piano Sonatas

74 Elements of Sonata Theory

in C, op. 53, “Waldstein,” and in F Minor, op. of the first movement: tonic PACs at mm. 14,

57, “Appassionata.” 22, 32, followed by a prolonged tonic pedal;

Of particular interest are instances of the sudden shift onto vi in m. 36.) Closely related other extreme: P-zones that consist of multiple examples in Mozart may be found in the openmodules, several of which end with a I:PAC or ings of Piano Concerto No. 14 1n E-flat, K. 449 IAC. In such cases the tonic is locally overde- (quasi-Sturm-und-Drang 'TR-displacement in vi termined, and the subsequent tonic-phrases after in m. 17 following redundant cadences) and the

the initial PAC can take on a number of dif- Quartet in D, “Hoffmeister,” K. 499 (urgent ferent characters. In the most typical instances TR-shift on vi in m. 23). It recurs also in the the varied modules of the overdetermined P are solo exposition of the first movement of Piano sounded at a relatively quiet dynamic. When Concerto No. 15 in B-flat, K. 450 (sudden shift there 1s a sudden forte or tutti-affirmation out- to vi, G minor, in mm. 86—87, the beginning burst elided with the downbeat of a I:PAC, it of that exposition’s TR). ‘The same procedure 1s is usually preferable, on rhetorical grounds, to also locatable in Haydn, as in the opening of the regard that as the onset of a TR (which itself | Quartets in D, op. 20 no. 4 (aggressive move might be overdetermined with regard to in1- onto viin m. 31), and in B-flat, op. 50 no. 1 tial-tonic cadences, as discussed in chapter 6). (explosive augmented-triad TR-ignition in m. Above all, what is clear is that the multiple 28). On the other hand, while Haydn’s QuarI:PACs within P produce an effect of local re- tet in B-flat, op. 64 no. 3, overdetermines the dundancy. In turn this suggests an unusually tonic at the outset through a non-sequitur P? powerful tonic field at the outset of P, one from interpolation (mm. 8—17), the onset of TR (m. which the narrative subject 1s either obstinately 18) comes as less of a shock, more of an efficient unwilling or (tragically?) unable to depart. Or —_— returning-to-business.

it may be that for purposes that will become Another important trope of I:PAC overdeclear as the sonata proceeds it will have proven termination occurs more frequently in lyrical important to lay down an unusually forceful lo- slow movements. The second movements to cal tonic at the beginning. All such situations Beethoven’s Piano Concerto No. 3, op. 37, and invite a hermeneutic interpretation. (For more his Symphony No. 6, “Pastoral,” op. 68, furnish on the terms “under-” and “overdetermination” the paradigmatic illustrations. In both cases P

see chapter 11, n. 32.) first unfolds as a lyrical theme closed off with

Within Allegro compositions perhaps the a structural I:PAC: in op. 37 we encounter an most common trope of P-overdetermination expansive multimodular theme that ends deciinvolves the suggestion of a temporizing, smug, sively with the tonic cadence at m. 17; in op. 68, or static reluctance to get the sonata moving off — the Szene am Bach, we find two circular statethe initial tonic. This holding all-too-fast to the ments of a sentential phrase, each concluding tonic-spot is typically corrected with a strong, with a I:PAC, in mm. 7 and 13. And in both we

forte TR-jolt that suddenly shifts the overdeter- find that once the essential structure of P has mined tonic onto a different tonal plane —as been attained, we are released into a specially an impatient non sequitur (“Enough! Let’s get highlighted, postcadential, codetta-like phrase on with it!”?)—and instantly redirects the in- or Nachsatz of uncommon satisfaction (in the decisive stasis into linear, forward motion. The concerto mm. 17-21 with cadential reiteramost well-known case occurs in the first move- tions, mm. 22—24; in the symphony— shown ment of Mozart’s Piano Sonata in F, K. 332 (ex- in example 5.4—mm. 14-15, 16-17, with ample 5.3). Following a soon-cloying succes- bobbing reiterations of the tonic, mm. 17-18; sion of self-satisfied tonic PACs (mm. 12, 20, in op. 68 the phrase also resurfaces 1n differing 21, and 22), the TR-jolt 1s provided with the contexts throughout the movement). locally D-minor (v1) Sturm und Drang outburst Such luminous P-codetta moments cry out at m. 23. (Compare this with the Allegro assai for an interpretation beyond facile labeling. finale of the same sonata, which in several ways They are particularly suitable, as here, for conrecomposes and rethinks the structural details veying moments of heightened reverence or

A

f ee Piano | P L a a EXAMPLE 5.3. Mozart, Piano Sonata in F, K. 332,1, mm. 1—26

EP” Aaes RSDS (Oe iFPN A Se a 2” ne«=| ee BFR “=e Allegro

ee eee een | | ee | A eeftie, Oe ImT QO OS OO ee” efi | UT eeTe ly ry? [KR Tsgma He

i| fFOE I Topo acc|] J;coe coe hA | | |DT fT fT Jjce PTT I[ JT 7; [fo |DS | | TT ~~ [ JT four *) ow» P | ~~ | Pea | + {| 7-T 7TifT,a | T_T [ [| | [yy gg | pl px |

5

i0 .

cy”. mh SS \ \ | QSOS ee gm GK —=—p i ae er a ee ee eS I ee ee ee a es el RO A a ee

p ee : a Tas St ee et Cee ce es Bs ey «Sr CN ee (QO

15 — aeoFe SD | ee fi by ~~ F&F » | Tm | | i» | | | fee | | 6 6 ge hh hTl Uhd|hmmmmLCUTUU

/

A CS CS FS 0 CC, OC GC 6 Pp a im t—(C | tC O—K a |j~Ff ¥ fywey [| a YT gg ga fF | | ¢e@ fa &: 3 | F § ££ ££: | | hv ¢ LS AD Ss = a A Ose OE ee Lee Doe eee

ee ee eeQO no =| sf. bteeee |ee| ee |eeJT |ec | ee | Tg Ec ee ee aaQO ned A aee.eeieeaa ee esee ee irra ea aaa |SeS eTedSS yf RwaA.sll) eeeeC =eeeeee ee es = eees ee nn ee RP” AS es es QS I Ok QO QO GD (OG S(O (I (OO (OS

pT

—¥ Fi fySy ef Ec ee es ee es| ss—_Z = SSTB ee ee = —_> _ oS

20

|fifra’Se Foe 2| fy || fg ff iee | | #¢ ee ;@=e #|Fr # ffsirPJO | OS #7{ . TT FFO_o |te-F#iJ{ lOU™T i tk¢ ee a ee eeiFFT

eS se A 0 0 SSeeRS

f\ aN 7 . . ee ee oe ee a i a rr ee eecseea

= -, -,

24 Y

Fk Seeea ee ceaBe eeaace A ee A DM De ea ae ae rs esASa=skSG ————— es a

er ee ee Cs GO GORY QO, QO ise l CUti“r0—ACA0AOUOwWwWwvWwW’w’---0”"n0'07-0ATTLATAT{-{jSN>NT?—@—@?—@?—?@?@?@—?@™@¢€9t:—I—r—lT1]]

aTe N07 SE ZSI 0—— CCeee ddlleA tT --—-—

|oyeJ FEJ wAOmMm246-™"—VVWOWVWWVWwWwWwy —_—-_ — - e I od

FF 5 —— TT t.—’".""-"\-[(._""[_ icles ie

29 SF; 2; TSP

pb et os? yo —] JJ! wnmnmtmvtwtwvwTv}

ype oo séa SiSS SS eT Per 4. LO —_ > F$+-TCThL_{te-w (et A——>— ee TSasi eS Oo

79

80 Elements of Sonata Theory

ally include even more striking complications. Mozartian “Loops”: Sometimes one encounters more obviously ge- A Specialized Variant of the Sentence neric TR activity (standard forte gestures, often

at the downbeat of a I:PAC, which one would One commonly encountered type of Mozarnormally take as a TR-launch) before the I: HC tian theme begins with a short module (two to tasi-MC effect. When this occurs—as “4 the six measures) — usually closing with a cadential fet movement of Haydn's Symphonies No. 96 progression— that is either elided or flush-jux1D. “Miracle.” and No. 101 in D “Clock . taposed with a repetition of itself before moving alonc with the Finale of Mozart’s ¢ wn hon forward into differing material. One touchstone No Uy in C, K. 551, “Jupiter” —the impression instance occurs at the opening (P-theme) of the

oe one Pi K. —. — i 2 6). ;

given is that of a TR-space decisively entered 3. onan C, ° (example -6) Le

, 1t] f th h 3-5.

following a clearly delimited P, then aborted oo Beate & Compact propecssion Sens

through a surprising or witty change of mind with a I:PAC in m. 3, simultaneously setting off

We refer to the section from the beginning of ‘ repetaon : ~ eam rare’ bars, mm. 3 °

the exposition through these brief, aborted TR The impression given is that of circular repetitextures (“false” TRs before the I: HC caesura tion, a “loop” of self-replication that could coneffect, then the I:HC proper) as a com lex orand tinue indefinitely unless something intervenes

Tec - eg. . h fth ;

antecedent | pee Pees to break the pattern. That “breakout” occurs at

The additional adjective indicates that the m. 5, at the moment ome theme s second PAC “a ntecedent-effect” may comprise more than Here the music breaks free of its initial circuone phrase (more chan one -roiog cadence larity and shoots forth with a differing idea in a which would not be the case with an anteced- more clearly linear vector. A slightly more expanded example may be found 1n the P-theme ent proper) and that, as a result, the TR-effect .

oo, . in E-flat, K. 271 le 5.7). Mm. 1-

before the I: HC had been more convincing or of the first movement of Piano Concerto No.

elaborate than in a normative grand anteced- 7 an . at / (examp 7: 9-7) m 7 4 ent (which is by definition an expanded single comprise a characteristically brief, Mozartian phrase). What follows the I:HC MC-effect is double-idea: an opening, annunciatory, triusually to be regarded as the real TR, which adic gesture all’unisono (mm. 1—2) followed by

can take the character of any of various TR a responding melodic module that ends with tvpes. The MC moment at the end ofa com a cadential progression and PAC at m. 4. This alex erand antecedent is conceptually close to four-bar idea 1s then elided with a literal repetithat ofa THC MC-declined situation. Can one tion of itself (mm. 4—7). In this case the exit distinguish between them? Assuming that the from the circular repetition is handled with a distinction promises to have hermeneutic value meres axed smoot a Instead of beginning (if not, the distinction is not worth bothering at once with a directly elided contrasting idea

ay. . r.ns. hinn K.; hi, m._i-

with), it may pivot on our personal assessment of 4 ; fownbeat tt x 399/ oo ape tel

the strength of the intervening TR-effect. If the “ an a . ism ” i" » he ¢ ; TR-effect before the I:HC MC-effect is more 4 1 , don : KOT /; c oh of elaborate even than that found in a complex on ae OSM AN EEE EEE erand antecedent—such as the one in Haydn’s also serves as a space separating the two ideas, a “Clock” Symphony —it may be better to regard eraceful slide into a new, lyrical module, whose

the situation as one of medial caesura declined. foment “he hol i m cul 1¢

Obviously, one should not frame a major her- . eee ey en eee

oo. . . ; in play h fi ill im-

meneutic point around any such presumed dis- repetitive loops is the primary expressive factinction in what 1s self-evidently an analytically ror in play here, Irom a secondary but still im

ambisuous situation portant perspective the format of such themes,

considered as wholes, is also in dialogue with the structural principle of the sentence: a presentation module appears twice in the manner of a potentially continuous loop and releases it-

/2a aDG ‘Oete erfr fey erey er eee YI

EXAMPLE 5.6 Mozart, Piano Sonatain C, K. 279, 1, mm. 1-20 Allegro

DT * 2 8?—___—— t 1 9? o-oo 87S——_=—_ - — — — es, ——— St ss... —_ oo oo tO 8 et 8 —_H>0“00wWw7000Wwvnw a —’--——-_ |)

” eine lard —_—

A ‘ tr !Fm —_eS - -es - oe sees

Oh. fe Fe — OS _smSmefs— eae, oo _ 9 OSS eewwTee TT Oot xTON2--2p>@@_NW|..-»—_— oT

3

te TS EE CO OHftSr 8 SS OO TD EEEOE — EEE |. |...

if se TP 8 CUL Y e.nOOWW2aDretOOeldvOlo" ?—_iaorn-0W.-——/— ai _ f+ rr | "en Pe

cy > Dad TY SaaS a ey a a aKS a ae Sa So

f)AA—A aOQ ae a58 ASSA 9A SSee A

5

Se oF ee. He au nnn tt i\—~—* a OO _—— OO OOOO on, ee OTB ———— EEO

FS SN AseeleeeeA2eg2elA Of eeSN ee eg Tg eg I Tes ee eeA Te Si)Py __@& —eg&__@__ esses ee

;y i,! 1 a

7 tra” t>ire Fg SS A DP A > i ee oo OD EDD 0 OD oO 8 OO OOo

|SESS OC 5 ee a SS aS OSS LSS ESS SS Fg SS A A SO CS SG SO A A 8 i i ee a a iA

fe ee ee eae Toe Toro oe (2/1. 8 oe oe Te Tre ral S|

f) royroy roy Pe Ya aa Sl TT eT? oe wo? ?_ono0we CF. T_T eee OO tO

— t YoY yr * 4 A SS SO A AI peAees eeAldA”ee —i— 2 _."#@€@am@a—m—m@—@—@—@—@—@—@—_ ——]3#ITIE=2E_IE=z_[_—_ EE OHOTATNOWWas

F — es Pee

ro OO0- ME a—_unnvnNwnNN OT ETTE—E—ETE—E eT —eE TS

cy scn7uUNTO-FONWAWVWVWVWV_ |"@—_ ——_ o_oa ee oo oh.meee ee FeouU]N}ttnNnTcT*Y*Y*--—=*——_—:"_[“XK{n...W TT ee

fey eee =e ee ee ——-_ —— —_— ae er ee en ee ae ee RT Rei A A YL A pm ant OT rr GC gt Il Sl PDTF—=>=PD_— Oo —— _ 8 8 ee OO OONn._ rNnu"7=2>T"V’[——

i—_— eee ot derNNuNW— oT ee OO Di OP OE F——> SS _— Op s—s--NX4§™~”—wNwWT#YNTN oye aD eeeeeFesy”rrP?’._€—_7-°--7™]7]’>"’"-"-"__|__3-N-_ 8 _ TT TEE eee

13 tr tr EXAMPLE 5.6 (continued)

LP” ee es srr ee ee )=— ee ey me TT et TlULmDlUlUL Oe e—=e, a terme a _ A—“‘“‘“‘cadential portion (P!:3) beginning at m. function, within this specific, readily recogniz- 9 and finally arriving at the desired structural able paradigm, as the continuation of a larger- I:PAC on the third beat of m. 12. (The remainscale sentence. This continuation function is der of this example will be discussed in chapter capable of undoing the full-closure effective- 6.) In K. 271/1, the “new sentence” breaking ness of any immediately preceding PAC, since free from the loops also begins with its own it indicates that we are to understand the earlier presentational modules, here still in a piano dyloops as only the first portion of a larger sen- namic, in mm. 8—9 and 10-11. In this case it tence, which is the real governing format at this is only with the “new sentence’s” continuation, point. Notwithstanding the two obvious PACs, m. 12, that the dynamic shifts suddenly to forte. the positionality of those cadences within the If the “P!-2” breakout-idea leads to its own larger sentential-thematic structure, along with I:PAC (as in K. 279/1, m. 12, beat 3), then its their subordination to the circular loops within status within P-space 1s normally secure, absent

which they are generated, weakens the usual any other factors that might lead us to think sense of a PAC as a sign of emphatic structural otherwise. Another option, though, would not closure and renders them incapable of function- lead the breakout-continuation to that PAC but ing as normative structural cadences. These per- drive it instead to a half-cadence arrival and fect authentic cadences, in short, cannot “end” subsequent MC. In these cases the breakout 1s the theme in question.!> They serve only as simultaneously the onset of the transition. The specialized openers to the larger theme, the P!-2 continuation will have been merged with sentence, within which they are embedded.!® TR, producing a “P>TR merger,” a special As a consequence, it is clarifying not to re- case often encountered with sentential P-themes gard such breakouts as we find in m. 5 of K. (as outlined in chapter 6). (More literally, this 279/1 and m. 7 of K. 271/1 as P? modules (fol- may be represented as P!-*=TR.) This is the lowing a structural, conclusive PAC) but rather situation in K. 271/1, where what begins as a as P!-2 modules— presuming here the conven- P!-2 breakout (m. 7), though initially in a piano tion whereby the continuation of the sentence dynamic, leads to a forte-driven half-cadence should normally be indicated by the 1.2 dec- and dominant-lock (in effect) in m. 14. A I:HC 15. We agree with Caplin’s more recent discussion of of mm. 28—32 and 32—36: see example 8.5 and its surthis issue (“The Classical Cadence,” p. 86) — which also rounding discussion.) Thus the general principle: notcited K. 279/1 as an example. Here Caplin also invoked withstanding any local PAC that it might contain, no the utility of “hierarchical perspective. For it 1s some- module that participates in a loop of normative size can

times valid to speak of cadential content having an ac- provide a cadence strong enough to be taken to mark tual cadential function at one level of structure while the end of a major structural unit. The “looped theme” also recognizing that this same content loses its function should always be considered as proceeding forward into at a higher level of structure. In these cases, it might be the breakout-continuation, however contrasting that useful to invoke the notion of limited cadential scope to continuation might be.

account for the effect of such cadences.” 17. Following this convention, however, can lead to 16. One consequence of this within S-themes that be- some complexities when dealing with a compound loop gin with loops is that no compound loop, ending with (b.1. + c.i.) in which one wishes to identify each smaller a PAC, should be considered to be capable of producing module with a separate label. One solution is to label the EEC. (The classic instance is found in the loops that b.i. and c.i. as P!-!8 and P!-!> (K. 271, mm. 1—2 and 3-4,

begin the secondary theme of Piano Concerto No. 21 in with upbeat). C, K. 467, e.g., in the opening ritornello, the two loops

86 Elements of Sonata Theory

MC will follow in m. 24. In other instances the opening module completely. Could the movemoment of the (“P!-2”) breakout may itself be- ment just as easily have begun with the existing gin with a sudden forte or at least an aggressive P!-le If so, or if that possibility at least seems plunge forward, even more clearly suggesting at reasonable to consider, then the designation P?

this point the effective onset of TR. may be taken to imply that judgment. On the There are many examples of similarly looped other hand, if it seems that P!! 1s set into mothemes in Mozart, and when dealing with that tion or otherwise “reacts” in a more dependent composer the analyst is well advised to be on way to the initial impulse —if one cannot really the lookout for them, particularly in order to imagine the piece starting with that P!-! gesbe able to interpret the presentational PACs ap- ture—then the label P!° is more appropriate, propriately. Other examples of pieces that open suggesting its function as a more “necessary” with this technique include the Quartet in preparation for the particular P!-! that follows it. B-flat, K. 172 (mm. 3—6, 7—10, following an The distinction between P°- and P!.°-modules initial set of separate, P° hammerstrokes on the is by no means absolute. Each label represents a tonic);!® Symphony No. 28 in C, K. 200 (an- broad span of modular types on a continuum of other classic instance, like K. 271/1, of the com- possibilities, shading into each other. In many pound loop, mm. 1—7, 7—13, with a forte break- cases it seems pointless to haggle in favor of the

out at m. 13);!° and Symphony No. 30 in D, K. one designator over the other. (At some level 2(0)2 (three separate, flush-juxtaposed loops—a one would assume that every P® conditions the triple-loop, each repetition of which 1s slightly — nature of what follows it.) P°-modules need not

varied, mm. 1—4, 5-8, 9-12, with an elided be (and are normally not) separated from P!:!

forte breakout at m. 12).?9 modules by a PAC. (See the discussions above of “zero-modules” and “exceptions in thematic

P°- and P!.°-Modules/Themes numberings. ) a:

Before identifying some common zero-types In some Allegro movements one gets the impres- within P, we should stress that not all prefatory

sion that the “real” P-theme (P!:!) begins two, Allegro gestures should be regarded as P9- or three, four, or more bars into the piece, and that P!.0-modules. Sometimes this initial gesture 1s this theme is preceded by a brief, different Al- merely a brief, in-tempo introduction, not a zerolegro idea, perhaps an opening flourish or other module proper. This would be the case with initializing gesture, that in some way prepares the opening chords of Beethoven’s Eroica Symus for it (typically in the sense of “get ready!” phony. Its generic forebears, the hammer-blow

or “attention!”). We identify such a preparatory openings of many mid-eighteenth-century gesture as either a P°- or a P!-°-module, depend- works, also usually fall into this category. But ing on one’s assessment of its conceptual sepa- when this initial gesture, with its undeniably rability from P!-!'—a P® idea being somewhat introductory feel, is included in any immedi-

more hypothetically “dispensable” than a P!° ate restatement of the P-theme, or when it is idea. In making such a distinction, one might included in the repeat of the exposition and pertry to imagine whether it would have been pos- haps also begins the recapitulation—as in the sible for the composer to have suppressed that quadruple-hammer-blow P® in mm. 1-2 of the 18. The loops of K. 172/1 are also notable in employ- initial-loop strategy may also be subjected to deformaing the circular 8—b7-6—47—8 motion discussed as a tion. The opening of Piano Sonata in C, K. 309, invokes

P-theme option later in this chapter. the paradigm only to block the normative completion 19. In K. 200/1 each loop ends not with a PAC proper (!) of the second compound loop at mm. 13-14. Mozart but rather with an obvious substitute, a 5-4—3-—2-1 then bursts impulsively through the staged blockage by

descent in the upper voice, mm. 3—7, 9-13. suddenly grasping onto a workable breakout-continua20. Other examples include the opening of Symphony tion, forte, atm. 15. Once the paradigm is grasped, one

“No. 1” in E-flat, K. 16, and what appears to be its might regard the opening of the Overture to Mozart’s much more sophisticated recomposition at the begin- Le nozze di Figaro, for instance, as a much-expanded ning of Piano Concerto No. 22 in E-flat, K. 482. The variant of the compound-loop technique.

The Primary Theme (P) 87

first movement of Mozart’s Quartet in B-flat, in the first movement of Schubert’s Quartet No. K. 172—then it seems grouped as part of the 13 in A Minor, D. 804: two bars of a twisting launching idea of the sonata proper. The musi- “background” pattern are soon overlaid with cal context tells us that it is conceptualized as the theme proper, P!!, in m. 3. By this time in

part of the P-theme complex. the nineteenth century —and certainly in subQualifications like these lead inevitably to sequent decades—the accompanimental opensituations that are open to interpretation. In ing became more common (for example, at the practice, such distinctions are not always so un- beginning of Mendelssohn’s Violin Concerto in equivocally made. It is not uncommon to find — E Minor and many other pieces). examples where an initial gesture hovers some-

where between an introduction and a P® theme. The Anacrusis-Module. Occasionally a comIn Haydn’s String Quartet in G, op. 33 no.5,for | posed-out initial gesture is elided with the onexample (example 5.1), one might initially sup- set of P!-!, functioning as a large upbeat to it. pose that the opening V’—I of mm. 1-2 con- The touchstone illustration of this type of P!-° stitutes a brief, in-tempo introduction: because (or perhaps P®) occurs in the opening movemim. 1—2 are not included in the expositional ment of Schubert’s Symphony No. 5 in B-flat, repeat, they seem introductory. On the other D. 485. Here one encounters a pianissimo P!°, hand, those measures are soon folded into expo- mm. 1-4, of considerable psychological subsitional space proper in mm. 9-10, suggesting tlety. P!-° serves as a gracefully expanded anatheir after-the-fact conceptual incorporation crusis, as if providing a corridor transporting us into P. They also reappear in mm. 182-83 to from our own worlds into that of the symphony launch the recapitulatory rotation. Thus mm. proper. P!-9 is included in the expositional re1—2 blend aspects of P° into what seems most peat, and while it does not begin the recapitulaintuitively to be a brief, in-tempo introduction. tion, the preceding development is essentially a A similar situation crops up in the opening large-scale expansion of it. In other words, the three hammer-blows of the Quartet in G, op. development marks the onset of a large second 76 no. 1. On the face of it, these would appear rotation that persists until the end of the recato constitute a brief, in-tempo introduction, pitulation. A related example may be consulted but a mild, P® function of the gesture is un- in the D-major Adagio non troppo movement derscored when the three-chord rhythm resur- of the Flute Concerto in G, K. 313 (a Type 5 faces in the lower parts near the opening of the sonata movement): one and a half bars of solemn development, mm. 89-90 (cello), mm. 93-94 “entry” or “invocation” (P® or P!-°) precede the (viola) —neatly linked, as it happens, with the onset of the actual P! theme (m. 2, beat 3). That three-chord gesture that closes the exposition. “invocation” reappears to begin the solo exposiThere are different types of P®- and P!.°- tion (m. 10) and the recapitulation (m. 38). modules. Among them: The Motto, Emblem, or Head-Motive. This is com-

Accompanimental Figuration (Rhythmic Stream). In mon in minor-mode works, where we often these cases one first hears a measure or two of — find—as a normal generic option—an abrupt,

accompaniment—as with the accompaniment peremptory initial stamp, a negative head moto a song (without words) — on top of which the tive, played forte, usually in octaves, before the theme proper is soon overlaid. The accompani- “real” theme (P!-') starts to flow forward. The ment or rhythmic stream is almost always best opening of Haydn’s Symphony No. 44 in E Miregarded as P!°. The opening bar of Mozart’s nor (“Trauer”) is typical (Example 5.8): the P®

Symphony No. 40 in G minor (Example 5.5) peremptory stamp (perhaps also construable as can be considered a brief P!-° of the accompa- P!.°) occurs, forte, in mm. 1—4; P!-! ensues, piano,

nimental type. A similar opening is provided inm.5. Here P® reappears in m. 13 and—probin the celebrated F-major Andante of his Pi- ably the central criterlon—is included in the ano Concerto in C, K. 467. A slightly more ex- expositional repeat. Similarly, at the beginning tended P!-° along these lines may be recognized of Haydn, Symphony No. 95 in C Minor: P?

gery Sa em a ( eg (( EXAMPLE 5.8 Haydn, Symphony No. 44 in E Minor (“Trauer”), 1, mm. 1—20

ew ee a ae a ae ' ' a “gt lee el pope ee LR eT a fee es es 7 -—

pO &— nem a QQ Allegro con brio

A, Re ee = SO CO (LS O_O (OO (OO (OR

5

a eS ee ee a i)eeapee st A a ee aO_O SQmC QO CO S(O aPRA esan 2”CO (OR GF CC PP ne ‘oy «© @2© a __ f° _0f_ 0” _0f 0? 0? "so @2 ff2 @ @© @@ mo ss | ~UUUUUUUUUUUUUUUdl l l0? le e0? le

A a es es ee es GO QO Once ae net ie ae ee el

= a 2 ae

16 mn —_ . . a . | ae st es, Fy fF Mo yt oof fy fed) gf © |

Hf ex 9 — te ya POO .—_ y. O'’SC‘CNOC‘iézs

FF —_— : id ~ 19 o——~ a

an end _ a yids | eee —,

Rey a [$j —_____"Ssssenal__epseesesssiaent ri} |

a Ae | a 2 GS DS (Os

The Primary Theme (P) 89

unisons, forte, mm. 1—2; P!-! proper, piano, with withdraw for the onset of what probably comes

the upbeat to m. 4. In a much more compressed to be regarded as P! proper. The openings of manner the opening of Beethoven’s Quartet in Schubert’s last two string quartets provide paraE Minor, op. 59 no. 2, also participates in this digms: in both instances P® suggests the preslogic (notice that the opening two hammer- ence of a calamitous situation to be confronted.

blows are recaptured in the exposition’s first In the Quartet No. 14 in D Minor, D. 810 ending), as does that of his Symphony No. 5 in (“Death and the Maiden”), mm. 1-14 burst C Minor, op. 67. In the latter case P!-°1s much forth at their outset, then recede into pianissimo less conceptually separated from P!!: P!-9, forte, and before long into stasis (fermata on V, m. mm. 1—5; P!-!, piano, begins with the upbeat to 14), as if parting the curtain for the main-theme m. 7. (It may be that peremptory, forte octave proper, P!, beginning in m. 15. Schubert made openings to minor-key works in general— cer- use of a structurally similar procedure in the tainly a commonly encountered feature — may Quartet No. 15 in G, D. 887. Here the modbe related to the P® concept. Or, of course, vice ally tormented initial declaration—like a tragic versa.) Much later in the century, Brahms would or defiant initial stamp, a bold illuminated iniopen his Symphony No. 3 with a latter-day vari- tial—flares up at once, urgently seizes the lisant of the P!-°-motto or emblem—1in this case tener, then recedes (mm. 11-14, again with also a kind of emblematic anacrusis—swelling fermata) to make way for the second launch,

dynamically into an explosive P!!. or the start of the “real” Hauptsatz in G major, Another example from Mozart illustrates an P!, pianissimo in m. 15. Perhaps in both cases instance in which P!-9 seems closely linked con- Schubert may have had in mind the opening ceptually with an otherwise sharply contrasting of Beethoven’s C-minor Coriolan Overture, P!-!: the opening gestures of his Violin Con- which had also begun with an analogous set of certo No. 4 in D, K. 218 (example 5.9). Here P° outbursts before P! proper gets underway: P®, the initial flourish leads to a contrasting, pia- _fortissimo, mm. 1—14 (a musical battering-ram, no-dynamic P!! theme with the upbeat to m. 5: breaking down the barriers to sonata-action); P! one could hardly imagine the concerto begin- begins, pianissimo, atm. 15. Here there is no rening with the music of mm. 5ff. It is also typical peat sign to guide us in distinguishing this from of such P!-! themes that they are structured as a brief, in-tempo introduction. The rhetorical sentence presentations. In the Violin Concerto recapitulation, though (beginning famously the opening four bars sound a stiff, march-like in the key of the subdominant), announces its fanfare in octaves. The contrasting module, m. presence with the P® theme. 5 (abandoning the all’unisono texture), is nonelided to the fanfare, although it is linked to Double Introductory Gestures. Is it possible to it by a sustained D4 in the horns and doubled imagine a situation in which the “real” P theme by the violas, which also articulate a Trommel- is preceded by not one but two introductory gesbass reiteration of that pitch. Mm. 5-12 are tures? This is extremely rare, but it does hapobviously sentential (mm. 5—6 and 7-8, a P!/! pen. The classic example occurs at the openpresentation; mm. 8 (beat 3)—12, P!-? continu- ing of Schubert’s Symphony No. 8 in B Minor ation — beginning with a bass recalling the fan- (“Unfinished”). In this case the best choices fare—and cadence). The first movement of the for labels are P® for the opening “motto” and Sonata for Two Solo Pianos in D, K. 448/375a, P!.° for the subsequent rhythmic stream. Thus

is also a member of this strategy-set. we have P® in the cellos and basses, mm. 1-8, At times a P® theme may be developed con- followed by P!-° with the subsequent rhythsiderably beyond the brief formulaic flourish. mic-stream accompaniment in the strings, mm. It can be more extended and assertive, only to 9—12. (Here it is useful to recall that that prepa21. On the opening of K. 172/1, see also n. 16. The The recapitulation also begins with this P°-module, “loops” referred to in that note start with the upbeat to mm. 72—73. m. 3, after the quadruple-stroke P® idea in mm. 1-2.

ee OT PS EXAMPLE 5.9 Mozart, Violin Concerto No. 4 in D, K. 218, 1, mm. 1-12

Yt, _ Nf dL Ummm LO _si—=n Allegro

aSS LOA6 > eSa Aa

eaueameneennasnaenen Pt eo _ 8 8S ro oS i CH oy.ov7.-——— eo adVvw#*"—NN—"—_ Io FH woo I oO

4

a a 1 y3’ &-—_~ ' Ids dsn TH, ~

Pp r— P~ ee eeTS eee Pe»WQ\ A O1O0 oc Ft eo er a "4 / pApf P~ AA if = P

T’' '— tr a i J ‘ 7 SS Se a SR A) 27 pd? ’ 7 oJ oF —* #f

On ot eS | __? _ ? ? e_ _ , ,'—_ S 2 , -—_ -— e

2 FO fEO Do. or .a . | ee

BF 2 aAR.A SS SS Sd

py —FC 7? 18 cy,Oo OA OTooo 7,

49

ay A ACY, OS OC CS A 00CT 7,SSSE "9 DS A nS eS SS S s- ae es Oo EE

Ln OA FD eS CC A rf tm OS —m =p ess a SS OE OO OH O_ o_O 7o co ay FO eeeeeSeEeeeeEeE—=E§ KO#8VYVVVY“‘ ——l oe a DO OO Po DT Eo EE TTT ETO YS .R5[T TT Sa@"@”"—“00W—N

ty? pF An. 2 > I“#W. (“#2 TOO i er _——oOO_—

The Transition (TR) 101

then extended via a dominant-lock into the me- Sonata in B-flat, K. 333, m. 11, and his Quartet dial caesura (m. 49). After a touch of caesura- in G, K. 387, m. 11. The restatement has the fill, S begins in m. 51. In sum, the re-entrance of — effect of reopening the closure provided by the P-material at m. 35 does not sound like a re- I:PAC. The opening of the P-theme 1s restated, beginning. Rather, the music bypasses the ex- but it now takes a different path, one that leads pected rebeginning in favor of impulsive motivic away from symmetry and the tonic closure of development and harmonic action. Here both P. The first movement of the Piano Sonata in A the agitato character and the manic overriding of Minor, K. 310, m. 9, is also in dialogue with this more common TR-generic norms contribute to TR type, although the elision of the restatement the expressive character of the music —its sense with the expected 1:PAC (suppressing the exof impatience, forward press, and dogged strug- pected tonic pitch in the upper voice) also serves gle to escape the minor mode. In this case it locally to suggest a destabilizing of full closure leads to an S that appears in the mediant minor, at this moment. The symphonic genre usually destabilized by a dominant pedal, in the manner marks the beginning of a dissolving-restatement of an S° theme. This suggests that S is “injured” transition in Allegro compositions with the geor subjected to a deformation, perhaps as a result neric tutti (or forte) affirmation, suggesting that of the impetuosity of the transition. See also the the full orchestra enthusiastically accepts the discussion of this piece —and of the principles proposed theme as the basis for a sonata, at the of minor-mode sonatas in general—1in chapters same time boosting the music’s energy in prepa-

7 and 14. ration for the drive to the medial caesura.

The tutti affirmation is a powerful generic signal within symphonies and overtures. In Common Transition Strategies: The most cases the listener should be aware at once

“Dissolving” Types that the music has entered the TR-zone. Even though the presence of this exuberant outburst Another frequently encountered strategy is to —_ can be a definer of TR-space that overrides other begin a thematic restatement, complement, or _ potential interpretations, its TR-authority is not

reprise that before long branches off into self- absolute. Complications with and exceptions to evident TR-elaboration. What is “dissolved” the tutti-affirmation principle are considered as along the way is the expectation of normative part of a separate section later in this chapter, thematic completion, which could have been “Premature or Delayed Tutti Affirmations.” fulfilled by attaining the cadence predicted by

the model begun in P.° These transition types . . may be subdivided according to the immediate The Dissolving Consequent structural impression or function provided by If P closes I: HC, in the manner of an anteced-

the opening of TR. ent (or, more commonly, grand antecedent), TR may begin as a parallel consequent that is

The Dissolving Restatement diverted before long into transitional processes.

In Allegro compositions this would be more ap-

When P closes I:PAC, TR may begin asarep- _ propriate in works that begin with larger, muletition or restatement of P, although it soon timodular, or grand antecedents, since shorter transforms into transitional activity. Examples ones run the risk of precipitating the work into include the first movements of Mozart, Piano TR too rapidly.®° Examples following grand an5. Cf., e.g., the similar concept in Marx, Die Lehre, e.g., 6. One might also recall that the more extended 3:259—61, Die Periode mit aufgeléstem Nachsatz. Marx’s P-theme cases that we call complex grand antecedents illustration, as it happens, 1s the opening of Beethoven’s (ch. 5) produce a “false” TR-effect (usually beginning Piano Sonata in F Minor, op. 2 no. 1. Later in this chap- with a local I:PAC) before the onset of the grand conseter we consider op. 2 no. 1 to illustrate instead the dis- quent. From certain perspectives, this might be considsolving hybrid—adapting here the terminology of Wil- ered a “TR°”-effect—since it leads directly into TR!1,

liam E. Caplin. the dissolving consequent.

102 Elements of Sonata Theory

tecedents abound in the literature, including the the upper voice moves through this cadence and first movements of Mozart’s Symphonies No. 40 begins to repeat the continuation module, as if in G Minor, K. 550, m. 21 (Example 5.5), and preparing to deliver a more satisfactory I: PAC No. 41 in C Major, “Jupiter,” K. 551, m. 24; and four bars later. M. 14, however, introduces 24 in Beethoven, Piano Sonatas in C major, op. 53, the bass, and leads to a I: HC in m. 15, followed “Waldstein,’ m. 14, and F Minor, op. 57, “Ap- by a dominant-lock —thus submitting the prepassionata,’ m. 17. Slow movements with mod- vious cadence to a deformation and leading to est proportions —at least in principle— permit self-evident TR-procedures, with a I:HC MC the application of the dissolving-consequent TR at the downbeat of m. 20. In this example P and to smaller, more normative antecedent phrases, TR have again been merged through what we although even here the period 1s more typically might call the “dissolving consequent-restate-

completed before TR proper (of a different ment,’ which in this case involves the defor-

type) is launched. mational repetition of the entire final portion As with the TR of the dissolving-restatement of a sentence. TR proper thus begins with the type, the dissolving consequent has the character anacrusis to m. 13. of a second launch— one that pushes the music

toward the medial caesura. In this case there is The Dissolving P-Codetta: Reiterated

no sense of reopening previous harmonic clo- . . .

. ee, Reinforcement)

sure. On the contrary, TR first suggests that it Cadences (Dissolving Cadential

will attempt to close the harmonic situation left

open by P but soon abandons this effort. Thus When P concludes with a I:PAC, it may be folthe hermeneutic implications of the dissolving lowed by repeated cadential units (often two of restatement and the dissolving consequent are them) designed to reinforce that cadence. Thus different. In the first an idea has been brought our initial impression is that of a P-codetta. At to authentic-cadence completion; in the second times it is best to regard it as that—as an emthe P-idea veers away (or decays away) from that phatically concluding portion of P-space, part PAC-completion in order to pursue different of the P-landing-strip, so to speak. This seems expositional aims, and P proper is thereby ton- unequivocally the case when the P-codetta

ally underdetermined. is bluntly separated off from what follows by

a full break, producing an unmistakable sense

Period with Dissolving-Consequent of closure (“Punkt!”) and a clear shift into

Restatement (or Sentence with the subsequent tRezone, rere there mu

. . . . TR-dissolution, and P is Restatement) merely supplied .with Dissolving-Continuation . a self-closing codetta of cadential reiterations. It sometimes happens that a complete period Such is the case in the finale of Mozart’s Piano

or sentence 1s sounded only to have the be- Sonata in C, K. 330, in which, following the sinning of its second portion— consequent or I:PAC of an unproblematically periodic P', we continuation —resounded (as if a full restate- are given two loops of repeated tonic cadences ment is to be expected) but then subjected to (mm. 16—20) concluding with the archetypally dissolution. Following the logic of the above Mozartian closure-figure 8—5—7 in octaves (m. TR types, we consider TR to begin with the 20) —nailing the door shut on P-space. TR folonset of the undermined restatement. For ex- lows in m. 21. ample, in Beethoven’s Piano Sonata in G, op. 49 But related analytical situations are rarely so no. 2, P is structured as a larger-scale sentence easily decided. It may happen that the (usually) with a presentation of 4 + 4 bars and a con- two looped cadences are not so clearly set off by tinuation-conclusion of four bars (mm. 9-12) a full break from what follows. It may be that ending in I:PAC. Because the size of the second the second reinforcing unit leads more directly portion of a sentence often balances (or even into a differing module that turns transitional in extends) that of the presentation, the cadence in function, often advancing rather quickly to the m. 12 may strike us as premature —and indeed, medial caesura (frequently I: HC). In this case

The Transition (TR) 103

the effect is that of undoing the I:PAC reitera- The Dissolving P-Codetta: Tonic Prolongation tions and transforming them into the half ca- via the “Circular” 8—b7-6—47-8 Model

dence thatprocedure marks the medial caesura. Moreover, ,, . .isThis operates oncadential a logic parallel it clear that the doubleness of the . ..j. . eg. with the cadential P-codetta discussion directly reiterations 1s in dialogue presentation ,, . one above.with In this the case, though, we are not given modules of a sentence, whose continua, , , . overt cadentialbyreiterations of follows P’s I:PAC. Intion is represented the music that the . i , stead, the now-attained tonic is held fast with. a

two looped. cadences. turn inthis can .. . pedal point,In usually thebegins bass,suggest while a twicethat we have a sentential TR that with a oo, . . . repeatedcadential holding-pattern figure In cycles above. .it tonic-reiterative, presentation. this . « TR-“sentence” a . . in one or more upper This creates sense the (if“in-and-out-of-focus” that 1svoices. the impres, , ca sta3the .sion . . .obtained) , impression of an is perceived as beginning with are . a: sis circling around theHere fixed tonic, a shift into the onset of the. .cadential reiterations. one . . . a gratified circular recursion—even a modest

should note the double-function of the the repeated . . .into . . celebration — following linear progress cadences: on the one hand, they first strike us as , . . P’s I:PAC. This may be accomplished with any

P-codetta; on the other hand, we come to un- , , , i. of several simple harmonic patterns, including derstand that were actually the beginning . . yi: . . they tonic-dominant oscillations.

of a [TR—possibly a sentential TR. Such a TR . . . . of , Joa: 4 _ Of particular interest is the prolongation is a subset of .the “dissolving” transition types. . . .pedal . the thing cadential stasis byenlocking onto aD2 tonic What starts out as one 1s converted route 4 A 1A LA 6 . . . above which is that rotated an 8—57-6—47-—8 into else is self-evidently more .—usu, me_ something . lodic motion in one of the upper voices

transitional infactor procedure. . pattern, I-V’/IV . . . . ally outlining chordal Obviously, the pivotal in athe decision 6. . . —IV4i—viie® (or VI71)—-I. (More rarely, the

. . 8 the —b7-6—47-8 model may be transferred cerning presence or absence of that potential A |A A LA Ato

here (P-codetta or TR?) is our judgment con- A |5 2 LAB

. . the bass.) The 8-—b7-6—-47-8 module is a full break after .the cadential reiterations. When a: stock contrapuntal pattern within the style. Its

the. full break seems absent or1s when what fol- , of . ca. . . most natural function the grounding lows. dences is clearly reactive to motion or continuative of the .... with a static of circular rotation cadential .figures, then we may speak properly . . . . of . around an attained tonic. The linear motion of a TR .of; .the the dissolving cadential reinforce. . phrase(s) convert here into a rement type. Thispreceding procedure occursFor with some . . . the . ; cursive, circular stasis. this reason figure frequency in the first movements of Mozart’s . . . . . . is sonatas. especially appropriate formodel postcadential clospiano One exemplary is that in ,in . . ing ideas (post-EEC C-space) and, as here, the Piano Sonata in to C, P-codettas. K. 330 (example 6.3). . the amount (Occasionally, Here .P;.1s.what structured as a sentence (aa', mm. 1-2, , . _ static figuremm. is used withvaried notable effect to launch 3—4; .continuation, 5—8, repetition , oe , . . . P-space: See the discussion in chapter 5.) ‘Typiof the continuation, mm. 9-12) ending with . . cally the figure makes two full cycles through a I:PAC in m. 12. Following a bar of fill we A |\A A LA BZ . . thereiterations 8—b7-6—47—8 pattern. a. . hear two. .varied ofthethe preceding . Within context the potential P-codetta cadence, mm. 13-14, 15-16. M.pedal 16ofand provides ,.. as . the figure’s tonic circular recursion no.full break. On the contrary, it is elided imi, . . . . . . give it a character that suggests immediate cadenmediately launching Fal reinf. .4of keen; cp P . . ialtowith reinforcement, keeping-open P-space. a drive the contrasting I:HC MC material in m.a18. We consider pine OP

1s so even if the pattern introducesi,a. more TR proper. .to(This begin at m. 13. Three sixteenth . . or less newlead figuration into theonset piece. In. this sense notes of caesura-fill (CF) into the it can also partake thematically in the strategy of the compound-sentence S, m. 19.transition, Another, _. . below.) . of the independent discussed more extravagant example of this type of TR . . . . if . , ,inAs was themovement case with cadential reiterations, occurs the opening of Beethoven’s 5S 1A A LA 6 . . thein8—b7-6—47-8 diQuartet C Minor, op. 18pattern no. 4, is m.not 13.merged ,,, rectly with a continuative or reactive subsequent

i

aa EXAMPLE 6.3. Mozart, Piano Sonata in C, K. 330, 1, mm. 1-20

hff >ee »2 F7«> tr TO |a ra», | #£ gg ¢f aOO ree, * | ee ree Piano elo os ee A a Sl eeee|lle Ra ee ee ee =eS eelll ||

era ym | 4h trYT - «a . _Li gf Allegro moderato

Pi | [| ffBe | | Ff wl #-| #3 jf |] gy [| | fFee @#ee | CewmAS te# ST ee[)yJ LT« #JTteoy C“‘“LS

— —__—— ————_ _— = eee

PU a eS eqrmy | |eA TT| | a a ee ee se

se —“—imes A cele| ee es COS emu ggesOM fT aesaAOD fg> JOF fpee a .ee - —s eeseG oy fey —P 18

oa a—5P ft

module —if it 1s clearly separated off from what 8-57 -6—-47—8 pattern (C,—Bb,—A,—B4,—C.),

follows by some sort of full break —then it gives which ground the C-tonic statically, then push the impression of belonging wholly to P-space. into linear motion on the third beat of m. 14, This type of P-codetta may be found in the first now driving toward the I:HC MC in m. 16. (Exmovements of Mozart’s Quartet in B-flat, K. ample 5.6 also shows the opening of an extremely 589, mm. 12—20 (with a full-break GP-gap in unusual S, a sentence that begins off-tonic, m. 17, m. 20 and TR proper beginning in m. 21, al- with a bar-by-bar circle-of-fifth descent through beit one that preserves a motivic figure from _ the presentation into a tonic continuation.)

what precedes it); and Beethoven’s Quartet in Similar examples of the dissolving D, op. 18 no. 3, mm. 27—35. Another example 8 —57-6—47-8 pattern may be found in the at least beginning with the 8—b7 —6—47 -8 pat- first movement of Mozart’s Sonata in D, K. 284

tern, here producing a tonally overdetermined (shown in example 3.1, TR at m. 9, beginning P followed by the characteristic TR-jolt to the with two loops of the pattern, literally in mm. submediant, occurs in the first movement of | 9-10, implicit in mm. 11-12) and the finale Haydn, Piano Sonata in E-flat, Hob. XVI:25 of the Sonata in G, K. 283 (TR at m. 24, two (P-codetta, m. 8; TR, second half of m. 12). loops, mm. 24-28, mm. 28—32). Some re-

It may also happen that the two lated passages include the first movement of the 8 —b7 -6-47-8 “codetta”-cycles are followed Sonata in B-flat, K. 281 (m. 8, initiating two without such a break into a contrasting mod- soundings of a I-IV%—I motion, but one clearly ule of continuation. Again, as with cadential related to the pattern); the finale of the Sonata reiterations, what originally seemed marked in F, K. 280 (beginning in m. 17, ‘TR is literally as P-codetta begins to take on the presenta- an independent sentence, yet it begins with two tion function of a [TR-sentence. At the same presentation modules over the familiar tonic time, the circular loops divert almost immedi- pedal); and the slow movement of the Sonata in ately into more recognizable TR-processes and B-flat, K. 333—whose initial harmonic swaytheir once-again linear implication. Following ing is restricted to tonic and dominant. the argument outlined above, in these cases we

. Module

again consider TR to have begun with the onset Sentence with Dissolving Continuation of the first cycle of the pattern. One comes across this issue in several of Mozart’s piano sonatas. The first movement of In this case the sonata begins in the manner of a the Sonata in C, K. 279 (see chapter 5, example sentence, generally some sort of a a’ b format. P 1s 5.6) provides an illustration. In the preceding restricted to the normative presentation modules chapter we have already discussed this P, mm. only, a a', sometimes a mere four bars (2 + 2), 1—12, as a touchstone example of the Mozartian while it is the continuation (b) that ramifies into “loop” variant of the classical sentence. Begin- TR-activity —rhythmic verve and accelerated ning with the cadence on the third beat of m. harmonic action that drives toward the medial 12 (the elided onset of TR) are two cycles of the caesura. TR is considered as beginning with b

106 Elements of Sonata Theory

(transition of the dissolving-continuation type). that leads toward a half cadence (m. 11) anda Before dealing with the locus classicus, the open- well-articulated I:HC medial caesura at its end. ing of Mozart’s sonate facile, the Piano Sonata There can be no doubt that what follows it is S in C, K. 545, it may be useful to consider one (an S!° vamp in m. 13, the onset of a sentential

larger issue. S-proper in m. 14). The opening four measures We have insisted that in nearly every case suggest a normative sentence presentation, aa’ we understand TR to begin with the onset of © (mm. 1—2, 3-4) with the characteristic hara new phrase. If we remain consistent with our monic oscillations around the tonic. The four definition of a phrase—a stretch of music end- bars do not end with a cadence. (This 1s also the ing in a cadence —this TR-situation is the ex- case in the other examples cited above, K. 155/ ception. This is because, by accepted definition, i1 and K. 156/1.) Instead they prolong the tonic a sentence’s presentation is so harmonically weak ~— with neighboring motions in the bass. There-

(perhaps only oscillations between I and V) that fore they do not constitute a phrase—only a

it is not considered normatively to end with module (or a complementary pair of two-bar a cadence. (True, oscillations to I or V might modules). The sentence’s continuation module strike us as protocadential— almost there — but begins in mm. 5-8 (b, with its typical sequenthe fact remains that they are not normally re- tial treatment of a shorter structural unit). Mm. garded as producing a “cadence” proper.) Thus 9—12 constitute the conclusion of the sentence by our preferred definition of a phrase, a stan- with the cadential module (essentially 1°—V in dard-length sentence is usually a single phrase the tonic key). But mm. 9-12 also take on the from presentation through continuation and ca- transitional features of a typical drive to a medence.’ Since it is the continuation that dissolves dial caesura, including a dominant lock at m. 11 here, in this case TR does not begin with the and a triple-hammer-blow gesture I: HC MC at onset of a phrase. On the other hand, there 1s m. 12. The continuation portion of P overlaps always a clear separation of thematic modules with TR, and P and TR thus merge (PS>TR). as one moves from presentation to (dissolving)

continuation. One might at least claim that TR The Dissolving Hybrid

begins with a new module.

One paradigmatic instance is found at the In Caplin’s Classical Form the classifications of opening of the A-major slow movement of the common phrase types are not limited to periods Quartet in D, K. 155 (shown in chapter 17, ex- and sentences. Also included are four types of ample 17.4a). Another occurs at the opening of = “hybrids” between the two, a set of categories Mozart’s Quartet in G, K. 156 (mm. 1-18, end- that, in order to avoid further complications, ing I:HC, S at m. 19). More familiar 1s his Piano we adopt within our own discussions here.® Sonata in C, K. 545, first movement (example According to that method, an antecedent, for 6.4): Here the opening appears structured as a example, may be followed not by a parallel consingle phrase (as strictly defined), mm. 1-12, sequent but by a nonparallel phrase more typi-

7. The most obvious exception to this involves the fundamental, with the result that any of the common loop-variant of the sentence, in which the initial loops deviations away from those models (weakened cadentypically end with local I:PACs, thereby qualifying as tial-effects or mere dominant arrivals, nonparallel “secphrases. In other words, a sentence designed in this ond halves,” and the like) should be understood as “hyloop-format will typically consist of three phrases: the brids” between the two. We would prefer to posit more first and second loops and the continuation, which will flexible, but still normative, phrase-models with which itself end with a cadence. See the discussion of Mozart- any individual exemplar is understood to be in dialogue, ian “loops” in ch. 5. Other types of large-scale sentences at times even deformationally. Under such a conception, may also provide exceptions to this general principle of an apparent deviation from the norm does not produce

the noncadential quality of the presentation. an example of a new, “hybrid” category but rather an 8. Cf. ch. 5, nn. 10 and 14. From a larger perspective, individualized realization—for particular expressive though, it is not self-evident why the sentence and the purposes—understood to interact conceptually with parallel period alone should be taken to be exclusively the most relevant background model.

EXAMPLE 6.4. Mozart, Piano Sonata in C, K. 545,1, mm. 1-17

a a SD CO ( | - | nn Allegro

iFeyAT 6 eeELes eeKX, «=ALLS QO|_OO NS EEE ps pps ISX, if, ¢* TJ JT TT E fayee Cle ee ee| TT ee |eefT ee ee ee ee ee ee *ee| SU tCeeOr? |eeFY ~~. ~*~ | |yyy iF es | *ee~_

4afAyY \)Atr aN nc Ne|_=e Se Ale a OS ano ema

a+)Apr,a __ NSpfSO rn iee AS eeireee a age re ee C= _le A ey. CT ”™~—“—;éié~sY:SC‘®SSNSC‘CEANYNN’NUN:UNNNNN | —‘_"_"# “=p Pe PCT CT CU dT Te ee

a|reeS ==ee A ee ss eee

NR eae

i A ee ee ee ee |A Cs OD | SD ne(DD nel es GD OD 6

Pfa. WE 2 ee ee ee 2 eee a 2 eee ee ee ee 2 eee

Ce Pr | rwwg [| | | jfge ff, _| Pr Pe, | Ti | __s7° gp© —6—hlvh)4ngW | {mw ge — ©.| ff| °°

i a ee ee ce Oe ecco _ oe

8

ee

a CS «a ele ei Fe ns DO Nk rrr a SD OO CS ee 2nl A 2| aE nl lec OD a AIan aS a ne Ee —(ii‘(‘(‘(‘(‘ In some of his sketches from around 1800 ... should not be in utter contrast to the primary theme proto-Hegelian one) as elaborated in Fred Ritzel, Die but should rather present a specific variation reminiscent Entwicklung der ‘Sonatenform’ im musiktheoretischen Schriftof it. In 1777, Carl Ludwig Junker stated [in his Tonkunst tum des 18. and 19. Jahrhunderts (Wiesbaden, Breitkopf &

(Bern, 1777)|, ‘Each secondary subject [Nebenthemata] is Hartel, 1968), p. 102—5. E.g. (p. 105): “Kontrast bleibt a consequence of the original ruling passion of the work fiir die formalen Konstellationen sekundar, nur ein Mit-

and so is related to it. Otherwise the whole piece will tel zur Erzielung der Mannigfaltigkeit, aber nicht wesbe incoherent, and the fundamental emotion will re- entliches Element eines dialektischen Formprozesses.” main hidden.’” See also the remarks on the eighteenth- 3. Early descriptions of the sonata are documented in century concepts of contrast and “antithesis” (then un- many sources. See, e.g., Fred Ritzel, Die Entwicklung der derstood more in a classically rhetorical sense than a ‘Sonatenform’ im musiktheoretischen Schrifttum des 18. and 19.

The Secondary Theme (S) and Essential Expositional Closure (EEC) 119

(the op. 18 quartets, the violin sonatas from op. ter 14 discusses the role of the minor mode in 30, the piano sonatas from p. 31), Beethoven eighteenth-century practice and also glances at seems to have referred to the S-theme with the some alternative choices for the key of S—such abbreviation, “m.g.” William Drabkin has sug- as VI— explored in the nineteenth century.) In gested that this (foreshadowing the terminology major-mode sonatas before 1800 moving to the of Czerny?) may have been an abbreviation for dominant key (V) for S was the only standard Mittel-Gedanke, an observation that seems in- choice. From time to time one might find the creasingly likely as further evidence from the beginning of S in the dominant minor (as in

later works comes to light.* Mozart’s Overture to Idomeneo), and this option is treated 1n a separate section below. Similarly, a major-mode S (in V) might contain unstable

Tonal Choices for S local inflections of the dominant minor or be subjected to a restatement in the dominant mi-

As indicated in chapters 2 and 3, minor-mode nor. These are significant gestures within the sonatas usually move to the major mediant (HI) — individualized narrative at hand, and they de-

for S and C. Considerably less frequently, S mand hermeneutic attention. and C are deployed in the minor dominant (v), Around 1800 Beethoven began to investigate with much more negative implications. (Chap- _ the deformation of moving to III for part 2, not Jahrhunderts; Birgitte Moyer, “Concepts of Musical p. 25. For Czerny and Marx, see Moyer, “Concepts of Form in the Nineteenth Century with Special Refer- Musical Form,” pp. 65 and 69-125. ence to A.B. Marx and Sonata Form,” Diss. Stanford, 4. William Drabkin, “Beethoven’s Understanding of 1969; William S. Newman, The Sonata in the Classic Era: ‘Sonata Form’: The Evidence of the Sketchbooks,” in The Second Volume of A History of the Sonata Idea, 2nd Beethoven’s Compositional Process, ed. William Kinder-

ed. (New York: Norton, 1972), pp. 19-42; Leonard G. man (Lincoln: University of Nebraska Press, 1991), pp. Ratner, Classic Music: Expression, Form, and Style (New 14-19 (“m.g.” on p. 18). Further evidence from the York: Schirmer, 1980), pp. 217-47; Ian Bent, Analysis sketches for a “secondary theme,” “similar to [a] second (New York: Norton, 1987), pp. 12—32; Hans-Joachim subject,” marked “m.g.” 1s found in the Benedictus of Hinrichsen, “Sonatenform, Sonatenhauptsatzform”’ the Missa Solemnis, as noted in William Kinderman, Ar[1996], in Hans Heinrich Eggebrecht, ed., Handwérter- taria 195: Beethoven’s Sketchbook for the Missa solemnis and buch der musikalischen Terminologie (Stuttgart: Steiner, the Piano Sonata in E Major, Opus 109 (Urbana and Chi-

n.d.), pp. 1-7. The above refer to virtually all of these cago: University of Illinois Press, 2003), vol. 1, Comfigures. More specifically: For Koch, see the Versuch einer mentary, pp. 109-10, which refers to the “m.g.” detail Anleitung zur Composition, vol. 3, sections 100—3 (“Von in the sketch itself, p. 93 (in vol. 2, Facsimile, and vol.

der Sinfonie’”’), 141, and 147 (1793; rpt. Hildesheim: 3, Transcription). In a personal communication to the Georg Olms, 1969), p. 301-11, 363-66, 381-86 ; in authors (November 22, 2004), Kinderman additionthis case the thematic descriptions are taken from sec- ally noted that “m.g.” or “Mittelgedanke” also “aptions 141 and 147, pp. 364, 385; translations of much of pears in Beethoven’s sketches for his unfinished Piano Koch are available in Koch, Introductory Essay on Com- Trio in F minor from 1816 in the Scheide Sketchbook position: The Mechanical Rules of Melody, Sections 3 and at Princeton, where the ‘second subject’ in D-flat major

4, trans. Nancy Kovaleff Baker (New Haven, Conn.: is so labeled.” In the same message Kinderman noted Yale University Press, 1983). For Galeazzi, see Bathia that Beethoven’s terminology also included “m.s.” for Churgin, “Francesco Galeazzi’s Description (1796) of “Mittelsatz,” meaning “a large internal section, often Sonata Form,” Journal of the American Musicological Soci- though not always corresponding to our ‘development

ety 21 (1968), 181-99. For Kollmann, see his An Essay section,” and cited evidence of the “m.s” indication on Practical Musical Composition [1799], (rpt New York: from several works, including the second movement of Da Capo, 1973), excerpted also in Music in the Western the Piano Sonata in E, op. 109 (also discussed in the ArWorld: A History in Documents, ed. Piero Weiss and Rich- taria 195 volumes). Beethoven’s term for exposition was

ard Taruskin (New York: Schirmer, 1984), pp. 316-19 normally “‘erster [or ‘1ter’] Theil,” while he referred to and Ratner, Classic Music, p. 219. For Reicha see Bent, the remainder of the movement, from the development Analysis, pp. 18—20 and especially Peter A. Hoyt, “The onward, as “2ter Theil.” Following the publication of Concept of développement in the Early Nineteenth Cen- the Artaria 195 facsimile, we now know that Beethoven tury,’ in Music Theory in the Age of Romanticism, ed. Ian continued to use this terminology in his later period. Bent (Cambridge: Cambridge University Press, 1996), We thank William Kinderman for drawing this to our pp. 141-62. For Birnbach and Gathy, see Moyer, “Con- attention. cepts of Musical Form,” pp. 56—57, and Bent, Analysis,

120 Elements of Sonata Theory

to V, within major-mode sonata-based struc- its S-theme moves sequentially from the “notures. This might have been suggested by anal- ble” E-flat major, through F minor, and into G ogy to the role of the mediant in minor-mode minor for the EEC and C-idea. (See the chapsonatas. Examples may be found in the first ter 14 discussion of this piece in the context of movements of his Piano Sonatas in G, op. 31 i—III—v motions in minor-mode expositions.) no. 1 (S first in B major, m. 66, then repeated in Tonally migratory S-areas are also found in the B minor, m. 74, which key persists for the rest of first movements of a few concertos of the pethe exposition) and in C, op. 53 (“Waldstein,” riod (chapters 20 and 21). Such practices can Sin E major), in the Leonore Overtures Nos. produce what have sometimes been regarded 2 and 3 (P in C major; S and C in E major), as three-stage expositions, in which, in effect, and in the finale of the Piano Trio in E-flat, op. part-2 space was divided into two tonal regions. 70 no. 2 (S and C in G major).° Moving to III In some cases, these provide deformational was not the only alternative that Beethoven ex- complications associated with apparent double plored around this time. In the first movement — medial caesuras (the trimodular-block effect, or of the String Quintet in C, op. 29, he deployed TMB), to be discussed in chapter 8. the secondary theme in VI. Here the S-theme begins in a bright A major, VI (mm. 41-51),

but soon decays to A minor, vi (m. 52, with an Essential Expositional Closure:

immediate, transient “escape attempt” toward The First-PAC Rule F major in mm. 54—56), for the remainder of the exposition. A similar procedure may be ob- One central feature of Sonata Theory is its emserved in the first movement of the Triple Con- phasis, after the onset of the secondary theme, certo in C, op. 56, whose solo exposition moves on the attainment of the first satisfactory perto A major, then, eventually, to A minor, while fect authentic cadence that proceeds onward to the first movement of the Piano Trio in B-flat, differing material. This is the moment that we op. 97, “Archduke,” deploys its S- and C-ideas term essential expositional closure (“the EEC”’). It entirely in G major. Schubert was another who _1s toward the accomplishing of this PAC, markexperimented with other expositional possibili- | ing the end of S-space, that we understand all of ties. By the mid-nineteenth century the move the preceding music to have been aiming. This to V, the key of generic tradition, remained a issue of “the first PAC” 1s a complex matter, and first-level default for expositions, but other tonal — we do not wish to minimize its difficulties. (We choices were also acceptable (especially various shall revisit it again below in “Some Schenkershades of mediants and submediants) as low- ian Implications” and at some length in chapter 8 er-level defaults for idiosyncratic structural im- —_ in our consideration of EEC deferrals — specific

plications — among which the unwillingness or situations in which it is a later PAC that effects staged “inability” to move to the traditionally | the EEC.) At stake 1s the issue of when S may be normative V was by no means the least telling. considered to have ended or, conversely, when Also conceivable around 1800, although the closing ideas (C) may be said to have begun. still rare, were “tonally migratory” S-themes ‘There are two general approaches to the questhat begin in one key and move into another tion. These two approaches— the larger versus to produce the EEC. Beethoven’s C-minor Co- the more restricted understanding of S (or the riolan Overture, op. 62, is the touchstone here: brief versus the broader concept of C)— may be 5. Animportant predecessor here is what may regarded is problematized by the persistence of TR-figuration as the (deformational) exposition of the slow movement through the end of the exposition, which also flows of Beethoven’s Piano Sonata in C, op. 2 no. 3 (E major through most of the developmental space. To undermoving, after a modal collapse to E minor, to G ma- stand this movement as a “special rondo form,” as jor). This movement undertakes a dialogue with cer- does Tovey, is inadequate. See Tovey, A Companion to tain sonata principles and expectations. It has both an Beethoven’s Pianoforte Sonatas (London, 1931), pp. 27-29. expositional and a recapitulatory rotation—suggesting Cf. William E. Caplin’s “large ternary form” solution in

the presence of a Type 3 sonata—although the whole Classical Form, pp. 216, 282, nn. 40 and 41.

The Secondary Theme (S) and Essential Expositional Closure (EEC) 121

represented by the positions of William E. Cap- plin’s sense, and many S-theme “groups” would lin and William Rothstein. We favor Rothstein’s extend far beyond their first PACs. While this

conclusions for reasons that emerge below. reasoning is internally consistent, it appears to Caplin’s view, in Classical Form (1998), was to have been decided by fiat, perhaps also partly in

restrict the concept of closing music to nonthe- resonance with mid-twentieth-century analytimatic materials only. Consequently he preferred cal assumptions: there is not much by way of an the term “closing section” to “closing theme” appeal to earlier historical evidence to support (which term he assessed as confusing, for exam- the claim. Caplin recognized, though only in ple as found in Charles Rosen’s Sonata Forms), passing, that Rothstein had come to a different and he distinguished it as a “group of codet- conclusion, but he presented no counterargutas” from the genuine “themes” of the “subor- ment to shore up his own position.’ dinate-theme group.” On this interpretation “‘a In claiming priority for the “first strongly arclosing section usually contains several different ticulated perfect cadence in the goal key” —at codettas,” often shortening in overall length. In the time (1989) a striking claim—Rothstein another description he defined the closing sec- had elaborated the central problem at a more tion as “a postcadential intrathematic function fundamental level and had stood on firmer hisfollowing a perfect authentic cadence. It con- torical ground: sists of a group of codettas, often featuring frag-

mentation and a recessive dynamic.” The per- There may be some question [once the second fect authentic cadence in question here 1s the group has begun] as to which of two or three final PAC of the “subordinate-theme group,” cadences is the closing cadence. Normally, it 1s which may display several successive PACs dis- the first perfect cadence in the key of the second posed in “multiple subordinate themes”: “each STOUP. + + The cadences that come later may be

; considered reinforcements of the closing cadence; one ends with atoperfect . ; , of 59these theirthemes purpose is often satisfyauthensome element of

IC cadence in the subordinate key. Thus Ca- closure left incomplete in the closing cadence itplin decided on behalf of the larger view of S, self. Elements of closure, beyond the harmonic an additive S-grouping principle that construed cadence, may involve the register of the bass or S-space as potentially occupied by a succession melody, the presence or absence of some imporof themes that each end with a PAC. This sub- tant melodic tone, ... the presence or absence ofa ordinate-theme group “almost always demands subdominant-type harmony in the cadential proa postcadential passage [of codettas] either to gression, or any of a number of other factors.® dissipate the accumulated energy or, sometimes,

to sustain that energy even further beyond the In support of this view Rothstein had also writ-

actual moment of cadential closure.”® ten:

In short, Caplin interpreted the end of Sspace to occur when the “themes” stop and the Actually, the subdivisions of the second group “codettas” begin, with their characteristically have been better described by older theorists such “postcadential function” and their “general sense as Koch and Reicha than by most of their succesof compression of musical material.” With few SOT. Following cadence Structure as usual, both exceptions closing sections would encompass of these theorists distinguish sharply between any

_ passages preceding the first perfect cadence in the

only the final bars of most EXPOSITIONS. They goal key of the exposition and any passages followwould comprise only those reinforcing sections ing that cadence. The former they consider part of that are clearly not graspable as “themes” in Ca- the main body of the exposition, the latter not. . . .

6. Caplin, Classical Form, pp. 121-22. ondary-theme group lasted through an expanded ca7. Caplin, Classical Form, pp. 122, 273 n. 82 (“fre- dential progression (ECP) —basically the articulation of quently identifies”). In an earlier article, “The ‘Ex- a 16—predominant—V4=3-1 progression spread out over panded Cadential Progression’: A Category for the a broader space of four or more bars. Cf. n. 15. Analysis of Classical Form,” Journal of Musicological Re- 8. Rothstein, Phrase Rhythm in Tonal Music, p. 116. search 7 (1987), 215—57, Caplin had argued that the sec-

122 Elements of Sonata Theory

Following the reasoning of these older theorists, F-major keyboard sonata by Franz Mezger, Vowe will term as closing theme, or in some cases gler wrote that the fourth section (our S) contincodetta, only the suffix or suffixes to the exposi- ued “until finally . . . a formal cadence in C ap-

Hone 8 ony wos P a of ne vecond pears,” a cadence that “receives its confirmation

fect cadence thefirst goal key...cadence . a ; 9 11 [schlussfallmasigen Periode|” (our C). Emphasis oninthe perfect in the

STOUP TOMOWINS UE HISE STONE TY ATTIC Wared: per | Bestatigung| by the following cadential section

Le the Such lines of argument us as-persuasive, second group- as critical point in its strike form corresponds closely to the ideas of Koch and Reicha. and we should add d few additional remarks reIt also conforms to the usual analytical practice garding our similar prioritizing of the first PAC

of Schenker 2 in our concept of essential expositional closure.

What is meant by “essential”? Within smaller, Here Rothstein’s appeal to Koch probably refers earlier-eighteenth-century binary Structures, to the latter’s remark in the Versuch of 1793 con-_ _—- the _ most notable ancestors of sonata-form cerning the role of cadences, or “formal phrase- movements, the first part, when modulatory, endings” in the latter portion of the exposition: had driven to a single PAC close —in major“TIn the first period (exposition) of a symphony] mode pieces, a close in the dominant. This ca-

following the cadence a clarifying period [ein dence represented the essential generic task of erklarender Periode| is often appended that con- the first section—the single thing that part 1 tinues and closes in the same key in which the needed to accomplish. At times, as with some preceding one had also closed. Thus it is noth- of Domenico Scarlatti’s sonatas, the first secing else than an appendix [Anhang] to the first tion contained two cadences near the end: the period and both united may quite properly be generically obligatory one and an extra one considered a single main period.’’!° Similar con- completing a brief, additional passage of coclusions might be drawn from an even earlier detta-like reinforcement, an added passage of description by Georg Joseph Vogler in 1778 in new-key stabilization that Ralph Kirkpatrick his periodical Betrachtungen der Mannheimer Ton- wittily compared to the effect of the coasting on

schule. In the process of pointing out what he a ‘landing strip” once an airplane has touched called the five Perioden (sections) of the “first down.!” part” (exposition) of the first movement of an As the binary structure expanded and was 9. Rothstein, Phrase Rhythm, pp. 116-17. In a footnote First-Movement Form,” Journal of the American MusicoRothstein adds that Schenker’s analysis of the Eroica logical Society 27 (1974), 32, from which we have adapted does not take the first PAC as the end of S: as such 1t 1s the above translation. Stevens also provides the relevant exceptional. (On the other hand, we do accept that first original German text in the context of a broader discus-

V:PAC here, m. 83, as the EEC. In part our differing sion of Vogler. Along the same lines, in 1796 Galeazzi interpretation brings up issues that we deal with below maintained that the passo caratteristico (our S) ends with in “Some Schenkerian Implications”; in addition, we its “final cadence” (following, perhaps, a thematic repinterpret what follows immediately, mm. 84ff, as a rare etition) after which one may “elegantly add a new pe-

instance of a C° theme, discussed in ch. 9.) riod, called a Coda [our C, of variable length], which is 10. English version as in Introductory Essay, trans. Nancy an addition or prolongation of the cadence.” Galeazzi’s Kovaleff Baker, p. 199. In Koch, Versuch, vol. 3, section “Coda” appears to be more than a mere “codetta,” espe101, p. 305, the original reads: “Oft ist zwar nach der cially since he goes on to suggest that it may refer back

Cadenz desselben noch ein erklarender Periode ange- to (or at least lead back into) the original expositional hangt, der aber in ebenderselben Tonart fortmodulirt ideas (perhaps suggesting our P-based C): “it serves very und schlieBt, in welcher der vorher gehende auch ge- well to link the ideas which end the first part with those schlossen hatte; daher k6nnen wir ihn ftir nichts anders, which have begun it, or with those with which the secals blos ftir einen Anhang des ersten Perioden erkliren, ond part begins.” See Churgin, “Francesco Galeazzi’s und kénnen gar fiiglich beyde vereinigt als einen einzi- Description (1796) of Sonata Form,” p. 194, from which gen Hauptperioden betrachten.” Baker’s translation ofa we have adapted the translation. “formal phrase-ending” is Koch’s “férmlichen Absatz” 12. Kirkpatrick, Domenico Scarlatti (Princeton, N.J.:

(p. 306). Princeton University Press, 1953), p. 255. Kirkpatrick

11. Quoted in Jane R. Stevens, “Theme, Harmony, and used the apt simile to refer to the “reiterated cadences” Texture in Classic-Romantic Descriptions of Concerto following the “final resolution into the closing tonal-

The Secondary Theme (S) and Essential Expositional Closure (EEC) 123

transformed through thematic differentia- that with this multiplication of cadences, the getion and specialization, the central feature of — neric question of “essential closure” (when does the generically essential PAC must surely have S end?) opened itself to ambiguity and the pobeen retained—the idea of that first necessary, tential for differing interpretations. And as we

“touch-down” cadence before any “landing have mentioned, analysts today disagree with strip” of additional ideas. (The tonally generic regard to this matter. point of an exposition, after all, 1s to secure the A close study of the music itself repeatedly new key in part 2.) PACs in the second key of an suggests that the first satisfactory V:PAC that exposition are never to be taken lightly. More- goes on to different material remained a generiover, the first of these PACs produces the sense cally significant moment. It was normally the that at least the minimal generic-harmonic re- end of the essential exposition, to which differquirements of the form have been satisfied. If | ent kinds of (as Reicha put it) “accessory ideas” only for this reason, the first satisfactory (non- (our C-zone) could be added to produce any

repeated or otherwise overridden) perfect au- number of effects. To be sure, the first-PAC thentic cadence within an exposition’s part 2 1s rule is more of a guideline than a “rule”: it 1s conceptually privileged. Generically it repre- anything but inflexible. Composers devised a sents the attainment, however rudimentary, of | number of strategies to override that implicathe essential exposition—the accomplishment of | tion—1in other words, to demonstrate that the

the one thing that all expositions are expected first PAC was being reopened—and to defer to do. Thus this PAC has a dual function. On the EEC to the next PAC by means of immethe one hand, viewed more locally, it closes the diate thematic repetition or variation, the conthematic materials of the S-zone. On the other __ tinuation of an accompaniment figure, the later

hand, viewed from the point of view of the placement of a nonelided cadence, and the like. function of the S-zone as a whole within the A few examples will be provided in this chapter, broader structure, it marks the successful arrival and the topic of EEC deferral will be dealt with and cadential securing of the secondary key, the in more detail in chapter 8. accomplishment of a guided trajectory that had In short, we have learned that evidence found been generically “in mind” from the first mo- in the music itself regularly reaffirms our prin-

ments of the piece. (See figure 2.1a.) ciple of the EEC. Our analytical encounter with In considering this issue, it is also significant “new” pieces seems continually to reinforce to observe that there 1s no generic requirement these conclusions —and only very rarely to chalfor a major-mode exposition to have more than lenge them (although this may be clearer only as one V:PAC, although many in fact do. Therein our discussion proceeds in subsequent chapters). lies the difficulty. The obvious complication is This reinforcing happens, for example, when a

that instead of producing merely one or two clear shift to certain readily identifiable C-types V:PACs (as in the earlier binary structure), the is made following the EEC-point: with the EEC typical exposition might have sounded four, nailed down, the music 1s liberated into a freer five, six, or more, depending on aspects of the- closing-space or “appendix” (Koch’s Anhang) of matic restatement, concatenation of diverse the- “accessory ideas.” One eventually comes to rematic modules, and so on. Moreover, it was pos- alize that certain types of C!-theme (the onset sible, even common, for these later V:PACs to of C) became normative. These C!-identifiers bring an even more emphatic closure (both dy- (especially the forte P-based C!) can help to supnamic and textural) to thematic phrases. Which port the general proposition about the role of of these was now to be considered the generi- the first PAC (or if deferred, its later countercally “essential” cadence? There is no denying part), even as they can help in making judgity” in the “second half” of the binary structure. Since this binary form, the ESC. Malcolm Boyd, Domenico Kirkpatrick did not specifically focus on the attainment Scarlatti: Master of Music (New York: Schirmer, 1986), p.

of cadences, this passage is a bit obscure, but it appears 168, also quoted the phrase by Kirkpatrick and applied that he was referring to the additional [quasi-C-space] it to the “codetta ... reinforcing the closing tonality of cadences following what we would call, even within the section.”

124 Elements of Sonata Theory

ments about the location of the EEC in difficult point of division separating the one from the cases. All of these things need to be considered other (S / C). There 1s no way of predicting what in assessing the possibility of EEC deferrals past proportion of part 2 will be occupied by S or C. the first PAC in the new key. But the bulk of | Sometimes an expanded S blocks out the possibilthe evidence overwhelmingly bolsters our gen- ity of any C; sometimes C is merely a brief codetta eral conclusion. Both historically and generi- to S; sometimes C is more elaborate, unfolded cally, the essential exposition— by no means the with two, three, or more thematic or cadential complete exposition—should be considered as modules, filling up more part-2-space than had normally extended to that first satisfactory PAC S. (Figure 7.1 provides a visual suggestion of some proceeding onward to differing material. That possibilities.) The only caveat—albeit an impor-

PAC is given a special priority. tant one—1s that there seems to be a contextuAs a word of caution, we should underscore ally informed limit below which S-space would this one more time: the EEC need not be —and reasonably be considered too brief to provide a often 1s not—the strongest cadence within sufficient sense of closure to the freshly launched the exposition. Stronger cadences— more em- zone. (This situation this is discussed 1n chapter 8.) phatic expanded-cadential-progression cadences Developing a feel for the manifold ways in which (I6—predominant—V4-3—]),'3 trill-cadences, and part-2-space may be rhetorically subdivided (each

so on—often occur as reinforcement-work in bearing its own expressive/proportional implicaC-space. One should not determine an EEC on tions) is an important step toward understanding the basis of what one imagines an EEC should sonatas as a whole. “feel” like in terms of force or unassailably conclusive implication. Nor should one assume that

we are making grand claims regarding either Structure the completeness or the degree of the closure implied by the EEC. Its “‘closure’”’ may not in S may be articulated in an abundance of differing fact be absolute or “fully satisfying” from the — shapes: period, repeated period, sentence, hybrid perspective of the larger proportions or other phrase, and so on. Sometimes one finds chains telling factors within the exposition asa whole. — of these characteristic shapes (as in the multiThis first PAC closing the essential exposition modular S) linked either by subverted or evaded 1S primarily an attainment of an important ge- cadences—thus avoiding the crucial PAC — or

neric requirement— nothing more and noth- by some sort of cadence- or phrase-repetition ing less. It may be composed, however, in such or recapturing of pre-existing S-material, both a way as virtually to demand the “accessory of which techniques defer the EEC to the next ideas” of a more decisive C-zone, which can PAC down the road. As with P-themes one ofcertainly include broadly thematic and ener- ten finds nested hierarchies of structure within gy-gaining modules. Our main point remains. S. A composer might construct the antecedent With the first satisfactory PAC the exposition | and consequent phrases of a parallel-period S has now accomplished what it set out essen- as sentences. Or an S may begin by seeming to tially to do: to cadence decisively in the second — announce itself as a period only to have its con-

key, thus setting up and forecasting the paral- sequent undermined, requiring the conversion lel point of essential structural closure (ESC) in the of the whole into a larger sentence. This hap-

recapitulation. pens when the antecedent and “unsuccessful” consequent— or the antecedent submitted only to a repetition —are reinterpreted as the charac-

Proportions of S- and C- Space teristic a + a' of a larger sentence’s presentation section. The next module, often a contrasting As indicated above, part 2 1s normally subdivided one, will begin the b portion, or continuation, into S- and C-space, with the EEC marking the of this broader sentence. 13. Seen. 7.

The Secondary Theme (S) and Essential Expositional Closure (EEC) 125

MC 9

S C (usually C!, C2, etc.)

MC

°S (often S!-!, S!2, etc.) C (usually C!, C2, etc.)

MC 9

MC 9

S (S!-!, S!2) etc.) C (codetta-type only?)

S (S11, S!2) etc.) (no C) FicureE 7.1 Varying Placements of the EEC within Part 2 of the Exposition

A good example of multiple structural hi- b-portion, the continuation. A V:PAC 1s finally erarchies occurs with the secondary theme of — declared—as if with finality—at m. 77, and Beethoven’s Symphony No. 1 in C, first move- we immediately assume that it will serve as the ment (example 7.1). Here S begins with what we EEC. Yet this assumption, too, is instantly unfirst might suppose will be an antecedent (mm. dermined. Instead of moving directly into C, 53-—60—itself sentential) and a consequent. S-material is retained with a sardonic, pianisThe expected consequent, however, fails to simo, after-the-fact back-reference to the openclose with a PAC (m. 68), and the music pushes ing of S. In effect, this is an unanticipated extra onward (m. 69), as if in a struggle to shake loose zone, a darkened interpolation dropping simulthe generically required cadence. With m. 69 taneously into the minor dominant, just as the (and in retrospect) one converts the earlier view bottom of the earlier V:PAC seems to drop out of mm. 53-60 and 61—68 into that of two broad as one starts to plunge flatward, down the cirpresentation modules, a+ a', of a larger sentence cle of fifths, before the V:PAC is reinstated, and (in this case, each ending with a half cadence), with it the EEC proper, at m. 88.'4 A generically while mm. 69ff mark the onset of the sentence’s standard P-based C!-theme follows. 14. The definition of the EEC as the first satisfactory quished with a shift into differing ideas. In this case V:PAC means both that the material producing the ca- the typical P-based C! that follows (m. 88) confirms dence is not immediately recycled in a repetition and this interpretation. Chapter 8 considers EEC deferrals that characteristic S-melodic-material is also relin- more generally.

a7 EXAMPLE 7.1. Beethoven, Symphony No. 1 in C, op. 21, 1, mm. 53-90

A SS SSS P —y ’ ' OO Fo i oO [Allegro con brio] oo 2 _

OF 0 2 A A A SS A A

tee, _——__—_4—______* #9 te ___ ey | — ) — __ 2 _ 2 ep | —WvAN7> 5 ———— —

p y - id y - - g -_ oo

— Fam Fes TY

f) Sg ror if dio _ ao PI P|SN Pl A YP| fideo

| | | — 1+} t = Sf Sf

60 — ~

Pe

Aw 2 } o eP he — i all _ sf —p DO 2 ew LOOT PO # t ow YY Y yy 7 :

8a (A Le : f Tou | | YP gre, jy {| £if_{ £4¥T|Y fj. "4 «| lg¥y 7

pepe Te ga gm em eee UL ele

ee nS oo Cy we tee | a_i

{) a ae * a” A cee nen). ae a A De QO es QDQoR———

ee Edees OUD iTEAPSD eeOSeeLeE*"EO es ee aelr—“em—™"e ne eyTP = OD _>00nnwvuvnvw’N-r.-._@ De tee OE ee ote

| ea ae ee eee i

A eeo__fn On ey I |__ o_o en _ 8 _8_ a se 7 ———$——————————————————

sy ge _ #8 | ™“ #» #» #@g shy » !Onh ae _|

|? 4 4 4 enn oS Fa _ e. at Ce ee ee ee ee ee ee ee ee ee ee eee A A cA | AC GS Ad GT A GO (CO pn rt—“CSsYYtSC‘“‘(SSd”rSC“‘(S STC Tl

se a es ee

‘6g %° | ge ig, $i Aa Ss 2» it i i=» | * 1 1 1&+&+&+&3«2~ fF gg Ff |

Nee” i i al A ry

127

Te I pn i eeee|

aoO . . eee ..e EXAMPLE 7.1 (continued)

————————————— P ——————S———————— ee O_

A eeedeeaee es PB = | ,A§aes aa eel aa in ie a ed el ed eeaeea ee if)’ §§+ & 8 @£@,8 @ ff ££ ,f | Jif 2 @. °° °»°&»-F °° &§;x »+=ST ff #28 06h6hk.DLmLUmULmULUmULUL™ChU CFeSC,

Cresce.

ne i

87 St tPM — = Cf a Ce * Ty —=ps oes ae — YS 8 Fs es _ et ed y.

| fs i |e ”A — . ——~ GO CC iF a ee a ee a —, en < te £ 4 a ER PS ce res

Cl ee ee — es —_ es ee This is a strong sign of closure, a stock more common choice. S is usually introduced formula upon which composers couldring many as a reduction of sonic forces, a drop to a piano changes. From the perspective of Sonata Theory, dynamic after the preceding TR, which had composers will sometimes postpone a full artic- maintained or gained intensity through the ulation of the ECP until the C-zone. In other MC. So much seems obvious to those familwords, once again, in these cases the first PAC iar with the style, and yet the observation can will not be the strongest PAC within the expo- prove misleading, particularly when undersition — only the first satisfactory one. The EEC stood through the lenses of later nineteenthcadence may well be less strongly articulated than and twentieth-century transformations of the one or more of the cadences within C-space: sonata. It is by no means the case that cigh-

indeed, this is not uncommonly the case. teenth-century S-themes are predominantly Many S-themes will not begin directly with lyrical and cantabile, although many of them are. 5 in the upper voice: in these situations there The temptation to apply reductive, sentimentalmust be an initial ascent up to that 5. These as- ized, or gender-ideological nineteenth-century

cents can be vividly dramatized —sometimes descriptions of S-themes to late-eighteenthturned into considerable struggles, as often in century sonatas— Gesangsthema, “feminine” Beethoven. Similarly, the descent downward theme, and so on—leads to unwarranted conto 1 can be problematized (sometimes frustrat- clusions within both analysis and hermeneutics. ingly), and one frequently sees the last-moment What we actually find in the music is a wide vaundermining of the implied PAC as a highly riety of thematic types. Additionally, individual

characteristic strategy of prolonging S-space. S-spaces are anything but consistent in charThe implicit drama involved in producing the acter. They often contain much inner surprise, first satisfactory linear fifth-descent in the new wit, change, and contrast: sudden outbursts key (a task of central importance to the sonata of forte, quick drops back to piano, unforeseen as a whole) is often the whole point of S—its changes of mode, unprepared interruptions, implied difficulty in fulfilling its mission, the concluding forte drives toward the cadence, and production of the EEC. One should be attentive the like. No single adjective or thumbnail charto the adventures of S on its way to the EEC. acterization does justice to such a wide range Some of these issues are rejoined at the end of — of volatile possibilities. Nervous energy and this chapter in the subsection, “Some Schenker- rapid changes of strategy are as much at home

ian Implications.” within S-space as is the gentler dolce theme.

15. See n. 7 and Caplin’s discussion of the ECP within of supporting a complete phrase (of at least four meahis later Classical Form, defined in the glossary, p. 255, as sures) or group of phrases.” On Caplin’s use of the term

“an expansion of the cadential progression to the extent “phrase” see ch. 5 above, n. 10.

132 Elements of Sonata Theory

We regard the common drop to piano for the changes of character en route. What seems clear, beginning of S as a first-level default. Continu- though, is this: in terms of its rhetorical char-

ing to bluster forward by maintaining an en- acter and motivic/thematic derivation the Sergetic forte on the other side of the MC 1s a zone normally participates in a generic system considerably less common, second-level default. of melodic S-conventions, many of which are The notable preference for the former, the piano in sharp contrast with each other. More helpS, suggests that the usual function for this mo- ful than trying to assert what S is like “in genment 1s that of relaunch—a second start midway eral” is being aware of some of the most recthrough the exposition. The piano convention ognizable options within this constellation of may have been devised as a means of setting this conventional options. These options are best relaunch into high relief. Within the piano con- understood as only heuristic categories, and bevention certain types of thematic modules, at cause they sometimes overlap they are not to be least at the opening of S, were more appropriate regarded as airtight. They include but are not than others. Such a view accounts for the sev- limited to the following. eral characteristic types of S-incipits. While no

eighteenth-century theorist gave the slightest The Bustling, Staccato, Energetically Galant,

thought to inventorying all of the possibilities, .

; or Jauntily Self-Confident S

several of them noticed the piano convention (which was anything but an ironclad rule) and This is a familiar eighteenth-century S type sought to give abbreviated descriptions of some marked not so much by a cantabile character as by

of the effects within it that they remembered. light, strutting steps, much motivic repetition, It would be naive to regard Koch’s descriptions perhaps a leaning toward a musical eccentricity, from 1793, for instance, as limiting the range of — and a strong forward momentum. Sometimes possibilities or as generally adequate as covering high-alert, nervous quirks animate the efferves-

concepts for all eighteenth-century S-themes cent figuration—frisky, clipped, or Lombard (most famously, cantabler Satz and einem mehr — rhythms, Schleifer-like decorative slides, and the singbaren, und gemeiniglich mit verminderter Starke like. Occasionally such “stylized” themes begin des Tons vorzutragenden Satze [literally, “a more with a forte impulse; more typically, they begin singable idea, one usually to be performed with — piano but may contain a few edgy outbursts of a diminished strength of sound”’])!© Perhaps the —_forte. In virtually all cases the expressive effect

most engaging cighteenth-century description is that of the opening’s high energy continuing was that of Galeazzi in 1796, who cited S asa into the exposition’s part 2, a retained sizzle of new theme (una nuova idea) of sharply defined excitation now only barely constrained under character, a “characteristic passage or inter- piano wraps and eager to erupt again into forte. mediate passage” (II Passo Caratteristico, 0 Passo This theme type is especially typical of briskly di mezzo) introduced “for the sake of greater animated, midcentury galant works, but one also beauty” (per maggior vaghezza). He went on to finds it as a frequently selected option in later insist that the relevant music be “gentle, expres- decades. sive, and tender in almost all kinds of composi- A good example of this familiar S-theme type tions” (deve questo esser dolce, espressivo, e tenero is provided in the first movement of Haydn’s

quasi in ogni genere di composizione).!" Symphony No. 83 in G Minor (“Hen” —this The evidence found in the music supports theme doubtless contributed to the symphony’s these generalizations in only a limited way. nickname), mm. 46—59.!8 (Example 7.3a preMore important is the suppleness of change and sents the initial statement of S only, omitting its surprise within S-space, its potential for agile slightly rescored repetition.) Another locus clas16. Koch, Versuch, vol. 3, pp. 306, 364, 385. See n. 3. ization—not shown in example 7.3a—of the “blocked 17. Churgin, “Francesco Galeazzi’s Description (1796) medial caesura” (see ch. 3), ending with a decline of

of Sonata Form,” p. 193. See n. 3. energy into a III:PAC at the downbeat of m. 45. 18. The S-theme of the “Hen” follows a touchstone real-

The Secondary Theme (S) and Essential Expositional Closure (EEC) 133 EXAMPLE 7.3a Haydn, Symphony No. 83 in G Minor (“Hen”), 1, mm. 46—52

f nNetx he N S he S ~~teKOo K, J” OE wt rr Plrestets _- el — aan ae ae [Allegro spiritoso] 46

[p]

Obed utes fede sl] terete 5a a $= , Sy eft $= ee elif§ elie oe= = ee = es

5] Y ¥ ¥ ¥ Y Lf y Lf yi —— JT. ITT: PN gy ea i gy Ny

Oe i — ey ey ey ee ST Coo ee —/

sicus of this theme type may be found in Mo- the variety found within this type itself. This zart’s Overture to The Marriage of Figaro, mm. “contrasting” S may be distinguished from the 59-107. (Example 7.3b shows the first module “bustling” type cited above by its broader lyriof this lengthy, multimodular theme.) Other cal lines, its tendency toward legato or slurred touchstone examples include the S-themes of articulation, its flow and continuity of texture the first movements of Mozart’s Symphonies (instead of being broken up by frequent rests or No. 22 in C, K. 162 (mm. 32—48), No. 25inG abrupt changes of topic), and its general sense of Minor, K. 183 (mm. 59—74, also an example of | contentment with its piano dynamic (as opposed a piano S-theme repeated forte), and No. 34 in C, to giving the impression of being eager to burst K. 338 (see the earlier example 7.2, especially out of this containment). The difficulty is that

from m. 44 onward). cantabile S types shade by degrees back into the

bustling types and vice versa. Many themes fall

, cae between the two heuristic categories.

one lly “’Singing™ or Gracefully The S-theme of the finale of Mozart’s Symphony No. 40 in G Minor shows some of the This is the S type long assumed to have been complexities potentially in play. (Its initial statestandard for sonatas in the decades around ment, mm. 71—85, 1s shown 1n example 7.4.) At 1800—singing melody, piano, over accom- first glance it would seem to be an archetypal panimental bass—and it is doubtless the type illustration of the cantabile type, and yet it simulalluded to briefly in Koch (cantabler Satz) and taneously suggests some intersections with the Galeazzi (dolce, espressivo, o tenero). It 1s also the preceding type as well as displaying a few curi-

type that would flower into the characteristi- osities in itself. Following a forte TR and HI:HC cally expansive, maximally contrasting Ge- MC (m. 70), S begins with a drop to piano at sangsthema of many nineteenth-century works. m. 71 and proceeds as an asymmetrical (8 + 7) Because our seeming familiarity with this cat- period with varied and compressed consequent egory is so ingrained— and often so lacking in (mm. 79-85). On the one hand, the legato texnuances—it 1s helpful to remind ourselves not tures throughout are characteristic, giving the only of the many other types of S but also of — impression of an easy flow through S-space, and

ee Fa.

A EXAMPLE 7.3b Mozart, Overture to The Marriage of Figaro, K. 492,

mm. 59-67

59 [Presto]

DP A FkJp ee (DoD cee Jp f CS._ oSy2 0727 SS ageee |sl a“Tite |eSaSS QO SS s7?.2_2s 27.22.” Cy Ae Fe se eo SS eews is 7777 eeGO ee sis

Ip Ip JP

62 — Dp ige eos ? Mm i e_ DT A Oc ” _ SSO SS pe TR a —_L AS

Jp

IP Sp

as 64 — hy op

we ™ | Spe aaee a Pw Ed wVW oo hy JP es ee “toy — : -_

a Jp 6 SSPha Ae

twee eee ? ——— fn i Wo _ _ 8 —_ Eo a ea eea Se a2 oer=|2

eB oR. > -UwwNW— _— —_ 8 --re oo oo ETE

660

Yate *—_¢_~ 6 _3_ 2.2 —l_ a EE© ESS SBS OO OOOTE —

Ley © _ © _ ae eee i a. toa} ae —_

po...

The Secondary Theme (S) and Essential Expositional Closure (EEC) 135

7 ns a

a Oe eT Be ao ereers EXAMPLE 7.4. Mozart, Symphony No. 40 in G Minor, K. 550, iv,

mm. 71-85

[Allegro assai]

Joy wy Oi o_o 7 7 Ld P

mfp

i ur , ij} eo = ey er io f) oN _ eT “y o o :

f_; eo s _9 os" ee > —?>—(_— ee Ce oe_*a"|eo i ee eee oe Soe

82 re ae as yo ee ea et (G Pp OM ge eRa a®ggl

Ne = NL »7

the piano dynamic 1s generally constant through- clearly content with its piano dynamic, the catout the passage. On the other hand, although like tread of its predominant light staccatos sugthe theme probably has a predominantly canta- gests less a lyrical dolce theme than a tender and bile basis, it also suggests features of the bustling elegant passo caratteristico.) type. The Allegro assai tempo and the residual retention of the Trommelbass impulses from TR

. . The P-Based S

in the descending bass (mm. 71—74, 79-80) . ase

continue a sense of nervous tension and forward In this type the incipit (at least) of the S theme 1s press from the preceding music. The compres- either identical to or an easily recognizable varision of the consequent (a sign of impatience?) ant of the P theme. At least in their openings P may also contribute to this “mixed” effect. In and S are often sonorously congruent: if the one addition, the extra push of the mfp up to 8 in m. had been sounded piano, so is the other—al73 (the onset of an “if only” sigh?), fortified by though exceptions do occur. ‘The main thing 1s two preparatory appoggiaturas on 5 (m. 72) be- that the onset of S is a recapturing or re-soundfore the leap and echoed with the fleet, iambic ing of the initial idea. The P-based S gives an alsteps of mm. 75—76, also suggest an underly- tered emphasis to the idea of a two-part exposiing dialogue with an all-propelling higher en- tion. Both parts set out from the same basic idea, ergy, more characteristic of the bustling type. even though what follows them will be differThe III:PAC in m. 85 would serve as the EEC ent. Consequently, this type of sonata suggests were the entire period not subjected to a varied that the “narrative subject” (the central musical repeat (mm. 87—101), which defers the EEC to character or idea of the “drama,” stated by P) is m. 101. A standard P-based C! follows in m. still in evidence and is now ready to undertake 102. (Compare, for example, the well-known the second phase of the exposition, S and the S-theme of the first movement of Eine kleine production of the EEC. Should the broader S / C Nachtmusik, K. 525, mm. 28—35, a clearer par- space continue, however spottily, to refer here

allel period, 4 + 4, with EEC at m. 35. While and there to ordered material in P-TR, as if

136 Elements of Sonata Theory

touching upon certain stations of fleeting refer- music in general: one different-sounding theme ence, one might suggest that the expositional is motivically derivable from what has preceded rotation as a whole is being conceived as two it. In the 1970s and 1980s Carl Dahlhaus revived subrotations, or two varied cycles through simi- the concept and argued that it was one of the lar materials. In such cases the second subrota- keys to understanding Beethoven.?? When this tion (starting with S) would be presented as a occurs, it is clear that P’s potential for “thought”

varied recasting of the first. has grown into the contrasting S: though conUnlike his contemporaries Haydn appears trasting, S is a response to P or an outgrowth of to have adopted the P-based S as a first-level ongoing motivic elaboration. It is by no means default (part of his individualized customiza- limited to the music of Beethoven, of course. tion of sonata practice), although contrasting A classic instance may be found in the similar S-themes are also found in his works.!? The intervallic content and general contour of the P-based S produces what has sometimes been opening of S (mm. 23-24) in the first movecalled the “monothematic” exposition (or so- ment of Mozart’s Piano Sonata in B-flat, K. 333, nata), although in most cases this is a misnomer: with that of the P-theme (mm. 1—2). Such subfollowing the EEC, Haydn normally presents an tle resemblances continue as S proceeds, invitaltogether new theme for the C-zone. As always ing an interpretation of the whole exposition as

in Haydn there are exceptions, and from time subrotational. to time one finds the same motive or incipit launching P, S, and C, as in the first movement The Forte S of the String Quartet in C, op. 74 no. 1. Such a dogged retention of opening material may be This is a lower-level default than the piano S. As referred to as the “emphatically monothematic”’ such, it is usually to be interpreted as an overexposition. Occasionally one finds the P-based riding of expectation, the surprising, threatenSin Mozart (finale of the Symphony No. 39 in ing, or desperate refusal to drop back to piano to E-flat, K. 543—which “emphatically” extends launch part 2. By late in the eighteenth century the principle to the C theme; first movements the forte S seems to have been an even less freof the Piano Sonata in B-flat, K. 570 and the quently selected option, although it certainly reString Quintet in D, K. 593), although Mozart mained as a possibility, especially when reacting selected this option considerably less frequently to a weakened or deformationally “flawed” TR than he did that of the contrasting S. See also and/or MC. In earlier instances the forte S seems the additional considerations below, “P- and to be an overflow of the principle grounding the

‘TR-material in S-zones.” energetic, bustling S into the forte dynamic (the brimming-over of sheer exhilaration beyond

S as “Contrasting Derivation” from P the MC, continuing with no dynamic letup).

In other pieces it suggests the relentless persisThis type of theme 1s related to all of the previ- tence of a Sturm und Drang texture refusing to ously mentioned types: contrasting and P-based. give way to the more generic piano S. It was Sometimes the materials for the “contrasting” S particularly useful in short-breathed, modestcan be understood as being motivically related to scale works (such as some of the briskly vigorP. This is what the musicologist Arnold Schmitz ous early symphonies of Mozart).

early in the twentieth century called the prin- In still other cases the forte S could take on ciple of “contrasting derivation” for S. Schmitz a broader, even rougher character, with the viewed this as a central feature of Beethoven’s implication of having sternly overridden (or 19. Cf. Somfai on “the ‘varied primary theme’ as sec- trans. J. Bradford Robinson (Berkeley: University of ondary subject’” in The Keyboard Sonatas of Joseph Haydn, California Press, 1989), p. 84; and Ludwig van Beethoven:

pp. 270-74. Approaches to His Music, trans. Mary Whittall (Oxford: 20. Schmitz, Beethoven’s ‘Zwei Prinzipe’ (Berlin and Clarendon Press, 1991), pp. 38, 51—52, 96, and elseBonn, 1927). Cf. Dahlhaus, Nineteenth-Century Music, where.

The Secondary Theme (S) and Essential Expositional Closure (EEC) 137

erased) the hope for a more normative, gentle S as Virtuosic Figuration

S. These passages usually suggest the onset of . .

an S-deformation. Such 1s the situation in the In some early keyboard sonatas S is occasionD-major Andante of Mozart’s Quartet in A ally elaborated with rapid, idiomatic figuration K. 169. Following a disturbingly unusual piano (broken chords, sixteenth- or thirty-secondand minor-mode TR (mm, 1—15 are built asa note figures, lightning-fast scalar or other passentence with dissolving continuation), a mi- sagework) rather than any sort of self-evidently

HC7» MC (m.ONO 15), and eo lyrical theme.mentioned In his study of Haydn’s keyboard pores de © i:we BE sonatas Somfai these as “contrast-

bar of silence (m. 16), S bursts out in an anxiety- . ever

wracked forte in the “wrong” key, F major (re- ing themes built on arpeggios.”?! Whatever

tine to th dine. | 6 ID Ye ‘ 7 the figuration-type, the result is that S takes on

The saitial ik rose oe tule i ccomeneed on G the character of a florid display of idiosyncratic minor (m, 21) and A minor (m 25) before lead technique, sometimes additionally at an indiing toa minor-mode “PAC (1) am. 34. which cated or implied forte dynamic. A good example should probably be understood 1s an EEC ac occurs in the first movement of the Sonata in

. . F, Hob. XVI:23. modules restore the major dominant. . .

complished under high strain. Subsequent C- - ob. XVE23 in Tesponse to a ViHC MC

, , in m. 20 S takes off in C major, m. 21, with a

| a ; 3 . A rel found i

As this example suggests, S-themes that be volley of thirty-second-note figuration, much gin forte often compensate for MCs that are un- of it scalar. A related example may be found in

sual Thi hen TR and the first movement of the Sonata in E-flat, Hob. ES I OCONEE an XVI:25, with V:HC MC at m. 14 and an ener-

the MC are submitted to a dynamic deforma- . . .

x" ' YM at m. 15.

tion —when they are articulated quietly (pi- getically figurational, forte S launched in B-flat ano) or when a diminuendo has been applied to

the drive to the preceding MC. In general, if S begins with a strong dynamic— particularly The “Learned-Style” or Fugal/Imitative S

forte—one should look for signs of decay, en- _ ervation, or faltering in the preceding TR and This is an infrequently selected S type reserved

Pad. j for special effects. Here the composer begins

MC (loss of nerve, inability to sustain a forte, S_space by plunsing into the added 4 reduction of dynamics, alteration of mode, and “SP y Pe ne te BEES EMETBY OB so on.) In these cases, as suggested also in chap- earnest resolve of fugal entries, which then usuter 3, the forte S takes on a compensatory role in ally run out to be short-lived 1s S proceeds toresponse to generic problems composed earlier ward its cadence. Quite unlike that of the canta“ato the viece. In the first movement of Movart’s bile S, the impression 1s that of following the MC

Piano co in F, K. 280 (example 7.5), the with a continuation of fidgety vibrancy, compopreceding MC in i 26 has been © amnieved to sitional self-display, or sparkling, expressive zest.

1 piano dynamic. im 27 is then blurted out At times these qualities suggest the character of

P y ep 4 the energetic S.only (They may even be combined with a measure of forte to be hushed back . ; . with it— or with the forte S.)28—30, Added to. a. to three bars of piano normativity, mm. .it,« how5 whereupon the dynamic process is repeat- ever, 1s the self-conscious, “learned-style” texed—blurt and hush. (Compare this with the ture, in part a memory of a now-eclipsed prac-

. 7: elaborations tice (in some contexts also witty of the same idea in an the age-old reca- _. i sign of pitulation.) S proceeds as a compound sentence compositional practice itself), revived under the wvith the contrasting continuation besinnine in aegis of the lighter and fleeter galant style. There

mm. 35 and : ) ducine th EC ° are a few illustrations of it in Mozart, some of Seo Be euemenay Prose is an them extremely celebrated. They include: the

43. Fourteen bars of closing material round out

the exposition second movement (Allegro) of the String Quar21. Somfai, The Keyboard Sonatas of Joseph Haydn, pp. 267—69.

|= Hee 32 f

EXAMPLE 7.5 Mozart, Piano Sonata in F, K. 280, 1, mm. 22-43

f) Re ee ee | , ery | Fa. Tl, CU aeT e—*FeeG TT eer.eT | ews) PN ea fi A a ——— aN -8P Piano en fee? Cm =( ay YY ee aeeA; NS A ey =O aenaSS ©eeeS DO cnc Hea? te Te ih i tt a rrPee ee ee Re ee a Nee’ Pp —~ | Allegro assai|

22

ee ae ee ee ED come OOee GO| fifiee Iter ewee |ce| GOD Ul Oe ei asI OO ll er

YS SF f) eu o> o>

25

fi hy | | “| | *riher .» | | i, ff ™ Ft fT fT | fT Jf fT fT tf fT | FF | | ws: | ae em lm t—“‘iYSCéidS:C“(RYRRRNUCSQSAS“C( (RU YC tC*dYTsi‘C( oe 2+ | ye es ce mK gm aT Ky EE a a ee ee ey ee SS | BA! AeyPtZé : ee ae

fe a eee BO eS LT €onNn’’’'’'’'-—"—z # ewe Poe fae i/o RN, NTNWWWWWTNANAC_:——"MTDWD.TVT_‘"—— foe ‘4 NWN02eE-@ 0nmaNn0W0wWwW>

The Secondary Theme (S) and Essential Expositional Closure (EEC) 139

tet in G, K. 80 (mm. 16ff, here merged with the musical narrative that the composer 1s unthe idea of the forte S); the finale of the String _—_ folding. S may give the impression, for example, Quartet in G, K. 387 (upbeat to mm. 52ff, again of arriving too early in the exposition— perhaps

urgent, forte); and the finale of the String Quin- entering hastily after a too-early MC or a I:HC tet in D, K. 593 (mm. 54, beat 2ff, a five-voice MC that 1s unsatisfactory in one way or another. set of fugal entrances proceeding from top to In such a situation the first S-module needs to bottom, perhaps more playful than earnest, here be rescued by later S-material. also with a drop to piano). Clear allusions to the The MMS strategy is readily found in Haydn, practice are also found in the finale of Mozart’s Mozart, and Beethoven, and the expressive aims Symphony No. 31 in D (“Paris”), K. 297 (mm. in play can differ from composer to composer 45ff), and in the first movement of the Clarinet and from piece to piece. In Beethoven, where Concerto in A, K. 622 (mm. 25ff, an imitative strenuous striving within the S-zone is common,

stretto within Ritornello 1). the strategy can suggest that accomplishing the requirement of the EEC can be a monumen-

The Multimodular $ (MMS): Lengthy tal or nearly impossible task. In one common

S-Themes (or S-Modular Groups) Beethoven scenario S begins more or less un-

problematically but then runs into difficulties, It often happens that S does not proceed to an represented by an ominous, murky, or threatenefficient PAC (EEC) at all, blossoming instead ing central module (often with a diminishedinto a bouquet of differing S-ideas (each an seventh coloration), and finally breaks through S-module), none of which, except the last (the to a decisive cadential module that “heroically”

cadential module), articulates a PAC. Some- claims the EEC. The first movement of the times the cadential completion of S 1s consis- Fifth Symphony is paradigmatic. Following the tently undermined for a long span of time. The emphatic hornsignal articulation of the MC in composer can stage S as frustrated in its pursuit mim. 59-62 (which deformationally combines of a decisive EEC, obliging it to pass through features of MC and caesura-fill), a dolce S sets differing thematic and expressive modules, asif —_ forth in m. 63, passes through a middle section seeking the right one to produce the EEC. This of considerable uneasiness (mm. 83-—c. 93), from produces a chain-like, multimodular S (MMS) out of which it pulls itself, in a grand crescendo

that can change topics, styles, or theme types of growing will, into a decisive module of hefrom link to link. If we wish to distinguish roic confidence (mm. 94—101). The moment of among the S-modules, all of which are normally the EEC (m. 110) is celebrated with a P-based C sounded before the first satisfactory PAC in the (mm. 111—22), here transforming the original, new key (for that 1s embedded in the definition fatalistic motive into a victory shout of tempoof S), we may employ the decimal designators rary Overcoming.

explained in chapter 5: S!!, S!2, S!-3, and so on. When S themes unfold in three modules, (An S? theme would be rare, since the designa- as with the S-zone in the first movement of tion implies a new or differing S-module that Beethoven’s Fifth Symphony, we sometimes persists, contrary to the definition of S, beyond speak of a trimodular S (TMS). The MMS- or that first PAC. This could happen only in cases TMS-effect can be produced in varying graof EEC deferral, a topic considered in chapter 8. dations of strength, clarity, and expansiveness. A cogent argument needs to be constructed in Sometimes the separate modules are distinct,

defense of each case.) easily distinguishable; sometimes the first modIt may also be that the first module of S ar- ule dissolves or decays almost imperceptibly into rives purposely weakened or flawed in the sense the second; and so on. In all cases they produce that it cannot (or chooses not to) produce the the adventures through which the S-zone tracks PAC/EEC. Obviously, such a situation does not on its way to the attainment of the EEC. (One betray a compositional defect or a problem that should not confuse the TMS with the trimoduthe composer 1s experiencing personally. Rather, lar-block effect, TMB, which is signaled by the such unsettling modules play intended roles in presence of apparent double medial caesuras. A

140 Elements of Sonata Theory

special S-complication, the TMB 1s dealt with ways, the contrasting S followed by the non-

in chapter 8.) contrasting S).

of S-Zones . . ,

This happens in the first movement of

a. , Haydn’s String Quartet in G, op. 33 no. 5. The

P- or TR-Material in the Interior V:HC MC occurs in m. 48 ‘followed by a fermata!), and S!! begins, dolce, in D major, in m.

The most common location in which one finds 49. The sentence structure appears headed for P-material in S-zones is at their openings. ‘This an unambiguous PAC (EEC) in m. 65, but at is the normative P-based S, and as already men- the crucial “tonic-chord” moment the upper tioned it 1s especially characteristic of Haydn. voice replaces the expected scale-degree T with Consider, however, those sonatas that do not 3. (Although the point may be arguable, we feature a P-based S. At least as a first-level de- shall assume here that this does not represent an fault, when S (more precisely, S!-') has declared implied PAC with elided C-theme beginning itself to be something other than P-based, it on 3: Haydn’s shift from the leading-tone, 7, to simultaneously agrees that obvious P-material 3 seems staged as a surprise, sforzando, with the is not likely to play a major structural role in upper voice veering away from the PAC impliS. (One exception: it may happen that a com- cation.) At the downbeat of m. 65 the motive mon phrase, counterpoint, or accompanimental that had grounded the P-theme proper bursts figure binds together both the P and S themes, in, apparently claiming its due. This P-based S!-? as in the famous bass figure, mm. 65—66ff., in (or S!-3, depending on how one had labeled the the first movement of Beethoven’s Fifth Sym- preceding sentence) launches an extension that phony.) Another way of considering this: If S eventually produces the EEC at m. 89.??

begins with a new theme, this generally signals Equally significant are those passages in that the return of the P-idea is being reserved which S proceeds in such a way as to suggest as a possibility for the onset of C, although that that it is either delaying the EEC or having

possibility need not be acted upon. difficulty in bringing it about. Into this situaStill, things are not always so clear. From tion P-material— especially its incipit— can intime to time one encounters a contrasting (not tervene to take control of the drive to the PAC. P-based) S-zone whose apparently normative This can be interpreted in a number of ways. S drive toward an EEC 1s thwarted, blocked, may simply “drift” back into a recollection of P, evaded, or otherwise attenuated, only to lead at which point the discourse gains in point and to an obviously P-based module—a surpris- focus, vectoring more clearly toward the EEC. ing change of course that extends the S-zone In other cases P breaks in as a genuine intervenbeyond its expected end-point. This interven- tion. In still other cases it may appear that Ping P-based module, then, ultimately brings S material is called upon at the end of S to effect into the EEC. The effect of this P-based S!-? the EEC in a recovery operation. When any of (or S!3, S!-4) varies from work to work, but it this happens, one might also get the impression typically suggests a fresh burst of compositional that the P-based material seemingly “reserved” energy and the onset of an extension or new as a potential for C! is summoned to appear prephase of S. In Haydn it can signify a change of | maturely (in the space originally set aside for Splans, seeming to decide late into the game to material proper) to direct the push to the EEC. fold in a reference to his own first-level-default A locus classicus of this procedure occurs in S-choice, the P-based-S (as if having it both the first movement of Mozart’s Quartet in A,

22. That all of this is to be understood within S-space ing to the third of the chord instead of the root. (Nois bolstered by additional musical evidence. The first tice also the back-reference to the abandoned S!:! in the violin in mm. 78—82 recaptures the melodic descent first violin, mm. 83-84.) A final, freer recapturing of of mm. 61—64, with some chromatic alterations, and the melodic descent in mm. 86—89 finally does provide

although it now does resolve 7—1, this time it is the scale degree | in both outer voices, thus effecting the bass that overrides the expected PAC (m. 83) by leap- EEC at m. 89.

The Secondary Theme (S) and Essential Expositional Closure (EEC) 141

K. 464. S begins with a new theme, a brief pe- Somewhat related to the unexpected appearriod at m. 37. A varied repetition ensues at m. ance of P toward the middle or end of S is the 45, but that restatement, expanded and spun out appearance of TR-material at that point— TRat length, postpones the EEC remarkably and based motives or themes that spur the music finally accomplishes it (m. 83) with a similarly onward to the EEC. Here the situation can expanded variant of P (mm. 69-83). Further be similar to that discussed above: a lengthy S examples may be found in the first movements may be presented as problematic or dilatory in of Mozart’s Quartet in B-flat, K. 589 (m. 61, its motion toward the EEC, and a part of TR a blurted intervention) and Beethoven’s Piano is called upon to complete that trajectory. An Sonata in C Minor, op. 10 no. 1 (m. 86). The example may be found in the first movement initial movement of Beethoven’s Piano Sonata of Beethoven, Piano Sonata in C, op. 2 no. 3 in D, op. 10 no. 3 (at m. 67, following an unex- (TR intervention, m. 61—taken from mm. pected lapse in S in m. 66) provides a generally 13ff— with EEC at m. 77). similar situation. One of the most dramatic illus-

trations occurs in the S-zone of the first move- Minor-Mode Modules within S ment of Beethoven’s Symphony No. 2 in D, op.

36, in which S seems literally to break down Sometimes the first S-module within a mabefore being rescued by P-based, “premature” — jor-mode work makes its appearance in the mi-

C-material (mm. 103ff, with the EEC finally nor dominant (v) with the implication of tragsecured at m. 112,?° followed by the P-based edy, malevolence, a sudden expressive reversal, C!, now in its “proper” place). An ominous, or an unexpected complication within the mupsychologically negative example may be found sical plot. This happens in Mozart’s D-major in the opening movement of Schubert’s Piano Overture to Idomeneo (S begins in A minor, upSonata in A Minor, D. 845. After the complete beat to m. 46, 1n the aftermath of a paradigmatic breakdown of the “new,” C-major S-theme, 1n- blocked MC), in the finale of Mozart’s Piano trusive, P-based material slithers into the empti- Sonata in F, K. 332 (m. 50; S begins in C miness in C minor (m. 64, pianissimo) and accom- nor), and in the first movements of Beethoven’s plishes the EEC on its own sinister terms at m. Piano Sonata in C, op. 2 no. 3 (m. 27; S begins 77. This is elided directly with a C-major clos- in G minor) and Quartet in A, op. 18 no. 5 ing passage of “C as S-aftermath,” mm. 77—90, (m. 25: S begins, somewhat “demonically,” in interlaced with disturbing interruptions from — E minor). In virtually all cases the minor-mode

the dominating P-head-motive. effect is corrected later in the exposition, often It may also happen that a motive froma later = within S-space itself, as in the F-major Adagio part of P (not its opening) makes a passing ap- movement of Haydn’s Symphony No. 102 in Bpearance in the middle of S, apparently as a kind flat, with S in C minor, m. 9, brightening to of ongoing reference— or that various motives C major only in m. 13 in preparation for the from P, including related accompanimental C-major EEC at m. 14. (The effect is replicated, ideas, keep rearing their heads within sectors of |= mm. 25-30, in the non-normative, varied reS, as, subtly, 1n the first movement of Mozart’s peat of the exposition.)

String Quintet in G Minor, K. 516. Sometimes This procedure may have been rather comthe “interior” P-figure changes the expressive mon in the middle of the eighteenth century. In track of S-space and moves on its own toward Sonata Forms Charles Rosen singled it out as the the EEC, as in the first movement of Mozart’s first of “three [sonata-form] stereotypes of the Quartet in D, K. 575 (mm. 49-50, a fill-figure 1750s and 1760 that were [later] to disappear.” 24

taken from P, mm. 7-8). The claim on behalf of its disappearance is exag23. The cadence at m. 112 may appear to be an imper- and most of the winds, is a PAC. Nonetheless, the presfect authentic cadence, with the third appearing above ence of the third in the flute does weaken, though not in the flute. More likely, the third is a cover tone, and nullify, the EEC-effect. the essential structural cadence, outlined in the strings 24. Rosen, Sonata Forms (Rev. ed.), pp. 153-54.

142 Elements of Sonata Theory

serated, but it is certain that the effect was well the first movement of his Symphony No. 31 in known among the contemporaries of young D, K. 297, “Paris,” m. 52 (S proper at m. 53). Haydn and young Mozart: it existed as a nor- Other types of S°- or S!-°-options are more mative expressive option within their concep- complicated. Following an MC, it sometimes tions of S themes. Whether or not its frequency happens that a thematic S-module sets out over

diminished toward the end of the century, it a prolonged dominant in the new key. This was used by Haydn, Mozart, and Beethoven to dominant typically lasts for several measures,

produce telling effects. then shifts to the tonic for the sounding of a difRelated to this are major-mode S themes ferent idea (or thematic module) that seems to that turn minor (“lights out”) with their rep- be more securely grounded within S space. Unetition, on the way, that is, to the EEC. This less it is sustained for an unusually long time, we is a technique sometimes adopted by Mozart, normally designate such an S-holding-pattern as in the first movement of the Symphony No. over the dominant as S!-°. Quite often the brief 38 in D, K. 504, “Prague” (S-collapse to A mi- holding-pattern appears twice, thus taking on nor, m. 105; restoration of A major, m. 112). the role of a sentence presentation. It is almost Also related are minor-mode intrusions midway always succeeded by a new S-module beginning into normally major-mode S-space (Beethoven, over the tonic—very commonly the continu-

Quartet in B-flat, op. 18 no. 6, first movement, ation of an ongoing sentence. In general, the S at m. 45, decaying into minor by m. 49, re- S!° strategy sets up the more decisive arrival of gaining the major at m. 62) and the technique the subsequent new idea: it 1s a way of staging of turning an S-theme minor with the onset not its entrance. Even though they are preparatory of an S-repetition but of a new thematic module ideas, S® and S!-° modules usually participate in within an MMS. For the larger implications of — the standard range of S-types. They normally all such minor-mode matters, see chapter 14. feature a drop to piano and display “thematic” S-features (perhaps energetic, perhaps lyrical). As themes they do launch S-space and yet seem

S° and S!.° Themes also to prepare the way for the arrival of something different that 1s more stable. Because the Each of the zones of an exposition—especially dominant prolongation usually lasts for several

P, S, and C—may begin with a preparatory bars, S°- and S!-°-ideas may be distinguished module that sets up or otherwise precedes what from S themes that happen to begin on a shortstrikes one as the “real” initial theme of the zone. lived V chord within the new key. For such initiatory moments we have devised the The immediate effect of an S® or S!-° module Zero superscript (either O or 1.0, the former for depends also on the type of medial caesura that stronger effects, modules that seem more clearly precedes it. Such a module can appear in two

separable from what follows) whose general different contexts, each of which carries difprinciple was discussed in chapter 5.2° The light- ferent connotations. It is to these that we now

est type of S!-°-effect occurs when an accom- turn. panimental figure, vamp, or rhythmic stream

is laid down 1n advance of the S-theme proper. . .

Illustrations may be found the first movements S° or SI Following a V:HC Medial Caesura

of Mozart’s Piano Sonatas in C, K. 309, mm. (or HI: in Minor-Mode Sonatas) 33-34 (also in dialogue with the norm of cae- Here the dominant underpinning the S® or sura-fill of the 5—1 type, even though it follows S!-0 theme retains the MC’s active dominant, an MC-gap; S proper ensues in m. 35) and again which continues to ring through the succeedin C, K. 545, m. 13 (S proper at m. 14); and in ing music as momentarily fixed or immobile. 25. Here we might emphasize once again that the labels of how “independent” the relevant thematic module S° and S!-° are for the most part interchangeable. The might be. The 1.0 label suggests a closer interconnectonly distinction between them lies in one’s assessment edness with the material that follows.

The Secondary Theme (S) and Essential Expositional Closure (EEC) 143

Consequently this type of zero-module func- In minor-mode sonatas the equivalent tions locally as a prolongation of the caesura- would be an S® or S!-° theme that begins (over dominant itself, that is, much like caesura-fill the dominant) following a III:HC MC. This of the 5-7 type. The distinction between the would include such extraordinary examples as two— expanded caesura-fill and a genuine S° that found in the first movement of Beethoven’s or S!° module—can be difficult to make, and Piano Sonata in C Minor, op. 13. (See also the perhaps in some cases we should not make it at discussion in the previous chapter, along with all. It may be that this kind of zero-theme 1s a example 6.2.) Here the presumed “S” theme thematically emphatic subset of that type of cae- occurs at considerable length, tonicizing E-flat sura-fill. The impression given by this context minor (!), over a prolonged dominant (albeit is that the caesura-gap has been held open: the one that is shifted about in the middle of the gears have been pulled apart, awaiting re-en- theme), mm. 51—88. After several measures this

sagement with S!! proper. is released to E-flat major (m. 89), and a new A locus classicus of a more prolonged, S° theme theme is produced over the tonic at the point of

occurs in the first movement of Beethoven’s an evaded PAC (the upper voice momentarily Eroica Symphony (example 7.6). The forte-dy- drops out). The lengthy E-flat-minor theme namic V:HC MC 1s reached with an almost (over V) is an unusually extended S°— perhaps disturbing abruptness at m. 45. A new, “quest- suggesting that, absent the immediate presence ing” theme, piano, is sounded over the dominant of the generically expected E-flat major, the 1n1of B-flat major with the upbeat to m. 46. This tial S-idea 1s wary of declaring itself as a fully dominant in the bass is prolonged for several confident or successful theme. The release into measures — with thematic material above —un- E-flat major, then (m. 89), launches what must

til the moment of its decisive tonic resolution be regarded as S!, even though its opportunity at m. 57, which then launches a new theme, S for thematic singularity has long since been proper: the gears re-engage. Mm. 46—57 con- ceded to S°. Such a claim may be easy to misstitute S°. Notice that, especially toward the end interpret. We are not suggesting that the sec(mm. 55—57), they behave much like a thematic ondary theme begins with the E-flat major idea caesura-fill of the 5-1 type. M. 57 begins S!-, in m. 89: this 1s indefensible. We do acknowlnow grounded on the tonic: this is one of the edge that the S-theme begins on E-flat minor few themes of the symphony that Beethoven in measure 51 but observe that in its manner of kept more or less invariant in his multiple con- deployment this theme 1s more closely related tinuity drafts, and it is also a theme alluded to to the concept of S°. S° themes belong emphatiand probably confirmed as S in the coda, m. cally to S-space.?’ 673.26 (Compare this also with the first move- It may occasionally happen that a prolongedment of Beethoven’s Ninth Symphony, with an dominant opening to S-space—an S?® or S!9 S° or S!-° theme —a briefer idea— that emerges identifier— does not so much shift decisively to in m. 74 after a stifled, somewhat deformational a clear, differing S!' module beginning on the MC-attempt. S!:! follows in m. 80 over the new new tonic as merge fluidly into more forward-

tonic, B-flat.) directed music, as if the zero principle comes 26. On the drafts, see Lewis Lockwood, “Eroica Per- in the web of transformations. Around it are shaped thespectives: Strategy and Design in the First Movement,” matic units that have much further to go before they Beethoven: Studies in the Creative Process (Cambridge, reach their final linear and harmonic form” (p. 131). It Mass.: Harvard University Press, 1992), pp. 118-33. is also instructive to compare the Eroica S° example with “One feature [of the continuity drafts] 1s particularly the nearly parallel situation —also in E-flat major—that striking. The passage that I am calling ‘Unit C’ in the occurs in the Quartet from Mozart’s Idomeneo, “Andro

exposition ...1is [apart from the very opening]... the ramingo e solo.” only one that is present in essentially fixed form in all 27. The S-theme of op. 13 no. 1/1 is also revisited in these drafts. . . . It maintains basically the same contour ch. 14, in the context of a larger discussion about tonal that it is to have in the final version, undergoing some and modal choices within S-zones in minor-mode sorefinement in rhythmic continuation and in its registral nata forms. position, but remaining fundamentally a fixed element

EXAMPLE 7.6 Beethoven, Symphony No. 3 in E-flat, op. 55 (“Eroica’), 1,

mm. 43-60

°| or a

o£ dd kd pte /|Fimaee . UN Oe | iFLipa eh Ke Dn i(i‘é“RC ee eS ET by es St [Allegro con brio]

CE y | FA ee La pf} __} _ié_

f i ———= —_ Lo-g-0-3 pepPeieer SP orbeSree -} $3. —____—- Ao vr + ot wi “Pd “EE “>FS, the centermm. 97—112, has been conceptually converted section of a continuous exposition), whereupon

to the presentation phase of a larger sentence, a new phrase begins—which we initially aswhose continuation phase (best labeled as post- sume to mark the onset of C—only to fall a cadential, S?) begins with the first PAC in m. few measures later into a perhaps slightly var112. Without the explicit S-retention in the bas- ied repetition of the music that had led to the soons m. 112 would have been considered to earlier, presumed-EEC cadence. Its effect can be C. (In this case, illustrative of Mozart’s ca- be that of a seeming C-module setting forth, pacity for astonishing combinations of deferral then changing its mind and driving to its catechniques, the second PAC in m. 121 1s itself = dence with recovered S- (or FS-) material. The undone by another, ‘TR-based S-appendix, ret- amount of music restated may vary, but it should rospectively understood as S$ [!], and the EEC is be enough to be recognized as an obvious quodelayed to m.129. Here the central factor 1s that tation or near-quotation, not merely another the V:PAC at m. 129 1s nonelided with its subse- cadence (because, from some perspectives, all quent P-based C! theme, m. 130—a character- PACs are formulaically similar).

istic C-theme identifier. This differing mode of This situation may be understood in two EEC deferral—later cadences not elided with ways, and good arguments exist on each side. subsequent material—is discussed in a separate It may be that the composer was suggesting that

section below.) the closing module (C!) was seeking a comple-

A possible example of the retained-accompa- mentary rhyming-cadence with S for rhetorinimental-figure situation may be found in the cal emphasis, in which case one might confirst movement of Beethoven’s “Waldstein” So- sider the EEC not to have been deferred and nata (example 8.2). S begins in III at m. 35—a the post-PAC module’s C-status to remain sebrief parallel period ending with a IIJ:PAC in cure. Although individual cases must be assessed

m. 42. A varied repetition (featuring triplet on their own terms, this seems to us less likely figuration) ensues in m. 43, arriving at a sec- than the alternative, namely that what had set ond PAC at m. 50—clearly a candidate for the out as a C-module relinquishes this status after EEC. What follows, however, is a long excursus a few bars, tucking itself back into S-space (or (potentially an S2, not all of which is shown in TR=FS-space in a continuous exposition) by

EXAMPLE 8.1. Mozart, Symphony No. 38 in D, K. 504 (“Prague”), 1,

mm. 97-133

ga “64 |TAives asad Oe a ee ee=eee a TN =e — pe All

P Nw! - *

100 mn ee DS a i a a a

| yo)ts OE ftfe» ee 5Fp) OgPT) ey Seta | wheeetose es

— @ PI we! aN Pr a ga ff Ee — ee ~ —~

103 f) iA| a| ne aa aa 06° >a f) 4 a | | UT

Fe ses YT eer | ter oH tey 0 ae am emt me | A el a ee ee ee a es a ls ee i Nee’

ee

She 2 ted, half ea™Nw ay| |4,fioe aeeee¢oePog ee See ,-dLio

aAo) mT | eg © gd sss Te SON ee Lt FOa=“ 88

ie Sf | na eae = Sd, eo2 || she, » gt : a notte aT PS a —) * } = © 3 @* f PAV ee, ee eee eee, ee 2 eee eee es, ee eee ee ee eee eee

EXAMPLE 8.1 (continued)

A tts

i i. | oo oe ne ae Pe, ee ret RE aoeeS ?—— —a— ee => or

109 —_—- —_ fey ti 3 —__—_§_*- $87 9? eg 5? ae > oo d_

rr ST A cS 1 5)See LD Aanna, A SG OS nd] aa D4

a | ee P72 Pa Se i A, ——— —_ .| Pu eee — fey tt teneFs— a NE wo 2- ITI Z P ——= al 112

A eS LN ED LP EE em SSeSLO DN EFPsErrer? Sy isa o_os |lies _ ¥_«oe —_*_ o_o

Og i? ee A SS Lis a, ae —~

8 ee 115

je 2? _ as es ee Ld oe OO OT ™=E Fo

on™ f) —_ 4 te: » |. J.

P

eee eee = p7777 s — — =| PY N pat ng fp | | —}—_~+ oS a 121 y 8 ea. -. | .J—“~ At oe ee Pen "Oe o_o Hey ot ea —o oo — |? or — — Oo

eye eMasa a Seeees, se

f— =

— | or? an Oe ae A Eo ee oNoOO, A Ee EF ce ee cs ec oN ae oe "eg Oe f —_— ed eg 154

PT EXAMPLE 8.1 (continued)

eee eee eee ee wee .Jlef «|, #*_ ~=_ Tf, —. | ww» 7» 1

125 c 2, Pt eo — Dees Cy cBTT 5SASS >SOre7 — >eg, ON ne |ec A tl >= eo —__ a ee 9 $$}

/p2,Aarenee a[Aa ee ee Q iP 4 SQ QO hn gp i ee =e 4 ‘ f a, ‘ rs C , ‘ CreSC. yt Hy a ts cor a,

39

pee a4?oe =—_ts aee2em8

Cee rs re es Rs Sn (nC QO a CO oo——_—_ l@e* Ute mm eri i s—“‘iad i OR

— ———————————————————————————————

atdee Ff ae eS

/PY 3 te ne oe ow | ont | Fe eT le 0 a ee nC et LL OT ee a

a” cs es cd ee ne Bhs 0 cnc ne |i sae EFA ce eeeeee ee se ey—«= 2”

Cy us ©. pd A ee oO an OO ys 0 ot on, QO ee

aa

‘i Ec ee a ee end 1

/AWaaee. eee —— oT] ole. we as St i St ee Oi li Oa iBN ee ee nn, Se Sen. ST ba = 2 sy) |el es EE co ee oo” DD nee OO A Pg

ee 2 fe fCe“NS ttg Bo: r . egy a“ TAa CM "Fk OS en f..

te. « ad ——. |ia)eewe Ts ma EE a a | ee rs ss QS ODSe OS (QO es ee Yh, iGOD nlyes $= f)AA—_____| aEP np, “EPO ik 0 Po, Cs es Xs Ge OO = J es, es a +—_—_—_——. 4>WN SY OO oO. t4XH

46

oo OO a” EP A err,QO Dr PA we —(—“(Cti‘“(;(“‘“‘“(“(“(s;sSCd* Ss ll—ep Ue OO a EC eC ae

ey Me ne Rn Ge

48

f

eS —T) Tl Tae |. ae CO nL. eo A eee nel ~, ee ee rr—CO A r——CCS A ge ye a La = | tL)

Toe a ee Ld Oe ee eULDOEe, #S Gt—“—‘“‘ -E 2 L eee eee eee oe oe =e

roth 2." eae See eee es ee neLo ee ee es ee

LF rm | tT to ee TU—=es gamega pT —"“e“—st

aAAeeene es ee es DS ee SOO I SD GS (OE a ee DO ( —————

i ae ee >’ A) See Lee) Oe? eee 2 een eee eee eee ee eee eee eee

|LL_. aPe TD QOes OSD DS ODa; nn lie. CUM Si —CCCCCdSC a ee

48 ee f) A2: eseeee EP Pd,——— i4 so a OO eea |

et ee TF bh | —= ee —m A ee P..ti—CC yk be Te PT em —“CsSCCCC‘C‘i*zCS

a aPee a on a is ff fF OS —S — aga

yctohee— se Pwee esae seemJei le aareaTeea ee i me

52 b ~~ —

i ri| —-» ——___ + ee —_feo— |YS fi fp JB Eeypee| feee| eee ee eee ee eee

me

ee ee S—sS OT TE i@ur | [| fT { @ ¢ ~_ #£#§;3%T,y Bf ae ee ee

aa : be ¢ 4g be

jE : el ee

| ef ise Ci Bom yer pM Po hhh SS

a_

——————————————————————————————————————— ©

60 aN a f i 7 bo! i 4 bal : a a (CO OO OO GO GO GC a a p crese.

Ec a ed pee” I ieee ne nooo nt oot net ne

a SG SS

aA CO ee GO ee CS ee RS oe GREE ee oe”) 6 a GYee OOee NSee GOS ORE 164

a es en

63 = : . De Me NaSOQO 8 a aaee GO 09 EXAMPLE 8.4 (continued)

Fa eee ee eg er ce Be Rs

a . oe aPp____ [sory @& OS QO ££ ££ oTGO eKNN ee eee eee CO eeeCO ee A a | aie a ee es GQ GO££ CO@CO YQ CO

ifey a —gs _ Pp, eSgy OR| ysgoffes pM l—= Ug hlLdlKgh OF Pe @ mm EE 66

eee es My PB O_O er a I carr Re es es ee es es ss OO Ol OR A DOR O_o 9 |

|ee Y Y Y y Y ; OP a 2 a, $s Oe Jy a eA RS (O(NA A

/ » >. |

69 = , oe aa LF a me | ee" Rae SE ee wy ee ee OS Te eT fey’ of ZO A a Se ee ee ee ee eee ee ee eee ..

wT a eC P

Pp mee ty Tee ee Pm

p a P| a P| Tey ge ge p ; a | > aoe ) oe ye oenr T= iPe —=™spe SP A FR a a A wre i re OS ee P. —_ a _ P| a —_ 4 —_ 5 4, —_ P

165

———__

166 Elements of Sonata Theory

and balance not to be captured by a simple rule, The classic instances of sentence-presentabut if one 1s dealing in generally broad propor- tional PACs—to be overridden and deferred tions within an ambitious work, an S consisting until later in the theme— occur in themes that of only a short, perfunctory phrase or “naive,” begin as “looped” phrases, relatively brief ideas problem-free period can give the impression of that cycle back immediately into self-repetition. a letdown or unexpectedly facile articulation of | (This Mozartian subtype of the sentence was a proposed EEC. To be sure, an S-theme may considered more generally in chapter 5. See esbe brief—and it certainly might be a single pecially n. 16 in that chapter, which introduces period—but given the proportions of any in- the EEC complications in such S-themes.) Such dividual piece, there seems to be a conceptual S-themes are uncommon in S-space, but they threshold short of which it cannot satisfactorily can occur there. A particularly instructive ileffect expositional closure. One cannot specify lustration— on several counts—1is provided by this length in absolute numbers of measures. the opening ritornello’s S in the first movement Obviously within an Allegro movement a PAC of Mozart’s Piano Concerto No. 21 in C, K. in the first or second measures— or probably in 467 (example 8.5). This S (still in the tonic C the third and fourth— would be an example of a major, of course, since this occurs in the conpremature PAC that could not serve as the EEC. certo’s initial tutti) begins with two statements But within a small-scale, midcentury work or of a compound (or binary) five-bar loop, mm. briefer sonata what may happily suffice as a sat- 28—32 and 32—36. Now in this particular pasisfactory S might seem out of place, dispropor- sage, unlike most other loop-theme situations, tionately simple or abbreviated within a larger- it may well be that one should not consider each

scale work. Tempo and metrical-notational loop—essentially built from a tonic-domi-

choice are also relevant factors. nant-tonic oscillation—to have produced a full, When a composer does provide a notably brief — bona-fide “cadence” at its end (since the module

S leading to an early PAC, one common strat- does not display a full cadential progression, inegy is to repeat (and perhaps vary) the theme, in cluding a predominant), but in this case a mild part to defer the EEC-moment to a more pro- PAC-“effect” is at least suggested at those ends. portionally acceptable position. This is one of | (Put another way, one can imagine that “effect” several motivations that could be implicated in being interpreted by some listeners—rightly or EEC deferral through thematic repetition. More wrongly —as a PAC.) As 1s typical, the first loop problematic are brief phrases or periods leading elides into its repetition (m. 32), and the secto a quickly secured PAC that are not followed ond loop elides directly into a continuation (m. by such a repetition. Sometimes the early PAC 36). In sum, cadence or not, what we have 1s a concludes the “presentation” portion (aa') of a looped presentation connected directly with (in longer sentence. But this proposition should be this case) a contrasting continuation. Moreover, viewed with caution: the presentation of a sen- the continuation— most unusually — picks up tence of normal thematic length, while occasion- the P-theme idée fixe from the piece’s opening, ally almost cadential, does not end with what we clearly beginning a new idea. In such a situation would usually consider to have risen to the level as this there are manifold temptations to conof a full-fledged cadence. When the presentation sider the EEC to occur at m. 36. And yet it does

is more extended, as in a longer or compound not. It is deferred past m. 36—which in any sentence, a cadence at its end becomes more case might not fully attain the level of a PAC conceivable. Nevertheless, if a PAC or seeming proper— well into the P-based continuation PAC is construed as “obviously” the conclusion and beyond. (We revisit this theme 1n chapter of a sentence-presentation, that PAC should not 20 and place the EEC, after several extraordibe taken as structural in terms of the broader nary deferrals, at m. 64.) theme. The general principle, in other words, 1s Any “early” PAC concluding a simple phrase that no PAC at the end of the presentation of a or period 1s not usually in itself a sufficiently desentence —even though such a cadence occurs terminative reason automatically to regard the only rarely—should ever be taken as the EEC. proposed EEC as inadequate. Once again, many

S-Complications 167 EXAMPLE 8.5 Mozart, Piano Concerto No. 21 in C, K. 467, 1,

28 , p = »P tdae - ee —it2# * A ee eee Pie | eee rrr reer @ ,te S> Eve . ada; EF 3 rer r fr eff mm. 28—39

[Allegro] Y yr ——

Cr

P

Y os y

OS ee OO SS S—— HHI Pi, Om pj —_@ __

34 —_ _ —

ee Me =acct eee—i. =: ee,:Pe 7 Ze : zee :

is yr al if g -_

fg._*eg, rT P\ se \ _N ;eroTOD ff Ft py —_ \/ =n p ¥ > ¥, S-spaces are filled with single periods alone, and mm. 34—41 may be heard as an S?, a welcome, we are not suggesting that a mere period 1s in- balancing complement or symmetrical reply to adequate to the demands of S-space. Context 1s the brief S!. Additionally, the onset of a vigoreverything, but when one comes across a sus- ous, completely different forte module at m. 41, piciously early PAC —and this is not uncom- fully plausible as an onset of C-space (though

mon in Mozart, particularly in the piano con- not in itself determinative of the “correct” certos—one should also be on the lookout for ©= = EEC-placement) bolsters this interpretation. In

any supporting signs of cadential attenuation or such difficult cases we are reminded that any EEC deferral. When they are to be found they analysis— even a simple labeling—is an act of might include sudden drops of dynamic, mode hermeneutics, not an uncovering of an objecchanges, or the immediate recovery of a charac- tively planted “fact” within the music. We are teristic accompaniment figure below an appar- dealing with readings, with interpretations, not ently new theme. This last aspect 1s relevant in with objectively verifiable “truth claims.” As a the first movement of Mozart’s Piano Concerto result, differing analysts might come up with No. 9 in E-flat, K. 271 (example 8.6), where different interpretations of underdetermined

S! within the initial tutti rotation is a brief, musical situations. suileless period concluding with a PAC, mm.

26-33. Is m. 33 the effective ritornello-EEC? oo. . Perhaps not. True. what follows. mm. 34—41 Substitution of an Imperfect Authentic "P ) ye oe Cadence (IAC) for the More Usual PAC provides a new, nonelided thematic sentence, but its accompanimental pattern by and large Although rare, it is possible for an EEC to be retains that of S!, giving the impression that more weakly secured by an IAC. Before one

Ce

26 A EXAMPLE 8.6 Mozart, Piano Concerto No. 9 in E-flat, K. 271, 1,

mm. 26-42

[Allegro]

fp, OTD aCeee eeeeeeeEFoeEe Fo| he a QQ —_2 Ee (DO ee a

P..*.*..*

a ee TF eg OT — 7 a, a fta,i ttayti+P—© i9— — —_ =z _ _z: _ _ Le _—_ —_ =z _ ~~ __

wy gy gy ff p

tS

5b >?_ A A 7??? 2 BS Jp

32 ip oy iy |, — anaN es SG GSgm GO”i PK ky Lg gg eee i or rT Tt tS

oo eo

pI ST 45 JP _

Rey? a |_| ¢ “Ei -| aa Fo | —_____p__}—__f gp, _

Jz.J4 7 * Oe —=— a Se es QO QO:|_| Qs +] 9 ——————— es ee ee 'p

A PR LP ty tMa a ian nn 2) a. nn

FT LES aac ae a

aa6Ae FR a ( Ip eee tee /) — mens — fy ob ‘ a m Ma rg a Pe a, mt gm ma a ee em ee ee ee

. . . ' . . . . * . *. . — a)

: Ly Ces = A DS

S-Complications 169

apA es PP ~ \) St vi i_ o~ vi PI . Fs Trew aA sy Fe ee a ce a tm ne CC 2 °—— . eseeCCerrs aesll 7—— , P |)SErr EXAMPLE 8.6 _— (continued)

TAY’ UCM hep op op | Gp es my el

A| aPL ee |(

_pees #£f0-— =ge ——* ——— |£aa yeet, ye A «eee ee es *GO ONaE OD CO CO ee weiy ie comes to this decision, the rhetorical signals The Evaded PAC 8

surrounding this EEC-moment— particularl . .

di * tat £ C—sh - b y A drive to an anticipated cadence may be unregarding status of C—shou e over-at 4 . the last moment. The . . . the ermined or evaded

whelming (sufficiently overwhelming to over- , 3 , sy

power the EEC-concept, among the strongest term “evaded cadence” appears in Caplin’s re-

c tions). Thi to be th cent Classical Formto to be referthe usually unexof conventions). This appears caseto .the 1 Beeth 6S h ae dinD 36 pected motion ofa cadential dominant chord to in Beethoven’s Symphony No. 2 in D, op. 36, . . . Gest y ve hich 3 Pe a m. I® (instead of the normatively irst :Movement, ,in which 3 appearscadential ina —I) .and . nonmelodic voice in the flute at the EEC. These the beginning of a new cadential progression,

....rn,

—— sometimes built around the aptly named “one the effect is that of a PAC in the literal or 1m- md . ome cases require interpretive subtlety. Frequently more time technique.” in hich viche composer

plied structural voices with a mere cover tone in repeats [immediately] previously heard ideas and

. .decorative leads them again tovoices a potential cadence. ... The one of the upper —as probably .. occurs at the EEC-moment in the first move- composer backs the music up, so to speak, in

- of of Beeth 5 Quartet in E-flat 74 order the listener hear. the ment Beethoven’s Quartet inforE-flat, op. to74, . aeimpending ca-

“Harp” thatbedthat nit th dentialthe arrival more time.”” From the standarp,’ m. m. 70. 70. Thus Thus itit be may despite . ofone oint Sonata Theory, of course, such an evasurface appearance of an IAC the clear sense of P th hadrovt h 16 chord would neve

. . sion through a dropthe to day. the I® chordone w never a structural-voice PAC carries AlternaP.

ene e roug e -effect in or- ,.

tively. the composer may have purposely weak. be considered even a proposal for an EEC, since

ah CEC th th TRC a 4 . the requisite PAC is still nowhere in view.

We also use the term “evaded cadence” for an-

der to cast more of a burden on the subsequent oo.

C-space. This pseudo-LAC situation should not other situation, one that more closely approaches

e confused with cases in which the -even .

b fy dwith a which the EEC : that of the perfect authentic cadence. This set of

lid thaCthwith cerine |atheme . circumstances occurs one structural voice elides entering inwhen another voice. . When that elided C th b 5 2 or & th drops out at the tonic-moment of the otherwise

en that elide eme begins on 3 or 5, the . . . . . . . absence on the downbeat measure highest-sounding voice. Obviously, of thethe implied . in eimoment of the EEC/PAC will not have 1 as the normative PAC, creating a momentary blank or

PAC is not undermined by these circumstances ther the treble or the bass. Typically, the abanOne must not rely on only a mechanical vertical doned voice will rebegin immediately with new reckoning above the bass to determine whether and lively figuration, perhaps relaunching on an

or not we have a ; . 14:

eweh PAC off-beat. Such evaded cadences are especially typical of Mozart. As a general guideline (while

2. William E. Caplin, Classical Form, pp. 101—7 (quo- Schmalfeldt. Caplin also used the term “evaded catation from p. 103). As Caplin noted, the colloquial dence” to refer to other possibilities that we house unterm “one more time technique” was coined by Janet der the concept of the attenuated cadence (p. 103).

170 Elements of Sonata Theory

admitting the possibility of an occasional excep- even though it proceeds onward to differing tion), we do not consider these evaded cadences material (our criterion for the determining of to serve as the EEC. Normally, the S-zone is the EEC)? In some cases it might, although the renewed into another phrase that will continue claim is more persuasive when buttressed with the drive to the EEC. Two examples from first additional evidence. In the first movement of movements by Mozart: the Piano Sonata in B- Haydn’s Symphony No. 46 in B, S! in F-sharp flat, K. 281 (evaded PACs within S at mm. 30, major (mm. 22—36), drifting comfortably to34; EEC delayed until the next PAC at m. 38); ward its cadence 1n a mild piano, shifts suddenly and the Piano Sonata in A Minor, K. 310 (the to F-sharp minor and erupts into a jolting forte upper voice drops out and resumes in a higher at the first PAC, m. 36 (as if confronted with a register, m. 35; the EEC is evaded again, with body blow, a sudden or desperate “No!”). In bass dropping out, at m. 40, postponing the EEC this case Haydn clarifies that this was an inef-

until the next PAC at m. 45). fective PAC a few measures later, when the S? idea recovers the original S-cadential module (m. 52 = m. 31) and leads to a less problematic

The Attenuated PAC PAC/EEC in due course.

A PAC can have a weak or attenuated effect Additional possibilities of PAC-attenua—something that instantly problematizes the tion occur when one encounters a radical regstrength of its potential EEC-status—1in more ister-shift in one or more voices at the tonicthan one way. This may happen, for example, moment, when the linear fifth-descent within when a forte cadential module (concluding S) S is not fully present prior to the PAC, or when drives toward the expected cadence but sounds the linear descent occurs in an inappropriate it with a sudden drop to piano at the precise register (too high or too low). These are intonic-moment of the PAC, as if signaling a variably difficult matters to evaluate, and they last-minute hesitation or failure of nerve. Ab- often invite a consideration of the techniques sent other motivic or thematic evidence, we and principles of Schenkerian analysis. When have tended to consider such attenuated PACs potentially structural PACs are problematized as EECs, albeit ones requiring reinforcement in or attenuated in one way or another, opinions the C-zone. Each case differs according to the may differ on the resulting implications for severity of the attenuation and the nature of the EEC-placement. Within the argument appeals surrounding material. At times the issue cannot will have to be made to additional material surbe decided with a strong degree of confidence. rounding that PAC. Of particular relevance are But producing this ambiguity must have been matters of motivic or thematic retention or the the composer’s point at that moment. We should generic thematic signals provided in the immesuppose that it plays into the dramatized musical diately following (presumed) C-space. Do clear

narrative being laid out in the exposition. C-identifiers, such as the P-based C, seem to Another possibility of PAC-attenuation oc- announce the onset of closing-space? The mucurs when at the PAC the mode unexpectedly sical context of the PAC considered along with switches from major to minor—obviously a the severity of the attenuation can help to decide dramatic signal that something has gone wrong the situation at hand. In some cases it may be at this crucial moment. What ought to have that any simple decision one way or the other is been a point of self-assurance and attainment is inappropriate. Rather, the point of an adequate plunged into anxiety and signs of the negative analysis might be to explicate the ambiguity. (the “lights-out” effect). To be sure, the expositional “attainment”’ still exists in a mechanical,

literal sense—a PAC has been produced in the Apparent Double Medial Caesuras: new key—but it is simultaneously undermined The Trimodular Block (TMB) by doubt, undercutting the generic expectation

of major-mode success. Does this mean that the Although the more normative two-part exPAC in question is no longer a satisfactory PAC, position is marked by a single medial caesura

S-Complications 171

somewhere in the center, it is not uncommon decades around 1800 one finds it occasionally to encounter the setup and execution of a sec- in Haydn but perhaps with greater frequency ond, additional medial caesura before the EEC. in Mozart and Beethoven. It is to be found in This can occur in a variety of contexts, but the all types of compositions— sonatas, quartets, invariable impression is that of apparent double symphonies, and concertos. It took on an esmedial caesuras, and, concomitantly, the effect pecially vital role in many of the solo exposiof two separate launches of new themes (pre- tions of Mozart’s piano concertos (that is, in the EEC themes) following those MCs. Depending Type 5 sonata), typically expanding and varyon the circumstances at hand, the second new ing the layout of the preceding tutti rotation, theme can seem to be something of a second S. which had usually been supplied with only one The first new theme, following the first MC, MC. (The details are laid out in chapter 21.) It will prove “unable” to move to the EEC and is also the foremost expositional strategy that led will instead be converted into the preparation to some of Schubert’s much-noted three-stage for anew MC, possibly including the establish- (sometimes three-key) expositions (or “double ment of a dominant-lock and other features of | second groups’’).

MC-preparation.° The issue of double MCs (and the resultIn these situations we find at least three ele- ing TMB) 1s a complicated, sometimes elusive ments: the first new theme after the first caesura; topic, and we have also dealt with it elsewhere.4 its dissolution and the setting up of the second Because on closer consideration the double-MC

caesura; and the onset of a differing S-theme, pattern can occur with differing S and/or TR starting its own, renewed journey toward the implications, it can be desirable in some analytiEEC. We also refer to this characteristic three- cal situations either to replace the perhaps-exphase pattern, with apparent medial caesuras pected S!!, S!-2, S!-3 numbers with TM!, TM2, before the first and third elements, as a trimodular and ‘T'M® or to use both in conjunction. The block (TMB). The presence of a TMB, a strategy exponential numbers of the TM-modules do for enriching and extending mid-expositional not refer to PACs accomplished. For the sake space, complicates the matter of determining the of simplicity (in an already sufficiently entanextent of the associated, potential S-zone. The sled topic) this exception to our general rule two MCs in question are usually different, and seems advisable. When TM? does not proceed they almost always follow the deployment se- efficiently to the EEC, we might find a need to quence of MC options. By far the most common subdivide it, however, into TM?-!, TM?-?, and pattern is: I:HC / V:HC. Also possible, though so on. In nearly all cases, TM? is self-evidently much less frequent are: I:HC / V:PAC; V:HC a different theme from TM!. Only on very rare

/ V:PAC; and the repetitive option, V:HC / occasions are the themes based on the same V:HC. Alterations of these patterns are also idea.°

available as deformations. One may distinguish among differing types Not only is the double-MC-effect a fairly of double-MC-eftects. The simplest TMB type common phenomenon, but it is also of consider- occurs entirely within an unequivocal S-space, able historical and structural importance. In the so that I'M! is unproblematically equivalent to

3. The phenomenon of seemingly “double second 4. Hepokoski and Darcy, “The Medial Caesura,” pp. themes” in Stamitz was mentioned by Eugene K. Wolf, 145—50. The Symphonies of Johann Stamitz: A Study in the For- 5. Foran example in which TM? returns to the leading mation of the Classic Style (Utrecht: Bohn, Scheltema & idea of TM!, see the first movement of Haydn’s Piano

Holkema, 1981), e.g., pp. 151, 199, and 272. Cf. pp. Sonata in C, Hob: XVI/50, mm. 20 and 30. In this 327-28 (“false [transition] sections”). Wolf described case, both TM! and TM? are also P-based ideas, and the the post-S! TR-texture as a “secondary transition.” second MC is deformational, with the full clarity of the On p. 200 he mentioned that “this design [including a MC-effect not completely attained. new, forte transition that leads to a second S theme] also appears with some frequency in Viennese symphonies (e.g., by Wagenseil and Dittersdorf).”

