Opera in Paris from the Empire to the Commune [1° ed.] 1138065161, 9781138065161

Studies in the history of French nineteenth-century stage music have blossomed in the last decade, encouraging a revisio

235 14 14MB

English Pages 488 Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents
List of examples
List of figures
List of tables
Acknowledgements
Ouverture: power, licence and technology
1 The music of power: Parisian opera and the politics of genre, 1806–1864
2 Grand opéra – petit opéra: Parisian opera and ballet from the Restoration to the Second Empire
3 Jacques Offenbach: the music of the past and the image of the present
4 The operas of François-Auguste Gevaert: the tour d’horizon
5 Between Opéra-Comique and Opéra-National: Scribe, Vaëz and Boisselot c1850
Premier entr’acte: les ultramontains
6 Beethoven and Rossini: opera and concert at the end of the Restoration
7 ‘Il n’y a qu’un Paris au monde, et j’y reviendrai planter mon drapeau!’: Rossini’s second grand opéra
8 A transalpine comedy: L’elisir d’amore and cultural transfer
9 Partners in rhyme: Alphonse Royer, Gustave Vaëz, and foreign opera in Paris during the July Monarchy
Second entr’acte: la musique allemande
10 Castil-Blaze and the reception of Weber in Paris, 1824–1857
11 Gluck, politics and the Second Empire press
12 Wagner and Paris: the case of Rienzi (1869)
Strette: Censors and others
Bibliography
Index
Recommend Papers

Opera in Paris from the Empire to the Commune [1° ed.]
 1138065161, 9781138065161

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

Opera in Paris from the Empire to the Commune

Studies in the history of French nineteenth-century stage music have blossomed in the last decade, encouraging a revision of the view of the primacy of Austro-German music during the period and rebalancing the scholarly field away from instrumental music (key to the Austro-German hegemony) and towards music for the stage. This change of emphasis is having an impact on the world of opera production, with new productions of works not heard since the nineteenth century taking their place in the modern repertory. This awakening of enthusiasm has come at something of a price. Selling French opera as little more than an important precursor to Verdi or Wagner has entailed a focus on works produced exclusively for the Paris Opéra at the expense of the vast range of other types of stage music produced in the capital: opéra comique, opérette, comédie-vaudeville and mélodrame, for example. The first part of this book therefore seeks to reintroduce a number of norms to the study of stage music in Paris: to re-­establish contexts and conventions that still remain obscure. The second and third parts acknowledge Paris as an importer and exporter of opera, and its focus moves towards the music of its closest neighbours, the Italian-speaking states, and of its most problematic partners, the ­German-speaking states, especially the music of Weber and Wagner. Prefaced by an introduction that develops the volume’s overriding intellectual drivers of cultural exchange, genre and institution, this collection brings together twelve of the author’s previously published articles and essays, fully updated for this volume and translated into English for the first time. Mark Everist is Professor of Music at the University of Southampton. His research focuses on the music of Western Europe in the period 1150–1330, opera in France in the nineteenth century, Mozart, reception theory and historiography. He is the author of Polyphonic Music in Thirteenth-Century France (1989), French Motets in the Thirteenth Century (1994), Music Drama at the Paris Odéon, 1824–1828 (2002), Giacomo Meyerbeer and Music Drama in Nineteenth-Century Paris (2005) and Mozart’s Ghosts: Haunting the Halls of Musical Culture (2013) as well as the editor of three volumes of the Magnus Liber Organi for Editions de l’Oiseau-Lyre (2001–2003). In addition, he has published over 80 articles in peer-reviewed journals and collections of essays. The recipient of the Solie (2010) and Slim (2011) awards of the American Musicological Society, he was elected a fellow of the Academia Europaea in 2012. Everist was president of the Royal Musical Association from 2011–2017 and was elected a corresponding member of the American Musicological Society in 2014. His monograph Discovering Medieval Song: Latin Poetry and Music in the Conductus was published with Cambridge University Press in 2018, as was The Cambridge History of Medieval Music, co-edited with Thomas Kelly. A monograph on the reception of Gluck in the nineteenth century has just been completed.

Opera in Paris from the Empire to the Commune

Mark Everist

First published 2019 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 Mark Everist The right of Mark Everist to be identified as author of this work has been asserted by them in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record has been requested for this book ISBN: 978-1-138-06516-1 (hbk) ISBN: 978-1-315-15993-5 (ebk) Typeset in Times New Roman by codeMantra

Contents

List of examples List of figures List of tables Acknowledgements

vii ix x xii

Ouverture: power, licence and technology 1 1 The music of power: Parisian opera and the politics of genre, 1806–1864 9 2 Grand opéra – petit opéra: Parisian opera and ballet from the Restoration to the Second Empire

53

3 Jacques Offenbach: the music of the past and the image of the present 104 4 The operas of François-Auguste Gevaert: the tour d’horizon 131 5 Between Opéra-Comique and Opéra-National: Scribe, Vaëz and Boisselot c1850 165 Premier entr’acte: les ultramontains

221

6 Beethoven and Rossini: opera and concert at the end of the Restoration 224 7 ‘Il n’y a qu’un Paris au monde, et j’y reviendrai planter mon drapeau!’: Rossini’s second grand opéra

242

8 A transalpine comedy: L’elisir d’amore and cultural transfer 287 9 Partners in rhyme: Alphonse Royer, Gustave Vaëz, and foreign opera in Paris during the July Monarchy 310

vi Contents

Second entr’acte: la musique allemande 329 10 Castil-Blaze and the reception of Weber in Paris, 1824–1857 331 11 Gluck, politics and the Second Empire press 355 12 Wagner and Paris: the case of Rienzi (1869) 370 Strette: Censors and others 411 Bibliography Index

415 439

Examples

4.1 Gevaert, Georgette, end of no. 5, Morceau d’ensemble ‘Attaquons’ 139 4.2 Gevaert, Le billet de Marguerite, extract from no. 5, act i finale ‘Partir! Qu’ai-je entendu?’ 140 4.3 Gevaert, Le billet de Marguerite, extract from no. 8bis, Mélodrame 142 4.4 Gevaert, Le billet de Marguerite, extract from act iii entr’acte 144 4.5 Gevaert, Les lavandières de Santarem, act i introduction, chœur des femmes ‘Joyeuse lavandière’ 147 4.6 Gevaert, Le château trompette, no. 13bis, Musique de scène 157 4.7 Gevaert, Les lavandières de Santarem, no. 2bis, Musique de scène 158 5.1 Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, Larghetto ‘Toi qui séduis mon cœur’, first draft of Allegro fieramente. F-Pn MS 4397 171 5.2 Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, ­L arghetto ‘Toi  qui séduis mon cœur’, second draft of Allegro fieramente. F-Pn ­MS 4397  172 5.3 Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, Larghetto ‘Toi qui séduis mon cœur’, final draft of Allegro fieramente. F-Pn MS 4397 and published score, pp. 81–82 174 5.4 Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, Larghetto ‘Toi qui séduis mon cœur’, Boisselot’s monstre and Scribe’s final version. F-Pn n.a.f. 22545, fols 125v 176 5.5 Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 9, Allegro ­moderato ‘Ciel de feu, beau ciel des Espagnes’, bars 1–7, original version with two alternative vocal lines. F-Pn MS 4397 178 5.6a Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 9, Allegro ­moderato ‘Ciel de feu, beau ciel des Espagnes’, original setting of ­final four lines. F-Pn MS 4397 (transposed from A major to G major to facilitate comparison with Example 5.6b and with original crossings-­through retained) 180

viii Examples 5.6b Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 9, Allegro moderato ‘Ciel de feu, beau ciel des Espagnes’, definitive setting of final four lines (published version) 184 5.7 Boisselot, Mosquita la sorcière, no. 10 act ii finale, Maestoso ‘Sa jalouse fureur’ (published version) p. 185 201 5.8 Verdi, Il proscritto (Ernani), no. 5 finale primo, bars 231–247 204 5.9 Verdi, Il proscritto (Ernani), no. 11 congiura, bars 106–109 208 5.10 Verdi, Il proscritto (Ernani), no. 11 congiura, bars 126–131 209 5.11 Boisselot, Mosquita la sorcière, no. 10 act ii finale, Maestoso ‘Sa ­jalouse fureur’ [page 186, bar 5 to page 187, bar 9] 211 6.1 Beethoven, Symphonie 6, 1: 5–16 234 6.2 Beethoven, Symphonie 6, 1: 289–300 235 6.3 Rossini, Ouverture de Guillaume Tell, 176–184 235 6.4 Rossini, Ouverture de Guillaume Tell, 209–217 236 8.1 Auber, Le philtre, no. 11: [Récit +] Duo (Guillaume, Jolicœur) ‘De désespoir je reste anéanti’ – ‘Si l’honneur a pour toi des charmes’, opening 307 8.2 Donizetti, L’elisir d’amore, no. 11: Scena e duetto (Nemorino, Belcore) ‘La donna è un animale’ – ‘Venti scudi’, opening 308 11.1 ARMIDE / DRAME HÉROIQUE / Mis en Musique / PAR / MR LE CHEVALIER GLUCK / Représenté pour la premiere [sic] fois, par l’Academie [sic] / Royale de Musique, le 23 Septembre 1777 / […] / A PARIS / AU BUREAU DU JOURNAL DE MUSIQUE, Rue Montmartre, / vis-à-vis celle des vieux Augustins / à l’Opéra, Et aux Adresses ordinaires de Musique. / A.P.D.R., 22 363

Figures

6.1 Salle du Conservatoire c1821 231 6.2 Salle du Conservatoire c1841 232 6.3 Orchestral layout of Société des Concerts du Conservatoire, 1828–40 (after Donna Di Grazia, ‘Rejected Traditions: Ensemble Placement in Nineteenth-Century Paris’, 19th-Century Music 22 (1998), 196) 233 7.1 Antoine Etex, Rossini (L’illustration, 13 June 1846). Bibliothèque nationale de France 278 7.2 THEATRE DE L’OPERA / Bustes dans le Foyer / Côté de la Rue en commençant à gauche’ F-Po Estampes: Salles, Opéra Le Peletier (23 February 1872). Bibliothèque nationale de France 280 8.1 Relationships between music, libretti and source texts for Le comte Ory and Le philtre 296

Tables

2.1 2.2 2.3 2.4 2.5 2.6 3.1

Structure of Rossini, Le comte Ory 73 Ballet and Opéra at the ARM, Aug-Sept 1828 77 Structure of Auber, Le philtre 79 Roles and singers in Le comte Ory and Le philtre 81 Ballet and Opéra at the ARM, June-July 1831 83 Repertory of petits opéras, 1828–1863 87 Comparison of Italian and French versions of Rossini, Il signor Bruschino 127 4.1 François-Auguste Gevaert: Operatic Output 133 4.2 One-act opéra comique 136 4.3 Les lavandières de Santarem introduction 159 4.4 Les lavandières de Santarem act ii outline 161 4.5 Le chateau trompette act ii outline 162 5.1 Number of acts in opéra comique, 1834–1854 193 5.2 Boisselot, Mosquita la sorcière, no. 10 act ii finale 199 5.3 Cahiers des charges, Opéra-National, 1 April 1851 and 12 January 1847 214 6.1 Lower strings: Orchestras of the Sociéte des Concerts du Conservatoire and of the l’Académie Royale de Musique, 1829 230 6.2 Programme of Beethoven’s sixth symphony: Paris 1829 and Vienna 1808 238 7.1 Sources for the Paris 1844 Othello 254 7.2 Sources for Robert Bruce (1846) 260 8.1 Italian opera libretti based on French opera libretti, 1800–1899 289 8.2 Casting details for L’elisir d’amore, Le philtre and cognate works 298 8.3 Comparison of Le philtre and L’elisir d’amore 302 9.1 Royer and Vaëz, French adaptations of Italian operas 312 9.2 Theatrical and literary works by Royer and/or Vaëz 314 10.1 Performances of Weber’s operas in Paris, 1824–c1870 335 10.2 The Repertory of Röckel’s troupe, 1829–1831 339 10.3 The Detail of Röckel’s 1831 Season 340

Tables  xi 10.4 10.5 10.6 12.1

Petit opéra during the Restoration and July Monarchy 344 Auguste Schumann’s German Opera Season, April-May 1842 347 Castil-Blaze: Writings on Weber, 1824–1842 350 The repertory of the Théâtre-Lyrique under the direction of Jules-Etienne Pasdeloup, 1868–1870 379 12.2 Wagner, Rienzi, act ii: keys and scoring of the ambassadors’ scene 408

Appendix 2.1 R  epertory of opera and ballet, Académie Royale de Musique, 1817 (titles and number of acts only; details of librettist, composer and premiere in notes to the main text) 58

Acknowledgements

In a collection of essays such as these, the main acknowledgements should go to the publishers who have granted permission to reproduce the articles. It is, therefore, a pleasure to thank the University of California Press (­Chapters 1, 2 and 12), the University of Chicago Press (Chapter 3), the Société belge de musicologie (Chapter 4), Brepols Publishers (Chapter 5), Éditions Mardaga (Chapters 6 and 8), Oxford University Press (Chapter 7), Cambridge University Press (Chapter 9) and Hollitzer Verlag (Chapter 11). A full list of citations is given at the beginning of the Bibliography. I was the recipient of awards from the British Academy that assisted with the research towards Chapters 3, 5 and 9 and from the Arts and Humanities Research Council towards Chapter 1. I thank both these organisations. The chapters in Opera in Paris from the Empire to the Commune originated in keynote addresses and invitations to speak at individual institutions or at conferences. The list of those who invited me would be too long to give here, and there would always be the risk of omission, but I thank all those colleagues who have extended invitations to speak during the last decade or so and to the audiences that provided the feedback that so greatly enriched the publications that reappear here. Two chapters have been the subject of awards: Chapter 2 was honoured by the American Musicological Society’s H. Colin Slim Award for the best article in musicology (beyond early stages) in 2010, and Chapter 3 appeared in a volume I edited with Annegret Fauser that won, again from the American Musicological Society, the Ruth A. Solie Award for the best collection of essays on a musicological topic in 2009. Perhaps my greatest pleasure, though, lies in thanking those colleagues with whom I have talked about theatre, music and culture over the years, which resulted in the publication of these efforts. There are simply too many to mention all by name, but there are those who have read complete chapters at one point or another – in addition to the serried ranks of anonymous peer r­ eviewers – who deserve mention and thanks: Olivier Bara, Thomas B ­ etzwieser, Michele Callala, Jeremy Coleman, Lorenzo Frassà, Philip Gossett, Thomas Grey Steven Huebner, Francesco Izzo, Lawrence Kramer, Ralph Locke, Robert Montemorra Marvin, Klaus Pietschmann, Hilary Porriss, Alban Ramaut, Anthony Sheppard, Henri Vanhulst, Benjamin Walton, William Weber and

Acknowledgements  xiii Kimberly White. Annie Vaughan at Routledge and Jeanine Furino at ­codeMantra provided immeasurable support. But those who deserve the greatest thanks are those who have lived with these projects for the last decade. ­My partner, Jeanice Brooks, has been there at every turn, and my daughter, Amelia-Brooks-Everist, endured two years in an inner-city school in Paris so that research for these essays could proceed. They have my thanks, love and, in the case of the latter, probably the royalties as well. The book is dedicated to the memory of my father: mentor, friend and co-conspiritor. Banister Park, Southampton St Vincent de Paul, Paris

In memoriam Egerton George Arthur Edwin Everist (1923–2018)

Ouverture Power, licence and technology

Getting about the Indian Ocean and the Java Sea in the 1860s could be a capricious business: scheduled shipping could be severely delayed, and journeys could be of unpredictable duration. Travelling opera companies were no exception to these sorts of interruptions, and the ‘Batavia Opera Company’, as the local press called it, found itself in Singapore, en route to ­Jakarta (known then as Batavia) in August 1863. The French opera company stayed for three weeks, waiting for storms to subside, and while waiting put on four productions in the Singapore Town Hall Theatre, only opened the previous year. The troupe, led by Alfred Maugard, consisted of around thirty artists, and their productions, although sung in French to a partly Anglophone audience, were warmly received.1 The company’s opening night on 14 August 1863 consisted of Victor Massé’s opéra comique classic, Galathée, and the comédie-vaudeville, Les erreurs du bel âge; both were fully staged.2 In between the two works were extracts from Meyerbeer’s Robert le diable, and the aria ‘Il va venir’ from Halévy’s La juive, sung by the composer’s probably illegitimate daughter, Marie-Clémence-­ Félicie Proche.3 The trio from Rossini’s Guillaume Tell was advertised but apparently not performed. Although an opéra comique premiered in 1852, a comédie-vaudeville from 1854 and extracts from the early days of grand opéra

1 The account of the first night was given by The Straits Times’ regular but anonymous reviewer (15 August 1863). When it came to the detail of the French stage works, he passed over to a colleague who completed the review in French. For the performance location and its recent construction, see Dhoraisingam S. Samuel, Singapore’s Heri­ tage: Through Places of Historical Interest (Singapore: Elixir Consultancy Service, 1991), 339. 2 The authors of Les erreurs du bel âge were Joseph-Xavier Boniface, (pseud. Xavier), Charles Varin and Léon Dumoustier. Julian-Pierre Nargeot, the orchestra director at the Théâtre des Variétés, where the work had been premiered in 1854, was responsible for the single new composition ‘Allons, mon vieux, il faut faire bombance’ as well as the arrangements of the borrowed music. 3 Diana Hallmann, Opera, Liberalism and Antisemitism in Nineteenth-Century France: The Politics of Halévy’s La Juive, Cambridge Studies in Opera (Cambridge: Cambridge University Press, 2002), 86, and note 50.

2  Ouverture on a historical subject offered a conspectus of the lyric theatre back in Paris,4 it was not until subsequent performances, on 18 and 21 August, that complete performances of grand opéra were given. And when they were, they consisted of Donizetti’s La favorite5 and of the same composer’s Lucia di Lammermoor in its version premiered at the Paris Opéra in 1846.6 The farewell performance was of Flotow’s Martha, given on 26 August.7 This repertorial microcosm, far from home, points to the central feature of Parisian musical culture developed in this book: the relationship of genre to questions of institution. The English part of the audience in Singapore would have known little of how the two Donizetti works had their origins in productions and early drafts of works at the short-lived Théâtre de la Renaissance, but their French colleagues would have been very sensitive to the generic differences between opéra comique, characterised by its use of spoken dialogue, and comédie-vaudeville, whose musical element – not ­significantly less than that found in opéra-comique – was largely borrowed. Generic and institutional sensibilities – even in Anglo-French Singapore – were but an echo of those that had been on display in Paris for half a century.8 All aspects of the Parisian theatre were regulated by licence in the period from when the relevant legislation was enacted – 1806–1807 – until it was repealed in 1864; the system thus underwrites almost all the essays in this book. Despite changes in political and administrative regimes, the licensing system was designed to minimise competition between organisations so that a particular genre – Italian opera or mélodrame for example – was restricted to a single theatre or small group. The system was inflected by a number of factors: state subvention, the predilection of certain regimes (the Restoration [1814–1830] and Second Empire [1852–1870], for example) for both direct control of, and for a privileged status for, the Opéra, and by the personalities of the ­directors themselves. The July Monarchy (1830–1848), led by its ‘­Citizen-King’ ­Louis-Philippe, preferred a system of managers who ­undertook their work at their own financial risk. But every aspect of the production and consumption of the theatre and its music was subject to the contest and competition that the implementation of the licensing system entailed. Recognising the importance of the relationship between genre and institution requires a recalibration of the too-comfortable structuring of 4 Les erreurs du bel âge was much less admired than the Victor Massé: ‘The Vaudeville itself was no doubt very good, as it proved highly amusing to the foreign part of the audience. Vaudevilles, however, are purely French, and do not suit English taste’ (The Straits Times, 15 August 1863). 5 Ibidem, 22 August 1863. 6 Ibidem, 15 August 1863. 7 Announced ibidem, 22 August 1863, but no review was forthcoming. 8 In fact, two of the eleven musical numbers in Les erreurs du bel âge are borrowed from opéra comique of the Restoration period. It included arias from La vieille by François-­Joseph ­Fétis (1826) and Ferdinand Hérold’s La clochette (1817). Of the remaining nine numbers, five were timbres, and four borrowed original airs from other comédies-vaudevilles.

Ouverture  3 work on ‘opera’ around composers who survive in the early twenty-first-­ century canon and the works of those composers that are regularly played today. Immediately, it is possible to identify a tension between this focus on the largely Italian and German focus of ‘opera studies’ in the early twenty-first century and the nineteenth-century position where music that originated in the Parisian theatre, or that was filtered through it, was heard as part of a Franco-Italian underpinning of music in the theatre across Europe and the wider world;9 the example from 1863 Singapore is emblematic. Terminology is central here, with the term ‘opera’ problematic in the sense that it directs attention to works that form part of the ‘repertory’ of the early twenty-first century, which in turn depends heavily on through-composed works from the 1830s and later, with Mozart, a little Handel, Monteverdi and some Rameau perhaps. Trying to encompass such works as these as well as opéra comique (from which modern ears are almost completely protected) and opérette together with its related genres, to say nothing of mélodrame, ballet-­pantomime or comédie-vaudeville, entails linguistic gymnastics that risk both contortion and pleonasm. It is clear, however, that the term ‘opera’ does not do sufficient justice to the complex generic networks found not only in Paris, but also in most large cities worldwide. To think of the relationship between genre and institution as a simple binary association is to misunderstand the complexity of artistic life in nineteenth-century Paris. It requires inflection through taking account of chronology and questions of power. Chronology is very much a false friend here. Received wisdom maps artistic change onto major ‘political’ change – ­‘great-date’ change – and for nineteenth-century France these are well known: the two Bourbon Restorations of 1814 and 1815, Les trois glorieuses of 1830, the 1848 Revolution and the Franco-Prussian War, Siege of Paris and the Commune in 1870–1871. In almost every case, music and theatre were largely able to ride out changes of government, and those agents responsible for words, notes, dance, costumes and staging created an entirely different set of caesurae in this history of nineteenth-century music in the theatre: the establishment of the licensing system in 1806–1807, the ­emergence of grand opéra between 1828 and the mid-1830s, changes in the organisation of the Opéra coupled to the registral dislocation of opéra comique which led to the emergence of opérette between 1854 and 1856, the repeal of the licensing legislation in 1864 and – all in 1873 – the burning of the Salle le Peletier (which had housed the Opéra since 1821), the collapse of Théâtre-Lyrique and that of the Théâtre Italien (which had given a home 9 For a dissection of the Franco-Italian background to much cultural transfer of music in the nineteenth-century theatre as a counterweight to the idea of an Italian ‘invasion’, see Mark Everist, ‘Cosmopolitanism and Music for the Theatre: Europe and Beyond, ­1800–1870’, Music History and Cosmopolitanism, ed. Anastasia Belina, Kaarina Kilpio and Derek Scott (London and New York: Routledge, forthcoming).

4  Ouverture to Italian-language opera since the beginning of the century).10 It is certainly true that events marked by ‘great dates’ resulted in the collapses of Adolphe Adam’s short-lived Opéra National in 1848, and the destruction of the Théâtre-Lyrique’s newly built home on the Place du Châtelet, only nine years after its construction in 1862, during the Commune on 25 May 1871. But Adam’s enterprise could well be viewed as just a false start in the history of the Théâtre-Lyrique, and the burning of Gabriel Davioud’s new building for the same institution resulted in an interruption of service for only thirteen weeks in the middle of the summer before operations were transferred to the Salle de l’Athénée.11 Chronologies of music in the nineteenth-century theatre demand a more nuanced understanding of both music and politics, to say nothing of an understanding of the polity itself. Chronology points away from the simple mapping of ‘great dates’ onto single works, then, and more towards the way in which politics may be genuinely understood in relation to the creation of music in the theatre: not from the crude charting of political events – the revolution of 1848, the Risorgimento – onto individual operas (as some studies of opera and politics would have it) but from understanding how technologies of power effect institutional structures (agents, actors and networks) that support them and the ways in which such technologies of power can be manipulated by individuals, mostly acting as part of a complex network of influence and behaviour. An obvious example of the latter is Jacques Offenbach’s handling of the terms of his licence to create more and more ambitious works which would eventually lead to the creation of Orphée aux enfers in 1858 (this is set forth in Chapter 3), but the history of music in the nineteenth-century Parisian theatre is shot through with frequently successful attempts to circumvent Napoléon’s 1806–1807 legislation, is a continuous thread in Chapter 1 and recurs throughout this book. Offenbach’s attempts to circumvent the licensing system in the 1850s by invoking the music of the long eighteenth century also point to the preoccupation in the culture of music in the Parisian theatre with questions of memory and the past. The traditional name of the Opéra – the Académie Royale de Musique – changed several times over the period, but the word ‘académie’ only disappeared in 1854: the responsibility of the Opéra, then, was at least in part one of the preservation of the past – the establishment of a collective memory.12 This concern with the reminiscence of previous

10 Amadée Verger attempted to resuscitate the Théâtre-Italien in early 1872 but was declared bankrupt the following year. Various individuals attempted to keep the institution alive for a few fitful years: Maurice Strakosch, Prosper Bagier (who had managed the ­T héâtre-Italien for the last decade of the Second Empire) and even Léon Escudier, who oversaw its eventual demise in 1878. 11 See Nicole Wild, Dictionnaire des théâtres parisiens au xixe siècle: les théâtres et la musique (Paris: Amateurs des Livres, 1989), 339 and 238. 12 As understood in Paul Ricœur, La mémoire, l’histoire, l’oubli, Collection Points; Série Essais (Paris: Seuil, 2000), 94–99.

Ouverture  5 generations was not limited to the Opéra, however, and despite modern preoccupations with tracking the re-inscription of ‘great dates’ in specific musical works in the theatre, Parisian musico-theatrical cultures were as much concerned with their musical and theatrical pasts, as they were with remembering les trois glorieuses, the hundred days or even the revolution of 1789.13 For the Opéra in the nineteenth century, this involved a concern with those works that inaugurated the period of grand opéra between 1828 and 1836 – La muette de Portici, Guillaume Tell, Robert le diable, La juive and Les Huguenots; all these works remained more or less continuously in the repertory up to the Franco-Prussian war and the Commune, and most right up to 1900 and beyond. The Opéra-Comique looked back further as it recalled the past of opéra comique of both the ancien régime and the revolution. While some works – Grétry’s Guillaume Tell (1791) for example – only reappeared as a counterweight to Rossini’s opera of the same name at the end of the 1820s, such works as Dalayrac’s Adolphe et Clara (1799) were performed continuously until the 1840s, just at the point when Grétry’s ­Richard cœur de lion (1784) began a spectacular decade-long run. The licensing system indeed forced some opera houses to align the past with the present: the Odéon, for example, was only allowed to present public domain opéra comique in the 1820s, which meant that exactly that repertory of the ancien régime and revolution were promoted vigorously; the same was true at the Théâtre-­Lyrique in the 1850s, although the progressive loosening of the licensing system meant that they were quickly able to appropriate new works in the genre. Even in the 1860s, as houses devoted to opérette multiplied, the past could be reimagined, as it was at the Théâtre des Fantaisies-Parisiennes from 1865 onwards, by performances of works from the 1820s onwards.14 The same was true at the Opéra-Comique and Théâtre-Lyrique; at the former house, Boieldieu’s La dame blanche (1825), for example, never left the stage for the entire century, to say nothing of its worldwide impact.15 Much of the contest and competition that resulted from the licensing system may be explained by an analysis of power. The licensing system, as discussed here, is a clear example of what Michel Foucault called a ‘technology of power’, and like all technologies of power, it is subject to Foucault’s claim that ‘power is everywhere’. In addition to those who deployed obvious technologies of power, it also resided in the hands of the creative 13 The 1789 revolution is a regular touchstone for studies of grand opéra in particular. See ­Sarah Hibberd, French Grand Opera and the Historical Imagination (Cambridge: ­Cambridge University Press, 2009). 14 See Louis-Henry Lecomte, Les Fantaisies-Parisiennes. L’Athénée. Le Théâtre Scribe. L’Athénée-Comique (1865–1911), Histoire des théâtres de Paris 10 (Paris: Daragon, 1912; R Geneva: Slatkine, 1973), 1–62 and 77–136. 15 For an outline of the astonishing geographical distribution of La dame blanche, see Alfred Loewenberg, Annals of Opera: 1597–1940, 3rd edn revised and corrected (London: Calder, 1978), 698–700. The opéra comique had been the first stage work to be performed in Jakarta (Batavia), on 10 October 1836 (ibidem, 700).

6  Ouverture agents – librettists, composers, choreographers, set designers and so on – as well as with those who held different forms of power – censors, reading committees, managers – and with the press and the audiences for whom, in principle at least, they spoke. Many of the essays in this book are framed by the idea that these different constituencies regularly interlocked, networked and contested in ways that left the technologies of power subtly altered in ways with which the legislative process could rarely keep up. Principle and practice came into conflict within a very few years of the original legislation, and Chapter 1 disentangles the networks of power that allowed the gradual unravelling of the system. The repeal of the legislation in 1864 and the proclamation of the so-called liberté des théâtres were the result of a system having been manipulated from the earliest years of the legislation’s existence to a point in the late 1850s when it was in danger of collapse. Opéra comique was at the centre of progressive attacks on the system – from the Odéon in the 1820s, from the Théâtre de la Renaissance in the late 1830s and especially from the Théâtre-Lyrique during most of the Second Empire. Tensions between the Opéra-Comique and the Théâtre-Lyrique ought to have been stronger than they actually were, but composers moved effortlessly from one to the other and back again, and Chapters 4 and 5 consider two composers – Francois-Auguste Gevaert and Xavier Boisselot – and the tools they used to dismantle the technologies of power that had been in place for between thirty and forty years, as they traded opéra comique between the theatre of the same name and the Théâtre-Lyrique. Opérette, emerging as it did in the last decade of the licensing period, was able to take off so fast because of the ease with which Offenbach, Hervé (Florimond Ronger) and others were able to break the letter of the law while respecting enough of its spirit to avoid the institutional tools of power that were designed to control them. Music was endemic in the nineteenth-century theatre, and even in those institutions which gave a home to what is variously called ‘straight’, ‘legitimate’ or ‘regular’ theatre, music played an important part, not merely as ‘incidental’ music at such theatres as the Théâtre Français (Comédie Française) or the Théâtre de l’Odéon (only an ‘opera house’ during the 1820s and the subordinate to the Théâtre Français for the rest of the century), but as a key element in such genres as mélodrame and comédie-vaudeville, that coexisted alongside grand opéra, ballet-pantomime, opéra comique, petit opéra, Italian opera and others. Each of these genres takes a different approach to the relationship between music, drama and literary register. Opéra comique, for example, is best understood as a play into which are introduced arias, duets, trios and ensembles and as an alternating discourse of spoken dialogue and newly composed music. Broadly speaking, over the course of the long nineteenth century, more and more kinetic action is moved into the domain of composed music – so that the component of spoken dialogue is reduced and replaced by accompanied recitative or mélodrame, but literary register varies a good deal from the farcical to the tragic (Bizet’s Carmen

Ouverture  7 is an opéra comique), with the largest proportion sitting in a less easily definable, semi-serious register. Comédie-vaudeville – with a much narrower dramatic register, focussed on what today would be called the comedy of manners – is a play with music, but where the music is mostly borrowed; the mix of borrowed music in Les erreurs du bel âge is typical for this genre in the nineteenth century, for example. Mélodrame, on the other hand, is most obviously characterised by musical interventions that underscore both the action and the dialogue in the play; it also makes use of pantomime and various sorts of dance. Central to the activities of the Paris Opéra, dance had been an integral part of tragédie lyrique as well as grand opéra; the so-called divertissement began as a series of scattered formal dance numbers in grand opéra, but over time coalesced into a single formal section, frequently large enough to dominate a single act. But the repertory of the Opéra was divided between these types of grand opéra and ballets nobles et gracieux. Together, these works were thought to encompass the entire repertory of the institution, but it has become clear that – after the emergence of grand opéra – it was increasingly difficult to perform both ballet and opera on the same evening because the operas had become so long; this resulted in the emergence of what was known as petit opéra, specially written to accompany a ballet in a single evening (Chapter 2 of this book is dedicated to the problem, its solutions and the emerging repertory of petit opéra), and in two short acts or one longer one. Formal divertissements were entirely absent, but a close reading of the surviving stage manuals for both petit opéra and opéra comique shows that dance was frequently introduced incidentally in numbers where it was indicated neither by the libretto nor by the score.16 If institution and genre are two of what Bruno Latour would call ‘actors’ in this complex network of activity, they can only be understood as part of an interlocking network of human agents,17 of which librettists, composers, impresarios and singers are all key members. The order is important, and the current preference for privileging the vocal artist above other agents is

16 This work is in its infancy, but see the section on dance in the account of Gevaert’s works for the Parisian stage, 159–163, and Mark Everist, ‘Choriste, danseur, comparse: Dance and Movement in opéra comique during the July Monarchy and Second Empire’, paper read at conference, ‘Opera as Spectacle’, London, June 2012. 17 Bruno Latour, Reassembling the Social: An Introduction to Actor-Network Theory, ­C larendon Lectures in Management Studies (Oxford and New York, NY: Oxford University Press, 2005), 54–58. Latour’s distinction between ‘actor’ and ‘actant’ is not clearly maintained even in this short passage of definition and is developed at greater length in idem, ‘Where Are the Missing Masses? The Sociology of a Few Mundane Artifacts’, in Technology and Society, Building Our Sociotechnical Future, eds. Deborah J. Johnson, and Jameson M. Wetmore (Cambridge, MA: MIT Press, 2008), 165–166. For human agents, in the sense in which they are understood here, see Howard Becker, Art Worlds (Berkeley: University of California Press, 1982; 25th anniversary edition, updated and expanded, 2008).

8  Ouverture one that is hard to square with the major shifts in cultural practice that the essays in this book describe and explain. Singers played no role in the emergence or the abolition of the licensing system; grand opéra grew out of a set of preoccupations largely driven by Eugène Scribe – usually called a ‘librettist’ in this context – but much better known to the wider world as an author of comédies-vaudevilles; opérette emerged with little reference to its performers; and the cultural exchange between German and Italian works and Paris, and between French works and the rest of the world, was similarly largely independent of singers’ work.18 While the emphasis given the prima donna or diva is central to modern close readings of any work for the nineteenth-century stage, the attempts made in this volume to distance the essays from such close readings result in a lesser focus on singers and a greater emphasis on other agents. A reluctance to align with the close reading of a small number of canonic works has an effect on methodology and the use of sources. Almost all of the material in the essays in this volume is newly recovered, as befits the novelty of the questions being posed. And that results in a vast amount of material consulted for the studies in question. For example, work on Chapter 1 entailed reading something around 5,000 archival documents, of which over 200 were formally transcribed and less than fifty were cited in the text. But yet another study of Verdi’s Falstaff, say, can take most of this groundwork for granted; in this volume, the infrastructure has to be rebuilt, piece-by-piece. In short, methods and sources are inexorably linked when the reconstruction of cultures and positions of individuals within them are central, and nowhere is this clearer than in the understanding of the institutional technologies of power that underwrite Parisian music in the nineteenth-­century theatre. This is where this book begins.

18 The Théâtre-Italien is an obvious exception, as is Hortense Schneider’s role in Offen­ bach’s successes in the 1860s. Some of the major works at the Opéra in the late 1850s were dictated by the presence or recruitment of single artists, but this was a rarity; see Mark ­Everist, ‘The Empire Strikes Back: Politics, Power and Music in Second Empire Opera (forthcoming).

1 The music of power Parisian opera and the politics of genre, 1806–1864

Music for the stage may be understood in terms of institution, power and genre. Nowhere is this truer than in the kaleidoscopic range of music drama produced in nineteenth-century Paris during the period 1806–1864, when institution and repertory were aligned according to a system of licences. ­Nineteenth-century Parisian stage music has been no stranger to the analysis of relations between state power and cultural artefact, and many works produced at the Paris Opéra during the middle third of the nineteenth century have been interpreted in a number of ways, ranging from Jane ­Fulcher’s attempts to read works in the light of contemporary historical events to Sarah Hibberd’s view that the historical events in such works as Auber’s La muette de Portici and Gustave III should be viewed more as a retelling of modern post-revolutionary history.1 Other examples of similar concerns have been Benjamin Walton’s interpretation of Rossini’s Le siège de Corinthe as a response to, and engagement with, Parisian philhellenic practices; Diana Hallman’s analysis of the anti-Semitic in Halévy’s La Juive; and Mary Ann Smart’s reading of Donizetti’s Don Sébastien as an act of mourning for the recently deceased Duc d’Orléans.2 Anselm Gerhard’s The Urbanization of Opera, perhaps for the first time, invited its readers to consider grand opéra in terms of its urban reception as the basis for readings of the classics of the

1 Jane Fulcher, The Nation’s Image: French Grand Opera as Politics and Politicized Art (Cambridge: Cambridge University Press, 1987); Sarah Hibberd, French Grand Opera and the Historical Imagination (London: Cambridge University Press, 2009). The Nation’s Image may be productively read in conjunction with Steven Huebner, Review of The Nation’s Image: French Grand Opera as Politics and Politicized Art, by Jane F. Fulcher, Music & Letters 70 (1989), 114–118 and Hugh MacDonald, ‘… and Politics’, Review of The Nation’s Image: French Grand Opera as Politics and Politicized Art, by Jane F. Fulcher, The Musical Times 129 (1988), 405–406. 2 Benjamin Walton, Rossini in Restoration Paris (London: Cambridge University Press, 2011), 108–153; Diana R. Hallman, Opera, Liberalism and Antisemitism in Nineteenth-­Century France: The Politics of Halévy’s La Juive, Cambridge Studies in Opera (London: Cambridge University Press, 2002); Mary Ann Smart, ‘Mourning the Duc d’Orléans: Donizetti’s Dom Sébastien and the Political Meanings of Grand Opera’, Reading Critics Reading, eds. Mary Ann Smart and Roger Parker (Oxford: Oxford University Press, 2001), 188–212.

10  The music of power Opéra, and Cormac Newark projects the same repertory across the history of the novel in the nineteenth century.3 The current study complements this work in two ways. First, it considers that an account of Parisian stage music based exclusively on the works premiered at the Opéra presents at best an incomplete picture of the operatic culture of nineteenth-century Paris.4 Second, it assumes a view of state power, and the works engendered by that power, less as a set of hierarchical pressures and more as a network of regulations, practices and negotiations. These two scholarly trajectories contribute to the analysis of the politics of genre.5 Aligning a wider institutional perspective on nineteenth-century stage music with a Foucauldian approach to the ways in which power structures interact – acknowledging his view that ‘power is everywhere’ – comes closer to a regime of truth that explains the wealth and complexity of the culture of stage music in nineteenth-century Paris.6

3 Anselm Gerhard, Die Verstädterung der Oper: Paris und das Musiktheater des 19 Jahrhunderts (Stuttgart: Metzlersche J.B. Verlagsb, 1992); Cormac Newark, Opera in the Novel from Balzac to Proust, Cambridge Studies in Opera (London: Cambridge University Press, 2011). 4 This is not to suggest that secondary theatres have been absent from contemporary scholarship. See, for example, the study of works based on Sir Walter Scott in Paris between the 1820s and 1840s in Hibberd, French Grand Opera, 21–26, and her work on dream phenomena on the Parisian stage (eadem, ‘“Dormez donc, mes chers amours”: Hérold’s “La Somnambule” (1827) and Dream Phenomena on the Parisian Lyric Stage’, Cambridge Opera Journal 16 (2004), 107–132), both of which draw on ballet-pantomime and comédie-­ vaudeville and the theatres that promoted them as well as grand opéra and opéra comique. Other parts of the culture have been studied in their own right: the ­T héâtre-Italien’s history during the Restoration is the subject of Janet Johnson, ‘The Théâtre Italien and Opera and Theatrical Life in Restoration Paris’, 3 vols [paginated consecutively] (PhD diss., University of Chicago, 1988), and that of the Opéra-Comique during the same period is examined in Oliver Bara, Le Théâtre de l’Opéra Comique sous la Restauration: enquête autour d’un genre moyen, Musikwissenschaftliche Publikationen 14 (Hildesheim, Zürich and New York: Olms, 2001). Beyond the Restoration, Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981) provides a documentary account of the work of the Théâtre-­ Lyrique from the end of the July Monarchy to the beginning of the Third Republic. 5 The concept of the politics of genre can be traced back to several sources. Here, it finds its point of origin in René Wellek and Austin Warren, Theory of Literature (London: Cape, 1949; 3rd rev. edn, 1966), 226, and further adumbrated by Jacques Derrida (‘The Law of Genre’, Critical Inquiry 7 (1980), 55–81) and Ralph Cohen (‘History and Genre’, New Literary History 17 (1986), 203–218) in the 1980s. 6 Michel Foucault’s comments on the dispersal of power across and around networks are glossed in terms of power having ‘the privilege of consolidating everything under its invincible unity’ and of being produced ‘from one moment to the next, at every point, or rather in every relation from one point to another’ (The History of Sexuality, trans. Robert Hurley (New York: Random House, 1978), 93). Power must be analysed, according to Foucault, as something which circulates, or rather as something which only functions in the form of a chain. It is never localised here or there, never in anybody’s hands, never appropriated as a commodity or piece of wealth. Power is employed and exercised through a net-like organisation

The music of power  11 Paris between the Empire and the Commune was home to a number of institutions – the Opéra,7 the Théâtre des Variétés, the Opéra-Comique, the Théâtre-Lyrique, among others – whose repertory, personnel, geographical location, audience and ambition were subtly distinguished.8 These organisations participated in complex networks that were subject to wide-ranging technologies of power: matrices of influence and benefit that changed not only according to government but also to personal and political preference.9

[emphasis added] (‘Two Lectures’, Power-Knowledge: Selected Interviews and Other Writings, 1972–1977, ed. Colin Gordon [Brighton: Harvester, 1980], 98). For a succinct account of Foucault’s views on the concept of the network of power, see Joseph Rouse, ‘Power/ Knowledge’, The Cambridge Companion to Foucault, ed. Gary Gutting (Cambridge: Cambridge University Press, 1994), 107–110. Foucault’s propositions on regimes of truth underpin not only the following paragraphs but the general thrust of this article. He concludes an interview entitled ‘Truth and Power’ with a written response: ‘“Truth” is to be understood as a system of ordered procedures for the production, regulation, distribution, circulation and operation of statements. “Truth” is linked in a circular relation with systems of power which produce and sustain it, and to effects of power which it induces and which extends it. A “regime of truth”’ (‘Truth and Power’, The Foucault Reader, ed. Paul Rabinow (New York: Random House, 1984), 74). 7 The study of the relationships between creative activity and institutional power in ­n ineteenth-century France has been bedevilled by a confusion between institution and physical location (the Salle Feydeau as opposed to the Opéra-Comique, for example). For reasons of clarity, institutions are referred to by their formal title in this study and not by the building they occupied. The only exception is that the term Opéra is used to encompass all the changes in title to which the organisation was subject during the period covered here: 1806–1864 (Académie Royale de Musique, Académie Impériale de Musique, Académie de Musique, Théâtre de l’Opéra, Théâtre de la Nation, Opéra-Théâtre de la Nation, Académie Nationale de Musique, Théâtre Impérial de l’Opéra). The genre opéra comique (lower-case) is distinguished from the opera house that mainly promoted it: Opéra-Comique. 8 The fundamental text for understanding the chronology, topography and administrative personnel of Parisian lyric institutions in the nineteenth century is Nicole Wild, Dictionnaire des théâtres parisiens au xixe siècle: les théâtres et la musique (Paris: Amateurs des Livres, 1989; 2nd edition, Lyon: Symétrie, 2012). The inventory in this work (31–431) is based on the second volume of eadem, ‘Musique et théâtres Parisiens face au pouvoir (­ 1807–1864)’, 3 vols (Doctorat d’État, Université de Paris IV, 1987). The ‘Introduction Historique’ to the Dictionnaire (9–27) is an abridgement of ‘Musique et théâtres Parisiens’, 1:1–31 (subsequent references to this work are to the 1989 publication). Important background sources for Wild’s study are Pierre Bossuet, Histoire administrative des rapports des théâtres et de l’état (Paris: Jouve, 1909) and idem, Histoire des théâtres nationaux (Paris: Jorel, 1909). 9 Technologies of power were identified by Michel Foucault as emerging in the seventeenth and eighteenth centuries and possessing a ‘concrete and precise character’, and they were designed to ‘undertake the administration, control, and direction of the accumulation of men’ (‘Truth and Power’, 66–67). The invocation of institutional networks with the ability to negotiate power is strongly reminiscent of the concept of the ‘artworld’ first constructed by Arthur Danto in 1964, and endorsed by George Dickie a decade later (Arthur Danto, ‘The Artworld’, The Journal of Philosophy (1964), 571–584; Dickie, Art and the Aesthetic: An Institutional Analysis (Ithaca, NY: Cornell University Press, 1974); see also Dickie’s revised position as expressed in idem, The Art Circle: A Theory of Art (Chicago, IL: Spectrum Press, 1997)). While the ‘artworld’ provided a context in contemporary art criticism for the distinction between art and non-art, it was subject to immediate criticism for its

12  The music of power Thus, the concept of genre as a contract between creator and performer that may be maintained, broken or subverted, and that functions so well for instrumental music, cannot entirely do justice to the variety of music-drama promoted in nineteenth-century Paris (comédie-vaudeville, grand opéra, ballet, opéra comique, petit opéra, mélodrame, opera buffa and opera semiseria) or to the culture that supported it. In short, while genre in instrumental music floats between competing forces (publishing, interpretation, instrumental technologies and performers, for example), genre in stage music cannot be fully understood independently of the institutional framework in which it was embedded.10 Power, as one recent commentator has remarked, was ‘at the centre of the French lyric system’.11 It was exercised not only by institutions and the state, but also by creative individuals who worked within, subverted, triumphed over, but could in turn be destroyed by, institutional networks.12 This chapter seeks to develop this axiomatic view and to examine the ­contexts  – of narrowness of concept (Anita Silvers, ‘The Artworld Discarded’, The Journal of Aesthetics and Arts Criticism 34 (1976), 441–454), but remains of value, immediately one moves outside the domain of art history, as a pointer to the importance not just of institutions but of the ways in which they interlock to create meaning. The first systematic attempt to engage with social questions and the ‘artworld’ was Howard Becker, Art Worlds (Berkeley: University of California Press, 1982), which had much in common with early Pierre Bourdieu (see the latter’s ‘The Production of Belief’, The Field of Cultural Production: Essays on Art and Literature, ed. Randal Johnson (Cambridge: Polity, 1993), 74–111 (based on essays originally written in 1977 and 1983); see also Martin Irvine, ‘Institutional Theory of Art and the Artworld’ (Accessed 20 July 2006) www.georgetown.edu/faculty/irvinem/ visualarts/Institutional-theory-artworld.html.). The emphasis in this study on such social issues as personnel, institution, audience and topography, as well as on more obvious sources of power, clearly owes much not only to Becker and Bourdieu but also to earlier attempts to engage with the concept of the ‘artworld’. 10 The clearest recent statements on genre in instrumental music as a series of negotiable contracts are found in Jefferey Kallberg, ‘Understanding Genre: A Reinterpretation of the Early Piano Nocturne’, Atti del xiv congresso della Società Internazionale di Musicologia: Trasmissione e recenzione delle forme di cultura musicale, eds. Angelo Pompilio, Donatella Restani, Lorenzo Bianconi and F. Alberto Gallo, 3 vols (Bologna: EDT, 1988), 2:775–779, and idem, ‘The Rhetoric of Genre: Chopin’s Nocturne in G Minor’, 19th Century Music 11 (1987–1988), 238–261. R Chopin at the Boundaries, Convergences: Inventories of the Present (Cambridge, MA.: Harvard University Press, 1996), 238–261. Key background texts to these statements, which also necessarily place their focus away from institutional networks, are Philippe Lejeune, Le pacte autobiographique, Collection Poétique (Paris: Seuil, 1975), and Heather Dubrow, Genre, The Critical Idiom 42 (London and New York: Methuen, 1982). 11 ‘Au centre du système lyrique français’ (Hervé Lacombe, ‘De la différenciation des genres: Réflexion sur la notion de genre lyrique français au début du xixe siècle’, Revue de musicologie 84 (1998), 260). 12 According to Clarissa Rile Hayward, ‘institutions shape what actors can do in particular social contexts, what they want to do, and the ways in which it is strategically rational for them to pursue particular aims, ends and interests’ (De-Facing Power (Cambridge: Cambridge University Press, 2000), 4). For a broader introduction to the ‘New Institutionalism’ that underpins this article, see Peter Hall and Rosemary Taylor, ‘Political Science and the Three New Institutionalisms’, Political Studies 44 (1996), 936–957, and Rogers Smith,

The music of power  13 ­ olitics, genre and power – in which nineteenth-century Parisian stage music p was produced and consumed. In doing so, it brings into question the premise that power is necessarily hierarchical and coercive and suggests an approach that views it as something that ‘comes from below’, that ‘bring[s] about redistributions, realignments … and convergences of the force relations’.13 In the context of nineteenth-century opera, the concept of power may be parsed in many different ways: the creative power of librettists, composers, managers, stage designers, costumiers and other agents who shaped the production of music drama; institutional power embodied in the network of organisations that allowed a dozen or more opera houses or theatres hosting stage music to coexist simultaneously in the same city; and the power of audiences, the censors and the press. Many examples – the emergence of a style of through-composed opera in five acts on a historical theme around 1830, the expansion of the scope of opéra comique from the late 1840s onwards and the development of opérette in the 1850s – may be better understood when examined within the networks of institutional power as described by Foucault in writings from the 1970s, and analysed here. The relationship between opera and the state ranges from a situation where creative power could be exercised almost without reference to institutional power – where artists could allow their aesthetic ambitions to take them where they wished within the limits of existing structures – to instances where institutional power could override creative power completely or at least force different compromises. In extreme cases – such as its attempt to create new genres at the Théâtre de la Renaissance in the late 1830s – the state could effectively suppress creative power by making decisions that affected the local practices of composers and librettists and by attempting to create genres that were by no means the product of the relationship between creative and institutional power. An alternative both to current understandings of opera in France in the nineteenth century and to the ways in which opera and culture interrelate more generally explains how technologies of operatic power – systems, mechanisms and regulators – might underpin future interpretative strategies. Such an account both encompasses the multiplicity of organisations that make up the polity in which power is negotiated, and effaces the distinction between creative and institutional power within which the creation of works and development of genres might be thought to be contained.14 ‘If Politics Matters: Implications for a “New Institutionalism”’, Studies in American Political Development 6 (1992), 1–36. 13 Foucault, History of Sexuality, 94. For a further analysis of the concept of ‘alignment’, see Thomas E. Wartenberg, The Forms of Power: From Domination to Transformation (Philadelphia: Temple University Press, 1990), 150. 14 The concept of the work is much less familiar for nineteenth-century France than it is at the beginning of the twenty-first century, and recourse to such – by now – well-known concepts as Werktreue is insufficient (see Lydia Goehr, The Imaginary Museum of Musical Works: An Essay in the Philosophy of Music (Oxford: Clarendon, 1992), and the

14  The music of power

Technologies of power The imperial rationalisation of Parisian opera houses and theatres, undertaken by a series of decrees and ministerial rulings during 1806 and 1807, changed the face of Parisian cultural life for over half a century. This regulatory framework was a vivid embodiment of the ‘administration and control’ animated by what Foucault identified as technologies of power. In response to the wide range of theatres spawned by the theatrical liberty unleashed by the revolution of 1789, the emergent empire classified eight institutions into four grands théâtres and four théâtres secondaires:15 the Opéra, the Comédie-Française, the Opéra-Comique, and what would shortly become the Théâtre-Italien, in the first group, and the Théâtre de la Gaîté, Théâtre de l’Ambigu-Comique, Théâtre des Variétés and Théâtre du Vaudeville in the second. Each institution was assigned a genre, largely based on existing practice, and all other theatres were ordered to close.16 The ‘technology’ remained in place until its repeal in 1864 and apparently controlled the central focus of European consumption and reception of stage music throughout the first half of the nineteenth century. cascade of texts that it prompted: among others, Horst-Peter Hesse, ‘Musikwerk und Aufführung: Reflexionen zum Begriff der Werktreue’, Aspekte historischer und systematischer Musikforschung: Zur Symphonie im 19. Jahrhundert, zu Fragen der Musiktheorie, der Wahrnehmung von Musik und Anderes, eds. Christoph-Hellmut Mahling and Kristina Pfarr, Schriften zur Musikwissenschaft 5 (Mainz: Are Musik Verlag, 2002), 517–522, and Hermann Danuser, ‘Werktreue und Textreue in der musikalischen Interpretation’, Europäische Musikgeschichte, eds. Giselher Schubert, Sabine Ehrmann-Herfort and Ludwig Finscher, 2 vols (Kassel: Bärenreiter; Stuttgart: Metzler, 2002), 2:1115–1165). The concentration on German nineteenth-century instrumental music is clear. Even in Horst Weber (ed.), Oper und Werktreue: Fünf Vorträge, Metzler Musik (Stuttgart: Metzler, 1994), the subjects are J.C Bach, Berg and Gluck. The only French example concerns modern stagings of Offenbach’s Les contes d’Hoffmann. While the textual instability of music drama in nineteenth-century Paris goes without saying, understanding this fluidity has to be tempered by the contemporary view that works were as much identified by their original subject matter as by their immediate form. 15 See the list of 51 theatres functioning with a licence by 1795, given in Hippolyte Hostein, La liberté des théâtres (Paris: Librairie des auteurs, 1867), 51–52. 16 The enactment of this legislation was complex. Napoléon’s decree which set out the distinction between grands théâtres and théâtres secondaires was promulgated on 8 June 1806 (Décret concernant les théâtres), and the règlement followed logically ten months later on 25 April 1807: Ministère de l’Intérieur. However, it is clear that this legislation – especially as it related to the identification of which theatres were secondaires, and particularly the fate of the unauthorised theatres – was met with imperial disapproval. It took a second décret from the emperor for the full impact on the rest of theatrical Paris to be felt: ‘All unauthorised theatres shall be closed by 15 August [1807]’ (‘Tous les théâtres non autorisés par l’article précédent seront fermés avant le 15 août [1807]’) (Décret impérial sur les théâtres, 2). A summary of the legislation is in Wild, Dictionnaire, 13–14; the full text was published and republished, throughout the licensing period and beyond, each time the law was subject to review and reinterpretation. See, for example, Edmond-Adolphe Blanc and Auguste Vivien, Traité de la législation des théâtres, ou Exposé complet et méthodique des lois et de la jurisprudence relativement aux théâtres et spectacles publics (Paris: ­Brissot-Thivars, 1830).

The music of power  15 It was an act of unbridled political power, and one that tolerated no compromise, no appeal. The Empire immediately established a series of structures to support the newly drawn map of Parisian opera as readily as it had redrawn the map of Europe from Austerlitz to Wagram. Central to imperial control of opera and theatre were the licenes granted to each, signed by the Emperor, in which the specific set of genres associated with the institution was identified, described and delimited. In the knowledge that more public money was spent on subventions to stage music and theatre in Paris than anywhere else in the world, the state was at pains to ensure that the complex network of lyric institutions was not hobbled by unnecessary overlap and dispersal of resources. Thus, the Opéra was the single institution permitted to produce ballet and music drama entirely in music (by which was meant continuous opera), the Théâtre-Italien works exclusively in Italian, the Comédie Française tragedy and the Opéra-Comique ‘all sorts of comedy and tragedy mixed with couplets, ariettes and ensembles’.17 Although ensuring discrete repertories at each of the secondary theatres was not as important as at the grands théâtres since none were in receipt of a state subsidy, the Gaîté was restricted to pantomimes and harlequinades and the other three theatres to vaudeville; in its commingling of spoken dialogue and music (importantly based on what were called airs connus, or pre-existent music), vaudeville was the closest point of contact – and hence the strongest competition – with opéra comique which mixed newly composed dialogue and music.18 This attempt at ensuring that single genres were assigned to particular institutions has recently been theorised into an opposition between co-répertoire – where genres are kept carefully distinct – and contre-répertoire – where genres overlap; in the first case, the licensing system is assumed to function without serious difficulty, whereas in the second, contre-répertoires have the effect of bringing the individual institutions within the network that constitute the licensing system into conflict in a way that may only be resolved through the exercise of power.19

17 ‘Toute espèce de comédies ou drames mêlés de couplets, d’ariettes et de morceaux d’ensemble’ (Ministère de l’Intérieur: Règlement pour les théâtres [25 April 1807], 2). 18 Ibidem, 3–4. 19 For the concepts of co-répertoire and contre-répertoire, see Maud Pouradier, ‘Opéra, Comédie-Française et Opéra-Comique: entre co-répertoires et contre-répertoires’, L’Opéra de Paris, la Comédie-Française et l’Opéra-Comique: approches comparées (1669– 2010), eds. Sabine Chaouce, Denis Herlin and Solveig Serre, études et rencontres de l’École des Chartes 38 (Paris: École des Chartes, 2012), 133–143. Useful for understanding the background to Pouradier’s approach is her ‘Le répertoire: enjeux musicaux d’une notation théâtrale (1680–1840)’, Le répertoire de l’Opéra de Paris, 1671–2009: analyse et interprétation, eds. Michel Noiray and Solveig Serre, études et rencontres de l’École des Chartes 32, 37–52 (Paris: École des Chartes, 2010), 37–52. The strength of Pouradier’s theoretical engagement lies more in its power as heuristic tool to examine specific source-based instances of the deployment of operatic power rather than in its ability to explain the complexity of a single incident, sequence of incidents or historical continuum. Its heuristic precepts underpin much of what is discussed in this article.

16  The music of power Alongside the eight theatres under central government control were a host of other entertainments that were controlled by the Prefect of Police. Ten years after the enactment of the Napoleonic legislation, there were five petites spectacles, eleven soirées amusantes (all of which involved music) and ­thirty-nine grands bals en exercice, which included the exhibition of a crocodile, optical illusions, ombres chinois, dance-halls, automata and waxworks.20 By and large, there was little competition between these entertainments and the four grands théâtres, but any mix of music and dialogue at the former could call down the wrath of any or all of the secondary theatres. The Prefect of Police also had an important supervisory role in theatrical activities, ranging from the Paris première of Tannhäuser in 1861 to the display of an elephant on the Boulevard Bonne-Nouvelle in 1826: maintaining public order and ensuring the safety and security of all the theatrical buildings in the city.21 He was able to shut down any performance if he considered the threat to the public excessive, and even the Opéra was not immune to such action, as when a performance of La muette de Portici was stopped on 6 February 1843, not because of any revolutionary qualities embodied in the work itself but because G ­ ilbert Duprez, who had been the Masaniello of choice since the 1840 revival, was not available, nor was his understudy, Claude-­Marie-Mécène Marié de l’Isle; the third in line, Placide-Alexandre-Guillaume Poultier, was sick. The fourth choice for the role that evening, Raguenet (only his family name has survived), met with so little approval that his performance triggered a riot that was only contained by the Prefect of Police clearing the building with the aid of armed force.22 The transfer of power from the state to the opera house was effected via the formal licence (privilège) which specified the duration of the manager’s tenure, his repertory, financial terms (especially if the licence involved a

20 During the second half of 1816, the Prefect of Police and the Minister of the Interior conducted an inquiry into the threats posed to the Opéra-Comique by such petits spectacles. Among the surviving documents is the État des théâtres secondaires, des petits spectacles, bals, soirées amusantes, curiosités et marionettes ambulantes, 1 October 1816 (Paris, ­A rchives Nationales (hereafter F-Pan) F21 1045), from which this account is taken. 21 The responsibilities of the police during the Restoration were outlined in Préfecture de Police: Ordonnance concernant la police intérieure et extérieure des spectacles, Paris, le 7 janvier 1818 (Paris: Lottin, 1818) supplemented by Préfecture de Police; Instruction pour messieurs les commissaires de police, officiers de paix, officiers de la gendarmerie royale de Pairs, et adjudants, relativement à la surveillance aux théâtres, Paris, ce 12 janvier 1818 (Paris: Lottin, 1818). 22 Report by Auguste Louis Joseph Casimir Gustave de Franquetot (1788–1865) Comte de Coigny, president of the Commission des théâtres, to Minister of the Interior, 10 ­February 1843, in reply to letter from the Prefet de Police (Gabriel Delessert) to the Minister (Charles Marie Tanneguy, duc de Duchâtel), 7 February 1843 (F-Pan AJ13 180 (X)). Fulcher (The Nation’s Image, 101) cites an extract from Delessert’s letter to Duchâtel, but goes on to argue, against all the evidence, that ‘the Opéra seemed to be becoming an arena where political tensions could be given expression through indirect, hence legally acceptable, symbolic means’ (ibidem, 101–102).

The music of power  17 23

state subvention) and other matters. From the early 1830s onwards, the licence took on a more perfunctory form and dealt simply with the legal status of the manager’s contract while questions relating to repertory and the day-to-day functioning of the opera house were contained in the cahier des charges.24 Here were detailed the nature of the repertory, the number of performances per year, the length of the season, the proportion of new work that the manager was obliged to showcase and, frequently, the nature of the performing forces themselves. The cahier des charges could also become a site of negotiation where the manager could seek to nuance the stipulations of the state to his own advantage.25 A particular feature of the management of operatic culture during the Bourbon Restoration, and one that created significant levels of rancour, was the redevance des théâtres secondaires, a levy on all the non-royal theatres at a rate of 5% on all receipts that was to be paid to the Opéra alone. This was instituted in 1811, and was unsuccessfully challenged in the courts between 1826 and 1828, but while the collection of the levy was supposed to be carried out by the Prefet de la Seine, he was frequently slow or reluctant to forward the money to the Opéra and also failed to punish those institutions that failed to pay. The sums that came to the Opéra, while an important contribution to its budget, were far smaller than the 300,000 francs per annum described by the Opéra’s manager, Louis-Désiré Véron.26 The secondary theatres refused to pay after the July Revolution in 1830, and Louis-Philippe abandoned the levy the following year.27 A picture of Parisian opera houses functioning efficiently within the bureaucratic structures of government was never a complete one during the licensing period; with very few exceptions, and from the beginning of the July Monarchy almost exclusively, an intermediary tier of management was installed, and it was here that much of the brokering of operatic power took place. The Commission des théâtres was central to the cartography of Parisian operatic life throughout the July Monarchy, Second Republic and the Second Empire. Even as early as 1817, the secondary theatres were beginning

23 See, for example, the licences for the Théâtre Royal de l’Odéon in Mark Everist, Music Drama at the Paris Odéon (Berkeley, Los Angeles and London: University of California Press, 2002), 44–53. 24 The first cahier des charges in which most of the detail from the formal licence appeared was the one granted Véron in February 1831, and immediately revised in May 1831 and again in May 1833 (F-Pan F21 4655(2)). 25 From 1831, any change of manager at a grand théâtre or théâtre secondaire would always be accompanied by a new cahier des charges, which would almost invariably entail a renegotiation of its terms. 26 Louis-Désiré Véron, Mémoires sur l’affaire du Constitutionnel, sur la transaction Mirès; mes relations avec la famille Aguado (Paris: Dupont, 1853), 1:164. 27 For a definitive account of the redevance des théâtres secondaires, see John Drysdale, Louis Véron and the Finances of the Académie Royale de Musique, Perspektiven der Opernforschung 9 (Frankfurt-am-Main: Peter Lang, 2003), 36–38.

18  The music of power to take too liberal a view of relationship between vaudeville and opéra comique with the result that officials within the Maison du Roi were beginning to notice that receipts at the Opéra-Comique were dropping. The consequence, according to Jules Jean-Baptiste François de Chardebœuf, Comte de Pradel, would be ‘the rapid decline in French dramatic art’ for which he blamed ‘the excessive number and the overly generous scope of licences’.28 The Opéra-Comique had been complaining about the way in which secondary theatres were breaking the terms of their licences and consequently i­ mpinging on its repertory almost as soon as the 1806/1807 legislation had been enacted, but a decade later, the Maison du Roi acted by setting up a Commission des théâtres, made up of five members of the Institut de France, with the task of identifying the reasons for the décadence de l’art dramatique et théâtral and finding ways of remedying it. The members were François-Juste-Marie Raynouard, François-Nicolas-Vincent Campenon, Henri-­Montan Berton, Antoine-Chrysostome Quatremère de Quincy and Hippolyte Auger.29 The commission fell apart because the Minister of the Interior considered that it ‘is useless. It would only be called on to debate something that is already judged… If there are attempts at trespass at the Vaudeville, at the Variétés, on the Boulevard, they will be immediately suppressed’.30 Given that the Commission des théâtres, in one form or another, was to be so important from the early 1830s until the end of the licensing period, it is reasonable to ask why such an idea was so quickly abandoned. Difficulties must have arisen because the grands théâtres were at that point under the supervision of the Maison du Roi, whereas the secondary theatres were loosely under the control of the Minister of the Interior. This meant that any disagreements between any of the secondary theatres and, for example, the Opéra-Comique, involved two government departments, each with their own administrative structure and their own minister. Under such circumstances, it comes as no surprise that the Maison du Roi would be enthusiastic about a commission to protect its own interests, whereas the Ministry of the Interior would have had more confidence in the existing system of licences to ensure the regulation of the opera houses and theatres of the capital. No supervisory committee was in place during the Restoration, a situation that was ameliorated, but by no means rendered satisfactory, by the fact that the Théâtre-Italien was placed under the same management

28 ‘La prompte décadence de l’art dramatique en France’; ‘le trop grand nombre et la trop grande étendue des privilèges’. Letter from Jules Jean-Baptiste François de Chardebœuf, Comte de Pradel to the Minister of the Interior, Joseph-Henri-Joachim Lainé, 5 August 1817 (F-Pan F21 1045). 29 Letter from the Comte de Pradel to Minister of the Interior, 22 September 1817 (F-Pan F21 960). 30 ‘est inutile. Elle ne serait appelée qu’à débattre une chose déjà jugée … Si des empiètements étaient tantés au Vaudeville, aux Variétés, au Boulevard, ils seraient réprimés aussitôt’ (F-Pan F21 960, cited in Wild, ‘Musique et théâtres’, 1:41).

The music of power  19 31

as the Opéra between 1819 and 1827. At the same time, however, the ­Opéra-­Comique was having to fight off competition for the genre that gave the institution its name from the newly titled ‘Théâtre-Royal de l’Odéon, Second Théâtre Français’.32 Because these often bitter arguments over which institution was permitted to play which parts of the repertory were confined to two royal theatres, they could therefore be settled by recourse to the M ­ aison du Roi alone. The opening of new theatres (by 1827, these included the Gymnase-Dramatique, the Théâtre de la Porte Saint-Martin and the Théâtre des Nouveautés), however, resulted in various challenges to the royal theatres that the 1806–1807 legislation had never envisaged. Louis-François Sosthène de la Rochefoucauld, in a ministerial arrêté, formalised a practice that had emerged in the legislative vacuum during the previous decade: individual directors and administrators of royal theatres were expected to watch over ‘secondary theatres [who] give themselves permission daily to overstep the generic boundaries that are assigned them’.33 While such practices had been common for most of the Restoration, formalising managers’ responsibility to inform government of such transgressions of licence ensured that this method of surveillance would continue throughout the licensing period. A Commission de surveillance établie près le Conservatoire et l’Académie royale de Musique was established in early 1831, at the point when the management of the Opéra was given to the directeur-entrepreneur, Véron. Thought to be necessary under these particular circumstances, the commission consisted of Pierre-Paul Royer-Collard (replaced when he became Chef du Division des Beaux Arts by Auguste Hilarion, comte de Kératry), Edmond Blanc, Armand Bertin, Jean-Baptiste-Roger Fauchon d’ Henneville, with Claude Antoine Gabriel, duc de Choiseul-Stainville as chair and Edmond-Ludovic-Auguste

31 Wild, Dictionnaire, 197. 32 Given the status of a royal theatre in 1818, by 1824 it had received a licence to play certain segments of the opéra comique repertory, and this put it in direct conflict with the Opéra-Comique itself. See Everist, Music Drama at the Paris Odéon, 199–225. 33 ‘théâtres secondaires [qui] se permettent journellement de sortir du genre qui leur a été taxativement assigné’. Managers were required to report institutions which ‘allowed themselves especially frequently to perform plays mixed with romances, ariettes, duets, trios, ensembles and choruses, whether newly composed or borrowed from scores belonging to works composed for the Opéra, Opéra-Italien or Opéra-Comique’ (‘se permettent surtout de jouer fréquemment des pièces mêlées de romances, ariettes, duos, trios, morceaux d’ensemble et chœurs, soit de musique nouvelle, soit empruntés des partitions appartenantes aux pièces composées pour le grand opéra français, l’Opéra Italien et l’Opéra Comique’). They were further enjoined to ensure that ‘librettists and composers of the latter works were not given permission to play or sing the airs or their accompaniments at any other theatre in the Capital’ (‘les auteurs et compositeurs des dites pièces ne pourront donner aucune autorisation d’en jouer ou chanter les airs et accompagnements sur aucun autre théâtre de la capital’). Arrêté promulgated by Louis-François Sosthène, Vicomte de La Rochefoucauld, 29 March 1827, and countersigned by the Chef du Département des Beaux Arts, François-Henri Hilaire, Comte de Tilly (F-Pan AJ13 109 (I)).

20  The music of power Cavé as secretary.34 At first, their work exclusively concerned the Conservatoire and Véron’s management of the Opéra. When Véron resigned in 1835 to be replaced by Charles-Edmond Duponchel, the committee was reconstituted to encompass all the royal theatres as well as the Conservatoire. With a new title of Commission spéciale près le Conservatoire et les théâtres royaux, it was a mediating force between the Minister of the Interior (to which all theatres were now subject) and the managers of the opera houses and theatres responsible to the ministry. The members of this committee wielded immense power and could arbitrate, almost without appeal, on all matters brought to their attention by either the minister or a manager of one of the institutions.35 The remit of the commission, even before it took over the supervision of all the royal theatres, was broad. It was responsible for reviewing candidates for directorships of opera houses and theatres in the capital; drafting of cahiers de charges and their renegotiation on managers’ behalf; approving libretti and synopses of ballets; receiving requests from the Ministry of the Interior, managers and third parties for action and report; and requests for retirement. Its meetings frequently involved interviews with managers, artists and representatives from Société des Auteurs et Compositeurs Dramatiques that frequently provided its consideration of requests and reports.36 In the context of drafting Léon Pillet’s cahier des charges at the Opéra in 1841, for example, it effectively defined those works that fell into the canon and those that fell out of it; the removal of, for example, Gluck, Sacchini, Salieri, and almost all Spontini, gave Pillet a repertory largely made up of works composed in the previous decade and – temporarily at least – severed the link with the past that had characterised the Académie Royale de Musique since the seventeenth century.37 The uncompromising minutes of their meetings and reports written to the Ministers of the Interior commented on

34 See the minutes of the Commission’s meetings from 1 March 1831 to 31 August 1835, in F-Pan F21 4633 (3–4). 35 A case in point concerned the commission’s revision to Pillet’s contract in early 1841, the wording of which was queried by the minister on 18 March. The Duc de Coigny’s chilly, and well-documented, response written on behalf of the commission shows quite clearly where he thought the balance of power resided, and the absence of any counter-response from the minister shows that he shared that view (F-Pan AJ13 180 (III)). 36 Although based on such secondary sources as Bossuet, Histoire administrative, Fulcher’s account of the power of the commission (The Nation’s Image, 57–64) is a fair reflection of its activities as fully documented both before 1835 and from 1836 onwards (for the latter, see F-Pan F21 4633 (5–7) and F-Pan F21 4634 (1–2)). It is difficult to agree entirely with Cras’ complaint that Fulcher claims for the commission ‘une autorité bien plus grande qu’un examen scrupuleux des sources ne permet, à mon sens, de la faire. Ainsi, elle illustre le pouvoir exercé par cet organisme à l’aide d’exemples pris dans les premiers mois de l’année 1831, à une époque où Véron n’avait pas encore pris la direction de l’Opéra’ (Anne-Sophie Cras, ‘L’exploitation de l’Opéra sous la monarchie de Juillet’ (PhD diss., École des Chartes, 1996), 87–88), since Fulcher’s evidence clearly ranges widely across the July Monarchy. 37 Commission des Théâtres. Registre des Procès-Verbaux des délibérations, 15 January 1841 (F-Pan F21 4633 (7)).

The music of power  21 the quality of the performances, mises en scène, casting and, in some cases, the quality of the libretto and its music. The success of the commission was incontestable and so great that, at the 1848 Revolution, it was retained and given a remit for all Parisian theatres. An important consequence of the establishment and functioning of the Commission spéciale was that it also provided a setting in which artists’ associations could make representations to government, an environment that had been denied them during the Restoration. Principal among these was the Société des Auteurs et Compositeurs Dramatiques (SACD), the organisation that had been founded by Beaumarchais in 1777.38 In the 1840s, among its other activities, it campaigned vigorously against the Opéra-Comique, which it claimed was not fulfilling the obligations of its cahier des charges in relation to the production of new works.39 And while this campaign might have been thought to have backfired, the SACD’s petitions to government (9 March 1820, 2 June 1830, July 1837, 24 April 1840 and 12 November 1850) for a third lyric theatre were important voices in the clamour that eventually resulted in the establishment of the Théâtre-Lyrique.40 When the Commission spéciale had been given its wider powers in 1835, there had been concerns in the Ministry of the Interior that it might not be able to serve the interests of all parties simultaneously, so the Minister, Adolphe Thiers, appointed a Commissaire royal près l’Académie Royale de Musique, le Conservatoire de Musique, le Théâtre Royal Italien et le Théâtre Royal de l’Opéra Comique to supervise the managers of royal theatres. For the first five years, this was Pillet, who in 1840 became manager of the Opéra itself. He was replaced by Édouard Monnais, who served in this capacity throughout the rest of the July Monarchy, the Second Republic and most of the Second Empire. He was still acting as Commissaire impérial in 1864 when the licensing system was abolished, after nearly a quarter of a century in the post.41 The Second Empire brought great changes to the management of entertainment in the capital. In many respects, it marked a return to the past, 38 See Jacques Boncompain, La Révolution des auteurs: naissance de la propriété intellectuelle (1773–1815) (Paris: Fayard, 2002). 39 The Opéra-Comique’s cahier des charges stipulated that it should mount at least four opéras comiques in three acts each of which should be newly composed. The SACD began by observing that Crosnier, during the 1841–1842 year, had only mounted fifteen out of the obligatory twenty acts and – given that it represented the interests of those who might have written the remaining five acts – argued that he should be required to fulfil the terms of his cahiers des charges. The Commission spéciale considered this question and, when it came to its decision four years later, reached the conclusion that the cahier des charges should be modified so as to require the manager to play sixteen acts a year only. Letter from Commission dramatique des auteurs et compositeurs français to Commission des théâtres, May 1842 (F21 4673(4 g)). 40 Wild, ‘Musique et théâtres’, 1:156. 41 There is currently very little on Edouard Monnais. See Marie-Hélène Coudroy-Saghai, ‘Edouard Monnais, critique à la Revue et Gazette musicale’, Paper read at conference Music in France, 1830–1914, Melbourne, 5–7 July 2005.

22  The music of power separating out opera houses and theatres in receipt of a subvention from those who were not. The former were the responsibility of the newly created Ministre de la Maison de l’Empereur, which appointed commissaires for each of the main organisations.42 The institutions without subvention remained in the hands of the Ministry of the Interior. When the theatres without subsidy were brought under the control of the Ministre d’État in 1854, they effectively fell under the control of the same individual: Achille Fould, who held the positions of minister of state and minister of the Maison de l’Empereur. And while there had been nothing to equal the Commission spéciale des théâtres during the first two years of the Second Empire, Fould established a Commission supérieure permanente institutée pour l’examen des affaires relatives à la gestion du Théâtre Impérial de l’Opéra on 30 June 1854. The extraordinary seniority of its membership coupled to the fact that they were now responsible exclusively for the Opéra (now called the Théâtre Impérial de l’Opéra), and the intrusive nature of their involvement, indicate the importance the state attached to this institution at this particular moment. Although Fould was to resign as both Ministre d’État and Ministre de la Maison de l’Empereur in 1860, the commission continued its work until the end of the licensing period in 1864.43 Three subsidiary elements in the network of operatic power are audiences, the press and the censor. The analysis of audience in nineteenth-­c entury Paris is fraught with difficulties, but significantly more productive than is the case in other European cities. Steven Huebner’s study of this question acknowledged the scattered nature of the evidence but was able – on the basis of a consideration of surviving subscription lists and ticket prices – to document, and speculate carefully on, the makeup of the audiences at the Opéra, Opéra-Comique, Théâtre-Italien and Théâtre-Lyrique.44 There is, however, little evidence of the audience having an impact elsewhere in the power network of which it formed part. Needless to say, the absence of ­audience – lack of interest – would play a role in determining the length of a work’s run, but quantifying such a process falls foul of a mismatch between methodological desiderata and surviving evidence. Surprisingly, in view of the sheer volume of the press’s activity in ­n ineteenth-century Paris and the impact of censorship on opera elsewhere in Europe, neither plays a particularly significant role in the politics of genre at this point in the history of Parisian stage music. Despite occasional calls for a state subvention for an unfunded but successful opera

42 The three commissioners with responsibility for single institutions were Arsène Housaye (Théâtre-Français), Gilbert de Voisins (Théâtre-Italien) and Perrot de Renneville (Odéon). Monnais remained responsible for all the lyric theatres and the Conservatoire. See Almanach impérial pour l’an MDCCCCLIII présenté à leurs majesties (Paris: Guyot & Scribe, 1853), 862–864. 43 The work of the Commission supérieure and its membership is discussed below, 32–34. 4 4 Steven Huebner, ‘Opera Audiences in Paris, 1839–1870’, Music & Letters 70 (1989), 206–225.

The music of power  23 house or pointing to self-evident failings of a manager to fulfil the obligations of his cahier des charges, the press’s influence was focused primarily on the exercise of creative power: the domains in which librettists, composers and interpreters functioned.45 As judge and jury in the cases of singers’ careers, the press was all-powerful. But it also freely exercised its opportunities to comment on the quality of work and libretto to the extent that it could not only call for cuts but also expect to see them carried out. Perhaps the most striking example of this is Halévy’s La Juive in which three consecutive numbers were excised from the beginning of act iii after the 1835 premiere, reducing the dimensions of the role of Eudoxie by about half,46 as a result of pressure from journalists. But these incursions into the creative power of librettists and composers were subtle and frequently convincing, as, for example, when Pier-Angelo Fiorentino writing as ‘A. de Rovray’ suggested that in François-Auguste Gevaert’s opera comique, Quentin Durward (1858), it is in the second act that the largest cuts must be made. Mlle B ­ oulart’s aria is not very striking; Barrielle’s couplets can be cut without ­regret  … But not a note should be touched in Faure’s romance; it is charming, and the artist performs it with perfect taste and an admirable simplicity. Fiorentino was not just arguing for shrinking the dimensions of an already very long work, but carefully pointing to strengths and weaknesses as he engaged with the creative power of the work’s authors.47

45 This is by no means to minimise the other ways in which the study of all types of the press contribute to a study of nineteenth-century stage music. The literature is by now large, and the field is changing from studies that concentrated on such well-known figures as Berlioz (Kerry Murphy, Hector Berlioz and the Development of French Music Criticism, Studies in Musicology 97 (Ann Arbor and London: UMI Research Press, 1988)) to those that subject individual titles to particular study (Katherine Ellis, Music Criticism in Nineteenth-­ Century France: La Revue et Gazette Musicale de Paris, 1834–80 (Cambridge: Cambridge ­University Press, 1995)), examination of such authors outside the canon as Joseph ­d’Ortigue (Sylvia L’Écuyer, Joseph d'Ortigue: Écrits sur la musique, 1827–1846, Publications de la Société française de musicologie 2:17 (Paris: Société française de musicologie, 2003)), and to wider studies of music journalism outside the specialist press (Emmanuel Reibel, L’écriture de la critique musicale au temps de Berlioz (Paris: Champion, 2005)). The establishment of an online resource to promote the study of writings on music in the nineteenth-century French press further extends this work (Francophone Music Criticism, 1789–1914. http:// music.sas.ac.uk/fmc). 46 Diana Hallman, Opera, Liberalism and Antisemitism, 228–230. 47 ‘C’est dans le second acte qu’il faut pratiquer les plus larges coupures. L’air de Mlle Boulart n’a rien de saillant; les couplets de Barrielle pourraient être supprimés sans regret…. Mais qu’on se garde bien de retrancher une note à la romance de Faure, elle est charmante, et l’artiste la dit avec un goût parfait et une admirable simplicité’ (Le moniteur universel, 28 March 1858). For an overview of Gevaert’s opéras comiques, see 131–164.

24  The music of power The reason why the censors seem to have had less impact in nineteenth-­ century Paris than elsewhere in Europe is that managers and librettists showed themselves extremely adept at ensuring that any issues that might interest the censors were resolved in advance. While the surviving censors’ libretti are a major source for the reconstruction of the final phase of ­revision for almost all Parisian stage music of the period, they usually reveal far more of the librettist’s and occasionally the composer’s creative activity rather than evidence of state interference.48 The very few occasions when this system went awry point to the truth of this claim. Eugène Scribe’s ­l ibretto of Meyerbeer’s Les Huguenots is a case in point. Scribe’s establishment of the libretto and Meyerbeer’s completion of the score fall into a period just after censorship had been abolished by the July Monarchy in 1830. But censorship was reintroduced in 1835, before Les Huguenots’ premiere, and at a point where the libretto placed an arquebus into the hands of the King of France and closed with a dispassionate Cathérine de Médicis overseeing the act v carnage. The censors came down very hard on these scenes because Scribe and the management of the Opéra had carefully constructed the libretto according to the principles of the period 1830–1835 and were temporarily wrong-footed by the change in legislation. Both scenes were excised from the version that received its premiere in 1836.49 It is therefore no surprise that studies on censorship in nineteenth-century Paris ­c entre on Victor Hugo and the ‘legitimate’ theatre rather than on opera.50

New theatres and opera houses It was inevitable, after the closure of so many petits spectacles by the ­ apoleonic decrees of 1806–1807, that there would be attempts to restart N theatrical and operatic enterprises despite the legislation, and this was the most immediately visible challenge to the technologies represented by the licensing system. In the first instance, the only means of gaining permission 48 The standard accounts of censorship in nineteenth-century Paris are Odile Krakovitch, Les Pièces de théâtre soumises à la censure (1800–1830): inventaire des pièces (F 18 581 à 668) et les procès-verbaux des censeurs (F 21 966–995) (Paris: Archives Nationales, 1982) and idem, Censure des répertoires des grands théâtres parisiens (1835–1906): Inventaire des manuscrits des pièces (F18 669à 1016) et des procès-verbaux des censeurs (F 21 966à 995) (Paris: Centre Historique des Archives Nationales, 2003). See also idem, ‘Les Romantiques et la censure au théâtre’, Romantisme: Revue de la Société des Études Romantiques 38 (1982), 33–43. 49 For the sources of the libretto and the changes that the censors forced onto the creative team for Les Huguenots, see Jeanice Brooks and Mark Everist, ‘Giacomo Meyerbeer’s Les Huguenots: Staging the History of the French Renaissance’, The Renaissance in the Nineteenth Century – Le XIXe siècle renaissant, eds. Yannick Portebois and Nicholas Terpstra, Publications of the Centre for Reformation and Renaissance Studies: Essays and Studies 2 (Toronto: Centre for Reformation and Renaissance Studies, 2003), 121–142. 50 Krakovitch, Hugo censuré: la liberté au théâtre au XIXe siècle (Paris: Calmann-Lévy, 1985).

The music of power  25 to open was via the Prefect of Police, which meant that the first reopenings were on a very modest scale.51 The Théâtre Comte opened as early as 1809, but was permitted only acts of ventriloquism, magic and trials of strength; similarly, when the Théâtre de Luxembourg started in 1816, it was exclusively for high-wire acts and marionettes.52 But when the Panorama-Dramatique began its series of performances that mixed the well-known landscapes of the Panorama itself with scenes in dialogue, members of the Conseil du Roi argued strongly that the Prefect of Police had gone beyond the bounds of his jurisdiction and that authorisation of the Panorama-Dramatique should have been granted (or more probably refused) only by royal ordinance.53 None of the petits spectacles that opened during the first two decades of the licensing period represented any threat to the grands théâtres, since their repertory was so far removed from the music drama promoted elsewhere. But from 1820 onwards, various pressures resulted in increasing the number of secondary theatres substantially beyond the four that had been agreed in the 1806–1807 legislation. Some of the most important theatres of the period from the Restoration to the Second Empire emerged: the Théâtre de la Porte-Saint-Martin, Gymnase-Dramatique, Théâtre des Nouveautés and the Théâtre du Palais Royal. Especially important for opera were the Odéon, Théâtre de la Renaissance, Opéra-National/Théâtre-Lyrique and finally the Théâtre des Bouffes-Parisiens. All these organisations fell outside the range of the two principal institutions promoting opera in French – the Opéra-Comique and the Opéra itself – and became serial candidates for the third lyric theatre in the capital. Given that the environment created by the Napoleonic legislation was one that promoted an ideal match between genre and institution, any new opera house had to make a very special case in order to be given royal permission or – after 1830 – ministerial approval to open its doors. Furthermore, it had to be prepared to compromise with the state in maintaining the equilibrium demanded by the politics of genre to which all changes of repertory – ­however slight – were subject after 1806. Claims fell into one of two categories: either the opera house would support the careers of young composers, especially Prix de Rome laureates when they returned to Paris after their sojourn in Italy and other European states, or they would serve as a training ground for dramatic and lyric artists for the benefit of the grands théâtres. The Odéon in 1824, the Renaissance in 1838 and the Bouffes-Parisiens in 1855 all argued that permission to play new opéras comiques, any music drama or a new genre that would be called opérette or opéra bouffe, respectively, would enhance the prospects of novice composers, and the series of negotiations and false starts 51 See the countless appeals to the Prefect of Police in F-Pan F21 1045–1046. 52 Wild, Dictionnaire, 105–106 and 233. 53 Louis-Henry Lecomte, Les Jeux gymniques (1810–1812) – Le Panorama dramatique (1821–1823), Histoire des théâtres de Paris 7 (Paris: Daragon, 1908; R Geneva: Slatkine, 1973), 69–74.

26  The music of power that resulted in the establishment of the Théâtre-Lyrique in 1851 was entirely predicated on the encouragement for, and was supported by petitions from, young composers.54 Both the opening of the Gymnase-Dramatique and the extension of the Odéon’s licence to include opera were underpinned by the appreciation of their roles as institutions that would provide an environment for the development of new artistic talent.55 Such claims could also be coupled to more obviously opportunistic motives: the multiple applications to open a second Opéra-Comique in 1829, none of which was successful, involved a ritual demonstration of support from younger composers, but had more to do with the disarray in which the Opéra-Comique found itself at the end of the Restoration and from which it would not recover until 1834.56 But there could never be enough genres to support the growth in the number of new opera houses while maintaining the balance of generic power that the 1806/1807 legislation had guaranteed, and the state had to use all its skill to limit repertories in more exacting ways. The Odéon, in addition to spoken tragedy and comedy, was restricted to opéra comique that had fallen into the public domain and to opera in translation, whereas the Gymnase-­Dramatique was only permitted to put on operas, again from the public domain, but reduced to a single act in the case of operas in more than one act and to a single scene for one-act works.57 When the Nouveautés was opened with a licence to perform vaudeville in 1827, it quickly moved to replace the couplets du domaine public with newly composed music – effectively opéra comique – in a way that clearly breached the terms of its licence.58 The consequences of such a light hand with its repertory immediately brought the Nouveautés into conflict with the Opéra-Comique itself.59 Unsuccessful bids would occasionally

54 For the Odéon, see Everist, Music Drama at the Paris Odéon, 104–105; for the Renaissance, see idem, ‘Theatres of Litigation: Stage Music at the Théâtre de la Renaissance, 1838–1840’, Cambridge Opera Journal 16 (2004), 136. 55 See the Gymnase Dramatique’s licence, dated 1 February 1820, in F-Pan F21 1137. 56 Four proposals to set up a second Opéra-Comique were presented in September 1829 by André Sourd, Adolphe Bossange, François-Victor-Armand d’Artois de Bourganville, Scribe and Anne-Honoré-Joseph Duveyrier (pseud. Melesville). The Opéra-Comique responded to the proposals with a printed pamphlet that outlined all possible objections to the scheme, which was not formally abandoned until the publication of a ministerial arrêté on 5 July 1830. All four of the presentations cited the need to support young composers. See F-Pan F21 1092 for all four proposals and the Opéra-Comique’s printed response. For this period of the Opéra-Comique see Bara, Le Théâtre de l’Opéra Comique. 57 Ibidem. See the summary of the repertory permitted at the Gymnase in Wild, Dictionnaire, 181. 58 The Nouveautés’ licence, dated 8 December 1825, specifically identified plays in one, two or three acts, ‘mêlées de couplets sur des airs du domaine public [emphasis added]’ as its central repertory. F-Pan F21 1141. Still useful is Lecomte, Les Nouveautés, Histoire des théâtres de Paris 5 (Paris: Daragon, 1907; R Geneva: Slatkine, 1973). 59 The dispute between the Opéra-Comique and the Nouveautés, much complicated by the one-sided support of ministers, is one of the earliest sites of genre negotiation in this period. See below, 38–40.

The music of power  27 come up with imaginative suggestions for repertory unprompted either by managers or the state; an attempt to open a second Italian opera house in 1839 offered a draft cahier des charges that restricted its repertory to national dance (Spanish and Italian) and Italian comic opera (opera buffa and farse), but was no more successful than any other similarly unprompted proposals.60 One of the consequences of the difficulties in the relationship between vaudeville and opéra comique in the late 1820s and 1830s was the immense care that the state took with the licence of the Théâtre de la Renaissance in the 1830s. The history of opera at the Théâtre de la Renaissance is coloured by the fact that music drama was never envisaged there: Victor Hugo and ­A lexandre Dumas were looking to set up a Second Théâtre Français where – as ever – there would have been a small orchestra for overtures, entractes and incidental music, but no opera. The Renaissance was immediately put under pressure from its obvious rival, the Théâtre Français (the Comédie Française), and the state was forced to act. Its initial response was to propose and then authorise the performance of what it called vaudeville avec airs nouveaux. Vaudeville, at any point in the nineteenth century, depended largely for its music on the use of timbres, pre-existing musical numbers, and this defined the genre as a mix of speech and borrowed song. While the idea of inventing a genre that took the outlines of vaudeville but insisted on the use of new music must have been seen as a possible solution to competition with the secondary theatres where vaudeville was promoted, it threw the Renaissance into competition with the Opéra-Comique. Finding that it had created an impossible situation, the Ministry of the Interior made things significantly worse by trying to find a third solution to the question of a repertory for the Renaissance that did not overlap with other opera houses. Its subsequent invention of opéra de genre was – had its definitions been adhered to by the Renaissance – just about workable: at the same time as using continuous recitative, opéra de genre was broken up into numbers, with clearly articulated pauses after each, in the fashion of contemporary Italian opera. But opéra de genre was simply too close to grand opéra in many respects for the Opéra to ignore, and like the Opéra-Comique saw a potential threat and took successful steps in the courts to neutralise it.61 Although the Opéra and Théâtre-Italien had clearly distinct repertories and personnel, they were further delineated by the fact that, during the winter, the Opéra performed on Mondays, Wednesdays and Fridays, the

60 Charles-Auguste Morny de Montfort was a retired cavalry officer when he made this unsuccessful but imaginative proposal. It was rejected by the minister on 20 August 1839. F-Pan F21 1038. 61 For a full account of the state’s creation of vaudeville avec airs nouveaux and opéra de genre, see Everist, ‘Theatres of Litigation’, and for the impact of the latter, idem, ‘Donizetti and Wagner: opéra de genre at the Théâtre de la Renaissance’, Giacomo Meyerbeer and 19th-Century Parisian Music Drama, Variorum Collected Studies Series CS805 (Aldershot: Ashgate, 2005), 309–341.

28  The music of power Théâtre-Italien on Tuesdays, Thursdays and Saturdays; when, in the 1840s, Sundays were brought into the planning frame, they were treated alternately with the Opéra one week, the Théâtre-Italien the next. This was a system that worked well for decades, and when the manager of the Théâtre-Italien, Watel, attempted in 1845 to hire out his building to John Mitchell’s English theatre troupe on the nights he was not allowed to perform opera (and therefore allowing the building to be used in competition with the Opéra), he was arraigned before the minister and fined 10,000 francs.62 The most striking exception to this general principle concerns Jacques Offenbach’s attempts to start his Théâtre des Bouffes-Parisiens in 1855. As usual, the new enterprise was controlled by a licence that attempted to distance it from potential competitors; in this case, Offenbach was allowed to play music drama in a single act with no more than four singing characters on stage.63 But another theatre had been given an almost identical licence. Hervé’s Théâtre des Folies-Nouvelles had originally been refused permission to have more than two people on stage at any one time; as soon as the Bouffes-Parisiens were given the right to have four, so was the Folies-­Nouvelles.64 The two theatres were in direct competition with each other, for the same artists, composers and librettists. Had they been in competition for the same audiences, one or the other would have been likely to collapse, but while Hervé was looking to the traditional audiences of the boulevard, Offenbach – with tenancies at the Salle Lacaze on the Champs Elysées in the summer and on the Salle Choiseul in the Passage Choiseul in the Winter – was not only setting out his stall in the different part of the city, but he was also aiming to amuse a very different type of audience.65 However agile Offenbach managed to be in finding ways of differentiating his work from that of Hervé, there was no escaping the radical change in attitude to the relationship between genre and institution represented by this state-controlled competition. Whether it was inadvertent, deliberate or predicated on the knowledge that the similarity of licence would be balanced by the different geographical and social environments of the two 62 Judgement was passed on 18 January 1845. The matter was raised by the director of the competing opera house, Léon Pillet, at the Opéra. F-Pan AJ13 1051. 63 The earliest version of Offenbach’s licence restricted him to three singing characters (F-Pan F21 1136, 4 June 1855; see Jean-Claude Yon, Jacques Offenbach, NRF biographies (Paris: Éditions Gallimard, 2000), 137); the loosening of this limitation to four was accomplished in the revision of Offenbach’s contract, 22 October 1855 (Wild, Dictionnaire, 63). 64 For Florimond Ronger (Hervé) and the Théâtre des Folies-Nouvelles, see Eugène Woestyn and Eugène Moreau, Les Folies-Nouvelles, Les théâtres de Paris (Paris: Martinon, 1855), and Lecomte, Les Folies-Nouvelles, Histoire des théâtres de Paris 4 (Paris: Daragon, 1909). 65 For an account of Offenbach’s deft negotiation of the politics of genre during the early years of the Théâtre des Bouffes-Parisiens, see Chapter 3.

The music of power  29 opera houses is difficult to explain. But this curious set of circumstances arose less than a decade before the end of the licensing system, and activities at Offenbach’s opera house represented one of several pressures on the system that were to bring about its decline and repeal.

Loups-cerviers, private enterprise and the aristocracy The July Monarchy gave the appearance of a coherent management of opera houses and theatres: a system of reporting to the Ministry of the Interior, of Commissaires royaux and of the commission itself. But in addition to the wide range of challenges to these technologies, much clandestine power resided in the so-called loups-cerviers, the financiers and bankers that supported the Opéra in the first instance, and other theatres subsequently.66 The position arose because every manager of an operatic enterprise had to put up more caution money than they could ever raise themselves (managers of the Opéra were required to find 250,000 francs, for example); this could only be achieved by recourse to a financial backer acceptable to the Ministry of the Interior.67 During the course of the period, managers of the Opéra turned to Alexandre-Joseph-Marie-Léon-Pierre-Paul-Raimond-Louis-­Gonzague Aguado, Marquis de Las Marismas del Gaudalquivir (1831–1842), A ­ ugusteJules-Edouard de Saint-Mars (1843–1845), Étienne Jean François d’Aligre (1845–1846) and Jean-François-Pierre, Baron Dudon (1847–1848).68 Beyond their formal financial support for the Opéra’s management, little is known of the involvement in the institution’s affairs of Saint-Mars, d’Aligre or Dudon, and it is entirely possible that their participation in the life of the 66 A loup-cervier is, in purely zoological terms, a lynx (Lupus cervarius; but note that the modern Latin term relates specifically to the Canadian lynx). For nineteenth-century France, however, the term was also applied to the mandarins of banking, industry and commerce who were able to do what they wished with their capital to the disadvantage of their neighbour. See Cras, ‘L’exploitation de l’Opéra’, 140, and the sources cited there. For a contemporary definition, see Pierre-Joseph Proudhon, De la création de l’ordre dans l’humanité, ou Principes d’organisation politique (Paris: Prévot, 1843), 449. 67 The caution money for the Opéra remained the same throughout the July Monarchy and into the Second Republic. In 1849, the cautionnement for the Théâtre Français was the same as the Opéra, after which the sums dropped off rapidly: 80,000 francs for the Opéra-Comique, 60,000 francs for the Théâtre-Italien. Most of the secondary theatres were priced around 30,000 francs, with the petits spectacles at 10,000 francs. See the complete listing in Conseil d’Etat. Section de legislation. Commission chargée de préparer la loi sur les théâtres. Enquête et documents officiels sur les théâtres (Paris: Imprimerie nationale, 1849), 200. 68 The claim made by Albéric Second (Les petits mystères de l’Opéra (Paris: Kugelmann, 1844), 223), and repeated by Cras (‘L’exploitation de l’Opéra’, 154) that the librettist SaintGeorges was successful at the Opéra because of his links with Saint Mars is not supported by an analysis of Saint-Georges’ activity before, during and after Saint-Mars’ period as guarantor of the Opéra. In any case, Second’s book was published only halfway through this period and therefore could never be considered a viable witness.

30  The music of power institution was limited to nothing more than financial backing. By contrast, Alexandre Aguado, Marquis de Las Marismas, left abundant traces of his financial support for the Opéra, and other opera houses both in Paris and elsewhere in Europe, which reveal his intervention in their day-to-day running and artistic management. Having settled in Paris at the beginning of the Restoration, by the beginning of the July Monarchy, Aguado was probably one of the wealthiest men in the city. His involvement with opera formed a large part of his life: he had invited Rossini to his château at ­Petit-Bourg to write most of Guillaume Tell and – the protector of the ballerina Alexandrine Fitz-James – he was a well-known figure around the Opéra. His town house was the old Hôtel d’Augny in the rue Grange-Batelière, almost part of the same block that housed the Opéra itself.69 From the moment the Opéra was opened up to private enterprise in 1831, Aguado was ready with the caution money that Véron so urgently needed to start the business. Aguado had already established an alliance with ­Domenico Barbaja in Naples, and shortly after underwriting the Opéra, he created a similar relationship with Thomas Monck-Mason in London; by 1832, he therefore had interests in three major opera houses: the Opéra, Teatro San Carlo in Naples and the King’s Theatre.70 Much of his activity during the 1830s may be viewed as an attempt to build a pan-European operatic power base centred on Paris. This was to be achieved first and foremost by the re-establishment of a joint management of the Opéra and Théâtre-Italien; there were two formal attempts in 1834 and 1836 which were rebuffed by the state, rejections in which the commission played a key role.71 In the second half of the 1830s, Aguado became progressively dissatisfied with Duponchel’s financial performance as manager of the Opéra, and it was claimed that although Duponchel was the titular head of the institution, Aguado was effectively in control. When Severini died in the fire at the Théâtre-Italien, Aguado was in a sufficiently strong position to summon Duponchel and the new manager of the Théâtre-Italien, Louis Viardot, to his holiday retreat in Dieppe to instruct them in his plans to merge the two institutions with Her Majesty’s theatre in London.72 Although Duponchel was in Aguado’s pocket, Viardot was a much tougher adversary, and while 69 See, for a summary of Aguado’s biography, Jean-Louis Tamvaco, Les cancans de l’Opéra: le journal d’une habilleuse, 1836–1848, 2 vols (Paris: CNRS, 2000), 2:889–890. There is useful contemporary material in Véron, Mémoires sur l’affaire du Constitutionnel, 33–45. There is disappointingly little on Aguado and opera in Jean-Philippe Luis, L’ivresse de la fortune: A. M. Aguado, génie des affaires (Paris: Payot, 2009). 70 Cras, ‘L’exploitation de l’Opéra’, 144. Aguado was well known as the individual behind the managerial soubriquet ‘Robert’, as evidenced by in a letter from the Prefect of Police (Henri-Joseph Gisquet) to the Minister of the Interior (Montalivet) ‘Renseignements sur le théâtre royal Italien’, May 1836. F-Pan F21 1113. 71 Cras, ‘L’exploitation de l’Opéra’, 146–147. 72 See the account from Louis Gentil given in Tamvaco, Les cancans de l’Opéra, 1:441 and 2:822.

The music of power  31 the fate of both institutions remained in the balance during the first half of 1839, the commission stepped in and stopped the merger in its tracks.73 Aguado was, however, responsible for the appointment of Édouard Monnais as ­Duponchel’s assistant in December 1839, and he also received permission from incoming Minister of the Interior Duchâtel to appoint Duponchel’s successor. Incredibly, and in a move that shows just how far-reaching ­Aguado’s control was over the higher echelons of government, the Commissaire royal, Léon Pillet, was appointed to the management of the Opéra as both Aguado’s creature and puppet, while Monnais found himself neatly swapped with Pillet as Commissaire royal, a position that he was to hold for a quarter of a century.74 At certain points during the July Monarchy, Aguado wielded as much power in Parisian operatic circles as he did on the international financial stage (he was the European agent for the Spanish Crown). His interests, however, were not confined to the purely financial security of the institutions he backed. He certainly ensured that Fitz-James was well placed in the corps de ballet, at least until he broke with her in 1838,75 but his more general influence on artistic policy at the Opéra was recognised by ­Meyerbeer, Halévy, Donizetti and the commission itself.76 His power was expressed in the selection of artists and in the commissioning of new works: when ­Donizetti reported that Aguado specifically wanted a four-act opera from him in addition to Le Duc d’Albe, the former was clearly looking to repeat the success of Les martyrs and La favorite, both in four acts.77 And when Kathinka Heinefetter was auditioned at the Opéra, Monnais reported to Meyerbeer that Aguado was enthusiastic about her for the role of Fidès in Le prophète, an extraordinary and ill-conceived suggestion, since this was the role that would eventually be given to the totally different voice of the mezzo-soprano Pauline Viardot.78 73 The plans were, however, sufficiently advanced for the preparation of a draft arrêté in September 1839 (F-Pan F21 1113). 74 Commission des Théâtres. Registre des Procès-Verbaux des délibérations, (19 June 1839; F-Pan F21 4633 (6)). See Cras’s finely textured account of this affair (‘L’exploitation de l’Opéra’, 149–151). 75 Tamvaco, Les cancans de l’Opéra, 1:279. 76 Commission des Théâtres. Registre des Procès-Verbaux des délibérations, 11 October 1839 (F-Pan F21 4633 (6)). 77 Letter from Donizetti to Giovanni Ricordi, 23 May 1841 (Guido Zavadini, Donizetti: vita, musiche, epistolario (Bergamo: Istituto Italiano d’arti grafiche, 1948), 540–541). Les martyrs, with a successful first run and an abortive 1843 revival, could never match La favorite’s sustained success up to the end of the century, however. 78 ‘I should add confidentially that when we heard [Kathinka Heinefetter] in the opéra house, M. Aguado, who was present at the audition, immediately said “I hope that we have here the woman that we need for Meyerbeer’s opera”’ (‘Je dois ajouter confidentiellement que lorsqu’elle [Kathinka Heinefetter] s’est fait entendre chez nous, au théâtre, M. Aguado, qui assistait à l’audition, a dit aussitôt “J’espère que voilà la femme qu’il fallait pour l’opéra de Meyerbeer”’). Letter from Edouard Monnais to Meyerbeer, 11 May 1840. See Heinz

32  The music of power The changes to the administration of the imperial theatres in the early years of the Second Empire threw the special status of the Opéra into relief: it was no longer run by a directeur-entrepreneur but was brought back directly under the financial control of the state, a return – in effect – to the situation before 1830. It was, however, placed under the supervision of a Commission supérieure permanente … du théâtre impérial de l’Opéra from 30 June 1854. Although the title of the commission recalled that of the July Monarchy and Second Republic, its membership was strikingly different. Whereas its predecessor had been largely made up of those with a professional interest in the arts, and theatre and opera in particular, the Second Empire’s Commission supérieure was made up of some of the most powerful men in the country.79 The Opéra had not been under such centralised power for a century, and no signal could have been stronger: it was as important to Napoléon III’s government as the railways, the military and probably the church. This would even have been the case if the power the commission gave itself had been delegated to others. Fould and his colleagues, however, took immense trouble themselves with the most routine of matters relating to the Opéra. The commission’s remit was wider than that of its predecessor and more relentlessly exploited: its members read libretti, attended ­rehearsals, arbitrated in cases of dispute and undertook detailed monitoring of individual performances, while at the same time arguing in the various seats of government the merits of universal suffrage or war with the Austrians in ­Italy.80 The members of the commission could manipulate the institution in

Becker and Gudrun Becker (eds.), Giacomo Meyerbeer: Briefwechsel und Tagebücher, 7 vols to date (Berlin: De Gruyter, 1960–), 3:711, cited in Cras, ‘L’exploitation de l’Opéra’, 154, note 110. For Heinefetter’s career, see Karl-Josef Kutsch and Leo Riemens, Großes Sängerlexikon, 4 vols (Bern and Stuttgart: Francke, 1987–1994); 4th ed. 7 vols (Munich: Saur, 2003), 3:2014–2015. Although she sang Valentine in Meyerbeer’s Les Huguenots, she never sang, nor could have sung, the role of Fidès in Le prophète. 79 As has already been seen, the commission’s president was Fould, again succeeded by Alexandre Florian Joseph, Count Colonna-Walewski, and they were assisted by Eugène Rohrer, minister for agriculture, commerce and public works (who in previous regimes would have been called the Minister of the Interior); Raymond-Théodore Troplong, president of the senate; Charles-Auguste, Comte de Morny, president of the Corps ­législatif; Jules Baroche, president of the Conseil d’État; Félix, Comte Baciocchi, first ­chamberlain, superintendent of court entertainments and of the chapel and chamber ­music; ­Gustave-Louis-Adolphe-Victor-Charles Chaix d’Est-Ange, vice-president of the Conseil d’État. The commission’s secretary was Alphonse Gautier, secretary general to the Imperial Household. 0 See the range of activities documented in F-Pan F21 1053. On 12 April 1856, Troplong 8 circulated members of the entire commission, noting that Gilbert Duprez’s opera Samson was to be rehearsed in Duprez’s home (11 rue Turgot) on the following day. Troplong considered the work ‘worthy of the administration’s [the choice of the term ‘administration’ rather than ‘commission’ is emblematic of the central position the Opéra now held for government (administration)] attention’ (‘digne de l’attention de l’administration’) and invited members of the committee to attend the rehearsal; Samson received a concert performance on 1 October the following year, and the rest of the commission clearly disagreed with Troplong’s view, hardly surprising given the work’s generic designation as

The music of power  33 order to support initiatives in international relations – with England, ­Russia, Austria and Italy. Furthermore, the commission was directly involved in the both the day-to-day and longer-term programming of the Opéra, frequently under the direct instructions of the emperor. During the 1850s and 1860s, the commission both regularly made ­general recommendations concerning repertory81 to the Opéra and insisted on ­specific changes.82 Even the pattern of new works at the Opéra changed ­radically as a result of the involvement of Fould and the commission. While the major works each year in the first half of the 1850s had been the product of such well-known names as Meyerbeer, Auber, Gounod, Niedermeyer, Halévy and Verdi,83 the repertory of the second half of the decade – by which time the commission had been able to intervene – consisted of reworkings of Bellini, Verdi, Rossini and Auber84 alongside grands opéras from the pens of Ernst II, Duke of Saxe-Coburg-Gotha (Sainte Claire, 1855), Józef-Michel-François-Xavier-Jean, Prince Poniatowski (Pierre de Médicis, 1860) and Richard Wagner (Le Tannhauser, 1861), that showed all the signs of government intervention.85

81

82

83

84 85

opérette, although he might be forgiven his enthusiasm on the basis of the work’s title. An undated memorandum from Crosnier (between 11 November 1854 and 30 June 1856 therefore) makes clear that the commission was responsible for reading libretti, and a further letter from Crosnier, 10 January 1855, asked the minister to allow Mme Miolon to leave the Opéra-Comique and move to the Opéra. This entire routine request was apparently rejected (the document is incomplete) because of the organisation’s budgetary position. A final example of the commission’s work is a complaint from the Minister of the Interior (Persigny, replacing Rohrer on the commission) to Royer, 6 February 1862, about the performance of Guillaume Tell the previous Sunday, commenting on the lack of discipline in the chorus and the poor selection of artists for the divertissemenents. Royer’s very full response (ibidem) demonstrates the poverty of Persigny’s understanding of the Opéra’s management but also the levels of power entrusted to such ignorance. In 1854, Fould had written to Louis-Victor-Nestor Roqueplan, the manager of the Opéra, complaining about the breadth of the repertory at the Opéra and suggesting works that might be revived. His recommendations would have seemed breathtakingly conservative to his contemporaries. See the letter from Fould to the manager of the Opéra [Roqueplan], 27 September 1854 (F-Pan AJ13 451 (I)). On Wednesday 5 March 1862, for example, Gautier wrote to the administrator of the Opéra that ‘the Minister, whom I have just consulted, wishes nothing less in the world for Lucie de Lammermoor to be played on Sunday. Replace it with either Le philtre or Le comte Ory’ (‘Le Ministre que je viens de voir ne veut pour rien au monde qu’on joue Lucie dimanche. Remplacez Lucie par Le philtre ou Le comte Ory’) (F-Pan AJ13 443 (III)). Meyerbeer, Le prophète (1849); Auber, L’enfant prodigue (1850); Gounod, Sapho (1851) and La nonne sanglante (1853); Halévy, Le Juif errant (1852); Niedermeyer, La fronde (1854); Verdi, Les vêpres Siciliennes (1855). Verdi, Le trouvère (1857); Auber, Le cheval de bronze (1858); Bellini, Roméo et Juliette (1859); Rossini, Sémiramis (1860). The period did not entirely see the eclipse of grand opéra by established composers: ­David’s Herculanum was premiered in 1859, and Gounod returned with La reine de Saba, but not until 1862. This account of the involvement of central government with the repertory of the Opéra cannot take account of the vast literature, for example, on Le Tannhauser. It

34  The music of power Both the members of the Commission supérieure de l’Opéra, in the decade after 1854, and Aguado, throughout the first decade of the July Monarchy, were able to exercise their power over the production, consumption and reception of stage music in Paris, particularly at its most prestigious venue; the exercise of this power was, however, in the context of the administrative framework that they had inherited. Aguado could no more change the generic profile of the Opéra than he could change the currency in use in his native Spain. But although Fould and his colleagues had to work within the existing technologies of power, they had also been instrumental in setting them up. As has been seen with the conflicting licences given to Offenbach and his colleagues in the 1850s, the commission’s actions paradoxically served to weaken the share of the network of power enjoyed by the state for nearly half a century.

