Communication in Eighteenth-Century Music [illustrated edition] 0521888298, 9780521888295

Citation preview

This page intentionally left blank

Communication in Eighteenth-Century Music

Written by ten leading scholars, this volume assembles studies of eighteenth- and early nineteenth-century music under the broad rubric of communication. That such an impulse motivates musical composition and performance in this period of European musical history is often acknowledged but seldom examined in depth. The book explores a broad set of issues, ranging from the exigencies of the market for books and music in the eighteenth century through to the deployment of dance topoi in musical composition. A number of close readings of individual works by Mozart, Haydn and Beethoven draw on a sophisticated body of historically appropriate technical resources to illuminate theories of form, metre, bass lines and dance topoi. Students and scholars of music history, theory and analysis will find in this volume a set of challenging, state-of-the-art essays that will stimulate debate about musical meaning and engender further study. da n u ta m i r k a is Senior Lecturer in Music at the University of Southampton. She is the author of The Sonoristic Structuralism of Krzysztof Penderecki (1997) and the forthcoming Playing with Meter: Metric Manipulations in Haydn and Mozart’s Chamber Music for Strings. Her articles have appeared in many publications including Journal of Music Theory, Journal of Musicology, The American Journal of Semiotics and Musical Quarterly. ko f i ag aw u is Professor of Music at Princeton University. His books include Playing with Signs: A Semiotic Interpretation of Classic Music (1991), which received the Young Scholar Award from the Society for Music Theory, Representing African Music: Postcolonial Notes, Queries, Positions (2003) and Music as Discourse: Semiotic Adventures in Romantic Music (2008). Author of numerous articles and reviews, his writing has appeared in a wide variety of scholarly journals, among them Music Theory Spectrum, Ethnomusicology, Journal of the American Musicological Society and 19th-Century Music.

Communication in Eighteenth-Century Music Edited by da n u ta m i r k a a n d ko f i ag aw u

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521888295 © Cambridge University Press 2008 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2008

ISBN-13 978-0-511-41396-4

eBook (EBL)

ISBN-13 978-0-521-88829-5

hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents

Notes on contributors [page vii] Introduction [1] da n u ta m i r k a pa rt i c o m m u n i c at i o n a n d t h e m a r k e t 1 Communication and verisimilitude in the eighteenth century [13] pau l c o b l e y 2 Listening to listeners [34] m a r k e va n b o n d s 3 ‘Mannichfaltige Abweichungen von der gew¨ohnlichen Sonaten-Form’: Beethoven’s ‘Piano Solo’ Op. 31 No. 1 and the challenge of communication [53] c l au d i a m au r e r z e n c k pa rt i i m u s i c a l g r a m m a r 4 Metre, phrase structure and manipulations of musical beginnings [83] da n u ta m i r k a 5 National metrical types in music of the eighteenth and early nineteenth centuries [112] w illiam rothstein 6 Schoenberg’s ‘second melody’, or, ‘Meyer-ed’ in the bass [160] william e. caplin

vi

Contents

pa rt i i i r h e t o r i c a l f o r m a n d t o p i c a l d e c o ru m 7 A metaphoric model of sonata form: two expositions by Mozart [189] michael spitzer 8 Beethoven’s Op. 18 No. 3, first movement: two readings, with a comment on analysis [230] ko f i ag aw u 9 Mozart’s k331, first movement: once more, with feeling [254] wye j. allanbrook 10 Dance topoi, sonic analogues and musical grammar: communicating with music in the eighteenth century [283] l aw r e n c e m . z b i ko ws k i Afterword [310] ko f i ag aw u Bibliography [318] Index of concepts [335] Index of names and works

[340]

Notes on contributors

ko f i ag aw u is Professor of Music at Princeton University. His books include Playing with Signs: A Semiotic Interpretation of Classic Music (1991), which received the Young Scholar Award from the Society for Music Theory, Representing African Music: Postcolonial Notes, Queries, Positions (2003) and Music as Discourse: Semiotic Adventures in Romantic Music (2008). Author of numerous articles and reviews, his writing has appeared in a wide variety of scholarly journals, among them Music Theory Spectrum, Ethnomusicology, Journal of the American Musicological Society and 19th-Century Music. w y e j . a l l a n b r o o k is Professor Emerita at the University of California, Berkeley. Her publications include Rhythmic Gesture in Mozart: Le nozze di Figaro and Don Giovanni (1983), ‘Two Threads Through the Labyrinth: Topic and Process in the First Movements of K. 332 and K. 333’, in Convention in Eighteenth- and Nineteenth-Century Music, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt (1992), and ‘Theorizing the Comic Surface’, in Music in the Mirror, ed. Thomas Mathiesen and Andreas Giger (2002). Her current research is for a book on expression in late eighteenth-century instrumental music tentatively entitled A Secular Commedia: Comic Mimesis in Late-Eighteenth-Century Music. m a r k e va n b o n d s is the Cory C. Boshamer Distinguished Professor of Music at the University of North Carolina at Chapel Hill. His books include Wordless Rhetoric: Musical Form and the Metaphor of the Oration (1991), After Beethoven: Imperatives of Originality in the Symphony (1996) and Music as Thought: Listening to the Symphony in the Age of Beethoven (2006). His articles have appeared in Music and Letters, Journal of Musicology, Journal of the American Musicological Society and Studi musicali. He is currently working on a study of the aesthetics of absolute music in the nineteenth century. w i l l i a m e . c a p l i n is James McGill Professor of Music Theory in the Schulich School of Music, McGill University. His research interests include the theory of musical form, history of harmonic and rhythmic theories and theories of rhythm and metre. His book Classical Form: A Theory of

viii

Notes on contributors

Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven (1998) won the Society for Music Theory’s Wallace Berry Award in 1999. He has published studies in The Cambridge History of Western Music Theory, Journal of the American Musicological Society, Eighteenth-Century Music, Musiktheorie, Music Theory Spectrum, Journal of Music Theory, Journal of Musicological Research, Beethoven Forum. He served as President of the Society for Music Theory in 2006–7. pau l c o b l e y is Reader in Communications at London Metropolitan University and author of a number of books, including The American Thriller (2000) and Narrative (2001). He is the editor of The Communication Theory Reader (1996), The Routledge Companion to Semiotics and Linguistics (2001) and Communication Theories (2006). He also co-edits two journals: Subject Matters and Social Semiotics. His articles have appeared in Media, Culture and Society, Sign Systems Studies and Semiotica. c l au d i a m au r e r z e n c k is Professor of Music at the University of Hamburg. Her books include Vom Takt: Untersuchungen zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert (2001) and Mozart’s Cos`ı fan tutte: dramma giocoso und deutsches Singspiel (2007). She has published on both historical and theoretical aspects of the late eighteenth-century musical repertory in such scholarly journals as Archiv f¨ur Musikwissenschaft, Musikforschung, Mozart-Jahrbuch and Acta Mozartiana. da n u ta m i r k a is Senior Lecturer in Music at the University of Southampton. She is the author of The Sonoristic Structuralism of Krzysztof Penderecki (1997) and the forthcoming Playing with Meter: Metric Manipulations in Haydn and Mozart’s Chamber Music for Strings. Her articles have appeared in many publications including Journal of Music Theory, Journal of Musicology, The American Journal of Semiotics and Musical Quarterly. w i l l i a m r o t h s t e i n is Professor at Queen’s College and The Graduate Center, City University of New York. His areas of interest include Schenkerian theory and analysis, theories of rhythm and theories of form. He has also written on performance practice and the relationship of analysis to performance. His landmark book, Phrase Rhythm in Tonal Music (1989), received both an ASCAP-Deems Taylor Award and the Society for Music Theory’s Young Scholar Award. He has published in Beethoven Forum, 19thCentury Music, Music Analysis, Int´egral and Music Theory Online. His current research concentrates on Italian Romantic opera.

Notes on contributors

m i c h a e l s p i t z e r is Reader in Music at Durham University. A theorist as well as a musicologist, he has written widely on music and metaphor, including the first ever monograph on the subject, Metaphor and Musical Thought (2004). He also researches on the Classical style and in aesthetics, interests which come together in his second book, Music as Philosophy: Adorno and Beethoven’s Late Style (2006). His articles have appeared in Music Analysis, Music and Letters, Beethoven Forum and Journal of the Royal Musical Association. He is currently writing a book on emotion and musical structure. l aw r e n c e m . z b i ko ws k i is an Associate Professor of Music at the University of Chicago. He is also the director of the Division of Humanities’ project on creativity and cognition. His principal research interests involve applying recent work in cognitive science to various problems of music theory and analysis. He is the author of Conceptualizing Music: Cognitive Structure, Theory, and Analysis (2002), which won the Society for Music Theory’s 2004 Wallace Berry Award. His articles have appeared in such scholarly journals as Music Perception, Music Analysis, Music Theory Spectrum, Music Theory Online and Journal of the Royal Musical Association.

ix

Introduction da n u ta m i r k a

In view of the diverse approaches to musical scholarship current today, the position that music’s function is communication is not an obvious one to take. For many, a musical piece is an object to be contemplated, an organism to be examined, a mechanism to be deconstructed or a product to be consumed. None of these metaphors allows one to speak sensibly of musical communication. Yet none of them was in use in the late eighteenth century. At that time theoretical and aesthetic discourses about music were based upon the metaphor of music as language.1 Within this metaphor, a composer or performer was compared to an orator, and a musical piece to an oration subdivided into parts, periods and sentences. Just as the art of rhetoric had its raison d’ˆetre in persuading the listener, so the art of composition consisted in arousing his sentiments. The musical repertory labelled by later generations as the ‘Classical style’ was thus an expression of the aesthetic stance which conceived of music as communication between composer and listener. And yet, the comparison with rhetoric, dating back to the beginning of the sixteenth century, does not fully explain the characteristics of musical communication in the late eighteenth century. Rather, these characteristics are related to transformations of musical life bearing fruit in this period. Little by little, from the early decades of the eighteenth century on, music becomes the favourite kind of entertainment for the middle classes. This brings about a rapid growth of musical literacy manifested in the growing number of amateurs learning how to perform music. Beside amateur music making in the household and private or semi-public musical performances in the circle of guests, the institution of public concerts comes into being. The increase of the public role of music has a direct influence on its vocabulary. In place of the rhetorical figures characteristic of the Baroque, Classic music uses topics based upon allusions to different styles, genres and types of music making as its main source of expression. Proliferation of musical venues and multiplication of musical centres that can be reached via prints, concerts or commissions lead further to a differentiation of audiences. In order to successfully address a given audience, it is necessary for composers to avail themselves of an appropriate set of conventions. Incidentally, this

2

da n u ta m i r k a

necessity is dictated not only by aesthetic demands of musical rhetoric but also by more pedestrian exigencies of the musical market. Apparently, a piece tailored for the ears of Viennese audiences might not be as effective in London or in Paris. The result is, on the one hand, a heightened awareness of the role played by conventions on all levels of musical communication and, on the other hand, the sense of their very conventionality, inciting composers to confound conventions by mixing or ‘misapplying’ them in a given piece. In addition to differences between venues and local traditions, the eighteenth-century composer had to take into account the distinction between amateurs (Liebhaber) and connoisseurs (Kenner). The category of listeners represented by Kenner reflected the growth of musical literacy not only in performing but also in composing music. Theoretical knowledge of Kenner allowed the composer to consider them not as passive receivers but as active partners in the process of communication, and thus to engage them in a game played upon the technical rules of composition. The wit (Witz) of this game suggests that, within the metaphor of language, the ancient model of oration gradually gave way to a more modern ideal of conversation.2 The issue of communication in reference to late eighteenth-century music has gained currency over the past decades. Perhaps the most noted manifestation of this process is the study of musical topics. After the pioneering work of Leonard Ratner, topics were further investigated by Wye Allanbrook, Kofi ´ Raymond Monelle Agawu, Elaine Sisman, Robert Hatten, M´arta Grabocz, and others.3 What these studies reveal is the profound sociability of Classic music. Musical conventions represented by topics rely upon more general conventions governing eighteenth-century social life and derive their meaning from various social contexts. Defined by Ratner as ‘subjects for musical discourse’,4 topics constitute the semantic level of musical communication in the late eighteenth century. Yet they are not its sole level. In this period, communication also takes place on the level of musical form. The analytical approach to form in which deviations from norms are considered meaningful has been represented more or less explicitly by many authors, not least by Charles Rosen, despite his nominal rejection of musical forms as norms. The most recent instantiation of this approach is the new ‘sonata theory’ offered by James Hepokoski and Warren Darcy.5 Its theoretical foundations were laid down by Leonard B. Meyer, who related meaning to expectation.6 As Meyer consistently emphasized, expectations concerning the course of a given composition are determined by formal schemata learned by the listener through exposure to a given musical repertory. Meyer was, in fact, the first twentieth-century author to conceive of music as communication. Studies in late eighteenth-century music inspired

Introduction

by him include those by Janet Levy, Gretchen Wheelock and Karol Berger.7 Symptomatic of this stream of research are references to literary theories which emphasize the role of the reader’s expectations in actualization of a verbal text: the reader-response theory by Wolfgang Iser (Wheelock) and the reception theory by Hans Robert Jauss (Berger). Note that Meyer’s ‘listenerresponse theory’ is significantly earlier than its literary counterparts. Yet the interest of these references resides in the fact that they illuminate certain historical parallels between music and literature. It is not by chance that Iser’s book, The Implied Reader, begins with the eighteenth-century novel.8 Apparently, in the eighteenth century communicative strategies started for the first time to pose a problem consciously confronted by both writers and composers. Comparisons such as those drawn at that time between Laurence Sterne and Joseph Haydn have their origins here.9 Another stream of research devoted to musical form which has a bearing on the study of musical communication proceeds from Ratner. Following in his footsteps, several younger authors began a search for descriptions of formal schemata in eighteenth-century composition handbooks. In Germany a similar process commenced under the influence of Wilhelm Seidel.10 In this way, the history of music theory gradually abandoned its antiquarian status as it engaged with music analysis. The objective behind this enterprise was to describe formal processes in eighteenth-century music from the point of view of the historical listener. This objective was perhaps most radically voiced by Karol Berger, who – echoing earlier ideas of Heinrich Besseler11 – postulated ‘a history of musical hearing’ as a field of investigation whose purpose is ‘to reconstruct the experience of music in the past’.12 Although this postulate remains as yet unrealized, it should be clear that historical awareness is indispensable for the study of musical communication. Since all communicative acts refer to the background knowledge of the receiver, communicative strategies developed in late eighteenth-century music imply a listener equipped with the theoretical knowledge and listening habits characteristic of that time. The same historical imperative informs the work of Mark Evan Bonds and Elaine Sisman, authors who deserve the credit for highlighting the connection between music and rhetoric, mentioned earlier.13 As amply documented by Bonds, in the late eighteenth century questions of musical form belong to the area of musical rhetoric in the proper sense of the word. One more level of musical communication in the late eighteenth century is grammar. In contradistinction to rhetoric, which deals with the organization of entire pieces or movements, musical grammar – like its linguistic counterpart – concerns internal organization of musical sentences

3

4

da n u ta m i r k a

(S¨atze). Grammatical rules of music, described in eighteenth-century composition handbooks, are thus rules of harmony and counterpoint. (Musical metre has its counterpart in poetic metre but it is related to grammar in the sense that strong beats, like strong syllables, are called ‘grammatical’ accents.) While these rules also give rise to expectations, analysis of such expectations must necessarily come closer to linguistics than to literary theory. Analysis of melodic expectations started with the model of implication-realization developed by Meyer and Eugene Narmour.14 Since then, syntactic expectations have been subject to theoretical and experimental study in the field of cognitive musicology and cognitively oriented music theory. One particularly interesting aspect of the late eighteenth-century repertory revealed by this study is the presence of conventional patterns or schemata shaping ‘Classic turns of phrase’. Following Meyer, substantial work was devoted to these schemata by Robert Gjerdingen.15 Incidentally, the alliance of cognitive and historical approaches, pursued by Gjerdingen, seems particularly promising due to the strong cognitive component of music theory in the late eighteenth century. In theoretical treatises of the time the ‘ears’ (Ohren) or ‘hearing’ (Geh¨or) of the listener are regularly referred to in order to substantiate rules formulated by their authors. Also important in eighteenth-century music theory is the notion of expectation. The earliest remarks devoted to this theme come from Joseph Riepel, Johann Philipp Kirnberger and Heinrich Christoph Koch. From the point of view of the metaphor of music as language it may be interesting to observe that manipulations of syntactic expectations occur only seldom in linguistic communication. In everyday speech they are represented by word games and puns; in poetry by limericks. In music, by contrast, such manipulations are very common. Here, however, they serve not only the purpose of communication but also the arousal of emotions (Meyer). Mutual interrelation of these two aspects constitutes one of the most interesting problems faced by contemporary music theory. As evident from this short survey, scholars of music who take the position of communication today, though numerous, do not form any league or camp. Rather, they are scattered over many different fields of musical studies ranging from historical musicology and music semiotics through modern and historical music theory to cognitive psychology of music. Though working with different methods, they share an explicitly listener-oriented approach and more or less explicit connections to various branches of communication sciences. The idea of bringing together scholars engaged in this kind of research materialized for the first time during the workshop ‘Communicative Strategies in Music of the Late Eighteenth Century’ which took place in Bad Sulzburg, Germany, from 5 to 9 July 2005 (see list of participants

Introduction

below). This volume presents articles based upon selected papers given in Bad Sulzburg. The hope is to continue the lively dialogue started there, and to foster systematic study of musical communication – undoubtedly one of the central issues raised by late eighteenth-century music. The first part of the volume is dedicated to complex interactions between musical communication and the market. Paul Cobley, a representative of communication sciences, considers publication markets for literature and for music as elements of the ‘public sphere’ (Habermas). As he observes, literature and music were among the main subjects of public sphere debates taking place in salons, taverns and coffeehouses throughout eighteenthcentury Europe. Supported by musical and literary criticism printed in newspapers, periodicals and magazines, these debates generated the ‘public opinion’ – in music this might be called ‘public taste’ – which, together with ‘rules of genre’, is taken by Cobley to underlie what he calls ‘verisimilitude’: a communicative transaction between the author and the audience about what is credible, or likely to happen, in the course of a musical or literary work. The influence exerted by the publication market upon strategies of musical communication is considered by Mark Evan Bonds. One consequence of the publication market was to confront composers with a heterogeneous audience including both connoisseurs (Kenner) and amateurs (Liebhaber). According to Bonds, the most effective strategy to address such an audience was to write for Liebhaber, from time to time inserting moments to be appreciated by Kenner. Bonds identifies such moments preeminently with sections of contrapuntal writing. His essay is an invitation to embrace the aesthetic and the commercial in analysis of late eighteenth-century music. An illuminating case study of the relation between publisher and composer is offered by Claudia Maurer Zenck. Beethoven’s Piano Sonata in G major Op. 31 No. 1 was commissioned by the Zurich publisher, Hans Georg N¨ageli, for a series (R´epertoire de clavecinistes) earmarked for piano sonatas featuring ‘many departures from usual sonata form’. Beethoven responded to this commission with a piece of ‘humorous music’: a parody of musical bungling by a bad composer. Jesting with the ‘rules of the genre’ in this piece represents another kind of communication with Kenner. The close analytic reading of Op. 31 No. 1 provided by Maurer Zenck forms a bridge to the second part of the volume dealing with aspects of musical grammar. In my own essay I analyse sophisticated manipulations of metre taking place at the beginning of musical pieces and impinging upon phrase structure. Since such manipulations rely upon theoretical knowledge of eighteenth-century listeners, they represent one more aspect of

5

6

da n u ta m i r k a

communicative strategies addressed to Kenner. The focus of William Rothstein’s essay is not on musical communication through metre but on verbal communication about metre. The preference for beginning-accented and end-accented concepts of metre by North American and German scholars respectively is traced by Rothstein back to different national traditions of metric notation originating in the eighteenth century. William Caplin’s contribution is the first attempt at a comprehensive theory of a tonal bass melody. The potential of this theory for the study of musical communication lies in providing the framework for the mechanism of melodic implication-realization in the bass as different from that in upper voices. Models of melodic bass progressions, proposed by Caplin, summarize implications arising in different types of phrases and allow one to analytically trace surprises caused by deviations and delays. The third group of chapters clusters around the issue of musical form and its topical decorum. Michael Spitzer proposes a new model of the sonataform exposition referring to metaphor theory and concentrating on the thematic contrast between first and second groups. According to Spitzer, this contrast arises not from different types of material but from presenting the same materials in two different ways: first absorbed into stereotypical patterns or schemata (Gjerdingen); then liberated from this patterning and turned into motives. The second presentation ‘realizes’ the musical material in that it makes its specificity perceivable for the listener. Ultimately, Spitzer’s model thus relies upon the mechanism of music perception. His ‘two-time rule’ is illustrated in analyses of expositions from first movements of Mozart’s k515 and k551. The point made by Kofi Agawu is an epistemological one. He presents two readings of the first movement of Beethoven’s String Quartet in D major, Op. 18 No. 3. The first, structural reading, expanding upon the paradigmatic analysis by Nicolas Ruwet and Jean-Jacques Nattiez, departs from models of tonal progressions (background). The second reading unearths the expressive dimension of this movement represented by topics (foreground). As Agawu points out, the two readings, while complete in themselves, are irreducible to one another since no rules of transformation ultimately bridge background and foreground. The tension between the structural and the expressive is framed in a slightly different manner by Wye Allanbrook. She dissects some classical analyses of one of music theory’s paradigm pieces: the theme of the first movement of Mozart’s k331. In her own analysis, several unusual features of the theme that are normally revealed – or concealed – in strictly ‘structural’ readings fall in place when seen through the lenses of an expressive sign: the topos of siciliano.

Introduction

Dance topoi are expressive signs not only by virtue of their associations with social contexts (explored by Allanbrook and other exponents of topic theory), but also because they represent sonic analogues of dynamic processes characteristic of the bodily experience of dancing. As such, they can not only be combined, juxtaposed and contrasted with other topics but also modified and subverted so as to yield new expressive possibilities. This level of topical signification is illuminated from a cognitive perspective by Lawrence Zbikowski. In his essay Zbikowski demonstrates how, departing from sonic analogues of dance steps and movements, a composer may arrive at constructs which have no analogy in dancing. As an illustration, he cites the bourr´ee from the finale of Haydn’s Op. 76 No. 4. Even if this process has to do with communication, it is communication of a special sort in that it goes beyond the metaphor of music as language. In this way Zbikowski’s essay rounds off the collection by bringing it from the original metaphor to its opposite pole. As he emphasizes, language and music are two distinct realities characterized by different functions in human culture. The former is designed for dealing with objects or concepts, the latter for dynamic processes associated with motion and emotion. This insight was already familiar in the eighteenth century. While the rhetorical tradition emphasized the similarity between language and music, it did not fail to recognize the differences between the verbal language of ideas and the musical language of sentiments. In the anthropological perspective adopted by such writers as Charles Batteaux, Etienne Bonnot de Condillac and Jean-Jacques Rousseau, this difference was interpreted as superiority of music over language, opening the path from the late eighteenth to the nineteenth century and towards Romanticism. I wish to thank all the participants of the Bad Sulzburg workshop – those represented in this volume and those whose papers for various reasons are not included in it – for their contribution to this unique event. The five summer days in the depths of the Black Forest are memorable not only for the intellectual rigour and density of scholarly discussions but also for an exceptional atmosphere of collegiality and mutual openness that gave these discussions a stamp of genuine interpersonal exchange. The workshop was sponsored by Deutsche Forschungsgemeinschaft and the University of Freiburg. I am particularly grateful to the Dean of the Philosophical Faculty, Professor Hermann Schwengel, and the Director of the Institute of Musicology, Professor Christian Berger, for financial and organizational support from their respective university divisions. Professor Berger was personally engaged in preparing the workshop as its co-organizer. My stay at his Institute was

7

8

da n u ta m i r k a

possible thanks to research scholarships from the Alexander von Humboldt Foundation and Deutsche Forschungsgemeinschaft. Among those who have directly contributed to the preparation of this volume I thank first and foremost Kofi Agawu, who agreed to share the editorial tasks with me. I thank further Guillermo Brachetta, who set the bulk of the musical examples, Susan Beer and Helen MacFarlane who copy-edited the manuscript, as well as Victoria Cooper, Rebecca Jones and Rosina Di Marzo, who took care of it on behalf of Cambridge University Press.

Notes Participants in workshop ‘Communicative Strategies in Music of the Late Eighteenth Century’, Bad Sulzburg, 5–9 July 2005: Kofi Agawu (Princeton University) Wye J. Allanbrook (University of California, Berkeley) Christian Berger (University of Freiburg) Mark Evan Bonds (University of North Carolina) William E. Caplin (McGill University) Paul Cobley (Metropolitan University, London) Ludwig Holtmeier (University of Music, Freiburg) Robert Levin (Harvard University) Claudia Maurer Zenck (University of Hamburg) Danuta Mirka (Music Academy, Katowice) William Rothstein (Queen’s College and The Graduate Center, City University of New York) Janet Schmalfeldt (Tufts University) Elaine Sisman (Columbia University) Michael Spitzer (University of Durham) Dean Sutcliffe (Cambridge University) James Webster (Cornell University) Gretchen A. Wheelock (Eastman School of Music) Lawrence M. Zbikowski (University of Chicago) 1 For a thorough discussion of this metaphor and its role in the late eighteenth century, see Mark Evan Bonds, Wordless Rhetoric: Musical Form and the Metaphor of the Oration (Cambridge, MA: Harvard University Press, 1991), 61–80, and Michael Spitzer, Metaphor and Musical Thought (Chicago: University of Chicago Press, 2004), 207–75. 2 At least since Goethe’s famous saying about Haydn’s string quartets, conversation has served as a model of interaction between performers in chamber music. See, for instance, J¨urgen Mainka, ‘Haydns Streichquartette: “Man h¨ort vier vern¨unftige Leute sich untereinander unterhalten”’, Musik und Gesellschaft 32

Introduction

3

4 5

6 7

(1982), 146–50, Barbara R. Hanning, ‘Conversation and Musical Style in the Late Eighteenth-Century Parisian Salon’, Eighteenth-Century Studies 22 (1989), 512–28 and Mara Parker, The String Quartet, 1750–1797: Four Types of Musical Conversation (Aldershot: Ashgate, 2002). The ideal of conversation as a manner to address the listener was proposed by Gretchen A. Wheelock, ‘Engaging Strategies in Haydn’s Opus 33 String Quartets’, Eighteenth-Century Studies 25 (1991), 1–30; Haydn’s Ingenious Jesting with Art (New York: Schirmer Books, 1992); and ‘The “Rhetorical Pause” and Metaphors of Conversation in Haydn’s Quartets’, in Haydn und das Streichquartett, Eisenst¨adter Haydn-Berichte 2, ed. Georg Feder and Walter Reicher (Tutzing: Schneider, 2003), 67–85. Leonard G. Ratner, Classic Music: Expression, Form, and Style (New York: Schirmer, 1980), 1–30; Wye J. Allanbrook, Rhythmic Gesture in Mozart: Le Nozze di Figaro and Don Giovanni (Chicago: University of Chicago Press, 1983); V. Kofi Agawu, Playing with Signs: A Semiotic Interpretation of Classic Music (Princeton: Princeton University Press, 1991); Elaine Sisman, Mozart: The ‘Jupiter’ Symphony No. 41 in C major, K. 551 (Cambridge: Cambridge University Press, 1993); Robert Hatten, Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation (Bloomington: Indiana University Press, ´ ‘A. J. Greimas’s Narrative Grammar and the Analysis of 1994); M´arta Grabocz, Sonata Form’, Int´egral 12 (1998), 1–24; Raymond Monelle, The Sense of Music: Semiotic Essays (Princeton: Princeton University Press, 2000), 2–80, and The Musical Topic: Hunt, Military and Pastoral (Bloomington: Indiana University Press, 2006). Ratner, Classic Music, 9. James Hepokoski and Warren Darcy, Elements of Sonata Theory: Norms, Types, and Deformations in the Late-Eighteenth-Century Sonata (New York: Oxford University Press, 2006). Leonard B. Meyer, Emotion and Meaning in Music (Chicago: University of Chicago Press, 1956). Janet M. Levy, ‘Texture as a Sign in Classic and Early Romantic Music’, Journal of the American Musicological Society 35 (1982), 482–531; Beethoven’s Compositional Choices: The Two Versions of Opus 18, No. 1, First Movement (Philadelphia: University of Pennsylvania Press, 1982); and ‘Gesture, Form and Syntax in Haydn’s Music’, in Haydn Studies: Proceedings of the International Haydn Conference, Washington, DC, 1975, ed. Jens Peter Larsen, Howard Server and James Webster (New York: Norton, 1981), 355–62; Wheelock, Haydn’s Ingenious Jesting; Karol Berger, ‘Toward a History of Hearing: The Classic Concerto, A Sample Case’, in Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt (Stuyvesant, NY: Pendragon, 1992), 405–29, and ‘The First Movement Punctuation Form in Mozart’s Piano Concertos’, in Mozart’s Piano Concertos: Text, Context, Interpretation, ed. Neal Zaslaw (Ann Arbor: University of Michigan Press, 1996), 239–59.

9

10

da n u ta m i r k a

8 Wolfgang Iser, The Implied Reader: Patterns of Communication in Prose Fiction from Bunyan to Beckett (Baltimore: Johns Hopkins University Press, 1974). 9 See Howard Irving, ‘Haydn and Laurence Sterne: Similarities in EighteenthCentury Literary and Musical Wit’, Current Musicology 40 (1985), 34–49; Mark Evan Bonds, ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’, Journal of the American Musicological Society 44 (1991), 57–91; Wheelock, Haydn’s Ingenious Jesting. ¨ Rhythmustheorien der Neuzeit, Neue Heidelberger Studien 10 Wilhelm Seidel, Uber zur Musikwissenschaft 7 (Bern: Francke, 1975). Among historically-oriented studies in the German language inspired by Seidel, most important are those by Hermann Forschner, Instrumentalmusik Joseph Haydns aus der Sicht Heinrich Christoph Kochs, Beitr¨age zur Musikforschung 13 (Munich: Musikverlag Emil Katzbichler, 1984); Wolfgang Budday, Grundlagen musikalischer Formen der Wiener Klassik: An Hand der zeitgen¨ossischen Theorie von Joseph Riepel und Heinrich Christoph Koch dargestellt an Menuetten und Sonatens¨atzen (1750–1790) ¨ (Kassel: B¨arenreiter, 1983); and Claudia Maurer Zenck, Vom Takt: Uberlegungen

11

12 13 14

15

zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert (Vienna: B¨ohlau, 2001). Heinrich Besseler, ‘Grundfragen des musikalischen H¨orens’, Jahrbuch Peters 1925, 35–52 and Das musikalische H¨oren der Neuzeit, Berichte u¨ ber die Verhandlungen der S¨achsischen Akademie der Wissenschaften zu Leipzig, Philologischhistorische Klasse 104/6 (Berlin: Akademie-Verlag, 1959). Berger, ‘Toward a History of Hearing’, 406. Bonds, Wordless Rhetoric; Elaine Sisman, Haydn and the Classical Variation (Cambridge, MA: Harvard University Press, 1993). Leonard B. Meyer, Explaining Music: Essays and Explorations (Berkeley: University of California Press, 1973); Eugene Narmour, Beyond Schenkerism: The Need for Alternatives in Music Analysis (Chicago: University of Chicago Press, 1977). Robert O. Gjerdingen, A Classic Turn of Phrase: Music and the Psychology of Conventions (Philadelphia: University of Pennsylvania Press, 1988); Music in the Galant Style: Being an Essay on Various Schemata Characteristic of EighteenthCentury Music for Courtly Chambers, Chapels, and Theaters, Including Tasteful Passages of Music Drawn from Most Excellent Chapel Masters in the Employ of Noble and Noteworthy Personages, Said Music All Collected for the Reader’s Delectation on the World Wide Web (New York: Oxford University Press, 2007).

pa rt i

Communication and the market

1

Communication and verisimilitude in the eighteenth century pau l c o b l e y

There are three intellectual catalysts to this chapter. The first is Daniel Defoe’s The Life and Strange Surprizing Adventures of Robinson Crusoe (1719). It could have been almost any cultural artefact, but the eighteenth century is a rich source of examples of this kind: that is, texts whose readings have radically changed from their first appearance. Like Jonathan Swift’s Gulliver’s Travels (Travels into Several Remote Nations of the World by Lemuel Gulliver, 1726), Robinson Crusoe is today often presented as a children’s book and its original political bearing is forgotten outside of scholarship. What it is taken to communicate has thus changed in time, despite the text retaining identical words in successive publications. If one wishes to begin to excavate what Robinson Crusoe communicated to its original audience, an audience no longer available for questioning, it is therefore necessary to examine the contemporary discourses that surrounded that text and which might have provided at least the co-ordinates for understanding what it communicated, especially regarding allegory, economy, industry and instructions for good living.1 The second catalyst is also articulated with reference to an eighteenthcentury text, Laurence Sterne’s The Life and Opinions of Tristram Shandy, Gentleman (1759–67). The issue is slightly different, here. What is in question is a representative device: clearly, at one level, Sterne’s novel is a reaction to proto-realism in the eighteenth century, a refusal to swallow the claims to verisimilitude of fictional representations which purport to parcel up time and space into equal, digestible portions. Yet, such ‘rupturing’, ‘alienating’ or ‘metafictional’ devices as Tristram Shandy employs have been more closely associated with the twentieth-century representation in ‘radical’ film, Brechtian theatre and postmodernist novels. In this case, the entire mode of representation can be understood as a vehicle whose efficacy and import is specific to the century in which it is employed. Tristram Shandy, then, does not embody a general humorous undermining of convention; rather it is a context-specific configuration of humour and departure from convention, referring to contemporary writing and likely to be recognized as such by its contemporary audience.

14

pau l c o b l e y

Finally, the third catalyst to this chapter has been writings on music about communicative strategies,2 some of which have also come to fruition in the current volume. Again, these have focused on the eighteenth century. But, like the other two, they are clearly dilemmas of how cultural history is to be constructed and each might be seen as asking similar questions. These questions are concerned with the responses to artefacts by contemporary audiences, as well as the ways in which it might be possible to reconstruct those responses now that the audiences have gone and, especially, identifying in the construction of artefacts themselves a sense of the audience(s) at which they were aimed. As such, these questions challenge the canonical imperative in which the ‘meaning’ of artefacts – music, novels, plays, and so forth – is fixed by retrospective discourses which pronounce for the contemporary period on what is ‘good’ and worthy of value or preservation and what is not. Perhaps central to these questions, too, is verisimilitude: what is found to be credible in different epochs. Undoubtedly, there is more to communication and verisimilitude than mere ‘intent’; such a fallacy was left behind some time ago and it has been recognized that there is a demand for attention to the technology of communication and the specific figures in texts of communication among a number of things. Other chapters in this volume contribute to the important task of elucidating these in music of the late eighteenth century. However, this chapter will focus on one small element of the broad context in which eighteenth-century European music communicated. It will suggest that this element of the communicative transaction, its verisimilitude (or general credibility), must be factored in to analyses of the strategies of address. As is common with theories of verisimilitude – although not explicitly within casual uses of the term – it will find that credibility derives from both artistic conventions and public opinion. The issue of verisimilitude is to be found in a plethora of places in eighteenth-century life, sometimes in unexpected environments. It also appears in a number of guises wherever communication and signs are discussed. It might be expeditious, then, to begin with a brief consideration of the concept of ‘communication’ and the roles played by context before addressing the more elusive signs of verisimilitude.

Modelling communication As is well known, the word ‘communication’ comes from the Latin communicare and shares Indo-European etymological roots with the words ‘common’,

Communication and verisimilitude

‘commune’ and ‘community’. It is often noted, then, that ‘communication’ is associated with ‘bringing together’. The study of communication, however, has seldom proceeded in a systematic way until modern times. John Durham Peters, for example, despite a commitment to discussion of the ‘idea of communication’ since the ancients, locates the genesis of communication study in intellectual currents of the 1920s.3 It was here that attention started to be explicitly given to the participants in the communication process, proceeding from a basic SMR model, or Sender → Message → Receiver

What are brought together in this model, in the most fundamental sense, are the sender and receiver, addresser and addressee. The message provides the basis for this action since it is encoded by the sender and decoded by the receiver. The fact that it requires coding and encoding, of course, indicates that the message is not a perfect, transparent vehicle for ‘meaning’; rather, the ‘message is a medium’, to invert McLuhan. It mediates meaning, or certainly will do as a result of being in a channel. It is for this reason that the era of ‘communication science’,4 in which models of communication were successively elaborated, often involved extended discussions of ‘interference’ or ‘distortion’ at the level of the message. Classically, for Claude E. Shannon and Warren Weaver, such ‘noise’ is a factor that might corrupt the implicitly mythical integrity of the message, the prime example being many conflicting signals in the same channel at once.5 It is a difficulty in the study of communication which has been revisited by numerous attempts to construct communication models.6 What this preoccupation in communication theory has achieved is an unsettling of straightforward relations of the sender and receiver. Initially, this involved a questioning of the extent to which an author – or, typically in cinematic communication, an auteur – is the seat of meaning, and a displacement of the authoritative genius from the pole position of all meaning creation in communications. At the very least, communication theory’s influence on conceptualization of the arts has been such that a model of ‘creator, work, reception’ can only be considered where due attention is paid to each component in the transaction. Nowhere has this been more apparent than in narrative theory. The traditional (though actually relatively new, Leavisite) conception of authorial genius demanding the rapt contemplation of the careful reader has been supplanted by a more variegated picture of (narrative) text production and consumption:7

15

16

pau l c o b l e y

Real Author → Implied Author → Narrator → (Narratee) → Implied Reader → Real Reader

Undoubtedly, this model is more complex at the level of creativity and reading, but it is also more informed about roles in the text. For example, the narratee (in brackets in this version of the model) plays a role which, with reference to epistolary novels especially, has sometimes been noted but seldom appreciated in narrative communication in general. The left hand side of this model of narrative levels reveals facts which have been known for a long time but never properly integrated into studies of the arts; namely, that the thoughts, proclivities and intentions of a ‘real author’ can never be fully known – even, in a post-Freudian world, to him/herself. As such, the realm of intentions is located in those places where an ‘implied author’ is seen to be operating, offering coherence where it cannot be automatically assumed that the empirical author had meticulously planned this all along. On the right hand side of the model, there is the distinction between a coherent set of imperatives with which a reader might engage – more accurately, this should be called an implied reading rather than an ‘implied reader’. There is also the ‘real reader’, the locus of attitudes, values, experiences – ideological baggage that is brought to the act of decoding. This issue has been one of the most prominent concerns of communication study in the last twenty years.8 If one is to consider communicative strategies, involving necessarily a sender and receiver or an implied author and implied reader, there comes a time when the broader situation understood by both participants has to be scrutinized and evaluated. Often, this is called ‘context’, a problematic term because it has an everyday usage which everyone is assumed to understand. For Roman Jakobson, in his noted communication model of 1960, context is simply the situation understood by addresser and addressee, such that messages can be orientated towards a referential function. For literary theory after the work of the Constanz School, this is understood as the ‘horizon of expectations’.9 For sociosemiotics, on the other hand, it is axiomatic that the signs used by humans are already thoroughly imbued with ‘motivation’ and, at the point of their use, can only ever be experienced as such.10 Thus, context is not outside the communication, guiding it somehow for its participants, but, instead, actually is the communication.11 Superimposing a communication model onto literate communication in the eighteenth century, one would have to augment the narrative levels schema outlined above. This is particularly the case if one were to adopt a political economic perspective which was at least cognizant of commerce.

Communication and verisimilitude

In an important essay, John Feather puts forward the following basic understanding of the communication situation in the period:12 Author → Publisher → Bookseller → Reader

But Feather also adds complications. Firstly, he notes that there is the potential for ‘noise’ in the channel from author to publisher: one has also to figure the negotiated communication between patron, editor and (initial) reader, each exercising discrete influences. Secondly, he points out that from publisher to bookseller there might be the imperative of negotiation with a printer, a binder, a manufacturer and the supplier of materials, as well as anticipation of different levels of bookselling. Thirdly, between bookseller and reader it is necessary to remember that various intermediaries might wield influence, such as reviewers and librarians, to which should also be added other word-of-mouth reports. Further to these suggestions, it would probably not be injudicious to add that such relations were not unknown to eighteenth-century readers. For example, it was well known that one of the bestselling novelists of the period in Britain, Samuel Richardson, was also a famous bookseller. To begin to understand ‘communication’, then, we need to have a broader understanding of contemporary practices of disseminating messages and of reading them. Necessarily, this involves some engagement with the idea of the public. This very notion has haunted communication study since the 1920s (see note 6) – but with good reason. In the case of the eighteenth century, it is well known as a period of boom in literacy and, as a consequence, a period in which the dissemination of communication took place on a scale never witnessed before. The printing press had been in use for over three hundred years before 1800. But in England from 1695, with the lapsing of the Licensing Act, a new era was opened. The Act had severely regulated and censored printed publications of all sorts until this time. Now that a range of printed artefacts could be produced for prospective markets, access to printing technology was sought by entrepreneurs.13 Booksellers grew in number. The subsequent growth of magazines, periodicals, newspapers and other printed materials also aided literacy, whether in schools or when taken up privately by the self-taught citizen. As Jack Goody and Ian Watt point out in a well known argument, the initial years of the advent of print in the preceding centuries – despite the continued practice of public readings – had made reading a more private, individual phenomenon and had promoted communications centred on the self, such as diaries and confessions.14 Yet, equally well known is the extent to which public life in Europe prospered after 1800.

17

18

pau l c o b l e y

Communication and the public Debate around J¨urgen Habermas’ groundbreaking thesis in The Structural Transformation of the Public Sphere continues.15 What is agreed by most scholars, however, is that the activities of public life took on a different complexion in the eighteenth century, developing into what Habermas calls the (bourgeois) public sphere. This sphere, derived from the meetings of the mercantile class, burghers and others, for talk and debate, developed in the seventeenth century but accelerated especially in the eighteenth century and was attendant on an explosive growth of coffeehouses in Britain and the German-speaking lands and, in France, the salons. The importance of the discourse in these arenas sits deeper than its surface appearance. As Habermas shows, the coffeehouse talk was mainly orientated towards questions of literature, even above politics. Yet, such talk, importantly, took place in a domain divorced from the family and the intimate sphere while, at the same time, being extraneous to reproduction (that is, the reproduction of the relations of production, commerce and business). As such, it exemplified the bourgeoisie and its ability to pursue leisure in the period, yet it also had another complexion deriving from the fact that the bourgeoisie had not become entrenched in the way that it was to become in the following century in Europe. This entailed that the discourse of the public sphere was political in quite a ‘pure’ sense, a ‘rational’ discourse, as described by Habermas, not simply dictated by, or the epiphenomenon of, the accumulation of capital. Public sphere discourse also had the power to proliferate. Habermas suggests that ‘the public sphere can best be described as a network for communicating information and points of view . . . the streams of communication are, in the process, filtered and synthesized in such a way that they coalesce into bundles of topically specified public opinions’.16 Not only was this discourse located in the coffeehouses, then, it was also replicated and disseminated in the journals of the day, specifically in Britain, The Tatler and The Spectator. As Angus Ross notes, the latter, with its address to the morning tea-table and the contemplative hours of civil servants, grew out of the former with its more publicly orientated coffeehouse tone.17 Most recently, Luke Goode has asserted that ‘there is an implicit logocentrism lurking in Habermas’s theoretical frameworks, an unproblematized communications hierarchy that privileges speech and the written word’.18 There can be no disputing this: even though conversation among the educated classes in eighteenth-century Europe assumed an importance which far surpasses that of the present day Western world of mobile telephony and wireless connection, and produced such metadiscourses as Swift’s celebrated

Communication and verisimilitude

‘Hints toward an essay on conversation’ (1758), verbality was not the only form of communication in the period. Work which would substantially ameliorate Habermas’ logocentrism would be centred, instead, on those public practices in the eighteenth century which could not only be revealed to be political at one level, but which were also unconfined to verbal discourse. Clearly, music – a predominantly public phenomenon – is one of those practices. If this point is given the serious attention that it demands, then current arguments about communicative strategies in late eighteenth-century music take on a political and historical importance that may not have been examined hitherto. In a set of articles published in quick succession, Howard Irving, Mark Evan Bonds and Gretchen Wheelock suggest that the communicative strategies in some of Haydn’s music were not only like conversations but were specially infused with wit, humour and, in the original German, the difficult to translate Laune.19 As Irving notes, even within a particular age and culture, some forms of wit and humor can be easily misunderstood because they tend to be hermetic by design: that is, they are calculated to be intelligible only to those possessing a specific body of knowledge or even a particular intellectual outlook.20

Such wit, then, needs to be excavated by those outside that wit’s immediate circle. Further, though, both Irving and Bonds suggest that the Laune in Haydn’s oeuvre has affinities with the wit of Sterne. Two important points for conceptualizing the public sphere arise from these comparisons, one offered by Irving, one by Bonds. Bonds argues that devices such as those found in Haydn’s String Quartet Op. 33 No. 2 (‘The Joke’) and the drumstroke in the ‘Surprise’ Symphony are delightful but also manifest a change in the traditional relationship between the composer, his work, and his audience. By violating the conventions of form so flagrantly, Haydn calls attention to them, thereby asserting his presence not only within but also outside the work. He establishes, in effect, a certain distance between himself and that which he has created. In calling attention to the craft of his art, the composer makes the listener aware of the very artificiality of that art, thereby emphasizing the gap between art as a technique and art as an aesthetic experience.21

Undoubtedly, the same kind of thing could be said of Sterne. In addition, though, Irving suggests that although neither Sterne nor Haydn were particularly rebellious in the strictly political sense, their ‘good-natured playfulness’ often ‘knowingly skirts the edge of what many in their time regarded as correct or proper’.22 These should be clear indications that the

19

20

pau l c o b l e y

rational-political discourse of the public sphere as Habermas outlines it in the world of eighteenth-century letters is also to be found in the nonverbal communication of music. The other issue proceeding from this is that the undermining of convention did not amount to outright rebellion but, rather, a more vulpine strategy of humour. For a period of approximately fifteen years at the end of the eighteenth and the beginning of the nineteenth century, humour became a frequent topic of discussion in European musical circles. As Claudia Maurer Zenck (this volume) shows, writers such as Weber, Koch, Rochlitz and Michaelis thrust humour towards the centre of the ‘musical public sphere’, especially when the topic of innovation was on the agenda. Indeed, as Maurer Zenck illustrates, such discourse repeatedly stressed that departure from the ‘usual rules’ or the ‘conventional’ should principally serve a humorous purpose. Clearly, this discourse, with its Sterne-like stress on principles of humour in the undermining of the conventional, provides crucial evidence to the task of excavating the communication that might have taken place in European music of the late eighteenth century.

Music and the public sphere in the later eighteenth century In addition to the quasi-conversational and humorous address of composers of the period such as Haydn, it is pertinent to ask where music might have been more directly a part of public sphere debate. If it was, what might musical debate in the period be like and how might it be possible to uncover its rational-political character? Schueller argues that music commentators in Britain during most of the eighteenth century were concerned chiefly with music’s ‘effect’ and saw it as serving a ‘universal harmony’ – rational, ethical, simple, clear, being the desired attributes by which to meet the heart of humans and, in turn, the mind of God.23 Indeed, it was not until later in the century that early Romanticism created the notion of music as having an intrinsic ‘effect’ specific to itself. The beginnings of Romanticism in Europe contributed, also, to the notion of the artist or author, a loading of the dice that could, potentially, downplay the receiver. At various times in the late eighteenth century, then, discourse related to music was caught between a manifestly rational imperative – in the sense of seeing a communicative logic and purpose in music – and a nascent Romantic spirit whose bearing was more inner-directed. Victor Anthony Rudowski suggests that eighteenth-century theorists of different stamps (from Burke to Lessing) were willing to embrace a ‘theory of signs’ which would enable debate on

Communication and verisimilitude

the ‘imitative’ character of communications artefacts. Indeed, the impetus of such reasoning was considerable; but, as Rudowski reveals, the debate about mimesis, spurred by revisiting the classics, rapidly fell off the intellectual agenda with the arrival of Romanticism and the emphasis on the expression of emotion.24 The public phenomenon of music, then, also experienced its own transformation in the late eighteenth century. Yet there is evidence to suggest that there had been significant public sphere activity before this took place. The case is put forward by Harold Love.25 In a powerfully argued essay, Love criticizes the very historical basis of the public sphere, arguing that there is no certainty that rational debate as Habermas describes it actually took place. He does aver that partisan debate frequently occurred and, importantly, he demonstrates that the role of music has been underestimated. Love points to the social fact of musical families in Europe where musical skill and debates about artistic merits were passed down in a domestic environment. He also focuses on the development of public concerts in taverns which metamorphosed into performances in dedicated concert rooms in Europe from around 1720 onwards. Concert promoters would work with tavernkeepers, caterers and proprietors of pleasure gardens to ensure music was performed in public. From the 1760s onwards, Love notes, regular orchestral concerts grew in number and, crucially, there was fervent encouragement from publishers to individuals to form amateur music clubs where sheet music could be sold to them. There were even ‘intermediate’ concerts falling between domestic musical activity and bona fide public concerts which took place at coffeehouses or taverns under the aegis of ‘catch clubs’. These flourished in the larger towns across Britain.26 All of these developments constitute the creation of a substantial market for music. Furthermore, they were not unconnected to other markets and other aspects of public discourse. From the beginning of the century, British newspapers advertised such guides for domestic musicians as The Compleat Tutor to the Violin,27 providing a link between the realms of public and private music making. Along with publishing in general, there was a massive growth of music publishing after 1750.28 In Britain, as elsewhere in Europe, musical societies took off, particularly choral musical societies.29 Furthermore, British musical societies were by no means concentrated in the capital; they were to be found in English provincial towns and in Scotland.30 One major institution was the Academy of Ancient Music which had been set up in 1726;31 by 1776, with the founding of the Concert of Antient Music, both institutions had gone some considerable way to establishing a canon in British musical life. William Weber argues that this canon, supplanting

21

22

pau l c o b l e y

a general neophilia in British musical consumption, was to have a heavy impact on the musical public sphere since ‘the repertories and ideas established around music by Purcell, Corelli, and Handel laid the basis for the pantheon of Haydn, Mozart, Beethoven, and J. S. Bach in the imposing classical-music world of the nineteenth century’.32 Contrary to the common idea of eighteenth-century Britain where ‘our literature was up, our music down’,33 which undoubtedly contributes to the emphasis on the world of letters in forging the public sphere, ‘it could be argued that England was the most musical country in Europe in the second half of the eighteenth century, judged by the amount of musical activity of all types’.34 Peter Holman refers to domestic music making but also notes that a music lover in London could choose between a number of rival concert series, the Italian opera, elaborate music in spoken theatres in plays, masques or pantomimes, and, in the summer, ambitious musical programmes in the various pleasure gardens . . . We must not forget, too, that the average Georgian gentleman regularly participated in a form of music-making, however humble, when he sang with his friends in the tavern. The craze for catches and glees reached such a pitch in the late eighteenth century that prizes were instituted for new compositions, and pubs even ran competitions for the best performances.35

Nor was the literate British music aficionado bereft of supporting discourse. Love notes that the famous gentleman scholars, Dr Charles Burney and Sir John Hawkins, with their histories, contributed to the canon of music in Britain – Burney in particular leaning towards Europe – but also to general discussion. For Love, the evolution from private to public is always accompanied by the creation of reading matter and, in addition to histories of music, he points to the growth of periodicals and their predilection for music reviews aimed at the connoisseur and amateur performer. ‘Reviewing’, he writes, ‘is the main manifestation within the musical public of the critical rationality identified by Habermas as the defining characteristic of the bourgeois public sphere’,36 stressing, nevertheless, that however much it is shaped by print culture, the public sphere is primarily a culture of face-to-face discussion. Love sees a public sphere in the growth of the music community in general, especially its debates in musical societies. He suggests that the music community in late eighteenth-century Europe was ‘an often fractious coalition of dispersed and competing viewpoints and interests’,37 a description of the public sphere par excellence. He also notes that while politics and religion were banned as topics for discussion in musical societies, taste was

Communication and verisimilitude

fiercely debated. Such debates should be considered ‘a way of addressing what were really ideological issues by direct means’.38 Love is keen to distinguish the theatre-going public from the musical public; to begin with, the theatre was a more established public entertainment than music by the late eighteenth century. There were no theatrical families and amateur theatrics were by no means as common as amateur music societies. But what might be important in the context of Europe and which Love does not discuss is the ‘universalist’ or ‘internationalist’ contribution to the public sphere. Above all, this depends on music’s status as a nonverbal communication. Plays could be translated, to be sure; however, as Thomas Tolley (2001) demonstrates, while the growth of a public for music in the eighteenth century was already predicated on international principles with the craze in Europe for Italian operas, such principles were to become exploited to their utmost with the promulgation of the ‘cult of Haydn’.39 As Tolley describes, Haydn’s music was promoted as a kind of ‘international language’, a strategy which was successful in Europe and beyond.40 Interestingly, too, Tolley’s thesis suggests that the growth of the music public took place in tandem with the growth of a public for the visual (as opposed to verbal) arts. Amid the growth of a public for music and its convergence with the development of the public sphere as described by Love, it is important to remember, with Habermas, that there is a structural transformation suffered by the public sphere from the late eighteenth century onwards. In the relatively short term, music, it can be argued, suffers this more acutely. The late eighteenth century saw a move from rational-ethical debate to an early Romantic basis for communication in music. The jests, humour, wit of Haydn (as well as other communication strategies in music) were caught up in the interregnum. Indeed, it may be that forms of audience address in music are symptomatic of the bleeding of rational discourse into innerdirected discourse. To begin with, a crucial basis for jesting, as Wheelock implies,41 is authorship. Not coincidentally, many attest to the fact that late eighteenth-century criticism began to focus on the originality and individuality of the author.42 Concurrently, as has been noted, debates about mimesis and imitation in aesthetics died out during this period. The status and activities of authors, or creators, of works were beginning to be viewed in a different light in Europe. Of course, encroaching Romanticism as part of the literary component of the later bourgeois public sphere is not alone in forging a reified sense of the author. As print allowed the commodification of artistic products, the establishment of copyright after the Statute of Anne, passed in England in 1710, had already gone some way in helping to create new understandings of the ‘ownership’ of artistic products,

23

24

pau l c o b l e y

introducing a fixed notion of authorship and readership.43 Furthermore, it would be remiss to ignore the paramount role of the rise of individualism in general.44 In tension with the notion of authorship that would become consolidated in Romanticism, the growth of the book trade, as has been seen, had exacerbated the knowledge of the extent to which intermediary relations are involved in communication. Rather than promoting an elevated sense of the author, vested interests were to be seen at each stage of the creative process. Faced with the possible charge of mercenary intent, a novelist such as Richardson, writing Clarissa (1747–8) at mid-century, would therefore find it crucial to make the instruction in his work as convincing as possible in order to at least appear to his readers to be circumventing the paraphernalia of publishing. As Love argues, the commercial situation of music was likewise as overdetermined as the literary field, with various interests prominent at each stage of the process of production and consumption. It would not be surprising, then, if late eighteenth century music, faced at one end with rational-political discourse and at the other with the growth of a consciousness of authorial concerns, sought to ‘bring together’ creators and audiences. If this is the case, then central to aesthetic debates is the question of ‘verisimilitude’.

Communication, genre and verisimilitude Clearly, genres can be considered as collections of specific kinds of text. These might include musical ‘texts’, too, such as string quartets or symphonies. Genre has customarily been discussed in this way, especially since the second part of the twentieth century. Yet, it would be a mistake to imagine, as has traditionally been the case, that genres can be identified simply by examining some formal qualities of texts. Rather, genre should be understood as a set of expectations on the part of audiences;45 these expectations are sufficiently prominent in audiences’ minds, and prompted by numerous cues, that they impute a great deal of flexibility in the process by which a text might be said to be part of a genre (although there may be some insuperable limiting factors, too, such as the mode of enunciation – a quartet of players would find it difficult to perform a symphony, for instance). Obviously, producers of a given text can play upon such expectations in the construction of the text from within, and the construction of it from without (through such factors as publicity), in order to address a putative audience. Indeed, even

Communication and verisimilitude

history might create expectations that neither construction from within or without would ever be able to anticipate. What is crucial to harnessing generic expectations, however, is verisimilitude. These consist of imperatives for construction both from within and from without. To put it another way, it can be argued that there are two kinds of norms by which a work or set of statements is said to have verisimilitude: the ‘rules of the genre’ and ‘public opinion’ or doxa.46 Characters bursting into song during a musical film do not commit indecorous acts when they sing: the song is part of a specific regime of verisimilitude and falls within a range of expectations on the part of the audience that such acts are legitimate within the bounds of the genre. Public opinion, on the other side, consists of a set of expectations and understandings of the world by audiences. Public opinion, of course, is not absolute reality; the doxa is, in fact, a regime of verisimilitude in itself, constantly shifting according to a complex set of checks and balances which characterize the world of discourse in general. Sometimes the ‘rules of the genre’ and the doxa meet; sometimes they do not. Genres are therefore a specific means of encoding in order to enable a similarly specific decoding. Verisimilitude is a means of bringing together a sender and receiver through genre, a means of invoking sets of expectations of generic norms in relation to discourses about the world. It is no coincidence, then, that Bonds argues, in reference to a comparison with the ironic strategies in Haydn’s instrumental music, that Sterne’s writing is symptomatic of the way in which ‘the reader was being drawn into the workings of genres’ in the late eighteenth century.47 In a later essay (see this volume), Bonds also discusses the way in which composers in the late eighteenth century were often considered as orators or rhetoricians. In a compelling account, he shows how composers such as Mozart, Haydn and others faced the dilemma of bringing together their music and their publics, not simply through envisaging an ideal listener, but through considering the marketplace and the less-than-ideal listener. Indubitably, success in this enterprise relied on accurate gauging of expectations of audiences and an appeal to mutual grounds of verisimilitude. As Bonds shows, this does not necessarily lead to consensus or quash innovation. Likewise, Sterne, Haydn and others were necessarily operating on doxological grounds at the time that they challenged the artistic parameters set by their peers or masters. Yet, the challenge could still be mounted because verisimilitude was a delicate concept in the eighteenth century. As now, it should not be considered as ‘veracity’ or ‘relation to truth’, but more to do with being rhetorically

25

26

pau l c o b l e y

convincing (see Bonds, this volume). Note, for example, the ambivalence of Johnson’s definition of the word in 1755:48 verisi′ militude verisi′ mility

} n.s. [verisimilitudo, Lat.] } Probability; likelihood; resemblance of truth.

Touching the verisimility or probable truth of this relation, several reasons seem to overthrow it. Brown A noble nation, upon who if not such verities, at least such verisimilities of fortitude were placed. Brown Verisimilitude and opinion are an easy purchase, but true knowledge is clear and difficult. Like a point, it requires an acuteness to its discovery: while verisimilitude, like the expanded superficies, is obvious, sensible, and affords a large and easy field for loose enquiry. Glanville The plot, the wit, the characters, the passions, are exalted as high as the imagination of the poet can carry them, with proportion to verisimility. Dryden Though Horace gives permission to painters and poets to dare everything, yet he encourages neither to make things out of nature and verisimility. Dryden

Compare, too, with Johnson’s entries on Verisimilar, Verisimilous and Putative: verisi′ milar verisi′ milous

} adj. [verisimilis, Lat.] } Probable; likely

Many erroneous doctrines of pontificians are, in our days, wholly supported by verisimilous and probable reasons. White put′ ative

} adj. [putatif, Fr. from Latin puto] } Supposed; reputed

If a wife commits adultery, she should lose her dower, though she be only a putative, and not a true and real wife. Ayliffe

There is a clear sense from the quotes that Johnson gives that verisimilitude is tied to public opinion and consensual understanding, even though it might serve as a kind of ‘truth of the day’. It is probable that this take on verisimilitude is derived from the importance of the vraisemblable which was so central to the French classical theatre of the previous century. As Jeffrey Peters argues, the seventeenth-century aesthetic of vraisemblable ultimately involved inviting audiences to witness

Communication and verisimilitude

a ‘social truth’ through the coherence of a communication. For him, this also leads to social actors mistaking the cultural values of verisimilitude for universal values.49 It is not too difficult to imagine how this might take place in the verbal world of the theatre, but possibly less easy to envisage in the communicative strategies of late eighteenth-century music. Nevertheless, those strategies are manifestly a means of placing the composition of music and particular audiences on the same firm grounds of evaluation. They are verisimilar in the sense that they recognize what is decorous, drawing attention to the very form, as Bonds suggests,50 while also softly violating it, invoking the audience in the process to enact its knowledge of the form and its interest in the partial contravention. Furthermore, the aspirations of music in the late eighteenth century to become an ‘international language’ imply the belief that the meeting of musical composition with a knowing audience is capable of eradicating local and national differences, even if only on a temporary basis. If verisimilitude is central to the generic expectations of the music public, then it is always pulled in two ways: towards the doxa, a realm of public opinion and rational, if not conclusively true discourse; and towards the ‘rules of the genre’, the expert knowledge of composition. In eighteenth-century European artistic appreciation, it is very likely that this dual compulsion would have specific resonances. Certainly, it is as likely as the specificity of references in, say, Tristram Shandy’s humorous undermining of literary convention or the social imperatives of economy and effort in Robinson Crusoe. In the process of being pulled in two ways, it is not the case that verisimilitude will be divided, however tempting it might be to see verisimilitude as constituting the preserve of highly specific audiences. Verisimilitude is likely to be generally located, for example, historically and regionally; but, for music, it is more difficult to posit verisimilitude as the preserve of minutely localized audiences. Bonds (this volume) refers to a ready-made, but not necessarily formally theorized, eighteenth-century division of audiences. The distinction between Kenner and Liebhaber, derived in part from C. P. E. Bach’s identification of connoisseur and amateur players, seems to imply also two different kinds of audiences for music. The Kenner audience would be most fixated on musical convention and artistry or, in the terms of verisimilitude, the rules of the genre. The Liebhaber audience would be less committed to technical matters and, as a group of listeners, would be involved in music generally at the same level of accomplishment at which, as amateur musicians, they could play.51 It would be convenient if the Kenner could map onto an appreciation of formal, ‘rule-based’ aspects of music while the Liebhaber were similarly congruent with an amateur audience interested mainly in

27

28

pau l c o b l e y

‘social truth’. Yet, as Bonds suggests, Haydn, and the music scene in which he flourished, was in some measure dedicated to producing music which might satisfy both Kenner and Liebhaber alike. In this way, the distinction is a fairly loose one. In addition, Elaine Sisman shows that there is evidence to suggest that Kenner and Liebhaber existed with a range of different music audience categories in the late eighteenth century, not just as a couplet.52 Most telling, though, is the fact that, even in the present day,53 it is difficult to clearly bisect a music audience in terms of connoisseurs and amateurs. In the eighteenth century, likewise, it would probably be impossible to identify an audience – or even individual audience members – according to who was concerned with either doxological or formal verisimilitude alone. Verisimilitude, then, is where doxa and generic rules converge in communication. Unsurprisingly, with these checks and balances, adjustments to verisimilitude are often required to proceed gently and with humour.

Conclusion Artistic life in the late eighteenth century was located in an interregnum. Of course, an interregnum can be claimed for many periods in history, particularly cultural history. Moreover, the specific nature of this interregnum is not always immediately apparent; it is buried from view and not always evident from a surface glance at cultural practices. Yet, work on communicative strategies in music of the late eighteenth century is very suggestive. What those strategies entailed at the time of their production and initial consumption is much different from what they can be discerned to entail in the present. Additionally, they are not unrelated to a contesting of issues of form in arts such as novel writing. Above all, though, these communicative strategies might constitute a more or less unwitting compromise position between two discursive–aesthetic currents. The first is associated with the ‘struggle’ (although not a desperate one) for music in the first years of the second half of the eighteenth century in Europe; that is, a period when musical societies and other agencies promoted an active discourse about music which could be understood as political in the ‘pure’ sense. Here, music contributed to the public sphere and certainly is to be associated with the doxa. The burgeoning exigencies of commerce and the growing public status of music dictated that those at the production end of the music supply chain be in close communion with their public. The second current is associated with the emergence of Romanticism in the arts, the consolidation of the notion of authorship and those moments

Communication and verisimilitude

when artists might forgo financial gain in order to celebrate their art for its own sake. In respect of the latter, Bonds (this volume) gives an instance of Mozart making just such a decision. Clearly, the Romantic conception of artistic production is different from the rational-political one. There is a greater sense of authorial talent, the expression of emotion and the purity of the ‘rules of the genre’. On this basis the second current, too, involves attempts to forge a communion of the production and consumption points in music supply. Yet it is a much different kind of communion, based on principles of artistry rather than principles of reference or imitation. Still, as Maurer Zenck (this volume) reveals, commerce cannot be occluded. N¨ageli’s commission of Beethoven’s Op. 31 sonatas was a financially painstaking transaction despite the innovative nature of the resulting music. Indeed, N¨ageli deliberately showcased the sonatas for the public as innovations in which to delight. The strategies of communication in late eighteenth-century European music, especially those featuring Laune, would seem to be an attempt to navigate a route between the Scylla and Charybdis described above. Knowing the importance of their art and their public, yet operating in an artistic environment where matters of form and purity of expression were beginning to take hold, it is hardly surprising that composers might wink at their audiences. As with Sterne’s Tristram Shandy, it can be concluded that these strategies amount to an injection of sanity into proceedings, whereby producer and consumer are momentarily brought together to ask each other whether they really want to settle for the current proposals of verisimilitude. Notes 1 Angus Ross, ‘Introduction’, in Daniel Defoe, Robinson Crusoe (Harmondsworth: Penguin, 1965), 7–21. 2 See below, note 19. 3 John Durham Peters, Speaking into the Air: A History of the Idea of Communication (Chicago: University of Chicago Press, 1999). 4 See Paul Cobley, ‘General Introduction’, in Communication Theories, ed. Paul Cobley, 4 volumes (London: Routledge, 2006), volume 1, 1–33. 5 Claude E. Shannon and Warren Weaver, The Mathematical Theory of Communication (Urbana: University of Illinois Press, 1949). 6 See, for example, George Gerbner, ‘Toward a General Model of Communication’, Audio-Visual Communication Review 4 (1956), 171–99; Bruce Westley and Malcolm MacLean, ‘A Conceptual Model for Communication Research’, Journalism Quarterly 34 (1957), 31–8. One point that is worth noting, though, in considering why the matter was revisited so frequently is that much

29

30

pau l c o b l e y

7

8

9 10

11

12 13

communication theory, as Peters, Speaking into the Air, also argues, proceeded from an agenda set by works of the 1920s which were concerned with issues of the ‘crowd’, ‘mass’ or ‘public’ behaviour. On the one hand, this represented an impetus to understand ‘successful’ propaganda; but, also, there was a desire to unravel the components of communication to determine which elements and which intersections of elements in the communication act might be susceptible to the cause of misunderstanding. This model is proposed in my Narrative (London: Routledge, 2001), 139, where I adapt it with some adjustments from Shlomith Rimmon-Kenan, Narrative Fiction: Contemporary Poetics (London: Methuen, 1983), 86. See also Walker Gibson, ‘Authors, Speakers, Readers and Mock Readers’, College English 11 (1950), 265–9, reprinted in Reader-Response Criticism: From Formalism to PostStructuralism, ed. Jane P. Tompkins (Baltimore and London: Johns Hopkins University Press, 1980), 1–6. For example, David Morley, The Nationwide Audience (London: British Film Institute, 1980); Dorothy Hobson, Crossroads: The Drama of a Soap Opera (London: Methuen, 1982); Janice A. Radway, Reading the Romance: Women, Patriarchy and Popular Literature (Chapel Hill and London: University of North Carolina Press, 1984); Klaus Bruhn Jensen and Karl Erik Rosengren, ‘Five Traditions in Search of the Audience’, European Journal of Communication, 5 (1990), 207–38; Joke Hermes, Reading Women’s Magazines: An Analysis of Everyday Media Use (Oxford: Polity, 1996); Ellen Seiter et al., eds., Remote Control: Television, Audiences and Cultural Power (London: Routledge, 1989); Sonia M. Livingstone, ‘The Challenge of Changing Audiences: Or, What is the Audience Researcher to Do in the Age of the Internet’, European Journal of Communication 19 (2004), 75–86. Hans Robert Jauss, Toward an Aesthetic of Reception (Minneapolis: University of Minnesota Press, 1982). Gunther R. Kress, ‘Against Arbitrariness: The Social Production of the Sign as a Foundational Issue in Critical Discourse Analysis’, Discourse and Society 4 (1993), 169–91. There is even a sense in which this might be found to have a bio-chemical dimension: Thomas A. Sebeok, A Sign is Just a Sign (Bloomington: Indiana University Press, 1991), 30, points out that ‘messages encoded in the chemicals isolaveric acid and methyl mercaptan are components, respectively, of human body malodor and halitosis. This notwithstanding, the same chemicals are responsible for some of the bouquet and flavor of cheese.’ John Feather, ‘The Commerce of Letters: The Study of the Eighteenth-Century Book Trade’, Eighteenth Century Studies 17 (1984), 405–24. Gordon Williams, ‘Newsbooks and Popular Narrative During the Middle of the Seventeenth Century’, in Narrative: From Malory to Motion Pictures, ed. Jeremy Hawthorn (London: Edward Arnold, 1985), 203–6; J. Paul Hunter, Before Novels:

Communication and verisimilitude

14

15

16

17

18 19

20 21 22 23 24 25 26

27 28

The Cultural Contexts of Eighteenth-Century Fiction (New York and London: Norton, 1990), 193. Jack Goody and Ian Watt, ‘The Consequences of Literacy’, in Literacy in Traditional Societies, ed. Jack Goody (Cambridge: Cambridge University Press, 1968), 27–68. J¨urgen Habermas, The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society, trans. Thomas Burger and Frederick Lawrence (Cambridge, MA: The MIT Press, 1989). J¨urgen Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans. William Rehg (Cambridge, MA: The MIT Press), 360. Ross, ‘Introduction’, in Selections from The Tatler and The Spectator, ed. Angus Ross (Harmondsworth: Penguin, 1982), 37. For criticism of Habermas’ thesis as a failure to see public sphere discourse in the context of its pragmatic Whig bearing, see Brian William Cowan, ‘Mr. Spectator and the Coffeehouse Public Sphere’, Eighteenth-Century Studies 37 (2004), 345–66, and for not being sufficiently aware of the workings of media, see John B. Thompson, ‘Social Theory and the Media’, in Communication Theory Today, ed. David Crowley and David Mitchell (Cambridge: Polity, 1992), 27–49, and Mark Poster, The Second Media Age (Cambridge: Polity, 1995). Luke Goode, J¨urgen Habermas: Democracy and the Public Sphere (London: Pluto, 2005), 89. Howard Irving, ‘Haydn and Laurence Sterne: Similarities in Eighteenth-Century Literary and Musical Wit’, Current Musicology 40 (1985), 34–49; Mark Evan Bonds, ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’, Journal of the American Musicological Society 44 (1991), 57–91; and Gretchen A. Wheelock, Haydn’s Ingenious Jesting with Art: Contexts of Musical Wit and Humor (New York: Schirmer Books, 1992). Irving, ‘Haydn and Laurence Sterne’, 34. Bonds, ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’, 71–2. Irving, ‘Haydn and Laurence Sterne’, 47. Herbert M. Schueller, ‘The Use and Decorum of Music as Described in British Literature, 1700–1780’, Journal of the History of Ideas 13 (1952), 73–93. Victor Anthony Rudowski, ‘The Theory of Signs in the Eighteenth Century’, Journal of the History of Ideas 35 (1974), 683–90. Harold Love, ‘How Music Created a Public’, Criticism 46 (2004), 257–71. Stanley Sadie, ‘Music in the Home ii’, in The Blackwell History of Music in Britain: The Eighteenth Century, ed. H. Diack Johnstone and Roger Fiske (Oxford: Blackwell, 1990), 320. Roger Fiske, ‘Music and Society’, in The Blackwell History of Music in Britain: The Eighteenth Century, 4. Sadie, ‘Music in the Home ii’, 313.

31

32

pau l c o b l e y

29 Love, ‘How Music Created a Public’; Peter Holman, ‘Eighteenth-Century English Music: Past, Present and Future’, in Music in Eighteenth-Century Britain, ed. David Wyn Jones (Aldershot: Ashgate, 2000), 4. 30 Fiske, ‘Music and Society’, 24; Holman, ‘Eighteenth-Century English Music’, 3. 31 Fiske, ‘Music and Society’, 18. 32 William Weber, The Rise of Musical Classics in Eighteenth-Century England: A Study in Canon, Ritual and Ideology (Oxford: Clarendon Press, 1996), 1–2. 33 Fiske, ‘Music and Society’, 3. 34 Holman, ‘Eighteenth-Century English Music’, 3. 35 Ibid., 3. 36 Love, ‘How Music Created a Public’, 265. 37 Ibid., 266. 38 Ibid., 267. 39 Thomas Tolley, Painting the Cannon’s Roar: Music, the Visual Arts and the Rise of an Attentive Public in the Age of Haydn (Aldershot: Ashgate, 2001). 40 Ibid., 23. 41 Wheelock, ‘Engaging Strategies in Haydn’s Opus 33 String Quartets’, Eighteenth Century Studies 25 (1991), 1–30. 42 For a summary, see Jean Marsden, ‘British Theory and Criticism: 2. Late Eighteenth Century’, in Johns Hopkins Guide to Literary Theory and Criticism, ed. Michael Groden and Martin Kreiswirth (Baltimore and London: Johns Hopkins University Press, 1997), 108–12. 43 David Saunders, Authorship and Copyright (London: Routledge, 1992); Michel Foucault, ‘What is an author?’, in The Foucault Reader, ed. Paul Rabinow (Harmondsworth: Penguin, 1986), 101–20. 44 As witnessed in the work of Ian P. Watt; see, for example, The Rise of the Novel: Studies in Defoe, Richardson and Fielding (Harmondsworth: Penguin, 1963) and Myths of Modern Individualism: Faust, Don Quixote, Don Juan, Robinson Crusoe (Cambridge: Cambridge University Press, 1997). 45 Rick Altman, Film/Genre (London: British Film Institute, 1999); Paul Cobley, ‘Objectivity and Immanence in Genre Theory’, in Genre Matters: Interdisciplinary Perspectives, ed. Garin Dowd et al. (London: Intellect, 2005), 41–54; Merja Makinen, Feminist Popular Fiction (London: Palgrave, 2001). 46 Tzvetan Todorov, The Poetics of Prose (Ithaca: Cornell University Press, 1977). 47 Bonds, ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’, 85. 48 Samuel Johnson, A Dictionary of the English Language in Which the Words Are Deduced from their Originals and Illustrated in their Different Significations by Examples from the Best Writers to Which are Prefixed a History of the Language and an English Grammar (London: Printed by W. Strahan, for J. and P. Knapton, T. and T. Longman, C. Hitch and L. Hawes, A. Millar, and R. and J. Dodsley, 1755). 49 Jeffrey N. Peters, ‘Ideology, Culture and the Threat of Allegory in Jean Chapelain’s Theory of la vraisemblance’, Romanic Review 89 (1998), 491–505.

Communication and verisimilitude

50 Bonds, ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’. 51 Matthew Riley, ‘Johann Nikolaus Forkel on the Listening Practices of “Kenner” and “Liebhaber”’, Music and Letters 84 (2003), 414–33 and Musical Listening in the German Enlightenment: Attention, Wonder and Astonishment (Aldershot: Ashgate Press, 2004); Annette Richards, ‘Listening for Kenner and Liebhaber’ [Review of Riley, Musical Listening in the German Enlightenment], Early Music 33 (2005), 332–4. 52 Elaine Sisman, ‘Haydn’s Career and the Idea of the Multiple Audience’, in The Cambridge Companion to Haydn, ed. Caryl Clark (Cambridge: Cambridge University Press, 2005), 3–16. 53 Evidence is offered by the audience feedback to any serious classical music radio station.

