You are on page 1of 395

The Cambridge Companion to Ballet

Ballet is a paradox: much loved but little studied. It is a beautiful fairy


tale; detached from its origins and unrelated to the men and women
who created it. Yet ballet has a history, little known and rarely
presented. These great works have dark sides and moral ambiguities,
not always nor immediately visible. The daring and challenging
quality of ballet as well as its perceived ‘safe’ nature is not only one of
its fascinations but one of the intriguing questions to be explored in
this Companion. The essays reveal the conception, intent and
underlying meaning of ballets and re-create the historical reality in
which they emerged. The reader will find new and unexpected
aspects of ballet, its history and its aesthetics, the evolution of plot
and narrative, new insights into the reality of training, the choice of
costume and the transformation of an old art in a modern world.
The Cambridge Companion to

BALLET
............

edited by
Marion Kant
c a m b r i d g e u n i ve r s i t y p re s s
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9780521539869


C Cambridge University Press 2007

This publication is in copyright. Subject to statutory exception


and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.

First published 2007

Printed in the United Kingdom at the University Press, Cambridge

A catalogue record for this publication is available from the British Library

ISBN 978-0-521-83221-2 hardback


ISBN 978-0-521-53986-9 paperback

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

List of illustrations [page vii]


Foreword Ivor Guest [xi]
Notes on contributors [xiv]
Chronology [xxi]

Introduction Marion Kant [1]

Part I r From the Renaissance to the baroque: royal power and


worldly display
1 The early dance manuals and the structure of ballet: a basis for Italian,
French and English ballet Jennifer Nevile [9]
2 Ballet de cour Marina Nordera [19]
3 English masques Barbara Ravelhofer [32]
4 The baroque body Mark Franko [42]

Part II r The eighteenth century: revolutions in technique and spirit


5 Choreography and narrative: the ballet d’action of the
eighteenth century Dorion Weickmann [53]
6 The rise of ballet technique and training: the professionalisation of
an art form Sandra Noll Hammond [65]
7 The making of history: John Weaver and the Enlightenment
Tim Blanning [78]
8 Jean-Georges Noverre: dance and reform Judith Chazin-Bennahum [87]
9 The French Revolution and its spectacles Inge Baxmann [98]

Part III r Romantic ballet: ballet is a woman


10 Romantic ballet in France: 1830–1850 Sarah Davis Cordova [113]
11 Deadly sylphs and decent mermaids: the women in the Danish romantic
world of August Bournonville Anne Middleboe Christensen [126]
12 The orchestra as translator: French nineteenth-century
ballet Marian E. Smith [138]
13 Russian ballet in the age of Petipa Lynn Garafola [151]

[v]
vi Contents

14 Opening the door to a fairy-tale world: Tchaikovsky’s ballet


music Thérèse Hurley [164]
15 The romantic ballet and its critics: dance goes public
Lucia Ruprecht [175]
16 The soul of the shoe Marion Kant [184]

Part IV r The twentieth century: tradition becomes modern


17 The ballet avant-garde I: the Ballets Suédois and its modernist
concept Erik Näslund [201]
18 The ballet avant-garde II: the ‘new’ Russian and Soviet dance in the
twentieth century Tim Scholl [212]
19 George Balanchine Matilde Butkas [224]
20 Balanchine and the deconstruction of classicism Juliet Bellow [237]
21 The Nutcracker: a cultural icon Jennifer Fisher [246]
22 From Swan Lake to Red Girl’s Regiment: ballet’s sinicisation
Zheng Yangwen [256]
23 Giselle in a Cuban accent Lester Tome [263]
24 European ballet in the age of ideologies Marion Kant [272]

Notes [291]
Bibliography and further reading [311]
Index of persons [331]
Index of ballets [339]
Subject index [342]
List of illustrations

1 Maestro dei Tornei di Santa Croce, cassone panel. La magnanimità di


Scipione, 1460. London, Victoria and Albert Museum, Inventory No:
5804.1859 [page 10]
2 Guillemine la Quinteuse, Fée de la musique. Ballet des fées de la forêt de
Saint-Germain (1625), drawing by Daniel Rabel, Musée du Louvre,
Inv. 32 604. Published in Marie Françoise Christout, Le ballet de cour au
XVII° siècle (Genève: Minkoff, 1987), p. 76 [22]
3 Androgynes, Ballet de la douairière de Billebahaut (1626), drawing by
Daniel Rabel, Musée du Louvre, Inv. 32 692. Published in Marie Françoise
Christout, Le ballet de cour au XVII° siècle (Genève: Minkoff, 1987), p. 148
[23]
4 Faun from Lully’s Le triomphe de l’amour, 1681. Malkin Collection
Pennsylvania State University [26]
5 Cover page, Coelum Britannicum 1634. Malkin Collection Pennsylvania
State University [35]
6 List of masquers, Coelum Britannicum, 1634. Malkin Collection
Pennsylvania State University [36]
7 Catherine Turocy, Artistic Director, The New York Baroque Dance
Company (photo Lois Greenfield) [44]
8 Mark Franko in Le Marbre tremble (photo Midori Shinye) [45]
9 Jason et Medée by Jean-Georges Noverre, music by Jean-Joseph Rodolphe
1781. Francesco Bertoluzzi (artist), Gaëtan Vestris as Jason. C V&A
Images/Victoria and Albert Museum, London [58]
10 Dancers from the Australian Ballet as hens in La Fille mal gardée, 1970s.
National Library of Australia. W. F. Stringer Collection (photo Walter
Stringer) [61]
11 ‘Modern grace, – or – the operatical finale to the ballet of Alonzo e Caro’
(Rose Didelot; Charles Louis Didelot; Madame Parisot) by James Gillray,
published by Hannah Humphrey, 1796. C National Portrait Gallery,
London [64]
12 The five positions of the feet, as depicted by Feuillet in his Chorégraphie,
1700 [67]
13 An example of the opening phrase of a theatrical duet, as notated by
Feuillet in his Recueil de Dances, 1700 [68]
14 From Carlo Blasis, The Code of Terpsichore, 1828 [76]
15 Festival of Federation 14 July 1790, copper engraving by Paul Jakob
[vii] Laminit, no date [99]
viii List of illustrations

16 The Festival of the Supreme Being, 8 June 1794 at the Champs de Mars in
Paris; painting by (Pierre) Antoine Demachy, no date. Paris, Musée
Carnavalet [108]
17 Carlotta Grisi as Giselle, Paris 1841, from Les Beautés de l’opéra, ou
Chefs-d’oeuvre lyriques, illustrés par les premiers artistes de Paris et de
Londres sous la direction de Giraldon, avec un texte explicatif rédigé par
Théophile Gautier, Jules Janin et Philaréte Chasles (Paris: Soulié, 1845).
Rare Books, University of Pennsylvania [115]
18 Mademoiselle Marquet in Le Dieu et la Bayadère, opera-ballet 1830.
Malkin Collection Pennsylvania State University [118]
19 and 20 Cut-out paper dolls, depicting Marie Taglioni and Fanny Elssler in
contemporary fashionable frocks, 1830s. Malkin Collection Pennsylvania
State University (formerly in the collection of Lincoln Kirstein) [120,121]
21 The Dream of a Ballerina, Léon F. Comerre, late nineteenth century.
Malkin Collection Pennsylvania State University [124]
22 The Royal Danish Ballet, Season 2003/4. August Bournonville, La
Sylphide. Gudrun Bojesen, Thomas Lund (photo: Martin Myntskov
Rønne) [127]
23 The Royal Danish Ballet, Season 2003/4. August Bournonville, La Sylphide,
rehearsal (photo: Martin Myntskov Rønne) [128]
24 The Royal Danish Ballet, Season 2004/5. August Bournonville, Far from
Denmark. Mads Blangstrup, Marie-Pierre Greve (photo: Martin Myntskov
Rønne) [131]
25 The Royal Danish Ballet, Season 2004/5. August Bournonville, scene from
Napoli, Act iii (photo: Martin Myntskov Rønne) [135]
26 Gala performance at Peterhof 11 July 1851, from Geirot’s Opisanie
Petergofa, 1868 (Carlotta Grisi and Jules Perrot in The Naiade and the
Fisherman) Malkin Collection Pennsylvania State University [154]
27 Swan Lake, St Petersburg 1910, choreography by Lev Ivanov and Marius
Petipa, music by Pyotr Ilyich Tchaikovsky. C V&A Images/Victoria and
Albert Museum, London [160]
28 Scene from Swan Lake, Bolshoi Ballet Moscow, 1959 (uncredited
photographer) [162]
29 Caricature by Cham from Le Charivari, 1 August, 1858. By permission of
the British Library [179]
30 Fanny Elssler’s pointe shoes. Austrian Theatre Museum Vienna [191]
31 Pointe shoes 1950–2005 from the Soviet Union, Bulgaria, Germany, Great
Britain and the United States [195]
32 Skating Rink, Les Ballets Suédois, 1931 (photo Isabey) from Les Ballets
Suédois dans l’art contemporaine (Paris: Trianon 1931). Malkin Collection
Pennsylvania State University [207]
ix List of illustrations

33 Within the Quota, Les Ballets Suédois, 1931 (photo Isabey) from Les Ballets
Suédois dans l’art contemporaine (Paris: Trianon 1931). Malkin Collection
Pennsylvania State University [209]
34 Anna Pavlova (photo Mishkin studio NY), from Troy and Margaret West
Kinney (“The Kinneys”), The Dance: Its Place in Art and Life (New York:
Frederick A. Stokes Company 1914, 1924) [215]
35 Vaclav Nijinsky in the title role of Petrushka, 1911, choreography by
Mikhail Fokine. Kiev State Historical Museum [217]
36 Pennsylvania Ballet Principal Dancer Riolama Lorenzo with Soloist Philip
Colucci in The Prodigal Son, choreography by George Balanchine (photo
Paul Kolnik) [229]
37 Suzanne Farrell and Peter Martins in Jewels – Diamonds, choreography by
George Balanchine (photo Paul Kolnik) [235]
38 Giorgio de Chirico, costume for a male guest in Le Bal, 1929. Collection of
the Wadsworth Atheneum [238]
39 Alice Nikitina, Felia Doubrovska, Lubov Tchernicheva, Serge Lifar in
Apollon musagète (Apollo, Leader of the Muses), 1928. Choreography by
George Balanchine C The George Balanchine Trust. Music by Igor
Stravinsky. Courtesy of New York City Ballet Archives, Ballet Society
Collection [242]
40 The American department store chain Lord & Taylor announces its
presentation of The Nutcracker in the 1974 catalogue. Malkin Collection
Pennsylvania State University [251]
41 Pennsylvania Ballet Company Member Heidi Cruz in The Nutcracker,
2005, choreography by George Balanchine (photo Paul Kolnik) [254]
42 Escape Scene from the ballet Red Girl’s Regiment [259]
43 Shooting Scene from the ballet Red Girl’s Regiment [260]
44 Alicia Alonso in Act i of Giselle (photo Tonatiuh Gutierrez) [264]
45 Alicia Alonso in Act ii of Giselle (photo Tito Alvarez) [265]
46 Galina Ulanova in Act ii of Giselle, Bolshoi Ballet Moscow, 1954
(uncredited photographer) [281]
47 Maja Plisetskaya as Odette in Swan Lake, Bolshoi Ballet Moscow, 1966
(uncredited photographer) [282]
48 Maina Gielgud in Forme et ligne (Squeaky door) by Maurice Béjart.
Australian Ballet 1974. W. F. Stringer Collection, National Library of
Australia. (photo Walter F. Stringer) [287]
49 Felia Doubrovska as Bride in Les Noces, choreography by Bronislava
Nijinska (photo Abbe), from Walter Archibald Propert, The Russian Ballet
1921–1929 (London: John Lane The Bodley Head, 1931) [288]
50 Three Atmospheric Studies by William Forsythe, the Forsythe Company
2005. (photo Joris-Jan Bos) [289]
Foreword

Ballet is a theatre art that, by virtue of its origins, is essentially and incontro-
vertibly European. Those origins are, in large measure, to be sought in the
Italian courts of the Renaissance in the fifteenth century, where it developed
as a means of displaying the splendour and power of the ruling prince. A
century or so later it crossed the Alps in the marriage train, so to speak, of
Catherine de’ Medici, the chosen bride of King Henri II of France, becoming
a dominant feature in the entertainments of the French court for more than
a century until well into the reign of Louis XIV. By that time professional
dancers were already being employed to add variety and brilliance through
a technique that far exceeded that of even the most talented of the courtiers.
However, halfway through Louis XIV’s reign the court ballet went into a
sudden decline, not so much on account of the king’s growing corpulence
as through the increasing demands on his treasury of the wars in which
France then became embroiled. Providentially a far more suitable and last-
ing future for the ballet was then provided by the king himself in creating
the Académie de Musique, which was to be the forerunner of the institution
now known as the Paris Opéra. Here professional dancers found an arena
from the very outset, and dance as spectacle was to play a conspicuous role
in the creation of what we now recognise as the art of classical ballet. At first
it took the form of an adjunct to opera, as in the opera-ballets of Rameau,
but from the mid-eighteenth century it became an independent theatre art,
in which the stage action was conveyed by the dancers themselves in pan-
tomime. This was one of the great theatrical turning-points that marked the
Age of the Enlightenment.
While Paris continued to be regarded as the prominent centre of this
new art form, ballet soon took root elsewhere in Europe. Italy, where the
infant art had been nurtured, became a fertile field as many opera houses
throughout the peninsula adopted ballet as a respected adjunct to the opera.
In Milan and Naples major ballets were being produced on the stages of those
cities’ celebrated opera houses, based upon plots that required powerful gifts
of pantomime in those players responsible for the dramatic roles. By the
end of the eighteenth century ballet was generally recognised throughout
Europe as a significant theatre art, and one with its own philosopher in the
distinguished figure of Jean-Georges Noverre, whose Letters on Dancing are
to this day still revered as a classic.
France retained its ascendancy notwithstanding the cataclysm of the
[xi] Revolution, a period that saw the emergence of the formidable figure of
xii Ivor Guest

Pierre Gardel, who was to dominate French ballet for some forty years.
When the Revolution and the Napoleonic wars had receded into history,
Paris was still regarded as the fountain-head of the theatrical dance, and
it was there within the august walls of the Opéra that the conflict between
the opposing trends of classicism and romanticism in the art of ballet was
finally and unequivocably resolved in the latter’s favour. Under the banner
of romanticism the choreographers Taglioni and Perrot produced those two
ballets that are treasured today as lasting classics, La Sylphide and Giselle.
The Paris ballet continued to enjoy a dominance that would remain virtually
unchallenged until the last decades of the nineteenth century, when the
world became increasingly aware that ballet had taken root in most fertile soil
in St Petersburg, underwritten by the vast wealth of the tsars but preserving
nevertheless a vital French connection in the person of the Marseilles-born
ballet master Marius Petipa.
In the early 1900s the first rumblings were felt of the revolution that was
to come, and it was in the last few years before the outbreak of the First World
War that Europe was given its first taste of the balletic riches that Russia had
to offer. This came about through a privately sponsored company of dancers
from the Imperial Theatres organised and directed by Serge Diaghilev that
conquered Paris literally overnight in the summer of 1909, presenting not
extracts from the works of Petipa, but a programme mainly produced by a
younger choreographer, Mikhail Fokine. This extraordinary enterprise was
continued, with a break of a few years resulting from the war, until 1929 when
Diaghilev died. The consequences of his demise was to prove the permanence
of the legacy left by that extraordinary company in the course of just two
decades – an eye-blink in the context of history – for the disappearance
of its guiding spirit let loose a younger brood of choreographers – notably
Nijinska, Massine, Balanchine and Lifar – to propagate a new vision of ballet
throughout Europe and America.
Not even the great conflict that then raged across the world in the mid-
century would hinder this process. In many of the lands that were ravaged
in those dark years, ballet provided a momentary release from the brutal
consequences of aerial bombardment, invasion and occupation, gaining
many converts in all ranks of society to the allurements of the dance. Such
seeds have continued to bear fruit ever since, and today, more than ever
before, the dance is regarded as a vital and major part of our artistic heritage.
In its theatrical manifestations, most notably the great heritage of classical
ballet but also in the multi-faceted complexity of freer disciplines, ballet
has earned its place as a major component of the arts of spectacle in the
modern world. Furthermore, it has been accepted as a subject for serious
scholarship, revealing, as it does, the possibilities of human movement as
a means of expression no less valid than human speech. Today a vast body
xiii Foreword

of literature, recorded reminiscences, musical sources and other material


of record is being assembled on the subject of dance in all its multiple
forms, opening up new vistas for study and research, and requiring works
of reference, such as is offered in this volume, to guide both the scholar and
the devotee.
These are the rich fields over which I have been privileged to wander these
past sixty years. In the beginning of my historical endeavours I luxuriated
in that of Second Empire Paris, unearthing the history of a period of French
ballet that had received scant attention from earlier scholars. The Library
of the Paris Opéra became, for me, virtually a second home on my regular
visits to Paris, and over the years my research wanderings were to lead me
to other fields, some that had seen earlier harvests and others that had been
more recently cultivated.
My quest to unravel the rich strands of the history of ballet has thus
led me to many libraries and archives, and I have thereby become deeply
conscious of my debt to the volumes of reference books which repose on
open shelves as friendly guides to those in search of information. I therefore
welcome – most specially, I may add, since I am a Cambridge graduate – this
new companion, which I am sure will find an honoured place not only on
the reference shelves of libraries, both specialised and general, but also on
those of countless devotees of an art that today holds so many in its thrall.