172 Elements of Sonata Theory

S'. This type of TMB might be regarded as will eventually manage to attain the PAC/EEC. a variant of the multimodular or trimodular =Any number of C-modules may then follow. S, one in which an additional MC-effect and Considered as a whole, the TMB situation “second” S have been planted somewhere in conveys the impression of a flawed or unsatisfacthe middle. (As a rule of thumb: If the dou- tory first S-idea, TM! (“No! This theme won’t ble-MC-effect 1s not present, we are not dealing do! This isn’t the one we had hoped for!”’), fol-

with a TMB.) This type of TMB begins with lowed by some sort of TR-texture-based coran initial caesura (usually a I:HC) that could rective action, TM?, anda “better” S idea, TM3. serve as an MC, followed by a TM! that ap- A good example of this type of TMB occurs in pears with acceptable S-rhetoric, characteristic, the first movement of Beethoven’s Piano Sonata lyrical, or cantabile, in the expected new key. In in C, op. 2 no. 3 (example 8.7). Here we find this situation TM! accepts the proposed MC and a I: HC medial caesura, with GP gap, in m. 26. launches S-space, although it might also strike What follows is an enormous TMB (with exus as weak or flawed in some way (minor-mode? panded third module) that stretches from m. 27 thematically problematic? too eager to accept a to the EEC in m. 77. TM!, or the first S-idea, premature or insufficient MC? drifting back to begins at m. 27 in an expressively “flawed” G

the original tonic?). minor, the dominant key having unexpectedly In any event, TM! proves in some way unsat- collapsed into minor (“lights out’’) at this point. isfactory, unable to secure the EEC. As a result This “flaw,” it seems, will have to be expunged

that theme is jettisoned, normally by dissolv- through the TMB strategy. Beginning in the ing into what may be regarded as TM?, the set- dominant minor, the troubled TM! either canting-up of another caesura, sometimes preceded not or chooses not to sustain its G minor, the by a move back to characteristically transitional mark of its imperfection. It begins to modulate (TR) texture. Sometimes one cannot distin- sequentially, rising by fifths to a restatement on suish any extended TM? module by texture and D minor (m. 33) to new material on A minor content alone. In other words, we might have a (m. 39). M. 39, starting the TTM?-phase, rein-

TM!>TM?2 merger, with the TM? aspect vigorates a more characteristic —[R-texture and marked only by the articulation of the new MC leads to a clear postmedial caesura, V:HC, at m. at its end. Since we are considering what pre- 45, followed by two bars of caesura fill. TM? receded TM! to have been the “real” MC open- sponds to this second MC in m. 47 —a new, caning up S-space, this second caesura, normally tabile theme, now in the radiantly sunlit G maa V:HC, may be considered one type of post- —sjor. After reinvoking TR-based material along medial caesura (PMC). Nonetheless, the larger the way (m. 61) the EEC is strenuously attained effect produced up to this point 1s that of appar- only with the V:PAC at m. 77 (not shown in ent double medial caesuras. ‘The second caesura the example). (the PMC) occurs in the middle of the already The above represents only the simplest type

launched S-zone. of TMB, the case in which the S!-effect of the The second MC 1s sometimes articulated first module is clear. But other factors can enmore weakly than the first, giving the impression ter into this apparent double-MC situation. ‘The that the “strong” MC energy had already been most obvious complication ensues when TM! is spent in the preparation for TM!. Whatever its not a satisfactory S-candidate. TM! may strike rhetorical strength — or lack thereof—its func- us as unacceptable for any number of reasons, tion is to restart S or to prepare for a more “‘suc- and we are then obliged to conclude that Scessful” S-theme. In this type of TMB that new space has not been genuinely opened by the first Sis TM>3, which emerges with characteristic S- apparent MC (which must thus be regarded as

rhetoric and represents a “second chance” for a “false MC”). Obviously, if the first MC is a S. Although it may encounter adventures along legitimately permissible medial caesura (usually the way —and although ITM? might prove mul- I:HC) and if it does not open S-space, it must timodular in its unfolding—this stretch of S- in some sense have been a declined offer. In space will produce no further MC effects and this respect the more complex instances of the

a

. tr o — =: SN a caeea —S= a a SC

a EXAMPLE 8.7 Beethoven: Piano Sonata in C, op. 2 no. 3, 1, mm. 23—56

Pe ESSE EES

23 [Allegro con brio]

Sx A 7eS Aee SH

A bs A A nA

Piano [Psu Cy. ££ +, @» 5» ct(a wo i OY =)© gpptesrusoeeraee A4 @A+, ct| °°» A 8A

24 ‘eo A - ——— SP —— ”- e 8 aoS OO OS”——— SYA [—— 6 —_

=||N 7 ‘ : "2 ,,°= i: |CS OY yo =" eeOS nc CC PY fe O_O ee Eo SS 8eo—Ee (C—O ow TP Pt

| Cy: # fs » °&xp £@*,8 8,0.“ .}.}.}.}#§»~+»S#*@2 TTT

iOu, eea ee |A ON

iASY (GS SOOSCC SS (SS a CS LS f) ES all Ds—s—o——_ Fr8 >Ee—=* ail Fe Fae ea =e eee "= |A pie SO fo fo | SC \—— (RC oe ee oeA fe a a CS (GS

oe eee?

29

a1Seea

iOS a oP rr a a a A aA6A OOAA

h aSS E SS t 3 CS ? ~,SC . . LL Fg

|os:Se nc A a A a CS Se 7 te | @ | ws #@wtse, © © | @ . © |, 2 | sf &@£~ 5» # naa rr |... 4 a - a ee ee ee es

f)A 4=D:GG bAae aA A :| Aff|CS SG,

ay Fa Oe| OL a 6eeKT A A|

| Se

A A SS A lS es Ss i — > Se

EEE PT re P| i a

é al | aA A aaCC I 4? i i é " aa b SF See we

38 , PP A eo ee _ FC SO A A 08oFaoD ee es“Sn ee | ee ~~ NS gg tteeieoetcr |oea SEeePa |OEeeeeTwtT-e_|_— | A

| es gy 7 Pt __ _g —g —___ gp —__ p> —__ gp —___|___ yg —_—_

!cS 8 ASC A 44

Fy a —eae—em a t —“e

JEPooa S".848-“00"-"”"”0N0nNW.-el ee ee ee es ee errs ee ee ee eeeeeeEE ee Oe ee

ee -RE —-a-A -eS-cl- -*-S de———"— ) - ea Se ay bed-wwe CO) se¢¢ee eeeeeee

pA 9p 1 "9 BF ANT » «dD j=eee=F To Teee" hae as >/) 20) eee eee ee ee eee ,—-f ——

P crese. f= NN —~ oa a as oe a a en aavA

96 p 1. J-> |

ee” . . . . ° / => ~ PP oye ol ot 2 ee __ ee es ee a oe ee a FS | a SS SD AS OO QO CO oS 188

A EXAMPLE 9.1 — (continued)

102

f) Ar * ee” ‘eee al= -ee*ee.) . aeee* eee. * *.ee *oe*ee . .eeeee *2 *eeee*22:ee” . eee . 2... CJOo * a*=eee *Oe DD Oo ee.Oe Oe Pf .4G ees eee ee Oe 4 2.ee

rey TFahe e™ a” ee a |Oe a |rree eeeedE ee eeeeee ee ee ee a OL ee 53335 149435 .OEa.A1aa2.seShaaLa ee er aAFF lS OO ee a a aSO, a Cs CFESC.

er 4 nail P y , P ha nalll

106 seit t 4 Dd ne ee| ee a aee aeeees I st22, J UT ggtits — aa 3arssfsOb ee a a BO a

mf ee fi lf | @ BR es f wor Ty |]lhlUmrCLUmPr Ree Ue Ome Um es ee

_ ws | 2. - ££ ff. | @ fF. £2. £ fF. | ae Ff 7 Tt TT ULULUDLULULULULERLUULULULSUUULULULULULD™*EUU

a ET FPR ec Gr OY ell a = ee ee ~ —~ _FPrPpi

110 , . . . $ i 4 : iPPB i aeea TT SSee“ 6”see 0

SP A FP PSee oreny QO. ee nee=e ee > ae a eePnesedee

Fy eT

PR a a a TT —f--—2, ___¢¢#§_________@____@ _—_________@ | ___@ @ __@# _

190 Elements of Sonata Theory

shows the muscular, plunging-octaves approach is to regard it as C°, an unusual, perhaps unique to the EEC (as if conquering it by driving it into deformation.

the ground) and the immediately succeeding This interpretation also plays into larger hermusic. Following this, the first thing to notice meneutic issues. If this sonata form (as suggested is the instant withdrawal into a hushed, piano by the “Eroica” subtitle) 1s metaphorically repdynamic and the utter change of texture (the resentational of the hero’s battlefield—that onto very things, of course, that had suggested its which the narrative subject is drawn into comcontrasting S-ness to commentators). But this bat—here that hero drops away from normamodule does not proceed to a perfect authentic tive sonata-action, rests after the conquest (the cadence. On the contrary, by m. 99 it locks onto EEC), withdraws from battle into the shad-

a dominant and begins a crescendo-build to a ows, and prepares for the next onslaught (Cnew, forte module at m. 109, an idea that draws space), which within a few bars is greeted with on the procedures of the standard, forte P-based renewed vigor. In a sense mm. 84ff represent C (or at least a C that alludes to the heroic triads a withdrawal after victory, just as their radical

that had characterized P). transformation into the much-discussed, miAt this point one needs to reason through the nor-mode new theme in the development (m. evidence, reconsidering all the options. So rec- 284) suggests a withdrawal in defeat.® ognizably C-like and conclusive in character, m.

109 could by no stretch of the imagination be

considered as the beginning of a “real” S. (Nor S© Themes: Apparent C-Zones in the

does it fit the normative conception of the con- Absence of an EEC cluding, cadential module of a multimodular S.

Clearly something different is beginning here, Particularly in sonatas after 1800 S may break q Cl.) If that is the case, then the piano passage down without producing a PAC. This inability that begins in m. 84 could not be S either, since is sometimes followed by a decisive, contrasting, it leads to no V:PAC before C!: S ideas do not potentially “C-lke” theme. In such instances merely prepare for and serve as grand anacruses the question inevitably arises as to whether the for C ideas. None of this need be a problem un- nineteenth-century C, as a by-now reified, sepless we insist on trying to turn m. 84 into the arable thematic concept, was capable of forging exposition’s secondary theme (or perhaps, alter- ahead on its own in the absence of an EEC. On natively, into a PAC-triggered TM3>). But there the one hand, this contradicts the definition of is no reason to do this, since the piece’s musical C as postcadential (post-EEC): at least within processes up to this point have already provided the eighteenth-century norm nothing should for an MC, an S-space, and a forceful EEC. If — exist conceptually as C until the EEC has been m. 109 is best heard as a “heroic” realization of a secured. On the other hand, one can imagine typical C! launch, then what are we to make of situations, especially after 1800, in which a its long-upbeat preparation, mm. 84—108? Un- composer might have intended to portray just der these circumstances our preferred decision | such an S-breakdown. While S fails in its mis-

8. Much in the first movement of the Eroica has been those in C°, mm. 86—87, 90—91; compare the contour

misunderstood. In part the development of the first of the 5—1 oboe descent, mm. 287-88, with that in movement cannot be properly construed without a clear the first violin, mm. 90-91, immediately restated in sense of its expansive dialogue with rotational prin- the winds; and so on. Part of the case for connecting ciples. Within this context m. 284 is most profitably these two ideas (although the former is referred to as heard as a minor-mode, anti-C° referent (this is prefer- the exposition’s second theme) has also been made by able to hearing it most essentially as a reworking of P), Robert P. Morgan, “Coda as Culmination: The First even as the immediately preceding E-minor cadence re- Movement of the ‘Eroica’ Symphony,” in Music Theory ferred back negatively, and with enormous strain, to the and the Exploration of the Past, ed. Christopher Hatch and

EEC-principle following a vast battle-excursus fugato David W. Bernstein (Chicago: University of Chicago on S°. Compare the bass pizzicatos in mm. 285ff with Press, 1993, pp. 357-76 [esp. p. 369]).

The Closing Zone (C) 191

sion, C is left waiting for its “scheduled” turn devised the label S*, which is intended to suggest to appear, and in fact, following the demands the presence of a theme literally in precadential, of unstoppable clock-time, it does so at the ex- S-space that in other respects sounds as though it pected moment regardless of S’s lapse. The cu- is more characteristically a closing theme. Thus rious thing about such themes is that they seem S* means “‘an S-theme, literally pre-EEC, in the to bestride both the S- and C-concepts. They style of a preplanned C-theme.” Its equivalent are emphatically precadential, pre-EEC (the es- in a continuous exposition (which contains no sence of S-space), and yet, in part because of the S) is Cpre-EEC (chapter 4). In any event, making a block-like layout of the exposition, one suspects clean, reductive decision about labels and term1ithat they are simultaneously implying the onset nology 1s less important than explicating the cri-

of what “should” be a C-idea. sis or ambiguity created by the breakdown of S. An early, perhaps defining instance occurs in the first movement of Beethoven’s Piano Sonata

in F Minor, op. 57, “Appassionata” (discussed in The Retransition chapter 14). Later examples may be found 1n the

first movements of Brahms’s Symphonies Nos. A retransition (RT) is a passage that prepares 1 and 3 and in several movements by Bruck- for and generally leads to the return of the priner and Mahler. Additionally, in several of these mary theme (P) in the tonic key. It does this by cases a major-mode S breaks down and is over- destabilizing the key in which it begins, then

taken by a parallel-minor-mode drive to ca- driving toward an active and frequently prodence. Here one might cautiously entertain the | longed dominant. The most natural home of option of defying the definition of C as post- the retransition may be at the end of the develcadential and considering the breakdown of S_ opmental space, where it sets up the onset of the to be an EEC-deformation. Such a deformation recapitulation (in a Type 3, 4, or 5 sonata). But would portray a manifest collapse of S before at- it can also occur at the end of the exposition, taining the EEC. Under this interpretive option — where its local function depends on the type of

(relying on the principle that even strong ge- sonata at hand. In a Type 1 sonata (with neineric norms can be overridden for extreme ef- ther a development nor an expositional repeat) fects) the S-breakdown would be followed by a it can prepare for the immediate recapitulation. precadential (pre-EEC!), rhetorical C that now In a Type 2 (“binary”) or Type 3 (“textbook”) has to take on the EEC-burden of S. The her- sonata it can look forward to the repeat of the meneutic implications of this situation are obvi- exposition. In a Type 4 sonata (sonata-rondo), ous. This interpretive option requires stressing it leads to the second statement of P. (In a Type the hazy notion of “C rhetoric” in what techni- 4 sonata it can also occur at the end of the reca-

cally remains S-space. pitulatory space, where it usually readies us for In order to describe such a situation we have the final return of P.) EXAMPLE 9.2a Mozart, String Quartet in C, K. 465 (“Dissonance”), 1,

23 Allegro a a ee ee ee oe ee i oe oe a See mm. 23-26

25 — i ’ Ferree peppy te PLETPERP? ee EPP PEEP

Se

sd 2 Sa a a a A EXAMPLE 9.2b Mozart, String Quartet in C, K. 465 (“Dissonance”), 1, mm. 88—106

AefiFF 6) QO ee eee i ae eg Te ff at [Allegro]

eee ae |FFj=tretooPp — fet— ee 9 fo

AO) A — 0 0 Fs (| yO Oe OF Oy OF OO

I — OO —

FAYE} — oo Ol _ te ear

treeea.oN aama: eee eee Fy — p

93 _—aN /) ee ee A. oe ere be @ ah Fo yp ote 6h Ug OTT sd GS Ee

PP es f = NN Sa DIN aoe,

ee“ 2a aN Se er ———_

See eee eee eee eee eee, ep 6See a———~a J) a———a | o_O OOO =e , 7:JJ]i

Hey Fg 0 ss — | — | — 8

wy VS N/ 2-2 Joo So eye es—s—eE eae ="

em ge ge Nu 2 2 @ PP ae

99 oN o~ OY mo" OY

ft| \—__£__1 sss = > 40AM] __e __ -——_ EXAMPLE 9.2b (continued)

Nee A P oo Wis, 2-5 | ——

apn femme ptt f a ———

jo ° ©a«ge! i) owe NaPiePs aaa Yd| 105

Ppmam — Crese. leer ys sly ll ee eC ;Cisdiz

When a retransition occurs at the end of the fect supported by triadic motion in the bass in exposition, it may follow the final closing mod- mm. 95-99, At m. 99 we cross a divide. The ule as an easily separable idea, often P-based, cadence-reinforcement stops—there is a sense or it may begin as a closing module (or its rep- that we have reached the cadential end of a Cetition) and dissolve into retransitional activity module—and even while the P-based material (C=>RT). In general, C extends through its last continues, the fadeout now shifts to its reverse, literal authentic cadence or cadence-reiteration a texture of dynamic and harmonic accumula(V—I motion, even if sounded over a pedal). tion. M. 99 thus provides the sense of beginning When a new module veers away from authen- anew module, a new thought. The melodic idea tic-cadential implication and toward a new is shifted to the bass and begins to rise sequendominant setup for the return of the tonic, the tially. M. 100 makes the crescendo explicit, and beginning of that module is considered the start in m. 102 the introduction of F4 (47 in the key of RT, regardless of the material on which it 1s of the dominant, 4 in the tonic) converts the based. As with TR, one should not consider RT G-major tonic into an active dominant and preas beginning in the middle of a phrase or self- pares for the repeat of the exposition. Two bars

standing concluding module. of piano fill, mm. 105—6, are appended to proThe end of the exposition of the first move- vide a graceful link back to the opening of the ment of Mozart’s String Quartet in C, K. 465, exposition. RT 1s best assessed as beginning in shown in example 9.2b, illustrates the basic idea. m. 99. Had the downbeat of m. 99 not given us Here a strong C-module deep into C-space ca- the sense of a cadentially reinforcing close to the dences V:PAC at m. 91, which is also marked earlier module, we would have considered the with a sharp drop to piano. At this point Mozart RT to have begun in m. 91. The example also introduced P-based material—the classic piano demonstrates the occasional relatedness of the P-based C—in a cadentially reinforcing mod- _ piano P-based-C concept to that of one type of ule that completes its first rhetorical span at m. RT. 99. Mm. 91-99 thus extend C-space: we find In a Type 4 sonata it is not uncommon for a tonic-dominant oscillations, first over a Trom- retransition to usurp the place of a closing zone. melbass G-tonic pedal (mm. 91-94), then fol- In the finale of Mozart’s Sonata in B-flat, K. lowed by a short-winded cadential-fadeout ef- 333, the EEC (V:PAC) occurs at m. 36 and 1s

194 Elements of Sonata Theory

elided with a retransition that reactivates the tonic and requires more energy to pull away dominant before the listener has much chance from it and lead to the active dominant. In this to construe it as a closing section. At the end of — case that dominant takes the form of a cadenthe recapitulatory rotation this same passage oc- tial ? chord that heralds an extended cadenza, a curs elided with the ESC (m. 163) and 1s greatly passage that stops sonata-time until the longextended. This expansion may be motivated by awaited return of P proper at m. 199. the fact that the retransition now begins in the

CHAPTER TEN © "ISLE 0

The Development (Developmental Space)

ihe familiar term “development” can be plot-“unfoldings,” to the close discussion of an misleading, since for today’s English speakidea, or to the setting-in of [intrigue or le noeud ers it can imply an omnipresent working-out of — (knot), leading eventually to its dénouement (un-

expositional material within that space of the tying) .? sonata. Although this is the most characteristic For German speakers this zone 1s the Durchprocedure, many development sections present _ftihrung, etymologically “a leading-through.” “new” material in individual sections, and a few This was originally a term used to describe fill that zone almost entirely with contrasting fugal or polyphonic processes or other intense material. (Considering it primarily in harmonic motivic or thematic treatment (motivische oder terms, Ratner suggested the synonym “X sec- thematische Arbeit), and in the nineteenth cention” for this musical space.)! ‘The word dével- tury 1t came to describe “developmental” acoppement was introduced into the discourse by tivity within differing portions of sonatas and Anton Reicha in 1814 in the Traité de mélodie scherzos as well. Along with such similar terms and, again, in 1826 in the Traité de haute composi- as Ausfiihrung, Durcharbeitung, and the like, it tion musicale, which referred to the premiere sec- seemed an apt description of what usually haption de la seconde partie .. . développement principal pened at the “second part” or “middle part” of en modulant sans cesse. Reicha’s meaning was not a sonata—that is, it could feature expositionwhat we might initially suppose it to have been. based “Durchfiihrung periods,” as Ernst Friedrich He considered everything after the exposition des Richter put it in 1852. Increasingly, particularly idées, including what we call the recapitulation, from the 1860s onward, it referred to the whole

to have been their “development.” Reicha in- section (as in the writings of Arrey von Domtermixed concepts taken from classical rhetoric mer, 1862, “Mittelsatz oder Durchfiihrung’). But and French dramatic theory in such a way that Durchftihrung came to be widely accepted as the développement was analogous to the process of — standard way to refer to the second sonata-part 1. Ratner, Classic Music, p. 225, “the develop- 2. Peter A. Hoyt, “The Concept of développement in the ment, or X section.” Cf. pp. 209, 213 (the X section Early Nineteenth Century,” in Jan Bent, ed., Music Theat the opening of the B-portion of small two-reprise ory in the Age of Romanticism (Cambridge: Cambridge forms—AABB— “whose function is to open the way University Press, 1996), pp. 141-62. to the final confirmation of the tonic 1n reprise II’). 195

196 Elements of Sonata Theory

only in the work of Hugo Leichtentritt (1911): the piece as a whole, should be taken as a com-

Exposition— Durchftihrung— Reprise.° positional claim of increased thought-content Neither of the ingrained terms, development and prestige. and Durchfiihrung, are likely to disappear from Longer developments usually displayed a vacurrent analytical discourse, nor do we suggest riety of differing, now-familiar developmental that they should. Still, when underscoring the strategies: frequent modulation; complete or idea that not all portions of developments are fragmented references to motivic or thematic necessarily “developmental” of expositional material from the exposition, typically shifted ideas (they may feature episodic passages or through different harmonic and major-minor other events), we sometimes refer to the develop- colors; occasionally interpolated episodes or mental space, signifying that portion of the sonata “new themes”; blocks of sequences; Sturm-undwhere a “traditional” [textbook] development Drang textures; large-scale intensification-drives;

would be placed, were that texture to be pres- surprises and interruptions; fugato or other conent at all. The term developmental space does trapuntal treatment; the “false-recapitulation” not imply the inescapable presence of thematic effect; and several others. Some of these will be back-reference and fragmentation, motivic ma- taken up as individual topics later in this chap-

nipulation, sequencing, and other commonly ter. accepted developmental techniques. Developmental spaces may be brief or much

expanded. Within ‘Type 3 sonatas a typical mid- Tonal Layout century development (the “first part of the sec-

ond section” of a large binary form) was nor- The Development as a Whole mally a modest affair, perhaps under half the length of the exposition. As sonatas, quartets, In more ambitiously realized works, especially symphonies, and concertos grew in their ambi- from the later eighteenth century onward, tions in the later eighteenth century, the size the development usually moves through variof their developments began to expand. The ous tonal areas, often by means of sequences or amount of space allotted to a development may other leveraged shifts. In all cases tonal motion be taken as an indication of the intended seri- is creatively interrelated with the selection and ousness of purpose, depth of thought, or con- ordering of thematic material, which is anynoisseur appeal of these works. In the last de- thing but haphazard. It is not uncommon for the cades of the eighteenth century developments original tonic to be visited early on, as nearly all commonly extended anywhere from about 25 of the theorists around 1800 not only remarked percent to 75 percent of the length of the pre- but specified, but in general the development ceding expositional rotation. In mature Mozart is characterized by a restless, modulatory plan and Haydn we can find more extended devel- that stakes out one or more nontonic local goals. opments, occasionally matching the breadth Although the plan is never random, by the later of the exposition itself. And as is well known, eighteenth century it was usually tailored to Beethoven was capable of writing even longer, the individual piece. Some keys are merely almore monumental developments, sometimes luded to, passed through fluidly; others are se(as in the first movement of the “Eroica” sym- cured with a cadence and thereby articulated as phony) exceeding the amount of space allotted momentarily “fixed in place,” more structurally to the exposition. This hyperexpansion of the highlighted.

development, along with the resulting size of Analyzing a development’s harmonic plan 3. See, e.g., Siegfried Schmalzriedt, “Durchfiihren, der musikalischen Formen und ihre Analyse (Leipzig: Georg

Durchfiihrung” [1979], in Hans Heinrich Eggebrecht, Wigand, 1852), pp. 43-44; Arrey von Dommer, Eled., Handwéorterbuch der musikalischen Terminologie (Stutt- emente der Musik (1862), p. 289; Hugo Leichtentritt, gart: Steiner, c. 1972ff), pp. 1—16 (here, pp. 9-10). The Musikalische Formenlehre (1911), p. 128. books cited are: Ernst Friedrich Richter, Die Grundziige

The Development (Developmental Space) 197

includes such things as tracking the bass and setting forth a succession of harmonic and rheupper-voice motion, observing the logic be- torical adventures. hind the sequential levels visited, noting the Following a motion or series of motions to tonicizations and modulations and the degrees different tonal planes, the development’s last of strength with which they are suggested, task is to prepare for the dramatized return of registering the expressive implications of ma- the tonic (retransition), usually by deploying an jor- and minor-mode transformations of earlier active dominant (locking onto the structural themes, and so on. One should be especially dominant, almost always V, of the tonic-toattentive to clearly articulated PACs, IACs, or come) and proceeding forward with it, often HCs, along with any developmental MC- or gaining energy in the process. In Schenkerian other caesura-effects, especially if preceded by terms the V, ending the development is then dominant-locks. Similarly, one expects to find normally subjected to a harmonic interruption individual stretches of common sequential pat- and the piece rebegins its governing linear-tonal terns: circle-of-fifth descents and ascents, tonal motion with the onset of the recapitulation. In motion by seconds and thirds, and the like. One Type 3, 4, and 5 sonatas what usually follows is should also be sensitive to a shift from one strat- a full recapitulatory rotation, beginning with P egy (perhaps a descending circle of fifths) to an- in the tonic. In the Type 2 sonata what follows other (such as sequential rises by whole steps via is normally S in the tonic. (The Type 1 sonata a chromatic bass line), and also to the tighten- has no developmental space.)

ing or foreshortening impression given by any The retransitional procedures just deincreases 1n harmonic rhythm or rate of chord- scribed— dominant-lock, energy-gain, and so

change. on—can recall those that precede the MC in It was always typical for the developments expositions. Indeed, what is usually produced of major-mode pieces to shift toward more at the end of the development is a prominent

“dramatic” minor-mode regions as an expres- HC caesura (although we reserve the term mesive contrast. The submediant, vi, was a com- dial caesura for the center of the layout first promon goal, frequently marked with a vi:PAC. In vided in the exposition), sometimes followed midcentury works this was sometimes the only by caesura-fill bridging the way to the subse-

tonal goal of “the first part of the second sec- quent relaunch of the tonic. Thus apart from tion.” Developments by J. C. Bach, early Mo- their similar local functions there can be 1mzart, and many others often drove efficiently portant parallels between the expositional MC toward it, then concentrated on preparing for and the caesura that typically occurs at the end the recapitulation.* Less often the mediant (111) of the development. One may go further: when was selected instead of the submediant. By the a development is laid out, as so many are, as a time of later Mozart and Haydn—not to men- half-rotation (based only on materials from P tion Beethoven— the tonal plans of these devel- or P + TR: see the discussion of developmenopments grew more complex, the options more tal rotations below), the end-of-development varied, the treatment of the standard move to caesura in some respects “stands for” the earv1 or 111 more flexible, more inventive. Devel- lier MC. The end-of-development caesura, of opments came to have multiple nontonic goals, course —unlike the exposition’s MC —does and the earlier, often single-minded motion to- not usually lead one to expect S (unless a Type ward vi could be displaced altogether. Ambi- 2 sonata 1s in play). On the contrary, it can contious development sections grew in length, re- vey asense of blockage, the impression that one sulting in a central, much-varied action-tableau cannot get beyond the MC-point into S (which 4. Ratner, Classic Music, pp. 225-26, referred to the gal motion, until it reached this goal of “furthest reprincipal “target” of these earlier developments, nor- move.” At this point one encounters “a change of harmonic

mally vi, as the “point of furthest remove” from the intention” —a harmonic divide—and all flows back tooriginal tonic. In his view the development at first ward the eventual recovery of the tonic, in a centripetal continued the process of motion away from the tonic motion. that had been initiated in the exposition, a centrifu-

198 Elements of Sonata Theory

in turn requires one to start over again, with P, plucked from relative darkness (the implication

and execute a full recapitulatory rotation). of impending minor) to the renewed brightness Notwithstanding the development’s typical of the major mode with the onset of the recamodulatory behavior, moving here andthere on _ pitulation. the local level, it 1s helpful to stand back from it This juxtaposition between different phrases and consider its larger tonal purpose in relation (V/vi [interrupted]—I) already had a long his-

to the end of the exposition. When a major- tory by the later eighteenth century. It had ocmode exposition ends, as it usually does, with curred, for example, when the minor-mode a tonicized dominant (V-,,), the entire develop- middle section of a da capo aria pauses on an ment may be heard as a prolongation of this V, HC fermata, V/vi, to return to the major-mode regrasping it and activating it as a chord (V,) opening or occasionally when a second (slow) at the end. The development as a whole at first movement gives way to the third in an early unsettles the exposition’s dominant key (usu- eighteenth-century concerto. Bach’s Brandenally in a set of related modulations) and even- burg Concerto No. 3 in G provides a familiar tually recrystallizes it at the end as a dominant example of the latter: the slow movement ends chord. Normally the intervening keys are to be on a B-major chord, V of E minor, and the last interpreted in relation to this process of pro- movement begins directly in G major. Almost longation. A move to vi can be understood as a any large-scale connecting-point in which an tonicization of the upper neighbor of the pro- interrupted dominant was to be succeeded by longed V; early tonics in the development may something new was susceptible to this “surprisbe construed as only “apparent tonics,” express- ing” chordal juxtaposition. This could include

ing IV of V or V of IV; and so on. The same i: HC MCs in minor-mode sonatas that burst reasoning would be operative for minor-mode into III for S. It could even be used—though works whose expositions ended in the dominant most unusually —as a means of passage from a minor (v). Most, however, end in the mediant symphony’s slow introduction to the Allegro major (III). In this case the V, at the end of the sonata form proper, as in the first movement of development completes a large-scale 1-HI-—V Haydn’s Symphony No. 103 in E-flat, “Drumbass arpeggiation that began with the onset of _ roll.” Here the introduction ends somberly on V

the exposition. of C minor (V/v1), and the sonata form proper takes offin E-flat major.

Substitutes for V, at the End of the Our present concern is the juncture-point Development (Lower-Level Defaults) — the seam where the develop ment ove to the recapitulation. Here too the juxtaposiLeaving behind an active V of the tonic in or- tion, V/vi—I, had a long history of occasional der to proceed to the recapitulatory relaunch use (second-level-default use, as an alternative was not the only way to negotiate the develop- to V,). Since Charles Rosen has dealt with this ment-recapitulation seam. One sometimes finds topic at some length and provided numerous exa replacement of the structural-dominant lock amples in Sonata Forms, we shall not repeat that (dominant preparation) on V, at the close of a discussion here, except to note that he underdevelopment with a seemingly “wrong” dom1i- stood this situation to be a variant of the stannant, most typically V/vi (for instance, an E- dard developmental motion toward vi, which major dominant chord, V of A minor, preced- stereotype, in his view, had become something ing a C-major recapitulation). The effect is that of a cliché by around 1800.°

of predicting a recapitulation that will begin The issue has also been discussed from on the submediant (“relative minor’) but that a Schenkerian viewpoint by David Beach, is actually followed by one that begins in the who—expanding on observations by both proper tonic, I. Expressively, this is like being Schenker and Ernst Oster— regarded the move5. Rosen, Sonata Forms, rev. ed., ch. 10, “Development,’ pp. 263, 267—70 et seq.

The Development (Developmental Space) 199

ment back to I through V/vi to exemplify a at its close. (In cases where the “correct V” is downward arpeggiation from the dominant se- sounded as V$ of I, that is, with 2 of the movecured at the end of the exposition: V—III (as ment’s tonic as its lowest voice, the bass motion V/vi)—I.° From a different perspective, recent from V/vi to the I that normally begins the reneo-Riemannian or transformational theory capitulation will be 3— 3-1 .) In these cases the might be attracted by the V/vi—I shift as one seam between the development and the recapituse —Iintersecting also with other, more tra- ulation is often negotiated in a three-stage patditionally functional understandings— of the tern: (1) a “wrong-dominant”’-lock (typically “PL” operation (a double-transformation of — V/vi, but as will be mentioned below, V/111 is a major triad: (1) major sonority into the par- also common); (2) a concomitant caesura-effect allel minor plus (2) a concomitant 5—6 shift, at the end of this lock; (3) a composed-out coror Leittonwechsel, reinforced by root-support in ridor of caesura-fill, often brief but sometimes the bass) or, within the terminology of Cohn’s extending for several measures, that eventually hexatonic cycles, of the shift between “next- leads, through some form of the dominant, to adjacencies” (or “modally matched harmonies”) the tonic key and the onset of the recapitula-

with “two pc [pitch-class] displacements.’’” tion The juxtaposition of V/vi and I, a phrase- The mediated move from V/vi to I may be ending followed by a phrase-beginning, could exemplified in the first movement of Mozart’s occur 1n either a mediated or an unmediated Piano Sonata in F, K. 280 (example 10.1). Here way. When it is unmediated, V/vi will be fol- the central portion of the development treats the lowed directly by the new phrase on I, with no S-theme through chords outlining a descending additional chordal activity between the two so- circle of fifths: D minor (m. 67), G minor (m. norities. Examples may be foundin the finale of | 69), C major (m. 71), and F major (m. 73, only Haydn’s Symphony No. 103 in E-flat, “Drum- an apparent tonic in this sequential context).

roll” (mm. 263-64, within a Type 4 sonata); Mm. 75-77 provide a neighbor-note cycling and in the first movement of Beethoven’s Vio- in the bass around V of D minor (V/vi) —auglin Sonata in F, op. 24, “Spring” (the develop- mented sixth, passing ¢, and vii? of V/vi—bement ends with a prolonged V of D minor, mm. fore a dominant-lock on V/vi is reached at m. 116-23, and the recapitulation begins directly 78 and extended into a caesura-effect at m. 80 in F major at m. 124). When it 1s mediated one (curiously, lacking its upper voice on the downfinds a passage of chordal fill separating the two beat). Strictly considered, the “wrong-dompoles, V/vi and I. The fill bridges the “wrong inant” caesura is the last structural moment dominant” with a brief channel of harmonic of the development. The implicit caesura-gap slippage that usually touches fleetingly upon (mm. 80-82) 1s filled with three bars of chrosome version of the “correct dominant,” V,, matic slippage, in this case a relatively brief pas-

6. Heinrich Schenker, Free Composition (Der freie Satz), 7. See, e.g., Richard Cohn, “Maximally Smooth Cycles, trans. and ed. Ernst Oster (New York: Longman, 1979), Hexatonic Systems, and the Analysis of Late-Romantic p. 69 (referring to fig. 69 in the volume of examples). Triadic Progressions,” Music Analysis 15 (1996), 9-40; Beach, “Schenker’s Theories: A Pedagogical View,” in Cohn, “As Wonderful as Star Clusters: Instruments for Beach, ed. Aspects of Schenkerian Theory, pp. 31—32; “A Gazing at Tonality in Schubert,” 19th-Century Music 22 Recurring Pattern in Mozart’s Music,” Journal of Music (1999), 213-32; the issue of Journal of Music Theory deTheory 27 (1983), 1-29; “Schubert’s Experiments with voted to neo-Riemannian theory, 42 (Fall 1998); Brian

Sonata Form: Formal-Tonal Design versus Underlying Hyer (who, following Lewin, helped to stabilize the Structure,” Music Theory Spectrum 15 (1993), 6. Beach’s letter-codes L, P, and R), “Tonal Intuitions in Tristan reading of Mozart’s Piano Sonata in F, K. 332, first und Isolde,” Ph.D. diss. Yale University, 1989, and Hyer, movement, has been challenged by David Gagné, “The “Reimag(in)ing Riemann, Journal of Music Theory 39 Compositional Use of Register in Three Piano Sonatas (1995), 101-38. The term “modally matched” stems from by Mozart,” in Trends in Schenkerian Research, ed. Allen Daniel Harrison, Harmonic Function in Chromatic Music: A Cadwallader (New York: Schirmer, 1990), pp. 29-30, Renewed Dualist Theory and an Account of Its Precedents (Chi-

38 n. 13. cago: University of Chicago Press, 1994), e.g., p. 52.

ai a re

EXAMPLE 10.1. Mozart, Piano Sonata in F, K. 280, 1, mm. 67—86

4C—O 3 | 1eee0| 8| «+4 tl, [Allegro assai]

67

Piano (on i Fe, A A A SS A CO, (SG

Ped lO Ooo F—@a29=zro™—“O0OnOnan’-—™ On CQOT J..6WwNW@ . 2 —___ ee eee TO OO

| 8 Sb a!AOyfpiowITD A SS = A SS |*aws LI p— < * , , a eS el .a i

PEE TFT h NOOO eee oom os re}.OtPed ft eOm#wd— FTNCeFNOTOnO0O0W——— eg Ft OOO OOS EE OE eeWwWwW,§ , «s USE OE “FT

$a > _pg> OP _ ee 2D Eeo? Lf O_O --] 7? 2 _ arre Oo S SOOE TSEoo B 70

SOOO eemr aS OE OO 0 EO OOOO em TTT ET OTOT yy *u_.? #!@™29”0#70"9#"-0000.0w47-.w4Ww"wI#wT.w..

—y i. [> ttSS|ToEoODn®™M._.IL |fey? ft las —— > >ODOT ——| gpae 0 gpHe ag Rr pr| SSS FO OS DO S81 A GS Seeeee 73

icy LeOTT a SY OP FO TTT Ss T”r0Sy 9i yO OO SE Mm _ re eonrewPePePereeeeeeeo De — ONO Fo 8 70q#q#48V/.q#C@o@a@aqontc1ctiolIIII.—_ oa EE ggg gh

77 b f 3 3 3 3 3 3 Se ee ee ee =e oe ee | 8

A SSS CO OS f) Tn an

po Po eee TTI TTT ae TTT OTTO TT

0A OA” SS _

8]

TTT OOOO rt.UCotronw™~—oo"-'.:-——| SP —>7— — TS *>FD>>N2_o_>—xa@anv. 0 eT EE —_————— a POON” [/-=z”sz0—>"’7-2... —™O TT _—TenDo>-"@"M0”MoMD m@=u™éé>iz2“—--——O_ OO ——"onon-=0>—~—-'--- OO

et ET ooo WQm SON —-PvAz#j(NYNTNOTTOTLrNwmNW

PS ——— — eSEhFhfeeFeeDENOoNo oi

atay?mld em md a LL CU gm _e._ __ Sf —* eo, ok SY ._-cC!MNNvwv7TwT7"-—"& >= i. 2 OO EE OOOO EE ooTWO O ON4@09>"#"9—"—"W"“4“n”-_ SP — 9-8 OO OO OOOO OO Ee _-§$F——E>=>™$™$]Jmjm-W.

| acit ©, «| §, § 2 of ;>—_2—_2—__ 2" - - - - « oHoN=00—0. 0 _u€M« MW — 9 O_O _ 9 _ _ __ 9? _ O_O _ O_O

i a CS SS CS SC EG GS LSC

DP ODDO SOTTO EEE eT —DT »;_—_—ooT—

The Development (Developmental Space) 201

sage moving smoothly through the characteris- ments whose structural processes proper contic V3 of lin m. 82, which tips the music into F clude not on V/vi but on V/i11, the dominant

major for the recapitulation at m. 83. of the mediant. Both shifts (V/vi to I and V/ Comparable illustrative passages— of many 111 to I) are mediant-related juxtapositions, and possible— may be consulted in the first move- both V/vi and V/ii1 contain the leading-tone ments of other piano sonatas of Mozart, such of the movement’s tonic. (In transformationalas that of K. 332 in F (lock onto V of D minor, theory terms, however, V/1i1 to I involves a min. 123-28, with shift to the minor dominant more radical shift, since there are no common in mm. 127-28 before being adjusted further tones between the two chords.) Here the sucto V3 and V7 of F, mm. 129-32; recapitulation cession, V/iii—I, is more likely to be mediated in F, m. 133) and K. 547a (with a longer, more with a passage of chordal fill (the last part of the composed-out span of mediation between V of _ three-stage pattern mentioned above) for obviD minor and the recapitulation in m. 119); and, ous reasons: the latter triad may also be heard as additionally, in that of Beethoven’s Symphony the former merely hoisted up a half-step. That No. 1 in GC, op. 21 (V/vi lock at m. 160-72; fill, at the end, touches on the generically exwind-tilt in octaves brightening to C major pected V, at the end of the development.? One and the recapitulation, mm. 174—78). More ex- instructive example occurs in the first movetravagant is the expanded fill found at the end ment of Mozart’s Symphony No. 35 in D, K.

of the development in the first movement of 385, “Haffner” (Example 10.3). Here the Mozart’s Piano Sonata in B-flat, K. 333/1 (ex- “wrong-dominant”-lock on V/iu is produced ample 10.2). This features what amounts to a in mm. 111-16. This V/iu1 is interrupted with a “wrong-dominant’’-lock onto V/vi at m. 81, in- caesura-effect at m. 116, and the remaining bars terrupted with a caesura-effect at m. 86, and a before the recapitulation are taken up with an prolonged, elaborately composed-out passage of | expanded passage of fill moving down the circle modulatory fill, touching affectively also on the of fifths (fF, C#7, F$7, B7, E7, A7, mm. 117-28) tonic minor, from the upbeat of m. 87 through and finally emptying out (not cadencing) on the

m. 94. In the case of K. 333/1 a passage of fill D-major octaves in m. 129 that launch the rethat is obviously relatable to the other instances capitulation. Additional examples can be found mentioned above is extended and given a mo- in the finale of Mozart’s C-major “Jupiter” tivic interest (thematic and tonal “reintegration” Symphony, mm. 210-24, with a miraculously after a minor-mode assault) to the point where brightening or clarifying effect (compare also it seems to take on a renewed role of retransi- the first movement, where within the develoption on its own. (Compare example 10.2 with ment the “false-recapitulation effect” in F, m. the similar occurrence in Piano Concerto No. 161, 1s preceded by several bars of V of A minor 19 in F, K. 459, example 22.4. Such illustrations that drift only at the end to the “proper” domare easily multiplied.)® Once the grounding pat- inant); and in the first and last movements of tern is recognized, we may perceive instances of | Beethoven’s Symphony No. 2 in D (first moveits further adaptation or deformation. Some of | ment, aggressive lock onto a prolongation of V

these are brought up in chapter 12. of F-sharp minor, mm. 198—214; sudden shift As suggested above, the issues at hand are to V of D, m. 215; recapitulation in D, m. 216; similar to those encountered with develop- finale, effective lock onto V of F-sharp minor,

8. Other Mozartian examples are mentioned in Da- Zaslaw volume published in Journal of the American Muvid Rosen, “‘Unexpectedness’ and ‘Inevitability’ in sicological Society 51 (1998), 373-84. Mozart’s Piano Concertos,” Mozart’s Piano Concertos: 9. This motion from V/i1i to V can be considered a Text, Context, Interpretation, ed. Neal Zaslaw (Ann Ar- chromatic 5—6 shift—an alteration that also avoids the bor: University of Michigan Press, 1996), pp. 261-84. parallels that would be created in a direct V/111—I proRosen speculated that the procedure may be more prev- eression. The 5—6 shift is clearest if the dominant ocalent in flat keys, although the claim was controversially curs in V® position but 1s only slightly obscured if it is in

disputed by Charles Rosen in the review of the entire root position V or V’ position.

EXAMPLE 10.2 Mozart, Piano Sonata in B-flat, K. 332, 1, mm. 81-97

aPN a ee I (ss nie eeOL ee [Allegro|

Piano ene FS (as Sl : _—_ a A ,. ee( —¢

ptite—_t Tryp, 7 TT otieeeee| eee a AV es AW ns ee OO... ee ee ee ee eeer ee ee eee eee

es.) . »+¢ yj _f[_|[_¢ 9 *} er ee ee¢.ee£f#£. ef, , Ts Qe

f2 )arGY ee a a ——— a“a are OO Oe|

ey 2 ee ee =e ee|eS|ee| ee

383

Ppa" EIEOE__—E> —_——— ULULULLCUTTTCCCS for“—*". _tefo©QQ@2 © wa * a2? ta? ,Os+.hhOOOO OT ==. ZT

ee ne a a A (A

TF yO —e—=— —e—pe

os. p__ COS CO*E Ee —E OS ——_ o7¥Xwr 85

SS "A ” A” A CO OO Oe ee

fy | Fe BleOeBere Poo FLesFOO Or ee ee ee__e_ ee ae ae ae ee 2? ae aow a ee ee es es QsBT > DO

| iVegan ——f dee , £__o OOONOST——WwANA______] ltl, ee ~« eles e.~i..-.. 2... ol? pgp7}@BPat’ os. «fp?1+o « L-—T @ lee | T_T |]tat.1 FS _ Vee 8S _ T_T Dt EOE LALO eo ¢ or of eo Affe OFoe—uNNNWNW

EE es ee Le Le Le eee

a Pata

4 H———_ 34 ____£ @«€- @ © SS

a a ae ee eee py jo ed ee fy FZ a APfA1h SC A Se” Ne

|EaPN aOOee SC

| Poe, fo PtOA rrr-.A $l- aJ—3 Te? «2. SK OOO. Ea, Se SS aSe Aeste SG— =

PY i (a ‘(a ative CC EN§NT §$N Ng EEE eT ——————————e_

PY em OS —= C—O — se ese

A A a YZ a Len, ee aa

fp _ Teepee oo Le. to 2? 7ee sr if »|__| POC | |OW” ll PP2a2—me “

The Development (Developmental Space) 203 EXAMPLE 10.2 (continued)

94 ~ —_ —

Pee > — a -(ce

96 — (ef ie

mm. 157-81; quiet shift to V8 of D, m. 183; first movement of Haydn’s Quartet in B-flat, op. recapitulation in D, m. 185, with the P-theme 64 no. 3 (vi:PAC, m. 120, followed by several

beginning on V”). bars of “grounded” G minor and texture disSomewhat analogous situations at the de- solution to the two common tones, Bb and D; velopment-recapitulation seam are provided recapitulation in B-flat, m. 126). An example of when the development ends with the conven- a brief, two-bar mediation-fill between vi:PAC tional vi:PAC (instead of the active dominant, and a rebeginning on | may be found in the first V/vi) and proceeds with little or no significant movement of Mozart’s Quartet in F, K. 168, mediation into the recapitulation. This pro- mm. 61—62. cedure occurs with some frequency in earlier, Also possible are developments that end by tomidcentury works. It can seem more elemen- nicizing the minor-mode mediant with aiu:PAC, tary, cruder, than the normal practice, since it then proceed to the tonic recapitulation by cuts out the need for bridging the submediant inflecting the fifth of 111 up a half-step (the famuilPAC to the tonic with an extended passage of iar 5—6 shift) to produce the tonic, thus bypassretransition. In part it may also recall the ef- ing a strong dominant, although a brief passage fect of da capo arias whose center-section ends of fill might allude en passant to the otherwise with full closure on a perfect authentic cadence “missing” dominant. The juxtaposition at the in the “relative minor.” Whatever its histori- seam is that of 11—I, transformation theory’s “L” cal memory, what is produced is the direct or operation (Leittonwechsel) —a process that 1s also nearly direct move from closure on vi to a rebe- understandable, like the V/vi—I entrance into sinning on I (connected, at most, by a one-bar the recapitulation, as part of a broad V—i-—I

link, perhaps gesturing at the “proper” domi- downward arpeggiation, reckoned from the nant en route)— essentially the “R” operation end of the exposition. A virtually pure example (to the “relative” major) in terms of transfor- (though mediated by a brief V3) may be found mational theory. This may be found in some of in the first movement of Haydn’s Quartet in C, the first movements of Mozart’s earliest violin op. 33 no. 3, “Bird,” mm. 108—11; another ocsonatas (in B-flat, K. 8, mm. 46—47; in G, K. curs in the first movement of his Quartet in C, 9, mm. 59-60; in F, K. 13, mm. 64-65), and op. 76 no. 3, mm. 65—79, in which the seeming in a scattering of other works. A more sophis- mediation, here by V®, 1s actually the upbeat to ticated example of a virtually unmediated case P itself. A more elaborately mediated instance of of vi—I juxtaposition at the seam occurs in the i11—I occurs in the finale of Mozart’s Symphony

ES eeYTN ee ee er ge KT eee Y OT Te a Y (oer em Foy Aa rl fl fi A i i EXAMPLE 10.3. Mozart, Symphony No. 35 in D, K. 385 (“Haffner”), 1,

mm. 111-33

Allegro pirito] con spirito WT [Alleg a, ae

ae eee ae ci > OP De 0 Or I mel” a| fFeStivSS esfFOO > ee ¥ J FD ga ~/esiFee | ¥ ee” [| gg 2“~~ | gf sO - #ga i ~ im fFf |i

| ‘1 PP P I14

Me TN ge yy TT —“e aFE ET] P y9Y,J =ee Y i y a Qo... o> “o> — Be QD ne acl (S(T Tf eS tTIfy gy eT, a T , pe tr——. af) "et ee

eee. eee eee eee eee a ee.) eee Jfifa Ne Ge. CCCCdC ey =ieelaheee t—i‘CS;COC;CéCS yd ee at 4ee 2 “ee iF. ae ri (ststi‘“‘“(‘=

rT y/ ST e ee —= 127 a ”

es es ee es | eS es es T

124

= 20 De, ee ee, | ee ee | ee ie TP nnnnnnrrrrrrrr—(—tsCCC(‘“‘“‘

rw em CdCl lmtti‘(Cisd,:ti‘(‘(C;(C(CS*‘“‘(‘;SST.)CUTTLLLUUti“‘(‘(C‘(CSCO(t+tYG OF Ulgti“‘“‘c(;LLCUdELCLCUhUDTLULULLLLCUN TC Cd)

—_—_— a —* o>

t ep of A Serato oe) fle ——~ Y eee wee ETT

ime" “ff A ff #|SY goeyp =| @qyp A SS 2SdJfqy OSP|CS | OY= #Jj|qy 2 CSsit|

Oy Pp rt—“—itCsdrSC“(‘(;SC~“t;éi‘iLSC‘( PT me —“(—ti‘“s*sSsdr:*tsi eC iii se OO lm est—“‘“‘(‘(‘(‘(‘( .es mT SS tfh Ibe©If Seege © D re pp — CFese.

||eeFm elel 4ID *« ooOA o9iZ_[nau| lO, I—“—sT SCnce ed aseeSdI Ii (| OD| QO Pee emt CCC (ited ey Tti—“‘Ci(‘CSCSsi‘(‘(‘(‘(‘(‘(‘i‘aL COC V@SxSXwKwPCOCOLLULUmLLcC h,LLULULmELmLmLlrlCUmE CGO —CsSCS

se —~ J J Jd be waa Ez PPS here oT we OO —— OJ]

hy CE ASD 2” OO a nce an

yet __ AGS hg A eeIaeOg YS A” Gn GO er aoOeS O iA

rl A

The Development (Developmental Space) 211

D / Py ; Lt OE es ee A|| Py Pe = DO 140 ;> ‘ p ;b ;‘p EXAMPLE 10.4 (continued)

1 Yoroy ro , oY 137 1 eaten —— 8gphn| _/I__ql74 s ba Po 2 2Pt a) Dm \ Jey “eT s Lf gy _i__@_i4 p = f Nee’ ri ‘| K i i i i A r FY A Ne rn ‘|

Lb ¥ T ¥ Lb Y ¥ ¥ .YY

yp rwllee) b® be hrc ae#nee elb#ebel en i.#ae Ao; Ao A — id Ai St L ¥ ¥ if , — ' ; St L T ¥ ¥ , — ' ;

RA — ra —

2 i / / rf DB r ’ / . f r r

ee Ee f) ; D i ee a a

afos) esaeeFr eeOO eee re ee gS 8eeSS Co eses|2ieerr Oe es 142 L ‘

a oA 4d | Ao od 4 2 a 4 bk kh kk kk

be o—~

ty tdTN!![WWHHftHf/72SY7Y7}]}a>}7W7}Y}?;7]Y‘/TWtYSTTST] TT IIT

(—2, m. 130, effectively a dominant-lock mov- emerges out of the exposition’s second ending

ing toward an eventual MC-effect in m. 142) (with A major, +1, being converted into an and a new section beginning on E-flat major RTJ-like V,) to begin the development proper (—3, m. 144). Here Beethoven takes a standard with P on D minor (0-level, m. 138—the tonic opening strategy and submits it to exaggeration return here being substantially colored by the to fill most of the first portion of the develop- lights-out effect). In turn, D minor gives way to

ment. an extraordinary string of fifth-descents —a slipSimilarly, when we find that a development page downward desperately seeking a foothold begins with an immediate return to the ton- to stop the sinking, all articulating P-material: ic—seeming to reinstate a tonic-return that G minor (—1, m. 146), C minor (—2, m. 148), F had been earlier associated with the expositional major (—3, m. 150), B-flat major (—4, m. 152),

repeat — what is doubtless implied is the sup- an E dyad (—5, m. 154), and so on, with the pression of the normative +1 (dominant) level chords now changing every bar. In cases where (which might be “prematurely” tucked into the the exposition had not been repeated —as 1n the second ending or even earlier) in order to begin first movements of Mozart’s Wind Serenade in directly with the 0-level. ‘This occurs in the first E-flat, K. 375, Haydn’s Piano Sonata in D, Hob. movements of Beethoven’s Piano Sonata in G, XVI:51, and Beethoven’s Quartet in F, op. 59 op. 31 no. 1, and his Symphony No. 2 in D, op. no. 1—the local impression can be that of an 36. In the symphony the developmental space expositional repeat begun, then aborted.!”

17. In finales from the decades immediately around one might also ascertain that what 1s being confronted 1800 (for which the Type 4 sonata was a real option), is not a “sonata-rondo.” For first movements and over-

212 Elements of Sonata Theory

While the initial tonal gesture of the devel- forced if the development proceeds into TR opment was not invariably associated with allu- or, in some developments, into part-2 material sions to the fifth-chain, it often was, and master (C, less often S), in this latter case producing a composers found creative ways to manipulate full rotation. What are we to make of developthe implied rapid changes of bright-to-dark ments that begin with essentially new material, tonal colors. Should a development juxtapose not with a self-evident restatement of any prethe major-mode end of the exposition with a ceding theme? (Although such episodes may be quick shift into an even-deeper subdominant — motivically related to earlier themes, one should direction, the effect can be somber indeed— the not overplay this hand. Within a style grounded plunging into an abyss. In the first movement of in scales, triads, and neighbor-note relations, it Mozart’s Quartet in D Minor, K. 421 (example is usually an easy matter to “derive” one theme

10.5), the exposition’s F-major end (III, —3) from another.) sinks through its lower fifth (—4, merely a Bb$ Episodic openings are met with fairly frechord in the second ending, m. 41) to launch quently —it may be a second-level default opthe development still another fifth lower, with tion — and it was also a practice recommended by

P-material on E-flat major (—5, m. 42—the Galeazzi in 1796. An episode suggests a changNeapolitan of the tonic, D minor). This chill- ing of the subject, a reluctance or unwillingness ing effect is rendered even more unnerving to do the more standard thing. It could also sugby the onset of a stepwise, flatward descent of gest, as often in Mozart, the sudden flowering the bass (suggesting another lower-fifth, A-flat of a new theme, the exuberant overflow of a minor, —6, m. 44), the shift to an ominously thematically abundant master.!® Or as Tovey put hushed piano dynamic in m. 45, and the bleak it in 1929, “Such an episode, which is generally deep-shadow of the pianissimo V¢ of A minor placed at the beginning, by no means always

in m. 46. indicates a lighter style and texture. It may be a

Once again, not all developments begin with relief from unusually concentrated figure-work fifth-descents. Our concern here 1s only to call in the exposition.”!? Episodes may also occur attention to this typical procedure and to re- further into the development—a topic taken up mind ourselves of the importance of the rela- in a separate section below.

tionship of the initial developmental key with When such an episode opens a developthat of the end of the exposition. Whenever the mental space, it is important to examine what development begins with a key different from happens in the remainder of it. If what follows that which ended the exposition, a first-level upholds the rotational convention —1f what foldefault option is being bypassed. As listeners lows brings back selections of expositional mawe are encouraged to inquire into what the terial in its original order (perhaps returning to intended effect might be. This is also the case P or TR, perhaps proceeding directly to S- or with developments that begin with other than C-ideas) —then two possibilities exist and may P-material, a topic to which we now turn. even overlap. The initial episode may be either an interpolation wedged into the work before

4. eye ee > the onset of the developmental rotation proper

Episodic Openings: “Writing Over or a substitution for the more standard P!. In the

A P-based opening invites the understanding former, less frequent case we would expect the that a new rotation 1s underway. This 1s rein- episode to be followed by a more or less normatures in all periods, the Type 4 format (sonata-rondo) 18. Cf. Caplin in n. 12. was never an option and should not be entertained as a 19. Donald Francis Tovey, “Sonata Forms,” written possibility — even in repertories as late as Brahms and for the fourteenth edition of the Encyclopaedia BritanMahler. Since the Type 4 option declined rapidly in nica (1929), rpt. in Musical Articles from the Encyclopaedia the nineteenth century, even within finales, one should Britannica (London: Oxford University Press, 1944), p. normally suppose that mid- and late-nineteenth-cen- 215; reissued (1956) under the title The Forms of Music. tury finales —as in Brahms—are in dialogue with Type 1 or Type 3 variants, not with Type 4s.

39 J oN -_ . os m1 poo ee, SS ee Ne’ . . tr yj EXAMPLE 10.5 Mozart, String Quartet in D Minor, K. 421, 1,

mm. 39-47

ae 6

[Allegro moderato] = Ly

|SFaPo1. | 2 Se a | FP pe gg i (Fee eee A arrr A a TS Bn A 6 a SS FE BP > 7

6 aw ~ Nee mandy As. FW, yf} —_ SZ Fb NN —p TE o—p TF

A eS Ar FO I060G_V“JK_~VC"_"1/!_—— ———— ———————.._ ?# —@ |{&KPeEp?)}-"—".17.—_ — i —_9_9@_

4] —— too. ; —— Lo. ,

Cresc eee tT tS a2J72

Dy _ -+- I:Ss 7: O° — Ft 41 A yo~N ee HN ees fh? eoNO _ eo,pc fF| ee op 7 CFESC. — — == — om ee” Py J- — I aa a | 2.

fea? pe peg & oy.” pe ew we et

££ ww oD OT’? Le

Oe ai py SS7 Se Cy. 2-—mD@V\LvNMNT1\![?——OOOO POT Oe OH$324 9A)sO SO —y—_—_+—____—________|______, A OO

(yo OOOO Eo eS Eo or ef Nu0W— st A TO -_ 98> a TO OI 2 aESeeeSSAS ES SS OA OS AT SSS A SS SD SS SS ——— SS A == OD OOS(#70 OS SOO ters _Oe| ___*_t TO > = oa _—e_qQUO Ee oe

EF YS Oe a. Te iy CS 2” 2 2 2 A eS RS nA S.C SS a aS A SS SS —— Se fo FiGS TF FDL PSS eso eS cu,otaeee GC ae,

94

Fe OOO sf0 A >|.‘na CC _a /.. p}

| Ve eee TTdd6envc OOOO —>?——— TT TTF uo™oO

96 ’ ’ N ’ ’ A a Ol nl Ec GS cl GO

ps —__ py | SC _—— —— ee A SC OC GC CC Ea, OC SC 8)Ne AA a A a A _ ~ Ne rsa

Oi 262 Elements of Sonata Theory

EXAMPLE 12.2a Mozart, String Quartet in C, K. 465 (“Dissonance’’), 1Vv,

pos a as oo = OOOO P —< mm. 1-4

Allegro molto

PJ a; oy - , tte be

on C 1s preceded by two false starts of the P!.!- starts in the repertory is to be found in the first idea— two successive hoists. The first begins movement of Schubert’s Piano Sonata in B-flat, on A-natural (A minor, m. 86, triggered by a D. 960, where the gentle P-idea vacillates achstrong v1:PAC concluding a rotational develop- ingly (mm. 188—215) between an unutterably ment); the second on B (m. 90, outlininga V’ of | despondent D minor and a seemingly distant C). In the second, from the Allegro molto finale B-flat major before being able to pull itself toof his Quartet in C, K. 465 (example 12.2), the gether sufficiently to decide for the latter and development proper ends with a strong cae- the full burden of the recapitulation proper (m.

CI 11. ;

sura on V/III (m. 180, V of E, the “wrong 216). key”). This 1s followed by three false-start, an-

is 3 elena imines bam lee ae Recapitulatory Rotations That Begin in IV if questioning, then aborted again); beginning A subdominant recapitulation is one that begins again on E minor, dissolving to a new retran- its recapitulatory rotation and initial set of corsition and a setup on the correct key, V of C respondence measures with the sounding of P!-! (mm. 190—98). The recapitulation itself begins in IV. Subdominant recapitulations typically ocin C major at m. 200, now fully corrected and cur within Type 3 sonatas with rhetorically full

prepared. and normative P-zones. This may seem self-evRelated cases in Haydn may be found in the ident, but dubious assertions have been made first movements of his Quartet in E-flat, op. that subdominant inflections of later modules 33 no. 2 (false start on vi in m. 59; tonic re- are also instances of this practice.!! Normally, capitulation in m. 63) and his Symphony No. one should not consider the sounding of any 43 in E-flat, “Mercury” (false starts on IV and later module in IV (even P!-2) to be the onset of 11, mm. 152 and 157; tonic recapitulation in m. a subdominant recapitulation. Such a situation 1s 162). One of the most moving series of false almost always better interpreted as a Type 2 so-

11. One of the most remarkable of these is that found pearance in IV, however, clearly belongs to developmenin John Irving, Mozart’s Piano Sonatas: Contexts, Sources, tal space. Confirming this assessment, S?-*, first heard

Style (Cambridge: Cambridge University Press, 1997), with the upbeat to m. 29, reappears within the later p. 102, which proposes that a subdominant appearance tonal resolution—and in the proper tonic key —starting of an interior module of S (we number it S?-7) within with the upbeat to m. 92. Irving’s explanation would the developmental space of the opening movement of have it appear twice, in two different keys, in the “rethe Sonata in D, K. 311—m. 58, or the upbeat to m. prise.” K. 311/1 is the most structurally problematic of 59—launches the “Reprise” and thus is “akin” to the Mozart’s piano-sonata movements—seemingly a Type practice of what we call the Type 2 sonata, now merged 3 sonata that converts midstream (m. 79 with S!) into a with the possibility of a subdominant recapitulation (as Type 2—and its puzzling structure has led more than suggested in the related p. 190, n.10). That module’s ap- one analyst to grasp at straws. See ch. 17.

met .2 pA a(Cea ee or oe }Filia... OS ee ie, —| V _ o— p os EXAMPLE 12.2b Mozart, String Quartet in C, K. 465 (“Dissonance”), iv, mm. 178-204 [Allegro molto]

178 _ women,

#, J:aa te ones D2 we

‘nee’ ° . ° p 184 a aoo~

. é — = hie a a (yigts eu se 2 a ligis | [2° aa)_| it.oe ff2s.aml ee / ad tye aee a a eS a ee ee ee 190 = . ea ———. —

()-$— o_o | Cotes/ ee | oi} ef ge—ef TTT yy ——— —_— — OF

o~ .>p—. — 196 _————

a a a Ce a ae

ys, 2; LY L/ y, p Nee’

P —,

—————————

201 J aN —~ Coo na —_—— ee ———— oS . . . . P< $$ ______________}____4 arr ee eee ee ee ee ee,

es es ee ee a; , ttete Lae a aTe ee

264 Elements of Sonata Theory

nata with early crux, as in the often-cited finale ennese connoisseur could interpret this device of Mozart’s Quartet in G, K. 387, with an em- as a bifocal recapitulation in the subdominant, phatically prepared and launched P!? crux in IV known from chamber and symphonic works at m. 175 (more clearly, m. 176 = m. 18). In this by such composers as Christoph Sonnleithner case P!-? continues the ongoing rotation with and [Florian] Gassmann.”!4 Adding to our percorrespondence measures: the preceding devel- spective, Bonds informs us that “unlike many opment, of course, had been based on P!:!. The of his contemporaries [including Dittersdorf, highly dramatized return of P!? in IV, however, Gassmann, Stamitz, and D’Ordonez]... Haydn does suggest the secondary infiltration of some was never particularly drawn to this procedure. aspects of Type 3 logic into the movement. He experimented with it only occasionally, and For most writers today the touchstone cases even then only on a very limited scale. There is —setting aside the perhaps misconstrued K. 387/ no counterpart in his entire output, for example, iv—are the first movement of Mozart’s Piano to the extended subdominant thematic return in Sonata in C, K. 545 (with recapitulation begin- Mozart’s Sonata in C Major, K. 545.”"!5 ning in F major), Beethoven’s Coriolan Overture, Charles Rosen noted the presence of the op. 62 (C minor, with recapitulation beginning subdominant return in the first movement of in F minor), and the first movement of Schubert’s an early (1757, non-Viennese) keyboard sonata Symphony No. 5 in B-flat, D. 485 (with “reca- (No. 4) of Giovanni Marco (Placido) Rutini pitulation” beginning in E-flat—but see our and proceeded to disparage it in high-moralisqualification of this assessment later in this sec- tic terms: the Rutini work contains a “full retion).!? Additional examples in Schubert— who capitulation of the exposition beginning on the was much attracted to this option, particularly subdominant, a form that was to become a lazy between the years 1814 and 1819— may be found mannerism only after 1800.” Rosen’s distaste in the first movements of the Symphony No. 2 in for the practice surfaced elsewhere in his Sonata B-flat, D. 125, the Piano Sonatas in A Minor, D. Forms, sometimes taking on a tone of shudder537, and B Major, D. 575, and the Piano Quintet ing censure: “There even arose a kind of de-

in A, D. 667, “Trout.”!5 generate recapitulation, which began not in the Although infrequent, recapitulations starting — tonic but in the subdominant, and which made on IV turn up consistently enough in the eigh- possible a literal reprise of the exposition, transteenth century that we consider it a lower-level posed down a fifth.” “For a massive recapitula-

default option within the genre, not a defor- tion starting at IV, see Hummel’s Piano Trio mation. Composers who exercised this option in Eb major, op. 96, sometimes also labeled op. were drawing on several precedents. George R. 93.” “The opening of a recapitulation in IV 1s Hill and A. Peter Brown have suggested that also used more frequently by Schubert than by the procedure is characteristically Viennese. Ac- any other composer.” !© And so on. cording to Brown “an eighteenth-century Vi- Some of C. P. E. Bach’s symphonies (which

12. Cf. also the odd tonal structure of the C-minor sec- “Trout” Quintet, D. 667. See also the discussion and ond movement of Beethoven’s Piano Sonata in E-flat, listing in Tobel, Die Formenwelt der klassischen Instrumen-

op. 81a, “Les adieux,” already discussed in ch. 11. This talmusik, pp. 170-75. is arare example of a Type 1 sonata (thus containing no 14. Brown, Joseph Haydn’s Keyboard Music: Sources and

development) whose recapitulatory rotation begins in Style, pp. 352—53. See also George R. Hill, “The Coniv, F minor (m. 21 = m. 5), a tonal situation made pos- cert Symphonies of Florian Leopold Gassmann,” Ph.D. sible by the diminished-seventh sonority launching the Diss., New York University, 1975, pp. 161-96. The

recapitulatory rotation in m. 21. Hill- and LaRue-grounded term “bifocal recapitula13. The dating claims are to be found in Martin Chu- tion” 1s avoided by Sonata Theory.

sid, “Schubert’s Chamber Music: Before and after 15. Bonds, “Haydn’s False Recapitulations,” pp. Beethoven,” The Cambridge Companion to Schubert, ed. 244-46. Christopher H. Gibbs (Cambridge: Cambridge Uni- 16. Rosen, Sonata Forms, rev. ed., pp. 144 (“full recaversity Press, 1997), p. 186. On this basis Chusid pre- pitulation”), 288 (“There even arose”), 289 n. 4 (“For ferred a dating earlier than 1823 or 1825 for the A Major a massive”), and 360 (“The opening”).

Non-Normative Openings of the Recapitulatory Rotation 265

foreground bizarre effects) contain subdomi- finales of op. 10 no. 21n D and op. 13 no. 5 in F nant entries into the section best regarded as (revised to include a recapitulation in IV in the the recapitulatory rotation. (Normative “de- ca. 1810 version!).!® velopments,” “recapitulations,’ and “codas” are Within major-mode works there is a self-evisometimes submitted to substantial thematic and dent logic behind the choice of a subdominant tonal deformations in these movements, whose recapitulation. Since the exposition had moved large-scale shapes are rendered coherent chiefly from I to V (modulating up a fifth), one could by a readily perceptible adherence to the rota- always produce a perfectly parallel recapitulational principle.) Examples occur in four of the tion, by-for-bar, that moves from IV to I (again first movements of the six “Hamburg” sympho- modulating up a fifth), thereby producing the nies, Wq 182 (H. 657-62, 1773): No. 1 in G necessary tonal resolution for the S and C zones. (recapitulatory rotation proper—amiudst many This is precisely the solution, for example, found other rotational complications — beginning in in the first movement of Clementi’s Sonata in BC,m. 59); No. 3 in C major (recapitulatory ro- flat, op. 10 no. 3 (whose recapitulation Plantinga tation starting on F, m. 69); No. 5 in B minor described as “proceed|[ing] to the end as an exact (recapitulatory rotation beginning in E minor, transposition of the exposition’’) and in several m. 35, after four false-start effects along the as- of Schubert’s works.!? And yet this easier transcending circle of fifths, C, G, D minor, and A positional route was not always taken. Mozart, minor, mm. 33—34); and No. 6 in E (recapitu- for instance, did not provide any such slavishly latory rotation starting in A, m. 53).!7 The sub- parallel recapitulation in the first movement of dominant recapitulation also surfaced in several K. 545. There the recapitulation contains an inof the piano sonatas of Muzio Clementi from terpolated four bars (mm. 50—53) that, at least 1780 onward. These include the first move- theoretically, with small modifications, could ments of op. 5 no. 3 in E-flat, op. 10 no. 3 in have been omitted).?° B-flat, and op. 13 no. 4in B-flat, along with the More likely, this penchant for the subdom117. Especially since these first movements combine an K. 455: An Uninterrupted Sonata-Form Movement?” oddity of syntax with the lack of the guidepost of an Theory and Practice 16 (1991), 64: “Schubert on several expositional repeat, their overall plans are easily mis- occasions begins the recapitulation of a major-key soconstrued. Usually the return of the P idea marks the nata-form movement in the subdominant— but always onset of a new rotation—development, recapitulation, with a preparation and in all cases the transition actually and coda—although the manner of approaching this modulates, so that the recapitulation is in fact a transpoP (not to mention issues concerning key patterns and sition of the exposition, measure for measure.” EEC attainment) can be extremely unorthodox. In the 20. Snyder, on the other hand (“Schenker and the First first movement of Symphony No. 3, for instance, the Movement of Mozart’s Sonata, K. 545,” 57), insisted main divisions are as follows: exposition, mm. 1—51; (mistakenly, in our view) that Mozart’s recomposition developmental space (half-rotational), mm. 51—68; re- of the TR (or bridge) was “stylistically imperative”: “In capitulatory rotation (beginning with P in IV), mm. the exposition, Mozart had not established the dominant 69—124; coda (incipient rotation, beginning with P in key prior to the appearance of the second subject; that is, the tonic), mm. 124—28 [broken off]. As in Haydn’s the bridge leads to a half cadence on the dominant, not Symphony No. 6/1 (mentioned in Ch. 11) the S!! idea in the dominant. In a ‘normal’ sonata-form movement, is no theme, registering only a space of thematic absence the tonic is re-established prior to the return of the first (mm. 24—27). An S!+ module brings about the EEC subject by a dominant preparation. Since that has not at m. 43; the C-theme is found in mm. 44—51. The happened here, it becomes stylistically imperative that recapitulatory precrux alterations occur at m. 76; the the tonic be firmly established before the restatement of crux at m. 80 (= m. 11); another eccentrically “wrong” the second subject, so that there will be no doubt about the entrance of S!-! (and still on the dominant!) at m. 93; a resolution of the ‘structural dissonance’ the second sub-

tonal correction with added postcrux alterations in S!:3, ject represents. A simple transposition of the bridge

mm. 105—8; the ESC at m. 116. from the exposition would have failed this task; the ex18. Leon Plantinga, Clementi: His Life and Music (Lon- tension of the bridge passage is therefore structurally don: Oxford University Press, 1977), pp. 75—78, 90, necessary.” But there is no “necessity” in such matters.

and 220. Other compositions with recapitulations beginning 19. Plantinga, Clementi, p. 90. Cf. John L. Snyder, in IV, such as the first movement of Clementi’s B-flat “Schenker and the First Movement of Mozart’s Sonata, Piano Sonata, op. 10 no. 3, mentioned above —and

266 Elements of Sonata Theory

nant could have arisen as an extension of the sure equivalent of m. 17 of the exposition.) As more common principle of moving toward the we continue to inch the transposed-crux point, subdominant in recapitulations shortly after now tracking in IV, toward the beginning of the the initial re-sounding of P in the tonic. As dis- rotation— and finally attain 1t—it would seem cussed in chapter 11, the subdominant-shift of | that what we have produced is a Type 2 sonata P!.2, P?, or TR material before the medial cae- with an unusually advanced crux-point. Or has sura 1s characteristic of recapitulations in gen- the sonata-type itself changed? If so, at exactly eral, doubtless because the fleeting toniciza- which point? Is it preferable to insist that when tion of [V in TR facilitates the impression of a the initial bar of the rotation is touched (or bar “modulation” up a fifth (corresponding to that 2? bar 3? bar 42), our conceptual assessment of of the exposition) that will bring the music back the structure should change to encourage into I. If the [V-convention were to be pushed stead the perception of a Type 3 sonata with back further toward the beginning of the rota- subdominant recapitulation? tion —to the point of coinciding with it— one There can be no denying that there are shadwould have produced a recapitulation that be- ows of Type 2 logic lurking behind aspects of

gan on the subdominant. the Type 3 sonata with subdominant recapituThis line of reasoning leads one into compli- lation. Still, one of the essential features of the cations. Consider, for example, the normal struc- Type 2 sonata, an expanded development after

ture of the second rotation of a Type 2 sonata an off-tonic treatment of P!, is missing here (chapter 17). Here the initial, nontonic P-ma- (although, to be sure, the TR may be “develterial billows out into a broader development opmentally” recomposed). More to the point, (usually based on P!, P!-, P?, or TR) and pro- within this Type 3 variant the P-incipit-launch ceeds eventually to rejoin the expositional track in IV is always preceded by material previously

at a point of crux shortly around or before the interpreted as development. In other words, medial caesura. When the expositional MC had were we to insist on interpreting the form pribeen articulated as a V:HC, that crux 1s likely to marily through Type 2 expectations, we would be sounded a fifth lower, in order to produce the have an “extra” or redundant developmental desired recapitulatory I: HC MC. For this reason space to account for—a separate development recurrences of correspondence measures a fifth in its own right—which may or may not have below the expositional model are common oc- begun with a tonic or nontonic sounding of P. currences in [ype 2 sonatas. In most cases origi- From this perspective the subdominant-recanally dominant-key ideas will be reappearing in pitulation-effect provides at best an extremely the tonic. Now let us suppose that the composer awkward variant of a Type 2 sonata. decides to slide that point of transposed-crux Thus the Type 3 based interpretation has the further back toward the beginning of the rota- upper hand in this potential blending of Type 2 tion. Once the composer has moved the trans- and Type 3 principles. In virtually all cases the posed-crux back into originally tonic material, two most obviously complete rotations (expothat material will appear in the subdominant in sition and recapitulation) are separated by dethe recapitulation. (As mentioned earlier, Mo- velopmental activity. But even having observed zart provides a convenient example in the Type this, things are not always so clear. The capacity 2 finale of the G-major String Quartet, K. 387. of the subdominant variant of Type 3 to enHere the subdominant crux, m. 175, is pushed ter into a dialogue with both types can at times back to the P!-? point, the correspondence-mea- be made a topic of the overall structure of the

several works of Schubert as well—do not follow any It overlooks the Type 2 option, in which that dominant such “imperative.” Moreover, Snyder’s premise about preparation did not normally occur. Finally, Snyder’s [longer passages of?] dominant preparation preceding discussion of what we would call a I: HC MC 1s irrelrecapitulations— very much in line with Schenkerian evant to the matter at hand: such MC’s could be—and principles of harmonic interruption on the tonic’s V—1s were—treated in a variety of ways in recapitulations. based entirely on normative Type 1, 3, 4, and 5 sonatas.

Non-Normative Openings of the Recapitulatory Rotation 267

piece. This can happen when for one reason or sic instance of a P!-° (mm. 1-4). The apparent another the material of the developmental space recapitulation begins in IV with the P! theme. is selected in such a way as to suggest that it But the preceding development had been a vast belongs conceptually to either the expositional expansion of the P!-° idea. Thus the entire strucor the recapitulatory rotation. In such cases the ture gives the impression of a large double-rotaanalytical task is not to decide whether a given tion (exposition + development-recapitulation), structure “is” exclusively a Type 2 or a Type 3 with the crux pushed back to the P! point—a sonata-variant. Rather, the aim would be to ex- moment that, because of the presence of P!-°, plore the compositional interaction between the does not correspond with the beginning of the two principles and to notice how those differing exposition proper. Notwithstanding its ternary principles are brought into productive tension impression, this movement unfolds much like a

within the work at hand. Type 2 sonata with an early crux.

In the first movement of Mozart’s Piano So- Finally, one might mention the clever adapnata in C, K. 545, for example, the development tation of the subdominant recapitulation in the section (mm. 29-41) is devoted to an elabora- finale of Beethoven’s Quartet in E-flat, op. 127. tion of the exposition’s brief C idea (sounded in In the exposition P stretches out as the first two mim. 26—28 after the EEC in m. 26). One point __ portions of a broad-spanning ternary plan, ABA’.

of the retention of C-material may be to suggest Following an initial P!-° invocation, P! (A) 1s a the persistent echoing presence of the final ele- closed parallel period, mm. 5-12, 13-20. P? ment of Rotation 1—as if Rotation 1 were still (B) is a briefer, thematically contrasting parallel “in the air,” even in the developmental space. On period in the tonic (mm. 21—24, 25-28), imthis interpretation the onset of the subdominant mediately subjected to a varied repetition (mm. recapitulation (m. 42) could be understood as the 29-36). The P! reprise (m. 37) seeks to close beginning of Rotation 2 of a double-rotational the ternary scheme with a I:PAC but 1s unable sonata that had included no self-standing, inde- to accomplish the task. Thus the expositional pendently rotational (or episodic) developmental P! comes to be understood as a TR of the disspace. Such a view would bolster the Type 2 as- solving-reprise type, barely touches on any MC pect of these Type 3 variants. This observation at all (perhaps an elided V:PAC MC at m. 55?), would be generally applicable for all Type 3s with and launches directly into what has usually been

subdominant recapitulations that also featured regarded as a blurted, forte Sin m. 55. preceding developmental spaces (or expanded The seeming recapitulation on IV, beginning retransitional links) based on the concluding in A-flat in m. 145, provides us with P! and P? element(s) of the exposition’s C-theme.?! complete (A—B of the ternary structure) in that The first movement of Schubert’s Symphony key, only undermining the expected cadence of No. 5 in B-flat, D. 485, addresses the issue in P? in m. 176 and thereafter modulating away an even more provocative manner. The expo- from A-flat. What one expects at this point is sition’s P! theme (m. 5) is preceded by a clas- an imminent reprise of P!, probably returning 21. Here one might recall that a retransitional link ‘recapitulation’ was not solidified until the nineteenth based on the last module of C connects the two ro- century [sic/]. In this case, therefore the Kopfton [5| will tations of the Type 2 sonata in the finale of Mozart’s be recovered with the restatement of the second theme

Quartet in G major, K. 387. Quite apart from these material, in the tonic, beginning at m. 59... . The rotational matters, John Snyder, noticing the lack of the ‘subdominant-key recapitulation’ —long recognized as

Schenker-required harmonic interruption on V of I be- an oddity—turns out not to be the recapitulation at fore the “recapitulation” in IV in K. 545, has suggested all, but merely a statement of the primary material in a differing solution, although one that we cannot en- that key, in the course of the development.” To be sure, dorse (“Schenker and the First Movement of Mozart’s Snyder may be correct about the Kopfton issue, but this Sonata, K. 545,” 64, 69): “My thesis is to consider the line of reasoning, overlooking the rotational principle restatement of the principal theme at measure 42 as part entirely (and operating from an at best hazy awareness of the development. It is useful in this regard to remem- of how “binary” or Type 2 sonatas characteristically unber that sonata form is ancestrally a binary form, and fold in the eighteenth century), runs aground from the that the distinction between the ‘development’ and the perspective of Sonata Theory.

268 Elements of Sonata Theory

as a recapitulatory TR. In fact, that P!—pre- was the case, the alternative key most often chosumably the final limb of the A—B—A melodic sen seems to have been that of the submediant, structure —recurs in the tonic, E-flat (m. 187, usually with the sense of momentarily locking sometimes regarded in the literature as the onto a set of P!! correspondence measures— the “true” recapitulation). But instead of providing usual signal of relaunch. The tonic is soon reonly a brief concluding reference to the main gained, and the rotation stabilized, three, four, theme, Beethoven now “backed up” the music five or more measures into the rotation, perhaps to furnish once again not only the full P! pe- even with its second thematic module. (‘The roriod (as if correcting the earlier A-flat recapitu- tation is not rebegun in the tonic from the P!! lation by retracing its steps, albeit with variants) point. If it is, the brief, preceding submediant but also a full restatement of P? (mm. 203-10, P!.1 would be a false start, and the recapitulation 211—18)—which then merges directly into “S” proper would begin with the tonic statement. in E-flat (m. 219). The whole procedure begin- We have already cited the opening movement ning at m. 145 has a false-start flavor (a recapitu- of Haydn’s Quartet in E-flat, op. 33 no. 2, as an lation begun “wrong” and set right by rebegin- example of this.) The correction from nontonic ning again), but it is carried on too long to be to tonic necessitates the altering or addition of a genuine false start, and it is also true that the one or more measures shortly into the rotation. ternary plan of the P-theme, suggesting possi- Such nontonic recapitulatory openings imply an bilities for the manipulation of the P-reprise (or expressive strain at the beginning of the rotaTR, mm. 187-218 of the recapitulation), plays tion, as though the tonic-track 1s still something a prominent role in the formal conception. Kin- to be achieved. derman is probably correct in regarding m. 145 Hill has pointed to some instances of this as an example of a subdominant recapitulation; submediant opening in the Viennese concert Kerman’s earlier suggestion that m. 145 begins symphonies of Florian Leopold Gassmann: “In “the most obstinate ‘false reprise’ in the whole a number of first movements that seem clearly

classic repertory” 1s less satisfactory.” to be in sonata form he begins the recapitulation in the subdominant or submediant (relative mi-

Recapitulatory Rotations That Begin in vi, nor); achieving a firm return to the tonic only

VI, or SVI with the reappearance of the second theme. When instances occur in the master composers, Very rarely, composers began a concluding the result has sometimes been confusion with Type 3 recapitulatory rotation inakey other — regard to where the recapitulation begins. One than the tonic or the subdominant.?5 When this familiar solution is to suppose that the recapit-

22. William Kinderman, Beethoven (Berkeley: Univer- 24. George R. Hill, ed., “Introduction,” Florian Leopold sity of California Press, 1995), pp. 291—92; Joseph Ker- Gassmann, 1729—1774: Seven Symphonies, in The Symman, The Beethoven Quartets (New York: Norton, 1966), phony 1720-1840, ed. Barry S. Brook, Series B, vol. 10

pp. 236-37. Note also that both exposition and reca- (New York: Garland, 1981), p. xix. Cited as examples pitulation become more stressful as they proceed. No were the Symphony in B-flat (Hill Thematic Index No. EEC 1s produced in the exposition; no ESC in the reca- 15), E (No. 63), and C (86). The last of these (from pitulation. Thus this movement features a nonresolving 1769), with a recapitulatory rotation clearly beginning recapitulation, and the elusive goal of tonal closure is in A minor, m. 59, may be consulted in Hill’s edition provided only late in the coda, probably with the I: PAC of seven Gassmann symphonies. The Symphony in C, at m. 289—and even then it barely seems to stay put. H86, appears on pp. 225—53. Cf. Hill’s similar remarks 23. On the surface there might appear to be certain in the New Grove Dictionary of Music and Musicians, 2nd ritornello-like influences at work here. So far as we are ed., 9:565 (originally found as well in the 1st ed.): “in aware, however, no such link between the two prin- a first movement, the recapitulation often occurs in the ciples (sonata and ritornello) — occasionally claimed in subdominant or relative minor.”); and Hill, “The Concasual analysis—has been rigorously demonstrated. On cert Symphonies of Florian Leopold Gassmann” (n. 14 the basis of the evidence available thus far, we are skep- above). tical about any invocation of ritornello (or concerto) principles in Type 1, 2, 3, or 4 sonata forms.

OO Non-Normative Openings of the Recapitulatory Rotation 269

EXAMPLE 12.3a Haydn, String Quartet in B-flat, op. 50 no. 1,1, mm.

1-7

Allegro dolce

| P\—J

— sede P

(“pe 99 @ | ® _@_@ 9 | 9 99 @__| » 9 @—@—

3 3 . i (>, J Jai nfosedijis «#24 pe me MT mm

Yo fa

— — i ee ee ee Se ae ae mf ?

ulation starts only when the tonic is regained, flat at m. 109. As a whole the modulatory mm. even if that occurs mid-phrase, thereby creat- 103—9 represent an expansion of mm. 1—4. The ing a purposeful ambiguity around this impor- twisting mm. 105-8 all flower from m. 3, to tant structural seam. However appealing such a ——swhich they are apparently seeking a clearer tonal

claim might be in combating the “textbook” or correspondence. Metaphorically, the progress of schematic view of sonata form, it ignores the ro- the music 1s stalled at a red light. The suggestion tational status of the moment and almost invari- of an impending move forward is attained in m.

ably misses the main point. An instructive in- 108, “almost” a correspondence measure with stance occurs in the first movement of Haydn’s m. 3, and the green light to proceed is provided Quartet in B-flat, op. 50 no. 1. Example 12.3a m. 109, which even more clearly corresponds shows the opening of P at the beginning of the with m. 4. Mm. 110-15, with small variants, exposition: two bars of tonic B-flat in the cello correspond to mm. 5—10. (Precrux alterations, followed by the famous “concluding”’-formula including an enormous ellipsis, soon follow.)

gesture in mm. 3—6 and a launch of triplets in Some English-language descriptions of m. 7. The development is fully rotational and this music have been predicated on the urge toward its end produces an unexpectedly strong to identify the point of the supposed recapitPAC in vi (G minor) at m. 103 (example 12.3b). ulation— usually a moment of strong articulaThe gestures of the opening of P follow at once tion— only with the arrival of the tonic. Thus (mm. 103-4 = mm. 1-2), for two measures, at Janet M. Levy and W. Dean Sutcliffe, in sep-

least, implying a rotational restart on vi. This arate studies, pointed to the “ambiguity” and merely implied submediant launch proves un- “freedom” of this recapitulation, which 1s atstable. It gives way at once to corrective mod- tained only “at a ‘mid-way point’ of the origiulatory shifts, leading P-material through the nal theme, at about bar 110.” Along the same 8-3 variant of a descending-fifth progression lines, Charles Rosen insisted that “the recapit(mm. 105-9), before tracking into the tonic B- ulation [in op. 50 no. 1/1] enters [in m. 108] 25. Janet M. Levy, “Gesture, Form, and Syntax in ternational Haydn Conference, Washington, D.C., 1975, ed. Haydn’s Music,” in Haydn Studies: Proceedings of the In- Jens Peter Larsen, Howard Serwer, and James Webster

I

ee se gg Example 12.3b_ Haydn, String Quartet in B-flat, op. 50 no. 1,1, mm. 100-12

NR | 2) el (ll crese. Jz i“ a J - . 100 [Allegro]

> A eee eee eee ee eee eee eee ee ee eee eee eee ee ee ee

Ep

fo ee 102 f) — 008 ees aN jog j J fe ed ewe

_ —— oe eet, SS

105» p — / ieee — | |] ee a

Lg} 2 =e eT —@ Fo # ., | @& °+;5>'*8 _, FF gf |

pet fo we tl

f) Oe) oP gjNe » ae O_o tt TY? EEE rr paOoSS ET ne Fy ed . Fy . . . .

5—

Non-Normative Openings of the Recapitulatory Rotation 271 EXAMPLE 12.3b (continued)

Ee CO

Dig, 2 ett fede | FFE mf

“without warning in the middle of a phrase... the vi:PAC is secured in m. 103 appears to leave so that the precise moment of the return to the the recapitulatory onset of P bludgeoned onto tonic is almost unnoticed.”?° But is Haydn’s an ominous, minor-mode vi, somewhat dizzily musical point really one of understatement and seeking its way back home. a sly, mid-phrase settling into a recapitulation? One would normally expect off-tonic reThis seems unlikely. An exclusive reliance on capitulations starting on vi to begin in minor, tonality alone to determine recapitulatory on- but Beethoven provided an engaging variant sets can lead to conclusions that are counterin- in the first movement of his Piano Sonata in

tuitive and countergeneric. F, op. 10 no. 2.2” The sonata’s plein-air, birdFrom the perspective of Sonata Theory what song-bright expositional opening is shown in is happening here seems clear. On rare occa- example 12.4a. The development, shot through sions one finds a development that ends with a with largely minor-mode tonal allusions, frames vi:PAC closely juxtaposed with the normative a stormy central episode (mm. 77—94) with beginning of the recapitulation on I—the “R” C2-based figuration (mm. 67—76, 95-113). M. operation in terms of transformational theory. 107 touches briefly on F minor—the tonic miA few instances were mentioned in chapter 10 nor (example 12.4b)—but by m. 112 the sub(in the section, “Substitutes for V, at the End sequent descending bass slides past the possibilof the Development’’): one was the first move- ity of retaining and further preparing this tonic, ment of another quartet by Haydn, op. 64 no. 3. eventually touching bottom instead on a domThe present, historically prior case, op. 50 no. inant-lock, V{=3 of D minor in mm. 113-17. 1, 1s similar, except that the opening bars of P The development concludes with V/vi, one of themselves— corresponding to mm. 1—41n the the familiar substitutes for the much more comexpositional model—take on the task of modu- mon interrupted V of I. One local implication, lating from vi to I, thus dovetailing a retransi- following the turbulent and minor-mode-sattional function with the beginning of the reca- urated development, is that the piece’s original pitulatory rotation. By the time that the tonic is F major (the sign of the positive) has decayed reached the rotation is already underway. The to one of its negative alter-images, D minor. initiating gesture started in m. 103, which 1s The fermata in m. 117 stands for the dilemma where we must consider the recapitulatory ro- of continuation: are we obliged to begin in D tation (or, less precisely, the recapitulation) to minor (as implied) or can the seemingly lost F have begun. All this, recall, is in response to major be plucked out of the ashes (the generia particularly aggressive concluding half of the cally customary solution)? development, and the sheer force with which In the face of these alternatives Beethoven

(New York: Norton, 1981), pp. 355-62. Levy’s discus- 27. Por a subtle dialogue with the minor-mode subsion is commented upon in W. Dean Sutcliffe, Haydn: mediant recapitulatory entry in Beethoven, see the String Quartets, op. 50 (Cambridge: Cambridge Univer- Presto finale of the Violin Sonata in A Minor, op. 47,

sity Press, 1992), p. 70. “Kreutzer.” 26. Rosen, The Classical Style, rev. ed., p. 124.

pT |eseS eee 272 Elements of Sonata Theory

ley? I OO

EXAMPLE 12.4a Beethoven, Piano Sonata in F, op. 10 no. 2,1, mm. 1—12

Allegro / 3 3 eee ee 9RA, yyrr 4). Sea a A ne ee

fposse im “|e |

ar re t& 6aee ~ —_. se ONA A nc Ch CC (GG Oe

, A= es ee > ee —— oe | ees ll ( e|LLCCCCdEC( CT

eee s

10 Fa a ae

f)CO — A. °hace aa .OY a OO cl A

es??? 7 Sr

staged a tonal surprise by splicing a two-sharp eral bars in this naively radiant D (mm. 118-29, signature into m. 117 and beginning the reca- P!.1 + the more lyrical P!.2), as if nothing were pitulatory rotation in D major, the major sub- in the slightest out of the ordinary. Following a mediant in m. 118. Many implications are folded PAC on D major (m. 129), the ground-currents into this choice. The signature change secks to of the “real” tonic start to assert themselves. A transform the notational and conceptual sense of | measure of pause—another question mark of the tonic altogether. What is produced 1s a dou- continuation— brings us to a distorted and exble-brightness. Not only does the music shift panded restart of P! (mm. 131-36), soon pullfrom the dominant of a minor-mode key into a ing the P-idea back to the proper F major (m. major key but also, from a generic perspective, 133). The escapism of the D-major signature the key selected is not the more conventional is extinguished in m. 136, and the music proF major (I) but an “escapist” D major (VI), a ceeds with a corrected, presumably chastened sharp-grounded key three notches higher on P!-2 (mm. 137ff). The whole passage also sugthe ascending circle of fifths —and also, in this gests resonances with the false-start technique

case, the parallel major of the implied minor (particularly in the “backing-up” recoverykey. This is a false-front mask, a wide-eyed de- effect at mm. 137—44, restating P!.? 1n the ver-

nial of the V/vi problem and the aftershocks of sion first provided in mm. 5—12), but in this the minor-mode agitation of most of the pre- case the recapitulation is not literally rebegun

ceding development. with P!:! tonic-key correspondence measures to The recapitulatory rotation proceeds for sev- mm. 1—4. Here it is preferable to conclude that

ie OO ie, nO 2pT A ee ee es | 2 TT f A Arr abe.

aa nDa pT EXAMPLE 12.4b Beethoven, Piano Sonata in F, op. 10 no. 2, 1, mm.

107-44

[Allegro]

107

A >Feeee 0Eeeeee OO ell Pf OP eee ee eeee ee ee1GO eeAD ee 2 esee ee eeee eei—_ eeel ee2see ee Pfa4G VPA! LES LAE AS! 2ee— --_

i OE A a | 60» Se

he EC > TF bf fa ee D b ban fi fh etee ee TC ee ee ee SQ QO ht (Oe Se er a ee ee Ca.QO GO. PA nles ee cee heea NDneCOED ie OS eeI OO ee i oe ee es rs ee reel es SS 2 QS OR el QOSel COee GOSce OO QT,

a bh, Dn, OD GO Di, OD GO GO 1 _e@ttesi ol » |»! ef el « P| # | £2 | | gf | xf | we] eR I

[fa LA” 6. ——_ 2 ee ee ee ee ee ee ee eee eee

i1l7 — ; —~ po Bt rot 2? Eo FO if, & = [ey =| | Qf CCClCULCUN l lT try’ ew of Te7 feTF Eg es SCE 8fyee | TyeS eS A ~?

Se 139 —— a, ee ose ee ee A a SS a yeee A eeee ————__. SS OS _, -b-~ BD OT TT TTT eKHeIO.'N"VN’NYH-WNMM__. _ TTT > OOOOO—F—DMT ____IT1.],,_—0_0.@@7TNNWwTAFv§q7w—--.0._ | —>—_——T_s«“_-_=s2z2”09”"9"9w0w0--00O0w0WT

ee Fg SSee Uc

FS CL aSOSC SQ A OO eee > t=

f Jie —oerae Fo or rer ._ >—n__] Bn Aieee| eS ~ GQ CC SP Fee esSNNeeeSOOo Or

Non-Normative Openings of the Recapitulatory Rotation 275

the recapitulation itself begins in VI, m. 118, immediately succeeding recapitulation reverses

and self-corrects en route.?8 this course, moving from V (for P) to I (for S An apparently later variant of the subme- and C), and in which there is no significant diant recapitulation is one in which the reca- developmental expansion within the recapitupitulatory rotation is launched on bVI before latory rotation. (If there were —for instance, adjusting itself into the tonic. A classic exam- if P!-2 or TR! were enlarged into a developple occurs in the B-flat major second move- ment— we would classify it as a Type 2 sonata.) ment, Allegretto vivace e sempre scherzando, Early instances of the structure are provided by of Beethoven’s Quartet in F, op. 59 no. 1. In a few late-eighteenth-century opera overtures, this deformational scherzo-sonata— nightmar- such as that to Gluck’s Alceste and some of the ishly distorted in tone and structure —the ro- operas of Salieri.2? The young Schubert, a pupil tation in question begins on G-flat in m. 239. of Salieri, occasionally made use of this variant, (Mm. 239-53, which include ascending-fifth as in the first and last movements of his Quartet shifts through D-flat and A-flat at mm. 246 and “No. 6” in D, D. 74 (in the first movement the 253, are largely referential to the material first recapitulation outlines a V—IV—I plan!), and in stated in mm. 1—16.) The tonic B-flat major 1s the tonally unusual finale of the Piano Sonata reattained only with the arrival of P!-? at m. 259 in A Minor, D. 537 (whose exposition moves

(= m. 23). i-IV—V, complemented in the immediate recapitulation by v—VII-—I). Doubtless related to

. . a this way of thinking 1s the finale of his Piano

Recapitulatory Rotations That Begin in V Quintet in A, D. 667, “Trout,” in which the

From time to time one encounters a recapitu- exposition moves from I to IV (!) and a literlatory rotation that begins in the dominant, a ally transposed recapitulation follows directly, tonality usually associated with S and C in the V to 1.38 exposition. In its simplest manifestations a dom- When a development section is present—as inant recapitulation can be a variant of the Type ina Type 3 sonata—the situation becomes more 1 sonata (lacking repeat signs and a development) complex. In the first place, the normative, tonic

in which the recapitulatory P begins directly in opening of the recapitulatory rotation is preand continues at length in V, not in I. (Such a ceded by a substantial dominant prolongation, variant is counterdefinitional to our view of the often a structural-dominant lock leading to the normative Type 1 sonata, which we regard as characteristic harmonic interruption before the identifying itself with a tonic return to P directly onset of the recapitulation. To begin that rotaafter a nonrepeated exposition. See chapter 16.) tion on (or in) the dominant is to superimpose This produces a bi-rotational scheme in which a musical procedures that are normally kept sepa-

two-part exposition moves from I to V and the rate. The recapitulatory P! (or P !!) is called

28. The wrong-key start to the recapitulatory P in the Beethoven,” pp. 175—76, noted Schubert’s indebtedness

op. 10 no. 2 may be intended to recall the wrong-key to Salieri and Cherubini in the forms of some of his onset of the recapitulatory S (on F major, IV) in the early works and identified Gluck and Salieri as overture first movement of the C-minor sonata, op. 10 no. 1. composers who sometimes omit the development and Both moments suggest an effect of escapism and generic follow the exposition with “recapitulations [that] do not irresponsibility. The pairing of the two works in this [always] begin in the tonic key.” By way of an explana-

respect may be complemented by the first movement tion Chusid cited “the older bipartite sonata form” as of the D-major sonata, op. 10 no. 3, featuring a ton- “the form... of a majority of opera overtures of the ally errant TM! in both the exposition (m. 23) and the time” and “fa most important [principle] for Schubert,” recapitulation (m. 205). (See the discussion of op. 10 but his account does not distinguish between what we

no. 3 in ch. 11.) would call the normative Type 1 sonata, this Type 1

29. Tobel, Die Formenwelt, p. 177, mentioned Alceste sonata variant, anda Type 2 sonata. as “an early example” of die dominantische Reprise be- 30. Cf. the similar situation in the finale of Schubert’s

fore moving on to Beethoven and Schubert. Martin Quartet in B-flat, D. 36, also noted in Rosen, Sonata Chusid, “Schubert’s Chamber Music: before and after Forms, rev. ed., p. 359.

276 Elements of Sonata Theory

upon to appear simultaneously with tonal resi- structure or event appears to be in dialogue with

dues of its own dominant preparation. several different generic possibilities, poised To be sure, it occasionally happens that oth- tantalizingly among them. Haydn, in particuerwise tonic-key recapitulations are ushered in lar, enjoyed teasing out and dwelling within above dominant pedal-points, as famously in these ambiguities. The first movement, Vivace the first movement of Beethoven’s Piano Sonata assal, of his Quartet in D, op. 33 no. 6, proin F Minor, op. 57, “Appassionata” (mm. 134ff). vides a challenging case that proves instructive. In these cases the normal point of harmonic in- The expositional layout may be schematized as

terruption on V at the end of the development follows: P! (mm. 1-4); varied repetition of P! is kept audibly open into recapitulatory space. (mm. 5—8); P? (mm. 9-18); nonmodulatory Consequently, the fresh rebeginning sets forth TR (mm. 19-26), leading to light, first MCin tandem with a sometimes-suspenseful pro- effect (I: HC, m. 26); TM! (m. 27-31), merging longation of what “in better circumstances” it into TM? (mm. 31—34) and leading to a secwould have left behind. As is the case with the ond apparent MC (V:HC, m. 34); TM? (mm. type of S° or S!-° themes that unfold over dom- 35-43, with a V:PAC in m. 43, probably not to inant pedals, the altered P-passage has a dou- be taken as the EEC); S-space extension based ble-function: the retention of the last element on TM? figures (mm. 44—49) with EEC at m. of an earlier preparatory situation coupled with 49; RT (mm. 50-58).

the beginning of something new. At this point Haydn provides us with a brief, The same psychology is pushed further P!-based development (mm. 59-70), starting when the P-idea is not merely a tonic module on F major and running aground with a ferover a dominant pedal but a theme beginning mata-held E-major chord, V of A minor. The in (or on) the dominant key itself. Here again fermata on a strongly implicative dominant can we find the splaying of two ideas normally kept suggest the end of a development section, alseparate — dominant preparation and thematic though in this instance the development seems return— only in a more extreme format. Be- to have closed down too early. What follows is cause such situations are exceptional to norma- a near-literal statement of the first P! on A mative practice, we consider them to be recapitu- _—_—sijor (that is, on V, mm. 71-74, corresponding

latory deformations. The dominant onsets of | to mm. 1—4)—replicating the initial texture these recapitulatory rotations fold into them- of the exposition— followed by three interselves aspects of unfinished tonal business from polated bars (prematurely introducing P?) that the developmental space. The recapitulatory shift back to the tonic for the varied repetition rotation does indeed begin, although it does so of P! in I, mm. 78—81 (corresponding to mm. prematurely, without the usual, fuller harmonic 5—8) and the first three bars of the subsequent preparation. It sometimes happens that a brief P?, mm. 82—84 (corresponding to mm. 9-11). patch of Fortspinnung 1s applied after P! in or- Notwithstanding the interpolated mm. 75-77, der to bring the ongoing rotation back onto the this is an impressive stretch of correspondence tonic track. Were this fleetingly renewed devel- measures, and following the twelve bars of deopmental texture to be succeeded by a return velopment and strong fermata one might wonto a tonic P! (and thus a new rotational begin- der whether something recapitulatory might ning), the dominant-key P should be considered be underway. And yet one must be cautious. within the general category of the false-recapit- We also recall (chapter 10, on varying degrees ulation effect—along with all the conceptual of the false-recapitulation effect) that earlyproblems that come with that category (chapter developmental dominant and tonic statements 10). But since it empties out instead on a tonic of P! are normative within the style and are usuP!2, P?, or TR, thereby continuing the rota- ally not recapitulatory at all—even though this tion, it is preferable to construe the whole as a one had been preceded by twelve bars of tonrecapitulatory rotation that begins on V. ally shifting development and a fermata. We are These considerations can open compositions thus situated between two possibilities, perhaps to interpretational ambiguities, in which a given tilting—so far—more toward the latter. Any

Non-Normative Openings of the Recapitulatory Rotation 277

further interpretation will depend on the musi- capitulatory practice. It is imprudent to make cal continuation. Will it proceed in the manner a decision on behalf of only one of them, a de-

of an ordered rotation? cision that would overlook the purposeful amIn fact, it does. The rotation begun with cor- biguity of the compositional situation. Further, respondence measures in V at m. 71 proceeds if the most obviously developmental bars, mm. in rigorous order—with some developmental 59—70, were not there at all, we would be lookexpansions—all the way to the coda. Having —_ ing at an example of a Type 2 sonata with early heard the two versions of P! and the onset of — crux. Thus Haydn provided his audience with a P? (and having recovered the D-major tonic), witty work cleverly suspended in the force fields Haydn now provides a Fortspinnung (develop- of at least three formal categories without demental) expansion of P?, touching on the Nea- claring definitively on behalf of any of them. politan E-flat (mm. 86—90), moving through a The structure 1s in dialogue with more than one series of shifting tonal levels (mm. 90—99), and hermeneutic structural type, caught in a web of finally locking back onto correspondence mea- differing interpretive possibilities.

sures with a first crux (mm. 100-4 = 14-18, Simpler, more schematic examples of recathe end of P?). As expected, TR arrives next pitulatory rotations that begin in V following (mm. 105-8 = 19-22), followed by more re- a development may be found in the first movecomposed material, designed to iron out the first ments of Clementi’s Piano Sonata in F Minor, apparent MC, TM!, and TM? (mm. 109-25). op. 13 no. 6, and Schubert’s Symphony No. 4 The “real” MC 1s now altered to a I:PAC (m. in C Minor, D. 417.3! Both composers were at125), and the original TM? and its extension tracted to unorthodox, sometimes flagrantly follow (mm. 126—44—now reinterpreted as S). transgressive, tonal layouts in their sonata forms, The beginning of this S marks the moment of — and in this case the structures of the two pieces

a second crux and set of correspondence mea- are (coincidentally?) similar. In the Clementi sures, with mm. 138—42 expanding the origi- movement one finds a C-minor recapitulatory nal single bar, m. 47, to provide the ESC at m. onset, m. 70, with a subsequent S beginning 144. An RT and first ending follow. Everything fleetingly in the “wrong key,” II—A-flat, m. from m. 71 onward has been strictly rotational, 84 (the same key in which it had appeared in the with individual modules of the rotation devel- exposition) — before falling into the fatalistic F

opmentally expanded here and there. minor. Schubert’s symphony-movement has a From a resolutely rotational perspective this more unusual key-plan for the exposition: its is an instance of a recapitulatory rotation that S is produced entirely in VI, A-flat (two closed begins on V and shortly thereafter modulates to statements and an expanded continuation, mm. I and includes some internal Fortspinnung jags, 67—76, 76—85, 85-130, the whole forming an perhaps to compensate (as impulses of develop- extremely large sentence). The recapitulation ment still lingering in the air?) for the unusu- begins in v, G minor, m. 177 (the P-idea seems ally brief development and premature entrance also to allude to the opening of Beethoven’s

of P!. Yet there 1s no denying that in mm. Quartet in C minor, op. 18 no. 4), and the ini71-81 Haydn was also playing on the tradition tial two statements of S (mm. 214—23, 223-32) of presenting P in the dominant and in the tonic are sounded, each with a closing PAC, in VI, Eearly on in a development. In part this implica- flat. Only with S’s expanded continuation, mm. tion 1s made possible by the double-sounding 232—68, is the tonic, C minor, regained to proof P!—tonic-tonic in the exposition, mm. 1—4 duce the proper ESC. The thematic segments and 5—8, but dominant-tonic in mm. 71—74 of the broader S—the initial statements— are

and 78-81. In mm. 71-81 Haydn conflates never sounded in the tonic. Because of this they two categories: Type 3 developmental and re- sienify a tonal alienation of this portion of S,

31. Another example may be found in the first movement of Schubert’s Quartet “No. 4” in C, D. 46.

278 Elements of Sonata Theory

demonstrating that certain features of post-MC TR). In Leonore No. 3, the tonic, C major, 1s space are forever nonassimilable into the tonic. regained only at the TR point, m. 378. And this They remain irrecoverably alienated from tonal —_is where most commentators, sometimes with resolution. Tonal alienation of this sort may be ingenious arguments, have placed the point of

found in many of Schubert’s pieces. recapitulation.°? Surely the most celebrated instance of a reca- The situation is more complex than is usually pitulatory rotation beginning in the dominant acknowledged. Merely to locate the recapitulaoccurs in Beethoven’s C-major Overture, Le- tion at the C-major passage in m. 378 —tonally onore No. 3. Following the trumpet-call break- the obvious place— overlooks the earlier (varthrough-interpolation into the development (m. ied) beginning of the recapitulatory rotation, 272 — thus forestalling the calamitous C-minor itself prepared by its own structural dominant recapitulation predicted), the development re- lock. The earlier passage has been marginalized settles onto what 1s clearly to be taken for the both because it is sounded in the dominant, generic re-entry dominant lock to the recapitu- not in the tonic, and because of its TR-version lation (m. 318)— only here on the wrong dom1- references. Here again is a situation where it nant, V of V (with D in the bass), not V of I. is useful to distinguish between the onset of a This leads to a G-major, emphatically thematic rhetorical recapitulation or recapitulatory rotastatement of a variant of P! in the flute (mm. tion (beginning in G in m. 330) and the later, 330ff) that, despite its obviously ‘TR-based me- emphatic return to the tonic (beginning with lodic continuation (mm. 334-38), must have TR in C in m. 378). In this case Beethoven was been intended to provide the rhetorical impres- doubtless expanding the idea of recapitulations sion of a recapitulation.*? Still, occurring in V, that begin over the dominant to suggest that in it is generically in the wrong key, which may this programmatic overture the sheer process of help to explain its “inability” to recover the ex- reattaining the tonic in the recapitulation— on act melodic contour of the original P-idea. (In the way to the ESC —1s going to be an uncomother words, its provisional status is highlighted monly arduous enterprise. through its more evident congruence with the A simular situation, perhaps modeled on LeTR version of the melody.) Over the course of —— onore No. 3, occurs in the first movement of the next several bars the flute melody, in dia- Berlioz’s Symphonie fantastique (C major). Genlogue with the bassoon, seeks a PAC in G major, eralizing about such an unusual composition 1s but at the last moment (m. 352) that cadence 1s a perilous procedure, but one productive way undermined and proceeds instead to an extended to consider this piece 1s to examine its rotacrescendo and decisive return to V’ of the gen- tional structure, a guiding thread through the erally proper key, C major (mm. 371-77). This purposely garish deformations. After a repeated crescendo corresponds gesturally to that located exposition and a substantial stretch of developin the later stages of P in the exposition (mm. ment one comes across a much-noted complete 49—68)—the sense of a newly begun rotation statement of the P-theme—the idée fixe—in V, is continuing—and as in the exposition, it dis- G major (m. 239). This G-major full statement charges its energy, sempre ff, onto an emphatic marks the beginning of another, varied rotation TR of the tutti affirmation type, now in C ma- of expositional materials (with alterations and jor (m. 378; cf. m. 69, the exposition’s onset of | expansions), and this rotation will lead mid-

32. This was also the view of Tobel, Die Formenwelt, p. room to grow; and so he continues his development at 177, who designated this moment as an example of the leisure, with a sunshiny passage in which the flute and

eroup of “dominant reprises.” bassoon give in G major the substance of the tutti that 33. Asin Tovey’s famous analysis, reprinted in Sym- followed the first subject. . . . This is the sublime and phonies and Other Orchestral Works (Oxford: Oxford unexpected use of the dominant to which I referred in University Press, 1989), p. 138: “We are now begin- connexion with the development of Leonora No. 2... . ning to learn a lesson in proportion. . . . [Because of his [The] fortissimo in the tonic .. . does duty for the recacompressions of Leonora No. 2, Beethoven] has thus left pitulation of the first subject.”

Non-Normative Openings of the Recapitulatory Rotation 279

course to C major and the tonal resolution of — (only) of P float past, minor-inflected, modulathe rotational layout. From this perspective we tory adjustments in mm. 79-82 bring us back might consider the G-major idée fixe as begin- to the “reality” of the tonic and achingly familning a bizarre rhetorical recapitulation in the lar correspondence measures, at first mistily in dominant. The final, riotous appearance of the minor, then clarifying into the major at m. 88. idée fixe in C major (m. 410) belongs more prop- After a provisional ESC in m. 93 the suppressed

erly to coda-space than to recapitulatory space, continuation of P (= mm. 30-35) returns in as has sometimes been claimed. Overlooking mm. 94—99 with the flavor of a benediction for the rotational aspects of the composition has led the whole movement.

commentators (beginning with Schumann) to In the Schubert “Great C-Major” finale the consider the movement to be most fundamen- development ends with a long, 84-bar dominant

tally arrayed as a symmetrical arch. pedal, V of I, beginning in m. 515. The first portion of this dominant-lock implies C ma-

Recapitulatory Rotations That Begin in Jor, a 91552, with a brief intermixture of

Other Keys C minor in mm. 531-37. Before long this expectant passage darkens to V of C minor more

Although nontonic recapitulatory launches permanently (mm. 553ff) and the texture thins happen most often in IV, VI, or V, it is pos- out to a single G4 in octaves (especially by mm. sible to come across more deformational, ad hoc 576—82), with continued anticipations of P jutnontonic choices. One of the strangest may be ting upward in the strings. A passing F4 in the found in the first movement of Clementi’s Piano bassoons and trombones (mm. 583—90) resolves Sonata in F, op. 13 no. 5: prompted by some downward in the succeeding bars and alters the tonal veerings in the development, its recapitu- sonority to bare, throbbing G/E}> dyad in mm. lation begins in E-flat, bVII, m. 63, and P is ar- 591-98. We are still hearing an incomplete V of ticulated fully in that key, even cadencing in it, C minor, but the lone dyad is tonally ambigu-

m. 70, before returning to F major. ous (5 and 3 of C minor or 3 and 7 of E-flat Only slightly less odd are recapitulations that major?) and in its insistence seems to be tipping begin in bIII. These are rare, but examples may in the direction of a potential E-flat major. At be found in two of the most well-known pieces m. 599 the full orchestra picks up the implicain the repertory: the F-major slow movement tion and begins the recapitulation vigorously in of Mozart’s Piano Concerto in C, K. 467; and E-flat, bIII of the tonic C. Here Schubert was finale of Schubert’s Symphony No. 9 in C, “The probably recalling one alternative treatment of Great,” D. 944. In the Mozart, after an initial the development-recapitulation seam, the move orchestral tutti in F (mm. 1—22 with an ap- from V/vi to I, and reinterpreted it unconvenpended bar of fill), the “expanded” exposition tionally as V/i leaping away to bIII, producing a proper, with soloist, begins in m. 23. Anew TR remarkable color-shift by pivoting on the comfollows, m. 35, leading to a very light V:HC mon tones G and Eb. MC -effect in m. 44 with a sustained dominant-

lock replicating material from the initial tutti Double-Recapitulation Effects? (minor giving way to major at the end) for the

next five bars—although the exposition could One of the strangest, and rarest, deformations also be understood as continuous. The EEC in occurs when something that starts out, postdeC major occurs atm. 55, and itissucceeded bya — velopmentally, as an apparently recapitulatory brief development closing on a languidly melan- rotation fails to accomplish its tonal mission and choly V of F minor, mm. 71. As if turning away is succeeded by what amounts to a fully recomfrom the negative implications of the minor | posed—and now more successful— “second retonic, the music “escapes” in a bar of linkage to capitulation,” rebeginning with P and proceedIll of F minor, A-flat major (bILI of F major). It ing all the way to S and C. Here the rotational is in that dream-like key that the recapitulation aspect underlying sonata form takes precedence begins, m. 73. Once the presentation modules over normative tonal expectations, although it 1s

280 Elements of Sonata Theory

also possible that any such second-recapitulation capitulation?), with a more clearly identifiable effect might allude generically to the possibil- restart of P (m. 195) and the remaining later ity of a second visiting of a recapitulation in the portion (only) of S (m. 228 = m. 63), now both

once-standard full repetition of the develop- in the proper tonic, G minor. The broad ex-

ment and recapitulation. panse of the original S-material is split between One prototype for this oddity occurs in the the two recapitulatory rotations, the “wrong” first movement of Clementi’s Piano Sonata in one and the “right” one.*4

G Minor, op. 34 no. 2 (1795). There are many The oddity of the Clementi piece would curious features in this movement, one of which perhaps not be worth remarking upon were a is the varied, major-mode return of the open- related strategy not employed—famously —in ing Largo at the end of the development, mm. one of Beethoven’s late quartets, the first move-

126-35. (See also the section of chapter 10, ment of the A-minor Quartet, op. 132. Here “Introductory Material in Developments.’) too we find what amounts to a tonally “wrong” What one initially presumes is the recapitula- recapitulatory rotation (mm. 121—94, begintion begins in IV, C minor, in m. 143. S!-° fol- ning in E minor, v, and moving to C major, lows in m. 162, but in the “wrong key,” VI, E- III) followed by a notably varied, “right” one flat major, and proceeds to stay there to sound (mm. 195—264) in the tonic.*> Needless to say, the first portions (only) of a multimodular S. the complications surrounding this composition Since this supposed recapitulation is in the pro- are profound—though not unique, since cercess of misfiring, Clementi sets things right by tain aspects of its procedures are foreshadowed dissolving the “flawed” rotation, mm. 183-94 in some of his earlier works.%° and giving us a corrected version (the “real” re-

34. Plantinga, Clementi, p. 173. leads to a first recapitulation, mm. 162—266 (notice the 35. The description here is similar to that in Rosen, false start in m. 152), in which P and S are brought back Sonata Forms, rev. ed., p. 355. The accompanying foot- in the proper, tonic key (S begins in D-flat, then adnote, however, citing movements in Haydn’s Sympho- justs itself to F), although no ESC 1s attained, paralleling nies Nos. 75 and 89, is questionable. What Rosen was the exposition’s lack of an EEC. This first nonresolving noticing in those cases were engaging adaptations of recapitulation immediately recycles back to a substantwo common procedures in Haydn sonata forms: ro- tially recomposed revisiting of the entire developmenttational developments and recomposed recapitulations. recapitulation complex (a composed-out substitute for For other discussions of the double-recapitulation effect the once-conventional repetition of the complex), mm. in op. 132/i and additional bibliography see Kerman, 267-355 (with a different false start in m. 346) and The Beethoven Quartets, pp. 247-50; Robin Wallace, 356-437. In the “second recapitulation” S is contained “Background and Expression 1n the First Movement of fully within the tonic, but once again the ESC 1s elusive, Beethoven’s op. 132,” The Journal of Musicology , 7 (1989), and F-major closure is obtained only in the coda, which

3-20; and Michael Steinberg, “The Late Quartets,” in begins in m. 439. Overlooking the obvious, Tovey’s The Beethoven Quartet Companion, ed. Robert Winter much-repeated claim (Symphonies and Other Orchestral and Robert Martin (Berkeley: University of California Works, p. 82) that mm. 267ff constitute “a coda that Press, 1994), p. 268. Cf. Kinderman, Beethoven , p. 292: is nearly as long as the whole body of the movement,” “The first movement of the A minor Quartet op. 132... is indefensible. One much more compressed precedent contains a recapitulation in the dominant, as does the for the Eighth’s written-out and recomposed develop-

Credo of the Missa Solemnis.” ment and recapitulation is the discursive “coda” —its 36. Op. 132/1 should not be considered apart from the compression makes it easier to defend as a coda—to the perhaps-related precedent of the wittily (and wildly) Symphony No. 3 in E-flat, op. 55, “Eroica,” mentioned “mechanistic” finale to Symphony No. 8 in F, op. 93. in this regard at the end of ch. 5 and in ch. 13. Cf. also Here the development, mm. 91—161 (itself beginning the related precedent in the finale of the Piano Trio in with P in the tonic after a nonrepeated exposition, re- E-flat, op. 70 no. 2 (two trackings through the recapitucalling the procedure in the Quartet in F, op. 59 no. 1), latory S and C), discussed in ch. 11, n. 20.

CHAPTER THIRTEEN © "ISLE 0

Parageneric Spaces Coda and Introduction

B’ ulated sonata-space we mean that space artic- The Coda by the generic sonata form proper:

normal treatments of the exposition, develop- , ,

mental space, and recapitulatory rotation. Some Definitions, Traditional Views sonata movements also feature parageneric = Codas may be conceptualized either rhetorically

spaces (or not-sonata-space), everything else (with regard to their positional relation to the in the movement that may set up, momentarily preceding thematic layout) or tonally (with restep outside of, or otherwise alter or frame the gard only to keys and tonal-contrapuntal “backpresentation of the sonata form. In such move- sround” resolutions, disregarding the thematic ments the most frequently encountered para- _ parallels of the recapitulation with the exposigeneric spaces are accretions that in the second tional Anlage).! Sonata Theory favors the rhehalf of the eighteenth century came to be in- torical approach for its definition of a coda. As creasingly attractive options as add-ons to the a rule of thumb the coda begins once the recabasic structure. The most common are codas __ pitulation has reached the point at which the and introductions— the subjects of this chapter. exposition’s closing materials, normally includOther, historically later parageneric spaces in- ing a final cadence, have been revisited in full.

clude interpolations within the movement that In analytical work one should identify the withdraw from the sonata-action, such as some referential or correspondence measures in the recurrences of a slow introduction (which re- recapitulation that recapture the way in which currences, however, might well be implicated the exposition had ended. In most cases once in larger rotational structures that stretch over we are past the point where the last exposi-

both sonata and not-sonata-space.) tional measure has been retraced in the reca1. Esther Cavett-Dunsby, for instance (“Mozart’s Co- the movement [what follows after ‘the arrival of the 1’], das,” Music Analysis 7 (1988), 31-51), sought to dis- rather than to its surface form. . . . Schenker’s concept tinguish between the “formal” and “structural” coda. of the ‘structural coda’ (as I shall call it) is a foil to the The formal coda begins “after double barlines and re- more familiar notion of the ‘formal coda.’ As a rule, peat marks towards the end of the movement” (p. 32). structural and formal codas in Mozart’s sonata forms do The structural coda is defined in explicitly Schenkerian not coincide” (p. 34). terms, “with reference to the background structure of 281

282 Elements of Sonata Theory

pitulation— assuming an otherwise straightfor- brief, postphrase module) 1n its own right. ward situation— we have moved into a coda. Its The coda is a parageneric space that stands specific treatment may be marked by localized outside the sonata form. It is what Schoenberg idiosyncrasies. It might happen that as the reca- famously — perhaps wryly — called “an extrinpitulation comes to its expected close one finds sic addition. The assumption that it serves to a last-instant deviation from a strict correspon- establish the tonality is hardly justified; 1t could dence with the end of the exposition: a shy- scarcely compensate for failure to establish the ing-away from the anticipated final cadence or tonality in the previous sections. In fact, it would some other alteration or expansion of it. This re- be difficult to give any other reason for the adcomposed recapitulatory conclusion might even dition of a coda than that the composer wants to pause on an unexpected chord—a dominant, say something more.’* Although its length may a diminished seventh, an applied dominant, or vary, shorter codas were the norm before longer some other harmonic turn—or it might merge ones began to appear. Sometimes codas are little into a transitional passage preparing for the coda more than emphatic, tonic-prolongational tags proper. One might also find smoothing or blur- of one, two, three, or four measures, following ring features peculiar to the area surrounding the conclusion of the recapitulation’s last phrase. the introduction of coda material.? One’s ana- In Classical Form Caplin, on the basis of formal lytical apparatus should be flexible enough to function, regards such a brief tag as a codetta

handle these variants with ease. rather than a coda, reserving the latter term for The situation may be complicated by the a longer, “relatively large unit.”> Our preference presence of repeat signs. When a repetition of 1s to use the term “codetta” to refer only to a the development and recapitulation is called for, final subsection housed within a larger formal as Rosen noted, a coda could either be included zone, such as S- or C-space. (One variety of C, within the repeat signs of the second part of the for example, 1s the “codetta type.) Under this sonata or it could be appended after that repeti- definition a codetta could not stand alone betion.* In other words, reckoning from the be- yond sonata-space (beyond the recapitulation’s sinning of the recapitulation, one might find articulation of the measure corresponding to the

{...PTR’S/Ci coda}, {...P TR S/ last bar of the exposition). In most cases our tenC coda ‘]?, or something like {...P TR’ S/ dency is to regard this separate section, short or C! C2—coda-like material— C3 |}. This third long, as a coda. possibility may or may not lead to a repeat sign Insisting on hairsplitting terminological disafter the last C-module returns. To distinguish tinctions is rarely relevant to the larger tasks this last procedure from a coda proper we call it of analytical hermeneutics. We see room for coda-rhetoric interpolation (CRI): it will be treated flexibility here. We admit the possibility, and at in a separate section below. Finally, one can dis- times even the desirability, of interpreting certinguish all of these options from another com- tain instances of a brief “extra” bar or two as bemon practice of the time, the expansion of a final ing a mere “coda-effect” broadening of the final or penultimate C-module by several bars in such chord of the concluding recapitulatory measure. a way as to suggest a wrap-up or coda-effect. This might be the more appealing option when Such a C-module expansion with coda-effect those bars articulate only the tonic chord and

(or with CRI-effect) is an interior broadening are not set off from the normatively final reof an existing phrase, not a separate phrase (or capitulatory bar by a rest or other break. We 2. A similar point was made in Caplin, Classical Form, Beethoven’s Egmont Overture): see Hepokoski, “Back

p. 181. and Porth from Egmont.” The argument of the coda ex3. Rosen, Sonata Forms, rev. ed., p. 297. isting “after-the-end” is also pursued in Caplin, Clas4. Schoenberg, Fundamentals of Musical Composition, p. sical Form, pp. 179—91—which also began by quoting

185. Schoenberg’s dismissal of a tonally compensatory Schoenberg (as, indeed, did Cavett-Dunsby, “Mozart’s coda is incorrect in the case of such sonata deformations Codas,” p. 32). as the nonresolving recapitulation (as in, for instance, 5. Caplin, Classical Music, p. 179.

Parageneric Spaces 283

also agree that many short codas have a codet- applicable to extended codas, Rosen noted that ta-function to the entire movement— or to the “the appearance of a coda always disturbs the

recapitulation proper. binary symmetry of a sonata form. ... One The more elaborate the coda, the higher a might say that the coda 1s a sign of dissatisfaccomposition’s claim to an enhanced prestige, to tion with the form, a declaration in each ina heightened “weight and seriousness.’’® In the dividual case that the symmetry is inadequate later-eighteenth and early-nineteenth centuries to the demands of the material, that the simple the coda normally accomplished the standard parallelism has become constraining.”® This functions of grounding further the secured ter- observation poses larger interpretive questions. ritory of the tonic and ending the movement In what respect would the preceding sonata be with an appropriate rhetorical flourish. Since the considered “inadequate” were it to have been full presence of the tonic key had been precip1- presented without a substantial coda? Why is tated or “made real” only at the moment of the the coda responding to it at such length? ESC (as discussed in the concluding portions of

chapter 1), the coda may also be understood as Thematic-Rhetorical Material in Codas:

. , oo. . Rotational Implications

confirming the reality of the fully secured ton- . . ic — celebrating it or basking in it—that it had

taken the exertion of the sonata process to ac- Rhetorically, there were two standard eighcomplish. This interpretation seems particularly teenth-century options. The composer could apt in longer, more discursive codas. Codas re- provide a coda that completes the movement spond to the preceding kinetic thrust of the so- briefly, and forte, with generic concluding modnata. As Ratner put it, one of the coda’s purposes ules (Ratner: “a few emphatic chords or caden-

is to provide “a stronger effect of closure... tial gestures”’).? Or the composer could begin to arrest the momentum generated throughout the coda with a restatement of the primary the movement,”’ even though some composers, theme (P) or an obvious adaptation thereof.!” in particularly “grand” compositions, occasion- So common is this practice that the reintroducally treated the coda as an extended postsonata tion of P-material at the end of the recapitula-

space in its own right. tion (proposing a sense of rebeginning, not a

Codas could be treated freely because of their mere P-based C) is a strong sign that the coda separateness from sonata conventions. The 1m- has begun. With the exception of the final “ronplications of this can be provocative. The mere do-theme” appearance of P in a Type 4 sonata, existence of a coda—especially one of greater any P following tonally resolved S- and C-zones length— can provide a challenge to the preced- signifies the onset of a coda. ing sonata, as though the normative bringing of Codas that begin with P-based music suggest sonata-space to completion at the end of the re- the onset of another rotation of the referential capitulation were being arraigned as insufficient materials. This final return-to-P may correto the expressive task at hand. In a passage most spond with the earlier move from the end of the

6. Rosen, Sonata Forms, rev. ed., p. 298, suggesting also 7. Ratner, Classic Music, p. 230. that it was in Haydn’s Quartets, op. 9, that codas were 8. Rosen, Sonata Forms, rev. ed., p. 297. initially attached to significant sonata-form structures. 9. Ratner, Classic Music, p. 230. Again, cf. Caplin’s difRosen, pp. 304—5, notes additionally the added “dig- fering terminology cited in n. 5. nity” given to a composition by means of a coda as well 10. The P-launched coda was also mentioned, particas the responsiveness of the coda to the content that had ularly with reference to Haydn and Clementi, as the preceded it in the sonata proper. Ratner, Classic Music, most common procedure in Rosen, Sonata Forms, rev. p. 231, made essentially the same point about the coda, ed., p. 311. Kerman suggested some caveats about the citing also Koch’s 1802 definition of it as “a more com- return-to-P procedure in Mozart (see n. 11), although

plete closing section, following the second reprise of such returns do occur here and there in the works of an allegro,” along with Reicha’s 1813 comparison of a that composer (for example, in several of the earlier broad coda—as in some of Beethoven’s music—as “an symphonies). oratorical peroration.”

284 Elements of Sonata Theory

exposition into the development, which often Ratner and Kerman that it 1s inappropriate to begins with P-based gestures. Complementa- consider such lengthy codas as “second” or “‘terrily, if the development had begun instead with minal developments” (the latter being the term new or episodic material— sometimes either as proposed, it seems, by Vincent d’Indy).!%

a substitute for the more normal P or an inter- When a coda is lengthy, we refer to it as a polation before the treatment of the P-theme discursive coda. This term conveys the sense that proper—the coda might start with a reference it unfolds a separate, often multisectional disto that idea, not to P. The recycling back to course beyond sonata space. While one’s first an idea from the opening of the development instincts might be to point toward discursive retains the underlying impulses of rotation, po- codas in some of Beethoven’s stormy Allegro tential repeat schemes, and the like.!! The coda compositions, we should recall that he was also can often be understood as an incomplete co- attracted, especially in the 1790s (as Plantinga da-rotation, based only on P or a proxy for it. pointed out), to appending lengthy extensions In most cases the rotation 1s soon stopped short to slow movements. One finds them in the P1with generically cadential (or other) modules ano Trios op. 1 no. 1/i1 and op. 1 no. 2/11, in that call an end to the rotational cycling. Lon- the Piano Sonatas op. 2 no. 1/11, op. 2 no. 3/11, ger, more discursive codas may have more elab- op. 7/11, and op. 10 no. 1/11; in the Piano Con-

orate rotational implications. certos No. 1, op. 15/u, and No. 2, op. 19; and Needless to say—to consider other thematic in several other compositions.!4+ A discursive options— one should take note of any coda that coda is a separate tableau, a surplus-conclusion begins with (or emphasizes) non-P material or after the main event. Especially in Beethoven, of any coda that introduces new themes or tex- it may momentarily “lose” the tonic secured at tures. When such things occur, we should in- the ESC by slipping into nontonic keys before quire why they are as they are and what effect regaining the tonic near the end. It is always of they have on the whole. One might also keep in central interest to come to terms with why this mind Kerman’s delicious phrase regarding one type of coda was added at all. characteristic type of later-Mozart coda: “It 1s While some longer codas are largely P-based as though after having left the party Mozart has (representing only an incomplete rotation), a final remark to make sur l’escalier—a grieving one occasionally finds cases in which they give remark or a wary one, a witticism, a complh- the impression of producing another full rota-

ment, or a retort.” !2 tion through the materials: P and S or P and C.

Classic instances occur in the first movements of

Discursive Codas Beethoven’s Piano Sonatas in C, op. 53, “Wald-

stein” (the coda begins with P on bII, m. 249: Although there are a handful of celebrated ex- S appears in the tonic, m. 284) and in F minor, ceptions, a coda in Mozart or Haydn 1s usually op. 57, “Appassionata” (the coda begins with P a brief or relatively modest affair. On the other in m. 204, soon shifting away from the tonic; hand, Beethoven often expanded his codas S appears on VI at m. 211). Since both sonata vastly and treated their materials with consider- movements also feature rotational developments able complexity, although we agree with both (in the “Waldstein” it is S? that appears there, 11. Cf. Kerman, “Notes on Beethoven’s Codas,” P-based C-theme], but the effect of this is quite differBeethoven Studies 3, ed. Alan Tyson (Cambridge: Cam- ent from that ofa full-scale return.” Cf. n. 10. bridge University Press, 1982), pp. 142, 146: “Mozart 12. Kerman, “Notes on Beethoven’s Codas,” p. 143.

often works without a coda. ... When he does have a 13. Kerman, “Notes on Beethoven’s Codas,” pp. coda it is short and carefully kept subsidiary to the rest 151-52, which also provides the reasoning behind the of the movement. And it is likely to echo the begin- objection; Ratner, Classic Music, p. 231, “not secondary ning of the development section. ... What is quite rare developments . . . but extended areas of arrival.” in Mozart is a coda containing a strong return of the 14. Leon Plantinga, Beethoven’s Concertos: History, Style, movement’s first theme. He sometimes makes reference Performance (New York: Norton, 1999), pp. 106—7. to the first theme in the cadential phrases, of course [our

Parageneric Spaces 285

not S!), both are examples of quadri-rotational of Mozart’s “Prague” Symphony and virtually movements. The “Waldste1n” movement is ad- all of his “Jupiter” Symphony.) In the opening ditionally rounded off with a further recycling movement of the Eroica Beethoven took on this of P—as if beginning a fifth rotation—at the challenge while renouncing the grand extension end of its coda (m. 295), while at the compara- automatically provided by the increasingly obble place the “Appassionata” refers first to S (m. solete option of a literal repetition of the devel240) and then, at the very end, to P (m. 257). opment and recapitulation. In compensation he Although Kerman did not discuss this issue in crafted a discursive coda that reviewed, in order, these terms, he did mention that Beethoven’s some of the central events of the development codas often begin with P-material (more so than and recapitulation. Presumably it is also for this do Mozart’s), that they are sometimes inflected reason that the first part of the coda is modulaat first into “one of the subdominant-area keys tory, invoking developmental procedures, while (subdominant, supertonic, flattened supertonic, the second part, referring to the recapitulation, even flattened submediant) before bringing it is grounded on the E-flat tonic.!6 round to the tonic,” and that they also often Once the fundamental conception 1s grasped, contain a “distant recollection” of the S-theme the ordering of the Eroica coda’s thematic delater on—a brief statement in which “the feel- tails fall readily into place. (Some other, ching 1s of a distant, nostalgic memory rather than mactic functions of this coda are noted in the of a firm restatement of reinterpretation.”!> next section of this chapter.) Following a brief, Another Beethovenian possibility is to em- P-based link with its famous chordal descents ploy a discursive coda to give the impression of a (mm. 551-63) some of the varied thematic refrecomposed or telescoped repeat of the develop- erences in the coda may be mapped, with some ment and recapitulation. Something along these flexibility (and adjusting for a sense of a properly lines might he behind aspects of the rotational “coda flavor”), as: mm. 564ff = 178ff (P-based); coda types mentioned above, but the clearest 582ff = 285ff (the famous “new theme” from the paradigm 1s provided by the huge coda to the development, whose reappearance here has profirst movement of Beethoven’s Eroica Symphony. voked much discussion);!’ 603ff = 338ff (domiOne of the central points of this movement was nant preparation; rebuilding for the recapitula-

to provide, in its early nineteenth-century con- tion); 631ff = 398ff (P, recapitulation); 673ff = text (presumably using the later Haydn sympho- 460ff (bolstering the case that Beethoven had nies as a standard of normative monumentality), indeed conceptualized this theme as S!!, as a sense of hypermonumentality, a heightened mentioned also in chapter 7).!8 Similarly insublimity of effort and expanse. (Its chief pre- structive instances— with their own variants of decessors in this regard were the first movement the idea— occur in the finales of Beethoven’s

15. Kerman, “Notes on Beethoven’s Codas,” pp. 154-55. 320) and a re-sounding of this “new theme” in that key 16. The essential details of the Eroica coda are also sum- at m. 323, that is, prior to the formal reactivation of the marized, more expansively, in Hepokoski, “Beyond the structural V at m. 338. For Cone’s remark, see Musical Sonata Principle,” p. 111. A similar observation, though Form and Musical Performance, p. 77.

in a different interpretive context, was provided by 18. Cf. Theodor W. Adorno, Beethoven: The Philosophy Robert P. Morgan, “Coda as Culmination: The First of the Music, ed. Rolf Tiedemann, trans. Edmund JephMovement of the ‘Eroica’ Symphony,” in Music Theory cott (Stanford: Stanford University Press, 1998), p. 104: and the Exploration of the Past, ed. Christopher Hatch and “At the very end of the first movement [of the Eroica] David W. Bernstein (Chicago: University of Chicago the original—then interrupted—idea of the second

Press, 1993), p. 371. subject group reappears as the last thematic event in the

17. The “new theme’s” coda-tracking through F mi- movement (apart from the chord syncopations). It 1s, nor (m. 582), then E-flat minor (m. 590, the tonic mi- as 1t were, redeemed, vindicated. Cf. Schoenberg’s nonor), is probably less a reaction to an imperative from tion of the obligation once contracted. — Moreover, this any presumed “sonata principle,” as Edward T. Cone theme already contains the kernel of the motif—the reclaimed in 1967-68, than it is a back-reference to the peated crotchets—of the consequent [sic] phrase which development’s touching on the tonic E-flat major at m. will follow it in the exposition, after the dramatic in316, followed by a quick collapse into E-flat minor (m. terruption.”

286 Elements of Sonata Theory

Symphony No. 8 in F, op. 93, and the String another example of a grandly discursive coda Quartet in B-flat, op. 130 (in which the “reca- with a final, appended codetta- or coda-like tag pitulatory” portion of the discursive coda is left at its end. incomplete). The larger idea behind all of these A “theatrical” discursive-coda effect was was recognized by Ratner: a longer coda could produced in Cherubini’s Overture to Les deux begin with “a harmonic digression (optional),” = journées (1800): at the end of a rather lengthy, “a firm return to the tonic, generally with the perorational coda (at an Allegro tempo) the last opening theme,” and “a set of emphatic caden- eighteen measures, a P-based “tag,” charge fortial gestures... . These events review those of | ward witha sudden increase in tempo, “sérrés le part II in compressed form— development, re- mouvement” (or Presto in some later editions).

capitulation, and closure.”’!? Beethoven called upon a similar effect more When confronted with a multisectional, than once —it would become a standard trope discursive coda, one often finds that its longer, of concluding excitement later in the centumain section itself ends with a shorter “coda,” ry—perhaps most famously in the determined a passage with the effect of a traditional coda tightening of the pulse found in the sempre piu proper or a separate coda to the coda. (In some allegro and Presto conclusion to the finale’s coda instances this might be reduced to a codetta to in the Fifth Symphony. The further possibility the coda.) In turn this suggests that the bulk of — of writing all or most of the discursive coda in the discursive tableau may also be understood an increased, more frenetic tempo was addition-

as an interpolated block wedged between the ally explored by Beethoven in other composiend of the recapitulatory rotation and its own tions: in some of his middle- and late-quartet “coda.” Sometimes introduced by a passageway movements, for instance, and in several of the or bridge from the end of the recapitulation, a overtures: Leonore 2 and 3, Egmont, and Fidelio. discursive coda typically subdivides into two main portions: (1) an extended tableau, perhaps Characteristic Functions of the

. Discursive Coda

multisectional, capable of serving a variety of . . purposes; and at the end (2) a briefer coda to the

coda, wrapping up the movement— something The many potential roles of larger codas have that may be P-based, suggesting the onset of a been remarked upon by several scholars. In his new incomplete rotation of materials (as in a 1982 study of Beethoven’s codas Joseph Ker-

more standard coda). man observed that many of them, especially

The general effect is by no means limited to in works after 1800, gave the impression of a Beethoven and later composers. Mozart’s Over- removal of difficulties or obstacles set up earture to The Marriage of Figaro is a Type 1 sonata lier in the movement—that such concluding with a discursive coda beginning in m. 236—a sections could serve the larger structural and “new,” P-related crescendo-module leading cu- expressive purposes of “‘normalisation, ‘resomulatively at mm. 250ff to a recapturing of an lution,’ “expansion, ‘release,’ ‘completion,’ and expositional TR-passage (cf. mm. 43ff), “lost” ‘fulfilment,’” diverse descriptions containable or omitted in the recapitulation. (As such, this under the general concept of “thematic ‘comdiscursive coda takes on a compensatory func- pletion.’”2° Especially in Beethoven’s hands, tion, one of the several types mentioned in the the coda could become the capstone or telos of section below.) A “codetta-like” coda-to-the- the entire movement, “providing an emotional coda effect—one last extension to the preced- resolution, or rather an apotheosis,” as Kerman ing excursus— occurs with the final grounding had written elsewhere about the coda to the first of the tonic in mm. 284—94. The famous coda movement of the Eroica, “so different in spirit to the finale of the “Jupiter” Symphony provides and form from the symmetrical resolutions of

19. Ratner, Classic Music, p. 230. 20. Kerman, “Notes on Beethoven’s Codas,” pp. 149, 151.

Parageneric Spaces 287

Haydn and Mozart.’?! (Mozart, of course, had cast aside once it is scaled and its utilitarian purprovided a few stunning precedents in his later pose of ascending used up—a generative matrix works, most notably in the “Jupiter” finale .)?? without which the crowning Klang of the codaRobert P. Morgan has similarly noted, particu- apotheosis would have been unattainable. larly with reference to the Eroica coda, the phe- Still another type of discursive coda somenomenon of “coda as culmination,” in this case times (though infrequently) found in a Type 3 manifesting itself as the goal of a long-range sonata form incorporates a written-out cadenza, “textural crescendo,” a “dynamically evolving suggesting a local intermixture with some of the process,” a culminating, tutti realization of reg- solo-display features of the Type 5 sonata. Other istral, textural, and dynamic space.” Focusing aspects of the Type 5 sonata, however— most on Beethoven’s deployment of the “heroic-style notably its reliance on ritornello pillars—are coda,’ Scott Burnham agreed with such views, not relevant to the other portions of such moveproposing that in the Eroica coda “the [P-based] ments. In its most emphatic realizations, the

melodic repetitions at the end... [suggest] “coda-cadenza” can even include a $ platform, the possibility of endless repetitions, endless some interior tempo changes, and a trill-cadence affirmation. ... Thus the final melodic utter- exit into a separate, additional coda-space. The ance of the opening theme has thematic stability locus classicus occurs after the recapitulation of but not thematic closure —again, what an ap- the first movement of Beethoven’s Cello So-

propriate way to signal an apotheosis.”4 nata in F, op. 5 no. 1. Related examples, better Another clear instance of the apotheosis classified as instances of coda-rhetoric interpolatype of coda occurs in the Egmont Overture, tion (CRI or CRI-effect), may be consulted in op. 84, to which Beethoven additionally shifted the first movement of his Piano Sonata in C, op.

the burden of tonal resolution, following an 2 no. 3—discussed in more detail below—and unorthodox, nonresolving recapitulation.?° in the birdsong cadenza in the second movePerhaps an even more idiosyncratic situation ment of the Pastoral Symphony.?°®

emerges in such pieces as Weber’s “Jubilee” Each discursive coda has its own role to play Overture (Jubel-Ouvertiire, 1818), a sonata form in the larger argument of the movement. Each rising in its coda to the fortissimo statement, An- needs to be studied individually and flexibly. dante, of a new, but pre-existing and commu- Caplin has outlined a number of categories of nally shared melody, in this case, “God Save the “compensatory functions” for such codas. His

King.” (Brahms produced much the same ef- listing includes: “recollection of main-theme fect decades later in his Academic Festival Over- ideas” (our incomplete coda-rotation); “restorature [1880], which ends with a peroration on tion of deleted material from the recapitulation” “Gaudeamus Igitur.”) Such examples suggest (as we have suggested occurs in Mozart’s Overthat the sonata process that precedes thistype of | ture to The Marriage of Figaro, restoring a TRcoda is not to be taken as a fully self-sufficient idea missing from the recapitulation — although

action (as most sonata forms normally were within the style such a restoration was by no throughout the eighteenth century) but rather — means obligatory or even expected); “reference as a preliminary exertion that must be under- to the development section” (such an event may taken before what happens in the coda 1s able to have rotational or repeat-scheme implications); emerge as a fully ripened possibility. This treats “shaping a new dynamic curve” (either apothe preceding sonata as a Wittgensteinian ladder theoses and climaxes, as in Beethoven’s Eroica 21. Kerman, “Theme and Variations” [rev. of Charles 23. Morgan, “Coda as Culmination: The First MoveRosen, Sonata Forms|,” The New York Review of Books, ment of the ‘Eroica’ Symphony,” pp. 357-76. 23 October 1980, 52. Cited in Morgan, “Coda as Cul- 24. Burnham, Beethoven Hero (Princeton, N.J.: Prince-

mination: The First Movement of the ‘Eroica’ Sym- ton University Press, 1995), p. 19. On the “heroic-style

phony,” p. 359. coda” see also pp. 53-60.

22. Cf. Rosen’s remark, “I presume that [Mozart] is 25. Hepokoski, “Back and Forth from Egmont.” Cf. n. 4.

the inventor of the contrapuntal coda” (Sonata Forms, 26. See also the discussion in Kerman, “Notes on

rev. ed., p. 314). Beethoven’s Codas,” pp. 154—55.

288 Elements of Sonata Theory

Symphony and Egmont Overture, or dissolu- expansion of C, which would overlook the retions, as in the Coriolan Overture); and “real- lationship of the interpolation with the kind of ization of unrealized implications” (remind- music commonly found in codas. ing one of Kerman’s “thematic completion” or What initially seems to be a straightforward “fulfillment”; Caplin’s examples are Beethoven’s concept turns out to be more intricate once Coriolan, once again, and the slow movements one examines it carefully. Since Sonata ‘Theory of Mozart’s Symphony No. 39 and Beethoven’s does not encourage the view that major func-

Piano Sonata, op. 10 no. 1).?7 tional zones can begin mid-phrase, it is more precise to distinguish between CRI proper and

Expanded C-Space and Coda-R hetoric a mid-phrase CRI-effect. The former implies

Interpolation (CRI) the conclusion of a preceding phrase with a ca-

dence and the beginning of a new one (CRI), As suggested above, a coda 1s normally con- perhaps elided or flush-juxtaposed with it. (The sidered to have begun once the rhetorical pat- P-based CRI is the most common option: rebetern of the exposition has been retraced (in the ginning P with a new phrase.) But sometimes tonic) to its end in the recapitulation. If there the coda-like expansion 1s not separated off or are three C-modules ending the pattern, one “enclosed” as a separate phrase between two cawould expect the recurrence of all three C- dences. Instead, it appears as an interior expanmodules before the coda proper begins. In this sion of a pre-existing module, as though that case we would find {C! C? C3 || coda}, in which module had been inflated in the middle to in|| represents the point at which the recapitula- corporate coda-like rhetoric. In these instances tion has attained the point parallel to the end of = we have a C-module expansion with CRI-

the exposition. effect (or S-, S©-, or TR=FS-module with Occasionally, though, a composer would CRI effect, when those situations apply). One interpolate a passage of coda-rhetoric material typical pattern might be something like {C! —

(strong closing cadences, perhaps a return to the C? with coda-effect expansions — C3 || coda}, normative P, and so on) before all of the final although, of course, the number of C-modules recapitulatory modules have been sounded. present varies from piece to piece. (The new expansion would normally occur in An additional complication is that not all C-space, although not all expositions and re- C-module expansions invariably produce a capitulations contain that zone. When the ex- CRI-effect. The modular expansion may also position and recapitulation feature an EEC and occur with other implications, sometimes that ESC occurring at or very near their ends, the of preparing for a CRI proper. This could proexpansion could be applied to the later stages of = duce a pattern like {C!—C?, expanded — CRI S, SS, or TRS FS.) This produced a concluding as new phrase, usually P-based — C3}. Obvi-

pattern on the order of {C! C? — expansion ously, the degree of CRI-effect in such a C? via coda-like material — C3], which reserves expansion (in this case) could vary from piece the final C-module, or sometimes the final two to piece, and different analysts might come C-modules, for the close of the work, thus pro- up with modestly divergent readings of the ducing a concluding musical “rhyme” with the evidence. A composer could deploy the CRIend of the exposition. We refer to this procedure effect in any number of clever and inventive as a coda-rhetoric interpolation (CRI). This termi- ways, and one should be prepared (especially nology steers clear of two misconceptions. The with the constantly astonishing Haydn) to confirst would consider the coda to begin after C?, front each instance on its own terms. thus ignoring the sonata-space, non-coda status There are several examples of the CRI in of C%, returned to as a set of correspondence the literature. One paradigmatic case occurs at measures with the exposition. The second would the end of the first movement of Mozart’s Symconstrue the CRI only as a functionally neutral phony No. 40 in G Minor, K. 550 (Examples 27. Caplin, Classical Form, pp. 186-91.

Parageneric Spaces 289

13.1a and b). Assuming that the last phrase of — tion we find that the last bar of the sentence’s the repetition of C? (m. 279 = the exposition’s presentation module (m. 295 = 138) veers from m. 91) has been expanded in such a way as to its model and launches a formal interior expanlead into it, the CRI proper is sounded with sion with a strong “coda” flavor. The equivalent the piano-dynamic P-material in mm. 287-93 of m. 139 is regained around eight bars later, (with an exquisitely dovetailed anacrusis begin- in m. 304, and the C? phrase ends with more ning in m. 286). As an elided conclusion, C? or less strict correspondence measures with the subsequently bursts forth, forte, to complete the exposition, mm. 304—7 (= 139-42). The final recapitulation, mm. 293—97 (= the exposition’s chord is expanded for two more measures (mm. final mm. 95-99). A brusque, two-bar exten- 308-9), which may be regarded either as a tiny, sion 1s appended at the very end (mm. 298-99): appended coda or as an example of final-Cthree pitiless tonic chords, perhaps construable module terminal expansion (with coda- or co-

as a small coda proper.”°® detta-effect). Something similar occurs in the Another instance is found in the first move- finale of Mozart’s Symphony No. 40, where the ment of Beethoven’s Piano Sonata in C, op. 2 interior expansion occurs within the repetition

no. 3. Here the recapitulatory C! is found in of the C! module (mm. 286-301), while the mm. 212—17 with suppressed cadence. Instead movement itself ends with a more literal transof cadencing, the still-active C-module plunges position of C?, mm. 301-8. suddenly into a lengthy extension beginning on C-module CRI-effect issues can be handled A flat (bVI, m. 218) and moving onto a $ with in varied ways. The situation near the end of the fermata (m. 232, beginning) that releases an 1n- finale of Haydn’s Quartet in B Minor, op. 33 terior cadenza and a V’ (m. 232, end) whose no. 1, is fairly straightforward: the P-based CRI effective resolution occurs on the downbeat of occurs in mm. 184-89, and it leads to a varied m. 233. Strictly considered, mm. 218—33, not a restatement of C2, mm. 190-94, which ends self-contained phrase (no phrase had concluded the movement. More challenging and diverse before it began), are best regarded as a rhetorical are the examples found in the first movements expansion of C!, but their purpose 1s to set up an of Mozart’s String Quintet in C, K. 515 (featurunmistakable CRI. This occurs with the elided ing an aborted drive to a I:PAC, mm. 320-21, onset of the next phrase, mm. 233-52 —featur- followed by an unsettling, “added” infiltration ing a telltale return to P!—elided again, at the of the C? module [and later, fragments of S]

end, with the as-yet-unsounded recapitulatory occupying a full thirty-two bars before the C2, which now surfaces to conclude the move- correspondence-measure rearticulation of C?

ment in mm. 252—57.7? proper and few concluding bars, mm. 353-68), A classic example of C-module expansion Haydn’s Symphony No. 103 in E-flat, ““Drumwith CRI-effect—in the absence of a CRI roll” (strictly considered, a complex S©-module proper— occurs 1n the first movement of Mo- expansion that includes a return of the introzart’s Symphony No. 39 (examples 13.2a and ductory Adagio), and Haydn’s String Quartet b). In the midst of the usual round of corre- in B-flat, op. 76 no. 4, “Sunrise” (in which the spondence measures the ESC occurs at m. 276, strong, P-based CR I-effect— harmonically, not and the tonic restatement of C! follows, mm. a separate phrase, despite the preceding ferma276-92. In the exposition the concluding C? ta—is launched over a subdominant chord, m. had been articulated as a straightforward eight- 175).

bar sentence, mm. 135-42. In the recapitula- Finally, one potentially problematic issue 28. Kerman, “Notes on Beethoven’s Codas,” p. 143, some of his own works (as in the first movement of the mentions “the characteristic calando effect” (as in the “Pathétique” sonata, op. 13, etc.). finale of the Piano Concerto No. 24 in C Minor, K. 29. Kerman, “Notes on Beethoven’s Codas,” p. 147, al491) or “a pathetic calando followed by some sort of fu- ludes to this procedure (without our terminology) in rioso conclusion” (as here) of some of Mozart’s codas as Beethoven’s op. 2 no. 3 and understands it as characterbeing an effect that Beethoven sought to replicate in istically Haydnesque.