The politics of genre While the Napoleonic legislation of 1806–1807 went as far as it could to map genre onto institution, the emergence of new opera houses and the more generous rewriting of licences meant that the concept of genre became a central site for dispute, negotiation and resolution. The inadequacy of the top-down model of power embodied in this imperial legislation cannot simply be countered by stressing the bottom-up power of individual opera houses; such a move only reveals part of the picture and risks failing to reflect the constant renegotiation of power within the polity.86 Alongside the legislation that set up the licensing system came an understanding of genres as property, certainly as far as the grands théâtres were concerned: opéra comique, continuous opera in French and Italian opera being regarded as the property of the Opéra-Comique, Opéra and Théâtre-Italien, respectively. For one institution to lay claim to what was thought of as a genre that belonged to another, either by simply attempting to mount performances or to request permission through government agency, was thought of, and indeed termed, trespass (empiètement). In some cases, modifications to institutional arrangements could result in genres passing from one institution to another. The joint management of the Théâtre Royal-Italien and Académie Royale de Musique during the Restoration, for example, meant that works that were the property of the former could appear on the stage of the latter: this was of importance for

forms the basis, however, of a forthcoming study that brings together government control, repertory and Gluck reception in the period 1855–1862. 8 6 The association of a polity and the arts has a pedigree going back to the middle of the eighteenth century. See John Brown, A Dissertation on the Rise, Union, and Power, the Progressions, Separations, and Corruptions of Poetry and Music (London: Davis and Reymers, 1763), 40.

The music of power  35 three key works by Rossini. Mosè in Egitto was reworked as Moïse, Maometto II as Le siège de Corinthe, and Il viaggio a Reims transferred in part from the Théâtre-Italien to the Opéra as Le comte Ory. None of these transformations represented an immediate threat to the Théâtre-Italien from the Opéra, since no trespass could exist within the same institution.87 Such other cases as the 1834 production of Don Giovanni appear at first sight to trespass on the rights of Théâtre-Italien, which was now back under separate management and had been playing Don Giovanni in Italian since 1811.88 But the Opéra had mounted a performance of the work as long ago as 1805 and could therefore claim ownership of Don Giovanni, at least to the extent of fending off any possible complaint from the Théâtre-Italien.89 The latter was by 1834 involved in an initiative of its own – bringing Italians to write new operas for the house (Donizetti’s Marino Faliero and Bellini’s I Puritani both premiered the following year; Mercadante’s I briganti the next) – and was therefore less likely to object in any case. Other instances where works passed from one institution to another ­reflect the direct imposition of political power over the opera houses themselves. Perhaps the most remarkable history is that of the opéra comique by Scribe and Auber entitled Le cheval de bronze, premiered in 1835 with limited success,90 which was brought back to life at the Opéra in 1857 with the spoken dialogue replaced with recitative, the composition of dance music and other

87 See, for the former, Paolo Isotta, ‘Da Mosè a Moïse’, Bolletino del Centro Rossiniano di Studi 1–3 (1971), 87–117; idem (ed.), Gioacchino Rossini: Mosè in Egitto, Azione tragico-­ sacra; Moïse et Pharaon, Opéra en quatre actes; Mosè, Melodramma sacro in quattro atti, Edizione critica delle opere di Gioachino Rossini 4 (Turin: Unione tipografico-editrice torinese, 1974); Marcello Conati, ‘Between Past and Future: The Dramatic World of Rossini in Mosè in Egitto and Moïse et Pharaon’, Nineteeth-Century Music 4 (1980–1981), 32–47; Richard Osborne, Rossini, The Master Musicians (London: Macmillan, 1986), 208–213 and 237–242; Giuseppe Ierolli, ‘Mosè e Maometto: da Napoli a Parigi’, Tesi di laurea, ­Università degli studi di Bologna, 1989–1990. For Le Siège de Corinthe, see Gerhard, Die Verstädterung der Oper, 68–84 [page numbers refer to English translation]. For the detail of the relationship between the music of Le comte Ory and Il viaggio a Reims, see Johnson, ‘A Lost Rossini Opera Recovered: Il Viaggio a Reims’ BCRS 23 (1983), 5–57; eadem (ed.), Il viaggio a Reims, 1:xliv–xlv. 88 The 1834 production was influential and has elicited much comment. See Sabine ­Henze-Döhring, ‘E.T.A. Hoffmann-‘Kult’ und ‘Don Giovanni’-Rezeption im Paris des 19. Jahrhunderts: Castil-Blazes ‘Don Juan’ im Théâtre de l’Académie Royale de Musique [sic] am 10 März 1834’, Mozart-Jahrbuch 1984/5 des Zentralinstitutes für Mozartforschung der Internationalen Stiftung Mozarteum Salzburg (Kassel, Basel and London: Bärenreiter, 1986), 39–51; Katherine Ellis, ‘Rewriting Don Giovanni, or “The Thieving Magpies”’, Journal of the Royal Musical Association 119 (1994), 212–250. 89 Laurent Marty, 1805: la création de Don Juan à l’Opéra de Paris, Univers musical (Paris: L’Harmattan, 2005). 90 Premiered on 23 March 1835, it received 84 performances that year and 22 the next. There were no more performances until it transferred to the Opéra in 1857. See Albert Soubies, Soixante-neuf ans à l’Opéra-Comique en deux pages de la première de ‘La dame blanche’ à la millième de ‘Mignon’ (1825–1894) (Paris: Fischbacher, 1894).

36  The music of power musical numbers, including an octet parlante which was much praised by Meyerbeer.91 This was the first time that an opéra comique had ever found its way onto the stage of the Opéra, and given the Opéra-­Comique’s redoubtable track record in attacking any possible challenge to its right to opéra comique, this is remarkable. Little evidence survives of any challenge, nor is there any explanation for where the idea for such a striking move came. Some clues come from correspondence between the Minister of State, Walewski, and the director of the Opéra, Royer, dating from five years later. In the context of a search for shorter operas that could be performed alongside ballets to complete an evening’s entertainment, which had resulted in a revival of Rossini’s Le comte Ory and two new works, Alary’s La voix humaine and Massé’s La mule de Pedro, the minister himself suggested that a revival of an opéra comique by Hérold from 1829 – L’illusion – ‘could usefully take its place in the Opéra’s repertory which includes so few works in one act to accompany large ballets’.92 Walewski had already approached Jules-Henry Vernoy, Comte de Saint-George, one of the two co-authors of the libretto, to make the necessary changes, and invited Royer to think about the casting with a view to performance in the following season. The fact that he failed to mention the musical changes might not be unconnected with the fact that Hérold, unlike Auber, had been dead for thirty years and could not undertake the essential changes in the music. This may well explain why the project came to nothing. It demonstrates that the Ministre de l’État had no need to refer to the Opéra-Comique in appropriating what earlier would have been considered its intellectual property and could then order the execution of the plan by the manager. Although L’illusion was an opéra comique in a single act, the story strongly suggests the sort of power that was exercised in the case of the three-act Cheval de Bronze five years earlier.93 One of the principal ways in which the state intervened in networks of power was in the arbitration in matters of dispute between opera houses concerning repertory, and this is one of the ways in which the politics of genre is most clearly negotiated. All the major opera houses were involved in some sort of dispute, either as plaintiff or defendant, and frequently their actions involved the secondary theatres as well. Inevitably, the secondary theatres, when they in turn saw their repertorial rights (particularly the 91 Becker and Becker (eds.), Giacomo Meyerbeer: Briefwechsel und Tagebücher, 7:618. 92 ‘pouvait prendre utilement sa place dans le répertoire de l’Opéra qui compte si peu ­d’ouvrages en un acte, pour accompagner les grands ballets’. Ministre d’État (Waleski) to Royer, 11 November 1862 (F-Pan AJ13 443 (III)). 93 This instance of the possible use of an adaptation of a one-act opéra comique to occupy an evening with a ballet is a remarkable perspective on the tradition of petit opéra. Most petits opéras – designed to accompany ballet after grands opéras had increased in size so as to make such a pairing impractical – were in two acts; the idea that a one-act work would be suitable for accompanying longer ballets is imaginative but not unknown. See Marliani’s La xacarilla (1839); Adam, La bouquétiaire (1847); Membrée, François Villon (1857). See 87–88.

The music of power  37 right to employ music in the public domain within vaudeville) under threat from the so-called petits spectacles, acted swiftly and decisively. Many new opera houses and theatres that had opened during the Restoration and July Monarchy, because their repertory was based on opéra comique or a close cognate such as vaudeville, came into direct competition with the Opéra-Comique itself, and, as has been seen, this institution was vociferous in its challenges to their very existence. When the Théâtre Royal de l’Odéon opened in 1824, like most theatres, its first night included a pièce de circonstance, an occasional work that alluded to its repertory and acted as a sort of cultural manifesto (and bears comparison with the Opéra National’s 1847 Les premiers pas). As it was allowed to play comedy, tragedy and opera, the Odéon’s opening piece was Les trois genres, which included an opera scene to a libretto by Scribe and music by Boïeldieu and Auber. Such a work as Les Trois Genres might seem to be breaking a condition of the licence of the theatre that it should not mount productions of new opéras comiques. The ministre de la Maison du Roi, Jacques-Alexandre-Bernard Law, Marquis de Lauriston, accepted the argument that this was a unique event not to be repeated. A precedent had, however, been set, although not the one the state anticipated. If new music was permitted in an occasional work such as Les trois genres, other occasional works could also include new music, and the Odéon assiduously promoted occasional works for the coronation of Charles X and for the monarch’s name days, all the time taking the opportunity to present new stage music in contravention of its licence. It was never challenged.94 Simultaneously, however, the Odéon was attempting to win the right to play new opéras comiques. Given that these new works were not in the Opéra-Comique’s repertory, but belonged to the genre of which it thought it had a monopoly, the royal theatres, led by the Opéra, contested the principle by recourse to the 1806/1807 legislation and to the example set by the Gymnase-Dramatique: that each opera house and theatre should have its own genre. The Gymnase was restricted to lyric works in one act, the Odéon to those in the public domain, and that was the way the Opéra-Comique thought it should stay. To its credit, the Odéon came up with a number of solutions to the problem – reducing the delay for entry into the public domain from ten to five years, for example – and at one point looked like it might just succeed, until a change of government took the entire set of negotiations back to the beginning.95 The Opéra-Comique’s relationship with the Odéon was problematic because both were royal theatres and both in receipt of a subvention. The brokering of power was therefore finely nuanced within the Maison du Roi and

94 For the history of Les trois genres and subsequent smuggling of new composition into occasional works at the Odéon, see Everist, Music Drama at the Paris Odéon, 143–170. 95 Ibidem, 65.

38  The music of power its replacement structures. The same was not true with other institutions that were allowed to perform comédies mêlées de musique/ariettes (the terminology varies) and that sprang up during the period: the G ­ ymnase-Dramatique, Nouveautés, Renaissance, Opéra National and Théâtre des Variétés. In two cases, those of the Nouveautés and the Renaissance, the Opéra-­Comique launched guerrilla campaigns that forced suspensions of performances and temporary closures as the managers of the two institutions tried to fight off the claims being made against their opera houses in the Maison du Roi, in the Ministry of the Interior and even in the courts. The case against each of the rival opera houses was subtly different in its context and claims, but ultimately the Opéra-Comique was pursuing a policy of protecting what it saw – until the emergence of the Théâtre-­Lyrique in the early 1850s – as its exclusive rights to opéra comique. During the Restoration, matters were complicated by direct royal and political patronage. The Gymnase-Dramatique was set up in 1820 on the Boulevard Bonne-Nouvelle, but the theatre gained the support of the Duchesse de Berry in 1824 after it had played for her in Dieppe that ­summer. It ­ adame la was then entitled to change its name to the Théâtre de S.A.R., M ­ ourbons in 1830 as duchesse de Berry, and was known until the fall of the B the Théâtre de Madame.96 Concomitant with this support was protection from the challenges from the Opéra-Comique as the Théâtre de ­Madame developed its repertory.97 Here, although the mechanisms of the state had been circumvented, liberated even, by royal patronage, creative power could dominate, as Charles-Gaspard Delestre-Poirson and Scribe manoeuvred within institutional structures to evolve a completely new genre, the comédie-vaudeville; although also permitted to play opéra ­comique, the Gymnase saw greater possibilities in purely artistic developments of comédie-vaudeville and was able to distance itself from challenges from other theatres.98 The story at the Théâtre des Nouveautés unfolded differently. The theatre was set up in March 1827 with the direct support of the Minster of the Interior, Jacques-Joseph-Guillaume-Pierre, Comte de Corbière, acting on the orders of the King.99 Until the beginning of 1828, it was able to make substantial inroads into repertories banned by its licence, performing more and more new music in the context of vaudeville to the louder and louder

96 Delestre-Poirson made his announcement of this patronage to the minister on 7 September 1824. It may be read very clearly as a response to threats to the Gymnase-Dramatique posed by the minister eighteen months early (20 February 1823). F-Pan F21 1137. 97 See the dossier of complaints about the founding of the Gymnase Dramatique in November 1819–March 1820 (F21 4673(1)). 98 Discussed in Wild, Dictionnaire, 181, and Jean-Claude Yon, Eugène Scribe: la fortune et la liberté (Saint-Genouph: Nizet, 2000), 57–96. 99 See Lecomte, Les Nouveautés, 1–5.

The music of power  39 100

complaints from the Opéra-Comique. With the support of Corbière, it was able to ride out these calls of protest, but when the second Corbière left office in January 1828, the climate changed almost overnight, and hostile reports found their way to the new minister’s desk. The head of the ­Bureau des théâtres, Antoine-Marie Coupart, drew up a long list of the Nouveautés’ sins within days of Corbière’s departure.101 Exactly how Corbière was able to continue his support for the theatre in the eighteen months after leaving office is not clear, but his refusal to sign the oath of allegiance to ­Louis-Philippe after the July Revolution in 1830 and his retirement to ­Brittany sealed the fate of the Nouveautés, and it closed at the end of 1831. If the temporary success of the Théâtre des Nouveautés was a result of the patronage of a single individual, disputes between the Opéra-Comique and the Opéra-National two decades later were the result of the imprecise drafting of cahiers des charges.102 The Opéra-National was given permission to play opéra comique but limited in a way that would protect the interests of the Opéra-Comique itself and its current manager, André-Alexandre ­Basset.103 Accordingly, the Opéra-National was permitted to play opéra comique that had fallen into the public domain but had not been played either 100 For example, the Minister of the Interior, Corbière, wrote to Duc d’Aumont in response to his complaints about the Nouveautés, 17 December 1827 (F-Pan F21 1141). 101 ‘Théâtre de Nouveautés’, 29 January 1828 (ibidem). 102 The account of the history of the Opéra National in Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981), 1–12, is an accurate chronicle of the events based on ­n ineteenth-century printed sources and the press. For a full account of the Théâtre-­ Lyrique which skillfully diagnoses systemic problems with the organisation, see K ­ atherine ­Ellis, ‘Systems Failure in Operatic Paris: The Acid Test of the Théâtre-Lyrique’, Music, Theater and Cultural Transfer: Paris, 1830–1914, eds. Mark Everist and ­A nnegret Fauser (­Chicago: Chicago University Press, 2009), 49–71 (the rest of Walsh, Second Empire Opera is still useful, however). The summary of activity at the Théâtre-Lyrique in Di Profio, ‘Le Théâtre-Lyrique’, 91–103, has little control of the primary sources, and even less of the secondary sources (it does not even cite Walsh, Second Empire Opera). 103 Negotiations towards the Opéra National’s licence were complex and tense, as Adam attempted to gain access to the modern repertory. The relevant documents are preserved in Commission des Théâtres. Registre des Procès-Verbaux des délibérations (F-Pan F21 4634 (1)). He argued that he should be allowed to perform such works if they had not been performed for three years, whereas Basset insisted on a period of ten years before they could be taken up by the Opéra-National; a draft of Adam’s cahier des charges presented a third possibility: that he could mount productions of works played before 1840 (a ­s even-year delay) but only if they had not been performed since then by the Opéra-Comique. The final recommendation from the commission was clear, however, and represented an exercise of state power rather than an attempt at legislation. The Opéra-National ‘shall not produce any work performed for the first time under the management of MM. Crosnier and Basset without the special authorisation of the Minister, the Commission des théâtres royaux having been consulted’ (‘ne pourra faire représenter aucun ouvrage joué pour la 1iere fois sous la direction de MM. Crosnier et Basset sans l’autorisation spéciale du Ministre, la Commission des théâtres royaux consultée’, ibidem). In other words, the state could rule on each and every case separately.

40  The music of power by Basset or his predecessor, Crosnier; in other words, the Opéra-National could only play public-domain opéra comique that had been off the stage since at least 1834. There remained a category of work – not in the public domain and not part of either Basset’s or Crosnier’s repertory – that was not addressed in the Opéra-National’s cahiers des charges. Not surprisingly, a work that fell into this category – Berton’s classic Aline, reine de Golconde of 1803 – was one of the first works to be programmed by the Opéra-­National. The two librettists, Jean-Baptiste-Charles Vial and Edmond de Favières, had both died in 1837, but Berton had only died in 1844, so while the libretto was just in the public domain, the music was not. Although the opéra comique had not been performed since 1830, it was so close to the public domain and otherwise within limits that Mirecour and Adam clearly thought it was worth trying to programme. When the matter came to the Commission des théâtres after Basset had indeed complained, the situation turned out to be even more complex, since both the Opéra-Comique and Opéra-National had individually contracted with the heirs of the composer and librettists, respectively. The final paragraph of the committee’s report made it clear: The Commission especially sees again with some regret that between the Opéra-Comique and the Opéra-National there is a perpetual conflict of ambition resulting in court cases, such as the disagreement to which the revival of the opera Aline will probably lead.104 The Opéra-National’s requests to mount performances of Adam’s Le brasseur de Preston and Donizetti’s La fille du régiment seem to have met with responses that betray attempts to permit breaches to its licence in order to ensure the survival of the ‘third lyric theatre’, the establishment of which had cost the state and the commission so much effort. Both works had been premiered during the period of Crosnier’s management (1838 and 1840, respectively), were therefore not in the public domain and under no circumstances could have formed part of the Opéra-National’s repertory according to its licence. Why the commission should have felt it necessary to compromise and allow the Opéra-National to perform the two works if Basset had not mounted his own performances by 1 October 1848 is far from clear.105 Even more remarkable are the series of performances of Le brasseur de Preston that took place from 22 January 1848 – in advance

104 ‘Ce que la Commission revoit surtout avec peine, c’est qu’il s’établit entre l’Opéra-Comique et l’Opéra National, un conflit perpétuel de prétentions aboutissant à des procès, tels, par exemple, que celui qu’amènera probablement la reprise de l’opéra d’Aline’. Commission des Théâtres. Registre des Procès-Verbaux des délibérations, 27 September 1847 (F-Pan F21 4634 (1)). This document identifies – in the use of the phrase ‘conflit ­p erpétuel’ – the consequences of institutional competition about as clearly as it is possible to be. 105 Commission des Théâtres. Registre des Procès-Verbaux des délibérations, 21 and 25 December 1847 (F-Pan F21 4634 (1)).

The music of power  41 of Basset’s deadline – and in blatant defiance of the commission’s ruling. Whether this laissez-faire attitude on the part of the commission was a desire to allow the Opéra-National more repertorial room for manoeuvre than its licence allowed or whether it was merely a consequence of the political instability at the end of the 1840s is an open question. Whatever the answer, this is a critical moment in the negotiation of power in ­n ineteenth-century Parisian stage music where – having spent nearly half a century attempting to avoid it – the state allowed the development of a clear competition between the Opéra-Comique and the Opéra-National by allowing both to mount productions of opéra comique. It is not d ­ ifficult – given the subsequent history of the Opéra National as the Théâtre-­ Lyrique – to regard the 1847–1848 season as a tipping point in the fortunes of the licensing system. In the light of these – at best inconsistent – defences of the Opéra-­Comique’s repertory, the extraordinary permission granted to the Théâtre-­Lyrique to produce opéra comique seems all the less remarkable, and the complex history of the Théâtre-Lyrique’s licence makes clear the Opéra-Comique’s impossible position in dealing with the predecessors of the Théâtre-Lyrique from the late-1840s onwards. But other organisations that promoted stage works with less ambitious musical establishments were watching events at the Théâtre-Lyrique with interest. In the summer of 1853, for example, the manager of the Théâtre des Variétés, Aimé Carpier, began a surreptitious campaign to enlarge his repertory of ‘petites pièces dans le genre grivois, poissard ou villageois, quelquefois mêlées de couplets également sur des airs connus’ by gently testing the waters with opéra comique.106 In late June, he requested permission from the Ministry of the Interior to put on a production of Favart’s Les trois sultanes. Although the work originated in 1762 and existed in a variety of forms with various musical configurations, it was undoubtedly the property of the Opéra-Comique, and the Minister of the ­Interior – who at this point in the Second Empire was responsible for the Variétés but not for the Opéra-Comique – asked the advice of the ­Minister of State, then responsible for the imperial theatres of which the Opéra-­ Comique was one. While there was no objection to the appropriation of Favart’s libretto at the Variétés, the author of the report written for the Minister of State had heard a rumour that Carpier was planning on including musical scenes in the work, possibly using artists from the Opéra-Comique and Opéra.107 Authorisation was given later in June, but Carpier came back to the Minister in August to request permission – as the author of the earlier ministerial report had anticipated – to include in the work ‘Les adieux de Marie Stuart’, from Niedermeyer’s Marie Stuart, and the Tyrolienne,

1 06 F-Pan F21 1133. 107 Undated report [shortly after 23 June 1853] for the Minister of State (ibidem).

42  The music of power from Donizetti’s Betly, ossia La capanna svizzera.108 The Minister of State could simply have asked whether or not these were airs connus, and therefore within the remit of the Variétés; the answer would have been that ‘Les adieux de Marie Stuart’ fell into this category (Marie Stuart had been premiered at the Opéra in 1844, and the individual aria had been frequently reprinted),109 but the case of the Tyrolienne from Betly was not so clear.110 However, rather than invoke what had up until the early 1850s been routine checks, Fould was perfectly happy to disallow the latter borrowing, because a performance of Betly was planned at the Opéra and indeed took place in December the same year.111 The instance of the Variétés is important for three reasons: it is evidence of a secondary theatre successfully broadening its repertory by appropriating a libretto from the Opéra-Comique and – within certain constraints – music from the Opéra. More important is that it shows the Minister of State turning his back on the mechanics of licences and repertory – the technologies of power – and acting ad hoc, while it was clear that he still understood the question of genre as a central issue in the politics of Parisian theatres; he wrote ‘Given the exceptional generosity of which Les trois sultanes has been the beneficiary, I have not the slightest intention of opening up the question of genre again [emphasis added]’.112 And finally, the exchange demonstrates the difficulty of authoritative decision – arbitration within the politics of genre – when secondary theatres, under the control of one ministry, were in competition with imperial theatres, under the control of another. Later in 1853, the Minister of State heard about a performance of Offenbach’s first opérette, Pépito, at the Variétés (premiered 28 October 1853). He wrote to his colleague at the Ministry of the Interior in terms that precisely focussed many of the issues in the relationship between operatic genre and power during the Second Empire: 108 Request from the Minister of the Interior to the Minister of State, 17 August 1853 (ibidem). 1 09 It is possible that Fould’s generosity in allowing ‘Les adieux de Marie Stuart’ was prompted by a confusion with settings of the romance of the same name (by Beranger and others) in circulation by 1853: Apolline Barrière wrote a setting of her own version of the poem in 1843, and Gabriel de Galembert set Beranger’s poem in 1845. Richard Wagner’s 1840 setting was not published until 1913 and was therefore unknown in 1853. 110 Betly had been premiered at the Teatro Nuovo in Naples on 21 August 1836, but there is no evidence that it was ever performed in Italian at Paris’ Théâtre-Italien. See Alfred Loewenberg, Annals of Opera: 1597–1940, 3rd revised and corrected ed. (London: Calder, 1978), 781–782. See also Soubies, Le Théâtre-Italien de 1801 à 1913 (Paris: Fischbacher, 1913), v. There had, however, been four performances at the Opéra, in French, in early 1847. 111 Response from the Minister of State to the Minister of the Interior, 18 August 1853 (F-Pan F21 1133). Fould seemed unaware of the 1847 performances, or – at least – he did not mention them in his response. Two performances took place in December 1853 and two in January 1854, both in the context as a petit opéra supporting ballet. 112 ‘attendu la faveur exceptionnelle dont la représentation des Trois sultanes a été l’objet, je n’ai nullement l’intention de soulever de nouveaux la question du genre’ (ibidem).

The music of power  43 The conservation of genres being of an equal interest to the prosperity of all Parisian theatres, whether they depend on your administration or whether they are attached to mine, I would be grateful, for my part, if you could, in the name of this common interest, ensure that managers are in future restrained within the limits imposed by their cahiers de charges, as a reciprocal guarantee, on each dramatic enterprise.113 Although Fould was only asking Jean-Gilbert Victor Fialin, Comte de Persigny, to ensure that ‘in the future’ managers of secondary theatres should be kept within the limits imposed by their cahiers des charges, the latter ensured that Carpier withdrew Pépito forthwith. Again, important in this extract from Fould’s request to Persigny is the very clear understanding that the organisation of operatic power in Second Empire Paris required two separate ministries to work in concert, cooperation that could not always be guaranteed and that always left open the opportunity for an ingenious manager to profit from the sort of ad hoc arrangements put in place for Les trois sultanes. Offenbach, who had seen these arrangements at work at the Variétés in 1853, has already been seen to have been the most ambitious and wily entrepreneur on the operatic stage in the late 1850s, and his manipulation of the politics of genre at the Théâtre des Bouffes-Parisiens in the late 1850s were to have great consequences. Within the domain of the grands théâtres, the most public exposure of the politics of genre related to the question of opera in translation and largely concerned attempts by the Opéra to mount French translations of ­Rossini and Verdi, works that the Théâtre-Italien considered its exclusive repertory. From 1844 until the end of the licensing period, successive managers of the Théâtre-Italien attempted to stop the inexorable migration of their repertory to the Opéra. The Théâtre-Italien had got used to the idea that its repertory could not be appropriated by others during the previous two decades. An otherwise legal translation of La gazza ladra at the Odéon in 1826 as La pie voleuse attempted to smuggle in a duet from Semiramide, which had yet to appear at the Théâtre-Italien; immediate complaints resulted in the prompt removal of the duet from the work, which – purged of its illegal elements – was allowed to continue its run.114 But later attempts to keep the Théâtre-Italien’s repertory to itself – Othello in 1844, Lucie de Lammermoor in 1846, Robert Bruce in 1846, Louise Miller in 1853, Le trouvère in 1857

113 ‘Le maintien des genres intéressant à un égal degré la prospérité de tous les théâtres de Paris, soit qu’ils dépendent de votre administration, soit qu’ils se rattachent à la mienne, je vous serait reconnaissant, pour ma part, de vouloir bien, au nom de cet intérêt commun, faire en sorte que les directeurs de spectacle soient retenus à l’avenir dans les limites que les cahiers de charges imposent, comme une garantie réciproque, à chacune des entreprises dramatiques’, Minister of State to the Minister of the Interior, 3 November 1853 (F-Pan F21 1133). 114 See Everist, Music Drama at the Paris Odéon, 239–241.

44  The music of power and  Sémiramis in 1860 – were all destined to failure, and the Opéra, for different reasons in almost every case, was able to triumph over the rights of its competitor. The question of Otello was critical, since this was the first time that the Opéra had attempted to mount a production of a work that had been central to the Théâtre-Italien’s repertory; Otello had been a mainstay of its repertory since 1821.115 The difficulty was rooted in the fact that since 1840, the two opera houses, the Théâtre-Italien and the Opéra, had licences that very slightly overlapped, and this overlap could only be separated by a strict – and modern – differentiation between ‘Italian model’ and ‘French translation’ that was completely foreign to Parisian operatic mores of the July Monarchy. In general, such an opera as Lucie de Lammermoor (the French translation of Lucia di Lammermoor, premiered at the Théâtre de la Renaissance in 1839) was considered within the domain of the Parisian politics of genre to be the same work as Lucia di Lammermoor itself, with Le trouvère the same as Il trovatore and Othello the same as Otello. In all cases, both the question of translation and the nature of any changes, omissions or substitutions that the translation may have entailed were irrelevant to this fundamental question of identity, which depended effectively on the literary source of the libretto. In this context, the arguments surrounding Otello and Othello in 1844 were close to specious. Pillet’s 1840 cahier des charges, in addition to stipulating the number of grands opéras and ballets he was required to put on each season, required ‘a second opera in two acts, or two operas in one act, of which one could be replaced by a translated work with foreign music [un ouvrage traduit avec de la musique étrangère]’.116 To many at the time, this would have seemed to run counter to the clause in Eugène Watel’s cahier des charges which guaranteed that the state ‘promises not to authorise at any Parisian or suburban opera house the performance of lyric works in a foreign language’.117 The decision as to where the ownership of the work lay, resided in the hands of the Commission des théâtres, in whose report the key words were: Indeed, if on the one hand, the manager of the Théâtre-Italien is granted by his cahier des charges the exclusive licence of lyric works in a foreign language, on the other, it is no less certain that the cahier des charges of

115 From 1821 to 1844, Otello had been off the boards of the Théâtre-Italien for a single year only, in 1842. Soubies, Le Théâtre-Italien, v. 116 ‘un second opéra en deux actes, ou deux opéras en un acte dont l’un pourrait remplacé par un ouvrage traduit avec de la musique étrangère’, Pillet’s cahier de charges, 1 June 1840 (F21 4655 (5)). 117 ‘s’engage à n’autoriser sur aucun Théâtre de Paris ou de la banlieue de Paris, la représentation d’ouvrages lyriques en langue étrangère’. This was exactly what Watel cited in his letter to the Minister of the Interior, 25 May 1844, noting incidentally the fact that this cahier des charges was less than a month old (F-Pan AJ13 183).

The music of power  45 the Académie Royale de Musique enshrines … past usage in the right of playing translations of foreign lyric works, and according to current knowledge, it is a question of a translation.118 Faced with an obvious overlap in the terms of the two licences, the Commission found a way out that tried to distinguish, for perhaps the first and last time during the licensing period, between ‘ouvrages lyriques en langue étrangère’ and ‘traductions d’ouvrages lyriques étrangers’. After Pillet had succeeded in mounting Lucie de Lammermoor at the Opéra, translations of foreign operas were restricted, in the cahier des charges given to the new team established in 1847, to works that had not yet been performed in France.119 Had this stipulation been in place a decade earlier, neither Otello nor Lucia di Lammermoor could have transferred from the Théâtre-Italien to the Opéra. But this new situation created its own difficulties. In late 1852, both institutions were racing to mount performances of Verdi’s Luisa Miller; if the Théâtre-Italien managed to mount the work first, then this would effectively have prevented the Opéra from profiting from its own performances of the opera in translation. Within a week of each other, both managers had written to the Minister of the Interior to complain about his rival’s plans to mount the opera. Arguments were finely balanced: although Luisa Miller had been approved by the minister for performance at the Théâtre-Italien on 2 November 1852, the Opéra had obtained the score in 1851 and contracted with the translator on 20 September 1852, and with the soprano Rita Bassa Borio, expressly for the production.120 While performances at the Théâtre-Italien had begun on 7 December 1852, rehearsals had started at the Opéra a month earlier. And finally, the Théâtre-Italien – as a result of misunderstanding – had only submitted the libretto to the ministry on 3 December 1852, whereas the Opéra had submitted theirs 48 hours earlier. Both organisations had already spent significant amounts of the Empire’s money on productions of the same work, and the only solution for the Minister of the Interior was to allow both productions to proceed, even though he knew that the experienced manager of the Opéra, Roqueplan, had got the better of the very inexperienced manager of the Théâtre-Italien, Alexandre Corti (only appointed eight weeks previously) at the expense of

118 ‘En effet, si, d’un côté, le directeur du Théâtre Italien est investi par son cahier des charges du privilège exclusif de représenter des ouvrages lyriques en langue étrangère, de l’autre, il n’est pas moins certain que le cahier des charges de l’Académie Royale de Musique consacre … l’usage passé en force de droit de jouer des traductions d’ouvrages lyriques étrangers, et, dans la connaissance actuelle, c’est d’une traduction qu’il s’agit’. Report from Commission spéciale des théâtres royaux, 8 July 1844 (F-Pan AJ13 183). 119 See the cahier des charges granted to Pillet, Duponchel and Roqueplan, 1 August 1847. F21 4655 (5). 120 The biographical account in Kutsch and Riemens, Großes Sängerlexikon, 1:276, mentions no activity in Paris.

46  The music of power the licensing system itself.121 The minister almost admitted as much when he told Corti that ‘exceptional circumstances have forced a decision on me that, however, appears just; but your repertory will not otherwise suffer because of it, and I shall watch out in the future for what should be reserved for you alone’.122 The Théâtre-Italien’s plans to produce Donizetti’s Poliuto in 1858 ought not to have become embroiled in the sorts of struggles experienced with Otello and Luisa Miller, but Royer at the Opéra was, however, able to manipulate the situation to his advantage. He wanted to mount a production of Semiramide in French translation, but since the work had been part of the repertory of the Théâtre-Italien since 1826, was prevented from doing so by his cahier des charges. To overcome this difficulty, he objected to the production of Poliuto and forced a compromise that permitted the Théâtre-Italien to put on Poliuto in return for permission to perform Semiramide. He was able to do this by claiming – and he most likely knew that this was false – that Poliuto (originally composed in 1838) was an Italian version of the original French Les martyrs (reworked by Donizetti from his music for Poliuto in 1840), based in turn on Corneille’s Polyeucte, and that the Théâtre-Italien was therefore trespassing on the artistic property of the Opéra.123 His ability to carry off this subterfuge (since it neatly avoided the reality of Poliuto being the musical source for most of Les martyrs) raises further questions about the integrity of a work when it existed in an Italian and a French version; here, it suited Royer’s purposes to claim that Poliuto and Les martyrs were identical, whereas it had suited his predecessor, Pillet, to argue that the French and Italian version of Rossini’s Otello were effectively different.124 In comparison with Royer’s virtuosic management of a set of power structures to his own ends (both Poliuto and Sémiramis, as the French version of Semiramide was known, found their way to the stage), the attempts in 1863 by the manager of the Théâtre-Italien, Prosper Bagier, to stop the Théâtre-­ Lyrique putting on Rigoletto in French translation seem positively ­infantile. In making his argument to the ministry that the Théâtre-Lyrique had to

121 The relevant documents are dated 10, 17 and 28 (the formal report from the Minister to the Commissaire Impérial près des Théâtres lyriques [Édouard Monnais]) December 1852 (F-Pan AJ13 183). 122 ‘des considérations exceptionnelles m’ont déterminé à une mesure qui, d’ailleurs, me paraissait juste; mais votre répertoire n’en suffira autrement et je veillerai dans l’avenir à ce qu’il vous soit conservé’, Minister of the Interior to Corti, 6 January 1853 (F-Pan F21 1116). 123 Poliuto was not performed in Italian until 1848, well after Italian translations of Les martyrs had been circulating in Italy for some time. 124 See the letter from Royer to Ministry (Gautier), 13 September 1858, F-Pan AJ13 451 (I), and the series of well-documented negotiations between September 1858 and March 1859 (F-Pan F21 1116). For Royer’s successful attempt to use the affair to permit him to play Semiramide in 1860, see the letters between Royer and Torribio Calzada at the ­T héâtre-Italien, 13 and 27 October 1859 (F-Pan AJ13 501).

The music of power  47 wait ten years after the Parisian premiere of a work, he was correctly quoting its original 1851 cahier des charges, but he had failed to recognise the key changes granted to Carvalho in his 1856 revision which made R ­ igoletto an entirely permissible production.125 The Opéra usually got its own way, especially after it received special status during the central part of the Second Empire. Exceptions arose when the licences and cahiers de charges were stretched to their limits. Dance, especially ballets du genre noble et gracieux, was a prized monopoly for the Opéra that it protected with care. When the Théâtre de la Porte Saint-­ Martin opened in 1814 with a licence that permitted mélodrames and pantomimes with dance, there was always a risk that it would attempt repertory that fell into the category of dance promoted at the Opéra; by 1822, this was such a problem that the Minister of the Interior, under pressure from the Maison du Roi, undertook to purge the secondary theatres and the Porte Saint-Martin in particular of this particular abuse, with clear success.126 Thirty years later, however, the Opéra would still be fighting to protect its monopoly over dance. Halévy had written his Italian opera La tempesta for a performance at Her Majesty’s Theatre in London in June 1850; given the venue for the performance, it included a substantial divertissement. In October 1850, Benjamin Lumley, the manager of Her Majesty’s from 1842 to 1858, took over the simultaneous management of the Théâtre-Italien (he remained for exactly two years).127 It was no surprise that he should want to bring back an Italian opera by Halévy, which would put one of the most important composers of grand opéra on the stage of its biggest rival for the first time since 1828. There was nothing the Opéra could do to stop this except to stress, in similar terms to those used in 1822, its monopoly on ‘noble and gracious ballets’ and to argue that the ballets in La tempesta should be removed.128 Jules Baroche replied, without any reference to the commission, that dance would be formally forbidden in the new cahier des charges negotiated with Lumley; Baroche also promised that there would be a clause that forbade the Théâtre-Italien from taking an artist from the Opéra unless

125 Bagier’s letter to the Superintendent of Theatres, 16 December 1863, and the latter’s response, 23 December 1863, are preserved in F-Pan F21 1116. 126 The prompt attention to this matter by the Maison du Roi is both impressive and unsurprising. See the correspondence between the Maison du Roi, the Minister of the Interior and the Théâtre de la Porte Saint-Martin, 13–29 June 1822 (F-Pan F21 1128). 127 The episode is discussed in Lumley’s entry in the DNB: L. M. Middleton, ‘Lumley, ­Benjamin (1811/12–1875)’, rev. John Rosselli, Oxford Dictionary of National Biography, eds. H. C. G. Matthew and Brian Harrison, Oxford: OUP, 2004 (Accessed 10 July 2006). www.oxforddnb.com/view/article/17174. 128 Roqueplan wrote to the Minister of the Interior (Jules Baroche), copied to the manager of the Théâtre-Italien (Lumley) on 3 October 1850. The problem was complicated by the fact that Lumley’s cahier des charges was still under negotiation (F-Pan AJ13 180 (X)) and that Baroche would leave the Ministry of the Interior for less than a year at the Ministry of Foreign Affairs on 24 January 1851.

48  The music of power they had already been discharged for at least two years.129 Almost none of these promises were kept, and the episode reveals a critical dislocation between the Minister of the Interior and the commission that was typical of the republican period. Baroche assumed powers that were technically his, but the detail of which were beyond his competence. This difficulty resulted in the members of the Commission des théâtres attending a dress rehearsal in order to arbitrate between the Opéra’s claims that the work breached the Théâtre-Italien’s cahier des charges and Lumley’s argument that La tempesta could not be mounted without dance. The solution was to permit a single dancer to act in La tempesta, but to forbid the participation of any corps de ballet; this satisfied the regulatory impulses of the Commission des théâtres and apparently both Lumley and Roqueplan.130 The negotiation of the politics of genre was a central concern for every Parisian operatic and theatrical enterprise, from the unpretentiously modest to those with pan-European ambitions. The state’s attempts to control the matrix of power ran the complete gamut, from inflexibility to caprice, from incompetence to closely argued compromise. Experienced managers could outmanoeuvre their rivals and in some cases circumvent state regulation entirely, renegotiating the balance of power and ensuring that it was as pervasive as Foucault was later to claim. For the state, the licensing system meant that it could ensure that overlapping activities that could sap the partially state-funded resources of the grands théâtres were avoided, and that those resources were not threatened by secondary theatres. Predictably, the managers of established opera houses were by and large happy with the arrangements and fought hard to protect their interests; even the managers of new opera houses, or theatres trying to break into the lyric repertory, never felt inclined to challenge the efficacy of the licensing system. But its cost to them could be great: it could be argued that the failure of the Odéon’s opera troupe in 1828 was a result of its failure to gain permission to mount productions of new opéra comique, a failure that could only be staved off by the – admittedly very successful – exploitation of Italian and German stage music. Even more starkly, the exploitation of the licensing system and courts by the Opéra-Comique and Opéra brought about a similar collapse of the opera troupe at the Renaissance in 1840, and the institution’s demise. When Napoléon III announced to the Chamber of Deputies on 5 November 1864 that the licensing system for the theatres was to end and the legislation finally promulgated, he was bowing to the inevitable, but only in the knowledge that the impact of the ‘liberty of the theatres’ had already

129 Baroche’s reply was sent the day Roqueplan’s letter was received (4 October 1850; (F-Pan AJ13 180 (X)). 130 Commission des Théâtres. Registre des Procès-Verbaux des délibérations, 18 and 21 February 1851 (Pan F21 4634 (2)).

The music of power  49 131

been vitiated. Part of the transformation of the city of Paris undertaken by Baron Georges-Eugène Haussmann, Prefect of the Seine from 1853 to 1870, was the construction of the Boulevard du Prince-Eugène (now Boulevard Voltaire)132 that involved the demolition of the northern part of the existing Boulevard du Temple and which destroyed all the theatres that had been housed there. The Délassements-Comiques, Folies-Dramatiques, Gaîté, Funambules, Petit-Lazari, Théâtre-Lyrique and the Cirque Olympique were removed from the map.133 While the Théâtre-Lyrique, Gaîté and Cirque-Olympique were rehoused at the expense of the city, the others simply lost their buildings. And these institutions were exactly those who might have benefitted from a genuine freedom to develop whatever sort of stage music they wanted, who might have gained much from what many thought would be a step change in the dispersal of operatic power. Napoléon III might well have been thought to be acting more liberally in 1863–1864 when he announced the freedom of the theatres, and this would certainly accord with the rest of his domestic policy, finely gauged to attract more and more supporters to the principles espoused by the Second Empire. But the 1850s had seen a progressive loosening of state control over the entire network of operatic institutions. The close, almost suffocating, attention given the Opéra at the expense of other institutions worked to the latters’ advantage: the Théâtre-Lyrique carefully profited from the fruits of a decade’s negotiation, and the consequence for the Opéra-Comique of having to share generic power with another institution was to allow the creative power of its librettists and composers to inflate the genre well beyond its traditional limits, which also had a later effect on the ambitions of opéra comique at the Théâtre-Lyrique. The creation of opéras comiques on such an unprecedented scale as Thomas’ Le songe d’un nuit d’été, Adam’s Giralda, or Halévy’s La dame de pique134 paved the way for Meyerbeer’s move into this generic field with L’étoile du nord in 1854.135 And this in turn gave space for Hervé and Offenbach to claim that there was now a generic gap – opéra

131 Hostein, La liberté des théâtres, 8. The imperial decree was promulgated on 6 January 1864 ibidem, 10–11). 132 The literature on Haussmann and the renovation (or destruction, depending on intellectual and artistic perspective) of Paris is extensive. See, for a changing view over three decades, Henri Malet, Le Baron Haussmann et la rénovation de Paris (Paris: Les Éditions municipales, 1973); Donald J. Olsen, The City as a Work of Art: London, Paris, Vienna (New Haven, CT: Yale University Press, 1988); David P. Jordan, Transforming Paris: the Life and Labors of Baron Haussmann (New York: Free Press, 1995). 133 See the chronological diagram of the closure of Parisian theatres in Wild, Dictionnaire, figure 3, p.[445]. For an account of the end of the licensing system, which for the first time attempts to assess its consequences both negative and positive, see Katherine Ellis, ‘Unintended Consequences: Theatre Deregulation and Opera in France (1864–1878)’, Cambridge Opera Journal 22 (2010), 327–352. 134 Premiered at the Opéra-Comique on 20 April 1850, 20 July 1850, and 28 December 1850. 135 L’étoile du nord received its premiere at the Opéra-Comique on 16 February 1854.

50  The music of power comique of the lighter sort – that needing filling. The fact that Offenbach’s ambitions lay in much larger works, and his stunning ability to circumvent his licence in the late 1850s, meant that within the space of a few months, such works as Gounod’s Faust at the Théâtre-Lyrique, Offenbach’s Orphée aux enfers at the Bouffes-Parisiens and Meyerbeer’s Le pardon de Ploërmel at the Opéra-Comique could be seen.136 All were works of ambition, and all merged spoken dialogue and music, as had been the traditional monopoly only a decade earlier of the Opéra-Comique. Older commentators would have been hard-pressed to separate out these three works in generic terms as they were understood before the 1848 Revolution, and the implicit breach of fundamental generic principles that had held sway since the Empire rendered those principles impossible to sustain. A single document is emblematic of the demise of the licensing system: the manuscript libretto to Georges Bizet’s Les pêcheurs de perles betrays a curious mixture of opéra comique and continuous opera, and in the months during which Napoléon III was considering the legislation that would end the licensing period, Bizet rewrote all the dialogue as recitative to create the first continuous opera at the ­Théâtre-Lyrique.137 He was in flagrant breach of the opera house’s licence, but with such a gesture, the politics of genre that had dominated the capital of the nineteenth-century became as much part of the past as the debris of the demolished theatres on the Boulevard du Temple. *** Networks of operatic power contained and disciplined the creative energies expended by librettists, composers, stage designers and managers in nineteenth-century Paris. The relationship between state bureaucracies, managers, creative artists and the commissions designed to broker various 136 See above 28–29 for Offenbach, and for the original version of Faust, Steven Huebner, The Operas of Charles Gounod (Oxford: Clarendon Press, 1990), 107–129. Although the version of Faust with recitatives was prepared for other opera houses as early as 1860, the Théâtre-Lyrique continued to perform the work in its original form with spoken dialogue as late as 1866 (ibidem, 129). 137 On the basis of the libretti (censors’ libretto, F-Pan F18 737, deposited 11 August 1863; and two others in F-Pan AJ13 1158), the Choudens 1863 piano-vocal score and the manuscript violin conductor score (F-Po Mat. 19. 1863), Hervé Lacombe has been able to show how the original form of the work consisted of a first act in the form of traditional opéra comique (four numbers alternating with dialogue) followed by two acts in which all the libretto is set as continuous music. Whether the document is interpreted correctly, and whether there might not have existed an earlier version of the entire work in the shape of an opéra comique, is impossible to tell from the surviving material. See Hervé Lacombe, ‘Les pêcheurs de perles de Bizet: contribution à l’étude de l’opéra français au xixe siècle’ (Thèse de Doctorat, Université François Rabelais de Tours, 1993), 46–52. The tabular material is reprinted in idem, Les voies de l’opéra français au xixe siècle, Les Chemins de la Musique (Paris: Fayard, 1997), 316–317. See also Lesley A. Wright, ‘Les pêcheurs de perles: Before the Premiere’, Studies in Music 20 (1986), 27–45.

The music of power  51 interests rendered the negotiation of the politics of genre as much a part of the operatic culture of the period as the Opéra’s use of continuous recitative and ballet or the Théâtre-Italien’s concern for musical events beyond the Alps. The underpinning technologies of power were no guarantee, however, against the involvement of such plutocrats as Aguado or such ambitious and cunning politicians as Fould or Walewski, or of the adroit management of individual opera houses and theatres. When the state came close to the superstructure of operatic creation – when it attempted to invent genres to solve particular problems – it was a striking failure. While the support for the Théâtre de la Renaissance in the late 1830s may be viewed as part of the response to the creation of the third lyric theatre that would eventually result in the creation of the Théâtre-­ Lyrique, the generic fabrications – vaudeville avec airs nouveaux and opéra de genre – merely threw the opera house into conflict with others. And the greatest continual problem at the Opéra was finding petits opéras in two acts to accompany a ballet; while creative energies were being expended elsewhere, it was the legislative requirement of the state that meant, now that the preferred form of operatic creation was the five-act work on a historical libretto, that ballet still had to be promoted, an unchanging orthodoxy which posed problems that the Opéra itself never really resolved. Even when enmeshed in the complexities of the politics of genre, the institutional structures of French opera – their governing technologies of power  – served as a context for the exercise of creative power. The ­decade-long emergence of grand opéra hurdled the July Revolution of 1830, as Scribe, ­Auber, Meyerbeer and Halévy forged what would constitute the ­c entre of European music drama for the middle third of the nineteenth ­c entury. And while it is unlikely that anyone sitting on the revolutionary committee discussing new theatrical law in 1849 could have envisaged opérette, the expansion of opéra comique or Gounod’s Faust, the creation of the ­Théâtre-Lyrique was an institutional catalyst for the creation of works as different as ­Meyerbeer’s Le pardon de Ploërmel, Bizet’s Les pêcheurs de perles and Offenbach’s La grande-duchesse de Gérolstein. As we move away from nineteenth-­c entury Paris, the politics of genre remains close to our critical practice. Technologies of power are as easily deployed by ­municipalities, free cities and ­commercial and charitable bodies as they are by constitutional monarchies or empires. But however it is manifested, power and opera are as negotiable in seventeenth-century Italy or twenty-first-century Hanoi as they are in nineteenth-century Paris, and Foucault’s view of power relations remains a central framework for their understanding. Whether the headline judgement on Parisian operatic history in the first two-thirds of the nineteenth century is one that identifies the coercive or creative is likely to be coloured by the critic’s ability to make sense of what little of the past is recoverable. It is certainly true that the pan-European success of grand opéra, opéra comique and comédie-vaudeville was a result of the perceived quality of the works being exported, not of the context in

52  The music of power which they were created, but it is equally the case that the dispersal of power outlined in this article created the environment that enabled those works to succeed. The politics of genre described in this article underpinned not only the production of work of transcontinental significance but also the reception of foreign works in Paris from Il barbiere di Siviglia to Tannhäuser. This cultural exchange marks out Parisian stage music as the central trigger for musical culture in the nineteenth century. An analysis of the politics of genre and of the ways in which technologies of power were negotiated suggests a view – largely as Foucault described – of power dispersed across the field ready to be seized by any agent willing to engage with, or work around, its technologies. And while it was clearly the grands opéras, opéras comiques and comédies-vaudevilles themselves that took Europe by storm in the middle third of the nineteenth century and beyond, the attraction of Parisian operatic culture to foreigners from Spontini to Wagner depended at least in part on the networks of power described here and the resources and prestige they underpinned. A perspective on the relationship between creative and institutional power recognises the need for an organisational framework which could keep an operatic culture that promoted continuous opera in French, opéra comique and Italian opera in equilibrium. At the same time, it also accepts that ad hoc arrangements, especially in the 1840s and 1850s, created opportunities for those that the technologies of power were supposed to discipline. To acknowledge both the limitations imposed and the opportunities created by the functioning of this institutional framework is key to reconstructing a regime of truth that promotes a fuller understanding of the position of Parisian stage music at the centre of European nineteenth-century experience.

2 Grand opéra – petit opéra Parisian opera and ballet from the Restoration to the Second Empire

In the short term, the Opéra continued using older, smaller-scale works to accompany ballet as part of the same evening’s entertainment, but despite the canonic pressures this exerted, it was a practice that could not be sustained indefinitely. A second alternative was to shorten up-todate grand opéra, to bring them down to dimensions at which they could be performed with ballets; the best known example of this procedure is the reduction of Rossini’s Guillaume Tell from four acts (1829) to three (1831). A third possibility was the process of morcellement: the extraction of ­i ndividual acts or pairs of acts from grand opéra and their performance alongside ballet. A longer-term strategy, and one so far entirely ignored in modern scholarship, was the development of a new type of stage music specifically to accompany ballet: petit opéra. Emerging from two closely related works at the same time as the birth of grand opéra – Rossini’s Le comte Ory (1828) and Auber’s Le philtre (1831) – petit opéra was the direct result of institutional pressure from the state for the Opéra to mount productions of both opera and ballet, the preference of the institution and its audiences for evenings with both opera and ballet, and the indirect aesthetic pressure engendered by the appearance of grand opéra. The tradition of composing petit opéra continued up to the end of the licensing period in 1864 and encompassed, among such foreign imports as Weber, Verdi and Donizetti, works by Halévy, Adam, Thomas, Auber and Rossini; two of the best-known casualties of the complexities surrounding the genre were Berlioz’s Benvenuto Cellini (1838) and Wagner’s scenario for Le hollandais volant. Petit opéra was characterised by libretti in two acts with a limited number of musical compositions, a clear distinction between composed number and following recitative (a decisive break with grand opéra), a limited number of characters and a comic register; it developed a set of conventions that remained consistent from its origins c1830 to its latest presentations in Alary’s La voix humaine (1861) and Massé’s La mule de Pedro (1863). French grand opéra always existed in a symbiotic relationship with dance. The divertissement within grand opéra itself defined the genre throughout the nineteenth century, and ballet-pantomime was recognised as equal in

54  Grand opéra – petit opéra importance to grand opéra in the licensing system that governed all Parisian and provincial theatres from 1806 to 1807 until 1864.1 Furthermore, the repertory of the Opéra, encompassing both opera and ballet, was enshrined in law.2 It was ‘made up of all the works, operas as much as ballets, that have appeared since its establishment in 1646. It alone may present works that are entirely musically through-composed and ballets of a noble and gracious type’.3 With appeals as much to history as to the legislative necessity to minimise the overlap between genre and institution – the principal aim of the 1806–1807 enactments – the Opéra was identified as the exclusive home of both ballets du genre noble et gracieux and pièces entièrement en musique. The latter term was glossed in the directors’ cahiers des charges as ‘grand opéra, with orchestral recitative’,4 a definition that would last not only until 1 The enactment of this legislation was complex. Napoléon’s decree which set out the distinction between grands théâtres and théâtres secondaires was promulgated on 8 June 1806 (Décret concernant les théâtres au Palais de Saint-Cloud, le 8 juin 1806 (Paris: Rondonneau et Decle, 1806), and the réglement followed logically ten months later on 25 April 1807: Ministère de l’Intérieur: Règlement pour les théâtres [25 April 1807] (Paris: Imprimerie Impériale, 1807). However, it is clear that this legislation – especially as it related to the identification of which theatres were secondaires and particularly the fate of the unauthorised theatres – met with imperial disapproval. It took a second décret from the emperor for the full impact on the rest of theatrical Paris to be felt: ‘All theatres unauthorised according to the preceding clause shall be closed by 15 August [1807]’ (‘Tous les théâtres non autorisés par l’article précédent seront fermés avant le 15 août [1807]’) (Décret impérial sur les théâtres au Palais de Saint-Cloud, le 29 juillet 1807 (Paris: Imprimerie Impériale, 1807), 2).   The term divertissement and its usage in this article require clarification. It is used exclusively to refer to a contiguous suite of dances (or rarely a single dance) within a grand opéra for which there is a formal choreography that is attributed in the same way as the music or libretto, for which the composer of the rest of the work has supplied new dance music, and for which the solo artists and corps de ballet are identified in the preface to the libretto. A divertissement, furthermore, is fully integrated into the drama and described in the libretto and also in the livret de mise-en-scène (this is a fuller account that accords with the summary in Nathalie Lecomte, ‘Divertissement d’opéra: 2’, Dictionnaire de la musique en France au xixe siècle, ed. Joël-Marie Fauquet (Paris: Fayard, 2003), 387–388). The term is used in opposition to informal dance for which the choreography is unattributed and unrecorded, for which the composer writes no ballet music and which leaves little trace in the documentary record, as witnessed in the examples noted elsewhere in this study. 2 During the period under review, the institution underwent eleven changes of name. In this study, the term ‘Opéra’ is used to subsume all titles where the discussion transcends the boundaries of a single regime. For a list of the changes of title for the Opéra, see Nicole Wild, Dictionnaire des théâtres parisiens au xixe siècle: les théâtres et la musique (Paris: Amateurs des Livres, 1989), 300. 3 ‘[Le] répertoire [du Théâtre de l’Opéra (Académie Impériale de Musique)] est composé de tous les ouvrages, tant opéras que ballets, qui ont paru depuis son établissement en 1646. Il peut seul représenter les pièces qui sont entièrement en musique et les ballets du genre noble et gracieux’ (Ministère de l’Intérieur: Règlement pour les théâtres [25 April 1807], 2). All translations in this article are my own, and the nineteenth-century orthography of the original French is retained throughout. 4 See, for example, Charles-Edmond Duponchel’s 1835 cahier des charges, which alludes directly to ‘Le grand opéra, avec récitatif à orchestre’ (Académie Royale – Cahier des Charges. M. Duponchel. 19 August 1835; Paris, Archives Nationales (hereafter F-Pan) F21 4655(3)). In

Grand opéra – petit opéra  55 the repeal of the licensing laws in 1864 (which marks the formal end-point of this study) but also to the end of the century itself.5 From at least the beginning of the century, the Opéra had taken the ­1806–1807 legislation literally: opéra and ballet had been performed during the same evening. However, the emergence during the decade around 1830 of five-act grand opéra, which itself included divertissements, made it impossible to continue this tradition by pairing ballet-pantomime with a grand opéra in the same way.6 A gap therefore emerged in the Opéra’s repertory for works that could accompany a ballet while grands opéras now occupied an entire evening. This gap was filled by a number of techniques, the most remarkable of which was the emergence of what amounted to a new kind of opera, specifically designed to accompany ballet-pantomime: in contexts ranging from the formal to the casual, it was called petit opéra.7 The absence of any discussion his attempts to dismiss grand opéra as a term without ‘secure historical credentials’, David Charlton gives insufficient weight to this type of administrative document, in which grand opéra is in common use as a synonym for the types of works promoted at the Opéra (‘Introduction’, The Cambridge Companion to Grand Opera, ed. David Charlton (Cambridge: Cambridge University Press, 2003), 3; the chapter in the same work on which he depends is reticent about the documentary trace of both institution and state (Herbert Schneider, ‘Scribe and Auber: Constructing Grand Opera’, ibidem, 168)).   While every opera house and theatre in Paris between 1806–1807 and 1864 operated under a licence drawn up by the state, the activities of each manager were controlled by a cahier des charges which circumscribed the detail of his operations. Although licences could remain unchanged from one decade to the next, a new cahier des charges emerged at each change of management. The term ‘licensing system’ is here a translation of ‘le régime du privilège’, with which it should be taken to be synonymous. The relationship between privilège, cahier des charges and mode of financial organisation changed between institutions and, frequently, within institutions themselves from one date to another. To argue, then, that the existence of manager who took financial risk himself (rather than the state) negates the idea of the licensing system is fundamentally to misunderstand the latter. See the clear statement of the position adopted in this paragraph in Nicole Wild, ‘Privilège’, Dictionnaire de la musique en France au xixe siècle, 1002; the fuller account in eadem, ‘Musique et théâtres parisiens face au pouvoir (1807–1864)’, 3 vols (Doctorat d’État: Université de Paris IV, 1987), 1–206, with a further summary in eadem, Dictionnaire des théâtres parisiens au xixe siècle, 9–19. 5 ‘Décret relatif à la liberté de l’industrie théâtrale, 6 janvier 1864’, Recueil des lois, décrets, arrêtés, règlements, circulaires, se rapportant aux théâtres et aux établissements d’enseignement musical et dramatique (Paris: Imprimerie Nationale, 1888), 11–16. 6 The standard texts on ballet-pantomime is Ivor Guest, The Romantic Ballet in Paris, 2nd edn (London: Dance Books, 1980; R 2008) and idem, The Ballet of the Second Empire, 1847– 1858 (London: Black, 1955), and on its music is Marian Smith, Ballet and Opera in the Age of Giselle, Princeton Studies in Opera (Princeton and Oxford: Princeton University Press, 2000). For ballet embedded in grand opéra, see Maribeth Clark, ‘Understanding French Grand Opera through Dance’ (PhD diss., University of Pennsylvania, 1998). In this study, the nineteenth-century difference in terminology is retained between b­ allet-pantomime to describe a free-standing work (synonymous with ballet en action or ballet d’action) and divertissements within the confines of grand opéra. See Smith, Ballet and Opera, xxi, and eadem, ‘The Orchestra as Translator: French Nineteenth-Century Ballet’, The Cambridge Companion to Ballet, ed. Marion Kant (Cambridge: Cambridge University Press, 2007), 138–150. 7 Whether petit opéra constitutes a genre is a question that may only be answered by interrogating the definitions of genre that are offered. If one adopts a modern definition from

56  Grand opéra – petit opéra of petit opéra in the modern scholarship on the Opéra is striking.8 The scope of this omission may be gauged by the number of grands opéras produced in the period 1828–1864 (44) and that of petits opéras and their surrogates (23). The two kinds account for 92% of all works mounted at the Opéra during the period;9 to its repertory, for so long understood in terms of two types only – ballet and grand opéra – may now be added a third: petit opéra.

Music, dance and drama, 1807–1828 The Opéra had aimed in its programming to mount productions of opera and ballet every night that the theatre was open up to c1830. Thus, an opera, even in three acts, would be combined with a ballet-pantomime, most frequently resulting in a total of five acts for the evening. In general, however, most operas were performed alone at the beginning of their first run, but even such highly regarded works as Spontini’s La vestale eventually lost this particular status; premiering in December 1807 and remaining in the

literary studies that stresses the polyvalence of the concept of genre, then petit genre certainly may be thus considered (see Alistair Fowler, Kinds of Literature: An Introduction to the Theory of Genres and Modes (Oxford: Oxford University Press, 1982), 55 and passim). Making the older, and largely discredited, equation between genre and typology would make it harder to justify such an identification (see Joël-Marie Fauquet, ‘Genre’, Dictionnaire de la musique en France au xixe siècle, 512–513). 8 Petit opéra fails to rate any mention in Jane Fulcher, The Nation’s Image: French Grand Opera as Politics and Politicized Art (Cambridge, etc.: Cambridge University Press, 1987) or in Anselm Gerhard, Die Verstädterung der Oper: Paris und das Musiktheater des 19. Jahrhundert (Stuttgart: Metzler, 1992) trans. Mary Whittall as The Urbanization of Opera: Music Theater in Paris in the Nineteenth Century (Chicago and London: University of ­Chicago Press, 1998). Neither is it discussed in any recent francophone account of the subject; see Jean Mongrédien, La Musique en France des Lumières au Romantisme, Harmoniques: La Musique en France (Paris: Flammarion, 1986); Nicole Wild, ‘Le Spectacle lyrique au temps du grand opéra’, La Musique en France à l’époque romantique ­(1830–1870), ­Harmoniques: La Musique en France (Paris: Flammarion, 1991), 21–57; Hervé Lacombe, Les Voies de l’opéra français au xixe siècle, Les Chemins de la Musique (Paris: Fayard, 1997); Wild, Dictionnaire des théâtres parisiens au xixe siècle, 299–321; Joël-Marie Fauquet (ed.), Dictionnaire de la musique en France au xixe siècle. The same is true of the older literature listed in Wild, Dictionnaire des théâtres parisiens au xixe siècle, 320–321. Most problematic is the attempt to recalibrate questions of genre in Parisian stage music around the idea of institution (itself largely uncontentious) without taking account of petit opéra, which leads to conclusions that are vitiated by failing to take account of the genre (Hervé Lacombe, ‘De la différenciation des genres: Réflexion sur la notion de genre lyrique français au début du xixe siècle’, Revue de musicologie 84 (1998), 247–262; see also idem, ‘Définitions des genres lyriques dans les dictionnaires français du xixe siècle’, Le Théâtre lyrique en France au xixe siècle, ed. Paul Prévost (Metz: Éditions Serpenoise, 1995), 297–334). Even in recent work dedicated to Rossini, whose Le comte Ory is central to the history of petit opéra, the genre receives no mention; see, for example, Damien Colas, Rossini: L’opéra de lumière, Decouvertes Gallimard (Paris: Gallimard, 1992), 84–85. 9 The remaining six works are sui generis and are discussed passim.

Grand opéra – petit opéra  57 repertory until 1830, it was paired with a ballet from 1817 onwards.10 Ballets, however, were never premiered alone.11 Some operas were exceptions to this general rule. Two of the best known works from the Empire, Lesueur’s Ossian and Persuis and Lesueur’s Le triomphe de Trajan, held the stage alone for a large part of their careers.12 That Ossian was in five acts was an important precursor of trends at the end of the Restoration, while the importance of Le triomphe de Trajan was directly related to its political connotations.13 Classics of the ancien régime were revived under the restored Bourbons, and new productions of Gluck’s Armide and Salieri’s Les Danaïdes, both in five acts, were – like Ossian – not accompanied by ballets.14 Examination of the programming for the year 1817 shows this traditional practice at work and is typical for the Imperial period after 1806–1807 nearly up to the end of the Restoration (Appendix 2.1). Entire months could elapse with no change to the pattern of presenting both opéra and ballet-pantomime during the same evening; almost the entire first half of the year was structured this way. Of the 158 performances given at the Opéra in 1817, 113 (72%) comprised both opéra and ballet-­pantomime; only 45 (28%) were dedicated to opéra alone. La vestale, which had started out being played alone, was now paired with the ballets Le carnaval de Venise, ou La constance à l’épreuve, Nina, La dansomanie and Flore et Zéphire.15 So too were Sacchini’s Œdipe à Colone, Grétry’s La caravane du Caire and Gluck’s 10 La vestale was premiered on 16 December 1807 and had received 200 performances by 1830 (Théodore de Lajarte, Bibliothèque musicale du Théâtre de l’Opéra: catalogue historique, chronologique, anecdotique, 2 vols (Paris: Bibliothèque Nationale, 1878; R Hildesheim: Olms, 1969), 2:63). 11 The evidence for the pairing of opera with ballet-pantomime and the shifting status of certain works is recoverable from the manuscript series entitled Journal de l’Opéra (Paris, Bibliothèque nationale de France, Bilbiothèque-Musée de l’Opéra (hereafter F-Po) MS without shelfmark) that runs throughout the nineteenth century. Whereas the Journal de l’Opéra records exactly which works were performed on a particular evening (taking account of last-minute changes that the press could only report retrospectively), it is less than reliable in cases where single acts of a work were performed; in these cases the press adds important detail. There are scattered comments on the sorts of performance patterns discussed here in Lajarte, Bibliothèque musicale, 2: passim. After completion of this article, the online resource ‘Chronopera’ became available (http://chronopera.free.fr (consulted 29 December 2008)), which provides some of the same information (mostly from the Journal de l’Opéra, although the documentation to the resource does not specify) but not, critically, details of the performance of individual acts. At the time of the final preparation of this article’s typescript, no data for the year 1817 were available (see Appendix 2.1). 12 Ossian, with a libretto by P. Dercy and Jacques-Marie Deschamps, was premiered on 10 July 1804, and Le triomphe de Trajan, libretto by Joseph Esménard, on 23 October 1807. 13 Despite conventional views on the work’s weaknesses, the political ideology of Le triomphe de Trajan cannot alone explain its success well after Napoleon’s exiles in 1814 and 1815. The work continued to be performed with success and enthusiasm almost until the July Revolution. Its final performance was on 18 February 1827. 14 There were new productions of Armide in 1811 and 1819 and of Les Danaïdes in 1817. The works had been premiered in 1771 and 1784, respectively. 15 Outlines of the authorship (music and choreography) and reception of these works is given in Lajarte, Bibliothèque musicale, 2:86–87, 81–82, 16–17 and 86.

Appendix 2.1  R  epertory of opera and ballet, Académie Royale de Musique, 1817 (titles and number of acts only; details of librettist, composer and premiere in notes to the main text) Date

Opera

Ballet

3 5 7 10 12

January January January January January

L’enfant prodigue (3)

14 16 17 19 24 26 28 31

January January January January January January January January

2 4 7 9 11 14 16 18 21 23 25 28 2 4 7 9 11 13 14 16 18 21 23 25 28 30 8 11 13 15 18 20 22 25 27

February February February February February February February Feburary February February February February March March March March March March March March March March March March March March April April April April April April April April April

Le rossignol (1) Armide (5) Œdipe à Colone (3) La caravane du Caire (3) Le rossignol (1); Le devin du village [1] Œdipe à Colone (3) Panurge (3) Le rossignol (1) Panurge (3) Panurge (3) La caravane du Caire (3) Aristippe (2) Le rossignol (1); Le devin du village [1] Panurge (3) Iphigénie en Aulide (3) Les bayadères (3) Le rossignol (1) Panurge (3) Œdipe à Colone (3) La caravane du Caire (3) Les rossignol (1) Les mystères d’Isis (4) Le rossignol (1) La caravane du Caire (3) La vestale (3) Le rossignol (1) Roger de Sicile (3) [premiere] Roger de Sicile (3) Alceste (3) Roger de Sicile (3) La mort d’Abel (3) La caravane du Caire (3) Le rossignol (1) Roger de Sicile (3) Œdipe à Colone (3) Les bayadères (3) Roger de Sicile (3) Iphigénie en Aulide (3) La caravane du Caire (3) La vestale (3) Panurge (3) Les bayadères (3) Roger de Sicile (3) Le rossignol (1) Panurge (3) Armide (5) Iphigénie en Tauride (4) Le rossignol (1)

Nina (2) Paul et Virginie (3) Les sauvages (1) La dansomanie (2) L’épreuve villageoise (2) Flore et Zéphire (2) Nina (2) Paul et Virginie (3) Les sauvages (1) Achille à Scyros (3) Les sauvages (1) Les noces de Gamache (2) Flore et Zéphire (2) L’enfant prodigue (3) Flore et Zéphire (2) Nina (2) Le carnaval de Venise (2) Les noces de Gamache (2) Le carnaval de Venise (2) Nina (2) Le carnaval de Venise (2) Flore et Zéphire (2) Le carnaval de Venise (2) Flore et Zéphire (2) Vénus et Adonis (1) L’épreuve villageoise (2) Le carnaval de Venise (2) Nina (2) Le bourgeois gentilhomme (1) Flore et Zéphire (2) Les noces de Gamache (2) La dansomanie (2) Paul et Virginie (3) Le carnaval de Venise (2) Flore et Zéphire (2) Nina (2) Flore et Zéphire (2) La dansomanie (2) Les sauvages (1) L’épreuve villageoise (2) Flore et Zéphire (2) Achille à Scyros (3) Nina (2) Flore et Zéphire (2) L’enfant prodigue (3)

Date

Opera

Ballet

29 2 5 7 9

April May May May May

Le carnaval de Venise (2) L’heureux retour (1) Flore et Zéphire (2) Nina (2)

12 14 16 19 21 23 26 28

May May May May May May May May

30 2 4 6 7 9 11 13 16 18 20 23 25 27 30 2 4 7 9 11 14 16 18 21 23 25 28 30 1 4 6 8 11 13 15

May June June June June June June June June June June June June June June July July July July July July July July July July July July July August August August August August August August

La vestale (3) Œdipe à Colone (3) Anacréon chez Polyeucte (3) Les bayadères (3) Le rossignol (1); La caravane du Caire (3) Iphigénie en Aulide (3) Aristippe (2); Le devin du village [1] Les bayadères (3) Didon (3) Le rossignol (1) Les prétendus (1) La vestale (3) Fernand Cortez (3) [new production] Fernand Cortez (3) Fernand Cortez (3) Le rossignol (1) La caravane du Caire (3) Fernand Cortez (3) Fernand Cortez (3) La caravane du Caire (3) Fernand Cortez (3) Orphée et Eurydice (3) Panurge (3) Fernand Cortez (3) Œdipe à Colone (3) Les mystères d’Isis (4) Fernand Cortez (3) La caravane du Caire (3) Le rossignol (1) Fernand Cortez (3) Les bayadères (3) La vestale (3) Panurge (3) La vestale (3) Fernand Cortez (3) La caravane du Caire (3) Le rossignol (1) Fernand Cortez (3) Anacréon chez Polyeucte (3) Armide (5) La caravane du Caire (3) Fernand Cortez (3) Castor et Pollux (5) Orphée et Eurydice (3) Fernand Cortez (3) Panurge (3) La vestale (3) La caravane du Caire (3)

Flore et Zéphire (2) Paul et Virginie (3) Le carnaval de Venise (2) Flore et Zéphire (2) La dansomanie (2) Flore et Zéphire (2) Nina (2)

Achille à Scyros (3) Nina (2)

Flore et Zéphire (2) Nina (2) L’épreuve villageoise (2) Psyché (3) Flore et Zéphire (2) Psyché (3) Psyché (3) Psyché (3) Flore et Zéphire (2) Psyché (3) Nina (2) Le carnaval de Venise (2) Psyché (3) Psyché (3) Flore et Zéphire (2)

Psyché (3) Le carnaval de Venise (2) Flore et Zéphire (2) Psyché (3) (Continued)

Date

Opera

18 20 22 24 26

August August August August August

27 29 1 3 5 8 10 12 15 17 19 22 24 26 29 1 3 6 8 10 13 15 17 20 22 24 27 29 31 3 5 7 10 12 14 17 19 21 24 26 28 1 3

August August September September September September September September September September September September September September September October October October October October October October October October October October October October October November November November November November November November November November November November November December December

Fernand Cortez (3) Castor et Pollux (5) La vestale (3) Fernand Cortez (3) L’épreuve villageoise (1); Le rossignol (1) La caravane du Caire (3) Fernand Cortez (3) Les bayadères (3) Panurge (3) Fernand Cortez (3) Le rossignol (1) Les mystères d’Isis (4) Fernand Cortez (3) La vestale (3) La caravane du Caire (3) Panurge (3) Les bayadères (3) Les rossignol (1) Didon (3) Ossian, ou Les bardes (5) La vestale (3) Ossian, ou Les bardes (5) Les bayadères (3) La caravane du Caire (3) Les rossignol (1) Panurge (3) Les mystères d’Isis (4) Les bayadères (3) Œdipe à Colone (3) Les Danaïdes (5) Les Danaïdes (5) Les Danaïdes (5) Le rossignol (1) Les Danaïdes (5) La caravane de Caire (3) Les Danaïdes (5) Armide (5) Les Danaïdes (5) La vestale (3) Les Danaïdes (5) Les Danaïdes (5) Armide (5) Les Danaïdes (5) Les Danaïdes (5) La caravane du Caire (3) Les Danaïdes (5) Les Danaïdes (5) Les bayadères (3)

Ballet

Flore et Zéphire (2) Psyché (3) Le carnaval de Venise (2) Achille à Scyros (3) Flore et Zéphire (2) Le carnaval de Venise (2) Psyché (3) Flore et Zéphire (2) Le carnaval de Venise (2) Les fiancés de Caserte (1) L’épreuve villageoise (2) Les fiancés de Caserte (1) Nina (2) Flore et Zéphire (2) Nina (2) Psyché (3) Flore et Zéphire (2) L’enfant prodigue (3) Nina (2) Psyché (3) Le carnaval de Venise (2) Flore et Zéphire (2)

L’enfant prodigue (3) Nina (2)

Flore et Zéphire (2)

Flore et Zéphire (2)

Psyché (3)

Grand opéra – petit opéra  61 Date

Opera

5 7 8 10 12 15 17 19 22 24 26 28

Les Danaïdes (5) Le rossignol (1) Panurge (3) Les Danaïdes (5) Œdipe à Colone (3) Les Danaïdes (5) Les bayadères (3) Fernand Cortez (3) Fernand Cortez (3) Iphigénie en Tauride (4) Fernand Cortez (3) Les Danaïdes (5); Le devin du village [1] Aristippe (2); Le rossignol (1) La caravane du Caire (3)

December December December December December December December December December December December December

29 December 31 December

Ballet Nina (2) Psyché (3) Paul et Virginie (3) L’épreuve villageoise (3)

Flore et Zéphire (2)

L’épreuve villageoise (3) Psyché (3)

Iphigénie en Aulide. By contrast, Spontini’s Fernand Cortez was given a new production and much new music in late May; its premiere and first sixteen performances were not accompanied by a separate ballet, but by September it appeared both alongside Le carnaval de Venise and alone. Gluck’s Armide, Lesueur’s Ossian and Salieri’s Les Danaïdes, as outlined above, were played on their own. The combination of new productions could mean, for example, that in November 1817, three quarters of the evenings were dedicated exclusively to opéra, and that only a quarter included ballet-pantomime. The balance between the two types of evening could be controlled by the artistic instinct of the manager choosing to insert a new operatic production and giving it repertorial emphasis, or maintaining the traditional balance between ­ballet-­pantomime and opera. The limits on duration practised by the Opéra cannot be illustrated by means of hours and minutes and are better expressed by the number of acts, in the same way in which operatic bureaucracy, the press and probably the audience calculated in the first half of the nineteenth century. More than half the evenings analysed in Appendix 2.1 featured five acts, with a tiny proportion (less than 10%) offering four acts, and a little more, six; about a fifth of the performances were of three acts only, almost all the result of the first run of the new production of Fernand Cortez. The preponderant five-act evening was made up either of a single five-act work, noted earlier, or more often as the combination of a three-act opera and a two-act ballet. The Opéra’s conventional preference for an evening consisting of a total of five acts, once five-act composition for new grand opéra became the norm, was of immense significance, as will be seen later in the chapter. A decade later, Rossini’s two reworkings of Italian serious opera at the Opéra illustrate both tradition in action and the beginnings of change. Maometto II in its guise as Le siège de Corinthe behaved much as any other

62  Grand opéra – petit opéra three-act opéra had done in the previous twenty years; it was premiered alone on 9 October 1826, but at its seventeenth performance, on 24 J­ anuary 1827, it was paired with Le carnaval de Venise and remained part of a composite entertainment until it fell from the repertory in 1844. Rossini’s Mosè in Egitto, produced at the Opéra on the 26 March 1827 in four acts as Moïse, took programming in the direction at which Lesueur’s Ossian had hinted, since its length now made it impossible to pair it with a ballet in a single evening.16 Although Moïse was often broken up into individual acts (see later in the chapter), it retained its ability to occupy an entire evening as late as the new productions it received in 1852 and 1863 and behaved in many ways akin to the grands opéras that immediately followed it.

Eugène scribe and the challenge of grand opéra When Scribe began reworking Germain Delavigne’s three-act libretto for Auber’s La muette de Portici into five acts (1828), he set Parisian music drama on a course that was as novel as it was epoch-making.17 Grand opéra after 1828 took many features of earlier opera (an attention to dance, costume and scenery), embedded them in a libretto taken from modern rather than ancient history and married them to compositional and vocal practices derived from what was understood of Rossini and his Italian colleagues from productions at the Théâtre-Italien and the Odéon.18 In so doing, the new genre broke with the past with consequences that were completely unforeseen. Whereas the manager of the Opéra had previously been able to choose whether or not to pair opera with ballet-pantomime – according to 16 The two Rossini arrangements for the Opéra have received significant scholarly attention, but such interest has focussed more on the works themselves and their significance for Rossini’s output rather than their performative context. See Philip Gossett, ‘Gioachino Rossini’s Moïse’, The Rosaleen Moldenhauer Memorial: Music History from Primary Sources--A Guide to the Moldenhauer Archives (Washington, DC: Library of Congress, 2000), 369–374; Giuseppe Ierolli, ‘Mosé e Maometto: Da Napoli a Parigi’ (Tesi di lauria, Università degli Studi di Bologna, 1989–1990); Marcello Conati, ‘Between Past and Future: the Dramatic World of Rossini in Mosè in Egitto and Moise et Pharaon’, 19th-­C entury Music 4 (1980), 32–47. 17 The presence of a version of La muette de Portici in three acts was first signalled by Anselm Gerhard (acknowledged in Jean Mongrédien, ‘Variations sur une thème – Masaniello: du héros de l’histoire à celui de La Muette de Portici’, Jahrbuch für Opernforschung 1 (1985), 101), but the identification of the hand – and by implication the authorship of Germain Delavigne for the three-act version – was made by Herbert Schneider and Nicole Wild, La Muette de Portici: kritische Ausgabe des Librettos und Dokumentation der ersten Inszenierung, Erlanger romantistische Dokumente und Arbeiten 11 (Tübingen: Stauffenberg, 1993), 3 and 9–51. 18 For Rossini at the Théâtre-Italien, see Janet Johnson, ‘The Théâtre Italien and Opera and Theatrical Life in Restoration Paris’, 3 vols [paginated consecutively] (PhD diss., University of Chicago, 1988), and at the Odéon, Mark Everist, Music Drama at the Paris Odéon, 1824–1828 (Berkeley, Los Angeles and London: University of California Press, 2002), 227–249.

Grand opéra – petit opéra  63 the conventions outlined earlier – an evening that included a five-act grand opéra did not now allow him that luxury, since five acts – occasionally six – were traditionally the optimal upper limit for performances at the Opéra. For reasons of length, inclusion of a formally composed and choreographed divertissement within the work and perhaps because of the related complexities of staging, five-act grands opéras held the stage alone and were never paired with ballet. The overwhelming success of La muette de Portici, Rossini’s Guillaume Tell (four acts, 1829) and Meyerbeer’s Robert le diable (1831) meant that an evening dedicated exclusively to a single work was henceforth a common feature on the landscape of the Opéra. This could mean, immediately after an important premiere, that ballet-pantomime could almost disappear from view: in 1843, for example, largely as the result of the premiere of Halévy’s Charles VI, audiences at the Opéra saw almost no ballet-pantomime whatsoever, and it took the creation of La péri, the ballet fantastique by Frédéric Burgmuller, Théophile Gautier and Jean Coralli, to bring the institution back into line with its governing statutes.19 Such moments as this did not mean that audiences at the Opéra were denied dance. Earlier opera, by convention, included both divertissements and danced choruses, and this was carried over into grand opéra of the late 1820s and beyond.20 Indeed, changes in the structure of grand opéra in the 1830s and 1840s may well be a consequence of the fact that they could not share an evening with a ballet. Dance, which had been scattered quite widely within individual works from La muette de Portici until the late 1830s, thereafter became more formalised. With the 1834 production of Mozart’s Don ­Giovanni as Don Juan and the 1835 premiere of Halévy’s La Juive, librettists and choreographers began to consolidate dance into single more elaborate divertissements, with the result that in grand opéra of the 1840s, the flexible and highly varied relationship between dance and other music-­dramatic discourses of the period around 1830 was replaced by a conventional practice which focused on a single place for dance.21 Coterminous with this ­solidification of dance into a single divertissement was the appearance of 19 The only evenings that were not the subject of a single work were 1 March (Le philtre and La jolie fille de Gand), 6 March (two acts of Fernand Cortez and La tarentule), 10 March (Le philtre and La jolie fille de Gand), 2 April (Le guérillero and La jolie fille de Gand), 12 April (Fernand Cortez and Giselle) and 24 April (Le freyschütz and Giselle). All operas mentioned in this note are discussed below. For La jolie fille de Gand, see Lajarte, Bibliothèque musicale, 2:171, and for Giselle, Smith, Ballet and Opera, 167–200. 20 This paragraph summarises a complex history of the changing function of divertissement in grand opéra that has yet to be written. The summary table given in Marian Smith, ‘Dance and Dancers’, The Cambridge Companion to Grand Opera, 100 (which gives twelve works out of a total in excess of 30), may be supplemented by those in Guest, Romantic Ballet, 444–445, and idem, Ballet of the Second Empire, 128. 21 The move towards single, complex divertissements as the century takes no account of the activities of supernumeraries who might engage in danced choruses or elsewhere, with the most marginal trace in the written or printed records.

64  Grand opéra – petit opéra the cortège as a generic feature of grand opéra. Beginning again in the mid1830s, with Don Juan and La Juive, almost every new grand opéra now included a cortège with the music to accompany it.22 Ballet-pantomime, apart from a few hybrid attempts in the very early 1830s, never underwent the same seismic shift as did grand opéra in the late 1820s and never took on the same dimensions.23 Its length, therefore, prevented its survival on its own without the support of opera to accompany it. However, the growth of grand opéra meant that newly composed works could no longer be used in conjunction with a ballet in the same way that they had before 1828. The change in grand opéra and the impact that it had on its relationship with ballet-pantomime resulted in one of the longest running problems faced by the Opéra, one that elicited a number of responses. The difficulty was caused by a collision of two interests: the state’s expectation of what the Opéra should deliver and the aesthetic interests of librettists and composers who saw the future in terms of grand opéra in five acts. There is no evidence, despite the claims made post hoc by Louis-Desirée Véron (manager from 1831 to 1835), that the management of the Opéra was involved in the aesthetic shifts taking place in music drama at the institution.24 The musical 22 These comments are based on the descriptions of cortèges in contemporary libretti. In general, however, cortèges have received little attention in modern scholarship. For two important exceptions, see Mary Ann Smart, ‘Mourning the Duc d’Orléans: Donizetti’s Dom Sébastien and the Political Meanings of Grand Opéra’, Reading Critics Reading: Opera and Ballet Criticism in France from the Revolution to 1848, eds. Roger Parker and Mary Ann Smart (Oxford: Oxford University Press, 2001), 206–212, and Cormack Newark, ‘Ceremony, Celebration and Spectacle in La Juive’, ibidem, 184–187. However, both authors depend for their descriptions of these cortèges not on contemporary libretti but on Palianti’s livrets de mise-en-scène (conveniently published in H. Robert Cohen (ed.), Douze livrets de mise en scène lyrique datant des créations parisiennes, Musical Life in 19th-Century France 3 (Stuyvesant, NY: Pendragon, 1991) and idem (ed.), The Original Staging Manuals for Ten Parisian Operatic Premieres, Musical Life in 19th-Century France 6 (Stuyvesant, NY: ­Pendragon, 1998)). See also Karin Pendle and Stephen Wilkins, ‘Paradise Found: The Salle le Peletier and French Grand Opera’, Opera in Context: Essays on Historical Staging from the Late Renaissance to the Time of Puccini, ed. Mark E. Radice (Portland, OR: Amadeus Press, 1998), 193–195. Arnold Jacobshagen has, however, recently shown how Palianti’s livrets de mise en scène for Dom Sébastien dated from the time of its premiere, although the same is not true for the text of La Juive; he has argued that the most likely date for a production that underpinned, for example, Palianti’s livret for La juive is 1866. See his ‘Analysing mise-en-scène: Halévy’s La Juive at the Salle Le Pelletier’, Music Theater and Cultural Transfer: Paris, 1830–1914, eds. Annegret Fauser and Mark Everist (Chicago, IL: Chicago University Press, 2009) 176–194. 23 Two important exceptions, Le dieu et la bayadère (libretto by Eugène Scribe, music by Auber, choreography by Filippo Taglioni) and La tentation (libretto by Edmond Cavé and Duponchel, music by Halévy and Casimir Gide, choreography by Jean Coralli) are part of the history discussed in this chapter and in Smith, Ballet and Opera, 138–166. 24 No cahier des charges issued to a manager of the Opéra during the licensing period involved itself with matters as detailed as the nature of the libretto, placement or dimensions of the divertissement, or the number of acts of a grand opéra. Rather, pompe et luxe were bywords for the institution’s productions. However, ‘pomp and luxury are not the only characteristics

Grand opéra – petit opéra  65 and dramatic moves were being made by c­ omposers – Auber in particular – and librettists, chief among them Scribe.25 Scribe’s role in the aesthetic experiments around 1830 cannot be exaggerated: not only was he responsible for moving La muette de Portici from three to five acts and for changing Robert le diable from an opéra comique into a grand opéra, he was the librettist for ­Auber’s Le dieu et la bayadère (1830), Le serment (1832) and Gustave III (1833),26 all of which called the relationship between music and dance into question in different ways. It is hardly surprising that Scribe should have been responsible for attempts to rebalance the equilibrium between opera and ballet with a type – petit opéra – where divertissements were absent and less formal dance liminal. His work on the libretti for both works that ushered in the genre, Rossini’s Le comte Ory and Auber’s Le philtre is entirely congruent with, and complementary to, his creation of the grand opéra libretto. Grand opéra based on Scribe’s libretti for the Opéra therefore represented a repertorial development for Émile-Timothée Lubbert (manager from 1827 to 1831) and his successors that could not remain without response if ­ballet-pantomime was to remain on an equal footing with grand opéra. There were four consequences: (1) to pair ballets composed after 1828 with operas composed before that date that were of smaller dimensions than contemporary of the genres attributed to the Académie Royale [de Musique]: the operas that are produced there must also obtain the approval of masters of art and men of taste, in terms of performance and of the merit of the musical compositions’ (‘La pompe et le luxe ne sont pas les seuls caractères du genre attribué à l’Académie Royale: il faut aussi que les opéras qui y sont représentés obtiennent l’approbation des maîtres de l’art et des hommes de goût, sous le rapport de l’exécution, et du mérite des compositions musicales’; ‘Appendice au cahier des charges de la direction de l’Opéra en régie intéressée, contenant les éclaircissements et additions jugées nécessaires à l’exécution du traité, et consenties par Mr Véron directeur’. 30 May 1831. F-Pan F21 4655(2)). Despite this fleeting recognition of the aesthetic consequences of works mounted at the Opéra, there is little sign of such a body as the Commission spéciale pour les théâtres being alive to the problems faced by the growth of grand opéra (the minutes of the Commission des théâtres are in F-Pan F21 4633(3–6) and F21 4634(1)). 25 The beginnings of work on the experiment may be seen in Rossini’s two adaptations of Italian originals in 1826 and 1827, in which Scribe played no part, and it is clear that he depended on this tradition in the period from the mid-1820s to the mid-1830s. 2 6 It is therefore difficult to assign an exact starting point to the history of grand opéra, pace Ludwig Finscher (‘Aubers La muette de Portici und die Anfänge der Grand-opéra’, Festschrift Heinz Becker zum 60. Geburtstag am 26. Juni 1982, eds. Jürgen Schläder and ­Reinhold Quandt ([Laaber]: Laaber, 1982), 87–105). The even more complex history of Robert le diable – as not only in three acts but as an opéra comique with spoken dialogue and written specifically for the Opéra-Comique’s vocal troupe – places it clearly in the same experimental tradition. See Mark Everist, ‘The Name of the Rose: Meyerbeer’s opéra comique, Robert le Diable’, Revue de Musicologie 80 (1994), 211–250. The particular ­c oncentration on dance, especially in the fifth act of Gustave (see below), by no means represents the end of the period of experimentation. Given that the grand opéra for the following year (1834) was a translation and adaptation of Don Giovanni, which posed its own set of generic problems, it could be argued that the sort of experimentation which allowed the emergence of grand opéra and engendered petit opéra did not really end until the definitive statement of La Juive in 1835.

66  Grand opéra – petit opéra grand opéra; (2) to reduce large-scale grands opéras, especially those that had not fared so well in their original form, to dimensions appropriate to an evening of mixed opera and ballet; (3) to take single acts or pairs of acts from their four- or five-act context and mix them with a new ballet; and (4) to encourage the composition of new two-act operas by the same composers who were writing five-act grands opéras; these began to appear in managers’ cahiers des charges and other administrative documents under the name petit opéra.27 For Lubbert, who was to remain at the Opéra only until the end of ­February 1831, using operas from the past was something to which he had been used since 1827 and that François-Antoine Habeneck and Raphaël de Frédot Duplantys, his predecessors (1821–1824 and 1824–1827, respectively), had been doing before.28 And given the regard in which opera from the ancien régime was held during the Restoration – Grétry in particular – it was a reasonable supposition that old works would continue to prove acceptable until such time as ‘old’ could be equated with ‘outdated’. Two of the works that were most popular during Lubbert’s administration and subsequently were operas that dated back to the Empire: Rodolphe Kreutzer’s Aristippe and Catel’s Les bayadères.29 The most important work from the past was Louis-Sébastien Lebrun’s Le rossignol.30 Aristippe remained on the stage until 1830, while Les bayadères was the subject of an important new production in 1827. Le rossignol, however, stayed in the repertory until 1852, with new productions in 1831, 1845 and 1850.31 Producing old opera on a smaller scale to accompany ballet became more of a problem as time went on. More and more five-act grands opéras were being composed and being successfully received, and audiences were welcoming new works in a way that they had not since Spontini; they were therefore less likely to tolerate old works that might fare rather badly alongside the new. Although attitudes to older works could still be enthusiastic, exclusive diets of works from the Empire and Restoration became less palatable during the July Monarchy. Véron felt this acutely and found two further ways of responding to the emerging question of how to accompany a ballet. Véron had taken over the Opéra on 1 March 1831 but formally began his management on 1 June that year. His opening night had something of the manifesto about it: it consisted of Guillaume Tell, reduced to three acts 27 The first draft of Véron’s cahier des charges puts grand opéra and petit opéra on the same footing in its list of permitted genres: ‘Le grand ou le petit opéra avec ou sans ballets’ (‘Cahier des charges de la direction de l’opéra en régie intéressée, arrêté en commission et approuvé par Monsieur le Ministre de l’Intérieur’. 28 February 1831. F-Pan F21 4655(2)). In his own words, Véron described Auber’s Le philtre as a petit opéra (Mémoires d’un bourgeois de Paris, 5 vols (Paris: Gouet, 1853–1856), 3:156). 28 For a summary of the careers of all the managers of the Opéra discussed in this study, see Wild, Dictionnaire des théâtres parisiens au xixe siècle, 305–307. 29 Aristippe was premiered on 24 May 1808, Le bayadère on 8 August 1810, both at the Opéra. 30 Le rossignol received its premiere on 23 April 1816. 31 See Lajarte, Bibliothèque musicale, 2:88.

Grand opéra – petit opéra  67 from four, coupled with Scribe’s and Hérold’s ballet-pantomime entitled La somnambule, ou L’arrivée d’un nouveau seigneur. Premiered in 1827, La somnambule was rapidly approaching classic status. In the twofold context of a solution to an emerging problem and of a new manager bursting onto the Parisian stage, reducing Guillaume Tell to three acts is understandable, a fact that Rossini himself recognised.32 The modifications consisted of reworking the act iv finale and moving Arnold’s act iv aria ‘Asile héréditaire’ to the beginning of act iii. Almost all the ballet music, with the exception of the act iii pas de trois, was removed.33 This general practice furthermore continued throughout the century: Niedermeyer’s Stradella of 1837 was reduced from five acts to two a year after it premiered and was paired with La fille mal gardée and La fille du Danube.34 Two operas in three acts – by now a rare practice, Ruolz’s La vendetta (1839) and Gounod’s Sapho (1851) – were similarly reduced to two acts in order to achieve the same purpose (in 1840 and 1858, respectively).35 In both cases, the works were first performed alone but soon after paired in their original form with a ballet (making for a long evening) and subsequently formally reconfigured in two acts. Véron’s second strategy for dealing with the question of how to construct an evening’s entertainment that included a ballet-pantomime was to extract individual acts from contemporary five-act grand opéra and set them alongside ballets. The process continued throughout the nineteenth and into the twentieth century. The majority of works selected for this process of what was called – even in the managers’ cahiers des charges – ‘morcellement’ dated from the late 1820s and 1830s, with the result that works that are today considered staples of the Opéra’s repertory were frequently broken up in this 32 Rossini’s letter to Bis, in which he permits almost any modification to the work, makes striking reading: ‘If there is any amputation to make to it, I give you carte blanche as well as to Lubbert’ (‘S’il y a quelque amputation à y faire, je vous donne carte blanche ainsi qu’à Lubbert’; M. Elizabeth C. Bartlet, Guillaume Tell, Opéra en quatre actes di Victor Etienne de Jouy e Hippolyte Louis Florent Bis; musica di Gioachino Rossini, 4 vols, Edizione critica delle opere di Gioachino Rossini, Sezione prima 39 (Pesaro: Fondazione Rossini, 1992), 1:xxxviii). 33 Ibidem, 1: xxliii–xl and especially 4:1972–2025 where the revisions to the act iv finale are edited. 34 The choreography of La fille mal gardée was originally composed in 1786 by Jean Bercher (pseud. Dauberval) and reworked for a new production on 17 November 1828 by Jean-Louis Aumer with music by Ferdinand Hérold. La fille du Danube (music by Adam and choreography by Taglioni) was premiered on 21 September 1836. Stradella’s compositional history (libretto by Émile Deschamps and Émilien Pacini) is complex, however. It started out in three acts with a contract to produce the work before September 1836, as witnessed by a complaint made by the composer and librettists after the five-act version’s withdrawal (F-Pan AJ13 183 (9 April 1838)). See Sarah Hibberd, ‘Murder in the Cathedral? Stradella, Musical Power, and Performing the Past in 1830s Paris’, Music & Letters 87 (2006), 577, note 81. 35 Henri de Ruolz’s La vendetta set a libretto by Léon Pillet and Adolphe Vannois and was premiered on 11 September 1839; for its reduction to two acts, see Lajarte, Bibliothèque musicale, 2:161. Gounod’s Sapho (libretto by Émile Augier and premiered on 16 April 1851) is discussed in Stephen Huebner, The Operas of Charles Gounod (Oxford: Oxford University Press, 1990), 26–34.

68  Grand opéra – petit opéra fashion. Although this was a practice to which opera was routinely subject, ballets-pantomimes were almost never so treated except in such exceptional cases as benefit performances. Perhaps the most famous and – because it concerns Berlioz – the example of morcellement that has attracted the most opprobrium is Benvenuto Cellini. Of its seven performances in late 1838 and early 1839, four occupied an entire evening and three consisted of act i alone, together with the ­ballets-pantomimes La gipsy and Le diable boîteux.36 In withdrawing his opera from the Opéra, Berlioz was flying in the face of what by 1839 was a firmly established practice,37 but had already acquiesced to the morcellement of Benvenuto Cellini since he wrote linking material as act i was reworked.38 Rossini’s Moïse, frequently performed in its entirety alone, was often represented just by its third act alongside a ballet; the second act of Spontini’s La vestale was also extracted to the same end. Other works were only really known in a fragmentary form. Auber’s Le serment was used in a variety of guises: its first act alone, second act alone, and the first two acts together, accompanying a ballet in all cases. Even the first two acts of Don Giovanni were taken from their grand opéra context to serve as the operatic component in an evening shared with ballet.39 The most remarkable instance of morcellement is surely that of Auber’s Gustave III. Although the second, fourth and fifth acts were all extracted from the work to accompany a ballet, the treatment of the fifth act is particularly revealing, since its history shows how the work emerges from the experimental mix of opera and ballet-pantomime c1830 that resulted in such works as Le dieu et la bayadère and La tentation. The fifth act is a ball scene and is therefore characterised by a cognate discourse to ballet-pantomime and, as has been much discussed, to masked balls held at the Opéra. Not only was Auber’s fifth act extracted from the work, but the composer completely reworked the act 36 The affair is fully documented in Hugh MacDonald (ed.), Benvenuto Cellini, 4 vols, H ­ ector Berlioz: New Edition of the Compete Works 1 (Kassel, etc.: Bärenreiter, 1994–2006), 1:xv–xvii. 37 There remains the residual question of whether Benvenuto Cellini was originally conceived as a petit opéra. The only letter in which Berlioz discussed the work in terms of ballet is to Franz Liszt, 22 January 1839 (Pierre Citron (ed.), Hector Berlioz: Correspondance Générale II: 1832–1842, Hector Berlioz Œuvres Littéraires (Paris: Flammarion [Nouvelle bibliothèque Romantique], 1975), 521), but it relates to the process of morcellement itself in early 1839, and not to any earlier state of the work. Auguste Barbier’s and Léon de Wailly’s libretto, however, falls well outside the dimensions of those that characterise a petit opéra, with two acts of 13 and 20 scenes, respectively. 38 See Macdonald, Benvenuto Cellini, 1:xv. 39 Such dismemberment was not restricted to the repertorial exigencies of the Paris Opéra. For an account of several single acts of Italian opera being reworked (although with significantly different aims) into a single evening’s entertainment, see Hilary Poriss, ‘To the Ear of the Amateur: Performing Operas Piecemeal’, Fashions and Legacies in Nineteenth-­ Century Italian Opera, eds. Roberta Montemarra Marvin and Hilary Poriss (Cambridge: Cambridge University Press, 2010), 111–131.

Grand opéra – petit opéra  69 expressly as a ballet; the revision was premiered on 1 March 1835, but as the ballet component in the entertainment. It therefore stands outside the pattern of using isolated acts from erstwhile grand opéra to eke out the programme.40 When Duponchel took over the direction of the Opéra from Véron in 1835, his new cahier des charges specifically acknowledged the process of morcellement, attempting to ban it while simultaneously giving comprehensive exceptional clauses that allowed the manager to continue with business as usual: The manager shall not break new works into pieces, nor old ones that have not been performed in parts, without our special authorisation, except the rights of authors and composers. This authorisation may be given on a permanent basis, or only for a certain number of times, according to the value of the works or their influence on the public.41 With no evidence of any example of morcellement being refused by the Commission des théâtres, this clause in Duponchel’s cahier des charges did little more than recognise and largely sanction the practice, despite the incomprehensible end to its first sentence. Although the 1830s saw the highest density of morcellement, and the practice waned after 1840, it did not die out. As late as 1845, the first act of La muette de Portici was performed with a ballet, and from 1847 its first four acts were played in conjunction with two-act ballets such as Paquita and La sylphide.42 And Guillaume Tell was not only subject to reduction to three acts for much of its nineteenth-century Parisian career, but the second act (unchanged in the three-act version) was also morcelé as a key part of the Opéra’s repertory.

40 There is a very full discussion of the significance of the fifth act of Gustave III in Maribeth Clark, ‘Understanding French Grand Opera through Dance’ (PhD diss., University of Pennsylvania, 1998), 118–179. For other studies of Gustave III that touch on the issue of the fifth act, see eadem, ‘The Role of Gustave, ou le bal masqué in Restraining the Bourgeois Body during the July Monarchy’, The Musical Quarterly 88 (2005), 204–231; Sarah ­Hibberd, ‘Auber’s Gustave III: History as Opera Music Theater and Cultural Transfer: Paris, 1830–1914, eds. Annegret Fauser and Mark Everist (Chicago, IL: Chicago University Press, 2009), 157–175; Anna McCready, ‘Assassination and the Historical Ball: Auber’s Gustave III and the Duc de Berry’, in ‘Gilding the Lily: Music and Monarchy in Paris (1814–1833)’ (PhD diss., King’s College, University of London, 2003), 157–207 (a shorter version of the chapter was delivered as a paper at the Symposium on Music in France (1830–1940), University of Melbourne, 14–17 July 2004). 41 ‘L’entrepreneur ne pourra morceler les ouvrages nouveaux, ni les anciens qui n’ont pas été joués par parties, sans notre autorisation spéciale, sauf le droit des auteurs et compositeurs. Cette autorisation pourra être accordée à toujours, ou seulement pour un certain nombre de fois, selon la valeur des ouvrages ou leur influence sur le public’ (Académie Royale de Musique – Cahier des Charges. M. Duponchel. 19 August 1835. F-Pan F21 4655(3), article 24). 42 Paquita (music by Edmé Deldevez, scenario by Paul Foucher and choreography by ­Mazilier) was premiered on 1 April 1846; La sylphide (music by Schneitzhœffer, scenario by Adolphe Nourrit and choreography by Taglioni) on 12 March 1832.

70  Grand opéra – petit opéra Maintaining older short operas in the repertory, shrinking four- and five-act works to lesser dimensions and the practice of morcellement were strategies that served Lubbert, Véron and their successors well. The fourth approach, the composition of new works, usually in two acts, was born out of a series of accidents. Two-act petit opéra seems to have found its origins in Rossini’s Le comte Ory of 1828 and Auber’s Le philtre of 1831. Emerging at the same time as the experimental works that led to the establishment of grand opéra, Le comte Ory and Le philtre were employed to accompany ballet from the date of their composition until the 1860s. More importantly, they served as a template for the composition of both libretti and music for petits opéras for nearly half a century and therefore formed the origins of an entirely new generic thread in the repertorial skein of the Opéra. From the late 1820s onwards, petit opéra was a form sui generis with its own conventions and modes of understanding. ‘Le goût français allié à l’allure vive et entraînante de l’opéra bouffe italien’: Le comte Ory, Le philtre and petit opéra Le comte Ory, premiered on 20 August 1828, was Rossini’s third opera in French for the Opéra.43 Like Moïse and Le siège de Corinthe, it was – for at least part of its music – based on an earlier Italian opera: Il viaggio a Reims, ossia L’albergo del giglio, an occasional work for the celebrations following the coronation of Charles X in 1825. Broadly speaking, all of the first act of Le comte Ory is borrowed from Il viaggio a Reims with the exception of the duet for the eponym and Isolier ‘Une dame de haut parage’ (no. 3), and all of the second act is newly composed except for the duet for Ory and La Comtesse ‘Ah quel respect ma dame’ (no. 7) and Raimbaud’s aria ‘Dans ce lieu solitaire’ (no. 9).44 For hostile critics, Rossini’s failure to produce a new opera for the French stage was an opportunity to catalogue the borrowings 43 Le comte Ory, as has been recently acknowledged (Benjamin Walton, Rossini in Restoration Paris: The Sound of Modern Life, Cambridge Studies in Opera [Cambridge: Cambridge University Press, 2007], 240), is ‘a difficult opera to account for in a convincing way’, and the current study resolves that difficulty without recourse to claims that it was a response to the discovery by a very few of Beethoven at the Conservatoire concerts (ibidem, 245) or as ‘reaffirming his connections with Mozart over the new world of Beethoven’ (ibidem, 246). But Le comte Ory also had a second, parallel life in the late 1830s as the justification for the emergence of a formalised type of opera called opéra de genre. See La France musicale, 17 January 1839, and Mark Everist, ‘Donizetti and Wagner: opéra de genre at the Théâtre de la Renaissance’, Giacomo Meyerbeer and 19th-Century Parisian Music Drama, Variorum Collected Studies Series CS805 (Aldershot: Ashgate, 2005), 312–313. 4 4 For the detail of the relationship between the music of Le comte Ory and Il viaggio a Reims, see Janet Johnson, ‘A Lost Rossini Opera Recovered: Il Viaggio a Reims’, Bollettino del Centro Rossiniano di studi n.v. Nos. 1–3 [bound as one] (1983), 5–57; eadem (ed.), Il viaggio a Reims, ossia L’albergo del giglio d’oro: dramma giocoso in un atto di Luigi Balochi, musica di Gioachino Rossini, 2 vols, Edizione critica delle opere di Gioachino Rossini 35 (Pesaro: Fondazione Rossini, 1999), 1:xliv–xlv.

Grand opéra – petit opéra  71 and to berate both composer and librettists for what some journalists chose to view as self-plagiarism.45 Even more important, since it concerns the tone and structure of the entire opera, the libretto was reworked by Scribe and his collaborator, CharlesGaspard Delestre-Poirson, from their vaudeville, also entitled Le comte Ory and premiered at the Théâtre de Vaudeville in 1816.46 While much of the tone of the vaudeville and its dependent opera is shared with such works as Boieldieu’s Jean de Paris (an opéra comique from 1812) and Scribe and Delavigne’s vaudeville entitled Thibault, comte de Champagne (1813), the structure of the libretto to which Rossini adapted and composed his music was a product of the late 1820s.47 Although the 1816 vaudeville provided most of the action for act ii of the 1828 Le comte Ory, the libretto to the petit opéra’s first act was new and based on the opening stanzas of the 1816 vaudeville, based in turn on music by Luc Guénée.48 In its Italianate musical origins and its textual background in vaudeville, Le comte Ory marked a major change in the repertory of works at the Opéra. In a long and carefully considered review of its premiere in the Gazette de France, Auguste Delaforest pointed to its radical nature. Having drawn attention to the routine nature of programming at the Opéra where ‘there are plenty of people who adore Le devin du village, Panurge and Les prétendus’,49 he turned to Le comte Ory:

45 See, for example, Le corsaire, 21 and 24 August 1828. But not all critics were hostile. François-Joseph Fétis admitted that he had been opposed to the idea of ‘an old ­vaudeville on which one was to parody the music of the occasional opera Il viaggio a Reims’ (‘un ancien vaudeville de M. Scribe sur lequel on parodierait la musique de l’opéra de circonstance Il viaggio à Reims’, Revue musicale 4 (1828–1829), 82), and slightly disingenuously in the face of the work’s obvious success claimed that he had been misled (although his original understanding is entirely fair); his review was entirely positive. 46 LE COMTE / ORY, ANCECDOTE DU XIe SIÈCLE, / VAUDEVILLE EN UN ACTE, Par MM. EUGÈNE SCRIBE ET DELESTRE-POIRSON; / Représenté pour la première fois sur le Théâtre / du Vaudeville, le 16 décembre 1816 / .. / DE L’IMPRIMERIE DE FEUGUERAY. / A PARIS / Chez BARBA, Libraire; au Palais-Royal, derrière le Théâtre français …/ = / 1817. 47 Jacques Joly, ‘Du gothique ‘troubadour’ a l’hommage à Mozart’, Rossini: Le v­ oyage à Reims – Le comte Ory, L’Avant-Scene Opéra 140 (Paris: L’Avant-Scène Opéra, 1991), 90. 48 See Herbert Schneider, ‘Le Comte Ory: Du vaudeville au livret d’opéra’, Les écrivains français et l’opéra, eds. Jean-Paul Capdevielle and Peter Eckhard Knabe, Kölner Schriften zur romanischen Kultur 7 (Cologne: DME, 1986), 174. 49 ‘Il y a beaucoup de gens que le Devin du village, Panurge et les Prétendus ravissent toujours’ (Gazette de France, 21 August 1828; R Cours de littérature dramatique, 2 vols (Paris: Allardin, 1836), 2:505 [page numbers refer to 1836 reprint]). Le devin du village (music and libretto by Jean-Jacques Rousseau) was premiered at Fontainebleau on 18 October 1752; Panurge (music by Grétry, libretto by Etienne Morel de Chédeville) was premiered at the Opéra on 25 January 1785; Les prétendus (music by Jean-Baptiste Lemoyne, libretto by Marc-Antoine-Jacques Rochon de Chabannes) premiered ibidem, 2 June 1789.