Ivor Guest
Notes on contributors

Inge Baxmann studied romance languages and literature at the universities of


Bochum and Paris. From 1981 to 1986 she was assistant professor at the Insti-
tute of French Literature of the Technical University of Berlin. In 1987 she com-
pleted her doctoral dissertation, “Die Feste der Französischen Revolution. Insze-
nierung von Gesellschaft als Natur”. She received several post-doctoral fellowships
and became a fellow of the prestigious Alexander-von-Humboldt-Stiftung. 1997
Habilitation at the Humboldt University Berlin with Mythos Gemeinschaft. Körper-
und Tanzkulturen in der Moderne. She holds a chair for Theatre Studies at Leipzig
University.
Her next book will be a study of Cultures in Movement. Life of the Transnation.
Kulturen in Bewegung: Leben in der Transnation (Munich: Fink-Verlag).
Juliet Bellow completed her dissertation, “Clothing the Corps: How the Avant-Garde
and the Ballets Russes Fashioned the Modern Body”, in 2005 at the University
of Pennsylvania. Her publications include “Reforming Dance: Auguste Rodin’s
‘Nijinsky’ and Vaslav Nijinsky’s ‘L’Après-Midi d’un Faune”’, in a special issue of
the Cantor Arts Center Journal, New Studies on Rodin.
She teaches art history at the University of Pennsylvania in Philadelphia and
at Hunter College New York.
Judith Chazin-Bennahum performed in many dance companies, including dancing
with Agnes de Mille, the Robert Joffrey Ballet Company and the Metropolitan
Opera Ballet Company as Principal Soloist. She also danced with the Santa Fe
Opera and toured Europe with Igor Stravinsky as the conductor and was invited by
George Balanchine to join the New York City Ballet. She received her Doctorate in
Romance Languages at the University of New Mexico and is the author of Dance in
the Shadow of the Guillotine (1988) a book on ballet during the French Revolution
and published The Ballets of Antony Tudor in 1995. The Lure of Perfection: Fashion
and Ballet 1780–1830 was published by Routledge in 2004. She is preparing a
volume, Teaching Dance Studies which will have essays on the pedagogy of teaching
dance courses in a university setting.
Bennahum has choreographed for the Santa Fe Opera, the Southwest Ballet
Company, the UNM Opera Studio and annually for the UNM Dance ensemble.
She re-created Jean-Georges Noverre’s ballet Medea (1780), which was filmed for
video and is now being distributed by Princeton Books.
Tim Blanning teaches at Cambridge University. His research interests are focused
on the history of continental Europe in the period 1660–1914. His early work
concentrated on the Holy Roman Empire and the Habsburg Monarchy during the
eighteenth century and he retains a strong interest in this area. During the 1980s
and 1990s his focus moved westwards to France during the Revolution, especially
to its foreign policy and its interaction with the rest of Europe. Most recently,
he has concentrated on the high culture of Europe and its relationship to state
[xiv]
xv Notes on contributors

power, which resulted in his prize-winning study the Culture of Power and Power of
Culture 1660–1789 (2002). He is currently working on why music progressed from
subordinate status in the early modern period to its present position of supremacy
among the creative arts. He is the general editor of The Oxford History of Modern
Europe (2002) and of The Short Oxford History of Europe, editing personally the
volumes on the eighteenth and nineteenth centuries in the latter series (2000). He
has been a Fellow of the British Academy since 1990.
Matilde Ann Butkas holds Master of Music Degrees in piano performance and piano
pedagogy from the University of Illinois, and is currently recording the complete
keyboard works of Johann Mattheson on harpsichord. She is a doctoral student
in musicology at the University of Oregon.
Anne Middelboe Christensen, born in Denmark, is a dance critic. She holds an MA
in Danish literature and theatre history from the University of Copenhagen, on
the subject of Danish dance and ballet criticism from 1771 to 1862. She also
studied at York University in Toronto. She has worked as a journalist for various
magazines since 1985, specialising in dance criticism and dance writing, and since
1995 writes for the daily Dagbladet Information. Since 1995 she has been assistant
professor at the University of Copenhagen. She is the author of Diversions of the
Royal Danish Ballet. Interviews with the Dancers about the Bournonville Tradition
(Schønberg, Copenhagen, 2002) and Backbone. Dansescenen, Denmark 1993–2003
(2003). She also has worked and works as a dramaturgue for choreographers Itzik
Galili (Holland) and Tim Rushton (Danish Dance Theatre).
Sarah Davies Cordova. Her interdisciplinary work in French and Francophone cul-
tures together with dance studies enables her to research texts of post-revolutionary
France, and colonial and postcolonial eras which incorporate the politics of (self)
representation, gender concerns, and geographical, topographical, diasporic and
historical (dis)placements of persons in terms of corporeality, movement styles as
well as bodily memory and traces of conforming, and resistance.
She has published Paris Dances: Textual Choreographies of the Nineteenth-
Century French Novel (1999) and a number of articles on nineteenth-century
ballet and literature; as well as on works by women authors from Guadeloupe,
Haiti and Algeria. She teaches French language and literatures at Marquette Uni-
versity, although she is currently the resident director for the Marquette University
service learning programme at the Desmond Tutu Peace Trust and in conjunction
with the University of the Western Cape in Cape Town, South Africa.
Mark Franko received his Ph.D. in French from Columbia University and danced
professionally before becoming a dance historian, theorist and choreographer.
He has taught at Princeton University, New York University, Columbia/Barnard,
Purdue University, Paris 8, the University of Nice, Montpellier 3, the Catholic
University of Leuven and is currently Professor of Dance and Chair of Theatre
Arts at the University of California, Santa Cruz.
He is the author of Excursion for Miracles: Paul Sanasardo, Donya Feuer, and
Studio for Dance (1955–64), The Work of Dance: Labor, Movement, and Identity in
the 1930s (CHOICE magazine “Outstanding Academic Title” for 2003), Dancing
Modernism/Performing Politics (1996 de la Torre Bueno prize Special Mention),
Dance as Text: Ideologies of the Baroque Body (1993; published in France by Editions
xvi Notes on contributors

Kargo and forthcoming in Italy by L’Epos) and The Dancing Body in Renaissance
Choreography (1986).
His choreography for NovAntiqua, the company he founded in 1985, has
been produced at Lincoln Center Out-of-Doors Festival, the Berlin Werkstatt
Festival, the Getty Center, the Montpellier Opera, Toulon Art Museum and in
many national venues.
Jennifer Fisher is Assistant Professor of Dance at the University of California Irvine
and teaches dance history, fieldwork, philosophy, aesthetics and criticism. She
holds a master’s degree in Dance from York University in Toronto and a Ph.D. in
Dance History and Theory from the University of California, Riverside. A former
dancer and actor, she has previously taught at York University and Pomona Col-
lege. Her book, Nutcracker Nation: How an Old World Ballet Became a Christmas
Tradition in the New World, was published by Yale University Press in 2003. She is a
regular contributor of dance criticism for the Los Angeles Times; other publications
include Dance Research Journal, Women & Performance, Queen’s Quarterly, the
International Dictionary of Ballet, Stuttgarter Zeitung, the Encyclopedia of Homo-
sexuality, Dance and Society in Canada, and several dance periodicals. She is a
contributing editor for the Society of Dance History Scholars Newsletter and serves
on that organisation’s Editorial Board (working in association with the University
of Wisconsin Press). In 2003, she staged Deborah Hay’s conceptual dance work,
Exit, for students and faculty for Pomona College’s spring dance concert.
Lynn Garafola is a dance critic and historian. She is the author of Diaghilev’s Ballets
Russes (1989) and the editor of several books, including André Levinson on Dance:
Writings from Paris in the Twenties (with Joan Acocella) (1991), The Diaries of
Marius Petipa (1992) (which she also translated), Rethinking the Sylph: New Per-
spectives on the Romantic Ballet (1997), José Limón: An Unfinished Memoir (1998)
and most recently The Ballets Russes and Its World (Kurt Weill Prize 2001). She is
the editor of Studies in Dance History, the book series published by the Society of
History Scholars, and a senior editor/New York critic for Dance Magazine, and her
essays and criticism have appeared in The Nation, Ballet Review, Dancing Times,
The Times Literary Supplement, New York Times Book Review, Los Angeles Times
book review, and many other publications.
She holds a Ph.D. in Comparative Literature from the Graduate Center-City
University of New York and is the curator of “On Dance”, a series of lectures and
public programmes sponsored by the Barnard Department of Dance.
Ivor Guest is the Doyen of dance studies in the United Kingdom. He studied law
at Trinity College Cambridge and worked as a solicitor for many years. He is
the internationally recognised authority on nineteenth-century romantic ballet
in France and England. His books The Romantic Ballet in England (1954) and
The Romantic Ballet in Paris (1966) are considered standard reference works. He
also wrote the official history of the Paris Opera Ballet, Le Ballet de l’Opéra de
Paris (1976), which has been published in English in 2006. He has written many
biographies of romantic ballerinas and choreographers, among others of Fanny
Cerrito, Fanny Elssler, Adeline Genée, Jules Perrot and Virginia Zucchi.
An extensive bibliography is included in Guest’s autobiographical Adventures
of a Ballet Historian: An Unfinished Memoir (1982). An updated bibliography
xvii Notes on contributors

appeared in Dance Research (Summer 1995), an issue published in celebration of


his seventy-fifth birthday.
Sandra Noll Hammond is a dancer and dance historian whose research was among
the first to explore the development of ballet technique and training of the eigh-
teenth and early nineteenth centuries. She has presented this material in many
international venues, at master classes and concerts as well as lectures and articles.
Recent publications include “Sor and the Ballet of his Time” in Estudios Sobre
Fernando Sor/Sor Studies (2003) and “International Elements of Dance Training
in the Late Eighteenth Century” in The Grotesque Dancer on the Eighteenth-century
Stage/Gennaro Magri and his World (2005). She was co-founder and first director
of the dance major at the University of Arizona and later professor and director
of dance at the University of Hawaii. She studied ballet with Antony Tudor and
Margaret Craske at the Juilliard School and the Metropolitan Opera Ballet, and
with Arthur Mahoney and Thalia Mara at the School of Ballet Repertory. As a
performer, she was a member of Pacific Ballet and Arizona Dance Theatre, and
she has appeared as guest artist in concerts of baroque dance.
Thérèse Hurley holds a BM and GPD from the Peabody Conservatory of the Johns
Hopkins University and an MM from Temple University. Her master’s thesis
“The Harp in Tchaikovsky’s Ballets” examines the composer’s skill at composing
idiomatic music for the harp and his use of the instrument to convey the super-
natural in his ballets. She is a doctoral student in musicology at the University of
Oregon.
Marion Kant earned her Ph.D. in musicology in 1986 at Humboldt University in
Berlin on the subject of “Romantic Ballet: An Inquiry into Gender”. She has taught
at German Universities, at Cambridge University, King’s College London and the
University of Surrey, Great Britain and was a Visiting Fellow at King’s College
Cambridge. She is presently teaching courses in cultural and dance history, per-
formance criticism and the history of secularism at the University of Pennsylvania,
Philadelphia. Her publications include a monograph on the German choreogra-
pher Jean Weidt (1984), several articles and books on modern German dance
under Nazism: Hitler’s Dancers appeared in 2003 with Berghahn Books Oxford
and New York. An essay on Giselle was commissioned by the State Opera, Berlin
in 2000. Her main research and subsequent publications focus on the problems of
exile, on dance and music history in the nineteenth and twentieth centuries and
on dance aesthetics.
Together with musicians Marshall Taylor and Samuel Hsu she has organised
and presented a series of concerts commemorating Entartete Musik, music for-
bidden by the Nazis.
Erik Näslund is a dance scholar and director of the Dance Museum in Stockholm.
He has written extensively on the history of the Swedish Ballet in the 1920s. He
has also contributed significantly to developing methods of documenting dance
and making dance collections accessible for research and scholarship.
He is the author of monographs on Birgit Cullberg and Carina Ari; from 1973
to 1981 he was the editor of the magazine Dans. He writes dance criticism for the
Svenska dagbladet in Stockholm. He is engaged in the dance folklore movement
of Sweden.
xviii Notes on contributors

He recently rewrote the libretto for The Nutcracker, together with Pär Isberg
for the Royal Swedish Ballet.
Jennifer Nevile took her undergraduate degree in music at Sydney University and
wrote her doctoral dissertation at the School of Music at University of New South
Wales on fifteenth-century Italian dance manuscripts. Since then she has con-
tinued her research into dance and music from the fifteenth to the seventeenth
centuries.
She produced a video, The Amiable Conqueror: Dancing at the Courts of Europe
(1999), that showcased several sixteenth-century dances and four baroque chore-
ographies. Her research interests lie in the examination of the various dance prac-
tices in their social and intellectual contexts, that is, how the intellectual ideas
current during the early modern period were transformed by the dance mas-
ters into choreographic practices. Her latest book, The Eloquent Body: Dance and
Humanist Culture in Fifteenth-Century Italy (2004), explores dance as a physical
expression of Renaissance humanism and analyses the dance practice of fifteenth-
century Italy in relation to issues of economic class, education and power, as well
as to the contemporary intellectual discussion on the meaning of the arts and
ideas on the body, including moral concepts of eloquent movement, nobility and
ethics.
She is currently investigating changes in the choreographic structure of Italian
dances from the mid-fifteenth century to the first few decades of the sixteenth
century.
Marina Nordera was born in Mantua and graduated in musicology at the University
of Venice in 1990. Her thesis topic was on the tradition of the dialogue in Della
danza by Luciano. From 1985 to 1994 she was a professional dancer in the follow-
ing companies: Il Ballarino, Ris et Danceries, Fêtes Galantes, L’Eventail. She also
performed in several dance groups specifically staging baroque dance.
Her Ph.D. thesis, written and defended at the History Department of the Euro-
pean University in Florence, focused on the female performer in early and modern
dance: “La donna in ballo. Danza e genere nella prima età moderna.”
Since 2003 she has taught dance history, aesthetics and analytical methodology
of dance at the University of Nice.
Barbara Ravelhofer is a lecturer in English Literature at the University of Durham
and a Research Associate of the Centre for History and Economics, King’s College,
Cambridge. She pursued her research at the Universities of Munich, Princeton,
Bologna and Cambridge, where she was a Junior Research Fellow in Renaissance
Studies.
Her latest book, The Early Stuart Masque: Dance, Costume, and Music (2006),
studies the complex impact of movements, costumes, words, scenes, music, and
special effects in English illusionistic theatre of the Renaissance. Drawing on a
massive amount of documentary evidence relating to English productions as well
as spectacle in France, Italy, Germany and the Ottoman Empire, the book eluci-
dates professional ballet, theatre management and dramatic performance at the
early Stuart court.
Lucia Ruprecht graduated from the Universities of Tübingen and Aix-en-Provence
and completed her Ph.D. in German Literature at Cambridge. She teaches
xix Notes on contributors

literature, thought and film on the nineteenth and twentieth centuries and on
modern comparative literature at the English Faculty of the University of Cam-
bridge. Her research to date has focused on the interaction between literature and
dance, especially on questions of embodiment and subjectivity. She has a strong
interest in literary and cultural theory and is co-editor of Performance and Per-
formativity in German Cultural Studies (2003). She is currently working on the
notion of charisma in early twentieth-century cultural theory, literature, film and
dance.
Tim Scholl, associate professor of Russian at Oberlin College and the director of
Oberlin’s Center for Russian, East European and Central Asian Studies, is the
author of From Petipa to Balanchine: Classical Revival and the Modernization of
Ballet (1994) and Sleeping Beauty: A Legend in Progress, released in 2004 by Yale
University Press. His perspective on the restaging of Sleeping Beauty is particularly
intimate: he helped lead officials of the Maryinsky (formerly Kirov) Ballet of St
Petersburg to some invaluable documentation of the original 1890 production’s
choreography. His chronicle of the reconstruction of the Maryinsky’s signal work is
a tale of historical sleuthing that illuminates the difficulty of interpreting historical
evidence as well as the political conflict that often surrounds and shapes cultural
production.
Marian Smith holds a Ph.D. degree from Yale University. She is Associate Professor
of Music at the University of Oregon. She has published articles and reviews in
both music and dance journals, including the Cambridge Opera Journal, Dance
Chronicle, Journal of the American Musicological Society and Dance Research.She
has contributed chapters to the volume Reading Critics Reading: Opera and Ballet
Criticism 1830–1848 (2001), The Cambridge Companion to Grand Opera and (with
a co-author) Rethinking the Sylph (1997). Her essays on opera and ballet appear
in programme books of the Royal Opera and Royal Ballet in London, and she has
also presented scholarly papers in Italy, England, Germany and Denmark. Her
book on the intersection of opera and ballet in nineteenth-century Paris, Ballet
and Opera in the Age of Giselle (2000), was awarded the De la Torre Bueno Prize
from the Dance Perspectives Foundation. She is currently working on projects
about the historiography of nineteenth-century ballet, and the history of the pas
de deux.
Lester Tome, a Cuban dancer, performer, dance scholar and journalist, has taught
salsa, dance history, ethnography and pedagogy at Temple University, Denison
University and the University of the Arts in Philadelphia. He is currently com-
pleting a Ph.D. in dance as a University Fellow at Temple. His dissertation is
on Cuban ballet. As a journalist, he has published for the Chilean newspaper
El Mercurio; in the United States, his articles and reviews have appeared in the
Durham Herald-Sun and on Dance magazine’s website. In Cuba he wrote for Cuba
en el Ballet, Evolución y Cultura, and CMBF-Radio Musical Nacional. He has been
a fellow of New York Times Foundation and the National Endowment for the
Arts. He performed for Ally Ink and is a member of Sprezzatura, the baroque
dance ensemble at Temple University. In 2003, he performed in a restaging of Paul
Taylor’s Esplanade. He also writes the notes to the programmes for Ballet de San-
tiago, in Chile.
xx Notes on contributors