88 p Zz t—12 =A #JT . J EXAMPLE 13.1a Mozart, Symphony No. 40 in G Minor, K. 550, 1, mm. 88—100

[Molto Allegro]

| 4 A | | ce Py rr rc eecmo,—CriVSOHPYsSséaéFésrLsCSC™*#F#3K, @ & FY = |

—— Left

[a

91 fe= | =| . iio a” A JJ) eefea. a eo

Ee ee ee I pence DO sD OD QO (OU oe neo P 6 eee ee ee lr, ttrti‘OSOCOCOCOCOCCC:CSC SC rr ZPLLlLUDmyll TT ee Ee

BF Feee esrree esES es ee |© A a es | (pace nee eeaOE es sO eAGSDO

94 . . *. .:os; . :. \ . —\

a A ” _ -s s wreee GO ec I (pene QO GO ee ————SS ————

Eee 4 Gi er ee ajie a ee$F es RS "QP FF OG 7rfp. a@p wo. feo. che. tt et | Pet TPP ey| [|yy iv, ft | | f|{ 7{ T{ T{ {[ {| _JT [pee et | ~ @ ~| YeUU yf gy frUU jy =f

LE SPLELSe ELne i 2 eee hail A PA nee nc 290

EXAMPLE 13.1b Mozart, Symphony No. 40 in G Minor, K. 550, 1, mm.

276 oy /) cd , rt —_— =ae te |

Zs a - a so pP

282 -—~ — TT D> ib)» he | Heo i i i ie cw” FA gq FT| oS -ee, a A1)- Ty a, fFoeff, oo

Pp ~ fr 220 ~ : ) ° ob aw, D Y a” A es le ee Pe el a oo irra’ v™s* ame a PN = TT z YUL =eT LL

oT Se er iprQD DD” «SS eerrrT E—E—ERDQ Re sf, ky hey — | @ Tie @¢c@ Pe | F- FP | Fly” | mT mh Ld ee Pp

Pp

289 9 sdTOded

o~ . m~ o~

~ 3 fea deed se

SN * et *e

EET A

se — 4: oyeALA t ee to

Sa

291

rey Te 296 fey 2a /Pw ‘ 2 ga wy eg I PEO gy© ly PIIIgeSg SY 292 Elements of Sonata Theory

293 'we"eo ) - Pa ‘ r a os Mt SS lg4 rne r ,"en .,‘ee aurneoeae PT a'I id we EXxampLe 13.1b (continued)

S on . . a, . . om . ‘ o_o . . om * . oo, . .

l#eeg io)7?. “sie J led. | pie fe gg er? er? er?is,

SP” A Pe ee GO GO GO CO (OF

ppreteiete, | * om~ . . on . .

id oe, wr

arises when the CRI in question is both P-based even only one, two, or three chords (perhaps the and occurs within a movement that is in dia- formulaic triple-hammer-blow) that, in contrast logue with the Type 2 (“binary”) sonata— that to more explicit P® or P!-° material, are genutype in which a full recapitulation beginning inely prefatory. Such a brief, in-tempo introduction

with P is not provided. (Chapter 17 lays out the serves as an initial spur for the entire moveargument more fully.) The error is in imagining ment, a gateway emblem or wluminated initial. the P-based CRI to represent a recapitulatory It is normally not involved either in the expogesture, one that produces some sort of “mirror sitional repeat (often the repeat sign is placed form” or “reversed recapitulation.” Such claims, after the introduction, before the P-theme) or for example, have been repeatedly made about in the launching of the recapitulatory rotation. the first movement of Mozart’s Piano Sonata in There are several examples of this kind of openD, K. 311—a deformational piece whose com- ing material in Haydn’s quartets (op. 71 no. 3/1 plications are not easily summarized in a few sen- [a single chord], op. 74 no. 1/1, op. 76 no. 1/1, tences. For the present, we might say only that and so on), although some of them are “slowed” we regard the return of the incipit (only) of P in or provided with other delays through one or m. 99 to be a passage of coda-rhetoric interpola- more fermatas. Doubtless the opening two meation lasting until m. 109, when it is elided with sures of Beethoven’s Eroica Symphony recalls a the onset of C!. In this case the CRI 1s wedged tradition of brief, in-tempo introductions. (See between S-space— with the ESC at m. 99 —and the discussion in chapter 5 concerning possible the beginning of C-space. A comparable situa- intermixtures of this kind of introduction with tion occurs in the related first movement of the the P® or P!-9 idea.)

Sonata for Violin and Piano in D, K. 306. Also characteristic are more extensive introductions, which usually provide a prolonged sense of anticipation and formal preparation for

The Introduction a rapid-tempo sonata-to-come. The present section will focus on these longer introductions. From time to time a piece or movement begins The longer and more complex an introduction, with material that precedes the sonata form. the more importance 1S being claimed for the This may consist of a brief cadential formula or piece as a whole. (This does not mean that pieces

Exampte 13.2a Mozart, Symphony No. 39 in E-flat, K. 543, 1, mm. 132-42

[Allegro] 2 ‘ 132 bo ie: Fe _@. PD f)AS ae he a7/2 call ‘T — —ll| aOF > AOG A 4o> oD“AE AG7~| —SO ———— at hooaktlak ™ aN —s bh cle: | “ be 5a oe L_ ye oroTTF eo eeoe™Door iSD OO) A OD ? y 2222 ]1 et ie,

f) eft — ee =~©— 4eT —OES —- — —OOOO =—F = A) ero ESE EEE Hea? ps EE eee _seesesese esses ,_

SS a J 7 3a 00 eoer? ter SOS

PO eee 7. oP oes a| —— ssoot es ore ee

oo oO >a OO TT Ooo NNT a OO — OE .,?— I @Mu8n90Thumo€—>snnnnw00]TIIIII-]

139 @ aw et, yo * Hf 9 hrEE 9 0ED nee Pe Ps iea oo a yet oF—— oea [I

— pares aT Ne

pg 0? nent —$§$ St ——

7... rr ee

14] . tf — : ef aN . — ft, ey? SS OO— eeSs—O—OFehMN TF So sss $ . ae oTSS .Que ‘ ms.CG . ymM.:qY oo Oy | ye ia A ——— es ee

i ExampLe 13.2b Mozart, Symphony No. 39 in E-flat, K. 543, 1, mm.

289-309

289 D ae oe el. = +e . i a ee ee 8 a se a a ee = ee 7A (ol [Allegro]

REY 7 | FA nnn nn (Un ——— GE

tr te——" A A i

>8=es— igPR . Zia —-— TT Dye Ev o_o i OT) DR 2RQ

a; fe... Ty ated,

ye ye eg yeOT a =e ne Me aa 295 . o P| 5 ud a de £ £ Iee | 1 Fy ]—__a___g _ a aE to wire wre L b 2a a PD” > —— 7-9 OE A a PS Ay gy I a a ee ee ee ee ee ee SS SS

aes ee EL Pree crrr A ee COGN QOaoe OSCO >" CO PRAAesGG oP

Ter

|Ec en |gg a _eg ei £ 2—J ¢—124 )mw —— —

a SS a a | I. a hE ee ns ne PyOT PPi gee ———S =e.

! TT ue: ioe i > —PPR @— — Po OO —4 — i ]o_] oe i hE| CE GG OO oelr OO

301 2 o “ oe o —}_ |_| FF @ 0, i se LF |, — letters ——|——|-—f@ Pom in ——— fn i nn ES ee SS ee i ——_—_——$ canes

loa | >—— i, —o | o_o —_f 7 Vr

f= Parageneric Spaces 295

a A —= J a 2 faze EXAMPLE 13.2b (continued)

Ce Te om ater sl). 7 5 cc ee ee ere —— a a a ee oeee ee 54d036s — CORR A

ON SG ee 7 7 a Rs DG CS SS OO A (QO OU

2 SS SS SS ee

A

without introductions are ipso facto less serious. Hofmann’s most important and original contriIt means only that a sizable introduction more butions to the symphony [c. 1760] was the adopovertly thematizes the claim of the formality of tion of a slow introduction to the opening movethe occasion and the seriousness of the compo- ment, anticipating Haydn.”%? Within Haydn’s

sition.) Almost invariably such broader intro- symphonic output—and even though a few ductions are in a slow tempo and end with an earlier examples may be found—they began to expectant pause, usually on the dominant, thus occur more frequently in a spate of symphonies setting up the trip-lever effect (“Go!’’) that sets between 1773 and 1775 (chronologically, 50, 54,

the subsequent sonata form into motion. 57, 60, 53); they continued a few years later with Symphonies Nos. 71, 73, and 75; and the prac-

. tice seems to have evencer more Introductions andbeen Genre 5 consolidated ,

Slow or Movement Types in the six Paris” symphonies (1785-86), three

of which (84, 85, 86) have slow introductions. Slow introductions are not equally available to all By the period of the twelve “London” symAllegro movements. Since they typically connote phonies—at the time the touchstones of grand, a heightened sense of dignity or grandeur, they monumental symphonies—these introductions were most appropriate in such “public” statements were normative in Haydn (except in the sole mi-

as symphonic first movements, overtures, and nor-mode symphony of the set, No. 95 in C misome festive serenades (Mozart’s Serenade in D, nor). Mozart also had his role to play, writing K. 320, “Posthorn’”). One mid-eighteenth-cen- substantial slow introductions for the first movetury pioneer in the production of orchestral slow = ments of the Symphony No. 36 in C, K. 425, introductions was Leopold Hofmann. The cur- “Linz” (1783), No. 38 in D, K. 504, “Prague” rent New Grove Dictionary reports that “among (1786), and No. 39 in E-flat, K. 543 (1788), for 30. Hermine Nicolussi-Prohaszka and Allan Badley, Journal of the American Musicological Society 34 (1981),

“Leopold Hofmann,” in The New Grove Dictionary of 562: “About 1760 [the Austrian] Leopold Hofmann Music and Musicians, 2nd ed., 11: 605. This sentence does experimented repeatedly with what became the final not appear in the preceding edition of the New Grove, in design of the symphony (slow introduction plus sonata-

which the Hofmann article was written by Nicolussi- allegro; andante; minuet/scherzo plus trio; substantial Prohaszka alone. Cf. Jan LaRue’s summary, in his re- finale) somewhat earlier than Haydn’s occasional pioview of Charles Rosen, Sonata Forms (1st ed., 1980), neering efforts in this framework.”

296 Elements of Sonata Theory

the Overtures to Don Giovanni, Cost fan tutte, Die one might mention the 29-measure Adagio lead-

Zauberflote, and so on. ing to the Allegro di molto in the first move-

Current discussions of slow introductions ment of Mozart’s Sonata in F for Four Hands, typically suggest two additional, practical func- K. 497 (1786). (The composer also appended a tions of such passages. On the one hand, as full coda to this ambitious movement.) With the pointed out by Laszl6 Somfai, in the eighteenth 1790s and early 1800s one finds these introduccentury they could have a “‘noise-killer’ ef- tions in an increased scattering of sonatas, 1nfect,’ calling a bustling audience’s attention to cluding a few by Clementi (such as the Sonatas the music at hand, particularly within large or in G Minor, op. 34, and B Minor, op. 40 no. 2) spacious rooms.?! On the other hand, as often in and Beethoven (Sonatas in F Minor, WoO 47 Haydn, they served to set up allegros that began no. 2;in C Minor, op. 13, “Pathétique,” in Equietly or with off-tonic openings. Orchestral flat, op. 81a, “Lebewohl,” and in C Minor, op. introductions were also explored by other cigh- 111). Even with such distinguished exemplars teenth-century composers. However important the practice remained somewhat exceptional. they may have been to the subsequent tradition, When it did occur, the implication was probHaydn and Mozart were expanding on the work ably that of providing an uncommon elevation

of predecessors.°? of aesthetic intention, as if one were producing Any slow introduction to the first movement the pianistic equivalent of a symphony.

of a trio, quartet, or quintet—a less familiar Finales of multimovement sonatas bepractice— must have reinforced the advancing fore 1800—normally lighter or more playprestige claims of the genre at hand, suggest- ful movements—are almost never prefaced ing that something of special importance was to with separate, slow introductions. Within the follow in both the movement and the piece as a generic practice most relevant to Haydn, Mowhole. The first movement of Haydn’s Quartet zart, and early-middle Beethoven that honor in D, op. 71 no. 2, prefaces its Allegro with a is generically reserved only for the first movelyrical, four-measure Adagio, bridging the sta- ment. Consequently, any slow introduction to sis of the external silence to the suddenly forte an eighteenth-century finale would have been agitation of the exposition. The fourteen-mea- regarded as unusual. ‘This is apparently the case sure Adagio preceding the Allegro assai in the in such works as Dittersdorf’s Symphony No. 6 first movement of Mozart’s Quartet in E-flat, in A on Ovid’s Metamorphoses, “Transformation K. 171, strikes an initial tone of unanticipated of the Lycian Peasants into Frogs” (c. 1781-82), earnestness in its 1773 context. (The introduc- in which the finale begins with a mysteriously tion also returns to conclude the movement.) reflective Adagio that both precedes and interMasterly examples of slow introductions from rupts a vigorously contrapuntal, quasi-fugal ViMozart’s later chamber works include the first vace, ma moderato (not in sonata form), before movements of the Serenade in B-flat for Thir- the movement dissolves into amphibian croak-

teen Winds, K. 361, the Quartet in C, K. 465 ing at the end; or in Boccherini’s Symphony (the celebrated “Dissonance” opening), and the in D Minor, op. 12 no. 4 (G. 506, ca. 1771), String Quintet in D, K. 593. Beethoven opened nicknamed “La casa del diavolo” (“In the devil’s his Piano Trio in G, op. 1 no. 2, and both of — house”; borrowings from Gluck suggest an unhis op. 5 cello sonatas with substantial, Ada- derlying Don Juan program), in which the same

210-tempo introductions. slow introduction precedes both the first moveSlow introductions to first movements of pi- ment and the last. ano sonatas are rare before the 1790s, although More well-known instances also occur in

31. Somfai, “The London Revision of Haydn’s Instru- ihre Herkunft und ihr Bau bei Haydn und Mozart, 2 vols. mental Style,” Proceedings of the Royal Musical Association (Tutzing: Hans Schneider, 1977). Cf. A. Peter Brown’s 100 (1973-74), 166, referring especially to Symphonies review of this book, Journal of the American Musicological

No. 93 and 104. Society 33 (1980), 200-204. 32. See Marianne Danckwardt, Die langsame Einleitung:

Parageneric Spaces 297

the late-eighteenth and early-nineteenth centu- might be a single phrase or two (often with ries. What happens in the astonishing G-minor erandly-robed, “regal” dotted rhythms or coups Adagio before the “false-front” G-major finale d’archet), a stiffly formal exordium (rhetorical of Mozart’s String Quintet in G Minor, K. formal sign of opening), a fanfare-like gesture 516 (1787) strikes one as extraordinary, within (Haydn, Symphony No. 104, “London”), a set a piece that has a multimovement logic all its of striking, cadence-like gestures (Beethoven, own. Early Beethoven 1s also relevant: his String Symphony No. 1), or even a single, imperious

Quartet, Op. 18 no. 6 (1798-1800), whose chord (Beethoven, Symphony No. 2, Egmont finale features the celebrated “La Malinconia” Overture). When the initial impulse 1s compact, Adagio introduction, comes immediately to it sometimes melts into gentler, zone-2 material, mind, as does the last movement of the Sym- and it may return to relaunch or invade some of phony No. 1 in C (1800), with its coy, witty the subsequent phrases as well. introduction. Once intensely serious slow intro- 2. Quieter material, often a brief, lyrical melody. ductions begin to be written for finales laterin — Especially when articulated with a lyrical melothe nineteenth century (as 1n Beethoven’s Piano dy —by no means the invariable case — this zone Sonata in B-flat, op. 106, “Hammerklavier,” his gives the impression of blocking out a roomy exSymphony No. 9 in D Minor, op. 125, and his panse of musical space, thus already predicting Quartet in F, Op. 135; in Mendelssohn’s Sym- the ample breadth of the sonata-to-come and phony No. 5 in D Minor, op. 107, “Reforma- the consequent proportions of the whole. This tion”; in Brahms’s Symphony No. 1 in C Minor, zone may be occupied by the piano aftermath op. 68; and in many other works), the effect is of zone 1, emerging as that which 1s released generally that of reopening a serious issue 1n1- by the initial forte impulse. Or it may exist on tially broached in the first movement: the issue its own, without being preceded by an earlier now has to be reconfronted in a “do-or-die” _forte. In a major-mode work it sometimes hap-

finale. pens that the major collapses here to minor or displays significant borrowings from the parallel

Characteristic Zones within a Slow minor. If so, the aeMOSP here ‘YP ically becomes

Introduction one of brooding or foreboding, one of “the fall

that must be restored in the subsequent sonata. While it is difficult to generalize about what In more advanced compositions this zone might can happen in slow introductions, a generically modulate fleetingly away from the tonic, pertypical introduction has available to it four ex- haps temporarily escaping the implications of a pressive or functional zones, which usually be- reigning minor mode, perhaps framing itself off

come accessible in order. By no means do all from its surroundings as a mini-tableau (as in introductions make use of all four zones. An Beethoven’s Overtures Leonore 2 and 3), or perintroduction may omit, elide, or intermix one haps tonicizing a different key for other expresor more zones for localized expressive purposes. sive reasons. This zone may also initiate a m1xOne may also encounter an idiosyncratic intro- ture of different “topics,” figures, and rhetorical duction in which the zone-concept seems inap- gestures, as suggested by Ratner.°? plicable or strained as a background interpretive 3. Sequences. These often give the impression device. The zones, which may be treated very of a “searching” — groping toward the attain-

flexibly, are: ment of the structural dominant. Once again, 1. Heraldic or annunciatory call to attention. This this often involves either a decay to the minor

is an initial forte impulse launching the entire mode or a continuation of an earlier such decay, work in a grand or “important” style, claim- thereby suggesting considerable anxiety. When ing and clearing space for what 1s to follow. It such sequences are greatly extended, they can 33. Ratner, Classic Music, pp. 104-6, provides a much-cited discussion of the introduction to Mozart’s “Prague” Symphony along these lines.

298 Elements of Sonata Theory

have the local effect of a fantasia, as in Mozart’s phony No. 97].”5° A further “dramatic” devel“Prague” Symphony and his Overture to Don opment in the ever-transforming concept of the Giovanni, K. 527, both of which also invoke introduction 1s the idea of bridging over the end

the darkened, ominous “ombra” (“shadow”) of the introduction to the beginning of the so-

topic.*4 nata proper with a special link, often accelerando, 4. Dominant preparation. At some point near giving the impression of a precipitously accumu-

its end the zone-2 or zone-3 music locks onto lating energy out of which the sonata is hurled a structural dominant (V of I) to begin one or __ forth like a javelin, as happens in Beethoven’s more measures of final dominant prolongation. Fourth Symphony, in his Leonore 1 and Egmont Occasionally, the V is not prolonged at length Overtures, and in several other works.

but merely attained at the introduction’s end. Although these four zones and their subHowever it is introduced, the last measure (V) types provide many of the basic patterns and ends generically with an expectant fermata—a expressive effects for eighteenth-century introharmonic interruption on the dominant— that ductions, they do not underpin them all. Some separates the introduction from the sonata form introductions might rework individual aspects that it has prepared. As part of deformational of individual zones to the exclusion of others. practice this zone can even prepare the “wrong The opening of Mozart’s Quartet in C, K. 465, dominant,” as, most famously, in Haydn’s two “Dissonance,” for example, may be understood late Symphonies 1n E-flat, Nos. 99 and 103, both as beginning at once with Zone 3—-sequential, of whose final, preparatory introduction-zones foreboding, and “searching.” The same might set up V of C minor (V/vi), although in No. 99 be said of an introduction that shares many of the “wrong dominant” 1s “artificially” corrected the same characteristics: that found at the open-

through the last-moment addition of an extra ing of Beethoven’s Quartet in C, op. 59 no. 3. measure, V7 of E-flat. But with Haydn—as al- In addition, as has been widely noted, from ways— the situation 1s even more unpredictable. the “Paris” Symphonies onward Haydn often ex-

As James Webster has pointed out, the mature perimented with introductions that motivically Haydn employed a number of “through-com- and harmonically foreshadow musical events positional” procedures to soften the virtually and themes of the sonata-to-follow. (As Webster inbuilt break between an introduction and the put it, “In general, the later the work, the closer sonata form that follows. These included: the and more pervasive the relations [to the followuse of a “harmonic progression [that] deviates ing fast movement].”)3° One classic instance ocfrom the customary -V|II- (Symphonies 60, 73, curs in the introduction to the first movement 88, 90, 92, 94, 99, and 103),” in the simplest of his Symphony No. 98 in B-flat, which starts of cases of which, “the introduction still ends with an Adagio, minor-mode version of the idea on the dominant, but the allegro begins off the that will become the sonata’s major-mode P (mm. tonic” (Nos. 60, 73, 86, 88, and 94); and, es- 16) at an Allegro tempo. Similarly, his Piano pecially in his later symphonies, the appeal to Trio in C, Hob. XV:21 (1795) begins with a “run-on” introductions, in which the introduc- six-bar Adagio pastorale, an idea in 6/8 meter tory music flows more directly into the sonata that springs to life at the opening of the enform, sometimes even with the suppression of — suing, 6/8 Vivace assa1. Introductions that are the normal fermata into the exposition, as es- generative or that serve as sources in other ways pecially in Nos. 86 and 97 (“No other Haydn for the material of the subsequent sonata will be introduction is run-on to [the] extent [of Sym- readdressed below in the section on “expressive

34. On the presence of “ombra” and a discussion of the 35. Webster, Haydn’s “Farewell” Symphony and the Idea of topic more generally see, e.g., Elaine Sisman, “Genre, Classical Style, pp. 162—65. Gesture, and Meaning in Mozart’s ‘Prague’ Symphony,” 36. Webster, Haydn’s “Farewell” Symphony and the Idea of

in Mozart Studies 2, ed. Cliff Eisen (Oxford: Clarendon Classical Style, p. 162. Press, 1997), pp. 27-84; cf. Ratner, Classic Music, pp. 24, 104-6.

Parageneric Spaces 299

or representational functions.” Obviously, they 109, a movement without an expositional reare also found in works by Mozart, Beethoven, peat. And in the first movement of Beethoven’s

and numerous other composers. String Quartet in B-flat, op. 130, the opening Finally, one should observe that some in- (and recurring) Adagio ma non troppo, prepartroductions, such as the opening, 22-measure ing a contrasting Allegro, is not only repeated Largo in Haydn’s Symphony No. 102 in B-flat, with the whole exposition but may also be unimply a truncated rounded-binary design, AA'B derstood simultaneously as both an introduc... (instead of AA'BA"). The structural dom1- tion— clearly its principal role—and the onset nant and fermata are reached at the end of B. As of a deformational P. The first movement of his a result one of the exposition’s functions 1s that earlier Piano Sonata in D Minor, op. 31, no. of beginning to supply (in extenso) the final limb 2 (“Tempest”), had provided a less self-evident of the rounded binary. This is especially notable forerunner of this procedure. when the P-theme is related to the introduction, Associated with the above are false-start sonaas is the case here. From a larger perspective this tas, a relatively late development in the genre. rounded binary structure, begun in the intro- In these cases the piece begins with a fast-tempo

duction, will not attain its own tonal closure flourish (often either the P-idea itself or someuntil the sonata form itself also does so—that thing related to it) only to be called up short,

is, at the moment of the ESC.°’ reined back into a more “proper” slow introduction or other slow passage. One’s impres-

Variants and Later Deformations of the sion is that of a sonata movement rashly “et

“Slow” Introduction to plunge forth, only to be stopped, set back,

for the generic formalities. Alternatively, one Much less frequently, perhaps deformationally, might regard such a procedure as a vivid anithe broad introduction can also be laid out in a mating of the characteristic “call-to-attention” more energetic tempo, as in Rossini’s Overture heraldic fanfare or illuminated initial encounto Il Signor Bruschino (1813), where a 38-measure tered so frequently at the openings of slow 1nAllegro precedes a Type 1 sonata in the same troductions—a manic impulse that writes over tempo (into which characteristic portions of the the standard, usually more majestic gesture. In introduction return to fill in the interstices); or Beethoven the touchstone case occurs in the Fiin the same composer’s Overture to La gazza delio Overture: an abrupt Allegro, mm. 1—4; a ladra (1817), which begins with a snappy Mae- reflective Adagio, mm. 5—12; a return of the stoso marziale to set up the principal Allegro. brief Allegro, now on the subdominant, mm. Even more deformationally, on very rare occa- 13-16; and then a longer slow introduction sions, the slow-introduction concept is merged eventually merging into the Allegro onset of the with the sonata proper, at least in such a way that sonata deformation. Rossini provided some reit participates in the expositional repeat as well. lated examples in his overtures, notably 1n those In the first movement of Mozart’s Violin Sonata to La Scala di Seta and Le siége de Corinthe.>°® in C, K. 303, a Type 1 sonata, P and TR of The main components of the false-start procethe exposition and recapitulation play out as an dure —impulsively fast and brief but aborted; Adagio, while S and C shoot forward in an Al- slow and more properly deliberate; and an ensulegro molto. A simular, though more elliptically ing fast sonata form—recur as a virtually norcompressed (and complex), version of this, with mative practice in several of Berlioz’s sonatatempos reversed (Vivace, then Adagio espres- deformational overtures, such as Benvenuto Celsivo), begins Beethoven’s Piano Sonata in E, op. lini, Le carnaval romain, and Le corsaire.

37. Itis thus not correct to assert, with Webster (Haydn’s 38. Cf. Rossini’s Overture to Semiramide (1823), in “Farewell” Symphony and the Idea of Classical Style, p. which the opening Allegro vivace is merely a large cre163), that the Allegro’s “first theme (or even the entire scendo over a tonic pedal and the space of the slow infirst group) is a large-scale ‘consequent’ to the introduc- troduction is taken up with an Andantino. tion as a whole.”

300 Elements of Sonata Theory

Obviously relatable to this procedure is its as open to more than one defensible reading. reverse: slow introduction; false first start with One should not discard such readings as emAllegro P-material, aborted; return to slow in- pirically unverifiable: confusing knowledge troduction; restart or continuation of the Alle- with interpretation, this strict view would only oro. The impression is of two or more attempts impoverish the musical experience. Each readto launch a P-theme, only the last of which suc- ing is limited, provisional, metaphorical, an act ceeds. In turn this suggests either a need to go of conjecture in dialogue with the music as a back to the “reflective” introduction to allow historically informed, inevitably personalized the faster theme to be gestated more sufficiently response.) The listing here of some commonly or a momentary indecision or reluctance to face recognizable functions is intended only as a set the task that is to follow. We have already men- of loose categorizations projected back from the tioned the opening of Beethoven’s String Quar- present onto a large and varied body of music. tet in B-flat, op. 130, in a slightly different con- The proposed types overlap considerably. In text. That of the Quartet in A minor, op. 132, some instances one category can be understood

is in many ways comparable.°° as a special case of another, a subset, or a slight rephrasing of a similar, more general kind of

Expressive or Representational Functions connoraron. In addition to carrying out their standard ge- Representation. Introductions could take on an neric duties, individual introductions took on imitative or pictorial function either by a genexpressive or representational roles vis a vis eralized implication or by various means supthe sonata-to-come, as though their implica- ported by a verbal or topical suggestion. Extions continued to resonate throughout the rest amples include the “sunrise” introduction of of the movement. The specific effects varied Haydn’s Symphony No. 6 in D, “Le matin” from piece to piece, although certain families (“Morning”); the prolepsis (flashforward) of of introduction types were normative by 1800 the appearance of the statue of the Commenand the ensuing decades. These supplementary datore in Mozart’s Overture to Don Giovanni; functions were not so much objectively “in the and the prolepsis of the imprisoned Florestan music” (as raw, tangible “facts” to be agreed in Beethoven’s Leonore Overtures Nos. 2 and upon by all commentators) as they were avail- 3. In each case the subsequent sonata form reable to imaginative and thoughtful listeners as sponds in one way or another to the illustration part of the spirit of interpretation that enveloped set forth in the introduction. More broadly, this the sonata enterprise. The composer invited a typically suggests: circle of listeners to discover or read such implications into the piece. From time to time addi- The World, Condition, or Field of Implications within tional clues (nicknames, titles, identifiable top- Which the Subsequent Sonata Form Plays Itself Out.

ics) were provided to assist in this process. This is clear within overtly or connotatively By no means was an introduction limited programmatic introductions and sonatas, such to conveying only one expressive implication. as in all the overtures mentioned in the cateViewed simultaneously from differing per- gory above, as well as in Beethoven’s Overture spectives—not usually a difficult interpretive to Egmont (a world of tyrannical oppression, task — an ambitious introduction could provide responded to by heroic actions of resistance), several at once. (We take it as axiomatic that Weber’s Overtures to Der Freischiitz (rustic vilin the realm of hermeneutics one should regard lage life) and Oberon (the idealized fairy-world most instrumental pieces, usually underdeter- of the forest), and many other works. Mozart’s mined in the specifics of their connotations, Overture to Cosi fan tutte asks us to imagine its

39. The procedure was obviously taken up again to- Franck’s Symphony in D Minor. (Cf. the first moveward the end of the nineteenth century in, for example, ments of Mahler’s Symphonies Nos. 3 and 7.)

Parageneric Spaces 301

presentation of a skeptical maxim about the in- despoiled, as we learn in the opera, through the stability of human erotic relationships (with an demonic intervention of Samiecl). However ac-

implied text-underlay toward the end, “Co- customed we may have become to such modal si fan tut-te”), then, one presumes, illustrates changes within introductions to standard-reperthe maxim with a madcap romp of constantly tory pieces, many of them were surely originally switched and intermixed rhetorical roles in the intended to convey the sense of something catamaterials of the subsequent sonata— whereupon strophic, and it is worth the effort to reawaken the maxim returns at the end (indeed, returns to the image of urgency, darkness, and alarm that accomplish the ESC) as the musical equivalent they present. That major-to-minor prefatory of “Q. E. D.” Mozart’s Overture to The Magic passage suggests that it 1s the task of the maFlute invites us to interpret the introduction as —_jor-mode sonata form that follows to repair the evoking a sanctified realm of Enlightenment problem, to restore a fully stable world of major. wisdom (and occasionally mysticism), then sug- This does not happen at once, with the forward gests that the goal of the ensuing sonata process, vector of the exposition’s major-mode P-idea. It and the following opera as a whole, 1s to prove can occur with full presence only at the end of worthy of entering “these sacred halls.” In all the essential sonata process, at the moment of such cases—even more abstract ones in nonpro- the ESC. When prefaced by this introduction erammatic (or less programmatic) works— one type, a sonata 1s portrayed as having significant

might understand the initial condition pro- tonal and modal work to accomplish throughposed by the introduction to persist conceptu- out its trayectory.*? ally throughout the sonata as its framing raison d’étre, that which, in the fiction or game at hand, The Negative (Minor-Mode) State of Affairs or Situmakes it possible and gives it a purpose for ex- ation to Overcome. This 1s similar to the above

isting at all. category, except that the introduction 1s entirely in the minor mode. The “fall” into minor—the “The Fall”/ “Fallen World.” Intersecting with the crisis from which one seeks to emerge — has

type above, this category presents an initially been set in place before the piece has begun. confident, positive, or serenely pastoral, ma- Once again, there may or may not be represenjor-mode world shattered by a sudden shift into tational implications, and once again the task the minor-mode negative (“lights-out’’), per- of the subsequent sonata— often in the major haps also with supernatural, ombra evocations. It mode—is to overturn the situation sketched is the overt demonstration of the loss of the ma- out in the introduction. Examples from Haydn jor mode that is crucial here. Additionally, the include the Symphonies No. 98 in B-flat, No.

minor mode is not a merely momentary expe- 101 in D, “Clock,” and No. 104 in D, “Lonrience within an otherwise major-mode intro- don.” Clementi prefaced his Piano Sonata in duction: once it enters it persists for the rest of | D, op. 40 no. 3 (1802), with a D-minor Molto the introduction. Classic examples include: Mo- Adagio. The starkest of examples (perhaps even zart’s Symphonies No. 34 in C, “Linz,” K. 425, involving the subsequent sonata’s denial of what and No. 38 in D, K. 504, “Prague”; Beethoven’s precedes it) is provided in the Adagio introducSymphony No. 2 in D, op. 36; and Weber (once tion to the finale to Mozart’s String Quintet in again), Overture to Der Freischiitz (perhaps the G Minor, K. 516.

archetypal representation of a “fallen world,” Before 1800 slow introductions to sonatas

40. Cf. the deformational variant of this category found form—a sonata that plays itself out in the submediant, in the first movement of Beethoven’s Cello Sonata No. A minor. In other words, op. 102 no. 1/1 presents a tonic

4in C, op. 102 no. 1. Here the Andante introduction introduction and a completely off-tonic sonata. The aim is sounded entirely in C major, ending emphatically of the entire piece is to work through a multimovement with the tonic chord in that key in m. 27. “The Fall” tonal process that will eventually restore the C major so away from this initial purity, however, happens in a flash astonishingly lost in the first movement. with the m. 28 downbeat of the Allegro vivace sonata

302 Elements of Sonata Theory

in the minor mode were less frequent, though resentation of preparing for and setting about they were not unheard of. In an earlier sec- the compositional act— although probably falltion we mentioned a handful of early examples ing short, in most cases (at least in this period), from piano sonatas in the 1790s by Clementi of being necessarily a direct self-representation. and Beethoven. (Is it significant that Haydn’s More obvious self-representations of precompoSymphony No. 95 in C Minor was both the sitional searchings for the proper theme, tone, only minor-key symphony in the twelve “Lon- and style were famously presented by Beethoven don” Symphonies and also the only to lack an at the openings of the finales of his “Hamintroduction?) After 1800, and certainly once merklavier” Sonata and Ninth Symphony. Selfplunged into more “Romantic” waters, the referentiality would become a familiar implicapractice becomes more normative. When the tion in several later nineteenth-century works sonata that follows is still in the same minor __ by other composers, sometimes with a sense of mode, the self-evident hope of that turbulent, self-questioning with regard to the aesthetic Allegro structure is to use the sonata process as obligation of the post-Beethovenian symphony, a vehicle to turn the minor mode into major and sonata, or quartet. A related variant, perhaps provide a major-mode ESC in the recapitula- worthy of a category in itself, is: tion. Should this fail, that hope is dashed, and the tonal and modal problem persists: 1t may be The Ceremonial Gathering or Assemblage of Forces deferred either into the coda or into subsequent Available for the Sonata Form Proper. The arche-

movements— especially a later finale—should typal example is found in Beethoven’s Seventh they exist. (Chapter 14 discusses this aspect of | Symphony. Here the introduction accomplishes the psychology of minor-mode sonatas.) The many tasks, laying out much of the harmonic, first movement of Beethoven’s “Pathétique” So- registral, and timbral space that will provide the nata, op. 13, provides a familiar example (one field and forces of the sonata that follows. (See without eventual resolution into the major), as also the category to this effect below.) Of interdoes his F-minor Egmont Overture (whose ton- est within this subcategory is its obvious impresic-major cadential resolution occurs only late sion of assembling and setting all of the avail-

into the F-major discursive coda). able material in order—by the end, everything is poised and ready, checked, double-checked, The Setting of a Tone of High Seriousness or Con- and triple-checked—before the composition 1s templative Absorption. A more generalized cat- sent off into the sonata form proper. Relatable

egory available in either major or minor, this to this general concept are the highly formalsuggests through its earnestness that what fol- ized, largely ceremonial introductions to many lows is the product of sober and significant of Rossini’s overtures. These often include a thought, as in Haydn’s Symphony No. 103 in lyrical, separate melody largely unrelated to the E-flat, “Drum Roll” (which at the opening sonata form that is to follow, with the whole ingives the impression of quoting the incipit of the troduction-formula giving the impression of us“Dies irae”). Here the implication can be purely ing ample time both to give a sense of the glit-

musical—reflecting on the act of composition tering prestige of the operatic occasion and to itself, music about the making of music. In such prepare the orchestra and the audience to settle cases the introduction may suggest the pres- down to undertake the onset of the real Allegro ence of the composer (either “in reality” or as action. One might imagine, for example, that a a staged aesthetic persona), absorbed in thought preliminary, first curtain lifts at the introducand about to produce not merely a sonata-move- tion; a second one at the start of the exposition; ment but a whole multimovement work. Under and a third one — the final one —at the opening these lights an introduction can be read as a rep- of the act.*!

41. Cf. Gossett, “The Overtures of Rossini,” esp. pp.

5-7.

Parageneric Spaces 303

The Searching, Foreboding, or “Mysterious” Intro- phony No. 103 in E-flat, “Drumroll,” are exduction. ‘This is another variant of the “contem- plicitly refashioned into an importantly situated

plative absorption” type, typically providing the rapid-thematic module of the later TR=FS impression that the ideas of the work — perhaps zone, mm. 74—79; and so on.)*? ‘These concerns

aspects of tonal assertion itself—are still in a are equally evident in Mozart and Beethoven. state of unclear formulation. The paradigms are Instead of belaboring the obvious, one might the openings of Mozart’s “Dissonance” String turn instead to some notable subtypes: Quartet, K. 465 and Beethoven’s Quartet in C, op. 59 no. 3. A related and instructive ex- The Generative Introduction (Producing the Pample (though not an introduction to a sonata) Theme). Here one might give a nod to those is the opening representation of “Chaos” from even more explicit situations in which the in-

Haydn’s The Creation. troduction, or at least the concluding portion of it, spawns, nurtures, or otherwise generates the The Source-Material for Much of What Follows. In theme that will be fully formed and launched this broad (and common) category with many with the sonata’s P-theme. Several of the Haydn

subtypes the introduction is perceived, ret- examples cited just above would qualify in any rospectively, as having furnished many of the broad sense, but sometimes the generative princentral ideas for the sonata-to-follow. As such it ciple seems to be more overtly thematized. exists as something of a caldron of motives, a re- Haydn’s Symphony No. 98 in B-flat provides a source or “inkwell” into which the composer’s classic example, mentioned earlier, as does the pen dips in the writing of the subsequent piece. finale to Beethoven’s First Symphony, whose Presented as such a broad category, this 1s an al- brief introduction “constructs” the P-theme, most-too-obvious introductory function. Many step by step, until it is ready to take flight. In-

introductions take on this task, which stresses troductions with a final accelerando link (also motivic consistency between it and the subse- mentioned earlier in this chapter) into the Alquent sonata, sometimes to the point of inau- legro—Beethoven’s Fourth Symphony, the surating a process of what Schoenberg would Leonore 1 Overture, and others—also usually later call “developing variation.” An introduc- participate in this general compositional logic. tion may foreshadow harmonic or tonal regions This type of introduction would become espethat will take on an added importance later in cially characteristic of later nineteenth-century the piece; it may anticipate certain features of | works: Schumann’s Symphony No. 4 in D Miregister, timbre, voice-leading, or other features nor, op. 120; Brahms’s Symphony No. 1 in C that will turn up in the sonata; or it may pre- Minor, op. 68; Franck’s Symphony in D Minor; sent embryonic versions of melodic contours, and many others. themes, or portions thereof that blossom more

fully in the ensuing Allegro. Direct Statements of Thematic Material to Come in Since demonstrating such relationships has the Interior of the Sonata. Here one is presented been among the most common staples of analy- with an idea (often, the “slow” version of an sis for several decades, we need not deal with it idea) that literally —or very nearly—will be at any length here. The procedure 1s certainly re-sounded more actively sometime after the characteristic of many of the introductions to onset of the exposition’s TR-space. When the Haydn’s twelve “London” Symphonies. (The theme in question is P, the introduction might opening contour of the introduction to Sym- be more properly considered generative of the phony No. 100 in G, “Miltary,” anticipates that idea that launches the sonata. Interestingly, the of the P-theme, as do those of the Symphonies theme in question is sometimes used later as S, No. 101 in D, “Clock,” and No. 102 in B-flat; as in Cherubini’s Overture to Eliza (1794); in the initial intervals of the introduction to Sym- Beethoven’s Overtures Leonore 2 and 3 (where

42. See n. 36 and the text to which it refers.

304 Elements of Sonata Theory

the incipit of S refers explicitly to the introduc- is especially characteristic of nationalistic works tory theme — Florestan’s imprisonment aria); of the later nineteenth century. Its larger impliand in Wagner’s Overture to Rienzi (where cation 1s obvious, standing ideologically for the the relevant melody, “Rienzi’s Prayer,” is de- deep and uncontaminated reservoir of ethnic esployed as S! in the exposition but suppressed in sence or Volksgeist—a quality to be understood the recapitulation). Similar instances occur in as generally untouched by (or otherwise precedHaydn’s Symphony No. 103 in E-flat, “Drum- ing) the present-day court-and-urban cultures roll,” mentioned above, and in Weber’s Over- of the powerful cities of Western Europe. As ture to Oberon, where the opening horn call re- such the vector of the motion from introduction appears in the sonata as the inauguration of a to Allegro is in part metaphorically both tem-

programmatic, expanded caesura-fill. poral and historical: “Out of our ancient and Introductory functions such as these per- ‘authentic’ reservoir will emerge the following sisted and were developed further in later nine- symphony, quartet, or sonata, genres that we 1n-

teenth-century works. Two offshoots of the tend to engage the urban contemporaneity and trope of the introduction’s establishment of the cultural prestige of the Western European pres“world” in which the sonata-to-follow plays it- ent. We, too, can now master and appropriate

self out might be appended here to provide a these things.” Such introductions are usually sense of what was to come much later. Neither —_ folk-like in one way or another, and may even of the two subcategories below are characteris- allude to a folk song (as does Gade’s Symphony

tic of the decades around 1800. They are typi- No. 1 in C Minor). Touchstone examples are cal, though, of the second half of the nineteenth obvious: Dvorak’s Symphony No. 8 in G, op. century and beyond, in the world of the sym- 88; Ichaikovsky’s Symphony No. 2 in C Miphonic poem, the (quasi-) programmatic sym- nor, op. 17; Glazunov’s Symphonies No. 2 in phony, and emerging “nationalistic” composi- F-sharp Minor, op. 16, and No. 4 in E-flat, op.

tion. 48; Sibelius’s Symphony No. 1 in E Minor, op. 39; and many others. Representation of the “Narrator” — or Perhaps the

Raison d’étre or the Animating Force— of the Tale .

Told in the Sonata. Something of shite is The Introduction-Coda Frame

present in Mendelssohn, Overture to A Mid- One striking deformation of normative practice summer Night’s Dream and, perhaps, in Berlioz’s was the introduction-coda frame, in which material Symphonie fantastique (the representation of the from the introduction returns as all or part of the

opium overdose as precondition for the rest of | coda. Examples before 1800 are sparse, but they the piece). But the concept reaches a different include the first movements of Mozart’s String level in symphonic poems that are about the Quartet in E-flat, K. 171 (Adagio— Allegro asrepresentation of well-known stories or the tell- sai— Adagio), his String Quintet in D, K. 593, ing of tales. One might sense the presence of a Haydn’s Symphony No. 103 in E-flat, “Drum represented narrator in several introductory sec- Roll” (here the recurrence is more of a defortions of Liszt’s symphonic poems (as if setting mationally ad hoc—and vast—expansion of a out to relate the narrative of, say, Tasso). It oc- C-module with pronounced CRI-effect than curs even more overtly in Tchaikovsky’s Romeo a literal coda, since the concluding portion of and Juliet (in which the strumming of the harp that last C-module, Allegro con spirito, returns even more characteristically calls up images to complete the movement), and Beethoven’s of a bard singing the age-old tale, even while “Pathetique” Sonata, op. 13 (whose Grave insurrounded by the Friar Laurence motive); in troduction had also preceded the opening of the Rimsky-Korsakov’s Sheherazade; and in Strauss’s development proper). Beethoven revisited the

Till Eulenspiegel. deformation in the first movement of his Piano Trio in E-flat, op. 70 no. 2 (1808).

The Lyrical-Expressive “Folk-Soul” or “Native The framing introduction-coda combination Landscape” Out of Which the Sonata Proceeds. This becomes more common in the later decades of

Parageneric Spaces 305

the nineteenth century. The first movement of — Night’s Dream (1826) is instructive with regard Schubert’s “Unfinished” Symphony comes close to the complications that could arise. The open-

to producing a perfect example, except that the ing four-chord introduction (and subsequent, in-tempo opening idea in question, mm. 1—8 sustained B4), mm. 1—7, function as a gateway (also occupying the bulk of the development to the sonata proper, which begins with the E-

and recurring, expanded, in the coda, mm. minor elfin scurrying at m. 8. The introduc328-68), is probably better regarded as a P? tion recurs as a bridge (mm. 394—403) famously module (discussed in chapter 5); nonetheless linking the C-sharp-minor end of the developthe psychology of the gesture 1s much the same. ment with the E-minor onset of the recapitulaEven clearer examples may be found in the first tion, a procedure also with rotational implicamovement of Schubert’s Symphony No. 9 1n C; tions. And it recurs as the overture’s final bars, in Mendelssohn’s Overtures to A Midsummer 682-86, providing an introduction-coda frame Night’s Dream and The Fair Melusine; in the first for the whole. What is curious, though, is that movement of Mendelssohn’s Symphony No. 3, the overture had already been provided with a “Scottish”; in some of Berlioz’s overtures, 1n- separate Allegro di molto discursive coda of its cluding Benvenuto Cellini; in Wagner’s Overture own, beginning in m. 620. Moreover, this coda to Tannhduser; and in many other works (in- has escaped outside of the “formally presented cluding such later “nationalistic” pieces as the piece” itself, which Mendelssohn apparently first movements of Tchaikovsky’s Symphony wants us to imagine has been brought to a conNo. 2 and 4, the finale of his Symphony No. 5, ventional (coda- or codetta-like) end with the and his Overture “1812,” in which the massive generically emphatic, “curtain-down” chords introduction and coda dwarf the relatively small of, say, mm. 608—20—as though, once the

but vigorous sonata deformation within). merely human drama and assorted love-tangles Whenever we find an introduction-coda have been sorted out and brought to a close, frame the interior sonata seems subordinated the attention shifts to the supernatural elves and to the outward container.*? The introduction fairies, who pre-existed the piece and will conand coda represent the higher reality, under __ tinue to thrive beyond the confines of whatwhose more immediate mode of existence — or ever the human personages imagine to be their under whose embracing auspices—the sonata own “story.” The broadening-out of the music form proper is laid out as a contingent process, at the end, and the final chords, suggest the cona demonstration of an artifice that unfolds only clusion of a series of multiple coda-sections or under the authority of the prior existence of the programmatic actions, rounded magically at the frame. Metaphors of narrativity are not inevita- end by the four chords with which the piece had bly implied —the external narrator and the tale begun. Mendelssohn’s later revisiting of much told— but in some cases they can spring to mind of this music in the complete incidental music and appear to be hermeneutically relevant.*4 from 1843—which includes a verbal text from The extraordinary conclusion of Men- Shakespeare — helps to clarify the expressive 1ndelssohn’s E-major Overture to A Midsummer tent of all of this.

43. The introduction-coda frame was also occasionally 44. The introduction-coda frame has also been menapplied to works that were not organized as sonatas. tioned with regard to later repertory in Hepokosk1, See, for instance, Weber’s Invitation to the Dance (Auffor- “Beethoven Reception: The Symphonic Tradition,” derung zum Tanze), rondo brillant in D-flat, J. 260 (1819, in Jim Samson, ed., The Cambridge History of Ninepubl. 1821); Beethoven, Bagatelle in E-flat, op. 126 no. teenth-Century Music (Cambridge: Cambridge University

6 (1823-24; here the normative tempi are reversed: an Press, 2002), p. 451; and treated at greater hermeneuimpetuous Presto frame surrounds an Andante amabile tic length in Hepokoski, “Framing Till Eulenspiegel,” e con moto interior); the Larghetto slow movement of 19th-Century Music 30 (2006), forthcoming. Chopin’s Piano Concerto No. 2 in F Minor, op. 21; and Mendelssohn’s Lied ohne Worte in A Major, op. 19b, no.

3 (1829, publ. 1832—the so-called Jagerlied).

CHAPTER FOURTEEN © "ESTO

Sonata Form in Minor Keys yY

The Extra Burden of Minor-Mode Sonatas reaffirm the minor mode at the ESC and beyond. What matters at this level of considerVirtually all of the guidelines for sonata-form ation is not whether the initial tonic minor 1s movements in the major mode hold true for converted into a stable tonic major at or around those whose initial tonic is minor. Here, too, the ESC but rather that minor-mode sonatas, the sonata process involves structural points unlike major-mode ones, are uniquely capable common to all sonatas: two-part or continuous of effecting this modal transformation. This 1s expositions; P TR’ S / C zones in two-part the extra burden under which minor-mode soexpositions; trajectories toward the EEC (expo- natas are placed, regardless of the results of the sition) and ESC (recapitulation); the possibility minor-major musical drama engaged.

of a dialogue with the rotational principle; and The possibility of a tonic-minor-to-tonicso on. In these respects the morphology of the major trajectory (or the represented inability to minor-mode sonata is analogous to that of the attain that transformation) 1s rich in metaphori-

major-mode sonata. cal implication. If we understand sonata form There is, however, a crucial difference be- as a metaphor for an idealized but nonspecific tween them. In addition to articulating the rhe- human action (chapter 11), minor-mode sonatas torical shape familiar from a major-mode sonata _ provide the means by which an initially negative form, a minor-mode sonata bears an additional state (the minor mode) is acted upon in order to burden. This is that of the minor mode itself, seek to overturn it by means of major-mode asgenerally interpretable within the sonata tradi- sertion at or around the ESC point, even though tion as a sign of a troubled condition seeking that quest might be unsuccessful. Minor-mode transformation (emancipation) into the parallel sonatas contend with the initial presence of the

major mode. Since many minor-mode sonata tonic minor—often a turbulent or threatenstructures do attain a major-mode ESC in the ing expressive field—either to overcome it or recapitulation and do sustain that major mode to be overcome by it. Composers turn to the for the rest of the composition, the sonata pro- minor-mode sonata to project an either successcess can function as a strategy capable of trans- ful or unsuccessful modal action— attainment forming tonic minor into tonic major. It 1s true, or failure—even though all other criteria for of course, that some minor-mode sonatas are the sonata process (such as the production of an dominated by the minor throughout or sternly adequate ESC) are to be satisfactorily met. From 306

Sonata Form in Minor Keys 307

this perspective, minor-mode sonatas present great composers of the eighteenth and nineteenth composers with additional opportunities for centuries believed in this association is evident to

varied realizations. anyone who studies their songs and other music they composed to texts.!

Major and Minor as Binary Opposites Once one takes an interest in the ways 1n which

(Positive and Negative) that mode has been understood or explained

over centuries, one may sidestep the vexed Perhaps because they are sometimes based on question of whether the minor mode 1s “really” unexamined assumptions — or perhaps because imbued with these expressive implications.? We the subjects are often handled in an overly el- —-T€™ain content to regard these affective properementary fashion—even modest claims regard- ties as the product of social custom, as cultural ing the affective qualities of the major and mi- matters or tacit agreements within interpretive

nor modes can raise warning flags within the communities, agreements that were reinforced analytical community. Surely it would be in- over many years. cautious to reduce the nuances of the major and On both historical and experiential grounds, minor modes into simplistic descriptions. Major —1¢ 1 absurd to suggest that within the post-1750 is not consistently collapsible into “happy” nor style the compositional usage of minor or major

minor into “sad.” Aldwell and Schachter ad- was objectively neutral, a matter of little condressed this issue and went on to make a more sequence, a creative choice made arbitrarily, in

compelling point: the abstract, “for (unexplained) reasons of con-

trast,” or for merely formalist reasons. On the Sophisticated musicians often question this asso- contrary, this modal dichotomy was one of the ciation, believing that it is a purely arbitrary one basic binary oppositions of tonal music. Daniel

based on nothing except, perhaps, habit. And of | Harrison recently explored this issue in a closely course it is true that the emotional character of a argued monograph on the topic, proposing “a

piece depends on many factors in combination. renewed dualist theory,” particularly with reLight and even comical pieces—some of Men- gard to the analysis of chromatic music of the delssohn’s scherzos, for instance —are in minor. late-nineteenth and early-twentieth centuries. And some very solemn pieces are In major, tor As Harrison put it, in Western tonal practice the wma le the Dead March . Handel s Saul But major and minor systems “can be [and typically

it isa mistake to ignore the likelihood that choice . ;

of mode is one of the factors that determine the were] organized and developed mm two opposed character of a piece. And sometimes it may be the directions, creating sets of dualistically paired

most important factor. tonal concepts”; “the dependence of tonal music For one thing the association of mode and and its theory on the contrariety of major and emotion is a very old one... . Writing in 1558, minor is far reaching”; music analysis should Gioseffo Zarlino, the greatest theorist of the late recognize them as “a basic duality,” “a virtual Renaissance, remarks that melodies (and modes) primitive of tonal music,” “a fundamental dufeaturing a major 3rd above the central tone sound alism” that led to dual modal “networks” or cheerful and that those with a minor 3rd sound “systems” possessing different implications for sad. Any cultural tradition that has persisted for their associated scales, chords, and tonal relaSO long takes ona certain importance even if it tionships. The affective impact of many pasis based on nothing more than custom. That the

1. Edward Aldwell and Carl Schachter, Harmony and ousticians. .. . [On the basis of this we conclude that] Voice Leading, 2nd ed. (San Diego, New York, etc.: in major-minor tonality, the major mode 1s normally Harcourt Brace Jovanovich, 1989 [1st ed., 1978]), pp. the positive, happy, bright one, and the minor 1s the

19-20. negative, sad, dark one. This again points to the greater

2. Cf. Aldwell and Schachter, Harmony and Voice-Lead- stability of the major triad.” ing, p. 26: “Nowadays most musicians would maintain 3. Daniel Harrison, Harmonic Function in Chromatic Muthat the foundations for music theory should le in the sic: A Renewed Dualist Theory and an Account of Its Precworks of great composers, not 1n the laboratories of ac- edents (Chicago: University of Chicago Press, 1994);

308 Elements of Sonata Theory

sages resides in the composer’s skillful use of the “unnaturally” within a psychologically negachiaroscuro shadings made possible through this tive or minor-mode context that seems othantithesis, coupled with a manipulation of the erwise overwhelming. In such circumstances musically narrative implications of those shad- one should be attuned to the possibility of a ings. Needless to say, musical paragraphs gov- chillingly ironized or “false major”: the poserned by one of the modes are often mixed with ture of sheer bluff or cruel deception; the piti-

borrowings from the other. able embracing of an illusion or an emotional No single emotive description suffices to mirage; the projection of something intensely describe the general sense of major or minor. desired but no longer available; the deep sigh According to the circumstances, the major of “if only” amidst a situation of profound loss. mode may underpin expressions of energetic Such expressive situations denaturalize the maassertion, accomplishment, brightness, joy, sta- = jor mode. The major is now framed ironically, bility, contentment, pastoral relaxation, eroti- “from outside,” as a false front not properly accized (or noneroticized) lyricism or “sensibil- cessible in the context supplied by the 1immeity,” confidence, exuberance, majestic splendor, diate musical or aesthetic surroundings. One marchlike vigor, radiant wonder, and many might interpret the much-discussed, seemingly

other emotional states as well. Similarly, the frivolous G-major finale of Mozart’s String minor mode may support representations of | Quintetin G Minor, K. 516, under these lights. melancholy, “sweet and tender” sadness,+ ma- The concept may also apply in varying degrees

jestic gravity, oppression or threatening cir- to major-mode slow movements within micumstances, anxiety, rage, defiance, storminess nor-mode-domiunated pieces. Localized appear(Sturm und Drang), sorrow, resignation, fatalism, ances of the denaturalized, even “pathetic” maerief, darkness or shadows (as with the ombra — jor mode are particularly associated with such topos), instability, the unnatural, demonic, an- psychologically troubled and complex music as tique, or exotic (as with certain tints of alla turca that of Schubert (as in the sudden turn toward

music), and so on. the “false major” in the final stanza of the first

But even these generalized descriptions can song from Winterreise, “Gute Nacht’), Brahms, be insufficient when coming to terms with the and Mahler. Considerations along these lines expressive impact of an individual passage— or also provide an approach to the classic illustra-

even an entire movement—1in major or mi- tion of the major mode deployed in sorrowful nor. Any specific effect is conditioned by the circumstances, “Che faro senza Euridice” from local context in which it appears. The charac- Gluck’s Orfeo ed Euridice, an example commonly

teristic moods of the minor mode, somewhat trotted out by critics and philosophers, from more consistent, even in their variedness, may Hanslick onward, skeptical of the proposition be less problematic in this regard. Within the that certain emotional qualities are properties e1ghteenth-century style the minor mode is inherent to the major and minor mode. typically a special negative condition operat- Nothwithstanding these caveats, on the most ing as a conventionalized exception to the more general level we refer to major and minor as binormatively prevailing major mode. Assessing nary signs of the positive and the negative (light the character of a passage in the major mode, and dark or, colloquially, “lights on” and “lights

though, can be more difficult when it appears out”). Eighteenth-century theorists (Christ-

quotations from p. 15 (“can be organized”); p. 17 (“the 4. As mentioned (citing Rameau and other eighteenthdependence,” “a basic duality,” and “a virtual primi- century theorists) in Gretchen A. Wheelock, “Schwarze

tive”); pp. 22—23 (“a fundamental dualism”, “net- Gredel and the Engendered Minor Mode in Mozart’s works,” and “systems’’). “Saying that a composition is Operas,” in Ruth A. Solie, Musicology and Difference: in ‘major, for example, references qualities of scale, of Gender and Sexuality in Music Scholarship (Berkeley: Uni-

primary chords, of modulatory structures, etc. Yet it versity of California, 1992), p. 207. also references something larger: a set of relationships that links these items into a system” (p. 23).

Sonata Form in Minor Keys 309

mann, Rousseau, Kirnberger) confronted this female, comic and serious characters alike, epibinary opposition in predictable ways. Summa- sodes in the minor mode expose the darker side of ries of some of these discussions — consistently human passions, memories, and actions. In reachcharacterizing the minor mode as depressive, ing to universal fears of irrationality and death,

; the elusive and mutable mode -. . . [had] audible sad, hollow, unsettling, lessto natural or complete . ; - powers destabilize tonal order and to evoke

than the major, hesitating, indecisive, so 8 disordered states of humanand consciousness. on— have been provided in studies by Ratner,

Steblin, and Wheelock.>® Ratner’s conclusion:distincoo, , . 7 The custom of making interpretive

From these comments and others, the rhetori- ; . , vides less the majorHelmholtz, in periodicOettingen, definitionRie.. . ., than (Hauptmann,

tions between the the opposites, major and cal implications area clear: minor mode . . minor, —— continued throughout the pronineteenth century

and confirmation key.’ . 19 .upon . . oo oo mann) into theof twentieth.? Adding to the mixand of implications, in hisExpanding ._.. . . the earlier tradition— still with an ear turned

Grundregeln zur Tonordnung insgemein of 1755 Jo- cots toward the “scientific” certainty of the overtone

seph Ruiepel referred— apparently with humor. . characseries —the two modes were sometimes ous intent— to the social (and gendered?) status , oo. gk . 7 , terized binaries “natural” (following of the parallel minor ofasathe “master” [Meyer] major « > ct aesthe .

. . law ofschwarze nature’) Gredel, and “artificial” (not exfound in tonic as. nature the or, in Ratner’s oe —and therefore problematic). Such was planation, “black Margaret—a local nickname . . .. Fa ‘ , lex the basis, forswarthy instance, complexion of the extended discussion or Swedish a Swedish queen whose . Mtlook . y P"like at the opening In of Schenker’s Harmony (1906): made her a man.”” Wheelock’s later « ..a . —- The minor mode springs from the originalinterpretation “the power of [Riepel’s] schwarze . —_ ity of the artist,. whereas the sources, at least, of Gredel is not far to find: unnaturally feminine/ , . the major mode flow, so to speak, spontaneously

masculine, the Master’s tonic, and perspective, 10 X72 , , . a . she fromshares Nature.”!° Viewed from this her chromatic mutability is capable of destabi. . _. . oo minor-mode sonatas Seeking begin under the condition lizing his natural domain.” to enlarge . . . 7 —_ . of artifice and, as part of their generic burden,

contemporary associationsinto of minor keys withcondioo. . . . seek an emancipation a more natural erieving passive gloom,” she de. ey: _ eerand tion, even though that which quest might fail.within scribed as “simply limited,” Wheelock explored ra .

7 ,“deeper any individual sonata narrative. the rich, ambivalence” of this often ae . . “fami nized” mode. th Writing in the mid-twentieth century, The-

eminized” mode, the _ . . . . jor-minor polarity in his discussion of a muchodor W. Adorno captured this condition of ma-

seemingly disparate constructions of the minor , ‘[inlater repertory, that of Mahler’s compositions. mode the eighteenth century|—as weak and a To be sure, in the passage quoted below Adorno passive in affect, but also as powerful and subver- born; 1 oath sive in function. ... [In Mozart’s dramatic works] was re erring to a ate stage in ¢ ¢ conceptuthe minor mode was a vehicle for often menacing alization and effectiveness of the minor mode: forces, the expression of which threatened to ex- in fact, he was insisting that the traditional unceed the bounds of the ‘natural’... . For male and derstanding of the minor mode (its generally

5. Leonard G. Ratner, Classic Music, pp. 50, 55-56; rad Wolff, trans. Paul Rosenfeld (New York: Norton, Rita Steblin, A History of Key Characteristics in the Eigh- 1946), p. 60.

teenth and Early Nineteenth Centuries (Ann Arbor: UMI 8. Wheelock, “Schwarze Gredel,” pp. 203, 205, 219Research Press, 1983); Wheelock, “Schwarze Gredel,” 20.

pp. 201-21. 9. Some of these “dualist” theoretical positions are re6. Ratner, Classic Music, pp. 55-56. viewed in Harrison, Harmonic Function in Chromatic Mu-

7. Ratner, Classic Music, p. 50. Compare these con- sic, pp. 215-322. siderations with the remark of Robert Schumann in 10. Heinrich Schenker, Harmony, ed. and annotated by 1834: “The difference between major and minor must Oswald Jonas, trans. Elisabeth Mann Borgese (Chicago: be allowed beforehand. The former 1s the active, virile University of Chicago, 1954, rpt. 1980), p. 52. Cf. n. principle; the latter, the passive, the feminine.” Quoted 2. in Robert Schumann: On Music and Musicians, ed. Kon-

310 Elements of Sonata Theory

implied effect within the tradition) had dete- second-level default key, minor v, discussed sepriorated and existed primarily as a loss that was arately below.) From a Schenkerian perspective, exposed in Mahler’s anguished music. Still, the the normative tonal task of the development is distinction between the modes and the interplay to move from this stabilized III to V,, (an “acwithin any structure that depends upon its pre- tive” V functioning as a dominant chord within sentation as the affective “divergence” or “de- the principal tonic), which will usually be harviation” [Abweichung| from the more normal- monically interrupted to rebegin the recapituized major 1s stated with uncommon force here. lation on the tonic minor (i). The large-scale Once again, we do not insist on a unitary or motion up to the structural harmonic interrupreductive meaning to the minor mode in any tion is still from tonic to dominant (i-III-V), period. Nevertheless, Adorno’s remark affords but the dominant is not reached until the end an insight into generic issues in minor-mode so- of the development, and the path from 1 to V natas written a century or more before Mahler: is bisected by the third-divider III, attained at the EEC and sustained throughout the rest of [In Mahler’s hands] the long neutralized minor, the exposition. The III at the close of the exsedimented as a formal element in the syntax of position is thus only a temporary point of rest, Western music, only becomes a symbol of mourn- a way station on the path to the more compeling when modally awakened by the contrasting —_ Jing tonal goal, the structural V. The resulting

ae | Its aa me oe everson "4 isolation arpeggiation 1—III—V outlines the tonic minor

EMO Tonger Procuced TAs erect. AS a ceviaon, triad. (See the section that concludes chapter 7,

the minor defines itselfSchenkerian equally as the not inte- ce , Lo, > _. . Some Implications.”)

erated, the unassimilated, the not yet established. Th eth ... Minor is the particular [that is, that which € customary CXPTESSIVE connotation oF the

strives to be assimilated], major the general [that exp ositional move from 1 to III in a minor-mode which assimilates or is taken as the norm]; the sonata is different from that of the standard move Other, the deviant, is, with truth, equated with | from I to V in a major-mode one. This is due suffering. In the major-minor relationship, there- to the negative implications of the movementfore, the expressive content is precipitated in sen- governing minor mode. (For the sake of clarity suous, musical form. .. . Mahler’s minor chords, let us confine our discussion to the most com-

disavowing the major triads, are masks of coming mon expositional rhetorical type, the two-part

dissonances. exposition. The principle at hand 1s easily extended to continuous expositions.) In normative

sions in th , major-mode sonatas both parts of the exposition EEC in IIT mode tonalities. These two parts (tonally, I and - . V) may be minor-mode modally characterized as major-maThe expositions of most sona-{+Ls+}, a even . jor, or positive-positive, though local Expositions in t € Minor Mode: (P TR’ and S / C) are underpinned by major-

ta-form; _movements inmusic the decades around 1800ofLo. .. stretches of might show signs minormove from ito III, the tonicized key of the ma- ; ;

mode mixture (which not infrequently happens jor mediant. C-minor sonatas go to E-flat 4: at the approach to the within major; TR —especially D-Minor ones to F major, and so on. ‘This move a:S). :; The psy. .isMC, or even within portions of tokevechoice III a strong first-level default with regard ; P 4; herProceeding k chology of first-level-default to key-choice. to any otherminor-mode key area sonaee .

y oy 8 y y tas is different. Here the two parts of the exposi-

for the exposition’s part 2 overrides the muchtion on.(1 and IIJ—and here by part 1 we generally

more ly normative choiceP,and suggests amodal set ofoptions un- . .are posforceful . mean since various commonly forceful expressive circumstances. . . 6) are arY a P — sible for TR, as mentioned in chapter (This observation includes an establishing of the

11. Theodor W. Adorno, Mahler: A Musical Physiog- Eine musikalische Physiognomik [1971], rpt. in Adorno, nomy, trans. Edmund Jephcott (Chicago: University of Die musikalischen Monographen (Frankfurt am Main: Chicago Press, 1991), p. 26. Orig. published as Mahler: Suhrkamp, 1986), pp. 174-75.

Sonata Form in Minor Keys 311

rayed as minor-major, or negative-positive, {— +}, tial field of the minor mode is represented as so

even though moments within the structural coercive that it seems not to permit the generic zones may include intermixtures with their op- modulation at all within part 1, resulting in a posite modes. While the I—V move in a ma- 1:HC or highly rare 1:PAC (or 1:1AC) MC, jor-mode exposition generates tonal tension as which defers the task of modulation either to a central expressive effect, the i—-III move in the a passage of caesura-fill or to the first modules

minor-mode exposition produces the impres- of S. Such is the case with the first movement sion of temporarily escaping from the troubled of Mozart’s String Quintet in G Minor, K. 516,

minor mode into the major. which may be compared with Schubert’s treatThe desire to be emancipated from minor ment of the tonic-key authentic cadence MC into major constitutes the basic narrative para- in the opening movement of the “Unfinished” digm—the extra burden— of minor-mode so- Symphony, in which the modulatory/modal nata form. Within this paradigm a minor-mode task is accomplhshed by caesura-fill. exposition can offer the promise of modal release Regardless of the manifold ways in which it by electing to follow the first-level tonal default, may be accomplished, the point of a typical exproceeding to the mediant major for part 2 and position within a minor-mode sonata is to proits EEC. A more stable emancipation from the duce the escape into major and to secure that tonic minor, however (a “truer” liberation), can major mode with the PAC that accomplishes be effected only by a lasting conversion into the the EEC (and, of course, to secure it even furtonic major. But within the genre that conver- ther— or to celebrate its being secured— with sion can only occur—if at all— within the re- C). This is the expressive goal of the exposition, capitulatory space, wherein the modal aspect at but since the EEC 1s sounded in III, not in I, the

and around the ESC will be telling. problem of the minor mode is not yet solved. This {— +} modal conversion in the exposi- It will be addressed only in the recapitulation, tion is accomplished with the generic knowl- when (hopefully) the parallel moment will proedge (the shared understanding of the genre duce the ESC in the tonic parallel major, I. In within the interpretive community) that, as a minor-mode sonata the exposition represents a nontonic space, the major mode supporting the building of a structure of promise, a strucS and C can exist only temporarily. In multi- ture that, when it reappears in the recapitulamovement works with expositional repeats the tion, will manage to do what the exposition exposition’s major III will almost immediately could not do: decisively emancipate the tonic vaporize —and thus be shown to be provision- minor by converting it into the parallel major. al—by being cast back at the repeat sign (some- The expressive promise of part 2 of such expositimes through an active dominant) to the tonic tions suggests something like: “This is how the minor, that is, to the original modal condition, victory/relief/liberation will be secured in the with all that that might imply metaphorically. recapitulation-to-come. This is how the sonata In addition, within the norms of the genre it narrative will rid itself of the ‘imperfect’ or “speis understood that the major III at the end of — cial-condition’ minor mode (although it still has the exposition will be replaced with other to- to pass through the development— which may nalities both in the development-to-come and, prove to be an arduous process).”

most important, in the recapitulation. Since the attaining of the major mode 1s How long the initial, negative-minor cond1- so important an expressive event, any erosion tion will prevail within part 1, the P and TR away from that major within the exposition’s zones, varies from piece to piece. In the nar- part 2 (S / C)—predictive of the material-toratives offered by some compositions the mi- come around the recapitulation’s ESC—1is a nor-mode conditions of the onset seem intol- slenificant occurrence within the narrative at erable, and the music seeks to leap out of the hand. Of particular consequence 1s any deflation minor as soon as possible {— +}—generally of the expected major-mode III to minor-mode around the onset of TR, as discussed in chapter iui— a “lights-out” effect or perhaps an unsta6. Conversely, in a few cases the threat or ini- ble, back-and-forth flickering between major

312 Elements of Sonata Theory

and minor— either at the beginning of or to- 2 (S / C) had been {-— +}, delaying the exward the end of the S theme (or S/C complex). pected generic production of the emancipatory In the first movement of Beethoven’s “Pathé- major mode, the “Appassionata” first movement tique” Sonata in C Minor, op. 13, the initial reverses the procedure into a more harrowing

minor-mode conditions (announced at the representation of hope extinguished: {+ —}. onset with a non-normative, perhaps funereal The “Appassionata” model— marked by a mod-

slow introduction, Grave) last considerably ally decaying S—would recur in later ninelonger than generically expected. Following a teenth-century works (with some additional IHI:HC MC in m. 49 (the actual mode at complications) —for instance, in part 2 of the the MC, whether III or 111, is left uncertain), exposition of the first movement of Brahms’s Beethoven provides us with a prolonged S® Symphony No. 1 in C Minor, op. 68, which theme, mm. 51—88 (over the dominant— also proceeds through E-flat major to close in E-flat discussed in chapter 7) that, taken as a whole, muinor.!? expresses V/ii1, E-flat minor. An evaded PAC at m. 89 (avoiding the more concretized EECeffect in minor) suddenly clarifies the mode into Recapitulations and Codas in Sonatas with

the “natural” or “emancipated” E-flat major Expositions of the i— III Type (III) for the next module of S, beginning in that

m. 89. After a few additional evaded HI:PACs During the recapitulation of the i—III type of the EEC is finally articulated in the major mode, minor-mode sonata the composer has two opIII:PAC, in m. 121. A brief C-theme (the ge- tions. Earlier mediant-major material may be neric forte P-based-C, often an indicator of the recapitulated either in the tonic major, thereby onset of the closing zone) is elided to this in preserving the original mode of the exposition’s m. 121. Here this C proves unable to confirm part 2 (S/C) and achieving the minor-major the shift to major mode with its own cadence. emancipation (one “wins’’), or in the tonic miInstead (pathétiquement), the major-mode EEC- nor, thereby altering the mode (and the entire promise runs aground and hurtles back to an character of the material) and denying the miexpositional repeat with its initial C minor. nor-major emancipation (one “loses”). Once The first movement of Beethoven’s “Appas- the recapitulatory S begins in the tonic major, sionata” Sonata in F Minor, op. 57, adopts a re- the local assumption 1s that the positive outlated but different strategy. S emerges confidently | come 1s in the offing, and indeed, if S is strictly

in II, A-flat major (m. 36) but soon runs into a transposed, a major-mode ESC will happen harmonic catastrophe involving mixtures with automatically. Many sonatas project a sense of the minor mode, mm. 42—43. This is the point confidence with the onset of a recapitulatory, at which the structure of promise—the intro- | major-mode S and are able to glide smoothly duction of the major mode — collapses like a into that ESC. Others suggest that the task of house of cards. What is probably best regarded — sustaining the major mode throughout S will as an S© theme (also mentioned in chapter 9) be difficult, perhaps impossible. The first moveregisters this breakdown in minor-mode iii, A- ment of Beethoven’s Ninth Symphony provides flat minor (m. 51ff) , thus predicting (accurately) a trenchant object lesson in the shattering of inia modally devastating recapitulation, which will tial recapitulatory hopes, displaying en route an fatalistically reconstruct the expositional calam- eventually permanent decay of the multimoduity and end (m. 204) in the tonic minor, com- lar S into the minor mode. On the other hand, if pleting a sonata-tale of stern failure. Whereas — the once-major S begins in the minor mode, the in the “Pathétique” the modal quality of part | expectation is that the entire S/C block is likely

12. In the nineteenth century the “Appassionata” Brahms’s Symphony No. 3 in F Major, op. 90, part 2 of model, as characterized above, was not limited to mi- whose exposition decays from A major to A minor. nor-mode works. See also, e.g., the first movement of

Sonata Form in Minor Keys 313

TABLE 14.1. Modal Options in Major- and Minor-Mode Sonatas + = major mode — = minor mode The two signs within the braces refer to parts 1 and 2 of a two-part exposition: especially the onset of P and

all of S/ C.

Exposition Recapitulation

1. Major-Mode Sonata {+ +} {+ +} 2. Minor-Mode Sonata {- +} {— +} I-V

i-III option with positive recapitulation

3. Minor-Mode Sonata {— +} {——} i-III option with negative recapitulation

4. Minor-Mode Sonata {— [- +]} {—[- +]} i-III option with mixtures in part 2:

positive outcome within sonata-space

5. Minor-Mode Sonata {— [+ —]} {—[+ -]} i-III option with mixtures in part 2: negative outcome within sonata-space

6. Minor-Mode Sonata {— +} {—[- +]} 7. Minor-Mode Sonata {— +} {—[+ -]} 8. Minor-Mode Sonata {—-—} {——} i-III option with mixtures only in recapitulatory space: positive outcome

i-III option with mixtures only in recapitulatory space: negative outcome I-v option

to be minorized throughout (although further emancipatory or redemptive coda. Conversely, modal changes may also occur). Table 14.1 fur- a recapitulation that concludes in major may nishes a fuller review of some of the possibili- be undercut by a negative, minor-mode co-

ties. da—darkly pessimistic in its implications —as In the negative, minor-mode case (minorized in the first movements of Beethoven’s String S, at least at the EEC-point, usually followed Quartet in C Minor, op. 18 no. 4; Piano Conby a minorized C), nos. 3, 5, and 7 in table certo No. 3 in C Minor, op. 37; String Quartet 14.1) the emancipatory paradigm has been in E Minor, op. 59 no. 2; Symphony No. 5 in C unfulfilled, and for this reason we may speak of | Minor, op. 67; String Quartet in F Minor, op. one type of sonata-process failure. (This is not 95 —and (to cite a non-Beethovenian example) a compositional shortcoming on the part of the Schubert’s Symphony No. 8 in B Minor, D. 759, composer. Rather, it indicates that the musical “Unfinished.” tale told is that of a tragedy, or at least one that There 1s little more powerful or more afends in failure or sorrow —an inability to over- fecting within minor-mode sonatas of the 1—III come the negative or special-effect conditions type than the bleak realization that all of part prevailing at the opening.) To adapt the lan- 2—sounded in major in the exposition— might guage of Kerman, such a recapitulation is sub- come back entirely in minor in the recapitulaject to “minor-mode saturation.”!3 A return to tion. To sound all of part 2 in minor 1s, beat-bythe major mode may take place, however, in an beat, to cancel out the hopes raised 1n the expo13. Kerman, “Beethoven’s Minority,” in Write All These Here Kerman was referring not to recapitulations but to Down: Essays on Music (Berkeley: University of Califor- minor-mode sonatas that move to the minor dominant

nia Press, 1994), pp. 217—37 (quotation from p. 222). in the exposition.

314 Elements of Sonata Theory

sition: a moving wave of despair passes through sued throughout for the exceptionalism of C this music, inexorably reversing former hopes. minor within Beethoven’s works. There the It is helpful to remember that part 2 represents minor-mode redemption paradigm often grew the “active” part of the sonata—the part that into the per aspera ad astra narrative trajectory should secure what the sonata sets out to accom- (heroic struggles concerned with moves from plish. Within the fictional narrative usually pre- darkness to light, sickness to health, suffering sented in a minor-mode sonata, one’s hopes are to redemption) so important also to later nineunderstood to reside in the major-mode part 2 teenth-century composition. (S and C). To witness these hopes dashed in the This is not to say that Beethoven inevitably recapitulation is one of the most telling things preferred to bring back S and C in major (or that a composer can show us. This negative vi- with some sort of expressive major-minor m1ixsion is the one invariably presented to us by Mo- ture, eventually releasing a minorized S into zart in his minor-key sonata form movements.!4 major or vice-versa). In some i—III composiHaydn, on the other hand, 1s usually eager to get tions of particularly tragic import S comes back to the major mode quickly within these reca- entirely, or nearly so, in the tonic minor. These pitulations—sometimes at the earliest available include the first movements of the Piano Soopportunity. To be sure, there are a few notable nata in F minor, WoO 47 no. 2 (along with its cases in which the minor mode persists or even last movement as well), the Piano Sonata in F prevails in Haydn, but they are less common. Minor, op. 2 no. 1, the String Trio in C Minor, (One instance occurs in the first movement of op. 9 no. 3, and the Piano Sonata in C Minor, the Symphony No. 44 in E Minor, “Trauer.’) op. 10 no. 1 (after an initial nontonic feint in IV, Beethoven presents us with a more compli- F major), and the Piano Sonata in G Minor, op. cated situation with regard to these matters. In 49 no. 1. Even bleaker and more minor-mode a recent study devoted to Beethoven’s key re- saturated, however, are the several Beethoven lations in minor-mode sonatas, Kerman noted works that have expositions of the 1—v type. that Beethoven is more inclined to use the 1—III

exposition type in C minor than he 1s in other minor keys, and that C-minor works nearly al- Expositions in the Minor Mode: EEC in v

ways feature a prominent section of C major (the Minor Dominant) somewhere near the end (either in the sonata itself or in the subsequent coda). “The tendency In the later eighteenth century and extending of works in C minor to break into C major 1S shortly into the nineteenth, the main alternasomething we take for granted, perhaps, because tive to the 1-III move in the exposition was one of the Fifth Symphony. It is still rather remark- into the minor dominant, i-—v: C minor to G able to see this tendency played out on one level | minor, D minor to A minor, and so on. Duror another in every single one of Beethoven’s ing this period this pattern may be regarded many works in this key. It is doubly remark- as a second-level default with regard to keyable when the other minor-mode works [which choice. It was still a recognized generic option, are less inclined to share this minor-major con- —- one with deep roots in the tonal past, but one cern] are also brought into consideration.” “At | much less frequently chosen. (Before too long

some point, early on, Beethoven hada... vi- into the nineteenth century it would become sion of troubled C minor ceding to serene C an even less standard choice, used only for exmajor. The vision haunted him.’!5 Kerman ar- traordinarily negative tonal narratives.!© At least 14. Data along these lines are provided in Rey M. 16. Foran example, see the first movement of MendelsLongyear, “Parallel Universes: Mozart’s Minor-Mode sohn’s Symphony No. 3 in A Minor, op. 56, “Scottish,” Reprises,” Mozart-Jahrbuch 1991 (Kassel: Barenreiter, in which the EEC occurs in E minor, m. 181, and is

1992), pp. 810-15. reinforced with an E-minor C-space.

15. Kerman, “Beethoven’s Minority,” pp. 223, 230. Part of the essay traces Beethoven’s increasing “clarification”

of this vision throughout his works.

Sonata Form in Minor Keys 315

as common by the 1820s and beyond would be modal “fate” is decreed in advance. All exits a modulation to the major-mode submediant, are blocked. We still encounter a structure of i—VI, discussed at the end of this chapter.) promise —all expositions seek to forecast the In some respects the i—v option mirrors the procedures of their recapitulations — but what tonal course of a major-mode sonata, whose is promised here is an inescapable recapitulaexposition conventionally moves from I to V. tory negativity. Typically, the remainder of the In both types the development normally deto- sonata experience unfolds as something to be nicizes this dominant and ends on V,; at this endured or struggled against, grimly, deterpoint, a large-scale harmonic interruption oc- minedly, or stoically. If this is a first movement, curs, and the recapitulation rebegins on I (or 1). subsequent movements either alter or confirm The difference between the major-mode expo- the fate decreed here. sition and the i—v minor-mode exposition hes Expositions of this type are rare in Mozart, in their prevailing modal qualities. While the who nearly always works with the 1—III option. former are major-mode saturated (and therefore A few examples do occur, though, in the early typically more positive in metaphorical implica- works, such as in the slow movements of the tion), the latter are minor-mode saturated.!” String Quartet in F, K. 168 (exposition keys: F This less common move to the minor dom1i- Minor to C minor) and the String Quartet in nant is not merely “just another option.” On the E-flat, K. 171 (C minor to G minor). Isolated contrary, it is a doggedly negative tonal choice. examples may also be found in Haydn, for inOnce we recall the extra burden of minor-key stance in the first movement of the “Farewell” sonatas —their generic will to explore the possi- Symphony, No. 45 in F-sharp Minor (F-sharp bilities of a transformation into the major mode, to C-sharp minor, discussed more closely beeven though that endeavor might fail—we rec- low).

ognize that the i—v expositional option pro- The move to the minor v instead of to the duces a chillingly dark, fatalistic, punishing, or more normative major III occurs with more frepessimistic referential layout. In the more com- quency in Beethoven, invariably with a grimly mon type of minor-mode exposition, i—III, the tragic or negative effect. As Kerman pointed point was to build a major-mode structure of | out, Beethoven never used this option within promise, leading to a major-mode mediant EEC multimovement C-minor works, preferring and thus constructing the possibility of a ma- instead the move to III (but see the discussion jor-mode ESC in the recapitulation-to-come. below of the Coriolan Overture). On the other In the i-III type the major mode offered a space hand, many of his minor-mode works in other of relief, brightness, or hope within the prevail- keys move to the minor dominant. These 1ning tonic minor. In the i—v type no such relief —_— clude the first movements of: the Violin Sonata

is permitted.!® Table 14.1, no. 8, summarizes in A Minor, op. 23; the Piano Sonata in D Mithis situation as {— —}. What 1s being predicted nor, op. 31 no. 2 (“Tempest”); the Violin Sowithin such an exposition is a tonic-minor ESC, nata in A Minor, op. 47 (“Kreutzer”); and the and with it the failure of that sonata to over- Piano Sonata in E Minor, op. 90. Beethoven come its initial modal conditions. The sonata’s seemed even more likely to employ this option

17. The phrase, again, is taken from Kerman. See n. example occurs in the first movement of Schubert’s

13. Piano Sonata in A Minor, D. 784, the second part of 18. Note, however, the unusual (and rare) deformation whose exposition plays out in E major. See also the first of an exposition that moves “illicitly” from a minor tonic movement of Schubert’s Quartet in D Minor, D. 810, to a (normatively unavailable) major dominant — that is, “Death and the Maiden”: here the exposition tracks fromito V. In such a case the major dominant impresses structurally through F major and A minor (III and v),

one as a delusion, a denial, a “false major” —patheti- although the initial, ominous presence of that A minor cally seeking to overturn the negative implications of (cf. mm. 90—99) is reacted to by a substantial passage of the initial tonic or to proceed “as if” the initial tonic “delusionary”’ A major, mm. 102—12, a stretch of modal had been in the major mode, “as if” the governing mi- denial torn to shreds in mm. 112-14. nor-mode circumstances did not exist. The touchstone

316 Elements of Sonata Theory

in non-C-minor finales, doubtless as negative Symphony No. 45 in F-sharp Minor, “Farewell.” signs of the “sealing of the fate” of the entire The famous Sturm und Drang opening (perhaps piece. Examples of such finales are: the Piano referring wittily, according to Griesinger’s faSonata in F Minor, op. 2 no. 1; the Violin So- mous report of Haydn’s underlying program,

nata in A Minor, op. 23; the Piano Sonata in to the “ardent” frustration of the musicians C-sharp Minor, op. 27 no. 2 (“Moonlight”); kept too long at Eszterhaza castle, away from the Piano Sonata in D Minor, op. 31 no. 2; the their homes and wives in Eisenstadt)?" proceeds Piano Sonata in F Minor, op. 57 (“Appassio- into a continuous exposition that in the vigor-

nata’); the String Quartet in E Minor, op. 59 ous TR at first seems to move (normatively) no. 2; the String Quartet in F Minor, op. 95; into III (mm. 21ff, although the new key 1s not and the String Quartet in A Minor, op. 132. yet secured by means of a PAC—it remains a Commented Kerman: “Both of these Beetho- promise) and thence into an encouraging V7 of venian syndromes—the hankering of C minor A major in m. 30 (albeit one that is not articufor its parallel major, and the tropism of other lated as an MC). But this promise of A major minor keys toward their minor dominants— are collapses at once to A minor (i11) in m. 38 with aberrant according to the norms of the Classic a return of a variant of the plunging Sturm und period. They are certainly not characteristic of | Drang theme. At this point, lacking any precedHaydn and Mozart, Count Waldstein’s anointed ing MC-effect, the continuous (non-two-part)

models for the young prodigy.”!” nature of the exposition declares itself. The supMinor-mode expositions of the i—v type, posedly “intended” A major and its presumed therefore, are appropriate only for special-ef- comforts are shown as lost, and the exposition fect negative statements. Moreover, since the storms onward to the only other tonal option, far more common first-level-default option 1s C-sharp minor (v), within which key the EEC the i-III exposition with its assertion of major is produced shortly before the first ending (m. mode {—+}, selecting the second-level default 65, whose bass C# is then held for eight meai—v option carries with it the connotation of sures). having bypassed (or somehow “lost”) the more Equally instructive is Beethoven’s Coriolan normative, major-mode option. Composers Overture, op. 62, which 1s also concerned with sometimes demonstrated this by constructing a quasi-programmatic matters, although comtonal path that first seeks a move to the posi- mentators have differed on the details of what tive III, then collapses en route or gets derailed is being represented. In the exposition the Pat some pivotal moment. The major III is thus theme (doubtless standing for Coriolanus or depicted as a vision that cannot be realized, the “heroic manner” in general) sets out deterleaving open only the minor-v option. Such a minedly in C minor. A II: HC MC-deformastrategy actualizes the collapse of modal “hope” tion (V® of III) is planted in m. 50, and two bars within the generic sonata: we literally hear the of elegant, falling caesura-fill lead to the emerpossibility of the major mode being liquidated. gence in m. 52 of a nobly contoured S-theme One should be attentive to any suggestions that in the normative E-flat major (III). This E-flat the more normative II]-option 1s being sought major proves unstable, incapable of producing

and lost within TR-space. an EEC in III. Its heroic major-mode promise The locus classicus of such a thematization of 1s turned into something grimmer—something loss occurs in the first movement of Haydn’s forecasting the ultimate demise of the hero.”! S’s 19. Kerman, “Beethoven’s Minority,” p. 220. Pp. 21. Many commentators have repeated Wagner’s spec218—28 provide a table of key relations in Beethoven’s ulation from 1852 that the loss of E-flat and subsequent

minor-mode sonata movements. rising sequences of the S-theme represents the plead20. See the discussion in James Webster, Haydn’s ing wife and mother of Coriolanus outside the city “Farewell” Symphony and the Idea of Classical Style: gates—S as a representation of the feminine. NotwithThrough-Composition and Cyclic Integration in His Instru- standing the tenacity of this belief no evidence suggests

mental Music (Cambridge: Cambridge University Press, that it is inescapably correct. See ch. 7, n. 31, and its

1991), pp. 1-2, 113-19. associated text.

Sonata Form in Minor Keys 317

once-confident E-flat major is placed through 132. One also encounters the 1—VI pattern in rising sequences— sequences of loss—through Schubert, as in the first movements of his (deF minor (m. 64) and into G minor (m. 72), in formational) Symphony No. 4 in C minor, D. which negative key, after several bars of resis- 417 (“Tragic”), his Piano Sonata in A Minor, tance, the EEC 1s sounded (m. 102). This is fol- D. 537, and his Symphony No. 8 in B Minor, lowed by a confirming closing zoneinG minor ~~ _D. 759 (“Unfinished”). Before long this option (mm. 102—18). In sum, the exposition first pre- became more frequently employed, and it might sents us with the normative i—III, {—+}, only to be argued that by the 1840s it had replaced the

demonstrate the extinguishing of major-mode i—v option as a second-level default. (By the hope through the III—1v—v sequences. The re- mid- and late-nineteenth century other “exsult is an unusual realization of the {—[+ —]} perimental” second-key areas also turn up with

exposition. some regularity.) Beethoven and Haydn were not alone in this This is not the place to enter into a discus-

thematizing of the “lost” key in the middle of — sion of the tonal implications of the submedithe exposition. Clementi’s Piano Sonata in F- ant as the secondary key within a minor-mode sharp Minor, op. 25 no. 5 (1788-90) displays sonata. (To some extent it may have been forean engaging adaptation of a trimodular block shadowed by a few celebrated slow movements

(TMB) that articulates two different second- in VI within minor-mode multimovement ary keys. An initial HI: HC MC, m. 24, leads works toward the end of the eighteenth cento TM! (“S”) in A major (III). But this A major tury—such as the E-flat Andantes of two celdoes not produce an EEC capable of being sus- ebrated G-minor Symphonies, Haydn’s No. 83 tained, although one is claimed provisionally, (“Hen”) and Mozart’s No. 40, K. 550. Chapter a III:PAC at m. 40. Instead the music proceeds 15 mentions additional examples.) It may suffice onward, moving toward C-sharp minor to set to point out that Ernst Oster’s commentary to up a second “apparent” MC, v:HC (m. 49), a Schenker’s Free Composition suggests that in such postmedial caesura (PMC) in what we had orig- cases (including their parallel appearances in inally supposed was C-space—a classic strategy major-mode works) “the VI, which ends the of EEC-deferral (see chapter 8). At this point exposition, acts as a third-divider within the another theme ensues (now to be regarded as descending fifth I-IV [which IV often appears TM) in C-sharp minor. It 1s in this key that the near the opening of the development]... . [In

EEC will occur. other cases] the VI probably has to be understood as a neighboring note of the eventual V, and not as a third-divider.”?* Notwithstanding

(Nineteenth-Century) Expositions in the the different tonal “color” and long-range lin-

Minor Mode: EEC in VI ear implications of VI (as opposed to III), these two nineteenth-century second-key options In music from the nineteenth century one en- do share at least one fundamental similarity: counters the occasional exposition that moves they both represent havens or escapes from the from a minor tonic to the key of the major sub- minor-mode tonic (havens that keep open the mediant, i— VI: for example, from D minor to possibility of a major-mode ESC). Both types B-flat major. This may be found in the first —_ of exposition, i—III and the later i- VI, may be movements of Beethoven’s String Quartet in regarded as examples of the {—+} exposition. F Minor, op. 95, his Piano Sonata in C Mi- Consequently, the remarks above about the typnor, op. 111, his Symphony No. 9 in D Minor, ical modal drama carried out in the normative op. 125, and his String Quartet in A minor, op. i—III exposition also apply here.

22. Schenker, Free Composition (Der freie Satz), p. 140.

CHAPTER FIFTEEN © "ESF 0

The Three- and Four-Movement Sonata Cycle

ihe idea of seeking to grasp whole successive thing analysis that makes such resistant to movements as a coherent ratherOne empirical 1s that theissues sought-for coherthan as a string of dissociated contrasts —that ence, or lack of it, is by no means a function is, to understand them as a planned cycle bear- only of the raw data provided by the acoustic ing intermovement implications—seems at surface of music—the sounds that one literally once self-evident and dauntingly difficult. For — hears. This sense of cohesiveness, construed as some the illusory “naturalness” of the cycle can a web of interrelations, is not to be conceived appear unproblematic through the very famul- exclusively as a pre-existing “object” to be dislarity of its patterns. Yet once we start asking covered through research. Instead, perceiving questions about it, the notion blurs into un- degrees of multimovement integration also becertainty. Are the movements among differing longs to the realm of interpretation. This aspect late-eighteenth-century works interchangeable? will be revisited at the end of this chapter, under What significance might there be in the order “The Role of the Listener.” in which the movements appear? Does it matter One’s sense of cyclic coherence 1s assisted by which of them, if any, is in the parallel mode (in focusing on demonstrable relationships among F minor, for instance, within an F-major sonata) the movements. Some relevant factors can be or in anontonic key? Why does tempo matter?! dealt with apart from an examination of indi-

1. The past few decades have seen a number of histori- burtstag, ed. Christoph-Hellmut Mahling (Tutzing: cal studies of the late-eighteenth-century sonata cycle Schneider, 1979), pp. 1-40, Wilhelm Seidel, “Schnell— in both theory and practice. Fundamental background Langsam—Schnell: Zur ‘klassischen’ Theorie des inreading includes the general, sometimes statistical sur- strumentalen Zyklus,” Musiktheorie 1 (1986), 205-16, veys of “the cycle as a whole” in William S. Newman, and Seidel, “Die altere Zyklustheorie, tiberdacht im The Sonata in the Classic Era (New York: Norton, 1972), Blick auf Beethovens Werk,” in Beitraige zu Beethovens

pp. 133-43; the useful, often Beethoven-oriented Kammermusik: Symposion Bonn 1984, ed. Sieghard summaries of Germanic eighteenth- and early-nine- Brandenburg and Helmut Loos, Veréffentlichungen teenth-century writers on theories of “unity within des Beethovenhauses in Bonn, Neue Folge, 4, Reihe: diversity” and “the older cycle-theory” in Chris- Schriften zur Beethovenforschung, 4 (Munich: Henle, toph-Hellmut Mahling, “Zur Frage der ‘Einheit’ der 1987), pp. 273-82; and the reflections on this problem Symphonie,” in Uber Sinfonien: Beitrage zu einer musi- with special regard to Haydn in James Webster, Haydn’s kalischen Gattung: Festschrift Walter Wiora zum 70. Ge- “Farewell” Symphony and the Idea of Classical Style. 318

The Three- and Four-Movement Sonata Cycle 319

vidual content. General features of style and event rarely happens in this period) but opened tempo can be perceived as creating abstractly the concepts to other “aspects of musical consatisfying balances among themselves— fast struction and technique (commonalities of maouter movements (often fast and faster) enclos- terial, tonal relations, and the like [among the ing different, contrasting movements, for exam- movements])” —something like the features of ple; or a multimovement work that increases in K. 332 mentioned above.? tempo as one movement gives way to the next. The utility of identifying such technical feaMore commonly, this line of discussion turns tures is self-evident. It 1s a necessary step in the to those observations that current music theory analysis of all works considered as wholes. Most conventionally teases out from single works: of this chapter, however, 1s concerned with a shared motives, keys, themes, harmonies, 1n- different matter: the coherence provided by the strumentation, texture, and the like. One may norms of the relevant background genre, against contend that Mozart’s three-movement Piano which the individual work throws itself. This Sonata in F, K. 332, is self-referentially balanced involves a look at conventional movement plans,

because the first and last movements share cer- common structural and tonal choices and their tain form-defining features, inviting the conclu- implications, movement order, and, toward the sion that the finale reworks procedures that had end, a broader consideration of what it might originally been set into relief at the opening. entail to grasp a multimovement work as a sinThe finale may be understood as re-presenting ole entity. them from a different vantage point (as with a sleeve drawn inside-out): both movements

overdetermine their P-zones with a series of Number of Movements unusually redundant I:PACs; both initiate TR

with an unprepared lurch into the submediant; In his inventories of “the sonata in the clasand both feature post-MC spaces that emphasize sic era” William S. Newman noted that mula prominent module that has decayed into the timovement sonatas in the decades before and minor mode (a later module in the first move- around 1800 normally contained two, three, or

ment; an initial one in the finale). four movements, with the most common numSome helpful thinking along these lines is ber being three. Within nonorchestral pieces found in James Webster’s 1991 study of Haydn’s labeled as sonatas (which usually begin with a “Parewell” Symphony and related works. Web- sonata form) ster identified two musical strategies capable of

affecting our perception of long-range, multi- there can be no question that by far the largmovement coherence. Normally the two are est number... are in three movements. .. . In found simultaneously and overlap in practice. the three-movement cycle, the most frequent One was through-composition, descriptive of “de- order of movements 1s P(ast)—S (low) —F, OF stabilizing techniques” within early movements F—M (oderate) ~F. oo Haydn uses a mainuer as

, cc , the middle or final movement in more than half

that defer resolution to later points, of “dynamicsonatas. his three M itonl is three-movement Mozart uses only or gestural phenomena (run-on movements, re- twice, as the middle movement, and Beethoven calls, unresolved instabilities, lack of closure not at all in his three-movement sonatas... . In and so forth).” The other was cyclic integration or centers other than Vienna, when the three-movecyclic organization, those aspects that contribute ment cycle is not built on the favorite F—S—F or toward the multimovement stabilization of the F—M-F plan (as it is so regularly in Germany), cycle 4s a whole. By these latter terms Webster it sometimes reverts to the late-Baroque plan of did not limit himself only to the resurfacing of S—F-F. Otherwise, the variety of movement earlier themes in later movements (which in any plans [can be] so great that often we shall be able

2. Webster, Haydn’s “Farewell” Symphony and the Idea of Classical Style, pp. 7-8.

320 Elements of Sonata Theory

to note only the frequency of the minuet finale, three-movement works; his next set, the “Vienthe fondness for rondos and variations, the need nese” cycle, K. 168-73, comprises four-movefor contrast between movements, and perhaps a ment works, as do the remainder of his quartets. tendency to step up the meter (or fractional time Most of Haydn’s quartets are also in four move-

signature) from one movement to the next.’ ments. The most typical four-movement plan

rh h b lat; is F—S—miunuet—F. (Instead of either a conven-

, ere eee .. a CEM a COFTE ne tional minuet-finale or a fast-finale after the

etween the number of movements in a Key- slow movement, one had both.) Occasionally, board sonata (or other nonorchestral work) and the minuet was placed in second-movement po-

the seriousness of the work, at least when the sition number deviated away from the standard three. There are currently two points of view with ‘Two-movement sonatas were almost certainly regard to the origin of the four-movement sym-

viewed aseich lighter works;e€four-movement ones h Th dard ed1n| nf a the h phony. more standard, as summarized (infrequent in the eighteenth century) as more 1986 by Wolf, is that four-movement symphoambitious, particularly toward the end of the nies were created “by the insertion of a minuet century and thereafter. Although several Of 3.4 trio before the finale,” a pattern encountered Haydn’s piano sonatas are in two movements, in some of the Mannheim composers —notain the eighteenth century the two-movement bly Johann Stamitz— “from approximately the sonata may have been associated with such Ital- a yaag. on 5 In 1991 Webster proposed an an keyboard COMTPOSCTS a's Alberti, Paradisi, alternative interpretation: “The development Galuppi, and Rutini (hence the term sometimes of the four-movement symphony may have en-

used for them, iealian vane a cist tailed, not so much the insertion of a minuet ey ee ninnce P wa Se ET FONGOs into the F—S—F pattern, as is usually assumed,

and r—miuinuet. . . 7 STOW movement In CWO- as the addition of a finale to the traditional movement sonatas 1s rather rare.’”+ (Needless to three-movement pattern ending with a minsay, late-Beethoven two-movement sonatas pro- uet.”6 While the three-movement symphony vide significant exceptions to the earlier norm continued to exist, often as an overture-sinfoof lightness.) Four movement eyboard pone nia or a brief, lighter composition, four-movetas are rare before the mid-1790s. Beethoven’s ment works became more the norm throughout frequent expansion of the piano sonata to four the eighteenth century, particularly among the movements is doubtless to be understood as an master composers. To borrow Sisman’s sumenriching of the genre, which had already seen mary, “By 1780, symphonies had long since some three-movement advances in the works of abandoned the three-movement format that

f lightness.) Four- k - of light, 3

Haydn and Mozart. | linked them to earlier Italian opera overtures. In the last three or four decades of the eigh- Haydn wrote no three-movement symphonies teenth century the highest compositional pres- after 1765, while Mozart's ‘Prague’ Symphony tige within instrumental penne was claimed by with its famously absent Minuet, remains the the Austro-Germanic string quartet and sym- exception that proves the rule.”7 On the other phony. Here one often finds the three-move- hand, concertos— which became a much-elabment plan exp anded to four movements. Mo- orated genre with Mozart— were almost invarizart’s quartets illustrate the shift. His “Italian ably three-movement works with their own excycle of six quartets, K. 155-60, contains only panded version of sonata form (Type 5) for the

3. Newman, The Sonata in the Classic Era, pp. 133, of Classical Style, p. 183, with references to Karl H.

135. Worner, Das Zeitalter der thematischen Prozesse in der Ge-

4. Newman, The Sonata in the Classic Era, pp. 134-35. schichte der Musik (Regensburg, 1969), chap. 1 “Final5. Eugene K. Wolf, “Symphony,” The New Harvard Dic- charakter”; and Bernd Sponheuer, “Haydns Arbeit am tionary of Music, ed. Don Michael Randel (Cambridge, Finalproblem,” Archiv fiir Musikwissenschaft 34 (1977),

Mass.: Harvard University Press, 1986), p. 823. 199-224. 6. Webster, Haydn’s “Farewell” Symphony and the Idea 7. Sisman, Mozart: The “Jupiter” Symphony, p. 7.

The Three- and Four-Movement Sonata Cycle 321

first movement, similar adaptations for whatever First Movement formal plans undergirded the slow movement and finale, and an overall F—S—F plan. (See When the first movement was in a rapid tem-

chapters 19-22.) po—as was the case most frequently — its struc-

Five- and six-movement works were also ture was obligatory. It was to be cast in “grand possible, especially in such “entertainment” binary” structure (Reicha’s “la grande coupe genres as serenades and divertimenti. The loose binaire’”’), which we now call sonata form.? The correlation between a greater number of move- opening movement declares the tonic that will ments (and their lengths) and the enhanced pres- govern the whole work. We find here an assertige-claim of the work breaks down when the tion of the tonic key 1n its full complexity within number of movements exceeds four.® This may a sophisticated structural trajectory aimed at the be a residue of earlier suite practice, and indeed, ESC. The first movement sets the emotional

several of Mozart’s such works contain an in- tone for the work, designed to match the sotroductory march, one or more minuets, and so cial prestige of the occasion or “ceremony” at on. To be sure, in Mozart’s hands the serenade which it is to be performed. (A symphony could with more than four movements could take on be understood as a marker of the grandeur, forprofound connotations, as in the seven-move- mality, or splendor of its own realization as a ment Serenade in B-flat for Thirteen Winds, K. public event; chamber music as something more 361, and the five-movement Serenade in E-flat intimate, appealing more explicitly to the indifor Eight Winds, K. 375. The unexpected ele- vidual performers or the connoisseur.)

ment is the concealed depth in what seems to Within a symphony this first-movement advertise itself as unpretentious entertainment. structure was to be carried out in the high or (On other occasions Mozart’s four-movement elevated style. In J. A. P. Schulz’s much-cited serenades took on the serious tone of higher words from 1774, the opening movement was genres, most notably in the Serenade in C Mi- to unfold in such a way as to become the initial nor for Eight Winds, K. 388, reworked into the statement of a work whose task it was to be String Quintet in C Minor, K. 406. The Serenade in G, K. 525, “Eine Kleine Nachtmusik,” the expression of the grand, the festive, and the is essentially an elegant string symphony with a sublime. Its purpose is to prepare the listeners

“lighter” tone throughout.) for an important musical work, or in a chamber What follows is an outline of the norms of concert to summon up all the splendor of instruthe most standard four-movement plan as found mental music. — The allegros of the best chamin Haydn, Mozart, Beethoven, and their im- ber symphonies contain great and bold ideas, free mediate successors. Most of the discussion here handling of composition, seeming disorder in the melody and harmony, strongly marked rhythms of

has the four-movement symphony and quar- different kinds, powerful bass melodies and unitet in mind, although its observations gener- sons, concerting middle voices, free imitations, ally apply equally well to other ambitious four- often a theme that is handled in the manner of movement works. It is also applicable to most a fugue, sudden transitions and digressions from three-movement works: concertos, sonatas, and one key to another . . . strong shadings of the forte so on: one merely has to set aside the absent and piano, and chiefly of the crescendo... . Such movement— most often (though not invariably) an allegro 1s to the symphony what a Pindaric ode

the minuct. is to poetry. Like the ode, it lifts and stirs the soul of the listener and requires the same spirit, the same sublime power of imagination, and the same aesthetics in order to be happy therein.!°

8. A similar conclusion is drawn in Michael Talbot, The 10. From J. A. P. Schulz’s article “Symphonie” in JoFinale in Western Instrumental Music (New York: Oxford hann Georg Sulzer, Allgemeine Theorie der schénen Kiinste

University Press, 2001), p. 29. (Leipzig, 1771-74). Cited here is Elaine Sisman’s adap9. Anton Reicha, Traité de haute composition musicale tation of Bathia Churgin’s translation [Churgin, “The

(Paris, 1826), 2:300. Symphony as Described by J. A. P. Schulz: A Com-

322 Elements of Sonata Theory

The first movement sounds the tone of im- mode within an otherwise major-mode work,

portance for the entire composition, which or vice versa. in a symphony is a celebration of instrumen- This movement is conventionally placed (as tal music and its expressive capabilities. This a first-level default) in the second-movement was particularly true of grander symphonies, position. Exceptions do occur, though. Occacharacterized by broader gestures and increas- sionally a two- or three-movement sonata will ingly considered to have attained the level of | begin with a slow movement (as in Haydn’s Pithe Pindaric sublime. Sonatas and chamber mu- ano Sonatas in D, Hob. XV1I:42; in C, Hob. sic made less public claims and unfolded in a XVI:48; and in D, Hob. XVI:51; it is rarer more private, elaborate, nuanced, or detailed to begin a four-movement work with a slow style. First-movement rhetoric was to be shaped movement, as in Symphony No. 49 in F Minor, according to the genre of the composition at “La Passione’). Within four-movement works hand.!! Prom another perspective — encom- one might also encounter the slow movement

passing also ambitious sonatas and chamber displaced to the third position (a second-level music—this movement, as a demonstration of — default), following the minuet instead of precompositional skill, lays out the aesthetic and ceding it. This happens, for example, in seven of expressive levels at which the remainder of the the twelve quartets constituting Haydn’s op. 20 “same” will be played. The first movement sets and op. 33. (Additionally, in op. 33 each of the the terms of understanding for the movements six “minuet” movements 1s labeled as either a to follow. The flexibility and implied drama “scherzando” or a “scherzo.”) This slow-movewithin the grand-binary structure are naturally ment/minuet exchange of positions can also be suited to this task. In all likelihood the struc- found in Haydn’s middle symphonies but not in ture was developed precisely to permit the ac- the late ones. Placing the slow movement in the

complishment of these things. second position of a four-movement scheme is almost invariable in Haydn’s symphonies from the mid-1760s onward. A few exceptions may

Slow Movement be found, however—for instance, in Nos. 37,

44 (“Trauer”), and 68, in all three of which the The slow movement presents a space of contrast minuet is placed second. The same later relucwithin the four-movement plan. Especially in tance to switch the more standard placements of major-mode works, the slow movement 1s often the slow movement and minuet may be found in sounded in a nontonic key. When itis, itusually = Mozart’s and Haydn’s later quartets. Although functions as the only escape from the govern- _ the practice does not disappear in these works, ing tonic of the whole. Other features can add it is less common.

to the impression of the second movement as a Unlike the situation with a first movement, foil to its predecessor: its more leisurely contrast the slow movement’s form is nonobligatory. to the energy and bustle of the first movement; = One cannot predict what the form will be in its more persistent lyricism; its frequent clarity advance of hearing it. It can be another sonataof texture and relative contrapuntal simplicity; form structure: a Type 3 (“textbook”) or, quite its tendency to favor less complex formal struc- often, a Type 1 (without development) sonata. tures; and its occasional selection of the minor Type 2s (“binary,” without a full recapitulation) mentary and Translation,” Current Musicology 29 (1980), seen as comparable to the persistent themes of opera 7-16], in Sisman, Mozart: The “Jupiter” Symphony, pp. seria itself: love versus honour.” Zaslaw, Mozart’s Sym9-10. Cf. Neal Zaslaw’s remarks in 1989: “The first phonies: Context, Performance Practice, Reception (Oxtord: movements represent the heroic, frequently with mar- Clarendon Press, 1989), p. 417: tial character... . Later [eighteenth-century symphonic 11. See, e.g., Michael Broyles, “The Two Instrumental first movements] contain contrasting lyrical ideas. Ap- Styles of Classicism,” Journal of the American Musicological

propriately, given the origins of the sinfonia in the opera Society 36 (1983), 220. Cf. Sisman, p. 10. pit, the two sorts of ideas—lyrical and martial— may be

The Three- and Four-Movement Sonata Cycle 323

and ‘Type 4s (“sonata rondos” of varying kinds) lighter expression of personal, galant sentiment also turn up from time to time. The slow move- and cantabile discourse. Still, it would be misment may also unfold as a ternary ABA’, asa __— leading to consign more ambitious slow movesimpler rondo structure, or as a theme and varia- ments only to the realm of conventional sentitions or a set of alternating variations.'2 On rare ment. With the increasing importance given to occasions, it is appropriate to understand a slow instrumental music in the later eighteenth cenmovement in different terms from those pro- tury —and especially with its transformation in vided by the more standard guidelines. Such 1s the hands of Mozart and Beethoven—the slow the case with the Andante of Beethoven’s Fifth movement would take on an inward turn. As Symphony, which is most profitably grasped as Margaret Notley noted, the formerly sentimenan early illustration of rotational form deployed tal ideal often gave way to the ideal of emoon its own, without significant intersection with tional depth, one typically associated, especially

other, pre-established formal patterns. (Such in Beethoven and afterward, with the Adagio an overriding rotational structure foreshadows tempo, which was now capable of becoming the some of those found 1n both slow and fast move- expressive centerpiece of an entire work.!© By

ments in Berlioz, Bruckner, Mahler, Sibelius, the mid-nineteenth century A. B. Marx (and

and others.)!9 others) would refer to such slow movements in In his 1774 discussion of the symphony as a quasi-spiritual terms: the slow movement repgenre, J. A. P. Schulz mentioned that “the an- resented the “sabbath-day rest” of the composidante or largo between the first and last alle- tion, Marx claimed. Elsewhere, he wrote that in ero has indeed not nearly so fixed a character | Adagio movements the composer often seems to [as the first movement], but is often of pleas- ask “Who am I[?”!” ant, or pathetic, or sad expression. Yet it must

have a style that 1s appropriate to the dignity of Key Choice in Slow Movements:

the symphony.”!4 Summarizing Mozart’s sym- ; phonic slow movements, Neal Zaslaw remarked

Majyor-Mode Sonatas

that “the andantes deal with the pastoral, as the Among the first things to notice about a slow origin of a few of Mozart’s in bucolic operatic movement 1s its choice of key: does it retain the scenes reveals.”!5 More generally, the pastoral or original tonic of the first movement— perhaps sentimental slow movement resonated with the with a change of mode— or (as is more common) 12. According to Sisman, Mozart: The “Jupiter” Sym- ways with “linear” sonata deformational procedures and phony, p. 8, Haydn introduced the slow variation move- the process of teleological genesis—successive nurturment into the symphony in 1772 with his Symphony ings of an underlying idea that will eventually be treated

No. 47. to a climactic statement. EEC and ESC attainment (or 13. Cf. Darcy, “Bruckner’s Sonata Deformations,” nonattainment) remain crucially important features of Bruckner Studies, ed. Timothy L. Jackson and Paul Hawk- all of these works, as does the sheer sense of Berlioz’s

shaw (Cambridge: Cambridge University Press, 1997), purposely frictional, transgressive treatment of sonata pp. 256-77; Darcy, “Rotational Form, Teleological norms, often in pursuit, it seems, of a swashbuckling Genesis, and Fantasy-Projection in the Slow Movement or ego-charged “Romantic” freedom, sometimes tied of Mahler’s Sixth Symphony,” 19th-Century Music 25 loosely to an extramusical image. (2001), 49-74; and Hepokoski, “Jean (Christian Ju- 14. Translation as quoted in Sisman, p. 10. lius) Sibelius,” in The New Grove Dictionary of Music and 15. Zaslaw, Mozart’s Symphonies, p. 417.

Musicians, 2nd ed., 23: 319-47. A clear instance of an 16. Notley, “Late Nineteenth-Century Chamber Muoverriding rotational structure in a rapid movement of sic and the Cult of the Classical Adagio,” 19th-Century Berlioz may be found in the Overture, Le carnaval romain Music 23 (1999), 33-61.

(which at best responds awkwardly to “sonata-form” 17. Marx, Ludwig van Beethoven: Leben und Schaffen analysis, not the least because it gives the impression of (1859), quoted in Ian Bent, ed., Music Analysis in the containing a repeated exposition, normally not available Nineteenth Century, vol. 2, Hermeneutic Approaches, (Cam-

in overtures). Elsewhere in Berlioz (the Overtures Ben- bridge: Cambridge University Press, 1994), p. 215; and venuto Cellini and Le corsaire, the first movements of the Notley, “Late Nineteenth-Century Chamber Music Symphonie fantastique and Harold en Italie) the “circular” and the Cult of the Classical Adagio,” p. 35. rotational principle can be intermixed in provocative

324 Elements of Sonata Theory

does it shift to a different key? Because the mem- dominant. This occurs as a matter of course ory of the first movement persists conceptually when we have a subdominant slow movement throughout the remainder of the piece, the key that is disposed 1n one of the sonata-form types, of the slow movement 1s to be interpreted with in all of which the S- and C-zones will appear regard to at least three factors. The first 1s its “ab- in V of that IV. This situation permits the real stract” relationship to the governing key of the tonic of the whole work to emerge as tempowhole piece. The second 1s the earlier role that rarily subordinate, locally under the sway of the slow-movement key had already played in the the reigning slow-movement tonic. Any such first movement: that key arrives in the slow move- real-tonic appearances refer both to an earlier ment with a history. The third is the role that the condition (the first movement) and to the condifirst--movement key is able to play, albeit second- tion-to-come (the third and fourth movements). arily, in the slow movement. All (re)appearances But within the slow movement they are “fated” of the large-scale governing tonic within the slow not to last. They exist on a different conceptual

movement are noteworthy as such. plane, bracketed, fragile, ephemeral. The composer had available several options When the composer presents the slow movefor the key of the slow movement. If we ar- ment in the second-level default key, the domirange them in order of frequency, we arrive at a nant, the effect is different. Moves to the domiset of default choices. During this period these nant are gestures toward increased brightness defaults might be ordered as follows, from most and tension—the opposite from that produced

to least common: by the choice of the subdominant. Any dominant-key slow movement aligns itself tonally

1. Subdominant (IV) with the S- and C-zones of the first movement’s

2. Dominant (V) exposition. (As the exposition’s dominant-key 5. Tonic minor (1) zones eventually resolved to the tonic in the re-

4. Submediant manor (vi) . a capitulation, so too will the slow movement’s >. Wee on of IIT or VI (available primarily prevailing dominant be brought back to the

6. Tonic major (1) tonic in subsequent movements, although with

7 Other differing thematic material.) In dominant-key

slow movements the original tonic of the first Cases 1 and 2: Slow Movement in a Nontonic Ma- movement is less “automatically” available, jor Mode, Subdominant or Dominant. This is a since it would have to appear as the subdomi-

swinging outward to a fifth-related key, one nant of its OWn V—a key that need not recur that retains the major mode. The escape into a generically Mm any of the sonata types (although nontonic cantabile or dreamlike elsewhere 1s par- it could certainly be touched upon in the deticularly clear in slow movements with pastoral velopment, recapitulation, or coda). What ‘s or Arcadian connotations or in those speaking MOTE likely 1s that this dominant key will push the erotic language of love, desire, or seduction. upward to its own dominant, rendering its SExamples are ubiquitous in the repertory. To and C-zone EVER TOT’ tensely distant from the give a sense of proportion: in Haydn’s major-key original first-movement tonic. symphonies after No. 70, sixteen move to IV for

the slow movement: twelve move to V.!8 Cases 3 and 4: Slow Movement in Tonic or NonThe first-level default, the subdominant-key tonic Minor Mode (i or vi). In both cases it is the

slow movement, lowers the first movement’s unprepared switch of mode to the minor—the tension toward the more relaxed IV. Here the “lights-out” effect—that provides the initial opening movement’s tonic is to be reapproached impact: the sudden precipitation of an ominous only from below, as the dominant of that sub- antitype to the first movement's type. (‘he pro-

18. Moving to the subdominant are Nos. 72, 73, 75, Moving to the dominant: Nos. 71, 74, 76, 77, 81, 84, 88, 79, 82, 85, 86, 87, 90, 93, 94, 96, 97, 100, 101, and 104. 89, 91, 92, 98, and 102.

The Three- and Four-Movement Sonata Cycle 325

cedure is also analogous to the conventional mi- In the case-4 option the minor-mode key nore variation in theme-and-variation sets, mi- chosen is not the tonic but the submediant, vi, nor-mode slow movements within major-mode which may be heard as a much-prolonged upper suites, and the like.) We refer to case 3, the pre- neighbor to the dominant of the multimovesentation of the slow movement in the tonic ment work’s real tonic. From another perspecminor, as the prison-house effect, as if one were tive, it may seem as though the first--movement shackled fast to an immovable tonic. Since the tonic has undergone a 5—6 shift with added nontonic escape normally occurs, if at all, in the bass support (C major to A minor, for instance), slow movement— and since it is not occurring which also suggests intersections with neo-Riehere—we are led to expect that there will be no mannian interpretations of chordal transformarelief from this tonic in any of the movements. tions. Under any reading the submediant minor This sense of no escape is redoubled through the is likely to be heard as an antitype to the previcollapse of mode into minor. Metaphorically, ously governing key—as the deep-sinking into when we hear the beginning of such a slow the gloomy, spectral, grotesque, or funereal unmovement, the coldest of shadows passes over the derside of the tonic (the relative minor, retaintonic; the prison-house door closes and locks. ing the key signature of the work’s real tonic). A few examples spring to mind at once: the A submediant-minor slow movement also carsroaning, D-minor Largo e mesto in Beethoven’s ries the potential of producing the original maPiano Sonatain D, op. 10 no. 3, the F-minorslow —_—jor-mode tonic as its most probable secondary

movement (in third-movement position) in his key, the mediant (III). This transient retouchString Quartet in F, op. 59 no. 1, the D-minor ing of the first-movement tonic within the loLargo assai ed espressivo movement from the cal control of the dark submediant can be used Piano Trio in D, op. 70 no. 1, “Ghost,” and to powerful effect—the passing vision of what the relentless, A-minor variation movement in once was (and may yet be again in subsequent his Symphony No. 7 in A, op. 92. This case-3 movements). option turns up with some frequency from the The most familiar example of this option may mid-eighteenth century onward. It is readily — be the C-minor funeral march from Beethoven’s found in early Mozart: in the tonic-minor slow (E-flat) Eroica Symphony. Or it may be the movements from the Violin Sonata in F, K. 13; F-sharp minor Andante, rocking back and forth the Piano Concerto “No. 4” in G, K. 41; the in grief, from Mozart’s Piano Concerto No. 23 Symphony in C, K. 96 (111b); the String Quar- in A, K. 488. Or perhaps the C-minor Andante tetsin C, K. 157, andin F, K. 168; the Piano So- piu tosto Allegretto of Haydn’s Symphony No. nata in F, K. 280; and elsewhere. Examples from 103 in E-flat. However celebrated such moveHaydn’s symphonies include the slow move- ments, there were several precedents for their ments of No. 3 in G, No. 4in D, No. 12 in E, adoption of the submediant minor. To mention No. 17 in F, No. 19 in D, and others, including only a few from early and early-middle Mozart: No. 63 in C and No. 70 in D—though not the the Symphony No. 1 in E-flat, K. 16; the String later symphonies, in which the composer pre- Quartet in G, K. 156, the String Quartet in ferred nontonic slow movements. Any appear- E-flat, K. 171, and the Piano Concerto No. 9 ance of the tonic major within these tonic-mi- in E-flat, K. 271. nor movements can strike us as a poignant flash of hope predestined to be reabsorbed back into Case 5: Major-Mode Keys a Third Away from the

the minor—the “if only” mirage of what one Tonic (III or VI). When such third-related keys cannot have, at least in this movement, as in the begin to crop up in several of Haydn’s slow A-major “dreamlike” alternative passages in the movements in the years around 1790 (they are A-minor slow movement of Beethoven’s Sev- only very rarely to be found in Mozart’s),!? they

enth Symphony, mm. 102ff, 225ff. are best regarded, at least in those early years, as 19. Some unusual examples may be found among Mo- seven-movement work with two slow movements (the zart’s early serenades. The Serenade in D, K. 185, is a second and the fifth). The first of these, an Andante, is

326 Elements of Sonata Theory

deformations of normative practice (that would with a Poco Adagio in D (III); and an A-major later be standardized as lower-level defaults) .*° Fourth with a Largo ed innocentemente in F The option could have been borrowed from the (4VI).21 Within Haydn’s output some impor-

idea of moving from a major key to its relative tant instances occur in Piano Trios. Examminor (or vice versa) for the slow movement, al- ples include Hob. XV:14 in A-flat (1789-90), though in this adaptation the third-relation typi- with an Adagio in E major (enharmonic pVI), cally involved two major keys, while the interval claimed by David Wyn Jones to be the earliest between them could be either a major or a minor such example in Haydn’s instrumental music;?? third. As with all such third-relations, they may Hob. XV:20 in B-flat (1794), with an Andante be interpreted according to either Schenkerian, cantabile in G (VI); and Hob. XV:29 in E-flat linear-contrapuntal principles or neo-Rieman- (1795-97), with slow movement, Andante et

nian chordal transformation theory. In most innocentemente, in B (enharmonic VI). A few cases the original, first-movement tonic will be late-quartet examples from the 1790s, and one unavailable — or will be at least more unlikely to from a symphony, may also be included: the be touched upon—1in the new key, which will F-sharp-major Largo, Cantabile e mesto (III) probably emphasize its own dominant. As a re- from the Quartet in D, op. 76 no. 5; the B-major sult the sense of skow-movement escape from the Fantasia-Adagio (enharmonic bVI) from the

tonic is more deeply registered here. Quartet in E-flat, op. 76 no. 6; the E-flat AdaThe idea of a slow movement in a major III gio (bVI) from the Quartet in G, op. 77 no. 1; or VI was not new to Haydn: the practice may the D-major Andante (VI) from the Quartet in have been indebted to eccentric key choices F, op. 77 no. 2; and the G-major Adagio (III) made earlier by C. P. E. Bach. Reviewing the of the Symphony No. 99 in E-flat.?° The rash latter’s six “Hamburg” String Symphonies, Wq of third-related slow movements spread to early 182/1—6 (H. 657-62, 1773), for example, one Beethoven almost at once: the E-major Adagio finds a G-major First with a highly unsettled, (III) of his Piano Sonata in C, op. 2 no. 3 (1795), Poco Adagio in E (VI); a B-flat-major Second and many other examples. in F (4III), the second in A (V). Curiously, the third beit not reflected in the key signature) 1s in the minor movement, an Allegro, is also in F. The Serenade in mode. Cf. also the first of his “Kenner und Liebhaber”’ D, K. 203, is an eight-movement work with two slow keyboard sonatas from the first book, Wq 55/1 (H. 244, movements (the second and the sixth). The first is in publ. 1779), in C major with a slow movement in E B-flat (bVI), the second in G (IV). The third move- minor. ment is a non-tonic minuet in F (with trio in B-flat), 22. “The Adagio [of “Trio No. 27 in A-flat’ = Hob. and the fourth movement (an Allegro) is in B-flat. Thus XV:14] is notated in E major (really F-flat major to the the overall key plan is: D-Bb—F—-Bb—G-—D-—D. The A-flat of the first movement), the first time in any inF—Bb—F of the third movement minucet and trio reverses strumental genre that Haydn does not choose the oppo-

the tonal pattern of movements 2—4 taken as a unit. If site mode or a directly related key. Though this colourmovements 2—4 are considered as a large expansion of ful choice of key had been presaged in the chain finales Bb, then the deep-level tonal plan is D—Bb—G—D, en- of La fedelta premiata, in Haydn’s instrumental music it acting a downward arpeggiation to the subdominant. erew naturally out of the extended tonal and harmonic 20. Cf. the discussion in Webster, Haydn’s “Farewell” vocabulary of the 1780s, and in this particular Trio from Symphony and the Idea of Classical Style, p. 212: “Beginning the harmonic adventures of the previous movement.

around 1790, and increasingly thereafter, Haydn em- Mediant relationships within and between movements ployed remote key relations between contiguous move- were to fascinate Haydn for the rest of his life, and it ments.” Cf. also David Wyn Jones’s treatment of Haydn, was a feature of his style that fired the imagination of early Beethoven, and third-related keys for slow move- the young Beethoven. In Haydn’s case it added a new ments in H. C. Robbins Landon and David Wyn Jones, sense of colour and drama to his music that compleHaydn: His Life and Music (Bloomington: University of mented and sometimes interacted with more traditional Indiana Press, 1988), pp. 296-97; and Ethan Haimo, relationships” (Robbins Landon and Wyn Jones, Haydn: “Remote Keys and Multi-movement Unity: Haydn in His Life and Music, p. 211). the 1790s,” The Musical Quarterly 74 (1990), 242-68. 23. Obviously related but more striking (since it sets 21. Also related, it seems, is the C-major Third with a out from a minor-mode first-movement tonic) is the much-anguished Adagio centered around E minor (111), E-major Largo assai (4V1) of the Quartet in G Minor although in this case the main slow-movement key (al- (“The Rider”), op. 74 no. 3 (1793).

The Three- and Four-Movement Sonata Cycle 327

Case 6: Major Tonic (I). Examples in which all Key Choice in Slow Movements:

four movements are in the same tonic-major Munor-Mode Sonatas

key (perhaps recalling earlier aspects of the ma- . . ,

~ (pethap nedance P Minor-mode sonatas differ from major-mode jor-mode suite) may be found, but they ,mi5, . . ones because of the “extra burden” of the are considerably less common thantothebechoices .;.. . . nor mode itself: its seeking emancipated discussed above. Here one is again locked into . ,. . . into the major without mode (chapter 14).1:This aspect the tonic throughout— escape —al. ; ; a 1s 3 of prevailing minor mode 1s played out not though there nothe “prison-house” collapse to removements , only within individual sonata but the tonic minor for the slow movement. Instead .

.unremittingly . also throughout thenever multimovement sonata. As the whole bright, swervLy:movements , , aisresult, mode choice within slow

ing itsonmajor-tonic security. In someThe cases ... a .from takes aofheightened importance. options this is a result placing the slow movement . . . . _.first-movement . for key choice are listed In this case in position, as inbelow. Haydn’s Sym. . . the . . . . ’ . ordering does not imply a hierarchy of options,

phonies No. 5 in A, 11 in E-flat, 18 in G, 21 in ,

7 , 5 (“Der particularly because Nos.. 1, 2, and 3 seem alA, and .22 in E-flat Philosoph”)—along , . most equally available. (One should also recall with, for instance, Mozart’s Quartet in G, K. that minor-mode works were much less plen-

80, and few other early works. Less frequently, . . . For ; . . a. tiful than major-mode ones at this time.) the major-tonic slow movement is placed in an . . . Lf . . most of the period under consideration. .no. 1 interior position: in Haydn’s Symphonies No. 62 . . D; . , may have beennumber more regularly selected in in2a(especially substantial of his piano sona.than . No. . . by Haydn), although its frequency tas (which sometimes feature , No. 3 , ; seems toalso decline after tonic-minor 1790. Similarly, slow movements) ;?4+ or in choice—probably Mozart’s Serenade inmore ho; babl : . was a common stanE-FPlat..for Eight Winds, K. 375 (all five move_?y , . dard than no. 2—particularly from the 1770s ments of which are in the major tonic). onward.

a z: a This earegory encompasses more 1. Tonic major (1)

explicit deformations, w uich demand to be con- 2. Mediant major (IIT) fronted on an ad hoc basis. ‘The obvious touch- 3. Submediant major (VI) stone 1s found in Haydn’s Piano Sonata in E-flat, 4. Tonic minor (i) Hob. XVI:52 (1794), whose center is occupied by 5. Other a much-noted slow movement in E major, an enharmonic bII of the original tonic (a key famously Case 1: Tonic Major (I). This is a common key prepared by an E-major passage in the first move- choice for the slow movement, particularly in ment). One precedent for this choice occurs in a the decades prior to 1790. The usual result is Symphony in D by C. P. E. Bach, Wg 183/1 (H. that all of the movements appear in the tonic 663, composed 1775-76, published 1780), which — key, with the slow movement providing the

features a central Largo in E-flat, selected by contrast by means of mode only. There is no Tovey for one of his Essays in Musical Analysis.*° escape from the tonic, but for at least one move-

24. Cf. Newman, The Sonata in the Classic Era, p. 138: 25. ‘Tovey, “Carl Philip Emanuel Bach, ‘Symphony in

“Of Haydn’s 39 keyboard sonatas with 3 or 4 move- D major,” Essays in Musical Analysis, vol. 6 (London: ments, over 40 per cent have no change either of key Oxford University Press, 1939), pp. 8—12—an essay, or mode in any of the movements.” See also Somfai’s according to E. Eugene Helm (Thematic Catalogue of the summary of key organization in The Keyboard Sonatas of Works of Carl Philipp Emanuel Bach [New Haven, Conn.: Joseph Haydn, pp. 205—6: “It 1s surprising how many of Yale University Press, 1989], p. 145), “containing hardly Haydn’s mature piano sonatas keep the same tonic for all a paragraph that is not misinformed.” Tovey’s interest in the movements... . [This might] suggest a more archaic the work was doubtless sparked by its slow movement’s taste in key relations between movements than that of tonal anticipation of that in Haydn’s later piano sonata (a

Mozart or other contemporaries. A better interpreta- comment that surfaces more than once in his essays). tion might be that Haydn liked a strong cohesion in the overall key structure.”

328 Elements of Sonata Theory

ment we experience its parallel other. The effect Concerto in C Minor, K. 491. Beethoven seis poignant: following an arduous first move- lected this option in his Piano Trio in C Miment we encounter a temporary liberation into nor, op. 1 no. 3. the emancipatory tonic, a premature brightening of that tonic into the major mode. Often Case 3: Submediant Major (VI). Like other mathe local impression 1s that of a dream, a false = jor-mode options for the slow movement in a hope— especially when we are given to expect minor-mode sonata, the submediant provides that we are under the “sentence” of a return to a temporary haven from the work’s prevail-

the tonic minor (the work’s “true reality”) in ing mode—1in this instance a comfortingly the minuet/scherzo or the finale. On the other cool shadow or short-lived respite (again, “if hand, if the finale is to end in the major mode, only”) from the tensions that surround it. As the slow movement can also suggest a prolepsis was the analogous case in major-mode works, of what is to come. These effects are produced the movement’s VI may be understood as a proregardless of whether the slow movement is in longed upper neighbor to an implied dom1second- or third-movement position. Examples nant (a relationship made especially clear in the from Haydn include: the Quartets in G Minor, link between the slow movement and finale of op. 20 no. 3, in F Minor, op. 20 no. 5, in B Mi- Beethoven’s Piano Concerto No. 5 in E-flat, op. nor, Op. 64 no. 2, andin D Minor, op. 76 no. 2 73, “Emperor’”) and as a coloristic chordal trans(“Quinten”); the Piano Trios in E Minor, Hob. formation of the reigning tonic. This option, XV:12, and in F-sharp Minor, Hob. XV:26; and while striking, was not as unusual at might be the Symphonies Nos. 44 1n E Minor (“Trauer’’) initially suspected in the period under considand 52 in C Minor. A Mozartian illustration 1s eration. One finds it in the works of Kleinmeister available in the Quartet in D Minor, K. 173. (Vanhal’s Symphony in G Minor [g?], probably Examples from Beethoven may be found in the written before 1771)?° as well as in those of the Piano Sonata in F Minor, op. 2 no. 1, and the master composers. From Haydn: the Piano So-

Violin Sonata in A Minor, op. 23. nata in C Minor, Hob. XVI:20; the Quartet in D Minor, op. 42; the Piano Trio in D Minor, Case 2: Mediant Major (III). The tonal logic of | Hob. XV:23; and the Symphonies No. 39 in G

moving to III for the slow movement requires Minor, No. 80 in D Minor, and No. 83 in G little comment. This is an obvious choice, Minor (“Hen”). From Mozart: the two G-micombining the desirable escape away from the nor Symphonies, No. 25, K. 183, and No. 40, tonic with the shift into the major mode. Ad- K. 550; the A-minor Piano Sonata, K. 310; the ditionally, the move to III for the slow move- G-minor String Quintet, K. 516; the D-minor ment activates tonal allusions to what may have Piano Concerto, K. 466; and several other been the key of the first movement’s S and C in works. Within Beethoven’s works, the Ninth the exposition. Examples from Haydn’s quar- Symphony has many precedents in this regard tets are to be found in those in B Minor, op. 33 (the “Pathétique” Sonata, the Fifth Symphony, no. 1, and in F-sharp Minor, op. 50 no. 4; from and several others). the symphonies, in No. 45 in F-sharp Minor (“Farewell”), No. 78 in C Minor, and No. 95 Case 4: Tonic Minor. When the slow movement in C Minor. Examples from Mozart include retains the tonic and mode of the first movethe String Quartet in D Minor, K. 421, the ment, this normally indicates not only that there Piano Sonata in C Minor, K. 457, the Piano is no nontonic “escape” to be experienced in Quartet in G Minor, K. 478, and the Piano this work but that all three or four movements

26. This dating was suggested by Paul Bryan in the to be “generally an unusual tonal-level relationship, but Preface to Johann Vanhal, Six Symphonies: Part 1 (Madi- one in fact frequently found in G-minor symphonies, son, Wisc.: A—R Editions, 1985), p. 1x. Bryan also con- such as Mozart’s Symphony K. 183 and Haydn’s Hob. sidered the use of the submediant for the slow movement [:39” (p. x).

The Three- and Four-Movement Sonata Cycle 329

will unfold in the same tonic minor. This is an as his Piano Sonata in A Minor, D. 537 (op. 164 intensely negative statement, suggesting a grief from 1817), with its second movement, Alleso profound that it cannot be shaken. One finds eretto quasi Andantino, in E major (major V), it only in self-consciously extreme works, such or his B-minor “Unfinished” Symphony, with as Haydn’s Symphony No. 49 in F Minor, “La its Andante con moto in E major (major IV), passione,” and Clementi’s Piano Sonata in G the first thing to be noticed 1s the sheer “imposMinor, op. 50 no. 3 (published in 1821), “Di- sibility” of their keys: these are key-selection

done abbandonata.” deformations, dreaming off into fantasy-spaces.

The usual sense of escape is compounded with Case 5: Unusual or Deformational Tonal Choices: a denial of the modal experience of the first v, ww, V, IV, and So On. By analogy with the movement. If one misses this, one misses ev-

slow movement in the relative major (which erything.?’ recalls the probable secondary key of the first Also relatable to these interpretations are such movement), one might suppose that the other cases as one finds in Beethoven’s Quartet in F normative secondary key, the minor dominant, Munor, op. 95, “Serioso.” Here movements 1, 3, might also turn up as the governing tonic of | and 4 are in the tonic F minor, while the “slow a slow movement. In fact, this 1s an extremely movement,” Allegretto ma non troppo, is in D rare choice. One may find it in Clementi’s Piano major. (The scherzo also contains a D-major Sonata in F Minor, op. 13 no. 6, which includes trio.) D major is not smoothly accessible via F a C-minor middle movement. (Curiously, how- minor, but it would have been available more ever, the first movement’s exposition had mod- easily as VI— even as relatively normative praculated to III, not to v.) Also unusual 1s a move tice in this period of Beethoven’s output— had to the minor subdominant (iv), which may also the governing key of the whole piece been F be found in Clementi’s Sonata in F-sharp Mi- major rather than F minor. Reflections along nor, op. 25 no. 5, with its B-minor slow move- these lines permit one to interpret the latent ment. One presumes that the subdominant mi- irony behind deformational key choices. nor is accessed by analogy with the major-mode sonata’s predilection for its (major-mode) sub-

dominant— suggesting a forlorn attempt to ap- Minuet/Scherzo ply major-mode psychology in an alien, minor-

mode landscape. It cannot be the purpose of this book to detail the

Such an impression is even more affecting practices of the normal and deformational minuet when the slow-movement key is a major key (or scherzo) and trio. Here we shall mention only that would be properly available only if the first a few things about its often-implicit relation to movement had itself been in major (which is the first movement and its larger role in the mulnot the case here). In this situation there is an — timovement sonata. A minuet (or later, scherzo) element of self-deception represented (“This is is normally in the tonic key and, especially from how the second movement might have been had the later 1780s onward, most regularly appears the circumstances of the first movement been after the slow movement. This makes it the third otherwise’). Classic instances of such “false ma- movement of a four-movement plan—its most jor” keys taking their bearings from the parallel familiar placement— or sometimes, particularly major of the first-movement tonic may be found in earlier or lighter pieces, the last movement of in Schubert. When we confront such examples —_ a three-movement plan. From time to time one

27. An almost perfect inverse of the Schubert “Un- “normal” major-mode IV is negatively inflected, castfinished” situation, more properly belonging under ma- ing a “Romantic” minor-mode shadow over this movejor-key sonata deformations, appears in Mendelssohn’s ment—something that will find resonances later in the

Symphony No. 4 1n A Major, op. 90, “Italian,” with work, which will end with an A-minor finale. its slow movement in D minor (minor iv). Here the

330 Elements of Sonata Theory

finds a minuet instead of a slow movement in a ment, but on different terms. One should be atthree-movement work; more typically, though, tentive to any thematic or motivic resemblances

it is the minuet that is absent, producing the that might exist between the minuet and the common F—S-—F pattern. As mentioned earlier, first movement—as though aspects of the first it is also common to find a four-movement work were re-emerging here as a new, transformed with a minuet in the second-movement position, beginning (perhaps with rotational implications

before the slow movement. on the multimovement level).

A few of Beethoven’s scherzos deviate from

Potential Correspondences with the the customary form. Some of them have a

First Movement problematized tonal plans than one might expect

to find in a comparable late-eighteenth-century

In its normative third-movement position the movement. On occasion Beethoven cast aside minuet or scherzo re-establishes a principle of — certain traditions altogether. The scherzo secschematic order after the typical “escape” of the tions of the Fifth Symphony —reworking the slow movement. When the slow movement had “Fate” rhythm of the first movement — articubeen in a nontonic key, the minuet normally _late an ad hoc, quasi-triple-rotational form subrestores the tonic. In addition, the minuet/ stantially removed from the normative rounded scherzo returns not merely to obligatory struc- binary structure with repeats. In comparison, ture but to a preformatted array of obligatory the ensuing structure of the trio (though not its binary forms. This produces a virtually invari- content) is more normative. able pattern, usually consisting of two smaller When Beethoven, and then later composers, rounded binaries, one of which, heard twice, began to write scherzos that as a whole movement encloses (contains) the other. The container is are in dialogue more with sonata form than the also referred to separately as the “minuet” or the usual scherzo-trio form (as in the second move“scherzo’’; what is contained is the “trio.” The ments of Beethoven’s Quartet in F, op. 59 no. 1, resulting form, minuet/trio/minuet da capo, 1s and Mendelssohn’s Symphony No. 3 in A M1-

a compound ternary structure (M—T-M or nor, op. 56, “Scottish”), this may be taken as A-—B-A). Minuets or scherzos with two dif- an extension of the often-implicit principle of ferent trios [M—T!—M-—T?-—M] are possible, regenerating certain features of the first move-

as in some of Mozart’s serenades, but the prac- ment. The “entire-scherzo-as-sonata” should tice is almost unheard of in higher-prestige, not be confused with the situation in which four-movement works. In Beethoven (most fa- only the outer “scherzo” (or “minuet”’) section mously, perhaps, in Symphonies No. 4 and 7) (not including the trio) is devised as a multiwe sometimes find the same trio being visited thematic sonata and subjected to a complete twice, S—T—S—T-S, but this is a less frequent repeat in the da capo. The still-compact minpattern to be perceived as an expansion, one uet-section of Mozart’s Quartet in G, K. 387,

with rotational implications. famously anticipates this latter condition, as do In short, the first movement’s ruling shape a number of other extended minuets. A more (“grand binary”) 1s multiply resuscitated here in expanded example may be found in the scherzo a compact succession of smaller binary formats. of Beethoven’s Ninth Symphony. (On the other This is underscored by the minuet’s sharing of — hand, the Ninth’s trio, while hardly orthodox

(or return to) the key of the first movement.?° in the eighteenth-century sense, more clearly The minuet/scherzo represents the return of alludes to the proportions found in the standard many of the main principles of the first move- rounded binary form.)

28. Beethoven sometimes problematized the difficulty closure) in the second half of the minuet (scherzo) secof reattaining this tonic. The scherzos of Symphonies tion often becomes a highly charged, dramatic effect in Nos. 1 and 3, for example, levitate up to the dominant Beethoven: not infrequently, the seeming reprises of the before solidifying the first “real” tonic of the movement. “scherzo”’ theme begin in the “wrong key.” Similarly, the attaining of the ESC (essential structural

The Three- and Four-Movement Sonata Cycle 331

Expressive Connotations: Minuets and ward those of a more universal public. With the

Scherzos; Major and Minor Modes increasing delegitimation of aristocratic privia is. normally lege ina movement the agegiven of over modernity, the minuet’s more This to an _ . traditionally formal mood would be the most

enhanced charm and. usually to a stylized . a. . vulnerablealso convention within a multimovedance, .rhythm, or mood. In the eighteenth _ ment work. Beethoven would transform it into century, 1a: the3 minuet movement is the one most 3,_ , aan “scherzo,” whichsocial harbors aspects of- .a, .critique saturated with “obligatory” connota. . of the traditional minuet and its social connotation. This centers around the social norms of . (“old-world”’) aristocraticminuets society;and breeding and ;.. . Major-mode scherzos are typielegance; public expression; controlled, ritual. _. . . - . .eroticism; cally bright, positivecontainments statements, often reveling ized formalized of the . . , . . in elegant compositional workmanship. Those

« oe . . . tions.

pairing sexes; andmode the like. Inquite Zaslaw’s . . . matter. ~ “the . of in the the minor are another view, minuets stand for the courtly side of i. aewhen _ . . . This impression isand especially striking a eighteenth-century life, an old-fashioned . ae first . . third movement restores the tonic minor and formal aspect of it at that. The trios, on . 1. . . with asserted in the first movement but put aside in the other hand, often deal the antic, thus

, . . . a major-mode slow return standing in relation to themovement: minuet asthe an anti- of . .a . 4: negative condition that had been temporarily

masque. >to. its masque, providingwith [an] nineteenth-cenele- , ; . kept at bay.and Particularly ment of caricature.”2? It is worth observing that . . a: . . ;atury works, the minor mode and itscall concom1within multimovement work the minuet’s . . . . tant treatment can result in such special effects back to.tonic-order occurs simultaneously with , _ sy cc to“delkas ; . predatory,” “trapped,” “nightmarish,” or that movement’s underscored assertion of the 45

. , , monic” scherzos (Beethoven, Schubert— much

. ater, Mahler).

privileged social norms of the aristocracy. later, Mahler) 3! On the other hand, once the minuet became

more of an abstraction in a multimovement work, it took on a life of its own. It became The Nontonic Minuet/Scherzo

a musical genreIt subjected to happens, the compositional . ae only 8 a come occasionally though initially

craft something . . . asofa style-variation, deformation toward the end to of be themaeigh-

nipulated with wit and skill. Thus arose some

. . the teenth century, the movement expected subtypes of minuet: thethat canonic, fugal,isordisplaced . 4. og Oe to reinstate tonic-key tonal order otherwise minuet displaythis of com. . we are - a, into“learned” a nontonic key.(a When happens positional or contrapuntal ingenuity in the . eqec: In the . usually facing one of three possibilities.

. first— within a three-movement work—the thetic minor-mode andmovement so on. More, ,have , . . minuetminuet; as middle might taken manner of a scholastic game); the stormy or pa- a

over, some late-eighteenth-century symphonic . .

. . . onresemble the role of and tonal escape fromton the, —_ tonic, as in minuets faster far less aristocratic , 3 oe, Mozart’s Piano (Clarinet) ‘Trio in E-flat, K. 498, German dances.’’° This signified a shift of the , ol , ,

. Kegelstatt” (with Menuetto in B-flat, includgenre away. jfrom aristocratic connotations to. ... ing a G-minor trio). In the second— consider-

29. Zaslaw, Mozart’s Symphonies, p. 417. includes a bibliography of related articles concerning the 30. William Malloch, “Toward a ‘New’ (Old) Minuet,” subsequent controversy over minuet tempos.

Opus 1/5 (1985), 14-21, 52. E.g., from p. 16: “a real 31. Cf. Schopenhauer’s nineteenth-century reactions to whirling dance; heady and athletic, with nothing lumpish both dance music and minor keys: “The short, intelliabout it, it moves like the wind... . The new Deutsche gible phrases of rapid dance music seem to speak only

carried with it a nice antiroyalty dig... . And there is of ordinary happiness which is easy of attainment... . plenty of evidence to indicate that the ‘minuets’ in the Dance music in the minor key seems to express the later symphonies of Haydn and Mozart were simply ur- failure of the trifling happiness that we ought rather ban German dances in minuet costume, and that these to disdain; it appears to speak of the attainment of a dances were played extremely fast by our standards.” Cf. low end with toil and trouble.” The World as Will and Malloch, “The Minuets of Haydn and Mozart: Goblins Representation, trans. E. F. J. Payne (New York: Dover, or Elephants?” Early Music 21 (1993), 437-45, which 1966), 1:260-—61.

332 Elements of Sonata Theory

ing now the four-movement work—it may be [VI]); and No. 104 in D, “London” (trio in Bthat the slow movement, in whichever move- flat [bVI]).33 Second, the trio could introduce a ment-position it might occur, 1s itself in the modal shift from the minuet’s tonic minor to the tonic. Here the minuet and the slow movement trio’s tonic major, or vice versa.°4 In minor-key will have exchanged their more normative tonal works, tonic-major trios forecast the major-mode roles: one finds a tonic slow movement and a emancipation that one hopes might be achieved

nontonic minuet or scherzo, as in Beethoven’s in the finale. Within the limited boundaries of A-major Seventh Symphony, whose scherzo is the minuet or scherzo, however, such a majorin F, and Schubert’s early (and incomplete) Pi- mode passage is fated to be obliterated by the ano Sonata in E, D. 157, with a minuet in B. return to the minor with the da capo. (In Mo(In both of these instances the preceding slow zart’s Symphony No. 40 in G Minor, K. 550, movement is in the tonic minor.) With only a the G-major trio dreams of a condition that will few exceptions, within four-movement works never come to pass in the finale; in Beethoven’s this deformation is limited to the nineteenth Fifth and Ninth Symphonies the tonic-major trio century.°? In the third, the slow movement and _ foreshadows the major mode to be secured in the the minuet are placed in two differing nontonic finales.) Third, the trio could be centered on a keys, as in Beethoven’s String Quartet in E-flat, contrasting tonic to that of the minuet (perhaps op. 74, “Harp,” whose four movements are in its subdominant). When this happens, the trio’s E-flat, A-flat (slow), C minor (scherzo), and key may be —although it need not be—that of E-flat, and Schubert’s Fifth Symphony, D. 485, the slow movement, thereby establishing conwhere the succession is B-flat, E-flat (slow), G nections to that movement. This happens occaminor (minuet), and B-flat. We shall return to sionally in earlier Haydn, as in the Symphonies this topic in the final section of this chapter in a No. 35 in B-flat (slow movement and trio in

review of the complete cycle as a whole. E-flat); No. 38 in C (slow movement and trio

; in F),9°of the Trio . . The Key and Character

in F); and No. 56 in C (slow movement and trio

Structurally, the trio normally mirrors the The choice of the key for the trio was not oblig- rounded binary format of the minuet that sur-

atory, but there were some standard options. rounds it, although the trio is often simpler, The most common was to present the trio in the even more compact in its phrase structure. The same tonic and mode as the minuet (and as the trio is usually more relaxed in mood than is the work as a whole, continuing to reaffirm the re- minuet: it is typically simpler, calmer, more newed authority of that tonic). Within Haydn’s rustic or folk-like. Many variants can be rung symphonies this was the almost invariable sym- on this, but if a trio turns away from its usual

phonic practice from Symphony No. 66 on- character—into something brusque, complex, ward. Two late-period exceptions are from the learned, or frenetic—we are probably dealLondon Symphonies: No. 99 in E-flat (trio in C ing with low-level default or a deformation. 32. One can find a few examples from the eighteenth related” key possibility for trios emerges in the 1790s: century, but typically in such out-of-the-way places as for instance, in the Quartets, op. 74 no. 1 in C (trio in program symphonies, httle-known repertory, and the A) and Op. 74, No. 2 in F (trio in D-flat). like. For example, the first of Dittersdorf’s Symphonies 34. Asin Haydn Symphonies No. 44, “Trauer” (E mion Ovid’s Metamorphoses (Symphony No. 1 in C, “The nor with E-major trio), No. 46 (B major with B-minor Four Ages of the World,” c. 1781) is a four-movement trio), 48, “Maria Theresia” (C major with C-minor work whose third movement is a “Minuetto con garbo” trio), 49 “La Passione” (F minor with F-major trio),

in A minor. (Its trio is also in A minor.) and No. 65 (A major with A-minor trio). 33. There are also a few odd instances of this in the 35. Again, this practice was anything but invariable. middle Haydn symphonies: No. 39 in G Minor (trio None of the contrasting-key trios mentioned in n. 33, as in B-flat); No. 43 in E-flat, “Mercury” (trio in C mi- well as those of Symphonies No. 99 and 104 (mentioned nor); and No. 62 in D (trio in G—the one escape from in the related text), reinstate the key of their preceding the tonic, since the slow movement is also in D). As slow movements. with his choice of keys for slow movements, the “third-

The Three- and Four-Movement Sonata Cycle 333

The most celebrated example 1s the trio in the in its second, post-trio statement will this same scherzo of Beethoven’s Fifth Symphony, with its I:PAC be justifiably regarded as the real ESC imitative, C-major fugato probably represent- of the entire movement. Conceivably, the ESC ing a panicked “flight” (Latin, fuga—punning potential of even this second appearance of the on the term) from the threat of the C-minor I:PAC might be deferred, if the piece proceeds to

cadence that sets it off. either a second trio or another statement of the first one, both of which strategies are ESC-de-

Deferral of the ESC in Compound laying tactics. In these instances the ESC-ettect is deferred again to its third statement, which must be regarded as the real ESC: there are no

Ternary (ABA) Forms

Because the rounded binary structure of the more trio options left within the genre. minuet- (or scherzo-) section of this movement,

considered apart from the trio, is a miniature

version of the “grand binary” or sonata-form Finale structure of the first, the concepts of the EEC and ESC are applicable to it. In the rounded-bi- = Tonally, the last movement of a multimove-

nary “minuct” the EEC-equivalent would typi- ment sonata serves as a final grounding or cally occur at the first satisfactory PAC toward reaffirmation of the tonic key. Like the slow the end of the first part, before the first repeat | movement, the finale has a nonobligatory strucsien. (This PAC may or may not be in the tonic, ture: 1t may be laid out in any of a number of and an extra phrase or “codetta-like” extension, formal patterns. Summarizing the standard analogous to an exposition’s C-space, might choices in nonorchestral sonatas over some s1x also be appended to it). Similarly, with the sec- or seven decades, Newman surveyed the opond-part return of all or some of this first-part tions as follows: music, the ESC would occur at the parallel point

at or near the end of minuet section, before the The minuet served as the finale in countless second repeat sign and just before the trio. (In [three-movement] pre-Classic and not a few this case, the ESC-equivalent is marked by an high-Classic sonatas. . . . Along with dance

authentic cadence in the tonic.) movements [including the polonaise, the march, During the first, pre-trio statement of the the gigue, and others], the most frequent finales “minuet/scherzo” this ESC-effect should be the rondeaux and rondos, the sets of brillant

_ — display variations, the incipient or larger-scale

regarded as only provisional. This is because of ‘sonata forms,’ about on a par with the first quick our awareness of the genre in which it is par- movements, and various combinations of these. ticipating, which encompasses the entire formal A relatively small number of fugal finales can be pattern, minuet/trio/minuet da capo. (Our fore- found in sonatas throughout the Classic Era.*° knowledge of the genre provides the guidelines for understanding any individual exemplar.) Be- Finales of all of these types may be found in

cause we know that this first section 1s not to the works of Haydn, Mozart, and Beethoven. provide the immediate end to the movement, Still, the choice among them was anything but we may arrive at two conclusions. First, how- _—_ arbitrary. The main issue at hand was the desired

ever resolute the minuet section’s ESC-effect, weightiness of the whole piece, to which the it is not strong enough—yet—to suppress the finale makes an obviously large contribution. trio and finish off the movement on the spot. Some finales are extremely light (dance-moveAnd second, because the genre dictates that this ment related); others are more substantial, preinitial section 1s to come back literally (or nearly senting sonata forms (and even the occasional so) as a da capo, we also understand that only fugue) of various complexities and implications.