72  Grand opéra – petit opéra Nothing in all that [eighteenth-century repertory] would scare away mediocrity and routine. But a work in two acts in the Italian style, with an unusual plot, original costumes, a dialogue that is less stupid that that of La caravane, and a denouement and twists in the plot that do not at all resemble what has already been seen, all that is frightening for routine, and there is at least reason to protest.50 Delaforest’s commentary on the work’s coupe italienne and tone is central to understanding the opera, its place in stage music at the end of the Restoration and petit opéra in general. When he spoke of dialogue, dénouement and détails that did not resemble anything that had preceded it, he was doing more than pointing out that Le comte Ory was a comic opera: he was alluding to a tradition of Italian opera buffa best known in Paris via the works produced at the Théâtre-Italien, with Il barbiere di Siviglia at their head.51 And his comments went straight to the heart of the differences between Le comte Ory and other works at the Opéra. Delaforest’s comments need to be read alongside an outline of the opera itself (Table 2.1). When he spoke about two acts coupés à l’Italienne, he was alluding to the overall construction in two acts with around ten scenes in each, the major finale placed at the end of the first act, an introduzione after the overture, and with the end of the work marked by a (usually) perfunctory chorus or rondò. Delaforest was also pointing to the limited number of characters and the carefully controlled way in which arias and duets were allotted to each. There are six characters only in Le comte Ory and the most elaborate single number (excluding introductions and finales) is the trio. But Delaforest’s comments also pointed to the internal structuring of individual numbers, arias and duets in particular, which – in the case of those borrowed from Il viaggio a Reims – reflected the convenienze that had characterised Rossini’s works before he left Italy and that continued to dominate stage music on the peninsula.52 50 ‘Rien dans tout cela n’effarouchait la médiocrité et la routine. Mais un ouvrage en deux actes coupés à l’italienne, sur une intrigue qui n’est pas commune, avec des costumes originaux; un dialogue qui n’est pas aussi bête que celui de la Caravane, et un dénouement et des détails qui ne ressemblent pas à tout ce qu’on a déjà vu; tout ceci est effrayant pour la routine, et il y a au moins lieu à protester’ (Cours de littérature dramatique, 2:505). La caravane du Caire (music by Grétry, libretto by Morel de Chédeville) was premiered on 30 October 1783 at Fontainebleau. 51 For an introduction to the repertory of Italian opera in Paris, and its comic component, see François-Henri-Joseph Blaze (Castil-Blaze), L’opéra Italien de 1548 à 1836, Théâtres lyriques de Paris [2] (Paris: Castil-Blaze, 1856), 281–435. 52 The conventions that govern Italian opera of the primo ottocento and that underpin what French musicians understood as Italian opera are outlined in Philip Gossett, ‘Gioachino Rossini and the Conventions of Composition’, Acta musicologica 42 (1970), 48–58; idem, ‘Verdi, Ghislanzoni, and Aida: The Uses of Convention’, Critical Inquiry 1 (1974–1975), 291–334; Harold S. Powers, ‘La solita forma’ and ‘The Uses of Convention’, Acta Musicologica 59 (1987), 65–90; Scott L Balthazar, ‘Evolving Conventions in Italian Serious Opera: Scene Structure in the Works of Rossini, Bellini, Donizetti and Verdi, 1810–1850’ (PhD diss,

Table 2.1  R  ossini, Le comte Ory: summary of sources and structure No

1

2

Title and cast ACTE I Introduction (Le Comte, Raimbaud, Ragonde, Alice, Chœur) Récit Air (Le Gouverneur)

Récit 3 4

Duo (Le Comte, Isolier) Récit [marche] Air (La Comtesse) Récit

5

Finale

6

ACTE II Introduction Récit

7

Duo (La Comtesse, Le Comte) Récit

8

Chœur Récit

9

Air (Raimbaud) Récit

10 11

Incipit

Scene

Sources and borrowings

‘Jouvencelles, venez vite’

i–iii

= Introduzione, IvaR no. 1 (except LcO, 1–3)

‘De grace encor un mot’ ‘Veiller sans cesse’

iv

‘Cet ermite, ma belle enfant’ ‘Une dame de haut parage’ ‘Isolier dans ces lieux’ ‘En proie à la tristesse’

v–vii

‘C’est bien, je suis content’ ‘Ciel! Ô terreur/ bonheur’

iv–v

vii viii viii viii–ix ix

‘Dans ce séjour’ i ‘Quand tomberont sur ii lui’ ‘Ah quel respect ma iii dame’ ‘Voici vos campagnons fidèles’ ‘Ah la bonne folie’ ‘Eh mais quelle triste observance’ ‘Dans ce lieu solitaire’

‘Du fruit de sa victoire’ Chœur ‘Buvons, buvons!’ Récit ‘Elle revient’ Trio (La Comtesse, ‘A la faveur de cette nuit’ Isolier, Le Comte)

iii

Tempo di mezzo and cabaletta in LcO (51–62) are taken from the corresponding sections of IvaR, no. 4. Newly composed = Aria contessa, IvaR no. 2. = Gran pezzo concertato, IvaR no. 5 Newly composed = duetto di Corinna e Cavalier Befiore, IvaR no. 5

iv iv–v

Newly composed

v

= Aria Don Profondo, IvaR no. 6

vi vi vi–ix x–xi

Newly composed Newly composed

(Continued)

74  Grand opéra – petit opéra No

12

Title and cast

Incipit

Récit

‘Vous qui faisiez la xi guerre’ ‘Ecoutez ces chants de xi victoire’

Finale

Scene

Sources and borrowings

Newly composed

IvaR: Janet Johnson, ed., Il viaggio a Reims, ossia L’albergo del giglio d’oro: dramma giocoso in un atto di Luigi Balochi, musica di Gioachino Rossini, 2 vols, Edizione critica delle opere di Gioachino Rossini 35 (Pesaro: Fondazione Rossini, 1999). LcO: LE COMTE ORY/Opéra en deux actes/Représenté pour la 1ère fois/sur le Théâtre de l’Academie Royale de Musique/le 20 Aout 1828/Paroles de MM. Scribe et Delestre-­ Poirson/Mis en musique et Dédié à/Monsieur Alexandre Aguado/Chevalier de l’Ordre de Charles III/PAR/G. ROSSINI/.../à Paris, chez E. Troupenas, Editeur du Répertoire des Operas Francais/Rue S t . Marc Feydeau No .23/à Londres, chez Latour/ (Philip ­E dward Gossett, ed., Le Comte Ory: Libretto by Eugène Scribe and Charles-Gaspard ­D elestre-Poirson, Music by Gioachino Rossini, Early Romantic Opera [16] (New York and London: Garland, 1978)).

Two duets in Le comte Ory clearly illustrate Rossini’s different working practices in the opera: the act ii duet for Ory and La Comtesse (‘Ah quel respect’), which is borrowed from Il viaggio a Reims, and the act i duet for Isolier and Ory (‘Une dame de haute parage’), which was newly composed in 1828. ‘Ah quel respect’ mirrors the pattern of primo tempo (separate statements of paired poetic stanzas), cantabile (atypically the same tempo as the primo tempo, ending with both soloists in mellifluous sixths), tempo di mezzo (change of tempo and move to kinetic action, in which Ory [in disguise] tells the Comtesse that he loves her and asks her how she would respond to a declaration of love) and cabaletta (paired stanzas, one for each soloist, with a third statement of the music to bring the soloists together) of the original Italian duet; little is changed apart from the language. By contrast, ‘Une dame de haute parage’ is newly conceived in a single tempo throughout with large-scale repetition creating recapitulatory structures completely foreign to an Italian duet. True, there are some local echoes of Italian practices in Rossini’s use of textures more familiar from the end of a traditional cantabile (except that the passage in sixths is repeated later University of Pennsylvania, 1985); idem, ‘Ritorni’s Ammaestramenti and the Conventions of Rossinian Melodramma’, Journal of Musicological Research 8 (1989), 281–311; idem, ‘The Primo Ottocento Duet and the Transformation of the Rossinian Code’, Journal of Musicology 7 (1989), 471–497; idem, ‘Analytic Contexts and Mediated Influences: The Rossinian convenienze and Verdi’s Middle and Late duets’, Journal of Musicological Research 10 (1990), 19–46. The attempt to discredit Powers’ use of Basevi’s 1859 text (Abramo Basevi, Studi sulle opere di Verdi (Florence: Tofanil, 1859)) as ‘an essential starting point for Verdian analysis and criticism’ leaves both Basevi’s (and Ritorni’s for that matter [Carlo Ritorni, Ammaestramenti alla composizione d’ogni poema e d’ogni opera appartenente all musica (Milan: Pirola, 1841)]) vocabulary intact both as a tool in the non-analytical discussion of primo ottocento opera (Roger Parker, Leonora’s Last Act: Essays in Verdian Discourse, Princeton Studies in Opera (Princeton, NJ: Princeton University Press, 1997), 56) and in the discursive language of composers of Italian opera and of those foreigners who sought to emulate them.

Grand opéra – petit opéra  75 in the duet) and the opening where the voices present paired stanzas – but then this is completely undercut by the recapitulation of the same material both one-third and two-thirds of the way through the duet. In this regard, Le Gouverneur’s aria in act i is intriguing, since Rossini rewrote the primo tempo of this aria to replace the original but left the music of the tempo di mezzo and cabaletta intact.53 While many French composers adopted the general principles of the convenienze that filtered through Il viaggio a Reims into Le comte Ory with enthusiasm, a third element of Delaforest’s coupe Italienne was adopted specifically by subsequent French composers of petits opéras; this is the practice of clearly articulating individual numbers with recitative in the Italian manner of the recitativo dopo l’aria or recitativo dopo il duetto and eschewing the permeable boundaries between numbers, as was common in grand opéra.54 This pattern of regularly placed recitative is as clear in Le comte Ory as it is in Il viaggio a Reims or any other Italian opera of the period, serious or comic.55 Although the Italian influence on petit opéra may be traced through the stylistic impulses behind Le comte Ory, such influences were as much at work in grand opéra from Meyerbeer onwards.56 Coupled to the fact that many of the artists at the Opéra around 1830 had careers either in Italian theatres or at the Parisian Théâtre-Italien, a shared Italian background to both grand 53 The two cavatine are around the same length (‘Invan strappar dal core’ from Il viaggio a Reims is 127 bars of Allegro maestoso 4/4 while ‘Veiller sans cesse’ from Le comte Ory is 73 bars of Andantino 6/8). 54 This is not to deny that recitative is found within individual numbers in Le comte Ory, especially the introduction and finales; it is, however, significantly rarer in petit opéra than in grand opéra. Nor is it to disagree with the idea that accompanied recitative was frequently introduced during numbers in Italian serious opera into the 1830s, but even there, numbers conclude emphatically with the following recitative clearly articulated. 55 The history of the move from recitativo dopo l’aria or recitativo dopo il duetto to scena ed aria or scena e duetto – whether the kinetic sections of the operatic discourse are separated from individual numbers or associated with their beginning – has yet to be written, but it may be located in the 1830s. Although scattered examples of the use of the scena are found as early as c1830, some composers and librettists were still using the dopo l’aria pattern of recitative as late as 1840. For French audiences and critics, even in the late 1830s, and certainly around the time of Le comte Ory and Le philtre, clearly articulated recitatives after the composition were what was understood as characteristic of Italian opera.   Whether the Italian genealogy of petit opéra is shared with opéra comique is an open question: certainly both Italian opera of the period and opéra comique conclude individual numbers decisively, and the kinetic sections (recitative or spoken dialogue, respectively) are discrete; there is also sometimes a correlation between the literary register of the libretti of petit opéra and opéra comique. But the question of whether this is sufficient – given the absence of other links to opéra comique in this genealogy – to argue for a significant relationship between opéra comique and petit opéra remains without resolution, and such claims can only be based on the mistaken assumption that the libretti of opéra comique are exclusively ‘light and sentimental’ 56 See the seminal study of Italianate forms in Meyerbeer in Steven Huebner, ‘Italianate Duets in Meyerbeer’s Grand Operas’, Journal of Musicological Research 8 (1989), 203–258.

76  Grand opéra – petit opéra and petit opéra was inevitable. As will be shown, this background remained as important for petit opéra as for its larger partner.57 Of all the characteristics of Le comte Ory and all the petits opéras that followed, the absence of formal dance – divertissement and danced chorus – is perhaps the most striking.58 Scribe and Delestre-Poirson included no such scenes in the libretto, and Rossini wrote no dance music. Although the names of members of the corps de ballet occasionally feature in the libretti of petits opéras, these individuals functioned as supernumeraries and danced only to accompany small parts of certain choruses and the instrumental postludes to arias. This certainly explains some existing confusion about these works and indeed about some grands opéras.59 As a general rule, petits opéras never include formal ballet, divertissements or danced choruses, but some sort of stage movement that could have included dance (with minimal trace in the documentary record) may well have taken place.60 Le comte Ory was programmed from the very start as an opera to accompany a ballet. In the first few weeks after its premiere (when it was paired with L’épreuve villageoise), it seems to have been tested out alongside almost every

57 As with petit opéra, Italian forms remained in grand opéra until the end of the period under discussion. For Gounod, see idem, The Operas of Charles Gounod (Oxford: Clarendon Press, 1990), 267–280, and for Bizet and Massenet, see idem, ‘Configurations ­musico-dramatiques dans les grands opéras de Jules Massenet: reflets français de modèles italiens’, L’opéra en France et en Italie (1791–1925): une scène privilégiée d’échanges littéraires et musicaux, ed. Hervé Lacombe, Publications de la Société Française de Musicologie 3/8 (Paris: Société Française de Musicologie; Klincksieck 2000), 117–139. 58 It is important to distinguish between a danced chorus, created as such by librettist and composer and usually titled in the score as chœur dansé, where the choreographer would have had a creative role as well (as in grand opéra), and a chorus which might be suitable for accompaniment by some sort of stage action including dance. In the first case (chœur dansé), dance was obligatory, in the second it was not. 59 See the hesitation in the table in Smith, Ballet and Opera, 201–212, and especially note 12. Her speculation that the choreography ‘perhaps … was not elaborate enough to warrant attribution’ is supported by the evidence of petits opéras in one and two acts, and her suggestion that ‘in some cases, perhaps, the dancers did not dance but carried out some sort of stage business that required special skills lacked by the chorus members [some sort of movement of which chorus members were incapable or unwilling presumably]…’ is not only apposite for petit opéra but also explains much of the introductory material in the libretti of grand opéra where dancers are listed but for whom there is no obvious or formal context (dance music or danced chorus) in the score. 60 Tracing dance at a level less formal than the divertissement, with its attributed choreography, named dancers and composed dance-music, is severely problematic not only for the Opéra but also for the Opéra-Comique (where its importance is barely recognised) and the Théâtre-Lyrique. Evidence has to be pieced together from libretti, livrets de mise-enscène and occasionally from musical sources, although the latter is patchy in the extreme. Particularly problematic in this regard are the livrets de mise-en-scène from the period of Le comte Ory and Le philtre. There survives no livret de mise-en-scène for Le philtre and those that survive for Le comte Ory are severely problematic. Two of the three survivors date from the twentieth century (Paris, Bibliothèque historique de la Ville de Paris, C 23 (II) and ibidem, Mes 62 (4); the third, ibidem, C 23 (I), is undated and is insuffiently detailed to indicate the levels of informal dance deployed at is 1828 premiere).

Grand opéra – petit opéra  77 Table 2.2  O  pera and ballet at the Académie Royale de Musique, August– September 1828 (titles only and number of acts; details of librettist, composer and premiere in notes to the main text) Date

Opera

1 August 4 August 6 August 8 August 11 August 13 August 15 August 20 August

La muette de Portici (5) Moïse (4) La muette de Portici (5) Aladin (3) La muette de Portici (5) Le siège de Corinthe (3) Le rossignol (1) Le comte Ory (2) [premiere] Le comte Ory (2) Le comte Ory (2) Le comte Ory (2) Le comte Ory (2) La muette de Portici (5) Le comte Ory (2) Le comte Ory (2) Le comte Ory (2) Le comte Ory (2) Le comte Ory (2) Aladin (3) Fernand Cortez (3) Le siège de Corinthe (3) La muette de Portici (5) Fernand Cortez (3) La muette de Portici (5) Aladin (3)

22 August 25 August 27 August 29 August 1 September 3 September 5 September 8 September 10 September 12 September 15 September 17 September 19 September 22 September 24 September 26 September 28  September

Ballet

Le carnaval de Venise (2) Les pages du Duc de Vendôme (1) La somnambule (3) L’épreuve villageoise (2) Mars et Vénus (4) Les pages du Duc de Vendôme (1) Astolphe et Joconde (2) Mars et Vénus (4) Mars et Vénus (4) La somnambule (3) Lydie (1) Astolphe et Joconde (2) La somnambule (3) Les pages du Duc de Vendôme (1) Mars et Vénus (4) La somnambule (3) Astolphe et Joconde (2) Les pages du Duc de Vendôme (1)

ballet-pantomime currently in the repertory: Mars et Vénus, Les pages du Duc de Vendôme, Astolphe et Joconde, La somnambule and Lydie (Table 2.2).61 Table 2.2 also illustrates the way in which La muette de Portici and Moïse – hardly surprising, since this is only a few months after the former’s premiere and still during the latter’s first run – occupy all the repertorial space allowed for single operas and how pairing three-act works with b ­ allets  – Fernand Cortez, even Le siège de Corinthe, Le rossignol and Aladin – still remained part of the administration’s artistic policy.62 But even this small segment demonstrates the exclusively symbiotic relationship between the newly premiered 61 Mars et Vénus, ou Les filets de Vulcain (music by Jean-Madeleine Schneitzhœffer, choreography by Jean-Baptiste Blache), premiered 29 May 1826; Les pages du Duc de Vendôme (music by Adalbert Gyrowetz, choreography by Aumer), premiered 18 October 1820; Astolphe et Joconde, ou Les coureurs d’aventures (music by Hérold, choreography by Aumer), premiered 29 January 1827; Lydie (music by Hérold, choreography by Aumer), premiered 2 July 1828. 62 Fernand Cortez, ou La conquête du Mexique (music by Spontini, libretto by d’Esménard and de Jouy) had been premiered on 28 November 1809, Aladin, ou La lampe merveilleuse (music by Nicolas Isouard, libretto by Charles-Guillaume Etienne) on 6 February 1822.

78  Grand opéra – petit opéra Comte Ory – that would characterise performance contexts for all petits opéras that were modelled on it – and ballet-pantomime.63 In short, Rossini’s petit opéra was never performed on its own during its life at the Opéra. While the configuration of Le comte Ory as the template for the composition of petits opéras may be regarded as the consequence of happenstance, its almost immediate successor shows all the signs of having been carefully constructed to replicate almost all of its features. Premiered on 20 June 1831, Scribe and Auber’s Le philtre stands in the same relationship to Robert le diable as does Le comte Ory to La muette de Portici: as the season’s petit-opéra foil to its grand opéra. Le philtre is today much less well known than Le comte Ory, although its libretto is celebrated as the basis for Donizetti’s L’elisir d’amore premiered in Milan the following year.64 When Véron took over the Opéra in 1831, the importance of the type of work represented by Le comte Ory was already understood, and Le philtre was already in rehearsal. Consequently, in all versions of Véron’s cahier des charges, his repertory is divided into two, according to the 1806–1807 decrees mentioned earlier: 1 Grand or petit opéra with or without ballets, and 2 Ballet-pantomime and while the next clause in the document specifies the annual production of a new grand opéra, it also requires the manager to produce two new petits opéras a year.65 As Scribe and his musical colleagues were evolving solutions to the difficulties created by grand opéra that resulted in petit opéra, the repertorial sea-change at the Opéra was formally recognised in the rules that governed the institution’s work and the manager’s room for manoeuvre. In Duponchel’s 1835 cahier des charges, although the distinction between grand and petit opéra was glossed over in favour of a bland description of works in one, two, three, four or five acts, later conditions made clear that petit opéra was key to the functioning of the theatre. Among the complex rules concerning the timing of new productions was the concession that ‘in the case of petits opéras and ballets or their substitutes, the manager may advance them or delay from the first year to the second, from the third to the fourth and so on’.66 With a libretto and score now composed completely ab initio, the physiognomy of Le philtre bears direct comparison with that of Le comte Ory (Table 2.3).67 63 Table 2.2 may be compared to the analogous listing for 1843 given, for other purposes, in Smith, ‘Dance and Dancers’, 94. 64 See Chapter 8 and the sources cited there. 65 ‘Cahier des charges …’ 28 February 1831. F-Pan F21 4655(2). 66 ‘Quant aux petits opéras et aux ballets ou à leurs remplacements, l’entrepreneur pourra les mettre en avance ou rester en retard de la première année sur la deuxième, de la troisième sur la quatrième, et ainsi de suite’ (‘Académie Royale – Cahier des Charges’. M. D ­ uponchel. 19 August 1835. F-Pan F21 4655(3)). 67 Table 2.3 is based on the archive full score found in F-Po A.498.a.I-IV; in terms of division into numbers it agrees with all published scores. See LE PHILTRE, / OPÉRA EN DEUX

Grand opéra – petit opéra  79 Table 2.3  A  uber, Le philtre No

Title and cast

Incipit

Scene

‘Amis, amis, sous cet épais feuillage’

i

‘Ah qu’un philtre’ ‘Je suis sergent’ ‘Je suis fière’ ‘La coquetterie fait mon seul bonheur’ ‘Guéris toi’ ‘Est-il possible d’être insensible’ ‘Quel bruit soudain’ ‘Quel brillant équipage’ ‘Puisque, pour nous guérir’ ‘Philtre divin’ ‘Quel délire nouveau’ ‘C’est Guillaume, allons du courage’ – ‘Non, il reste’

i ii ii–iii iii iv iv iv–v v v vi vi vii

Ouverture  1

 2  3  4  5  6  7  8

 9

10 11 12

13

14

ACTE I Introduction, Ballade (Chorus, Guillaume, Térézine, Jeannette) Récit Marche et air (Jolicœur) Récit Air (Térézine) Récit Chorus Récit Air (Fontanarose) Récit Air (Guillaume) Récit [Récit +] Duo (Térézine, Guillaume) [attaca] [Récit +] Trio (Térézine, Guillaume, Jolicœur) et Finale ACTE II Entracte et chœur, couplets, barcarolle (Chorus, Jeannette, Térézine, Fontanarose) Récitatif et chœur Récit [Récit +] Duo (Guillaume, Jolicœur) Récit Morceau d’ensemble (Térézine, Jeannette, Guillaume, Jolicœur, Fontanarose, Chorus) Récit [Récit +] Duo (Guillaume, Térézine) Récit Finale

‘Que vois-je, et pour quelle joie’ – viii–ix ‘Dedans le cours de mes conquestes’

‘Chantons, chantons’ – ‘Habitans des bords de l’Ardour’ – ‘Je suis riche et vous êtes belle’

i

‘O doux aspect’ – ‘Chantons, chantons’ ‘Voici le soir, l’heure s’avance’ ‘De désespoir je reste anéanti’ – ‘Si l’honneur a pour toi des charmes’ ‘Signe ou bien fais ta croix’ ‘Grand Dieux, quelle nouvelle’

i ii iii

‘Comme il a l’air heureux’ ‘Effet miraculeux’ – ‘Je voulais partir pour la guerre’ ‘De mon art ce sont là les effets ordinaires’ ‘Je lui dois ma maîtresse’

vii viii

iii iv–vi

viii viii

ACTES, / PAROLES DE M. SCRIBE, / MUSIQUE DE M. AUBER / REPRÉSENTÉ POUR LA PREMIÈRE FOIS / SUR LE THEATRE DE L’ACADÉMIE ROYALE DE MUSIQUE, / LE 15 [sic] JUIN 1831 / … / PARIS. / BEZOU, LIBRAIRE, / BOULEVARD S.-MARTIN, NO 29.

80  Grand opéra – petit opéra When Scribe wrote the libretto to Le philtre, he modelled its coupe closely on that of Le comte Ory: two acts with nine scenes in the first and eight in the second (as opposed to nine and eleven in Le comte Ory).68 This translates into approximately a dozen compositions in each, mostly airs and duos with nothing more elaborate than the occasional trio outside finales and introductions. Both operas clearly separate composed number from recitative, and there is no divertissement. With a single exception, the cast of characters is similarly limited to five (Table 2.4). Roles map onto voices in largely similar ways. Jawurek played a trousers role in Le comte Ory as opposed to the more straightforward comprimaria in Le philtre, and Rossini therefore had a single additional female character, Ragonde, to make good this shortfall. Ory and Guillaume posed similar challenges to Nourrit, although the two roles written for Levasseur – Ory’s long-suffering governor and the charlatan Fontanarose – ­d iffer in tone. Dabadie must have run the risk of serious confusion playing the similar figures of Raimbaud and Jolicœur, although the former enjoyed the status of a knight while the latter held the rank of a mere sergeant. And this last difference points to the complementary registers of the two libretti: in Le comte Ory, all the protagonists are noble, even the comprimario/a Rimbaud and Ragonde, whereas in Le philtre, the characters are largely peasant, with even the military represented by relatively low-born characters. The social status of Fontanarose, despite – or perhaps because of – his 68 Scribe’s original notes on the libretto to Le philtre are found in his carnet that was concluded in 1826. There is a short scenario which is lined through twice, once to indicate that part was used for Le philtre and a second time to show that the rest was used in Haydée of 1848 (Paris, Bibliothèque nationale de France (hereafter F-Pn) n.a.f. 22584 (8), fols 47r-48r). Claims have, however, been made that the libretto to Le philtre was based on an Italian original by Silvio Malaperta (Corneloup, ‘De Scribe à Romani, d’Auber à Donizetti’, 5) that had been published by Stendhal, entitled Le philtre, the previous year in the Revue de Paris 15 (1830), 24–40 (see Karin Pendle, Eugène Scribe and French Opera of the Nineteenth Century, Studies in Musicology 6 (Ann Arbor, MI: UMI Research Press, 1979), 457). However, it is certain that the Silvia Valaperta (Malaperta is a transcription error that emerged in the 1867 reprint of the text) is Stendhal’s invention (Henri Martineau (ed.), Stendhal: Romans et nouvelles II, Bibliothèque de la Pléiade 13 (Paris: Gallimard, 1948; R 2002), 1478), and based on a story entitled ‘L’adultère innocent’ in Paul Scarron, Les nouvelles tragicomiques, 2 vols [bound in one] (Paris: David, 1701) (ibidem, 1062) which itself is based on a story by Maria de Zayas y Sotomayor, Al fin se paga todo (Yves Ansel and Philippe Berthier (eds.), Stendhal: Œuvres romanesques complètes I, Bibliothèque de la Pléiade 4 (Paris: Gallimard, 2005), 955). See also Louis Royer, ‘Stendhal imitateur de Scarron’, Mercure de France 255 (1934), 251–268, and Maurice Bardèche, Stendhal romancier (Paris: La Table Ronde, 1947), 168–173. Earlier accounts of Le philtre make no attempt to comment on the origin of the libretto; see, for example, Ebenezer Prout, ‘Auber’s Le philtre and Donizetti’s L’elisir d’amore’. Even when this detail is understood, the attraction of the Stendhalesque background to both Le philtre and L’elisir d’amore can be irresistible; see Anselm Gerhard, ‘Ein missverstandener Schabernack: Gaetano Donizettis eigenwilliger Umgang mit Felice Romanis L’elisir d’amore’, ‘Una piacente estate di San Martino’: Studi e ricerche per i settant’anni di Marcello Conati, ed. Marco Capra, Quaderni di musica/Realtà: Supplemento 1 (Lucca: Libreria Editrice Musicale, 2000), 117–126.

Grand opéra – petit opéra  81 Table 2.4  Roles and artists in Rossini, Le comte Ory and Auber, Le philtre, with related works Artist

Le comte Ory

Le philtre

La muette de Portici

Guillaume Tell

Robert le diable

Adolphe Nourrit HenriBernard Dabadie NicolasProsper Levassseur Laure CintiDamoureau Constance Jawurek Mori

Le comte Ory

Guillaume

Masaniello Arnold

Robert

Raimbaud

Jolicœur

Pietro

Guillaume



Le Gouverneur Fontanarose —

Walter

Bertram

La comtesse

Térézine

Elvire

Mathilde

Isabelle

Isolier

Jeanette







Ragonde





Hedwige



fraudulent claims (not least about his birth) is impossible to clarify. Deception of the sort that requires theatrical suspension of belief – the credibility of Fontanarose and the multiple disguise and cross-dressing of Ory, Isolier and their colleagues, for example – is central to the two comic plots, and a simple moral is easy to identify as the characteristic of the denouement. The registral distinction between the gothique troubadour of Le comte Ory and the absorption of the commedia dell’arte characters into Le philtre points to the wide range of comic plots supported by petit opéra. Narratives extend from the most frivolous buffo libretti of, for example, Marliani’s La xacarilla to the darker colours of Dietsch’s Le vaisseau fantôme. Such a wide range of comic register was also identified by theoreticians of comedy in the early nineteenth century. In his Lectures on the English Comic Writers, originally published in 1819 but frequently reprinted during the nineteenth century, William Hazlitt deplored the disappearance of the consistency of what he called ‘conversation and dress’ that characterised the comedy of the previous century and was suspicious of the ‘critical and analytical’ comedy of his own time.69 For Hazlitt, the disappearance of period dress – a feature of eighteenth-century comedy and which circumscribed its register – in favour of modern dress – which could encompass a variety of comic registers  – was a retrogressive step. However, it corresponds exactly with the variety of topics treated not only in petit opéra but also in early nineteenth-century opera buffa.70 Charles Lamb, in his essay ‘On the Artificial Comedy of the 69 William Hazlitt, Lectures on the English Comic Writers, 3rd ed. (London: Templeman, 1841), 330–331. 70 For comic register in nineteenth-century opera buffa, see Francesco Izzo, ‘Donizetti’s Don Pasquale and the Conventions of Mid-Nineteenth-Century Opera Buffa’, Studi musicali 33 (2004), 387–431.

82  Grand opéra – petit opéra Last Century’, almost contemporary with Hazlitt, took exactly the same view.71 Such a position was also found in German aestheticians of the same period. Only slightly earlier than Hazlitt and Lamb, Jean Paul (Johann Paul Friedrich Richter) had already drawn a distinction between dramatic and lyric comedy and – tellingly for Le philtre with its commedia dell’arte cast of characters – identified the figure of Harlequin as ‘no better intermediary spirit’ between the different registers of comedy.72 To be sure, Jean Paul’s framework would not encompass a narrative as dark as that of Le vaisseau fantôme (which was recognised in 1842 as inappropriate for petit opéra in any case) but serves well for works in the genre from Le comte Ory to Victor Massé’s La mule de Pedro of 1863. Much of the Italian colour of Le comte Ory and Le philtre, to say nothing of contemporary grand opéra, was enabled, if not propelled forward, by the fact that two of the key artists at the Opéra around 1830 had extensive careers in Italian opera behind them. Although the claim that Dabadie sang in both Le philtre and L’elisir d’amore in successive years has recently been shown to be false, both Levasseur and Cinti-Damoreau brought wide-­ ranging experience of Italian opera to the institution.73 Cinti-­Damoreau had enjoyed a successful career at the Théâtre-Italien from 1816 to 1825 when she moved to the Opéra, while Levasseur had a similar career, not only at the Théâtre-Italien but also at the Italian Opera in London, and had premiered Meyerbeer’s Margherita d’Anjou in Milan in 1820.74 With a very few explicable exceptions, then, the physiognomy of Le philtre is very close to that of Le comte Ory. In terms of its programming, it is also very similar. Véron’s formal opening was on 1 June 1831, with Guillaume Tell in its revised, three-act, version. The premiere of Le philtre was three weeks later (Table 2.5). Guillaume Tell is paired with four different ballets, and other works similarly played with ballets are – hardly surprisingly – Le comte Ory and Le rossignol. Among the ballet performances is the premiere of L’orgie, a ­ballet-pantomime by the omnipresent Scribe.75 The only five-act grand opéra at this point remains La muette de Portici, a situation to be changed only with the premiere of Robert le diable in November 1831, and it is 71 The essay appears in Lamb’s Elia (London: Taylor and Hessey, 1823; R Oxford: Woodstock Books [A Woodstock Facsimile], 1991), 323–337. 72 Jean Paul (Johann Paul Friedrich Richter), Introduction to Aesthetics, trans. Lee Chadeayne, Theories of Comedy: Critical Writings from the Classical Period to the Present, ed. Paul Lauter (Garden City, NY: Doubleday, 1964), 322. 73 The Dabadie at the 1832 premiere of L’elisir d’amore was almost certainly Dabadie’s younger brother, Justin Dabadie, for which see 297–301. 74 For the fullest summaries of the careers of Levasseur and Cinti-Damoreau, see Karl-­Josef Kutsch and Leo Riemens, Großes Sängerlexikon, 4 vols (Bern and Stuttgart: Francke, 1987–1994; 3rd edn, 7 vols, Bern: Saur, 1997–2002), 3:2056 and 2:644. 75 L’orgie (music by Carafa, scenario by Scribe and choreography by Coralli) was premiered on 18 July 1831.

Grand opéra – petit opéra  83 Table 2.5  Opera and ballet at the Académie Royale de Musique, June–July 1831 (titles only and number of acts; details of librettist, composer and premiere in notes to the main text) Date

Opera

Ballet

1 June 3 June 6 June 8 June 10 June 13 June 15 June 17 June 20 June 22 June 24 June 27 June 29 June 1 July 4 July 6 July 8 July 11 July 13 July 15 July 18 July 20 July

Guillaume Tell (3) La muette de Portici (5) Guillaume Tell (3) Le comte Ory (2) La muette de Portici (5) Le rossignol (1) Le comte Ory (2) Le dieu et la bayadère (2) Le philtre (2) [premiere] Le philtre (2) Le philtre (2) Le philtre (2) Le philtre (2) Le philtre (2) Guillaume Tell (3) Le philtre (2) Le philtre (2) Le philtre (2) La muette de Portici (5) Le philtre (2) Le rossignol (1) L’orgie (3)

La somnambule (3)

22 July 25 July

Le dieu et la bayadère (2) Le comte Ory (2)

La fille mal gardée (2) La belle au bois dormant (4) La belle au bois dormant (4) Mars et Vénus (4) Mars et Vénus (4) La somnambule (3) La belle au bois dormant (4) Manon Lescaut (3) La servante justifiée (1) [new production] Manon Lescaut (3) La belle au bois dormant (4) La servante justifiée (1) Mars et Vénus (4) Manon Lescaut (3) Flore et Zéphire (2) La belle au bois dormant (4) L’orgie (3) [premiere] Le comte Ory (2) (opera and ballet reversed) L’orgie (3) L’orgie (3)

s­ triking that grand opéra alone occupies three nights only in the summer of 1831. Just as in the case of Le comte Ory, Le philtre is paired with almost every ­ballet-pantomime in the repertory, although it does not accompany the newly premiered L’orgie until mid-August, and the latter’s premiere is one of the main reasons why ballet-pantomime and petit opéra are so important in this period and grand opéra so much in the background.76 A ­further work that figures in this careful mix of opéra and ballet-­pantomime is Le dieu et la bayadère, a piece that serves as a prompt for a review of the relationship between music, drama, mime and dance in the period in which petit opéra emerged.

76 Other works in Table 2.5 are La belle au bois dormant (music by Michele Carafa, scenario by Eugène de Planard, choreography by Pierre-Gabriel Gardel) which was premiered on 2 March 1825; Manon Lescaut (music by Halévy, scenario by Scribe and choreography by Aumer) appeared on 3 May 1830. La servante justifiée (music by Rodolphe Kreutzer and choreography by Gardel) had been premiered on 3 September 1818.

84  Grand opéra – petit opéra

Dance and drama – music and mime A consideration of all the ingredients in the recipe of music drama around 1830 points to a constantly shifting attitude to the place of dance in the various established or emergent genres. At one extreme were the works that, with hindsight, would be considered among the originators of grand opéra: La muette de Portici, Robert le diable and – despite the subsequent treatment it received – Guillaume Tell. At the other were such ballets-pantomimes as La somnambule, La fille mal gardée, La belle au bois dormant, Manon Lescaut, L’orgie, La sylphide and Nathalie, ou La laitière suisse whose structure and use of borrowed material are now beginning to be fully understood.77 But falling between these two extremes were generic experiments, some of which remained without any real consequence. For example, the effect of the elaborate ballet in the fifth act of Gustave had little impact either on subsequent ballet-pantomime or on divertissements in grand opéra, which had more to do with the pattern established in La muette de Portici and Robert le diable. Similarly, there were no attempts to follow up the careful balance between opéra and ballet-pantomime in a single work as found in Le dieu et la bayadère and La tentation. Le muette de Portici cannot be understood without a consideration of the role that gave the work its name and of the mime that it embodied. The ways in which the role of Fenella was embedded in the drama, and how mime in general was made to interact with music and dance, have been the subject of several recent studies.78 When Lise Noblet, who took the role of Fenella at the premiere was replaced by Marie Taglioni, this served as a prompt for Scribe and Auber more fully to exploit the talents of the latter (in ten of the eleven numbers) in Le dieu et la bayadère. This work, nominally entitled opéra, and in two acts, was of the dimensions of a petit opéra; but its use of both Noblet and Taglioni in mimed and danced roles gives it a central position in the laboratory of experimental music drama

77 Nathalie, ou La laitière suisse (music by Gyrowetz and Carafa, choreography by Taglioni) on 7 November 1832. See Smith, Ballet and Opera, 6–18 for the differences between music for pantomime and dance music and eadem, ‘Borrowings and Original Music: A Dilemma for the Ballet-Pantomime Composer’, Dance Research 6 (1988), 3–29, for an exploration of the question of borrowed and new material. 78 Emilio Sala, ‘‘‘Que ses gestes parlants ont de grâce et de charmes”: motivi ‘‘melo” nella Muette de Portici’, Atti del xiv congresso della Società Internazionale di Musicologia: Trasmissione e recenzione delle forme di cultura musicale, eds. Angelo Pompilio, Donatella Restani, Lorenzo Bianconi and F. Alberto Gallo, 3 vols (Bologna: EDT, 1988), 1:504–520; Mary Ann Smart, Mimomania: Music and Gesture in Nineteenth-Century Opera, California studies in 19th-century music 13 (Berkeley: University of California Press, 2004), 32–68; Sarah Hibberd, ‘La muette and her Context’, The Cambridge Companion to Grand Opera, ed. David Charlton (Cambridge: Cambridge University Press, 2003), 149–167; ­Maribeth Clark, ‘The Body and the Voice in La muette de Portici’, 19th-Century Music 27 (2003), 116–131.

Grand opéra – petit opéra  85 around 1830 and, critically, at a distance from petit opéra.79 Whether calling it a ‘hybrid’ does the work justice is an open question, but – in the context of the difficulty of finding works to accompany ballets and the emergence of petits opéras – Le dieu et la bayadère was able to define its genre by reference to other works and throughout the nineteenth century was paired both with ballets-pantomime and petits opéras or their surrogates: in the first case, it acted as a petit opéra itself, and in the second, as a ballet-pantomime. The fate of the remaining work that attempted to balance dance, music and drama in a radically different fashion has already been described. Before being a target for morcellement, La tentation wore its generic ambiguity on its sleeve; an opéra-ballet, its nomenclature underpinned a structure that was rarely reproduced. Its main attraction to the Opéra may well have been that the corps de ballet and the chorus had such large roles; while this may have solved the purely local difficulty Véron may have had with indispositions during the 1832 cholera outbreak, it may also have been part of its attraction as a work à morceler during the coming decades.80 Within the context of experimentation with dance, mime, drama and music, the appearance of petit opéra becomes that much more explicable as one of the more extreme types (in that it expresses itself generically without the aid of formalised dance) to have successfully emerged. Scribe’s role was central. Of the grands opéras, ballets pantomimes, opéras, opéras-ballets, petits opéras and ballets féeries that have been discussed in this context, Scribe was involved with all except Guillaume Tell, La tentation and three ballets-­ pantomimes: La fille mal gardée, La sylphide and Nathalie. It is difficult to escape the view that the determining features of experimental stage music around 1830 were their libretti, and that Scribe had written or adapted most of them. Among those involved in the emergence of both grand and petit opéra, few of the musicians, Auber, Halévy, Rossini or Meyerbeer, or the choreographers and other librettists – Aumer, Taglioni, Coralli, Soumet, Balocchi, Jouy, Bis, Cavé, Duponchel or Germain Delavigne – had as much involvement or as much impact as Scribe.81

Petit opéra, 1828–1863 The solutions proposed around 1830 to the problems posed by the creation of grand opéra in five acts continued in use throughout much of the 79 See Smith, Ballet and Opera, 138–149. 80 Ibidem, 149–166. For the 1832 cholera outbreak, see Ange-Pierre Leca, Et le choléra s’abattit sur Paris: 1832 (Paris: Albin Michel, 1982). 81 Scribe had played a central role in the recasting of La muette de Portici from three acts to five for the Opéra, and in the reworking of Robert le diable from an opéra comique in three acts to a grand opéra in five acts. See the sources cited on page 62, note 17.

86  Grand opéra – petit opéra nineteenth century. Morcellement, the reduction of larger works to smaller dimensions and – to a lesser extent – the use of works from before 1828 were all strategies used by Lubbert, Véron and their successors. These remained, however, essentially ad hoc reactions to the problem, and the emergence of a pair of new works, Le comte Ory and Le philtre, by composers currently garnering critical acclaim, Rossini and Auber, that could accompany ­ballet-pantomime was an obvious further answer to the question of how to discharge the obligations of the Opéra towards ballet-pantomime in the age of grand opéra. Le comte Ory and Le philtre remained immensely valuable commodities up to the mid-1860s; the former enjoyed 373 performances up to 1866, was revived in 1837, 1844, 1857 and 1863 and was performed every year except 1850–1852 and 1855–1856; the latter was performed 242 times up to 1862, received new productions in 1842 and 1861 and similarly was performed every year except 1850–1851, 1853 and 1858–1860.82 But the role of these two works was not restricted to performances: together, they acted as a catalyst for the composition of an entire repertory of petits opéras during the next forty years (Table 2.6). Table 2.6 gives details of all eighteen petits opéras composed during the period together with a small but significant handful of foreign works which, to a greater or lesser degree, functioned as surrogates for petit opéra. It has already been seen how the two types of grand opéra and petit opéra account for more than 90% of the repertory of the Opéra between 1828 and 1864. The handful of remaining three-act works constitutes no more than a negligible part of this history. Auber’s Le serment has been shown almost never to have been performed except in performances of individual acts morcelé to accompany ballet. Gounod’s Sapho and Ruolz’s La vendetta were both swiftly reduced to two acts and made to function like petits opéras during their short lifetimes. Apart from the reworking of Auber’s opéra comique, Le cheval de bronze (originally 1835), in 1857 as an opéra-ballet, only two other three-act operas were produced during the entire period. Halévy’s Le drapier of 1840 and Auber’s Zerline (1851), both to libretti by Scribe, failed lamentably (eight performances for the former and fourteen for the latter). Most composers of petit opéra after 1831 were composers who had yet to make their mark at the Opéra with a major work in five acts. Apart from Rossini and Auber, the only composer with any previous experience at the Opéra was Halévy, whose Le lazzarone of 1844 was produced only after success in grand opéra with La Juive, Guido et Ginevra, La reine de Chypre and

82 See Lajarte, Bibliothèque de l’Opéra, 2:131–132 and 138–139; Albert Soubies, Soixante-sept ans à l’Opéra en une page du ‘Siège de Corinthe’ à ‘La Walkyrie’ (1826–1893) (Paris: ­Fischbacher, 1893).

Table 2.6  Repertory of petit opéra at the Académie Royale de Musique, 1828–1863 Composer Gioachino Rossini

Librettist

Eugène Scribe and CharlesGaspard Delestre-Poirson [Daniel-François-Esprit Auber Eugène Scribe Daniel-François-Esprit Auber Eugène Scribe Marco Aurelio, comte Eugène Scribe Marliani Ambroise Thomas Eugène Scribe Carl-Maria von Weber, arr. Friedrich Kind, arr Émilien Hector Berlioz Pacini and Berlioz Fromental Halévy Théodore Anne Pierre-Louis-Philippe Dietsch Paul Foucher and Bénédict-Henry Révoil Fromental Halévy Jules-Henry Vernoy de Saint-Georges Gaetano Donizetti Salvadore Cammarano, arr. Alphonse Royer and Gustave Vaëz Friedrich Adolf Ferdinand Jules-Henry Vernoy de Freiherr von Flotow Saint-Georges Adolphe Adam Hippolyte Lucas François Benoist Germain Delavigne Adolphe Adam Jules-Henry Vernoy de Saint-Georges Jacob Rosenhaim arr. Jean-François-Alfred Bayard and Etienne Arago

Petit opéra

Foreign import

Date

Acts

Le comte Ory

1828

2

Le dieu et la bayadère Le philtre La xacarilla

1830 1831 1839

2] 2 1

1841 1841

2 3

1842 1842

2 2

1844

2 4

L’âme en peine

Lucie de Lammermoor 1846 (Lucia di Lammermoor) 1846

La bouquetière L’apparition Le fanal

1847 1848 1849

1 2 2

1851

2

Le comte de Carmagnola Le guérillero Le vaisseau fantôme

Le freyschütz (Der Freischütz)

Le lazzarone

Le démon de la nuit (Liswenna)

2

(Continued)

Composer

Librettist

Giuseppe Verdi

Salvadore Cammarano, arr. Benjamin Alaffre and Emilien Pacini Henri Trianon

Armand-Marie-Ghislain, baron Limnander de Nieuwenhove Gaetano Donizetti, arr. Adolphe Adam Théodore Labarre Emanuele Biletta Edmond Membrée Jules Alary Victor Massé

Gaetano Donizetti, arr. Hippolyte Lucas Henri Trianon Jules-Henry Vernoy de Saint-Georges Edmond Got Mélesville (pseud. of AnneHonoré-Joseph Duveyrier) Philippe Dumanoir (pseud. of Philippe-François Pinel)

Petit opéra

Foreign import

Date

Acts

Louise Miller (Luisa Miller)

1853

4

1853

2

1853

2

Pantagruel La rose de Florence

1855 1856

2 2

François Villon La voix humaine

1857 1861

1 2

La mule de Pedro

1863

2

Le maître chanteur Betly (Betly)

Grand opéra – petit opéra  89 83

Charles VI. This is not to say, however, that composers of petits opéras were novices. Looking at their track records before their appearance at the Opéra, it is clear that the composition of petit opéra was viewed as a logical career progression for a composer from opéra-comique and, later in the period opérette, towards a major work there. To a degree, Italian music drama was also considered a plausible background, an interesting consideration given the Italianate features of Le comte Ory and its residue in Le philtre and later works. For example, although Limnander’s first petit opéra was ­­Le maîtrechanteur, he had already enjoyed success at the Opéra-Comique with Les monténégrins (1849) and Le château de Barbe-bleu (1851). Similarly, the first petit opéra produced after Le philtre, Marliani’s La xacarilla (1839), followed two works from the composer for the Théâtre-Italien, Il bravo and Ildegonda, and his Le marchand forain at the Opéra-Comique in 1834. Thomas, too, had enjoyed four annual productions at the Opéra-Comique before his first petit opéra: Le comte de Carmagnola; the earlier careers of Flotow and Alary were similar. And while Adam’s La bouquetière was technically his ­ alestine – had second work for the Opéra, its predecessor – Richard en P been such an abject failure that it was his earlier successes at the Opéra-­ Comique and elsewhere that formed the background to his more successful petit opéra. The only exception to selecting composers with a background in opéra comique for working with petit opéra was the choice of Dietsch, whose Le ­vaisseau fantôme was his only stage work, written while he was simultaneously chef de chant at the Opéra and maître de chapelle at the Madeleine. Initially trained at Choron’s Ecole Royale et spéciale de chant, 83 La lazzarone poses a number of issues. The piano-vocal score (LE / LAZZARONE / Opéra en deux actes, / Paroles de / Mr de St Georges, / Musique de / F. HALÉVY / Membre de l’Institut / … / A PARIS, chez MCE SCHLESINGER, Rue Richelieu, 97 / Berlin. A. M. Schlesinger. / Prope des Éditeurs) appears to have many more numbers than the Comte Ory / Philtre pattern might suggest. However, some of the numbers – had they appeared in ­ umbers either of the two exemplars – would have been run together, and three of Halévy’s n (7, 13 and 15) are in fact the articulatory recitatives that do not receive a separate number in the earlier works (or most petits opéras listed in Table 2.6). The libretto (LE / LAZZARONE, / OU / LE BIEN VIENT EN DORMANT, / OPÉRA EN DEUX ACTES. / PAROLES DE M. DE SAINT-GEORGES, / MUSIQUE DE / F. HALÉVY / - / PARIS. / Mme Ve JONAS, LIRAIRE ÉDITEUR DE L’OPÉRA, / TRESSE, LIBRAIRE, PALAISROYAL, / Galerie de Chartres, 2 et 3, 10 and 13) records two places where dance might be employed. The second of these comes at the end of the act ii tarentule, a duo sung by ­Rosine Stolz and Julie Dorus-Gras. Press references make it clear that Stolz and ­Dorus-Gras themselves danced the tarentule without the assistance of anyone from the corps de ballet (Le corsaire, 31 March 1844 [a hostile review that found no redeeming features in the work]). In none of the voluminous music materials for La lazzarone (Halévy’s autograph sketch material [F-Po Rés. A. 541.a]; the archive full score [F-Po A. 542. b. 1–ii, 2:151–176]; the violon principal [F-Po Mat. 19. 345 (139) which has massive cuts in the tarentule]; the printed piano-vocal score [LE / LAZZARONE / Opéra en deux actes, 145–148]) is there any reference to dance, however, and one is forced to conclude – in accordance with the general principles given above – that any dance must have been very much an ad hoc arrangement.

90  Grand opéra – petit opéra he had ­previously been Maître de chapelle at Saint-Eustache. The consequences of the collision between this background largely in sacred music and petit opéra, to say nothing of the scenario for the libretto by Richard Wagner, are discussed later. From the late 1830s to the 1860s, petits opéras appeared at the rate of around one every two years. The clear gap between Le comte Ory and Le philtre and the next work, Marliani’s La xacarilla, is explained by the immense success of the two earlier works between 1831 and 1839 and the continuing predilection for morcellement which was most preferred during the 1830s. Significantly, the first two petits opéras after Le comte Ory and Le philtre were both to libretti by Scribe. Although Thomas’ Le comte de Carmagnola is nominally in two acts, only eight numbers were composed for it (the lowest number of any petit opéra); paradoxically, Marliani’s La xacarilla, although nominally in one act (although split into two tableaux), contains eleven numbers, much in line with the template outlined in Le comte Ory and Le philtre. Adam’s La bouquetière, also nominally in one act, is similar to other two-act petits opéras in terms of its dimensions.84 Table 2.6 includes a single work, Benoist’s L’apparition, with a libretto by Germain Delavigne – almost his only libretto written without the collaboration of Scribe (and it is difficult to believe that Scribe was unaware of the work of his exact contemporary and classmate) – that shows all the signs of having been written as a petit opéra in two acts, with eight scenes in the first and nine in the second. The music, however, runs to twenty-two numbers and contains a full divertissement entitled Zapateado, which would not have been out of place in a grand opéra, and other dances throughout. In this, it far outstripped the generic range of petit opéra. Its fate is difficult to judge: premiered on 16 June 1848, it received two performances only. Paired with La sylphide and Le diable à quatre, it folded when the Opéra closed its doors as revolution swept Paris.85 84 A single act of 16 scenes and nine composed numbers. 85 It would be too easy to dismiss L’apparition as an aberration, the work of an unknown composer unfamiliar with practices at the Opéra. But Benoist had been troisième chef de chant since 1835, and since 1840 charged with coaching the soloists. He was also the composer of the music for the ballets-pantomimes La gipsy (the first act only, Thomas wrote the second and Marliani the third; scenario by Saint-Georges and choreography by Mazilier), premiered 29 January 1839; Le diable amoureux (the second act was by Henri Reber, scenario by Saint-Georges and choreography by Mazilier), premiered 23 September 1840; Nisida, ou Les amazones des Açores (scenario by Eugène Deligny with choreography by Auguste Mabille), premiered on 21 August 1848; Pâquerette (scenario by Théophile Gautier and choreography by Charles-Victor-Arthur Saint-Léon), on 15 January 1851. Benoist’s other stage work before L’apparition was his opéra comique, Léonore et Félix (libretto by Jean-Victor Fontanès de Marcelin), premiered shortly after his return from his laureate’s tour of Italy on 27 November 1821. He was, however, also responsible for the extensive reworking of Rossini’s Otello premiered at the Opéra in 1844. This epoch-making production (the first of an Italian work by a living Italian composer during the period) is discussed in Chapter 9 and idem, Rossini’s Second grand opéra’ ‘Rossini at the Paris Opéra,

Grand opéra – petit opéra  91 Two examples, one from 1849 and one from 1863, illustrate the consistency of petit opéra across the period. Adam’s second work in the genre, Le fanal, was a setting of a libretto by Saint-Georges in two acts with eleven and eight scenes, respectively, and four characters. Victor Massé’s La mule de Pedro, to a libretto by the prolific Philippe Dumanoir, consists again of two acts, this time with eight and nine scenes, respectively, and five characters. There is no divertissement in either work, and the subject matter is comic throughout.86 Both then follow the structural patterns established at the end of the 1820s by Scribe for Le comte Ory and Le philtre in terms of scene disposition and number of compositions.87 Of the two Italianate trends in petit opéra – the predilection for Italian formal types and the careful distinction between recitative and composed ­number – the first, shared with grand opéra, was enhanced by the Italian Comte Marliani, most of whose operas before and after La xacarilla were for Italian opera houses.88 The survival of Italian forms in the hands of such quintessentially French composers of opéra comique (where such Italian forms are rarer) as Thomas, Adam and Massé is perhaps less expected. However, Massé’s La mule de Pedro includes a duo for Pedro and Tebaldo entitled ‘Le voilà, ce rival’ that is nothing less than a four-section Italianate duo with a cabaletta, and even Adam cannot resist an Italianate primo tempo-cabaletta pair for Martial in the first act of Le fanal that exploits Placide-­A lexandre-Guillaume Poultier’s effortless legato high c in a the first movement and a structure, if not the éclat of a cabaletta, in the second.89

86

87

88

89

1843–1847: Translation, Arrangement, Pasticcio’, Librettoübersetzung als Phänomen der Interkulturalität im europäischen Musiktheater, Musikwissenschaftliche Publikationen 32, ed. Herbert Schneider and Rainer Schmusch (Hildesheim: Olms, 2009), 131–163. The only slight exception to the comic tone of the two works might be thought to be the abduction of Gilda by Pedro in La mule de Pedro. While several different readings of this section of the opera would today be possible, if not desirable, the immediate comic effect of Gilda locking Pedro in the cellar after luring him into drunkenness and the entirely fortuitous arrival of Tebaldo (brought thence by Pedro’s mule) serves to maintain what for 1863 clearly fell into a comic register. The differences in tone between Le fanal and La mule de Pedro again indicate the wide range of comic register at work in petit opéra. See Genevieve Chinn, ‘The Académie Impériale de Musique: a Study of its Administration and Repertory from 1862–1870’ (PhD diss., Columbia University, 1969), 98–111, for an account of Le mule de Pedro and notes on the new production of Le comte Ory, which, however, is silent on the history of petit opéra before 1861. Marliani’s La xacarilla clearly betrays the Italian origins of its composer and his earlier Italian works, most obviously in two successive numbers in the work. The aria ‘Mon Dieu, que faut-il faire?’ is a two-tempo primo tempo-cabaletta pair, and the duo ‘Oh délice suprême’ exhibits an uncompromising four-section structure on the traditional Rossinian model. On the other hand, the last number before the finale is a ronde avec chœur, a type of number that could not be more quintessentially French. Poultier was an important tenor at the Opéra from 1842 until 1856, and took part in the premieres of Halévy’s Charles VI, Benoist’s L’apparition and David’s Eden as well as ­Adam’s Le fanal. See Kutsch and Riemens, Großes Sängerlexikon, 4:2793.

92  Grand opéra – petit opéra The more general concern, which may be traced back via Le comte Ory to Italian practices of the 1820s, to follow each number with a discrete recitative characterises petit opéra from its inception up to the end of the licensing period and continuously differentiates it from grand opéra. In both Le fanal and La mule de Pedro, almost every number is followed by a separate recitative in the manner of Le comte Ory, Le philtre and their ­Italian predecessors. The only exceptions concern the penultimate numbers in each act where there is no recitative and the numbers run on without a break, a principle that goes back to the eighteenth century. This tendency has already been observed in the two numbers that close the first act of Le philtre and is furthermore an obvious response to the need to accelerate the action towards the end of the act. Although Massé calls these ‘scène’, ‘scène et récitatif’ or even ‘scènes et récitatifs’, there is no difference either in placement or function between these sections in La mule de Pedro and any other previous petit opéra. Where they do differ is in their ambition. Massé, even when writing opéra comique, had been criticised for stretching the boundaries of the genre, and Galathée was cited as a good example.90 And when he turned to writing his first work for the Opéra, he embellished the sections between numbers by including fragments of choral recitative in the section that follows his introduction. In the recitative after the chanson de la mule, he builds the music around a fully composed four-bar passage which he then repeats and fragments during the course of the scene. ­Furthermore, in the recitative that follows Tebaldo’s act i aria, Massé invokes a reminiscence of the romance that had preceded the aria, a practice with a long history in Italian opera. Large parts of the kinetic sections of the score also exhibit a discursive mode that would not have been out of place thirty years earlier. A sense of how composers themselves, or at least their copyists, thought of the distinction between recitative and composed number within the domain of petit opéra is revealed in the Opéra’s manuscript copy of Adam’s La bouquetière (1847), a work that in general conforms clearly to the general

90 In two of his recitatives, Massé interpolates additional arias that are printed in the ­piano-vocal score and that are marked to be cut in the opera house. Whether this is simply a residue of a routine working practice that involved over-composition or evidence of some further tension with the genre is not clear from the surviving documents. See A MADAME LA DUCHESSE DE TARENTE / LA / MULE DE PEDRO / OPÉRA EN 2 ACTES / DE M. DUMANOIR MUSIQUE DE / VICTOR MASSÉ , Partition PIANO et CHANT arrangée par VAUTHROT / … / Paris, au Magasin de Musique du Bazar de l’Industrie française, / O. LEGOUIX, Éditeur, Boulev t. Poissonnière, 27, 148, 195 and 248–251 and 252–259. Whatever the reason for the two excisions, they are congruent with Massé’s ambitions in the recitative sections of the work in general. For the view that Massé’s Galathée went beyond the acceptable dimensions of opéra comique, see Jacques Offenbach’s manifesto on the state of the genre in the mid-1850s (discussed in the next chapter).

Grand opéra – petit opéra  93 patterning of recitative and composed number found in petit opéra.91 In two cases, the recitatives that follow airs for Le vicomte and Nanette (numbers 2 and 4, respectively) are titled in the manuscript ‘Récitatif après le no. 2’ and ‘Récitatif après le no. 4’. Here is a word for word translation, twenty years after the emergence of petit opéra, of the commonplace Italian terms, recitativo dopo l’aria, il duetto etc., alluded to in the context of Le comte Ory. La bouquetière not only provides evidence of the continuing consistency of petit opéra but also of the longevity of its Italian origins. The programming of Adam’s Le fanal provides evidence of a slightly different approach to the combination of opera and ballet than to that of the 1830s, where the new petit opéra would be paired with almost every ballet currently in the repertory. Le fanal was premiered alongside Le violon du diable exclusively for its first seven performances.92 Apart from one performance alongside La vivandière, all its remaining performances were with a further single ballet: Stella, ou Les contrabandiers.93 Similarly, Alary’s petit opéra, La voix humaine, was premiered in 1861 alongside L’étoile de Messine and hardly ever appeared in combination with any other ballet.94 Examining the overall structure of libretto and music of petit opera points to a consistency of practice throughout the half century under discussion, with an origin for these practices in Le comte Ory and Le philtre. Libretti were of similar dimensions and dramatis personae modelled on the original patterns. Divertissements were completely excluded, and petits opéras were marked out from grand opéra by a clear demarcation between recitative and composed number, even if the orchestral resources for the recitatives were largely the same. Programming practices, using petits opéras to balance out evenings with ballet and as an alternative to morcellement, the use of old works or the modification of grand opéra, were largely consistent. Although it emerged as a pair of creative experiments in response to changing repertorial circumstances at the end of the Restoration, by 1840, petit opéra was understood as much as a clearly defined genre as grand opéra and remained so until the end of the Second Empire.

Foreign music drama The Opéra was well known for the welcome it gave to foreign stage works by composers alive and dead. Mozart and Weber were early beneficiaries 91 La Bouquetière / Opéra / en un Acte / paroles de Mr. Hippolyte Lucas / Musique / de Mr. Adam. / Représenté pour la 1ere fois sur le théâtre de l’Académie / Royale de Musique, le lundi 31 mai / 1847 (F-Po A.556 (i–ii)). 92 The music of Le violon du diable was by Cesare Pugni, and choreography by Saint-Léon, premiered 19 January 1849. 93 La vivandière and Stella were both also by Pugni and Saint-Léon, premiered on 20 October 1848 and 22 February 1850, respectively. 94 Music by Comte Niccolò Gabrielli, scenario by Foucher and choreography by Pasquale Borri, premiered 20 November 1861.

94  Grand opéra – petit opéra at the beginning of the July Monarchy and Bellini in the dying days of the licensing system.95 But the welcome for living composers came at a cost: significant reworking of the kinetic sections (usually recitative) of their works, especially in the earlier part of the period, in order to exploit the conventions of récitatif simple, accompagné and obligé was required, as well as judicious adjustment of formal types (although more true for Italian than German works) and the composition of ballet music for the divertissement. Most composers thought that the price was worth paying: Donizetti (Les martyrs), Verdi (Jérusalem) and of course Rossini (Othello, Sémiramis and – to an extent – the pasticcio Robert Bruce).96 For both Donizetti and Verdi, it was a step towards La favorite and Dom Sébastien, and Les vêpres siciliennes, respectively. Verdi’s sometimes turbulent relationship with the Opéra continued with another grand opéra, Le trouvère of 1857. In all these cases – and in some, Le trouvère, for example, there was no ballet – the works enjoyed the same status as four- and five-act grands opéras by francophone composers and as new grand opéra by foreigners. Both Verdi and Donizetti also contributed works to the Opéra that functioned as surrogate petits opéras. Louise Miller, which received its Paris premiere on 2 February 1853 but without Verdi’s sanction, was an unalloyed failure and only received eight performances. Each of these was in combination with a ballet-pantomime: Giselle, Orfa and the new production of La fille mal gardée.97 Lucie de Lammermoor, however, was probably one of the most successful French adaptations of an Italian opera throughout the whole of the nineteenth century.98 But apart from its first four performances in 1846, 95 See the productions of Euryanthe in 1831 (Mark Everist, ‘Translating Weber’s Euryanthe: German Romanticism at the Dawn of French Grand Opéra’, Revue de Musicologie 87 (2001), 67–105), and of Don Giovanni in 1834 (Sabine Henze-Döhring, ‘E.T.A. Hoffmann-‘Kult’ und ‘Don Giovanni’- Rezeption im Paris des 19. Jahrhunderts: ­Castil-Blazes ‘Don Juan’ im Théâtre de l’Académie Royale de Musique [sic] am 10 März 1834’, ­Mozart-Jahrbuch 1984/5 des Zentralinstitutes für Mozartforschung der Internationalen Stiftung Mozarteum Salzburg (Kassel, Basel and London: Bärenreiter, 1986), 39–51; ­Katharine Ellis, ‘Rewriting Don Giovanni, or “The Thieving Magpies”’, Journal of the Royal Musical Association 119 (1994), 212–250). The production of Bellini’s I Capuleti e i Montecchi as Roméo et Juliette in 1859 remains as yet unexplored, but see Janet Johnson, ‘“Vieni a veder Montecchi e ­Cappelletti”: Bellini’s Roméo et Juliette, grand opéra’, paper at conference ‘The Institu­ niversity of tions of Opera in Paris: From the July Revolution to the Dreyfus Affair’, U North Carolina at Chapel Hill and Duke University, 24–26 September 2004. 96 Composers played different roles in all the productions noted here: Donizetti was responsible for the changes to Les martyrs from Poliuto, and Verdi’s changes to I Lombardi alla prima crociata almost amount to a new work. Although Rossini’s involvement in the 1844 Othello and the 1860 Sémiramis was non-existent, his sanction – especially as witnessed by his work on Robert Bruce – was clear. 97 Giselle, ou Les willis (music by Adam, scenario by Gautier and Saint-Georges with choreography by Coralli) was premiered on 28 June 1841, Orfa (music also by Adam, scenario by Henri Trianon and choreography by Mazilier) on 29 December 1852. 98 The French version had originally been mounted at the Théâtre de la Renaissance on 6 August 1839. See Everist, ‘Donizetti and Wagner’, 323–329.

Grand opéra – petit opéra  95 it was always part of a mixed evening’s entertainment. Its form, however, created problems for the management of the Opéra, and at the end of the composer’s life, the management was pairing off Lucie de Lammermoor in various combinations of ballets-pantomimes morcelés – single acts of ballets, a practice that was almost without precedent. Lucie de Lammermoor appeared with the first acts of La Péri and Le dieu et la bayadère (the latter fulfilling its role as a ballet on this occasion); neither of these solutions was successful, and the answer to the question of how to programme Lucie de Lammermoor was to bring back the starting time of performances by 15 or 30 minutes and to pair it with a complete ballet.99 It was performed this way for nearly a decade until it too began to be the subject of morcellement in February 1855, when its first two acts were played with a ballet. By this time, a translation of Donizetti’s Betly (his dramma giocoso of 1836) had also appeared, again in company with ballets, but had sunk without trace. Two works by German composers, Weber’s Der Freischütz and ­Rosenhaim’s Liswenna, were also treated as surrogate petits opéras. Le ­freyschütz, its libretto by Émilien Pacini with ballet music arranged and recitatives composed by Berlioz, was played alone for its first dozen performances between June and October 1841; it was then pressed into service to accompany ballet and enjoyed one of the most successful careers of any work of its type.100 Inevitably, later in its career, it was morcelé, and its last two acts performed as part of a mixed evening’s entertainment.101 Rosenhaim’s Le démon de la nuit, the revision of Liswenna for the Opéra ten years later, did not enjoy the same fate. Paired with a ballet at its premiere and clearly conceived from the beginning as a surrogate petit opéra, it only survived for four performances accompanying Paquita and Le diable à quatre.   99 L a péri (music by Frédéric Burgmuller, scenario by Gautier and choreography by Coralli) was premiered on 17 July 1843. 100 Berlioz’s recitatives and his arrangement of Weber’s Aufforderung zum Tanz which formed part of the act iii ballet are edited in Ian Rumbold (ed.), Arrangements of Works by Other Composers (II), Hector Berlioz: New Edition of the Complete Works 22b (­Kassel etc.: Bärenreiter, 2004), 78–199. Why the divertissement is described as ‘another of its [the Opéra] conventions, a ballet to be performed as an entr’acte [emphasis added]’ when it is clearly embedded in the third act is not explained (ibidem, xii). See the unequivocal placement of the divertissement in the printed libretto (LE FREYSCHUTZ / OPÉRA ROMANTIQUE EN 3 ACTES / Paroles de M. Émilien Pacini, / (TRADUCTION DE L’ALLEMAND) / MUSIQUE DE CARL MARIA DE WEBER / DIVERTISSEMENTS DE M. MAZILIER, Décors de MM. Philastre et Cambon. / REPRÉSENTÉ POUR LA PREMIÈRE FOIS / SUR LE THEATRE DE L’ACADÉMIE ROYALE DE MUSIQUE, / LE 7 juin 1841 / La musique des récitatifs est de M. Hector Berlioz / PARIS / CH. TRESSE, SUCCESSEUR DE J. N. BARBA ET BEZOU / …/ 1841, 22. The remaining three numbers of the divertissement are merely noted by reference to Berlioz’s Mémoires: they were taken from Oberon and Preciosa. 101 Berlioz’s participation in later revisions to the recitatives he had composed in 1841 is outlined ibidem, xiv. This particular instance of morcellement meant that the petit opéra surrogate retained its ballet alongside the ballet-pantomime it accompanied. Le freyschütz was the only foreign opera functioning as a surrogate petit opéra that included ballet.

96  Grand opéra – petit opéra Why the ­thoroughgoing supernatural colour of Le freyschütz should fit so well and why the more insipid tints of Le démon de la nuit should fail so comprehensively are questions that lack obvious answers.102 But the success of Le freyschütz certainly points to the registral inclusivity, if not of petit opéra itself, at least of its surrogates, and may have triggered a pair of unsuccessful attempts to elide significantly more serious subjects with petit opéra: Dietsch’s Le vaisseau fantôme and Thomas’ Le guérillero (both 1842).

Phantom ships and phantom libretti Wagner is perhaps the last individual whom his devotees would wish to be considered alongside such composers as Adam, Massé or – especially ­perhaps – Halévy. But one of the most lucrative consequences of Wagner’s first sojourn in Paris was the sale of the scenario that became both Der fliegende Holländer and Dietsch’s Le vaisseau fantôme. In all the discussions of whether or not Wagner sold the scenario to what was then called Le hollandais volant, exactly what the director of the Opéra, Léon Pillet, was buying has been paramount.103 It now seems that the object of the purchase was a prose draft that Wagner had sent to Scribe in May 1840, in the hopes that the latter might turn it into a libretto that Wagner might in turn be able to compose for the Opéra.104 The reasons why the composer thought that Pillet and the librettists who might benefit from the sale, Paul Foucher and Bénédict-Henry Révoil, might have been at all interested in the scenario for Le hollandais volant may now be explained by reference to the tradition of petit opéra and its particular state in 1841–1842. Wagner well described the position of the Opéra when he wrote to Scribe: I have already given you a general idea of my project, Monsieur (which is to procure the libretto of a petit opéra in one act) both because I happen to know that the administration of the Opéra is in need of such a piece and because I believe that such an opera could be accepted and performed much sooner than a large work. (emphasis, except on petit opéra, added).105 102 In 1861, Rosenhaim argued that the failure of the work was a result of Roger’s indisposition followed by two months leave, which meant that the tenor for whom the principal role had been destined was unavailable for most of the season. Letter from Rosenhaim to the Minster of State (Comte Colonna Walewski), 26 September 1861 (F-Pan F21 1053). 103 See the remarkably dense bibliography on the subject assembled in Peter Anthony Bloom, ‘The Fortunes of the Flying Dutchman in France: Wagner’s Hollandais Volant and Dietsch’s Vaisseau Fantôme’, Wagner 8 (1987), 42, note 1. 104 F-Pn n.a.f. 22552, edited in Bloom, ‘Fortunes of the Flying Dutchman’, 63–66. 105 ‘Je vous ai fait entrevoir, monsieur, ce qui est mon dessein (c’est d’avoir le libretto d’un petit opéra en un acte), parce que je sais que l’Opéra en a besoin et parce qu’il faut présumer qu’un tel opéra pourrait être accepté et donné beaucoup plutôt qu’un grand

Grand opéra – petit opéra  97 Exactly how Wagner knew that the Opéra was in need of petit opéra is open to discussion, but with Le comte Ory and Le philtre now a decade old, and with only La xacarilla in the repertory to complement them, the matter of finding a type of work to accompany a ballet had again become pressing. And indeed, Wagner’s elision of petit opéra with works in one act, when the genre encompassed works in both one and two acts but in practice would ultimately emerge in two (including Le vaisseau fantôme), may well have been prompted by his knowledge of La xacarilla (technically in one act but of the scope of one in two), performances of which he could have heard during the time he was in Paris. In terms of the outline of Foucher’s and Révoil’s libretto, the version of the story set by Dietsch fits well into the pattern that was established by Scribe for Rossini, Auber and Marliani and would be followed by librettists up to 1864. The pattern is confused by the division of the first act into two tableaux, but the basic pattern of two acts with a total of seventeen scenes (divided asymmetrically 11 and 6) was closer to Scribe’s original libretti for Auber and Rossini than to his more recent one for Marliani.106 Dietsch’s score, however, bears little relationship to the physiognomy of petit opéra as exemplified in those works, consisting as it does of eight extended numbers, each of which involves multiple changes of soloists, tempo and key.107 This division into discrete but multipartite musical compositions has more in common with ­Wagner’s 1843 score for Der fliegende Holländer than it does with the template for petit opéra. This may prompt a further consideration of the meetings between Wagner and Dietsch, probably brokered by Meyerbeer in late 1839 and early 1840; these could well have involved the exchange of musical and dramatic ideas on the basis of the scénario and have engendered the idea of both an approach to Scribe and the structural similarities of the music.108 Whatever the links between Dietsch’s and Wagner’s music, Le vaisseau fantôme was only performed eleven times, but like the petits opéras that had gone before, it was paired with almost all the ballets then in the repertory: La fille mal gardée, La tarentule, Giselle, La jolie fille de Gand, Le diable ouvrage’ (Gertrud Strobel and Werner Wolf (eds.), Richard Wagner: sämtliche Briefe, 12 vols (Leipzig: VEB Deutscher Verlag für Musik, 1967–2001), 1:390), cited in English translation in Bloom, ‘Fortunes of the Flying Dutchman’, 46–47. 106 LE / VAISSEAU FANTOME / OPÉRA EN DEUX ACTES, / PAR / M. Paul Foucher / MUSIQUE DE M. DIETCH, / DÉCORATIONS DE MM. PHILASTRE ET ­CAMBON / REPRÉSENTÉ POUR LA PREMIÈRE FOIS, A PARIS, SUR LE THÉÂTRE DE L’ACADÉMIE ROYALE DE MUSIQUE, / LE 9 NOVEMBRE 1842. / PARIS. / ­M ARCHANT, ÉDITEUR DU MAGASIN THÉATRAL, / BOULEVART SAINT-MARTIN, 12. / - / 1842. 107 Le Vaisseau / Fantôme / Opéra en 2 Actes / Paroles / de Mr. Paul Foucher / Musique / de Mr. Dietsch / Représenté pour la 1iere fois sur le théâtre de l’Académie / Royale de Musique, le Mercredi 9 novembre / 1842 (F-Po A.537.i–ii). 108 See the entries in Meyerbeer’s diary for October 1839 and January 1840 (Heinz, and Gudrun Becker (eds.), Giacomo Meyerbeer: Briefwechsel und Tagebücher, 5 vols to date (Berlin: De Gruyter, 1960-), 3:201 and 205).

98  Grand opéra – petit opéra amoureux, La révolte au sérail and La Gipsy.109 But putting a petit opéra on such a serious subject together with ballet-pantomime led to censure from all sides. For the Commission Spéciale des Théâtres Royaux, there was clear mismatch between the presentation of the work and the requirement for Pillet to maintain the Opéra ‘in the state of luxury that distinguishes it from others, in terms of the richness of scenery and costumes’.110 But for the press, the friction between form and content was the opera’s undoing: the attempt to compress serious action into the confines of petit opéra that had previously been characterised by its comic plots had failed. Two quotations from among many contemporary critics speak eloquently: It is difficult to maintain a complete and striking interest when one is obliged to contain a plot within narrow limits. This is why the Opéra ought to reserve the serious, dramatic and passionate genre for works in four or five acts, and why it should only permit, for works in one or two acts, such light subjects and gracious fantasies as Le philtre and Le comte Ory.111 and We can blame only the management of the Académie Royale de Musique for welcoming serious opera in two acts. It is to offer a trap to librettists and composers: constrained by narrow dimensions, they are obliged to strangle their plot and their score. These operas, of the same color as large works, do not make a contrast between themselves; furthermore, deprived of the luxury and development that may be found in five acts, they cannot bear comparison with their large siblings, who necessarily suffocate them…. The success obtained by Le comte Ory, Le philtre and even La xacarilla ought to have already indicated the route to follow.112 109 La tarentule (music by Gide, scenario by Scribe and choreography by Coralli) was premiered on 24 June 1839, La jolie fille de Gand (music by Adam, scenario by Saint-Georges and choreography by Albert Decombe) on 22 June 1842, and La révolte au sérail (music by Théodore Labarre, choreography by Taglioni) on 4 December 1833). See the complete listing in Bloom, ‘Fortunes of the Flying Dutchman’, 61. 110 ‘son théâtre dans l’état de luxe qui le distingue des autres, sous le rapport de la richesse des décorations et des costumes’ (F-Pan F21 4156). 111 ‘Il est difficile d’arriver à un intérêt saisissant et complet quand on est obligé de renfermer une action dans des limites étroites. Voilà pourquoi l’Opéra devrait réserver le genre sérieux, dramatique et passionné pour les ouvrages en quatre ou cinq actes, et n’admettre pour les pièces en un ou deux actes que de sujets légers et de gracieuses fantaisies comme Le philtre et Le comte Ory’ (Eugène Guinot in Le courrier français, 14 November 1842). 112 ‘Nous ne pouvons que blâmer la direction de l’Académie royale de Musique d’accueillir des opéras sérieux en deux actes. C’est tendre un piège aux auteurs et aux musiciens: restreints dans des dimensions étroites, ils sont obligés d’étrangler leur sujet et leur partition. Ces opéras, de même couleur que les grands ouvrages, ne font pas contraste entre

Grand opéra – petit opéra  99 The message was simple and fully understood by the Opéra. Le vaisseau fantôme and its 1842 companion, Le guérillero, stand alone as the only petits opéras based on serious subjects (and Le guérillero was significantly less serious than Le vaisseau fantôme) and commanded little or no support. It is at least possible that the idea of pairing a serious subject with petit opéra may have been prompted by the success of Le freyschütz as a surrogate petit opéra just before and during this period. The fact that the plots of other foreign operas that functioned as surrogates for petits opéras (Lucia di Lammermoor, Luisa Miller) were similarly serious points to key distinctions between petit opéra and foreign substitutes. However, the immediate successor to Le vaisseau fantôme and Le guérillero, Halévy and Saint-Georges’ Le lazzarone of 1844, returned to the comic register of original petit opéra and of those works identified by the press as the ‘route to follow’. Une sorte de vermout artistique Almost simultaneously with the appearance of the last petit opéra before the abolition of the licensing system, Massé’s La mule de Pedro, Le comte Ory was given a new production that prompted a good deal of retrospective reflection on the question of petit opéra and ballet-pantomime.113 A new chance to hear Le comte Ory reminded critics of the difficulty of s­ ustaining the compositional tradition of writing petit opéra.114 By the early 1860s, critical attitudes to the process of morcellement had hardened considerably, and this gave an added piquancy to the commentary. Whereas in the 1830s, ­using individual acts of a larger opera to accompany a ballet was as acceptable to most critics and composers as using a new petit opéra, by the 1860s, both voices were succumbing to the aesthetic demands of the ‘work’: petit opéra was an acceptable way forward, morcellement an aberration. When ­Johannès Weber commented on Alphonse Royer’s attempts to commission new petit opéra at the end of the 1850s, he prefaced his comments with a witty but punitive diatribe against morcellement. And when he turned to the discussion of the two works that were the products of Royer’s ideas, Alary’s La voix humaine and Massé’s La mule de Pedro, it was not the nature of petit opéra he found wanting but the quality of the works themselves. Weber’s comments are an important set of clues to the ways in which petit opéra was regarded at the end of the period under discussion:

eux; de plus, privés de la pompe et des développements que l’on trouve dans les cinq actes, ils ne peuvent soutenir la comparaison avec leurs grands confrères, qui nécessairement les écrasent …. Le succès obtenu par Le comte Ory, Le philtre et même La xacarilla aurait dû déjà indiquer la route à suivre’ (A. L. in La Patrie, 14 November 1842). 113 10 May 1863. 114 New works written for the recently emerged Théâtre-Lyrique, characterised by their use of spoken dialogue, ally themselves more closely with contemporary opéra comique than with petit opéra notwithstanding the shared comic register of their libretti.

100  Grand opéra – petit opéra Very recently, M. Royer appeared to have taken up the idea of having some works written that could precede a ballet, but he has executed this idea rather badly. He began by giving us La voix humaine. On what, then, was the choice of M. Alary based? On Le tre nozze, on La beauté du diable or on L’orgue de Barbarie? As for La mule de Pedro, M. Massé had previously produced a good number of works at other opera houses, and one would have had no reason to submit him to a new trial in order to know that his talent was better suited to the Opéra-Comique … What we demand is that composers who show some aptitude for grand opéra be tested, and that they prepare themselves to write, if necessary, a work of larger dimensions later.115 There had been no new petits opéras premiered since the trilogy of Pantagruel, La rose de Florence and François Villon in the mid-1850s, and none of these had proved successful. Weber’s complaints against Alary and Massé, when set in the context of the entire history of petit opéra, run counter to experience. Whereas it had been the case that almost all composers of petit opéra had not already written for the Opéra but had experience in some sort of lighter work (most obviously at the Opéra-Comique), for Weber this was now seen as a reason to condemn. Massé’s experience at the Opéra-­ Comique (and Théâtre Lyrique, although Weber does not mention that) is now a reason not to commission a work for the Opéra, and Weber seems not to feel the need to justify his contempt for Alary’s earlier works. But Weber and Alary agreed on one of the fundamental aesthetic traits of petit opéra towards the end of the Second Empire. Alary had been approached by Roqueplan as long ago as 1854 to write a work in two acts for the Opéra. Alary’s response shows that he considered works in two acts ‘merely curtain-raisers’ and that they were destined for composers’ ‘first attempts’; pointing out the details of his multifaceted career, he therefore asked permission to write in three acts, a request that came to nothing.116 And this potential difficulty had been clarified a decade earlier when Le 115 ‘Tout récemment, M. Royer paraissait avoir repris l’idée de faire écrire quelques ouvrages pouvant précéder un ballet; mais cette idée, il l’a assez mal exécutée. Il a commencé par nous donner la Voix humaine. Sur quoi donc se fondait le choix qu’il a fait de M. Alary? Sur le Tre nozze [Théâtre-Italien, 1830], sur la Beauté du diable [Opéra-Comique, 1861] ou sur L’orgue de Barbarie [Bouffes-Parisiens, 1856]? Quant à la Mule de Pedro, M. Massé avait donné précédemment bon nombre d’ouvrages à d’autres théâtres, et l’on n’aurait eu nul besoin de le soumettre à un nouvel essai, pour savoir que son talent est plus spécialement apte à l’Opéra-Comique…. Ce que nous réclamons, c’est qu’on mette à l’épreuve des compositeurs qui montrent quelque capacité pour le style du grand opéra, et qui se prépareraient ainsi à écrire plus tard, s’il y a lieu, une de plus vastes dimensions’ (Johannès Weber in Le temps, 4 June 1863). 116 ‘You know, Monsieur le Comte, that two acts are only a curtain raiser, destined for a first attempt. Now, I am no longer in this situation, since I have already produced many works [which are listed]’ (‘Vous savez, Monsieur le Comte, que 2 actes ne sont qu’un lever de rideau, dessiné toujours pour un premier essai. Or, je ne suis plus dans cette situation,

Grand opéra – petit opéra  101 guérillero had been premiered the same night as the premiere of La jolie fille de Gand. The Minister of the Interior wrote to Monnais, in his capacity as Commissaire des théâtres royaux, to draw his attention to the difficulties faced by the simultaneous premieres of two works, one of which was destined to support the other: The simultaneity of two new works must necessarily be a hindrance, each of which trying the attention of the public; furthermore, it has the disadvantage of making them think that one or the other, or both, as works of a mediocre importance.117 More explicit insight into the aesthetics of petit opéra also comes from the premiere of Halévy’s Le lazzarone of 1844. When the President of the Commission des théâtres, Duc de Coigny, reported to the Minster of the Interior, he pointed to a number of the work’s shortcomings that reveal institutional desiderata for the genre that are important to its understanding. Central to Coigny’s account was the opposition between on the one hand gaîté and intérêt and on the other simplicité vulgaire; Le lazzarone had sacrificed the former to the latter, in the Commission’s view.118 And while recognising that pompe et éclat were not called for by the genre, it stressed the importance of the Opéra’s more general mission (shared with ballet and grand opéra) ‘to offer to the ear and the eye the most délicat and the most élévé that art could offer all together’.119 The finale of the first act (the baptism of the harbour bell) came under close scrutiny in which the Commission looked for ‘something more soigné, better characterised, more picturesque’.120 Petit opéra therefore had to maintain a difficult equilibrium: achieving a tone that – while clearly comic – did not trespass on the vulgarity that might be expected at, for example, the Théâtre de Vaudeville or the Gaîté. Furthermore, the genre had to achieve that balance at an institution where puisque j’ai produit déjà plusieurs œuvres’). See the letter from Alary to Comte Bacciochi, surintendant de la musique, 1 October 1854 (F-Pan AJ13 221). 117 ‘La simultanéité de deux ouvrages nouveaux doit nécessairement nuire, chacun d’eux en fatiguant l’attention du public; elle a de plus l’inconvénient de les faire considérer l’un ou l’autre et peut-être l’un et l’autre, comme des œuvres d’une médiocre importance.’ Letter from Ministre, secrétaire de l’État de l’Intérieur [Duchâtel] to Commissaire de Roi [­Monnais], [19 June 1842] (F-Pan AJ13 183). 118 ‘The plot of Le lazzarone is made up of elements whose vulgar simplicity is capable of producing neither gaîté [which has both the sense of mirthfulness and good humor] nor interest’ (‘l’action du Lazzarone se compose d’éléments dont la simplicité vulgaire n’est susceptible de produire ni gaîté ni intérêt’). Report from the President of the Commission des théâtres (Duc de Coigny) to Minister of the Interior (Duchâtel), 1 May 1844 (F-Pan AJ13 183). 119 ‘d’offrir à l’oreille et aux yeux ce que l’art peut donner tout ensemble de plus délicat et de plus élevé’ (ibidem). 120 ‘quelque chose de plus soigné, de mieux caractérisé, de plus pittoresque’ (ibidem).

102  Grand opéra – petit opéra other musico-choreographic types could call forth massive scenic and other material resources that gave room for dramatic experiment from which petit opéra could not profit; such comic subject matter did not call for use of immense material resources, but the mechanisms of surveillance still required the presentation of the ‘elegance and distinction’ that were expected at the Opéra. For Weber and his contemporaries at the height of the Second Empire, the concern was no longer to produce petit opéra that worked well with ­ballet-pantomime; it was as much ‘to discover this phoenix that is becoming rarer and rarer: a French composer capable of writing a grand opéra worthy of taking its place alongside La muette de Portici and of La Juive’; the genre could offer a degree of preparation for larger works – grand opéra on the Meyerbeerian model.121 Weber’s view of petit opéra was that ‘one or two acts of an opera are for the public a sort or artistic vermouth, which helps them taste the charms of the corps de ballet better’.122 Given that Weber had been Meyerbeer’s secretary from 1845 to 1855, during which time he had witnessed at close quarters the premiere of Le prophète, much work on ­L’africaine and the explosion of dimensions at the Opéra-Comique with L’étoile du nord, this concern was hardly surprising and had been a recurrent issue in Parisian musical circles for a decade. In Weber’s commentary can be seen the beginning of the end for such works as Le comte Ory, Le philtre and those others that had followed them. After the proclamation of the ‘liberté des théâtres’ in 1864 – however disappointing its outcome – smaller-scale music drama at the Opéra became more diffuse and the clearly chiselled distinction between petit opéra and grand opéra less easy to identify. To peer over the precipice of the 1864 legislation and the abolition of the licensing system is to witness such works as Ernest-Henri-­Alexandre Boulanger’s Le docteur magnus (1864)123 and ­Jules-Laurent-Anicharsis Duprato’s La fiancée de Corinthe (1867) – so

121 ‘découvrir ce phénix de plus en plus rare: un compositeur français, capable d’écrire un grand opéra, qui fût digne de prendre rang à coté de la Muette et de la Juive’ (Johannès Weber in Le temps, 4 June 1863). 122 ‘un ou deux actes d’opéra sont pour le public une sorte de vermout artistique, qui lui fait mieux goûter les charmes du corps de ballet’ (ibidem). 123 Le docteur Magnus consisted of a single act in 11 scenes with seven numbers only. See the libretto, LE DOCTEUR / MAGNUS / OPÉRA EN UN ACTE / PAR / MM. EUGÈNE CORMON ET MICHEL CARRÉ / MUSIQUE DE M. ERNEST BOULANGER  / Représenté, pour la première fois, à Paris, à l’Académie Impériale de / Musique, le 9 mars 1864 / PARIS / MICHEL LÉVY FRÈRES, LIBRAIRES ÉDITEURS / 1864. The score was not published; see the manuscript full score: No 72 / Le Docteur / Magnus / = / Opéra en un acte de Messieurs Cormon / et Michel Carré; Musique de M E. Boulanger / Représenté pour la 1ere fois sur le Théâtre Impérial / de l’Opéra le mercredi 9 mars / 1864 (F-Po A. 613.i.-ii and ibidem, Rés. 31). While the work was performed with a ballet, it opened up the possibility, explored on 13 May 1864 of an evening consisting of two ballets and an opera (the two ballets were Giselle and Le marché des Innocents).

Grand opéra – petit opéra  103 different to the petits opéras that preceded them – and the entirely new world of small-scale works at the Opéra that they were able to open up.124 *** By the time of the catastrophe of Sedan, free-standing ballets d’action or ballets pantomimes had been a feature of the Opéra for a century. While the emergence of five-act grand opéra around 1830 set French stage music on a course that would last until the First World War, it created serious problems in terms of the relationship between opéra and ballet-pantomime, since no longer could the former be programmed alongside the latter in the same evening. Ad hoc solutions were to maintain old works in the repertory longer than their popularity might allow, to reduce larger operas to manageable proportions and to use individual acts or pairs of acts from longer works. The most creative solution was the emergence – just as the problem began to arise in the late 1820s – of an operatic genre, that would become formalised over the next half-century, specifically destined to accompany ballet: petit opéra. Grand opéra was the product of a network of like-minded librettists, composers, managers, choreographers and scene-designers who were working together within the already-established generic limits of the Opéra. Petit opéra, by contrast, was the result of two forces external to the genre: the state’s requirement that the institution mount productions of both opéra and ballet and the purely aesthetic development of another genre: grand opéra itself. Taken together, the two forces created a repertorial vacuum that could be filled by a number of ad hoc techniques: the use of older and small works, morcellement and the reduction in size of larger works. But the most striking consequence of the conjunction of the aesthetic forces creating grand opéra and the institutional pressures of the state was the creation of petit opéra. Born out a serious of fortunate coincidences emerging from the generic hothouse of the period around 1830, petit opéra remained a feature on the landscape of operatic history for most of the nineteenth century.

124 La fiancée de Corinthe consisted of six scenes and six numbers (LA FIANCÉE / DE CORINTHE / OPÉRA EN UN ACTE / PAROLES / CAMILE DU LOCLE / MUSIQUE DE / J. DUPRATO / REPRÉSENTÉ / Pour la première fois, à Paris, sur le théâtre impérial de l’Opéra, / le 21 octobre 1867 / PARIS / LIBRAIRE DRAMATIQUE / 10, RUE DE LA BOURSE. 10; the published score is ACADÉMIE IMPÉRIALE DE MUSIQUE / LA FIANCÉE / DE / CORINTHE / OPÉRA EN UN ACTE / PAROLES DE CAMILLE DU LOCLE / MUSIQUE DE / J. DUPRATO / … / PARIS / AU MÉNESTREL, 2 BIS , RUE VIVIENNE. It was performed exclusively alongside the ballet Le corsaire (music by Adam, scenario by Saint-Georges and choreography by Mazilier).

3 Jacques Offenbach The music of the past and the image of the present

The Théâtre des Bouffes-Parisiens opened in 1855 under less than ideal circumstances for its composer-manager.1 Jacques Offenbach was restricted by his licence to the performance of music drama in a single act with no more than four singing characters on stage.2 Furthermore, another theatre with an almost identical licence, Hervé’s Théâtre des Folies-Nouvelles, could have been a direct competitor for the same audiences, composers and librettist as the Bouffes-Parisiens, despite the fact that such competition was what the licensing system itself had been designed to avoid.3 And finally, the artistic space between the opérette that Offenbach was permitted to produce and the works of the subsidised theatres – the Académie Impériale de Musique, Opéra-Comique and Théâtre-Italien – had largely been filled five years previously by the Théâtre-Lyrique.4 However, Offenbach held the lease on two theatres, one on the Champs-Elysées and one – sufficiently far from the boulevard home of the Folies-Nouvelles to reduce direct c­ ompetition – in the Passage Choiseul; he also enjoyed the support of important state officials connected to the aristocracy and the imperial family and of a bourgeois audience developed during his time as a salon musician. 1 For an introduction to the early years of the Théâtre des Bouffes-Parisiens, see J­ ean-Claude Yon, ‘La création du Théâtre des Bouffes-Parisiens (1855–1862) ou La difficile naissance de l’opérette’, Revue d’histoire moderne et contemporaine 39 (1992), 575–600, see also idem, Jacques Offenbach (Paris: Gallimard, 2000), 128–165. For the repertory of, and other material concerning, the Bouffes-Parisiens, see Annie Ledout, ‘Le théâtre des Bouffes-­ Parisiens, historique et programmes, 1855–1880’ (PhD diss., Université de Paris IV, 2001). 2 The earliest version of Offenbach’s licence restricted him to three singing characters (Paris, Archives Nationales (hereafter F-Pan), F21 1136, 4 June 1855; see Yon, Jacques ­Offenbach, 137); the loosening of this limitation to four was accomplished in the revision of Offenbach’s contract, 22 October 1855 (Nicole Wild, Dictionnaire des théâtres parisiens au xixe siècle: les théâtres et la musique (Paris: Amateurs des Livres, 1989), 63). 3 For Florimond Ronger (Hervé) and the Théâtre des Folies-Nouvelles, see Eugène Woestyn and Eugène Moreau, Les Folies-Nouvelles, Les théâtres de Paris (Paris: Martinon, [1855]) and Louis-Henry Lecomte, Les Folies-Nouvelles, Histoire des théâtres de Paris 4 (Paris: Daragon, 1909). 4 Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981).

Jacques Offenbach  105 Offenbach’s artistic and professional aims were clear: he wanted to develop the Bouffes-Parisiens into a theatre with a radically different mission and profile to that of the Folies-Nouvelles. He could do this by developing the bourgeois rather than the popular elements of his audience – easy at the Salle Choiseul and the Salle Lacaze – and by setting himself two more complex objectives: to support young composers, especially Prix de Rome laureates returning to the capital, and to position the Bouffes-Parisiens and its emergent genre of opérette squarely within the tradition of international comic opera of the past and specifically within that of eighteenth-century opéra comique. His first objective tapped into a tradition of aspirant opera managers claiming that their new enterprises would support young composers that dated back at least to the beginning of the licensing system in 1806–1807: it was a claim well known to anyone who remembered the attempts to promote music drama at the Odéon in the 1820s or at the Théâtre de la Renaissance in the 1830s.5 Offenbach’s second objective – to return to the music of the past – was an ambitious and idiosyncratic undertaking that had a significant impact on the success of the Théâtre des Bouffes-Parisiens in its first phase. It forms the subject of the present inquiry.6

Sources for the music of the past Between 1850 and 1860, Offenbach engaged with earlier music in a variety of ways. As conductor at the Comédie-Française, he was involved in A ­ rsène Houssaye’s attempts to revive Molière in ways that might have been familiar to the seventeenth century by removing two centuries of musical tradition.7 In his articles published in the journal L’artiste in the early part of 1855, and in the announcement of his competition for a new opéra bouffe the following year, he exploited the music of the past for his own ends. In addition to attempting to mount productions of largely eighteenth-century comic opera 5 See Mark Everist, Music Drama at the Paris Odéon, 1824–1828 (Berkeley, Los Angeles and London: University of California Press, 2002), 210–211 and idem, ‘Theatres of Litigation: Stage Music at the Théâtre de la Renaissance, 1838–1840’, Cambridge Opera Journal 16 (2004), 136. 6 Despite its title, Gerard Loubinoux, ‘Le chercheur d’esprit, ou Offenbach et la mémoire du xviiie’, Retour au xviiie siècle, eds. Roland Morier and Hervé Hasquin, Études sur le xviiie siècle 22 (Brussels: Éditions de l’Université de Bruxelles, 1994), 63–76, addresses none of the sources or repertories discussed in this article. It reads those libretti of Offenbach’s opérettes that depend on eighteenth-century settings: Le chanson de Fortunio (1861) and La Foire Saint-Laurent (1877), Mesdames de la Halle (1858; but wrongly assigned by Loubinoux to 1868), Madame Favart (1878) and attributes the interest in the eighteenth century to Offenbach alone rather than to his librettists. 7 Yon, Jacques Offenbach, 101, 107, and 111. In his bid for the management of the Théâtre-­ Lyrique in 1854, however, Offenbach made no mention of any interest in the music of the seventeenth or eighteenth centuries (pace Wild, Dictionnaire, 238–239) nor in the fate of Prix de Rome laureates. See the letter from Offenbach to Camille Doucet [Directeur de l’administration des théâtres], 3 July 1854, F-Pan F21 1120/2.

106  Jacques Offenbach (with variable degrees of success), he used composers from the past to pillory aspects of contemporary musical culture in his Le carnaval des revues of 1860. Offenbach wrote four causeries musicales in the journal L’artiste in the six months before the Théâtre des Bouffes-Parisiens opened in the Salle Lacaze.8 One of the tasks they accomplished was to appease those whom Offenbach thought key players in the world of stage music in the mid-1850s. Accordingly, revivals of Halévy’s La Juive and Auber’s La muette de Portici are praised without reserve, as is the Parisian première of Verdi’s Il trovatore;9 Émile Perrin’s joint direction of the Opéra-Comique and Théâtre-­ Lyrique comes in for approving comments, and Offenbach throws his weight behind the contemporary vogue for the works of Adolphe Adam.10 Berlioz is praised for his L’enfance du Christ, offered sympathy for being beaten by Louis Clapisson in the competition for the most recent musical nomination to the Institut and praised vicariously by allusions to the shortcomings of the recent revival of Der Freischütz – in Castil-Blaze’s 1824 version as Robin des bois – at the Théâtre-Lyrique.11 Such attempts to curry favour need to be read alongside the production of such works as the Décaméron dramatique: album du Théâtre-Français, published in October 1854, in which each of the major actresses at the Comédie-Française was presented with a dance for piano composed by Offenbach, a portrait by Hermann Raunheim and a quatrain by a noted poet of the day: Gautier, Dumas, Musset and others.12 The articles in L’artiste also undertook significant cultural work by praising smaller-scale contemporary opéras comiques in terms that stressed their similarities to eighteenth-century classics of the genre. In doing so, they contributed to a view of contemporary opéra comique that set it apart from more ambitious contemporary works: the eighteenth-century was made to contribute to Offenbach’s aesthetic position as he tried to promote his own artistic programme. A year after the publication of the articles in L’artiste, and during the first year of the Bouffes-Parisiens’ operations, Offenbach advertised a ‘Concours pour une opérette en un acte’ for the composition of an opéra bouffe aimed at composers who had not been performed on any Parisian stage.13 As part of the announcement for the ‘Concours’, Offenbach published what amounted 8 L’artiste, 14 January 1855, 4 February 1855, 25 February 1855, 25 March 1855. 9 Halévy: ibidem, 25 March 1855, 178, Auber: ibidem, 14 January 1855, 39, Verdi: ibidem, 39–40. 10 Perrin: ibidem, 4 February 1855, 78, and 25 March 1855, 178, Adam: 14 January 1855, 40. 11 Ibidem, 14 January 1855, 40, 4 February 1855, 78. 12 Yon, Jacques Offenbach, 131–132. 13 Jacques Offenbach, ‘Concours pour une opérette en un acte’, Revue et Gazette musicale de Paris, 20 July 1856, 230–231. The advertisement was also printed in Le ménestrel, 27 July 1856, with minimal introduction and the articles only in La France musicale, 20 July 1856. Subsequent reference will be made to the version published in the Revue et Gazette musicale de Paris.

Jacques Offenbach  107 to a history of opéra comique that was slanted so as to throw into relief his own activities at the Bouffes-Parisiens. The announcement constituted not only a partial view of the history of opéra comique up to the present but also a manifesto for the artistic ambitions of the Bouffes-Parisiens, its manager and principal composer. Supported by a series of carefully placed allusions to the music of the past, Offenbach could cast about for works that fell within the limitations of his licence, could amplify the repertory of his theatre and could contribute to its cultural capital. He identified four comic operas from the p ­ revious ­c entury – Rousseau’s Le devin du village, Pergolesi’s La serva padrona, ­Mozart’s Der Schauspieldirektor and Rossini’s Il signor Bruschino – that might be of use to the Bouffes-Parisiens.14 Of these, Le devin du village and La serva padrona fell foul of the controls placed on the theatre’s licence and were never performed (although the Rousseau may well have got as far as rehearsal); Der Schauspieldirektor and Il signor Bruschino, however, served as important parts of the repertory and key aesthetic statements for Offenbach and the Bouffes-Parisiens in the second half of the 1850s. If putting words into the mouths of dead composers was one of the tricks that Offenbach was using in his manifesto on opéra comique that accompanied the advertisement for the ‘Concours’, it was a short step to having composers of the past speak on stage. Offenbach, together with his collaborators Eugène Grangé and Philippe Gille, took his chance in his Le carnaval des revues of 1860, a review of the previous year’s events.15 One of the main musical targets was the series of concerts that Wagner had mounted at the Théâtre-Italien – feet from Offenbach’s own Salle Choiseul in the Salle Ventadour – but there were three other targets: the inquiry into pitch that had reported during 1859, Meyerbeer’s continuing success and – close to Offenbach’s heart – the question of the fate of young composers. The sixth tableau of Le carnaval des revues begins in the musical corner of the Elysian Fields, a neat classical allusion to Orphée aux enfers that would have been missed by no one.16 Grétry enters humming the aria ‘Et zig et zog’ from one of the works Offenbach had endorsed four years earlier in his articles in

14 The inclusion of Rossini’s Il signor Bruschino (1813) raises the issue of what repertories Offenbach considered appropriate to his definition of the past. The concept of the ‘long eighteenth century’ does not perhaps allow a sufficiently finely-textured view of Offenbach’s position, conditioned as it was by his view of the history of opéra comique (for which see below). Although Rossini’s early farse clearly fall into this category – and perhaps ­Weber’s too – the music of the 1820s is much more closely contested, as the discussion of both opéra comique and opera semiseria in the ‘Concours’ article suggests (ibidem, 230, 1:5–8 [reference to the ‘Concours’ article is via page, section:paragraph number as published in the Revue et Gazette musicale de Paris]). 15 Eugène Grangé and Philippe Gilles, Le carnaval des revues: revue de carnaval en 2 actes et 9 tableaux; Les souper de mardi-gras, prologue, Paris, Bouffes-parisiens, le 10 février 1860 … Musique de Jacques Offenbach ([Paris]: Michel Lévy frères, [1860]). 16 Ibidem, 14–17.

108  Jacques Offenbach L’artiste (see below), Richard, cœur de lion; from the other side of the stage enters Gluck, singing ‘J’ai perdu mon Eurydice’. The resulting banter encompasses the revival of Gluck’s Orphée et Eurydice at the Théâtre-Lyrique and the contralto who took the title role (Pauline Viardot); the two then propose a game of dominos. Before they can start, Mozart enters, and Grétry and Gluck greet him by singing ‘Mon cœur soupier’ – the French version of ‘Voi che sapete’. Before they can start their three-way game of dominos, Weber arrives, and the other three welcome him with a performance of ‘Chasseur diligent’, the French version of the Huntsmen’s chorus from Der Freischütz. This quartet of posthumous composers – Grétry, Gluck, Mozart and Weber – serves as the commenting chorus against which the objects of Offenbach’s satire are projected. Apart from the attention given to the compositeur de l’avenir in Le carnaval des revues – almost too easy a target – Offenbach’s most important victims were Meyerbeer and the quartet of commenting composers themselves. All served as important ways of promoting Offenbach’s own aesthetic agenda, especially the support to young composers and the appeal to the musical and dramatic values of the past, exactly those elements that he was trying to develop at his own institution. Offenbach’s quartet of composers served here as a fixed point of reference to whom he could have recourse in his constant invocation of earlier values. Their posthumous repudiation of a particular feature of the present functioned as a pointer – as much as Offenbach’s explicit statements in the ‘Concours’ article – to the values of eighteenth-­ century opéra comique that, by 1860, Offenbach would have claimed were well and truly re-established in Paris on the boards of his own theatre.

The French tradition Offenbach’s competition to encourage recent laureates of the Prix de Rome was launched in July 1856; at face value it sought to develop his mission to help young composers, which he had set forth in his requests to open a theatre the previous year.17 To have chosen Georges Bizet and Charles Lecocq as winners can be, with hindsight, considered a successful outcome. But when the ‘Concours’ was advertised, Offenbach gave it a context that was no less than the entire history of opéra comique; he took advantage of the opportunity to produce a manifesto for the aesthetic project hosted by his own theatre and the music drama supported there. He set the competition in the context of a view of contemporary opéra comique that valued most the sort of music drama that he himself was allowed and able to play at 17 There is a tension in all of Offenbach’s comments on young composers between Prix de Rome laureates and those novice composers trained outside the Conservatoire, and therefore not eligible to enter the state competition. These included Offenbach himself, not to mention such competitors as Clapisson or Gevaërt. Such a tension is clear in the ‘Concours’ text itself (Revue et Gazette musicale de Paris, 20 July 1856: 231, 3:1–3 and 7).

Jacques Offenbach  109 the Bouffes-Parisiens: works in one act with a limited number of soloists; conversely, he was suspicious and critical of both the serious elements that had infiltrated opéra comique at least from the 1780s and of its more recent growth. Offenbach’s view of the history of opéra comique was neatly encapsulated as he tried to sum up towards the end of his introduction to the ‘Concours’: One may easily follow the progress of opéra comique from its origins to the present. At first, a little brook with limpid water, with new banks it develops little by little as it advances, until it becomes what we see today, a wide river, with imposing waves on its vast surface.18 His metaphor draws an unbroken teleological line from Philidor’s Blaise, le sauvetier of 1759 to Meyerbeer’s L’étoile du nord of 1854. Offenbach o ­ utlined three phases in the history of the genre. Phase one ran from ­Philidor – or from Pergolesi and Duni, since Offenbach was anxious to stress the Italian origins of the genre – to Dalayrac and Grétry. But Offenbach proposed a clear distinction between pre- and post-revolutionary works and pointed to a sudden enlargement of the genre after 1789 that was only accomplished ‘en se dénaturant’;19 he saw this as the product of the influence of ideas of ‘­political and artistic renovation’. While this gave a neat point of articulation between his first and second phases, it also gave Offenbach the chance to point to works by Dalayrac and Grétry – ‘the two most illustrious representatives of opéra comique’20 – and to accuse them of abjuring the genre that had served them so well with the result that they had embarked on the composition of Camille, ou Le souterrain and of Pierre le grand and Guillaume Tell, respectively. The works of his first phase that Offenbach pointed to with real approval were Monsigny’s Le déserteur, Gossec’s La fête du village, three works by Dalayrac (Adolphe et Clara, Maison à vendre and Picaros et Diego) and Gretry’s Le tableau parlant, Zémire et Azor, L’amant jaloux and Richard, cœur de lion. The last of the four works by Grétry came in for special praise ‘applauded every day as if it were written yesterday, although it dates from 1785 [sic; recte 1784]’, and in Offenbach’s discourse on the eighteenth-century Richard, cœur de lion takes on the role of a signature work for its composer.21

18 ‘On peut suivre aisément la marche de l’opéra comique depuis son origine jusqu’à nos jours. D’abord petit ruisseau aux eaux limpides, au frais rivage, il s’étend peu à peu, à mesure qu’il avance, jusqu’à devenir, ce que nous le voyons aujourd’hui, un large fleuve, roulant dans son vaste lit ses ondes imposantes’ (ibidem, 2:12). 19 Ibidem, 1:2. 20 ‘Les deux plus illustres représentants de l’opéra comique’ (ibidem, 1:2). 21 ‘que nous applaudissons tous les jours, comme s’il était d’hier, bien qu’il date de 1785’ (ibidem, 1:1).

110  Jacques Offenbach Throughout Offenbach’s appropriation of the eighteenth century during the 1850s, Grétry remained a key figure. In Le carnaval des revues, he is perhaps the central figure among the four composers who comment on the subjects brought before them, but five years before, Offenbach had also made extensive use of Grétry in his articles in L’artiste. In his review of Albert Grisar’s Le chien du jardinier at the Opéra-Comique, Offenbach opened up praise for contemporary composers via analogy with his central figure of the eighteenth century.22 Grisar is identified as ‘a real comic composer’, and in the work under discussion, Offenbach takes the opportunity to impute a love of Grétry to the composer. ‘In this recent score’, he writes, ‘there is often a reflection of the style of Grétry, much beloved of Grisar’.23 It is impossible to know what Offenbach’s evidence for Grisar’s love of Grétry might be, but that is barely the point. Offenbach takes the opportunity to praise those contemporary works that are easy for him to associate with his own project of self-identification with the music of the eighteenth century, and eliding Grisar with Grétry accomplishes that task very effectively. Grisar was to be included in those works in the third phase of the ‘Concours’ article that were in general terms subject to censure, but Offenbach uses analogy with Grétry to protect the work under review that, for him, stands apart from the rest of that phase in the genre’s history.24 In his double-edged appreciation of Victor Massé in the ‘Concours’ article, Grétry is used as a point of reference to praise the former’s Les noces de Jeanette – ‘a jewel – [that] proves the aptitude of the young composer for writing in the old and frank manner of Grétry’ – at the expense of Massé’s other works.25 Again, when savaging Clapisson in 1855, he uses Grétry as an exemplum of the best of the past (see later in this chapter). But even Grétry did not escape Offenbach’s wrath for having occasionally stepped outside the boundaries that he had set for opéra comique of the past. When Offenbach refers to Grétry, it needs to be remembered that his enthusiasm for such works as Le tableau parlant, Zémire et Azor, L’amant jaloux and Richard, cœur de lion needs to be read alongside his adverse comments on Pierre le grand and Guillaume Tell, which he thought betrayed the modishness of post-revolutionary composition.26

22 L’artiste, 4 February 1855: 79. 23 ‘un véritable compositeur bouffe…. Dans cette dernière partition, il y a souvent un reflet du style de Grétry, que Grisar affectionne, du reste, particulièrement’ (ibidem). 24 In his commentary on Grisar in the ‘Concours’ article, Offenbach praises, in addition to Le chien du jardinier, Les porcherons and Bonsoir, monsieur Pantalon!. With regard to the other works, he writes that Grisar there ‘preferred to become the rival of Italian masters’ (‘a préféré devenir le rival des maîtres italiens’, Revue et Gazette musicale de Paris, 20 July 1856, 231, 2:7). 25 ‘un bijou, - prouvent l’aptitude du jeune compositeur à écrire dans la vieille et franche manière de Grétry’ (ibidem, 2:8). 26 ‘Pour se faire pardonner ses premiers succès [this is ironic], … l’auteur du Tableau parlant composa Pierre le grand et Guillaume Tell’ (ibidem, 1:2).

Jacques Offenbach  111 Offenbach’s objections to the over-inflated nature of contemporary opéra comique were clarified by further comments on the genre which lead straight back to the previous century. A reworking of Charles-Simon Favart’s La chercheuse d’esprit by Jules Lecomte had just been mounted at the Théâtre du Vaudeville with music by Jean-Baptiste Montaubry that, for Offenbach, could have been played on the harpsichord of the time of Jean-Jacques [Rousseau], in the salons of Mme de Pompadour, decorated with pastels by Latour and Boucher. If Mme d’Épinay could be resuscitated, she would speak of it in her posthumous memoirs.27 Perhaps the most colourful of Offenbach’s causeries, this is a remarkable inclusion in its own right since music at the Vaudeville rarely came in for any comment, favourable or adverse, in his columns; its inclusion here betrayed the importance of the position its author was attempting to develop. It was a marginal event in the Parisian theatrical world of early 1855, but Offenbach’s commentary stresses exclusively its eighteenth-century credentials by allusion to the composer of Le devin du village, two of the key artists of the ancien régime, Maurice de la Tour and François Boucher, Mme de Pompadour and Mme d’Épinay. The contemporary nature of Lecomte’s and Montaubry’s arrangement holds no interest for Offenbach in this literary pastiche of the eighteenth century. Offenbach called the second phase of his history of opéra comique ‘the reign of the harmonistes’, those, he said, who exhibited ‘a sovereign contempt for la petite musique’ – a term that recalled the language (petites ouvrages, petites pièces) of his articles in L’artiste the previous year.28 Monarchs in this reign were Nicolo, Berton, Méhul, Catel, Boïeldieu, Lesueur, ­Cherubini, and the early works of both Hérold and Halévy. As he attempted to steer a course between avoiding the antagonisation of the musical establishment and promoting his own cause, his argument ran the risk of collapse. Having chastised the generation of his second phase, he then turned to his audience and claimed that these ‘masters … have no less created works of which the French stage can justly be proud’.29 Hérold, however, is held up as a figure of transition, with his Le pré aux clercs and Zampa, representing – again according to Offenbach – ‘the transition … between opéra comique 27 ‘pourrait se jouer sur les clavecins du temps de Jean Jacques, dans des salons Pompadour, constellés de pastels de Latour et de Boucher. Si madame d’Épinay pouvait ressusciter, elle en parlerait dans ses mémoires posthumes’ (L’artiste, 4 February 1855, 80). 28 ‘le règne des harmonistes …. un souverain mépris pour la petite musique’ (Revue et Gazette musicale de Paris, 20 July 1856, 230, 1:2). In his approving comments on the current repertory of one-act opéras comiques at the Théâtre-Lyrique and Opéra-Comique, Offenbach alludes to ‘les petites pieces’, ‘petits opéras comiques’ and ‘ce repertoiricule charmant et amusant’ (L’artiste, 25 March 1855, 178). 29 ‘maîtres … n’en ont pas moins créé des œuvres dont la scène française doit justement s’enorgueillir’ (Revue et Gazette musicale de Paris, 20 July 1856, 230, 1:4).