Dorion Weickmann studied social and economic history and political science at the
University of Hamburg, Germany. Her Ph.D. thesis focused on cultural aspects
of dance and ballet history. It was published in 2002 as Der dressierte Leib. Kul-
turgeschichte des Balletts (1580–1870). She is presently preparing a book on the
history of German dance in the twentieth century. She writes for several journals,
and also Die Zeit and Süddeutsche Zeitung and lives in Berlin.
Zheng Yangwen is a Research Fellow at the Asia Research Institute at the National
University of Singapore. She received her Ph.D. from Cambridge University (King’s
College). Her Ph.D. and postdoctoral work resulted in The Social Life of Opium in
China, 1483–1999 (2005).
Chronology

1279 The Mongols conquered China.


1283 Teutonic Order completed subjection to Prussia.
1292 Dante Alighieri, La Vita Nuova.
1302 Bull Unam Sanctam pronounced highest papal claims to
supremacy.
1321 Dante completed La Divina Commedia.
1321 Founding of minstrels’ guild.
1323 Thomas Aquinas canonised.
1327 Marsillius of Padua wrote Defensor Pacis.
1337 Giotto (painter) died.
1347–51 Black Death devastated Europe.
1348–53 Giovanni Boccaccio, Decamerone.
1358 Revolt of French peasants (Jacquerie).
1362 William Langland, Piers Plowman.
1368 Mongul Yuan dynasty in China overthrown by national Ming
dynasty.
1377 Guillaume de Machaut (composer) died.
1387–1400 Geoffrey Chaucer, The Canterbury Tales.
15th century Dance treatises recorded choreographies for the first time in
Western Europe and laid the structural foundations of ballet.
1415 Jan Hus burnt as a heretic in Prague.
1418 Thomas à Kempis De Imitatione Christi.
1421 Peking made capital of China.
1431 Joan of Arc burnt at the stake in Rouen.
1453–5 Johannes Gutenberg printed the Mazarin Bible in Mainz.
1453 The Turks conquered Constantinopol.
1455 Tristano Sforza’s wedding celebrations in Milan choreographed by
Domenico da Piacenza.
Domenico da Piacenza De arte saltandj & choreas ducendj De la
arte di ballare et danzare.
c. 1455 Antonio Cornazano Libro dell’arte del danzare.
1463 Guglielmo Ebreo da Pesaro Guilielmi Hebraei pisauriensis de
pratica seu arte tripudii vulgare opusculum incipit.
1469 Letter of Filippus Bussus to Lorenzo de’ Medici offering to come to
Florence in order to teach Lorenzo and his siblings “some elegant,
beautiful and dignified balli and bassadanze”.
Marsilio Ficino Commentary on Plato’s Symposium on Love.
1474 William Caxton printed the first English book.
1480 Ferdinand and Isabella of Spain appointed Inquisitor against
heresy.
[xxi]
xxii Chronology

1489 Marsilio Ficino, De vita.


1492 Jews expelled from Spain.
Christopher Columbus sailed for America.
1494 The Venetian Press of Aldus Manutius issued its first book.
Charles VIII invaded Italy and expelled the Medici.
1495–7 Leonardo da Vinci painted The last supper.
1503 Leonardo da Vinci painted the Mona Lisa.
1506 Albrecht Dürer from Milan: “I set to work to learn dancing and
twice went to the school. There I had to pay the master a ducat.
Nobody would make me go there again. I would have to pay out all
that I earned, and at the end I still wouldn’t know how to dance!”
1508–12 Michelangelo painted the ceiling of the Sistine Chapel in Rome.
1509 Henry VIII become King of England.
1512 Henry VIII celebrated epiphany with a masque.
1513 Niccolo Machiavelli, The Prince.
1517 Martin Luther affixed 95 Theses to the door of Wittenberg church.
1524–5 Hans Holbein the Younger painted The Dance of Death.
1528 Balthasar Castiglione, The Courtier.
1531 Thomas Elyot, The Boke Named the Governour.
1534 Jesuit Order founded in Paris.
1536 John Calvin went to Geneva and issued The Institute of the
Christian Religion.
1551 Giovanni Pierluigi Palestrina appointed conductor at St Peter’s in
Rome.
1558 Elizabeth I Queen of England.
1570 Andrea Palladio Treatise on Architecture.
Académie de Poésie et Musique founded by Jean Antoine de Baı̈f.
1572 St Bartholomew massacre in France.
1573 Ballet des ambassadeurs.
Torquato Tasso Aminta.
1580 Michel Eyquem de Montaigne, Essays.
1581 Balet comique de la Royne (Allegorie of Circé).
Fabritio Caroso, Il Ballarino.
1586 The war of the three Henrys in France.
end 16th c. Emergence of ballet de cour.
1590 Edmund Spenser, The Faery Queene.
1593 Henry IV of France converted to Roman Catholicism.
1597 William Shakespeare, Romeo and Juliet.
1599 Opening of the Globe Theatre, London.
1600 Fabritio Caroso, Nobiltà di dame.
1602 Cesare Negri, Le gratie d’amore.
1602–4 Galileo Galilei discovered laws of gravitation.
1603 William Shakespeare, Hamlet.
1603 Elizabeth I died.
1605 The Masque of Blackness.
Miguel de Cervantes Saavedra, Don Quixote.
xxiii Chronology

1607 Claudio Monteverdi, Orfeo.


1608 The Masque of Beauty.
1609 The Masque of Queens.
1609 Johannes Kepler, Astronomia Nova.
1610 Ballet de Monseigneur le Duc de Vandosme.
1611 Oberon.
William Shakespeare, The Tempest.
1613 The Masque of the Inner Temple and Gray’s Inn.
1617 Ballet de la déliverance de Renaud.
1618–48 Thirty Years War.
1620 Puritans set up Plymouth Colony in New England.
1623 François de Lauze, Apologie de la danse.
1626 Ballet de la douairière de Billebahaut.
1629 Pierre Corneille, Mélite.
1632 First female professional singers in English theatre history appear
in the masque Tempe restored.
1634 A Masque Presented at Ludlow Castle.
The Triumph of Peace.
John Milton, Comus.
1635 Académie Française founded by Cardinal Richelieu.
1637 Réné Descartes, Discours de la méthode.
1639 Nicolas Poussin appointed French court painter.
1639 Francesco Cavalli, Le nozze di Peleo e di Teti, Venice.
c. 1640 Stage for the ballet de cour is elevated.
1640 Salmacida Spolia.
English Civil War broke out.
1641 Ballet de la prosperité des armes de la France.
1642–60 English theatres closed.
1642 Rembrandt Harmensz van Rijn, Night Watch.
1644 The Manchus conquered China.
1649 Charles I of England beheaded.
1650 Il tabacco.
1651 Thomas Hobbes, Leviathan.
1653 Cupid and Death.
Ballet de la Nuit
1653 Il Gridelino.
1654 Carlo Caproli, Le nozze di Peleo e di Teti (Les Noces de Pelée et
Thétis), Paris.
1654 Louis XIV crowned.
1661 Académie Royale de Danse founded in Paris.
1662 Building of Versailles begun.
1664 Molière, Tartuffe.
1666 Molière, Le Misanthrope.
1667 John Milton, Paradise Lost.
Jean Racine, Andromaque.
1669 Académie Royale de Musique founded in Paris.
xxiv Chronology

1670 Louis XIV gives up dancing in leading roles of the ballet de cour.
Baruch Spinoza, Tractatus Theologico-Politicus.
1672 First journal for light reading: Le Mercure galant.
1673 William Wycherley, The Gentleman Dancing Master.
Molière, Le Malade imaginaire.
1674 William Wycherley, The country wife, The plain dealer.
1675 Calisto (John Crowne).
1677 Racine, Phèdre.
1680 Pierre Beauchamps, second director of Académie Royale de Danse.
Comédie française established.
Henry Purcell, Dido and Aeneas.
1687 Charles Perrault, The Age of Louis the Great.
Isaac Newton, Philosophiae Naturalis Principia Mathematica.
1688 Glorious revolution: William of Orange King of England.
1689–97 Nine years of war between England and France.
1690 John Locke, Essay concerning human understanding.
1695 William Congreve, Love for Love.
1700 Raoul Auger Feuillet, Chorégraphie ou l’art de décrire la danse par
caractères, figures et signes démonstratifs.
1702 Charles Le Brun, Méthode pour apprendre à dessiner les passions.
John Weaver, The Tavern Bilkers.
1702–13 War of the Spanish Succession.
1704 Jonathan Swift, The Tale of a Tub.
1704 Isaac Newton, The Corpuscular Theory of Light.
1705 Peter the Great founded Moscow University.
1706 Mr Isaac, A Collection of Ball-Dances perform’d at Court: viz. The
Richmond.
John Weaver, A Small Treatise of Time and Cadence in Dancing,
Reduc’d to an Easy and Exact Method, Shewing how Steps, and their
Movements, agree with the Notes, and Division of Notes, in each
Measure.
John Weaver’s translation of Raoul Auger Feuillet, Orchesography
or the Art of Dancing, by Characters and Demonstrative Figures, By
which any Person, who understands Dancing, may of himself easily
learn all manner of Dances.
1707 Johann Pasch, Beschreibung wahrer Tantz-Kunst.
1709–11 Sir Richard Steele founded The Tatler.
1711–14 The Spectator, editors Joseph Addison and Sir Richard Steele.
1712 John Weaver, An Essay Towards an History of Dancing, In which the
whole Art and its Various Excellencies are in some Measure
Explain’d, containing the several sorts of Dancing, Antique and
Modern, Serious, Scenical, Grotesque, etc. with the Use of it as an
Exercise, Qualification, Diversion etc.
1712 Alexander Pope, The Rape of the Lock.
1713 the Paris Opéra gave formal recognition to its dance constituents
by establishing a permanent troupe of twenty dancers, ten women
and ten men.
xxv Chronology

Prince Eugene of Austria built the Belvedere Palace.


1717 John Weaver, The Loves of Mars and Venus.
1718 John Weaver, Orpheus and Eurydice.
1719 Claude Ballon director of Académie Royale de la Danse.
Daniel Defoe, Robinson Crusoe.
1720 Christian Wolff Rational thought on God, the world and the human
soul.
1721 John Weaver, Anatomical and Mechanical Lectures upon Dancing,
wherein Rules and Institutions for that Art are laid down and
demonstrated.
1722 Johann Sebastian Bach, The Well-Tempered Clavier.
1725 Pierre Rameau emphasised the vertical, balanced stance and
outward turn of feet of the dancer.
Pierre Rameau, Le maı̂tre à danser.
C. Sol, Méthode très facile et fort nécessaire, pour montrer à la
jeunesse de l’un et l’autre sexe la manière de bien dancer.
Giovanni Battista Vico Scienza Nuova Intorno alla Natura.
1726 Jonathan Swift, Gulliver’s Travels.
1728 John Weaver, Perseus and Andromeda.
John Weaver, The History of the Mimes and Pantomimes, with an
Historical Account of several performers in Dancing, living in the
Time of the Roman Emperors.
John Essex, The Dancing-Master, English translation of Pierre
Rameau’s Le maı̂tre à danser.
Giambattista Dufort, Trattato del ballo nobile.
1729 Soame Jenyns, The Art of Dancing. A Poem.
1730 Johann Christoph Gottsched, Critical art of poetry for the Germans.
1733 John Weaver, The Judgment of Paris, A Dramatic Entertainment in
Dancing and Singing, After the Manner of the Ancient Greeks and
Romans.
Marie Sallé in Pygmalion.
Antoine François Prévost, Manon Lescaut.
1738 First spinning machines patented in England.
First ballet school in Russia under Jean-Baptiste Landé.
1740 Samuel Richardson, Pamela.
1740s Franz Hilverding produced dance dramas after Jean Racine’s
Britannicus, Crébillon’s Idoméneo and Voltaire’s Alzira.
1741 George Frederick Handel, Messiah.
First German translation of a Shakespeare play (Julius Caesar) by
von Borcke.
1743 Jean-Georges Noverre on stage for the first time in Favart’s
vaudeville Le Coq du village.
1747 Jean-Georges Noverre appointed ballet master in Marseilles, first
choreography there Les Fêtes chinoises.
1748 First silk factory in Berlin.
Carlo Gozzi, Turandot.
Samuel Richardson, Clarissa.
xxvi Chronology

1750 Jean-Georges Noverre Le Jugement de Paris (first ballet


pantomime).
1751 First volume of the Encyclopédie, ed. Denis Diderot and Jean
d’Alembert; entry on dance by Louis de Cahusac.
Jean-Georges Noverre, Fêtes chinoises (Lyon).
1752 Benjamin Franklin invented lightning rod.
Guerre des buffons.
1754 Louis de Cahusac, La Danse ancienne et moderne ou Traité
historique de la Danse.
1754 Jean-Georges Noverre, Fêtes chinoises (Paris).
1755 Johann Joachim Winckelmann, Gedanken über die Nachahmung
der griechischen Werke in der Mahlerey und Bildhauer-Kunst.
Gotthold Ephraim Lessing, Miss Sara Sampson.
Jean-Georges Noverre Fêtes chinoises (London).
Samuel Johnson Dictionary of the English language.
1756–63 Seven Years War.
1756 Imperial Theatres as a state system founded by decree of Catherine
the Great.
1757 Jean-Georges Noverre, La Toilette de Vénus.
1758 Jean-Georges Noverre, La Mort d’Ajax.
Denis Diderot, Entretiens sur le fils naturel.
1759 Voltaire, Candide.
1760 Jean-Georges Noverre, Lettres sur la danse, et sur les ballets, Lyon
and Stuttgart.
1760s The ballet d’action arrived in Russia with Franz Hilverding and
Gasparo Angiolini.
1761 Gasparo Angiolini, Le Festin de Pierre, or Don Juan.
Jean-Jacques Rousseau, La Nouvelle Héloı̈se.
1762 Giovanni-Andrea Gallini, A Treatise on the Art of Dancing.
Christoph Willibald Gluck, Orfeo ed Euridice (Orphée et Eurydice).
1763 Jean-Georges Noverre, Jason et Médée.
1764 James Hargreaves invented Spinning Jenny.
1765 Gasparo Angiolini, Sémiramide.
1766 Johann Joachim Winkelmann’s Gedanken über die Nachahmung
der griechischen Werke in der Mahlerey und Bildhauer-Kunst
translated into French.
1770 Christoph Willibald Gluck, Paride ed Elena (Pâris et Hélène).
Malpied, Traité sur l’art de la danse.
1772 Jean-Georges Noverre, Iphigénie en Tauride.
Pierre Gardel refused to appear in full dress and decorative wig in
the entrée of the opera-ballet Castor et Pollux by Jean-Philippe
Rameau.
Jacques Cazotte, Le Diable amoureux (novel).
1773 Johann Gottfried Herder, Von deutscher Art und Kunst.
Jean-Georges Noverre, Apelle et Campaspe.
Jean-Georges Noverre, Adèle de Ponthieu.
xxvii Chronology

1774 Johann Wolfgang von Goethe, The Sorrows of Young Werther


Joseph Priestley discovered oxygen.
Gasparo Angiolini, Thésée en Crète.
Jean-Georges Noverre, Horaces et des Curiaces.
1775 James Watt constructed first efficient steam engine.
1776 Jean-Georges Noverre, Les Caprices de Galathée.
American Revolution.
Adam Smith, The Wealth of Nations.
Christoph Willibald Gluck, Alceste.
1778 Jean-Georges Noverre, Les Petit Riens.
Jean-Georges Noverre, Anette et Lubin.
1779 Gennaro Magri, Trattato teorico-prattico di ballo.
1780 Jacques-François Deshayes appointed director of Académie Royale
de la Danse.
Jean-Georges Noverre, Jason et Medée.
1781 Jean-Jacques Rousseau, Confessions.
1784 André-Ernest-Modest Grétry, Richard Coeur de Lion (opera) Paris.
1785 Jacques-Louis David, Oath of the Horatii.
Wolfgang Amadeus Mozart, The Marriage of Figaro.
1789 Abbé Emmanuel Joseph Sieyès, Qu’est-ce que le tiers état?
Jean Dauberval/music arr. La Fille mal gardée (Bordeaux).
French Revolution; Declaration of rights of man and of the citizen.
Giovanni Paisiello, Nina ou la folle par amour.
1790 (14 July) Fête de la Fédération (Festival of Federation).
Pierre Gardel, Psyche.
Johann Wolfgang von Goethe, Faust I.
Pierre Gardel, Télémaque.
1792 (30 September) Pierre Gardel, Offrande à la Liberté performed in
Paris Opera.
1793 Pierre Gardel, Le Judgement de Paris.
(10 November) Fête de la Raison.
1793–4 Reign of Terror instituted in France under Maximilien
Robespierre.
1794 (8 June)Fête de l’Etre Suprême.
Pierre Gardel, La Réunion du 10 âout (The reunion of 10 August)
1796 Charles Didelot/Cesare Bossi, Flore et Zéphire (London).
Edward Jenner used vaccination for the first time.
1797 Johann Christian Friedrich Hölderlin, Hyperion.
1798 Thomas Malthus, Essay on the Principle of Population.
1799 Napoleon Bonaparte seized power.
1800 Pierre Gardel, La Dansomanie (Paris).
1803 Republication of Jean-Georges Noverre, Lettres sur la Danse, et sur
les Ballets, St Petersburg.
1804 Pierre Gardel, Une demi-heure de Caprice (Paris).
Bonaparte crowned Emperor as Napoleon I.
1807 Joseph Mallord William Turner, painting Sun Rising in a Mist.
xxviii Chronology

Georg Friedrich Wilhelm Hegel, Phenomenology of the Spirit.