36. Newman, The Sonata in the Classic Era, pp. 161, 164.

334 Elements of Sonata Theory

Rondo finales occupied the broad middle of this sometimes mark the arrival of an expressive scale. Simpler rondos of alternating themes oc- world that is more elemental, rustic, direct, cupied the lhghter side; rondos intermixed with or folk-like— more fundamentally “natural” the sonata (sometimes producing Type 4 sonatas or stable—than was that of the first moveof uncommon subtlety) made higher claims. ment. In general, as Michael Talbot observed in

But even such generalizations need a recent monograph on concluding movements, qualification. Seemingly simpler generic choices “Whereas ‘long’ metres such as 4/4 and 3/4 are could be composed on a grand scale or be placed characteristic of opening movements, ‘short’ me-

in compositions whose preceding movements tres such as 2/2, 2/4, 3/8, and 6/8 belong more were conceived along the most elevated and to finales.”°8 Moreover, finale themes often also searching lines, all of which enhanced their own give the impression of what Talbot called “reimplications. One thinks of the substantial vari- eression”: a “going back” that “often reaches ations concluding Mozart’s Quartet in D Minor, back beyond previously heard material to strive K. 421, or Piano Concerto No. 24 in C Minor, for something more basic. It is, so to speak, deK. 491—not to mention those in Beethoven’s velopment by stripping down rather than by the Eroica Symphony. On the other hand, many more usual process of elaboration.’’? eighteenth-century sonata-form finales present When the finale was a sonata form (usually a lighter, more entertaining tone—suggesting a Type 3 or Type 4 sonata; much more rarely a playful wrap-up to the whole piece, an ef- a Type 2, as in Haydn’s Symphony No. 44 in fervescent display of wit, charm, or skill that, E Minor, “Trauer,” and Mozart’s Eine Kleine while appropriate to the large-scale trajectory Nachtmusik), it provided a sense of recasting of the whole work, only rarely seemed to bear certain features of the first movement. The two the same conceptual weight as that offered by tonic-key “bookends” of such a multimovement

the first movement. work invite us to discern a relationship or po-

Characteristically lighter choices of tempo, tential balance between them. This is especially meter, and style for the thematic subjects of so- apparent with Type 3 sonata finales, which also nata and rondo finales became second nature restore the usual repeat-sign conventions found among cighteenth-century composers. Zaslaw’s the first movement.?9 This concluding return to generalizations about Mozart’s symphonies as a a reaffirmation of sumptuous balances and arwhole are typical: “The finales [like those of — chitectural symmetries is an important element many other composers] are generally based on of many finales. rustic or popular dances: gavottes, contredanses,

. . . of the Whole Work

Jee, OF quick SEEPS: ~” Finales of this period also The Role of the Finale in the Trajectory often have light, jocular, or humorous conno-

tations. Dance-like, triple-time finales (3/8, 6/8) are particularly common, especially in Much scholarly reflection has been applied to

the mid-eighteenth century, and toward the the issue of the comparative weights of the end of the century (especially in Haydn) finales first and last movements in eighteenth-century

37. Zaslaw, Mozart’s Symphonies, p. 417. which simple repetition is preferrred to elaboration or 38. Talbot, The Finale in Western Instrumental Music, p. complex development . . . an effusion of virtuosic bril55, also crediting Wilhelm Seidel with the observation liance .. . [and/or] a humorous or quirky tone” (p. 8).

in the latter’s “Schnell—Langsam—Schnell” (see n. 1 40. We might mention once again (see chapter 2, n.

above). 12) that even though in 1826 Reicha wrote that some

39. Talbot, The Finale in Western Instrumental Music, p. “erand-binary” finales —sonata-form finales —can dis-

38. Cf. p. 8: “the stripping down of thematic material pense with the expositional repeat, in practice among into even more basic shapes, a kind of ‘reverse’ develop- the major composers this is rather infrequent, though

ment.” The traditional sense of lightness was also typi- not nonexistent: see, for example, Haydn’s Quarcally achieved “by a quicker tempo...ashorter metre... tets in F, op. 74 no. 2, andin D Minor, op. 76 no. 2, a simpler texture .. . a more quadratic, or at least more “Quinten.”). transparent, phrase structure .. . a formal structure in

The Three- and Four-Movement Sonata Cycle 335

three- and four-movement works. In large part of fugal (or otherwise highly contrapuntal) this is because of the early-nineteenth-century finales with their “effect of culmination,” as in shift of weight toward the apotheosis or cli- Haydn’s Symphony No. 40 in F, in three of the mactic finale found in Beethoven and subse- Quartets from op. 20 (no. 21n C; no. 5 in F Miquent composers—the rise of the “finale sym- nor; no. 6 in A), and in the Quartet in F-sharp, phony.’#! ‘The orthodox view has been that the op. 50 no. 4; and in Mozart’s Quartets in F, K. multimovement work began its historical course 168, in D Minor, K. 173, and in G, K. 387, and in the mid-eighteenth century as something his Symphony No. 41 in C, K. 551, “Jupiter.”* weighted toward its first movement (resulting

in a sequence of movements lightening toward Major- and Minor-Mode Finales in

. Munor-Mode Works

a mannered playfulness at its end) and by deerees, and sporadically, ratcheted up the heft of

its finale to a position of rough equality with Turning things brighter in the finale, followthe first movement—say, by the period of later ing a minor-mode first movement and minuet/ Haydn and Mozart—and, eventually, to that of | scherzo, is a commonly selected option (and one the high point to which an entire composition of Webster’s indicators of “the finale as culmi-

ascended. nation”). Haydn’s Symphony No. 95 in C Mi-

Although the general outlines of this account nor ends with a C-major finale, as does, most are accurate, one should neither minimize the famously, Beethoven’s Fifth Symphony. Even contributions of Beethoven’s predecessors nor if the finale to a minor-mode work begins in overlook the substance of many eighteenth-cen- the tonic minor, it may well end in the major tury finales. Webster has mounted a telling ar- mode, spurred onward by the psychology outcument that the idea of “the finale as a culmi- lined in chapter 14. But it 1s also normative to nation” (the result of a “through-composed find a finale to a minor-key work that both beeffect”) was explored at length by Haydn, Mo- gins and ends in the minor mode, thus carrying zart, and other composers from the 1760s on- the negative connotations throughout the whole ward. “The paradigmatic example remains work. This procedure stages the finale as a negaBeethoven’s Symphony No. 5... . But the fea- tive culmination or an expression of unrelieved tures on which such effects depend were by no despair, as often in Mozart. Although Mozart means unknown during the eighteenth cen- occasionally does write major-mode finales to tury.” The features that Webster had in mind minor-mode works (most controversially in (and proceeded to illustrate in several analyses the G-minor String Quintet, K. 516), his much of Haydn) included “run-on movements,” “re- more frequent practice is to retain the minor calls of earlier movements,” “the transformation mode all the way to the bitter end. To be sure, of minor into major,” “prominent and unusual a small number of his obsessively minor-mode tonal relations,” “an impression of incomplete- finales are released into the major mode at their ness of unfulfilled potential before the finale,” conclusions (Piano Concerto in D Minor, K. and “a mood of tension or irresolution [in ear- 466). This turn to major, however, typically lier movements].” Not surprisingly, Webster occurs in the coda, not in the sonata or sonaeranted a special importance to the emergence ta-rondo form proper. 41. A brief review of this perennial “motif” within dis- aiming “to sum up the cycle as a whole,” including “a cussions of multimovement works is provided in Tal- more overt thematic (and hermeneutic) relationship to bot, The Finale in Western Instrumental Music, pp. 12-15, the earlier movements” [p. 50]; or “valedictory” (char-

170-71. acterized by a slow tempo, where the sense of “[psycho-

42. Webster, Haydn’s “Farewell” Symphony and the Idea of logical] homecoming is pushed back still further —beClassical Style, pp. 184-85. Cf. Michael Talbot’s three- yond the boundary of the work, in fact” [p. 198]. Talbot

fold categorization of finale types circa 1700-1900 (and (p. 64), apparently contra Webster, was more likely to beyond) in The Finale in Western Instrumental Music. Por regard eighteenth-century fugal finales as “relaxant”

Talbot, finales may be classified primarily as “relaxant” rather than “summative” on the basis of what he re(“inducing relaxation” [p. 50]; “summative” (weightier, garded as their ironic or parodistic tone.

336 Elements of Sonata Theory

Is it conceivable to conclude a major-mode poral space. In part this is because the movework with a minor-mode finale, suggesting an ment successions themselves suggest possibilities

unanticipated reversal of fortune? There are a for aesthetically pleasing inner relationships, few (very few) early instances of this, occasion- self-references, and balances. ally underpinned with programmatic implications. Some are liberated into the major mode The Three-Movement Pattern toward the very end. Such is the case with Dittersdorf’s Symphony No. 6 in A after Ovid’s In a three-movement cycle, fast—slow-—fast Metamorphoses, “The Transformation of the Ly- (F—S—F), the obvious implication is that of a cian Peasants into Frogs” (c. 1781-82), in which balanced, arch-like shape with rapid-tempo the unfortunate metamorphosis is reflected upon outer sections and a slower, more lyrical cenin an A-minor finale that turns, at the end, into ter providing a space of contrast. The first and an A-major, diminuendo fade-out. The situation last movements are consequently thrown into may also be found in Haydn’s Quartet in G, op. an inevitable correspondence as comparable 76 no. 1, in which the finale’s G-minor sonata, “bookends” —tempo, length, beginning and Allegro ma non troppo, converts into the ma- ending positions. The relationship between the

jor mode for the recapitulation. On the other two movements may be made more palpably hand, Boccherint’s spectral Symphony in D Mi- present through musical interconnections, innor, op. 12 no. 4 (G. 506, 1771), “La casa del cluding similar structural shapes. diavolo,” is a three-movement work in which The chiastic (and as such static) A-B-—A' a D-minor introduction leads to a D-major so- implication of the three-movement plan 1s ennata form for the first movement,* while the riched by the simultaneous overlay of linear finale both begins and ends in D minor. There concerns. If the finale’s tempo 1s faster than that are also some parallel examples from the nine- of the first movement, the effect is that of a pasteenth century. One 1s Berlioz’s Harold en Italie sage through an affective, slow center—and (a G-major work with a G-minor finale enti- sometimes also an “escape” tonality —in order tled “Orgie de brigands” —although one end- to arrive at an increased exhilaration. The slow ing with a blazing conclusion in G major). Oth- movement can be understood a site of transforers, more pessimistically, conclude in the minor mation: a process to pass through in order to mode, as in Mendelssohn’s Symphony No. 4 arrive at the heightened spirits of the finale. If in A, op. 90, “Italian” (the finale is the famous that finale is also lighter in tone, style, or strucsaltarello, Presto, a frenzied dance — or better, ture, more dance-like or nimble, then the lin-

death-dance—in A minor that persists to the ear effect of a multimovement pathway into end) and Brahms’s Piano ‘Trio No. 1 in B, Op. an enhanced vivacity is underscored further. 8, which in both versions features a B-minor |§When the three-movement pattern deploys a

finale that also ends in that key. minuet or other dance movement as its finale (fast—slow—minuct), the linear aspect takes on

a different connotation: now the whole piece

The Multimovement Cycle stages its progress from a dynamic sonata-form as a Complete Gesture world into one of stylized, ancien-régime grace and short-winded galant symmetries. In this Apart from their realizations in any specific case the pathway brings us to the affirmation work, the familiar three- and four-movement of high-prestige (if slightly empty) aristocratic patterns can be viewed as generic wholes, satis- stabilities. fying ways of building a musical span over tem-

43. This would suggest that the work would more normally be called a Symphony in D Major, although its minor-mode finale may have overturned that option.

The Three- and Four-Movement Sonata Cycle 337

Multimovement ESC Deferral ment is in a nontonic key, one can perceive

; ; 1: ;of two complementary tonalotherwise arcs: tonic. to nonAnother aspect linearity within the P , .to ,rounded” .. ; tonic three-movement (first and secondplan movements) and tonic (A—B-—A') . . . clear . . tonic (minuet and finale). Even without —an aspect applicable to all multimovement .

. thematicour correspondences between the; slow successions—is awareness that no first . movement and the finale, the tonal trajectory movement can provide large-scale closure on . .of .itsthe whole suggests the displacement-correction own. Since we know that other of movements .general. . ,aware pattern typical sonata form in Par-. are to follow, we are that the entire piece . -., .is .not ticularly if all of this is reinforced by thematic yet brought fully, or to rest. Indeed, the tonic .there . may be broad interconnections, secured by. rotational the first other movement’s ESC may be ee . implications in this complementarity. lost or .abandoned in the second. In these cases .; . In any event, the minuet/scherzo often provides the finaleawill be of obliged to recover the dis-(tonic, « , imfaster , sense a “return to business” placedne tonic with its own ESC. Consequently, . .calm . multimovement tempo, binary schemes) afterfirst the relative in any work the move, of a , . . of the second movement—the beginning ment’s ESC 1s only attainment, . . broad second half atoprovisional the whole work. In this revalid for that movement only. (The same point .

. spect the finale is relatable to both the first and was co made regarding the first ESC-effect 3 .above second movements: with regard. to the former,

. . ntar r.

of the “minuet” [pre-trio] section of the min- , ae . as one of the “bookends”; with regard to the . latter, as the second part of a two-movement

uet movement as a whole.) From the multimovement perspective, early-movement ESCs complementary pai

are continually deferred until they find their P YP

resting-point in the ESC of the finale, even if that finale 1s not structured in one of the sonata Alterations of the Normative Movement- or types. Only at this point are no further options Key-Order Scheme

for additional movements available. As a group, . . Movement-order and key-succession are .immultiple movements imply broader structural ; . , ; portant factors in athe multimovement

work. trajectory toward thefolast movement’s ‘This .movement _ a: Changing theESC. traditional order and

feature through-composition built into theproduces .. . keyof(or simply selecting1s alternatives)

multimovement genre qua genre. ere differing implications. We shall take up some of the options and relate each of them to the fol-

The Standard Four-Movement Pattern lowing normal-order diagram:

All of the generalizations regarding Lan . >often Slow. : egro, the tonic ~~ Z£, Sow, nontonicof three-movement pattern and ESC deferral are( escape .c(! P) readily adaptable to the four-movement pattern. 3. Minuet, tonic ~~ 4, Finale, tonic The latter also merges aspects of static roundedness (the “bookend” outer movements, of- Tonic Minuet in Second Position. Assuming a non-

ten featuring other similarities) with aspects tonic slow movement (which is usually, but not of multimovement linearity. When we con- always the case), this produces the following sider the most common four-movement pattern pattern: (fast—slow—minuet—finale), an additional factor of large-scale, 2 + 2 symmetry can be in- 1. Allegro, tonic —~ 2. Minuet (or scherzo), tonic troduced into the whole. This is suggested in 3. Slow, nontonic —~ 4. Finale, tonic figure 15.1.

In each half of the 2 + 2 format an obliga- Here one of the main features is the persistence

tory (binary-based) structure gives way CO 4 of the tonic into the second movement (the nonobligatory one, although the second Pair more normal place to move away from that key). of movements usually accomplishes this in a This results in a heavily weighted tonic-balance lighter tone. Moreover, when the second move- in the work’s first half. At times the impres-

338 Elements of Sonata Theory

First Movement Slow Movement Establishes Obligatory Establishes Often Nonobligatory | “Otherness”

tonic binary the importance nontonic structure in lyrical mood

structure of its own or tone discourse —_ (If so, a (more

temporary freedom) (Relative

“escape” from simplicity mvmt. 1’s key) becomes a norm)

balanced and “resolved” by the complementary gesture:

Minuet Finale

Reasserts Obligatory _| Aristocratic and Reaffirms Nonobligatory Reasserts

tonic-control binary erotic social tonic structure principle of

structures connotations moreflexibility personal [plural] (declines to (but some (highly ———* | indulge in the common and wit after

(more rigid, formalized, aberration of choices) the minuet?

but far ritualized) the slow less complex movement) than mvmt. 1)

NOTE: the finale often also looks back in some respects (often structural) to the first movement.

FicgurE 15.1 2+ 2 Symmetries in the Four-Movement Cycle

sion is that the tonic cannot be escaped from, “Haydn” Quartets: those in G, K. 387, in B-flat, as though it were insistently exerting its author- K. 458 (“Hunt”), andin A, K. 464, as well as in ity. In minor-mode works (with a tonic-minor, the Quartet in D, K. 499 (“Hoffmeister’’), the second-movement minuet or scherzo) this effect String Quintet in G Minor, K. 516, and several can be ominous or menacing. The possibility other works. The most “colossal” example of of an expressive escape from the tonic is conse- this option is found in Beethoven’s Ninth Symquently deferred to the (slow) third movement, phony, and it occurs in some of his other works and the burden of resecuring the tonic falls as well, such as the Piano Trio in B-flat, op. 97,

squarely on the shoulders of the finale. “Archduke.” Since this was a common option—a second-level default—1in the eighteenth and ear- Tonic Slow Movement in (Normal) Second-Movely-nineteenth centuries, it is anything but de- ment Position; Nontonic Minuet/Scherzo in (Norformational. As mentioned earlier, three of — mal) Third Movement Position. ‘This option is Haydn’s six quartets from op. 20, feature the rare — virtually nonexistent— 1n the eighteenth minuet/slow-movement switch, No. 1 in E-flat, century. In the nineteenth century it was introNo. 3 in G Minor, and No. 5 in F Minor—al- duced as a deformation but before long became

though in two of these, No. 3 and No. 5, the a lower-level default choice: succeeding slow movement 1s also in the tonic key (the parallel major in both cases) — while in 1. Allegro, tonic ~~ 2. Slow, tonic (perhaps with a

the op. 33 set the first four place the “scherzo” switch of mode) or scherzando” in the second-movement posi- 3. Scherzo, nontonic — 4. Finale, tonic tion. One also finds it in three of Mozart’s six

The Three- and Four-Movement Sonata Cycle 339

Here the normative movement order is retained 1. Allegro, tonic ~~” 2. Scherzo, nontonic

with regard to tempo and character, but the es- 3. Slow, tonic —~ 4. Finale, tonic cape from the tonic key is assigned — unusual- (perhaps with a ly —to the scherzo, in third-movement position. switch of mode) This nontonic aspect can add to the scherzo’s flavor of caprice, although it may also suggest In this case the key-plan of the four movements

a fleeing from a heavy-handed tonic. Since the is normative (tonic/nontonic; tonic/tonic), musical material of scherzos, rhetorically, often but the tempos and styles of the two middle reinstates features from the first movement, the movements have switched positions. The secconnotations of placing the scherzo in a non- ond-movement slot retains its tonal role of protonic key are especially subtle. The tonic bur- viding the “escape” key but is occupied by the den on the first half of the work is heavy, with scherzo instead of the slow movement. In short, the corresponding obligation upon the finale to the scherzo has usurped the slow movement’s

resecure the tonic adequately. tonal role. The third movement restores the One well-known example occurs in Bee- tonic, but now with the only remaining movethoven’s Symphony No. 7 in A, op. 92. In this ment type available, the slow movement. case the second movement’s collapse to A mi- One example occurs in Beethoven’s Quartet nor is additionally telling (the “prison-house”’ in F, op. 59 no. 1, in which the second-movetonic minor). The third movement is in F major, ment scherzo is in B-flat, while the third-move-

but the first section of the scherzo—up to the ment Adagio molto e mesto is in F minor. In first repeat sign— modulates back to the work’s such a situation the slow movement, pushed to tonic, A major, as if recalling the key in which third position (and with little hope of escaping it “ought” to have been. (The sonority of that A to anontonic key, since that option has already will also pervade the trio [as 5 of D] as a station- been taken up by the scherzo), is obliged to ary, suspended vision.) Similar logic may be per- take on the responsibility of returning to the ceived in Brahms’s Symphony No. 4 in E Minor, tonic after its absence in a preceding moveop. 98. Here the Andante moderato is situated ment. Considerations of this sort may suggest in its normal, second-movement position. Span- why in this case Beethoven collapsed this key ning outward from an unharmonized E4 (thus to the “prison-house” tonic-minor— charging sustaining via linkage the “E”’-sonority of the the théme-russe finale with the task of restoring preceding movement), its famous opening four F major. The opposite situation is found in Menbars suggest that it is about to unfold in C major delssohn’s Symphony No. 3 in A Minor, “Scot(with Ch, D4, F4, and G4), in which the E is to tish.”” Here we have a minor-mode work with serve as 3 of a non-tonic slow movement in VI. second-movement scherzo in F major, followed (This would be a normative procedure: one need by a lyrical slow movement in A major, before not insist on a “phrygian” reading here.) In the A minor 1s restored in the finale (although an second half of m. 4, however, the potential for epilogue in A major is subsequently added to C major is rejected. The tonal center of the first this). In this third-movement position, the Amovement asserts itself, as though the normative major slow movement seems like a lost dream, slow-movement pull away from the tonic had a fragile major within an overriding minor-key not been strong enough, and the movement’s context, an “if only” effect. key instantly transforms into the tonic major, E major. The “lost” C major is reinstated in the Slow Movement and Scherzo in Normal Order, but third movement, the nontonic Allegro giocoso. Each is in a Different Nontonic Key. Minuet/Scherzo and Slow Movement Exchange Po- 1. Allegro, tonic — 2. Slow, first nontonic key

sitions; Nontonic Minuet and Tonic Slow Movement. . .

Again, this is rare nonexistent in the eigh- 3. Scherzo, | en a Lsorsecond nontonic key teenth century, and it is to be regarded as a deformation in the period of Beethoven’s works.

340 Elements of Sonata Theory

This situation is much more a nineteenth- or any portion thereof, into an integrated whole century phenomenon than an eighteenth-cen- is largely the task —and to a significant extent tury one. It places the burden of restoring the a creation— of the listener. Any consideration tonic entirely on the finale, while the interior of the “coherence” problem that does not acmovements, sometimes retaining their nor- knowledge this 1s inadequate. This takes us out mative slow/scherzo order (as in the schema of the empirical realm (scientific knowledge) above), each occupy differing nontonic keys. and into that of hermeneutics (interpretation), a Third-relations among the movements are com- different mode of thinking altogether.

mon, either surrounding the initial tonic key Put another way (drawing upon strands of above and below or building one third onto an- phenomenology, Gestalt psychology, and curother, resulting in an upward arpeggiation to rent studies of cognition), human perception 1s the dominant or a downward one toward the influenced by a drive to make wholes, cohersubdominant. In all instances one is invited to ent shapes and continuities, out of otherwise speculate on the central issue at hand: what set merely successive, scattered, disparate, or parof musical or conceptual circumstances permits tial information. We seek to fill gaps, to fashion (or encourages) the scherzo not to return to the incompleteness into a recognizable totality, to

tonic? find meaningful patterns in what might otherAs mentioned earlier, one example occurs in wise be random—yin short, to make the coheBeethoven’s Quartet in E-flat, op. 74, “Harp,” in siveness that we crave. Perceptual integration 1s which the four movements outline a -IV—vi-I as much a function of the perceiver as it 1s of the scheme. The practice is more commonly found musical object. This is all the more true when in Schubert, as in Symphony No. 4 1n C Minor, our perception 1s to operate within a guidelines D. 417, “Tragic” (i—VI-III—-1), and Symphony of a genre system—such as the varying types

No. 5 in B-flat, D. 485 (I-IV—vi-I). And it is of sonata form and multimovement construcespecially frequent in Mendelssohn, especially tion—that encourages us to find the coherin the chamber music, as 1n the Octet in E-flat, ence that is presupposed by the system in the op. 20 (I—-vi—il1—I) and the Quintet No. 1 in first place. Within the enabling and constrainA, op. 18 (I-4VI-iv—I). It is somewhat more ing conditions of any genre system (the “rules common in Mendelssohn to find it coupled of the game’’) we are not to do this arbitrarily. with a scherzo displaced to second-movement We perceive (or create) this music’s coherence position. This occurs, for instance, in Mendels- in large part because we are expected to do so. sohn’s Quartets No. 1 in E-flat, op. 12, and No. Several factors can assist us in our willing5 in E-flat, op. 44 no. 3, and in the Quintet No. ness to precipitate a multimovement Gestalt. 2 in B-flat, op. 87. In the first the tonal plan 1s While many of the important internal relaI—i1— V—(vi—I); in the second, I—vi-—IV—I; in tionships within a piece can be dealt with in

the third I—vi—i1-—I. standard theoretical terms, many other factors transcend the specificities of the acoustic surface

considered alone. These factors include back-

The Role of the Listener sround expectations, relevant presuppositions, prior knowledge of normative procedures, an It would be short-sighted to presume that locat- awareness of extramusical details pertaining to ing coherence within a multimovement work is _an individual piece, and so on. In some respects only a matter of being able to locate properties _ these recall the impulses behind Ingarden’s dethought to be objectively “in” that work. Co- scriptions of the “nonsounding” elements of a herence 1s not primarily a property of “the notes work of music.44 themselves.” On the contrary, making the piece, Of these we might single out two. The first

44. Roman Ingarden, The Work of Music and the Prob- Harrell (Berkeley: University of California Press, 1986), lem of Its Identity, trans. Adam Czerniawski, ed. Jean G. esp. pp. 83-115.

The Three- and Four-Movement Sonata Cycle 341

is the imposition of any verbal or visual program sive ranges and limits established by the culture onto the work. A program may be supplied by of the galant, the culture of Enlightenment senthe composer (in varying degrees of concrete- sibility, the culture of burgeoning “genius” and

ness), as with Dittersdorf’s Symphonies on Sturm und Drang. So long as those conceptual Ovid’s Metamorphoses; it may be inferable from — boundaries are not perceived as transgressed, other available evidence (letters, anecdotes, per- virtually any work may be perceived as intersonal communications from the composer, sub- nally consistent, even in its variety.

titles, nicknames, use of characteristic musical Reflections like these can help us to reframe topics, and so on); or it may be a metaphorical what was emphasized by Zaslaw in his 1989 narrative of images or emotions (even a poetic study of Mozart’s Symphonies. Observing the idea or a Marxian Grundidee) projected onto the sparse eighteenth-century evidence that would acoustic details of the work by either a single encourage us to look for “unity and high purlistener or a community of listeners. The pre- pose as criteria for symphonies,” Zaslaw offered sumption of a background narrative can trump us more modest conclusions: “Most eighteenththe purely technical expectations of work-1m- century composers of symphonies... appear manent musical coherence. What might other- to have been less interested in. . . philosophiwise be perceived as a non sequitur or a generi- cal concerns and more in pragmatic estimates of cally transgressive event tends to be absorbed how best to entertain their audiences. For them, and interpreted as illustrative of that implicit the symphony may have worked simply by jux-

narrative. taposing movements so that changes in tempo The second is the assumption of a consistent and mood from movement to movement— and (or consistently implied) field of psychological as the century wore on, increasingly within affects that suffuses the otherwise diverse, mul- movements— offered a pleasing variety of aural timovement work, bringing the piece’s contrasts experiences.’’46

together as belonging to the same family of feel- This is the baseline, the bare minimum, of ings. This is the network of “inner relations” one’s experiences with these works. But the examong the musical ideas, sometimes presumed periences of generations with Haydn, Mozart, to be governed by a “single dominant feeling,” Beethoven, and others urge us to be alert for that was occasionally mentioned by contem- more. Three- and four-movement works, esporaries of Haydn, Mozart, or Beethoven.* pecially the most ambitious of them, invite us Minor-mode multimovement works, such as to wonder about such utterances more deeply, Haydn’s Symphony No. 44 1n E Minor, “Trauer”’ to seek more compelling implications within (“Mourning”), are among the clearest ilustra- them. In part this can be done by looking for intions of this. As an initial premise of listening, ternal consistencies and cross-references among one is invited to assume that there must be an the movements, by finding evidence (motivic, interrelatedness among the changeable affects, tonal, linear, or other) that the musical argua kinship among their differences. The listener ment is proceeding “logically,” perhaps through is encouraged to presuppose that the differing a process of developing variation, perhaps contents of the disparate movements will inhabit through strategies of linkage connecting one the same psychological world. This presupposi- idea to the next, or perhaps through some other tion is even clearer when that music’s original means demonstrable within traditional music world differs from our own in many of its cul- theory. Those even more interpretively inclined tural assumptions, as is the case with music from might go further, with Adorno and others, and the “‘classical” period. Situated in a substantially assert that traces of cultural processes are inaltered world of thought and feelings, the cur- scribed on musical technique itself, which we rent listener may “automatically” bracket what experience as proceeding through linear time he or she hears as circumscribed by the expres- in each individual work. However we regard it, 45. See, e.g., the summary in Webster, Haydn’s “Farewell” 46. Zaslaw, Mozart’s Symphonies, p. 416. Symphony and the Idea of Classical Style, pp. 179-80.

342 Elements of Sonata Theory

a fully text-adequate coherence can be drawn In literary texts, not only is the sequence full forth only by a thoughtful listener familiar with of surprising twists and turns, but indeed we ex-

the norms of the genre at hand. pect it to be so—even to the extent that if there If we take seriously the expectation that there is a continuous flow, we will look for an ulterior must be a long-range sense of coherence obtain-

; motive.*/

able from what we hear, wecorrespond should also fore-it ais ye These remarks with what

; nderin it mi

sround. .our awareness that individual moments : like to listen closely to a multimovement work of the music do not die away into loss once they ‘call i derine h te mich

have been replaced by the next audible module. (or any musical work), po WGering How If Mgnt

a be putintogether as a meaningful whole. EnterThey continue exist our memory, ; ; : ing the to acoustic surface ofcreata second movement,

ing an ongoing ofmemory contexts, thefirst condi. ; weexistence can string drawofthe of the into it: tions for;the what is currently being ; From . ; the first movement’s ideas and groundingtotonalsounded. this perspective, listening a musical work is a process of accumulation over ity remain present as a tacit backdrop against

; : which theWolfgang otherwise self-contained processes of time. In 1976 Iser described such a PROCES - ; ” the second movement can be read. (This is one situation in the reading of a literary text as “thetonic _. a %» reason why appearances of the original

synthesizing process”: —

within a nontonic movement can be so impornt—such as that electrifying moment in the

The whole text can never be perceived at any one - tf suc . ast Sh. — oy “ © , h of

time. ... The ‘object’ [that is the whole] text can Ugato-section of the © -munor funeral march O only be imagined by way of different consecutive the Eroica Symphony, mm. 135—39, when E-flat phases of reading. .. . The relation between text major, led by fortissimo French horns, briefly rises and reader is therefore quite different from that up with incomparable eftect, then with a single between object and observer: instead of a sub- blow is bludgeoned back into C-minor grief.) ject-object relationship, there is a moving view- Entering the third movement, we can draw the point which travels along inside that which it has first and second into it; and it is possible to make

to apprehend. te . the first, second, and third movements dwell The synthesizing process, however, is not spo- tacitly in the sounding fourth, which may be radic—it continues throughout every phase of ; . . understood as a reaction or response to what the journey of the wandering viewpoint. ... h ded it. Th hout all ph € th

Throughout the reading process there is a con- as Lee ce POMS Ou abe Paases ; ©

tinual interplay between [the reader’s|] modified WOT » we Can trace an ongoing conceptua nar expectations and transformed memories. .. . Each rative —a master thread— not so much in what sentence correlate contains what one might calla we literally hear as in our reconstructions of the hollow section, which looks forward to the next work’s ongoing dialogue, moment by moment, correlate, and a retrospective section, which an- with a pre-existing, flexible, and constellated swers the expectations of the preceding sentence network of generic norms—norms not only for (now part of the remembered background). Thus individual zones and individual movements, but every moment of reading 1s a dialectic of [what for multimovement works as a whole. Husserl called] protension and retention... .

47. Wolfgang Iser, The Act of Reading: A Theory of Aes-

thetic Response (Baltimore: Johns Hopkins University Press, 1978 [orig. German, Der Akt des Lesens, 1976],

pp. 108-12.

CHAPTER SIXTEEN © "ISLE 0

Sonata Types and the Type 1 Sonata

Five Sonata-Form Types nata form depends on the interpretive purposes one has in mind for doing so. There 1s no reality

Much energy has been expended in the litera- question at stake here. ture seeking to declare which eighteenth-cen- Once one takes a more sophisticated view of tury structure does and which does not qualify a genre (or a form) not as a concrete thing to as a sonata form. The label was unknown to the be found in the music proper but as regulative age of Haydn and Mozart. It has been retrojected idea guiding analytical interpretation, many of into music of that period from German-lan- the problems associated with this terminologisuage music-theoretical discourse apparently cal concern become less pressing.* What we find generated around the years 1824—40.'! Thus we as we analyze eighteenth-century instrumental have narrower definitions (such as James Web- works (along with some kinds of arias and other ster’s in the New Grove Dictionary, proposing the vocal pieces) are interrelated families of musicriterion of the “double return,” the return of — cal processes that are generically appropriate both the tonic key and main theme at the begin- for simular types of compositional situations. In ning of the recapitulation)? and more expansive some cases the related processes are functionally definitions (such as Charles Rosen’s in Sonata interchangeable: one “type” of sonata form can Forms, whose plural title, following Tovey, was stand in for another. Finales, for instance, are pointedly chosen). Such terminological ques- typically cast either as sonatas or sonata-rondos;

tions reflect nothing more (and nothing less) single-movement overtures may or may not than larger heuristic intentions. Given the his- have a development section; what we call Type tory of the term and its self-consciously ahis- 2s below are almost as common as Type 3s in torical application to eighteenth-century works, the Allegro movements of the symphonies and what one chooses to call a sonata type or a so- keyboard sonatas of J. C. Bach; and so on. 1. For the dates of the earliest usages of the term, see what we call the “Type 2 sonata,” which he regarded in-

the opening of ch. 2, especially n. 1. stead as a “binary variant of sonata form” (see ch. 17) or 2. Webster, “Sonata Form,’ New Grove Dictionary of “expanded binary” form (cf. Webster, “Sonata Form,” Music and Musicians, 2nd ed., ed. Stanley Sadie and John New Grove, p. 690).

Tyrrell (London: Macmillan, 2001), 23:688. What 3. On the regulative principle concept, see appendix 1, Webster sought to disallow as a sonata form proper is “Some Grounding Principles of Sonata Theory.” 343

344 Elements of Sonata Theory

We acknowledge this relatedness of family episodic substitutions for certain expected elresemblances by housing them all under the idea ements (especially early ones, such as P-based of differing sonata-form types. All of the types material). Unless one keeps in mind the undershare similar structural principles. These include: lying rotational basis of the structure, along with a modulatory, expositional layout consisting of | amemory of prototypical examples, it is easy to

functionally differentiated modules; a struc- misunderstand the architecture of this sonata ture-determining dialogue with the principle type. At their conclusions, Type 2 sonatas may of large-scale rotation; and the need for a qua- also be provided with a post-tonal-resolution si-symmetrical tonal resolution in the last sona- (post-second-rotation) coda based on P, which ta-space rotation. From this perspective, sonata- can give rise to the misconstrued impression of form-related structures may be partitioned into a “reversed recapitulation” or a “mirror form” five broad categories, five different types. To sonata. This issue is discussed more thoroughly

avoid the sometimes unhelpful connotations in chapter 17. of prior terminology, we designate these types Type 3 sonatas are the standard “textbook” only with numbers. In brief (putting aside sub- structures, with expositions, developments, and

types and internal complications): recapitulations that normally begin with P in Type 1 sonatas are those that contain only an the tonic. (At times Type 1s with modestly exexposition and a recapitulation, with no link or panded retransitional links connecting the exonly a minimal link between them. These have position to the recapitulation become virtually been referred to as “sonatas without develop- indistinguishable from Type 3s with small dement” (or instances of “exposition-recapitula- velopment sections. In these instances the cattion form,” “slow-movement sonata form,” or egories of Type 1 and Type 3 shade into each the “sonatina’’). Type 1s normally lack internal other.) Because the Type 3 is the most familiar repeats. Fast-tempo examples of the Type 1 so- type of sonata, and because the preceding chapnata include Mozart’s Overture to The Marriage ters have dealt with its many possible realiza-

of Figaro and most of Rossini’s overtures. tions and deformations, it does not require adType 2 sonatas are (to use terms that we shall ditional treatment here.

replace in chapter 17) those “binary” (or “bi- Type 4 sonatas are the differing types of sonary variant’) structures in which what others nata-rondos. Along with that of the Type 3 sohave called the “recapitulation” begins not with nata, the sonata-rondo format was a frequently

the onset of the primary theme (P) but sub- selected option in many symphony, concerto, stantially after that point, most commonly at or chamber-music, and solo-sonata finales, as well around the secondary theme (S). Like Type 1s, as in some slow movements. The rondo theme at they are double-rotational sonatas (two cycles the beginning is the opening gesture of an i1nithrough an extended thematic pattern, the first tial rotation laid out as a sonata exposition, usu-

of which constitutes the exposition), but the ally complete with energy-gaining transition, treatment of their second rotation differs from medial caesura, secondary theme, EEC, and that found in Type 1s. In a Type 2 format that so on. Thus the Type 4 sonata begins with an rotation begins as a more normatively develop- expositional rotation that traverses the usual P mental section in a nontonic key. Type 2s may TR’S/C pattern. Additionally, that rotation’s or may not call for internal repeats: both prac- last module normally dissolves into a retransitices are represented in the literature. Examples tion leading, without expositional repeat, to the of the Type 2 sonata include the first movements next tonic statement of the rondo theme, or P. of Mozart’s Symphony “No. 1” in E-flat, K. 16, The expositional rotation is usually balanced by “No. 5” in B-flat, K. 22, and Piano Sonata in E- a recapitulatory rotation with the same features. flat, K. 282; and the finale of Eine Kleine Nacht- What happens in the substantial space separat-

musik, K. 525. ing the expositional and the recapitulatory ro-

There are many variants of the Type 2 so- tations can differ from piece to piece (episode? nata, many different options for realization. In development?). The traditional, seven-partthe second rotation, for instance, one may find rondo letter-scheme, ABACABA, is inadequate

Sonata Types and the Type 1 Sonata 345

to describe sonata-rondo structures.4 While the 5 sonatas, thematic layouts within correspondletter-format suggests juxtaposed blocks (which ing rotations can become volatile, provisional does occur in some rondos), the sonata-rondo constructs. Understanding Type 5s adequately proper, the Type 4 sonata, is more strongly in presupposes a thorough grasp of the options and dialogue with the expositional-rhetorical norms alternative (less normative) procedures available that underpin all of the sonata formats. Such in Sonata Types 1-3.

distinctions are elaborated in chapter 18. The numbering for the five sonata-form Type 5 sonatas encompass concerto-sonata types may seem arbitrary, but there is a logic adaptations. These are blends between earlier behind it. Type 1s are the simplest, most probritornello (or tutti-solo) principles and other lem-free sonatas. Type 2s are also double-rotasonata types—most commonly the Type 3 so- tional sonatas—the next step 1n expansiveness nata. Type 5s are marked by an initial Anlage and complexity after Type 1s (since they contain (layout) normally given to the orchestra alone a developmental space). By a happy coincidence, (Ritornello 1 or Rotation 1). The opening or- Type 2s have often been called “binary” strucchestral ritornello is almost always stated en- tures (emphasizing “two-ness’’); similarly, Type tirely in the tonic key—or it at least begins and 3s, the most common of the types, are someends with strong, extended thematic statements times thought of as containing an emphatic terin that key.» A modulatory solo exposition fol- nary layout.® Type 4s are the typical remaining lows, one whose materials should be heard in sonata-alternative in nonconcerto instrumental relation to what had been sounded in the pre- compositions (especially for finales). And Type ceding Ritornello 1. Other Type 5 conventions 5s, capable of becoming the most complex of are found in the remainder of the movement: the structures, are special treatments reserved, later ritornello/tutti punctuations and cadential at least in purely instrumental practice, for conconfirmations, a solo cadenza near the end, and certos. so on. The definitions of the Type 5 exposition and recapitulation require special care and nu-

ance. These definitions are elaborated in chap- The Type 1 Sonata ters 19-22. In Mozart’s concertos the Type 5 format came Sometimes referred to as a sonata form without

to be a laboratory of formal surprise, complex- development, this pattern is the most elemenity, and personalized experimentation. It be- tary type of double-rotational sonata. The escame the most complicated of the sonata types, sence of the Type 1 sonata hes in the minimal particularly because of the modular multiplic- retransitional-link (or lack of a link) between ity and variety that he presented in them, along the two large-structural blocks: the expositional with his frequently employed thematic-modu- and recapitulatory rotations. In this type of solar substitutions, omissions, or reorderings in _nata the second rotation begins immediately or all of the rotations after the first one. In Type very shortly after the end of the first with the

4. Is Aalways to be confined only to P? Or can it some- (New York: Norton, 1986), p. 766. Wolf did not sugtimes encompass P + TR? Or is it B that always begins gest that these were the only three types of early sowith TR (in which case B extends from TR to the end nata design. Indeed, the contrary is true: “All these and of S/C)? If so, within B is no distinction to be made be- many other designs existed side by side at approximately tween TR and S materials? Or does B begin only with mid-century” (p. 766). Still, by singling out only these S? For all of its familiarity, the traditional letter-appara- three, the implication 1s that they were the clearest and tus 1s too coarse a filter to serve as an adequate descrip- most helpful ones to perceive, even though individual

tion of the sonata-rondo. variants abounded. (The quoted passages were unaltered 5. Exceptions are noted in ch. 20. in Wolf’s slight revisions to the entry “Sonata Form”

6. Our first three types also correspond to the three in the most recent edition of that reference book, The types of “large-scale configuration” singled out by Eu- Harvard Dictionary of Music, 4th ed., ed. Don Michael gene K. Wolf in his entry “Sonata Form” in The New Randel [Cambridge, Mass.: Harvard University Press, Harvard Dictionary of Music, ed. Don Michael Randel 2003], pp. 799-802.)

346 Elements of Sonata Theory

sounding of P! in the original tonic. This im- we allow for ellipses, expansions, and recompomediate rejoining of the tonic and a recapitula- sitions, the impression given by the second rotatory P! is a cardinal feature of the Type 1 so- tion is that of an immediately undertaken, comnata.’ Additionally, the strong first-level default plementary rotation that balances and resolves (almost invariable option) for Type 1 sonatas is the expositional layout. (Expanded Type 1s, a to dispense with repeats. Normally, neither the special subset, are treated separately below.)

expositional rotation nor the recapitulatory ro- Lacking both repeat signs and a developtation 1s repeated.’ Perhaps for this reason the ment, the Type 1 sonata, often a succession of Type 1 sonata is particularly suitable for over- entertaining melodies or contrasting topics, has tures and slow movements. (As a matter of ge- connotations of lightness, economy, simplicity neric principle, all overtures lack repeats,’ but of elaboration, and relative brevity. This is esnot all overtures are Type 1 sonatas: they can pecially true of fast-movement Type 1s. Slow

also be Type 2s or Type 3s.) movements, while still usually renouncing deMost typically, the second rotation is a close, velopment and complexity, typically favor a minimally adjusted replica of the first. The only broader lyricism or eighteenth-century sentirequired adjustment is the transposition needed ment. And yet at times the Type 1 slow moveto produce the recapitulation’s tonal resolution. ment could suggest deeper or darker things in It is not unusual, particularly in a slow move- its tone and character. This deepening is related ment from a smaller-scale example, to be able to the general expressive shift in some late-cighto map the second rotation (the recapitulation) teenth-century slow movements from a stylized bar-for-bar, onto the first (the exposition). Still, | galant sentimentality to more personalized inthis feature is to be understood flexibly. Not trospection, a move associated with later Mozart uncommonly, one finds alterations here and and, especially, with Beethoven. there, especially in the TR zone, which can show signs of recomposition, especially (but not Terminology only) if in the exposition it had led to a V:HC medial caesura. Moreover, as with recapitula- A. B. Marx may have been the first to use the tions in general, a second rotation will some- term “sonatina” for this format. In his system times be compressed, perhaps omitting repeti- from the 1840s it was viewed as a way-station tions of individual modules or even suppressing on the teleological path to “sonata form.” !° some modules altogether. Nevertheless, evenif |= Within music-theory discourse, Marx’s term 7. Note, however, that the anomalous onset of what movement) of the Serenade in D, K. 185, and in some amounts to the recapitulatory rotation in the subdomi- of the early string quartets. In such cases the presence of nant in the C-minor Andante espressivo (“Abswesen- repeats may suggest the underlying idea of a more deheit” or “L’Absence”) of Beethoven’s Piano Sonata in veloped, Type 3 sonata, even though the developmental E-flat, op. 81a, “Das Lebewohl” (“Les Adieux”’), m. space is attenuated. From slightly altered points of view, 21 (= m. 5 of the exposition!)—1in dialogue with the though, some of these might also be heard as Type 1s Type 1 sonata with P-based coda—is clearly deforma- with repeats. tional. In its seemingly aimless, circular loss, it doubt- 9. Only a few rare exceptions crop up in the major litless reflects the underlying program. See also the similar erature. See ch. 2, n. 7, which cites in this regard the observations regarding op. 81a/i1 in ch. 11, n. 21 (and “Intrada/Prologus” to Mozart’s earliest opera, Apollo et

related text), and ch. 12, n. 12. Hyacinthus, K. 38. There may also be some exceptions to 8. If the simplest Type 1s (no linkage between the ex- this overture-principle in a few of the earliest Neapolipositional and recapitulatory rotations) were provided tan opera overtures of the 1730s and 1740s, concerning with a double-set of repeats, we would confront four which Wolf (“Sonata Form,” The New Harvard Dictioconsecutive statements of the basic modular layout with nary of Music, p. 766) asserted that “no development sec-

little or nothing separating them. And yet at times one tion is present, and repeat signs are nearly always omitfinds slow movements with both halves repeated and ted” (our emphasis: the full quotation is provided later only minimal linkage between the two rotations, per- in this chapter). See also the more general discussion of haps eight or fewer bars of primarily retransitional ma- repeats in ch. 2. terial. This happens, for example, in a few of Mozart’s 10. Marx considered the sonatina to be a sonata form early works, such as the A-major Andante grazioso (fifth with what he presumed was a subtracted middle part.

Sonata Types and the Type 1 Sonata 347

seems instantly to have come to mean “sonata Rosen preferred to call the design “slowform without development.” It is with that movement form,” because it 1s often found meaning that the term has survived in both the there in the canonic works of the Austro-GerAustro-Germanic and the English-speaking manic tradition. Still, that format also surfaced music-theory world for the past century and in rapid first movements (for example, in the a half. The term “sonatina” [Sonatine] may be first movements of those “early” Neapolitan traced through the writings of Marx, Riemann, opera overtures of the 1730s and 1740s—Leo, Schoenberg, Leichtentritt, and so on—and, in Jommelli, and so on), and it was very much at the Anglophone tradition, in those of Prout, home, as mentioned above, in single-movement Goetschius, and others. Among many music overtures from the decades around 1800 and theorists today, this is still the term of choice, slightly beyond: Mozart, The Marriage of Figaro; one reflected in the American textbook on form Beethoven, Prometheus; almost all of Rossini’s by Douglass M. Green.!! The principal dissi- overtures (with only a few exceptions, such as dent in this regard was Heinrich Schenker, who that to La scala di seta, a Type 3 sonata with an in Der freie Satz called the structure “Four-Part S-based development); and many others. Hence

Form” (A,—B, : A,-B,).!? LaRue’s initial objection to Rosen: “‘ExposiSince the middle of the last century (in the tion-recap form, ... although admittedly mulwake of writings by Leonard G. Ratner, Jan tipedalian, at least has the virtue of descriptive LaRue, and others), musicologists have been accuracy. All the other terms that have been adamant in avoiding the term “sonatina” be- suggested are partly misleading: ‘slow-movecause of its historically faulty connotations. ‘The ment’ form occurs in many fast movements and characteristic LaRue argument was that “‘so- by no means all slow movements.”!> So far as it nata form without development’ and ‘abridged goes, this makes sense. But the term “exposisonata form’ suggest incompleteness ina form tion-recapitulation form” is not widely recogcomposers obviously felt could stand on its own; nized, particularly among music theorists, and it and ‘sonatina form’ sounds ridiculous when ap- is identified with a specific and outdated methplied to large symphonic movements.”!5 In the odology with which we are in little sympathy. revision of Sonata Forms Charles Rosen objected Our term, the “Type 1 sonata,’ is more conno-

to the term on similar grounds.!4 tationally neutral. This view may be consulted in Marx, Musical Form in the form, too, is guaranteed only by the fundamental line Age of Beethoven: Selected Writings on Theory and Method, and the bass arpeggiation. B, rests upon V, or is at least

ed. and trans. Scott Burnham (Cambridge: Cambridge moving toward it, whereas B, is based on the I.” University Press, 1997), pp. 82 (a “poorer and lighter” 13. LaRue, rev. of Charles Rosen, Sonata Forms, in Journal form but one that is “more unified’”’), 93 (also known as of the American Musicological Society 34 (1981), 563, n. 5.

the “small sonata form’), 94. More broadly, Marx un- 14. Rosen, Sonata Forms, rev. ed., p. 29n. LaRue’s own derstood the “sonatina” as a transitional form between explication of the term—cited as “Exposition-Recap the “fifth rondo form” and the “sonata form” proper. Form’”—may be found in Guidelines for Style Analysis, This mode of establishing the form corresponded to his 2nd ed. (Warren, Mich.: Harmonie Park Press, 1992), own conceptual derivation of the structure, not to the p. 188. Within this discussion La Rue locates the struchistorical appearance of the form, of which Marx knew ture in a questionable “evolutionary series” of forms (pp.

little or nothing. 187-90). These range from “primitive binary form” 11. Green, Form in Tonal Music: An Introduction to Analy- [sic] through “polythematic binary” and “large binary sis, 2nd ed., p. 230: “From the point of view of musical with full thematic differentiation” to “early sonata form

structure the sonatina form is a sonata form without a with incomplete recapitulation after a strongly differdevelopment section, its place being taken by a link or entiated exposition and fairly evolved development”’

transition leading to the recapitulation.” [roughly equivalent to what we are calling Type 2 here] 12. Schenker, Free Composition, 1:141. “It is most often and “full sonata form in its most evolved state” [similar

considered to be a sort of sonata form, in some way to our Type 3]. altered or mutilated. In actuality, the four-part form is 15. LaRue, review of Rosen, Journal of the American Mujust as independent as the two- or three-part forms. It sicological Society 34 (1981), 563, n. 5. The text continues is found especially in the slow movements of sonatas, directly with the quotation from LaRue just cited a few

chamber works, or symphonies. ... The unity of this lines above.

348 Elements of Sonata Theory

Historical Origin We do not take sides in this controversy, and

. in any event theof two positions hardly; seem muThe. historical sources the Type 1 design . P » . tually exclusive. What does seem to be the case

na : .

have also been disputed. one hand, we s that T 1 tas (“ ti cul . 1SOna the ec sONnatas exposition-recapitula-

have Eugene K. Wolf,. following the work YP ) and T (Tull, 3 P tas (full i 10n Orms) anofection” sONnatas triparHelmut Hell. Thus Wolf in The New Harvard wit ta f ; yP least in the; P 1 to, ; < 1te SONnata forms) appecar—al least 1n C1r Carll—-

Dictionary of Music (1986), under the entry “So- PP . . ea Form” est prototypes—to have emerged side-by-side in the 1720s or 1730s. So far as we currently . know, neither of them seems The Neapolitan overture mentioned above pre- ; ;to be historically . derivable as a variant of the other. Once Types sents a third large-scale configuration. ... In most 43 had 4 ved first movements of these [F—S—F] works from the 1 and 3 Lach Appeared as concreuze opuons—

late 1730s on, a full recapitulation enters in the by the mid-eighteenth century i was possible tonic either immediately after the close of the ex- to create a large-scale form poised tantalizingly position or after a brief retransitional passage. No between the two. Such a structure would be a development section is present, and repeat signs Type 1 sonata with a modestly expanded link are nearly always omitted (an option in the other between the rotations (“almost”’ BY developtypes, as well). Historically, this ‘exposition-re- ment), or, from the opposite point of view, a capitulation form (Jan LaRue) was derived from Type 3 sonata with small, rather insignificant tri-ritornello form [as tound, for ol be, “ m the development (so small that it seems virtually to contemporaneous NP eno concerto] y reaucrron be classifiable as a mere link).!? Although histor-

or elimination of the middle section (see Hell , ; , 1971) 16 ically Types 1 and 3 are two different things, in practice, we might find sonata forms that seem

Charles Rosen disagreed, perhaps observing suspended between the two possibilities. .

. . How much does it take to expand a service-

that this passage could serve as the basis for a ble link ; iI devel > I. ;

challenge to his contention in the first edition ans ch 1nCO her of eve han e - "hone of Sonata Forms (1980) that “slow-movement ter of the numbcr o MEASUTES INVOIVEG, OF COES

.) . ;was it depend on what happens in those _s measures? form” derived primarily from ritornello - In coni: ; This 1s not always an easy decision.?9 structures within the eighteenth-century opera 49 . ;Accordingly, : fronting this problem of classification, the presaria.!’ in the revised edition from b ¢ cht also hel 1988 Rosen added a new paragraph to expand EBEE OF ABSENCE O repeat signs mas ‘ aso ep oo.original 18 us,argument as mighteven the genre his further.of the composition. Or we might be content to let the conceptual suspen-

16. Wolf, “Sonata Form, p. 766 (minimally altered in on—also abounded at midcentury. Putting aside such the same entry in the updated The Harvard Dictionary of complications, however, one may observe that da capo Music, 4th ed., p. 801, where “the Neapolitan overture” arias [A—B—A] usually set two stanzas of text (1, 2). The

was changed to “the Italian overture’). The entry un- initial and final A section typically presented stanza 1 der “sonatina” (p. 767)—not by Wolf, and defined only twice: thus A (1, 1) B (2) A (1, 1). The double-presentaas a multimovement sonata, “but on a smaller scale and tion of the first stanza, each usually preceded by a ritoroften less technically demanding’—1is not relevant to nello figure, often took the format of a Type 1 sonata: this discussion. The “sonatina” entry makes no refer- the first stanza modulated to V and confirmed that key; ence to the music-theoretical use of this term to signify the second rotation of that pattern—and the text—rea single-movement structure. Wolf’s Helmut Hell ref- solved back to I. When the second stanza of the A secerence 1s to Die neapolitanische Opernsinfonie in der ersten tion began with “P” in the tonic, what was produced in Halfte des 18. Jahrhunderts (Tutzing: Schneider, 1971). the A section alone was very much like a Type 1 sonata. 17. Rosen, Sonata Forms, rev. ed., pp. 28-70, 106-12, (In other words, if the aria was a full da capo—with its 136. An adaptation of Rosen’s argument with slightly final A section a literal reprise of the first [as opposed

different terminology could proceed as follows. The to a shortening thereof]—each aria would present the first “two parts” of a standard midcentury five-part da Type 1 effect twice.) capo aria were often organized as what might be called 18. Rosen, Sonata Forms, rev. ed., p. 44. a sonata-without-development, although other possi- 19. Seen. 8. bilities and clever variants—including compressions of 20. Rosen, Sonata Forms, rev. ed., pp. 106-12. the final section, adaptations of Type 2 sonatas, and so

Sonata Types and the Type 1 Sonata 349

sion stand. When the decision one way or the The Expanded Type 1 Sonata other does not matter for more important issues

of hermeneutics, there 1s no need to make it. The pure Type 1 sonata contains no development or extensive elaboration in the second

The ‘Type 1 Sonata with P-Based rotation (the recapitulation). Whenever such

Discursive Coda an elaboration does occur, we may speak of an

expanded Type 1 sonata: an expositional rotaWhen a Type 1 sonata is provided with a sub- tion followed by an expanded restatement. Such stantial, P-based coda in the tonic, it can re- an expansion is typically to be found in the re-

semble what in Chapter 18 we call the “Type composed P-TR zones of the recapitulation, 1 sonata-rondo mixture.” Under such circum- Rotation 2.7! This expansion/recomposition stances it can be tempting to consider P to be a produces a billowing-out of one section of the rondo theme and the expanded coda as its last referential rotational layout. ‘This may be a relaappearance. The two structures are close and tively small matter, but it can also result in the can easily shade into one another, but the cur- impression of a more thoroughgoing, implanted rent category may be distinguished from that of | “development” section.

the Type 4 sonata-rondo (without development Relatively modest in size (but still compoor central episode) in two ways. First, the end of _ sitionally intense) expansions of P-ITR may be Rotation 2, the recapitulatory rotation, will not found in some of Mozart’s string quartets, as in have a significant retransition (RT) to set up the the slow movements of the Quartets in G, K. last return of P. (This RT is a central feature of | 387, and B-flat, K. 589. A slightly larger expanrondo forms and sonata-rondo mixtures.) Sec- sion occurs in the otherwise much-compressed,

ond, the last appearance of that P-idea may be strikingly non-normative second movement, presented in dissolution or only fragmentarily, Allegro molto vivace, of Beethoven’s Quartet in as opposed to a more literal appearance of an C-sharp Minor, op. 131; and expansions emerge earmarked rondo theme. The locus classicus 1s even more decisively in the finale of Schubert’s the A-flat major Adagio molto of Beethoven’s String Quintet in C, D. 956, and in many works Piano Sonata in C Minor, op. 10 no. 1. Here of Brahms, an early example of which 1s the first the exposition (mm. 1-44) is linked by a single movement of the Piano Quartet in G Minor, bar (V’, forte) to the recapitulation (mm. 46— op. 25. The expanded Type 1 format, which in91). Instead of providing a retransition back to terpolates a substantial P- or ‘TR-based developthe P idea (in the manner of a sonata-rondo), ment shortly into the recapitulation of a Type the final cadence of the recapitulation elides di- 1 sonata, has often been considered a typically rectly with it (m. 91)—a version with an altered Brahmsian procedure, and it has been much continuation. As a result, this movement is best discussed—using different terminology—in understood as a Type 1 sonata with an extended, the literature on that composer.?? Brahms’s op. P-based coda—one that, like many discursive 25/1 1s extraordinary in employing this format codas, is divided into sections and also features for a first movement; Beethoven’s op. 131/11

a final coda-to-the-coda (mm. 102-12). aside, rapid-tempo expanded ‘Type 1s are much 21. A much rarer option in the decades around 1800 deformation: a more or less normative rotational layout was to expand the “tonal resolution” section of the sec- followed a subsequent, decidedly non-normative probond rotation: its S and C areas. Still, by overriding the lematization and expansion of the initial pattern en route norm this could happen. The F-major slow movement to the ESC. of Mozart’s String Quartet in C, K. 465 (“Dissonance”), 22. Green discusses the structure as an “enlarged sonafor example, is a Type 1 sonata in which the second tina” in Form in Tonal Music, 2nd ed., pp. 231-32. See rotation (recapitulation) turns out at several important also the discussion, e.g., in John Daverio, “From “Conpoints to be an intensified, expanded restatement of ma- certante Rondo’ to ‘Lyric Sonata’: A Commentary on terial from the first. Not only is TR expanded, qua- Brahms’s Reception of Mozart,” Brahms Studies, vol. 1, si-developmentally, from thirteen to eighteen measures, ed. David Brodbeck (Lincoln and London: University the S theme is also enlarged from fourteen to twenty- of Nebraska Press, 1994), pp. 111-36. Daverio’s article seven measures. This is an extraordinary Type 1 sonata (p. 114, 116) provides lists of examples in Mozart and

350 Elements of Sonata Theory

more commonly found as finales. The finales of Because of the possibilities for variants and Brahms’s Symphony No. 1 in C Minor, op. 68, deformations, all of this can be confusing unand Symphony No. 3 in F, op. 90, may stand as less fundamentals are kept in mind. As a ruletouchstone examples, while other familiar in- of-thumb, if one 1s encountering a double-rotastances include those of his Piano Quintet in F tional sonata in which P! is returned to directly Munor, op. 34, and Quartet in C Minor, op. 51 after a nonrepeated exposition, one should ask no. 1. Brahms’s Tragic Overture, op. 81, is also whether P! is sounded in the original tonic. If laid out as an expanded Type 1 sonata form. so, the double-rotational sonata belongs to the The postexpositional thematic pattern of an Type 1 category, even though subsequent deforexpanded Type 1 resembles that of a typical sec- mations and expansions might occur. If not—if ond rotation of a Type 2. The only difference P! is sounded in anything other than I—then the between it and that of a standard Type 2 1s the structure is best considered within the confines tonic relaunching of P in Rotation 2. The second of Type 2. rotation of a Type 2, on the other hand, begins in the key in which the exposition ended ons The Type 3 Sonata with Expositional-Repeat

some other nontonic key. From perspec- . Loe: . . Feint: A Related butthis Differing Structure

tive the expanded Type 1 pattern mixes certain features of Type 1 (the tonic P-incipit-launch at = We should add one caveat to this discussion of

the outset of Rotation 2) and Type 2 (the de- the expanded Type 1 sonata. Because of the velopmental billowing-out shortly thereafter). immediate restatement of the P-theme in the In addition, as 1s the situation with Type 2s, ex- tonic once the exposition has ended, one might panded Type 1 formats rejoin the expositional conclude that it would be conceivable for a pattern (they begin to display correspondence composer to play momentarily on the idea of measures) at some later, post-P! crux-point, an expositional repeat. In the later nineteenth usually in TR, though occasionally earlier.?3 In century this possibility may be relevant, for exan expanded Type 1 design all of Rotation 2 ample, to certain types of Brahmsian, blended (unlike the case with the analogous portion of | sonata structures—ones in which, by midcenType 2 sonatas) participates in the psychology tury, the availability of an expositional repeat of recapitulation because the rotation begins was an option, not an obligation. In some inwith the P-theme in the tonic. It is a recapitula- stances it might be part of the sonata-game at tion, but one with a developmentally expanded hand initially to take the immediate return of

P-T'R zone.?+4 the P theme in the tonic for an expositional reBrahms of what he calls the “amplified binary form,” as 24. In this, too, the terminology differs from that well as an engaging discussion of the issue. We, how- which we use for Type 2 sonatas. For Type 2s we no ever, come to somewhat different conclusions and re- longer speak of a recapitulation at all. As discussed in ch. frame the issue in different ways. (See also our treatment 17, this 1s because the term, by definition and historical of these and related issues—and Daverio’s analyses—in tradition, carries the connotation of a “full recapitula-

ch. 18.) Cf. the discussion in Robert Pascall, “Some tion.” For this reason, in Type 2 formats—which do not Special Uses of Sonata Form by Brahms,” Soundings, ed. begin their second rotations with statements of P in the Arnold Whittall, No. 4 (1974), 58-63; Cf. also, e.g., tonic—we dispense with the term “recapitulation” and Walter Frisch, Brahms: The Four Symphonies (New York: identify only a “tonal resolution” beginning around the

Schirmer, 1996), p. 61; David Brodbeck, “Medium and S-point. The tonal resolution, the statement of the S/C Meaning: New Aspects of the Chamber Music,” in The complex in the tonic, occurs in all sonata types. (See ch. Cambridge Companion to Brahms, ed. Michael Musgrave 2.) Only in the Type 2 sonata, though, does it emerge

(Cambridge: Cambridge University Press, 1999), pp. in the absence of the normative tonic-key P launching 98-132 (especially p. 111 and the corresponding p. 297, a recapitulation proper. (Occasional exceptions to the

n. 23). tonic-P recaiptulatory launch within Type 3 sonatas are 23. One might suggest further, as with Type 2s, that considered in ch. 12.) it might be possible to substitute new material (an episode) into the space normally reserved for developmental elaboration in the expanded Type 1.

Sonata Types and the Type 1 Sonata 351

peat. This false implication may be part of the set of Rotation 2 might be the beginning of a piece’s logic at this point, even though the rest recapitulatory rotation does not emerge. These of the structure might prove to be a more or less are situations in which Type 1 or expanded

normative expanded Type 1. Type 1 formats are extremely rare: as such, they Here one needs to proceed with caution: do not rise to the level of a significant compoconfronting the issues at hand requires that one sitional option for first movements (despite the make careful distinctions. First, we should re- deformational case of Brahms’s op. 25/1). Nor, call that such a repeat is generically inappropri- under these circumstances, should one mistake ate in the Type 1 (“exposition-recapitulation” ) the “op. 59 no. 1” variant of a Type 3 sonata for design. Hence that repeat-feint reference could a Type 4 design (sonata-rondo): Type 4 sonatas be made only in situations in which a listener are historically and generically unavailable for might expect a sonata form with expositional first movements.

repeats. This expectation applies neither to In a nineteenth-century finale, however, the overtures (which lack expositional repeats) nor, issue is less clear. The Type 4 format was very later, to symphonic poems. At least in principle, much at home here, and the expanded Type 1 however, it might apply to either of the outer was also to become a viable option, especially

movements of a multimovement work. with Brahms. The local impression of an exIn late-eighteenth century and early-nine- position-repeat feint in a concluding moveteenth-century Type 3 movements, in which ment, for instance, might also be interpreted a repeated exposition was strongly normative, under some circumstances as the first return of this tactic of making a feint toward a generic a rondo theme. Type 4s, that is, can also disrepeat, then abandoning it, was rare. The clas- play a nonrepeated exposition followed immesic instance occurs in the first movement of diately (though usually after a clear retransition) Beethoven’s Quartet in F, op. 59 no. 1.2° The by a restatement of P or a portion thereof in the “op. 59 no. 1” variant of the Type 3 sonata con- tonic that soon dissolves into development. Still, sists of the following elements: a nonrepeated if the relevant P-theme does not have an obviexposition; an expositional-repeat feint with a ous “rondo character” 7° or if the movement is few bars of P in the tonic, soon merging into not labeled by the composer as a rondo—or if development; and, later on, a generally full re- other signals suggest that assigning the finale at capitulatory rotation beginning with another this point to the sonata-rondo category is not statement of P in the tonic. (Later examples in- the most informed choice*/—the encountering clude the first movements of Beethoven’s Sym- of what seems to be an exposition-repeat feint phony No. 9 in D Minor, op. 125, Brahms’s can be construed along different, non-Iype-4 Violin Sonata No. 1 in G, op. 78, and Brahms’s lines. At the point of the feint we should remain Symphony No. 41in E Minor, op. 98. The open- open to two other possibilities for interpretaing movement of Mahler’s Symphony No. 4 is tion. Deciding between the two will depend on also in dialogue with this paradigm.) When this whether or not P returns once again in the tonic occurs 1n a first movement, the issue of consid- to trigger a recapitulatory rotation proper. If it ering the possibility that the tonic-P at the out- does, the finale’s structure 1s to be understood as

25. ‘Two precedents from Mozart and Haydn, the first ment opening Rotation 2 dissolves away rather quickly, movements of the former’s Serenade in E-flat for Eight before proceeding through very much of its entirety. Winds, K. 375, and the latter’s Piano Sonata in D, Hob. In the case of Brahms, the contextualizing awareness

XVI:51, were also mentioned in ch. 2, n. 9. of other expanded Type 1 finales within his ceuvre 1s 26. As, for example, in the last movement of Beethoven’s obviously also a factor. Chapter 18 outlines the several Symphony No. 2 in D, op. 36, which, as a finale, is more different signals of Type 4 (sonata-rondo) behavior in

likely to impress us as a Type 4 sonata. finales, which, of course, also feature a tonic-P-led sec27. This impression is fortified: (1) if the finale is pre- ond rotation following a nonrepeated exposition. It also ceded by a weighty, slow introduction (uncharacteris- considers the hybrid that we call “the expanded Type 1

tic of Type 4 movements), or (2) if the tonic-P state- sonata-rondo mixture.”

352 Elements of Sonata Theory

the “op. 59 no. 1” variant of the Type 3 sonata. rotational pattern continues to be pursued at the This is what happens in the finale of Brahms’s crux-point after the development), the finale is Symphony No. 2 in D, op. 73 (whose design, best regarded as an expanded ‘Type 1. As menin other words, is also similar to that of the first tioned earlier, examples of such finales include movements of Brahms’s opp. 78 and 98, men- those of Brahms’s First and ‘Third Symphonies, tioned above). But if P does not return again in along with those of his F-Minor Piano Quartet, the tonic, and if the other criteria for the struc- op. 34, and String Quartet No. 1 in C Minor, ture are met (most importantly, if the ongoing op. 51 no. 1.

CHAPTER SEVENTEEN © "ISLE 0

The Type 2 Sonata

Be Type 1 and Type 2sonatas: sonata both formsprovide are ondary theme along with anyinclosing matedouble-rotational rials (C), will(S), then be resolved the tonic, just two cycles through the governing layout (P TR’ as they would be in the other sonata types. In S / C) with lttle or nothing separating them. all sonata types we refer to this tonic-groundThe differing characteristics of their second ro- ing presentation of the S + C block (part 2 of tations distinguish the Type 1 from the Type the rotation) as the tonal resolution. S, of course, 2 format. In a Type 1 sonata (“sonata without accomplishes the ESC. Figure 17.1 lays out the development”) the second rotation, following a most basic scheme, although many variants of it nonrepeated exposition, begins with an intact are found in the music of this period. P sounded in the tonic. In other words, it begins and continues as a normative recapitulation. In a Type 2 sonata, however (sometimes thought of | Inappropriateness of the Term “Recapitulation” as a “binary variant” of a Type 3), the exposition may or may not be repeated, andthe second While casually one might suggest that the secrotation begins as a developmental space; only — ond rotation of a Type 2 sonata form begins as

in its second half—often from S onward—does a development and turns into a recapitulation, it take on “recapitulatory” characteristics. Asis | sucha claim can be misleading. As will be elabcommon within developments in general, Type orated below, most sonata-form discussions in 2’s Rotation 2 normally begins with the first the eighteenth, nineteenth, and twentieth centheme (P) sounded either as an explicit thematic turies overlooked the Type 2 option. Within this reference or in an immediate developmental discourse the term “recapitulation” (German, elaboration but in a nontonic key, most often the Reprise) was devised to describe the normative key in which the exposition had ended. Shortly situation in the postexpositional spaces of what thereafter, this off-tonic P-reference broadens we call Types 1, 3, 4, and 5 sonatas, namely, that into more explicit, modulatory developmental space usually begun by the simultaneous arrival activity (often based on P and/or TR-material) of P and the tonic key and proceeding onward that usually drives toward a crux-point andthe _ to include S and C. Even while variants and deexplicit tracking through correspondence mea- formations of this practice do crop up, a crucial

sures at or shortly before S. Continuing (or component of the recapitulation-concept as it sometimes beginning) these measures, the sec- emerged historically in the music-theoretical 353

5 354 Elements of Sonata Theory

Internal repeats of each of the rotations are optional: some Type 2 sonata forms call for both rotations to be repeated, while others do not.

Rotation 1 Rotation 2 (Not-Sonata-Space) Exposition Development Tonal Resolution Coda (optional) P TR °S /C Poecccmep TRivcmnmp?S SE

I Vv V V? modulatory [Va] | I I I (Episodic (Ifisthis section substitutions provided, it

for P are possible) typicallywith begins Pll)

FicurE 17.1. The Basic Pattern of the Type 2 Sonata

literature was the initiating function of P, that “no.” What does begin with the arrival of the marker that launches what we called in chapters tonic-key S—and this is by no means to mini11 and 12 the recapitulatory rotation. One of — mize its structural importance—1s the tonal P’s central functions in all of the sonata-form resolution, the second portion of the second rotypes is to signal the onset of a structural rota- tation. There is no reason to consider this the tion (exposition, development, recapitulation, beginning of a “recapitulation,” as if the precedor coda). But the same cannot be said of S. On ing P-I'R material, the first half of the second the contrary, S’s role—above all in the expo- rotation, setting up the tonal-resolution funcsitional and recapitulatory rotations—is within tion of this S, were an insignificant matter. Here an ongoing rotation to proceed from the me- the primacy of the rotational principle—obvidial caesura to drive toward and secure the EEC ous enough for those who choose to observe or ESC. Whenever it is also participating in a it—trumps traditional, erroneous terminology. larger rotation, S never begins a large structural Type 2 sonatas do not have recapitulations at all, unit but continues one already in progress, one in the strict sense of the term. Instead, their secthat has been preparing for its arrival. This is a ond rotations have developmental spaces (P-TR fundamental characteristic of S qua S, one that or, sometimes, their episodic substitutes) grafted is established in the expositional Anlage. (The onto tonal resolutions (S—C). This topic will be relatively infrequent S-based development, con- revisited and expanded later in this chapter, in

sidered in chapter 10, 1s a special case. Such an our treatment of the fallacy of the “reversed S-initiation would not succeed a pre-established recapitulation” (a structure more accurately P and/or TR, or their substitutes, earlier in the understood as a Type 2 sonata with a P-based

rotation.) coda).

For this reason, it is inappropriate to claim Because the Type 2 sonata has been underthat the “recapitulation” in a Type 2 sonata investigated or misconstrued in the literature, it “begins with S.” Such an assertion, still com- bears reflecting on more deliberately here. ‘The monly encountered, is one of several unfortu- only secure way to approach Type 2 sonatas 1s not nate consequences arising from the eagerness in to begin with advanced examples of them—Mo-

the mid-twentieth century to define a sonata zart’s much-cited Piano Sonata in D, K. 311/1, only in tonal terms, pushing to the side impor- for instance—but rather to follow a chain of tant considerations of thematic function and historical examples from a few of Domenico arrangement. After all, the reasoning seems to Scarlatti’s binary-form sonatas through Mozart have gone, when the tonic is finally regained and beyond, noting the incremental changes, and stabilized, that must be the beginning of — additions, and deformations that can accrue to the “recapitulation,” must it not? The answer is the basic schema as one proceeds decade by de-

The Type 2 Sonata 355

cade, from composer to composer. Unless one from I to V (A! + B?) is followed by a fortykeeps in mind how these accretions and altera- one-bar, repeated second part beginning in V tions came to enter the Type 2 sonata in the first and moving back to I (A? + B!). The second place, it 1s easy to slip off the rails and revert to part begins (mm. 38-49) by replicating mm. such misjudgments as reversed recapitulations, 1-12 in V (the beginning of the A segment,

“mirror forms,’ and the like. down a fourth), then diverging slightly from exact correspondence measures—but retaining the model figuration—for modest alterations and a

Historical Considerations brief extension that regains the G-major tonic. After some twelve freer bars (mm. 50-61 vary

Reconstructing the genealogical lines of “clas- mm. 13-20), the strict correspondence measical” structures is a perennially controversial sures resume at m. 62, the crux, and persist to endeavor. Still, some of the historical anteced- the end of the piece (mm. 62—78 = mm. 21-37, ents of the full-fledged Type 2 sonata seem clear in effect the B segment now transposed to the enough. At least in part, it seems to have affinities tonic).2 The last correspondence bar (m. 78 = with the less fully elaborated, earlier-eigh- —__m. 37) extinguishes the musical process: there is teenth-century binary dance forms, particularly no coda. The double-rotational aspect of such a those in which the two structural parts conclude structure is self-evident, and we may refer to it similarly (though in different keys) with rhym- as a straightforward parallel binary form. ing material. The binary sonatas of Domenico A second, more complex (and more common) Scarlatti are also instructive along these lines. type of Scarlatti sonata involved the crystallizAlthough one hesitates to generalize over such a ing of more distinct thematic ideas throughout varied repertory, typical Scarlatti sonatas (as has and the suppression of a literal return of the A

been remarked more than once) include at least module at the beginning of the second part. two main types, disposed in a generous diversity Instead of presenting the A idea at that point of realizations.! The first, less common, is a sim- of “new beginning,” as in K. 2, Scarlatti frepler pattern in which the second part tracks the quently substituted a succession of new, usually melodic material presented in the first, while modulatory ideas. This new material eventually reversing its original tonic/nontonic motion —_ joined up, at a crux-point, with a pronounced (here represented by superscripts). In the lit- thematic—and harmonically resolving—parerature this has been described as the pattern allelism, sometimes of substantial length, with

Il: At + B2: []: A2+Bt:Il,in which Bis not the end of the first part. This pattern is usutypically a theme or even a distinct melodic idea _ally represented as something along the lines of:

but merely a “clause” or a continuation of A. I]: At + B2: [I]: C+ Bt: 11.3 Following the A touchstone example may be found in terminology of Douglass M. Green, this may Scarlatti’s Sonata in G, K. 2 (example 17.1, one be regarded as a balanced binary structure, a of the Essercizi published in 1738 but doubt- generically available alternative to or variant less composed several years before this). Here a of the parallel binary, or vice versa.* Particuthirty-seven-bar, repeated first part modulating larly if one keeps the simpler K. 2 pattern in

1. Here we follow, with simplifications, the discus- 2. For a discussion of our slight reinterpretation of sions found in Ralph Kirkpatrick, “The Anatomy of Kirkpatrick’s term “crux,” see ch. 11, n. 11. the Scarlatti Sonata,’ Domenico Scarlatti (Princeton, 3. Sutcliffe, The Keyboard Sonatas of Domenico Scarlatti, N.J.: Princeton University Press, 1953), pp. 251-79; pp. 320-75 (ch. 7, “Formal Dynamic’) takes pains to and Malcolm Boyd, Domenico Scarlatti—Master of Music emphasize the degree of inventiveness that was possible

(New York, Schirmer, 1986), pp. 166-78. See also the within this format. more recent discussion of Scarlatti’s binary forms in W. 4. Green, Form in Tonal Music: An Introduction to AnalyDean Sutcliffe, The Keyboard Sonatas of Domenico Scarlatti sis, 2nd ed. (New York: Holt, Rinehart, and Winston, and Eighteenth-Century Musical Style (Cambridge: Cam- 1979), pp. 78-79. The term “parallel binary,” so far as we

bridge University Press, 2003), pp. 320-75. are aware, has not been used before. The term was suggested by the similar formal category, “parallel period.”

I

| pg gy (ed ce Ay lel. — N ei. EXAMPLE 17.1 Scarlatti, Sonata in G, K. 2, mm. 1-78

trQO EEee ———— i& f ) “a es * a hy he. Mam “eg —=“—pss EEN TT TT iBOF a es SS I Ge” DQ ee “3 =e oe CE © ee ee ee ee a ed ns “T §=.Tn ANS © 2 A A Oh a CC A A O_O Presto

(Hee aEF ef ma OTT oT eT FT eo(QO ee AE SSFD QOaa (OS OS—ega accel

ee ,ai3 ee(QS | |aeSS ( || |a —_— |r : : a Qa Qe ec Qe Qe ( [| LJ te a >= _g aaA eS (S(O (DQ ( ( ( f=al=|. ia. ee _-_ ———n

OF A ee AO_O a eS _ a ree OO nee-OO

ee Ae a ee a eS ES RS ace (eee GO Oe irri ~~ i ¢~g~~ | zx ¢¢g¢~ i|{~_ fq F-‘' jryy iF | #i*” ge |* yf |

ee —“—s a A GD QO (ee—“e GO ee a | TT. oD eee ram an 2 neni =A a es nnn Me hi Pee ae Oe QO A CD QO QO (QO(

iPfTo a)eee TOT EE EE] Ani"es Laeae eeeae Oe+ eeeee eeeaeee Oe eee 22 EEE eee Oe "eee

ees Se ae ee eee =es#|ee||se Re ees = —— ape aaat ee es © PO GG iryf) 4g#*"., ee ee es Oe eS Oe ee A” || i» wtee |(GD iF DO |||# Jgi -i g@*+,., ) el h| fg".e7lee =e

| aSE , J. ee T.i{n

QS ( a J] sia

°° aoFoe Lo"—a|2 aeer eRa2TTgd Cea gd °0.N797}”)_

es es Se ee ae Oe ee (a ee ia ww Ve eee eee 2 eee eee eee eee Jee 2 eee eee eee eee

iV Urr.DLDLDLUCUDE“DCi—iayy’esi(i:”es .=e IED,a——— SS a De xa) “iSCer se

44

Cy ee Se OO @9MAUVWHVI—__U—— — —>*——_———mh

fa IN De

46 ——— p Ps J TT. . i 3 4 ee Te rr a —__—__ | § 2 eeee t¢-———"

372 Elements of Sonata Theory

of the generic structure of the entire movement. tional pattern, normally describable as P!-!, in It would also have to assume, mistakenly, that a the key in which the exposition had just ended: near-literal restatement of P in the dominant— in V for a major-key sonata, or in III (less often, such as is found in K. 22/1, mm. 39—47—could _ v) for a minor-key work. In some major-key normatively appear as the concluding element instances the opening nontonic restatement of of a series of otherwise contrasting C-modules P-material within Type 2s—as was also the case (perhaps as a C? or C4), something that seems in many Type 3 developments—is immediately virtually never to be the case. (As pointed out restated down a fifth, that is, on the tonic-level, in chapter 9, vigorous P-based C-ideas are more before proceeding to development proper. (The commonly sounded at the opening of C-space, once-common, though inaccurate, term for this not at its end. Moreover, as example 17.3 sug- much-discussed tonic possibility within develgests, what is often heard in these Type 2 cases opments, “premature reprise,” should of course is less a potential P-based C than a literal, some- be avoided here, just as it is in Type 3 developwhat extensive restatement of P in V. This, too, ments.) *#! would be non-normative within an exposition’s When it occurred, this bar-for-bar, nontonic C-space and would provide substantial rein- statement of the opening of P—assuming here forcement for the interpretation of this P as the the simpler case of a single P-reference, not a sebeginning of Rotation 2, not the end of Rota- quenced one—typically lasted only a few meation 1.) In short, this misreading, in part facili- sures before merging into a freer development, tated by the absence of the “corrective” indica- as happens in the Type 2 first movement of tions of repeat signs, would misinterpret sucha —_‘ J. C. Bach’s Symphony in B-flat, op. 3 no. 4, and return of P as is found in K. 22/1, m. 39, as still the finale of Mozart’s Symphony “No. 11” in D,

belonging to expositional space.*° Similarly, if © K. 84. A second way was to begin the second

the effect were to be replicated in the tonic at rotation not with a literal restatement but dithe movement’s end, it would be misconstrued rectly with a nontonic developmental variant of as existing in recapitulatory space. In those in- P-material. One example may be found in Mostances, as in the concluding bars of K. 22/1, zart’s Piano Sonata in E-flat, K. 282, a Type 2

mim. 89-98, what is actually occurring 1s the sonata in Adagio tempo, where in m. 16 the appearance of a P-based coda—a normative op- original version of P proper is suppressed, and

tion for codas within all sonata types. the second rotation begins with a troubled, clouded allusion to P. Still another option—at

. least within extended P-complexes whose con-

P-Based Openings to the Developmental Space trasting modules may be identified as Pll, P12, As discussed in chapter 10, in the eighteenth P!-3, and so on—was to begin the second rotacentury the developmental space usually began tion not with P!-! but with one of the later modwith a nontonic reference to P, although this ules. Found in Stamiutz and others, this strategy reference could be made in various ways. Type suppressed an initial module (or two), some2s are no different from Type 3s in this respect. times reserving the right to restore it later as the The most direct of P-references would be a re- onset of a coda-rotation.

statement of the opening measures of the rota- Our focus here, though, is on the more ge40. Here a similar matter regarding binary movements ment shows it to be in an asymmetrical binary form.... lacking an interior repeat sign was brought up by Wolf, [Writes Riepel:] “At P I have written Beginning of the The Symphonies of Johann Stamitz, p. 141. Wolf cited a Second Part; for there could have appeared at that point passage from Joseph Riepel’s Tonordnung (1755, pp. 69- a major repetition with a :||: sign. But I have noticed 70) in which Riepel underscores that the presence or that this repetition is used only very rarely these days. It lack of a repeat sign does not alter the basic structure [the repetition] may perhaps show a composer’s poverty of a binary movement. “Riepel prints a symphonic first of ideas.” Wolf noted additionally that it is Stamitz’s movement without repeats, yet at the letter P labels the middle and late symphonies that are most likely to lack return of the primary theme in the dominant ‘Anfang repeat signs, not the earlier ones. des zweyten Theils’ (p. 69). The remainder of the move- 41. Cf. the discussion in ch. 10, n. 14.

The Type 2 Sonata 373

nerically normative P-incipit-launch option: a lit- flat, K. 16 (this time, with expositional repeats).

eral P!! in V or another nontonic key. (Recall Sometimes the near-verbatim statement can that an initial recurrence of P in the tonic key proceed into a few measure of TR. This situawould trigger the expectation of a Type 1 so- tion sometimes arises in the works of J. C. Bach, nata.) The mere presence of this option by no whose relatively unelaborated sonata structures means marks a sonata as belonging to the Type often feature short P-themes, as happens in the 2 category, since the more frequent Type 3 so- first movement of the Sonata in B-flat, op. 5 nata-practice also includes this possibility as one no. 1. Here the closed P-theme proper occupies of its most common choices. Since by the mid- mm. 1—8 (interpretable as a two-loop presentaand late-cighteenth century (and certainly later tion), eliding its local, concluding PAC -effect as well) Type 3s were more commonly encoun- into a TR-continuation of the dissolving-contered than Type 2s, the most reasonable assump- tinuation variety, mm. 8-14. Rotation 2 begins tion for listeners to have made as this type of — with a full sounding of P in F major (rhetori-

postexpositional material proceeded would be cally, mm. 35-42 = m. 1-8), and the first five that they were 1n the midst of a Type 3 dis- measures of TR (mm. 42—46, with a slight varicourse. Within the implied generic rules of the ant in the last sixteenth note of m. 46) are transgame, if the postexpositional material contin- posed directly from mm. 8-12. At this point the ued more or less in rotational order (PTR), a music departs from the expositional pattern to conceptual conversion into a Type 2 sonata was steer into quasi-episodic elaboration, moving always possible at or around the S-point. On through the expected minor keys (G minor and the other hand, any deviations from rotational C minor), and so on. order within the initial sections of developmen- A second-level default for the opening of Rotal-space material decreased the clarity and rel- tation 2 was the presentation of P-material—eievance of the rotational principle, and with it, ther literal or developmental—shifting away at the possibility that the listener would continue once from the key in which the exposition had to consider the Type 2 option—grounded in the ended. Here the P-grounded music sets out in perception of rotational practice—to be still vi- a key other than V (for major-mode sonatas) or able. Put another way, nonrotational develop- III or v (for minor-mode ones). This option 1s ments seem, ipso facto, to be a sign of canceling found in the Type 2 first movement of Mozart’s

the Type 2 option, which 1s so reliant on the Symphony “No. 6” in F, K. 43, in which the concept of a double-rotational structure. En- development (m. 50) begins with TR textures countering one would only reinforce the con- (based on P) suddenly presented on an A-chord, viction that one is dealing with a Type 3 so- which immediately serves as V of D minor (V of

nata. vi). So long as the P- or ‘TR-derived material is In cases in which the exposition-pattern’s P- not sounded in the tonic, which would suggest a zone 1s brief—perhaps only a phrase or two— Type 1 structure, we may consider the result to the recycled theme at the opening of Rotation be the launching of a normative developmental 2 may be sounded in full and proceed directly space, characteristic of both Type 2 and Type 3

into a development of P- or TR-material. In sonatas. this situation the whole P-theme (and sometimes the incipit of the transition as well) can be Episodic Openings to the cited verbatim—or nearly so—before veering into freer thematic elaboration. This means also that the developmental space can take its initial =A third-level default within the ‘Type 2 format bearings from a complete or nearly complete re- was to override the normal P-related launch with statement of P before proceeding to more recog- more or less new, “episodic” material—ain effect, nizably developmental or modulatory textures to blank it out with differing music that we are proper. This 1s what happens in K. 22/1 (ex- invited to understand as standing in structurally ample 17.3b). Another illustration may be found for P. This procedure seems to have been less in the initial movement of the Symphony in E- common in the mid-eighteenth century, but it

. . . Developmental Space

374 Elements of Sonata Theory

did occur in several works of Mozart. One of — mains implicit, even though a literal restatement the clearest illustrations occurs in the A-major of P has been suppressed at the onset of Rotation Andante of the Quartet in D, K. 155. Example 2. (A more extended discussion of writing over 17.4a shows the first eight bars of the movement, within developments was presented in chapter a single sentence leading to a half cadence that 10.) K. 155/11 has one final complication that serves as aI: HC MC at the end of m. 8. (Mm. needs to be addressed here. As if by way of com-

1-8 constitute a PS>TR merger with the TR pensation, Mozart converted the end of Rotaportion, the sentence’s continuation, beginning tion 2, m. 44, into an I:IAC, not a I:PAC, and in m. 5.) Example 17.4b provides the beginning appended six bars of coda at the end—the first of the modestly sized developmental space and four of which restore the “missing” P—finally

ensuing bars (mm. 21-32). Here the norma- achieving a I:PAC and articulating a delayed tive P proper, the presentation-idea from mm. ESC in the final bar, m. 50. The last six bars thus 1—4, is suppressed for the first eight bars of the bring back the “lost” P, mm. 1-4, only after the postexpositional space, although the sequential last bar of the second rotation has been sounded. (and sentential) mini-“episode” that replaces it This 1s a common occurrence in mid- and latis clearly related to P—as if the “silent” four-bar er-Mozartian Type 2 practice. As the rotational P in generic expectation were pressing into the principle makes clear, these six measures are in eight-bar fabric of the episodic substitute, which coda-space (post-tonal-resolution space), not in is efficiently brought to V of the tonic A minor __ recapitulatory space. (Returning to P 1s a sign

in m. 28. M. 29 is a crux-point: mm. 29-32 of beginning a new rotation, not ending one.) recover and retrack the TR materials from the We find this especially persuasive when coupled exposition (mm. 29-32 = mm. 5-8, here at the with the awareness that some Type 2s in early same pitch level), thereby producing the same Mozart that suppress a literal P at the onset of I: HC MC—which 1s then followed by the tonal the developmental space do not restore it at the

resolution (the original S now in I). Strictly end, after the completion of the last module of considered, this brief developmental space oc- Rotation 2. This situation will be dealt with

cupies only mm. 21-28. further below.

What is crucial to recognize here is that a Another clear instance of a developmental more straightforward Type 2 second rotation space beginning with a writing-over of the exwould have presented its modular materials in pected, more literal allusion to P occurs in the the order, P TR’ S. In this case, beginning “de- first movement of the Violin Sonata in D, K. velopmentally” in m. 21 and extending past the 306. Here an ecight-measure episode, mm. 75— MC (m. 32) into the tonal resolution through 82 (whose bass may have been derived from that

the end of S in m. 44, what we have is the of P), is introduced after the expositional reexposition’s materials in the order, X TR’ S. peat sign and 1s followed immediately in m. 83 Thus the P-alluding, modulatory X (mm. 21—___— by a development of the rotationally “proper” 28) substitutes for the more standard statement TR material (based motivically on P).4? In such of P. Apart from this, all of the other elements a clear case (provided that a sufficient number of the rotation are in place and presented in or- of later elements of Rotation 2 remain intact der. Only the first, P, has been erased or writ- to render the resultant large-scale pattern recten over by something else. (The procedure and ognizable as a rotation—as happens in K. 306) its related interpretive problem recall those of | we may speak of a substitution or proxy for a the typical Scarlatti-sonata binary structure, generically expected individual element of the

I]: At+ B2: []: C+ B11, discussed above.) pattern. But if such proxies multiply as RotaIn these cases the background or conceptual tion 2 proceeds, it becomes more difficult to double-rotational aspect from m. 21 onward re- perceive the rotational nature of the whole. At 42. In this case the suppression of P at the outset of 158 as the onset of a stretch of coda-rhetoric interpolaRotation 2 is compensated for toward the end of the tion (CRI). movement, when the head motive of P returns in m.

a ———

EXAMPLE 17.4a Mozart, String Quartet in D, K. 155, 1, mm. 1-8

eer cre TES | BE ooo eee Andante >

edd agatdd ic 7 PF err ele

gp ST eeNe gg ) jf rer Os (Sa SSS SS ES EE EE EE ————————

a

Sed dt AS

— — slieseetcs; eee —_—— LOor p

ddedd a

OO 25 f A tr

ee

eS _ ( 376 Elements of Sonata Theory

EXxAmpLe 17.4b Mozart, String Quartet in D, K. 155, 11, mm. 21—32

21 tous | f_P ,, f~P fy +pP N —— \oor 4tot. = | ———— | Andante |

cr =(Cam td ate | i$ ue 0 5 $a ——

At Ste 9 HO gw eoF | * o e Cw lt e g wee WO we ge om

ZA.

ya 4 om 7 mm, e- “s .|T4

tgdddo es ==yoVed PE Pra ———

| a op. —$§| mf ——} ff ff __}__{ —_}__4___gff ff

29 pe) tS PRas llee— e 2o—~ ft TT eo P

OO Ee a ao fiaads) OO

ss a te —____ ff we rrrer ire - ie rrr iad

rete. ieee eeeerr) fly a Le ft

some point, that sense disappears altogether, and S or C would seem to be a declaration that rotait is more difficult to speak convincingly of the tional principles were being abandoned for the

presence of a rotation. development. And abandoned with them, by

ded in th inciples. S- or C-Based.Openings to the Developmental » Be. PNP .°. as > . And yet, once thus audibly “thrown away”

extension, would be the Type 2 option, which

Space: Type 3=>lype 2 Conversions .; an option, a non-normative Type 2 sonata deRelated to this third-level, “P-substitute” de- formation could apparently be recuperated by fault was the overriding of an expected element simple fiat. This appears to be the case in the (such as the P-related launch) with a thematic curiously asymmetrical—and exceptional—first or motivic element taken from the S or C zones movement of Mozart’s Symphony “No. 4” in (the markers of rotational endings, not begin- D, K. 19, whose exposition manages to occupy nings), which are then immediately treated de- substantially more than half of the movement. velopmentally. This option probably belongs Even stranger, neither P nor TR are heard again more to a theory of Type 2 structural deforma- after the nonrepeated exposition (mm. 1—46). tion than to a theory of standard practice. Such The development (syncopated upbeat to m. a procedure flouts the notion of rotation, which 47) instantly veers into modulatory regions on is conceptually linked with recurrences of a S-based material (m. 49, for instance, is related more or less fixed thematic ordering. To begin to S°, m. 21), and no significant P or TR modpostexpositional work by developing aspects of — ules are presented at all. Under these nonrota-

The Type 2 Sonata 377

tional circumstances, one would assume that the activity—preparing us for the Type 3 tempoType 2 option is nowhere in sight—that Mozart ral breadth that the rest of the piece is likely is obviously in the midst of a Type 3 sonata with to occupy—only to compress that remainder an S-based development, and that a return to a into a Type 2 format (as in K. 19/1) results in a tonic P and a recapitulatory rotation will soon dramatic collapse of anticipated sonata-space, a ensue. The surprise, however, is that this devel- substantial foreshortening of the expected latter opment soon leads to a crux-I:HC:MC-effect in portion of a work. Mozart revisited the “K. 19” m. 59 (= the MC in m. 20) followed not by P but conversion-effect (in which a predicted Type by the tonal resolution only, a slightly recom- 3 shifts into a Type 2) in the much-discussed posed S- and C-space in the tonic. Moreover, Piano Sonata in D, K. 311, whose developthe movement provides no compensatory coda ment begins with continued reiterations of the to restore the “lost” P. The movement ends in a module that had ended the exposition, C?. An manner parallel to the end of exposition—with equally “extreme” example from Haydn may be no added coda at all. As will be revisited later, K. found in the second movement, Presto, of the 19/1, along with the first movement of the Sym- Symphony No. 21 in A. Here the development phony “No. 6” in F, K. 43, demonstrates that at is dominated by a treatment of the TR-based S. this time there was still no generic requirement The subsequent tonal resolution, led into with to restore in an appended coda a P that 1s “miss- the related TR-module-crux at m. 66 (= m. 10)

ing” in postexpositional space. is thoroughly overhauled to “omit” this version One way to grasp the unusual structure of K. of that S-module.

19/1 1s to suppose that, for reasons of high play, Type 3=Type 2 conversions are not reyoung Mozart presented us impishly with ge- stricted only to pieces that feature nonrotaneric signals that would be workable only within tional developmental spaces, such as those just Type 3 sonatas (the presentation of a nonrota- discussed. Related cases occur where developtional development) and then staged the music as ments that start out as P-based “go past” the impulsively “changing its mind” at the MC-ef- P- and TR-zones to refer, however briefly, to

fect and tonic-S-launch point (mm. 59-60). a module of S or C. This would render the deWhat we had assumed en route was a Type 3 so- velopment, or at least a portion of it, fully ronata is suddenly yanked into Type 2 status. More tational. This situation also seems to rule out

precisely, this sonata form endsin the manner the expectation of a normative Type 2 conof a Type 2, with a tonal resolution only (S and tinuation, since that option usually reserves the C), not a recapitulation beginning with P, even second-rotation S and C for the tonal resolu-

though the literal Type 2 status of the result is tion and its correspondence measures. But if not fully realized, since the development proper what follows the developmental rotation rejoins cannot reasonably be construed as contributing the expositional pattern at a post-P!-! idea that to the first half of an ongoing second rotation. (It is earlier in the rotation than the last one alwould be difficult to make a case that the nor- ready heard in the development—probably at a mative P in the development was being written TR- or S-module, thus providing only a tonal

over in toto by S-based modules.) resolution, not a recapitulation—we would seem This example demonstrates the possibility of | once again to have a Type 3=> Type 2 convera conceptual interplay between two different sion. This arrangement of events may be found, sonata types—a wrenching of expectations as together with a few additional complications, the listener shifts from those of the one model in the finale of Haydn’s Symphony No. 21 in A to those of the other. For a composer to invite and in the earlier, D-major slow movement of us to predict the proportions of a Type 3 so- Symphony No. 14 in A.*# nata by means of nonrotational developmental The reverse effect could also occur, in which 43. Asis clear from the discussion, we currently regard 2 one (by definition), leads only to a tonal resolution, all of the above situations as Type 3=Type 2 hybrids, not to a full recapitulation. These are instances where in which a seemingly Type 3 development, not a Type the Type 2 tonal resolution is present (a “full recapitula-

378 Elements of Sonata Theory

Type 2 signals, following a seemingly obvious sessed according to the proportions predicted in crux point within Rotation 2 (a Type 2 conven- the exposition. tion discussed below), disintegrate in favor of a The strong first-level default is for a Type 2 P-based recapitulation. This produces a sense development to be based on elements of P or of sudden spatial extension: a predicted Type 2 TR—that is, on elements from only part 1 of

is made to change its course to convert into a the exposition—in order to preserve the diaerander, broader Type 3 sonata. Such an effect chronic unfolding of the rotational pattern. plays a central role in the first movement of the The development may also mix materials from Symphony No. 38 in D, K. 504, “Prague,” in P and TR in an unpredictable order, as though which a mid-developmental joining of TR mod- one were free to reassemble the part-1 modules ules in the tonic (a crux, mm. 177—89-mm. 59— within a generalized developmental space that 71) initially suggest that the piece will proceed still may be understood as preserving the larger as a Type 2 sonata form, only to see it fall apart rotational ordering. This emphasis on P or TR at the “first MC” point (m. 189) 1n order to turn modules is usually what one finds in Type 2 into a Type 3 sonata with recapitulation begin- sonatas, although relatively new episodes or ning with P (m. 208). (Even more rarefied are the thematic digressions may also be encountered. extremely uncommon instances 1n which one or Conversely, within this developmental portion more of an exposition’s S-materials appear briefly of Rotation 2—whether literally developmental in the tonic—something of a “false tonal resolu- or quasi-episodic—one would normally expect tion’—1in what then turns out to be the center of | no elements of S or C (part 2) to intrude. When the developmental space of a Type 3 sonata.)*4 they do, we would be dealing with an exceptional procedure (an expressive deformation)

Developmental Activity Proper that would call for a special explanation. Such

an exception occurs in the Type 2 first moveFollowing the P-material-launch (or that of its ment of J. C. Bach’s Sonata in G, op. 5 no. 3, proxy), we expect to find a freeing of the the- into whose development an element of C-texmatic rhetoric into more recognizable devel- ture—identifiable as C? from m. 31—1is introopmental processes. This occurs at the point at duced and expanded, mm. 57-62. which the second rotation’s P- or P-T R-mate- Another unusual deformation, already disrial is no longer mappable as correspondence cussed in a different context in chapter 10, measures onto that of the first’s. All of this is may be found in Mozart’s three-movement similar to the common developmental practice Symphony No. 32 in G, K. 318, the entirety discussed in chapter 10, which need not be re- of which articulates a Type 2 sonata. The first peated here. The length of the development in a movement proceeds into a largely P-based deType 2 sonata 1s not predictable. In some cases it velopment, which 1s interrupted by an unrelated is quite short (asin K. 155/11, example 17.4b—if = slow movement. The finale begins with a rethat passage may be regarded as developmental transitional link and preparation for the tonal

at all); in others, particularly later in the cen- resolution (not for a full recapitulation) —the tury, itis remarkably extended. The composer’s sounding of the original S and C in the tonic. choice in this matter bears on the larger ques- As a whole, K. 318 merges the episodic variant tion of how much time the sonata 1s striving of the Type 2 sonata with what has been called to span. Lengthier developments need to be as- the da capo symphony or reprise overture.

tion” 1s lacking) but the span from the onset of the de- cussion at the end of this chapter, “Confronting Hard velopment through the end of the tonal resolution does Cases: Flexibility in Sonata-Type Recognition.” not constitute a single, clear rotation. It may also be, 44. For a discussion of this atypical situation see Hehowever, that these Type 3=> Type 2 hybrids constitute pokoski, “Beyond the Sonata Principle,” pp. 130-39, what others might regard as a separate, “early” sonata which also locates an exceptional example in the first type found in some midcentury works. See also the dis- movement of Haydn’s Quartet in B-flat, op. 64 no. 3.

The Type 2 Sonata 379

The Crux Since we have dealt with issues surrounding . . the crux in chapter 11—differing transposition After a span of development, the music will eeand P so . . levels, the possibility of a double crux, (sometimes almostmatters imperceptibly) lock onto . a on—these will not be revisited some. .middle portion of the expositional pattern . here. In ae basis . , most cases, the on a bar-by-bar either at thelocating original pitchcrux . . in a Type 2 sonata is unproblematic. It usually occurs near level or transposed to an appropriate key to lead . the center of Rotation 2, at. or shortly before the to or produce the tonal resolution. These cor. medial caesura that opens the path to S and the

respondence measures are thenresolution. pursued, In forthis thecase, . . the P and/or TR , tonal

most part, for the rest of the rotation. As out- .

. . . modules are11,given over entirely, or nearly so, lined in chapter and adapting the terminolyo .. . . to development. Still, it is impossible to predict ogy .devised by Ralph Kirkpatrick for Scarlatti .. where the crux will situated. In unusual cirsonatas, we. a.call moment theforbe crux.*> Within , _. this _. cumstances and special expressive purposes,

a Type 2 sonata, by definition, the expositional . . . -

may occur relatively early in the expositional pattern was. itnever rejoined here atlocus its opening . of , , , pattern, with, say, P!.?. The classicus point, P!-!.46 If it were, it would be an example , , fa aType : th a full tulat athis occurs in the finale .ofa Mozart’s Quartet of e 3P3sonata—with full recapitulation. e > at m. yP in G, K. 387, with “early” P!-? crux 175 In a Type 2i,sonata the crux must occur at some . . , (= m. 17, now in the subdominant), although

. option.P _. ordering.

later, point the thematic a . selected . i. .post-P!! this seems toinhave been an pattern, infrequently point that also serves to maintain the rotational Kon 47

In sonatas whose expositions had featured a The moment that aI:HC development based on 4. in. . og. MC, the crux is likely to be sounded P- or TR-modules or on antonic, appropriate episodic . . .the extent that original not in IV. To proxy _. (athe writing-over) establishes a crux-event . . a prolonged P!2 or TR module is sounded. in

on .a. subsequent module ofanthe rotational pat, ,1s, 3folthe tonic (following “early” crux) and tern 1s the moment when the ‘Type 2 intentions . lowed by correspondence measures only slightly of the sonata form are declared. This can hap- , . adjusted to lead to the proper S-resolution, that pen, for example, .atwhole P!?, tonic-based at TR!!, atsection TR42,approaches at , the MC, or even at the. S-point. As(which we process oea Type , 4, , 3, recapitulation” would normally begin

. . with P!-!).48 In other words, the further back

the moment-to-moment events of the piece, we .

notice that eg the ,crux-event occurs of any 3 ; the crux 1sinstead pushed, the’ more

that the Type-3-identifying “recapitulatory” tonic arrival . . Type 2 sonata begins to take on the attributes of a Type 3. of P!! and thereby declares that we are no longer , , , , 4g a, _ In such situations—“recapitulations” beginning

to.expect that arrival at all. As or a result, this is the . . of the P-pat. i. with the second third module point at which the composer us to un-encountered . . . tern (aninvites occasionally occurrence derstand that the Type 3 option (signaled by the . ae ae P!.! . in mid-eighteenth centurydiscarded sonata forms)—one appearance of that return) is being ,a3

. might be tempted to speak of a “mixed” so-

in favor the more compact, double-rotational ttYP : “ti +i_. . . .ofnata ora existing onattaina continuum alternative of thee, Type 2e sonata. With the YP somewhere between Types 2 and 3. Additionment. .ofally, a Type 2 crux, wethese are to abandon the might ... to what extent considerations expectation of a P!:!-launched recapitulation. Lt. . . . implicate recapitulations that omit P proper to 45. Cf. n. 2. 48. Again: the standard alternative for an early tonic-

46. See, however, the complications and ambiguities crux would be to sound P!-? in the subdominant. Here raised in ch. 12 dealing with Type 3 variants: recapitu- one needs to recall that within Type 3 sonatas beginlatory rotations that begin in nontonic keys, especially ning a recapitulation in IV was always an option, albeit

in those beginning in IV. not a frequently chosen one. See the relevant discussion 47. Perhaps related to it, though, are the cases described in ch. 12, which again brings up the K. 387/iv issue. in Rosen, Sonata Forms, rev. ed., p. 157. Cf. also, e.g., Rosen’s discussion of C. P. E. Bach’s Sonata in F-sharp minor [1763] on pp. 285-86.

380. Elements of Sonata Theory

begin with (P-based) TR material—a feature of — 1n vi, leading a vi:PAC in m. 41, tonally correcapitulatory compression—is a vexed question rected in the next several measures and leading that admits of no easy solution, and we shall not to a second crux, m. 45 = m. 18 [S!-]). address it here.

The most common Type 2 crux-strategy, The Tonal Resolution (S+C) however, was less problematic: to rejoin the pattern at a point midway through the transi- Normatively, at this point the second rotation’s tion, perhaps around TR! or TR!-. It was also tonic-oriented S- and C-material would now possible to delay the crux until or immediately retrack, bar-for-bar, the analogous music of Roafter the medial caesura. In this case the devel- tation 1, and this is very often the case. As 1s also opment presents a recomposition of P-T'R and true within the other sonata types, this principle proceeds to the medial caesura in a manner that of mere transposition and correspondence meais not a direct restatement of the correspond- sures 1s not inflexible. Slight variants, occasional ing zone of the expositional pattern. When this reorderings of modules, and expansions or comhappens, the crux 1s usually attained with the pressions were always possible. However these tonic onset of S, the launch of the tonal resolu- S and C themes might be varied, they still give

tion. This occurs, for instance, in J. C. Bach’s the effect of being correspondence or referenSonata in B-flat, op. 5 no. 1, first movement, tial measures. To this section of tonal resolution

m. 63 (= m. 15). is given the task of articulating essential strucIn extreme cases—apparently rare—a com- tural closure (the ESC) and, usually, the task poser could begin to engage S-material (or its of following this with rhetorical, postcadential equivalent in a continuous exposition) outside confirmation. of the tonic, without a sense of tonic-key crux, As elaborated at the beginning of this chapter, regaining the crux-effect of correspondence ina Type 2 sonata the tonal resolution normally measures only later in the composition, perhaps does indeed “recapitulate” (restate and resolve around the area of the ESC. Here an easily con- tonally) the S+C portion of the exposition, but sulted example is the exceptional Type 2 finale it is a conceptual error to think of it as the start of Mozart’s Symphony “No. 6” in F, K. 43. Fol- of any sort of “recapitulation.” Rather, it is the lowing the repeated, unproblematic exposition continuation of a rotationally ordered series of (mm. 1—47), the remainder of the finale unfolds events—P TR’ S / C—that had begun in the as an obvious second rotation. The P-based de- developmental space. The S-event is an action velopment, however, leads to the “wrong” MC, that occurs midway through the governing roon V of D minor (V/vi) at m. 67, whose triplets tational pattern at hand. From that perspective recall that of the exposition’s I: HC MC at m. there is no sense of large-sectional structural re16. This 1s a crux on the wrong tonal level, and beginning at the onset of the tonal resolution. the sentential S sounds its presentation modules, The real structural rebeginning or recycling had S!1, for eight measures on that nontonic D mi- already occurred with Rotation 2’s P-material nor, mm. 68-75 (= mm. 17-24). As if register- launch (or that of its proxy). To posit the exing the “comically tragic” impossibility of the istence of a rebeginning or “recapitulation” at situation, Mozart now interpolates twelve bars this S-point is conceptually to erase the funwrenching the music into the proper F major damental rhetorical principle guiding the dou(mm. 76-87), and the S-continuation at m. 88 ble-rotational structure as a whole—thus miss(S!-2), finally a tonally proper crux (= m. 25), ing the expressive and architectural point of the proceeds with the appropriate correspondence Type 2 sonata. measures 1n the tonic, to the ESC (m. 95 = m. 52) and beyond, MEO closing material. As omen The “Recapitulation” Question and Tonal what similar instance in of Continuous a “wrong-key” crux .... . ; Resolutions Second Rotations may be found in the G-major slow movement,

Adagio, of Haydn’s Symphony No. 24 in D One reason that it has been tempting to mis(brief first crux, m. 36 = m. 14 [S!!], only now construe the onset of the tonic S as the begin-

The Type 2 Sonata 381

ning of a “recapitulation” in Type 2 or “binary” Type 2 Sonata Forms without Codas

sonata forms isOnce that in any two-part exposition ; . the . ay 2 P Rotation 2 has completed its cycle, or recapitulation (that is, one with a medial cae- . . piece may be over. such. cases—the most elsura) S articulates a structural relaunchIn halfway . . ementary instances—the sonata has. 4. no. coda. In through the Something does indeed . ,begin” oy. aa.with thisrotation. familiar option the end of the piece rhymes S: athe highlighted second part of . a. prolonged set . _ . . with endisof Rotation 1 after the rotation. But this not the impression given . . by no . . _ of correspondence measures, a situation

by sonata forms with continuous . . 7. . . means infrequent in expositions— the mid-eighteenth cen-

those lacking a medial caesura and an S-theme. . .

. .with tury.continuous The coda-free ‘Type 2 1s common, for 1nAs result, Typein2s exposi, .tions .astance, the Allegro movements of J. C. Bach, demonstrate more clearly whyof posit, ogyeven occurring in two the ,three Type 2 first move-

ing .a. supposed in the center neither of . . . of , ments of“recapitulation” the six op.shaky. 3 symphonies, Rotation 2 is conceptually If the sonata’s ... which, as it happens, calls for block-repetitions second rotation—development plus reso.. . . . . of thealso rotations—Nos. 5it in Ftonal and 6 in G—and lution—is continuous, will provide no . . . in all of the three Type 2 first mid-rotation break and relaunch on movements which to . .in the

«term . ae six op. 5 Keyboard Sonatas—Nos. 1 in hang the “recapitulation.” Instead, one . .B-flat, a . . 3 in G, and 5 in E. One also finds this simpler will experience only a P-based development, a . . procedure someof ofcrux the early Mozart in symphomerger into TR, ainpoint somewhere . «1”5) , nies, such as in the first movements of “No. the central expansion section, a continuation of “No. , 7in , E-flat, K. 16 (with each rotation repeated), all of this with the normative TR=>FS conver- a oa.

. 4”(chapter in D, K. (with no repeats indicated),°” sion 4), 19 and, eventually, a drive to the oc i , , re, and “No. 6” in F, K. 43 (with each rotation ESC and whatever C-space might follow. There . . . oo. j . peated). The basic structure without coda also is no point in the central Fortspinnung that mightkind oo. .of. Type . seems to be the historically earliest reasonably be isolated as the take-off point of . . . qd“ ulation.” 2 sonata form, with roots in such pieces as the yA SUPP P keyboard sonatas ofmay Domenico Scarlatti and the perfect,example of suchsonatas a movement , 4 E. Bach. (See examPrussian” of C. P. be consulted in the Type 2 finale of Haydn’s . .

any supposed “recapitulation. . . , 3in ,Eple 17.2 and the surrounding Symphony No. .44 Minor, “Trauer,” which ..

discussion.)

. i . It isthe this coda-free format for Type the also contains characteristic complication . . 2,that . simplest version of the generic norm, of a. .substantially recomposed latter portionvarietof . . should be kept in mind as one confronts

Rotation 2, along with aadditions P-based coda.*? Butif - . . In stan. . ies with and interpolations. Type sonata forms with continuous recapitu7. the , , , 52such dard practice once lations asmid-eighteenth-century that in the “Trauer” finale can, , , , end of the second rotation had been reached—

notogy be processed as having a point “recapitu, the conclusion ofof the Ssecond + .C area, the moment lation” in the center of their rotations, -_ . parallel to the ending of the first (expositional) why should we be eager towas grant existence . . -. to . _. rotation—there no the generic obligation of such a point in two-part expositions? Rather .This _. . . . add anything further to the composition. than trying to devise dodges aroundP this .. . . is the case regardless of whether or a subP-variant

stantial problem, it makes more sense to realize .

ce , ogy ee had launchedisRotation 2 or whether the normathat the, term “recapitulation” inappropriate . . . tive P had been suppressed 1n favor of a modular also at S-points in two-part second rotations.

49. Another instructive instance occurs in the first 50. In K. 19 the single-chord end of Rotation 1, m. movement of Vanhal’s Symphony in F [F3, c. 1768-71). 46—immediately undermined by a syncopated “epiIn this case, however, the continuous exposition is not sode” launching the second rotation—is slightly exof the expansion-section subtype but rather of the “sec- panded to a more typical triple hammerstroke at the ond” or alternative type, that with an early V:PAC fol- end of Rotation 2, m. 78 lowed by repetitions of the cadence. Nonetheless, such a structure poses the same questions to the analyst as does Haydn’s “Trauer”’ finale.

382 Elements of Sonata Theory

proxy or other non-P idea (as in the coda-less efficient Allegro in 2/4. At the movement’s end Type 2 first movements of Mozart’s K. 19 and K. we find an appended four bars after the last cor43). From the standpoint of midcentury generic respondence measure of the tonal resolution. practice, at least, there was no widely shared, — Those four bars recapture the initial gesture of compelling urge to shore up any “missing” Pin the finale: a perfunctory march up and down the second rotation with a P-based coda. The 1s- the tonic arpeggio in octaves. (In the exposisue of whether or not a coda is appended to the tion, mm. 1—4, this module had functioned as second rotation (following the tonal resolution) the first half of an eight-bar antecedent.) Here is also a conceptually separate matter both from we find perhaps the most modest instance of a the issue of internal repetitions and from the ___P-related coda rather than a mere cadential exmanner in which the exposition yields to the _ pansion. Redeploying the P-incipit at this point onset of Rotation 2, the developmental space. was no surprise. The principle of rotation suggests that what follows the close of the C-zone Type 2 Sonata Forms with P-Based Codas and should be the reopening of the P-zone and the

the “Reversed Recapitulation” Fallacy implied onset of a new cycle. This produces

the common situation of a coda that begins as From time to time—and increasingly after the —_an incipient rotation (in this case, an incipient 1760s, as a first-level detault—major compos- third rotation). In the K. 84/iii coda, though, ers chose to append elaborations or coda-exten- _ the postsonata rotation is cut short after only sions of differing sorts and lengths onto the end _ four bars.

of Rotation 2. The coda typically extends be- Codas that begin rotationally (with P-mateyond the final bar of Rotation 2 (andthe sonata _rial) are common occurrences in all of the sonata proper) into postsonata-space. Another option types as a first-level default for coda treatment. was to provide not a literal coda but a coda rhet- (See chapter 13.) This is worth underscoring. As

oric interpolation (CRI), wedged in before the Charles Rosen put it, “Ending a symphony or final module of Rotation 2 proper. (The CRI sonata with the first theme forte was too comalternative in general is discussed in chapter 13. mon a practice for me to cite examples: if the It also occupies a space that is not part of the reader cannot remember any, he can amuse himsonata form proper, although by definition it is self by looking—he will find them with ease.’’>!

inserted before the full completion of the tonal It is not reasonable to claim that when such a resolution.) None of these parageneric additions tonic P-restoration occurs in a Type 3 sonata it jeopardized the essential Type 2 structure, but _is self-evidently a coda, while when it is found each provided a more emphatic or rounded end- in a Type 2 sonata it is to be considered part ing. Sometimes this took the shape merely ofa of a presumed “reversed recapitulation.” Recall more elaborated final cadence that was scarcely again that the “reversed-recapitulation” misuncoda-like at all, as in the transformation of the derstanding relies on the unfounded premise final correspondence measure into two bars that a supposed recapitulation had in fact begun found in the first movement of J. C. Bach’s at the tonic-S-point. This claim rides roughSymphony in B-flat, op. 3 no. 4, a movement shod over the actually governing double rotathat is otherwise a straightforward, double-ro- tion. References to P in the tonic at the ends of

tational ‘Type 2 sonata. Type 2 sonatas are more accurately understood An unassuming final-cadence enhance- as codas existing in an extra space beyond the ment could be expanded by degrees into some- sonata form proper. thing more substantial. A significant step in this As it happens, one may appeal to evidence process may be observed in the finale of Mo- beyond generic reconstruction to fortify this zart’s Symphony “No. 11” in D, K. 84, a brisk, claim. In the first movement of Mozart’s Piano

51. Rosen, Sonata Forms, rev. ed., p. 97. Rosen’s accu- the classic “reversed recapitulation” fallacy for Type 2 rate remark is followed, however, by a reinscribing of sonatas. See n. 59 below.

The Type 2 Sonata 383

Sonata in E-flat, K. 282, an Adagio, each of the been made about this movement’s “reversed re-

two rotations is provided with a repeat sign. capitulation.” Once past the repetition of the second rotation Needless to say, even a fleeting return to or (development and tonal resolution, ending in paraphrase of a version of P that had been unm. 33) the sonata proper is over, but Mozart sounded since the repeat of m. 1 can have a poappended three extra bars (mm. 34—36) whose etic effect—the recovery of the original launchinitial five beats provide an only slightly varied ing-impulse for the sonata process, nothing less recapturing of the opening of P in the tonic—a than the Hauptgedanke. Nevertheless, that is no return to an eloquent version of P that had not reason to include the add-on coda-statement of been heard 1n that manner since m. 1. Above m. P within sonata-space. (This line of reasoning, 34 Mozart wrote the word “Coda,” thus clari- it appears, is also defensible within much later, fying how he was thinking of such P-returns.°” mid- or late-nineteenth-century Type 2 sona(Supporting evidence along these lines may also tas.)>4+ In rejecting the concept of the reversed be drawn from the arrangement of text-blocks recapitulation—along with any claims of mirin such vocal works as the Type 2 Quartet from — ror form asa primary formal construct—we also the third act of Idomeneo, K. 366.)°° In the case align ourselves with Wolf’s general position in of K. 282/1 it is likely that because of the repeat- 1981 on this matter with regard to the Allesign interventions in the text, few analysts today sro movements of Stamitz’s symphonies, whose would suggest that this return of the P-incipit multimodular designs and second-rotation re-

in the tonic is the tail-end of a “reversed re- orderings often render the simplistic notion of capitulation.” Still, as we have seen, the repeat a mere thematic reversal indefensible: “There structure of a movement has no necessary re- seems little doubt that the idea of mirror [or lationship with its Type 2 or Type 3 structure. reversed] recapitulation stems in this instance Mozart might just as readily have elected to from the inappropriate application of a dualistic omit the repeats to provide a briefer Type 2 for- nineteenth-century conception of sonata form, mat. Had he done so—and had he not written one in which expositions consist ‘essentially’ of “Coda” above m. 34—one might expect that Pand S,”>5 inaccurate analytical observations would have 52. Our observations here are based on the edition of degré de frénésie, par une descente chromatique sur the sonatas found in the Neue Mozart Ausgabe, ed. Wolf- la pédale de si, opérée par la derniére répétition de la gang Plath and Wolfgang Rehm (Kassel: Barenreiter, phrase corollaire.” Rpt. in Liszt, Lohengrin et Tannhduser

1986), p. 42. de Richard Wagner (Paris: Adef-Albatros, 1980), p. 158

53. In the Quartet from Idomeneo the provided text is (for the identification of the “phrase corollaire,” see set—as a first run-through—in the exposition. In paral- pp. 156-57). We thank Michael Puri for calling this to lel fashion, the second rotation—mm. 68-153, encom- our attention in a seminar paper on that overture. On passing the developmental space and tonal resolution— the other hand, cf. Thomas S. Grey’s identificaton of lays out the complete text for a second time. With the this overture as featuring a reversed recapitulation, in return of the P-incipit in the tonic at the end, following “Wagner, the Overture, and the Aesthetics of Musical an “undermined” conclusion to the second rotation on Form,” 19th-Century Music 12 (1988), 3-22, e.g., the a held V’ chord (m. 153), Idamante returns to the ini- diagram of the overture on p. 15 and the discussion on tial textual line of the piece, “Andro ramingo e solo” pp. 16-17. (m. 154). This restoration beyond the now-twice-fully- 55. Wolf, The Symphonies of Johann Stamitz, pp. 155-56.

consumed text of the Quartet (as if a new text-rota- To this statement Wolf added the following footnoted tion were beginning) identifies the motto-incipit as an remark, p. 162, n. 47: “The idea of mirror or reversed add-on coda-accretion to the Type 2 format proper. recapitulation evidently arose in the early nineteenth 54. The idea of interpreting a return of P in a Type 2 century; see Robert Schumann, ‘Sinfonie von H. Bersonata as a coda evidently persisted in some quarters at lioz,’ in his Gesammelte Schriften, ed. Martin Kreisig (5th

least into the middle of the nineteenth century—and ed., Leipzig, 1914), 1:73, in which Schumann attempts perhaps beyond. Liszt’s 1849 discussion in the Journal to show that the first movement of the Symphonie fantasdes débats of Wagner’s Tannhduser identifies the final re- tique is an arch form (A BC D CB A); he specifically turn of the P-idea in the Overture, the Venusberg music compares this scheme with dualistic sonata form.” A at m. 273, as a coda: “La Coda résume les principaux convenient translation of Schumann’s puzzling analysis dessins du début de l’Allégro, et arrive 4 son plus haut may be found in Berlioz, Fantastic Symphony, ed. Ed-

384 Elements of Sonata Theor Y

The P-Based Coda: The Potential for a postulate is the production of a double-rotaSecondary “Completion-Effect”’? tional format. We are given an expositional lay-

whil 4 h eth out, usually of the P TR’ S/ C type, followed : ; we ° om es “ee or the re- by a second rotation of that Anlage starting off VETSE ve hea. can kee we a with a development based on P and/or TR (or, ian at ¥ TP ase co can © 5 an ace sometimes, a proxy for the one or the other) and

ional sct o connotations ina Type ~ sonata, leading to an MC and a tonal resolution of S and as opposed to its presence in a Type 3 work. In C. To this primary structure a tonic-P-based Type 2 formats P-based codas provide us with coda may or may not be added, the onset of the potential for perceiving a particularly suit- an incipient, soon aborted third rotation. Since

able compretion eee that s he typically that P-based coda is often thematic, in the sense

P ee 5 i © YS: 1 ad the a DL call that, however fleetingly, it recaptures (even cel-

ous y ne © ONE MANS me ype sonata, 1 ebrates) the supposedly “missing” tonic return

any historical and generic sense, 1s not properly of the P-idea—an absence only perceivable by viewed as an incomplete structure. It does not importing the perhaps inapplicable Type 3 perrequire anything beyond itself for completion spective—the temptation is to regard that reson is own terms. On the other hand, from the toration as a fundamental aspect of the Type 2 perspective of the Type 3 sonata, which had structure. which it is not grown to become the preponderant norm after In most works in which it occurs. the once-

the 1770s, it could have been interpreted as a d “mirror” quality of such a P-cod

format that could be enhanced, made more sat- Deane Tr ne eY OF SHED a SCOR! Tes"

oe —? toration 1s also questionable. Here one would

istying, by means of restorative-P-coda supple- ae tg argue that, in terms of the ordering of

nen der th P-based cod broad sonata spaces according to this perspective, Ne net Mhese terms, a bpasec coda we have: (1) an exposition that may be regarded

within a Type 2 sonata form should not be con- as pre-MC material followed by post-MC mate-

meee a of any prestmed recapieay ation. rial; (2) a developmental space (which from this

(This judgment responds to a PONTE PANTO strained point of view 1s to be regarded as nonthat stems from a fundamental misunderstand- participatory); (3) the return of the post-MC nS CONCCTNINE the role of modular arrange- material in the tonic (the tonal resolution) folment within sonatas.) Still, the P-based coda in lowed by the tonic return of the P-incipit from a Type 2 sonata furnishes an unmistakable sense the exposition’s pre-MC material (as coda). This of Wrapping things up at the end with a high- argument would insist that the pattern of matelighted restoration of the piece’s main idea, its rial that is most obviously “thematic” (not deHauptgedanke, in the tonic. This impression 1s velopmental), the exposition’s P (tonic) =>postmade all the more telling because of the lack of MC (nontonic), seems recaptured and reversed any preceding Type 3 structure, in which a tonic- at the movement’s end: post-MC (tonic) =P key P, unheard since the exposition, would have (tonic). But the supposed “reversal” (or mirror)

recurred to launch a bona fide recapitulation. as Wolf pointed out, is ultimately grounded in

The important thing . ro Sse) Oe an unacceptably simplistic and outdated view of

tween primary structural Principles Chose modular arrangement within rotational layouts. governing the sonata form proper—and the po- S- and C-spaces are often multimodular in a tential for secondary effects (or even illusions), way that renders claims about simple “mirror however vivid those effects might initially seem effects” naive, particularly if they are taken as to be. In a Type 2 sonata the primary guiding fandamental structural features.

ward T. Cone (New York: Norton, 1971), pp. 220—48 mention of this piece in ch. 12. Schumann’s remark does (“A Symphony by Berlioz”). The first movement of play into the general fallacy of “reversal,” but it remains the Symphonie fantastique, however, is no Type 2 sonata. unclear where the misconstrual of the “reversed recapitRather, it is a deformational Type 3 whose recapitula- ulation” within Type 2 format was codified in analytical

tory rotation begins in the unusual key of V: see our practice. Cf. our remarks on Tobel, n. 19.

The Type 2 Sonata 385

That said, there can be little doubt that the tonic as a genuine, full-blown coda in the first restoration of P-as-coda, one source of the “re- movement of Symphony No. 20 in D, K. 133.97 versed recapitulation” fallacy, was doubtless P-material codas may also be found in the slow viewed by composers, early on, as a particularly movement of the Quartet in D, K. 155 (cited in attractive option within Type 2 sonata forms. example 17.4); in the Overture to Il re pastore, The concluding return of P in the tonic, re- K. 208; and in the Piano Sonata in E-flat, K. animating the “lost” tonic opening, however 282, also discussed above. Similarly, the tonic briefly, before merging into the final coda-ca- P-incipit returns as coda rhetoric interpolation dences, does fill what might be felt from the (CRD) in the slow movement to the Flute QuarType 3 perspective as a purposeful absence. Not tet in G, K. 2854; and in the two obviously reonly do P-incipit-based codas in Type 2 sonata _lated first movements that are often cited in this forms become more frequent as the century pro- regard, that of the Violin Sonata in D, K. 306, eressed, but composers also increasingly staged and the Piano Sonata in D, K. 311 (which, we

them as spotlighted, significant returns. recall, also displays a conversion from Type 3 to Within Type 2 sonatas the P-based coda-op- Type 2 principles midstream).

tion became more and more attractive in the A P-incipit-based coda or CRI may therelast decades of the eighteenth century, to the fore be regarded as in some senses compensapoint where it became a strong first-level de- tory within the Type 2 sonata form.°® Because fault. It was deployed more often than not, espe- the P-incipit had not been sounded in the tonic cially in Allegro movements. P-based codas are since the opening of the piece, it was at least fairly common in Stamitz symphonies, for in- pleasing to bring it back in this way in the coda. stance, and within the mature works of Haydn The restorative effect is even more gratifying (for whom the Type 2 option was generally less if the developmental space—the beginning of attractive) a classic example may be located, as Rotation 2—had suppressed a more obvious rementioned above, in the finale of his “Trauer” turn of P-material in favor of something freer Symphony, No. 44. For Mozart P-based codas or more episodic that had written over that P, became a strong first-level default in Type 2s asin K. 133/1, K. 155/11, K. 285a/i, K. 306/1, soon after his earliest symphonies. Two early and K. 311/1. In these cases one might conclude instances—amiudst several counter-examples that what was to be most essentially perceived without codas—can be found in the first move- at the opening of the developmental space was ments of his Symphony “No. 5” in B-flat, K. 22, the absence of the more normative P, particufrom 1765 (some thematic material for which is larly in a sonata form that was ultimately to be shown in example 17.3) and the unnumbered interpreted through Type 2, double-rotational Symphony in B-flat, K. Anh. 214 (45), perhaps lenses. Such sonatas can give the impression of a

from 1767.°° P shunted aside 1n the initial bars of Rotation 2. By the early 1770s it had become Mozart’s From that vantage-point, such a developmental most common practice within Type 2s to pro- opening can seem like a crucially placed rotavide a P-based coda (or CRI). P returns in the tional blank. One purpose of this could be to 56. Here we follow the suggested dating of K. Anh 214 final bars of the second rotation in a grand gesture of in-

(45b) in Zaslaw, Mozart’s Symphonies, p. 95. clusion. It is also helpful to hear K. 133/1 as one of a set 57. As noted also in Zaslaw, Mozart’s Symphonies, p. of similarly structured works (with crescendo-themes 238, with an allusion to “a kind of mirror form.” P’s for P) that include Stamitz’s Symphony in D, op. 3 no. return as coda also “corrects” the deceptive cadence in 2/1, and Mozart’s own Symphony “No. 5” in B-flat, the original theme (the sentential mm. 3-10) to a PAC K. 22/1. in m. 169—literally the first time that this theme is ever 58. For the concept of compensatory effects or brought to authentic-cadence completion. It may be this “fulfillments” within codas in general—especially in long-sought resolution to which the full tutti responds Beethoven’s Type 3 structures—see ch. 13. There (n. with such an ecstatic, forte repetition, mm. 170-77 (pure 30) we also note Caplin’s listing of compensatory funcaffirmation, with another I:PAC), followed by a “coda tions for codas in his Classical Form, pp. 186-91. to the coda,” mm. 177-82, that in part returns to the

386 Elements of Sonata Theory

keep the more normative P in reserve to be re- bars (mm. 162-63) and promptly fizzles into a stored in the tonic, and satisfyingly so, in the compressed and recomposed TR. The effect 1s post-Rotation-2 coda—as Zaslaw proposed was that of an original P-I'R space subjected here to the case with regard to the thematic arrange- an unanticipated collapse—producing a lack or ment of K. 133/1. In terms of the historical tra- “failure” at the onset of the recapitulatory rotadition, a compensatory return of P in the coda tion (which is nonetheless recognizable as such). was not originally generically obligatory, even Following that rotation, the initial moments of in cases where references to P had been omit- a more expansive, though still incomplete and ted entirely in Rotation 2 (K. 19/1, K. 43/1). recomposed, P return, now as compensatory It was principally in later years, especially af- discursive coda, beginning with an elision at m. ter 1770, that it advanced to become a strong 237. Here we get ten bars of the initial theme first-level default, while never attaining the sta- subjected to ellipses: mm. 237-46 reanimate tus, it seems, of becoming absolutely necessary mim. 1—6 but move directly to the major-minor or a fixed part of the essential form itself. decay of mm. 11-14 before dissolving into freer coda material, mm. 247-53, with a “coda to the

P-Compensatory Codas in coda” added at mm. 253-64.°° Non-Type 2 Sonatas

Finally, with regard to compensatory codas that Confronting Hard Cases: Flexibility in

restore a fuller tonic sense of P, we might only Sonata-Type Recognition point out that these are not limited to Type 2 sonatas. It can happen in a Type 3 sonata, for As categories of analysis, and from our perspec-

instance, that the moment of recapitulation tive today, Types 1, 2, 3, 4, and 5 formats are makes only a perfunctory gesture at P in the heuristic tools, historically defensible ways of tonic—a mere bar or two, as if merely mark- construing familiar repertories that make posing an obligatory structural station, before sible an unusually robust processing of what we proceeding into recapitulatory recomposition. now call sonata-form movements. The sonata Here a touchstone example is the first move- types are not “real” in the normal sense of the ment of Mozart’s Symphony No. 34 in C, K. word, and we do not seek to reify them or make 338. The exposition’s P opens majestically and them rigid here. In the mid- and late-eighteenth is quite expansive, mm. 1-20. The exposition century there is no reason to think that they had is not repeated, and the developmental space any concrete and separate existence apart from begins with a link in m. 112 and is noticeably the structure- and meaning-effects that they dominated by episodic (nonrotational) mate- | made possible—or that they make possible for rial. Toward the development’s end the music us today. Instead, they are regulative ideas (in grasps onto a prolonged dominant-lock, V of | the Kantian sense), reconstructed modes of apthe tonic C, at mm. 148-57, clearly a prepara- prehension that make certain types of perception for the recapitulation. That recapitulation tual coherence and meaning possible within this begins at once, with mm. 158-61 reanimating and later repertories. (Appendix 1 has more to the grand opening of P, mm. 1—4, in the tonic. say on this aspect of Sonata Theory’s “GroundAt this point, though, the music stalls into a —_—sing Principles.”)

slightly varied repetition of the preceding two Nor are the five sonata types absolutely in-

59. On discursive codas unfolding in two or more of this partially restorative coda that led Charles Rosen stages, the first part of which often accomplishes some- to mistake K. 338/1, more than once, for a sonata with thing more dramatically participatory vis a vis the pre- “the appearance of a recapitulation in reverse order” ceding sonata form and the last part of which 1s a “coda [sic|—our Type 2 with P-based coda—in Sonata Forms, to the coda,” see ch. 13. It was probably the collapsing P rev. ed., pp. 97 and 286. (Cf. n. 51 above.) at the onset of the recapitulation and the discursiveness

The Type 2 Sonata 387

dependent of each other. Here and in other We have devised the sonata types primarily chapters we have suggested potential points of | to serve as modes of processing the works of intersection between some of them, occasion- the most historically influential composers in ally gray areas where the Type 1 sonata with the decades after 1770—Haydn, Mozart, and an expanded retransitional link might come to Beethoven in particular (although our system 1s resemble a Type 3 with only a brief develop- also applicable, with only modest adjustments, mental space; where Type 3s appear to convert to most nineteenth-century work as well). As midstream into Type 2s, or the reverse; where many scholars have noted, mid-eighteenth-censome Type 2 norms are mixed with others from tury sonata formats and those sometimes found Type 3; where the Type 1 sonata with an ex- in the lesser masters of the second half of that tended P-based coda can come to resemble one century often seem less regularized, looser in kind of Type 4 sonata; and so on. Blends and their architectonic realizations—as though what overlaps among the sonata types are possible, came to be the various “standard types” of the even if they are not regularly encountered in form had not yet been fully crystallized in the standard-repertory works. In confronting hard —_ wide-ranging field of compositional practice. In

cases, we should be advised not to assume that our own work we have tried to bring several of the point of our analysis is to shoehorn a prob- these other composers into the discussion from lematic composition into only one category time to time, even while concentrating on the to the exclusion of another. When hybrids or — three most celebrated figures. This second- or ambiguities occur—as perceived through the third-tier repertory—encompassing thousands conceptual grids of the five sonata types—we of less ambitious and now largely forgotten should be prepared to accept them and even to works—1is where, from the perspective of the use the resultant structural friction as a potential five sonata types, numerous hard cases are likely enhancement of our own analytical work. Ex- to be found. Even here, though, employing the plicating the analytical problem 1s always pref- sonata types as tools for analysis can help us to erable to pretending to solve it through a deci- realize just what 1s and what is not “unusual” in

sionistic categorization. these compositions.

CHAPTER EIGHTEEN © "ISLE 0

Rondos and the ‘Type 4 Sonata

ihe 4rondo sonatas comprise a variety of sonataEach of these usually arrayed mixtures. The “unmixed” rondo,recapitulation. in the customary P TR’ S/C is pattern, although along with its simpler predecessor, the rondeau, a continuous format (AB = P TR=>FS / C) is a structure built primarily by the juxtaposi- is also possible, especially in some of Haydn’s tion of discrete sections, each of which 1s nor- Type 4 sonata forms. (As will be discussed tomally marked by memorably tuneful ideas. Its ward the end of this chapter, Haydn also often defining feature is the recurrence of a tonic-key treats the recapitulatory spaces of his Type 4s refrain (or “rondo theme’’) separating the ap- freely, sometimes in asymmetrical or recompearances of differing or contrasting episodes posed ways that challenge their expected “reca(or “couplets”), which are often, though by no pitulatory” functions. In these cases it is the exmeans always, in nontonic keys.! The rondo istence of at least an unequivocal exposition that proper encompasses formats traditionally de- is to be regarded as the hallmark of the Type 4 scribed by such letter-schemes as ABACA, sonata.) In other respects the Type 4 recalls the ABACADA, ABACABA, and the like, within older, more traditional rondo practice, which which thematic variations or shortenings of the remained viable as an alternative even after the A-idea, the refrain, are also possible in its later rise of the sonata-rondo in the early 1770s. This appearances. The “mixed” Type 4 sonata (sona- is most notably the case in the near-inevitabilta-rondo) is a rondo that has been shaped to be ity of its retransitions (RT) and tonic-key ronin dialogue with a Type 1 or Type 3 sonata or, do-theme refrains at pre-established locations. from another perspective, a Type 1 or Type 3 so- Within multimovement works, especially in nata that 1s also in dialogue with the rondo prin- pieces from the 1760s and 1770s onward, rondos ciple. In standard practice this usually means that and sonata-rondos (Type 4s) may be found as a the stark AB juxtapositions of the simpler rondo typical option for fast finales and slow moveformats (such as ABACABA) are converted into ments. As also noted by Malcolm S. Cole, the

a nonrepeated exposition and—at least in the rondo format was used only “rarely as the first most standard cases—a generally symmetrical movement (Haydn, Piano Sonata HXVI: 48).” 1. Any nontonic appearance of the refrain—qua “re- rondos and Beethoven’s occasional practice of sounding frain”—would be deformational to the norm. See the the rondo theme in the “wrong key” near the beginning later discussions of C. P. E. Bach’s Kenner und Liebhaber of the coda rotation.

388

Rondos and the Type 4 Sonata 389

Moroever, “it had limited use in chamber mu- rondo in the revised, 2001 New Grove Dictionary sic and the symphony; it was more freely em- reproduced his earlier one from the 1980 ediployed in sonatas and serenades, but only in the tion: “One of the most fundamental designs in concerto was it the almost invariable choice for music, the rondo is a structure consisting of a

finales.” series of sections, the first of which (the main Before proceeding to examine the Type 4 section or refrain) recurs, normally in the home mixture, it will be helpful to consider the rondo key, between subsidiary sections (couplets, epiconcept generally and then to look more closely sodes) before returning finally to conclude, or

at the unmixed rondeau and rondo formats. As round off, the composition (ABAC... A).”4 will be seen, the basic idea of the sonata-rondo Essentially the same definition was provided by (Type 4) is not particularly challenging. At one Douglass M. Green 1n 1979.° And in 1998 Willevel of inquiry, we need ask only: does the pre- liam E. Caplin introduced the form in a similar sumed “rondo” begin with a nonrepeated sonata fashion, noting additionally that “in the classiexposition—including a TR-idea following the cal era, however, most rondos can be situated in refrain—or does it not? But such simplicity 1s one of two main categories—the five-part rondo deceptive. The real analytical difficulty is that (ABACA) and the sonata-rondo (ABACABA). of attaining the ability to navigate through the Variants of each type create a number of other many variants and overlapping subtypes of ron- formal designs (e.g., ABACADA, ABACBA).’¢

deaux, rondos, and Type 4s that one finds 1n the It is not our intention to trace the historilate-eighteenth-century repertory. Once one cal origins and development of this form or the investigates the details of movements or indi- many varieties of its treatment by eighteenthvidual pieces labeled as rondos, one finds that and nineteenth-century composers.’ As has they often differ from each other in such mat- been noted, the term “rondo” 1s sometimes used ters as the structure and scope of the refrain and to designate three different manifestations of the the contrasting sections, the presence or absence general principle outlined above. Moving from of transitions and (especially) retransitions, the the simplest to the most complex, these are: (1) presence or absence of developmental and reca- the rondeau; (2) the rondo; and (3) the sonata-

pitulatory features, and so on. rondo (which we refer to as the Type 4 sonata). These formal categories differ in size, scope, and

internal elaboration. Their separateness from

Definitions one another, while generally clear, is not absolute. The categories blend into one another on a There is no disagreement on the basic principle — continuum of possibilities. Some of the problem

underpinning the rondo. As Schoenberg put is bound up with unstable terminology from it, “the rondo forms are characterized by the the eighteenth century. “Rondo, from the Italrepetition of one or more themes, separated by ian, became the standard term in classic times intervening contrasts.”’> Cole’s definition of the to cover all versions of this form,” while “the

2. Cole, “Rondo,” The New Grove Dictionary of Music the rondo and reappears at least twice. The passages beand Musicians, 2nd ed., ed. Stanley Sadie (London: Mac- tween appearances of the refrain are called episodes or

millan, 2001), 21:651. couplets.”

3. Arnold Schoenberg, Fundamentals of Musical Compo- 6. Caplin, Classical Form, p. 231. As will emerge, we do

sition, ed. Gerald Strang and Leonard Stein (London: not define the Type 4 sonata (sonata-rondo) with the

Faber and Faber, 1967), p. 190. familiar ABACABA design—which, more appropri-

4. Cole, “Rondo,” p. 649. ately, describes the seven-part symmetrical rondo. For 5. Green, Form in Tonal Music: An Introduction to Analy- an exposition and critique of Schenker’s view of rondo

sis, 2nd ed. (New York: Holt, Rinehart, and Winston, form, see Joel Galand, “Form, Genre, and Style in the 1979), p. 153: “The word ‘rondo’ is a generic name re- Eighteenth-Century Rondo,” Music Theory Spectrum 17 ferring to those compositions that are distinguished by (1995), 27-52. frequent recurrence of a refrain. The refrain is normally 7. See also, e. g., Cole, “Rondo,” and Ratner, Classic a self-contained, harmonically ‘closed’ passage. It begins Music, pp. 248-59 (“Couplet Forms’’).

390. Elements of Sonata Theory

prescription for a rondeau [the French term] was Rondeaux, Rondos, and Rotations:

[more] explicit.”8 Even so, present-day assess- A Preliminary Note ments of individual pieces usually fall more or less comfortably into one category or another. Rondeau, rondo, and sonata-rondo formats are Those assessments have different implications in dialogue with the rotational principle. Each for analytical interpretation. For the purposes successive appearance of the refrain (“A” in the of analysis and hermeneutic clarity, it is help- usual schematic letters appropriate for rondeaux ful to recognize distinctions among these three and rondos) begins a new rotation. Each rotation

rondo formats. Our presentations of the ron- is marked by a similar opening, even though deau, rondo, and Type 4 are heuristic constructs what follows in the remainder can differ from or “ideal types’—conceptual formats that are one rotation to the next: AB—AC-—AD (if apuseful in sharpening current analytical preci- plicable)—and so on, usually concluding either sion. In brief overviews we shall define and de- with a single A (a half-rotation) or A + coda limit each of these structures separately. (a full coda rotation). In the cases of the symBefore doing so we might underscore a few metrical or tonally resolved formats, two of the things. First, while the customary alphabetic la- rotations will contain similar materials: an inibeling (ABACA, and so on) may be useful for tial rotation, “AB” with B in the dominant or the simpler rondeau and the rondo, it proves other nontonic key, may be recycled toward the counterproductive when applied to the elabora- end, with B at that point sounded in the tonic. tion and complexity of the Type 4 sonata. This Assuming for the moment a four-rotation prois also the case with the terms “couplet” or “epi- cedure, this could produce a rondo format on sode,” which last term has been used almost in- the order of: AB—AC-—AB'-A + coda. A simpler,

variably in prior discussions of the sonata-rondo. three-rotation (or two-and-a-half-rotation) We use “couplets” for the contrasting sections of structure could be schematized as: AB—AC-—A. the rondeau, but “episodes” for the usually more The rondo refrain is always a beginning, never elaborate ones found in the rondo. In the sonata- an ending. It always begins something new.

rondo we switch instead to the PTR ’S/C To construe rondo structures in this way labels. ‘To be sure, in the Type 4 sonata P also might at first seem unusual, since most analysts

functions as a refrain—a characteristically have been accustomed to the undivided string “rondo-style” theme recurring in the tonic at of letters found in such textbook formulas as predesignated spots. For this reason we usually ABACA or ABACABI'A. And yet our rotadesignate it as the P-refrain or P'f. But we discard tional divisions find support in the writings of the term “episode” as unhelpful when applied to two early-nineteenth-century theorists, Anton the expositional and recapitulatory rotations. Reicha and Carl Czerny. In the second volume It might be useful at the outset to suggest of his Traité de haute composition musicale (1824— diagrammatically some of the distinctions that 25), Reicha supplemented his somewhat curious we observe among the three types—rondeau, discussion of what we refer to as a symmetrical rondo, and sonata-rondo. Once again, the pos- rondo (or perhaps a sonata-rondo proper) with sibilities exist on a continuum, potentially shad- an explanatory diagram that segmented the ing into one another. Yet, for the most part, they movement in a way that bolsters a rotational unremain conceptually and heuristically separable, derstanding of the form.’ Reicha’s paradigm in

as suggested in table 18.1. this “Coupe du Rondo” diagram, however, was not the ABACABI'A pattern commonly associated with the sonata-rondo but rather a lengthier and non-normative one, ABACADAB' coda

8. Ratner, Classic Music, p. 249. This article is the source for our discussions of Reicha 9. Reicha’s discussion is summarized in Malcolm Cole, and Czerny on rondos. The original Reicha source, “Sonata-Rondo: The Formulation of a Theoretical as listed by Cole, 1s Traité de haute Composition Musicale Concept,” The Musical Quarterly 55 (1969), pp. 185-86. (Paris: 1824-25), 2:301-3.

Rondos and the Type 4 Sonata 391

TABLE 18.1 From Rondeau to Sonata-Rondo: A Continuum of Formal Categories

Rondeau Rondo Sonata-rondo (Type 4) |Sonata| “Refrain” and “couplets” “Refrain” and “episodes” Sonata terminology 1s preterable; PTR.’ S/ ©

Alternation of simple Alternation, but of TR-zone follows the

melodic structures somewhat more initial “refrain”

(period, hybrid, or complex/expanded (refrain = P or P*t) group of phrases) structures (esp. binary, rounded binary, sometimes with repeats)

Few or no links or More elaborate First rotation is explicitly

retransitions between retransitions as the structured as the

the sections episodes return to the exposition of a sonata, refrain, but no TR with KT at its end

between A and B

Refrain usually Returns of refrain are A later rotation returns literally sometimes truncated recapitulates and resolves the expositional rotation May be in dialogue with the Type 1, the expanded Type 1, the Type 3, or the Type 5 sonata

(with no return of A prior to the coda). Never- He then went on, however (as noted by Cole), theless, his larger point is clear. He unequivo- to describe another possibility: cally divided his theoretical major-mode rondo

into four “sections”: (1) AB; (2) AC; (3) AD; If this Rondo had been written on a greater (4) AB'—followed by the coda. (He also indi- scale, a longer and more decided middle subject

cated that B was to be in V; C in IV; D ini must have been interwoven, which would then

(tonic minor); and B' in [.)'° have been repeated in the third principal period Similarly, in his School of Practical Composition Ga oe ee yea pee aaah (1849) Czerny first devoted considerable space have followed, which would have formed a more , extended and brilliant conclusion, and a longer

; ; ; ;into ; Oda. rience, “periods,” a iWondo, according to aits vided three principal cach be- ae er ee en ee extent, ee weof ; os —s consists of three or four principal periods, and to a rondo of the ABACA type, which he di- Gti. Piste. o Rando. akesilineae ie este, pn with the refrain: AB—AC—A + coda. as many repetitions of the principal subject.!!

10. In his treatment of Reicha on the “sonata-rondo” too much the ternary style [i.c., sonata form].” If one (“Sonata-Rondo: The Formulation of a Theoretical were to suppress Reicha’s Section 2, the result would Concept,” pp. 185-86), Cole also pointed out that the be AB—AD-AB'-coda, which would indeed bring the theorist’s quirky treatment of the form indicated that the form close to cither our symmetrical seven-part rondo fourth section was to be “the longest and the most im- or our Type 3 sonata-rondo mixture (what we call the portant,” with its B’ section “recalling that which 1s the Type 4 sonata), especially if the coda were based upon, most salient in the three preceding sections.” Cole ad- or preceded by a return of, the main idea (A). ditionally noted that “all sections are connected by tran- 11. Quoted in Cole, “Sonata-Rondo: The Formulation sitions and retransitions” and quoted Reicha’s remark ofa Theoretical Concept,” p. 187. Czerny’s discussion of that his prolix pattern could “be abridged by suppress- rondos, as cited by Cole, appeared in his School of Practiing the second section; but in this case it will resemble cal Composition, Opus 600 (London, 1849), 1:67-81.