112  Jacques Offenbach of a light allure and musical drama of lugubrious effect’.30 At this moment, for ­Offenbach, the original genre of opéra comique disappeared in favour of larger works; this was not yet grand opéra but a mixed genre along the lines of Italian opera semi-seria and the German-language works that Offenbach thought derived from it: Weber’s Der Freischütz, Oberon and Euryanthe, Mozart’s Die Zauberflöte and Weigl’s Die Schweizerfamilie. Offenbach reserved his greatest censure for works of his third phase: those of Auber, Halévy, Thomas, Reber, some works of Massé and Grisar, Gevaërt and Meyerbeer. He considered that: The scores of many of our contemporary composers resemble elegant women on the boulevard: they wear too much crinoline. In the daylight, they constitute quite a substantial outfit, and beautifully coloured. Close up, en déshabillé, at the piano, they are phantoms inflated by wind and sound.31 The overuse of crinoline echoes one of Offenbach’s recurring views on contemporary music: that it is overblown beyond the appropriate confines of its genre, especially in the case of opéra comique, and it seems clear from his refusal to extend this particular polemic to works at the Académie Impériale de Musique that it is opéra comique at both the Théâtre-Lyrique and the Opéra-Comique itself that is his target. As has already been seen, Offenbach did his best not to antagonise those of whom he was critical. In his careful commentary on the works of Thomas, for example, even the pieces on which he was hardest – Le songe d’une nuit d’été and Raymond – he made sure that their composer knew that he thought them ‘œuvres magistrales’;32 Offenbach saw redemption for Thomas, however, in La double échelle and Mina. It is almost possible to smell the fragrance of the salons Offenbach was frequenting in the 1840s and early 1850s in this careful stroking of the culturally rich and powerful in his attempts to position his own works. Massé is praised on several occasions for his Les noces de Jeanette, while in his work on La chanteuse voilée and Galathée he was accused of trying to faire grand – an expression that almost defies translation but well expresses the activity that Offenbach found so reprehensible.33 With Meyerbeer’s recent L’étoile du nord, Offenbach came to the end of his history and the point at which opéra comique had reached

30 ‘une transition … entre l’opéra comique aux allures légères et le drame musicales aux ­lugubres effets’ (ibidem, 1:5). 31 ‘les partitions de beaucoup de nos compositeurs du jour rassemblent aux élégantes du boulevard, elles portent trop de crinoline. A la lumière, elles forment un ensemble assez substantiel et d’un beau coloris. De près, en déshabillé, au piano, ce sont des fantômes gonflés de vent et de son’ (L’artiste, 14 January 1855, 40). 32 Revue et Gazette musicale de Paris, 20 July 1856, 231, 2:5. 33 Ibidem, 2:8.

Jacques Offenbach  113 its lowest ebb. He set his criticism in the context of the stranglehold of the box office over any flexibility that an opera-house manager might wish to exercise. Émile Perrin, the manager of the Opéra-Comique (and until recently the manager of the Théâtre-Lyrique as well), is praised for mounting productions of Massé’s Les noces de Jeannette and Grisar’s Le chien du jardinier but is portrayed in his promotion of Meyerbeer’s L’étoile du nord as the victim of the preferences of his public, dragged towards productions at the Opéra-Comique of what Offenbach was certain was ‘un grand opéra tout à fait!’.34 Offenbach also pointed towards the libretti of such opéra comique as La dame blanche, Fra Diavolo, L’éclair, La fille du regiment and Les mousquetaires de la reine as literary works that existed independently of their music, but also towards the transformation of the libretto into something that had become indissolubly linked to its music in works of both the second and third phases of his history of the genre.35 In his very last causerie musicale, Offenbach points with approval to the vogue for what he calls petites pièces, and singles out such works as Ernest Boulanger’s Les sabots de la marquise, Prosper Pascal’s Le roman de la rose, Victor Massé’s Miss Fauvette and Ferdinand Poise’s Les charmeurs.36 Poise, of course, is one of the early composers recruited to Offenbach’s own ­theatre, and Massé forms an important link in the historical chain that ­Offenbach forges between himself and the eighteenth century in his ‘Concours’ article. Émile Perrin, the director of both institutions able to mount productions of opéra comique, is credited with the creation of this répertoiricule in the 25 March 1855 causerie musicale but also – in the 14 January article – with the productions of Hérold’s Le pré aux clercs and Massé’s Galathée (­Meyerbeer’s L’étoile du nord is mentioned in parenthesis). Now, these three works would come in for harsh criticism in the ‘Concours’ article: Le pré aux clercs, alongside Zampa, as a key work in the transition from Offenbach’s second to third phase in his history, and Galathée as an example of its composer attempting to stretch the aesthetic boundaries of opéra comique beyond what Offenbach claimed was acceptable.37 In his attitude to Perrin’s repertory at the Opéra-Comique and Théâtre-­ Lyrique, Offenbach’s position was hopelessly ambiguous: praising almost without reserve both the repertory that he would shortly be promoting as the future of opéra comique and those works that he would then be damning as its worst excesses.38 These ambiguities remain in his accounts of the works of Adam and Clapisson. Again praising Perrin, Offenbach points to productions of Adam’s Le muletier de Tolède, A Clichy, and the revival of 34 35 36 37 38

Ibidem, 2:17. Ibidem, 2:15. L’artiste, 25 March 1855: 178. Revue et Gazette musicale de Paris, 20 July 1856, 230, 1:5, and ibidem, 231, 2:8. Compare, for example, the two causeries musicales, L’artiste, 14 January 1855: 40 and ibidem, 25 March 1855: 178.

114  Jacques Offenbach La reine d’un jour. In each case, the work is attributed to ‘l’auteur du Chalet’, a conventional form of describing the composer, but Offenbach’s use of the convention three times in three lines drives home his implicit point that Adam – a composer at the peak of his career – was perhaps getting more than his fair share of the stage at the Théâtre-Lyrique; the point is made explicitly when Offenbach suggests that although he is very happy to hear Adam’s charmante musique, he would also like to hear a work by one of his own young colleagues, a category that would necessarily include Offenbach himself.39 Strangely absent from the account of contemporary opéra comique in the ‘Concours’ article is the work of Clapisson, who had come in for heavy criticism in the articles in L’artiste the previous year: first as the successful rival to Berlioz at the Institut and second as the composer of Dans les vignes, ­recently premiered at the Théâtre-Lyrique.40 Offenbach’s allegiance to ­Berlioz might well be explained by his hostility towards Clapisson. Much of Offenbach’s and Clapisson’s early careers were mirror images of each other in many respects; both were born outside France (Clapisson’s father had been the principal horn in the Teatro San Carlo in Naples), and both came to Paris as string players (Clapisson was a violinist first at the Théâtre-Italien and then at the Académie Royale de Musique). Both then turned to composition with opéra comique as their target, but Clapisson was significantly more successful during his first decade than Offenbach. Offenbach’s critique of Clapisson was partial and one-sided. He constantly referred to him as merely a composer of romances, and claimed in his commentary on Clapisson’s election to the Institut that it ‘had need of a symphonist, but it was a romancier [un faiseur de romances] who prevailed’.41 He went further and built a play on words around the title of one of Clapisson’s romances ‘Le postillon de Madame Ablou’, so that he could claim that both Clapisson and Adam were elected to the Institut because of ‘Le postillon’, elections that showed that the Académie was ‘a stickler [‘à cheval’] for the principles of art’.42 In his critique of Clapisson’s Dans les vignes, Offenbach accuses the composer of celebrating his nomination to the Institut by producing a few outdated melodies. He continues by suggesting that, in making such an observation, he is not reproaching the composer; but he piles on insult by excusing Clapisson because the melodies are not his own: ‘he is content to borrow them from the repertory of old French songs, such as “Le sultan

39 Ibidem, 14 January 1855: 40. 40 In the editorial commentary that preceded the copy of Offenbach’s advertisement for his ‘Concours’, the exclusion of Clapisson was explicitly recognised with a degree of censure (Revue et Gazette musicale de Paris, 20 July 1856: 230). 41 ‘On avait besoin d’un symphoniste, ce fut un faiseur de romances qui l’obtint’ (ibidem). 42 The full text of this play on words is ‘Aussi M. Clapisson a-t-il été nommé au même titre que M. Adam, c’est-à-dire pour cause de Postillon, ce qui prouve que l’Académie est à cheval sur ses principes d’art’ (ibidem).

Jacques Offenbach  115 43

Saladin”, “Le père Trinquefort”, etc.’ Offenbach here combines contempt for Clapisson with a subtle promotion of his own cause; accusations of simple borrowing from the past are coupled to a choice of works that is explicitly signalled: to invoke ‘Le sultan Saladin’ is yet again to memorialise Offenbach’s key work from the eighteenth century, Grétry’s Richard, cœur de lion, in which the crusader song ‘Le sultan Saladin’ is quoted at the end of the first act. Offenbach’s primary target in the history of opéra comique, Giacomo Meyerbeer, resurfaced four years later, when he figured as an object of satire in Le carnaval des revues. Offenbach’s immediate pretext for the introduction of Meyerbeer into a parody of the events of 1859 was the centenary of the birth of Friedrich von Schiller (born 10 November 1759), for which Meyerbeer had written a Festmarsch and a Festgesang to a text by Ludwig Pfau; both had been premiered in 1859. The fourth scene of Le carnaval des revues begins with the entry of the Master of Ceremonies – who speaks exclusively in a German fit only for the tourist industry and which is consistently misunderstood by Grétry – followed by an individual bearing a banner on which is written Fête de Schiller. Weber kindly translates and explains how the four composers are to be given a performance of Meyerbeer’s Festmarsch; Grétry retorts that he has known the march for a decade, declaring that ‘C’est la Marche du Prophète’, a convenient swipe at the generic nature of this Meyerbeerian form.44 The Master of Ceremonies then collapses into a panegyric to Meyerbeer with which he expects his audience to concur. As he names in chronological order each of Meyerbeer’s operas, all of which he describes as ‘sublime’, he waits for the four composers to agree. They readily concur with this description for Robert le diable and Les Huguenots but are progressively less and less enthusiastic as Le prophète, L’étoile du nord and finally Le pardon de Ploërmel (premiered in April 1859) are named. The composers’ response to the invitation to express enthusiasm for Le pardon de Ploërmel is so muted that it is entirely silent, and the Master of Ceremonies is obliged to repeat his invitation to comment; the response is the same. He takes his banner with him as he leaves, and the composers sing what is, to judge from the scansion of the text, probably a parody of the opening of the septuor du duel ‘En mon droit j’ai confiance’ from act iii of Les Huguenots to the following poetry: En Meyerbeer j’ai confiance; L’Africaine enfin paraitra, Et le succès lui reviendra!…. Oui!….

I have trust in Meyerbeer; L’africaine will finally appear, And success will return to him! Yes!

43 ‘Il s’est contenté de les emprunter au répertoire des vieux airs français, tels que le Sultan Saladin, le Père Trinquefort, etc.’ (ibidem). 4 4 Grangé and Gille, Le carnaval des revues, 16.

116  Jacques Offenbach Offenbach’s treatment of this scene echoed the criticism of the turn taken by opéra comique in the 1850s so clearly expressed in the 1856 ‘Concours’ article. There, Offenbach had identified Meyerbeer’s L’étoile du nord – then enjoying its first run after its 1854 premiere – as the nadir of the recent tendency to compose opéra comique that tended towards the type of work more suitable for the Académie Impériale de Musique. By 1860, and the recent premiere of Meyerbeer’s Le pardon de Ploërmel, not only was the trend still visible but the same composer was responsible for perhaps the most high-profile works in that genre. The alignment of Meyerbeer with Schiller in this scene had the effect of placing the composer in the same national frame as Wagner and reminding the audience that – despite his triumphs in French stage music – he was as much a Prussian as Wagner was a Saxon. But in turn, that threw the spotlight onto the four composers who serve as the commenting chorus in this tableau, none of whom were born in France. In the articles in L’artiste, in the ‘Concours’ text, and even in his commentary on the music of the past in Le carnaval des revues, Offenbach subtly rewrote the history of opéra comique in a way that threw his own work into relief. He could do this partly because there was almost no competing view of the historiography of opéra comique in the middle of the 1850s. There was plenty of commentary on the status of the genre and its place in Parisian and provincial musical cultures, but in those texts where one might expect some clear view on the history of opéra comique, even if it be no more than the ‘origins and progress’ summary that would have been likely, such a commentary was missing. The two texts that one might have looked to in order to find some context or competition for Offenbach’s comments are remarkably reticent. The two histories by Blondeau (1847) and de la Fage (1844) are either confused in the first case or do not reach the modern period in the second.45 Fétis’ later history of music (1869–1876) stops well short of the beginnings of opéra comique.46 The only history of opéra comique from the 1850s was that of Castil-Blaze which remains unpublished to this day.47 With Castil-Blaze, we find a very different history of opéra comique, one that takes its historical credentials very seriously, and sees the origins of the genre in the Jeu de Robin et de Marion from the end of the thirteenth century. 45 [Pierre] Auguste L[ouis] Blondeau, Histoire de la musique moderne, depuis le premier siècle de l’ère chrétienne jusqu’à nos jours, 2 vols, (Paris: Tantenstein et Cordel, 1847), JusteAdrien Lenoir de La Fage, Histoire générale de la musique et de la danse, 2 vols and atlas (Paris: Comptoir des imprimeurs unis, 1844). 46 François-Joseph Fétis, Histoire générale de la musique depuis les temps les plus anciens jusqu’à nos jours, 5 vols (Paris: Firmin-Didot, 1869–1876). 47 François-Henri-Joseph Blaze [Castil-Blaze], Histoire de l’opéra comique (MS [1856], Paris, Biblothèque-Musée de l’Opéra, MS Rés 660). The text was presumably the third in the trilogy of which the first two works were idem, L’Académie Imperiale de Musique; histoire littéraire, musicale, choréographique, pittoresque, morale, critique, facétieuse, politique et galante de ce théâtre, 2 vols, Théâtres lyriques de Paris [1] (Paris: Author, 1855); idem, L’Opéra-Italien de 1548 à 1856, Théâtres lyriques de Paris [2] (Paris: Author, 1856).

Jacques Offenbach  117 But Castil-Blaze’s text remained unknown, with Offenbach’s 1856 ‘Concours’ sketch left commanding the field. The latter’s influence was more than obvious in a series of thirty articles published by Louis Meneau in Le Ménestrel between 19 August 1860 and 14 April 1861. This is a much less ambitious undertaking than Castil-Blaze’s, and in its final article reaches a conclusion that owes a great deal to Offenbach’s essay written five years earlier. Here is Meneau’s position ironically encapsulated in a single paragraph: Today, the Opéra-Comique has become a branch of the Opéra. One is sometimes astonished to hear, in certain scores which have been the most played in recent years, the orchestra interrupt itself to allow the dialogue to take the place of the music; since, if this dialogue were replaced by recitative, one would have nothing short of a grand opéra in which no-one would recognise the original genre. It appears that our compatriots compensate with the Opéra-Comique since our premier lyric stage – which ought to be national par excellence, since the whole nation contributes from its sous to its splendour – serves as a pedestal for foreign composers.48 To characterise the Opéra-Comique as a branch [succursale] of the Académie Impériale de Musique, and to claim that the music of opéras comiques had almost eclipsed their spoken dialogue, were rhetorical moves that would not have been out of place in Offenbach’s ‘Concours’ text. And while Offenbach had stopped short of formal criticism of the presence of foreign composers at the Académie Impériale de Musique (indeed, he had been positive about Verdi’s Les vêpres siciliennes in 1855), his commentary on le compositeur de l’avenir in Le carnaval des revues has much in common with Meneau’s complaint; indeed, it is entirely possible that this edge to Meneau’s article was a product of the Paris premiere of Tannhaüser that had taken place a month before publication. With this construction of history in place, Offenbach was able to situate the activities of the Théâtre des Bouffes-Parisiens within it as he sought to justify his announcement of the ‘Concours’. His theatre’s task was, he said, ‘to resuscitate the original and true genre’ of opéra comique.49 Its success in 48 ‘Aujourd’hui, l’Opéra-Comique est devenu une succursale du grand Opéra. On est parfois étonné, à l’audition de certaines des partitions qui ont été le plus jouées dans ces temps derniers, d’entendre l’orchestre s’interrompre pour laisser le dialogue prendre la place du chant; car si on remplaçait ce dialogue par un récitatif, on aurait bel et bien un grand opéra dans lequel personne ne reconnaîtrait le genre primitif. Il semble que nos compatriotes se dédommagent sur l’opéra-comique de ce que notre première scène lyrique, - qui devrait être nationale par excellence, puisque la nation tout entière contribue de ses deniers à sa splendeur, - serve de piédestal aux compositeurs étrangers’ (Louis Meneau, ‘L’opéra-­ comique, sa naissance, ses progrès, sa trop grande extension’, Le ménestrel, 14 April 1861). 49 ‘De ressusciter le genre primitif et vrai’ (Revue et Gazette musicale de Paris, 20 July 1856, 231, 3:5).

118  Jacques Offenbach its first year had been to elide the opéra comique of the past with the farse of Cimarosa and the early Italian masters. When Offenbach claimed that the repertory of his theatre had no other ambition than to ‘faire court’, he was simply juxtaposing its works with those such as Massé’s Galathée and La chanteuse voilée in which, as has been seen, Offenbach identified a composer who tried to ‘faire grand’.50 Works at the Théâtre des Bouffes-­Parisiens were, according to Offenbach, but the first step on the generic ladder; how, he asked, could composers step up the ladder if that first step did not exist? ‘Opéras comiques in three acts’, he argued, remained the domain of the proven master, and were an unequal challenge to the debutant composer.51 With this claim, he was able to merge the interests of the young composer with the works, limited by its licence, of his own institution. In his account of the French eighteenth-century tradition of opéra comique, Offenbach was not only subtly (re)writing history but also reconfiguring the present for his own ends by pointing to the betrayal of true opéra comique by his colleagues and their immediate predecessors, and by praising works that seemed to embody the characteristics that he considered representative of the truth of opéra comique. In practical terms, he was forbidden to mount productions of classic opéra comique, but that did not stop him from attempting to produce an eighteenth-century French classic: Rousseau’s Le devin du village. The fate of Rousseau’s Le devin du village at the Bouffes-Parisiens explains just how much difficulty Offenbach experienced conducting negotiations with the Minister of State, and reciprocally, how much difficulty the state encountered while trying to fend off Offenbach’s advances. Offenbach had approached the manager of the Académie Impériale de Musique, FrançoisLouis Crosnier, and obtained permission to mount Le devin du village at the Bouffes-Parisiens. While seeking permission from the manager of the institution at which the work had originally been performed, Offenbach was quite rightly doubtful that such permission lay in Crosnier’s gift; he accordingly then wrote to the Minister of State to confirm this permission in March 1856. In his request, he noted that rehearsals had already started, that the production had received the institution’s greatest care and that it could not fail to please the public. The work, in one act and with three characters only, appeared to fit the terms of the Bouffes-Parisiens’ licence perfectly.52 The exchange of reports and draft letters prepared within the ministry goes a long way to showing just how close the system of exclusive licences was to collapse, as various civil servants attempted to justify refusing the permission that Offenbach sought. An internal report, dated March 1856, and 50 ‘Il n’a d’autre ambition que celle de faire court’ (ibidem; emphasis original). 51 Ibidem, 3:6. 52 The four groups of documents relating to the negotiations between the ministry and the Bouffes-Parisiens on which this paragraph and the following one are based are found in F-Pan F21 1136; they are identified numerically in the discussion that follows.

Jacques Offenbach  119 presumably prepared within just a couple of days of the receipt of Offenbach’s request, points to a reluctance to authorise ‘the reduction to the proportions of his theatre the great works of the lyric stage’.53 There is a strong sense in this comment that its author took as axiomatic that a work that was originally destined for the Académie Royale de Musique would be similar to one which that individual could hear at the Salle Le Peletier in the 1850s: such works as Meyerbeer’s Le prophète or Verdi’s Les vêpres siciliennes, grands opéras in five acts on historical themes with ballets, elaborate sets, a colossal cast and ostentatious costumes. That Le devin du village was in fact a very different sort of work became progressively clearer as the civil servants in the ministry worked on the case, and finally admitted that it was perhaps a work ‘that did not have a very great importance’.54 On the other hand, in its use of fourteen numbers and closely integrated recitative, it would have been a significantly more complex undertaking had it been reworked for the Bouffes-Parisiens, and the outcome would have been in marked contrast to other works at the theatre. In this initial report, a second argument was advanced for refusal: the composer in question was particularly worthy of respect. There can be little doubt that Jean-Jacques Rousseau’s reputation was as strong in the mid1850s as it ever had been, and that he might qualify for perhaps greater protection than other composers active a hundred years before, but this was a strange reason to offer for refusing to allow a production of his best-known work. Attached to this first internal report to the minister was a draft letter which, although it ought to have summarised the two reasons given earlier, simply introduced a third reason for refusal: that ‘the work belong[s] to a repertory too different to the one you [are permitted to] exploit’.55 This first report and its summarising draft letter were clearly met with a mixed reception, since a second version of the letter to Offenbach, dated 12 March 1856 (the date is crossed through), contains very different reasoning. The relevant part reads: The works of dead masters being placed under the protection of the higher authority, I can not allow the opera in question to be withdrawn from the theatre to which it belongs in order to be changed into an opéra comique and transferred to a stage of a lower order.56 Here, the reduction of Le devin du village to the proportions of a work for the Bouffes-Parisiens was described in different, more specific terms: it was 53 ‘de réduire aux proportions de son théâtre les grands ouvrages de la scène lyrique’ (F-Pan F21 1136/2). 54 See below, note 57. 55 ‘Cet ouvrage appart[ient] à un répertoire trop différent de celui que vous exploitez’ (ibidem). 56 ‘Les œuvres des maîtres morts étant placées sous la protection de l’administration supérieure, je ne puis permettre que l’opéra dont il s’agit soit retiré du théâtre auquel il appartient pour être changé en opéra comique et transporté sur une scène d’ordre inférieure’ (F-Pan F21 1136/3).

120  Jacques Offenbach the change into an opéra comique that is so objectionable, coupled to the fact that such a change would be accompanied by a move to a stage of a lower order. Plausible reasons, indeed, but they represented subtle changes in emphasis from the earlier report. It was the first of the three reasons given in this document that ought to have been the most important: quite simply that Le devin du village ‘could [not] be withdrawn from the theatre to which it belongs [auquel il appartient; emphasis added]’. One of the tasks of the legislation surrounding the assignment of licences was to avoid just the sort of appropriation that Offenbach was proposing. However, the opening clause in this paragraph invoking the protection of ‘l’administration supérieure’ for dead composers betrayed uneasiness on the part of the ministry in exercising this particular power. The matter was made even worse in the subsequent and final draft of the letter of refusal which included the paragraph cited earlier with a marginal addition that qualified the works of such dead composers as ‘meriting respect, even though they are without great importance’.57 While this finally acknowledged the relatively slender dimensions of Le devin du village (so poorly judged in the first ministerial report), the effect of this further clause was to bury the question of theatrical property more deeply into the text and to minimise its effect. Offenbach was no stranger to these sorts of refusals. His first work for an established Parisian theatre had been Pépito, produced at the Théâtre des Variétés two and a half years earlier. This had broken the terms of the Variétés’ licence and had threatened to impinge on the repertory of the Opéra-Comique. The minister’s action was swift and decisive: Pépito came off the stage at the Variétés within a month of the complaint being made (probably by Perrin, since he had most to lose from the competition).58 Pépito was revived (within the terms of Offenbach’s licence) at the Bouffes-­ Parisiens on 10 March 1856, only five days after his exploratory letter to the minister concerning Le devin du village, and two days before the first internal report. Offenbach must have been struck by the change in attitude between Pépito’s rapid and summary rejection in 1853 and the refusal of permission for a revival of Le devin du village in 1856. In 1853, the argument was simple, and familiar to theatre managers since 1807: that they should ‘be restrained in the future within the limits imposed by their cahier de charges as a reciprocal guarantee to each dramatic enterprise’.59 But in 1856, Offenbach was offered a mixture of excuses, with the central one buried in a welter of obfuscation. It is difficult to ignore the equivocal position of the ministry in the affair of Le devin du village which may well have offered encouragement 57 ‘méritant des égards, alors même qu’ils n’ont pas une très grande importance’ (F-Pan F21 1136/4). 58 Yon, Jacques Offenbach, 118–119. 59 ‘soient retenus à l’avenir dans les limites que les cahiers des charges imposent, comme une garantie réciproque, à chacune des entreprises dramatiques’ (F-Pan F21 1133, cited ibidem, 119).

Jacques Offenbach  121 as Offenbach launched a campaign of transgression against his licence that was ultimately to prove successful.

Foreign traditions Offenbach’s attempt to exploit French traditions of the ancien régime had fallen foul of official interference and the limitations of his licence. But only weeks after the debacle with Le devin du village, the Bouffes-Parisiens was to have one of the biggest successes of its early years with an eighteenth-century stage work. The successful production of Mozart’s Der Schauspieldirektor as L’impresario in May 1856, however, invoked an entirely different repertory and set of references to those evolved in Offenbach’s engagement with the music of the ancien régime: German opera, and opera by ­German-speaking composers of the previous century.60 In nineteenth-century German-speaking states, the four numbers and overture of Der Schauspieldirektor were regarded as a problematic constellation of compositions that did not constitute what was then thought of as opera. Of the various attempts to rework the music, the most successful was that of Louis Schneider and Wilhelm Taubert. This version took the music of Der Schauspieldirektor, supplemented it with orchestrations of Mozart’s Lieder and other vocal works and provided it with a comic libretto set during preparations for Die Zauberflöte. The resulting production was originally entitled Mozart und Schikaneder (later revivals of this version returned to the original title, however), and was given for the first time in Berlin in 1845. Offenbach took this version and commissioned Léon Battu and Ludovic Halévy to write a new libretto which kept that action loosely within the domain of opera production, familiar from Der Schauspieldirektor and Mozart und Schikaneder, but turned it into a simple story of duping a guardian into agreeing to a love-match to which he is initially opposed. Musically, L’impresario was very close to Schneider and Taubert’s 1845 reworking: one number was substantially cut and another resequenced. The production was successful well beyond what Offenbach might have expected. Its immediate effect was to distance the Bouffes-Parisiens from Hervé’s Folies-Nouvelles, an act that was all the more desirable in the light of some adverse press in the first couple of months of 1856 that Offenbach was anxious to counter before the launch of his ‘Concours’. The wider impact of L’impresario was even more remarkable: commentators borrowed critical discourses from the then-commonplace eulogies to Don Giovanni 60 The relationship between Der Schauspieldirektor, Mozart und Schikaneder and L’impresario is outlined in Mark Everist, ‘Mozart and L’impresario’, ‘L’esprit français’ und die Musik Europas: Entstehung, Einfluß und Grenzen einer ästhetischen Doktrin, eds. Rainer Schmusch and Michelle Biget-Mainfroy (Hildesheim, Zürich and New York: Olms, 2006), 398–411 on which the following two paragraphs are based.

122  Jacques Offenbach and Le nozze di Figaro to praise this still-slender work further than might have been thought reasonable given the ambitions of the original Gelegenheitsstück, in the same way that Bruschino would be compared favourably with Il barbiere di Siviglia and Guillaume Tell in 1857. Offenbach’s successful appropriation of the past had been a financial and managerial success, and he had struck gold in associating his opera house with arguably one of the most revered composers in Second-Empire Paris. The only point at which Weber and Gluck make any impact on Offenbach’s engagement with the past is in the role they play in Le carnaval des revues. It could quite reasonably be argued that their presence in this scene is triggered by the fact that their music had been the subject of recent revival at the Théâtre-Lyrique. Weber’s presence in the cast might also have been a result of needing a German-speaking character to explain the plays on words in the scenes that involved Meyerbeer and Wagner. Neither Gluck nor Weber figure in Offenbach’s reconfiguration of the comic opera of the past, and it is an open question whether Offenbach knew of the existence of Gluck’s opéras comiques. At the beginning of the 1856 manifesto on opéra comique presented in the ‘Concours’ article, Offenbach had drawn attention to the important role that Pergolesi had played in the birth of the genre and had drawn fiercely nationalistic distinctions between Pergolesi and Grétry, and – for a later ­generation – Cimarosa and Boïeldieu.61 It is no surprise then to read in ­Albert de la Salle’s 1860 account of the first five years of the Bouffes-­ Parisiens’ history that Gevaërt had been commissioned to arrange Pergolesi’s La serva padrona for Offenbach’s theatre; it failed to be produced at the Bouffes-­Parisiens but was revived at the Opéra-Comique in 1862.62 In its one-act format and its use of a small number of soloists, La serva padrona seemed to be a close fit with the works permitted by Offenbach’s licence; it was at least as suitable as the two works, Der Schauspieldirektor and Il signor Bruschino, that did enjoy productions at the Bouffes-Parisiens. The evidence is elusive, but various factors explain the work’s redirection. There already existed a perfectly serviceable adaptation of Pergolesi’s score as an opéra comique that dated from 1754. La serva padrona had been translated by Pierre Baurans as La servante maîtresse and had enjoyed immense success at the Comédie-Italienne that year. There were therefore questions concerning the rights to the work in its guise as an opéra comique that would have posed insuperable problems for the Bouffes-Parisiens if faced with a challenge from the Opéra-Comique: the Comédie-Italienne was viewed as the historical predecessor of the Opéra-Comique, and La servante maîtresse as part of its repertory. By the time La servante maîtresse was revived at the Opéra-Comique in 1862, any claims to it by the Bouffes-Parisiens had been completely forgotten, and the press was pleased to identify the opera as an 61 Revue et Gazette musicale de Paris, 20 July 1856: 230. 62 Albert de Lasalle, Histoire des Bouffes-Parisiens (Paris: Bourdillat, 1860), 44.

Jacques Offenbach  123 important landmark in the Opéra-Comique’s history and a logical inclusion in its repertory.63 As Offenbach cast around for other Italian works, he was drawn to the repertory of farse that Rossini had written for Venice’s San Moisè theatre between 1810 and 1813. The last of these, Il signor Bruschino, formed the basis of the second great foreign success of the Bouffes-Parisiens’ early years.64 Much of the success garnered by L’impresario was replicated in Bruschino, as Rossini’s farsa giocosa was known in Paris. Unlike L’impresario, however, Offenbach was not only returning to the past (Il signor Bruschino had been premiered in 1813), but of course the work’s composer was alive, living in Paris, and by the 1850s, a national monument. Furthermore, Bruschino was the first musical event of any significance publicly associated with the composer since his return to Paris in 1855. The production was surrounded by a series of discursive moves that associated Rossini with the production, that reminded readers of journals of the early history of the composer of Il barbiere di Siviglia and Guillaume Tell and that praised the Bouffes-Parisiens for having been the first theatre to mount a ‘new’ opera of Rossini since 1829 (pasticci were excepted).65

63 Lasalle also wrote a ‘Notice historique sur la Servante maîtresse’ published with the piano-­ vocal score of the work that accompanied the 1862 production at the Opéra-Comique; it is dated 16 August that year (ÉDITION DÉDIÉ A MONSIEUR ÉMILE PERRIN / LA / SERVANTE / MAITRESSE / OPÉRA COMIQUE EN DEUX ACTES / PAROLES FRANÇAISES DE / BAURANS / MUSIQUE DE / PERGOLESE / PARTITION ­RÉDUITE POUR PIANO ET CHANT / PAR SOUMIS / Seule édition conforme aux représentations de l’Opéra-Comique et précédée d’un / notice historique / PAR / ALBERT DE LASALLE / … / PARIS / E. GIROD, ÉDITEUR / 16, BOULEVARD MONTMARTRE, 16 / - / 1862.). In this text, he curiously made no reference to the planned performance at the Bouffes-Parisiens, as he had in his book published two years earlier. 64 The work was premiered under the title Bruschino at the Bouffes-Parisiens on 28 D ­ ecember 1857. The piano-vocal score was published as BRUSCHINO / Opéra Bouffe en deux actes / Poème de / Mr A de Forges / Musique de / ROSSINI / Acct de Piano par H Salomon / … / Paris, LÉON ESCUDIER, 21 Rue Choiseul, / Propriétaire pour tous Pays. Given that we know very little of the chronology of the work on La servante maîtresse, it is entirely possible that Offenbach’s searches for Italian opera that resulted in La serva padrona and Il signor Bruschino were coterminous. 65 Claims that Rossini had given the production his blessing were common, and Jules Lovy was typical: ‘The illustrious composer has not only authorised the performance of his work at the Bouffes-Parisiens, but he has again promised M. Offenbach his assistance and advice, “considering himself happy”, he added graciously, “to do something for the one he calls The Mozart of the Champs-Elysées”’ (‘L’illustre compositeur a non-­s eulement autorisé l’exécution de son œuvre aux Bouffes-Parisiens, mais encore il a promis son assistance et ses conseils à M. Offenbach, ‘s’estimant heureux, a-t-il ajouté gracieusement, de faire quelque chose pour celui qu’il appelle le Mozart des Champs-Élysées’) (‘Un opéra de Rossini aux Bouffes-Parisiens’, Le Ménestrel, 1 November 1857). In his eulogy to the composer on the occasion of Bruschino’s premiere, Henry Boisseaux introduced the ­c omposer: ‘This man, it is Rossini, the composer of Il barbiere, the composer of Guillaume Tell’ (‘Cet homme, c’est Rossini, l’auteur d’Il barbiere, l’auteur de Guillaume Tell’) (‘Premières représentations – Bouffes-Parisiens: Bruschino’, Revue et gazette des théâtres, 31 December

124  Jacques Offenbach Unlike ­Offenbach’s other appeals to the past, with Bruschino, and thanks to Rossini’s longevity and very early success, he was able genuinely to embody the past on the boards of his own theatre. The effect of this embodiment cannot be overstated. In comparison with the very real cultural capital that Offenbach earned through his production of L’impresario, his musical and theatrical profit from Bruschino was immense. In stark contrast to the Bouffes-Parisiens’ production of L’impresario, which came as something of a surprise to Parisian theatrical connoisseurs, Bruschino was heavily trailed in the press. The piano-vocal score was ­published by the Escudier brothers who had advertised the forthcoming production incessantly in their house journal, La France musicale. Not only were there weekly bulletins on the progress of rehearsals for the six weeks before the premiere on 28 December 1857, but during this period, the journal offered the published score of Bruschino free to all subscribers from 15 December that same year.66 The plan was predicated on an early-December premiere for the work, but the fact that subscribers to La France musicale possessed copies of the work two weeks before its premiere does not seem to have damaged its reception. Enthusiasm for Bruschino well before its ­premiere was not restricted to those who had a vested interest in its success. Jules Lovy, writing in Le ménestrel, was typical in pointing to a number of qualities to be expected in the production: the original libretto was not only ‘une délicieuse bouffonnerie’ but had been arranged by Philippe-­AugusteAlfred Pittaud de Forges in a way that had astonished Rossini himself.67 Finally, Offenbach – as he had in the case of L’impresario  – had engaged new ­soloists, one of whom – Charles Duvernoy – had, like his Mozartian predecessors, just won the premier prix at the Conservatoire.68 The importance of Rossini, and the imprimatur that he gave to the production, cannot be overestimated. A recurring trope in the pre-­production commentaries was the juxtaposition of the ‘high-priest of modern music’ with ‘the little temple of the Passage Choiseul’ (the Théâtre des Bouffes-­ Parisiens).69 This went very well with similar types of hagiography that had

66 67

68 69

1857). Journalists had to look back as far as the Stabat mater (known in Paris since 7 ­January 1842) for comparable Rossini premieres: ‘Not since the Stabat have we seen public curiosity excited in so lively a manner’ (‘Jamais, depuis le Stabat, on n’a vu la curiosité publique aussi vivement excite’) (‘Nouvelles’, La France musicale, 15 November 1857). The advertising campaign began in La France musicale, 15 November 1857. ‘Le poème primitif est une délicieuse bouffonnerie italienne que M. De Forges a fait entrer dans un cadre nouveau, avec une habilité qui a étonné Rossini lui-même’ (Lovy, ‘Opéra de Rossini’). Ibidem. ‘But M. Offenbach is conscious of the obligation and the double responsibility that he incurs in having the name of the High-Priest of modern music enter his little temple in the Passage Choiseul’ (‘Mais M. Offenbach a le sentiment du devoir et de la double responsabilité qu’il encourt en faisant entrer le nom de grand-prêtre de la musique moderne dans son petit temple du passage Choiseul’) (‘Semaine théâtrale’, Le ménestrel, 20 December 1857).

Jacques Offenbach  125 accompanied the production of L’impresario the previous year.70 The fact that Rossini – ‘the high-priest of modern music’ – could be seen in the neighbourhood (he lived no more than a couple of blocks away from the Passage Choiseul on the Chaussée d’Antin) simply enhanced the prestige that the institution could reap from the production. It is inevitable that Bruschino would have been praised to the skies in La France musicale, even if the Escudier frères did manage to resist the temptation to review the work themselves and handed the task over to Achille de Lauzières who signed himself ‘A. Aldini’.71 But others were just as quick to praise. Both Jacques-Léopold Heugel in Le ménestrel and Henry Boisseaux in the Revue et gazette de théâtres – writing for very different audiences – could not have been more positive; they pointed to the quality of the work, its arrangement by de Forges and Offenbach and the performance and the number of celebrities – musical, political and social – in the audience.72 In the same way that L’impresario had benefitted from analogy with Don Giovanni and Le nozze di Figaro, Bruschino was praised by association with either the Il barbiere di Siviglia-La cenerentola pairing or that of Il barbiere and Guillaume Tell. Boisseaux went further and used the work as an opportunity to renew Paris’s praise for the ‘high-priest of modern music’. He addresses his review directly to Rossini himself: One no longer knows what place you should be assigned in art. To affirm that you are a powerful colourist… what would Weber say? To propose that your orchestration is splendid… what would Beethoven say? To take witness from the School that you have infused with science… what would M. Fétis say? To swear in front of all that melody flows in buckets from your work… what would Grétry, Mozart and Auber say? No, a hundred times no! I shall not judge you. I shall simply repeat, alongside the intelligent critic of Le Figaro that you are only a musician, nothing but a musician, but the most perfect of musicians!73 70 See Everist, ‘Mozart and L’impresario’, 430–433. 71 A. Aldini, ‘Premières Représentations I: Bruschino – Rossini (Bouffes-Parisiens)’, La France musicale, 3 January 1858. For the pseudonym, see Charles Joliet, Les pseudonymes du jour, new edn (Paris: Dentu 1884), 33–34. 72 Jacques-Léopold Heugel, ‘Un partition de Rossini aux Bouffes-Parisiens’, Le ménestrel, 3 January 1858; Boisseaux, ‘Premières représentations’. 73 ‘On ne sait plus quelle place il faut vous assigner dans l’art. Affirmer que vous êtes un puissant coloriste…, que dirait Weber?… Soutenir que votre orchestration est splendidement combinée… que dirait Beethoven?… Prendre à témoin l’École que vous avez la science infuse, que dirait M. Fétis?… Jurer devant tous que la mélodie s’échappe à flots de votre œuvre… que diraient Grétry, Mozart et Auber? Non, cent fois non! je ne vous jugerai pas. Je répéterai seulement, avec l’intelligent critique du Figaro que vous n’êtes qu’un musicien, rien qu’un musicien, mais le plus parfait des musiciens!’ (ibidem). This group of composers, with the exception of Auber, reinforces the view of Rossini as a living composer of the past. By 1858, Auber was in the middle of a career that would last for another 15 years but that had started 35 years previously; he is now allied with composers of the past: Grétry,

126  Jacques Offenbach The unquestioning praise for Rossini is here hardly the issue; but the consumption of Bruschino as a work that could serve as a springboard for such eulogy – by analogy with core repertory of the Théâtre-Italien and Académie Impériale de Musique – is testimony to its success. Offenbach’s unsuccessful jousts with the state over Pépito at the Variétés and over Le devin du village would have made him realise how important state control over his activities – no matter how well connected he was ­personally – could be. His success in negotiating a slight change in his ­licence – increasing the number of soloists in musical works from three to four – when he moved operations to the Salle Choiseul in October 1855, and the ambivalent nature of the refusal to allow him to play Le devin du village, would also have made him realise that however important such control, it could be manipulated legally and by other means. Lasalle carefully documented no less than six changes in Offenbach’s treatment of his licence between 1855 and the premiere of Orphée aux enfers in 1858.74 The first change  – already ­mentioned – was entirely legal. The remaining breaches of Offenbach’s ­licence were illegal, and of these Bruschino represented one of the most ­i mportant stages. The work demonstrates how Offenbach could use the power of a past work – especially when its venerable composer was still alive to authorise it – to generate so much publicity that he could smuggle in generic changes to which he could subsequently point as legitimation of the enlargement of the works mounted as his own theatre.75 The success of Bruschino, like that of L’impresario, had not come without effort. Bruschino was originally made up of an overture and eight numbers separated by recitativo semplice, and this entailed not only the reworking of the libretto for the sung numbers but also the replacement of the recitative by spoken dialogue in the same way deployed by Castil-Blaze in his adaptations of Rossini for the Théâtre Royal de l’Odéon in the 1820s, versions that were still in use at the Théâtre-Lyrique in the 1850s (Table 3.1). The librettist, de Forges, effectively wrote a new narrative around the characters and the music he found in the original; this explains the reassignment

Mozart, Beethoven and Weber. If Heugel had been thinking of Rossini as a modern composer (rather than as ‘the high priest of modern music’ – which is a different concept entirely) – he would have chosen Meyerbeer, Halévy and Verdi as points of reference. 74 Albert de Lasalle, Histoire des Bouffes-Parisiens: 15–17. 75 Lasalle’s careful diagnosis of the six ‘époques’ in the development of the Bouffes-Parisiens’s repertory is distorted by Yon when the latter claims that Bruschino ‘officialise le droit au cinquième personnage et autorise des pièces d’une plus grande dimension’ (Jacques Offenbach, 199). This fundamentally misunderstands the dynamic of the relationship between the theatre and the state. The ‘époques’ identified by Lasalle correspond to progressive transgressions of Offenbach’s licence that were – as Lasalle observed – ‘tolerances’ (Histoire des Bouffes-Parisiens, 16). When Offenbach sought to renew his licence in October 1860, the terms of the revision were exactly the same as they had been five years previously (F-Pan F21 1136). If Offenbach’s licence had indeed been subject to change during the late 1850s, the renewal of the licence would have had to have taken account of any such changes.

Jacques Offenbach  127 Table 3.1  Comparison of Italian and French versions of Rossini, Il signor Bruschino Il signor Bruschino (1813)

Bruschino (1857)

Sinfonia

Ouverture Act I

1 Introduzione ‘Deh tu m’assisti amore’ 2 Duetto ‘Io danari vi darò’ (Florville, Filiberto) 3 Cavatina ‘Nel teatro del gran mondo’ (Gaudenzio) 4 Terzetto ‘Per un figlio già pentito’ (Florville, Bruschino, Gaudenzio) 5 Recitativo ed aria ‘Ah voi condur volete’ – ‘Ah donate il caro sposo’ (Sofia) 6 Aria ‘Ho la testa’ (Bruschino, Gaudenzio, Commisario, Sofia, Florville) 7 Duetto ‘È un bel nodo’ (Sofia, Gaudenzio) 8 Finale ‘Ebben, ragion, dovere’

1 Introduction et Duettino ‘Hélas! Je désespère’ (Corilla, Flavio) 2 Duetto ‘C’est moi qui vous rembourserai’ (Flavio, Giuseppe) 3 Air ‘Tout va bien dans ma citadelle’ (Bruschino) 4 Trio ‘Que d’un père’ (Flavio, Bruschino, Bombarda) Act II 5 Récitatif et air ‘O père inexorable’ – Ah! Pour vous quand l’espoir’ (Corilla) [repeat of cabaletta cut] 6 Quintette [sic] ‘L’imposture est bien complète’ (Bruschino, Bombarda, Giuseppe, Corilla) 7 Rondo ‘Oui grâce à cette ruse’ [= cabaletta only of source duetto] (Corilla) 8 Finale ‘Hélas! Plus d’espérance’ (Bruschino, Bombarda, Giuseppe, Corilla, Flavio)

of the cavatina ‘Nel teatro del gran mondo’ from Gaudenzio Strappapuppole to Bruschino in the 1857 French version, as well as other changes in the aria ‘Ho la testa’ and in the finale. At face value, Il signor Bruschino should have been an easy work to have adjusted in line with the Bouffes-Parisiens’ licence. In one act and consisting of eight numbers, its physiognomy resembled most works produced at the theatre in 1857. Its original cast ran to more that the Bouffes’ allotted limit, but reworking of numbers with more than four characters should have posed no problem. But rather than reworking Rossini to conform exactly to the requirement of his licence, Offenbach’s strategy was to test its limits, to see how far he could go beyond the terms of his licence before the state – or his theatrical competitors – took action. He did this, however, in ways that would not damage the work if complaints were made – in contrast to the experience with Pépito and Le devin du village, when they had been either taken off after a curtailed run or withdrawn during rehearsal. Two examples show how Offenbach accomplished this administrative high-wire act. One of his ambitions was to put himself in a position where he could ultimately mount opérettes in two acts – the format of Orphée

128  Jacques Offenbach aux enfers that was to be premiered the following year. Il signor Bruschino was in its original form a single act, and so too were all the works with seven or eight numbers then in the repertory of the Bouffes-Parisiens.76 Offenbach chose deliberately and unnecessarily to divide the work into two acts, for which there was no musical, dramatic or practical justification.77 A work in two acts would have broken the terms of Offenbach’s licence, but if he had been challenged, he could simply have abandoned the division with no change to the content of the work. If unchallenged, he would have had case law on his side as he sought to defend future works in two acts that – like Orphée aux enfers – were significantly more ambitious. Similarly, the ‘aria Bruschino’ originally entitled ‘Ho la testa’ was apparently transformed into a quintette, again breaking the terms of Offenbach’s licence. Here the example is even more subtle than that of the act division: Rossini’s original in Il signor Bruschino certainly consists of an aria with four additional ancillary parts; although much of the aria consists of duet and trio textures, it certainly employs all five voices in the stretta, and removed from its original context might well be titled a quintet.78 But in Bruschino, there are only four characters in the number, and the five-part music in Rossini’s original is reworked as a quartet (the roles of Florville and the Commissario de polizia are elided): the music therefore falls comfortably within the terms of the theatre’s licence.79 But Offenbach titled it quintette in the score (nothing more than an over-technical description of the scoring of Rossini’s original) on the page in question and in the catalogue de morceaux. He knew that all would recognise that a number with such a title was an infringement of his licence, but it would have constituted case-law when he mounted future productions with larger ensembles, and furthermore he could defend the number, if required, by showing that it was really a quatuor in which no more than four soloists sang at any one time.80 In the listing of voice-types for the number (in both the score and the index), Offenbach adds a tenor to the soprano, baritone and pair of basses; although the tenor Flavio/Florville sings in Rossini’s original, he is entirely absent from Offenbach’s arrangement of the number. In the libretto, however, Offenbach was at the same time more cautious and even more subtle; he called ‘Ho la testa’ / ‘L’imposture est bien faite’ a morceau d’ensemble, a terminology that would

76 Even those works composed just before and after Bruschino that aggressively attempted to transgress the theatre’s licence in terms of the number of characters retained a single act: Les petits prodiges (19 November 1857) and Mesdames de la halle (3 March 1858). 77 As Table 3.1 shows, each act consists of four numbers each. 78 Arrigo Gazzaniga (ed.), Il signor Bruschino, ossia Il figlio per azzardo: farsa giocosa per musica in un atto di Giuseppe Foppa, musica di Gioachino Rossini, Edizione critica delle opere di Gioachino Rossini 1:9 (Pesaro: Fondazione Rossini, 1986), 303–317. 79 BRUSCHINO / Opéra Bouffe en deux actes: 73–76, 79–85. 80 Ibidem: 68.

Jacques Offenbach  129 include a perfectly legal quartet but also any further ensembles that might include more characters.81 Bruschino was a perfect context in which to explore these duplicitous manoeuvres. The noise generated by the work was enough to allow enough people to overlook what might be thought of as minor transgressions. The timing was also fortunate. The source of any objection to breaches of ­Offenbach’s licence was likely to be the director of the Opéra-Comique, and Emile Perrin had probably been behind the complaints about Pépito, but Perrin had stepped down as director of the Opéra-Comique right in the mid­ ruschino dle of the public excitement about the impending production of B (20 November 1857), and despite the experience he had from his work at the Académie Royale de Musique and Académie Impériale de ­Musique from 1847 to 1854, Perrin’s successor, Nestor Roqueplan, might have been thought to be not quite as ready to take on the protégé of the emperor’s halfbrother as he might have been later in his tenure of the management of the Opéra-Comique. In appropriating foreign opera in French translation for the benefit of the Bouffes-Parisiens and in order to circumvent the limitations of his licence, Offenbach was inscribing his activities in a tradition that was common to a wide range of Parisian institutions that supported music drama; the only two exceptions were the Théâtre-Italien – whose licence expressly limited its managers to opera in Italian – and the Opéra-Comique. Since 1800, the Académie Impériale de Musique had mounted translations of Die ­Zauberflöte, Don Giovanni, Euryanthe, Der Freischütz, Mosè in Egitto, ­Maometto II, Poliuto, Otello, Lucia di Lammermoor and I Lombardi alla prima crociata. While these were occasional productions, other theatres – the Gymnase-Dramatique, the Théâtre Royal de l’Odéon and the Théâtre de la Renaissance – also mounted such productions, and in the case of the Odéon largely depended on them for their survival. The practice would of course continue with the production of Tannhaüser at the Académie Impériale de Musique in 1861, and the process of translation would fall out of the control of the state with the repeal of the licensing laws in 1864. *** Offenbach saw the appropriation of the stage music of the ‘long eighteenth century’ as one of several strategies that he could use for the aesthetic positioning of his Théâtre des Bouffes-Parisiens, alongside sycophantic praise for contemporary composers who had little to fear from him, the development of the bourgeois and aristocratic contacts that he had built up in the previous decade and the claims that he could support young composers in

81 Lasalle’s commentary (Histoire des Bouffes-Parisiens, 17) makes it clear that, in this case (and perhaps in Yon’s case as well), Offenbach’s strategy of deception had worked well.

130  Jacques Offenbach a way that no other theatre could. In his critical writing in L’artiste and in his manifesto that accompanied the announcement of the ‘Concours’, he praised the opéra comique of the eighteenth century and those contemporaries who emulated it. Even in the context of such a parody as Le ­carnaval des revues, he could harness the power of eighteenth-century composers to lampoon those, such as the composer of the future, who had little sympathy with opéra comique at all. And finally, although plans to produce ­eighteenth-century music drama at the Bouffes-Parisiens were difficult to bring to completion, Offenbach identified perhaps the four key works in the long eighteenth century that seemed to fall within the remit of his theatre’s licence, and brought productions of two of these to successful conclusions. Offenbach always had his sights set higher than his own theatre, and even if his first experience at the Opéra-Comique itself was a disaster, he was to abandon the sole management of the Bouffes-Parisiens in 1862 in pursuit of more ambitious goals. By that time, the aesthetic identity of the Bouffes-­ Parisiens was well established, with clear artistic space separating it from the Folies-Nouvelles, which had always been Offenbach’s ambition. The consistent and sustained allusion to music drama from the past had successfully done its work.

4 The operas of FrançoisAuguste Gevaert The tour d’horizon

François-Auguste Gevaert’s operatic output has been dwarfed by his later scholarly and administrative career.1 Given that he wrote eleven operas that received performances and was involved in at least four more that failed to come to the stage, this is a sizeable – and to judge from the press – a successful corpus that demands scholarly attention if he is to be given a rightful place in the history of opera during the Second Empire, and if his biography is to be judged in its entirety.2 In terms of the spread of institutions for which he wrote, Gevaert was a significant figure in opéra comique during the 1 The account in GroveOnline dismisses Gevaert’s operatic output in a single line (Anne-­ Marie Riessauw and Jean Hargot, ‘Gevaert, François-Auguste’, Grove Music Online. Oxford Music Online, www.oxfordmusiconline.com/subscriber/article/grove/music/11003 (Accessed 20 April 2009)), whereas the one in The New Grove Dictionary of Opera – scarcely more ambitious – states that ‘His stage works are light in character, in keeping with the political and cultural climate of the time; most are vaudevilles in which simple and memorable songs feature prominently’ (Jean Hargot, ‘Gevaert, François-Auguste’, The New Grove Dictionary of Opera, ed. Stanley Sadie. Grove Music Online. Oxford Music Online, www.oxfordmusiconline.com/subscriber/article/grove/music/O007688 (Accessed 20 April 2009)). Lightness of character will be seen to be exactly not how his opéras comiques were viewed, and the idea that they have anything to do with vaudeville is incomprehensible. The entry in Die Musik in Geschichte und Gegenwart: allgemeine Enzklopädie der Musik, ed. Ludwig Finscher, 26 vols (Kassel: etc.: Bärenreiter, 1994–), Personenteil 7:854–855 goes no further than the entries discussed earlier. Older biographical sources are still useful. See Ernest Closson, ‘Gevaert’, Revue catholique des idées et des faits, 21 and 28 ­December 1928, 4 January 1929; R as Gevaert (Brussels: Lesigne, 1929); Fr. Dufour, Le Baron ­F r.-Aug. Gevaert, Lettres et arts belges: Collection Diamant, Série Artistique 1 (Brussels: Société belge de librairie, 1909); L. Dubois, ‘Notice sur François-Auguste ­Gevaert’, Annuaire de l’Académie royale des sciences, les lettres et les beaux-arts de Belgique 96 (1930), 9­ 7–147. An essential bio-bibliographical guide is Jean Hargot, ‘François-Auguste Gevaert, 1­ 828–1908, compositeur de opéras bouffons et directeur du Conservatoire de Bruxelles: étude bio-­ bibliographique et contribution à l’histoire de la vie musicale belge au xixe siècle’ (Mémoire de licence, Université Catholique de Louvain, 1987). 2 The advantage of Fétis’s account of the composer is its date. Published in 1862, it devoted most of its text to Gevaert’s operatic output, at a point where his scholarly work was still largely in the future. See François-Joseph Fétis, Biographie universelle des musiciens et bibliographie générale de la musique, 8 vols (2nd edn. [with supplement in two vols] Paris: Firmin Didot, 1860–1865), 3, 470–472.

132  The operas of François-Auguste Gevaert ­ econd Empire. There is also evidence that suggests that by the mid-1850s, S he was viewed as one of the next generation of composers for the Opéra after Meyerbeer and Halévy. Of particular interest are the three works that Gevaert wrote for the ­Théâtre-Lyrique between 1853 and 1855 and the four works for the Opéra-Comique between 1858 and 1860 and again in 1864,3 because the 1850s represent the first period in the history of nineteenth-century Parisian stage music in which two institutions were permitted by licence to mount productions of new opéra comique without the risk of triggering the administrative or legal dispute that had characterised such efforts at the Odéon and Nouveautés in the 1820s, or the Théâtre de la Renaissance in the 1830s.4 The consequences of this change in state policy were far-reaching,  and ­Gevaert’s opéras comiques in the 1850s give answers to these particular questions in a particularly revealing way. Gevaert’s stage career encompassed a wide range of genres, librettists and institutions (Table 4.1). He began with two radically different works for the Grand Théâtre in ­Ghent – a three-act serious opera, Hugues de Zomerghem, and a one-act opéra comique, La comédie à la ville. While travelling all over Europe as a Belgian Prix de Rome laureate, he wrote another one-act opéra comique, Les empiriques. Settling in Paris in 1852 led to the sequence of three opéras comiques for the Théâtre-Lyrique, and then three more for the Opéra-­Comique itself. The 1860s were marked by a return to the Théâtre-Lyrique, but only in a collaboration on La poularde de Caux with Clapisson, Gautier, Poise, Bazille and Mangeant – all his contemporaries, an opera for BadenBaden which falls outside the scope of this study, and a final return to the Opéra-Comique in 1864 with Le capitaine Henriot which has been awaiting performance for the best part of a decade. In addition to these works that reached performance are four further works, two opéras comiques and two works apparently for the Théâtre Impérial de l’Opéra. After establishing himself with three successes at the 3 The chronological spread of works for the Opéra-Comique is deceptive, since the 1864 piece, Le capitaine Henriot, was mooted as early as 1856, and had it gone ahead, would have been the first work for the Opéra-Comique, not the last; see later in the chapter. 4 For the Théâtre Royal de l’Odéon, see Mark Everist, Music Drama at the Paris Odéon, 1824–1828 (Berkeley, Los Angeles and London: University of California Press, 2002); for the Théâtre de la Renaissance, see idem, ‘Theatres of Litigation: Stage Music at the Théâtre de la Renaissance, 1838–1840’, Cambridge Opera Journal 16 (2004), 133–162 and idem, ‘Donizetti and Wagner: opéra de genre at the Théâtre de la Renaissance’, Giacomo Meyerbeer and 19th-Century Parisian Music Drama, Variorum Collected Studies Series CS805 (Aldershot: Ashgate, 2005), 309–341. The attempts at mounting operatic productions at the Théâtre des Nouveautés remain without a convincing modern account. See Louis-Henry Lecomte, Les Nouveautés, Histoire des théâtres de Paris 5 (Paris: Daragon, 1907; R Geneva: Slatkine, 1973) and Wild, Dictionnaire, 279–281.

The operas of François-Auguste Gevaert  133 Table 4.1  François-Auguste Gevaert: stage works Title

Institution

Librettist(s)

Hugues de Zomerghem (23 March 1848; 3) La comédie à la ville (5 January 1849; 1) Les empiriques (1851, unperformed; 1) Georgette, ou Le moulin de Fontenoy (27 November 1853; 1) Le billet de Marguerite (7 October 1854; 3) Les lavandières de Santarem (25 October 1855; 3) Untitled opéra (complete September 1856; scheduled for performance September 1857 according to Sardou; 3 originally in 5) Untitled opéra comique described by Fétis and Sardou (before 1857) La fille d’Erwin (never set; September 1857) Quentin Durward (25 March 1858; 3)

Ghent

Victor Prilleux

Ghent

Prilleux

?TL TL

Opéra

Gustave Vaëz Vaëz [and Alphonse Royer / Eugène Scribe] Adolphe de Leuven and Léon-Lévy Brunswick Adolphe d’Ennery and Eugène Grangé Scribe [and Vaëz?]

?TL/OC

Léon Battu

?TL/OC

Victorien Sardou

OC

Eugène Cormon and Michel Carré Cormon and Carré Cormon et Carré

Le diable au moulin (13 May 1859; 1) Le Chateau Trompette (23 April 1860; 3) La poularde de Caux (17 May 1861; 1) Les deux amours (31 July 1861; 2)

TL TL

OC OC TL Baden-Baden

Le Cid (?1861–1862, unperformed and Opéra incomplete; 3) OC Le capitaine Henriot (29 December 1864; 3) [libretto completed before September 1856]

De Leuven and Prilleux Cormon and Amadée Achard Sardou Sardou and Vaëz

Théâtre-­Lyrique, it seems as if Gevaert took to heart some of the press commentaries that had alluded to the suitability of his style and talents for the Opéra and moved in exactly that direction. The evidence of his work, however, is slender and contradictory. Victorien Sardou wrote a good deal about the early history of his libretto of Le capitaine Henriot and his attempts to persuade Gevaert to set it. Sardou’s libretto was complete by September 1856, and he described the situation in a memoir dated the 26th of that month: Gevaert confessed to me yesterday that Vaez was pressing him for the opera that he wrote with Scribe, in five acts and now reduced to three. The hope of getting an opera played quickly this winter and to profit rapidly from these brutes is too fine for Gevaert to hesitate for a moment.

134  The operas of François-Auguste Gevaert To compose two operas at a time is impossible. He will have to lose his turn in December and put back our project [Le capitaine Henriot].5 It seems at this point in 1856 that Sardou had completed the libretto to Le capitaine Henriot and that the Opéra-Comique had a notional date for its performance at the end of the year, but that Gevaert’s plans were now for a work for the Opéra whose libretto – by Gustave Vaëz (Jean-Nicolas-­Gustave van Nieuwenhuysen) and Eugène Scribe – was now complete and for which he had hopes that it would shortly be performed.6 Sardou’s frustration may be gauged by the fact that he had a second opéra comique libretto for ­Gevaert ready for the 1857 season, La fille d’Erwin.7 It became clear by December 1857 that ­Gevaert’s opera was going to be performed after the French translation of Verdi’s Il trovatore as Le trouvère. The Verdi was premiered on 12 January 1857,8 and the last that was heard of Gevaert’s grand opéra was that it was planned for performance in September 1857,9 although Fétis was clear that it was complete.10 From February 1857, the situation became even more confused since it then became clear that Gevaert had also been dealing with Léon Battu for another libretto for the Opéra-Comique that Sardou thought was promised for the Opéra-Comique for September that year. This came to nothing, quite possibly, as Fétis claimed, as a result of Battu’s illness and death later in 1857.11 This leaves three possible incomplete projects between Les lavandières de Santarem and Quentin Durward, to which has to be added a further libretto by Sardou, described by Albert Soubies and Charles ­Malherbe: Le Cid. The date is far from clear, but it is difficult to work out 5 ‘Gevaert m’a avoué hier que Vaez [sic] le pressait pour l’opéra qu’il fait avec Scribe et qui, de cinq actes est maintenant réduit à trois. L’espérance de faire jouer bien vite un opéra cet hiver et de profiter rapidement de la bonne fortune de ces brutes, est trop belle pour que Gevaert puisse hésiter un seul moment. / Faire les deux opéras à la fois est une chose impossible. Il faut donc perdre son tour de décembre et reculer notre affaire [Le capitaine Henriot]’ (Georges Mouly (ed.), Les papiers de Victorien Sardou (Paris: Albin Michel, 1934), 149). 6 Although it is not clear from the citation given earlier, subsequent reference makes it clear that Vaëz is referring to a libretto that he and Scribe had composed, and Vaëz was sending text directly to Gevaert for setting as it was written (‘L’opéra est prêt, le grand. Vaez [sic] est pressé. Ils n’ont rien là-bas, car il lui envoie les paroles tous les jours à mesure qu’elles sont faites. Il m’a montré le manuscrit. Il faut que dans un mois au plus tard, les deux premiers actes soient prêts à être répétés [emphasis added]’; ibidem, 156). Fétis also refers to an unperformed three-act work for the Opéra in 1862 (Biographie universelle des musiciens, 3:472). 7 Mouly, Papiers de Victorien Sardou, 157. 8 Théodore de Lajarte, Bibliothèque musicale du Théâtre de l’Opéra: catalogue historique, chronologique, anecdotique, 2 vols (Paris: Bibliothèque Nationale, 1878; R Hildesheim: Olms, 1969), 2:221–222. 9 Ibidem, 210. 10 François-Joseph Fétis, Biographie universelle des musiciens et bibliographie générale de la musique, 8 vols (2nd edn. [with supplement in two vols] Paris: Firmin Didot, 1860–1865), 3:472. 11 Ibidem.

The operas of François-Auguste Gevaert  135 even if this is for the Opéra or for the Opéra-Comique; it seems difficult to rationalise it with the piece for the Opéra to a libretto by Scribe and Vaëz.12 Table 4.1 also shows the librettists with whom Gevaert worked. Both ­Ghent operas were to libretti by the prolific Victor Prilleux – who will recur in the discussion of the Belgian network in Paris – and the first three operas for the Opéra-Comique were to the first three libretti on which Cormon and Carré collaborated. This was a collaboration that would last for over a decade and include the libretti to Maillart’s Lara and Bizet’s Les pêcheurs de perles. Sardou’s collaborator on Le capitaine Henriot was again Vaëz, who had written the libretto to Les empiriques which apparently remains lost. Gevaert was also lucky to have Vaëz as his librettist for his first Parisian opera, Georgette. Sources differ as to whether Vaëz’s collaborator was Auguste Royer or Scribe, but in either case, these were experienced teams: Scribe and Vaëz had been collaborating since 1847, but Royer and Vaëz had begun collaboration as long ago as 1838.13 But Le billet de Marguerite marked the first of a series of different collaborations with two other well-known teams. De Leuven and Brunswick had been writing libretti together for nearly twenty years – they started with Adam’s Le postillon de Lonjumeau in 1836 – and d’Ennery and Grangé had been writing for just as long. Exactly why Gevaert was so promiscuous with his literary collaborators at the Théâtre-Lyrique and so consistent at the Opéra-Comique may betray something of his working relationships. With well-established librettists at the Théâtre-Lyrique, if Gevaert wanted a particular type of libretto, he would have to hunt around to find a team with one he could use, whereas with the younger team at the Opéra-Comique, it was entirely possible that the composer had more say on the types of scenario that the librettists developed; he was also better established in Paris by 1856. But not only did G ­ evaert work consistently with Cormon and Carré: other librettists – ­Prilleux and obviously Vaëz – showed up at other points in his career, Prilleux most bizarrely with De Leuven, for the multi-authored La poularde de Caux. *** 12 See Albert Soubies and Charles Malherbe, Histoire de l’Opéra Comique: la seconde salle Favart, 1840–1887, 2 vols (Paris: Flammarion, 1892–1893), 2:77, where they print what purports to be a letter from Gevaert to the authors in which Le Cid is an important agent in Gevaert’s withdrawal from the stage. 13 The title page of the piano-vocal score (A LA VILLE DE GAND. / GEORGETTE / ou le Moulin de Fontenoy, / Opéra bouffe en un acte, / (représenté à Paris le 27 9bre 1853) / Poème de Mr / Gustave Vaëz / Musique de / F.A. GEVAERT. / … / PARIS / FÉLIX MACKAR, Editeur Commissaire, / 22 Passage des Panoramas / A GAND, chez V. GEVAERT, Rue Digue de Brabant, 36 / Propriété de l‘Auteur pour la Belgique) mentions Vaëz only as does the libretto (BIBLIOTHEQUE DRAMATIQUE / Théâtre moderne. / GEORGETTE / OPÉRA COMIQUE EN 1 ACTE / PAR M. GUSTAVE VAEZ / MUSIQUE DE / M. GEVAERT / … / MICHEL LÉVY FRÈRES, LIBRAIRES-ÉDITEURS RUE VIVIENNE, 2 BIS,). Gevaert’s autograph gives both Scribe and Vaëz (Brussels, Bibliothèque du Conservatoire (hereafter B-Bc) 38633). For Scribe and Vaëz’s collaborations on Boisselot’s two opéras comiques, Ne touchez pas à la reine! and Mosquita la sorcière, see Chapter 5; for Royer and Vaëz, see Chapter 9.

136  The operas of François-Auguste Gevaert

One-act opéra comique Georgette, ou Le moulin de Fontenoy was certainly the first opéra comique that Gevaert wrote for Paris, and it was the only piece in one act that he wrote for the Théâtre-Lyrique. In many respects, this is entirely predictable. The tradition behind the emergence of what was known as the Troisième théâtre lyrique, which went back at least a quarter of a century, was that it should be an opera house where new composers – and Prix de Rome laureates in particular – could have their first works produced. Now, Gevaert was a Belgian and a winner of the Belgian Prix de Rome, and Georgette, although not his first opera, was his first for Paris, so that the manager of the Théâtre-Lyrique, Jules Seveste, could claim that he was discharging one of the obligations that formed the centre of his licence, although – as the evidence shows quite clearly – much less central to his artistic policy. There is, however, a broader context to Gevaert’s first Parisian triumph, and Georgette has to be understood alongside Gevaert’s previous two oneact opéras comiques – La comédie à la ville and Les empiriques – and as the end of a campaign to write such works that had lasted four years. A comparison of all three works is presented in Table 4.2. Table 4.2  François-Auguste Gevaert: Early one-act opéra comique La comédie à la ville (1849)

Les empiriques (1851)

Georgette, ou Le moulin de Fontenoy (1853)

Ouverture 1 Couplets (Durosier): ‘En réponse à votre lettre’ 2 Trio (Isabelle, Angéline, Durosier): ‘Vous le savez’

Ouverture 1 Trio (Pedrille, Robert, Gaspard): ‘Arrivez donc’

Ouverture 1 Couplets (Georgette): ‘C’est moi qui suis la meunière’ 2 Romance (André): ‘Celle que j’aime va venir’

2 Morceau d’enemble (Estrella, Robert-Gregorio, Tufiador + chorus): ‘Chantons deux époux formant des nœuds’ 3 Romance (Estrella): ‘C’était 3 Trio (Renard, Corbin, l’ami de mon enfance’ Clovis): ‘Pour couronner un si beau feu’

3 Quintetto (Isabelle, Angéline, Flavigny, Durosier, Grandval): ‘La fortune nous a fait’ 4 Air (Isabelle): ‘O Dieu 4 Air avec chœur (Gaspar): de l’harmonie’ ‘Quel est le personnage’ 5 Rondo (Flavigny): 5 Trio (Robert, Gaspard, ‘C’est moi que partout’ L’alcade): ‘Bon, les voilà’ 6 Couplets (Grandval): ‘Je suis le fils’ 7 Trio (Flavigny, Durosier, Grandval): ‘Sa franchise me plaît’ 8 Finale: ‘Une telle insolence’

6 Duo (Estrella, Gaspard): ‘Ah, quel transport’ 7 Final: ‘Voici donc l’instant du miracle

4 Duo (Georgette, André): ‘Le cœur me bat’ 5 Morceau d’ensemble (Georgette, Renard, Corbin, Clovis): ‘Attaquons’ 6 Duetto (Georgette, Corbin): ‘Fidèle j’accours’ 7 Chanson (Clovis): ‘Vive la goguette’ 8 Finale: ‘Ma vengeance est complète’

The operas of François-Auguste Gevaert  137 There are problems with the sources for these three operas: no libretto of Les empiriques exists, and this makes assessing its tone difficult. Similarly, the fact that La comédie à la ville was never performed in Paris means that its footprint in the press is significantly smaller than, say, for Georgette. According to Fétis, Gevaert attempted to have Les empiriques performed at the Théâtre-Lyrique, but it ran up against competition from another work with a similar subject, and was dropped in favour of Georgette.14 But Ernest Reyer, reviewing the premiere, stated that the work had originally been intended for the Opéra-Comique, but its manager, Emile Perrin, had simply received the score, and had then done nothing with it; Gevaert had then taken the opera to the Théâtre-Lyrique.15 Les empiriques was composed during Gevaert’s period in Italy during the travels incumbent upon a Prix de Rome laureate: the manuscript is in an Italian oblong format largely unknown in Paris, and the dates at the end of certain compositions attest to this.16 But despite the fact that Georgette was produced in Paris, the manuscript itself, in its mixture of formats, suggests that Gevaert wrote the first seven numbers in Italy (or at least in an Italian manuscript format) and the finale in Paris, thus strengthening the links – visible in Table 4.2 – between the two works.17 The anatomy of the three works is very similar. Most of the kinetic action takes place in the spoken dialogue amongst which the composed music is placed, and there is no formalised dance anywhere. All three works end with finales made up of multiple sections which are significantly more ambitious than many finales in one-act compositions at the Théâtre-­Lyrique or Opéra-Comique that consist merely of a chœur général. Two of the three operas have elaborate, multisectional morceaux d’ensemble, whereas Georgette’s morceau d’ensemble is monosectional but exploits complex mélodrame at its end. There is no freestanding instrumental music, and all three works balance out solos and small ensembles (duets and trios) in different ways, reflecting the dramaturgy of the libretto.

14 Fétis, Biographie universelle des musiciens, 3:471–472. Given the lack of a libretto for Les empiriques, judging which of the one-act operas at the Théâtre-Lyrique might have forced it out because the subject matter was similar is a dangerous undertaking. Possible candidates are Déjazet’s Le mariage en l’air (26 January 1852), Adam’s La poupée de Nuremberg (21 February 1852), Gautier’s Flore et Zéphire (2 October 1852) and his Choisy-le-Roi (14 October 1852). See Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981), 301. 15 ‘Ce début devait avoir lieu à l’Opéra-Comique, mais M. Perrin, auquel son cahier des charges n’impose pas à l’égard des lauréats de l’Institut de Belgique les obligations qu’il a à remplir envers les lauréats français, se borna à recevoir la partition que lui présenta M. Gevaert, et en ajourna indéfiniment la représentation. M. Gevaert, las de passer ses journées dans l’antichambre de M. le directeur, s’adressa à la bienveillance plus hospitalière de M. Seveste’ (La revue de Paris, 15 December 1853). 16 B-Bc 38633. 17 B-Bc 80833.

138  The operas of François-Auguste Gevaert Reaction to Georgette was in general positive. For Reyer, it was a Parisian premiere that not only demonstrated Gevaert’s talent, but also his ability to judge the tone of opéra comique perfectly: One finds in Georgette all the essential qualities of the genre bouffe: verve, simplicity, melodic grace and rhythmic clarity; none of which stops, under this charming and delicate tissue, the recognition of the powerful touch, the elevated style, the breadth and the majesty of a great dramatic composer.18 Reyer’s opposition between a ‘tissu charmant et délicat’ and ‘la touche puissante, le style élevé’ framed a tension that recurs throughout commentaries on Gevaert’s music for the stage. For some critics, even in Georgette, ‘la touche puissant, le style élevé’ – embodied in orchestration and the amount of music in the score – could be seen as too much. Although Jules Lovy was able to say that ‘the music of Georgette lively, melodious and emerges from the Italian school’, and that ‘the orchestration seems excellent to us’, he also suggested that it ‘sins through too much sonority, since it often crushes the voice. Let us also say that there is too much music for a single act, and that deft sacrifices would have made everybody happy’.19 The two reservations are important ones: the risks of too heavy orchestration and too great density of musical compositions. Adolphe Giacomelli even argued for cuts in the duo ‘Le cœur me bat’.20 But even this level of balance between ‘style élevé’ and the ‘tissu charmant et délicat’ had cost Gevaert a good deal of effort in the later stages of work on Georgette. The autograph is riddled with cuts:21 complete preludes and postludes excised in Georgette’s opening couplets ‘C’est moi qui suis la meunière’ and in Clovis’ chanson ‘Vive la goguette’. All the musique de scène in the finale was rewritten and cuts introduced, and the morceau d’ensemble ‘Attaquons, attaquons’ (a composition that was much admired) was similarly treated.22 A single cut just after the very first exclamation ‘Attaquons!’ is of ten bars, but at the end of the first ensemble, just before Clovis’ words

18 ‘On trouve dans Georgette toutes les qualités essentielles au genre bouffe: la verve, la simplicité, la grâce mélodique et la franchise du rhythme; ce qui n’empêche pas, sous ce tissu charmant et délicat, de reconnaître la touche puissante, le style élevé, l’ampleur et la maëstria d’un grand compositeur dramatique’ (La revue de Paris, 15 December 1853). 19 ‘La musique de Georgette est vive, mélodieuse et procède de l’école italienne…. l’orchestration nous semble excellente…., pèche par trop de sonorité, puisque souvent elle écrasé les voix. Disons aussi qu’il y a trop de musique pour un simple acte, et que d’habiles sacrifices eussent contenté tout le monde’ (Le ménestrel, 4 December 1853). 20 ‘Le duo: Le cœur me bat, est un morceau intéressant, la mélodie y est douce, conduite par le goût et les accompagnements con sordini y produisent un effet charmant. La deuxième reprise sur ces mots: Faut-il déjà partir est habilement amenée; la troisième qui ramène la phrase principale pour la cinquième fois est de trop; nous engageons l’auteur à la supprimer’ (L’Europe artiste, 4 December 1853). 21 B-Bc 80833. 22 For praise for this number, see L’Europe artiste, 4 December 1853; see A LA VILLE DE GAND. / GEORGETTE / ou le Moulin de Fontenoy, 55–63.

The operas of François-Auguste Gevaert  139 ‘Je n’entends rien’ is a cut of forty-five bars. More significantly, the mélodrame at the end of the number is so short that it is difficult to imagine how two full pages of dialogue in the libretto could be made to fit over or between the music composed by Gevaert (Example 4.1).23

Example 4.1  Gevaert, Georgette, end of no. 5, Morceau d’ensemble ‘Attaquons’.

The great difficulties in this passage are amplified when it is realised that the last eight bars (a repetition of the previous eight) are an addition that are 23 BIBLIOTHEQUE DRAMATIQUE / Théâtre moderne. / GEORGETTE, 22–24.

140  The operas of François-Auguste Gevaert not found in the autograph. Although such contradictory treatment of this passage brings musique de scène and mélodrame into focus as key points of inquiry into Gevaert’s later works, the general point here is clear. When Gevaert arrived in Paris with Les empiriques and probably large parts of Georgette in his portfolio, he was starting a career at the Théâtre-Lyrique with works in a style in which he was already expert, and on which – in terms of the construction of their finales at least – he had already imprinted a very personal stamp.

Musique de scène and Mélodrame: the Théâtre-Lyrique Gevaert followed the success of Georgette with two works in three acts at the Théâtre-Lyrique. The first of these was Le billet de Marguerite, an opéra comique in which the eponym is repeatedly abandoned by family and ­fiancée, despite the written promise of the latter – the billet of the title – but which all turns out happily in the end, with a long-lost half-sister found along the way. Gevaert develops the character of Marguerite through the use of reminiscence motives, an understanding of which is important not only for the internal functioning of Le billet de Marguerite but also for the history of Gevaert’s operas during the 1850s. In the act i finale of Le billet de Marguerite, Marguerite remembers her miserable fate as an orphan twice: the first time to celebrate her joy at finding a lover, Reinhold, and the second to lament his departure to seek his fortune. Both are identical, and Example 4.2 gives the second of the two.24

Example 4.2  Gevaert, Le billet de Marguerite, extract from no. 5, act i finale ‘Partir! Qu’ai-je entendu?’ 24 A M. FERDINAND DU CHÈNE / LE BILLET / DE / MARGUERITE / OPÉRA-­ COMIQUE EN 3 ACTES / Représenté pour la première fois, à Paris, au Théâtre-Lyrique, le 7 Octobre 1854 / PAROLES DE / MM. LEUVEN ET BRUNSWICK / MUSIQUE DE / F.A. GEVAERT / … / PARIS / PROPRIÉTÉ DES ÉDITEURS POUR TOUS PAYS / CHEZ HENRY LEMOINE RUE SAINT-HONORÉ, 256 / MAISON LEMOINE AINÉ, HARAND, Sr / RUE DE L’ANCIENNE-COMÉDIE, 20, 95–96.

The operas of François-Auguste Gevaert  141 The text to the passage is ‘Depuis que j’ai perdu ma mère / Trésor de soins et d’amitié / Lui seul m’a témoigné sur terre / De l’intérêt de la pitié’. This is a powerful reminiscence within the act i finale, but Gevaert brings it back with even greater effect in the middle of the second act, in a mélodrame which features the central confrontation between Marguerite and Berta, her rival for Reinhold’s affections (Example 4.3).25 The same music as in Example 4.2 is clearly visible. But the text, ‘Depuis j’ai perdu ma mère’, is missing, and the music becomes the underscore for the mélodrame. The spoken text supported by this music is significantly longer than the published score suggests – it just gives– as is normal – enough of the text to coordinate words and music.26 So underneath this musical reminiscence is the following monologue as the servant Marguerite realises that Berta is Reinhold’s new aristocratic lover (parts of the text indicated in the published score are underscored): Mam’zelle, je vous demande bien pardon… je ne savais pas… j’ignorais… oh! Mais ne craignez rien, je ne veux pas vous inquiéter… vous faire souffrir… moi, j’en ai l’habitude… je vais partir, sortir de la ville… j’irai bien loin… il ne me verra pas. Adieu, mamzelle, adieu… gardez ce papier… déchirez-le… moi partie, rien ne vous rappellera qu’avant vous, il a aimé la pauvre Marguerite Muller.27 Naming herself at the end of the passage triggers the recognition of Berta and Marguerite as half-sisters, but more importantly, the self-pitying tone of Marguerite’s monologue just quoted fits perfectly under the music whose original text is similarly self-pitying. Finally, before the third act, the entr’acte makes use of the same music, but this time in an entirely instrumental context (Example 4.4).28 To summarise the dramaturgical use to which Gevaert puts this musical passage: it occurs twice in the act i finale, sung to a text by the eponym; it then appears as the underscore to an extended mélodrame in act ii and introduces the third act in a purely instrumental guise. It invites comparison with one of the best-known opéras comiques of the decade. Gevaert’s Le billet de Marguerite was premiered at the Théâtre-Lyrique in October 1854; the act i finale was completed in August that year and the rest of the work between August and the premiere in October. February of the same year had seen the biggest event – in every sense of the word – at the Opéra-Comique for a very long time. Meyerbeer, in his first opéra comique, L’étoile du nord, had thrown down the gauntlet to the genre by

25 Ibidem, 144–145. 26 This difficulty also underpins the musique de scène in Example 4.1. 27 LE BILLET / DE / MARGUERITE / OPÉRA-COMIQUE EN TROIS ACTES / PAR MM. A. DE LEUVEN ET BRUNSWICK / MUSIQUE DE / M. GEVAËRT / Représenté pour la première fois, à Paris, sur le Théâtre-Lyrique / le 7 octobre 1854 / … / PARIS / MICHEL LEVY FRÈRES, LIBRAIRES-ÉDITEURS / RUE VIVIENNE, 2 bis. / - / 1854, 45. 28 A M. FERDINAND DU CHÈNE / LE BILLET / DE / MARGUERITE, 196–197.

Example 4.3  Gevaert, Le billet de Marguerite, extract from no. 8bis, Mélodrame.

The operas of François-Auguste Gevaert  143

Example 4.3  (Continued)

writing a work that went beyond its generic boundaries in just about every way possible. One of these was in the use of reminiscence. This is not to say that reminiscence is found nowhere else in opéra comique, but in L’étoile du nord is found not only a very large level of reminiscence, but also exactly the same changes of discursive mode found in Gevaert’s Le billet de Marguerite. Meyerbeer identifies Catherine’s recollections of her mother’s words with a passage of mélodrame which he then reprises at the end of the act iii finale. In the meantime, however, exactly the same music is used – but this time with a text – for Catherine’s prayer in the act i finale, and it is also found – exclusively instrumentally – in the overture. In other words, Meyerbeer – in the same way as does Gevaert – translates his musical reminiscence from one discursive mode to another – from aria, to underscore for mélodrame – to instrumental music. The fact that both sets of reminiscences are about mothers, and that Gevaert started writing Le billet de Marguerite in the middle of L’étoile du nord’s first stupendous run (88 performances in 1854,

Example 4.4  G  evaert, Le billet de Marguerite, extract from act iii entr’acte.

The operas of François-Auguste Gevaert  145

Example 4.4  (Continued)

76 in 1855), makes it impossible to divorce Gevaert’s dramaturgy from his response to Meyerbeer’s L’étoile du nord. Gevaert had managed to keep the score of Le billet de Marguerite down to manageable proportions. Criticism of any mismatch between generic and institutional frame was rare. Only Théophile Gautier declared that ‘His opera is much too long’,29 although Jules de Prémary suggested that ‘M. Gevaert, breaking the narrow frame in which one wanted to keep him, has written broad and dramatic music, which revealed his male qualities, but which made the error of not being appropriate’, but went on to observe that ‘but what is that to M. Gevaert? A musician possessed by the demon of his art would write the fourth act of Les Huguenots based on the memoirs of a laundress’.30 De Prémary went on to make a more specific allusion to Meyerbeer when he praised ‘a number that Meyerbeer would have willingly signed, and it is astonishing that a composer, at the beginning of his career, could put as much science to the service of his originality’.31 What neither Gautier nor de Prémary knew was that the original version of Le billet de Marguerite had already been extensively cut by the composer. In addition to tightening of several numbers (Jacobus’ act iii 29 ‘Son opéra est beaucoup trop long’ (La presse, 17 October 1854). 30 ‘M. Gevaert, brisant le cadre étroit dans lequel on voulait l’enfermer, avait écrit une musique large et dramatique, qui faisait deviner ses mâles qualités, mais qui avait le tort de ne pas être en situation’, ‘Mais qu’est-ce que cela faisait à M. Gevaert ? Un musicien possédé du démon de son art écrirait le quatrième acte des Huguenots sur les mémoires de sa blanchisseuse’ (La patrie, 16 October 1854). 31 ‘un morceau que M. Meyerbeer signerait des deux mains, et on s’étonne qu’un compositeur, au début de sa carrière, puisse mettre autant de science au service de son originalité’ (ibidem). The reference must be to the trio bouffe for the three Anabaptists in Meyerbeer’s 1849 Le prophète, still in the middle of an unbroken first run that would last until 1863.

146  The operas of François-Auguste Gevaert couplets ‘Allez, mon jeune compère’, for example), two complete numbers were excised from the score, and survive only in the autograph: a romance for Tobias in act ii ‘Ah! Laissez moi’ and couplets for Berta in act iii ‘Evitez pour le mariage’. Texts for these numbers are absent from the printed libretto, and it may be assumed either that de Leuven and Brunswick had these numbers in their original libretto and then excised after Gevaert had composed (in which case experienced librettists would have misjudged the dimensions of their work), or that Gevaert had requested these numbers himself. In either case, the evidence points to a composer creating greater amounts of music than needed for the genre in which he was writing, a feature that would surface again, but with greater significance, in Les lavandières de Santarem and especially Quentin Durward. The only exception to this practice lies in exactly those areas to which Gevaert attached so much importance: musique de scène and mélodrame. The musique de scène in the act i finale (Example 4.2) consisted in Gevaert’s original version of two bars for flute and violas and was replaced by the music cited in Example 4.2.32

Towards the Opéra-Comique As Gevaert was moving from writing one-act opéra comique to more ambitious three-act works for the Théâtre-Lyrique, he was at the same time looking along the Boulevard to the Salle Favart and at the most elaborate and ambitious opéra comique of its generation, seeking to emulate it. It is hardly surprising that when Gevaert wrote his first work for the Opéra-Comique four years later, Quentin Durward, it was very much in the mould of ­Meyerbeer’s L’étoile du nord. The move from writing one-act opéra comique to works in three acts –a typical move for every successful debutant composer – entailed a change of dimensions, but not so obviously a change of tone. This is clear from the fact that in his third opera for the Théâtre-Lyrique, Les lavandières de Santarem, Gevaert reused material from his now-abandoned Les empiriques. A passage from the morceau d’ensemble (no. 2) ‘Chantons deux époux formant des nœuds si doux’ of Les empiriques 33 finds its way into the act i introduction to Les lavandières de Santarem (Example 4.5).34 Changed only to accommodate the new text, the music seems to move effortlessly between genres of a radically different cut, but of a similarly musical if not dramatic register.

32 The autograph of Le billet de Marguerite, where all information concerning the composer’s original version is found, is B-Bc 80822. 33 B-Bc 38633. 34 LES / LAVANDIÈRES / DE / SANTAREM / Opéra comique en trois actes, / Paroles de MM. / Dennery et Grangé / Musique de / F.A. GEVAERT. / … / PARIS, Alex GRUS ainé, éditeur / pour la France et l’Etranger, 31 Boulevart Bonne Nouvelle, 23.

The operas of François-Auguste Gevaert  147

Example 4.5  G  evaert, Les lavandières de Santarem, act i introduction, chœur des femmes ‘Joyeuse lavandière’.

Such continuities bring into focus Gevaert’s works for the Opéra-­Comique. Between Quentin Durward and Le chateau trompette, his two three-act works for the Opéra-Comique, Gevaert returned to the composition of opéra comique in one act with Le diable au moulin. In some respects, Gevaert backs away from some of the more ambitious features of his earlier one-act opéra comique, especially in the way he handles the finale, which in Le diable au

148  The operas of François-Auguste Gevaert moulin is a single uncomplicated ensemble général.35 But in other ways, even this one-act opéra comique exhibits a transgressive ambition more familiar from, say, Quentin Durward. There are eleven numbers, as opposed to the seven or eight in earlier works. Only four of these are solos, and there are no less than three large concerted numbers: two quartets and a quintet, whereas in the three early works there was never more than one. There is a definite break with past practice here, and Le diable au moulin may be read with profit not only against the other three-act works for the Opéra-­Comique, but also against the three-act compositions for the Théâtre-Lyrique. When placed against a contemporary one-act opéra comique, for example, the position becomes even clearer. In such a work as Potier’s Le rosier of August 1859, premiered just a few weeks after the premiere of Le diable au moulin,36 duos and trios constitute half the eight numbers in the work, and the overture and finale are ambitious, multisectional compositions. But the eight numbers in Le Rosier come nowhere close to the eleven of Le diable au moulin, and in this regard the latter stands out against the repertory of the early 1860s as Georgette, Les empiriques and La comédie à la ville were projected against the backcloth of opéra comique from the early 1850s.

Towards the Opéra Almost as soon as Les lavandières de Santarem was completed, it seems as if Gevaert was looking towards the epicentre of Parisian operatic culture, the Théâtre Impérial de l’Opéra, and he was approaching this institution in two ways. Most obviously, if Sardou’s account is to be trusted, he had written an opera based on a libretto by Vaëz and Scribe, originally in five acts and reduced to three, which was complete by September 1856 with the expectation of performance at the Opéra in September 1857.37 Together with the two other opéras comiques on which he was apparently engaged, this explains the apparent gap of two and a half years between the premieres of Les lavandières de Santarem and Quentin Durward. Nothing apparently survives of either libretto or score, but it may be assumed that it was a work with continuous recitative, and even in a version reduced to three acts, a substantial divertissement. At the same time, however, Gevaert was continuing to take opéra comique in the direction that Meyerbeer had established with L’étoile du nord by expanding its dimensions. Quentin Durward represents the apogee of this development and resembles exactly what one might expect from a composer writing his first work for the Opéra-Comique at the same time as his first

35 B-Bc 80832. 36 THÉÂTRE IMPÉRIAL DE L’OPÉRA COMIQUE / LE ROSIER / OPÉRA COMIQUE EN UN ACTE / PAROLES DE MM / AUGIN CHALAMEL & *** / Musique de / HENRI POTIER / PARIS / AU AMAGASIN DE MUSIQUE DU CONSERVATOIRE / 11, faubg Poissonnière, 11 [sic]. 37 See the sources cited earlier, 135.

The operas of François-Auguste Gevaert  149 38

work for the Opéra. Throughout the commentaries on Les lavandières de Santarem, the press had commented in the abstract on Gevaert’s’ science and specifically how this had an effect on his orchestration. Henry Boisseaux was convinced that Gevaert, ‘sure of all his ingenious and brilliant effects, manipulates his orchestra in a magisterial fashion’,39 and this view was seconded by the Revue et Gazette musicale de Paris.40 Even the correspondent for an exclusively theatrical journal was impressed by Gevaert’s scoring: Gevaert’s orchestration is rich, elegant and well-nourished. He knows, and he makes it seen without indulging in attempts at new effects, which are often understood only by their author. It is sparkling and grandiose when it must be; above all, it is clear and pleasing.41 And while Marie Escudier declared Gevaert ‘at the same time a composer of science and instinct’,42 de Prémary was concerned that Gevaert could become too concerned with orchestral effect: He plays the orchestra as securely as Meyerbeer. But already he seems to be much too preoccupied with the effect on the public. It’s a danger I’m warning him of. By being subservient to the poorly enlightened taste of the masses, one can have great success, earn money, but one risks losing one’s originality. As for orchestral science, the mysteries of harmony, Mr. Gevaert is a master, Mr. Gevaert is initiated. But let him not forget this beautiful muse called melody, which was the smiling and sweet fairy of his artist’s cradle.43 Whether de Prémary also thought that Meyerbeer was at risk of similarly being preoccupied with orchestral effect is not clear, but he was in no doubt 38 For an account of the detail of the score of Quentin Durward, and its generic ambiguity, see Mary Jean Speare, ‘Gevaert’s Quentin Durward: Opéra comique or opéra?’, Revue belge de musicologie, 64 (2010) 49–56. 39 ‘sûr de tous ses effets, ingénieux et brillant, il manie son orchestre d’une façon magistrale’ (Le messager des théâtres et des arts, 28 October 1855). 40 ‘His orchestra is full of ingenious effects’ (‘Son orchestre et rempli d’effets ingénieux’; Revue et Gazette musicale de Paris, 28 October 1855). 41 ‘L’orchestration de Gevaert est riche, pimpante et bien nourrie. Il sait, et il le fait voir sans se livrer à plaisir à des tentatives d’effets nouveaux, qui souvent ne sont compris que par leur auteur. Il est sémillant et grandiose quand il le faut; avant tout, il est clair et fait plaisir’ (Revue et Gazette des théâtres, 28 October 1855). 42 ‘à la fois un compositeur de science et d’instinct’ (La France musicale, 28 October 1855). 43 ‘Il joue de l’orchestre avec autant de sûreté que Meyerbeer. Mais déjà il semble se préoccuper beaucoup trop de l’effet à produire sur le public. C’est un danger que je lui signale. En s’asservissant au goût peu éclairé des masses on peut avoir de grands succès, gagner de l’argent, mais on risque de perdre son originalité. Pour ce qui est de la science orchestrale, des mystères de l’harmonie, M. Gevaert est passé maître, M. Gevaert est initié. Mais qu’il n’oublie pas cette belle muse qu’on appelle la mélodie, et qui a été la fée souriante et attendrie de son berceau d’artiste’ (La Patrie, 5 November 1855).

150  The operas of François-Auguste Gevaert about Gevaert’s skills in this direction. His reservations concerning the ‘belle muse qu’on appelle la mélodie’ were ones that inevitably followed praise for a skill – or science – in orchestration. So if Gevaert’s science and skill in orchestration were much praised after the premiere of Les lavandières de Santarem, there was little indication that it did not belong at the Théâtre-Lyrique. In praising the variety of styles and registers found in the work, however, Jules Lovy pointed to three elements: ‘the dramatic éclat of grand opéra, the coquettish allure of opéra comique and the voluptuous insouciance of ballet’. Ballet will be discussed later, but even here Lovy was only hinting at the presence of the style of grand opéra in Les lavandières de Santarem. This was not the case for his first attempt at the Opéra-Comique, Quentin Durward, which was widely recognised as a work either better suited to the Opéra or one that clearly identified the composer as a candidate for the institution. Many critics, Alberic Second, Lovy, Gustave Héquet and Etienne Saint-Sylvain, simply noted the fact that Quentin Durward was a work that should have been played at the Opéra.44 Lovy counted no less than thirty compositions in the work: And what is honourable to say, what is sad to say, is that all these numbers, all these choruse, all these duets, all these trios, all these romances, all these airs, are written with conscientious care, developed with consummate skill. I would regret the slightest cut [emphasis added].45 Relatively few saw the disproportionate size of Quentin Durward as a disadvantage. The critic calling himself ‘Faust’ noted that the performance had started at 7.30 p.m., that it had not finished until 1.00 a.m. and that it would have been too long even if it had been a question of a grand opéra with divertissements and visual splendour.46 Frank-Marie (Franco-Maria Pedorlini), writing in La patrie, thought it all too much: There is no action, or it is too slow. Many recitatives, replacing dialogue, slow it down further. The actors speak little and sing a lot; they sing too much. It is not longer thus an opéra comique, as the title suggests, but a genuine opéra.47 4 4 L’entr’acte, 27 March 1858; L’illustration, 3 April 1858; L’union, 8 April 1858. 45 ‘Et ce qui est honorable à dire, ce qui est triste à dire, c’est que tous ces morceaux, tous ces chœurs, tous ces duos, tous ces trios, toutes ces romances, tous ces airs, sont écrits avec un soin consciencieux, développés avec une habileté consommée. Je regretterais la moindre coupure [emphasis added]’ (Le ménestrel, 28 March 1858). 46 Le théâtre, 4 April 1858. 47 ‘L’action ne marche pas, ou ne marche pas assez vite. De nombreux récitatifs substitués au dialogue, la retardent encore. Les acteurs parlent peu et chantent beaucoup; ils chantent trop. Ce n’est plus ainsi un opéra-comique, comme l’indique le titre, mais un opéra véritable’ (La patrie, 30 March 1858).

The operas of François-Auguste Gevaert  151 Pier-Angelo Fiorentino, writing as A. de Rovray in the Moniteur universel, saw the lack of humour as a decisive factor in the misalignment between work and institution: ‘Nowhere in these three acts is there the tiniest word to make you laugh’.48 Others, hardly surprisingly, saw the size of Quentin Durward as an advantage, and the possibility of success on the rue Le Peletier as something to await with interest.49 Various reasons were advanced for Quentin Durward’s suitability for the Opéra that went beyond mere size. The quantity of recitative was noted both by Franck-Marie (who saw it simply as a substitute for dialogue)50 and by Léon Leroy who considered it ‘handled with a breadth and a majesty worthy of the Opéra’.51 Fiorentino pointed, in addition to the choruses and morceaux d’ensemble, to Quentin Durward as types of composition with no real home at the Opéra-Comique.52 He also saw the act i finale as a key moment that pointed to the Opéra.53 But there was a sense that such a subject as Quentin Durward could only be treated in a form suitable for the Opéra. Gustave Chadeuil was in no doubt: It was impossible for M. Gevaert not to write a grand opéra. His imagination had to be ignited by the memory of these characters from a fine book. He had to evoke to the sombre walls of Plessis-lez-Tours; he had to represent Louis XI with his devious look, and Tristan l’Ermite with his hands stained with blood; and the Ducesse de Croy, frail dove fallen in the clutches of a vulture; and Crèvecour, of the haughty demeanour and Quentin Durwald with a heart taken in love. So, it was necessary to take his most colorful notes, and to extend them on his palette, and to paint canvases with broad strokes. That’s why this opéra comique is an opéra. The dialogue barely holds the place usually occupied by recitatives, ­music abounds, the phrases have breath, everything in vast proportions.54 48 ‘Il n’y a point dans ces trois actes le plus petit mot pour rire’ (Le moniteur universel, 28 March 1858). 49 See Le constitutionnel, 5–6 April 1858 (Fiorentino); Le courrier français, 4 April 1858 (Adolphe Huard). 50 La patrie, 30 March 1858. 51 ‘traité avec une ampleur et une majesté digne de l’Opéra’ (L’europe artiste, 4 April 1858). 52 Le moniteur universel, 28 March 1858. 53 ‘Le finale est très bien fait, puissamment orchestré, et prouve que M. Gevaërt a de très heureuses dispositions pour le grand Opéra’ (Le constitutionnel, 5–6 April 1858). 54 ‘Il devenait impossible à M. Gevaërt de ne pas écrire un grand opéra. Son imagination devait s’échauffer au souvenir de ces personnages d’un beau livre. Il devait évoquer les sombres murailles de Plessis-lez-Tours; il devait se représenter Louis XI au regard sournois, et Tristan l’Hermite [l’Ermite] avec ses mains teintes de sang; et la duchesse de Croy, frêle colombe tombée dans les serres d’un vautour; et Crèvecœur, le front hautain, et Quentin Durward, le cœur épris. Alors, il fallait prendre ses notes les plus colorées, et les étendre sur sa palette, et peindre des tableaux à larges traits. Voilà pourquoi cet opéra-comique est un opéra. Le dialogue y tient à peine la place habituellement occupée par les récitatifs, la musique abonde, les phrases ont de l’haleine, tout à de vastes proportions’ (Le siècle, 30 March 1858).

152  The operas of François-Auguste Gevaert There is a complex mix of images in play here, but the argument is logical: the personalities of Scott’s novel are simply too colourful to be framed by an opéra comique, and dialogue (here Chadeuil echoes Leroy) of necessity has to be replaced by recitative. Gevaert and his librettists were, however, highly sensitive to the conventions that governed the Opéra-Comique, and only two years later, critics identified Le château trompette as a return to the real spirit of opéra comique, which had been much under pressure during the 1850s as a result of such works as L’étoile du nord and Quentin Durward itself. Saint-Sylvain put the position clearly: The Opéra-Comique, temporarily diverted from its path, seems to want to returnto it from now on, to judge by the new play given to us by Messrs. Carré, Cormon and Gevaert. We will not complain about it. We have protested enough, in the past, against spectacular pieces L’ Etoile du Nord, Le Carillonneur de Bruges, Raymond, and so many others that seemed to want to compete with the méldorame of the boulevard…. This time, Mr. Gevaert was at his ease; he had not been given these energetic and powerful situations that require a great deal of sonic forces and violent tactics. Also his music is coquette, pleasant, lively, cheerful…. It is the real music of opéra comique. Of a pleasant and easy construction, no heaviness in the melody nor in the instrumentation, often wit, always knowledge.55 Critics muddled plot locations with musical style, so keen were they to bring Gevaert within the fold of traditional opéra comique, ‘the sanctuary of Grétry, de Méhul et d’Auber’ as Léon Escudier put it.56 Thus, Lovy pointed to Le billet de Marguerite as a ‘Lied Allemand’, the ‘cantilènes de ­l’Estrémadure’ in Les lavandières de Santarem and Quentin Durward’s ‘larges accents’ as the cosmopolitan background to ‘Château Trompette, un opéra-comique pur sang, une œuvre toute sémillante, toute française, toute bordelaise’. Later in the same review, the critic’s pleasure seemed without limit: ‘Speak to me of the M. Gevaert’s score! It is, as I told you, an opéra comique in all the grace, all the freshness, or the meaning of the term’.57

55 ‘L’Opéra-Comique, un moment déraillé de sa voie, paraît vouloir désormais y rentrer, si nous en jugeons par la nouvelle pièce que nous ont donnée MM. Carré, Cormon et ­Gevaert. Ce n’est pas nous qui nous en plaindrons. Nous avons assez protesté, dans le temps, contre des pièces à grands fracas, telles que L’Etoile du nord, Le Carillonneur de Bruges, Raymond, et tant d’autres qui semblaient vouloir faire concurrence aux mélodrames du boulevard…. Cette fois, M. Gevaert était à son aise; on ne lui avait pas donné de ces situations énergiques et puissantes qui demandent un grand déploiement de forces sonores, de procédés violens. Aussi sa musique est coquette, agréable, vive, enjouée…. C’est une vraie musique d’opéra-comique. Facture agréable et facile, pas de lourdeur dans le chant ni dans l’instrumentation, de l’esprit souvent, du savoir toujours’ (L’union, 4 May 1860). 56 ‘Le sanctuaire de Grétry, de Méhul et d’Auber’ (La France musicale, 29 April 1860). 57 ‘Parlez-moi de la partition de M. Gevaert! C’est, comme je vous l’ai dit, un opéra-­c omique dans toute la grâce, dans toute la fraîcheur de son acception. C’est vif, c’est gai, c’est

The operas of François-Auguste Gevaert  153 There were also voices that expressed dismay at this stylistic turn, but only those who were disillusioned with opéra comique of the previous decade in general,58 but there were also those who understood the dynamic of comic opera in Paris around 1860 much better. Their point of reference was not the opéra comique of Auber or even Grétry, but the opérette of Offenbach. Chadeuil thought ‘to hear the best Offenbach. This has been successful beyond expectations. It is that instead of addressing a few, like Meyerbeer’s L’étoile du nord; Victor Massé’s Les saisons; Halévy’s Valentine d’ Aubigny, Le carnival de Venise by Ambroise Thomas; Les trois Nicolas; and what else? It is addressed to the masses, like Ba-ta-clan, like Orphée aux enfers, like Le carnaval des revues, like Daphnis et Chloé, but let us hurry to say it, in a higher order’.59 This was a very perspicacious view, given Offenbach’s attempts to appropriate eighteenth-century comic opera as a way of legitimising the repertory of the Théâtre des Bouffes-Parisiens, and the recognition that Gevaert might have more in common with the Salle Choiseul than with the Salle Favart was prescient in the extreme.60 But Charles Delosme, while praising Offenbach who ‘although German, knows how to write cheerful music’ encouraged Gevaert to give ‘new sisters to the Lavandières’ [Les lavandières de Santarem], to Quentin Durwald, and since he has the energy, let him leave to others lyric movement, the delicacy of appearance, the sparklin side’. For Delosme, Gevaert had taken a wrong track because only Italians were able to ‘faire de l’esprit français’ and that ­Gevaert’s personal nature – a Belgian – would always work to his disadvantage.61

The Belgian network Gevaert’s nationality was an important issue for the reception of his operas in the Paris of the 1850s. His humble origins and background as an autodidact were so constantly rewritten, especially in the wake of the success of Georgette, that they took on the status of myth. It took little to develop the few scattered facts that were known about Gevaert’s early career and turn them into legend. De Prémary spoke for many: ‘He [Gevaert] was born

58 59

60 61

mélodieux et coloré’ (Le ménestrel, 29 April 1860). Similar strategies were employed by Gaston de Saint-Valry (Le pays, 30 April 1860) and Chadeuil (Le siècle, 1 May 1860). The single example is an anonymous review in the Journal des débats, 5 May 1860. ‘entendre du meilleur Offenbach. Cela a réussi au-delà même des prévisions. C’est qu’au lieu de s’adresser à quelques-uns, comme l’Etoile du nord, de Meyerbeer; Les saisons, de Victor Massé; Valentine d’Aubigny, d’Halévy; le carnaval de Venise, d’Ambroise Thomas; Les trois Nicolas; que sais-je encore? cela s’adresse aux masses, comme Ba-ta-clan, comme Orphée aux Enfers, comme le Carnaval des revues, comme Daphnis et Chloé, mais hâtonsnous de le dire, dans un ordre plus élevé’ (Le siècle, 1 May 1860). For Offenbach’s attempts at the appropriation of comic opera from the eighteenth century, both opéra comique and works from abroad, see earlier, Chapter 3. ‘sait écrire de la musique gaie quoique Allemand…. ‘de nouvelles sœurs aux Lavandières, à Quentin Durward, et puisqu’il a l’énergie, le mouvement lyrique, qu’il laisse à d’autres la grâce sémillante, la délicatesse d’allure, le côté pétillant’ (L’Europe artiste, 29 April 1860).

154  The operas of François-Auguste Gevaert as a musician like warblers and gossipy roe deer; he proved to be an artist like Giotto and Lantara, in the middle of nature, in the middle of fields, sometimes guarding the sheep, more often than not losing them by giving himself over to his poet’s daydreams, sometimes leading the plough while stuttering some naïve melody’. The analogies with Giotto and Lantara are telling: ­Giotto’s early life – as related by Vasari – is well known.62 Less well known today is the almost identical early career of Simon-Mathurin Lantara, a young shepherd whose susceptibility to nature was the basis for an artistic career.63 But the image of the self-taught Belgian, lonely in Paris, fits ill with the well-developed network of Belgian musicians in the city which Gevaert himself may well have helped foster.64 When he arrived, his most noteworthy compatriots were the composers Limnander, Grisar and Fauconnier, instrumentalists Bériot, Vieuxtemps, Servais and Batta,65 to say nothing of the soprano Marie-Josèphe Cabel and her ex-husband, the baritone Louis-­Joseph Cabel.66 Between them, they appeared in three of Gevaert’s operas: L ­ ouis-Joseph as Corbin in Georgette and Jean V in Les lavandières de ­Santarem, and ­Marie-Josèphe as Lise in Le chateau trompette. But, for Gevaert, more important than any of these was Vaëz, resident in Paris since the late 1830s, but also the librettist for two of his three one-act opéras comiques. This inner circle of Belgians was soon amplified by the arrival of Prilleux and Pauline Deligne-Lauters.67 Prilleux had been the librettist for the first of Gevaert’s 62 ‘Il [Gevaert] est né musicien comme les fauvettes naissent cantatrices et les bouvreuils bavards; il s’est révélé artiste comme le Giotto et Lantara, en pleine nature, en pleins champs, tantôt gardant les moutons, le plus souvent les perdant en se laissant aller à ses rêveries de poète, tantôt conduisant la charrue en bégayant quelque naïve mélodie’ (La patrie, 16 October 1854). For a sense of how Giotto’s childhood and early career were received in the mid-nineteenth century, see Grand Dictionnaire universel du xixe siècle français, historique, géographique, biographique, mythologique, bibliographique, littéraire, artistique, scientifique, etc., 17 vols (Paris: Grand Dictionnaire Universel, 1866), 8:1265. 63 See, for an account of Lantara that would have been known to de Prémary, Émile ­Bellier de la Chavignerie, Recherches historiques, biographiques et littéraires sur le peintre ­L antara, avec la liste de ses ouvrages, son portrait et une lettre apologétique de M. Couder (Paris: ­Dumoulin, 1852), and for a modern account of the mythology of his birth, George ­L evitine, ‘Les origines du mythe de l’artiste bohème en France: Lantara’, Gazette des Beaux-arts 113 (September 1975), 49–60. 64 Soubies and Malherbe noted that in 1859 there were four premieres at the Opéra-Comique alone by Belgian composers. In addition to Gevaert’s Le diable au moulin were Grisar’s Le voyage autour de ma chambre, Limnander’s Yvonne and Fauconnier’s La pagode (Histoire de l’Opéra Comique, 2:318). 65 There were endless such listings in the press. See, for example, La revue contemporaine, 1 October 1854 (Alphonse de Calonne); L’artiste, 18 April 1858 (Philarète Charles); Le Figaro, 26 April 1860 (Benoît Jouvin). 66 Fétis, Biographie universelle des musiciens, suppl. 1:137–138. 67 There is no entry for Deligne-Lauters ibidem. For a summary account, see Elizabeth Forbes, ‘Guéymard-Lauters, Pauline’, The New Grove Dictionary of Opera, ed.  ­Stanley  Sadie. Grove Music Online. Oxford Music Online, www.oxfordmusiconline.com/subscriber/­ article/grove/music/O901980 (Accessed 17 April 2009).

The operas of François-Auguste Gevaert  155 one-act opéras comiques and was to take roles in every work by Gevaert at the Théâtre-Lyrique and Opéra-Comique with the exception of Georgette and Le billet de Marguerite.68 Deligne-Lauters took the roles of Marguerite in Le billet de Marguerite and of Margarida in Les lavandières de Santarem. This group was amplified by a vociferous circle of Belgians who acted as an informal claque at Gevaert’s premieres.69 This could work to Gevaert’s disadvantage, as in the case of the premiere of Quentin Durward where Charles could say of Poniatowski’s Don Desiderio that ‘La partition a fait plaisir sans claqueurs et sans Belges. Ce n’est pas ce que l’on peut dire de M. Gevaërt et de Quentin Durward. Il y avait dans la salle de M. Gevaërt bien des Belges et bien des claqueurs’, and this was the trigger for an unbridled, but largely defensive, attack on the composer: For my part, I feel very strongly about Belgium, which has given us great artists of all kinds… But I would not want Belgium to trample us for too long and flatten us. I will kneel before his glory on my knees, but I will not bow my face to the earth. Let Mr. Gevaërt be remembered with loud cries, let him be carried in triumph, let him be serenaded, let his street be lit up, I consent to that; but I absolutely refuse to uncouple the horses from his carriage.70 This was not a unique backlash against the composer. His very appearance at the Théâtre-Lyrique had proved controversial because of his nationality. During the premiere of Le billet de Marguerite, according to Jules Collin, it was repeated often during the evening, at the same time as acknowledging the incontestable merit of the work, that the Théâtre-Lyrique had been created to represent the works of young French composers, and that it had strayed from its mission a little, since M. Gevaert is Belgian. 68 He took the roles of Pablo (Les lavandières de Santarem), Pavillon (Quentin Durward), Boniface (Le diable au moulin), Turcant (Le Chateau Trompette) and Pastorel (Le capitaine Henriot). Little is known of his biography, but see the correspondence with the director of the Opéra-Comique in Paris, Archives nationales, AJ13 1137. His first appearance at the Théâtre-Lyrique was in a revival of Auber’s La sirène, 19 June 1855 (see The Musical World, 30 June 1855, cited in Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981), 61), and his last, the revival of Adam’s Le sourd on 18 January 1856 (ibidem, 308). 69 ‘Presque tous les morceaux d’ailleurs, qu’ils fussent bons ou médiocres, ont fait éclater des applaudissements bruyants et des bis chaleureux, qui prouvent l’esprit national dont les Belges sont animés’ (L’illustration, 14 October 1854). 70 ‘Pour ma part, j’estime fort la Belgique qui nous a donné de grands artistes en tout genre … Seulement je ne voudrais point que la Belgique nous marchât trop longtemps sur la tête et nous aplatît à grands coups de talon. Je veux bien me mettre à genoux devant ses gloires, mais je ne veux point me prosterner la face contre terre. Qu’on rappelle M. Gevaërt à grands cris, qu’on le porte en triomphe, qu’on lui donne des sérénades, qu’on illumine sa rue, j’y consens; mais je me refuse absolument de dételer les chevaux de son fiacre’ (­L’artiste, 18 April 1858).