1810 Walter Scott, Lady of the Lake.
Johann Wolfgang von Goethe, Colour Theory.
Foundation of Krupp works at Essen.
1812 Salvatore Taglioni and Louis Henry founded the school of ballet at
the Teatro San Carlo in Naples.
1813 Academy of dancing established in Milan at La Scala.
Jane Austen, Pride and Prejudice.
Louis-Jacques Milon, Nina ou la folle par amour.
Battle of Leipzig, Napoleon I defeated.
1814 Pierre Gardel, Le Retour des Lys (Paris).
Congress of Vienna opened.
1815 Battle of Waterloo, Napoleon I finally defeated.
1816 Ernst Theodor Amadeus Hoffmann’s short story The Nutcracker
and the Mouse King published.
1818–1910 Marius Petipa.
1818 John Keats, Endymion.
François Decombe Albert/Jean-Madeleine Schneitzhoeffer, Le
Séducteur du village.
1819 Lord Gordon George Byron, Don Juan.
1820 Carlo Blasis, Traité élémentaire théorique et pratique de l’art de la
danse.
1821 Carl Maria von Weber, Der Freischütz.
1822 Gas lighting to illuminate the stage was introduced at Paris Opéra.
Jean Louis Aumer/Count Robert Gallenberg/Gustave Dugazon,
Alfred le grand (Paris).
1824 Jacques-François Deshayes/Jean-Madeleine Schneitzhoeffer,
Zémire et Azor.
Auguste Baron, Lettres et entretiens sur la danse.
1825 Alexander Pushkin, Boris Godunov.
Marius Petipa stage debut in Brussels.
1826 Jean Baptiste Blache/Jean-Madeleine Schneitzhoeffer, Mars et
Vénus ou Les Filets de Vulcain (Paris).
Felix Mendelssohn-Bartholdy, Overture to A Midsummer Night’s
Dream.
1827 Jean Louis Aumer/ Louis Joseph Ferdinand Hérold, La
Somnambule ou L’Arrivée d’un nouveau seigneur (Paris).
Alessandro Manzoni, I Promessi Sposi – first novel in Italian.
Heinrich Heine, The Book of Songs.
Franz Schubert, The ‘Trout’ Quintett.
Vincenzo Bellini/Felice Romani, La Somnambule (opera) Paris.
1828 August Bournonville, Nytaarsgave for Dandseyndere (A New Year’s
Gift for Dance Lovers).
Daniel-François-Esprit Auber, La Muette de Portici.
Jean Dauberval/Louis Joseph Ferdinand Hérold, La Fille mal
gardée (revised) (Paris).
xxix Chronology

1828–30 Carlo Blasis, The Code of Terpsichore.


1829 St Matthew Passion by Johann Sebastian Bach performed in the
Singakademie Berlin on March 11, 1829 under the direction of
Felix Mendelssohn-Bartholdy.
Gioacchino Rossini Guillaume Tell (opera) (Paris).
Victor Hugo, Fantômes (poem).
Jean Louis Aumer/Ferdinand Hérold, La Belle au bois dormant
(Paris).
1830 Revolutions in France and Italy.
Adolphe Adam, La Chatte blanche.
Filippo Taglioni, Le Dieu et la bayadère.
Jean Louis Aumer/Fromental Halévy, Manon Lescaut.
Alexander Pushkin, Eugene Onegin.
1830–77 August Bournonville ballet master at the Royal Danish Ballet.
1831 Dimming of house lights introduced to Paris Opéra.
Jean Coralli/ Michel Enrico Carafo/Jean-Madeleine
Schneitzhoeffer, L’Orgie (Paris).
Giacomo Meyerbeer, Robert le diable (opera with the Ballet of the
nuns).
1831–35 Louis Véron director of the Paris Opéra.
1832 A.E. Théleur, Letters on dancing.
Filippo Taglioni/ Jean-Madeleine Schneitzhoeffer, La Sylphide
(Paris).
Jacques-Fromental Halévy, La Tentation (opera ballet) (Paris).
Filippo Taglioni/Adalbert Gyrowetz/Michel Enrico Carafa
Nathalie, ou la Laitière Suisse (Paris).
1833 Filippo Taglioni, La Révolt au Sérail (La Révolte des femmes) (Paris).
André Deshayes/Adolphe Adam Faust (London).
1834 Jean Coralli/ Jean-Madeleine Schneitzhoeffer, La Tempête ou l’ı̂le
des génies (Paris).
Heinrich Heine, De l’Allemagne.
1835 Filippo Taglioni/Count Robert Gallenberg, Brézilia ou la tribu des
femmes (Paris).
Georg Büchner, Danton’s Death.
1836 Giacomo Meyerbeer, Les Huguenots (opera) (Paris).
Jean Coralli/Casimir Gide, Le Diable boiteux.
August Bournonville, La Sylphide (Copenhagen).
Filippo Taglioni/Adolphe Adam, La Fille du Danube.
1837 Carlo Blasis and his wife, Annunziata Rammaccini, directors of
the Academy of dancing in Milan at La Scala.
Nicola Guerra/Adolphe Adam, Les Mohicans (Paris).
1838 Therese Elssler/Casimir Gide, La Volière ou les oiseaux de
Boccace.
1839 Joseph Mazilier/ François Benoist/ Marco Aurelio Marliani/
Ambroise Thomas, La Gypsy.
1839–42 The Opium War in China.
xxx Chronology

1840 Joseph Mazilier/ François Benoist/Napoléon-Henri Reber, Le


Diable amoureux.
Filippo Taglioni/Adolphe Adam, Die Hamadryaden (Berlin).
Adolphe Adam, L’Écumeur de mer (St Petersburg).
1841 Gioacchino Rossini, Moı̈se (opera).
Jean Coralli/Jules Perrot/Adolphe Adam/Frederich Burgmüller,
Giselle.
During a performance of Toreadoren August Bournonville was
ordered off stage by the Danish King Christian VIII.
Incandescent electrical light bulb patented.
1842–1911 Late Qing period in China.
1842 François Decombe Albert/Adolphe Adam, La Jolie Fille de Gand
(Paris).
August Bournonville, Napoli (Copenhagen).
1843 Richard Wagner, The Flying Dutchman.
Jean Coralli/Frederich Burgmüller, La Péri.
Gaetano Donizetti, Dom Sébastien (opera) (Paris).
1844 Joseph Mazilier/Friedrich von Flotow/Edouardo Deldevez
Frederich Burgmüller, Lady Henriette, ou La Servante de
Greenwich.
Jean Coralli/Edouardo Deldevez, Eucharis (Paris).
Arthur Saint-Léon/Cesare Pugni, La Vivandiére (London).
1845 Jules Perrot, Pas de quatre with Marie Taglioni, Carlotta Grisi,
Fanny Cerrito, Lucile Grahn in London.
Joseph Mazilier, Le Diable à quatre (Paris).
François Decombe Albert/Adolphe Adam, The Marble Maiden
(London).
1846 Joseph Mazilier/Edouardo Deldevez, Paquita (Paris).
1847 Charlotte Brontë, Jane Eyre; Emily Brontë, Wuthering Heights.
Marius Petipa settled in St Petersburg.
Carlo Blasis, Notes Upon Dancing, Historical and Practical.
Jean Coralli, Ozaı̈ (Paris).
Arthur Saint-Léon/Cesare Pugni, La Fille de marbre.
The sisters Elssler, La Salamandrine (London).
Charles Baudelaire, La Fanfarlo.
1848 Revolutions all over Europe except Britain, Belgium and Russia.
Joseph Mazilier/Adolphe Adam, Griseldis ou les cinq senses (Paris).
Lucien Petipa, Nisida ou les Amazones des Açores.
August Bournonville, My Theatre Life.
1849 Giacomo Meyerbeer, Le Prophète (opera) (Paris).
Jules Perrot/Adolphe Adam, La filleule des fées (Paris).
August Bournonville, The Conservatory.
1850 Richard Wagner, Lohengrin.
1850 Carlotta Grisi’s debut as Giselle in Russia.
1851 August Bournonville, The Kermesse in Bruges.
Giuseppe Verdi, Rigoletto.
xxxi Chronology

Joseph Mazilier/Jean Baptiste Tolbecque/Edouardo Delvedez,


Vert-Vert (Paris).
1852 Coup d’Etat by Louis Napoleon Bonaparte, the French president
who became Napoleon III.
Joseph Mazilier/Adolphe Adam, Orfa (Paris).
1853 Giuseppe Verdi, Il Trovatore; Richard Wagner, Der Ring des
Nibelungen.
1854 August Bournonville, A Folk Tale (Copenhagen).
Dogma of the Immaculate Conception made an article of faith.
1854–6 Crimean War.
1855 François Henri Joseph Castil-Blaze, L’Académie Impériale de
Musique. Histoire littéraire, musicale, chorégraphique, pittoresque,
morale, critique et galante de ce théâtre de 1645 à 1855.
August Bournonville, Abdallah.
Giuseppe Verdi, Les Vêpres siciliennes (opera) (Paris).
1856 August Bournonville, La Ventana.
Joseph Mazilier/Adolphe Adam, Le Corsaire (Paris).
1857 Gustave Flaubert, Madame Bovary.
Marius Petipa/Ricardo Drigo/Ludwig Minkus/Cesare Pugni with a
new divertissement by Delibes, Le Corsaire.
1858 August Bournonville, The Flower Festival in Genzano.
Suez Canal Company (Compagnie Universelle du Canal Maritime
de Suez) founded.
1858–9 Théophile Gautier, Histoire de l’art dramatique en France depuis
vingt-cinq ans.
1858 and 1867 Théophile Gautier travelled to Russia (Voyage en Russie).
1859 Charles Darwin, Origin of Species by means of Natural Selection.
France and Piedmont defeated Austria.
Arthur Saint-Léon appointed ballet master at Imperial Theatre in
Russia for eleven seasons.
G. Léopold Adice, Théorie de la gymnastique de la danse théâtrale.
Jacques Offenbach, Orpheus in the Underworld.
First oil well discovered in the United States.
1860 Marie Taglioni, Le Papillon.
August Bournonville, Far from Denmark.
1861 Victor Emanuel proclaimed King of Italy.
Richard Wagner’s Tannhäuser failure in Paris.
Hans Christian Andersen completed the Fairy Tales.
1861–5 American Civil War.
1862 Marius Petipa, The Daughter of Pharaoh (St Petersburg).
1863 Arthur Saint-Léon/Cesare Pugni, Diavolina, Imperial Theatre St
Petersburg.
1864 Arthur Saint-Léon/Cesare Pugni, The Little Humpbacked Horse,
Imperial Theatre St Petersburg.
Pius IX condemned all forms of liberalism in the Syllabus of Errors.
1866 Prussia defeated Austria and Northern Germany united.
xxxii Chronology

Arthur Saint-Léon/Leo Délibes, La Source (Paris).


Fyodor Dostoyevsky, Crime and Punishment.
Ambroise Thomas, Mignon (opera) (Paris).
1867 Giuseppe Verdi, Don Carlos (opera) (Paris).
Karl Marx, Das Kapital I.
1868 Richard Wagner, Die Meistersinger (Munich).
1869 August Bournonville founded a special pension fund for dancers.
Peter Tchaikovsky, Undine (opera).
Marius Petipa/Ludwig Minkus, Don Quixote St Petersburg.
After eleven years of work the Suez Canal opened in November.
1869–1903 Marius Petipa director of the Imperial Ballet in St Petersburg.
1870 Arthur Saint-Léon/Léo Délibes, Coppélia.
Prussia defeated French Empire, Napoleon III captured and
French Republic proclaimed.
Dogma of Papal Infallibility declared by Vatican Council.
1871 Friedrich Nietzsche, The Birth of Tragedy out of the Spirit of Music.
William I proclaimed German Emperor at Versailles.
August Bournonville, The King’s Volunteers on Amager.
Charles Darwin, Descent of Man.
1871 First impressionist exhibition at Paris.
1872 Marius Petipa/Ludwig Minkus, Camargo.
1873 Severe economic crisis in Europe, America and Australia.
1875 Georges Bizet, Carmen (opera) Paris.
1876 August Bournonville, From Siberia to Moscow.
Louis Mérante/Léo Délibes, Sylphia ou La Nymphe de Diane
(Paris).
Richard Wagner’s Festspielhaus opened in Bayreuth.
1877 Leo Tolstoy, Anna Karenina.
Pyotr Tchaikovsky/Julius Reisinger, Swan Lake.
Marius Petipa/Ludwig Minkus, La Bayadère.
1878 André Messager, Fleur d’oranger, Folies Bergère.
1879 Thomas Edison perfected the electric bulb.
André Messager, Les Vins de France and Mignons et villains, Folies
Bergère.
1880 Marius Petipa, Le Corsaire.
Fyodor Dostoyevsky, The Brothers Karamazov.
Louis Mérante/Charles-Marie Widor, La Korrigane (Paris).
1881 Alexander II assassinated.
1881–99 Ivan Alexandrovich Vsevolozksky Director of Imperial Theatres.
1882 Richard Wagner, Parsifal.
Monopoly of the Imperial Theatres in St Petersburg abolished,
which made visits of foreign companies possible.
Lucien Petipa/Éduard Lalo, Namouna (Paris).
1883 Daimler-Benz factories established in Germany.
Friedrich Nietzsche, Thus spake Zarathustra.
1884 Marius Petipa/Adolphe Adam, Giselle (revised) St Petersburg.
xxxiii Chronology

1885 Edmond de Goncourt, Chérie (novel).


1886 Louis Mérante/André Messager, Les Deux Pigeons (Paris)
1888 Heinrich-Rudolf Hertz detected electro-magnetic waves.
1889 Ambroise Thomas, La Tempête (Paris).
Gustave Eiffel built the Eiffel tower in Paris.
1890 Peter Tchaikovsky/Marius Petipa, Sleeping Beauty.
Knut Hamsun, Hunger.
1891 Leo XIII issued papal encyclical Rerum novarum on the rights of
labour.
Henrik Ibsen, Hedda Gabler.
1892 Peter Tchaikovsky/Lev Ivanov/Marius Petipa, The Nutcracker.
1893 Giacomo Puccini, Manon Lescaut.
1894 Trial of Alfred Dreyfus in Paris.
1895 Emile Durkheim, The Rules of Sociological Method.
Revival Pyotr Tchaikovsky/Marius Petipa/Lev Ivanov, Swan Lake.
Guglielmo Marconi invented wireless telegraphy.
1895 Wilhelm Conrad Röntgen discovered X-Rays.
Gustave le Bon, Psychology of the Crowd.
1898 radium discovered by Marie and Pierre Curie.
Emile Zola, J’accuse.
Marius Petipa/Alexander Glazunov, Raymonda (St Petersburg).
1899 Aleksandr Gorsky’s production of Sleeping Beauty in Moscow.
Houston Stewart Chamberlain, Foundations of the Nineteenth
Century.
1900 Sigmund Freud, The Interpretation of Dreams.
c. 1900–1940 The Mountain of Truth, reform and artists colony in Ascona,
Switzerland.
Wilhelm Wundt, Comparative Psychology.
Aleksandr Gorsky’s production of Don Quixote in Moscow.
Isadora Duncan moved to France, gave first performances in
London.
1901 Aleksandr Gorsky staged Giselle in Moscow (and again in 1907,
1918 and 1922).
Aleksandr Gorsky/Frédéric Chopin, Valse fantaisie (Moscow).
1902 Maxim Gorky, Nights Lodging.
Aleksandr Gorsky, Don Quixote (restaging of Petipa’s ballet at
Imperial Theatre).
1903 George Bernard Shaw, Man and Superman.
Introduction of ballet to China by Yu Ronglin.
Wright brothers flew the first aeroplane.
Marius Petipa, The Magic Mirror, St Petersburg, a failure.
1903–4 The Imperial Ballet ballet employed 122 female and 92 male
dancers.
1904 Russo-Japanese War.
Isadora Duncan’s first tour of Russia.
1904–83 George Balanchine.
xxxiv Chronology