392 Elements of Sonata Theory

Czerny’s rondo of four principal periods has been applied to a body of seventeenth- and (AB—AC-AB'-A + coda) is either the symmetri- early eighteenth-century French compositions cal seven-part rondo or the Type 3 sonata-rondo characterized by the alternation of a harmonimixture proper. The main point, though, 1s that cally closed refrain (A) in the tonic key with his segmented layout of the form 1s congruent contrasting couplets (B, C, and so on), usually with our rotational view of these structures. At in different keys.!4 The rondeau appeared in a least some nineteenth-century theorists rec- variety of contexts in the decades before 1750: ognized the rotational aspect of this form, an ballet, opera, orchestral music, violin sonatas, insight that was apparently lost in the middle and harpsichord pieces.!> The number of couof the nineteenth century, perhaps as a result plets varied from one to as many as eight, but of A. B. Marx’s evolutionary view concerning most rondeaux contained two (the usual numhis five rondo types as progressive steps of the ber in Rameau’s harpsichord works), three, or “spirit,” striving ultimately to attain the greater four. Differentiations may therefore be made

cohesiveness and “ternary” symmetry provided among the single-couplet rondeau (ABA), by sonata form.!* However the swerve away the two-couplet rondeau (AB—AC-A), and from the quasi-rotational views of Reicha and the multicouplet rondeau or “chain rondeau” Czerny occurred, its loss was perpetuated dur- (AB—AC-AD-... A). ing the production of twentieth-century Formen-

lehre, which sometimes, misleadingly, viewed Traditional and Expanded Rondeaux the symmetrical rondo types as large-scale ter-

nary structures.!9 Especially when confronting pieces from multimovement works written in the last several decades of the century, one might also make the

The Rondeau distinction between the traditional and the ex-

panded rondeau. The traditional rondeau—asAs presented by Malcolm S. Cole (whose New sociated with the simple letter-schemes suitable Grove descriptions of earlier formats we sum- for works of Couperin and Rameau—is charmarize and adapt slightly here), the term “ron- acterized by the straightforward alternation of deau,” suggesting a round or circular motion, _ brief melodic structures. With its short-winded 12. Marx’s view may be found, e.g., in Die Lehre von der was more florid and bound up with the Germanic phiMuikalischen Komposition, 2nd ed. (Leipzig: Breitkopf & losophies of the time (Musical Form, pp. 77-78): “Every

Hartel, 1848), 3:94—200 and 307-13. A translation of form is a restraint, a fetter for the spirit that has come to the most relevant material is available in Marx, Musical belong to it. With every succeeding form, the spirit is Form in the Age of Beethoven: Selected Writings on Theory released into a new perspective. The spirit is free only

and Method, ed. and trans. by Scott Burnham (Cam- when it possesses all the forms, as well as the complete bridge: Cambridge University Press, 1997), pp. 78-82. power to build them—and, in requisite cases, to build Marx’s fourth rondo form, for instance, is basically an new ones. Every form is an expression of formative reaABACAB’ structure, similar to the sonata-rondo form. son, which finds its complete justification only in the Rather than calling attention to its rotational underpin- sum [Inbegriff| of all forms.” nings, Marx sought to call attention to its “clearly de- 13. Green, for example (Form in Tonal Music, 2nd ed., p. lineated ternary structure’—that is, the outer relation- 161), divided one type of symmetrical rondo into three ships between AB and AB’ (the two Hauptsatz-Seitensatz parts, ABA—C-ABVA, thereby obscuring the rotational

pairs, the second of which is tonally resolved). (Marx, impulse that underlies this form. Musical Form, pp. 80-81.) As pointed out in Cole, “So- 14. On the historical etymology and principle of circunata-Rondo: The Formulation of a Theoretical Con- larity or roundedness embedded in the term “rondeau” cept,” p. 188, Marx’s (“historically inaccurate”) point (“rondo”), stretching back to poetic formats of thirwas to suggest that each rondo type showed an increased teenth and fourteenth centuries, see also Fritz Reckow, striving toward the goal of sonata form, with its exposi- “Rondellus/rondeau, rota,” in Hans Heinrich Eggetion and symmetrical recapitulation. Thus each rondo brecht, ed. Handwérterbuch der musikalischen Terminologie

type, he argued, developed (in Cole’s words) “when (Stuttgart: Steiner, n.d.), pp. 1-7. composers sensed something lacking in the type imme- 15. Cole, “Rondo,” pp. 649-50. diately preceding.” Marx’s own way of expressing this

Rondos and the Type 4 Sonata 393

themes and nonpostponed recurrences of the tion is sixteen bars long. The tonic refrain (A) refrain, the traditional rondeau establishes the is a compound period with cadence pattern base-line of assessment when confronting later IAC/PAC. The first couplet (B) 1s a modulating adaptations. The refrain is usually a simple pe- compound period that begins in G major (III) riod or a similar, comparably concise musical and leads to a PAC in B minor (v). The second idea. Each couplet is also brief, generally either couplet is a compound sentence that begins in a period or a group of phrases. Sometimes—es- A minor (iv) and leads to a PAC in G major pecially in earlier examples—the refrain 1s writ- (III). The strongest cadences articulate the tonal

ten out only once, with directions to repeat it design i-v—i-III—i, a pattern that reverses the after every couplet. Only the briefest of inks Gf ©=Rameau minor-mode design. Each return of the

any) leads from the end of a couplet into the re- refrain is written out, and there are no transiturn of the refrain.!° Any couplet may be tonally tions or links between sections. The basic princlosed in a single (contrasting) key, or it may ciple is one of simple, sectional juxtaposition. be harmonically progressive, modulating from one key to another. As Cole reported, Rameau S The Multicouplet Rondeau two-couplet rondeaux (AB—AC-A) display a fairly uniform key scheme. His major-mode ron- Even as Cole noted that the “assimilation [of the deaux typically traverse the keys, I-V—I-vi—I; rondeau] into the music of other nations and its

the minor-mode ones, i-IJJ—i-v—1.!” transformation into the rondo of the Classical What we call the expanded rondeau, which period have not been adequately investigated,”!®

will be treated further below, builds on and it does seems clear that in the early 1770s the complicates the traditional rondeau principle young Mozart was writing them, even though either by turning some or all of the couplets he may have labeled a piece or movement in into a string of small melodic-harmonic forms, this format as a “rondo” instead of a “rondeau.” thereby postponing the return to the melodi- Neither a simple respelling nor the adoption of cally brief A, or by expanding one or more of its the Italian version of a formal term amounts to a sections into a larger format, perhaps a rounded transformation of essential formal structure. binary structure or something similar—an ex- Consider the finale, Allegro grazioso, of Mopansion that nudges the whole piece in the di- zart’s Quartet in B-flat, K. 159, written in 1773. rection of a rondo. For the moment we shall Although marked “rondo,” this is a clear fourset these issues aside to confine ourselves to the couplet rondeau (AB—AC—AD-—AE-A). The re-

traditional rondeau. frain (A) recurs each time in the tonic B-flat, while the four couplets are tonally closed in F

The Two-Couplet Rondeau (V), G minor (vi), B-flat minor (i), and B-flat

major (1), respectively. ‘The overall harmonic

The qualification “en rondeau” (“in the form plan is thus: -V—I-vi—I-1—I-I—I. (Here one of a rondeau”’) might be appended to any dance notices that the first five sections replicate the title: “gigue en rondeau,” “menuet en rondeau,” Rameau major-mode model: I-V—I-vi—I.) and the like. J. S. Bach’s Fifth English Suite in The brief refrain (A, mm. 1—8) is most simply E Minor (BWV 810) contains a movement en- regarded as a clipped, eight-bar period with an titled “Passepied en Rondeau,” which serves as only slightly altered consequent. On its first a clear exemplar of the form (and demonstrates two appearances (the second is in mm. 25-32), that neither the form nor the term was confined it is literally repeated, while its third and fourth to France). This dance movement unfolds as a statements (mm. 49-64, 97-112) are both subtwo-couplet rondeau: AB—AC-A. Each sec- jected to written-out, varied repeats. Couplet 1 16. While normative, the transitionless rondeau 1s not plet with an ensuing return of the refrain (op. 2 no. 4, without exception. Cole, “Rondo,” p. 650, notes that Aria).” in the later eighteenth century “Leclair... was among 17. Cole, “Rondo,” p. 650. the first to compose a linking passage to connect a cou- 18. Cole, “Rondo,” p. 650.

394 Elements of Sonata Theory

(B, mm. 9-24), the most “loosely” structured couplet 3 suggests such an expansion (a sectional of the four contrasting sections, is a sixteen- simple binary); otherwise, each section is a twobar phrase comprising four different four-bar phrase period or hybrid, or (in the case of coumodules, the last of which drives to a V:PAC. plet 1) a single phrase. Couplet 2 (C) is a sixteen-bar compound pe-

riod in the minor submediant. Couplet ° D, The Symmetrical Three-Couplet Rondeau mm. 65—96) 1s a thirty-two -bar sectional simple

binary that comprises two compound periods, Two other rondeaux by Mozart are worth exboth tonally closed in the tonic minor. Cou- amining, especially since these two pieces have plet 4 (E, mm. 113-28) is a sixteen-bar com- been cited as the first two sonata-rondos: the pound “hybrid 1” (antecedent + continuation). finales of the Quartet in C, K. 157 (1772-73) The form 1s complicated by the insertion of a and the Symphony No. 23 in D, K. 181 (1773).!° coda-rhetoric interpolation (CRI, mm. 129— __— But if we define our subtypes carefully, such a 36) before the final return of the refrain (mm. matter-of-fact statement claims too much. Nei137-44), as well as by the addition of a genuine ther movement contains a sufficient number of coda following this return (mm. 145-59). This features to be considered a genuine sonata-rondo piece may thus be more precisely schematized (or Type 4 sonata). Indeed, in their structural as: AB—AC-A'D—A'E [CRI]-A + coda. simplicity they even fall short of being properly Several factors place this movement in the classified as normative rondos. Instead, each 1s a category or tradition of the rondeau rather than rondeau that has been submitted to one or more that of the more elaborated, normative rondo additional factors (especially tonal factors) that (notwithstanding Mozart’s designation of the suggest a conceptual slide in the direction of movement). First, there are no transitions or what would eventually become a full-scale so(especially) retransitions between the sections. nata-rondo. Both pieces exemplify a three-cou-

Second, the refrain is laid out as a simple eight- plet rondeau of the AB—AC—AB'—A bar, two-phrase design. Finally, couplets 1, 2, variety, where couplet 3 (B') 1s a transposiand 4 are symmetrical sixteen-bar structures, tion to the tonic key of the nontonic couplet 1 each containing either one or two phrases and (B). This procedure can hardly be said to proonly one PAC. Couplet 3 is the only section that duce a full-scale sonata-rondo, since it lacks the

contains two periodic structures (16 + 16 bars) defining feature of the Type 4, that of beginand consequently two PACs. Additionally, as we ning with a generic expositional rotation that shall see, a distinguishing feature of the classical includes a section of TR. At most we might obrondo is the expansion of the contrasting sec- serve that each of these two rondeaux registers tions, and sometimes the refrain as well, into the effects of modest sonata urgings. We refer tonally closed binary forms of either the simple to such a structure as a symmetrical three-couplet or rounded variety. In the finale of K. 159 only rondeau—a structure defined by its featuring of 19. Cole, “Rondo,” p. 653: “Mozart, in the String tett 157, D-Symphonie 181”), see his Die Formenwelt Quartet K157 (1772-73), composed the first known so- der klassischen Instrumentalmusik (Bern and Leipzig: Paul

nata-rondo. ... Haydn adopted it somewhat later (Sym- Haupt, 1935), p. 183. Internal structure aside, arguphonies nos. 64, 66, and 69 of the 1770s in one view; ments against the absolute priority of Mozart for the soSymphony no. 77, 1782, in another).” Cf. the earlier nata-rondo concept are presented in Stephen C. Fisher, remarks of Cole from 1969, “Sonata-Rondo, The For- “Purther Thoughts on Haydn’s Symphonic Rondo Fimulation of a Theoretical Concept,” p. 182: “In 1924, nales,” Haydn Yearbook XVII [1992], ed. H. C. Robbins Wilhelm Fischer ascribed the invention of the sona- Landon, Otto Biba, I. M. Bruce, and David Wyn Jones ta-rondo to Haydn. In 1935, Rudolf von Tobel stated (Eisenstadt: Joseph Haydn Stiftung, 1992), pp. 85-107, decisively: “The sonata-rondo 1s created [by Mozart] in e.g., at p. 85 (“Haydn’s first symphonic sonata-rondo 1773, C Major Quartet K. 157, D Major Symphony K. by the usual formulation is the orphan finale Hob. Ia:4, 181.’ Tobel’s view is the prevailing one today, and my which may well pre-date his first meeting with Moinvestigations so far tend to confirm his statement.” For zart,’ and at p. 106, “The sonata-rondo concept was the original claim and its context in Tobel (“Mit dem plainly in the air, though, and Mozart has lost his claim

Jahre 1773 ist das Sonatenrondo geschaffen: C-Quar- to priority.”

Rondos and the Type 4 Sonata 395

a tonal resolution of couplet 1 later in the piece, 1 material. To be sure, if one were to invoke the in the space traditionally allotted to couplet earlier image of the continuum of formal types 3. (When we find the same sectional pattern, (table 18.1), with the “pure” rondeau at one end AB-—AC-AB'-A, with B' as a tonic transposi- and the fully-formed sonata-rondo (Type 4 sotion of the nontonic B, in the classical rondo, nata) at the other, these two tendencies in K. which typically features more elaborate internal 181/111 might push it a small distance along the structures than that of the rondeau as well as scale away from the rondeau end. But bluntly to clear retransitions back to A, we refer to it as a call it one of the first sonata-rondos ever comsymmetrical seven-part rondo, a structure that has posed is exaggerated.

still not yet attained Type 4 status.) The same 1s true of the slightly earlier piece, Consider first the Presto assai finale of Mo- the Presto finale of Mozart’s Quartet in C, K. zart’s Symphony No. 23 in D, K. 181, which may 157. Here yet another “sonata tendency” might be schematized as AB—AC (RT)—AB'-A + coda, be observed in that the refrain (A, mm. 1-16) is with a tonal scheme of I-V—I-1—I-I—I. The followed by a four-bar transition (TR, mm. 17D-major refrain is a sixteen-bar compound pe- 20) that modulates to the dominant 1n preparariod stated by the full orchestra, forte; 1ts second tion for couplet 1 (B). Following the third stateand third appearances (mm. 41—56, 81-96) are ment of the refrain, however (mm. 65-80), this exact, while its final statement (mm. 121-52) transition 1s eliminated before the tonic transpois repeated. During the repeat, mm. 137-52, sition of the B material as couplet 3 (B'). Couplet the antecedent phrase is played piano by strings 2 is sounded in the minor tonic (i). The entire alone, while the consequent phrase is again formal design is A (TR) B—AC—AB'-A + coda, played forte by full orchestra, to which a brief — while the tonal design is -V—I-i—I-I—I. Both coda is appended (mm. 153-66). Couplet 1 (B, the refrain (A) and couplet 2 (C, mm. 49-64) mm. 17—40) shifts to the dominant key with a are sixteen-bar periods, while couplet 1 (B, twenty-four-bar hybrid comprising a presenta- mm. 21-32) is a sentence with repeated contintional-antecedent phrase followed by a repeated uation motivated by a deceptive cadence. Durand varied continuation phrase. The continu- ing couplet 3 (B', mm. 81-100) this sentence 1s ation is first played forte by the full orchestra; stated twice, the first time without the decepits varied repetition is sounded piano by the tive cadence and the repetition of the continuastrings alone. The B theme 1s grounded in A tion. The formal simplicity of each section and major (V). When it recurs as couplet 3 (B', mm. the absence of retransitions urge us not to con97-120), the antecedent phrase is rewritten to sider this piece within the category of a rondo remain in the tonic key, while the continuation proper. The presence of a modulatory transiphrase and its repetition are transposed to the tion, though, however slight, does increase the tonic and revoiced. Couplet 2 (C, mm. 57-72) “sonata tendency’ evident in the dominant-tois a sixteen-bar period in the tonic minor. An tonic transposition of the B material. We might eight-bar dominant preparation (mm. 73-80) therefore nudge this movement further along ushers in the third appearance of the refrain. the table 18.1 continuum than we did K. 181/1u. As in the case of the multicouplet rondeau Nevertheless, it is still not a full-scale sonatamentioned earlier, K. 159/111, the relatively el- rondo. It is specifically to cover such situations ementary structure of the refrain and the cou- that we use the term “symmetrical three-couplets, as well as the general lack of retransitions, plet rondeau.”

situate this piece squarely within the tradition With the consideration of the symmetrical of the rondeau, not that of the more elaborate three-couplet rondeau the rotational aspect of rondo. On this interpretation, the only “rondo rondeau (and rondo) practice becomes more tendency” is the extended dominant retransition evident. Rather than viewing it as a simple al(mm. 73-80) that prepares the third statement ternation of seven contrasting sections (eight if of the refrain, while the only “sonata tendency” one includes the optional coda), it may also be is the transposition to the tonic (with slight re- heard as a series of four rotations, each of which composition) of the dominant-oriented Couplet begins with the tonic refrain:

396 Elements of Sonata Theory

Rotation 1: A B Ends V:PAC and rondos, produces a string of differing but Rotation 2: AC Ends X:PAC often complementary melodies, each of which

Rotation 4: A Coda Ends I: PAC . . . less self-sufficient couplet in a traditional ronKoraton 9: AB Ends :PAC on its own might have qualified as a more or

, . deau.earlier Especially when this principle isycarried As .discussed 1n this chapter, such a rotacee P Pis .an . ae out in more than one couplet, the result tional reading of this format (which 1s also trans- . oo: . extended chain of differing melodies, one af-

ferable to. ter other rondeau and rondo variants)“decide 1s , re another, which only occasionally also consistent with early treatments of this and . y: — oa to return to the A-refrain as a melodic stabilizer

ourth r - ; . .

. for the whole succession.

similar structures by the early theorists, includ- ,

ing Reicha and Czerny. We might also observe

A touchstone case of the expanded rondeau that in the symmetrical three-couplet rondo the .

fourth refrain is the only one that follows a rota can be found in the finale, Allegro molto, of

with ends PAC a th key | Mozart’s Serenade inkey. B-flat for Thirteen Winds, tionh dsthat with a in tonic In . . marked . y K. 361the (which the composer himself as other words, only the final refrain takes place cc 3 , — a “rondo’”’). we find a series of eightand within a tonic thatHere hassimple been fully actualized, . . bars , , sixteen-measure structures (sixteen

made into a concrete reality.?° . .

for the refrain, eight bars for each contrasting melody), each closed with a PAC in the relevant

The Expanded Rondeau key and each (excepting the final appearance of

, . . . the refrain) individually subjected tocited a literal reThe traditional rondeau, including those . peat by means of a notated repeat sign. Thus above, alternates. aeach briefmelodic refrain (sentence, pe. _from what precedes unit 1s set off riod, hybrid) with a series comparably brief, . . . . and what follows it. Theofrecurring B-flat-tonic single couplets, in the AB—AC-—... A pattern. a, ,directly refrainjuxtaposed, (A, mm.back-and-forth 1-16, a compound period),. the In this me. . . . tonic-key multiple A-refrain short melodies, the complete ablodic .practice, the returns 7and sence of transitions mark the movement as beimmediately after each short-lived couplet, _ . longing to the the rondeau (not the rondo) and.transitions between melodic blocks arestructure . - cate. sory. In this mushrooming additive nonexistent or rare. The effect is that of a rapid one should probably regard all of the repeated alternation of memorable themes, everyforming other . . sets . or , eight-bar melodies as couplets, one of which 1s the A-refrain. If one or more of .this .. . clusters between the sixteen-bar refrain. On the couplet sections, however, were be en- . . as : reading, the movement could be to schematized

larged tooo, include a series or cluster two“letter” or , ce1svoheard . follows (remembering thatofeach more which stillexcept .. , . characteristic twice viaideas—each a literalofrepeat, maintained the completeness and short-wind- A):

for the final

edness of a typical couplet theme—we could ) speak of an expanded rondeau. The B couplets, nm 1 17 28 33 41.57 68 73 31 89 10446

for instance (and/or later ones as well), could ABB BAC! OC Cs A Coda be expanded into a chain of differing short mel- [ LV—l vil Wio—vi vii tod odies, B!, B?, B%, and so on, before returning

to the A-refrain.*! This perhaps “playful” sce- Looming behind such a structure is the tranario, capitalizing on the connotations of light- ditional two-couplet rondeau (AB—AC-A), alness and freedom surrounding many rondeaux _ though the latter’s normatively single-strain

20. On the concept of the ontological status of the tonic 21. Cf. the description of rondos (and episodes) that within modulatory structures such as sonatas, see the sec- may contain multiple “sections,” in A. F. C. Kollmann, tion, “Tonal Potential, Tonic Presence,” near the con- An Essay on Practical Musical Composition (London, 1799), clusion of chapter 11, along with the section at the end [chapter 1, section 10] (rpt. New York: Da Capo Press, of this chapter, “ESC Issues in Type 4 Sonatas: the Dou- 1973), p. 4. ble Perspective.”

Rondos and the Type 4 Sonata 397

B and C couplets have now multiplied into a binary or rounded binary format for A [or P| succession of contrasting, self-enclosed tunes. in a rondo [or Type 4] proper.) Similarly, the Moreover, in K. 361/vii the otherwise differ- C section is a sectional balanced binary, again ing couplet-ideas B?, C2, C3, and (possibly) C4 with internal repeats, whose second part, a senare interrelated with each other through their tentially designed “bbc—c,” features 1n its 1mshifts to the minor mode and their deployment mediately reiterated continuation, c, a rhyming of the “Turkish” idiom—a favorite topical al- cadence with that of the first part, a. lusion in many rondeaux, rondos, and ‘Type 4s While most closely related to the expanded (mildly evoking a sense of the Near-Eastern rondeau in its emphatic, transitionless sectional“Other,” apparently with an “insider,” witty ism and additive string of related “B” melodies, intent from an assumed position of implied cul- Haydn’s Rondo all’ongarese 1s poised somewhere

tural superiority). In other words, the B and C between that format and the rondo proper. sectionally additive couplet-strings begin in a In addition, all sections of this finale remain “Western”-style major mode (essentially the fixedly in the tonic G (unlike the normative tonic in B!-B?, the subdominant in C!) only rondeau or rondo), although shifting from time to “turn Turkish” as they proceed—more em- to time to G minor and back again in a clear phatically so in the C-cluster. The mock-exotic maggiore-minore alternation that is doubtless part implication 1s also reinforced by the Serenade’s of the Hungarian flavor, along with the vigorthick-wind timbres, also a component of ‘Turk- ously folk-like, perhaps Romany topoi of the

ish janissary music. melody types.?? Another example 1s Haydn’s famous Rondo all’Ongarese, the Presto finale of the Piano Trio

in G, Hob. XV:25—like K. 361/vii, expand- The Rondo ing the AB—AC-A rondeau format, albeit in different ways and with differing implications. Influence of Earlier Dance-Suite

Like K. 361/vii, this piece features closed, Binary Formats?

“catchy” melodic sections, several repeat signs, _

and no transitions or retransitions at any point One way that the rondo PFOPCr May be distin(although there are sometimes patches of con- euished from the rondeau is that it expands the nective fill at the joints). Moreover, its B-space latter's relatively simple phrase structures (pecomprises an additive series of brief, though un- tiods, hybrids, and so on) into MOTE complex repeated, melodic forms (B', B2, B?, B4). These forms, usually various types of binary forms. features would seem to place it in the general This expansion may have been influenced by category of the expanded rondeau. But in this the Baroque da capo procedure in suite MOVE instance two of its major sections are not mere ments, 10 which 4 COMPOSCT could write, for ex-

one-strain tunes but larger structures—binary ample, two minuets, entitle them “Menuet I” forms—which slide this expanded-rondeau and “Menuet II,” and follow the second with movement in the direction of a rondo proper. the instruction “Menuet I da capo.” This is of In short, the movement is an expanded ron- course the origin of the Minuet and ‘Trio form deau with significant leanings toward becom- within multimovement sonata practice, but in ing a rondo. Moreover, the A-refrain 1s itself earlier times it was also applied to gavottes, one of these larger units: a sectional rounded bourrées, passepieds, and so on. The second binary, a b+a' with internal repeats (the first of dance usually employed the same tonic as the which is written out), mm. 1-34; the melody is first, p erhap s with a change of mode, although dl shortened to aba, without repeats, in later ap- contrasting key was a possibility. But the tTPOrpearances. (Haydn—and others—often prefer a tant structural point is that each dance was itself

22. Another variant of the expanded rondeau, that movement of the Piano Sonata in D, K. 284, is menfound in Mozart’s “Polonaise en rondeau,” the second tioned in n. 45.

398 Elements of Sonata Theory

a binary form (complete with double bars and Refrains of “Rondo Character”; Occasional repeats); the result was a compound ternary form. “False Predictions” and Composers’ Similarly, the refrain (A) of normative later- “Muislabelings”

. In in AllegroAB-AC-AD-A, tempo or faster in therondos formats AB—AC-A, } " po(encoun© ken tered in finales), the A-refrain 1s usually nimble, eighteenth-century rondos (for example, those .

or AB—AC-—AB-A) was oftenThere constructed as a , 3 , , . oo playful, or “tuneful.” is a characteristic, . . lighterrecalling feel to many thesepractice. rondo themes— and repeats), perhaps daofcapo .._ something popular in flavor, something inSubsequent. .statements of the refrain were often 7 . stantly memorable, aor“contredanse character.” ?>should . i. . Notorevery “A-theme” statement be con-

calling earlier ritornello practice. Charles Rosen . . .

,them , sidered a refrain proper. Mm. 21-24, in ,G—a assessed as “essentially modulating fanta5 55 . first, “wrong-key” attempt at a local reprise— sias.”°3 In so subordinated doing, they would appear to fallwithin . . - y: the tothe their position under re the. are implicit censure of theorist Koll. . block, _ 1799 . larger, tonic-centered rounded-binary mann, who in (citing works of Bach) dis. . 1-42. , , > 3 cc , adapted and varied rondos as a whole inwhich mm. tinguished between “proper” (“in . . of . i. To regard mm. 21—24 as a genuine refrain the .first section always returns in theweight principal . one found —_ . Sasincomparable structural as the key, either its original form, or varied”) and . .in. cc: 5 cc , in ones m. 1 (“in would missthe thesubject point. Moreover, improper” which or first ... . . . H. 260 some later nontonic refrain-surfacings section also3 appears in keys to which a digres1 . . (momen. seem like impulsive local intrusions

sion may be made’’).%4 cot to pe , oo, C. P. E. Bach’s unusual rondos are anoma- ; a: _. . .

tary “‘visits’”’) inside predominantly episodic do-

. mains, all within a compositional style in which

lous, and they are not to be used as paradigmsalternatives . . - og. . . refrain material and quasi-episodic

of form. (As might be expected, they alsonot . . in . . athat . interpenetrate and are treated liberally, differ one from another in interesting ways.) . . . the customarily schematic way thatthat we associate By way of example, weIn might note only ... ,in, the with other rondos. sum, while refrain-mafirst of the thirteen rondos—C major, H. . . .it-_ . a,Liebhaber terial does appear in several differentidea, keys, 1s 260, Kenner und 2/i—the refrain . . anything but clear that all of the nontonic, often a simple eight-bar period,soundings recurs several times , , in ogy . . . incomplete are “refrains” any ap-

in the piece, often incompletely, sometimes dis- .

propriately structural sense.%

33. Rosen, Sonata Forms, 2nd ed., p. 126. an eight-bar period, is heard seven times—1in the keys of 34. A. F. C. Kollmann, An Essay on Practical Musical G, D, B, G, C, E-flat, and G.” Leisinger then proceeds Composition, p. 6 (chapter 1, section 12). Kollmann’s at- to recall Forkel’s 1778 “theory of the rondo” (Musikatitude, of course, is not necessarily that of censure: he lisch-kritische Bibliothek, vol. 2), which somewhat sternly could merely have been registering an irregular prac- advises composers that any episode should be related to tice. See also the discussion in Malcolm S. Cole, “Ron- and spring naturally from the rondo theme itself, often dos, Proper and Improper,” Music & Letters 51 (1970), in the manner of a paraphrase. As a result, “the fre388-99; and Fisher, “Further Thoughts,” p. 89 n. 14. quently read claim that in his rondos C. P. E. Bach made

35. Cf. the remarks of Ulrich Leisinger in the entry use of the refrain in keys other than the tonic . . . is to “Rondeau—Rondo” in Die Musik in Geschichte und Ge- be modified, if we assess the situation on the basis of genwart, 2nd ed., ed. Ludwig Finscher, Sachteil, vol. 8, Forkel’s theory, since the principal idea [Hauptgedanke] col. 554. According to Leisinger, “With the finale of [C. can also serve as the foundation for the episodes. There, P. E. Bach’s] Clavier Trio of 1775, Wq 90.2, the rondo is however, it is not perceived to have the function of a found in a new, already nearly ripened form for the first refrain.” See also Galand, “Form, Genre, and Style,” time [1n Bach’s works]. The principal theme [Hauptthema], p. 37.

404 Elements of Sonata Theory

The Sonata-Rondo (The Type 4 Sonata) (usually in a contrasting key), the piece is probably a rondeau, particularly if the refrain itself

is comparably If it is structured What. Qualifies as abrief. Sonata-Rondo? . .as. a more elaborate binary form, the piece is a rondo. But In the symmetrical three-couplet rondeau and if it begins as a recognizable expositional transithe symmetrical seven-part rondo (both sche- tion, then we are dealing with a sonata-rondo or matized as AB—AC-AB'-A), we noticed the Type 4 sonata. That TR typically leads to other principle of tonal resolution characteristic of | features characteristic of sonata expositions. In the sonata: the transposition to the tonic key — most cases, it will drive to a medial caesura, fol(as episode 3 or B') of material that was origi- lowed by a secondary-theme zone with EEC, nally stated in an off-tonic key (as episode 1 or — and perhaps a closing zone as well before iniB). This carries with it the implication that a _ tiating a retransition back to the refrain. It is V:PAC (or III:PAC, v:PAC, VI:PAC, and so on) also possible, especially in several of Haydn’s that closes episode 1 will eventually become a Type 4s, for the TR not to produce an MC at I:PAC (or i:PAC) to close episode 3. We also all but to carry outa TR=>FS process producing commented upon its rotational implications, a continuous exposition with EEC and possible

along with those of the three-couplet/sev- closing zone. Whatever its internal format, the en-part pattern (AB—AC-—AB'-A-coda), which exposition of a standard Type 4 sonata is never are particularly telling in the parallels between repeated. The absence of any repeat sign at this Rotation 1 and Rotation 3. Both the tonal and point 1s often the clearest “early” notational sigrotational aspects of these rondeau and rondo nal that the movement is a Type 4, not a Type 3. formats display an obvious kinship to the tonal Instead of a literal repeat, the exposition, ending and rhetorical features of sonata form. This sug- | with an obvious retransition back to an active gests that they may be understood as formal hy- dominant, will proceed ahead to the reappearbrids—as mixtures of the rondo (or rondeau) —_ ance, full or partial, of the rondo theme.°°

and the sonata. While this is relevant to any Especially with full-scale sonata-rondos, analyses of pieces organized by means of these the traditional letter designations (A, B, and structures, it may also be said that they do not — so forth) become inadequate and should be rise fully to the level of the “sonata-rondo” in dispensed with. In dealing with the thematic the strictest sense of the term. For that reason modules of the Type 4 sonata, it is more accuwe do not consider them to be sonata-rondos rate to use the terminology of the sonata, not

(Type 4 sonata forms). the rondo—with only a few adaptations. In a A piece or movement should not qualify as a ‘Type 4 exposition we consider the initial, closed (full-scale) sonata-rondo, or Type 4 sonata, un- = rondo theme not as “A” but as a P-refrain (P"), a less its first rotation is structured as the exposi- primary theme with “rondo-character,’ whose tion of a sonata (P TR’ S / C), and a later rota- simultaneous function also as a rondo refrain retion either recapitulates this expositional pattern mains intact throughout the piece. Following (the strong norm) or recomposes the pattern in — the I:PAC close of P"f we have the customary what may still be reasonably (and flexibly) con- _— sonata designators, TR, MC (or TR=>FS), S

sidered to be recapitulatory space (as often in (not “B” or “episode 1”), EEC, and C. the exceptional practice of later Haydn, as will Type 4 sonatas usually mix the rondo prinbe seen). To consider the matter in broad gener- ciple with the Type 3 sonata, but a less common alizations: if the material immediately following hybrid with the Type 1 design is also possible.

the closing PAC of the refrain’s opening state- Accordingly, we will jettison such traditional ment is structured as a simple period or hybrid terms as “seven-part sonata-rondo” and “five36. A few exceptions in Haydn’s works are noted in the section below on “Haydn’s Treatments of Type 4 Finales.”

Rondos and the Type 4 Sonata 405

part sonata-rondo” and regard these as Type 3 mal hybrid always ends with a partial fourth rosonata-rondo mixtures (Type 4°—or, more simply, tation that comprises either P alone or P + coda,

Type 4 without the superscripted numeral) and and may be referred to as the “coda rotation” Type 1 sonata-rondo mixtures (Type 4"). In addi- (“the rotation that contains the coda’). tion, the Type 1 sonata-rondo 1s often treated The normative ‘Type 3 sonata-rondo mixture to a special type of expansion, fairly often en- may be diagrammed as: countered in Mozart’s works, which we call the expanded Type 1 sonata-rondo mixture (Type Rotation 1: P** TR’ S/ CRT EEC on V:PAC or HI:PAC,

4l-exp), For most analysts, though, the term so- then Vy

nata-rondo immediately conjures up the m1x- Rotation 2: P:f development Ends V,

ture with the Type 3 sonata—so familiar from or episode RT many of Beethoven’s finales, where many of the Rotation 3: Pf TR’S/C=>RT ESC on I:PAC or i:PAC,

clearest examples (sometimes with small vari- then Vi,

ants) are to be found. We shall deal with that — Rettion 4 P+ optional coda Ends I:PAC format first.

This form exists on a continuum someThe Standard Type 4 Sonata (The ‘Type 3 where between the “pure” Type 3 sonata and

Sonata-Rondo Mixture) the pure symmetrical seven-part rondo. In

composing such a hybrid the composer may This hybrid typically begins with a light, emphasize either the sonata or the rondo aspect square-cut, and memorable “rondo-style” by structuring the post-P*f portion of Rotation opening theme, P*f, which 1s almost always har- 2 as either a development or an episode. A Type monically closed. It may be a simple period or 4 with a central development tilts more in the sentence, but in larger pieces it is often shaped direction of the most normative Type 3 proper as a rounded binary or simple binary structure, than does one with a closed, interior episode sometimes with internal repeats. In a Type 4 (although one should also recall that such an epsonata this “rondo-identifier” P*f proceeds di- isode may also occur within the developmental rectly into TR-space and thence further into the space of a Type 3 sonata). Here again, however, exposition, either at the onset of the next bar it is not the presence or absence of a developfollowing the concluding I:PAC of P™, or some- ment proper that makes the larger structure a times directly elided with that PAC, often with sonata-rondo (or Type 4 sonata) but rather the a sudden forte affirmation. The nonrepeated ex- expositional layout of Rotation 1 and the norpositional rotation concludes with either an RT matively symmetrical tonal resolution of Rota-

ora C>RT passage that reactivates the toni- tion 3. cized dominant (or, 1n the case of a minor-mode The Allegro grazioso finale of Mozart’s Pipiece, that leads from the tonicized III or v to ano Sonata in B-flat, K. 333, provides a clear exan active dominant). ‘The second (developmen- ample of the Type 4°—although, as is common, tal-space) rotation begins with a tonic statement one that also displays a few unusual features. In of P (full or partial) and then proceeds either _ this case P'fis structured not in a more expanto a development of the expositional material sive binary format but only as a repeated parallel or to an episode; this episode may be a closed period (mm. 1—8) whose forte restatement (mm. binary form similar to the typical episode 2 (C) 9-16) might suggest the orchestral tutti repetiof a symmetrical seven-part rondo. The devel- tion commonly heard at this point in some of opmental rotation ends with a retransition and his concerto finales. This is followed by a moddominant preparation. The third (recapitula- ulating sentence with an obvious TR-function tory) rotation also concludes with a retransition (mm. 17—24)—a central marker of a Type 4 soand dominant preparation that leads to a fourth nata (as opposed to a “pure” rondo). This TR tonic statement of all or part of P; this P may or leads to a rhetorically weak V:HC medial caemay not be followed by a coda. Hence this for- sura in m. 24. An S-theme in the dominant fol-

406 Elements of Sonata Theory

lows, a sentence with a varied repetition of its 1—an element that helps to predict the “inevicontinuation=>cadential module (upbeat to m. table” return of P"f. The modulatory aspect of 25-36; the recomposed repetition is motivated Rotation 1’s entire post-P" material, as well as by an evaded cadence in m. 32). The second- its harmonic incompleteness, suggests its larger ary theme’s V:PAC EEC (m. 36) 1s elided with role as that of filling out a developmental space a five-bar RT that ends on an active V’ chord. within the Type 4 sonata. Rotation 2 as a whole

Thus Rotation 1 follows the P TR ’S/ RT moves I-vi,-IV,-V,. pattern. There is no closing zone. What we Beginning in m. 112, Rotation 3 follows have heard so far (perhaps with the adjustment a typical recapitulatory path. Pf returns unof the RT into a more normative C-zone) could changed, while TR is recomposed (and exequally well have been the exposition of a Type panded) in order to close with a I: HC MC (m. 3 sonata. It is only with the absence of an expo- 147), one whose rhetorical features are much sitional repeat, the presence of an RT, and the stronger than those of the relatively unassumtonic-key return of P'fas the onset of the next ing V:HC MC of the exposition. S 1s transposed rotation that the Type 4 status of this movement to the tonic (upbeat to m. 148), and its repeated

becomes clear. continuation is dramatically expanded to make Rotation 2 begins with a literal restatement the arrival at the I:PAC ESC especially strong of the grazioso P*f (mm. 41-56), followed by the (m. 163). The RT passage from Rotation 1 1s

same presentation module that had initiated also expanded and leads to the most surprisTR (upbeat to m. 57). This time the TR-based ing feature of this movement—and certainly to passage leads via an augmented-sixth chord to something exceptional within Type 4 norms. the dominant of G minor (V/vi, m. 64). A new This RT, most unusually, leads to a cadential ¢ theme follows, initiating a central episode. This chord sustained by a fermata (m. 170), and Mois an eight-bar sentence (upbeat to m. 65-72) in zart follows it with a fully composed “Cadenza G minor (v1) that ends on another half cadence in tempo” (upbeat to m. 171-98) that ends on in that key (V/vi, m. 72). A brief, chromatic an active dominant (V’, m. 198). This cadenza transitional link (mm. 72-75) leads unexpect- setup and interpolation suggests yet another hyedly to the key of E-flat (IV) and to the “sec- brid feature of this movement—a mixture with ond stage” of this central episode. (It sometimes at least this element of the Type 5 sonata (conhappens that interior episodes are subdivided certo first movement), outlined in chapters 19into two or more stages—unseparated by any 22. (One might also recall the “concerto-like” sounding of P'—each giving the impression of _forte restatement of the initial Pf period at the an individualized “episodic” theme and tonal- opening, mm. 9-16. In this case we may say ity.) We are now given another new theme be- that while Mozart mixed rondo norms here prisinning in m. 76, a sentence whose attempts to marily with Type 3 format-structures, at one cadence are continually frustrated. As a result, or two points, and especially toward the end, this second stage of the central episode 1s left he alluded in a separate, ad hoc way, to Type 5 harmonically unclosed. Just at the point of its practice as well.) potential closure, though (an explicit cadential Following the non-normative cadenza, Roformula in E-flat, IV, in m. 89), its completion tation 4, the coda rotation, starts with a single is starkly undercut by a return of P* based ma- statement of Ptf whose consequent phrase is terial on C minor (m. 90)—entering incisively ereatly expanded through evaded cadences and to block the predicted episodic cadence in IV subsequent extensions. (Even its “definitive” with a chill and also to initiate a Ptf-based re- trill cadence, mm. 212-13, is undermined transition, one that also recalls the “false-start” by 3 in the bass at the downbeat of m. 213.) procedure (“wrong-key” attempts to start the The “struggling” consequent 1s followed by recapitulatory rotation) outlined in chapter 12. a Ptf-“fade-out” coda proper (mm. 214-24), This RT leads ultimately to a prolonged domi- which ends, finally, with a conclusive, forte canant preparation (mm. 102-10) that incorpo- dence. rates RT-material from the end of Rotation The finales to several of Beethoven’s piano

Rondos and the Type 4 Sonata 407

sonatas also provide instructive examples of the as S (m. 25) 1s abrupt and deformational—no standard Type 4 sonata, though usually—and effective MC 1s articulated—and, famously, the most typically—with individual quirks that are movement’s final gesture once again features a of special interest. The otherwise paradigmatic by-now-typically Beethovenian “wrong-key Sonata in C, op. 2 no. 3/iv, for instance (with feint” of the opening bars of P*fin VI, the key a closed, F-major central episode), is marked of the central episode (and the slow movement), by a dramatically “failed” exposition and non- before the aggressive C-minor tonic correction

resolving recapitulation (no EEC or ESC in (“no escape!”) in the final bars.9/ sonata-space; in both spots the repetition of S In the hands of Haydn, Mozart, Beethoven collapses into minor and proves unable to reca- and others, both the simple rondo and the Type dence); and by witty, “wrong-key” references 4 sonata-rondo were invitations to high-formal

(in A major, m. 298, then A minor, m. 302, extravagance and wit—a propensity for struceach halted with tentative pauses of reflection) tural play congruent with the “popular,” often

to Pf at nearly the end of the coda-rotation. insouciant simplicity of the refrains that proThe Sonata in E-flat, op. 7/iv (again with closed pelled them forward. “Surprising” deviations central episode, here in C minor) introduces the from an implied heuristic norm were part of the coda-rotation’s P™ on the “wrong” tonal level, game. E-major (a half-step too high), m. 157—as if

the tonal processes have momentarily slipped .

off-track—before correcting the key x mn. {6 The Type 1 Sonata-Rondo Mixture (Type 4)) and subsequently appending a tonic coda proper, This hybrid, Type 4', merges the rondo principle m. 166, that also looks back to the central ep- and the Type 1 sonata. As discussed in chapter isode. The Sonata in D, op. 10 no. 3/1v—an 16, the Type 1 sonata, lacking a development, is edgy, nervous movement throughout—provides a double-rotational structure—an expositional a disturbingly compressed exposition (with S at rotation followed by a recapitulatory rotation. m. 17); no EEC; a tonally shifting, rapidly dis- An optional coda begun with Pf may suggest a solving central episode that starts in bVI (m. 33); perhaps partial third rotation. When the Type 1 a “wrong-key” attempt to restart the recapitu- sonata 1s blended with the rondo, both the exlatory rotation (m. 46, returning to bVI), with positional and the recapitulatory rotations end tonal correction at m. 56; and a radical defor- with a retransition (or C=>RT) and dominant mation of what one expects to the be recapitula- preparation for a tonic return of Pf. After the tory rotation, which, in its fidgety tension, finds recapitulation is completed, its subsequent RT itself “unable” to reprise the original S at all, leads once again to a third tonic statement of much less in the tonic, and instead strays off into Pf —or to a slightly varied allusion to it. This differing tonal areas, failing also in the process refrain may or may not be followed by a coda to secure an ESC. The Sonata in C Minor, op. proper. Hence this formal hybrid always ends 13/111, is more normative in its overall effect, but with at least the onset of a third rotation (the its preparation for what must surely be regarded coda rotation) that comprises either Pf alone or

37. The final moments of op. 13/111 are discussed, in Type 4 procedure in op. 10 no. 3/iv, this piece provides the context of Edward T. Cone’s misleading “sonata a ““wrong-key” start to what listeners might at first take principle” claims about them, in Hepokoski, “Beyond to be a recapitulatory rotation (Pin A-flat, m. 76), folthe Sonata Principle,” Journal of the American Musicologi- lowed by a dissolution of that theme and its subsequent cal Society 55 (2002), pp. 115-18. Beethoven’s penchant correction into the “right key,” the tonic (m. 92), anda

for witty, “wrong-key” entrances of Pf does not undo “failed” recapitulatory remainder that is unable to pro-

the norm that P'f is to make its reappearances in the duce the original S idea at all. Thus while the piece tonic. On the contrary, it is precisely on the strength does display an exposition—the most crucial marker of of that norm that the effect of Beethoven’s transgres- a Type 4 sonata—it does not provide a full recapitulasions of it relies. Another instance from the period in tion. See also the discussion above of C. P. E. Bach’s question, among several, occurs in his C-major Rondo keyboard rondos with seemingly modulating refrains. for Piano, op. 51 no. 1. Not unlike the deformational

408 Elements of Sonata Theory

Prf + coda. The Type 1 sonata-rondo mixture the presence of the RT setup for the final ap-

may be laid out as follows: pearance of that P¥.

Examples of this Type 1 sonata-rondo mixRotation 1: Pf TR’S/C=>RT ~~ EEC on V:PAC or III:PAC, ture may be found in the finale of Mozart’s

Rotation 2: PiF'TR* S ) Cos ae o ePAC Quartet in E-flat, K. 428, in the slow movement

then V, ) of Piano Sonata in D, K. 311 (though with very

Rotation 3: P*f (+ coda) Ends :PAC short, one-bar RTs), and in the slow movement

of his Symphony No. 39 in E-flat, K. 543. In K.

Until we arrive at the end of the second rota- 543/11 (Andante con moto) the opening theme tion, it may be difficult to distinguish a Type 4! (which we eventually realize is P') is structured from a “pure” Type 1, since many Type 1 ex- as a 27-bar rounded binary with internal repeats positions also end with an RT ora C>RT in in the tonic key of A-flat. Because the primaryorder to prepare the recapitulation. In addition, theme zone of a sonata exposition does not normany Type 1 sonatas conclude with a P-based mally contain double bars and repeat signs, the coda. The distinguishing feature of the Type 1 listener might well anticipate a more clearly secsonata-rondo mixture, Type 4", is the extensive tional form, perhaps something on the order of a retransition that also concludes Rotation 2, as compound ternary or a five part rondo. Followwell as the relatively literal restatement of P that ing this theme, however, a two-measure link follows. Thus the retransition that leads from (mm. 28-29) bursts open into a forte, nine-bar the recapitulation back into a relatively intact Sturm und Drang TR (m. 30)—the first of many Pf for the beginning of the third refrain-state- dramatic contrasts in this movement. This TR ment and coda rotation is a crucial indicator of — begins off-tonic in F minor (vi)38 and leads to

this format. (This distinction is also discussed a first-level default medial caesura (V:HC, m. in chapter 16.) The composer may additionally 38): at this point, the sonata aspect of the deemphasize the rondo aspect of this structure by sign comes to the fore. The MC is followed by fashioning P* as a “refrain-like” rounded bi- a trimodular S that may also be construed as nary form with internal repeats. Because most —a sentence. We first hear a triple presentation Type 4 sonatas exist on a continuum, situated of the “b” idea from Pf that “holds open” the somewhere between the “pure” sonata and the dominant harmony of the MC, in the manner “pure” rondo, it is occasionally difficult to de- of an S° (or S!°) module (mm. 39-40, 41-42, termine whether the sonata or the rondo ele- 43-45), giving the impression, perhaps, that ments predominate 1n a given piece. Sometimes the movement’s aesthetic protagonist—marked the interplay between the sonata and the rondo by his or her own thematic module—is reactconventions forms the expressive core of the ing to the storm that was so violently and unpiece. Insisting that hard cases must be decided expectedly unleashed in the preceding TR.°? one way or the other might miss the point. ‘The This 1s succeeded by the forte unleashing of a (nonexpanded) Type 4! is an infrequently en- four-bar continuation (mm. 46—49) featuring countered form. The Type 1 with P-as-coda, fragmentation and a descending fifth sequence; easily mistaken for the Type 4!,seemstocropup and finally by yet another quieter reaction, pimore often. Generally considered, one should ano, a five-bar cadential unit (mm. 50—54) that not give the nod to the Type 4! category unless brings about the V:PAC EEC. This cadence at two factors are also present: a “rondo-block” —_m. 54 elides with the onset of a CS>RT zone. A structure, or at least character, to the P theme (a seven-bar sentence, at first suggesting an obviconvincingly “rondo-like” P'f) and, especially, | ous drone-pastoral mood (mm. 54-57, the calm

38. TRs that begin with a sudden plunge into vi, typi- On that reading the CF flowers into a set of differing cally following a P-theme that overdetermines the modules, and no S-theme proper is ever fully launched.

tonic, are discussed in chapter 6. From this perspective what we have is one type of con39. An alternative interpretation might regard the bars tinuous exposition, “unable” to produce a genuine S. following the MC-effect as “place-holding” caesura-fill.

Rondos and the Type 4 Sonata 409

after the storm?), reinforces the V:PAC with a parallel period, as if to compensate for the tonal forte, cadential module (mm. 58—60; a spontane- disruption in Rotation 2. ous outpouring of heartfelt gratitude?), which

itself clides with a slightly varied repeat of the The Expanded ‘Type 1 Sonata-Rondo presentation-drone (m. 60). This time the re- Mixture (Type 41-RT, which this time first tonicizes (mm. It should strike one as something beyond what 137—42), then reactivates the dominant. Rota- one might normatively expect at this point.4? We tion 3 (m. 144) restates only the first part of P™ — indicate this important, expanded variant of the (the a section of the rounded binary), greatly Type 4! sonata form—found with some frequency expanding the consequent phrase of the original in Mozart’s works—as ‘Type 4!-«p.4!

40. In the second movement of Mozart’s Symphony example is the Allegro finale (marked “rondeau’’) of No. 39 in E-flat, K. 543, for example, analyzed above, Mozart’s Piano Sonata in B-flat, K. 281. In this piece, the Sturm und Drang transition is expanded from nine the post-P"f portion of Rotation 2 comprises a continuto thirteen bars. As a reaction to the preceding tonal ous simple binary in G minor (vi, mm. 52-67) followed lurch, this expansion is certainly noteworthy, but is by a brief retransition (mm. 68—70). Rotation 3 begins insufficient to lift the piece out of the category of the with a restatement of P'f(m. 72) followed by a sudden “pure” Type 4! sonata-rondo. The same is true of the inflection toward E-flat (IV) and a new, closed theme finale (Allegro vivace) of Mozart’s Quartet in E-flat, in that key (mm. 90-101). A lengthy retransition (mm. K. 428. Here TR is expanded from twenty-five bars in 102—23—1ncluding an internal reference to P™f)—leads Rotation 1 to thirty-three bars in Rotation 2. Because to an extended dominant preparation that seems to prethis sort of expanded recomposition may also be en- pare yet another return of P*f, but instead leads to the countered in the recapitulation of any sonata type, it is delayed tonal resolution (S in the tonic key, m. 124). Afinsufficient to warrant a consideration of the movement ter the ESC (m. 136), a retransition prepares Rotation 4,

under the norms of a separate category. which is limited to a final return of P'f (m. 143). (One 41. One might wonder whether the Type 3 sonata- should not misinterpret this finale as a “nine-part sorondo mixture (Type 4%) may be treated to a similar nata-rondo” that lacks a fourth return of A.) Another sort of expansion—that is, whether its third (recapitu- example may be found in the finale of Violin Concerto latory) rotation may feature an interpolated develop- No. 3 in G, K. 216. ment or episode. Although rare, instances do exist. One

410 Elements of Sonata Theory

For inexperienced analysts the Type 4!-exp with the second tonic statement of Pf (the onset sonata contains a number of traps into which it of Rotation 2). This is readily perceived if one is all too easy to fall. The most tempting ana- examines the music on either side of the indilytical missteps, though, are readily avoided if — vidualized episode or developmental expansion, one approaches the structure with a firm grasp looking in particular for evidence of the linear, of the rotational practice guiding the large-scale rotational ordering of the modules first preevents. To illustrate (here we adapt an argument sented in the exposition. Disregarding the epiput forth by John Daverio):*# 1n one typical sce- sode itself (assuming here the most challenging nario, an episodic or developmental “billow- case of a “new,” nontonic episode that functions ing-out” can occur when the tonic return of P™, as a self-contained interpolation), the question launching Rotation 2 and the recapitulation, is becomes: is the rotational ordering otherwise followed by either a closed episode or a genu- preserved throughout? If so—and if the other ine development (often of Pf or TR material) Type 4 “rondo” signals are also in play (such as or by both. This episodic and/or developmen- the RT functions and the third statement of P*f tal expansion has frequently been mistaken for to launch the coda rotation)—then one 1s con“Episode 2 (C)” of a seven-part sonata-rondo fronting a Type 4!-«xP sonata. in which the third statement of the refrain (A) The central thing to observe—the main feais eliminated. In other words, some analysts ture that undermines the erroneous ABACB'A have parsed this familiarly Mozartian pattern as view of this form, with its faulty presumption of ABACB'A, suggesting that it arises as an “in- a missing rondo element—1s that the recapitulacomplete” ABACABIA design.*? Once again we tory S is often prepared in the same manner (by see the pitfalls of reducing the Type 4 sonata to the same MC) as it had been in the exposition. a mere string of alphabetic symbols. Type 4'-«P, The Rotation 2 passage of episodic or developthe expanded Type 1 sonata-rondo mixture, mental expansion does not normally bring us may more meaningfully be conceived as a ro- to a standard RT that would lead us to expect

tational structure: a tonic return of Pf. On the contrary, the mu-

| | sic from the TR-crux-point onward indicates

Rotation Tr PETRUS (CST pac. then, that the next anticipated rotational event will Rotation 2: P*f (TR) [development or ESC on I:PAC or i:PAC, be not Px! but the tonic return of S. Once this is

episode]...TR? S / CORT then V, erasped, any sense of a supposedly “missing” Prt Rotation 3: Pt (+ coda) Ends LPAC disappears. The attentive and informed listener should not expect the return of P*f until after It is essential to notice that in a Type 4!-¢xP so- the RT that concludes Rotation 2.44

nata form the Rotation 2 interpolation, even One of the most common places to find if it is a new and separate episode, frequently this structure 1s in the rondo-finales of several links up at its end with the end of the original (though not all) of Mozart’s concertos, where TR, now transposed to the tonic key. The Ro- an additional feature of “ritornello” or “tuttitation 2 expansion often leads to a crux-point solo” contrast is also thrown into the mix. that slips onto correspondence measures in Mozart’s concerto adaptations of this and the TR-space. This means that it rejoins an ongo- Type 4 structure are discussed toward the end ing rotation-in-progress, one that had begun of the present chapter. Both straightforward

42. Daverio, “From ‘Concertante Rondo’ to ‘Lyric So- 44. Daverio, “From ‘Concertante Rondo’ to ‘Lyric Sonata’: A Commentary on Brahms’s Reception of Mo- nata,’” pp. 115-17, also made this point, taking to task zart,’ Brahms Studies, vol. 1, ed. David Brodbeck (Lin- Malcolm Cole and Charles Rosen in the process. The coln: University of Nebraska Press, 1994), pp. 115-17. latter had made a bad situation worse by referring to ex43. As stillin Cole’s revised “Rondo” entry in The New amples of this form as (in Daverio’s words) “sonata-ronGrove, 2nd ed., vol. 21, p. 653. See, also, e.g., n. 60 dos with reverse recapitulations.” (See Rosen, Sonata

below. Forms, rev. ed., pp. 123-32.)

Rondos and the Type 4 Sonata 411

and adapted versions of the Type 4!-«*P are also ing correspondence measures, by the true tonal found, however, in several finales of some of his resolution (S in the tonic). As expected, the renon-concerto works: it became a favored format capitulation is followed by RT (m. 164) leading for the mature Mozart sonata-rondo finale. It oc- to a return of Pf (cello, m. 182) and an ensuing curs, for instance, in Piano Sonatas in C, K. 309/ coda. In this case Rotation 2, the recapitulatory 11, in C Minor, K. 457/11, and in D, K. 576/10; rotation—with its extra, “wrong-key” expanin the Sonata in D for Two Pianos, K. 448/111; sion—might be diagrammed as:

in the Piano Quartet in G Minor, K. 478/iv; in the String Quintets in C, K. 515/iv, and in Pf TR [——(’) S——JendofTR ’S/ C RT G Minor, K. 516/iv; in the Quartet in D, K. 575/iv; and in several other last movements. The interpolation in K. 575/i11 is clearly of Particularly in non-Mozartian finales, however the developmental type—a feature that tilts the (and especially in later works, such as those of | Sonata-rondo hybrid more in the direction of its Schubert and Brahms), the analyst should take “sonata” elements. By contrast, the interpolacare not to confuse the Type 4!-exP with the ex- tion in the recapitulatory rotation of the Molto panded Type 1 proper, (the nonrondo version allegro finale of the Sonata in D for Two Pianos, of the form, discussed in chapter 16), whatever K. 448, is an obvious subdominant episode surtheir apparent similarities. As with the Type 4!, rounded by “referential” TR material on both the rondo-identifiers of Type 4!-exP (apart from sides. Rotation 2 begins with a nonrepeated the “suggestive” rondo-character of the theme) Prfin the tonic at m. 139. Its I:PAC (m. 154) are the presence of the RTs leading back to P*, leads to a five-bar transitional link based on and the sounding of P as a satisfactory ron- the exposition’s opening TR! !-module (mm.

do-refrain at the outset of a coda rotation. 16-20) but this time tipping toward the subIf we look at the Type 4!-exP Allegretto finale dominant. This triggers the central episode in of Mozart’s Quartet in D, K. 575, for example, G major, IV, a closed rounded-binary structure we find a nonrepeated exposition (normally a | With written-out repeats (mm. 159-206). Afsonata-rondo signifier) that features a P-based S ter the episode’s [V:PAC close the music pulls (m. 32) and a contrasting C (upbeat to m. 59).46 efficiently toward a crux-point that resuscitates A retransition follows, mm. 67-71. In Rotation modules from the exposition’s TR (m. 215 = m. 2, the start of the recapitulation, P*f (m. 72) is fol- 26), now sounded at the tonic-key pitch level lowed first by a reworked version of TR (m. 91) and leading toward the corresponding MC in that leads to a “false MC” on V of F major (411 m. 229 (= m. 40). Considered rotationally, what of D, m. 104), then by a “wrong-key,” F-major we have 1s an ongoing T'R-space that is momenstatement of the beginning of S (m. 105). This tarily “stopped” by the interpolated episode (or is evidently to be understood as a witty anomaly — Whose central modules are written over by it). that needs correction. And that is precisely what | Nonetheless, the rotational implications of the happens. Mozart soon leads the music into the larger P—T'R succession could hardly be clearer. point of TR-crux (m. 124 = m. 28, varied) to K. 448/111 is as a locus classicus of the Type 4!-exp

the “correct” MC on V of D (m. 127 = m. 31, with an internal-episode expansion in the recanow in the tonic) and follows this, in continu- _ pitulatory space.

45. The list of Mozartian works in this and Type 4 not to be regarded as a Type 4!-©xP sonata-rondo: it is structure provided in Daverio, “From ‘Concertante better understood as a merely sectional, less developed Rondo’ to ‘Lyric Sonata,’” p. 114, is more complete— “rondeau” forerunner of the Type 4! proper. although not every movement on this list rises to the 46. Particularly in his later works, Mozart may have category of a Type 4!-exp. We regard the Quartet, K. considered the P-basis of S in some of his finales—sug428/iv, for example, as merely a Type 4!; and the Po- gesting recurrence—to be a feature of the “rondo-charlonaise en rondeau slow-movement center of Piano Sonata acter” of the whole, though by no means is the P-based

in D, K. 284, lacks an exposition proper and is hence S a unique feature of his Type 4 movements.

412 Elements of Sonata Theory

Before leaving this topic we might provide dissolving instead into an “early” RT around an indication of how understanding a back- mm. 178-79. Shortly thereafter, and in advance sround heuristic model helps one to grapple of the C-modules that would close out the rotawith hard cases. The Allegretto grazioso finale tion proper, Mozart (prematurely!) brings back of Mozart’s Piano Sonata in C, K. 309, is an Prf(m. 189), as if beginning a coda rotation. intricate Type 4!-°xP sonata (which the com- Strictly considered, this would be an instance of poser labeled a “rondeau”’) with extraordinary coda-rhetoric interpolation (CRI, Chapter 13 recapitulatory complications, among which the above)—coda-like behavior, including the sugcentral subdominant episode is by no means the gestion of a rotational restart—wedged into the most provocative. Following the recapitulation’s concluding sections of a recapitulatory rotation opening, Pf (mm. 93-111), we plunge into a before it has completed itself. This is therefore brief passage of S material (S!4, mm. 111-15, a premature interpolation of Rotation 3 behavbased on figures from mm. 58—62—the first of ior into the still-unfinished Rotation 2, as if P*f a series of unexpected, “nonrotational” hap- were appearing punctually “on schedule,” blisspenings) that serves as a transition to F major fully unaware that no ESC and C-theme had (IV). At this point, the composer stitched in a been sounded.*” With the I:PAC that concludes closed episode in F, a repeated parallel period this Pf (m. 204), a version of the only apparently (mm. 116-31). Moving out of the episode, he “missing” S!-2 tumbles in (m. 204), eventually provided a retransition (mm. 131-42) that pre- setting up a “corrective” return to the recapitpares for the tonal resolution (S in the tonic, ulatory rotation and the return of the original which begins at m. 143). In this case, unusually, C-idea (m. 214), though in a context in which the RT passage is not based on anything from the ESC has yet to be sounded. Within this conthe exposition’s TR—which 1s suppressed alto- text of modular dislocation—comparable to the eether—but rather, from m. 137 onward, on S!2 head-spinning confusions of identity and order (!), thus not producing the usual TR-crux-point typical of opera buffa, the giddy defiance of sober and ensuing correspondence measures. This is a but artificial convention*®—C gives way sudcurious moment. The seeming disorder of the denly to S!4 (m. 221), which, this time, does referential modules on either side of the episode manage to bring the much-abused Rotation 2 to does not suggest a rotational pursuit. The impli- the ESC (m. 236). A coda proper follows, based cation of a rotational structure (and a Type 4!-exp on TR (the P' ticket having been spent premadesign) is revived only with the return to S at turely as an interpolation into Rotation 2), anda m. 143—as if one were grasping ata principle of final view of a piano, “expiring” P* incipit drags

order to pull together the whole. its way across the finish line in the concluding Even though we might feel ourselves “safely” bars. In sum, the deformational recapitulatory arrived at S (where a succession of correspon- rotation of K. 309/111, Rotation 2, can be repdence measures is the norm), the surprising resented as follows (with the bold-printed letters events continue to unpin our sense of security. suggesting the rotationally governing features): The just-heard S!-2 1s omitted in favor of a third, varied repetition of S!-', for instance (mm. 151- Prf (S!4—— (’) Episode S'2——] (?) S14 $13) St4=> RT! (’) 56), and S!4 fails to produce the expected ESC, [PE St2I]TC) CC [StL]

47. From a different perspective, Rotation 2 (the reca- has come before.” He also interpreted these “terminal pitulatory rotation) and Rotation 3 (the coda rotation) references” as a “summarizing coda.” It is uncertain

are composed as overlapping at this point. whether Daverio’s claim that such scattering 1s typical 48. Daverio, “From ‘Concertante Rondo’ to ‘Lyric So- of Mozart’s Type 4!-exP sonata-rondo movements can be nata,”’ p. 120, referred to cases in which “material from fully supported. The more important point would be to

the first group is redistributed or ‘scattered’ through- come to terms with each individual instance from the out the responsive portion of the design,” noting that perspective of rotation theory, standard coda practices, “Mozart’s amplified binary movements feature terminal coda-rhetoric interpolation, and so on. references to the whole of the principal material that

Rondos and the Type 4 Sonata 413

Haydn’s Treatments of Type 4 Finales listeners, the same quality makes him a difficult composer to use as a frequent source for paraHaydn adapted the Type 4 sonata with remark- digms on form. The more or less standard forable freedom. In itself this 1s hardly surprising, mal options used by his contemporaries seem evsince he, more than any other composer of the er-present in Haydn, but more often principally period, sought a pervasive originality of content as benchmarks or background concepts that he and design in his works, as though he were re- persistently tweaks, overrides, and alters on the melting at each compositional moment the crys- acoustic surface of the music (what one actually tallizing forms and procedures that had come hears). While this 1s certainly true of his Type to be normative, even schematic, in the hands 3 sonatas—with their often-thoroughly recomof others into a persistent volatility of instan- posed recapitulatory spaces, for example—it taneousness, an unpredictable malleability that is even more the case with his Type 4, sonaoften eludes a clean capture by the standard, ta-rondo finales. Here the “rondo-invitation”

heuristic formal categories. However “sim- to high-spirited play in rapid tempos spurred ply” Haydn might begin a movement, each of | Haydn to compose even more exuberant transits subsequent moments—once past the initial formations of the formal options at hand. As has idea—bursts with an energy of ongoing inven- been recognized by all who have attempted the tion, a spontaneous sense of “generation on the task, those who approach most of these Type 4 spot” that can skew the compositional pathway finales with only the standard schemata in hand into unforeseen, sometimes asymmetrical direc- (much less with only the inappropriately reductions. One of the central paradoxes of Haydn’s tive letter-scheme, ABACAB'A) will find themstyle is the vast gulf that separates the seeming selves challenged at nearly every turn, particusimplicity of his thematic materials (the “naive” larly as one gets past the “expositional” portion or “problem-free” manner in which they are of- of the form. The result has been a tangle of muten first stated) from the densely complex, diz- sicological definitions and debates about what zyingly vitalistic treatment to which they are ought to count as a bona-fide sonata-rondo in almost immediately put.4? Ultimately, this as- Haydn.°? pect may point to a foundational incongruity at In confronting a fast-tempo Haydn finale that the heart of the Haydn style—one that drives to is not a variation movement, one might expect the core of his persona as a composer—but it is to find that its main lines will be dominated by surely also a feature of high connoisseurship that one of three formal options: the Type 3 sonata the composer invited his most adequate lsten- (the “standard” sonata form also characteristic

ers to relish. of his first movements); the rondo (or rondeau); While Haydn’s high-pressure, bar-to-bar or—with increasing frequency only from the originality is a source of delight for attentive late 1770s onward—the Type 4 sonata (the so-

49. On Haydn and our metaphor of the vitalistic musi- “Rondos, Proper and Improper” (n. 34). For Fisher, see cal particle, constantly in pursuit of growth and trans- his “Sonata Procedures in Haydn’s Symphonic Rondo formation, see the section of chapter 11 with subtitle, Finales of the 1770s,” Haydn Studies: Proceedings of the “Recompositions, Reorderings, Interpolations.” International Haydn Conference Washington, D.C., 1975, 50. Readers wishing to familiarize themselves with ed. Jens Peter Larsen, Howard Serwer, and James Websome of the musicological discussions and conflicting ster (New York: Norton, 1981), pp. 481-87; and his interpretations regarding this uncommonly complicated “Further Thoughts on Haydn’s Symphonic Rondo Fitopic might begin by reviewing some central articles in nales” (n. 19). In this debate we prefer Fisher over Cole, the exchanges between Malcolm S. Cole and Stephen even though our own approach to these finales differs C. Fisher in the 1970s, 1980s, and 1990s. Both writ- from his in some central features. Cf. also the extended ers presented a wealth of historical and analytical data, studies of Bernhard Moosbauer, Tonart und Form in den seeking to find defensible paths through Haydn’s ev- Finali der Sinfonien von Joseph Haydn zwischen 1766 und er-original finales. Their analyses, including a reliance 1774 (Tutzing: Hans Schneider, 1998); and Reiner Leison (schematic-letter) systems that we have abandoned, ter, Das Finale in der Sinfonik Joseph Haydns (Stuttgart: are sometimes different from ours. See, e.g., Cole, ibidem-Verlag, 1999). Each of these studies leads one to “Haydn’s Symphonic Rondo Finales” (n. 23); and his further bibliography on the topic.

414 Elements of Sonata Theory

nata-rondo mixture). In fact, though, many of in the style of its P-theme.°+ Only two of the Haydn’s finales from the 1780s and 1790s are finales from the twelve “London Symphonies” ingenious hybrids among these options.°! Under have this repeat sign (No. 98 in B-flat and No. such circumstances any simple classification 1s 104 in D); as do two of the three quartets from insufficient. A number of the finales—includ- op. 74 (No. 1 in C and No. 3 in G Minor), three ing several in his most celebrated works—wit- of the six quartets from op. 76 (No. 1 in G, No. tily present conflicting generic signals in dif- 3 in C, and No. 6 in E-flat), and both of the ferent portions of their structures, as if musical quartets from op. 77. processes that begin within one category sud- When a sonata-exposition 1s present (as opdenly shift (or “change their mind’) to allude posed to the merely sectionalized portions of to procedures in another. Movements that begin a “pure,” transitionless rondo, as in the finales as fairly clear Type 4s can change midstream to of the Quartet in B-flat, op. 76 no. 4, and the take on more telling characteristics of either the Symphony No. 96 in D) and an expositional rerondo proper or the Type 3 sonata. Conversely, peat sign is lacking, this should be regarded as a even those finales that are most clearly Type 3s signal that the governing principle of the moveoften have a strong rondo or Type 4 charac- ment is at least initially announced to be that of ter. A central aspect of Haydn’s game-like ap- the Type 4 sonata-rondo. The Type 4 impresproach to the finales was to ride the dividing sion 1s typically reinforced by the light or playful, lines among the different formal options as the contredanse-character of P* along with Haydn’s movement proceeds, now tilting this way, now characteristic structuring of it as a binary or that—ultimately to produce individualized and rounded-binary structure, often with notated ingenious syntheses among the “purer” (heuris- internal repeats. (In Haydn’s finales, the prestic) options.°? The task of the analyst is to expli- ence of such notated repeats within a binary or cate the connotations of Haydn’s compositional rounded-binary Pf is invariably a Type 4 signal, choices as they unfold in time, not to provide although it is one that 1s sometimes encountered the end product with a reductive, monodimen- in other composers in works that turn out to be

sional classification. Type 3s.)°> Another important Type 4 marker is

When considering any Haydn finale from the presence of a retransition (RT) at the end of about 1773 onward,°? one should initially ex- the nonrepeated exposition along with the subamine the procedures of the first third of the sequent tonic restatement of at least the openmovement, those through its first rotation. If | ing module of P™ as a refrain-identifer. This last the movement contains a sonata-exposition signal, however, 1s sometimes lacking, as will be (two-part or continuous) that includes a no- outlined below. In any event, if all of these signals tated, full repeat—the most decisive generic are in place, the predominantly Type 4 status of marker—that movement is to be regarded as the movement is assured, regardless of the potengoverned primarily by the norms of a Type 3 tially unorthodox treatment of what might folsonata, whatever other rondo or Type 4 ten- low, especially in the space that we anticipate will dencies it might display elsewhere, for instance be occupied by a recomposed recapitulation.

51. As Fisher put it in “Further Thoughts,” p. 101, and rondo elements in a symphonic finale no later than “Haydn is exploring a spectrum of possibilities that does 1773.”

not take into account the conventional [present-day] 54. Here again we agree with Fisher, “Further distinction between the rondo and the sonata-rondo.” Thoughts,” e.g., pp. 95, 96, 97, and 100. Cf. p. 86: “The Also quoted in Cole, “Rondo,” The New Grove, 2nd position of repetition marks is always an important indi-

ed., p. 653. cation of Haydn’s thinking.”

52. Cf. Fisher, “Further Thoughts,” p. 86: “Haydn is 55. See, e.g., chapter 5 with regard to Type 3 P-themes fond of hybrids incorporating elements of one design in binary or rounded-binary formats, a few of which into movements that have the overall structure of an- contain internally notated repeats: the finales of Mo-

other.” zart’s String Quintet in D, K. 593, and Schubert’s Sym-

53. Fisher, “Further Thoughts,” p. 85: “Haydn actually phony No. 5 in B-flat. began experimenting with the combination of sonata

Rondos and the Type 4 Sonata 415

In sum: Haydn’s Type 4 movements are to be tions that produce the effect, unfolding through identified as such primarily by their binary or specific events in time, of a thoroughgoing forrounded-binary themes coupled with the pres- mal hybridity that eludes simple classifications.

ence of a nonrepeated exposition and the 1m- Among the complicating procedures that one mediate tonic return of at least the beginning can find in Haydn are the following. of Prf.56 While some version of P*f nearly always

does begin what we presume will be a recapitu- Type 4! Variants latory rotation, a convincingly symmetrical re-

capitulation is not always forthcoming, nor does The Type 4! sonata, like the more standard the concluding rotation (the coda rotation) al- Type 49, follows a nonrepeated exposition with ways begin with a literal return of P'f. To be the return of the primary theme, which in this sure, in several ‘Type 4 finales Haydn additionally case simultaneously functions as the onset of the provided at least a quasi-symmetrical recapitula- recapitulation. Since both forms invite the retion subjected to his usual thorough recomposi- turn of Pf at this point, the potential for the

tion. When this happens, the result is that the slippage from the one to the other is present, normative Type 4 practice is at least more clearly and the only way to distinguish them is to experceptible amidst the recapitulatory recastings: amine what follows in the rest of the movement. in those, for example, of Symphonies No. 94 (Will there be a recapitulation after the ensuing in G, No. 99 in E-flat, No. 102 in B-flat, and development or episode, as in the Type 4°? Will No. 103 in E-flat. In these recapitulatory spaces that full-blown development or episode—if it Haydn revisited a sufficient number of (often exists as such at all—be patched into the already substantially recomposed) thematic modules begun P—TR space of the recapitulatory rotawithin the tonal resolution portion to provide a tion, as in the Type 4!-¢xp? Or will the rotation general sense of their quasi-symmetrical return. display not so much a developmental or episodic In other instances Haydn merely touched lightly “expansion” as a comprehensive recomposition, on an individual referential module or two with as in the Haydnesque variant of the Type 4!?) a much-compressed tonal resolution, as in the The generically challenging last movement of finales of Symphonies No. 85 in B-flat (mm. Symphony No. 95 in C Minor, for instance, 1s 190-97 = mm. 39-46, “S!-°”) and No. 88 in G perhaps best regarded as an unusual example of (mm. 195-97 = mm. 66-68, “S!-2,” varied and a Type 4! sonata among these finales.°’ In this

given a different continuation). case, within Rotation 2 the ,TR=>FS portion of Some of Haydn’s finales, however, begin in the continuous exposition returns significantly the manner of ‘Type 4s—at least through the non- recomposed—fully reconceived—including the repeated exposition—but then pursue other op- breakout of a quasi-episodic “C-minor storm”

56. This is also the view of Fisher, contra Cole (“Further ing authentic cadence at m. 78 is particularly provoca-

Thoughts,” p. 86). tive. This is not the way that Haydn’s developmental 57. The formal ambiguities called forth in Symphony spaces typically begin. On the other hand, such an eliNo. 95/iv are illustrative of Haydn at his most structur- sion is a characteristic way to launch a closing zone, C. ally playful. While by no means minimizing the ambi- That said, however, this C at m. 78 seems to aspire to euities, our preferred interpretation regards m. 78, the certain features of development (the descending fifths) emphatic resumption of the TR>FS fugato first heard but is soon cut off from that pursuit with the harmonic at m. 33, asa “celebratory” onset of C-space rather than discharge onto V/viin mm. 100—5. This suggests that as the beginning of a brief development. (If it were a what begins as an apparent C at m. 78 flirts with the idea development it would be either one of a Type 3 sonata of development but ultimately turns into a retransition.

without an expositional repeat and whose exposition Thus mm. 78-105 may be regarded as a C>RT link was elided directly into the development—virtually that has been uncommonly expanded—under which unheard of in Haydn’s multimovement works—or of interpretation the formal category of choice, lacking a Type 4 sonata that suppresses an appearance of Pf at a clearly separable development, shrinks to a Type 4 this point, a Type 4 deformation that does sometimes format. Nonetheless, its witty interplay with the exoccur in late Haydn, as mentioned in the next para- pectations of developmental activity at this point in the graph.) The direct elision of the fugato with the trigger- structure should not be overlooked.

416 Elements of Sonata Theory

(m. 152). A variant of Pf returns only to launch its nonrepeated exposition. In this instance the the coda rotation at m. 186. Somewhat simi- large-scale Type 3 impression 1s heightened by larly—though with different details—the finale omitting any reference to P™ after the recapituof Symphony No. 93 in D, with its developmen- lation. (On the other hand, if we choose to retally enlarged second rotation (recapitulatory gard this finale as a Type 3 tout court, it would be rotation, beginning with the opening phrases an isolated example in Haydn of a Type 3 whose of Pf at m. 172), seems most at home as a Type P-theme contains internal repeats and that does

4l-exp sonata-rondo. not display an expositional repeat.) In op. 76 no. 5/iv Haydn took the procedure a step further:

Omission of RT and Any Tonic Reference to this time the presto P-theme, while certainly

contredanse-like to serve. as Pr after_. sufficiently the Exposition _—. a rondo refrain, is not structured in a binary format. Since one of the defining features of the sona- Still, the exposition 1s not repeated (mm. 1ta-rondo 1s the immediate return of the refrain 120)—again, something that never happens in after the exposition, its pointed omission— Haydn’s “pure” Type 3 finales—and the music moving instead immediately into the develop- moves at once into the development at m. 126. ment—is a strong indicator of Type 3, not Type

4, behavior. Consider, for instance, the Presto More Radically Recomposed

, 7 9 , Recapitulatory Spaces

finale of the Quartet in F, op. 74 no. 2. This .

begins with a “rondo-style” rounded-binary theme with internal repeats (mm. 1-34), clearly While it was Haydn’s general practice to vary suggesting a Type 4 to come. The nonrepeated and recompose his recapitulatory spaces, in some exposition concludes with an emphatic full-stop of these finales the recomposition is so substanin m. 104. This final caesura shuts down any tial that the concept of “recapitulation” becomes

potential for the appearance of the expected strained. At what point of alteration does an exRT. As a consequence, the music omits the re- pected “recapitulation” turn into something diffrain proper to plunge directly into a P'™-based, ferent—suggesting, perhaps, yet another “epirotational development that begins at m. 105 sode” of an ongoing rondo, albeit one that might with the head-motive P'f treated in imitation, be construed as “standing in” for a more norbeginning on V and proceeding through a se- mative recapitulation? In the Vivace finale of ries of modulations. A recast recapitulation (at Symphony No. 101 in D the Type 4 outlines are first compressed, but then recomposed and en- clear through the developmental rotation (which larged triumphantly toward its end) begins with begins with a variation of Pf in m. 103 and pera tonic P'fin m. 146. Unlike what happens in sists through m. 188). The presumed recapitulathe exposition, this recapitulation does feature tory space, however, 1s completely reconceived an RT at its end (effectively, mm 259-71), one as a brilliant fugato based on the Pf head-motive that leads to a brief allusion to the refrain at the (m. 189), with, at best, only the most passing beginning of the coda rotation, m. 272. In sum: of potential allusions, if any at all, to figuration once past the exposition, this finale behaves like originally heard in the later parts of the exposia Type 3 sonata with a P-based coda, all the tion. This “recapitulation-substitute” section 1s while sporting a high-spirited, Presto-driven broadly symmetrical with the exposition in its

“rondo character’ in its themes. normative position within the movement, in its Haydn revisited the procedure of op. 74 no. P-references, and in its thorough grounding of 2/iv—the Type 4 that strives to become a Type the tonic, D. But as something that can give the 3 en route—in the last movements of two quar- impression of a “new” fugal episode, it is certainly tets from op. 76: no. 2 in D Minor and no. 5 not a normative recapitulation. While Symphony in D. Op. 76 no. 2 is a clear Type 3 in all fea- No. 101/tv is an extreme case, radicalized recomtures except for two decisive markers of the positions of the recapitulatory space might also Type 4: its binary, “rondo-like” P-theme (V1- be observed in the last movements of Symphovace assai), with notated internal repeats; and nies No. 97 in C and 100 in G.

Rondos and the Type 4 Sonata 417

The concept of a movement that starts, at sic back to the normative appearance of Prfas least potentially, as a Type 4 (with a sonata-style a recognizable (or unequivocally “thematic’’) exposition) but falls short of a full recapitulatory final refrain. In its place, typically in the gensymmetry as it continues is one of the source eral vicinity of the expected start of the coda— problems underlying the ongoing debates about though sometimes before or after it—may inthe structures of these finales. One way of re- deed be P'!-references, perhaps variants of the solving this problem is to take a more supple, ro- rondo theme’s incipit, but not enough of the tational approach to these movements. Whatever idea qua “melody” to consider it a refrain proper Haydn’s “recapitulatory” alterations to a poten- —_—_as encountered in more standard Type 4 sonatas. tial Type 4° sonata, the compositional space in — Haydn’s Type 4s often suppress the literalness of

question, leading off with Pf, is most emphati- _ the last P"-return in favor of procedures that are cally another rotation, one that usually high- _ better regarded as more typical of a normative lights the final solidifying of the tonic key. As Type 3 coda or CRI (which may or may not be pointed out in the discussion of rondo-rotations P-based). Examples are legion: at the ends of the

toward the opening of this chapter, in rondos finales, for instance, of Symphonies No. 93 in and ‘Type 4 sonatas each rotation is initiated by D (m. 292), No. 94 in G (m. 234, P-reference Prf but may then spin outward to produce dif- in the “wrong-key” E-flat), No. 95 in C Mifering continuations or alternatives, which can nor (m. 186), No. 97 in C (mm. 291, 313), and be quite freely conceived. Thus a Type 4 so- many other works. This alteration of more stannata is one with three rotations plus (usually) a dard practice helps further to produce the effinal coda rotation. Each of the three rotations fect of blending between the Type 4 and Type 3 begins with Pf and ends with an RT. In order formats. As was the case with several of Haydn’s minimally to qualify as a Type 4 sonata-rondo, —_ adaptations mentioned above, it suggests that Rotation 1 must be recognizable as a sonata ex- the most commonly encountered trajectory of

position (nonrepeated), in either a two-part or generic interaction in these finales is for them a continuous format. Rotation 2 often begins to start out as more or less regular Type 4s but with a shortened version of Pf (sometimes only oradually to sift in features more typical of Type the “a” limb) and proceeds either to an episode 3 practice. or to a development. (Occasionally this central section may consist of two subrotational cycles, with an intervening P*f and further develop- Type 4 Procedures in Mozart’s Concerto Finales ment, as in Symphony No. 94/iv, mm. 104 and 146.) In Haydn’s hands Rotation 3, the reca- Like those of many of his contemporaries—and pitulatory rotation, is the most variably treated like Beethoven’s—Mozart’s concerto finales are and often the most puzzling. To construe what almost always sonata-rondos, Type 4 sonatas. happens here primarily as another rotation in a The most notable exceptions are those written process of delineating an ongoing formal free- as variations, most famously in the last movedom—a rotation at least standing in for norma- ments of Piano Concertos Nos. 17 in G, K. 453, tive recapitulatory space—goes a long way to- and No. 24 1n C Minor, K. 491.°° Mozart’s conward facilitating our approach to these finales. certo finales comprise a collection of stunning individuals, not only in expressive tone and con-

No Return to Pfas a Literal Refrain tent but also in the local details of their designs.

. . While certain broad principles. are after the Recapitulation . . shared among them, his creative reshapings of the Type 4 con-

It often happens in Haydn’s Type 4 variants cept from piece to piece remain a challenge to that the recapitulation does not recycle the mu- anyone who secks to generalize about them. 58. As noted by Joel Galand, “The Large-Scale Formal 450, n. 25 (pp. 445-46): “The exceptions: the finale of Role of the Solo Entry Theme in the Eighteenth-Cen- the Concertone for Two Violins in C, K. 190 is a mintury Concerto,” Journal of Music Theory 44 (2000), 381— uet and trio with a written out da capo,” and the origi-

418 Elements of Sonata Theory

By the later 1770s and early 1780s Mozart Mozart’s concerto-finale Type 4s constitute a had developed his own customizations of Type second order of hybridization among the sonata 4 practice. These customizations were also in types that Sonata Theory proposes heuristically. dialogue with the Type 5, ritornello-grounded The non-concerto Type 4 sonata is already a structures that he was developing concomitantly (first-order) hybrid between the rondo and one in the concertos’ first movements. As a result, of three other types of sonatas: the Type 3, the it is difficult to confront Mozart’s Type 4 con- Type 1, and the expanded Type 1. Mozart’s certo finales without presupposing an awareness concerto finales usually sift in yet another eleof the principles of his Type 5 first movements, ment of hybridization, an intermixing, in varythe subject of chapters 19-22. This is especially ing degrees from piece to piece, with the comtrue of the concertos that most listeners and an- poser’s own idiosyncratic treatment of the Type alysts are likely to confront, the seventeen Vien- 5 sonata, itself a mixture between the earlier, nese plano concertos written between 1782 and ritornello-based concerto and sonata form. One 1791 (“Nos. 11-27”—No. 12 in A, K. 414, is the effect of this intermingling of structural assoearliest of the set) and the Clarinet Concerto in ciations is the occasional blurring in concerto A, K. 622 (1791), although it applies to several Type 4s of some traditional functions that are others as well. These features proceed beyond clearer in the “purer” sonata formats—includfeatures of local texture—beyond the expected ing Type 5 formats.

dialogical interaction of solo and tutti. Despite these complications, it is possible to For this reason, in the present discussion of — cut an efficient path through the topic of MoMozart’s concerto Type 4s we are obliged to al- zart’s concerto finales by keeping nonconcerto lude to concepts and terminology that will be Type 4 designs firmly in mind—those discussed laid out in full only in the following four chap- earlier in this chapter—and then looking at the ters. These include references to such things as portions of the concerto finales that are most ritornello designs and functions, thematic la- likely to differ from those simpler designs. This beling conventions within broad zones (such as entails a glance at each of three main topics: R1:\C to designate an apparent closing theme (1) the varieties of Type 4 designs found within that initially appears within a first orchestral © Mozart’s concerto finales; (2) the texture and tutti or ritornello), P-theme prefaces, and sujet- structure of the opening refrain (P*) and its tutlibre styled transition themes. Readers coming ti-extensions, if any; (3) the normative addition to this discussion from the concerto chapters of a “new solo re-entry theme” starting in the that follow should have no difficulty in relating tonic after the full closure of those P-space tutthese terms to Mozart’s Type 4 concerto-finale ti-extensions. Once these things are clarified, movements. Readers who have not consulted the remaining features of the individual Type 4

chapters 19-22 might wish to work through movements become more navigable, notwiththem first, or at least to consult those chapters standing their ad hoc character and localized difnow and again for more expansive definitions ferences. Our aim here can be only to present and examples of the new terms here. To be sure, an overview of the central issues, along with an this discussion of the concerto finales could have explanation of how these matters appear from been placed at the end of the Type 5 discussion. the perspective of Sonata Theory. And yet the rondo-logic of these movements makes them more at home in this chap eT wee Type 4 Formats in Mozart’s Concerto Finales as we are thereby compelled to anticipate things

to come in the remainder of this book. Mozart’s Type 4 concerto finales fall into two One should also underscore another aspect broad categories, each of which in any individual that deepens the problem at hand. The mature case 1s usually submitted to additional concerto nal finales of Piano Concerto No. 5 in D, K. 175, and “the finales to the Sinfonia Concertante in Eb, K. 297b Violin Concerto No. 1 in B-flat, K. 207, are differing (a doubtful work), and the Concertos in G, K. 453 and variants of what we call the Type 5 sonata. In addition, C Minor, K. 491, are variation sets.”

Rondos and the Type 4 Sonata 419

enrichment. The first is the Type 3 sonata-rondo “normative” central episode and the recapitumixture, the standard Type 4, or Type 43, witha — lation. This occurs in Piano Concerto No. 9 central development or episode—or both—fol- in E-flat, K. 271/111, which includes a suddenly lowed by a full recapitulation. The second isthe _ static, interpolated minuet in A-flat, IV, after expanded Type 1 sonata-rondo mixture, Type 4!-&?, the central episode proper and before the recain which the recapitulatory rotation begins im- _ pitulation; and in an earlier work, Violin Conmediately after the completion of the exposition certo No. 5 in A, K. 219/in, with its extended, and subsequent retransition but contains inner = A-minor “Turkish” episode, preceded and foldevelopmental or episodic expansions once P*f — lowed by P'*—which results in nothing less than has been sounded. Following its central expan- _—_— the unexpected “invasion” of an entire, “extra”

sion, this second rotation, in most cases, will —_ rotation before the onset of the recapitulation. merge back into correspondence measures—the | Another place to interpolate an “extra” epicrux—somewhere in the TR-zone (sometimes, sode was between a statement of the recapitulathough by no means always, at the convenient tory P™and the onset of the “new theme” TR point of the solo’s “new-theme TR opening”) proper. This may be found in Violin Concerto

before proceeding into the tonal resolution. No. 3 in G, K. 216/111, and Concerto “No. 7” One source of concerto enrichment in both in F for Three Pianos, K. 242/111. formats 1s the highlighted interplay of solo-tutti Much notice has been taken of Mozart’s dialogue (theme-sharing, back-and-forth re- other concerto-finale format, the Type 4!-«x, sponses, yieldings and collaborations), always a double-rotational pattern plus coda-rotation a pronounced feature of concerto practice, and = with only three appearances of Pf (beginning treated by Mozart in an unpredictable, indi- the expositional, recapitulatory, and coda rotavidualized manner from finale to finale. Oth- tions), not four as in the Type 4° pattern (which ers include: the potential expansion of the ton- _also includes an interior developmental rotation ally closed P'f theme proper into a fuller block — followed by a recapitulatory restart with P*). by means of a series of tutti-extensions, some Following a thorough study of Mozart’s concerwith codetta functions; the almost invariable tos, Joel Galand concluded that what we call the beginning of TR proper with a new theme Type 4!-exp design “was Mozart’s favorite procefor the soloist alone; the possibility for two- or | dure for composing finales that incorporated the three-stage central sections, perhaps juxtapos- _—sritornello and recapitulation techniques he was ing developmental passages with episodes; the | using in his concerto first movements, while frequent preparation for and execution of a —_ also exploiting the popular fashion for ronhigh-display cadenza, normally led into by the — dos.”°? Examples may be found in the finales last retransition (set forth as an emphatic tutti § of twelve concertos: the Oboe Concerto in C after the completion of the recapitulation), and — (K. 314); the Flute and Harp Concerto in C (K. followed by the last sounding of P'f, which 299); the Sinfonia Concertante in E-flat for Vi-

launches the coda-rotation. olin and Viola (K. 364): the Clarinet Concerto

Many of Mozart’s concerto finales present the in A (K. 622); and the finales of numerous piano Type 4 pattern in full. Examples may be found = concertos—Nos. 11 in F (K. 413); 12 in A (K. in Piano Concertos No. 10 in E-flat (K. 365, for 414); 13 in C (K. 415, though with significant Two Pianos), 14 in E-flat (K. 449), 15 in B-flat complications, discussed further below); 18 in (K. 450), 16 in D (K. 451), 21 in C (K. 467), — B-flat (K. 456); 19 in F (K. 459); 20 in D Minor

22 in E-flat (K. 482), and 25 in C (K. 503). (K. 466); 23 in A (K. 488); 26 in D (K. 537); Additionally, some of Mozart’s earlier concerto and 27 in B-flat (K. 595).

finales display Type 4 architectonic formats but As might be expected, the main confusion with an “extra” episode or expansion at some regarding this latter structure—as with its nonpoint after the central episode. One place for | concerto version—has been the unhelpful apsuch an additional expansion was between the _ plication of schematic letters to it, ABACB'A— 59. Galand, “The Large-Scale Formal Role,” p. 408.

420. Elements of Sonata Theory

along with the incorrect assumption that the lights participation by the soloist in an engaging format should be regarded as a seven-part sona- interplay with the orchestra. In the third volume ta-rondo with a supposedly expected “third A” (1793) of the Introductory Essay on Composition, omitted after C.° This misconstruction, which Koch mentioned that “it is more usual that the sometimes multiplies the analytical missteps by solo part performs the rondo theme first, before invoking the erroneous concept of a “reversed it is repeated as a ritornello by the orchestra.’’® recapitulation,” has been undermined in the —= Apart from the problematic word “ritornello” past fifteen years in work by Joel Galand (an 1n- (which creates more difficulties than it solves in dispensable source in this regard, who correctly many of these finales), Koch’s description 1s inreferred to the Type 4!-*P as one type of “ex- deed often the case in Mozart’s concerto finales. panded binary” pattern—the “exposition-re- It is typical for the soloist to lead off with the capitulation pattern... [here also] subjected first module of the rondo theme proper, which to the ritornello principle”), by John Daverio may then be continued and brought to its con(calling it the concerto version of an “amplified clusion through a back-and-forth dialogue bebinary” pattern, our expanded Type 1), and in tween tutti and solo. But neither this familiar earlier portions of this book.®! Galand’s position solo-onset nor the common solo-tutti interhas been reaffirmed in recent work on Mozart’s change within P*fis invariable. A second-level concerto finales by David Grayson and John Ir- textural default was to sound the entire P-theme ving. (Irving also provides a succinct overview complex, including all of Pf, in the orchestra

of Mozart’s concerto-finale procedures.) © and to bring in the soloist only at the beginning of—or even slightly into—TIR. P-theme

Prf and the Possibility of Tutti-Extensions rom lexes sounded entirely b y the tutti may be

(“Ritornello 1”) found in the finales of Violin Concerto No. 3 in G, K. 216, mm. 1-40 (in which the orches-

In a nonconcerto Type 4 rondo, P*f may be a tral violins are also doubled by the violino princirelatively brief theme (a period, sentence, or pale), and Piano Concertos No. 11 in F, K. 413 hybrid) or a slightly more developed structure, (mm. 1-32), No. 12 in A, K. 414 (mm. 1-20), such as a simple binary or rounded binary form. No. 14 in E-flat, K. 449 (mm. 1-32), No. 16 Whatever the choice, the self-sufficient Ptf — in D, K. 451 (with TR also beginning in the the rondo theme proper—comes to a caden- orchestra in m. 17, into which the soloist soon tial close, I:PAC, before proceeding onward, in merges), and No. 25 in C, K. 503 (mm. 1-32). most cases, into one or more P-block extensions (No. 21 in C, K. 467/i11, also begins with the confirming that cadence. In Mozart’s concerto orchestra, though the soloist enters to conclude finales Pf typically (though not always) high- Prfin mm. 21-28.) A less common option 1s to

60. With regard to the concerto, this is implied, for 62. Grayson, Mozart: Piano Concertos Nos. 20 in D Miexample, in Green, Form in Tonal Music, 2nd. ed., pp. nor, K. 466, and No. 21 in C Major, K. 467 (Cambridge: 251-52 (“The third refrain of the concerto-rondo 1s fre- Cambridge University Press, 1998), pp. 73-92 (esp. p. quently omitted [in Mozart]’’); cf. n. 43 above for the 82—on K. 466/i11— following Galand and insisting that

same issue in nonconcerto versions of Type 41-«%P. it is only “a preconceived (and wrong-headed) notion 61. Galand, “The Large-Scale Formal Role,” p. 408. Ga- of the form” that results in the erroneous impression land’s most extensive study is “Rondo-Form Problems in “that something [an A refrain] may seem to have been Eighteenth- and Nineteenth-Century Instrumental Mu- omitted, suppressed, or bypassed”’). Cf. the similar consic, with Reference to the Application of Schenker’s Form clusions in Irving, Mozart’s Piano Concertos (Aldershot:

Theory to Historical Context,” Ph. D. diss., Yale Uni- Ashgate, 2003), pp. 73-93 (chapter 5; Movement Forms versity, 1990. The central conclusions are summarized in III: Finales,” esp. pp. 77-79). Galand, “Form, Genre, and Style in the Eighteenth-Cen- 63. Koch, Introductory Essay on Composition, trans. Nancy

tury Rondo” (n. 6 above) and “The Large-Scale Formal Kovaleff Baker (New Haven, Conn.: Yale University Role” (n. 58 above). See also Daverio, “From “Concer- Press, 1983), p. 172, n. 50. Also cited in Irving, p. 81, tante Rondo’ to ‘Lyric Sonata,” esp. pp. 113-19, with a who provided examples and variants in Mozart. list on p. 114 of examples that include concertos.

Rondos and the Type 4 Sonata 421

have P-space occupied entirely by the soloist (as finales contain brief codetta-extensions that are in Piano Concerto No. 9 in E-flat, K. 271/111, not played exclusively by the tutti, and P-space

mm. 1-35). is concluded with music led by the soloist. This Even more striking is that one often finds— happens in the finales of Violin Concertos No. especially in the Viennese piano concertos— 4in D, K. 218 (P? and P?, mm. 19-22, 23-30, that the rondo theme proper (the initial, I: PAC- then linked with a sudden tutti impulse into a closed P*f, in which the soloist 1s often a par- solo-led TR proper) and No. 51n A, K. 219 (P?, ticipant) is only the first element of a more mim. 17—22); and in those of Piano Concertos

broadly conceived P-space. The square-cut, No. 7 in F, K. 242 (P*f, a solo-tutti anteced“rondo-character” P*f proper is almost always ent and consequent, mm. 1-8, 9-16, followed followed by several additional, reinforcing tonic by P?2, solo-led cadential codettas, mm. 17—22) modules, normally played by the orchestra alone and No. 9 in E-flat, K. 271 (an which the piano as a tutti. Sometimes these tutti-extensions to solo occupies all of P-space, mm. 1-35, and the Prf (P2, P3, and so on) are brief—only a codetta tutti that follows at m. 35 is a straightforward

module or two. On other occasions the tut- TR of the dissolving restatement type). In such ti-extensions are much longer. Typically, they cases, the concept of any “ritornello” intervenclose with a rhetorically emphatic flourish in the tion proper (in the Type 5 sense) is hazy—apart

tonic, I:PAC, followed by a final caesura and from an appeal to the loose, Kochian sense of re-entry of the soloist with a new theme (the a tutti restatement of or brief response to a soonset of TR-space, TR!, as we shall propose be- lo-presented module within an otherwise musilow). These multimodular extensions can take cally closed P. on roles that recall those found in the opening Most often, however, one finds a closed Prf ritornellos of Mozart’s ‘Type 5 sonatas, such as a with solo participation that comes to a I:PAC closing-theme or codetta-like function.°+ This (with or without the soloist at that point) that is hybridization (of Type 4 and Type 5) presents elided with a set of vigorous tutti modules, all in analytical and conceptual problems that must be the tonic. Examples are plentiful in the Type 4 confronted head-on, andit appears to have been _ finales of Mozart’s piano concertos. What varies a Mozartian innovation. As Galand concluded, from one instance to the next are a number of referring to the more extended instances of this factors. As already mentioned, there 1s the unpossibility, “Mozart appears to be the first to predictable degree to which the soloist will parhave developed the initial rondo refrain into a ticipate in the rondo theme proper, along with full-fledged ritornello section that occupies al- the varying ways in which the orchestra can remost the same proportions as those found in first — peat or complete P'. Also variable are other fac-

movements.’’> tors that pertain to the post-P" tutti-extensions When the tutti-extensions are short, they that are appended to it while still occupying an present few if any difficulties under the more expanded P-space: its breadth and extent; the normative Type 4 concept. As modest tutti-co- differing quasi-functional characters of its moddettas within a clearly demarcated P-space, they ules (locally TR-like? S-lhke? C-like?); and the seem little more than examples of more or less uses to which these modules will be put (or not standard (non-concerto) Type 4 P™ formats en- put) in the remainder of the movement.

riched through the juxtaposition of solo-tutti In the finale of Piano Concerto No. 27 in contrasts. Some of Mozart’s earliest concerto B-flat, K. 595, for instance, Mozart laid out P'f

64. Koch’s writing from the late-eighteenth and ear- tutti extensions were not unheard of in the concertos ly-nineteenth centuries is of little help with regard to of other composers “that Mozart might have known— this “ritornello” issue in Mozart’s concerto rondos. Cf. principally ... Viennese and Mannheim composers, and Irving on Koch, Mozart’s Piano Concertos, p. 91. those of the French violin school, [in whose concerto 65. Galand, “The Large-Scale Formal Role,” p. 419. rondos| the rondo refrain may be followed by some perGaland did note on the same page, however, that shorter functory tutti closing measures.”

422 Elements of Sonata Theory

as a sectional rounded binary, aaba, in which low). There are several ways that this can occur. the soloist plays the entire theme except for its One of the most obvious 1s the presence of a second limb (the second a, mm. 9-16), which 1s nonelided “C-style” tag at the end of the orgiven over to an affirmative repetition from the chestral extensions—a tag more set off and sepaorchestra). Pf is an athletic, sprinting theme in rately stylized than those found in K. 467/111 6/8 that comes to a “first conclusion,” I: PAC in and K. 595/111, mentioned above.®® The end m. 39. At this point the piano solo drops away, of the P-space in the finale of Piano Concerto and the tutti-extensions occupy mm. 40—64. No. 22 in E-flat, K. 482, for instance, seems These orchestral extensions consist mostly of — to cross this line into “C-character.” FollowP-related material (keeping P™ proper alive and ing an aaba rounded-binary P* (led mostly by developing in the tutti), expanding P-space fur- the soloist, with the orchestra delivering only ther outward. They begin with varied recap- the second limb, mm. 9-16, as would also be the turings of material related to the previous ca- case in K. 595/111), the tutti-extensions begin dence (P?:!, mm. 40—51, as if backing up for a with a forte P*, interpretable, as in the above varied repetition), a decisive cadential module cases, as a P-codetta (mm. 41-51). M. 51 1s es(P2-2, m. 51-59, with internal self-repetition), sentially a filled-in gap, and a nonelided, impuand a quiet closing tag, P'-related at the end dently buffa tag now enters, piano, as an “extra” (P3, mm. 59-64). As is typical, the soloist then P31, mm. 52-59, followed by a P?-? continuare-enters to begin a new theme in the tonic (m. tion (mm. 59—67)—precisely the sort of cheeky, 65), which we shall interpret below as the onset “afterthought” theme that would be at home in of TR-space. None of the tutti-extensions to P C-space in Type 5 first movement ritornellos. In will reappear in the movement until the final addition, a forte P+ follows, mm. 67-71, soundstatement of the refrain following the cadenza. ing much like the concluding emphatic flourish The general procedure of the finale of Pi- of a normative ritornello C-space. (In K. 482/111

ano Concerto No. 21 in C, K. 467, is some- most of these “C-character” ideas will be rewhat similar. Here, however, as mentioned ear- heard only in the final statement of the refrain, lier, the rounded-binary P*f theme is given over after the cadenza. This is another feature, one mostly to the orchestra, with the piano sound- of subsequent modular postponement, that reing only its final limb, the reprise (the last a, calls Type 5 first-movement practice. Only P>-?, mim. 21-28). Its I:PAC at m. 28 elides directly placed into a different context, will also resurwith the forte-affirmation tutti-extensions. The face earlier, in the recapitulation proper.) Only solo piano now withdraws, and the orchestral after this set of tutti-extensions does the piano ideas, only modestly extended in this case, are re-enter in m. 74 with a “new” ‘TR—but one readily construed as vigorous P-codetta mod- that is obviously reacting to the curious “extra,” ules (with final tag). After the tutti’s full stop P> by recasting it in a different light, as if pickin m. 57, I:PAC, the solo piano re-enters with ing it up and musing on it.

a tonic theme new to this movement, launch- Another way that more substantial, post-P'f

ing TR. tutti-extensions can take on a local ritornello Interpretive issues arise, though, when the cast is by following the completed refrain proper, P-space tutti-extensions are more expansive one usually led by the soloist, with the onset of or contain modules that convey a TR, S, or C a forte orchestral restatement that dissolves into flavor, as though the P'f + tonic-key tutti-ex- more characteristic, energy-gaining TR activity tensions were being more explicitly composed before proceeding to its tonic cadences. (Such a to take on some of the attributes of a compressed procedure also recalls Koch’s 1793 claim, men-

Type 5 Ritornello 1, even though what follows tioned above, of having the soloist sound the seems to begin TR-space (as we shall argue be- refrain first, then calling upon the orchestra to 66. Cf. Irving, Mozart’s Piano Concertos, p. 82: “A re- material... . In taking such an approach, Mozart was curring characteristic of [the orchestral continuations importing a feature more familiar from the ritornello of] Mozart’s refrains is the clear segmentation of closing design of his concerto first movements.”

Rondos and the Type 4 Sonata 423

repeat it as a “ritornello.”)®’ This happens in the the exposition, first as TR? (m. 39, following last movement of Piano Concerto in D Minor, the “new-theme,” solo-entry TR! in the tonic, K. 466. Here the strikingly brief P*fis reduced mim. 21—38),°8 then, most surprisingly, as S, in

to a blunt period with expanded consequent, V, mm. 56-80. played by the soloist alone, mm. 1-13. The or- The last movement of Piano Concerto No. chestra sounds what begins as a forte restatement 13 in C, K. 415, provides another illustration— with the upbeat to m. 14, perhaps at first sug- an elegantly styled, seemingly carefree surface gesting the more normative second limb of an that engages substantial conceptual problems of aaba rounded-binary theme. Whatever our ini- formal type and structural ambiguity. What we tial impression, we soon realize that the orches- presume, according to the concerto-finale Type tral entrance instead begins the tutti-extensions. 4 norm, is the opening P"™!-block, mm. 1-48 (in Within only a few bars Mozart steers the orches- this case with solo participation only in mm. tra’s P'frestatement into developmental activity 1-8), articulates an aabc + codetta structure. that evokes the texture of a TR, one leading to Considered by itself, this melodic shape presents a dominant-lock at m. 30, to further dramatic no analytical difficulties: P™=* (mm. 1-8, 9-16), activity in the measures that follow, and finally = Pt’ (mm. 17-30), Ptf* (mm. 31-44, ending to ringingly declarative final modules that set I:PAC), and the nonclosed Prfcodetta (mm. 44— the seal on the extended “ritornello” effect with 48). What follows, however, is a non-normatwo grim and authoritative PACs in D minor _ tive Adagio solo interpolation in C minor, mm. (mm. 59, 62) before the re-entry of the piano 50-64, with more of an S1:\Pprreface effect than

with its new theme in m. 63. that of a typical solo-entry “new theme.’’®? MoAny implicit dialogue with the Type 5 zart then follows this unusual passage (which Ritornello 1—one of whose roles is to provide has apparently ground Type 4 “progress” to a a potential source of secondary and closing ideas halt) with a restart of the exposition proper— to come in the ensuing solo exposition (chap- an extremely atypical event within concerto ter 20)—1is made considerably more provoca- finales—featuring a restoration of the tonic mative when seemingly P-completing materials —_ jor for a full restatement of Ptf (mm. 65-72) from the tutti-extensions resurface many bars leading to the onset of P*#>, now expanded and down the road to occupy S-space, now in the treated as the modulatory portion of the transidominant. In such cases the “ritornello role” of — tion (mm. 72—91). P'** now appears intact as the opening P-block is even more suggestive, Sin V, mm. 92-101, and new closing material even to the point of taking on quasi-rotational (C) follows. In this virtually unique case, the (or subrotational) connotations. What appears opening tutti, mm. 1—48, serves also as an inito be a simple P* module in the finale of Piano tial rotation in the manner of a Type 5 sonata. Concerto No. 12 in A, K. 414, for instance—a One might even argue that K. 415/111 begins as serpentine, coiling figure, mm. 9—16—slithers a Type 5 movement (or at least with a Type 5 its way throughout much of the remainder of | movement in mind) but converts into a Type 4 67. At issue here is the dissolving restatement follow- 68. The sometimes-problematic TR! status of such seping the completion of the fully closed P™ rondo theme arate, closed themes in the tonic (sujets libres) 1s treated proper. The orchestral restatement of the initial “a” idea in the following section of this chapter. midway through P* within a binary or rounded-binary 69. Cf. the similar S1:\Ppreface strategy in Violin Con-

context, asin K. 488/iii and K. 595/iii, is a different certo No. 5 in A, K. 219/i—in that case part of an unmatter. K. 488/111, however, is notable in that the soloist equivocal Type 5 sonata, as mentioned also in chapter

does not return in the second portion of the binary P*. 21. On such labels as R1:\S, RI:\EEC, R1:\C, and so Once the orchestra enters with its forte repetition of the on, see the discussion in chapter 19. In brief, these labels initial period, mm. 9-16, it continues to take over the refer to typical expositional-rhetoric zone or functions texture completely, supplying the closing cadence for within tonic-key Ritornello 1-space in a Type 5 sonata Prf proper (m. 40), and proceeding to plunge into the (concerto first movement). tutti-extensions as well (mm. 40—61)— thus producing, texturally, an extended “ritornello” (or at least “tutti’’) effect in mm. 9—61.

424 Elements of Sonata Theory

sonata-rondo (of the expanded Type 1 subtype) be understood as filling out a much-broadened only with the retransition at the end of the solo P-space, one to be followed by the soloist’s exposition. At this point of conversion—or of — re-entry with TR. From another perspective, the full declaration of its truer Type 4 status—it though, Mozart crafted those 119 bars to sound drives home its point by suppressing a crucial like a Type 5 Ritornello 1, as if the orchestra’s marker of the ‘Type 5 sonata, the trill-cadence normative tutti-extensions “take off” wittily on plunge into an unmistakably formalized Ritor- their own, vaulting past the more modest duties

nello 2 (chapter 20). of normative Type 4 Pt“responses, now with

The extraordinary finale of Piano Concerto impulsive aspirations of shaping themselves into No. 19 in F, K. 459, is another, more fully de- a bona-fide Type 5 opening tutti (notwithstandveloped case along these lines. Once past the ing the soloist’s participation in Pf). The tutcontinuous binary P*f concluding with the usual ti-extensions appear to wish to change the basic I:PAC (aabb, mm. 1-32, in which the piano and format of the piece en route from a Type 4 to a the orchestra alternate phrases), the aggressively Type 5—and they very nearly succeed, until the forte tutti-extensions elided at m. 32 (presum- plano reappears to restore “Type 4 order” and ably P?) instantly take on the character of an settle the issue, in this case with its new-theme independent, imitative-contrapuntal TR (or TR in m. 120. The aftershocks of the orchestra’s R1ATR), one that even leads to a I:HC dom- almost-successful generic coup, however, coninant-lock (mm. 62—65) anda I:HC MC at m. tinue to be felt both with the reappearance of 65. Even more astonishingly, what follows (per- the tutti-extension’s P?-! (= quasi-R1:\P-based haps also understandable as P+, even though it S!1, mm. 67-74) in the exposition’s S proper at responds to an MC -effect) has the unmistakable m. 167 and with the assertive invasions of P? character of a P-based S (mm. 67—98, similar to (= quasi-R1:\TR!“) in the C-space display epia Type 5 R1:\S in the tonic). And indeed, Mo- sode, mm. 228-33 and 236-41. zart would reuse its presentation modules, mm. There is thus a hybrid-like double perspective 67-74, as the initial idea of the dominant-key at play in several of the opening sections of MoS in the remainder of the exposition that fol- zart’s concerto finales. Whatever the extent of lows (mm. 167—74).729 An R1:\EEC-effect the “Type 5 ritornello impression” at the outset, (I:PAC) occurs in the tutti-extensions at m. 98, though, the broader course of the movement 1s and what seems to be an R1:\C! of the charac- soverned by Type 4 norms. Virtually all of the teristic 8—57—6—47-8 type follows: two short tutti-extensions, no matter how expansive they (P+) subcycles, mm. 99-106 and 107-114. These might be, are in the final analysis to be reckoned are rounded off with a short, emphatic close (P°) as operating inside an expanded P-space, albeit in the manner of an R1:\C? in mm. 115-19— one that might be staged as exhibiting local TR, leading to a full-stop I:PAC anda re-entry of | S, or C behavior. When the quasi-ritornello efthe soloist, as usual, with the new, independent fect surfaces, it is usually nested as a secondary theme that is best regarded as the onset, finally, phenomenon within a stretched-out P-space

of TR-space proper (m. 120).7! that seeks to open a dialogue with the principle The analytical challenge posed by the ini- of the Ritornello 1 within a Type 5 concerto tial 119 bars of K. 459/111 (as with several first movement. Since such a movement had beother movements described above) is that from sun the concerto—often with a grandly outone perspective—that of the movement as a lined Ritornello 1—the Pf + tutti-extensions whole, especially as construed with Mozart’s of the finale’s beginning can provide a balancing other, more typical exemplars of the Type 4 complement between the openings of the outer

concerto-finale genre in mind—they are to movements. 70. M. 167 may also be construed as TM!, the onset of 18 (instead, for example, of consigning mm. 203-18 to a trimodular block, if we are willing to interpret the C-space, which might be the preferable view). V:PAC at m. 202 not as the EEC but as a V:PAC MC 71. Once again, regarding the R1:\ and S1:\ labels, see

followed by a characteristic, periodic TM3, mm. 203- n. 69.

Rondos and the Type 4 Sonata 425

But the fact remains that, unlike the case in “extra” theme in the tonic, a surplus idea that the first movement, the expanded, tonic-key one does not encounter in nonconcerto Type tutti of the finale is to be grasped as a specially 4 movements. This means that Mozart’s sonastylized filling-out of the exposition’s P-space, ta-rondo concerto finales begin with a series of regardless of its internal contents. Instead of lay- challenging signals to interpret within any siming down a complete referential rotation (first pler Type 4 context. Not only is the concept of rotation) for the rest of the movement, as hap- a closed refrain problematized with the hybridpens in Type 5s (see chapters 19 and 20), it al- ity and ritornello-like aspect of the tutti-extenmost always occupies only the opening zone (P) sions, but the solo re-entry theme seems also of the actual first rotation, which extends from like a surplus element that lingers further in the

the beginning of the movement through the already overdetermined tonic. These features end of the exposition proper. What follows the have made analytical descriptions of the confinal module of the tutti-extensions, with their certo finales difficult. emphatic closure in the tonic, I:PAC, is to be One familiar solution has been to consider regarded as the onset of TR, as will be discussed the new theme in the solo to be the start of below. For this reason, and despite the potential “episode 1” of the sonata rondo. This approach,

for the P+ tutti-extensions to remind us lo- however, 1s hobbled by the use of “episodic” cally of Type 5 Ritornello 1 activity, there is no terminology within an ongoing sonata exposineed to identify its contents as “R1:\ themes,” tion. (While the term “episode” and its accomas we shall do in Type 5s (chapters 19 and 20). panying schematic letter-formats—A, B, and so The thematic modules of the opening zone may on—are workable in simple rondos, they are inbe identified merely with such P-superscripts appropriate for describing the behavior of Type as P! (or Pf), P2-!, P22, and so on, although to 4 sonata-rondo expositions and recapitulations.) suggest the full implications of a module one Green went further than this in the 1960s and might wish to note, for instance, such things as 1970s, maintaining that the new solo theme in “P2 (= quasi-T'R!-!),” as we have done in K. the tonic began not only “Episode 1” but the

459/iu, m. 32. movement’s exposition as well, in a manner somewhat comparable to the post-Ritornello-1

The Solo Re-entry in the Tonic solo exposition of what we calla Type 5 sonata./9

(the Sujet-Libre TR!) One can sympathize with the difficulty of the

problem of coming to terms with the opening A much-noticed idiosyncratic aspect of Mozart’s events of Mozart’s Type 4 concerto finales, but concerto finales is the immediate contrast pro- this explanation only muddied the waters furvided after the tutti close with the re-entry of ther. It relegated the complex activity preceding the soloist (or, less often, his or her initial entry, the solo re-entry to a refrain (or ritornello-like) if the soloist had not participated in P), usually status that stands apart as a separate entity, out-

with a “new theme” that begins (and some- side of the sonata form proper. (In part this was times also closes) in the tonic before beginning an inappropriate transference of a classical probto modulate to the secondary key of the move- lem regarding the sonata-form status of Ritorment.’? Particularly after the multiple tonic ca- nello 1 ina Type 5 sonata, a problem taken up in dences of the preceding P*f + tutti-extensions chapters 19, 20, and 21.) For Green and several

(tonal overdetermination), this new theme— others, the exposition of a Mozartian concerto marked also by a suddenly contrasting solo tex- finale gets underway only when the opening ture—can give the impression of still another, rondo-gestures and their tutti-continuations

72. Only in rare cases, such as in the finale of Piano in D, K. 451/11, provides another exception to the “new

Concerto No. 14 in E-flat, K. 449, does the soloist theme” norm. Here the pianist merges (m. 21) into a re-enter with a repetition of P™—in this case launching TR already begun by the orchestra (m. 17). a TR of the dissolving restatement type (m. 33) follow- 73. Green, Form in Tonal Music, 2nd ed., pp. 251-53. ing a Pf that had included no solo participation. No. 16

426 Elements of Sonata Theory

have been decisively completed. This seems 48-64. When this happens it can easily give both counterintuitive and unlikely. More to the the impression—doubtless purposefully—of a point, it excludes the Hauptgedanke of the move- brief, separately closed idea (or episode) in the ment, the rondo theme proper (P*), from par- tonic before more normatively transitional maticipation in the exposition, something that does terial (here considered as TR2) 1s allowed to not happen in normative nonconcerto Type 4s. proceed forward: a momentary halting of the As we have argued above, however, Pf and = forward-vector to permit the soloist a neatly its tutti-extensions are most profitably under- framed, individualized statement on his or her stood as occupying a specially crafted and ex- own. On this reading, the fully closed TR! panded P-space within the Type 4 sonata form. would not display overtly transitional behavior. Since the solo re-entry almost never beginsa ___ Yet it could still be regarded as occupying the new rotation (by reverting back to P*‘),”4 but beginning segment of the TR-zone, generically instead continues the ongoing rotation with a considered. texturally contrasting new theme that begins in This “TR” understanding is reinforced when the tonic, it is most reasonable to assume that we consider an alternative procedure, one in this solo theme operates as the onset of an in- which the solo re-entry theme in the tonic 1s

dependent, thematic TR-space. That TR can not fully closed with a perfect authentic cabegin melodically—and with a new melody— dence. In these instances, the sujet libre moduwas established in chapter 6. There is no ge- lates away from the tonic before any such formal neric reason to disallow TR-status to a “new I:PAC is achieved (even though a I:HC might theme” beginning in the tonic at this point. have been articulated before modulating away And indeed, as will be elaborated in chapter from the tonic). It thus replaces a sense of local 21, “new-theme” TRs are a commonly selected tonic closure with an opening outward toward option in the solo expositions of Type 5 con- more “typical” TR behavior. This occurs in the certo first movements. In that format, we call Violin Concertos K. 216/iii, mm. 41ff, and K. these familiar TR openings (adapting terminol- 219/i11, mm. 23ff; in the Piano Concertos K. ogy from Saint-Foix) sujet-libre (“free-subject’’) 456/i1, mm. 58ff, K. 482/11, mm. 74ff, with an SIANTRs, or SI:\TRs of the sujet-libre type. evaded cadence in m. 89, K. 595/111, mm. 65ff, The “new theme” solo re-entries in Type which begins with what might be regarded as 4 concerto finales (TR!) share a number of _ two elided, five-bar presentation modules prodefining characteristics also found in the Type longing the tonic over a tonic pedal and lacking 5 sujet-libre S1:\TR!. They are independent a full-closure PAC at the end of each; and in the themes assigned exclusively to the soloist fol- Clarinet Concerto, K. 622/111, mm. 57ff. lowing some type of full tutti close. They thus The sujet-libre TR', like its sujet-libre S1:\TRI

represent a handing-over of the action from counterpart in Type 5 first movements, may the orchestra to the soloist. Correspondingly, either reappear or not reappear elsewhere in they provide a sense of initiating a new section the movement. It may happen, though less frewith a new melody (“onward!”’), or with a new quently, that the sujet-libre TR! is a one-time variant of or clear response to something heard event, as in the Piano Concertos K. 413/iu, K. earlier. Quite often the sujet libre is a complete 450/111 and 459/11. More often, though, it does melody—period, sentence, or hybrid—brought play a role elsewhere—perhaps as something to to a close in the tonic, I:PAC, before proceed- be revisited or varied in the development or cening onward to a modulation. Examples may be tral episode (K. 466/111, K. 503/111, K. 537/111), found in the Piano Concertos K. 450/111, mm. situated in a similar TR-position in the reca43-62; K. 459, mm. 120-42; K. 466/111, mm. pitulation (K. 467/111) or TR-crux-point within 63-73; K. 467/111, mm. 58-74; K. 488/111, the familiar Type 4!-exp variant (K. 456/111; K. mim. 62-77; K. 503, mm. 33-48; K. 537, mm. 488/iu1, K. 595/11),”° or alluded to briefly or in 74. One exception, as noted above, is K. 415/111, with 75. The return of the sujet libre TR! in K. 595/111, m.

its expositional restart in m. 65. 208, is complicated further by the fermata-prepared re-

Rondos and the Type 4 Sonata 427

varied fashion after the cadenza, as part of the instances the solo re-entry back-reference could

coda-rotation (K. 414/11, K. 482/111). be to something presented in the preceding Prf Interpreting the “new” solo re-entry theme + tutti-extensions, as if the soloist were enterin a Type 4 concerto finale as a TR of the sujet ing into the discussion by picking up an already libre type familiar from Type 5 first movementsis | sounded idea and adapting it in an individuthe most persuasive interpretation, one that also alized way. One of the most obvious of these,

brings clarity to the otherwise numerous pos- already mentioned, is that in Piano Concerto sibilities for misunderstanding the generic reso- = No. 22 in E-flat, K. 482/i11, mm. 74ff, which nances of this theme. To be sure, the re-entry muses reflectively on the rhythm and contour theme might also suggest some allegiances tothe — of the highly stylized, “buffa” or “C-character”’ S1:\Ppreface possibility within Mozartian Type 5s. theme that had appeared in the tutti-extensions

That, too, is a “new theme” following an em- atm. 52. phatic, cadential tutti close (in the Type 5 case, More provocative are cases in which the suthe close of Ritornello 1). But as will be out- —_fet-libre ‘TR reaches back to resuscitate aspects of lined in chapter 21, a Type 5 S1:\Ppreface leads provocative material from an earlier movement. invariably to an emphatic restatement of R1:\P Reshaping something heard in the first movein the orchestra and the clear start ofa new rota- = ment not only pulls that earlier gesture into a tion (the solo exposition). In the Type 4 concerto finale-context at a crucial structural point, but finale, on the other hand, even where motives it also suggests outer-movement relationships— from the refrain, P™!, might follow fragmentarily contributions to a quasi-cyclic integrity—that after the sounding of the sujet-libre TR (as in K. help us conceptually to bind the entire work to466/iii and 467/ii1, for instance), nobody could gether as a coherent statement. Classic instances sustain the claim that a new rotation was being — occur in Piano Concertos Nos. 20 in D Minor, launched at that point. On the contrary, the sujet K. 466/111, and No. 21 in C, K. 467/111. In the libre’s claim to exist fully in TR-space—continu- former the finale’s solo re-entry, m. 63, reaches ing to pursue an expositional trajectory begun at back to the first movement’s plaintive S1:\Ppreface, the movement’s opening—seems incontestable. m. 77, helping to establish parallels between the

Any Pf fragments that persist or reintrude after two movements. In the latter the finale’s sujet-li“new-theme” solo re-entry—either in the or- bre TR, starting in m. 58, shoots forth with the chestra or in the soloists’ part itself—mark a shift | same upward-striving rhythm as that found in a in strategy from a largely independent transition § famously surprising moment of the first move(the sujet libre) to a dependent one with back-ref- ment, the sudden G-minor solo entry, S1:\TM!,

erences to earlier material.7® in m. 109, an idea that had appeared only as Finally, it might not be amiss to point out the — a One-time event in that movement.’7 About obvious. Not all of the sujet-libre TR solo themes such possibilities for obvious interconnections are entirely “new” or fully “independent.” As a in other concertos, however, one cannot genresult of the numerous modular interconnections eralize, except to reaffirm the general principle

that Mozart composed into each concerto— that the workings of each of these concertos is contributing mightily to their senses of internal to be explored not only generically, in terms of coherence, individuality, and long-range dra- what it shares with other works of its kind, but matic argzument—many of them have obvious also individually, in terms of its own patterns of resonances with material already heard. In some memorability and uniqueness. currence of the head-motive of Pf on IV, m. 182. Be- 76. Aslightly different view of this 1s provided in Graycause of the similarities of its ensuing procedures with son, Mozart: Piano Concertos Nos. 20 in D Minor, K. 466, those of mm. 131ff (tonic-key P'f recapitulation-gesture and No. 21 in C Major, K. 467, p. 79.

proper leading into development, in the manner of a 77. For more on these interconnections, see again Type 41-«xP), it is probably best to include mm. 182-207 Grayson, Mozart: Piano Concertos Nos. 20 in D Minor, K.

within the sphere of the developmental space as well, 466, and No. 21 in C Major, K. 467, pp. 79-80, 88-89. as a second subrotation (though one with the idea of a On the $1:\TM! status of K. 467/1, m. 109, see chapter

subdominant recapitulation on its mind). 21 below.

428 Elements of Sonata Theory

ESC Issues in Type 4 Sonatas: S theme, and so on) and developments. In others

The Double Perspective the tilt toward the sonata rhetoric is not so pronounced. Whatever the balance between sonata

If the ESC marks the moment of the attainment and rondo, the “sonata” aspect will ask for the of the full reality of the tonic in sonata forms presence of an ESC at the end of the recapitula(a fundamental principle of Sonata Theory out- tion’s S theme. And this is precisely how we have lined at the end of chapter 11), what may be been using that term in the preceding pages: in said of rondo structures? When rondos are not a manner analogous to other sonata types, the intermixed with sonata formats, the issue pre- ESC may be regarded as occurring normally, in sents little difficulty: the corresponding point the recapitulatory space. As mentioned in the of essential structural closure (ESC) must ar- preceding paragraph, however, a simple rondo’s rive with the concluding I:PAC of the final ap- ESC is delayed until the moment of the PACpearance of the rondo theme—with the I:PAC, closure of its final thematic statement—which that is, that concludes its thematic block. Earlier in a sonata-rondo occurs after the recapitulaI:PAC’s in prior rondo-theme appearances will tion is completed. Thus a sonata-rondo presents have proven to be overridden (or merely provi- us with the possibility of two conflicting ESC sional), since they proceeded onward to (usu- claims (figure 18.1).

ally) nontonic episodes. But a sonata-rondo 1s something that by

Sonata-rondo structures, however (Type 4 definition can be viewed simultaneously from sonatas), present a more complicated conceptual the perspectives of two different structural situation. As hybrid forms they can be viewed principles. Viewed through the rondo lens, the from two different perspectives: from that of the final statement of P*f is part of the form: it exsonata and that of the rondo. In some sonata- ists unequivocally within rondo-space and may rondos the “sonata” aspect strongly outweighs followed by a coda (outside of rondo-space). that of the rondo in structural (governing) im- Viewed through the sonata lens, through, the portance. These would include compositions final Pf stands outside of sonata-space: it initiwith sonata-like melodic material, thoroughly ates the (possibly lengthy or discursive) coda, a elaborated exposition- and recapitulation-lay- parageneric space within a sonata (chapter 13). outs (including transitions, a medial caesura and In this sense two ESCs might be said to co-exist,

mS % Sonata Aspect

Exposition Development ] Recapitulation

P™TR > S / C pt Va P@? TR ? S / C_ Prt /

a (nontonic) ESC (in I) ESC? Rondo Aspect

FiGurE 18.1. The Type 4 Sonata: A Doubled ESC?

Rondos and the Type 4 Sonata 429

although depending on the degree of “sonata- pitulatory rotation. In that sense, only the final ness” of the individual piece (its tilt—or lack of — statement of the rondo theme (P')—following it—in the sonata direction) we might argue that the tonic-securing ESC of the recapitulation— the two ESC moments might not be understood is fully sounded “in” the stable tonic key. All

as equally weighted in the overall structure. prior statements of P'f (rondo) have been suborThis formal ambiguity may have important dinated to other tonal and musical processes that hermeneutic ramifications. Obviously, more follow it. Here in the coda alone can it exist on sonata-like pieces will favor a heavier weight- its own, in a tonic that has, finally, been fully ing of the sonata’s ESC-claim; more rondo-like validated by the preceding musical processes. pieces will place their conceptual weight more With its own PAC-closure (the “rondo’s” ESC) on the rondo’s ESC-claim. Viewed from only it completes the tonic-stabilizing process of the the sonata perspective, the final rondo statement movement from the point of view of the rondo. is reduced to the post-sonata-space function of a And, of course, its own coda may follow. thematic “discursive coda” following the reca-

CHAPTER NINETEEN © "ISLE 0

The Type 5 Sonata Fundamentals

Aw feature of many concerto move- found in other sonata types. It sets up and then ments, the Type 5 sonata combines ritor- gives way to a solo entry that normally launches nello formats and procedures passed down from a sonata-form oriented structure, one punctuearlier eighteenth-century concerto and aria ated and framed by additional, reinforcing ortraditions (dramatized tutti-solo alternations) chestral appearances, often, of appropriate porwith aspects of sonata form. While this concer- tions of the initial ritornello.

to-sonata blend includes a number of identify- Making use of traditions grounded in the ing features throughout its structure, its most history of the genre, this sonata type is not to prominent difference from the other sonata be derived exclusively from sonata practice (as types lies in its quasi-introductory opening sec- a variant of it), as was generally supposed some tion, an anticipatory rotation of thematic mod- decades ago. On the contrary, the historically ules that precedes the onset of the sonata adapta- separate ritornello formats of earlier concertos, tion proper, which may then expand, recast, or especially around the middle of the eighteenth otherwise react to those modules in engaging century, were instead increasingly informed by ways. This first section is an initial orchestral formal layouts characteristic of the new symritornello of varying length and complexity. Un- phonic writing of the period. (The same may like a sonata exposition (which in other respects be said of eighteenth-century opera seria arias, it often resembles), it usually begins and ends which at least through the 1770s followed much

in the tonic, and it often remains in the tonic the same historical path.)! All who have dealt throughout. This opening, tonic-centered tutti with the formal structures of concertos in the is an important “extra” in the ‘Type 5 sonata not period 1730-1820 are aware of the complexi-

1. See Martha Feldman, “Staging the Virtuoso: Ritor- applied to Mozart, however, is that the formal similarinello Procedure in Mozart, from Aria to Concerto,” in ties entailed affect primarily seria arias before 1780, not Mozart’s Piano Concertos: Text, Context, Interpretation, ed. buffa arias, nor the majority of seria arias after 1780. As

Neal Zaslaw (Ann Arbor: University of Michigan Press, the century progressed, the aria and the concerto in1996), pp. 149-86. In another essay from the same vol- creasingly diverged” (Webster, “Are Mozart’s Concerume James Webster provided an important nuance (in tos ‘Dramatic’? Concerto Ritornellos versus Aria Intropart taking issue with remarks by Charles Rosen): “A ductions in the 1780s,” Mozart’s Piano Concertos, p. 109. serious problem with the aria-concerto hypothesis as

430

The Type 5 Sonata: Fundamentals 431

ties—and the perils—involved in writing even exemplary case is the Adagio of Mozart’s Viothe most basic things about this topic. If sonatas lin Concerto No. 51in A, K. 219. Additionally, in general (Types 1-4) present us with challenges of Mozart’s seventeen Viennese piano concerof understanding, the concerto-sonata combi- tos, from No. 11 in F, K. 413 through No. 27 nations typical of later eighteenth-century con- in B-flat, K. 595—around half are cast in Type certo first (and sometimes second) movements 5 formats, sometimes without a developmenredouble those challenges, seeding the field of tal space.) Ritornello formats in concerto first analysis with conceptual and terminological and second movements may be in dialogue with landmines. Merely to select a label for describ- Type 1, 2, or 3 sonatas. The Type 5 adaptation ing certain events or zones (“tutti,” “ritornello,” of the Type 1 sonata (no developmental space) “transition,” “secondary theme,” “exposition,” is found primarily in slow movements: one lo“recapitulation,” “episode,” “sonata form,’ and cus classicus is the Andante cantabile of Mozart’s so on) is to wade into a morass of previously es- Violin Concerto No. 41n D, K. 218.4 The infretablished connotations, each of which has been quent adaptation of the Type 2 sonata (“binary” ardently defended and just as ardently opposed, without a full recapitulation) may be found in particularly within music-historical scholarship the first movement of the same concerto.° Most

of the last forty years. often, especially in first movements, we are

The first movements of later-eighteenth-cen- dealing with the Type 3 version of the Type tury concertos are almost invariably built around 5 sonata. This features a structural merger of the Type 5 idea. It also shows up in two of Mo- the ritornello principle with the “textbook” sozart’s early concerto finales (Piano Concerto nata, including an exposition (without repeat), No. 5 in D, K. 175/111 [original finale], and Vi- a developmental space, and a full recapitulation. olin Concerto No. 1 in B-flat, K. 207/111) and Ritornello blends with the Type 4 sonata (soin two interior movements of each of the two nata rondo) are also encountered—in the ingeSerenades in D, K. 203/11 and iv, and K. 204/11 nious finales of Mozart’s concertos, for example: and 111—a slow movement and a fast movement an overview of these has been provided toward in each case.? Examples of it are also encoun- the end of chapter 18.°

tered in many concerto slow movements. (An To complicate the matter further, the most 2. Cf. the exceptional—and anomalous—Type 5-Type Mozart’s Piano Concertos, pp. 112-14. Using a different 3 hybrid in the Andante from the Serenade in D, K. set of definitional guidelines, Webster classified the slow 185/i1. This is a Type 3 sonata (with repeats!) that fea- movement of K. 467 as being in “concerto/rounded bi-

tures a solo violin and that includes an eleven-bar open- nary” form, while he judged that of K. 449 to be “a ing ritornello and a concluding ritornello with interior more or less unclassifiable ritornello-rondo hybrid.”

cadenza. Most recently, John Irving, Mozart’s Piano Concertos (Al-

3. These include the slow movements of: No. 11 in F, dershot: Ashgate, 2003), pp. 66-69, considered what K. 413; No. 12 in A, K. 414; No. 13 in C, K. 415; No. we regard as Type 5 slow movements to be relatable 14 in E-flat, K. 449; No. 17 in G, K. 453; No. 19 in primarily to vocal music formats, that is, to “ritornello F, K. 459; No. 21 in C, K. 467; and No. 25 in C, K. (aria) forms.” 503. Four of these, K. 414, K. 449, K. 453, and K. 467 4. For other examples see n. 3. are Type 5s in dialogue with Type 3 guidelines—that 5. Other examples of Type 2 adaptations within Type is, with a developmental space. (K. 449 and 467 also 5 sonatas include the first movements of J. C. Bach’s feature unusual tonal plans at various points in their Keyboard Concerto in F, op. 7 no. 2, Mozart’s pastiche structures. For K. 449 see Hepokoski, “Back and Forth Piano Concerto in G, K. 107 no. 2, and the Andante from Egmont: Beethoven, Mozart, and the Nonresolv- moderato of Mozart’s Serenade in D, K. 204/11. The ing Recapitulation, 19th-Century Music, 25 (2001-2), first movement of the Piano Concerto No. 24 in C Mi149.) Four others, K. 413, K. 415, K. 459, and K. 503 nor, K. 491, also features some Type 2 aspects. Joel Gaare Type 5s in dialogue with Type 1 guidelines—that is, land, “The Large-Scale Formal Role of the Solo Entry without a developmental space. It might be added that Theme in the Eighteenth-Century Concerto,” Journal of the slow movement of No. 16, K. 451, is a Type 5 vari- Music Theory 44 (2000), p. 401, also cites examples from ant of a rondo: although its initial section does unfold in Johann Samuel Schréter’s op. 3.

the manner of a sonata exposition, however, there is no 6. Another discussion of these movements was under-

tonal resolution of the presumed S material. taken in “Movement Forms III: Finales” in Irving, MoSee also the discussion and table in James Webster, zart’s Piano Concertos, pp. 73-92. “Are Mozart’s Concertos ‘Dramatic’?,” in Zaslaw, ed.,

432 Elements of Sonata Theory

impressive examples at hand, some three dozen labeling systems to register the disappearances, or more mature concertos by Mozart, present us reappearances, and rearrangements of the mulwith his most intricate formal layouts. And each tiple thematic modules juggled about in suclayout is strongly individualized, the result, it ap- cessive tutti and solo passages. Merely keeping pears, of a playful approach to the formal possi- track of what happens to the thematic and texbilities of concerto form. In Mozart (and others) tural ideas as the concerto movement proceeds we typically find unique “solutions” tailored to is a complicated business. There 1s no avoiding individual concertos—a differently played chess such complexity of description. As will be evigame each time—even while other broad struc- dent in what follows, we, too, shall not be able tural guidelines are held relatively constant. This to avoid it. makes generalizing over this repertory difficult. Some of the most often-cited writing (SteIt may be that the earlier, flexible traditions of | vens, Davis) has examined the matter historically the Baroque concerto ritornello movement led by compiling relevant discussions of the topic by to the continuing formal freedom of concerto eighteenth- and early-nineteenth-century comforms in the last half of the eighteenth century. mentators and theorists, enriched with synoptic Even while acknowledging the many elasticities overviews of formal plans of concertos besides within ‘Types 1-4, it does appear that the Type those of Mozart (for example, those precedents

5 sonata was given an even freer formal rein, found in C. H. and J. G. Graun, J. Stamitz, encouraging a “looser” approach to individual C. P. E. Bach, C. F. Abel, J. C. Bach, and oth-

realizations of structure. ers).’ Much of the music-historical work has

Often centering around the Mozart concer- sought to displace ahistorical assumptions emtos, the bibliography on this topic is large. The bedded in previous habits of concerto descripliterature is riddled with differently formatted tion, especially those grounded in nonprobdiagrams of standard and individualized con- lematized sonata-form terminology. Related certo procedures; with contrasting descriptions writings (Leeson and Levin, Kiister, Forster, and interpretations of intricate structural layouts Brtick, Webster, Berger, Galand) have proand “standard operating procedures”; and with vided statistical or analytical overviews of stanconvoluted numerical or alphabetical (or both) dard practice in Mozart’s concertos.® Others 7. Jane R. Stevens, “An 18th-Century Description of “H. C. Koch, the Classic Concerto, and the SonaConcerto First-Movement Form,” Journal of the Amer- ta-PForm Retransition,” Journal of Musicology 2 (1983), ican Musicological Society 24 (1971), 85-95; Stevens, 45-61; and Davis, “C. P. E. Bach and the Early History

“Theme, Harmony, and Texture in Classic-Romantic of the Recapitulatory Tutti in North Germany,” in C. Descriptions of Concerto First--Movement Form,” Jour- P. E. Bach Studies, ed. Stephen L. Clark (Oxford: Clarnal of the American Musicological Society 27 (1974), 25—60; endon Press, 1988), pp. 65-82. See also the useful sur-

Stevens, “Formal Design in C. P. E. Bach’s Harpsichord veys by Michael Talbot, “The Instrumental Concerto: Concertos,” Studi musicali, 15 (1986), 257-97; Stevens, Origins to 1750,” and Cliff Eisen, “The Classical Pe“Patterns of Recapitulation in the First Movements of riod,” within the general entry for “Concerto” in The Mozart’s Piano Concertos,” in Musical Humanism and Its New Grove Dictionary of Music and Musicians, 2nd ed., ed.

Legacy: Essays in Honor of Claude V. Palisca, ed. Nancy Stanley Sadie and John Tyrrell (London: Macmillan, Kovaleft Baker and Barbara Russano Hanning (Stuyve- 2001), 6:242—46 and 246-51. sant, N.Y.: Pendragon, 1992), pp. 397-418; and Stevens, 8. The classic essay in English is Daniel N. Leeson and

“The Importance of C. P. E. Bach for Mozart’s Piano Robert D. Levin, “On the Authenticity of K. Anh. Concertos,” in Mozart’s Piano Concertos, ed. Zaslaw, pp. C 14.01 (297b), a Symphonia Concertante for Four 211-36 (in which Stevens characteristically noted, p. Winds and Orchestra,” Mozart-Jahrbuch 1976/77 (Kas212, that she has “profound objections (some of which sel: Barenreiter, 1978), pp. 70-96. This was preceded are shared by many others) to this [sonata-form-based by Levin, “Das Konzert ftir Klavier und Violine D-Dur sectional] framework,” which by 1996 she assumed to KV Anh. 56/315f und das Klarinettenquintett B-Dur,

be so discredited as to decline to “take time to beat KV Anh. 91/516: Ein Erginzungsversuch,” Mozartthis already rather feeble horse”). Somewhat related Jahrbuch 1968/70 (Salzburg: Internationale Stiftung Moin style and intent are: Edwin J. Simon, “Sonata into zarteum, 1970), pp. 304-26. See also the three monoConcerto: A Study of Mozart’s First Seven Concer- graphs: Konrad Kiister, Formale Aspekte des ersten Allegros tos,” Acta musicologica 31 (1959), 170-85; Shelley Davis, in Mozarts Konzerten (Kassel: Barenreiter, 1991); Robert

The Type 5 Sonata: Fundamentals 433

(Charles Rosen) have continued the Toveyan the EEC and ESC in a Type 5 sonata, concepts style of commonsense overview coupled with central to the other sonata types. always-provocative aesthetic pronouncement.? Still others, seeking to offer syntheses of prior scholarship along with individual interpretations Historical Overview and Initial Questions

and commentary, provide studies of selected of Terminology works.!° Most recently, William E. Caplin has provided a review of Mozart’s concerto practice Like ‘Type 4s, Type 5s are hybrid constructions. congruent with his own system of Classical Form One aspect of Type 5 structure looks back to the (1998), while John Irving has devoted an entire older Baroque tradition, which exerted demands book to a broad consideration of Mozart’s Piano on the architecture separate from those pressed Concertos (2003).!! Perusing the literature on — by sonata form proper. No one should confront concertos, one soon learns not only that there — the Type 5 sonata without a sufficient historiare no simple ways to track through this topic cal consciousness of the origin of the concerto but also that there are no disciplinarily neutral | and an awareness of earlier formal patterns and

systems of description. their continued claims in the later eighteenth

This and the following three chapters do not century. For this reason, we need to spend some present a full account of concerto structure. Our time summarizing the literature on this topic. plan is more limited. Keeping our focus on the By the early eighteenth century the concerto concertos of Mozart, with a few glances forward idea was tied up with the principle of alternatto Beethoven, we offer an interpretation of their —_— ing tutti-solo contrasts: the dramatized juxtapo-

analytical features from the perspective of So- sition of a group and an individual (or smaller nata Theory. Our interests foreground three = group) emerging out of that group. From Torelli problems, which may not always be kept sepa- and Vivaldi onward, this ritornello principle rate: (1) that of coming to a broad understanding —_ typically involved tonic-centered introducof the large-scale structure and its implications, tory and closing ritornellos that enclosed an ofwith an emphasis on the historical residues of | ten-modulatory inner series of lightly accompatutti-solo alternation as it secks also to unfolda _ nied, virtuosic solo passages. Each of these was sonata structure; (2) that of rotational and ref- affirmationally punctuated by a briefer, fullerential layout (the relationships among the the- orchestra tutti (usually elided with emphatic matic modules of the initial ritornello, the ex- solo cadences) that usually reanimated selected position, and the recapitulation); and (3) that of | modules from the initial ritornello.' Forster, Die Kopfsatze der Klavierkonzerte Mozarts und in Sonata Forms, rev. ed. (1988; orig. 1980), pp. 71-97. Beethovens: Gesamtaufbau, Solokadenz und Schlupbildung Here the most relevant backdrop is Tovey’s essay from (Munich: Fink, 1992); and Marion Brtick, Die langsa- 1903, “The Classical Concerto,” Essays in Musical Analmen Sdtze in Mozarts Klavierkonzerten: Untersuchungen zur ysis, vol. 3, “Concertos” (London: Oxford University Form und zum musikalischen Satz (Munich: Fink, 1994). Press, 1936), pp. 3-27, which established some positions

Cf. the cumbersome insistence on retaining late-eigh- with regard to the concerto often reiterated or adapted teenth-century terminology carried out—then fused in English-language musical writing. with philosophical reflections—in Karol Berger, “The 10. Recent examples include David Rosen, “The ComFirst-Movement Punctuation Form in Mozart’s Piano poser’s ‘Standard Operating Procedure’ as Evidence of Concertos,” in Mozart’s Piano Concertos, ed. Zaslaw, pp. Intention: The Case of a Formal Quirk in Mozart’s K. 239-59. Webster, “Are Mozart’s Concertos ‘Dramatic’ ?” 595,” Journal of Musicology 5 (1987), 79-90; David Gray-

in the same volume, pp. 107-37 (see n. 1 above), pro- son, Mozart: Piano Concertos No. 20 in D minor, K. 466, vides an English-language study and inventory of slow and No. 21 in C major, K. 467 (Cambridge: Cambridge movements. An inventory of various treatments of an University Press, 1998); and Leon Plantinga, Beethoven’s individual moment within concerto movements from a Concertos: History, Style, Performance (New York: Nor-

music-theoretical point of view is provided in Galand, ton, 1999).

“The Large-Scale Formal Role,” 381-450. 11. Caplin, ch. 17, “Concerto Form,” in Classical Form, 9. Charles Rosen, “The Concerto,” in The Classical pp. 243-51; Irving, Mozart’s Piano Concertos (see n. 3) Style, expanded ed. (1997, this section basically unal- 12. As Michael Talbot (“The Instrumental Concerto: tered from the 1971 edition), pp. 185-263; “Concerto,” Origins to 1750,” in the entry “Concerto,” The New

434 Elements of Sonata Theory

For the first decades of the “Vivaldian” con- For some writers over the past decades it has certo, the number of tutti and solo passages in been a point of honor to confront these strucany ritornello-based movement was not stan- tures—all the way through Mozart and bedardized. Toward the second third of the eigh- yond—primarily in late-eighteenth-century teenth century the concerto gradually came to terms, that is, with a suspicion of, if not a conbe informed by aspects of the emerging galant tempt for, “modern” sonata-form terminology. style and by the sonata plan used for sympho- Defamiliarizingly “historical” descriptions of nies and sonatas. In part this was due (as Cliff — these works are readily found in the musicologiEisen put it in the recent New Grove entry on the cal literature. Their positive effect is to remind concerto) to “the surprisingly rapid replacement us how these concertos might be addressed usof the concerto by the concert symphony as the ing only surviving late-eighteenth-century de-

dominant orchestral genre in the middle of scriptions of form.!® Still, certainly by the conthe 18th century.”!? This slow acclimatizing of | certos of Mozart—where convergences with the concerto to what we now call sonata form is the sonata-form ideas found in his other works a complicated matter. The process of rapproche- could hardly be more apparent—such aversions ment remains a contested topic within musico- to later, more efficient terminology become ex-

logical research.'4 cessive and counterproductive. To be sure, we Considered generally, we may sce the ritor- should never marginalize the few eighteenthnello-solo traditions being bent incrementally century and early-nineteenth-century written in the direction of the symphonic. The figures views of the concerto that we have—Vogler, in the center of this historical process (C. P. E. Koch, Kollmann, Galeazzi, and so on. NeverBach, C. H. and J. G. Graun, the Mannheim theless, the recent riposte of David Rosen to any composers, J. C. Bach, and others) furnish methodologically hobbled insistence on limiting examples of this ongoing historical conver- ourselves to eighteenth-century, “historical” gence, which is anything but linear: “old fash- terminology cannot be improved upon: “Moioned” concertos, without much sonata-form zart would not have recognized the sonata-form influence, occasionally cropped up as late as the terminology, but we need not make the usual 1770s—such as Josef Mysliveéek’s violin con- ritual apology for that (or worse, arbitrarily to certos, which “may have influenced Mozart, invent new terminology): I suspect that it would who became acquainted with them in Vienna have taken less than ninety seconds to explain it in 1773.”'!> The most telling syntheses of tradi- to his satisfaction.” !”

tional ritornello and sonata-form practices oc- Mozart’s concerto-sonata syntheses were curred in the hands of Mozart, who balanced continued by Beethoven and others. Eventuthese complementary force-fields in ever more ally, with Mendelssohn especially, the initial

monumentalized ways. ritornello of the Type 5 concerto came to seem Grove Dictionary, 2nd ed., 6: 243) recently summarized mains an insufficiently explored area. Unlike the symone common type of Vivaldian practice from op. 3 (1711) phony, the concerto did not adopt sonata form but inonward, “the ritornello—one or more ideas constituting stead continued in the second half of the century to rely a refrain played by the full ensemble—is used to establish on its tried and tested ritornello form, although certain the opening tonality and subsequently to affirm the vari- increasingly common features such as the reprise of the ous other tonalities reached in the course of the move- material of the first solo towards the end of the move-

ment; the alternate sections (episodes), scored for the ment are evidence of convergence between the two solo instrument with a generally light accompaniment, forms. In fact, the division between Baroque and Classiaccomplish the structurally important modulations and cal is invisible, structurally speaking, in the concerto.”

supply contrasting themes or figurations.” 15. Eisen, “The Classical Period,” 6:247. 13. Eisen, “The Classical Period,” in “Concerto,” The 16. The ne plus ultra is Berger’s Koch-based “The

New Grove Dictionary, 2nd ed., 6:246. First-Movement Punctuation Form in Mozart’s Con14. A still-conflicted middle ground is suggested in certos.” (See n. 8 above.) Eisen’s summary (“The Classical Period,” 6:246): “In- 17. David Rosen, “’Unexpectedness’ and ‘Inevitability’ deed, the nature of the relationship between sinfonia in Mozart’s Piano Concertos,” in Mozart’s Piano Concer-

(symphony) and concerto between 1700 and 1750 re- tos, ed. Zaslaw, p. 281 n. 2.

The Type 5 Sonata: Fundamentals 435

redundant, old-fashioned, something that had tween the first and second parts the instruments outworn its original raison d’étre. With its exci- execute a prelude, postlude, and interlude [Vor-, sion, what had been the favored format for con- Nach-, und Zwischenspiel].'® certo first movements—the Type 5 sonata—col-

lapsed into the Type 3 pattern. At this point the Here Vogler addressed the matter in terms absorption of the concerto idea into sonata form of compositional practice. By suggesting that became complete. The history of the concerto the solo passages were to be composed first, he in the eighteenth century and beyond, develop- implied that this gewéhnliche Sonate (our Type ing alongside the symphony, is that of gradually | 3 Sonata) was the most essential part of the being attracted to the latter’s principles, finding movement. The ritornellos were afterthoughts, ways of adapting itself to them while retaining add-ons, mere framing mechanisms for a more important features of its own identity, but even- telling, often pre-existing structure, though one tually (around the fourth decade of the nine- now shorn of its customary repeats. (Especially teenth century) succumbing rather totally to since Koch in 1793 shared this view—and since them. Later concerto reinstatings of the open- — YOUNg Mozart seemed to put it into practice a ing ritornello, as in Brahms’s concertos, are best. f€w times—this is a point to which we shall reregarded as archaizing or retrospective efforts, turn.) Along the same lines, Vogler also advised recalling largely eclipsed traditions of enhanced that the opening prelude—which he told us al-

monumentalization. most always began and ended in the tonic!?—be kept less “beautiful,” and by implication shorter, than the first solo (the exposition of the gewéhn-

Vogler, Koch, and Others: liche Sonate) in order not to draw attention frivo-

Six Tutti-Solo Layouts lously from the latter, the principal item of interest.70

Vogler was describing a five-part structure.