156  The operas of François-Auguste Gevaert Collin attempted to distance himself from such latent xenophobia by pointing to ‘the gorious collaboration that foreign artists had given our national art’.71 Boisseaux pointed to the same problem: that the Théâtre-­ Lyrique was ­destined for young composers and Prix de Rome laureates, not necessarily Belgians. And in his attempts to wish away the problem by pointing to the quality of Le billet de Marguerite, he could not fail to point to how when ‘young musicians who waited, in the hope that the door of the Théâtre-­Lyrique would end up by opening to them, see M. Gevaert pass before them, they have to murmur about it’.72 In the light of these difficulties, the ritual claim that all Belgians were good at writing choruses would have counted for little, and even Lovy’s quip about ‘Belgian’ could not neutralise the force of his comments.73

Institution, repertory and dance Opéra comique still remains one of the worst understood genres in nineteenth-­ century opera. Although most today acknowledge its mixture of spoken ­dialogue and composed music (and this is fundamental to the criticism of the dialogue and recitative in Quentin Durward discussed in the previous section), the residual sense, in use even as late as the 1850s, as a play with music is still strongly resisted today. Even worse understood is the range of music-dramatic discourses present in the genre. The use of accompanied recitative to precede a number is an important element in the genre’s discursive make-up, and this has long been recognised, but this has to be seen alongside mélodrame in a variety of guises, musique de scène, danced choruses, semi-­improvised dance and divertissements. And these discourses are given here in the full knowledge that – in most accounts – dance never figures in descriptions of opéra comique. For works at the Théâtre-Lyrique, however, all levels of dance are of real importance, and this importance throws light back onto the place of dance in opéra comique at the institution that bears the same name. Examples from the reminiscence scenes in Le billet de Marguerite have shown how mélodrame functions, but these gave no sense of just how enthusiastically Gevaert develops this discursive mode in his three-act opéras 71 ‘On a répété souvent dans la soirée, tout en s’inclinant devant le mérite incontestable de l’œuvre, que le Théâtre-Lyrique avait été créé pour représenter les ouvrages des jeunes auteurs Français, et qu’il sortait tant soit peu de son programme, puisque M [Gevaert] est belge …. la glorieuse collaboration qu’ont prise les artistes étrangers à notre art national’ (L’Europe artiste, 15 October 1854). 72 ‘les jeunes musiciens qui faisaient antichambre, dans l’espoir que la porte du Théâtre-­ Lyrique finirait par s’ouvrir pour eux, virent passer avant eux M. Gevaërt [Gevaert], ils durent en murmurer’ (Le messager des théâtres et des arts, 8 October 1854). 73 ‘C’est particulièrement dans les chœurs qu’excelle le maëstro Gevaert. Déjà la Belgique a eu cent échantillons de sa puissante aptitude pour les masses vocales. Il s’entend merveilleusement à grouper les timbres, à faire vibrer leur sonorité. Il est peu d’artistes en France, —et même en Belgique—qui manient et pétrissent la matière chorale mieux que lui, savez-vous? (pour parler belge)’ (Le ménestrel, 28 March 1858).

The operas of François-Auguste Gevaert  157 comiques. Musique de scène and Mélodrame are titles given to individual numbers in Gevaert’s output. The passage described in Example 4.3 is titled ‘No. 8bis Mélodrame’ in all sources for the opera, and, as the previous discussion makes clear, extensive dialogue takes place over the music. Furthermore, mélodrame is a technique that may surface in a wide range of contexts, as the earlier discussion of the finale of Georgette shows. The use of the term ‘musique de scène’ is not quite so clear. In the sense purely as music to accompany stage action, it appears in Le chateau trompette (no. 13bis); there is no dialogue, barely any stage action, and thus there is no need to attempt to coordinate stage action with the music (Example 4.6) beyond the réplique ‘Suis moi, Champagne, j’ai des ordres à donner’.74

Example 4.6  G  evaert, Le château trompette, no. 13bis, Musique de scène.

But in Les lavandières de Santarem, where there are no less than five separate pieces of musique de scène (2bis; 4bis; 4ter; 9bis; 11bis), in every case, the 74 A MONSIEUR DE COMTE DE MORNY / LE / CHATEAU TROMPETTE / OPÉRA COMIQUE / en trois Actes, / PAROLES DE / M.M. CORMON & M. CARRÉ / M ­ USIQUE / de / F.A. GEVAERT / … / PARIS A. GRUS, EDITEUR / Boulevart Bonne Nouvelle, No.31, 160. The dramatic context for this passage of musique de scène can only be recovered from the libretto (LE / CHATEAU-TROMPETTE. OPÉRA-COMIQUE EN TROIS ACTES / PAR / MM. CORMON ET MICHEL CARRÉ / MUSIQUE DE M. GEVAERT / … / PARIS / MICHEL LÉVY FRÈRES, LIBRAIRES-ÉDITEURS / RUE VIVIENNE, 2 BIS / - / 1860, 58.

158  The operas of François-Auguste Gevaert passage in question involves not only stage action but also the presentation of dialogue over the music – in short, mélodrame. The musique de scène from act i of Les lavandières de Santarem is a case in point (Example 4.7).75

Example 4.7  Gevaert, Les lavandières de Santarem, no. 2bis, Musique de scène.

As in the case of the mélodrame from Le billet de Marguerite, the piano-­ vocal score gives only a fraction of the story. The réplique is visible on the score: ‘Justement voici la fermière’. The stage direction states that Don Luiz ‘s’approche d’elle et l’embrasse’ and the fortissimo flourish at the beginning of the musique de scène are to coordinate with the kiss, since an identical 75 LES / LAVANDIÈRES / DE / SANTAREM, 79.

The operas of François-Auguste Gevaert  159 flourish is explicitly coordinated with a kiss in the penultimate bar. There is a significant amount of text that fits into the section marked ‘on parle’, which includes a scene change and the introduction of new characters. The music in bars two to six supports not only stage directions but spoken dialogue, as in mélodrame. The stage directions are ‘Bruit de voitures en dehors. Remontant et regardant à droite et à gauche’. And the dialogue, ‘Et tenez, les voilà qui descendent de voiture… Je cours donner des ordres pour qu’ils soient bien reçus’ is probably to be delivered above the sustained dominant seventh at the end of the third system, although the exact coordination is not clear from the libretto. The Duke then enters – and this is marked in the score, but the fragments of text above the bottom line of the score are merely for the purposes of coordination. Again, there is a significant amount of text delivered over the gently oscillating A major Andante which ends when the Duke kisses Térésa in the penultimate bar, triggering a second exclamation from her husband: ‘Qui est-ce qui embrasse ma femme?’.76 This explanation makes the point that Gevaert’s use of musique de scène does not merely provide a background for stage movement, but introduces ­elements of mélodrame, spoken dialogue, sometimes over sophisticated music, with the effect that the two discourses – musique de scène and ­mélodrame – become permeable. In addition to questions of mélodrame, musique de scène and their combinations, Les lavandières de Santarem raises questions concerning dance that are far-reaching in their significance. Its introduction consists of two choruses, a solo for Manoël and a duet for Margarida and Térésa (Table 4.3). Given the local colour of the libretto, the seguidilla and fandango are almost obligatory. But what is so important are the stage directions in the libretto that accompany the chœur de soldats and the fandango for Margarida Table 4.3  Gevaert, Les lavandières de Santarem – Introduction 1. Introduction A chœur des soldats B Cavatine

Manoël

C Chœur des femmes D Fandango à 2 voix

Margarida Térésa

3/4 Eb major, mouvement de séguidille ‘Dieu de la guerre’ 4/4 Ab major, moderato; 2/4 d minor, alltto grazioso ‘Joyeuse lavandière’ 2/4 d minor, alltto grazioso; D major) ‘Jadis la fermière 3/8 Bb major, mouvement Paquette’ de fandango + repetition of opening chorus ‘Pour nous guider’

76 LES LAVANDIÈRES / DE SANTAREM / OPÉRA-COMIQUE EN TROIS ACTES  / PAR / MM. D’ENNERY ET GRANGÉ / MUSIQUE DE M. A. GEVAERT / MISE EN SCÈNE DE MM. GRIGNON ET ARSÈNE. / Représenté pour la première fois, à Paris, sur le théâtre Lyrique / le 15 octobre 1855 / … / PARIS / MICHEL LÉVY FRÈRES, ­LIBRAIRES-ÉDITEURS / RUE VIVIENNE, 2 bis / - / 1855, 11.

160  The operas of François-Auguste Gevaert and Térésa. In the case of the first, ‘Les fifres et les vivandières dansent pendant le chœur’, and in the second, ‘Pendant le refrain on a repris les danses’.77 In other words, the introduction is framed by a danced chorus and a fandango in which the same artists dance again. The first of the two points that need to be made about dance is that what is visible here is endemic in opéra comique, both at the Théâtre-Lyrique and at the Opéra-Comique itself. What is required for the entire repertory of ­nineteenth-century opéra comique is a full account of all the stage directions that relate to dance, either in libretti or in the livrets de mise en scène. Although such an account is a long way off, it is clear that, despite the fact that divertissements were never composed or choreographed at the Opéra-Comique (and indeed were forbidden by licence), large portions of certain opéras comiques – choruses, the instrumental postludes to musical numbers of all types – were indeed danced, but in ways that were more informal, without a named choreographer and which were documented only in the livret de mise en scène.78 The second of the two points to make about dance and Gevaert’s opéras comiques concerns divertissements. The real danger here is assuming that the institutional control of repertory at the Opéra-Comique and the Théâtre-­ Lyrique was the same. This was simply not the case. Most particularly, the Théâtre-Lyrique was permitted the same generic licence as the Opéra in terms of dance, and this can be seen readily in the divertissements Gevaert wrote for Les lavandières de Santarem. Here is the outline of act ii (Table 4.4): There is much to consider here: an example of musique de scène bleeding over into mélodrame, the cortège 8bis, but most important, the ballet – which consists of three dances employing local colour and in the case of the aragonaise, a wordless chorus. This is fundamentally different to the types of danced ­chorus already witnessed. This ballet is fully choreographed, and ­Gevaert wrote music explicitly for the dance numbers – and they are extensive. The divertissement in act ii of Les lavandières de Santarem was praised for its generic qualities – elegance and originality79 – but also for its comic elements,80

77 Ibidem, 3 and 6. 78 See the inventory of the livrets de mise en scène in H. Robert Cohen, Cent ans de mise en scène lyrique en France (env. 1830–1930): catalogue descriptif des livrets de mise en scène, des libretti annotés et des partitions annotées dans la Bibliothèque de l’Association de la régie théâtrale (Paris), Musical Life in 19th-century France 2 (New York: Pendragon Press, 1986). Facsimiles are given in idem, Douze livrets de mise en scène lyrique datant des créations parisiennes, Musical Life in 19th-Century France 3 (Stuyvesant, N.Y.: Pendragon, 1991) and The Original Staging Manuals for Ten Parisian Operatic Premieres, Musical Life in 19th-Century France 6 (Stuyvesant, N.Y.: Pendragon, 1998). 79 Lovy’s praise (Le ménestrel, 28 October 1855) has already been discussed (138). See also Héquet (L’illustration, 3 November 1855). 80 ‘Les danses occupent une large place dan le deuxième acte; elles sont amusantes, réglées avec intelligence et suffiraient seules à sauver la pièce, qui ne brille ni par la variété ni par la nouveauté des situations. On a surtout applaudi une sorte de bourrée où s’agite et se trémousse de la façon la plus bouffonne un danseur qui ne serait pas déplacé sur la scène de l’Opéra. Rien

The operas of François-Auguste Gevaert  161 Table 4.4  G  evaert, Les lavandières de Santarem – Act ii 7 A Chœur des femmes B Air

Margarida

8 A Chœur des chasseurs B Ballet 1. La Valencienne 2. La Galicienne 3. L’aragonaise 8bis. Chaconne: 9 Trio

Margarida, Le Roi, Le Baron

9bis. Musique de scène 10 Couplets 11 Quatuor

11bis. Musique de scène 12. Final

Pablo D. Luiz, Manoël, Le Baron, le Duc

Margarida, D. Luiz, Manoël, Le Roi, le duc, chœur

‘Sous cette charmille’ ‘Où suis-je?’

3/8 e minor / E major, allegro agitato Recitative. 12/8 C major, andante non troppo lento; 4/4 a minor, allegro; 4/4 E major, plus lent ‘Le cor sonne’ Chœur et entrée du roi. 6/8 F major allegro con brio. 3/8 d minor, allegro con brio 6/8 a minor, allegretto 3/8 D major, allegro con brio + wordless chorus

Instrumental. Sortie des seigneurs ‘Approchez Begins with mélodrame. 2/4, D major, mon enfant’ andante sostenuto; G major, moderato; A major tempo primo; 3/4 d minor, allegro agitato assai; D major Instrumental. Mélodrame; dialogue and stage directions ‘Ah quel sort’ 6/8 D major, allegro con brio ‘Vous ne 4/4 C major, allegro; moderato; 3/8 a sortirez pas’ minor andantino; C major; allegro agitato; moderato

‘C’est elle’

Instrumental – tiny mélodrame at end (two words). Preceding réplique is ‘Que je la sache sauvée et je pars’ 4/4 E major, maestoso; a minor, allegro agitato; D major; E major, allegro vivace; allegro moderato

its exoticisms and its medievalisms.81 Intriguingly, the anonymous critic for the Revue et Gazette musicale de Paris praised both act i and act ii dances together (‘the dances in the first and second acts are lively and well ordered’), although the dance in act i was unchoreographed and without any formal divertissement.82

de plus gracieux, de plus vif et de plus original à la fois que la musique de ce pas, dont le succès a été très-grand et très-légitime’ (La France musicale, 28 October 1855 [Marie Escudier]). 81 Exoticisms were praised widely (La patrie, 5 November 1855 [De Prémary]; Le moniteur dramatique, 1 November 1855 [Théophile Deschamps]; Le moniteur universel, 28 October 1855 [De Rovray]; La revue française, 1 November 1855 [Reyer]). L. Girard noted that ‘the oboe motif that accompanies the dancer appears to belong to a medieval tonality, something that gives it a strongly original physiognomy’ (‘Le motif de hautbois qui accompagne le pas du danseur paraît appartenir à la tonalité du moyen-âge, ce qui lui donne une physionomie fort originale’; La revue de Paris, 15 November 1855). 82 ‘Les danses du premier acte et du second acte sont vives et bien réglées’ (Revue et Gazette musicale de Paris, 28 October 1855). The comments about act i must clearly relate to the

162  The operas of François-Auguste Gevaert Table 4.5  G  evaert, Le Chateau Trompette – Act ii 9 Entracte 10 Couplets

Richelieu

10bis. Musette Air de danse sur la scène 11 Air Lise 12 Duo

Lise, Richelieu

12bis. Menuet

Air de danse sur la scène

‘A tout propos je t’entends geindre’ [small offstage orchestra] ‘Non, non, non, je ne suis plus Lisette’ ‘Elle est vraiment jolie’ [Small offstage orchestra]

13 Quintette Lise, Cadichonne, ‘Ah! Ah! Ah!’ Richelieu, Champagne, Turcant 13bis. Musique de scène 14 Couplets

Cadichonne

14bis. Rigodon

Air de danse

15 Final

Lise, Richelieu, Champagne, Raffé

Allegretto, 2/4: d, D, d for each couplet Allegro vivo, 6/8: C Allegro leggiero, 4/4: E; Andante, 3/8: G; Tempo 1Allegro, 4/4: E Allegretto moderato, 3/4: C, Eb. 16 February 1860. Mouvement de menuet, 3/4: E [24 bars menuet; 8 bars trio + DC] Andante sostenuto, 4/4: Eb; Presto, 2/4: Eb; 20 bars of end of quintette (to accompany a comic pantomime)

‘Que je fasse bien ou mal’ [small offstage Music taken from no. 9 orchestra, strings (entracte) only muted] ‘A table ma chère’

Other commentators were more ambitious for the theatre and queried the absence of serious ballet83 and of a première danseuse.84 Any move to the Opéra-Comique would entail a reduction in the composition of dance music. Quentin Durward, of course, had no divertissement, although it did have danced choruses in the same way that did so much opéra comique. Gevaert’s second work for the Opéra-Comique, Le chateau trompette, raises interesting questions in this regard. Table 4.5 is an outline of its second act. introduction which includes two choruses, both of which, according to the libretto, are danced. See above. 83 ‘Le divertissement chorégraphique introduit dans la pièce, a plutôt amusé les yeux que captivé l’attention. Le premier sujet masculin est un comique étonnant. Accroupi sur ses pointes, il tourne comme une toupie, sans qu’on puisse comprendre comment il conserve son centre de gravité. Pour notre part, nous nous attendions à chaque instant à le voir tomber. Tomber! ah! bien oui; s’il tombe, c’est assis, avec une intention scénique qui donne des craintes sérieuses pour sa colonne vertébrale. Il se relève pour reprendre sa danse burlesque et ses drôlatiques mouvemens. C’est une bonne acquisition pour ce théâtre, qui ne saurait attaquer le ballet sérieux’ (Chadeuil, Le siècle, 30 October 1855). 4 ‘Le ballet du second acte a été vivement et justement applaudi. L’ensemble est excel8 lent. Mais il manquait au corps de ballet une première danseuse, et nous apprenons que M.  ­Pellegrin s’est empressé d’engager Mlle Galby, la charmante sylphide dont la danse correcte a toujours été remarquée et qui fut tant applaudie à la Porte Saint-Martin ainsi que sur plusieurs scènes étrangères’ (Revue et gazette des théâtres, 28 October 1855).

The operas of François-Auguste Gevaert  163 This seems to contradict everything that is conventionally understood about the relationship between dance and opéra comique; here are no fewer than three airs de danse that Gevaert expertly distributed across the act, and in apparent contravention of the monopoly on divertissement that the Opéra had enjoyed for centuries. Why this does not contradict everything that is understood about dance, and why there is no contravention of the Opéra’s licence, is because all the music is offstage and there is no visible dance – and indeed no dancers – involved.85 But what this indicates is – for Gevaert – ballet music, even in an environment where divertissement was effectively banned, was an interest that he would attempt to introduce whenever the opportunity arose. In this respect, there is a very clear continuity between Gevaert’s approach to opéra comique at the Théâtre-Lyrique and at the Opéra-Comique itself. Most of Gevaert’s opéras comiques had successful first runs but were not revived in Paris. According to Soubies, all Gevaert’s works ran from forty to sixty performances within the first year or so of their premiere.86 The exceptions were Les lavandières de Santarem, an exception that is explained by the critical disgust at the libretto (which largely overshadowed the music), and Le château-trompette. Soubies and Malherbe allude in the early 1890s to possible revival of the Opéra-Comique works: of Le château-trompette, Le capitaine Henriot and – if Gevaert would consent to prune some outmoded Italianisms out of it, Quentin Durward.87 The reception received by Gevaert’s operas when they returned to Brussels was uneven but significantly more enthusiastic for revival in Paris. Georgette, Quentin Durward and Le diable au moulin returned to the Théâtre de la Monnaie less than six months after their respective Paris premieres, and Le capitaine Henriot took not much longer.88 Le billet de Marguerite was not produced there until ­February 1871, and there is no evidence that any Brussels performance of Les lavandières de Santarem ever took place.89 There was, however, a production of Le billet de Marguerite at the Théâtre du Cirque, which opened on 19 July 1856 but closed on 8 August the same year.90 Ghent was almost as fast in its reception of Gevaert’s operas producing the same ones as in

85 The offstage music figures in act ii, scenes 1, 6 and 10; see A MONSIEUR DE COMTE DE MORNY / LE / CHATEAU TROMPETTE, 42–43; 51; 62. 86 See the figures given in Albert Soubies, Soixante-neuf ans à l’Opéra-Comique en deux pages de la première de ‘La dame blanche’ à la millième de ‘Mignon’ (1825–1894) (Paris: Fischbacher, 1894) and idem, Histoire du Théâtre-Lyrique (1851–1870) (Paris: Fischbacher, 1899). 87 Soubies and Malherbe, Histoire de l’Opéra Comique, 2:6, 2:77 and 1:306. 88 Jacques Isnardon, Le Théâtre de la Monnaie depuis sa fondation jusqu’à nos jours (Brussels etc.: Schott; Leipzig: Jeune, 1890), 397, 436, 444 and 482. 89 Ibidem, 516. According to Lionel Renieu, (Histoire des théâtres de Bruxelles, 2 vols (Paris: Duchartre and Van Buggenhoudt, 1928), 2:816) the musical text of the 1871 production of Le billet de Marguerite was subject to significant change. 90 Ibidem, 2:424.

164  The operas of François-Auguste Gevaert Brussels but adding in both Le billet de Marguerite and Le chateau trompette to its repertory rapidly after their Parisian premieres.91 Studying the opéras comiques of François-Auguste, Gevaert opens a window on a new world where operas of that sort were – for the first time – ­permitted at both the Théâtre-Lyrique and the Opéra-Comique. Working with tried and tested librettists and teams of librettists, Gevaert profited from his experience in one-act composition in order to gain a foothold and the Théâtre-Lyrique, and developed an ambitious approach to opéra comique at the Théâtre-Lyrique that continued traditions set in place at the beginning of the 1850s. Moving to the Opéra-Comique, he carried over much of what he had learned at the Théâtre-Lyrique, and with such works as Quentin Durward, Le chateau trompette and his final work, Le capitaine Henriot, he continued the tradition of the large-scale opéra comique, most obviously associated with Meyerbeer. What will never be known is how much he learned at both the Théâtre-Lyrique and Opéra-Comique would have been redeployed at the Opéra.

91 Prosper Claeys, Histoire du théâtre à Gand, 3 vols (Ghent: Vuylsteke, 1892), 3:117–201.

5 Between Opéra-Comique and Opéra-National Scribe, Vaëz and Boisselot c1850

The repertory of the Opéra-Comique underwent significant change in the ten years on either side of 1850.1 Two impulses, still insufficiently understood, were at the heart of this transformation. On the one hand was the gradual increase in size, scope and ambition of particularly three-act opéra comique during the period, largely driven by aesthetic concerns and animated by librettists and composers. On the other was the emergence of a Troisième théâtre lyrique – which may be called by the name it retained either side of the 1848 Revolution, the Opéra-National – that was allowed for the first time since the establishment of the licensing system in 1806–1807 to put on productions of new opéra comique, and therefore enter into competition with the Opéra-Comique for the first time.2 This article therefore seeks 1 The history of the Opéra-Comique at mid-century is poorly served. See Albert Soubies and Charles Malherbe, Histoire de l’Opéra Comique: la seconde salle Favart, 1840–1887, 2 vols (Paris: Flammarion, 1892–1893). Émile Genest, L’Opéra Comique connu et inconnu: son histoire depuis l’origine jusqu’à nos jours (Paris: Fischbacher, 1925) is useful in reconstructing this history. For listing of the repertory, see David Charlton and Nicole Wild, Théâtre de l’Opéra-Comique, Paris: Répertoire, 1762–1972, Musique-Musicologie (Sprimont: Mardaga, 2005). Still important, and an essential complement to the work previously cited, for tracing the fortunes of the institution is Albert Soubies, Soixante-neuf ans à l’Opéra-Comique en deux pages de la première de ‘La dame blanche’ à la millième de ‘Mignon’ (1825–1894) (Paris: Fischbacher, 1894). 2 Terminological and institutional problems abound in the description of the Opéra-­ National and its successor, the Théâtre-Lyrique. The Opéra-National, under the d ­ irection of Charles-Achille Tranchant (pseud. Mirecour) with Adolphe Adam as co-director, opened on 15 November 1847 at 66 Boulevard du Temple, previously the home of the Cirque Olympique; the operation folded in the wake of the 1848 Revolution on 29 March 1848. The Opéra-National reopened under the direction of Edmond Seveste on 27 ­September 1851 at 72 Boulevard du Temple (previously the home of the Théâtre­-Historique) and changed its name to Théâtre-Lyrique on 12 April 1852, by which time its direction was in the hands of the Jules Seveste. For the relationship between the licences granted to the Opéra-­National in its 1847 and 1851 guises, see 214. See Albert Soubies, Histoire du ThéâtreLyrique (1851–1870) (Paris: Fischbacher, 1899); Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981); Nicole Wild, Dictionnaire des théâtres parisiens au xixe siècle: les théâtres et la musique (Paris: Amateurs des Livres, 1989), 339–342 (for the Opéra-­ National) and 237–244 (for the Théâtre-Lyrique).

166  Between Opéra-Comique and Opéra-National to interrogate the changes in the poetry, drama and music of the genre in the period between c1840 and c1860, and to investigate the consequences of the expansion of the Opéra-National’s licence to encompass new opéra comique. It is hardly surprising that composers and librettists who worked both for the Opéra and Opéra-Comique should blur the generic boundaries that characterised the two institutions. Such composers as Auber, Halévy, ­Meyerbeer, Donizetti, Clapisson and Balfe moved, if not effortlessly at least with some frequency, between opéra comique and grand opéra and the institutions that supported the two genres. To a lesser extent, opéra comique was changed by librettists who straddled both genres. Jules-Henri Vernoy de Saint-Georges, in addition to a voluminous output of opéra comique, wrote the libretti to Halévy’s La reine de Chypre (1841) and Noë (produced posthumously),3 but it was Eugène Scribe who was largely responsible for weakening the boundaries between the two genres. Scribe’s contribution to the composition of libretti of both types requires a degree of revision: his musical reputation takes too little account of his work as a vaudevilliste, producing works for the Gymnase-Dramatique and the Théâtre des Variétés, at least as extensive as his work for the two major Parisian opera houses.4 Furthermore, Scribe’s contribution to the grand opéra libretto was largely complete by 1843; the libretti of almost all the works for the Opéra – from 1843 up to Meyerbeer’s L’africaine of 1865 – had been written by the ­m id-1840s or before.5 But in the period here under discussion, 3 For the parlous state of our knowledge of the career and aesthetics of Saint-Georges, see Francis Claudon, ‘Un collaborateur de F. Halévy: Henri de Saint-Georges’, Actes du colloque, Fromental Halévy, Paris, novembre 2000, eds. Francis Claudon, Gilles de Van and Karl Leich-Galland, Etudes sur l’opéra français du xixe siècle 5 (Weinsberg: Lucie ­Galland, 2003), 19–25. 4 Contrast Karin Pendle, Eugène Scribe and French Opera of the Nineteenth Century, Studies in Musicology 6 (Ann Arbor, MI: UMI Research Press, 1979), who treats Scribe exclusively as a librettist with Jean-Claude Yon, Eugène Scribe: la fortune et la liberté (Saint-­ Genouph: Nizet, 2000), who gives a balanced view of Scribe’s life and output. 5 This is a point insufficiently understood by either Pendle or Yon. The latter’s table of Scribe’s output (ibidem, [337]–[353] and of Scribe’s libretti (‘Scribe, Augustin-Eugène’, Fau­ ayard, quet, Joël-Marie, ed., Dictionnaire de la musique en France au xixe siècle (Paris: F 2003), 1136–1140) give only dates of premieres with occasional notes on settings of his libretti that significantly postdate the composition of the libretto. After Dom Sébastien of 1843, Scribe wrote almost no new libretti for five-act grand opéra: Clapisson’s Jeanne la folle of 1848 dates from 1843 (and passed through Rossini’s and Donizetti’s hands before reaching Clapisson); the contract for Meyerbeer’s Le prophète (1849) dated originally from 1838; Halévy’s La tempesta (1850) and Thalberg’s Florinde (1851) were Italian translations for Her Majesty’s Theatre in London (and Florinde had originally been written for Gomis in 1836); Auber’s Zerline, a work in three acts designed to support ballet-pantomime, Gounod’s La nonne sanglante (1854) and Verdi’s Les vêpres siciliennes (1855) were based on libretti from the 1830s for Berlioz and Donizetti, respectively; and Meyerbeer’s L’africaine (1865) was based on a libretto twenty-eight years old. Scribe’s collaboration with SaintGeorges on Halévy’s Le juif errant (1852) was reluctant, leaving Auber’s biblical L’enfant prodigue (1850) as the only instance of Scribe writing a new ­l ibretto for the Paris Opéra after 1843, a work that in any case stands apart from the historical five-act libretto.

Between Opéra-Comique and Opéra-National  167 c1840 to c 1860, Scribe’s output of opéra comique, especially in the largest, three-act form, rivalled his output of vaudevilles, and in the 1850s threatened to eclipse it.6 The emergence of a Troisième théâtre lyrique in 1847 marked yet another milestone on the road that stretched back to the 1820s, when the Théâtre Royal de l’Odéon had tried and failed to add new opéra comique to its repertory, and that encompassed similar attempts at the Théâtre des Nouveautés and the Théâtre de la Renaissance.7 Despite the energy expended on trying to re-establish the third lyric theatre during the 1840s, there was no concrete outcome until the appearance of the Opéra-National in 1847. But when this precursor to the Théâtre-Lyrique burst onto the stage, it did so with permission to play new opéra comique. This was the first time that the licensing system had been reconfigured to allow competition between the Opéra-Comique and another opera house.8 As in the case of the blurring of boundaries between grand opéra and opéra comique, composers writing for both the Opéra-Comique and the Opéra-­ National were important participants in the generic transfer between the two institutions. Three composers come into focus: Adolphe Adam, Aimé Maillart and Xavier Boisselot.9 Of these, however, only Boisselot worked with the same

6 The competition between libretti for opéra comique and vaudeville is easily traced in Yon, Eugène Scribe, [337]–[353]. 7 For the Théâtre Royal de l’Odéon, see Mark Everist, Music Drama at the Paris Odéon, 1824–1828 (Berkeley, Los Angeles and London: University of California Press, 2002); for the Théâtre de la Renaissance, see idem, ‘Theatres of Litigation: Stage Music at the Théâtre de la Renaissance, 1838–1840’, Cambridge opera Journal 16 (2004) 133–162, and idem, ‘Donizetti and Wagner: opéra de genre at the Théâtre de la Renaissance’. Giacomo Meyerbeer and 19th -Century Parisian Music Drama, Variorum Collected Studies Series CS805 (Aldershot: Ashgate, 2005), 309–341. The attempts at mounting operatic productions at the Théâtre des Nouveautés remain without a convincing modern account. See Louis-Henry Lecomte, Les Nouveautés, Histoire des théâtres de Paris 5 (Paris: Daragon, 1907; R Geneva: Slatkine, 1973) and Wild, Dictionnaire, 279–281. 8 In the 1820s, the Odéon had been permitted to play opéra comique but only works that had fallen into the public domain. The consequences were a disruption to the process of ­c anon-formation and unseemly fast reactions to the deaths of composers and librettists. See Everist, Music Drama at the Paris Odéon, 199–226, especially (for the impact of the death of Dalayrac’s librettist, Benoiît-Joseph Marsollier), 201. 9 Of these three composers, Adam hardly needs any introduction. Maillart, the c­ omposer of the evergreen Les dragons de Villars, died at the age of fifty-four in mid-career and has left little stamp on modern understandings of mid-century opéra comique. ­ roveOnline is significantly less impressive than the one given by Fétis The entry in G and Pougin (­‘Maillart, Aimé’. In Grove Music Online. Oxford Music Online, www.­ oxfordmusiconline.com/subscriber/article/grove/music/17474 (consulted 13 November 2008); François-­Joseph Fétis, Biographie universelle des musiciens et bibliographie générale de la m ­ usique, 8 vols (2nd edn. [with supplement by Arthur Pougin in two vols] Paris: Firmin Didot, ­1860–1865), Suppl. 2:148–149. Boisselot receives no mention in GroveOnline and – although called a ‘compositeur de renom’ – is only considered as a publisher in Anik Devriès-Lesure, ‘Boisselot, Xavier’, Dictionnaire de la musique en France au xixe siècle, ed. Joel-Marie Fauquet (Paris: Fayard, 2003), 159.

168  Between Opéra-Comique and Opéra-National librettist for both opera houses in this period.10 Not only did he work consistently with Scribe, but Scribe’s collaborator for both libretti was Gustave Vaëz, also a veteran of work at the Opéra but also the librettist of Donizetti’s Rita, ou Le mari battu (composed in 1841 for the Opéra-­Comique but performed only posthumously in 1860).11 The two works that Boisselot wrote to libretti by Scribe and Vaëz were Ne touchez pas à la reine! (Opéra-Comique, 1847) and Mosquita la sorcière (Opéra-­National, 1851). Analysis of the relationship between the three men, the two works and the two institutions is enhanced by, in addition to the usual range of surviving sources, the extensive correspondence that survives between Vaëz, Scribe and Boisselot, and the revealing quantity of sketch material that ­documents Boisselot’s working practices. Xavier Boisselot was the second son of Jean-Louis Boisselot, who had set up a music shop in Marseilles in 1823 and begun piano manufacture in 1830 with his elder son Louis-Constantin. By the late 1840s, the company had been responsible for pioneering the sostenuto pedal and was producing pianos in numbers and of a quality to rival both Pleyel and Erard.12 Xavier had studied with Fétis at the Conservatoire and in Lesueur’s composition class from which he won the Prix de Rome in 1836. On his return from Rome in 1838, he suffered the same difficulties as most of his laureate colleagues as he attempted to establish himself as an opera composer.13

Ne touchez pas à la reine!: Compositional History When Ne touchez pas à la reine! was premiered in 1847, most critics pointed to the decade that separated the composer’s return to Paris in 1838 from the premiere of the opera. Accounts in the press varied between an eight-year wait that the composer had to endure even before receiving a libretto and an opera that had in fact been written around 1840 and that Boisselot had kept from

10 Maillart worked with both Cormon and Carré at both houses, however – Les pêcheurs de Catane at the Théâtre-Lyrique and Lara at the Opéra-Comique – but only in the 1860s; Les pêcheurs de Catane dates from 1860 and Lara from 1864. 11 In collaboration with Alphonse Royer, Gustave Vaëz had been responsible for the translations and adaptations of Donizetti’s Lucia di Lammermoor as Lucie de Lammermoor (1839); Rossini’s Otello as Othello (1844); the Rossini pasticcio, Robert Bruce; and Verdi’s I Lombardi alla prima crociata as Jérusalem (1847). See Chapter 9. 12 The literature on the Boisselot piano company has focussed almost exclusively on the development of the sostenuto pedal and Liszt’s enthusiasm for the instrument. For the former, see Peter Roggenkamp, ‘Anmerkungen zum Tonhaltepedal’, 20 Jahre EPTE: Beiträge des Kongresses in Wiesbaden 1999 und des Seminars in Leipzig 2000, no ed. (Dusseldorf: Staccato, 2001), 74–94; idem, ‘Das Tonhaltepedal: eine wichtige Erfindung der Brahms-Zeit’, Brahms-Studien 11 (1997), 83–96. For the latter, see Mária Eckhardt, ‘Liszt à Marseille’, Studia Musicologica Academiae Scientiarum Hungaricae 24 (1982), 163–197; Robert M. Stevenson, ‘Liszt in Andalusia’, Journal of the American Liszt Society 26 (July–December 1989), 33–36; idem, ‘Liszt in the Iberian Peninsula, 1844–1845’, Inter-­ American Music Review 7/2 (Spring–Summer 1986), 3–22. 13 Almost everything that is known of Boisselot’s biography comes from Fétis. See his Biographie universelle des musiciens, 2:10 and Suppl. 1:105 (the latter authored by Alexis Rostand).

Between Opéra-Comique and Opéra-National  169 performance for the best part of decade.14 The compositional chronology is rather more complex. Boisselot was negotiating with Scribe over the rights to his music as early as April 1842, and a little over a year later was rejoicing at the successful reading of the libretto to, and its acceptance by, François-Louis Crosnier, the manager of the Opéra-Comique.15 Boisselot spent the next eighteen months writing at least a draft of the score which he described to Scribe in February 1845.16 The work then had to wait another eighteen months until August 1846 before it went into rehearsal, with a premiere on 16 January the following year.17 Ne touchez pas à la reine! was set in the Kingdom of Léon, but with no date specified in the libretto, and hinged on a putative tradition that insisted that any individual making contact with the Queen’s person should be put to death. In the context of Vaëz’s and Scribe’s libretto, the solution to the difficulty posed by the Queen being saved from certain death by Don ­Fernand d’Aguilar (but touching the Queen’s person in the process) was that the Queen and Fernand should marry.18 The long gestation of Ne touchez pas à la reine! masked a very real political context for this particular libretto. Pier-Angelo Fiorentino was one who recognised how the opera of 1847 carefully read the diplomatic environment of 1841: This piece, skilfully contrived and enhanced by piquant details obviously dates back to 1841, when European diplomacy was occupied marrying a young queen who was still playing with her dolls. By a multitude of circumstances that would be too long to enumerate, this work, full of timely allusions, was played six years later, when there is no longer a single infanta to be established in Europe.19 Fiorentino’s allusion would have conjured up the complex diplomacy of the early 1840s which had a real resonance for Louis-Philippe’s July ­Monarchy. Isabella II of Spain had come to the throne at the age of six in 1836 after 14 See the reviews in Le national, 24 January 1848 (Gustave Hecquet); Le ménestrel, 24 ­January 1848 (Edmond Viel); Le moniteur universel, 18 January 1847 (Hippolyte Prévost); Le journal des débats, 24 January 1847 (Hector Berlioz). 15 See the letters from Boisselot to Scribe, 24 April 1842 (Paris, Bibliothèque nationale de France [hereafter F-Pn] n.a.f. 22545, fols 120r-120v and 2 June 1843, ibidem, fols 122r-123r. 16 Boisselot asked Scribe to accept the dedication of the score which was complete in draft at that stage (letter, 17 February 1845; ibidem, fols 124r-124v). 17 Boisselot described rehearsals a progressing well in a letter to Scribe, 16 August 1846 (­ibidem, fols 125r-126v). 18 The best synopsis of Ne touchez pas à la reine! is in Félix Clément and Pierre Larousse, Dictionnaire des opéras (Dictionnaire lyrique)…, ed. Arthur Pougin (Paris: Larousse, [1897]), 772. 19 ‘Cette pièce, habilement conduite et relevée par de piquans détails date évidemment de 1841, époque à laquelle la diplomatie européenne s’occupait déjà de marier une jeune reine qui jouait encore de la poupée. Par une foule de circonstances qu’il serait trop long d’énumérer, cet ouvrage, plein d’à-propos, a été joué six ans plus tard, lorsqu’il n’y a plus en Europe une seule infante à établir’ (Le constitutionnel, 19 January 1847).

170  Between Opéra-Comique and Opéra-National three years under the Regency of her mother, Maria Christina. ­Isabella’s younger sister, María Luísa Fernanda, was two years her junior and remained heiress-presumptive until the birth of Isabella’s first child in 1851. The marriage of both daughters was critical to the French Crown: since ­Isabella was engaged to Francisco de Asís de Borbón (known to be homosexual, thought to be impotent, from whom any issue would likely be illegitimate), the marriage of her younger sister was of interest to all of ­Europe.20 Her marriage (which took place on the same day as her elder sister’s, 10 October 1846) to Louis-Philippe’s younger son, the Duc de Montpensier, was diplomatically advantageous to the French Crown since it left open the possibility of the throne of Spain falling to one of Louis Philippe’s grandchildren.21 In January 1847, with both the French Revolution of 1848 and the Spanish one of 1868 still in the future, the à propos to which Fiorentino alluded may still have had a delicious resonance for those in the audience at the Opéra-Comique with a good memory and an interest in current affairs. Boisselot’s surviving autograph working papers include a complete set of sketches and drafts for Ne touchez pas à la reine!.22 In at least one case, when taken together with the evidence from the composer’s correspondence with Scribe, it is possible to identify exactly how the two individuals worked together and to gain a greater precision than, for example, has been gained by consideration of Scribe’s relationship with Auber.23 The working relationship between Scribe and Boisselot may be gauged from Don Fadrique’s act ii aria. In its published version, it consists of the following sections:24 Recitative: ‘C’est contre mon amour trop longtemps défendre’ A.  Larghetto (3/8; Ab major):  ‘Toi qui séduis mon cœur’ Recitative: ‘Estrella, toi que j’aime’ B.  Allegro fieramente (4/4; Eb major):  ‘C’est moi, c’est moi, c’est toujours moi’ C.  Moderato (2/4; Bb major):  ‘De moi seul dépendent les places’ B.  Allegro fieramente (4/4; Eb major):  ‘C’est moi, c’est moi, c’est toujours moi’

giving an ABCB structure to the aria, a rondeau, as Boisselot himself called it.25

20 Raymond Carr, Spain, 1808–1975, Oxford History of Modern Europe, 2nd edn (Oxford: Clarendon, 1982) 241–242. 21 For the details of the diplomacy surrounding both marriages, see Ernest Jones Parry, The Spanish Marriages, 1841–1846 (London: Macmillan, 1936). 22 F-Pn MS 4397. 23 See, for example, the exemplary study in Herbert Schneider and Nicole Wild (eds.), La Muette de Portici: kritische Ausgabe des Librettos und Dokumentation der ersten Inszenierung, Erlanger romantistische Dokumente und Arbeiten 11 (Tübingen: Stauffenberg, 1993). 24 Ne touchez pas / À LA REINE / Opéra Comique en 3 actes, / Poeme de MM / E. Scribe et ­G ustave Vaëz, / Musique de / X. BOISSELOT / PARTITION PIANO ET CHANT / … / A PARIS, chez J. MEISSONNIER et FILS, Rue Dauphine, 22 / J.M. 221. / Leipsig Breitkopf et Hartel, 78–86. 25 Letter from Boisselot to Scribe, 16 August 1846. F-Pn n.a.f. 22545, fols 125r-126v

Between Opéra-Comique and Opéra-National  171 The ritournelle for this number brought forth a number of attempts from the composer. The first attempt is notated in Eb major, 4/4 on two staves (Example 5.1).

Example 5.1  Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, Larghetto ‘Toi qui séduis mon cœur’, first draft of Allegro fieramente. F-Pn MS 4397.

Fully worked out, with the exception of the inner voices, the vocal line exhibits some awkwardness over the problematic word ‘ministre’ and the following words in the third and fourth bars and deploys an attractive Neapolitan approach to the tirade at the end of the sketch. But Boisselot abandoned this version completely and drafted a replacement that was completely different (Example 5.2). The clef has been changed to a treble, which seems to suggest a change of voice type from baritone to tenor, and the key is G major. However, the sketched melody, and a good deal of its continuation (the first thirteen bars), feeds directly into the final version (Example 5.3). The Larghetto clearly gave both librettist and composer some difficulty, and Boisselot had to resort to writing a monstre in order to show Scribe ­exactly what verse structure and rhymes he wanted, and where internal stresses in the line were supposed to fall. The principle behind the monstre was the composer would use language functionally and without poetic ambition to indicate the patterns required, and then the librettist would write more ambitious poetry around it.26 Comparing the A section of Boisselot’s monstre with Scribe’s final version shows the system working well (Example 5.4). 26 See Herbert Schneider’s useful introduction to the monstre (Dictionnaire de la musique en France au xixe siècle, 813–814).

Example 5.2  B  oisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, ­L arghetto ‘Toi  qui séduis mon cœur’, second draft of Allegro fieramente. F-Pn ­MS 4397. 

Example 5.2  ( Continued)

Example 5.3  B  oisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, Larghetto ‘Toi qui séduis mon cœur’, final draft of Allegro fieramente. F-Pn MS 4397 and published score pp. 81–82.

Example 5.3  ( Continued)

176  Between Opéra-Comique and Opéra-National Boisselot’s monstre Viens, toi que j’aime tant Estrella mon ivresse Adorable maitresse Eve aux yeux de serpent Sirène enchanteresse Enfin ton cœur se rend. Recherchant le mystère et l’ombre Les cœurs remplie d’un doux émoi Quant la nuit régnera plus sombre L’amour te guidera vers moi.

Scribe’s final version Toi qui séduis mon cœur Sirène enchanteresse, Toi qui fais mon ivresse Je veux par mon ardeur Ô gentille maitresse Désarmer ta rigueur! Recherchant l’ombre et la mystère Le cœur rempli d’un doux émoi Estrella quelque jour moins fièvre L’amour doit te livrer à moi

Example 5.4  B  oisselot/Scribe, Ne touchez pas à la reine!, air, no. 6, Larghetto ‘Toi qui séduis mon cœur’, Boisselot’s monstre and Scribe’s final version. F-Pn n.a.f. 22545, fols 125v.

Scribe reworked parts of the monstre, retaining the sequence of six hexasyllables followed by four octosyllables as well as the rhyme scheme and one (‘…resse’) exact rhyme. But Boisselot’s monstre also included those parts of the lyric on which he and Scribe had already agreed, the B and C sections, but in this original version the number began with the B section and included an extra section creating a structure (using the same letters as previously with supplements) as follows: Recitative: ‘La reine prie! Et moi, veillant sur sa jeunesse’ B.  Allegro fieramente (4/4; Eb major): 

‘C’est moi, c’est moi, c’est toujours

moi’ C.  Moderato (2/4; Bb major):  ‘De moi seul dépendent les places’ D.  [No surviving setting] ‘Et jusqu’aux femmes même’ A.  Larghetto (3/8; Ab major):  ‘Toi qui séduis mon cœur’ B.  Allegro fieramente (4/4; Eb major): 

‘C’est moi, c’est moi, c’est toujours

moi’ Scribe clearly rewrote the entire recitative text, and even in Boisselot’s monstre, he noted at the end of the text for the D section that ‘Following your advice, I have deleted these lines to move promptly to the andante or rather to the cantabile’.27 Boisselot’s letter to Scribe, written in August 1846 during the rehearsals for the premiere, when the former could report that most of the music was known by heart and that the spoken dialogue had been rehearsed for the first time the day before, shows that this aria at least was still in a

27 ‘Suivant votre [Scribe’s] conseil, j’ai supprimé ces vers pour passer de suite à l’andante ou plutôt au cantabile….’ (ibidem).

Between Opéra-Comique and Opéra-National  177 significant state of flux. Exactly how long the situation continued is not at all clear. Despite the composer’s conviction that the number was ‘reasonably successful; it is one of the less bad numbers in my piece; on hearing it, the regent was entirely satisfied’28; the only voice in the press to comment on the quality of the piece was Marie-Pierre-Pascal Escudier’s review in La France musicale, which described it as trivial and much too long.29 It is entirely possible, then, that the change in structure from the recitative and five sections (described in Boisselot’s letter to Scribe) to recitative and four sections in the published version of the aria might have taken place after the premiere, although some explanation would have to be found for the absence of the D section in the first version among Boisselot’s otherwise very full working papers. More conventional were the changes Boisselot made to La reine’s act iii aria. He reported to Scribe in October 1846 that he had come under significant pressure from Anne-Benoîte-Louise Lavoye who was rehearsing the role to make changes, and despite his initial reluctance, he agreed to modify the aria apparently to the delight of the singer; it was complete by the beginning of the following month.30 The overall structure of the aria changed little during the course of the compositional process: Instrumental introduction Moderato (4/4; g minor); Andantino (6/8; g minor); Allegro (6/8; g minor) Recitative: ‘L’effroi que je combats’ Adagio (4/4; Bb major): ‘Hélas! Qui m’aimera!’ Andatino pastorale (3/4; d minor): ‘Quel est ce bruit!’ Recitative: ‘Elles s’éloignent’ Allegro moderato (4/4; G major): ‘Ciel de feu, beau ciel des Espagnes’ Even in its earliest stages of composition, the final section of the aria, ‘Ciel de feu, beau ciel des Espagnes’, posed problems for the composer. As ­Example 5.5 shows, it was conceived with two very different types of vocal line. Despite its appearance, this is not an example of an ossia, or a passage marked for facilité, since the lower of the two lines exploits virtuosic Italianate arpeggios with pauses for cadenzas, while the upper line displays a 28 ‘assez bien réussi; c’est l’un des moins mauvais morceaux de ma pièce; en l’entendant le régent a été satisfait entièrement’ (ibidem). 29 ‘Le principal mérite de M. Boisselot est d’être clair et distingué. Si ses mélodies ne sont pas toujours neuves, au moins sont-elles empreintes du meilleur goût. Nous devons cependant faire une exception pour l’air de basse chanté par Herman Léon [sic] qui est trivial et beaucoup trop long’ (La France musicale, 24 January 1847). 30 For the biography of Lavoye, notwithstanding Fauquet, ‘Lavoye, Anne-Benoîte-Louise’, Dictionnaire de la musique en France au xixe siècle, see the fuller account in Fétis, Biographie universelle des musiciens, Suppl. 1:83.

178  Between Opéra-Comique and Opéra-National

Example 5.5  Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 9, Allegro ­moderato ‘Ciel de feu, beau ciel des Espagnes’, bars 1–7, original version with two alternative vocal lines. F-Pn MS 4397.

preference for powerful triplets, motor rhythms and dotted figures. For a composer writing years in advance of the premiere, this was a predictable and pragmatic approach. Even given the fact that Lavoye had been singing successfully at the Opéra-Comique since 1843, Boisselot could not at this

Between Opéra-Comique and Opéra-National  179 stage guarantee the identity, and hence the musical character, of the artist for whom he was writing. Furthermore, it was clearly the same section to which Lavoye took exception; its text is as follows: Ciel de feu, beau ciel des Espagnes, Rayons dorés, parfums enivrants de nos fleurs, Air sauvage, air pur des montagnes, Vous apportez la vie et l’amour dans les cœurs! Pauvre Reine, triste et plaintive, Je suis seule, seule et captive! Je ne sais quels vagues désirs Font monter vers Dieu mes soupirs. Ciel de feu, beau ciel des Espagnes, Rayons dorés, parfums enivrants de nos fleurs, Air sauvage, air pur des montagnes, Vous apportez la vie et l’amour dans les cœurs! Nul espoir ne me vient sourire, La tristesse remplit ma cour; Je n’ai qu’un trône où je soupire, Et je pourrais avoir le bonheur et l’amour! Ciel de feu, beau ciel des Espagnes, Rayons dorés, parfums enivrants de nos fleurs, Air sauvage, air pur des montagnes, Vous apportez la vie et l’amour dans les cœurs! With routine exceptions, the final four lines of the aria resemble closely the upper-stave alternative presented in F-Pn MS 4397 (Example 5.5), and it would appear that Lavoye was largely content with everything except the very end of the final section. Example 5.6a gives this final passage in Boisselot’s original version. As in the case of the rest of the aria, Boisselot’s draft gives fragments of two vocal alternatives, and while the original certainly varies from the ritornello, it clearly did not go far enough for Lavoye, as comparison with the final published version shows: With hindsight, and the knowledge of the two key roles that Lavoye created either side of La Reine in Ne touchez pas à la reine! – Athénaïs de Solange in Halévy’s Les mousquetaires des la reine (3 February 1846) and the title role in Auber’s Haydée (28 December 1847) – the demands that she made on Boisselot were predictable. Comparing the changes in ‘Ciel de feu, beau ciel des Espagnes’ with Lavoye’s act iii aria in Haydée, ‘Je suis dans son palais’, and her act i aria in Les mousquetaires de la reine, ‘Me voilà seule enfin’, points very clearly to the types of ornamentation missing in Boisselot’s original version (Example 5.6a) but preferred by the

Example 5.6a  Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 9, Allegro ­moderato ‘Ciel de feu, beau ciel des Espagnes’, original setting of fi ­ nal four lines. F-Pn MS 4397 (transposed from A major to G major to facilitate comparison with Example 5.6b and with original crossings-­ through retained).

Example 5.6a  (Continued)

Example 5.6a  (Continued)

Between Opéra-Comique and Opéra-National  183

Example 5.6a  (Continued)

singer (Example 5.6b). All three arias end with an allegro akin to ‘Ciel de feu, beau ciel des Espagnes’, and in each of the three, preferences are discernible: a predilection for rising chains of trills (modest in Ne touchez pas à la reine!), swooping coloratura arpeggios and scales, and ambitious chromatic runs. There was no doubt about the success either of the aria or its performance. Viel could hardly have been more generous when he described the aria as ‘as piquant as it is varied, and whose riches Mlle Lavoye rendered with a dazzling perfection’, and Georges Bousquet delighted in all parts of the aria: ‘The queen’s air, at the beginning of the third act, is very well designed; the theme of the andante is of a naïve, original, charming distinction; between the two parts of the air we notice details of a successful instrumentation of a

Example 5.6b  Boisselot/Scribe, Ne touchez pas à la reine!, air, no. 9, Allegro moderato ‘Ciel de feu, beau ciel des Espagnes’, definitive setting of final four lines (published version).

Example 5.6b  (Continued)

Example 5.6b  (Continued)

Between Opéra-Comique and Opéra-National  187 picturesque colour; the strette [the section just under discussion, ‘Ciel de feu, beau ciel des Espagnes’] is brilliant in its effect’.31

Ne touchez pas à la reine!: Register and Status Boisselot’s musical language came in for much comment in the wake of his first opera’s premiere. Many of the observations centred on the question of the recherches from which his music seemed to profit. Although Hecquet could praise the composer because ‘he does not seek out bizarre or unexpected modulations, nor overcomplicated accompaniments’,32 he also pointed to his use of unexpected rhythms, a view seconded by the anonymous critic of La sylphide who wrote that ‘The melody is not exempt from learnedness; it often avoids straightforward rhythm and falls into a manner that seeks for distinction’.33 Viel, on the other hand, accused the composer of ‘seeking after originality which results in pretention’.34 There were, however, critics such as Bousquet who thought that the innovations wrought by the composer merited exceptional attention from the critical community.35 Other voices located Boisselot’s innovations in his response to the demands of dramaturgy. Fiorentino, having taken a less generous view of Boisselot’s recherches and having attributed them not only to questions of rhythm but also to those of harmony and orchestration, stated the case plainly: His main concern, his exclusive preoccupation, is expression and sentiment; he translates not only the poet’s word with scrupulous fidelity, but the spirit of the situation, the colour of the scene, the image of the character, and when he finds or believes he has found the motive that should render his thought literally, he does not worry about the rest.36

31 ‘aussi piquant que varié, et dont Mlle Lavoye a rendu les riches détails avec une éblouissante perfection’ (Le ménestrel, 24 January 1847); ‘L’air de la reine, au commencement du troisième acte, est très bien coupé; le thème de l’andante est d’une distinction naïve, originale, ravissante; entre les deux parties de l’air on remarque des détails d’instrumentation d’une couleur pittoresque bien réussie; la strette [the section just under discussion, ‘Ciel de feu, beau ciel des Espagnes’] est brillante et à effet’ (Le commerce, 27 January 1847). 32 ‘Il ne recherche pas les modulations bizarres et imprévues, ni les accompagnemens trop compliqués’ (Le national, 24 January 1847). 33 ‘La mélodie n’est pas exempte de recherche; elle évite souvent le rythme franc et net pour tomber dans la manière en courant après la distinction’ (La sylphide, 24 January 1847). 34 ‘une recherche d’originalité qui aboutit à la prétention’ (Le ménestrel, 24 January 1847). 35 ‘But there is in the success of Ne touchez pas à la reine!, a particular circumstance, a sort of innovation, very bold in what is claimed, which deserves on the part of the critics an exceptional attention’ (‘Mais il y a, dans le succès de Ne touchez pas à la reine!, une circonstance particulière, une sorte d’innovation, très hardie à ce qu’on prétend, qui mérite de la part la critique une attention exceptionnelle’; Le commerce, 27 January 1847). 36 ‘Son principal souci, sa préoccupation exclusive, c’est l’expression et le sentiment; il ­traduit  avec une fidélité scrupuleuse non seulement le mot du poète, mais l’esprit de la situation, la couleur de la scène, le caractère du personnage, et lorsqu’il trouve ou croit

188  Between Opéra-Comique and Opéra-National These comments were made with the conviction that the composer, ‘far from defending himself against them, will regard some of these accusations as the highest praise that one can give music’, and Fiorentino thought – u ­ nlike many of his colleagues - that this was a consequence not of B ­ oisselot’s inexperience but of a genuine conviction.37 But again, the anonymous critic of La sylphide thought that the characteristics identified by Fiorentino were ‘inquiétants pour l’avenir’, and spun the latter’s idea of ‘fidélité scrupuleuse’ a totally different way: By wanting to translate the meaning of each word literally, the composer goes astray, his style becomes obscure; his ideas are often vague, elusive, for lack of a fairly fixed form; his music sometimes has the effect of a painting that lacks lines and contours.38 The difference between the anonymous author and Fiorentino was ­dramatised starkly by the latter when he asked, ‘Is this a bold innovation? is this a return to the old school?’, and this strikes at the heart of the position of Ne touchez pas la reine! in the context of opéra comique around 1850.39 The concepts of ‘ancienne école’, ‘usage ancien’ or ‘vieille école’ are key to understanding discussions of the trajectories of opéra comique throughout the nineteenth century. Such terms evoke a relationship between music, poetry and drama that held sway before the appearance of such works as Auber’s Fra Diavolo and Hérold’s Le pré aux clercs around 1830. The critic of La gazette de France, Charles-Louis de Sévelinges, could claim that such a respect for poetry and drama was new but that it marked a return to the past: It is something very praiseworthy and quite new; the poem is usually sacrificed to music in things as in words; it seems that M. Boisselot wanted to establish the opposite principle and revive the ancient usage which saw in music a brilliant accessory given to speech, a harmonious form for the setting aside of feeling or idea, but not a kind of abstract pleasure, tasted independently of the circumstances surrounding it, and to which one does not ask to lend the slightest relief to the scenic effects of the scene [emphasis added].40

37 38

39 4 0

avoir trouvé le motif qui doit rendre littéralement sa pensée, il ne s’inquiète pas du reste’ (Le constitutionnel, 19 January 1847). ‘loin de s’en défendre, regardera une partie de ces accusations comme le meilleur éloge qu’on puisse faire de sa musique’ (ibidem). ‘A force de vouloir traduire mot à mot le sens de chaque parole, le compositeur s’égare, son style devient obscur; ses idées sont souvent vagues, insaisissables, faute d’une forme assez arrêtée; sa musique fait quelquefois l’effet d’une peinture qui manquerait de lignes et de contours’ (La sylphide, 24 January 1847). ‘Est-ce là une innovation téméraire? est-ce un retour à l’ancienne école?’ (Le ­constitutionnel, 19 January 1847). ‘C’est quelque chose de très louable et d’assez nouveau; le poëme ordinairement est sacrifié à la musique dans les choses comme dans les mots; il semblerait que M. Boisselot a

Between Opéra-Comique and Opéra-National  189 Such attention to dramatic and poetic detail could therefore be harnessed in support of an enthusiasm for a return to the past, but other voices, taking a broader view of Boisselot’s style, would take a different view of the same observations. The knowledge that Boisselot had taken anything between three and five years to bring his score to the opera house gave some an explanation for why such powerful music might have little verve: One would think, to hear them, that these melodies have been touched, retouched, revised, corrected, increased so many times, that they have lost their primitive character, the one that had given them inspiration; it is music in some way distilled drop by drop, where almost nothing came from this clearly fertile and energetically powerful source that we call verve.41 For others, however, such careful compositional industry could lead to an elevated style unwarranted in current opéra comique. ‘More severe perhaps and of a higher style than the current genre of opéra comique’ was how B. Carr of La tintamarre described the score,42 and Hecquet wondered if ‘Perhaps the work would have gained much if the author had treated the comic parts with as much care and love as the passionate scenes’, but  – unlike Fiorentino  – attributed this imbalance to inexperience.43 Even ­G alignani’s Messenger noted that ‘the music [of Ne touchez pas à la reine! is] of a more elevated character than is generally heard at this theatre [the  Opéra-­Comique]’.44  But  Berlioz  attributed such an elevated character more to the musical situations, which – as many other critics noted – were the creation of the librettists: ‘The musical situations in this piece are numerous, varied, and of a more elevated type than those of most opéras comiques’.45 Escudier, on the other hand, took such an elevated style to be undesirable:

41

42 43 4 4 45

voulu établir le principe opposé et faire revivre l’usage ancien qui dans la musique voyait un accessoire brillant donné à la parole, une forme harmonieuse pour la mise au dehors du sentiment ou de l’idée, mais non une sorte de plaisir abstrait, goûté indépendamment des circonstances qui l’entourent, et auquel on ne demande pas de prêter le moindre relief aux effets de la scène [emphasis added]’ (La gazette de France, 21 January 1847). ‘On croirait, à les entendre, que ces mélodies ont été touchées, retouchées, revues, corrigées, augmentées tant de fois, qu’elles ont perdu leur caractère primitif, celui que leur avait donné l’inspiration; c’est de la musique en quelque sorte distillée goutte à goutte, où presque rien n’est venu de ce jet nettement fécond et énergiquement puissant qu’on appelle la verve’ (La sylphide, 24 January 1847). ‘Plus sévère peut-être et d’un style plus élevé que ne le comporte le genre actuel de l’opéra-comique’ (La tintamarre, 24–30 January 1847). ‘Peut-être l’ouvrage aurait-il beaucoup gagné si l’auteur en eût traité les parties comiques avec autant de soin et d’amour que les scènes passionnées’ (Le national, 24 January 1847). Galignani’s Messenger, 20 January 1847. ‘Les situations musicales, dans cette pièce, sont nombreuses, variées, et d’un genre plus élevé que celles de la plupart des opéras-comiques’ (Le journal des débats, 24 January 1847).

190  Between Opéra-Comique and Opéra-National The music of Ne touchez pas à la Reine does not have too much the character of opéra comique. One can feel that the composer has allowed himself to be drawn into a sphere that is too high; opéra comique does not have large, widely developed pieces. It is necessary to know how to comply with the demands of this bastard theatre, which wants above all a place for its dialogue and its little songs.46 Escudier’s mild antipathy to the genre as a whole explains his term ‘théâtre bâtard’, and his apparent preference for dialogues and ‘petites chansons’ over ‘grands morceaux largement développés’ was shared by many. Prévost, Charles Monselet and the anonymous critic of Le charivari all expressed the same view47 – that composers for the Opéra-Comique had to limit their activities to the boundaries imposed by the institution – but it was left to Escudier to drive the point home most clearly: ‘Well, frankly, in this score, there are too many of the beauties of grand opéra and too few beauties of opéra comique’.48 Suggestions that the Opéra-Comique might be trespassing on the territory of the Opéra were serious indeed, although the presence of ‘de grands morceaux largement développés’ was unlikely to trouble the Commission des théâtres charged with monitoring the licensing system. The question of movement, however, was a more pressing issue, and it is surprising that more was not made of it as the elevated register of Ne touchez pas à la reine! was picked over. There had always been a number of supernumeraries in opéra comique who took the responsibility for stage action – platoons of soldiers, crowds of peasants and so on – and also possibly for danced choruses. While this is still an area of practice in opéra comique that is insufficiently explored, it is certain that the Opéra-Comique never commissioned a choreographer for a divertissement and no composer ever wrote ballet

46 ‘La musique de Ne touchez pas à la Reine, n’a pas trop le caractère de la musique d’opéra comique. On sent que le compositeur s’est laissé entraîner vers une sphère trop élevée; l’opéra comique ne comporte pas de grands morceaux largement développés. Il faut savoir se plier aux exigences de ce théâtre bâtard, qui veut avant tout place pour son dialogue et pour ses petites chansons’ (La France musicale, 24 January 1847). 47 ‘Ainsi, ses accompagnements sont en général trop laborieux, les dessins des parties secondaires s’écartent trop savamment de ceux de la mélodie vocale. Ce procédé ingénieux, dont la symphonie tire de curieux et magnifiques effets, doit être employé avec plus de sobriété au théâtre, et surtout à l’Opéra-Comique’ (Le moniteur universel, 18 January 1847 [Prévost]); ‘Nous sommes à l’Opéra-Comique, et les auteurs ont su se renfermer dans les limites du genre. Nous avons affaire à une comédie tout à fait coquette, et non à un drame échevelé’ (L’artiste, 24 January 1847 [Monselet]); ‘Ses idées musicales semblent même se trouver parfois à l’étroit dans le cadet de l’Opéra-Comique, nous citerons, par exemple le final du second, qui n’est pas évidemment dans les proportions du genre’ (Le charivari, 18 January 1847). 48 ‘Eh! bien, franchement, dans cette partition, il y a trop de beautés de grand opéra et pas assez de beautés d’opéra comique’ (La France musicale, 24 January 1847).

Between Opéra-Comique and Opéra-National  191 music, as was normal at the Opéra.49 The adjunct to the divertissement at the Opéra, the cortège, was not part of the tradition at the Opéra Comique. Except until Ne touchez pas à la reine!. The act one finale consists of an opening chorus, ‘Le maure approche’, which is repeated after a series of exchanges between Estrella and Don Fadrique. There is then an address by the latter announcing the arrival of the queen followed by another chorus. The stage directions for the second part of this chorus are remarkable: Pendant la seconde partie du chœur, entre par la porte qui conduit aux appartements de la Reine, un cortège ouvert par DES HALLEBARDIERS suivis par DES OFFICIERS DU PALAIS, portant des bannières, et qui vont les ranger dans la galerie extérieure, auprès des pennons des chevaliers. Paraissent ensuite LES ALCADES, LES HAUTS JUSTICIERS, puis LES DAMES DE LA COUR avec DES JEUNES FILLES vêtues de blanc, et qui portent des corbeilles de fleurs. Elles se rangent pour laisser passer LA REINE avec LE RÉGENT, marchant à sa gauche. FERNAND est confondu dans la foule qui le cache. ESTRELLA, pour voir la Reine, s’avance curieusement derrière les dames.50 This is clearly not a cortège of the same dimensions as those inaugurated by La juive and that by the mid-1840s were as much a conventional feature of grand opéra as a formally choreographed divertissement. For Escudier, this was yet another piece of evidence that Ne touchez pas à la reine! had ambitions that went beyond the limits of opéra comique: ‘An excellent finale! But what do you want us to do with a finale with all the splendours, all its cortege of chorus and orchestra? [emphasis added]’.51

49 In a domain in which composers do not write ballet music for divertissements, tracing movement (which could range from simple movement of the foot to fully-developed dance steps) is severely problematic. Occasional indications are given in libretti, but much greater detail is available for those works whose published livrets de mise en scène genuinely reflect practice during the first run of the opera. The collection of livrets de mise en scène published by Louis-Pierre-Marie Palianti is particularly valuable for the Opéra-Comique, since it was there that he was sous-régisseur from c1836 and régisseur from 1849 until 1872. See his Collection de Mises-en-scène rédigées et publiées par M. L[ouis] Palianti, 3 vols (Paris: chez l'auteur et MM. les correspondants des théâtres, n.d.); for the chronology of this collection see Arne Langer, Der Regisseur und die Aufzeichnungspraxis der Opernregie im 19. Jahrhundert, Perspektiven der Opernforschung 4 (Frankfurt am Main, Berlin and Paris: Lang, 1997), 292–322 and 349–356. 50 THÉÂTRE ROYAL DE L’OPÉRA-COMIQUE. / - / NE TOUCHEZ PAS / A / LA REINE / OPÉRA-COMIQUE EN TROIS ACTES, / Paroles de MM. SCRIBE et GUSTAVE VAËZ, / Musique de M. XAVIER BOISSELOT, / (MISE EN SCÈNE DE M. HENRI.) / Représenté pour la première fois, à Paris, sur le théâtre royal de l’OPÉRA-COMIQUE, / le 16 janvier 1847 / PARIS / BECK, ÉDITEUR / … / 1847, 8. 51 ‘Un beau final! mais que voulez-vous que nous fassions d’un final avec toutes ses splendeurs, tout son cortége de chœurs et d’orchestre [emphasis added]’ (La France musicale, 24 January 1847).

192  Between Opéra-Comique and Opéra-National For critics and admirers alike, Ne touchez pas à la reine! had ambitions in terms of the severity of its style and the extent of individual compositions that went beyond the confines of opéra comique. For a debutant composer to produce a work in three acts at the Opéra-Comique was almost unheard of. Bousquet explained how the system worked normally: A use consecrated in recent times, a use which had become, in a way, the fundamental law of the administration of the Opéra-Comique, rigorously established this point: that every beginner... had to be satisfied with a work in one act,52 and Théophile Gautier typically dramatised the extraordinary position Boisselot enjoyed in having a three-act success at his first attempt at the Opéra-Comique: What! a royal theatre finally opened up sincerely, frankly, to a new composer, French and laureate of the Institute! He was given three acts all at once, proper actors, rich costumes and fresh sets, and M. Scribe deigned to put in the poem the finesse and spirit of his heyday! This overturns all preconceived ideas in the theatre.53 But neither Bousquet or Gautier drew attention to the fact that, although Ne touchez pas à la reine! was in three acts, it only consisted of eleven numbers, many fewer than the sixteen – each as complex as anything in Ne touchez pas à la reine! – in Auber’s Haydée or Halévy’s Les mousquetaires de la reine, both exact comparators and contemporaries. In a largely hostile review, the anonymous critic of L’illustration observed that if Scribe and Vaëz had cast Ne touchez pas à la reine! in a single act, it would have become ‘one of the most pleasant works in the repertory, old or modern’.54 Overstatement, certainly, but it gives a context to Fiorentino’s better-judged comment to the effect that ‘Indeed, it is M. Basset [André-­ Alexandre Basset, Crosnier’s successor as manager of the Opéra-Comique] who wet the work in three acts. During the reign of M. Crosnier, it would

52 ‘Un usage consacré dans ces derniers temps, usage qui était devenu, en quelque sorte, la loi fondamentale de l’administration du théâtre royal de l’Opéra-Comique, établissait rigoureusement ce point: que tout débutant … devait se contenter d’une pièce en un acte’ (Le commerce, 27 January 1847). 53 ‘Quoi! un théâtre royal s’est ouvert enfin sincèrement, franchement, à un compositeur nouveau, français et lauréat de l’Institut! On lui a donné trois actes tout d’un coup, des acteurs convenables, des costumes riches et des décors frais, et M. Scribe a daigné semer dans le poème la finesse et l’esprit de ses bons jours! Ceci renverse toutes les idées reçues au théâtre’ (La presse, 18 January 1847). 54 ‘l’un des ouvrages les plus agréables du répertoire, ancien ou moderne’ (L’illustration, 23 January 1847).

Between Opéra-Comique and Opéra-National  193 Table 5.1  Number of acts in opéra comique, 1834–1854 Dir.

Crosnier

Acts

1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845

II III I

2 1 6

1 4 4

2 4 6

Basset

0 4 3

1 4 4

3 3 4

1 4 4

2 3 6

0 3 3

0 4 4

0 4 4

0 3 4

Perrin

1846

1847

1848

1849

1850

1851

1852

1853

1854

0 2 4

0 4 4

0 3 4

2 3 1

0

0 2 2

2 5 2

2 3 4

0 2 3

3

have only had two’.55 This is a provocative statement that can be tested simply: Table 5.1 gives numbers of opéras comiques for a twenty-year period from 1834 to 1854 (the year of Meyerbeer’s L’étoile du nord) (Table 5.1).56 This covers all of Crosnier’s administration (1834–1845), all of Basset’s (1845–1848) and the first six years of Emile-Charles-Victor Perrin’s management of the Opéra-Comique, and gives the number of works in one, two and three acts. It mostly confirms Fiorentino’s observation, but while it is true that, during Basset’s administration there were no opéras comiques in two acts, this had also been the case since 1841, under Crosnier. In other words, two-act opéra comique was more or less unknown during the 1840s but was revived under Perrin in the 1850s. In the light of the data presented in Table 5.1, the compositional chronology of both Ne touchez pas à la reine!’s libretto and music is critical: as has been seen, Scribe and Vaëz were working on the libretto in 1842–1843, at the very nadir of enthusiasm for two-act works. By the time Boisselot came to write the score (1844–1845), and especially by the time of the premiere (1847), configuring or reconfiguring the work in two acts would have occurred to nobody except such a critic with as good a memory as Fiorentino. The dimensions and poetic, dramatic and musical ambitions of Ne touchez pas à la reine! bear comparison with the work that Scribe, Vaëz and Boisselot wrote immediately afterwards, Mosquita la sorcière. 55 ‘En effet, c’est M. Basset qui a fait mettre la pièce en trois actes. Sous le règne de M. ­Crosnier elle n’en avait que deux’ (Le constitutionnel, 19 January 1847). 56 Table 5.1 is largely based on the information in Charlton and Wild, Théâtre de l’Opéra-­ Comique and therefore suffers from the same weakness: that it does not take account of revivals of earlier works during the period. The period 1834–1854 saw revivals of large numbers of opéras comiques originally premiered before 1830 (and many before the Revolution), especially during the period after 1840 which frequently saw two revivals of such repertory a year. 1841, for example, saw revivals of Grétry’s Richard cœur de lion (1784); Kreutzer’s Paul et Virginie and Dalayrac’s Camille (1791).

194  Between Opéra-Comique and Opéra-National

Mosquita la sorcière: Compositional History If January 1847 and the premiere of Ne touchez pas à la reine! represented a high point in Boisselot’s career, the death of his father shortly afterwards cut short any prolonged celebration. At Jean-Louis’ death in 1847, his elder son, Louis-Constantin, took over the manufacture of pianos, but the latter died in 1850, leaving Xavier to run the business until Louis-Constantin’s son, Franz, reached the age of majority and was able to take over in 1863. The death of his father just after the Ne touchez pas à la reine! premiere touched the composer greatly. Scribe and Vaëz took the view that a new libretto would reanimate the composer, and were writing Mosquita la sorcière between May and July 1847.57 Boisselot received the libretto around 15 July 1847, and by the time that he wrote to Scribe to acknowledge receipt, he had written the couplets for the aunt, two thirds of Mosquita’s air, (which I would like to preface with a chorus of bohemians whose idea I have in my head) and a large part of the drinking chorus that completes this act. Besides that, I found two or three motifs for cavatines or small melodies that I’m happy with.58 He concluded that if this rate continued, he would finish the score by 15 October the same year. There is no doubt that Mosquita la sorcière was intended for performance, not at the Opéra-National – where it eventually emerged – but at the Opéra-­ Comique, as an immediate successor to Ne touchez pas à la reine! Whether Boisselot, Scribe and Vaëz were able to present their opéra comique to ­Basset before the end of his management in May 1848 is not clear but – from what can be observed in the press – it seems as if Boisselot and Basset’s successor, Perrin, failed to agree on mounting Mosquita la sorcière at the Opéra-­Comique. Eugène de Fresne, writing in Le Pays just after the premiere of ­Mosquita la ­sorcière, ­summarised what was known three years after the event: Mosquita la Sorcière was written for the Théâtre Favart; it was to be performed there when the February Revolution delayed it. Tired of the difficulties that arose between M. Basset, the outgoing director, and M. Emile Perrin, the current director, M. Boisselot, who had taken part in this war, withdrew his work.59 57 ‘Boisselot, toujours à Marseille, aspire après son poème pour trouver des consolations dans le travail. Je vais lui écrire pour lui annoncer l’envoi prochain de ses deux premiers actes’ (letter from Vaëz to Scribe, 27 June 1847, F-Pn n.a.f. 22552, fols 98r-99v). 58 ‘les couplets de la tante entièrement, les deux tiers de l’air de Mosquita (que j’ai bien envie de faire précéder d’un chœur de bohémiens dont j’ai l’idée dans la tête) et une grande partie du chœur des buveurs qui termine cet acte. Outre cela, j’ai trouvé deux ou trois motifs pour des cavatines ou des petites mélodies dont je suis content’ (Letter from Boisselot to Scribe, 22 July 1847. F-Pn n.a.f. 22545, fols 142r-143v). 59 ‘Mosquita la Sorcière a été écrite pour le théâtre Favart; elle allait y être représentée ­lorsque la révolution de février la fit retarder. Las des difficultés qui s’élevèrent entre M. Basset,

Between Opéra-Comique and Opéra-National  195 According to de Fresne’s account, it looks very much as if Boisselot may have come very close to his October 1847 deadline for the completion of the score if it was ready for performance in February 1848. Basset’s bankruptcy and departure from the Opéra-Comique in May 1848 triggered a lawsuit with his successor that lasted until February 1851, and if it is true that Boisselot had taken sides in the dispute,60 this might well have had consequences for the fate of Mosquita la sorcière at the Opéra-Comique.61 There may also have been other difficulties with Mosquita la sorcière relating to its dimensions – a work in three acts comprising fifteen numbers and therefore congruent with the larger works (Haydée and Les mousquetaires de la reine) that represented the apogee of Basset’s achievement at the Opéra-Comique. To return to Table 5.1, and to examine the fate of two-act opéra comique in the 1840s, is to witness a recrudescence of the two-act form during Perrin’s administration, with two new two-act works, each receiving their premieres in 1849, 1852 and 1853. This move away from the emphasis on large-scale three-act works – temporary as it was – accords with Perrin’s requests to Scribe in a letter of 10 January 1851, only months before Mosquita la sorcière would premiere at a rival opera house: I have already expressed to you several times the desire for a work of a cheerful and lighter genre than those I have given for some time. For I fear that no one has abused great works, of great dimensions, with too much music, I fear especially that lassitude comes to the public... you alone have the hand strong enough and skilful enough to put the Opéra-Comique back in its true path from which it ends up too distant.62 Perrin acknowledged both that the Opéra-Comique had been mounting works ‘des grandes dimensions’ even under his own management and that Scribe had been writing such libretti, but he feared that such works may have been overused and may have damaged the Opéra-Comique’s success with its audience. This was a key moment for Perrin, at a point where the future repertory of the Opéra-National was emerging, and the new works there were looking very much like the larger products of the Opéra-­Comique

directeur sortant, et M. Emile Perrin, directeur actuel, M. Boisselot, qui avait pris parti dans cette guerre, retira son ouvrage’ (Le pays, 30 September 1851). 60 As suggested by de Fresne’s comment to the effect that the composer ‘avait pris parti dans cette guerre’ (ibidem). 61 The dispute between Basset and Perrin is detailed in the documents preserved in Paris, Archives nationales (hereafter F-Pan) F21 4673 (6c). 62 ‘Je vous ai témoigné déjà plusieurs fois le désir d’un ouvrage d’un genre gai et plus léger que ceux que j’ai donnés depuis quelques temps. Car je crains bien qu’on n’ait abusé des grands ouvrages, des grandes dimensions, du trop de musique, je crains surtout que la lassitude ne vienne au public … vous seul avez la main assez forte et assez habile pour remettre l’Opéra-Comique dans sa véritable voie dont il finit par trop s’écarter’ (Letter from Perrin to Scribe, 10 January 1851. F-Pn n.a.f. 22550, fols 162r-163v (partially cited in Yon, Eugène Scribe, 199)).

196  Between Opéra-Comique and Opéra-National from the late 1840s. His comments may have been an attempt to position his opera house in a way that would have avoided head-to-head competition with the newcomer. Under such circumstances, it is hardly surprising that – leaving aside any legal disputes with which Boisselot might have become involved – on purely aesthetic grounds, Mosquita la sorcière might not have been the sort of work that Perrin would have wanted to encourage at the Opéra-Comique at the moment at which Boisselot presented it. When Mosquita la sorcière was finally premiered at the Opéra-National in 1851, critics were quick to identify the background to the libretto. Viel pointed to Auber’s Les diamants de la couronne and La sirène, Adam’s ­Giralda and Mengozzi’s La dame voilée, to which Henri Blanchard added Don Juan and Le comte Ory.63 Such observations were tempered by Pommereux, The libretto... is an opéra comique and not a lyric work, as we expect to see on the boulevards.... This is a comic subject that cannot take root on the boulevard du Temple.... The country’s terrain of drama and melodrama requires more vigorous planting, with a much more energetic sap. What suits it best are pieces with strong situations, mixed with a good comedy, adorned with large choral masses. To try to acclimatize in this country comedies à ariettes is to waste time, to prepare its ruin,64 whose commentary takes the discussion in a completely new direction. Opposing an opéra comique to an ouvrage lyrique seems to be drawing a distinction between a clearly understood genre (opéra comique) and something much less clearly defined – an ouvrage lyrique, which is glossed first as ‘a Boulevard work’, then again as ‘drame’ and ‘mélodrame’. When Pommereux counterposed ouvrage lyrique with comédie à ariettes, it seems very much as if he was arguing for a fully sung genre that would have resulted in the immediate punishment of the Opéra-National for breaking the limits of its licence. This may simply be the consequence of a review appearing in the Revue et gazette des théâtres, rather than in the generalised or specialised music press, but it may also be the beginnings of the articulation of a tension between works with spoken dialogue and through-composed works that would not be resolved until Les pêcheurs de perles of 1863.65 63 Le ménestrel, 2 October 1851; Revue et Gazette musicale de Paris, 5 October 1851. 64 ‘Le poème … est un opéra-comique et non un ouvrage lyrique, comme nous comptions en voir représenter aux boulevards….. C’est là un sujet de comédie qui ne saurait prendre racine sur le boulevard du Temple….. Le terrain de ce pays du drame et du mélodrame veut des plantations plus vigoureuses, d’une sève bien autrement énergique. Ce qui lui convient, ce sont les pièces à situations fortes, mélangées d’un bon comique, ornées de grandes masses chorales. Chercher à acclimater dans ce pays des comédies à ariettes, c’est perdre son temps, préparer sa ruine’ (Revue et gazette des théâtres, 2 October 1851). 65 The manuscript libretto to Georges Bizet’s Les pêcheurs de perles betrays a curious mixture of opéra comique and continuous opera, and in the months during which Napoléon III was considering the legislation that would end the licensing period, Bizet rewrote all the

Between Opéra-Comique and Opéra-National  197 In the light of the concerns around the length of three-act opéra comique circulating in the late 1840s, the compositional history of the Mosquita la sorcière’s libretto takes on a remarkable significance. In the middle of this well-documented process, when the first two acts had been agreed by Scribe and Vaëz and sent to Boisselot, Vaëz sent a draft of the third act to Scribe: I’m sending you our third act. I have indicated where music should be placed in the places that I found favourable, but there are only four numbers, including the finale. We will see later whether we should add more. Sometimes in the ensemble[s] I did not prepare the verse of all the characters because our friend Boisselot often changes the rhythms, so he has the lyrics re-done.66 Apart from the general points of importance – that it was Vaëz and not Scribe who decided where the music should go, and that Boisselot was given to changing rhythms frequently – Vaëz’s concern that there were only four musical compositions in the act at this stage should not go unremarked. In the final version of the score, the third act is as follows: [Entracte] Dialogue 11. Duettino (Benita, Peblo):  ‘Celui qu’ici ton cœur préfère’ Dialogue 12. Sérénade (Benita, Peblo and chorus):  ‘Ah! Le ciel se colère’ Dialogue 13. Trio (Galliardo, Don Manoël, Carasco):  ‘Allons! Il faut me *Dialogue

suivre’

dialogue as recitative to create the first continuous opera at the opera house. On the basis of the libretti (censor’s libretto, F-Pan F18 737, deposited 11 August 1863; and two others in AJ13 1158), the Choudens 1863 piano-vocal score and the manuscript violin conductor score (F-Po Mat. 19. 1863), Hervé Lacombe has been able to show how the original form of the work consisted of a first act in the form of traditional opéra comique (four numbers alternating with dialogue) followed by two acts in which all the libretto is set as continuous music. Whether the document is interpreted correctly, and whether there might not have existed an earlier version of the entire work in the shape of an opéra comique, is impossible to tell from the surviving material. See Hervé Lacombe, ‘Les pêcheurs de perles de Bizet: contribution à l’étude de l’opéra français au xixe siècle’ (Thèse de Doctorat, Université François Rabelais de Tours, 1993), 46–52. The tabular material is reprinted in idem, Les voies de l’opéra, 316–317. See also Lesley A. Wright, ‘Les pêcheurs de perles: before the premiere’, Studies in Music 20 (1986) 27–45. 6 6 ‘Je vous envoie notre troisième acte. J’ai mis de la musique à faire dans les endroits qui m’ont paru favorables, mais il n’y a que quatre morceaux, y compris le final. Nous verrons plus tard si nous devons en ajouter. Quelquefois dans les ensemble[s] je n’ai pas fait le couplet de tous les personnages parce-que notre ami Boisselot change très souvent les rythmes, il fait alors refaire les paroles’ (Letter from Vaëz to Scribe, 27 June 1847; F-Pn n.a.f. 22552, fols 100r-101r).

198  Between Opéra-Comique and Opéra-National *14. Chœur et scène (Mosquita and chorus):  ‘Qu’il meure!’ *Dialogue *15. Duo final (Mosquita, Don Manoël and chorus):  ‘Que me

veux-tu’67

There are five compositions, of which one must have been added later than the date of Vaëz’s letter. This cannot have been the duo final, since Vaëz describes it as already written. Furthermore, his note that he used a form of shorthand in the ensembles because of Boisselot’s tendency to adjust rhythms post hoc suggests that the duettino and trio were probably already written. This leaves the chœur et scène as a number added later in the compositional process.68 Two weeks later, Vaëz again wrote to Scribe: I think I can reassure you of your fear that the situation will be monotonous in the last part. The scene between Mosquita and D. Manoël will gain some relief by the piece of music that we will place there, it only needs very slight alterations. Unless a more in-depth examination makes me change my opinion, it is a duo that will have to be put in. It does not seem to me that we have too many, and remember that there are none in this third act, if my memory serves me well.69 Unlike Scribe, Vaëz was clearly concerned about the second half of the act, dominated as it is by Mosquita and Don Manoël. This corresponds to the section from the dialogue after the trio (no. 13) to the end of the opera (marked with * in the plan above). So, when Vaëz alluded to ‘le morceau de musique que nous y ferons entrer’, he must have been referring to the chœur et scène (no 14). But even at this stage, he was thinking that this should be a duo (presumably between Mosquita and Don Manoël), although his parting comment – ‘si ma mémoire est fidèle’ – proved to be over-optimistic, since the act already at that stage had a duo final at the end and a duettino at the beginning.70 Vaëz’s further comments, and the overall context, are revealing. His view at this stage was that ‘Il ne me semble pas que nous en [de la musique] ayons trop’, and his concerns here were triggered in the first place by the fact that the act had only four compositions rather than five with which it ended up,

67 A la Mémoire de mon Maître / J. F. Le Sueur. / - / MOSQUITA / LA SORCIÈRE / OPÉRA-­ COMIQUE EN 3 ACTES, / POËME DE / MM. E. SCRIBE ET GUSTAVE VAËZ / Musique de / X. BOISSELOT. / Partition pour Chant et Piano / … / PARIS, / J. M ­ EISSONNIER FILS, 18, RUE DAUPHINE, / Mayence, B. Schott – Londres, Schott et Cie., 193–244. 68 Ibidem, 221–223. 69 ‘Je crois pouvoir vous rassurer sur votre crainte de voir monotone la situation dans la dernière partie. La scène entre Mosquita et D. Manoël prendra du relief par le morceau de musique que nous y ferons entrer, elle n’a besoin que de très légères retouches. – A moins qu’un examen plus approfondi ne me fasse changer d’opinion, c’est un duo qu’il faudra y mettre. Il ne me semble pas que nous en ayons trop, et souvenez qu’il n’y en [a] pas dans ce troisième acte, si ma mémoire est fidèle’ (Letter from Vaëz to Scribe, 7 July 1847. F-Pn n.a.f. 22552, fols 102r-103v). 70 Ibidem.

Between Opéra-Comique and Opéra-National  199 and presumably which Vaëz favoured. In other words, the optimal size of the act for Vaëz in the middle of 1847 (and it was he who was responsible for settling where the music went) entailed five compositions, and this is exactly the structure of the first and second acts of Mosquita la sorcière, and of most other three-act opéra comique of the period. There were other respects in which Mosquita la sorcière was reflecting current operatic practices. Boisselot was already being described as a composer who might write for the Opéra, but one of the most striking contexts for Mosquita was recent events at the Théâtre-Italien: the arrival of the works of Giuseppe Verdi in Paris. Among the most admired compositions in ­Mosquita la sorcière was the trio that closed the second act.71 The finale consists of a series of interrelated numbers (Table 5.2): Table 5.2  Boisselot, Mosquita la sorcière, no. 10 act ii finale A. Allegro (Chorus with interpolations for Morellos and Don Manoël; 6/8 E major): ‘Ah! c’est impayable’ B. Andante (Morellos, Peblo, Mosquita + chorus; 6/8 G major): ‘Silence!’ Recitative (Mosquita): ‘Pour toi’ C. Allegro (2/4 G major): ‘Nuit d’amour’ D. Andantino (Mosquita; 6/8 g minor): ‘Où vas-tu?’ E. Moderato (Mosquita; 3/4 g minor): ‘Ah! ah! ah!’ B. Andante (Mosquita, Don Manoël; 6/8 G major): ‘Monsieur Peblo!’ Recitative (Mosquita, Peblo): ‘Tu le vois’ D. Andantino (Mosquita, Don Manoël; 6/8 g minor): ‘Notre sort est lié’ E. Moderato (Mosquita; 3/4 g minor): ‘Ah! ah! ah!’ Recitative (Mosquita,; ‘Ah! c’est vous!’ F. Allegro agitato (Mosquita; 4/4 e minor): ‘Un amant, un époux’ G. Allegro moderato (Mosquita; 3/8 E major): ‘Mais tout me l’atteste’ H. Maestoso (Mosquita, Don Manoël, Peblo; 12/8 E major): ‘Sa jalouse fureur’ F. Allegro agitato (Mosquita; 4/4 e minor): ‘Cet esclave’ G. Allegro moderato (Mosquita; 3/8 E major): ‘Mais tout me l’atteste’ H. Maestoso (Mosquita, Don Manoël, Peblo; 12/8 E major): ‘Sa jalouse fureur’

71 MOSQUITA / LA SORCIÈRE / OPÉRA-COMIQUE EN 3 ACTES, 156–192.

200  Between Opéra-Comique and Opéra-National The complex of interlocking sections is indicated in Table 5.2, as is the structure of the trio (sections F-H which are repeated), which was the section selected for particular attention by the press. Léon Escudier found echoes of Verdi in the trio. The trio at the end of the act reminds us of Verdi. If this melody could have been given or repeated by the chorus, the effect would have been even more striking. As we can see, our young composers who follow the law of progress have some admiration for the compositions of the new school; and they are right, Verdi is one of those masters whose triumphs can be stopped momentarily by interested or systematic critics; but for us he is the star of the future.72 He might perhaps have argued that the effect would have been more like Verdi if the passage in question had been repeated by the chorus, but his view of the composer as ‘l’étoile de l’avenir’ – far from clear in 1851 – was congruent with his position as the composer’s major Parisian publisher and advocate. Escudier’s view was endorsed by Fiorentino and Prévost,73 but it was Blanchard who made clear exactly which passage it was that everyone thought was so Verdian. He wrote: It’s also a little bit like Verdi’s facile unisons; but the effect is no less grandiose and powerful. The modulation to the lower third major that intervenes at the end of this piece, from E major to C major, as far as we remember, this modulation, although very well known and even a little overused, is well placed here and has a very dramatic effect.74 and the passage in question is Example 5.7. This is the first statement of the passage. On its return, it is repeated exactly but extended to twice its length at a Largo tempo and made to feed into 72 ‘Le trio qui finit l’acte rappelle tout à fait la manière de Verdi. Si ce chant avait pu être dit ou redit par les chœurs, l’effet en eût été plus saisissant encore. Comme on le voit, nos jeunes compositeurs qui suivent la loi du progrès, ont quelque admiration pour les compositions de la nouvelle école; et ils ont raison, Verdi est un de ces maîtres dont on peut arrêter les triomphes momentanément par des critiques intéressées ou systématiques; mais pour nous il est l’étoile de l’avenir’ (La France musicale, 5 October 1851). 73 Le constitutionnel, 30 September 1851; Le moniteur universel, 7 October 1851. 74 ‘C’est bien un peu aussi dans la manière des faciles unissons à la Verdi; mais l’effet n’en est pas moins grandiose et puissant. La modulation à la tierce majeure inférieure qui intervient à la fin de ce morceau, de mi majeur en ut majeur, autant qu’il nous en souvient, cette modulation, bien que très-connue et même un peu usée, est bien placée en cet e­ ndroit et d’un effet très-dramatique’ (Revue et Gazette musicale de Paris, 5 October 1851).

Between Opéra-Comique and Opéra-National  201

Example 5.7  Boisselot, Mosquita la sorcière, no. 10 act ii finale, Maestoso ‘Sa ­jalouse fureur’ (published version) p. 185.

a short orage that ends the act. But Blanchard’s description of the passage is exact in all its details: the abrupt modulation to bVI (Blanchard’s major third below) is ubiquitous, but its power in this passage is undeniable, and even Verdi’s ‘facile’ unisons might not have met with disapproval in some quarters of the Parisian musical world. This is an important moment both for Verdi reception in Paris and for the status of opéra comique both at the institution of the same name (for where Mosquita was written) and for the Opéra-National (where it was performed).

202  Between Opéra-Comique and Opéra-National

Example 5.7  (Continued)

By 1851, or even by 1847, four operas by Verdi had been promoted at the ­Théâtre-Italien: Nabucco, Ernani – entitled Il proscritto, I due Foscari and those parts of I Lombardi alla prima crociata that had found their way into ­Jérusalem, premiered in Paris between October 1845 and November 1847.75 Whether Bois 75 Fundamental to any study of Verdi in Paris in the 1840s is Hervé Gartioux, ‘La réception des opéras de Verdi en France, entre 1845 et 1867, a travers une analyse de la presse’ (PhD diss., Université de Paris IV, 1999); see also idem, ed., La réception de Verdi en France: anthologie de la presse, 1845–1894, La musique en France au xixème siècle 4 (Weinsberg: Music-Edition Lucie Galland, 2001).

Between Opéra-Comique and Opéra-National  203 selot heard any of the performances of Jérusalem in 1847 is doubtful, since he was in Marseille from early in the summer, but evidence of the influence of the other three works is abundant in Mosquita la sorcière. Although unison part-writing for chorus is almost endemic in the three works he must have known, and although Verdi’s voicing of ensembles frequently involved parts in unison or octaves, the type of exposed unison writing for soloists is rarer. It is found, for example, in the duet and quartet in act ii of Il due Foscari, in the act i finale of Nabucco and the trios in acts i, ii and iv of Il proscritto. The act i trio of Il proscritto bears comparison with Example 5.7 (Example 5.8), where all three soloists are found in unison, and – as in the example from Mosquita la sorcière – presenting different texts. Also absent in this example is any alternation of gesture between the unison soloists and the orchestra. But the most striking point of comparison between Mosquita la sorcière and those Verdi works known in Paris is the act iii congiura from Il proscritto. This is an extended presentation of a single theme growing from pp at its opening to a climactic tutta forza at the end (Example 5.9) At its climax, Verdi uses gestures that Boisselot was keen to emulate. The chromatic scales linking phrases in the congiura (Example 5.10) are echoed in the very end of Boisselot’s trio and in the transition from the end of the H section to the reprise of the F section (Example 5.11). But such daring could also have its critics. Having praised Mosquita for the same qualities found in Ne touchez pa à la reine!, Bousquet also noted the same faults in Boisselot’s second opera: ‘the worst of these faults is an abuse of chromatic progressions throughout, where the melodic idea must necessarily yield to the exigencies of the situation’; his conclusion was that certain passages suffered from ‘a shining and annoying hue’.76 The particular qualities of the congiura from Il proscritto had called forth a good deal of favourable comment from Parisian critics. Gautier, in La presse, had written that The conspiracy scene and the conspirators’ chorus are of a dark, muffled colour, of a grandiose and terrible mystery that reminds us without imitating the oath of the daggers in Les Huguenots. The final trio is treated in a pathetic way; the anguish of the two lovers, who touched happiness and found on the threshold of their paradise the terrible and threatening figure of the avenger, could not be rendered with a deeper feeling.77 76 ‘le plus grave de ces défauts est un abus des successions chromatiques partout où l’idée mélodique doit nécessairement céder aux exigences de la situation … ‘une teinte criarde et agaçante’ (L’illustration, 4 October 1851). 77 ‘La scène de la conjuration et les chœurs des conspirateurs sont d’une couleur étouffée et sombre, d’un mystère grandiose et terrible qui rappelle sans imitation le serment des poignards dans les Huguenots. Le trio final est traité d’une manière pathétique; l’angoisse des deux amans, qui touchaient au bonheur et qui trouvent sur le seuil de leur paradis la figure terrible et menaçante du vengeur, ne pouvait être rendue avec un sentiment plus profond’ (Gartioux, Réception de Verdi en France, 79).

Example 5.8  Verdi, Il proscritto (Ernani), no. 5 finale primo, bars 231–247.

Example 5.8  (Continued)

Example 5.8  (Continued)

Example 5.8  (Continued)

Example 5.9  Verdi, Il proscritto (Ernani), no. 11 congiura, bars 106–109.

Example 5.10  Verdi, Il proscritto (Ernani), no. 11 congiura, bars 126–131.

210  Between Opéra-Comique and Opéra-National

Example 5.10  (Continued)

The obvious background in act iv of Les Huguenots to the congiura from Il proscritto may explain Jules de Premaray’s comment to the effect that ‘the second act alone is a complete work. There, there is … a splendid trio that Meyerbeer could have signed’, which raises the question as to the real origins of the final pages of Mosquita la sorcière’s act ii trio.78 There is at least the possibility that the hostile journalistic discourses surrounding Verdi’s use of unisons meant that any composition that employed such resources would be tagged ‘Verdian’, and such authors might have then blinded themselves to the more local background for the Mosquita trio in Meyerbeer’s justly celebrated fourth act. The claims that Boisselot was writing opéra comique that threatened to trespass on the generic domain of grand opéra when he produced Ne touchez pas à la reine! were reiterated in the reviews of Mosquita la sorcière; the act ii trio was the main piece of evidence. Fiorentino’s view of the finale was that act ii closed with

78 ‘Le 2e acte, à lui seul, est une œuvre complète. Il y a là … un trio splendide que pourrait signer Meyerbeer’ (La patrie, 29 September 1851).

Between Opéra-Comique and Opéra-National  211

Example 5.11  Boisselot, Mosquita la sorcière, no. 10 act ii finale, Maestoso ‘Sa ­jalouse fureur’ [page 186, bar 5 to page 187, bar 9].

a big dramatic trio that received a triple round of applause. This piece, broadly treated, goes beyond the framework of opéra de genre and aims at grand opéra. What impressed the audience was one of the great sound effects that Verdi has brought into fashion [emphasis added]. Blanchard prefaced his comments, cited earlier, with the line ‘the final trio of this act is in a broad and dramatic style and resembles grand opéra’.79 79 ‘un grand trio dramatique qu’on a couvert d’une triple salve d’applaudissemens. Ce morceau, largement traité, sort du cadre de l’opéra de genre et vise au grand-opéra. Ce qui

Example 5.11  (Continued)

Between Opéra-Comique and Opéra-National  213 With references to Verdi – whose Jérusalem had only recently been heard at the Paris Opéra – and Meyerbeer, the institution’s mainstay, it was inevitable that such comparisons would have been drawn. However, in the case of Mosquita la sorcière, the comments were rather less heavily loaded than in the case of Ne touchez pas à la reine! since in the latter case it was a question of a composer at the Opéra-Comique writing in the style of the Opéra – which was part of a critical discourse that went back several years and would continue until the mid-1850s at least – whereas for Mosquita – now transferred to the O ­ péra-National – the generic boundaries were significantly more permeable and the identification of grand opéra characteristics more a question of praise than blame. Boisselot’s position vis à vis opéra comique c1850 is characterised by ambivalences and paradoxes that serve to illuminate a critical moment for the genre at mid-century. On the one hand, Ne touchez pas à la reine! was both admired and criticised for its ambitious style, and a register that brought it into close alignment with grand opéra; on the other, its dimensions were those of a classic opéra comique, perhaps better suited to a work in two acts (or even one), with its eleven numbers setting it apart from its larger contemporaries, Auber’s Haydée and Halévy’s Les mousquetaires de la reine. Reasons for this paradox are not hard to find. Boisselot’s score for Ne touchez pas à la reine! was old by the time it came to performance, and the libretto even older. The chronological frame for the work, limited by the origins of its libretto and the lengthy gestation of the score – especially when this coincided with the first performances of Verdi in Paris – explains its ambitious and well-worked nature. Mosquita la sorcière, by contrast, had more in common with the two opéras comiques that were its near contemporaries, sharing almost exactly the same number of compositions across three acts as both Haydée and Les mousquetaires de la reine, and had it survived the changeover of management in the spring of 1848, would almost certainly have been very well received at the Opéra-Comique. But its transfer to the OpéraNational raised other key questions about the new opera house’s repertory. Critics were unanimous in their view that the strengths and perceived weaknesses of Boisselot’s style were audible in both operas; they pointed to the composer’s skill as a melodist. Premaray could write of ‘a richness of broad and original melody’ in Mosquita la sorcière80 as easily as Prévost could attribute the clarity of melodic writing in Ne touchez pas à la reine! to the composer’s Conservatoire tuition with Lesueur.81 Similarly, while all commentaries on Ne touchez pas à la reine! pointed to Boisselot’s adroit and generous use of the chorus in individual numbers, this skill was elevated to a general cause for praise in the reception of Mosquita la sorcière. a frappé le public c’est un de ces grands effets de sonorité que Verdi a mis à la mode’ (Le constitutionnel, 30 September 1851); ‘Le trio final de cet acte est d’un style large et dramatique, et sent de grand opéra’ (Revue et Gazette musicale de Paris, 5 October 1851). 8 0 ‘une richesse de mélodies larges et originales’ (La patrie, 29 September 1851). 81 Le moniteur universel, 18 January 1847.

214  Between Opéra-Comique and Opéra-National Mosquita la sorcière commands an important position in the history of opéra comique c1850 as a work that was conceived for the Opéra-Comique but premiered at the Opéra-National. According to Fresne, it only moved from the Opéra-Comique to the Opéra-National ‘having been subject to a few changes’, but Boisselot’s working papers for the opera betray nothing of what these changes might have been.82 The Opéra-National, in the form it took in 1851, stood in a tradition that originated not only in the institution of the same name that had run from 1847–1848 but in a long line of attempts to establish a Troisième théâtre lyrique that went back at least to the attempts to mount opéra comique at the Odéon in the 1820s. The licence that the Opéra-National enjoyed in 1851 was essentially the one that it had received in 1847, and this was the point at which opéra comique was permitted to form part of the Opéra-National’s repertory. A comparison of the cahiers de charges for the two theatres makes the point clearly (Table 5.3).83 Table 5.3  Cahiers des charges, Opéra-National, 1 April 1851 and 12 January 1847 Cahier des charges, 1 April 1851

Cahier des charges, 12 January 1847

Le répertoire se composera d’ouvrages nouveaux en un ou plusieurs actes, mêlés de musique nouvelle, avec chœurs, airs duos, trios morceaux d’ensemble, et en général, tous les développements que comporte le genre lyrique. Il aura la faculté d’y ajouter des divertissements chorégraphiques. Il pourra faire représenter des traductions d’ouvrages lyriques étrangers à la condition, toutefois, que le nombre de ces ouvrages n’excède pas deux par année. Il pourra représenter les ouvrages des auteurs vivants appartenant aux répertoires des autres théâtres lyriques, dix années seulement après leur première représentation sur ces théâtres. Il en sera de même à l’égard des ouvrages morts et non encore tombés dans le domaine public. Il pourra représenter les ouvrages tombés dans le domaine public de manière, toutefois, que le nombre d’actes résultant de ces divers emprunts faits aux théâtres français et étrangers serait dans la proportion d’un tiers avec le nombre des ouvrages nouveaux. Ce compte sera fait par période de deux années.

Le répertoire se composera 1 d’ouvrages nouveaux en un ou plusieurs actes, en prose ou en vers, mêlés de musique nouvelle, avec chœurs, airs, duos, trios, morceaux d’ensemble, et en général, avec tous les développements que comporte le genre de l’opéra comique. 2 Des ouvrages de l’ancien répertoire de l’Opéra-Comique tombés dans le domaine public. 3 Des ouvrages du répertoire moderne de l’Opéra-Comique, joués antérieurement à 1840, et délaissés par ce théâtre. 4 Des traductions des opéras et ouvrages étrangers, mêlé de musique et de chant. Des divertissements de danse, et de petites ballets en un acte.

82 ‘après avoir subi quelques modifications’ (Le pays, 30 September 1851). 83 The two documents are found in F-Pan F21 1119 and 1120 respectively.

Between Opéra-Comique and Opéra-National  215 The 1847 cahier des charges was the result of hard-won negotiations that went back at least three years and represented a major breakthrough for supporters of the idea of a Troisième théâtre lyrique. But as Table 5.3 shows, not only did the 1851 cahier des charges replicate much of the detail of its 1847 predecessor, but it subtly switched the formulation from ‘tous les développements que comporte le genre de l’opéra comique’ to ‘tous les développements que comporte le genre lyrique’. As far as existing operas were concerned, the 1851 document put in place a ten-year moving wall to replace the fixed date of 1840 before which existing opéras comiques could be performed in the 1847 document. The opportunities represented by the 1847 cahier des charges are exemplified by the only new three-act work that was mounted in the short period before the Opéra-National collapsed: Maillart’s Gastibelza, ou Le fou de Tolède, to a libretto by Adolphe d’Ennery and Eugène Cormon. In terms of its dimensions, it has much in common with the larger types of contemporary opéra comique: three acts with six, six and five compositions, respectively, and every number separated by dialogue. It distances itself from opéra comique promoted at the institution of the same name in that there were only two compositions based on couplets or similar small forms, and both are assigned to a minor character; both make use of dialect and onomatopoeia. Also unlike contemporary opéra comique, Gastibelza presented Opéra-National audiences with a divertissement consisting of two characteristic Spanish dances: La Toledana and La saltarella. Furthermore, the act iii chœur des moines is the backdrop for a cortège, which – as has been seen in the discussion of the maverick inclusion in Ne touchez pas à la reine! – is more a characteristic of grand opéra.84 In the context of such an ambitious work as Gastibelza, even Mosquita la sorcière begins to look anomalous, especially when compared to the works specifically written for the Opéra-National after it reopened in 1851. The three works in question, David’s La perle de Brésil, Adrien Boieldieu’s La butte des moulins and Adam’s Si j’étais roi, all exhibited the same ambitious scale as Gastibelza and all included fully developed choreography. Although Si j’étais roi’s divertissement was modest, the one for La perle de Brésil consisted of three pieces of dance music and La butte des moulins no less than seven. Both La perle de Brésil and La butte des moulins also featured cortèges. Fiorentino refers specifically to airs de danse in his review of Mosquita la sorcière, referring especially to the chœur des jeunes filles in act i (‘Voici la ­bienheureuse vierge’). In its published form, this is clearly nothing more than a danced chorus, of the sort routinely promoted at the Opéra-Comique; 84 Bibliothèque Populaire / GASTIBELZA / OPÉRA DRAMATIQUE EN TROIS ACTES, / Poème de MM. Dennery & Cormon, / MUSIQUE / DE / A. MAILLART. / Prix net : 12 fr. / - / PARIS / AU BUREAU CENTRAL DE MUSIQUE, 95 RUE RICHELIEU. / MILAN, RICORDI [B.C. 1073].

216  Between Opéra-Comique and Opéra-National it is therefore possible that Fiorentino’s airs de danse might fall into the category of ‘quelques modifications’ to which de Fresne alluded in his review of the work. But nothing in Mosquita la sorcière comes close to the elaborate divertissements that characterise newly composed works for the Opéra-National. In terms of the complexity of individual compositions, Mosquita bears comparison with such works as Haydée and Les mousquetaires de la reine, the jewels in the crown of the Opéra-Comique, and its migration from that institution to the Opéra-National was greatly facilitated by the presence there of soloists with experience at the Opéra-Comique itself, in particular Mlle Rouvray and M. Grignon.85 Nevertheless, in comparison with works composed for the Opéra-National it stands apart as a critical index to the ways in which opéra comique had diverged generically between the Opéra-National and Opéra-Comique. The relationship between the Opéra-National and Opéra-Comique was one that came in for much discussion as the former premiered Mosquita la sorcière in September 1851. Viel expressed the view that the Opéra-National was to the Opéra-Comique what the Théâtre de la Porte Saint-Martin was to the Comédie-Française.86 Exactly what he meant by this is not entirely clear. Certainly, during the 1830s and 1840s, the Porte Saint-Martin had been Le citadelle du romantisme where much Dumas and Hugo found a home, and if this meant that the Opéra-National was to mount more ambitious works than the Opéra-Comique, this would have been congruent with Perrin’s concerns to rein in large three-act works at the Opéra-Comique that have already been noted. Other voices were dismayed by the fact that it seemed that Mosquita la sorcière had threatened, in Pommereux’s words, ‘the spectacle of a useless rivalry with the Opéra-Comique’.87 If there was at least a certain amount of clear blue water between La perle de Brésil, Si j’étais roi and La butte des moulins, and those works by contemporaries at the Opéra-Comique, some were calling for a greater generic differentiation. Julien Lemer, for example, came up with a remarkable proposal: …. we seriously believed, this time, that we were going to try to create a new genre and bring the feeling and taste of music into the audience of this neighbourhood dedicated to mélodrame. In order to do this, we believe that it is absolutely essential to adopt a transitional genre that holds a considerable place, where music is for some time still only the accessory. Nothing could be easier; it is enough to address one of the clever authors that the public of mélodrame knows 85 See Blanchard’s comments in the Revue et Gazette musicale de Paris, 5 October 1851. 86 Le ménestrel, 2 October 1851. 87 ‘le spectacle d’une inutile rivalité avec l’Opéra-Comique’ (Revue et gazette des théâtres, 2 October 1851).

Between Opéra-Comique and Opéra-National  217 and reveres, to commission from him a drama conceived, written as if he were to be represented as it is; then to make a composer choose the three or four musical situations that can not fail to find themselves in any well-made drama, and finally to play an interesting, pathetic piece, highlighted by two or three choruses, a finale, at most a duet, couplets and above all an easy chanson, written in one of these rhythms destined to become rapidly popular. Music, served in this wise and prudent way, would be received with favour, thanks to the interest. Gradually, little by little, it would be given a greater place among the works that followed, and soon the day would come where the public of the quartier of the Temple would understand and would like opéra comique and even grand opéra. The taste for opéra comique as we see it today has not penetrated the bourgeoisie in any other way. It started with comédies à ­ariettes sixty years ago, it was still there; from the ariettes, it came to airs, nocturnes, duets, then quartets, choruses, ensembles, and finally today, it understands and applauds a grand finale at the Salle Favart just as willingly as at the Rue Lepelletier [emphasis added].88 Lemer’s proposal is worth quoting at length, since it stands at the end of a sequence of other attempts to create new genres that might be supported by the Troisième théâtre lyrique. Lemer’s genre transitoire would have taken a spoken play as its basis, to which a composer would have added a highly limited number of musical compositions of a type that would have appealed to Boulevard audiences. At one level, this is akin to the creation

88 …. nous avons cru sérieusement, cette fois, qu’on allait essayer de créer un genre nouveau et de faire pénétrer dans le public de ce quartier voué au mélodrame le sentiment et le goût de la musique.   Pour ce faire, nous pensons qu’il est absolument indispensable d’adopter un genre transitoire qui tienne une place considérable, où la musique ne soit, pendant quelque temps encore que l’accessoire. Rien n’est plus facile; il suffit de s’adresser à l’un des faiseurs habiles que le public des mélodrames connaît et révère, de lui commander un drame conçu, agencé, écrit comme s’il devait être représenté tel quel; puis de faire choisir par un compositeur les trois ou quatre situations musicales qui ne peuvent manquer de se trouver dans tout drame bien fait, enfin de jouer une pièce intéressante, pathétique, relevée par deux ou trois chœurs, un final, un duo tout au plus, des couplets et surtout un chanson facile, écrite dans un de ces rythmes destinés à devenir promptement populaires. La musique, servie dans cette sage et prudente mesure, serait reçue avec faveur, grâce à l’intérêt. Peu à peu, de proche en proche, on lui donnerait une place plus grande dans les ouvrages qui succéderaient, et bientôt un jour viendrait où le public du quartier du Temple comprendrait et aimerait l’opéra comique et même le grand opéra. Le goût de l’opéra comique tel que nous le voyons aujourd’hui n’a point pénétré d’une autre façon dans la bourgeoisie. Elle a commencé par la comédie à ariettes, il y a soixante ans, elle en était encore là; des ariettes, elle en est venue aux airs, aux nocturnes, aux duos, puis aux quatuors, aux chœurs, aux ensembles, et enfin aujourd’hui, elle comprend et applaudit un grand final à la salle Favart tout aussi volontiers qu’à la rue Lepelletier [emphasis added] (La sylphide, 10 October 1851).