33

2

Listening to listeners m a r k e va n b o n d s

It is one of the foundational principles of rhetoric that a speaker must take into account the audience being addressed. Every manual of rhetoric ever written includes some exhortation along the lines that a speech must be appropriate to the occasion, the place and, above all, to the audience. When Aristotle identified three elements in speech-making – the speaker, the subject and the audience – he declared the last of these to be the one that determines the speech’s purpose and object.1 And ever since, rhetoricians have crafted their presentations according to the backgrounds, interests, and competencies of their listeners. In preparing a speech for a specific audience – be it the Roman senate, a class of school children or a conference of music scholars – a speaker will fashion the speech at hand according to the level of knowledge, predisposition and purpose of those assembled. But how to address a more heterogeneous audience? Here, the rhetorical manuals tend to be of little help, for speeches are by their very nature events of a particular time and place, directed toward a particular audience. The few commentators on rhetoric who do address this problem advocate what amounts to little more than common sense. Orators, we are told, must seek a middle path, neither too high nor too low, neither too demanding nor too simple. Don’t make your speech too sophisticated, but don’t make it too plain, either. Make it just right. We might well call this the Goldilocks school of rhetoric. An excerpt from George Campbell’s widely read Philosophy of Rhetoric (1776) gives a good example of this kind of approach. The Scottish cleric recommended that whatever is advanced in a speech shall be within the reach of every class of hearers, and shall not unnecessarily shock the innocent prejudices of any. This is still, however, to be understood with the exception of mere children, fools, and a few others who, through the total neglect of parents or guardians in their education, are grossly ignorant. Such, though in the audience, are not to be considered as constituting a part of it. But how great is the attention requisite in the speaker in such an assembly, that, whilst on the one hand he avoids, either in style or in sentiment, soaring above the capacity of the lower class, he may not, on the other, sink below the regard of the higher. To attain simplicity without flatness, delicacy without refinement, perspicuity without recurring to low idioms and similitudes, will require his utmost care.2

Listening to listeners

Composers of the late eighteenth century, often likened to orators in their time, frequently wrote with specific performers and audiences in mind. But more often than not, they faced the challenge of writing for an undefined, abstract audience, diverse in its competencies. This was particularly true in the case of music written to be sold in the commercial marketplace and performed by the very individuals purchasing that music. For composers of this era were necessarily writing for a double audience of both performers and listeners: not only could the music not sound too difficult, it also had to be within the technical grasp of those most likely to perform it. And in the case of chamber music, performers – which is to say, the buying public – consisted largely of non-professional musicians. Composers, in short, had to be conscious of both the aesthetic and technical limitations of consumers. This is not the kind of challenge composers of the late eighteenth century were given to talk about, at least not publicly. It was such an accepted part of the composer’s lot that it scarcely required discussion. But in private correspondence, in contemporary discourse about music, and in the music itself, we can find enough scattered clues to suggest at least some of the ways in which these considerations shaped what composers wrote. In Mozart’s case, the question of how to accommodate public taste is something of a recurring theme in the correspondence with his father. Leopold Mozart, writing to his son in Paris in 1778, spelled out the inherent tensions in the conflicting demands between art and commerce and urged his son to pursue a middle path by creating works of artistic merit that nevertheless appeal to a heterogeneous public: If you have no pupils at the moment, then compose something more, even if you have to let your work go for a smaller sum; for God’s sake you have to make yourself known. Only short – easy – popular. Talk with an engraver about what he would most like to have; perhaps easy quartets for two violins, viola and basso. Do you believe perhaps that you lower yourself by such things? In no way! Did Bach in London [J. C. Bach] ever write anything other than such trifles [Kleinigkeiten]? The small is great if it is natural – if it is flowing and easy and soundly composed. It is harder to do this than to write all those artificial harmonic progressions that are incomprehensible to most people and melodies that are difficult to perform. Did Bach lower himself by doing this? In no way! Good composition and ordering, il filo – these distinguish the master from the bungler, even in trifling works . . . You must now make an effort to sell something to one engraver or another. Must you not have money in order to live? And how can you make money otherwise when all your pupils are off in the countryside?3

35

36

m a r k e va n b o n d s

We can see these same ideas reflected in the son’s subsequent, oft-quoted letter of 28 December 1782, in which he assures Leopold that his three latest piano concertos (k413–15) are precisely the mean between that which is too difficult and that which is too easy – they are very brilliant, pleasing to the ear, natural without falling into emptiness. Although here and there connoisseurs alone can obtain satisfaction, the non-connoisseurs will be satisfied, without knowing why. The mean – that which is true in all things – is never recognized and valued nowadays. In order to win applause one has to write things that are so clear a coachman could sing it by ear, or so unclear – that it pleases people precisely because no rational person can understand it.4

As Elaine Sisman has pointed out, both of these passages are part of an ongoing conversation between father and son about finding ‘the mean’ (das Mittelding), about negotiating the fine line between artistic integrity and commercial appeal, about writing works that can be performed and appreciated by a wide range of musicians and listeners and yet at the same time rise to the highest standards of art.5 Mozart does not spell out here just what he means by ‘too difficult’ or ‘too easy’, but it would seem from other contemporary sources that listeners of the time categorized the relative ease or difficulty of works based on the technical demands placed upon the performers, the complexity or simplicity of the harmonies, the number and nature of the musical ideas, the general trajectory of form and, above all, the extent and nature of the counterpoint. Imitative counterpoint in particular, by all accounts, provided something of a litmus test dividing Kenner from Liebhaber. Yet even with these categories in mind, Mozart’s comments are in the end no more substantive than Campbell’s, and we would be hard pressed to find in these three particular concertos anything that makes them different in kind from those Mozart had written before or from those he would write later. Still, the framework of the comment is worth taking seriously. Even if Mozart was characterizing these concertos in this way simply because he knew that this was the kind of thing his father wanted to hear, he nevertheless recognized a basic compositional imperative of writing in a way that would appeal as much to the masses as to the elite. Mozart’s awareness of this challenge to reconcile the high and the low clearly shaped his approach toward his art in general and toward certain works in particular. In another oft-quoted passage from a letter to his father, written from Paris in 1778, he describes in detail the way in which he had calculated the expectations of Parisian audiences while writing his Symphony in D major, k297/300a:

Listening to listeners

The symphony [k297] began . . . and in the middle of the opening Allegro there was a passage that I knew quite well would please; all the listeners were swept up in it, and there was great applause. But because I knew exactly how I had done this and knew what kind of effect it would make, I brought it back again at the end [of the movement], and there it was all over again. The Andante and the Allegro finale also pleased, but particularly the latter, because I had heard that all finales here begin with all the instruments together, and usually in unison, and so I began with just two violins alone, playing softly for eight measures; then came a forte, followed at once by a piano, and all the listeners (as I expected) said ‘shhhhh’ at that moment. Then the forte came back, and when they heard that, the sounds of the forte and of the applause were one.6

Mozart was not alone in anticipating audience reactions as he composed. Haydn was particularly anxious to appeal to a broad listenership even while cultivating the approval of connoisseurs. ‘I would only wish and hope that the critics do not deal too severely with my Creation’, he wrote in confidence to the publisher Christoph Gottlob Breitkopf in 1799. ‘They might perhaps object a little to the musical orthography of certain passages, and possibly some other minor points elsewhere; but every true connoisseur will see the reasons for these as readily as I do and will push aside this stumbling block.’7 The same concerns would surface in a conversation with Griesinger a few years later. ‘I was never a fast writer’, Haydn confessed, ‘and always composed with circumspection and diligence. But at the same time, such works are for the long run, and a connoisseur [Kenner] will discover this at once from the score.’ Haydn seems to have been particularly proud that when Cherubini examined some of his scores he always recognized those particular passages that Haydn himself felt were worthy of praise.8 The musical public also recognized and responded to these efforts to reach Kenner and Liebhaber alike in one and the same work. Haydn won repeated praise during his lifetime for writing music that in spite of its technical sophistication was immediately comprehensible by a wide range of listeners. Ernst Ludwig Gerber, for example, observed in 1790 that for all its contrapuntal sophistication, Haydn’s music ‘often seems familiar’.9 Johann Karl Friedrich Triest asserted in 1801 that ‘If one had to distill the essence of Haydn’s art into two words, they would be “artful popularity” [kunstvolle Popularit¨at] or “popular artfulness” [popul¨are Kunstf¨ulle]’.10 And in 1812 Giuseppe Carpani expounded at length on Haydn’s ability to write music that was ‘clear yet erudite’, ‘learned but not dry’ and ‘melodious but not trivial’. In this respect, Carpani noted, Haydn distinguished himself from composers like Ignaz Pleyel, whose melodies and harmonies were ‘too simple’, and composers like Mozart and Beethoven, whose works were

37

38

m a r k e va n b o n d s

overly difficult – ‘full of the most erudite, intricate confusions’ and ‘strange modulations’ and therefore ‘lacking in effect’.11 Such responses reflect the eighteenth-century belief that art and entertainment were by no means incompatible. As James Webster has demonstrated, more recent attempts to equate Haydn’s ‘popular’ style with a ‘pandering’ to the lowest common denominator are fundamentally anachronistic, in that they rest on nineteenth- and especially twentieth-century aesthetic notions that value the learned over the popular, the challenging over the readily accessible. Webster has convincingly argued that ‘there is no reason why an approach to composition as entertainment should be considered either morally suspect or inherently incompatible with the production of great art’.12 But while the artful and the popular may not have been mutually exclusive, neither did they go hand in glove, and the fact that critics of the late eighteenth and early nineteenth centuries singled out Haydn so pointedly for synthesizing the two underscores the uncommon nature of his achievement. Haydn himself seems to have struggled in this respect at least on occasion, as witnessed by his cryptic remark next to a cancelled passage in the autograph score of the slow movement of Symphony No. 42: ‘Dieses war vor gar zu gelehrten Ohren’ – ‘This was for far too learned ears’.13 This comment may well record a process that routinely went unrecorded – namely, Haydn making his music more accessible by making it less intricate. This same kind of struggle to reconcile artistry and accessibility can be found in accounts of other composers of the late eighteenth century as well. Carl Philipp Emanuel Bach privately acknowledged to Alexander Reinagle, a fellow composer, that he had incorporated rondos into various of his keyboard publications aimed expressly at ‘Kenner und Liebhaber’ in order to ‘promote sales’. ‘I know from experience’, he confided, ‘that many persons buy my collections merely because of the rondos’.14 But this attention to popular taste did not always sit well with the musical elite. Charles Burney reported that when he visited Bach in Hamburg in 1772, the composer played for him ‘his last six concertos, lately published by subscription, in which he has studied to be easy, frequently I think at the expense of his usual originality’.15 Bach himself confessed to Burney that because he had composed most of his works ‘for specific individuals or for the public’, he had always been ‘more constrained’ than when fashioning those ‘few pieces I have written merely for myself’. Yet he acknowledged that it was possible that ‘these not exactly pleasant circumstances’ might also have ‘challenged’ his ‘genius’ to ‘certain novelties’ that would not otherwise have come to him.16 Carl Friedrich Abel similarly recognized that a composer’s awareness

Listening to listeners

of the listening public could act as a positive force, and Burney concurred with Abel’s observation that ‘the genius of Emanuel Bach would have been more expanded and of more general use, if, instead of being confined to the Hans town of Hambro’ [Hamburg] without a rival, he had resided in a great capital, such as Paris, Naples, or London, where he would have been stimulated to diligence and exertion by competitors, and obliged to study and respect the public taste’.17 From time to time this ‘public taste’ is unmistakably inscribed into the music itself: the multiple false endings and the true ending that sounds like a false ending in the finale of Haydn’s String Quartet, Op. 33 No. 2, the unprepared fortissimo chord early on in the slow movement of the ‘Surprise’ Symphony, the reference to ‘God Save the King’ at the beginning of the slow movement to Symphony No. 98, and the multiple formal and harmonic ‘errors’ in Mozart’s Ein musikalischer Spaß, k522. Haydn and Mozart have calculated listeners’ expectations into these and other compositions in ways that even the most tin-eared of audience members can appreciate. But what about other works – the large majority of works, in fact – that do not play to the galleries quite so obviously, or that lack the kind of commentary Mozart provided for k297? When we analyse these works, we typically posit a kind of ideal listener of the time, a retrospective projection of ourselves, as it were, a listener with a wide knowledge of repertory, genres, styles and techniques. There can be no doubt that this sort of approach has deepened our understanding of these works; yet it suffers from a serious shortcoming in so far as it takes only Kenner into account, for our ideal listener is invariably a connoisseur and not an amateur. This scarcely does justice to the full range of the ‘public taste’ that C. P. E. Bach, Haydn, and Mozart had in mind when they contemplated the challenge of writing a work that would appeal to Kenner and Liebhaber alike. The imperative of appealing to a diverse audience of performers and listeners would become all the more acute in the second half of the eighteenth century when the rapid growth of music publishing created an unprecedented financial incentive for reaching the broadest possible market.18 When publishing firms in Vienna began to commit themselves in a serious way to music for the first time in the late 1770s and early 1780s, Haydn and Mozart were quick to establish working relationships with several of them, including Hoffmeister, Torricella and, above all, Artaria. From 1781 onward, Mozart published the overwhelming majority of the first editions of his music with the last of these firms.19 Most of the music published, not surprisingly, consisted of works that could be performed in the home: solo and chamber genres, and even concertos, which could be played domestically with a

39

40

m a r k e va n b o n d s

reduced orchestra of one player to a part. When Mozart wrote ad libitum oboe and horn parts for his Piano Concertos k449, k450 and k451, he was casting an eye toward a wider commercial market. And on one of several occasions when he requested a loan from his Masonic brother Michael Puchberg, he pointedly called attention to the fact that he was in the process of writing a set of ‘easy’ keyboard sonatas – the clear implication being that these would do better in the marketplace than more technically demanding works.20 But the question remains: How do we get beyond such general and fairly obvious considerations in analysing this music? Scholars who have examined the works of Haydn, Mozart and Beethoven as commercial commodities have added much to our understanding of how these and other composers approached their art in the late eighteenth century,21 yet economic expediency and analytical interest rarely overlap, save at the most superficial of levels. Generally, the dynamic runs along the following lines. We acknowledge that the composer shapes the work at hand to accommodate one or more various external parameters, such as the nature of a particular singer’s voice, the tastes of a dedicatee, the technical abilities of the average amateur performer, the plan to market a concerto to a domestic audience, or the desire to make a quick profit. And as analysts, we identify the various manifestations of these motivations: the alterations for the Viennese production of Don Giovanni, the prominence of the cello part in the ‘Prussian’ Quartets, the presence of ad libitum wind parts in a piano concerto or the technically undemanding nature of certain piano sonatas. But in the end, these external considerations remain outside the roodscreen that separates the sanctuary of analysis from the mundane demands of economic life. External considerations, when they can be documented, are readily acknowledged but rarely embraced. They are perceived as a kind of necessary evil, an uncomfortable reminder that the artwork at hand also happens to have been a commercial commodity in the mind of the composer who created it. The tendency to see aesthetic and commercial demands as opposite sides of a single coin does not do justice to the full range of compositional choices that make a given composition what it is.22 If one of the purposes of analysis is to reconstruct, in our minds, the creative process within the mind of the composer – why this harmonic turn at this particular moment? why a new theme here? and so on – then it behooves us to take into account the full range of plausible contingencies that might have come to bear on that creative process. And the competencies of a typical Liebhaber, as the composers of this time repeatedly acknowledged, were clearly among the elements that shaped the compositional matrix. In this sense, our ideal listener, posited

Listening to listeners

for purposes of analysis, needs to be even more of an abstraction, in so far as that construct must embrace both Kenner and Liebhaber at once. This may seem like something of a contradiction, but it is the kind of paradox with which composers writing for an abstract, heterogeneous market dealt with on a regular basis. This seemingly less-than-ideal ideal listener can in fact help us better understand the ways in which composers went about their art. Let us consider, as a case in point, Mozart’s ‘Haydn’ Quartets, generally regarded as one of the most Liebhaber-unfriendly set of works published in the second half of the eighteenth century. This reputation is well deserved, for the set poses a number of challenges to performers and listeners alike. The quasi-fugal finale of k387, the extreme chromaticism of the slow movement of k428, the opening of the ‘Dissonance’ Quartet, k465 – all these are well-known examples of a challenging idiom and were recognized as such in Mozart’s time. An anonymous reviewer in Carl Friedrich Cramer’s Magazin der Musik declared in 1787 that these ‘new quartets . . . are certainly much too heavily seasoned – and what palate can tolerate this for very long?’ Carl Dittersdorf, in turn, was well aware of the set’s limited appeal in the marketplace when he wrote to Artaria in 1787 to offer his own set of six string quartets, promising that these would do better financially than Mozart’s had two years before. ‘I . . . am certain that you will recover with mine the profit you have forgone through your investment in Mozart’s quartets’, Dittersdorf assured the firm, noting that even though Mozart’s quartets ‘deserve the highest regard from me and from far greater theorists, they are not everyone’s kind of purchase, because of the overwhelming art that predominates in them’.23 It is of course precisely this ‘overwhelming art’ that we value so highly today in Mozart’s ‘Haydn’ Quartets. Dittersdorf recognized and praised this quality but emphasized the problems it raised in the marketplace. So did Mozart simply forgo any attempt to write for a broader audience here? Many scholars have assumed that he did – in part because of the supposed nature of the genre, in part because of anachronistic attitudes about the relationship between art and commerce. But in the 1780s the string quartet was no more or less prestigious than any other genre, and no more or less immune to the challenge of reaching a heterogeneous music-loving public.24 It was the string quartet, after all, that Leopold had urged his son to take up a few years before in order to gain both fame and fortune. The notion that Mozart might have abandoned all consideration of public taste in writing these quartets, moreover, is fundamentally suspect. If we look again at the wording of Mozart’s letter of 28 December 1782 about the three concertos, k413–15, we can see that he constructs the problem of reaching a

41

42

m a r k e va n b o n d s

heterogeneous audience not as a matter of simplifying that which is complex, but of making more complex that which is already accessible. Finding the mean is not in the end a matter of splitting the difference between learned and popular so much as tempering a fundamentally popular style. ‘Hie und da – k¨onnen auch kenner allein satisfaction erhalten – doch so – daß die nicht-kenner damit zufrieden seyn m¨ussen, ohne zu wissen warum.’ Emily Anderson’s widely used translation – ‘there are passages here and there from which the connoisseurs alone can derive satisfaction; but these passages are written in such a way that the less learned cannot fail to be pleased, though without knowing why’ – ignores that little word ‘auch’ that appears just before ‘Kenner allein’, thereby eliminating the faint sense of defensiveness Mozart betrays in the original. The ‘auch’ is an abbreviated way of saying, ‘in spite of the fact that certain passages are written in a more demanding fashion . . .’ What Mozart is being defensive about are a few passages ‘here and there’, not any of the three concertos as a whole. The default mode of writing, by Mozart’s own account, is to appeal first and foremost to the non-connoisseurs and to create moments from which connoisseurs alone will obtain satisfaction. Mozart’s formulation of the matter – his emphasis on specific, individual moments – resonates with Haydn’s account of what Cherubini found in Haydn’s scores. The wording Griesinger transmits is: ‘so traf er immer auf die Stellen, welche Auszeichnung verdienen’ – which is to say, in effect, those passages ‘here and there’ deserving special praise. What these accounts suggest is a sense that moments aimed specifically at connoisseurs tend to be just that: moments, few in number and of a highly concentrated quality. Although Mozart was writing about piano concertos in late 1782 when describing the anticipated responses of connoisseurs and non-connoisseurs, we should remind ourselves that he was very much occupied at the time with another genre as well: the string quartet, the very genre his father had held up as exemplary for reaching a heterogeneous audience. The date of Mozart’s letter is significant: 28 December 1782. Three days later, on the last day of the year, he completed the String Quartet in G major, k387, the first of the six ‘Haydn’ Quartets. The finale of this work famously juxtaposes fugal and homophonic passages in dazzling succession, alternating quite systematically between what might be thought of as music for Kenner and for Nicht-Kenner – the homophonic passages, the last of which in the exposition is marked semplice, a term that should always raise a red flag in our minds. If the distinction between the ‘too easy’ and the ‘too difficult’ is not so readily discernible in the three concertos k413–15, they are glaringly obvious in the finale of k387; indeed, the very contrast between the two styles becomes an object of discourse itself. This movement is usually read as a commentary on

Listening to listeners

the disparate styles of polyphony and homophony, and rightly so, but if we accept the premise that Mozart was also trying to connect with listeners of widely varying competencies, then we might also think of this movement as an attempt to balance, through a series of rapid and repeated juxtapositions, contrasting modes of writing that will appeal to connoisseurs and amateurs alike. A strictly fugal finale, following the lead of Haydn’s Op. 20 No. 5 (on two subjects), No. 6 (on three subjects) or No. 2 (on four subjects) would perhaps have been more ostentatiously artful yet also decidedly less palatable for listeners. It is scarcely coincidental that Haydn’s Opus 33, a set of quartets written with an eye toward publication – which is to say, with an eye toward a heterogeneous audience – contains not a single fugue of any kind. Considerations for what a heterogeneous audience could and could not endure may thus have played a far more significant role in the shape of the finale of k387 than most analysts would probably care to acknowledge. Yet we are reluctant to let philistines into the composer’s workshop and have them looking over the shoulder of the artist. And we are equally reluctant to let these Nicht-Kenner look over our own shoulders as we analyse this music. Haydn, Mozart and Beethoven all created music that was superfluously imaginative (to use Robert Levin’s happy phrase), far more rich and intricate than it really needed to be. This superabundance of creativity nevertheless created problems of a different kind, and the challenge of writing for a heterogeneous public without compromising artistic integrity remained very much in Mozart’s mind as he continued work on the set of six quartets eventually published by Artaria as Opus 10 and dedicated to Haydn. The languidly chromatic slow movement of k428 is followed by the foot-stomping minuet, the most folk-like of all the minuets in the set. And by far the most difficult of all passages in all the ‘Haydn’ Quartets, the dissonant opening of k465, is followed immediately by the brightest, most lyrical C major theme imaginable. Again, the difficult and the clear are juxtaposed, inscribing both Kenner and Liebhaber into these works, just not necessarily at the same time. A similar pattern can be seen in the finale of the ‘Jupiter’ Symphony, k551, which also juxtaposes learned and free styles. By April 1784, Mozart had completed three of the six ‘Haydn’ Quartets (k387, k421 and k428) and at least some, if not all, of k458 as well. It was at this point that he wrote to his father the following lines: Some quartets by a certain Pleyel have recently appeared; he is a student of Joseph Haydn. If you do not yet know these, then you should try to get hold of them; it will be worth the effort. They are very well written and very pleasing. You will at once recognize in them his master. Fine – and it will be fortunate for music if Pleyel in his time is capable of replacing Haydn for us!25

43

44

m a r k e va n b o n d s

This would seem to be the only time in all their voluminous correspondence that Wolfgang encouraged Leopold to acquaint himself with the published work of a particular composer. And this is no idle recommendation: ‘Try to get hold of them; it will be worth the effort.’ Yet it is difficult to accept this endorsement on face value.26 Pleyel’s quartets represent precisely the kind of music aimed squarely at Liebhaber, with very little in the way of harmonic complexity or counterpoint. They are exactly the kind of ‘short, easy, popular’ quartets Leopold had outlined to his son six years earlier, and the music-consuming public responded to them with gusto. Over the next three decades Pleyel’s Opus 1 would be reissued in an astonishing number of editions by publishers throughout Europe: two in Amsterdam, four in Paris, six in London and one each in Mannheim and Offenbach. And this represents only complete sets in the original scoring: many dozens of arrangements of individual works from Opus 1 appeared, for winds and strings, keyboard and strings, solo keyboard, duos for strings and winds, keyboard and winds, and even as songs.27 Mozart’s recommendation is an oblique continuation of the running dialogue between father and son about the demands of commerce and the demands of art, for if Leopold did in fact take the trouble to secure a copy of these new quartets, he would also have seen the set’s letter of dedication to Count Ladislaus Erd¨ody, which addresses precisely this point.28 Pleyel begins this letter, appropriately enough, by thanking the Count for his generous financial support, which had enabled the young composer to study with Haydn in the mid 1770s and then later to travel extensively in Italy. Pleyel then goes on to describe the origin and nature of these new quartets. As if by way of apology, he explains that he had ‘written these quartets in Italy and therefore according to the predominant taste there; they are not as difficult to perform, nor as profound in their art as my previous [quartets]’.29 We do not know what these earlier quartets might have been, but Opus 1 is relentlessly homophonic, and the rare passages of counterpoint might be charitably described as awkward. Pleyel cleverly turns this textural simplicity to his advantage, however, noting that he had composed them this way ‘deliberately in order to make them more commonplace and pleasing’. No ‘mean’ for Pleyel between the too easy and the too difficult: he went straight for the commonplace, and the public could not get enough of it. Mozart’s ‘Haydn’ Quartets, as noted before, fared less well in the marketplace. Admittedly, the set was not quite the commercial failure it is sometimes made out to be – the plates apparently received enough wear for Artaria to re-engrave the set in its entirety and issue a new edition in 1791, and Sieber published its own edition in Paris sometime in the late 1780s or early

Listening to listeners

1790s – but it scarcely compares with the spectacular success of Pleyel’s Opus 1. And critics, as we have seen, tacitly condemned Mozart’s Opus 10 by acknowledging its high level of artistry. The full context of the anonymous critique that had called these quartets ‘too highly seasoned’ is particularly revealing, for it opens with effusive praise of Pleyel: I now respond to your friendly letter and other musical questions. Our beloved dear Pleyel now finds himself, as you know, in Strasbourg, and he wins much applause there. The Parisians, in particular, pay him a great deal for his compositions; he will make his fortune in those regions, even more so than in Italy. His unforgettable teacher Haydn has been here for several weeks now; just the day before yesterday I heard three new, quite divine symphonies by him, which he has written for Paris. Storace is here, and she is still our favorite female singer, but she is leaving Vienna at the beginning of Lent and will be difficult to replace. A few weeks ago Mozart began a musical journey to Prague, Berlin, and (it is said) even London. I hope that this will turn out to his advantage and delight. He is the most accomplished, best keyboard player I have ever heard; unfortunate, though, that in his artistic and really beautiful writing he climbs too high in order to become a new creator, and in the process, it must be said, the sentiments and the heart gain little by it. His new quartets for two violins, viola, and cello, which he has dedicated to Haydn, are certainly much too heavily seasoned – and what palate can tolerate this for very long? Forgive this imagery from the cookbook.30

We should not put too much weight on the opinions of a single critic, but the framework of this account certainly conforms to the basic parameters for writing quartets that Leopold Mozart laid down to his son in 1778 and helps explain the disparity between the commercial success of Pleyel’s Opus 1 and Mozart’s ‘Haydn’ Quartets. Nor was this particular critic isolated in comparing the works of the two composers in these terms. Another correspondent in the same journal, reporting from Italy a few months later, noted that Pleyel, on his recent visit there, had ‘composed some beautiful keyboard sonatas, which are impatiently requested here. His agreeable melodies in them are not as difficult as Clementi’s and Mozart’s sorceries; he is more true to nature, without, however, violating the rules of composition.’31 Dittersdorf ’s damning of Mozart’s ‘Haydn’ Quartets with strong praise has already been noted. And lest we think Dittersdorf too pedestrian in the matter, we should remind ourselves that Haydn would express similar reservations to Beethoven about the potential reception of the young composer’s as yet unpublished Piano Trio in C minor, Op. 1 No. 3. To his credit, Haydn reportedly later acknowledged that he ‘had not imagined that this trio would be so quickly and easily understood nor so favourably received by the public’.32 In any event, we see once again an anticipation of public competencies as part of the

45

46

m a r k e va n b o n d s

compositional matrix. We will never understand or even recognize every decision that goes into the making of a musical work; but if we can take into account the demands of listening to listeners, we are likely to appreciate all the more the works we analyse. The presumption that a composer take into account an anticipated audience began to fade in the early nineteenth century as the paradigm of rhetoric gave way to a new paradigm of artistic self-expression. In the 1813 revision of his celebrated essay on Beethoven’s Fifth Symphony, E. T. A. Hoffmann was quick to dismiss charges of the work’s incomprehensibility and turn the tables on listeners, placing the onus of blame on them if they did not understand this new and very challenging work: Yet how does the matter stand if it is your feeble observation alone that the deep inner continuity of Beethoven’s every composition eludes? If it is your fault alone that you do not understand the master’s language as the initiated understand it, that the portals of the innermost sanctuary remain closed to you?33

We are in a different world here: art has portals that lead to innermost sanctuaries, and mere Liebhaber need not bother to approach, for only the initiated may enter. Hoffmann’s assault on the listener is, admittedly, an extreme case, but the sentiment behind it is reflected more graciously but no less tellingly in the many journals and books being published at the time with the express purpose of raising the level of musical understanding among amateurs. Audiences were beginning to accommodate themselves to what they heard in the concert hall – or to put it more precisely, audiences were beginning to accept the idea that from time to time they might need to accommodate themselves to certain works they heard. Composers have never stopped listening to listeners, even in those instances in which they seem to be going out of their way to antagonize those very audiences. Still, the notion that the burden of comprehensibility might not lie entirely with the composer laid the foundation for a new and very different approach to composition. The Leipzig Allgemeine musikalische Zeitung was one of the landmark enterprises in this attempt to raise standards of listening, and its founding editor, Friedrich Rochlitz, was quite clear on this point. In an essay of 1813 entitled ‘The Fugue: An Essay Directed Primarily toward Dilettantes and Lay-persons’, Rochlitz argued that composers normally supplied what audiences demanded, and that the general level of composition could therefore rise only if listeners’ demands rose accordingly. Dilettantes and lay-persons, he insisted, would have to come to terms with that hallmark of musical

Listening to listeners

connoisseurship, the fugue. And although it does not deal specifically with the fugue, the passage below points to Mozart as a harbinger of this new attitude, in which the composer abandons his listeners and forgoes monetary gain to preserve a supposed ‘purity’ of his art: The artist and the public stand in a perpetual relationship of mutual effect: what the latter stubbornly does not wish to have, the former will stubbornly not supply. The artist would otherwise have to sacrifice himself, like a hero, to his idea of art, as in the case of Mozart, who, upon being told by one of his publishers (Hoffmeister): ‘Write in a more popular manner, otherwise I cannot publish anything more by you and pay you!’ gave the answer: ‘Well, then I won’t earn anything more, I’ll go hungry, and the devil take me!’ It need not be said to how few this would or might apply.34

This is a quintessentially nineteenth-century perspective. The words Rochlitz puts into Mozart’s mouth about refusing to write in a popular manner are transparently fictitious, yet the sentiment was believable for readers in 1813 in a way it would not have been a generation before. Critics like Rochlitz held a vested interest in the problems composers like Mozart and Beethoven posed for general listeners: the greater the perception that this new music was complicated, the greater the perceived need for the kind of interpretive commentary critics like Rochlitz and Hoffmann could provide. What the public perceived as a widening gap between audience and composer was filled with an essentially new type of literature that emerged in response at the end of the eighteenth century: that of the guide for the educated listener eager to come to terms with new and less-than-readilyaccessible repertory.35 The fault-lines in the division between high and low had been in place long before. What changed in the decades after 1800 was the growing acceptability of the idea that composers might not be writing for a universal audience. They could compose for such an audience if they chose to, of course, and they often did just that. But when they did so, they now ran the risk of being accused of ‘pandering’ to the public taste, as opposed to creating great art. It is a division that continues to shape the ways in which we listen to – and analyse – music today. Notes 1 Aristotle, The ‘Art’ of Rhetoric, trans. John Henry Freese (Cambridge, MA: Harvard University Press, 1947), book 1 ch. 3; see also Aristotle’s treatment of various types of audiences in book 2. 2 George Campbell, The Philosophy of Rhetoric, ed. Lloyd F. Bitzer (Carbondale: Southern Illinois University Press, 1963), 102–3 (ch. 10 section 2).

47

48

m a r k e va n b o n d s

3 Leopold Mozart, letter to his son, 13 August 1778, in Mozart: Briefe und Aufzeichnungen, ed. Wilhelm A. Bauer and Otto Erich Deutsch, 7 vols. (Kassel: B¨arenreiter, 1962–75), vol. 2, 444. Emphasis in the original. The translation here is based on but substantially modified from that given in The Letters of Mozart and his Family, ed. and trans. Emily Anderson (New York: Norton, 1985), 599. 4 Mozart, letter to his father, 28 December 1782, in Mozart: Briefe und Aufzeichnungen, vol. 3, 245–6. 5 Elaine Sisman, ‘Observations on the First Phase of Mozart’s “Haydn” Quartets’, in Words About Mozart: Essays in Honour of Stanley Sadie, ed. Dorothea Link and Judith Nagley (Woodbridge: Boydell, 2005), 33–58; see especially 35–43. 6 Mozart, letter to his father, 3 July 1778, in Mozart: Briefe und Aufzeichnungen, vol. 2, 388–9. 7 Haydn, letter to Christoph Gottlob Breitkopf, 12 June 1799, in his Gesammelte Briefe und Aufzeichnungen, ed. H. C. Robbins Landon and D´enes Bartha (Kassel: B¨arenreiter, 1965), 320. The translation here is slightly modified from that given in Haydn, The Collected Correspondence and London Notebooks, ed. H. C. Robbins Landon (London: Barrie and Rockliff, 1959), 154. 8 Georg August Griesinger, Biographische Notizen u¨ ber Joseph Haydn (Leipzig: Breitkopf und H¨artel, 1810), ed. Karl-Heinz K¨ohler (Leipzig: Reclam, 1975), 79. 9 Ernst Ludwig Gerber, Historisch-biographisches Lexikon der Tonk¨unstler, vol. 1 (Leipzig: J. G. I. Breitkopf, 1790), col. 610. 10 Johann Karl Friedrich Triest, ‘Bemerkungen u¨ ber die Ausbildung der Tonkunst in Deutschland im achtzehnten Jahrhundert’, Allgemeine musikalische Zeitung 3 (March 1801), col. 407 (my translation). Triest’s essay has been translated in its entirety by Susan Gillespie as ‘Remarks on the Development of the Art of Music in Germany in the Eighteenth Century’, in Haydn and His World, ed. Elaine Sisman (Princeton: Princeton University Press, 1997), 321–94. 11 Giuseppe Carpani, Le Haydine (Milan: Candido Buccinelli, 1812), 11–12: ‘O cercarono essi di semplificare la melod`ıa collo scemare gli accordi e scarseggiare di transizioni; e fecero come il Pleyel, de’ lavori meno dignitosi e robusti; o come il Mozart ed il Beethoven, accumularono i numeri e le idee, e la quantit`a e la stranezza ricercarono delle modulazioni, e non produssero allora [che] delle erudite intricatissime confusioni, piene di ricercatezza e di studio, ma prive d’effetto; ove all’incontro riscossero lodi somme e ben giuste, quando seguirono la luce Haydiniana. Cos`e e` . Egli solo, il nostro Haydn, seppe senza eccezione essere sempre perfettamente chiaro ed erudito, energico e naturale, armonioso e non confuso, dotto e non arido, melodioso e non triviale, magistralmente tessuto e non mai intralciato, spontaneo e regolare, e cio` in mezzo alla bizzarria delle idee non mai piu` sentite, alla novit`a de’ mezzi, all’ardire de’ ripieghi, alla ferocit`a dell invenzioni.’ 12 James Webster, ‘Haydn’s Symphonies Between Sturm und Drang and “Classical Style”: Art and Entertainment’, in Haydn Studies, ed. W. Dean Sutcliffe (Cambridge: Cambridge University Press, 1998), 231.

Listening to listeners

13 See Joseph Haydn, Werke, series 1 vol. 6, Sinfonien, 1767–1772: Kritischer Bericht, ed. Carl-Gabriel Stellan M¨orner (Munich: Henle, 1969), 39. On Haydn’s approach to audiences in general, see Elaine Sisman, ‘Haydn’s Career and the Idea of the Multiple Audience’, in The Cambridge Companion to Haydn, ed. Caryl Clark (Cambridge: Cambridge University Press, 2005), 3–16. 14 Carl Philipp Emanuel Bach, letter to Alexander Reinagle, c. 1785–1786, in Carl Philipp Emanuel Bach, Briefe und Dokumente: Kritische Gesamtausgabe, ed. Ernst Suchalla, 2 vols. (G¨ottingen: ‘Vandenhoeck und Ruprecht, 1994), vol. 2, 1130: ‘Die Liebhabere¨y zu den Rondos ist hier [in Hamburg] so groß, wie in London, und ich habe sie deswegen eingemischt, um meinen Verkauf zu bef¨ordern. Ich weiß aus der Erfahrung, daß sehr viele meine Sam[m]lungen blos wegen der Rondos kaufen.’ 15 Charles Burney, An Eighteenth-Century Musical Tour in Central Europe and the Netherlands, ed. Percy A. Scholes (London: Oxford University Press, 1959), 220. Burney would go on to qualify his misgivings with the comment that ‘the great musician [nevertheless] appears in every movement, and these productions will probably be the better received, for resembling the music of this world more than his former pieces, which seem made for another region, or at least another century, when what is now thought difficult and far fetched will, perhaps, be familiar and natural’. 16 See C. P. E. Bach’s autobiographical account interpolated into the Germanlanguage edition of Burney’s diary published in 1772, Tagebuch einer musikalischen Reise durch Frankreich und Italien, trans. Christoph Daniel Ebeling, ed. Eberhardt Klemm (Wilhelmshaven: Heinrichshofen, 1980), 455: ‘Weil ich meine meisten Arbeiten f¨ur gewisse Personen und f¨urs Publikum habe machen m¨ussen so bin ich dadurch allezeit mehr gebunden gewesen als bei den wenigen St¨ucken, welche ich bloß f¨ur mich verfertigt habe . . . Indessen kann es sein, daß dergleichen nicht eben angenehme Umst¨ande mein Genie zu gewissen Erfindungen aufgefodert haben, worauf ich vielleicht außerdem nicht w¨urde gefallen sein.’ 17 Charles Burney, Memoirs of the Life and Writings of the Abate Metastasio, 3 vols., (London: G. G. and J. Robinson, 1796), vol. 3, 309. 18 On the comparatively feeble state of music publishing in German-speaking lands in the first half of the eighteenth century, see Hans Lenneberg, On the Publishing and Dissemination of Music, 1500–1850 (Hillsdale, NY: Pendragon, 2003), 8 and chs. 3 and 4; Axel Beer, ‘Composers and Publishers: Germany, 1700–1830’, in Music Publishing in Europe, 1600–1900: Concept and Issues, Bibliography, ed. Rudolf Rasch (Berlin: Berliner Wissenschafts-Verlag, 2005), 160–81. On the rise of music publishing in Vienna in the last three decades of the eighteenth century, with special emphasis on the firm of Artaria, see Rupert Ridgewell, ‘Economic Aspects: The Artaria Case’, in Music Publishing in Europe, 90–113. On the relationship of composers, publishers, and the public in a slightly later period, see Axel Beer, Musik zwischen Komponist, Verlag und Publikum: Die Rahmenbedingungen des Musikschaffens in Deutschland im ersten Drittel des 19. Jahrhunderts (Tutzing: Schneider, 2000).

49

50

m a r k e va n b o n d s

19 See the chronology of published first editions in Gertraut Haberkamp, Die Erstdrucke der Werke von Wolfgang Amadeus Mozart, 2 vols. (Tutzing: Schneider, 1986), vol. 1, 419–25. 20 Mozart, letter of 12 July 1789 to Michael Puchberg, in Mozart: Briefe und Aufzeichnungen, vol. 4, 93. 21 Important recent contributions include Julia Moore, ‘Mozart in the MarketPlace’, Journal of the Royal Musical Association 114 (1989): 18–42; William J. Baumol and Hilda Baumol, ‘On the Economics of Musical Composition in Mozart’s Vienna’, in On Mozart, ed. James M. Morris (Washington, DC: Woodrow Wilson Center Press, 1994), 72–101; Neal Zaslaw, ‘Mozart as a Working Stiff’, in On Mozart, ed. James M. Morris (Washington, DC: Woodrow Wilson Center Press, 1994), 102–12; Tia de Nora, Beethoven and the Construction of Genius: Musical Politics in Vienna, 1792–1803 (Berkeley: University of California Press, 1995); Peggy Daub, ‘The Publications Process and Audience for C. P. E. Bach’s “Sonaten f¨ur Kenner und Liebhaber”’, Bach Perspectives 2 (1996), 65–83; Rupert Magnus Ridgewell, ‘Mozart and the Artaria Publishing House: Studies in the Inventory Ledgers, 1784–1793’ (Ph.D. dissertation, Royal Holloway, University of London, 1999). 22 In the case of Mozart, Daniel K. L. Chua (‘Myth: Mozart, Money, Music’, in Mozart Studies, ed. Simon P. Keefe (Cambridge: Cambridge University Press, 2006), 193–213) insightfully identifies the motivations behind longstanding resistance to constructing the composer’s music as a commercial commodity. 23 Carl Dittersdorf, letter of 1788 to Artaria, quoted in Eva Badura-Skoda, ‘Dittersdorf u¨ ber Haydns und Mozarts Quartette’, in Collectanea Mozartiana, ed. Cordula Roleff (Tutzing: Schneider, 1988), 47: ‘Ich offerire Ihnen also das Originalmanuscript oder besser zu sagen: meine eigenh¨andige Partitur davon um den nemlichen Preiß, wie sie die Mozartischen bezahlt haben und bitte mir extra noch die ersten 10 Abdr¨ucke oder Exemplar aus, und . . . bin sicher, daß Sie sich bey meinen wegen den lucrum cessans der Mozartischen (welche zwar bey mir so wie bey noch gr¨oßern Theoretiquern alle Hochachtung verdienen, aber wegen der allzugroßen darinen best¨andig herrschenden Kunst nicht Jedermanns Kauf seyn) erhollen werden.’ 24 See W. Dean Sutcliffe, ‘Haydn, Mozart and their Contemporaries’, in The Cambridge Companion to the String Quartet, ed. Robin Stowell (Cambridge: Cambridge University Press, 2003), 185–209. 25 Mozart, letter of 24 April 1784 to his father, in Mozart: Briefe und Aufzeichnungen, vol. 3, 311. 26 For a fuller examination of this passage and the inter-relationship of the quartets of Haydn, Pleyel and Mozart, see Mark Evan Bonds, ‘Replacing Haydn: Mozart’s “Pleyel” Quartets’, Music and Letters 88 (2007): 201–25. 27 See Rita Benton, Ignace Pleyel: A Thematic Catalogue of his Compositions (New York: Pendragon, 1977), 99–105.

Listening to listeners

28 No copy of these or any other quartets by Pleyel is to be found in the library of Leopold Mozart, though other works known to have been in his collection have since been dispersed. See Cliff Eisen, ‘The Mozarts’ Salzburg Music Library’, in Mozart Studies 2, ed. Cliff Eisen (Oxford: Clarendon, 1997), 85–138. 29 ‘Scrissi questi quartetti in Italia, e quindi secondo il gusto dominante di col`a; non sono n`e si difficili nell’esecuzione, n`e si profondi nell’arte, come i miei precedenti, ma composti cosi a bella posta, accio si rendano piu` comuni, e piacevoli.’ A facsimile of the letter of dedication is given in Julius Zsako, ‘The String Quartets of Ignace J. Pleyel’ (Ph.D. dissertation, Columbia University, 1975), 539. I am grateful to Massimo Ossi (Indiana University) for help on the finer points of this and other translations from the Italian. 30 Carl Friedrich Cramer, Magazin der Musik 2 (23 April 1787), 1273–4: ‘Ich komme nun zur Beantwortung Ihres freundschaftlichen Briefes und u¨ brigen musikalischen Fragen. Unser lieber theurer Pleyel befindet sich, wie Sie wissen, in Straßburg, und hat dort vielen Beifall; besonders bezahlen ihm die Pariser seine Compositionen außerordentlich theuer; er wird in jenen Gegenden sein Gl¨uck machen, noch mehr als in Italien. Sein unver[g]aeßlicher Lehrer Haydn ist seit einigen Wochen hier; vorgestern habe ich eben von ihm 3 neue, ganz g¨ottliche Symphonien geh¨ort, die er f¨ur Paris geschrieben hat. – Storace ist hier, und noch immer unsere Lieblingss¨angerin, wird aber mit Anfang der Fasten Wien verlassen, und schwer zu ersetzen seyn. Mozart hat vor einigen Wochen eine musikalische Reise nach Prag, Berlin, und man sagt, sogar nach London angetreten. Ich w¨unsche, daß sie [die Reise] zu seinem Vortheil und Vergn¨ugen ausschlagen m¨oge. Er ist der fertigste, beste Clavierspieler, den ich je geh¨ort habe; nur Schade, daß er sich in seinem k¨unstlichen und wirklich sch¨onen Satz, um ein neuer Sch¨opfer zu werden, zu hoch versteiget, wobey freilich Empfindung und Herz wenig gewinnen, seine neuen Quartetten f¨ur 2 Violin, Viole und Baß, die er Haydn dedicirt hat, sind doch wohl zu stark gew¨urzt – und welcher Gaum kann das lange aushalten. Verzeihen Sie dieses Gleichniß aus dem Kochbuche.’ 31 Cramer, Magazin der Musik 2 (26 July 1787), 1378–9: ‘Er [Pleyel] hat verschiedene sch¨one Clavier-Sonaten gesetzt, darnach man hier mit Ungeduld verlangt. Sein gef¨alliger Gesang darauf ist nicht so schwer wie Clementis und Mozarts Hexereyen; er bleibt dabey mehr der Natur getreu, ohne desfals [sic] in die Regelen des reinen Satzes hintenanzusetzen.’ The translation here differs from that given in Cliff Eisen, New Mozart Documents (London: Macmillan, 1991), 48. 32 Franz Gerhard Wegeler and Ferdinand Ries, Biographische Notizen u¨ ber Ludwig van Beethoven (Koblenz: K. B¨adeker, 1838), 85. Translation from Remembering Beethoven, trans. Frederick Noonan (London: Andr´e Deutsch, 1988), 74. 33 E. T. A. Hoffmann, ‘Beethovens Instrumentalmusik’, Kreisleriana (1813), reprinted in his S¨amtliche Werke, ed. Hartmut Steinecke and Wulf Segebrecht, 6 vols. (Frankfurt-am-Main: Deutscher Klassiker Verlag, 1993), vol. 2 part 1, 55, 59–60; emphasis in the original. For further commentary on Hoffmann’s

51

52

m a r k e va n b o n d s

views about the changing relationship between composer and listener in the early nineteenth century, see Mark Evan Bonds, Music as Thought: Listening to the Symphony in the Age of Beethoven (Princeton: Princeton University Press, 2006), ch. 3. 34 Friedrich Rochlitz, ‘Die Fuge: Zun¨achst an Dilettanten und Layen’, Allgemeine musikalische Zeitung 15 (12 May 1813), 313: ‘K¨unstler und Publicum stehen stets in Wechselwirkung: was dies beharrlich nicht will, wird jener beharrlich nicht liefern – er m¨usste sich denn seiner Idee von Kunst als ein Held aufopfern wollen, wie Mozart, der einem seiner Verlege (Hoffmeister’n) auf dessen Eindringen: Schreib populairer, sonst kann ich nichts mehr von Dir drucken und bezahlen! zur Antwort gab: “Nun, so verdien’ ich nichts mehr, und hungere, und scher’ mich doch den Teufel drum!” Wie Weniger Sache dies aber sey, und seyn k¨onne, braucht nicht erst gesagt zu werden.’ Thus presumably was born the myth of Hoffmeister and the broken contract for a set of three piano quartets. There is no documentary evidence to substantiate Rochlitz’s account. In fact, Hoffmeister would go on to publish no fewer than eleven further works by the composer, including what Maynard Solomon aptly characterizes as ‘several of Mozart’s more uncompromising compositions’, such as the String Quartet in D major, k499, the Fugue in C minor for two pianos, k427, and the Sonata in C major for piano duet, k521. See Maynard Solomon, ‘The Rochlitz Anecdotes’, in Mozart Studies, ed. Cliff Eisen (Oxford: Oxford University Press, 1991), 1–59. On the trope of Mozart’s unconcern with monetary reward, see Chua, ‘Myth: Mozart, Money, Music’. 35 Leon Botstein, ‘Listening Through Reading’, 19th-Century Music 16 (1992), 129–45.

3

‘Mannichfaltige Abweichungen von der gew¨ohnlichen Sonaten-Form’: Beethoven’s ‘Piano Solo’ Op. 31 No. 1 and the challenge of communication c l au d i a m au r e r z e n c k

The genesis of Opus 31 It was noticed early on that Beethoven, contrary to his usual practice, did not dedicate the three piano sonatas Opus 31 to a patron or backer; yet, the notion that they might have been composed without a commission is still doubted today.1 The circumstances that brought forth those sonatas thus remain unclear, and only the time-frame in which they were composed can be determined with some accuracy. Of the incomplete correspondence surrounding Beethoven, only one letter has survived that deals specifically with the piano sonatas Op. 31. The letter, dated 18 July 1802, was sent by N¨ageli, the Zurich publisher, to Horner, his business partner in Paris. In it he reports that Beethoven has demanded payment of 100 ducats – not 100 guldens – for three sonatas. Beethoven’s brother had advised N¨ageli to ask for a reduction in fee, but N¨ageli himself would have preferred to request a fourth piano sonata, so that he could publish two complete Beethoven volumes in his series, R´epertoire des clavecinistes. In the letter, N¨ageli adds that he expects the three sonatas to arrive by post on 17 August. From the foregoing information, the following deductions can be made: First, N¨ageli must have written to Beethoven in early summer 1802, requesting that the composer contribute several sonatas to the planned series.2 Second, Beethoven accepted by letter and demanded a fee, the currency of which N¨ageli initially misunderstood, until Beethoven’s brother, who often handled his business correspondence, corrected him. (Karl van Beethoven, however, had informed the Leipzig publisher Breitkopf and H¨artel, on 22 April 1802, that his brother would demand fifty ducats for a grand sonata and 130 ducats for three sonatas, with or without accompanying instrument.)3 Third, from the letter to Horner, it cannot be deduced whether N¨ageli also suggested that Beethoven contribute three sonatas, or whether the composer first mentioned that number. It must be presumed, however, that N¨ageli

54

c l au d i a m au r e r z e n c k

informed Beethoven, either prior to writing to Horner or shortly thereafter, that he wanted to publish just two (not all three) sonatas together in one volume. Presumably, he expected works of uncommon scope from Beethoven.4 Fourth, even if Beethoven suggested the figure himself, the sonatas had not been composed at the time of the earliest correspondence relating to them. It has been known since at least 1985 that Karl van Beethoven, in the aforementioned letter of 22 April 1802 to Breitkopf, did not offer three completed (or nearly completed) piano sonatas but rather violin sonatas (Op. 29).5 (This chronology is important, as the thesis of this chapter to a certain extent relies on it.) And fifth, if one bears in mind the fact that N¨ageli was expecting the completed sonatas on 17 August, that is, only one month after his letter (Beethoven’s brother could have informed him of the date of completion), and that, according to Ferdinand Ries, Beethoven typically put off working on pieces due by a certain date,6 then one must agree with the editors of the Beethoven Sketchbooks, who date the last entries in the Kessler Sketchbook – to which those of the first two movements of Op. 31 No. 1 belong – from July or early August 1802.7 It is not known whether Beethoven actually had the sonatas sent to Zurich on time. That was possibly arranged by Ries – if his story is correct – but only after some vehement arguments between Beethoven and his brother; for Karl Beethoven would rather have sent the sonatas to the Leipzig publisher with whom he was negotiating and from whom he had demanded a larger fee than Beethoven had done. Beethoven, however, wanted to keep the commitment he had made to N¨ageli.8 Yet, it is clear that he only sent numbers one and two, which were to be published together in a single volume. It is unknown whether he had assured N¨ageli of an additional, fourth sonata and had asked for more time to compose it. In any case, Op. 31 No. 3 was first published in November 1804 in Zurich, together with Op. 13 (which had first appeared in print in Vienna in 1799) – more than one and a half years after the first two sonatas. (The first sketches of the next piano sonata to be composed, Op. 53, date from December 1803.) In mid August 1802, Beethoven would certainly have needed more time to write a fourth sonata for N¨ageli’s series, as well as for the long-planned third sonata: the first sketches for Op. 31 No. 3 appear in the Wielhorsky Sketchbook (pages 1–11), in which the earliest entries date from autumn 1802. Hence, the first two sonatas of Opus 31 were composed no earlier than summer of 1802, the third in autumn (and possibly not until the beginning of 1803). All three were evidently written at N¨ageli’s request that Beethoven contribute to the R´epertoire series. The publisher’s concept for the latter may account for the distinctiveness of the sonatas in question. N¨ageli announced

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

the series in the Leipzig Allgemeine musikalische Zeitung in August 1803, in words similar to those he had used in coaxing Beethoven to accept the commission. He boasted not only of the brilliance, compositional integrity and virtuosity of the works, but also about their unusual features: First of all, the series is about piano solos in the grand style, of great scope, with many departures from the usual sonata form [Abweichungen von der gew¨ohnlichen Sonaten-Form]. These works should be distinguished by their detail, abundance of ideas, and fullness of texture. Contrapuntal phrases must be interwoven with piano displays of florid extravagance. Whoever is not an accomplished piano virtuoso will hardly be able to achieve anything noteworthy here.9

My thesis, therefore, is that Beethoven was attracted by N¨ageli’s request that he contribute such sophisticated sonatas to the series – substantial, brilliant, virtuosic, and formally unusual – and that in them he realized, in various ways, the ideal of composing ‘with many departures from the usual sonata form’.

The humorous in music around 1800 For about fifteen years, in the late eighteenth and early nineteenth centuries, the ‘humorous’ in music was a frequent subject of discussion. In 1792 an article entitled ‘Comisch’ appeared in the fourth edition of Sulzer’s Allgemeine Theorie der sch¨onen K¨unste; also published in the same year was Friedrich ¨ August Weber’s letter, ‘Uber komische Charakteristik und Karrikatur in praktischen Musikwerken’ (reprinted in 1800). Heinrich Christoph Koch mentioned the subject in his Musikalisches Lexikon of 1802. Friedrich Rochlitz submitted an anonymous article that appeared in several instalments in the Leipzig Allgemeine musikalische Zeitung, in the last of which he also treated the idea of the ‘Niedliche’ (niedrig Komische, low humour) in music.10 In 1807, Christian Friedrich Michaelis published an essay entitled ¨ ‘Uber das Humoristische oder Launige in der musikalischen Komposition’,11 in which he went so far as to claim the following: However, our newest music is for the most part humorous, especially since Joseph Haydn, as the greatest master in this genre, . . . set the tone . . . Haydn was the first to do this in overall effect, and awakened a large number of famous composers of the most recent times to write in this manner.12

All of the authors cited above unanimously emphasized that any ‘departure from the usual rules’,13 that is, from the ‘conventional’, should serve

55

56

c l au d i a m au r e r z e n c k

a humorous purpose; that such departures could be a ‘genuine or only apparent irregularity between the presented object and the means of the art’.14 In short, purpose and means should ‘oppose each other abruptly’.15 As Weber maintained in 1792, the composer should accomplish that goal primarily through a ‘particular application of the rules of harmony and melody, whereby a feeling of the comical should be awakened in the listener whose hearing is attuned to it’.16 Fifteen years later, Michaelis expanded the ways in which the humorous might reveal itself, by also seeing the ‘Launige’ (the witty, or the jocular) in the character of a particular rhythm or in the ‘unexpected entrance of certain voices or instruments’.17 Weber and Michaelis occupied themselves with the humorous in music in ways that to us are informative: Weber classified various types of humorous composition, while Michaelis provided detailed descriptions of musical techniques by which a composer could achieve the humorous.

Friedrich August Weber 1792/1800: species of the humorous Examples of the ‘humorous’ genre were first and foremost to be found in musical representation and imitation, and then in the ‘witty application of the great or small gift of humour with which the composer is endowed, and thereafter in musical parody and intentionally undertaken musical bungling’.18 Whether it be a song or instrumental piece, a composition can be comical without descending into caricature. This is the case with musical Mahlerey, which may be merely ‘droll’ (possirlich), as in the fine arts; that is to say, when a humorous composer who does not want or has no need to descend into caricature grants himself only moderate departures from the usual rules, whose observance at the same time creates an artistic product of the beautiful and serious genre, and connects phrases into a whole, which, through this process, receives a layer of paradox and, through an occasional combination of idea, ludicrousness of a lesser degree.19

Musical imitation (Mimik) goes one step further, by connecting tones, chords, phrases, and melodies in such a way that they take on the character of ‘modest comedy, or of extravagant caricature’. Here, Weber points only to composers of op´eras-comiques, opere buffe, and Singspiele, and ultimately exemplifies caricature with an aria by Philidor.20 By contrast, the musically witty consists of the ‘discovery of unexpected similarities between two musical thoughts and of their turning out to be ingenious and purposeful through the element of surprise’.21 According to

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

Weber, composers ‘who possess the gift of musical wit abundantly’ include, among others, Dittersdorf (in his operas), Mozart, and ‘Papa Haydn’.22 Weber recognizes a particular variety of wit in musical humour (musikalische Laune) that is carried to the extremes of ‘singularity and paradox’ (Singularit¨at und Paradoxie). This summit of wit is not to be confused with musical nonsense, for humour is light years away from the latter: nonsense is that which is produced ‘by the newest and recent corrupters of good musical taste’.23 On the other hand, parody – the ‘reversed form of usage’ (umgekehrte Art des Verbrauches) – can only occur in music in a humorous, not in a serious way; for if humorous thought is employed seriously, it becomes distasteful. On the other hand, a serious thought can be put into a humorous context. The object of parody can even be another composer. It also occurs, however, that a composer might not be aware of precisely whom and what he is parodying; i.e., parody does not necessarily require a direct object. As an example of good parody, Weber points to the aria ‘Der Gott, der Herzen bindet’ in Hiller’s Liebe auf dem Lande, ‘in which the old-fashioned opera arias with long runs and droning passages are masterfully drawn into the ridiculous by the rising and falling sevenths’.24 ‘Artfully imitated musical bungling’, as the final species of the musically humorous, cannot be defined more closely; it evades theory, for by nature it brings forth ‘a certain disorderliness’.25 In order to intimate fake bungling, and/or require it of the performers, the composer must be well-acquainted with the ways in which composers typically botch the writing of vocal and instrumental parts.

Carl Friedrich Michaelis 1807: means of the humorous Weber’s classification already makes reference to certain compositional methods, three of which he relates to the opera stage: representation (Malerei), as the paradoxical connection between periods and phrases; imitation (Mimik), understood as character contrasts; and parody, the reversal of the serious into the humorous. In Weber’s discussion, other methods are referred to in a more general way; those include the witty invention of similarity in thoughts that seem dissimilar, and the random nature of dilettantism used for compositional purposes. Michaelis, on the other hand (and similar to Rochlitz, who had recently published in the same journal), does not classify the humorous types of music, but instead explores in more detail the craft of composition:

57

58

c l au d i a m au r e r z e n c k

There appear melodies, figures, runs, heavy chords; one does not know where they come from; the music begins in such a peculiar way, perhaps extremely simply, with some notes which appear insignificant, that one would not have suspected the interesting and amusing work which develops. The composer gives the conventional figures a new shape through their reversal, or arranges the notes and accents against the beat, entangles the voices so wonderfully in one another, slips so boldly into other keys, and just as unexpectedly returns quickly to the previous key and principal melody, closing in such an individual way that all this cannot be explained by the usual practice of the musical system, by the typical rules of composition, or by a natural, regular process, but appears just as surprising as the ideas of a witty or humorous narrator . . . Humorous music is sometimes comical and naive, sometimes serious and sublime. The departure from the conventional – the curious combination of the strange and remote, the reversal of figures, the unusual form of the beginning, the transitions, or the endings – initially has an appearance of absurdity; because the latter, however contrary to expectation, is connected with brilliant music; because in the course of the music and in view of the whole, it ceases to be absurd, the music gives the impression of humour and tempts one into laughter.26

The composer’s intention, therefore, is to produce a surprising range of musical expression, from the sublime to the naive, from the comical to the serious. According to Michaelis, occasionally complemented by Rochlitz, that intention is achieved as follows: first, by unexpected turns at the level of thematic construction: (melodic) inversion of the ‘usual figures’, (rhythmic) division of the notes and dynamic accents ‘against the beat’ or simply in rapid alternation; by surprising ‘complexity’ of a part (e.g., through unexpected counterpoint or unexpected entrance of instruments); second, by sudden interjections, incongruous connections and continuations at the level of syntax, or, put historically, of phrase construction (Satzbau); for example, through the unexpected insertion of melodies, figures, passages, or heavy chords; through complex development of simple beginnings, and vice versa, through simple development of interesting beginnings, as can be observed in many scherzi; and through surprising returns of the theme or ‘leaps into popular, light, frivolous ideas’;27 and third, by unexpected harmonic turns, such as the sudden entries of foreign keys (or simply of dissonances) and their equally sudden disappearance; by ‘audacious’, that is to say, distant and rapid modulations and equally rapid returns to the opening tonality, as well as idiosyncratic closures, which certainly act in an idiosyncratic way not only on the harmonic level, but also on the level of syntax.28 Those unexpected features are fundamentally all the harder to recognize and comprehend, the less one is acquainted with the ‘particular conventions’, the ‘usual practice of the musical system’, and its ‘natural, regular

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

process’.29 Acquaintance with those norms also helps with a second fundamental problem: even when we recognize the unexpected as such, we have to be able to tell when it is meant seriously, i.e., when it is an innovative violation of ‘convention’ (Herkommen). A further complication is the fact that our compositional and analytical ‘wit’, 200 years after the composition of Op. 31 No. 1, is no longer the same as the ‘wit’ of that time. The problem can be further exacerbated when the work in question is not a musico-dramatic composition, i.e., when a humorous text is lacking. As early as 1792, Weber had already complained: It’s a pity that musical wit, applied in mere instrumental music, practically always requires a commentary which only the composer can deliver, and therefore its purpose to astonish and amuse is often not completely achieved!30

Unfortunately, Beethoven left no commentary on Opus 31 and can no longer provide one. Hence, in the following analysis of the first movement of Op. 31 No. 1, I make recourse to the music theory of the latter eighteenth century, in supporting my thesis that Beethoven conceived the ‘mannichfaltige Abweichungen von der gew¨ohnlichen Sonatenform’ of that opus (exemplified here by the first movement), from start to finish, not in a serious way, but with comic intent. He applied all the means of creating humour, as described by Michaelis and occasionally Rochlitz (see Table 3.1). Moreover, Beethoven combined all those means with techniques for achieving simplicity, so as to present the composer persona of an amateur and bungler. That is not to say, however, that the composer himself drew no serious consequences from this humorous work. That he did so can be seen in his subsequent compositions, above all the piano sonatas. For that reason, the latter are shown, in the course of my argument, to constitute a ‘serious’ boundary to this ‘humorous’ sonata, Op. 31 No. 1. For example, consider the question-and-answer play toward the end of the finale of the G major sonata (bars 224/5–242/3). There bars 1–2 and 5–6 of the quite ‘normal’ allegretto rondo-theme spar with bars 3–4 and 7–8 of the Adagio theme, to which must be added the delay of the Grundkadenz through repetition and prolongation of the dominant. That kind of alternation has consequences for the first movement of the subsequent D minor sonata, Op. 31 No. 2, in the opening contrasts of Largo (which by repetition is extended into an instrumental recitative) and Allegro. On the other hand, there are evident links with another genre: in the same last movement of Op. 31 No. 1, the music played in the left hand, bars 239–41, is an expanded version of bars 3–4 of the opening theme of the Sixth Symphony, which Beethoven began to compose just a short time later.

59

60

c l au d i a m au r e r z e n c k

¨ Table 3.1 Humorous musical means after Michaelis, ‘Uber das ¨ Humoristische oder Launige’ (M), and Rochlitz, ‘Uber den zweckm¨assigen Gebrauch’ (R). A 1 2 3

Themes: unexpected shape ‘ungewohnte Art des Anfanges’ (M; see also b1) ‘[launige] Beschaffenheit . . . des Rhythmus’ (M) ‘[Ueberh¨upfen in] popul¨are, leichte, t¨andelnde Ideen’ (R; see also b6) 4 ‘hergebrachte(n) Figuren . . . Umkehrung’ (M) 5 ‘unerwartetes Eintreten gewisser Stimmen’ (M) 6 ‘schneller, aber scherzhaft angebrachter Wechsel des gem¨assigten Forte und Piano’ (R) B 1 2 3 4 5 6 7

the unusual form of the beginning jocular formation of rhythm leaps into popular, light, frivolous ideas reversal of conventional figures unexpected entrance of certain voices quick but playfully applied alternation of forte and piano

Syntax: sudden interruptions, incongruous transitions, continuations, developments ‘ungewohnte Art des Anfanges’ (M; see also a1) the unusual form of the beginning ‘unerwartete Wendung’ (M) unexpected turn ‘¨ofteres und u¨ berraschendes Wiederkehren des frequent and surprising returns of the theme Thema’ (R) ‘ungewohnte . . . Ueberg¨ange’ (M) unusual transitions ‘sonderbar zusammengesetzte Figuren’ (M) curiously compounded figures ‘vert¨andelnde Behandlung eines ernsthaften frivolous treatment of a serious theme Thema’ (R; see also a3) ‘schliesst auf eine . . . eigenth¨umliche Art’ (M; see closes in an individual way also c5)

C Harmony: unexpected turns 1 gegen den ‘nat¨urlichen Gang(e) der Harmonie’ (M) 2 ‘pl¨otzliches . . . Eintreten fremder Tonarten’ (R) 3 ‘k¨uhne Ausweichungen in andere Tonarten’ (M) 4 ‘unerwartet schnell wieder [R¨uckkehr] in die vorige Tonart und Hauptmelodie’ (M) 5 ‘schliesst auf eine . . . eigenth¨umliche Art’ (M; see also b7)

contrary to the natural course of harmony sudden entrance of foreign keys bold [= at wrong place] modulations into other keys unexpected quick return to the previous key and principal melody closes in an individual way

Op. 31 No. 1: analysis of the first movement The first movement seems to begin with an upbeat in the right hand, but that note is held, and the following downbeat is instead articulated by a chord in the left hand, a truly ‘unusual form of the beginning’ (a1, b1).31 This ‘asynchronous’ relationship between the hands is continued in the following bars, without yet revealing its significance. The held note g2 descends rapidly

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

Example 3.1 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 1–11

to g1 , then the music moves forward unexpectedly with arpeggiations of the tonic triad – from g1 to b1 and d2 up to g2 – before reaching a cadence. The continuation of the rapid opening run by the repeated chords and the ‘individual way’ (b7) of closing the phrase present an ‘unexpected turn’ (b2): this latter phrase is so brief in comparison with the preceding passage that it invites comparison with the typical Beethovenian scherzo more than with a piano sonata; furthermore, it does not cadence in the tonic but in the dominant key. Is it not a little early for a Quintkadenz? How many bars has Beethoven composed up to this point (bar 11)? Put differently, how many bars does the listener actually hear? Contrary to appearances, when listening it quickly becomes clear that this is in fact a simple, symmetrical construct (see Example 3.1): an expanded phrase (erweiteter Satz) of six bars in alla breve time, which Beethoven actually writes as 2/4,32 every two-bar group resulting in one ‘double measure’. It is far from the ‘natural harmonic progression’ (c1) to begin a substantial sonata allegro with a closing phrase (Schlußsatz) moving directly to the dominant key. Rather, it is an instance of wit, since the earliest moment for a Quintkadenz in a substantial sonata allegro is with good reason immediately before the start of the subsidiary subject, and better still, at the end of the first main period (Hauptperioden). (Even in a small musical binary in which both sections contain cadences, the latter occur at the end of the second phrase, not the first.) Here, the slightness of the thematic substance is particularly noticeable, comparable in its triadic focus to the opening of the sonata Op. 57. In the Appassionata, the hesitant upbeat leads to rhythmically differentiated arpeggios and a brief and imperfect cadence on the dominant (Quintabsatz),

61

62

c l au d i a m au r e r z e n c k

which alternates between minor and major third and furthermore presents the smallest motivic unit of the sonata. On the other hand, a ‘thematic’ relationship connects the first movement of Op. 31 No. 1 with the opening of the sonata Op. 53: there, too, Beethoven opens with a single note (C) and follows it with a variety of triads, immediately and persistently repeated. That is just as unusual as the opening of the G major sonata, but any aspect of humour in Op. 53 disappears after two bars: above the repeated chords an initially slight, melodic substance appears, which then is subjected to serious development when metrically shifted and explored melodically, rhythmically (semiquavers from bar 10 onward), and harmonically (arpeggios starting in bar 12). Thus far, humorousness might be detected, given the slightness of thematic substance, the not-quite-fitting incises (Einschnitte) of the Satz, the premature closing phrase, the unmotivated dynamic changes (‘quick but playfully applied alternation of forte and piano’ [a6])33 and the brisk Allegro vivace tempo (the ‘wrong’ time signature, which makes the opening seem like an unusual three-bar Einschnitt, that is, a Dreyer, cannot yet be defined as humorous).34 Still, the suspicion arises that those surprising moments are not intended as simply humorous, but also meant to represent the inept ‘fumblings’ of the fictive composer-persona – one that apparently is not particularly ingenious nor adequately schooled in the rules of composition. On this view, one must therefore understand the repeated upbeat differently: it is far less about a true upbeat than about revealing the inability of the hands to play together, their asynchronicity, which either parodies the less than luminous brilliance of contemporary piano virtuosi or their custom of beginning figures such as runs or arpeggios with a short appoggiatura. In any case, the effect is to focus the listener’s attention on the limited material, derived more from improvisation than from composition. That Beethoven’s persona composes in alla breve metre while writing in 2/4 proves useful at the start of the second phrase (bar 12). In terms of the composed ‘double measures’, it actually begins half a bar too early, a fact which does not occur to the fictional composer precisely because he has written 2/4 bars. Here a slightly varied form of the ‘theme’ is introduced, not in the dominant key of D major, which was prematurely reached, but in F major. This harmonic shift does not seem to make sense after the previous brief modulation to D. Rather, it functions in a surprising and humorous way (‘sudden . . . entrance of foreign keys’ [c2]), referring once more to the fictional composer’s lack of skill. William Kinderman has noted that this shift is meaningful in relationship to Op. 53 and Op. 57.35 Indeed, considering the fact that the ‘theme’ here is

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

transposed down two fifths, the beginning of the sonata Op. 53 immediately springs to mind. There, after ending with an imperfect cadence on the dominant, Beethoven displaces the first four-bar phrase two fifths lower, shifting down two more fifths to F minor, before returning to the theme in C major via the third fifth, presented in the parallel minor. This kind of harmonic motion also characterizes the second thematic area, which lies four fifths above the home key of C major, and the further course of the movement, in which the upper fifth area is gradually scaled down; i.e., the second theme is constantly being taken to flatter key areas. In Op. 53, Beethoven unfolds a harmonic idea in creating the design of an entire sonata movement, something which cannot be said of Op. 31 No. 1. In Op. 57, after the first four-bar phrase in F minor, Beethoven moves the theme towards G flat major and then immediately back to F minor. This shift, too, has not only motivic meaning but formal consequences as well. First, what has already been heard melodically by bar 4, in three of the four voices of the phrase, is expressed harmonically: the alternation between neighbouring minor seconds (which will also be used in bar 23, in the fourth voice). In Op. 57, what is applied at the level of small detail will later define the harmonic plan of the whole movement: the second Hauptperiode begins in E major, in semitone relation to the tonic of the first and third Hauptperioden. The variant of the ‘theme’ in Op. 31 No. 1 consists in the opening run being presented in both hands together, with the upbeat ‘lost’ as a consequence of asynchronicity in the left hand. In any case, the music finally manages to arrive at a cadence in the tonic (bar 26), via a transitory cadence on C major in bar 22 (a ‘logical’ cadence on the dominant of F major). All that, along with the double repetition of the cadence in the tonic key, suggests that this is the one which the composer unsuccessfully tried to reach at the beginning. At the same time, the uppermost line in the right hand (g2 –b2 –d3 ), recalling bars 4–8 and 15–23, ascends as aimlessly as they did (see Example 3.2). The strong cadence in bar 30 is followed by a Passaggie based upon the second bar of the ‘theme’. A close look shows that it also integrates material from the first bar, whose notes g2 –f♯2 –e2 appear exactly in the corresponding positions. In this way, the composer presents a second variant of the ‘thematic’ head, but one which retrospectively takes even more from its already sparse essence. It appears as an ‘embellishment of no substance’. In the seventh bar of the Passaggie, which has brought the pianist to the lower-most edge of the keyboard, a reversal begins, three bars long and rising up to d/d1 . The listener likely expects the figure to continue for another bar – up until now the runs have been grouped into orderly, two-bar units – but

63

64

c l au d i a m au r e r z e n c k

Example 3.2 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 12–45

the ‘virtuoso’ instead veers into oscillating D major arpeggios for another five bars. He initially stops on the octave D/d at bars 44–5, the dominant, which he particularly emphasizes with a sforzato and an extra-long fermata, as if it were of particular importance. After the seemingly endless figuration, which constitutes a third of all the music up to this point, the presentation of the subsidiary subject in the dominant is now expected, not necessarily through modulation, but perhaps directly – a simple method sometimes used by Mozart and by the younger Beethoven. But that is too simple for the fictional composer of Op. 31 No. 1; he would like to use a proper dominant pivot, and therefore unexpectedly returns from the dominant ‘quickly to the previous key and principal melody’ (c4), that is, for the third time into that which Rochlitz called ‘frequent and surprising returns of the theme’ (b3).36

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

Example 3.3 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 45–65

This time he is smarter, and tries to cadence in a different way. Though he does not modulate to the normally-expected supertonic A, via E major, he nonetheless does reach F sharp via E minor (bars 53–4) and subsequently underlines this move via the insertion of a weak secondary dominant (bars 55–6). He is so satisfied with this accomplishment that he immediately repeats the modulation and the weak cadence, and then the ‘cadence’ twice by itself (see Example 3.3). A two-bar trill ensues, which serves to increase the tension of the new tonality and the subsidiary subject. It flows into a clumsy theme in B major forming two simple Vierer – but in 2/4 time! The left hand has a typical dance bass, striking the deepest bass note at the beginning of the bar and chords in the middle register afterwards (see Example 3.4). The harmonic course is constructed on the basis of varying two bars of tonic, then two of dominant, then tonic again, and finally one bar of dominant and a closing tonic. All this reminds the listener of the fatal trio in the ‘Pastoral’ Symphony, the first sketches of which, as already mentioned, originate in the year following the composition of this piano sonata. Beethoven did not write his first ‘wrong’ entry for the oboist of the trio, but already in this sonata he is foisting a probable blunder onto the composing performer. Indifferent as to whether he had picked up this simple ‘folk dance’ on the street, which perhaps only later made its name as the krakowiak,37 or if he had a round-dance in mind

65

66

c l au d i a m au r e r z e n c k

Example 3.4 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 65–72

and now had to fit its odd metre into the even metre of this sonata through lengthening of the second note in the melody, here he achieves – easily but enervatingly – through the syncopation in the melody and the repetitions of the accompaniment a ‘[jocular] formation . . . of rhythm’ (a2). If one were certain that, at this time, the agogic accent of the Viennese waltz was the custom, with its prematurely sounded and lengthened second beat, then it might be presumed that Beethoven had parodied that mannerism here. In other respects also, the melody of this subsidiary subject is simple; what is striking are the semiquavers which do not conform to the ‘German’ dance character, presented as a Verkettung in the fourth bar and as decoration in the seventh (‘unusual . . . transitions’, ‘curiously compounded figures’ [b4, b5]), and above all the integration of a dance theme in the first allegro of a sonata (‘leaps into popular, light, frivolous ideas’ [a3 = b6]). Immediately following, the melody is transferred to the left hand in the minor and ‘developed’, during which it largely loses its dance-like character. Initially, the two final bars (slightly modified) are developed, then the composer concentrates on large leaps, reversing their direction (‘reversal of conventional figures’ [a4]), and moving them by step from a to d (bars 83–6). In this way the composer modulates in the ‘wrong place’ down the circle of fifths by means of secondary dominants (‘bold slips into other keys’ [c3]). In the process, he continues to touch on scale degrees (Nebenstufen) of B minor and – quite out of place here – the tonic G major (bar 86), from which he cadences back towards B minor (‘unexpected quick return’ [c4]). With the melody back in the right hand (bar 88), there follows a sequential repetition of the last bar with leaps (bars 93–6) retracing the same modulating pattern as before (bars 83–6).38 This time the leaps are elaborated:

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

the top and bottom notes are held as voices and continued in alternating and intertwining ‘counterpoint’ (‘unexpected entrance of certain voices’ [c4]). In the heat of the moment, it goes unnoticed that these lines actually only comprise neighbour-notes and repetition; that is to say, what Rochlitz in 1806 characterized as ‘a more implied than implemented contrapuntal treatment of a very light theme’ and categorized as ‘lovely’ (niedlich).39 Example 3.5 shows that the B minor, when reached again in bars 97–8, receives a very long Anhang, in which the cadential motion v–i is repeated seven times (!). The temporary tonic is presented once very imaginatively with a major third, followed by the B minor triad, heard twice again piano (‘quick but playfully applied alternation of forte and piano’ [a6], and ‘closes in an . . . individual way’ [b7]). Much ado about nothing . . . Beethoven’s composer, without further ado, moves from B minor back to G major for the repeat of the first Hauptperioden. Since he was in no position to write (or did not think of writing) a second ending with a different harmonic progression, after the repeat of the whole first exposition (Hauptperioden), he arrives again at the G major of the beginning (contrary to the ‘natural course of harmony’ [c1]) and from there tries to shape the second, modulating Hauptperioden (development). Even here he initially does not get beyond three keys, which he reaches by downward root motion by fifths: C minor, F major and B flat major. This is accomplished with the chordal repetitions from bars 4–9, whose triadic structures are raised to unforeseen and numerously repeated heights (from bar 125). In this way he reaches the key of B flat major, so as to bring the running part of his ‘main subject’ into play, by which means he gets from B flat major via C minor to D minor. After some prevarication in the lower register of the piano, the latter is turned into D major (bar 158), which is complemented by and alternates with the Neapolitan E flat major (the first note of which, the fifth of b♭/B♭ in place of the expected d1 /d, is emphasized with a sforzato). As that alternation is accomplished by triadic arpeggios up and down, the virtuoso recalls the D major rolls of bars 39–44, which are extended by three bars in order for one to savour the new note, c, introduced in bar 166, and hence the transformation of the D major triad into a seventh chord (Example 3.6). As a sweetener for an intelligent public, which presumably has a taste for this reinterpretation, the seventh chord notes are now slowed down to crotchets, again set surprisingly in the left hand. From there, they are again driven upwards, against dramatically dotted repetitions of d in the right hand (or is this again the clumsiness of the virtuoso?), in three sweeps enriched with added thirds and fourths and by crossing of the right hand, up to the minor ninth, e♭3 . Thirteen years before the composition of this

67

68

c l au d i a m au r e r z e n c k

Example 3.5 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 74–113

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

Example 3.6 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 162–92

sonata, ‘crossing of the hands’ was considered worthy of a short digression in Daniel Gottlob T¨urk’s piano method: It [hand-crossing] was formerly considered a most significant piece of artistry. One would have very much doubted the skilfulness of a keyboard performer who could not play a Menuet with the hands crossed over one another. In the present time, one also finds pieces set by good masters where this technique occurs; only it is used less often, and not before it becomes necessary due to a particular thought. However, many recent composers sometimes require the hands to cross certainly only as a kind of horse-play.40

Needless to say, the crossing of the hands by our fictional composer does not concern a ‘particular thought’, as it does in Beethoven’s Sonata Op. 13, but is rather a kind of horsing or playing around (T¨andeley). The passage is extended further with slower D7 chords, some with the ninth added. The horse-play continues for ten bars, falling to pianissimo, until – completely by surprise, of course – in fortissimo come the upbeat g2 in the right hand and the asynchronous G in the left, ringing in the third Hauptperioden and the umpteenth repetition of the theme.

69

70

c l au d i a m au r e r z e n c k

But at least the composer is smart enough not to overstretch the patience of his audience, and he greatly shortens the return of the main ‘subject’ section. In effect, the music jumps from bars 8–11 (= bars 201–4), the cadence of which he repeats as a Grundkadenz (bars 205–8), straight to bar 53 (= bar 209). Here A minor moves through a weak cadence to B minor, and, in analogy to the earlier passage, this cadence is also immediately repeated. Correspondingly, the subsidiary subject appears a fifth lower than in the first Hauptperioden (in E major, bar 218). Since it played no role in the modulating second Hauptperioden, it is here presented in full and even expanded to forty-nine bars from its original thirty-three. The change starts in bar 23141 (= bar 78) and serves as a return to the tonic key (G major). Here, the second subject is presented once more, as in the first Hauptperioden, the sole difference being that here, after eight bars, there is no darkening to the minor mode: the clumsy ‘theme’ remains in G major, and the sequences by fifth play on secondary tones of G major, a process that continues for the thirty-three bars, up until the cadence. Only after that, in the coda (bars 266–70), is the play of major and minor, previously heard at bars 98–102, taken up again. The ending of the first Hauptperioden is also constructed in a similar way, such that the Hauptgedanke is heard again (for the fifth time!) from the upbeat to bar 280. This time it does not flow into the chords as it did from bar 4 onward, but continues with octave runs, which present its third variant (from bar 30), and ends correspondingly at bars 294–5 (as in bars 44–5). As a result, the coda – which the music has now reached – presents the largest part of the section which was omitted at the start of the third Hauptperioden. It again involves descending and ascending passagework and the subsequent, oscillating arpeggios; correspondingly, the section again ends on the note D/d. Why does the composer write a second coda? Firstly, he cannot and does not want to end on the fermata, the fifth of G major – the octave does not ‘resound’ sufficiently for him. Secondly, he does not want to continue as in the first Hauptperioden at bar 46 – that is, by resuming the beginning of the theme. Thirdly, he knows that one function of a coda is to confirm the cadence. He begins the second coda with two, initially separate moments: the asynchronous rhythm and a cadential turn, variants of which were previously heard countless times within the ‘main subject’ (Example 3.7). Because the chord is held on the first beat, those elements are meant to introduce a moment of reflection. In this regard, it is fitting that the d2 and d3 in the right hand, which repeats the bass of the dominant (bars 301 and 309) and/or the fifth of the tonic (bars 303 and 311), is preceded by a sentimental

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

Example 3.7 Beethoven, Piano Sonata in G major, Op. 31 No. 1/i, bars 294–end

(bars 300 and 302) then arpeggiated triplet (bars 308 and 310) embellishment that replaces the normal, appoggiatura-like semiquavers. From bar 312 on, those two moments merge, the asynchronous chords forming the final cadence, which consists simply of v7 –v7 –i. The cadence is played three times, the last time fortissimo, but even that is not a sufficient punchline for the composer: as a surprise, he follows up with a general pause, then two G major chords, played piano. That final, playful stroke, joined with all the foregoing analysis, suggests that, in Op. 31 No. 1, Beethoven applied all the means which Michaelis and Rochlitz described as producing the humorous. I hope that my analysis of the first movement has demonstrated concretely what Michaelis could have had in mind, when he later referred to Beethoven as one of the most humorous composers of the newest music,42 who was occasionally in a disposition to ‘rave with wisdom in such a manner’.43 Even today, to compose both competently and humorously at the same time is no simple matter. Nor is it an easy task for listeners and performers to distinguish between the ‘jocular’ and the serious-minded in musical compositions, especially those of earlier times.

71

72

c l au d i a m au r e r z e n c k

The reception of Op. 31 No. 1 As early as in the nineteenth century, the first sonata of Opus 31 encountered astonishing resistance. Carl Czerny characterized the first movement of Op. 31 No.1 as nothing more than ‘jocular and spiritedly lively’.44 In 1860, Wilhelm von Lenz also spoke of the ‘jocular, fantastic drive of the first movement’ but nevertheless considered the sonata as ‘nobly conceived’.45 Lenz glossed Adolf Bernhard Marx’s ‘masterful analysis’46 of the first movement, which he described as a spirited dialogue based on a musical idea (bars 1–10 [sic]) that, in itself, is ‘not very significant’. Marx had described the piece as a logical succession of events, based on material that made great demands of the composer.47 To Lenz, the second movement of the sonata appeared as a ‘fantasia-like outpouring in the elegant style’, in which the composer revealed himself as an ‘elegant and fashionable composer who surrenders to fulsome euphony, one might even say, to Italian passions’.48 Even such a glowing admirer of Beethoven as Wilibald Nagel found the movements ‘harmless’ (harmlos), ‘insubstantial’ (gehaltlos), and philistine,49 while such a detail-obsessed interpreter as Hugo Riemann could not say ‘conclusively . . . what is revolutionary about Op. 31 or what was new about it in comparison with Beethoven’s previous practice’.50 Paul Bekker was the first to stress the humorous and parodistic traits of the sonata, particularly of the first movement.51 But he also fell into the trap of the ‘noble virtuoso at the grand piano’ when, regarding the central part of the second movement, he ‘almost wanted to ask’, why ‘these far-flung modulations and this transition rich with dissonance were really necessary’.52 Theodor Veidl also discussed the Sonata Op. 31 No. 1, and in 1929 dedicated an entire study to Beethoven’s musical humour.53 Unlike Gustav Becking, Veidl did not confine himself to the scherzi;54 still, he only pointed to characteristics of the work that were relevant to his particular argument. It seems that his study has remained relatively unknown and of no consequence for Beethoven interpretation. A certain embarrassment is still noticeable in more recent interpretations of Opus 31, including those by pianists, as to how one might explain and perform the work, which seems curiously old-fashioned, especially in the light of N¨ageli’s exacting demand for virtuosity. For the most part, authors seem content to deal only with a few details of that opus. Konrad K¨uster dedicated a brief analysis to the sonata in 1994, in which he declared the asynchronous attacks of the chords in the left and right hands to be the ‘essential motivic principle of the first movement’.55 In the same year, in his analysis in Beethoven-Interpretationen, Joel Shapiro asserted that Beethoven did not meet N¨ageli’s demands in that sonata, but that the latter was nonetheless

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

‘extremely original, indeed . . . daring’.56 Even the well-known fact that the performance indication of the second movement, Adagio grazioso, is unique in Beethoven57 (and certainly not only in Beethoven) seems of no concern to Shapiro – otherwise it would have occurred to him that the slow tempo of ‘adagio’ cannot be coupled with ‘grazioso’, an expression always used in conjunction with lively movement (e.g. allegretto grazioso). Shapiro’s description is clearly not aimed at interpreting the originality of this music. That was first undertaken, if in a rudimentary way, by William Kinderman, who seems to take seriously Czerny’s reference to the close of the third movement as ‘very humorous, somewhat baroque’ (sehr humoristisch, etwas barock).58 Kinderman views the first movement, following an observation by Alfred Brendel, as proceeding with an air of paradox and comedy, also recognizing in it a touch of the bizarre. Particularly illuminating is Kinderman’s remark that the key to understanding the movement lies in ‘Beethoven’s ironic attitude toward the unbalanced, somewhat commonplace nature of his basic material’ and that ‘what unfolds in the development with startling vehemence later dissolves into coy, understated accents in the coda’.59 Kinderman’s interpretation swings between the revelation of ‘mockery’, which he finds in the second movement, and what he takes as Beethoven’s ‘ingenious demands on the musical tradition’ in the closing Rondo: He renders the expected repetitions of the main theme in the rondo unpredictable through variations in texture and rhythmic intensification, but reserves the most extraordinary events for the coda. Here, once more, he sees through, and beyond, the surface of his thematic material.60

Kinderman’s notice of Beethoven’s ironic stance towards musical tradition is on target, even if he backs away from its consequences and presents the facts as if Beethoven had ‘found’ his material and then worked on it, as if ‘material’ were a quasi-autonomous element. Beethoven, however, did not ‘find’ his material – he ‘invented’ it. Therefore it is not about the ‘demands on the musical tradition’ made by an ingenious composer; but rather the ‘many departures from the usual sonata form’ that in this work were intended to achieve something different, something which was typically acted out in the scherzo genre: a departure from the usual for humorous purposes, carried out under the guise of musical bungling.

Final remarks The reception of Beethoven’s sonata Op. 31 No. 1 illustrates the principal difficulty of distinguishing between serious and jocular compositions. This

73

74

c l au d i a m au r e r z e n c k

difficulty is enhanced by the period that separates the listener from the composer (as mentioned above). How can later generations, with their different musical means and their different mentality and wit, identify the transgression of norms (‘Abweichungen von der gew¨ohnlichen Sonaten-Form’) with certainty as intentionally humorous and not, for example, as daring ideas of a genius or, on the other hand, as sheer mistakes of a composer possibly not of first rank? The comparison with two other sonatas, Op. 53 and Op. 57, written shortly after Op. 31 No. 1 and making use of some ideas already displayed in the earlier work, was one of the ways to show the borderline between the serious and the jocular; the other one was to study what contemporary theorists wrote about humorous musical means. There are no general criteria for drawing this borderline, however. Already Beethoven’s contemporaries certainly found that to be a difficult task. To support that point, I conclude by quoting a footnote from Rochlitz’s article in the Allgemeine musikalische Zeitung, mentioned at the beginning of the present study: It is very difficult to make this sufficiently clear in words – it is very difficult, because it rests upon so many small and fine nuances; but it will be clear enough, when they [the listeners] pay close heed to the impression which the most accomplished masterworks in the genre make upon them, and how that effect is brought forth.61

Notes 1 Alfred Kalischer, in a note included in his edition of Franz Gerhard Wegeler and Ferdinand Ries, Biographische Notizen u¨ ber Ludwig van Beethoven (Berlin and Leipzig: Schuster und Loeffler, 1906), 107–8. Indeed, with the exception of the three sonatinas, Op. 49 and Op. 54, the only other piece without a dedication, besides Op. 31, is Op. 110; see Douglas P. Johnson, Alan Tyson and Robert Winter, The Beethoven Sketchbooks: History, Reconstruction, Inventory (Oxford: Oxford University Press, 1985), 128. 2 Sieghard Brandenburg presumes, calculating the times involved in postal deliveries, that N¨ageli had first written in May (Ludwig van Beethoven, Briefwechsel: Gesamtausgabe, ed. Sieghard Brandenburg, vol. 1: 1783–1807 (Munich: Henle, 1996), 108, note 1 on letter No. 99). That it should have taken from seven to eleven weeks (up to the letter to Horner in mid-July) for the post to deliver four letters between Vienna and Zurich seems a rather liberal estimate. 3 Beethoven, Briefwechsel, vol. 1, 107, No. 85. 4 The uncommon scope was noted in a contemporary review: ‘die erste Sonate ist die originellste’ (Sp [Johann Gottlieb Spazier], Review of ‘Lieferung I’ of the R´epertoire des clavecinistes, in Zeitung f¨ur die elegante Welt 3 (1803), col. 611).

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

5 The reading error was explained by Brandenburg in the correspondence (Beethoven, Briefwechsel, 108, note 4 on letter No. 85). 6 Wegeler and Ries, Biographische Notizen, 97. 7 Johnson, Tyson and Winter, The Beethoven Sketchbooks, 128. The few sketches for Op. 31 No. 2 are to be found on fo. 65v; those for Op. 31 No. 1 on fo. 88r and fos. 91v–96v. 8 From this it is also clear that the negotiations with N¨ageli took place first. 9 Quoted in Martin Staehelin, Hans Georg N¨ageli und Ludwig van Beethoven (Zurich: Hug, 1982), 19: ‘Es ist mir zun¨achst um Klavier-Solos in grossem Styl, von grossem Umfang, in mannichfaltigen Abweichungen von der gew¨ohnlichen Sonaten-Form zu thun. Ausf¨uhrlichkeit, Reichhaltigkeit, Vollstimmigkeit soll diese Produkte auszeichnen. Contrapunktische S¨atze m¨ussen mit k¨unstlichen Klavierspieler-Touren verwebt seyn. Wer . . . nicht zugleich Klavier-Virtuos ist, wird hier kaum etwas namhaftes leisten k¨onnen.’ 10 Johann Georg Sulzer, Allgemeine Theorie der sch¨onen K¨unste, 4th edn, vol. 1 (Leipzig: In der Weidmannschen Buchhandlung, 1792; reprinted Hildesheim: ¨ Olms, 1967), 485; [Friedrich August] Weber, ‘Uber komische Charakteristik und Karrikatur in praktischen Musikwerken’, Allgemeine musikalische Zeitung 3, No. 9 (26 November 1800), cols. 138–43, No. 10 (13 December 1800), cols. 157–62; Heinrich Christoph Koch, ‘Komisch’, Musikalisches Lexikon (Frankfurtam-Main: August Hermann der J¨ungere, 1802; reprinted Hildesheim: Olms, ¨ 1964), cols. 872–3; [Friedrich Rochlitz,] ‘Uber den zweckm¨assigen Gebrauch der Mittel der Tonkunst’, Allgemeine musikalische Zeitung 8, No. 1 (2 October 1805), cols. 3–10; No. 4 (23 October 1805), cols. 49–59; No. 13 (25 December 1805), cols. 193–201; No. 16 (15 January 1806), cols. 241–9. ¨ 11 Christian Friedrich Michaelis, ‘Uber das Humoristische oder Launige in der musikalischen Komposition’, Allgemeine musikalische Zeitung 9, No. 46 (12 August 1807), cols. 725–9. ¨ 12 Michaelis, ‘Uber das Humoristische oder Launige’, col. 729: ‘Hingegen ist unsere neueste Musik grossentheils humoristisch, besonders seitdem Joseph Haydn, als der gr¨osste Meister in dieser Gattung, . . . den Ton dazu angab . . . Haydn that es zuerst mit dem allgemeinen Effekt, und weckte eine Menge ber¨uhmter Tonk¨unstler der neuesten Zeit, in diesem Charakter zu schreiben.’ ¨ 13 Weber, ‘Uber komische Charakteristik und Karrikatur’, cols. 139–40: ‘Abweichung von der allgemeinen Regel’. 14 Koch, ‘Komisch’, col. 872: ‘eines wirklichen oder auch nur scheinbaren Mißstand[es] . . . zwischen dem dargestellten Objekte und zwischen den Kunstmitteln’. ¨ 15 Rochlitz, ‘Uber den zweckm¨assigen Gebrauch’, col. 7: ‘einander schroff entgegenstehet’. ¨ 16 Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 139: ‘specielle Anwendung der Regeln der Harmonik und Melodik, wodurch bey dem Zuh¨orer, dessen Geh¨or dazu gestimmt ist, ein Gef¨uhl des L¨acherlichen erweckt wird’.