1905 Revolution in Russia.


Herbert George Wells, Modern Utopia.
Richard Strauss, Salome.
Albert Einstein, Special Theory of Relativity.
Michel Fokine, Acis and Galatea.
Isadora Duncan established her school of modern dance in
Berlin.
1907 Shell Oil Trust founded.
Henri Bergson, L’Evolution créatrice.
Michel Fokine/Camille Saint-Saëns, The Swan for Anna Pavlova.
Michel Fokine, Pavillon d’Armide.
Rainer Maria Rilke, New Poems.
Michel Fokine/Frédéric Chopin, Chopiniana (St Petersburg).
1908 Pablo Picasso and Georges Braque founded cubism.
1909 Serge Diaghilev presented Russian dancers in five ballets in his
Saisons russes in Paris.
Michel Fokine/Frédéric Chopin, Les Sylphides, Ballets Russes
(Paris).
Filippo Tommaso Marinetti published the Founding and Manifesto
of Futurism.
Arnold Schøenberg, Three Piano Pieces.
Vasily Kandinsky announced absolute painting.
1910 Michel Fokine, Schéhérazade and Firebird, Ballets Russes.
1911 Michel Fokine, Le Spectre de la rose and Petrushka, Ballets Russes.
Nationalist Revolution in China.
Vasily Kandinsky, Concerning the Spiritual In Art.
Richard Strauss, Der Rosenkavalier.
1912 China declared republic.
Thomas Mann, Der Tod in Venedig (Death in Venice).
1912–13 The three Balkan wars.
1913 Edmund Husserl, Phenomenology.
1913 Vaclav Nijinsky/Igor Stravinsky, Le Sacre du printemps, Ballets
Russes (Paris).
1913–14 Mikhail Fokine in Stockholm.
1913–18 Rudolf von Laban lived in Ascona, on the Mountain of Truth and
founded Modern German Dance there.
1914 (July) Michel Fokine’s manifesto of the new ballet first appeared in
The Times.
1914 (28 June) Archduke Franz Ferdinand assassinated in Sarajevo.
1914 Heinrich Mann, Der Untertan.
1914–18 First World War.
1915 First use of poison gas by German army.
1915 (February – December) Battle of Verdun.
1915 Denishawn dance school in Los Angeles founded.
1917 Germany proclaimed unrestricted submarine warfare.
Filippo Tommaso Marinetti, Manifesto on Futurist Dance.
xxxv Chronology

Jean Cocteau/Pablo Picasso/Eric Satie, Parade, Ballets Russes


(Paris).
(November) Bolshevik Revolution in Russia.
United States declared war on Germany and the Austro-
Hungarian Empire.
1918 President Woodrow Wilson issued The 14 Points (programme for a
just peace).
William II abdicated, Germany declared a republic.
(November) armistice ended the First World War, revolution in
Germany and other European countries.
1919 Peace Conference began at Versailles.
Bauhaus founded in Germany.
Alexander Sacharoff, Au temps du grand siècle/Pavane royale.
Sociedad Pro-Arte Musical founded in Havana.
1920 Léonide Massine, Pulcinella.
Vaclav Nijinsky/Claude Debussy, Jeux, Jean Börlin/Isaac Albeniz,
Iberia, Jean Börlin/Alfvén, La Nuit de Saint-Jean, Jean
Börlin/Alexander Glazunov, Derviches.
Jean Börlin, La Maison de fous, Jean Cocteau/Darius Milhaud, Le
Boeuf sur le toit, Ballets Suédois (Paris).
1920–5 The Ballets Suédois in Paris.
1921 Jean Börlin/Jean Cocteau/Les Six, Les Mariés de la Tour Eiffel, Jean
Börlin/Darius Milhaud, L’Homme et son désir.
Agrippina Vaganova began teaching career at the Leningrad State
Choreographic School.
1922 Anna Pavlova brought The Dying Swan to Shanghai.
Jean Börlin, Skating Rink, Ballets Suédois (Paris).
Oskar Schlemmer, Triadic ballet (Triadisches Ballett) (Stuttgart).
Benito Mussolini and Fascists came to power in Italy.
1922–3 Hyper-inflation in Germany and Austria.
1923 Bronislava Nijinska, Les Noces (Paris).
Rudolf von Laban began his Kammertanz Theatre in Hamburg.
Adolf Hitler and Nazis attempted coup d’état in Munich.
Fyodor Lopukhov/Ludwig van Beethoven dance-symphony,
Magnificence of the Universe.
Jean Börlin/Darius Milhaud, La Creátion du monde, Jean
Börlin/Cole Porter, Within the Quota, Ballets Suédois (Paris).
1924 Death of Vladimir I. Lenin, leader of Russian Bolshevik
Revolution.
Bronislava Nijinska, Les Fâcheux.
Ballets Suédois last evening: Jean Börlin/Francis Picabia/Eric Satie,
Relâche, Jean Börlin/Arthur Honegger/Fernand Léger, Skating
Rink, Jean Börlin/Darius Milhaud, La Création du monde.
1925 Fyodor Lopukhov, Paths of a Ballet-master.
Léo Staats/Léo Delibes, Soir de fête.
Kasyan Goleizovsky, The Legend of Joseph the Beautiful (Moscow).
xxxvi Chronology

1926 George Balanchine, La Pastorale.


George Balanchine, The Triumph of Neptune.
Marie Rambert founded the Marie Rambert Dancers, later called
the Ballet Club, Ballet Rambert (1935–87) and finally Rambert
Dance Company.
Martha Graham gave her first New York performance.
Fritz Lang, Metropolis (film).
1927 George Balanchine, La Chatte.
Isadora Duncan, Autobiography.
Kasyan Goleizovsky, The Whirlwind, led to his resignation.
Death of Isadora Duncan.
Lev Lashchilin, Vasili Tikhomirov/Reinhold Glière, The Red Poppy,
Bolshoi Theatre Moscow.
1928 George Balanchine/Igor Stravinsky, Apollon musagète.
1928 Ninette de Valois engaged as director of Sadlers Well’s Ballet
(VicWells Ballet); Sadler’s Wells Theatre Ballet eventually split in
1956 into the Covent Garden’s Royal Ballet and Birmingham Royal
Ballet.
1929 Kurt Jooss, Pavane on the Death of an Infanta.
George Balanchine, Le Bal.
George Balanchine, Le Fils prodigue.
(May) Soviet Union adopted first Five-Year-Plan.
(October) US stock market crashed on Black Friday; world
depression began.
1931 Several Austrian and German banks collapsed.
Ninette de Valois, Job, Camargo Society (London).
(September)Japan invaded Manchuria.
1932 Kurt Jooss/Fritz Cohen, The Green Table (Paris).
The Ballets Russes de Monte Carlo founded as a fusion of the
Ballets de l’Opéra de Monte Carlo and the Ballet de l’Opéra Russe
à Paris, with Colonel de Basil as director and René Blum as artistic
director.
Vasily Vainonen/Boris Asafiev, The Flame of Paris, Kirov Theatre
(Leningrad).
1933 (January) Adolf Hitler chancellor of Germany – Nazi seizure of
power.
(March) Franklin D. Roosevelt inaugurated as President of the
United States; banks closed for three days throughout the country.
George Balanchine arrived in New York.
1934 The School of American Ballet officially opened.
The Nutcracker (Tchaikovsky) staged at Vic-Wells in London with
help of Nikolay Sergeyev.
Rostislav Zakharov/Boris Asafiev, The Fountain of Bakhchisarai.
George Balanchine/Pyotr Tchaikovsky, Serenade (New York).
1935 Martha Graham, Imperial Gesture.
Wu Xiaobang choreographed first modern ballet in China.
xxxvii Chronology

1936 Broadway musical, On Your Toes (Rodgers/Hart), choreography


George Balanchine.
Spanish Civil War.
The Ballets de Monte Carlo, founded by René Blum.
1937 George Balanchine/Igor Stravinsky, Apollo, Le Baiser de la fée, Jeu
de Carte (New York).
1938 The Ballet Russe de Monte Carlo (a descendant of René Blum’s
Ballets de Monte Carlo) opened in 1938 and lasted into the 1950s.
Léonide Massine artistic director.
Ivo Váňa Psota/Sergey Prokofiev, Romeo and Juliet premiere in
Brno, Czechoslovakia.
(November) ‘Kristallnacht’ – violent anti-semitic pogrom
organised by Nazi party.
1939 Valborg Borchsenius/ Harald Lander, La Sylphide.
(September) Britain and France declared war on Germany,
beginning of the Second World War.
1940 First retrospective exhibition of Cuban art at the University of
Havana.
Leonid Lavrovsky/Sergey Prokofiev, Romeo and Juliet (Moscow).
Alexandra Federova’s version of The Nutcracker (Tchaikovsky) for
the Ballet Russe de Monte Carlo tour in South America.
Walt Disney film Fantasia (with music from Tchaikovsky’s
Nutcracker).
1941 (June) Germany invaded the Soviet Union.
Japanese aircraft destroyed US American fleet at Pearl Harbor,
United States entered the Second World War.
1942 Nazi state agreed on “Final Solution of the Jewish Question”.
US Navy defeated Imperial Japanese navy at Battle of
Midway.
1942–1943 (September – February) Soviet Red Army defeated the German
Wehrmacht in Battle of Stalingrad; German retreat from Russian
territory began.
1943 Alicia Alonso danced Giselle in a performance of the Ballet
Theatre in New York.
1944 Broadway musical Song of Norway (Robert Wright/George
Forrest), choreography George Balanchine.
(June) Allied landings on beaches of Normandy.
First Nutcracker production in the United States by Willem
Christensen in San Francisco.
1945 (8 May) German armed forces surrendered to allied forces
unconditionally.
(June) Charter of the United Nations signed.
(August) first atomic bomb dropped on Hiroshima.
(August) Imperial Japan surrendered to allied forces
unconditionally.
George Orwell, Animal Farm.
xxxviii Chronology

1946 Ballet Society, a subscription only company founded by George


Balanchine and Lincoln Kirstein.
George Balanchine/Paul Hindemith, The Four Temperaments
(New York).
1947–52 Marshall Plan or European Recovery Program.
1947–89 The Cold War in Europe.
1948 William Shockley invented transistor radio.
Ballet Society became New York City Ballet, first season: Concerto
Barocco, Orpheus and Symphony in C.
Alicia Alonso founded the Ballet Alicia Alonso in Havana; staged
Giselle for it.
1949 Roland Petit, Carmen.
José Limón, The Moor’s Pavane.
Peoples Republic of China declared.
George Balanchine/Igor Stravinsky, Fire Bird (New York).
1950 National School of Ballet Alicia Alonso opened (Harana).
1953 Harald Lander, La Sylphide Grand Ballet du Marquis de Cuevas.
Death of Joseph Stalin.
1954 Beijing Academy of Dance founded.
George Balanchine/Pyotr Tchaikovsky, The Nutcracker New York
(performed annually ever since).
1955 Maurice Béjart/Pierre Henry and Pierre Schaeffer, Symphonie pour
un homme seul.
Frederick Ashton/Sergey Prokofiev, Romeo and Juliet, the Royal
Danish Ballet.
Serge Lifar/Sergey Prokofiev, Romeo and Juliet (Paris).
1956 Bolshoi Ballet in London with Romeo and Juliet.
Leonid Jacobson/Aram Khatchaturian Spartacus, Kirov Theatre
(Leningrad).
1957 George Balanchine, Square Dance (Antonio Vivaldi/Arcangelo
Corelli), Gounod Symphony, Stars and Stripes (Philip Sousa), Agon
(Igor Stravinsky).
Treaty of Rome established European Economic Community.
1958 John Cranko, Romeo and Juliet (Milan).
Alvin Ailey founded American Dance Theatre.
1959 Fidel Castro Prime Minister of Cuba.
George Balanchine/Martha Graham/Anton von Webern,
Episodes.
1960 Frederick Ashton/John Lanchberry arr. Hérold, La Fille mal
gardée, Royal Ballet.
Elsa Marianne von Rosen, La Sylphide (Stockholm).
Maurice Béjart founded the Ballet de XXe siècle.
George Balanchine, Donizetti Variations, New York City Ballet.
1961 Frederick Ashton/Edouardo Deldevez, Les deux pigeons, Royal
Ballet (London).
(13 August) East Germany sealed off Berlin by Wall.
xxxix Chronology

1962 Cuban Missile Crisis.


George Balanchine/Felix Mendelssohn-Bartholdy, A Midsummer
Night’s Dream.
New City Ballet tour to Soviet Union.
1963 George Balanchine/Igor Stravinsky, Movements for Piano and
Orchestra.
1963–75 Vietnam War.
1964 Hongse Nianzijun or Red Girl’s Regiment performed in Bejing.
1965 White Haired Girl performed at Shanghai Academy of Dance.
Kenneth MacMillan, Romeo and Juliet, Royal Ballet with Margot
Fonteyn and Rudolf Nureyev.
George Balanchine/ Nicolas Nabokov, Don Quixote (New York).
Yvonne Rainer, Parts of some sextets.
1966–76 Culture Revolution in China.
1967 Hans Brenaa, La Sylphide (Copenhagen).
George Balanchine/Gabriel Fauré/Igor Stravinsky/Pyotr
Tchaikovsky, Jewels.
Maurice Béjart, Messe pour les temps presents, text: Buddha, Song
of songs, Friedrich Nietzsche, music: Pierre Henry, military
marches, traditional Indian and Japanese music.
1968 George Balanchine/Leo Délibes, La Source, New York City Ballet.
Yury Grigorovich/Aram Khachaturian, Spartacus, Bolshoi Theatre
(Moscow).
Student Revolution in Paris.
Prague Spring (Czechoslovakia dismantled Stalinist Regime).
Warsaw forces suppressed Czechoslovak liberal regime.
1969 Leonid Jacobson, Vestris.
1970 George Balanchine/George Gershwin, Who Cares?
1971 Elsa Marianne von Rosen and Allan Fridericia, Napoli,
Gothenburg Ballet, Kirov Ballet in St Petersburg and the Royal
Swedish Ballet in Stockholm.
1972 Alicia Alonso’s Giselle production staged at the Paris Opéra.
1974 Kenneth MacMillan, Manon, Royal Ballet, Covent Garden.
1975 Kenneth MacMillan, The Four Seasons, Royal Ballet.
1976 New York Baroque Dance Company founded by Catherine Turocy
and Ann Jacobi.
1977 Rudolf Nureyev, Romeo and Juliet, London Festival Ballet.
1978 Jiřı́ Kylián, Symphony of Psalms.
1979 First Bournonville Festival in Copenhagen.
Peter Schaufuss, La Sylphide, London Festival Ballet.
Maurice Béjart, Un instant dans la vie d’autrui.
1980 Ris et danceries founded by Francine Lancelot.
1980s Francine Lancelot decodes Raoul Anger Feuillet’s dance
notation.
1981 George Balanchine/Tchaikovsky, Mozartiana (revised).
1983 L’Eclat des Muses founded by Christine Bayle.
xl Chronology

1984 William Forsythe director of Ballett Frankfurt; Artifact.


Flemming Flindt, La Sylphide, Dallas Ballet.
1985 Mikhail Gorbachev became General Secretary of the Communist
Party of the Soviet Union.
L’autre pas founded by Klaus Abromeit.
Toni Lander and Bruce Marks reconstruction of Abdallah by
August Bournonville.
1986 Mark Franko, Le Marbre tremble.
1987 Atys by Lully (baroque revival by Arts Florissants).
1988 Henning Kronstam, La Sylphide (Copenhagen).
François Raffinot, Caprice.
Guy Debord, Society of the Spectacle.
1989 Ayatollah Khomeini issued Fatwa demanding execution of Salman
Rushdie, author of the Satanic Verses.
Mark Morris, Dido and Aeneas.
Berlin Wall fell, disintegration of Eastern European Socialist bloc.
1990 Maurice Béjart/Richard Wagner, Ring um den Ring.
Federal Republic of Germany absorbed German Democratic
Republic.
World Wide Web originated at CERN, European Organisation for
Nuclear Research.
1991 Jiřı́ Kylián, Petite Mort.
Frank Andersen, A Folk Tale (together with Anne Marie Vessel
Schlüter).
1992 Second Bournonville Festival in Copenhagen.
Dinna Bjørn and Frank Andersen, Napoli, Royal Danish Ballet
(also 1998, 2005).
1993 Fêtes Galantes founded by Beatrice Massin.
1994 The ZIP drive with removable computer storage developed.
Maurice Béjart, Le Ballet des mots.
1995 William Forsythe, Eidos Tellos.
1997 Dinna Bjørn, La Sylphide (Copenhagen).
DVD went on sale for first time.
1998 Maurice Béjart, Nutcracker.
1999 Soviet Union abolished.
Frank Andersen, La Sylphide, Chinese National Ballet, Inoue Ballet
and the Royal Swedish Ballet.
Beatrice Massin, Le Roi danse.
2003 Tim Rushton Napoli – den nye by (Napoli – The New Town) New
Danish Dance Theatre, Copenhagen.
Nikolaj Hübbe, La Sylphide (Copenhagen).
Hans van Manen/Johann Sebastian Bach, Monologue, Dialogue.
2004 Thomas Lund and Johan Holten, En anden akt (Another Act)
Copenhagen International Ballet, Bellevue Teatret.
Jiřı́ Kylián/Dirk Haubrich, Sleepless for Nederlands Dans
Theater II.
xli Chronology

2005 Jiřı́ Kylián/Dirk Haubrich, Toss of a Dice for Nederlands Dans


Theater I.
Dinna Bjørn and Frank Anderson, Napoli Finnish National Ballet.
Third Bournonville Festival at the Royal Theatre in Copenhagen.
Ulrik Wivel, Jeg Dig Elsker (I love you) (film on La Sylphide).
Dinna Bjørn and Frank Andersen, Napoli Finnish National Ballet
in 2005.
2006 Peoples Republic of China declared Ghengis Khan Chinese.
Introduction

m a r i o n ka n t

Another ballet book?