218  Between Opéra-Comique and Opéra-National of the vaudeville avec airs nouveaux and opéra de genre demanded by the licences for the Théâtre de la Renaissance in the late 1830s.89 A decade later the position was rather different, for while the attempts to create two new genres for the Théâtre de la Renaissance were designed to avoid competition with the Opéra and the Opéra-Comique (although they failed in both cases and triggered endless litigation), ­L emer’s proposal was made in the context of an environment where cultivation of the same genre was entirely legitimate. And in his proposal may be seen a significant change of focus, away from avoiding competition between opera houses and towards satisfying the aspirations, abilities and interests of new audiences. The impact of Boisselot’s two opéras comiques differs in the two instances. Although Mosquita la sorcière was clearly a work on the boundaries of what could fit into an opéra comique, there were those – Taxile Delord, for ­example – who thought that in parts it ‘even appears perhaps stronger than the situation’,90 while Berlioz pointed how Boisselot, ‘... instead of making a large piece,... links together several different small pieces, on which the listener scatters and wears themself out in sheer loss’.91 Berlioz had already commented on how Ne touchez pas à la reine! enjoyed ‘musical situations in this piece that are numerous, varied and of a more elevated genre than those in most opéras comiques’,92 and most other critics – Escudier, M ­ onselet and Bousquet – had agreed, despite the fact that Ne touchez pas à la reine! was in many respects much less ambitious than its contemporaries at the Opéra-Comique. On the other hand, others took the view that Boisselot was reviving the ancient usage which saw in music a brilliant accessory given to speech, a harmonious form for the setting aside of feeling or idea, but not a kind of abstract pleasure, tasted regardless of the circumstances surrounding it, and to which one does not ask to lend the slightest relief to the effects of the scene.93 Here, de Sévelinges was arguing that Boisselot was returning to a view of opéra comique that predated the 1789 Revolution, where the genre was 89 See above, 167 and note 7. 90 ‘semble même peut-être plus fort que la situation’ (Le charivari, 29 September 1851). 91 ‘au lieu de faire un grand morceau, … enchaîne ensemble plusieurs petits morceaux ­d ifférens, sur lesquels l’auditeur s’éparpille et se fatigue en pure perte’ (Journal des débats, 30 September 1851). 92 ‘situations musicales, dans cette pièce, [qui] sont nombreuses, variées, et d’un genre plus élevé que celles de la plupart des opéras-comiques’ (ibidem, 24 January 1847). 93 ‘l’usage ancien qui dans la musique voyait un accessoire brillant donné à la parole, une forme harmonieuse pour la mise au dehors du sentiment ou de l’idée, mais non une sorte de plaisir abstrait, goûté indépendamment des circonstances qui l’entourent, et auquel on ne demande pas de prêter le moindre relief aux effets de la scène’ (La gazette de France, 21 January 1847).

Between Opéra-Comique and Opéra-National  219 indeed a play with music, a comédie mêlée d’ariettes. And when Gautier said that ‘the music of M. Boisselot is of the old French school and sticks closely to the verse so as not to sustain mediocrity!’, he was largely saying the same thing.94 The tendencies towards grand opéra that had been identified in both Mosquita la sorcière and in Ne touchez pas à la reine! were crystallised by Premaray when he wrote that ‘M. Boisselot’s place is marked at the Opéra where one day he will become the saviour and make the fortune’.95 And while Boisselot’s surviving sketches contain fragments of at least three ­further works, the composer’s career was derailed first by the death of his father in 1847 and then by that of his elder brother in 1850.96 What successes he had enjoyed at the Opéra-Comique and Opéra-National at mid-century were largely forgotten as he helped turn the Boisselot piano company into one of France’s finest.

94 ‘la musique de M. Boisselot est de la vieille école française et se colle étroitement au vers de façon à n’en pas supporter la médiocrité!’ (La Presse, 18 January 1847). 95 ‘La place de M. Boisselot est marquée au grand Opéra dont un jour il deviendra la providence et fera la fortune’ (La patrie, 29 September 1851). 96 Apart from the piano-vocal score of a ballet entitled Miranda (F-Pn MS 4376), there are groups of material that seem to relate to three different works: F-Pn MS 4387, 4379, 4366–4367 all relate to a single composition, while MS 4354 and 4375 could be either part of the same further work or parts of two other operas.

Premier entr’acte Les ultramontains

Most major cities in Europe enjoyed, in addition to venues for theatre and staged music in the vernacular, either a French theatre or an Italian opera house, and the Parisian Théâtre-Italien was a magnet for singers and composers from the peninsula throughout the nineteenth century. Indeed, in at least two cases, Rossini and Donizetti, Parisian careers that had started at the Théâtre-Italien rapidly moved the composers into careers writing for other houses. The near-annexation of the Théâtre-Italien by Rossini started with L’italiana in Algeri in 1817, with other comic works following in 1819 (Il barbiere di Siviglia) and 1820 (Il turco in Italia), by which time Rossini’s serious operas were taking hold, if anything, more quickly:1 Otello in 1821, Tancredi, Elisabetta, regina d’Inghilterra and Mosè in Egitto all in 1822. By the middle of the 1820s, Rossini could hardly keep up with the demand: Zelmira and Semiramide were only months old before they arrived in Paris.2 Donizetti, Bellini and Mercadante were all courted by Paris around the same time: Marino Faliero and I Puritani were both premiered during the 1834–1835 season at the Théâtre Italien, and Mercadante’s I briganti – ­although not premiered there until l836 – was a continuation of the same campaign to recruit what were considered the decade’s three leading Italian composers.3 Bellini, of course, died in Puteaux, just outside Paris, in later

1 For a critique of the too-casual claims of a Rossini craze in the early nineteenth century, see Everist, ‘Cosmopolitanism and Music for the Theatre’. 2 For the history of the Théâtre-Italien up to 1830, see Janet Johnson, ‘The Théâtre Italien and Opera and Theatrical Life in Restoration Paris’ 3 vols [paginated consecutively] (PhD diss., University of Chicago, 1988), and for the 1830s, Philip Gossett, ‘Music at the Théâtre Italien’, Music in Paris in the Eighteen-Thirties, ed. Peter Bloom, La Vie musicale en France au xixe siècle 4 (Stuyvesant, NY: Pendragon, 1987) 327–364. There is little modern study of the house in the 1840s and later, unless it concerns specific composers, mostly Verdi. See Octave Fouque, Histoire du Théâtre-Ventadour, 1829–1879: Opéra-­Comique-­ Théâtre de la Renaissance-théâtre italien (Paris: Fischbacher, 1881); and Albert Soubies, Le ­T héâtre-Italien de 1801 à 1913 (Paris: Fischbacher, 1913). 3 The history of Donizetti and Bellini at the Théâtre-Italien and the relative success of Marino Faliero and I Puritani are well known. The last two chapters (of six) of William Ashbrook, Donizetti and his Operas (Cambridge: Cambridge University Press, 1982) are

222  Premier entr’acte 1835, and Mercadante never really expressed much interest in the city after I briganti. Donizetti returned to Paris in 1838, spent most of the last years of his life there (with extended periods in Vienna) and assimilated to Paris and the musical culture of its theatre. Rossini left Paris in 1836, to return and spend the rest of his life there in 1855. But he had also to return in 1844, and the resulting torrent of rumour and wishful thinking around a second grand opéra to follow Guillaume Tell grew to astonishing proportions, with the result being the first of a series of Italian works translated for the Opéra: Otello, followed by Donizetti’s Lucie de Lammermoor in 1846, a Rossini pasticcio, Robert Bruce, in the same year and Verdi’s reworking of I Lombardi alla prima crociata as Jérusalem the next year. By the mid- to late 1840s, the Opéra had works by Rossini, Donizetti and Verdi in performance, but of these only Lucie de Lammermoor lasted into the 1850s, by which time the first of Verdi’s Italian operas translated into French had arrived – Louise Miller appeared in 1853. Even Verdi’s first original work for the Opéra – the 855 Vêpres siciliennes – could not match the enduring success of Il ­trovatore – Le trouvère – which started an unbroken run of nearly twenty years in 1857. To suggest, then, that the Théâtre Italien had a monopoly on Italian opera in Paris would be a mistake and to misunderstand a history that was fitful but not discontinuous. Italian works in French translation had been one of the ways in which the Théâtre de l’Odéon had managed to carve out an artistic niche in the 1820s with versions mostly of Rossini. The same was true of both the Théâtre de la Renaissance in the late 1830s and the ­Théâtre-Lyrique throughout the Second Empire. Also during the 1820s, while the Opéra and Théâtre Italien were under shared management, Rossini was able to mount productions of Mosè in Egitto and Maometto II in heavily revised versions in French and adjusted to the conventions of the Opéra. The production of Otello in 1844 and Rossini’s visit to France therefore take on a particular importance and are the subject of Chapter 7. Cultural exchange characterises all four essays in this section. The position is complicated by the fact that – for perhaps one of Rossini’s most famous single numbers, the ouverture to Guillaume Tell – not only was an Italian composer writing his first completely original work for the Opéra, but he was doing it in the light of one of Beethoven’s symphonies in turn promoted at a Parisian concert series with French players. Such a cultural fusion of Italian and ‘German’ elements within a Parisian context would dominate the next few decades at the Opéra with the works of Giacomo

devoted to the composer’s years in Paris; see also Herbert Weinstock, Vincenzo Bellini: His Life and Operas (New York: Knopf, 1971). For Mercadante, see Francesca Placanica, ‘Saverio Mercadante and France (1823–1836)’ (PhD diss., University of Southampton, 2013), 133–184.

Premier entr’acte  223 4

Meyerbeer, beginning with Robert le diable in 1831. By contrast, Chapter 8 examines an instance of the export of a Parisian opera libretto – Auber’s 1831 Le philtre – to Italy to serve as the basis for Donizetti’s L’elisir d’amore (which in turn would return to Paris in 1839). The early twenty-first ­c entury would have no difficulty attributing value to Donizetti’s melodramma ­giocoso while expressing complete ignorance of its French counterpart. But not only was Le philtre a key part of the tradition of petit opéra (already discussed in ­Chapter 2), and continued in a more or less unbroken tradition for over thirty years, but as late as 1900 it was still being advantageously compared with the Donizetti.5 Chapter 8 switches focus to two of the principal agents in the cultural transfer of Italian music into the Parisian theatre: Alphonse Royer and Gustav Vaëz. Responsible for many of the Frenchified Italian works just discussed, their work went well beyond the domains of what might be considered translation or even ‘opera’, and this background is important for their collaborations with Rossini, Donizetti and Verdi.

4 By the time of his arrival in Paris, Meyerbeer was already recognised as a Prussian composer who had spent nearly a decade in Italy, and for whom his Italian output was the most important part of his œuvre. For an outline of the composer’s sojourn in Italy, see Andrew Everett, ‘“Bewitched in a Magic Garden”: Giacomo Meyerbeer in Italy’, Donizetti Society Journal 6 (1988) 163–192, and for the influence of Italian melodic types on Meyerbeer’s early works for the French theatre, see Steven Huebner, ‘Italianate Duets in Meyerbeer’s Grand Operas’, Journal of Musicological Research 8 (1989) 203–258. 5 Ebenezer Prout, ‘Auber’s Le philtre and Donizetti’s L’elisir d’amore: A Comparison’, Monthly Musical Record 30 (1900), 25–27, 49–53, 73–76.

6 Beethoven and Rossini Opera and concert at the end of the Restoration

Paris: Monday 3 August 1829, early evening. For the world premiere of a new work to take place in August, when most of Parisian society should have been somewhere cooler, was entirely out of the ordinary. But tout Paris was all in the city, ready and waiting – as they had been for the best part of a year – for Rossini’s first serious production for the Paris Opéra. Yes, there had been three previous appearances, but none had delivered what Parisian audiences were expecting: a new opera, written to a new French libretto, that was untainted by any origin in Italian works – as the previous three had been.1 And the very public disagreements between Rossini and the management of the Paris Opéra had kept the new work very much in the public mind – of course, that was why it had been delayed until the most sultry part of the summer.2 When something approaching silence fell in the Salle Le Peletier, the audience looked expectantly to the orchestra – at this point still fully visible in front of the stage – to begin the overture. They could reasonably have expected an overture to begin, much as had the previous year’s major presentation, Auber’s La muette de Portici with dramatic fortissimo diminished seventh chords, or maybe with the sort of tutti call to attention that one associates with Rossini’s overtures to Italian operas, or – as in the previous year’s Comte Ory – an introduction replacing the overture itself. Looking over at the orchestra, something seemed wrong: François-­ Antoine Habeneck, the distinguished conductor of the opera orchestra, was sitting with his violin bow on his knees showing no sign of action, and

1 These works were Le siège de Corinthe (9 October 1826), Moïse et Pharaon, ou Le passage de la Mer Rouge (26 March 1827), et Le comte Ory (20 August 1828). 2 On the dispute between the administration and the Maison du roi, see the documents in Bruno Cagli and Sergio Ragni (eds.), Gioachino Rossini: Lettere e documenti III: 17 ottobre 1826 – 31 dicembre 1830 (Pesaro: Fondazione Rossini, 2000), 444–467, and the summary in Richard Osborne, Rossini, The Master Musicians (London: Macmillan, 1986; 2nd edn 2007), 119.

Beethoven and Rossini  225 no one else seemed like they were about to perform.3 The only exception seemed to be Louis Norblin, the principal cellist, who was ready to play as soon as Habeneck gave the signal. When that signal came, the audience had little idea what had hit them. The best part of five hours later, as the audience came out of the Salle Le Peletier, they were still dumbstruck by Rossini’s overture. The overture to Guillaume Tell is too well known to allow an understanding of the impact the work made on the audience’s horizon of expectations in 1829 – whether it’s from the movie Happy Days (1929), filmed exactly 100 years after the opera’s premiere, or from more recent uses in Hearts in Atlantis (2001), the 2006 remake of The Pink Panther and even in a 2005 episode of a television series entitled Joey. Added to this are all 169 episodes screened between 1949 and 1957 of the TV series The Lone Ranger, which indelibly engraved the final section of the overture to Guillaume Tell on the memories of anyone who saw it. Rossini scholars will need no reminding that the overture to Guillaume Tell consists of four sections played without a break: the ensemble for solo cellos, one of Rossini’s finest storm sequences, an elaborate pastoral section and a closing allegro vivace described variously in the contemporary press as a galop, a march or a pas redoublé. The overture lasts around twelve minutes.4 Nothing had prepared the 1829 audience for the overture to Guillaume Tell. Nothing in any of the works prepared for the Paris Opéra was remotely like it, and neither was any opera overture that had been performed at the Théâtre-Italien (and there were plenty of those by 1829). Rather than demonstrate this very obvious negative, a couple of contemporary critics may be given voice. The reporter for La gazette de France stressed the overture’s originality, and he singled out the opening and closing movements as particularly innovative. Le courrier français spoke as follows: The overture … genuinely electrified the audience. The allegro of this symphonie is of an incredible originality and verve. Never has the maestro been better inspired. A thunder of applause broke out in the house after this wonderful piece.5 3 On Habeneck’s career, see Nicolas Southon, ‘L’émergence de la figure du chef d’orchestre et ses composantes socio-artistiques: François-Antoine Habeneck (1781–1849), La naissance du professionalisme musical’ (PhD diss., Université François-Rabelais Tours, 2008). 4 The overture is edited as part of the critical edition in M. Elizabeth C. Bartlet (ed.), Guillaume Tell, Opéra en quarter actes di Victor Etienne de Jouy e Hippolyte Louis Florent Bis; musica di Gioachino Rossini, 4 vols, Edizione critica delle opere di Gioachino Rossini, Sezione prima 39 (Pesaro: Fondazione Rossini, 1992). 5 ‘L’ouverture … a véritablement électrisé l’assemblée. L’allegro de cette symphonie est d’une originalité, d’une verve incroyables. Jamais le Maestro ne fut mieux inspiré. Un tonnerre d’applaudissemens a éclaté dans la salle après cet admirable morceau’ (Le courrier français, 5 August 1829).

226  Beethoven and Rossini Other accounts claim that the last pages of the overture were not in fact audible because the applause had begun well before the end and probably obscured Rossini’s final coup de théâtre: the famous general pause ­twenty-four measures before the end.6 Of course, there were complaints: some objected to the use of what was effectively chamber music in the opera house,7 others simply complained that the overture was too long (and most claimed that the whole opera was too long triggering Rossini’s cuts after the premiere) but general opinion was that the overture was new, original and electrifying in its effect. But the critics had one serious difficulty, and that was identifying any sort of precedent or origin for the overture, and this sort of discursive attack was conventional in any review of a premiere in Paris during the Restoration or July Monarchy. The veteran critic Castil-Blaze gamely pointed to the overture to Méhul’s 1799 Ariodant which employed a quartet of cellos – a work that had failed at its premiere – and of which Rossini, together with most in the Salle le Peletier on that hot August night, was probably unaware.8 And the difficulty for audiences and critics in 1829 is one for us too. For a work so deeply embedded in popular culture of such varying forms, it is a strange paradox to find such a gap in our understanding of this key work. The questions are clear: where exactly does the overture to Guillaume Tell come from? What are its stylistic antecedents and – perhaps more interestingly given that they will appear so obvious – why have they been so comprehensively ignored, apparently, for the best part of 200 years? When speaking of the work as a whole in 1829, critics were at pains to point out how Guillaume Tell was unlike any of Rossini’s Italian operas, and invoked comparisons with Mozart, Haydn and Beethoven. As far as the orchestration was concerned, the Moniteur universel thought that ‘it must be compared … with the orchestration – so energetic and so grandiose – of Beethoven’.9 And it was to Beethoven that one critic reached out as he looked for predecessors for the overture to Guillaume Tell. The anonymous critic for Le censeur dramatique was unstinting in his praise and specific in his identification: The overture to Guillaume Tell is on its own, if I dare say it, a masterpiece of art and genius. This overture … was composed in the same style as the fine symphony in f  by Beethoven.… 10

6 Le moniteur universel, 5 August 1829. 7 La revue encyclopédique, 5 August 1829. 8 Le journal des débats, 11 August 1829. 9 ‘Il faudrait le comparer … aux orchestres si énergiques, si grandioses de Beethoven’ (Le moniteur universel, 5 August 1829). 10 ‘L’ouverture de Guillaume Tell est à elle seule, j’ose le dire, un chef-d’œuvre d’art et de génie. Cette ouverture … est composée à peu près dans le genre de la belle symphonie en fa de Beethoven .…’ (Le censeur dramatique, 9 August 1829). The text reads: ‘de la belle

Beethoven and Rossini  227 An intriguing suggestion: that the origins of the overture to Guillaume Tell might lie in Beethoven’s Sixth Symphony Op. 68, the Pastoral, and a simple question immediately has to be posed: how likely is it that Rossini, or the critic for Le censeur dramatique, or anyone else in Paris, knew ­Beethoven’s Pastoral Symphony in 1829? Rossini met Beethoven, as everyone knows, in Vienna in 1822 during preparations for the premiere of Zelmira.11 But there he heard only some quartets, the andante of the seventh symphony and the entire Eroica symphony, as well, of course, as meeting Beethoven which he related to Wagner in 1860. This seems an unlikely point of contact.12 The print transmission of the Pastoral Symphony is a no more likely point of contact between Rossini and Beethoven. The Pastoral Symphony was preserved in a bewildering variety of arrangements before Rossini got remotely close to it. It was published in versions for septet (two violins, two violas, flute, cello and double bass) and string sextet as well as in a version for keyboard with either flute or violin in 1810. The version for piano four hands was published in 1816, and the two-hand version – as well as the arrangement for flute, violin, cello and keyboard – appeared in 1829, the year of the Guilliaume Tell premiere. Critically, the full score, which Rossini would have needed for understanding so many of the instrumental effects, was not published until May 1826.13 Rossini almost certainly wrote the overture to Guillaume Tell during the month of July 1829 while he was undertaking the orchestration of acts iii and iv of the opera, and while it was already in rehearsal.14 And it was only three months before when Beethoven’s Sixth Symphony had received its Paris premiere at the Société des Concerts du Conservatoire.15 This had taken place at the Conservatoire itself, a short stone’s throw from where Rossini lived on the rue Montmartre. No lists of audiences for the two concerts that included the Pastoral Symphony can tell us whether Rossini was symphonie en la de Bethowen [sic]’, and although the Symphony No. 7 in A had been given on 1 and 29 March 1829, the comparison with the overture to Guillaume Tell makes little sense in comparison with that of Symphony No. 6. 11 Giuseppe Radiciotti, Gioacchino Rossini: vita documentata opere ed influenza su l’arte, 3 vols (Tivoli: Chicca, 1927–1929), 2: 96. 12 Edmond Michotte, Souvenirs personnels: la viste de R. Wagner à Rossini (Paris 1860): détails inédits et commentaires (Paris: Fischbacher, 1906), 25. 13 Georg Kinsky, Das Werk Beethovens: thematisch-bibliographisches Verzeichnis seiner sämtlicher vollendeten Kompositionen, complété et édité par Hans Halm (Munich and Duisberg: Henle, 1955). It is possibly the publication of the complete score that triggered Castil-Blaze’s commentaries on Beethoven’s symphonies (Journal des débats, 1 June 1827), but the text cited by Jacques-Gabriel Prod’homme (Les symphonies de Beethoven (1800– 1827) (Paris: Delagrave, 1906), 269, note 1) does not contain this extract. 14 Bartlet (ed.), Guillaume Tell, Opéra en quatre actes, 1: xxxvii. 15 For the programmes of the Société des Concerts du Conservatoire, see D. Kern Holoman, ‘The Société des Concerts du Conservatoire, 1828–1967’ (http://hector.ucdavis.edu/sdc/; consulted 28 June 2010).

228  Beethoven and Rossini present, but there is no doubt that he was in the city, proposing change after change to the contract that would result in his pension from the French government.16 There are other reasons to think that Rossini must have been at the two concerts on 15 March and 12 April 1829 (the concerts always took place on Sunday afternoons). Rumours circulated that Rossini had much to do with the entire project of bringing Beethoven’s symphonies to Paris, together with Habeneck, who was not only the conductor of the Opéra orchestra but also of the Société des Concerts du Conservatoire – and from 1829 its president.17 It seems unlikely that an individual as closely involved in the development of the Beethoven performances at the Société des Concerts du Conservatoire as Rossini, and living less than a quarter of a mile away, would have not been present at performances and quite possibly for rehearsals. A further compelling reason for a relationship between the genesis of the overture to Guillaume Tell and the Pastoral Symphony lies in the fact that many members of the two orchestras who premiered the two works in Paris overlapped. The orchestra of the Société des Concerts du Conservatoire was drawn from the best players at all the opera houses in the city – the Opéra itself, the Opéra-Comique, the Théâtre-Italien and some smaller houses. In other words, the sonic effect of the two performances – Beethoven’s Pastoral Symphony on 15 March and 12 April and of the overture to Guillaume Tell on 3 August – was very much the same. Questions of personnel are also key to the musical similarities between the two works: because they are important and revealing, they benefit from being considered together. In all the commentary that the Pastoral Symphony has elicited in recent years, one thing is common: the conjunction of pastoral and temporale – or storm topics – in a symphonic work was rare. As Richard Will has shown, there are abundant examples of symphonies from the period 1770 to 1820 that employ pastoral modes, and plenty that develop the temporale topic, 16 See above, note 2. 17 The implication of Rossini in the project to bring all the Beethoven symphonies to Paris may be determined from a number of sources. According to Jacques-Gabriel Prod’homme, Rossini studies Beethoven’s scores ‘under Habeneck’ (‘Rossini and His Works in France’, Musical Quarterly 17 (1931), 129). Prod’homme depended on recently published documents that had belonged to Julius Stockhausen (Julia Wirth (ed.), Julius Stockhausen: der Sänger des deutschen Liedes nach Dokumenten seiner Zeit (Frankfurt am Main: Englert und Schlosser, 1927), 9) in which Charlotte Kester wrote to Stockhausen’s father, Franz Anton, that she thought him responsible for having brought the works to Paris. Kester’s preoccupation was aroused by the claim according to which Rossini was implicated in the Parisian Beethoven concerts, in an article published in the [Augsburger] allgemeine Zeitung, 15 November 1861: ‘The program of yesterday’s concert also contained a beautiful Pièce by Rossini, so we were glad to remember how Rossini introduced our Beethoven to Paris with this symphony (‘Das Programm des gestrigen Concerts enthielt auch eine schöne Pièce von Rossini, und so liessen wir uns gern daran errinnern wie Rossini gerade mit dieser Sinfonie unsern Beethoven in Paris einführte’).

Beethoven and Rossini  229 18

but combining the two is almost unheard of. The best-known exception is Justin Heinrich Knecht’s Le portrait musical de la nature which has been acknowledged as a direct source of the Pastoral Symphony since 1861.19 This puts the relationship between the overture to Guillaume Tell – which also combines pastoral and temporale styles – and the Pastoral Symphony into a very close alignment indeed. There is no doubt that there are any number of sources for Rossini to have chosen for his pastoral and temporale topics – dozens alone in his own works, there is abundant material within Guillaume Tell itself. But compositions that mix the two – apart from the Pastoral Symphony, fresh in Rossini’s mind in July 1829 – were unknown. More explicitly, the Guillaume Tell overture and the Pastoral Symphony integrate their temporale and pastoral sequences in similar ways: by alternating fragments of the storm with the preceding movement at its beginning, and treating the ending in similar ways. Here’s the transition from Rossini’s storm into the ranz des vaches, alternating storm topics with a more peaceful descending passage in half notes. Approaches to the temporale sequence are harmonically similar: Beethoven simply interrupts a dominant onto a flat sixth chord and Rossini interrupts a tonic onto a 4/2, then treating the seventh as the flattened sixth of f# minor in the first of several steps towards the dominant of e minor. Rhetorically, it is anacoluthon: the disturbance of – in this case – musical grammar, and its effect one of simple interruption. Caution needs to be exercised in drawing too many specific comparisons between the exact temporale topics in the two works: their range is too great in general terms to argue for any meaningful comparison or contrast in specific terms.20 Too much importance should not be attached to the differences between bowed tremulandi in the Rossini and fingered tremulandi in the Beethoven, since both are conventional topics, found in a range of works. The tiny melodic flashes at the end of both storms, where Beethoven echoes material from the storm itself while Rossini anticipates his next section, are, however, interesting. There are a large number of points of strictly musical comparison, each of which could be read as a simple acknowledgement of, in some cases, an attempt by Rossini to outdo Beethoven (the handling of the piccolo and trombones is a good example).21 There is space here just for two: the sound world of the lower strings, and the shared use of periphrastic variation. 18 Richard Will, The Characteristic Symphony in the Age of Haydn and Beethoven, New Perspectives in Music History and Criticism (Cambridge: Cambridge University Press, 2002), 294–298. 19 François-Joseph Fétis, ‘Deux symphonies pastorales’, Revue et Gazette musicale de Paris [28 October 1866] 33 (1866), 337–338. See also Will, Characteristic Symphony, 175–177. 20 Ibidem, 157–158. 21 Beethoven’s use of the piccolo in bars 82–119 of the fourth movement of the sixth symphony (in which it doubles the flute for most of the time) may be compared with Rossini’s, among others, between bars 52–92 of the overture to Guillaume Tell. It is there employed conventionally and in octaves with the oboe, then (bar 152) at the unison. Similarly,

230  Beethoven and Rossini Table 6.1  L  ower strings: Orchestras of the Sociéte des Concerts du Conservatoire and of the l’Académie Royale de Musique, 1829 Nom

Fin 1829

Autre

Norblin, Louis

Opéra

Vaslin, Olive-Charlier

Opéra

Huber[t], Chrétien Chaft, Louis-François Franchomme, Auguste-Joseph Déjazet, Alphonse Desnos, Auguste-Prosper Rogé Mercadier, André-Adolphe Chevillard, Pierre-François Tilmant, Alexandre-Théophile-Joseph Thomas, Charles-Louis

Opéra (Cb) Théâtre-Italien Opéra-Comique Opéra Opéra-Comique Opéra-Comique Gymnase Dramatique Variétés

Prof. @ Conservatoire [Classe III] Prof. @ Conservatoire [Classe IV] 1st Prix1824 Accessit 1806 1st Prix1825 1st Prix1826 Accessit 1820

Opéra-Comique

1st Prix1822 1st Prix1827 1st Prix1829 2nd Prix1828

It is striking that Rossini’s use of a large ensemble of solo cellos should emerge only weeks after the Paris premiere of perhaps Beethoven’s unique employment of solo cellos.22 The whole of the slow movement of the Pastoral (up until the bird calls at the end) is characterised by the presence of two solo cellos, for most of the time playing different parts. The very opening of the movement gives some idea of the type of writing both in eighth and sixteenth notes in the Beethoven which evokes the same sound world as the opening of Rossini’s overture. The overlap between the two orchestras that premiered the two works has already been mentioned. Looking just at the lower strings shows just how much this is the case. This slide also gives a sense of just how impressive the orchestra of the Société des Concerts du Conservatoire actually was.23 Few orchestras could field anything close to twelve cellos, and the quality of the players is astonish-

Beethoven’s trombones are deployed in exactly the same passages as the piccolo, and for a similar reason, while in Rossini, the trombones are restricted to a melodic responsibility (92–136 and 399–422). 22 The use of two solo ‘cellos in the ‘Et incarnatus’ of the Missa solemnis falls more into the category of scoring for reduced string section than solo writing per se. 23 Table 6.1 is constructed from Jean-Michel Nectoux, ‘Trois orchestres parisiens en 1830: L’Académie Royale de Musique, le Théâtre-Italien et le Société des Concerts du ­Conservatoire’, Music in Paris in the Eighteen-Thirties, ed. Peter Bloom, La Vie musicale en France au xixe siècle 4 (Stuyvesant, NY.: Pendragon, 1987), 502–505; Pierre Constant, Le ­Conservatoire National de Musique et de Déclamation: Documents historiques et administratives (Paris: Imprimerie Nationale, 1900), passim, as well as the listings in Holoman, ‘The ­Société des Concerts du Conservatoire’.

Beethoven and Rossini  231 ing: Norblin, and his desk-partner, Olive-Charlier Vaslin, were both professors at the Conservatoire, and shared the same desk in the Opéra orchestra. The rest of the team were taken either from the rest of the Opéra orchestra, the orchestra at the Opéra-Comique or the principals at the orchestras at the Théâtre-Italien, Gymnase-Dramatique or Théâtre des Variétés. Almost all had won the premier prix during their career, and the younger players were the winners from 1825, 1826, 1827 and 1829. What this means, then, was that Norblin and Vaslin were both responsible for the solo cello lines in the Pastoral Symphony performances and for the top two parts of the Rossini ensemble, and Chaft and Déjazet took near-identical roles – as the bottom part in the ensemble – in both performances. Not only was the sound – instruments and players – identical, but Norblin and Vaslin were highly visible in both instances. It goes without saying that the orchestra at the Paris Opéra – in front of the stage with house-lights up – was clearly in sight, and the cello soloists particularly visible. Three months earlier, they had been even more obvious, as an engraving of the hall of the Société des Concerts du Conservatoire shows (Figure 6.1). It shows clearly how the space for the seating of the orchestra is raked at the back of the hall. In a modern auditorium, this would be for the chorus, but in the 1820s, the chorus is in front of the orchestra, as is visible in this illustration of the same space but with people in it: The raked platforms were for lower strings, wind and brass, and this means that Norblin and Vaslin were some of the most visually prominent

Figure 6.1  Salle du Conservatoire c1821.

232  Beethoven and Rossini

Figure 6.2  Salle du Conservatoire c1841.

members of the ensemble. A ground plan of the orchestra indicates Norblin and Vaslin (Figure 6.3). And finally, it is important not to exaggerate the difference in size between the Société des Concerts du Conservatoire and the Opéra; the Opéra hall is certainly bigger, seating 1,900 people, but the Conservatoire hall was not that much smaller – well over half its size at around 1,200 seats. And given that the Conservatoire hall was designed like a ­theatre, the acoustic phe­ etween the two – except nomenon cannot have been significantly different b for the resonance of the Salle du Conservatoire which was much praised. Anyone who was at both events would have heard the music played by the same players, on the same instruments and even using the same strings in spaces that – if not identical – were at least comparable, and that certainly afforded similar visual experiences. All the movements of the Pastoral Symphony except the scherzo and storm are characterised by the preponderance of what Elaine Sisman has called periphrastic variation – variation where ‘the original notes [are] replaced by a more ornate line, though with sufficient resemblance to the original, especially at cadences’;24 this could be called ‘decorative variation’ or even to leap back a century or more ‘division’. Outside of variation 24 Elaine Sisman. ’Variations’, Grove Music Online. Oxford Music Online. 28 June 2010 www. oxfordmusiconline.com/subscriber/article/grove/music/29050pg3.

Beethoven and Rossini  233

Figure 6.3  Orchestral layout of Société des Concerts du Conservatoire, 1828–40 (after Donna Di Grazia, ‘Rejected Traditions: Ensemble Placement in Nineteenth-Century Paris’, 19th -Century Music 22 (1998), 196).

sets and such movements as the slow movement of the seventh symphony, this sort of variation is rare in Beethoven, especially rare in sonata structures, so it’s particularly striking to find it (a) used with such emphasis in the Pastoral Symphony and (b) in such quantity also in the overture to Guillaume Tell. Here is the handling of a passage from very close to the opening of the first movement of the Pastoral and its variation in the recapitulation.

234  Beethoven and Rossini

Example 6.1  B  eethoven, Symphonie 6, 1: 5–16.

And this may be compared with the beginning of the third section of the Guillaume Tell overture and its variation. In both cases, the original material is simply repeated with a decorative counterpoint above; in the case of the Beethoven, by combining violins one and two into a single line and placing an ornamental counterpoint in triplets above it, and in the case of the Rossini by adding an elaborate flute counterpoint to the cor anglais melody from the beginning of the section. Of course, such periphrastic variation is common coin in Rossini’s vocal writing – where would the rondò with variations be without it? – but it is as rare in his instrumental lines (outside of dance music) as it is in Beethoven’s.

Beethoven and Rossini  235

Example 6.2  B  eethoven, Symphonie 6, 1: 289–300.

Example 6.3  R  ossini, Ouverture de Guillaume Tell, 176–184.

Be that as it may, that last example was of a passage that allowed Rossini to deploy probably the finest oboist (cor anglais) and flautist in Europe: Gustave Vogt and Jean-Louis Tulou. The case that Rossini modelled the overture to Guillaume Tell on the sixth symphony of Beethoven is unassailable. But it is possible that this

236  Beethoven and Rossini

Example 6.4  Rossini, Ouverture de Guillaume Tell, 209–217.

was the second time that he modelled an instrumental work on the Pastoral Symphony. Unlike the case of Guillaume Tell, the first occasion is fraught with chronological difficulty. Rossini was in Milan on 26 December 1813 for the prima of Aureliano in Palmira; only five months earlier, Giuseppe Moller had given concerts in the city that included not only Mozart’s symphonies thirty-five, thirty-nine and forty-one, but also Beethoven’s fourth, fifth and – critically – sixth symphonies.25 Rossini could not possibly have heard the performances in Milan since he was in Venice at least as late as 6 August and only arrived in Milan some time before 14 October 1813.26 It is, however, more than possible that Rossini wrote his La notte, temporale, preghiera, caccia for two flutes, clarinet and string quartet in the wake of 25 See the review, of which Pietro Lichtenthal is possibly the author: ‘Nachrichten. Mayland (Beschluss aus der 31sten No.)’, Allgemeine musikalische Zeitung 15/32 (11 August 1813). 26 See the letters from Rossini to his father, 6 August 1813, and to his mother, 4 October 1813, in Bruno Cagli and Sergio Ragni (eds.), Gioachino Rossini: Lettere e documenti IIIa: lettere al genitori 18 febbraio 1812 – 22 giugno 1830 (Pesaro: Fondazione Rossini, 2004), 41–44.

Beethoven and Rossini  237 the Beethoven performances in Milan, which must still have been important subjects for discussion well after the July performances were over.27 A date of c1813 has been suggested for the work on the basis of the similarity between the handwriting of the autograph and that of the Andante, et Tema con variazioni for flute, clarinet, horn and bassoon from the previous year. A link between La notte… and the Andante, et Tema con variazioni exists in the nature of the clarinet writing which might well link both works to Giuseppe Adami for whom Rossini had written the elaborate solo parts in La pietra del paragona, also in 1812.28 The evidence is nothing like as strong as in the case of the events of 1829, but the possible adjacency of a performance of Beethoven’s Pastoral Symphony and La notte… is an intriguing foretaste of the creation of the overture to Guillaume Tell. Although Beethoven provided titles for his movements in 1808, by the time they got to the Société des Concerts du Conservatoire in 1829, they had already accumulated some significant accretion. When the two versions are viewed side by side, Beethoven’s Lande in the first movement is glossed as agreste, which translates as uncultivated, the second movement is described more fully – but with good reason – but the finale has a fully developed programme that completely replaces Beethoven’s original (Table 6.2). The 1829 Paris premiere of the Pastoral Symphony elicited one of Fétis’ finest essays on what he called ‘la musique descriptive’. For him, the best parts of the work were the storm and the scherzo – where questions of imitation were simply answered. The most problematic were the first two and the final movements. He pointed to those scenes that he felt responded well to ‘imitation by music… a storm, a tempest, etc.’ For other styles, including the pastoral – what he called ‘the calm of ideas … may express again up to a certain point that of nature, but what a difference in the results!’.29 The ‘certain point’ to which Fétis pointed was well illustrated by the hermeneutic ambivalence exhibited in the reviews of the overture to Guillaume Tell in the wake of its premiere. In the light of his scepticism, it is not surprising that Fétis’ own review was so neutral: The beginning is a cantabile for five violoncellos, where the beauty of the melody and the piquancy of the harmony compete for the laurels. 27 The piece is edited in Martina Grempler and Daniela Macchione (eds.), Gioachino Rossini: Chamber Music without Piano: Musica da camera senza pianoforte, Works of Gioachino Rossini/Opere di Gioachino Rossini 1 (Kassel, etc.: Bärenreiter, 2007), 15–34. 28 Ibidem, xxvi–xxviii. See also Daniela Macchione, ‘La caccia dopo una notte tempestosa: quel vero che produce suono’, Occasional Essays from C[entro] I[talo-] A[mericano per l’] O[pera], http://humanities.uchicago.edu/orgs/ciao/introductory/essays%20from%20ciao/ la%20caccia%20dopo%20una%20notte%20tempestosa.htm. 29 ‘l’imitation par la musique… un orage, une tempête, etc.’ ‘le calme des idées … peut encore exprimer jusqu’à certain point celui de la nature; mais quelle différence dans les résultats!’ (‘François-Joseph Fétis, ‘Nouvelles de Paris : École Royale de Musique – Société des Concerts’, Revue musicale 2 (1828), 173–176).

238  Beethoven and Rossini Table 6.2  Programme of Beethoven’s sixth symphony: Paris 1829 and Vienna 1808 Paris, 1829

Vienna, 1808

    I S  ensation de plaisir à l’aspect d’une campagne agreste    II Scène au bords d’un ruisseau, langage des oiseaux III  Scène joyeuse et danse de campagnards

Erwachen heiterer Gefühle bei der Ankunft auf dem Lande Szene am Bach

IV Orage…     V Le calme renaît. Les pâtres rappellent leurs troupeaux. Chant pastoral en action de grâces à l’Eternel.

Lustiges Zusammensein der Landleute Gewitter, Sturm Hirtengesang: Frohe, dankbare Gefühle nach dem Sturm

After this cantabile come the delicious details of the cor anglais and the flute on a theme imprinted with a mountainous character. Then comes the allegro, a piece of an extraordinary warmth, whose effect provoked a general enthusiasm in the audience30 There was not a word attempting to describe the expressive power of any part of the overture. But La revue encyclopédique undertook exactly the same cultural work as those who amplified Beethoven’s programme, offering a hermeneutic reading of the overture: [The opening of the work exhibits] grace and nobility. The cor anglais solo, which then follows, is full of wave and abandon: it is a young shepherd, walking haphazardly and without a clear aim, but very sure of returning eventually to the place where his companions are found. However, when the flute comes to mingle its melodies with those of the cor anglais, the orchestra is animated and alights on a sort of pas redoublé, entirely in Rossini’s manner.31 And the anonymous account in Le globe mixes up metaphor and topos to argue that the first movement of the overture is characterised by the ranz des 30 ‘Le début est un cantabile pour cinq violoncelles, où la beauté du chant et le piquant de l’harmonie se disputent la palme. A ce cantabile succèdent des détails délicieux de cor anglais et de flûte sur un thème empreint d’un caractère montagnard. Puis vient l’allegro, morceau d’une chaleur extraordinaire, dont l’effet a produit un enthousiasme général dans l’assemblée’ (La revue musicale, 3 (1829), 103–104). 31 ‘la grâce et de la noblesse. Le solo de cor anglais, qui vient ensuite, est plein de vague et d’abandon: c’est une jeune bergère du pays, marchant au hasard et sans but déterminé, mais bien sûr de revenir à tems au lieu où se trouvent ses compagnes. Cependant, lorsque la flûte est venue mêler ses mélodies à celles du cor anglais, l’orchestre s’anime et se décide enfin sur une sorte de pas redoublé, tout-à-fait dans la manière de Rossini’ (La revue encyclopédique, 5 August 1829).

Beethoven and Rossini  239 vaches – when it is in fact found in the third movement – and the outbreak of the storm in the second movement marked bizarrely by the appearance of the sun.32 Even the veteran Castil-Blaze – locating all the conventional topics with his customary precision in the second and third movements – was inclined to invoke the image of ‘some rustic Swiss chapel’ for the first movement, and for the fourth movement, he argued that the trumpet sounds, it calls to combat the shepherds and the hunters. Liberty must be won: all run to arms; they hurry, they march, they throw themselves at the enemy. A pas redoublé with a proud rhythm and full of vivacity follows the striking clarion calls. Everything that music possesses of the most brilliant and the most vigorous is used in this ­finale of an enthralling warmth.33 So even for Castil-Blaze, the hermeneutic ambiguity in the overture to Guillaume Tell was an opportunity both for the demonstration of his legendary musical skills and his ability to interpret. Rossini’s engagement with Beethoven’s Pastoral Symphony raises the question of the degree to which all of Guillaume Tell has a Beethovenian dimension to it. When the German troupe put on a short season of operas at the Salle Favart between May and June 1830, Fidelio appeared on 30 May and the 6 June; this was almost certainly too late to have any serious impact on Rossini at such a late stage in the gestation of his own opera.34 32 ‘A ‘cello prelude is heard; it is a melody in the style of the Ranz de vaches; on might say a shepherd’s song at first light; soon the violins, then the flutes, are introduced; a new more lively melody, and worked with a charming coquettishness, follows the first; the day appears to advance; the music becomes clearer and more luminous; finally suddenly the sun appears, or at least we hear the timpani, trombones and the bass drum. Whether this is dawn on the Swiss mountains that M. Rossini has tried to paint, or whether it is something else entirely, it matters little (‘un prélude de violoncelle se fait entendre; c’est une mélodie dans le style des Ranz de vaches; on dirait le chant des pâtres dès la première aube du jour; bientôt les violons, puis les flûtes se mettent de la partie; une nouvelle mélodie plus vive, et travaillée avec une coquetterie charmante, succède à la première; le jour semble s’avancer; la musique devient de plus en plus claire et lumineuse; enfin tout-à-coup le soleil a paru, ou du moins nous entendons retentir les tymbales, les trombones et la grosse caisse. Que ce soit le lever du jour sur les montages de l’Helvétie que M. Rossini ait voulu peindre, que ce soit toute autre chose, peu importe’; Le globe, 5 August 1829). 33 ‘quelque chapelle rustique de la Suisse’ for the first movement and for the fourth movement, he argued that ‘la trompette sonne, elle appelle aux combats les pâtres et les chasseurs. Il faut conquérir la liberté : tout le peuple court aux armes; on se presse, on marche, on se précipite vers l’ennemi. Un pas redoublé d’un rythme fier et plein de vivacité succède aux appels éclatans du clairon. Tout ce que la musique a de plus brillant et de plus vigoureux est employé dans ce finale d’une chaleur entraînante’ (Le journal des débats, 11 August 1829). 34 For the exact chronology, see Mark Everist, ‘Translating Weber’s Euryanthe: German Romanticism at the Dawn of French Grand Opéra’, Revue de Musicologie 87 (2001), 67–105.

240  Beethoven and Rossini But a more likely route for an influence of Fidelio on Beethoven was the aborted production of the work in French at the Théâtre Royal de l’Odéon in 1826. The production, which got as far as rehearsal, was probably aborted because the original libretto to Fidelio and its precursors was by Victor-­ Joseph-Étienne de Jouy who was both French and still alive in 1826 (he died in 1846). Issues relating to copyright may well have stalled the production indefinitely.35 But Rossini, at this point, was closely involved with the Théâtre Royal de l’Odéon.36 The pasticcio of his works, Ivanhoé, was performed the following year, and Rossini’s compositional involvement is well known; he may also have been involved in the production itself. It is inconceivable that Rossini did not know the detail of the projected Fidelio performance, and that he did not have access to the Farrenc edition of the full score published in the same year. This is clearly a subject for further inquiry that falls outside the scope of this article. In conclusion, this chapter makes five points. 1 The remarkable chronology of the Paris premiere of the Pastoral Symphony and the composition of the Guillaume Tell overture 2 The particular qualities shared by Beethoven’s Sixth Symphony and the overture to Rossini’s opera: the similar contiguity of pastoral and temporale topics 3 A similarly ambivalent hermeneutic impulse in the reception of the two works 4 The use of similar musical procedures (moves from one movement to another, periphrastic variation) 5 The sound world in the two premieres generated by the same players and the same instruments in performing spaces they and Rossini knew only too well. A number of interpretative strategies could be adopted at this point. Crudely, it could be claimed that Rossini modelled his overture on what he understood – or took away – from Beethoven’s symphony. Or it could be argued that the similarities are just coincidence (that would be difficult). More cautiously, it could be suggested that without Beethoven, the overture to Guillaume Tell might have been a very different sort of piece. But the links between Beethoven and Rossini are so much more obvious here than the spurious application of Dahlhaus’ Stildualismus would suggest. The English-language translation of Dahlhaus’ term as ‘The Twin Styles’ 35 When the Odéon opened as an opera house in April 1824, Fidelio was envisaged as part of its repertory. Towards the middle of 1825, it became clear that a production was imminent, but it was abandoned between December 1825 and January 1826. See Mark Everist, Music Drama at the Paris Odéon, 1824–1828 (Berkeley, Los Angeles and London: University of California Press, 2002), 271–277. 36 Ibidem, 177–188.

Beethoven and Rossini  241 with its crude binary opposition does no justice to Kiesewetter’s original idea (on which Dahlhaus based his concept) or on Dahlhaus’ exemplification of Stildualismus, where his Beethoven example is juxtaposed, not with an example of Italian opera but with a passage from the fourth act of ­Meyerbeer’s Les Huguenots.37 The difficulty of rationalising this example – from French grand opéra written almost a decade after Beethoven’s death – with Dahlhaus’ general claims, remains unexamined but widely parroted. The ­Rossini-Beethoven relationship is much better understood as a series of cultural transfers akin to those that saw Weber appear in Paris in 1824 or Wagner in 1861.

37 See Mark Everist, ‘In Search of the Waters of Oblivion’, Journal of the American Musicological Society 62 (2009), 703, and the sources cited there.

7 ‘Il n’y a qu’un Paris au monde, et j’y reviendrai planter mon drapeau!’ Rossini’s second grand opéra

Rossini’s retreat from the stage after 1829 remains one of the unresolved mysteries in the history of nineteenth-century music. For the Paris of the July Monarchy, it was explicable in any number of ways: laziness, fear of competition from Meyerbeer, ill health, an ungrateful disdain for the city that had given him a home for over a decade;1 there seemed to be as many explanations as there were authors prepared to offer them, and little pause in the attempts to understand this silence in the years after Rossini returned to Bologna in 1836. The central question for Parisian audiences and critics was why Rossini had not written a second grand opéra to follow Guillaume Tell.2 Five years had elapsed between Rossini’s arrival in Paris and the première of Tell, and by 1834 it was agreed that Rossini’s second grand opéra was overdue. By the time ­Rossini left two years later, his creative silence was a matter of serious discussion. In the early 1840s, two events brought the question of Rossini’s silence back centre stage. The first was the Paris première of the revised version of the ‘Stabat mater’, which was given multiple performances in January 1842, and which was given even more publicity by the lawsuit over the rights to the work which were not settled until August the same year (in favour of Troupenas and against Aulagnier and Schlesinger).3 The second was news 1 The published literature on the explanations for Rossini’s silence runs from H. ­Sutherland Edwards’s 1869 biography of the composer (The Life of Rossini (London: Hurst and ­Blackett, 1869), 319–327), up to the current edition of The New Grove Dictionary of Music and Musicians (Philip Gossett, ‘Rossini, Gioachino, 6: Retirement’, Grove Music Online, ed. Laura Macy (consulted 7 May 2008), www.grovemusic.com). 2 Rossini’s interest in composing a second work for the Opéra immediately following Guillaume Tell is well known. The account given by Ferdinand Hiller in 1855 makes explicit reference to a setting of Faust and has recently been confirmed in an undated letter from Rossini to the manager of the Opéra, Émile-Timothée Lubbert. See Ferdinand Hiller, Aus dem Tonleben unserer Zeit, Gelegentliches von Ferdinand Hiller, 2 vols (Leipzig: Mendelssohn, 1868), 2:67; translated in Herbert Weinstock, Rossini: A Biography (London, Melbourne and Cape Town: Oxford University Press, 1968), 171–172, and Janet Johnson, ‘Rossini in Bologna and Paris during the Early 1830s: New Letters’, Revue de musicologie 79 (1993), 77–79. 3 For details of the lawsuit over the rights to the ‘Stabat mater’, see Anik Devriès and François Lesure, Dictionnaire des éditeurs de musique français, 2 vols [vol. 1 in 2 parts], Archives de l’édition musicale française 4 (Geneva: Minkoff, 1979–88), 2:420. Schlesinger’s

Rossini’s second grand opéra  243 circulated in early 1843 that Rossini was returning to Paris.4 Those making assumptions about the immediate delivery of a new work for the Opéra were to be disappointed, but Parisian musical discourse for the next five years was nevertheless characterised by the questions of Rossini’s silence and of a successor to Guillaume Tell. Despite the wide range of new Rossinian endeavours planned for the Opéra in 1843 – translations of La donna del lago, Semiramide and L’italiana in Algeri – none came to fruition in the form in which they were planned.5 Although the question of Rossini’s second grand opéra continued in the background to various rumours of new works – an opera based on Jeanne d’Arc perhaps based on a new libretto by Scribe – the concrete results were a French translation of his Othello (1844) and the pasticcio Robert Bruce (1846), neither of which have formed the subject of inquiry in modern scholarship. The composer’s status in the city was further enhanced when, in 1844, a statue of Rossini was proposed for the foyer of the Opéra in the Salle Le Peletier and was inaugurated in 1846. Rossini was the only living composer to be so honoured, a fact that resonated in a number of texts – formal and informal – that had their origins in the question of Rossini’s second grand opéra. These various responses – performances of existing operas, both planned and aborted, new works, statuary and various types of literary text – together constitute a coherent discourse on Rossini’s creative silence and, more generally, on creative and corporeal death.

Rossini returns to Paris Rossini’s reasons for returning to Paris in 1843 were entirely medical and entirely logical. The composer had suffered from gonorrhoea and haemorrhoids for most of his adult life.6 From the mid-1830s, he also began to exhibit fear of a contraction of the urethra, which he attempted to counteract by introducing a catheter into the bladder. By 1838, it was clear that he was suffering from a urethral disorder that his doctors in Bologna were unable to

defeat largely coloured the coverage of matters relating to Rossini in the Schlesinger house journal, the Revue et Gazette musicale de Paris, during the following years, which explains the largely negative judgements on the composer that were so much out of step with the rest of the music press and of Parisian musical culture. 4 Revue et Gazette musicale de Paris, 15 January and 2 April 1843. 5 La donna del lago would resurface as an important component in the pasticcio Robert Bruce in 1847, for which see 258–264. 6 Franz Hermann Franken, ‘Gioacchino Rossini (1792–1868)’, Krankheit und Tod großer Komponisten (Baden-Baden, Cologne and New York: Witzstrock, 1979), 233–266; Bruno Riboli, ‘Profilo medico-psicologico di Gioacchino Rossini’, Rassegna musicale 24 (1954), 292–303; Peter Volk, ‘Der Krankheitsverlauf bei Rossini’, Rossini in Paris; ­Tagungsband, eds. Bernd-Rüdiger Kern and Reto Müller, Schriftenreihe der Deutschen ­Rossini ­Gesellschaft 4; Veröffentlichungen des Frankreich-Zentrums 8 (Leipzig: Leipziger ­Universitätsverlag, 2002), 35–49.

244  Rossini’s second grand opéra treat successfully. A chance visit by Carl Gustave Carus, doctor to the King of Saxony, in April 1841, resulted in a consultation by Rossini that probably only confused the issue. It could, however, have been Carus who suggested the possibility of a cure in Paris by the world-famous Jean Civiale.7 By 1843, Civiale’s reputation was based on the development of a non-­ invasive treatment for bladder calculi, acknowledged as one of the most painful conditions known to man or woman, and for which existing surgical cures had a 25% mortality rate.8 The effort expended on travel to Paris would have been more than outweighed by the possibility of successful treatment. Rossini stayed in Paris from May to September and was treated daily by his doctors until midday. From noon until 4.00 p.m., Rossini walked in the  neighbourhood of his apartment on the Place de la Madeleine (part of the treatment), but from six in the evening until midnight his home became one of the most lively in the city.9 Criticism of Rossini’s inability to produce his second grand opéra was fuelled by his physical absence from the opera house; the situation was rendered worse by the energy with which he apparently entertained musical Paris at home. Although Weinstock claims that ‘for about three months Rossini was kept in almost total isolation’, La France musicale stated in late September 1843, that his apartment had been visited by nearly 2,000 people, and had previously observed that friends of the composer met with him every evening.10 Artists from all the Parisian opera houses visited and performed, Pleyel provided a piano every time one was needed and on 16 August ­Rossini’s salon was the venue for a complete performance of Mendelssohn’s Octet op. 20. A widely disseminated story was of Gilbert Duprez, who sang ‘Asile héréditaire’ from Guillaume Tell in Rossini’s apartment; Rossini responded ‘with a visible sense of emotion …’ and ‘said to him in Italian,

7 Olympe Pélissier had been in contact with her friend, Hector Couvert, who suggested she had a report drafted on Rossini’s health that could be shown to Parisian specialists. The report survives and forms the basis of almost all accounts of Rossini’s health in the early 1840s. See Bruno Riboli, ‘Documentazione clinica: malattia di Gioacchino Rossini secondo una relazione medica del 1842’, Note e riviste di psichiatria: periodico trimestrale di clinica di biologia del sistema nervoso 48 (1955), 429–435 (R Pesaro: Federici, 1956), translated in Weinstock, Rossini, 379–381. 8 Civiale’s De la lithotritie ou broiement de la pierre dans la vessie (Paris: Béchet jeune, 1827) triggered an extensive international correspondence, published in idem, Lettre à M. le cher Vincent de Kern… (Paris: Béchet jeune, 1827–48); see also idem, ‘Lettre sur la lithotritie’, L’esculape, 12 May 1840 (R Paris: Baudouin, [1840]). The early 1840s saw further publications that may well have had an effect on Rossini’s and his advisors’ decision to come to Paris: Jean Civiale, Du traitement médical et préservatif de la pierre et de la gravelle, avec un mémoire sur les calculs de cystine (Paris: Crochard, 1840); idem, Mémoire sur l’anatomie pathologique des rétrécissements de l’urètre (Paris: Renouard, 1842); idem, Mémoire sur l'emploi des caustiques dans quelques maladies de l’urètre (Paris: Renouard, 1842). 9 Rossini’s timetable was well known and frequently published. See, for an example, La France musicale, 24 September 1843. 10 Ibidem; Weinstock, Rossini, 230.

Rossini’s second grand opéra  245 pressing his hand: “the grateful author!”. These two words will remain in the heart of the artist for a long time, since it is the greatest praise for which he could hope’.11

A second grand opéra: Scribe and Jeanne d’Arc Although there had been warnings that Rossini’s visit would not result in a new work as soon as he arrived, even shortly before his departure from the city disappointment was still being voiced at his failure to produce a second grand opéra. In May 1843, the Revue et Gazette musicale de Paris wrote that ‘the illustrious composer of Guillaume Tell and of so many masterpieces, Rossini, has just arrived in Paris. Unfortunately, he does not come to produce a new work, awaited for so long and so impatiently’.12 As Rossini’s sojourn in the city came to an end, the Parisian musical world had to try and make the best of the inevitable. In September of the same year, Le ménestrel regretted that ‘the great man has resisted all prayers, all requests, has foiled all the ruses to obtain a new score from him’.13 Picking up on a trope whose importance will be explored later, La France musicale declared that ‘Rossini is not dead for music’, and that after his illness, he would recover and ‘you will see the sublime singer of enslaved Switzerland take up his pen and drown the entire world with these joyous songs, with his noisy melodies’.14 Other valedictory accounts cited Rossini’s own words, such as these, related by an unattributed correspondent for Le ménestrel: Other indiscretions go further, and guarantee a new work from the maestro, who exclaimed in an unguarded moment: ‘there is only one Paris in the world, and I shall return there to raise my standard!’15 These comments echoed throughout the next three years. Claims that Rossini was to make another visit to Paris in 1844 were accompanied by wistful

11 ‘avec un sentiment visible d’émotion, …’ … ‘lui a dit en italien, en lui pressant la main: “l’auteur reconnaissant!” Ces deux mots retentiront longtemps dans le cœur de l’artiste, car c’est le plus bel éloge qu’il pût ambitionner’ (La France musicale, 17 September 1843). 12 ‘L’illustre auteur de Guillaume Tell et de tant de chefs-d’œuvre, Rossini, vient d’arriver à Paris. Malheureusement, il ne vient pas pour faire représenter une nouvelle partition, depuis si longtemps et si impatiemment attendue’ (Revue et Gazette musicale de Paris, 28 May 1843). 13 ‘Le grand homme a résisté à toutes les prières, à toutes les sollicitations, a déjoué toutes les ruses mises en œuvre pour obtenir de lui une nouvelle partition’ (Le ménestrel, 3 September 1843). 14 ‘Rossini n’est pas mort pour la musique’ … ‘vous verrez le chantre sublime de l’Helvétie esclave reprendre sa plume et inonder le monde entier de ces chansons joyeuses, des ses mélodies bruyantes’ (La France musicale, 24 September 1843). 15 ‘D’autres indiscrétions vont plus loin, et garantissent une œuvre nouvelle du maestro, qui se serait écrié dans un moment d’oubli: “Il n’y a qu’un Paris au monde, et j’y reviendrai planter mon drapeau!”’ (Le ménestrel, 1 October 1843).

246  Rossini’s second grand opéra thoughts about a new work that had little basis in fact. When the news of Rossini’s imminent arrival was wrongly announced, it was asked, ‘is he coming to supervise the rehearsals of Othello, to add some new numbers to this score, or finally to endow us with a new opera [emphasis added]?’.16 And as will be seen, the administration of the Opéra was anxious that any work with Rossini’s name attached – translation, arrangement or pasticcio – should be considered a new grand opéra. So, when Léon Pillet, the manager of the Opéra, was defending his decision to mount Robert Bruce at the end of 1846, he claimed that ‘today, after five years of struggle and disagreeable efforts against difficulties that I had not at all created, I have succeeded in obtaining an opera from Rossini’.17 Although most reports of Rossini’s second grand opéra lacked substance and detail, two recurring threads ran through the skein of Parisian journalism: Rossini was writing an opera to a libretto by Scribe, and he was planning one based on the history of Jeanne d’Arc.18 The critic of Le ménestrel put it clearly: ‘No-one speaks any more of Meyerbeer or of Donizetti; Rossini’s Jeanne d’Arc alone seeks to gain credibility with the public’, but went on to express scepticism: ‘but the latter is so distrustful of the Swan of Pesaro that it places no faith in this rumour’.19 Schiller’s Die Jungfrau von Orleans was an obvious source that could have been dreamed up by the Parisian press for a libretto based on Jeanne d’Arc given that the same author had served similarly for Guillaume Tell. By the 1840s, Schiller was again in vogue: Théodore Anne’s libretto for Marie Stuart was set by Niedermeyer in 1843, at exactly the time Rossini was in Paris. Die Jungfrau von Orleans had elicited more interest on Italian operatic stages than on French ones with works by Vaccai (1827), Pacini (1830) and Verdi (1845); its single reflection on the Parisian operatic stage had been Carafa’s Jeanne d’Arc

16 ‘Y vient-il pour surveiller les répétitions d’Othello, ajouter à cette partition quelques nouveaux morceaux, ou enfin nous doter d’un opéra inédit [emphasis added]?’ (Ibidem, 9 June 1844). 17 ‘Aujourd’hui, après cinq ans de luttes et d’efforts pénibles contre des difficultés que je ne n’avais point crées, je suis parvenu à obtenir un opéra de Rossini’ (Le ménestrel, 11 October 1846). Between the composition and premiere of Robert Bruce, much of the correspondence between those responsible for the work (Gustave Vaëz, Pillet, Louis Niedermeyer and Rossini) was published in the press, together with a summary by Pillet from which this quotation is taken. 18 For a valuable account of the background to pictorial and stage presentations of Jeanne d’Arc in the 1820s, see Sarah Hibberd, ‘Marianne: Mystic or Madwoman? Representations of Jeanne d’Arc on the Parisian Stage in the 1820s’, Prose Studies 23 (2000), 87–98; R Medievalism and the Quest for the ‘Real’ Middle Ages, ed. Clare A. Simmons (London and Portland, OR: Cass, 2001), 87–98 [page numbers refer to 2001 reprint but are identical to 2000 original]. For a broader account of the operatic history of Jeanne d’Arc, see Alberto Rizzuti, ‘Music for a Risorgimento Myth: Joan of Arc, 1789–1849’ (PhD diss., University of Chicago, 2001); idem, ‘Viganò’s Giovanna d’Arco and Manzoni’s March 1821 in the Storm of 1821 Italy’, Music & Letters 86 (2005), 186–201; idem, ‘Joan of Arc’s Operatic Debut: Vicenza, 1789 – Venice 1797’, Il saggiatore musicale 11 (2004), 51–75. 19 ‘On n’entend plus parler ni de Meyerbeer, ni de Donizetti; la Jeanne d’Arc de Rossini seule cherche à prendre consistance dans le public’… ‘mais celui-ci est si défiant à l’endroit du cygne de Pesaro, qu’il n’ajoute aucune foi à ce bruit’ (Le ménestrel, 11 May 1844).

Rossini’s second grand opéra  247 à Orléans, premiered at the Opéra-Comique in 1821. Apart from Désiré Beaulieu’s cantata of 1815 and Leborne’s Prix de Rome cantata of 1818, Paris had to wait until the Second Empire for Duprez’s grand opéra on the subject (1865) and until the Third Republic for Mermet’s (1876). Were Rossini to have so written, it would have been an important feature on the landscape of opera during the July Monarchy, and the first work on the subject at the Opéra.20 A number of factors may have played into the rumour of a grand opéra by Rossini based on Jeanne d’Arc. Commentators could well have been confused with the early stages of the planning of Verdi’s Giovanna d’Arco – another Italian composer writing an opera based on the story.21 A more convincing explanation for the idea that Rossini might be working on a libretto by Scribe embodying the story of Jeanne d’Arc is possible confusion with Scribe’s libretto for what eventually turned into Clapisson’s opera Jeanne la folle. Although premiered in 1848, Jeanne la folle was drafted by Scribe exactly during the period of Rossini’s 1843 sojourn in Paris. Confusing Joanna the Mad of Castile with Joan of Arc would have been easy in a context where opera titles were frequently reduced to a single word, so an amplification of Jeanne to Jeanne d’Arc rather than Jeanne la folle would have been a plausible response from most Parisian critics. Scribe’s notebook for 1843 contains a clear reference to Jeanne la folle with notes on its sources: ‘A fine operatic subject / Consult Mme Clémence Robert’s novel in La presse 21 June [sic] 1843 / Read La ligue d’Avila by Victor Du Hamel published by Delloye’.22 Du Hamel’s Ligue d’Avila was published in 1840, and Madame Robert’s Jeanne de Castille was serialised in La presse in May 1843.23 It 20 There was, however, a libretto based on Jeanne d’Arc submitted to the Paris Opéra in 1828 by Edouard-Hubert-Scipion d’Anglemont and de Feuillade that was rejected by the Comité de Lecture (Paris, Archives nationales de France [hereafter F-Pan], AJ13 120; cited in Hibberd, ‘Marianne: Mystic or Madwoman?’, 95 and note 29) and a scenario for a ­ballet-pantomime-historique on the same subject was rejected in 1825 (F-Pan, AJ13 115; cited in Benjamin Walton, Rossini in Restoration Paris: The Sound of Modern Life, ­Cambridge Studies in Opera (Paris: Cambridge University Press, 2007), 94. 21 The Parisian rumours of an opera on Jeanne d’Arc were certainly known in Milan as Verdi prepared Giovanna d’Arco for La Scala: ‘If it is true that Rossini spends his honoured peace composing a Giovanna d’Arco, as is often stated in the French press, we should be struck by seeing both the Swan of Pesaro and the author of Nabucco, Crociati [sic] etc., walking the same path’ (‘S’egli è vero che Rossini spenda i suoi ozi onorati nel comporre una Giovanna d’Arco, siccome i giornali francesi più volte affermarono, stupir dovremo veggendo ad un tempo avventurarsi per un’istessa via il Cigno Pesarese l’autore del Nabucco, de’ Crociati [sic] etc.’) (La Fama, 2 September 1844; cited in Rizzuti, ‘Music for a Risorgimento Myth’, 447, and note 32 [translation lightly modified]. 22 ‘Beau sujet d’opéra / Consulter la nouvelle de Mad. Clémence Robert insérée dans La presse 21 juin [sic] 1843 / Lire La Ligue d’Avila par Victor Du Hamel chez Delloye ­l ibraire’ (Paris, Bibliothèque nationale de France [hereafter F-Pn], nouvelles acquisitions ­franchises 22584/8, fols 47v–48r). 23 Victor Du Hamel, La ligue d’Avila, ou l’Espagne en 1520 (Paris: Delloye, 1840). Clémence Robert’s Jeanne de Castille appeared between 13 and 29 May 1843 in La presse and was published as Jeanne la folle (Paris: Boisgard, 1852; R 1875).

248  Rossini’s second grand opéra seems likely that Scribe offered Jeanne la folle to Rossini, who kept it until 1844, at which point it was passed via Donizetti to Clapisson.24 It triggered the most specific claims that Rossini was engaged on a second grand opéra but, like many such rumours, proved impossible to confirm or deny.25 The myth of Rossini’s Jeanne d’Arc adhered to Parisian operatic discourse for well over a decade, however. When Philippe Royer translated Oettinger’s biography of Rossini, he added an epilogue dated March 1858: Everyone knows that since the summer of 1855, Rossini returned to Paris, where he has definitively established himself. May the illustrious maestro, to take revenge on the injustices of which he complains, exhume from his files his – Jeanne d’Arc – and allow us to applaud it during his lifetime.26 Although clearly ironic, the comment seems to assume the existence of at least a draft of the work as late as 1858. And Oettinger, in the main body of his text, relates a conversation with Rossini that he claims to have taken place before 1849, and not only suggests the composer had indeed set Scribe’s libretto, but takes this idea of exhumation further by alluding to Rossini’s conditions for the work’s performance: [Oettinger]: There have been rumours for years that you composed a new opera – Jeanne d’Arc – that may only be performed after your death. Is that true? / [Rossini]: Yes, my friend, the entire score has been in my desk for three years / [Oettinger]: And why do you wish to ­deprive your admirers of the pleasure of hearing your work for so long? / ­[Rossini]: Because I hope that the world will be more just towards the dead than towards the living. My time is past, I know. Today other idols are ­worshipped, against which my mediocre talent cannot struggle.27 24 For Donizetti’s involvement in Jeanne la folle, see William Ashbrook, Donizetti and his Operas (Cambridge: Cambridge University Press, 1982), 580. 25 It is unlikely that either the cantata Giovanna d’Arco (Paris, 1832) or the recitatives for Lucio Campiani’s Giovanna d’Arco (Bologna, 1845) have any relevance for the mythology of the opera based on the subject circulating in Paris in the 1840s, but see Rizzuti, ‘Music for a Risorgimento Myth’, 447, note 32 and ibidem, 408–409. 26 ‘Tout le monde sait que, depuis l’été de 1855, Rossini est revenu à Paris, où il est fixé définitivement. Puisse l’illustre maestro, pour se venger des injustices dont il se plaint, exhumer bientôt de ses cartons sa – Jeanne d’Arc – et nous permettre de l’applaudir de son vivant’ (Edouard Maria Oettinger, Rossini: l’homme et l’artiste, traduit de l’allemand avec l’autorisation de l’auteur par Philippe Royer, 3 vols (Paris: Bohné; Brussels: Schnée, 1858) 3:167). 27 ‘[Oettinger]: On y colporte depuis des années le bruit que vous avez composé un nouvel opéra – Jeanne d’Arc – qui ne doit être représenté qu’après votre mort. Est-ce que cela est vrai ? / ­[Rossini]: Oui, mon ami, la partition entière est depuis trois ans dans mon bureau. / ­[Oettinger]: Et pourquoi voulez-vous priver si longtemps vos admirateurs du plaisir d ­ ’entendre votre ­ouvrage? / [Rossini]: Parce que j’espère que le monde sera plus juste envers un mort qu’envers un vivant. Mon temps est passé, je le sais. Aujourd’hui on adore d’autres idoles, avec lesquelles mon médiocre talent est incapable de lutter’ (ibidem, 3:155).

Rossini’s second grand opéra  249 Oettinger’s text is as problematic as it is unreliable and cannot be taken seriously as a matter of historical record. However, it seems to suggest composition of the music for the complete work by 1846 at the very latest, which – however fictional – fits well with the non-fictional compositional chronology of Scribe’s libretto for Jeanne la folle. More important, the idea that the work should be performed only posthumously, and that even Rossini would need to ‘exhume’ it, plays directly in the discourse of creative death, as does Rossini’s hope that the world would be more just to the dead than to the living.

‘… ses conseils à ce sujet’ Rossini’s presence in Paris, even if it did not result in the much-anticipated new opera, sparked an interest in performances of French translations and arrangements of his Italian stage works at the Opéra. Despite continuing performances at the Théâtre-Italien, it was Rossini’s position at the Opéra and the legacy of Guillaume Tell that were paramount. Even as late as the 1840s, arranging a foreign opera for the Opéra – as the production of Le freyschütz in 1841 had made clear – was still a viable repertorial choice and an acceptable procedure in the eyes of the state, audiences and critics. Furthermore, at the end of Rossini’s period in Paris, he publicly declared his support for French adaptations of his Italian operas and offered advice on how they might be made; such declarations undercut the views of those journalists who were uncritically hostile to the idea of arrangement and suggest that their criticism was motivated more by a dislike of the managerial regime at the Opéra than by an aesthetic objection to the modification of what they may have considered inviolable works of art. Rossini’s support for the campaign of translation and arrangement of his Italian operas is found in a letter to Pillet, published in Le ménestrel: ‘The same letter teaches us that one has not yet given up producing a translated work by Rossini at the Opéra, and that he himself has given advice on the subject’.28 Such a practice would also produce not only Donizetti’s Lucie de Lammermoor and Verdi’s Jérusalem at the Opéra (1846 and 1847, respectively) but also Rossini’s Othello and Robert Bruce, a pasticcio based on the same composer’s works.29 In an operatic world in which the director of the Opéra was waiting not only for a second grand opéra by Rossini but also for two works by Meyerbeer – Le prophète and L’africaine – it was no surprise that Pillet would have recourse to such repertory, especially if it could respond to the publicly expressed need of the institution to honour the presence of the composer in the city.30 28 ‘La même lettre nous apprend qu’on n’a pas encore renoncé à monter à l’Opéra un ouvrage traduit de Rossini, et que lui-même a fourni ses conseils à ce sujet’. Le ménestrel, 1 October 1843. 29 Lucie de Lammermoor had originally been premiered in French translation at the Théâtre de la Renaissance in 1839. 30 Despite the much later dates of the premieres of Le prophète and L’africaine (1849 and 1865 respectively), both were complete in draft by 1843.

250  Rossini’s second grand opéra Plans to put on a gala performance of the composer’s works to celebrate Rossini’s arrival in Paris were promptly abandoned, probably because it was swiftly recognised that his state of health would not permit his presence at the Opéra.31 Alternatives were immediately adopted. Adolphe Adam was invited to arrange La donna del lago, but this was soon rejected because of the Opéra’s inability to find suitable tenors for the roles of Giacomo and Rodrigo.32 No sooner was this plan abandoned than a production of Semiramide – again arranged by Adam – was proposed. Little evidence survives of either of these proposals, and no librettists are named, although Rossini’s supervision had been claimed for at least the first of the two.33 The idea of a composite work consisting of parts of Semiramide and Tancredi, again arranged by Adam and again under Rossini’s supervision, was also floated.34 A more substantial ­u ndertaking – the adaptation and translation of L’italiana in Algeri, rumoured in the third week of June 1843 – was not finally abandoned until the end of the following month.35 What is striking about this embryonic production is that the team assembled consisted of Auguste Royer and Gustave Vaëz for the translation of the libretto and François Benoist for the musical arrangement, exactly the same personnel who would be responsible for the first successful arrangement of the campaign, the 1844 production of Othello. Two reasons were given publicly for abandoning L’italienne à Alger. The first was that rehearsals were stopped because of Donizetti’s arrival in Paris and the immediate preparations for his Dom Sébastien;36 the second was that Paul Barhoillet would not be available to take part in the production until after Rossini had returned to Bologna.37 A third reason may have been the most telling, however. The manager of the Théâtre-Italien, Eugène Watel, had complained to the Commission des théâtres that the production of L’italienne à Alger trespassed on his repertory. The commission’s report focused on law – which it found in favour of the Opéra (it was indeed permitted to play translations), but in terms of convenance could not reach a conclusion. The final line of the minutes of its deliberations,

31 Revue et Gazette musicale de Paris, 7 June 1843. 32 Le ménestrel, 11 June 1843. 33 Ibidem. 34 La France musicale, 11 June 1843; Le ménestrel, 18 June 1843. Much later, in 1860, a production of Semiramide would be mounted in Paris with recitatives added, and ballet music written, by Michele Carafa. See ACADEMIE IMPERIALE DE MUSIQUE / SÉMIRAMIS / OPÉRA EN QUATRE ACTES / DE / G. ROSSINI / AVEC / ILLUSTRATIONS, POINTS D’ORGUE, TEXTE ITALIEN / et / TRADUCTION FRANÇAISE / DE / MÉRY / … / PARIS / AU MÉNESTREL, 2 BIS, RUE VIVIENNE. 35 La France musicale, 18 June 1843; Le ménestrel, 25 June 1843, 2 July 1843 and passim. 36 Le ménestrel, 30 July 1843. 37 Ibidem, 16 July 1843.

Rossini’s second grand opéra  251 As for the suitability, the Commission believes that it should abstain from examining it at the moment, the project of performing L’italienne à Alger being postponed perhaps indefinitely, and the resolution of the question depending on circumstances,38 was a guarantee of placing the work in limbo for long enough to ensure that any impact it may have had would have been lost. The ill-starred production nevertheless had its echoes in the extracts from L’italiana in Algeri that found their way into the 1844 production of Othello.

Rossini at the Opéra The two successful attempts at projecting Rossini’s Italian operas onto the stage of the Opéra were Othello and Robert Bruce. While clearly attempting to satisfy – however inadequately – the desires of Parisian audiences to hear a second grand opéra by Rossini, the two works were also responding to a larger agenda that centred on British subjects at the Opéra. Among seventeen grands opéras and ballets, no fewer than seven works based on British subjects were mounted there between 1842 and 1846 – including Dietsch’s Le vaisseau fantôme (1842), Adam’s Richard en Palestine, Niedermeyer’s Marie Stuart (both 1844), Lucie de Lammermoor (1846) and the ballet Lady ­Henriette (1844) – and of the eleven operas, six were derived from British subjects.39 There is no other period in the history of the Paris Opéra when such a concentration was so intensely developed.40 Othello – as perhaps Shakespeare’s best-known work – and Robert Bruce – which not only played off a Scottish historical subject but also a background in Scott’s The Lady of the Lake via La donna del lago – both adhered closely to these cultural priorities.41 Although the administrators of the Opéra tried hard to convince their audience that Othello and Robert Bruce were new works that brought Rossini

38 ‘Quant à la convenance, la Commission croit devoir s’abstenir de l’examiner en ce moment, le projet de représenter L’italienne à Alger étant ajourné peut-être indéfiniment, et la solution de la question dépendant des circonstances (Commission des Théâtres. Registre des Procès-Verbaux des délibérations, 26 July 1843 (F-Pan F21 4633 [7])). 39 See Théodore de Lajarte, Bibliothèque musicale du Théâtre de l’Opéra: catalogue historique, chronologique, anecdotique, 2 vols (Paris: Bibliothèque Nationale, 1878; R Hildesheim: Olms, 1969), 2:170–184. 40 Viewed more broadly, of course, the interest in Scott goes back to 1819 and Rossini’s La donna del lago, although English operas based on Scott predated Rossini by nearly a decade. See Jerome Mitchell, The Walter Scott Operas (Tuscaloosa: University of Alabama Press, 1977) and idem, More Scott Operas: Further Analyses of Operas Based on the Works of Sir Walter Scott (New York: University Press of America, 1996). This trend demands further investigation beyond the large lyric theatres. 41 The Opéra-Comique was also susceptible to the same trend. See Monpou and Adam’s Lambert Simnel (1843); de Leuven, Brunswick and Clapisson’s Gibby le cornemeuse (1846); and Planard’s and Boulanger’s La cachette (1847).

252  Rossini’s second grand opéra back to the home of Guillaume Tell, they failed. In the case of Othello, the press simply ranged it alongside other French translations of Rossini’s operas from the 1820s: Moïse, Le siège de Corinthe and – perhaps a little unfairly since a good half of it had been new – Le comte Ory,42 or pointed to other foreign works performed in translation there: Don Juan and Le freyschütz.43 This commentary resuscitated a debate, which was never far from the surface of Parisian musical discourse, over the suitability of the repertory of the Théâtre-Italien for the Opéra. Even though the poster for Robert Bruce claimed it as a new work by Rossini, this strategy backfired badly once it was realised that the work was a pasticcio,44 and commentaries subsequently focussed on the place of the genre in the culture of the 1840s, and its origins.45 Almost no one was convinced that either Othello or Robert Bruce embodied Rossini’s second grand opéra, although enough commentators took the view that the two works were better than having no Rossini at all. Indeed, the anonymous correspondent of the Revue et gazette des théâtres spoke for many when he wrote, in the wake of the Othello premiere, ‘Ah! If Othello could be worth a pendant to Guillaume Tell to us…!’46

Othello The surviving sources for the 1844 Othello not only permit the reconstruction of this version of the opera but also provide indications of the librettists’ and musical arranger’s working practices and give some idea of the music and poetry that was subsequently rejected. As was customary, a full 42 La gazette de France, 6 September 1844; L’illustration, 14 September 1844; La France, 9 September 1844 (Théodore Anne); Le constitutionnel, 7 September 1844. 43 Le corsaire, 7 September 1844. 4 4 See, among others, Georges Bousquet’s commentary: ‘But this word ‘pasticcio’ must sound disagreeable to the ears of certain French people, since the latter manifest such a great repugnance at hearing or seeing it written. One cannot attribute to any other obstinate, illusory cause, the Opéra’s poster, carrying in black and white: ROBERT BRUCE, new opera; although there is not a single new melody in the piece, expressly written for the libretto, and although not only is one limited to borrowing from the Italian repertory of Rossini for the edifice of amalgamated ideas, but borrowings have even been made from the music that he composed for the French stage’ (‘Mais il faut que ce mot de pastiche sonne bien désagréablement aux oreilles de certaines personnes en France, pour que celles-ci éprouvent une si grande répugnance à l’entendre prononcer et à l’écrire. On ne peut attribuer à aucune autre cause obstinée, illusoire, de l’affiche de l’Opéra, portant en toutes lettres: ROBERT BRUCE, opéra nouveau; tandis qu’il n’y a pas dans toute la pièce une seule mélodie nouvelle, expressément écrite pour le libretto, et que non seulement on ne s’est pas borné à puiser dans le répertoire italien de Rossini pour construire cet édifice d’idées amalgamées, mais qu’on a fait des emprunts même à la musique qu’il a composée pour la scène française’) (Le commerce, 5 January 1847). 45 Opposing views are found in L’illustration, 19 January 1847, and in Théophile Gautier’s column in La presse, 4 January 1847. 46 ‘Ah! si Othello pouvait nous valoir un pendant à Guillaume Tell … !’ (Revue et gazette des théâtres, 5 September 1844).

Rossini’s second grand opéra  253 47

score was published, based directly on the complete archived copy and other material that was retained by the Opéra; as is the case for almost all Opéra productions of the period, all these sources survive.48 In addition are two precious documents: the first is Royer and Vaëz’s working copy of the French translation of the work, based on the Pacini and Launer Italian vocal scores of the Italian version;49 to this are added Benoist’s translations and adaptations of the recitatives and additional arias, also based on manuscript French translations of Italian printed originals,50 from L’italiana in Algeri, La donna del lago and Persiani’s Inès de Castro. The second additional document is Benoist’s autograph of the inserted aria from L’italiana in Algeri, one of the ballet movements, a rejected draft of the recitative after Desdémone’s and Emilia’s act i duet (‘Je lis dans ta pensée / Tu veux tromper mon cœur’, no. 5 [‘Inutile è quel pianto’ – ‘Vorrei, che il tuo pensiero’, no. 4 in the Italian original]) and Iago’s act ii aria.51 The anatomy of the 1844 Othello is given in Table 7.1. From these sources, it is possible to reconstruct an arrangement that follows Rossini’s Italian original both in its overall outline (three acts with similar locations of act divisions) and in the inclusion and ordering of numbers. Numbers from other works, however, also make a significant impact on the 1844 version of the opera. The interpolations of ‘Pensa alla patria’ scena e rondò, no. 17 from L’italiana in Algeri into act i as ‘O jour prospère’ after the duo for Emilia and Desdémone, together with its preceding recitative, newly composed (‘A Venise, ce jour prospère’; Expl. ‘O mon âme, vois donc plus forte’) and the inclusion of two dance numbers into the act i 47 Seule Edition conforme à l’exécution de l’Opéra / - / OTHELLO / Grand Opéra en trois Actes / Paroles de MM / Alphonse Royer et Gustave Waez / Musique / DE / G. R ­ OSSINI / PRIX 10f. net / PARIS / AU MÉNESTREL, 2 bis rue Vivienne, HEUGEL et FILS / ­Editeurs-Propriétaires pour tous Pays. 48 The archived full score is Paris, Bibliothèque nationale de France, Bibliothèque-Musée de l’Opéra (hereafter F-Po) A. 544. a. i–iii, entitled ‘Arrangement avec additions d’airs de La donna de lago, L’italienne à Alger, Armide, Mathilde de Sabran, Ermione, par Benoist’. 49 F-Pn MS 1335. OTELLO, Opera Seria / Posta in Musica / dal Maestro / GIOACHINO ROSSINI / Nuovo Editione / Prezzo: 36f / IN PARIGI / Nel Negozio di Musica / di PACINI, Boulevard des Italiens, No. 11. / 900; OTELLO / Musique de / ROSSINI / Opéra Complet / Partition de Piano et Chant, / PAROLES ITALIENNES / EDITION DE LUXE / Publiée par / MME VE LAUNER, / ÉDITEUR, Mde DE MUSIQUE ET DE PIANOS, / 14, boulevart Montmartre. 50 L’ITALIANA IN ALGIERI / Opera Buffa / del Maestro / G. ROSSINI. / Nuova edizione. / Prix 36 fr. / à PARIS, au Magasin de Musique de PACINI, Boulevard des Italiens, No 11, / ou l’on trouve tous les Ouvrages de Beethoven et Rossini: Scena e Rondò, no. 17 ‘Amici, in ogni evento’ – ‘Pensa alla patria’; (No.2) / QUANDO IL CORE IN TE RAPITO / SCENA E CAVATINA / Cantata dall Signora MALIBRAN. / Nell’ INES DE CASTRO. / Del maestro GIUSEPPE PERSIANI / PRIX Ff. 50 c. / A PARIS, Chez BENARD-LATTE, Éditeur de Musique, Boulevard des Italiens, et passage de l’opera No 2. / (Propriété des Editeurs). 51 ‘Otello. Opéra. Fragments. MS frç., 1ère moitié XIXe s. 5 fasc. / Copié de la main de Benoist’ (F-Po Rés. 23).

Table 7.1  S  ources for the Paris 1844 Othello 1844 Heugel print

F-Pn MS 1335

Ouverture 1–11

[Sinfonia] 1– 66

ACT I 1. Introduction ‘Vive Othello’ 12–21 Pacini 901, 15–19 [end of ritornello missing] 2. Marche 22–23 Pacini 902, 20–21 [Recitative] 24–26 MS: Inc. ‘[adop]-té / Toi qui vengeas nos armes’; expl. ‘Couronne de laurier d’or décérnée au vainqueur’ 3. Cavatine ‘A vous mon bras’ Pacini 904, 25–32 [end of ritornello missing] 26–36 [Recitative] 37–40 MS: Inc. ‘Écoute moi’. Expl. ‘tombera sous nos coups’ 4. Duo 41–47 ‘Va, soir sans crainte’ Pacini 905, 37–44 [Recitative] 48–51 MS: Inc. ‘O ma chère maîtresse’. Expl. ‘A vos genoux’ 5. Duo ‘Je lis dans ta pensée / Tu Pacini 906, 49–53 [transposed from G to F] veux tromper mon cœur’ 52–55 Persiani’s ‘Quando il core in te rapito’ from Inès de Castro. B.L. 725, 1–10. MS: ‘Scène et cavatine avec chœurs / Chantée dans l’Othello de Rossini par Madame De Lagrange; Paroles françaises de Mr. C. de Charlemagne’. ‘L’heure s’avance, il ne vient pas’ – ‘Quant mon âme en toi ravive’ [Recitative] 56 6. Air ‘O jour prospère’ 57–66 7. Chœur ‘Chaste hymen’ 67–75 No. 1 Pas de deux 76–81 No. 2 Pas de deux 82–87 [Recitative] 88–95

Other Sources (Critical Edition unless noted)

Pacini 908, 55–69 [entitled ‘Final du 1er acte’] MS 1336 MS 1336

1. [Introduzione] 67–124 [Recitativo dopo l’introduzione] 125–131 2. Cavatina di Otello 132–187 [Recitativo] dopo la cavatina d’Otello 188–195 3. [Duetto Rodrigo – Iago] 196–245 4. [Scena e duettino Desdemona – Emilia] 246–262 [with two important excisions] and 264 4. [Scena e duettino Desdemona – Emilia] 265–286

= F-Po Rés. 23/IV? = F-Po Rés. 23/II fragments of section with women’s chorus. Air de Desdemona avec chœur en mi b’ 5. Coro, e Finale dell’Atto Primo 302–347 F-Po Rés. 23/III 5. Coro, e Finale dell’Atto Primo 348–379

9 [sic; recte 8]. Trio ‘Que Dieu me soutienne’ 95–100 9. Final. 101–124

Vve L.5571, 61–65

5. Coro, e Finale dell’Atto Primo 380–398

Pacini 910, 75–96 [end of ritornello is missing]

5. Coro, e Finale dell’Atto Primo 399–469

ACT II 10. [Récitatif et] Air d’Iago ‘Dans le cœur d’Otello’ 125–126 10. [Récitatif et] Air d’Iago ‘Ah! Manque Air de Iago / Tiré de La donna del lago Mon cœur serait encore noble’ 126–131 [Recitative 132–135] 11. Duo ‘Plus de doute’ 136–146 [Recitative] 147–148 12. Trio 148–168 [Recitiative] 168–169 13. Final 170–183 ACT III [Recitative] 184–186 14. Barcarolle ‘Il n’est pas sur la terre’ 187–188 [Recitative] 188–190 15. Romance ‘Au pied d’un saule’190–195 16. Prière ‘Grand Dieu ma voix t’implore’ 196–197 17. Scène et duo 197–203 17. Scène et duo ‘Viens, frappe’ 203–210

Vve L.5571, 100–101; MS: Inc. ‘Morne et sombre’. Expl. ‘Et de sa main’. Pacini 915, 113–123 MS: Inc. ‘Oh l’accabler et la confondre’. Expl. ‘Ô transport’. Pacini 916, 125–144 MS: Inc. ‘Desdémone, grand Dieu’. Expl. ‘Est-il tombé sur moi’ Pacini 917, 145–158

[Recitativo] 471–475 = F-Po Rés. 23/I (fragments of cabaletta only) Last two movements of Rodrigo’s sortita in La donna del lago ‘Eccomi a voi, miei prodi’ (pages 391–420 of edition. Replaces Otello no. 6 ‘Che ascoloto’ (also for Rodrigo) 7. [Scena e duetto Otello – Iago] 503–518 7. [Scena e duetto Otello – Iago] 519–571 [Recitativo] dopo il duetto [Otello – Iago] 573–575 8. Terzetto [Desdemona- Rodrigo – Otello] 576–673 [Recitativo] dopo il terzetto 674–677 9. Finale secondo 678–760

Vve L.5571, 151–153 Pacini 919, 163

10. Scena prima 761–774 10. Scena prima [Canzone del Gondoliero] 775–778

Pacini 919, 164–165 Pacini 920, 166–172

10. Scena prima 778–785 [Canzone del Salice] 786–812

Pacini 920, 172–173

[Preghiera] 813–816

Pacini 921, 174–179 Pacini 921, 180–190

Scena III 817–840 [Duetto] 840–896 and 925–926 (897–924 om.) (Continued)

Key: Heugel 1844 Pacini 901–921 VE L.5571

B.L.725

Critical Edition A. 544.i–iii Rés. 23/I–IV MS 1336

Seule Edition conforme à l’exécution de l’Opéra / - / OTHELLO / Grand Opéra en trois Actes / Paroles de MM / Alphonse Royer et Gustave Waez / Musique / DE / G. ROSSINI / PRIX 10f. net / PARIS / AU MÉNESTREL, 2 bis rue Vivienne, HEUGEL et FILS / Editeurs-Propriétaires pour tous Pays [1844; H.5718] OTELLO, Opera Seria / Posta in Musica / dal Maestro / GIOACHINO ROSSINI / Nuovo Editione / Prezzo: 36f / IN PARIGI / Nel Negozio di Musica / di PACINI, Boulevard des Italiens, No . 11. / 900 [F-Pn L.2067; date is uncertain] OTELLO / Musique de / ROSSINI / Opéra Complet / Partition de Piano et Chant, / PAROLES ITALIENNES / EDITION DE LUXE / Publiée par / MME VE LAUNER, / ÉDITEUR, Mde DE MUSIQUE ET DE PIANOS, / 14, boulevart Montmartre [Vve L. 3371; F-Pn Vm4. 412; 1843] L’ITALIANA IN ALGIERI / Opera Buffa / del Maestro / G. ROSSINI. / Nuova edizione. / Prix 36 fr. / à PARIS, au Magasin de Musique de PACINI, Boulevard des Italiens, No 11, / ou l’on trouve tous les Ouvages de Beethoven et Rossini [1231–1249; F-Pn Vma. 4212; c 1820]: Scena e Rondò, no. 17 ‘Amici, in ogni evento’ – ‘Pensa alla patria’. (No.2) / QUANDO IL CORE IN TE RAPITO / SCENA E CAVATINA / Cantata dall Signora MALIBRAN. / Nell’ INES DE CASTRO. / Del maestro GIUSEPPE PERSIANI / PRIX Ff. 50 c. / A PARIS, Chez BENARD-LATTE, Éditeur de Musique, Boulevard des Italiens, et passsage de l’opera No 2. / (Propriété des Editeurs) [B.L.725]. On version in F-Pn MS 1335 is the alternative title in blue ink MS before printed title: Collins, Michael, ed., Otello, ossia Il moro di Venezia: dramma per musica in tre atti di Francesco Berio di Salsa, musica di Gioachino Rossini, Edizione critica delle opere di Gioachino 1/19 (Pesaro: Fondazione Rossini, 1994) F-Po A. 544. a. i–iii. The archive score for the 1844 performance at ARM: ‘Arrangement avec additions d’airs de la Donna de lago, L’italienne à Alger, Armide, Mathilde de Sabran, Ermione, par Benoist’ F-Po Rés. 23/I–IV. ‘Otello. Opéra. Fragments. MS frç., 1ère moitié XIXe s. 5 fasc. / Copié de la main de Benoist. F-Pn MS 1336

Rossini’s second grand opéra  257 finale, almost all of which finds its way verbatim into the published version, lengthen the act considerably. By contrast, Iago’s air ‘Ah! Mon cœur serait encore noble’, based on the last two movements of Rodrigo’s sortita in La donna del lago, ‘Eccomi a voi, miei prodi’ (but retaining the original recitative in French translation), merely replaces Rodrigo’s act ii aria from Otello ‘Che ascolto’ (no. 6). Another interpolation, not found in the printed score of the work and presumably abandoned during rehearsals, is an aria from Persiani’s Inès de Castro, ‘Trascorsa è l’ora’ – ‘Quando il core in te rapito’, interleaved into F-Pn MS 1335 after the Emilia-Desdémone duet. A manuscript note on the print states that it was ‘chantée dans l’Othello de Rossini par Madame De Lagrange’ and furnished with ‘paroles françaises de Mr. C. de Charlemagne’ (the French translation is written into the score above the Italian text: ‘L’heure s’avance, il ne vient pas’ – ‘Quant mon âme en toi ravive’). Whether this was conceived as an alternative to ‘Pensa alla patria’ or as an addition to it, the textual disturbance of F-Pn MS 1335 at this point makes it impossible to offer a definitive reading.52 Rossini himself had nothing to do with the production, although in the light of the comments he was reported to have made in late 1843, and the fact that the editorial team of Benoist, Royer and Vaëz had been active during the same year and must have been known to him, it might be surmised that he had at least a vicarious involvement in the project.53 The broader context for Vaëz’s and Royer’s work is discussed elsewhere in this volume.54 For the arias and ensembles, they simply wrote their French words onto printed copies of the music in question; all the recitatives, however, were rewritten and in some cases composed by Benoist (this is all clear in F-Pn MS 1335). Contemporary voices clamoured in praise of his discretion: journalists claimed that they could easily recognise the original Italian recitatives in the French version and noted that Benoist’s original additions were restricted to a chord here and there.55 This is a slight overstatement. Benoist was responsible for all the new recitative around the act i substitution of the replacement aria from L’italiana in Algeri, but given that Rossini may have had no direct participation in this production, claiming fidelity to the composer’s score was as much as could be anticipated. Central to the commentaries on the 1844 Othello was the fact that all the additions were taken from works by Rossini himself and not, for example, from works by Donizetti or Bellini, 52 Anna-Carolina de Lagrange certainly performed ‘Quando il core in te rapito’ in a ­p erformance of Otello at La Fenice in 1848. It is found in the following libretto: Otello / ossia / L’Africano di Venezia / dramma tragico in tre atti / parole di Leone Totola / musica del maestro cav. Gioachino Rossini / da rappresentarsi / nel Gran Teatro la Fenice / nel ­c arnovale / 1847–48 / … / in Venezia / dalla tipografia Rizzi / 1848. 53 The Parisian press, however, knowing that Rossini was not present in the city, was convinced that the composer had little to do with the production. See, for example, L’illustration, 14 September 1844. 54 See Chapter 9. 55 La France, 9 September 1844 (Théodore Anne).

258  Rossini’s second grand opéra as might well have been the case.56 This might be a further reason why the Persiani aria in F-Pn MS 1335 never reached performance. Furthermore, Benoist was praised for having selected the music for the divertissements from works that were largely unknown in Paris, particularly Armida.57

Robert Bruce Many of the objections to the 1844 production of Othello were insurmountable. The absence of Rossini’s direct participation, however, was something that the Opéra could rectify, and it went to great lengths to impart to Othello’s successor, Robert Bruce, the cachet of Rossini’s own hand.58 To do this, Pillet, Vaëz and the musical arranger, Louis Niedermeyer, travelled from Paris to Bologna to work directly with Rossini during July 1846. Pillet apparently stayed only long enough to set up the working conditions with Rossini and his two colleagues. Vaëz and Niedermeyer remained in Bologna – albeit reluctantly – for three weeks.59

56 ‘Rossini’s work has not only been respected by M. Benoist, but all the additions have been taken from the works of the maestro, even the dance music’ (‘L’œuvre de Rossini a été non seulement respectée par M. Benoist, mais toutes les additions ont été empruntées aux œuvres du maëstro, tout, jusqu’aux airs de danse’) (ibidem). 57 ‘M. Benoist had finally to compose the divertissement for the first act. Instead of making use of his own resources, the composer of La gypsy and Le diable amoureux preferred to have us hear the dance music from Armida from which is only known in Paris the fine duet “Amor! Possente nume”’ (‘M. Benoist avait enfin à composer le divertissement du premier acte. Au lieu de puiser dans son propre fonds, l’auteur de la Gypsy et du Diable amoureux a préféré nous faire entendre les divertissements de l’Armida dont on ne connaissait à Paris que le beau duo: Amor! possente nume’) (Le constitutionnel, 7 September 1844). 58 In contrast to Othello, Robert Bruce has received a certain amount of admittedly superficial attention in modern Rossini scholarship. Built largely on accounts in Louis-Alfred Niedermeyer, Louis Niedermeyer, son œuvre et son école (Paris: Repos, 1867), 11 and idem, Vie d’un compositeur moderne, 1802–1861 [Louis Niedermeyer] (Fontainebleau: Bourges, 1892; R [with introduction by Camille Saint-Saëns] Paris: Fischbacher, 1893), 78–79 are accounts in Giuseppe Radiciotti, Gioacchino Rossini: vita documentata opere ed influenza su l’arte, 3 vols (Tivoli: Chicca, 1927–29), 2:292–303; Weinstock, Rossini, 237–240; and Richard Osborne, Rossini, The Master Musicians (London: Macmillan, 1986), 98; when it is stated in the second edition of the latter (New York: Oxford University Press, 2007), 136–137, that ‘Unfortunately, pasticcio, once so popular, was now more or less taboo’ can be shown from the analysis presented here, to be an oversimplification. 59 The reluctance had little to do with working with Rossini, which was satisfactory on all sides, but much more to do with the cosmopolitan Niedermeyer’s and Vaëz’s being stranded in provincial Bologna and a suspicious chief of police. In an undated letter to an undisclosed recipient in Paris, Niedermeyer explained how ‘Bologna holds no attraction, and I am in a hurry to be finished with my work. There is a complete sympathy between the Bologna police and myself. The Director of Police is kind enough to enquire very often of us if we are going to leave soon’ (‘Bologne n’a rien d’attrayant, et j’ai hâte d’avoir terminé mon travail. Il y a sympathie complète entre la police de Bologne et moi. Le directeur de la police a la bonté d’envoyer fort souvent chez nous pour savoir si nous ne partirons pas bientôt’) (cited in Niedermeyer, Vie d’un compositeur moderne, 79).

Rossini’s second grand opéra  259 The extant sources for Robert Bruce, in particular the correspondence between the participating artists (from Rossini to Pillet, Vaëz to Pillet, and Niedermeyer to an anonymous correspondent) and Niedermeyer’s working score of the opera, permit a vision not only of how Rossini, Niedermeyer and Vaëz worked together, but also of how much work had already been undertaken in Paris and brought to Italy, and are thus central to understanding the compositional history of the work.60 The musical source is a three-volume oblong-format holograph, copied anonymously and almost certainly in Bologna, of all the music Rossini selected for inclusion in the pasticcio.61 Within the document are Niedermeyer’s annotations, based on his work with Rossini. Whether Niedermeyer merely copied originals dictated by Rossini, or whether he wrote them himself is difficult to settle; the high level of erasure and correction certainly suggests, in those parts at least, that the modifications originate with Niedermeyer. Together with the traces habitually left by productions at the Opéra, this score allows a clear view of the work as it was assembled by Rossini and his collaborators and as it was premiered in Paris (Table 7.2).62 The absence of Royer in Bologna could be read in a number of ways: that despite his name on the title page of the libretto and score, his role was largely supervisory, or that he and Vaëz had settled most of the libretto before the latter left Paris in July 1846; he could equally have been involved in the very late stages of polishing the libretto after the latter had returned to Paris. Whatever Royer’s role, it seems clear that not only was the libretto largely complete by the time the Parisian artists arrived in Bologna but also that Pillet had some idea about which musical numbers might form the basis of the work; in other words, the Parisian team – between them – had a text and a series of proposals about which musical numbers to use that they were prepared to submit to Rossini for his approval. In assembling the beginning of the third act, for example, Vaëz found himself having to arbitrate between instructions given by Pillet and artistic judgements made by Rossini. It is not entirely clear what form Pillet’s plan took, but when Rossini contributed the chorus ‘Buvons, il faut saisir amis’, which Vaëz told Pillet was unpublished, 60 The correspondence between Vaëz and Pillet is F-Pn L.a. Vaëz (G.) 4–5, 276–283 and 284– 287, and between Rossini and Pillet is published in Rossini à Paris: exposition au Musée Carnavalet, 27 octobre – 31 décembre 1992, ed. Jean-Marie Bruson (Paris: Société des Amis du Musée Carnavalet, 1992), 127. 61 Niedermeyer’s working score is now F-Po A.554.a.I–III. 62 Other sources are the untitled archived full score (F-Po A.554.b.I–III) and the published full score: ROBERT BRUCE / Opéra en Trois Actes / Paroles / DE / MM. ALPH. ROYER ET GUSTAVE VAEZ / Musique / DE / G. ROSSINI / Représenté pour la première fois à Paris / sur le Théâtre de l’Académie Royale de Musique / Le 30 Décembre 1846 / Vendu comme Manuscrit / A. Lafont / Paris, chez E. Troupenas & Cie, Rue Nve Vivienne, 4 / Leipsic, chez Breitkopf & Haertel Londres, chez Mitchell Milan, chez Ricordi. Partial identifications of the sources for Robert Bruce are found in Radiciotti, Gioacchino Rossini, 2:297–298.

Table 7.2  S  ources for Robert Bruce (1846) Robert Bruce (1846)

Sources

Ouverture ACTE I 1. Introduction ‘Robert! Mille angoisses mortelles’ 1–20 2. Récit et Chœur ‘A ma fille je l’espère’ 3. Barcarolle ‘Marie à ce doux nom comme mon cœur palpite’ 32–36 4. Récit, duo et chœur ‘Est-ce moi qui fait naître’ 37–49 ‘Ciel, un anglais!’, 50–52 5. Air et chœur ‘Ce sont les clans de la montagne’ 53–57 6. Couplets et Chœur ‘Alerte, fille’ 58–62 Ier air de danse; pas de cinq 63–67 IIième air de danse: Ecossaise 68–71 7. Marche, scène et cavatine ‘Pourquoi cesser vos jeux’ 72 ‘La gloire est belle’ 73–75 8. Finale ‘Sire! Douglas le Noir’ 76–124 ACTE II 9. Entr’acte et scène ‘Le roi sommeille’125–126 10. Air ‘Que ton âme, si noble si bonne’ 127–129 11. Cavatine ‘Ô noble père’ 130–138 Récitatif ‘Les Anglais! Ils cernent le château’ 139–143

Zelmira, no. 1 (introduzione), pp.14–85 La donna del lago, no. 1 (introduzione), pp.79–125 La donna del lago, no. 1, pp.41–54 ‘O mattutini albori’ (Elena)’ La donna del lago, no. 2 (duo Elena Uberto), pp. 184–247 = F-Po A.554.a.I, pp.183–4 (Instrumental ending taken from La donna del lago, recitativo dopo l’introduzione, pp.126–127) La donna del lago, no. 2 (Duo, Elena-Uberto), pp.145–166 Bianca e Faliero, no. 9 (cavatina Bianca), pp.91–99 (cabaletta only) Moïse, air de danse no.2, pp.192–198 Possibly new acording to Vaëz La donna del lago, no. 6 (Coro e Cavatina Rodrigo), pp. 353–356 = F-Po A.554.a.I, pp.225–7 La donna del lago, no. 8 (Cavatina Uberto), pp. 545–580 Zelmira, no. 8 (Finale I), pp.589–678 Zelmira, no. 7 (Duettino Zelmira-Elivira), pp.474–479 Torvldo e Dorliska, no. 3 (cavatina Torvaldo), pp.29–34 (cavatina only) La donna del lago, no. 3 (Cavatina Malcolm), pp.248–292 F-Po A.554.a.II (Continued)

Robert Bruce (1846)

Sources

12. Duo ‘Loyale famille!’ sois fière de ta fille’ 144–150 Récitatif ‘On m’a dit vrai’ 151

Zelmira, no. 3 (terzetto), pp.119–150 Not in Niedermeyer’s hand in F-Po A.554.a.II

13. Trio ‘Sort funeste, mon Dieu j’appelle les dangers’152–155 ‘Vous me tromper’ 155–157 ‘Suis moi’ 157–169

Zelmira, no. 10 (Quintetto), pp.869–888 La donna del lago, no. 9 (Trio), pp.636–643 La donna del lago, no. 9 (Trio), pp.669–706

14. Scène et chœur ‘Restez qu’allez vous faire’ 170–178 15. Finale ‘La guerre sans trève’ 179–184 ‘Sire, montrez vous’ 185–186 ‘Que l’hymne du barde’ 186–9 ‘Salut! Gloire, 189–202

La donna del lago, no. 7 (Finale I), pp. 494–505 and 531–544 La donna del lago, no. 7 (Finale I), pp.477–493 Zelmira no. 8bis (Coro et aria Emma), pp.716–732 La donna del lago, no. 7 (Finale I), pp.512–525 La donna del lago, no. 7 (Finale I), pp. 464–467

ACTE III 16. ‘Oui demain l’Ecossais libre’ 203–205 Cavatine ‘Anges sur moi penchés’ 205–206 17. Scène et chœur ‘Oui, cette route souterraine’ 207–8 Chœur ‘Point de bruit’ 208–222 18. Chœur ‘Buvons, il faut saisir amis’214–222 IIIe Air de danse: Pas de trois 223–229 IVe Air de danse: Pas de deux 230–234 19. Chœur dansé ‘Jeune fille de la montagne’ 235–241 ‘Sire, les Ecossais’, 242–245 20. Sextuor et chœur ‘Puisqu’un destin barbare à jamais’ 246–258 259–278 21. Finale ‘Arrête et pour ta vie tremble a ton bourreau’ 279–285

F-Po A.554.a.III, p.1 Zelmira, no 2 (cavatine Polidoro), pp.106–115 F-Po A.554.a.III, pp. 11–12 Armida, no. 7 (coro di furie), pp.118–125 F-Po A.554.a.III, pp.37–51 Moïse, air de danse no. 1, pp.186–191 F-Po A.554.a.III, pp.117–133 Maometto II, no. 11 (introduzione [act ii]), 145–153 F-Po A.554.a.III, pp.149–60 Bianca e Faliero, no. 29 (canone del quartetto), pp.316–323; Bianco e Faliero, no. 30 (seguito e stretta del quartetto), pp. 324–335 [excludes the stretta] Armida, no. 3 (quartetto), pp.39–51 [stretta only] La donna del lago, no. 7 (Finale I), pp.512–525

Key: Robert Bruce (1846)

Répértoire / DES OPÉRAS FRANÇAIS. / - / ROBERT BRUCE / OPÉRA EN TROIS ACTES / PAROLES DE MM ALPH. ROYER ET GUSTAVE VAËZ, / MUSIQUE DE / G. ROSSINI, / AVEC ACCOMPAGNEMENT DE PIANO / Par Niedermeyer. / … / A PARIS / CHEZ E. TROUPENAS ET CIE. / RUE NEUVE-VIVIENNE, 40.