75

76

c l au d i a m au r e r z e n c k

¨ 17 Michaelis, ‘Uber das Humoristische oder Launige’, col. 726: ‘unerwartetes Eintreten gewisser Stimmen oder Instrumente’. ¨ 18 Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 139: ‘launigte[n] Anwendung der großen oder kleinen Gabe von Witz, die dem Tonsetzer verliehen ist, dann auch in der musikalischen Parodie und absichtlich unternommenen musikalischen St¨umperey’. ¨ 19 Weber, ‘Uber komische Charakteristik und Karrikatur’, cols. 139–40: ‘der komische Tonsetzer, der nicht bis zur Karrikatur herabsinken will, oder muß, [sich] nur m¨aßige Abweichungen von der allgemeinen Regel [verg¨onnt], deren Befolgung ein Kunstprodukt der sch¨onen und ernsthaften Gattung zugleich erzeuget, und . . . S¨atze und Phrasen zu einem Ganzen [verbindet], welches dadurch den Anstrich von Paradoxie und durch eine veranlasste Ideenverbindung [den Anstrich] von L¨acherlichkeit in minderem Grade bekommt’. ¨ 20 Weber, ‘Uber komische Charakteristik und Karrikatur’, cols. 140–1: ‘sittsamer L¨acherlichkeit, oder von extravaganter Karrikatur’. His reference to more imitation and caricature than representation in Papagena’s arias before her rejuvenation in Die Zauberfl¨ote (col. 141) suggests that either he had made a mistake or that he had heard humorous inserted arias. Before the publication of his essay, that could only have been in the Freihaustheater in Vienna or perhaps in Prague, where the opera was premiered in October 1792. ¨ 21 Weber, ‘Uber komische Charakteristik und Karrikatur’, cols. 141–2: ‘Erfindung von nicht erwarteter Aehnlichkeit zwischen zwey musikalischen Gedanken und ihrer durch das Ueberraschende sich als geschickt und zwekm¨aßig ank¨undigenden Verbindung’. ¨ 22 Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 142: ‘die Gabe des musikalischen Witzes in reichem Maaße besitzen’. ¨ 23 Weber, ‘Uber komische Charakteristik und Karrikatur’, cols. 141–2: ‘von den neuern und neuesten Verderber[n] des guten musikalischen Geschmacks’. ¨ 24 Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 143: ‘wo die altmodischen Opernarien mit langen Passagen und Geleyer mit der aufsteigenden und absteigenden Septime meisterlich in’s L¨acherliche gezogen werden’. ¨ 25 Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 143: ‘k¨unstlich nachgemachte musikalische St¨umperey’; ‘eine gewisse Regellosigkeit’. ¨ 26 Michaelis, ‘Uber das Humoristische oder Launige’, cols. 726–7: ‘Da erscheinen Melodieen, Figuren, Passagen, kr¨aftige Accorde, man weiss nicht, wo sie herkommen; die Musik f¨angt auf eine so eigne Art, vielleicht h¨ochst simpel, mit einigen unbedeutend scheinenden Noten an, dass man gar nicht das interessante, unterhaltende St¨uck h¨atte vermuthen sollen, was sich daraus entspinnt. Der Komponist giebt den hergebrachten Figuren durch Umkehrung eine neue Gestalt, oder vertheilt die Noten und Accente so wider den Takt, verwickelt die Stimmen so wunderbar unter einander, weicht so k¨uhn in andere Tonarten aus, und kehrt eben so unerwartet schnell wieder in die vorige Tonart und Hauptmelodie zur¨uck, und schliesst auf eine so eigenth¨umliche Art, dass alles dies aus

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

27 28

29

30

31 32

33

der gewohnten Aus¨ubung des musikalischen Systems, aus den u¨ blichen Formeln der Komposition, und aus einem nat¨urlichen, regelm¨assigen Verfahren sich nicht erkl¨aren l¨asst, sondern eben so u¨ berraschend vorkommt, wie die Einf¨alle eines launigen oder humoristischen Erz¨ahlers . . . Die humoristische Musik ist bald komisch und naiv, bald ernsthaft und erhaben. Die Abweichung vom Hergebrachten, die sonderbare Verbindung des Fremdartigen und Entfernten, die Umkehrung der Figuren, die ungewohnte Art des Anfanges, der Ueberg¨ange oder Schl¨usse u. d. gl. hat Anfangs den Anschein von Ungereimtheit; weil diese aber doch wider Erwarten mit einer geistreichen Musik verbunden ist, weil sie im Verfolg und im Ueberblick des Ganzen sogleich aufh¨ort, Ungereimtheit zu seyn; so macht diese Musik den Eindruck des Komischen und kann zum Lachen reizen.’ ¨ Rochlitz, ‘Uber den zweckm¨assigen Gebrauch’, col. 247: ‘Ueberh¨upfen in popul¨are, leichte, t¨andelnde Ideen’. Some of these elements can be found in the systematic study by Theodor Veidl, published 130 years later (Der musikalische Humor bei Beethoven, Leipzig: Breitkopf und H¨artel, 1929), for instance, exaggerated dynamics (to emphasize a typical Quintabsatz, 38), rhythmic confusion (through anticipated or delayed octaves, 94–5), or a noticeable disruption of musical syntax (104). ¨ Michaelis, ‘Uber das Humoristische oder Launige’, cols. 725, 727: ‘gewisse[s] Herkommen’; ‘nat¨urliche[s], regelm¨assige[s] Verfahren’, ‘gewohnte Aus¨ubung des musikalischen Systems’. ¨ Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 142: ‘Schade, daß musikalische Witzeley [in] bloßer Instrumentalmusik angebracht, beynahe allezeit eines Commentars bedarf, den nur der Tonsetzer selbst liefern kann, und daher oft seinen Zweck zu frappiren und zu belustigen, nicht vollkommen erreicht!’ The letters and numbers here and in the following text refer to the listings in Table 3.1, above. The choice of a ‘wrong’ metre occasionally happens in Beethoven; for instance, in the first movement of the Fifth Symphony. It is therefore hard to decide whether the discrepancy between an apparent 2/4-metre and a composed Cmetre is here intended humorously or simply slipped in, as Heinrich Christoph Koch had observed to be the case in the work of numerous composers. See note 34 below. Early editions present a multitude of variants of these indications; but since the autograph is lost, it cannot be said with certainty what Beethoven wrote. Yet, if one considers the present musical context, then the thematic entries in bars 12, 112, 194 and 205 (which help to clarify the opening and reveal the common pattern of dynamic alternations, forte (bars 1–3) – piano (bars 4–7) – forte (bars 8–9) – piano (bars 10–11)), suggest that, with the brief piano cadential turn, Beethoven was again possibly taking a dig at the composer-persona’s straining to be original.

77

78

c l au d i a m au r e r z e n c k

34 Koch had already noted that even famous composers sometimes confused the correct metre of their work, notating an incorrect C instead of common-time metre. See Heinrich Christoph Koch, Versuch einer Anleitung zur Composition, vol. 2 (Leipzig: Adam Friedrich B¨ohme, 1787; reprinted Hildesheim: Olms, 1969), 295–6. The same can be said with regard to other metres and also to Beethoven, especially in his scherzo movements, where they are employed as humorous means. Several examples of this sort are discussed in my book Vom ¨ Takt: Uberlegungen zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert (Vienna: B¨ohlau, 2001). 35 William Kinderman, Beethoven (Oxford: Oxford University Press, 1995), 75. ¨ 36 Rochlitz, ‘Uber den zweckm¨assigen Gebrauch’, col. 247: ‘¨ofteres und u¨ berraschendes Wiederkehren des Thema’. Rochlitz adds that these repeats are always paired with ‘teasing’ (neckenden) and ‘unaffected variations’ (ungek¨unstelten Ver¨anderungen), leading to the genre that was ‘created and solicitously cultivated through this spirit of loveliness’ (durch diesen Niedlichkeitsgeist geschaffen, gehegt und gepflegt worden, col. 248): the rondo. 37 I am grateful to Ralf Stierlen for calling my attention to this dance, which appears in the published collection by Oskar Kolberg, Krakowskie, vol. 2 (Cracow, 1873); the emphasis on the second beat can be found there, for example, in dances Nos. 640–1, 663–5, 693–5. 38 There is a small divergence between bars 95–6 and bars 85–6: the G major chord is replaced in bar 95 by one that is increasingly perceived as B minor. ¨ 39 Rochlitz, ‘Uber den zweckm¨assigen Gebrauch’, col. 247: ‘mehr angedeutete als ausgef¨uhrte, kontrapunktische Behandlung eines sehr leichten . . . Thema’. 40 Daniel Gottlob T¨urk, Klavierschule oder Anweisung zum Klavierspielen f¨ur Lehrer und Lernende (Leipzig and Halle: Schwickert und Hemmerde, 1789; reprinted Kassel: B¨arenreiter, 1997), 190. My translation differs from that by Raymond H. Haggh in T¨urk, School of Clavier Playing (Lincoln: University of Nebraska Press, 1982), 182. 41 Bar 230 is here an additional bar, inserted for reasons of syntax: in bar 230 a Vierer starts with the return to the first bar of the subsidiary subject. ¨ 42 Michaelis, ‘Uber das Humoristische oder Launige’, col. 729. This statement was made in 1807, four years after the publication of Op. 31 No. 1. ¨ 43 Weber, ‘Uber komische Charakteristik und Karrikatur’, col. 142: ‘auf diese Weise mit Weisheit zu delirieren’. 44 Carl Czerny, Die Kunst des Vortrags der a¨ ltern und neuern Claviercompositionen ¨ (Vienna: Diabelli, 1842), ch. 2: ‘Uber den richtigen Vortrag der s¨ammtlichen Beethoven’schen Werke f¨ur das Piano allein’, 54: ‘launig und geistreich lebhaft’. In this passage Czerny refers to Op. 31 No. 1 as ‘Op. 29 No. 1’. At the time it was composed, Czerny had possibly suspended lessons with his teacher, and only came to know the piece later; not until 1804 did he return to Beethoven, who immediately presented him with the score of his brand new sonata, Opus 53, for sight-reading.

Beethoven’s ‘Piano Solo’ Op. 31 No. 1

45 Wilhelm von Lenz, Kritischer Katalog s¨ammtlicher Werke Ludwig van Beethovens mit Analysen derselben, 2nd edn, part 2 (Hamburg: Hoffmann und Campe, 1860), 160, 155: ‘launig-gaukelnden Getreibe des Hauptsatzes’; ‘edel geborene’. 46 Lenz, Kritischer Katalog, 156. Marx’s analysis is on pages 156–60. 47 See the oft-recurring words ‘nothwendig’, ‘also’, ‘folglich’ and the formulations ‘Komponist . . . Schuldner jenes Motivs’, ‘er muß es endlich gew¨ahren lassen’ (Lenz, Kritischer Katalog, 159). 48 Lenz, Kritischer Katalog, 161, 160: ‘fantasieartiger Erguss im eleganten Style’; ‘eleganter Modekomponist, der sich ges¨attigtem Wohlklange, man m¨ochte sagen, italienischen Gel¨usten u¨ berl¨asst’. 49 Wilibald Nagel, Beethoven und seine Klaviersonaten, vol. 2 (Langensalza: Beyer, 1905), 4, 12, 15. 50 Hugo Riemann, L. van Beethovens s¨amtliche Klavier-Solosonaten, vol. l part 2 (Berlin: Max Hesse, 1919), 319–20. 51 Paul Bekker, Beethoven (Berlin: Schuster und Loeffler, 1912), 150. He also discovered the ‘almost paradoxically sounding characterization’ (‘fast paradox anmutende Charakteristik’) of the slow movement; discussed below, in main text. 52 Bekker, Beethoven, 150: ‘erlauchte[r] Virtuose am Fl¨ugel’; ‘Fast m¨ochte man fragen, wozu diese umfangreichen Modulationen, diese dissonanzenreiche ¨ Uberleitung eigentlich n¨otig ist.’ 53 Veidl, Der musikalische Humor bei Beethoven (see note 28). 54 Gustav Becking, Studien zu Beethovens Personalstil: Das Scherzothema (Leipzig: Breitkopf und H¨artel, 1921). 55 Konrad K¨uster, Beethoven (Stuttgart: Deutsche Verlags-Anstalt, 1994), 158–61, here 158. 56 Joel Shapiro and Kenneth Drake, ‘3 Klaviersonaten G-Dur, d-Moll “Sturmsonate” und Es-Dur Op. 31’, in Beethoven: Interpretationen seiner Werke, ed. Albrecht Riethm¨uller, Carl Dahlhaus, Alexander Ringer, 2nd edn, vol. 1 (Laaber: Laaber-Verlag, 1996), 251. The analyses of Op. 31 No. 1 and No. 3 are by Shapiro; of Op. 31 No. 2 by Drake. 57 Shapiro and Drake, ‘3 Klaviersonaten’, 254. 58 Czerny, Die Kunst des Vortrags, 55. 59 Kinderman, Beethoven, 74–5. 60 Ibid., 74–5. ¨ 61 Rochlitz, ‘Uber den zweckm¨assigen Gebrauch’, cols. 247–8: ‘Es ist sehr schwer, dies . . . durch Worte deutlich genug zu machen – es ist sehr schwer, weil es auf so vielen kleinen, feinen N¨uancen beruht; es wird aber . . . anschaulich genug werden, wenn sie [die H¨orer] mit Genauigkeit auf die Wirkung achten wollen, welche die vollkommensten Meisterwerke dieser Gattung . . . auf sie gemacht haben, und wodurch diese Wirkung hervorgebracht worden.’

79

pa rt i i

Musical grammar

4

Metre, phrase structure and manipulations of musical beginnings da n u ta m i r k a

Changing metre In the eighteenth century the constancy of notated metre was taken for granted by both composers and listeners. Virtually every piece or movement was written with one time signature from the beginning to the end. Exceptions to this rule occur in movements preceded by a slow introduction, which normally differs from the following main part of a movement not only as regards tempo but metre as well. However, within the same notated metre, the composed metre could change. One class of changes consisted in varying the period or phase of a composed metre. The former could be caused by imbroglio and hemiolas, the latter by syncopations.1 In contemporaneous music theory such metric changes were described as ‘confusions’ (Verwirrungen) by Joseph Riepel and ‘shifts’ (R¨uckungen) by Heinrich Christoph Koch.2 Because they referred to the mechanism of metric perception, they were accessible to all attentive listeners including less cultivated ones (Liebhaber). Another class of eighteenth-century metric manipulations relied upon the theoretical knowledge of the listener and hence could be appreciated only by Kenner able to properly interpret cues provided by the composer. Such manipulations involve a change of the level of the metrical hierarchy corresponding with the so-called ‘parts of the measure’ (Taktteile). The most important cues used by eighteenth-century composers to indicate the metrical level of Taktteile are ending formulas represented by Einschnitte, Abs¨atze and Kadenze.3 As repeatedly emphasized by authors of composition handbooks, caesura notes of these formulas must fall on a strong Taktteil (downbeat).4 Friedrich Wilhelm Marpurg and Koch explicitly refer to the location of caesuras as the criterion that allows one to distinguish between simple and compound metres characterized by the same size of notated measures but different rhythmical values of Taktteile. In the late eighteenth century this distinction pertains primarily to two pairs of metres: on the one hand, compound 4/4 (c) with crotchet Taktteil, versus simple 2/2 or alla breve (C) with minim Taktteil; on the other hand, compound 6/8 with quaver Taktteil, versus simple 6/8 with dotted-crotchet Taktteil. Because

84

da n u ta m i r k a

Example 4.1 Riepel, De Rhythmopoeia, 52

measures of simple metres contain only one downbeat, in these metres a caesura can fall only at the beginning of a notated measure. Not so in compound metres. As Koch explains, because every measure of a compound [4/4 or 6/8] meter consists of two measures of a simple [2/4 or 3/8] meter, it must necessarily contain two strong and also two weak Taktteile, and hence the caesuras of the resting points of the spirit [Ruhepuncte des Geistes] in every compound meter must be allowed to fall both on the first and on the second half of the measure; and this last [location] is particularly the indication which beginners must follow in order to distinguish compound meters from simple ones.5

From the fact that the level of Taktteile is indicated by the metrical location of the caesura note, it follows that the change of this location makes it possible to recognize a change of metre between compound and simple. Such changes occur frequently in the course of eighteenth-century compositions and are acknowledged by authors of composition handbooks. The earliest examples of switching between 4/4 and 2/2 in the course of a piece are shown by Riepel. Although Riepel does not explicitly refer to the location of caesuras, it may be observed that the Absatz of the first melodic section in Example 4.1 falls in the middle of bar 4, indicating compound 4/4 metre (c). The Kadenz falling at the beginning of bar 12 and elided with the subsequent section (NB) testifies to the change from 4/4 to 2/2 metre (C). After the return to 4/4, the Absatz

Manipulations of musical beginnings

Example 4.2 Koch, Versuch, vol. 3 (Leipzig: Adam Friedrich B¨ohme, 1793), 224–5

in bar 15 falls again in the middle of the notated measure.6 Koch mentions changes of metre occasioned by changes of the metrical level of Taktteile as one type of interpolation (Parenthese) consisting in ‘the insertion of melodic sections of a simple meter in a piece composed in a compound meter’.7 His example, reproduced above (Example 4.2), also demonstrates a switch from 4/4, indicated by the time signature c, to 2/2. Caesura notes of Abs¨atze and Einschnitte are marked by Koch with squares and triangles respectively. In bars 1–6, composed in 4/4 metre, Abs¨atze fall in the middle of measures. In

85

86

da n u ta m i r k a

the phrase beginning at bar 7, however, Einschnitte (triangles) and Abs¨atze (squares) occur at the beginnings of bars 8, 10 and 12, with their caesura notes apparently shifted to weak positions by appoggiaturas. In bar 14 the caesura note of the Kadenz, falling similarly on the first beat, is elided with the beginning of the subsequent phrase. Elsewhere in his handbook Koch admits that the direction of metrical change can be reversed so that a piece can start in simple metre and change to compound metre.8 It is worth noting that Koch’s objective in discussing this example is not so much to identify metrical change but to explain its consequences for phrase structure. He emphasizes that ‘one must be careful not to take sections of an even number of measures for those of uneven number, nor simple meters for compound, nor incomplete segments or incises (Einschnitte) for phrases (Abs¨atze)’.9 The same objective stands behind Riepel’s example. By means of small and large numerals, he demonstrates that the shift of Taktteile from crotchets to minims has consequences for counting measures.10 Changes of the metrical level of Taktteile discussed by Riepel and Koch have only seldom been recognized by modern scholars. The first to address this issue was Wye J. Allanbrook.11 According to her, such changes often accompany changes of musical topoi. She demonstrates this in the finale of Mozart’s String Quartet in G major, k387, in which the metre 4/2 should properly be assigned to the opening fugal section, and 2/4 to the swift contredanse. The time signature C, which Mozart actually uses in the score, lies half way between these extremes and relates to the alla breve metre of the closing idea identified by Allanbrook as a bourr´ee.12 More recently, Claudia Maurer Zenck found changes of the metrical level of Taktteile in several pieces by Mozart, Beethoven and Schubert.13 Her most extensive example comes from the opening movement of Mozart’s D minor Quartet, k421, where the principal theme (bars 1–8) is set in 2/2 (C) and the subsidiary theme (bars 24–32) in 4/4 metre. Although Maurer Zenck does not attempt to identify their topics, the former theme represents what Allanbrook would call ‘exalted march’ while the latter is an example of ‘singing style’. As the author convincingly demonstrates, sophisticated decorations obliterate the location of Abs¨atze and Einschnitte in the course of the transition section, thereby allowing Mozart to shape a sort of metric modulation parallel to the harmonic modulation between the themes.14 Common to the examples analysed by Allanbrook and Maurer Zenck is that metrical changes take place between longer sections. In this way metrical changes participate in the articulation of musical form. By contrast, in this essay I discuss changes of the metrical level of Taktteile that take place at the very beginning of a given piece or movement, inside its main theme. Such changes have not been

Manipulations of musical beginnings

Example 4.3 Haydn, String Quartet in B flat major, Op. 50 No. 1/ii, bars 1–6

investigated until now. My examples come from Haydn and Mozart’s chamber music for strings. Because a change of metre entails an adjustment in the counting of measures, every instance is of consequence for phrase structure.

Changes of Taktteile in compound measures Two particularly obvious examples of such changes can be found in slow movements of Haydn’s String Quartets in B flat major, Op. 50 No. 1 and in D major Op. 50 No. 6. As in the examples shown by Riepel and Koch, the change of metre between compound and simple can be recognized after the location of caesuras. In Op. 50 No. 1 (Example 4.3), the 6/8 metre is simple at first, with Einschnitt and Absatz falling, respectively, at the beginning of notated bars 2 and 4. In either case the caesura note is followed by a feminine ¨ ending, or overhang (Uberhang), which reaches to the following weak beat. In bars 5–6 the metre changes to compound 6/8, so that the Absatz at the end of the first reprise arrives in the middle of bar 6, after four measures of the composed 3/8 metre. The change of metre makes the theme curiously asymmetrical; yet, in fact, both phrases are four measures long, having been realized at two different metrical scales.15 A reverse succession of metres takes place in the slow movement of Op. 50 No. 6 (Example 4.4). In the two regular four-bar phrases which make up its theme, the Abs¨atze – Quintabsatz on the dominant and Grundabsatz on the tonic – fall clearly in the middle of notated bars 2 and 4, indicating compound 6/8. The phrase starting in bar 5, which seems at first to be a repetition of the theme in the higher octave, then takes a different course and reaches the Quintabsatz on the downbeat of the notated bar 8, thus after four measures of simple 6/8. The length of two

87

88

da n u ta m i r k a

Example 4.4 Haydn, String Quartet in D major, Op. 50 No. 6/ii, bars 1–10

four-bar phrases in 3/8, notated as compound 6/8 metre, is thus balanced by one phrase in simple 6/8 metre. Even so, the number of measures in these two melodic sections is not equal (8+4). Interestingly, the third phrase is extended beyond the caesura note by means of a two-measure overhang. As will be explained below, Koch considered an overhang of this size to be incorrect. It represents Haydn’s further manipulation of the phrase structure beyond the change of metre. Even more intriguing is the opening movement of the String Quartet in E flat major, Op. 64 No. 6 (Example 4.5). In this example the level of

Manipulations of musical beginnings

Example 4.5 Haydn, String Quartet in E flat major, Op. 64 No. 6/i, bars 1–8

Taktteile can be determined by a knowledgeable listener already before the caesura on the basis of another criterion derived from the eighteenth-century principle concerning the articulation of Taktteile at the musical surface.16 This principle, systematically elaborated by Koch, was related to his concept of Metrum. According to Koch, the Metrum of a given piece is always based on a certain metrical level.17 This level is normally that of Taktteile, although in slow tempo Metrum is often based on the lower metrical level of Taktglieder. In either case, all beats of the metrical level constitutive of Metrum must be consistently articulated by attacks: When the melody of the main part does not preserve the accepted Metrum perceptibly enough or deviates from its movement too much, then either the bass or one of the subsidiary parts present can and must maintain the similarity of this movement.18

Because at the beginning of Example 4.5 the lowest metrical level consistently articulated by attacks is represented by minims, these rhythmical values are to be interpreted as Taktteile of 2/2 or alla breve metre in compliance with Haydn’s time signature, C. Metrical interpretation in the compound 4/4 metre (c) is not possible because in bar 1 crotchets are not articulated. When they occur in bar 2 and following, they do so only as Taktglieder. But in alla breve, caesuras of Abs¨atze and Kadenze are allowed to fall exclusively on the beginnings of notated measures. Instead, the Quintabsatz occurs in the middle of bar 4, and the same is true of the Kadenz in bar 8. What happens here is a sort of metric modulation, which Haydn carries out in the course of the first phrase, and then again in the course of the second phrase, from simple 2/2 to compound 4/4 metre. As a result of this modulation, the

89

90

da n u ta m i r k a

(a)

(b)

Example 4.6 Haydn, String Quartet in B minor, Op. 64 No. 2/i, (a) bars 1–4 (b) recomposition

crotchets are advanced from the initial role of Taktglieder to that of Taktteile, and the minims, originally taken for Taktteile, turn out to represent the metrical level corresponding with real 2/4 measures (Takte) contained in every notated measure of 4/4. The shift of Taktteile to a lower metrical level is clearly perceivable not only because of the diminution of rhythmical values at the musical surface but also because of the gradual acceleration of the harmonic rhythm. Note that articulation by attacks as a criterion for the identification of Taktteile in the above example allows for ‘metric modulation’ only from simple to compound metre. It is not effective for metrical change in the other direction. This is because the level of Taktteile in a simple metre is higher than in its equivalent compound metre. If Taktteile of the compound metre are articulated consistently by attacks, so are also Taktteile of the simple metre. In such cases, interpretation of a given passage in both metres 2/2 or 4/4 is equally possible and the caesura, rather than changing the metre, clarifies it in the first place.19 In light of these remarks one can better

Manipulations of musical beginnings

appreciate the exceptionality of the manipulation accomplished by Haydn in his B minor Quartet, Op. 64 No. 2 (Example 4.6a), in which the first phrase begins in compound 4/4 metre but ends in simple alla breve (2/2). Haydn is able to realize this direction of metric modulation because he takes his cue from the phrase structure. After the opening D, whose function remains at first unclear for the listener, the repeated quaver motive (indicated by the brackets) will be most naturally understood as an imperfect incise. One of the most popular types of four-bar phrase in eighteenth-century compositions starts precisely with a repetition – exact or varied – of an imperfect incise (unvollkommener Einschnitt) of one measure followed by a perfect incise (vollkommener Einschnitt) of two measures.20 If this scenario was realized by Haydn, the phrase could likely receive a continuation similar to that shown in Example 4.6b. The size of the incomplete incises allows a knowledgeable listener to determine the metre as 2/4 which can be notated as compound 4/4 in keeping with the time signature c Haydn used in the score. Yet the actual continuation of the phrase takes a different course and forces the listener to revise his conjecture concerning the metre. The Quintabsatz of bar 4, with the appoggiatura falling on the first half of the measure and its resolution reaching to the second half, clearly indicates alla breve metre. Further cues pointing to the change of metre in the course of the phrase are the two minim values in bars 3 and 4, which interrupt the articulation of crotchets as alleged Taktteile of the compound c metre. (Note that, in and of themselves, these interruptions would not testify to the change of metre because the articulation of crotchets at the beginning of the phrase does not unequivocally indicate 4/4. Rather, as explained above, it admits of either 2/2 or 4/4.) It is interesting to observe that the articulation of minims was initiated already by the opening D before the articulation of crotchets. Whereas, in the context of the two following incises, they were interpreted as spanning entire measures of the 2/4 metre, in 2/2 metre minims represent the level of Taktteile ultimately explained only at the end of the phrase by its Absatz. As a further consequence, the change of metre attested to by the Absatz clarifies also the role of the opening note D as the actual beginning of a four-bar phrase in 2/2 metre. (The fact that this clarification comes only in retrospect is indicated by left-pointing arrows in Example 4.6a.) The structure of this phrase, highly unusual by eighteenth-century standards, serves the purpose of the metric modulation and therefore can be understood only in that light.21 Another peculiarity of this phrase is the change of key from D major, suggested at its beginning, to B minor at the end. Whereas this tonal

91

92

da n u ta m i r k a

Example 4.7 Haydn, String Quartet in B flat major, Op. 55 No. 3/iv, bars 1–8

manipulation has been recognized by many authors writing about Haydn’s string quartets, the metrical change from 4/4 to 2/2 apparently escaped their attention. In point of fact, these two manipulations constitute two sides of the same trick executed by Haydn in the opening movement of Op. 64 No. 2, which aims at producing a surprising twist in the expressive character of the theme: from its humorous opening in D major and 2/4 – announcing a thematic idea proper for a quick finale rather than an opening movement – to the grave reflective ending in B minor and 2/2. The moment of this twist can be precisely indicated. It happens with the entrance of the dominant seventh in the middle of the phrase (bars 2/3), which interrupts the realization of the scenario reconstructed in Example 4.6b at once in its metrical and tonal dimension. The effect of this chord is enhanced by the sudden change from solo to tutti and from piano to forte as well as the drastic interruption of Metrum, previously established on the level of quavers as Taktglieder of the alleged 2/4 metre. A particularly complex manipulation takes place in the finale of Haydn’s String Quartet in B flat major, Op. 55 No. 3 (Example 4.7). The movement

Manipulations of musical beginnings

starts from a typical Verwirrung which yields the phase of the metrical structure shifted in perception by a quaver in comparison with the notated metre. After the perceived metrical structure is aligned with the notation in bar 2, Haydn takes advantage of the Quintabsatz in bar 4 in order to playfully shift the level of Taktteile. Because this Absatz is felt as the closure of a regular four-bar phrase (Vierer) initiated by the D in the middle of the notated bar 2, it reveals that the 6/8 metre is compound, each of its notated measures comprising two real measures of simple 3/8. That the caesura note of this Absatz falls on the first half of the notated measure and not on the second half, as normally is the case with regular phrases in compound metres, has to do with the fact that the Vierer begins in the middle of bar 2, being preceded by a huge anacrusis of more than three real measures of 3/8, which is not counted to its length and whose function remains unclear for the listener at this stage of his understanding. This understanding is immediately challenged, however, by the material following the caesura. Whereas one would expect a single stroke of the dominant chord in bar 4, this chord is repeated many times and, in the course of these repetitions, provided with a double appoggiatura which ultimately resolves to the plain dominant on the second half of the notated measure. If the real metre were 3/8, this resolution would fall on the subsequent downbeat after the deleted barline. This means, in turn, that the size of the material following the caesura would go beyond the limit of one measure permitted for an overhang. According to Koch, it is beyond doubt that the caesura of an Absatz must fall only on one strong Taktteil. Although the Absatz itself can end with a feminine ending, that means, be prolonged until the weak Taktteil, . . . the ending of an Absatz can never take two measures, that is, two strong and two weak Taktteile.22

The manipulation performed by Haydn corresponds, in fact, to an example of incorrect notation provided by Koch (Example 4.8a). The conclusion he draws is that this example is not really in 3/8, but rather in (simple) 6/8 derived from 2/4 through triple subdivision of Taktteile into Taktglieder, and its correct notation should appear as in Example 4.8b. The same conclusion also applies to Haydn, who, however, does not commit a notational mistake. Although the 3/8 metre, which the listener recognizes at first on the basis of the caesura, must be corrected to 6/8 in the face of its overhang, the composer writes the whole passage in 6/8, which provides a comfortable framework for both metres. Observe that the caesura falls not only in the fourth bar of 3/8, counting from the D in bar 2, but also in the fourth bar of the simple 6/8 counted from

93

94

da n u ta m i r k a

(a)

(b)

Example 4.8 Koch, Versuch, vol. 2 (a) 320 (b) 324

the beginning of the theme. Because of the revision of metre forced by the overhang, the entire first phrase of the theme turns into a Vierer aligned with the notated barlines in the score and incorporating in its structure the initial semiquaver runs. (As in the previous example, the retrospective character of this recognition is indicated by leftward-pointing arrows.) From this point of view it is a true stroke of genius which, in the end, brings all the puzzles of this peculiar beginning into a logical pattern. Again, as in Op. 64 No. 2, the complexity of this pattern reflects the complexity of the process through which the phrase comes into being. But this is not yet the end of Haydn’s play with the listener’s expectations. The final stage of this game takes place during the repetition of the phrase in bars 5–8. Because this time the phrase ends not with the Quintabsatz but with a Kadenz, it includes one more chord – the tonic – which falls in the middle of bar 8, in this way unequivocally attesting to 3/8 metre notated in compound 6/8. Simple 6/8 does not arise here at all. The entire second phrase, starting with the D in bar 6, is five bars long (F¨unfer), and is preceded by an eccentric three-bar anacrusis which does not count in its length.

Manipulations of musical beginnings

Changes of Taktteile in double measures In all the examples analysed thus far, the metre characterized by the smaller rhythmical value of Taktteile is notated in compound measures of 4/4 (c) or 6/8. Because the size of such measures is identical with that of equivalent simple metres characterized by larger Taktteile, switches from one of them to the other do not change the location of bar lines. Metric notation is correct in either case. Yet changes of the metrical level of Taktteile can also take place in pieces in which the metre characterized by the smaller rhythmical value of Taktteile is notated as simple, for instance 2/4 or 3/8. Such changes result in switches between 2/4 and 2/2 or 3/8 and 6/8. Because the size of notated measures in every pair of metres is different, each measure of the ‘larger’ metre must be notated as two measures of the ‘smaller’ metre. This type of notation, labelled by Maurer Zenck as ‘double measures’ (Doppeltakte),23 was considered incorrect by theorists of the time. Riepel points out that it is a mistake to notate 6/8 as 3/8 or 2/2 as 2/4.24 Koch subjects double measures to an extended critique in many passages of his treatise. The primary reason for his critique is that this notation misrepresents the relationship of strong and weak beats and in this way contradicts the nature of metre. It does so because ‘both main parts of each measure, thesis and arsis, are separated from each other by the barline and occur in the outer form of two bars’.25 What can be seen between two barlines on the score of such incorrectly notated pieces is not a measure (Takt) but only Taktteil of the composed metre. Yet, when every Taktteil is notated as Takt, all beats are downbeats and the differences between them are eliminated. As a further consequence, the notation in double measures misrepresents the phrase structure. Because the caesura note of Kadenz or Absatz indicates the last measure of the phrase, a regular four-bar phrase, such as that in Example 4.9a, appears irregular in this notation, its length being expressed in an uneven number of measures. Koch illustrates this in Example 4.9b. In order to achieve a four-measure phrase, it is necessary to notate the weak beat of the fourth measure as a quaver bar following the caesura. While not to be counted in the preceding phrase, this bar does not yet start a new phrase. In the melody it is thus either empty or, as in Example 4.9a, filled with an incorrect overhang exceeding the length of one bar. The phenomena of the empty bar and incorrect overhang can serve as indicators of double measures in eighteenth-century compositions. They are related to the criterion of caesura because an empty bar or incorrect overhang always arises as a result of the location of the caesura note of the Kadenz or Absatz in the preceding notated measure. Referring to this criterion, Maurer

95

96

da n u ta m i r k a

(a)

(b)

Example 4.9 Koch, Versuch, vol. 2, 398, (a) Figure 13 (b) Figure 14

Zenck finds double measures in numerous pieces by Mozart and Beethoven. Even if, for Koch, notation in double measures is always a mistake – ‘a mistake which one encounters almost daily in modern compositions, and which is promoted by the tyranny of habit, but which remains a mistake, whether committed by a beginner or by an already trained composer’26 – it certainly does not occur by mistake in the pieces of masters. Observe that Koch’s examples of double measures concern exclusively an improper use of time signatures 2/4 for alla breve and 3/8 for simple 6/8 metre. In either case, the time signature of the composed metre was available and could be used in the properly rewritten versions of such examples. Instead, the double measures identified by Maurer Zenck result from the use of old metres whose time signatures had already been dropped from the notational conventions of the late eighteenth century. They occur mostly in minuets and scherzos, which are often composed in 6/4, although notated in 3/4. In certain cases composed 6/4 and 3/4 metres alternate, resulting in changes of the level of Taktteile between the crotchet and the dotted minim. Two particularly curious examples of such changes, not discussed by Maurer Zenck, can be found in minuets from Mozart’s String Quintets k515 (Example 4.10) and k614 (Example 4.11). In k515 the caesura arrives with the dominant at the beginning of bar 3, although the accompaniment enters only one crotchet later. The tonic occurring on the last beat of bar 3 forms an auxiliary chord within the prolongation of the dominant. According to Koch, such auxiliary chords are introduced ‘for more harmonic variety’ and

Manipulations of musical beginnings

Example 4.10 Mozart, String Quintet in C major, k515/iii, bars 1–10

result in an ‘improper’ harmonization of ending formulas.27 In his example (quoted in Example 4.12), the structural tonic of the Grundabsatz is varied by means of the auxiliary dominant. The squares indicate that the position of the caesura remains unchanged. In Example 4.10 the auxiliary tonic arises from an improper harmonization of the melodic tone E in the overhang following the structural dominant. The incorrect size of this overhang, reaching over the barline to bar 4, betrays the fact that the composed metre is 6/4. An analogous situation can be observed in k614 (Example 4.11). There too, the caesura arrives in bar 3 and is followed by an overhang reaching to bar 4. The prolongation of the dominant is even more evident due to the ties in the middle voices. Because these voices do not move, the E♭ played by the second viola – corresponding with the C of the cello in Example 4.10 – does not result in an improper harmonization but merely constitutes an auxiliary note in the bass. In both examples the 6/4 metre subsequently changes to 3/4. In Example 4.10 this is evidenced by the Kadenz in bar 10, closing the regular four-bar phrase initiated in bar 7. (If the 6/4 metre continued until the end of the first reprise, the caesura note of this Kadenz would fall incorrectly in the middle of a composed measure.) Similarly, in Example 4.11 the 3/4 metre is confirmed by the Kadenz in bar 12. As a result of the metrical change, the phrase structure becomes highly complicated. In both examples the first four notated measures are to be

97

98

da n u ta m i r k a

Example 4.11 Mozart, String Quintet in E flat major, k614/iii, bars 1–12

understood as a perfect incise in the composed 6/4 metre. However, in both of them the listener achieves this understanding in retrospect. On the one hand, the 6/4 metre is explained only at the end of the incise by its caesura. In this way the caesura determines the size of composed measures to be counted by the listener. On the other hand, it determines the moment when counting should start. Because it arrives at the end of the incise as its second measure, it renders the foregoing passage in parallel thirds as the first measure, even if the listener would have originally been inclined to hear an unaccompanied melody as an ‘elongated upbeat’ or anacrusis, such as that seen earlier

Manipulations of musical beginnings

(a)

(b)

Example 4.12 Koch, Versuch, vol. 2, 400

in Example 4.7, which does not count in the length of the phrase.28 The understanding of the first melodic section, achieved at the caesura, guides the listener’s interpretation of the melodic section after the caesura. Since in both examples this section begins in an analogous way, the listener will count it analogously by taking bars 5–6 for the first measure of a new incise. Yet this interpretation must be revised in light of the different continuation containing the change of metre. The revision concerns not only the size of measures but also the identification of measures to be counted as first. In k515 the notated bars 5–6, containing the unaccompanied melody, turn out to be an elongated upbeat to a regular Vierer in 3/4 started at bar 7, and hence not to be counted as part of its length. The introductory function of this melody, belied in the first instance, is thus confirmed in the second. In this way Mozart dupes his listener twice by means of one and the same musical device. He plays out this trick once again in k614, with the sole difference that the structure of the following phrase (bars 7–12) is more complicated due to the internal repetition of bars 7–8 in bars 9–10. Although the phrase is six bars long, the repetition determines that it is not a Sechser but rather an expanded Vierer. Repetitions were a standard means of phrase expansion in the eighteenth century and expanded phrases (erweiterter S¨atze)

99

100

da n u ta m i r k a

starting with repeated perfect incises of two measures were most common of all. Koch’s statement that ‘such a four-measure phrase, which has been extended to six measures by the repetition of two, is always considered as a four-measure unit with respect to the rhythmic relations of phrases’29 implies that the repeated measures are supposed to be counted by the listener with repeated numerals. This is why, in Example 4.11, bars 9–10 are numbered once again as the first and the second.

Change of Taktteile and perceptual factors It is worth noting that, in the minuets from k515 and k614, the change of metre from 6/4 to 3/4 is signalled to the listener already before the caesura by the change of the harmonic rhythm. In k515 the acceleration of the harmonic rhythm takes place from bar 7, in k614 from bar 10. In both examples the effect of the metrical change is confirmed by caesuras, but not produced by them. A similar observation can be made in reference to the earlier examples. In all of them changes of metre are accompanied by changes of the harmonic rhythm. In some of them the changes are additionally enhanced by other factors including prevalent rhythmical values, the size of melodic sections, and the lowest metrical level consistently articulated by attacks (Koch’s level of Metrum). Although these factors do not represent absolutely reliable criteria for the identification of Taktteile, and their changes often take place without any metrical changes, they are indispensable if metrical changes are to be perceived by the listener rather than merely recognized.30 Furthermore, perceptual factors play an important role in indicating changes of Taktteile in the course of phrases that do not display any subdivision into incises or in which such a subdivision is not unequivocal. The position of Abs¨atze at the end of such phrases does not necessarily determine the metre at the beginning if this metre is not confirmed by an earlier Einschnitt. That it is not always possible to unequivocally identify Einschnitte in the course of phrases has to do with two different problems. The first is related to decorations of ending formulas, which may obliterate the position of caesuras. While these decorations, divided by Koch into appoggiaturas, overhangs and lead-ins,31 are unproblematic insofar as the harmonic goal is achieved in a straightforward way, they may cause problems when receiving an ‘improper’ harmonization. Since Koch offers no criteria to distinguish chords resulting from an improper harmonization and prolonging the harmonic goal of a given melodic section from structural chords preceding

Manipulations of musical beginnings

Example 4.13 Koch, Versuch, vol. 2, 408

such a goal, in certain cases the position of the caesura is not easily determined. While this problem is less evident in Example 4.12 – which shows an improper harmonization of an overhang – because of its triple metre, it becomes critically important in another example, kept in duple metre, with which Koch illustrates an improper harmonization of an appoggiatura (Example 4.13). As observed by Maurer Zenck, if the size of measures and rhythmical values in this example were doubled, it would be difficult to decide whether it illustrates an improper harmonization of an appoggiatura in simple 2/2 or, rather, an undecorated caesura in compound 4/4 metre.32 The second problem concerning the identification of Einschnitte is that such caesuras, or ‘resting points of the spirit’ (Ruhepunkte des Geistes), are marked by several different formulas which can otherwise occur in the course of melodic sections without any closing effect. As Koch remarks, the ending formulas of these sections are so various and can be formed in such manifold ways that it would be very questionable to decide, by means of these figures, where resting points are present in the melody; not to mention that such figures in the melody also can be used where there is no resting point. In short, nothing concrete can determine the places where they are in the melody.33

Both problems are illustrated in Example 4.14 from the minuet of Haydn’s String Quartet in E flat major, Op. 64 No. 6. The Absatz, falling in bar 6, indicates that the metre at the end of the first phrase is 3/4. But what is the metre at the beginning? If the answer to this question were to be given on the basis of a caesura occurring earlier in the course of the phrase, it would not

101

102

da n u ta m i r k a

Example 4.14 Haydn, String Quartet in E flat major, Op. 64 No. 6/iii, bars 1–14

be unequivocal. Is there any caesura here at all? And if so, does it fall in bar 3 or 4? Theoretically, both interpretations are possible. Provided the caesura is achieved with the subdominant in bar 4, the metre is 3/4 throughout the phrase and the chord of bar 3 has a structural function of an applied dominant. In this case, however, the caesura does not represent an Einschnitt but an Absatz. This is because, by arriving in the fourth measure, it closes a four-measure phrase.34 The two following measures (bars 5–6) are to be taken for an appendix (Anhang) closed with another Absatz on the tonic. Although not quite common, appendices closing on a different harmony than the proper phrase are discussed by Koch as an available compositional option.35 Yet it is also possible to interpret the first four notated measures as a perfect incise in 6/4. In this interpretation the caesura falls in bar 3 and is decorated by an appoggiatura which delays the proper caesura note C to bar 4. The chord in bar 3 is auxiliary, resulting from an ‘improper harmonization’ of the appoggiatura note B in the melody accompanied by another appoggiatura G in the bass. The evidence for the auxiliary status of this chord – arising from the superposition of two appoggiaturas within the underlying subdominant harmony – is the common tone E♭ in the middle

Manipulations of musical beginnings

Example 4.15 Haydn, String Quartet in F major, Op. 50 No. 5/iv, bars 1–12

voice (bars 3–4). Of these two theoretically possible interpretations, only the latter is supported by perceptual factors. Between bars 4 and 5 of the first phrase the harmonic rhythm, the rate of bass attacks, and the shortest rhythmical values in the melody all shift downward in the metrical hierarchy. This clearly suggests a switch from 6/4, notated in double measures, to 3/4 metre. The same switch happens in the second phrase between bars 10 and 11, even if not marked by any caesura. As a result, both phrases are asymmetrical (4+2) and have an irregular length of six measures. If the change of metre is taken into account, however, each of them counts as four measures long. Another example of a perceived metrical change not unequivocally endorsed by a caesura occurs in the finale of Op. 50 No. 5 (Example 4.15). At the beginning of the theme every composed measure embraces two notated measures, each of them filled with a homogeneous motivic substance. This suggests that the metre is 12/8 as a tripled version of alla breve.36 Even in this composed metre, harmonic changes are very slow because they take place only every second bar. In the notation, one harmony covers four measures, which make up a perfect incise. Although this incise is clearly separated from the next one, the location of the caesura is not clear because

103

104

da n u ta m i r k a

Example 4.16 Haydn, String Quartet in E flat major, Op. 50 No. 3/iv, bars 1–12

no harmonic changes take place within the incises. The harmonic rhythm is drastically accelerated in bars 9–12, where harmony changes with every notated measure. At the same time the size of melodic units is reduced to one measure. The unit of bars 1–2, which formed half of the perfect incise in 12/8, corresponds to twice the shorter unit of bar 9 as an imperfect incise in 6/8. This last metre, reflected in notation, is ultimately confirmed by the caesura in bar 12. Again, the change of metre in the course of the phrase results in its irregular length of twelve measures and in the asymmetry of its component elements (8+4) apparent in the notation. If the change of metre is considered, the phrase turns out to be eight measures long and perfectly symmetrical, containing four measures of 12/8 and four measures of 6/8. It is precisely the outer irregularity and asymmetry of phrase structure that most eloquently speaks for the change of metre in the two above examples. When phrase structure is regular, the change of Taktteile may be less evident. For example, consider the theme of the finale of Op. 50 No. 3, shown in Example 4.16. At the beginning of the theme (bars 1–4) one harmony covers two notated measures and delineates the size of incises, thus suggesting double measures. This suggestion is confirmed by the Absatz which falls in bar 11 but reaches over the bar line to bar 12 containing the resolution of the

Manipulations of musical beginnings

appoggiatura. This means that the phrase starts and ends in 2/2 metre. But does it stay in this metre the whole time? From bar 5 on harmony changes every notated bar, and the size of motives, indicated by slurs, is contracted accordingly. This might suggest a switch to 2/4, the metre Haydn indicates in the time signature. Even so, this change does not cause any outer asymmetry. One perfect incise in 2/4 (bars 5–6) equals the length of one imperfect incise in 2/2 (bars 1–2). Together with its repetition in bars 7–8, it takes the same amount of time as the repetition of the imperfect incise of bars 1–2 in bars 3–4. But a reading of bars 5–8 in 2/2 metre is possible as well, since the caesuras in bars 5 and 7 are not unequivocal. If the only caesuras fall in bars 6 and 8, the melodic units of bars 5–6 and 7–8 can be considered as imperfect incises in 2/2.37 The acceleration of harmonic rhythm is in itself no firm proof of a metrical change. It acquires greater weight from bar 9, where the harmonic rhythm is even more accelerated. Although one might not be inclined to assume that the metre changes twice in the course of this phrase from 2/2 to 2/4 and from 2/4 to 2/8 – in the late eighteenth century this last metre is ‘not in use’ even according to the conservative Kirnberger38 – at least one change from 2/2 to 2/4 is likely to have taken place because the harmonic rhythm in bars 9–10 is uncharacteristically fast for alla breve.

Playing with metre The manipulations of metre and phrase structure analysed in this essay stand out in the repertory of late eighteenth-century music as examples of the highest sophistication. The most obvious difference between them and the metrical changes in k387 and k421, discussed by Allanbrook and Maurer Zenck, is that they serve a different purpose. Rather than enhancing the articulation of the formal structure, they obscure this structure by yielding asymmetrical or otherwise hybrid phrases. Furthermore, metrical changes observed in these examples display no consistent correlation with changes of topics. The only example in which such correlation can be clearly observed is Op. 64 No. 2 (Example 4.6), where the change from 2/4 and D major to 2/2 and B minor results in a twist from contredanse to ‘exalted march’. The change of metre and key is thus semanticized, inviting the listener to develop expressive associations along the lines of eighteenth-century conventions. Even so, after the lighthearted contredanse, the ‘exalted march’ does not really evoke an exalted expression, as it otherwise would, if firmly assigned to an entire theme or section. Rather, it appears like a buffoon in eighteenth-century opera, whose

105

106

da n u ta m i r k a

exaggerated seriousness is one aspect of his overall clownery. Somewhat less definite semanticization of the metrical change takes place in the example from the first movement of Op. 64 No. 6 (Example 4.5). The white-note rhythm and the chordal texture in 2/2 suggest a chorale, whereas short chords and dotted rhythms in 4/4 arouse associations with the march. In the slow movements of Op. 50 No. 1 (Example 4.3) and Op. 50 No. 6 (Example 4.4) the original topic of siciliano is dissolved after the change of metre, when its characteristic accompaniment and rhythm disappear.39 In other examples the original topic does not seem to change. In the finales of Op. 55 No. 3 (Example 4.7) and Op. 50 No. 3 (Example 4.16) it can be identified as a quick contredanse in, respectively, triple and duple metre; in Op. 50 No. 5 (Example 4.15) it is a tarantella. In the minuets of k515 (Example 4.10), k614 (Example 4.11) and Op. 64 No. 6 (Example 4.14) the topic is determined by the movement type.40 In these last examples the introduction of the church style, customarily associated with 6/4 as tripled alla breve, is not only not observed but would even be improper. The composer’s main aim lies apparently in playing with metre by shifting the gear of Taktteile. The fact that changes of metre do not always go together with changes of topics testifies to the new status of metre in the late eighteenth century. In the Affektenlehre of the early part of the century, every metre was ascribed a unique affect, which belonged to its very nature and defined its expressive quality by association with a specific musical genre or style.41 The high or low status of an affect was indicated by the rhythmical values of Taktteile. Consequently, changes of the metrical level of Taktteile resulted in changes of Affekte. Toward the end of the eighteenth century these associations between metres and Affekte are progressively dissolved. To be sure, they continue to be described by conservative theorists, most notably by Kirnberger, and form an important part of the musical tradition, yet they are increasingly ignored by composers. As a result of this process – observed with disapproval by some eighteenth-century authors, with sympathy by others – metre emerges gradually as an independent dimension of musical compositions orthogonal to expression.42 This new status of metre is reflected in Koch’s writings. For Koch the rhythmical values of Taktteile do not matter because metre is nothing more than a grid ordering the temporal course of a piece.43 Clearly, topics are determined not only by metre but, above all, by the character of musical material. If this character does not change, the play with metre remains a purely intellectual exercise appealing to the listener’s abstract sense of humour rather than to his expressive associations. Indeed, as suggested in the course of the analyses, the above examples represent complex games between composers and listeners played upon the compositional rules of the time.

Manipulations of musical beginnings

To be sure, these games are not limited to metrical tricks at the beginnings of movements. Charles Rosen’s general remark about the beginnings of Haydn’s compositions, that ‘they express an immediate conflict, and the full play and resolution of the conflict is the work’,44 proves particularly illuminating in reference to beginnings containing a conflict of metres. Such metrical manipulations are never left without consequences by the composers (not only by Haydn, for that matter, but also by Mozart, even if this last introduces them far more seldom). Rather, they form announcements of more or less elaborated metrical strategies realized throughout the rest of the movements, during which the same conflict is usually exposed more clearly. While the analysis of the further course of strategies announced by the earlier analysed examples goes beyond the scope of this essay, one general observation concerning such an undertaking may be made: although metrical strategies can be analysed for their own sake, they are not isolated from other aspects of eighteenth-century compositions but bear some relation to the musical form on the one hand and its topical decorum on the other. To disclose the nature of this relation is vital for their understanding.

Notes 1 In the recent theory of metrical dissonance by Harald Krebs, Fantasy Pieces: Metrical Dissonance in the Music of Robert Schumann (New York: Oxford University Press, 1999), these phenomena are called, respectively, ‘grouping dissonance’ and ‘displacement dissonance’. The connection between metrical dissonance and the eighteenth-century concept of imbroglio was drawn by Floyd K. Grave, ‘Metrical Dissonance in Haydn’, Journal of Musicology 13 (1995), 168–202. The term ‘hemiola’ did not exist in the late eighteenth century. 2 Joseph Riepel, Anfangsgr¨unde zur musicalischen Setzkunst, 5 vols., vol. 1: De Rhythmopoeia oder von der Tactordnung (Regensburg and Vienna: Emerich Felix Bader, 1752); Heinrich Christoph Koch, Musikalisches Lexikon (Frankfurt-amMain: August Hermann der J¨ungere, 1802; reprinted Hildesheim: Olms, 1964). 3 Although equivalent English terms exist, in this essay I retain the original German terms for ending formulas. It is worth noting that, in eighteenth-century German terminology, an ending formula and a section closed with such a formula are denoted by the same word. Absatz refers to an internal phrase and its ending. Einschnitt or incise denotes a component part of a phrase and its ending formula. Kadenz is an ending formula of a closing phrase (Schlusssatz) and hence of an entire period (Periode). 4 This rule is exposed, among others, in Johann Mattheson, Der vollkommene Capellmeister (Hamburg: Herold, 1739); Riepel, De Rhythmopoeia; Friedrich Wilhelm Marpurg, Kritische Briefe u¨ ber die Tonkunst, 2 vols. (Berlin: Friedrich

107

108

da n u ta m i r k a

5 6

7 8 9 10

11 12

13

14

Wilhelm Birnstiel, 1759–1764; reprinted Hildesheim: Olms, 1974); Johann Philipp Kirnberger, Die Kunst des reinen Satzes, vol. 2 part 1 (Berlin and K¨onigsberg: Decker und Hartung, 1776); included in Kirnberger, The Art of Strict Musical Composition, trans. David Beach and Jurgen Thym (New Haven: Yale University Press, 1982); Heinrich Christoph Koch, Versuch einer Anleitung zur Composition, vol. 2 (Leipzig: Adam Friedrich B¨ohme, 1787; reprinted Hildesheim: Olms, 1969); partly included in Koch, Introductory Essay on Composition, trans. Nancy Kovaleff Baker (New Haven: Yale University Press, 1983). The term Taktteile, as well as Taktglieder, referring to subdivisions of Taktteile, is characteristic of Marpurg and Koch. Other eighteenth-century authors refer to Taktteile with a number of equivalent terms such as Zeiten, Zahlzeiten, Hauptzeiten or Schl¨age. For the sake of uniformity, throughout this chapter I follow Koch’s terminology. This also concerns terminology related to melodic sections and ending formulas, discussed in the previous note. Koch, Versuch, vol. 2, 333. In this passage Koch retraces the argument used earlier by Marpurg, Kritische Briefe, vol. 1, 108. A longer example of a similar phenomenon follows on pages 53–4 of Riepel’s treatise. Locations of caesuras in sections of 4/4 and 2/2 metre correspond to those in Example 4.1. Koch, Introductory Essay, 162–3. Ibid., 19. Ibid., 162. While Riepel and Koch describe the counting of measures by the composer, I deal, instead, with the counting of measures by listeners. In the late eighteenth century this issue was addressed only by Kirnberger, Art, 408: ‘Just as the ear soon perceives the meter (Takt) in every composition and wants it to be retained for the entire piece, the ear is also soon influenced by the rhythmic organization (Rhythmus) and is always inclined to count the same number of measures for each phrase (Einschnitt); it is actually somewhat offended if this uniformity is broken.’ The fact that Kirnberger writes about ‘the ear’, not about ‘the listener’, suggests that he takes the mechanism of counting to be unconscious. Even so, metric manipulations described below do involve consciousness since they refer to the theoretical knowledge of the listener. Wye J. Allanbrook, Rhythmic Gesture in Mozart: Le Nozze di Figaro and Don Giovanni (Chicago: University of Chicago Press, 1983). Allanbrook, Rhythmic Gesture, 24–5. The state of affairs described by Allanbrook is more complicated than it might seem from this short summary due to the fact that individual time signatures should be realized in different tempos. ¨ Claudia Maurer Zenck, Vom Takt: Uberlegungen zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert (Vienna: B¨ohlau, 2001). Maurer Zenck, Vom Takt, 251–3.

Manipulations of musical beginnings

15 William Caplin, in Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven (New York: Oxford University Press), 48, interprets it as an example of what he calls ‘compression of the continuation phrase’. More properly, one should speak here about compression of measures in the continuation phrase. 16 This principle is already implied by Riepel. As he emphasizes, De Rhythmopoeia, 17, ‘in music motion must be continuously heard, that is, when one or two voices rest, the others must move’ (‘in der Musik muss stehts eine Bewegung vernommen werden, das ist: wenn eine oder zwei Stimmen ruhen, so m¨ussen sich die u¨ brige r¨uhren’). In examples with which Riepel illustrates his remark the regular motion takes place on the metrical level corresponding either with Taktteile or Taktglieder. Because the latter are subdivisions of the former, all Taktteile are articulated by attacks in either case. 17 In the eighteenth century Metrum was traditionally related to poetic feet (F¨usse) and their musical equivalents (Klangf¨usse). Continuity of Metrum was considered necessary to secure the aesthetic unity of a composition. Koch follows this tradition but reinterprets the idea of continuity. Rather than repetitions of a given foot, he describes it as a consistent articulation of a given metrical level. 18 Koch, Introductory Essay, 70. I have amended the translation by retaining the original term Metrum, which Nancy Kovaleff Baker renders as ‘metre’, thus blurring the distinction drawn between Metrum and Takt in eighteenth-century German music theory. 19 This is confirmed by Koch, Versuch, vol. 2, 294, who shows an example containing four crotchets in every notated measure. In the commentary he points out that these rhythmical values can be interpreted either as Taktteile of compound 4/4 metre or as Taktglieder of simple 2/2 metre. 20 See Koch, Introductory Essay, 14. Kovaleff Baker translates vollkommener Einschnitt as ‘complete incise’. This translation is not only linguistically imprecise but it also obliterates the fact that, for Koch, all incises are by definition incomplete (unvollst¨andig). 21 This is why the otherwise valuable analysis by William Rothstein, Phrase Rhythm in Tonal Music (New York: Schirmer Books, 1989), 170, remains incomplete. 22 Koch, Versuch, vol. 2, 305–6. 23 Maurer Zenck, Vom Takt, 43–4. 24 Joseph Riepel, Anfangsgr¨unde zur musicalischen Setzkunst, vol. 4: Erl¨auterung der betr¨uglichen Tonordnung (Augsburg: Johann Jacob Lotter, 1765), 77–8. In De Rhythmopoeia, 66, Riepel shows a passage in alla breve notated in 2/4 without any critical commentary. 25 Koch, Versuch, vol. 2, 302. 26 Ibid., 319. 27 Koch, Introductory Essay, 28. 28 The term ‘elongated upbeat’ was introduced by Rothstein, Phrase Rhythm, 39– 40. In his example from the trio of the minuet from Mozart’s String Quartet

109

110

da n u ta m i r k a

29 30

31 32

33 34

35 36

37

in D major, k575, the elongated upbeat is represented by an unaccompanied melody played by two violins. In the minuets of k515 and k614 the situation is further complicated by the fact that the melody offers no perceptual cues that would allow the listener to infer the grouping of crotchets. Also this grouping must be inferred in retrospect on the basis of the following caesura. Koch, Introductory Essay, 43. In terms of Fred Lerdahl and Ray Jackendoff, A Generative Theory of Tonal Music (Cambridge, MA: The MIT Press, 1983), perceived changes of metre resulting from changes of the metrical level of Taktteile can be described as changes of tactus. Koch, Introductory Essay, 19–34. See Maurer Zenck, Vom Takt, 187. Koch’s example is problematic because the dominant chord is introduced for the first time only on the downbeat. In most examples from the musical literature, the dominant used for the ‘improper harmonization’ of a Grundabsatz is reached already on the upbeat of the preceding bar and repeated on the subsequent downbeat. Because such a repetition was otherwise prohibited in the Harmonielehre of the time – see, for instance, Kirnberger, Art, 307 – it indicates clearly that the second dominant counts as a harmonic appoggiatura to the tonic. Koch, Introductory Essay, 4. In Koch’s theory the distinction between incises (Einschnitte) and phrases (S¨atze) is based upon length. A given melodic section is a phrase if it is at least four measures long. (But length alone does not suffice to determine a phrase’s completeness. While most phrases are complete melodic sections, a phrase can be incomplete if, for instance, it closes with a dissonant sonority. I am grateful to William Rothstein for a thorough critical discussion of this aspect of Koch’s theory.) Although caesuras of phrases (Abs¨atze) are made usually on the dominant (Quintabsatz) or the tonic (Grundabsatz), phrases ending on the subdominant are not entirely uncommon. See Koch, Introductory Essay, 36. See ibid., 47–8. Triple subdivision applies here not to Taktteile but to Taktglieder of the alla breve metre. The resulting 12/8 metre was not known in eighteenth-century theory. For Koch, 12/8 is always a compound metre derived from 4/4. Riepel speaks, likewise, about 12/8 as tripled gemeiner Takt (4/4). Nevertheless, simple 12/8 derived from 2/2 occurs frequently in eighteenth-century compositions. Compare the discussion of Beethoven’s ‘Appassionata’ by Maurer Zenck, Vom Takt, 47–8. These caesuras are secured by repetition. As observed independently by Maury Yeston, The Stratification of Musical Rhythm (New Haven: Yale University Press, 1976), 15, and Wolfgang Budday, Grundlagen musikalischer Formen der Wiener Klassik: An Hand der zeitgen¨ossischen Theorie von Joseph Riepel und Heinrich Christoph Koch dargestellt an Menuetten und Sonatens¨atzen (1750–1790) (Kassel: B¨arenreiter, 1983), 37, repetitions unequivocally delineate the size of melodic

Manipulations of musical beginnings

38 39

40

41 42

43 44

units even in the absence of pauses or longer rhythmical values, which otherwise make caesuras more noticeable. Kirnberger, Art, 387. Apparently, siciliano could be either in simple or in compound 6/8 metre. Accents to be observed in Examples 4.3 and 4.4 seem characteristic of this topic. Compare the theme of the first movement of k331, in compound 6/8 metre, analysed by Wye Allanbrook elsewhere in this volume. Allanbrook offers a more specific explanation of the function performed by the two sf in bars 4 and 7. Leonard Ratner, in Classic Music: Expression, Form, and Style (New York: Schirmer Books, 1980), 9, defines types as topics that ‘appear as fully worked-out pieces’. For a useful survey, see Allanbrook, Rhythmic Gesture, 13–23. Directly related to Affekte was the manner of execution and the tempo proper for a given metre: the so-called tempo giusto. For more information about the tradition of tempo giusto and its gradual dissolution in the course of the eighteenth century, see Maurer Zenck, Vom Takt, 14–16. See Koch, Versuch, vol. 2, 291–3, 311–13. Charles Rosen, The Classical Style: Haydn, Mozart, Beethoven (New York: Norton, 1971), 120.

111

5

National metrical types in music of the eighteenth and early nineteenth centuries w illiam rothstein

The first metric organization . . . which seems to coincide with that of the Italians, is [also] preferred in France. That in (Q) has from time immemorial been at home in Germany, Hungary, Croatia, Moravia and so on; that in (R) in Poland. Isn’t it foolish to recast the ears and nerves of entire kingdoms and to try to reduce them to a single rule? Joseph Riepel1 Perhaps it is the same with rhythm as with so many things about which it is impossible to come to an understanding, because different meanings are given to the same word. Camille Saint-Sa¨ens2

I Rhythm and metre are among the chief means of communication in music, but communication about rhythm and metre is fraught with difficulty. Communication between North American and European scholars, for example, is made vastly more difficult by the different theoretical traditions from which they spring. Continental musicologists are often imprinted, whether they realize it or not, with the legacy of Hugo Riemann, for whom the musical phrase was invariably end-accented, with the cadence marking its strongest beat. American theorists tend to follow – again, often unthinkingly – the tradition of Gottfried Weber and Heinrich Schenker, for whom phrases are normally beginning-accented, allowing cadences to fall on relatively weak metrical positions, either a weak beat or a weak bar within a larger metrical unit, the hypermeasure.3 This study grew out of my frustration over this impasse, which seemed to render impossible further progress in the theory of rhythm. Gradually I realized that competing theories of metre correspond to real differences in musical style, but this fact has gone largely unrecognized. I had long been aware that the structure and function of the musical bar changed between the early eighteenth and the early nineteenth centuries,4 but studying nineteenth-century opera has convinced me that this change included a

National metrical types

strongly national component: metrical habits that are rare in German music after 1820 survive undisturbed in music by Italian and French composers. By ‘metrical habits’ I mean, above all, the habits of composers in placing their barlines. Broadly speaking, Italian and French composers were much more likely in the nineteenth century to place cadences on the first beat of a bar, whereas German composers often placed them later. Conversely, phrases in German music were less likely to begin in mid-bar, beginning instead on the downbeat or with a short anacrusis, one-third of a bar or less in length. The rule that a cadence must fall on the strongest beat of a bar was widespread, indeed ubiquitous, in eighteenth-century theory. Exceptions were allowed in certain cases, usually for certain dance types (the polonaise is often mentioned).5 Significantly, the downbeat-cadence rule was not a staple of seventeenth-century theory or practice: weak-beat cadences are easy to find in the music of such composers as Chambonni`eres, Carissimi, Froberger, Buxtehude, Corelli and Purcell, to name a few. Sarabandes, for example, routinely cadenced on the third beat, while gavottes sometimes began on the downbeat and cadenced in mid-bar. Far from being a matter of nature, then, the association of harmonic cadence with the downbeat was a matter of eighteenth-century convention. When in 1721 Johann Mattheson objected to a weak-beat cadence in Friedrich Niedt’s Handleitung zur Variation (1706), he expressed a shift from a late seventeenth-century aesthetic to a fully eighteenth-century one.6 A gradual move away from the eighteenth-century theoretical consensus regarding metre began in Germany in the last quarter of the century, compositional change having begun somewhat earlier. The fracturing of the theoretical consensus was linked to, and perhaps partly caused by, a change in the conception of the 4/4 bar. Although Kirnberger recognized a simple quadruple metre, most eighteenth-century theorists regarded 4/4 as a compound metre, composed of two bars of 2/4 with one barline removed. In most accounts, the two 2/4 bars were precisely equal, so that the third beat in 4/4 bore the same accentual weight as the first beat. At least three theorists departed from this consensus, however: Kirnberger and Vogler in the late 1770s, and T¨urk in the late 1780s.7 To Kirnberger and T¨urk, the compound 4/4 bar (which Kirnberger distinguishes from simple quadruple metre) still contains two 2/4 bars, but the second of these bars is subordinate to the first, so that the third beat – the second ‘downbeat’ – is weaker than the first. Whereas Kirnberger also recognizes the older type of compound 4/4 bar, containing two equal ‘downbeats’, T¨urk does not. Vogler goes still further, calling the third beat in 4/4 an ‘upbeat’ (Aufstreich) even where the

113

114

w illiam rothstein

metre cannot reasonably be mistaken for alla breve.8 These theorists anticipate metrical theories of the early nineteenth century, especially that of Gottfried Weber. Kirnberger and Vogler even developed a modest theory of hypermetre, speaking explicitly of strong and weak bars that relate to each other like strong and weak beats within a single bar.9 The reconception of the 4/4 bar as having only one strongest beat, not two, coincides with a phenomenon noted by Floyd K. Grave: a gradual decrease in the incidence of half-bar displacements in 4/4 metre. Such a decrease is to be expected if the two halves of the bar are perceived as being different in quality. Vogler even ‘corrected’ displacements in Pergolesi’s Stabat mater, so that repetitions of musical themes remain in the same position with respect to the barlines. All such corrections affect movements in 4/4.10 Where the music of Haydn, Mozart and Beethoven is concerned, my perspective is close to Grave’s and is often opposed to that of Claudia Maurer Zenck.11 Maurer Zenck, who follows the metrical theories of Koch – the last important representative of the eighteenth-century consensus – sometimes lapses into pure Riemannism, as when she regards as ‘downbeats’ the thirdbeat cadences in the Andante of Beethoven’s Piano Sonata Op. 79 (1809).12 She even adopts Riemann’s notation, adding a dotted barline before each cadential ‘downbeat’. She also regards Koch’s theories, which fail to recognize any distinction between the two halves of the compound bar, as binding on the music of the three Viennese – so that, for example, mid-bar cadences in the 6/8 finale of Beethoven’s Piano Sonata Op. 28 (1801) are regarded as ‘really’ falling on the downbeats of 3/8 bars, with no hierarchy existing among these bars. First-beat and mid-bar cadences are regarded as equivalent, both marking ‘downbeats’ of equal strength.13 If Maurer Zenck’s analyses are accepted, even the modest hypermetres of Kirnberger and Vogler must be rejected, and the emergence of deeper metrical hierarchies in the writings of Weber, who wrote during Beethoven’s lifetime, cannot be accounted for at all. Although hypermetre is a fascinating topic, in order to keep this study within reasonable limits I will focus on metric phenomena at levels up to the notated bar. Most of my examples concern metres typically described as compound in the eighteenth century: 4/4, 6/8 and 12/8.

II That change was in the air in the decades surrounding 1800 is suggested by cases in which German composers re-barred either their own music or music

National metrical types

Example 5.1 Mozart, Piano Variations in G major on ‘Unser dummer P¨obel meint’ from Gluck’s Die Pilger von Mekka, k455, Theme

by a predecessor. Mozart’s re-barring of his duet ‘Bei M¨annern welche Liebe f¨uhlen’, from Die Zauberfl¨ote, is only the most celebrated example.14 Perhaps more instructive is Mozart’s uncertainty regarding the proper barring of an aria he had heard in performance but probably never saw in score: ‘Unser dummer P¨obel meint’ (originally ‘Les hommes pieusement’) from Gluck’s Die Pilger von Mekka (originally La rencontre impr´evue); Mozart used a distorted (half-remembered?) version of this aria as the theme of his Piano Variations k455 (see Example 5.1). Mozart’s first instinct was to place Gluck’s cadences on downbeats, as he generally did in his own vocal music. He soon revised the barring, however, to place phrase-beginnings on downbeats and phrase-endings in mid-bar – which, as it happens, is the way Gluck barred his aria in the first place (see Example 5.2, which includes the original French text).15 Mozart’s revised barring emphasizes the foursquareness of Gluck’s theme, giving it a simplicity well suited to the aria’s German title (‘Our stupid rabble believes’). Soon this style of barring – with phrases beginning on or just before a downbeat, and with cadences falling in mid-bar or later – became closely associated with German melodies im Volkston. Two further cases of re-barring transformed vocal pieces by Haydn and Mozart into the kinds of foursquare ditties against which late nineteenthcentury composers (Wagner) and theorists (Rudolph Westphal and Hugo Riemann) would rebel. When, in 1841, the German nationalist poet August Heinrich Hoffmann von Fallersleben converted Haydn’s Kaiserhymne (‘Gott erhalte Franz den Kaiser’) into the Lied der Deutschen (‘Deutschland, Deutschland u¨ ber alles’), he changed Haydn’s barring to the blunter style that Mozart used for ‘Unser dummer P¨obel meint’ (see Example 5.3). I will call this style of barline placement ‘German barring’.

115

116

w illiam rothstein

Example 5.2 Gluck, La rencontre impr´evue, Aria ‘Les hommes pieusement’, bars 1–8

Beethoven acted similarly to Fallersleben when he transformed Leporello’s opening solo in Don Giovanni, ‘Notte e giorno faticar’, into Variation 22 of his Diabelli Variations (Example 5.4). Beethoven’s humour is broader and more Germanic than Mozart’s. His barring corresponds not to the rhythm of a pre-existent Italian text but to the abstracted pitch motion of Mozart’s melody: two bars of the tonic note, then two bars of the second degree and so on.16 German barring often causes musical bars to coincide with the textless rhythms of pitch events; Mozart’s barring of ‘Notte e giorno faticar’ syncopates these rhythms. A style of barring in which cadences are always placed just to the right of the barline, and in which phrases begin at least half a bar to the left of the barline, will here be termed ‘Italian barring’. In this study, a cadence is located at the arrival of its bass note. The identifying feature of German barring is the regular placement of cadences on relatively weak beats; a lack of long upbeats is a secondary feature, necessary but not sufficient for a pattern of barring to be identified as German. The identifying characteristic of Italian barring is the frequent use of upbeats lasting half a bar or longer; the regular placement of cadences on downbeats is a secondary feature, necessary but not sufficient for a pattern of barring to be identified as Italian. If a piece of music has a simple notated metre (duple or triple), has cadences placed on downbeats and is without long upbeats its barring will be described as ‘neutral’. Neutral barring is extremely common

National metrical types

(a)

(b)

Example 5.3 (a) Haydn, Kaiserhymne (b) Fallersleben, Lied der Deutschen

in vocal music of the late eighteenth century; it is almost as common in instrumental music of that time. In 4/4, 6/8 or 12/8, the second half of the bar must be felt as weaker than the first (as described by Kirnberger, Vogler and T¨urk) for a cadence on the first beat of the second half-bar to be described as ‘German’. (I refer here to the third crotchet in 4/4, the fourth quaver in 6/8, and the seventh quaver in 12/8.) Compound metres will be considered in considerable detail in Part V of this study.

117

118

w illiam rothstein

(a)

(b)

Example 5.4 (a) Mozart, Don Giovanni, No. 1 Introduzione, Leporello, bars 10–16 (b) Beethoven, Thirty-Three Variations on a Waltz by Diabelli, Op. 120, Variation 22, bars 1–6

Writing in 1880, Westphal advised that a piece such as ‘Notte e giorno faticar’ should be counted 1 – 2 – 3 – 4, with the musical bar conceived not as Mozart wrote it but starting with an upbeat and ending with a downbeat. (Notice that Westphal’s bar is twice as long as Mozart’s.)17 Although Westphal believed that in some contexts ‘1’ might denote a strong beat and ‘2’ a weak one, his successor Riemann developed a theory of metre in which, as for Momigny seventy-five years earlier, the ‘true’ bar necessarily begins with its weakest and ends with its strongest beat. Barlines, in short, do not show where bars begin and end, only where their strongest accent falls. Example 5.5 gives relevant illustrations from writings by Momigny and Riemann. Many composers of the late eighteenth and early nineteenth centuries failed to follow Riemann’s theory when barring their music. German composers were among the worst offenders, but Chopin was equally unruly. In 1884 Riemann took aim at one of Chopin’s most famous works, the Nocturne in E flat major, Op. 9 No. 2 (Example 5.6): The entire Nocturne has the barlines in the wrong place; instead of cutting through the middle of the motives, as here, the barlines frame the motives. That such barlines, incorrect according to the theory developed here, are a frequent occurrence should not be taken by anyone as evidence against the theory itself; nor should anyone reproach me for giving barlines a meaning they have not had traditionally. As proof

(a)

(b)

Example 5.5 Illustrations of musical metre from (a) J´erˆome-Joseph de Momigny, Le seule vraie th´eorie de la musique (Paris, 1821), 115, and (b) Hugo Riemann, Elementar-Schulbuch der Harmonielehre (Berlin: Max Hesse, 1906), 180, transcribed after William Caplin, ‘Hugo Riemann’s Theory of “Dynamic Shading”: A Theory of Musical Meter?’, Theoria 1 (1985), 24

120

w illiam rothstein

(a)

(b)

Example 5.6 (a) Riemann, Musikalische Dynamik und Agogik, 240 (b) Chopin, Nocturne in E flat major, Op. 9 No. 2, bars 1–11

I point to Koch’s . . . Introduction to Musical Composition . . . from which one can learn more about metre, rhythm and phrasing than from all the subsequent literature put together. There one will . . . find the clearest formulation of the meaning of the barline.18

Riemann does not frame Chopin’s motives with his barlines, but he frames the issue with admirable clarity. He contrasts two styles of barline placement: one, used by Chopin, in which the compound bar coincides, more or less, with the motive or small phrase; another, his own, in which barlines indicate the point of greatest emphasis or weight within the motive.19 Chopin

National metrical types

Example 5.7 Mattheson, Der vollkommene Capellmeister, 147: examples of incorrect and correct barring, composer unknown

often used German barring, especially when writing in compound metres. Riemann, following Westphal, shows a strong preference for Italian barring, because both theorists believed that the barline exists to show where the principal accents fall within a motive or phrase.20 Hence Riemann’s notorious re-barring of music by Beethoven, Chopin, and others, a practice which Schenker’s pupil Oswald Jonas memorably characterized as ‘literally chas[ing] the accents with his barlines’.21 Riemann’s reference to Koch is interesting because it explicitly links Riemann to what I have called the eighteenth-century consensus – the idea that there can be no stronger beat within the notated bar than that on which the cadence falls.22 In the nineteenth century, this principle was followed more often by French and Italian composers than by German ones.

III In criticizing Chopin’s barlines, Riemann surely knew that he was doing much as Koch had done a century earlier, and Mattheson half a century before that (albeit in reference to simple, not compound, metres). Mattheson shows a melody by ‘an otherwise good master’ (Example 5.7) in which the cadence falls on the fourth crotchet of bar 2 (the metre is alla breve). This is incorrect, says Mattheson; he shows the correct barring immediately beneath. The composer’s error is, in effect, the error that Riemann imputes to Chopin (‘instead of cutting through the middle of the motive . . . the barlines frame the motives’), except that here we might better speak of incises (Einschnitte), using Koch’s terminology, rather than ‘motives’, a nineteenthcentury term. In Koch’s example (Example 5.8), the beginning of a vocal melody is given first without any barlines, then with incorrectly placed barlines and finally with the correct barring. The composer of this melody is Georg Gebel, one

121

122

w illiam rothstein

Example 5.8 Koch, Versuch, vol. 2, 310–11: examples of incorrect and correct barring, melody by Georg Gebel

of Koch’s predecessors as Kapellmeister in Rudolstadt.23 Gebel died in 1753, having been Kapellmeister since 1750, so his melody was at least thirtyfour years old when Koch used it as a model. Like Mattheson, Koch points out that the incorrect barring places the caesura of the phrase on a weak beat.24 Through their examples, Mattheson and Koch assert that what I have called Italian barring is how one ought to notate music, at least in simple metres. (Remember that Italian barring entails not only downbeat cadences but long anacruses as well.) Metre is neither culturally determined nor culturally circumscribed; it is simply a matter of correctness. That ‘otherwise good masters’ did not always bar their music ‘correctly’ indicates that practice was not as straightforward as theory, even in Mattheson’s day. So to what extent was the theory followed by Koch’s contemporaries during the century’s final quarter? In music composed to Italian and French texts, the rule regarding the placement of cadences still held (Example 5.2 is an exception). In Italian operas by Johann Christian Bach and Mozart, for example, the final accent in each verse – the so-called accento comune – falls almost invariably on the first beat of the notated bar, whether that bar is simple or compound. (The Appendix lists all of the exceptions to this rule that I have

National metrical types

found in Mozart’s operas.) This practice continues without change into the nineteenth century.25 By contrast, composers before 1760 freely placed the accento comune on either of the two strong beats within a compound bar. This practice is evidence for one of my main contentions: that the status of the compound bar changed in the late eighteenth century. If only the first ‘good’ beat in a 4/4 bar was good enough for the accento comune, then the two halves of the bar were no longer regarded as equal. In other words, Koch’s conception of the compound bar – with both halves equally weighted – no longer reflected musical reality. The greatest change in metrical practice occurred in German music, both vocal and instrumental. This change seems to have been connected not only to the changing status of the compound bar, but also to changing styles in the setting of German texts. In German, accentual patterns on a small scale are relatively fixed, but the main accent in a verse is based on meaning; there is no accento comune. Hence there is often no inherent need for a vocal phrase to move towards an accented cadence. Italianate setting of German remains the norm for the first two-thirds of the century; a good example is the final chorus from Graun’s Der Tod Jesu (Example 5.9). Nevertheless, the metrically weak cadences that Mattheson and Koch condemned are in fact appropriate to the setting of much German verse. In Example 5.10, I have set Mattheson’s ‘incorrectly’ barred melody to the chorale text ‘Was Gott tut, das ist wohlgetan’. Only a short upbeat (one quaver) is needed to make the setting plausible, although it is more plausible in 4/4 than in Mattheson’s alla breve. I am not the first to notice a difference in late eighteenth-century settings of German and Italian. In a 1978 article, Friedrich Lippmann distinguishes melodic rhythms suitable to the setting of poetry in each language.26 Although some rhythms fit both languages, one metre common to German verse, iambic tetrameter, has no Italian counterpart, and hence corresponds to no melodic rhythm commonly found in Italian opera. Furthermore, according to Lippmann, iambic tetrameter often causes cadences to fall in mid-bar. Lippmann cites Beethoven’s ‘Ode to Joy’, the Lutheran chorale, and German song generally as evidence of a specifically German rhythmic style, one that finds no parallel (Lippmann claims) even in Mozart’s German operas. J. S. Bach’s chorale harmonizations support Lippmann’s thesis. Even if we accept, in principle, Robert Marshall’s conclusion that there is no intrinsic difference between the first and third beats in Bach’s 4/4 bars,27 that leaves chorales such as shown in Example 5.11, in which Bach places cadences on the fourth beat, in apparent violation of Mattheson’s rule. (The first

123

124

w illiam rothstein

Example 5.9 Graun, Der Tod Jesu, final chorus, bars 1–15, string parts omitted

National metrical types

Example 5.10 Mattheson’s ‘incorrect’ barring from Example 5.7 set to a chorale text

weak-beat cadence, in bar 2, falls appropriately on the word Schwachen (‘weak ones’).) In fact, Mattheson was well aware of stylistic differences: he specifically permits weak-beat cadences in ‘a few choraic and melismatic things’, including dances (his example is a minuet), chorales and strophic songs. Mattheson specifies that such cadences must occur on the final beat of the bar, and that their purpose is to secure ‘a special uniformity’ in phrase rhythm.28 Example 5.12 gives the beginning of Bach’s setting of the text given in Example 5.10. Cadences fall on the third beat of the 4/4 bar, with the strongest syllable falling on the preceding downbeat. Given the textual accents, it is questionable whether the third beat is equal in weight to the first. Even if the first and third beats are understood as metrically equal, the third beat is rhythmically weak by virtue of its weaker syllable, at least in the cadential bars. One might reasonably expect this rhythmic weakness to be reflected in performance. Strophic songs in German were only beginning to make a comeback in Mattheson’s time. An impressive recent study by William H. Youngren gives a detailed account of the relevant history, with examples by numerous composers from the 1730s through the 1780s.29 Especially interesting are settings of the same poem by different composers. Often not only musical metres but styles of barring differ, indicating again that mid-century Germany was a time of flux in metrical matters. Youngren reproduces four settings of Friedrich von Hagedorn’s poem An den Schlaf; three of these, all from the early 1740s, are given here as Example 5.13. A composer named de Giovannini set the poem in a 4/4 metre that is clearly compound (twice 2/4). Each phrase starts on an upbeat and cadences on the (strong) third beat. Some of the poem’s most accented syllables, including Freu-den, nei-den, and wei-den, are relegated to the weaker second beat so that cadences may be metrically strong. Telemann’s setting of the poem is similar in that cadences fall on strong beats – in this case, the notated downbeat – with phrases beginning in mid-bar; Telemann’s slower harmonic rhythm suggests some differentiation between beats 1 and 3. A setting by Johann Valentin G¨orner, in a 4/4 metre that often resembles alla

125

126

w illiam rothstein

Example 5.11 J. S. Bach, final chorale from Cantata No. 78, ‘Jesu, der du meine Seele’, bwv78

National metrical types

Example 5.12 J. S. Bach, final chorale from Cantata No. 99, ‘Was Gott tut, das ist wohlgetan’, bwv99

breve, is the only one to begin on a downbeat and to set words such as Freu-den according to their natural accentuation. Not surprisingly, historians have regarded G¨orner’s setting as the most galant of the three.30 It looks forward to the German 4/4 metre of the late eighteenth century – the type of 4/4 described by Kirnberger, Vogler and T¨urk.

IV Example 5.14 shows the opening of Mozart’s Singspiel Zaide (1780), a work apparently overlooked by Lippmann. This introductory chorus begins with a solo sung by a low-born European slave; accordingly, Mozart affects the style of German peasant music, rather like J. S. Bach in his ‘Peasant’ Cantata bwv212. The two halves of Mozart’s bars are not strongly differentiated, but caesuras occur exclusively, and regularly, in the second half of the bar. Koch would probably have considered each two-bar phrase as a four-bar phrase in 2/4 metre; the caesura would thus fall on the downbeat of the fourth bar.