Not exactly. A Companion, a book to accompany you when you go to see
ballet or when you want to know something, find a name, a date, a work; it
is a book to read and a book to consult, not on everything in ballet, not a
complete book, not even a complete history. A book with essays that revisit
aspects of this beloved and detested art form, a book that is needed as much
as ballet is needed. Yes, “needed”, as we need our bodies. Ballet shows us
what the trained human body can do to make flesh become art. Ballet, the
art of the body, puts our physical presence into form, into fantasy and into a
deeper reality. “Needed” because in the nineteenth century ballet became
a uniquely feminine enterprise, and to an extent it has retained this aspect
in the twenty-first century. The great ballerina floats before us, telling us
something about ourselves, our genders, our fears, hopes and, above all,
prejudices. The artificiality and conventions of ballet protect our nervous
sensibilities but also indirectly assault them.
Ballet is part of our history and our heritage and if we deny our past we will
not comprehend the present nor grasp the future. Either in life or in dance.
Ballet has a history, which reflects and refracts the social order in which it
arises. A courtly society demands an art form in which the king can dance
and his court revolve around him. A revolutionary society proclaims its
ideals in dance; a conservative one does the same. Ballet, adored and reviled
often for similar reasons, belongs to our contemporary cultural landscape
as much as any other performance art. Ballet tells us about ourselves and
the world we inhabit; it holds up a mirror and projects wishes and desires;
it expresses our ideals and mocks our vanities; it demolishes certainties and
tests limits; it creates values and sets standards; it invents the past as much
as the future.
Constant reflection, introspection and evaluation are part of a process
which helps us assess critically where we are and how we got there; then
we can make choices for the future and think about alternatives. Such a
process of thinking must never end; it is a vital component of our artis-
tic consciousness. Thinking about ballet poses special problems because it
creates its meanings without words. It literally “embodies” meaning. It lets
us know in its own way what being human, having arms and legs, a gait,
[1]
2 Marion Kant

a posture, a gesture, means or could mean. Ballet has changed but never
gave up its history, its link to the past: the invention of the “white ballet”
in the nineteenth century revolutionised and transformed a courtier’s code
into the “woman question”. Every time a “classical” ballet is performed, it
reminds us that the “woman question” still has no satisfactory answer. But
the shape in which it is presented has significantly been transformed. Those
new and old forms transport new and old philosophical questions alike.
The authors gathered together in this book come from many parts of
the world; they stand for a truly international crowd, as international as
ballet itself. The methods with which they work reflect different attitudes
in research and highlight the fact that ballet is not one homogenous thing
but a flexible art as well as a serious academic research subject. Many of the
contributors combine practice and theory of dance and ballet and can thus
offer a wide range of experience.
This Companion to Ballet consists of twenty-four chapters gathered in
four parts on various aspects of the art form’s historical evolution and on
its aesthetic properties. They consider the most important developmental
stages since its “origins” in the fifteenth century and travel to the end of the
twentieth century, sometimes stopping to consider a particular movement,
sometimes to contemplate a specific moment of importance, sometimes to
review a particular artist’s contribution or a specific choreography.
All the essays consider an art of the highest public interest and discuss
ballet as an expression of “modern” ideas at various times. We never intended
to provide a complete history or a full theoretical framework of aesthetic
ideas. Instead we offer an impression of the most important elements, which
demonstrate continuity and gradual evolution as well as those which mark
sudden changes of artistic direction. This Companion is not a history book
in the strict sense; yet it cannot ignore the importance of historical develop-
ments. Thus it has been organised along historical axes, for ballet is an art
form, which has grown over time and peculiarly and intimately symbolises
the age in which it arises.
In the first chapter we begin our journey in northern Italy to which
Jennifer Nevile takes us and introduces us to the dance manuals of the
fifteenth century, their philosophical ideas and the structure of an art form
which they describe. We then make a leap to France and England and their
royal courts. The genres of the ballet de cour and the masque took up the
challenge of the ballet that was born a couple of centuries before in free
city states and had been modified to celebrate the glory of emperors and
empresses. Marina Nordera and Barbara Ravelhofer present court ballet as
an integral part of a social order. Mark Franko introduces the “baroque
body” – a concept through which a historical moment is preserved today; a
3 Introduction

corporeal reality we can no longer recall but have to imagine and invent in
order to understand the past.
In the second part we follow the transformation of court ballet to a
bourgeois public expression. The modernisation of ballet in the eighteenth
century lay in its opening up of spaces and its new approach to the human
body and human movement. Dorion Weickmann supplies an overview of
the concept of the ballet d’action and Sandra Noll Hammond explains the
training principles that led to a complete professionalisation of dance. Tim
Blanning and Judith Chazin-Bennahum write about two important and
innovative choreographers and dance theoreticians: John Weaver and Jean-
Georges Noverre. We leave the eighteenth century with the French Revo-
lution, the event that changed all European societies and countries for the
next 200 years. With its radical conceptions of a “new man” it also redefined
the place, the order, the structure and aesthetics of human movement.
The third part looks at Europe and Russia after the French Revolution.
Romanticism had descended upon France and Germany and quickly spread.
Sarah Davies Cordova and Anne Middleboe Christensen evoke romantic
representations in France and Denmark and the emergence of the ballerina
as the ultimate embodiment of romantic ideas. Marian Smith reminds us of
the close interaction between music and dance and recalls the often forgotten
practices of composing for ballet. Lynn Garafola and Thérèse Hurley take us
to Russia where the Italian-French Marius Petipa dominated ballet for two
generations and together with the composer Tchaikovsky created ballets
that belong to the canon of the art form today. Lucia Ruprecht argues that
ballet as an art form in the public sphere very much depended on the
critic; ballet entered the consciousness of the bourgeois audience through
its written reflections in news-papers and journals as much as through
its nightly performances. The critic was then, as now, an institution – an
advocate for or against, a propagandist who used ballet to advance more
than only a personal opinion. The chapter ends with my investigation of
the ballet costume, especially skirts and shoes, in their cultural context. I
interpret their formal properties and meaning: the full white skirt and the
pointe shoes still symbolise ballet and the power of the female dancer.
The last and most extensive part examines the twentieth century. Ballet
had become a well-established art form, no longer accessible only to an
elite but to many – high and low. It also carried historical baggage with it,
had over the centuries modernised and reinvented itself several times. With
choreographies and performance styles, dance philosophies had developed.
Dance had been “done” but also written; it had become memory of physi-
cality and incarnation and had entered the intellectual sphere of European
culture. By the early twentieth century ballet had to cope with its own history
and the stereotypes and tropes it had created. With the Russian and Swedish
4 Marion Kant

Ballets of the early twentieth century – both companies as well as aesthetic


principles – Tim Scholl and Erik Näslund recall important instances of
another modernisation process in ballet. The Russian as well as the Swedish
Ballet soon were regarded as revolutionary breaks with tradition, yet both
were also firmly connected with ballet’s tradition as well as to other social
and artistic developments of the time. The Russian Revolution of 1917,
very much like its French predecessor, fundamentally shaped the European
landscape and also affected ballet. From post-revolutionary Russia came
the man who is synonymous with ballet – George Balanchine. Matilde
Butkas traces Balanchine’s career from Russia to Paris and eventually to the
United States, whereas Juliet Bellow pauses in Paris to show how Balanchine
broke down the notion of classicism in the 1930s before reinstating it in
the 1960s. Jennifer Fisher focuses on one ballet – The Nutcracker – and
describes how it was transformed from near failure to the most successful
and most often performed piece of our contemporary era. Every small com-
pany can today realise its ambitions by staging a Nutcracker and integrate
references to regional politics, cater to local tastes and satisfy the native
community’s demands. With the chapters by Zheng Yangwen on ballet in
China and Lester Tome on Alicia Alonso’s Cuban Giselle we see ballet in
its worldwide context and understand the politics of internationalisation.
Ballet from its very beginning had been an international affair. In the first
century of its existence it had crossed the European continent and when the
colonial powers spread their cultural ideals they took with them their value
systems. Ballet too suited the needs of these powers to proclaim their ideas
of rule and order. But it was neither a simple nor a one-sided relationship.
Ballet, as other arts, is never just a tool in the hands of a regime to control nor
a means to suppress indigenous interests. Thus for Zheng Yangwen ballet,
like opium or communism, offered itself as a vehicle to translate contradic-
tory beliefs in a society in full transformation. In China and in Cuba this
initially foreign artistic articulation offered itself as an agent to formulate
those national principles which it was supposed to help replace. My chapter
ends the part by revisiting the ideological challenges of the entire century.
I too emphasise that art and ballet have never been static but served the
diverse needs of those who took up the challenge of expressing through
movement the problems and tensions of their contemporary world.
The division in the twentieth century between ballet and modern dance
has produced two very different types of movement art, often fiercely and
destructively hostile to each other. The final chapter in Part IV tells the
story of modern dance as a counterpoint to ballet in the same period. It
is to be hoped that in due course modern dance will be granted the full
treatment which it deserves in a Companion of its own. As editor I had to
take another, difficult decision: did Martha Graham, Merce Cunningham
5 Introduction

or Siobhan Davies belong in this book? I decided that these three as all other
modernists certainly played an important part in the story of theatrical
dance in the twentieth century but that they represent a different genre of
dance. Their motivations in dance, their philosophies, movement codes,
structural and formal features have to be treated independently from ballet.
The Companion does not provide a guide to becoming a ballet dancer.
Problems of professional training, anatomical requirements and medical
advice for dancers or nutritional questions have deliberately been excluded.
Neither has the history of companies and theatre institutions, nor the rich
history of ballet design been observed. Many names will not appear and
many works are only mentioned in passing. But the omissions should make
you, the reader, want to find out more, search for that volume which will
solve your question and lead you to the next level of understanding. This
collection of chapters should whet your appetite and rouse your curiosity.
We hope the Companion will live up to its name – that it will accompany
you as a useful guide and open up the complex, fascinating world of theatre
and ballet. We hope that it will be a friendly and reliable escort during a
first encounter with ballet as well as a good interpreter during future visits.
My fondest wish is that at some future performance you, the reader, will
remember a fact, a name or an evolution of technique that you first read
in this book and now see on stage. If the pleasure of recognition or the
application of information sharpens your enjoyment of the work before
you, the Companion will have done its job.
pa rt i

From the Renaissance to the baroque: royal


power and worldly display
1 The early dance manuals and the structure of
ballet: a basis for Italian, French and
English ballet
j e n n i f e r n ev i l e

In the dance treatises of the fifteenth century choreographies were first


recorded in Western Europe: dance became literary and philosophical as
well as a physical skill and oral tradition. These treatises laid the foundation
for the future structure of European dance. The main dance genres recorded
in the treatises, ballo, bassadanza, and basse danse in the fifteenth century,
pavane, galliard, branle, almain, balletto, bassa, brando and cascarda in the
sixteenth century, were the dances of the upper levels of society: the courtiers
and nobility, those who wielded power, as well as the wealthy merchants and
trading families. Important state occasions, marriage celebrations, official
visits by neighbouring rulers or ambassadors, annual religious festivities
and theatrical events were all marked by formal balls or dancing at which
members of the elite performed. Often these dance events took place in
public spaces, on a stage erected in the main piazza in front of thousands
of spectators. At other times the space in which the dancing was conducted
was more private, being the main hall of a palace. But even on these occa-
sions the dances performed were part of the official ceremonies and rituals,
contributing to the presentation of the image of a ruler as a powerful and
magnificent prince, whose authority could not be challenged. When a ruler
and the leading members of his court danced in public before his subjects
he was displaying his magnificence, and in doing so he was displaying his
power. The Italians in particular were obsessed with protocol and ceremony,
and one of the chief means of indicating rank was by spatial relationships
among people. Thus dance, an art form with spatial relationships as its basis,
was a significant tool in this presentation of power and rank through rituals
and ceremonies.
The dances recorded in the treatises are overwhelmingly for both male
and female performers (see Fig. 1). Many of the dances are for one or
two couples, or for three performers, two men and one woman or vice
versa. Some dance genres were processional in nature, for example the basse
danse, pavane and almain, during which a line of couples paraded around
the hall, exhibiting not only their skill at dancing, but also their sump-
tuous clothes, hairstyles and jewellery. Other genres chronicled the social
[9]
10 Jennifer Nevile

Figure 1 Maestro dei Tornei di Santa Croce, cassone panel, La magnanimità di Scipione, 1460.
London, Victoria and Albert Museum, Inventory No: 5804.1859.

interactions between the men and women. The choreographic sequences


and floor patterns of the Italian balli, for example, emphasised typical inter-
actions that occurred on a daily basis at court. Some of the balli enact
themes of fidelity, fickleness or jealousy. Sobria is a ballo for one woman and
five men where the sole woman remains faithful to her partner despite the
advances and pleading of the other four unattached men. Merçantia, for one
woman and three men, presents the opposite scenario, as the woman is all
too ready to abandon her partner and flirt with the other two men. Gelosia
(jealousy), a ballo for three couples, is a dance in which the men constantly
change partners, thereby providing many opportunities for the display of
this emotion. In the sixteenth century the confrontation between the sexes
became more explicit with dances entitled Barriera, La Battaglia and Torneo
Amoroso. Often these dances started with two lines of men and women who
advanced and retreated before clashing (often striking hands that echoed
swords hitting shields) and the final reconciliation. Other dance genres such
as the galliard were explicitly choreographed for a display of virtuosity and
athleticism, especially on the part of the man, who was expected to perform
sequences of complicated variations that could involve kicks, leaps and turns
in the air. Hundreds of these variations were recorded in the dance treatises,
and competent dancers were expected to memorise many of them, to be
used at will during a performance. By the sixteenth century the necessity
for a courtier to be skilled in the art of dance was without question. The
ability to perform gracefully, seemingly without any effort, was one of the
11 The early dance manuals and the structure of ballet

distinguishing marks of a courtier and the absence of this ability exposed a


gentleman or lady to ridicule and derision from colleagues.
The dance treatises from the fifteenth and sixteenth centuries (often
dedicated to members of the leading families) contain hundreds of chore-
ographies – a substantial body of material. Perhaps the most obvious con-
tribution of these treatises to European dance practice is the idea of a chore-
ography as a unique arrangement of steps, floor patterns and music. Dances
such as the pavane, in which a simple sequence of steps was repeated until
the end of the music, continued throughout this period, but the vast major-
ity of dances recorded in the treatises were individual choreographies. A
dance was therefore a specific creation: it needed a creator – usually a dance
master – who also often wrote or arranged the music to fit the step sequences
of each dance. The fifteenth-century Italian dance master Guglielmo Ebreo
recognised dancing as an innate, natural human activity.