Sources:

Slim, H. Colin, ed., La donna del lago, melo-dramma in due atti de Andrea Leone Tottola, musica di Gioachino Rossini, 3 vols, Edizione critica delle opere di Gioachino Rossini, Sezione prima xx (Pesaro: Fondazione Rossini, 1990) Greenwald, Helen and Kathleen Kuzmick Hansell, eds, Zelmira,dramma per musica de Andrea Leone Tottola, musica di Gioachino Rossini, 3 vols, Edizione critica delle opere di Gioachino Rossini, Sezione prima xxxiii (Pesaro: Fondazione Rossini, 2005) BIANCA E FALIERO / O SIA / IL CONSIGLIO DEI TRE / MELODRAMMA IN DUE ATTI DI FELICE ROMANI / POSTO IN MUSICA DA / GIOACCHINO ROSSINI / Rappresentato per la prima volta nel R. Teatro ala Scala in Milano il Carnevale del 1820 / RIDUZIONE PER CANTO E PIANOFORTE DI L. TRUZZI / … / MILANO – NAPOLI / FIRENZE, Ricori Jouhaud. TORINO, Giudici e Strada. – MENDRISIO, Bustelli-Rossi MOÏSE / OPÉRA / en / quatre actes, / Musique de / G. ROSSINI / PARTITION PIANO ET CHANT / … / PARIS, BRANDUS et CIE., Editeurs, / Successeurs de MSS SCHLESINGER et de TROUPENAS et CIE., / propriétaires de l’ancien fonds de Musique du Conservatoire, / Rue Richelieu, 87 et Rue Vivienne, 40 ARMIDA / Opera Seria / in tre Atti / Del Signor / G. ROSSINI / Ridotto / Per il Piano-Forte / … / A PARIS / Chez JANET ET COTELLE, Editeurs Mdes de Musique, du ROI, au Mont d’Or, Rue St . Honoré, No . 123, Hôtel d’Aligre / Et Rue de Richelieu, No. 92, près celle Feydeau MAOMETTO SECONDO / Opera Seria / in due Atti / Musica del Signor / G. ROSSINI / Ridotto / per il Piano-Forte / … / A PARIS / Chez JANET et COTELLE, Editeurs Mdes de Musique, du ROI, au Mont d’Or, Rue St . Honoré, No . 123, Hôtel d’Aligre / Et Rue de Richelieu, No. 92, près celle Feydeau TORVALDO E DORLISKA / Melodramma semiserio in due atti di Sterbini / POSTO IN MUSICA DA / GIOACHINO ROSSINI / Rappresentato per la prima volta al Teatro Valle in Roma / il Carnevale del 1816 / Riduzione per CANTO con accompagnamento di Pianoforte di LUIGI TRUZZI /.. / MILANO – NAPOLI / FIRENZE, Ricori Jouhaud. TORINO, Giudici e Strada. – MENDRISIO, Bustelli-Rossi F-Po A.554.a.I–III

Rossini’s second grand opéra  263 and suggested the inclusion of the cavatine for Bruce ‘Point de bruit’63 Vaëz was at pains to point out to Pillet that, although they were leaving aside ‘the third act as you have sketched it out [emphasis added]’ after Rossini’s additions, what then followed was ‘your third act, faithfully in accordance with the plan that you showed me and which will be good [emphasis added]’. Finally, Vaëz attempted to excuse this act of enforced lèse-majesté, by observing that to approach, for the first time, a plan, clear to whomever conceived it, is to enter into a darkened room; at first, one sees nothing, then the eye gets used to this darkness, and then one distinguishes the outlines, at first confused and indistinguishable.64 Vaëz’s communications with Pillet do nothing but praise Rossini’s commitment to the project; he confesses that Rossini gives us complete and frankly willing assistance that I was far from expecting. He is daily at our disposal, he has meetings as long as we wish; before showing us a piece of music, he explains the situation well, and it is because he has completely understood the emotions of the characters, their passions, at a given moment, that he decides on the choice of a such an andante or such an allegro.65 Rossini’s choices of music unknown in Paris were to prove valuable in the press reception of Robert Bruce’s premiere, but his involvement with the structure of the work did not stop there. Vaëz reports occasions when Rossini agreed to the use of particular numbers pencilled in by Pillet but insisted on changing the order; this proved particularly problematic when it involved reorganising the music planned for Rosine Stolz, Pillet’s spouse; Vaëz’s explanation is a model of subtle diplomacy.66 To describe Robert Bruce as a pasticcio does not adequately take account of its background. While Rossini was in Paris in 1843, the first of the Opéra’s unsuccessful attempts to mount his Italian operas had been a translation of 63 Letter from Vaëz to Pillet, 10 July 1846 (F-Pn L.a. Vaëz [G.] 4, 277). 64 ‘votre troisième acte tel que vous l’avez crayonné [emphasis added]’ … ‘ce qui viendra ensuite sera votre troisième acte, fidèlement conforme au plan que vous m’avez indiqué et qui sera bien [emphasis added]’…. ‘Aborder, pour la première fois, un plan, clair pour celui qui l’a conçu, c’est entrer dan une chambre sans lumière; au premier moment, on ne voit rien, puis l’œil s’accoutume à cette obscurité et l’on distingue ensuite nettement les lignes d’abord confuses et uniformes’ (ibidem). 65 ‘nous prête une assistance que j’étais loin d’attendre aussi complète et aussi franchement volontaire. Il est tous les jours à notre disposition, il nous donne des séances aussi longues que nous les voulons; avant de nous désigner un morceau de musique, il me fait bien expliquer la situation, et c’est lorsqu’il a bien saisi le sentiment des personnages, leurs passions dans ce moment donné qu’il détermine le choix de tel andante ou de tel allegro’ (ibidem, 276). 66 Ibidem, 279–281.

264  Rossini’s second grand opéra La donna del lago, and in many respects, Robert Bruce may be seen as the partial fulfilment of that plan. The Scottish (in the geographical as well as the authorial sense of the word) context of Robert Bruce clearly relates the two works, but – as Table 7.2 shows – La donna del lago was Robert Bruce’s main source; its music appears in ten of the twenty-one numbers in the score. The next most frequently employed opera is Zelmira, which appears in six numbers. Armida, Torvaldo e Dorliska, Moïse or Maometto II are each used once only, and Bianca e Falliero is used twice. The emphasis given La donna del lago and Zelmira proved an important point of departure, both in the critical reception of Robert Bruce and in more general discussions of the new and the old as the July Monarchy came to a close.

Rights, fidelity and reception Both Othello and Robert Bruce created ripples on the surface of the Parisian theatrical lake. The Opéra had been expressly permitted to mount foreign opera in translation throughout the July Monarchy, and as long as works by composers who were no longer living (Don Juan or Le freyschütz) were involved, this was largely unproblematic. As soon as works by living composers, that formed part of the repertory of the Théâtre-Italien, came into question, the situation changed. The productions of Rossini at the Opéra in the 1840s not only stand at the very beginning of conflicts between the Opéra and Théâtre-Italien, but also – together with the 1846 production of Lucie de Lammermoor and the 1847 Jérusalem – exemplify many of the characteristics of the interaction of French and Italian traditions in Paris during the Second Empire, characteristics that would be most clearly dramatised in the legal difficulties surrounding the reception of Verdi.67 Pillet’s 1840 cahier des charges, in addition to stipulating the number of grands opéras and ballets each season, required ‘a second opera in two acts, or 67 Jérusalem has been extensively discussed: David R. B. Kimbell, ‘Verdi’s First Rifacimento: “Il Lombardi” and “Jérusalem”’, Music & Letters 60 (1979), 1–36; Julian Budden, The Operas of Verdi, 3 vols (Oxford: Oxford University Press, 1978–81), 1:339–359; Ursula Günther, ‘Giuseppe Verdis erster Erfolg in Paris’, Lendemains 31–32 (1983), 53–62; Arrigo Quattrocchi, ‘Da Milano a Parigi: Jérusalem, la prima revisione di Verdi’, Studi verdiani 10 (1994–5), 13–60. It should be noted that enthusiasm for Jérusalem only really dated from the 1980s; see the critique of earlier views in Budden, The Operas of Verdi, 1:342. For Lucie de Lammermoor, see Everist, ‘Donizetti and Wagner: opéra de genre at the Théâtre de la Renaissance’, Giacomo Meyerbeer and Music Drama in Nineteenth-Century Paris, Variorum Collected Studies Series CS805 (Aldershot: Ashgate, 2005), 323–329. For Verdi’s later legal difficulties, see Anik Devriès-Lesure, ‘Les démêlés de Verdi avec le Théâtre-Italien sous la direction de Torribio Calzado (1855–63)’, L’opéra en France et en Italie (1791–1925): une scène privilégiée d’échanges littéraires et musicaux, ed. Hervé Lacombe, Publications de la Société Française de Musicologie 3/8 (Paris: Société Française de Musicologie; Klincksieck 2000), 235–261; Christian Sprang, Grand Opéra vor Gericht, Schriftenreihe des Archivs für Urheber-, Film-, Funk- und Theaterrecht 105 (Baden-Baden: Nomos, 1993).

Rossini’s second grand opéra  265 two operas in one act of which one may be replaced by a translated work with foreign music’.68 To many at the time, this would have seemed to run counter to the clause in Eugène Watel’s cahier des charges at the Théâtre-Italien in which the state ‘guarantees not to authorise at any Parisian or suburban theatre the production of lyric works in a foreign language’.69 The difference had already proved contentious in the case of the aborted production of L’italiana in Algeri.70 The decision as to where the ownership of the work lay resided in the hands of the Commission des théâtres, in whose report the key words were Indeed, if, on the one hand, the manager of the Théâtre-Italien is granted through his cahier des charges the exclusive licence to produce lyric works in a foreign language, on the other hand, it is no less certain that the cahier des charges of the Académie Royale de Musique enshrines past practice in the matter of playing translations of foreign lyric works.71 Faced with an obvious overlap in the terms of the two licences, the commission found a way out that distinguished, for perhaps the first and last time during the licensing period, between ‘ouvrages lyriques en langue étrangère’ and ‘traductions d’ouvrages lyriques étrangers’. While the commission allowed Pillet to continue with the production of Othello, it attempted to mollify Watel with a stipulation that Pillet should consult him, via the commission, when Pillet next wanted to put on an opera in which the Théâtre-Italien might have an ­interest.72 The question of Lucie de Lammermoor’s transfer from the ThéâtreItalien to the Opéra was confused because not only had its Italian original ‘belonged’ to the Théâtre-Italien, but its French translation had originated at the Théâtre de la Renaissance at the end of the previous decade. The commission was, however, far more concerned about whether Lucie would count as one of Pillet’s new works for the season than with any question of aesthetics or ownership.73 Robert Bruce elicited the same sort of outrage that Othello had generated, but Pillet managed to circumvent the difficulties despite having failed to notify the 68 ‘un second opéra en deux actes, ou deux opéras en un acte dont l’un pourrait être remplacé par un ouvrage traduit avec de la musique étrangère’ (Pillet’s cahier de charges, 1 June 1840 (F-Pan F21 4655 [5])). 69 ‘s’engage à n’autoriser sur aucun Théâtre de Paris ou de la banlieue de Paris, la représentation d’ouvrages lyriques en langue étrangère’. This was exactly what Watel cited in his letter to the Minister of the Interior, 25 May 1844, citing incidentally the fact that this cahier des charges was less than a month old (F-Pan AJ13 183). 70 See 250. 71 ‘En effet, si, d’un côté, le directeur du Théâtre Italien est investi par son cahier des charges du privilège exclusif de représenter des ouvrages lyriques en langue étrangère, de l’autre, il n’est pas moins certain que le cahier des charges de l’Académie Royale de Musique consacre … l’usage passé en force de droit de jouer des traductions d’ouvrages lyriques étrangers’ (Report from Commission spéciale des théâtres royaux, 8 July 1844 (F-Pan AJ13 183)). 72 Ibidem. 73 Commission des théâtres. Registre des Procès-Verbaux des délibérations, 19 January 1846 (F-Pan F21 4634 [1]).

266  Rossini’s second grand opéra commission that he was mounting a translation. His defence was simply that Robert Bruce was not a translation in the same manner as Othello or Lucie de Lammermoor since its libretto was completely new and the music had been chosen by the composer himself.74 Othello and Robert Bruce brought into focus questions of novelty, fidelity and canonicity on which much contemporary musical discourse focussed. Although many critics pointed to the sources of Robert Bruce as evidence of a lack of novelty, Julien Lemer questioned just how well known were the operas on which it was based: We, who have never heard ZELMIRE, nor BIANCA E FALIERO, nor ARMIDE, nor ERMIONE, nor TORVALDO E DORLISKA, nor perhaps LA DONNA DEL LAGO, we who, all at least deprived of this opera for ten months, wish to judge ourselves the effect of these pieces that are said to be so disparate.75 Théodore Anne, writing in La France, argued similarly that ‘seriously interrogating the 1,900 people present at Wednesday’s performance, one would not find 100, perhaps not even 50, initiated into the mystery of this antiquity so proposed’.76 In other words, there was some doubt as to the currency of the works that made up the pasticcio: for Lemer and Anne, speaking not for the critical fraternity but for the rest of the audience, the presence of pre-­ existing numbers that were known only to a few cognoscenti could not vitiate the power of a work based on a new libretto. Notwithstanding any controversy as to the exposure of Parisian audiences to Rossini’s Italian operas – especially La donna del lago and Zelmira – R ­ obert Bruce focussed questions of originality and unity in a particularly intense way.77 Although Lemer’s view effectively neutralised the question for his 74 The correspondence between Pillet and Antoine Passy (undersecretary of state at the Ministry of the Interior) is a model of evasion and prevarication (3 and 9 October 1846). The Commission’s response (6 November 1846) was to permit the production while chiding Pillet with preferring a pasticcio to a new work – itself problematising anew the issue of Rossini’s creative silence. The exchange and report are both in F-Pan AJ13 183. 75 ‘nous qui n’avons jamais entendu ni ZELMIRA, ni BIANCA Et FALIERO, ni ARMIDE, ni ERMIONE, ni TORVALDO ET DORLISKA, ni même peut-être la DONNA DEL LAGO, nous qui, tout au moins sommes privés de cet opéra depuis dix ans, nous voulons juger par nous-mêmes de l’effet de ces morceaux qu’on dit si disparates’ (La sylphide, 10 January 1847). 76 ‘en interrogeant sérieusement les 1,900 personnes qui assistaient à la représentation de mercredi, on n’en trouverait pas 100, peut-être même pas 50, initiées au mystère de cette antiquité qu’on met en avant’ (La France, 4 January 1847). 77 Although La donna del lago had been a staple of the Théâtre Italien’s repertory with performances running from 1824–1829, 1833–1834 and 1838–1841, Zelmira only appeared in 1826–1827, in 1829 and a single performance in 1831. The Théâtre Italien responded to ­Robert Bruce with a revival of La donna del lago in 1848. See Albert Soubies, Le ­T héâtre-Italien de 1801 à 1913 (Paris: Fischbacher, 1913).

Rossini’s second grand opéra  267 readers, other critics took radically opposed views. Pier-Angelo Fiorentino’s critique evoked the authority of eighteenth-century masters who – he claimed – would not have engaged in such actions: Where are you, peaceful and venerated shades of Paisiello, Cimarosa and Mozart? Never would your name have served as the sign for such a cheap second-hand trade! It is you who possess genius and faith, the holy love of art and the respect of others and yourselves. You do not yield to importunities, to seductions nor to flattery; your force and your glory did not fall under Delilah’s hand!78 Fiorentino chose his examples badly, of course, since – probably unbeknown to him – all three named composers wrote substitute arias or themselves contributed to pasticci.79 For Fiorentino, it seemed, not only were pasticci beyond the pale but their history had been so comprehensively overlooked by the middle of the nineteenth century that their eighteenth-century practitioners could be called upon to chastise contemporary ones. Although his review is noteworthy for its splenetic tone, he rarely goes beyond personal criticism of Rossini for having supervised the work and uses the advantage that Pillet thought that he had in securing Rossini’s participation as an instrument of censure. Henri Blanchard went further in the Revue et Gazette musicale de Paris to say that ‘A pasticcio, composed of pieces taken here and there from the scores of the same master, is no more an opera than a pile of stones taken from the same quarry is not a house, a palace’; in other words, it had no redeeming features at all. For Blanchard, and – according to him – the entire nation, ‘we do not only want a score but a drama; where the unity of emotion, of colour, is placed at the highest level of essential conditions’, a demand that Pillet might quite reasonably have countered with the observation that Royer and Vaëz’s libretto for Robert Bruce was just such a drame.80 78 ‘Où êtes-vous, ombres paisibles et vénérées de Paësiello, de Cimarosa, de Mozart? [J]amais votre nom n’eût servi d’enseigne à un pareil brocantage! C’est que vous aviez le génie et la foi, le saint amour de l’art et le respect des autres et de vous-mêmes. Vous ne cédiez ni aux importunités, ni aux séductions, ni aux flatteries; votre force et votre gloire ne tombaient pas sous la main de Dalila!’ (Le constitutionnel, 3 January 1847). 79 Mozart wrote substitute arias ranging from his setting of ‘Warum, o Liebe’ for a German translation of Gozzi’s Li due notti affannose in Salzburg in December 1780 (K. 365a) to two arias inserted into Martin y Soler’s Il burbero di buon core in Vienna a decade later (K. 582 and 583); Paisiello was responsible for the pasticcio Lo sposo burlato (1778); Cimarosa’s Le donne rivali (1780) was modified with contributions from other composers in Moscow as Le due fidanzate in 1789, and his Il pittore parigino (1781) was reworked with additions by Francesco Cipolla six years later. Much the same could be said of any composer of Italian opera in the last quarter of the eighteenth century, suggesting that such a tradition had largely fallen outside the knowledge of even the best-informed critics of the 1840s. 80 ‘Un pasticcio, composé de morceaux pris çà et là dans les partitions du même maître, n’est pas plus un opéra qu’un amas de pierres tirées de la même carrière n’est une maison, un pâlais….’, ‘nous voulons non seulement une partition, mais un drame; où l’unité de

268  Rossini’s second grand opéra Such unnuanced criticism might sound familiar to the modern reader for whom works rather than events still form the subjects of music history. But in early 1847, panegyric was also a response. Léon Escudier could not contain his enthusiasm: Rossini! Rossini! Oh, the great name of this musician; so many shiny sonic nuggets fallen from his lyre, so many masterpieces. What is there, then, and what do you want from me to call me thus with all your trumpets? Once again, what do you want from me? Has some new sprit flapped his melodious wings in your abandoned temple? It is Rossini, Rossini who reappears: make way for the sun, let the king of harmony enter… What music, and what admirable music, in Robert Bruce. I know that all the pieces in this score are known; but what does that matter to me, they are no less admirable. Rossini’s music may be heard often and for a long time. Rossini belongs to that race of geniuses of which it may be said: ‘my eyes and my ears see and follow him everywhere’.81 Whether Escudier’s final quotation is loosely based on Molière or the Bible, his unquestioning acceptance of Rossini’s genius and his explanation – excuse even – for the fact that most of the music was known bears comparison with other voices speaking in praise of Othello.82 Neither Fiorentino and Blanchard nor Escudier represents a typical view of pasticcio in the mid-1840s, however: most opinions fell between the two extremes. A central consideration in favour of Robert Bruce was Rossini’s authorship, as when Gustave Héquet wrote in L’illustration: A new score would undoubtedly have been better: but when one cannot have what one wants, one has to be satisfied with what one finds, and a pasticcio by Rossini appears to us preferable to new works by many composers,83 sentiment, de couleur est placée au premier rang des conditions essentielles!’ (Revue et Gazette musicale de Paris, 10 January 1847). 81 ‘Rossini! Rossini! Oh le grand nom que celui de ce musicien; autant de paillettes sonores tombées de sa lyre, autant de chefs-d’œuvre. Qu’y a-t-il donc et que me voulez-vous pour m’appeler ainsi avec toutes vos trompettes? Que me voulez-vous encore une fois? Quelque génie nouveau vient-il abattre ses ailes mélodieuses dans votre temple délaissé? C’est Rossini, Rossini qui reparaît: place au soleil, laissez entrer le roi de l’harmonie…. Que de musique et d’admirable musique dans Robert Bruce. Je sais bien que tous les morceaux de cette partition sont connus; mais que m’importe, ils n’en sont pas moins admirables. La musique de Rossini, on peut l’entendre souvent, longtemps. Rossini appartient à cette race de génies dont on peut dire: mes yeux et mes oreilles le voient et le suivent partout’ (La France musicale, 3 January 1847). 82 Possible sources for the concluding quotation are Molière’s La princesse d’Elide, act iii, sc. 2 and Deuteronomy 29:4. Other sources praising the production were La France, 4 January 1847; Le presse, 4 January 1847; La sylphide, 3 January 1847; Le ménestrel, 3 January 1847; Le moniteur universel, 4 January 1847; Le siècle, 11 January 1847; Revue et gazette des théâtres, 14 January 1847. 83 ‘Mieux eût valu sans contredit une partition originale: mais quand on n’a pas ce qu’on désire, il faut savoir se contenter de ce qu’on trouve, et un pastiche de Rossini nous paraît

Rossini’s second grand opéra  269 a view echoed by the journalist Alphonse Duchesne.84 Despite a tendency to explain away pasticci by eliding them with simple translation (that at one point would lead Lemer to claim erroneously that Le comte Ory was a pasticcio), views ranged from those that divorced Rossini’s music from the theatre – in other words, turned Robert Bruce into a type of staged concert – or that claimed that popular success was the only test of a work’s value. Much greater emphasis was placed on the deceitful way in which the Opéra had claimed on its poster that Robert Bruce was an opéra nouveau. No one was fooled for a moment, and while Pillet’s claims of Rossini’s involvement were accepted and widely reported, it was also stressed that Robert Bruce was no more Rossini’s second grand opéra than had been Othello. Georges Bousquet, in his review of Robert Bruce, sought reasons for why the word pasticcio should be so uncongenial; he concluded that there was nothing innately disadvantageous about pasticcio, but that pasticci on the stage of the Opéra (and it has to be assumed that he was including translations in making this comment) symbolised an institution in distress and betrayed the ‘apathy into which music and the management entrusted with maintaining national dignity has currently fallen’.85 Much of the commentary on Robert Bruce justified its presence on the Parisian stage by pointing to the composer’s canonic status.86 Two years earlier, the Revue et Gazette musicale de Paris had privileged a single work, Othello, in the same way: Rossini’s score is admirable, it cannot be denied, and it is always good to offer to one’s public works of a recognised excellence; it is no less than accustoming artists to feel and give back the beauty of these rare creations…. It has been so often said that we know the entire work by heart: it has enjoyed, in this same space, the happiness that the classics of our theatre have not obtained in two centuries, that of encountering for each of its principal roles three or four superior artists, who have indelibly marked them.87

encore préférable aux œuvres originales de bien des compositeurs’ (L’illustration, 19 ­January 1847). 84 ‘It was thought to be saying a good deal by saying the following: Robert Bruce is a pasticcio, nothing more. Pasticcio, so be it, since that is what you want. But a pasticcio, when it is such a master as Rossini who bestows one on us, is still excellent fortune, and the bequest of a foreigner is sufficiently rare for us to make a little more of the event’ (‘On a cru dire beaucoup en disant ceci: Robert Bruce est un pasticcio, rien de plus. Pasticcio, soit, puisque vous le voulez. Mais un pastiche, quand c’est un maître comme Rossini qui nous en gratifie, est encore une excellente fortune, et l’aubaine est assez peu commune pour qu’on en fasse un peu plus de cas’) (Journal des femmes, January 1839, 39). 85 ‘marasme où était [sic] momentanément tombé [sic] l’art musical et l’administration chargée d’en maintenir la dignité nationale’ (Le commerce, 5 January 1847). 86 See for example Le moniteur universel, 4 January 1847; Le national, 7 January 1847; La quotidienne, 11 January 1847. 87 ‘La partition de Rossini est admirable, on ne saurait le nier, et il est toujours bon d’offrir à son public des ouvrages d’une excellence reconnue [emphasis added]; il ne l’est pas moins

270  Rossini’s second grand opéra Othello’s status was guaranteed, it was therefore argued, because everyone knew the music by heart and because it had accumulated a tradition of performance in which every role had been performed by three or four artists of the highest rank. This suggests that, in 1840s’ Paris, both novelty and canonic status could impart value to a work; indeed, a lack of novelty which was not balanced by such renown could be a serious detriment. This type of aesthetic fluidity created precisely the environment in which radically different views of the same work or production could coexist and be considered with equal attention. But one of the canonic attributes of Otello was also one of the objections to the 1844 Othello: in contrast to Lucia di Lammermoor, which was only three years old when it was first rendered into French, by 1844 Otello was nearly ten times as old. Hippolyte Prévost, while vaunting the difficulties of modern French opera in performance, argued that Rossinian singing styles were no longer taught, and it was therefore all the more difficult for modern artists to give as much of an account of the work as any of their justly celebrated predecessors.88 The view was even more robustly articulated in reviews of Robert Bruce, when even Théophile Gautier could bluntly state that ‘the lack of unity could have been masked by a superior performance; unfortunately, the tradition of Rossini’s music was lost at the Opéra a long time ago’.89 d’accoutumer les artistes à sentir et à rendre les beautés de ces rares productions…. Il a été si souvent donné, que nous en savons par cœur toutes les notes: il a joui, dans ce même espace, du bonheur que les chefs-d’œuvre classiques de notre théâtre n’ont obtenu qu’en deux siècles, celui de rencontrer pour chacun de ses rôles principaux trois ou quatre artistes supérieurs, qui les ont marqués à leur empreinte’ (Revue et Gazette musicale de Paris, 8 September 1844). 88 ‘The requirements of French lyric drama are more severe that those of Italian opera. In order to be less ornamented, to allow fewer embellishments, French singing is no less difficult to execute; the difficulty is just not the same. The composer, freer in his charms, in his choice of movements, of forms of his melodies, makes fewer advances with us, however renders his tasks more laborious. The latter is the slave of musical thought, when he has completely yielded his inspiration. In Italy, the singer only has to allow himself to slip, as it were, into the melodic clothing fitted to his size; all the difficulty is the composer’s. Hence the uniformity of the melodic gestures of Donizetti, Bellini and even of Rossini himself. The music of the composer of Le barbier is no longer found written in modern conditions; it requires an agility which is no longer the subject of vocal study, even in Italy’ (‘Les exigences du drame lyrique français sont plus sévères que celles de l’opéra italien. Pour être moins orné, pour permettre moins de fioritures, le chant français n’en est pas moins d’une exécution très-difficile; seulement la difficulté n’est pas la même. Le compositeur, plus libre dans ses allures, dans le choix de ses mouvemens, des formules de ses mélodies, fait chez nous moins d’avances au chanteur, pourtant rend sa tâche plus laborieuse. Celui-ci est l’esclave de la pensée musicale que lorsqu’il a rendu complètement son inspiration. En Italie, le chanteur n’a qu’à se laisser glisser, en quelque sorte, dans le vêtement mélodique ajusté sur sa taille; toute la peine est pour le compositeur. De là l’uniformité des coupes mélodiques de Donizetti, de Bellini et de Rossini lui-même. La musique de l’auteur du Barbier ne se trouve plus écrite dans les conditions modernes; elle exige une agilité qui n’est plus l’objet des études vocales, même en Italie’) (La sylphide, 7 September 1844). 89 ‘le défaut d’unité aurait pu être masqué par une exécution supérieure; par malheur, la tradition de la musique de Rossini est perdue à l’Opéra depuis longtemps’ (La presse, 4 January 1847).

Rossini’s second grand opéra  271

‘Reports of my death….’ Across a wide range of Parisian discourses surrounding Rossini during the 1840s, the composer’s creative silence was equated with creative death, a practice which developed effortlessly into an elision with death itself. Even Fiorentino, in his diatribe against Robert Bruce, recognised that Rossini ‘already saw posterity on its knees erecting statues and setting up altars’, and Gautier mused on the possibility of the composer ‘descending from this serene position; alive, he is present at his glory, and in his own presence is part of posterity’.90 This discursive range includes light-hearted accounts that dramatise attempts to convince Rossini to compose, which refer to putting a pistol to his throat, a remarkable anecdote associated with Mozart’s Requiem, stories based on Rossini’s place in the cemetery by Jules Janin and the sculptor Antoine Etex, a fictional dialogue with Adolphe Nourrit, and monumental and funerary discourses surrounding Etex’s statue of Rossini; it was enhanced by widespread metaphors of mourning and exhumation, already witnessed in the context of Jeanne d’Arc, Othello and Robert Bruce, and which resurfaces in the reception of the ‘Stabat mater’.91 Each of these cultural moves is discussed in detail later in the chapter. Casual lexical choices that invoke death, or threats of death, are found in texts that might be considered the most innocuously hagiographical. In Alexandre Dumas Père’s encounter with Rossini at the 1831 premiere of Meyerbeer’s Robert le diable, the following words are placed into Rossini’s mouth (in a pseudo-Italian dialect): Well, I think that, if my best friend was waiting for me at the corner of a wood with a pistol, put this pistol to my throat and said to me: “Rossini, you are going to compose your best opera!” I would do it.92 In Dumas’ ‘Un diner chez Rossini’, in the context of a discussion of Rossini’s apparently self-confessed laziness, Dumas asks ‘if a manager was waiting for

90 ‘voyait déjà la postérité à ses genoux lui élevant des statues et lui dressant des autels’ (Le constitutionnel, 3 January 1847); ‘descendre de cette position sereine; vivant, il assiste à sa gloire, et, vis à vis de lui-même, fait partie de la postérité’ (La presse, 4 January 1847). 91 Such texts receive almost no attention in the literature on Rossini and are usually dismissed as unreliable or without significance for the study of Rossini’s life and works. For an exception, see Benjamin Walton, ‘Rossini and France’, The Cambridge Companion to Rossini, ed. Emanuele Senici (Cambridge: Cambridge University Press, 2004), 33–34. 92 ‘Eh bien, ze pense que, si mon meillour ami m’attendait au coin d’oun bois avec un pistolet, et me mettait ce pistolet sour la gorze en me disant: “Rossini, tou vas faire ton meillour opéra!” ze le ferais’ (Alexandre Dumas, Mes mémoires 1830–1833 (Paris: Lafont, 1989), 613). Dumas’ Mémoires were originally published in 1852–1855 and were written between 1847 and 1852, thus nearly twenty years after the event but at almost exactly the same moment in the 1840s when such discourses of creative death were beginning to emerge around Rossini.

272  Rossini’s second grand opéra you at the corner of a theatre and put a pistol to your throat…’.93 A similar, but less extreme, measure is found in Dumas’ ‘Comment fut composé l’Otello de Rossini’, set in Naples in 1816 (around the composition of the original Italian version of Otello) but written in 1843 when Rossini was again in Paris and just before the 1844 revival of the same opera. The article takes as its central conceit the impresario Domenico Barbaja’s ambush (guet-apens) of Rossini after the composer has declared that he will not compose anymore; Rossini is then locked in Barbaja’s palace until he delivers, act by act, the score of Otello. Kidnapping is a less serious crime than the attempted murder described by Dumas, and the Barbaja anecdote overlaps with the much more widely distributed trope of Rossini’s leaving compositions until the last minute, but in the 1843 ‘Otello’ article the fictional account is executed rather than simply remaining a threat.94 Rossini had barely left Paris for Bologna when Antonio Zanolini tried to come to terms with the question of the composer’s creative silence. In his Rossini et la musique, published in Paris in 1836, Zanolini compares Rossini with Antonio Canova. Rossini is called the gran maestro, as Canova is called the prince of modern sculpture. These two men serve to delineate our century. Canova merits the name because he knew, better than anyone else, how to approach the ancients; Rossini, because, distancing himself from every model, he effaces the ancients and the moderns. But Canova’s works, which are a perfect imitation of the beautiful, will enjoy their glory as long as the arts bestow honour; Rossini’s works, which are the expression of the century, will die with the century [emphasis added].95 Zanolini had good reasons for comparing Rossini and Canova: both artists were Italians who spent significant time in France, and – it could be argued – changed artistic mores in their adopted country, although, as Zanolini suggests, in different ways. The comparison also has its darker side: Canova had 93 ‘… si un directeur vous attendait à l’angle d’un théâtre et vous mettait un pistolet sur la gorge…’ (‘Un diner chez Rossini’, Les mille et un fantômes, 3 vols (Brussels and Livorno: Meline, Cans; Leizpig: Meline, 1849), 2:87). 94 Alexandre Dumas, ‘Comment fut composé l’Othello de Rossini’, Le livre de feuilletons: recueil de nouvelles, contes, épisodes, anecdotes, etc., extraits de la presse contemporaine 1 (1843), 35–38. Le livre de feuilletons is a collection of texts, mostly reprinted from other publications, by Sand, Balzac, Hugo, Alphonse Karr, Scribe and so on. The Dumas text is not attributed to another journal (as are most of the others), so was probably written expressly for the publication in 1843. 95 ‘Rossini est appelé le gran maestro, comme Canova fut appelé le prince de la sculpture moderne. Ces deux hommes servent à désigner notre siècle. Canova mérita ce nom, parce qu’il sut, mieux que tout autre, se rapprocher des anciens; Rossini, parce que, s’éloignant de tout modèle, il effaça les anciens et les modernes. Mais les ouvrages de Canova, qui sont une imitation parfaite du beau, jouiront de leur gloire tant que les arts seront en honneur; les œuvres de Rossini, qui sont l’expression du siècle, mourront avec le siècle [emphasis added]’ (Antonio Zanolini, Rossini et sa musique (Paris: Bettoni, 1836), 12).

Rossini’s second grand opéra  273 died in Venice in 1822, at almost exactly the same moment as Rossini was moving to Paris, and Zanolini’s suggestion that Canova’s works would have value ‘tant que les arts seront en honneur’ while Rossini’s would die with the century seems to place a term on Rossini’s creative life.96 Zanolini does this, it seems, almost to threaten the composer, to goad him into action. He continues: Rossini might, with his new compositions, achieve more; but it appears that it is enough for him to be the first. If he saw someone follow him closely perhaps he would get back on track. While waiting for this, he laughs at those who tire themselves following him, attaching themselves to him to gain advantage.97 Whatever Zanolini’s motives, the matter of Rossini’s creative death was thrown into question in a way that would be developed beyond recognition during the following decade. Blurring boundaries between creative and physical death is central to a nearly contemporary fictional dialogue between Rossini and Nourrit, purporting to have taken place in Milan in 1839 and found in Félix Joseph Delhasse’s Vie de Rossini published the same year. The dialogue stresses the similar situations of both Rossini and Nourrit: both have voluntarily finished careers in Paris and find themselves in Italy. There is an undercurrent of retreat: Nourrit from Gilbert Duprez’s successes and Rossini’s from those of Meyerbeer. The dialogue directly addresses the question of creative and performative silence on the part of two individuals who were thought to have left the public eye in their late 30s (Rossini was 38 at the premiere of Guillaume Tell in 1829 and Nourrit was 37 in 1839). Nourrit is circumspect about his reasons for leaving France and his artistic silence, but Rossini is enthusiastically envoiced to admit to laziness and to retirement in Italy: ‘the people imagine that I am jealous of Meyerbeer and you of Duprez; they do not look further, and do not admit that one could get tired of working for them’;98 Delhasse has Rossini continue: ‘The cemetery, 96 Rossini wrote a cantata Alla memoria di Canova: omaggio pastorale, based on his La riconscenza, for the unveiling of a bust of the artist on 1 April 1823 at the invitation of the Ateneo e la Filodrammatica of Treviso. See Patricia Brauner (ed.), La riconoscenza, cantata pastorale di Giulio Genoino, musica di Gioachino Rossini…; Il vero omaggio, cantata di Giulio Genoio e Gaetano Rossi, musica di Gioachino Rossini, Edizione critica delle opere di Gioachino Rossini, Sezione secunda 5 (Pesaro: Fondazione Rossini, 2003), xliv–xlvii. Whether Zanolini was aware of the Canova cantata is an open question. 97 ‘Peut-être Rossini pourrait-il, par de nouvelles compositions, s’élever plus haut; mais il semble qu’il lui suffise d’être le premier. S’il voyait quelqu’un le suivre de près, peut-être se remettrait-il en chemin. En attendant, il se rit de ceux qui fatiguent à le suivre, se rattachant à lui pour s’élever’ (ibidem). 98 ‘la foule s’imagine que je suis jaloux de Meyerbeer et vous de Duprez; elle n’en cherche davantage, et n’admet pas qu’on puisse se fatiguer de travailler pour elle’ ([Félix Joseph Delhasse, (pseud. van Damme)], Vie de Rossini, célèbre compositeur … dédiée aux vrais adorateurs du célèbre maitre par un dilettante (Antwerp: Librairie nationale et étranger, 1839), 196).

274  Rossini’s second grand opéra it is a good name, and here is why I like it, since rest is my supreme joy, my passion, my life’.99 The undercurrent of retreat is supported by a range of military metaphors, made explicit from the beginning of the dialogue. Rossini describes both himself and Nourrit as ‘ayant pris nos invalides’, in other words, as receiving a pension for those wounded in battle and invalided out of the army.100 Nourrit replies in kind: ‘Allow me tell you, dear maestro, that with pensioners like us, one could still raise a fine army’. Nourrit’s next line makes clear not only the links between the two artists, but also the construction that could be put on their actions: ‘And we left the theatre of our joint exploits to the great astonishment of the people who understand nothing of abdication and insist on regarding ours as a suicide [emphasis added]’.101 At this point, the text’s complexity dissolves into its own history. It is not known if the book, published in 1839, was written in the knowledge of Nourrit’s own suicide (8 March 1839). The fact that Nourrit threw himself from a fourth-floor window in the same year as the publication of the text throws an entirely new light on its reception, on the reading one might give to Rossini’s part in the dialogue and on the way in which contemporaries would have consumed this fictional dialogue. The shadow of physical death over creative death looms even more strongly in another text from the late 1830s. Jules Janin’s Voyage en Italie is a travelogue of his journey in Italy in the second half of 1838. Like Dumas eighteen months later,102 Janin was impressed by the Fountain of Neptune in Bologna, but there the similarities between the two visits end. Janin goes to the cemetery of La Certosa, a place that he finds identical to the city: The town and the cemetery are the same thing; they shelter under the same roof, one goes there easily and despite oneself, by the very force of the road the leads there. To enter the town is to enter these tombs; the town and the tombs have the same form, they are surrounded by the same silence, they are more or less inhabited by the same people. To sleep here or to live there, it is the same thing.103 99  ‘Le champ de repos, c’est bien dit, et voilà pourquoi je l’aime, car le repos, c’est ma suprême félicité, ma passion, ma vie’ (ibidem). The jeu de mots depends on the substantive ‘champ de repos’ representing a cemetery, and ‘repos’ itself alone having the wider sense of rest. 100  Ibidem, 195. 101 ‘Permettez-moi de vous dire, cher maëstro, qu’avec des invalides tels que nous on lèverait encore une belle armée’ … ‘Et nous avons quitté le théâtre de nos communs exploits au grand étonnement de la foule qui ne comprend rien à l’abdication et s’obstine à regarder la nôtre comme un suicide’ [emphasis added]’ (ibidem, 196). 102 Dumas’ ‘Un diner chez Rossini’ is set in Bologna in 1840. 103 ‘La ville et le cimetière, c’est même chose; ils sont abrités par le même toit, on y va de plain-pied et malgré soi, et par la force même de la route qui y conduit. Entrer dans la ville, c’est entrer dans ces tombeaux; la ville et les tombeaux, ils ont la même forme, ils sont entourés du même silence, ils sont habités à peu près par le même peuple. Dormir ici ou vivre là-bas, c’est même chose’ (Jules-Gabriel Janin, Voyage en Italie (Paris: Bourdin, 1839), 228–229).

Rossini’s second grand opéra  275 Developing the analogy between the living in the city and the dead in the cemetery, Janin finds a gravedigger who draws his attention to the finest death that the city is going to know. Without mentioning a name, the gravedigger not only claims the importance of the man who is going to die in Bologna, but also goes on to give a full account of the statuary that will adorn the individual’s grave. All Europe in coalition will pay their last respects to this man who has charmed them so much. Italy, his native land, will place the first stone of the monument; France, his pupil, will furnish the marble and the sculptor; Russia, for whom he wrote war marches, will send the bronze; Germany which he dragged out of her troubled sadness, will write the funerary inscription.104 Janin is ostensibly kept in ignorance of the recipient of this funerary largesse, and his reader is similarly treated, until the gravedigger starts singing to himself the Willow Song from Otello, the grand air from Il barbiere di Siviglia and the finale from Mosè in Egitto. Astonished, Janin observes that the gravedigger takes charge not only of the tomb but also of the funeral oration. His astonishment is matched only as he leaves the cemetery, when passing by a fine house, he hears Rossini himself call his name.105 There is a remarkable echo of Janin’s Voyage, most obviously in the account of Etex’s May 1845 visit to Bologna to cast Rossini’s hands. The account is in the words of Etex’s anonymous travelling companion and author of the account (possibly Joaquín Espín y Guillén) in the Spanish journal La Iberia musical [y literaria]: [Etex] especially insisted that we visit the cemetery (La Certosa), which is one of the most noteworthy in Italy. The urge with which he recommended it made a lively impression on me. The cemetery in Bologna encloses the bodies of the city’s most famous men; has the Raphael of music [Rossini] perhaps determined the place he himself should occupy in that dark shelter of death?106

104 ‘Toute l’Europe coalisée rendra les derniers honneurs à cet homme qui l’aura tant charmée. L’Italie, sa terre natale, posera la première pierre du monument; la France, son élève, fournira le marbre et le statuaire; la Russie, dont il a écrit les marches guerrières, enverra le bronze; l’Allemagne qu’il a tirée de sa tristesse maladive, écrira l’inscription funèbre’ (ibidem, 237). 105 I bidem, 238. 106 ‘[Etex] ha particolarmente insistito che andassimo a visitare il Camposanto (la Certosa), che è uno dei più notevoli d’Italia. La premura con la quale ce la ha raccomandato, mi ha fatto una viva sensazione. Il cimitero di Bologna racchiude le salme di tutti gli uomini illustri della città: i Rafaello [sic] della musica avrebbe forse fissato il posto che dovrà occupare egli stesso in quel cupo asil della morte?’ (Radiciotti, Gioacchino Rossini, 2:289–290). Parts of the article are translated in Weinstock, 235–236, but not the extract cited here. Given that the editor of La Iberia musical y literaria, Joaquín Espín y Guillén, may have been

276  Rossini’s second grand opéra While Etex’s interest in the funerary statuary is explicable in purely professional terms, his speculation as to Rossini’s interest in the location of his own grave resonates remarkably with the similar lines given to the gravedigger in Janin’s Voyage. The textual and linguistic history of this incident (an English translation of Radiciotti’s Italian translation of a Spanish text relating the oral account of Etex’s words in French) makes interpretation difficult, but it is unlikely that Guillén knew Janin’s Voyage, and both ­authors – Janin and Etex/Guillén – may have been participating in the broader funerary discourse on Rossini prompted by his creative silence.

‘… are greatly exaggerated’ By the time Rossini arrived in Paris in 1843, then, a progressively morbid discourse surrounded the question of the composer’s creative silence. Promoted among others by Zanolini, Oettinger and Janin, at the very least it hinted at the equation between creative silence and creative death, and such an elision was easy to make in the context of early nineteenth-century French attitudes to la belle mort and creative failure.107 In the years after Rossini’s 1843 visit to Paris, and alongside preparations for the 1844 Othello and the 1846 Robert Bruce, the implicit juncture of Rossini’s creative silence with creative death reached its apogee. Even before the composer’s arrival, funerary allusions had been made with regard to the 1842 performances of the ‘Stabat mater’: introducing a ­movement-by-movement critique of the work, the critic for Le ménestrel noted that ‘a second hearing immediately permits the making of some judgement on the beauty and the weaknesses of this Rossinian exhumation [emphasis added]’.108 Such a commentary may be placed side-by-side with that on Jeanne d’Arc, which used exactly the same locutions and images.109 This was nothing, however, in comparison with the anecdote published in the Revue et Gazette musicale de Paris on 16 June 1844. Described precisely as a pendant to the story of the mysterious individual who commissioned Mozart’s Requiem, the 1844 report told how – on 8 June – a thin and pale young man had sought an interview with Rossini.110 The

responsible for most of its content, it is equally possible that he was Etex’s travelling companion. See Belén Vargas, ‘La Iberia Musical [IBM] (Madrid, 1842)’, Retrospective Index to Music Periodicals www.ripm.org/journal_info.php5?ABB=IBM consulted 15 May 2008. 107  See 283–285. 108 ‘une seconde audition permet présentement de porter quelque jugement sur les beautés et les défauts de cette exhumation rossinienne [emphasis added]’ (Le ménestrel, 23 January 1842). 109  See above. 110 T he anecdote returned to an early nineteenth-century view of Mozart’s last year and the commissioning of the Requiem. There is no evidence that the careful sifting of evidence found in Oulibicheff’s 1843 biography of the composer (in French and which subjected Nissen’s anecdotes and Gottfried Weber’s analysis to criticism) had any impact on the telling of this story. See Alexandre Oulibicheff, Nouvelle biographie de Mozart, suivie d’un aperçu sur l’histoire générale de la musique et de l’analyse des principales œuvres de

Rossini’s second grand opéra  277 former accused Rossini of having betrayed his gifts by his reluctance to compose and went on to declare that his love of music had collapsed into a mere mixture of sounds; he had decided on suicide, but had also decided that his death would be more bearable if his suicide coincided with the premiere of Rossini’s new opera. The composer was given a year in which to complete this work, but was warned by his tormentor: ‘You may be able to flee Italy, even leave Europe, I shall be able to reach you!’ Rather more blunt than even the most elaborate retelling of the story of Mozart’s commission for the Requiem, the exchange had an effect on Rossini: ‘since that day, the great composer has become anxious and abandons himself to melancholy’, but the editorial team at the Revue et Gazette musicale de Paris added ‘If it was there only the symptom of the birth of a masterpiece, no-one would have to complain of a cause that led to such a fine result’.111 The most remarkable way in which Rossini’s creative silence was reinterpreted as death was treating him as if he were indeed already dead by placing Etex’s statue of the composer in the foyer of the Opéra. It has already been seen how Etex’s interest in the composer resonated with textual discourses surrounding the composer’s death, and his statue undertook similar cultural work. By the mid-1840s and although still only in his mid-thirties, Etex was a major force in Parisian sculpture. Ever since receiving the commission from Adolphe Thiers for part of the decoration on the Arc de Triomphe in 1833, he had been regularly exhibited at the Salon and was contributing works to most major buildings in Paris; by the end of his career, his work could be seen in the Musée de Versailles, the Madeleine, the Palais de Luxembourg, Grand-Trianon, on the Place de la Nation, in Saint-Eustache, the Invalides and the Hôpital de Lariboisière. His private work included medallions and busts of Alfred de Vigny, the Duc d’Orléans, Pierre Leroux, Augustin Thierry, Chateaubriand, Raspail, Auguste Comte, Judith Gautier, Delacroix, Ingres, Ampère and Liszt.112 During Rossini’s 1843 sojourn in Paris, the composer had sat for Etex, and the bust that had been originally envisaged grew into a full-size statue. Mozart (Moscow: Auguste Sener, 1843), R as Mozart (Paris: Séguier, 1991), 185–231 and 723–725. The fictions surrounding Mozart’s Requiem have been dealt with in summary fashion in modern scholarship, so anxious to establish facts and discredit speculation. See Christoph Wolff, Mozart’s Requiem: Historical and Analytical Studies, Documents, Score (Oxford: Clarendon, 1994), 1–2, and Stafford, William, The Mozart Myths: A Critical Reassessment (Stanford: Stanford University Press, 1991), 78–79 and passim. 111 ‘Vous pourrez fuir l’Italie, quitter même l’Europe, je saurai vous atteindre!’ … ‘depuis ce jour le grand compositeur est devenu soucieux et s’abandonne à la mélancolie’ .. ‘Si ce n’était là que le symptôme de l’enfantement d’un chef-d’œuvre, personne n’aurait à se plaindre d’une cause qui amènerait un si beau résultat’ (Revue et Gazette musicale de Paris, 16 June 1844). 112 Stanislas Lami, Dictionnaire des sculpteurs de l’école française au dix-neuvième siècle, 4 vols (Paris: Champion, 1914–21; R Nendeln: Kraus, 1970), 2:294–307.

278  Rossini’s second grand opéra Admirers of the two men set up a subscription committee in 1844, led by Ludovic Vitet, which raised the money to have the statue executed in marble and set in place. The statue was completed and put in position as part of a solemn ceremony in June 1846 (Figure 7.1).113 Visual accounts of the full suite of statuary in the foyer of the Opéra are rare, but the best surviving witness is a pencil sketch made in February 1872 just before the fire that destroyed the Salle Le Peletier.114 While this,

Figure 7.1  A ntoine Etex, Rossini (L’illustration, 13 June 1846). Bibliothèque nationale de France. 113 T he fullest contemporary account of the commissioning of the bust and its translation into a full-size statue is in L’illustration, 13 June 1846, which also provided the image that serves here as Figure 7.1. 114 T here is also a bronze model of the Etex statue that survives today in the Louvre, inventaire RF 3668. See Jean-René Gaborit , Sculpture française II: Renaissance et Temps modernes (Paris: Réunion des Musées Nationaux, 1998).

Rossini’s second grand opéra  279 logically by 1872, has busts of Halévy (d. 1862) and Meyerbeer (d. 1864), it also records a bust of Verdi, suggesting that the statue of the living Rossini prompted the inclusion of another living but foreign composer (Figure 7.2).115 As far as can be established, when Rossini’s statue was placed in the foyer of the Opéra in 1846, he joined Paër, Rameau, Philidor, Gluck, Méhul, Quinault, Lully, probably Le Sueur and possibly Cherubini. Not only was no other living composer represented, but even the dead were immortalised by no more than a bust; Rossini was the only composer to be honoured with a full-size statue. Rossini’s image, then, stood out from among all others in terms of its size, the living status of its subject, the material in which it was cast and the celebrity of its artist. Public verdicts on the statue of Rossini were divided. There were those who, seeing that the project had the support of a large number of the rich and powerful – the Minister of the Interior, Charles-Marie ­Tanneguy Duchâtel and Vitet – saw little difficulty in celebrating the life and work of the composer. Vitet’s and Duchâtel’s enthusiasm for the Rossini statue may well have been related to their families having been the subject of  works  by Etex in 1843.116 Analogies were made with the statues of Voltaire in the  Théâtre Français (Etex had also produced a bust of André Chénier for the same institution in 1839) and of Grétry in the Opéra-Comique. When the statue of Rossini was unveiled in 1846, its supporters argued that: whoever sees this statue indeed sees Rossini himself. The resemblance is perfect. The sculptor has been able to reproduce all the features of the body as faithfully as the elements and the expression on the face. It is simple, it is true, and its aspect is however elegant, noble and completely monumental. It is worthy, in a word, of the man that it represents.117

115  ‘THEATRE DE L’OPERA / Bustes dans le Foyer / Côté de la Rue en commençant à gauche’ (F-Po Estampes: Salles, Opéra Le Peletier). The document is dated 23 ­February 1872 and was probably part of an inventory of the Salle le Peletier made in advance of the planned move to the Palais Garnier. The bust of Verdi was by JeanPierre Dantan and dates from 1866, in the wake of the Parisian Macbeth and during the composer’s work on Don Carlos. In addition, Dantan had created one of his famous caricatures which, according to a letter from Verdi to Tito Ricordi (28 January 1866), found its way to Sant’Agata. See Mary Jane Phillips-Matz, Verdi: A Biography (Oxford: Oxford University Press, 1993), 509. 116 Lami, Dictionnaire des sculpteurs, 2:299. 117 ‘…qui voit en effet cette statue croit voir Rossini lui-même. La ressemblance est parfaite. Le sculpteur a su reproduire toutes les habitudes du corps aussi fidèlement que les traits et l’expression du visage. Elle est simple, elle est vraie, et pourtant l’aspect en est élégant, noble et tout à fait monumental. Elle est digne, en un mot, de l’homme qu’elle représente’ (L’illustration, 13 June 1846).

Figure 7.2  THEATRE DE L’OPERA / Bustes dans le Foyer / Côté de la Rue en commençant à gauche’ F-Po Estampes: Salles, Opéra Le Peletier (23 February 1872). Bibliothèque nationale de France.

Rossini’s second grand opéra  281 Benoît Jouvin, in La France musicale, went even further: In this natural pose, insouciant even, that the sculptor has given to the musician, one can very quickly recognise this Rossini who defends himself against glory with the same enthusiasm that others seek it out; who does not pose more in the presence of the public than in that of his friends; who has finally left to artists, tormented by the thirst for renown, the futile pleasure of playing the great man with contested works, as one sees charlatans playing at science, with the great reinforcement of drums, brass chains and strokes on the bass drum!118 The public presentation of the statue was accompanied by a concert consisting of the ‘Stabat mater’ and act ii of Guillaume Tell, performed by current artists at the Opéra and some distinguished visitors: Antonio Tamburini and Laure Cinti-Damoureau, among others. The performance, according to supporters of the project, was heard in reverent silence: ‘At each chord, at each note that escaped this harmonious and overstretched lyre, the entire vast audience shuddered’, and at the end of the performance ‘each greeted, while going down the staircase, the image of the great artist to whom his contemporaries owe such pure joys’.119 Detractors were imaginative, witty and aimed at a number of targets, one of which was the verisimilitude of the statue itself. Although proponents thought that those who saw the statue saw Rossini, there were those, such as the correspondent for the Revue et Gazette musicale de Paris, who wished with all my heart to be able to say that it is a masterpiece, that it resembles the person whose features it is supposed to represent, that it reproduces all the vigour and all the finesse of his physiognomy, all the majesty of his forehead, all the witty mockery wandering around his mouth, or at least we would like to be able to praise the general aspect, the details of the costume and other accessories, but in truth we are forced to repeat what we have heard those say who know the author of Le barbier and Guillaume Tell: It is not him! It is even someone completely different to him! Without the name, which is on the base of the statue, no one would doubt that it is the effigy of a great man. One might 118 ‘Dans cette pose naturelle, insouciante même, que le statuaire a donné au musicien, on reconnaît bien vite ce Rossini qui se défend contre la gloire avec la même ardeur que d’autres la recherchent; qui ne pose pas plus en présence du public que devant ses amis; qui a laissé enfin, aux artistes, tourmentés par la soif de la renommée, le futile plaisir de jouer au grand homme avec des œuvres contestables, comme on voit les charlatans jouer à la science, à grand renfort de breloques, de chaînes de similor et de coups de grosse caisse!’ (La France musicale, 14 June 1846). 119 ‘A chaque accord, à chaque note qui s’échappait de cette lyre harmonieuse et trop tôt détendue, la vaste salle frémissait tout entière’ … ‘chacun salua, en descendant, l’image de ce grand artiste à qui ses contemporains doivent de si pures jouissances’ (L’illustration, 13 June 1846).

282  Rossini’s second grand opéra be tempted to take the statue for that of a retired administrator who amuses himself by checking the accounts of his successors.120 Other opponents concentrated largely on Rossini’s living status. An editorial in Le ménestrel could not help making a comparison with the tenor, PierreJean-Baptiste-François Elleviou, who had died in 1842 after nearly thirty years in retirement. Elleviou had been Mayor of Villefranche in later life, and Le ménestrel parodied this with an article entitled ‘Petite Chronique: Rossini Conseiller municipal’, which made the point that at least when Elleviou took on municipal duties (the sort that might lead to commemoration in marble), he had lost his voice and therefore had little with which to please the public. ‘In contrast to Rossini’ hardly needed to be said.121 More hard-hitting was the list of other composers who might also have been feted in such a way: Gluck, Piccini, Sacchini, Mozart and Beethoven were proposed by the Revue et Gazette musicale de Paris – ‘nos tous illustres morts’ – which similarly selected carefully foreign composers with a following in Paris. The same journal also asked why Meyerbeer, Auber and Halévy were not worthy of the same honour, especially – as the author pointed out – Rossini had written only a single original work for the Opéra, claiming with some reason that Le siège de Corinthe, Moïse and – perhaps unfairly – Le Comte Ory were arrangements.122 But on both sides of the argument, Rossini’s corporeal status was key. In Jouvin’s laudatory account of the inauguration of the statue, he made use of language which brought Rossini’s body to the centre of the discussion: ‘In disappearing from this stage that laments him, Rossini has become a symbol: it is no longer a body, it is a spirit’.123 ‘Disparaître’ is a routine synonym for dying, and ‘pleurer’ a heightened synonym for grieving, and if such lexical choice were insufficient, Jouvin states the case boldly: ‘ce n’est plus un corps, c’est un génie’. And as he goes on to praise Etex and his artistic skill, it becomes clear that Rossini’s spirit resides in the Etex marble while his body

120 ‘de grand cœur pouvoir dire que c’est un chef-d’œuvre, qu’elle ressemble à celui dont elle est censée représenter les traits, qu’elle reproduit toute la vigueur et toute la finesse de sa physionomie, toute la majesté de son front, toute la spirituelle moquerie errante autour de sa bouche, ou du moins nous voudrions en louer l‘ensemble général, les détails de costume et autres accessoires, mais, en vérité nous sommes forcé de répéter ce qu’ont dit tous ceux qui connaissent l’auteur du Barbier et de Guillaume Tell: Ce n’est pas lui ! C’est même tout autre beaucoup plutôt que lui! Sans le nom, qui est au bas de la statue, personne ne se douterait que c’est l’effigie d’un grand homme. On serait tenté de prendre la statue pour celle d’un contrôleur retiré qui s’amuse à vérifier les comptes de ses successeurs’ (Revue et Gazette musicale de Paris, 14 June 1846). 121 L e ménestrel, 24 March 1844. 122  Revue et Gazette musicale de Paris, 14 June 1846. Meyerbeer and Halévy would eventually be accorded this honour posthumously as the 1872 document discussed earlier shows. 123 ‘En disparaissant de cette scène qui le pleure, Rossini est devenu un symbole: ce n’est plus un corps, c’est un génie’ (La France musicale, 14 June 1846).

Rossini’s second grand opéra  283 is largely effaced from the discourse. Similarly, the hostile accounts in the Revue et Gazette musicale de Paris – despite a radically different conclusion – culminate in a clear elision of creative silence with death, expressed in blunt and unambiguous terms: ‘Why this statue of a living man who wishes to be taken for dead?’124 Going on to ask whether the statue was designed to elicit the elusive second grand opéra from the composer, the author argued that if no immediate interest was attached to the statue, why not wait for ‘l’heure qui consacre toutes les gloires’ – in other words, Rossini’s corporeal death?125 The irony that underpins the entire episode of the Etex statue of ­Rossini and the premature artistic, and to some physical, death that it apparently granted was that the statue outlasted the composer by only five years. ­Rossini died on 13 November 1868 and the statue, together with the rest of the Salle Le Peletier, was incinerated during the night of 28 October 1873. The episode, however, importantly brings Rossini reception into alignment with those of other composers, since Rossini’s profile outlined here puts him very much alongside others who were similarly monumentalised before their death. The literature on the posthumous monuments to Mozart and Beethoven is extensive, but two figures bear particular comparison with Rossini: Handel and Haydn.126 Needless to say, the impulses behind the monumentalisation of the two latter composers were radically d ­ ifferent – Handel’s a key moment in his change of self-image and Haydn’s part of an ongoing attempt to contain the composer within the confines of the eighteenth century.127 All three composers, however, share the distinction of memorialisation avant la lettre.

‘…la scène …dans un tombeau illuminé…’ Such a treatment of a living composer as witnessed in the texts discussed in the previous sections of this article conflicts with modern understandings of death in the Western world: highly medicalised, unambiguously to be regretted and a subject that is silenced in most conventional discourses. For the 1840s, however, an interest in death and creativity was a current concern, to the extent that it is difficult to imagine Rossini’s creative silence not forming the subject of a discourse of creative death. 124 ‘Pourquoi cette statue à un homme vivant qui voulut qu’on le tînt pour mort…?’ (Revue et Gazette musicale de Paris, 24 May 1846). 125 I bidem. 126 For Mozart, see Gernot Gruber, Mozart und die Nachwelt (Salzburg: Residenz Verlag, 1985) trans. R. S. Furness as Mozart and Posterity (London: Quartet, 1991), and for Beethoven, Alessandro Comini, The Changing Image of Beethoven: A Study in Mythmaking (New York: Rizzoli, 1987). 127 Handel’s monumentalisation is outlined in Suzanne Aspden, ‘“Fam’d Handel Breathing, tho’ Transformed to Stone”: The Composer as Monument’, Journal of the American Musicological Society 55 (2002), 39–90, and that of Haydn in Matthew Head, ‘Music with “No Past?”: Archeologies of Joseph Haydn and “The Creation”’, 19th-Century Music 23 (2000), 191–217.

284  Rossini’s second grand opéra The early nineteenth century saw a growth in the building of cemeteries: Paris’ Père-Lachaise was laid out in 1804, and Bologna’s Certosa cemetery two years later; London’s Highgate Cemetery was opened in 1839. All provided a context for what has been called a ‘cult of tombs’ within which sculptors of all abilities (and Etex was already a national leader) could develop styles and techniques to serve most strata of society.128 The period also saw the emergence of the concept of la belle mort: death in old age after a successful life, painless, prosperous and surrounded by adoring family.129 This was contrasted later in the century with la mort laide, most explicitly represented in fictional accounts with Flaubert’s Emma Bovary and Zola’s Nana the best-known examples.130 Artistic death was the inverse of la belle mort: an early death, frequently in poverty and often solitary. Fictional accounts would also focus on a failed career, as in the case of Balzac’s anti-heroes: Frenhofer in Le chef d’œuvre inconnu, Lucien in Splendeurs et misères des courtisanes or the eponym in Sarrasine. Balzac was perhaps the author who most systematically established the link between death and creation, although he consistently identified death as a condition for creation: ‘One is only born to art at the cost of a victorious crossing of the spaces of death, one’s own death’.131 Such a view complements the equation of creative silence with death that characterises discourses of the period on Rossini. It is impossible to exaggerate the difference between attitudes to death in the 1840s and those that prevail at the beginning of the twenty-first century. The historian of death, Philippe Ariès, makes the point forcefully: ‘Because death [in the nineteenth century] is not the end of the beloved, however hard the pain of the survivor, it is not ugly or to be feared. It is beautiful, and the corpse is beautiful’. But he goes further to argue that the presence at the deathbed in the nineteenth century is more than the usual participation at a ritual social ceremony, it is to be present at a comforting and exalting performance; visiting the house of the dead is rather like a visit to the museum: how beautiful it is!132

128 ‘Introduction’, Birth and Death in Nineteenth-Century French Culture, ed., Nigel Harkness, Lisa Downing, Sonya Stephens and Timothy Unwin (New York and Amsterdam: Rodopi, 2007), 10–11. 129 Philippe Ariès, L’homme devant la mort, L’univers historique (Paris: Seuil, 1977), 403. 130 Idem, Essais sur l’histoire de la mort en occident du moyen âge à nos jours (Paris: Seuil, 1975), 68 and 80. 131 ‘On ne naît à l’art qu’au prix d’une traversée victorieuse des espaces de la mort, mort à soimême’ (Isabelle Michelot, ‘Figures de l’artiste et comédiens du réel: de la difficile naissance à l’implacable mort dans La comédie humaine’, Birth and Death in N ­ ineteenth-Century French Culture, 103). 132 ‘Parce que la mort n’est pas la fin de l’être cher, si dure que soit la peine du survivant, elle n’est ni laide ni redoutable. Elle est belle, et le mort est beau’… ‘La présence au lit de mort est au xixe siècle plus que la participation coutumière à une cérémonie sociale rituelle, elle

Rossini’s second grand opéra  285 The idea of death as spectacle, analogous to a visit to a museum, although not for the purposes of instruction but for the aesthetic pleasures it affords, brings the argument very close to those surrounding Rossini in the theatre, and is paralleled by a commentary on the correspondence between two illfated lovers, Albert de La Ferronays and his fiancé, Alexandrine. In 1834, the two spend a last night at the theatre where Alexandrine declares her admiration for ‘the auditorium, the light, the stage…, it suddenly felt like being in an illuminated tomb’.133 Both Ariès and Michelot point to the masking or concealment of death as a consequence of its aestheticisation: death, according to the former, ‘hides itself behind its beauty’.134 It is, therefore, hardly surprising that the same period should see what has been described as a ‘statuomanie’. The statesman and man of letters, Jean-Pons-Guillaume Viennet, wrote in 1842 that ‘the mania for statues develops like an epidemic’,135 although it is not clear whether he is referring to what Maurice Agulhon has identified as ­‘monument-signal’, ‘monument-tombe’ or both.136 Even in this description, a year before Rossini arrived in Paris, Viennet embeds his comments in a discourse on death as he uses the word ‘épidémie’ to describe the enthusiasm for funerary statuary: he was referring not to the type of low-level outbreak familiar to modern ­Western medicine but probably to the 1832 cholera epidemic which not only killed 20,000 inhabitants of Paris (one in thirty of the population), but denied most of those killed the characteristics of la belle mort.137 Rossini’s pre-posthumous memorialisation takes on an entirely logical aspect when viewed in the context of a nineteenth-century fetishisation of death. The opposition between la belle mort and la mort laide gave an acceptable environment to such discourses as those discussed in this article, and the concept of the death-bed as spectacle – with the equivalence of a museum or especially a stage – simply refined that context as a background to introducing the living Rossini therein.

Conclusion Rossini’s creative silence after 1829 called forth a number of commentaries on the composer in an attempt to come to terms with what was considered variously retirement, laziness or an early creative death. His return, in est assistance à un spectacle réconfortant et exaltant; la visite à la maison du mort tient un peu de la visite au musée: comme il est beau!’ (Ariès, L’homme devant la mort, 466). 133 ‘la salle, la lumière, la scène …, il me sembla tout d’un coup être dans un tombeau illuminé’ (Pauline Craven, Récit d’une sœur: souvenir de famille, 2 vols (Paris: 1867), 1:61–63, cited in Ariès, L’homme devant la mort, 409). 134 ‘… se cache sous sa beauté’ (ibidem, 466). 135 ‘La manie de statues se propage comme une épidémie’ (Journal de Viennet, pair de France, témoin de trois règnes, 1817–1848 (Paris: Amiot-Dumont, 1955), 266). 136 Maurice Agulhon, ‘La ‘Statuomanie’ et l’histoire’, Ethnologie française 8 (1978), 149. 137 A nge-Pierre Leca, Et le choléra s’abattit sur Paris: 1832 (Paris: Albin Michel, 1982).

286  Rossini’s second grand opéra 1843, after an absence of nearly a decade, to the site of the Parisian triumph of Guillaume Tell brought many of these discourses into sharp focus and brought back the idea of a successor to Guillaume Tell, a second grand opéra. At the centre of a long period of creative silence, Rossini competed with Donizetti, Meyerbeer and Halévy for the centre ground of Parisian stage music without writing a note. Rossini’s presence in Paris, planned exclusively as a visit to medical specialists, triggered a large number of planned productions of his works – almost all French translations of his Italian operas. Despite the energy with which such endeavours were promoted, none came to fruition during Rossini’s four-month sojourn in the city. The concrete outcomes of the increase in interest in the composer’s works were two substitutes for a new grand opéra: a translation of Otello in 1844 and the pasticcio Robert Bruce in 1846; although both productions offended against the emergent enthusiasm for works untainted by their event-based nature, they were well received by the majority of critics and audiences who, seeing and hearing an alternative to a new work, were pleased to consume Rossini’s operas in a different guise. All practical activity that attempted to return Rossini to the Parisian operatic stage during the 1840s took place in the context of a series of discourses about creative death. These took a variety of forms, were in a wide range of media and found different degrees of intensity. Vocabulary that invoked the funerary provided a backdrop to travelogues that focussed on cemeteries and Rossini’s possible place in them, and there were rumours that Rossini had indeed written a second grand opéra that was only to be played posthumously. The erection of a monument to the composer in the foyer of the Paris Opéra set the composer among a range of canonic figures, all of whom were dead, and unleashed further commentaries on the status of a composer who was physically alive but – so the argument went – could be considered creatively dead. Anecdotes that elided Rossini’s position with those concerning Mozart’s last year, in this context, became almost inevitable and established a clear link between creative and corporeal death. The development of Parisian discourses on Rossini in the decade after 1838 points to the importance of the city’s culture in the reception of the composer and in the formation of a canon of his works. His supporters sustained a series of discourses – textual, performative and iconographical – that contributed immeasurably to his burgeoning reputation. But such a reputation came at a cost, and Rossini was variously read as retired, creatively dead or even physically dead. Such a reading of his reputation would not begin to change until his triumphant and final return to Paris in the spring of 1855.138 138 If Rossini’s creative death might have been postponed by his return to Paris in 1855, any expectations that he might be creating a new grand opéra were dashed when, in a parallel move to those of the mid 1840s, Semiramide was produced at the Opéra with ballets by Carafa in 1860. See ACADEMIE IMPERIALE DE MUSIQUE / SÉMIRAMIS /.

8 A transalpine comedy L’elisir d’amore and cultural transfer

Operatic relationships between France and Italian states during the first half of the nineteenth century functioned in both directions: the Parisian careers of Italians from Cherubini to Verdi are well known, and, as the descriptive text that introduced this conference suggested, French theatrical traditions were critical models for the growth of opera in the primo ottocento: To adapt a French source for the Italian stage, it was therefore necessary to extract its essence in order to retranscribe it according to the representational codes typical of Italian opera.1 Such a statement raises two questions. First, there are various types of French ‘source’: drames, ballets-pantomimes, comédies, opera libretti, comédie-­ vaudevilles, mélodrames; second, each offers different opportunities to those who seek to retranscribe and therefore to subject them to different degrees of interpretation.

Cultural transfer, Italian Opera and the Livret d’opéra Comédies or drames, works written for the Théâtre Français, the Théâtre de la Porte Saint-Martin and the Théâtre Royal de l’Odéon, had a musical ­dimension – overtures, entr’actes and possibly choruses – but in the printed text of the play the evidence of the music was concealed.2 And whereas mélodrame – one of the central sources for Italian serious libretti of the primo ottocento – was conceived as work in which music played as important a role as in opéra 1 ‘Pour adapter une source française sur la scène italienne, il était donc nécessaire d’en extraire la matière pour la retranscrire selon les codes de représentation typiques de l’opéra italien’ (‘Appel à Communication’, Hernani – Ernani: 4 siècles d’opéra italien inspirés par la littérature française, Colloque International, 14–18 February 2007, 2). 2 Still the best set of clues to the musical dimensions of what might be considered ‘spoken’ theatre in nineteenth-century Paris are Nicole Wild’s introductions to the key institutions: the Théâtre Français, the Théâtre de la Porte Saint-Martin and the Théâtre Royal de l’Odéon. See her Dictionnaire des théâtres parisiens au xixe siècle: les théâtres et la musique (Paris: Amateurs des Livres, 1989), 96–99, 368–372 and 291–296.

288  A transalpine comedy comique, the printed texts also carried only the sketchiest indications of the music’s participation and – to a potential Italian consumer of the text – would convey no more information about the careful equilibrium between music and drama than printed comédies or drames.3 What remains an epistemological question for modern scholarship – what sources might be used to reconstruct mélodrame to give appropriate weight to the music – was also a conditioning factor in the cultural transfer between mélodrame and melodramma. When a French opera libretto was taken as a source, however, many more decisions had already been taken, and these were manifest in the printed text. In the case of an Italian libretto based on opéra comique, the position of the numbers was already determined; the poetry for the sung items was already in place, and in the case of those numbers prefaced with a recitative; the different types of poetry for aria and recitative (to be translated into versi lirici and versi sciolti, respectively) had already been identified. In the case of through-composed opéra, the kinetic parts of the work were already configured in recitative poetry. In such cases, the activity of the Italian librettist was much less ‘extraire la matière pour la retranscrire’ but much more to review decisions already taken, and then either to accept or reject them as the basis for translation. In both through-composed opéra and opéra comique, the printed libretto made explicit such decisions and presented them clearly for the consumption of any Italian librettist. Bibliographical control over French sources for Italian libretti, with the exception of the works of some major named composers, is at best patchy. Table 8.1 gives a list that is sufficiently extensive to allow some preliminary general remarks without making any claims to exhaustivity. Table 8.1 skirts some serious methodological difficulties concerning the question of common sources for more than one libretto and of multiple sources for a single libretto (the table attempts to avoid the former and marks the latter with an asterisk). It suggests that libretti of opéra comique outweigh those of opéra by a factor of three to one as sources for Italian music drama, and the number of French opéra settings is dominated by a relatively small number of works: La vestale, La muette de Portici, Gustave III and La reine de Chypre. In the context of a much wider range of opéra comique, some preferences are visible, particularly in the use of what used to be called ‘rescue opera’: Camille ou le souterrain, Lodoïska, Raoul, sire de Créqui, Béniowski and others. It also gives a broader context to the appropriation of Eugene Scribe’s libretto for Auber’s Le philtre, the source for Romani and Donizetti’s L’elisir d’amore. 3 Even in mélodrame, the musical dimensions still remain little explored. See Emilio Sala, ‘Tra melodrame e dramma borghese: dal Pasteur di Souvestre-Bourgeois allo Stiffelio di Verdi-Piave’, Tornando a ‘Stiffelio’: popolarità, refacimenti, messinscena, effettismo e altre ‘cure’ nella drammaturgia del Verdi romantico, ed. Giovanni Morelli (Florence: Olschki, 1987), 97–106; idem, ‘Melodrame: Definitions et metamorphoses d’un genre quasi-­ operatique’, Revue de Musicologie 84 (1998), 235–246; Nicole Wild, ‘La Musique dans le mélodrame des théâtres parisiens,’ Music in Paris in the Eighteen-Thirties, ed. Peter Bloom, La Vie musicale en France au xixe siècle 4 (Stuyvesant, NY: Pendragon, 1987), 589–610.

Table 8.1  I talian opera libretti based on French opera libretti, 1800–1899 Title

Composer

Librettist

Date

Source

Source Librettist

Source composer

Source Date

La Lodoiska Camilla

Mayr Paer

Gonella Carpani

1796 1799

1791 1791

Fioravanti, Caravita Valentino Paer Gonella Paer Schmidt Pucitta Romanelli Mayr Romanelli Morlacchi Perotti Fioravanti Tottola Valentino Rossini Foppa Rossini Sterbini Rossini Ferretti Morlacchi Romani Meyerbeer Rossi Pacini Romani

Loraux B.-J. Marsollier des Vivetières B.-J. Marsollier des Vivetières Loraux Bouilly Jouy Boutet de Monval Boutet de Monval Boutet de Monval

Cherubini Dalayrac

Camilla, ossia La forza del giuramento Lodoiska Leonora La vestale Raùl di Créqui Raoul di Crequy Raoul signore di Créqui

Lodoiska Camille, ou Le souterrain Camille, ou Le souterrain Lodoiska Léonore La vestale Raoul, sire de Créqui Raoul, sire de Créqui Raoul, sire de Créqui

Dalayrac

1791

Cherubini Gaveaux Spontini Dalayrac Dalayrac Dalayrac

1791 1798 1807 1789 1789 1789

1812 1815 1817 1818 1819 1820

Planard Loraux Etienne Claude de Saint-Just Bouilly Saint-Marcellin

Gaveaux Cherubini Isouard Boieldieu Méhul Catel

1808 1791 1810 1812 1803 1817

Rossini

Ferretti

1821

L’échelle de soie Lodoiska Cendrillon Jean de Paris Hélena Wallace, ou Le ménestrel écosssais Euphrosine

Hoffmann

Méhul

1790

Pacini Donizetti

Romanelli Romani

1823 1828

Fioravanti, Vincenzo Donizetti

?

1829

Tottola

Pavesi

Rossi

La scala di sieta Torvaldo e Dorliska La cenerentola Gianni di Parigi Emma di Resburgo Vallace, o sia L’eroe scozzese Matilde di Shabran, ossia Bellezza, e cuor di ferro La vestale Alina, regina di Golconda La conquista del Messico Elisabetta al castello di Kenilworth La donna bianca di Avenello

1804 1804 1804 1810 1810 1811 1811

Jouy Spontini Vial and de Favières Berton1

1807 1803

Jouy and d’Esmenard Scribe

Spontini

1809

1829

La vestale Aline, reine de Golconde Fernand Cortez, ou La conqête de Mexique Leicester

Auber

1823

1830

La dame blanche

Scribe

Boieldieu

1825 (Continued)

Title

Composer

Librettist

Date

Source

Source Librettist

Source composer

Source Date

Beniowski, ossia Gli esiliati in Siberia

Generali

Rossi

1831

Duval

Boieldieu

1800

La vendetta Fenella L’elisir d’amore Leocadia Betly Il postiglione di Lonjumeau La prigione di Edimburgo La vestale Clemenza di Valois Il duca d’Alba

Pugni Pavesi Donizetti Rossi Donizetti Coppola

Bassi Rossi Romani ? Donizetti Bassi

1831 1831 1832 1835 1836 1838

1828 1828 1831 1824 1834 1836

Carafa Spontini Auber Donizetti

1833 1807 1833 1839

1843 1843 1844 1846 1847

Gustave III Léocadie La reine de Chypre La reine de Chypre Le brasseur de Preston

Auber Auber Halevy Halevy Adam

1833 1824 1841 1841 1838

Rubino Artusi

1849 1851

Le domino noir Raoul, sire de Créqui

Scribe Scribe Scribe Scribe Scribe De Leuven and Brunswick Scribe and Jouy Jouy Scribe Scribe and Duveyrier Scribe Scribe Saint-Georges Saint-Georges De Leuven and Brunswick Scribe Boutet de Monval

Auber Auber Auber Auber Adam Adam

1838 1840 1841 1842

Béniowski, ou Les exilés du Kamtchatka La muette de Portici La muette de Portici Le philtre Léocadie Le chalet Le postillon de Lonjumeau La prison d’Edimbourg La vestale Gustave III Le duc d’Albe

Auber Dalayrac

1837 1789

1851 1852

Saint-Georges Scribe

Halévy Adam

1848 1850

1859 1861

Le val d’Andorre Giralda, ou La nouvelle Psyché Gustave III Le muletier de Tolède

Auber Adam

1833 1854

1880

L’enfant prodigue

Scribe Ennery and Clairville Scribe

Auber

1850

Il reggente Leocadie Caterina Cornaro La regina di Cipro Il birraio di Preston

Ricci, F Mercadante Gabussi Pacini

Rossi Cammerano Rossi Peruzzini and Piave Mercadante Cammarano Mazza Romani Donizetti Sacchèro Pacini Guidi Ricci, Luigi Guidi

La valle d’Andorra Giralda

Rossi Fioravanti, Vincenzo Cagnoni Cagnoni

Un ballo in maschera Il mulattiere di Toledo

Verdi Pacini

Giachetti Giachetti and R. Berninzone Somma Cencetti

Il figliuol prodigo

Ponchielli

A. Zanardini

Il domino nero Raoul di Créqui

1

Based in turn on Sedaine’s 1766 ballet

A transalpine comedy  291 The second question that the opening quotation raises is the relationship between French source and Italian convention, a question that may be pursued through the exploration of the concept of cultural transfer. ‘Extraction’ and ‘retranscription’ are simple concepts that are underwritten by discrete histories of French theatre and Italian opera; in this regard, they are analogous to the comparative use of national histories. They lead to the study of originals and copies, to readings of one text against another. But such ­c omparative studies are the direct target of those who have sought to p ­ romote a study of cultural transfer as an alternative, and in the study of opera and its sources, a grasp of the transfer of culture can assist in reconfiguring the ways in which national traditions and their interaction are understood. The work on Franco-German relationships by Michel ­Espagne and Michael Werner from the mid-1980s onwards is paradigmatic in this regard and offers much that may benefit attempts to explain the p ­ an-­Europeanisation of opera in the nineteenth century.4 While much of the ­detail that ­u nderpins the project of cultural transfer has been recovered from research into ­Franco-Prussian and Franco-Saxon transfers, the general trajectories of this mode of inquiry help explain the relationships between Italian and French music drama between 1800 and 1850. It is not difficult to see how the two questions that the quotation with which this article began elicit two contradictory tendencies: the straightforward identification of ‘source’ and ‘copy’, on the one hand, and the suspicion aroused by concepts associated with cultural transfer and by the closed comparative reading that such identifications seem to invite, on the other. Maintaining both questions in equilibrium is essential in contemplating one of the most remarkable instances of cultural transfer within the domain of opera of the primo ottocento – Romani and Donizetti’s L’elisir d’amore – and the complex, multilingual, literary background that supports it. The structures that gave rise to the French source for the libretto of L’elisir d’amore and to the ways in which the latter came into being depend on complex relationships between theatrical subjects, composers, librettists, singers, operatic conventions and institutional practices.

4 This is not the place to engage in a full description of the possible repercussions of recent studies of cultural transfer on the analysis of the internationalisation of opera during the primo ottocento. But a number of key themes emerge: the focus on individuals rather than texts; the concept of métissage: cross-breeding rather than confrontation; the disassociation of cultural-historic events from traditional political or military ones and the idea that identity and difference are not contradictory but complementary. For a useful introduction that self-consciously reviews much work of the 1990s, see Michel Espagne, Les transferts culturels franco-allemands, Perspectives Germaniques (Paris: Presses Universitaires de France, 1999), especially the two chapters entitled ‘Les limites d’une notion’ and ‘Au-delà du comparatisme’ (ibidem, 17–49).

292  A transalpine comedy

L’elisir d’amore and Le philtre Until very recently, the theatrical background to L’elisir d’amore was thought to be well understood. Donizetti and Romani, working under pressure of time for the premiere on 12 May 1832, took the libretto to Scribe’s Le philtre that had been set by Auber the previous year at the Paris Opéra and used this as a model for the new work; the Auber premiere had taken place only eleven months earlier on 15 June 1831.5 It was also thought that the libretto to Le philtre was in turn based on a novella by an otherwise unknown writer called Malaperta entitled Il filtro, apparently translated from the Italian by Stendhal.6 If this had been true, it would have represented an extraordinarily interesting case for two reasons: first, because the libretto for L’elisir d’amore was based on a French libretto, which was rarer than the more general absorption of French literary, poetic and dramatic culture in otttocento stage music; and second, because that French tradition was in turn based on an Italian one. There would have been here a fascinating bidirectional transfer of culture, with Scribe and Auber’s Le philtre mediating between the novella of Malaperta and Donizetti’s L’elisir d’amore. Much of this understanding of the background to L’elisir d’amore was, however, wrong. It is doubtful that any novelist called Malaperta ever existed, and Stendhal’s Il filtro was most likely a pseudo-Italian confection that he wrote himself. There is no correlation between the content of S ­ tendhal’s novella and that of the libretto to Le philtre. And despite the demolition of the first part of the story by most Stendhal scholars in the 1950s and the destruction of the second by Anselm Gerhard, this set of relationships is still alluded to regularly in accounts of L’elisir d’amore, in programmes, brochures and electronic media.7 There is also evidence that suggests that Scribe invented the scenario for Le philtre himself, independently of other sources. In his notebook that covers the period 1812–1826 are a series of notes on various scenarios; most of these are lined through, with the work in which they finally figured identified 5 Standard texts on L’elisir d’amore are Herbert Weinstock, Donizetti, and the World of Opera in Italy, Paris, and Vienna in the First Half of the Nineteenth Century (London: Methuen, 1964), 82–83 and 332–333; William Ashbrook, Donizetti (London: Cassell, 1965), 380, 401, 415–416, 441 and 453. 6 William Ashbrook, Donizetti and his Operas (Cambridge: Cambridge University Press, 1982), 72, note 53. The claim that Le philtre ‘later entered the repertory of the Opéra-­ Comique’ (ibidem) is not supported by evidence. See also Gérard Corneloup, ‘De Scribe à Romani, d’Auber à Donizetti: un elixir franco-italien’, Donizetti: L’elixir d’amour, L’avantscène opéra 95 (Paris: L’avant-scène Opéra, 1987), 5; Karin Pendle, Eugène Scribe and French Opera of the Nineteenth Century, Studies in Musicology 6 (Ann Arbor, Mich.: UMI Research Press, 1979), 683. 7 Anselm Gerhard, ‘Ein missverstandener Schabernack: Gaetano Donizettis eigenwilliger Umgang mit Felice Romanis L’elisir d’amore’, ‘Una piacente estate di San Martino’: Studi e ricerche per i settant’anni di Marcello Conati, ed. Marco Capra, Quaderni di musica/ Realtà: Supplemento 1 (Lucca: Libreria Editrice Musicale, 2000), 117–126.

A transalpine comedy  293 in the margin. According to this document, different parts of one such text served for Le philtre (1831) and for Haydée (1848); the following text is entitled Le sorcier and was the basis for act two of Gustave III (1833). The fact that all these fragmentary scenarios date from before 1826, and the absence of any alternative candidate for the source of Le philtre, argues strongly for Scribe’s authorship based on the characters of the commedia dell’arte.8 This series of misunderstandings is of importance because it masks a very real Italian origin for Le philtre’s libretto, one that has nothing to do with Stendhal but that has everything to do with the complex mix of stylistic impulses in Parisian libretti around 1830 and the consequent cultural transfer thus implied. Not only does the libretto of Le philtre travel to Italy as the basis for L’elisir d’amore, but it is itself the product of cultural transfer from Italy to France. This therefore returns to L’elisir d’amore a complex, bidirectional transfer of culture. Invoking Scribe and Auber in the context of Romani and Donizetti risks falling prey to modern prejudices about the relative value of the two works. Recent accounts of the relationship between Romani and Donizetti’s ­L’elisir d’amore and Scribe and Auber’s Le philtre take as axiomatic the almost self-evident higher quality of Donizetti over Auber.9 A return to the century in which L’elisir d’amore was composed, however, results in a very different picture. The unassuaged popularity of L’elisir d’amore in Italy meant that Le philtre was entirely bypassed as French opéra comique was imported into Italy, and no performances of Le philtre were heard on the peninsula until 1900.10 Both works, however, were regularly performed in London throughout the nineteenth century, and whether we listen to Henry Chorley writing in 1862 or Ebenezer Prout writing in 1900, the story is the same: Auber is a superior composer to Donizetti and Le philtre a superior work to L’elisir d’amore.11 We can, of course, choose to disregard these opinions as the ravings of the spokespersons for the Land ohne Musik, but it might just make us pause and reflect on the possibility of an even-handed attempt at understanding the two works, the ties that bind them and the context in which they were each created.

8 Scribe’s notes are in Paris, Bibliothèque nationale de France (hereafter F-Pn) n.a.f. 22584 (8), fols 47r–48r. 9 A view reinforced, or perhaps conditioned, by Ashbrook’s and Weinstock’s publications of the 1960s. See, for example, Ashbrook, Donizetti, 453, where Scribe and Romani are played off against each in a competition where the outcome (Romani’s superiority) is assured from before the beginning. 10 Alfred Loewenberg, Annals of Opera: 1597–1940, 3rd edn revised and corrected (London: Calder, 1978), 734. 11 ‘If … I must decidedly give the preference to the French work, I think that few who know both well will be inclined to disagree with me’ (Ebenezer Prout, ‘Auber’s Le philtre and Donizetti’s L’elisir d’amore: A Comparison’, Monthly Musical Record 30 (1900), 76); Henry Chorley, Thirty Years’ Musical Recollections, 2 vols (London: Hurst and Blackett, 1862), 2:212.

294  A transalpine comedy In Paris, similarly, L’elisir d’amore was as much a staple of the Théâtre-­ Italien as Le philtre was of the Opéra. The 1839 Paris premiere of the ­Donizetti work was an opportunity to compare the libretti and music of the two compositions that no critic could ignore. The very close relationship between Romani’s and Scribe’s texts proved problematic in a critical environment where it was expected that there would be something new in a libretto to interest audiences and critics alike. The absence of any novelty threw the weight of the comparison directly onto the two scores. The very best that Donizetti could receive was an honourable draw; as the critic in L’echo français wrote: ‘Ce duel artistique ne se terminera donc point à la défaite d’aucun des deux combattans: ni la bannière de la France, ni celle de l’Italie ne se baissera humble et humiliée’.12 While Auguste Morel in the Revue et gazette des théâtres made clear that ‘Nous craignions donc que tout autre compositeur, Donizetti lui-même, venant à se rencontrer avec [Auber] sur le même terrain, ne dut forcement s’avouer vaincu’ but then went on to argue that the two composers were not on the same ground since Auber was writing opéra comique and Donizetti opéra bouffe.13 The inaccurate terminology probably convinced contemporary readers as little as it does modern commentators. Le philtre is described as an opéra in all its public forms and referred to as petit opéra in administrative documents; L’elisir d’amore is called melodramma giocoso.14 Other critics were less concerned to avoid confrontation. The critic for Le constitutionnel was clear: If we had to decide between M. Auber’s score and Donizetti’s, our preference would not hesitate; it would be entirely for Le philtre against l’Elissir [sic]. One seems to us to have superiority over the other; the spirit is more frank and fertile, and we are struck more by the abundance of ideas, the clarity of expression, the flexibility, and the polish of tones and nuances, the charming variety with which the composer’s talent adapts to situations, feelings, characters, and ingeniously characterizes them.15

12 L’echo français, 22 January 1839, signed ‘Alf. D.-S.’. 13 Revue et gazette des théâtres, 20 January 1839. See Joël-Marie Fauquet (ed.), Dictionnaire de la musique en France au xixe siècle (Paris: Fayard, 2003), s.v. ‘Morel, Auguste’. I am grateful to Katharine Ellis for an exchange on the subject of the pseudonyms in the texts cited in this article. 14 Melodramma giocoso is one only of a wide range of terms that might be subsumed under the generic term opera buffa: farsa, farsa comica, opera buffa, dramma giocoso, melodramma giocoso, dramma buffo, burletta per musica, commedia, commedia lirica, melodramma buffo and melodramma comico. 15 ‘Si nous avions à nous décider entre la partition de M. Auber et celle de M. Donizetti, notre préférence n’hésiterait pas long temps; elle serait tout entière pour le Philtre contre ­l’Elissir [sic]. L’un nous semble avoir sur l’autre une supériorité; l’esprit y est plus franc et plus fertile; on y est frappé davantage de l’abondance des idées, de la netteté de l’expression, de

A transalpine comedy  295 And even when attempting to offer an even-handed account of the two works – giving the first act to Auber and the second to Donizetti – the anonymous critic of Le cabinet de lecture could not quite come up with words that admitted that the second act of L’elisir d’amore was superior to that of Le philtre: To draw a parallel between the two operas in a few lines, we say that Auber’s first act is superior, and that there is nothing in Donizetti’s that approaches the air of the sergeant, the key and so original number for Fontanarose and the air of Nourrit, Philtre charmant. But in the second act of Italian opera, there is a delightful chorus of young girls; then there is a delightful barcarolle: Yo son Ricco, imaginatively brought back to the end of the work; and finally a romance full of charms: Una furtiva Lagrima, which excited the applause of the whole auditorium.16

Le philtre, Le comte Ory and petit opéra The generic origins in petit opéra of, and the very close relationship between, the libretti of Le philtre and Le comte Ory reveal the Italianate background to Le philtre and therefore the bidirectional cultural transfer of L’elisir d’amore. Le comte Ory was a work for the Paris Opéra by an Italian composer, but it is half-based on Italian music: Rossini’s Il viaggio a Reims of 1825. Furthermore, Luigi Balocchi’s Italian libretto of Il viaggio based in part on Mme de Staël’s French novel, Corinne, is itself a text with much to say about Italy.17 Rather than any kind of straightforward transfer of French culture to Italy, the pre-history of L’elisir d’amore is a complex maze of references (Figure 8.1). la souplesse, et du fini des tons et des nuances, de la charmante variété avec laquelle le talent du compositeur s’accommode aux situations, aux sentiments, aux personnages, et les caractérise ingénieusement’ (Le constitutionnel, 24 January 1839, signed ‘A.’). The author remains unidentified in Kerry Murphy, ‘La Critique musicale dans les grands quotidiens parisiens de 1830–1839,’ Revue internationale de la musique française 17 (June 1985), 20. 16 ‘Pour établir en quelques lignes un parallèle entre les deux opéras, nous dirons que le premier acte d’Auber est supérieur, et qu’on ne trouve dans celui de Donizetti rien qui approche de l’air du sergent, du morceau capital et si original de Fontanarose et de l’air de Nourrit, Philtre charmant. Mais il y a au deuxième acte de l’opéra italien un délicieux chœur de jeunes filles; puis une ravissante barcarolle: Yo son Ricco, spirituellement ramenée à la fin de l’ouvrage; et enfin une romance pleine de charmes: Una furtiva Lagrima, qui a enlevé les applaudissements de la salle entière’ (Le cabinet de lecture, 25 January 1839). 17 For the relationship between Il viaggio de Reims and the background to the libretto in Mme de Staël, see Janet Johnson, ‘A Lost Rossini Opera Recovered: Il Viaggio a Reims,’ Bollettino del Centro Rossiniano di studi n.v. Nos 1–3 [bound as one] (1983), 5–57; eadem (ed.), Il viaggio a Reims, ossia L’albergo del giglio d’oro: dramma giocoso in un atto di Luigi Balochi [sic], music di Gioachino Rossini, 2 vols, Edizione critica delle opere di Gioachino Rossini 35 (Pesaro: Fondazione Rossini, 1999).

296  A transalpine comedy Corinne, ou L’Italie (de Staël, 1807)

Le comte Ory (Scribe and Delestre-Poirson, 1817)

Il viaggio a Reims (Balocchi and Rossini, 1825)

Le comte Ory (Scribe, Delestre-Poirson and Rossini, 1828)

Le philtre (Scribe and Auber, 1831)

L’elisir d’amore (Romani and Donizetti, 1832)

Figure 8.1  Relationships between music, libretti and source texts for Le comte Ory and Le philtre.

Figure 8.1 shows the links between the six works mentioned so far. In all cases except one, the links concern the reuse of material from one to another and are identified with a solid line. The single exception is the relationship between Le comte Ory and Le philtre marked with a broken line. No musical or poetic material is shared between the two operas, and the two plots are entirely different; indeed, Scribe and Delestre-Poirson’s style troubadour is entirely effaced as Scribe returns to the well-known characters of the commedia dell’arte for the libretto of Le philtre. But the generic association between Le comte Ory and Le philtre is equally as strong as the network of material that links the other works in Figure 8.1. The generic characteristics of the libretti of the two petits opéras, and their successors, may be enumerated as follows: • • • • •

The libretto is comic; The dramatis personae consist of four or five characters only; There are two acts with around ten scenes in each act; There is no dance; There are rarely more than twelve to fourteen separate musical compositions in the work;

A transalpine comedy  297 • • • •

Ensembles larger than a trio are rare; Petits opéras are never programmed alone and always accompany a ballet and have the following musical consequences: Numbers are articulated by a clear cadence and a following recitative in the manner of contemporary Italian opera; Italian forms are common ( finale, multi-movement aria).

There are even closer links between Le comte Ory and Le philtre laid out elsewhere.18 However, Rossini’s Le comte Ory shows all the signs of ad hoc solutions to its multiple sources that happened to have provided exactly the right generic mix at just the right time. Once this was recognised – and it has to be remembered that the pace of operatic change in Paris between 1828 and 1831 was colossal and the need for petit opéra becoming more pressing – for Scribe to write a new libretto that imitated Le comte Ory as closely as possible and for Auber also to replicate the music became not only essential but inevitable. Given the very clear Italianate features of both the music and libretto of Le philtre, its attraction to Romani and Donizetti was obvious: the use of a libretto with such a clear Italian pedigree that had elicited such unambiguous Italian musical responses meant that rapid appropriation was a plausibility that pragmatic librettists and composers could not overlook. The take-up of the libretto was very quick indeed: no more than eight or nine months probably elapsed between its Parisian premiere and the decision to use it as the basis for the new opera at the Canobbiana.

The Dabadie brothers The rapid absorption of a brand-new French opera libretto south of the Alps requires some sort of explanation. One explanation has centred on Henri-Bernard Dabadie, who had taken the role of the sergeant Jolicœur in Le philtre in Paris and who has been assumed to have taken the corresponding role of Belcore in L’elisir d’amore.19 The cast of L’elisir d’amore and of Le philtre, together with further detail for some of the artists, both at the Paris Opéra and at the Canobbiana’s spring season of 1832, are in Table 8.2.20 18 See Chapter 2. 19 The assumption is endemic. Weinstock, Donizetti and the World of Opera, 83; idem, Donizetti and his Operas, 72; Annalisa Bini and Jeremy Commons, Le prime rappresentazioni dell opere di Donizetti nella stampa coeva, L’arte harmonica serie III, Studi e testi 3 (Rome: Accademia Nazionale; Milan: Skira, 1997), 300; Norbert Miller, ‘L’elisir d’amore’ Pipers Enzyklopädie des Musiktheaters: Oper, Operette, Musical, Ballett, eds. Carl Dahlhaus and Sieghart Döhring, 7 vols (Munich and Zurich: Piper, 1986–1997), 1:747. 20 Pompeo Cambiasi, Rappresentazioni date nei Reali Teatri di Milano, 1778–1872, 2nd edition (Milan: Ricordi, 1872) R, Bibliotheca musica bononiensis 3:33 (Bologna: Forni, 1969), 104–105; Théodore de Lajarte, Bibliothèque musicale du Théâtre de l’Opéra: catalogue historique, chronologique, anecdotique, 2 vols (Paris: Bibliothèque Nationale, 1878; R Hildesheim: Olms, 1969), 2:138–139.

298  A transalpine comedy Table 8.2  Casting details for L’elisir d’amore, Le philtre and cognate works L’orfanello L’incognito L’elisir di Ginevra d’amore Sabine Heinefetter x Giovanni Battista x Genero Giuseppe x Frezzolini Henri-Bernard Dabadie Adolphe Nourrit Nicolas-Prosper Levassseur Laure CintiDamoureau

x x

Adina Nemorino

x

Dulcamara ?Belcore?

Le philtre

Le comte Ory

Jolicœur

x

Guillaume x Fontanarose x Térézine

x

Dabadie was the only member of the cast of L’elisir d’amore thought to have also been involved in Le philtre, and the possibility of his position in Milan in early 1832 deserves some attention. He took no role in the Parisian premiere of Robert le diable in late November 1831 and – more than most of his colleagues  – regularly absented himself from the Paris Opéra to tour the provinces and abroad. While archival sources clearly describe his annual tours to the French provinces – frequently including his native Pau – during the 1820s, the period around 1830 is frustratingly poorly documented.21 Nevertheless, there is no doubt that of all the artists of his stature at the Académie Royale de Musique, he travelled most, and – even after retirement – he was in dispute with the Opéra administration concerning the terms of his congé for the year 1834.22 In all the tours that he made in the French provinces during the 1820s, however, there is no evidence that Dabadie left the country, and although his participation in the premiere of L’elisir d’amore is an attractive story, it is largely false. Henri-Bernard Dabadie did not take the role of Belcore in the premiere of L’elisir d’amore. He was in Paris throughout the rehearsal period for the work and sang the role of Pietro in La muette de Portici the day before the premiere of L’elisir d’amore; he even sang the role of Jolicœur in Le philtre four days after the Donizetti premiere.23 How, then, could the libretto and the entire Milanese press talk about a Signor Dabadie participating in the premiere of L’elisir d’amore? The answer is that there was a second Dabadie, known in French sources from the late 1820s as Dabadie jeune: Henri-Bernard’s younger brother. A single reference in the Milanese press account of the premiere of 21 The litany of Dabadie’s negotiations with the Académie Royale de Musique are Paris, Archives Nationales (hereafter F-Pan) AJ13 111/VIII, AJ13 112/III, AJ13 113/V, AJ13 114/II, AJ13 114/IV, AJ13 115, AJ13 116/III and AJ13 186/III. 22 F-Pan F21 1056. 23 11 and 16 May 1832 respectively.

A transalpine comedy  299 L’elisir d’amore refers to ‘un esordiente basso-cantante francese, il sig. Giustino Dabadie’ reveals the Christian name of Dabadie jeune as ­Justin; furthermore, the descriptor ‘esordiente’ matches not at all the career of ­Henri-Bernard, who by 1832 had ten years’ experience at the Académie R ­ oyale de Musique, four of those as premier sujet.24 Further descriptions in the Milanese press point to Justin Dabadie’s inexperience and, as was habitual in accounts of the performance of debutants, offered explanations and advice: a newcomer, a basso-cantante (Dabadie), endowed with a beautiful voice and uncommon agility, beside himself with fear, that he never knew how to overcome during the whole course of the opera. He is to be flattered that in progress of time, reborn, he will deserve more and more the favour of that public, which encouraged him with kind applause.25 Justin was admitted to the Académie Royale de Musique in 1827, but his debuts were postponed because of sickness, and in early 1828 he was granted permission to take a congé to re-establish his health.26 He finally made his debut on 4 June 1828 as a prisoner in Spontini’s Fernand Cortez and, on 25 July of the same year, took the role of Cinna in La vestale. The performances were catastrophic, although he remained as a double on the staff of the Académie Royale de Musique during 1829 and 1830.27 He then moved to the Théâtre des Arts in Rouen, where he sang until the opera house closed in August 1831; he does not figure on the staff lists when the opera house reformed in October 1831, and it may be assumed that he then began as series of moves that took him to Milan by May 1832 for the premiere of L’elisir d’amore.28 His visit to Italy may be compared with those made by a number of other male artists from the French capital. It is well known that 24 Il censore universale dei Teatri, 23 May 1832, cited in Bini and Commons, Le prime rappresentazioni, 309. 25 un nuovo acquisito, un basso-cantante (Dabadie), appalesò bella voce e non comune agilità, frammezzo ad una temenza, che non seppe mai superare durante il dorso tutto dell’opera. È da lusingarsi che in progresso di tempo, rinascitosi, si meriterà sempre più il favore di quel pubblico, che lo ha incoraggiato con plausi gentili; Gazetta privilegiata di Milano, 14 May 1832 (Gian Jacopo Pezzi) cited ibidem, 305. 26 Justin Dabadie was originally proposed as a replacement for Henri-Etienne Dérivis who retired in 1827; see F-Pan AJ13 119 (letter from Louis-François Sosthène de La Rochefoucauld to Emile-Timothée Lubbert, 6 August 1827). By the following February, it was clear that his debut was to be made later that year, so he accordingly requested leave; see F-Pan AJ13 120 (letter from Justin Dabadie to La Rochefoucauld, 20 February 1828). 27 To take a single example, see the review in Le courrier des théâtres, 26 July 1828: ‘M.  Dabadie jeune continuait ses pâles débuts en récitant le rôle de Cinna, joué d’ordinaire par le ­premier numéro de ces trois unités qui ne forment point de total, et c’était le cas de se soigner, comme on dit dans l’endroit. Les claquetins avaient aussi leur consigne, mais ils n’ont pu que très-faiblement l’exécuter, tant M. Dabadie jeune a été nul et glacé’. Dabadie is listed as double at the Académie Royale de Musique in Almanach des spectacles pour 1829 (Paris: Barba, 1829), 43 and Almanach des spectacles pour 1830 (Paris: Barba, 1830), 42. 28 Jules-Edward Bouteiller, Histoire complète et méthodique des théâtres de Rouen, 4 vols (Rouen: Giroux et Renaux, 1860–1880), 3: 424 and 460.

300  A transalpine comedy Gilbert Duprez, having sung at the Odéon during the 1820s, spent the period up to 1837 in Italy, and that his near contemporary, Paul Baroillhet, left the Conservatoire in 1830 for a decade beyond the Alps. Similarly, Prosper Derivis sojourned in Italy between 1840 and 1845. Adolphe Nourrit’s tragic and terminal Italian period was perhaps less typical. None of these, however, match Dabadie’s stay in Italy; Duprez and Baroillhet effectively began careers in Italy and saw this period as, at least in part, one of training, and Nourrit’s sojourn was sui generis. But Nicolas-Prosper Levasseur left the Théâtre-Italien in 1820 for as short a period as Dabadie over a decade later, in the former’s case for the premiere of his friend Meyerbeer’s opera Margherita d’Anjou at La Scala. He returned to the Théâtre-Italien more or less immediately where he remained until transferring to the Académie Royale de Musique in 1828 for a career that lasted almost a quarter of a century.29 As in the case of Levasseur, there is no indication that Dabadie took part in any other production – either at La Scala or at the Canobbiana in early 1832. With the example of Levasseur in mind, it is entirely reasonable to conclude that when Justin Dabadie made the journey from Rouen to Milan, it was expressly for the premiere of L’elisir d’amore and for no other purpose.30 Almost everything that is known of the genesis of L’elisir d’amore comes from accounts by Federico Albroghetti and Michelangelo Galli (1875) and Emilia Branca (1882). Neither source mentions Dabadie at any length except to confirm his inexperience, and Branca’s account of the selection of the source for the libretto – of course designed to vaunt the fame of her late husband and dating from a half century after the events themselves – excludes any mention of how Romani or Donizetti might have gained access to Scribe’s text: The dialogue between the two gentlemen started with jokes, but turned serious. It is a case of an audacious oddness, and its peculiarity tickles the two men’s stubbornness. They think, they laugh, they argue, finally, they could square the circle. After leafing through several books and libretti in order to choose the plot the two focused on Scribe’s Filtro to imitate, and the imitation was such a good one that it was so far, so far better than the original. It was an effort of genius!31 29 Summary biographies of all the figures discussed in this paragraph are in Karl-Josef Kutsch and Leo Riemens, Großes Sängerlexikon, 4 vols (Bern and Stuttgart: Francke, 1987–1994; 4th edn, 7 vols, Munich: Saur, 2003), passim. 30 Dabadie jeune had returned to Rouen in time for the local premiere of Auber’s Gustave III on 25 January 1834. See Joann Elart, Catalogue des fonds musicaux conservés en Haute-Normandie: Tome 1: Bibliothèque municipale de Rouen; volume 1, Patrimoine musical régional (Rouen: Publications de l’Université de Rouen, 2004), 154 (erroneous references [ibidem 384] to Henri-Bernard Dabadie refer to his brother). 31 ‘Il colloquio incominciato fra le celie si fa serio, si tratta di una bizzarria audace, e la stranezza del caso solletica il puntiglio dei due valentuomi. Si pensa, si ride, si discute; infine la quadratura del circolo strano fu trovata. Dopo sfogliazato libri et libretti per scegliere l’argomento, i due artisti si fermarono sul FILTRO di Scribe, da imitare, e l’imitazione riusci si felice da lasciare di molto, ma di molto, indietro l’originale. Fu uno sforzo d’ingegno!’ (Emilia Branca, Felice Romani ed i più riputati maestri di musice del suo tempo:

A transalpine comedy  301 The description implies a choice from a wide range of material. The nature of the text, closer to the hagiographical than to the documentary, leaves more questions open than answered, however. Unlike his brother, Henri-Bernard, Justin never sang in Auber’s Le philtre, or at least not before the premiere of L’elisir d’amore, but he must have known the work well since his brother sang the role, and Justin was a member of staff at the Académie Royale de Musique until just before its premiere. Furthermore, Le philtre was going into rehearsal in Rouen just as he left, so it is entirely possible – as the brother of one of the artists at the Parisian premiere – that he was involved in the early stages of the Rouen production: the premiere there was on 16 January 1832.32 Nevertheless, as a medium by which the libretto of Le philtre travelled so quickly from Paris to Milan, Justin – and not his brother – remains the prime candidate. The transfer of the libretto of Le philtre to Milan in early 1832 might be thought to be complicated by the first publication of Italian translations of Scribe’s works the same year. None of this publication project affects Le philtre, however, since Luigi Stella’s translation includes two opéras comiques only – Léocadie and La dame blanche – and they appear in volumes 7 and 11 of the publication, published in 1833 and 1834, respectively; furthermore, the lyric items in the two works are paraphrased in prose with the result that any sense of Scribe’s musical decisions – where musical numbers appear, as outlined at the beginning of this article – is entirely effaced.33

Scribe and Romani The evidence presented so far strongly suggests that the libretto of L’elisir d’amore deserves a series of readings in two related contexts: those of the conventions of opera buffa after Rossini and of the conventions of petit opéra; it is clear, furthermore, that the formal principles of petit opéra are greatly indebted to what Francesco Izzo has called – following Powers – La comica forma.34 The overall dimensions of Le philtre and L’elisir d’amore are analogous (Table 8.3).35

32 33 34 35

cenni biografici ed aneddotici (Turin, Florence and Rome: Ermanno Loescher, 1882), 218). See also Federico Alborghetti and Michelangelo Galli, Gaetano Donizetti e G. S. Mayr. Notizie e documenti (Bergamo: Guffuri e Gatti, 1875), 77–78. Elart, Catalogue, 272. Luigi Stella (trans.), Eugène Scribe: Teatro, tradotto dal francese, 12 vols (Bologna: Genio, 1832–1834). Francesco Izzo, ‘Donizetti’s Don Pasquale and the Conventions of Mid-Nineteenth-­ Century Opera Buffa’, Studi musicali 33 (2004), 387–431. There is a critical edition of neither L’elisir d’amore nor Le philtre. Table 8.2 and subsequent direct reference to the score are based on Mario Parenti (ed.), Gaetano Donizetti: L’elisir d’amore – Opera completa per canto e pianoforte (Milan: Ricordi, 1971) and LE / PHLTRE / Opéra en deux actes, / Paroles / de Mr E. Scribe, / Musique de / D.F.E. AUBER / PARTITIONS PIANO ET CHANT / … / PARIS, BRANDUS ET CIE Editeurs, / Success. de Mce SCHLESINGER et d TROUPENAS et Cie / Rue Richelieu, 87 et Rue Vivienne, 40.

Table 8.3  Comparison of Le philtre and L’elisir d’amore Auber, Le philtre

Donizetti, l’elisir d’amore

Ouverture ACTE I 1

2

3 4 5 6

7

ATTO PRIMO

Introduction inc. short section (p. 17) for Guillaume

‘Amis, amis, sous cet épais feuillage’ ‘La voilà Quelle est jolie’

Ballade (Térézine) Récit (Gullaume) Marche et air (Jolicœur)

‘La reine Yseult a blanches mains’ ‘Ah qu’un philtre’ ‘Je suis sergent’

Récit (Térézine, Guillaume, Jolicœur) Air (Térézine)

‘Je suis fière’ including a reprise of the opening chorus of the Introduction and recitative (Guillaume, Térézine) ‘Un mot, Térézine’ ‘La coquetterie fait mon seul bonheur’

Récit (Guillaume, Jeanette) Chorus Récit Air (Fontanarose) Récit Air (Guillaume) Récit

‘Guéris toi, me dit-elle’ ‘Est-il possible d’être insensible’ ‘Quel bruit soudain’ ‘Quel brillant équipage’ ‘Puisque, pour nous guérir’ ‘Philtre divin’ ‘Quel délire nouveau’

[Récit +] Duo (Térézine, Guillaume)

‘C’est Guillaume, allons du courage’ – ‘Non, il reste’

1

Introduzione ‘Bel conforto al mietitore’ Cavatina (Nemorino) ‘Quanto è bella, quanto è cara!’ [Reprise of opening chorus] Cavatina (Adina) ‘Della crudele Isotta’ [Marcia e] Cavatina (Belcore) ‘Come Paride vezzoso’ [Recitativo] dopo li introduzione (Belcore, Adina) ‘Intanto, o mia ragazza’

2

Scena e Duetto (Nemorino, Adina) ‘Una parola, Adina’ – ‘Chiedi all’aura lusinghiera’

3

Coro ‘Che vuol dire cotesta suonata?’