127

128

w illiam rothstein

(a)

(b)

(c)

Example 5.13 Three settings of An den Schlaf (Friedrich von Hagedorn): (a) [first name unknown] de Giovannini (b) Georg Philipp Telemann (c) Johann Valentin G¨orner

National metrical types

Example 5.14 Mozart, Zaide, k344, No. 1, Coro ‘Br¨uder, laßt uns lustig sein’, bars 1–20

This is a plausible hearing, but Mozart very rarely sets Italian in this way. The closest example I have found is from another unfinished work, the comic opera L’oca del Cairo (Example 5.15). In this duet, two servants are trying, without much success, to hatch a rescue plot. Mozart underscores the pompous incompetence of the basso, a Haushofmeister, by setting his first two verses to a march-like theme that partly contradicts the accentuation of

129

130

w illiam rothstein

Example 5.15 Mozart, L’oca del Cairo, k422, Duetto ‘Ho un pensiero’, bars 1–17, accenti comuni underlined

the text. The next two verses, which are repeated, are barred and accented in Mozart’s usual Italian manner. In effect, the first couplet is set in German 4/4, with phrases beginning on the notated downbeat, which also carries the main musical (though not textual) accent. Phrases in the second couplet begin with long upbeats and move toward downbeats; in other words, their barring is Italian.

National metrical types

(a)

(b)

(c)

Example 5.16 C. P. E. Bach: (a) Sonata in B flat major, h282/ii, bars 1–19 (b) Rondo in E major, h265, bars 1–11 (c) same, final four bars

Example 5.16 gives excerpts from two late keyboard pieces by C. P. E. Bach, an important figure in the development of German metre. The first excerpt (Example 5.16a) opens the slow movement of a sonata published in 1785. Here, changes of harmony on the second beat may cause the listener to hear the first beat as an upbeat and the second beat as the downbeat. In Riemann’s

131

132

w illiam rothstein

Example 5.17 Couperin, L’Art de toucher le clavecin (Paris, 1716), 28

terms, the barlines incorrectly frame the motives rather than indicating their ‘heavy’ points (Schwerpunkte). At bar 16, a strong-beat cadence with Tacterstickung (the suppression of a bar) makes Bach’s metre clear to the listener. Throughout his long career, Emanuel Bach often used a contrast between weak-beat and strong-beat cadences to create a hierarchy of cadential functions in his instrumental music. Strong-beat cadences end the main periods of a movement, as they do in this Largo; weak-beat cadences mark lesser divisions.31 This is, however, not so in the E major Rondo h265, published in 1781 (Example 5.16b). This piece is in compound 4/4; four notated bars clearly stand for eight bars of 2/4. Nevertheless, every caesura falls on a weak beat, the fourth beat of the notated bar. Each 4/4 bar forms a complete incise; the barlines frame these incises, in the manner condemned by Riemann. In the entire Rondo, not a single formal cadence falls on a strong beat.32 It was not always thus with the keyboard rondo. Franc¸ois Couperin, who popularized the genre, was such a principled adherent of upbeat beginnings that, as Ido Abravaya has recently noted, even Couperin’s keyboard exercises start with long upbeats (see Example 5.17).33 Example 5.18 gives the beginnings of two Couperin rondeaux, both in compound metres. Couperin always places his caesuras and cadences on the first half of the compound

National metrical types

(a)

(b)

Example 5.18 Couperin, beginnings of two rondeaux: (a) ‘La S´eduisante’ from Ordre No. 9 (b) ‘Les Bergeries’ from Ordre No. 6

bar. Consequently, he begins many pieces with upbeats of half a bar or more. In ‘La S´eduisante’, Couperin’s barlines clearly indicate the Schwerpunkte of his incises and phrases. In ‘Les Bergeries’, the root-position tonic harmony at the beginning is placed on the upbeat, not the downbeat; the repetition of a half-bar motive in bars 2–3 goes from upbeat to downbeat, not the reverse. Such are the lengths to which Couperin goes to place his cadences on downbeats. Example 5.19 shows a well-known piece in French style by J. S. Bach. Today, most casual listeners probably hear this piece as beginning on a

133

134

w illiam rothstein

Example 5.19 J. S. Bach, Badinerie from Orchestral Suite No. 2 in B minor, bwv1067, bars 1–16

downbeat. From the perspective of German metre, the opening tonic arpeggio, four beats long, would fit perfectly within a 4/4 bar. Bach’s apparent intent, however, is that the piece begins not only with an upbeat but with a long upbeat of three crotchets; the strongest downbeats are those of evennumbered bars. This pattern, which is clearest in the bass, becomes gradually more audible as the piece proceeds.

V Let us return to the question of the compound bar. I have already distinguished three types of barring: German, Italian and neutral. I also distinguish three types of compound metre. The first type is Couperin’s, in which caesuras and cadences fall exclusively on the first half of the bar; incises and phrases begin with long upbeats. I will call this ‘French compound metre’; it corresponds to Italian barring and is possible only where phrases consistently occupy an even number of half-bars. Bach’s ‘Badinerie’ could have been notated in a French 4/4, starting on the second crotchet.

National metrical types

French compound metre seems to be relatively rare in the second half of the eighteenth century. I have not found it, for example, in Gluck’s operas, although I have not examined all of them. The second type is ‘German compound metre’, which corresponds to German barring. Here caesuras and cadences fall mostly, if not exclusively, in the second half of the bar. Upbeats, where present, are rarely more than one beat in length. Consequently, complete incises are more or less framed by barlines. Like the French type, German compound metre is possible only where phrases and incises contain an even number of half-bars. Odd lengths are possible, however, if the composer compensates by using Tacterstickung, the suppression of a half-bar at a cadence. The third type is ‘Italian compound metre’, which has no precise correlate but might be related to neutral barring. Italian compound metre is the type described by Marpurg, Koch, and other theorists of the eighteenth-century consensus; it is common in both Italian and German music of the early and mid eighteenth century. Example 5.20 illustrates Italian compound metre with one of the passages in Pergolesi’s Stabat mater to which Vogler objected.34 The words ‘et afflicta’ are repeated after three half-bars, the words ‘mater unigeniti’ after five. There is no regular alternation of half-bar units, and thus no possibility of perceiving an alternation of strong and weak half-bars. Pergolesi’s compound 4/4 is best regarded as a 2/4 metre in which 2/4 bars may be grouped in any way the composer desires. Caesuras and cadences fall in either half of the compound bar; there is little or no perceived difference between the halves. Handel generally used Italian compound metre in his operas, but his oratorio Messiah contains many passages that suggest the German type. Example 5.21 gives three excerpts from the aria ‘O thou that tellest good tidings to Zion’. The metre is mostly compound 6/8. In the first excerpt (Example 5.21a), exchanges between voice and violins suggest, in addition, a doubly compound metre of 12/8. (Handel’s autograph reinforces this impression by omitting every second barline; the lengthened barlines in the example show those he did write in.) Notice in bar 18 the setting of ‘mountain’, with a half cadence on the weak syllable. The second excerpt (Example 5.21b), beginning with the words ‘behold your God’, has an even stronger feeling of two beats to the bar; it might be heard in simple duple metre, what Marpurg and Koch would later term ‘mixed’ 6/8 (2/4 subdivided in triplets). The final excerpt (Example 5.21c) shows the beginning of the aria’s choral section. Here, regularly spaced entries of the theme reinforce the Schwung of the German 6/8 metre, which is again compound, but with the first half-bar noticeably stronger than the second. Overall, Handel’s treatment of compound metre in this aria seems decidedly more modern

135

136

w illiam rothstein

Example 5.20 Pergolesi, Stabat mater, No. 3, Duetto ‘O quam tristis’, bars 1–12

than Pergolesi’s. There is almost always an audible difference between the first and second halves of the 6/8 bar – enough so that an experienced listener should have no difficulty telling which half of the bar she is hearing at any moment. Wherever this is true, the compound metre is either German or French. Where there is frequent doubt on this point, the compound metre is Italian, or (much the same thing) the metre is not perceived as compound. In the instrumental music of Haydn and Mozart one finds all three types of compound metre, but mostly the Italian and German types. Sometimes one finds more than one type in the same piece – plus, in many cases, passages in simple metre, either alla breve (for pieces notated in 4/4) or mixed 6/8 (for pieces notated in 6/8). The distinction between simple and compound

National metrical types

(a)

(b)

Example 5.21 Handel, Messiah HWV56, Aria ‘O thou that tellest good tidings to Zion’ (a) bars 11–25 (b) bars 53–67 (c) bars 107–17

137

138

w illiam rothstein

(c)

Example 5.21 (cont.)

metres is fundamental in the eighteenth century, and it is thoroughly treated in Maurer Zenck’s book. Unlike Maurer Zenck, however, I do not believe that the two halves of a compound bar are always equal in music of this period. Italian compound metre is common in Haydn’s string quartets up to and including Op. 33, but it is also prominent in some later quartets, such as the first movement of Op. 64 No. 2 (see the analysis by Danuta Mirka

National metrical types

Example 5.22 Haydn, String Quartet in B flat major, Op. 1 No. 1/i, bars 1–24

elsewhere in this volume). German compound metre appears, however, as early as Haydn’s Op. 1 No. 1 (Example 5.22). The style of the first movement is rustic, with suggestions of hunting music (much like Mozart’s ‘Hunt’ Quartet, which is in the same key). All caesuras fall on the second half of the 6/8 bar until the final, two-bar codetta places the new tonic, F major, on the downbeat for the first time. (By ‘downbeat’ I mean exclusively the first beat of the notated bar.) A progression from weak-beat to strong-beat caesuras may also be found in a much later work, the finale of Haydn’s Quartet Op. 71 No. 3 (Example 5.23). The mid-bar caesuras and cadences in bars 1–8 (Example 5.23a) identify the metre as compound, but a strong-weak alternation is

139

140

w illiam rothstein

(a)

(b)

Example 5.23 Haydn, String Quartet in E flat major, Op. 71 No. 3/iv (a) bars 1–14 (b) bars 131–42

National metrical types

suggested at the very beginning, and the half cadence in bar 14 confirms the location of the downbeat. The theme is treated imitatively later in the movement, but its position within the bar remains constant, as in the choral passage by Handel. This further confirms the stability of Haydn’s German 6/8. Only in two codettas, both over tonic pedals, is the theme’s head-motive shifted metrically; one of these codettas is shown in Example 5.23b. The first five bars of the excerpt are in mixed 6/8. Beginning at bar 136, the theme’s head-motive appears twice, shifted so that it ends on a downbeat. The shift in the motive’s metrical position is easy to hear, and it reinforces the feeling of closure for the movement as a whole. Vogler already writes of such metric shifts and their ability to promote closure; David Temperley has recently written of similar phenomena in music by Haydn, Mozart, Beethoven and others.35 Contrasts of metrical type are also found in Mozart’s ‘Haydn’ Quartets. The first movement of the Quartet in D minor, k421 – the subject of a fine analysis by Maurer Zenck – contains a passage that Schoenberg praised as an example of ‘musical prose’.36 In this passage, half-bars are grouped in an irregular manner, indicating Italian compound metre. By contrast, the first movement of the Quartet in G major, k387, is perhaps the most ‘German’ of all Mozart’s first movements. In the first theme (Example 5.24), Mozart consistently accents the third beat, either with appoggiaturas or, in bars 8 and 10, with crescendos that partly counteract the effect of his slurs (which end there). In the instrumental music of Mozart and Beethoven, it is extremely common for the second half of the German compound bar, or the second bar of a two-bar unit, to receive an explicit accent of some kind. The metric units are delineated, in other words, by means other than explicit accents. The metre may be indicated by changes of dynamics (as at the beginning of k387) or by changes of harmony. Thus, in these German metres, there is a sharp distinction between the beginning of a bar (Taktbeginn) and its ‘heaviest’ point (Taktschwerpunkt). This is precisely the distinction that bothered Riemann in Chopin’s Nocturne in E flat major. As Maurer Zenck has noted, it also bothered Carl Dahlhaus in a Beethoven analysis by Diether de la Motte.37 Beethoven’s middle-period quartets are frequently shaped by metric play, sometimes involving contrasts between metric types. Space permits only one example, the Andante con moto from Op. 59 No. 3 (Example 5.25). The first theme (Example 5.25a) is composed in an Italian 2/4 with triplet subdivisions, starting with a single forte bass note that recalls similar beginnings in Baroque music.38 The second theme (Example 5.25b), built over a repeated cadential pattern, is German, whether one considers the metre simple

141

142

w illiam rothstein

Example 5.24 Mozart, String Quartet in G major, k387/i, bars 1–10

or – more likely – compound. The weak metrical status of the cadential tonics in bars 43, 45 and 47 is underscored by Beethoven’s slurs in the inner voices. After this series of relatively weak cadences, the theme ends with a strong-beat cadence at bar 50, marked with the subito piano that Beethoven habitually reserves for large-scale downbeats. German metre is the norm in Schubert’s Lieder; the closer a song is to folk style, the more likely it is that its metre will be German. Schubert’s Wiegenlied, d867 (Example 5.26), is a good example for several reasons. First, each of Schubert’s bars precisely frames one line of verse and one melodic incise; Riemann would have regarded Schubert’s barring as incorrect. Secondly, the third crotchet of each bar is accented with an appoggiatura, much as in Mozart’s k387; again we have the apparent paradox that Taktbeginn and Taktschwerpunkt fall in different places, at least in the vocal line. Thirdly, the poetic metre superficially resembles the Italian quinario, which one finds, for example, in Mozart’s arias ‘Vedrai carino’ and ‘Finch’ han dal vino’ from Don Giovanni, and in ‘Voi, che sapete’ from Le nozze di Figaro. In those arias, however, Mozart bars neutrally, writing in very short bars so that the accento comune always falls on a notated downbeat. In setting German, this is unnecessary. Finally, the melodic rhythm in Schubert’s Wiegenlied

National metrical types

(a)

(b)

Example 5.25 Beethoven, String Quartet in C major, Op. 59 No. 3/ii (a) bars 1–13 (b) bars 40–53

143

144

w illiam rothstein

Example 5.26 Schubert, Wiegenlied, d867, bars 1–16

recalls that of the ‘Allegretto’ from Beethoven’s Seventh Symphony, which is notated in short bars (neutral barring) but which proceeds at a similar tempo. The comparison is suggestive, but I cannot pursue it here. By 1820, Italian compound metre had largely disappeared; the third beat in 4/4 is now definitively weaker than the first beat, although Weber continues to describe 4/4 metre as ‘compound’. The concepts ‘Italian barring’ and ‘French compound metre’ seem to merge, leaving a compound metre that I will call ‘Franco-Italian’. I offer two examples, which could easily be added to. Example 5.27 gives the beginning of Hal´evy’s celebrated aria ‘Rachel, quand du Seigneur’ from La Juive. The 4/4 setting is Franco-Italian, following closely the rhythm of Scribe’s alexandrines. A German setting in 4/4 would have set each hemistich within a single bar, with the Schwerpunkt on the third beat, as in Chopin’s Nocturne (or his Pr´elude in C minor). As we have seen, a strong opening tonic is no reason for a Franco-Italian composer to begin on a downbeat. Many opera-lovers would be surprised to see that the ‘Anvil Chorus’ from Verdi’s Il trovatore begins on an upbeat (see Example 5.28). Only when the orchestra reaches the relative major (bar 13) does the notated 4/4 metre correspond to what, I think, most

National metrical types

Example 5.27 Hal´evy, La Juive, Aria ‘Rachel, quand du Seigneur’, bars 1–28, tonic accents underlined

listeners hear. When the chorus enters, still in G major, their tonic harmonies also fall on upbeats. The poetic metre is quinario doppio, with the accento comune falling on the penultimate syllable in each five-syllable hemistich.

VI What is metre? What is the meaning of the barline? I hope I have shown that these questions cannot receive a single answer, not even for the single century

145

146

w illiam rothstein

Example 5.28 Verdi, Il trovatore, ‘Vedi! le fosche notturne spoglie’ (Coro di Zingari), bars 1–22

that separates Haydn’s birth from Schubert’s death. It is not only modernday theorists who have held opposing conceptions of metre; eighteenth- and nineteenth-century composers did too. Sometimes opposing conceptions were held alternately, or even simultaneously, by the same composer, as I have suggested was the case for Haydn, Mozart and Beethoven – composers who lived in a Vienna that was still strongly influenced by Italy, at a time of transition in metrical theory and practice. By 1820, German metre was well established as a compositional framework, not only in Germany but further afield, while Franco-Italian metre continued to be used, at least in France and Italy.

National metrical types

Example 5.29 Koch, Versuch, vol. 2, 400

The theory of metre developed after 1875 by Westphal and Riemann corresponds well to Franco-Italian metre,39 but it is ill-equipped to deal with German metre. Their theoretical framework made it impossible for these theorists to conceive of German metre as anything but a mistake, or a phenomenon that needed to be explained away. Despite evidence from composers and theorists (including Kirnberger, T¨urk and Czerny) that many cadences were regarded as light rather than heavy,40 Riemann and his present-day successors distort any music in which weak-beat cadences occur. The weak-beat cadence must be explained as ‘really’ strong. Such distortions may be found not only in Riemann’s Phrasirungsausgaben and Maurer Zenck’s book but even in Lippmann’s ‘Mozart und der Vers’, where the author claims that the ‘real’ downbeats in Mozart’s ‘Der Vogelf¨anger bin ich ja’ (from Die Zauberfl¨ote) fall on the third quaver, because that’s where cadences are heard.41 Similar distortions may be found in Koch, who upheld the eighteenth-century consensus after German composers had begun to ignore it. Because Koch equates the downbeat of the bar with the Ruhepunct or caesura note, phrase-endings such as those in Example 5.29 are regarded as anomalous, requiring explanation.42 Koch’s explanation is that the first two beats of these 3/4 bars represent long appoggiaturas which are given their own harmonic accompaniment. The true caesura in each case, according to Koch, remains on the first beat – even though the harmonic cadences would disappear if the so-called appoggiaturas were removed!43 I doubt that I am the only reader who finds Koch’s explanation forced and unconvincing. The term metre refers to measure. German metre and Franco-Italian metre measure different things. Italian metre is organized around accents, usually explicit accents, whether textual or otherwise. The barlines set these accents in relief; they indicate the Schwerpunkte. What does German metre measure? Judging from the writings of theorists from Gottfried Weber to Carl Schachter and Christopher Hasty, German metre measures the number

147

148

w illiam rothstein

of pulses between the beginnings of rhythmic cycles, so long as those cycles are regular; the cycles themselves may be the musical equivalents of poetic feet, hemistichs, verses, couplets, or quatrains, with the qualification that short anacruses (such as the first syllable of an iamb) are disregarded. One hierarchical level is chosen as the notated bar; as Theodor Wiehmayer pointed out, which level is chosen is partly arbitrary.44 German metre presupposes that metrical accent is different, because more purely psychological, than other kinds of accent; it refuses to reduce the concept of ‘accent’ to a single dimension. Saint-Sa¨ens said it well when he argued, against Vincent d’Indy, that the ‘emphasis or accent’ in a bar may fall anywhere.45 When Riemann claimed that his conception of the barline is the traditional one, he was half-right, but he failed to see that there are at least two traditions, and they are incommensurate; one definition of metre cannot be translated into the other. North American scholars have been slow to recognize FrancoItalian metre,46 yet it is ironic that German scholars have resisted a concept of metre whose origins are almost certainly German, tied to the nature of the German language, and whose traces can be found throughout the great nineteenth-century collections of German folk song. Example 5.30 offers for comparison two Italian folk songs (from a twentieth-century collection) and an eighteenth-century French vaudeville. It might be argued that the distinction between German and FrancoItalian metre is merely that between downbeat (abtaktig) and upbeat (auftaktig) phrase beginnings – or, in terms of Lerdahl and Jackendoff ’s theory, between ‘in phase’ and ‘out of phase’ relations between grouping and metre.47 In such a view, metre itself would remain unitary, its nature fixed and immutable. That this view is flawed is suggested by the existence of cultural differences far greater than that between France and Germany. As is well known, Indonesian gamelan music has a deep metrical hierarchy in which all beat-cycles, of whatever length, are end-accented. When transcribed into Western notation, however, these cycles end not on the last but on the first beat of a bar.48 It is easy to see that the conventions of Western notation do violence to the Indonesian concept of metre. Here one can only sympathize with the suggestion of Momigny and Westphal that the strong beat be counted not as ‘1’ but as ‘2’ or ‘4’; but this is not how most Western musicians have regarded the structure of the bar. Nevertheless, under the influence of French and Italian verse structures, Western composers have sometimes treated metre in a way analogous to that of gamelan music, as demonstrated in this study by Couperin’s rondeaux and Hal´evy’s aria (Examples 5.18 and 5.27 respectively). An exploration of ‘German’ and ‘Franco-Italian’ hypermetre, while

National metrical types

(a)

(b)

(c)

Example 5.30 (a) Three German folk songs from Andreas Kretzschmer and Anton Wilhelm von Zuccalmaglio, Deutsche Volkslieder mit ihren Original-Weisen. 2 vols. (Berlin: Vereins-Buchhandlung, 1838 and 1840) (b) two Italian folk songs from Roberto Leydi, I canti popolari italiani: 120 testi e musiche con la collaborazione di Sandra Mantovani e Cristina Pederiva (Verona: Mondadori, 1973) (c) an eighteenth-century French vaudeville from Le chansonnier fran¸cois; ou, Recueil de chansons, ariettes, vaudevilles et autres couplets choisis. 16 vols. (Paris, 1760–2), only the first quatrain of text given

149

150

w illiam rothstein

Example 5.31 Riepel, De Rhythmopoeia, 50

Example 5.32 Riepel, ‘Des Harmonischen Sylbenmaßes dritter Teil’, 123

beyond the scope of the present study, would confirm and extend the conclusions I have drawn here. In short, ‘German metre’ and ‘Franco-Italian metre’ represent two different ways of listening. Which way a listener chooses to hear is surely a matter of acculturation, but there is much to be gained by being able to listen in either way. I should like to give the last word to Joseph Riepel, whom I quoted in my first epigraph. Riepel, who travelled through Eastern Europe as a young man, was unusually aware of, and open-minded about, differences in musical and metrical styles. When Riepel contrasts French compound metre, which he considers outmoded, to the newer German type with cadences in mid-bar, his hypothetical student asks if die Alten were wrong to write as they did. No, replies the master; they were not wrong.49 And when the master shows his student two different barrings of the same melody (Example 5.31), representing how the melody is heard respectively by Polish merchants and German peasants, the student asks which barring is the correct one. ‘Both’, replies the master.50 Similarly, for the melody discussed in my epigraph, Riepel notes that ‘the Italian metric organization [Example 5.32, first line] is held to be the most natural, because we originally got it from them; but one must not blindly condemn as errors liberties committed

National metrical types

against it, because we ourselves take many liberties . . . To oppose the liberties of masters is presumptuous.’51

Appendix: exceptions in Mozart to the ‘downbeat rule’ for the accento comune All of Mozart’s Italian operas, serenatas, and concert arias were surveyed, using the text of the Neue Mozart-Ausgabe. I focused mainly on passages in 6/8, 2/2, 4/4 and 12/8. The following were excluded: (1) an isolated phrase with ‘misplaced’ accent; (2) polyphonic passages with overlapping texts; (3) passages (usually comic) in which the text is delivered twice as fast as normally, so that two verses are sung within one bar. For each number listed, the following information is given: tempo; notated metre; poetic metre; bars in which the exception is found. Further comment is offered for selected examples. Identifications of ‘German’, ‘French’ and ‘Italian’ compound metres refer to Mozart’s notation, not to the effect on a listener. No examples were found in Mozart’s concert arias (four volumes in the NMA). No examples were found in two operas by J. C. Bach: Carattaco (1767) and La clemenza di Scipione (1778).

La finta semplice, K51 (1768) No. 2 (Aria Simone). Tempo ordinario, 4/4, ottonario; bars 70–5 (revised version only) No. 4 (Aria Cassandro). Allegro non molto, 6/8, ottonario; bars 60–74. The earlier mixed 6/8 shifts by half a bar. No. 19 (Duet Fracasso/Cassandro). Allegro, 4/4, ottonario; bars 65–85. Caesuras fall mostly on beat 3, sometimes on 1.

Mitridate, Re di Ponto, K87 (1770) No. 8 (Cavata Mitridate), the third of four sketches printed in NMA (p. 277). [No tempo given], 4/4, ottonario. The first and third main periods begin in ‘German’ 4/4 but modulate to ‘French’ 4/4, ending on a downbeat. The second main period is also written in ‘German’ 4/4; its closing phrase is expanded to end on a downbeat. No. 24 (Aria Farnace). Andante, 4/4, ottonario; main section of da capo aria. ‘German’ 4/4, with closing phrases expanded to end on a downbeat.

151

152

w illiam rothstein

Ascanio in Alba, K111 (1771) No. 24 (Coro di Pastorelle). Allegro, 4/4, quinario; bars 7–12. ‘German’ 4/4 follows a polyphonic opening; the last half-bar is repeated after a deceptive cadence, causing the final cadence to fall on the downbeat of bar 13.

Il sogno di Scipione, K126 (1771) No. 10 (Aria Scipione). Primo tempo = Un poco Adagio e Maestoso, 4/4, quinario. Five bars of ‘Italian’ 4/4. Secondo tempo = Allegro, 4/4, quinario. First two verses (bar 7 and bar 9) in ‘German’ 4/4 over a tonic pedal.

Lucio Silla, K135 (1772) No. 6 (Coro). Adagio, alla breve, ottonario; bars 21–4 and 29–30 (bars 25–8 are polyphonic). ‘German’ 4/4. Surrounding passages are normal.

La finta giardiniera, K196 (1774–5) No. 5 (Aria Nardo). Secondo tempo = Allegro, 4/4, settenario; bars 132–5. An apparent change to ternario sdrucciolo is set with two four-syllable ‘verses’ per bar, with the accento comune falling on beat 3. No. 12 (Finale i). Bars 110–34 = Adagio ma non molto, alla breve, ottonario; ‘Italian’ 4/4. Bars 213–18 = Maestoso, 4/4, quinario; ‘German’ 4/4. No. 19 (Aria Contino). Adagio, 4/4, ottonario; bars 59–61. No. 23 (Finale ii). Primo tempo = Andante sostenuto, 4/4, ottonario; bars 3–9, 23–4, 35–9, 46–54; some other passages are double-timed (two verses in one bar).

Il re pastore, K208 (1775) No examples.

Idomeneo, K366 (1781, revised 1786) No. 18 (Coro). Allegro assai, 12/8 [extremely rare in Mozart], senario. Begins in ‘French’ 12/8; includes short passages of ‘German’ 12/8.

National metrical types

No. 20b (Duet Ilia/Idamante). Substitute duet for 1786 Vienna production. Larghetto, 4/4, settenario. ‘Italian’ 4/4 throughout; very unusual for Mozart!

L’oca del Cairo, K422 (1783) No. 2 (Aria Auretta). Andante, 4/4, settenario; bars 12–33 and 47–76. ‘German’ 4/4; remaining passages are written quasi alla breve. No. 6 (Finale). Bars 301–10 = Andante maestoso, 4/4, quinario. Parlando style; accento comune falls on beat 3 except at the end, where it falls on 1. Duet Auretta/Chichibio (in Anhang). Allegretto vivo, 4/4, ottonario; bars 8–11 and 18–21 (transposed repetition). ‘German’ 4/4. Introductory and final ritornellos use the same music.

Lo sposo deluso, K430 (1783) No examples.

Le nozze di Figaro, K492 (1786) No examples.

Don Giovanni, K527 (1787) No. 17 (Aria Don Giovanni). Andante con moto, alla breve, settenario; bars 1–6 (repeated later in the aria). Written like ‘German’ compound 4/4.

Cos`ı fan tutte, K588 (1789) No. 25 (Rondo` Fiordiligi). Adagio, 4/4, ottonario; bars 11–21 (middle section of Adagio). ‘Italian’ 4/4.

La clemenza di Tito, K621 (1791) No. 19 (Rondo` Sesto). Primo tempo = Adagio, 4/4, ottonario; bars 6–9 and 26–31. ‘German’ 4/4. In the autograph, Mozart shifted the barlines by half a bar, making most of the primo tempo conform to the usual rule regarding placement of the accento comune.

153

154

w illiam rothstein

Notes 1 Joseph Riepel, ‘Des Harmonischen Sylbenmaßes dritter Teil’, manuscript transcribed in Riepel, S¨amtliche Schriften zur Musiktheorie, ed. Thomas Emmerig, 2 vols. (Vienna: B¨ohlau, 1996), vol. 2, 119–75, here 123. 2 Camille Saint-Sa¨ens, Outspoken Essays on Music, trans. Fred Rothwell (London: Kegan Paul, 1922), 12–13 (from the essay ‘The Ideas of M. Vincent d’Indy’). 3 The term ‘hypermeasure’ was coined by Edward T. Cone in Musical Form and Musical Performance (New York: Norton, 1968), but Weber gave the classic explanation of the concept in his Versuch einer geordneten Theorie der Tonsetzkunst, 3 vols. (Mainz: Schott, 1817–21 and later editions). 4 This subject is treated in Cone, Musical Form, 59–82. Its earliest treatment, however, seems to be in two essays by Thrasybulus Georgiades, ‘Aus der Musiksprache des Mozart-Theaters’ (1950) and ‘Zur Musiksprache der Wiener Klassiker’ (1951); both are reprinted in Georgiades, Kleine Schriften (Tutzing: Schneider, 1997), 9–43. 5 Theorists who permit exceptions to the downbeat-cadence rule, while clearly regarding them as exceptions, include Jean-Philippe Rameau, G´en´eration harmonique ou trait´e de musique th´eorique et pratique (Paris: Prault fils, 1737); Johann Mattheson, Der vollkommene Capellmeister (Hamburg: Herold, 1739); and Johann Philipp Kirnberger (Die Kunst des reinen Satzes in der Musik, vol. 2 part 1 (Berlin and K¨onigsberg: G. J. Decker und G. L. Hartung, 1776)). Rameau and Kirnberger refer specifically to cadences on the upbeat in triple metre, Mattheson to cadences on the final beat in any metre. 6 Mattheson’s critical commentary is added in a new edition of Niedt’s Handleitung zur Variation, the second volume of his Musicalische Handleitung, first published in 1706. The published English translation includes Mattheson’s commentary. See Friedrich Niedt, The Musical Guide, trans. and ed. Pamela L. Poulin and Irmgard C. Taylor (Oxford: Clarendon Press, 1989), 91. 7 Kirnberger, Kunst, vol. 2 part 1; included in Kirnberger, The Art of Strict Musical Composition, trans. David Beach and Jurgen Thym (New Haven: Yale University Press, 1982); Georg Joseph Vogler, Tonwissenschaft und Tonsetzkunst (Mannheim, 1776) and Betrachtungen der Mannheimer Tonschule, 3 vols. (Mannheim, 1778–81); Daniel Gottlob T¨urk, Klavierschule oder Anweisung zum Klavierspielen f¨ur Lehrer und Lernende (Leipzig and Halle: Schwickert und Hemmerde, 1789). 8 See, for instance, Betrachtungen, vol. 2, 115, quoted in Floyd K. Grave, ‘Abb´e Vogler’s Revision of Pergolesi’s Stabat Mater’, Journal of the American Musicological Society 30 (1977), 43–71, 50n. 9 For Kirnberger, see Art, 398 (Kunst, vol. 2 part 1, 131); for Vogler, see Tonwissenschaft, 37.

National metrical types

10 Grave, ‘Abb´e Vogler’s Revision’. 11 Floyd K. Grave, ‘Common-Time Displacement in Mozart’, Journal of Musicology 3 (1984), 423–42, and ‘Metrical Displacement and the Compound Measure in Eighteenth-Century Theory and Practice’, Theoria 1 (1985), 25–60; Clau¨ dia Maurer Zenck, Vom Takt: Uberlegungen zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert (Vienna: B¨ohlau, 2001). 12 Maurer Zenck, Vom Takt, 244–50. Compare Riemann’s analysis in L. van ¨ Beethovens s¨amtliche Klavier-Solosonaten: Asthetische und formal-technische Analyse mit historischen Notizen, 3 vols., 2nd edn (Berlin: Max Hesse, 1919), vol. 3, 177–81. 13 Maurer Zenck, Vom Takt, 37–8. 14 The re-barring of this duet has been much discussed in recent years. See, for instance, Larry Laskowski, ‘Voice Leading and Metre: An Unusual Mozart Autograph’, in Trends in Schenkerian Research, ed. Allen Cadwallader (New York: Schirmer Books, 1990), 41–50; Arnold Feil, ‘Zum Rhythmus der Wiener Klassiker’, Internationaler Musikwissenschaftlicher Kongreß zum Mozartjahr 1991, Baden–Wien, ed. Ingrid Fuchs (Tutzing: Schneider, 1993), 241–54; Rudolf Nowotny, ‘Das Duett Nr. 7 aus der Zauberfl¨ote: Periode und Takt’, MozartJahrbuch 1996, 85–126; Benjamin Perl, ‘Wo soll man bei 6/8 die Taktstriche ziehen? Weitere Gedanken zum Duett aus der Zauberfl¨ote’, Mozart-Jahrbuch 1998, 85–101. 15 Two partial autographs are reproduced in Mozart, Neue Ausgabe s¨amtlicher Werke, ix: 26, xviii–xix; the altered barlines appear on the latter page. In Gluck’s aria, only the final ritornello is barred in the way that Mozart initially used for the aria’s beginning. 16 Compare Diabelli’s waltz, with its four bars of tonic, four of dominant, and so forth. 17 Rudoph Westphal, Allgemeine Theorie der musikalischen Rhythmik seit J. S. Bach (Leipzig: Breitkopf und H¨artel, 1880), 60–4. Westphal applies this counting more directly to another number from Don Giovanni, Zerlina’s aria ‘Batti, batti’, but he discusses ‘Notte e giorno faticar’ in the same passage, and in very similar terms. 18 Hugo Riemann, Musikalische Dynamik und Agogik (Hamburg: Rahter, 1884), 240n–241n. 19 In a slightly later work, Katechismus des Klavierspiels (Leipzig: Max Hesse, 1888), Riemann indicates that the correct metre for the Nocturne is actually 6/8: ‘Here the unit of measurement is the [dotted crotchet], therefore the bars indicated by Chopin are compound; but unhappily Chopin has not placed the barline so that it might indicate the strong accent [Schwerpunkt] of the accented (responsive) bar, but rather the barline is placed throughout so as to indicate the accent of the unaccented bar, as is obvious from the fact that all the cadences fall in the middle of the bar.’ From an anonymous English translation, Catechism of

155

156

w illiam rothstein

20

21

22

23 24

25

26

Pianoforte Playing (London: Augener, 1892), 71. The passage is found on page 61 of the original German edition. Thus Westphal, Allgemeine Theorie, lix: ‘In instrumental music, the composer places the barlines following the same principles as in vocal music: before the most strongly stressed theses [strong syllables or beats], not to designate the boundaries of rhythmic groups, which coincide with the barlines just as infrequently [here] as they do in vocal music.’ Oswald Jonas, Introduction to the Theory of Heinrich Schenker: The Nature of the Musical Work of Art, trans. and ed. John Rothgeb (New York: Schirmer Books, 1982), 15. Ironically, Koch would have had no argument with Chopin’s barring in Op. 9 No. 2, because he would have regarded the third beat as equal to the first, and thus equally appropriate for a cadence. I am assuming that Chopin, like most nineteenth-century musicians, regarded the first and third beats in 4/4 and 12/8 as unequal. Koch served as Kapellmeister for only one year before relinquishing the post voluntarily. Heinrich Christoph Koch, Versuch einer Anleitung zur Composition, 3 vols., vol. 2 (Leipzig: Adam Friedrich B¨ohme, 1787), 310–11. Interestingly, Schenker uses the term Einschnitt (‘caesura’) to describe the undesirable effect of writing an untied minim following two crotchets in a bar of fifth-species counterpoint. Fux also discourages this practice, although his reasoning differs from Schenker’s. See Johann Joseph Fux, Gradus ad Parnassum (Vienna: van Ghelen, 1725), 80–1; the partial translation by Alfred Mann (Steps to Parnassus: The Study of Counterpoint (New York: Norton, 1943)) is inaccurate in the passage in question (67). For Schenker, see his Neue musikalische Theorien und Phantasien, vol. 2: Kontrapunkt, part 1: Cantus Firmus und zweistimmiger Satz (Vienna: Universal Edition, 1910), 407–9; Eng. trans. (Counterpoint) by John Rothgeb and J¨urgen Thym (New York: Schirmer Books, 1987), vol. 1, 316–17. The rule received perhaps its clearest articulation in Rameau’s Treatise on Harmony, trans. Philip Gossett (New York: Dover, 1971), 176–7: ‘Verses appropriate for airs de mouvement are those which contain an equal number of syllables or feet and whose meaning is on the whole complete with the end of each line. In setting such verses to music, the last syllable of each line must occur at the beginning of the first beat of a measure. Notice two things: First, the penultimate syllable of a line whose rhyme is feminine may follow the preceding rule; second, whenever possible, we must avoid using a final cadence on the last syllables of lines which do not conclude the meaning of the sentence.’ Friedrich Lippmann, ‘Mozart und der Vers’, in Colloquium ‘Mozart und Italien’ (Rom 1974), ed. Friedrich Lippmann, Analecta musicologica 18 (Cologne: Arno Volk, 1978), 107–37.

National metrical types

27 Robert Marshall, The Compositional Process of J. S. Bach: A Study of the Autograph Scores of the Vocal Works, 2 vols. (Princeton: Princeton University Press, 1972), vol. 1, 71–5. 28 Johann Mattheson, Der vollkommene Capellmeister: A Revised Translation with Critical Commentary, trans. Ernest C. Harriss (Ann Arbor: UMI Research Press, 1981), 322. For Mattheson’s definitions of the choraic and melismatic genres, see 223–4. 29 William H. Youngren, C. P. E. Bach and the Rebirth of the Strophic Song (Lanham, MD: Scarecrow, 2003). 30 Youngren, C. P. E. Bach and the Rebirth of the Strophic Song, 73–83. 31 For a striking early example, see the second movement of Bach’s Keyboard Sonata in B minor, h32.5 (1743). Such functional distinctions between strongand weak-beat cadences are mentioned by Marpurg as a possible justification for using weak-beat cadences; the relevant passage is translated in David A. Sheldon, Marpurg’s Thoroughbass and Composition Handbook: A Narrative Translation and Critical Study (Stuyvesant, NY: Pendragon, 1989), 172–5. But Marpurg allows this usage only for half or ‘imperfect’ cadences, not for the perfect cadences used by Bach in the examples cited here. 32 An interrupted (deceptive) cadence falls on the downbeat of bar 46. 33 Ido Abravaya, ‘The Baroque Upbeat: Outline of its Typology and Evolution’, in Bach Studies from Dublin, ed. Anne Leahy and Yo Tomita, Irish Musical Studies 8 (Dublin: Four Courts, 2004), 17–28. Abravaya reproduces a different example than the one given here. 34 Vogler, Betrachtungen, vol. 1, 107–9; Grave, ‘Abb´e Vogler’s Revision’, 52–4. 35 See Grave, ‘Metrical Displacement’, 40–2; David Temperley, ‘End-Accented Phrases: An Analytical Exploration’, Journal of Music Theory 47 (2003), 125– 54. 36 Arnold Schoenberg, Style and Idea, ed. Leonard Stein (London: Faber and Faber, 1975), 414–16 (from the essay ‘Brahms the Progressive’); Maurer Zenck, Vom Takt, 251–4. 37 Maurer Zenck, Vom Takt, 27–8. 38 See Channan Willner, ‘Durational Pacing in Handel’s Instrumental Works: The Nature of Expansion in the Music of the High Baroque’ (Ph.D. dissertation, City University of New York, 2004), especially 163–5. 39 Striking confirmation of this claim may be found in Bonifazio Asioli, Il maestro di composizione, 3 vols. (Milan: G. Ricordi, 1832), vol. 3, 16, where Asioli re-bars in 12/8 the opening of the Andantino grazioso from Haydn’s String Quartet in C major, Op. 74 No. 1, originally in 3/8. Reasoning precisely as Riemann does in his discussion of Chopin’s E flat major Nocturne, Asioli begins the movement with a half-bar upbeat so that the accento comune (Riemann’s Schwerpunkt) will fall on the downbeat of the 12/8 bar. 40 See Kirnberger, Art, 398–402 (Kunst, vol. 2 part 1, 131–5), where it is explained that cadences fall on the second half of the compound 4/4 bar and that the

157

158

w illiam rothstein

41

42 43

44

45 46 47 48

49

50 51

second half should be performed ‘more lightly than the first’; Daniel Gottlob T¨urk, School of Clavier Playing, trans. Raymond H. Haggh (Lincoln: University of Nebraska Press, 1982), 325–6 and 330–1 (Klavierschule, 335–7 and 340–2); Carl Czerny, Vollst¨andige theoretisch-practische Pianoforte-Schule, Op. 500, vol. 3: Von dem Vortrage (Vienna: Diabelli, 1839), 5–6. Czerny’s example suggests that he regarded cadences in the fourth and eighth bars of an eight-bar period as light, not heavy, in contrast to Riemann’s theory of the eight-bar period. Lippmann, ‘Mozart und der Vers’, 130–5. According to Lippmann, Mozart’s setting of Papageno’s song presents ‘stronger rhythmic surprises’ (‘st¨arkere ¨ rhythmische Uberraschungsmomente’) than Italian operas by Mozart and his contemporaries. Thus, apparently, it is precisely in its simplest, most folk-like passages that Die Zauberfl¨ote outdoes, for example, Don Giovanni in ‘rhythmic surprises’. This is a curious conclusion, to say the least. See also Example 4.12 in Danuta Mirka’s essay in this volume. Indeed, the model phrase from which Example 5.29 is allegedly derived (Koch, Versuch, vol. 2, 400; reproduced in this volume as the first system of Example 4.12) contains no dominant and, hence, no cadence. Theodor Wiehmayer, ‘Der “Große Takt” und die Analyse der c-moll-Symphonie von Beethoven’, Zeitschrift f¨ur Musikwissenschaft 4 (1922), 417–27. Earlier, in his book Musikalische Rhythmik und Metrik (Magdeburg: Heinrichshofen, 1917), 50–5, Wiehmayer had spoken of ‘normal’, ‘large’ and ‘small’ bars. William Caplin makes a similar distinction in Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven (New York: Oxford University Press, 1998), 35. Saint-Sa¨ens, Outspoken Essays, 12. James Webster is an exception; see his ‘The Analysis of Mozart’s Arias’ in Mozart Studies, ed. Cliff Eisen (Cambridge: Cambridge University Press, 1991), 101–99. Fred Lerdahl and Ray Jackendoff, A Generative Theory of Tonal Music (Cambridge, MA: The MIT Press, 1983), especially 25–35. Harold Powers (personal communication). Lerdahl and Jackendoff mention gamelan music as a counter-example to their theory (A Generative Theory of Tonal Music, 150–1). I use the terms ‘French’ and ‘German’ here to describe metrical types, not musical styles. Nevertheless, Riepel comments that ‘only a few French contredanses remain bound to the old cadence on the downbeat’; Joseph Riepel, Anfangsgr¨unde zur musicalischen Setzkunst, 5 vols., vol. 1: De Rhythmopoeia oder von der Tactordnung (Regensburg and Vienna: Emerich Felix Bader, 1752), 49–50; reprinted in Riepel, S¨amtliche Schriften, vol. 1, 69–70. Riepel, De Rhythmopoeia, 50 (S¨amtliche Schriften, vol. 1, 70). Riepel, ‘Des Harmonischen Sylbenmaßes dritter Teil’, in S¨amtliche Schriften, vol. 2, 123. In my translation I have reversed the order of Riepel’s two sentences; the

National metrical types

first refers to an earlier example of weak-beat caesuras. The melody in Example 5.32 is so devised as to permit the final cadence to be heard on a downbeat regardless of the barring used; what varies is not, therefore, the cadence but the location of lesser caesuras and the relation of the melodic phrase to the barlines.

159

6

Schoenberg’s ‘second melody’, or, ‘Meyer-ed’ in the bass william e. caplin In memory of Leonard B. Meyer (1918–2007)

At the end of a chapter of his Fundamentals of Musical Composition entitled ‘Advice for Self Criticism’, Arnold Schoenberg admonishes the beginning composer to: ‘watch the harmony; watch the root progressions; watch the bass line’.1 It is no surprise that he would direct the composer to harmony and root motion, the bread and butter of compositional training. But striking is his highlighting of the bass part, in effect saying nothing about the upper voice, to which listeners normally direct their hearing. Indeed, most composition treatises – from Riepel and Koch in the eighteenth century, through Marx, Lobe and Riemann in the nineteenth – employ musical examples that mostly show the soprano melody alone, thus supposing that readers will intuit the bass line on their own. Yet Schoenberg wants to focus on the lowest voice of the musical texture; and in so doing, he taps into another important source of compositional pedagogy, one reflected, for example, by the numerous partimenti treatises produced throughout those same two centuries, works that provide stock bass lines for learning improvisation and composition.2 Now what does Schoenberg actually want the composer to watch for in the bass line? Earlier in the chapter, he briefly explains: The bass was previously described as a ‘second melody’. This means that it is subject to somewhat the same requirements as the principal melody. It should be rhythmically balanced, should avoid the monotony of unnecessary repetitions, should have some variety of contour and should make full use of inversions (especially of seventh chords).3

Schoenberg’s remarks are suggestive enough, but they constitute little more than a starting point for understanding how bass lines are structured and, more specifically, how they can attain the status of a ‘second melody’. As important as the bass undoubtedly is, music theorists have been strangely silent on just how to analyse that voice as a melodic construct. The few writings devoted to the topic are pedagogically oriented studies largely concerned with how melodic patterns in the bass relate to the harmonies supported by that line.4 One would assume, of course, that Schenkerian

Schoenberg’s ‘second melody’

theory would provide important information about bass-line organization, especially since Schenker defines the outer voices as fundamentally different entities: the upper voice conforms to the demands of the Urlinie’s stepwise descent, while the bass supports that descent through an arpeggiation of the essential tones of the overriding tonic. But Schenker tends to associate linear progressions, or Z¨uge, arising at later structural levels with motion in the upper voices, and whereas the bass may acquire a melodic contour through embellishments of its fundamental harmonic tones, its Z¨uge are typically seen as driven by, or supportive of, those in the upper parts. All too often, important melodic details of the bass are not represented in a Schenkerian graph, except perhaps at the most foreground levels, and Schenker’s general characterization of that voice as the Bassbrechung, the ‘bass-arpeggiation’, seems to inhibit him from defining independent principles of bass melody.5 Leonard B. Meyer offers a different view of melody.6 With his emphasis on the manifold implications that melodic motions may engender, along with their specific realizations, Meyer attends to the individuality of a given melody without necessarily considering its relation – either harmonic or contrapuntal – to the other parts. But Meyer analyses soprano lines almost exclusively, and nothing in his theory suggests that the bass would differ melodically from the upper voices. In short, theorists have not yet provided a framework for comprehending the bass as a melody in its own right, as a melodic voice distinct from others. This study is a step in that direction. I want to take seriously the idea that the bass line can be heard as an independent object of aesthetic attention, without necessarily referring to its contrapuntal interaction with the soprano. I entirely concede that this is a highly partial mode of hearing, but I believe that interesting results can be obtained by such a focused listening experience. I begin by proposing some fundamental models underlying many bass melodies and then illustrate a variety of melodic techniques in selected passages from Haydn, Mozart and Beethoven. I conclude with a brief discussion of how bass-line melodic organization can be thought of as a communication strategy employed by late eighteenth-century composers. As will be obvious, I am influenced throughout both by Schenker’s Bassbrechung and by Meyer’s ideas of melodic implication and realization, and I employ modified aspects of each of their analytic notations. What I find especially attractive in Meyer’s approach is the potential for uncovering relationships that cross over boundaries defined by harmonic or contrapuntal prolongation, thus allowing for the representation of melodic patterning that might be deemed invalid from a strictly Schenkerian perspective.

161

162

william e. caplin

gap

fill

Example 6.1 Melodic analysis of ‘Twinkle, Twinkle’

Example 6.2 C major harmony

Let us begin by reviewing some well-known distinctions between melodic and harmonic relationships. A melodic relationship is established in reference to a scalar collection of pitches, and the basic unit of melodic motion is the individual step. If the melodic interval between two pitches exceeds a second, we sense that one or more notes of the scale have been skipped over. Indeed, the ‘gap’ thus created may then be ‘filled-in’ by stepwise motion, as in the ‘Twinkle, Twinkle Little Star’ melody of Example 6.1.7 A harmonic relationship, on the contrary, is established in reference to a triadic collection. Thus the basic units of harmony are the intervals of the fifth and third (and their inversions). Unlike a melodic interval, however, a harmonic one does not necessarily imply the presence of intervening notes: the interval between, say, the root and fifth of a harmony does not create a gap implying a subsequent fill. Harmonic relationships distinguish themselves from melodic ones in another important way, one less often considered by theorists. The pitches of a harmony are normally understood to reside in different voices; thus in the C major harmony of Example 6.2, the C and G, comprising the harmonic fifth, reside in the bass and alto voices respectively; the E lies in the tenor, and the doubled C appears in the soprano. Each of these notes has the potential of creating melodic relationships with other notes in the same voice, as one harmony succeeds another, as shown in Example 6.3. Thus the connections of pitches within one voice are normally melodic, while those between voices are harmonic. But there exists, theoretically, one voice whose pitches are exclusively harmonic: this is Rameau’s basse fondamental,

Schoenberg’s ‘second melody’

163

Example 6.3 Simple harmonic progression in C major

prolongational stream

prolongational stream descending ascending = member of prolongational stream = member of cadential stream

= focal pitch of stream cadential stream

Example 6.4 Prolongational stream and cadential stream under the rule of the octave

which he conceptualizes as a single, distinct voice.8 But what we have been considering thus far suggests that the pitches of the fundamental bass, as harmonic entities, should logically reside in different voices. In other words, as the fundamental bass progresses, its notes form a series of harmonic connections, and as such, each note can be thought to leap from one voice to another. The fundamental bass, of course, is a theoretical construct. But there does exist a real sounding voice that normally includes elements of the fundamental bass: this is what Rameau calls the basso continuo, or, more simply, our regular ‘bass voice’.9 What I am thus proposing is that the bass voice, as distinct from the upper voices, is, in principle, a two-voiced structure.10 Following Schenker, we can say that the bass is anchored in the harmonic relationship of tonic and dominant. But as harmonic constituents, these two notes can be thought to reside in two different voices, and each note can become the focal point of melodic activity within its own voice. To avoid terminological confusion, I will now refer to these internal voices (within the single bass voice) as streams.11 Example 6.4, a modified r`egle de l’octave, shows the pitches normally occurring within the two streams. As well, I have indicated typical harmonic progressions associated with ascending or descending melodic motions within each stream. In reference to these harmonies, I label the stream focused around the tonic as prolongational; that focused around the

164

william e. caplin

= cadential linking of streams = linear motion in prolongational stream = linear motion in cadential stream = prolongational harmonic progression = cadential harmonic progression

Example 6.5 Basic model

dominant, as cadential.12 Observe that the third, fourth and sixth scaledegrees belong to both streams: they can thus function as pivot notes linking the streams in an apparently melodic manner. For a given thematic unit within a movement, such as a main theme or a subordinate theme, the bass line behaves roughly as follows. It begins with the tonic and then explores various melodic motions within the prolongational stream. Eventually, it attains the third degree, which then pivots the line into the cadential stream for further ascent to the dominant, the melodic goal of that stream. Bass-line closure is achieved when, in the context of a formal authentic cadence, the dominant leaps back to the tonic, thus creating a harmonic connection linking the two streams and purging the bass of all melodic tendencies that might generate further continuation. The simplest manifestation of this process is shown in Example 6.5, which I offer as a basic model for bass melodies.13 Comparing this model to the ‘Twinkle, Twinkle’ configuration of Example 6.1, we see that the melodic activity is entirely different:14 the soprano line opens up an ascending gap which motivates a subsequent descending fill that brings closure to the line; all of this melodic motion takes place within a single voice. By contrast, the linear ascent of the bass traverses two streams and closes with a harmonic leap, one that creates no sense of melodic gap for a subsequent fill. To be sure, the ascending motion is so linear that it does not necessarily give the impression of changing from one stream to the other. In actual musical realizations, however, the composer often articulates this shift by means of register, texture, grouping mechanisms, and so forth. Let me now lay out some variants to the basic model. The first ones arise with themes that modulate to the dominant or mediant regions, as shown in Example 6.6. Another variant, especially common in Mozart, underlies a periodic formal design (see Example 6.7).15 Here, the bass of the antecedent phrase ascends to scale-degree three, which has the potential of shifting the melody into the cadential stream (as indicated by the descending dotted stem); a further rise to scale-degree four reinforces that implication. But the melody returns to the initial tonic to complete a broad prolongation. The

Schoenberg’s ‘second melody’

(a) dominant modulating

165

(b) mediant modulating

or

or

Example 6.6 Modulating models

antecedent

consequent = potentially cadential

(a) HC

PAC

= neighbouring motion

(b) implication

Example 6.7 Period model

or

(a) g

(b)

ˆ Example 6.8 2-gapped model

line then leaps directly to the dominant, thus bypassing cadential melodic activity, to create a formal half cadence. The consequent phrase reproduces the opening ascent; but this time, the potential for scale-degrees three and four to be cadential is fully realized, and the line continues up to the goal dominant, whose leap back to the tonic creates authentic cadential closure for the period. The notation in staff b shows the implication/realization (or I/R) relationships of the third and fourth degrees.16 Another common ˆ variant, shown in Example 6.8, omits scale-degree 2. With this ‘2-gapped’ variant, it is often interesting to see whether the lack of the second degree is offset at some point within the theme itself, or, more likely, after the theme has closed. Let us now consider these abstract models in relation to actual music. Example 6.9 shows the opening of the slow movement of Mozart’s Piano Trio in B flat major, k502.17 The bass line of the four-bar antecedent phrase conforms

realization

166

Larghetto

william e. caplin

continuation

antecedent

5

6

V-modulating model

basic model

(a) m. 5

m. 7

(b)

Example 6.9 Mozart, Piano Trio in B flat major, k502/ii, bars 1–8

entirely to the basic model. Instead of a following consequent, Mozart writes at bar 5 a continuation, a phrase conventionally associated with the second half of the sentence (Satz) theme-type.18 The phrase begins with an ˆ which I interpret as operating within the unusual move from 5ˆ down to 3, prolongational stream, since nothing of the formal and harmonic context suggests cadence.19 At bar 6, the bass returns to tonic via the neighbouring leading-tone, and from here to the end of the continuation, we can recognize the dominant-modulating model. Staff b of the analysis shows a possible I/R relationship generated by the unusual bass descent of bar 5, whose continuation with the second-degree F, at bar 7, is an essential element of the modulation. The period model is well illustrated by the opening of the finale of Mozart’s Piano Sonata in B flat major, k281 (Example 6.10). The antecedent begins with a double neighbour-note configuration embellishing the initial tonic of the prolongational stream.20 The leap to 3ˆ at the end of bar 2 can be heard to

Schoenberg’s ‘second melody’

Rondeau Allegro antecedent

167

consequent 2

4

p

f

x

x

x

PAC

HC

x

x

x

(a) (b)

(c)

(d) g

g

(e)

Example 6.10 Mozart, Piano Sonata in B flat major, k281/iii, bars 1–8

initiate the cadential stream, and the continuation up to 4ˆ further supports ˆ which then pushes back down that presumption. But when 4ˆ returns to 3, ˆ to 1, we recognize that all of this melodic activity actually takes place within the prolongational stream. The leap to the dominant at bar 4 supports the half cadence that closes the phrase. The opening of the consequent brings back the same basic pitches of the antecedent, but now with some upperthird embellishments (labelled motive x). The second embellishment in bar ˆ whose repetition and ascent to 4ˆ fully realizes the cadential 6 achieves 3, implications from the prior phrase; staff b shows this I/R relationship. As ˆ seen in staff c, the consequent exhibits the 2-gapped variant of the basic model. But when we take the double-neighbour note configuration into account (see staff d), we can discern a latent expression of the complete basic model.21 Finally, staff e highlights a number of gap-fill relationships that arise in the course of the theme. Another example of the period model, this time in Beethoven’s Bagatelle in E flat major, Op. 126 No. 3, shows a more complex interaction of prolongational and cadential streams (Example 6.11). Following a long

g

168

Andante

william e. caplin

antecedent 4

5

6

7

expanded cadential progression (E.C.P.)

HC

consequent 9

13

crescendo

-

-

-

-

-

-

15

cresc.

p

E.C.P.

p

PAC

(a)

Example 6.11 Beethoven, Bagatelle in E flat major, Op. 126 No. 3, bars 1–16

pedal, the tonic is briefly embellished by its lower neighbour. The line then begins its ascent, one that leads up an entire octave to restore the tonic prior to leaping down to 4ˆ at bar 7, as pre-dominant for the half cadence.22 Let us consider this octave line in greater detail. The quick rise from 1ˆ to 3ˆ at the end of bar 4 is obviously prolongational, but the lingering on 3ˆ in bar 5 strongly signals the start of an expanded cadential progression (abbreviated E. C. P.), which is continued by the iv and v7 harmonies of bar 6.23 It is odd, of course, for the dominant seventh to be placed on the weak third beat, and when iv6 arrives on the final semiquaver of the bar, we recognize that the dominant is functioning not so much cadentially, but rather as a passing chord prolonging the subdominant. iv6 could have returned to a cadential dominant, but instead, it presses up to v6/5, in order to regain a root-position tonic just prior to the half cadence, as called for in the model. Whereas the bass line traverses the cadential stream from 3ˆ up to 5ˆ , 6ˆ proˆ vides the pivot for restoring the prolongational stream in its rise up to 1.

Schoenberg’s ‘second melody’

Thus within this broad ascent, which, on the surface, presents a unified melodic gesture, we can recognize a subtle shifting in and out of the prolongational and cadential streams. In the consequent, the arrival on 3ˆ at bar 13 again proposes an expanded cadential progression. But this one is now fully accepted when the cadential dominant is shifted to the downbeat of bar 15 and holds its root position until resolving to tonic for the perfect authentic cadence.24 Example 6.12 reveals how elements of the period model can enter into a non-periodic formal design. The bass line opening Haydn’s String Quartet in G major, Op. 54 No. 1, begins with the tonic and its neighbouring ˆ leading-tone. At bar 4, the bass shifts up to 3ˆ and ascends very quickly to 6, thus running through pitches typical of the cadential stream. But this extended anacrusis in no way suggests cadential function, and so it is not surprising that the next two bars see the bass moving back down, in an entirely prolongational manner. The arrival on 3ˆ at the end of bar 6, however, strongly signals an impending cadence, and the theme could have continued with bars 7 and 8 bringing a standard cadential progression, say ii6–v7–i, thus allowing the bass to ascend again, this time fully expressing the cadential stream. But Haydn does not let the theme take its standard course; instead, the bass leaps back down to the leading-tone for further elaborations of the tonic in bars 7–10. This restoration of 1ˆ after a strong implication for cadence reminds us of what happens in the antecedent of the period model. Bars 7–10 now build a high degree of tension: having thwarted an expected cadence and now seeming to get stuck on root-position tonic so late in the game, the theme seems desperately in need of closure. So a great sense of release obtains when the bass finally leaps down to 3ˆ in the middle of bar 10 to initiate an expanded cadential progression, which closes the theme at bar 13.25 Apparently sensing that even this expanded cadence was insufficient to balance the preceding ten bars of tonic prolongation, Haydn repeats the cadential phrase in bars 14–16 and even brings a one-bar codetta to conclude the main theme as whole. Note how register plays a supporting role in helping to differentiate the streams. Though the basic model shows the cadential stream lying above the prolongational one, many basses leap down to capture the essential pitches of the cadential stream. ˆ The underlying structure of this main theme’s bass line features the 2gapped variant of the basic model, as shown in staff b. Indeed, scale-degree two is absent from the bass line of the entire theme. In that light, it is interesting to observe what happens at the beginning of the transition, at bar 17. After four bars of tonic pedal at a piano dynamic, the bass dramatically

169

170

william e. caplin

MAIN THEME

Allegro con brio. 4

f

staccato

6

sf

sf

sf 7

10

E.C.P.

sf 11

13

14

sf

16

f

sf

PAC

TRANSITION

sf

sf

17

21

sf sf

sf

f

p

(a) (b) g

g

(c)

Example 6.12 Haydn, String Quartet in G major, Op. 54 No. 1/i, bars 1–23

p

codetta

p

Schoenberg’s ‘second melody’

ˆ now supporting a v ascends at bar 21, with a forte outburst, directly to 2, of v harmony to signal a modulation to the dominant region. Though the ˆ 2ˆ appears in the background structure of many moduprogression from 1– lating transitions, such an explicit move directly at the foreground is rather unusual and can perhaps be attributed to the conspicuous avoidance of the second degree throughout the preceding main theme. Staff c in the analysis ˆ whose implied fills are deferred until shows two local gaps between 1ˆ and 3, the transition; such details help motivate the prominent appearance of the passed-over supertonic.26 I want now to turn to a theme in which, like the 2ˆ of the previous example, the fate of an individual pitch of the underlying model – this time 4ˆ – generates interesting compositional play. The slow movement of Mozart’s Piano Sonata in C major, k330, shown in Example 6.13, opens with a four-bar antecedent, whose disjunct bass line is most complex.27 As staff b shows, ˆ ˆ D, replaces 4, ˆ I see here a variant of the 2-gapped model, in which 6, ˆ B♭. Note how 6 is already emphasized prior to its appearance in the half cadence (see the asterisks in staff a). The following continuation phrase resides entirely in the dominant key of C major and clearly projects the basic model in that new key.28 As a result, scale-degree four (of the home key), B♭, does not appear here, just as it was largely absent in the prior phrase. Following the double barline, the second part of the theme brings a series of descending lines leading from dominant back to tonic and followed by a brief half-cadence at bar 12. As suggested in staff b, this phrase can be seen to ˆ express the 2-gapped model, though 2ˆ does appear as an upper neighbour to the initial tonic. Within these bars, 4ˆ emerges from its prior obscurity to find a role in the structure of the bass line. But a more dramatic exposure of this degree is yet to come: for bars 13–14 see the bass climb back up to B♭ in a broad ascending sweep that counterbalances the prior descending motions. The v4/2 harmony supported by this B♭ temporarily pushes it back down to A, in the following bar. But with this A now supporting a v6/5 of iv, the bass is forced up again to B♭ at bar 16. Scale-degree four thus appears here with considerable emphasis and in a manner that fully compensates for its absence in the first part of the theme. As shown in my analysis, I read the melodic activity of bars 13–16 as residing in the cadential stream, and following the subdominant harmony of that last bar, we could very well expect the bass to rise up to the cadential dominant. But in the following bars, Mozart surprises us by having the bass line leap into the prolongational stream for further melodic action around the tonic. Finally, bar 19 sees a leap back to the cadential dominant to bring authentic cadential

171

172

Andante cantabile

william e. caplin

antecedent

continuation

f

dolce

p

*

*

* C:

PAC

HC

10

12

13

14

[ ]

cresc.

p cresc.

dolce

f

p

HC

15

16

17

19

cresc.

f

p

PAC

*

*

*

(a) HC

(b) "Romanesca"

(c)

(d) (m.15)

Example 6.13 Mozart, Piano Sonata in C major, k330/ii, bars 1–20

closure for the theme. Since it is difficult to interpret the sudden shift to the prolongational stream in terms of our bass-line models, I regard the passage from bar 17 to the downbeat of bar 19 as parenthetical, thus interrupting the broad cadential line begun with 3ˆ in bar 15.29 It is especially in complicated cases like this, that the differentiation of a bass melody into two streams proves to be of particular analytical utility. Up to now, I have identified the basic models underlying Example 6.13 at the level of the various phrases making up the theme, as shown in staff b.

Schoenberg’s ‘second melody’

We might speculate, though, that the entire theme is anchored in the basic model; see staff d. Here, 2ˆ is brought in conjunction with the modulation to the dominant, somewhat akin to what we saw in the prior Haydn quartet example (at bar 21). Scale-degree three is then picked up at bar 15, while the remainder of the model continues as already discussed. Whether it is legitimate and useful to extend the basic model to such a broad-scale level of analysis is a topic for further research and reflection. A set of examples from the slow movement of Beethoven’s Piano Sonata in E flat major, Op. 7, shows how unusual features of a bass melody can play themselves out over the course of an entire movement. Let us begin with the main theme (Example 6.14), which takes the form of a small ternary. The bass line begins normally enough with lower-neighbouring motion around the tonic, after which the bass would typically begin its ascent. In a striking departure from the norm, however, the line suddenly leaps down a tritone to ♯4ˆ in bar 3, a pitch that almost always appears within the cadential stream, as indicated by the dotted downward stem. But instead of resolving up to the dominant, the F♯ presses down to F♮, which, by supporting a v4/2, must itself move down to 3ˆ at bar 5. Descending motion continues back to the opening tonic, thus suggesting that the entire linear descent is best interpreted as prolongational. Insofar as the leap from C to F♯ initially transfers motion from the prolongational to the cadential stream, a melodic fill is not necessarily implied by that gap. Yet since the line continues to descend in a prolongational manner, the possibility arises that this gap may eventually be filled. At bar 6, the bass leaps up to 4ˆ to signal the cadential progression that closes the theme. As shown in staff b, the underlying bass motion of the A section omits both scale-degrees two and three of the basic model. We might wonder, then, whether the complete model may yet appear at some later point in the movement. The B section of the ternary form prolongs dominant via the upperˆ Note the I/R relationship shown in staff c, suggesting that the neighbour 6. ˆ 6– ˆ 5ˆ continues the linear descent initiated by the move from motion from 6–♭ ˆ1 to 7ˆ at the opening of the piece. Perhaps we could even perceive here the intimation of a filling-in of the initial C–F♯ gap. The subsequent A′ section develops materials of the A section in such a way as to forge a bass line that is entirely different and considerably more complex. Following the opening embellishment of tonic in bar 15–16, the line now forgoes the descending leap and instead strives chromatically ˆ The harmonies supported by this ascent are decidedly sequenupwards to 6. tial, though for our purposes here, I have analysed the line as belonging to the prolongational stream.30 The cadential stream takes over at bar 19 with ♯4ˆ

173

174

william e. caplin

Largo, con gran espressione A

3

5

B

6

sf

a

sf

p

tenute

rinf.

PAC

11

15

fp

A'

16

pp

rinf.

sf

sf

f (sequential)

x

18

19

20

22

sf pp

ff

pp

PAC

(dec. cad.)

A

B

pp

x

A'

(sequential)

(a)

(b)

(c)

Example 6.14 Beethoven, Piano Sonata in E flat major, Op. 7/ii, bars 1–24

as pre-dominant for what then becomes, at bar 20, a deceptive cadence, whose sixth degree is initially substituted by its upper neighbour, B♭. Further sequential activity, with emphasis on the supertonic, leads eventually to a final run at the cadence. Note that this second time, the pre-dominant at bar ˆ which neutralizes the prior alteration of that degree: 22 is supported by ♮4, we thus see how the local F♯–F♮ prolongational connection from early in the A section is now transferred to a broader, cadential level of structure (see the downward pointing arrows on staff a). As for the underlying organization of A′ itself, staff b shows that we could discern elements of the complete basic

Schoenberg’s ‘second melody’

25

sempre tenuto

sf

false recapitulation ten. 36

37

sf

f

ten.

41

pp f

pp

sf

p

pp ten. pp

a

(a) =

Example 6.15 Beethoven, Piano Sonata in E flat major, Op. 7/ii, bars 25–8, 36–43

model; yet the powerful sense of sequence associated with scale-degrees two and three somewhat obscures a full expression of that model, and so we might still expect a more obvious appearance of it at some later point. The development section, which immediately follows the main theme, shows that we do not have to wait long for the basic model to appear.31 For the opening phrase in Example 6.15 brings an unembellished form of the model, albeit in the submediant key of A flat major. Further elaboration of this material eventually leads to the home-key dominant at bar 37, at which point 5ˆ is embellished by its upper and lower chromatic neighbours. In bars 41–2, F♯ moves down to F♮, thus seeming to reproduce that unusual move from the beginning of the movement. Here, however, the melodic progression is actually somewhat different, because the F♯ now functions enharmonically, and cadentially, as G♭, whose resolution to F, as a cadential dominant, leads the music into B flat major, for a false recapitulation.

175

176

william e. caplin

Coda 74

72

pp

p

f

sf

ff

f

f

sf

ff

x x PAC

codetta 78

81

84

82

pp

sf

x

PAC

HC

85

86

88

pp

ffp

pp

pp

ffp

PAC

x

x

(a)

(b)

Example 6.16 Beethoven, Piano Sonata in E flat major, Op. 7/ii, bars 72–end

I skip over the rest of the development section along with most of the real recapitulation in order to consider the magnificent coda (see Example 6.16). Eliding with the cadence of the A′ section at bar 74, the coda begins by recapitulating the opening phrase of the development. This not only fulfils a compensatory formal function, but also, finally, realizes the basic model in the home key, though, to be sure, leading only so far as the half cadence at bar 78. From here to the end of the piece, the bass line exhibits a broad descending motion, perhaps to complement the upward progressions featured throughout the A′ section of the recapitulation and the opening of ˆ After the coda. The line first moves prolongationally from 4ˆ down to 1.

Schoenberg’s ‘second melody’

repeating this move in bars 81–2, the line shifts to the cadential stream for an authentic cadence at bar 84. Note that the cadential progression feaˆ 6– ˆ 5ˆ , thus referencing the prior half cadence, as tures the bass motion 6–♭ well as some earlier appearances of that motive, which I have labelled as x. When the cadence is followed by a one-bar codetta, which is immediately repeated, we could well believe that the piece is finished at this point.32 But one unresolved detail of the bass line has yet to be worked through, namely, the gap between C and F♯ from the beginning of the piece. So at bar 86, Beethoven realizes the implication that this gap could be filled in, by leading the bass down chromatically to F♯, whose arrival is emphasized by the fortissimo-piano dynamic. This is not, of course, a standard ‘gap-fill’ of the type we have seen thus far, whereby the fill reverses the direction of the gap; rather, it is what might be called a unidirectional fill, a melodic technique observable in other examples from the literature. What allows us especially to connect this fill back to the opening gap is the manner in which Beethoven allows the chromatically descending bass to support a final return of the main theme’s opening material. Finally, the coda closes with one of the most unusual cadences in the classical literature: having arrived on ♯4ˆ at the end of bar 88, which brings the pre-dominant v6 of v, the bass descends to F♮ to support another predominant – ii6. When these pre-dominants are used together, ii6 almost always precedes the secondary dominant, so that ♮4ˆ becomes chromatically raised on its way to the dominant. That Beethoven unconventionally reverses the harmonies is owing, of course, to his creating a final appearance of the unusual F♯–F♮ motive. But more than just referencing this motive, he assimilates it entirely to the cadential stream, thus allowing the F♯ to realize a major implication arising from its very first appearance in bar 3. Can we say that the theoretical approach to bass-line melodies developed in this study reflects a communication strategy for eighteenth-century instrumental music? Given the highly implicative nature of the melodic processes represented in the models, a likely answer would be yes. The evidence from the analyses above suggests that once a bass line begins to move melodically, particular pitches (or pitch-classes) are expected to occur as continuations of that melodic motion, be they linear continuations or the filling-in of an opened gap. Such pitches may appear exactly at the point they are expected, and these immediate realizations – for example, when the motion from 1ˆ to 2ˆ is followed directly by 3ˆ – are so numerous as not to require further discussion. But when an implied pitch is thwarted in its appearance, its realization may then be deferred to some later, indeterminate point in time.33

177

178

william e. caplin

Such longer-term realizations were observed in a number of the examples discussed earlier.34 For instance, the gap opened up between 1ˆ and 3ˆ early on in the Haydn quartet (see Example 6.12, bar 4) is neither filled-in immediately nor at any other point within the main theme; only when the transition is already under way does the move from 1ˆ to 2ˆ (bars 20–1) bring the longexpected scale degree. Similarly, when 6ˆ replaces 4ˆ within the basic model underlying the opening phrase of Mozart’s k330/ii (see Example 6.13, bars 1–4), expectations are aroused that the omitted degree will arise at some later point in time, though just where that melodic completion will occur cannot be predicted. The various immediate and deferred realizations just cited are part and parcel of Meyer’s general implication–realization approach to melodic analysis. But the dual-stream construct hypothesized here for bass melodies gives rise to implications that are not inherent in Meyer’s model; namely, that the bass line will move between streams in the course of a theme. For example, a potential for compositional play is strongly engendered by our expectation that a prominent arrival on 3ˆ will shift the bass melody from the prolongation stream to the cadential one. But this potential can be initially thwarted and then eventually achieved in many ingenious ways, as discussed in connection with both the Beethoven Bagatelle (Example 6.11) and the Haydn quartet (Example 6.12). In another case, the opening of Beethoven’s piano sonata (Example 6.14) leads to the expectation that the leap from 1ˆ to ♯4ˆ would witness a transfer in streams, but the realization is deferred to the very end of the movement, when ♯4ˆ finally fulfils its normal role within the cadential stream. Finally, an even more abstract expectation involves the appearance, or lack thereof, of the entire bass-line model. As discussed in connection with the Beethoven sonata, the main theme does not conform to the basic model or to any of its variants. Only at the beginning of the development section does the basic model first appear (Example 6.15, bars 25–8), and even then, in the wrong key. It is not until the coda (Example 6.16, bars 74–8) that the basic model finally appears in the home key, albeit not fully complete. One final example summarizes well the various types of implications and realizations arising from the analysis of bass melody. The opening theme for variations of Mozart’s Piano Sonata in A major, k331 (Example 6.17) – perhaps the most widely analysed passage in the entire classical repertory – is constructed as a small ternary (A–B–A′ ), whose A section takes the form of a period.35 The bass line of the opening antecedent already presents a rather complicated set of melodic implications.36 The initial ascent

Schoenberg’s ‘second melody’

Andante grazioso

179

A

consequent

antecedent

p

sf

p

sf

p

PAC

HC

B

A'

9

12

sf

sf

sf

sf

HC

extension 16

17

f

p

IAC

PAC

A

B

antecedent

consequent

A'

extension

(a) g

(b)

(c)

(d)

Example 6.17 Mozart, Piano Sonata in A major, k331/i, bars 1–18

ˆ is very briefly deferred until from 1ˆ to 3ˆ in bar 1 opens up a gap whose fill (2) the second half of bar 2, and whose return to 1ˆ is further delayed until the downbeat of bar 4 (see staff b). At a broader level of motion, the move from 1ˆ to 7ˆ on the downbeats of bars 1 and 2 would most likely imply a return ˆ since this lower-neighbour configuration is highly typical of opening to 1, bass-melodic gestures (see Examples 6.11, 6.12 and 6.14). But as shown in staff c, the return to tonic is deferred until the downbeat of bar 4, when the ˆ 37 At the same leading-tone is embellished by its own lower neighbour (6).

180

william e. caplin

time, the opening descent from 1ˆ to 7ˆ implies a linear continuation, one that is immediately realized at the level of crotchet motion by the appearance of 6ˆ on the downbeat of bar 3 (see staff d ).38 Further descent to 5ˆ is implied, but then deferred until the second half of bar 4. A similar melodic implication arises in the course of the B section, when the move from 1ˆ to ˆ which is quickly realized in bar 12 and continued 7ˆ proposes a descent to 6, on to 5ˆ in that same bar.39 Here, however, another level of expectation – one involving the two streams – comes into play, namely, the expectation that 6ˆ would eventually participate in the cadential stream, as support for a pre-dominant harmony. In the earlier descent (bars 1–3), 6ˆ could only be interpreted as prolongational; at bar 12, its function as cadential is fully realized. Beyond the various complexities of melodic implication arising within the antecedent phrase is the curious fact that the entire A section bears no relation to the ‘period model’ found in most of Mozart’s period forms: neither the antecedent nor the consequent allude to the basic model, in that a structural ascent from 1ˆ to 3ˆ is entirely missing from both phrases. Moreover, the bass melody of the consequent phrase conforms almost identically to the antecedent, which, as discussed earlier, is atypical of period forms.40 The absence of both basic and period models from the A section provokes powerful expectations that elements of one or both models will appear later. Following the B section,41 the opening phrase of the A′ section conforms entirely to the consequent of the A section, thus avoiding yet again the basic model (or any variant thereof). The imperfect authentic cadence closing this phrase motivates an extension in bars 17–18, and it is only here that the bass line finally brings the complete basic model in its simplest, unembellished form. As a result, this extension, rather than being a mere appendage, becomes a central structural element of the theme.42 We thus see that this theme, despite its extensive use in the analytical literature as a representative of ‘normal’ practice, is actually quite deviant from the perspective of bass melody developed in this study.43 One would like to believe that Mozart’s ‘ideal’ listener would perceive (albeit at an intuitive, unconscious level of experience) the absence of the basic model throughout the theme and would then relish its last-minute appearance in the extension. If so, then we can confidently speak of a powerful communicative strategy at play on the part of the composer. The theory of bass melody proposed in this study is still at a nascent stage of development. A number of important issues touched upon here call for additional research. Let me conclude by outlining five topics that require further

Schoenberg’s ‘second melody’

work. (1) The role of register needs to be made considerably more precise: at times, the analysis needs to respect the exact pitch of a bass note, especially when helping to clarify the different streams; at other times, however, it seems that a ‘pitch-class’ representation makes the melodic relationships more evident. (2) It remains unclear into which stream the bass notes of sequential harmonic progressions should be placed. Since sequences normally play no role in cadential articulations in the classical style (though they sometimes do so in romantic styles), their supporting bass notes would seem not to belong to the cadential stream. Unless we are to posit a separate, third stream to hold these pitches, they seem best to belong to the prolongational stream. (3) More work needs to be done to clarify just what constitutes a bass motive and how such motives might function within the line. On the basis of preliminary research, motivic play appears infrequently in works by Mozart, somewhat more often in those by Haydn, and is decidedly more present in Beethoven. The validation and significance of this observation obviously requires additional study. (4) Basses with considerable leaping motion do not assimilate well to my models, so more satisfactory ways of accounting for such melodies need to be found. I have observed some situations in which leaping bass lines can be seen to use up, in a systematic way, the seven pitches of the diatonic scale. Perhaps some modified notion of ‘aggregate completion’ could be a useful concept to invoke in such cases.44 (5) Limitations of space prevented me from presenting and illustrating a second basic model – one that features descending motion in the prolongational stream (from 1ˆ down to 5ˆ ). Such a model seems particularly appropriate to bass lines of baroque styles, though some classical works exhibit features of this model as well.45 I opened with Schoenberg’s advice to the budding composer and used his remarks as a stimulus for exploring an avenue of research largely ignored by music theorists. Let me close with my own brief words for the budding theorist: allow yourself at times to become ‘Meyer-ed’ in the bass line; and always keep a watchful eye on that ‘second melody’.46

Notes 1 Arnold Schoenberg, Fundamentals of Musical Composition, ed. Gerald Strang and Leonard Stein (London: Faber and Faber, 1967), 118. 2 Important research on partimenti treatises of earlier centuries is currently being conducted by Robert O. Gjerdingen, Giorgio Sanguinetti and Ludwig Holtmeier. Gjerdingen has already begun a large-scale project of collecting

181

182

william e. caplin

3 4

5

6

7

8

9 10

11

these treatises on a dedicated website (http://faculty-web.at.northwestern.edu/ music/gjerdingen/index.htm). Schoenberg, Fundamentals, 117. Reed J. Hoyt, ‘Harmonic Function and the Motion of the Bassline’, Journal of Music Theory Pedagogy 4 (1990), 147–90; David Pacun, ‘Scanning Bass Patterns: A Middleground Path to Analysis’, Journal of Music Theory Pedagogy 17 (2003), 59–77. I am referring here to Schenker’s mature theory, as normally practised by theorists today. Schenker’s conception and analysis of the bass line evolved over time, however. As William Rothstein notes (in a personal communication): ‘In the Teens and early Twenties, Schenker’s paradigm of bass motion is the Stufengang, the literal or implied chain of harmonic roots – essentially a fundamental bass – that derives from Rameau via Sechter. From the middle Twenties, his underlying model becomes increasingly the Bassbrechung, as described in Free Composition. In Der Tonwille, which is where the two conceptions rub against each other, Schenker seems closest to your [Caplin’s] way of thinking about bass lines.’ A full study of Schenker’s views on bass melody would make a fascinating research topic of its own. Meyer presents the basics of his theory of melody in ‘Part Two – Explorations: Implication in Tonal Melody’, Explaining Music: Essays and Explorations (Berkeley: University of California Press, 1973). The notion of ‘gap-fill’ melodic organization is advanced prominently in Meyer’s melodic theory; see Explaining Music, 145–57. A recent perceptual/cognitive study has challenged the validity of the concept as a model for melodic classification (Paul von Hippel, ‘Questioning a Melodic Archetype: Do Listeners Use Gap-Fill to Classify Melodies?’ Music Perception 18 (2000), 139–53), but this work does not necessarily invalidate the use of gap-fill principles in the analysis (as opposed to the classification) of melodic structures. Throughout his writings, Rameau notates the fundamental bass as a series of pitches residing in a single staff; see, for instance, Example ii.10 from JeanPhilippe Rameau, Treatise on Harmony, trans. Philip Gossett (New York: Dover Publications, 1971), 86. The example cited in the previous note clearly distinguishes basso continuo from fundamental bass as separate voices. The idea of a single voice projecting multiple implied voices is akin to the notion of compound melody, as exemplified by the subject of the C minor Fugue in Bach’s Well-Tempered Clavier, book 1. Most instances of compound melody see the line regularly leaping back and forth from one implied voice to the next, as in this fugue subject. The kind of two-voiced structure I propose for the bass, however, sees a less frequent alternation between the implied voices compared to most compound melodies. Though there are some similarities, my use of ‘stream’ does not refer to how that term is employed by Albert Bregman in his theory of ‘auditory stream

Schoenberg’s ‘second melody’

12

13

14

15

16

17

18

segregation’ (Auditory Scene Analysis: The Perceptual Organization of Sound (Cambridge, MA: The MIT Press, 1990)). This labelling conforms to a fundamental distinction, which I have been promoting throughout my writings on musical form, between prolongational harmonic progressions, used to begin thematic units, and cadential progressions, used to close such units. (A third category includes sequential harmonic progressions.) See William E. Caplin, Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven (New York: Oxford University Press, 1998), ch. 2; William E. Caplin, ‘The Classical Cadence: Conceptions and Misconceptions’, Journal of the American Musicological Society 57 (2004), 69–70. This model of bass melody resembles some of the first-level middleground patterns of bass motion given by Schenker in Free Composition, trans. and ed. Ernst Oster (New York: Longman, 1979), Figure 14.2. Schenker’s figure suggests that the melodic motion in the bass is an embellishment of a more structural bass ˆ 3– ˆ 5ˆ ). Furthermore, the figure does not distinguish between the arpeggiation (1– ˆ 3ˆ (as, say, prolongational) and 3– ˆ 5ˆ (as cadential). motion 1– That the two patterns stand in a retrograde relation to each other seems, to me, incidental, though perhaps there is some further theoretical significance to this relationship that others may wish to explore. My definition of period form, derived from Schoenberg, can roughly be stated thus: an antecedent phrase with weak cadential closure (usually a half cadence) is followed by a consequent phrase, which begins like the antecedent, but which closes with a stronger cadence, usually a perfect authentic cadence; see Caplin, Classical Form, ch. 4. That the bass line of the antecedent differs considerably from that of the consequent reveals that the two phrases are not as parallel in construction as they often appear to be when focusing attention on the soprano melody alone. The model also reveals that the half cadential progression of the antecedent is normally shorter than the authentic cadential progression of the consequent; see Classical Form, 53. In the examples that follow, a basic analysis of the bass-line melody is placed below the musical score; this analysis is reproduced, along with additional bass-line analyses where necessary, in separate staves (lettered a, b, etc.). Unlike a consequent, which brings back the opening two-bar basic idea of the antecedent (thus suggesting a repeat of that earlier phrase), a continuation typically brings a combination of characteristics including fragmentation into shorter units (usually one-bar groups), an acceleration of the harmonic rhythm, faster surface durational patterns and sequential harmonic progression; Classical Form, 40–2. The combination of an opening antecedent phrase with a following continuation creates a hybrid theme-type (one that combines elements of both period and sentence); see Classical Form, ch. 5.