[I]f eight or ten people are dancing without music, [but] with steps that are
harmonised and measured together, then it is a natural thing. And when a
musician plays and those dancing harmonise and measure their steps to the
music, then it is an acquired skill.1

But when music was played and the dancers adjusted their steps to fit the
music, then dance became an art, a product of human ingenuity and skill. In
this latter scenario the “natural” product of dance was ordered and perfected
by the addition of human application and skill, and training and education
in the dance.
Each new dance was a unique combination of steps drawn from the exist-
ing step vocabulary. The number of different steps available to the choreog-
rapher increased dramatically in the sixteenth century. The step vocabulary
of the fifteenth-century Italian ballo and bassadanza were nine “natural”
steps and three “accidental” steps, with variety obtained from performing
the steps of one misura (a specific combination of metre and speed) to the
music of a second misura, and from adding the quick “accidental” steps
to the “natural” steps.2 By contrast, the late sixteenth-century Italian dance
treatises of Fabritio Caroso contain descriptions of fifty-eight different steps
(Il Ballarino, 1581) and seventy-four (Nobiltà di dame, 1600) respectively.
Cesare Negri in Le gratie d’amore (1602) describes fifty-one widely used
steps, as well as forty-two variants on the galliard cinque passi and thirty-
four different galliard mutanze, twenty-seven salti (jumps), thirty capriole,
and ten zurli (spinning turns).3 With this many steps the possibilities for
new combinations of step sequences were vast, even without the addition
of improvised passages and added ornamental steps.4
The structure of these individually choreographed dances enhanced the
importance of memory in European ballet, and led to the requirement for
12 Jennifer Nevile

sustained rehearsal. A good memory was crucial for anyone who wished
to perform in public, as one had to commit to memory each different
choreography. Dances were subject to fashionable trends, and those in the
elite level of society had no wish to be seen performing last year’s dances,
let alone those of five or ten years ago, which had now filtered down to a
lower level of society. Therefore, new dances had to be continually learnt
and mastered. For example, in 1469 the dance master Filippus Bussus wrote
to Lorenzo de’ Medici offering to come to Florence in order to teach Lorenzo
and his siblings “some elegant, beautiful and dignified balli and bassadanze”.
According to Bussus, the performance of such new and elegant dances would
bring “honour and fame” to Lorenzo and his family.5 The letter from Bussus
highlights the need for rehearsal before these dances were performed in
public. Thus dance education began at an early age for the children of the
nobility. Ippolita Sforza was only ten when she danced at Tristano Sforza’s
wedding celebrations in Milan in 1455, while Isabella d’Este started her
public performances from the age of six.
A high level of skill was needed in order to perform gracefully in public,
without error. A dancer had to be able to learn the correct carriage of the
body, to master the steps and their variants and to memorise the choreogra-
phies. Furthermore, he or she had to possess a thorough understanding of
the interaction between the dance and the music, the ability to adapt the
patterns of each dance to the available space, the wit and invention to subtly
vary each step so that it was not performed the same way several times in a
row, a knowledge of the gestures and movements of the body which accom-
panied the steps, an awareness of the phrasing of each step as well as the
agility to cope with the speed changes in the choreographies. An example of
the difficulty of mastering the mechanics of the dance practice as an adult,
let alone its subtleties, is illustrated by a letter from the German painter
Albrecht Dürer. Dürer, while on a visit to Venice in 1506, wished to improve
his social standing. Apart from buying new and luxurious clothes, Dürer
also enrolled in a dancing class. He found this part of the process of social
ascension much more difficult than just purchasing expensive clothes, as
the somewhat complaining tone of his letter reveals. “I set to work to learn
dancing and twice went to the school. There I had to pay the master a ducat.
Nobody would make me go there again. I would have to pay out all that I
earned, and at the end I still wouldn’t know how to dance!”6
Dürer’s letter vividly illustrates that dancing was a social marker, a means
of distinguishing those who belonged to an elite group from those who
did not. In the thirteenth and fourteenth centuries dance had always been
regarded as a normal aristocratic pastime. But from the fifteenth century
onwards the art of dance as described in the dance treatises became a sign
of membership in the upper levels of society. The rules and postural codes
13 The early dance manuals and the structure of ballet

as taught by the dance masters were part of the mechanism by which the
court made itself appear superior and inaccessible to the rest of society. The
courtiers believed that their superiority should be demonstrated to the rest
of society by the different way in which they moved, walked, danced and even
stood in repose. Their carriage and demeanour when on the dance floor did
not change once they finished dancing: it remained with them as it became
their normal posture. Thus the instruction the young children received
from the dance masters was extremely important socially and ethically, as
it not only allowed them to obtain approbation when they exhibited their
skills in the dance, but it also trained them in the patterns of behaviour
and deportment essential for membership in the social elite. If you moved
ungracefully you immediately demonstrated to others that you did not
belong to the right class of society, as you could not perform the movement
patterns appropriate to that class. Dancing taught the chosen members of
society control over their body and over all their actions, both when dancing
and in day-to-day interactions with their colleagues and superiors. It was
visible evidence that a person was capable of appearing in public without
making an exhibition of herself or himself. If a person could control his or
her outward bodily movements, then they were capable of controlling their
inner emotions as well. Dancing, therefore, functioned as a social marker,
as one of the ways a certain group in society defined itself and excluded
others. A stark example of how dance was used to define the elite in society
comes from Nuremberg in 1521. In this year those who held political power
wished to limit further the numbers of citizens entitled to vote. Therefore
they designated the voting elite as “those families who used to dance in the
Rathaus in the olden days, and who still dance there”.7 In Nuremberg it was
the ability to dance that was used as a tool to exclude people from the group
who exercised political power.
Dance as an elite activity was strengthened in the Renaissance because
it became a form of consumption: a consumption of both time and money.
The dance practice recorded in the treatises needed many hours of teaching
and practice from a young age, and only those who were wealthy enough to
have the leisure time to devote to this activity were able to participate.
The fifteenth-century Italian dance treatises were more than just a com-
pilation of choreographies.8 These manuscripts also contained a theoretical
section in which the steps were briefly described, rules for the mastery of the
dance were given, as well as the essential principles of the art and its philo-
sophical basis. In their treatises the dance masters argued for dance to be
included among the liberal arts. The authors of the treatises, Domenico da
Piacenza, Antonio Cornazano and Guglielmo Ebreo, were fully aware that
for dance to be included in the liberal arts through its association with music,
it had to be understood both on a physical and intellectual level. If dance was
14 Jennifer Nevile

a liberal art it could then lay claim to be a demonstration of eternal truths,


and a path to understanding the nature of God and the universe. The dance
masters, especially Domenico, devoted a great deal of attention to setting
out the philosophical basis of the art of dance that had the same numerical
basis as the other mathematical arts of the quadrivium. These numerical
proportions that were believed to order the cosmos found expression in the
ratios of the relative speeds of the four misure – bassadanza, quaternaria,
saltarello and piva misura – out of which the balli were constructed.9
Dance masters were also concerned with eloquent movement. Ever since
the late fourteenth century, Italian humanists had been passionately con-
cerned with eloquence in spoken and written text. The humanists’ profes-
sional activity was the use of words, and so the production of elegant prose
or poetry was one of the chief aims. The professional métier of the dance
masters was movement of the human body, and this was where they strove
to inculcate eloquence. For Domenico and his colleagues a person’s ges-
tures, deportment, facial expressions and manner of walking were a silent
language that carried a rich treasury of meaning.
Given the concern of the dance masters with the performance of elegant
patterns of movement, it is not surprising that a large part of the specialised
technical vocabulary developed in their treatises dealt with nuances of these
eloquent movements. Maniera, aiere, gratia (grace), ondeggiare, campeg-
giare10 and fantasmata11 were all terms developed by the dance masters
in their attempts to describe elegant movements, all of which involved a
fluidity and flexibility in the dancer’s body. For movements to be eloquent
they also had to be in harmony with the gestures and steps performed (a
constant refrain in Guglielmo’s treatise), as well as with the clothes worn by
the dancer. Movements that would look dignified and seemly when dancing
in a long garment would appear slightly ridiculous when dancing in a short
tunic. Similarly, the jumps, turns and flourishes that appear elegant when
wearing a short garment would have the opposite effect if observed on a
dancer in a long tunic. Elegant movements, therefore, were not always slow
or stately. They could be vigorous and lively when appropriate.
The engagement of the dance masters with contemporary intellectual
concerns continued in the sixteenth century. From the mid-sixteenth cen-
tury onwards the intellectual climate in both continental Europe and Eng-
land fostered an interest in symbols. Indeed, the “manipulation and inter-
pretation of symbols became a popular intellectual sport in the sixteenth
century”.12 Symbols were seen as a great force both to draw heavenly power
down to earth, and to help raise human understanding closer to a knowledge
of the divine. Marsilio Ficino, in his treatise De vita (1489) is explicit on the
power of “figures” (figurae) to influence human activity. And these “figures”,
or magical symbols that contained a hidden power, included music, people’s
15 The early dance manuals and the structure of ballet

gestures, facial expressions, movements and dance.13 Ficino’s writings were


very popular in France in the sixteenth century, not only De vita, but also his
Commentary on Plato’s Symposium on Love, which influenced many poets
including Ronsard and Baı̈f, both of whom were involved in the danced
spectacles at the French court. In Ronsard’s own writings he says that dance
is divine and has the power to bring those divine effects down to the earthly
sphere to transform those who perform and watch it.14

The evening that Love enticed you down into the ballroom to dance with
skill, the wonderful dance of Love . . . The dance was divine . . . now it was
circular, now long and then narrow, now pointed, as a triangle . . . I mistake
myself; you did not dance, but rather your foot touched the summits of the
earth; and your body was transformed, for that night, into a divine nature.15

The aim of the dance master when choreographing a dance for a masque
or fête was to present to their educated audience of the court a series of
symbols and images that communicated messages to the viewers just as did
the spoken or sung portions of the spectacle. To them dance was a lan-
guage analogous to writing, a form of “moving script” and the geometric
figures created by the physical bodies of the dancers were the equivalent of
the words and sentences of a spoken or written text. Thus choreographers
built the dance around a series of discrete geometric figures or patterns that
continually changed over the course of the dance. Each different rearrange-
ment of the geometric shapes of a circle, square, triangle, or symbol such
as a cross, produced a new dance figure with a different meaning. Dance
masters saw themselves as architects responsible for the design of a dance,
in geometric and proportional terms, and for these choreographers the true
beauty of their dances resided in the geometric figures. Geometrically pat-
terned choreographies were one way in which cosmic influences could be
magically (that is, in a hidden or occult manner) transported to earth and,
once there, could induce the same cosmic harmony to operate on earth.16
Dance was seen as a form of alchemy: an alchemy that acted upon perform-
ers and viewers, refining and transforming them into a purer state, closer
to that of the divine nature. Therefore the notebook of a French dancing
master who worked in Brussels c.1614–19 does not contain written descrip-
tions of steps as do the treatises of Caroso, Negri and Arbeau, but rather
over 450 drawings of figures for five to sixteen dancers.17 Many of the fig-
ures are simple geometric shapes – squares, circles, triangles, pyramids and
lozenges – as well as composite figures in which discrete geometric figures
are combined to form larger figures. Other named shapes are alchemical
images: the salamander, serpent, tortoise, dart, sun, moon and star.18
Thus the interest in images and symbols that prevailed in society in gen-
eral affected the choreographic structure of the theatrical dances in the late
16 Jennifer Nevile

1500s in France and England. By this time alchemical images and emblem
books circulated widely, and by the beginning of the seventeenth century
the language, images and metaphors of alchemy “were available to anyone
who cared to read or listen”.19 It is inarguable that the texts of the masques
and the French fêtes were written by the leading poets of the time, all of
whom would certainly have been well aware of alchemical ideas and images,
but the choreographers would also have had access to this imagery through
the large number of printed books available on the subject, as alchemical
books were part of the staple literature of the reading public.20 By the end of
the sixteenth century the corporeal rhetoric and eloquent movement of the
previous century had given way to a language of symbols – alchemical and
mystical – that spoke directly to those watching who had the intelligence to
comprehend and interpret what they saw.
A further contribution of the early dance manuals to European ballet
was the ideal of dance as morally virtuous behaviour. In quattrocento Italy
it was a commonly held belief that movements of the body were an out-
ward manifestation of the movements of a person’s soul.21 This belief was
part of the reason for the dance masters’ cultivation of, and insistence on,
elegant movements, and the distancing of their choreographies from the
dances of the peasants. Thus those who moved in an ungraceful and inele-
gant manner in public exposed their inner nature for all to see. There was
more at stake than momentary ridicule for one’s clumsiness. Vulgar move-
ments that were not eloquent would be a clear sign that a person’s soul
was not virtuous. Dancers could move those who saw their performance to
sorrow, anger, happiness or laughter, as the emotions of the dancers were
made visible through the movements of their body. This gave them both
a tremendous power and responsibility: a power to affect the emotions of
those who watched, and the responsibility to represent only morally edifying
emotions.
Thus dance had the ability to teach ethical behaviour. A virtuous person
when dancing would be imitating in his or her movements various positive
ethical states, these would then be recognised by the spectators, who could
themselves learn to imitate these virtues in their own lives. Naturally, the
reverse position was also true; that is, a dancer’s movements could represent
negative emotional states. Guglielmo did not seek to deny that the art of
dance could be abused and used for immoral or improper purposes. But
he also argued that when used by virtuous, noble and moral men, it could
have a positive ethical effect on its practitioners and on those who observed
it. Even though Guglielmo devoted an entire chapter on the behaviour and
demeanour expected of young women of gentle birth, which he said had
to be more moderate and virtuous than that expected of young men, he
was not excusing a low level of behaviour from the latter. They still were
17 The early dance manuals and the structure of ballet

expected to exhibit courteous and virtuous behaviour. Therefore, the dance


masters had an interest in promoting the moral virtues of the art of dance,
and in emphasising its benefits for society as a whole, as is illustrated by the
following passage from Guglielmo’s treatise:

But when it is practised by noble, virtuous and honest men, I say that this
science and art is good, virtuous and worthy of commendation and praise.
And moreover not only does it turn virtuous and upright men into noble
and refined persons, but it also makes those men who are ill-mannered and
boorish and born into a low station into a sufficiently noble person. The
character of everyone is made known by the dance.22

In the sixteenth century the ability of dance to teach moral truths


was carried further by northern European writers. The status of dance
as a medium of moral instruction was elevated by the publication of
such works as Thomas Elyot’s The Boke Named the Governour (London,
1531). Elyot’s book was an educational treatise for young boys who were
destined for careers in the country’s administration. According to Elyot,
dance was a noble and virtuous pastime, as it provided both recreation
and a means to learn and comprehend the virtues necessary for adult
life. Through the study and practice of the basse danse children could
learn the important moral truths, as each step of the basse danse sig-
nified a different aspect of prudence.23 For example, the reverence that
begins every basse danse signified the honour due to God that is the basis
of prudence, and which should be the starting point for all of mankind’s
actions.24
The written history of Western European ballet begins in the fifteenth
century with the production of the dance treatises. These manuscripts docu-
ment a dance practice many of whose characteristics continued in the ballet
of later centuries. Each dance was individually choreographed: a unique
arrangement of steps, floor patterns and music. The treatises record hun-
dreds of individual dances, as well as describing a large and sophisticated
step vocabulary. These dances required a creator, or choreographer, and
their complexity demanded that dancers learn, memorise and rehearse them
before performing in public.
The choreographies were performed by members of the aristocracy,
those who participated in the government and wielded power. These per-
formances were often part of official state ceremonies and festivities, as well
as for private entertainment. Dance was part of the display of wealth and
power of the elite, as its practice involved the consumption of leisure time
and money. The dance practice as described in the treatises, with its spe-
cific movement patterns, gestures and carriage of the body, acted as a social
marker, identifying those who belonged to the elite level of society.
18 Jennifer Nevile

The dances that are recorded in the treatises are overwhelmingly chore-
ographed for both men and women, rather than for one sex only. Thus
the dances served as a means of expressing common social interactions and
relationships between men and women. The dances also embodied societal
norms, such as the need to control the movements of one’s physical body and
inner emotions. The dances performed by the elite in society were also seen
as teaching ethical behaviour to those who watched through the external
representation of moral virtues.
Most importantly, the dance practice as described in the treatises was
part of the contemporary intellectual culture. The choreographers, either
through the corporeal eloquence of the dancers’ movements, or through the
figures and symbols formed by groups of dancers, communicated important
truths to the viewers who had the knowledge to comprehend what they
were seeing. Dance was a liberal art, and as such was one path towards
understanding the nature of God.
2 Ballet de cour
m a r i na n o rd e r a