4

Cavatina (Dulcamara) ‘Udite! Udite! O rustici’

5

Scena e duetto (Nemorino, Dulcamara) ‘Ardir! Ha forse il cielo’ – ‘Voglio dire’ Scena e duetto (Nemorino, Adina) ‘Caro elisir!’ – ‘Esulti pur la barbara’

6

8

9

attaca [Récit +] Trio (Térézine, Guillaume, Jolicœur) et Finale

‘Que vois-je, et pour quelle joie’ – ‘Dedans le cours de mes conquestes’

7

ACTE II Entracte et chœur, couplets, barcarolle (Chorus, Jeannette, Térézine, Fontanarose)

‘Chantons, chantons’ – ‘Habitans des bords de l’Ardour’ – ‘Je suis riche et vous êtes belle’

9/10

10 Récitatif et chœur

‘O doux aspect’ – ‘Chantons, chantons’

Récit 11 [Récit +] Duo (Guillaume, Jolicœur) Récit (Jolicœur, Guillaume) 12 Morceau d’ensemble (Térézine, Jeannette, Guillaume, Jolicœur, Fontanarose, Chorus)

‘Voici le soir, l’heure s’avance’ ‘De désespoir je reste anéanti’ – ‘Si l’honneur a pour toi des charmes’ ‘Signe our bien fais ta croix’ ‘Grand Dieux, quelle nouvelle’

Récit (Térézine, Fontanarose)

13 [Récit +] Duo (Guillaume, Térézine) Récit 14 Finale (Chorus)

‘Comme il a l’air heureux’

‘Effet miraculeux’ – ‘Je voulais partir pour la guerre’ ‘De mon art ce sont là les effets ordinaires’ ‘Je lui dois ma maîtresse’

8

11

attacca Terzetto (Adina, Nemorino, Belcore) ‘In guerra ed in amore’ Quartetto e stretta del finale I (Adina, Nemorino, Belcore, Gianneetta) ‘Adina, credimi’ ATTO SECONDO Introduzione e Barcarola a due voci   9: Introduzione ‘Cantiamo’   10: Barcarola a due voci (Adina, Dulcamara) ‘Io son ricco’ Recitativo [e coro] (Belcore, Dulcamara, Adina) ‘Silenzio! È qua il notaro’ + reprise of chorus ‘Cantiamo’ Scena e duetto (Nemorino, Belcore) ‘La donna è un animale’ – ‘Venti scudi’

12/13 Coro e quartetto Coro ‘Saria possibile?’ Quartetto (Nemorino, Giannetta, Adina, Dulcamara) ‘Dell’elisir mirabile’ 14 Recitativo (Adina, Dulcamara) ‘Come sen va contento’ e duetto (Adina, Dulcamara) ‘Quanto amore!’ 15 Romanza (Nemorino) ‘Una furtiva lagrima’ 16 Recitativo (Nemorino, Adina) ‘Eccola… Oh! Qual le accresce’ Aria (Adina) ‘Prendi, per me sei libero’ Recitativo (Belcore et al) ‘Alto! Fronte!’ 17 Aria finale (Dulcamara) ‘Ei corregge ogni diffetto’

304  A transalpine comedy The four main characters in both operas are those of the commedia dell’arte: the two lovers, the blustering soldier and the farcical charlatan, matched to their traditional voice-types – soprano, tenor, basso cantante and basso buffo.36 Each opera consists of two acts with around ten scenes in each act. This overall structure in Le philtre supports fourteen musical compositions, and both works have the same number of compositions in the first act. Much has been made in Donizetti scholarship of the interpolation of ‘Una furtiva lagrima’ into the second act of L’elisir d’amore, but it has to be recognised as a simple transposition of Guillaume’s act i aria ‘Philtre divin’ to the second act; a duet replaces it in act i.37 This lengthens the second act, and a further recitative and duet for Adina and Dulcamara, no. 14 in the table, make this even more marked. The disparity in length of the two second acts returns the discussion to the generic function of petit opéra: keeping the works’ dimensions down to suitable proportions was critical since the evening’s entertainment at the Paris Opéra was also to include a ballet. At the Canobbiana the following year, L’elisir d’amore occupied the entire evening, and Romani and Donizetti had significantly more scope for the inclusion of such numbers as ‘Una furtiva lacrima’ that would have so great a role in the subsequent reception of the work. In general, and in keeping with the traditions of la comica forma, Romani and Donizetti prefer duets and ensembles to arias.38 While in Le philtre there are three arias in the first act and one in the second, in L’elisir d’amore this is greatly reduced. The two so-called ‘arias’ at the end of Donizetti’s second act are problematic: the aria finale has much more in common with a vaudeville, and Adina’s ‘Prendi, per me sei libero’ has a significant role for Nemorino – that goes well beyond the use of pertichini – not only in its tempo di mezzo but also in its cantabile and cabaletta; indeed the end of the composition could well be mistaken for the end of a duet cabaletta. This leaves only ‘Una furtiva lagrima’ – a romanza in any case – and Dulcamara’s show-stopping first aria whose omission would have been dramatic suicide on Romani’s and Donizetti’s part.

Auber and Donizetti If the similarities between Scribe’s and Romani’s libretti are clear, and the means by which the latter encountered the former plausible, the extent to which Auber’s musical innovations in Le philtre were carried over into L’elisir d’amore is difficult to judge. A remarkable feature of Le philtre is the massive level of musical reminiscence. Some of this is carried over into 36 Mario Rinaldi: Felice Romani: dal melodramma classico al melodramma romantico (Rome: Edizioni de Santis, 1965), 287. 37 Alessandro Roccatagliati, Felice Romani librettista, Quaderni di Musica / Realtà 37 (Lucca: Libreria musicale italiana, 1996), 209, note 7. 38 Izzo, ‘Donizetti’s Don Pasquale’, 402–403 and 406.

A transalpine comedy  305 L’elisir d’amore; the brief recall of the barcarolla in Dulcamara’s reappearance in recitative is directly modelled on an identical practice in the libretto of Le philtre,39 but its reappearance as the basis of the vaudeville at the end of the opera is an invention probably of Donizetti’s alone, and that did not depend on Romani’s intervention.40 However, Auber’s act ii finale is based on the chorus that opens Fontanarose’s act i aria; both operas, then, end with a reminiscence of the music associated with the charlatan: Auber chooses Fontanarose’s first appearance, whereas Donizetti focuses on Dulcamara’s involvement with the celebrations that open the second act.41 It is entirely plausible that Donizetti’s decision was prompted by Dabadie’s description of Auber’s opera and the reminiscence contained therein. What does not appear in L’elisir d’amore is the high level of reminiscence in the trio et final that ends the first act of Le philtre; here much of the musical material is shared with the overture, and the stretta to the finale is written in such a way – in the same key, tempo and rhythm – as to make possible an extraordinary metamorphosis of the music into the chorus that opened the introduction.42 This is omitted by Donizetti since it would have collided too much with the conventions of opera buffa and was more part of a project that was occupying Parisian composers as grand opéra was being developed that was concerned with the repetition of choruses to impart some degree of cohesion to long first acts; the locus classicus here is the first act of Les Huguenots, and the project has no impact on L’elisir d’amore. It furthermore gives a glimpse of some reasons why Romani and Donizetti elided Scribe and Auber’s first two numbers into a single introduzione whose length has attracted a good deal of commentary.43 Almost every single composition in L’elisir d’amore benefits from being read either against its opposite number in Le philtre, more broadly against the intersection of the conventions of opera buffa and petit opéra, or as part of a complex network of relationships governed by an idea of cultural transfer. A possible context for ‘Una furtiva lagrima’ and many of the problems 39 Parenti (ed.), Gaetano Donizetti: L’elisir d’amore, 162; LE / PHILTRE, 175. 40 The barcarolla is Parenti (ed.), Gaetano Donizetti: L’elisir d’amore, 156–160 and its reprise, ibidem, 247–252. The nature of the barcarolla has proved problematic since it exploits none of the conventional features of the type; see Pierluigi Petrobelli, ‘Dulcamara e Berta: storia di una canzone’, Liedstudien: Wolfgang Osthoff zum 60. Geburtstag, eds. Martin Just and Reinhard Wiesend (Tutzing: Schneider, 1989), 308. The appearance of a perfectly conventional barcarolle in Auber’s Le philtre as the basis for Romani’s and Donizetti’s duo – had they known enough of the work to recognise it as conventional barcarolle – might well have been the trigger for avoiding a simple barcarolla and possible comparisons, and the reason for selecting musical topics that, as Petrobelli shows (ibidem, 310–312), had more in common with Berta’s ‘Il vechiotto cerca moglie’ from Rossini’s Il barbiere di Siviglia. 41 LE / PHILTRE, 58–59 and 230–234. 42 Ibidem, 3 and 99–100; 12–18 and 146–155. 43 The poetry of Romani’s introduzione runs to 100 lines in comparison with the 30, for example, of Il pirata (Roccatagliati, Felice Romani librettista, 219–220).

306  A transalpine comedy around the act ii barcarolla may be explained by reference to Le philtre. Both the act i introductions and finales may be read similarly, and arias in Le philtre, which are turned into duets in L’elisir d’amore, make sense when considered bifocally from the standpoint of petit opéra and opera buffa. There is a single number that is worthy of attention in the context of this discussion of the transmission of the libretto from Paris to Milan, since it is closely related to the musical interests of the family responsible for the transfer. This is the duet for Belcore and Nemorino in L’elisir d’amore – the equivalent to Jolicœur and Guillaume in Le philtre – in the second act of both works; in both it is no. 11.44 The beginning of the second act proceeds almost identically in both works, with introductory choruses and the famous barcarolle for Dulcamara and Adina. This is followed by recitative in both operas, in which the notary arrives, and in which Adina expresses her regret that Nemorino is missing, thus thwarting her vengeance; the recitative is followed by a reprise of the opening chorus in both works. The following recitative between Nemorino and Dulcamara, which includes – as already said, in both operas – a reminiscence of the barcarolla, is critical for explaining Nemorino’s need for money to buy a second dose of Dulcamara’s elixir in order for it to work fast enough to ensnare Adina before her marriage to Belcore. Nemorino despairs and Dulcamara leaves. The stage is set for the encounter between Belcore and Nemorino. Scribe and Auber set the duo as a single-tempo movement preceded by a short accompanied recitative. Guillaume’s agreement to enlist is divided into two separate moments: one within the body of the duo, where he agrees to allow Jolicœur to prepare the paper for signature, and the second, where Guillaume signs in the short and generically obligatory recitative at the end of the duo. Romani and Donizetti’s duet is more ambitious. As in Le philtre, it begins with a short scena whose poetry does little more than translate Scribe’s text. But Romani’s libretto is clearly designed for a four-movement duet, and that is what Donizetti creates. As so often in buffo duets, the slow movement is characterised by rapid parlante writing for Belcore, and Nemorino’s single signature is the kinetic moment in the tempo di mezzo which triggers the cabaletta.45 The musical structures of the two duets are correspondingly different, and an instance of where Romani and Donizetti put a higher price on the conventions of opera buffa than on the speed of production that could have been achieved by continuing to imitate the poetic and musical moves of Le philtre. But in these very different duets is embedded music that is very similar and which invites comment on its relationship with the frères Dabadie. Example 1 is the beginning of the duet in Le philtre (Example 8.1).

4 4 Parenti (ed.), Gaetano Donizetti: L’elisir d’amore, 165–178; LE / PHILTRE, 178–189. 45 Izzo, ‘Donizetti’s Don Pasquale’, 408.

Example 8.1 Auber, Le philtre, no. 11: [Récit +] Duo (Guillaume, Jolicœur) ‘De désespoir je reste anéanti’ – ‘Si l’honneur a pour toi des charmes’, opening.

308  A transalpine comedy Critical points here are the Italianate martial dotted rhythms in the vocal part; the descending fifth from the sixth to the second degree of the scale on the first and second beats of the second bar; and the string accompaniment of chords on the dotted quaver and single notes on the semiquaver. This example may be compared with the beginning of Donizetti’s cabaletta (Example 8.2). The vocal line not only exploits almost exactly the same dotted rhythms found in Le philtre but also deploys the same descending fifth from the ­degree to the second in an almost identical rhythmic configuration. These  are

Example 8.2  Donizetti, L’elisir d’amore, no. 11: Scena e duetto (Nemorino, Belcore) ‘La donna è un animale’ – ‘Venti scudi’, opening.

A transalpine comedy  309 marked with an upper-case ‘A’ in both Examples 8.1 and 8.2. The beginning of the cabaletta also uses accompaniment figures almost identical to those in the duet from Le philtre. In Donizetti’s cabaletta, Belcore’s stanza exploits exactly the same range of musical topics found in the duet from Le ­philtre, but Nemorino’s stanza lurches into the tonic minor with impassioned legato phrases to which Belcore is finally joined. Although the repeat of the cabaletta is massively truncated, the only part that is repeated exactly is the opening music that has so much in common with the Auber passage. Donizetti anchors his work in Le philtre even when Romani’s libretto is at its greatest distance from Scribe and Auber’s original; he can only have done this with some account of the music of Le philtre, and the most likely source for that information was Justin Dabadie. And that Justin should encourage Donizetti to write music that so much resembled the highly Italianate music that Auber had written for his older brother is entirely congruent with the parallel – but imbalanced – careers that the two had in Paris before Justin’s departure for Italy. *** The complex network of Franco-Italian relationships that underpins the history of L’elisir d’amore encompasses dramatic, poetic and musical materials deployed by some of the best-known composers, librettists and singers of the period. It has been complicated by the continuing presence of errors of fact in scholarly and more popular literature on the subject: the erroneous involvement of Stendhal and the confusion of the two Dabadie brothers have made criticism of the cultural transfer from Italy to France and back again harder to analyse. The modern unchallenged preference for L’elisir d’amore over Le philtre has also been a barrier to understanding the relationship between the two works. The generic context of Scribe and Auber’s Le philtre demands a place in the network of practices that underpins L’elisir d’amore, a context that invokes Le comte Ory and its Italian origins. And the more general context of the French opera libretto as a source for Italian opera libretti is a further more elusive concern. In such a discussion, oppositions between import and export, foreign and native and even centre and periphery become difficult to sustain. And alternative views that see the process of cultural transfer as a series of interconnected networks do greater justice to such works as L’elisir d’amore and Le philtre; they both emerge as complex products of a wide range of French and Italian works, composers, librettists and singers, and almost every number in both operas responds to a reading that accepts its position in a modern understanding of cultural transfer and in the ­pan-European complexities of melodramma giocoso and petit opéra.

9 Partners in rhyme Alphonse Royer, Gustave Vaëz, and foreign opera in Paris during the July Monarchy

Starting in 1839, the translation of foreign works at the Académie Royale de Musique took on a different aspect, with a single team devoting nearly a decade to translating foreign opera for the French stage: Alphonse Royer and Gustave Vaëz. Alongside Scribe’s Dom Sébastien (the French version of Donizetti’s Poliuto, 1840) and the Pacini-Berlioz arrangement of Der ­Freischütz (1841), Royer and Vaëz produced translations of Donizetti’s ­Lucia di Lammermoor and Don Pasquale, Verdi’s I Lombardi alla prima crociata and Rossini’s Otello. They also prepared the libretto for the ­Rossini ­pasticcio Robert le Bruce (1847), for Donizetti’s L’ange de Nisida and its ­better-known revision, La favorite (1840). Royer and Vaëz’s activity has to be seen in the context of their work – in mutual collaboration and in association with other colleagues – for almost every theatre in Paris and in almost every genre. Their musical interests led them to write libretti for ­Boisselot and Gevaert, with Vaëz writing the libretto for Donizetti’s Rita ou Le mari battu and Royer translating Flotow’s Stradella for Brussels. Both wrote the libretto for the collaborative Les premiers pas that opened the Opéra ­National in 1847. Despite the close relationship between Royer and Vaëz, almost every production was subject to slightly different artistic pressures, and the balance between differing circumstance and a continuing collaboration marked out the products of their work between 1839 and 1847. The epicentre of European stage music, Paris’ Académie Royale de ­Musique, was not immune to the tradition of naturalising both German and Italian opera. The beginning of the century saw adaptations there of ­Mozart’s Die Zauberflöte as Les mystères d’Isis (1801), of Don Giovanni (1805) and of Winter’s Il trionfo dell’amor fraterno as Castor et Pollux (1806), an ­international soundtrack to Napoleonic triumphs all over Europe.1 1 All productions at the Académie Royale de Musique up to 1876 are given a summary listing in Théodore de Lajarte, Bibliothèque musicale du Théâtre de l’Opéra: catalogue historique, chronologique, anecdotique, 2 vols (Paris: Bibliothèque Nationale, 1878; R Hildesheim: Olms, 1969). For Les mystères d’Isis, see Jean Mongrédien, ‘Les Mystères d’Isis (1801) and Reflections on Mozart from the Parisian Press at the Beginning of the Nineteenth Century’, Music in the Classic Period: Essays in Honor of Barry S. Brook, ed. Allan W. A ­ tlas (New York: Pendragon, 1985), 195–211; Rudolph Angermüller, ‘“Les Mystères d’Isis”

Partners in rhyme  311 A ­second phase of adaptation involved Rossini’s assimilation of French operatic practices during the 1820s, with reworkings of Maometto II, Mosè in Egitto and (partially) Il viaggio a Reims,2 and this was followed by a production of Weber’s Euryanthe in 1831, an epoch-making version of Don Giovanni in 1834 and Berlioz’s reworking of Der Freischütz in 1841.3 These artistic undertakings were all products of different teams of librettists and composers. Although Rossini clearly naturalized the music of Maometto II and Mosè in Egitto himself, he worked with different combinations of librettists: Luigi Balocchi and Alexandre Soumet for the former, and Balocchi, this time with Victor-Joseph-Étienne de Jouy, for the latter. For the partial reworking of Il viaggio a Reims as Le comte Ory, Rossini worked alongside Eugène Scribe and Charles-Gaspard Delestre-Poirson. The production of Euryanthe was from the veteran Castil-Blaze (both the translation of the text and the adaptation of the music), and of Don Giovanni from a team than involved both Castil-Blaze and his son Ange-Henri Blaze (Blaze de Bury) as well as Émile Deschamps. When Berlioz naturalised Der Freischütz, the libretto was translated by Emilien Pacini.

(1801) und “Don Juan” (1805, 1834) auf der Bühne der Pariser Oper’, Mozart-Jahrbuch 1980– 83 des Zentralinstitutes für Mozartforschung der Internationalen Stiftung Mozarteum Salzburg (Kassel, Basel and London: Bärenreiter, 1983), 32–97. In addition to Angermüller’s article, for the 1805 Don Juan, see Laurent Marty, 1805: la création de Don Juan à l’Opéra de Paris, Univers musical (Paris: L’Harmattan, 2005). 2 Paolo Isotta, ‘Da Mosè a Moïse’, Bolletino del Centro Rossiniano di Studi 1–3 (1971), 87–117; idem (ed.), Gioacchino Rossini: Mosè in Egitto, Azione tragico-sacra; Moïse et Pharaon, Opéra en quatre actes; Mosè, Melodramma sacro in quattro atti, Opera 1a, 4 (Turin: Unione tipografico-editrice torinese, 1974); Marcello Conati, ‘Between Past and Future: The Dramatic World of Rossini in Mosè in Egitto and Moïse et Pharaon’, 19th Century Music 4 (1980–81), 32–47; Richard Osborne, Rossini, The Master Musicians (London: Macmillan, 1986), 208–213 and 237–242; Giuseppe Ierolli, ‘Mosè e Maometto: da Napoli a Parigi’ (Tesi di laurea, Università degli studi di Bologna, 1989–90). For Le siège de Corinthe, see Anselm Gerhard, Die Verstädterung der Oper: Paris und das Musiktheater des 19. Jahrhundert (Stuttgart: Metzler, 1992) trans. Mary Whittall as The Urbanization of Opera: Music Theatre in Paris in the Nineteenth Century (Chicago and London: University of Chicago Press, 1998), 68–84 [page numbers refer to English translation]. Janet Johnson, ‘A Lost Rossini Opera Recovered: Il Viaggio a Reims’, Bollettino del Centro Rossiniano di studi n.v. Nos 1–3 (1983), 5–57; eadem (ed.), Il viaggio a Reims, ossia L’albergo del giglio d’oro: dramma giocoso in un atto di Luigi Balocchi, music di Gioachino Rossini, 2 vols, Edizione critica delle opere di Gioachino Rossini 35 (Pesaro: Fondazione Rossini, 1999). 3 Mark Everist, ‘Translating Weber’s Euryanthe: German Romanticism at the Dawn of French Grand Opéra’, Revue de Musicologie 87 (2001), 67–105; Sabine Henze-Döhring, ‘E.T.A. ­Hoffmann-“Kult” und “Don Giovanni”-Rezeption im Paris des 19. Jahrhunderts: Castil-­ Blazes “Don Juan” im Théâtre de l’Académie Royale de Musique [sic] am 10 März 1834’, ­Mozart-Jahrbuch 1984/5 des Zentralinstitutes für Mozartforschung der Internationalen Stiftung Mozarteum Salzburg (Kassel, Basel and London: Bärenreiter, 1986), 39–51; ­Katharine Ellis, ‘Rewriting Don Giovanni, or “The Thieving Magpies”’, Journal of the Royal Musical Association 119 (1994), 212–250. The best introduction to Berlioz’s naturalisation of Der Freischütz is in Ian Rumbold (ed.), Arrangements of Works by Other Composers (II), ­Hector Berlioz: New Edition of the Complete Works 22b (Kassel etc.: Bärenreiter, 2004), xii–xiv.

312  Partners in rhyme Table 9.1  Royer and Vaëz, French adaptations of Italian operas Title

Paris premiere

Lucie de 10 August Lammermoor 1839 Othello 2 September 1844

Musical collaborators

Original work(s)

Premiere

Donizetti

Lucia di Lammermoor

Benoist

Otello [L’italiana in Alger; La donna del lago] Lucia di Lammermoor

26 September 1835; Naples, San Carlo 4 December 1816; Naples, Fondo

Lucie de 20 February Donizetti 26 September 1835; Lammermoor 1846 Naples, San Carlo Robert Bruce 30 December Rossini; Zelmira;, La donna del 16 Feb 1822; Naples, 1846 Niedermeyer lago [Bianca e Faliero; San Carlo [Zelmira]; Torvaldo e Dorliska; 24 Oct 1819; Naples, Armida; Mosè in San Carlo [La donna Egitto; Maometto II] del lago]; etc. Jérusalem 26 November Verdi I Lombardi alla prima 11 February 1843; 1847 crociata Milan, Scala

During the second half of the July Monarchy, however, two librettists – Alphonse Royer and Gustave Vaëz (a pseudonym of Jean Nicolas Gustave van Nieuwenhuysen) – assumed a near monopoly position in the naturalisation of Italian opera at the Académie Royale de Musique (Table 9.1). They were responsible for the four main productions of Italian opera in French at the Académie Royale de Musique from the late 1830s until the Revolution of 1848, with key works by Rossini (Otello), Donizetti (Lucia di Lammermoor) and Verdi (I Lombardi alla prima crociata); they were also responsible for the Rossini pasticcio Robert Bruce.4 Although all were ultimately destined for the Académie Royale de Musique, Lucie de Lammermoor had started out at the Théâtre de la Renaissance in 1839 and moved to Académie Royale de Musique in 1846. In addition to translations, Royer and Vaëz’s most enduring triumph was their libretto to Donizetti’s L’ange de Nisida which metamorphosed into La Favorite.5 Together, they were also the translators of Donizetti’s Don Pasquale for Brussels the same year as its Parisian premiere at the Théâtre-Italien, and the authors of an occasional work, Les premiers pas, for the opening of Adam’s ill-fated Opéra-National 4 The exception to Royer and Vaëz’s near monopoly was the translation and arrangement of Donizetti’s Poliuto as Les martyrs (1840), for which Scribe worked on the translation of the libretto. 5 For L’ange de Nisida, see Everist, ‘Donizetti and Wagner’, 335–338, Rebecca Harris-­ Warwick, ‘The Parisian Sources of Donizetti’s French Operas: The Case of La Favorite’, L’opere teatrale di Gaetano Donizetti: atti del convegno internazionale di studio, Bergamo, 17–20 settembre 1992, ed. Francesco Bellotto (Bergamo: Commune di Bergamo, 1993), 77–92, and eadem (ed.), Gaetano Donizetti: La Favorite, 2 vols [paginated consecutively], Edizione critica delle opere di Gaetano Donizetti (Milan: Ricordi, 1997), 1:xiii–xix. For the reasons why Scribe’s name was frequently, but much later in the century, added to the list of librettists for La favorite, see ibidem, 1:xxi–xxii.

Partners in rhyme  313 in 1847. Posterity has dealt unevenly with Royer and Vaëz’s adaptations of Italian opera. Jérusalem has been exhaustively discussed, as has Lucie de Lammermoor.6 A recent recording of Robert Bruce gave it a degree of life, but the work had been treated harshly in older literature.7 The translation of Othello, critical for the reception of Italian opera in French, has been hardly discussed in modern times.8 Both Royer and Vaëz had careers in the theatre before turning their hand to music drama. Royer had been closely involved in liberal and romantic circles in the 1820s and had made the seemingly obligatory tour to the East. His early career as a playwright brought him success at the Nouveautés, the Gaîté and the Porte Saint-Martin, alongside the publication of novels. Vaëz, on the other hand, was Bruxellois, and started his theatrical career in the city, but his move to Paris was marked by success at the Gaîté, the ­Gymnase-Dramatique and the Théâtre du Vaudeville.9 Table 9.2 gives a full list of Royer and Vaëz’s collaborations, both within the domain of music drama and in other dramatic genres.

6 Jérusalem has been extensively discussed: David R. B. Kimbell, ‘Verdi’s First R ­ ifacimento: “Il Lombardi” and “Jérusalem”’, Music & Letters 60 (1979), 1–36; Julian Budden, The Operas of Verdi, 3 vols (Oxford: Oxford University Press, 1978–81), 1:339–359; Ursula Günther, ‘Giuseppe Verdis erster Erfolg in Paris’, Lendemains 31–32 8 (1983), 53–62; ­A rrigo Quattrocchi, ‘Da Milano a Parigi: Jérusalem, la prima revisione di Verdi’, Studi verdiani 10 (1994–5), 13–60. It should be noted that enthusiasm for Jérusalem only really dated from the 1980s; see the critique of earlier views in Budden, The Operas of Verdi, 1:342. For Lucie de Lammermoor, see Everist, ‘Donizetti and Wagner’, 323–329. 7 Gioachino Rossini: Robert Bruce, dir Paolo Arrivabene (Dynamic, CDS 421/1–2, 2002); the review by William Ashbrook, ‘Review of Gioachino Rossini: Robert Bruce’, Opera Quarterly 20 (2004), 330–331 is useful. But Jacques-Gabriel Prod’homme, ‘Rossini and His Works in France’, Musical Quarterly 17 (1931), 135, only mentions the work in passing, and despite an account in Jerome Mitchell, The Walter Scott Operas (University, Alabama: University of Alabama Press, 1977), 350–357, this was dismissed in a review by Winton Dean (‘Review of The Walter Scott Operas’, Music & Letters 59 (1978), 460). Félix Clément and Pierre Larousse, Dictionnaire des opéras (Dictionnaire lyrique), ed. Arthur Pougin (Paris: Larousse, [1897]), 958 is a more sympathetic view of the work, originally published in 1869. 8 See the purely factual statements based on Lajarte in Prod’homme, ‘Rossini and his Works in France’, 135. For a fuller account of the circumstances that led to the production of Othello and Robert Bruce, as well as the detail of the libretti, music and their sources, see Chapter 7. 9 Reconstructing biographies for the two individuals is problematic. Neither yet appears in the Dictionnaire de biographie française, eds. Michel-Prévost and Jean-Charles Roman d’Amat (Paris: Letouzey et Ané, 1933– [in progress]). The best accounts of both lives are in Gustave Vapereau, Dictionnaire universel des contemporains contenant toutes les personnes notables de la France et des pays étrangers, 5th edn. (Paris: Hachette, 1880), 1588 and 1760, and in Dantès [pseud. of Charles-Victoire-Alfred Langue], Dictionnaire biographique et bibliographique alphabétique et méthodique des hommes les plus remarquables dans les lettres, les sciences et les arts, chez tous les peuples, à toutes les époques (Paris: Boyer, 1875), 881 and 1009. Royer alone is given a partial entry in François-Joseph Fétis, Biographie universelle des musiciens et bibliographie générale de la musique, 8 vols (2nd edn. [with supplement in two vols] Paris: Firmin Didot, 1860–1865), suppl. 2:437.

Table 9.2  Theatrical and literary works by Royer and/or Vaëz (locations of premieres are given in parentheses for theatrical works; place and date of publication appear in parentheses for literary works; names in square brackets are explanations of pseudonyms; ‘with’ indicates a collaborator) Date Royer and Vaëz (Music Drama)

Royer and Vaëz (Other)

1830 1834 1836 1837 1838 1839

Lucie de Lammermoor

1840

La favorite

1841 1842 1843

Don Pasquale

Le voyage à Pontoise (Odéon); Le bourgeois grand seigneur (Odéon) Mademoiselle Rose (Odéon)

Royer

Vaëz

Les mauvais garçons, 2 vols (Paris: Renduel, 1830) with Auguste Barbier; Henri V et ses compagnons (Nouveautés) Manoel (Paris: Ledoux, 1834); Un divan Le cheval de Grammont (Brussels, (Paris: Ledoux, 1834) Théâtre Royal) L’auberge des trois pins (Paris: Dumont, La belle écaillère (Gaîté; Gabriel de 1836) with Roger de Beauvoir Lurieu; Théaulon de Lambert); Il signor Barilli (Gaîté) Petit Pierre (Gaîté; Auguste Jouhaud) Lellou (Saint-Antoine; August Jouhaud); Le connêtable de Bourbon (Paris: Werdet, 1838) Timoléon le fashionable (Porte SaintLes brodequins de Lise (GymnaseAntoine; Auguste Jouhaud); Le camp de Dramatique; Laurencin [P-A Fontainebleau (Fontainebleau; Auguste Chapelle]; Desvergers [A Chapeau]); Jouhaud); Les beaux hommes de Paris Le coffre-fort (Vaudeville) (Panthéon; Auguste Jouhaud) Robert Macaire en Orient (Paris: Dumont, 1840); Mademoiselle Beata (Paris: Dumont, 1840) Rita, ou Le mari battu (OpéraComique; Donizetti†; 1860) Mon parrain de Pontoise (Théâtre du Palais-Royal)

(Continued)

Date Royer and Vaëz (Music Drama) 1844 1846 1847 1847

1849 1850

1851 1852

1854 1855 1859 1860 1864 1866

Othello Robert Bruce Jérusalem Les premiers pas (Auber, Halévy, Carafa, Adam; Opéra National)

Royer and Vaëz (Other)

Royer

Vaëz

La comtesse d’Altenberg (Odéon) Les janissaires, 2 vols (Paris; Dumont, 1844)

Ne touchez pas à la reine! (OpéraComique; Scribe, Boisselot); Nouvelles d’Espagne (Odéon) Les bourgeois des métiers (Odéon) Le jour et la nuit (Variétés); Les fantaisies de Milord (Variétés; Charles Narrey); La dame de trefle (Vaudeville; Charles Narrey); Chodruc-Duclos (Gaîté; Michel Duporte); Un ami malheureux (Vaudeville; A Roger)

Le jeu d’amour et de la cravache (Montansier; Anicet, and Charles Narrey)

Mosquita la sorcière (Opéra National; Scribe; Boisselot) Dans une armoire (Foliesdramatiques); Grandeur et décadence de M. Joseph Prudhomme (Odéon; Henry Monnier) Georgette (Théâtre-Lyrique; Gevaert)

Déménagé d’hier (Variétés; Charles Narrey)

Jane Eyre (Bruxelles, Galeries Hubert; Victor Lefèvre) Stradella (Flotow, Brussels, Monnaie; with Gustave Oppelt) Georges Brummell (Bruxelles, Galeries Hubert) Cadet le perle (Gaîté; Théodore de Langeac)

Le capitaine Henriot (Opéra-Comique; Victorien Sardou; Gevaert)

316  Partners in rhyme Table 9.2 also gives Royer and Vaëz’s individual operatic and dramatic contributions: Royer collaborated with Gustave Oppelt on the 1859 translation of Flotow’s Stradella for Brussels, and Vaëz was also the librettist for a number of new works: Donizetti’s Rita, ou Le mari battu, two works for Gevaert and two – in collaboration with Scribe – for Boisselot, including the strikingly successful Ne touchez pas à la reine!. The operatic careers of Royer and Vaëz are marked by paradox. Although they consistently functioned as the ambassadors of Italian opera at the Académie Royale de Musique from around 1840 onwards, the process of naturalising these works encompassed a wide range of practices and approaches to their originals. These could range from their faithful arrangement of Lucia di Lammermoor to the 1847 Robert Bruce, which veered between the type of Rossini pasticcio well known from the 1820s and a naturalisation of La donna del lago. Their revisions of Verdi’s I Lombardi alla prima crociata as Jérusalem and of Rossini’s Otello fall somewhere between these two extremes. Much of the explanation for this wide range of practices may be found in the musical collaborators with whom Royer and Vaëz worked (noted in Table 9.1). Donizetti worked closely with Royer and Vaëz on Lucie de Lammermoor, and Verdi did the same for Jérusalem. Rossini, however, took no part in work on the 1844 Othello, since he was not in Paris. A key player in the Othello production, therefore, was the widely experienced François Benoist. Chef du chant at the Académie Royale de Musique, organist in the royal chapel and professor of organ at the Conservatoire, he was also a veteran ballet composer.10 And since Rossini was in Bologna while preparations were in hand for Robert Bruce, Vaëz himself went there to work on the project together with the musical arranger for the project, Louis Niedermeyer. By the late 1840s, the latter was an experienced actor on the stage of Parisian music drama, with Stradella and Marie Stuart (1837 and 1844, respectively) behind him and the École Niedermeyer and a permanent place in the history of musical pedagogy in the future.11 Each of Royer’s and Vaëz’s four major enterprises for the Académie ­Royale de Musique involved a different type of collaboration with composers. In Lucie de Lammermoor, they worked closely with Donizetti to produce a work that was as close to a simple literary translation and musical arrangement of the Italian original as the changes to the drama permitted, but when working with Verdi on essentially the same terms, they produced almost a completely new libretto for Jérusalem to which Verdi wrote large 10 Fétis, Biographie universelle, 1:347. Benoist’s four ballets at the Académie Royale de Musique extended from 1839 to 1851 and involved collaboration with Ambroise Thomas, Henri Reber and Marco Aurleio Marliani, to scenarios by Théophile Gautier and ­Jules-Henry Vernoy, comte de Saint-George, with choreography by Joseph Mazilier and Arthur Saint-Léon. 11 Ibidem, 6:318–321.

Partners in rhyme  317 quantities of new music. Paradoxically perhaps, their collaboration with Benoist on Othello followed similar lines to those practices employed in Lucie de Lammermoor, while, for Robert Bruce, they wrote an entirely new libretto loosely based on themes from Scott to which Niedermeyer and Rossini had to fit pre-existing music.

Drama Drama was central to the concerns of Royer and Vaëz as they contemplated the naturalisation of dramma – both tragico and lirico – and melodramma. The broader context for their work was the general feeling expressed in the press that, although the Italian language was easier to set and in general worked better for stage music than French, the dramatic structure of Italian originals left much to be desired when measured against the yardstick of established French drama, whether by Hugo, Racine or Scribe. Royer and Vaëz made this explicit when they wrote a dedicatory letter addressed to Donizetti at the beginning of their adaptation of Lucia di Lammermoor and pointed to both general and specific changes that French theatrical and operatic culture demanded: We have selected Lucia di Lammermoor as the most poetic and impassioned work to which your musical genius has given birth, and we have tried to adapt from it a form and words that permit the theatres of our great cities to popularize it in France…. Furthermore, we have again simplified the performance of the drama, avoiding scene changes in the middle of acts that French dramatic forms do not accept willingly; the new scenes that you have composed with us, in order to appropriate this arrangement of the libretto to the requirements of our stage, are for your opera a true naturalization.12 Much of the Parisian press agreed with Royer and Vaëz when they pointed to Lucia di Lammermoor as a good choice for a work of Donizetti to naturalise. Félix Bonnaire, writing for the Revue de Paris, observed that:

12 ‘Nous avons choisi la Lucia di Lammermoor comme l’œuvre la plus poétique et la plus passionnée qu’ait enfantée votre génie musical, et nous avons essayé de lui adapter une forme et des paroles qui permissent aux théâtres de nos grandes villes de la populariser en France.….. Nous avons d’ailleurs simplifié encore la représentation de la pièce, en évitant, au milieu des actes, les changements de décors que la forme dramatique française n’accepte pas volontiers; les scènes nouvelles que vous avez composées avec nous, pour approprier cette imitation du libretto aux exigences de notre théâtre, soient pour votre opéra une véritable naturalisation’ (Guido Zavadini, Donizetti: vita, musiche, epistolario (Bergamo: Istituto Italiano d’arti grafiche, 1948), 882–883, based – with many errors – on Charles Malherbe, Centenaire de Gaetano Donizetti: catalogue bibliographique de la section française à l’exposition de Bergame (Paris: n.p., 1897), 135–136). The dedicatory letter appears only in the version of the libretto put on sale in February 1839 and listed in the Bibliographie de la France, 9 February 1839.

318  Partners in rhyme Among Donizetti’s operas, Lucia was the one that without doubt suited us the best. The numbers in this score, although they are developed with a certain generosity, have nothing to do with the vast dimensions with which ears used to French ariettes reproach Italian music so much. Choruses follow one another quickly, simple motifs abound. If one excepts the finale of the first act or the Ravenswood scene in the second, compositions in a grand style with a high register of expression, it is almost everywhere music of the sort we like at the French Opera.13 By contrast, Otello was deemed a poor choice according to Théophile Gautier in La presse: We believe that the choice of Othello is a poor choice, not because the score does not sparkle with sublime beauty, but in general the work is in that Italian style full of thoughtlessness for the dramatic situation and which worries little if the melody agrees with the sense of the words, provided that the musical phrase is lively, agile, sparkling.14 In other words, those things that marked out a fine Italian libretto were exactly the things that had to be avoided on the French stage. Furthermore, it was claimed in the press that Cammarano’s original libretto had abbreviated and disfigured Scott’s novel on which it was based, and that Royer and Vaëz had gone some way to restoring the authority of the 1819 Bride of Lammermoor. While this hardly seems a fair criticism of the libretto, it is clear that French theatrical culture viewed it with a degree of suspicion and was only too happy to see this position apparently remedied by French paroliers. 15 Royer and Vaëz were similarly praised for having well understood

13 ‘Parmi les opéras de Donizetti, la Lucia était sans contredit celui qui nous convenait le mieux. Les morceaux de cette partition, bien qu’ils se développent avec une certaine ampleur, n’ont rien de ces vastes dimensions que les oreilles habituées aux ariettes françaises reprochent tant à la musique italienne. Les chœurs se succèdent avec rapidité, les motifs faciles abondent. Si l’on excepte le finale du premier acte et la scène de Ravenswood, au second, compositions d’un grand style et d’une très haute expression, c’est un peu partout de la musique comme on l’aime à l’Opéra français’ (Revue de Paris 8 (1839), 146). 14 ‘Nous croyons que le choix d’Othello est un choix malheureux, non que la partition n’étincelle de sublimes beautés, mais l’œuvre, en général, est entendue dans ce style italien plein d’insouciance de la situation et qui s’inquiète peu si la mélodie concorde avec le sens des paroles, pourvu que la phrase soit vive, alerte, étincelante’ (La presse, 9 September 1844). 15 ‘The translation of the libretto, with all due deference to several of our colleagues, must lay claim to a very small part in the success of this drama; one knows how Walter Scott’s admirable novel has been mangled and disfigured by the Italian poet’ (‘La traduction du ­l ibretto, n’en déplaise à plusieurs de nos confrères, doit revendiquer une très faible part dans le succès de ce drame; on sait comment l’admirable roman de Walter Scott a été tronqué et défiguré par le poète italien’ (Le ménestrel, August 1839)).

Partners in rhyme  319 Shakespeare – again in a veiled criticism of Berio di Salsa’s Italian libretto – in their reworking of Otello.16

Language If the dramatic framework of Italian melodramma was intolerable on the stage of the Académie Royale de Musique, its linguistic and rhythmic structure was recognised as ideal for musical setting and a serious challenge for anyone attempting to translate it into French. Indeed, there was no shortage of critics prepared to dilate on the relative merits of the French and Italian ­ ossibly languages and their relationship with music. The anonymous critic – p Gustave Hequet – of L’illustration was explicit on this matter shortly after the 1844 premiere of Othello: ‘By virtue of work and skill, MM Royer and Vaëz have forced our language, so cold and so unmalleable, so constrained by consonants, so loaded with epithets, to enter, without too many cuts or bruises into this narrow and flexible mould of Italian poetry’.17 The almost entirely positive response to Royer and Vaëz’s translations and adaptations of the 1840s provides a key to understanding judgements of value made on these works. To be sure, some terms – élegance, fraîcheur, pureté and even originalité – are too vague to decode with precision at nearly 200 years distance, but others tell us a good deal about the qualities that were sought and praised in these naturalisations of Italian opera.18

16 ‘The Italian libretto of Otello is indeed a pale reflection of Shakespeare’s admirable genius, a thoroughly enfeebled echo of the sublime poetry of the great English author, honored by the entire world; MM Alphonse Royer and Gustave Vaëz have understood him better than anyone; they have also dignified the libretto as much as they have been able’ (‘Le poème italien d’Othello est une reflet bien pâle de l’admirable génie de Shakespeare, un écho bien affaibli de la poésie sublime du grand écrivain de l’Angleterre, dont le monde entier s’­honore; MM. Alphonse Royer et Gustave Vaëz l’ont compris mieux que personne, aussi ont-ils relevé le libretto autant qu’ils l’ont pu’. (Le siècle, 19 September 1844; Hippolyte Lucas)). 17 ‘A force de travail et d’habileté, MM. Royer et Vaëz ont forcé notre langue si froide, si peu ductile, si embarrassée de consonnes, si chargée d’épithètes, à entrer sans trop de meurtrissures ni d’avaries dans ce moule étroit et flexible de la poésie italienne’ (L’illustration, 14 September 1844). 18 The general approval with which Royer and Vaëz’s work was greeted may be suggested by two quotations, one from the beginning of their career and one from the end: ‘MM Alphonse Royer and Gustave Vaëz have made their translation with ability and conscience; the Italian libretto has only been subject to light modification that has completely been to its advantage …’ (‘MM. Alphonse Royer et G. Vaës [sic] ont fait leur traduction avec talent et conscience; le libretto italien n’a subi que de légères modifications qui toutes lui ont été avantageuses’ (Journal des débats, 6 August 1839; Hector Berlioz)); ‘These are the same authors who have originated these two reconstructions, and one must be deeply grateful to them for having twice [Robert Bruce and Jérusalem] come out of this sort of undertaking, surrounded with so many great difficulties, with honor’ (‘Ce sont les mêmes auteurs qui ont procédé à ces deux reconstructions, et on doit leur savoir gré d’être sortis deux fois avec honneur de cette sorte d’entreprise, entourée d’assez grandes difficultés’ (Le siècle, 28 November 1847; Louis Desnoyers)).

320  Partners in rhyme Royer and Vaëz were admired for the way in which they had responded to ‘le mouvement et les rhythmes’ of the music and, in general, for overcoming the difficulties of writing poetry to pre-existing melodies.19 At least one critic, Hippolyte Lucas, devoted an entire article to the technical merits of what he called this ‘travail souterrain’ – the underground techniques of translating Italian poetry into French that could be sung at the Académie Royale de Musique.20 Specifically, their translations were praised for their qualities as sung texts, an important consideration given the serious issues that arose with singers with technique designed for grand opéra of the 1830s and 1840s confronting Italian opera of – in some cases – previous generations.21 All the poetry in the translations of the 1840s was praised for the richness and sonority of its versification – a wide range of rhymes and their happy juxtaposition in the new texts. A recurring feature of the praise for Royer and Vaëz’s translations was their fidelity to their original, and it is unclear what this might mean in the context of critical writing in the 1840s. In some cases, it was clearly generic approbation not based on first-hand comparison of, say, the 1844 Othello and its 1816 original. But for the critic of La France musicale, Marie Escudier, ‘They have even pushed fidelity and respect for musical phraseology as far as choosing only those French words that have the greatest consonant affinity with the Italian words’ and went on to claim that ‘they have furthermore, as in their previous work of this type, sought to render the feeling of the melodies rather than the sense of the Italian words, so often trifling and gibberish’.22 ­Following the cast of the melody at the expense of words, even if thought to be trifling and gibberish, hardly fits with what might today be considered a faithful translation, but in the context of French views on the sense of Italian libretti – as opposed to the eminent suitability of the language for music – such fidelity begins to develop an explicable context. Nevertheless, a reading of the press that commented on these issues reveals a wide range of tensions, not about whether Royer and Vaëz had produced faithful translations – it was

19 La quotidienne, 23 February 1846. 20 Le siècle, 19 November 1844. 21 ‘[Paul-Bernard] Barroilhet performed the aria from La donna del lago, intercalated into his role [Iago, in Othello] in a very brilliant manner; but from his style of performance, a completely modern selection of some of his cadenzas, one would have said a cavatina by Donizetti; the Rossinian color had disappeared in the translation’ (‘Barroilhet a exécuté d’une manière très brillante l’air de la Donna del Lago, intercalé dans son rôle; mais à son style d’exécution, au choix tout moderne de quelques-unes de ses cadences, on eût dit d’une cavatine de Donizetti; la couleur rossinienne avait disparu dans la traduction’ (Le commerce, 5 September 1844)). 22 ‘Ils ont même poussé la fidélité et le respect envers la phraséologie musicale de Rossini jusqu’à ne choisir que ceux des mots français qui ont le plus d’affinité consonante avec les mots italiens’ … ‘Ils ont d’ailleurs, comme dans leurs précédents travaux de ce genre, cherché plutôt à rendre le sentiment des mélodies que le sens des paroles italiennes souvent oiseuses et amphigouriques’ (La France musicale, 8 September 1844).

Partners in rhyme  321 almost always considered that they had – but about what might constitute fidelity in translation as late as the eve of the 1848 Revolutions. Fidelity to an original in translation was a very different concept in Paris at the end of the July Monarchy to what it is today where the original sequence of numbers coupled to the integrity of the drama might be thought to be central – with the possible acknowledgment of the use of arie di baule. And while Jérusalem is today considered a satisfactory reworking of I Lombardi alla prima crociata, this is largely because the musical changes were new and were authored by Verdi himself. To criticise the 1844 Othello, largely as a result of the interpolation of pre-existing works from elsewhere in the Rossini canon, is a more typical modern view. For Royer, Vaëz and their mid-century admirers, such a consideration was clearly not as important as faithfulness to the original text of the libretto, which meant ­Shakespeare and Scott rather than Berio di Salsa or Cammarano, and musical ­considerations – reflecting the caesura and accentual structure of Italian poetry in the translation, for example – were more important than the accurate translation of the words themselves.

Musical Collaborators Royer and Vaëz did not work in a vacuum; when they assembled the l­ ibretti for Lucie de Lammermoor and Othello or wrote them for Robert Bruce and Jérusalem, they were working in collaboration with composers. In the case of Lucie de Lammermoor, the conditions for making the arrangement were ideal: Donizetti had recently arrived in Paris with an ambition to conquer all four active opera houses and with the clear u ­ nderstanding that reworking his existing operas in French would require significant ­effort on his part. The collaboration clearly worked well: Royer and Vaëz’s translation required little change in the score, and Donizetti willingly made the required modifications. In the area where most change was required, the recitatives, the work was made easier by the fact that – ­u nlike such earlier endeavours as Don Giovanni or Rossini’s reworkings from the 1820s – Donizetti’s original recitative was fully scored, and much of it could be allowed to stand, or at least serve as the basis for translation. Although Royer and Vaëz’s almost complete removal of Raimondo from the French version of the opera triggered the removal of his act ii aria, the substitution of Lucia’s ‘Regnava nel silenzio’ with a translation of ‘Perchè non ho del vento’ from Rosamonda d’Inghilterra resulted simply from the use of that aria di baule in Parisian performances at the Théâtre-Italien during the previous two years. Royer and Vaëz’s collaboration with ­Donizetti is one instance where claims of fidélité match modern expectations closely. Although Royer and Vaëz’s collaboration with Verdi was equally happy, the type of work involved was of a different order altogether. The libretto for Jérusalem can only be related to Solera’s I Lombardi alla prima crociata very

322  Partners in rhyme indirectly, with some scenes and character relationships remaining. Similarly, the musical consequences for Verdi represented a step change over those demanded of Donizetti a decade earlier for Lucie de Lammermoor or for the musical arranger of Othello; hardly a single number from I Lombardi figures in Jérusalem without internal reworking or resequencing, and the amount of new material that Verdi wrote, so much more than for any other translations here under review, has been constantly noted by subsequent commentators. But the press was in some doubt about the degree to which Jérusalem was based on I Lombardi: some thought it little more than a translation of the Italian original,23 while others claimed that there was no more of I Lombardi in it than there was of La donna del lago (rather under half) in Robert Bruce.24 Despite significantly different levels of intervention and change, Lucie de Lammermoor and Jérusalem have in common the fact that both were the composers’ first works in French and had been anticipated by productions at the Théâtre-Italien: Nabucco and I due Foscari for Verdi and no less than seven works for Donizetti.25 Rossini, in comparison, was a different matter, and Royer and Vaëz’s engagement with his works of a significantly higher degree of complexity. The overriding context for the 1844 Othello and 1847 Robert Bruce pasticcio was Parisian impatience with Rossini’s silence for over a decade. Parisian audiences were used to delays with foreign composers: they had to wait five years between Meyerbeer’s Robert le diable and his Les Huguenots, and the 1840s were marked by the knowledge that the composer had completed Le prophète at the beginning of the decade but that the Parisian public would have to wait until 1849 for its premiere. Rossini’s visit to Paris in 1843 was greeted with some excitement, although such enthusiasm that would be only partially satisfied by the 1844 Othello, a work that had been well known in Paris since its premiere at the Théâtre-Italien in 1821.26

23 The commentary in the Revue et Gazette musicale de Paris is even more remarkable: while it argues that ‘at first glance, the additions that the composer has made do not seem very extensive’, it admits the possibility that – as in the case of Othello and Robert Bruce – the additions made by the composer were ‘either new compositions or borrowed from his other works’ (‘Les additions qu’y a faites l’auteur, soit en morceaux nouveaux, soit en emprunts à ses autres ouvrages, nous semblent au premier coup d’œil peu importantes’ (Revue et ­G azette musicale de Paris, 28 November 1847) [emphasis added]). 24 ‘[Jérusalem] has been constructed with the material of I Lombardi, in the same way as Robert Bruce was largely built out of fragments of La donna del lago’ (‘Elle a été bâtie en très grande partie avec les matériaux des Lombards, I Lombardi, comme Robert Bruce a été composé en grande partie des fragments de la Dame du lac, la Donna del lago’ (Le siècle, 28 November 1847; Desnoyers)). 25 John Black, ‘The Contract for Paris’, Donizetti and France, eds., Alexander Weatherson and Fulvio Stefano Lo Presti [The Donizetti Society Journal 7 (2002) ([London]: Donizetti Society; [Bergamo]: Fondazione Donizetti, 2002), 17. 26 Giuseppe Radiciotti, Gioacchino Rossini: vita documentata opere ed influenza su l’arte, 3 vols (Tivoli: Chicca, 1927–29) 2:292; Rossini à Paris: exposition au Musée Carnavalet, 27 octobre – 31 décembre 1992, ed. Jean-Marie Bruson (Paris: Société des Amis du Musée Carnavalet, 1992), 125.

Partners in rhyme  323 Royer and Vaëz were able to profit from the current view that Berio di Salsa’s version of Othello was merely a pale imitation of Shakespeare. This was bound to be the case for any libretto based on an early-modern play, but it gave Royer and Vaëz room to manoeuvre as they put together their adaptation. The very complex set of surviving sources for Othello shows how Rossini’s original music was largely retained, with only Rodrigo’s act ii aria being cut, but also how interpolations from his other operas – L’italiana in Algeri, La donna del lago and others – were necessary for this threeact adaptation.27 While Royer and Vaëz simply wrote their French words for the arias and ensembles onto printed copies of the music in question (whether from Otello or elsewhere), all the recitatives were rewritten and in some cases composed by Benoist. Contemporary voices clamoured in praise of his discretion, and journalists claimed that they could easily recognise the original Italian recitatives in the French version and noted that Benoist’s original additions were restricted to a chord here and there.28 This is a slight overstatement, since he was responsible for all the new recitative around the substitution of the aria from La donna del lago in act ii, but given that Rossini had no hand in this production, claiming fidelity to the composer’s score was as much as could be anticipated. Central to the commentaries on the 1844 Othello was the fact that all the additions were taken from works by Rossini himself and not, for example, from works by Donizetti or Bellini, as might have been the case.29 Furthermore, Benoist was praised for having selected the music for the divertissements from works that were largely unknown in Paris, particularly Armida.30 The distinction between borrowing from Rossini himself as opposed to other composers and between known and little-known works gives an additional perspective on the concept of fidelity c1840 about which Royer and Vaëz’s activities reveal so much. 27 The sources for Othello are discussed in chapter 7. 28 ‘M. Benoist also has had to respect the work of the master, although there were in some parts a few notes to add to the recitative in order to link the pieces together’ (‘M. ­Benoist devait lui aussi, respecter l’œuvre du maître, et cependant il y avait dans quelques p ­ arties, quelques notes à ajouter au récitatif, pour lier les morceaux entre eux’ (La France, 9 ­September 1844; Théodore Anne)). 29 ‘Rossini’s work has not only been respected by M. Benoist, but all the additions have been borrowed from works by the Maestro, all, even the dance music’ (‘L’œuvre de R ­ ossini a été non seulement respectée par M. Benoist, mais toutes les additions ont été empruntées aux œuvres du maëstro, tout, jusqu’aux airs de danse’ (La France, 4 September 1844; Anne)). 30 ‘M. Benoist had then to compose the ballet music in the first act. Instead of drawing on his own sources, the composer of La gypsy and of Le diable amoureux preferred to have us hear the ballet music from Armida, from which is only recognized in Paris the fine duet ‘Amor! Possente nume’. (‘M. Benoist avait enfin à composer le divertissement du premier acte. Au lieu de puiser dans son propre fonds, l’auteur de la Gypsy et du Diable amoureux a préféré nous faire entendre les divertissements de l’Armida dont on ne connaissait à Paris que le beau duo: Amor! possente nume’ (Le constitutionnel, 7 September 1844)).

324  Partners in rhyme Pasticcio However praiseworthy the work of Benoist, Royer and Vaëz, Paris still felt cheated of its new opera by Rossini; Othello – as the press put it – ‘suffered from [Rossini]’s absence’.31 Such a sense that Paris was Rossini’s natural home – and in the middle of the nineteenth century, Parisian critics thought of Paris as the entire musical world’s natural home – the administration of the Académie Royale de Musique took steps to bring, if not a new work by Rossini, at least a major adaptation under his direction, to the stage. A completely new work looked less and less likely as the 1840s progressed, but serious consideration had been given as early as 1843 to a production of La donna del lago at the Opéra; this plan was rejected since – like Otello – La donna del lago had been an important part of the repertory of the Théâtre-Italien since its premiere there in 1824. But the residue of La donna del lago found its way into the work to which Rossini did finally contribute to the Académie Royale de Musique, the pasticcio, Robert Bruce. The Académie Royale de Musique was taking no chances with Robert Bruce; its director, Léon Pillet, sent both Vaëz and Niedermeyer to Bologna for a month from mid-June to mid-July 1846.32 It seems certain that the libretto was complete before Vaëz and Niedermeyer left Paris and that all the musical materials were assembled in Bologna; Niedermeyer’s working score is based on an oblong-format holograph, copied anonymously and ­almost certainly in Bologna, of all the music selected by Rossini.33 Vaëz alone must therefore have fitted and adjusted the new French libretto to the music selected by Rossini. As in the case of Othello, most of the ­musical changes fell in the recitatives, and these are copied in Niedermeyer’s hand in the working score. Whether Niedermeyer merely copied originals dictated by Rossini, or whether he wrote them himself is difficult to settle; certainly the high level of erasure and correction suggests that these are from Niedermeyer’s hand. Rossini’s involvement with Robert Bruce was a great coup for the Académie Royale de Musique. Even without a sequel to Guillaume Tell, Paris was pleased to have such a work, a view summarised by Gustave Hequet in L’illustration: Without doubt it would have been better to have had an original score; but when one cannot have what one wishes, one has to be content with 31 ‘Rossini is in Bologna, forgetting all his scores, disdaining glory, and the translation of Otello necessarily suffered from his absence’. (‘Rossini est à Bologne, oubliant toutes ses partitions, dédaignant la gloire, et la traduction d’Othello a nécessairement souffert de son absence’ (ibidem)). 32 Sources for Robert Bruce are discussed in chapter 7. 33 Niedermeyer’s working score is now Paris, Bibliothèque nationale de France, ­BibliothèqueMusée de l’Opéra, A.554.a.I-III.

Partners in rhyme  325 what one finds, and a pasticcio by Rossini appears to us preferable to the original works of many composers.34 And this was a typical view notwithstanding those of Fiorentino and ­Berlioz.35 So while Robert Bruce problematises the issue of joint authorship, pasticcio and artistic intention, in the late 1840s, such a stage work could be viewed within the same critical framework as both an entirely new opera and the types of reworkings that Royer and Vaëz had promoted for the previous decade. *** To take an overview of opera in translation at the Académie Royale de Musique in the first half of the nineteenth century is to give a central position to Royer and Vaëz’s work in terms of both chronology and continuity. Although there had been occasional overlaps in personnel, for example, between Les mystères d’Isis and Castor et Pollux (Étienne Morel de Chedeville was the translator of both librettos), Le siège de Corinthe and Moïse (Balocchi) and Euriante and the 1834 Don Juan (Castil-Blaze), these are relatively rare. And after Royer and Vaëz had stopped working on opera in translation, Emilien Pacini (who had translated the libretto to Le ­Freyschutz) was responsible both for the translations of both Louise Miller (1853 in collaboration with Benjamin Alaffre) and the 1857 Le trouvère; similarly, Charles Nuitter translated both Bellini’s I Capuleti e il Montecchi (1859) and Tannhäuser (1861) as Roméo et Juliette and Le Tannhauser, respectively.36 But despite these occasional moments when a single translator and arranger became involved in more than one work, none of these endeavours matches the consistent coverage of the field of translating Italian opera that Royer and Vaëz achieved between 1839 and 1847. There is as wide a variety of working practices at the middle of the century as at the beginning. For both the 1801 Mystères d’Isis and the 1834 Don Giovanni production, for example, it was the norm to recast works radically – either for the original composers, deputies or others – and frequently to include pre-existing music from other works or even other composers. And such freedom was common as late as 1859 (Roméo et Juliette

34 ‘Mieux eût valu sans contredit une partition originale: mais quand on n’a pas ce qu’on désire, il faut savoir se contenter de ce qu’on trouve, et un pastiche de Rossini nous paraît encore préférable aux œuvres originales de bien des compositeurs’ (L’illustration, 19 January 1847). 35 Their reviews appear, respectively, in the Journal des débats and Le constitutionnel, both for 3 January 1847. 36 Louise Miller and Le trouvère are discussed in Hervé Gartioux, ‘La reception des operas de Verdi en France entre 1845 et 1867, a travers une analyse de la presse’ (PhD diss., Université de Paris III, 1999).

326  Partners in rhyme was an amalgam of Bellini and Vaccai).37 Similarly, translations that largely respected the structure of their originals, with minimal substitution and resequencing, were common, although paradoxically they are found in Castil-Blaze’s much-criticised translations of Rossini’s Il barbiere di Siviglia (1819) and of Donizetti’s Anna Bolena (1835), which adapt their originals little more than contemporary Italian productions of the same works.38 Royer and Vaëz adapted the concept both of translations that involved significant adaptation by the original composer and of those that did not; they also continued the tradition of pasticcio that, although popular elsewhere, had lain dormant at the Académie Royale de Musique since 1813.39 Shifting focus away from histories of music centred on great composers and enduring works and towards musical cultures in which other individuals and cultural artefacts might play a role, views of the activities at the Académie Royale de Musique are incomplete if such agents of change as Royer and Vaëz are ignored. Their achievements demonstrate the continued acceptance of what today are considered high levels of change in the translation and arrangement of Italian opera for the French stage as late as the 1850s, as well as the importance of the wide range of composers, both of original works and musical arrangers, with whom they worked. Their adaptations were widely praised during the 1840s for what was then understood as fidelity to the original work they were translating, and such praise points to the importance of contextualising of such artistic endeavours in order to understand them. Finally, Royer and Vaëz’s work on what are now, with the exception of Robert Bruce, all regarded as classics points to the very high level of detailed analysis in which contemporary critics engaged, communicating an understanding of both the detail of Royer and Vaëz’s work and its background to their voracious readers.

37 This important production is too little known, but see Janet Johnson, ‘“Vieni a veder ­Montecchi e Cappulletti”: Bellini’s Roméo et Juliette, grand opéra’ (paper presented at the international symposium ‘The Institutions of Opera in Paris from the July Revolution to the Dreyfus Affair’, Duke University and University of North Carolina at Chapel Hill, 24–26 September 2004) and Lajarte, Bibliothèque musicale, 2:226–227. 38 Castil-Blaze incorporated ‘Di tanti palpiti’ into the lesson scene of Il barbiere di Siviglia in common with many Italian productions all over Europe; see Everist, ‘Lindoro in Lyon: Rossini’s Le Barbier de Séville’, Acta musicologica 44 (1992), 50–85. Whether the inclusion of ‘Una furtiva lagrima’ in his translation of Anne de Boulen, for Le Havre in 1835, was personal choice or a wider practice remains to be determined. 39 The last pasticcio to be performed at the Académie Royale de Musique before Robert Bruce was Le laboureur chinois, compiled from works by Haydn and Mozart and arranged by Ludwig Wenzel Lachnith and Henri-Montan Berton. See Peter Revers, ‘Mozart und China: Henri-Montan-Bertons Pasticcio Le laboureur chinois: ein Beitrag zur französischen Mozart-Rezeption des frühen 19. Jahrhunderts’, Bericht über den Internationalen ­Mozart-Kongreβ, Salzburg, 1991, 2 vols [paginated consecutively], eds. Rudolph ­Angermüller, Dietrich Berke, Ulrike Hofmann and Wolfgang Rehm, Mozart-Jahrbuch 1991 des Zentralinstitutes für Mozartforschung der Internationalen Stiftung Mozarteum Salzburg (Kassel, Basel and London: Bärenreiter, 1992) 2:777–786.

Partners in rhyme  327 Royer and Vaëz emerge as key players in the transfer of operatic ­cultures from Italy to France and central to the subsequent careers of Rossini, Donizetti and Verdi. Uncovering the ways in which opera developed as a pan-European phenomenon during the nineteenth century, the work of Royer, Vaëz and their colleagues who did so much for the internationalization of music drama provides a point of entry into a world that challenges so many modern assumptions about the fashion and legacy of nineteenth-century Italian opera.

Second entr’acte La musique allemande

While Italian composers had a relatively easy time assimilating to the ­musical culture of the nineteenth-century Parisian stage, the same cannot be said for their German colleagues. Almost none came in person to the city to work on their productions and, in many cases, gave the impression that they would rather not have their works produced there at all. Weber’s concern that his ‘intellectual property’ had been somehow stolen when Thomas Sauvage’s and François-Henri-Joseph Blaze (Castil-Blaze)’s arrangement of Der Freischütz as Robin des bois had taken Paris by storm in 1824 was distorted by Berlioz’s attempt to tell the story in a way that threw his current projects into a favourable light (when he was responsible for exactly the same project at the Opéra (1841)). But in fact, Weber’s view of music in the Parisian theatre was such that – at his death – he was planning performances not merely in Paris but at the very opera house that had premiered Robin des bois. Weber’s place in nineteenth-century Paris is fragmented, and Chapter 10 brings into focus the disparate elements of the composer’s reception in the city through the lens of Castil-Blaze’s critical writing. The press is a fundamental part of the epistemology of nineteenth-century French culture of all sorts, and in this chapter, not only is Weber foregrounded but so too is one of the most important critics of the period. There is perhaps a sad irony in that the chapter stops with Castil-Blaze’s death in 1857, just as the most successful series of Weber performances was starting at the Théâtre-Lyrique. Performances of Robin des bois, Oberon and Euryanthe took place in 1855 and 1857, but it is not clear how many of these the critic saw (he certainly never reviewed them), and he did not live long enough to see Preciosa and Abou Hassan in the later 1850s.1 If Weber’s place in the history of nineteenth-century Paris is complicated by the fact that his reception was largely posthumous, it is even more true of Gluck’s works for the theatre. Not only that, but there is a real doubt as to whether Gluck should even be placed alongside Weber and Wagner as he is here. Gluck’s national status is as variable as that of his Upper 1 Thomas Joseph Walsh, Second Empire Opera: The Théâtre Lyrique, Paris, 1851–1870, The History of Opera (London: Calder; New York: Riverrun, 1981).

330  Second entr’acte Palatinate place of birth. Claimed as German by the middle third of the twentieth ­c entury, he was considered by French musicians as not only a native but also the originator of grand opéra on the Meyerbeerian model.2 In nineteenth-­c entury Gluck reception, the composer is frequently inscribed as ­Meyerbeer’s ‘father’. Attempts to bring Gluck’s works back to the stage after their disappearance c1830 were inflected in a large number of ways. Here, in Chapter 11, the nineteenth century re-enacted one of the cultural predilections of the eighteenth: the pamphlet war, with figures central to Napoléon III’s third cabinet jousting in the press not over the Italian and French music of their forefathers but over the music of the living and the dead. Just at the same time as Napoléon’s ministers were arguing over the past and the present, the art of the future was arriving in Paris. More ink has probably been spilt over the disastrous 1861 production of Wagner’s ­Tannhäuser in Paris than over any other example of cultural transfer in music. Whatever the reason for its catastrophic failure, whomever is blamed, and whatever this says about Wagner, Parisian musical culture or la musique de l’avenir, Le Tannhauser, as the production was called, was a cultural, political and aesthetic controversy of near-biblical proportions. And to judge from the literature today, the reasons for, and significance of, the failure still remain contentious territory. For those who take Wagner’s Mein Leben at face value, the matter is easy: Wagner’s genius was rejected by a corrupt and decadent Parisian musical culture led by the hooligans of the Jockey Club.3 The truth is much more complex and requires careful analysis of competing pro-Wagnerian factions, different trends in the press and the repertory at the Opéra c1860, to say nothing of the disdain felt by many for a production that was merely forced on the Opéra as a diplomatic concession. All this is summarised in Chapter 12 as an introduction to a second Wagner production in Paris, of Rienzi in 1869, where every effort was made to efface the controversy around the 1861 Le Tannhauser. Attempts by all the agents concerned – even Wagner himself – to minimise the contentious were largely successful, and the production allowed for the first time a level-headed assessment of one of the composer’s – albeit early – compositions. 2 For the appropriation of Gluck as a German composer, and the origins of the Sämtliche Werke of the composer, see Michael Custodis, ‘Tradition, Pragmatik, Ideologie: Rudolf ­Gerber und die Gründung der Gluck-Gesamtausgabe (1939–1953)’, Christoph Willibald Gluck (1714–2014): Bilder Mythen Diskurse, eds. Thomas Betzwieser, Michele Calella, and Klaus Pietschmann (Vienna: Hollitzer, forthcoming). For his place in French culture, which goes beyond Berlioz’s engagement with the revivals of the composer’s music, see Mark E ­ verist, Genealogies of Music and Memory: Gluck in the Nineteenth-Century Parisian Imagination (forthcoming). 3 References to the activities of the Jockey Club at the Paris premiere of Tannhäuser receive far more attention than they deserve. Of the tens of thousands of words spilled in the Parisian press in 1861, a tiny handful mention the Jockey Club, and all can be traced either to Wagner himself or to quotations of German-speaking publications which may well have been overly well disposed to the products of their compatriots or directly quoting Wagner. The only study of the Jockey Club (Joseph-Antoine Roy, Histoire du Jockey Club de Paris (Paris: Rivière, 1958), 13) that mentions Tannhäuser simply repeats text from Mein Leben.

10 Castil-Blaze and the reception of Weber in Paris, 1824–1857

François-Henri-Joseph Blaze was known as Castil-Blaze for his entire professional life.1 He was a central actor on the stage of Parisian music drama during the Restoration, the July Monarchy and the first half of the Second Empire. A writer on music, a journalist and a publisher, he was a critical agent in the importation and assimilation of Italian and German opera into both Parisian and provincial musical circles. In his lifetime, he was respected, and his work was valued and understood. Largely as the result of intemperate comments in Berlioz’s Memoirs, he has come to be vilified by those who fail – or refuse – to understand the culture of nineteenth-century French opera.2 However, his journalism is now understood as the first attempts at

1 Still the best account of Castil-Blaze is that of Pierre Larousse, ‘Blaze (Fr.-Henri-­Joseph, dit Castil-)’, Grand Dictionnaire universel du xixe siècle français, historique, géographique, biographique, mythologique, bibliographique, littéraire, artistique, scientifique, etc., 17 vol. (Paris: Grand Dictionnaire Universel, 1866–1877), 2:812–813 (volume 2 dates from 1867, exactly ten years after Castil-Blaze’s death). The two articles in GroveOnline, while updated, are of more modest scope: Cormac Newark, ‘Castil-Blaze’, Grove Music Online. Oxford Music Online, 8 October 2012, www.oxfordmusiconline.com/subscriber/ article/grove/music/05133 and Mark Everist, ‘Castil-Blaze’, The New Grove Dictionary of ­Opera, ed. Stanley Sadie, Grove Music Online. Oxford Music Online. 8 October 2012 www.oxfordmusiconline.com/subscriber/article/grove/music/O003340. François-Joseph Fétis’ two articles, Biographie universelle des musiciens et bibliographie générale de la musique, 8 vols (2e ed. [with supplement in two volumes] Paris: Firmin Didot, 1860–1865), 1:440–443 and Supplément 1:99 are still useful. See also Die Musik in Geschichte und Gegenwart: allgemeine Enzyklopädie der Musik, ed. Ludwig Finscher, 29 vols (Kassel, etc.: Bärenreiter, 1995–2007), Personenteil 4:416–417. 2 Berlioz’s most virulent diatribe (written decades after the event and further complicated by his own engagement in a similar project) is found in the Mémoires d’Hector Berlioz, Membre de l’Institut (Paris: n.p., 1870), translated and edited by David Cairns under the title The Memoirs of Berlioz, Member of the French Institute, Including His Travels in Italy, Germany, Russia and England: 1803–1865 (London: Gollancz, 1969; Panther, 1970), 99–106 (page numbers refer to the 1970 edition). An uncritical reading of the Mémoires, coupled to an outdated view of music for the nineteenth-century stage, is found in Catherine Nazloglou, ‘Castil-Blaze [Blaze, François-Henri-Joseph]’, The New Grove Dictionary of Music and Musicians, 20 vol., ed. Stanley Sadie (London: Macmillan, 1980), 3:872.

332  Castil-Blaze and the reception of Weber a sophisticated discourse on music in nineteenth-century France,3 his relationship with Berlioz is coming under greater scrutiny,4 and his work on the naturalisation of foreign opera is now being read in a way that takes the complexities of early nineteenth-century France more seriously.5 Castil-Blaze’s writings on music are the most sophisticated of their generation. From the 1850s are a series of weighty volumes on the history of opera in France: two – on the Opéra and Théâtre-Italien – were published and remained fundamental texts until well into the twentieth century, and a third, on the Opéra-Comique, remained in manuscript in the Bibliothèque-Musée de l’Opéra.6 Other works of a similar range and ambition were his Molière musicien of 1852 and an important work on versification and libretti, his L’art des vers lyriques, published in the year of his death.7 It was probably his De l’opéra en France of 1820 and his Dictionnaire de musique moderne of the following year that really put him on the map in the first place.8 These texts were well received across the musical world, and largely led to his appointment as what might be considered the first professional musical journalist in Paris. He published what he called a ‘Chronique musicale’ in Le journal des débats from 1820 to 1831, the first sustained campaign of professionally competent musical journalism in a major daily newspaper in Paris and perhaps in Europe. The levels of discourse that will be discussed later in this paper will be seen to be an impor-

3 A fundamental advance in the understanding of Castil-Blaze’s journalistic output was made in Donald Garth Gislason, ‘Castil-Blaze, De l’Opéra in France and the Feuilletons of the Journal des Débats (1820–1832)’ (PhD diss., University of British Columbia, 1992). But even in this study, and at a point where the critical edition of the Berlioz’s criticism is near completion, very little of Castil-Blaze’s output is easily available. See also I. H., Turnevicius, ‘The Feuilletons of Castil-Blaze in the Journal des Débats (1820–1832): A Critical Examination’ (MA diss., McMaster University, 1994). 4 See Mark Everist, ‘Gluck, Berlioz and Castil-Blaze: The Poetics and Reception of French Opera’, Reading Critics Reading: French Music Criticism, 1789–1848, ed. Roger Parker and Mary Ann Smart (Oxford: Oxford University Press, 2001), 86–108. 5 Idem, Music Drama at the Paris Odéon, 1824–1828 (Berkeley, Los Angeles and London: University of California Press, 2002), passim. 6 François-Henri-Joseph Blaze (Castil-Blaze), L’académie impériale de musique: histoire littéraire, musicale, chorégraphique, pittoresque, morale, critique, facétieuse, politique et galante de ce théâtre, 2 vols, Théâtres lyriques de Paris [1] (Paris: Castil-Blaze, 1855); idem, L’opéra Italien de 1548 à 1856, Théâtres lyriques de Paris [2] (Paris: Castil-Blaze, 1856); Castil-Blaze’s Histoire de l’opéra comique de Castil-Blaze has been the subject of new edition: Castil-Blaze: Histoire de l’opéra comique, eds. Alexandre Dratwicki and Patrick Taïeb (Lyon: Symétrie, 2012). 7 François-Henri-Joseph Blaze (Castil-Blaze), Molière musicien: notes sur les oeuvres de cet illustre maître et sur les drames de Corneille ... Beaumarchais, etc, où se mêlent des considerations sur l’harmonie de la langue française, 2 vol. (Paris: Castil-Blaze, 1852); idem, L’Art des vers lyriques (Paris: Delahaye, 1858). 8 Idem, De l’opéra en France, 2 vols (Paris: Janet et Cotelle, 1820; 2nd edn Paris: Sautelot, 1826); idem, Dictionnaire de musique moderne, 2 vol. (Paris: A la Lyre Moderne, 1821; 2nd ed. 1825).

Castil-Blaze and the Reception of Weber  333 tant precursor of the appearance of such specialist music journals as the Revue musicale in 1827.9 With the exception of his work in the Journal des débats, Castil-Blaze’s journalistic career remains relatively unknown. Frequent reviews appeared in Le constitutionnel up to January 1833, at which point Castil-Blaze appears to have dedicated himself exclusively to the Revue de Paris, where his name appears up to 1840. He also wrote for La Revue et Gazette musicale de Paris in the 1830s and took on some editorial responsibilities there.10 His later journalistic work, more scattered, is found in La revue de la musique ancienne et moderne, Le ménestrel, L’europe littéraire, Le musée des familles, Le magasin pittoresque, Le dictionnaire de la conversation and La revue française. At the end of the 1840s, finally, Lamennais’ La reforme was the frame for a later but sustained period of work: a series of ‘chroniques musicales’ appeared there between October 1848 and June 1849.11 And if this level of professional engagement was not enough, Castil-Blaze was engaged in the publication of music and texts about music from an early stage. In the 1820s, he was commercially involved with Charles Laffillé, and from as early as 1822 managed his own publishing house.12 Not only did he publish scores of works he arranged for the French stage, but also libretti, collections of operatic excerpts arranged in albums for piano – Le Dalayrac des concerts, for example – and, of course, the major works of scholarship that he wrote himself. But Castil-Blaze was at least as well known for his commitment to the naturalisation of foreign opera on French soil. And it is this part of his work that is probably least well understood. To begin with, a distinction has to be drawn between his arrangements of complete works – Le nozze di ­Figaro, L’italiana in Algeri, Oberon, for example – and his four pasticci: La fausse Agnès, Les folies amoureuses, Monsieur de Porceaugnac and La forêt de Sénart.13 For the most part, his arrangements were made for provincial opera houses: Lyon, Lille, Bruxelles, Nîmes, Bordeaux and so on.14 9 On the importance of Castil-Blaze for the birth of the Revue musicale in 1827, see Katharine Ellis, Music Criticism in Nineteenth-Century France: La Revue et Gazette Musicale de Paris, 1834– 80 (Cambridge: Cambridge University Press, 1995), 27–32. 10 Ibidem, xxx. 11 Castil-Blaze’s series of reviews in La réforme are found in Paris, Bibliothèque nationale de France, 4-B Pièce 13. See Lise Devreux and Philippe Mezzasalma (eds.), Des sources pour l’histoire de la presse (Paris: Bibliothèque Nationale de France, 2011), 117. Further research is necessary to determine the importance of Castil-Blaze’s work for La réforme. 12 For both activities, see Anik Devriès and François Lesure, Dictionnaire des éditeurs de musique français, 2 vol. [vol. 1 in 2 parts], Archives de l’édition musicale française 4 (Geneva: Minkoff, 1979–1988), 2:91–92. 13 For discussion of these pasticci, see Everist, Music Drama at the Paris Odéon, 193–198. 14 Until there is a coherent and comparative account of music for the French provincial stage, it will remain difficult to establish with precision the importance of Castil-Blaze’s contribution. For Lyon, however, see Everist, ‘Lindoro in Lyon: Rossini’s Le Barbier de Séville’, Acta musicologica 44 (1992), 50–85.

334  Castil-Blaze and the reception of Weber The only work was the one that is central to our concerns today, the arrangement of Weber’s Der Freischütz as Robin des bois which was premiered at the Odéon in December 1824. On the other hand, two of his four pasticci were written expressly for the Odéon, while a third was written for Paris’ Gymnase-Dramatique. Underpinning all of Castil-Blaze’s commitment naturalising foreign opera for the French stage was a dual ideology: of the relationship between institution and genre and of the particular preferences of French audiences that lay below the surface of institutional control. On the one hand, he was sensitive to the licensing arrangements that controlled the importation of foreign opera onto the ­ apoleonic reforms of 1806–1807. French stage that had been in place since the N This meant that foreign opera had to be translated into French at the Odéon, that recitativo semplice had to be replaced by spoken dialogue and that at the Opéra accompanied recitative had to be retained as the libretto was translated into French. For Italian opera, there was a continuing tension between his endeavours to produce Cimarosa, Mozart and Rossini in French at the same time as the works of those composers were being performed in their original language at the Théâtre-Italien. But in addition to such institutional pressures was the second element of ­Castil-Blaze’s musical ideology. The idea that even once barriers of institution and genre had been overcome, there were issues relating to scenario, to the inclusion and exclusion of numbers, to the choice of work in the first place, that had to be calibrated against the constantly changing preferences on Parisian audiences and their tribunes in the press. Modern opinions of Castil-Blaze’s efforts to naturalise foreign opera in France are viewed through the prism of Berlioz’s less than generous account in the sixteenth chapter of his Mémoires. Berlioz’s animosity to Castil-Blaze can be traced back to the 1820s and to an exchange between Castil-Blaze in the flagship Le journal des débats and Berlioz in the marginal Le corsaire; ­Berlioz can only have been infuriated by the effortless grace with which Castil-­Blaze largely ignored the letters he wrote in response to the other’s articles on Gluck; the opportunity to settle old scores when he came to write his Mémoires must have been irresistible.15 There were other reasons for ­Berlioz to fulminate about the success of Robin des bois; when he began writing his Mémoires in 1848, his own 1841 arrangement of Der Freischütz had been off the boards of the Académie Royale de Musique for two years.16 And while it would resurface for a couple of years in 1853, not only was its lifetime at the Opéra at an end, but Berlioz had to contend with the fact that ­Castil-Blaze’s Robin des bois was still very much in circulation and would be the basis of stunningly successful series of productions at the Théâtre-­Lyrique in the middle of the 1850s. Berlioz’s avant garde view of the sanctity of the work 15 See Everist, ‘Gluck, Berlioz and Castil-Blaze’. 16 For the performance history of the Berlioz-Weber Le Freyschutz, see Albert Soubies, ­Soixante-sept ans à l’Opéra en une page du ‘Siège de Corinthe’ à ‘La Walkyrie’ (1826–1893) (Paris: Fischbacher, 1893), and the discussion below.

Castil-Blaze and the Reception of Weber  335 and the unshakeable authority of the composer which was so much at odds with early nineteenth-century experience obviously had a much greater resonance when his Mémoires were eventually published in 1870, at which point the autonomy of the artwork was a more comfortable concept. There are two contexts in which Castil-Blaze and his work on behalf of the reception of Weber in France may be set. The first of these is the performances of Weber’s operas in Parisian opera houses during the period from the composer’s first appearance in 1824 up to Castil-Blaze’s death. The second is Castil-Blaze’s journalistic commentary on Weber which itself requires contextualisation within the world of music criticism in Paris during the Restoration and July Monarchy.

Weber in Paris Between the Restoration and the Franco-Prussian war may be seen at least ten attempts to promote Weber at various opera houses in Paris (Table 10.1). The production of Der Freischütz as Robin des bois at the Théâtre Royal de l’Odéon in December 1824 must stand as one of the most misunderstood productions in nineteenth-century France. Only a year after the Odéon closed in the 1828, the so-called troupe allemande, directed by Joseph ­August Röckel, Table 10.1  P  erformances of Weber’s operas in Paris, 1824–c1870 Date

Work

Opera House

Version

?1823

[Le Freischutz]

?Schlesinger

1823

[Le chasseur noir]

1824

Robin des bois

1826

Les bohémiens

1826

La forêt de Sénart

1829–1831 1831

Der Freischütz; Oberon; Euryanthe Euriante

Académie Royale de Musique Gymnase Dramatique Théâtre Royal de l’Odéon Théâtre Royal de l’Odéon Théâtre Royal de l’Odéon Théâtre-Italien

1835

Robin des bois

1841

Le freyschutz

1842 1855

Der Freischütz et Euryanthe Robin des bois; Oberon; Euryanthe; Preciosa and Abou Hassan. Le Freyschütz

1866

Académie Royale de Musique Théâtre Royal de l’Opéra-Comique Académie Royale de Musique Théâtre allemand Théâtre-Lyrique Théâtre-Lyrique

Castil-Blaze Sauvage et Castil-Blaze Sauvage et Crémont Castil-Blaze Röckel Castil-Blaze Castil-Blaze Pacini et Berlioz Schumann Castil-Blaze Trianon et Gautier

336  Castil-Blaze and the reception of Weber arrived in the capital for three seasons dedicated to opera in German, which ran from 1829 to 1831. The last of these collided with ­Castil-Blaze’s arrangement of Euryanthe for the Académie Royale de Musique; however difficult the competition, the moment gave Parisian audiences the chance to hear Euryanthe both in German and in French, and with German and French singers. Four years later, there was a successful production of Der ­Freischütz at the Opéra-Comique, a production that ran to 60 performances in the year. This was then followed by Berlioz’s arrangement of Der Freischütz as Le Freyschütz in 1841, performances of which continued from 1841 to 1846 and from 1850 to 1853. At this point attention shifted to the newly founded Théâtre-Lyrique, where a revival of Robin des bois was produced in 1855 with spectacular success but was eventually replaced by a new production by Henry Trianon and Jean-François-Eugène Gautier in 1866. The 1824 Robin des bois at the Odéon was the most spectacularly successful production of the season, vying with Boieldieu’s La dame blanche at the Opéra-Comique for the most highly regarded work of the year. But it was only one of three potential productions of Weber’s opera that year. Correspondence between Maurice Schlesinger and François Antoine Habeneck makes it clear that the Académie Royale de Musique was considering a production but also that Schlesinger was unenthusiastic about the chances of the work’s success, writing ‘My opinion of this opera is always the same, and I cannot persuade myself that the Parisian public will ever enjoy the libretto of this opera’.17 Schlesinger was of course right, and it would be Castil-Blaze’s achievement that his arrangement of the work did allow the Parisian public to ‘goûter’ Weber’s opera. Castil-Blaze had originally been contemplating mounting a French version of Der Freischütz at the Gymnase-Dramatique in 1823; this was to be called Le chasseur noir. Very little survives, but it seems likely that to conform to the strictures of the licensing system, Weber’s three acts would have had to have been reduced to one. Even as late as the opening of the Odéon as an opera house in April 1824, Castil-Blaze still referred to Le chasseur noir as a work for the Gymnase-Dramatique. During the course of the second half of 1824, the attraction of a three-act version of Der Freischütz, coupled with the emerging excitement about the productions at the new opera house at the Odéon, encouraged Castil-Blaze to withdraw the work from the Gymnase-­ Dramatique and to move it to the Odéon.18 Premiered at the Odéon in December 1824, by April 1825 Robin des bois was so successful that the management was having to turn away m ­ embers of the audience in their hundreds. The production reached its 100th performance within nine months and its 150th within thirteen months. Other new

17 ‘Mes vues sur cet opéra sont toujours encore les mêmes, et je ne puis me persuader que le public parisien goûtera le poème de cet opéra’ (Jacques-Gabriel Prod’homme, ‘“Robin des Bois” et “le Freyschütz”’, Le Ménestrel 88 (1926) 437). 18 See Everist, Music Drama at the Paris Odéon, 255–258.

Castil-Blaze and the Reception of Weber  337 productions queued up behind Robin des bois, waiting for its success to wane so that they could have their turn at the Odéon; not only did they have a long wait, but the audience would frequently reject whatever work was promised for performance, clamouring for another performance of Robin des bois.19 Der Freischütz, in its original German form, was not viable on any ­Parisian stage, as Schlesinger had made clear. In exactly the same way as Berlioz ­seventeen years later, Castil-Blaze was obliged to rework the opera to conform not only to the institutional restrictions of the Parisian licensing system, but also to respond to what was seen as the ‘savant’ and ‘compliqué’ nature of Weber’s music and drama. The modifications to Kind’s libretto and Weber’s score have been pored over by critics and musicologists many times. The levels of change are no greater than that found in any of ­Castil-Blaze’s arrangements of Cimarosa, Rossini and Mozart for provincial theatres that were also produced at the Odéon during the 1820s. Two examples may stand for many, however. Much has been made by later critics of moving the action from Kind’s Bohemia to Castil-Blaze’s Yorkshire. So easy to sneer at for the critic with no understanding of the cultural milieu of the 1820s, such a move played directly into the current enthusiasm for Sir Walter ­ orthern-English and Scottish romanticism. So for later critics, Scott and n such a change was an aberration, but for the 1820s, it was not only appropriate but secured the success of the production. Similarly, the absence of any duet for the two main characters in Der Freischütz was a serious problem for ­Parisian music drama during the Restoration, and Castil-Blaze’s response was clear, and he supplied the requisite duet, basing it on the duet ‘Hin nimm die Seele mein’ from Euryanthe. Such changes are analogous to the transpositions, additions of recitative and ballet music executed by Berlioz in 1841.20 Much has been made of Weber’s complaints about Robin des bois, most of which have to be explained by his lack of understanding of intellectual property, copyright and its international dimension. Many of these complaints are in fact Schlesinger’s, as is borne out by the fact that before Weber’s visit to Paris (where he allegedly refused to have anything to do with the Odeon), he had been in contact with the management of the Odéon itself for a production of a pasticcio based on his opera Preciosa, as Les bohémiens, which would have premiered in 1826. Had Weber not died in London, the plan was to have the composer come to Paris to supervise the performances. The documentary sources are clear about Weber’s intentions, and sit badly with the views imputed to him by others (Berlioz most notably) about the Odéon and Robin des bois.21 Only a few months after the closure of the Odéon, and the last performances of Robin des bois there, Röckel brought his German troupe to 19 Ibidem, 266–267. 20 Ibidem, 258–266. 21 Les bohémiens formed part of a project designed to mount productions of pasticci not only by Weber but also by Rossini and Meyerbeer. See ibidem, 173–186.

338  Castil-Blaze and the reception of Weber ­ aris.22 The greatest strength of the troupe in the 1829, 1830 and 1831 seaP sons was its singers: Röckel himself had been Beethoven’s Florestan in 1806, Anton Haizinger had been Weber’s first Adolar in Euryanthe and Wilhelmine Schroeder-Devrient (who joined the troupe for the 1830 season) was then at the height of her powers.23 The three seasons had each an individual character, and their impact on Parisian musical culture was quite different. The qualities of the three seasons underpin the reception of Weber in the capital in the years around 1830. Although Röckel’s first year in Paris was something of an experiment, it laid down the basic pattern for the next three seasons. The troupe hired the Salle Favart, usually the home of the Théâtre-Italien, for Tuesdays, ­Thursdays and Saturdays, and mounted productions of opera with libretti in their original German.24 In 1829, the season lasted four weeks and c­ onsisted of thirteen performances. Only three works were mounted: Der Freischütz (7 performances), Die Zauberflöte (2) and Fidelio (4).25 The season may have been an experiment, but it encouraged the troupe to return the following year with a much more ambitious schedule. The 1830 season ran for eleven weeks (thirty-three performances), and its repertory was significantly more elaborate (Table 10.2).26 If the number of performances in 1829 can be taken as an index of a work’s success, Die Zauberflöte seems to have been poorly received, dropped from the 1830 repertory and replaced with another work by Mozart, Die Entführung aus dem Serail. Fidelio was retained, but most of the other operas received only a couple of performances. Three of the works that did less 22 An account of the three seasons, viewed through the career of Wilhelmine Schröder-­ Devrient, is found in Jean Mongrédien, ‘Les Débuts parisiens de Wilhelmine Schröder-­ Devrient’, Festschrift Christoph-Hellmut Mahling zum 65. Geburtstag, 3 vols, eds. Axel Beer et al. (Tützing: Schneider, 1997), 2:935–946. 23 The fullest account of the careers of these individuals is given in Karl-Josef Kutsch and Leo Riemens, Großes Sängerlexikon, 4 vols (Bern and Stuttgart: Francke, 1987–1994), 2:2487 [Röckel]; 1:1221–1222 [Haizinger]; II, 2670–2672 [Schroeder-Devrient]. Other artists in the German Opera mentioned by Kutsch and Riemens were the married couple Friedrich Fischer and Caroline Fischer-Achten (ibidem, 1:940–941 and 943–944), Eduard Genast (ibidem, 1:169), Maria Schmidt (ibidem, 2:644) and Franz Zaver Vetter (ibidem, 2:3086–3087). 24 The claim (Heidlberger, Carl Maria von Weber und Hector Berlioz: Studien zur französischen Weber-Rezeption, Würzburger musikhistorische Beiträge 14 (Tützing: Schneider, 1994), 371) that the German Opera performed on Wednesdays, Thursdays and Saturdays therefore needs revision. 25 The productions made little impact, however, on French operatic life. For an exception, see the enthusiastic account (as might be expected in a journal that so firmly adhered to the aesthetic imperatives of De l’Allemagne) in Le Globe, 16 May 1829, cited in Heidlberger, Carl Maria von Weber und Hector Berlioz, 376. 26 The summary of the repertoire ibidem, p. 372, conflates all three of the seasons and is consequently misleading. The data given in this chapter’s tables are taken from the contemporary press. For a tabular listing of the box-office receipts for 1830, see Mongrédien, ‘Les Débuts parisiens de Wilhelmine Schröder-Devrient’, 2:936–937.

Castil-Blaze and the Reception of Weber  339 Table 10.2  The Repertory of Röckel’s troupe, 1829–1831 Date

Composer

Work

1829

Weber Mozart Beethoven Weber Spohr Winter Pixis Beethoven Weber Weigl Kreutzer Mozart Weber Beethoven Mozart Weber Mozart Weber

Der Freischütz Die Zauberflöte Fidelio Der Freischütz Faust Das unterbrochene Opferfest Bibiana Fidelio Oberon Die Schweizerfamile Cordelia Die Entführung aus dem Serail Der Freischütz Fidelio Die Entführung aus dem Serail Oberon Don Juan Euryanthe

1830

1831

well, Das unterbrochene Opferfest, Die Schweizerfamilie and Cordelia, had been known in French translations at the Odéon earlier in the 1820s;27 the latter two works, known as Emmeline, ou La Famille Suisse and La Folle de Glaris, had not been outstandingly successful at the Odéon (with seven and five performances, respectively), but Das unterbrochene Opferfest, as La Sacrifice Interrompue, had been the first German work at the Odéon and had blazed the trail so successfully followed by Robin des Bois with no less than thirty-one performances in 1824 and 1825. The 1831 season was only slightly shorter than the 1830 one. Röckel had learned his lesson the previous year and brought a repertory of six works instead of nine. The season followed the same pattern as the first two and ran for nine weeks with around thirty performances (Table 10.3) Der Freischütz opened the third and final season of the Opéra Allemand in exactly the same way as it had for the previous two years. Despite the efforts of the Opéra Allemand to promote his works, by the end of the 1820s, Weber had a serious rival in Paris: Giacomo Meyerbeer.

27 Mongrédien’s statement (ibidem, 2:941) that ‘cette dernière pièce [Kreutzer’s Cordelia], aussi bien que son auteur d’ailleurs, sont – est-il besoin de le dire? – totalement inconnus alors du public parisien’ should be reconsidered in the light of the earlier Odéon performances of Cordelia as La Folle de Glaris. Kreutzer was the composer of an opéra comique, also mounted at the Odéon, L’Eau de Jouvence (a rarity, given that the Odéon’s licence did not permit such works). See Everist, Music Drama at the Paris Odéon, 1824–1828, 277–278. However unsuccessful the performances of Kreutzer’s works in mid-1820s Paris, the impact they made on the contemporary press was not inconsiderable.

340  Castil-Blaze and the reception of Weber Table 10.3  The Detail of Röckel’s 1831 Season Title

Premiere

Number of Performances

Der Freischütz Fidelio Die Entführung aus dem Serail Oberon Don Giovanni (in German) Euryanthe

  5 May 1831 12 May 1831 14 May 1831 23 May 1831 26 May 1831 14 June 1831

3 11 1 2 6 5

Although Paris hardly knew anything of his music beyond revivals of his Italian operas, Meyerbeer was known as a ‘German’ composer and subject to many of the same prejudices as Weber. By the end of 1830, works by both composers were being considered for production at the Académie Royale de Musique. In many respects, Castil-Blaze had achieved with his 1831 arrangement of Euryanthe one of his life’s ambitions: he was about to produce one of his arrangements of foreign opera, so far restricted to provincial opera houses or to such theatres as the Odéon, at the Académie Royale de Musique; not only that, but the work was the natural successor to Robin des Bois, the most successful by far of all his arrangements, and it should by all accounts have resulted in similar levels of success. Such achievement was now marred by the fact that the work had to go into production at an unprecedented speed because of the contractual relationships between the Académie Royale de Musique and Meyerbeer for the premiere of Robert le diable. Euryanthe ought to have fitted well into the operatic culture of Paris in the 1820s and 1830s: it was a ritter-romantische opera, and it was through-­ composed. Operatic subjects based on medieval French romance enjoyed a particular vogue in the early 1820s: Schubert’s Fierrebras was planned, as was Euryanthe, for the Kärntnertortheater in Vienna (Schubert’s opera was never produced) and was based on Carolingian legend, while Euryanthe was based on the thirteenth-century Le Roman de la Violette. Only five months after the Viennese premiere of Euryanthe, Meyerbeer and Gaetano Rossi produced their own response to the same fashion with Il crociato in Egitto. Although the work’s background was very different to the two ­Viennese compositions, many of the surface modalities were similar: particularly striking is the central plot design in which a number of knights are captured, sentenced to death and therefore are given the opportunity to display qualities of nobility in adversity.28 In Paris, the young Liszt’s Dom Sanche, 28 Meyerbeer’s opera was not based on any medieval source; it was based on a mélodrame entitled Les Chevaliers de Malte, dating from 1813, which in turn was based on early modern history (although Rossi’s libretto for Il crociato returns the action to the thirteenth century); see Everist, ‘Meyerbeer’s Il crociato in Egitto: Mélodrame, Opera, Orientalism’,

Castil-Blaze and the Reception of Weber  341 ou Le château de l’amour was based on a story with its origins in the middle ages reworked by Claris de Florian in the eighteenth century; the work was started in the same year as Il crociato in Egitto and premiered at the Académie Royale de Musique in 1825. A further work that developed a similar theme but was based on exactly the same source as Euryanthe was Carafa and Leborne’s La violette, premiered at the Opéra-Comique in 1828.29 Castil-Blaze’s version of Euryanthe failed to capture the Parisian imagination in the same way that Röckel’s troupe allemande had in the previous three years. But it was the best known of Castil-Blaze’s arrangements that was the next Weberian landmark on the Parisian operatic map. Robin des bois reappeared, this time at the Opéra-Comique in 1835 where it ran for series of 60 performances during the year. In terms of genre, the Odéon version of Robin des bois, which retained the spoken dialogue that had characterised Weber’s original in a French adaptation, was perfectly suited to the Opéra-Comique whose institutional relationship to genre mapped closely onto that of the Odéon. Indeed this close relationship had been a source of significant difficulty for the two opera houses during the 1820s, and had in part led to the collapse of the Odéon in 1828. The production at the Opéra-Comique in 1835 was therefore based directly on the printed materials from the 1824 Odéon production.30 But the success of the 1835 Robin des bois at the Opéra-Comique points to some remarkable questions concerning Castil-Blaze’s arrangement and its reception in both the Restoration and the July Monarchy. By and large, the 1835 version replicated the 1824 Odéon production. There were two differences, however. The first is the romance et air ‘Un soir rêvant’ – a translation of ‘Einst träumte meiner sel’gen Base’ – which was retained in the later stages of the Odéon productions according to the critic of Le ménestrel in 1820s but was cut from the libretto. This was accordingly omitted from the 1835 Opéra-Comique production. Conversely, the Opéra-Comique also removed the duet ‘Non, plus d’alarmes’ that had figured in act ii of Robin des bois but had originally been taken from Euryanthe. The anonymous critic of Le ménestrel took a less than positive view of the two decisions:

Cambridge Opera Journal VIII (1996), 217–218. This throws into question the perhaps too-confident claims made by Francis Claudon (‘Style troubadour et grand opéra’, Euryanthe: Livret, traduction, analyses, discographie (Paris: L’Avant-Scène Opéra, [1993]; L’Avant Scène Opéra CLIII), 90) that Il crociato in Egitto forms part of what he calls the genre troubadour. For the art-historical background to this subject, see François Pupil, Le Style troubadour, ou La Nostalgie du bon vieux temps (Nancy: Presses Universitaires de Nancy, 1985). 29 For a complete account of the details and context of the Weber-Castil-Blaze Euriante in 1831, see Everist, ‘Translating Weber’s Euryanthe: German Romanticism at the Dawn of French Grand Opéra’, Revue de Musicologie 87 (2001), 67–105. 3 0 Corinne Schneider, ‘Le Weber français du Théâtre-Lyrique: enjeux et modalités de la réception de l’opéra allemand traduit sous le Second Empire’ (PhD diss., Université François-Rabelais, Tours, 2002), 229–231.

342  Castil-Blaze and the reception of Weber We regret that the act iii ballade … that the Odéon did not omit ten years ago, has been suppressed in the new arrangement of the score. We were also deprived of the charming duet between the two lovers, ‘Non, plus d’alarmes!’, which found a perfect place at the end of Robin des bois, although it had been borrowed from the score of Euryanthe.31 This quotation suggests two things: first, that the modification of the score of Der Freischütz was a process that continued throughout Castil-Blaze’s lifetime, with additional changes being made every time the work was produced, and second, contrary to Berlioz’s view, that Castil-Blaze’s changes to the work in 1824 were more or less universally endorsed explicitly by critics and implicitly by the audiences both at the Odéon and the Opéra-Comique. In other words, the claims that Castil-Blaze made about the essential nature of these changes proved to be almost exactly correct. And when we turn to the production of Robin des bois at the Théâtre-Lyrique twenty years later, we see this view inscribed even more clearly.32 Perhaps more has been written about Berlioz’ arrangement of Der ­Freischütz than any other aspect of Weber reception in nineteenth-century France. And presumably this is because of Berlioz’s authorial imprimatur, rather more than the impact of his arrangement. Le Freyschütz enjoyed 78 performances between 1841 and 1853, as opposed to the 60 of Robin des bois at the Opéra-Comique in 1835 alone, the several hundred of Robin des bois at the Odéon in the 1820s and the 150 of Robin des bois at the Théâtre-Lyrique between 1855 and 1866 before the premiere of the Trianon-Gautier production. Berlioz’s arrangement has been specially privileged in other ways as a result of his authorship. For the revival of the Berlioz-Weber Freischütz at the Opéra-Comique in 2012, its website described Berlioz’s endeavours: The coherence of his [Berlioz’s] work, accepted after mature reflection, illuminates the familiarity of Weber with our repertory of opéra comique as well as the extraordinary posterity of his work in France.33 This is wrong on several counts. As already suggested, the various productions of Castil-Blaze’s Robin des bois were probably much more significant than those of Berlioz. Furthermore, the quotation perpetuates an entirely misleading view of Berlioz’s arrangement of Der Freischütz as possessing

31 ‘Nous regrettons que la ballade du 3e acte … que l’Odéon n’a pas négligé il y a dix ans, ait été supprimée dans l’arrangement nouveau de la partition. On nous a aussi privé du charmant duo entre les deux amants, ‘Non, plus d’alarmes!’ qui trouvait parfaitement sa place dans Robin des bois, bien qu’il ait été emprunté à la partition d’Euryanthe’ (Le ménestrel, 25 January 1835). 32 Schneider, ‘Le Weber français’, 255–258. 33 ‘L’intégrité de son travail, accepté après mûre réflexion, éclaire la familiarité de Weber avec notre répertoire d’opéra-comique ainsi que l’extraordinaire postérité de son œuvre en France’ (www.opera-comique.com, consulted 5 June 2014).

Castil-Blaze and the Reception of Weber  343 an almost moral advantage over Castil-Blaze’s when it argues that Berlioz exercised ‘an uncompromising respect, hardly yet established’ (‘un respect sourcilleux guère encore de mise’). But in truth, both arrangements were aiming to do similar things: to bring Weber’s romantische Oper into line with prevailing aesthetic and institutional imperatives. 1824 – with its concern for the romanticism of Sir Walter Scott, for example – was very different to 1841, and the Odéon during the Restoration was a very different institutional proposition to the Académie Royale de Musique in the July Monarchy. But the pressures were equal: where Castil-Blaze needed to move the action from Bohemia to Yorkshire and to add in a love duet taken from another opera, Berlioz had to add recitatives and to adapt some of Weber’s other music for the ballet score that he wrote. It is difficult to escape the view that, in terms of the double standards applied to Berlioz and Castil-Blaze, a good deal of the evidence has been ignored at best and slanted at worst. Nowhere is the application of this double standard clearer than in the edition of Berlioz’s additions to Le Freyschütz in the New ­Berlioz Edition.34 There is another important way in which the Berlioz-Weber arrangement of 1841 needs to be understood that marks it out from Robin des bois and other arrangements of Der Freischütz. Berlioz’s arrangement of Der ­F reischütz was not a work similar to Les Huguenots, La juive or any other of the five-act grands opéras on historical themes that dominated the stage of the Académie Royale de Musique during the July Monarchy. Le Freyschütz falls fairly and squarely into the genre of petit opéra that was critical to supporting the repertory of the Académie Royale de Musique throughout the nineteenth century.35 Two foreign works functioned as petits opéras in exactly the same as Le comte Ory, Le philtre and other parts of the repertory: the translation of Donizetti’s Lucia di Lammermoor and Berlioz’s arrangement of Der Freischütz (Table 10.4). Ironically, Berlioz’s arrangement of Der Freischütz suffered exactly the same fate as had Benvenuto Cellini three years earlier – before Berlioz withdrew the work, it had been commandeered by the management of the Académie Royale de Musique to accompany the ballets La gipsy and Le diable boiteux. Le Freyschütz was used in exactly the same way. After a run of only twelve performances on its own – and this was entirely normal for petit opéra – it was paired with Giselle, La tarentule, La peri, La sylphide, Lady Henriette, Eucharis, Les danseuses, Le diable à quatre – all before 1850. Again, this symbiotic relationship with ballet that characterised the history of Le Freyschütz means that any impact that might be claimed for the work has to be understood in comparison with Robin des bois which was always conceived as a freestanding opera at the Odéon.

34 Ian Rumbold, ed., Arrangements of Works by Other Composers (II), Hector Berlioz: New Edition of the Complete Works 22b (Kassel etc.: Bärenreiter, 2004). 35 See Chapter 2.

344  Castil-Blaze and the reception of Weber Table 10.4  Petit opéra during the Restoration and July Monarchy Composer

Librettist

Petit opéra

Gioachino Rossini

Eugène Scribe et Charles-Gaspard Delestre-Poirson Eugène Scribe

Le comte Ory

1828

2

1830

2

Eugène Scribe

Le dieu et la bayadère Le philtre

1831

2

Eugène Scribe

La xacarilla

1839

1

Eugène Scribe

Le comte de Carmagnola

1841

2

1841

3

1842 1842

2 2

Daniel-FrançoisEsprit Auber Daniel-FrançoisEsprit Auber Marco Aurelio, comte Marliani Ambroise Thomas Carl-Maria von Weber, arr. d’Hector Berlioz Fromental Halévy Pierre-LouisPhilippe Dietsch Fromental Halévy Gaetano Donizetti

Friedrich Adolf Ferdinand Freiherr von Flotow Adolphe Adam François Benoist

Friedrich Kind, arr d’Émilien Pacini et Berlioz Théodore Anne Paul Foucher et Bénédict-Henry Révoil Jules-Henry Vernoy de Saint-Georges Salvadore Cammarano, arr. Alphonse Royer et Gustave Vaëz Jules-Henry Vernoy de Saint-Georges Hippolyte Lucas Germain Delavigne

Le guérillero Le vaisseau fantôme Le lazzarone

Production étrangère

Le Freyschütz (Der Freischütz)

Date Actes

1844 2

Lucie de 1846 4 Lammermoor (Lucia di Lammermoor) L’âme en peine 1846 2

La bouquetière L’apparition

1847 1 1848 2

It was inevitable that reviews of the Berlioz’s 1841 arrangement would draw comparisons between his version and Castil-Blaze’s Robin des bois. Henri Blanchard drew attention, in his review in La Revue et Gazette musicale de Paris, to the literary debate between those who saw greater qualities in the libretto to Robin des bois than to Pacini’s translation of Kind,36 and Eugène Ponchard – in Le ménestrel – recognised both Pacini’s skill but also noted that ‘since today we can better perceive the various nuances that separate our habits from German ones, one would be almost tempted to grant an indemnity to the libretto of Robin des bois’.37 Similarly, although Berlioz’s 36 La Revue et Gazette musicale de Paris, 13 June 1841. 37 ‘aujourd’hui que l’on saisit mieux les diverses nuances qui séparent nos mœurs des habitudes germaniques, on serait presque tenté d’accorder un bill d’indemnité au poëme de Robin des bois’ (Le ménestrel, 13 June 1841).

Castil-Blaze and the Reception of Weber  345 recitatives were thought to have slowed the action somewhat, he was c­ redited by both critics with having acted, in Ponchard’s words, ‘with the most praiseworthy restraint. He sought to align himself with the style and colour of German music, without striving for effect, without any thought of eccentric supplement’.38 Berlioz’s friends rallied round, and Théophile Gautier read Pacini’s and Berlioz’s introduction to the printed libretto, took it at face value and simply paraphrased Berlioz’s view that ‘genius should only be touched with respectful hands like the priest when he holds the host. Each note is sacred, and the words which have awakened the inspiration of the master should not be changed lightly’.39 These were perhaps predictable sentiments for someone whose ballet Giselle was scheduled for performance at the same institution thirteen days later, and his review unsurprisingly challenged none of the claims made by the arrangers. Berlioz’s detractors were better prepared and had better arguments. Blaze de Bury – it must be remembered that he is Castil-Blaze’s son, so as parti pris as Gautier – wrote that With what pleasure does Berlioz give himself over to his fantasy, never finding enough space between two numbers to spread out completely at his will his languorous and fastidious melody, surrounding this lively and enthralling music with the bonds of his confused thought, tiring the audience, exhausting the singers, and throwing a lead mantle over this fantastic and marvellous conception.40 This is a tiny fragment of a much longer but no less hostile review. It is, however, gentle in comparison with the one published by Louis-Amadée Achard in La France musicale under the pseudonym B[aron] de Grimm. His attitude to the relative merits of Castil-Blaze’s Robin des Bois and Berlioz’s Le Freyschütz was clear: Monday night only served to convince a large number of people, that, furthermore, were doubtful about it for a long time, that M. Castil-Blaze was a man full of spirit and taste, who had perfectly understood that what might be a masterpiece for the Germans might not be one for the 38 ‘avec la plus louable abnégation. Il a cherché à se rapprocher du style et du coloris de la musique allemande, sans viser à l’effet, sans aucune arrière-pensée de surérogation excentrique’ (ibidem). 39 ‘On ne doit toucher le génie qu’avec des mains respectueuses comme le prêtre quand il tient l’hostie. Chaque note est sacrée, et les paroles qui ont éveillé l’inspiration du maître ne doivent pas être changées à la légère’ (La presse, 15 June 1841). 40 ‘Avec quelle complaisance [Berlioz] se laisse aller à sa fantaisie, ne trouvant jamais l’espace assez vaste entre deux morceaux pour étaler tout à son aise sa traînante et ­fastidieuse mélopée, entourant cette musique vive et saisissante des liens de sa pensée confuse, fatiguant l’auditoire, écrasant les chanteurs, et jetant un manteau de plomb sur cette fantastique et merveilleuse conception’ (La revue des deux mondes, 15 June 1841).

346  Castil-Blaze and the reception of Weber French, and that one could, by means of certain modifications, again reignite the merit of a work. In effect, not one of the beauties of Le Freyschutz was left out of Robin des bois, many longeurs had disappeared…, the libretto had been rendered intelligible, and all the changes to the numbers had been made in the happiest manner.41 Achard goes on at length to praise every element of Robin des bois and to decry every part of Berlioz’s arrangement of Der Freischütz. In the light of such criticism, it is hardly surprising that Berlioz himself wrote a long article in his feuilleton in Le journal des débats that is long on self-congratulation and short on critical precision.42 The claims to integrity and to the intentions of the composer are repeated here among a repetition of Berlioz’s futile attempts to meet Weber in 1826. It is, however, instructive to place Berlioz’s article alongside Castil-Blaze’s text – both in Le journal des débats, of course; both articles act as apologiae for their respective attempts to naturalise Weber’s opera.43 Berlioz’s arrangement of Der Freischütz ran as a petit opéra until 1853, but the opera was not to be heard again at the Théâtre national de l’Opéra – as the Opéra was then called – until 1870. Almost immediately after the premiere – in 1841 – of Berlioz’ arrangement of Der Freischütz at the Académie Royale de Musique, a new peripatetic troupe arrived in Paris for a disastrous season, despite its plan to repeat the success of Röckel’s troupe a decade earlier. Alongside German versions of Gluck’s Iphigénie en Aulide and Iphigénie en Tauride, of Spontini’s La vestale, Mozart’s Le nozze di Figaro, Die ­Zauberflöte and Don Giovanni, the director Auguste Schumann aimed to include performances of Fidelio, Der Freischütz and Euryanthe, as well such more recent works as Marschner’s Hans Heiling and Der Templer und die Jüdin, Conradin Kreutzer’s Libussa and Das Nachtlager von Grenada, Franz Gläsner’s Des Adlers Horst and Lortzing’s Hans Sachs and Zar und Zimmermann.44

41 ‘La soirée de lundi n’a servi qu’à convaincre un grand nombre de personnes qui, du reste, s’en doutaient depuis longtemps, que M. Castil-Blaze était un homme plein d’esprit et de goût, qui avait parfaitement compris que ce qui était un chef-d’œuvre pour des Allemands, pourrait très bien ne pas l’être pour des Français, et qu’on pouvait, à l’aide de certaines modifications, rehausser encore le mérite d’un ouvrage. En effet, pas une des beautés du Freyschutz n’avait été omise dans le Robin-des-bois, bien de longueurs avaient disparu…., la pièce avait été rendue intelligible, et toutes les transpositions de morceaux avaient été faites de la manière la plus heureuse’ (La France musicale, 13 June 1841). For the identification of Grimm as Achard, see Emmanuel Reibel, L’écriture de la critique musicale au temps de Berlioz (Paris: Champion, 2005), 302. 42 Le journal des débats, 13 June 1841. 43 Castil-Blaze’s article on his own arrangement of Der Freischütz is found in the Journal des débats, 28 September 1824. 4 4 Castil-Blaze, ‘Théâtre Allemand de Paris’, La France musicale, 10 April 1842. See also Paris, Archives nationales (hereafter F-Pan) F21 1116 for Die Vestalin.

Castil-Blaze and the Reception of Weber  347 The projects described by Castil-Blaze omitted most importantly one of the most important works in the repertory of Schumann’s troupe: Spohr’s ­Jessonda. The German theatres lasted scarcely more than two weeks (Table 10.5). The planned performances of the two Gluck Iphigénie operas were probably stopped by the Minister of the Interior because both technically formed part of the repertory of the Académie Royale de Musique, although he seemed to have missed the fac