183

184

william e. caplin

19 Though 5ˆ normally appears in the cadential stream, it may sometimes participate in prolongational activity when supporting a passing chord between the harmonies placed over 4ˆ and 6ˆ or when supporting a tonic harmony in second inversion (as here). 20 The opening B♭, though not literally present in the score, is obviously implied as the bass support for the initial F in the soprano voice. 21 Note that the linear relationship between 2ˆ and 3ˆ shown on staff d clearly crosses over a natural prolongational boundary (as shown on staff c), an issue I alluded to in the introduction. Observe, furthermore, that if the double-neighbour configuration had been inverted, that is, B♭–A–C–B♭, then there would be no ˆ Perhaps this directed motion from 1ˆ to 2ˆ that would imply a continuation to 3. ˆ 2– ˆ 7ˆ –8ˆ is why the opening double neighbour of so many classical bass lines is 1– 22 23

24

25

26

and not the inverse. The use of a pre-dominant prior to the dominant of the half cadence represents a variant of the period model shown in Example 6.7. On the formal significance of an expanded cadential progression, see William E. Caplin, ‘The “Expanded Cadential Progression”: A Category for the Analysis of Classical Form’, Journal of Musicological Research 7 (1987), 215–57. Note that the voicing of the passing v7 appearing on the third quaver of bar 6 and the cadential v7 on the downbeat of bar 15 are practically identical, thus allowing us to hear a strong connection between these chords despite their differing harmonic functions. The formal organization of this theme is based on a hybrid theme-type, one consisting of a compound basic idea followed by continuation; see Caplin, Classical Form, 61. Had the theme cadenced in bar 8, as proposed above, it would have conventionally realized this theme-type. But the extension of continuation function in bars 7–10 and the expansion of the cadential function in bars 10–13 render the theme considerably looser and less conventional in design. All modern editions of this quartet indicate that bars 24–5, which follow the passage shown in Example 6.12, bring back the two-bar idea from the opening bars of the piece, which is then followed by a variant of bar 3, now harmonized with v7 of v. As a result, the bass line reproduces, indeed reinforces, the effect ˆ But in a personal communication, of bars 17–21, where 1ˆ moves directly to 2. James Webster, editor of this quartet for the Joseph Haydn Werke (Munich: Henle Verlag), reveals that the original autograph (as well as many contemporary prints) brings the note E, as a series of quavers, in the violoncello part of bars 24–5 (instead of G, as in the modern editions); thus the bass motion would ˆ This ‘original’ version see a more conventional foreground move from 6ˆ to 2. is not entirely unproblematic, however, since it results in a rather unusual reharmonization (with vi) of the opening two-bar melodic idea, such that it is understandable that subsequent editions could be seen to correct a perceived mistake of the original.

Schoenberg’s ‘second melody’ 27 Staff b suggests a broad prolongation of 1ˆ throughout bars 1–2 until the leap to 3ˆ at bar 3. Within these opening three bars, it may also be possible to discern elements of a Romanesca (or ‘Pachelbel canon’) bass-line pattern, as shown in staff c. 28 See Caplin, Classical Form, 59–60 (Example 5.1), for a formal analysis of the opening eight bars of Example 6.13 as a hybrid theme. 29 I am considering this parenthesis largely from the melodic perspective of the bass line itself; interpreting the whole of bars 17–18 as parenthetical is less compelling, though a direct connection can be made by changing the downbeat of bar 19 into a cadential six-four and linking the music from this point back to the end of bar 16. 30 I briefly return to the issue of sequential progressions in the conclusion to this study. 31 The overall movement is in large-ternary form (see Caplin, Classical Form, ch. 14); the second part, which typically is formed as an interior theme (Classical Form, 212–13), is replaced by a genuine development section (Classical Form, ch. 10). 32 Though everything is concluded from a purely formal point of view, we might be dissatisfied that the movement is ending in such a high register, given its much lower opening registral position. 33 Of course, a given implication for continuation might never be realized within the musical work; see Meyer, Explaining Music, 117. 34 I thank Danuta Mirka for pointing out (in a personal communication) this distinction between what I am calling immediate and deferred realizations. She further speculates that this distinction may be rooted in different modes of music cognition. 35 Like Wye Allanbrook (in her essay in this volume), I believe that much can still be said about this theme; and like her, though for different reasons to be explored below, I regard the theme as an exceptional representative of the ‘period’ theme-type and hardly to be taken as a model. 36 In order to avoid an overly detailed description, I am overlooking the embellishing upper-neighbour motions in the first half of bars 1 and 2. 37 This deferral can be related to the influence of the siciliano topos, as identified by Allanbrook. 38 The passage features other immediate realizations, such as the resolution of the leading-tone to tonic on the downbeat of bar 4 and the cadential move from 4ˆ to 5ˆ in that same bar. Such realizations, however, are so direct as to attract little attention. For that reason, immediate realizations are less a source of communication than deferred ones. 39 To be sure, the expectation of 7ˆ moving to 6ˆ is conceived here from a purely melodic point of view. When the broader harmonic context is taken into account, the v6/5 supported by 7ˆ demands resolution to i, which immediately occurs, ˆ thus fulfilling the stronger expectation that 7ˆ will return to 1.

185

186

william e. caplin

40 Recall again that unlike the soprano melody, whose consequent more or less repeats the antecedent, the bass line of most periods is usually altered in the consequent relative to the earlier phrase. 41 The B section is also unusual in that it emphasizes tonic in the bass, rather than dominant (see Caplin, Classical Form, 75). 42 Indeed, this extension then goes on to play a major role in the following variations. For Mozart frequently sets up a textural and dynamic contrast between the antecedent and consequent phrases of the A section, which he then matches with a comparable contrast between the opening phrase of the A′ section and the extension (see Variations 1, 2, 3 and 6). 43 See Allanbrook’s essay for an extensive discussion of the use of k331 as an analytical paradigm. 44 James Baker develops the idea that aggregate completion can play a role in music of this style period; see ‘Chromaticism in Classical Music’, in Music Theory and the Exploration of the Past, ed. David Bernstein and Christopher Hatch (Chicago: University of Chicago Press, 1993), 233–307. 45 See, for example, Mozart, Piano Sonata in A minor, k310/i, bars 1–9. 46 Support for this research was provided by the Social Sciences and Humanities Research Council of Canada. I want to thank Erin Helyard for his help in preparing the musical examples.

pa rt i i i

Rhetorical form and topical decorum

7

A metaphoric model of sonata form: two expositions by Mozart michael spitzer

This chapter proposes a new theory of the sonata-form exposition from the viewpoint of contrast rather than from the more common perspectives of unity, symmetry or balance. Expositions end differently to how they begin, and so do sonata forms. Why, then, does music analysis, with an eye to tonal return, celebrate sonata form chiefly as a paradigm of conflict resolution, as epitomized in the dynamics of recapitulation?1 An equally pertinent issue is surely the formal contrast within an exposition, and moreover, the kind of contrast which survives into the recapitulation, when the modulation to the second group is diverted into the tonic. Thus a level of contrast more fundamental than the tonal rhythm of modulation and resolution which has occupied sonata theory since Tovey and Rosen. At this level, I will argue that an exposition divides into two groups because it needs to say the same thing twice – it must present its material in two ways, the first time conventionally, the second time distinctively. I call this the ‘two-time rule’, a rule which attends to the dialectics of Enlightenment semiotics, whereby convention is equally the vehicle for communication and its straitjacket. Given that communication involves schemata, it entails what psychologists identify as a dilemma between gist and detail in schematic perception. Much of this is common sense. On the one hand, something will not be perceived or remembered if it cannot be fitted into a stereotypical pattern.2 On the other hand, the better things do fit a pattern the less well they will be perceived or remembered. In apprehending the features as mere variables, we don’t attend to them as specific events. The data’s specificity must therefore be grasped later, via a second glance. This ‘second glance’ is afforded by the second group. My objective in this chapter is to suggest ways in which we can gauge the functional differences between first and second groups in the classical style – these first and second ‘glances’. I will focus on Mozart, since thematic contrast is particularly characteristic of his expositions.

The theory Metaphor theory is a rich, multi-dimensional field. The final section of this chapter describes what I see as the three most vital aspects of this theory

190

michael spitzer

for music. Deferring background discussion till the end, I will propose that expositional contrast in sonata form can be illuminated from the following three distinct yet interdependent ‘metaphorical’ perspectives: (1) The second glance is figurative in itself, in its lyrical or expressive character. (2) The second glance is also figurative relationally, because it reflects upon previous music in such a way as to bring out its ‘materiality’. That is, material is initially assimilated into function, so that it is ‘transparent’. It is then showcased as an object of interest in its own right, becoming ‘opaque’. (3) This second glance involves a shift in grammatical or conceptual structure. The latter, cognitive, side to metaphor is interdependent with the first two: it is this very shift in structure which occasions them. A metaphorical standpoint transforms how we view the exposition. My central idea is that the second subject marks the intersection of two great arcs of motion: it is the goal of the first group, and the source of the second group. These two arcs are both processes of emergence: respectively, of thematic substance and cadential structure. Borrowing terms from Schoenberg, I title them ‘thematic liquidation’ and ‘cadential liquidation’. I endeavour to keep in play three historical time-frames: eighteenth-century formal theory (particularly Koch); Schoenbergian Formenlehre (particularly its most recent manifestation, Caplin); and contemporary cognitive approaches to metaphor. There is also the key question of generality: how widely applicable is my metaphoric model? A general survey of sonata practice is well beyond the scope of this chapter. My model of the exposition is thus nearer to a hypothesis than a theory. What, for instance, about Haydn, whose famously irregular expositions favour monothematicism over thematic contrast? I would predict that a metaphoric approach would highlight the way that similar materials are disposed differently across the two halves of the exposition. Even so, the model proposed in this chapter fits best the type of sonata form exposition cultivated by Mozart. Moreover, my chapter pertains only to a late subset of Mozart’s music; to a strategy he invents in 1787. The strategy I will describe (its aspects include unifying the tonic group into a sentence phrase, and ‘flipping’ the dyad figures which end subphrases in the tonic group into second-group motives) clarifies and tightens up the metaphoric model. But it is not foundational to this model. Rather, we may keep in mind the paradox that simplicity or parsimony in music of the first rank is often an emergent property – an extreme case of a general principle.

A metaphoric model of sonata form

1 Articulation Theoretical sources such as Heinrich Koch’s Versuch einer Anleitung zur Composition (1782–93) explain how classical form was organized around quasi-metrical patterns of articulation, enshrined in phrase-endings or caesuras.3 According to Koch, a motivic idea, or Einschnitt, was derivative of the phrase, rather than a germinal starting-point. Indeed, the process whereby a composer gradually extracts the melodic substance from a (normatively four-bar-) phrase by reducing it from four bars to one, thereby eliminating redundancy, is the origin of Schoenberg’s notion of liquidation within the sentence structure.4 While this held for the first group of sonata form, and thus prototypically for the entire style, the practice in the second group was marked against this norm. Here, Einschnitte assume a generative, seminal function, subject to various techniques of repetition and transformation. To a much greater extent than first groups, classical second groups are typically dedicated to the repetition and elaboration of a characteristic motive.5 It is thus possible to theorize a qualitative difference in the way themes operate in the two halves of an exposition, roughly akin to the cognitivist distinction between ‘top-down’ and ‘bottom-up’ processes (or ‘concept-driven’ and ‘data-driven’). This qualitative difference operates irrespective of specific thematic correspondences within the piece. Nevertheless, I would argue that a second group is ‘figurative’ also in this reflective sense, presuming an over-arching thematic discourse within the exposition. If so, how do we square this thesis with Mozart’s liking for thematic contrast, compared with Haydn’s monothematicism? Furthermore, Mozart’s critics have tended to downplay the role of motivic development in favour of a more fragmentary design, suggestive of collage or mosaic, following a traditional opposition between Entwicklung and Fortspinnung. Whereas Haydn generally submits his material to teleological development (Entwicklung), Mozart prefers to re-arrange independent fragments in a musical kaleidoscope (Fortspinnung).6 But why should the kaleidoscope not turn in such a way as to resolve or bring out aspects of the first group? Thinking of Mozart’s ‘themes’ as force-fields of competing materials, rather than as monolithic objects, re-admits a dimension of purposive change, ostensibly taken away when we rule out notions of ‘development’. If the first group arranges its materials in a harmonious conflict, then the function of the second group is to change the terms of this conflict; to reverse the balance of power; to flip the figure/ground relationship. In a nutshell, the second group lets us ‘see’ aspects which the first group assimilates into conventional function, thus giving us a second ‘glance’.

191

192

michael spitzer

Example 7.1 Caplin’s analysis of Beethoven’s Piano Sonata in F minor, Op. 2 No.1/i, bars 1–8 (Classical Form, 10, Example 1.1)

2

Sentence and liquidation

With a language as hierarchical as the classical style, it is no surprise that these principles are already operative at the level of the phrase; they are even implicit in the way Schoenbergian Formenlehre defines its succession of structural functions. I want to focus on two elements of this formal tradition: sentence phrase-structure, and its attendant dynamics of liquidation. I think that the enormous power of these concepts has yet to be fully exploited. A good starting point is William Caplin’s Schoenbergian analysis of the opening sentence from Beethoven’s Piano Sonata in F minor, Op. 2 No.1, a locus classicus which rewards still further scrutiny (see Example 7.1).7 Divided into the three functions of ‘presentation’, ‘continuation’ and ‘cadential idea’, the analysis at first seems to refute the ‘two-time rule’, according instead with the ternary rhetorical succession of beginning–middle–end.8 In fact, the narrative is more binary than it appears, since Caplin stipulates that ‘the two functions of continuation and cadential normally fuse into a single “continuation phrase” in the eight-measure sentence’.9 Although bars 7–8 in Beethoven’s phrase are perfectly conventional, and taken in themselves could well be imagined to close ‘any number of themes from different works’, taken in context the cadential idea ‘grows naturally out of the preceding measures’.10 Beethoven perpetrates the illusion that the end of the phrase is both conventional and thematic. From this angle, Schoenberg’s concept of ‘liquidation’ is similarly dual, the better-known definition of ‘gradually eliminating characteristic features until only uncharacteristic ones remain’11 completed by the rider that the material is not so much stripped away as condensed: ‘Condensation implies compressing the content of the model, whereby even the order of the features may be somewhat condensed.’12 Beethoven’s illusion, then, is a semiotic one: the material may be ‘eliminated’ in practice, but it is ‘condensed’ in principle into a motivic detail – what Schoenberg terms a ‘residue’ – which functions as a musical sign. The signification of a phrase by a single note is an axiom of Riepel

A metaphoric model of sonata form

Example 7.2 Mozart, Piano Sonata in G major, k283/i, bars 1–10

and Koch’s theories of phrase-ending, whereby the meaning of a phrase was expressed by the tonal orientation of its final note (Koch labels a phrase either a Grundabsatz or a Quintabsatz, depending on its ending).13 Schoenbergian semiosis is of a more formal sort: a motivic detail is felt to represent an extended time-span of material. The resultant compaction of a whole into its part is responsible for the vertiginous intensity induced by liquidation. But it also allows the liquidation process to continue into the cadential part of the theme; the quavers in bar 7 of Beethoven’s sentence function simultaneously as steps of a stereotypical cadential descent, and a climax of the liquidation’s acceleration of phrase rhythm. The passage’s dual nature (cadential, motivic) gives the sentence a circular trajectory, returning to the stable and conventional (‘tight-knit’) form of the presentation phase, as much as it continues the free (‘loose’) form of the continuation. At its goal, the sentence filters liquidation through conventionality; it is a ‘second glance’ on the opening, extracting and recontextualizing its motivic kernel.14 It is in this respect that Schoenberg’s term ‘residue’ is a misnomer, since it does not express the sense that the motivic kernel was always there, at the start of the process. Only, it takes the completion of the sentential process to reveal it. This notion of kernel as starting-point, rather than outcome, is clearer when we turn from Beethoven to Mozart. The role of conventionality is predictably stronger when we go back in time and examine another locus classicus of style analysis, the theme of Mozart’s Piano Sonata in G major, k283 (see Example 7.2).15 The theme’s conventionality is clearest in the first four bars, which instantiate what Gjerdingen calls a ‘1–7 . . . 4–3’ schema, by virtue of the two semitone dyads which articulate the subphrases. By investing so much power in the signification of the dyads, Gjerdingen’s modern theory recalls Riepel and Koch’s ethos of punctuation – their notion of ‘phrase-ending formulas’. Mozart’s dyads

193

194

michael spitzer

function not motivically, like the turn figures in Beethoven’s theme, but as ˆ 7ˆ to cognitive cues, so that a competent listener would be primed by the 1– ˆ ˆ expect a 4–3 consequent. Strikingly, their function reverses with the onset of the continuation phrase at bar 4, when the dyads are reinterpreted as accented appoggiaturas. They are now heard in themselves, as expressive pianti figures, rather than as abstract signs.16 Technically, what Mozart has done is scramble the schema’s ‘congruence’ and ‘conformance’: parameters which should march together are sent their separate ways, and the questionanswer periodicity is dissolved.17 Thus the melodic F♯ at bar 5 clashes with the harmony’s G; the dyad is repeated directly, at bar 6, transforming its function from phrase-ending to a seed of a motivic sequence; the pattern of falling tenths across bars 5–6 is broken when the bass continues its descent to A while the melody suspends the D for a further bar and a half, yielding to C on the weak beat of bar 8; the irregularity of a six-bar continuation/cadence to a four-bar presentation is compounded when the entire consequent is repeated in bars 10–16. The effect of this general disorder is to reveal features which were immanent within the schema of bars 1–4, but assimilated into its syntax, and thus transparent: these features, in addition to the dyads, include the descending scale, repeated notes and arpeggiations (see Example 7.3). If, by definition, a schema comprises a coordinated set of melodic, harmonic, and metrical movements, then to ‘realize’ the material means to liberate it from this patterning. Leonard Meyer, the theorist who first brought Mozart’s schema to critical attention, confines ‘realization’ within his system of pattern-continuation and gap-fill motion. But I would argue that the undoubted gap-fill across bars 5–10 falls against a broader critical horizon. Paramount here are the ‘functional flips’, as it were, of the dyads from phrase-endings to motives, and the recuperation of musical material from schematic function (see Example 7.3b). This dialectic of convention and particularity is far weaker in Beethoven’s sentence, which is driven primarily through musical logic. For instance, the turn figures at bars 2 and 4, which generate Beethoven’s liquidation, are not phrase-endings but motives – the process is motivic from the start. If Beethoven’s liquidation is geared toward dramatic intensification, then Mozart’s serves a more leisurely, albeit equally radical, aesthetic of unmasking.

3 Mapping from phrase-level to exposition Working our way back up from phrase-level to large-scale form, it is possible to show how these processes unfold across the exposition as a whole. The key to doing so is provided by the affinities Caplin elegantly demonstrates

A metaphoric model of sonata form

(a)

(b)

Example 7.3 Materials of the theme from Mozart’s Piano Sonata in G major, k283/i, (a) the presentation phrase (b) the continuation phrase

between intra- and inter-thematic functions – the functional constituents of individual themes and the arrangement of those themes – both of which are governed by a syntax of presentation, continuation and cadence. Three levels of syntax are immediately evident: the theme; the group; the exposition. The two groups each express the three functions of presentation, continuation and cadence within themselves. The transition, which can divide into three parts on its own account, also mediates between first and second groups as large-scale versions of these functions (thus first group = presentation; transition = continuation; second group = cadential). Whereas this functional scheme is implicit in the sonata principle in general, it becomes explicit in Mozart’s mature instrumental music. To this end, Mozart invents two formal gambits: (i) He casts the first group as a single phrase – a sentence – which can thus function as the presentation phrase to the transition’s continuation; (ii) Mozart clarifies the second group’s threepart structure. (i) As we will see, the expositions of k515 and k551 expand their first group into a single sentence structure, directed to a single imperfect cadence. By the same token, the transition becomes a modulating consequent

195

196

michael spitzer

phrase: it answers the tonic-group ‘sentence’ like the consequent of a higher-order period, although as another sentence, which this time modulates. The breadth of this ‘phrase rhythm’ makes the parallel between liquidation and transition especially audible. Strikingly, none of Mozart’s string quartets or piano sonatas do this, and none of the symphonies prior to the last three of 1788, although earlier pieces anticipate certain aspects. For instance, the tonic group of the ‘Dissonance’ Quartet, k465 (bars 23–44) comprises two sentences, disposed as antecedent and consequent; the tonic group of k387 does comprise a single sentence (bars 1–10), but one which ends with a perfect cadence. Mozart’s later discovery that tonic group and transition could be coupled as antecedent and consequent was one of the most important lessons he taught Beethoven about formal expansion.18 (ii) In k515 and k551, Mozart clarifies the isomorphism between first and second groups by bringing out the latter’s three-part structure. Accordingly, the second subject (the presentation phase of the second group) compactly echoes the first group as a whole by ending on an imperfect cadence, and is answered by a disproportionately long consequent. The middle phase of the second group (its ‘continuation’) thus corresponds to the transition, the difference being that it doesn’t modulate. It compensates for this tonal stasis by being the most dissonant and asymmetrical stretch of music in the exposition as a whole – a far cry from the first group’s schematic transparency. A juncture as distinctive as this is in need of a name: I christen it the ‘expositional nadir’. The second group’s third phase brings in a new theme, which combines aspects both of the transition’s ‘standing on the dominant’ and the second subject itself. This theme evinces the descending linear progression characteristic of cadential passages.

4 Structural liquidation The thrust of these formal gambits, I would argue, is to commute the first half of the exposition (first group and transition) into a higher-order sentence. This sentence climaxes with what I term ‘structural liquidation’, which unfolds in the formal space of the transition/standing-on-the-dominant, before being re-packaged in the presentation phrase of the second group – the second subject. I propose that a complementary process of ‘structural liquidation’ takes place across the second group as a whole. Unlike ‘thematic liquidation’, which embraces the foreground motivic-thematic substance of the theme, ‘structural liquidation’ involves plotting middleground

A metaphoric model of sonata form

harmonic structure on a series of cadential progressions. I thus see the exposition as two complementary liquidations, intersecting on the second subject. In other words, the second subject has a ‘dual function’ as both the goal of the first group, and the starting-point of the second group. As the goal of the first group, it is the culmination of thematic liquidation. As the head of the second group, it initiates a process of cadential liquidation. The second subject’s dual aspect extrapolates the sentence’s own hybridity, its fusion of continuation and cadence functions. It also expresses the transition’s double trajectory, both journeying into new regions, and re-visiting the material of the exposition’s opening.

(i) Second subject as goal of ‘thematic liquidation’ In its capacity as a large-scale liquidation, the transition ‘realizes’ the material of the first group, before this material is repackaged as a well-formed (‘tight-knit’) theme in the presentation phase of the second group. Otherwise put, the liquidation process actually runs on into the second subject, and is responsible for its characteristic dual aspect of busy activity combined with relaxed (‘lyrical’) phraseology. This continuity is generally overlooked by theorists, who have focused rather on the formal and tonal separation between the two junctures. Thus the goal of a transition is, in the first instance, the dominant plateau which Caplin terms the ‘standing on the dominant’.19 This local goal then hands over to the higher-order goal of the second group, when the ‘standing on the dominant’ resolves into the new tonic.

(ii) Second subject as head of ‘cadential liquidation’ Like the second half of a sentence, a second subject submits primary material to a ‘functional flip’, transforming phrase-endings into motives. These consequently behave as seeds for a thematic process, a process governed nevertheless not by development so much as the variational, repetitive logic of cadential emphasis. The popular image of the second group as an expanded cadence has recently come under fire, not least from Caplin.20 Laying this debate to one side for the moment, one must surely concede that a variational logic is inevitable, given that the second group can’t properly go anywhere: unlike the first group, it doesn’t modulate. Historically, this variational logic was modelled as a gradation of cadential closures, according to the eighteenth century’s linguistic metaphor of a series of punctuation marks.21 Echoing the venerable topos of the ‘stream of rhetoric’, theorists

197

198

michael spitzer

imagined the musical discourse as flowing towards the full stop of a perfect cadence, interrupted on the way by the ‘colons’, ‘semicolons’ and ‘commas’ of weaker cadences. Against this tradition, Caplin is anxious to safeguard the cadence’s positional status as coming at the end of a theme, and, by the same token, to preserve the functional distinction between themes and cadences proper.22 Without contesting this distinction, I would nevertheless add that the tonal models which underlie the phrases of a second group may be plotted on the same functional spectrum as cadences, since they involve the same grammatical harmonies, albeit at a middleground level. The emergence of foreground cadence out of middleground tonal models can thus be conceived by analogy to how sentential liquidation realizes thematic material. We can term this process ‘cadential liquidation’, in contrast to ‘thematic liquidation’. The advantage of this concept is that it makes sense of those ‘post-cadential’ v–i repetitions which permeate codas as residual kernels. Even though these v–i repetitions typically elide subdominant-type auxiliary functions, thus forfeiting their cadential identity, they continue the cadential function through the logic of liquidation. The concept of ‘cadential liquidation’ allows us to re-configure the ‘drive to the cadence’ according to a ‘depth’ (rather than kinetic) model. That is, we are talking now about the emergence rather than the approach of cadence. The dispute about whether or not the second group is an expanded cadence thereby admits of a new solution. The functional burden of the group begins as thematic but ‘modulates’ to the cadential.

5

Summary: ‘material’ and ‘syntactic’ second glances

Viewing the exposition as a complementary pair of liquidation processes, arcing across the first and second groups, highlights the dynamic and transformative trajectory of sonata form. What, in particular, is ‘transformed’? Change is multi-levelled: it occurs within the two groups; between them as a whole; and between various ‘nodal’ junctures (first and second subjects; second subject and coda theme; coda theme and da capo of first subject). More broadly, change is both ‘material’ and ‘syntactic’.

(i) Material transformations The second group in k515 and k551 recuperates various aspects of the first group at different junctures, according to the following pattern. The presentation phrase (second subject) takes up the ‘residue’ dyads from the transition, and submits them to a ‘functional flip’, transforming punctuation

A metaphoric model of sonata form

phrase-endings into motivic line. The continuation phrase (or expositional nadir), being the loosest and most noncongruent part, showcases the first subjects’ ‘buffer’ material (arpeggios in the quintet; fanfares in the symphony). Conventionalized material is rendered ‘opaque’. Finally, the coda themes recall the surface of the first subjects, but re-configured according to second-group syntax. The sudden switch from second group to first, after the da capo, creates a jolt of syntactic type. The overall impression is twofold: of the emergence of suppressed material, according to a depth model; and of reflection upon opening material across great distance.

(ii) Syntactic transformations We can now think of sonata form as a process through which opening material is gradually weaned off a backward-facing thematic function and onto a forward-facing cadential end.23 This functional reorientation from retrospective to prospective is itself the trace for a still richer qualitative difference between the workings of the two groups. The difference concerns the nature of musical syntax. Under the umbrella of its schematic opening, the first group moves teleologically from source to goal along a conventionalized pathway (paradoxically, flexing away from the opening). ‘What happens next’ is much less predictable in the second group, which is ruled instead by the additive, chain-like logic of variation, with the source (or ‘prototype’) of the cadential model emerging only at the end. Although the second group is broadly goal-orientated, progress towards this goal is not a matter of ‘motion’ so much as focalization around material, stripping away and unmasking. Technically, the first group is ‘schematic’ and ‘hypotactic’, whereas the second group is ‘radial’ (around prototypes) and ‘paratactic’ (I will explore these terms later).24 This, then, is how the second group is a ‘second glance’ at the first group, superseding the latter’s abstract, schematic syntax with a syntax of materiality, circling around categorical prototypes. In practical terms, the second group as a whole is a kind of variation, or ritornello, on the first group.25 Mozart’s ‘metaphoric model’ of the exposition can be summarized in the diagram on page 200 (Figure 7.1). The figure indicates the isomorphism between the two tripartite groups, each of which traces the pattern of ‘presentation-continuation-cadential’. It shows the second subject to be a pivot between a ‘thematic liquidation’ (orientated backwards, to the opening material) and a ‘cadential liquidation’ (orientated forwards, to the final authentic cadence). And it represents the function of the two groups as complementary between ‘hypotaxis’ and ‘parataxis’.

199

200

michael spitzer

Figure 7.1 Mozart’s ‘metaphoric model’ of the sonata-form exposition

Mozart’s interest in casting the first and second groups along a similar pattern can be related to his love of ritornello form. It is a sign of the feedback of his concerto experience into mainstream sonata forms after actual concerto composition began to wane (after 1786). It also signals the expansion of his Fortspinnung principle to embrace larger blocks of musical structure. Written between 1787 and 1788, k515 and k551 are amongst the finest fruits of this crossover.

String Quintet in C major, k515/i This is Mozart’s longest sonata-form exposition, and the longest before Beethoven’s ‘Eroica’. The breadth of conception is surely dependent on the simplicity of material: Mozart has boiled things down to their fundamentals. The first subject comprises an unusually transparent schematic sequence, ˆ 3ˆ figure. I articulated by dyads; the second group is based on a repeated 4– will show that the difference between the two groups consists of opposite perspectives on phrase-ending dyads. Dyads as articulation of a hypotactic schematic progression yield to dyads repeated paratactically as a ‘cantus ˆ 3ˆ 4– ˆ 3ˆ chain’. The joke – or genius – of firmus’ idea – an idea I will call the ‘4– ˆ 7ˆ in C Mozart’s strategy is that it pivots on reinterpreting the same pitches: 1– ˆ 3ˆ in G. Another extraordinarily neat major comprise the same pitches as 4– conceit is that Mozart underscores the second group with pastoral topics, in the guise of musette pedal points.26 Given that the task of the second group is to realize the ‘nature’ of first-group material, what could be more natural than performing this as pastoral? Now, a pastoral topic is not mandatory here (the parallel point in the ‘Jupiter’ is made in ‘singing style’); it is merely

A metaphoric model of sonata form

Example 7.4 Mozart, String Quintet in C major, k515/i, bars 1–20

appropriate. But this does suggest that, occasionally, the choice of topic is motivated by syntax, rather than being an arbitrary condition of ‘play’.27 The motivation in k515 is another function of its breadth. To undergird the enormous arcs of tonal motion, Mozart needs to dig into the foundations of his language.

Sentence The tonic group is integrated into an expansive twenty-bar (nineteenbar, with one-bar rest) sentence, with the transition starting bar 21 (see Example 7.4).

201

202

michael spitzer

With its transparent dialogue between cello arpeggios (‘Mannheim skyrockets’) and phrase-ending dyads, the phrase is a supreme example of schematic regularity and predictability. At a broader level, it epitomizes the goal-oriented syntax of the exposition’s first half, the tonic group-cumtransition. Likewise, the sentence’s liquidation foreshadows processes which will unfold across the exposition as a whole. (Note that the onset of the sentence continuation is actually ambiguous, since sequence is already intrinsic to the presentation phase itself.) These take the form of three reversals – subverting the music’s syntax so as to reinterpret its material. 1. IRREGULARITY: The symmetrical, ‘question-and-answer’, phrase rhythm breaks down. The liquidation could be heard to begin with the cello F at bar 15, which interrupts the end of the third subphrase (the violin’s statement from bar 14). The premature entry of the cello F (a bar early – a two-bar rest had been the pattern), the extreme chromaticism of the harmony, and the break up of the sequence all signal liquidation. The sequence does partially continue, with the F–E of bar 17 answering bar 14’s E–D, the interval of imitation tightened from five bars to two. But in all sorts of ways, this F constitutes a crisis: a registral reversal (F an octave higher had been implied) which, nonetheless, picks up the register of the anacrusic Gs of bars 4, 9 and 14; metrically displaced to the middle of the bar; accented with a mfp dynamic; and marked with a chromatic G♯ in the bass. 2. ‘FUNCTIONAL FLIPS’: The function of the dyads – which so clearly punctuate the three subphrases in bars 1–15 – ‘flips’ from phrase-ending cues to steps of a thematic line. See the emergent scale figure of bars 16–17, first violin. 3. PARATAXIS: If the harmonic goal of the phrase is the G half-close of bar 19, then this path is deflected by two v4/2 – i cadences, at bars 16 and 18, ˆ 3ˆ melodic apex at bar featuring F–E steps in the bass. The first violin’s 4– 17 – the crux of the schematic sequence – adds a third instance of F–E repetitions. These three F–E harmonic/melodic repetitions, unfolding a ˆ 3ˆ 4– ˆ 3ˆ chain’, stamp the sentence’s continuation with parataxis. ‘4–

Transition The most interesting point of the transition is its liquidation phrase, bars 69–85, which comprises a remarkable reinterpretation of the tonic group. It is a ‘structural liquidation’ of the entire first half of the exposition, where the burden of the tonic-group material comes to a head. Then comes the pivot of the exposition, when the liquidation hands over to the second subject, which

A metaphoric model of sonata form

Example 7.5 Mozart, String Quintet in C major, k515/i, bars 34–41

initiates the next great arc of music. The transition is also remarkable for its size and scope. It occupies 65 bars (bars 21–85), and has its own tripartite structure, with its three phrases assuming the local functions of presentation (bars 21–37), continuation (38–57) and cadence (69–85). Furthermore, the transition is an expanded and modulating ‘answer’ to the opening sentence. Beginning as a minore consequent phrase to bars 1–20, it recruits the tonic group into a complex, higher-level, period/sentence hybrid. The transition’s three phrases afford three fresh ‘glances’ at the material of the tonic group. phrase 1 : The phrase’s liquidation eliminates the schematic ‘buffer’ (the arpeggio), leaving just the dyad cue, which is taken around the circle of fifths: A–D–G–C (see Example 7.5). Formally, this sequence ends with the cello’s F–E dyad at bar 37. Harmonically, it actually continues to the second violin’s B♭–A in bars 37–38, which sounds like a final compressed swing of the cycle (the implied swing to B♭ major is shortcircuited by the cadence on F). The latter creates a kind of ‘textural modulation’ into the second phrase, whose melody elaborates the F–E figure (given by the first viola in full).

203

204

michael spitzer

Example 7.6 Mozart, String Quintet in C major, k515/i, bars 69–81

phrase 2 : This middle phrase counters the dynamism of the preceding fifth cycles with static parataxis; it takes up the F–E repetitions of bars 15–19 and expands them into an oscillating iv–i pattern (or a i–v in F major, given that the B♭s of bar 37 make the tonality ambiguous: C major is only clarified with the powerful cadence at bars 55–57, after repeated, and increasingly elaborate, evasions, including an exquisite Neapolitan ‘purple patch’ in bars 48–51). If the F/C major melody sounds like a premature second subject (which it actually anticipates), then the cadence at bars 59–60 expresses the confirmatory rhetoric of a codetta, a bluff which Mozart calls when the cadence is suddenly extended into a compressed harmonic sequence leading into the D major ‘standing on the dominant’. phrase 3 : Tonally a ‘standing on the dominant’, the phrase also completes the cycle of three ritornelli of the cello/violin dialogue (Example 7.6). Crucially, it makes some striking changes to this ritornello, which confirm it as a ‘meta-liquidation’ of the opening sentence. Mozart now ‘cashes in’ the tonic group’s potential, condensing its breadth in two ways. The five-bar phrases are normalized to four. The sequence repeats ˆ 3ˆ dyads, rather than unfolding a rising scale. These 4– ˆ 3ˆ dyads refer 4– of course to the apex of the original sequence, the F–E of bar 17, as well

A metaphoric model of sonata form

as the paratactic v4/3–i repetitions in bars 15–19. It is as if Mozart has scooped out the music in between bars 5 and 17, answering a phraseˆ 7ˆ with one on 4– ˆ 3, ˆ compressing seventeen bars into nine. ending on 1– This endows the passage with a beguiling double perspective. From one ˆ 7ˆ . . . 4– ˆ 3’s ˆ are absolutely normative, as angle, taken as themselves, the 1– is their four-bar phrase structure. From another, taken in context, they are semiotic liquidations of the opening sentence, distilling its content into an abstract sign (as with the liquidations in k283 and Beethoven’s Op. 2 No.1, but on a larger scale). Hence the feeling of summary and speed, an impression heightened with the stretto at bar 77. As at bars 34–7, the stretto – a ‘liquidation of a liquidation’ – chops off the arpeggio ‘buffer’ of the schema, reserving the dyad cue. These reiterated cues will become thematic in the second group. The buffer will return in the ‘expositional nadir’.

Second Group Mozart has cast his second group in three main sections, bringing it into line with the structure of the first group (see Example 7.7). It begins with an eight-bar sentence (bars 86–93), which is a compressed variant of the first-group sentence. Just as that was followed by the transition’s expanded consequent, the second subject is completed by an extended answering phrase (bars 94–115). Although tonally static, this middle phrase – the expositional nadir – contains the freest and most irregular material in the entire exposition. Arguably, the complexity of this material compensates for its lack of modulation. The third, cadential, part begins at bar 115, with a repeated cadential theme, leading to a post-cadential coda – a musette emphasizing the subdominant.28 The presentation-phrase of the second group – the second subject of bars 86–93 – is pivotal between two liquidations. The thematic liquidation follows through the transition liquidation of bars 77–85. The tonal liquidation culminates with the Perfect Authentic Cadence (PAC) of bar 131, which emerges from the ‘cadential theme’ of bar 115.

1

Thematic liquidation

The stretto from bar 77 telescopes the interval of imitation to single bars, and the acceleration is carried through into the quaver motion of bar 82 (see Example 7.8).

205

206

michael spitzer

Example 7.7 Mozart, String Quintet in C major, k515/i, bars 86–115

Although the last dyad of the sequence is ostensibly C♯–D, this is transposed down to C♮–B, elaborated by the quavers. The quaver diminution facilitates the second subject’s slow-fast double rhythm, when the C resolves to the theme’s B at bar 86. That is, the music from bar 86 is split between the flurry of quavers in the first violin and the whole notes in the middle strings; the

A metaphoric model of sonata form

Example 7.7 (cont.)

ˆ 4– ˆ 5ˆ figure pointing to the cadence of bars 115–18, and thus in former’s 3– ˆ 29 tension with the latter’s axial prolongation of 3. There is consequently a lot more to these inner parts than mere harmonic and textural ‘padding’. Focusing on the second violin’s lowest notes, its B–C alternations continue the dyads from the end of the transition, confirming the emancipation of phrase-ending from phrase, and the ‘functional flip’ from dyad to motive. A third transformation is a reversal of ˆ 3ˆ to 3– ˆ 4; ˆ that is, a displacement of the oscillation the pattern from 4– so as to begin with the B, rather than the C. That is how the oscillations become ‘axial’ prolongations of the third, within a tonic harmony. ˆ 4, ˆ the consequent dyad Another outcome of this reversal is that, as 3– ˆ ˆ can be superimposed upon the antecedent 1–7 dyad, rather than formally answering. That is, the enjambment of the two conformant phrase-endings continues the compression process of the preceding stretto. Rather than ˆ 7ˆ and 4– ˆ 3ˆ pull against each other in a subtle marching in step, the 1– counter-metre – a device Mozart reproduces in Symphony No. 40 (see Example 7.9).

207

208

michael spitzer

Example 7.8 Mozart, String Quintet in C major, k515/i, bars 76–89

This counter-metre, besides enhancing the feeling of speed and compression, also sets in motion a hypermetrical modulation, whereby the stress is successively displaced from the B (bar 86) back to the C (bar 132), via the pointed piano-forte syncopations of bars 107–10 (which emphasize Cs on the weak bars of the phrase). Although thematic liquidation reaches its climax with the second subject, the most probing analysis actually occurs ‘off-centre’ in the exposition, in the second group’s middle phrase, the expositional nadir (bars 94–115). The unusual status of this juncture, as the freest and most irregular point in the exposition, has not been recognized in the sonata literature. Strictly speaking, the effect here is not of ‘liquidation’, since the material is expanded rather than compressed. Interestingly, both here and in k551, the juncture is a forum for a key aspect which has so far been neglected by the music: the ‘buffer’ of the first subject; the arpeggio figures which had been framed by the phrase-ending dyads. They now come into their own. What had been transparent is now rendered ‘opaque’. The opening arpeggiations (bars 1–13) were ‘transparent’ because perfectly assimilated into the sentence’s schematic structure. Beginning with the first violin’s D at bar 101 (see Example 7.7 above), third cycles now dramatically sweep down the registers, continuing mechanically right the way to the cello’s C at bar 106 (D–B–G–E–C–A–F♯– D–B–G–E–C). Descending thirds

A metaphoric model of sonata form

Example 7.9 Mozart, Symphony No. 40 in G minor, k550/i, bars 70–7, string parts only

are intrinsically more destabilizing than rising arpeggios, since they tend to modulate away from their starting-point. Thus the three cascades (bars 101–3; bars 103–5; and bars 105–6) start and end on different harmonies, with the third and last cycle creating a clash between B and C, verticalized as ˆ 3ˆ dyad, is the group’s a iv7 (bar 106). This verticalizing of the B and C, the 4– most dissonant moment.

2

Cadential liquidation

From one angle, syntactical parataxis is evinced in the second group’s variation structure, articulated by a series of increasingly strong cadences, culminating with the PACs of bars 119 and 131. I contend, however, that it is artificial to separate the variation from the cadential process; the two are intimately linked in a process of ‘cadential liquidation’. That is, the second subject adumbrates a cadential model which emerges in the ‘cadential theme’ of bar 115. Cadential liquidation thus endows the group’s (linear) ‘drive to the cadence’ with a dimension of depth: cadence not only arrives, it also rises to the surface. At the end of this chapter, I will explain, from the perspective of cognitive semantics, how the realization of this cadential ‘prototype’ concludes a pathway from an abstract to a concrete category.

209

210

michael spitzer

Example 7.10 Mozart, String Quintet in C major, k515/i, bars 113–31

The cadence at bar 119 (elaborated at bar 131) is a PAC because it concludes a genuine theme and incorporates a subdominant – Caplin’s prerequisites for a PAC (see Example 7.10).30 This theme outlines a distinctive cadential model: a rise from 3ˆ to 5ˆ , ˆ 4– ˆ 5ˆ 5ˆ –4– ˆ 3– ˆ 2– ˆ 1). ˆ The emergence followed by a linear descent to the tonic (3– of this model in the course of the second group can be illustrated by means of a three-stage middleground reduction (see Example 7.11).

A metaphoric model of sonata form

Example 7.11 Three-stage middleground reduction of the second group: (a) presentation phrase, bars 86–93 (b) continuation phrase, bars 94–115 (c) cadential phrase, bars 115–31

The model is implicit in the harmonic middleground of the second group’s presentation phrase, bars 86–92 (see Example 7.11a). Although 5ˆ (D) can be inferred as a background presence, it is overshadowed by B, and its diaˆ 4– ˆ 5ˆ is subsumed into a B–C–B logue with its neighbour, C.31 That is, the 3– oscillation, spread out over two octaves. A cadential descent from 5ˆ is equally down-played. By contrast, the second graph, referring to bars 94–115 (see Example 7.11b), projects 5ˆ (D) as prominently as the presentation ˆ at this broad level, a 3– ˆ 5ˆ rise is unfolded across the phrase had prolonged 3: two phrases. Since this middle phrase states the i–iv–v–i cadential model twice (the second time, more elaborately), it could be viewed as a twofold variant. From a different angle, however, the first half (bars 94–101)

211

212

michael spitzer

Example 7.12 The three pedals of the exposition

prolongs the 5ˆ , and the second half (bars 101–15) unfolds the linear descent ˆ Hence prior to the cadential theme of bar 115, the presentation to 1. ˆ and continuation phrases of the second group together elaborate the 3– ˆ 5ˆ 5ˆ –4– ˆ 3– ˆ 2– ˆ 1ˆ model at a middleground level. It is left for the caden4– tial theme to draw this model out into the foreground. The third graph (see Example 7.11c) is full of the repetitions which characterize a liquidation. Like the middle phrase, it also divides in two. The first half repeats a ˆ 4– ˆ 5ˆ ascent (bars 115–19 and 119–24) – the first ascent immediately 3– ˆ 2– ˆ 1, ˆ the second sitting on the 5ˆ (and it is notable that grounded by a 3– neither of these cadences feature a iv). The second half, similarly, repeats a descent from 5ˆ (bars 125–27 and 128–31) – the first descent interrupted on ˆ the second reaching the tonic. It is this last event which constitutes the Per2, fect Authentic Cadence towards which the entire group – and exposition – gravitates, as its defining moment. The coda is so expansive, and marked with such thematic import, as to offer a counter-narrative for the group, one oriented towards the cadence in its musette (bar 139) (see Example 7.12). From this angle, the second half of the exposition is articulated by three eight-bar pedals, each compounded of three repetitions of B–C plus half close on A: B–C B–C B–C B–A. This model is explicit in the first (bars 86– 93) and second (bars 107–114) statements; with the third, the violin’s E and the D are cover-tones to the viola’s C and B. Hence the process is paratactic on two levels: repetitions of an eight-bar model which is itself based on

A metaphoric model of sonata form

(a)

(b)

(c)

Example 7.13 Transformations of the opening material in the coda: (a) bars 131–3, Violin I (b) bars 4–5, Violin I (c) bars 322–4, Violin II

ˆ 3ˆ 4– ˆ 3ˆ chain’. repetitions of the B–C dyad. I call this paratactic model a ‘4– In terms of topic theory, it is notable that the first and third statements are a musette (and perhaps, for contextual reasons, the middle one ought to be heard as an implicit musette too). For this reason, I term the three statements a ‘musette narrative’. It is striking, then, that Mozart ‘counterpoints’ two processes emanating from the second group – treating it both as a tonal model and as axial repetitions around the schematic B–C dyad. Perhaps this interplay extrapolates the cadential/thematic fusions intrinsic to sentential liquidations. And just as the tonal model gravitated towards the cadential theme, the ‘musette narrative’ unfolds an equally striking set of transformations, whose object is to gradually recall the (1) harmony; (2) metre; and (3) melodic profile of the quintet’s opening (see Example 7.13).

(1) Harmony: The i–v–i alternations (especially bars 86–90) become a i– iv–i pendulum (bars 131–7). Why? Because iv in G major is C, yielding a C–G alternation which rhymes with that of the opening sentence (Examples 7.13a and b). (2) Metre: The phrasing is shifted so as to begin with the C rather than the B, displacing the pattern from B–C–B–C to C–B–C–B (in a complementary move to the earlier displacement at bars 85–6). The underlying pedal allows the end of the cadential phrase (bar 131) to subtly overlap with the start of the coda phrase, which begins properly with the C of bar 132. A natural consequence of this metrical displacement is that the phrase ends with a full rather than a half close at bar 139 (by contrast to the half closes at bars 93 and 113).

213

214

michael spitzer

(3) Melodic profile: The G–C–B line in the first violin (bars 131–3) echoes the incipit of the quintet, the violin’s opening G–C–B figure (bars 4–5). This absolute-pitch correspondence must have meant a lot to Mozart. Why else would he have worked it into, astonishingly, also his recapitulation? That is, at bar 322, Mozart brings back the coda material more or less at pitch, starting with the G–C–B figure, while harmonizing it in C major (see Example 7.13c). It comes back, more normally, in the codetta, bars 353–end, with C–F–E. (The coda of the G minor Symphony affords another striking comparison (see ˆ 7ˆ /4– ˆ 3ˆ overlaps are followed by an echo Example 7.9, above). The 1– of the G minor opening, at pitch, so that the schematic dyads are juxtaposed with the E♭–D appoggiaturas. The ‘category slide’ between two levels of descending minor-second encapsulates pitch-memory across distance.) Finally, these harmonic, metrical, and melodic correspondences are all thrown into relief by the basic syntactic change unfolded across the entire exposition. The original C–B dyad is part of a hypotactic sequence; the terminal C–B figure is submitted to paratactic repetitions. The rhyme thus serves to profile a radical transformation in how we hear the material; as in the G minor Symphony, the effect underscores the great distance travelled. Across this distance, the second group affords us a ‘long gaze’. A further corollary of the metaphoric gaze is to reveal the second group’s strikingly countervailing motion: if the start of the group moves away from the first subject, then the coda returns towards it. The function of the second subject is to realize the suppressed ‘depth’ of the first-subject schema, whereas the coda theme is responsible to its ‘surface’.

The ‘Jupiter’ Symphony, k551/i Thematic liquidation The symphony’s opening sentence is ostensibly more regular and systematic than the quintet’s (see Example 7.14). Its presentation phrase divides cleanly into two conformant subphrases, articulated respectively on v and i (whereas k515’s presentation extended to a third statement). These G–F and A–G phrase-endings outline another common schema: ˆ 5ˆ (as in the first subject of Mozart’s ‘Dissonance’ Quarthe 5ˆ –4ˆ . . . 6– tet, k465). Nor is there any ambiguity about the continuation’s beginning at bar 9, with a liquidation which geometrically accelerates the phrase

A metaphoric model of sonata form

Example 7.14 Mozart, Symphony No. 41 in C major (‘Jupiter’), k551/i, bars 1–24

rhythm from four bars (bars 9–12) to two (bars 13–14) to one (bars 15 and 16) to half (bar 17) to a quarter (the four quaver pairs in bar 18). At its climax (bar 21), the sentence homes in on the last phrase-ending dyad in the implied sequence (that is, G–F → A–G → [B–A] → C–B) much more pointedly than in the quintet. The schema’s liquidation into its dyad could hardly be more dramatic. This semblance of order comports with the plainer, more block-like materials characteristic of the symphonic style, compared with the sharper focus typical of chamber music. Nevertheless, the semblance is deceptive, since Mozart composes against the symphonic style. The antecedent-consequent logic of bars 1–8 cuts across a symphonic-chamber dialogue within the subphrases: military-drum-like Schleifer answered by affective pianti.32 And the pianti themselves are arranged into two-bar groups: the minim phraseending dyads (bars 4 and 8) are miniaturized and sequenced as faster ‘secondary dyads’ in the previous bar (compare with the more normative k465, which approaches its two dyads in quaver scales). A third subtlety

215

216

michael spitzer

Example 7.15 Rising scales in the first theme of k551/i

Example 7.16 Mozart, Symphony No. 41 in C major (‘Jupiter’), k551/i, bars 38–44

is the rising scale (G–A–B–C) tucked into the Schleifer themselves. Having established three levels, or ‘spaces’, of syntax – symphonic opening; lyrical phrase-ending; motivic secondary dyads – Mozart proceeds to mix them. Even more so than in the quintet, the ‘Jupiter’’s kaleidoscope works through category slippage. Each of these three ‘spaces’ unfolds a rising scale (see Example 7.15). The phrase-ending dyads imply a rise of a fourth from G to C, the boundary sounded by the top C at bar 9, and then filled in by the continuation. This G–C scale is of course the same as the Schleifer’s roll, although because of absolute differences of metre (between triplet semiquavers on the one hand, and a scale staked out by widely-separated minims on the other), it is as yet a dormant relationship. The key to ‘awakening’ this relationship is provided by the third of these scales, begun on the ‘secondary dyads’ of bar 3. The three levels fuse together in the transition section, which is saturated with the dyads, configured in sequential groups of fourth scales (beginning G–A–B–C, bars 39–40), and scored tutti (see Example 7.16). The fusion involves an ingenious cross-over of pitch and tonal function, taking up the dotted rhythms of the secondary dyads, plus their tonic orientation, but simultaneously tonicizing the G orientation of the minim phrase-ending dyads. This cross-over is instrumental in making us hear the G scale of the second subject as a projection from the tonic sentence’s phraseendings.

A metaphoric model of sonata form

Example 7.17 Mozart, Symphony No. 41 in C major (‘Jupiter’), k551/i, bars 52–66

As in the quintet, the transition (bars 24–55) mediates between the first and second groups by weaning the phrase-endings off their phrases – the pianti from their Schleifer. Most dramatically, the second half of the sentence, the ritornello-like block in bars 9–23, is largely eliminated, displaced at bars 31–4 by a sequential extension of the pianti, and then its symphonic sound absorbed into a second sequence at bars 39–48. The transition is marked off with a fragment of the ritornello, the v–i cadential repetitions of bars 49–55. With the liquidation from bar 39 concentrating on the dyads, ˆ 2– ˆ 3ˆ ascent in G major the ‘standing on the dominant’ focuses on the 1– (G–A–B), welding the crucial link between the C and G scales, and setting up the second subject (see Example 7.17). The presentation phrase of the second group, bars 56–61, is a hybrid fusion on an architectonic level – assimilating the sequential climax of the transition into the periodicity of a lyrical theme. Just as in k515, it sounds

217

218

michael spitzer

Example 7.18 Permutation of schema pitches

both fast and relaxed at the same time. The theme clearly flows out from the preceding liquidation. In particular, the v7 of bar 55 sounds active across the first bar of the theme, which regains the A at bar 57 (this is partly because the preceding tutti makes the piano bar 56 sound anacrusic). The phrase ends (bars 59–60) with a compressed reference to the transition’s E–G cadential descent (44–7). But the most interesting aspect of the theme is its contrapuntal design. At the top, the first violin permutates the pitches ˆ 5ˆ schema into a 5ˆ –6ˆ . . . 4– ˆ 5ˆ (G–A . . . F♯–G), or of the original 5ˆ –4ˆ . . . 6– ˆ 2ˆ . . . 7ˆ –1ˆ in G major, while ‘flipping’ their function from schematic to 1– motivic (see Example 7.18). At the same time, the texture leads the ear from the violins’ G–A to the cellos’ A–B. This ‘counterpoint’ between static axial prolongation and ˆ rise dynamic rise echoes that in the quintet (k515: prolongation around 3, ˆ3–4– ˆ 5ˆ ; k551: prolongation around 1; ˆ rise 1– ˆ 2– ˆ 3). ˆ

Cadential liquidation The second group gravitates towards not one but two cadential themes in the coda, respectively at bars 101 and 111.33 The latter in particular comprises a ˆ 3– ˆ 2– ˆ 1ˆ cadential descent. The group’s course from second prototypical 5ˆ –4– subject to cadential theme pivots on a third contrapuntal strand (in addition ˆ 2ˆ . . . 7ˆ –1ˆ and 1– ˆ 2– ˆ 3ˆ progressions): a descent from D is implicit in to the 1– the upper voice of the second violins’ tremolo at bar 56, and emerges at the end when the D is transferred to the melody at bar 60 and starts to fall. We can track the path from second subject to coda as a series of realizations, which bring out its melodic (bars 73 and 77) and cadential (bars 87–94) aspects. One function of the striking C minor/major episode (bars 56–61) is to add a crucial subdominant to the descent, a chord signally lacking in bars 56–61 (see Example 7.19).

A metaphoric model of sonata form

Example 7.19 Mozart, Symphony No. 41 in C major (‘Jupiter’), k551/i, bars 81–9

The powerful subdominant digression is channelled into the cadence at bars 87–9, and then encapsulated upon the cadence’s subsequent repetitions (bars 92, 97, 107, 113, 116). The episode is notable also because it is the expositional nadir. Its irregularity is due not simply to the dramatic C minor explosion at bar 81, but to its subsumption of both opening gestures (the C major fanfare of bars 85–6) and closural cadences (the PAC’s of bars 87–9 and 93–4) in medias res. Indeed, the blast of C major fanfare is an incongruent – and strangely inaccurate – echo of the symphony’s very opening. That is, the ‘Jupiter’ began with a Schleifer, not a fanfare. Nevertheless, by picking out the triadic lineaments underlying this opening, the fanfare could be heard as realizing the topic’s implications. As in k515, the symphony recuperates a triadic buffer. Equally remarkable is the progression – both melodic and metrical – charted by the second group’s three themes, which recalls that of k515’s three musettes, and which similarly plays in counterpoint against the group’s three phrases (that is, theme 2 comes in the coda, not in the middle). Theme 1 (bar 56) starts on G, theme 2 (bar 101) on B, and theme 3 (bar 111) on D, the three onsets arpeggiating a rise to D which is miniaturized at key moments of the foreground (for instance, bars 71–2, cello), and which perhaps originate in bars 3–4. Metrically, themes 2 and 3 are diminutions of theme 1, keeping the i–v v–i harmonic structure but halving its phrase rhythm (Example 7.20). Whereas the acceleration from quavers to semiquavers across themes 2 and 3 evokes liquidation texturally, a simple rhythmic reduction also shows that theme 2 accelerates the units of the i–v v–i pattern from whole bars to

219

220

michael spitzer

(a)

(b)

Example 7.20 Mozart, Symphony No. 41 in C major (‘Jupiter’), k551/i, (a) bars 101–3 (b) 111–13

minims, a pattern ‘squashed’ in theme 3 into a syncopated dotted rhythm. Textural details connect the two coda themes with the second subject: theme 2’s Alberti bass evokes the earlier tremolo, and theme 3’s melodic cello writing (particularly the B–A–G descent at bars 111–13) recalls that at bars 58–9. The third theme sounds climactic also because Mozart blends in a yet faster liquidation of the dyads from the end of the original sentence (bar 21). This last liquidation is more complete because of its semiquavers (bar 21 has quavers), and because Mozart shifts the neighbour-note figures so as to stress the respective notes of resolution (bar 21 emphasized the leadingnote B). The three panels of the coda – themes 2 and 3, as well as the codetta’s miniature Schleifer ritornello – all feature ternary schemes. Theme 2 states the i–v v–i model not once but thrice; theme 3 is a repeated three-bar phrase; the codetta (bars 117–20) is a three-bar three-fold v–i cadence; and we can also include the three tonic hammer-blows in the last bar as both a semiotic kernel (‘residue’) of the cadential liquidation as a whole, and a motto of its curiously ternary logic (see Example 7.21).

A metaphoric model of sonata form

Example 7.21 Mozart, Symphony No. 41 in C major (‘Jupiter’), k551/i, bars 117–20 and 1–2

Why three? The number becomes pointed at the da capo, when the final three G major crotchets get juxtaposed with the three C major minims which began the journey at bars 1–2. Listened to da capo, the tonic group is rife with all sorts of metrical tensions between ternary and duple groupings – too complex to be addressed here. Enough to say that they are subsumed within the prevailing binary logic of the antecedent-consequent presentation phrase (bars 1–8), just as they seem to become emancipated in the second group. Thought-provokingly, the ‘Jupiter’ implicitly associates hypotaxis with duple patterns, and parataxis with triple. Perhaps this two-againstthree dialogue has something to do with the binary versus ternary tensions in sonata form (a binary form yet with a central development). At least on a harmonic level, the rhyme simply sounds between two grammars of i–v relationship: the i–v v–i of conformant subphrases (bars 1–8), and the i–v–i iterations of the codetta which, though stated only three times, could symbolically continue ad infinitum. The ‘Jupiter’ takes the listener on a path – a circular path – from the sublunar to the eternal, as if Mozart were saying that in his world the two are much the same.

Of metaphor: rhetoric, philosophy, cognition This study is necessarily open-ended, since the exposition leaves us only half, or even a third, of the way through the sonata form. Applying my ‘two-time rule’ to the entire form will be part of a larger future study. But, towards that end, I would like to conjecture that the two-time rule’s recursiveness (from sentence through group to exposition and its da capo) climaxes at the level of recapitulation. The conflicts between the sonata’s binary and ternary aspects, which turn on the role of the development, similarly extrapolate

221

222

michael spitzer

tensions in the distribution of a sentence’s three functions: specifically, the fusion of continuation and cadence within the liquidation process. Rather than pursuing this line here, I would like to conclude by returning to a question ventured at the beginning of this chapter: what do these formal inquiries have to do with the concept of ‘metaphor’? Metaphor, as I have shown at length elsewhere,34 is a composite, portmanteau category, yoking together a multitude of practices. I will outline three of these practices here: the rhetorical, the philosophical and the cognitive. Metaphor in discourse was classically associated with the rhetorical arts of persuasion, albeit more with the eloquence of the legal chamber or the pulpit than with poetic literature. Eclipsed in the eighteenth century by rationalist philosophy in mainland Europe, rhetoric persisted longest in Britain, where it comported with the pragmatist aesthetics of the Scottish Enlightenment. Nevertheless, the counter-flows between British empiricism and European rationalism were complex. From music’s standpoint, Johann Sulzer, the philosophical godfather of Koch’s Versuch, was strongly indebted to Hugh Blair’s Lectures on Rhetoric and Belles Lettres, the most influential and oft-published rhetorical treatise of the century.35 Blair’s method reveals its modernity through a commitment to the binary opposition between what he terms ‘perspicuity’ and ‘ornament’: All the qualities of a good style may be ranged under two heads – perspicuity and ornament. For all that can possibly be required of language is to convey our ideas clearly to the minds of others, and, at the same time, in such dress, as by pleasing and interesting them, shall most effectually strengthen the impressions which we seek to make.36

This opposition is operative beneath the ostensibly old-fashioned six-stage sequence of the discourse, which Blair details in chapters 31 and 32: (1) ‘Introduction’; (2) ‘Division’; (3) ‘Narration and Explication’; (4) ‘The Argumentative Part’; (5) ‘The Pathetic Part’; (6) ‘The Peroration’. The pattern actually breaks down into two pairs of oppositions, each reflecting the basic perspicuity/ornament dualism, framed by the introduction and peroration. That is, the ‘Division’ gives ‘the proposition, or enunciation, of the subject’, and thus evinces the rational principle of perspicuity; the concepts of the discourse are then ‘ornamented’ by ‘Narration and Explication’, which by adding detail and colour serve ‘to illustrate the cause, or the subject of which the orator treats’.37 The second pair, ‘The Argumentative Part’ and ‘The Pathetic Part’, recapitulate this opposition at a more dramatic level, resorting by turns to forensic debate and impassioned feeling in order to persuade, first by reason, then emotion. Metaphor, the topic of Blair’s

A metaphoric model of sonata form

‘Lecture XV’, is the premier figure of the rhetorical language which dominates the second element of each pair. Blair’s rhetoric thus unfolds the ‘two-time rule’ recursively, persuading the audience by saying the same thing twice, the first time logically, the second time figuratively, and at rising levels of intensity. Blairist rhetoric may well capture the variational character of the ‘twotime rule’ in the discrete succession of schematic and figurative modules. But it has little to say about the rule’s cognitive dimension, which concerns changes in the way we hear. For that, we can look to the philosophical mainstream in Germany. The cardinal difference between British rhetoric and German aesthetics is that the latter undergoes a psychological turn, and becomes preoccupied with the evolution of consciousness, from brute sensation to refined conceptualization. While Blair looks at styles of expression, seeing them as schematic or figurative, philosophers such as Christian Wolff explore modes of thinking, classed as ‘clear’ or ‘obscure’. Metaphor comes back into play to help sort out a paradox in the evolution of mind. Postulating conceptual clarity as the goal of enlightened thought was all very well, but how could this square with more sensuous modes of thought, such as aesthetics? Aesthetic experience – including literature and music – is curious, in that it seems to return to the ostensibly primitive realm of sensation, while holding on to rational principles of artistic technique. The need for such a mediated return is hard-wired into the constructivism of thought: if thought constructs reality, by regulating perception through concepts and ideas, then it might equally be felt to block reality, in which case there is a need for reality to be rescued. Aesthetics met that need, harmonizing the faculties of reason and imagination. If art mediates sensation through concepts, then German aesthetics endows the ‘two-time rule’ with both its tension and its circularity. Crucially, in the realm of discourse, philosophers identified this ‘return to nature’ with the workings of metaphor. Whereas the conceptuality of prose was epitomized by the arbitrariness of the sign, language attained the expressive ideal of poetry when signs were ‘naturalized’ into metaphors. For instance, Moses Mendelssohn, a contemporary of Kant, thought that metaphor made ‘discourse sensate’, and described the process in terms of a transformation of cognition from ‘symbolic’ to ‘intuitive’. Of course, unmediated perception of the object ‘in itself’ was impossible, as Kant proved. Yet, contra Kant, Mendelssohn suggests that, with metaphoric intuition, ‘the lower faculties of the soul experience an illusion in that they often forget the signs and believe that they are viewing the object itself’.38 The resulting illusion is a mediated immediacy, a ‘second nature’, an expressiveness filtered through

223

224

michael spitzer

convention. The trajectory from arbitrary sign to the picture-like, ‘naturalized’ signs of metaphor is circular. The link with musical metaphor is straightforward, on the basis of the analogy between the signs of language and musical convention. Simply put, Mozart’s signs are those dyads which punctuate his phrase-endings. Their ‘functional flip’ into motives in the second group parallels the naturalization of the arbitrary sign. More broadly, given the ‘sign-character’ of musical form as a whole, scrambling that form (its congruence and conformance) emancipates its materiality, so that the music is ‘naturalized’. The second group gives us the illusion of an immediate perception of the first group’s material. By Mendelssohn’s lights, our cognition of the two groups is respectively ‘symbolic’ and ‘intuitive’. Metaphor theory has itself undergone a kind of mediated return, putting German conceptualization on an Anglo-American empiricist footing reminiscent of Blair’s rhetoric. The rationalist-empirical hybrid emerged through recent orientations in cognitive science.39 Metaphor, for cognitive linguists such as George Lakoff,40 is a category of thought, structured through ‘mapping’ from patterns of bodily experience called ‘image schemata’. Different patterns of experience (traversing a path, balancing, rising or falling and so forth) yield different sorts of categories, and these have the potential to loosen up our models of musical structure. Rather than thinking of form as monolithically hierarchical or tree-like, we can be relaxed about the possibility of one kind of form handing over to another. Two types of category seem ideally suited to music. The s o u r c e - pat h - g oa l schema enshrines our experience of directed motion through space, and Lakoff has shown how it maps onto the flow of reason in an argument, and of syntax in that argument’s expression. I would venture that this schema also maps onto the course of the tonic group in a sonata-form exposition. The variational character of the second group is explained by another kind of schema: the c e n t r e - p e r i p h e ry . While flowing, in the first instance, from our experience of being centred, and of gauging the world as deviations from our centre, the schema also maps onto our categorization of objects, qualities and concepts in relation to prototypes or basic-levels. There are prototypical cases of birds (robins, rather than penguins), just as there are prototypes of the colour red shared across cultures. We are more liable to relate to basic-level categories such as ‘chair’, than to superordinate ‘furniture’ or subordinate ‘Chippendales’, because ‘chair’ is what we sit on: it represents an optimal level of engagement with physical reality. Thus, given a prototypical centre, we gauge reality in terms of distance from that centre. So far as I know, the c e n t r e - p e r i p h e ry schema has not been related to musical discourse in such a way before. But its affinity with variational

A metaphoric model of sonata form

Figure 7.2 The source–path–goal schema and the centre– periphery schema combined in the sonata-form exposition

structure – indeed with the general diminution process – seems plausible. A cadence is prototypical when it is a cadence proper (as Caplin would define it), coming at the end of a theme and at the end of a second group, complete with subdominant-type harmonies, root-position dominant and final tonic, together with a descending line. I want to suggest two other connections. First, the c e n t r e - p e r i p h e ry schema accords with my earlier account of how musical expressiveness ‘focalizes’ around a musical detail; that is, it names the structure of old-style expressive metaphor. Second, I suggest that c e n t r e - p e r i p h e ry naturally comes in second position, after s o u r c e - pat h - g oa l ; that the two schemas constitute the ‘twotime rule’. The exposition can thereby be pictured as a succession of two schemas (see Figure 7.2). We would need another picture, however, to express the sense that the second group is itself a ‘goal’ to the first; that the c e n t r e - p e r i p h e ry schema is the destination of the s o u r c e - pat h - g oa l . And from a third standpoint, we could easily hear the s o u r c e - pat h - g oa l of the tonic group as a ‘centre’, from which the c e n t r e - p e r i p h e ry schema deviates. In plain English, the second group can be heard as both secondary (a deviation) and primary (a goal); as the first group’s elaboration and as its realization. This is just another variant of the venerable dispute as to whether metaphors are ornaments or concepts; whether they concern expression or thought. The crux to seize on is, perhaps, that the status of the second subject – the hinge between the exposition’s two liquidation processes – keeps the two possibilities in play; this metaphorical play is kindred, surely, with the well-chronicled play of the ars combinatoria, of which Mozart was master. I will end with two other arrows. The first arrow leads us back to Mozart’s character. In an extraordinary chapter of his biography, ‘The Zoroastran Riddles’,41 Solomon relates Mozart’s Spieltraub to his love of carnival masks and to the interplay in his character between mask and reality. ‘Mozart’s

225

226

michael spitzer

riddles’, he writes, ‘oscillate between concealing and revealing, between mystery and clarity, between asking and waiting for an answer’.42 This oscillation is dramatized, according to Solomon, as much in the various disguises and unmaskings of the operas, as in Mozart’s psychological games with his father. I would add that it is also composed out in the dynamic of his musical materials: in the oscillation between concealing and revealing within the second subject of Mozart’s sonata forms. My second arrow points forwards and outwards to the general condition of discourse. Not to lose sight of the gambit which began this investigation, we should remember that Mozart’s play is bisected by Time’s Arrow: sonata form ends differently to how it starts. Music, like life, always moves forward, and time can never be regained. Recapitulation is never repetition, because even repetition is never repetition. The second time is always experienced differently to the first. It may look the same on paper, but, as with the liquidation principle, the details encapsulate memory of the past, as a sign. I have elsewhere, following Paul Ricoeur’s hermeneutic theory of metaphor, modelled music’s ‘one-way street’ as a ‘trajectory towards the (metaphorically) dense’.43 For now, it is sufficient to conclude that Mozart’s oscillation between concealing and revealing floats downstream on Heraclitus’ river of life.

Notes 1 This is not the place for a full treatment of James Hepokoski and Warren Darcy, Elements of Sonata Theory (New York: Oxford University Press, 2006), the most substantial recent study of sonata form. I consider their theory in Michael Spitzer, ‘The New Formal Theory’, Beethoven Forum, forthcoming. 2 See Jean Mandler, ‘Categorical and Schematic Organization in Memory’, in Memory Organization and Structure, ed. Richard Puff (New York: Academic Press, 1979), 259–99. 3 On Koch’s punctuation model of classical form, see Michael Spitzer, Metaphor and Musical Thought (Chicago: University of Chicago Press, 2004), ch. 5, espe¨ Rhythmustheorien der Neuzeit, Neue cially 243–60. See also Wilhelm Seidel, Uber Heidelberger Studien zur Musikwissenschaft 7 (Bern: Francke, 1975); Wolfgang Budday, Grundlagen musikalischer Formen der Wiener Klassik: An Hand der zeitgen¨ossischen Theorie von Joseph Riepel und Heinrich Christoph Koch dargestellt an Menuetten und Sonatens¨atzen (1750–1790) (Kassel: B¨arenreiter, 1983); and Hermann Forschner, Instrumentalmusik Joseph Haydns aus der Sicht Heinrich Christoph Kochs, Beitr¨age zur Musikforschung 13 (Munich: Musikverlag Emil Katzbichler, 1984). 4 Arnold Schoenberg, Fundamentals of Musical Composition, ed. Gerald Strang and Leonard Stein (London: Faber and Faber, 1967), 58–62. William Rothstein

A metaphoric model of sonata form

5

6

7

8

9 10 11 12 13 14

15

16

also discusses the term in relation to sentence form in Phrase Rhythm in Tonal Music (New York: Schirmer Books, 1989), 26. I have elsewhere (Metaphor and Musical Thought, 219) designated the functions of the first and second groups as ‘rhythmic’ and ‘logical’, respectively, following ¨ Seidel, Uber Rhythmustheorien der Neuzeit, and Forschner, Instrumentalmusik Joseph Haydns. ‘Logical’ may sound like a misnomer, since the lyricism typical of the second group suggests an expressivity normally thought of as the contrary of logic. Nevertheless, it refers to the structure of the group – its dedication to the repetition and elaboration of a motive – rather than to its affective tone. The Entwicklung/Fortspinnung distinction originates with Friedrich Blume, ‘Fortspinnung und Entwicklung: Ein Beitrag zur musikalischen Begriffsbildung’, Jahrbuch Peters 1929, 51–70. See also Carl Dahlhaus, ‘Some Models of Unity in Musical Form’, Journal of Music Theory 19 (1975), 2–30, and Rudolph ¨ Klein, ‘Wo kann die Analyse von Haydns Symphonik ansetzen?’, Osterreichische Musikzeitschrift 37 (1982), 234–41. William E. Caplin, Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven (New York: Oxford University Press, 1998): 9–12; see Schoenberg, Fundamentals of Musical Composition, 59, 63. Kofi Agawu lays out the ‘Beginning–Middle–End Paradigm’ in Playing with Signs: A Semiotic Interpretation of Classic Music (Princeton: Princeton University Press, 1991), 51–79. Classical Form, 11. Ibid., 11. Schoenberg, Fundamentals of Musical Composition, 59. Ibid., 59n. See Spitzer, Metaphor and Musical Thought, 216–18. The functional fusion at the end of Beethoven’s sentence also suggests a Hegelian synthesis. In Music as Philosophy: Adorno and Beethoven’s Late Style (Bloomington: Indiana University Press, 2006), 244–5, I consider Caplin’s analysis of this phrase in the context of what Janet Schmalfeldt terms the ‘Beethoven–Hegelian tradition’ (Schmalfeldt, ‘Form as the Process of Becoming: The Beethoven–Hegelian Tradition and the “Tempest” Sonata’, Beethoven Forum 4 (1995), 37–71). Leonard Meyer calls it an ‘archetype’ in his 1980 essay ‘Exploiting Limits: Creation, Archetypes, and Style Change’, Daedalus 109 (1980), 177–205; reprinted in Meyer, The Spheres of Music: A Gathering of Essays (Chicago: University of Chicago Press, 2000), 189–225. Robert O. Gjerdingen, in A Classic Turn of Phrase: Music and the Psychology of Convention (Philadelphia: University of Pennsylvania Press, 1988), re-named it a ‘schema’ and dedicated a whole book to a statistical survey of its historical evolution. Spitzer, Metaphor and Musical Thought, 46–50, relates the schema to metaphor theory. For a survey of the pianto topic, see Raymond Monelle, The Sense of Music: Semiotic Essays (Princeton: Princeton University Press, 2000), 66–73.

227

228

michael spitzer

17 See Leonard B. Meyer, Explaining Music: Essays and Explorations (Berkeley: University of California Press, 1973), 44 and 81, for definitions of ‘conformance’ and ‘congruence’. 18 A classic example of this technique is the exposition of the ‘Waldstein’ Sonata, Op. 53. 19 Classical Form, 16. 20 William E. Caplin, ‘The Classical Cadence: Conceptions and Misconceptions’, Journal of the American Musicology Society 57 (2004), 51–117. 21 The classic example is Johann Mattheson’s punctuation of a sixteen-bar minuet in his Der vollkommene Capellmeister (Hamburg: Herold, 1739), 224, discussed in Spitzer, Metaphor and Musical Thought, 211. Caplin comes down hard on the eighteenth-century metaphor of musical punctuation in ‘The Classical Cadence’, 103–6. But his critique overly simplifies the philosophical and cognitive issues involved (see Spitzer, Metaphor and Musical Thought, 211–43). 22 Caplin, ‘The Classical Cadence’, 59. 23 See the time-span reduction of a sonata form in Fred Lerdahl and Ray Jackendoff, A Generative Theory of Tonal Music (Cambridge, MA: The MIT Press, 1983), 246–7, which models these ‘tensing’ and ‘relaxing’ functions as ‘left’ and ‘right’ branches of a Chomskyan tree diagram. 24 I explain the hypotactic-paratactic distinction in Spitzer, Metaphor and Musical Thought, 228, and more extensively in Spitzer, Music as Philosophy, 114–15. Klein, ‘Wo kann die Analyse von Haydns Symphonik ansetzen?’, and Elaine Sisman, Haydn and the Classical Variation (Cambridge, MA: Harvard University Press, 1993), also discuss these terms, with particular reference to Haydn. For the cognitive dimension, see Spitzer, Metaphor and Musical Thought, especially 10 (schemas) and 20–1 (prototypes). See also Lawrence M. Zbikowski, Conceptualizing Music: Cognitive Structure, Theory, and Analysis (New York: Oxford University Press, 2002), 41–4, for a different interpretation of prototype theory. 25 For Mozart and the variation principle, see also V. Kofi Agawu, ‘Mozart’s Art of Variation: Remarks on the First Movement of K. 503’, in Mozart’s Piano Concertos: Text, Context, Interpretation, ed. Neal Zaslaw (Ann Arbor: University of Michigan Press, 1996), 303–13. 26 See Agawu’s extensive analysis of k515 in Playing with Signs, 80–99, which identifies the musette topics. 27 And hence a qualification to Agawu’s argument about the essential arbitrariness of topical ‘play’ (Playing with Signs, 49, 78–9). 28 Caplin explains that the musette’s tonic pedals promote tonal stasis, and thus render it ‘inappropriate for cadential articulation’ (‘On the Relation of Musical Topoi to Formal Function’, Eighteenth-Century Music 2 (2005), 118). This is why ‘the topic is also associated with post-cadential formal function’ (117–18). 29 Meyer, Explaining Music, 183, defines ‘axial’ melodies as consisting of ‘a main or axis-tone embellished by neighbor-notes above and below’. Strictly speaking,

A metaphoric model of sonata form

30

31

32 33

34 35

36 37 38 39 40 41 42 43

Mozart’s B–C alternation lacks a lower neighbour until the A at the end of the phrase. I would argue, nevertheless, that the functions are similar. Caplin, ‘On the Relation of Musical Topoi to Formal Function’, stresses that a terminal v–i is not enough to constitute a classical cadence, which ‘more typically brings three or four harmonies’ (83). The same criterion defines the v–i repetitions which often permeate codettas as ‘post-cadential’ (89–96). It is revealing that Agawu, in his own Schenkerian interpretation of this movement (Playing with Signs, 92), gives equal credence to the 3ˆ and the 5ˆ as head tones of both groups (although he settles for 3ˆ in the tonic group). I follow Monelle’s designation of these topics (The Sense of Music, 43, 70). I don’t think the fact that the first of these themes is re-used in Mozart’s concert aria Un bacio di mano, k541, alters the argument. The ‘Jupiter’ may well have been in his mind during or even before the aria’s composition, so questions of priority or influence do not necessarily come into play (but see also Monelle, The Sense of Music, 31). Spitzer, Metaphor and Musical Thought, and ‘Tovey’s Evolutionary Metaphors’, Music Analysis 24 (2006), 437–70. Although not published till 1783, unauthorized copies were in circulation since the time Blair began lecturing at Edinburgh University in 1759. Sulzer’s Allgemeine Theorie der sch¨onen K¨unste (1771–4) must have drawn on one of these unauthorized pre-publication copies. Hugh Blair, Lectures on Rhetoric and Belles Lettres, third edition (Edinburgh: J. Murray, 1787), 116. Blair, Lectures on Rhetoric and Belles Lettres, 424. David E. Wellbery, Lessing’s Laocoon: Semiotics and Aesthetics in the Age of Reason (Cambridge: Cambridge University Press, 1984), 79. I review these new orientations in Spitzer, Metaphor and Musical Thought, especially 54–67. George Lakoff, Women, Fire, and Dangerous Things: What Categories Reveal about the Mind (Chicago: University of Chicago Press, 1987). Maynard Solomon, Mozart: A Life (London: Pimlico, 1996), 337–52. Solomon, Mozart: A Life, 350. See Spitzer, Metaphor and Musical Thought, 93–111. Ricoeur’s ‘tension theory’ is particularly appropriate to model Mozart’s interplay between masking and revealing, as it deals with the ‘stereoscopic’ relationship between material in its original and recontextualized form. The emphasis in Ricoeur’s hermeneutics increasingly falls on how metaphorical discourse unfolds against the horizon of time.

229

8

Beethoven’s Op. 18 No. 3, first movement: two readings, with a comment on analysis ko f i ag aw u

In this chapter, I aim to stimulate discussion of the practice of music analysis. Although analysis is basic to the training of most academic musicians, it remains a contested discipline. Some are put off by its technical language, its failure to address certain parameters (like affect), its supposed blindness to salience, its uneasy alliance with composing and performing, and its use of ostensibly anachronistic methods. These charges are obviously too complex and multifaceted to be dealt with adequately in a short paper.1 Since the emphasis here is on practice, I offer for discussion and debate the beginnings of two readings of the opening movement of Beethoven’s String Quartet in D major, Op. 18 No. 3, the first of the Opus 18 quartets to be composed (1801), and one that has not (yet) received an excessive amount of scholarly commentary.2 Beethoven’s music is, of course, central to the definition of a Euro-American analytic tradition, so this choice should be neither surprising nor controversial. The first reading conveys aspects of structure through a paradigmatic analysis inspired by the semiological method associated with Nicolas Ruwet, Jean-Jacques Nattiez and their followers.3 A second reading seeks access to the movement’s expressive dimension by exploring its topical content in the manner of Leonard Ratner.4 Each reading is self-sustaining and yet ultimately partial; both are at once autonomous and complementary. No strenuous effort is made to seek a synthesis. The emphasis, rather, is on what each analytical proceeding makes possible. As Example 8.1 shows, Op. 18 No. 3 begins with two violin whole notes, and continues to a cadence in bar 10 either in whole notes (the lower strings) or in diminutions of whole notes (the first violin). The texture suggests alla breve style, an archaic style domesticated within the pedagogy of counterpoint. From this simple intuition, we might go on to reason that if the first ten bars can be represented in something like a two-voice first-species contrapuntal exercise (as shown in the reduction in Example 8.1), the entire movement might respond similarly. The material displayed in Examples 8.3, 8.5, 8.7, 8.9, 8.12 and 8.15 concretizes this intuition by reproducing the content of the whole movement in arrangements that demonstrate affinity among units rather than chronology. For each unit, I have written out the outer voices, added figures between the staves to register intervallic content

Beethoven’s Op. 18 No. 3, first movement

Example 8.1 Beethoven, Op. 18 No. 3/i, bars 1–10, with two-voice reduction

in reference to the sounding bass, and simplified the rhythmic texture to make the analysis more manageable. Although some of them may be self-evident, the criteria for dividing up the movement into meaningful units (sense units, utterances, ideas, phrases, periods) require comment. The challenge of segmentation remains at the heart of many analytical endeavours, and while it seems more acute in relation to twentieth-century works (as Christopher Hasty and others have shown5 ) it is by no means straightforward in Beethoven. In this analysis, I have opted for the most basic of conventional models of tonal motion based on the i–v–i nexus as the principal criterion. Some support for this proceeding may be garnered from Adorno, who argues as follows: To understand Beethoven means to understand tonality. It is fundamental to his music not only as its ‘material’ but as its principle, its essence: his music utters the secret of tonality; the limitations set by tonality are his own – and at the same time the driving force of his productivity.6

Whatever else this might mean, it engenders an idea that seems tautological at first sight, but actually fuses a historical insight with a systematic one, namely, foundations as essences. The implication for Beethoven analysis is that the most fundamental modes and strategies of tonal definition constitute (part of) the essence of his music. The analytical task then is to discover or uncover ways in which these essences are thematized.