Vogler’s Description (1779) of the Concerto At least preliminarily, we may schematize it Because of the persisting imprint of the older (and interpolate the more common term “ritorBaroque and midcentury traditions in Mozart’s nello,” though “tutti” would work as well) as concertos—our main concern here—t 1s help- follows: ful to look at the historical state of the genre as that composer may have found it. There is no Ritornello 1. Orchestral prelude, tonic.

better place to start than with one of the ear- Solo 1. Sonata exposition, modulating to and liest descriptions of the increasing interaction confirming a secondary key. between the ritornello-movement concerto and Ritornello 2. Orchestral interlude, “a short excerpt of the first” ritornello.

sonata form. In 1779, in a now-famous passage Solo 2. Sonata development and recapitulation. from his periodical, Betrachtungen der Mannheimer Ritornello 3. Orchestral postlude. Tonschule, Georg Joseph Vogler provided a sim-

ple recipe for producing a concerto: Implicit in such a description, to judge from the musical literature, is that Ritornello 1 is norWhoever wishes to compose a concerto does well mally the longest of the three and was usually if he first writes [macht] an ordinary sonata [eine separated from Solo 1 by a I:PAC and clear caegewohnliche Sonate|. The first part of it is used for sura. On the other hand, Solo 1’s and Solo 2’s

[giebt] the first solo, the other part for the second ,

solo. Before the first, after the second [and] be- concluding PACs were generally elided with the ritornellos that followed them. In all cases, the

18. Vogler, Betrachtungen der Mannheimer Tonschule, H, in Stevens, “Theme, Harmony, and Texture,” p. 33 n.

36. The translation is adapted from Stevens, “Theme, 25.

Harmony, and Texture,” p. 33. 20. Vogler, quoted in Stevens, “Theme, Harmony, and 19. “Only very seldom do those [preludes] occur that Texture,” p. 33 n. 24. conclude on the dominant,” Betrachtungen, 11, 38, quoted

436 Elements of Sonata Theory

overall sonorous impact of the later ritornello los, could be treated freely in individual comentrances was aptly described well over a cen- positions—via occasional suppressions, displacetury later by Tovey, who in 1903 laid out an es- ments, unusual textural treatments, “extra” tutti sential feature of what he was calling the “con- interjections, and so on.

certo principle”: “The solo is probably more Adapting the work of Stevens, Davis, and active, as well as more personal and eloquent, Galand, we identify six heuristic subtypes that, than the orchestra, and can therefore make a with some flexibility, are commonly found in brilliant climax if it chooses; but 1t cannot make mid- and late-eighteenth-century Type 5 sonaits climax very powerful in sound as compared tas. These are laid out, with particular reference with what the orchestra can obviously do with to Mozart’s works, in table 19.1 as Subtypes A— ease; and so this one missing element may be F.23 (What may seem to be the unusual numsupplied, and the design rounded off, by bring- bering of the ritornellos—with some additions ing in the ritornello forte on the last note of the and deletions from strict integer sequence—will

solo, thus ending the piece.’”?! be explained in due course.) Of these, the two seven-part (four-ritornello)

Koch (1793) and the Two Four-Ritornello formats (A and B in the table) are particularly

Layouts: Subtypes A and B . ,

important. Not only are they are the ones most frequently encountered in Mozart’s concertos,

Vogler’s five-part (three-ritornello) plan was but one of the two plans, famously—our Subonly one of several that had sprung up in the type A—was also described by Koch in his Vermid-eighteenth century, and, if we take Vo- such of 1793—one grounded, he mentioned, in gler’s description literally, it was by no means C. P. E. Bach’s concerto practice.24+ Koch’s exthe most common subtype. Sonata-like proce- planation provided the basis for much revisiondures could be combined with concerto pat- ist scholarly work on the concerto toward the terns featuring three, four, or even five ritor- end of the twentieth century, and it described nello appearances at major structural hinges the older of the two seven-part plans. As Davis within the movement—something onthe order has noted, Koch’s Versuch model most accurately of ritornello pillars,?? identifying the movement corresponds to the pattern found especially in

emphatically as a concerto. These pillars help many of C. P. E. Bach’s earlier concertos— to precipitate and intensify important struc- those of the 1730s and 1740s. “Actually, Bach tural stations in the form. The most important favoured several formal plans; however, the point is the adaptability of the whole system. one that Koch described was used with relative Even while today we may extrapolate certain consistency by Bach and was also the one most structural paradigms as useful heuristic models, frequently employed in the general repertory it is Wiser to suppose that eighteenth-century bridging the generation between C. P. E. Bach composers operated under the assumption that and Mozart.”?° It is particularly characteristic, orchestral tuttis, especially the internal ritornel- for instance, of the Mannheim composers and 21. Tovey, “The Classical Concerto,” pp. 9-10. over some of Galand’s tabular conventions, which he 22. “Pillars” is taken from James Webster, “Are Mo- described on p. 399: “The parenthesized Roman nuzart’s Concertos “‘Dramatic?’” in Mozart’s Piano Concer- merals indicate variants not specified by Koch but often

tos, ed. Zaslaw, p. 111: “The later ritornellos are stable encountered in C. P. E. Bach’s concertos, which Koch pillars, confirming the structural cadences that end each regarded as paradigmatic. For the five other concerto of the major solo sections.” As will emerge, Webster types in the table, only major-mode plans are indicated, and others (including Leeson-Levin and David Rosen) and subsidiary keys are designated ‘x.’” do not consider one of our frequently invoked “pillars,” 24. Two useful introductions to Koch’s view of the just before or at the onset of the recapitulation (our R3), concerto are Stevens’s now-classic “Theme, Harmony,

to be a ritornello. and Texture” (n. 6) and Irving, Mozart’s Piano Concertos,

23. The principal model for the diagrams, on which our pp. 1-16 (n. 3) differing (and somewhat reordered) versions are based, 25. Davis, “C. P. E. Bach and the Early History of the is in Galand, “The Large-Scale Formal Role of the Solo Recapitulatory Tutti,” pp. 65, 66. Entry Theme,” pp. 400—401. In table 19.1 we also take

TABLE 19.1 Six Subtypes of the Eighteenth-Century Type 5 Sonata, with Particular Reference to Mozart’s Concertos (Adapted from Galand 2000: 400-401)

Subtype A Seven-part (four ritornello) format: cf. Koch’s 1793 concerto plan. Koch: S1 material; (in later Mozart: R1, S1 material to end of R4)

R1 S1 R2 S2BO R3 S3 R4 soloist BO BO orchestra 3 ; a —_ cadenza —

I I ——V Vv V — vi vi ——— I I I (ii, iii)

I i —(v,Illiv)Il(III, lil—— iv) v v ——!1 1 1

AA soloist BO BO bo

Subtype B

Seven-part (four ritornello) format: variant of Koch 1793: (vestigial?) “Ritornello 3”-effect launches the tonic return (=recapitulation); the most common format in Mozart’s mature concertos. (Material from R1 and $1)

R1 S1 R2 S2 R3>_ “83 R4

orchestra 3 ptV ;Vv —_ I I ——V —cadenza x —— 1— II

Aa soloist BG FT

Subtype C

Three-ritornello variant, suppressing or minimizing the potential tutti-effect (““R3” of Subtype B) at the tonic return; cf. Vogler 1779, Koch 1802; Leeson-Levin 1978. (Material from R1 and $1)

R1 S1 R2 S2 [“*S3”] R4

orchestra _ pt —_ cadenza — I | —V Vv Vv — x ——— | if I

TABLE 19.1 (continued)

Subtype D

a

A

Nine-part (five-ritornello) format, with an interior ritornello in the developmental space; “C. P. E. Bach’s four-solo plan.”

(Material from R1 and S1)

R1 S1 R2 S2.1 R2.2 $2.2 R3 => 83 R4 soloist po Ho Po pS orchestra i io / ‘cadenza —

I | —— V Vv Vv — x x —— | I I Subtype E

Type 5 adaptation of the Type 2 (“binary”) Sonata (in this variant, suppressing a clear tutti-effect around the area of the tonal resolution).

S1 R2 S2oo => “S3” R4 . R1 tonal resolution soloist — orchestra ___! ot cadenza — Another rotation of material from R1 and S1

I I ——V Vvusually Vvsimilar —— x —— I! I I usually parallel to to the opening the “second half”

of S1 of S1 or R1/S1 synthesis

Subtype F

nen

Type 5 adaptation of the Type 1 Sonata (no development, i.e., suppression of S2); R2 proper may also be suppressed in favor of an “R3”-effect. (Material from R1 and S1)

RI S1 R2 => R3 S3 R4

loi or' retransition only “R3”- § soloist orchestra es a: — [omit cadenza? ] —

I | —V Vv ——— 1 I I usually the same may conclude

as the opening the rotation or

of S1 function as a coda

The Type 5 Sonata: Fundamentals 439

is found in many of J. C. Bach’s concertos.?°® Solo 1. This is a nonrepeated sonata exposition, This “older” format 1s also retained in a handful modulating to and confirming a secondary key. of concerto movements in Mozart. In Koch’s As the first of the three Hauptperioden it has a description three solos (corresponding roughly higher conceptual status than does the opening to exposition, development, and recapitulation) ritornello. At pivotal cadential or caesura points

were set off by four ritornellos: “the melody of the main part 1s sometimes interrupted by the orchestra with short passages, The first allegro of the concerto contains three which consist either of repeated segments of the main periods [Hauptperioden|] performed by the principal melody or of phrases which occurred soloist, which are enclosed [eingeschlossen| by four only in the ritornello.” subsidiary periods [Nebenperioden| performed by

the orenestra as neomnenos: mn mode lout a Ritornello 2. Elided with the closing cadence FOS, The HISE TIFOPNENO IS Benerally WOMKEG OUL at of Solo 1, this shorter ritornello “repeats a few length. It consists of the principal melodic sec,,,; . melodic sections which already were contained tions of the plan [Anlage| of the allegro, which are a brought ht j ininto thea first ritornello andand closes likewise with different connection extended ,

through other means than in the solo of the con- formal cadence in the fifth. certo part. ... Nothing remains to be noted in connection Solo 2. This is equivalent to a sonata developwith the three main periods of the solo part, for ment (“the second main period of the first althey have the same external arrangement | dusser- legro of the symphony”), though often starting liche Einrichtung| and the same course of modula- with a new theme and typically closing with a tion as the three main periods in the first allegro cadence on the minor submediant. the minor of a symphony. The type of melody, on the other mediant, or the minor supertonic. hand, is very similar to that of a sonata. . . .27

Koch also furnished brief d a eth Ritornello 3. This is a short ritornello of retranee SO ISIS OS escripaons oF ene sition that “modulates back into the main key,

harmonic structure of each of the ritornellos . 41: .»

ith hand. A Ith; in which it closes with a [half cadence].” It may

793 © Vartous oe at th - a result As be worth noticing that while Koch mentioned Sah ae ( our-ritorne! 0) ort tetasted that Ritornellos 2 and 4 are normally based on

' ee ; We ma Expansive - na passages from Ritornello 1, he made no such

than Vogler’s. We may summarize Koch’s Ver- remark about Ritornello 3. such plan as follows:

. | . . Solo 3. isThis is equivalent toritora sonata recapitulaRitornello 1. This a third substantial orchestral . . . first i , tion (“the main period of. «the allegro nello “in modern concertos.” Although it re- ,» .

, heidi ; of the symphony”). It typically ends, however,

mans conceptua y a sttpstageny section, merely with an orchestral passage—which Koch, un-

aén Nebenperiode, it does anticipate at least the . . incipal melod; ons” of th b like most modern commentators (as reflected in ee me © The nea © t © - sequent table 19.1) does not consider a separate ritor-

S© ° exposition. © Openings ritorne ° 's fon- nello—that leads to a held $ chord and fermata,

ic-centered: .it. lody mayinbe sounded entirely in the h he solo; f d aN the whereupon the soloist performs a cadenza: “At a or may nn an. eee hen bey © h the caesura tone of this period [= the final caomunant (a cantabile seconcary theme), with or dence of the third solo], the ripieno parts usu-

a ally introduce, by means a few measures, in either case the willof be to . s» a . fermata onritornello the six-four ofled theback keynote.

without a formal cadence in that key—although . the tonic for a conclusion.

26. Davis, “C. P. E. Bach and the Early History of the troductory Essay, trans. Nancy Kovaleff Baker, pp. 210,

Recapitulatory Tutti,” pp. 70-71. 211. 27. Koch, Versuch einer Anleitung zur Composition [1793],

3:333, 336. The translation is that found in Koch, In-

440 Elements of Sonata Theory

Ritornello 4. “With the caesura tone of this ignition for Koch’s third Hauptperiode, the recaso-called cadenza, which always ends with a __ pitulation. Normally, the soloist joins in after formal cadence, the last ritornello begins. This a few bars, perhaps in a second or third phrase generally consists of the last melodic sections of | or module: we label this merger with Solo 3 as

the initial ritornello, with which the entire first R383. This “recapitulatory tutti” procedure,

allegro concludes.”?8 far and away the norm in Mozart, is presented When confronting Mozart’s or Beethoven’s as Subtype B in table 19.1, the second of the two

concertos with Koch’s 1793 descriptions in seven-part (four-ritornello) formats. mind, one must provide several nuances and qualifications. The most Important thing 1s The Problematic Status of the Recapitulatory

. . . . Tutti, R3 (R3=583)

to avoid falling into the historicist trap of re- . garding Koch’s outlines as a sufficient guide to

the examination of these much more complex There has been some debate in recent decades works. Koch’s simplistic descriptions and Mo- about whether the R3 in the Subtype-B case zart’s compositional depth and polish operate —_‘ just mentioned—or even in the preceding Subon vastly disparate conceptual levels. We concur type-A situation described in Koch’s Versuch—is with John Irving’s recent conclusion about this a genuine “ritornello” or only a “tutti interjec-

matter: “Reflecting on Koch’s detailed assess- tion” marking the juncture between two juxment of the concerto, one is struck by the gulf —_ taposed solo sections, both because of its occabetween his theory and Mozart’s practice.”?? sional brevity and because it does not reinforce One technical issue concerns the much-dis- a cadential close led by the soloist.59 The frecussed status of the third ritornello (R3), which quently encountered tutti-impulse at the onset

Koch situated at the end of the development as of the recapitulation, to the extent that it 1s a a passage of retransition. In fact, the composi- ritornello at all, is substantially different in functional treatment of this brief ritornello was vari- tion and effect from R2 and R4, both of which able. At times it does take on this retransitional solidify strong perfect authentic cadences at the (sometimes modulatory) role, preparing for a end of an extended solo passage. What is needed recapitulation begun by the soloist, as in the is a sense of the variability of the recapitulatory opening movements of Mozart’s first two vio- R3, acknowledging its functional difference lin concertos, K. 207 and 211, and those of the from the other ritornellos and its uncertainty Piano Concerto No. 19 1n F, K. 459, the Horn with regard to genuine ritornello status. Concerto No. 1 in D, K. 412, and the Clarinet One factor involved in this question appears Concerto in A, K. 622 (see chapter 22). More to have been the wish to keep the ritornello often in Mozart’s concerto first movements, portions of the form conceptually separate from though, the retransitional function at the end the three solo sections, within which alone the of Solo 2 1s provided by the soloist. In this case sonata form is regarded as playing itself out. In the third “ritornello”—if we may still regard it part, this decision 1s based on the descriptions as that—1is used to provide an opening charge of Vogler, Koch, and other late-eighteenth-cento the recapitulation with a tutti, tonic restate- tury theorists. In recent decades some writers ment of the original, Ritornello 1 version of — have followed the influential lead of Leeson and the P-idea, or at least its incipit. In this latter Levin, who in 1978 categorized this moment case “Ritornello” 3 has left behind its retransi- as only the onset of the “recapitulation,” which tional function in favor of becoming a point of — they considered exclusively as the third solo sec-

28. Quotations from Koch, Introductory Essay, trans. though he changed a few details of his description—

Baker, pp. 210-12. to be discussed in the text below—the main points of 29. Irving, Mozart’s Piano Concertos, p. 12, where the Irving’s assessment still stand. Cf. Irving on the 1802 idea is developed further. As is well known, Koch revis- Koch, Mozart’s Piano Concertos, p. xv.

ited concerto form, now with an awareness of Mozart’s 30. Cf., e.g., n. 22. concertos, in the Musikalisches Lexikon of 1802, but al-

The Type 5 Sonata: Fundamentals 441

tion, arriving directly after the second without quasi-“da capo” impulse to be at least a ritornelritornello-intervention, thus steering clear of a lo-effect, reanimating the opening bars of R1 confrontation with this spot as a potential ritor- and having a specialized quasi-ritornello funcnello.3! (Cf. Subtype C in table 19.1.) Similarly, tion of calling attention to an important structhe current tendency in Germanic scholarship 1s tural moment in the movement—in this case the that of not granting ritornello status to this mo- beginning of the recapitulatory rotation. (We ment. Robert Forster, for instance, has argued revisit some of these terminological questions in

that because this “recapitulatory tutti” (Re- a subsequent section, “More on Terminology: prisentutti), is not really comparable in function ‘Ritornello’ or “Tutti’?”’) to that which orchestrally extends the soloist’s

final cadences of Solo 1 (exposition) and Solo Tutti-Solo Layouts, Subtypes C-F 3 (recapitulation), it is dominated more by the “recapitulation principle” than by the ritornello It may be this ambiguity surrounding a recapituprinciple.°? (This might involve, however, too latory R3—only a tutti impulse that soon yields narrow a definition of “ritornello.” In pursuit of — to an R3=S3 merger in the opening moments precedents for the recapitulatory “rebeginning”’ of the recapitulation—that led some late-eigh-

of R3, one might recall the possibility of “da teenth-century theorists to appear to agree capo” ritornello-effects toward the latter por- with Vogler’s earlier five-part (three-ritornello, tions of certain Baroque concerto movements, two-solo) plan, which had passed over this posas in the first movements of J. S. Bach’s Vio- sible “third ritornello,” and in which the “seclin Concerto No. 2 in E, BWV 1042, and the ond solo” comprised both the developmental Fourth Brandenburg Concerto, BWV 1049.)°3 space and the recapitulation. Such was the case On the other hand, other writers, doubtless in Francesco Galeazzi’s description of the form with an adaptation of the 1793 Koch in mind, in 1796 and August Kollmann’s in 1799—and do consider this tutti moment to represent a even Koch, in his Musikalisches Lexikon of 1802, third ritornello. For Shelley Davis in 1988, it only nine years after the Versuch (but now with belonged to the “tutti restatement: four-ritor- Mozart’s concertos more in mind), omitted any nello plan” that he found with some regularity discussion of an orchestral ritornello around the in C. P. E. Bach’s and occasionally in J. C. Bach’s area of the recapitulation.°° If so, they were proconcertos—a position also discussed (and prob- viding a different description of our Subtype B

lematized) with considerable nuance by Jane R. in table 19.1. On the other hand, it 1s heuristiStevens in 1996.54 For Karol Berger in 1996, cally prudent to devise a Subtype C, similar to the sometimes-brief tutti impulse at this point —B but with no ritornello claim at the onset of in Mozart’s concertos represented “the vestigial the recapitulation. This provides a more literal third tutti”—a useful turn of phrase—which representation of what these writers claimed. normally begins the “fourth period” (that 1s, Assuming that R3 is not suppressed altogether, the recapitulation) and 1s typically joined to as occasionally in J. C. Bach’s concertos (for inthe third solo.*> For the present, granting the stance, in the first movement of the E-flat Keycomplexity of the issue, we shall consider this board Concerto, op. 7 no. 5)3/and (perhaps) in 31. Leeson and Levin, “On the Authenticity of K. Anh. 34. Davis, “C. P. E. Bach and the Early History of the

C 14.01 (297b),” pp. 90-96; Cf. David Rosen, “The Recapitulatory Tutti in North Germany,” pp. 71-80; Composer’s ‘Standard Operating Procedure,” p. 81, and Stevens, “The Importance of C. P. E. Bach for Mozart’s “’Unexpectedness’ and ‘Inevitability,” p. 281, n. 2. Cf. Piano Concertos,” in Mozart’s Piano Concertos, ed. Za-

n. 22. slaw, pp. 211-36.

32. Forster, Die Kopfsatze, p. 49-50. A similar system 35. Berger, “The First-Movement Punctuation Form,” of categories is found in Kiister, Formale Aspekte, p. 7 in Mozart’s Piano Concertos, ed. Zaslaw, p. 248.

et seq. 36. Stevens, “Theme, Harmony, and Texture,” pp.

33. Other examples in J. S. Bach: the finale of the Sixth 41-43, provides the basic citations from Galeazzi and

Brandenburg Concerto, BWV 1051, and the first and Koch. third movements of the Harpsichord Concerto No. 2 37. Cited by Davis, “H.C. Koch, the Classic Concerto,

in E, BWV 1053. and the Sonata-Form Retransition,” p. 50.

442 Elements of Sonata Theory

the first movements of Mozart’s Violin Con- tutti-effect in the area of the recapitulation. R3 certo No. 5 in A, K. 219, and Piano Concerto normally signifies either a retransitional funcNo. 13 in C, K. 415,°8 today’s analysts might tion, preparing for a solo-launched recapitufind Subtype C to be helpful in cases in which lation, or a Reprisentutti, an orchestral recapthe R3 effect is unusually abbreviated (demoted turing of the opening sounds of R1 to begin

to the status of a mere tutti interjection). the recapitulatory rotation. Our R4 stands for We should add a word about our label- the ritornello that follows the last solo section, ing of the Subtype C diagram and subsequent concludes the movement, and is normally dischemes. Since our goal is to confront the con- vided into two parts surrounding the cadenza, certos of Mozart (as opposed to devising a more R4! and R4*. (Some recent writers, most notaencompassing, umbrella theory intended to bly Leeson and Levin, have categorized these cover their predecessors and contemporaries as as two separate tuttis or ritornellos.)*° If in any well), we adopt as a heuristic standard the most given movement the R3-effect is missing (as in commonly encountered Type 5 format found the Subtype-C format, table 19.1), we still label in their works. We thus propose a generic back- the closing ritornello as R4, even though it is drop of a seven-part (four-ritornello) frame- literally only the third ritornello heard in that work (Subtypes A and B—the latter is much movement. In this case R3 is present only in its more frequent in Mozart—in table 19.1), even absence—something possible only if for pracwhen all seven parts might not be fully articu- tical reasons we take Subtypes A and B as the lated on the acoustic surface of the music. Our norm for labeling. labels R1, R2, R3, and R4 are not connotation- We may deal with the remaining subtypes ally neutral designations. They refer to situated quickly. Found with some frequency in the

functions within the structure. mid-eighteenth century (it has been much noted Our R1 means “the opening tutti,” which in studies of C. P. E. Bach),*! Subtype D 1s a is also always the longest, most complete ritor- nine-part (five-ritornello) format. It is similar to nello. The R2 function is to follow the solo (or Subtype B—Mozart’s norm—except that from smaller) exposition and precede a developmental the perspective of Subtype B an “extra” ritorspace: some writers refer to this with such de- nello has been provided to reinforce a central scriptions as “the “subordinate-key ritornello,” developmental PAC, usually in vi, i111, or 1.47

or the “medial tutti in the secondary key.’ The retransition from this mid-developmental Our R3 label, the most flexible of them, refers ritornello to the Reprisentutti (an R3 function) is to that (vestigial) “ritornello” or pronounced led by the soloist. Since it is normally situated in

38. Discussions of these works are provided in ch. 22. the “final ritornello.” Similar labels are found in David Davis, “H. C. Koch, the Classic Concerto,” pp. 48— Rosen, “The Composer’s ‘Standard Operating Proce49, mentions other examples in J. C. Bach and Johann dure,” and “‘Unexpectedness’ and ‘Inevitability.’” Cf. Samuel Schroeter but also claims that the first move- n. 20 above. For Forster, Die Kopfsatze, pp. 50, 55, the ment of Mozart’s Violin Concerto No. 4 in D, K. 218, remainder of the movement after the third solo section exemplifies Vogler’s 1779 and Koch’s 1803 format. is regarded as a “closing complex” (SchluBkomplex), conWhile R3 1s lacking in that Mozart movement, the sisting of the “cadenza tutti” (Kadenztutti) often lacking whole structure is better regarded as a Type 2 adaptation a clear or genuine “ritornello” function (thus becoming of the Type 5 sonata (Subtype E in table 19.1). a merely “secondary tutti), the solo cadenza itself, and 39. Forster, Die Kopfsdtze, p. 49, “Miutteltutti”; Kiister, the “closing ritornello” (SchluBritornell).

Formale Aspekte, e.g., p. 181, “Mittentutti,”’ Caplin, 41. E.g., in Davis, “C. P. E. Bach and the Early HisClassical Form, p. 248, “subordinate-key ritornello”; tory of the Recapitulatory Tutti,” pp. 67-69; cf Galand, Harold S. Powers, “Reading Mozart’s Music: Text and “The Large-Scale Formal Role,” p. 402. Topic, Syntax and Sense,” Current Musicology 57 (1995), 42. Stevens, “The Importance of C. P. E. Bach,” pp. 8, “medial tutti (partial ritornello) in the second key”; 214-16, touched on the problematics of such a structure,

and so on. and, at least initially, regarded some models in C. P. E. 40. Leeson and Levin (“On the Authenticity of K. Anh. Bach as instances of the four-tutti model with a mere C 14.01 (297b),” p. 96) named these, influentially, as the tutti interjection at the moment of the “return to the “ritornello to cadenza” (or precadenza ritornello) and tonic” (the reprise).

The Type 5 Sonata: Fundamentals 443

the interior of S2, that is, in the developmental of these things are to be expected within the space, midway between what is functionally R2 only modestly constrained aesthetic play of the and R3, this “extra” ritornello can be labeled as craft. The goal of the models is only to provide a R2.2 (“another ritornello, sounded after R2 but starting-point for analysis, a lens through which before the R3-effect’’). It would be followed, the individual realizations may be perceived. of course, by S2.2 (“the resumption of $2, now with retransition function ). This solution leads Overview: Sonata Form, Mozart, and the

to a more general as will emerge in Chap- . Framework op: . .point: . Seven-Part (Four-Ritornello)

ters 21 and 22, brief interior tutti interjections or reinforcements in the exposition and recapit- As outlined above, our practice is to consider ulation, common in many concertos, need not Subtypes A and B as the most normative in Mo-

be provided with R-numbers at all. zart’s works and to use the four ritornello pillars Subtype E shows the Type 2 (“binary’’) vari- as a standard for labeling: R1, R2, R3, and R4, ant of the Type 5 sonata—the one lackinga full with S1, S2, and S3 as extended solo sections berecapitulation.*> Here the R2 function is nor- tween them. The question now becomes, how, mally complete (as is that of R4), and S2 merges in Mozart’s concrete practice, are these zones directly into the crux and tonal resolution, by- placed into a rapprochement with sonata form?

passing any R3 effect. This situation may be First—to consider a position with which described as an S283 fusion, although indi- we do not concur (at least for Mozart’s concervidual circumstances might call for an ad hoc tos)—some writers consider it a convenience, if

explanation. not a historical imperative (interpreting Vogler’s

Finally, Subtype F illustrates the Type 5 ad- and Koch’s words strictly), to grant sonata status aptation of the Type 1 idea, something normally only to the solo sections. On this understanding, found only in slow movements.** In this sub- the ritornello pillars stand as separate conceptual type the movement’s sonata aspects are limited entities, cleanly outside of the sonata-unfolding.

only to an exposition and a recapitulation: the This position construes R2 as a separate ritordevelopmental space 1s suppressed, replaced by nello space that does not participate in the expoonly a brief orchestral retransition (in Mozart, sition—which is thus reduced to only the “solo sometimes additionally decorated by the solo- exposition” of $1. Similarly, the recapitulation ist!). In other words, following Solo 1’s final ca- is restricted to only S3; R4 1s not interpreted as dence, the next ritornello is usually given over participating in the sonata process. Such a view primarily (or exclusively) to the retransitional has its merits: it is understandable and logically function. In this case, it is the R2 function (the consistent on the basis of the theoretical writ“medial ritornello,” preceding the development) ings; 1t is sometimes helpful 1n coming to terms that is lacking, and accordingly we label this with earlier concertos by other composers, such ritornello as R3. In cases in which the ritornello as C. P. E. Bach; it acknowledges the (non-soseems to begin as a genuine extension to Solo 1, nata-form) historical origins of the ritornello then soon to lapse into retransition activity, one procedure, whose residual traditions are clearly

could label it as an R2=>R3 merger. in evidence in the later eighteenth-century and It is worth repeating that the six ritornello early nineteenth-century concerto; and rather formats are presented here only as heuristic than reducing the concerto-movement strucmodels, not as analytical templates for all con- ture to only one thing, it emphasizes the procertos. One should expect to find adaptations of | vocative interplay between sonata and ritornello these models within individual concertos: later | form as central features of this hybrid genre. All ritornello displacements, suppressions, curtailed this is to the good.

allusions, additional tutti or solo interjections, But this viewpoint also has its shortcomings, unusual tutti dialogues with the soloist—all ones that are all the more palpable as we con-

43. Examples are provided in n. 5. 44. Por examples in Mozart, see n. 3, along with the passage in the text to which it refers.

444 Elements of Sonata Theory

front Mozart’s concerto formats. What it argues en route, or that, instead of closing, converts is that it is preferable, when considering the “so- at its end into a setup for what follows.

nata-form” question, to bracket off S1 from ev- Notwithstanding the variability of treaterything that precedes and follows it. To be sure, ment found in R2, what all of this means 1s that when we do this, we find a sufficient and coher- ina Type 5 sonata by Mozart, one encounters ent exposition that exemplifies all of the norms two differing sizes of exposition, depending on associated with that structural section, those that which of the two “bracketing” perspectives one

Sonata Theory describes through its PTR’S/C is attending to. Neither is “the” (only) exposisystem. The problem is that as the Type 5 so- tion, but both are present from different pernata proceeds in real time, S1 does not stand spectives. In Mozart’s concertos, once past the alone. On the contrary, it emerges in a context. opening ritornello in the tonic, S1 lays out a solo Most importantly, that context emphasizes its exposition, characteristically closing with a trill sharing of material with the preceding R1, and cadence. On the other hand, S1 and the brief, this sharing 1s a conspicuous part of S1’s essence. though emphatic R2 (even when R2 1s kept inWith regard to modular material R1 and S1, complete or otherwise problematized) usually while not identical, are interdependent. Because strive to complete a nonrepeated larger exposition.

this modular interdependence usually also en- Again, from the perspective of the solo alone, compasses at least the later portions of R2 (and the solo exposition, S1, is the governing one of consequently brings up the concept of govern- the movement: it presents “everything the soing rotations), it is methodologically inappropri- loist wishes to say” along these lines. But from ate to isolate S1 from the context within which, the more inclusive perspective of the orchestral in significant part, it derives its meaning. complex within which the soloist is embedded, For the moment, we might anticipate some a broader, more inclusive rotational complex 1s conclusions that will be buttressed in later sec- working itself out (S1 + R2, even if kept nontions. While a “bracketed” S1 does indeed fur- closed or incomplete), and it often does so in nish a satisfactory exposition, it is also the case, ways that are in line with the norms of a larger in Mozart’s (and Beethoven’s) concertos, thatif | exposition.*> This double perspective, retaining we were place our brackets differently, encom- the interplay and coexistence of two different passing both S1 and R2, we would frequently generic structural fields, is a crucial aspect of come up with something larger than S1 that also Sonata Theory’s conception of the Type 5 solooks like a satisfactory exposition. But some nata (in Mozart’s hands). We revisit this feature clarification is needed before proceeding further. more closely later in this chapter, “Adapting R2 may arrive ata terminal cadence,in which — J. C. Bach into K. 107,” and in chapter 21.

case its larger function is manifest: it serves as S2 is normally given over to the modulaan expositional appendix to the solo exposition tory developmental space. Harking back to its ori(usually invigorating, toward its end, modules gins in the Baroque concerto, S2 is frequently

from R1 that had not been heard in $1). But in more episodic—or given over to new, virtuoMozart it can also happen that R2 1s rotation- sic figuration—than it 1s developmental in any ally or cadentially incomplete in one way or an- standard sense of the term.*® In Forster’s words, other, perhaps breaking off before its anticipated “In contrast to the concepts of exposition and final cadence or dissolving into a dominant or recapitulation (for the first and third solos), the

other preparation for S2. In the latter cases designation of the second solo as a development the impression given 1s that of a larger exposi- is only really justified in a few cases.’4” Develtional appendix or conclusion that is interrupted opmental spaces given over to “new” material 45. As will be outlined in ch. 21, multimodular R2s within R2, inert>participatory, may be paralleled in often subdivide into a rotationally inert first module or the two halves of R4. two, proceeding directly into a rotationally participatory 46. Hence the aptness of the term developmental space:

set of modules taken from the end of R1 that had not see the beginning of ch. 10. been sounded in S1. The common functional succession 47. Forster, Die Kopfsdtze, p. 49. Cf. Marius Flothuis,

The Type 5 Sonata: Fundamentals 445

are especially common in Mozart’s earlier con- a preparatory # platform and fermata; R4? is the certos. Not surprisingly, the principle of devel- postcadenza portion, usually completing the opment crops up more in the later works, and _ larger recapitulation in a rotationally participaBeethoven’s S2 spaces, as might be expected, are tory manner and sometimes adding on a coda usually more self-consciously developmental. at the end. These Ritornello 4 concepts will be As discussed above, the brief R3 is variable more fully explicated both later in this chapter in function (Subtypes A and B differ in this re- and in chapter 22. gard), but in mature Mozart it usually begins

the recapitulation (Subtype B). The double More on Terminology: “Ritornello” or perspective’ encountered in the exposition(s) Tei???

also applies here, as do its caveats. The R383

portion constitutes a separate, solo recapitulation, To what extent are our R1-4 stations really referentially analogous to but perhaps differing “ritornellos’? That they are “tuttis” is clear, in some aspects of content from the S1 “solo ex- since their characteristic feature is that of suborposition.” The larger complex, R383 + R4, dinating the soloist to the lead of the orchestra. constitutes the larger recapitulation (which may (The impression will be that the “solo” has e1append a coda at the end). In Type 5s in which ther not yet started, as with R1, or has come toa S1 had differed substantially from R1 (com- temporary resting-point in its own ideas—even mon in the first movements of Mozart’s concer- if, as in some performance traditions, the soloist tos), the larger recapitulation—and especially continued playing a subordinate, basso continuo its post-MC modules—serves as a synthesis or role or became reabsorbed back into the tutti.)*® fusion of the two earlier layouts. When this But are these tuttis also ritornellos? The latter happens, the larger recapitulation will be more term claims more, both by way of historical rescomplete than either layout found in the first onance with concerto traditions and by the etytwo rotations. The successive rotations of the mological suggestion of the defining feature of a whole movement grow in their inclusiveness, return to material heard earlier (ritornello: “little culminating in the grand totality provided in return’’). In this sense “ritornello” is a subset of the larger recapitulation. One function of the the broader notion of “tutti.”

recapitulatory rotation 1s to serve as the har- Some recent writers, such as Plantinga in monic and rhetorical telos of the ongoing pro- his recent study of Beethoven’s Concertos, have

cess of modular accumulation. decided to avoid the connotatively charged Following most modern commentators, our term “ritornello” altogether, replacing it with R4 differs slightly from Koch’s, who restricted “the more neutral tutti.”4? (Our R1, R2, and it to only that orchestral tutti following the ca- so on, are generally equivalent to his T1, T2. denza. But an orchestral tutti also precedes the ...) Others, such as Forster, prefer to describe cadenza (Koch regarded it as a mere orchestral our R1, R2, and R4 with the “neutral coversupplement to S3), and today that, too, is usu- ing-concept tutti” —as in Anfangstutti, Mitteltutti,

ally grouped under the R4 concept. This un- and Schluftutti—while also admitting that in derstanding of R4 always splits it into two parts. most cases these moments, and some others, are

What we label as R4! 1s the precadenza por- also ritornello-lke in function: “The concept tion (often rotationally inert, as will be seen) of ritornello 1s used here only in the literal sense of this two-part ritornello, typically leading to of the word (ritornare) to refer to sections that Mozart’s Piano Concertos: A Study (Amsterdam/Atlanta: continuo accompaniment during the orchestral ritorRodopi, 2001), p. 4: “The development section is the nellos and perhaps also during the shorter tutti passages only one reminiscent sometimes of the ‘episode’ of the within solo sections. .. . Musical evidence also seems baroque concerto; in the early concertos (before 1777) it to support the use of continuo” (Mozart: Piano Concertos

is neither thematic nor motivic, but freely invented.” Nos. 20 and 21, pp. 104-8, at 104). 48. Grayson opens his summary of differing views as 49. See, e.g., Leon Plantinga, Beethoven’s Concertos, p. follows: “A variety of evidence suggests that Mozart 1n- 17. tended and expected the piano soloist to provide basso

446 Elements of Sonata Theory

would be taken over unaltered from the open- now call the Baroque concerto. In 1779, as we ing tutti. But ritornellos and ritornello-like sec- have seen, Vogler called these sections only the tions can also appear inside the solo [portions]. prelude, interlude, and postlude. In 1789 Ttirk Conversely, a tutti-part need not uncondition- mentioned that the opening section was “called ally be a ritornello. Instead it can seem to have the ritornello (tutti)”—so was the final statebeen built, as it were, on the spot.”°° The pres- ment at the end—but he used a slightly different ence of such tutti interjections within the solo sec- term for “interlude” (Zwischensatz, not Vogler’s tions complicates the matter. Occasionally they Zwischenspiel) for interior orchestral statements

reinvigorate a module, ritornello-like, from the following the solo sections. In 1793 and 1802 opening R1. (As will be discussed in chapter 21, Koch referred explicitly to ritornellos for all of many of Mozart’s concertos feature such a tutti these. Galeazzi (1796) and Kollmann (1799) interjection in the vicinity of the P-TR seam used the term tutti, as did Czerny (ca. 1840). It within Solo 1.) Others appear in an ad hoc way, is difficult to generalize from these writers, but part of an ongoing “dramatic dialogue” of “co- one might suppose that the older connotations operation,’ “competition,” or “confrontation” of the term ritornello lingered on, however hap-

with the soloist.>! hazardly, in some musical circles.

While acknowledging the terminological As for the second issue—current underproblem, our preference 1s to retain the famil- standings of the connotations of the term—we iar term ritornello for the four structural pillars. should recall that the tutti sections that we norThe passing, ad hoc, or dialogic tuttis we refer = mally label as “ritornellos” in Baroque concerto merely as tutti interjections or unnumbered tos also serve diverse functions in those pieces. tuttis. In making this choice we hope to under- Clearly, an opening ritornello has something of score the historical roots of the structure (the an expository or initiatory function (the begincontinued persistence of Baroque norms into ning of a process), something that in some conthe classical concerto) and to suggest that the certo movements can be recaptured with a “da sections that we designate as ritornello pillars capo” effect somewhere around the last third of 1—4 are specialized adaptations of older tradi- the movement.°? Interior tuttis most often feations, crafted to mark the most crucial struc- ture the merger of a modular-return function

tural hinges of the Type 5 movement. (the archetypal ritorno of something from the This terminological question involves two opening) with, at least at their outsets, that of broader issues: what were the words used in a cadence-confirming function. But some later the eighteenth century for these sections? and tuttis, for example in some Vivaldi concertos, what should be the broader connotations of the state material not provided at all in the opencurrent use of the term ritornello? As for the ing ritornello, and the current term for these first, erghteenth-century usage was inconsis- passages 1s also ritornello. Well-known instances tent. In Scheibe’s Critischer Musicus (1739) and of “blank” or new-material ritornellos may be Quantz’s Flute Versuch (1752), we findthe term — found in the first movements of the “Spring” ritornello (occasionally exchanged with tutti, as in and “Summer” concertos from Vivaldi’s The Quantz) but with reference to what we would Four Seasons, op. 8 nos. 1—2.°3 Such occurrences

50. Forster, Die Kopfsdtze, p. 48. circle-of-fifths passage moving from C-sharp minor 51. For an extended study of this, see Simon P. Keefe, to V of E. Although its rhythm is based on that found Mozart’s Piano Concertos: Dramatic Dialogue in the Age of in Ritornello 1, the actual modular (“thematic”) conEnlightenment (Woodbridge, U.K.: Boydell Press, 2001), tent of the ritornello differs from that original tutti.

from which the phrases in quotation marks have been The programmatic opening ritornello of the G-minor

drawn. “Summer” concerto, op. 8 no. 2, represents “exhaus52. See, e.g., the J. S. Bach examples cited in n. 33, tion caused by the heat,” even beginning at a slower

along with the text to which it refers. tempo than much of the rest of the movement. It re53. In the first movement of the E-major “Spring” turns as Ritornello 2, mm. 52—58, but subsequent ritor-

concerto, op. 8 no. 1, Ritornello 5 consists of a nellos move on to new programmatic material: Ritor-

The Type 5 Sonata: Fundamentals 447

open the door to the concept of a ritornello R1 and the Subsequent Exposition: The with merely a tutti function. In short, one needs Question of Conceptual Primacy

a broad the Vogler severalsuggested things that thatin.cre_. P . conception on In 1779, weof recall, the .traditional ritornello function could entail, , oq: _ ating a concerto movement one could begin including the possibility of writing over tutti . 3 sonata—for ae by composing an ordinary Type spaces. .of.ggexample, normative “return” with other, new . a plano sonata movement—to which modules. With a more flexible understanding . . . one could then add supplementary orchestral of the. varieties of ritornello spaces ainmodest, earlier .. material, encompassing anticipatory

4..

concertos, the apparent controversy involved in .

lectine th cerminolosy for Type terminology 5 prelude (R1) alongfor with a later interlude and a selecting 5 so.on such 6 Propetthe BY proper YP postlude (our R2 and R4). Ine passing

natas loses some of its urgency. . . . .

advice Vogler may have had in mind a tradi-

tional exercise or rule of thumb given to novice

The

; composers.°* Under conception R1, since it Status of Ritornello 1 this (R1) OPO . P

Type 5 sonata

is subordinated to S1, points forward to the more

. . important expositionMozart’s that followsPit,Pinstead of movements—especially a

. the otherones—are way around. Those marked wishing by to adapt more expansive normally . Vogler to the more complex practice of Motwo. aimportant features: their R1 layouts are . - zart similar would conclude the primary referential rhetorically to sonata that expositions in all layout is the larger solo exposition, $1 + R2,

_. ticipation.

ways except tonality (R1 normally does not . ; to which R1 is a mere (and thus secondary) anmodulate away from the tonic); and their larger eo

expositions (St + R2,view now including - ler’s v: © of th . the ¢ thesolo nn —

pm a IOc I ig A eeTP eG LS S(O Ge Aa ES CC

a nondissolving repetition—including of its ca- Although it is sometimes observed that dential module only—normally remains with Mozart’s concertos often begin in a piano dy-

the R1:\P-zone. namic—more frequently, at any rate, than do More compact or smaller-scale opening his symphonies—this impression should be ritornellos sometimes reduce the R1:\P-zone qualified. While twelve of the seventeen Viento a mere head motive, a matter of a few the- nese piano concertos from 1782 onward (“Nos. matically stamped bars that proceed efficiently 11-27”) do begin quietly (starting with No. 12 into an R1:\TR-zone before any such I:PAC in A, K. 414; the two concertos illustrated in is attained. In these instances we are usually Examples 20.1 and 20.2, both from 1784, are dealing with PS>TR mergers, in which a sen- typical in this respect), five others do not.!> tence-presentation or some other analogous unit Moreover, as Kiister has pointed out, “Up until leads immediately to a dissolving continuation his second trip to Paris [1778] Mozart always or succeeding module that 1n effect takes onan __ began his concertos in a forte dynamic, with e1-

R1ATR function. (One may find examples in ther a melodic or a pronounced harmonic forthe first movements of Violin Concerto No. 1 mula.’”!° This would include all of the concertos in B-flat, K. 207; Violin Concerto No. 2 in D, up to the Piano Concerto in E-flat, K. 271, from K. 211; Violin Concerto No. 5in A, K. 219; and 1777 and the Flute and Oboe Concertos, K. 313

a few other works.)!4 and 314, from 1778.'” Mozart’s forte openings 14. These include but are not limited to the Flute Con- K. 466; No. 21 in C, K. 467; No. 23 in A, K. 488; No. certo in G, K. 313, and the last three of the four Horn 24 in C minor, K. 491; No. 26 in D, K. 537; and No. Concertos: No. 2 in E-flat, K. 417; No. 3 in E-flat, K. 27 in B-flat, K. 595.

447; and No. 4 in E-flat, K. 495. 16. Kiister, Formale Aspekte, p. 39 (our translation). 15. The composition of No. 12 (K. 414/385p) chron- Ktister observed that the first concerto movement to ologically precedes that of No. 11 in F, K. 413/387a, begin piano was the fragmentary oboe-concerto moveand No. 13 in C, K. 415/387b, both from 1782-83. ment in F, K. 293. Some of our descriptions below of (The three were published together in 1785 as “op. 4,” thematic types within R1:\P is also indebted to Ktister, the first of Mozart’s Viennese piano concertos. No. 11 pp. 35-40. in F, in any event, begins not piano but with a largely 17. Note also the retention of the forte opening in forte, all’unisono scalar slide, mm. 1—4. Quiet openings the Sinfonia Concertante in E-flat, K. 364, and the are found in Nos. 12 in A, K. 414; No. 13 in C, K. 415; Two-Piano Concerto “No. 10” in E-flat, K. 365, both No. 15 in B-flat, K. 450; No. 17 in G, K. 453; No. 18 in from 1779. B-flat, K. 456; No. 19 in F, K. 459; No. 20 in D minor,

482 Elements of Sonata Theory

were frequently of two types. The first relies on sort or another. 2! We shall deal with this issue a vigorously driving initial module, sometimes in a later section.

featuring a syncopation on the second beat of In several piano concertos, especially those the first bar (especially characteristic of the ear- from 1784—5, Mozart favored the idea of a quiet lier concertos), whose forward vector might be opening in the style of a common-time march, also propelled by a pulsing Trommelbass (Bas- normally with a dotted-eighth/sixteenth figure

soon Concerto in B-flat, K. 191; Violin Con- on the second beat of the first bar. March-like certo No. 3 in G, K. 216; Flute Concerto in R1:\Ps initiate Nos. 13 in C, K. 415 (a “preG, K. 313).!8 The second is the familiar forte- decessor” from 1782-83); 16 in D, K. 451 (the piano juxtaposition, featuring a brief, declara- only forte opening in this group); 17 in G, K. tive basic idea—often an all’unisono module, a 453 (example 20.1); 18 in B-flat, K. 456 (examtriadic arpeggiation or “fanfare,” or some other ple 20.2); 19 in F, K. 459; and 21 in C, K. 467 “curtain-raising” effect that announces the pre- (lacking the dotted-eighth/sixteenth stamp). vailing tonality—to which a quieter contrast- Other quiet openings feature more lyrical, caning idea responds (Violin Concerto No. 4 in D, tabile melodies: Nos. 12 in A, K. 414; 23 in K. 218 [example 5.9]; Sinfonia Concertante in A, K. 488; and 27 in B-flat, K. 595—as well E-flat, K. 364; Piano Concerto No. 11 in F, K. as the Clarinet Concerto in A, K. 622 from 413).19 Occasionally the compressed forte-piano 1791.22 “Special-effect,” ominous piano openjuxtaposition creates the familiar effect of ini- ings launch the two minor-mode piano concertial presentation-“loops,” a self-replicating cir- tos, Nos. 20 in D minor, K. 466, and 24 in C cularity eventually released into a “breakout” minor, K. 491. continuation or dissolving continuation (Piano

1 - . . Wild Card

Concertos No. 7 in Py 242; No. 9 in E-flat, “Motto” R1:\P as Idée Fixe or Later

K. 271 [example 5.7], in which the contrasting cay: 3 idea is non-normatively, and famously, supplied

as an interjection by the soloist; and No. 22 in Sometimes the initial module or general rhyth-

E-flat, K. 482).?° mic figuration or contour of the piano R1:\P Many of the most celebrated piano concer- opening recurs repeatedly within the movement, tos, though, feature quiet openings, frequently spreading out at various locations throughout ones that burst into a forte R1:\TR with an Type 5 sonata-space. When this happens, the elided I:PAC. As many writers have noted, the distinctively stamped R1:\P becomes a motto or choice of a piano or forte opening creates a preset idée fixe. This motto can then function as a wild

situation to be revisited at the moment of the card, an often rotationally inert card that may solo piano entry, after the conclusion of R1, be placed onto the sonata-table at any number of since the solo exposition (or the second rota- later occasions, turning up, so to speak, at nearly tion) most often begins with a reiteration of the every available opportunity. It may be suitable R1:\P theme, albeit one that might be immedi- for filling in not only the onset of a P-based ately preceded by an “extra” solo entry of one C-space but also of insinuating itself into vari-

18. Others: Violin Concerto No. 1 in B-flat, K. 207; see also, e.g., Kiister, Formale Aspekte, pp. 39, 84-88;

Oboe Concerto in C, K. 314. and David Rosen, “‘Unexpectedness’ and ‘Inevitability’ 19. See also the Violin Concerto No. 2 in D, K. 211; in Mozart’s Piano Concertos,” in Zaslaw, ed., Mozart’s the Two-Piano Concerto, “No. 10,” in E-flat, K. 365; Piano Concertos, esp. pp. 270-78. and several others, culminating in the grand-style ad- 22. Notice within this category the presence of three aptation opening the Piano Concerto No. 25 1n CG, K. works in A major. Cf. the piano-dynamic, “weak-launch”

503. openings of, e.g., the Symphonies No. 14 in A, K. 114,

20. See the discussion of modular loops in ch. 5. and No. 29 in A, K. 201. On the other hand, the open21. See also the remarks on dynamics in David Gray- ing of the Violin Concerto No. 5 in A, K. 219, is set son, Mozart: Piano Concertos Nos. 20 and 21 (Cambridge: forth with a single, propulsive forte chord turning inCambridge University Press, 1998), p. 31. On the rela- stantly into a piano, nervously bustling “rocket” arpegtionship of the opening of R1 to the initial solo entry, elation punctuated with sudden forte flashes.

The Type 5 Sonata: The Opening Ritornello 483 ous “soft spots” of the sonata-structure to fol- In most of these latter cases—and here 1s where low: as medial-caesura fill, as overlay in the con- the “concerto” aspect of this moment most fully cluding display episode of Solo 1, as an internal applies—the forte R1:\TR1! will be the one used module within a phrase-chain or multimodular to begin Ritornello 2 and/or Ritornello 4! (the S, as a filler for a rotationally neutral opening precadenza portion of R4). Accordingly, in its portion of R2 space, as a “substitute” element energy-level and immediate impact of celebrawithin a recapitulatory transition, and the like. tory élan, it is often crafted so that it may also be At the same time, the regular resurfacings of used as the precipitating head-motive of either the idée fixe motto serve as threads binding to- or both of those later ritornellos. gether a highly varied discourse. The three clas- Three transition types are most commonly

sic instances of this “wild-card,” scattered-re- found within Type 5s with a large-scale R1 currence technique—a special and sophisticated (one whose primary theme normally ends with effect within the later Mozart concertos—are an elided I:PAC). As in Type 1-4 sonatas, the found in Piano Concertos Nos. 19 in F, K. 459; TR-type is identified by the module selected 21 in C, K. 467; and (with fatalistically negative for the TR-opening, R1:\TR!! (see chapter 6). connotations) 24 in C minor, K. 491. We shall The first, and the most common, is the indepen-

revisit some of them in chapter 21. dent transition, which sets forth a new thematic module. (Piano Concertos Nos. 17 in G, K. 453

The Transition: R1:\TR [Example 20.1, m. 16], and No. 18 in B-flat,

K. 456 [Example 20.2, m. 18] may serve as 1l-

In most respects the transition-zone within lustrations. Additional instances abound in the Ritornello 1 corresponds to characteristic TR concertos.)?3 The second is the reinforced, varbehavior and options within Types 1—4 sona- ied, and quickly dissolving restatement (Piano tas. Within a two-part R1, its main purpose 1s Concertos No. 21 in C, K. 467, m. 12; and No. to gain energy and drive toward a conventional 24 in C Minor, K. 491, m. 13). The third, and medial caesura, in most cases a I:HC MC. In least frequent, is the developmental transition or smaller-scale Allegro works, as mentioned transition that arrives as the motivically related above, the transition can emerge early on in the culmination of R1:\P (Piano Concerto No. 20 context of a P>TR merger following only a in D minor, K. 466, m. 16, with the effect of few declarative bars of P that are not pursued an unnervingly demonic “shock of a thunderto a fully closed I:PAC. This situation also ap- clap”’).24 plies to TR continuations following presenta- Occasionally one comes across a lower-level tional P-zones, including those structured as default procedure to open R1:\TR-space. Piano double-loops, as in the Piano Concerto No. 9 Concerto No. 14 in E-flat, K. 449, provides a

in E-flat, K. 271 (example 5.7). classic instance of a TR opening with a sudden, Within large-scale Type 5s—including es- stormy shift to vi (C minor, upbeat to m. 17) pecially most of the Viennese piano concer- in response to a tonally overdetermined P-zone tos—it is most often launched with a strong (one with obstinately multiple authentic caforte affirmation, elided with the I:PAC that dences in the tonic).?° And two of the Viennese concludes R1:\P. This produces a sudden surge piano concertos sound R1:\TR!! in an uncharof vigor, decisively accepting the offered sona- acteristically subdued piano dynamic, indicating ta-contract and propelling the structure onward. that some other module will have to be used

23. E.g., the Two-Piano Concerto “No. 10” in E-flat, a close); No. 23 in A, K. 488, m. 18; No. 25 in C, K. K. 365, m. 14; Piano Concertos Nos. 11 in F, K. 413, 503, m. 26; No. 26 in D, K. 537, m. 13; and the Clarinet m. 12; No. 12 in A, K. 414, m. 17; No. 13 in C, K. 415, Concerto in A, K. 622, m. 16. m. 10; No. 15 in C, K. 415, m. 10; No. 16in D, K. 451, 24. Grayson, Mozart: Piano Concertos Nos. 20 and 21, m. 10; No. 22 in E-flat, K. 482, m. 31 (probably a bet- p. 32. ter choice than the forte in m. 29, which impatiently— 25. See the discussion of P-overdetermination in ch.

and wittily—brings a blandly self-reiterating R1:\P to 5.

484 Elements of Sonata Theory

as the energetic opener of the later R2 and/or I:HC MC occurs, with a wind-figure of internal R41. One occurs in No. 19 in F, K. 459, m. 17 decoration, in mm. 29-31 (the downbeat of this (nonelided to the preceding I:PAC)—perhaps a last measure 1s the MC proper). Instead of prostaged reaction to the forte restatement of R1:\P ceeding directly to R1:\S, we find a continued in mm. 9-16; or perhaps a fleetingly local sug- sounding of the preceding wind-figure, moving gestion of the onset of a b section to a larger, downward through scale-steps 5—4-3-—2-1—a and ultimately not-accomplished aaba’ (rounded classic instance of “juggernaut” caesura-fill, binary) opening. The other is found No. 27 in coming to rest, as is common, on scale-step 1 B-flat, K. 595, m. 16, which follows an unusual, (m. 35) to release the opening of the secondary forte R1:\P?2 flourish-module in mm. 13—16.?° theme, here R1:\S!-!.28 Other ilustrations of an expanded R1:\CF may be found in Piano Con-

, , certos No. 16 in D, K. 451, mm. 26-35, No. 18

R1: Medial Caesura and Caesura-Fill in B-flat, K. 456, mm, 28-39 (with a sudden In most cases R1:\TR eventually produces a “lights-out” chill to minor), No. 22 in E-flat, dominant-lock and MC, dividing R1 into two K. 482, mm. 46—51; and No. 24 1n C Minor, K. parts. Since R1 does not normally modulate, 491, mm. 35—44 (P-motto-related, and in this in both major- and minor-mode Type 5s all of — case, in part similar to that in K. 482/1, a some-

this is most often built around an HC in the what uncommon ascending caesura-fill, with tonic (although modulatory feints are possible, R1:\S!+ beginning at m. 44).?° as will be discussed below). From time to time Although nearly all R1:\MCs are half caone comes across a widening of the MC-gap dences in the tonic, we should point out the with an expanded caesura-fill (CF) of several notably exceptional, locally emphatic V:PAC measures. Once again, Piano Concerto No. 17 MC found in Piano Concerto No. 25 in C, K. in G, K. 453, provides an illustration (example 503, mm. 48—50 (example 20.3). At least at this 20.1).27 RINNTRI! (m. 16, the presentation of a moment the music makes a surprising gesture sentence) and TR!2 (m. 22, the beginning of | toward a secured modulation to the dominant. the continuation) do not modulate but push di- Opinions have differed regarding whether we

rectly into a dominant-lock, I:HC, in m. 25. are really “in” or merely “on” the dominant at The triple hammer-blow effect announcing the this point.°9 It is surely relevant to notice, for 26. The presence in K. 595 of a second, differing P-mod- mm. 32-38 (the last bar, the conclusion of an “apparent”

ule within R1 1s a non-normative event. In this case PAC, elides with R1:\S). the “extra” P? flourish is the kind of gesture that Mo- 28. On “juggernaut” CF, see ch. 3. The procedure in zart had used in many earlier concertos at the very end K. 453 is similar to that found in the first movement of of Ritornello 1, one designed to be repeated almost Symphony No. 39 in E-flat, K. 543, cited there as the at once as the first tutti interpolation within Solo 1 as a touchstone example. characteristic confirmation of the initial I1:PAC of the 29. In all such instances the thing to bear in mind 1s soloist’s statement of the primary theme. In K. 595 this that any CF exists to bridge an otherwise empty space confirmational orchestra-flourish is doubtless also re- between two standard zones (TR and S). Although at lated to the brief wind interjections straddling the bar its end the CF often (though not in K. 453) produces the lines at mm. 6 and 10. It appears after P! in both Ritor- effect of an elided PAC when it joins up with S-space, nello 1 and Solo 1 (mm. 92-95) but 1s absent from its this authentic cadence is merely local, not structural. It normal position at the end of R1. In short, we have is not to be taken for either a PAC:MC or an EEC. Asa a characteristically Solo 1 P-concluding procedure ad- general rule, should any CF measures recur in a differvanced into the equivalent position within R14; or, from ent position later in the composition, they are to be reanother perspective, Mozart has displaced a typically garded as similarly incapable of articulating an S$1:\EEC

concluding gesture within R1 to become an appendix or S3:\ESC. Instead, what follows any later revisiting to the R1:\P-zone. Whatever the interpretation, it is of this CF-PAC will normally be understood to exist probably in response to this unusual feature that the within S-space. Realizing this can help to clarify EECsubsequent R1:\TR!! falls back to a piano dynamic, re- or ESC-deferral issues later in the composition. nouncing the transition’s customary opening forte. 30. Tovey, “The Classical Concerto,” p. 18, provided 27. Another example of expanded caesura-fill may be a paragraph of instruction that no real modulation has found in the R1 of Piano Concerto No. 26 in D, K. 537, occurred at this point (we have only “paused on the

EXAMPLE 20.3. Mozart, Piano Concerto No. 25 in C, K. 503, 1, mm. 36—58

PO a [Allegro maestoso]

% [add-—J aad J J] d — S) arAG +, —— —— SD —— ——— —$_—_— —— ee PF F 6 ee ee _ eeee n. ? nn? oa a al ne ny a laOe ie fallen Wwe Oe Oe OY eee, eee. | ee ae, a): ee Oe Oe Oe”

[fl ee ae ee ff a, |Pe | | CY Pt J a A a a a CD SN GO GOD (OG ee ee tt —m | e—m a SD QO OGLT GOD GQ | eee ee eee pM CUT OS kee Oe Oe tee ee etree OO Oe

ee t 2 2 + C2 o iaQO CG (OO EEGO ee ee( a eeee TJ 7 7 7 Oey | ye he ud = aA, ssAQQ es occ 0 i A A es A cr

OF A| ea Es eei a el apMaOTT |”Pf |—_ _ a“ yp ={_ — — Lad > (_ — — — a eo

PF a I nce (= | ee 1) Pie of oo —— he i Oe a |"rr ee2 Pf. eee Oe eee Oe Tf owe Oe | | | ———— | M~ | | [| |See | | | [abel | | ~~ a PM Mm | | FT 8 2\\ (eee ——— a2 a| |—— —— ——

|ea — : oe e| ¢ a Ee es es es nn ee a= iA— eet—“(‘“‘ MM Re QO Dae SO

43 * . a — ence -_ ff P| —_? »——_*;__,_@ _## | , 2) } |__| | ioe— asl |__| | _{ {|y| Pr’ hig Poa es SSS 7 2 eOa? _ SO ye TT fey | | — Tt TT TT | Ff gr | Ff Py yy Rl UU UUULULULULULULULULUml|m.|hUm6WRMmhUmw_e LCT CTC

45

8 a a wa a Ds a OT I SP a 8 fan ee ee eee ee ee ee ee ee ee eee 2 ee ee eee eee eee eee eee = eee

l@er |estaee Bm ay A” PF SSyx esGC —_ s OO EsSS ne esACY OD

ee /) "| | ee A=I

a AP

aPNee a a > RP *”* |||| EXAMPLE 20.3 (continued)

A Fca sD SeOC ee CO ee.CO SsOO PSCO A

48

A ey SY CO ON QO

5] a Oe gm leer FBT "=m

-o}-—_f : P__-#__ P| @m —————————————————

| gf $< gp —_____ __| 174 y__¢ _ _*. __4 li i... An) Oe §=e

Pp . . . . . i . . . . . . . . * .

54 — ’ oe el— | ao. my b wri Fl Le Pe SS

fo FFeeehte FB TT SoeeeoO Ef... Ee eee ee)oe ee ee eee eee es _hr? eee. eee) eee

lob 4 5A EESY eei a CS |eeaerol LS aeepe eS

ng eee a Ss _ SE -g 2 2 FG peg_

57

a nd w FP ee

-o-}-—_ GA

The Type 5 Sonata: The Opening Ritornello 487

example, the dominant-lock on V of V in mm. Any interior modulation or near-modulation 36-40, along with an “attempted gap” at m. within a normally tonic-centered R1 is an im40 in several voices. This might well be taken portant event. When one does occur, though, locally as an initial attempt to offer a V:HC that ephemeral visit to a nontonic key is more MC. On one line of interpretation, we could typically found around the beginning of R1:\S maintain that that MC-offer is strenuously de- than, asin K. 503/1, at the end of R1:\TR. (Exclined (“No!”) with the three-note upbeat to amples will be provided below.) In other words, m. 41, which unleashes a three-bar torrent of such nontonic feints are more characteristic of “learned-style” contrapuntal activity, descend- the second part of R1 (post-MC) than they are ing down the circle of fifths, then sprinting for- of the first. This is an important distinction. K. ward cleanly to sound a doubly stated, maxi- 503’s unusual move into (or onto) V exclusively mally emphatic V:PAC (mm. 46, 48) followed at the MC point, only to drop it at once for by aftershocks (mm. 49-50), which themselves the tonic, suggests a wittily overenthusiastic and reinforce the new MC-effect. From another premature grabbing after V-as-key in the wrong perspective—simultaneously relevant—tt 1s also spot. This is a purposely staged generic “error” possible to hear mm. 41—48 as an unusually ag- (deformation) at the end of the first part of R1, eressive, composed-out version of the sort of | and its consequences ripple forward into the rest

“juggernaut” caesura-fill, moving essentially of the initial, tonic-grounded tutti. 5—4~-3—3-1, that we just identified in K. 453/i.

Under this interpretation the 5-4-3 —2-1 fill No R15 Produced: The Continuous R1 apparently becomes so vehement 1n context that

it either shifts the earlier, hght MC-offer deci- Not all R1s are divided into two parts with an sively to m. 50 or stuns the expected, much-qui- MC in the center, followed by an S-idea: the eter R1:\S-theme to the point that it seems hes- continuous R1 1s also a possibility. It can hapitant to show its face at the expected spot. pen that no MC-effect 1s produced at all (Piano However one decides the matter (both “in” Concerto No. 19 in F, K. 459) or that we find a and “on” are integral aspects of this connota- parallel with the second type of continuous extively charged moment), Mozart immediately — position in Types 1—4 sonatas (early PAC with corrects the planted perceptual problem in the successive, varied recapturings of that cadence:

caesura-gap that follows. The G-major chord Violin Concerto No. 1 in B-flat, K. 207). Of is reiterated insistently in mm. 48—50, as if the particular interest are the few R1s that sound music were either registering its own astonish- an MC and begin an expanded caesura-fill (as if ment at the V:PAC or peremptorily summoning an S-theme were imminent) but then allow the the now-cowed R1:\S-theme. It is thereupon CF to spread out at a much greater length than taken as an active dominant, and we proceed expected. This results in the CF writing over directly to the march-like R1:\S! in the proper the possibility of a “real” S-theme and accomtonic key, at first in C minor, however (mm. plishing itself what is likely to be interpreted as 51-58: a fearful or timid entrance? a deflated the R1:\EEC. The sequence of events is: I: HC reaction to the improper MC? a registering of | MC — much-expanded CF, coming to take on disapproval?), then in C major (the reinflated a Fortspinnung identity of its own — R1:\EEC

mm. 59-67, followed, though, by an R1:\S? and onset of R1:\C. In effect, this is another that seems to have those earlier MC-aftershocks type of continuous (proto-)exposition, since

still on its mind). an S-theme proper is left unsounded. Examples

dominant. .. . [notwithstanding the cadence] it here n. 10). Both writers missed the essential point, namely, sounds only like very strong emphasis on the dominant this moment’s non-normative MC-function—someof C”). Plantinga disagreed, Beethoven’s Concertos, p. 72, thing substantially different from a seeming (and tranand pp. 329-30, n. 10: this passage in the tutti “modu- sient) modulation within R1:\S-space, which is a more lates unequivocally to the dominant” (p. 72) because of common ploy within an initial tutt1. “the very long V/V preparation in mm. 36ff” (p. 330,

488 Elements of Sonata Theory

may be found in Flute Concerto in G, K. 313 and modular texture initially produced as a de(MC-effect at m. 12, reinforced at m. 14; ex- ceptive cadence. panded CF throughout, leading to the R1:\EEC The very strength of this norm, its virtual at m. 23) and Piano Concerto No. 13 in C, K. inevitability, throws into vivid relief the few 415 (MC at m. 24; expanded CF at that point, exceptions to it: those instances where R1:\S

accomplishing the R1:\EEC at m. 36).+! begins in the nontonic key that would be appropriate were this a sonata exposition rather

The Secondary-Thematic Zone: RAS than Ritornello 1 of a Type 5 sonata. (This

(Tonality) procedure 1s different from that of the seemingly tonicized-V MC-deformation in K. 503/1,

In terms of its rhetorical shape and purpose, example 20.3.) In all cases, this purposeful R1:\S is analogous to Sin Types 1-4 sonatas. It “S-musstep” 1s corrected en route, and the R1:\S almost always opens piano, usually with a can- soon restabilizes back to the original tonic. In

tabile theme; more vigorous S-modules may this situation, up to the point of the tonic coror may not follow. As opposed to what hap- rection, R1 resembles the modulatory exposipens in a normative sonata exposition, however tion of the first movement of a symphony (since

(in which S is to be stated and completed in the soloist has not yet been heard from). The a nontonic key), R1:\S is almost always stated tonal correction is a definitive declaration of the

in the tonic key throughout. More precisely: genre at hand: “No! This is not a symphony. in Mozart’s Type 5s the R1:\EEC always se- This is a concerto!” The nontonic feint and cures the original tonic—declaring R1, con- “decision”’-aspect of the subsequent correction sidered in toto, to be nonmodulatory. And in are mechanisms that draw attention to what this the large majority of cases R1:\S also begins piece is—to its very “concerto-ness.” in the tonic key, following a I:HC MC. Thus Not counting the K. 503/1 variant, there are one expects as a strong first-level default that three celebrated instances of this in Mozart’s R1:\S will remain in the tonic throughout. Viennese Piano Concertos, each with differ(This does not exclude the common possibility ent expressive implications. In Piano Concerto of a fleeting interior plunge onto a surprising No. 11 in F, K. 413—see example 20.4—a I:HC nontonic chord—such as bVI—that might even MC (m. 23) leads to two bars of anacrusis-fill be evanescently tonicized before recollapsing and the beginning of a sentential idea proper, S, back into functionality within an unequivocal that begins in the dominant key, C (m. 26). At tonic-key center.) The R1:\S-zone of Piano the point of the continuation (mm. 30-31) the Concerto No. 17 in G, K. 453, mm. 35-57, generic “mistake” is noted: the C-major conshown in example 20.1, is normative: it begins tinuation is aborted (“Wait! This is not a symand ends in the tonic, G. The G-centricity, of | phony!”’), and a corrective modulation is made course, is never significantly challenged by the back to the tonic F. The whole process then sudden local oscillation around the dominant’s backs up and rebegins, only now tracked propupper neighbor, E-flat, bVI, in mm. 49-53 (the erly in the tonic: a restatement of the two-bar onset of R1:S!-2), a dramatic shift of tonal color upbeat fill (mm. 32-33) and the tonally chas-

31. The procedure of K. 415, in particular, should be might be understood as conventional C-rhetoric jumpcompared to the related but much more problematic ing in early and, in retrospect (especially after the new situation found in the first movement of Symphony No. “arrival” and subsequent new-C “wind-up”m. 74?), 35 in D, K. 385, “Haffner,” mm. 48-58. (See ch. 3 on being reconverted into S-space. In any case, the EEC this type of expanded caesura-fill.) In the latter piece we is best regarded as deferred until m. 74, where, in effind a local V:IAC at m. 58, which might initially tempt fect, an entirely new C is wittily conjured up to extend one to suppose that it could serve as a light EEC-effect the exposition. This effect is probably compensatory in or EEC-substitute. Moreover, this leads to what cer- function, in order to avoiding giving the impression of tainly seems to be emphatic C-rhetoric (suggesting a having concluded “too soon.” P-based C?) at m. 59. Under some interpretations m. 59

EXAMPLE 20.4 Mozart, Piano Concerto No. 11 in F, K. 413,1, mm. 22-41

[Allegro] =

22 y— es | $I F ot, wr. a. 244————~ ae oo —— EE 9 J iy" a

dee ddd P

Fi infest t§ es 2» 2» rrFesrrree a | @ Mm @¢ =| if ee, Jf we if ea jf | coy, | ri

} id Y oe Pet TT

A f) a | |

|Pige | OPT= gy gsagy ly SC bpieeing ss asa ga MM MBwg (ZI

260 a i ile ? N tr = — = \Iplee| PN me Pe ee eto Tet fty Cb ooo

DD A eee eee eee eee ee eee ee ee eee, eee Oe eee eee =

a SS SS Sc sl a Se ee

Sm 23: Dido es a ss es ee Ss ee > CB ee"A ee *sO ay CO A

| - by 4 ——————— Ee eee ee ee —————Eoa Ee See ee

fojaad 2Okdd ebi a”eta Pe 9 esa CeFe ig Dea aOe ee Baeee Rae|Cee nO Dae

Ce aa 4f)py)Biy Ssouvtisyn le le — —£ ———,

So —?— 2 7

—— rh eB eS SS GF J pn mee a eeDeOIWO eh OEP

a, -ee-| a4a-eeiase ota- SS A| [|a +fAP" _f, pete. (Oo OE [rf v8, >A Pa A —

oo,

’ ™ ons

ee S —

on f rs ig

490 Elements of Sonata Theory

tened sentential S!-idea (m. 34, with R1:\EEC The Secondary-Thematic Zone: at m. 41). Mozart carried out the staged generic R1:AS (R1:\EEC Issues)

further in P1 1 . . . .

n 7E; n nOYS-factory not I:PAC, on to be regarded as the R1:\EEC. ee NOE

ae K 119 _ which » sence we 7 The main task of R1:\S is to drive to a satis-

begins in V may (B-flat,happen m. 37) but is actually per-span: ae . R1:\S . . . This within a brief mitted to continue for some eighteen measures ou. . K. 482, , , in Piano Concerto No. 22 in E-flat, and produce a V:PAC in m. 54. It is apparently . . , is a mere eight-bar sentence, mm. 51—58, one

too late to back up and restart this theme in of the shortest in the larce-_scale first move ments. More often, R1:\S is longer, and it is

the tonic, but Mozart does reinstate that tonic 8 - ’ 7

in the subsequent module multimodular (mm. 54-62), duti; or . frequently (R1:\S!-!, S!-2, S'S, fully detonicizing the dominant on and behalf of . . so on, including suddenly forte modules). the concerto-genre actually. at hand and even. Returning to our. larger example 20.1, we may

tually proceeding to the appropriately generic oo, .

R1\EEC (I:PAC) d th d (m, 84) see the situation in the multimodular R1:\S of ) owe moa we SST: » K. 453/1. Following the expanded caesura-fill,

Surely the most famous “tonally exceptional .

. . , ing ina PA . 35-42. A

case in Mozart occurs in Piano Concerto No mm. 31-35, R1:\S begins as a cantabile sentence

20 in D Minor, K. 466 (example 20.5). Follow- ending ae C; mm. 994 mentioned

“ae the stormilv demonic R1-\P and R1:\TR earlier, this 1s elided to a rescored repetition, hich cnraches ite wav to a HC MC in m1 39. mm. 42—49, but one that is undercut with a deR1AS1 sets out. sur on 1 “1 F maior (III, ceptive cadence at the end, m. 49. Experiencing

e 33) 4 ch a 1 et ” this, we now reconstrue our understanding of oe ki " - - on thy ae ee i en “the repeated sentence as R1:\S!!. The blustery

Thic F ‘how hi oon lin in rake 4 one R1:\S!-2, alternating piano and forte dynamics at

rabl - he oe, > : and its outset, occupies mm. 49—57 and produces the

ee Oe ee R1:\EEC at m. 57. (R1:\C-space follows, mm. within the ground rules of this genre—such

* 57-74.)

an R1 escape, we understand, 1s incapable of It can also happen that an S!:! module is im

being accomplished. After only two bars on a . PP“loop, eg > , Low. . mediately repeated as a characteristic hoped-for F major, the initial module is restated . in ascending sequences, though G minor (mm a structural strategy found commonly in Mo-

35-36) aA ae ( 37-38) before b , zart’s works. As discussed in chapters 5 and 8, ing drawn back into the clutches of the tonic D such loops are best understood as a special type minor and producing a first cadence there (mm of presentation within a larger sentence. Even

8 a though an individual loop might might 39-44). (The theme 1s PAC a sentence , (or , that . . . whole not) have a light -effectwith at itsaend, lo-

non-normative triply reiterative presentation.) . .

As is well known. Beethoven. too. would make cal PAC-effect is not a sign of zonal closure use of the nontonic Ping “a the opening to (R1:\EEC), since it occupies only the first part RAS in each of his first three piano concertos of a broader sentential idea. Such presentation Moroever ‘Cin the Fourth Concerto m. 29 . loops are elided with the next module, which regarded 1s the onset of an unstable R1ASt functions as a “breakout” continuation, R1:\S!.2

. . . i, ee 1:\S? hich migh iffer-

beginning, like the situation in K. 466, with et 7. , " o ok oun “ “ial a triple-statement presentation. on three tonal of see SEEPS ENE INSin R1:\S-moments Sinfonia Concertante

levels, E-flat, (only more broadly)—our preferred 1n-to. a conK. 364, mm. 38—46 (leading terpretation—the list should be expanded to include this work as well trasting crescendo module at m. 46); and Piano

Concertos No. 16 in D, K. 451 (mm. 35-43),

No. 18 in B-flat, K. 456 (mm. 39—47, with the

32. This secondary theme will always begin on a hope- opening of this theme is tonally invariant: it is never ful F in all of its later appearances, including that of the sounded in D minor. recapitulation (where it appears in m. 288). Thus the

EXAMPLE 20.5 Mozart, Piano Concerto No. 20 in D Minor, K. 466, 1,

30 . . fggHs — Hi i OE OT e—= fi _= —_ mm. 30-44

[Allegro]

> A Se” OS” - -Q-"x9“4"V.y]7—-— 2 O_—_ OY —_—"—€C@"X[>MQ~—~Vuonw OE I=." DH

ad s i A id

PS SFTTaeaIai eeeeeeEeFeoheT Cy? NTT wTIS£ mm”. —_ 2]

rg Sse FFTT fF ll | PP 9 oo¢SC Ooo OETT a OTTO A>SS (a. Oo a» CS

f) | | So TOO SY P52 — > f ° Y || ft

eee emer e—-=_oaeeremr,e_—»=>=>=_Se sn ey eS OO OOO FOO OO :--—““=>

492 Elements of Sonata Theory

loop over a sustained, “rustic” tonic pedal), and the material leading up to those cadences difNo. 21 in C, K. 467 (mm. 28-36, provided in fers.25 This seems to be the case in Piano Con-

example 8.5). certo No. 11 in F, K. 413 (refrain cadences in One should be especially alert for other cir- the tonic at mm. 40—41, 44-45, 52-53 [cf. mm. cumstances in which the R1:\EEC can be re- 50-51], and resulting in R1:\S!, S? [m. 41], and

garded as deferred past the first S-space PAC S3 [m. 45], with the R1:\EEC probably best or PAC -effect. Situations in which closure 1s considered to occur at m. 53.) The potential for postponed through direct repetition or through refrain cadences can occur at varying degrees of S!!_loops normally present few difficulties of in- strength: just how much 1s needed to suggest a terpretation. But things are not always so clear. clear “recapturing” of an earlier cadence? InterIn some instances, opinions can legitimately pretations could differ, for example, about the

differ with regard to the proposing of a best R1:\EEC moment in Piano Concerto No. 20 location for the R1:\EEC. Depending on the in D Minor, K. 466.54

circumstances, it may be wiser to make out the Another factor surrounding potential case for the alternative interpretations than to R1:\EEC deferrals is Mozart’s frequent appenddeclare an R1:\EEC with brash confidence: the ing of a nonelided piano “afterthought” at or toexpressive point of this phase of some Ris might — ward the very end of R1. Often this nonelided be the uncertainty of just when this sort of clo- idea occurs far into what we had been presumsure is attained. More generally, the EEC-effect ing was R1:\C-space. Does the nonelision sugin Ritornello 1 should be thought of as a more gest retrospectively an undoing of the earliloosely construed expectation, one that admits er-presumed R1:\EEC and the incorporation or even encourages more ambiguity than one into R1:\S-space of that which had preceded expects to find in a normative Type 1-4 sonata the nonelided cadence (in which case R1:\C exposition. This might be because of the multi- will now begin with the afterthought)? The 1sple and contrasting modules that usually follow sues surrounding this situation in Types 1—4 sothe R1:\MC in large-scale Type 5s. Not only natas have been discussed in chapter 8.°° In Type are there often several I: PACs in the second part 5s, however, the issue 1s less clear: a concluding of R1, but the whole section is usually also un- —_—spiano-afterthought C-idea (R1:\C?? C??—cor-

folded as nonmodulatory and tonic-stable—a responding to Leeson and Levin’s “theme 6”) prolonged string of “second-part”’ ideas. is a common “special” option within Mozart’s When R1:\S is very brief and lyrical and mature concertos. As such, it alone need not be is followed directly by a second, complemen- taken as automatically indicating an EEC-detary cantabile theme (instead of proceeding to a ferral up to that point. In these circumstances, first-level-default forte R1:\C), it is most herme- any consideration of deferral should be bolstered neutically reasonable to consider the two themes by other evidence as well. as R1:\S! and S?. (See the discussion surround-

ing Piano Concerto No. 9 in E-flat, K. 271, The Secondary-Thematic Zone: RAS

mim.oo, 26-41, example 8.6 of chapter 8.) Another .. aq: (Implications for Later Rotations) complication 1s the possibility of refrain cadences,

that is, the recapturing of a cadential formula The piano R1A:\S theme (or its opening idea, S!, or stamp (usually both melodic and harmonic) if R1:\S is a multimodular zone) normally also that suggests the backing-up to and recovery of — appears in the solo exposition’s S-space, although

the PAC of the preceding module, even when there it frequently becomes only one element of

33. See ch. 8, “Revitalization of a Portion of S- (or FS-) maintain this, in which case the R1:\EEC would be

Material after Stating a New Module.” considered to occur at m. 44. Our current preference is 34. In K. 466/1 does the similar bass motion of mm. to regard m. 44 as the R1:\EEC, although a reasonable 42—44, 51-53, and 56-58 suggest the conceptual trig- case could also be made on behalf of m. 71. gering of a refrain cadence and consequent R1:\EEC 35. “Production of an Additional MC-effect or Nondeferral to m. 71? Or is the similarity insufficient to elided Cadence Shortly into Presumed C-space.”

The Type 5 Sonata: The Opening Ritornello 493 a trimodular block, most often the final element, sionificant ramifications for one’s interpretation TM?, where TM! is a new theme for the soloist of the relative strength and placement of such

(see chapter 21). Later R1:\S-modules (which later structural points as S1:\EEC, R2:\EEC, can carry such labels as R1:\S!-2, S!-3, S2, and S3:\ESC, and R4:\ESC. In turn, this underso on) might also appear in Solo 1 (S1), as part standing invites one to larger hermeneutic conof the solo exposition, although it sometimes siderations of structural completion or nonhappens that they are suppressed there, mak- completion at the various points of the Type 5 ing their next appearance(s) only in R2 and/or movement. R4?, to be followed, usually, by R1:\C-modules

in the expected order. Put another way, even The Closing Zone: R1\C

while S1 interlards, and sometimes replaces, R1’s materials with new ideas of its own, to the Once one has decided upon the location of the extent that it does refer intermittently to the R1 R1:\EEC, whatever remains within R1, up to succession, it does so in modular order—while its concluding perfect authentic cadence, will retaining the option of tracking them through occupy R1:\C-space.°6 R1:\C zones, especially only part of R1’s S-space. Thus Rotation 1’s in the later concertos, are variable in their use of original R1:\EEC might well not be “reached” ~—s juxtaposed forte and piano modules. While most

in S1, even though S1 provides a differing, and Ris come to a reinforced, forte close in the tonic suitable, S1:\EEC on its own. When this hap- (however brief), others end R1 with a sudden pens, Rotation 2 (S1 + R2, providing the sense diminuendo (Piano Concerto No. 24 in C Minor, of the larger exposition, as opposed to the S1 K. 491, with fatalistic implications) or a brief solo exposition) will typically be completed, or — piano module sustained all the way to the end at least advanced further (almost always to the (Piano Concertos No. 11 in F, K. 413; No. 15

equivalent of the R1:\EEC-point), in the lat- in B-flat, K. 450; No. 20 in D Minor, K. 466; ter portions of R2. The full rotation, including No. 27 in B-flat, K. 595). all R1:\C-material, will always be completed The beginning of R1:\C can also be handled in final, “synthesis” rotation: the R383 + R4 in different ways. When the cadential material

complex, or the larger recapitulation. for R1:\S is delivered in a piano dynamic and Most of the rotational implications of this elided with R1:\C, the latter zone often begins situation have been introduced toward the end with an abrupt and vigorous forte. (Occasionof the preceding chapter, in the context of the ally the presence of a strategically placed forte last of Mozart’s concerto arrangements of J. C. module in R1, elided with a PAC, can be helpBach, K. 107 no. 3. That discussion will not be ful evidence on behalf of the preferable locarepeated here, particularly since it will be revis- tion of the R1:\EEC when other factors seem ited also in chapter 21. We might only recall the insufficient or non-normative.) In most cases, central principle: modules that were understood the forte R1:\C is a new theme—1in other words,

as appearing in S-space within R1 will also be it is not P-based. In part, this may be because considered to recur in S-space in later rotations, the R1:\C! idea is the second most common even if they reappear only in R2 (thus reopen- modular choice (after R1:\TR'-!) to provide ing S-space within the larger exposition, even the opening idea for Ritornello 2—thus from though S1 will have declared an S1:\EEC within the outset of R2 continuing an otherwise “inits own spatial limitations) or R4 (reopening complete” larger rotation (see chapter 21). To S-space within the larger recapitulation, not- preserve this possibility and its sense of rotawithstanding an earlier S3:\ESC). Deciding tional clarity, it might have been desirable to where to situate the R1:\EEC, therefore, has avoid a sense of P-redundancy at the onset of

36. An unusual concluding option may be found in several measures of caesura-fill as a bridge that elides Violin Concerto No. 3 in G, K. 216, which, in effect, into Solo 1 (m. 38). comes to a half cadence close in m. 34, then provides

494 Elements of Sonata Theory

R2. Only a few forte R1:\C! ideas are P-based. torical punctuation. (In several early concerThese include those of Piano Concertos No. 21 tos—and in some later ones, such as Clarinet in C, K. 467, m. 64, and No. 24 in C Minor, Concerto in A, K. 622—this module will be K. 491, m. 63, both of which are unusually shot reheard shortly into Solo 1, as a tutti-interpolathrough with regular, idée fixe recurrences of — tion confirming gesture marking the end of the

their P-based mottos. P-zone and releasing the sonata process into TR But the forte R1:\C! is hardly a necessity. concerns.) Normally R1 is fully closed off from When a discursive R1:\S-space has proceeded the nonelided S1 that directly follows it. Occathrough two or more dynamically contrast- sionally one finds an “early” solo entry overlaid ing modules, it can end with a decisive I:PAC onto the final bars of R1, especially if R1 ends (R1:\EEC), nonelided with a piano response or piano. Such a procedure dramatically shifts the “afterthought,” often witty, wistful, or rumi- spotlight to the soloist, whose own construcnative. The presence of a nonelided quiet mod- tion of Rotation 2 is about to dominate the next ule at or near the end of R1 1s a typical Mo- phase of action. These and other related initial zart fingerprint (Leeson and Levin’s theme 6), solo entries are dealt with in chapter 21. though it, too, is not absolutely required—and is sometimes elided to a concluding forte flourish

(their theme 7). That this module can serve as Recurrences of R1 Modules

R1:\C! is evident from such circumstances in Later Rotations

as those in Piano Concertos No. 17 in G, K. 453, m. 58, with succeeding flourish module, One expects that every module within R1 will R1:\C?2, at mm. 69-74 (example 20-1); No. 19 normally appear (or at least be partially reprein F, K. 459, m. 62, with its famous effect of | sented) in a properly situated location somebuffa chuckling, reinforced with two conclud- where in the later rotations. To be sure, some ing forte chords, mm. 70-71; and No. 25 in C, modules (R1:\P) will almost always appear in K. 503, m. 82, with similar forte strengthening every subsequent rotation (exposition, recapituat the end, mm. 89-90. On the other hand, lation). Other R1 modules, however, may fail one might also conclude that the nonelided pi- to appear in the recrafted Rotation 2 (the larger ano module is tucked within an already estab- exposition) but resurface only in the synthesis lished R1:\C-space. In these situations one will — provided by the final rotation (the larger recahave decided to regard an earlier cadence asthe __ pitulation). This can happen when S is comR1:AEEC, following a strict application of the pletely reconceived in Solo 1, or when Solo 1’s

first-PAC guideline. (Even so, once again we S-space does not fully track through all of the admit that such situations offer temptations to R41:\S-modules. If the last few S- or C-modmake a case that that moment of closure may ules of R1 do not re-emerge to conclude R2, also be regarded as deferred through the earlier or if only some of them do (that is, if R2 1s I:PAC up to the point of the nonelided piano rotationally incomplete), the others are likely

theme.)37 be withheld until R4—and often until R4?, afIt is quite common to find a concluding forte ter the cadenza, when all of the previously unflourish that rounds out R1 with fortified rhe- sounded concluding modules finally return, in

37. Examples include Piano Concertos No. 23 in A, den, nonelided piano drop to the tonic minor (“lghts K. 488, m. 63 (probably best heard as a concluding out”), m. 46, and that of K. 595 had opened with still R1:\C?), and No. 27 in B-flat, K. 595, m. 77 (R1:\C°9), another piano possibility—among several others: an both perhaps with the sense of a “blessing” over what elided, crescendo-“loop” piano opening of R1:\C!, m. has preceded it, and Clarinet Concerto in A, K. 622, m. 39. Still, in each of these three cases one locates the 50 (R1I:\C?), with a forte flourish in mm. 55—56. Assum- earlier, “literal” R1:\EEC with some uncertainty, realing the strict first--PAC guideline, one would observe izing that other interpretations are possible. that K. 488’s R1:\C-space had also opened with a sud-

The Type 5 Sonata: The Opening Ritornello 495

their proper order, to produce a complete ro- terior module of R1:\TR. Only rarely will an tation. (In this last case the final R1 modules interior R1:\S-module fail to resurface later in will never have been heard 1n a nontonic key.) the piece. This does happen, though, in ClariFrom time to time one comes across a mod- net Concerto in A, K. 622, in which R1:\S2, ule in R1 that appears only there—one that 1s mim. 31-39, a reinforcement of the preceding abandoned after R1. This is especially possible theme, is unique to Ritornello 1.98 within pre-MC modules, for example, an in-

38. Cf. Piano Concerto No. 14 in E-flat, K. 449, in Concerto No. 2, where the first-proposed “wrong-key” which the emphatic R1:\S?2 appears in both Ritornello R1:\S—the R1:\P!2-based mm. 43-circa 57 (begin1 (mm. 63-70) and Ritornello 2, which it opens (mm. ning in bII])—does reappear in the development, mm. 169-76, in V), but fails to appear at any later point in 236-39, but is omitted from the interior of both the exthe movement: it 1s never recapitulated. Perhaps some- position and recapitulation, although similar harmonic what similar, strictly considered, is Beethoven’s Piano shifts may be found in each structural space.

,. CHAPTER TWENTY-ONE © PRES FLT O

The Type 5 Sonata

Mozart’s Concertos (Solo and Larger Expositions: Solo 1 + Ritornello 2)

Corre(R1) 20 outlined aspects of Ritornello 1 derstood not of primarily “by itself” in the reconstruction in Mozart’s concertos, lationships its particulars to the but Ritornello 1 the initial, nonmodulating rotation of the Type precedent. This 1s a fundamental difference, one 5 Sonata. Especially in large-scale pieces, that that complicates the consideration of every feainitial tutti establishes the movement’s prevail- ture of Sl + R2. R1 is the model against which ing character along with its subsequent interplay all that ensues is to be understood. It 1s already of contrasting topical nuances. Even while of- what has indelibly happened in the movement’s ten retaining a preparatory feel, R1 also deter- past. mines the rotational sequence of modular events likely to prove decisive for the rest of the move-

ment. The task now is to set forth a “real” so- The Soloist Enters: The Interaction Begins nata exposition that will accomplish an obliga-

tory modulation and secure a nontonic EEC. . _.

H STN tic I ESTES - Individual G At the end of R1, the orchestra hands over the ONTOS!and LEASES. beginning of that task to the soloist, who will In a Type 5 sonata the initial solo entry is a drabe spotlighted in most of what follows. For its matically charged moment. After the prolonged part, the soloist is expected to respond to what preparation found in the orchestral R1, the so-

R1 has already made concrete. loist qua individual—from whom the audience This back-referential feature makes a Type has been waiting to hear—steps forth to inter5 exposition different from those in Type 1—4 act with the modular introduction/proto-exsonatas. In all of the others, the exposition is position/rotation just furnished by the group.! autonomous, setting forth the particulars of its Very rarely, the soloist might already have parAnlage (layout) exclusively on its own terms. By ticipated with thematic touches—solo interjec-

contrast, in the Type 5 exposition (normally tions?—in brief spots of R1, as, famously, near S1 + R2) the expositional layout is to be un- the opening of Piano Concerto No. 9 in E-flat,

1. Cf. once again Charles Rosen’s remark cited in ch. insertion within a broader ritornello pillar (R1, R2,

19, n. 64 and related text. [R3], or R4, apart from the standard cadenza)— or, 2. A word on terminology: we regard any brief solo conversely, any orchestral insertion, coinciding with 496

The Type 5 Sonata: Solo 1 + Ritornello 2. 497

K. 271, mm. 3-4, 6-7, including upbeats. (In turing interruptions, moments of submission Beethoven we also have the exceptional be- before authority, dissolutions of texture, and cinnings of Piano Concertos Nos. 4 and 5.) In the like.4 these instances, hearing from the soloist “early To which larger social ends were these soin the game” could suggest any of anumber of — lo-group interactions staged? Were Mozart’s authings: an initiatory demonstration of the sonic diences (and are we today) invited most centrally forces from which the rest of the Type 5 sonata to identify with the emotional stance of the solowill be constructed; a poetic solo-setting of R1 ist, thus helping to construct their (our) “modinto motion; or a representation of the soloist’s ern” sense of subjectivity around the eloquent impulsive or egoistic impatience to begin a pro- presence of an individual voice? Or were they cess of self-assertion before the generically ap- (and we) also identifying with the more com-

pointed moment. posite sonic image of the claims of both the inBecause the “classical” repertory arose dividual and the group? That there is an implied within the social context of a rapidly emerging discourse between group and individual within sense of subjectivity, it has always been entic- each of Mozart’s concertos can scarcely be deing when confronting concertos to speculate on nied. This is also a social discourse whose inner the broader significance of the interaction be- tensions have resonances with such things as tween the individualized soloist and the orches- the rational and ordered basis of Enlightenment tra, interpretable as the social group that makes thought; the charged interplay of old-world (anthe soloist’s utterances possible, the group out cien-régime) and newer-world values; the beginof which the soloist emerges and with which nings of the “structural transformation of the he or she subsequently engages. In 1793 Koch public sphere”’;° the rise of the modern concepwrote that while a solo sonata might impress tion of the highly personalized, individual artus as “a monologue in passionate tones,” in a ist-as-genius, bolstered by a complementary sup“well-worked-out concerto” we find instead port group of culturally elite connoisseurs; and, “a passionate dialogue between the concerto within the concomitantly emerging philosophy player and the accompanying orchestra,” an of Austro-Germanic Idealism, the ideology of interchange of feelings, “something simular to an increasingly “autonomous” art music that was the tragedy of the ancients, where the actor ex- starting to claim inroads into higher expressive pressed his feelings not towards the pit, but to truths. All of this is fertile ground for close inthe chorus.’”? Starting up with the initial solo terpretive work within the concertos, either a entry, this dialogue can range from various de- sympathetically grounded hermeneutic inquiry grees of mutual support and reactive affirmation or a more skeptical ideology critique—or a proto more self-assertive or tense exchanges, fea- vocative combination of the two.°® the dropping-out of the soloist, into one of the solo tor-chorus thesis in Irving, Mozart’s Piano Concertos, pp.

blocks—as an interjection, something momentarily 2-6. inlaid into a broad section governed by a contrasting 4. Cf. ch. 19, n. 51 (the monograph on this topic by textural-generic principle. Much of K. 271/1, for in- Simon P. Keefe). stance, is dominated by a “solo-interjection” game, in 5. The reference is to Jiirgen Habermas, The Structural which the pianist repeatedly intrudes into space norma- Transformation of the Public Sphere, trans. T. Burger and

tively given over fully to the orchestra. This happens F. Lawrence (Cambridge, Mass.: MIT Press, 1989); in R1, R2, R3 (the recapitulation, with the orchestra Habermas is also mentioned in this context in Grayson, and piano parts famously switched around), and R4. As Mozart’s Piano Concertos Nos. 20 and 21, pp. 5-6. will emerge in this chapter, it is also standard practice to 6. The relevant literature on this is vast. For a review of include a few orchestral interjections within the S1, S2, some of the positions taken within Mozart’s concertos,

or S3 stretches. We describe these as brief “tutti inter- see, e.g., Joel Galand, “The Large-Scale Formal Role jections” rather than “ritornello interjections,” in order of the Solo Entry Theme in the Eighteenth-Century better to distinguish them from the structurally more Concerto,” Journal of Music Theory, 44 (2000), 381-85, significant ritornellos, R1, R2, [R3], and R4.) 440—41; and Grayson, Mozart’s Piano Concertos Nos. 20 3. Koch, Introductory Essay on Composition, trans. Baker, and 21, pp. 5-7. p. 209. See also the expanded discussion of Koch’s ac-

498 Elements of Sonata Theory

Option 1: Solo Enters with R1:\P Material open with an IAC, not a PAC. The expansion, (Onset of Solo Exposition and Rotation 2) mim. 91—94, backs up to sound a decorative rep-

etitionsolo of the preceding four bars, this of time Theo..initial entry after atthe completion P(m. 6proP . . 1: ducing a I:PAC the end. This cadence 94) R1 can treatedto inthe a variety of ways. Within oa, , cok tranat . isbeelided opening of the “original” Mozart’s concertos the first-level-default proce- _ , ; , , sition,inRi:\TR!! Gm.concertos—is 94), only nowfor sounded dure—especially the earlier . piano. Other examples of P-expansions be the orchestral R1 to,come toPiano a full close, No. with12oo, , K.may found in Concertos in A, 414; final caesura, whereupon the soloist enters (not . . . considering he cated. | sed and No. 26 in D, K. 537. In one instance, Piano

; - n- . .

nsiderin re any non-notated, improvise .

Einoan «lend in”)? by soundin he ope Concerto No. 23 in A, K. 488, the Solo 1 ver-

, Sans , the y , SR1:\P . P sion of R1:\P, mm.tonic. 67-82,Inisthis shorter than its R1 ing of theme in the case . . . oy tes > ; presence” version, lacking its earlier two-bar reinforcethe soloist’s “subjective announces it- the cadence. . Lod . ment of self by the initial idea of R1.inSince . . con. . . isaccepting Intheother cases—particularly earlier R1:\P!-! normative marker of the onset . . .rotation, . . certos this or those with smaller-scale R1s—the of a new results in a clear articula,brief ;,, .tion . . ofmodel furnished in Rotation 1’s R1:\P is the beginning of the been solo exposition and, a: . .it, and may also have subjected almost atofonce with Rotation 2—in most cases a recasting . . . to1,an R1i:\PS>TR merger. Soloand 1 may begin. by Rotation with substitutions, alterations, , , tracking through the opening bars of the R1:\P 4. . theme only to proceed to an immediate expanThe soloist, lightly accompanied by the or. , . , sion by merging into new material on its. own— chestra, may sound the R1:\P theme . . from the . ; the firstwith of many soloisticintact, deviations bar-for-bar, although often personalized . La: . . . . R1 early within or perdecorations and,model—either occasionally, a more active dia- - P-space .

expansions. .

oo. . r re isn e piano Concerto No. 1 in D, K. 412; Piano Concertos ,

. 2. haps at the original point of the P>TR merger.

logical participation of the orchestra toward the While this procedure is not typical of the p;

end (Violin Concerto No. 3 in G, K. 216; Horn P , yP , P , . concertos, instances of 1t may be seen in Violin

Nos. in in B-flat, K.K. 207; No. 2 in No.,8, Concertos in C, K. 246; No.114 E-flat, 449; , , , D, K. 211; No. 4 1n D, K. 218 Gin which the

. . :\P-tracking for K. e1595). ars,5 mm. 42—49, No. 19 F, K. 459; 27lasts in B-flat, mene IB rein, , inbefore theNo. P-deviations set 1n); and,BOE in an No. 16 in D, K. 451; No. 18 in B-flat, K. 456; R1\P-trackine lasts for eicht b 49-49

Sometimes _this Solo variant, 1 restatement is expanded . onset . genious the non-normative of at the end, or subjected to a at varied repetition _ the Allegro aperto solo exposition, m. 46, of of its final The extended example 21.1is. .not . . No.module. 5 in A, K. 219 (which, exceptionally,

. . works.

. . e soloist’s first entrance);® and in several other

includes this moment of Piano Concerto No. 17 the soloist’s first ent ):8 and i loth

in G, K. 453, which can be compared with the R1 version of the same theme at the opening of example 20.1. The “added-upbeat” figure in the piano, m. 74, probably recalling the practice Option 2: Solo Enters with New Material of improvised Eingdnge at this point, bridges the Preceding the Onset of R1:\P: Links, Bridges, caesura-gap at the end of R1—an exception to and Prefaces (“Rotation 2” Ambiguities)

beginning thediscussed downbeatinofthe theliterature, ... 8 , 8 y Asdirectly has beenwith much theme itself. Mm. 75—90 present mm. 1-16 of . several of Mozart’s concertos call for the soloR1:\P, a compound sentence, but m. 90 1s kept . Lo. . ist to make an initial appearance with a new 7. On Ejingdange, see Grayson, Mozart’s Piano Concertos preceded by an interpolated Adagio preface led by the

Nos. 20 and 21, pp. 101-4, and Irving, Mozart’s Piano soloist (to be revisited under option 2) but it also fea-

Concertos, pp. 162—64. tures a new theme in the solo violin (in dialogue with

8. K. 219 presents a different but relatable situation to option 3) soaring on top of the orchestral replication, those found in the first, second, and fourth violin con- mm. 46—54, of the first nine bars of R1:\P. certos. Not only is the solo exposition proper, m. 46,

EXAMPLE 21.1 Mozart, Piano Concerto No. 17 in G, K. 453, 1,

mm. 71-187

A BS Oe |g Poet i tg Gy 71

[Allegro]

Piano [fase

Orchestra [Hey 1€ #0 08 hesFeHhHs hee

- 7 = Pot | zr! F Pho. AaOF ee oO » try|pmM a aaOO Oe /4

ps _ === ! os

aA eGa

8 a ee Orch. [es Os ) ed

f) P te eo a aN ——_

77

"el spr! eff(mm a

Pno. _————_——————— a a

i A en ee ee fo} ee eee 2 Lee 2

PN | UP

79 rae & — —— | iy oe

oN DD Pno. O_Ae ee eee ee ee, ee ee he

NR a

Orch. [hx 1 dod. x Q \ \ = Poo. [A SS or? rr etl =i ie rir 7 ~~

PN f) +

——~

, =, stadt

Orch. |e" _$ ¢ § ot _}—__ ¢—__—

aOrch.a| AyraoS EXAMPLE 21.1 (continued)

34 —— ps ,Se LeTT » te3 2, ; ees ee

a 86a= o _——__ ns

A 4 ———._\ i

p

SS rr F A 4||, 4_9\2 ee Orch. 8 8B Oe wa: 3S \ = * P| \ 2 * \ . Ea ee a US (O(n

eo—= gg 2+ gee Se # # i PEF

Pno ——————————— Pe" —“g sd —"“—ms

= SS eee j jf =. 7" es P

9 — iatneatnl) =

Pno. fh. os ++_#_2__§ ——-_—_3 d>S__§€--rt.-+fNjTNRHo.7.~”.....--.....------nNn._ (ROM 8 dswe OO; eh O)dHJHaTTd2]>}Y?TY7]y,-N>}r7TANTNY—

RUF 8 ___ |

a a: a ns Pno. re S Roe Que ee. Ns[Zzg es aes oe. lg gf Orch. [AS 88 8S 8 -=;*==— A “~———~ 7 p—~ 0 ee nn ee Ca | | aT

LETTS peeJotSse (Ye 9oooaaaeeeew ee, __— Pan e-eyoos,JZ" So Po EE ee "OTT

. Ec ee =; es ee ae —" :nn r r ran ge.ns —_—— a—~

500

EXAMPLE 21.1 (continued) 94

peo

Pno. 1) — 9

(8). _l ——_, dT

Och. |. === 4 SET aaa bee ——— P

96 —— 1 ———= FT. is « « «

a ee Pho a so es i. ee i i a.

p ho At ____,, ___________|__|___|_}_} #4 J, ———————__~—————_o-—__._—_-# y.ti‘“‘CSCSCSC‘CSYT

:[)&=\“Zs 2 ——i,4.

100 a ——— =

AAP4 eeJ FS Dias —— iD ee ee>= ee eepee ee yu! |@,& » | | ti

Ig i

P no. ey r= ow ie i quan Orch Gs) EF OF OO(“LS QO OD . CP ae COeeese—“‘“‘

PY fp — fp 501

EXAMPLE 21.1 (continued)

§ — — 8 ce FC

102 — —— SgPEEPnD| ai hy. Pre |e 2e,, = ote wp _ Poe tk Pno ; pxe/a2 Mg | gglm y_—}_/ _}ié A 1

ub | |

Orch. | ————_ — o tr?

$ $ z —__ . oS Orch, [2 SS 2 7 fp i i

104 /) » = —_—_—_ Ze. == — Je d = a a2f_.ai Pno. LF pg 2 gg , an Bf gg a ID 42 eee ee ee eee Pg | jt ef lit gy @ yr - «cc Py

v 4 he a i LY +pi’E Fo ee — “ 7~*« 5, s+ —_ J a0V¥W—_q_ lr. _

gue | Sv. 106 ——— ae ——, ———

re te Beg Orch. |. 4 — — O nel ligeyt "eel >. 4>(72‘e$ tt— 72 2 108 _ ele. >" AP: ee eea—_ Pho. [45 te Orn, [4 O_O .SS Ec ey Serres EEE I a SS es ee J) “i -e r i 4 - —

i APHe eee eeet=eSs ee ee ee 0He ee eetreeOe eee Pno. a eeoreeieeoe

\, , PBS late Jefe2is

10 | AN Pho. | A. Orch —=eee . ie Le ry=Me {) #

A eS

hyo \ 2. \ TT) ade an ee a a = | ne 8s a avreeeet ee es itewn twYs , Au~ 4 4 _ *

502

EXAMPLE 21.1 (continued)

Pno. se ae Orch. | ey 116 ae —, °. Sieide oe=Tere Ne Pno. OE a Z id am am wwEE Nea ee Nee Orch. ee, Nee” ed ey . 113 Zz Je, — —_—. A— | Pri

Ne A as tie = 6#

33.

|e — 9 r=

119 ‘ oO — tr Pno. NE eg «zat |

fis ) a es 3

ts esees—oS— sess _ }—_ — __ ote ——__ 45

y—* +144 oa —— a ;— ie — 2

. fe) —__. OO —— ——— 122

Pho [fem fe} —i™=*es—siC |

‘bug ¢ ge | | | J¢s yo} 5, ti‘“Cti‘zr

Orch. ey y..—rt= OO ©OR ry| Hay ma me er FEotOO OO fy + ——T lL Cl

ee ae NS

wh: — t 7 Sy ir rr ot * . ° Ls f * a

™~

Orch PMS oreo ren. (faye OOTF FO OO > — FTF Ooo OS. Ho Fo Eo in Le, Fd aeneeiaenenne ||| ee” . , — ‘ , " a 1 OFCS—ASt--"-=—A~n”. CS 7 cry °°» °°» ”””©§=3=6=6hplcgF oF.lULULULULULULUPT

136

Pho. (HS dc _ a hYsv_'_Te [) #

a _ i 4:2. s

QO heee hy |CC eSwos>a ren.pO | CA tei fatto e@___ ti ie Stett or Se | ee P|ea| 8OY or yt pO +a —— 2 _ /)"“"vv"_ = 2 E §$*—*—-"“”X@#@{]WDa-o~Nv-Hododo-xp-pNw{TNVT?—— a oF

* . . Nee’ * * tS 3 - r

139 22a 4

glPte?—_ a Pno.pt gs PS ee ee (tao@ eo tao _? _ # eee ta we _*?__ —*_ee ot

agt| Pd CG SC ||=a EE all “=p _ Qe eee eee eee ee eee eee

Orch. 4 _ 2 o “g “=pe . ta a O |—————eeeeee eee ON. “= OO i) iw |

i” er a SS a a A a A

142 PI > om . . ~~ . . fp 4 — | | ee ee ee ee eee P OE oO’ ooo eee ye O_O ~

ae em Ae:Lt aebe aa id = 9).

no. [tay _@ | 4a a le O2—OTSN’\”””_ OS —~—— oa lor ©@ Og 2?

pS ET = SS

ee OO, | Oren,hTF Ef i eeFEeeEFE A.2uCe eeeeeeeees esSe, ra”~*—eS eeeFoeeeeeeoSeeFeFsam—~¥°éesass ee ee ee s, f) 44 |

[|

505

EXAMPLE 21.1 (continued)

ous,pm .. Ep ey __§__, Pno. | Hey et oa es ooo NS A A a 2eo __ — — | 145

ae sp Th oTOS O_o ———ee Orch. OO Ee, Eee OO F=D Er

iS t =

147 Poo. (es >? OO OT

DOO |

pedGdd2,fe=— ir

OOD Ti 58's Ss odHHd>HdSH,_—S—STSTyTyf.y.y.y._y._———



=e,

Orch. | hay sy — | — — THT r US P74 ar

47 ~ om oe ee ee cA CS SC CC SO SS A A a es, = 4 - S —

Poo. .)s a _])NYYSTW?YS—_awz_ SP [es TW eeeeeeeeeeeeeeeeeeeeeeeeEeheoeeeeeeeeeeeeee

ane ey | 4 |__ | ¢

Orch. Th ‘2 . @ | @ #olt¥,e | @ [rls | @#@ (Tt — »@

a . . i5] : : os —~ A. 2 0 Lo. (RR OOo i?ee 8 er om

" ee

ft r™"]}]pNRNp>NNNNNWW----— Eo eo —a«@”T To CO_r3./CNvNn"™>’>|)>>"0"c"-— — ™

fT Teese eeeeeeeea—eNNNeeNNNNeeEeEE

os . ’

+

SS 2 a A A”

tp 5or____—_¥H4 4 Pe ph gp — rerNNWN Foe iT Orch. |HASy—e NNO Oooo oo |

506

EXAMPLE 21.1 (continued)

153 ESttetee SSFF a ee SSIdle = oe Tepe he Jee

Pno. AT IT EF |

” ; | aw —— Orch, 42-8 gg pp ie

y

NF NN 155 eee ee Dy ST atetttetees | Tlie tesco) | TTT = eT eee Pno. OC

~ —_ A \ , oo _—_ | Orch. |} 4g , ro 4 ———n LE cee =e ee eee See eea “es

a Ne ee 157 a Ss hu >

, a” Pe es es Pe Se ee Aes

Pno. A a T J A y eo F Fr 2 : 78 y ‘De, = eS a Se ete

508

EXAMPLE 21.1 (continued)

ew ae Tian yo | ff = Pro. Ug ye o_O OD] ______________

SS ae tLe a ee

pe EE OT EE FeTTtONnw—lONOSTSNTSTaavNF}NjDRNERTERHTEOHETET o-oo]

SS —_— ee ey—ee—ey—_— ey Re 1c—a—_— A wo a ri —

Yow! Oe a yy» a a fea aee

a

Orch. [H#és;—270TST02~2—d™@]>Hq)Ss_ ne ot

ce re pe Per TEE FP PP PP

Ef _,_§_, ww fT 8ee 2 ~~ cS ~~; Se 17]

Pno. | f{ag3——_—_s—- 4v¥4 JdOoNJ)dJ})})YY |, JT72—_.>)N)NY>])NNy—}|

~

7 AP ee eeeheEeoe ne oe oc inOoo ee aeo A Orch. [Ay

174 ! my Ap al 5 J ore oO i OO Oro? Orch. fAy3———_ +2. Todt |— ——- i Ca

opt fF p--= f= == fo pe

177 a ] . ] pt ym tN tg , _

Ta OS 7>7D>->->2-(AATADDT]J]

rr g =

fA> a0 — A a 919 ee———— ee ———— Pp ___~_ 5 [yy re no. (fey? o_o. Oe @ OT24 EEE 61

ppt

Ate|,rlSr ——— St or ee TT re ro

63

no. OO ———————_—= STS Le

/ eS Om 0 . eye Se ne A SS CA 4 oc 4 “A iPO a a ( — ~ A a. See POO eee” eee EXAMPLE 21.2 (continued)

65 — ss ————— i =F, — _wip ftfftsfthdfttftt.s | ls | ffl ftwh6hTLr[a Ba lia *wawrta iswaso?

Pno ‘Pf. 5b a Fa iaii*azaws”"wa,aws | a; Ff Y*sa*,, * "oso? °° &@,@2,@

Orch ORD 7 OOOO i“ Of U@# OUT

a a,

AT ST zetatatata Pno He ae ,-0_# «eee . Ne a> ae -_.___._ a ee_—_ ee ienaAAne ol ee (Se > =ee vip i !1tsgyaer §#igzger fy i°* 4 yf 2a 2 ee

a OR 9 ae SS

—E—E———EE—————E———————— Orch. OP me

. a CS ce | | t*}

69 a—— WF cn—_ ;4, @o~ .=[tp . oN .awe. f , = fo tt gy _sf ip? se» °° °+«+ *2_Z tT | Wits ezwewlte © Pno hp — Er , es nie © SFrat'se * era = PP rey y ao . . ?Ne! * *Pere Nee| way .*

0 PCC

La . i . i] Orch "see? *¢ee «=| MT | Y- Bll i+ ona °° | °° «I

rCn. | TF ha OO Te eee i _ 4S ——sdT CTt—“‘i*T]SSC*STCOC*d,:CO ON OMeOOOONOCOCNONOCONCNCOCOCOCSCSCSCSCSCSCSCSCSCSCSCSCSCSCSCS*SY

“’ -lee = ars ——8 oo Fl,‘ss 3,—* ee

a

, vyutea 2. oe 2- [ng Ff |?—_* f= §@6§ * a,2 © @ lithe. Ka || Pno,fT ¢ FF= Cart | iCU#4 |S & &FF @FF @h’ Ss -us fey» Tt te O™~—CCCCCCC MN OOOOOOOOOOSSOSSOSSSSSSSSCSC—~*dC =e Nowe!

514

64 = , 2, y i fet

EXAMPLE 21.3. Mozart, Piano Concerto No. 21 in C, K. 467, 1,

[Allegro] ~a eo = A713 ee ee ee ee 8 ee EE 8 Ce eo f = too y we Y t

mm. 64-83

a A ee a a a) — a eea

Orchestra ai tl ~~. — || _y 1 WT. _S— Trea

67 4 : ¥ pl a -_ P Dig a eo = |or fi @@-2 » oo |? oO ae Orch. LO 7 Po r —_ PN 3 ¥ — a 0 PR Leet se a

PP icing 0 JTTF sote geet e et Orch. fa 3Nae OO eee7 Dig OOO

ta a Ne”

, ——— Pno. | Ao er e=eg——S Sr eg Oe ge 73

Grete a A [Ha A SY CS8CS OCS OD Orch. A@ YH? FTE 8 STE”

2‘

76 f 4

ry |oo, “wf,——— vw? |,EES oe toESS ee Pno. 4 reio. EEesEES | Orch MK 3S 8 78 a — = a a a

Pno. oA oA :see Z Z f 2 ek ry TV o-0>>>-—-"™$™9oomwMWw~WN"nnnnz—o—oO2TSMS"]"]"SS" 1098 oao0“T*C$®0uOu0O0Qo oe]

fon

Cl eeeE———————————————————————

ge |

SY eeeLSee OO . oO TI OSuio.a0n]

Orch. SS ——

EE YY a eneONWWOWOWVwVY’0vT__ eeaneraneneteserenenene eee 515

A 0 LS 8 f’ a3 |e

f) D| bee Pro. 8 A 0 eb, ids sae 0 : PF” A tr DESY Oe 516 Elements of Sonata Theory

EXAMPLE 21.3 (continued)

Orch. | if¢@9—kK— >>? > P

p

Orch. | Hféys—— 1H) ——————>/ FF? i Fr

hardly be more conclusive, and we are to ex- the action, spinning out a clearly preparatory pect an immediate solo-entrance at this point. “warmup” elaboration on the dominant, ending But instead the orchestra seems to “realize” on the V7 in mm. 78-79, held by a fermata—an that the soloist, for whatever reason, is not yet invitation for an improvised Eingang.'® More or ready to make that appointed entrance. Bridg- less normal business resumes in m. 80, the being over the unexpected absence with consum- sinning of Solo 1, although the opening four mate grace, the orchestra, in the now-prolonged bars of the statement of R1:\P are taken over by gap between R1 and S1, moves forward at the the orchestra, while the piano (even here still upbeat to m. 69 with elegant “delaying tactics” hesitant to participate?) is temporarily stalled on in search of the apparently missing (or inatten- a dominant-trill above. tive?) soloist—or perhaps, as Grayson has suggested, the orchestra seeks to “cajole the [reluc- Brief but “Thematic” Fill. It can happen that the

tant] soloist into entering.”!”? Contact 1s made short-lived fill-music preceding the soloist’s in m. 74 (following an augmented-sixth-chord embrace of R1:\P can take on a thematic conapproach to the dominant), and, led back by the tour that, more than mere figuration, presents supportive “group,” the soloist is drawn into a fleeting but separate thematic signature. This issues laid out in ch. 20.) As proposed earlier, double- itself: the music drops suddenly to piano after a vigorous loops of this sort are always to be regarded as incapable forte close; mm. 54—56 and the expanded variant at mm.

of producing a structurally closing cadence on their 58—64 suggest a return to a shoring-up paraphrase of own. Instead, they lead to a “breakout” that is norma- the cadence at mm. 48-52; and the emphatic, P-based tively to be read as the continuation of a special kind of material at m. 64 seems designed to clarify the “real” sentence. That breakout occurs at m. 36 (R1:\S!:2)—a onset of C-space. return to the march-like head-motive of R1:\P, which 17. Grayson, Mozart: Piano Concertos Nos. 20 and 21, p.

might initially be mistaken for a P-based C (an am- 48. biguity that Mozart probably wished us to register at 18. Grayson, Mozart’s Piano Concertos Nos. 20 and 21, this point). (In this movement, the head-motive recurs p. 48: “[This held V7 1s] the point of departure for an regularly, rushing into any potentially otherwise unoc- Eingang, which must be inserted to provide a link to cupied space as an idée fixe or “wild card.”) This contin- the beginning of the solo exposition. In the autograph uation proceeds until the next I:PAC at m. 52. Is this the score Mozart left space for this Eingang—a full page, in R1:\EEC? Probably not. Mm. 52—64 are best regarded fact—but, regrettably, he never filled it in. This blank as R1:\S? (and not R1:\C!) on the conjoined evidence page remains a tantalizing reminder of our loss.” of three features, no one of which is determinative in

a /\ ;A> A A P...‘

EXAMPLE 21.4. Mozart, Piano Concerto No. 9 in E-flat, K. 271, 1,

mm. 54-67

54 [Allegro] —— Piano [He5?—b(€ S$ ——————

.a 8a , 5 ,... tg JT,

Sa s

PP A eee eee Oe Oe Oe ae1eae Orchestra | Hf@s?-pIC 1 9EeleOelpOetLOoatOe1

cp 2 7 o_o * ot tt Le —_=—_*. eee

re steel Biadaad

f) > 4» “Nu o 7 | | » by

ee "=n ol we Se we E—EE Orch.fp |Hf¢3?-bp-———"—_\e— os # SLC # @SRO; 1goS ¢—_¢ | _ SC hh PAB e—“‘CCCsCsC‘“(‘(CSCC:Ct*LC OP, et6TC

r RPPRRO EP EP | —————— yy, /

“J ws ~ e = JF Pro ee Seaa Se eeeee er



Sg es NT —s.—"_1 Ore

TA 4 yb} —=, kh, | 9 2 tate a Pno. Tl EE

)OT re A | SS SC —e TT ft ona

’A| Se =

Orch. |[H{és?-p— 4 4+- o_ +H#_ 2 a ,W+-SST_..

Poo. |} AS 2 66

. Oy

Orch. | fey ?=p— | — 7 _ 9-1 0) 0 TR Merger

stances in Mozart, each with its o idiosyn, ,S1:\P-space opens ne . a we ia As discussed earlier, when crasies, mayabe found in(S1:\Ppref), the first movements of . ge . . . with preface it is common at the Piano Concertos No. 8 in C, K. 246, No. 18 in soloist 4 at for the soloist to d and -andth . cadence-point for the to drop out the B-flat, K. 456, No. 19 in F, K. 459, and No. 23 P ! P

, , A, orchestra to re-enter with the also R1:\P-theme, as in K. 488. is aRotation procedure favored . 4. . ifThis starting 2 proper. Once again,.one

by Beethoven, as in the Violin Concerto in D, ee Cobeing ae . . gets the sense of a brief “tutti-obligation” op. 61, and Piano Concertos No. 1 in C, op. 15, oe

fulfilled afterop. an 37, initial solo but ,here No., 3, ,,in Minor, 4 that inentrance, G,theop. 58, theCcircumstances areNo. such tutti serves

and inthis E-flat, op. 73. becinn; eth OF" tional 4. ceo, as aofrebeginning theween expositional rotation. TheNo. effect5of kind transition 1softhat poserotat; Low. ae 45 , As with the situation just described, the initial of ,the first genuine “partnership” of the piece.soon . . after ; R1:\P-tracking usually dissolves the Having just begun TR, the orchestra graciously . . the mu. soloist re-enters and eventually directs hands over the remaining control of it to the . ae . . soloist. For its part. the soloist’s interventio, sic, sometimes with either a new continuation

. For rt, ntervention _—— . pam . . . or differing material altogether (S1:\TR!), now suggests an individualized florification, a . .

. 4. , ., into clear TR-behavior. Reckoned the self-display within a newly “free space,” or a on . . the from oo , , , tutti entrance, the music takes character blossoming into lyrical or virtuosic figuration. .an orchesa. . of a P=>TR merger: what begins as The events found within this freshly opened . . _ .

n n r rches- . .

tral P-statement slidesThey into [R-activity withspace vary. .from piece to piece. may re- . on.

fer back to certain R1ATR moments (as refer- out an intervening I:PAC. This occurs in Piano ence points sometimes omoted by the orches Concertos No. 9 in E-flat, K. 271 (R1:\P at m.

° promp y 63, follow merger at m. 69);37 No. different 20 in D Minor, tra)P or they may a course quite i. K. 466 (R1:\P at m. 91, merger beginning at mm. fromberthe R1-pattern. (As always, one should . . that MC material this “ 3 99-104, with this last bar as unequivocally new

remember that as pre- material this “new . . TR mune P - cotuen in th material); No. 22 in E-flat, K. 482 (R1:\P at m.

music may or may not return the . . K. . aad ; . 94, merger at m. 106);°° No.in24 1n recaC Minor, pitulation: it is under no structural obligation . , , 491 (R1:\P at m. 118—alsofor suggesting aspects to do so.. K. 453’s new S1:TR!-2, instance, is , 4 one-time event that does not reappear.) The of R1i:\TR, merger at m. 124); and No. 25 in neven not r r. } | PP C, K. 503 (R1:\P at m. 112, full merger c. m. medial caesura may be articulated in a number 136)

of ways. The soloist may drop out at the last ,

moment, permitting a new tutti interjection

to articulate the MC with , to S1A\EEC (S1:\TI?) ) to ar ve WI PPEOP! From theappropriate (First) MC structural force. Alternatively, the soloist may (S- or TMB-Space) sound the MC, either alone or reinforced by the P ae a SOME CASES * secon TN ichted dach Regardless of the transition type selected, Solo

ns Ca ae “eae © “109 0 re i. 1’s TR drives toward a I: HC, V:HC, or V:PAC 51 1K 218/; (K. 65. 66). Ties EXATNPNC MC—or, in minor-mode Type 5s, toward their

a 1, mm. 65-66). analogues. Often articulated with a separate or-

chestral gesture (Tutti Interjection 2, SI:\TI, which may or may not have been borrowed from the original R1:\TR), the MC opens the 36. In general, the idea of sounding the presentational 37. Measure 69 is a new breakout-continuation to the modules of a sentence initially heard in R1 but follow- preceding presentational loops. Cf. nn. 16 and 36. ing them with a new continuation—at least temporar- 38. Again, m. 106 is a new breakout-continuation to ily suppressing the original continuation—is a common the preceding compound presentational loops. In this procedure that may crop up in Mozart at virtually any case, 1t is probably also in dialogue with the sujet-libre

point within an S1. strategy. Cf. the preceding note.

534 Elements of Sonata Theory

door onto a new, nontonic zone of sonata activ- R1:\S-module sounded after it within Solo 1. ity, one almost invariably started by the soloist. Any recurrence of any R1:\S-module within This new zone is normally larger in span than the broader post-MC space of Solo 1 is to be was its predecessor in the R1 model. It thereby understood as existing in S-space, even if it had gives the impression of a purposeful expansion been preceded by an apparent S1:\EEC-effect of that model, opening up into new ideas or (a satisfactory V:PAC that goes on to differing

decorative variants. material). To sound any R1:\S-module after a

seemingly decisive S1:\EEC reopens that PAC,

The S1:\EEC: A New Requirement in the defers the S1:\EEC to the next satisfactory PAC,

Type 5 Sonata and converts whatbe one had initially assumed to C-space into S-space. The structural-harmonic purpose of this newly S1:\S-space may take one all the way to the opened space, as with Sonata Types 1-4, 1s to trill-cadence close of Solo 1. In this option the attain and secure the nontonic point of es- considerably delayed S1:\EEC discharges into sential expositional closure within Solo 1, the the orchestral R2. Here the final portions of S S1:\EEC. But here the situation is complicated will probably also have merged into a new, cliby the existence of a previous model-rotation. mactic, and virtuosic display episode (S1:\DE) Unique to the Type 5 sonata, the solo exposi- —another standard, solo-tailored feature of tion (Solo 1) is not as free to declare its own Type 5s (discussed in a separate section below). uncontested EEC, because the most funda- Such a DE 1s also to be considered as playmental aspect of S1 is that its events are heard ing out within S-space. Alternatively, Solo 1’s only in relation to the orchestral model-rota- S1:\EEC may elide with the display episode, now tion that preceded it. Most important, R1 had to be regarded as S1:\C-space; or it may trigger articulated a tonic-key R1:\EEC following its a C-idea that merges into DE virtuosity, plungown proposal for S-space. In so doing, it had ing toward the final trill cadence. Wherever the defined the closing boundary of what is to be display episode might be placed—either within considered S-space in the movement’s later (and or after S-space—in one way or another, the larger) exposition and recapitulation. Simply solo exposition, $1, will invariably articulate its put, any module that the analyst decides is to be own S1:\EEC. (Again, however, if any module considered within R1’s secondary-theme zone from R1:\S appears in Ritornello 2—after the (pre-R1:\EEC) remains as “S-space defining” S1:\EEC and perhaps even after an expanse of (or occupies S-space) within his or her analyti- S1:\C—larger S-space is reopened in that R2. cal reading no matter where it might be found As outlined in chapter 19, a differing R2:\EEC

later in the movement.°? can be provided in the larger exposition, the one As a result, the moment of Solo 1’s essential that continues to be pursued, and is often conexpositional closure, the $1:\EEC, has not one cluded, in Ritornello 2. When it is, the solo exbut two defining features. It is not only the first position and the larger exposition that contains satisfactory perfect authentic cadence that goes it will have differing EEC-points, as discussed

on to differing material (as in Types 1-4) but below in the section on Ritornello 2 procealso the first V:PAC (or III:PAC) that has no dures.)

39. The nuance regarding the analyst’s decision is advis- although some R1:\EECs are unequivocal and obvious able and appropriately timed at this point for at least two (to those using this hermeneutic lens), others are subject

reasons. On the one hand, the analyst does not “find” to interpretation and reasonable disagreement. In part conceptual “objective facts” (such as an EEC) that are this is because of the extravagantly multimodular aspect there in the music for the taking. On the contrary, he or of so many R1:\post-MC spaces that often characterizes she makes those “facts” through the act of reading the Mozart’s opening tuttis—an aspect that invites one to music through a particular conceptual lens, such as So- speculate about a broader or freer treatment of the stannata Theory. On the other hand, as discussed in ch. 20, dard EEC-deferral strategies (outlined in ch. 8).

The Type 5 Sonata: Solo 1 + Ritornello 2. 535

The Musical Content of $1:\S-Space to be taken neither as rigid prescriptions nor as

Oo . . self-sufficient explanations of Mozart’s S-spaces.

Here the principal issue 1s the thematic relation- Instead, they are only groupings of relatable

ni ores RI 8 sae uy Neen 2) - choices, the close details of whose separate reat Ouran. ( soration ) an an alizations must also be assessed within the conmoas matter ‘ch » ditfor waar prow! an oh texts of the movements at hand. Such categories

ae we a ey. ven on ne have the primary function not of explanation

te NTOSTS ran eminent Yo BBteal Oe Na but of suggesting S-contextual information to few instances R1’s proposed S-material is re- the analyst inquiring into any single moveplaced by new S1:\S-material throughout. ment. (Which other movements present similar Sometimes only some of R's S-music appears things? How common an S1:\S-format, broadly mn SL:\S-space, placed into a differing CONEEXT. construed, is this particular choice?) It may be While it is possible to fill ST:\S-space with only convenient to divide these S-solutions into two a single S-idea (either R1:\S or something new), broad groups: those that include some or all of SI's secondary-theme zone is more frequently R1:\S and those whose expositional S-space 1s multimodular and carried on at some length. entirely new. Within each we shall identify subMost often, it stages a mini-drama of contrast- oroups

ing poses, featuring a freer dialogue and more )

flexible back-and-forth exchange with the orchestra.40 Multimodular space of this type can Options in Which All or Some of R1:\S impress one as an extraverted display of thematic | Appears in Solo 1

assemblage, as if Mozart were secking ro stun Mozart adopts this general strategy for most of his audiences with one astonishing idea after an- his Type 5 Allegro movements. It provides an other. Most commonly, Mozart laid out SEAS important connection between the second parts space as a trimodular block (S1:\ TMB), with a of Ritornello 1 and Solo 1, implying that the

second MC-effect in the middle, followed by latter continues to be a free expansion of the a strongly profiled lyrical theme (most often former. This option also underscores the intro-

RIASI, wee oe other, has been faye ductory function embedded to a greater or lesser

in SCEVE a8 b ) i S1:\DE). In another option, elided with a contrasting display episode. Here one might enter into the display episode assumthe suddenly “clicked-on” virtuosity concludes ing that it serves as S1:\C only to find motives the solo exposition as a self-enclosed interpola- or modules of earlier R1:\S-material placed into tion occupying S1:\C-space, even though one the accompaniment. When either of these situmight find within it fleeting references to ideas ations occurs, the S1:\EEC 1s to be regarded as

from R1:\C or other, non-S portions of R1. deferred until the final trill-cadence. ‘This kind (Recurrences of R1:\S material here would of solo exposition will contain no C-space, and keep open S1:\S-space.) There 1s no formulafor the S1:\EEC occurs at the downbeat of the producing a display episode, and Mozart’s solu- elided R2. This 1s the best solution even when tions are increasingly clever from at least K. 453 the display episode contains internal PACs and onward. The general situation, though, occurs shifts of motivic material: the point of the DE, in many Piano Concertos: Nos. 5, K. 175; 8, after all, is that it is a single stretch of “similar” K. 246; 10, K. 365; 13, K. 415; 14, K. 449; 17, music pointed at the final trill-cadence. ExamK. 453; 19, K. 459; 20, K. 466; and 22, K. 482. ples may be found in Piano Concertos Nos. 9, From time to time, as in No. 18, K. 456, the K. 271; 11, K. 413 (an which refrain cadences soloist’s virtuoso figuration has already started keep S-space alive); 12, K. 414; 16, K. 451; 21, toward the end of S1:\S-space (m. 146, with the K. 467 (R1IAS!2 [sic/] at m. 143=S1:\DE), No. S1:\EEC at m. 149), creating a dovetailed effect. 24, K. 491 (again, with persistent refrain caOther examples may be found in Nos. 15, K. dences); and 26, K. 537 (launching the DE with

450, and 25, K. 503. a V:IAC, a quasi-S-appendix, at m. 193). One alternative 1s to begin S$1:\C-space with

a thematic statement of R1:\C!!, which before ,

long merges into the display episode (R1:\C! => The Morphology of Display Episodes S1:\DE). Identifying such cases is highly inter- Smaller-scale works and many wind concerpretive. Much depends on the analyst’s prior de- tos make only perfunctory gestures at the discision about the location of the R1:\EEC and play-episode technique. The Bassoon Concerto,

subsequent R1:\C-space—not always an easy K. 191, and the four horn concertos feature decision. Here we suggest six examples. The merely a slightly activated conclusion of S1, only first two, smaller-scale works, have much re- minimal or no DE-activity. The violin concerduced display episodes, if they are to be counted tos provide more along these lines, but the most as such at all: the Flute and Harp Concerto, K. extended examples are to be found in the piano 299 (S1:\EEC at m. 107) and the Flute Con- concertos. Display episodes are structured in an certo, K. 313 (S1:\EEC at m. 79). The remain- ad hoc manner, appropriate for the mood and ing four are: Piano Concertos Nos. 7, K. 242 local circumstances of the movement in which (S1:\EEC at m. 101), No. 23, K. 488 (S1:\EEC they are embedded. Still, a few generalizations at m. 114, although the following minor-mode about them might be helpful. module could also be regarded here, as in R1, as Most typically, the longer pianistic display an extension of S-space following an attenuated episodes are multisectional, divided into dePAC), and No. 27, K. 595 (S1:\EEC at m. 153); marcated segments often separated by PACs. and the Clarinet Concerto, K. 622 (S1:\EEC One commonly finds a chain of lightly accom-

at m. 134). panied but soloistically vigorous DE-phrases—

The remaining option is to begin the DE stock figuration in excelsis—one or more of within ongoing, and now much-expanded, which might conclude with a provisional (but S-space. Thus it may happen that the thematic “not good enough”) trill-cadence and be di-

The Type 5 Sonata: Solo 1 + Ritornello 2 545

rectly elided with another phrase, producing the final trill (on V’, m. 170) and PAC-triggera “nonstop” effect. The effect is one of inex- ing of R2 (m. 171). orable accumulation, that of approaching the The technique of model-statement and exfinal trill-cadence through multiple stages, each panded recasting, as here, is common in the DEs

either leveraging up the floridity of its prede- of the piano concertos (as indeed it is throughcessor or simply adding another element to an out Mozart’s nonconcerto expositional zones ever-proliferating virtuosity. Throughout all of | as well). These sometimes display a high stanthis the orchestra typically plays a merely sup- — dard of contrapuntal ingenuity, brandishing a portive role, occasionally intervening to spur masterly display of compositional originality the action onward toward the trill-cadence, al- wrested out of seemingly stock materials.°? Simthough from time to time in the later concertos ilarly unpredictable—but much to be looked Mozart elevated some of the orchestral inter- | for—is any evidence that rotationally participaventions or overlays into special-effect partici- tory or other motivic materials might be conpation or (usually witty) “side-commentary.” cealed (or even openly expressed) within the The display episode in K. 453/1 (example soloistic display or its accompaniment. As men21.1) is one of modest size for the later piano tioned earlier, if these materials are identifiable concertos and is divided only into a single sen- _—_—as belonging to R1:\S, then S1:\S-space is being

tential phrase and its expanded recasting. The | extended into the display episode. On the other model phrase is a standard cight barslong(mm. hand, the appearance of R1:\C suggestions can 153-60, launching S$1:\C-space) and features both confirm the DE as existing in $1:\C-space a characteristically nonthematic, exercise-like | and help one to interpret the continued rotarunning-pattern in the right hand. It concludes __ tional material likely to be found in R2. with a V:PAC at mm. 159-60: a familiar cadence-formula decorated lightly with a trill, al- Display-Episode Variants: Playing with Signs

beit one predictive of the grander one to follow at mm. 170-71. The elided second phrase is built That the obligatory display episode is recognizon the harmonic scaffolding of the first, but it is able by its grasping onto stock virtuoso-figuration

now expanded from eight to twelve bars (mm. makes it an ideal playground for the manipula160-71). Overlaid onto the presentation mod- tion of those formulas into local surprises and ules (mm. 160-63, an oscillation of I, V7, I, V7) ad hoc adaptations. Mozart’s occasional decione finds a buffa, Figaro-like gesture in the winds sions to make the DE rotationally participatory (“amusing activity underway”)—an example of — through a partial sharing of R1:\S or C modules

layered orchestral commentary—that recedes illustrate one possibility. Another is the mixing into a more typical subordinate role with the of more thematic orchestral material into some expanded continuational windup to the cadence —-DEs, as with the four-bar buffa-commentary in (m. 164). As for the piano, it changes figuration -K. 453/1, mentioned above. In another example, from “new” arpeggios shared between the hands — Piano Concerto No. 19 in F, K. 459, the first (mm. 160-63), to rapid ascending-scales (mm. portion of the DE (m. 149) is shot through with 164-67) soon converting back to arpeggios (m. = sprouting R1:\P “march-motto” references, 168) to the inevitable plunge into the unmistak- which amusingly seek to close the first phrase able signals of full conclusion: the grand caden- —on their own terms (mm. 160-62), though the tial § of arrival (m. 169), formulaically igniting attempt is wittily undermined by evasion (I°,

59. Fora closer discussion of the wide array of differing Process of Variation in the Music of Mozart” (Ph.D. techniques that Mozart used to produce internal vari- diss., Yale University, 2003). In part, the subtitle for ants in piano-concerto display episodes—imaginative the next section, “playing with signs,” was suggested by elaborations of standard voice-leading structures, the Ivanovitch’s citation, in a different context, of V. Kofi “swapping around of hands” and textures, the regular Agawu’s much-noted Playing with Signs (Princeton, N_J.: patterning of cadences, and so on—see Roman Ivano- Princeton University Press, 1991). vitch, “A Practical Theory of Variation,” ch. 3 of “The 60. Cf. the preceding note.

546 Elements of Sonata Theory

m. 163) and an unexpected “backup” recovery dynamic, and on the “wrong” pitch, 1 (Eb,) inof R1I:\TR! (taken from m. 32). The sudden stead of the customary 2 (here, F,). This permits withdrawal of the soloist for two tutti interven- the audible staging of an adjustment in the trill, tions in Piano Concerto No. 25 in C, K. 503, nudged up chromatically to F,, 5, in mm. 222mm. 195-98 and 202—4—participatory spurs 23, prompted by the orders of the inlaid R1:\P toward closure—is also countergeneric, as 1s the march head-motive in the winds. The mysteriinterruption of the soloist for the gleamingly ously subdued piano dynamics throughout all of lyrical moment for the winds 1n Piano Concerto this provide a sense of unnatural quict, of some-

No. 27 in B-flat, mm. 164-70. thing powerful being held back in what ought The most conventional sign of all is the prepa- to be a climactic spot. At m. 225 the dam bursts. ration for and execution of the expanded trill-ca- Suddenly fortissimo—the moment of decision— dence. This is the final, forte signature-flourish the solo breaks out of its dynamic confinement

that wraps up the Sl-package and triggers the for a spectacular four-octave plunge downward start of R2. (A similar flourish 1s expected at the into the cadence at m. 227, a dramatic gesends of S3 and the cadenza in R4.) As such the ture well described by Plantinga as one of the onset of a pronounced cadential trill serves as a “highly distinctive dive-bomber-like cadences generic announcement that the soloist is now ending the three big solo sections of the first finishing and Ritornello 2 is being cued to en- movement.’¢! What follows is R2 at m. 227, ter on schedule. Mozart’s frequent practice of — here a major-mode restatement of R1:\C3 (from providing “false,” insufficient, or undermined m. 98), preserving the ongoing sense of exposi“early” trill-cadences within the DE—only to tional rotation, since that module had not been pursue that display-episode section into a var- sounded in S1. ied repetition or new module—has already been The idea of a quietly prolonged trill at this remarked upon as one aspect of his strategy of | moment, changing pitches and leading finally building intensity through the frustration of ex- to a decisive concluding gesture, resurfaced in pectations. One amusing deformation occurs in Beethoven’s Violin Concerto in D, op. 61 (mm. Piano Concerto No. 13 in C, K. 415, in which 205-24, with the added twist of a deceptive cathe piano delivers a fully executed, obviously dence into Ritornello 2, m. 224). An extraor“final” trill-cadence (m. 132-33) only to find dinary variant of it would appear, more deforthat the orchestra fails to enter with an elided mationally, in Piano Concerto No. 4 in G, op. R2. (Apparently the trill-cadence was sounded 58 (example 21.8), in which a notable trill (here an octave too low?) The piano finds itself a double-trill) is first attained in m. 166, forte— stranded in silence and 1s obliged to crank up as expected—but within four bars is choked the DE-engine on its own one more time (using down to a diminuendo fading-away for a rapt R1:\C> material) at m. 133, leading finally to high-register piano restatement, dolce e con espresthe more effective trill-cadence (in the generi- sione, of R1:\S!-3. Thus the soloist, at first plung-

cally correct octave) at mm. 147-48. ing efficiently toward Ritornello 2, undergoes Beethoven, particularly in his middle-pe- a change of mind, seeking now to stop linear riod concertos, sought ways of either heighten- time, reluctant to bring such beauty to an end ing this trill-cadence moment or submitting it and wishing to back up for one more statement to surprising deformations. In Piano Concerto of R1:\S!3 in a wondrous recovery (“Wait! Did No. 3 in C Minor, op. 37 (example 21.7), one you hear that theme? Did you realize what it hears a solo-trill for a full seven bars, mm. 219- — meant?”’). Eager to press forward—notice also

25. More to the point, the trill is first broached the crescendo-pressure in mm. 172—73—the “early,” at the moment that the cadential ? of | impatient orchestra, uncucd by a trill (!), cuts E-flat major is attained (m. 219—normally the the reverie short after four bars with R2, rushtrill sounds above a subsequent V7”), at a piano ing in at m. 174, mid-phrase, to complete on

61. Plantinga, Beethoven’s Concertos, p. 156.

EXAMPLE 21.7 Beethoven, Piano Concerto No. 3 in C Minor, op. 37, 1,

mm. 217-31 [Allegro con brio]

aede Viele idles ee . nyNefpNadde ig oe Se eo D> Ps > PA GO > |, © SO

Piano |Hf@s?—pi€_ pee 7-—_____ 0 . —/™ |“ , it bs yt

Sk NS P F sf =_ = =

2190 @& a= = e ie yb —————— es Pno. Hf es Pp) ot i tt

NS NR tT ded ADDAAAAAAAAAAAAAAAAAAAAADAAAADAAAADAAAAAADADAAAAAAAADAAAAAAAAAAAAAADAAAAAAAAAAAADAAAAAAAAADAAAA

|, | —_4 __-#: _$ y © ws = FF wo = Ff a f-

Sy * “2 eee Oe eee eee

eae Og ee ig

223 2: 2. 7 ——y | wf i 3 ry > re 3 3 3 5|

Pao. es =I ddNYdSY YYTY])o9)r7}NY—C_|}WH#_—— 99

a A PR a a

Orch. | H@s?-b—_- 3. Btw dT. ’

Pod aa FEeea P

22) = 2s — Pno. |ifes?->09 Ne a Poe | eee RP A FP a L257]

o

Orch. Cy Pp

Pno. | cA 4 ee p é t x < JJ, . De a ee a a ~~ ee 227

Orch AE —=_=__— OF? —oa—

he he oe ws > tT ez TT

548 Elements of Sonata Theory

ne PA (SO ? -/Gemeente 7 / 1. eeeR? = Pno. EXAMPLE 21.7 (continued) 229

Yee %o

Oreh. > Sg__§4 SF S| pg a[[Gb jo 4 Left _ | eet a . ew ee 2

aad creer ? :

its own terms what the soloist was unwilling to corresponding to R1’s concluding modules have end. The result is an R2 that is not articulated not yet been attained.

as a separate, post-solo-expositional block. In- From a textural point of view, as Tovey stead, the solo exposition does not close at all underscored in 1903, one goal of R2, beyond but moves directly into a larger-expositional marking an arrival-point in the form and movR2 completion. This may also be understood ing us onward to the next phase,®* is to provide as a structural ellipsis at the moment of the an enhancement of the merely “brilliant” cashift, one occasioned by the soloist’s stalling on dential conclusion of Si—as if this display of

R1AS!-3,62 the “powerful in sound” had been built up to by the display episode and triggered by the trill-cadence.® R2 provides the unmistakable impres-

Ritornello 2 and Its Role in the sion of an ardently supportive orchestral celebra-

Larger Exposition tion of the final V:PAC simultancously achieved by the soloist. Because of the near-invariability Whatever its divergences from the modular lay- of this affirmational éclat (““Yes!’’), the relatively out provided in R14, the solo exposition provides —_ infrequent instances in which the beginning of

a complementary commentary on or reaction R2 instead undermines the soloist’s PAC with to that earlier succession. As it does so, S1 also a deceptive cadence, either sternly or amusingly traces its way through a rotation (Rotation (“No!)—usually onto bVI—are worth noting: 2) that should be compared, moment to mo- _in Viotti’s once-famous Violin Concerto No. ment, with R1’s layout, the model to which it 22 in A Minor, for example, or in Beethoven’s responds. Here one finds as a virtually invari- Triple Concerto in C, op. 56 (m. 225) and his able norm that S1 stops short of executing a full Violin Concerto in D, op. 61 (m. 224). rotation. (The relevant concept of fullness was And yet beyond cadential affirmation it 1s also established in R1, bordered by recognizable be- a task of Ritornello 2 to seek to complete the ginning- and ending-modules.) S1’s display epi- _still-ongoing Rotation 2 by supplying some or sode and final trill-cadence, signaling the end of —_all of the missing modules from the end of R1.

the solo exposition, bring us rotationally only It may do this either beginning directly at the to a point that is short of the real conclusion start of R2 (for example, with an R1:\C module laid down by the Rotation 1 model.® Passages not reached in S1) or only after the interpolation

62. One might add, for the sake of completeness, that 63. K. 107 no. 2, discussed in ch. 19, is a rare and odd the Fourth Concerto’s successor, No. 5 in E-flat, op. exception. 73, provides a monumentalized display-episode drive 64. Rosen’s characterization of R2 as essentially a trantoward the end of S1 but replaces the expected trill-ca- sition, Sonata Forms, rev. ed., p. 85, is potentially misdence with a spectacular contrary-motion flourish in leading. the piano, flaring outward toward both extremes of the 65. Tovey, “The Classical Concerto,” pp. 9-10. Cf. ch.

piano’s register (mm. 225-27). 19, n. 21 and related text.

A A

EXAMPLE 21.8 Beethoven, Piano Concerto No. 4 in G, op. 58, 1,

mm. 164-82

——_ =, .ee —— a | rT betel ’ A .Piano a164 ee ee eee ee 2 ee | A betel wi ene ee4eee

a A ee [Allegro Moderato]

Soe” 3 -

A4K Orchestra | f{@y— Cpt) ———— b

—— t+ee/. 0bEE bOT:eS # Os 4ls | etf)Eo le UU ULLUwal Oe Pno. |ites—— pg a A PS ——— ———— 165

Orch. teg3— ee 8 a|© ~*~ P

Y

AA |DR Orch. {Qo F Mp DAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAAADAAAAAAAAAAAAAAADADAAAAAAAADADAADAAAAAADAAAAAAAA ADE

166 re

YE? ete , EP Pno. | if¢);——— = ot 00 sss ee ee ee Pe gh» _ _—————S—CSC—CSs

lel 6ASccdade

i)

h 4 $s

167 fr

Poo A otetecre te fet, rr eS SS Cr ES eee

169 +3 tg ie o ; _f ie te ee Pho. [és oe i i — Be — ——————————————

dolce e con espressione

pNe. . OO

Orch . yr Le TT TT ao aT 4A = J7N7vwa—a 77777

io aa ee ne 2 8 a oa ms _ eee Ao tr ee ee rit, oe , cCrese. Ze— POO aS a — a — eo —

EXAMPLE 21.8 (continued)

PF A ae = a ae -ggiis gr

Pno. ian Ee Eee ee eee eee ee eee 622

173 : — = = = - c= 2 — 2 Pno. | if@s——t eer ie og ir

A a CC GQ ee ae a ee OD ( Orch. [fe eee ee ee ee

A A_ i)

4 | fé5_t}———— eee Pno. 174 ¢

ly) — i or If (a tempo) toe, =. Ayue— fandte.A pees?

Ley eT i Oe.MBB A ee OR 175 2 7o.a—“oo = _—

5A, a Re _ a _ Shh i? 2 _ HF 9g Se or =

Orch | LEEP LEE LEE ter —_—

Sf

The Type 5 Sonata: Solo 1 + Ritornello 2 551 EXAMPLE 21.8 (continued)

he 7ee ee 7 (mm. 73-76) but had sup- This alternative is the one most frequently enpressed the expected R1:\S!-2, a frequent ploy countered in Mozart’s concertos. Here the roin Mozart, in favor of a differing continuation tationally participatory modules do not begin (mm. 77-81, producing the S1:\EEC at m. 81). at the outset, although they do appear several Elided with the DE’s trill-cadence, R2 jumps _ bars later. One presumption would be that in in at once with the missing S1:\S'? at m. 91, style or tone these later modules, perhaps more which produces its own V:PAC at m. 97, clos- continuational or codetta-like, could not serve ing off that newly revived S-space.® This isan _as effective R2-igniters. Since their appearance EEC -effect pertaining to the larger exposition, 1S delayed, the ongoing rotation 1s temporarily overriding that of the earlier S1:\EEC. We refer suspended, while the orchestra sounds an appro-

to such a moment as the R2:\EEC. priately spring-loaded, forte outburst to launch 67. A larger account of the logic behind this was pro- 69. Cf. the more analytically challenging situation in vided in the K. 107 analyses at the end of ch. 19. Piano Concerto No. 14 in E-flat, K. 449. 68. Unusually, a “new C” afterthought-tag, not heard in R1, follows this to conclude R2, mm. 97-99,

The Type 5 Sonata: Solo 1 + Ritornello 2. 553

R2 in proper style. This type of R2 subdivides light “the antithesis between one and many,” of into two portions. The first is a rotationally inert making “the best effect expressible by opposed or neutral space that in principle could be filled and unequal masses of instruments or voices” in with anything, although the material most — through the climactic vitality of its local, emoften interpolated here is the assertive initial phatic onset along with, we might add, some module of R1:\PR, devised in advance within implication of a command to drive the music Ritornello 1 to take on this role. This merges, forward to the next stage.’! sometimes mid-phrase, into the second portion, Although Mozart usually mixes complicaa rotationally participatory section, a recovery tions into each of his R2s, a sense of the normaof R1:\S- or C-modules, which may or may _ tive may be gotten, paradoxically, from an exnot bring Rotation 2 to a full completion.” pressively extraordinary first movement, that of When confronting this kind of R2, one needs the Piano Concerto No. 20 in D Minor, K. 466 to pose some basic questions. Is the rotation- (example 21.9). As is customary, Ritornello 2, ally inert module that begins Ritornello 2 the in F major, elides with the final trill-cadence of standard one, R1:\TR!+, or is it something else? the display episode (m. 174): a dovetailed tripWhere is the precise point at which it shifts into let-upbeat in the bass gives the elision an exrotational R1:\S- or C-material? How is this tra push. Mm. 174-84 revisit most of R1i:V\TR merger accomplished? What R1:\S- or C-ma- —which in this case had not appeared in the terial is it? And is the remainder of the rotation solo exposition—turning its originally “deparallel or nonparallel with the ending of R1:? monic” minor-mode model (mm. 16—26) into a That this use of R1:\VTR!-!—or other alter- still-turbulent major (though notice the “grindnatives—contravenes the expectations of nor- ing” minor-mode mixture in mm. 179-80). At mal rotational succession (“out of order’) is not the downbeat of m. 185 the music splices to a caan interpretive problem. One should neither re- dential figure that had been typical of the end of gard its appearance at the opening of R2 as puz- what we regard as R1:\C!3: mm. 185-86 (comzling or random nor suppose that it challenges pare with mm. 64—65, 69-71). Since some of the larger theory of rotations in any fundamen- the preceding display episode had been based on tal way. Instead, one comes to realize that we R1A\C!! (m. 153 = m. 44), this R1:\C!-4 module are confronting specialized modular behavior 1s one of the next modules in line: shifting into within the Type 5 sonata. In some of its realiza- even part of it atm. 185 changes what had been tions the Type 5 sonata 1s outfitted with a few rotationally inert in R2 into something rotastructural slots—a zone of expanded caesura-fill, tionally participatory. The correspondence meaperhaps, or the opening sections of R2, R4! sures are now pursued further, bar-for-bar. Mm. and R4?—that may be filled in with music that 186-92 provide a major-mode version of R1’s carries no immediate rotational implications. last module, the final “afterthought” R1:\C? Within the Ritornello 2 option under consider- (originally in minor, mm. 71-77), a close to R2 ation here, the first portion of R2 1s to be con- that is rotationally complete and rhymes with strued as a free zone or blank whose contents the ending of R1. (In this case, the corresponneed not be held up to the scrutiny of rotational dence bars continue beyond R2, as the developexpectations. Its main purpose would seem to mental-space S2 opens with a redeployment of be textural, along the lines of what Tovey called the S1:\Prref, m. 192.)

“the concerto principle’—that of engineering A more complicated situation—arguably things in order to display with self-evident de- more “typical” in its idiosyncrasy—may be 70. Irving, Mozart’s Piano Concertos, p. 46, sensed the 71. Tovey, “The Classical Concerto,” pp. 6-7. Our general pattern but was unable to make much of it: view would make something less of the notions of “an“The choice of material [for the second tutti] is impos- tithesis” and “opposition’’—or at least to downplay their sible to categorise exactly, although there is a tendency potentially hostile implications, since, one presumes, to ‘telescope’ phrases from near the beginning and near cooperation and staged interplay of differences are also

the end of the first tutti.” expressive options.

8

a = ) =ae, ° | A ee PM EXAMPLE 21.9 Mozart, Piano Concerto No. 20 in D Minor, K. 466, 1, mm. 171-95

/PB | | 1aia J e« + * a a a aT [Allegro]

pee @ Po ee [EAee€62@IeeoF Orchestra I7l ———— —————

. Tw — lO ESESE__E__ LL BQ ubkzubzS5EEEEM Piano [Hfes2{€}_--74 he ?ejirpgbgeEEE i a rr|

eee eee * | — — — _ — ,_ — e

Po. | (ey? _ fF ——————————————————— 1 _

et CL PL CL

: = = yy === t= Pho. [Hea? oF 7;

Orch. [ta?—44N7T-TTTTNYNY77W7N72#1 OWT

174

=z ff. ! a” |Se A ee Psieee ee i, > Orch. Hésy? eeees —e—s—ss Y

Bp tre 3 ce

Pp “es wl | Oe Si ee a el l—“em

177 a nn p>}, —, 9 Seg Orch. fay? tf -8

— — BT 2 ttre eS eR ES | eo, (~~ 4 oT 4d y ce “ST SO, DE SD Orch. PSyt oeCre -- 7re De —_ i we oe 7oe 4

eo a o SX?

180 ( vy 4 pr = Se

—_amo P ——_ ——_ a. po a =~ Mg pre gr dre A afay? Seet A Orch. a a |7 pr pry a ae , NP id

Neg | —=em | O—ee Ore Ur lOc rFTTetts—*— @ otoe, *; oe ee Orch. OE sd 233

e eeettes Fr ore ee rs

8 a Gy

236

———Pro.f)ACee —— — - rgeee “ss(on 5 ¢gCea io a: ‘

Orch, | p4R ett ue er Y) 4

» =, AS Pho. 0ey esrSF? OYwo, 0 7)ye 7a, FSst 6|” 2s (Oy |[Axe FRR, Pho. eee RO) tHs fo _ft2ee yyEF dd TY) —_ — & r F F A FS | A 0 a Orch. [Re 0 eo = eer ow He). O4x°- 4 ——~ 7 72 a ys } : — — i b —_ A ie ro J |

eC 241

244 r i 7} a Wa | oom | or bh | he | oT.