First reading: tonal models7 Based on a normative principle of departure and return, the fundamental mode of tonal definition may be glossed as a i–v–i harmonic progression

231

232

ko f i ag aw u

(obtained by setting the chord of nature into motion, giving it a temporal or horizontal or ‘worldly’ disposition). This progression is in turn subject to various concrete or idiomatic modes of expression and transformation. I could, for example, tonicize v, prolong it by introducing a predominant element, or prolong it even further by tonicizing the predominant. More radically, I might truncate the model by deleting part of its frame. Thus, instead of the complete i–v–i chain, I may opt for an auxiliary v–i progression, in which form it still attains closure but relinquishes the security of a beginning tonic. Or I may prefer a i–v progression, in which form the model is understood as open and incomplete. Modifications are understood in reference to the larger, complete progression. Stated so abstractly, these procedures sound obvious, mechanical and uninspiring. Indeed they belong at the most elementary level of tonal expression. And yet without some sense of what is possible at this level, one cannot properly appreciate the ecology (or economy, or horizon) that sustains a given composition. If music is language, a language based in conventional codes, then it is trivial to have to point out that by 1800 Beethoven the pianist, improviser, composer, student of counterpoint and copier of other people’s scores thoroughly spoke a language enabled by what might be called the first utterances of the tonal system: i–v–i progressions and what they make possible. What, then, are these models, these first utterances of the tonal system, and how are they composed out in the course of the first movement of Op. 18 No. 3? Figure 8.1 provides a broad orientation to the two readings that I will be offering by summarizing a segmentation of the movement into forty sense units (left column) and indicating some of the topical references that might be inferred from the sounding surfaces (right column). Figure 8.2 lists the same forty units, only now in reference to tonal models and a sonata form outline. As will be immediately clear from the examples that follow, each of the forty units belongs to one of three models, which I have labelled 1 (with subdivisions 1a, 1b, 1c, 1d), 2 and 3. Model 1 expresses a closed i–v–i progression, while Models 2 and 3 express open and closed i–v and v–i progressions respectively. In the following commentary on the models, I begin with a simple melodic-harmonic progression that encapsulates the essential motion (by ‘origins’ I refer to conceptual origins), and follow this with a gathering of all the units in the movement that express the particular motion. Readers are encouraged to play through the models and their offspring at the piano in order to grasp the main outlines of the more detailed description that follows.8

Beethoven’s Op. 18 No. 3, first movement

Unit 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Bars 1-10 11-27 27-31 31-35 36-39 40-43 43-45 45-57 57-67 68-71 72-75 76-82 82-90 90-94 94-103 104-110 108-116 116-122 123-126 127-28 129-132 132-134 134-138 138-142 143-147 147-156 158-188 188-198 199-202 203-206 207-213 213-221 221-225 225-234 235-241 239-247 247-250 251-255 255-259 259-269

Topical references alla breve, galant alla breve, cadenza, messanza figures alla breve, stretto, galant bourrée bourrée bourrée bourrée Sturm und Drang, brilliant style, march brilliant style, fantasia, march march, alla zoppa march, alla zoppa fanfare fanfare musette musette fantasia alla breve alla breve, fantasia bourrée bourrée bourrée bourrée alla breve, furioso, Sturm und Drang alla breve, furioso, Sturm und Drang alla breve, furioso, Sturm und Drang Sturm und Drang, concitato style alla breve, ricercar style, march brilliant style, march march, alla zoppa march, alla zoppa fanfare fanfare musette musette fantasia alla breve march, alla zoppa march, alla zoppa alla breve, stretto, messanza figures alla breve

Figure 8.1 Structural units and topical references in Beethoven, Op. 18 No. 3/i

Model 1a, shown in abstract in Example 8.2 and concretely in Example 8.3, features a straightforward closed progression, i–ii 6/5–v–i, perhaps the most basic of tonal progressions. Example 8.3 aligns all seven units affiliated with the model. Its first occurrence is in the tonic D major as unit 5 in a portion of the movement that some have labelled a bridge passage (bars 36–45). The melodic profile of this unit – ending on scale-degree 3ˆ rather than 1ˆ – confers a degree of openness appropriate for transitional function, this despite the closed harmonic form. Conflicts like this – between harmonic

233

234

ko f i ag aw u

Models

1c 1 2

1b

1a

1d

2

3

Exposition 3 4

5 6 7

8

9 10 11

12 13 14 15 16

Development

17 18 19 20 21 22 23 24 25 26

Recapitulation

27 28 29 30 31 32 33 34 35

Coda

36 37

38

39 40 Figure 8.2 Paradigmatic display of all forty units in Beethoven, Op. 18 No. 3/i

tendency (closed) and melodic tendency (open) – are important sources of expressive effect in this style. Other instantiations of the model are units 6 (in B minor), 19 (in B flat major), and 21 (in G minor), all of which display the same thematic profile. Units 7 in A minor and 22 in G minor abbreviate the progression slightly,

Beethoven’s Op. 18 No. 3, first movement

Example 8.2 Origins of Model 1a units

beginning on first inversion rather than root position chords. And unit 20 in B flat major withholds the normative continuation of the unit after the tonic extension of the model’s first two bars. All seven units – 5, 6, 7, 19, 20, 21, 22 – belong to a single paradigmatic class. They are harmonically equivalent, tonally disparate (a factor bracketed at this level), thematically affiliated and melodically similar. Of course they occur at different moments in this sonata-form movement. Unit 5, for example, begins the transition to the second key, but it also is imbued with a sense of closing by virtue of its harmony. Unit 6 signals tonal departure, while unit 7 intensifies the sense of departure with a shift from the B minor of the previous unit to A minor. These units thus perform different syntagmatic functions while sharing a network of (harmonic) features. Ultimately, it is the tension between their equivalence at this level – their resistance to change, so to speak – and the non-equivalence of their temporal positioning and formal functioning that accounts for their meaning. These are purely musical meanings in so far as they emerge from the native tendencies of tones, not from association with extra-musical usage. Hanslick’s ‘forms set in motion by sounding’, Roman Jakobson’s ‘introversive semiosis’, and Lawrence Zbikowski’s ‘dynamic processes’ index roughly the same group of qualities.9 Before proceeding to a consideration of the other tonal models, I should address the potential criticism that these units are too small, that they chop the music up into bits, and therefore that they obscure the larger flow of the movement. The point of extracting equivalent structural units and aligning them as in Example 8.3 is to reinforce the common knowledge that repetition is the basic constructional force in Beethoven’s music. Most discussions of repetition in Beethoven – and for that matter, the canonical demonstrations of the paradigmatic method by Ruwet and Nattiez – refer, however, to rhythmic and melodic motives, not the melodic-harmonic progressions being exposed here. These progressions represent another dimension of Beethovenian repetition that may be understood in dialogue with the more

235

236

ko f i ag aw u

Example 8.3 Model 1a units

‘surface’ repetitions usually acknowledged. Later, I will suggest that repetition confers a circular feel on the movement as a whole, and that such circularity contrasts with the linearity expressed in the tonal dimension. At this stage of the analytical proceeding, however, we leave the larger progression to take care of itself while we focus on the succession of small units. Model 1b expresses a complete i–iv–v–i progression (Example 8.4), but unlike 1a, includes an explicit and prominent flattened seventh (the pitch C) which effectively tonicizes the subdominant, and thus intensifies the

Beethoven’s Op. 18 No. 3, first movement

Example 8.4 Origins of Model 1b units

Example 8.5 Model 1b units

push towards the authentic cadence. Example 8.5 shows six passages totalling forty-two bars that express this model. First is unit 3, followed immediately by a registral variant, unit 4. Notice how unit 33, which recapitulates unit 14, incorporates the flattened seventh within a walking-bass pattern. The thematic profile of Model 1b is more varied than that of Model 1a, all of whose units belong topically to the sphere of bourr´ee. Here, units 3 and 4 are alla breve material (that is to say, they occupy an ecclesiastical register), whereas the others (15, 34, 14 and 33) invoke the galant style with an initial musette

237

238

ko f i ag aw u

Example 8.6 Origins of Model 1c units

inflection (this signals a secular register). Note also that units 15 and 34 exceed the normative five-bar length of the other Model 1b units. This is because I have incorporated the reiterated cadential figures that emphasize the dominant key, A major, in the exposition (unit 15) and its rhyme, D major, in the recapitulation (unit 34). Again remember that all Model 1b units are harmonically equivalent but thematically differentiated. With Model 1c (whose origins are displayed in Example 8.6) we come – again, as it turns out – to the central thematic material of the movement, the opening ten-bar alla breve and its subsequent appearances. Like Models 1a and 1b, its units (gathered in Example 8.7) might be heard as tonally closed (all except 17 and 36), but its journey is a little more protracted. Moreover, the model begins with a prefix (shown in Example 8.6) that does both thematic and harmonic work. Thematic work is evident in the recurrence of the rising minor seventh (sometimes in arpeggiated form) throughout the movement; harmonic work is evident in the tonicization of individual scale degrees that the seventh facilitates as part of a dominant-seventh sonority. I have identified seven passages of differing lengths – ranging from five to thirty-one bars – that express Model 1c. First is unit 1, the opening of the quartet, whose parallel minor version opens the development (unit 17). In a further transposition, the minor version reappears in the coda (unit 36). Incidentally, units 17 and 36 both close deceptively, but if we read vi as a substitute for i, then their membership of this model is valid. Units 39 and 40 are the closing music of the movement; unit 39 is a truncated form of unit 1 (first five notes only), while unit 40 is inflected to convey a broad cadential feel by the incorporation of three bars of six-four harmony found in no other Model 1c unit. Unit 40 also has an extra cadential tag in its last two bars that performs a confirmatory function comparable to that of the five-bar cadential reiterations of Model 1b units 15 and 34 (Example 8.5). The two remaining units of this model, 2 and 27, are expansions of 1. Unit 2 begins in the manner of unit 1 only in a lower register, but delays the drive to the cadence by pausing on a subdominant-functioning chord for the first violin to do cadenza work. An even bigger expansion occurs as

Beethoven’s Op. 18 No. 3, first movement

Example 8.7 Model 1c units

unit 27, the period that opens the recapitulation. Beginning as in unit 1, unit 27 closes deceptively after ten bars, and is then subject to a subdominanttinged internal expansion (bars 168 to roughly 176), before attaining a broad cadence (bars 183–8). Note especially the rising bass line beginning in bar 169, which starts out diatonically (in the orbit of G major) and finishes chromatically (in the orbit of D major). This is early Beethoven at his most organic, postponing the cadence for as long as possible, and expanding the phrase from the middle.

239

240

ko f i ag aw u

Example 8.8 Origins of Model 1d units

Example 8.9 Model 1d units

The fourth and last variant of Model 1, 1d, shares with 1a, 1b and 1c a closed harmonic progression featuring an expanded cadence (Example 8.8 establishes its origins, while Example 8.9 displays its six units). Model 1d, however, incorporates many more chromatic elements. Thematically, its material is in the galant sphere, not the alla breve. Notice that the relative sense of autonomy enshrined in the earlier models – the closed nature of the

Beethoven’s Op. 18 No. 3, first movement

Example 8.10 Origins of Model 1d unit 8

building block, which in turn confers an additive feeling on the succession of units – is undermined, but not entirely erased, by the overlap between each unit and its immediate successor; this also subtracts from the impression of closure within each unit. Indeed, the mobility of Model 1d units contributes not a little to the movement’s organicism. The most disguised of the units in Example 8.9 is unit 8, a Sturm und Drang passage within the second key area. In Example 8.10, I read it as a i–v–i progression, the opening i being minor, its closing counterpart being in major, and the long dominant prolonged by a German-sixth chord. Harmonically, Example 8.10 is plausible, but it is the rhetorically marked three-fold repetition of the augmented-sixth-to-dominant progression that contributes to the disguise. Note, also, that in the second half of unit 8, the treble arpeggiates up a minor seventh before beginning its descent to the tonic. The association with the opening of unit 1 (Model 1c) is thus reinforced. In an important sense, all of the units gathered under Models 1a, 1b, 1c and 1d do more or less the same harmonic – but not tonal – work. A certain degree of redundancy results, and this in turn contributes a feeling of circularity to the formal process. It is as if these 26 units, which comprise 196 out of a total 269 bars (72.86 per cent of the movement) offer a consistent succession of small worlds. The process is additive and circular. The selfcontainment and potential autonomy of the building blocks counters the normative goal-oriented dynamic of sonata form. Models 2 and 3 express i–v and v–i progressions respectively (see the abstracts in Examples 8.11 and 8.14). These progressions are not syntactically defective; rather, they are notionally incomplete. Playing through the Model2 units displayed in Example 8.12 immediately makes evident a progression from a local tonic to its dominant (units 10, 11, 29, 30, 37, 26), or from a local tonic to a non-tonic ‘other’ (units 9 and 28). Unit 10, which heads Model-2 units, is a four-bar antecedent phrase – a question – that might have been answered by a four-bar consequent. The ‘answer’, however, is a transposition of this same unit from C major into its relative, A minor. Thus unit 11 simultaneously embodies contradictory qualities of question and answer at different levels of structure. Units 29 and 30 replay this drama

241

242

ko f i ag aw u

Example 8.11 Origins of Model 2

Example 8.12 Model 2 units

Beethoven’s Op. 18 No. 3, first movement

Example 8.13 Origins of Model 2 unit 26

exactly, now in F major and D minor respectively, while unit 37, initiator of the coda, is answered by a unit that belongs to Model 3 (unit 38). Other Model-2 units are only tenuously related to the model. Unit 26, for example, whose abstract is given as Example 8.13, is the striking passage at the end of the development that closes on a C sharp major chord (where C♯ functions as v of F sharp minor); it marks what Ratner calls the ‘point of furthest remove’.10 Although it is shown as a i–v progression, its beginning (bars 147–8) is less strongly articulated as a phrase-structural beginning and more a part of an already unfolding process. Nevertheless, it is possible to extract a i–v progression from it, the v prolonged by an augmented-sixth chord. Units 9 and 28 are i–v progressions only in a metaphorical sense; essentially, they both resist synopsis. Unit 9 begins in a stable A major with a bass arpeggiation that dissipates its dominant seventh over four bars, and then, by means of chromatic voice-leading, enharmonic substitution and common-tone linkage, reaches the dominant seventh of C major. Similarly, the transpositionally equivalent unit 28 moves from D major to the dominant of F major. When I say that these passages resist synopsis, I mean that they resist a prolongational explanation. Each step in the progression is necessary. It is as if the narrator stepped forward to announce the destination of the story, rather than letting it unfold according to its own principles. The seams of Beethoven’s craft show in these moments; we are reminded of a speaking subject. (Note, incidentally, that evidence for narrativity in Beethoven is typically lodged in transitional sections, where the utterance is often prose-like (as in units 9 and 28) rather than stanzaic or poetic.) The third and last of our models, Model 3, reverses the process in Model 2, progressing from v (sometimes preceded by a predominant sonority) to i. The conceptual origins displayed in Example 8.14 show a passing seventh over a dominant and a chromaticization of that motion. Stated so abstractly, the process sounds straightforward enough, but listening, for example, to the transpositionally equivalent units 16 and 35 (see Example 8.15) suggests that the effect may sometimes be complex. An underlying v–i progression is hearable in retrospect, but because the individual chords sound one per bar, followed in each case by loud rests, the listener’s ability to subsume the entire

243

244

ko f i ag aw u

Example 8.14 Origins of Model 3 units

succession under a single prolongational span becomes difficult. Ratner refers to ‘a series of peremptory, disembodied chords, a play of cut against flow. These chords represent total disorientation.’11 This may be slightly overstated, although Ratner is surely right to draw attention to their effect. Unit 23 features a normative predominant-dominant-tonic progression in A minor, which becomes the model for a sequence encompassing units 24 and 25. The latter feature two and three bars of predominant activity before closing on B minor and F sharp minor respectively. To summarize: the first movement of Op. 18 No. 3 is made from a number of simple two-voice progressions common in the eighteenth century and embodying the basic utterances of the tonal system. By our segmentation, there are essentially three of these models: i–v–i, i–v and v–i. For the complete i–v–i progression we identified four variants: Model 1a with seven units, Model 1b with six units, Model 1c with seven units and Model 1d with six units. For the open or i–v Model 2, we identified eight units, while the closing but not closed v–i Model 3 has six. We have thus heard the entire first movement of Op. 18 No. 3; there are no remainders. We have heard it, however, not in Beethoven’s temporal or real-time order but in a new conceptual order. The question naturally arises: What is one to do with all these data about the forty structural units of the movement? Before answering, I should point out that this kind of question is not always productive. Analysis, like performance or composition, is a mode of doing, not a mechanism for producing summarizable results. Our paradigmatic analysis allowed us to scan the movement for certain elementary progressions, and in the process put us in touch with the detail of Beethoven’s composition. For some, this oneness with Beethoven is what analysis makes possible; and this is reward enough. No grand pronouncements are necessary, and no summary remarks about the ‘meaning’ of the movement as a whole will be adequate.12 This argument will not satisfy all readers, especially those who expect an analysis to produce results that might be summarized. Although the effort to frame such summaries produces limited results, it will be well to point to some of what might be done with the data assembled about the forty units. If we refer back to Figure 8.2, we see the succession of models at

Beethoven’s Op. 18 No. 3, first movement

Example 8.15 Model 3 units

a glance and some of the patterns formed by the forty units. I have also indicated affinities with sonata form in order to facilitate an assessment of the relationship between the paradigmatic process conveyed by repetition and the linear or syntagmatic process implicit in sonata form’s normative narrative. This data can be interpreted in different ways. For example, Model 1c, the main alla-breve material, begins and ends the movement and also marks the onset of both development and recapitulation. It acts as a fulcrum; its seven-fold occurrence may even be read as gesturing towards rondo form. Unit 1a, the little bourr´ee tune, occurs only in clusters (units 5, 6, 7 and 19,

245

246

ko f i ag aw u

20, 21, 22). In this it resembles unit 2, the opening-out i–v progression. Unit 1b, the alla breve in stretto, is totally absent from the development section. Model 3, an expanded cadence, functions like 1c in that it occurs roughly equidistantly in the exposition, development and recapitulation. On another level, one could argue that the essential procedure of the movement is that of variation since, in a deep sense, Models 1a, 1b, 1c, and 1d are variants of the same basic harmonic-contrapuntal progression. This is not a theme and variations in the sense in which a profiled theme provides the basis for thematic, harmonic, modal and figurative exploration. Rather, variation technique is used within the more evident sonata form process. Obviously, then, the paradigmatic analysis can support several different narratives.13

Bridges to free composition In one sense, the essential analytical process is over with this distribution of material into its contrapuntal constituents. We have, as it were, defined the movement’s conditions of possibility. It is now up to individuals to tell their own stories about the first movement of Op. 18 No. 3. Indeed, in an extreme formulation of this stance, it would be irresponsible, perhaps even unethical, to publicize a personal and particular reading. Why not let the enabling structures do their work? Why not confine one’s narratives to the private or semi-private spheres? Strategic reticence – holding back from naming – runs against the grain of much contemporary thinking about analysis. Some of the so-called ‘new musicologists’ wanted to tell stories, specific stories about specific compositions, or moments within those compositions; they wanted to say not how but what those works meant, and they blasted music theorists for confining their analyses to formal technique, to the means of production. Not for them the studied reticence of a parent, elder or mentor, one who seeks to nourish, to enable rather than impress, to teach you how to fish rather than offer you a fish. Fired by the impetuousness and zealousness of adolescence, they insisted on inscribing the worldliness and political rebellion of a Bach, the masculinity and aggressiveness of a Beethoven, the purposeful indirection of a Tchaikovsky and the femininity of a Schubert. I do not intend to stage yet another confrontation with musicology’s selfproclaimed progressives. I allude to their putative critique of the analysis we have developed so far because, like them, many academic musicians probably want something more than the mere enumeration of possibilities. Caught up in the teleology of a composition, some are unwilling – or unable – to imagine a different ordering of its elements; the hypothetical or the

Beethoven’s Op. 18 No. 3, first movement

might-have-been are simply not part of their critical equipment. The intertextual resonances of our models will seem overly general and banal. Why not provide an account that names the thematic substance precisely, lays out the elements of periodicity, and begins at the beginning, goes through a middle, and ends at the ending? Why not, in short, confirm other listeners’ views of the particularity of Op. 18 No. 3/i? Although the play of models that constituted our first reading can help characterize the movement in the ways demanded by these counterassertions, I will respond only to two aspects of these challenges. First, we have to come to terms with the gestures that comprise the work’s surface; in other words, we have to construct bridges from strict counterpoint to free composition. Our models function at a subsurface level, and as revealing as they might be, they are incompletely appreciated and understood if their more immediate features are ignored. Second, we have to take up the issue of form by restoring the actual order in which the units unfold. In both cases we must attempt to reach the foreground. Are such bridges possible? Yes, if the foreground is regarded as a repository of certain diminutions. No, if the same foreground, read as a historicallysituated plane of rhythmic and textural activity (among other dimensional activities), is understood as a repository of diffuse and heterogeneous gestures whose particularity cannot necessarily be inferred from an underlying counterpoint. In the conceptual journey from strict counterpoint to free composition, one invariably reaches a moment when one has to make a leap, a leap of faith, perhaps. You cannot, in other words, get from background to free composition without fudging something. Backgrounds do not generate foregrounds in the consistent way that Schenkerian theory might imply. Foregrounds – or at least the foregrounds made from a plurality of stylistic allusion – participate in other discourses; ontologically, they seem to exist only on one level. In general, the less heterogeneous the foreground, the less jarring is its generative capability from a certain background.14 Again, a lot depends on what one means by the foreground. Suppose we understand it in terms of a signifying surface, one animated by stylistic allusion. We may then return to Op. 18 No. 3 and offer a description of its rich topical content. (Access to the score would be handy for the following description.)

Second reading: topics Topics, Leonard Ratner tells us, are subjects to be incorporated into a musical discourse. Where previous scholars read Haydn, Mozart, Beethoven and

247

248

ko f i ag aw u

their contemporaries in terms of ‘pure’ forms and processes, Ratner sought to remind us of the remarkable extent to which these composers utilized the ordinary musics of the eighteenth century in their art-music compositions. Dance in Mozart, learnedness and comedy in Haydn, and invocations of march, the pastoral and various forms of fantasia in Beethoven: these are among numerous signifying poses adopted by composers. Being alert to topics brings a new and lively dimension to our appreciation of Classic music.15 A rapid topical tour of Op. 18 No. 3, first movement might go as follows (refer again to Figure 8.1). Unit 1 features the alla breve style; unit 2 is also in alla breve style but incorporates messanza figures (bars 17–20) within a cadenza-like gesture. (Messanza denotes a ‘passage in quick even notes with mixed figures, rather than with systematic repetitions of figures’.)16 Also at work in unit 2 is an interplay of duple and triple rhythmic groups. Units 3 and 4 treat the main motivic idea in stretto as part of a confirming progression. There is an emerging sense of the galant here, although the underlying alla breve remains in force. A decisive change takes place at the beginning of unit 5 (bar 36), where a bourr´ee tune in galant style announces a change of stylistic and expressive register. Sturm und Drang and march elements (unit 8) punctuate the move to E major as v/v. Hints of brilliant style mark the preliminary attainment of A major (unit 9); then a short fantasia passage serves as transition to a four-bar march tune (unit 10, bar 68) with a touch of the alla zoppa or limping style. Bursts of fanfare (unit 12) reinforce the proper arrival of A major in bar 76 after which galant and brilliant style elements give profile to the close of the exposition. At the very end (unit 16), a series of chords punctuated by rests functions dually as transition to either the development or the repeat of the exposition. These chords, too, project a fantasia manner. The development begins in alla breve style (unit 17), but the sentiments are darkened by the minor mode. Then the bourr´ee tune returns (in unit 19), followed in units 23, 24 and 25 by a furioso that amplifies the exposition’s Sturm und Drang. Intensification grows until the end of the development (unit 26), which is maximally contrasted with the opening of the recapitulation. Texture, spacing, key and dynamics are all subject to discontinuities between the end of unit 26 and the beginning of unit 27. Like the exposition, the recapitulation begins in alla breve style (unit 27), but the expanded periodicity invokes ricercar style while the change of design in bars 182–7 alludes to march. Brilliant style elements with march still in tow (unit 28) and a fantasia-like section take us to the march tune with alla zoppa (unit 29). The rest of the recapitulation follows the topical order established in the exposition. Finally, the coda (unit 37), too, takes the

Beethoven’s Op. 18 No. 3, first movement

alla-zoppa march tune as point of departure and concludes with the alla breve material, messanza and brilliant style in a modest celebration. Overall, one might say that there is an interplay between alla breve or old style (sacred, ecclesiastical) and the modern, galant style (secular). Because the two topical areas share motivic elements, the movement as a whole may be heard as featuring continuity on one level and discontinuity on another. Does this mean that old and new are equally functional, or that the question of topical dominance is undecidable? I would say not; alla breve seems to have greater phenomenal and symbolic presence – it begins and ends the movement, and functions as a kind of topical or affective or stylistic home, so to speak. Topics are hearable immediately. Their intertextual resonance is available to those who know more than one piece. Their boundaries are fluid, as are their identities, but for first-time listeners, or the students we teach in our music appreciation classes, topics provide a mechanism for anchoring the expression; they make possible a kind of story-telling, not the stories told subjectively by new musicologists, but stories that emanate from the play of style and gesture within an eighteenth-century compositional aesthetic practice, a practice that is rigorously committed to stylized, outwardly-directed play, not the explication or probing of inner feelings. In this sense, there is a difference between early, middle-period and late Beethoven understood from a topical point of view. In early Beethoven, topics signify in roughly the same way as they do in Haydn and Mozart – as part of a public utterance dedicated to stylized play; in middle-period Beethoven, topics recede into the background even as forces of expression generated by organic impulses and a more explicit subjectivism take over; in late Beethoven, topics return with a vengeance, this time shaped as objects with a self-consciously archaic or classicizing feel. Not everyone hears topics, or I should say, not everyone enjoys hearing Beethoven’s sounding surfaces in terms of such stylistic allusion. For some, this is simply an annoying taxonomic exercise, one that breaks the music up into too many little units. Others find the exercise entirely speculative, grounded – they grudgingly admit – in the stylistic world of the eighteenth-century styles and gestures, but not in the reported listening habits of listeners of the time. For some, topics are merely surface elements; they lack a syntax (rules of succession); they are not endowed with formgenerating capability; indeed, like the historically prior affections, they exist only at one level. They are flat, like the narrative moments we encountered in units 9 and 28. And this marks a basic qualitative distinction between the topical mode and the contrapuntal mode discussed earlier.17

249

250

ko f i ag aw u

It is not possible to address each one of these objections here. Let us just say that it is possible to describe the surface of Op. 18 No. 3 using a value-free designation such as ‘design’ instead of quaint terms like alla breve, brilliant style, alla zoppa, and so on. The term ‘design’ allows you to acknowledge textural and thematic changes without committing yourself to a historically specific mode of designation.18 So, to ask the question I asked of our models: what is the value of distributing the reality of Op. 18 No. 3 into topical categories? I have already alluded to a possible plot for the movement: the play of style between strict and galant impulses. But if this kind of plot seems overly general, even banal, perhaps – banal because it is applicable to many other compositions, although it is in the nature of plots that they be widely applicable – we might pursue the designating function of topics rather than their semantics. This allows us to restore ideas from the earlier analysis according to models. The movement may be heard as dramatizing a basic tension between linear and circular tendencies. Linearity is normatively expressed in the sonata form process, whereby a dramatized tonal contrast is followed by a strategic postponement of resolution, and then a stylized reconciliation of the previously opposed elements, concluding sometimes with an afterword. Circularity is inherent in the tonal models we discussed earlier, many of which go round the same progression.19 Consider, finally, bars 1–36 from the points of view of a paradigmatic and a topical analysis. These opening bars correspond to units 1, 2, 3 and 4, and comprise a mini-form within the larger form. In terms of the rhetorical functions of our contrapuntal models, we might say that unit 1 inaugurates the movement, unit 2 expands or amplifies that inaugural statement, while units 3 and 4 confirm the movement’s premise. So while all four units are variants of the same tonal expression, the passage as a whole is not static but dynamic. In terms of topic, one could point to the gradual shift from alla breve (units 1 and 2) to galant style (units 3 and 4). This progression of topics is not as clear cut, however, because alla breve and galant coexist from the beginning, and it is only the gradual and subtle ascendancy of the galant that suggests a shift in topical orientation. In one sense, the succession of topics parallels the progression of sense units, giving profile to the latter; in another sense, the two paths proceed according to their own internal rules. There is a third way of hearing bars 1–36. Leonard Ratner hears a working through the successive phases of an oration: narratio (unit 1), confutatio– distributio (unit 2) and peroratio (units 3 and 4). And yet, understood on another hierarchic level, the entire succession (units 1–4) functions again as a narratio. As the movement continues, however, we may revise these

Beethoven’s Op. 18 No. 3, first movement

attributions retroactively. Units 3 and 4, for example, not only confirm the proceedings in units 1 and 2 but act as agents of change in the pursuit of the movement’s remaining tonal and expressive goals. We might hypothesize a circular futurity as contributing an essential ingredient to the drama of the movement.

Conclusion I have pursued two kinds of analysis, or better, two modes of entry into the world of Beethoven’s Op. 18 No. 3, first movement. First was a paradigmatic analysis which reduced the movement to forty units or little progressions, grouped them on the basis of similarity, and pointed to some of the ways in which one might narrate musical meaning on the basis of their association. Second was a topical or referential analysis which identified certain textures and styles as topics, and sought to open up the immediate and signifying potential of the work’s ostensibly ‘surface’ gestures. With this second proceeding, too, the aim was not to concretize a specific meaning for the movement but to suggest paths to exploration. Although I hinted at some possibilities for synthesis, I have also stressed the prospect of a counterpoint between the two modes. If a synthesis seems difficult to achieve, it may be a sign of a genuine difficulty, namely, the impossibility of building genuine bridges between the world of our tonal models and the heterogeneous world of signifying topics. We should welcome such impasses rather than wish them away, for they may well embody an essential aspect of the work’s ontology. Notes 1 See, however, my ‘How We Got out of Analysis, and How to Get Back in Again’, Music Analysis 23 (2004), 267–86 for a view on the nature of analysis. 2 A sampling of analyses of the first movement of Op. 18 No. 3 would include Joseph Kerman, The Beethoven Quartets (New York: Alfred A. Knopf, 1967), 16–20; Leonard G. Ratner, The Beethoven String Quartets: Compositional Strategies and Rhetoric (Stanford, CA: Stanford Bookstore, 1995), 47–52; and Poundie L. Burstein, ‘Surprising Returns: The vii♯ in Beethoven’s Op. 18, No. 3, and its Antecedents in Haydn’, Music Analysis 17 (1998), 295–312. 3 Nicolas Ruwet, ‘Methods of Analysis in Musicology’, trans. Mark Everist, Music Analysis 6 (1987), 11–36; Jean-Jacques Nattiez, ‘Var`ese’s “Density 21.5”: A Study in Semiological Method’, trans. Anna Barry, Music Analysis 1 (1982), 243–340.

251

252

ko f i ag aw u

4 Leonard G. Ratner, Classic Music: Expression, Form, and Style (New York: Schirmer Books, 1980). 5 Christopher Hasty, ‘Segmentation and Process in Post-Tonal Music’, Music Theory Spectrum 3 (1981), 54–73. 6 Theodor Adorno, Beethoven: The Philosophy of Music, trans. Edmund Jephcott, ed. Rolf Tiedemann (Stanford, CA: Stanford University Press, 1998), 49. 7 On tonal models as determinants of style, see Christopher Wintle, ‘Corelli’s Tonal Models: The Trio Sonata Op. 3, No. 1’, in Nuovissimi Studi Corelliani: Atti del Terzo Congresso Internazionale, Fusignano, 1980, ed. Sergio Durante and Pierluigi Petrobelli (Florence: Olschki, 1982), 29–69 and my ‘Haydn’s Tonal Models: The First Movement of the Piano Sonata in E-flat Major, Hob. XVI:52’, in Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt (Stuyvesant, NY: Pendragon, 1992), 3–22. 8 It goes without saying that alternative segmentations of the movement are possible – even desirable. My interest here, however, is in the plausibility of the units isolated, and not (primarily) in the relative merits of competing segmentations. Whenever my chosen criteria seem especially fragile, however, I offer some explanation. 9 Eduard Hanslick, ‘On the Musically Beautiful’ (Vom Musikalisch-Sch¨onen), trans. Martin Cooper, in Music in European Thought 1851–1912, ed. Bojan Buji´c (Cambridge: Cambridge University Press, 1988), 19; Roman Jakobson, ‘Language in Relation to Other Communication Systems’, in Jakobson, Selected Writings, vol. 2 (The Hague: Mouton, 1971), 704–5, adapted in my Playing with Signs: A Semiotic Interpretation of Classic Music (Princeton: Princeton University Press, 1991), 51–79; and Lawrence M. Zbikowski, ‘Dance Topoi, Sonic Analogues and Musical Grammar: Communicating with Music in the Eighteenth Century’, elsewhere in this volume. 10 Ratner, The Beethoven String Quartets, 51. 11 Ibid., 50. 12 For an elaboration of this view of analysis, with emphasis on the affinities between analysis and performance on the one hand, and analysis and composition on the other, see my ‘How We Got Out of Analysis, and How to Get Back in Again’. 13 On the interpenetration of sonata and variation impulses, see Walter Frisch, Brahms and the Principle of Developing Variation (Berkeley: University of California Press, 1984), 19. In the opening chapter of Bach and the Patterns of Invention (Cambridge, MA: Harvard University Press, 1996), Laurence Dreyfus addresses the paradigmatic-syntagmatic issue by contrasting an organicist reading with a mechanistic one of the first of J. S. Bach’s two-part inventions. An organicist reading follows the course of the composition from beginning to end, noting the journey of keys and the thematic substance. A mechanistic reading by contrast dwells on the subject of the invention, explores and makes explicit

Beethoven’s Op. 18 No. 3, first movement

14

15

16 17

18

19

its structural possibilities, and consigns the task of disposition to a secondary level. On bridges from strict counterpoint to free composition, see Heinrich Schenker, Counterpoint: A Translation of Kontrapunkt, trans. John Rothgeb and J¨urgen Thym (New York: Schirmer Books, 1987). On topics see Leonard G. Ratner, Classic Music: Expression, Form, and Style (New York: Schirmer Books, 1980), 1–30; Wye J. Allanbrook, Rhythmic Gesture in Mozart: Le nozze di Figaro and Don Giovanni (Chicago: University of Chicago Press, 1983); Agawu, Playing with Signs, 26–50; Robert Hatten, Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation (Bloomington: Indiana University Press, 1994) and Interpreting Musical Gestures, Topics, and Tropes: Mozart, Beethoven, Schubert (Bloomington: Indiana University Press, 2004); Carol L. Krumhansl, ‘Topic in Music: An Empirical Study of Memorability, Openness, and Emotion in Mozart’s String Quintet in C major and Beethoven’s String Quartet in A minor’, Music Perception 16 (1998), 119–34; Raymond Monelle, The Sense of Music: Semiotic Essays (Princeton: Princeton University Press, 2000), 14–80, and The Musical Topic: Hunt, Military and Pastoral (Bloomington: Indiana University Press, 2007). For an identification of topics in the Beethoven quartets, see Ratner, The Beethoven String Quartets. My description of Op. 18 No. 3/i follows Ratner’s. Ratner, The Beethoven String Quartets, 330. No one so far has come out against the idea of topic as such, but a number of reservations have been expressed about the enterprise of topical analysis. Lewis Lockwood, for example, denies the relevance of topic to the understanding of late Beethoven: ‘I have never been able to adjust my mind to this way of hearing, by “topics” and allusions, except in cases in which the allusions are obviously and specifically planted for programmatic purposes. This is sometimes the case in the music of Ives, but it is not the case in the late style of Beethoven.’ (Lockwood, ‘Recent Writings on Beethoven’s Late Quartets’, Beethoven Forum 9 (2002), 84– 97.) Susan McClary likewise finds ‘formalist semiotics and the topics mania’ frustrating. (McClary, Review of Monelle, The Sense of Music, Notes 58 (2001), 326.) On the distinction between design and structure, see John Rothgeb, ‘Design as a Key to Structure in Tonal Music’, in Readings in Schenker Analysis and Other Approaches, ed. Maury Yeston (New Haven: Yale University Press, 1977), 72–93. For a valuable study of circularity in a canonical tonal composition, see Robert P. Morgan, ‘Circular Form in the Tristan Prelude’, Journal of the American Musicological Society 53 (2000), 69–104.

253

9

Mozart’s k331, first movement: once more, with feeling wye j. allanbrook

. . . one of the most overanalyzed pieces in the history of music theory . . . Eugene Narmour1 . . . that most overanalyzed piece of music . . . Carl Schachter2 . . . the most-analysed piece in music history . . . V. Kofi Agawu3

Pity the poor Theme! Opinion has it (in the academy at any rate) that the Theme of the Theme and Variations of Mozart’s Piano Sonata in A major, k331, has been talked to death and should be barred indefinitely from use as a musical example. While my epigraphs date from the 1980s, opinion has only hardened in recent years. Fast-forward to 2004: when she called for suggestions for pieces to be discussed in evening study sessions, the organizer of the stimulating conference that gave shape to this volume gently suggested that k331/i and others like it, which ‘have traditionally been debated in music-theoretical writings and handbooks of analysis’, should be avoided at those sessions.4 Rhetoric about the Theme’s overexposure can take a comic turn. For an adroit little article entitled ‘Dismembering Mozart’, Carolyn Abbate and Roger Parker borrowed the title of their piece from the exasperated hyperbole of an unnamed critic: Mozart’s . . . A major piano sonata . . . receives a well-nigh weekly deluge of leaden analytical prose, so much so that one prominent critic publicly voiced the opinion that the sonata now recalls for him nothing less than the appalling fate of Lavinia in Shakespeare’s Titus Andronicus. (Lavinia is violated by two brothers who – in order to ensure her silence – cut out her tongue – and – in case she might write their names – cut off her hands. The brothers, of course, receive their just de(s)serts: all too literally, in fact, as they end up baked in a pie and eaten by their mother.)5

The critic who made this off-the-wall comparison gifted Parker and Abbate with a happy opening for a comic point about obsessive analytical behaviour.6 Certain of Mozart’s works, they argued, suffer grievously at the

Mozart’s K331, first movement

hands of professional theorizers, whose repeated returns to the same source are one with the tired ideas they draw from it, blinded to its actual configurations by the long custom of their a priori assumptions. This compulsion leaves their hapless victims ‘battered and dismembered . . . , brutally denied an opportunity to speak out against those who have assailed them’.7 The prose is hyperbolic, but it merely gilds the truth. If the Theme is not the most frequently analysed piece in music history (Beethoven’s Ninth would be a serious contender), it must qualify as the shortest to be thus eviscerated. And while it has not in truth been physically abused (except, as we shall see, by Hugo Riemann), it has been persistently misunderstood.8 Its expressive voice – a voice that would have sounded out loud and clear to its late eighteenth-century Viennese audience – has been silenced by the torrent of misguided commentary that has constituted its twentieth-century reception history. The perpetrators have by and large been music theorists, who, finding the simple-seeming Theme an apt specimen for experimentation in their sonic laboratories, have adopted it as a primary paradigm. (Music historians have had little of substance to say about it, presumably because they find its apparent simplicity unremarkable.)9 The Theme has been canonized as the archetype of absolute music – a true ‘classical’ phrase, if by ‘classical’ one means simple, formally balanced, and non-referential, coming as close to abstraction as music can, and still remain music. These qualifications allow it to stand in for all music (at least after 1730 or so), and to suffer being pinched and poked at to reveal general truths about the habits of the canonical repertoire. Its chaste symmetries and pellucid surfaces appear transparent, presenting no immediately arresting obstacles to the prober of depths, the revealer of skeletal lines – although a close look may reveal a few odd bones here and there to give the musical anatomist a challenge. Most writers see only the Theme’s simplicity. Edward Cone described the eight-measure period as entirely without ‘unresolved musical thought’. If there is any inconclusiveness to be found, he ringingly states, ‘I have seen no account of it’.10 Some go so far as to blame the confusion of arguments about the Theme on the present condition of music theory: it reflects badly on the state of the discipline that such an uncontroversial example could have spawned this swarm of disagreements.11 Only Fred Lerdahl and Ray Jackendoff, who give the Theme more page-room than any other example in their copiously illustrated A Generative Theory of Tonal Music, hint at possible anatomical anomalies (in defending their use of such a shopworn example): ‘We discuss this already over-discussed passage’, they remark, ‘because, beneath its brevity and apparent simplicity, it is unusually useful in presenting issues of general significance’.12 Yet simplicity remains the

255

256

wye j. allanbrook

Example 9.1 Mozart, Piano Sonata in A major, k331/i, the Theme, bars 1–8

starting point of their analysis.13 Complexities, it seems, do not break the serenity of the paradigmatic surface. As a music historian who has written about k332 and k333, I decided to ignore the ominous warnings about their sister sonata and see what all the fuss was about.14 Since these three sonatas have long been considered a trio (even after Alan Tyson displaced them from one decade into another, leaving the precise dates and cities of composition of k331 and k332 conjectural), there is something to be said for completing the set.15 And it seemed high time to break the tacit taboo against a music-historical analysis. My first task was to scan the existing evidence – to gather a sheaf of examples from this avalanche of Theme-directed prose. Without overmuch effort I have assembled over a dozen fairly extensive (primarily English-language) analyses, plus several brief critiques, or mini-‘meta-analyses’. (I suspect I have barely scratched the surface, and welcome further contributions. The list is attached as an Appendix.) In what follows I will review the theorists’ arguments, I hope not tediously; I find the history of their struggles with the Theme fascinating. Then I will propose a resolution – a cutting of the Gordian knot, if that is not too presumptuous. To be fully envoiced, the Theme must lose its paradigmatic status. The historically and socially grounded features that inform its surface resist the generalizing power that a paradigm requires.

I Just after the turn of the century Hugo Riemann worried over the Theme’s barlines, and two decades later Heinrich Schenker scattered examples from it throughout his work, examples that were assembled by the Italian

Mozart’s K331, first movement

Schenkerian Susanna Pasticci in a 1995 essay that connects the Theme’s Schenkerian dots.16 But the preponderance of the analyses – the body of work I suspect Agawu and company had in mind in their remarks about the Theme’s ubiquity – consists of more recent American researches, by Schenkerians (Robert P. Morgan, David Neumeyer), theorists using modified Schenkerian techniques (Edward T. Cone, Leonard B. Meyer), and theorists offering alternatives to his practices (Joel Lester, Fred Lerdahl and Ray Jackendoff, Eugene Narmour). The agendas of these analyses can be described as ‘structural’. The Theme, imagined as floating in an ideal, ahistorical musical space, unlimited by conventional constraints, is used for systematic probes aimed to discover general principles of music’s mechanics. There is a linear conversation of sorts among these analyses, which spawned the meta-critiques by theorists and semioticians in my collection. Altogether a remarkable amount of attention for such a modest phrase! Schenker’s shadow hangs over this later conversation even though he used the Theme selectively, when it suited his case. It is invoked to illustrate a special condition of the Urlinie in which the Kopfton is 5ˆ , or E, rather than ˆ and also for a phrase that exemplifies Urlinie interrupta, as it were 3ˆ or 8, – the standard middle-ground shape of an antecedent-consequent period: ˆ 3– ˆ 2, ˆ 5ˆ –4– ˆ 3– ˆ 2– ˆ 1ˆ (the first member ends on the all-but tonic, the 5ˆ –4– second goes to ground; see Schenker’s middleground reduction of the first eight bars of the Theme in Example 9.2). In their confident assumptions of what befits a paradigm both examples hint at difficulties to come. Given its weak metrical position, Schenker’s choice of 5ˆ as Kopfton seemed odd enough in practical terms (no matter how much it may please theoretically) ˆ 17 And it to convince analysts of all three persuasions to opt for the 3. little matters to his diagrams that B, the 2ˆ on which the Urlinie pauses for the half-cadence, occurs on the weakest beat of the Theme’s compound measure (bar 4, beat 6), and the 1ˆ on a beat almost as weak (bar 8, beat 4). Because he abstracts pitch from rhythm, Schenker can choose tones conforming to the Ursatz paradigm without concern for a work’s rhythmic anomalies.18 Hence quarrels with his absolutist reductions are beside the point here: a radical dismemberment of its specimens is inherent in pure pitch-structure analysis. It is when theorists grow interested in the study of rhythmic structure, entering the domain of time and history, that musicology has a stake. Music-historical analyses begin with time, in their consideration both of the context provided by history’s longue dur´ee and of habits in the musical parsing of duration, where history is also a determinant. Here the Theme’s functional separation of pitch and metre (manifested in Schenker’s placing of structural nodes on the period’s weakest beats) becomes a major stumbling block.

257

258

wye j. allanbrook

Example 9.2 Schenker’s middleground reduction of the Theme, bars 1–8 (Free Composition, Figure 87.5)

These ‘rhythmists’, as I have dubbed them – Cone, Leonard Meyer, Robert Morgan, and Lerdahl and Jackendoff – in their different ways address a common question: which rhythmic events in the Theme have structural significance? Their aim is to integrate rhythm into the Schenkerian approach. For Cone, musical form is essentially rhythmic (‘not, as conventional analysis would have it, thematic, nor, pace Schenker, harmonic’19 ). Meyer starts his discussion of the Theme with a ‘data-based’ rhythmic analysis of the sort that he set forth in his early work with Grosvenor Cooper.20 Robert Morgan frames the question as an attempt to expand Schenker’s teachings from pitch structure to a rhythmic analogue. Lerdahl and Jackendoff take significant issue with Schenker by asserting the importance of rhythmic placement (‘time-span reduction’) in pitch-structure analysis; they begin on the rhythmic surface and work down. But eventually they call in pitchstructure (‘prolongational’) analysis to make their account workable. Ultimately all these accounts combine modified Schenkerian techniques with their rhythmic investigations, with seriously misleading results. Simply in the process of bringing rhythm into their graphs and diagrams the rhythmists run into heavy weather. Since time cannot suffer the same condensation as pitches, the foreground of the Theme becomes their canvas, and their graphs turn iconic, compelling a closer relation to the contingent than theory might prefer. Diachronic mappings of the score rather than synchronic, symbolic summaries, these equivocal reductions lose much of their Schenkerian purity: some do away with pitches altogether, employing the conventional poetic notation of rhythmic feet instead.21 Example 9.3 reproduces Robert Morgan’s comparison of four rhythmic diagrams of the Theme with his own version. A and B outline hypothetical trochaic and iambic readings, A after the metric theories of Gottfried Weber (1824), who posited all metre as beginning-accented or trochaic, and B after Riemann, who argued the opposite.22 C is Cone’s scansion and reduction, D Meyer’s scansion and E Morgan’s reduction. Whether in poetic or pitch notation, as

Mozart’s K331, first movement

Example 9.3 The Theme, bars 1–8, a` la Weber, Riemann, Cone, Meyer and Morgan (from Morgan, ‘The Theory and Analysis of Tonal Rhythm’, 442–3)

these representations show, where rhythm is the subject the surface cannot be teased out of the picture in the radical way it is in Schenker graphs. Strikingly Morgan, the committed Schenkerian, is more wedded to barlines than Cone. In Morgan’s diagram the last measure’s barline throws a discrepant stress on the dominant – discrepant because the surface marks it piano. This is a sign of trouble ahead. The problem that the theorists tackle with their multi-tasking graphs is a long-standing puzzle about periodic structure, dubbed by theorist Michael Spitzer the ‘head/end ambiguity’: ‘Articulation as metrical accent . . . is orientated toward the head of a phrase’, ‘articulation as punctuation . . .

259

260

wye j. allanbrook

toward the end of a phrase’.23 The generic period’s structural beginning is a downbeat, metrically trochaic, or strong–weak. Its half and full cadences must turn end-oriented, iambic, to achieve convincing rhythmic closure. In other words, the metric tick-tock of strong–weak measures that gets the period going must be reversed in order to drive to the tonal close. Scott Burnham neatly summarizes the problem in Beethoven Hero – ‘the trick of ending with the material conditions of the beginning – on a strong beat and with the tonic harmony – and yet with the opposite effect of closing rather than opening’.24 Burnham offers this formulation in the course of a discussion of Beethoven’s massive Egmont Overture; the generic eight-measure period provides a handily downsized specimen of the problem. Spitzer puts the conflict in historical perspective, suggesting that the late eighteenth-century pedagogue Heinrich Christoph Koch had in his composition treatise already proposed the classic solution. A downbeat can be scanned with what precedes or follows – can be made iambic or trochaic. When a melodic move to 1ˆ coincides with the rhythmic downbeat of measure 8, as happens in ‘normal’ eight-measure periods, the necessary cadential accent results. The first four measures are heard as arsic, the second four as thetic and at a higher level of organization. The Minuet from Eine kleine Nachtmusik (Example 9.4), as so often, can serve as a paradigm.25 But Mozart’s Theme does not behave like Mozart’s Minuet. Compared to ‘normal’ periods, it sports enough anomalous features to seem a woefully unparadigmatic paradigm.26 The rhythmists are searching for organization at a higher formal level: as Cone puts it, for ‘a more organic rhythmic principle [more organic than low-level metric groupings] that supports the melodic and harmonic shape of the phrase and justifies its acceptance as a formal unit’.27 But this quest runs aground because, as I pointed out when discussing Schenker’s example, the Theme’s structural closes are placed on weak beats – beat 6 of measure 4 (the close staved off by the appoggiatura) and beat 4 of measure 8. Cone clearly senses the difficulty this will cause for his ‘organic principle’: in an anxious rhetorical outburst he declares that these weak-beat closes create ‘doubly feminine cadences – nay, in the first phrase, because of its postponed resolution, a triply feminine one!’28 Some solutions to this problem do have rather a desperate air about them. Convinced that duration means accent, Riemann (in the ‘abuse’ I referred to earlier) decided that Mozart’s notation was ‘false’ and took the radical step of re-barring the Theme, moving the barline over by one dotted quarter in order to allow the downbeat to fall on these awkward cadence notes and on the prominent Es and Ds of bars 1, 2, 5 and 6 (Example 9.5; dotted lines indicate Riemann’s altered barrings).29 While later theorists would never be

Mozart’s K331, first movement

Example 9.4 Mozart, Eine kleine Nachtmusik, k525/iii, Menuetto, bars 1–8

so disrespectful of a Classic, Morgan feels the Theme falls more naturally into a 3/8 measure (‘all larger units . . . are ultimately contradicted’), and was forced to the hypothesis that Mozart chose 6/8 only in order ‘to avoid an excessive number of bar lines’.30 Worrying about whether the sixth beat should be slurred across the bar, he was unwilling to commit himself to an iambic or trochaic articulation, finally deciding to leave the measure-tomeasure scansion ambiguous, ‘floating in a state of equilibrium’.31 Meyer, on the other hand, was certain – and rightly so – that the Theme is trochaic, or phrased within the bar, rather than iambic, anacrustic, no matter what the apparent contradictions (he calls the iambic version ‘the aberrant phrasing’ and praises the ‘subtle cool’ of Mozart’s phrase). But as we shall see, he had to work hard to prove something that with a little help from a contextualizing analysis would have been self-evident.32 As for the cadence in bar 8, four out of the five models of Example 9.3, Meyer’s included, disguise rather than resolve its ambiguities. If a cadence occurs, they reason, it must be stressed, and their analyses are – frankly – doctored to reflect this certainty; their arguments claim melodic and

261

262

wye j. allanbrook

Example 9.5 Riemann’s rebarring of the Theme, bars 1–4 (Vademecum der Phrasierung, 60, Example 104a)

rhythmic coincidence willy-nilly, treating the cadential rhythms as though they were their counterparts in the Minuet. The ‘metricists’, Weber and Riemann, organize a simple succession of measures into pairs as trochees (Weber) or iambs (Riemann), mechanically compounding them into increasingly longer spans up to the period, which is itself a replication of the theorist’s chosen measure-type – strong/weak or weak/strong. Weber’s trochees go with the Mozartian flow, which without winning him any particular merit makes him the exception I mentioned above. Riemann, however, sacrifices a strong metric downbeat for an iambic close. At the opposite end of the spectrum, Morgan lets Schenker call the shots, assuming that rhythm will naturally conform to pitch structure. ‘What happens’, he asks, ‘if we accept [Schenker’s] points of structural origin and termination, rather than an a priori metrical schema, as the points of principal accent through which the larger rhythm is regulated?’33 What indeed? The Ursatz is installed as a higher-level a priori schema, and the Theme’s rhythmic surface is diagrammed to coincide with this ‘fundamental structure’. Lerdahl and Jackendoff make the same move. Beginning with rhythmic groupings, they early on subsume their time-span reductions into pitch-structure, or ‘prolongational’ reductions. Time-span reductions, they argue, are ultimately inadequate to musical intuition, that is, to the ‘intuitive judgment that patterns of tension and relaxation are at the heart of musical understanding’.34 In the case of the Theme musical intuition is assumed to stipulate a structural pillar in bar 4 – the ‘first real structural movement’ – and therefore one is posited. While rhythmically grouped as 2 + 2 measures, harmonically, they argue, the first three measures are a unit, and one in which not much happens (mere ‘neighbouring motion’; see Example 9.6). The arc of tension stretches from the opening downbeat to the tonic beginning of bar 4, where it relaxes (they are silent about the ‘doubly feminine’ rhythms that carry the harmony to the half-cadence). Again the ‘more organic rhythmic principle’ turns out to be a function of pitch structure, which asserts its claims over time-span analysis and the maintenance of the troubling trochees by means

Mozart’s K331, first movement

Example 9.6 Lerdahl and Jackendoff ’s introduction to prolongational reduction of the Theme, bars 1–4 (A Generative Theory of Tonal Music, 122, Example 5.14)

of an imported Ursatz-like fundamental skeleton.35 And although one might have thought that the complexity in the Theme hinted at by the authors (see above) was caused by these very trochees, it turns out in their judgment to be a function of the ‘acute noncongruence’ between time-span and prolongational reductions (baldly put, the 3 + 1 harmonic grouping is at war with the rhythmic 2 + 2). The Theme wins points for this: noncongruent phrases ‘have a more complex, elastic quality’.36 But the real problem is obscured. These are the extremes. While Morgan would lump Cone and Meyer together with the two metricists, Weber and Riemann, the two seem to me to hold the middle ground. Given Cone’s confidence in a higher rhythmic principle it is hard to see him as a metricist, and Meyer turns out to be driven by the same impulse. While Weber was not troubled by the lack of a strong terminal accent, and Riemann’s commitment to the holy iamb denied the metric downbeat, Meyer and Cone want them both. Hence they build into their diagrams a higher-level reversal that turns the downbeat trochee into a terminal iamb. Even Meyer succumbs to the structuralists’ imperative.37 As his iambic rhythmic units accumulate they overwhelm the metric ticking of the trochees, turning them iambic to compel the cadence in bar 8 (see the closing trochee-become-iamb in the top level of his metrical scheme, Example 9.3d).38 Cone, on the other hand, although he includes a rhythmic scansion like Meyer’s, is halfway to Morgan’s prioritizing of pitch-structure analysis in that he plucks that (‘highly abstract’) principle of rhythmic energy out of the air, and forces the cadential accents of his middleground analysis to conform; the rhythmic feet in Example 9.3c are after the fact, not causal like Meyer’s. Despite their differences, however, all these schemes impose on the Theme obvious a priori agendas that (with the exception of Weber’s fortunate choice of trochees a priori) force strong closing accents squarely on those ‘triply and doubly feminine cadence tones’. Identifying these tones as strong, despite

263

264

wye j. allanbrook

the inescapable fact of their weak rhythmic placement, they posit a dynamic arc that stretches from the opening downbeat to bar 8’s closing cadence – Cone’s ‘more organic rhythmic principle’.39 Cone characterizes the energy of this principle in the image of throwing a ball, Lerdahl and Jackendoff in the stretching of a rubber band, which in bar 4 (and bar 8 as well, of course) ‘belatedly springs loose’ to home in on the target cadence.40 As I pointed out earlier, Cone was so satisfied with the solution as to declare these eight measures clear and distinct – entirely without ‘unresolved musical thought’. But this is demonstrably not so. It is difficult if not impossible to think away the uncomfortable non-coincidence of rhythm and pitch in the blurred and hesitant cadences of the first strain; they leave the period without an energetic, confirming close. And Mozart made this clear by placing sforzandi on the subdominant members of both the half and the full cadences, both on weak beats, followed subito by piano markings for the cadence tones that are the putative goals of the rhythmic arc (see Example 9.1). Cone explains away the sforzando in bar 7, claming that it was intended to ‘throw the emphasis forward in such a way as to suggest that the concluding measure, even though piano, is cadentially strong’, and Meyer argues essentially the same thing.41 But the sforzandi work to prevent such a hearing, guaranteeing that the piano v and v–i cadences that follow them will remain trochaic dying falls rather than compelling iambic closes. ‘Throwing the emphasis forward’ would involve a slur over the barline (or across the bar-half in the case of the half cadence), which as we have seen is counter-indicated both by the inertial weight of the ongoing trochees and by Mozart’s instructions: he places a slur only over the i6/4–v, with in effect, because of the preceding sforzandi, subito pianos on the i6/4s that thwart the urge to give the cadential chord its conventional home-coming thump. Since the sforzandi are not slurred with this complex, they must be played slightly d´etach´e, cut off in mid outcry. Hence the lulling trochaic rhythms dominate through to the close – no iambs here! Accustomed to homing-in on dominants and tonics, most performers audibly find these accents a struggle. It is difficult to resist the impulse to cheat by emphasizing instead bar 4’s six-four chord or the tonic in bar 8. But given Mozart’s careful placement of these dynamic accents, even Riemann’s violent re-barring would not license such a rendition. Ironically, only Weber’s simple metric model, naively mechanical as it may be, reflects the simple truth of Mozart’s ongoing, enervating trochees. At the Theme’s first authentic cadence the dynamic arc fails to fall to its foreordained close; the ball slips into the catcher’s mitt without a satisfying clunk.

Mozart’s K331, first movement

II It is astonishing how many of these difficulties fall away if one situates the Theme in its historical context and examines its particular brand of play with expressive signs. If the shapely periods of the Theme are approached with a topical analysis in mind, even this paradigm of ‘Classic’ purity reveals a mimetic nature, and topicality ‘saves the appearances’ of its apparent rhythmic anomalies. Not one of the analyses makes note of the fact that the presumptively minimalist Theme is cast in the familiar rhythms of an idealized peasant dance – the siciliano.42 But 6/8 is a pastoral metre, and its patterns here exhibit the leisurely dottings characteristic of the siciliano topos, a subset of the pastorale (supported by the Andante grazioso marking).43 Hence right at the start Mozart’s musical landscape is peopled, if only with an imaginary – Arcadian – race stepping decorously to its nostalgic cadences.44 Each of the Theme’s first eight measures displays the characteristic siciliano pattern of hesitation and release; every measure bears its ‘feminine’ (trochaic) accent pattern – the strong–weak metric arrangement that so troubled the rhythmists. The continuity of the pastoral affect here requires self-contained 6/8 measures; had Meyer identified the dance topos, it would have provided him with a conclusive argument against slurring over the barline. The only justification for such a slur would be if the Theme began with an eighthnote upbeat, and in other sicilianos even that pattern will change within the phrase. For example, in the variation movement of Mozart’s Sonata for Piano and Violin in F major, k377 (composed around the time conjectured for k331), Variation vi, a ‘Siciliana’, begins with an upbeat (see Example 9.7). But the upbeat only provides rhythmic impetus at the start: circling melodic patterns and occasional large intervallic gaps between measures, crunchy chords, and the almost complete absence of notated across-the-barline slurs (one in the X-section, bars 11–12, over a super-upbeat) rule out any effort to set up a running anacrustic pattern. All the more reason why the highest notes of the opening measures of our upbeat-less Theme – the repeated Es in bars 1 and 5 and repeated Ds in 2 and 6, that so troubled Riemann and Morgan and gave Schenker his Kopfton – must be performed without accent, and with a slight articulation before the next measure begins, in order to maintain separation. In fact, Mozart rarely phrased over the bar line, and would add slurs or other markings to indicate those occasional instances when he wanted to go against his established practice.45 The slurs added in some editions are the impulse of a romanticized lyric imagination, uncomfortable with the prim circumspection of the dance measures.46

265

266

wye j. allanbrook

Example 9.7 Mozart, Sonata for Piano and Violin in F major, k377/ii, Tema con Variazioni, Variation vi (‘Siciliana’), bars 1–8

As for its cadences, the Theme tends to the extreme even among sicilianos. In other siciliano movements by Mozart – the siciliano in Example 9.7, the ravishing siciliano in the second movement of the Piano Concerto in A major, k488/ii, the terse theme of the Theme and Variations of the String Quartet in D minor, k421/iv, or the melancholy Rondo in A minor, k505 – Mozart tended to tidy up, or ‘galantize’ the siciliano, planting its full cadences, and often its half-cadences, neatly on a downbeat. The Theme’s mid-measure cadences, however, are in the tradition of Handel’s instrumental sicilianos and the siciliano movements in Haydn’s string quartets (for instance, Op. 20 No. 5/iii and Op. 33 No. 5/iv). But the Theme’s harmonies squeeze the cadential function to the limit; nowhere in these other examples does there exist a cadence like the half-cadence in the Theme’s bar 4 (Cone’s ‘triply feminine’ cadence), where beat 1 is a tonic, and the cadential 6/4 does not occur till beat 4. This particular awkwardness is special to the Theme, for affective reasons. For rhythm is not the only casualty of the theorists’ discussions. Prolongational analysis must also suppress the only striking harmonic phenomenon in the placid first reprise – the ‘vi7 -ish’ configuration that Lerdahl and Jackendoff term ‘hardly a chord, in the normal sense’.47 It too is a precipitate

Mozart’s K331, first movement

Example 9.8 Mozart, k331/i, normative versions of bars 3–4 and 7–8

of the pastoral topos. The compound metre and the dotted ‘hesitation steps’ on the first beats of the first two measures create a rocking, lulling rhythm, so that the listener does not resist when the parallel thirds continue their hypnotic descent, reaching down in bar 3 beyond the conventional tonal G♯–B to an analogous but modal F♯–A. The descent is tethered, so to speak, by the inner E pedal to which the A, G♯ and F♯ are pendant; it controls this affective slipping. The line’s stepwise motion rationalizes its termination in a modal degree. Colouristic rather than functional, this formation is hors de combat in structural analyses, banished at the middleground to ensure the clean, right-angled horn fifths of the Ursatz. Although the emblematic descent and return comprise three-quarters of each phrase member, on the theorists’ deeper strata, as we have seen, this ‘affective prolongation’ leaves no trace.48 To a theorizer of the surface, however, the modal chord’s colouristic quality evokes the never-never-land of Arcadia; this sepia modality is part of the palette of the nostalgic pastoral affect. Lerdahl and Jackendoff were right about the ‘noncongruence’ of the rhythmic and tonal domains in the Theme, but for the wrong reasons. Bars 1–41 – precisely those bars that they dismissed as ‘mere neighbouring motion’ (see Example 9.6) – are the working body of the phrase, in that they are responsible for projecting the affect. The affective field stretches from the downbeat tonic of bar 1 to the downbeat tonic of bar 4, leaving a scant three and one-half beats for the entire half-cadence formula – ii6 –i 6/4–v. In other words, an unusually long non-cadential passage luxuriates in the siciliano affect, and the cadence, a continuation of the inertial trochaic rhythms, is tossed off – an unstressed closural sign. I say ‘unusually’ because even the most murmuring, lulling sicilianos tend to maintain a more equal ratio between macro-rhythm and harmony.49 A less evocative but more ‘well-formed’ version of the Theme might move to tonic harmony on the first beat of bar 3 (in place of the suggestive F♯-based chord), to a i 6/4 on the first beat of bar 4, and to a solid tonic on the first beat of bar 8 (see Example 9.8). The Theme’s ‘belated’ arrival at its cadence points allows an unusually extended presentational space for

267

268

wye j. allanbrook

the pastoral affect, and explains the sforzando-piano markings, especially the one in bar 4: they protect the unstressed cadences from overeager accentuation – from closural dramatization of a cadence that is meant barely to register. A cadence is a conventional introversive sign that does not always require dramatic staging to do its work; even when tossed off, its pitches will still close the deal. The theorists’ metaphors of the thrown ball and taut elastic invert the priorities of these first eight bars. They are intended to evoke the expressive climate, not to achieve a sturdy close; the priority of the first reprise is continuity of affect. And the displaced accents of the sforzandi perform an expressive, colouristic function in addition to their rhythmic ˆ on the ‘down side’, the ‘dark side’ of tonality, so the one. 4ˆ is related to 6, subdominant-function sforzandi invoke the nostalgic colour of the descent to 6ˆ as they deliberately downplay the predictable v and v–i harmony of the close. Only briefly do the tentative cadences pull the listener back from the pastoral penumbra to a more conventional syntactic world. But cadencing is ever a priority in this style; Mozart did not plan to depend on those languid chords for ultimate closure. They are the weakest members in a hierarchy of cadences that Mozart develops over the Theme’s eighteen bars. As I pointed out earlier, the rhythmists seem to have forgotten that the Theme does not end at bar 8; their analyses are predicated on an eight-bar whole. Conveniently, the end of the Theme’s first reprise closes on the tonic rather than on the more usual dominant, fostering this illusion. But tonic closure at the halfway mark is historically a perfectly acceptable practice for the first reprise of short binary pieces. Koch attests to this in his Versuch, and it can be seen in countless late eighteenth-century minuets.50 Mozart’s solution can already be intuited in the total of the Theme’s bars – eighteen rather than the theorists’ eight or the sixteen of simple symmetry (Example 9.9). There is another reprise to come, with a two-bar extension in which an increased intensity in the topical drama will enable the issue of the tossedoff cadences to be revisited and resolved. The four-bar ‘X section’ following the double bar packs a real punch: it brings the Theme to an expressive crux by introducing a new topos, which with minimal rearrangement of the Theme’s opening elements causes maximal topical upheaval. The rise of the leading voice from 3ˆ to 5ˆ and the introduction of an Alberti bass (the sign of a ‘singing allegro’ topos) briefly transform the placid siciliano material, dots and all, into a veritable lyric outcry. Usually the singing allegro – a figure-and-ground configuration consisting of a song-like treble line over a bass of repeated notes or broken chord figures (the ‘Alberti bass’) and set in a moderately lively tempo – is a force for stabilization. Hence its frequent employment as a consolidating

Mozart’s K331, first movement

Example 9.9 The Theme, bars 1–18

‘second theme’. But after the lulling pastoral rhythms, it stirs a strong sense of urgency, aided by the halving of the rate of harmonic change. In bar 10, on an earnest subdominant over the Alberti’s tonic pedal, the singer reaches up to the melodic apex (a2 ) for the first time, via an expressive Schleifer that suggests improvisatory abandon – the lyric mode finally released from the prim paces of the dance. The Schleifer is a gap-filling ornament; here it fills in the interval between f♯2 and a2 . The interpolated, ‘extra’ notes convey emotional urgency. Completing a rising line in direct contrast to the slow descent of the opening, this melodic climax occurs on a downbeat. The lulling single-bar trochees of the first strain give way momentarily to a more complex rhythmic organization in which bar 9 functions as an upbeat to the melodic high of bar 10. Emphatic downbeat sforzando accents in bars 11 and 12 and a chromatically altered subdominant allow the straightforward passion unleashed by the two previous bars to play itself out in the ‘aria’s’ half cadence, where the sforzandi momentarily resolve the conflict between harmony and rhythm. The performer should not overstress the first reprise’s sforzandi, but – a hint of the passion to come – let them flash out momentarily in the regular siciliano motion. After the muted, hypnotic affect of the first reprise, the second reprise comes as a full-throated emotional outcry. The shepherdess sings out. The dancer becomes a diva.

269

270

wye j. allanbrook

This expressive ripple in the smooth surface of the siciliano provides a disturbance that drives through to an emphatic rhythmic climax at the close. Set off against the lulling isometric motion of the first reprise, the lyrical heartfeltness of the singing style introduces a strongly marked rhythmic energy. Consequently when the second member of the original phrase returns almost unaltered, it is not sufficient for closure, and the full authentic cadence is deflected by a ‘feminine’ double-appoggiatura cadence in bar ˆ This discreetly inconclusive articulation 16 (3ˆ in the treble instead of 1). calls out a more emphatic two-bar extension to bring the Theme to firm closure on a sturdy tonic downbeat. Here at last closure receives its proper staging, made all the more dramatic by contrast with the unstressed closures of the first reprise. Robust thick chords in a choral-like texture and a straightforward subdominant (on the last beat of bar 17) take over from the more ‘poetic’ ii6 chords of the original cadence formulae. They are bolstered by a sustained forte dynamic (the only hint of forte earlier came in the offbeat sforzandi accents). The subdominant Schleifer figure returns, its new offbeat position making it an iambic upbeat to the octave’s fall to the i 6/4 on beat 1 of bar 18, finally allowing that all-important antepenultimate chord its proper semiotic weight. In this second ascent to the high point the X-section’s poignant cry finds both rhythmic and melodic resolution, and, across the broader arc of the period, so do the ruminative cadence-denying rhythms of the first reprise. No longer is there any obstacle to the performer’s hammering out the remainder of the cadence in a conventionally celebratory mode. Schenker and others call these two bars a ‘coda’, an afterthought (oddly, since their analyses locate the third fundamental tone in the final bars of the Theme).51 But when rhythm enters the picture, it is clear that these two bars are critical to proper closure and hence indispensable (should I say ‘organic’?). If there is a dynamic trajectory to the Theme – the arc of a pitched ball or a stretched elastic – it is from bar 1 to the forthright forte cadence of bars 17–18, and it is engineered not by the powerful gears of an underground Ursatz, but by delicate balances wrought by expressive topoi on the surface. Before I made this study I might have thought that, given the stated intent of their analyses, it would not be fair to chastise the theorists for letting even such vivid imitations go unremarked. But I have come away with the strong conviction that as a result of their principled neglect of its historically grounded features, they ever but slenderly knew the piece. For truly ‘knowing’ it to any significant extent has required the re-marrying of the two perspectives. I cannot offer a universal solution to the conundrum of

Mozart’s K331, first movement

how opening trochees become closing iambs, but I hope to have shown how the filters introduced by investigators bent on their structural inquiries sifted out details that are inherent to the nature of their chosen object. As a result, their conclusions are not simply non-generalizable (as might befit a paradigm that revealed just a few too many underground complexities). Rather, I think it fair to say that they are erroneous. My quarrel is not with the point the theorists were trying to make: there are countless eight-bar periods that neatly transform beginning trochees into final iambs, the problem being perhaps no more complicated than Koch made it out to be. But the Theme is not one of those periods, at least until its final cadence – a cadence the theorists left out. The Theme is a special case; the little phrase has caused a major musico-theoretical ruckus essentially by mistake. The folly of making a paradigm out of an exception has its comic side, but it is also, as I hope to have shown, extremely revealing. Music is not merely pitch-structure; structural cadences are not the universal goal of every phrase. The conviction that this must be so has forced the exclusion of a rich vein of evidence on the grounds of its mere contingency – the taint it bears of the world. So, we must ask, what kinds of knowledge are admissible in music analysis? Is anything ‘extra-musical’? Why the apparent embarrassment with the fact that music has a history? Historical, socially mediated knowledge is not mere ‘soft’ knowledge, not less constitutive of a piece than its ‘natural’ tonal and rhythmic configurations. Musical topoi, for example, despite (or more correctly, because of) their ‘uniquely historicist discourse’, their saturation with reflections of their particular social world, are not extra-musical after the fashion of an imaginary programme.52 They are musical in a wholly non-contingent sense. Not only are they responsible for the expressive difference of the piece – what distinguishes the movement from others labelled ‘Andante grazioso’ – but they are as likely as any other musical parameters to effect, and in retrospect to explicate, seemingly abstract structural moves. Given the almost universal lack of recognition of the Theme’s topos in the musicological literature, it may come as a surprise even to music historians and friendly theorists that the ‘simple’ Theme is no more transparent in affect than it is in rhythm. We have all been guilty of viewing it as a bland blank canvas on which Mozart embroidered the ensuing variations; we have all ignored the topoi so perspicuously marked at the start. It is my hope that the uncovering of substantive topical references even there will help us begin to comprehend that expression – social communication – is always a parameter in late eighteenth-century music, although the expressive gestures may be less salient at one moment than another. Hence analysts should attend to it with the same constancy and rigour that they give to a work’s tonal plan, formal nodes, or structural dissonances. Not all topics are equally

271

272

wye j. allanbrook

marked, it is true. We can say about one passage that it is a military march or a sarabande, about another only that it is brilliant or lyrical. But ‘brilliant’ and ‘lyrical’ name considerable differences in expressive temperature. No moment in the persistent chiaroscuro of this music is ever expressively neutral; when it ceases to be A, it must be B or C or a cadence (topical markedness often falls off when it comes time for closure despite the regular interplay between topic and syntax). The topoi articulate each other’s differences in the same manner that modern linguists understand phonic units as delimiting each other – by juxtaposition and opposition, rubbing shoulders, ‘jostling each other about’. And just as in language, it is often far easier for the performer to execute and the listener to comprehend these differences than it is for the critic to analyse them. Ferdinand de Saussure famously articulated this semiotic problem for language: ‘Doubtless speakers are unaware of the practical difficulties of delimiting units . . . It is one thing to feel the quick, delicate interplay of units and quite another to account for them through methodical analysis.’53 Substitute ‘performers’ or ‘listeners’ for ‘speakers’ in this quotation and the problem is restated in terms of the expressive code of late eighteenth-century music. But despite the difficulty of articulating the units of musical topoi, their expressive stances saturate this music; they are never not a parameter, even in the absence of a convenient style name or obvious historical association.54 To articulate these differentiations is an indispensable analytical task. I do not think we can claim to know ‘how the music goes’ until we erase the line our disciplines have drawn in the sand between putatively pure relationships and the historical and cultural constraints that informed compositional choices and form the bond of communication between composer and audience. The barrier we so often erect between music’s theory and music’s history brings to mind another classic tale of mutilation, more universal in application than the silencing of the fair Lavinia because it attempts to explain the Fall. In Plato’s Symposium the comic poet Aristophanes describes heterosexual lovers as originally hermaphrodites, two genders bound in one body, split apart by Zeus in order to vitiate the powers they possessed as wholes, and forced to wander the world forlornly searching for their other halves. The unnatural sundering of music theory and history is academic music’s own Fall from grace, a point for which the mistreatment of the little Theme is compelling evidence. Leonard Meyer concluded his memorable 1988 SMT keynote address (see note 38) with a plea to his fellow theorists to reunite themselves with their missing complement: ‘It is the nature of human beings to be nurtured by cultural constraints. And the time has come, this Walrus thinks, for music theorists and psychologists to consider seriously the claims of culture, and of history.’55 Nearly two decades later, the Walrus

Mozart’s K331, first movement

has still not been given much reason to be gratified: a few theorists have taken up Meyer’s invitation, but they are notable exceptions.56 Failure to consult the historical context at worst falsifies analysis, and at best leaves it unedifyingly incomplete. To study a work by Mozart – or Haydn or Salieri or Beethoven – without bringing these ‘superficial’ gestures to consciousness is truly an act of dismemberment.57

Appendix: Analyses of the theme of k331/i (in chronological order, grouped by author) Hugo Riemann, Vademecum der Phrasierung (Berlin: Max Hesse, 1912), 57–61. Heinrich Schenker, Das Meisterwerk in der Musik, volume 1 (Munich and Vienna: Drei Masken Verlag, 1925); translated by Ian Bent et al. as The Masterwork in Music, ed. William Drabkin, Cambridge Studies in Music Theory and Analysis 5, vol. 1 (Cambridge: Cambridge University Press, 1994), 105; and Neue musikalische Theorien und Phantasien, vol. 3: Der freie Satz (Vienna: Universal Edition, 1935); translated by Ernest Oster as Free Composition (New York: Longman, 1979), Figures 72.3, 87.5, 95.b4, 132.6, 141, 157. Edward T. Cone, Musical Form and Musical Performance (New York: Norton, 1968), 28–31, 40–1, 43–4, 75. Leonard B. Meyer, Emotion and Meaning in Music (Chicago: University of Chicago Press, 1956), 109–10; Music, the Arts, and Ideas (Chicago: University of Chicago Press, 1967), 309–10; Explaining Music: Essays and Explorations (Berkeley: University of California Press, 1973), 26– 43, 94–5. Robert P. Morgan, ‘The Theory and Analysis of Tonal Rhythm’, The Musical Quarterly 64 (1978), 435–73. Walter Piston, Harmony, fourth edition, revised and expanded by Mark DeVoto (New York: Norton, 1978), 92–3, 101–3. Joel Lester, ‘Articulation of Tonal Structures as a Criterion for Analytic Choices’, Music Theory Spectrum 1 (1979), 73–9. Fred Lerdahl and Ray Jackendoff, A Generative Theory of Tonal Music (Cambridge, MA: The MIT Press, 1983), 32–3, 63–4, 70–1, 88, 118–23, 134–5, 138–41, 156–8, 162–5, 171–4, 194, 227–33, 276–7, 338, 339, 341. See also Jackendoff, Review of Leonard Bernstein’s The Unanswered Question, Language 53 (1977), 883–94; Lerdahl and Jackendoff, ‘Generative Music Theory and its Relation to Psychology’, Journal of Music Theory 25 (1981), 45–90.

273

274

wye j. allanbrook

Eugene Narmour, ‘Some Major Theoretical Problems Concerning the Concept of Hierarchy in the Analysis of Tonal Music’, Music Perception 1 (1983), 160–98. Edward T. Cone, Musical Form and Musical Performance (New York: Norton, 1968), 28–31, 40–1, 43–4, 75; ‘Twelfth Night’, Journal of Musicological Research 7 (1987), 136–41. David Neumeyer, ‘The Three-Part Ursatz’, In Theory Only 10 (1987), 26–7. William Drabkin, Susanna Pasticci and Egidio Pozzi, Analisi schenkeriana: Per un’interpretazione organica della struttura musicale (Lucca: Libreria Musicale Italiana, 1995), 57–72 and Table 2. David Epstein, Shaping Time: Music, the Brain, and Performance (New York: Schirmer Books, 1995), 47–50. ¨ Claudia Maurer Zenck, Vom Takt: Uberlegungen zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert (Vienna: B¨ohlau, 2001), 311–15. This list omits the numerous uses of the Theme in scholarship on music perception and cognition (for example, music psychologist Alf Gabrielsson has done two studies measuring deviations from the notational norm in performance of the first eight measures of the Theme; they do not constitute an analysis, instead relying on Meyer, Lerdahl and Jackendoff, and Narmour to explain their findings) and also its frequent appearances in basic music history and theory texts (except for DeVoto’s radically incorrect Schenkerian analysis in his fourth edition of Piston – misconstruing the troublesome bar 3 F♯, his skeleton shows two full cadences; it figures in Narmour, 176–81; the Schenker graph was removed from the fifth edition). Worth mentioning is the fact that k331/i wins the prize for highest page count among the examples in Allen Forte and Stephen E. Gilbert’s textbook Introduction to Schenkerian Analysis (New York: Norton, 1982). Alf Gabrielsson, ‘Once Again: The Theme from Mozart’s Piano Sonata in A major (K. 331): A Comparison of Five Performances’, in Action and Perception in Rhythm and Music, ed. Alf Gabrielsson (Stockholm: Royal Swedish Academy of Music, 1987), 81–103.

Meta-analyses Kofi Agawu and Jean-Jacques Nattiez each briefly critique Cone’s and Meyer’s earlier accounts of the Theme, Lawrence Zbikowski includes a stimulating analysis of the Cone, Meyer and Morgan discussions in his

Mozart’s K331, first movement

Conceptualizing Music (see note 39), and most recently Michael Spitzer has used the Theme to question the validity of the eight-measure period as a paradigm. In reviews by John Peel and Wayne Slawson and by David Harvey, Lerdahl and Jackendoff ’s reading of the Theme is compared with Schenker’s. David Beach makes special mention of the constellation of discussions of the movement in his overview of the field, and lists most of them in his bibliography. Carl Schachter opens an essay on metre in Music Forum with a brief characterization of five of the major contributions to the Question of the Theme. Jean-Jacques Nattiez, ‘The Concepts of Plot and Seriation Process in Music Analysis’, trans. Catherine Dale, Music Analysis 4 (1985), 112–14. John Peel and Wayne Slawson, Review of Lerdahl and Jackendoff ’s A Generative Theory of Tonal Music, Journal of Music Theory 28 (1984), 282–7. David Harvey, Review of Lerdahl and Jackendoff’s A Generative Theory of Tonal Music, Music Analysis 4 (1985), 296–9. Carl Schachter, ‘Rhythm and Linear Analysis: Aspects of Meter’, in Music Forum 6, ed. Felix Salzer and Carl Schachter (New York: Columbia University Press, 1987), 1–2. Kofi Agawu, ‘Schenkerian Notation in Theory and Practice’, Music Analysis 8 (1989), 279. David Beach, ‘The Current State of Schenkerian Research’, Acta Musicologica 57 (1985), 294–5; ‘Schenkerian Theory’, Music Theory Spectrum 11 (1989), 5, 12. Lawrence Zbikowski, Conceptualizing Music: Cognitive Structure, Theory, and Analysis (Oxford: Oxford University Press, 2002), 306–10, 321–4. Michael Spitzer, Metaphor and Musical Thought (Chicago: University of Chicago Press, 2004), 73–6.

Notes 1 Narmour proceeded nonetheless to add a thirty-nine-page discussion of the Theme (one of the longest) to the ever-growing total (Narmour, ‘Some Major Theoretical Problems’, 160–98). Where a full citation does not accompany a first reference, that citation can be found in the Appendix. 2 Schachter, ‘Aspects of Meter’, 2. 3 Agawu, ‘Schenkerian Notation in Theory and Practice’, 279. 4 Danuta Mirka, email to conference participants, 12 December 2004. Lest my decision to make the Theme the theme of my conference paper should seem to have been intentionally provocative, I hasten to add that it was taken before

275

276

wye j. allanbrook

5 6

7 8 9

10

11 12

13

14

15

Mirka made her request. Only later did I hesitatingly confess that I was going to sneak the Theme in by the front door. Carolyn Abbate and Roger Parker, ‘Dismembering Mozart’, Cambridge Opera Journal 2 (1990), 190. He turns out to be Carl Schachter, in the article that produced my second epigraph (Schachter, ‘Aspects of Meter’). I am grateful to Danuta Mirka for informing me of the source for Parker and Abbate’s Shakespearean riff. Abbate and Parker, ‘Dismembering Mozart’, 190. For Riemann’s ‘abuses’, see below, pages 260 and 262. Friedrich Blume, for example, summarily dismissed the Theme as a mere ‘example of alignment [the repetition of the strain?] without important harmonic, rhythmic, dynamic, or coloristic contrasts’ (Friedrich Blume, Classic and Romantic Music: A Comprehensive Survey, trans. M. D. Herter Norton (New York: Norton, 1970), 50). There is, however, one recent exception to the Theme’s musicological neglect in Matthew Head’s Orientalism, Masquerade and Mozart’s Turkish Music, Royal Musical Association Monographs No. 9 (London: Royal Musical Association, 2000), 112–32. Head makes up for previous inattention to the Theme – not only the musicologists’ sin of omission, but also the theorists’ failure to contextualize. See note 44 for a brief account of his discussion. Cone, ‘Twelfth Night’, 137. Actually, as the alert reader will note, the Theme consists not of eight but of eighteen measures. Most of the analyses in question, however, treat it as if it had only eight. More on this to follow. Schachter, ‘Aspects of Meter’, 2; Narmour, ‘Some Major Theoretical Problems’, 176, 196. Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 338, note 7; emphasis mine. The dominance of the Theme in this book is not merely statistical: it is used at the opening of almost every new discussion (second only to the chorale ‘O Haupt voll Blut und Wunden’ from St Matthew Passion). The tacit premise of the authors’ curiously elliptical statement appears to be something like ‘simplicity concealing complexity is a requirement for a paradigm’. For a brief discussion of what may constitute a paradigm, see note 26. For my discussion of k332/i and k333/i, see Wye J. Allanbrook, ‘Two Threads Through the Labyrinth: Topic and Process in the First Movements of K. 332 and K. 333’, in Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt (New York: Pendragon, 1992), 125–72. Alfred Einstein had dated the three sonatas to Paris 1778, a decision for which Alan Tyson says there is no evidence. Working from his paper studies, Tyson argues that k333 is the unidentified ‘Linz’ sonata, written in that city in November 1783, and that k331 and k332 were probably composed in Salzburg earlier in the same year. (Alan Tyson, Mozart: Studies of the Autograph Scores (Cambridge, MA: Harvard University Press, 1987), 20, 29–30 and 73–81.)

Mozart’s K331, first movement

16 Pasticci tidied up for the master by shaping into essay form the scattered hints he dropped about the Theme in Free Composition, ‘deepen[ing] them, and arrang[ing] them in a unitary framework’ (Drabkin, Pasticci and Pozzi, Analisi schenkeriana, 58–72 and Table 2). 17 Cone, Musical Form and Musical Performance, 28; Morgan, ‘The Theory and Analysis of Tonal Rhythm’, 449 and Example 3E; Lester, ‘Articulation of Tonal Structures’, 73; Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 162; Narmour, ‘Some Major Theoretical Problems’, 185. Schenker’s version does have its passionate defenders, however, for instance David Harvey, who feels Schenker ‘has been done an injustice’ by his critics (Review of Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 298). And David Beach is disposed to negate Lerdahl and Jackendoff ’s entire theory simply because it makes the master’s choice of Kopfton impossible! (Beach, ‘The Current State of Schenkerian Research’, 294.) In a clever albeit non-Schenkerian compromise, David Neumeyer tinkers with the Theme’s Ursatz to turn it into a three-part rather than a two-part counterpoint. The Urlinie can then begin on both 3ˆ and 5ˆ (Neumeyer, ‘The Three-Part Ursatz’, 26–7). If the choice of 5ˆ is defensible in Schenkerian terms, this is one more piece of evidence for the distance between Schenkerian analysis and attempts to give an account of the palpabilities of the surface. More about this issue below. 18 ‘[Schenker’s] criteria for pitch stability alone are “free floating” – they can connect up events anywhere’ (Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 119). 19 Cone, Musical Form and Musical Performance, 25. 20 Grosvenor Cooper and Leonard B. Meyer, The Rhythmic Structure of Music (Chicago: University of Chicago Press, 1960). Eugene Narmour makes the distinction between ‘data-based’ and ‘concept-based’ analyses in his discussion of the Theme (‘Some Major Theoretical Problems’, 181). Meyer’s ‘data’ are the rhythmic groupings of the surface. 21 Kofi Agawu articulated this problem in his 1989 discussion of Schenkerian notation, where he distinguished three types of reductions: ‘pure’ graphs (showing Schenkerian middlegrounds and backgrounds); mixed graphs (usually only foregrounds), that contain a combination of recognizable music notation and Schenkerian signs; and rhythmic reductions, to which he accords an ‘ambivalent status’ (Agawu, ‘Schenkerian Notation in Theory and Practice’, 286). 22 Weber’s schema is hypothetical, a representation of his theory rather than an analysis of the Theme, while Riemann’s iambic schema is hypothesized by Morgan extrapolating from Riemann’s theories and his remarks on the Theme’s rhythmic patterns. 23 Spitzer, Metaphor and Musical Thought, 247–8. 24 Scott Burnham, Beethoven Hero (Princeton: Princeton University Press, 1995), 128.

277

278

wye j. allanbrook

25 In Metaphor and Music Thought, 249–60, Spitzer goes on to argue that this switch from the metrical to the terminal accent makes the second key area of a two-reprise form different in quality from the first, in a ‘song/dance dialectic’ (246). Here I part company with him. This distinction, crucial to Spitzer’s theory of musical metaphor (‘Metaphor breaks out in the lyrical second group’ (253)), is not far distant from the old-fashioned premise of an opposition in character between first and second theme, now sensibly rejected as simplistic. Given the prodigious variations in both shape and materials in Mozart’s and Haydn’s expositions, Spitzer’s generalization seems rather a stretch. 26 The broader question of whether a paradigm need be simple floats in the background of this essay. Kofi Agawu raised it in a private communication, asking if a paradigm could be an example that, ‘marked by something unusual’, offers ‘resistance to ready explanation’. I would be inclined to answer in the negative. Too much of the unusual must transform any chosen example into a singleton – the very opposite of a paradigm, which, one assumes, ought to be all-embracing. But the question does not need to be answered here, because most of the writers I have scanned assume the transparency of their chosen object. 27 Cone, Musical Form and Musical Performance, 26. 28 Ibid., 29. By ‘first phrase’ Cone means the half-cadence in bar 4. 29 Riemann, Vademecum der Phrasierung, 61. The re-barring supports Schenker’s choice of E as Kopfton. 30 Morgan, ‘The Theory and Analysis of Tonal Rhythm’, 446, 447. In a late eighteenth-century composer’s choice of metre, expressive intent generally trumped such practical concerns. See my Rhythmic Gesture in Mozart: Le Nozze di Figaro and Don Giovanni (Chicago: University of Chicago Press, 1983), 1–30. 31 Morgan, ‘The Theory and Analysis of Tonal Rhythm’, 450. 32 Meyer, Explaining Music, 31. In Emotion and Meaning in Music, 109, Meyer makes the important point that the phrasing within the barline in the Theme is supported by ‘the absence of any prior anacrustic organization’. In other words, to treat successive beat-6 eighth notes as upbeats, a composition in 6/8 metre would normally start with an eighth-note upbeat – a point one regularly makes with beginning theory students. Lerdahl and Jackendoff make the same point (A Generative Theory of Tonal Music, 63), to which I will return below. 33 Morgan, ‘The Theory and Analysis of Tonal Rhythm’, 445. 34 Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 123. 35 The difference, they argue, is that this Ursatz is ‘an effect, not a cause, of tonal principles’ (Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 139–40). Nonetheless, it is neither an effect nor a cause of the rhythmic practices that are operative here. 36 Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 121–3. For their trochee time-span analysis, see Example 5.11 on page 120. 37 Narmour (‘Some Major Theoretical Problems’, 192) astutely points out that ‘[Meyer’s] method is epistemologically similar to Schenker’s in that there is a

Mozart’s K331, first movement

strong a priori bias toward certain foreordained patterns on higher levels – namely, end-accented groups (iambs, dactyls)’. In the quotation ‘anapaests’ should obviously be substituted for ‘dactyls’. 38 Because Meyer was certain that the ambiguities of the 6/8 resolve over the barline from bar 7 to bar 8, he warned of the ‘great technical control and interpretative finesse’ that a good performance of the early measures of the Theme requires in order not to disguise ‘a potential which was latent, but not actualized, in the preceding weak beats’ (Explaining Music, 42; emphasis the author’s). It should be noted, however, that Meyer’s study of the Theme serves a different purpose than those of his colleagues; he is not properly a member of this group. He uses the Theme not to buttress a general theory of rhythmic structure, but to make an epistemological point about the possibility of interpretative certainty in performance. To do so he compares two editions of the Theme (Broder and Peters), in which the latter phrases over the barline and the former does not, and he argues for music-theoretical reasons that the former is the only possible performance of the phrase. Meyer is not a Schenkerian; although he includes simple middleground reductions of the Theme in this discussion (37–8), his eye is on the surface, not on the higher-level rhythmic phenomena that concern the others. In a 1988 keynote address to the SMT, he memorably metaphorized his preference for the upper strata: ‘I am not a denizen of obscure, abstract depths – a diver after cosmic conceptions and unconfirmable hypotheses. I am content to snorkel along the surface, peering down just a bit to be bewitched by the pleasing patterns of luminous fish and the quiet swaying of colourful coral.’ But his analysis of the Theme in the earlier Explaining Music did not call on the contextualizing evidence – the luminous fish and swaying coral – which, I will argue below, would have settled the issue unequivocally. (Meyer’s keynote address was published, slightly altered, as ‘A Pride of Prejudices; Or, Delight in Diversity’, Music Theory Spectrum 13 (1991), 241–51, and reprinted in Leonard B. Meyer, The Spheres of Music: A Gathering of Essays (Chicago: University of Chicago Press, 2000), 262–78; the quoted passage occurs on page 262.) 39 See Zbikowski, Conceptualizing Music, 306–24, for an interesting discussion of the three schemata (Meyer, Morgan and Cone), in which he distributes them between two poles – ‘atomistic’ and ‘chain-of-being’ hierarchies. Represented by Meyer and Morgan respectively, the former he describes as an ordering that begins with surface detail, while the latter posits a mysterious Ursatz-like force as a starting place (310). Zbikowski praises Cone’s account as being a ‘conceptual blend’ of the two, beholden to the constraints of neither. I would argue that in giving final priority to end-oriented groupings in bar 8 all three have made the same error, while Meyer’s account is ultimately more faithful to the surface than either Morgan’s or Cone’s. 40 Cone, Musical Form and Musical Performance, 27; Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 123. ‘Belatedly’ because of the Theme’s

279

280

wye j. allanbrook

41

42

43 44

45

noncongruence of rhythm and pitch. The rubber band is overstretched because bars 1–3 are harmonically static. Cone, Musical Form and Musical Performance, 29. The rhythmists used editions that incorrectly included dynamic markings only on the authentic cadence, as can be seen in Examples 9.3A and 9.6. This probably helped to skew their understanding of the purpose of the markings. Only one page of the k331 autograph exists – of the end of the last movement. The source used for recent Urtext editions of the sonata is the Artaria print of k331–333, published in Vienna in 1784. Reviewing the Henle edition of 1966, edited by Walter Lampe, which lacks the sf over bar 4, beat 3, Alan Tyson reported that over time some Artaria plates became cracked and were replaced by ‘rather carelessly engraved ones’, containing many errors, ‘principally in the omission of dynamics’, which were used as sources for some editions. The sf, he attested, is found in the original plates, and is certainly Mozart’s marking (Tyson, Review of Mozart’s Piano Sonatas, Musical Times 107 (1966), 888). The most recent Henle edition (ed. Ernst Herttrich, 1977) has returned the sf p marking to bar 4, beats 3–4, and the Neue Mozart-Ausgabe follows suit. Example 9.1 is based upon the NMA. As far as I know, the only musicologists to mention this are Head (see notes 9 and 44) and Agawu (‘Il ritmo’, trans. Fulvia De Colle, in Enciclopedia della Musica II: Il sapere musicale, ed. Jean-Jacques Nattiez (Torino: Einaudi, 2002), 48). See Allanbrook, Rhythmic Gesture in Mozart, 40–5, for a discussion of the types of 6/8 pastorale – siciliano, eponymous pastoral, and gigue. In his account Matthew Head over-rusticizes the siciliano topos so that it can enter into a political dialectic of the oppressed with the brutally percussive Coda of the ‘Rondo alla turca’. As a manifestation of Volkst¨umlichkeit, he argues, the siciliano partakes of the same Otherness as the Turkish finale (Head, Orientalism, Masquerade and Mozart’s Turkish Music, 126). But this is to ignore the long tradition of the siciliano as an aria type and instrumental movement, frequent already in the operas of Alessandro Scarlatti, and a commonplace for Bach and Handel. By the 1780s the siciliano had lost any dewy country freshness it might once have possessed. No longer representing the radical rustic, it had been appropriated to express the city dweller’s nostalgic gaze at Arcadia. In 1752 Quantz was already distancing it as ‘an imitation of a Sicilian shepherd’s dance’ (Johann Joachim Quantz, On Playing the Flute, trans. Edward R. Reilly (New York: Schirmer Books, 1966), 168; emphasis mine), and Koch repeats the sentiment (Heinrich Christoph Koch, Musikalisches Lexikon (Frankfurt-amMain: August Hermann der J¨ungere, 1802; reprinted Hildesheim: Olms, 1964), cols. 1382–3). Only one eighteenth-century choreography is extant, reinforcing this notion of the siciliano as part of a courtly pastoral imaginary (Meredith Ellis Little, ‘Siciliana’, in The New Grove Dictionary of Music and Musicians, 2nd edn, ed. Stanley Sadie and John Tyrell (London: Macmillan, 2002), vol. 23, 350–2). Malcolm Bilson enjoins strict attention to Mozart’s careful use of slurs in his ‘Execution and Expression in the Sonata in E Flat, K. 282’, Early Music 20 (1992),

Mozart’s K331, first movement

46 47

48

49

50

51 52 53

54

237–43. His discussion of the attack and release of the slurs in each measure of the opening phrase of k332/i (243) is particularly apposite here. Meyer uses as his example a Peters edition edited by Louis K¨ohler and Adolf Ruthardt. Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 120. Joel Lester dismisses the chord as ‘clearly a foreground manifestation’, and F♯ as a neighbour to the bass’s G♯ (Lester, ‘Articulation of Tonal Structures’, 76). With one exception, rather a confused one. Surprisingly, again citing ‘musical intuition’, Lerdahl and Jackendoff leave the ‘vi7 ’ in their middleground reduction, arguing that the F♯ implies the E in bar 4 to produce a descending bass line A–G♯–F♯–E (Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 164–5). But this reduction is pretty much unsupportable for either camp (for example, Schenkerian David Harvey sensibly criticizes it (Harvey, Review of Lerdahl and Jackendoff, A Generative Theory of Tonal Music, 297)), and contrary to any musical intuition of the Theme that I can imagine. The F♯ is the end of the descending line, and it is not until the line pulls back from the ‘distant’ F♯ to the tonic at the beginning of bar 4 that extroversive semiosis yields to the snapped-off introversive sign of the ii7 –i 6/4–v half cadence. Laudably sensing the importance of the quasi-chord, Lerdahl and Jackendoff misconstrue its function, fabricating a skeleton that does justice to neither the historical surface nor the hypothetical depths. See, for example, the first four measures of the finale of Haydn’s String Quartet Op. 33 No. 5, where the harmony moves in stately fashion from i (bar 1) to iv–v7 (bar 2) to i–ii6 (bar 3) to i 6/4–v (bar 4). Heinrich Christoph Koch, Versuch einer Anleitung zur Composition, vol. 3 (Leipzig: Adam Friedrich B¨ohme, 1793; reprinted Hildesheim: Georg Olms, 1969), 41–2; Eng. trans. (Introductory Essay on Composition) by Nancy Kovaleff Baker (New Haven: Yale University Press, 1983), 79, Example 227. In this example Koch shows a gavotte by Christian Gotthelf Scheinpflug. See Drabkin, Pasticci and Pozzi, Analisi schenkeriana, Figure 2; Lester, ‘Articulation of Tonal Structures’, 74. V. Kofi Agawu, Playing with Signs: A Semiotic Interpretation of Classic Music (Princeton: Princeton University Press, 1991), 45. Ferdinand de Saussure, Course in General Linguistics, ed. Charles Bally and Albert Sechehaye, trans. Wade Baskin (New York: McGraw-Hill, 1959), 106; emphasis mine. See Kofi Agawu’s formulation of the same point in Playing with Signs, 45: ‘Let us not forget that the substantial domain of topics, like that of language, remains open, allowing for the possibility of discovering more topics . . . References to an area of “neutral” topical activity indicate not necessarily the absence of topic, but, rather, the absence of an appropriate label within the restricted domain of our topical universe. All Classic music, then, is conceptually laden with topical signification – although the extent to which that signification is made perceptually evident varies from context to context.’

281

282

wye j. allanbrook

55 Meyer, ‘A Pride of Prejudices; Or, Delight in Diversity’, in The Spheres of Music, 277. 56 Leonard Meyer and Kofi Agawu are theorists by claim, but their domain is all of music. Scott Burnham, cited earlier in this article, is an excellent example of a theorist whose work continually spans the divide. And theorists in this volume also consistently incorporate historical evidence into their investigations. But this volume constitutes an act of selection: it makes sense that the work of such theorists would be gathered here. 57 I would like to thank Pierpaolo Polzonetti for adding Analisi schenkeriana to my list of texts, and Mary Hunter for directing me to Alan Tyson’s comments on the editions of k331.

10

Dance topoi, sonic analogues and musical grammar: communicating with music in the eighteenth century l aw r e n c e m . z b i ko ws k i

For Etienne Bonnot de Condillac, music was a solution to a problem. The problem arose in connection with the radical empiricism he advocated in his 1746 Essai sur l’origine des connaissances humaines, which sought to derive all knowledge from sensation. Stated simply, the problem was one of explaining how such sensory information could be organized into knowledge. Music – and in particular the theoretical account of musical organization proposed by Jean-Philippe Rameau – offered a solution, for it provided a link to the passions that motivated the formation of knowledge as well as a rational structure through which knowledge could be organized. As he surveyed the basic capacities upon which knowledge was grounded Condillac noted that humans shared most of them with brute animals. What humans had that animals did not was a memory guided and facilitated by what Condillac called instituted signs. The two other sorts of signs Condillac recognized, accidental signs and natural signs, were shared with animals: accidental signs are simply the perceptual traces of the objects we behold; natural signs are the involuntary cries associated with passions such as joy or grief. Instituted signs, by contrast, are wholly volitional: they are signs that we choose to represent some aspect of knowledge. Once knowledge is represented it can be remembered and made available for contemplation – it becomes truly knowledge, rather than transitory sensory information. Condillac proposed that the establishment and organization of instituted signs – a process that led, ultimately, to language – proceeded gradually and over many human generations. At the beginning of this process was the natural sign, which was not limited to cries alone but included various gestures. The immediacy of these gestures stimulated an empathy between pre-linguistic individuals that led to a form of communication: For example, he who suffered, by being deprived of an object which his wants had rendered necessary to him, did not confine himself to cries or sounds only; he used some endeavours to obtain it, he moved his head, his arms, and every part of his body. The other [person] struck with this sight, fixed his eye on the same object, and perceiving some inward emotions which he was not yet able to account for, he suffered in seeing his companion suffer.1

284

l aw r e n c e m . z b i ko ws k i

With repetition these gestures and actions became familiar and were remembered, giving rise to what Condillac called a langage d’action. Directly related to the passions, it had great power, but in its unrefined form there was a good deal of variability among its signs and a consequent ambiguity about what they meant. Over time the gestures of the langage d’action were refined into dance, the inflections of its various sounds were refined into music, and the images and situations it called forth became the stuff of poetry. Key to the development of music and poetry (which Condillac believed were unified at this originary moment) was the discovery by our ancestors of the natural order of musical sounds. This order was none other than the one revealed by Rameau’s theory of harmonic generation, and it led the ancients to organize the widely varying inflections of their proto-language in accordance with the ut, mi, sol and ut with which every sonorous body resounded. Once this was accomplished, the true expressive and rational potential of language could be realized.2 Connections between music and language of the sort entertained by Condillac have been remarked upon since antiquity. What was new in the eighteenth century was the exploration of this connection within the context of inquiries into what it meant to be human. As Downing Thomas has noted, music served as an anthropological ‘missing link’, allowing Enlightenment writers such as Condillac ‘to trace semiosis to its origin, to pinpoint the semiotic moment which separates culture from nature, and human beings from animals’.3 There was also during this period a growing appreciation of the expressive potential of purely instrumental music, which theories like Rameau’s helped to explain. As a consequence of these developments, music and language came to be seen as parts of a unified system of communication, striving toward the common goal of expressing uniquely human knowledge. This view is one that has survived, largely intact, to the present day: music is commonly regarded as a communicative medium that functions as a kind of language, a language that, on the one hand, is well adapted for the expression of emotions and, on the other hand, has a natural rational basis which is captured and studied by theories of music. I would like to complicate this view substantially by proposing that language and music have different functions in human culture, that the function of music is reflected in its grammar, and that composers of the eighteenth century understood this. Where Condillac saw language as the successor to music, I suggest that the two are substantially independent, developed by humans for distinctly different communicative goals. More broadly, musical communication – whether in the eighteenth century or the twenty-first – is not like linguistic communication, but must be understood on its own terms.

Dance topoi, sonic analogues and musical grammar

My argument has two basic parts. In the first, I develop a theoretical approach to musical grammar based on research in cognitive linguistics and on nineteenth-century accounts of the proper way to compose dance music. The latter reflects the influence of Wye Allanbrook’s pathbreaking work on rhythmic gesture and dance topoi – as does she, I propose that music for dance is crucial for understanding music of the late eighteenth century, although my approach is ultimately somewhat more prosaic than Allanbrook’s. The second part of the argument involves an analysis of the Finale from Haydn’s Op. 76 No. 4 quartet, which builds off Allanbrook’s observation that the opening of this movement has many of the characteristics of a bourr´ee. My analysis suggests that not only was Haydn aware of the bourr´ee topic, but that he also drew on the musical grammar basic to the dance as a resource for shaping the tonal rhetoric of his Finale.

Musical grammar and composing for the dance Function and form in language and music The notion that music and language are different but related aspects of a single unified system of communication flows naturally from their shared features. Both are unique to the human species, both unfold over time, both have syntactic properties, and both make use of sound. There are also, however, a number of differences between music and language: musical meaning is on the whole much less precise than linguistic meaning; music often involves simultaneous events, where language does not; and there is more of a sense of play in ordinary music than there is in ordinary language.4 I propose that these differences reflect the different functions of music and language. Drawing on the work of the developmental psychologist Michael Tomasello, I take the position that the primary function of language within human culture is to direct the attention of another person to objects or concepts within a shared referential frame.5 The primary function of music, by contrast, is to represent through patterned sound various dynamic processes that are common in human experience. Chief among these dynamic processes are those associated with the emotions (which, following recent work by Antonio Damasio, can be construed as sequences of physiological and psychological events that subtend feelings6 ) and the movements of bodies – including our own – through space. The difference in function between these two media is matched by a difference in the forms through which the functions are realized. In the

285

286

l aw r e n c e m . z b i ko ws k i

case of language, these basic functions are realized through parts of speech. Edward Sapir, reflecting on the basic elements required for languages, wrote that in language there must be something to talk about and something must be said about this subject of discourse once it is selected. The distinction is of such fundamental importance that the vast majority of languages have emphasized it by creating some sort of formal barrier between the two terms of the proposition. The subject of discourse is a noun. As the most common subject of discourse is either a person or a thing, the noun clusters about concrete concepts of that order. As the thing predicated of a subject is generally an activity in the widest sense of the word, a passage from one moment of existence to another, the form which has been set aside for the business of predicating, in other words, the verb, clusters about concepts of activity. No language wholly fails to distinguish noun and verb, though in particular cases the nature of the distinction may be an elusive one.7

The basic parts of speech, then, are symbolic units that correlate with functions such as those represented by nouns and verbs, as well as the many other parts of speech recognized by grammarians. In the case of music, the basic unit of grammar is what I call a sonic analogue, which represents through patterned sound the central features of some dynamic process.8 One of the places sonic analogues are most evident is in music for dance.

Analogizing dance Adolf Bernhard Marx, in his instructions to student composers on how to compose a waltz, provided insight into how music can analogize dynamic processes. Marx begins by describing the basic gestures of the dance: ‘The waltz has two movements: first each pair of dancers turns itself in a circle around its own centre; second the pair progresses with these continuous turns in a greater circumference until it reaches its starting place and the circle is closed. Each little circle is performed in two-times-three steps and is, as it were, the motive of the dance.’9 A diagram of the basic pattern described by Marx is given by Thomas Wilson in his 1816 treatise on the waltz, shown in Figure 10.1: here the small circles summarize the ‘two-times-three steps’ gesture noted by Marx, and the large circle shows the path the dancers take around the dance floor.10 In truth, and as Sevin Yaraman has noted, the ‘circles’ described by Marx and diagrammed by Wilson are spirals, which connect to form the large circle shown in Figure 10.2.11 These circular motions are perhaps the most important feature of the dance, and one that Marx believes should be clearly supported by the music:

Dance topoi, sonic analogues and musical grammar

Figure 10.1 Diagram for waltzing, reproduced from Wilson, A Description of the Correct Method of Waltzing, facing page 46

‘At the very least the waltz must bring into prominence this basic motive of movement. Each bar, or, better, each phrase of two bars, must answer to the dance motive marking the first step firmly, and also the swinging turn of the dance. Where the bars do not point it out they must still favour it, by a melody which spiritedly turns away from the first note.’12 Having laid out these desiderata, Marx then provides two examples. The first, shown in Example 10.1, is a waltz from the first act of Carl Maria von Weber’s Der Freisch¨utz. The second, shown in Example 10.2, is his own confection; he remarks that the example demonstrates the three basic steps in a simpler but less swinging way. Marx then observes: ‘we see in the above pieces auxiliary tones placed before the chord tones in the melody in order to set the first step in relief; every other melodic, harmonic and rhythmic sharpening – the assistance of a forzato, an accented first note in the bass [as shown in Example

287

288

l aw r e n c e m . z b i ko ws k i

Figure 10.2 Diagram showing the path of the dancers in waltzing, reproduced from Yaraman, Revolving Embrace, 19

Example 10.1 Waltz from C. M. von Weber’s Der Freisch¨utz (Marx, Die Lehre von der musikalischen Komposition, vol. 2, 56)

Example 10.2 A. B. Marx’s fabricated waltz (Die Lehre von der musikalischen Komposition, vol. 2, 56)

10.3], and so forth – serves the same purpose’.13 For Marx, then, the function of waltz music is to support, as thoroughly as possible, the movements of the dancers. The grammar of the waltz – that is, the way its materials are organized – has to realize this function or, implicitly, there is no waltz.14 From the perspective of the early nineteenth century the music for a waltz had to correlate with the steps and movements of the dance: it provided the dancer with a sonic image of the various gestures required for the dance as well as a series of sonic events onto which she could map specific bodily motions. The music thus offered a sonic analogue for the dynamic process of the dance. More specifically, regular repetitions of the two-bar waltz motive

Dance topoi, sonic analogues and musical grammar

Example 10.3 A. B. Marx’s basic accompaniment pattern for the waltz (Die Lehre von der musikalischen Komposition, vol. 2, 56)

are the basis for the structure of the dance as a whole (typically thirty-two bars long), which is divided at regular intervals by cadential gestures. Tonal syntax – which is often thought of as the syntax of music – provides a way to reinforce this grammar, but the grammar of the waltz is first and foremost about the dance, rather than about tonal organization.15

Musical construction grammar The approach to grammar that I have adopted here, in which the form of a grammatical element is in an intimate relationship with its function, conforms with what cognitive linguists call construction grammar. In the case of language, constructions are defined as ‘stored pairings of form and function, including morphemes, words, idioms, partially lexically filled and fully general linguistic patterns’.16 In the case of music, constructions are in each case sonic analogues for dynamic processes. The most basic of these are immediate and of limited extent. Marx’s waltz motive is one example; other examples are the basic-level musical categories I have discussed elsewhere.17 As with linguistic constructions, musical constructions are stored pairings of form and function. Basic constructions are organized into larger structures through syntactic processes. Repetition – which can be defined as successive occurrences of equivalent constructions, and to which I referred in my sketch of the grammar of the waltz – is one such process. Repetitions of the waltz motive create the larger structures associated with the dancers’ movement through the large circle diagrammed by Wilson. Another process evident in Marx’s examples is cadence. In Weber’s waltz cadence is effected by an arrival on 1ˆ in the melody on the first beat of bar 8, which coincides with a momentary cessation of the running quavers and the oom-pah-pah accompaniment (which is not included in Marx’s citation). Cadence in the example formulated by Marx is much less marked: while there is an arrival on 1ˆ in bar 4 the accompaniment pattern continues uninterrupted, suggesting that further bars of music are to follow.

289

290

l aw r e n c e m . z b i ko ws k i

In keeping with the construction grammar approach, in which all grammatical elements are pairings of form and function, syntactic processes are themselves sonic analogues for dynamic processes. Repetition is, in the first instance, an analogue for cyclic processes. Cyclic processes can suggest nested hierarchical structures but can also suggest stasis or stagnation – that is, the negation of process. Cadence is an analogue for the process of arrival, although there are elements within cadence that are akin to interruption, as is evident in the last bar of Weber’s waltz. Either repetition or cadence can be applied recursively; although the former is quite apparent in Marx’s examples (with repetitions of the waltz motive practically defining the dance), the latter is less obvious – indeed, even in the complete version of Weber’s waltz the sense that there is a final cadence is thwarted when the music for the waltz melts into the continuation of the scene. In general, it is through the recursive application of syntactic processes, and the application of these processes at different levels of musical organization, that ever larger and more complex musical structures are created. (Further instances of syntactic processes will be discussed below in the analysis of Haydn’s Finale.) One consequence of this approach to musical grammar is a changed view of musical communication. ‘Communication’ is typically understood as having something to do with imparting knowledge to or sharing information with someone else. The framework of logic behind this intuition is what Michael Reddy called the conduit metaphor, according to which language functions like a conduit to transfer thoughts from one person to another.18 While such a perspective works well for language, with its focus on objects or concepts within a shared referential frame, it works less well for music. Our first engagement with music is on the level of the dynamic processes analogized through music’s patterned sound. Engagement of this sort, akin to empathy or sympathy, is not really captured by the idea of receiving something transferred through a conduit. Musical communication – if we even want to apply the term to the sort of engagement with dynamic processes with which I am concerned here – will in consequence be rather different from communication through language. Two important qualifications of this view of musical communication are in order. First, the starting point for my characterization of musical grammar was what I took to be the primary function of music, which is to represent through patterned sound various dynamic processes common in human experience. The perspective is similar to that of Tomasello, who proposed that the primary function of language within human culture is to direct the attention of another person to objects or concepts within a shared referential frame. But these are not the only roles of music or language within human

Dance topoi, sonic analogues and musical grammar

culture. In the case of music, either musical or non-musical events can prompt us to step back from our engagement with the dynamic processes analogized through sound and become aware of the means through which this engagement is accomplished. In such cases, reflection on musical objects replaces engagement with musical processes and something closer to the sort of communication we associate with language commences. The second qualification concerns the relative closeness of the relationship between music and dance. Although both Marx and Wilson, in their discussion of the waltz, emphasize the tightness of this connection, in most dances the relationship is much looser: gestures or steps of the dance may have no correlate in the music, and aspects of the music will not be represented in the dance.19 This variability of correspondence suggests that the combination of music and dance is best thought of as what Nicholas Cook calls an instance of multimedia: the gestures of the dance are one medium, the patterned sound of music another. As Cook has observed, correlations between the constituent media of an instance of multimedia can be many or few. Where there are many correlations the media are conformant with one another; if there is absolute conformance the boundaries between media may disappear altogether. Where there are few correlations the media are in a state of contest; if there is absolute contest any sense that the media participate in a common communicative function may be lost.20 Various dances could be situated along a continuum, the ends of which are marked by the two extremes of conformance and contest. Toward one end would be Marx’s waltz and its brethren, distinguished by a tight correlation between dance and music; toward the other would be cases where dance and music are so dissimilar that each threatens to go its own way. For my part, I will assume that most social dances fall somewhere in the middle: the connection between the gestures of dance and the materials of music is not quite as tight as that described by Marx, but still close enough that the music serves as a sonic analogy for the movements essential to the dance.

Musical communication in the eighteenth century Marx’s teaching of how to write waltzes and other dances fell early in his overall course of instruction as part of a thorough treatment of what he called periodic forms. The assumption was that dances exemplified the regular building blocks of musical expression, and that these building blocks could be exploited in more advanced forms. A similar perspective, informed by the composition treatises of Joseph Riepel and Heinrich Christoph Koch, led Wolfgang Budday to propose that the formal principles evident in dance

291

292

l aw r e n c e m . z b i ko ws k i

forms like the minuet are the basis of Viennese classical style.21 Musical composition in the latter half of the eighteenth century and first half of the nineteenth century involved more than simply the expansion or re-use of dance forms, but given the ubiquity of these forms in the social milieu inhabited by composers as well as the expectation that a well-trained Kapellmeister would be familiar with all the common dances of the court, it seems safe to say that a knowledge of dance forms was part of the basic equipment of a composer during this period.22 This knowledge included not only the rhythmic characteristics typical of any given dance but also the actual steps of the dance, since it was upon these that the rhythmic organization was predicated. Even when the dances were no longer danced the choreography suggested by their rhythmic patterns provided an expressive resource for the composer: as Allanbrook has shown, dance forms could serve as topics around which musical discourse is organized. In her exploration of Mozart’s musical language Allanbrook was primarily interested in the way various topics derived from dance forms – topics that instantiate the rhythmic gestures of dance – are combined, juxtaposed and contrasted for expressive ends. My focus is on a more basic, and perhaps more mechanical, level: I propose that composers of the late eighteenth century could not only use dance forms for their topical associations but could also modify and subvert the grammar of dance to create new expressive possibilities.23 Musical communication thus began with sonic analogies for familiar processes (such as the steps of a dance) but was developed through the introduction of new patterns or the modification of existing ones. As I shall try to argue in the next section, it is this sort of development of compositional strategy that is evident in the Finale of Haydn’s String Quartet Op. 76 No. 4. Although Haydn begins the movement with an evocation of a bourr´ee, before half a minute is up he effects a reinterpretation of the gestures of the dance, transforming elegant choreography into the bumptious antics and pratfalls of comic theatre.

Joseph Haydn, String Quartet Op. 76 No. 4, Finale The bourr´ee of the French court and eighteenth-century instrumental music In a survey of French court dances of the seventeenth and eighteenth centuries, Meredith Little and Natalie Jenne begin with the bourr´ee, reckoning it to be among the least complex of French baroque dances. The bourr´ee

Dance topoi, sonic analogues and musical grammar

is a moderately fast dance in duple time, the affect of which is invariably described as easygoing and gay. Little and Jenne summarize the basic rhythmic structure of the bourr´ee as follows: ‘In the bourr´ee dance rhythm the rhythmic–harmonic phrase is eight beats in length (four measures), preceded by an upbeat, usually two eighth notes or a quarter note. Beat 7 and the first half of 8 [which make up the fourth bar] constitute the primary repose, or thesis; beat 3 and the first half of 4 provide a preliminary resting point, or secondary thesis.’24 One of the characteristic rhythmic figures of the bourr´ee is a quick anapaestic rhythm of two quavers moving onto a strong beat (the same figure that Little and Jenne observe is often used for the upbeat to the dance). Another is a syncopated rhythm – crotchet/minim/crotchet – that typically appears coincident with the preliminary resting point associated with beats 3 and 4. The rhythmic instability of this syncopation tends to highlight the contingency of this resting point, since the rhythmic arrival on the second main beat of the bar is not correlated with a similar arrival in the music. Pierre Rameau, in a dance tutor published in 1725, described the basic step of the bourr´ee as composed of two movements. The first is a demi-coup´e (a pli´e – gently bending the knees – followed by an ´elev´e – gently stretching the knees to rise) succeeded by a pas march´e sur la demi-pointe (a simple step, but without lowering the heel); the second is a demi-jet´e (a short hop).25 Rameau notes that current fashion has replaced the demi-jet´e with another plain step; this combination of a demi-coup´e and two plain steps is generally known as the pas de bourr´ee. In either case the basic step divides the musical bar into four parts: parts 1 and 2 are taken up by the pli´e and ´elev´e of the demi-coup´e, and parts 3 and 4 with the two plain steps (or step and demijet´e). The fall and rise of the demi-coup´e suggest movement through parts 1 and 2; the second of the two plain steps that complete the dance figure suggests that part 4 marks a point of arrival. As conventionally rendered in musical notation, part 1 occupies an anacrusis, and part 2 begins with the downbeat of the bar; parts 2 and 4 thus correspond with the beginnings of minim durations. The dancers are consequently in motion through the first minim (the ‘strong beat’ of the bar) and arrive at the beginning of the second minim (the ‘weak beat’). A richer sense of the relationship between the music and steps of the bourr´ee is provided by Little’s analysis of a choreography published by Louis P´ecour in 1700. Little uses five symbols for the steps specified by the Feuillet notation P´ecour employed, which are shown in Figure 10.3: included are the pli´e, the ´elev´e, the jett´e, the plain step and the gliss´e; this last is a gradual sliding of the foot along the path of the step. In Little’s transcription, which

293

294

l aw r e n c e m . z b i ko ws k i

Figure 10.3 Meredith Little’s notation for dance steps (‘The Contribution of Dance Steps’, 112)

Example 10.4 Meredith Little’s analysis of P´ecour’s choreography for a bourr´ee (‘The Contribution of Dance Steps’, 114–15)

is given in Example 10.4 together with the music supplied by P´ecour, the brackets underneath these symbols show which steps were joined together in the original notation – that is, how the dancer was to group the steps – and the letters R and L above the symbols show which foot is being used.26 As can be seen, both the music and the choreography are in two parts, corresponding to bars 1–8 and bars 9–16 of the score. In the music, the first and second part are very similar; what changes there are have to do with the process of ending on the dominant in bar 8 and that of ending on the tonic in bar 16. The choreography, however, changes not at all: the steps for bars 1–8 are the same as those for bars 9–16. Unadorned pas de bourr´ee can be seen in bars 2, 4 and 7 (and correspondingly in bars 10, 12 and 15). Interspersed between these are somewhat more complicated steps that involve either jet´es or a

Dance topoi, sonic analogues and musical grammar

Example 10.5 J. S. Bach, Bourr´ee from Suite for Lute in E minor, bwv996, bars 1–8

decrease in the number of steps (something that can be seen in bars 5 and 6). But in every case the dance figure begins with a pli´e on the preceding crotchet upbeat which then entails a rise onto the downbeat, a movement that contributes much to the buoyancy associated with the dance. Another thing that contributes to the gentle animation of the dance are syncopations of the sort that occur in bars 2 and 10. Here the music and choreography diverge, with the dancers marking the second beat with an arrival that is withheld by the music. A different sort of divergence occurs in bars 4 and 12, when the melody makes its arrival on the first beat and the dancers continue through to the second beat. Both sorts of divergence bear witness to contest between the medium of music and the medium of dance. That is, the non-alignment of music and dance make P´ecour’s bourr´ee a true instance of multimedia. This contest plays out, however, against a backdrop of conformance: the music is organized in a clear pattern of two bars plus two bars plus four bars, as is the dance; the four-bar unit is itself distinguished by a two-bar sequence (in bars 5 and 6) which accompanies the deceleration of steps noted earlier; and the termination of each eight-bar section is clearly marked by simultaneous arrivals by both the music and the dancers (with the arrival of the dancers sustained by the gliss´e). In sum, the overall effect of the dance is accomplished through the way each of its constituent media is set out: P´ecour’s bourr´ee is found between the choreography and the music. A useful contrast to P´ecour’s bourr´ee, as well as a bridge to Haydn’s Finale, is provided by a bourr´ee for the lute written by Johann Sebastian Bach, the first eight bars of which are given in Example 10.5. The date of the work is uncertain, but it is most likely from after 1712. This portion of the dance has an AA′ form similar to that of P´ecour’s bourr´ee, but the sections are four rather than eight bars long. One of the most prominent features of the dance is Bach’s nearly relentless use of the anapaestic figure of two quavers moving to a strong beat. This antic activity is articulated by harmonic arrivals at

295

296

l aw r e n c e m . z b i ko ws k i

regular two-bar intervals: on E minor in bars 2, 4 and 6, and on G major in bar 8. Indeed, Bach’s bourr´ee has a far more regular rhythmic surface than does that of P´ecour. I would like to suggest that, pace Marx, the dance owes its rhythmic regularity to the fact that it was not intended for dancing. It was instead meant to evoke, through a single medium, the multimedia of music and dance. To accomplish this, Bach restricted his materials to those that evoke the basic gesture of the dance: a lift onto the first beat of the bar, followed by a continuation that moves to and is concluded on the third crotchet of the bar. The fluidity accomplished in P´ecour’s dance by the non-alignment of music and choreography is accomplished by Bach through harmonic and rhythmic means: each successive two-bar unit enacts a motion from dominant to tonic, and the relatively strong arrival on E minor in bar 4 is undercut by the running quavers (first in the melody, then in the bass) that push the music ahead to the reprise of the A section initiated by bar 5. Although the Finale of Haydn’s Op. 76 No. 4 quartet dates from the end of the century (it was first published in 1799), its opening eight bars, which are given along with the rest of the first main section of the movement in Example 10.6, bear a striking number of similarities to Bach’s bourr´ee. Within the framework of an AA′ form there is a predominance of the anapaestic figure of two quavers moving to a strong beat, although here the figure is distributed among the members of the quartet. Haydn, with more expansive instrumental resources at his disposal, summons a sense of forward motion by somewhat more subtle means than does Bach: in bar 1 both Violin II and the Viola move in quavers onto the third beat; in bar 2 only Violin II moves onto the third beat in quavers but Violin I does its own version of a jet´e with its turn around f1 ; in bar 3 the third crotchet is approached through a wedge of contrary motion in the upper and lower strings that pulls the music all the way through to the arrival on the dominant in bar 4; and in both bar 3 and bar 7 motion toward a strong beat is suggested by the accentual destabilization provided by the forzato marks on the preceding weak beat. Although not named a bourr´ee, the opening bars of Haydn’s Finale do most of the things expected of an instrumental bourr´ee in the eighteenth century, not the least of which is to project in duple time the gay, lively affect of the dance. In keeping with the basic perspective on dance music that I proposed above, the music for the seventeenth- and eighteenth-century bourr´ee serves as a sonic analogue for the steps of the dance. It does not, however, do so in a simple way: in the dance practice of the period, these sonic analogues

Dance topoi, sonic analogues and musical grammar

Example 10.6 Haydn, String Quartet in B flat major, Op. 76 No. 4/iv, bars 1–34

construct a framework for, rather than a simple mirror of, the choreography for the bourr´ee. The framework as a whole is an analogue for the dynamic shape of the dance, reckoned not only from the basic unit of the pas de bourr´ee but also in the way this figure is concatenated with others to create a larger pattern of eight- and sixteen-bar units. This dynamic shape informs the multimedia structure of the dance at every turn, to the extent that even when the music departs from the surface pattern of the pas de bourr´ee (departures we saw in the version of the dance that accompanied P´ecour’s choreography) it does so with the expectation that the dynamic shape continues in the steps of the dance. In instrumental forms intended to summon this multimedia structure, the dynamic shape is represented more explicitly – thus Bach and Haydn’s more regular bourr´ees. Having noted the similarities between the opening of Haydn’s Finale and a bourr´ee, I should also want to note two important differences. First, each half

297

298

l aw r e n c e m . z b i ko ws k i

Example 10.6 (cont.)

of Haydn’s opening eight-bar unit is stamped with a strong overall dynamic shape: there is no doubt that the downbeat of bar 4 and the downbeat of bar 8 are points of arrival for the melody, harmony and phrase. Second, these large-scale dynamic shapes inform relationships among statements of their constituent bourr´ee figures: in each half, the second statement of the figure carries more force than the first, and the third carries more force than the second. (The fourth statement carries rather less weight, owing to the effect of the cadence.) This kind of dynamic shaping is not a typical feature of the bourr´ee – the dance tends to be more about a subtle interplay of dynamic shapes rather than the projection of a single overarching shape. This departure suggests that the opening of Haydn’s Finale is, at best, a not very typical bourr´ee; it also suggests that Haydn has found a way to redeploy the basic dynamic shape of the dance to create new possibilities for musical communication.

Dance topoi, sonic analogues and musical grammar

Haydn’s Finale As can be seen in Example 10.6, the orderly presentation of the bourr´ee begun in the opening eight bars continues through bar 12, only to be destabilized by the imitative entry of the Viola in the middle of bar 13. Even as Violin I and Violin II continue the basic pas de bourr´ee, moving from the fourth crotchet of the bar through to the third, the Viola enters out of step, its pas de bourr´ee moving from the second crotchet of the bar through to the first. The effect is somewhat comical, rather as though a pair of dancers lost the beat and got out of phase with the rest of the troupe. This bit of comedy, however, comes at a price, for it subverts the rhythmic and gestural frame of the dance. This threat of rhythmic ambiguity is answered in bar 16 with a brusque gesture that reasserts the bourr´ee figure and leads to a somewhat overemphatic cadence on the dominant in bars 17–18. Although this cadence restores order, it also disrupts the forward motion characteristic of the bourr´ee figure that was just recalled. The cadential disruption of the basic step of the bourr´ee is, of course, something that marks each of the main formal divisions of the dance. Nonetheless, coming so close on the heels of the rhythmic ambiguity presented by bars 13–15 and stated with such emphasis, the interruption seems less a matter of articulating the form of the dance and more about calling a halt to regroup forces and reassemble the rhythmic frame. The passage that follows the cadence begins this process with an attempt to recover the forward motion, first with the accentuated dominant ninths of bars 20 and 21, and then with a melodic figure that, in bars 22–4, spins itself into a frenzy before running headlong into the beginning of the bourr´ee. The dynamic shape of the passage in bars 22–4 is worthy of a bit of comment, for it bears witness to Haydn’s larger compositional strategy. These bars begin with a compression of the melodic figure of bars 20 and 21, which is then repeated in running quavers. As shown by the upper brackets of Example 10.7, were we to follow Haydn’s bowings we could extract four patterns, each with a duration of three crotchets; together, these fill out the crotchet durations available in three bars of cut time. Ignoring the bowings and concentrating only on the basic figure (again, derived from bars 20 and 21) would yield the eight patterns shown by the lower brackets, each with a duration of three quavers. The end of neither of these patterns coincides with the end of a bar until bar 24; either singly or together they thus lead to the downbeat of bar 25. Running quavers like these often serve as sonic analogues for forward motion – indeed, up to this point in Haydn’s Finale that is just how they have operated. Here, however, they coalesce into

299

300

l aw r e n c e m . z b i ko ws k i

Example 10.7 Haydn, String Quartet in B flat major, Op. 76 No. 4/iv, bars 20–4

repeated figures that, for a brief bit of time, go nowhere. It is only when these figures finally align with the downbeat that the music moves forward once again, an event coincident with the return to the music of the opening bars. Let me pause for a moment to consider the syntactic processes that shape Haydn’s Finale up to bar 25. As was the case with Marx’s examples of the waltz, the syntactic processes of repetition and cadence are important for Haydn’s instrumental evocation of a bourr´ee. Repetition of the sonic analogue for the pas de bourr´ee creates the sense of forward motion, built up in hierarchical levels, that typifies these bars. Cadences (in bars 4 and 8) temporarily interrupt this forward motion, marking off main sections of the dance (which correspond with combinations of dance steps). These processes can, however, be used for other purposes. In bars 17–18, for instance, repeated statements of the arrival on the dominant create a comic effect: a harmony that has been, up to this point, a cooperative member of the ensemble suddenly pushes to the front of the stage and demands to be noticed. The result is a cadence, but one that marks not the completion of a sequence of dance steps but the end of a digression. Repetition plays a slightly different role in bars 22–24, for here it leads to a remarkable increase in tension as the forward motion of the anapaestic figure is turned back on itself, hinting at a stasis broken only by the downbeat of bar 25. Another syntactic process seen in these bars is imitation, through which materials that have appeared in one strand of music are restated, oftentimes with some modification, in another strand. It is imitation, of course, that disrupts the regular rhythms of the bourr´ee in bars 13–15. The subsequent return of the bourr´ee figure with the pickup to bar 16 manifests the syntactic process of displacement, in which an ongoing process is replaced by another contrasting process. Displacement can be regarded as a meta-syntactic process, in that it takes as its material other syntactic processes rather than sonic analogues proper. One of the consequences of Haydn’s employment of these various syntactic processes is a confirmation of the change in emphasis I observed in connection with bars 1–8: the listener’s attention is shifted away from the dynamic shape associated with the pas de bourr´ee and redirected toward the rhetoric of tonal forms. Haydn will continue to revisit the bourr´ee figure

Dance topoi, sonic analogues and musical grammar

throughout the remainder of the Finale; with each subsequent appearance, however, its importance as a sonic analogue for the steps of the dance will diminish. What will remain, to be refracted through the whole of Haydn’s discourse, are its broader attributes – its lively affect, its physicality, and its compelling sense of forward motion. In the following, I want to consider briefly three moments that illustrate this process and suggest that, for Haydn, musical communication was centred around the active shaping of dynamic processes such as that represented by the bourr´ee figure.

Bars 25–34 of the Finale Bar 25 begins the formal reprise that concludes the first main section of the Finale. Accordingly, bars 25–8 are almost identical to bars 1–4; the most significant departure from the version heard earlier is the dominant pedal begun when the Cello holds on to the F of bar 28. What happens next, however, completely unseats the bourr´ee motive: the melody for the second four-bar unit, begun in Violin II and accompanied by the Viola, is imitated at the interval of a crotchet by Violin I beginning on the downbeat of bar 29. On the last beat of bar 30 (just prior to what would have been the beginning of the cadential pattern, had the pattern of bars 5–8 been followed) Violin II joins with the Cello for paired quavers of the sort that typically mark the upbeat. These quavers are immediately echoed on the downbeat of bar 31 by Violin I and the Viola, and then on the second beat by the Cello. This profusion of paired quavers is followed, beginning in the second half of bar 31, with a complete absence of quavers, a change of surface rhythm that brings the succession of harmonies (which give voice to an expanded cadential pattern) to the foreground. The Cello’s pedal F has thus initiated an expansion of the dominant that reduces the nominal reprise of the tonic in bar 29 to a bit of idle chatter, the ineffectuality of which is emphasized by Violin I’s equally vapid imitation. It is as though the dancers were thrown into disarray after having become caught on the obstinate F of the Cello, and are rescued only by the march-like cadence of bars 32–4. Haydn’s introduction of the pedal has the effect of once more bringing tonal rhetoric to the foreground. Because a pedal is a syntactic process in which rhythmic activity in one part of a musical texture ceases, giving way to a sustained note, while continuing in the remaining parts, Haydn’s strategy also gives rise to a profusion of rhythmic activity (something emphasized by the close-order imitation of the bourr´ee figure by Violin I). Although the dynamic character of this rhythmic activity owes much to the bourr´ee figure, it no longer conforms to the grammar of the dance. It serves here

301

302

l aw r e n c e m . z b i ko ws k i

to contribute to an increasing sense of instability, which is then resolved through the cadential process of bars 32–4. It is just this process that provides a firm arrival on the tonic to answer the prolonged dominant initiated by the Cello’s pedal F.

Bars 64–82 of the Finale Although the contrasting music in B flat minor that Haydn introduces in bar 35 owes something to the character and dynamic shape of the bourr´ee, he does not present the bourr´ee motive with anything like the clarity of bars 1–8. One could dance a bourr´ee to this section of the Finale, but it would be without much help from the music. Indeed, by the end of this section, the last bit of which is shown at the beginning of Example 10.8, Haydn seems to have lost his way. It is only after a succession of ever-more hesitant musings in Violin I that a return to the first main section and B flat major is effected. These musings, which begin with the upbeat to bar 69, are cast in the shape of a bourr´ee figure, but without the melodic or harmonic progression that has typified Haydn’s instrumental version of the dance. This is not music that dances, but music that is trying to remember how to dance. The solution comes only on the last beat of bar 74, when Violin I finally tries a different starting pitch that, with Violin II’s help, recovers the bourr´ee once more. The passage reveals the close relationship between the bourr´ee figure and the tonal process of the Finale. By itself, the figure only weakly suggests the forward motion that typifies the dance. Indeed, when it is repeated without alteration or accompaniment, as it is in bars 69–74, it can be used to summon a kind of frustrated stasis. In the present case, when the figure is finally fleshed out as it is in bars 75–8, it marks not only the return to the dance but also the reprise of B flat major and the opening material of the movement. Its forward motion thus serves to summon not only the dance, but also the directed motion of tonal rhetoric. This is made all the more clear in bars 79–82 when the second phrase of the dance is embellished with running quavers over a syncopated accompaniment: here the overall motion of the four-bar unit toward the tonic becomes much more important than the succession of bourr´ee figures that, in bars 5–8, set out a sonic analogue for a sequence of dance steps.

Bars 106–132 of the Finale Aspects of the compositional strategy manifest in bars 64–85 can be seen in the passage that leads to the final reappearance of the bourr´ee, which is

Dance topoi, sonic analogues and musical grammar

Example 10.8 Haydn, String Quartet in B flat major, Op. 76 No. 4/iv, bars 64–82

shown in Example 10.9. Just prior to bar 106 there was a reprise of the music of bars 29–31. As happened in bars 75–82, the second phrase of the bourr´ee is embellished with running quavers. In the case of bars 103–5 (which are analogous to bars 29–31), the effect of the running quavers is to mitigate the instability introduced by the dominant pedal, such that the commencement of the cadential process in bar 106 (which is a literal repetition of bar 32) makes for a somewhat less marked contrast. When the running quavers are picked up again in bar 110 (after the augmentation of the material of bar 33 into bars 107–9, and the consequent change of the perfect authentic cadence into a semi-cadence) they sound as a continuation of the forwardmoving gesture that preceded the cadence, but now pi`u allegro. Although this continuation, taken up only by Violin I, is at first somewhat tentative it is soon distributed throughout the various voices of the ensemble and

303

304

l aw r e n c e m . z b i ko ws k i

Example 10.9 Haydn, String Quartet in B flat major, Op. 76 No. 4/iv, bars 106–32

Dance topoi, sonic analogues and musical grammar

developed through sequence and inversion. The gesture – now only vaguely reminiscent of the bourr´ee figure – gradually gains energy, and then jumps forward with the Cello entrance and forte of bar 122. It gets momentarily stuck in bars 125–8, until the reprise of the bourr´ee (still embellished in running quavers, but now pi`u presto) pushes the music toward the conclusion of the movement some forty-odd bars later. There is, at the beginning of this passage, only a faint vestige of the dance that began the movement; by its end the dance has been turned into a manic parody, rushing breathlessly forward.27 The transformation culminated in these bars, but which began as early as the out-of-phase imitative entrance of the Viola in bar 13, has shifted the listener’s attention away from the dynamic shape associated with the pas de bourr´ee and redirected it toward the rhetoric of tonal forms. It should be emphasized, however, that the sonic analogue for the steps of the dance is nonetheless of crucial importance for Haydn’s Finale, for it provides the basic grammar upon which his musical communication is based.

Conclusion Condillac was careful to insist that whatever the similarities between the langage d’action and music, there was a boundary between the two. Although the modulated pitches of the langage d’action bore a semblance to music, they were still too disorderly to constitute anything like a rational mode of communicating. Reflecting on the distinction between the two, Condillac wrote, ‘It is not sufficient for music, that the sounds succeed each other by distinct degrees, they must likewise be sustained, so as to let their harmony be perceived, and the intervals must be such as can be measured.’28 It was, of course, Rameau’s theory of harmony that provided the means to make such measurements, and thereby establish the boundary between music and barely articulate expressions of the passions. Further refinement eventually led to language, but this refinement also represented a further distance from the basic passions that had motivated expression in the first place. The development of language was thus set within a context of loss, a context that came to be thematized in Jean-Jacques Rousseau’s account of the relationship between music and language.29 What was gained, however, was the possibility of rational discourse: in the case of language, this discourse was based on instituted signs; in the case of music, it was based on the natural principles set out in Rameau’s theory.

305

306

l aw r e n c e m . z b i ko ws k i

Although it is still common to construe music and language as modes of discourse that are linked in some fundamental way, composers and theorists of the eighteenth century recognized the tenuousness of the connection. Heinrich Christoph Koch, for instance, after considering in some detail the notion of a musical grammar analogous to that of language, turned away from the idea, suggesting that music needed to be considered in its own terms.30 The perspective is similar to the one I have adopted here: whatever their surface similarities, music and language have distinctly different functions in human culture, and these are reflected in their grammars. The primary function of music is to represent through patterned sound dynamic processes that are common in human experience, a type of representation that is accomplished through sonic analogues for these processes. My point of departure for this view of musical grammar has been the sonic structures that analogize dance steps for the waltz and the bourr´ee. These structures are organized into familiar dance forms through syntactic processes which, in conformance with the construction grammar approach I adopt, are also sonic analogues for dynamic processes. It must be acknowledged that there is scant written documentation to support this perspective on eighteenth-century compositional practice. Although we know that dance forms were important to composers – as social structures, as models for composition, and as musical topoi – I am aware of no account from the period suggesting that anything like sonic analogues are at the centre of musical grammar. If, however, we turn to the music, we can see ample evidence for an exquisite attention to the details of dynamic process. It is this that I endeavoured to trace in my analysis of the Finale of Haydn’s Op. 76 No. 4 string quartet. There, Haydn gradually reshaped an evocation of the bourr´ee by adapting the dynamic gestures basic to the dance to the requirements of tonal rhetoric. For Haydn, musical communication was about many things, but it was first and foremost a matter of creating a soundscape with which listeners could engage and, through a sympathy or empathy with the dynamic processes shaping that soundscape, be moved. So moved, they might recall emotions and feelings, but they might equally recall the motions of their bodies through space or the path of an object across their vision. Gustave Flaubert, reflecting on the sometimes slender resources language offers for the expression of emotions, wrote that ‘human language is like a cracked kettledrum on which we beat out tunes for bears to dance to, when what we long to do is make music that will move the stars to pity’.31 Flaubert’s analogy suggests the limitations of language, as well as the resources of music. These are not the resources of language, nor are they the resources of rational

Dance topoi, sonic analogues and musical grammar

discourse. They are resources felt rather than known, enacted rather than objectified; they are the resources that resonate deep within the heart of what it is to be human.

Notes 1 Etienne Bonnot de Condillac, An Essay on the Origin of Human Knowledge, trans. Thomas Nugent (London: J. Nourse, 1756). Reprinted with introduction by James H. Stam in the series Language, Man and Society (New York: AMS, 1974), 172. 2 Condillac, An Essay on the Origin of Human Knowledge, 209. Rameau’s theory of harmonic generation was first stated in his G´en´eration harmonique ou trait´e de musique th´eorique et pratique (Paris: Prault fils, 1737) reprinted in Monuments of Music and Music Literature in Facsimile. Second Series, Music Literature (New York: Broude Brothers, 1966). 3 Downing A. Thomas, Music and the Origins of Language: Theories from the French Enlightenment (Cambridge: Cambridge University Press, 1995), 9. 4 For a discussion see V. Kofi Agawu, ‘The Challenge of Semiotics’, in Rethinking Music, ed. Nicholas Cook and Mark Everist (Oxford: Oxford University Press, 1999), 138–60. A similar distinction between language and music can be seen in Susanne K. Langer, Philosophy in a New Key: A Study in the Symbolism of Reason, Rite, and Art (Cambridge, MA: Harvard University Press, 1953), ch. 8. 5 Michael Tomasello, The Cultural Origins of Human Cognition (Cambridge, MA: Harvard University Press, 1999), ch. 5. 6 Antonio R. Damasio, The Feeling of What Happens: Body and Emotion in the Making of Consciousness (New York: Harcourt Brace, 1999); Antonio R. Damasio, Looking for Spinoza: Joy, Sorrow, and the Feeling Brain (Orlando, FL: Harcourt, 2003). 7 Edward Sapir, Language, an Introduction to the Study of Speech (New York: Harcourt, Brace and Company, 1921), 126. 8 As I use the term, ‘representation’ does not have to entail a full semiotic system. This perspective is similar to that adopted by Naomi Cumming in The Sonic Self: Musical Subjectivity and Signification (Bloomington: Indiana University Press, 2000), ch. 3. A dynamic process may be provisionally defined as a coherent sequence of phenomena that is distributed over time and typified by parametric modulation or change. 9 Adolf Bernhard Marx, Die Lehre von der musikalischen Komposition, praktischtheoretisch, vol. 2 (Leipzig: Breitkopf und H¨artel, 1838), 55. 10 Thomas Wilson, A Description of the Correct Method of Waltzing, the Truly Fashionable Species of Dancing (London: Printed for the author, published by Sherwood, Neely, and Jones, 1816), 46.

307

308

l aw r e n c e m . z b i ko ws k i

11 Sevin Yaraman, Revolving Embrace: The Waltz as Sex, Steps, and Sound (Hillsdale, NJ: Pendragon, 2002), 18–19. 12 Marx, Die Lehre von der musikalischen Komposition, vol. 2, 55. 13 Marx, Die Lehre von der musikalischen Komposition, vol. 2, 56. See also Wilson, A Description of the Correct Method of Waltzing, lv–lvi. 14 The idea that more is required for a waltz than simply music in triple time is also emphasized by Wilson in his treatise on the dance, who observes that ‘v e ry f e w musical composers have a sufficient knowledge of da n c i n g , to enable them to understand the nature or composition of Waltzing; yet, their productions, if composed in three-eight or three-four, whether adapted to Waltzing or not, are almost invariably without hesitation denominated wa lt z e s .’ Wilson, A Description of the Correct Method of Waltzing, lv. 15 This de-emphasis of tonal syntax can be seen in any number of the waltzes of Franz Schubert, which at times offer quite striking juxtapositions of key centre in the two halves of the waltz. See for instance his Sechzehn L¨andler und zwei Ecossaisen (genannt ‘Wiener Daimler-L¨andler’), Op. 67 (1822; d734): No. 1 (G major and E major), No. 6 (A major and F major), No. 8 (C major and A flat major) and No. 14 (B major, cadencing on C♯ minor, followed by D major and concluding in B major). 16 Adele E. Goldberg, ‘Constructions: A New Theoretical Approach to Language’, Trends in Cognitive Science 7 (2003), 219. 17 Lawrence M. Zbikowski, Conceptualizing Music: Cognitive Structure, Theory, and Analysis (New York: Oxford University Press, 2002), ch. 1. 18 Michael J. Reddy, ‘The Conduit Metaphor: A Case of Frame Conflict in Our Language About Language’, in Metaphor and Thought, second edition, ed. Andrew Ortony (Cambridge: Cambridge University Press, 1993), 170. 19 Meredith Little and Natalie Jenne, writing about French dance of the seventeenth and early eighteenth centuries, observe that, ‘In all choreographies, the rhythms of music and dance form counter-rhythms at least some of the time.’ That is, the choreography will imply one set of rhythmic structures, and the music will imply another. While these structures will align at important points of articulation, at other points they may be at quite a variance. For further discussion see Little and Jenne, Dance and the Music of J. S. Bach, revised edition (Bloomington: Indiana University Press, 2001), 24–5. 20 Nicholas Cook, Analysing Musical Multimedia (Oxford: Clarendon, 1998), 98– 104. 21 Wolfgang Budday, Grundlagen musikalischer Formen der Wiener Klassik: An Hand der zeitgen¨ossischen Theorie von Joseph Riepel und Heinrich Christoph Koch dargestellt an Menuetten und Sonatens¨atzen (1750–1790) (Kassel: B¨arenreiter, 1983). 22 Allanbrook makes just this argument with respect to Mozart; see her Rhythmic Gesture in Mozart: Le Nozze di Figaro and Don Giovanni (Chicago: University of Chicago Press, 1983), 32.

Dance topoi, sonic analogues and musical grammar

23 For a similar perspective, see David Lidov, Is Language a Music? Writings on Musical Form and Signification (Bloomington: Indiana University Press, 2005), ch. 8 and Cumming, The Sonic Self, ch. 5, especially 163–5. 24 Little and Jenne, Dance and the Music of J. S. Bach, 37. Little and Jenne’s characterization is slightly inaccurate in that the gesture they call an upbeat occupies only one half a beat: where the beat is a minim, for instance, the ‘upbeat’ is a crotchet. 25 Pierre Rameau, The Dancing Master, trans. Cyril W. Beaumont (New York: Dance Horizons, 1970), 78. See also Wendy Hilton, Dance of Court & Theater: The French Noble Style, 1690–1725, ed. Caroline Gaynor (Princeton: Princeton Book Company, 1981), 183. 26 Meredith Ellis Little, ‘The Contribution of Dance Steps to Musical Analysis and Performance: “La Bourgogne”’, Journal of the American Musicological Society 28 (1975), 112–14. 27 To the extent that this parody is heard as comedic it conforms with the perspective on Haydn’s compositional style presented in Gretchen A. Wheelock, Haydn’s Ingenious Jesting with Art: Contexts of Musical Wit and Humor (New York: Schirmer Books, 1992). 28 Condillac, An Essay on the Origin of Human Knowledge, 181. 29 Rousseau’s account of the development of language, and the relationship between music and language, is discussed in Thomas, Music and the Origins of Language, ch. 4. 30 Heinrich Christoph Koch, Introductory Essay on Composition: The Mechanical Rules of Melody, Sections 3 and 4, trans. Nancy Kovaleff Baker (New Haven: Yale University Press, 1983), 6. 31 Gustave Flaubert, Madame Bovary: Provincial Manners, trans. Margaret Mauldon, introduction by Malcolm Bowie (Oxford: Oxford University Press, 2004), 170.

309

Afterword ko f i ag aw u

Perhaps the most fundamental question to ask about the organizing rubric of this volume is whether music is a system of communication. The answer is almost certainly no. Equally important – and necessary – is to ask whether music has the capacity to communicate. That answer is probably yes. Thus we are caught in a paradoxical situation. Unlike language, music (specifically the art music of the late eighteenth and early nineteenth centuries) does not function primarily as a system of communication. (Even language, for its part, does not function exclusively as such.) Like myth and religion, however, music was sometimes pressed into communicative service, saddled with a communicative function by diktat, made to bear the weight of assigned meanings. Clearly, then, any attempt to read the instrumental works of Haydn, Mozart or Beethoven as continuously meaningful sites for the cultivation of a specifiable, non-trivial content is bound to encounter resistances. At the same time, there is no question that certain aspects of the making and rendering of this repertoire reinforce the belief that there are musical codes to be decoded and signs to be recognized or understood; what is ‘said’ during the course of a composition can at the very least be pointed to, even if it cannot be named precisely.1 To frame the concerns of these ten essays in terms of a ruling paradox – the simultaneous communicability and non-communicability of eighteenthcentury music – is to place the collective enterprise in a venerable tradition of aesthetic, philosophical, music-theoretical and music-historical debate reaching back at least to the eighteenth century. Writers have asked whether music has any content to speak of, whether it is a language, and whether it is translatable. When Hans-Georg N¨ageli, for example, says that music ‘has no content of any kind that men have tried to adduce from and give to it. It simply comprises forms, regulated combinations of sounds and sequences of sounds . . . Its essence is play, through and through,’2 he implies that there is no message-laden content to be communicated, no code to be deciphered, no truths to be unveiled; there is only the challenge of playing and singing, and playing again and singing again. That sentiment is echoed throughout the history of thought about music. In our own day, it appears, for example, in the musical philosophy of Vladimir Jank´el´evitch:

Afterword

Directly, in itself, music signifies nothing, unless by convention or association. Music means nothing and yet means everything. One can make notes say what one will, grant them any power of analogy: they do not protest. In the very measure that one is inclined to attribute a metaphysical significance to musical discourse, music (which expresses no communicable sense) lends itself, complaisant and docile, to the most complex dialectical interpretations.3

But this is only one side of the story. The expressive figures and topoi that animate eighteenth-century music provide opportunities for kinds of interpretation that, whether deluded or not, rest on what many authors believe to be concrete, specifiable content. Indeed, the hermeneutic tradition of musical interpretation, from E. T. A. Hoffman on, sponsors readings in which talk of angels, demons and dance, peasants, the Orient and armies, greetings, farewells and protests are said to be directly inferable from the musical works themselves.4 To speak of communicative strategies is thus to embrace both idea and illusion, and it is probably because of this doubleness – this ultimate undecidability – that the pursuit of a verbally mediated musical meaning has become something of a permanent temptation for writers on eighteenthcentury music.5 The key word here is writing, for without the problematics of writing we would probably be spared many of the less productive aspects of the debate about musical meaning. Were our interests concentrated, for example, on performing (including improvising), composing or even recomposing, we would not necessarily escape the challenge of deciphering codes, but those aspects of interpretation that are not organically linked with what we used to call ‘the music itself’, or those issues that are generated directly by the promiscuous and unavoidably intertextual nature of (written and spoken) language, would probably play a subsidiary role. Insofar as it returns again and again to haunt our discussion of music, it is best if we make our peace with language. Debates about musical meaning and musical communication, then, will almost certainly be with us for some time, especially given the repetitive way in which institutional knowledge is produced, consumed and reproduced in the modern (musical) academy. Such debates profit less from being formulated in general, ideologically partisan or extreme terms, however; they are better approached analytically and with rigorous attention to context. The analytical orientation of the majority of essays gathered in this volume is therefore both necessary and wise. If music’s communicative capacity cannot be assumed but must be postulated or demonstrated, then close reading of a phrase of Haydn’s, or the theme of a variation movement by Mozart, or

311

312

ko f i ag aw u

a string quartet by Beethoven, is a necessary step towards jumpstarting the discussion. We all know, however – and here we remain with the theme of language – that reading analyses of even relatively well-known compositions comes with its own challenges, requiring the reader to consult text, score and corroborating sound sources regularly and repeatedly, often in a non-linear manner. There is, moreover, the challenge of metalanguage, specifically of technical designation, made more formidable if readers do not all belong to the same interpretive community. Unfortunately, there is no way to avoid these hurdles. Foremost among the methodological challenges faced by the authors of the foregoing essays is the question of authority in analysis and the related (non-)issue of anachronism. It is now generally acknowledged that some eighteenth- and early nineteenth-century writers – ‘contemporaneous’ writers, some call them – had valuable things to say about the repertories studied in this volume, and therefore that attention to the writings of figures like Koch, Kirnberger and Riepel is not only desirable but even indispensable. William Rothstein’s attempt to impose a set of national identities on metrical notational practice draws productively on this heritage, including its manifestation in recent secondary literature. In this way, the oft-implicit authority of precedent is properly acknowledged. Similarly, Mirka’s demonstration of a set of (intentionally playful) manipulations of metre at the beginnings of chamber works by Haydn and Mozart, manipulations that are heard and felt but not necessarily seen in scores, is enriched by dialogue with some of the same figures.6 Mirka is especially concerned not to dissolve the nuances of period terminology into bland, functional translations (of mainly German terms), but to capture the sound of theorizing and thus to extend the effective realms in which ‘communicative strategies’ are deployed into the relatively unassailable domain of theory-making. In a similar vein, Claudia Maurer Zenck’s recovery of wit, humour, the irrational and the illogical in piano sonatas by Beethoven is based on a frank application of a set of contemporaneous categories postulated by Michaelis, Rochlitz and others. Exotic at first, this dialogue with eighteenthand early nineteenth-century ‘others’ soon reveals that our worlds and theirs are intertwined rather than separate, that our ostensible deafness to some of the subtleties of comic construction in Beethoven is a sign of the natural deficits that accrue over two centuries, not the severance of an umbilical cord. Again, in reconstructing the nature of a typically heterogeneous audience of the late eighteenth century, Mark Evan Bonds reaches for such contemporaneous sources as manuals on rhetoric, letters, and reviews of music in order to establish the nature of composers’ priorities, and to demonstrate an

Afterword

awareness – specifically on the part of Haydn and Mozart – of the consumption patterns and expectations of listeners classified as Kenner or Liebhaber. As with other authors, Bonds does not merely report what was said but enters speculatively into the compositional domestication of these ideas in order to correlate theory with practice. Many who subscribe to historicist methodologies often say that our perspectives are enlivened by them, our horizons of understanding expanded and enriched, sometimes even transformed. Of course, we can only understand better if we do not ignore what theorists closer to the time said. But what about analyses informed by more recent sensibilities; sensibilities that are presumably shaped by the wisdom of hindsight? Are such analyses anachronistic? We have already seen that Rothstein and Mirka, for example, bring period and modern perspectives into dialogue, even if the period sediment seems especially significant in their essays. The balance is reversed in William Caplin’s demonstration of the melodic potential of bass lines. Departing from the thought of twentieth-century figures like Schoenberg, Leonard Meyer and, to a lesser extent, Schenker in absentia, Caplin identifies certain archetypal bass patterns, explores their voice-leading tendencies, and establishes for the bass voice both compositional and analytical pertinence as a ‘second melody’. Koch would never have heard this way, and would almost certainly not have used this particular metalanguage. This says nothing, however, to undermine the suggestiveness of Caplin’s demonstration for appreciating the web of voice-leading connections in this repertoire. Michael Spitzer’s theory of sonata form is likewise heavy on recent theory – and here we are justified in referring to theory with a capital T – but the dialogue with contemporaneous theorists is not so much abandoned as transfigured. Spitzer’s bold thesis is that the trajectories of the two key areas of a sonata-form exposition map directly onto one another, that the second key area is in effect a recomposition of the first. The theory is so obviously a product of our post-structuralist age that it seems pointless to embark on a search for historical anticipations of its guiding insight. It is up to the listener to test the aural validity of this theory; this, to be sure, requires more effort than testing some of the claims made in the other essays, yet listening to Mozart in the age of Schenker and psychoanalysis is not a prospect to be shunned (it is too late for that, anyway!) but a mode of existence to be acknowledged and eventually celebrated. Progress in musical thought is impossible without the kinds of adventures that Spitzer embarks upon. In his detailed analyses of the first movements of Mozart’s C major Quintet, k515, and the ‘Jupiter’ Symphony, Spitzer implicitly rejects the limitations imposed by the normative compositional or pedagogical orientation of

313

314

ko f i ag aw u

eighteenth-century writers, but allows the production of initially unclaimed observations about his chosen works. The analytical act thus produces a surplus, an excess. Many more relations – deriving from the manipulation of schemata and from motivic affiliating – are produced than are hearable. The fact, however, that not all relations produced in the course of analysis are immediately hearable does not equip us to refute their reality; perhaps the obscurity of such relations is only temporary. No, the analyst must retain these findings as data, to be used if and when the need arises. Part of my own study of Beethoven’s Op. 18 No. 3, first movement, takes the recurring i–v–i progression (referred to as a ‘tonal model’) as a point of entry into the harmonic-contrapuntal, and thus formal, layer of this early quartet. Harmonic awareness is of course crucial to an appreciation of the Classic repertoire. Ratner, Dahlhaus, Tovey and Rosen, among others, have reminded us that it is the dramatization of harmonic relations that embodies a key aspect of the music’s rhetoric. In terms of the historiographical, ancient-versus-modern dichotomy that we have been invoking, I can think of no eighteenth- or early nineteenth-century writer who would seek to track the successive occurrences of a i–v–i progression in quite this way, arrange the results in the form of paradigmatic charts, and conclude that there is something circular about the dynamic process of Beethoven’s music. The meta-theoretical dialogue, rather, is with musical semiologists like Ruwet and Nattiez, who, among other things, pursued systematic methods of paradigmatic analysis, though not in reference to Beethoven. In the end, we would have to say that if the listening experience supports a given insight, then the charge of anachronism – a charge that often ignores music’s essence as a performed art – may well be beside the point. It would seem, then, that the ideal methodological stance is one that accepts the imperatives of both ancient and modern worlds; of both ‘contemporaneous’ writers and present-day theorists. Our challenge is to fuse these varied horizons. Although not framed in precisely these terms, several of the essays that employ notions of topos or topic shed light on this need. The idea of topic seems always already made for anyone anxious to prove that the Classic repertoire manifests an essential communicative tendency. Widespread affirmation of the importance of topic to an understanding of the expressive and dramatic aspects of Classic music has come in the last quarter of a century, even though the basic taxonomic project remains to be completed.7 As one who always thought that a case needed to be made for reading Classic works in reference to a learned, singing or brilliant style, a bourr´ee, gavotte or contredanse, the hunt, military or pastoral styles, I was pleasantly surprised to discover during the 2005 Bad Sulzburg gathering that

Afterword

notions of topic were being invoked as if they were a normal part of what we do when we analyse this repertoire, as if they were an accepted frame of reference whose constituent terms and concepts could be deployed without enclosing them in quotation marks. That tendency is localized in a number of essays in this volume. Lawrence Zbikowski, for example, follows Allanbrook in hearing the finale of Haydn’s Op. 76 No. 4 as a bourr´ee. Zbikowski’s interest is in exploring a nexus of relationships involving music, language and dance. Without the speculative projection of a stylized dance onto this finale, the task of demonstrating the peculiar nature of musical communication as a succession of sonic analogues to material units in other domains would have been made much harder. In her own essay on the well-known theme of Mozart’s Piano Sonata in A major, k331, Allanbrook unveils a siciliano topic hitherto overlooked by most theorists and analysts. This motivating insight enables a highly nuanced exploration of the expressive modes that Mozart’s theme makes possible. In my own reading of Beethoven’s Op. 18 No. 3, I follow Ratner’s identification of topics such as alla breve, brilliant style, march and fantasia in order to forge a reading that seeks to capture the socio-historical and stylistic specificity of Beethoven’s textures. Indeed, during the informal analytical sessions held in the evenings at Bad Sulzburg, the idea of topic was never absent from discussions of compositions by Mozart and Beethoven. Because topical categories are for the most part drawn from contemporaneous theory, there is an immediate sense in which they seem to promise a period view. But the actual application of these categories – or, formulated in the most extreme form, the wringing of topical significance out of every last textural event in a composition – is very much a preoccupation of our own time. While the pioneering research of Ratner and Allanbrook directed us to the relevant contemporaneous theoretical sources, fully fledged topical analyses as separate endeavours are almost without precedent. Once again, the dichotomy between a supposedly authentic eighteenth-century view and an anachronistic modern one is rendered problematic. Topical analysis may bring us close to the sound world of the eighteenth century and to the priorities of its composers, but hearing a work as a succession of topics is, if not an invention of our own time, then almost certainly an activity influenced by the imperatives of formal designation enshrined in the tradition of Formenlehre that dominated twentieth-century teaching of form and analysis. The perspectives opened up in these ten essays, then, promise well for the future of research into eighteenth-century music. The value of transcending disciplinary boundaries is evident, on the one hand, in the implicit

315

316

ko f i ag aw u

dialogue between Cobley’s reflections on communication in the eighteenth century and Bonds’ reconstruction of contemporary audiences, and, on the other, in the forays into dance, cognition and communication theory that serve as a backdrop to Zbikowski’s analysis of the finale to Haydn’s Op. 76 No. 4. The case for a historically mediated analysis is well made by Maurer Zenck, Mirka and Allanbrook; indeed their analyses contribute to another border-crossing project, that of ‘de-exoticizing’ the eighteenth century, or at least problematizing the distinction between its world view and ours. Ongoing fascination with the transmission of theoretical ideas is helped along by Rothstein’s essay which, among other things, dissuades us from wilfully reinventing wheels without at the same time denying us the opportunity to say some things again – for ourselves, this time. Finally, we are reminded again of the sheer pleasure in engaging directly with a living and vibrant repertory. The analytical offerings by Spitzer, Caplin, Zbikowski and Allanbrook may further fuel the suspicion, already set in motion by a significant spate of recent theorizing, that our understanding of late eighteenth-century music has reached a uniquely sophisticated stage in its combination of historical theory (including aesthetics) and systematic analysis. If in closing we pose once again the questions, ‘Is music a system of communication?’ and ‘Does music (have the capacity to) communicate?’, we see that the respective negative and affirmative answers given earlier, answers that then suggested a paradox, were not merely provisional but, in a sense, ultimately beside the point. As motivating or organizing questions, they have, I believe, done their work by inspiring a collection of essays that demonstrates the sheer meaningfulness of the contexts and attendant procedures that shaped – and continue to shape – the production and consumption of eighteenth-century music.8 Notes 1

2

For a recent affirmation of music’s communicative capacity, drawing on perspectives from psychology, education and communication studies, see Musical Communication, ed. Dorothy Miel, Raymond MacDonald and David Hargreaves (Oxford: Oxford University Press, 2005), which may be fruitfully contrasted with an earlier, more cautious semiological study by Jean-Jacques Nattiez, Music and Discourse: Toward a Semiology of Music, trans. Carolyn Abbate (Princeton: Princeton University Press, 1990), 3–37. Hans-Georg N¨ageli, Vorlesungen u¨ ber Musik mit Ber¨ucksichtigung der Dilettanten (Stuttgart and T¨ubingen: Cotta, 1826), 32; quoted and translated in Music and Aesthetics in the Eighteenth and Early-Nineteenth Centuries, ed. Peter le Huray and James Day (Cambridge: Cambridge University Press, 1981), 398.

Afterword

3 4

5

6

7

8

Vladimir Jank´el´evitch, Music and the Ineffable, trans. Carolyn Abbate (Princeton: Princeton University Press, 2003), 11. A sampling of the hermeneutic tradition may be found in Music Analysis in the Nineteenth Century, vol. 2: Hermeneutic Approaches, ed. Ian Bent (Cambridge: Cambridge University Press, 1994). Among recent reflections on musical meaning (not confined to the eighteenth century), Daniel K. L. Chua’s Absolute Music and the Construction of Meaning (Cambridge: Cambridge University Press, 1999) offers much food for thought. See also Craig Harwood, ‘Subversive Strategies: Conventions and Manipulation of Gesture and Syntax in Mozart’ (Ph.D. dissertation, Yale University, 2002) which, however, draws mostly on recent theory. Foundational works in topic theory include Leonard G. Ratner, Classic Music: Expression, Form, and Style (New York: Schirmer Books, 1980); Wye J. Allanbrook, Rhythmic Gesture in Mozart: Le Nozze di Figaro and Don Giovanni (Chicago: University of Chicago Press, 1983); Robert Hatten, Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation (Bloomington: Indiana University Press, 1994) and Raymond Monelle, The Musical Topic: Hunt, Military and Pastoral (Bloomington: Indiana University Press, 2006). Many thanks to Rowland Moseley for superb editorial assistance, and to Johanna Frymoyer for help in the preparation of this work.

317

Bibliography

Abbate, Carolyn and Roger Parker. ‘Dismembering Mozart’, Cambridge Opera Journal 2 (1990), 187–95. Abravaya, Ido. ‘The Baroque Upbeat: Outline of its Typology and Evolution’, in Bach Studies from Dublin, ed. Anne Leahy and Yo Tomita, 17–28. Irish Musical Studies 8. Dublin: Four Courts, 2004. Adorno, Theodor. Beethoven: The Philosophy of Music, trans. Edmund Jephcott, ed. Rolf Tiedemann. Stanford, CA: Stanford University Press, 1998. Agawu, V. Kofi. ‘Schenkerian Notation in Theory and Practice’, Music Analysis 8 (1989), 275–301. Playing with Signs: A Semiotic Interpretation of Classic Music. Princeton: Princeton University Press, 1991. ‘Haydn’s Tonal Models: The First Movement of the Piano Sonata in E-flat Major, Hob. XVI:52’, in Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt, 3–22. Stuyvesant, NY: Pendragon, 1992. ‘Mozart’s Art of Variation: Remarks on the First Movement of K. 503’, in Mozart’s Piano Concertos: Text, Context, Interpretation, ed. Neal Zaslaw, 303–13. Ann Arbor: University of Michigan Press, 1996. ‘The Challenge of Semiotics’, in Rethinking Music, ed. Nicholas Cook and Mark Everist, 138–60. Oxford: Oxford University Press, 1999. ‘Il ritmo’, trans. Fulvia De Colle, in Enciclopedia della Musica II: Il sapere musicale, ed. Jean-Jacques Nattiez, 45–71. Torino: Einaudi, 2002. ‘How We Got out of Analysis, and How to Get Back in Again’, Music Analysis 23 (2004), 267–86. Allanbrook, Wye J. Rhythmic Gesture in Mozart: Le Nozze di Figaro and Don Giovanni. Chicago: University of Chicago Press, 1983. ‘Two Threads Through the Labyrinth: Topic and Process in the First Movements of K. 332 and K. 333’, in Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt, 125–72. Stuyvesant, NY: Pendragon, 1992. Altman, Rick. Film/Genre. London: British Film Institute, 1999. Anonymous, ed. Le chansonnier fran¸cois; ou, Recueil de chansons, ariettes, vaudevilles et autres couplets choisis. 16 vols. Paris, 1760–1762. Reprinted Geneva: Slatkine, 1971.

Bibliography

Aristotle. The ‘Art’ of Rhetoric, trans. John Henry Freese. Cambridge, MA: Harvard University Press, 1947. Asioli, Bonifazio. Il maestro di composizione. 3 vols. Milan: G. Ricordi, 1832. Bach, Carl Philipp Emanuel. Briefe und Dokumente: Kritische Gesamtausgabe, ed. Ernst Suchalla. 2 vols. G¨ottingen: ‘Vandenhoeck und Ruprecht, 1994. Badura-Skoda, Eva. ‘Dittersdorf u¨ ber Haydns und Mozarts Quartette’, in Collectanea Mozartiana, ed. Cordula Roleff, 41–50. Tutzing: Schneider, 1988. Baker, James. ‘Chromaticism in Classical Music’, in Music Theory and the Exploration of the Past, ed. David Bernstein and Christopher Hatch, 233–307. Chicago: University of Chicago Press, 1993. Baumol, William J. and Hilda Baumol, ‘On the Economics of Musical Composition in Mozart’s Vienna’, in On Mozart, ed. James M. Morris, 72–101. Washington, DC: Woodrow Wilson Center Press, 1994. Beach, David. ‘The Current State of Schenkerian Research’, Acta Musicologica 57 (1985), 275–307. ‘Schenkerian Theory’, Music Theory Spectrum 11 (1989), 3–14. Becking, Gustav. Studien zu Beethovens Personalstil: Das Scherzothema. Leipzig: Breitkopf und H¨artel, 1921. Beer, Axel. Musik zwischen Komponist, Verlag und Publikum: Die Rahmenbedingungen des Musikschaffens in Deutschland im ersten Drittel des 19. Jahrhunderts. Tutzing: Schneider, 2000. ‘Composers and Publishers: Germany, 1700–1830’, in Music Publishing in Europe, 1600–1900: Concept and Issues, Bibliography, ed. Rudolf Rasch, 160–81. Berlin: Berliner Wissenschafts-Verlag, 2005. Beethoven, Ludwig van. Briefwechsel: Gesamtausgabe, ed. Sieghard Brandenburg. 7 vols. Munich: Henle, 1996–8. Bekker, Paul. Beethoven. Berlin: Schuster und Loeffler, 1912. Bent, Ian ed. Music Analysis in the Nineteenth Century, vol. 2: Hermeneutic Approaches. Cambridge: Cambridge University Press, 1994. Benton, Rita. Ignace Pleyel: A Thematic Catalogue of his Compositions. New York: Pendragon, 1977. Berger, Karol. ‘Toward a History of Hearing: The Classic Concerto, A Sample Case’, in Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. Wye J. Allanbrook, Janet M. Levy and William P. Mahrt, 405–29. Stuyvesant, NY: Pendragon, 1992. ‘The First Movement Punctuation Form in Mozart’s Piano Concertos’, in Mozart’s Piano Concertos: Text, Context, Interpretation, ed. Neal Zaslaw, 239–59. Ann Arbor: University of Michigan Press, 1996. Besseler, Heinrich. ‘Grundfragen des musikalischen H¨orens’, Jahrbuch Peters 1925, 35–52. Das musikalische H¨oren der Neuzeit. Berichte u¨ ber die Verhandlungen der S¨achsischen Akademie der Wissenschaften zu Leipzig, Philologisch-historische Klasse 104/6. Berlin: Akademie-Verlag, 1959.

319

320

Bibliography

Bilson, Malcolm. ‘Execution and Expression in the Sonata in E Flat, k282’, Early Music 20 (1992), 237–43. Blair, Hugh. Lectures on Rhetoric and Belles Lettres. 3rd edn. Edinburgh: J. Murray, 1787. Blume, Friedrich. ‘Fortspinnung und Entwicklung: Ein Beitrag zur musikalischen Begriffsbildung’, Jahrbuch Peters 1929, 51–70. Classic and Romantic Music: A Comprehensive Survey, trans. M. D. Herter Norton. New York: Norton, 1970. Bonds, Mark Evan. Wordless Rhetoric: Musical Form and the Metaphor of the Oration. Cambridge, MA: Harvard University Press, 1991. ‘Haydn, Laurence Sterne, and the Origins of Musical Irony’, Journal of the American Musicological Society 44 (1991), 57–91. Music as Thought: Listening to the Symphony in the Age of Beethoven. Princeton: Princeton University Press, 2006. ‘Replacing Haydn: Mozart’s “Pleyel” Quartets’, Music and Letters 88 (2007), 201– 25. Botstein, Leon. ‘Listening Through Reading’, 19th-Century Music 16 (1992), 129–45. Bregman, Albert. Auditory Scene Analysis: The Perceptual Organization of Sound. Cambridge, MA: The MIT Press, 1990. Budday, Wolfgang. Grundlagen musikalischer Formen der Wiener Klassik: An Hand der zeitgen¨ossischen Theorie von Joseph Riepel und Heinrich Christoph Koch dargestellt an Menuetten und Sonatens¨atzen (1750–1790). Kassel: B¨arenreiter, 1983. Burney, Charles. Tagebuch einer musikalischen Reise durch Frankreich und Italien, trans. Christoph Daniel Ebeling. Hamburg: Bode, 1772. Ed. Eberhardt Klemm. Wilhelmshaven: Heinrichshofen, 1980. Memoirs of the Life and Writings of the Abate Metastasio. 3 vols. London: G. G. and J. Robinson, 1796. An Eighteenth-Century Musical Tour in Central Europe and the Netherlands, ed. Percy A. Scholes. London: Oxford University Press, 1959. Burnham, Scott. Beethoven Hero. Princeton: Princeton University Press, 1995. Burstein, Poundie L. ‘Surprising Returns: The VII♯ in Beethoven’s Op. 18, No. 3, and its Antecedents in Haydn’, Music Analysis 17 (1998), 295–312. Campbell, George. The Philosophy of Rhetoric, ed. Lloyd F. Bitzer. Carbondale: Southern Illinois University Press, 1963. Caplin, William E. ‘Hugo Riemann’s Theory of “Dynamic Shading”: A Theory of Musical Meter?’, Theoria 1 (1985), 1–24. ‘The “Expanded Cadential Progression”: A Category for the Analysis of Classical Form’, Journal of Musicological Research 7 (1987), 215–57. Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven. New York: Oxford University Press, 1998. ‘The Classical Cadence: Conceptions and Misconceptions’, Journal of the American Musicology Society 57 (2004), 51–117.