Y) 4 — 4 — :

- i —_ te S_" bee i?‘ bol ie | ' r Orch, |} Arete ue ae 247 he — | dba a — 4 ne A > Fhe — he. Pno. f¢3—t eee EEE

Pno. ey on eo Se ow ee COS OO

LE an a rr a ae a f) 44

Po OT Taw SK © 2 0D = ~~ —_ r Pr i 7 25] Pl t -_ EXAMPLE 22.2 (continued)

0” J,121 55@ oF -#2. oe ry ©2, J, grate dl tetbetee fe OO "irr r—c-+-——_* @CI) @* ==

SS OT Sn Tort == Sa

ees? __, , @ 0 “FP ot, EE —ee——————

aa

Pho. ([KA sh —_ 1 — _* er]l0NY™ OT en —oT—*£$#xX” Hea A) PE se t+ 8: ; oeeee 9 goee©*~ $ $eeee 3 94 © -ee £*ee se

ee a, ee CS eee A Qe in

97 —~ | ;

Bsn. | G42 oo oda ———— fon

~ Or ~ y. a om Fs ms Sa a a gs Orch. Fe fp —S m+ ae Od 5 Oo} — LS lpr rrrrrry were? P.-C

f)© 2a. a ' o ss SfCEEP CEP Ceres oa"

100 Solo — Bsn. (EAS aH dy

tr SOLO ) | | le es eee a 2 Pe a ee «= ee ee ee a es eeee es es:es es es BA = ee nt LT ee EY Pe Oe Oe Oe ee ~

peracer Torrespt ryefe

103 . o 4 Sa a er DS ee a Gh DO

oN, mo A. f) ee TT ff Orch 7 we oo oad de

: r : r bw : t ge :

EXAMPLE 22.5b Mozart, Concertone in C, K. 190, mm. 152-62

oo. 4sPRE) f4e, fa - |] [Allegro Spiritoso]

ES | } af AQe he cee 154 TUTTI pt > }-__—__ : esate pe JT «ce, ereRD geGO = | ee ee a. BFOPH ea Orch PD OO

Principal Violin] & Il [b4gge-?———-* +_# -o—_$ + o_o

——_—,z ri —. 2) 5

Orchestra [Has t€+H4NNTIY—O NH dpdAH——J!._-TY————

Sa

Vin. 1 & Uo K6SspaSadv>d[]NTNYNY TTT, © _&_ 6 _#__, pS

t

SF cr, (Oe ary ota all(pe OD - o- all all

fp| &K y+II+PE g T Vin. 156

)

aa MBB A

a! #4 io.+¢~ we J re____ ee P. eee ee+1«o-< Orch.| fm £" +eee _@__| 7 4eee P| “#4 ~ Fete A |a

159 fr 3 Ae ; ore ee Vin.a1 & Ul fey ese eee ee = en ieeg Solo

3

ta OE ey | | CUmB$§lCtB$§$lTCLCeS =

> -L! FFI r

” SOLO Orch. [Hf#es—-uHHM#]>=FTOTFROETEReeeoOEHNuHVA_OSFOF/H—HO|T

. z= >."

|==

161 ’ _ Vin. 1 & U [iffy | He EO f

3

& 3 3 tr 3 ~~

Orch. LL

The Type 5 Sonata: Solo 2 through Ritornello 4 585

and in that of Horn Concerto No. 2 in E-flat, within a bar or two of the onset of R3, beginK. 417 (m. 117, orchestra; m. 126, soloist). ning to participate at once in the now-joint and fully shared project of recapitulation. Examples

. may be found in Piano Concertos No. 6 in

KRS=S9 Merger Openings B-flat, K. 238 (mm. 131, 133), No. 15 in B-flat,

In this procedure the solo recapitulation begins K. 450 (mm. 197, 199, initiating an opening with a decisive tutti pillar replicating the begin- recapitulatory dialogue), and No. 16 in D, K. ning of opening ritornello theme, R1:\P, but 451 (m. 219, 220, with nonthematic, affirmative usually within a few bars the soloist re-enters scalar flourishes in the solo).

to assist with or to take over its continuation. Once the decision has been made to merge This strategy avoids the redundancy of the dou- the soloist into a recapitulation begun with an ble-start recapitulation, with which it 1s con- R3 gesture, this may be accomplished in any ceptually related.28 Here, instead of a backup number of ways. It is unproductive to try to and restart, an initial R3 texture soon merges categorize them. Some solo entrances are noninto an S3 one. This is the most common option thematic, providing a decorative overlay to the for the middle and later Mozart concertos, from thematic material still continuing in the orPiano Concerto No. 6 in B-flat, K. 238 (1776), chestra (a sustained trill, with upbeat, in Piano onward. In its many variants and realizations, it Concerto No. 18 in B-flat, K. 456, m. 240; scasoon became the first-level-default option, es- lar passagework in Piano Concerto No. 22 in

pecially in the piano concertos. E-flat, K. 482, m. 272). Others participate in a As already indicated, this R3=S3 merger decorative or agitated doubling of the melody has been noted and discussed in the literature, (Piano Concerto No. 20 in D Minor, K. 466, with much debate about whether the R3 open- m. 262) or a brief back-and-forth dialogue with ing portion should qualify as a real ritornello or the tutti (K. 450, mentioned above). Still othas merely a tutti interjection. We regard it as a ers enter by taking over the melodic lead of the ritornello pillar, R3, for reasons already laid out P-idea, thereby completing the theme begun by in chapter 19 (the structural reprise or da capo the orchestra. The soloist provides the complefunction of this kind of ritornello), reasons that mentary consequent to an orchestral antecedent, are not dependent on the number of bars for for instance, in Piano Concertos No. 12 in A, which the soloist is absent from the onset of the K. 414 (m. 204), and No. 23 in A, K. 488 (m. recapitulation—usually the central topic around 206). In No. 25 in C, K. 503, m. 298, the sowhich this debate has focused. Obviously, the loist provides the complementary reiteration of longer the tutti, the more stable the R3 impres- a broad sentence-presentation module. In other sion and the sense of unfolding a reanimation of cases the piano enters into the thematic current

the opening of R1. mid-phrase, at more unpredictable moments It is unusual for the recapitulatory tutti (R3) (Concerto for Two Pianos in E-flat, K. 365, to be sustained without the soloist for more than m. 209, reiterating the opening unisono floureight bars.2? The longest R3s, frequently cited ish in the orchestra with a modal shift; Piano in this regard, are those found in Piano Concer- Concerto No. 17 in G, K. 453, m. 237, midtos No. 21 in C, K. 467 (a non-normative twen- way through the sentence-continuation; No. ty-three bars, mm. 274—96, extending well into 21 in C, K. 467, m. 297, suddenly picking up R1IA\TR [m. 285] and the onset of a surprising, on the freshly interpolated wild-card, idée-fixe new idée fixe interpolation on IV [m. 295]), No. march that had just intervened in the orches27 in B-flat, K. 595 (fifteen bars, mm. 242-56), tra, m. 295; No. 24 in C minor, m. 368, repliand No. 17 in G, K. 453 (ten bars with extended cating one of its early entries in the exposition

upbeat, mm. 227-36). At the other extreme, [= m. 124]). the shortest are those in which the soloist enters 28. See also the similar conclusion in Grayson, Mozart’s 29. Irving, Mozart’s Piano Concertos, p. 50, regarded this

Piano Concertos Nos. 20 and 21, p. 27. as more or less the norm.

586 Elements of Sonata Theory

The Solo Recapitulatory Transition into the tonic, down a fifth from its appearance in the exposition.)

In purely harmonic terms, recapitulatory transi- As Mozart’s career developed, he explored tions in Type 5 sonatas follow the same norms as ways of treating the recapitulatory TR more those in other types. In order to prepare for the imaginatively. One approach was to subject recurrence of the S-theme(s) now in the tonic, the traditional S1:\[R-modeled recapitulatory these TRs conclude with either a I:HC ora transition to a significant shortening (typically I:PAC MC (the expositional option, V:HC, following an expansion of P and the suppresis no longer available), and they often effect a sion of the first tutti interjection) through the move toward the subdominant that might occur surprising decision to omit its most memorable either at a sudden shift-point or through a more module, the head-motive sujet-libre TR'!. In eradual modulation. Because this topic has been these cases, S1:\T'R!! became a one-time event,

dealt with earlier (chapter 11), the only mat- not heard again after the exposition. We find ters that need to be touched upon here concern this solution in the Oboe Concerto in C, K. 314 whatever issues might be unique to Type 5s. (expositional sujet libre at m. 50; cf. 1ts omission Those issues center around modular content— ca.m. 133 [= m. 61]),°! and in the Concerto for particularly the degree to which this TR might Two Pianos in E-flat, K. 365 (sujet libre at m. 84; reintroduce modules from R1 as part of a pro- cf. its omission c. m. 225, a crux with S1:\TR!

cess of recapitulatory synthesis. [= m. 96]), and Piano Concertos No. 16 in D, Young Mozart’s norm was to base the reca- K. 451 (sujet libre at m. 98; cf. c. m. 235), and pitulatory TR on S1:\TR materials, without No. 26 in D, K. 537 (sujet libre at m. 103; cf. any “new” infiltration from R1:\TR.°° In these the radical P-T'R telescoping perceptible at the concertos S1:\TR was often a solo-led sujet-libre crux, mm. 305-6 [= mm. 121-22]). transition that, following a brief tutti interjec- Another set of solutions entailed the replaction, served as a substitute for R1:\TR!!. That ing of certain S1:\TR modules with ones from sujet libre also led off the recapitulatory TR, at R1:\?R—basing the recapitulatory TR more times starting in the original key (Piano Con- on the one in the initial ritornello than on the certo No. 5 in D, K. 175, m. 161), at times replacement for it found in the exposition. Any shifted at once onto the subdominant (Bassoon such resubstitution of R1:\TR modules deviConcerto in B-flat, K. 191, m. 112; Violin Con- ated markedly from the more traditional praccerto No. 1 in B-flat, K. 207, m. 115). When tice of composing an S1:\TR-modeled recathese S1:\TR-modeled recapitulatory transi- pitulatory transition. These strategies emerged tions began in the tonic, they usually led to a especially in Mozart’s Viennese piano concertos passage of mild precrux alterations almost at from 1784 to 1785. Within the concept a numonce (as in K. 175/1), a practice much later re- ber of solutions were possible. One of the earlistored in Piano Concerto No. 27 in B-flat, K. est, a “both-and” or “double-start” procedure, 595 (linkage-technique sujet-libre S1:\TR!# at appears in No. 15 in B-flat, K. 450. In the exm. 256 [= m. 95], precrux alterations at m. 260; position we had heard, at the end of P, a brief, full crux with TI?, m. 266 [= m. 104], produc- modulatory TI! (mm. 86-87), elided with a suing an MC at the same pitch-level as that of the —_jet-libre S1:\ TR. beginning in G minor (m. 87). exposition, I: HC, at m. 268). On rare occasions In the recapitulation, the original TI! 1s absent,

no alterations were made to the recapitulatory and the music plunges instead into a tonic-key TR, which was therefore identical, bar-for-bar, recovery of R1:\TR!! in m. 210 (= m. 14)—a with the expositional TR, leading to the same module that had also been used to begin R2 in I: HC MC, as in Violin Concerto No. 5 in A, V (m. 137). After four bars the continuation of K. 219. (The next bar, S$1:\T'M!, is then shifted R1iA\TR!! destabilizes and is broken off (mm. 30. Cf. the discussion of the recapitulatory TR in tory TR also includes five newly interpolated bars, mm.

Ktister, Formale Aspekte, pp. 154-59. 137-41 (cf. the figuration in R1:\P!2, mm. 44-45). 31. As if by way of compensation, K. 314’s recapitula-

The Type 5 Sonata: Solo 2 through Ritornello 4 587

214-15) with a modulation toward C minor the solo exposition with a full restoration of the (11), whereupon the exposition’s sujet libre recurs, iron-willed, negative R1:\TR (m. 269), only with some variants, on C minor in m. 216 (cf. the MC conclusion of which had been replim. 87). The recovery of the original R1:\TR!! cated in the expositional transition.

module at m. 210—a move toward Rotation In the expositions of other Viennese piano 1-Rotation 2 synthesis—serves as a longer re- concertos (such as No. 17 in G, K. 453, and No. placement for the exposition’s modulatory TI. 18 in B-flat, K. 456) Mozart pursued the differFrom K. 450 onward Mozart’s solutions be- ing, more “symphonic” strategy of beginning the come more inventive. He now began to explore S1 transition with R1:\TR!! before dissolving the idea of a recapitulatory TR that leaned more into a differing solo continuation, whose modtoward that of the initial ritornello. Each so- ules we normally label as S1:\TR!2, S1:\TR!, lution is idiosyncratic and requires exploration and so on. In these cases he usually retained the on its own terms. In one of the most compli- R1A\TR!-! opening for the recapitulatory transicated of these, No. 21 in C, K. 467, he replaced tion but replaced the soloist’s original continuthe original sujet libre (m. 91) with a return to ation with a restoration of that of R1I:\TR. K. the forte R1:\TR!! (m. 285 = m. 12), continu- 453’s RAI:\TR restoration 1s complete and asing for some ten bars. In this case even stranger signed to the tutti throughout (mm. 242-57 = things follow: an unexpected interpolation of | mm. 16-31, though with the solo piano supplythe wild-card idée-fixe march (in its R1:\S!-? ing the caesura-fill in mm. 257-60). Thus, as version) in the subdominant, m. 295 (= m. 36) Irving noted, in K. 453 “virtually all of the first and a merger into S1:\TM? (!) at m. 304 (cf. solo transition is discarded.”>* In the similar K. m. 121), setting up a I: HC MC with expanded 456 the restoration proceeds as far as R1:\TR!2 fill (mm. 307-12) and leading to a completely (mm. 255-58 = mm. 24-27), after which two reordered TMB beginning with S1:\T'M3> (im. bars of solo-fill (mm. 258-59 = mm. 101-2) 313).52 Somewhat related is the situation in lead directly to the recapitulation of S1:\TM! No. 22 in E-flat, K. 482. Here what had been (m. 260 = m. 103). (The remainder of R1:\VTR the “new” breakout-continuation from $1 (m. —its much-expanded, minor-mode-ridden cae106, a passage probably also with the sujet-libre sura-fill—is cleverly tucked, as in the exposiprocedure in mind) is replaced by the return of tion, into S$1:\T'M?: m. 274 = m. 117 = m. 28.) R1IATR!! (mm. 31, 294),35 following a broad In all such cases from the later works, the P-space that had resubstituted the lengthy R1:\P crucial analytical point is to perceive the varymusic for its Solo 1 version. No. 20 in D Mi- ing degrees to which the more traditional pracnor, K. 466, avoids the briefer P=>TR merger in tice—retaining the modular materials of the S1

32. The concepts of wild cards and idées fixes, along with S-space will continue to define itself and its surround-

that of their functions—a crucial aspect of this move- ings as S-material regardless of where it appears. Acment, the touchstone example of them—are elaborated cordingly this principle holds only for S-material that in ch. 20. For another view of the unique interpolation appears after the MC, that is, within the tonally active, of the march figure in IV, see Grayson, Mozart’s Piano structural-cadence-attaining portions of the recapitulaConcertos Nos. 20 and 21, p. 54. Also to be noted is the tion. unusual feature of moving what had been in S-space 33. Our recapitulatory measure-numbering of K. 482/1 earlier into recapitulatory TR-space—shifting it left- follows that of the current critical edition, which reward, to a position before the MC. A similar occurrence stores two measures following m. 281 that are lacking in may be found in Piano Concerto No. 12 in A, K. 414, some other editions. See K. 482/1 1n Neue Ausgabe samin which what may be construed as an S$1:\S° module tlicher Werke, Serie V, Werkgruppe 15, vol. 6, ed. Hans (from m. 98) is more clearly in TR-space in the reca- Engel and Horst Heussner (Kassel: Barenreiter, 1961, pitulation, m. 224. Mozart apparently considered such pp. xv—-xvi, 206. Thus what had appeared as m. 282 in a move to neutralize the S-aspects of these modules, in some editions is now m. 284, and two-bar numbering effect deactivating them as S-material. As such it rep- discrepancies persist for the rest of the movement. resents the exception to what we have regarded as the 34. Irving, Mozart’s Piano Concertos, p. 53. general principle that what was initially sounded within

588 Elements of Sonata Theory

transition—is overridden by the urge to reinsert with only small adjustments in register, melodic alternative R1 ideas into the musical discourse. line, or scoring. The obvious benefit of this soluSuch recapitulatory syntheses display creative tion was one much valued in the late-eighteenth interactions, stage a negotiated play of forces, century: the attaining of a gratifying, “classical” between the group and the individual, the dual symmetry between the most structurally 1mclaims of R1 and S1. Especially in the mature portant portions of the solo exposition and recaworks, once such interactions have begun in the pitulation. Its chief drawback, at least from the recapitulatory P- and TR-spaces, they are likely perspective of later compositional practice, was to continue throughout the remainder of the re- the largely mechanical aspect of the mere trans-

capitulation. position. While this procedure is not invariable in Mozart’s earliest concertos, it occurs often enough in them to suggest that any deviations

Solo Recapitulatory S- and C-Space from it, including those in several of the more

(S3:\ESC) complex later concertos, are to be perceived as

overridings of that norm for local purposes that Following the first MC, the solo recapitulation —_ invite individual investigation. Our first conproceeds into the zones of tonal resolution, re- cern, then—option 1—is that more generically visiting in the tonic ideas associated previously formulaic type of solo recapitulation that does with S- and (when it had existed) C-material. _ not reinstate R1-materials that had been “‘miss-

As in the solo exposition, this complementary ing” in the solo exposition. post-MC block will normally end with a display

episode finished off with an emphatic trill-ca- . .

dence plunging into the next ritornello pillar, Option I: No Restoration of RU-\Post-MC here R4. Solo recapitulatory S-space persists as Material Suppressed in the Larger Exposition long as it includes modules associated with either One may find the simplest solution, bar-for-bar RA or S1:\S-space. It ends only when a satis- ~— correspondences with the solo exposition, in

factory perfect authentic cadence is sounded— Violin Concerto No. 3 in G, K. 216, and in proceeding onward to differing material—and Flute Concerto in G, K. 313. More typically, S-modules are definitively abandoned for the the solo recapitulation will tweak a solo-exporemainder of the solo recapitulation. The ton- sitional moment or two, lengthening or shortic-key S3:\ESC-point is usually the recapitula- ening it slightly. Thus in the Concertone in C, tory equivalent of the nontonic S1:\EEC, but —_K. 190, Mozart interpolated a “new” two-bar in rearranged or otherwise altered recapitula- tutti interjection—yet another “false R4-cftions the two points may not be parallel. The fect” within C-space—into the ongoing corS3:\ESC may be articulated only with the fi- — respondence measures at mm. 223-24, while nal trill-cadence (in which case there will be in the Bassoon Concerto in B-flat, K. 191, the no $3:\C-space; the display episode will be part recapitulatory equivalent of the exposition’s of S-space) or it may occur earlier (in which mm. 69-70, the final drive to the trill-cadence, case the display episode will have been played is intensified and stretched to four bars, mm. in C-space, perhaps even after the reappearance 148—51.3° The procedure found in K. 191, an

of a separate C-theme). expansion only at the end, is found also in VioAs with other portions of solo-recapitula- lin Concerto No. 5 in A, K. 219, and in Piano tory space, the background default option—a Concerto No. 6 in B-flat, K. 238. generically offered template—was to revisit the The operative principle in both K. 191 and post-MC materials from S1, bar-for-bar (or very K. 238 is that of keeping the emphatically menearly so), now transposing them to the tonic lodic materials intact and expanding or altering 35. In K. 191 notice also the switching of the parts in recapitulation it is played by the soloist, upbeat to m. the sounding of R1:\S. In the exposition the theme had 139. been carried by the orchestra, upbeat to m. 60; in the

The Type 5 Sonata: Solo 2 through Ritornello 4 589

things only at the nonthematic display episode. 151-63 [cf. mm. 33—45]) into the solo recapituThis principle may be found in several of the lation as an S-appendix, slightly varied, at mm.

Viennese piano concertos, such as Nos. 11 in 315-35, now accomplishing the $3:\ESC and F, K. 413 (display-episode expansion at mim. leading to a new display episode.°’ 341-51; cf. mm. 160-63), No. 13 in C, K. 415 Four comparable situations involving the (one added measure in the display episode, m. displacement of R1:\S-materials from R2 into 281), No. 14 in E-flat, K. 449 (expanded and the end of the solo recapitulation crop up in the partially recomposed portion of the display epi- first movements of Violin Concerto No. 2 in sode, mm. 309-18; cf. mm. 162-67), and No. D, K. 211, and Piano Concertos No. 16 in D, 20 in D Minor, K. 466 (recomposed first por- K. 451, No. 18 in B-flat, K. 456, and No. 23 in tion of the display episode, mm. 318-30; further A, K. 488. In the solo recapitulation of K. 211 expansions in mm. 333-38 [cf. mm. 156-59]; we find two additional bars of R1:\S!-3, mm. and a recomposed and lengthened final portion, 109-10 (= mm. 13-14), which in the exposimm. 343-56 [cf. mm. 164—74]). It also occurs tion had been placed into R2 (mm. 56-57). In in Horn Concerto No. 3 in E-flat, K. 447, al- K. 451 the newly imported materials (heard in though here one also finds a substantially re- R2 at mm. 179-91) consist of a march variant composed S1:\S!-3 (m. 147 [cf. m. 59]) merging of R1:\S!3 (mm. 280-83 [cf. mm. 57—60]) and into a much-lengthened and reconceived display a much-expanded R1:\S!-4 (mm. 283-307, inepisode.°° Two instances 1n which the exposi- cluding a merger into recomposed display eptional display episode is altered by being made isode; [cf. mm. 60—68]). In K. 456 the final, shorter rather than longer occur in Piano Con- fanfare module of the S-theme of Ritornello 1, certos No. 17 in G, K. 453 (mm. 315-16 pro- R1:S!-4, mm. 61-67, is omitted in the solo exvide a compressed recomposition of mm. 164— position (in which R1:\S!-!-!-3 serves as an ex68), and No. 27 in B-flat, K. 595 (mm. 323-24 tended TM3, mm. 128-49) but does resurface telescope mm. 161—63; another bar 1s omitted in a shortened version in Ritornello 2 to effect

between mm. 332 and 333). the R2:\EEC, mm. 185-89. In the solo recaOccasionally, one finds more idiosyncratic pitulation, whose post-MC-space is otherwise changes within option-1 solo recapitulations. parallel to S1, an “extra” R1:\S!4 fanfare, mm. In the first movement of Mozart’s earliest origi- 331-35, is interpolated cheekily at the end of nal piano concerto, No. 5 in D, K. 175, the ex- the display episode, thereby converting what we position’s S-modules (R1:\S9, R1A\S!!, SUAS!2, had presumed to be C-space (as in the solo ex-

S1:\S!3, in mm. 66, 68, 72, and 76) appear in position) into S-space. R1:\S!-4 also reappears a different order in the recapitulation as part of | once again in R4? (mm. 355-61), producing the a whirligig reassembling of fast-paced materials R4°\ESC that overrides the S3:\ESC at m. 337 (R1IAS®, RANS!, SINS!ES, and S1:\S!-2 in mm. (S3’s concluding trill-cadence). 183, 185, 188, and 195). Additionally, in K. 175/1, The situation in K. 488 is more unusual—a one could make the case that the original S1:\C, masterstroke of imagination. As mentioned earm. 83, is folded into recapitulatory S-space at m. lier, this movement’s opening sections present

199, since it now functions as the continuation a poised, purposely unadventurous reliance on of a “new” sentence begun with the presenta- symmetry and balance: no new modular ideas tional S1:\S!-2 at m. 195. In the finale of Violin except the display episode are introduced in Concerto No. 1 in B-flat, K. 207, whose solo Si. After an sudden interruption in the middle exposition had provided a new S, bypassing the of R2 (m. 142: “Stop! I have something else one proposed in R1, Mozart shifted the R1:\S to say!”) comes the unexpected feature of this that had reappeared only in R2-space (mm. movement. Into that interruption-gap Mozart 36. Cf. the recomposed S1:\S!-2 in the Second Horn tory space. The difference is that the R1:\S-material

Concerto, K. 417, mm. 155-63. missing in the solo exposition had been sounded in the 37. This procedure is related to that in option 2, which larger exposition, within R2. typically restores R1:\S material into solo-recapitula-

590 Elements of Sonata Theory

introduced a radical change of topic, a wistful our understanding of it as what one does hear. new piano theme (m. 143, beat 3) blossoming Whatever the internal plenitude of such an S1 + up out of nowhere—like a cherished after- R2, in its post-MC differences from R1 it also thought-idea, a special gift, that had not been registered a lack, an absence, something that thought of within R1. This previously unheard would have to be restored in the larger recapittheme not only dominates the ensuing develop- ulatory rotation. Our option 2 addresses those ment but is interpolated into the solo recapitula- instances in which significant aspects of that restion’s C-space, mm. 261—75, where it 1s treated toration were placed into post-MC positions in

wondrously, as a precious gem. the solo recapitulation. (Restorations made after the solo recapitulation—that 1s, within R4,

Option 2: Restoration of R1:\Post-MC concluding the larger recapitulation—are dealt

Material Suppressed in the Larger Exposition with separately below.) . i,

We find anticipations of this practice in

By the early and mid-1780s Mozart’s under- two of Mozart’s earlier, Salzburg concertos, alstanding of the structural and expressive pos- though in each early case the restored module

sibilities within the Type 5 sonata had been is not one from the crucial R1:\S-space. The much deepened. In around a dozen concerto earlier occurs in Violin Concerto No. 41in D, K. first movements, nearly all of them strongly in- 218 (1774), where a previously missing R1:\C!2 dividualized piano concertos from the center module, mm. 177 (beat 3)—181 (= mm. 30-34,

of his Vienna period—especially from No. 15 originally the continuation of a sentence), is in B-flat, K. 450 (1784), onward—one finds a spliced into the end of TM?, part of a trimodmore complex handling of the solo recapitu- ular block that is otherwise fully parallel with lation. This centered around the recrafting of — that of the exposition. The second is that found S- or C-space in such a way as to restore one or in Piano Concerto No. 9 in E-flat, K. 271. Here more R1:\S-modules that had been omitted in the solo recapitulatory display episode, still in the larger exposition—in other words, modules S-space, includes an interpolated restoration, that had not been heard since R1. We regard introduced by the orchestra, of an “out-of-orsuch procedures as exercising a set of “option der” R1:\PR! (mm. 251-58), the module that 2” realizations. Most importantly, they fortified had previously been heard only as the breakthe role of the post-MC portion of the solo re- out-continuation (P=>TR merger, mm. 7-11) capitulation (as opposed to the R4-block of the of the R1:\P!-! binary loops at the opening of larger recapitulation) in the task of providing the piece (see example 5.7). In this instance the a recapitulatory synthesis of Rotation 1 (R1) tutti interpolation at m. 251 suggests the witty and Rotation 2 (S1 + R2). Such an opportu- interjection of a premature, “false R4,” even in nity arose in any Type 5 structure in which the absence of the characteristic trill-cadence in the larger exposition had failed to include an the solo part. R1:\post-MC module. Moreover, any exposi- Solo recapitulations that restore a module tional omissions of all of R1:\S or of at least its from R1:\S first appear with that of the Sinfonia head-motive, R1:\S!!, would have been espe- Concertante in E-flat, K. 364 (1779; example cially noticeable. This means that what one does 22.6).°8 Here the “lost” R1:\S!!, originally a not hear in the larger exposition is as central to binary-loop presentation module of a sentence 38. Cf. also the restoration of a “lost” R1:\S!! on the zart’s mind before being rejected as a viable option at tonic, but more accurately expressing the active domi- m. 199, with the sudden deflation to E-flat minor. The nant, near the end of the developmental space of the solo recapitulation of this movement also shows other Concerto for Two Pianos (No. 10), in E-flat, K. 365/1, “experimental” anomalies, including a second quasimm. 187-99 (cf. mm. 30—42). This moment functions redundant appearance of R1:\P!:! at the opening of primarily as an expanded caesura-fill—a prolonged mo- S3:\C-space, m. 253 (from m. 5, following R1:\P! at ment of static suspension. It appears also to suggest a m. 1)—perhaps compensating for its earlier minor-mode fleeting dialogue with the Type 2 variant of the Type 5 collapse at m. 209. sonata, as if that idea were momentarily crossing Mo-



EXAMPLE 22.6 Mozart, Sinfonia Concertante in E-flat, K. 364, 1,

mm. 285-302

. . . pt

[Allegro maestoso] 285

|| f Ly 4) y+ |—=

Solo Violin and Viola [H4¢y5?—b/¢tH_fH._# t ps bg 9 ps =

looa'auaiwu, Orchestra telede ee SSS 2S =F —

ns rs re re

287 yg pu i? . a= rire et Se Se

Vin. & Via. [H¢s3?-b——hshsFH —/ ee er?shK3KK hh 7 FS

Op - . matt SG, ro

| §£©4S£ ee,eesSehFsSeese eee nl ee =| ~~ Orch. |-E4*54.— OO > °‘oa* A? Eas, = RA_ovNv000Wy

NN y, || v t Nn

289 2. a a a. . habe: JJ 7 i” A > ee er a Ae a 7 An Ce —_ . . on ~ ~

Vin. & Vila. [AS eee eeeEeEeEFheeeeeeee

t a a ee a | ee

a” A Pe —_ es es-=_ CS GD > A8 (OS Orch. | Hea? 9 GO,” 29 8GO82 _ OO ot pS i/ i 2 2 ClUflrlrllUDlhlCUlCOLL™™ChlULlUlCUmDOti tO 0 ae tf de Aa A AADAAAAADAAAAAAAAAAAAAAAAADAAAAAAAAAAAAAAAAAAAAAAAAAAAAAL

; ! i pK ge. Hse TJ

A ( aa ee a

| ed[f4es?—b poy Am CC“‘#$NSCOC‘é ee) = “ne gg ~~“— e Orch. er?i_—§_ i $y ee ————

STEP Str Sher ats

297 ) s «cy EP” AeePR a (

Vin. & Via. [Kes )0SNINMY_—l_rd].]_ os TT To OP y.ti‘OCOCSCéC;d ,¥—-# 4 ye éséfTé See SEED IEEE SEER “EEE SEE GEEK SEE

pre Peref ere? eF e id our SS Ne NLL

299 FS ee a” 3 a Vin. & Via. | Haye?-p Ss ri I F add daaAAADAADAAAAAADDAAADDAAAAADAAAADAAAADAAAAADADAAAAAD DD

o~. —~ —_ fp . Ipdt .. } J | pga. lise

|g Poy te mC (CC C“‘#$UUCOCiéié CG y. Ss rt rc eC OFC ( ‘ ‘ COOCOUO'TOUCOUC —_#—_$ , | aw “ SF @ ¢ |} 3g —_§ FT

(mm. 38—46), reappears as TM? in a newly fash- stated the “forgotten” R1:\S? as an S-appendix

ioned trimodular block. In the solo exposition (upbeat tom. 253 [cf. upbeat to m. 51, where S1:\S, a replacement for the secondary theme it had also served as an R1:\S-appendix]) and proposed in R1, had been brief, a mere eight-bar within a few bars merged it into the display epiperiod (mm. 126-33). In the solo recapitulation sode (ca. mm. 261—64). The S-appendix status

S1:\S returns, now in the tonic (m. 285) and of the restored R1:\S is even clearer in No. 15 with the order of its solo parts reversed, but its in B-flat, K. 450 (1784). Here R1:\S, a balanced earlier, concluding PAC (m. 133, the S1:\EEC) sentential period, is suppressed in the exposi-

is now reshaped into a half cadence and sec- tional Rotation 2 but is plugged back into the ond MC-effect (mm. 291-92, a brief S3:\VTM? sonata in the solo recapitulation (mm. 249-64 merely articulating the caesura). At this point [= mm. 26—41]), nonelided after a full restateMozart interpolated the expositionally absent ment of the “alternative” S1:\S, mm. 233-48, R1:S!! loops, mm. 293-301 (“Don’t forget ending with an emphatic I:PAC. me!”’), which at m. 301 rejoin measures that are From this point onward Mozart became even referential to the exposition’s display episode (cf. more fully committed to overriding the earlier m. 133), now functioning as the breakout-con- norm of keeping his solo recapitulations largely tinuation of the presentational loops. The re- parallel with the corresponding solo expositions. composition also defers the expected S3:\ESC. Anticipated especially by the recapitulatory While in the solo exposition the display episode procedures of K. 364 and K. 414, K. 450 marks had occupied C-space (m. 133), itis now drawn the point at which Mozart was determined to into a much-expanded, though largely nonthe- make the relationships among R1, the solo and matic S-space. The S3:\ESC 1s attained only larger expositions, and the solo and larger re-

with the concluding trill-cadence at m. 328, capitulations more complex. Often central to

elided with R4. this enhanced richness was the production of Three years later Mozart, now in Vienna, one or more thematic R1:\S-module absences

employed a similar solution in Piano Concerto within the solo and larger expositions—concepNo. 12 in A, K. 414. This time he expanded the tual blanks that demanded refilling later in the end the S-sentence found in the solo exposition movement—and then reinstating the sidelined (mm. 115-22, 123-31, R1:\S!!+S1:\S2-7)%? and material at unanticipated moments within the brought it to an unequivocal PAC close in the solo recapitulation, or perhaps even deferring tonic (mm. 233-52). Following this, he rein- the reappearances of that material until R4 (as

39. Many portions of K. 414/1 present problems of clas- a close with the V:PAC in m. 114. M. 115 then starts a

sification. (Cf. n. 32.) If in its exposition m. 98 is re- second S-sentence, one whose presentation recaptures garded as the start of S1:\S (perhaps S1:\S° on V? of R1:AS!"1. Its continuation at m. 122, however, is new to V), then that portion of the secondary theme comes to Solo 1: hence the $1:\S?-? label.

594 Elements of Sonata Theory

in K. 453, revisited at the end of this chapter). are preoccupied with the further multiplication By K. 450 this practice had become Mozart’s of the ever-proliferating, idée-fixe march motive norm rather than the exception. It was now a that dominates much of this movement. Shiftpersonally customized first-level default that ing our attention now to the music prior to the emphasized the solo and larger recapitulations as recapitulatory display episode, we notice also spaces of unpredictable synthesis, reassembling that the solo exposition’s TMB (whose conand fusing the disparate materials of Ritornello stituent parts had begun in mm. 96, 103, and 1 and the larger exposition. Whenever Mozart 131) is shorn of its TM! module—mm. 96-102, did return to the earlier, simpler norm of keep- which, as a consequence, never recurs anywhere ing the solo exposition and solo recapitulation else in the movement. This means that much of essentially parallel, as in Piano Concerto No. 27 the exposition’s idée-fixe-grounded TM?, for the in B-flat, K. 595, or the Clarinet Concerto in A, most part an expanded caesura-fill ending with K. 622, he produced the effect of a smooth, un- the usual PAC, now appears in recapitulatory corrupted balance of parts, a classical equipoise TR-space. As a result, the exposition’s S1:\TM> existing “transcendently” beyond any forces (mm. 131-49) is restated as the beginning of that might introduce unwelcome elements of — the solo recapitulation’s S-space (mm. 298-316).

disequilibrium. From the perspective of the movement’s over-

Even as they beguiled listeners with one all proportions, the recapitulation’s “shortened” astonishing idea after another, the first move- thematic-S-space 1s compensated by the R1 inments of Mozart’s concertos had now become terpolation and later expansion within the disstunningly complex. Notwithstanding all of its play episode.

inbuilt archaic rigidities, the Type 5 sonata— When interpolating a thematically marked with its added requirement of assessing all of the but expositionally absent R1:\S-module into sonata-spaces 1n relation not only to each other the solo recapitulation, Mozart’s nearly invaribut also to the proto-expositional model pro- able practice was to restore it at or very near vided in Ri—had been transformed into a field the end of the reconfigured S-block. The recaof abundant structural possibilities, the ne plus pitulatory pattern thus provided was typically ultra demonstration of ingenious sonata-organ1- that of first sounding whatever S1:\S- or TMB zational technique. Mozart’s synthesis-solutions modules that he chose to include (from time to were now so individualized that the only way to time, the ever-“fragile” S1:\TM! was dropped do justice to them is to examine each of them out entirely, which sometimes had the effect separately and in detail, a task beyond the scope of pushing whatever remained of the original

of this chapter. S1:\TM? into recapitulatory TR-space), then The solo recapitulation of Piano Concerto appending the R1-module as the concluding

No. 19 in F, K. 459, for instance, is so intricate thematic gesture of the S-block (even though in as virtually to defy a quick description. The re- some instances S-space continues with a largely stored R1-module at issue here—beginning as a nonthematic display episode.) The missing set of melodically descending thirds—recurs as a R1:\S-module returns at the end both as a clever “surprise” interpolation, mm. 341-47, wedged restoration of something that one might have into the display episode. But that passage is not assumed was long lost (“Remember this?”) and really an R1:\S-module, since K. 459 is the only as a large-scale wraparound gesture, in which piano concerto to contain not a two-part but a the R1 model rotation’s S-space 1s demonstrated

continuous proto-exposition in R1. Thus the as being ultimately decisive in the rotational restored module is extracted from the latter part proceedings: all of the thematically tagged of the central portion (that is, of TR=>FS) of — S1:\S-substitutions are to be construed as havR1, mm. 43-49. Shortly after this, the reca- ing emerged conceptually within R1:\S’s appitulatory display episode features a large ex- parently expandable interior. Such is the case pansion from the solo expositional model. An in Piano Concerto No. 21 in C, K. 467. In this original seven bars (mm. 181-87) are inflated ingenious solo recapitulation, S1:\VIT'M! is banto twenty-two (mm. 356-77), many of which ished altogether (cf. its expositional appearance

The Type 5 Sonata: Solo 2 through Ritornello 4 595

on G minor, m. 109), and the newly recrafted nor-mode movement. The “classical” principle TMB-space proceeds in the order, S1:\T'M? (m. of an ordered tidiness is constantly undermined 313, hence now functioning in the role of TM!), by the psychological disturbance represented by idée-fixe march merging into a slightly varied the movement’s opening idée fixe.

recurrence of the display episode, all of which Mozart’s general “option 2” principle that serves as a new and expanded TM? (m. 328, the expositionally missing R1:\S-module is to originally sounded at m. 143 as what is best con- be brought in as the last thematic S-module of strued as an S-appendix to $1:\T'M3), and—fi- the solo recapitulation was most clearly aban-

nally—the restoration of R1:\S, begun by the doned in Piano Concerto No. 22 in E-flat, K.

orchestra, at m. 351. 482.40 Here, uniquely, it is shifted to the first

No. 24 1n C Minor, K. 491, presents a simi- S-position. As in K. 467, Solo 1 had sidelined lar but even more dizzyingly complicated situ- R1:\S in order to present an all-new TMB, thus ation. The solo exposition’s S-block had omit- providing an extended maximal contrast with ted R1AS (first heard in mm. 44-63, preceded the secondary theme proposed in Ritornello 1. by orchestral caesura-fill, mm. 35-44) in favor Such a procedure yields an overabundance or of a disturbingly lengthy, seemingly unstop- excess of S-material to confront in the solo repable S-phrase-chain in III, E-flat major, with capitulation. Mozart’s solution was to rewrite multiple refrain PACs (mm. 147-265). (As dis- the transition in such a way as to restore much of cussed in chapter 21, these may be labeled as R1:\TR and CF, thus permitting him to omit S1:\S! through the largely nonthematic conclu- the first two modules of S1:\S-space: S1:\TM! sion of the display-episode S1:\S°.) In the solo (sounded in m. 128, an aggressive and surprisrecapitulation—apart from several other altera- ing gesture on B-flat minor; cf. the $1:\VT'M! in tions of local detail—S1:\S° and S1:\S4 do not K. 467) and the transitional S$1:\VT'M? (c. m. 139)

reappear at all, and S° 1s only briefly alluded to and second MC (m. 149). All of that was now at the end of the display episode (mm. 469-73 = replaced with a gratifying restoration of R1:\S

mm. 261-65). The opening two, “thematic” (m. 314), ending with a I:PAC elided to the S1:\S-modules are retained as central players— original S1:\T'M3? as an S-appendix (m. 330 [cf.

now wearily in the tonic, C minor—but they m. 152]). are sounded in reverse order (mm. 391, 410), as Finally, in two extraordinary first movethe first two parts of a strained TMB. They are ments from 1786 and 1788, those of No. 25 in followed—again, finally—by the C-minor rein- C, K. 503, and No. 26 in D, K. 537—his crestating of R1:\S (m. 444 [cf. m.44], preceded by ative imagination now running at full tilt—Moseveral bars of its own R1:\CF (m. 435 [cf. m. zart produced a solo-recapitulatory synthesis by 35]). K. 491’s many substitutions, reorderings, means not of a mere trimodular block (with the and asymmetries among Ritornello 1, the solo standard two MC-effects) but rather of a virtuexposition, and the solo recapitulation, were ally unprecedented five-module block containdoubtless intended to convey a nightmarish 1r- ing three MC-effects. The simpler of the two, rationality coursing through this obsessive, mi- K. 503, first brings back 1n the solo recapitula-

40. Depending on one’s interpretation of the piece, an- related to R1:\C (cf., e.g., mm. 20-21), and the soloother exception might be found in the solo recapitula- ist initiates a largely new display episode, mm. 131-37, tion of a smaller-scale work, Horn Concerto No. 1 in which might, however, recall aspects of the solo-expoD, K. 412 (1791). Most unusually, recapitulatory S-space sitional continuation, S$1:\S!-2 and S1:\S!3 (mm. 42-46,

begins with a brief, interpolated module, mm. 114-17, 47-51). If that is the preferred reading of this passage, that had been heard nowhere before in the movement— the missing R1 module would have been interpolated a scale-and-arpeggio signal-figure idiomatic for the solo into the middle of material taken or adapted from S1:\Shorn. This elides to $1:\S!!, the presentation of a sen- space. On the other hand, if the display episode is re-

tence, in mm. 117-20 (= mm. 38-41). At this point garded as essentially figurational, nonthematic, then Mozart restored R1:\S and its repetition, mm. 121-24, R1:\S would have been reinstated as the last self-evi125-28, now serving as the sentence’s continuation. A dently thematic moment of recapitulatory S-space. two-bar tutti interjection follows, mm. 129-30, perhaps

596 Elements of Sonata Theory

tion the all-new S1:\T'MB from the solo expo- position—to restore any R1 modules (especially sition. (In other words, in this concerto, unlike R1:\post-MC modules) that had not been heard the case in K. 467, 482, and a few others, the since the opening ritornello. Many of the first original S1:\I'M! is not suppressed. Provoca- movements of Mozart’s concertos feature one tively, though, it still begins in the “wrong key,” or more such reinstatements at the end. This E-flat major, m. 326, as it had done in m. 148. means that some of the most memorable, theThe nontonic aspect of this module is doubtless matic modules of these movements, quite often relevant to the decision to produce a five-mod- their tag-conclusions, occur only in R1 and R4, ule block.) Instead of closing S1:\T'M3? (m. 345 and only in the tonic. [cf. m. 170]) with a PAC, the composer altered the ending of its repetition to set up a third MC RA! and R42 Subdivisions at m. 364. Thus invited, the major-mode version of the march-like R1:\S finally strides in In almost all cases, Ritornello 4 is subdivided at the end, cheerily and dutifully bringing this into two parts: an orchestral R4!, pressing efthematic parade to a congenial close (m. 365). ficiently toward the formulaic, grand ? chord,

The situation in K. 537 1s complicated by the pinned into stasis with a fermata, that opens existence of an amusing, three-element S-chain the path to a solo cadenza; and an orchestral in Ritornello 1: R1:\S! (m. 38), S? (m. 50), and R42, elided with the soloist’s last trill-cadence S? (m. 59). Of these, only R1:\S! appears in the and completing whatever leftover modular-roexposition, serving as the head-motive of TM> tational business remains to be addressed. (In

(m. 164). The same TMB returns in the solo the piano concertos K. 271 and 491 the solorecapitulation, but, as in K. 503, its TM? veers ist, exceptionally, returns in R4?. In K. 595 it away from closure to produce a third MC (im. intervenes in what initially seems to announce 383), permitting the restoration of the insouci- itself as R4', the passage beginning at m. 335,

ant R1:\S3 (upbeat to m. 384). elided with the normative trill-cadence.) Only three of Mozart’s concerto first movements lack

a cadenza within R4. Consequently, these in-

Ritornello 4 and the Conclusion stances do not subdivide into the normative R4!

of the Larger Recapitulation and R42. All come from Vienna-period wind concertos, and two of them emerged 1n the last

Even as the final trill-cadence of S3 closes the year of Mozart’s life: Horn Concerto No. 2 in solo recapitulation, the larger recapitulation —_E-flat, K. 417 (1783), Horn Concerto “No. 1” continues with the onset of the fourth and last in D (1791), K. 412, and the Clarinet Concerto ritornello pillar, R4. As with R2, normally the in A, K. 622 (1791). Similarly, Beethoven’s Pimodel against which the R4 events should be ano Concerto No. 5 in E-flat, op. 73, may be interpreted, Ritornello 4 may begin either with said to lack a cadenza proper in its R4, although rotationally inert or rotationally participatory the formulaically held ? platform (m. 496) is still modules. However it begins, R4 always com- present, splitting the section into R4! and R4?, pletes the recapitulatory rotation by bringing and it is followed by an intensifying passage back the concluding modules of RlI—modules _ for solo piano, one that the composer famously that were not included in the solo recapitulation marked, however, “non si fa una Cadenza, ma and that may or may not have been attainedin _s’attacca subito il seguente.” This cadenza-like R2, the (often incomplete) end of the exposi- “non-cadenza” almost immediately softens tional rotation. A primary task of R4, beyond into a revisiting of the minor-mode beginning that of framing its central event, the non-no- of R1:\S!! (m. 508; [cf. m. 41]), and it 1s soon tated and implicitly improvised cadenza,‘! is to joined by the orchestra sounding the theme’s pick up any remaining loose ends of the com- major-mode “correction” (m. 516) and flowing, 41. We also possess, of course, numerous separately no- ers. See below on “the cadenza” as treated generally in

tated cadenzas written by Mozart, Beethoven, and oth- Sonata Theory.

The Type 5 Sonata: Solo 2 through Ritornello 4 597

with the soloist, into a vast review of prior so- anacrusis (mm. 487—88), a bar of fill, and a re-

nata material. storing of an extended passage from R1:\C, unIn Mozart’s works, once the final rotational heard since the initial ritornello (from R1:\C!?, module is attained in R4*—with the sounding mim. 490—97 [cf. mm. 80-87]; R1:\C!3, mm.

of the last module laid out in Ri—the move- 498-501 [= mm. 88-91]; and all of R1:\C?, ment usually comes to a close. The end of R4? mm. 501-09 [= mm. 91-99, the end of R1]). typically rhymes with the end of R1 (oninfre- — At the point of the final structural cadence endquent occasions with an additional bar ortwoof ing the recapitulatory rotation, m. 509, Mozart emphasis), unless, for whatever reason, R1 had appended a fifteen-bar passage that he marked been left unclosed by merging into the solo ex- explicitly as the “Coda.” Here, perhaps recallposition, as in Violin Concerto No. 3 in G, K. ing the unusual solution in K. 271, Mozart un216, and the Sinfonia Concertante in E-flat, K. expectedly brings back the soloist, mm. 509-23, 364. Along with the Flute and Harp Concerto who participates in a spectral, sotto-voce fade-out

in C, K. 299 (the addition of mm. 262-65, with rippling, legato arpeggios coursing through R1:\P-based), however, eight of Mozart’s piano a final visiting of R1:\P, idée-fixe shivers. This concertos append an extension or coda (or CRI final passage 1s likely to have been the model equivalent) after the end of the recapitulatory — for the parallel coda with soloist, in the first rotation: No. 9 in E-flat, K. 271 (mm. 302-07, movement of Beethoven’s Piano Concerto No. extension of a restored R1:\C3 figure, includ- 3 in C Minor, op. 37, mm. 417—43.4 In this ing the non-normative return of the soloist, as case Beethoven’s piano-enhanced coda subalso in R4!),42 No. 15 in B-flat, K. 450 (mm. stitutes entirely for the more traditional R4?. 305-08, another extension to a final C-module, Non-normatively, Beethoven choked back the though here more normatively, without soloist), cadenza’s final trill-cadence to a piano/pianisNo. 17 in G, K. 453 (mm. 344-49, coda-return simo dynamic at the start of the “R4?-position” of an earlier, jocular MC-fill idea), No. 20 in coda and undermined its resolution through the D Minor, K. 466 (390-97, ominously quiet, substitution of a V’/iv chord (m. 417). This 1s minor-mode R1:\P-figuration), No. 21 in C, the start of a broad, fatalistic crescendo to the K. 467 (mm. 414-17, piano-dynamic echoes of | end of the movement, at first pushed forward the idée fixe, probably heard here as rebounding ominously by the idée-fixe march-motive in the

echoes of the preceding, restored R1:\C), No. timpani (mm. 417-18, 419-20, and so on) and 22 in E-flat, K. 482 (mm. 380-83, forte rein- reacted to with “K. 491-style arpeggio-shivers” forcement of the preceding piano cadence), No. in the piano. As might be expected, Beethoven 23 in A, K. 488 (mm. 310-14, again a forte rein- broadened the concept of retaining the soloist forcement of an otherwise piano close), and No. throughout all of the postcadenza space in his

24 in C Minor, K. 491. later concertos.

The last of these, K. 491, is the most extraordinary. No notated cadenza by Mozart R2 Material within Parts of R4:

. . . rer Three Options

survives for this movement, and the autograph

is unique in not showing any indication of a final trill-cadence emerging out of an impro- In Mozart’s concertos, the R4! and R4 subdivised one. However it 1s to have been prepared, visions, each with its own generic task, virtuR4 starts with two sweeping bars of C-minor — ally mandate that R4 will normally be longer 42. In K. 271 the repeated reappearances of the soloist in volve the abrupt, recurring R1:\P!-! loops. Cf. the disR4!, R42, and the coda—all tutti spaces that are almost cussion of K. 491 below. invariably occupied by the orchestra alone—provide a 43. See the similar conclusion in Plantinga, Beethoven’s good example of an R4 textural deformation. In this Concertos, p. 158, 165-66. Extended, solo-reinforced instance these recurrences play into the established tex- R4?s and/or codas would also recur prominently in

tural “game” that pervades that movement, in which Beethoven’s Violin Concerto and in his Fourth and the soloist and orchestra exchange expected positions Fifth Piano Concertos. with amusing regularity, especially in moments that in-

598 Elements of Sonata Theory

and incorporate more modular material than 219; in the Flute and Harp Concerto, K. 299; in did R2. The modular content of these subdi- the Oboe Concerto, K. 314, in Horn Concerto visions is flexible: indeed, Robert Forster has No. 4, K. 495; and in the piano concertos K. recently devoted the larger part of a full mono- 175, 238, 242, 246, 365, 413, 414, and 595. In eraph to cataloguing Mozart’s many realizations a slight variant, the first module of R2 may be of R4', the cadenza, and R4?.44 In almost all suppressed in order to start R4? with its second, cases, significant parts of R4 revisit and tonally asin K. 449 (m. 330, with R1:\S%, music from resolve the music of R2, a factor of structural m. 70) and K. 450 (m. 295, with R1:\C!-2, from importance with regard to the rotationally par- m. 45). Additionally, three other piano concerticipatory modules. R2 1s normally the most tos begin R4? with a differing module but proimmediate model—an underlying basis—to ceed at the end to R2 material, which thereby be expanded upon in the R4'-R4? complex. seems embedded within it: K. 415, 453, and Movements in which neither R4! nor R4? cor- 491. Of all of these R4?’s, ten provide the familrespond to R2 are rare. Even when both largely iar additional concluding modules, beyond the differ from R2, one or the other (usually R4?) R2-model proper, a restoration of “lost”? modwill include a brief, anchoring module near the ules or measures from the end of R1 that allow end to solidify the R2 connection. This can oc- them to complete the rotation in a manner parcur 1n pieces in which R2’s rotationally partici- allel to the model Rotation 1: K. 218, 219, 246, patory modules are minimized or prematurely 299, 365, 413, 449, 453, 495, and 595.

cut short. One such instance found in Piano Within this most frequently deployed, first Concerto No. 17 in G, K. 453, in which only option, it is also possible for the ad hoc, preRitornello 2’s interrupted attempt to sound an cadenza R4! to reinstate one or more expoR1:\C? module, mm. 178-81, returns—and is sitionally suppressed R1 modules on its own. concluded—in a portion of R47, mm. 340-43. When it does, it most often restores an otherThe remainder of R4! and R4?is taken up with wise abandoned R1:\TR module (not necesthe tonic restoration of modules unheard since sarily R1I:\VTR!!), which is always rotationally R1. A similar situation occurs in No. 13 in C, inert in this position. A classic instance occurs K. 415, while in the Concertone in C, K. 190, in the brief R4! of Violin Concerto No. 5, K. it had been only R4! that touches on material 219, m. 216-19, which brings back the “lost”

from R2. pre-MC module from R1 (mm. 16-17); its R4?, Just where the now-tonicized R2 material like its R2, begins with R1:\C (m. 220). Other will be relocated varies from concerto to con- R1:\TR-material restorations in R4! are found certo. Generally considered, there were three in the piano concertos K. 175, 365, 413, 415, available options. Arranged in order of fre- and 449 (m. 320, with its sudden jolt to vi, as quency, these are: R2 material returns primar- in R1i:\TR!!, m. 16). When such an R4! beily in R42; R2 material is divided between R4! gins with now-restored R1:\S-material, it is and R4?; or R2 material reappears only in R4. both rotationally participatory and reopens the

We shall take up each of these in turn. solo recapitulation’s S3:\ESC. This happens in two piano concertos, K. 414 (m. 283, R1:\S!-2

Option 1. Most commonly, Mozart outfitted [cf. m. 41]) and K. 453 (m. 319, R1:\S!? [cfR4! with differing material—another module m. 49], ushered in with an unanticipated dethat is usually, though not always, rotationally ceptive cadence undermining the soloist’s final inert—and brought back the R2-referential trill-cadence). Another possibility is to reinstate music, nearly always fully intact, after the ca- “missing,” rotationally participatory R1:\C denza, in R47. R4? begins in the manner of R2 modules at the opening of R4?, as happens in four violin concertos, K. 207, 216, 218, and in the piano concertos K. 246, 450, and 491. 44. Forster, Die Kopfsatze der Klavierkonzerte Mozarts und Beethovens: Gesamtaufbau, Solokadenz und Schlupbildung

(Munich: Fink, 1992).

The Type 5 Sonata: Solo 2 through Ritornello 4 599

Still another was to interpolate essentially new, sequent cadenza, and all or most of the remain“stock” material into R4'—merely serviceable der of R2 will resume in R4? after the cadenza,

tutti-activity that gets one efficiently to the interpolating or adding additional modules as fermata-pinned $ platform, as in Violin Con- desired, especially when needed to complete the certos Nos. 1 and 3, K. 207 and 211 (in which rotation left unfinished in R2. Mozart’s earliest the new material soon merges with R1:\TR!!). approach to this double-span procedure is found While this “free” alternative seems more suit- in the Bassoon Concerto, K. 191, although that able to the early, more purely “generic” concer- first movement is atypical in that it also presents tos, it does reappear in Horn Concerto No. 4, a reshuffling of the relevant modules in order K. 495. In three instances Mozart filled R4! not to keep the opening of R4 parallel with that of with material that was lacking in the solo expo- R2.46 More normative examples, with the R2 sition, but with R1:\P!! material that had not opening-module being placed into R4! and the reappeared in R2. This produced the effect of | subsequent modules into R42, occur in Violin a spliced-in “false start” to a new rotation that Concerto No. 2, K. 211; in the Flute Concerto, never materializes: in the Concerto for Three K. 313; in the Third Horn Concerto, K. 447; Pianos (No. 7), K. 242 (m. 245); in the Flute and in two piano concertos, K. 271 and 482. and Harp Concerto, K. 299 (m. 243); and in In Piano Concerto No. 26, K. 537, the procethe Oboe Concerto, K. 314 (m. 174, an R1:\P dure is similar, but R4! omits the first module of

variant). R2 in order to begin with the second (m. 409, R:ATR!? [cf. m. 21], while R4? completes the Option 2. If Mozart most frequently introduced succession (m. 416, R1:\C!, C2). And in three the earlier R2 material into R4? space, as de- piano concertos, K. 456, 466, and 467, R4! is scribed above, his next most common choice normative, beginning with the initial module

was to split the R2 music between R4! and of R2, while the remainder of R2, and any R4?. This strategy demonstrates the concept of subsequent modules required to produce the Ritornello 4 as a single formal entity, despite its full rotation, appear only after R4? has begun cadenza-separated subdivisions (which are now with differing material. In K. 456, R4? starts bound together 1n their sharing of R2 ideas). In by backing up to revisit a redundant R1:\S!-? this procedure R4! usually begins with the same and S!3 (mm. 349, 352, already heard in the module as R2, only now in the tonic. As out- solo recapitulation, mm. 293, 299); in K. 466 lined in chapter 21, this module may be either and 467, it begins with the restoration of one or rotationally inert (as with the commonly en- more modules unheard since R1—less expancountered, “out-of-order” R1:\TR! module) sive presentations of the R1 version of R1:\C! or rotationally participatory (for example, a later and of R1:\C2 in K. 466 (mm. 366 and 375 [cf. R1AS or C module). Sounding it again at the mim. 44 and 58]);#7 R1:\S!3 in K. 467 (m. 397 opening of R4! recaptures the same tutti affir- [cf. m. 44]). mation that had reinforced the soloist’s trill-ca-

dence at the end of S1. Normally, R2’s first Option 3. The remaining possibility was to module will now be pushed without delay to- include R2 material only (and often incomward R4"’s obligatory, held 9 platform and sub- pletely) in R4!, leaving R4? free to restore pre45. These cases are to be distinguished from those in ing the cadenza, we first hear R1:\TR (mm. 161-63), which R2 had begun with R1:\P!! or the R1:\P-motto, quickly spliced to two modules completing the rotation,

an effect replicated also in the R2-material-launched unheard since R1: the restorations of R1:\S!-? (mm. R4!, as in the Flute Concerto, K. 313, and Piano Con- 163—68, thus producing the R4:\ESC at m. 168) and certo No. 19, K. 459—both mentioned in this regard R1:\C? (mm. 168-70).

in ch. 21. 47. In K. 491 the S1 and S3 display episodes had also 46. In K. 191 the R2 succession is: R1:\TR (mm. 71— touched on recast versions of R1:\C!, as in mm. 153 73, rotationally inert) and R1:\C! (mm. 73-80, rota- and 330. In this sense the R1:\C! idea 1s not literally tionally participatory). R4! begins, exceptionally, with “restored” in R42. What is brought back is its characthe second of these, R1:\C! (mm. 152-60). Follow- teristically R1 version.

600 Elements of Sonata Theory

viously “lost” material from R1. This third op- now-celebrated cadenzas for many of his contion was appropriate in cases either where the certo movements—often more than one for an larger exposition had differed remarkably from individual movement—most likely, it seems, for the opening ritornello or where R2 had bro- the use or instruction of others.*® ken off before the final module(s) of the com- Three points regarding the late-cighteenthplete rotation. Thus the final touches of the century cadenza are of interest to our concerns larger-recapitulatory synthesis can come to oc- here. First, this space of improvisatory freedom cupy the whole of R4?, or very nearly so, and was a near-obligatory feature and climactic are not delayed to its final bars only, as in some event of the Type 5 sonata. Unlike the other instances of the other R4 strategies. While an types, this one contained a built-in generic moearly approach to this option may be found in ment, an expandable zone of performative freethe Concertone, K. 190 (although R4? here dom, in which the typical “group” constraints restores nothing, merely bringing back R1:\P of sonata practice were shown to be temporarily for six bars in the manner of a coda), Mozart lifted. (That the suspension of sonata activity was most interested in exploring this solution is allowed only temporarily, and as the soloist’s in four piano concertos from 1784 to 1786: K. “last word” in the movement, is doubtless her451, 459, 488, and 503. K. 488 differs from the meneutically significant.) Second, the contents other three. Its R4? begins not with “missing” of the cadenza, virtually by definition, were material but with the return of the fuller, less to vary from one performance to another, in decorative, R1 version of the otherwise redun- large part to exemplify that provisional freedant R1:\C! module (= R1, m. 56), also heard dom. And third, with the fermata-pause at the in a variant in the solo recapitulation, m. 254. end of R4'—along with its formulaic suspension The “lost” module restored is the one that fol- of harmonic motion on the cadential 3 chord, lows, R1:\C?, mm. 307-10, serving here, as in resolved only with the V7—I trill-cadence that R1, like a “final blessing” of the whole. finishes off the cadenza—the sonata clock stops, only to resume once the cadenza 1s finished.

The Cadenza In Mozart's concertos the cadenza is best erasped as an idiosyncratic, solo-performative From some perspectives the cadenza may be event operating outside of the structural proregarded as the central presentational event of — cesses of sonata form proper. It was a specialized

R4. Its appearance could hardly be more dra- bubble interpolated into the broader Type 5 matically staged: R4! exists largely to set it up; structure, a substructural parenthesis that simulR4? responds to it, as if all of the requisite so- taneously, and paradoxically, was temporarily loistic business has now been finished; and in to hold at bay the forward motion of the larger it, the orchestra recedes completely in order to formal demands. The silence of the now-stilled permit the soloist, the concerto’s central focus orchestra only highlights this aspect, as does the of attention, to stand forth on his or her own virtuosic flair with which the cadenza was to be terms, spotlighted and released into a relative improvised and delivered, widening the strucfreedom. It was fundamentally an improvisatory tural gap with each successive module that it event, even though Mozart did notate several presents. The fundamental structural processes 48. Cf. also n. 41. Overviews of many of the issues sur- Auszierungen zu Klavierkonzerten von Wolfgang Amadeus

rounding cadenzas may be found in Irving, Mozart’s Mozart (Kassel: Barenreiter, 1967); Philip Whitmore, Piano Concertos, pp. 152-62, Grayson, Mozart: Piano Unpremeditated Art: The Cadenza in the Classical Keyboard

Concertos Nos. 20 and 21, pp. 101-4, and Plantinga, Concerto (Oxford: Clarendon Press, 1991); Christoph Beethoven’s Concertos (where the written cadenzas for each Wolff, “Zur Chronologie der Klavierkonzert-Kadenzen of these works are discussed individually). More details Mozarts,” Mozart-Jahrbuch 1978-79, 235—46; and Wolff,

about Mozart’s cadenzas are available in Eva and Paul “Cadenzas and Styles of Improvisation in Mozart’s PiBadura-Skoda, Interpreting Mozart on the Keyboard, trans. ano Concertos,” in R. Larry Todd and Peter Williams,

Leo Black (New York and London: Barrie & Rock- eds., Perspectives on Mozart Performance (Cambridge: liffe, 1962); Paul Badura-Skoda, Kadenzen, Eingdange und Cambridge University Press, 1991), pp. 228-38.

The Type 5 Sonata: Solo 2 through Ritornello 4 601

of Type 5 sonata form were unaffected by the to R1:\S? (an S-appendix first heard in R1, upvariable contents of this separable cadenza-mo- beat to m. 51; this module had been “lost” in ment interpolated into one of its interstices, a the solo exposition but restored in the solo reca“musical moment” that is expandable in “real” pitulation, upbeat to m. 253). A handful of Moclock-time. And yet, because the production of zart’s other surviving cadenzas are more fully this solo-filled gap is one of the Type 5’s generic rotational in that at various points they embed obligations, one might wish to pursue some of citations of two or more modules, in order, with

its potential implications. at least one from each side of the MC-divide.>°? While it might seem strained to interpret the These include cadenzas written for K. 271 content of an improvisatory and variable “mo- (R1AA\TR!I [from m. 14] + R1:\S2 [from m. ment,” it is also clear that Mozart’s notated ca- 34]); K. 453 (variants of R1:\P + R1A:\S [from denzas, especially those written for the piano m. 35] + the last half of the S1:\T'M! antecedent concertos from K. 271 onward, typically include [from m. 114]); K. 456 (a linkage-technique, passages that recover, or reflect upon, some of nonrotational R1:\TR!? [originally from mm. the melodic material of the movement proper, 24-27] leads to R1:\P + R1IA\S!° [from m. 54] sometimes in a manner that can seem “tightly + R1:\S!-2 [from m. 47], these last two “out of structured.’*? That the notated cadenzas nor- order”); and K. 459 (the idée-fixe motto, based mally display an “opening-middle-closing” pat- on R1:\P, + St:\TM®? [from the variant at m. tern of organization, as noted by Eva and Paul 139]). Beethoven also seems to have preferred Badura-Skoda, far from being surprising, is pre- extended cadenzas with rotational implicacisely what we would expect.°° Nor is it sur- tions. prising that some of Mozart’s surviving piano Do such rotational flickerings within what cadenzas, like many of his sujet-libre S1:\TRs appear to be otherwise improvisatorily “arbiand Solo 2 developmental spaces, begin with trary’ modular selections carry hermeneutic

the linkage-technique of taking up material connotations? One might propose that even that had just been stated in the orchestra (as in within this structural parenthesis, while the his cadenzas for K. 415, 449, 450, 451, and 456), rest of the composition is put on hold, the linwhile others begin—as with developmental ear current of the modular successions provided spaces—in either a purely virtuosic or explicitly in R1 and S1 + R2 continues to exert its in-

thematic way. fluence. Or it might be that some cadenzas are More interesting are the potential rotational written to take on the role of “freely” reflecting implications of the modular content within “from outside” on previously heard material— those cadenzas that reconfigure previously heard thus providing a moment in which the sonata material. Cadenzas that cite or rework only one becomes self-reflexive, pondering some of the previously heard theme generally evade these is- modules from which the “real” structure has

sues, although they might be heard as making a been built. In so doing, a rotational cadenza prohalf-rotational gesture.>! Of the two surviving vides an ordered, if abbreviated, revisiting of the cadenzas for K. 414, for instance, one presents a concept of rotation itself, one of sonata form’s variant of R1:\P only in its center, m. 12, while most essential principles—thereby interpolating the other, much briefer, begins with a reference a telescoped, “last-glance,” nonstructural rota-

49. Additional summary-descriptions are provided Ir- into his autograph score—which opens with a reference ving, Mozart’s Piano Concertos, pp. 153 and 159. to amoment in the development (mm. 158, 162), not to 50. Eva and Paul Badura-Skoda, Interpreting Mozart on any R1 or expositional module proper. (See the sumthe Keyboard, pp. 215-16. Cf. Paul Badura-Skoda, Kaden- mary of this ““beginning—middle—end’ strategy” in zen, Eingdnge. See also the summary of this strategy in Irving, Mozart’s Piano Concertos, p. 160, and pp. 169-70,

Irving, Mozart’s Piano Concertos, pp. 160-61. n. 85.) 51. The exception among the surviving cadenzas for the 52. The appearance of “secondary material” in the inlater concertos would appear to be that for K. 488—as terior of cadenzas is also noted in Irving, Mozart’s Piano it happens, the only cadenza that Mozart actually wrote Concertos, p. 161.

602 Elements of Sonata Theory

tion-within-a-rotation. Still, one might be ad- module of R2 (m. 169) but is thereafter dropped vised from forwarding such speculations with- from the rest of the composition. Thus it does out the proper caveats: the cadenza, after all, not figure in the larger recapitulation and 1s remains a structurally free space. Moreover, in never resolved back into the tonic. Mozart’s surat least one case from the end of Mozart’s career, viving cadenza for the movement restores the the surviving cadenza seems, if anything, coun- missing module prominently, and in the tonic, terrotational, citing no fewer than four modules in m. 15—reminding his listeners of what the in reverse rotational order. This happens in the recapitulation had “forgotten.” cadenza for K. 595, in which the succession is: Finally, we might open the issue of what it R1IAC [from m. 39] + R1IA\TR!+ [from m. 16] could mean for a cadenza to end with a trill-ca+ an interior module of R1:\P (cf. mm. 5-6, dence that recaptures the one that had ended S3 although the reference may be to its version in not too long before. It 1s a principle of Sonata the development, mm. 194—96) + R1:\P proper Theory that an apparent ESC can be regarded

[cf. m. 1]). as deferred if its most essential cadence-defining

From time to time one encounters discus- particle is revisited as a refrain cadence concludsions of such cadenzas as “developmental” or as ing a later, differing module. One might there“secondary developments,” particularly if their — fore ask, in cases where the S3:\ESC has been interiors deviate momentarily onto nontonic ar- articulated at the downbeat of R4!, whether the eas.° As was the case in altered or expanded re- return of the same trill-cadence at the end of capitulatory transitions (see chapter 11), this ter- the cadenza implies that the S3:\ESC has been minology is unhelpful, since the “development” shifted to that point—thus passing over the or “developmental space” is a zone-specific term new-zone implications of R4! or at least creating within sonatas. To be sure, the attraction of such a cadence-structural interlock between R4! and a description 1s clear: there is no doubt that such the cadenza. This interpretation would grant a cadenzas can sometimes bring into play a dif- genuinely structural potential to the otherwise fering style of thematische Arbeit (thematic work). “arbitrary” cadenza: that of trumping the end Still, there is little or any essentially structural of some S3s by producing the “real” S3:\ESC. sonata work that is implied in this procedure, Rather than deciding the issue one way or the

however much individual modules might be other, it is preferable to explicate the ambigu-

improvisatorially reshaped. ity, which is inextricable from this moment of A larger question would be whether it 1s pos- the Type 5 sonata. From another perspective, sible for a cadenza to restore or compensate for one might wonder whether the explicit return otherwise “lost” or understated material from of the trill-cadence could also be understood as the sonata proper, thereby providing a balance implying a “backing-up” of the composition or completion lacking 1n the rest of the move- to the concluding S3-point, erasing whatever ment. Here one would have to posit the compo- structural work might have been accomplished sition of an absence or incompleteness into the in R4'. This underscores the generic fragility of sonata proper, which could then be addressed as the R4! position, which under this interpretaa conceptual topic in the cadenza-improvisation. tion 1s conceptually marginalized through such Remote as such a possibility might seem, there a backup maneuver. Obviously relevant in such is at least one case in Mozart where it seems a reading would be whether R4! had been rotato have happened. In Piano Concerto No. 14 tionally inert, rotationally participatory, or per-

in E-flat, K. 449, a prominent dotted-rhythm haps both. figure, R1:\S?-2 (m. 63), returns as the opening

53. Cf. Irving, Mozart’s Piano Concertos, pp. 161 and 169,

n. 83.

APPENDIX 1 © "ISLE 0

Some Groundingg Principles Principl

of Sonata Theory

The Intellectual Backdrop readings of individual works unsupported by adequate music analysis are all too easily proSonata Theory aims to facilitate issues of musi- duced and ring hollow. Conversely, even though

cal and cultural meaning. We regard the mul- analysis, with all of its technical terminology tiple meaning-systems present within musi- and lumber-room mechanisms, looms large at cal compositions as conceptually divisible into the initial stage of one’s inquiry, that first stage these two tracks, which are often treated singly is no end in itself. Rather, all analysis should or separately. Toward the end of bringing them be directed toward the larger goal of a hermetogether—or at least much closer—the theory neutic understanding of music as a communicaseeks in its vectored language and terminology tive system, a cultural discourse implicated in to ensure that its analytical observations are con- issues of humanness, worldview, and ideology, notationally rich with interpretive implications. widely construed—the second stage of the proIn part it does this by considering each obser- cess. Once one 1s sufficiently comfortable with vation against the backdrop of a proposed sys- the analytical system, the two stages proceed sitem concerning “how sonatas work.” Elements multaneously.

of Sonata Theory provides the music-analytical Sonata Theory is grounded in a blend of outlines of that backdrop, along with glimpses many strains of later-twentieth-century thought.

of what lies ahead in more advanced work. Seeking maximal flexibility, it is methodologiSonata Theory 1s a method attentive to the cally pluralistic, a hybrid between the rigorous details of individual compositions, but it is also precision of current English-language analytical more than that. If at first Sonata Theory seems practice and several registers of the broad-gauge almost exclusively formalistic in its concerns, interpretation and imaginative sweep encoun-

this is because obtaining an adequate back- tered in much continental thought of the past eround in analysis is the sine qua non of the century (thought often pursued also in recent larger system in which we are interested. Music Anglophone writing in the humanities). In this analysis is a first stage that cannot be dispensed respect Sonata Theory brings together aspects with. In any discussion of music, insufficient or of traditional music-disciplinary work—espedefective analysis undermines the legitimacy of cially some of the newer developments in recent broader interpretive claims and calls the com- music theory and musicology—with bolder inmentator’s competence into question; cultural terpretive considerations often sidelined in for603

604 Elements of Sonata Theory

mer decades as extra-disciplinary. Most prom1i- lard, Slavoj Zizek, and others. The wellsprings nently, these latter include: genre theory (work and potential resonances of the theory are many. by Mikhail Bakhtin, E. H. Gombrich, Alistair To reconstruct and defend them in any metatheFowler, Tzvetan Todorov, Hans Robert Jauss, oretical detail would produce a different book Adena Rosmarin, Fredric Jameson, Thomas O. altogether.!

Beebee, Margaret Cohen, and others); certain Sonata Theory may also be described as the features of phenomenology (aspects of Edmund style of analysis and hermeneutics resulting from Husserl, Martin Heidegger, Roman Ingarden, the flexibility provided by that particular blend. and others—Sartre, Merleau-Ponty—on one’s One of its convictions is that in order to arrive perception and processing of the artwork); the at an adequate sense of meaning within a work, explorations of hermeneutics by Hans-Georg Ga- we must reconstruct a sufficiently detailed gedamer and that method’s immediate successors; neric and cultural backdrop against which such and reader-response theory (particularly the issues individual works sought to play themselves out. raised by the Constance school, including Wolf- Genre theory, flexibly construed, is not necesgang Iser and, once again, Hans Robert Jauss). sarily the primus inter pares of the extra-disciplinAs the occasion demands—though admittedly ary sources of the Sonata Theory blend, but benot much in the more explicitly “music-theo- cause of the vastness of detail that it encourages retical” Elements itself—we are also prepared to as an initial step, it is something that one bumps fold into our interpretations ideas suggested by up against immediately and frequently. A suba number of prominent sociological theories (Pierre stantial part of the Elements stems from it.

Bourdieu on the field of cultural production, Genre theory is a complex and contested conAnthony Giddens on structuration, Niklas Luh- stellation of interests. It comprises many things mann on modern social differentiation and art at once: theories of how genres may be said to as an autopoictic system, Jiirgen Habermas on exist at all; speculations on the way that they are modernism and the public sphere). And we re- formed and how and for which social purposes main open to personal intermixtures or accents they are sustained; studies of the ramifications from critical theory. These include theories of | of the proposal that individual works can excultural materialism and the theory of ideology and in- emplify, illustrate, or contend with the genres;

stitutions—for instance, those of Raymond Wil- more current reinterpretations of genres as soliams, Theodor W. Adorno, Fredric Jameson, cial contracts, social relations, or Bourdieuian Peter and Christa Biirger, or Terry Eagleton—as “pnosition-takings” (prises de position) in a conwell as a broad array of postmodernist-poststructur- tested cultural field;? and so on. In addition, alist concerns including work by Roland Barthes, many would argue—as would we—that genres Michel Foucault, Jacques Derrida, Jean Baudril- are hermeneutic tools: rules-of-thumb or regu-

1. There are those, we realize, who would view the social, cultural, ideological, and so on—none of which intermixing of such disparate intellectual paths (which should be regarded as illegitimate. Different questions we would characterize as the strategic tacking from one require different sets of tools in pursuit of answers. (See set of influences to another) to be a dubious enterprise. proposition 9.)

Nevertheless, we agree with Carl Dahlhaus’s observa- 2. On genres as social contracts see, e.g., Fredric tion—originally formulated in response to East-West Jameson, The Political Unconscious: Narrative as a Socially

Germanic methodological challenges of the 1960s and Symbolic Act (Ithaca, N.Y.: Cornell University Press, 1970s (Foundations of Music History, trans. J. B. Robin- 1981), p. 104, and Jameson, “Beyond the Cave: Demys-

son [Cambridge: Cambridge University Press, 1983], tifying the Ideology of Modernism,” in The Ideologies e.g., pp. 24, 116, 122)—that the only other available of Theory: Essays, 1971-1986 (Minneapolis: University alternative today, a dogmatic or arbitrary adherence to of Minnesota Press, 1988), 2:115-32, esp. p. 116. Fora one system or one set of interests alone, is not only less recent endorsement of Jameson’s view coupled with an attractive but also indicts itself, especially now in the interpretation of genres as cultural prises de position, see twenty-first century, as naively reductive and outdated. Margaret Cohen, The Sentimental Education of the Novel The analyst must be free to use whatever tools will assist (Princteon, N.J.: Princeton University Press, 1999), pp.

him or her to address the questions at hand. There are 16-26. many questions to ask of a piece of music—analytical,

Some Grounding Principles of Sonata Theory 605

lative principles (in the Kantian sense) to guide a reading) in the mind of the informed listener.4 interpretation. (For the Kantian aspect, see It exists less substantially in the manuscript score

Proposition No. 3 below.) From this perspec- or printed notation; much less in the fleeting tive, choosing a genre to serve as an interpre- representations of any individual performance tive lens predetermines the patterns of what one or diachronic set of performances. The essenis likely to find in the individual object under __ tial character of this dialogue is not self-evident

scrutiny. on the music’s acoustic surface alone (its literal sounds). Therefore the central task of analysis 1s to reanimate this implicit dialogue in a way that

“Texts Always Take Place on the Level of is historically and musically sensitive.

Their Reader’s Abilities” 2. Toward that end, an intellectually responsible, culturally aware reading should seck to Rather than presenting a philosophical/literary- reconstruct the historical norms and variable critical argument to buttress each point of the options of the relevant genre. Absent a masSonata Theory blend, it might be more helpful ter-set of detailed instructions known to have at this stage merely to present some of the con- been carried out by composers, this must be victions and conclusions that helped to gener- done inductively, leading to constantly testate the theory throughout the 1990s and early ed-and-retested conclusions (a procedure simi2000s.° These are provided in the following set lar to that of formulating scientific hypotheses, of propositions, albeit in a compressed and un- _ consistently open to amendment and revision)5 der-argued format: to take even the first steps that emerge from the study of the works of the toward a sufficient documentation would be more influential composers of the period. To be cumbersome in the extreme. Instead, we hope sure, original theoretical writings—Koch, Gathat, taken together, the collection of proposals leazzi, Reicha, and so on—are to be taken into might serve to suggest a style of thought that is account, but as massively reductive generalizacharacteristic of the theory when it shifts into a tions they ultimately prove to be of secondary

self-reflective mode. importance. A more robust quality of informa-

tion is to be gained by the close study of actual 1. A piece of music may be said to exist on —_ musical practice. One goes directly to the musimany conceptual levels. (The mode of existence cal sources to learn what the masters do in real

of a work of art is a perennial issue in philoso- composition. phy.) For the purposes of structural analysis it 3. The genre thus reconstructed is to be reexists most substantially in the ongoing dia- garded as an implicit and necessary backdrop logue that it may be understood to pursue with _ that functions heuristically. In other words, it its stated or implied genre—a dialogue that exists not literally but rather as something like may be recreated (more accurately, proposed as a (Kantian) regulative principle, a rule for in-

3. The section title quotes Wolfgang Iser’s paraphrase structure of the text?” Iser, The Act of Reading: A Theory of a remark made by Jean-Paul Sartre. See proposition of Aesthetic Response (Baltimore: Johns Hopkins, 1978

11 and n. 13. [orig. German, 1976], pp. 34-35.

4. Cf. the remarks of Wolfgang Iser concerning the role 5. These hypotheses should be stated as “falsifiable of the “implied reader” within literary texts: “No mat- propositions,” open to and inviting “intersubjective ter who or what he may be, the real reader is always criticism,” in Karl Popper’s sense. See, e.g., Popper, offered a particular role to play, and it is this role that “The Problem of Demarcation” [1974] in Popper Selecconstitutes the concept of the implied reader. . . . The tions, ed. David Miller (Princeton, N.J.: Princeton Unitext must... bring about a standpoint from which the versity Press, 1985), pp. 118-30. reader will be able to view things that would never have 6. In Kant, Critique of Pure Reason, trans. J. M. D. Meicome into focus as long as his own habitual dispositions klejohn (London: Dent[Everyman], 1934), pp. 305, were determining his orientation, and what is more, 388-89 (from book II, “Transcendental Dialectic,” ch. this standpoint must be able to accommodate all kinds 2, section 8; and ch. 3, “The Ideal of Pure Reason,” of different readers. How, then, can it evolve from the section 7), the philosopher distinguished a “regulative”

606 Elements of Sonata Theory

terpretation that enables and constrains the choices made by numerous pivotal individuals production and subsequent reading of compo- over a span of time and ratified by communities sitionally placed musical events. Sonata form— of listeners to suit their own purposes. Musical the topic of this book—is one such regulative genres inevitably implicate communities of lisprinciple. The genre-system 1s the decoder of an teners.® For this reason genres contain social and otherwise unintelligible or free-floating musical ideological connotations that may also be teased message. For such reasons the most substantial out—or proposed—by means of hermeneutic existence of a piece of music (No. 1) may be inquiry.? Genres transform over time and differ called forth only by an act of hermeneutics—a from place to place. They are not static enticonceptual concretization of the work—on the ties. Rather, they are elaborate constellations of part of the informed listener/analyst. Once the norms and traditions. Generic forces are fluid, composition 1s out of the composer’s hands, it systems-in-motion. is the hstener/analyst who creates or reawakens 5. Thus a genre is not an autonomous, sepathe substantial work of art. This concretization rate organism existing apart from society. In-

can also impact the performance of that work. stead, a genre (such as the network of con4. As we construe them, musical genres (such ceptual forces that we call sonata form) is an as “sonata form” or “the multimovement so- agreed-upon set of guidelines devised and used nata”) are to be distinguished from mere forms by producers and receivers in a given time and insofar as they also carry an implicit social or place in order to permit certain kinds of meanideological content.’ A schematic form be- ing to happen. Genres exist only insofar as procomes a genre when we also attend to its social duction and reception communities agree to and cultural ramifications—among, which 1s its act as if they really did exist, as sets of rules, decisive position-taking on a contested social assumptions, or expectations. In that sense a field of cultural production. Musical genres are genre system resembles a game. In order to play usually not the conceptual products of isolated that game as opposed to a different one (in the individuals. Instead, they are socially consti- former case, to approach the originally desired tuted and reinforced, the results of hundreds of sorts of meaning), one has to be willing to ac-

principle from a “constitutive” principle: only the lat- a social institution, something like a social contract in ter, he believed, existed in reality. Thus adapting from which we agree to respect certain rules about the approKant’s first Critique: “|The regulative principle] is a priate use of the piece of language [or music] in quesprinciple of reason, which, as a rule, dictates how we tion” (“Beyond the Cave,” p. 116). ought to proceed in our empirical [inquiry], but is un- 9. Ideology, in this sense, is not a “thing,” no simple able to anticipate or indicate prior to the empirical [in- listing of commonly shared ideas; it 1s perhaps closer to quiry] what is given in the [analytical] object itself... . Gramsci’s much-cited concept of “hegemony,” perhaps [It] is valid only asarule.... It cannot tell us what the as elaborated by Raymond Williams. Another treatment object is, but only how the empirical [inquiry] is to be pro- of such a position may be found in Thomas O. Beebee, ceeded with 1n order to attain to the complete conception The Ideology of Genre: A Comparative Study of Generic In-

of the object.” Or: “The [regulative] idea is properly a stability (University Park: University of Pennsylvania heuristic, and not an ostensive conception; it does not Press, 1994), pp. 18-19: “Ideology is no longer somegive us any information respecting the constitution of thing that can be represented or paraphrased. Instead, it an object, it merely indicates how, under the guidance becomes something like a magnetic field that arranges of the idea, we ought to investigate the constitution and a chaotic mass of iron filings into intriguing, ordered the relations of objects in the world of experience. ... curves on a piece of paper. Ideology itself is usually in[Regulative ideas] cannot, therefore, be admitted to be visible. . . . It is only in the deformations and contrareal in themselves; they can only possess a comparative dictions of writing and thinking that we can recognize reality. ... They are to be regarded not as actual things, ideology; genre is one of those observable deformations.

but as in some measure analogous to them.” ... Asa form of ideology, genre is also never fully iden7. This point was also proposed in Hepokoski, “Genre tical with itself, nor are texts fully identical with their and Content in Mid-Century Verdi: “Addio, del pas- genres. Furthermore, if genre is a form of ideology, then sato’ (La traviata, Act III),” Cambridge Opera Journal 1 the struggle against or deviation from genre are ideo-

(1989), 249-76. logical struggles. .. . The generic classification of a text

8. Once again to invoke Jameson (n. 2), “genre is itself determines its meaning(s) and exposes its ideology.”

Some Grounding Principles of Sonata Theory 607

cept the regulative and constitutive rules. More- or her composition. Those aspects are socially over, participants within the game can come to predetermined (although the composer may armodify or influence some of the appropriated range pre-existing genre-blocks into individual

rules, gradually and over time, through indi- patterns).'! Thus in any composition there are vidual choices and interactions with the sys- at least two voices: the composer’s voice and the tem. Whenever such devised agreements (even genre’s voice. (To complicate matters further, if tacit) were widely shared in the past (or even the genre’s voice 1s itself'a pluralistic site of social assumed to have been shared in most of their tensions.) There is no reason to assume that what

essentials), genre study is open to historical in- the composer seeks to say and what the genre quiry. We may also study their diachronic trans- secks to say will be in concord. formations, and we may speak, however loosely, 7. In any hermeneutic study of a work it is of genres as having arisen, flourished, or de- crucial to distinguish between genres and indi-

clined.!° vidual exemplars of genres (individual compo-

6. When a composer creates an individual sitions). This leads at once to the possibility of work in dialogue with a genre, many compo- making a distinction between cultural critique sitional decisions are pre-given socially by that and aesthetics and affirming the need for both genre. Beethoven was by no means the only perspectives. On the one hand, the potential composer of the Eroica: he cannot lay exclu- sociohistorical content of a musical artifact resive claim to the totality of the work’s impli- sides most purely in its genre, not in any indications. Many of the compositional features of | vidual exemplar of it. Genres are social sites of that piece are more accurately regarded as dra- discourse. For an interpreter to decenter a work matized affirmations of (or dialogues with) pre- into a social text is, in part, to override its claim existing, culturally produced norms that were to individuality in order to dissolve the utterexternal to Beethoven: the very concept of the ance back into a concern for the social embed“symphony”; expectations regarding standard dedness of its genre. One important aspect of designs, lengths, tonal norms, sonata-form con- Beethoven’s Eroica, in other words—an aspect ventions, and orchestral practice; the inbuilt that is social, cultural, and ideological—is its awareness of probable occasions for performance; affirmative participation in the game of “symthe presence of the existing symphonic tradition phony-ness,” a game of cultural prestige socially

and well-known remembered works; and so devised in the pursuit of specifically advantaon. Because the tacit societal aspects inscribed geous social positions. On the other hand, the within a genre-constellation were also given and potential within a given reception community accepted as self-evident, they cannot be made to discern differing degrees of aesthetic content subject to an act of personal intentionality. No resides most purely in the particularities of the composer composes the generic aspects of his exemplar.!? By a sense of aesthetic presence, we 10. The term “transformation” is used here to sidestep to ‘translate’ is itself only a ready-formed dictionary, the more loaded (and usually much misunderstood) term its words only explainable through other words, and so “evolution,” with its organic and natural-selection impli- on indefinitely.” (“The Death of the Author” [1968], cations. (Cf. “transform” instead of “evolve” in no. 4.) in Image Music Text, trans. Stephen Heath [New York: 11. Cf. the famous remarks of Roland Barthes (which in Noonday, 1977], p. 146.) hindsight may be best regarded as a performative exag- 12. There are no ontological claims here: these contents geration on Barthes’s part, even as the remarks continue are not objectively “in” the piece, except to the extent to serve as a challenging corrective on more commonly that communities agree—socially—to believe in their

held views): “|The text is] a multi-dimensional space existence. In this proposal we are not sketching out a in which a variety of writings, none of them origi- rigid dichotomy in which only genre is cultural and only nal, blend and clash. The text is a tissue of quotations the individual utterance is aesthetic. Rather, we are redrawn from the innumerable centres of culture... . ferring to general tendencies, and there is much ambi[The author’s] only power is to mix writings, to counter euity, overlap, and interplay between them. The seemthe ones with the others, in such a way as never to rest ing separation here of the aesthetic and the cultural is on any one of them. Did he wish to express himself, he principally to furnish a conceptual clarity at this stage ought at least to know that the inner ‘thing’ he thinks of the discussion.

608 Elements of Sonata Theory

mean the invitation to experience the thing individual or group listeners. Obtaining the 1maesthetically—to “delight” in [delectare] an indi- pression of a meaning depends on the eftectivevidual object’s “made-ness”’ [ poiesis], or through ness of the hermeneutic genre that one uses to

sustained attention to call forth, savor, or be- process the work—and hermeneutic genres are come self-absorbed in the being and details of — legion. A work will generally seem to provide that made-ness. At this point the object 1s con- an answer to any question posed of it. While in sidered to exist primarily—though not neces- the abstract there may be no “wrong questions,” sarily exclusively—for aesthetic contemplation. the abstract claim can be misleading—or inapFrom this perspective, other aspects of potential propriately comforting. Shallow questions call meaning or connotation within the object are forth shallow answers. Some questions are more not sidelined altogether but are rendered sec- informed, more text-adequate, more historically ondary to the immediate purpose at hand. The relevant, and more appropriate than others. ‘The reader or listener chooses to isolate the specific questions that we ask of a musical work depend qualities and distinctions of the individual ex- on our own interests in doing so. Where one emplar, considered in dialogue with the back- situates oneself in asking the questions 1s imporsround genre that regulates the class of objects tant: we do not ask questions “from nowhere.”

to which it belongs. The perspectives from which these questions

8. Reinforcing No. 7: Because of their par- may be asked seem endless. Realizing the comticipation in genres and differing patterns of re- plexity of the issues at hand invites each of us to ception, musical works are both ideological and interrogate—and possibly to modify—the hidaesthetic. Any claim that wishes methodologi- den interests of our own subject-positions as we cally to invalidate either side by trumping the devise our own sets of questions to pursue. one with the other is incomplete and misguided. 10. One should be cautious in reconstructing The either/or issue commonly insisted upon by the internal anatomy and details of the formal partisans is a false dualism. One need not decide aspects of musical genres. The characteristic erwhich side 1s correct. Both are, since the answer ror is to reconstruct these things too sparsely depends entirely on the kinds of questions one or too stiffly. Far from being rigidly prescrip-

finds interesting and wishes to pose regarding tive, genres, properly construed, provide for a the work. A sufficient awareness of the nature flexible set of options at any given point in the of both the genre and the exemplar—both the realization of any individual exemplar. (‘This 1s preponderantly social and the potentially aes- certainly true of sonata form.) In practice, some thetic—is necessary to produce an adequate generic options were more frequently favored discussion of any composition. On this side of — than others. Within any “zone” of the genre the new challenges in the humanities, one-sided there were hierarchies of choices, hierarchies of discussions—from either faction of the partisan norms that we may consider to have been ardivide—are likely to seem inadequate, simplis- ranged into first-level defaults (the most comtic, or reductive. There is no returning to sim- mon options, the standard choices pre-made by

pler times and simpler methodologies. the genre unless they were overridden), second9. More generally, musical works should not level defaults, and so on. (As suggested whimbe supposed to contain only one correct mean- sically in ch. 1, the compositional situation 1s ing (what the piece “really” means) to be un- analogous to a selection of preranked formatting covered, in the manner of a lost object or thing, choices arrayed on “wizard” help feature within by the analyst. Instead, they house multiple, a computer program.) Genres are preformatted

sometimes conflicting strata of meaning(s) along lines of social preference, although the to be drawn forth through differing readings favored choices may be (and are) altered with (which include current performances). Works time. Reconstructing the genre involves recreare therefore fields of expressive possibility that ating the specifics of this flexible set of weighted

harbor differing, potential configurations of | default-choices for each interior zone. meaning, configurations that may be concret- 11. At any point in producing the individized only in their awakenings (constructions) by ual composition, a composer may realize with

Some Grounding Principles of Sonata Theory 609

personally crafted music the preformatted sug- conceptual field of implication and normative gestions of the various default options provided practices within which the acoustically heard to him or her by the genre (elegantly or dra- music asks to be construed. The notation on the matically realizing, for example, the first-level, page 1s not self-sufficient 1n the production of second-level, or third-level formatting default). any adequate interpretation of the music. While On the other hand, if desired, the composer several other theories of large-scale form seek to may strain any of the preformatted options to account only for what happens in an individual produce, in extreme cases, a deformation of the piece or passage, Sonata Theory intertwines this originally provided generic suggestion. Or the with a concern to observe also what does not composer may override all of the default op- happen or what is kept from happening. This is tions entirely, thus refusing to follow any of — a necessary consequence of understanding inthe options that were socially provided (that dividual works dialogically—as dialogues with were likely to have been provided, that 1s, in existing genres (No. 1).

the hypothetical “wizard” help feature in the 13. The preceding propositions may be recomputer analogy suggested above)—presum- inforced with a quotation from Wolfgang Iser ably in order to provide a strong, surprising, or (1976) about “blanks and minus functions” that individualized effect. This, too, is a deformation we endorse. This is a differing articulation of a of the genre. (On the term “deformation,” see general concept central to our theory. appendix 2.)

12. It is also possible that in any composi- If we say that an unfulfilled function can become tional zone a composer could choose to suppress a background, we are presupposing familiarity a strongly expected or normative event within with literary [for us, musical] texts. As Sartre

the genre. In these cases—which are by no has rightly pointed out, texts always take place means as infrequent as one might suppose—it on the level of their reader's abilities. Now if a is the absence of a certain occurrence, norm, or literary text does not fulfill its traditionally ex-

; ; pected functions, but instead uses its technique

PFOCESS that is the important factor. (The norm to transform expected functions into “minus can be written over by other material or options functions’—which is the deliberate omission of on the acoustic surface of the music.) Thus any a generic technique [as Lotman has argued]—in thoroughgoing analytical system should have order to invoke their nonfulfillment in the cona procedure to come to terms with the pres- scious mind of the reader, anyone who is not faence of absence—with what is clearly implied miliar with these traditional functions will autobut does not literally happen (blocked or unat- matically miss the communicatory intention of tained goals, ellipses of supposedly obligatory this technique widely applied in modern literaevents, non-normative articulations of certain ture. He will experience a sense of disorientation zones overriding a now-silenced norm, and so and may react accordingly, thus involuntarily reon). Many current music-theoretical approaches vealing the exp ectatons Fo which he app cars FO

, ; designed , be irrevocably committed. But the more to form were to deal only with what isthe_— . that familiar the reader is with functions are now be-

printed on the page. But what occurs notation- ing “nonfulfilled,” the more definite will be his ally—or does not occur—can make sense or cre- expectations, and so the more responsive will he ate an impression only within a backdrop-field be to their frustration ./3 charged with generic expectation: the larger 13. Iser, The Act of Reading, pp. 207-8. Iser’s Sartrean to him as inexhaustible and opaque as things.” (‘What reference—undocumented in The Act of Reading—al- is Literature?’ and Other Essays, trans. anon. [Cambridge, ludes to that author’s set of 1947 essays, What is Litera- Mass.: Harvard University Press, 1988], p. 54. Cf. the

ture?, especially to the section, “Why Write?” There original text from its first publication as a complete Sartre had written: “Thus, for the reader, all is to do book, Qu’est-ce que la littérature [Paris: Gallimard, 1948], and all is already done; the work exists only at the exact p. 58: “Ainsi, pour le lecteur, tout est a faire et tout est level of his capacities; while he reads and creates, he déja fait; ’ceuvre n’existe qu’au niveau exact de ses caknows that he can always go further in his reading, can pacités.’’) always create more profoundly, and thus the work seems

610 Elements of Sonata Theory

14. When individual occurrences happen his own standard practice—was different from within a piece that are extrageneric or con- Mozart’s. Broad, idiosyncratic patterns of custravene the generic norms, this 1n itself does tomization are in dialogue with the larger, more not alter or change the genre within which it generalized generic norms (with the normal or is participating. That genre continues to exist preformatted style generally obtained by all). as a background rule for interpretation. In or- When one finds a seemingly unusual effect in der to be considered a sonata, a work does not Haydn, for example, one might ask: is the effect have to fulfill a certain number of tonal, me- only part of Haydn’s customized deformation lodic, or structural criteria before being admit- of standard practice? (is it a norm for Haydn, ted under that classification. Instead, to call a though not for Mozart or J. C. Bach?); or is the work a sonata is to conclude that, on the basis effect a deformation even of his customization?; of the evidence at hand, it does indeed invite to what degree? and so on. us (in any number of ways) to use our generic 16. In any adequate genre-based analysis, conception of a sonata as the regulative princi- the goal must not be merely to identify patple of interpretation by which to understand its terns and to assign labels to them. To be sure, events. (As a corollary, there would have to be a a background taxonomy of labels is necessary sufficient set of reasons within the work—sonic to cover the basic set of possibilities within a signals, movement-or-work titles, positional genre: such labels provide a shorthand way of placements within multimovement construc- talking about these things. But any analysis that tions, and so on—to justify why we would be stops after the mere labeling is no analysis at tempted to consider it as a sonata at all. Ob- all. Under no circumstances should an analyviously, many pieces in different genres would sis seek to normalize unusual occurrences and not even begin to extend this invitation.) What anomalies: one should acquire a healthy distrust literally happens on the piece’s acoustic surface of all systems and catch-phrases that work in may or may not be normative for sonatas, but this direction. Rather, in confronting potenthose events are not the ultimately determining tially ambiguous situations—and sonatas are factors in defining what a sonata 1s. If the devia- filled with them—the proper goal of analysis is tions in any given work are extreme, we may be to explicate the ambiguities, to reawaken the

better off, however, referring to it instead as a strains and uncertainties within the text, not sonata deformation—as an unusual, perhaps ad to suppress them or filter them out. Moreover, hoc structure still dependent on sonata norms for analyses that seek a facile closure of explanation

interpretation. are invariably short-sighted. Unless a musical 15. Master composers typically create “cus- text is problematized—or brought to a deeper tomized” versions of the socially produced level of questioning and inquiry, where lurking genre for their own use. Such customizations and troublesome questions still remain (or are involve the arraying of characteristic choices finally glimpsed at a more proper level) —then that they prefer to make as part of their own that analysis 1s inadequate. The goal of analysis individualized styles. (This is analogous to the must never be to explain away the difficulties of setting-up of personalized formatting options, a musical work but rather to call forth a work’s involving selective modifications of what is problems, tensions, and larger implications. pregiven.) Haydn’s customization—his sense of

APPENDIX 2 © "ESRI 0 ‘

Terminology “Rotation” and “Deformation”

Rotation Another, perhaps more sophisticated, metaphor is that of tracking a large spiral through Although they differ in their degrees of sub- two or more cycles. No set of events that untlety and strictness, sonata movements are en- folds in nonrecoverable, ever-elapsing time can gaged in a dialogue with a more basic archi- exist In a condition of complete identity to any tectural principle of large-scale recurrence that similar set that has preceded it. An essential we call rotation. Rotational structures are those feature of all such constructions is the tension that extend through musical space by recycling generated between the blank linearity of nonone or more times—with appropriate altera- repeatable time and the quasi-ceremonial cirtions and adjustments—a referential thematic cularity of any repeatable events or structures pattern established as an ordered succession at that are inlaid into it. Rotational procedures are the piece’s outset. In each case the implication is srounded in a dialectic of persistent loss (the that once we have arrived at the end of the the- permanent death of each instant as it lapses into

matic pattern, the next step will bring us back the next) and the impulse to seek a temporal to its opening, or to a variant thereof, in or- “return to the origin,” a cyclical renewal and der to initiate another (often modified) move rebeginning. And indeed, quite apart from the through the configuration. The end leads into issue of merely inhabiting a different temporal the next beginning. This produces the impres- space, successive rotations in music are often sion of circularity or cycling in all formal types subjected to telling variation: portions of them that we regard as rotational. One metaphorical may dwell longer on individual modules of the image that might be invoked here is that of a original musical arrangement; they may omit clock-hand sweeping through multiple hours, some of the ordered modules along the way; or with the face of the clock representing the suc- they may be shortened, truncated, telescoped, cessive stages of the thematic pattern. 11:59:59 expanded, developed, decorated, or altered with leads inevitably to 12:00:00 (= 0:00:00) and ad hoc internal substitutions or episodic interpoanother round through the cycle. Simularly, lations. Not infrequently these varied multiple the regeneration of day upon day, calendar year recyclings build cumulatively toward a longerupon calendar year, suggests how strongly this range goal. In addition, within any individual perception of circular recurrence has been 1m- rotation an internal, smaller-pattern cycling can

pressed upon our experience. give the impression of a local subrotation. These 611

612 Elements of Sonata Theory

include such things as thematic-block restate- sonata movement—in each of the five sonata ments, the altered recurrences of larger sequen- types—are omnipresent in the Elements.

tial blocks or zones, and the like. Within a sonata, tonality is irrelevant to The rotational idea is an archetypal prin- the task of identifying the rotational principle. ciple of musical structure: a referential model The central thing is an implied or actualized followed by (usually varied) recyclings or re- ordered sweep through a temporal sequence of statements. It underpins a generous diversity thematic modules, along with the assumption of forms that may be distinguished from one that the most “natural” or expected continuaanother on more surface-oriented levels: theme tion of the layout’s last module will be to lead to and variations; strophic songs; strophic varia- a relaunching of the initial module of the next, tion; rondos (chapter 18); different types of | thus producing the characteristic spiral or circuostinato-grounded works; and the like.! Any lar effect. Rotation is what we call a rhetorical form that emphasizes return and rebeginning principle rather than a tonal one: it 1s governed

is in dialogue with the rotational principle.? by the expectation of a temporal presentationOne of the defining features of a sonata 1s the sequence of thematic-modular elements, not by particular way in which it textures and shapes harmonic procedures, even though, on another the underlying rotational idea. In a sonata-form plane of analysis, those harmonic features have composition the referential pattern laid down their own structures to articulate.

at the beginning is typically much longer and The underlying principle of recycling or remore internally differentiated than that found statement has also been widely noticed by othin the smaller strophic or variation forms. Here ers. In Sonata Forms, for instance, Charles Rosen the relevant pattern is the exposition, the musi- pointed out the relevant feature particularly cal configuration provided, in a two-part ex- with regard to what we would call triple-roposition, by P TR’ S / C, including the subdi- tational Type 3 sonatas: “The need for a balvisions of each zone, if any. Discussions of the anced symmetry always remained essential to rotational implications of the remainder of the any conception of sonata in all its forms. (Many 1. Our additional discussions of the rotational princi- 2. Baroque ritornello structures, for instance, suggest ple—especially as applied to music from later decades— a set of varied refrain-like recurrences—the first ritormay be found in the following: Hepokosk1, Sibelius: Sym- nello (or tutti) often first sounded as a patterned “beginphony No. 5 (Cambridge: Cambridge University Press, ning’’—that spin off into freer, diverse episodes for the 1993), pp. 23-26, 58-84; Darcy, “The Metaphysics of remainder of each rotation. (Or, conversely, once into the Annihilation: Wagner, Schopenhauer, and the Ending piece we could also construe the rotations as comprising of the Ring,” Music Theory Spectrum, 16 (1994), 1-40 freer, episodic beginnings—the “solo” passages—each (see esp. pp. 10ff); Hepokoski, “The Essence of Sibelius: of which is concluded by a refrain-like reference to the

Creation Myths and Rotational Cycles in Luonnotar,” ritornello or a portion thereof.) Extending such ideas The Sibelius Companion, ed. Glenda Dawn Goss (West- to varied strophic songs with refrains and to rondos port, Conn.: Greenwood, 1996), pp. 121-46; Darcy, is an easy matter. Similarly, da capo (large-formatted “Bruckner’s Sonata Deformations,” Bruckner Studies, ed. ABA’s) or other emphatically ternary structures could Timothy L. Jackson and Paul Hawkshaw (Cambridge: be interpreted as two “external” rotations of a musiCambridge University Press, 1997), pp. 256-77; He- cal pattern, separated by a contrasting interpolation in pokoski, “Rotations, Sketches, and [Sibelius’s] Sixth the middle. Extending the metaphor further, one might Symphony,” Sibelius Studies, ed. Timothy L. Jackson and wonder whether we should entertain the possibility that

Veyo Murtomiki (Cambridge: Cambridge University the contrasting central section of such forms might even Press, 2001), pp. 322-51; Darcy, “Rotational Form, Te- be understood as a substitution for an unsounded middle leological Genesis, and Fantasy-Projection in the Slow rotation: an erasing or writing over a rotation that is Movement of Mahler’s Sixth Symphony, 19th-Century potentially conceptually present but rendered tacit by Music 24 (2001), 49-74; Hepokoski, “Beethoven Re- the events sounded on the acoustic surface. However we ception: The Symphonic Tradition,” in Jim Samson, might seek to assess these “ternary” considerations and ed., The Cambridge History of Nineteenth-Century Music interpretive possibilities, they are relevant to the ways (Cambridge: Cambridge University Press, 2002), pp. in which we choose to understand the developmental 424-59; and Hepokoski, “Structure, Implication, and spaces of sonata forms. the End of Suor Angelica,” Studi pucciniani 3 (2004), 241-64.

Terminology: “Rotation” and “Deformation” 613

development sections reveal this, as they take up along Procrustean-rotational lines. For such the complete thematic pattern of the exposition, developments the main hermeneutic problem and develop each theme in turn.)”? So much 1s might be to wonder whether (and to what deobvious. But even while endorsing such prior gree) what is actually presented on the music’s statements, Sonata Theory distinguishes itself | acoustic surface as a nonrotational event is to be from them in three ways. First, we conceive the erasped as blanking-out or writing over a more restatement-symmetry postulate as wedded to normatively rotational option. The question rethe notion of circularity (as opposed to mere mains open: Is the rotational norm for developcarriage-return repetition.) Second, we treat the ments sufficiently powerful to suggest its tacit idea more flexibly. We include into the general presence (perhaps as “a choice not made’) even concept of circular repetition the related idea of — in cases when it is replaced by something else?

substantially altered restatements, such as devel- Devising a term for a previously unlabeled opmental half-rotations, truncated rotations, ro- but generally recognizable practice 1s not easy. tations with episodic substitutes “writing over” We use “rotation” in the familiar sense provided some of the expected individual elements, ro- in definition 2a of the Oxford English Dictionary: tations with newly included interpolations, in- “the fact of coming round again in succession; ternal digressions from the governing rotational a recurring series or period.’4 This meaning of thread, occasional reorderings of the modules, the word is virtually identical with two of the and the like. In general, the rotational character OED definitions of “cycle”: “a recurrent round of the whole sonata movement is underscored or course (of successive events, phenomena, whenever a development section begins with a etc.); a regular order or succession in which treatment of the primary theme (P) or whenever things recur; a round or series which returns a coda is added that is based on P-theme mate- upon itself” [definition 3]; or “a round, course, rial—regardless of what follows that material. or period through which anything runs in or(Interpreting “freer” thematic patterns that only der to its completion; a single complete period begin rotationally is a challenge called forth by or series of successive events, etc.” [definition the theory.) Third, because the rotational idea 4]. In the abstract, then, another term for rowas so important as an underlying assumption tational form would be cyclical form or cyclic orin the historical formation of the genre of so- —s ganization. The problem here is that that term nata form (and because it persisted so palpably already means something different not merely in so many later sonatas, extending through the in formal analysis but also in analytical work nineteenth century and beyond), Sonata Theory applied directly to sonata-based structures. It urges the elevation of the rotational principle to refers to a compositional strategy in which 1mbecome a foundational axiom of interpretation portant or motto themes or motives from an that in one way or another is implicated in ev- initial movement return, however transformed, ery sonata, even when it is apparently absent or in later movements.° “Rotation” and “rotational deeply obscured in developments. To be sure, form” do not have these prior (and 1n this situliterally nonrotational developments are an op- ation, misleading) connotations. This 1s not to tion within the style—though they are not as say that “rotation” is unused for other purposes common as might be supposed—and we hope in music theory. The term has a specific meanthat we are not misread as encouraging the in- ing in the analysis of serial practice and ordered terpretational forcing of any such development musical sets—as well as in the techniques of cer-

3. Rosen, Sonata Forms, p. 157. application of “rotation” is definition 2b: “Regular and 4. The Compact Edition of the Oxford English Dictionary recurring succession in office, duties, etc., of a number

[1971]. Definition 1 of “rotation,” not surprisingly, of persons. Freq. In phr. by or in rotation.” invokes the presence of an axis, which 1s less unprob- 5. See, e.g., the discussions of cyclic integration, cyclic lematically applicable to our use of the term in Sonata organization, and cyclic form—with nuanced definiTheory: “The action of moving round a centre, or of tions—in Webster, Haydn’s “Farewell” Symphony,” pp.

turning round (and round) on an axis; also, the action 7-9, 179-82, 246-47, 250-51, 252, 254-57, 258-59, of producing a motion of this kind.” Closer again to our 262-67, 280-300, 308-13, 318-20, and 327-34.

614 Elements of Sonata Theory

tain kinds of minimalist patterns that are subse- the compositional choices presented in the indiquently subjected to an organized reordering. (It vidual work confirm, extend, or override those can refer, for example, to the shifting of the last options as we move from phrase to phrase. The element of a succession to become the first—a desired goal is to be able to read the momentsingle operation that is itself usually considered to-moment action of a piece through the lenses a “rotation,” although, properly considered, the of (reconstructed) generic expectation and flexfull rotation would not have occurred until all ible generic possibility. of the elements in play have gone through one We use the term “deformation” to mean the cycle of these reorderings.) These other usages stretching of a normative procedure to its max1of “rotation,” though, refer to different reperto- mally expected limits or even beyond them—or ries and different kinds of discussions from the the overriding of that norm altogether in order one in question here. Confusion among these to produce a calculated expressive effect. It 1s

uses of “rotation” seems unlikely. precisely the strain, the distortion of the norm What other substitutions might there be (elegantly? beautifully? wittily? cleverly? stormfor “rotation”? “Strophic form” carries verbal ily? despairingly? shockingly?) for which the and poetic-textual connotations not appropri- composer strives at the deformational moment. ate here. “Theme and variations” suggests ir- The expressive or narrative point lies in the tenrelevant generic connotations and traditions of | sion between the limits of a competent listener’s a different kind. “Varied repetitions” or “varied field of generic expectations and what is made restatements’—when used apart from the rota- to occur—or not occur—in actual sound at that tional concept as the guiding backdrop—seem moment. Within any individual exemplar (such bland, unimaginatively divorced from the im- as a single musical composition) operating under

pled circularity of the procedure. Robert P. the shaping influence of a community-shared Morgan’s recent treatment of what he called genre-system, any exceptional occurrence along “circular form” displaying successive “cycles” these lines calls attention to itself as a strong exand “cyclic renewal” (for example, in the Tristan pressive effect. As such it marks an important Prelude) closely approximates our use of “rota- event of the composition at hand. A deformation,” although the methodology and interests tion may occur either locally, producing a moat stake in that description differ from our own.® mentary or short-range effect, or broadly, over Of the available terms, we have come to favor the large-scale architecture of a piece of music “rotation,” although by no means do we seek to as a whole. banish parallel descriptions from our own work:

we somermes also refer 0 ont TOLALIONS as CY Connotations of “Deformation”: Paradoxes of

cles, varied repetitions, or varied restatements, , 4, , 45the , , the “Normative” and “Non-Normative’’;

. Need for Nuance

as the occasion suggests. In referring to the

larger structure and musical process, however, the term “rotational form” is to be preferred. Since the concept of deformation 1s a central feature of Sonata Theory, we have tried to be careful in selecting and applying that term. While

Deformation we do intend “deformation” to imply a strain and distortion of the norm—the composer’s apSonata Theory V1ews compositions as individual- plication of uncommon creative force toward ized dialogues with an intricate system of norms the production of a singular aesthetic effect— and standard options. We seek to illuminate the we do not use this term in its looser, more colexpressive, dramatic, and contextual meanings loquial sense, one that can connote a negative of single compositions, in part by inquiring how assessment of aesthetic defectiveness, imperfec6. Morgan, “Circular Form in the Tristan Prelude,” Journal of the American Musicological Society 53 (2000), 69-103.

Terminology: “Rotation” and “Deformation” 615

tion, or ugliness. Here our definitions must be mation” is cooler, more detached—hopefully, explicit. Within our system, “deformation” is more connotationally “technical.” It marks only a technical term referring to a striking way of — our noticing (and often relishing) of a remarkstretching or overriding a norm. As a techni- ably unusual compositional choice; it 1s not cal term it is intended to carry no judgmentally = judgmental. negative connotation, as in some popular usages Nevertheless, we also recognize that howof the word.’ We understand that other scholars, ever carefully one might insist upon one’s intenfor other purposes, may have used the term to tions to provide only a “technical” definition of suggest some of these negative connotations.® any term, words have connotational, lateral slipBut that is not our intention. We are suggest- pages and past histories that can escape our coning neither that a sonata deformation is an unat- trol. And it may still be that some readers, for tractive structure (as opposed to any supposedly whatever purposes, might mistakenly read into more attractive or socially preferable norm) nor it only unintended implications of the negative that it is the result of a misguided execution on or the critical. For such readers—and for any the part of the composer. Nor, more locally, are readers curious about a more expanded treat-

we implying that the deformation of a medial ment of our view of the term and its connocaesura, for example, results in something that tations—we pause here to examine the matter is aesthetically negative.? To avoid encouraging (and related issues) with more patience and nusuch connotations in our own writing, we steer ance. This initially entails a backing-up to some clear of the verb “to deform” along with (espe- fundamental aspects of how we understand socially) the related word “deformed” (let alone nata form.

“deformity” !) to describe the effect of a defor- As indicated in chapter 1, our view of somation. Instead of a “deformed recapitulation” nata form is essentially dialogic, not conformawe prefer to write of a “recapitulation subjected tional. It is not the task of a sonata merely to to a deformation.”!° The abstract noun “defor- “conform” to a pre-existing template. On the 7. The Compact Edition of the Oxford English Dictionary 9. Complementarily—to suggest a clarifying reductio ad [1971] provides three definitions for the word “defor- absurdum—no serious scholar could maintain, within an

mation.” The first two have negative implications that analytical situation, that merely recognizing a familiar do not reflect how we are using the term: (1) “The ac- compositional choice as a “norm” or as “normative” tion (or result) of deforming or marring the form or inevitably connotes a tacit personal approval or moral beauty of; disfigurement, defacement”; (2) “Alteration endorsement of that norm. On the contrary (of course), of form for the worse; esp. in controversial use, the op- it is merely an acknowledgment of standard operating posite of reformation. ... An altered form of a word in procedure within the quasi-formulaic genre under obwhich its proper form is for some purpose perverted [as servation (as in, e.g., “in tonal practice the norm has God to ‘od].” Instead, our adoption of the term is analo- been to resolve dissonant sevenths downward”), an gous to the third, more technical (and nonjudgmen- awareness of what usually occurs, for whatever historital) definition: (3) “Physics. Alteration of form or shape; cal reasons, under certain circumstances and traditions relative displacement of the parts of a body or surface of musical manufacturing.

without breach of continuity; an altered form of.” 10. OED, “Deformed,” first definition: “marred in ap8. As perhaps in Renato Poggioli, The Theory of the pearance; disfigured, defaced.” Under the entry “deAvant-Garde, trans. Gerald Fitzgerald (Cambridge: Har- formed” there is no technical or neutral use that would

vard University Press, 1968), esp. pp. 176-79, in which be applicable to its potential relevance to the field of Poggioli refers, among other things, to the effects of physics (see n. 7, definition 3 above—closer to our usage “neoprimitivist deformation,” “ritual and allegrorical of “deformation”’). Nonetheless, the term “to deform” deformation,” “stylistic deformation,” and “avant-garde is encountered in scientific writing, as in the New Encydeformation” (“shocking” or nonrealistic representa- clopedia Brittanica: Micropedia (1994) entry, “Deformation tions) in early twentieth-century modernist art.—Pi- and Flow,” cited in the text above. Still, its absence from casso, Braque, and others. Poggioli initially played up the OED helps to confirm our sense that “deformed” the negative connotations of these things in the lan- has stronger negative connotations, a situation that we euage of their critics (“the principle of dehumaniza- do not believe is true for “deformation.” tion,” etc.) but soon turned to something of an apologia: “Avant-garde deformation . . . also becomes a tradition and a stylistic convention.”)

616 Elements of Sonata Theory

contrary, a sonata is a musical utterance that is a “thing,” a self-realizing verb, unspooling itself set into dialogue with generic options that are in time, not a static noun. themselves taken socially to be sonata-defining Nonetheless, once a compositional (modular) (establishing the guidelines for composition and decision has been made (via notation or perforreception under the categories of the genre). In mance in the case of a sonata), it is now part of this dialogic process the flash-point “now” of the piece’s history, fixed in place, unalterable. the unfolding structure moves progressively This is because chronological time is irreversible through various action-spaces. Within a sonata, on the work’s acoustic surface: one cannot take what we regard as an action-space—a flexible back in reality what has already been sounded.!!

concept—can be quite small (P, TR, MC, S, From this perspective, the already sounded 1s C), rather broad (exposition, development, re- converted into fixed units of modular space. In capitulation, coda), or very broad indeed (sin- chapter 2 we compared the process of composigle movement, multimovement plan). The dia- tion to that of constructing a sonic bridge over logue inherent in any sonata form may include ever larger stretches of otherwise empty time or the occasional stretching or overriding of the to the laying-down of “one appropriately styloptions on offer from the genre. These decisions ized musical tile after another’ —in which each bring those moments into the realm of what we tile, more fundamentally, was also described as a call deformation. But how can we characterize “space of action.” In this way decisions regarding more carefully what it is that is being subjected the filling of offered spans of time lay out real-

to a deformation? ized spaces, which in turn are analogous to con-

Each action-space of the sonata is generically ceptual spaces. The result is an individualized present to make possible an ongoing dialogue “shape”: an array of modules that has produced of compositional decisions with a background a fixed musical idea in time. The “shape” proconstellation of standard or traditional options duced is metaphorically analogous to a “shaped” (norms). This is nothing less than what it means vessel or container: one can perceive the musical to work within a genre—any genre—one that result sculpturally, in terms of how it realizes furnishes an ongoing horizon of expectations melodic pattern, meter, tempo, articulation, dyfor the receiver. All genres (indeed, all familar namics, timbre, density, drive-to-cadence, and actions) involve systems of norms and guide- the like. One might regard music as sculpted lines, typical and expected procedures. In the time, as a temporal sculpture in sound. case of music these are grounded in increments All genres of music presuppose genre-deof clapsing time. We are inescapably involved fining guidelines for the production of typical with the results of laying down compositional or more or less standardized “shapes” (modular decision after decision, module after module, arrays in each of the available action-zones). In in time, presumably with a larger purpose or the case of sonata form, with all of its complexigrander coherence in view all the while. The ties and possibilities, these guidelines are maniresult is a temporal process of ongoing dia- fold and varied. Within the sonata we have not logue—successive modular decisions that invite merely one or two but numerous standard prous to understand them, one by one (and then cedures available, which in turn means that the conceptually joined together in groups or clus- expected contours or “energy-shapes” of any ters), according to the guidelines of a backdrop individual work are supple in their realizations. of a set of implied norms for the genre, which (There is no single standard “shape” for expothe reception community is assumed to share. sitions, for instance, nor for its internal zones, Since the basic, initial process here is temporal, P, TR, S, and C, although in a more general the fundamental concept is that of a process, not sense one is invited to recognize any individ-

11. As pointed out elsewhere in this book, however, in nying, repeating, re-experiencing, and so on) is typimusic the fictive artifice of “psychological” time (sug- cally counterpointed against the inescapable “reality” of

gesting the possibility of backing up, recapturing, de- chronological time.

Terminology: “Rotation” and “Deformation” 617

ual exemplar of the possibilities when one hears likely to be sidelined by historical consensus as

it.) unimaginative, composition-by-the-numbers, a Under these circumstances, it is composition- boiler-plate product. This means that in the case ally possible—and was even doubtless encour- of sonata form—and certainly in the hands of aged—to submit such generically received, stan- classical masters—it was perfectly “normative” dardized “energy-shapes” to significant strain, to intersperse into the individual work instances stretching, and overriding. The term deformation of the “non-normative” or the rivetingly deforrefers to such situations. Musical deformations mational. Within the artifice of art the concept are purposeful distortions of the standardized of the “non-normative” or “nonconforming” is “action- or texture-shapes” on offer to the com- housed under a broader concept of what one is poser from the ordered complex of pre-exist- generically prepared to accept as standard proing generic expectations and traditional proce- cedure. Simply put, what is “non-normative” dures. Structural deformations are the results of | on one level of understanding becomes “normaapplications of compositional tension and force tive” under a wider span of consideration.!% to produce a surprising, tension-provoking, or In this more expansive sense, instances of aesengaging result. More to the point, on both thetic deformation are indications of normality the production and reception side of things, as within strong works of art. It is both historically part of the compositional “game” it was expected | inaccurate and simple-minded to understand (“normative”) that, within the then-current deformations as ipso facto violating a fundamenboundaries of taste and decorum, a composer tal premise of the genre at hand or introducwould apply conceptual force here and there ing illicitly foreign, unpleasant, or moralistically to strain or alter what is otherwise a bland or tainted elements of the “abnormal” or “disfigneutral set of conventional options and proce- ured” into an originally idyllic, positive model. dures—mere starting-points for the mature and Indeed, the reverse is true. Deformation, strain, experienced artist. As has been observed over and conceptual distortion were standard stratethe decades by virtually all commentators on gies within the sonata game, which was played the sonata repertory, applying such forces and increasingly under the auspices of a growing purposeful generic “misshapings” is just what demand for originality and apparent “depth” can give a composition personality, memorabil- of the compositional idea. (What would have

ity, appeal, interest, expressive power. been aesthetically “abnormal,” if not amateurThis is a crucial point. Deformations are ish, would be to shy away from all signs of them compositional surprises, engaging forays into altogether.) the unanticipated. But the paradox of art is that Within the sonata structures of the period the nature of the game at hand also and always in question “progressive” connoisseurs have includes the idea that we are to expect the unex- typically taken such thematic or structural depected. If deviations from the merely expected formations—in Johann Stamitz, C. P. E. Bach, never happen within an individual work, that is Haydn, Mozart, Beethoven, and so on—as signs no sign of aesthetic health or integrity.!? On the of creative vigor, not of any debilitating negacontrary, if expressively charged stretchings or _ tive. The distortions and localized reshapings transgressions of standardized shapes and pro- of these composers have historically been ascedures are not present at all, the work 1s more sessed as positive features, marks of original-

12. Nor, of course, do health and integrity (compo- 13. See also the final subsection of chapter 15, “The sitional “excellence” or community “legitimacy’’) lie Role of the Listener,” especially n. 47, which references only in the deformational moments. The central thing Wolfgang Iser (The Act of Reading, p. 112) on the role is the clever interplay—the dialogue—between a work’s of the reader in expecting the unexpected within the adherence to and departures from conventional expec- plots of novels. tation. Both have their roles to play, and the genre qua genre has things that it (socially) “wants to say” through the vagaries of the individual work. See appendix 1.

618 Elements of Sonata Theory

ity and personal voice. With the exception of | misses the complexity of what the term, with all a few grumbling caveats from waning sectors of its purposely implied connotations of strain of “old-world” traditionalists circa 1780-1820, and distortion, seeks to convey. Above all, such they have not been regarded as off-putting dis- an interpretation bypasses the crucial distincfigurements or “disabilities” to be contrasted tion, central to the philosophy of art for centuwith some tacitly posited concept of the sup- ries, between life and art. As Aristotle noted in

posedly “normal” or “well formed” of how the Poetics with regard to the effects of staged exemplars of the genre “ought” to proceed.!4 tragedy, what would displease us in life—terSuch a viewpoint—which would see in struc- rors and sorrows, violence and tears, brutal and tural deformations and innovative procedures unhappy outcomes—can be profoundly moving only implications of the exceeding of proper or in the displaced realm of art. We often savor and socially acceptable limits, only transgressions of | applaud in art what we do not in life. Judgments good taste—s historically associated principally that we might make in life-situations are not with social and aesthetic conservatives wield- properly transferable to the world of distanced ing their eroding claims of authority to cling to artifice, to the world of artificiality that is the the way things once were (or were imagined to most basic presupposition of the art-situation. be). Nonetheless, what becomes clear, especially Similarly, terminology that can carry nega-

as one moves further into the nineteenth cen- tive connotations when applied to assessments tury, 1s that a primarily technical local strategy in life can carry neutral or positive ones when within an individual piece—what we are calling applied to the very different situation of art. a structural deformation (Witz and originality Potential negatives in life can be reversed into in various degrees of strength)—could be seized positives in the reception-worlds of their metaupon by both proponents and opponents of any morphosed analogues within art. It may be for “new” or “developing” art form as evidence of reasons along these lines that musical distortions either, on the one hand, a brilliant display of — or intentional “misshapings” of a generically rebreathtaking creativity or, on the other, a lapse ceived action- and texture-space within music of compositional judgment from a composer of — (“deformations”) have so often been hailed as

questionable taste or talent.!° attributes of genius and originality—indicaFor all of these reasons one should not call tors of aesthetic seriousness and pleasure. Our attention to only the potentially negative slip- term “deformation,” with its charged edginess pages of the word “deformation” or conflate and flavor of aesthetic risk, seeks to convey this them inappropriately with concepts of defor- richer, more complex world of connotation. mity or disability. Such a one-sided view, in his-

torical terms, promotes the blinkered views of aesthetic and social reactionaries. Moreover, it

14. This claim has been made recently, and mistakenly pin, Schumann, Liszt, Wagner, Tchaikovsky, Strauss, attributed to us as a latent implication within the terms Mahler, and so on. To be sure, each of these composers “norm” and “deformation,” by Joseph N. Straus, “Nor- had their legions of champions. On the other hand, in malizing the Abnormal: Disability in Music and Music varying degrees, some of the works of these composers Theory,” Journal of the American Musicological Society 59 came to be placed under a quasi-moralistic suspicion (2006), forthcoming. In our view the arbitrary and ex- by conservative, masculinist, or aggressively nationclusive binaries driving Straus’s argument—the catego- alist (sometimes racialist) critics as deviant, dilettantries that he offered his readers were limited to only “well ish, socially undesirable, decadent, or degenerate. As formed” vs. “deformed” and “normal” vs. “abnormal” strategies within an aesthetic genre, structural-defor—are false choices, too narrowly drawn to engage the mational strategies may be “morally neutral” in them-

complexities of the topics at hand. selves, but history has also shown us—often much to 15. This obviously touches on the historical-recep- our consternation—that they may also be seized upon tion issues surrounding certain later, more “extreme” and denounced by cultural critics for their own cultural

or flamboyant rhetorical- or structural-deformational purposes. procedures—those found in Schubert, Berlioz, Cho-

Terminology: “Rotation” and “Deformation” 619

Precedents in the Use of the Term by norms—the application of force that would

. . « aeconsidered, subject it to or reshaping Strictly thedistortion term “deformation J . The . would poeact be .is. not . the creative will of the composer.!” new with us,producing although prior to our use . Lg. of . composition, something in _dialogue

of it, it has not been much applied to musi. . . could . . withadapted norms as they are socially received, cal. _. works. We have it primarily from ,certain ”a, . ; be understood as producing a “flow” in leading ofthose twentieth-century literary ._ _— . . .traditions portions norms—perhaps a straining, criticism (Russianof formalism, narrative theory, . . . a stretching, an expansion, or a bulging-out;

genre theory, .reader-response theory) but also, _. . . perhaps the omission of an important procemetaphorically, from its usages in physics and i. unexpected 4: . dure; perhaps the substitution of an mechanics.. event Within latterone. fields thecase term 1s . for anthe expected InEncyclopedia the of musical emphatically technical. The New . ; . norms the composer does not alter the genre itBritannica: 15th ed., vol. 3Existing (Chicago, ... . . for selfMicropedia, through such deformations. outside 1994), instance, contains an entry under ., «Deformation . , the composition proper, the genre 1s that which and Flow,” one that even uses the , aoa. , oe 3 provides the guidelines for understanding what

verb “to deform” neutrally: i. oye . occurs inside the individual piece.

. . . Within the humanities the alteration notion of; ,expresIn physics, [deformation referssivetoalterations the] ; applied to a conventional model

in shape or size of a body under the influence of hich] ; < famil;

mechanical forces. Flow is a change in deforma- (in our case, a highly complex one) 1s amit a tion that continues as long as the force 1s applied. Here one encounters the term “deformation ... Under normal conditions . . . solids deform with some regularity, often as a technical term when they are subjected to forces. Most solids 1n1- without a judgmentally negative connotation. tially deform elastically; that is to say, they return In its individual appearances the term has signifto their original shape when the load is removed. icant, sometimes broad intersections with our ... Eventually, plastic flow will come to an end: use of it. Those appearances help to illuminate deformation will ultimately tend to concentrate the history behind our selection of the term.

tile materials No astica u also Oaqd . cc . ) “mee oO Ny Ca "to always total. It is true18 thatTun“deformation

i. one area, Win we ore ae ty nae But in the literature those intersections are not

damental to their usefulnesshasin engineering. ; ; sometimes been used to describe effects (or

oye , , to address issues)ofthat are different from those Here»“deformation” is descriptive aincertain . _—Theory. . that we have in mind Sonata It can state of a solid of shape, a de- . of these usages. . . beobject—a helpful change to remind ourselves

parture. .from original, normal, or customary . Someits derive from the early-twentieth-censtate resulting from the application of force.!¢ -and . . refer ; . , tury tradition of Russian formalism No. judgment is made the deformation in , kc , . . especially to the that “enstrangement” or “defamulquestion (say, the bending of a steel bar, which ekwithin gy . oy.po. . . . larization” of ordinary language

might be a desirable end) is a negative disfig-

that- extends to the very defiurement or the result of .aetry—a marringconcern of the way .

, . nition of art itself.

that the object in question ought really to be. If,

metaphorically, wish to imagine the genre . . P y ; 6 The basicwe concepts of formalism—“transrational

of dial sonata form (or any portion thereof, such as — og a EEC lid object language,” “deautomatization,” “deformation,

a MECCTAL CACSUTA OF An A ) as a solid objec “deliberately difficult form” ...and others—are such things are at least solid concepts governed merely negations corresponding to various indices

16. Cf. n. 7, definition 3. marily to extreme strainings of the norms, or to their 17. If the scientific metaphor were applied strictly to abandonment altogether—a situation sometimes similar musical composition, every creative shaping of the ge- to the “breaking of plastic flow” in physics. One reason neric norms, however minimal, would be considered a for this is that aesthetic norms are flexible concepts, redeformation. Such a metaphor could be sustained, but maining essentially “themselves” even while permitting we prefer to use the term “deformation” to refer pri- much variation in their realizations.

620 Elements of Sonata Theory

of practical, communicative language. (Mikhail But the perception of those of a more conservaBakhtin, 1928 [critical of Russian formalism])!® tive persuasion continues to be determined by the Dynamic form is not generated by means of old canons; they will accordingly interpret any combination or merger (the often-used concept deformation of these canons by a new movement of “correspondence’’), but by means of interac- as a rejection of the principle of verisimilitude, as tion, and, consequently, the pushing forward of a deviation from realism. . . . [Thus in the “realone group of factors at the expense of another. In ism” debate one needs to consider the basic issue:] so doing, the advanced factor deforms the subor- The tendency to deform given artistic norms conceived dinate ones. The sensation of form 1s always the as an approximation of reality. . . . [Some anti-realsensation of the flow (and, consequently of the ists may take as an artistic principle:| I rebel against alteration) of correlation between the subordinat- a given artistic code and view its deformation as a more

ing, constructive factor and the subordinated fac- accurate rendition of reality .. . [while critics may tors.... Art lives by means of this interaction and argue:|] I am conservative and view the deformation of struggle. Without this sensation of subordination the artistic code, to which I subscribe, as a distortion of and deformation of all factors by the one factor reality. .. . The conservative, of course, fails to playing the constructive role, there is no fact of recognize the self-sufficient value of deformation.

art.” (Yuri Tynianov, 1924) !° (Roman Jakobson, 1921)?!

Or, as explained in an overview of suchthought — A simular position vis-a-vis the “art” question by two more recent writers (and extended here has been taken in reader-response criticism, for to include the idea of narrative, which brings us instance, in the work of Wolfgang Iser (tak-

close to Sonata Theory concerns): ing off, here, from a related idea that he wishes to adapt and improve upon, E. H. Gombrich’s The many Formalist studies in this tradition [plot well-known principle of “schema and correcstudies, narratology] describe how narratives are tion”).2? In this case Iser’s remarks, both in their “made” by “deforming” everyday narrative much content and in their general tone, are close 1nas poetry 1s “made” by deforming everyday lan- deed—virtually identical—to the concepts that guage. They developed an arsenal of techniques underpin our sense of deformation: and concepts that are by now familiar: fabula, siu-

zhet, repenon, parallelism, morphology. substi- Thus [in Gombrich’s view] the act of represen-

tution, motivation, and baring the device. (Gary os Saul Morson and Caryl Emerson, 1990)29 tation is seen as a continual process of modify-

ing traditional schemata, the correction of which provides an ever more “suitable” representation

In the 1920s Roman Jakobson applied the term of the world. .. . What is important for our pur“deformation” to the larger issue of the validity poses, however, is the fact that the correction vioof anti- or nonrealistic representation. His argu- lates a norm of expectation contained within the ment on behalf of the “deformation” of “artistic picture itself. In this way, the act of representanorms” also comes close to serving as a clearer tion creates its own conditions of reception. . . . precedent for the (perhaps more limited) way in [The observer of a work of visual art] is guided

which we use the term: by the correction to the extent that he will try

18. P. N. Medvedev/M. M. Bakhtin, The Formal Method 21. Jakobson, “On Realism in Art” [orig. 1921] in Readin Literary Scholarship: A Critical Introduction to Sociological ings in Russian Poetics: Formalist and Structuralist Views, ed.

Poetics [1928], trans. Albert J. Wehrle (Baltimore: Johns Ladislav Matejka and Krystyna Pomorska (Ann Arbor: Hopkins University Press, 1978), p. 87. Cf. pp. 89, 97. Michigan Slavic Publications, 1978), pp. 41, 43. 19. Tynianov, The Problem of Verse Language [orig. 1924], 22. E.g.,in Gombrich, Art and Illusion: A Study of Pic-

ed. and trans. Michael Sosa and Brent Harvey (Ann Ar- torial Representation (Oxford: Phaidon, 1960; 5th ed.

bor, Mich.: Ardis, 1981), p. 33. 1977); and The Image and the Eye (Ithaca, N.Y.: Cornell 20. Morson and Emerson, Mikhail Bakhtin: Creation of University Press, 1982). a Prosaics (Stanford: Stanford University Press, 1990), p. 19.

Terminology: “Rotation” and “Deformation” 621

to discover the motive behind the change in the What is more, this deviation may come into

schema. play on every level, in relation to the types, the It is in this sense that the concepts of schema genres, even to the formal principle of concordant and correction have a heuristic value as regards discordance. The first type of deviation [“types”], the strategies of literary texts. . . . Herein lies the it would seem, is constitutive of every individual particular function of the literary schemata [“the work. Each work stands apart from every other

repertoire of social norms and literary conven- work. .. . Rule-governed deformation constitions” ]|—in themselves they are elements of the tutes the axis around which the various changes text, and yet they are neither aspect nor part of of paradigm through application are arranged. It the aesthetic object. The aesthetic object signal- is this variety of applications that confers a hisizes its presence through the deformations of the tory on the productive imagination and that, in schemata, and the reader, in recognizing these de- counterpoint to sedimentation, makes a narrative formations, 1s stimulated into giving the aesthetic tradition possible. (Ricoeur, 1983)?4 object its shape. . . . It is here that the strategies

play their part, in laying down the lines along Sonata Theory, too, is concerned with “rulewhich the imagination is to run. (Iser, 1976)” governed deformation,” and, with Iser, we af-

. firm that the text’s “reader’—in our case the Also close to our use of the term is a passage listener to the composition, the analyst, the in-

from Paul Ricocur. terpreter—needs to be familiar both with so-

. . . nata norms and with the standard principles and

Innovation remains a form of behavior governed strategies of their deformation. As for the term by rules. The labor of Mag Mayon 1s Nt born “deformation,” it has both a solid and an hon-

from nothing. It is bound in one way or another ; a: ; ;

to the tradition’s paradigms. But the range of so- orable history within several disparate fields in lutions of vast. It is deployed between the two the twentieth century. For us, no substitute for poles of servile application and calculated devia- it “transformation”? “alteration”? “variant’’?) tion, passing through every degree of “rule-gov- carries as historically rich—or, more impor-

erned deformation” .... tantly, as proper—a connotation.

23. Iser, The Act of Reading: A Theory of Aesthetic Re- 24. Ricoeur, Time and Narrative [orig. 1983], trans. sponse [orig. 1976] (Baltimore: Johns Hopkins Univer- Kathleen McLaughlin and David Pellauer (Chicago:

sity Press, 1978), pp. 91-92. University of Chicago Press, 1984), 1:69—70.

This page intentionally left blank

Bibliography

Adorno, Theodor W. Beethoven: The Philosophy of the In Aspects of Schenkerian Theory, ed. Beach, 1Music, ed. Rolf Tiedemann; trans. Edmund Jeph- 38. New Haven, Conn.: Yale University Press,

cott. Stanford: Stanford University Press, 1998. 1983. —. Mahler: A Musical Physiognomy, trans. Ed- ——.. “Schubert’s Experiments with Sonata Form: mund Jephcott. Chicago: University of Chicago Formal-Tonal Design versus Underlying Struc-

Press, 1991. ture.” Music Theory Spectrum 15 (1993): 1-18.

Adrian, Jack. “The Function of the Apparent Tonic Bent, Ian, with William Drabkin. Analysis. The at the Beginning of Development Sections.” Inté- Norton/Grove Handbooks in Music. New York:

gral 5 (1991): 1-53. Norton, 1987.

Agawu, V. Kofi. Playing with Signs: A Semiotic Interpre- Berger, Karol. “The First-Movement Punctuation tation of Classic Music. Princeton, N.J.: Princeton Form in Mozart’s Piano Concertos.” In Mozart’s

University Press, 1991. Piano Concertos: Text, Context, Interpretation, ed. Aldwell, Edward and Carl Schachter. Harmony and Neal Zaslaw, 239-59. Ann Arbor: University of Voice Leading. 2nd ed. New York: Harcourt Brace Michigan Press, 1996.

Jovanovich, 1989. —. “Toward a History of Hearing: The Clas-

——.. “The Ternary-Sonata Form.” Journal of Music sic Concerto, A Sample Case.” In Convention in

Theory 34 (1991): 57-80. Eighteenth- and Nineteenth-Century Music: Essays Anson-Cartwright, Mark. “Chromatic Features of in honor of Leonard G. Ratner, ed. Wye J. AllanEb-Major Works of the Classical Period.” Music brook, Janet M. Levy, and William Mahrt, 405-—

Theory Spectrum 22 (2000): 177-204. 29. Stuyvesant, N.Y.: Pendragon, 1992. Badura-Skoda, Eva and Paul. Interpreting Mozart on Bonds, Mark Evan. “Haydn’s False Recapitulations

the Keyboard, trans. Leo Black. New York and and the Perception of Sonata Form in the Eigh-

London: Barrie & Rockliffe, 1962. teenth Century.” Ph.D. diss., Harvard University, Badura-Skoda, Paul. Kadenzen, Eingdange und Auszier- 1988. ungen zu Klavierkonzerten von Wolfgang Amadeus ——.. “The Paradox of Musical Form.” In Wordless

Mozart. Kassel: Barenreiter, 1967. Rhetoric: Musical Form and the Metaphor of the OraBarford, Philip T. “The Sonata-Principle: A Study tion, 13-52. Cambridge, Mass.: Harvard Univerof Musical Thought in the Eighteenth Century.” sity Press, 1991.

The Music Review 13 (1952): 255-63. Brown, A. Peter. Joseph Haydn’s Keyboard Music: Beach, David. “A Recurring Pattern in Mozart’s Sources and Style. Bloomington: Indiana UniverMusic.” Journal of Music Theory 27 (1983): 1-29. sity Press, 1986. —.. “Schenker’s Theories: A Pedagogical View” —.. Review of Danckwardt, Die langsame Einlei623

624 Bibliography tung: ihre Herkunft und ihr Bau bei Haydn und Mo- Cavett-Dunsby, Esther. “Mozart’s Codas.” Music zart. Journal of the American Musicological Society 33 Analysis 7 (1988): 31-51.

(1980): 200-4. ——. “Mozart’s ‘Haydn’ Quartets: Composing Up

Broyles Michael. “Organic Form and the Binary Re- and Down without Rules.” Journal of the Royal peat.” The Musical Quarterly 66 (1980): 339-60. Musical Association 113 (1988): 57-80. —. “The Two Instrumental Styles of Classi- Churgin Bathia. “Francesco Galeazzi’s Description cism.” Journal of the American Musicological Society (1796) of Sonata Form.” Journal of the American

36 (1983): 210-42. Musicological Society 21 (1968): 181-99.

Boyd, Malcolm. Domenico Scarlatti: Master of Music. —.. “Harmonic and Tonal Instability in the Sec-

New York: Schirmer, 1986. ond Key Area of Classic Sonata Form.” In ConBrodbeck, David. “Medium and Meaning: New As- vention in Eighteenth- and Nineteenth-Century Music: pects of the Chamber Music.” In The Cambridge Essays in Honor of Leonard G. Ratner, ed. Wye J. AlCompanion to Brahms, ed. Michael Musgrave, 98— lanbrook, Janet M. Levy, and William P. Mahrt,

132. Cambridge: Cambridge University Press, 23-57. Stuyvesant, N.Y.: Pendragon, 1992.

1999, —. “The Symphony as Described by J. A. P.

Brown, A. Peter. Joseph Haydn’s Keyboard Music: Schulz: A Commentary and Translation.” CurSources and Style. Bloomington: Indiana Univer- rent Musicology 29 (1980): 7-16.

sity Press, 1986. Chusid, Martin. “Schubert’s Chamber Music: Be-

Briick, Marion. Die langsamen Sdatze in Mozarts Kla- fore and after Beethoven.” In The Cambridge vierkonzerten: Untersuchungen zur Form und zum Companion to Schubert, ed. Christopher H. Gibbs,

musikalischen Satz. Munich: Fink, 1994. 174-92. Cambridge: Cambridge University Press, Burnett, Henry and Shaugn O’Donnell. “Linear 1997. Ordering of the Chromatic Aggregate in Classi- Cohn, Richard L. “As Wonderful as Star Clusters: cal Symphonic Music.” Music Theory Spectrum 18 Instruments for Gazing at Tonality in Schubert.”

(1996): 22-50. 19th-Century Music 22 (1999): 213-32.

Burnham, Scott. “A.B. Marx and the Gendering —.. “Maximally Smooth Cycles, Hexatonic Sysof Sonata Form.” In Music Theory in the Age of tems, and the Analysis of Late-Romantic ProRomanticism, ed. Ian Bent, 163-86. Cambridge: gressions.” Music Analysis 15 (1996): 9-40.

Cambridge University Press, 1996. Cole, Malcolm S. “Czerny’s Illustrated Description —.. Beethoven Hero. Princeton, N.J.: Princeton of the Rondo or Finale.” The Music Review 36

University Press, 1995. (1975): 5-16.

Bushler, David. “Harmonic Structure in Mozart’s —.. “Haydn’s Symphonic Rondo Finales: Their Sonata-Form Developments.” Mozart-Jahrbuch Structural and Stylistic Evolution.” Haydn Year1984/85, 15-24. Kassel: Birenreiter, 1986. book XIII 1982, ed. H. C. Robbins Landon, I. M. Cadwallader, Allen and David Gagné. Analysis of Bruce, and David Wyn Jones, 113-42. Cardiff: Tonal Music: A Schenkerian Approach. New York: University College Cardiff Press, 1983.

Oxford University Press, 1998. —.. “Sonata-Rondo: The Formulation of a TheCaplin, William E. “The Classical Cadence: Con- oretical Concept in the 18th and 19th Centuries.” ceptions and Misconceptions.” Journal of the Amer- The Musical Quarterly 55 (1969): 180-92.

ican Musicological Society 57 (2004): 51-117. —.. “Rondo.” The New Grove Dictionary of Mu——. Classical Form: A Theory of Formal Functions sic and Musicians. 2nd edition, ed. Stanley Sadie for the Instrumental Music of Haydn, Mozart, and and John Tyrell, 21: 649-56. London: MacmilBeethoven. New York: Oxford University Press, lan, 2001.

1998. —.. “Rondos, Proper and Improper.” Music &

——.. “The ‘Expanded Cadential Progression’: A Letters 51 (1970): 388-99. Category for the Analysis of Classical Form.” Cone, Edward T. Musical Form and Musical Perfor-

Journal of Musicological Research 7 (1987): 215-57. mance. New York: Norton, 1968. ——. “Hybrid Themes: Toward a Refinement Czerny, Carl. School of Practical Composition, trans. in the Classification of Classical Theme Types.” John Bishop. 3 vols, ca. 1848. Reprint, New York:

Beethoven Forum 3 (1994): 151-65. Da Capo, 1979. —.. “Structural Expansion in Beethoven’s Sym- Dahlhaus, Carl. Ludwig van Beethoven: Approaches to phonic Forms.” In Beethoven’s Compositional Pro- His Music, trans. Mary Whittall. Oxford: Claren-

cess, ed. William Kinderman, 27-54. Lincoln: don Press, 1991.

University of Nebraska Press, 1991. —. Nineteenth-Century Music, trans. J. Bradford

Bibliography 625 Robinson. Berkeley: University of California Mozart’s Piano Concertos: Text, Context, Interpreta-

Press, 1989. tion, ed. Neal Zaslaw, 149-86. Ann Arbor: Uni-

Danckwardt, Marianne. Die langsame Einleitung: ihre versity of Michigan Press, 1996. Herkunft und ihr Bau bei Haydn und Mozart. 2 vols. Fillion, Michelle. “Sonata Exposition Procedures in

Tutzing: Hans Schneider, 1977. Haydn’s Keyboard Sonatas.” In Haydn Studies. ProDarcy, Warren. “Bruckner’s Sonata Deformations.” ceedings of the International Haydn Congress, WashIn Bruckner Studies, ed. Timothy L. Jackson and ington, D.C., 1975, ed. Jens Peter Larsen, Howard Paul Hawkshaw, 256-77. Cambridge: Cambridge Serwer, and James Webster, 475-81. New York:

University Press, 1997. Norton, 1981.

——.. “The Metaphysics of Annihilation: Wagner, Fisher, Stephen C. “Further Thoughts on Haydn’s Schopenhauer, and the Ending of the Ring.” Mu- Symphonic Rondo Finales.” Haydn Yearbook XVII

sic Theory Spectrum, 16 (1994): 1-40. [1992], ed. H. C. Robbins Landon, Otto Biba, —. Review of William E. Caplin, Classical Form. I. M. Bruce, and David Wyn Jones, 85-107. Music Theory Spectrum 22 (2000): 122-25. Eisenstadt: Joseph Haydn Stiftung, 1992. —.. “Rotational Form, Teleological Genesis, ———. “Haydn’s Overtures and Their Adaptations and Fantasy-Projection in the Slow Movement of as Concert Orchestral Works.” Ph.D. diss., UniMahler’s Sixth Symphony.” 19th-Century Music 25 versity of Pennsylvania, 1985.

(2001): 49-74. —.. “Sonata Procedures in Haydn’s Symphonic

Daverio, John. “From ‘Concertante Rondo’ to ‘Lyric Rondo Finales of the 1770s.” In Haydn Studies: Sonata’: A Commentary on Brahms’s Reception Proceedings of the International Haydn Conference of Mozart.” In Brahms Studies, vol. 1, ed. David Washington, D.C., 1975, ed. Jens Peter Larsen, Brodbeck, 111-36. Lincoln and London: Univer- Howard Serwer, and James Webster, 481-87.

sity of Nebraska Press, 1994. New York and London: Norton, 1981.

Davis, Shelley. “C. P. E. Bach and the Early History Flothuis, Marius. Mozart’s Piano Concertos: A Study.

of the Recapitulatory Tutti in North Germany.” Amsterdam/Atlanta: Rodopi, 2001. In C. P. E. Bach Studies, ed. Stephen L. Clark, Forman, Denis. Mozart’s Concerto Form: The First 65-82. Oxford: Clarendon Press, 1988. Movements of the Piano Concertos. London: Rupert

—. “H. C. Koch, the Classic Concerto, and the Hart-Davis, 1971. Sonata-Form Retransition.” Journal of Musicology Forster, Robert. Die Kopfsatze der Klavierkonzerte Mo-

2 (1983): 45-61. zarts und Beethovens: Gesamtaufbau, Solokadenz und Drabkin, William. “Beethoven’s Understanding of Schlufbildung. Munich: Fink, 1992. ‘Sonata Form’: The Evidence of the Sketchbooks” —.. “Zur Funktion von Anfangsritornell und ReIn Beethoven’s Compositional Process, ed. William prise in den Kopfsadtzen einiger Klavierkonzerte Kinderman, 14-19. Lincoln: University of Ne- Mozarts.” Mozart-Jahrbuch 1986, 74-89. Kassel:

braska Press, 1991. Birenreiter, 1987.

Dunsby, Jonathan. “The Formal Repeat.” Journal of Gagné, David. “The Compositional Use of Register the Royal Musical Association 112/2 (1987): 196— in Three Piano Sonatas by Mozart.” In Trends in

207. Schenkerian Research, ed. Allen Cadwallader, 23-—

Einstein, Alfred. Mozart: His Character, His Work. 39. New York: Schirmer Books, 1990.

Oxford: Oxford University Press, 1945. Galand, Joel. “Form, Genre, and Style in the EighEisen, Cliff. “The Classical Period” (subsection of teenth-Century Rondo.” Music Theory Spectrum “Concerto”). The New Grove Dictionary of Music 17 (1995): 27-52. and Musicians. 2nd ed., ed. Stanley Sadie and John —.. “The Large-Scale Formal Role of the Solo En-

Tyrell, 6: 246-51. London: Macmillan, 2001. try Theme in the Eighteenth-Century Concerto.” Engel, Hans. Die Entwicklung des deutschen Kla- Journal of Music Theory 44 (2000): 381-450. vierkonzerts von Mozart bis Liszt. Leipzig: Breitkopf —.. “Rondo-Form Problems 1n Eighteenth- and

& Hartel, 1927. Nineteenth-Century Instrumental Music, with —.. Das Instrumentalkonzert. Wiesbaden: Breit- Reference to the Application of Schenker’s Form

kopf & Hartel, 1974. Theory to Historical Context.” Ph.D. diss., Yale

Everett, Paul. “Vivaldi: The Four Seasons and Other University, 1990. Concertos, op. 8. Cambridge: Cambridge Univer- Galeazzi, Francesco. Elementi teorico-pratici di musica.

sity Press, 1996. Vol. 2. Rome: Puccinelli, 1796.

Feldman, Martha. “Staging the Virtuoso: Ritornello Gossett, Philip. “The Overtures of Rossini.” 19thProcedure in Mozart, from Aria to Concerto.” In Century Music 3 (1979): 3-31.

626 Bibliography

Gratzer, Wolfgang. “Mozart, oder? Der Unisono- —.. “Rotations, Sketches, and [Sibelius’s] Sixth Beginn in Streichquartetten der Wiener Klas- Symphony.” In Sibelius Studies, ed. Timothy L. sik: Fragment zu einer Poetik des musikalischen Jackson and Veijo Murtomiaki, 322-51. CamAnfangs.” Mozart-Jahrbuch 1991, 641-49. Kassel: bridge: Cambridge University Press, 2001.

Barenreiter, 1992. ——. Sibelius: Symphony No. 5. Cambridge: Cam-

Grave, Margaret G. “First-Movement Form as a bridge University Press, 1993. Measure of Dittersdorf’s Symphonic Develop- —.. “Structure, Implication, and the End of Suor ment.” Ph.D. diss. New York University, 1977. Angelica.” Studi pucciniani 3 (2004): 241-64. Grayson, David. Mozart: Piano Concertos No. 20 in Hepokoski, James, and Warren Darcy. “The Medial D Minor, K. 466, and No. 21 in C Major, K. 467. Caesura and its Role in the Eighteenth-Century Cambridge: Cambridge University Press, 1999. Sonata Exposition.” Music Theory Spectrum 19 Green Douglass M. Form in Tonal Music: An Introduc- (1997): 115-54. tion to Analysis. 2nd ed. New York: Holt, Rine- Hill, George R. “The Concert Symphonies of Flo-

hart, and Winston, 1979. rian Leopold Gassmann.” Ph.D. diss., New York

Grey, Thomas S. “Wagner, the Overture, and the University, 1975. Aesthetics of Musical Form.” 19th-Century Music ——. “Introduction” to Florian Leopold Gassmann,

12 (1988): 3-22. 1729-1774: Seven Symphonies, in The Symphony

Haimo, Ethan. “Haydn’s Altered Reprise.” Journal of 1720-1840, ed. Barry S. Brook, Series B, 10:x111—

Music Theory 32 (1988): 335-51. xxii. New York: Garland, 1981. —. Haydn’s Symphonic Forms: Essays in Composi- Hinrichsen, Hans-Joachim. “Sonatenform, Sonaten-

tional Logic. Oxford: Clarendon Press, 1995. hauptsatzform,” In Handwérterbuch der musika—.. “Remote Keys and Multi-movement Unity: lischen Terminologie, ed. Hans Heinrich Eggebrecht, Haydn in the 1790s.” The Musical Quarterly 74 1-20. Stuttgart: Steiner, c. 1972ff.