Bibliography

‘On the Relation of Musical Topoi to Formal Function’, Eighteenth-Century Music 2 (2005), 113–24. Carpani, Giuseppe. Le Haydine. Milan: Candido Buccinelli, 1812. Chua, Daniel K. L. Absolute Music and the Construction of Meaning. Cambridge: Cambridge University Press, 1999. ‘Myth: Mozart, Money, Music’, in Mozart Studies, ed. Simon P. Keefe, 193–213. Cambridge: Cambridge University Press, 2006. Cobley, Paul. Narrative. London: Routledge, 2001. ‘Objectivity and Immanence in Genre Theory’, in Genre Matters: Interdisciplinary Perspectives, ed. Garin Dowd et al. London: Intellect, 2005. ‘General Introduction’, in Communication Theories, ed. Paul Cobley, vol. 1, 1–33. London: Routledge, 2006. Condillac, Etienne Bonnot de. An Essay on the Origin of Human Knowledge. Trans. Thomas Nugent. London: Printed for J. Nourse, 1756. Reprinted New York: AMS, 1974. Cone, Edward T. Musical Form and Musical Performance. New York: Norton, 1968. ‘Twelfth Night’, Journal of Musicological Research 7 (1987), 136–41. Cook, Nicholas. Analysing Musical Multimedia. Oxford: Clarendon Press, 1998. Cooper, Grosvenor and Leonard B. Meyer. The Rhythmic Structure of Music. Chicago: University of Chicago Press, 1960. Couperin, Franc¸ois. L’Art de toucher le clavecin. Paris, 1716. Cowan, Brian William. ‘Mr. Spectator and the Coffeehouse Public Sphere’, Eighteenth-Century Studies 37 (2004), 345–66. Cramer, Carl Friedrich, ed. Magazin der Musik 2. Hamburg: Musicalische Niederlage, 1787. Cumming, Naomi. The Sonic Self: Musical Subjectivity and Signification. Bloomington: Indiana University Press, 2000. Czerny, Carl. Vollst¨andige theoretisch-practische Pianoforte-Schule, Op. 500, vol. 3: Von dem Vortrage. Vienna: Diabelli, 1839. Die Kunst des Vortrags der a¨ ltern und neuern Claviercompositionen. Vienna: Diabelli, 1842. Dahlhaus, Carl. ‘Some Models of Unity in Musical Form’, Journal of Music Theory 19 (1975), 2–30. Damasio, Antonio R. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Harcourt Brace, 1999. Looking for Spinoza: Joy, Sorrow, and the Feeling Brain. Orlando, FL: Harcourt, 2003. Daub, Peggy. ‘The Publications Process and Audience for C. P. E. Bach’s “Sonaten f¨ur Kenner und Liebhaber”’, Bach Perspectives 2 (1996), 65–83. Drabkin, William, Susanna Pasticci and Egidio Pozzi. Analisi schenkeriana: Per un’interpretazione organica della struttura musicale. Lucca: Libreria Musicale Italiana, 1995. Dreyfus, Laurence. Bach and the Patterns of Invention. Cambridge: Harvard University Press, 1996.

321

322

Bibliography

Eisen, Cliff. New Mozart Documents. London: Macmillan, 1991. ‘The Mozarts’ Salzburg Music Library’, in Mozart Studies 2, ed. Cliff Eisen, 85–138. Oxford: Clarendon, 1997. Epstein, David. Shaping Time: Music, the Brain, and Performance. New York: Schirmer Books, 1995. Feather, John. ‘The Commerce of Letters: The Study of the Eighteenth-Century Book Trade’, Eighteenth Century Studies 17 (1984), 405–24. Feil, Arnold. ‘Zum Rhythmus der Wiener Klassiker’, in Internationaler Musikwissenschaftlicher Kongreß zum Mozartjahr 1991, Baden–Wien, ed. Ingrid Fuchs, 241–54. Tutzing: Schneider, 1993. Fiske, Roger. ‘Music and Society’, in The Blackwell History of Music in Britain: The Eighteenth Century, ed. H. Diack Johnstone and Roger Fiske, 3–27. Oxford: Blackwell, 1990. Flaubert, Gustave. Madame Bovary: Provincial Manners, trans. Margaret Mauldon, with introduction by Malcolm Bowie. Oxford: Oxford University Press, 2004. Forschner, Hermann. Instrumentalmusik Joseph Haydns aus der Sicht Heinrich Christoph Kochs. Beitr¨age zur Musikforschung 13. Munich: Musikverlag Emil Katzbichler, 1984. Foucault, Michel. ‘What is an Author?’, in The Foucault Reader, ed. Paul Rabinow, 101–20. Harmondsworth: Penguin, 1986. Frisch, Walter. Brahms and the Principle of Developing Variation. Berkeley: University of California Press, 1984. Fux, Johann Joseph. Gradus ad Parnassum. Vienna: van Ghelen, 1725. Partial translation by Alfred Mann as Steps to Parnassus: The Study of Counterpoint. New York: Norton, 1943. Gabrielsson, Alf. ‘Once Again: The Theme from Mozart’s Piano Sonata in A Major (K. 331): A Comparison of Five Performances’, in Action and Perception in Rhythm and Music, ed. Alf Gabrielsson, 81–103. Stockholm: Royal Swedish Academy of Music, 1987. Georgiades, Thrasybulus. Kleine Schriften. Tutzing: Schneider, 1997. Gerber, Ernst Ludwig. Historisch-biographisches Lexikon der Tonk¨unstler. 2 vols. Leipzig: J. G. I. Breitkopf, 1790–2. Gerbner, George. ‘Toward a General Model of Communication’, Audio-Visual Communication Review 4 (1956), 171–99. Gibson, Walker. ‘Authors, Speakers, Readers and Mock Readers’, College English 11 (1950), 265–9. Reprinted in Reader-Response Criticism: From Formalism to Post-Structuralism, ed. Jane P. Tompkins, 1–6. Baltimore and London: Johns Hopkins University Press, 1980. Gjerdingen, Robert O. A Classic Turn of Phrase: Music and the Psychology of Convention. Philadelphia: University of Pennsylvania Press, 1988. Music in the Galant Style: Being an Essay on Various Schemata Characteristic of Eighteenth-Century Music for Courtly Chambers, Chapels, and Theaters, Including Tasteful Passages of Music Drawn from Most Excellent Chapel Masters in

Bibliography

the Employ of Noble and Noteworthy Personages, Said Music All Collected for the Reader’s Delectation on the World Wide Web. New York: Oxford University Press, 2007. Goldberg, Adele E. ‘Constructions: A New Theoretical Approach to Language’, Trends in Cognitive Science 7 (2003), 219–24. Goode, Luke. J¨urgen Habermas: Democracy and the Public Sphere. London: Pluto, 2005. Goody, Jack and Ian Watt. ‘The Consequences of Literacy’, in Literacy in Traditional Societies, ed. Jack Goody, 27–68. Cambridge: Cambridge University Press, 1968. ´ Grabocz, M´arta. ‘A. J. Greimas’s Narrative Grammar and the Analysis of Sonata Form’, Int´egral 12 (1998), 1–24. Grave, Floyd K. ‘Abb´e Vogler’s Revision of Pergolesi’s Stabat Mater’, Journal of the American Musicological Society 30 (1977), 43–71. ‘Common-Time Displacement in Mozart’, Journal of Musicology 3 (1984), 423–42. ‘Metrical Displacement and the Compound Measure in Eighteenth-Century Theory and Practice’, Theoria 1 (1985), 25–60. ‘Metrical Dissonance in Haydn’, Journal of Musicology 13 (1995), 168–202. Griesinger, Georg August. Biographische Notizen u¨ ber Joseph Haydn. Leipzig: Breitkopf und H¨artel, 1810. Ed. Karl-Heinz K¨ohler. Leipzig: Reclam, 1975. Haberkamp, Gertraut. Die Erstdrucke der Werke von Wolfgang Amadeus Mozart. 2 vols. Tutzing: Schneider, 1986. Habermas, J¨urgen. The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society, trans. Thomas Burger and Frederick Lawrence. Cambridge, MA: The MIT Press, 1989. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans. William Rehg. Cambridge, MA: The MIT Press, 1998. Hanning, Barbara R. ‘Conversation and Musical Style in the Late EighteenthCentury Parisian Salon’, Eighteenth-Century Studies 22 (1989), 512–28. Hanslick, Eduard. Vom Musikalisch-Sch¨onen. Leipzig, 1854. Translated by Martin Cooper as ‘On the Musically Beautiful’, in Music in European Thought 1851–1912, ed. Bojan Buji´c, 12–39. Cambridge: Cambridge University Press, 1988. Harvey, David. Review of Lerdahl and Jackendoff, A Generative Theory of Tonal Music, Music Analysis 4 (1985), 292–303. Harwood, Craig. ‘Subversive Strategies: Conventions and Manipulation of Gesture and Syntax in Mozart’, Ph.D. dissertation, Yale University, 2002. Hasty, Christopher. ‘Segmentation and Process in Post-Tonal Music’, Music Theory Spectrum 3 (1981), 54–73. Hatten, Robert. Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation. Bloomington: Indiana University Press, 1994. Interpreting Musical Gestures, Topics, and Tropes: Mozart, Beethoven, Schubert. Bloomington: Indiana University Press, 2004.

323

324

Bibliography

Haydn, Joseph. Gesammelte Briefe und Aufzeichnungen, ed. H. C. Robbins Landon and D´enes Bartha. Kassel: B¨arenreiter, 1965. The Collected Correspondence and London Notebooks, ed. H. C. Robbins Landon. London: Barrie and Rockliff, 1959. Head, Matthew. Orientalism, Masquerade and Mozart’s Turkish Music. Royal Musical Association Monographs 9. London: Royal Musical Association, 2000. Hepokoski, James and Warren Darcy. Elements of Sonata Theory: Norms, Types, and Deformations in the Late-Eighteenth-Century Sonata. New York: Oxford University Press, 2006. Hermes, Joke. Reading Women’s Magazines: An Analysis of Everyday Media Use. Oxford: Polity, 1996. Hilton, Wendy. Dance of Court and Theater: The French Noble Style, 1690–1725, ed. Caroline Gaynor. Princeton: Princeton Book Company, 1981. Hippel, Paul von. ‘Questioning a Melodic Archetype: Do Listeners Use Gap-Fill to Classify Melodies?’, Music Perception 18 (2000), 139–53. Hobson, Dorothy. Crossroads: The Drama of a Soap Opera. London: Methuen, 1982. Hoffmann, Ernst Theodor Amadeus. S¨amtliche Werke, ed. Hartmut Steinecke and Wulf Segebrecht. 6 vols. Frankfurt-am-Main: Deutscher Klassiker Verlag, 1993. Holman, Peter. ‘Eighteenth-Century English Music: Past, Present and Future’, in Music in Eighteenth-Century Britain, ed. David Wyn Jones, 1–13. Aldershot: Ashgate, 2000. Hoyt, Reed J. ‘Harmonic Function and the Motion of the Bassline’, Journal of Music Theory Pedagogy 4 (1990), 147–90. Hunter, J. Paul. Before Novels: The Cultural Contexts of Eighteenth-Century Fiction. New York and London: Norton, 1990. Huray, Peter le and James Day, eds. Music and Aesthetics in the Eighteenth and Early-Nineteenth Centuries. Cambridge: Cambridge University Press, 1981. Irving, Howard. ‘Haydn and Laurence Sterne: Similarities in Eighteenth-Century Literary and Musical Wit’, Current Musicology 40 (1985), 34–49. Iser, Wolfgang. The Implied Reader: Patterns of Communication in Prose Fiction from Bunyan to Beckett. Baltimore: Johns Hopkins University Press, 1974. Jackendoff, Ray. Review of Leonard Bernstein, The Unanswered Question, Language 53 (1977), 883–94. Jakobson, Roman. ‘Linguistics and Poetics: Closing Statement’, in Style in Language, ed. Thomas A. Sebeok, 350–77. Cambridge, MA: The MIT Press, 1960. ‘Language in Relation to Other Communication Systems’, in Jakobson, Selected Writings, vol. 2, 697–708. The Hague: Mouton, 1971. Jank´el´evitch, Vladimir. Music and the Ineffable, trans. Carolyn Abbate. Princeton: Princeton University Press, 2003. Jauss, Hans Robert. Toward an Aesthetic of Reception. Minneapolis: University of Minnesota Press, 1982. Jensen, Klaus Bruhn and Karl Erik Rosengren. ‘Five Traditions in Search of the Audience’, European Journal of Communication 5 (1990), 207–38.

Bibliography

Johnson, Douglas P., Alan Tyson and Robert Winter. The Beethoven Sketchbooks: History, Reconstruction, Inventory. Oxford: Oxford University Press, 1985. Johnson, Samuel. A Dictionary of the English Language in Which the Words Are Deduced from their Originals and Illustrated in their Different Significations by Examples from the Best Writers to Which are Prefixed a History of the Language and an English Grammar. London: Printed by W. Strahan, for J. and P. Knapton, T. and T. Longman, C. Hitch and L. Hawes, A. Millar, and R. and J. Dodsley, 1755. Jonas, Oswald. Einf¨uhrung in die Lehre Heinrich Schenkers: Das Wesen des musikalischen Kunstwerkes. 2nd edn. Vienna: Universal Edition, 1972. Translated by John Rothgeb as Introduction to the Theory of Heinrich Schenker: The Nature of the Musical Work of Art. New York: Longman, 1982. Kerman, Joseph. The Beethoven Quartets. New York: Alfred A. Knopf, 1967. Kinderman, William. Beethoven. Oxford: Oxford University Press, 1995. Kirnberger, Johann Philipp. Die Kunst des reinen Satzes in der Musik, vol. 2 part 1. Berlin and K¨onigsberg: Decker and Hartung, 1776. Translated by David Beach and Jurgen Thym along with vol. 1 of the same work as The Art of Strict Musical Composition. New Haven: Yale University Press, 1982. Klein, Rudolph. ‘Wo kann die Analyse von Haydns Symphonik ansetzen?’, ¨ Osterreichische Musikzeitschrift 37 (1982), 234–41. Koch, Heinrich Christoph. Versuch einer Anleitung zur Composition, vols. 2 and 3. Leipzig: Adam Friedrich B¨ohme, 1787 and 1793. Reprinted Hildesheim: Olms, 1969. Partly translated by Nancy Kovaleff Baker as Introductory Essay on Composition. New Haven: Yale University Press, 1983. Musikalisches Lexikon. Frankfurt-am-Main: August Hermann der J¨ungere, 1802. Reprinted Hildesheim: Olms, 1964. Kolberg, Oskar. Krakowskie, vol. 2. Cracow, 1873. Krebs, Harald. Fantasy Pieces: Metrical Dissonance in the Music of Robert Schumann. New York: Oxford University Press, 1999. Kress, Gunther R. ‘Against Arbitrariness: The Social Production of the Sign as a Foundational Issue in Critical Discourse Analysis’, Discourse and Society 4 (1993), 169–91. Kretzschmer, Andreas and Anton Wilhelm von Zuccalmaglio. Deutsche Volkslieder mit ihren Original-Weisen. 2 vols. Berlin: Vereins-Buchhandlung, 1838–40. Reprinted Hildesheim: Olms, 1969. Krumhansl, Carol L. ‘Topic in Music: An Empirical Study of Memorability, Openness, and Emotion in Mozart’s String Quintet in C major and Beethoven’s String Quartet in A minor’, Music Perception 16 (1998), 119–34. K¨uster, Konrad. Beethoven. Stuttgart: Deutsche Verlags-Anstalt, 1994. Lakoff, George. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago: Chicago University Press, 1987. Langer, Susanne K. Philosophy in a New Key: A Study in the Symbolism of Reason, Rite, and Art. Cambridge, MA: Harvard University Press, 1953.

325

326

Bibliography

Laskowski, Larry. ‘Voice Leading and Meter: An Unusual Mozart Autograph’, in Trends in Schenkerian Research, ed. Allen Cadwallader, 41–50. New York: Schirmer Books, 1990. Lenneberg, Hans. On the Publishing and Dissemination of Music, 1500–1850. Hillsdale, NJ: Pendragon, 2003. Lenz, Wilhelm von. Kritischer Katalog s¨ammtlicher Werke Ludwig van Beethovens mit Analysen derselben. 2nd edn. 4 vols. Hamburg: Hoffman und Campe, 1860. Lerdahl, Fred and Ray Jackendoff. ‘Generative Music Theory and its Relation to Psychology’, Journal of Music Theory 25 (1981), 45–90. A Generative Theory of Tonal Music. Cambridge, MA: The MIT Press, 1983. Lester, Joel. ‘Articulation of Tonal Structures as a Criterion for Analytic Choices’, Music Theory Spectrum 1 (1979), 67–79. Levy, Janet M. ‘Texture as a Sign in Classic and Early Romantic Music’, Journal of the American Musicological Society 35 (1982), 482–531. Beethoven’s Compositional Choices: The Two Versions of Opus 18, No. 1, First Movement. Philadelphia: University of Pennsylvania Press, 1982. ‘Gesture, Form and Syntax in Haydn’s Music’, in Haydn Studies: Proceedings of the International Haydn Conference, Washington, D.C., 1975, ed. Jens Peter Larsen, Howard Server and James Webster, 355–62. New York: Norton, 1981. Leydi, Roberto. I canti popolari italiani: 120 testi e musiche con la collaborazione di Sandra Mantovani e Cristina Pederiva. Verona: Mondadori, 1973. Lidov, David. Is Language a Music? Writings on Musical Form and Signification. Bloomington: Indiana University Press, 2005. Lippmann, Friedrich. ‘Mozart und der Vers’, in Colloquium ‘Mozart und Italien’ (Rom 1974), ed. Friedrich Lippmann, 107–37. Analecta musicologica 18. Cologne: Arno Volk, 1978. Little, Meredith Ellis. ‘The Contribution of Dance Steps to Musical Analysis and Performance: “La Bourgogne”’, Journal of the American Musicological Society 28 (1975), 112–24. ‘Siciliana’, in The New Grove Dictionary of Music and Musicians. 2nd edn, ed. Stanley Sadie and John Tyrell, vol. 23, 350–2. London: Macmillan, 2002. Little, Meredith and Natalie Jenne. Dance and the Music of J. S. Bach. Revised edn. Bloomington: Indiana University Press, 2001. Livingstone, Sonia M. ‘The Challenge of Changing Audiences: Or, what is the Audience Researcher to Do in the Age of the Internet?’, European Journal of Communication 19 (2004), 75–86. Lockwood, Lewis. ‘Recent Writings on Beethoven’s Late Quartets’, Beethoven Forum 9 (2002), 84–97. Love, Harold. ‘How Music Created a Public’, Criticism 46 (2004), 257–71. Mainka, J¨urgen. ‘Haydns Streichquartette: “Man h¨ort vier vern¨unftige Leute sich untereinander unterhalten”’, Musik und Gesellschaft 32 (1982), 146–50. Makinen, Merja. Feminist Popular Fiction. London: Palgrave, 2001.

Bibliography

Mandler, Jean. ‘Categorical and Schematic Organization in Memory’, in Memory Organization and Structure, ed. Richard Puff, 259–99. New York: Academic Press, 1979. Marpurg, Friedrich Wilhelm. Kritische Briefe u¨ ber die Tonkunst. 2 vols. Berlin: Friedrich Wilhelm Birnstiel, 1759–64. Reprinted Hildesheim: Olms, 1974. Marsden, Jean. ‘British Theory and Criticism: 2. Late Eighteenth Century’, in Johns Hopkins Guide to Literary Theory and Criticism, ed. Michael Groden and Martin Kreiswirth, 108–12. Baltimore and London: Johns Hopkins University Press, 1997. Marshall, Robert. The Compositional Process of J. S. Bach: A Study of the Autograph Scores of the Vocal Works. 2 vols. Princeton: Princeton University Press, 1972. Marx, Adolf Bernhard. Die Lehre von der musikalischen Komposition, praktischtheoretisch, vol. 2. Leipzig: Breitkopf und H¨artel, 1838. Mattheson, Johann. Der vollkommene Capellmeister. Hamburg: Herold, 1739. Translated by Ernest C. Harriss. Ann Arbor: UMI Research Press, 1981. ¨ Maurer Zenck, Claudia. Vom Takt: Uberlegungen zur Theorie und kompositorischen Praxis im ausgehenden 18. und beginnenden 19. Jahrhundert. Vienna: B¨ohlau, 2001. McClary, Susan. Review of Monelle, The Sense of Music: Semiotic Essays, Notes 58 (2001), 326–8. Meyer, Leonard B. Emotion and Meaning in Music. Chicago: University of Chicago Press, 1956. Music, The Arts, and Ideas. Chicago: University of Chicago Press, 1967. Explaining Music: Essays and Explorations. Berkeley: University of California Press, 1973. Style and Music. Philadelphia: University of Pennsylvania Press, 1988. ‘A Pride of Prejudices; Or, Delight in Diversity’, Music Theory Spectrum 13 (1991), 241–51. Reprinted in Meyer, The Spheres of Music, 262–78. ‘Exploiting Limits: Creation, Archetypes and Change’, Daedalus 109 (1980), 177– 205. Reprinted in Meyer, The Spheres of Music, 189–225. The Spheres of Music: A Gathering of Essays. Chicago: University of Chicago Press, 2000. ¨ Michaelis, Christian Friedrich. ‘Uber das Humoristische oder Launige in der musikalischen Komposition’, Allgemeine musikalische Zeitung 9 (12 August 1807), cols. 725–9. Miel, Dorothy, Raymond MacDonald and David Hargreaves, eds. Musical Commnication. Oxford: Oxford University Press, 2005. ˆ Momigny, J´erome-Joseph de. La seule vraie th´eorie de la musique. Paris: Printed for the author, 1821. Monelle, Raymond. The Sense of Music: Semiotic Essays. Princeton: Princeton University Press, 2000.

327

328

Bibliography

The Musical Topic: Hunt, Military and Pastoral. Bloomington: Indiana University Press, 2006. Moore, Julia. ‘Mozart in the Market-Place’, Journal of the Royal Musical Association 114 (1989), 18–42. Morgan, Robert P. ‘The Theory and Analysis of Tonal Rhythm’, The Musical Quarterly 64 (1978), 435–73. ‘Circular Form in the Tristan Prelude’, Journal of the American Musicological Society 53 (2000), 69–104. Morley, Dave. The Nationwide Audience. London: British Film Institute, 1980. [Mozart, Wolfgang Amadeus.] Mozart: Briefe und Aufzeichnungen, ed. Wilhelm A. Bauer and Otto Erich Deutsch. 7 vols. Kassel: B¨arenreiter, 1962–75. The Letters of Mozart and his Family, ed. and trans. Emily Anderson. New York: Norton, 1985. Nagel, Wilibald. Beethoven und seine Klaviersonaten. 2 vols. Langensalza: Beyer, 1903–5. N¨ageli, Hans-Georg. Vorlesungen u¨ ber Musik mit Ber¨ucksichtigung der Dilettanten. Stuttgart and T¨ubingen: Cotta, 1826. Narmour, Eugene. Beyond Schenkerism: The Need for Alternatives in Music Analysis. Chicago: University of Chicago Press, 1977. ‘Some Major Theoretical Problems Concerning the Concept of Hierarchy in the Analysis of Tonal Music’, Music Perception 1 (1983–4), 129–99. Nattiez, Jean-Jacques. ‘Var`ese’s “Density 21.5”: A Study in Semiological Method’, trans. Anna Barry, Music Analysis 1 (1982), 243–340. ‘The Concepts of Plot and Seriation Process in Music Analysis’, trans. Catherine Dale, Music Analysis 4 (1985), 107–18. Music and Discourse: Toward a Semiology of Music, trans. Carolyn Abbate. Princeton: Princeton University Press, 1990. Neumeyer, David. ‘The Three-Part Ursatz’, In Theory Only 10 (1987), 3–29. Nora, Tia de. Beethoven and the Construction of Genius: Musical Politics in Vienna, 1792–1803. Berkeley: University of California Press, 1995. Nowotny, Rudolf. ‘Das Duett Nr. 7 aus der Zauberfl¨ote: Periode und Takt’, MozartJahrbuch 1996, 85–126. Pacun, David. ‘Scanning Bass Patterns: A Middleground Path to Analysis’, Journal of Music Theory Pedagogy 17 (2003), 59–77. Parker, Mara. The String Quartet, 1750–1797: Four Types of Musical Conversation. Aldershot: Ashgate, 2002. Peel, John and Wayne Slawson. Review of Lerdahl and Jackendoff, A Generative Theory of Tonal Music, Journal of Music Theory 28 (1984), 282–7. Perl, Benjamin. ‘Wo soll man bei 6/8 die Taktstriche ziehen? Weitere Gedanken zum Duett aus der Zauberfl¨ote’, Mozart-Jahrbuch 1998, 85–101. Peters, John Durham. Speaking into the Air: A History of the Idea of Communication. Chicago: University of Chicago Press, 1999.

Bibliography

Peters, Jeffrey N. ‘Ideology, Culture and the Threat of Allegory in Jean Chapelain’s Theory of La Vraisemblance’, Romanic Review 89 (1998), 491– 505. Piston, Walter. Harmony. 4th edn, revised and expanded by Mark DeVoto. New York: Norton, 1978. Poster, Mark. The Second Media Age. Cambridge: Polity, 1995. Quantz, Johann Joachim. On Playing the Flute, trans. Edward R. Reilly. New York: Schirmer Books, 1966. Radway, Janice A. Reading the Romance: Women, Patriarchy and Popular Literature. Chapel Hill and London: University of North Carolina Press, 1984. Rameau, Jean-Philippe. G´en´eration harmonique ou trait´e de musique th´eorique et pratique. Paris: Prault fils, 1737. Reprinted New York: Broude Brothers, 1966. Treatise on Harmony, trans. Philip Gossett. New York: Dover Publications, 1971. Rameau, Pierre. The Dancing Master, trans. Cyril W. Beaumont. New York: Dance Horizons, 1970. Ratner, Leonard G. Classic Music: Expression, Form, and Style. New York: Schirmer Books, 1980. The Beethoven String Quartets: Compositional Strategies and Rhetoric. Stanford, CA: Stanford Bookstore, 1995. Reddy, Michael J. ‘The Conduit Metaphor: A Case of Frame Conflict in Our Language About Language’, in Metaphor and Thought, ed. Andrew Ortony, 2nd edn, 164– 201. Cambridge: Cambridge University Press, 1993. Richards, Annette. ‘Listening for Kenner and Liebhaber ’, review of Riley, Musical Listening in the German Enlightenment: Attention, Wonder and Astonishment, Early Music 33 (2005), 332–4. Ridgewell, Rupert Magnus. ‘Mozart and the Artaria Publishing House: Studies in the Inventory Ledgers, 1784–1793’. Ph.D. dissertation, Royal Holloway, University of London, 1999. ‘Economic Aspects: The Artaria Case’, in Music Publishing in Europe, 1600– 1900: Concept and Issues, Bibliography, ed. Rudolf Rasch. Berlin: Berliner Wissenschafts-Verlag, 2005. ¨ Riemann, Hugo. L. van Beethovens s¨amtliche Klavier-Solosonaten: Asthetische und formal-technische Analyse mit historischen Notizen. 2nd edn. 3 vols. Berlin: Max Hesse, 1919. Musikalische Dynamik und Agogik: Lehrbuch der musikalische Phrasirung auf Grund einer Revision der Lehre von der musikalischen Metrik und Rhythmik. Hamburg: Rahter, 1884. Katechismus des Klavierspiels. Leipzig: Max Hesse, 1888. Translated anonymously as Catechism of pianoforte playing. London: Augener, 1892. Elementar-Schulbuch der Harmonielehre. Berlin: Max Hesse, 1906. Vademecum der Phrasierung. Berlin: Max Hesse, 1912. L. van Beethovens s¨amtliche Klavier-Solosonaten. vol. 2. Berlin: Max Hesse, 1919.

329

330

Bibliography

Riepel, Joseph. Anfangsgr¨unde zur musicalischen Setzkunst, vol. 1: De Rhythmopoeia oder von der Tactordnung. Regensburg and Vienna: Emerich Felix Bader, 1752; vol. 4: Erl¨auterung der betr¨uglichen Tonordnung. Augsburg: Johann Jacob Lotter, 1765. Both reprinted in Riepel, S¨amtliche Schriften, vol. 1, 17–99 and 333–439. ‘Des Harmonischen Sylbenmaßes dritter Teil’, manuscript transcribed in Riepel, S¨amtliche Schriften, vol. 2, 119–75. S¨amtliche Schriften zur Musiktheorie, ed. Thomas Emmerig. 2 vols. Vienna: B¨ohlau, 1996. Riley, Matthew. ‘Johann Nikolaus Forkel on the listening practices of “Kenner” and “Liebhaber”’, Music and Letters 84 (2003), 414–33. Musical Listening in the German Enlightenment: Attention, Wonder and Astonishment. Aldershot: Ashgate Press, 2004. Rimmon-Kenan, Shlomith. Narrative Fiction: Contemporary Poetics. London: Methuen, 1983. ¨ [Rochlitz, Friedrich.] ‘Uber den zweckm¨assigen Gebrauch der Mittel der Tonkunst’, Allgemeine musikalische Zeitung 8 (2 October 1805–15 January 1806), cols. 3–10, 49–59, 193–201 and 241–9. Rochlitz, Friedrich. ‘Die Fuge: Zun¨achst an Dilettanten und Layen’, Allgemeine musikalische Zeitung 15 (12–19 May 1813), cols. 309–17 and 325–33. Rosen, Charles. The Classical Style: Haydn, Mozart, Beethoven. New York: Norton, 1971. Ross, Angus. ‘Introduction’, in Daniel Defoe, Robinson Crusoe, 7–21. Harmondsworth: Penguin, 1965. ‘Introduction’, in Selections from The Tatler and The Spectator, ed. Angus Ross, 21–55. Harmondsworth: Penguin, 1982. Rothgeb, John. ‘Design as a Key to Structure in Tonal Music’, in Readings in Schenker Analysis and Other Approaches, ed. Maury Yeston, 72–93. New Haven: Yale University Press, 1977. Rothstein, William. Phrase Rhythm in Tonal Music. New York: Schirmer Books, 1989. Rudowski, Victor Anthony. ‘The Theory of Signs in the Eighteenth Century’, Journal of the History of Ideas 35 (1974), 683–90. Ruwet, Nicholas. ‘Methods of Analysis in Musicology’, trans. Mark Everist, Music Analysis 6 (1987), 11–36. Sadie, Stanley. ‘Music in the Home ii’, in The Blackwell History of Music in Britain: The Eighteenth Century, ed. H. Diack Johnstone and Roger Fiske, 313–54. Oxford: Blackwell, 1990. Saint-Sa¨ens, Camille. Outspoken Essays on Music, trans. Fred Rothwell. London: Kegan Paul, 1922. Sapir, Edward. Language, an Introduction to the Study of Speech. New York: Harcourt, Brace and Company, 1921. Saunders, David. Authorship and Copyright. London: Routledge, 1992. Saussure, Ferdinand de. Course in General Linguistics, trans. Wade Baskin, ed. Charles Bally and Albert Sechehaye. New York: McGraw-Hill, 1959.

Bibliography

Schachter, Carl. ‘Rhythm and Linear Analysis: Aspects of Meter’, in Music Forum 6, ed. Felix Salzer and Carl Schachter, 1–60. New York: Columbia University Press, 1987. Schenker, Heinrich. Neue musikalische Theorien und Phantasien, vol. 2: Kontrapunkt, part 1: Cantus firmus und zweistimmiger Satz. Vienna: Universal Edition, 1910. Trans. John Rothgeb and J¨urgen Thym as Counterpoint, Book 1: Cantus Firmus and Two-Voice Counterpoint. New York: Schirmer Books, 1987. Das Meisterwerk in der Musik, vol. 1. Munich and Vienna: Drei Masken Verlag, 1925. Trans. Ian Bent et al. as The Masterwork in Music, ed. William Drabkin, vol. 1. Cambridge Studies in Music Theory and Analysis 5. Cambridge: Cambridge University Press, 1994. Neue musikalische Theorien und Phantasien, vol. 3: Der freie Satz. Vienna: Universal Edition, 1935. Trans. Ernest Oster as Free Composition. New York: Longman, 1979. Schmalfeldt, Janet. ‘Form as a Process of Becoming: The Beethoven–Hegelian Tradition and the “Tempest” Sonata’, Beethoven Forum 4 (1995), 37–71. Schoenberg, Arnold. Style and Idea, ed. Leonard Stein. London: Faber and Faber, 1975. Fundamentals of Musical Composition, ed. Gerald Strang and Leonard Stein. London: Faber and Faber, 1967. Schueller, Herbert M. ‘The Use and Decorum of Music as Described in British Literature, 1700–1780’, Journal of the History of Ideas 13 (1952), 73–93. Sebeok, Thomas A. A Sign is Just a Sign. Bloomington: Indiana University Press, 1991. ¨ Seidel, Wilhelm. Uber Rhythmustheorien der Neuzeit. Neue Heidelberger Studien zur Musikwissenschaft 7. Bern: Francke, 1975. Seiter, Ellen et al., eds. Remote Control: Television, Audiences and Cultural Power. London: Routledge, 1989. Shannon, Claude E. and Warren Weaver. The Mathematical Theory of Communication. Urbana: University of Illinois Press, 1949. Shapiro, Joel and Kenneth Drake. ‘3 Klaviersonaten G-Dur, d-Moll “Sturmsonate” und Es-Dur op. 31’, in Beethoven: Interpretationen seiner Werke, ed. Albrecht Riethm¨uller, Carl Dahlhaus and Alexander Ringer, 2nd edn, vol. 1, 250–63. Laaber: Laaber-Verlag, 1996. Sheldon, David A. Marpurg’s Thoroughbass and Composition Handbook: A Narrative Translation and Critical Study. Stuyvesant, NY: Pendragon, 1989. Sisman, Elaine. Haydn and the Classical Variation. Cambridge, MA: Harvard University Press, 1993. Mozart: The ‘Jupiter’ Symphony No. 41 in C Major, K. 551. Cambridge: Cambridge University Press, 1993. ‘Haydn’s Career and the Idea of the Multiple Audience’, in The Cambridge Companion to Haydn, ed. Caryl Clark, 3–16. Cambridge: Cambridge University Press, 2005.

331

332

Bibliography

‘Observations on the First Phase of Mozart’s “Haydn” Quartets’, in Words About Mozart: Essays in Honour of Stanley Sadie, ed. Dorothea Link and Judith Nagley, 33–58. Woodbridge: Boydell, 2005. Solomon, Maynard. ‘The Rochlitz Anecdotes’, in Mozart Studies, ed. Cliff Eisen, 1–59. Oxford: Oxford University Press, 1991. Mozart: A Life. London: Pimlico, 1996. [Spazier, Johann Gottlieb.] Review of ‘Lieferung I’ of the R´epertoire des clavecinistes, Zeitung f¨ur die elegante Welt 3 (1803), col. 612. Spitzer, Michael. Metaphor and Musical Thought. Chicago: University of Chicago Press, 2004. Music as Philosophy: Adorno and Beethoven’s Late Style. Bloomington: Indiana University Press, 2006. ‘Tovey’s Evolutionary Metaphors’, Music Analysis 24 (2006), 437–70. ‘The New Formal Theory’, Beethoven Forum, forthcoming. Staehelin, Martin. Hans Georg N¨ageli und Ludwig van Beethoven. Zurich: Hug, 1982. Sulzer, Johann Georg. Allgemeine Theorie der sch¨onen K¨unste. 4th edn, vol. 1. Leipzig: In der Weidmannschen Buchhandlung, 1792. Reprinted Hildesheim: Olms, 1967. Sutcliffe, W. Dean. ‘Haydn, Mozart and their Contemporaries’, in The Cambridge Companion to the String Quartet, ed. Robin Stowell, 185–209. Cambridge: Cambridge University Press, 2003. Temperley, David. ‘End-Accented Phrases: An Analytical Exploration’, Journal of Music Theory 47 (2003), 125–54. Thomas, Downing A. Music and the Origins of Language: Theories from the French Enlightenment. Cambridge: Cambridge University Press, 1995. Thompson, John B. ‘Social Theory and the Media’, in Communication Theory Today, ed. David Crowley and David Mitchell, 27–49. Cambridge: Polity, 1992. Todorov, Tzvetan. The Poetics of Prose. Ithaca: Cornell University Press, 1977. Tolley, Thomas. Painting the Cannon’s Roar: Music, the Visual Arts and the Rise of an Attentive Public in the Age of Haydn. Aldershot: Ashgate, 2001. Tomasello, Michael. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press, 1999. Triest, Johann Karl Friedrich. ‘Bemerkungen u¨ ber die Ausbildung der Tonkunst in Deutschland im achtzehnten Jahrhundert’, Allgemeine musikalische Zeitung 3 (1 January–25 March 1801), cols. 225–35, 241–9, 257–64, 273–86, 297–308, 321– 31, 369–79, 389–401, 405–10, 421–32 and 437–45. English translation by Susan Gillespie as ‘Remarks on the Development of the Art of Music in Germany in the Eighteenth Century’, in Haydn and his World, ed. Elaine Sisman, 321–94. Princeton: Princeton University Press, 1997. T¨urk, Daniel Gottlob. Klavierschule oder Anweisung zum Klavierspielen f¨ur Lehrer und Lernende. Leipzig: Schwickert und Hemmerde, 1789. Reprinted Kassel: B¨arenreiter, 1997. Trans. Raymond H. Haggh as School of Clavier Playing. Lincoln: University of Nebraska Press, 1982.

Bibliography

Tyson, Alan. Review of Mozart’s Piano Sonatas, ed. Walter Lampe (Munich: Henle, 1966), Musical Times 107 (1966), 888. Mozart: Studies of the Autograph Scores. Cambridge, MA: Harvard University Press, 1987. Veidl, Theodor. Der musikalische Humor bei Beethoven. Leipzig: Breitkopf und H¨artel, 1929. Vogler, Georg Joseph. Tonwissenschaft und Tonsetzkunst. Mannheim, 1776. Betrachtungen der Mannheimer Tonschule. 3 vols. Mannheim, 1778–81. Watt, Ian P. The Rise of the Novel: Studies in Defoe, Richardson and Fielding. Harmondsworth: Penguin, 1963. Myths of Modern Individualism: Faust, Don Quixote, Don Juan, Robinson Crusoe. Cambridge: Cambridge University Press, 1997. ¨ Weber, [Friedrich August]. ‘Uber komische Charakteristik und Karrikatur in praktischen Musikwerken’, Allgemeine musikalische Zeitung 3 (26 November 1800– 13 December 1800), cols. 138–43 and 157–62. Weber, Gottfried. Versuch einer geordneten Theorie der Tonsetzkunst. 2 vols. Mainz: Schott, 1817–18. Weber, William. The Rise of Musical Classics in Eighteenth-Century England: A Study in Canon, Ritual and Ideology. Oxford: Clarendon Press, 1996. Webster, James. ‘Haydn’s Symphonies Between Sturm und Drang and “Classical Style”: Art and Entertainment’, in Haydn Studies, ed. W. Dean Sutcliffe, 218– 45. Cambridge: Cambridge University Press, 1998. ‘The Analysis of Mozart’s Arias’, in Mozart Studies, ed. Cliff Eisen, 101–99. Cambridge: Cambridge University Press, 1991. Wegeler, Franz Gerhard and Ferdinand Ries. Biographische Notizen u¨ ber Ludwig van Beethoven. Koblenz: K. B¨adeker, 1838. Ed. Alfred Kalischer. Berlin and Leipzig: Schuster und Loeffler, 1906. Trans. Frederick Noonan as Remembering Beethoven: The Biographical Notes of Franz Wegeler and Ferdinand Ries. London: Andr´e Deutsch, 1988. Wellbery, David E. Lessing’s Laocoon: Semiotics and Aesthetics in the Age of Reason. Cambridge: Cambridge University Press, 1984. Westley, Bruce and Malcolm MacLean. ‘A Conceptual Model for Communication Research’, Journalism Quarterly 34 (1957), 31–8. Westphal, Rudoph. Allgemeine Theorie der musikalischen Rhythmik seit J. S. Bach. Leipzig: Breitkopf und H¨artel, 1880. Wheelock, Gretchen A. Haydn’s Ingenious Jesting with Art: Contexts of Musical Wit and Humor. New York: Schirmer Books, 1992. ‘Engaging Strategies in Haydn’s Opus 33 String Quartets’, Eighteenth Century Studies 25 (1991), 1–30. ‘The “Rhetorical Pause” and Metaphors of Conversation in Haydn’s Quartets’, in Haydn und das Streichquartett, Eisenst¨adter Haydn-Berichte 2, ed. Georg Feder and Walter Reicher, 67–85. Tutzing: Schneider, 2003.

333

334

Bibliography

Wiehmayer, Theodor. ‘Der “Große Takt” und die Analyse der c-moll-Symphonie von Beethoven’, Zeitschrift f¨ur Musikwissenschaft 4 (1922), 417–27. Musikalische Rhythmik und Metrik. Magdeburg: Heinrichshofen, 1917. Williams, Gordon. ‘Newsbooks and Popular Narrative during the Middle of the Seventeenth Century’, in Narrative: From Malory to Motion Pictures, ed. Jeremy Hawthorn, 57–70. London: Edward Arnold, 1985. Willner, Channan. ‘Durational Pacing in Handel’s Instrumental Works: The Nature of Expansion in the Music of the High Baroque’. Ph.D. dissertation, City University of New York, 2004. Wilson, Thomas. A Description of the Correct Method of Waltzing, the Truly Fashionable Species of Dancing. London: Printed for the author, published by Sherwood, Neely and Jones, 1816. Wintle, Christopher. ‘Corelli’s Tonal Models: The Trio Sonata Op. 3, No. 1’, in Nuovissimi Studi Corelliani: Atti del Terzo Congresso Internazionale, Fusignano, 1980, ed. Sergio Durante and Pierluigi Petrobelli, 29–69. Florence: Olschki, 1982. Yaraman, Sevin. Revolving Embrace: The Waltz as Sex, Steps, and Sound. Hillsdale, NJ: Pendragon, 2002. Yeston, Maury. The Stratification of Musical Rhythm. New Haven: Yale University Press, 1976. Youngren, William H. C. P. E. Bach and the Rebirth of the Strophic Song. Lanham, MD: Scarecrow, 2003. Zaslaw, Neal. ‘Mozart as a Working Stiff’, in On Mozart, ed. James M. Morris, 102–12. Washington, DC: Woodrow Wilson Center, 1994. Zbikowski, Lawrence M. Conceptualizing Music: Cognitive Structure, Theory, and Analysis. New York: Oxford University Press, 2002. Zsako, Julius. ‘The String Quartets of Ignace J. Pleyel’. Ph.D. dissertation, Columbia University, 1975.

Index of concepts

Absatz; see ending formulas accento comune 122–3, 142, 145, 151, 157n.39 addresser vs. addressee, see sender vs. receiver affect, see expression Affektenlehre 106, 111n.42 alla breve style 230, 237, 245–6, 248–9, 250 alla zoppa 248, 249 analysis; see music analysis analytical paradigm 6, 186n.43, 255, 256, 260, 271, 276n.13, 278n.26 ars combinatoria 225 artificiality; see conventionality artificial vs. natural 35 association 105, 292, 311 audience 34 heterogeneous 1–2, 5, 34, 39, 42–3, 312 historical of the eighteenth century 13, 14, 23, 316 of the nineteenth century 46–7 see also Kenner and Liebhaber auteur 15 author 15, 20 author–publisher relationship 17; see also composer–publisher relationship ‘implied’ 13–16 ‘real’ 16 authorial intent 14, 16, 59 authorship 23–4, 28 barline, function of; see metre, concepts of barring 113, 158n.44 correct vs. incorrect 118–22 German 115–17, 121, 135 Italian 116, 121, 122, 134 neutral 116–17, 135 see also metric notation; re-barring basse fondamental 162–3, 182n.5, 182n.8, 182n.9 bass models basic 164 dominant-modulating 164 periodic 164–5, 183n.16 ˆ 2-gapped 165

basso continuo 163, 182n.9 ‘Beethoven–Hegelian tradition’ 227n.14 beginning–middle–end paradigm 192, 227n.8 bourr´ee 7, 86, 237, 245, 248, 285, 292–305, 315 choreography of 293–5 music for 292–3 brilliant style 248, 249, 315 bungling 5, 56, 57 cadence and topics 266, 267–8, 270, 272 as defined by Caplin 197–8, 225, 229n.30 as humorous means 61 strong- vs. weak-beat 112, 113, 123, 147 and hierarchy of cadential closures 132, 157n.31 see also caesura; ending formulas; formal functions, cadential; liquidation, cadential; syntactic processes cadence-placement rule 83–4, 107–8n.4, 113, 122, 156n.25 exceptions from 113, 123–5, 154n.5, 157n.31 cadential accent 260, 261–4; see also phrases, beginning- vs. end-accented; cadence, strong- vs. weak-beat cadenza 238, 248 caesura 84–6, 90, 93, 110n.34; see also cadence; ending formulas caesura-placement rule; see cadence-placement rule caricature 56 channel 15, 17 chorale 106 church style 106; see also alla breve style circularity vs. linearity 250, 253n.19 code 272, 310, 311; see also communication, system of coding; see decoding; encoding cognition 185n.34, 190, 223–4 cognitive studies 4, 7, 224, 274, 285, 289, 316 comical; see humorous

336

Index of concepts

commerce 16, 24, 28, 29, 35, 39–40, 45, 50n.21, 50n.22 art and 5, 35–6, 40, 41, 44 see also market communication 13, 14 concept of 14–15 linguistic 4, 284 musical 4, 20, 269, 272, 284, 301, 305, 306, 311 study of; see communication sciences system of 285, 310, 316 communication models 15–17 communication sciences 4, 15–17, 316n.1 communicative strategies 3, 5, 6, 14, 16, 19, 28–9, 177–8 composer composer–publisher relationship 5, 47, 53–5; see also author–publisher relationship fictional 59, 62, 64–5, 67, 70–1, 77n.33 ‘concept-based’ vs. ‘data-based’; see ‘top-down’ vs. ‘bottom-up’ processes concerts; see musical life; performances content; see meaning context 13, 14, 16 social 2; see also ‘public sphere’ contredanse 86, 105, 106 convention 1–2, 4, 14, 105, 113, 189, 224, 232, 311 departures from 19–20, 58–9 as means of the humorous 13, 27, 55–6, 58, 73 see also national conventions; schema; topics conventionality 19, 192–3 and particularity 191, 194; see also artificial vs. natural; signs, natural vs. arbitrary conversation 2, 8–9n.2, 18, 19 counterpoint 5, 36, 44 as means of the humorous 58, 67 see also fugue, fugal technique counting of measures 86, 87, 98, 100, 108n.10 criticism 5, 22, 23, 47, 312 crossing of hands 69 dance 7, 285, 291, 292, 306, 315 cadence placement in 113, 125 in composition instruction 291–2 origins of music, poetry and 284 decoding 15, 25, 310 design 250, 253n.18 ‘discourse, rational’ (Habermas) 18–19 about music 20–1, 22–3, 28 double measures 61, 62, 77n.32, 78n.34, 95–6, 104; see also metric notation

dynamic processes 7, 285, 286, 289, 290–1, 306 definition of 307n.8 syntactic processes and 290 Einschnitt; see ending formulas; motive elongated upbeat 98, 99, 109–10n.28 emotion 4, 7, 21, 29, 283, 284, 285, 305; see also expression encoding 15, 25 ending formulas 83, 101, 107n.3, 191, 193, 194 decorations of 86, 100 improper harmonization of 97, 100–1, 102, 110n.32 see also cadences; caesuras; punctuation entertainment 1, 38 Entwicklung and Fortspinnung 191, 200, 227n.6 expectations 2, 3, 36–7, 38, 94, 178, 180, 313 generic 24–5, 27 ‘horizon of’ 16 syntactic 4 see also implication-realization; unexpected expression 1, 6–7, 29, 105, 106, 249, 265–70, 271–2, 291, 314 see also emotion; topics fanfare 219, 248 fantasia 248, 315 feminine cadence 260, 262, 263 feminine ending 87, 93; see also overhang foreground vs. background 6, 247 form 2, 3 and metre 86, 105, 107 see also sonata form formal functions cadential 192, 195, 199 continuation 183n.18, 192, 195, 199 post-cadential 228n.28, 229n.30 presentation 192, 195, 199 formal schemata 2, 3 fugue, fugal technique 43, 46–7, 86; see also counterpoint furioso 248 galant style 237, 240, 248–9, 250 gavotte 113, 281n.50 genre 24–5, 106 gestures 247, 251, 283–4, 286, 296; see also dynamic processes; expression; topics gigue 280n.43 grammar construction grammar 289–90 musical 3–4, 5–6, 190 of dance 292, 305 of music and language 284–5, 286, 306 see also syntax

Index of concepts

harmonic generation 284, 307n.2 harmonic progressions cadential 183n.12, 197 prolongational 183n.12 sequential 181, 183n.12 harmonic relationships vs. melodic relationships 162 ‘head/end ambiguity’ 259 hemiola 83, 107n.1 hermeneutics 226, 229n.43, 311, 317n.4 humorous 5, 13, 19, 20, 23, 28, 55–9, 92, 312 distinguishing between humorous and serious 59, 71, 73–4 low humour 55 reversals of the serious into the humorous 57, 106; see also parody serious consequences of humorous techniques 59 see also bungling; imitation; representation, musical (Malerei); parody; wit hunting 139 hypermetre 112, 114, 154n.3, 208 hypotactic vs. paratactic 199, 228n.24 imbroglio 83, 107n.1 imitation, mimesis (Mimik) 21, 23, 56, 57, 265 implication–realization 4, 6, 161, 165, 173, 177–8, 185n.33, 194 continuation 194 deferred realization 177–8, 185n.34, 185n.38 gap-fill 162, 164, 173, 177, 178, 182n.7, 194 unidirectional fill 177 immediate realization 177, 185n.34, 185n.38 Kadenz; see ending formulas Kenner and Liebhaber 2, 5, 6, 22, 27–8, 36–7, 39, 40–1, 42–3, 83, 313 krakowiak (Polish dance) 65 langage d’action 284, 305 Laune 19, 29, 56, 57 learned style 43; see also alla breve style; church style; counterpoint; fugue linguistics 4 liquidation 191, 192–3, 194 cadential 190, 196–8, 199, 209 structural 196–7 thematic 190, 196–7, 198, 199 listener historical of the eighteenth century 3 ideal 25, 39, 40–1, 180 see also audience literary theory 4; see also reader-response theory; reception theory local traditions; see national conventions

Mannheim sky-rocket 202 march 106, 129, 248–9, 301, 315 exalted 86, 105 market 2, 5, 17, 18–19, 21, 35, 39–40, 41, 44 ‘mean’ 36, 42, 44 meaning 2, 14, 15, 235, 285, 310–11, 317n.5 and music analysis 244, 246, 251 language and the study of musical 311, 312 see also signification message 15 messanza 248, 249 metaphor 222–6 ‘conduit metaphor’ (Reddy) 290 of music as language 4, 7, 8n.1, 197, 232 see also sonata form, metaphoric model of metaphor theory 6, 189–90, 224, 227n.15 metre 4, 5–6, 105, 106, 110n.30, 110n.36, 278n.30, 312 compound 113–14, 117, 123, 134–45 Franco-Italian 144, 146–8, 150 French 134–5, 136, 144, 151, 158n.49 German 127, 135, 136, 146–8, 150, 151, 158n.49 Italian 135, 136, 144, 151 compound and simple 83–6, 90, 121–2, 136, 261 changes between 84–6, 87–94 concepts of 118, 121, 145 see also barring metrical dissonance 107n.1 metrical types, see barring metric displacement 114, 141 metric modulation 86, 89, 90 metric notation 6, 93, 95; see also barring; double measures Metrum 89, 100, 109n.17, 109n.18 minuet (topic) 125 motive metric notation of 118–20, 134, 135 role in the phrase structure 191, 193 motivic vs. schematic 6, 194, 224 multimedia 291, 295, 296, 297–8 musette 200, 205, 212–13, 219, 228n.26, 228n.28, 237 musical life 1–2, 21–2; see also ‘public sphere’ music analysis 40, 41, 230, 244, 246–7, 252n.12, 311–16 charge of anachronism in 312, 313, 314, 315 historical vs. theoretical perspective in 270–3 historicist analysis 3, 4, 312–13, 316 paradigmatic analysis 6, 230, 235, 244, 246, 251, 314 rhythmic-metric analysis 257–9, 264

337

338

Index of concepts

music analysis (cont.) rhythm–pitch correlation in 257, 261–3, 264, 267, 280n.40 Schenkerian analysis 160–1, 229n.31, 256–7, 258–9, 262, 274, 277n.17, 277n.18, 277n.21 topical analysis 230, 251, 253n.17, 265, 315 music and language 306–8, 309n.29, 310 dissimilarities between 7, 272, 284, 285–6, 289, 290–1, 307n.4 origins of 284 similarities between 283, 284, 285 see also metaphor, of music as language; music as ‘international language’; settings of texts music and literature 3, 5 music as ‘international language’ 23, 27 narrative texts 3, 13, 15–16, 24; see also settings of text narrative theory 15–16 national conventions 2, 6, 27, 113, 148–51; see also metre, compound; barring ‘new musicology’ 246 ‘noise’ 15, 17 norms 2; see also convention; rules, of the genre; public opinion (doxa) oration; see rhetoric overhang 87, 93, 94, 101 incorrect 88, 95 paradigmatic vs. syntagmatic 235, 245, 252n.13 parenthesis (Parenthese) 85, 172, 185n.29 parody 56, 57, 309n.27 partimenti 160, 181n.2 passions; see emotion pastoral 200–1, 248, 265, 267–8, 269, 280n.43, 280n.44 perception melodic 6, 189 metric 90, 93, 100, 110n.28, 110n.30 performances private vs. public 1, 21, 22 performers 8n.2, 35 fictional performer 62, 63–4 performing 265, 269, 270, 279n.38, 281n.48, 311 phrase completeness 110n.34 phrase-endings; see ending formulas phrase rhythm 125, 196 phrase structure 58, 91, 191 and metre 5, 86, 87, 91, 95, 97, 104

hybrid theme-type 183n.18, 184n.25, 185n.28 period 164–5, 183n.15, 183n.18, 185n.35, 259–60 sentence 183n.18, 191, 192, 195–7, 227n.4 see also syntax phrases beginning- vs. end-accented 6, 112, 148, 258, 279n.37; see also trochee vs. iamb; ‘head/end ambiguity’ pianto 194, 215, 217, 227n.16 play 2, 19, 171, 178, 225–6, 285, 310 with metre 94, 99, 106–7, 141, 312 with topics 201, 228n.27, 249, 250, 265 plot 250 poetic feet 109n.17; see also trochee vs. iamb poetic metre quinario doppio 145 iambic tetrameter 123 polonaise 113 popular 35, 47, 58, 65–6 artful and 37, 38, 42; see also commerce, art and printing 1, 17, 23; see also publishing prose, musical 141, 243 prototype 199, 209, 224, 225, 228n.24 public 17; see also audience public opinion (doxa) 5, 14, 18, 25, 26, 27, 28 ‘public sphere’ (Habermas) 5, 17, 18–19, 28, 31n.17 musical 20, 21–3 public taste 5, 22, 35, 39, 41, 47 publishing 5, 17, 21, 22, 24, 39, 49n.18 punctuation, musical 193; see also phrase endings reader ‘implied’ 16 see also listener, ideal reader-response theory 3 re-barring 114–16, 155n.14, 157n.39, 260, 264 reception of Beethoven’s Op. 31 No. 1 72–3 of Mozart’s K331/i 255–6 reception theory 3 r`egle de l’octave 163 repetition 91, 99–100, 110–11n.37, 191, 227n.5, 235–6, 245; see also syntactic processes representation fictional 13 musical (Malerei) 56, 57 rhetoric 34, 222–3 musical 1, 2, 3, 7, 25–6, 46, 197, 250, 300, 301, 302, 305, 306

Index of concepts

rhetorical figures 1 ricercar (topic) 248 Ritornello 199, 200 Romanesca (‘Pachelbel canon’) 185n.27 Romanticism 7, 20–1, 23–4, 28–9, 46 rondo 245 R¨uckung (shift) 83 rules of composition 2, 4, 56, 58, 106 departures from 56; see convention, departures from of the genre 5, 25, 27–8, 29 see also convention; norms sarabande 113 schema 4, 6, 189, 194, 224 ‘1–7 . . . 4–3’ 193–4 s o u r c e - pat h - g oa l 224, 225 c e n t r e - p e r i p h e ry 224–5 see also formal schemata Schleifer 215, 216, 217, 219, 220, 269, 270 segmentation 231, 252n.8 semantics 2 semiology; see semiotics semiosis extroversive 249 introversive 235, 268 origins of 284 semiotics 4, 20, 189, 192–3, 230, 272, 316n.1 semiotic system; see communication, system of sender vs. receiver 15, 25; see also communication models settings of texts German 123–30 French and Italian 122–3 siciliano 6, 106, 111n.39, 185n.37, 265–8, 269–70, 280n.43, 280n.44, 315 signification 7, 192–3, 311; see also meaning signs 6–7, 192, 310 accidental 283 instituted 283, 305 natural 283 natural vs. arbitrary 223–4 see also semiosis; topics ‘singing allegro’; see singing style 200 singing style 86, 200, 268–9, 270 sonata form 55, 61, 189, 191, 232, 241, 245, 246, 250, 252n.13 metaphoric model of 6, 190, 194–200, 278n.25, 313 ‘sonata theory’ (Hepokoski/Darcy) 2, 226n.1 sonic analogues 7, 286, 289, 290, 292, 297, 300, 306, 315 steps of dance; see gestures

streams 163–4, 178 cadential 164, 178, 181, 184n.19 prolongational 163–4, 178, 181, 184n.19 strict style; see alla breve style Sturm und Drang (topic) 241, 248 sublime 58 surface, musical 251, 259, 262, 269, 271, 279n.38, 281n.48; see also foreground vs. background surprise; see unexpected syntactic processes 289–90, 300, 301 cadence 289–90, 299, 300 displacement 300 imitation 300 pedal 301 repetition 289, 300 syntax 58, 77n.28, 194, 195, 199, 201, 249, 285, 289; see also grammar tactus 110n.30 Taktbeginn vs. Taktschwerpunkt 141, 142, 144, 147, 155n.19 Taktglieder (members of measure) 87–94, 108n.4, 109n.16, 109n.19, 110n.36 Taktteile (parts of measure) 83–4, 89, 90, 95, 100, 108n.4, 109n.16, 109n.19, 110n.30, 110n.36 tarantella 106 tempo giusto 111n.42 thematic contrast 6, 189–90, 191 tonal models 231–46 ‘top-down’ vs. ‘bottom-up’ processes 191, 258, 277n.20 topics 1, 2, 6–7, 247–8, 251, 271–2, 281n.54, 314–15 and metre 86, 105–6, 107 dance topoi 285, 292, 306 see also music analysis, topical analysis topoi; see topics trochee vs. iamb 260, 261–5, 269, 270, 271, 279n.37 Turkish style 280n.44 ‘two-time rule’ 6, 189, 192, 223, 225 unexpected 56, 58–9 variation 199, 228n.25, 246, 252n.13 verisimilitude 5, 13, 14, 24, 25–8, 29 Verwirrung (confusion) 83 waltz 66, 291 choreography of 286 music for 286–9, 308n.14 wit (Witz) 2, 19, 23, 56–7, 59, 312

339

Index of names and works

Abbate, Carolyn 254–5, 276n.6 Abel, Carl Friedrich 38–9 Abravaya, Ido 132 Adorno, Theodor 231 Agawu, Kofi 2, 6, 227n.8, 228n.25, 228n.26, 228n.27, 229n.31, 274, 277n.21, 278n.26, 280n.42, 281n.54, 307n.4 Allanbrook, Wye J. 2, 6–7, 86, 111n.39, 111n.41, 185n.35, 185n.37, 186n.43, 285, 292, 308n.22, 315 Allgemeine musikalische Zeitung 46, 55, 74, 309n.22 Anderson, Emily 42 Aristotle 34 Artaria (publishing firm) 39, 41, 43–4, 49n.18, 280n.41 Asioli, Bonifazio 157n.39 Bach, Carl Philipp Emanuel 27, 38–9 Rondo in E major, h265 132 Sonata in B flat major, h282/ii 131–2 Sonata in B minor, h32.5/ii 157n.31 Bach, Johann Christian 35, 122 Carattaco 151 La clemenza di Scipione 151 Bach, Johann Sebastian 22, 246, 280n.44 Cantata No. 78, ‘Jesu, der du meine Seele’, bwv78 No. 7 ‘Herr, ich glaube, hilf mir Schwachen’ (Chorale) 123–5 Cantata No. 99, ‘Was Gott tut, das ist wohlgetan’, bwv99 No. 6 ‘Was Gott tut, das ist wohlgetan’ (Chorale) 123–5 Fugue in C minor, bwv847 (Das Wohltemperierte Klavier I) 182n.10 Invention No. 1 in C major, bwv772 252n.13 Orchestral Suite No. 2 in B minor, bwv1067 Badinerie 133–4 ‘Peasant’ Cantata, bwv212 127 St Matthew Passion, bwv244 No. 54 ‘O Haupt voll Blut und Wunden’ 276n.12

Suite for lute in E minor, bwv996 Bourr´ee 296, 298 Baker, James 186n.44 Batteaux, Charles 7 Beach, David 275, 277n.17 Becking, Gustav 72 Beethoven, Karl van 53, 54 Beethoven, Ludwig van 37, 46, 64, 78n.34, 141, 181, 235–6, 243, 246, 312, 314 and tonal models 231, 232 and topics 248, 249, 253n.17 Bagatelle Op. 126 No. 3 167–9, 178 Diabelli Variations 116 Egmont 260 Kessler Sketchbook 54 piano sonatas Op. 2 No. 1/i 192–3, 194, 205 Op. 7/ii 173–7, 178 Op. 13 54, 69 Op. 28/iv 114 Op. 31 29, 53–5; No. 1 5, 54, 74n.4, 78n.42; i 59–74, 77n.33, ii 73, 79n.51; No. 2 54; i 59; No. 3 54 Op. 53 (‘Waldstein’) 54, 74, 78n.44; i 62–3, 228n.18 Op. 57 (‘Appassionata’) 62, 74, 110n.36; i 63 Op. 79/ii 114 Op. 110 74n.1 Piano Sonatinas Op. 49 74n.1 Piano Sonatina Op. 54 74n.1 Piano Trio Op. 1 No. 3 45 string quartets Op. 18 No. 3/i 6, 230–1, 232–46, 247, 248–9, 250–1, 314, 315 Op. 59 No. 3/ii 141–2 symphonies No. 3 (‘Eroica’) 200 No. 5 i 46, 77n.32 No. 6/i 59, iii 65 No. 7/ii 144 No. 9 255, iv 123 Violin Sonatas Op. 29 54 Wielhorsky Sketchbook 54

Index of names and works

Bekker, Paul 72 Berger, Karol 3 Besseler, Heinrich 3 Bilson, Malcolm 280n.45 Blair, Hugh 223, 224, 229n.35 Lectures on Rhetoric and Belles Lettres 222–3 Blume, Friedrich 227n.6, 276n.9 Bonds, Mark Evan 3, 5, 8n.1, 19, 25–6, 27–8, 29, 312–13, 316 Brandenburg, Sieghard 74n.2, 75n.5 Bregman, Albert 182n.11 Breitkopf, Christoph Gottlob (publisher) 37 Breitkopf and H¨artel (publishing firm) 53, 54 Brendel, Alfred 73 Budday, Wolfgang 10n.10, 110n.37, 226n.3, 292 Burke, Edmund 20 Burney, Charles 22, 38–9, 49n.15, 49n.16 Burnham, Scott 260 Buxtehude, Dietrich 113 Campbell, George 34, 36 Caplin, William E. 6, 109n.15, 158n.44, 192, 194, 197–8, 228n.21, 228n.28, 229n.30, 313 Carissimi, Giacomo 113 Carpani, Giuseppe 37–8 Chambonni`eres, Jacques Champion de 113 Cherubini, Luigi 37, 42 Chopin, Fryderyk 118–21 Nocturne in E flat major, Op. 9 No. 2 118–20, 141, 144, 155n.19, 156n.22, 157n.39 Pr´elude in C minor Op. 28 No. 20 144 Clementi, Muzio 45 Cobley, Paul 5, 316 Condillac, Etienne Bonnot de 7, 283–4, 305 Cone, Edward T. 154n.3, 154n.4, 255, 257, 259, 260, 263–4, 273, 274–5, 279n.39 Constanz School 16; see also Iser, Wolfgang; Jauss, Hans Robert Cook, Nicholas 291 Cooper, Grosvenor 258 Corelli, Archangelo 22, 113 Couperin, Franc¸ois 132–3, 134 Ordre No. 6, ‘Les Bergeries’ 133, 148 Ordre No. 9, ‘La S´eduisante’ 133, 148 Cramer, Carl Friedrich 41, 51n.30, 51n.31 Cumming, Naomi 307n.8, 309n.23 Czerny, Carl 72, 73, 78n.44, 147, 158n.40 Dahlhaus, Carl 141, 314 Damasio, Antonio 285

Darcy, Warren 2, 226n.1 Defoe, Daniel The Life and Strange Surprizing Adventures of Robinson Crusoe 13, 27 de Giovannini, [first name unknown] An den Schlaf 125 de la Motte, Dieter 141 DeVoto, Mark 274 d’Indy, Vincent 148 Dittersdorf, Carl Ditters von 41, 45, 57 Drabkin, William 274 Dreyfus, Laurence 252n.13 Einstein, Alfred 276n.15 Epstein, David 274 Erd¨ody, Ladislaus 44 Fallersleben, August Heinrich Hoffmann von Das Lied der Deutschen (‘Deutschland, Deutschland u¨ ber alles’) 115–16 Feather, John 17 Flaubert, Gustave 306 Forschner, Hermann 10n.10, 226n.3 Forte, Allen 274 Froberger, Johann Jacob 113 Fux, Johann Joseph 156n.24 Gabrielsson, Alf 274 Gebel, Georg 121–2 Georgiades, Thrasybulus 154n.4 Gerber, Ernst Ludwig 37 Gilbert, Stephen E. 274 Gjerdingen, Robert O. 4, 6, 181–2n.2, 193–4, 227n.15, 274 Gluck, Christoph Willibald 135 Die Pilger von Mekka (La rencontre impr´evue) No. 3 ‘Unser dummer P¨obel meint’ (‘Les hommes pieusement’, Aria Calender) 115, 155n.15 Goethe, Johann Wolfgang von 8n.2 Goode, Luke 18 Goody, Jack 17 G¨orner, Johann Valentin An den Schlaf 125–7 ´ M´arta 2 Grabocz, Graun, Carl Heinrich Der Tod Jesu 123 Grave, Floyd 107n.1, 114 Griesinger, Georg August 37, 42, 310 Habermas, J¨urgen 5, 18–19, 20, 21, 22, 23, 31n.17 Hagedorn, Friedrich von An den Schlaf 125

341

342

Index of names and works

Hal´evy, J.-F.-F.-E. La Juive No. 19 ‘Rachel, quand du Seigneur’ (Aria Eleazar) 144, 148 Handel, George Frideric 22, 266, 280n.44 Messiah No. 9 ‘O thou that tellest good tidings to Zion’ 135–6, 141 Hanslick, Eduard 235 Harvey, David 275, 277n.17, 281n.48 Hasty, Christopher 147, 231 Hatten, Robert 2 Hawkins, John 22 Haydn, Joseph 8n.2, 39, 107, 136, 181, 228n.24 and topics 248, 249 as Beethoven’s teacher 45 as Pleyel’s teacher 43, 44, 45 compared to Sterne 3, 19, 25 composing for Kenner and Liebhaber 28, 38, 39, 42 ‘cult of ’ 23 humour in 23, 55, 57 Kaiserhymne (‘Gott erhalte Franz den Kaiser’) 115 monothematicism 190, 191 string quartets Op. 1 ; No. 1/i 139 Op. 20 ; No. 2/iv 43; No. 5/iii 266; iv 43; No. 6/iv 43 Op. 33 43, 138; No. 2 (‘The Joke’) 19, iv 39; No. 5/iv 266, 281n.49 Op. 50 ; No. 1/ii 87, 106; No. 3/iv 104–5, 106; No. 5/iv 103–4, 106; No. 6/ii 87–8, 106 Op. 54 ; No. 1/i 169–71, 173, 178, 184n.26 Op. 55 ; No. 3/iv 92–4, 106 Op. 64 ; No. 2/i 91–2, 94, 105–6, 138; No. 6/i 88–90, 106, iii 101–3, 106 Op. 71 ; No. 3/iv 139–41 Op. 74 ; No. 1/ii 157n.39 Op. 76 ; No. 4/iv 7, 285, 290, 292, 296–7, 296–306, 315, 316 symphonies No. 42/ii 38 No. 94 (‘Surprise’), ii 19, 39 No. 98/ii 39 The Creation 37 Head, Matthew 276n.9, 280n.42, 280n.44 Hegel, Georg Wilhelm Friedrich 227n.14 Hepokoski, James 2, 226n.1 Hiller, Johann Adam Die Liebe auf dem Lande 57

Hoffmann, Ernst Theodor Amadeus 46, 47, 311 Hoffmeister, Franz Anton (composer and publisher) 39, 47, 52n.34 Holman, Peter 22 Holtmeier, Ludwig 181n.2 Horner, Johann Jakob 53–4, 74n.2 Irving, Howard 19 Iser, Wolfgang 3 Jakobson, Roman 16, 235 Jank´el´evitch, Vladimir 310 Jauss, Hans Robert 3 Jenne, Natalie 293, 308n.19, 309n.24 Johnson, Samuel 26 Jonas, Oswald 121 Kalischer, Alfred 74n.1 Kant, Immanuel 223 Kinderman, William 62, 73 Kirnberger, Johann Philipp 4, 105, 106, 108n.10, 110n.32, 114 cadence placement 108n.4, 147, 154n.5, 157n.40 compound metre 113–14, 117, 127, 157n.40 Koch, Heinrich Christoph 4, 55, 83, 106, 108n.10, 110n.32, 110n.34, 110n.36, 135, 156n.23, 222, 260, 268, 280n.44, 306 cadence placement 108n.4, 114, 121–2, 123, 147, 156n.22 compound metre 94, 123, 127, 135 compound vs. simple metre 83–4, 85–6, 109n.19 double measures 77n.32, 78n.34, 94, 95, 96 Metrum 89, 100, 109n.17, 109n.19 phrase-endings 88, 93, 96–7, 99, 100–1, 107n.3, 108n.4, 110n.34, 191, 193 phrase structure 100, 102, 109n.19, 110n.34 Kolberg, Oskar 78n.37 Kovaleff Baker, Nancy 109n.18, 109n.20 Krebs, Harald 107n.1 K¨uster, Konrad 72 Lakoff, George 224 Langer, Susanne K. 307n.4 Lenz, Wilhelm von 72 Lerdahl, Fred and Ray Jackendoff 110n.30, 148, 158n.48, 228n.23, 255–6, 257, 258, 262–3, 264, 266, 267, 273, 274, 275, 277n.17, 278n.32, 281n.48 Lessing, Gotthold Ephraim 20 Lester, Joel 257, 273, 281n.47

Index of names and works

Levy, Janet 3 Lidov, David 309n.23 Lippmann, Friedrich 123, 127, 147, 158n.41 Little, Meredith 292–3, 294, 308n.19, 309n.24 Lockwood, Lewis 253n.17 Love, Harold 21, 22–3, 24 Magazin der Musik 41, 51n.30, 51n.31 Marpurg, Friedrich Wilhelm 83, 107n.4, 108n.4, 108n.5, 135, 157n.31 Marshall, Robert 123 Marx, Adolf Bernhard 72, 286–8, 289–90, 291, 301 Mattheson, Johann 107n.4, 113, 121–2, 123–5, 154n.5, 154n.6, 228n.21 Maurer Zenck, Claudia 5, 10n.10, 20, 29, 86, 95–6, 101, 110n.36, 111n.42, 114, 138, 141, 274, 312 McClary, Susan 253n.17 McLuhan, Marshall 15 Mendelssohn, Moses 223, 224 Meyer, Leonard B. 2–3, 182n.6, 182n.7, 227n.15, 228n.17, 228n.29, 259 implication-realization 4, 161–2, 178, 194 rhythmic-metric analysis 257, 258, 261, 263–4, 272–3, 274–5, 278n.32, 278n.37, 279n.38, 279n.39 Michaelis, Christian Friedrich 55, 56, 57–9, 60, 71, 78n.42 Mirka, Danuta 185n.34, 312, 313 ˆ Momigny, J´erome-Joseph de 118, 119, 148 Monelle, Raymond 2, 227n.16, 229n.32, 229n.33 Morgan, Robert P. 253n.19, 257, 258–9, 261, 262, 263, 273, 274, 279n.39 Mozart, Leopold 35–6, 41, 43–4, 45, 51n.28 Mozart, Wolfgang Amadeus 39, 47, 52n.34, 57, 64, 107, 123, 136, 151, 164, 181, 225–6, 228n.25, 229n.43 between art and commerce 35–6, 39, 40, 44 composing for Kenner and Liebhaber 41–2 dance topics 248, 292, 308n.22 and Pleyel 37, 43–4, 45 sonata form expositions 190, 191, 195–6, 199–200, 224 and topics 249 Ascanio in Alba, k111 No. 24 ‘Che strano evento’ (Coro di Pastorelle) 152 Cos`ı fan tutte, k588 No. 25 ‘Per piet`a, ben mio, perdona’ (Rondo` Fiordiligi) 153 Die Zauberfl¨ote, k620 76n.20, 115, 158n.41

No. 2 ‘Der Vogelf¨anger bin ich ja’ (Aria Papageno) 147, 158n.41 No. 7 ‘Bei M¨annern welche Liebe f¨uhlen’ (Duet Pamina/Papageno) 115, 155n.14 Don Giovanni, k527 40, 158n.41 No. 1 ‘Notte e giorno faticar’ (Introduzione, Leporello) 116, 118, 155n.17 No. 11 ‘Finch’ han dal vino’ (Aria Don Giovanni) 142 No. 12 ‘Batti, batti, o bel Masetto’ (Aria Zerlina) 155n.17 No. 17 ‘Met`a di voi qua vadano’ (Aria Don Giovanni) 153 No. 18 ‘Vedrai carino’ (Aria Zerlina) 142 Ein musikalischer Spaß k522 39 Eine kleine Nachtmusik k525/iii 260, 262 Fugue for two pianos k427 52n.34 ‘Haydn’ Quartets 41, 43, 44–5, 141 k387 43, i 141, 142, 196, iv 41, 42–3, 86, 105 k421 43, i 86, 105, 141, iv 266 k428 43, ii 41, 43 k458 (‘Hunt’) 43, 139 k465 (‘Dissonance’), i 41, 43, 196, 214, 215 Idomeneo, k366 No. 18 ‘Corriamo, fuggiamo’ (Coro) 152 No. 20b ‘Spiegarti non poss’io’ (Duet Ilia/Idamante) 153 Il re pastore, k208 152 Il sogno di Scipione, k126 No. 10 ‘D`ı che sei l’arbitra’ (Aria Scipione) 152 La clemenza di Tito, k621 No. 19 ‘Deh per questo istante solo’ (Rondo` Sesto) 153 La finta semplice, k51 No. 2 ‘Troppa briga a prender moglie’ (Aria Simone) 151 No. 4 ‘Non c’`e al mondo altro che donne’ (Aria Cassandro) 151 No. 19 ‘Cospetton, cospettonaccio!’ (Duet Fracasso/Cassandro) 151 La finta giardiniera, k196 No. 5 ‘A forza di martelli’ (Aria Nardo) 152 No. 12 (Finale I) 152 No. 19 ‘Gi`a divento freddo’ (Aria Contino) 152 No. 23 (Finale II) 152 Le nozze di Figaro, k492 153

343

344

Index of names and works

Mozart, (cont.) No. 12 ‘Voi, che sapete’ (Arietta Cherubino) 142 L’oca del Cairo, k422 No. 2 ‘Se fosse qui nascoso’ (Aria Auretta) 153 No. 6 (Finale) 153 Anh. ‘Ho un pensiero’ (Duet Auretta/Chichibio) 129–30, 153 Lo sposo deluso, k430 153 Lucio Silla, k135 No. 6 ‘Fuor di queste urne dolenti’ (Coro) 152 Mitridate, Re di Ponto, k87 No. 8 ‘Se di lauri il crine adorno’ (Cavata Mitridate) 151 No. 24 ‘Gi`a dagli occhi il velo e` tolto’ (Aria Farnace) 151 piano concertos k413–415 36, 41, 42 k449–451 40 k488/ii 266 piano sonatas k281/iii 166–7 k283/i 193–4, 205 k310/i 186n.45 k330/ii 171–3, 178 k331/i 6, 111n.39, 178–80, 186n.42, 254–70, 271, 273–5, 276n.15, 280n.41, 315, iii 280n.44 k332 256, 276n.15, 280n.41, i 281n.45 k333 256, 276n.15, 280n.41 Piano Trio k502/ii 165–6 Piano Variations k455 115, 155n.15 ‘Prussian’ Quartets 40 k575/iii 109–10n.28 Rondo in A minor k505 266 Sonata for piano and violin k377/ii 265 Sonata for piano duet k521 52n.34 String Quartet k499 52n.34 string quintets k515/i 6, 110n.28, 195, 196, 198–9, 200–14, 217, 218, 219, 313, iii 96–9, 100, 106, 110n.28 k614/iii 96, 97–100, 106, 110n.28 symphonies No. 31, k297/300a (‘Paris’) 36–7, 39 No. 40, k550/i 207, 214 No. 41, k551 (‘Jupiter’), i 6, 195, 196, 198–9, 200, 208, 214–21, 229n.33, 313, iv 43 Un bacio di mano k541 229n.33

Zaide k344/336b No. 1 ‘Br¨uder, laßt uns lustig sein’ (Coro) 127–9 Nagel, Wilibald 72 N¨ageli, Hans-Georg 5, 29, 53–5, 72, 74n.2, 75n.8, 310 Narmour, Eugene 4, 257, 274, 275n.1, 277n.17, 277n.20, 278n.37 Nattiez, Jean-Jacques 6, 230, 235, 274, 314, 316n.1 Neumeyer, David 257, 274, 277n.17 Niedt, Friedrich 113, 154n.6 Parker, Roger 254–5 Pasticci, Susanna 257, 274, 277n.16 P´ecour, Louis 294–6, 298 Peel, John 275 Pergolesi, Giovanni Battista 136 Stabat Mater 114 No. 3 ‘O quam tristis’ 135 Peters, Jeffrey 26 Peters, John Durham 15, 30n.6 Philidor, Franc¸ois-Andr´e 56 Piston, Walter 273, 274 Plato 272 Pleyel, Ignaz 37, 45 String Quartets Op. 1 43–5, 51n.28 Pozzi, Egidio 274 Puchberg, Michael 40 Purcell, Henry 22, 113 Quantz, Johann Joachim 280n.44 Rameau, Jean-Philippe 154n.5, 156n.25, 162–3, 182n.5, 182n.8, 283, 284, 305, 307n.2 Rameau, Pierre 293–4 Ratner, Leonard 2, 3, 111n.40, 230, 244, 247–8, 250, 314, 315 Reddy, Michael 290 Reinagle, Alexander 38 Richardson, Samuel 17 Clarissa 24 Ricoeur, Paul 226, 229n.43 Riemann, Hugo 72, 112, 118–21, 141, 147, 148, 155n.19, 256, 258, 259, 260, 262, 263, 273, 277n.22 Riepel, Joseph 4, 83, 108n.10, 109n.16, 110n.36, 112, 150–1, 158n.49 cadence placement 107n.4 compound vs. simple metre 84–5, 86, 108n.6

Index of names and works

double measures 95, 109n.24 phrase-endings 192, 193 Ries, Ferdinand 54 Rochlitz, Friedrich 46–7, 52n.34, 55, 57, 58, 59, 60, 67, 74 Rosen, Charles 2, 107, 189, 314 Ross, Angus 18 Rothstein, William 6, 109n.21, 109n.28, 110n.34, 182n.5, 226n.4, 312, 313, 316 Rousseau, Jean-Jacques 7, 306, 309n.29 Rudowski, Victor Anthony 20–1 Ruwet, Nicolas 6, 230, 235, 314 Saint-Sa¨ens, Camille 112, 148 Sanguinetti, Giorgio 181n.2 Sapir, Edward 286 Saussure, Ferdinand de 272 Scarlatti, Alessandro 280n.44 Schachter, Carl 147, 275, 276n.6 Scheinpflug, Christian Gotthelf 281n.50 Schenker, Heinrich 112, 156n.24, 161, 182n.5, 183n.13, 253n.14, 256, 257, 258, 270, 273, 275, 278n.37 Schmalfeldt, Janet 227n.14 Schoenberg, Arnold 141, 160, 181, 183n.15, 190, 191, 192, 193 Schubert, Franz 86, 246 Sechzehn L¨andler und zwei Ecossaisen (genannt ‘Wiener Daimler-L¨andler’) Op. 67 No. 1 308n.15 No. 6 308n.15 No. 8 308n.15 No. 14 308n.15 Wiegenlied, D867 142–4 Schueller, Herbert M. 20 Sebeok, Thomas A. 30n.11 Seidel, Wilhelm 3, 10n.10, 226n.3, 227n.5 Shannon, Claude E. 15 Shapiro, Joel 72–3 Sieber, Jean-Georges (publisher) 44 Sisman, Elaine 2, 3, 28, 36, 228n.24 Slawson, Wayne 275 Solomon, Maynard 52n.34, 225–6 Spitzer, Michael 6, 8n.1, 259–60, 275, 278n.25, 313–14 Sterne, Laurence 3, 19, 20, 25 The Life and Opinions of Tristram Shandy, Gentleman 13, 27, 29 Storace, Nancy 45 Sulzer, Johann 55, 222, 229n.35 Swift, Jonathan

Travels into Several Remote Nations of the World by Lemuel Gulliver 13 ‘Hints toward an essay on conversation’ 19 Tchaikovsky, Pyotr 246 Telemann, Georg Philipp An den Schlaf 125 Temperley, David 141 The Spectator 18 The Tatler 18 Thomas, Downing 284 Tolley, Thomas 23 Tomasello, Michael 285, 290 Torricella, Christoph (publisher) 39 Tovey, Donald 189, 314 Triest, Johann Karl Friedrich 37 T¨urk, Daniel Gottlob 69, 113, 117, 127, 147, 158n.40 Tyson, Alan 256, 276n.15, 280n.41 Veidl, Theodor 72, 77n.28 Verdi, Giuseppe Il trovatore No. 4 ‘Vedi! le fosche notturne spoglie’ (‘Anvil Chorus’) 144–5 Vogler, Georg Joseph 113–14, 117, 127, 135, 141 Watt, Ian P. 1, 17, 32n.44 Weaver, Warren 15 Weber, Carl Maria von Der Freisch¨utz No. 3 Waltz 287, 289, 290 Weber, Friedrich August 55, 56–7, 59, 76n.20 Weber, Gottfried 112, 114, 144, 154n.3, 258, 259, 262, 263, 264, 277n.22 Weber, William 21 Webster, James 38, 158n.46, 184n.26 Westphal, Rudolph 118, 121, 147, 148, 155n.17, 156n.20 Wheelock, Gretchen A. 3, 9n.2, 19, 23, 309n.27 Wiehmayer, Theodor 148, 158n.44 Wilson, Thomas 286, 287, 289, 291, 308n.14 Wolff, Christian 223 Yaraman, Sevin 286, 288 Yeston, Maury 110n.37 Youngren, William H. 125–7 Zbikowski, Lawrence M. 7, 228n.24, 235, 279n.39, 315, 316

345