(1990): 242-68. Hoyt, Peter A. “The Concept of développement in the

Harrison, Daniel. Harmonic Function in Chromatic Mu- Early Nineteenth Century.” In Music Theory in the sic: A Renewed Dualist Theory and an Account of Its Age of Romanticism, ed. Ian Bent, 141-62. CamPrecedents. Chicago: University of Chicago Press, bridge: Cambridge University Press, 1996.

1994. —.. “Haydn’s ‘False Recapitulations,’ Late Eigh-

Hell, Helmut. Die Neapolitanische Opernsinfonie in der teenth-Century Theory, and Modern Paradigms ersten Halfte des 18. Jahrhunderts. Tutzing: Schnei- of Sonata Form.” Unpublished paper.

der, 1971. —. Review of Bonds, Wordless Rhetoric: Musical

Hepokoski, James. “Back and Forth from Egmont: Form and the Metaphor of the Oration. Journal of Mu-

Beethoven, Mozart, and the Nonresolving Reca- sic Theory 38 (1994): 123-43. pitulation.” 19th-Century Music 25 (2002): 127— Hyer, Brian. “Reimag(in)ing Riemann. Journal of

53. Music Theory 39 (1995): 101-38.

—. “Beethoven Reception: The Symphonic —. “Tonal Intuitions in Tristan und Isolde.” Ph.D. Tradition.” In The Cambridge History of Nine- diss., Yale University, 1989. teenth-Century Music, ed. Jim Samson, 424-59. Irving, John. Mozart’s Piano Concertos. Aldershot:

Cambridge: Cambridge University Press, 2002. Ashgate, 2003. —. “Beyond the Sonata Principle.” Journal of —. Mozart’s Piano Sonatas: Contexts, Sources,

154. 1997.

the American Musicological Society, 55 (2002): 91— Style. Cambridge: Cambridge University Press,

—. “The Essence of Sibelius: Creation Myths ——. Mozart: The “Haydn” Quartets. Cambridge: and Rotational Cycles in Luonnotar.” In The Sibe- Cambridge University Press, 1998. lius Companion, ed. Glenda Dawn Goss, 121-46. Ivanovitch, Roman. “The Process of Variation in the

Westport, Conn.: Greenwood, 1996. Music of Mozart.” Ph.D. diss. Yale University, —. “Framing Till Eulenspiegel.” 19th-Century 2003. Music 30 (2006): forthcoming. Jackson, Timothy L. “The Finale of Bruckner’s Sev—.. “Masculine/Feminine.” The Musical Times enth Symphony and the Tragic Reversed Sonata

135 (August 1994): 494-99. Form.” In Bruckner Studies, ed. Timothy L. Jack—.. “Ottocento Opera as Cultural Drama: Generic son and Paul Hawkshaw, 140-208. Cambridge: Mixtures in II trovatore.” In Verdi’s Middle Period Cambridge University Press, 1997. 1849-1859: Source Studies, Analysis, and Perfor- Kamien, Roger. “The Opening Sonata-Allegro mance Practice, ed. Martin Chusid, 147-96. Chi- Movements in a Randomly Selected Sample of

cago: University of Chicago Press, 1997. Solo Keyboard Sonatas Published in the Years

Bibliography 627

1742-1774 (Inclusive).” Ph.D. diss., Princeton Style, trans. Ulrich Kramer, 269-79. Ann Arbor:

University, 1964. UMI Research Press, 1988. First published as

Keefe, Simon P. Mozart’s Piano Concertos: Dramatic “Sonatenform-Probleme,” in Festschrift Friedrich Dialogue in the Age of Enlightenment. Woodbridge, Blume zum 70 Geburtstag, ed. Anna Amalie Abert

U.K.: Boydell Press, 2001. and Wilhelm Pfannkuch, 221-30. Kassel: BarenKeller, Hans. “The Chamber Music.” In The Mozart reiter, 1963. Companion, ed. H. C. Robbins Landon and Don- —. “The Symphonies.” In The Mozart Comald Mitchell, 90-137. New York: Norton, 1956. panion, ed. H. C. Robbins Landon and Donald ——. The Great Haydn Quartets: Their Interpretation. Mitchell [1956], 156-99. Reprint, New York:

London: Dent, 1986. Norton, 1969.

Kerman, Joseph. The Beethoven Quartets. New York: LaRue, Jan. Guidelines for Style Analysis. 2nd ed. War-

Norton, 1966. ren, Mich.: Harmonie Park, 1992.

—. “Beethoven’s Minority.” In Kerman, Write — Review of Charles Rosen, Sonata Forms (1st All These Down: Essays on Music, 217-37. Berke- ed., 1980). Journal of the American Musicological So-

ley: University of California Press, 1994. ciety 34 (1981): 557-66. —. “Notes on Beethoven’s Codas.” In Beethoven LaRue, Jan and Eugene K. Wolf. “Symphony. I. 18th Studies 3, ed. Alan Tyson, 141-59. Cambridge: Century.” The New Grove Dictionary of Music and

Cambridge University Press, 1982. Musicians. 2nd ed., ed. Stanley Sadie and John —.. “Tovey’s Beethoven.” In Kerman, Write All Tyrell, 24:812-33. London: Macmillan, 2001. These Down: Essays on Music, 155-72. Berkeley Leeson, Daniel N., and Robert D. Levin, “On the and Los Angeles: University of California Press, Authenticity of K.Anh C.14.01 (279b), a Sym-

1994. phonia Concertante for Four Winds and Orches-

Kinderman, William. Artaria 195: Beethoven’s Sketch- tra.” Mozart-Jahrbuch 1976/77, 70-96. Kassel: book for the Missa solemnis and the Piano Sonata in Biarenreiter, 1978. E Major, Opus 109. Urbana: University of Illinois Leichtentritt, Hugo. Musical Form. Cambridge, Mass.:

Press, 2003. Harvard University Press, 1951.

—. Beethoven. Berkeley: University of California Leister, Reiner. Das Finale in der Sinfonik Joseph

Press, 1995. Haydns. Stuttgart: ibidem-Verlag, 1999.

Kirkpatrick, Ralph. Domenico Scarlatti. Princeton, Levin, Robert D. “Das Konzert ftir Klavier und Vi-

N.J.: Princeton University Press, 1953. oline D-Dur KV Anh. 56/315f und das KlariKollmann, Augustus Frederic Christopher. An Es- nettenquintett B-Dur, KV Anh. 91/516c: Ein say on Practical Musical Composition, according to the Erginzungsversuch.” Mozart-Jahrbuch 1968/70, Nature of That Science and the Principles of the Great- 304-26. Salzburg: Internationale Stiftung Moest Musical Authors. London: Kollmann, 1799. Re- zarteum. 1970.

print, New York: Da Capo Press, 1973. Levy, Janet M. “Gesture, Form, and Syntax in Koch, Heinrich Christoph. Introductory Essay on Com- Haydn’s Music.” In Haydn Studies: Proceedings of position: The Mechanical Rules of Melody, Sections 3 the International Haydn Conference, Washington,

and 4, trans. Nancy Kovaleff Baker. New Haven, D.C., 1975, ed. Jens Peter Larsen, Howard Ser-

Conn.: Yale University Press, 1983. wer, and James Webster, 355-62. New York: —. Versuch einer Anleitung zur Composition Norton, 1981. [Leipzig: Adam Friedrich Béhme, 1793]. Re- ———. “Texture as a Sign in Classic and Early Roprint, Hildesheim: Georg Olms, 1969. mantic Music.” Journal of the American Musicological Kramer, Lawrence. “The Strange Case of Beethoven’s Society 35 (1982): 482-531. Coriolan: Romantic Aesthetics, Modern Subjec- Lockwood, Lewis. “Eroica Perspectives: Strategy and tivity, and the Cult of Shakespeare.” The Musical Design in the First Movement.” In Lockwood,

Quarterly 79 (1995): 265-80. Beethoven: Studies in the Creative Process, 118-33. Kucaba, John. “The Symphonies of Georg Christoph Cambridge, Mass.: Harvard University Press, Wagenseil.” Ph.D diss., Boston University, 1967. 1992. Kiister, Konrad. Formale Aspekte des ersten Allegros in Longyear, Rey M. “Binary Variants of Early Classic

Mozarts Konzerten. Kassel: Barenreiter, 1991. Sonata Form.” Journal of Music Theory 13 (1969):

Landon, H. C. Robbins, and David Wyn Jones. 162-83. Haydn: His Life and Music. Bloomington: Univer- —.. “Parallel Universes: Mozart’s Minor-Mode

sity of Indiana Press, 1988. Reprises.” Mozart-Jahrbuch 1991, 810-15. Kassel: Larsen, Jens Peter. “Sonata Form Problems.” In Barenreiter, 1992. Larsen, Handel, Haydn, and the Viennese Classical Longyear, Rey M., and Kate R. Covington. “Sources

628 Bibliography of the Three-Key Exposition.” Journal of Musicol- zart-Jahrbuch 1962/63, 256-79. Salzburg: Zentra-

ogy 6 (1988): 448-70. linstitut ftir Mozartforschung der Internationalen

McLamore, Laura Alyson. “Symphonic Conventions Stiftung Mozarteum, 1964. in London’s Concert Rooms, circa 1755-1790.” Pascall, Robert. “Some Special Uses of Sonata Form Ph.D. diss., University of California, Los Ange- by Brahms.” In Soundings, ed. Arnold Whittall,

les, 1991. No. 4 (1974): 58-63.

Mahling, Christoph-Hellmut. “Zur Frage der “Ein- Plantinga, Leon. Beethoven’s Concertos: History, Style, heit’ der Symphonie.” In Uber Sinfonien: Beitrige Performance. New York: Norton, 1999. zu einer musikalischen Gattung: Festschrift Walter ——. Clementi: His Life and Music. London: Oxford Wiora zum 70. Geburtstag, ed. Christoph-Hellmut University Press, 1977.

Mahling, 1-40. Tutzing: Schneider, 1979. Plath, Wolfgang. “Beitrage zur Mozart-AutograMalloch, William. “The Minuets of Haydn and phie II: Schriftchronologie, 1770-1780,” MozartMozart: Goblins or Elephants?” Early Music 21 Jahrbuch 1976/77, 131-73. Kassel: Barenreiter,

(1993): 437-45. 1978.

—.. “Toward a ‘New’ (Old) Minuet.” Opus 1/5 Powers, Harold. “Reading Mozart’s Music: Text and

(1985): 14-21, 52. Topic, Syntax and Sense.” Current Musicology 57

Marston, Nicholas. “The Recapitulation Transition (1995): 5-44. in Mozart’s Music,” Mozart-Jahrbuch 1991, 793- Ratner, Leonard G. Classic Music: Expression, Form,

809. Kassel: Barenreiter, 1992. and Style. New York: Schirmer, 1980. Marx, Adolph Bernhard. Die Lehre von der musika- —.. “Harmonic Aspects of Classic Form.” Jourlischen Komposition, praktisch-theoretisch. Vols. 2 and nal of the American Musicological Society 11 (1949):

3, Ist eds. Leipzig, 1838 and 1845. 159-68.

—. Die Lehre von der musikalischen Komposition, Ratz, Erwin. Einfiihrung in die musikalische Formenpraktisch theoretisch. 4th ed. Leipzig: Breitkopf und lehre: Uber Formprinzipien in den Inventionen und

Hartel, 1868 Fugen J. S. Bachs und ihre Bedeutung fiir die Kom-

—.. Ludwig van Beethoven: Leben und Schaffen. positionstechnik Beethovens. 3rd ed., enl. Vienna:

Berlin: Janke, 1859. Universal, 1973.

—. Musical Form in the Age of Beethoven: Selected Reicha, Anton. Traité de haute composition musicale. Writings on Theory and Method, ed. and trans. Scott Vol. 2. Paris, 1826. Burnham. Cambridge: Cambridge University Réti, Rudolph. The Thematic Process in Music. New

Press, 1997. York: Macmillan, 1951.

Mellers, Wilfrid. The Sonata Principle (from c. 1750). Ritzel, Fred. Die Entwicklung der ‘Sonatenform’ im

Fair Lawn, N.J.: Essential Books, 1957. musiktheoretischen Schrifttum des 18. and 19. JahrMomigny, Jér6me-Joseph de. Cours complet @’harmonie hunderts. Wiesbaden: Breitkopf & Hartel, 1968. et de composition. 3 vols. Paris: Momigny, 1803-— Rosen, Charles. The Classical Style: Haydn, Mozart,

6). Beethoven. Revised ed. New York: Norton, 1997.

Moosbauer, Bernhard. Tonart und Form in den Finali —. The Romantic Generation. Cambridge, Mass.: der Sinfonien von Joseph Haydn zwischen 1766 und Harvard University Press, 1995.

1774. Tutzing: Hans Schneider, 1998. —.. “Schubert’s Inflections of Classical Form.” In Morgan, Robert P. “Coda as Culmination: The First The Cambridge Companion to Schubert, ed. ChrisMovement of the ‘Eroica’ Symphony.” In Music topher H. Gibbs, 72-98. Cambridge: Cambridge Theory and the Exploration of the Past, ed. Chris- University Press, 1997.

topher Hatch and David W. Bernstein, 357-76. —.. Sonata Forms. Rev. ed. New York: Norton,

Chicago: University of Chicago Press, 1993. 1988. Moyer, Birgitte. “Concepts of Musical Form in the Rosen, David. “The Composer’s ‘Standard Oper-

Nineteenth Century with Special Reference to ating Procedure’ as Evidence of Intention: The A.B. Marx and Sonata Form.” Ph.D. diss., Stan- Case of a Formal Quirk in Mozart’s K. 595.” Jour-

ford University, 1969. nal of Musicology 5 (1987): 79-90.

Newman, William S. The Sonata in the Classic Era. —. “‘Unexpectedness’ and ‘Inevitability’ in Mo-

Revised ed. New York: Norton, 1972. zart’s Piano Concertos.” In Mozart’s Piano ConNotley, Margaret. “Late Nineteenth-Century Cham- certos: Text, Context, Interpretation, ed. Neal Zaber Music and the Cult of the Classical Adagio.” slaw, 261-84. Ann Arbor: University of Michigan

19th-Century Music 23 (1999): 33-61. Press, 1996. Palm, Albert. “Mozarts Streichquartett D-moll, Rothstein, William. Phrase Rhythm in Tonal Music. KV 421, in der Interpretation Momignys.” Mo- New York: Schirmer, 1989.

Bibliography 629 Schenker, Heinrich. Free Composition, ed. and trans. —.. “Genre, Gesture, and Meaning in Mozart’s Ernst Oster. New York: Longman, 1979. First ‘Prague’ Symphony.” In Mozart Studies 2, ed. Cliff

published as Der Freie Satz (Vienna, 1935). Eisen, 27-84. Oxford: Clarendon Press, 1997. —. Harmony, ed. and annotated by Oswald Jo- —. Haydn and the Classical Variation. Cambridge, nas; trans. Elisabeth Mann Borgese. Chicago: Mass.: Harvard University Press, 1993. University of Chicago, 1954, rpt. 1980. ——.. Mozart: The “Jupiter” Symphony. Cambridge: Schoenberg, Arnold. Fundamentals of Musical Compo- Cambridge University Press, 1993. sition, ed. Gerald Strang and Leonard Stein. Lon- Snyder, John L. “Schenker and the First Movement

don: Faber & Faber, 1967. of Mozart’s Sonata, K. 545: An Uninterrupted

Schmalfeldt, Janet. “Cadential Processes: The Evaded Sonata-Form Movement?” Theory and Practice 16

Cadence and the ‘One More Time’ Technique.” (1991): 51-78. Journal of Musicological Research 12 (1992): 1-51. Somfai, Laszl6. “The London Revision of Haydn’s

—.. “Towards a Reconciliation of Schenkerian Instrumental Style.” Proceedings of the Royal MusiConcepts with Traditional and Recent Theories cal Association 100 (1973-74): 159-74. of Form.” Music Analysis 10 (1991): 233-87. —. The Keyboard Sonatas of Joseph Haydn: InstruSchmalzriedt, Siegfried. “Durchftihren, Durchftih- ments and Performance Practice, Genres and Styles, rung” [1979]. In Handwérterbuch der musikalischen trans. Somfai and Charlotte Greenspan. Chicago: Terminologie, ed. Hans Heinrich Eggebrecht, 1-16. University of Chicago Press, 1995.

Stuttgart: Steiner, c. 1972ff. Spies, Claudio. “‘Form’ and the Tragic Overture: An —. “Reprise/ripresa (nach 1600) [1981]” In Adjuration.” In Brahms Studies: Analytical and HisHandwéorterbuch der musikalischen Terminologie, torical Perspectives, ed. George S. Bozarth, 391-98.

ed. Hans Heinrich Eggebrecht, 1-16. Stuttgart: Oxford: Clarendon, 1990.

Steiner, c. 1972ff. Stevens, Jane R. “An 18th-Century Description of Schmitz, Arnold. Beethoven’s ‘Zwei Prinzipe.’ Berlin Concerto First-Movement Form.” Journal of the

and Bonn, 1927. American Musicological Society 24 (1971): 85-95.

Schoenberg, Arnold. Fundamentals of Musical Compo- —.. “Formal Design in C. P. E. Bach’s Harpsition, ed. Gerald Strang and Leonard Stein. Lon- sichord Concertos.” Studi musicali, 15 (1986):

don: Faber & Faber, 1967. 257-97.

Schulenberg, David. The Instrumental Music of Carl —. “The Importance of C. P. E. Bach for MoPhilipp Emanuel Bach. Ann Arbor: UMI Research zart’s Piano Concertos.” In Mozart’s Piano Con-

Press, 1984. certos: Text, Context, Interpretation, ed. Neal Za-

Seidel, Wilhelm. “Die altere Zyklustheorie, tiber- slaw, 211-36. Ann Arbor: University of Michigan dacht im Blick auf Beethovens Werk. ” In Beitrége Press, 1996. zu Beethovens Kammermusik: Symposion Bonn 1984, —. “Patterns of Recapitulation in the First

ed. Sieghard Brandenburg and Helmut Loos. Movements of Mozart’s Piano Concertos.” In Verdffentlichungen des Beethovenhauses in Bonn, Musical Humanism and Its Legacy: Essays in Honor Neue Folge, 4, Reihe: Schriften zur Beethoven- of Claude V. Palisca, ed. Nancy Kovaleff Baker and

forschung, 4, 273-82. Munich: Henle, 1987. Barbara Russano Hanning, 397—418. Stuyvesant,

—.. “Schnell—Langsam—Schnell: Zur ‘klas- N.Y.: Pendragon, 1992. sischen’ Theorie des instrumentalen Zyklus.” —. “Theme, Harmony, and Texture in Classic-

Musiktheorie 1 (1986): 205-16. Romantic Descriptions of Concerto First-MoveSeiffert, Wolf-Dieter. Mozarts friihe Streichquartette. ment Form.” Journal of the American Musicological

Munich: Fink, 1992. Society 27 (1974): 25-60.

Simon, Edwin J. “Sonata into Concerto: A Study of Steinberg, Michael. “The Late Quartets.” In The Mozart’s First Seven Concertos.” Acta musicologica Beethoven Quartet Companion, ed. Robert Winter

31 (1959): 170-85. and Robert Martin, 215-82. Berkeley: Univer-

Sisman, Elaine R. “Brahms’s Slow Movements: Re- sity of California Press, 1994. inventing the ‘Closed’ Forms.” In Brahms Stud- Straus, Joseph N. “Normalizing the Abnormal: Dis-

1990. coming.

ies: Analytical and Historical Perspectives, ed. George ability in Music and Music Theory.” Journal of the

L. Bozarth, 79-103. Oxford: Clarendon Press, American Musicological Society 59 (2006): forth——. “C. P. E. Bach, Beethoven, and the Laby- Strunk, Oliver. “Haydn’s Divertimenti for Baryton, rinth of Melancholy.” Unpublished paper deliv- Viola, and Bass.” The Musical Quarterly 18 (1932): ered at the American Musicological Society, To- 216-51. Reprint, Strunk, Essays on Music in the

ronto, November 2, 2000. Western World, 126-70. New York: Norton, 1974.

630 Bibliography Sutcliffe, W. Dean. The Keyboard Sonatas of Domenico Interpretation, ed. Neal Zaslaw, 107-37. Ann ArScarlatti and Eighteenth-Century Musical Style. Cam- bor: University of Michigan Press, 1996.

bridge: Cambridge University Press, 2003. —. “Binary Variants of Sonata Form in Early —. Haydn: String Quartets, op. 50. Cambridge: Haydn Instrumental Music.” In Joseph Haydn: Cambridge University Press, 1992. Bericht iiber den Internationalen Joseph Haydn Kon—. Review of Charles Rosen, The Classical Style gress, Wien, Hofburg 5-12 September 1982, ed. (rev. ed.). Music & Letters 79 (1998): 601-4. Eva Barura-Skoda, 127-35. Munich: G. Henle, Talbot, Michael. The Finale in Western Instrumen- 1986. tal Music. New York: Oxford University Press, —.. “The General and the Particular in Brahms’s

2001. Later Sonata Forms.” In Brahms Studies: Analytical

—. “The Instrumental Concerto: Origins to and Historical Perspectives, ed. George S. Bozarth, 1750” (subsection of “Concerto”’). The New Grove 49-78. Oxford: Clarendon, 1990. Dictionary of Music and Musicians. 2nd ed., ed. —. Haydn’s “Farewell” Symphony and the Idea of Stanley Sadie and John Tyrell, 6:242—46. Lon- Classical Style: Through-Composition and Cyclic In-

don: Macmillan, 2001. tegration in His Instrumental Music. Cambridge:

Tobel, Rudolf von. Die Formenwelt der klassischen In- Cambridge University Press, 1991. strumentalmusik. Bern and Leipzig: Paul Haupt, —.. “Schubert’s Sonata Form and Brahms’ First

1935. Maturity.” 19th-Century Music 2 (1978): 18-35

Tovey, Donald Francis. “Brahms’s Chamber Music. and 3 (1979), 52-71. The Main Stream of Music and Other Essays, 220- ——. “Sonata Form.” The New Grove Dictionary of

70. Oxford: Oxford University Press, 1949. Music and Musicians. 2nd ed., ed. Stanley Sadie —. A Companion to Beethoven’s Pianoforte Sonatas. and John Tyrell, 23:687—701. London: Macmil-

London: Royal Schools of Music, 1931. lan, 2001. —.. Essays in Musical Analysis. 7 vols. London: Wheelock, Gretchen A. Haydn’s Ingenious Jesting with Oxford University Press, 1935-39. Partially re- Art: Contexts of Musical Wit and Humor. New York: printed in Symphonies and Other Orchestral Works. Schirmer, 1992.

Oxford: Oxford University Press, 1989. ——. “Schwarze Gredel and the Engendered Minor ——.. “Haydn’s Chamber Music.” The Main Stream Mode in Mozart’s Operas.” In Musicology and Dif-

of Music and Other Essays, 1-64. ference: Gender and Sexuality in Music Scholarship, ——.. “Musical Form and Matter.” The Main Stream ed. Ruth A. Solie, 201-21. Berkeley: University of Music and Other Essays, 160-82. Oxford: Ox- of California Press, 1992.

ford University Press, 1949. Whitmore, Philip. Unpremeditated Art: The Cadenza —.. “Some Aspects of Beethoven’s Art Forms.” in the Classical Keyboard Concerto. Oxford: ClarenThe Main Stream of Music and Other Essays, 271-97. don Press, 1991. Oxford: Oxford University Press, 1949. Will, Richard. The Characteristic Symphony in the Age —. “Sonata Forms.” Musical Articles from the En- of Haydn and Beethoven. Cambridge: Cambridge cyclopaedia Britannica, 208-32. London: Oxford University Press, 2002. University Press, 1944. Reissued (1956) under Winter, Robert S. “The Bifocal Close and the Evolu-

the title The Forms of Music. tion of the Viennese Classical Style.” Journal of the Tyson, Alan. “The Mozart Fragments in the Mo- American Musicological Society 42 (1989): 275-337. zarteum, Salzburg: A Preliminary Study of Their Wolf, Eugene K. “The Recapitulations in Haydn’s Chronology and Their Significance.” In Tyson. London Symphonies.” The Musical Quarterly 52 Mozart: Studies of the Autograph Scores, 125-61. (1966): 71-89.

Cambridge, Mass.: Harvard University Press, —.. “Sonata Form.” In The New Harvard Dictio-

1987. nary of Music, ed. Don Michael Randel, 764-67.

Wallace, Robin. “Background and Expression in Cambridge, Mass.: Harvard University Press, the First Movement of Beethoven’s op. 132.” The 1986. Journal of Musicology, 7 (1989): 3-20. ——. “Symphony,” The New Harvard Dictionary of Webster, James. “The Analysis of Mozart’s Arias.” Music, ed. Don Michael Randel, 822—27. CamIn Mozart Studies, ed. Cliff Eisen, 101-99. New bridge, Mass.: Harvard University Press, 1986. York: Oxford University Press, 1991. —. The Symphonies of Johann Stamitz: A Study in —. “Are Mozart’s Concertos ‘Dramatic’? Con- the Formation of the Classic Style. Utrecht: Bohn, certo Ritornellos versus Aria Introductions in the Scheltema & Holkema, 1981. 1780s.” In Mozart’s Piano Concertos: Text, Context, Wolff, Christoph. “Cadenzas and Styles of Improvi-

Bibliography 631 sation in Mozart’s Piano Concertos.” In Perspec- W.-A Mozart: Sa vie musicale et son ceuvre. 5 vols. tives on Mozart Performance, ed. R. Larry Todd and Paris: Desclée, de Brouwer et Cie., 1912-46. Peter Willams, 228-38. Cambridge: Cambridge Zaslaw, Neal, ed. Mozart’s Piano Concertos: Text,

University Press, 1991. Context, Interpretation. Ann Arbor: University of ——.. “Zur Chronologie der Klavierkonzert-Kaden- Michigan Press, 1996. zen Mozarts.” Mozart-Jahrbuch 1978-79, 235-46. ——. Mozart’s Symphonies: Context, Performance

Kassel: Barenreiter, 1979. Practice, Reception. Oxford: Oxford University Wyzewa, Théodore de, and Georges de Saint Foix. Press, 1989.

This page intentionally left blank

Index of Names

These references identify appearances of names within the text as well as expanded discussions of those persons in footnotes. Brief footnote-citation appearances are not listed. For individual composers listed here, see also the Index of Works: this index of names cites only references that are not linked in the text to specific compositions. For individual issues and subsections of sonata form listed below, see also the Index of Concepts.

Abel, Carl Friedrich, 432 Badura-Skoda, Paul, 600 n. 48, 601 Berger, Karol Adorno, Theodor W., 285 n. 18, Baker, Nancy Kovaleff, 448 n. 57 concerto analysis, method, 432,

309-10, 341, 604 Bakhtin, Mikhail, 604, 620 433 n. 8, 434 n. 16

Agawu, Kofi V., 228, 545 n. 59 Barthes, Roland, 604, 607 n. 11 “vestigial third tutti,’ 441, 442,

Alberti, Domenico, 320 Baudrillard, Jean, 252 n. 32, 604 574 n. 15 Aldwell, Edward, 307 Beach, David, 198-99 Berlioz, Hector, 220, 323, 618 n. 15 Allanbrook, Wye Jamison, 228 Beebee, Thomas O., 604, 606 n. 9 Birnbach, Heinrich Joseph, 14-15

Althusser, Louis, 252 n. 32 Beethoven, Ludwig van, 3, 8, 9, 12, n. 3, 118

Aristotle, 618 14, 15, 16, 158, 366, 387 Boccherini, Luigi, 472 n. 6 caesura-fill, 40 Bonds, Mark Evan Bach, Carl Philip Emanuel concertos, 433, 434, 440, 445, 490 conformational and generative

concertos, 432, 434, 436, 438, continuous exposition, 52 form, 8, 10, 615 441, 442, 443, 475 C-themes (closing zone), 181 “false recapitulation,” views of,

deformation, 617 deformation, 11, 617 222-23

expositions, 54 n. 6 development, 196, 197, 230 recapitulation, issues in, 256-58, rondos, 388 n. 1, 403, 407 n. 37 expositional repeats, 22 259, 260, 363 n. 11

Bach, Johann Christian hermeneutics, 341 Bourdieu, Pierre, 604

concertos, 432, 434, 439, 442 n. hypermonumentality of Boyd, Malcolm, 240 n. 11

38, 472 n. 6 symphonies, 16, 285 Brahms, Johannes, 49, 212 n. 17,

developments, 197 introduction, 302 349-50, 435

Mozart’s adaptations of (K. 107), number of movements Brown, A. Peter, 51 n. 2, 264, 363

444, 454-68, 493 (multimovement work), 320 n. 10

P-themes, 66 P-theme, 68 Briick, Marion, 432

Type 2 sonatas, 343, 359, 366 n. recapitulation, 255 Bruckner, Anton, 92, 191, 220, 323

21, 367 rondo, 388 n. 1, 417 Biirger, Christa, 604

Badley, Allan, 295 n. 30 scherzo, 331 Biirger, Peter, 604

Badura-Skoda, Eva, 368 n. 34, 600 S-theme, 36, 119, 131, 142, 185 Burnham, Scott, 287

n. 48, 601 TMB (trimodular block), 171 Bryan, Paul, 328 n. 26 633

634 Index of Names

Cadwallader, Allen, 30 n. 9 Derrida, Jacques, 604 Girdlestone, Cuthbert, 571

Caplin, William E., 3 d’Indy, Vincent, 284 Gluck, Christoph Willibald, 296 “basic idea/contrasting idea,” Dittersdorf, Carl Ditters von, 171 n. — Goetschius, Percy, 347

71-72 3, 264, 363, 472 n. 6 Gombrich, E. H., 604, 620

cadences, definitions, 24 nn. 1-2, © Dommer, Arrey von, 195, 205 Gramsci, Antonio, 606 n. 9

28 n. 6, 31 n. 11, 66-67 n. 5, Drabkin, William, 119 Graun, Carl Heinrich, 432, 434

85 n. 15, 131, 169, 184 nn. 2-3 Graun, Johann Gottlieb, 432, 434

Classical Form, 6 Eagleton, Terry, 604 Grave, Margaret G., 363 closing section (codettas), 184 Eckard, Johann Gottfried, 448 Grayson, David

nn. 2-3, 121-22 Eisen, Cliff, 367, 434 basso continuo (piano concertos), codas and codettas, 282, 287, Emerson, Caryl, 620 445 n. 48 385 n. 58. See also Index of Engel, Hans, 543 cadenza, 600 n. 48 Concepts, closing zone Everett, Paul, 447 n. 53 Eingang, 498 n. 7, 516

concertos, 433, 511, 577, 579 recapitulation (concertos), 585 n. developments, 207 n. 12, 222, Fischer, Wilhelm, 394 n. 19 28, 587 n. 32 228-29, 572 Fisher, Stephen C. Ritornello 1 (concertos) 482 n. evaded cadence, 169 reprise overture, 221 n. 26 21, 483 n. 24

hybrid themes, 69 n. 10, 77, 101 rondo and sonata-rondo, 394 Ritornello 2 and development

n. 5, 106 (incl. n. 8), 108, 129 n. 19, 399 n. 26, 413-15 nn. (concertos), 559 n. 76,

period, 106 n. 8 50—54, 56 570 n. 6

phrase, definition, 69 n. 10 Forkel, Johann Nikolaus, 404 n. 35 soloist-group interaction “postcadential,” term, 28 n. 6, 31 Forman, Denis, 511, 543, 571 (concertos), 497 nn. 5—6

n. 11 Forster, Robert, 432, 441, 442 n. sonata-rondo (Mozart’s

rondo, sonata-rondo, 389, 40, 444, 445-46, 598 concertos), 420, 427 nn. 76-77

401-02 Foucault, Michel, 604 Green, Douglass M., 21 n. 11, 347,

sentence, 84 n. 14, 106 n. 8 Fowler, Alistair, 604 349 n. 22, 355, 389, 392 n. 13, “standing on the dominant,” 31 Freud, Sigmund, 252 n. 32 420 n. 60, 425-26, 511

n. 11 Grey, Thomas S., 383 n. 54

subordinate theme, end of, Gadamer, Hans-Georg, 604

121-22 Gagné, David, 30 n. 9 Habermas, Jiirgen, 497 n. 5, 604

Cavett-Dunsby, Esther, 241, 281 Galand, Joel Haimo, Ethan, 258-59 n. 5

n. 1 concerto, 417 n. 58, 419 n. 59, Handel, George Frideric, 307

Chopin, Fryderyk, 618 n. 15 420, 431 n. 5, 432, 433 n. 8, Hanslick, Eduard, 308 Churgin, Bathia, 358, 359 n. 5, 365 436-38, 497 n. 6, 511 Harrison, Daniel, 199 n. 7, 307-08

n. 18 rondo and sonata-rondo, 389 n. Hauptmann, Moritz, 309

Chusid, Martin, 264 n. 13, 275 n. 6, 399 n. 27 Haydn, Franz Joseph, 3, 9, 12, 14,

29, 364 n. 15 Galeazzi, Francesco 16, 51, 158, 320, 387

Christmann, Johann F., 308—09 “authority,” assessments of, 5, coda, 284

Clementi, Muzio, 115, 302 434, 605 coherence in multimovement Cohen, Margaret, 604 concerto, descriptions of, 441, works, 341

Cohn, Richard, 199 446, 449 concertos, 472 n. 6 Cole, Malcolm S. 388-94, 398 nn. development section, 207, 208, continuous exposition, 51—61

23-24, 403 n. 34, 413 n. 50 212, 215 C-themes (closing zone), 181,

Cone, Edward T., 242-45, 407 n. passo caratteristico (S-theme), 118, 186

37 132, 133 developments, 196, 197

Couperin, Francois, 392 ripresa (recapitulation), 231 n. 1 inappropriate choice for sonata Czerny, Carl, 118, 365, 390-92, sonata form, description of, 14 paradigms, 363, 413

446, 543 Type 2 sonata, description of, introductions, slow, 295-304

364-65 key choices, slow movements,

Dahlhaus, Carl, 4 n. 1, 136, 543, Galupp1, Baldassare, 320 323-29

604 n. 1 Gassmann, Florian, 264, 268, 399 monumentality of symphonies,

Daverio, John, 349-50 n. 22, 410 n. n. 27 16, 285

44,411 n. 45, 412, n. 48, 420 Gathy, August, 118, 205, 231 n. 1 originality and surprise, 8, 11, 15,

Davis, Shelley, 432, 436, 441 Giddens, Anthony, 604 113, 288, 363, 413, 610, 617

Index of Names 635 recapitulations, recomposed, 233, “subdominant reprise” (K. 311/ closing group (display episode,

236, 241, 243, 244-45, 255, i), 262 n. 11 concerto), 543

388, 404, 416-17 Iser, Wolfgang, 252 n. 32, 342, 604, development (concerto), 564,

repeat conventions in, 22 605 nn. 3-4, 609, 617 n. 13, 571-72

S, minor-mode modules in, 142 620-21 recapitulation (concerto), 577 n. S-theme, P-based, 49, 136, 140 Ivanovitch, Roman, 545 n. 59 19, 580 n. 21, 586 n. 30

sonata-rondos, 413-17 Ritornello 1, modular layout TMB (trimodular block), 171 Jackson, Timothy L., 368 n. 38 (concerto), 471, 473-74, 481

TR (transition) launch (forte Jakobson, Roman, 620 solo entry (concerto), 511 affirmation), 66 Jameson, Fredric, 604, 606 n. 8 tutti interjections (concerto), Type 2 sonata, 358, 362-63, 366, Jauss, Hans Robert, 604 521-22

367 Jones, David Wyn, 326

“witalism” effect in, 16 n. 4, 233, Larsen, Jens Peter, 4 n. 1

413 Kamien, Roger, 367 “three-part” exposition, term

V:PAC (perfect authentic Kant, Immanuel, 386, 605, 606 (continuous exposition), cadence) in expositional Keefe, Simon P., 446 n. 51, 497 n. 4 51-52, 54, 59

center, 29 Keller, Hans, 3, 399 n. 25 Type 2 sonata, term for, 365-66

wit, techniques of, 39, 40, 55,58, Kerman, Joseph, 3, 4, 268 LaRue, Jan, 3, 4-5

233, 277, 407 codas, 284-89 “bifocal recapitulation,”

Heidegger, Martin, 604 minor mode, 313-17 264 n. 14

Hell, Helmut, 348 Kinderman, William, 119 n. 4, 268 “da capo symphony (overture)”,

Helm, E. Eugene, 327 n. 25 Kirkpatrick, Ralph, 122-23, 221 Helmholtz, Hermann, 309 240 n. 11, 355 nn. 1-2, 379 “exposition-recapitulation” form

Hill, George R., 264, 268 Kirnberger, Johann P., 309 (Type 1 sonata), 347, 348 Hoffmann, E. T. A., 231 n. 1 Koch, Heinrich Christoph introductions, historical source

Hofmann, Leopold, 295 Anhang (closing zone), 121-22, of, 295 n. 30

Honauer, Leontzi, 448 180 Type 2 sonata, approach to, 363

Hoyt, Peter A. Anlage, 448 n. 11, 367 n. 33

développement (Reicha), concept “authority,” assessments of, 5,14, | Latrobe, Christian, 368 n. 35

of, 195 n. 2 121-22, 132, 440, 475, 579, Leclair, Jean-Marie, 393 n. 16

“false recapitulation,” view of, 605 Leeson, Daniel N., 187 n. 6, 432,

208, 222, 223, 226 coda, 283 n. 6 437 (table 19.1, subtype C),

Husserl, Edmund, 342, 604 concerto format, description of, 440-41, 442, 471-74, 492, 434, 435-50, 454, 462, 466, 494, 543 Ingarden, Roman, 340, 604 572, 573-74, 576 n. 16, 580, Leichtentritt, Hugo, 196, 205, 347

Irving, John, 433 581 Leisinger, Ulrich, 403-04 n. 35 cadenza, 600 n. 48, 601 nn. 49, development, 207, 208 Lesser, Simon O., 252 n. 32

51, 602 n. 53 ein cantabler Satz (S-theme), 118, Levin, Robert D., 187 n. 6, 432, development (concertos), 511, 121-22, 123, 132, 133 440-41, 442, 437 (table 19.1,

563 n. 2,564 n. 4 Nachschlag, 34 subtype C), 471-74, 492, 494,

Eingang, 498 n. 7 recapitulation, 231 n. 1, 235 526 n. 28

K. 107, 448 n. 58 soloist-group interaction (actor- Levy, Janet M., 269 Koch, assessment of, 440, chorus, concerto), 497 Liszt, Franz, 11, 383 n. 54,

449 n. 63, 497 n. 3 sonata form, general description 618 n. 15 recapitulation (concertos), 577 n. of, 14, 366 n. 24 Lobe, Johann Christian, 231 n. 1 17, 585 n. 29, 587 sonata-rondo (concerto), 420—23 Lockwood, Lewis, 143 n. 26 Ritornello 1 (concertos), 470 n. Type 2 sonata, absence of Loewe, Carl, 15 n. 2

2,472 discussion, 364 Lohlein, Georg Simon, 208

Ritornello 2 (concertos), 551, Kollmann, Augustus Frederic Longyear, Rey M., 367

553 n. 70, 559 n. 75 Christopher, 5, 14, 118, 364, Luhmann, Niklas, 604 slow movements (concertos), 431 396 n. 21, 403, 434, 441, 446

n. 3 Kramer, Lawrence, 147 n. 31 Mahler, Gustav, 191, 212 n. 17,

sonata-rondo (concertos), Kucaba, John, 363 233 n. 2, 309-10, 323, 331,

399 n. 27, 420 Kiister, Konrad, 432 618 n. 15

636 Index of Names

Malloch, William, 331 n. 30 Mysliveéek, Josef, 434 Reicha, Anton Marston, Nicholas, 236 Newman, William S., 3, 4-5, 318 “authority,” assessments of, 5,

Marx, Adolph Bernhard n. 1, 319-20, 327 n. 24, 333, 121-22, 605

exposition themes (P and S), 367, 368 coda, 283 n. 6 gendered, 118, 145-47 Nicolussi-Prohaszka, Hermine, développement, 195, 208

Grundidee, 341 295 n. 30 exposition, layout, 118, 121-22, “Leonore 1, program for, 68 Notley, Margaret, 323 123

motion” as central to sonata idées accessoires (closing zone), form, 205 Oettingen, Arthur von, 309 123, 150, 180, 243

recapitulation, 231 n. 1 Oster, Ernst, 198, 317 recapitulation, view of, 231 n. 1

slow movements, 323 rondo, 390-92

“sonata form,” coining of term, Plantinga, Leon repeats in overtures and finales,

14-15 cadenzas, Beethoven’s, 21n. 12, 334 n. 38 7 Clementi, 220 n. 23, 265, 448 discussion, 365 transition (exposition), 93, 101 n. display episode, 543, 546 Réti, Rudolph, 3 5 K. 491, coda, 597 n. 43 Richards, I. A., 252 n. 32

“sonatina’ (Type 1 sonata), 346— 600 n. 48 Type 2 sonata, absence of

Type 2 sonata, absence of K. 503, Ritornello 1, 487 n. 30 Richter, Ernst Friedrich, 195

discussion, 365 solo entry (concerto), 518 Ricoeur, Paul, 621

Mendelssohn, Felix, 21, 307, 434 “tutti” or “ritornello” (term), Riemann, Hugo, 309, 347. See

Merleau-Ponty, Maurice, 604 445 also Index of Concepts, neoMezger, Franz, 122 Plath, Wolfgang, 448 n. 58 Riemannian theory (R, P, L Momigny, Jér6me-Joseph de, 111, Poggioli, Renato, 615 n. 8 operation)

231 n. 1, 365 Popper, Karl, 605 n. 5 Riepel, Joseph, 118, 309, 372 n. 40

Morgan, Robert P., 190 n. 8, 287 Portmann, Johann Gottlieb, 208, Rosen, Charles, 3

Morson, Gary Saul, 620 364 “closing theme,” 121

Mozart, Leopold, 366 n. 21 Prout, Ebenezer, 347 codas, 282, 283 (including n. 6), Mozart, Wolfgang Amadeus, 3, 8, Puri, Michael, 383 n. 54 287 n. 22, 382

9, 12, 14, 15, 16, 40, 387, 610, concertos, 433, 450, 496 n. 1, 617 Quantz, Johann Joachim, 446, 548 n. 64, 577-78 n. 19

C-theme (closing zone), 181, 511 n. 10 dissonance of expositional

185, 186 nontonic material, 243—44 codas, 284, 285 Rameau, Jean-Philippe, 308 n. 4, “false recapitulation,” 222, 225

concerto, approaches to, 417-27, 392-93 tonal “polarization” (opposition),

432, 433, 434, 439, 440, 441, Ratner, Leonard G., 3, 347 4,117 442, 443-45, 448, 451, 453- Classic Music, 4-5 “recapitulation” in Haydn, op. 50

602 codas, 283, 284, 286 no. 1, 269-71

continuous exposition, 52, 61-63 development (“X section”), 195, “reversed recapitulation” error,

developments, 196, 197, 212 197 n. 4, 205, 207 n. 12, 386 n. 59

EEC (essential expositional 216 n. 22, 228 rondos (C. P. E. Bach), 403 closure) deferral, frequency of, introductions, 297 “slow movement form” (Type 1

151 minor mode, 309 sonata), 347—48

“false recapitulation” effects, rhetorical figures, 5, 228, 297 sonata form, conception of, 4,

222, 226-28 rotational developments, 205

“loops,” 80, 84-86 anticipations of concept, 207 Sonata Forms, 4, 343

multimovement formats, 320, n. 12 “stereotypes,” historical (sonata

341 “Scarlatti sonata form,” 367 form, 1750s, 1760s), 141-42,

P-themes, 66, 68 topics, 65, 228, 253 (including 259

recapitulations, issues in, 234, 255 n. 33), 297, 334, 398 n. 23 subdominant (flat-side) in

repeat conventions in, 22 Ratz, Erwin, 3, 6, 31 n. 11, recapitulations, 235-37,

S-themes, 36, 136, 142 228-29 237 n. 9

TMB (trimodular block), 171 Raupach, Hermann Friedrich, subdominant return in

Type 2 sonata, 358, 359, 362, 366 448 recapitulation, 264

Index of Names 637

symmetries in sonata form Simon, Edwin G., 432 n. 7, 448 opposition to, 6, 7, 205,

(anticipation of rotational n. 58 471 n. 3

form), 612-13 Sisman, Elaine, 227 n. 37, 228, varied repeats, 21 Type 2 sonata, 368 320, 323 n. 12, 399 n. 27, 401 Tynianov, Yuri, 620

V/vi at end of development, 198, n. 29 Vanhal, Johann, 328 n. 26 201 n. 8 Snyder, John L., 265-66 nn. 19-20, —- Viotti, Giovanni Battista, 472 n. 6

Rosen, David, 201 n. 8, 433 n. 10, 267 n. 21 Vivaldi, Antonio, 433-34, 446 434, 442 n. 40, 449, 482 n. 21, Somfai, Laszl6, 68—69, 117-18 n. (including n. 53)

511 n. 9 2, 136 n. 19, 137, 296, 327 n. Vogler, Georg Joseph (Abbé) Rosetti, Francesco Antonio, 366 24, 367 “authority,” assessments of, 434,

n. 21 Sonnleithner, Christoph, 264 579

Rosmarin, Adena, 604 Spies, Claudio, 6 composer, 472 n. 6

Rossini, Gioachino, 186 Stamitz, Johann concerto movement, layout, 435— Rothstein, William, 121-22, 149 concertos, 432 37, 439-40, 442 n. 38, 446-50, Rousseau, Jean-Jacques, 309, 398 deformations, 617 454, 462, 580, 581

n. 23 four-movement symphonies, 320 exposition (themes, cadences),

Rutini, Giovanni Marco (Placido), recapitulations, reordered, 233 118, 122

264, 320, 368 n. 35 repeat schemes in, 20 subdominant recapitulation, 264 Wagenseil, Georg Christoph,

Sadie, Stanley, 367 Type 2 sonata, 359, 366, 367, 171 n. 3, 363, 579 n. 19 Saint-Foix, Georges de, 426, 372, 383, 385 Wagner, Richard, 11, 316 n. 21,

526-27, 539 n. 51 Stamitz, Karl, 472 n. 6 618 n. 15 Salieri, Antonio, 275 Steblin, Rita, 309 Webster, James

Sartre, Jean-Paul, 604, 605 n. 3, Stevens, Jane R., 122 n. 11, 432, arias, Mozart, 249 n. 26

609 n. 13 436, 441, 448 n. 57, 451, 471, “binary variants,” 353, 363 n. 11,

Scarlatti, Domenico, 122, 240 n. 11 474-75, 511, 574 n. 14, 580 366-67, 368

Schachter, Carl, 307 n. 22 concertos, 431 n. 3, 432, 436 n. Scheibe, Johann Adolph, 364, 446 Straus, Joseph N., 618 n. 14 22, 450 Schenker, Heinrich, 122, 198. Strauss, Richard, 221, 618 n. 15 cyclic integration/organization, See also Index of Concepts, Strunk, Oliver, 207-08 n. 14 318 n. 1, 319, 613 n. 5 Schenkerian analysis Sutcliffe, W. Dean, 269, 355 nn. 1, “double return,” 343, 366—67

coda, 281 n. 1 3, 399 n. 25 “false recapitulation,” 222

Der freie Satz, 5, 13 introductions, 298, 299 n. 37 minor mode, 309 Talbot, Michael, 334-35, 433-34 key relations, Haydn, 326 n. 20

sonata form, view of, 5, 147-49, n. 12 multimovement sonata, 319, 320, 205, 366 n. 24 Tchaikovsky, Pyotr Il’yich, 618 335, 341 n. 45

Type 1 sonata, view of, 347 n. 15 “pillars” (ritornello), 436 n. 22 Schmalfeldt, Janet, 169 n. 2 Tobel, Rudolf von, 221 n. 26, 259, sonata form, description of,

Schmitz, Arnold, 136 275 n. 29, 278 n. 32, 365-66, 362 n. 7, 366-67

Schobert, Johann, 259, 448 394 n. 19 “textbook” formal models, Schoenberg, Arnold, 3, 6, 205, 214, Todorov, Tzvetan, 604 objection to, 6—7 282, 285 n. 18, 303, 347, 389 Torelli, Giuseppe, 433 “three-part” exposition, 51 n. 2 Schopenhauer, Arthur, 331 n. 31 Tovey, Donald Francis, 3, 120 n. 5, throughcomposition, 319

Schr6ter, Johann Samuel, 442 n. 38, 147 n. 31, 327 Wheelock, Gretchen A., 309

431n.5 “concerto principle,” 436, 553 Williams, Raymond, 604, 606 n. 9

Schubert, Franz, 3, 171, 177, 265, concertos, 433, 471 n. 3, 484 n. Wolf, Eugene K.

331, 618 n. 15 30, 542, 548, 577, 579 n. 19 da capo symphony, 221 n. 26 Schulz, Johann Abraham Peter, developmental episodes, 212 “double second themes” (TMB

321, 323 errors in analysis, 327 n. 25 in Stamitz), 171 n. 3

Schumann, Robert, 11, 21, 49, 279, proportion, lesson in, 278 n. 33 reordered recapitulations

383 n. 55, 618 n. 15 “rotation,” use of term, 247 (Stamitz), 233 n. 3 Shklovsky, Viktor, 12 sonata “forms” (plural), 4, 343 repeats in Type 1 sonatas, absence

Sibelius, Jean, 323 “textbook” forms (molds), of, 346 n. 9

638 Index of Names

Wolf, Eugene K., (continued) Stamitz symphonies, 320 Zarlino, Giosefto, 307 “reversed recapitulation,” Type 2 sonata (Stamitz), 359 n. Zaslaw, Neal, 220, 322 n. 10, 323,

objection to concept, 383-84 6, 362 n. 8, 363 n. 14, 366, 331, 334, 362 n. 7, 385 nn.

(including n. 55) 367-68, 370 n. 39, 372 n. 40, 56-57 “sonatina,” 348 383 Zizek, Slavoj, 604 Wyzewa, Théodore de, 526

Index of Works

Bach, Carl Philip Emanuel op. 5 no. 1 in B-flat, 1, 362, 369, No. 4in G, BWV 1049, 1, 441

Concerto for Harpsichord in B 373, 380, 381 No. 6 in B-flat, BWV 1051, 111, Minor, H. 440, 1, 511 op. 5 no. 2 in D, i, 237, 454-63 441 n. 33 Rondo in C, H. 260 (Kenner und (ex. 19.1), 466-67 English Suite No. 5 in E Minor, Liebhaber 2/1), 403 op. 5 no. 3 in G, 1, 362, 378, 381, BWV 810, v (“Passepied en

sonatas, keyboard 454, 464-67 rondeau’’), 393

Prussian Sonatas (H. 24-29), 54, op. 5 no. 41in E-flat, 454, 463-— French Suite No. 4 in E-flat, BWV

64 n. 13, 381 64, 466-67 815, 111 (sarabande), 91

No. 1 in F, H. 24, iii, 359-61 op. 5 no. 5 in E, 1, 362, 381 n. 24

(Ex. 17.2), 362, 369 op. 5 no. 6in C Minor, 1, 362 n. Partita No. 1 in B-flat, BWV 825,

No. 5 in C, H. 28, i, 359 9 iv (sarabande), 91 n. 24 Sonata in F Minor, H. 173 op. 17 no. 1 in G, 1, 362 (Kenner und Liebhaber, 3/111), op. 17 no. 3 in E-flat, 1, 362 Beethoven, Ludwig van

111, 399 op. 17 no. 6 in B-flat, 1, 362 Bagatelle in E-flat, op. 126 no. 6,

Sonaten mit veranderten Reprisen symphonies 305 n. 43

(H. 136-39, 126, 140), 21. op. 3 no. 1 in D, 1, 237, 362 n. 9; Concerto for Piano, Violin, and

Wiirttemberg Sonatas (H. 30-34, 1, 362 Cello (“Triple Concerto’)

36), 54 op. 3 no. 2 in C, 1, 237 in C, op. 56, 1, 120, 548,

symphonies, “Hamburg” op. 3 no. 3 in E-flat, 11, 115 556

326 382 i, 512, 521, 533, 546, 548,

No. 1 in G, H. 657, 1, 265; i, op. 3 no. 4 in B-flat, 1, 362, 369, Concerto for Violin in D, op. 61,

No. 2 in B-flat, H. 658, 11, 326 op. 3 no. 5 in F, 1, 362, 381 556, 558, 559 n. 75, 597

No. 3 in C, H. 659, i, 265 op. 3 no. 6 in G, 1, 362 n. 9, 381 n. 43

No. 4 in A, H. 660, 11, 326 concertos, piano

No. 5 in B Minor, H. 661, i, 265 Bach, Johann Sebastian No. 1 in C, op. 15, 1, 490, 520, No. 6in E, H. 662, 1, 265 Concerto No. 2 for Harpsichord 521, 533, 558; 11, 249, 284 Symphony in D, H. 663, 11, 327 in E, BWV 1053, 1 and 111, No. 2 in B-flat, op. 19, 1, 490,

441 n. 33 495 n. 38, 520, 525, 557,

Bach, Johann Christian Concerto No. 2 for Violin in E, 570; 11, 284

concertos, keyboard, 439, 441 BWV 1042, 1, 441 No. 3 in C Minor, op. 37, 1, 313, op. 7 no. 2 in F,1, 431 n. 5 concertos, Brandenburg 490, 512, 521, 533, 546-49 op. 7 no. 5 in E-flat. i, 441 No. 3 in G, 198, BWV 1048, (ex. 21.7), 558, 597; 11, 48,

sonatas, keyboard 1i-i1i, 198 74 639

640 Index of Works

concertos, piano, (continued) Quintet for Strings in C, op. 29, 1, op. 90 in E Minor, 1, 315

No. 4 in G, op. 58,1, 471 n. 4, 120, 185 op. 106 in B-flat,

490, 497, 520, 537, 538 n. Rondo for Piano in C, op. 51 no. 1, ‘“Hammerklavier,” i, 237 n.

50, 546-51 (ex. 21.8), 573 407 n. 37 9; iv, 297, 302

n. 13, 597 n. 43 Septet in E-flat, op. 20, 11, 249 op. 109 in E, 1, 299

No. 5 in E-flat, op. 73, sonatas, cello op. 111 in C Minor, i, 296, 317 “Emperor,” 1, 471 n. 4, 497, No. 1 in F, op. 5 no. 1, 1, 287, WoO 47 no. 2 in F Minor, 1, 220,

512, 533, 548 n. 62, 558, 296 296, 314 570, 596-97 (including n. No. 2 in G Minor, op. 5 no. 2, 1, sonatas, violin

43); 11, 328; 111; 328 296 op. 23 in A Minor, 1, 219, 315; 11, overtures No. 4 in C, op. 102 no. 1,1, 15 n. 328; 111, 316 Coriolan, op. 62, 26, 89, 120, 147, 2,301 n. 40 op. 24 in F, “Spring,” 1, 46, 199

179, 215, 264, 288, 315, sonatas, plano op. 30 no. 2 in C Minor, 1, 20, 21 316-17 op. 2 no. 1 in F Minor, 1, 31 n. op. 47 in A Minor, “Kreutzer,” 1,

Egmont, op. 84, 68, 147, 179 n. 10, 101 n. 5, 108, 145, 229, 315; 11, 271 n. 27 9, 245, 247, 253, 286, 287, 314; 11, 284, 328; 1v, 214, 316 symphonies

288, 297, 298, 300, 302 op. 2 no. 2in A, 1, 186, 217 No. 1 in CG, op. 21, 1, 36, 125-28

Fidelio, op. 72, 178, 299 op. 2 no. 3 in C, 1, 141, 172-75 (ex. 7.1), 152, 185, 201, Leonore No. 1, op. 138, 68, 146, (ex. 8.7), 287, 289; 11, 120 n. 208-11 (ex. 10.4), 218, 231,

221, 298, 303 5, 284, 326; iv, 407 297; i1, 163; iv, 238, 297, 303

Leonore No. 2, op. 72, 120, 146, op. 7 in E-flat, 11, 284; iv, 407 No. 2 in D, op. 36, 1, 25-26, 45, 248-49, 259, 278 n. 32, 286, op. 10 no. 1 in C Minor, 1, 141, 141, 169, 201, 211, 297, 301;

297, 300, 303-04 238, 275 n. 28, 314; 11. 284, iv, 45—46, 176, 178, 201,

Leonore No. 3, op. 72, 120, 146, 288, 349; 111, 27 203, 246

248 n. 22, 278, 286, 297, op. 10 no. 2 in F, 1, 84 n. 14, 175, No. 3 in E-flat, op. 55, “Eroica,”

300, 303-04 215, 271-75 (ex. 12.4a—b) 1, 22, 86, 92, 111, 143-44

Prometheus, op. 43, 347 op. 10 no. 3 in D, 1, 141, 176, 275 (ex. 7.6), 187-90 (ex. 9.1),

quartets, string n. 28; 1, 27 n. 5, 325; iv, 216, 243, 253, 280 n. 36,

op. 18 no. 3 in D, 1, 105 407 285, 286-87, 292, 607; 11, op. 18 no. 4in C Minor, i, 27, op. 13 in C Minor, “Pathétique,” 325, 342; iv, 334

46, 103, 176, 207 n. 12, 313; i, 97-100 (ex. 6.2), 143, 217, No. 4 in B-flat, op. 60, 1, 68,

iv, 401 n. 31, 402 220, 289 n. 28, 296, 302, 298, 303; i11, 330

op. 18 no. 5 in A, 1, 141 304, 312; 11, 328; 111, 245, No. 5 in C Minor, op. 67, 1, 89,

op. 18 no. 6 in B-flat, 1, 142; iv, 407 139, 140, 217, 313; 11, 323, 297, op. 27 no. 2 in C-sharp Minor, 328; 111, 332, 333; iv, 77,

op. 59 no. 1 in F,1, 20 n. 9, 20, “Moonlight,” iii, 97, 316 178, 217, 246, 251, 286, 21, 211, 218, 226, 280 n. 35, op. 31 no. 1 in G, 1, 120, 211, 335 351, 352; 11, 275, 330, 339; 217, 237 n. 9, 238 No. 6 in F, op. 68, “Pastoral,” 1,

il1, 216, 325, 339 op. 31 no. 2 in D Minor, 50, 253; 11, 74, 76-77 (ex.

op. 59 no. 2 in E Minor, 1, 89, “Tempest,” 1, 299, 315; 111, 5.4), 218, 287

313; iv, 316 316 No. 7 in A, op. 92, 1, 61, 302; 11,

op. 59 no. 3 in C, 1, 298, 303; iv, op. 49 no. 1 in G Minor, 1, 183, 325, 339; i11, 330, 332, 339

246 314 No. 8 in F, op. 93, iv, 49-50,

op. 74 in E-flat, “Harp,” 1, 92, op. 49 no. 2 in G, 1, 30 n. 9, 102 117 n. 1, 246, 280 n. 36,

169; 11-111, 332, 340 op. 53 in C, “Waldstein,” 1, 285-86; iv, 178 op. 95 in F Minor, “Serioso,” 1, 73-74, 97, 102, 120, 152, No. 9 in D Minor, op. 125, 1, 92, 238, 313, 317; 11-111, 329; iv, 156-57 (ex. 8.2), 235, 237 143, 312, 317, 351; 11, 47-48,

316 n. 9, 238, 284-85 258, 330, 332, 338; i11, 338;

op. 127 in E-flat, iv, 267-68 op. 57 in F Minor, iv, 297, 302

op. 130 in B-flat, 1, 237 n. 9, 299, “Appassionata,” 1, 20, 21, Trio for Violin, Viola, and Cello

300; vi, 285-86 73-74, 102, 191, 207, 276, in C Minor, op. 9 no. 3,1,

op. 131 in C-sharp Minor, i1, 349 284-85, 312; 111, 316 49, 314 op. 132 in A Minor, 1, 237 n. 9, op. 81a in E-flat (“Les Adieux”; trios, plano

280, 300, 317; v, 316 “Das Lebewohl”’), i, 296; i1, op. 1 no. 1 in E-flat, 1, 41 n. 17,

op. 135 in F, iv, 297 247, 346 n. 7 91-92, 183; 11, 284

Index of Works 641

op. 1 no. 2 in G,1, 41 n. 17, 296; Cherubini, Luigi Franck, César

11, 179, 246-47, 284 overtures Symphony in D Minor, 1, 68, 303

325 286 Gade, Niels

op. 70 no. 1 in D, “Ghost,” 11, Les deux journées, 68, 177, 179,

op. 70 no. 2 in E-flat, 1, 304; iv, Eliza, 303, Symphony No. 1 in C Minor, op.

120, 247 n. 20, 280 n. 36 Meédeée, 70, 111, 364 5,1, 304

op. 97 in B-flat, “Archduke,” 1,

120; 11-111, 338 Chopin, Fryderyk Gassmann, Florian Leopold Concerto No. 2 for Piano in FP symphonies

Berlioz, Hector Minor, op. 21, 11, 305 H. 15 in B-flat, 1, 268 n. 24 Harold in Italy, 1, 323 n. 13; iv, 336 sonatas, piano H. 63 in E, i, 268 n. 24

overtures No. 2 in B-flat Minor, op. 35, i, H. 86 in C, 1, 268 n. 24 Benvenuto Cellini, 299, 305, 323 364

n. 13 No. 3 in B Minor, op. 58, 1, Glazunov, Aleksandr Le carnaval romain, 20 n. 7, 299, 364 symphonies

323 n. 13 No. 2 in F-sharp Minor, op. 16,

Le corsaire, 219, 299, 323 n. 13 Clementi, Muzio 1, 304

Symphonie fantastique, i, 147, 278-79, — sonatas, piano/harpsichord No. 4 in E-flat, op. 48, 304 304, 323 n. 13, 383-84 n. 55 op. 5 no. 3 in E-flat, 1, 265

op. 10 no. 2 1n D, 111, 265 Glinka, Mikhail

Boccherini, Luigi op. 10 no. 3 in B-flat, 1, 265 Ruslan and Lyudmila, overture, 247 Symphony in D Minor, op. 12 no. op. 13 no. 4 in B-flat, 1, 265 2, “La casa del diavolo,” 111, op. 13 no. 5 in F, i, 279; 11, 265 Gluck, Christoph Willibald

296, 336 op. 13 no. 6 in F Minor, 1, 277; Alceste, overture, 275 11, 329 Orfeo ed Euridice (aria, “Che faro

Brahms, Johannes op. 22 no. 3 in C, 220 senza Euridice’’), 308 overtures op. 25 no. 5 in F-sharp minor, 1, Academic Festival, op. 80, 287 317; 11, 329 Haydn, Franz Joseph Tragic, op. 81, 350, 364 op. 33 no. 1 in A, 448 The Creation, 303

Piano Quartet in G Minor, op. 25, op. 34 no. 2 in G Minor, 1, 280, La fedelta premiata, 326 n. 22

1, 349, 351 296 quartets, string

Piano Quintet in F Minor, op. 34, op. 40 no. 2 in B Minor, 11, 220, op. 1 no. 2 in E-flat, i, 256

iv, 350, 352 296 op. 17 no. 1 in E, 222, 223 n. 34

Piano Trio No. 1 for Piano, Violin, op. 40 no. 3 in D, 1, 301 op. 20 no. 1, in E-flat, 11-11, 338 and Cello in B, op. 8, iv, 336 op. 50 no. 3 in G Minor, op. 20 no. 2 in C, 1v, 335 Schicksalslied (Song of Destiny), op. “Didone abbandonata,” op. 20 no. 3 in G Minor, 1, 177-

54, 77 329 78; 1-11, 328, 338

String Quartet in C Minor, op. 51 Sonatina in C for piano, op. 36 no. op. 20 no. 4 in D, 1, 74, 223 n.

no. 1, iv, 350, 352, 364 3,1, 116, 237 34, 225

symphonies op. 20 no. 5 in F Minor, 1-111, No. 1 in C Minor, op. 68,1, 191, Dittersdorf, Carl Ditters von 328; iv, 335, 338 303, 312; iv, 297, 350, 351 symphonies on Ovid’s op. 20 no. 6in A, iv, 335

n. 27, 352 Metaphorphoses, 253 op. 33 no. 1 in B Minor, 11, 328; No. 2 in D, op. 73, 1, 44; 1, 249; No. 1 in C, “The Four Ages of iv, 52, 54, 242, 289,

iv, 352 the World,” ii, 158; i111, 332 op. 33 no. 2 in E-flat, “Joke,” i,

No. 3 in F, op. 90, 1, 44, 89, 191, n. 32 52, 54-57 (ex. 4.1), 58, 262, 247, 312 n. 12; 11, 249; 1Vv, No. 6in A, “Transformation 268; iv, 401

350, 352 of the Lycian Peasants into op. 33 no. 3 in C, “Bird,” 1, 203, No. 4in E Minor, op. 98, 1, 351, Frogs,” iv, 296, 336 218, 239, 243, 244; 111, 21; 352; 11, 339; 111, 339 Symphony in E-flat (3), 1, 368 iv, 401

Violin Sonata No. 1 in G, op. 78, 1, n. 34 op. 33 no. 4 in B-flat, iv, 400

351, 352 op.Dvorak, 33 no. 5 in G, 1, 66-67 (ex. Antonin 5.1), 87, 140; 11, 249

Bruckner, Anton Symphony No. 8 in G, op. 88, 1, op. 33 no. 6 in D, 1, 276-77; 1,

Symphony No. 7 in E, iv, 364 304 249; iv, 401

642 Index of Works

quartets, string, (continued) No. 28 in E-flat, i, 68 No. 45 in F-sharp Minor,

op. 42 in D Minor, iti, 328 No. 29 in F, 1, 68 “Farewell,” i, 52, 114, 221, op. 50 no. 1 in B-flat, 1, 74, 269-— No. 35 in C, i, 68; 11, 363 315, 316, 319; 11, 328 71 (ex. 12.3a—b); iv, 223 n. No. 41 in B-flat, i, 68 No. 46 in B, 1, 223 n. 34, 225;

34, 225 No. 42 in D, i, 322 111, 332 n. 34

op. 50 no. 3 in E-flat, 1, 239 No. 45 in E-flat, 1, 68 No. 48 in C, “Maria Theresia,” op. 50 no. 41n F-sharp Minor, i, No. 46 in B, 1, 170 111, 332 n. 34; iv, 222, 223 n.

328 No. 48 in C, 1, 322, 388 34, 225

op. 50 no. 6 in D, 1, 68, 244 No. 49 in E-flat, 1, 68—69 No. 49 in F Minor, “La op. 54 no. 3 in E, iv, 223 n. 34, No. 50 in C, 1, 68, 171 n. 5 passione,” 329; 1, 322; i11,

225 No. 51 in D, 1, 20 n. 9, 211, 322 332 n. 34

op. 64 no. 2 in B Minor, 11, 328 No. 52 in E-flat, i, 49, 68, 91; ii, No. 50 in C, 1, 295; iv, 226

op. 64 no. 3 in B-flat, 1, 74, 177, 327 No. 52 in C Minor, 11, 328 203, 238 n. 10, 244, 271, symphonies (Hob. I) No. 53 in D, i, 295

378 n. 44 No. 1 in D, 11, 259 No. 54 in G, 1, 295

op. 71 no. 2 in D, 1, 296 No. 3 in G, 11, 325 No. 55 in E-flat, i, 222, 223

op. 71 no. 3 in E-flat, 1, 292 No. 41in D, 11, 325 n. 34, 225

op. 74 no. 1 in C, 1, 67-68, 136, No. 5 in A, i, 327 No. 56 in C, ii-i1i, 332 292; 111, 332 n. 33; iv, 414 No. 6in D, “Le matin,” i, 239, No. 57 in D, 1, 295,

op. 74 no. 2 in F, 111, 332 n. 33; 300 No. 60 in C, 1, 295, 298

iv, 334 n. 38, 416 No. 9 in C, 11, 363 No. 62 in D, 11, 327; i11, 332

op. 74 no. 31n G Minor, “The No. 10 in D, i, 27 n. 33

Rider,” 11, 326 n. 23; iv, 414 No. 11 in E-flat, 1, 327; 11, 223 n. No. 63 in C, 11, 325

op. 76 no. 1 in G, 1, 87, 292; iv, 34 No. 65 in A, 1, 256; iti, 332

336, 414 No. 12 in E, 11, 325 n. 34

op. 76 no. 2 in D Minor, No. 13 in D, 1, 52, 54 No. 67 in F, iv, 221 “Quinten,” 11, 328; iv, 334 No. 14 in A, 11, 377 No. 68 in B-flat, 11-111, 322

n. 38, 416 No. 17 in F, ii, 325 No. 70 in D, 11, 325

op. 76 no. 3 in C, 1, 203, iv, 414 No. 18 in G, 1, 256, 327 No. 71 in B-flat, 1, 223 n. 34, op. 76 no. 4 in B-flat, “Sunrise,” No. 19 in D, 1, 256; 11, 325 225, 295; 11, 324 n. 18

i, 289; iv, 414 No. 21 in A, 1, 327; 11, 377 No. 72 in D, i, 256; 11, 324 n. 18

op. 76 no. 5 in D, 11, 326; iv, 416 No. 22 in E-flat, “Der No. 73 in D, 1, 295, 298; 11, 324 op. 76 no. 6 in E-flat, 11, 326; iv, Philosoph,” i, 327; 11, 223 n. n. 18,

414 34 No. 74 in E-flat, 11, 324 n. 18 op. 77 no. 1 in G, 223 n. 34, 244; No. 24 in D, 1, 259; 111, 380 No. 75 in D, 1, 280 n. 35, 295; 11, ii, 326; iv, 414 No. 35 in B-flat, 11-111, 332 324 n. 18

op. 77 no. 2 in F, 111, 326; iv, 414 No. 36 in E-flat, 1, 256; iv, 223 n. No. 76 in E-flat, 11, 324 n. 18; iv,

sonatas, piano/keyboard (Hob. 34 71, 399

XVI) No. 37 in C, i, 256; 11-111, 322 No. 77 in B-flat, 11, 324 n. 18; iv,

No. 3 in C, 1, 259 No. 38 in C, ii-i11, 332; iv, 223 71, 399 No. 5 in A (doubtful attribution), n. 34 No. 78 in C Minor, 11, 328 i, 363, 369 No. 39 in G Minor, 11, 328; 111, No. 79 in F, 11, 324 n. 18

No. 6 in G, 1, 363 332 n. 33 No. 80 in D Minor, ii, 328 No. 13 in E, 1, 68 No. 40 in F, iv, 335 No. 81 in G, 11, 324 n. 18

No. 14 in D, i, 68 No. 41 in C, 1, 222, 223 n. 34, No. 82 in C, “Bear,” i, 39 n. 15,

No. 18 in B-flat, i, 68 225, 114; 11, 324 n. 18

No. 19 in D, 1, 68 No. 42 in D, 1, 223 n. 34, 225 No. 83 in G Minor, “Hen,” 1, 47, No. 20 in C Minor, 1, 59, 68; 11, No. 43 in E-flat, “Mercury,” 1, 132-33 (ex. 7.3a); 1, 317, 328

328 223 n. 34, 225, 262; i111, 332 No. 84 in E-flat, 1, 295; 11, 324 n. No. 21 in C, i, 68 n. 33 18 No. 22 in E, 1, 68 No. 44 in E Minor, “Trauer,” i, No. 85 in B-flat, i, 295; i1, 324 n.

No. 23 in F, i, 68, 137 52, 87-88 (ex. 5.8), 111-12, 18; iv, 415

No. 24 in D, i, 68 314; 11-111, 322, 328, 332 n. No. 86 in D, i, 295, 298; 11, 324 No. 25 in E-flat, 1, 105, 137, 177 34; iv, 334, 363, 381, 385 n. 18

Index of Works 643 No. 87 in A, ii, 324 n. 18 No. 29 in E-flat, 11, 326; i11, 234, Concerto for Bassoon in B-flat, K.

No. 88 in G, i, 52, 61, 298; 11, 258 n. 4 191/186e, 453; 1, 482, 525 324 n. 18; iv, 415 Keiser, Reinhard (including n. 26), 526 n. 29,

n. 18 571 n. 10, 580, 582-84

No. 89 in F, 1, 280 n. 35; 11, 324 Croesus, sinfonia (overture), 221 527, 536, 540, 544, 556,

No. 90 in C, i, 298; 11, 324 n. 18 Leo, Leonardo (ex. 22.5a—b), 586, 588, No. 91 in E-flat, 222, 223 n. 34, L’Olimpiade, sinfonia (overture), 599

225; 11, 324 n. 18 221 Concerto for Clarinet in A, K. 622, No. 92 in G, “Oxford,” 1, 298; 11, i, 139, 440, 482, 483 n. 23,

324 n. 18 Liszt, Franz 494, 494 n. 37, 495, 528,

No. 93 in D, 11, 324 n. 18; iv, 416 Piano Sonata in B Minor, 221 537 n. 47, 538 (including

No. 94 in G, “Surprise,” 1, 68, symphonic poems m. 50), 544, 570, 574, 576, 29, 298; 11, 324 n. 18; iv, 415 Les préludes, 364 580 n. 21, 581, 594, 596; 111,

No. 95 in C Minor, 1, 87, 89, Tasso, 304 419, 426 111-12, 295, 302; 11, 328; iv, Concerto for Flute and Harp in C, 335, 415-16 Mahler, Gustav K. 299/297c, 1, 537 n. 47, No. 96 in D, “Miracle,” i, 52, symphonies 540, 544, 556, 562, 580 n. 55-58 (ex. 4.2), 80; 11, 324 No. 1 in D, 1, 68; iv, 364 21, 597, 598, 599; 111, 419

n. 18; iv, 414 No. 4, 1, 351 Concerto for Flute in G, K.

No. 97 in C, i, 52, 58-59, 298; 11, No. 6in A Minor, i, 183, 247; iii 313/285e, i, 44 n. 19, 481

324 n. 18; iv, 416 (Andante moderato), 77 (including n. 14), 482, 488,

No. 98 in B-flat, i, 298, 301, 303; 529-32 (ex. 21.6), 537 n.

i, 324 n. 18; iv, 414 Mendelssohn, Felix 47, 544, 557, 580 n. 21, 588,

No. 99 in E-flat, 1, 36, 114, 215 Concerto for Violin in E Minor, 599 (including n. 45); 11,

n. 20, 298; 11, 326; i11, 332; op. 64, 1, 87; 11, 77 87

iv, 415 Lied ohne Worte in A, op. 19b no. 3, Concerto for Flute in D, K. No. 100 in G, “Military,” 1, 26, “Jagerlied,” 305 n. 43 314 (revision of Oboe 41, 95, 215-16, 303; i1, 324 Octet in E-flat, op. 20, 1-111, 340, Concerto). See Concerto for n. 18; iv, 416 364 Oboe in C, K., 314/285d.

No. 101 in D, “Clock, 1, 80, 301, overtures Concerto for Oboe in C, K. 303; 11, 324 n. 18; iv, 416 The Hebrides, 27 314/285d, i, 481, 482 n. 18,

No. 102 in B-flat, i, 84 n. 13, 97, A Midsummer Night’s Dream, 145, 525 (including n. 26), 526 n.

222, 223, 227-28, 299, 303; 186, 234, 304, 305 27, 536 nn. 42 and 43, 557,

ui, 21, 141, 324 n. 18; iv, Die schéne Melusine, 305, 571 n. 8, 580 n. 21, 586,

415 quartets, string 598, 599; 111, 419

No. 103 in E-flat, “Drumroll,” i, No. 1 in E-flat, op. 12, 11-111, 340 = Concertone for Two Violins in C,

52, 95, 198, 215 n. 20, 289, No. 5 in E-flat, op. 44 no. 3, K. 190, 1, 454 n. 71, 520,

298, 302, 303, 304; 11, 325; ii—i1i, 340 537, 542 n. 54, 556, 559, 572

iv, 199, 415 quintets, string n. 12, 582, 588, 598, 600;

No. 104 in D, “London,” i, No. 1 in A, op. 18, 11-111, 340 11, 417 n. 58 36-37 (ex. 3.3), 95, 297, No. 2 in B-flat, op. 87, u—i11, 340 concertos, horn

301; 11, 324 n. 18; 111, 332; symphonies No. 1 in D, K. 412/386), 1, 440,

iv, 414 No. 3 in A Minor, op. 56, 498, 527 n. 33, 541, 544,

trios, plano (Hob. XV) “Scottish,” 1, 305, 314 n. 16; 558, 570-71, 574, 581, 595

No. 12 in E Minor, ii, 328 ii, 330 n. 40, 596

No. 14 in A-flat, 11, 326 No. 41n A, op. 90, “Italian,” 11, No. 2 in E-flat, K. 417, 1, 481 n.

No. 20 in B-flat, 11, 326 329 n. 27, 339; ii, 339; iv, 14, 520, 528, 536, 544, 572,

No. 21 in C, i, 298 336 585, 589 n. 36, 596

No. 23 in D Minor, 11, 328 No. 5 in D Minor, op. 107, No. 3 in E-flat, K. 447, 1, 481 n.

No. 25 in G, i11 (Rondo “Reformation,” iv, 297 14, 528, 536 nn. 42 and 43,

all’ongarese) , 397 544, 562, 572, 582, 589, 599

No. 26 in F-sharp Minor, 1, Mozart, Wolfgang Amadeus No. 4 in E-flat, K. 495, i, 481 n.

328 “Un bacio di mano,” arietta, K. 14, 544, 562, 563-64, 574,

No. 27 in C, iii, 54, 399 541, 227 576, 581, 598, 599

644 Index of Works

concertos, piano (traditional No. 13 in C, K. 415/387, 1, 44, nn. 47 and 48, 540, 542, numbers). See also Concertos 442, 481 n. 15, 482, 483 544, 545-46, 557, 562, 564,

for Piano, K. 107 n. 23, 488, 519, 526 n. 27, 571 n. 9, 572 (including n. “Nos. 1-3” (pastiche concertos), 527 nn. 31 and 34, 528, 541 12), 573, 577-78 (ex. 22.4),

K. 37, 39, and 40, 448 (including n. 52), 544, 546, 581, 594, 599 n. 45, 600, “No. 4” in G (pastiche concerto), 557, 571 nn. 9 and 10, 572 601; 11, 431 n. 3; 111, 419,

K. 41, 448, 11, 325 n. 12,576 n. 16, 580 n. 21, 424, 426

No. 5 in D, K. 175, 1, 453, 528, 582, 598, 601; 11, 431 n. 3; No. 20 in D Minor, K. 466, 1, 536, 544, 551, 557, 580, 582, ili, 419, 423-24, 426 n. 74 27, 187, 218, 481 n. 15, 482, 586, 589, 598; i111, 399 n. 27, No. 14 in E-flat, K. 449, i, 74, 483, 490-91 (ex. 20.5), 492,

418 n. 58, 431 483, 495 n. 38, 498, 526 n. 493, 519-20, 521, 533, 537

No. 6 in B-flat, K. 238, 1, 525 n. 27, 528, 537 n. 47, 538 n. nn. 47 and 48, 540, 544, 26, 536, 551, 556, 571 n. 10, 49, 540, 544, 552 n. 69, 558, 551, 553-55 (ex. 21.9), 556, 572 n. 12, 585, 588-89, 598 559, 566, 571 n. 9, 572, 572 564, 570, 571 n. 9,572 n. No. 7 in F (three pianos), K. 242, n. 12, 573, 589, 598, 601; 11, 12, 585, 587, 589, 597, 599; i, 482, 537 n. 47, 540, 544, 178-79, 431 n. 3, 556 n. 73; 11, 328; 111, 335, 419, 423,

551, 556, 598, 599; iii, 419, ili, 419, 420, 425 n. 72 426, 427

421 No. 15 in B-flat, K. 450, i, 74, No. 21 in C, K. 467, i, 26, 85 n. No. 8 in C, K. 246, i, 498, 533, 472, 481 n. 15, 483 n. 23, 16, 166-67 (ex. 8.5), 175537 n. 47, 538 n. 49, 540, 493, 512-14 (ex. 21.2), 526 76, 481 n. 15, 482, 483, 492, 544, 552, 557, 559, 580 n. nn. 27 and 30, 529, 544, 494, 512, 515-16 (ex. 21.3),

21, 598 556, 571 n. 9,572 n. 12, 528, 537 nn. 47 and 48, 539

No. 9 in E-flat, K. 271, 1, 80, 83 573, 585, 586-87, 590, 593, n. 51, 540, 542, 544, 556, (ex. 5.7), 85, 86, 167-69 594, 597, 598, 601, 602; 111, 559-62 (ex. 21.10), 564-66

(ex. 8.6), 471 n. 4, 481, 482, 419, (ex. 22.1b), 570 n. 6, 571 n.

483, 492, 496-97, 517-18 No. 16 in D, K. 451, 1, 482, 483 9,572,572 n. 12, 573, 580 (ex. 21.4), 533, 544, 557, n. 23, 484, 490, 498, 526 n. n. 21, 581, 585, 587, 594— 558, 570, 571 n. 9, 572 n. 27, 529, 536, 552, 556, 559, 95, 597, 599; 11, 87, 279, 431 12, 573, 581 n. 25, 590, 596, 571 n. 9, 572 n. 12, 580 n. n. 3; 111, 419, 420, 422, 426,

597, 599, 601; 11, 325; 111, 21, 585, 586, 589, 600, 601; 427

419, 421 11; 431 n. 3; 111, 419, 420, 425 No. 22 in E-flat, K. 482, 1, 86

No. 10 in E-flat (two pianos), n. 72 n. 20; 482, 483 n. 23, 484,

K. 365/316a, 1, 481 n. 17, No. 17 in G, K. 453, 1, 92, 490, 518, 537 nn. 47 and 48, 482 n. 19, 483 n. 23, 525 475-79 (ex. 20.1), 481 n. 540, 542, 544, 551, 557, 571 (including n. 26), 526, 540, 15, 482, 483, 484, 488, 490, n. 9, 572 (including n. 12), 541, 544, 557, 571 n. 9, 572 494, 498-510 (ex. 21.1), 512, 580 n. 21, 585, 587, 595, n. 12, 580 n. 21, 585, 586, 532-33, 537 n. 47, 538-39, 597, 599; 111, 419, 422, 426,

590 n. 38, 598; iti, 419 543, 544, 545, 555-56, 556, 427

No. 11 in F, K. 413/387a, i, 158-— 559, 570 n. 6, 571 nn. 9 and No. 23 in A, K. 488, i, 187, 481

59, 481 n. 15, 482, 483 n. 10, 572 n. 12, 580 n. 21, n. 15, 482, 483 n. 23, 494 n. 23, 488-90 (ex. 20.4), 492, 585, 587, 589, 594, 597, 598, 37, 498, 521, 533, 536, 544, 493, 519, 525 n. 26, 526 nn. 601; 11; 431 n. 3, 557 n. 73; 556, 558, 566, 570, 571 n. 9,

27 and 30, 528, 536 n. 42, ili, 417 572 n. 12, 573, 585, 589-90, 541, 558, 571 n. 9, 572 n. No. 18 in B-flat, K. 456, 1, 475, 597, 600; 11, 77, 249, 325;

12, 576, 582, 589, 598; 11, 480-81 (ex. 20.2), 481 n. ili, 419, 423 n. 67, 426 431 n. 3; 111, 419, 420, 426 15, 482, 483, 484, 490, No. 24 in C Minor, K. 491, i,

No. 12 in A, K. 414/385p, 1, 498, 533, 537 n. 47, 539 44, 289 n. 28, 481 n. 15, 481, 481 n. 15, 482, 483 n. (including n. 51), 540, 551, 482, 483, 484, 493, 494, 23, 498, 525, 527, 536-37, 556, 572 n. 12, 573, 585, 520 (including n. 21), 521, 544, 551, 557, 564-66 (ex. 587, 589, 599, 601; 111, 419, 533, 541-42, 544, 551, 556,

22.1a), 571 n. 9,572, 572 n. 426 564, 570, 571 n. 9, 572 n.

12, 573, 585, 593, 598, 601; No. 19 in F, K. 459, 1, 114, 201, 12, 573, 580 n. 21, 581, 585,

11, 431 n. 3; 111, 419, 420, 440, 481 n. 15, 482, 483, 595, 596, 597, 598, 599 n.

423, 427 484, 487, 494, 498, 533, 537 47; 11, 328; 111, 334, 417

Index of Works 645 No. 25 in C, K. 503, 1, 187, 482 518, 525, 526 n. 29, 527, 538 K. 157 in C, ii, 325; i11, 394-95

n. 19, 483 n. 23, 484-87 n. 49, 540, 557, 559, 572 n. K. 159 in B-flat, iti, 393-94 (ex. 20.3), 488, 494, 511 n. 12, 576, 580 n. 21, 581, 581, K. 160/159a in E-flat, 1, 236 11, 518, 526 n. 27, 533, 537 586, 588, 598; 11, 431; 111, K. 168 in F, i, 203; 11, 315, 325;

nn. 47 and 48, 539 n. 51, 401 n. 30, 419, 421, 423 iv, 335 540, 542, 544, 546, 558, n. 69, 426 K. 169 in A, 1, 29, 216 ; 11, 137 570, 571 n. 9,572 n. 12, Concertos for Piano, K. 107 K. 171 in E-flat, 1, 296, 304; 111,

573, 580 n. 21, 585, 595-96, (pastiche arrangements of 315 600; 11, 431 n. 3; iti, 419, J. C. Bach keyboard sonatas), K. 172 in B-flat, i, 39, 86, 87

426, 431 n. 5, 444, 448, 453-68, K. 173 in D Minor, ii, 328; iv,

No. 26 in D, K. 537, 552 n. 67, 556, 563 n. 1 335

“Coronation,” 1, 481 n. 15, Cosi fan tutte, K. 588, “Un aura K. 387 in G, 1, 101, 218, 226; 11,

483 n. 23, 484 n. 27, 498, amorosa,’ 249 n. 26. See also 330, 338; 111, 338, 349; iv,

527 n. 31, 528, 537 n. 47, under Mozart, overtures. 139, 163, 264, 266, 267 n. 539, 540, 544, 551, 556, Don Giovanni, K. 527, “Tl mio 21, 335, 369, 379 564, 566, 567—69 (ex. 22.2), tesoro,’ 249 n. 26. See also K. 421/4176 in D Minor, 1, 27,

571 n. 9,572 n. 12, 580 n. under Mozart, overtures. 61—63 (ex. 4.3), 111, 21, 586, 595-96, 599; 111, Eine kleine Nachtmusik. See under 212-13 (ex. 10.5); 11, 328;

419, 420, 426 serenades iv, 334

No. 27 in B-flat, K. 595, i, 187, Idomeneo, K. 366, Quartet, “Andro K. 428/421b in E-flat, 1, 27; iv,

481 n. 15, 482, 484, 493, ramingo e solo,” 143 n. 408, 409 n. 40, 411 n. 45 494 n. 37, 498, 526 n. 27, 26, 383. See also overture, K. 458 in B-flat, “Hunt,” 1, 61,

527 n. 34, 528, 537 n. 47, Idomeneo. 63, 108-11 (ex. 6.5), 214; 11,

540, 544, 546, 557, 559,570, overtures 338; 111, 338

571 n. 9, 572 (including n. Apollo et Hyacinthus, K. 38, 20 n. K. 464 in A, 1, 112-13, 140-41;

12), 580 n. 21, 585, 586, 7, 346 n. 9, 362 11, 338; i111, 338

589, 594, 596, 598, 602; 111, Cosi fan tutte, K. 588, 296, K. 465 in C, “Dissonance,” i, 47,

419, 421-22, 423 n. 67, 426 300-01 185, 191-93 (ex. 9.2a—b),

(including n. 75) Don Giovanni, K. 527, 223, 226— 217, 296, 298, 303, iv, 163,

concertos, violin, 454 27, 296, 298, 300 262-63 (ex. 12.2a—b)

No. 1 in B-flat, K. 207, i, 440, Die Entfiihrung aus dem Serail, K. K. 499 in D, “Hoffmeister,” i, 74,

481, 482 n. 18, 487, 498, 384, 219, 220-21 95; 11, 338; iti, 338

525, 556, 572 n. 12, 574-76 La clemenza di Tito, K. 621, 26 K. 575 in D, 1, 141; 11, 249; iv,

(ex. 22.3), 581, 586, 589, Idomeneo, K. 366, 47, 119, 141, 248 411 598, 599; 111, 418 n. 58, 431, Le nozze di Figaro, K. 492, 86 n. K. 589 in B-flat, i, 27, 105, 141;

573, 574, 581 20, 133, 134 (ex. 7.3b), 221, 11, 349

No. 2in D, K. 211, i, 440, 481, 226, 286, 287, 344, 347 Quintet for Clarinet and Strings in 482 n. 19, 498, 556, 572 n. Il re pastore, K. 208, 362, 385 A, K. 581, 1, 185 12, 576, 581, 589, 599 Die Zauberfléte, K. 620, 145, quintets, string No. 3 in G, K. 216, i, 482, 493 n. 219-20, 231, 296, 301 K. 406/516b in C Minor, 321 36, 498, 525, 526 n. 29,556, Quartet for Flute, Violin, Viola, K. 515 in C, 1, 207 n. 12, 289; iv,

572 n. 12, 562, 576, 580 and Cello in G, K. 285a, 1, 411 (including n. 21), 581, 588, 362, 385 K. 516 in G Minor, 1, 29, 141,

597, 598, 599; i111, 410 n. 41, quartets, piano 311; 11, 338; 111, 328, 338; iv,

419, 420, 426 K. 478 in G Minor, 11, 328; iv, 297, 301, 308, 335, 411

No. 4in D, K. 218, i, 89-90 (ex. 411 K. 593 in D, i, 136, 296, 304; iv, 5.9), 474, 482, 498, 522-25 K. 493 in E-flat, i, 91 70-71, 139, 414 n. 55 (ex. 21.5), 526 n. 29, 528, quartets, string Rondo in A Minor, K. 511, 400

539, 540, 557, 559, 572 n. K. 80/73fin G, u, 137, 139 Rondo in D, K. 485, 399 12, 580 n. 21, 581 n. 24, K. 155/134a4 in D, 1, 27, 28; 11, serenades 590, 598; 11, 431; 111, 58 n. 7, 106, 362, 369, 374-76 K. 185/167a in D, ii, 325-26 n.

421 (Ex. 17.4a—b), 385 19, 431 n. 2, 454 n. 71; v,

No. 5 in A, K. 219, “Turkish,” K. 156/1346 in G, i, 106; 11, 325-26 n. 19, 346 n. 8; vii,

i, 442, 481, 482 n. 22, 498, 325 259

646 Index of Works

serenades, (continued) K. 310/300d in A Minor, i, 101, No. 22 in C, K. 162, i, 133

K. 203/1896 in D, 11, 431, 454 n. 170; 11, 328 No. 23 in D, K. 181/1626, 111, 71; iv, 431, 454 n. 71, 557 n. K. 311/284c in D, 1, 34-35 (ex. 394-95

74, 572 n. 12, 582 3.2), 159, 187, 262 n. 11, No. 25 in G Minor, K. K. 204/213a in D, ii, 431 292, 354, 362, 369, 377, 183/173dB, 1, 111, 133; 11,

(including n. 5), 454 n. 71; 385; 11, 408 328 (including n. 26)

ili, 431, 454 n. 71, 557 n. 74, K. 330/300h in C, i, 103—05 (ex. No. 28 in C, K. 200/189k, i, 86,

558 n. 75, 582 n. 26 6.3); ii, 102 217

K. 320 in D, “Posthorn,” i, 20 n. K. 332/300k in F, i, 74, 75 (ex. No. 29 in A, K. 201/186a, 1, 66,

8, 295 5.3), 92, 95, 159-62 (ex. 217, 482 n. 22

K. 361/370a in B-flat for 8.3), 201, 319; 11, 115; 111, No. 30 in D, K. 202/186), 1, 86

Thirteen Winds (Gran 74, 95, 141, 234, 319 No. 31 in D, K. 297/300a, partita), 321; 1, 296; 11, 91 n. K. 333/315c in B-flat, i, 66, 101, “Paris,” i, 20 n. 8, 142; 111,

22; vii, 396-97 136, 159, 201, 219; ii, 105; 139

K. 375 in E-flat for Eight Winds, ii, 193-94, 405-07 No. 32 in G, K. 318, i, 186-87, 321; 1, 20, nn. 8 and 9, 211; K. 457 in C Minor, 1, 112; 11, 221, 378

111, 327 328; 111, 411 No. 33 in B-flat, K. 319, 1, 145

K. 388/384a in C Minor for K. 545 in C, 1, 95, 106—08 (ex. No. 34 in C, K. 338, 1, 20, n. 8,

Eight Winds, 321 6.4), 142, 215, 264, 265-66, 129-31 (ex. 7.2), 133, 187,

K. 525 in G, “Eine kleine 267; i11, 400 233, 243, 368 n. 38, 386

Nachtmusik,” 321; 1, 84 n. K. 547a in F, i, 201 No. 35 in D, K. 385, “Haffner,” 14, 135, 206; iv, 334, 344, K. 570 in B-flat, 1, 136; 11, 400; i, 20, 44, 201, 204—05 (ex.

362, 399 ili, 400-01 10.3), 217, 488 n. 31, 541 n.

Sinfonia Concertante for Violin K. 576 in D, i, 233-34; 111, 411 52:11, 108; iv, 113

and Viola in E-flat, K. sonatas, violin No. 36 in C, K. 425, “Linz,” 1,

364/320d, 1, 44, 481 n. 17, K. 8 in B-flat, 1, 91 n. 23, 203 47, 95, 295, 301 482, 490, 520, 529, 540, K. 9 in G, 1, 203, 259 n. 8 No. 38 in D, K. 504, “Prague,”

541, 556, 562, 582, 590-93 K. 13 in F, 203; 11, 325 320; 1, 47, 92, 142, 152-55 (ex. 22.6), 597; 111, 419, K. 303/293¢ in C, i, 299 (ex. 8.1), 162-63, 185, 285, Sonata for Piano (Four Hands) in K. 305/293d in A, i, 84n. 13 295, 298, 301, 378; 111, 320 F, K. 497, i, 296 K. 306/300/ in D, i, 157-58, 292, No. 39 in E-flat, K. 543, i, 45,

Sonata for Two Pianos in D, K. 362, 368, 369, 374, 385 69-70 (ex. 5.2), 73, 95, 113, 448/375a, 1, 41, 42—43 Symphonia Concertante in E-flat, 114, 186, 215, 289, 293-95

(ex. 3.5), 89; 111, 411 K. 2970, iti, 418 n. 58, 526 (ex. 13.2a—b), 295, 484 n.

sonatas, plano n. 27 28; 11, 288, 408—09; iv, 113, K. 279/189d in C, 1, 80-82 (ex. symphonies (standard numberings) 136, 163, 205 5.6), 85, 105, 217; 11, 233 No. 1 in E-flat, K. 16, 1, 86 n. No. 40 in G Minor, K. 550, i,

K. 280/189e in F, 1, 95-97 (ex. 20, 344, 362, 365, 369, 370, 36, 38 (ex. 3.4), 45, 77-79

6.1), 137-38 (ex. 7.5), 199— 373, 381; 11, 325 (ex. 5.5), 87, 102, 112, 113, 201 (ex. 10.1); 11, 325; 111, 105 No. 4in D, K. 19, 1, 362, 365, 163, 164-65 (ex. 8.4), 217,

K. 281/189f in B-flat, i, 105, 170; 376-77, 381, 382, 386 288-92 (ex. 13.1a—b); i, ili, 41, 409-10 n. 41 No. 5 in B-flat, K. 22, 1, 145, 317, 328; 111, 332; iv, 71, 97, K. 282/189¢ in E-flat, 1, 344, 344, 362, 369, 370-72 (Ex. 133, 135 (ex. 7.4), 288-89

362, 372, 382-83, 385 17.3a—b), 373, 385 No. 41 in C, K. 551, “Jupiter,” 1, K. 283/189h in G, 1, 95, 105, No. 6 in F, K. 43, i, 362, 373, 77, 102, 159, 223, 227, 285; 182; 11, 30; 111, 182, 105 377, 381, 382, 386; iv, 380 ii, 22; 111, 66-67 n. 5; iv, 22, K. 284/205b in D, i, 24, 30, No. 7 in D, K. 45, i, 29, 362 n. 7 80, 201, 285, 286, 287, 335 32-33 (ex. 3.1), 105; 11, 397 No. 8 in D, K. 48, i, 362 n. 7 Symphony in B-flat, K. Anh.

n. 22, 411 n. 45 No. 11 in D, K. 84/734, 111, 372, 214/45b, 1, 385

K. 309/284b in C, 1, 30, 86 n. 382 Symphony in C, K. 96/1110, 11, 325 20, 95, 142, 206, 260-62 No. 14 in A, K. 114, i, 482 n. 22 Symphony in D, K. 81/73], ii, 362 (ex. 12.1); 111, 95, 411, No. 20 in D, K. 133, 1, 47, 362, Symphony in D, K. 95/73n, iv, 362

412 369, 385-86 Trio for Piano, Clarinet, and

Index of Works 647 Viola in E-flat, K. 498, Sonata (Sonatina) for Violin and Symphony in E-flat (Wolf Eb—4), 1,

“Kegelstatt,” 11, 331; 111, 401 Piano in A Minor, D. 385, i, 362, 370

Trio for Piano, Violin, and Cello in 116, 141 Strauss, Richard G, K. 564, 111, 401-02 Sonata for Piano Four Hands in C, Till Eulenspiegels lustige Streiche, 304,

Rimsky-Korsakov, Nikolay D. 812, “Grand Duo,” 1, 237 305 n. 44

Sheherazade, 304 n. 9: iv, 237 n. 9, 259 Tchaikovsky, Pyotr Il’yich sonatas, plano Overture, “1512,” 305 Rossini, Gioachino D. 157 in E, iii, 332 Romeo and Juliet, 304 overtures, 186, 302, 347 D. 279 in C, 1, 29-30, n. 8 symphonies

La gazza ladra, 299 D. 537 in A Minor, i, 264, 317; No. 2 in C Minor, op. 17, 1, 304,

Semiramide, 299 11, 329; 111, 275 305

La scala di seta, 299, 347 D. 575 in B, i, 264 No. 4in F Minor, op. 36, 1, 305, Le Siége de Corinthe, 299 D. 784 in A Minor, i, 315 n. 18 364 Il signor Bruschino, 299 D. 845 in A Minor, i, 50 No. 5 in E Minor, op. 64, 1Vv,

D. 960 in B-flat, i, 262 305

Saint-Saéns, Camille symphonies

Symphony No. 3 in C Minor, op. No. 1 in D, D. 82, 1, 220 Vanhal, Johann 78, “Organ,” 1, 247 No. 2 in B-flat, D. 125, 1, 264 Symphony in D Minor [d1], v, 111

No. 41in C Minor, D. 417, n. 13

Sammartini, Giovanni Battista “Tragic,” 1, 277-78, 317; Symphony in F [F3], 1, 158 n. 1,

Symphony “No. 3” in D, 1, 54 n. 6, 1-111, 340 381 n. 49

358-59 No. 5 in B-flat, D. 485, i, 87, Symphony in G Minor [g2], 11, 220, 264, 267; 11-111, 332, 328

Scarlatti, Domenico, 340; iv, 71, 414 n. 55

Sonata, K. 2 in G, 355-58 (ex. 17.1) No. 8 in B Minor, D. 759, Verdi, Giuseppe sonatas, 122, 240 n. 11, 354, 368, “Unfinished,” 1, 29, 89, 91, Luisa Miller, overture, 364

374, 381 183, 220, 305, 311, 313, 317; 11, 329 Viotti, Giovanni Battista

Schobert, Johann No. 9 in C, D. 944, “Great,” 1, Concerto No. 22 for Violin in

Quartet in E-flat for Piano and 305; iv, 279 A Minor, W22/G97, 1, Strings, op. 7, 259 Winterreise, D. 911 (Lied, “Gute 548

Cello, 259Schumann, Vivaldi, Antonio Robert concertos, op. 8 nos. 1—2 (“Spring”

Trio in F for Piano, Violin, and Nacht’), 308

Schubert, Franz Symphony No. 4 in D Minor, op. and Summer” from The Four

quartets, string 120, 1, 44, 48, 68, 303; iv, Seasons), 446 (and 446—47

No. 3 in B-flat, D. 36, iv, 275 n. 364 n. 53) 30

No. 41n C, D. 46, 1, 277 n. 31 Sibelius, Jean Wagenseil, Georg Christoph No. 6in D, D. 74, i and iv, 275 symphonies concertos, keyboard, 579 n. 19 No. 13 in A Minor, D. 804, 1, No. 1 in E Minor, op. 39, 1, 304, symphonies

87 No. 4in A Minor, op. 63 1, 364 WV 374 in D, i, 235

No. 141in D Minor, D. 810, WV 393 in E,1, 64 n. 13 “Death and the Maiden,” i, Spohr, Ludwig (Louis) WV 413 in G, i, 235

89, 258, 315 n. 18 overtures WV 438 in B-flat, 1, 234-35

364 overtures

No. 15 in G, D. 887, 1, 89 Faust, 259

Quartettsatz in C Minor, D. 703, Jessonda, 364 Wagner, Richard Quintet for Piano and Strings in Stamitz, Johann Der fliegende Hollander (The Flying A, D. 667, “Trout,” i, 264; Symphony in D, op. 3 no. 2 (Wolf Dutchman), 147

v, 275 D-3), 1, 385 n. 57 Rienzi, 304

Quintet for Strings in C, D. 956, 1, Symphony in E-flat (Wolf Eb-1), 1, Tannhduser, 305, 364, 383 n. 54

70, 111, 182; iv, 349 362 n. 8

648 Index of Works

Weber, Carl Maria von Der Freischiitz, J. 277, 147, 300, “Aufforderung zum Tanze”’

overtures 301 (“Invitation to the Dance’’), Der Berherrscher der Geister, J. 122, Jubel (Jubilee), J. 245, 287 305 n. 43 364 Oberon, J. 306, 300, 304

Euryanthe, J. 291, 221 Rondo brillant in D-flat, J. 260,

Index of Concepts

This index does not locate every occurrence of the entries listed below. (Several of the terms recur regularly throughout the book.) Instead, it is a guide to their initial or central discussions, along with selected other appearances. Definitional and other key passages are indicated in italics.

8—b7-6—47-8 pattern. See under parallel, 355 (including n. 4), 359 perfect authentic (PAC), xxv, 17

closing zone (C); primary- rounded, 108-11, 330, 397-99 (fig. 2.1), 18, 66-67 n. 5 theme zone (P); transition sonata form, “binary” aspect of, attenuated, 170, 215 n. 20

(TR) 16, 19, 147-49, 366 n. 24 evaded, 169-70, 215 n. 20 medial caesura, serving as,

action-space (action-zone), 9, 16, cadence 27-29

18, 19, 23-24, 180, 250, attenuated. See under cadence, reopening (undoing closure-

616-18 perfect authentic effect), 60, 123-24, 151-63

aesthetic presence, 607—08 deceptive (DC), xxv refrain (method of EEC deferral), Anlage. See exposition, Anlage definitions, controversies over, 158-59, 492 (referential layout, model 24 n. 1-2, 28 n. 6, 66-67 n. trill, 176 n. 6, 470, 534, 546-48,

rotation), role as; Type 5 5, 106 596, 602

sonata, Ritornello 1 (R1), evaded. See under cadence, cadential span. See under phrase

Anlage (referential layout), perfect authentic cadenza. See under Type 5 sonata,

role as expanded cadential progression Ritornello 4 (R4)

anti-recapitulation. See under (ECP), 63, 121 n. 7, 131 caesura, 12, 34 recapitulation half (HC), xxv, 19, 24 (including development, end of, 197—98,

aria formats, 343, 348 n. 2) 217 da capo, 198, 203, 348 n. 17 dominant-arrival effect (MC), fill. See caesura-fill (CF) Mozart, 249 n. 26 24 final, 12 Type 5 sonata, similarity to, 430 energetic prolongation via medial (MC), xxv, 12, 17 (fig. dominant-lock (MC), 24, 31 2.1), 18, 23—50 (fig. 3.1), 117

bait-and-switch tactic (in (including n. 11) blocked, 44, 47-48, 49, 116, continuous expositions), half-cadence effect, 24, 25 132 n. 18

54—60 medial caesura (“built around” compensatory, 58

binary form. See also Type 2 sonata an HC), 24-27, 31 declined, 27, 45-47, 48, 53,

(“binary”) “tonicized,” 30-31 n. 9 175, 176 lyric, 111 opening gesture of P, 66-68 25-29

balanced, 355, 358 imperfect authentic (IAC), xxv defaults, harmonic, in major, 649

650 Index of Concepts

caesura, (continued) idées accessoires (Reicha), 123, 150, minor-mode sonatas, 312—14

defaults, harmonic, in minor, 180, 243 monumentality, 285

26-27 linkage technique (Kniipftechnik) P-onset of, 206-07, 349 definitions, 24-36 with preceding S, 182 rotational implications, 283-86 deformation, 26, 47, 49, 137 “monothematic” expositions, Schenkerian view of (“structural

deployment sequence, 25, role in. See exposition, two- coda”), 281 n. 1 36-40 (fig. 3.2), 176-77 part, “monothematic” telescoped development and double medial caesuras, multimodular, 180-81, 186 recapitulation, 22, 285 apparent. See trimodular nonelision axiom, 162, 163, 181 tonal resolutions (ESC) within,

block (TMB; apparent “non-S-ness” of, 181-83 243-47

double medial caesuras) normative compression within, Type 2 sonata, 382-86

dynamics, 31-34 principle of, 159 Type 5 sonata (concerto), 445, 597 hammer-blow convention, 34, onset of, determining, 117, 150 coda-rhetoric interpolation (CRI),

49 PMC axiom, 159 xxv, 242, 282, 288-92, 304

Nachschlag, 34 postcadential (post-EEC) aspect codetta. See under closing zone (C),

nonredundancy feature, 40 of, 180, 184 n. 2, 243 thematic types; secondarynot fully realized (ambiguity), retransition, merger into theme zone (S)

63-64 (C=>RT), 18, 193. See also computer (metaphors), 10, 452 n.

overridden, 28 retransition (RT) 69, 608-09

proportions of movement, S¢. See under secondary-theme concertos (Mozart, Beethoven). See

relation to, 25-29, 36-40, zone (S) Type 5 sonata (concerto),

53 S-aftermath, 141, 182-83 considered as a whole

proposed, 45 S references within, 181, 182-83 continuous exposition. See triple (five-module complex), thematic types, 183—87 exposition, continuous

542 “afterthought” codetta module conventions, 7-9, 614-18

troubleshooting problematic or appendix at end (piano), correspondence measures. See under

identifications, 48—50 159, 181, 187 recapitulation

postmedial (PMC), xxvi, 159, codetta, 162, 180-81, 184, 282 crux, 239-42, 258, 267

192 crescendo-module, 186—87 double, 240 40-45 “S rhetoric’), 186 postcrux alterations, 241—42

caesura-fill (CF), xxv, 34-36, new theme (sometimes with false, 240

5-1 descent type, 41, 44-45 P-based C (identifier of onset precrux alterations, 240—42 energy-loss, representation of, of C-space), 123-24, 152, customized (individualized)

40-41 162, 184-85, 372 versions of sonata-form

expanded, 47, 49, 50, 487-88 TR-based, 185-86 options, 136, 413-14, 418, (including n. 31) “zero” modules (C°, C1.°), 72-73, 610 juggernaut, 41, 44-45, 114, 538 187-90 (ex. 9.1) closing zone (C), xxv, 18, 180-94 coda (coda-space), 20, 281-92. de-energizing transition. See under

8—b7-6—47-8 pattern, 91-92, See also coda-rhetoric transition 184 interpolation (CRI); sonata- _— default, 8, 70, 181, 608—09. See also

absence of nonelided cadences space caesura, medial, defaults

or caesuras within early apotheosis, 287 deformation, 8, 10-11, 159, 529, C-space, 159-63, 181 beginning, determining, 281—83, 609, 614-21

Anhang (appendix; Koch), 180 288 connotations of term within

Cpre-EEC | xxv, 59-60, 191 cadenza-effect within, 287 Sonata Theory, 11, 674-18

cadential aspect of codetta-type, coda (or codetta) to the coda, medial caesura, 26, 47, 49, 137

184 n. 2 286, 386 n. 59 precedents of term within 150, 180 384-86 precedents of term within physics

definition, 17 (fig. 3.1), 18, 117, completion, elements of, 287-88, literary theory, 619-21

discursive. See closing zone, defined, 231, 281-83 and mechanics, 619

multimodular “development,” recapitulatory, 243, 245—49, 254,

elided (or flush-juxtaposed) inappropriateness of term, 259, 260, 279-80

cadences in, 181 284 “rule-governed,” 621

Haydn, characteristic, 59, 136, discursive, 22, 206—07, 284-88, secondary theme, 48, 117 n. 1,

146, 162, 186, 215 n. 20 349, 386 n. 59 177-79

Index of Concepts 651

deployment sequence. See caesura, Type 5 sonata, Solo 2 (S2, disability theory, 617-18

medial, deployment developmental space), display episode (DE). See under

sequence episodic Type 5 sonata, Solo 1 (S1)

development, 18-19, 195—230. See nonrotational, 207, 214-15 dominant-lock, 19, 24 (including n.

also rotation, development P-references at opening, 207-08, 2), 27, 28 n. 6, 30-34, 39

cadences within, 196—97 211, 223, 224-25, 350-52, n. 14. See also development, caesura, end of development, 372-73. See also primary- dominant-locks within

197-98, 217 theme zone (P), initiator of abandoning “(“unfreezing’’), 28

C-based openings, 215-16 rotations n. 6, 37-39, 39 n. 14

central action zone, 230 P-TR basis most common, 205, approaching, 30-31, 34 circle-of-fifths descent, opening, 217

207-12, 223-26, 372 “point of furthest remove,” 197 EEC. See essential expositional

developmental space (term), 196 n. 4 closure (EEC)

“disjunct recapitulation” “precursory recapitulation” ESC. See essential structural closure

(Bonds), role in, 223 n. 34, (Bonds), role in, 223 n. 34, (ESC) 256 256 essential expositional closure dominant-locks within, 197, “premature reprise” (EEC), xxvi, 12, 17 (fig. 198-203, 217, 224, 275-76 misconception, 207 (and 2.1a—b), 18, 120-24, 147-49

drama, topical, 228 207-08 n. 14), 223, 224, concertos, special issues in.

entry, 229-30 225, 372 See under Type 5 sonata,

episodic, 218-221 Ratz-Caplin model, 228-29, 572 Ritornello 1 (R1); Type 5 concertos. See Type 5 sonata, repeat (development- sonata, Solo 1 (S1) Solo 2 (S2, developmental recapitulation), 19-20, 21 defined, 23-24, 117, 120, 125 n.

space), episodic retransition. See retransition 14, 151-52 introductory material (from (RT) deferral, techniques of, 60-62, slow introduction) in, rotation. See rotation, 63, 123-24, 150-70, 184

219-20 developmental first PAC rule, 18 n. 6, 120—24

opening of development, S, relative infrequency of “early” PACs, 163, 166—67

212-15, 225, 373-78 appearances, 205—06 exceptions, 150—63

“slow-movement”’ episodes in S-based openings, 216-17 flexibility of, 123, 151

allegro movements, 220-21 seam (development- strength of PAC, 124, 151 tonic-centered, 219 recapitulation), 198, 203 historical support for, 121—24

exit, 230 sequence-blocks in, 196, 228-30 IAC substitutes for, 167—69 expositional repeat, beginning as strategies, typical, 196 MC (V:PAC), potential

aborted, 211, 350-52 term, history of, 195-96 confusion with, 27—29

“false-recapitulation” effect, thematic layout, 205-17, 229-30 minor-mode sonatas, 310-12,

221-28, 260, 276 tonal layout, 196-205 314-17

fill (concluding link mediating tonic key appearances near essential structural closure (ESC),

final dominant to recapitu- beginning of, 196, 207-08, xxvi, 12, 17 (fig. 2.1b), lation), 197, 199, 201, 203 211, 223-26, 256-57. See 20, 124, 232-33. See also

fugato, 259 also development, circle-of- minuet/scherzo and trio

interruption on V, at end, fifths descent, opening (as part of a multimovemnt normative, 17 (fig. 3.1), 19, tonic key in center of work); multimovement

147-49, 197-98 development, 205—06, sonata cycle ii: PAC at end, 203, 205 244-45 goal of entire structure, 17 (fig.

V/iii at end, 201-03 writing over, 207, 212-15, 358, 2.1.b), 20, 232, 250-54

V/viat end, 198-201 373-74 incongruence with EEC point, vi: PAC at end, 203 zones within, 229-30. See also occasional, 232, 234

length, connotations of, 196 Type 5 sonata, Solo 2 (S2, minor-mode sonatas, 306, 311,

link to entry, 229 developmental space), 312-17

modal shifts (minor mode), 197 event-zones role in securing (precipitating) modulatory pattern, assessing, developmental space. See under the tonic, 232. See also tonic

198 development presence (full “reality” of

“new themes” in, 196. See also dialogic form. See Sonata Theory, tonic)

development, episodic; dialogic form Type 4 sonatas, 428-29

652 Index of Concepts

essential structural closure, “three-part division” (Larsen), contract, generic, 65, 246, 311,

(continued) 39 n. 15, 51, 59. See also 337, 604 n. 2 sonata, Ritornello 4 (R4); two-part, 23-50 606-08

Type 5 sonatas. See under Type 5 exposition, continuous “form,” distinguished from,

Type 5 sonata, Solo 3 (S3, “Dutchman” type, 147 hermeneutic, 228, 608 solo recapitulation) “monothematic,” 49, 135-36, prise de position (Bourdieu), expansion section. See exposition, 140, 146, 181, 182, 185, 186 understood as, 604 continuous, subtype 1 Type 5 sonatas (concertos). Sonata Theory’s concept of,

(expansion section with See under Type 5 sonata, 9-12, 251-52, 319, 603-10

Fortspinnung) Ritornello 1 (R1), sonata Gestalt psychology, 340

exposition, 16—18. See also closing exposition, rhetorical erand antecedent, 45, 77—80, 97,

zone (C); primary-theme similarity to; Type 5 101 (including n. 6) zone (P); secondary-theme Sonata, Solo 1 (S1), larger erand consequent, 45 zone (S); transition (TR) exposition (S1 + R2) and

Anlage (referential layout, model solo exposition hierarchies of structure, nested

rotation), role as, 16—18, (sentences, periods), 69-70, 206, 231, 281, 370, 610-14. finale, 333—36. See also 124-25

See also Type 5 sonata, multimovement sonata hybrid theme (usage of term Ritornello 1(R1), Anlage cycle; Type 4 sonata (sonata- [Caplin] in Sonata Theory),

(referential layout), role as rondo) 69 n. 10, 106—08

continuous, 23, 39 n. 15, 49, culmination (finale symphony),

51-54, 117, 316 335 ideal type, 8

subtype 1 (expansion section fugal or contrapuntal, 335 idée fixe modules. See Type 5 sonata with Fortspinnung), 52—60 modal options (major/minor), (concerto), motto modules in

subtype 2 (reiterated cadences), 335-36 inert modules (rotationally) in

60-63 nonobligatory form, choices, concertos. See under Type 5

“failed,” 177-79, 251, 552 333-34 sonata, Ritornello 2 (R2), inert zones (tonally) within (P, role within the work, 334-35 rotational aspects of TR, C, uninvolved with styles, thematic and rhythmic inert zones (tonally). See under

EEC/ESC production), 216, (topics), 334 exposition

234, 243, 257,587 n. 32 first movement (as opening of a interruption, harmonic, melodic

“monothematic” (P-based). multimovement work), 321- (end of development), See exposition, two-part, 22. See also multimovement 19, 147-49. See also

“monothematic”’ sonata cycle development, interruption

nonrepeated in overtures, 20, forms/genres, social content of, 15, on V, at end, normative

346 251-54, 606-08 introduction, 292-305

referential layout. See exposition, _forte affirmation. See transition brief, in-tempo, 67 n. 6, 86-87,

Anlage (referential layout, (TR), tutti affirmation (or 292 model rotation), role as; forte affirmation). ceremonial assemblage, 302 Type 5 sonata, Ritornello Fortspinnung (FS) in continuous expressive functions, 300—04

1 (R1), Anlage, (referential expositions, xxvi. See also “false-start,” 299

layout), role as exposition, continuous, generative type, 68, 298-99,

repeat convention in, 20-22 subtype 1 (expansion section 303-04

repeat-feint variant of Type 3 with Fortspinnung); vitalism introduction-coda frame,

sonata exposition, 350—52 metaphor (in Haydn) 304-05

role, structural (harmonic, length, 292, 295 rhetorical), 16-17. See gendering of thematic modules. See less common in chamber music,

also exposition, Anlage primary-theme zone (P), solo sonatas, finales, 295—96 (referential layout, model “masculine”; secondary- merged with sonata proper rotation), role as; rotation theme zone (S), “feminine” (included in expositional structure of promise, 17 (fig. genre, 7-12. See also listener repeat), 299

2.1b), 18, 310-17 (analyst), role of; Sonata minor-mode within (“the fall”;

177 as dialogue) 301-02

“three-key exposition,” 120, 171, Theory, dialogic form (form “fallen world”), 297,

Index of Concepts 653

representational, 300-01 at the moment of PAC, 170 part of a multimovement representation of composer (self- “the fall” in slow introductions, work)

referential), 302 297-98, 301-02 balances, shared motives among

representation of projected metaphorical implication movements, 319

narrator, 304 (minor-mode sonata), broader ESC deferral to finale, “run-on,” 68, 298 306-07 337

slow, 68, 198, 219-20, 295-305 “modal failure” in exposition and cyclic integration/organization,

topics within, 297 recapitulation, 179, 306—07, 319

Volksgeist type (nineteenth 313, 315-16 four-movement pattern, 337-40 century), 304 ombra topic, 298, 308 narrative implications, 336—42

zones within, 297-99 saturation, 313-14, 315 number of movements (two, Sturm und Drang, 196, 228, 308 three, four, or more), K6échel numbers, traditional, viii, transformation into major (modal 319-21

469-70 emancipation; per aspera ad origins of four-movement astra), 306—07, 309, 310-14, “norm, 320

labeling/numbering, thematic 315, 327 run-on movements, 319, 335 (thematic/modular minuet/scherzo and trio (as three-movement pattern, 336

designations), 71—73, 171. part of a multimovement through-composition, 319,

See also Type 5 sonata work), 329-33. See also 335 (concerto), labeling/ multimovement sonata cycle

numbering modules, Sonata connotations, expressive/ narrative implications. See

Theory method of cultural, 331 multimovement sonata

“lights-out” effect. See minor correspondences with first cycle, narrative implications;

mode, “lights-out effect” movement, 330 sonata form, narrative

listener (analyst), role of, 53, 60 deformational, 275, 329, 330 character of n. 12, 246, 251-54, 300, ESC, deferral of, 333 neo-Riemannian theory (R, P, L

311, 318, 319, 337, 340-42, finale, 329-30 operations), 27 n. 4, 199,

606-08 “German dance,” 331 203

“loops,” Mozartian, 80-86, 91, 455 minor-mode, 331 norm, 7—8, 11, 469, 614-18 cadential (and noncadential) nontonic, 331-32, 338-40 “normative” (connotations of the

implications within, 80, (proto-) sonata form in, 330 term), 614-18 84-85 (including n. 16), 106 restorer of tonal order (return to

n. 7, 166, 490-92, 512 n. 16 tonic key), 330 off-tonic sonata. See sonata form,

compound and simple, 84 scherzo as social critique, 331 off-tonic styles, differing, 331 overdetermination, tonal, 73—77, meaning, multiple strata of (within switched to second (movement) 95, 114, 234, 251, 252 n.

musical works), 608 position, 320, 330, 337-39 32, 425, 483, 512. See also medial caesura (MC). See caesura, trio, key and character of, underdetermination

medial (MC) 332-33. overture, da capo (or reprise), 221,

minor mode, 306-17 “mirror form.” See under Type 2 378. See also repeat signs, affective properties and sonata (“binary”), “reversed large-scale, overtures

implications, 307-10 recapitulation” fallacy absence of repeats in

binary opposite of major, 307—08 module, 15-16, 69 n. 10, 94

burden, extra, 306-07, 309, 315, movements. See multimovement parageneric spaces, 251-305. See

327 sonata cycle; first also coda (coda-space);

dualist theory, 307—08 movement; slow movement; introduction

EEC keys in minor-mode sonata minuet/scherzo; finale period, 69 expositions, 310-17 multimovement sonata cycle, dissolving consequent (beginning

finale, 335-36 318-42. See also finale; first TR), 102 procedures, 111 a multimovement work); 106 n. 8

license for unusual formal movement (as opening of relation to hybrid, problematic, “lights-out” effect, 25-26, 31, 50 minuet/scherzo and trio (as sentential, 69-70. See also

n. 23, 208, 234, 297, 308, part of a multimovement hierarchies of structure,

311-12 work);slow movement (as nested (sentences, periods)

654 Index of Concepts

phenomenology, 340, 604, 616-18. multimovement work), “precursory” (Bonds). See under

See also listener (analyst), tonic minor (“prison-house development

role of effect”’) P-TR alterations, 235-37

phrase R1. See Type 5 sonata, Ritornello qualifications for (onset of new

cadential span, 250 n. 28 1 (R1) rotation), 231-32

definition, controversies over, recapitulation, 19-20, 231-80 recapitulatory rotation. See

66-67 n. 5, 69 n. 10 “alienated” S-modules (tonally), rotation, recapitulatory definition in Sonata Theory, 69 245-47, 277-78 recapitulatory space (term), 232 n. 10, 250 (including n. 28) anti-recapitulation (deformation), recomposed (Haydn), 233, 241,

in sentences, concept of, 106 249, 259 243, 388, 404, 416-17

point of conversion (in continuous correspondence measures, referential measures, 241—42

expositions), 52—60 (fig. 4.1) 239-42, 255, 258 reordering of modules, 233-34

primary-theme zone (P), xxvi, 18, crux. See crux response to “problems” in

23-24, 65-92 defined, 19-20, 231-32 exposition, 238-39

8—b7—6—47-8 pattern, 91-92 “disjunct” and “nondisjunct” “reversed recapitulation”

function (role in sonata), 65 (Bonds), 256-58 fallacy. See Type 2 sonata erand antecedent, 77-80 double-recapitulation effects, (“binary”), “reversed

idée fixe modules in Mozart’s 279-80 recapitulation” fallacy concertos. See under Type “double return” claim (P and “secondary development” (in

5 sonata (concerto), motto tonic key), 260, 343, 367 TR), inappropriateness of

modules in exposition type altered, 238 term, 236-37

initiator of rotations, 65, 207, fallacy of out-of-tonic S- “sonata principle” (Cone),

214, 231-32, 256, 354, 382- “resolution” (“closer 242-45, 400 n. 28

86, 511 relation,” “fifth- structure of accomplishment, 17

launch types (strong and weak), transposition”), 245 (fig. 2.1b), 19 65-66 “false starts” of recapitulation, subdominant (or flat-side) shift, “loops,” Mozartian, 80-86, 91. 206, 260-62, 268, 406 P or TR, 235-37, 239-40 See also “loops,” Mozartian “false starts” of S, 238 synecdochic strategy, 233

“masculine,” 145—47 interpolations in, 234-35 tonal resolution in. See tonal “narrative subject” or dramatic MC alterations, 237-39 resolution situation, 65, 251-54 minor-mode sonatas, 306—07, tonic-minor openings (in major-

overdetermined (tonally), 73-77. 312-14 key sonatas), 258-59 See also overdetermination, “mirror” recapitulation transition. See under

tonal fallacy. See Type 2 sonata recapitulation, P-TR

P=>TR merger, 85-86, 95, (“binary’’), “reversed alterations; transition (TR),

115 recapitulation” fallacy recapitulatory

P? codettas, 74-77 non-normative openings, 255—8&0 truncated, 232, 247-49, 255

Pgen, Prel, 92 nonresolving (“failed”’), 177-79, Type 2 sonata, inappropriate Prf (in Type 4 sonatas), 390, 245-49, 251, 254, 255, term within, 232, 353-55,

404-05 279-80 See also exposition, 359. See also Type 2 sonata

repeat signs in, 70-71 “failed” (“binary’’), “reversed

structure, 69-71 nontonic (off-tonic) openings, recapitulation” fallacy

topics, 65 260-79 “wrong-key” starts for S, 238

types in Haydn’s piano sonatas, in IV (subdominant), 262-68 refrain cadence. See cadence, refrain

68—69 in V (dominant), 275-79 (method of EEC deferral)

types in Mozart’s concertos, in VI (submediant), 268—75 regulative ideas/principles, 8,

481-82 in other keys, 279 605-06. See also forms/

underdetermined (tonally), nontonic S or “ESC- genres, social content of; 73-74. See also substitute” (deformation), genre; Sonata Theory, forms

underdetermination, tonal 245-47 (genres)

“zero” modules (P°, P1-°), 66, 67 postcrux alterations. See under repeat signs, large-scale, 20-22

n. 6, 71, 72-73, 86-91. crux codas, including, 282

“prison-house effect.” See slow precrux alterations. See under overtures, absence of repeats in,

movement (as part of a crux 20, 346, 351

Index of Concepts 655

Type 1 sonatas, absence of repeats retransitions (to refrain), Tovey, use of term by, 247

in, 346 importance of, 388, 398 (including n. 21).

Type 3 sonata with expositional- rondo-variation (variation- writing over, 207, 212-15, 358,

repeat feint, 350—52 rondo), 401 n. 29 373-74, 613

retransition (RT), xxvii. See also rotational basis, 390-92, 402-03, rotational form, 16 n. 5, 323. See

Type 4 sonata (sonata- 404 also rotation rondo); Type 5 sonata, rounded binary format of refrain,

Ritornello 3 (R3) common, 391, 397-99, 404 S! (distinguished from $1), 452 development, end of, 19, seven-part, 401-02 S1. See Type 5 sonata, Solo 1 (S1) 197-205, 230 symmetrical seven-part, 402—03, Schenkerian analysis, 3, 5, 21, 112

exposition, end of, 18, 180, 404 n. 14, 120

191-94 tonic refrain (normative), 388, coda, structural, 281 n. 1

“reversed recapitulation” fallacy. 403 Eroica, 122 n. 9 See under Type 2 sonata rotation, 16-18 (including n. 5), implications for Sonata Theory, (“binary”) 611-14. See also Type 4 147—49 (ex. 7.7)

rhetorical form, 23 sonata (sonata-rondo); Type linear fifth progression (Zug)

“ritornello” (inappropriateness of 5 sonata within S, 129, 131, 147-49, term in Types 1, 2, 3, and archetypal principle of musical 170

4 sonatas), 268 n. 23, 370 structure, 612 narrative implications, 251 nn. n. 39 coda, 283-84 29 and 31

ritornello (in Type 5 sonata), 345, concertos. See under Type 5 repeats, expositional, 21 430-31. See also under the sonata [Ritornello 2 (R2), sonata form, conception of, 5,

several separate entries of Type Ritornello 4 (R4), Solo 147-49, 197 5 sonata (Ritornello 1, 1 (S1), and Solo 2 (S2, subdominant recapitulation,

Ritornello 2, etc.) developmental space)] 265-67 nn. 20-21

rondeau, 95, 389-91 (table 18.1), “cyclic[al] form/organization,” V1 as third divider in minor-

392-97 613-14 mode sonata expositions,

brevity of refrain, 391, 392-93, developmental, 19, 205-20, 230 317

404 double or triple half-rotation, V/vi at end of development,

couplet (term), 390, 392, 398 218 198-99

expanded, 393, 396-97 double rotation, 218, 226 “ZPAC,” 147-49 multicouplet, 393-94 full rotation, definition, scherzo. See minuet/scherzo rotational basis, 390-92, 395-96, 206-07, 217, 229 and trio (as part of a

404 half-rotation, 207, 214, 217, multimovement work)

symmetrical three-couplet, 229 sculpture, music as temporal, 616 394—96, 404 incomplete rotation (blocked), secondary-theme zone (S), xxvi, 18, two-couplet, 393 217 23-24, 117-79 rondo, 388-92, 397-403 rotational norm for appendix (codetta: S? or S3),

character of Pf rondo theme developments, 206—07 152, 157-58, 162, 163, 183, (playful, popular, contredanse, expositional, 16-18, 23 282

etc.), 351, 398-99 foundational axiom of boundaries of S-space in the definitions, standard textbook, interpretation, 613 recapitulatory rotation,

389-90 recapitulatory, 19, 227, 228, determining 234, 237

elimination of one midmovement 231-32, 255-58, 260-70. See breakdown (collapse), 190-91

refrain, 401 also recapitulation, nontonic concertos. See under Type 5 episode (term), 390, 398 (off-tonic) openings sonata [Ritornello 1 (R141),

five-part, 400—01 referential, 23. See also Type 5 S-zone (R1:\S); Solo 1(S1), form as invitation to wit, high sonata, Ritornello 1 (R1), larger exposition (S1 + R2);

play, 407, 413 Anlage (referential layout), Solo 1 (S1), S-zone; Solo

“mislabeled” works (“rondos” role as 3 (S3, solo recapitulation), that are not rondos), 399 reverse direction unlikely, 216 larger recapitulation (S3 + nontonic recurrences of refrain term (“rotation”), 613-14 R4, the final rotation)| (non-normative), 388 n. 1, tonality irrelevant in deformation, 48, 117 n. 1,

403, 407 (including n. 37) determining, 612 119-20, 120 n. 5, 136-37

656 Index of Concepts

secondary-theme zone, (continued) P-based S, 135-36, 146. See early descriptions of, 14-15

dynamics, 36, 131-32, 136-37 also exposition, two-part, “game” aspect of, 9-10, 432,

“feminine,” 117-18, 131, 145-47 “monothematic”’ 606—07, 617

function (active zone producing tonal choices for, 119-20, 310-17 generative conception of, 8. See

the EEC), 117-18, 216, 217, tonally migratory, 120 also sonata form, process,

234, 314, 354, 463 trimodular (TMS), xxviti, 139, considered as

historical discussions of, 118-19 172 “meaning, ’ underdetermination length, variable (proportions of “zero” modules (S°, S1-°), 72—73, of. See underdetermination,

S and C space), 124-25 142-45 imagery or programmatic (fig. 7.1) sentence, 69, 105-06 analogues (“meaning”’)

linear descent (fifth progression), concept (Sonata Theory metaphor for human action, 15, 5— 4— 3-3-7, 129, 131. See alternative), 84 n. 14, 106 n. 177-78, 251-54

also Schenkerian analysis, 8 modular assembly, 15—16

linear fifth progression (Zug) “loops,” 80-86. See also “loops,” “multimovement work in a single

within S Mozartian movement.” See sonata

major S in minor-mode sonatas, presentation, 39 n. 14, 105-06, form, “double function”

315 n. 18 166 (multimovement work in a

mediant key (III) in major-mode — slow movement (as part of a single movement”), source of

sonatas, 119-20 multimovement work), narrative character of, 15,

m.g. (Mittelgedanke), 119 322-29. See also multi- 250-54, 306-07, 312-16,

(including n. 4) movement sonata cycle 336-42, 606—08

minor-mode modules in, 141—42 early descriptions of, 323 network of conceptual forces,

multimodular (MMS), xxvi, mediant (III) or submediant (VI), 606

139-40, 172, 190 325-26, 328 off-tonic, 301 n. 40

P-based. See under secondary- nonobligatory form, 322—23 origins of term, 14-15, 343 theme zone (S), thematic nontonic major (“escape”’ key, process, considered as, 10-11,

types usually IV or V), 322, 616-18

P-material appearances within, 323-24, 327, 328, 329 “programmatic” readings,

140-41 switched to third (movement) 251-54, 259, 341-42.

reappearances of S-material position, 320, 330, 337-39 See also Sonata theory,

midway into a presumed tonic major, 327 hermeneutic readings

C, 234. See also Type 5 tonic minor (“prison-house (metaphors for flexible sonata [Ritornello 1 (R1), effect”), 325, 327, 328-29, narrative actions)

S-zone (R1:\S), definer 339 textbooks, twentieth-century of subsequent S-space; truncated recapitulations in, war against the, 6—9

Solo 2 (S2, developmental 249 twentieth-century approaches space); Solo 3 (S3, solo sonata deformation. See sonata to, 1-9. See also Index of

recapitulation), larger form, deformation Names: Caplin, William

recapitulation (S3 + R4, the “sonata failure.” See recapitulation, E.; LaRue, Jan; Ratner,

final rotation) | nonresolving (“failed”) Leonard G.; Rosen, Charles;

restatements, enhanced, 129 sonata form Schenker, Heinrich;

Ss, xxvii, 59 n. 11, 190-91, conformational conception of, 8, Schoenberg, Arnold; Tovey,

215-16, 312 10, 615-16 Donald Francis

structure, 124-31 current concepts of, 3-6 world view of (Enlightenment), submediant key (v1 or VI) in deformation, 17, 177-79, 245-49, 15, 21 major mode, 120 254, 259, 260, 614-18. See “sonata principle.” See under

thematic types, 131-42 also deformation recapitulation, “sonata bustling, energetic, 132-33 dialogic form. See Sonata principle” (Cone)

contrasting derivation from P, Theory, dialogic form (form —sonata-rondo. See Type 4 sonata

136 as dialogue) (sonata-rondo)

forte S, 36, 136-37 “double function” sonata-space, 243, 251, 281-83, fugal (imitative, learned style), (“multimovement work in a 288, 382, 600-02

137-39 single movement’’), source Sonata Theory lyrical, cantabile, 133-35 of, 221 analysis alone insufficient, 603

Index of Concepts 657

blending of differing strains of rotational Type 4s in Haydn, topics, 65, 228, 253, 297, 308, 334,

thought, 603—05 417 346, 481-82

cadences, importance of, 12-13, short-circuiting analytical Hungarian, 397

18, 20, 73 thought with a prefabricated Romany (“gypsy”), 397

dialogic form (form as dialogue), term, avoiding, 224, 244 Sturm und Drang. See minor 9, 10-11, 214, 244, 251-54, sonata-type assessments, 346, mode, Sturm und Drang

306, 319, 340-42, 343, 605, 348 “Turkish,” 397, 401 n. 30

609, 614-18 flexibility of compositional transition (TR), xxviii, 18, 23-24,

form as process, 10-11, 616-18 procedures, 15 93-116

forms (genres) not “real” but genre theory, 604-10 8—b7—*6—-47-8 pattern heuristic, regulative guides hermeneutic readings (metaphors (dissolving), 103—05

for interpretation, 8, 343, for flexible narrative beginning, determination of, 386-87, 390, 436, 443, 469, actions), 15, 177-78, 246, 94-95 535, 604-07, 610 248-49, 251-54, 306-07, de-energizing (energy loss), 25,

flexibility in analysis, 315, 341-42, 429 31, 44, 48, 49, 116. See also

terminology, and labeling/numbering of themes caesura, medial (MC), blocked

interpretation and modules. See labeling/ developmental, 95-101

analyzing developmental numbering, thematic dissolving types, 70, 73, 95,

rotations, 206 (thematic/modular 101-11

coda and CRI, onsets, designations) energy-gain, 25, 93-94

282-83, 288 language of (vectored), 603 forte affirmation. See transition

differing analyses possible, 72, phenomenology, 340-42, 604 (TR), tutti affirmation (or 87, 167, 258, 408, 492, 534 reader-response theory, 604, 620. forte affirmation)

n. 20, 608 See also Index of Names: Fortspinnung merger. See EEC ambiguities, sensitivity Iser, Wolfgang transition (TR), TR=>FS to, 151, 167, 492 sociological theory, 604, 606-08 (in continuous expositions) explicating the ambiguity or sonata types, 344-45. See also under independent, 95

multiplicity of implications, Type 1 sonata (without merged and non-merged, 95. See 112, 159, 170, 224, 244, 267, development); Type 2 sonata also primary-theme zone

277, 348-49, 387, 429, 519, (“binary”); Type 3 sonata (P), P=>TR merger

602, 610 (“textbook”); Type 4 sonata minor-mode intermixtures,

“facts” not objectively within (sonata-rondo); Type 5 25-26, 31

the music (constructed sonata modulation not a necessary hermeneutically by analyst), | symphonic poem, 20, 221, 304, 351 feature of, 93-94 167, 253-54, 300, 318, 340, symphony, da capo. See overture, da off-tonic beginning (no need to

343, 534n. 39 capo (or reprise) resolve in recapitulation),

individuality of works, at- 74, 95, 111-13, 243, 529 tentiveness to, 427, 607-08 theme (usage of term), 65 n. 1. problems with term, 93-94

intersecting heuristic See also closing zone (C); recapitulatory, 235-37 categories of classification, primary-theme zone (P); sujet libre (in sonata-rondo

132, 297, 408 secondary-theme zone (S) concerto movements),

labeling alone, insufficiency tonal form, 23 425-27. See also under Type of, 112, 151, 167, 224, 241 tonal potential, 250—51. See also 5 sonata, Solo 1 (S1), sujet

mechanical or simplistic tonic presence (full “reality” libre (TR type)

labeling, avoiding, 7, 12-13, of tonic) TR=>FS (in continuous 59 n. 9, 63-64, 72, 87, 112, tonal resolution, 19, 117, 242-47, expositions), xxviii, 52-60

151, 159, 453, 608, 610 255, 353-54, 380 (fig. 4.1)

nuance (hermeneutic art), tonic. See also tonic presence (full tutti affirmation (or forte

12-13, 59 n. 9, 63-64, 167, “reality” of tonic) affirmation), 94, 95, 97, 101, 224, 241, 300, 340-42, 364, provisional, 73, 232, 250-51 113-14, 232

603-05 secured via cadence, 73-74 trimodular block (TMB; apparent

phrase analysis (sentences, tonic presence (full “reality” of double medial caesuras),

periods, hybrids, modules), tonic), 73, 232-33, 246, xxvii, 27, 40, 46 n. 21, 48,

69 n. 10, 106 n. 8 250-51, 283, 337, 396, 429 111 n. 13, 113, 120, 170-77.

658 Index of Concepts

trimodular block, (continued) completion-effect, 377, 398-99, 411 n. 46, 414, 416,

See also Type 5 sonata, Solo 384-86 421

1 (S1), TMB (trimodular coda (or CRI), P-based, 362, deformation, 412, 413-17

block) options 363, 369, 372, 382-86 differences from rondeau and differences from trimodular S crux-point in, 267, 353, 355, 359, rondo, 388, 391 (table 18.1),

(TMS), 139-40 379-80 394, 402, 404

five-module block (triple MCs, defining moment in determining ESC issues (double perspective),

concerto recapitulations), Type 4 format, 365, 369, 428-29

542, 595-96 379 expanded Type 1 sonata-rondo

importance in Mozart’s piano deformation, 364, 376 mixture (Type 4!-exp), 405, concertos, 171, 453, 535, developmental options (first half 409-12

537-40 of Rotation 2), 372-80 exposition format as defining

labeling/numbering. See differences from Type 1 and feature, 344, 388, 404-05 labeling/numbering, Type 3, 350, 353, 369 exposition never repeated, 388, thematic (thematic/modular double-rotational, 353—55, 362 n. 404, 414, 415

designations) 9, 369, 378, 384 Haydn’s treatments of, 413-17

MC declined, similarity to, 175, episodes in Rotation 2, 373-76 Mozart’s concerto finales, 417—27

176 generic options, 369-86 expanded Type 1 sonata-rondo

“S” question within, 172, 175 less common than Type 3, 366, mixture in, 419-20

“three-key expositions” 373 hybridization, 418, 421 (Schubert, etc.), relation to, mixtures with type 3, 258, 262 “ritornello” (term) 420-25

171, 177 n. 11, 266-67, 377-78 sujet-libre transition (solo re-

TM!=>TM? merger, 172 musicological treatments of, entry), 425-27

tutti affirmation. See transition prior, 365—69 tutti-extensions to Pf (TR), tutti affirmation (or not an abbreviated Type 3 sonata, (“ritornello” character), forte affirmation) 366 420-25

tutti interjection (TI), xxvii. See also “polythematic binary form,” 368 Type 3 sonata-rondo mixture

under Type 5 sonata, Solo 1 “recapitulation,” in, 418-19

(S1); Type 5 sonata, Solo 2 inappropriateness of term, Type 5 intermixture, 418,

(S2, developmental space) 350 n. 24, 353-55, 359, 420-25

two-part exposition. See exposition, 380-81, 581 n. 24 P-refrain (P* label), 390, 404-05

two-part repeat signs optional, 369-70 retransition (RT) as sonata-rondo

Type 1 sonata (without “reversed recapitulation” fallacy, marker, 191, 193-94, 405,

development), 191, 249, 232, 292, 344, 353-55, 414 344, 345-52 368-69, 380, 382-86 rotational basis, 390-92, 404-05,

aria form, resemblance to, 348 “Scarlatti sonata form,” 367 408, 410 differences from Type 2, 350, tonal resolution in, 350 n. 24, rounded binary format of P*,

353 353, 380 common. See rondo,

double-rotational, 345—46, 407 Type 2=>Type 3 conversion, rounded binary format of

expanded, 257, 349-50, 409 377-78, 387 refrain, common

origin, historical, 348—49 Type 3=>Type 2 conversion, Type 1 sonata-rondo mixture

P-based coda, 349, 408 373, 376-77, 387 (Type 41), 401, 405, 407-09 repeat signs, absence of, 346 Type 3 sonata (“textbook”), 16, Type 3 sonata-rondo mixture,

“slow-movement form” (Rosen), 344. See also repeat signs, 405-07

347 large scale, Type 3 sonata Type 4/Type 3 hybrids (Haydn),

“sonatina form,” 346—47 with expositional-repeat 413-17 terminology, historical, 346—47 feint Type 5 sonata (concerto),

Type 2 sonata (“binary”), 191,197, | Type 4 sonata (sonata-rondo), considered as a whole, 234,

244 (including n. 16), 256, 45-46, 243, 344-45, 388- 345, 430-602

257, 262, 264, 344, 353-87 92, 404-29 cadenza. See under Type 5 sonata, “binary’/” binary variant” label, alphabetic notation (e.g., Ritornello 4 (R4) 353, 355, 363 n. 11, 365-69 ABACABA) inappropriate, coda. See under Type 5 sonata,

coda, generically unnecessary, 344-45, 390, 404-05, 410, Ritornello 4 (R4)

362, 381-82, 386 413, 419-20 formal freedom in, 432, 436

coda, compensatory or character of P' rondo theme, 351, four-ritornello (seven-part)

Index of Concepts 659

structure as heuristic norm Ritornello 2 (R2), interdependent with a following in Sonata Theory, 442, rotational aspects of; Type Solo 1, 444, 452-53, 462,

443-45 5 sonata, Ritornello 4 (R4), 496, 534

historical origin, 430-31, 433-35, rotational aspects of Ktister model, 473-74

443, 444 participatory slots (rotational Leeson-Levin model, 472-74

hybrid (concerto and sonata), passages). See under Type 5 length, normative, 468, 471

430, 435 sonata, Ritornello 2 (R2), MC (R1:\MC) and caesura-fill,

idée fixe. See Type 5 sonata, motto rotational aspects of; Type 484-88

modules in 5 sonata, Ritornello 4 (R4), modular descriptions of, prior,

individualized structures in rotational aspects of 471-75

Mozart, 432, 470-72, 594 sonata form terminology modulations (modulatory feints)

individual vs. group (social) applicability of in Sonata within

connotations, 446, 496-97, Theory, 431, 434-35, 443- R1:\S-space, in (nontonic

526, 529, 533, 552, 600-02 45 openings of), 439, 488-90 inert slots. See under Type 5 inappropriateness of restricting R1:\TR-space, end of (K.

sonata, Ritornello 2, only to solo sections, 440— 503/1), 484-87 (ex. 20.3) rotational aspects of 41, 443-44, 462, 473 nonmodulating character of, 345, Koch’s descriptions (1793, 1802) twentieth-century 430, 435, 439, 447, 450-51, of, 436—50, table 19.1 musicological suspicions of, 453, 496. See also Type 5

actor-chorus thesis, 497 434, 440-41, 472, 474-75 sonata, Ritornello 1 (R1), inadequacy of, 440, 475, 579 (including n. 13) modulations (modulatory labeling/numbering modules, teleological genesis toward feints) within Sonata Theory method of, “synthesis” of recapitulatory omitted in mid- and later-

431, 432, 451-53 rotation (S3 + R4), 445, nineteenth century, motto modules in (idée fixes; 493. See also Type 5 434-35

“wild cards’), 166, 482-83. sonata, Solo 3 (S3, solo P-zone (R1:\P), 475-83 See also under Type 5 sonata, recapitulation), larger initial module, types, 482 Ritornello 1 (R1), P-zone, recapitulation (S3 + R4) “loops,” 482. See also “loops,”

motto (idée fixe; “wild “tutti (synonym for ritornello), Mozartian

card’’) 45—47 motto (idée fixe; “wild card”),

Mozart’s K. 107 adaptations of Type 1 sonata, mixtures with, 482-83, 587, 594, 595, 597 J. C. Bach keyboard sonatas, 431, 438 (table 19.1, subtype overdetermined tonally, 483

453-68 P), 443, 563 rotation initiator in subsequent

paradox of Mozart’s practice, Type 2 sonata, mixtures with, sections, 511, 518-20. See

470-71, 594 431, 438 (table 19.1, subtype also primary-theme zone

participatory slots within. E), 443, 467, 468, 563 (P), initiator of rotations See under Type 5 sonata, Vogler’s description (1779) of, R1:\ label, 452-53

Ritornello 2 (R2), 435-36, 437 (table 19.1, solo introduction to (or rotational aspects of; Type subtype C), 443. See also participation in), 471 n. 4 5 sonata, Ritornello 4 (R4), Index of Names: Vogler, sonata exposition, rhetorical

rotational aspects of Georg Joseph (Abbé) similarity to, 447, 450-517, ritornellos as a defining feature Type 5 sonata, Ritornello 1 (R41), 471-72

of, 430, 436-50, 469 430, 469-95 Stevens model, 474-75

formats (tutti-solo layouts), Anlage (referential layout), role structural functions, three,

differing, 436-50 as, 345, 447-49, 451, 452, 449-51, 471

“pillars,” 436 (including n. 22), 462, 471, 496, 535, 577. structural status (conceptually

443, 446, 470, 472, 573, See also exposition, Anlage prior to S1 or derivative

574, 576, 585, 596 (referential layout, model from it?), 447-49

“ritornello” (term), 430, 445-47 rotation), role as S-zone (R1:\S)

rotational aspects of, 451, 469, continuous-exposition format definer of subsequent S-space

579 (less common), 487-88 (in later solo and ritornello

inert slots (areas for inserted C-zone (R1:\C), 493-94 sections), 234, 463, 493, 534,

nonrotational passages), EEC-effect within (R1:\EEC, 587 n. 32 482-83, 525, 552-58. including deferrals), 453, implications for later rotations,

See also Type 5 sonata, 455, 462-64, 469, 488-92 492-93

660 Index of Concepts

Type 5 sonata, Ritornello 1, subtype A), 439, 440-41, replacement themes (for those in

(continued) 572, 573-577 R1), 470, 536-42

“loops,” 490-92. See also status of, problematic (R3=>S3), Rotation 2 (S1 + R2)

“loops,” Mozartian 440-41, 585 ambiguous onsets of, 498, 511,

multimodular, 461, 468, 490 “vestigial,” description as 518-20

nontonic opening. See under (Berger), 441, 574, 576 inert (nonrotational) slots, 525. Type 5 sonata, Ritornello Type 5 sonata, Ritornello 4 (R4), See also under Type 5 sonata

1 (R1), modulations 596-602. See also Type 5 subentries, rotational aspects

(modulatory feints) within sonata, Solo 3 (S3, solo of

refrain cadences within, 492. recapitulation), larger thematic (modular) freedom

See also cadence, refrain recapitulation (S3 + R4, the of, 521

(method of EEC deferral) final rotation) S1:\ (Solo 1) label, 452-53,

TR-zone (R1:\TR), 483-84 cadenza, 439, 442, 468, 470, 596, 518 n. 19

Type 5 sonata, Ritornello 2 (R2), 600-02 solo entry, 496-521

548-62 coda, 445, 597 anacruses, expanded, 512

affirmational aspect of, 548 divided into R4!, R4 (pre- and fill, “thematic,” 516-18 deceptive cadence at end, 559 postcadenza), 445, 468, preface, 498 n. 8, 511, 518-20

deceptive cadence at opening 596-97 R1:\P (restatement), 498 (rare), 548 ESC issues, 469, 493, 534, replacement theme

deformational, 558 596-602 (suppression of R1:\P), 520 “dialogue” with soloist, implicit R2 material, relationship to, 596, “warm-up, nonthematic, 512,

(social discourse), 548, 597-600 516

552 restoration of previously solo exposition, 345, 444, 462, EEC issues (R2:\EEC), 462-64, suppressed modules, 596, 468, 498, 548, 579. See also

468, 552 597-600 Type 5 sonata, Solo 1 (S1),

half-cadence endings, 559-62 rotational aspects of larger exposition (S1 + R2) larger exposition (S1 + R2), 444, inert modules within, 444 n. “sonata-clock,” stopping, 512,

462, 548-52, 579 45, 467, 468, 596, 5 518-19 modulatory, 559-62 97-600 sujet libre (TR type), 522, new-material openings, 556—57 participatory modules within, 525-29, 537

R1:\C-material openings, 557 444 n. 45, 468, 596 S-zone

R1:\P openings, 556—57 six-four platform, 439, 467, 470, multimodular zones, frequency

R1:\S-material openings, 557-58 596, 600-02 of, 468, 535

552-56 496-548 533 n. 36, 536 rotational aspects of “Devisen”-Ritornell (inappropriate “new themes” (replacement

R1A\TR!! openings (common), Type 5 sonata, Solo 1 (S1), 430, new continuation of R1:\S?1,

complete or incomplete, term for S1:\TH), 521 n. 24 themes), 535-42

551-52 dialogue with orchestra (social RAS absent from S1, inert modules within, 444 n. discourse), 446, 496—97, 540-42 45, 465-66, 468, 552-58 522, 533 relationships with R1:\Sparticipatory modules within, display episode (DE), xxv—xxvi, space, varying, 535—42, 587

444 n. 45, 468, 552-58 470, 534, 542-48 n. 32

Type 5 sonata, Ritornello 3 (R3) EEC issues (S1:\EEC), 462-64, S-chain, 535, 541-42

recapitulatory launch option, 469, 534, 543-44 TMB (trimodular block). See 437 (table 19.1 subtype B), Eingang (“lead-in”), 498, 511, 516 Type 5 sonata, Solo 1 (S1),

440-41, 445 interdependent conceptually TMB (trimodular block)

double-start recapitulatory with a preceding R1, 444, options

openings (R3+S3), 582-85 452-53, 462, 496, 534, TMB (trimodular block) options,

R3=>S3 merger 535-42, 580 535, 537-40, 542

recapitulatory openings, larger exposition (S1 + R2), 444, R1AS as TM! (unusual),

440-41, 585 462, 548-62, 579 540

retransition (RT) option at end medial caesura (S1:\MC), 525, R1:\S as TM3?, 537-40

of development (Koch 533-34. See also Type 5 R1:\S suppressed completely,

1793), 437 (table 19.1 sonata, Solo 1, TMB 542

Index of Concepts 661

transition (S1:\TR) complexity, conceptual, 594 underdetermination dissolving P=>TR merger, double-start recapitulatory imagery or programmatic

533 openings (R3+S3), analogues (“meaning”),

dissolving R1I:\TR, 532-33 582-85 252-54 (including n. 32), 300 sujet libre type, 522, 525-29, ESC issues (S3:\EEC), 588 tonal, 73—74, 102, 251, 252 n. 32

537 five-module block, exceptional

trill-cadence close, 470, 534, cases of, 595-96 vitalism metaphor (in Haydn), 16 n.

546-48 larger recapitulation (S3+R4, the 4, 233, 413

tutti interjection (S1:\TT'), 446, final rotation), 445, 577. See 470, 521-25, 528, 529 also Type 5 sonata, Solo 1 wit (Haydn). See under index of

Type 5 sonata, Solo 2 (S2, (S1), larger exposition names: Haydn, Franz Joseph developmental space), 444— (S1 + R2) writing over. See under rotation

45, 563-73 rotation, completion of, 577

“development themes” (Ktister), synthesis function, 493, “zero” modules, 71, 72-73. See

564 494-95, 521, 563, 577-81, also under closing zone (C);

episodic, 563—64 587-88, 590 primary-theme zone (P); event zones, 571—/3 R3=>S3 merger. See under Type secondary-theme zone (S) linkage-technique openings, 5 sonata, Ritornello 3 (R3), zone, 9-10, 17 (fig. 2.1), 23-24,

564-70 recapitulatory launch 616-18. See also action-space

retransition (within or following option (action-zone); closing zone S2), 572. See also Type 5 relationships to $1, modular, (C); primary-theme zone

sonata, Ritornello 3 (R3), 579-81 (P); secondary-theme zone retransition (RT) restorations of R1 material (S); transition (TR)

rotational implications of, 570-71 suppressed in S1, three

tutti interjections, 573 options, 588—96

Type 5 sonata, Solo 3 (S3, solo solo recapitulation, 445, 577

recapitulation) transition-zone options, 586-88