The ballet de cour is a type of composite theatre performance, made up


of instrumental and vocal music, texts declaimed in verse and prose, stage
design, scenic accessories, costumes, masks and, not least, dance. Its “poten-
tially chaotic”1 structure includes a series of successive entries, variable in
number and type, divided into acts and culminating in a final grand bal-
let. The term “ballet” is used, therefore, to indicate both the whole piece
as an entity as well as the danced portions of it. The definition ballet de
cour was coined only in the nineteenth-century historiography2 in order to
legitimate – although that was not explicitly expressed – classical ballet by
associating it with the noble context in which it had allegedly been born.
The ballet de cour was born and developed at the French court in the
last decades of the sixteenth century and lasted – with a variety of ups and
downs, among which was a crisis of popularity during the regency of Anne
of Austria – right to the end of the seventeenth century. Its evolution tracked
the parallel consolidation of the French monarchy from the first Valois king,
Henry III, through the Bourbons Henry IV and Louis XIII. It reached its
apogee at the beginning of the reign of Louis XIV (1638–1715) in the years
after 1643 when he became king. Louis, who until 1670 regularly danced
leading roles himself, attracted the attention of his courtiers, both French
and those from foreign courts, to the ballet de cour. The progressive decline
of the form began when the king retired from active performance but also
as a consequence of altered conditions in court politics and in the relations
between the crown and society as a whole.
The evidence by which we can reconstruct the ballet de cour comes from
a variety of types of sources; among them are libretti, with the poetic texts
or the subject matter, the musical scores and the many printed illustrations,
which reproduce the staging and costumes, and descriptions that are con-
tained either in official chronicles or private documents. As a result we can
state that the corpus of works so documented runs to hundreds of ballets,3
which in spite of their great variety and differences, display certain common
features.
The narrative plot of the ballet de cour is simple and schematic, drawn
from the mythological universe or medieval romances. It is built with the
rhetorical tropes of allegory and metaphor. From the dramaturgical point of
[19] view, within the vast, authenticated repertory, there are some more and some
20 Marina Nordera

less coherent examples. Where some ballets de cour contain presentation,


development, resolution and conclusion of an intrigue or plot, others reduce
themselves to a series of episodes linked by a pretext.4 Each ballet de cour has
its origin in a celebration of a specific event or occasion in the life of the court
and in particular the royal family or a military victory5 or notable success
in international relations. Most are secular but occasionally we find that a
piece has a religious pretext. The allegorical and mythological narratives
are elaborated in such a way as to convey the significant elements which
recall the precise social or political events. As a consequence the themes
chosen, though masked by a rhetorical apparatus or sometimes by irony or
the primacy of pure spectacle, always reveal a direct, immediate and easily
recognisable relation to actuality.6 While in the major part of the cases, the
intent, well understood in such coded signs, is to serve the propaganda of
the sovereign power, there are often ironic nuances, touches of self-criticism
and even subversion.
The ballet de cour took place in the chambers of the royal palace, in
various royal residences in Paris and in the provinces. The public sat on
three sides. The sovereign took the central position on a raised throne above
the stage level on which the ballet unfolded. At the conclusion of the grand
ballet, as the final episode, the public took the stage following a practice
which had begun to be popular during the sixteenth century. There they
joined the dancers on the stage to execute a variety of social dances like the
branle. The final bal united the public and the stage dancers on the same
plane “in an ambiguous area between participation and presentation”.7 In
fact, one or more members of the royal family often danced in the produc-
tion, joined by members of the court entourage, generally male, and from
time to time accompanied by professional dancers or tumblers who took
particularly complicated or special roles. So called “hommes de qualité”,
privileged courtiers, were not supposed to measure their abilities against
the technique and virtuosity of the dance masters, as François De Lauze
made explicit in his chapter on the courante, a dance considered especially
noble, in his treatise on dance for the gentlemen and ladies of court:
n’aymerois point qu’ils meslassent parmy leurs compositions des pas qui
sentissent son baladin, comme fleurets, frisoteries, ou branslements de
pieds, piroüetes (j’entens à plusieurs tours violens ou forcez), caprioles, pas
mesmes des demy caprioles, si ce n’est en tournant ou finissant, et tout plain
d’autres petites actions ennemies du vray air qu’on y doit observer.
([I] do not want them to include in their compositions any steps that
evoke a clown, like fleurets, frisoteries, or shaking of feet, pirouettes (I mean
those with violent or unnatural turns), caprioles, not even half-caprioles,
except if turning or finishing, and many other little movements that go
against the gracious appearance one must respect).8
21 Ballet de cour

Thanks to the involvement of real members of the court, the ballet acted
as a mirror set between spectators and dancers in which court society staged
itself, its particular worries and political projections. The role assigned to
each performer, as indeed the role assigned to, and occupied by, each specta-
tor, embodied a precise meaning in the order of society. Allegory transposed
that mirroring into the relationship between the individual and the universe
in which the social order became symbolised as cosmic reality and at the
same time the universal stood for the social. In the game of symmetry,
which kept the balance between the realisation of the work on stage and
the realisation of the pattern of court society, the ballet de cour became a
way to consolidate a consensus in society, which reflected it and on which
it reflected.
The ordonnateur des plaisirs royales furnished the theme and coordinated
the diverse trades and crafts needed in the production of a ballet de cour.
In some cases the ordonnateur had at his side or occupied himself the posi-
tion called by de Pure Poëte, who incorporated the functions of “autheur,
inventeur, dessinateur, entrepreneur”,9 a notion of the poet, which seems to
return to the original Greek meaning of the word poiein (to make).
Court artisans carried out the staging of the scenes, the operation of the
theatrical machinery and the realisation of the accessories, costumes and
masks. Many of them were Italians, who had come to Paris either in the
retinue of Catherine de’ Medici or were summoned on commission by her.
They contributed to the transmission of technical skills in the field of theatre
elaborated for decades in the Italian courts, especially those in Florence. The
scenic dressing of the stage of the ballet de cour in their hands aimed at pure
spectacle, made up of the ingredients of surprise, of the marvellous and
exotic. For example, in the Ballet de la délivrance de Renaud of 1617, the
Italian stage designer Tommaso Francini provided four changes of scene,
renewing the perspective in the Italian manner while the audience watched.
When the economic situation permitted it, the sovereign would place huge
sums of money at the disposal of his stage designers in order to meet such
challenges as would exalt his magnificent lavishness. In periods of austerity,
sets and costumes would be reused or readapted for the most disparate
thematic contexts and varied narratives.10
The costumes and accessories formed a combination that identified the
characters, but they became overloaded symbolic indicators that reached the
level of the absurd and achieved an “overkill” of meaning, so to speak. One
example of such overbearing symbols was the female giant who appeared in
the entrée of the music fairie (from the French féerie) in the Ballet des fées de
la forêt de Saint-Germain of 1625, portrayed in a drawing by Daniel Rabel:11
she carries a director’s baton and a musical score; she wears as hairdo a fiddle;
lutes and theorbos are attached to her skirt, a triangle serves as an earring and
22 Marina Nordera

Figure 2 Guillemine la Quinteuse, Fée de la musique. Ballet des fées de la forêt de Saint-Germain
(1625), drawing by Daniel Rabel, Musée du Louvre, Inv. 32 604.

small cymbals as ornaments of her corset (see Fig. 2). Another Rabel drawing
portrays hermaphrodites in the Ballet de la douairière de Billebahaut of 1626.
Their bodies, seen from the front, are divided vertically in two parts. The
right-hand side from the spectator’s point of view is feminine. The bosom
is uncovered, the hair is long, the figure wears a skirt and holds a spindle
in her hand. The left side is masculine. The leg is exposed and muscular,
he has a moustache, a plumed hat and a sword in hand.12 (see Fig. 3). All
the characters – grotesque or not – always wear masks which accentuate the
characterisation of the individual or render a group entirely homogeneous.
The masks according to de Pure serve to beautify and ornament the character
but above all to render it recognisable as such for the public. The mask must
eradicate the personal identity of the dancer in such a way that the spectator
sees in it only the character which he or she incarnates.
In spite of the collaboration of the various artistic authorities, the prin-
cipal role in the production of a ballet de cour remains that of the poet,
which confirms the dominance of the text in the hierarchy of the arts. The
most prolific poet who produced ballets de cour and whose career covered
thirty years (1651–81) was Isaac de Benserade, a member of the Académie
Française from 1674, holder of pensions from Richelieu and Mazarin and
author of successful libretti in the second half of the seventeenth century.
Benserade was admired for his ability to let the real figures of the courtiers
of the epoch and actual episodes from their world shine through the actions
of allegorical or mythological characters who appeared on stage.
23 Ballet de cour

Figure 3 Androgynes, Ballet de la douairière de Billebahaut (1626), drawing by Daniel Rabel, Musée
du Louvre, Inv. 32 692.

As to the form, structure, themes and modes of representation, the ballet


de cour continued the tradition of the Italian intermedii and the French
masquerades,13 divertissements that were more or less extemporaneous and
which delighted the private and public social life of the court beginning at
the end of the sixteenth century. It continued through the entire seventeenth
century and into the eighteenth.14 The elements of that inheritance were
reorganised in order to adhere to a coherent theoretical system put forward
by the Académie de Poésie et Musique, founded in 1570 by Jean Antoine de
Baı̈f. He was a poet and member of the Pléiade, a group of poets who had
come together for the common purpose to give new life to French poetic art.
The Académie intended to restore the unity and synthesis of the arts of the
ancient Greek drama. Poetry, music, dance and song were to be subordinate
to the same aesthetic principles and the same rhythmic and melodic laws.
The model created for dance was that of choral movement based at the same
time on the spatial geometry, which organised the evolution of the groups
on the stage and on the expressiveness of gestures capable of rendering a
wide range of meanings not expressible in words. In spite of the attention
dedicated to the stage arts, for the Academicians the poetic text remained
the principal motor of the action and the main ingredient of the aesthetic
synthesis.
This first attempt to provide norms for the ballet failed to impose a system
of rigid rules based on the unity of time, place and action or subdivisions
into acts derived from Aristotle and frequently imposed on the other forms
of theatrical performance. Its flexible forms lent themselves easily to various
24 Marina Nordera

types of experiment.15 Seventeenth-century theorists – de Pure, Ménestrier


and Saint-Hubert16 – limited themselves to description of the characteristics
of the ballet without codifying precise rules. Indeed Michel de Pure insisted
on emphasising the advantages brought to the ballet by the very absence of
codes and rules:

Soit que jusqu’icy les Loix du Balet n’ayent pas esté publiées, ou que le Ciel
et sa bonne fortune l’ayent préservé des chicaneuses et ridicules inquiétudes
des Maistres-ez-arts, il n’est tenu que de plaire aux yeux, de leur fournir des
objets agréables et dont l’apparence et le dehors impriment dans l’esprit des
fortes et de belles images.

(Whether it is that up until now the Laws of Ballet have not been published,
or that the benevolence of Heaven has preserved it from the ridiculous and
petty concerns of the masters of arts, Ballet must only please the eyes, it
must provide them with pleasant objects whose appearance and forms
imprint strong and beautiful images in the mind.)17

De Pure was a court intellectual and “aumonier du roi” (king’s chaplain).


He was an assiduous spectator of performances at court and was writing
at a time when the ballet de cour was at its height. The long chapter which
he dedicated to this type of performance contains descriptions and critical
observations of current practices and another attempt to systematise the
art. After a definition of the word “ballet”, he examines the elements which
come together in its creation and realisation: subject, title, subdivisions in
parts and entrées, music, texts in prose and poetry, instruments, dancers,
costumes, masks and above all le pas de balet, the true, proper choreography
of the piece. Dance therefore joins together with all the other arts in the
constitution of the ballet de cour and, even though it gives its name to
the entire composite performance, it is not dominant: song, declamation,
surprises in staging and scenery join the choreography on equal terms. The
choreutic vocabulary used is the same as that in social dances, even if the
technical level varies according to the competence of the interpreters and
the type of dance. In the ballet de cour, in fact, diverse forms of choreography
coexisted and alternated: figured dance, dances drawn from the repertory
of society, pantomime, solos of great virtuosity, masked processions and
dressing-up.
Some social dances were integrated into the ballet de cour because of their
expressive or functional value, for example, the pavane was used for royal
entries or divine interventions, or some branles offered the basic structures
for group dances. An example is given in one of the entrées of the ballet La
Douairière de Billebahaut:

Les Donzelles qui la suivent luy marchent sur les tallons cherchant et ne
trouvant pas la cadence des Bransles de Bocan; et leurs pas plutost de balle
25 Ballet de cour

que de ballets tesmoignent qu’elles ont grand tort de venir estudier dans la
Salle du Louvre pour aller danser ailleurs.

(The women who follow him step on their heels attempting in vain to find
the rhythm of Bocan’s branles; and their steps, more proper to a ball than a
ballet, testify to the fact that they were very wrong to come and study in the
Salle du Louvre in order to dance elsewhere.)18

This passage would seem to indicate the ironic deformation of a pre-existing


branle that had been composed by Bocan, a famous ballet master of the
epoch. In the solos, great importance was given to the symbolic significance
of the character, produced by the richness of the costume and the social rank
of the interpreter, which concentrated the attention on him or her, as in the
case of ballet de cour, in which the sovereign himself took part. In the Ballet
de la nuit, presented in the Hall of the Petit Bourbon on 23 February 1653,
with music by the young Jean-Baptiste Lully, Louis himself played the rising
sun. In the following year Le nozze di Peleo e di Teti19 became a political
event of fundamental importance after the Fronde crisis. Louis XIV, the
son of Louis XIII and Anne of Austria, was crowned in 1654 but prevented
by rebellion, the so-called two frondes, one the fronde of the nobles, the
other fronde of the parliamentarians, from assuming the full powers of the
crown. The young Louis XIV danced the role of Apollo and appeared as
the victor, pacifier and central figure of re-established, political harmony.
The use of expressive dance for such ends was documented by theoretical
texts and the iconography of the time. When gods or allegorical figures were
set in the dance, such elevated figures drew their gestures and moves from
the declamatory arts and rhetoric, which young nobles had learned at the
Jesuit colleges. In the case of grotesque or burlesque characters, the skills
of acrobats and of comedians furnished a vast range of gags or acrobatic
jests. Finally, in addition to every sort of armed combat, certain types of
“professional” moves and gestures were choreographed as in the case of
gardeners, carpenters and wood-cutters who utilised the tools of their trades
in cadence and in fantastic formations.
In the so called “figured dance”, sometimes called “horizontal”, dancers
took positions to create geometric designs, letters of the alphabet or symbols,
all legible only by the spectator who saw them from above and from a central
position relative to the stage. The king, his family and the dignitaries closest
to them were among the few privileged spectators who would have been in a
position to decipher the message correctly. No source reveals to us the rules
of composition for the “figured dance”. The descriptive elements which have
survived to our day come from libretti distributed before the show, from
celebratory texts edited afterwards and from contemporary descriptions.
Some of these libretti contain the geometrical figures which the choreo-
graphic compositions intended to “design” (possibly because on stage such
26 Marina Nordera

Figure 4 Faun from Lully’s Le Triomphe de l’amour, 1681.

figures were not really legible). The figures that were composed of knights
in the final ballet of the Ballet de Monseigneur le Duc de Vandosme (1610)
represented letters drawn from an ancient druidic alphabet, which is written
out and illustrated in the libretto. Each figure represents a concept: “pouvoir
supreme, verité connue, peine agréable, ambitieux désir” (“supreme power,
known truth, sweet sorrow, ambitious desire”) and so on. The succession of
the figures on stage constructs the phrasing of an abstract, verbal discourse,
transmitted in code to those who were in a position to decipher it (and
openly revealed to readers of the libretto!). The degree of comprehension of
the message depended on the position which the spectator occupied in the
hierarchical organisation of space. The figures in the dance could be static
or in movement. The transition from one figure to another occurred in a
kind of internal fluctuation of the dance or in passages apparently neutral.
Mark Franko indicates that in these transitions of spaces an independence
from the text of the body emerges, which “loses its human resonance when
it becomes a marker of geometrical position”,20 and hence achieves its own
27 Ballet de cour

political autonomy. The body acquires in this way a strong power to evoke
abstract concepts, cosmic and astral movements, which overcome its cor-
porality. A dialectical game begins as a consequence between the autonomy
of the dancing body and the text, in which the body, itself in process of
becoming text, assumes its own political significance that is independent of
the text and sometimes in open contradiction to it.21 Around 1640, under
the influence of Italian staging practices, the stage became elevated and it
was no longer possible to perceive such figures, which therefore disappeared
progressively from the compositions.
De Pure noted in 1660 with a certain irony that the “figured dance” had
established itself to show certain courtiers who could not do more than
walk in time to the music in a good light and called that practice a techni-
cal short cut, which showed clearly that the art of ballet had degenerated.
For de Pure, the practice of dancing was not unbecoming to a “homme de
qualité” who therefore should apply himself with greater commitment. By
acquiring greater technical and expressive competence the amateur inter-
preters of ballet de cour would be able to execute the complex compositions
proposed by dance masters.22 A distinction seemed to emerge thus between
“simple dance” and ballet. This was neatly expressed by Claude-François
Ménestrier, a Jesuit historian, who published two works on ancient and
modern theatrical ideas:

Et cela nous apprend la différance qu’il y a entre les Ballets, et que la simple
danse est un mouvement qui n’exprime rien, et observe seulement une juste
cadence avec le son des instruments par des pas et des passages simples ou
figurés, au lieu que le Ballet exprime selon Aristotle les actions des hommes,
leurs mœurs, et leurs passions.

(And that teaches us the difference that there is between ballets and simple
dance: that simple dance is a movement that expresses nothing and observes
only an exact rhythm with the sounds of instruments, by steps and simple
passages or step sequences, whereas the ballet according to Aristotle
expresses the actions of human beings, their customs and their passions.)23

Though there was a common vocabulary between social dance and bal-
let, the mode of executing both varied. The production of a ballet de cour
required the presence of two professional figures, whom we would nowadays
describe with the anachronistic terms director and choreographer. These
observations make us grasp how the quantity and quality of dance con-
tained in a ballet de cour depended on the type of interpreters who staged it,
on the type of dance employed and on the notion of dance which it implied.
Unfortunately, the scarcity of specific sources does not permit us to define
exactly the nature and range of these differences.
28 Marina Nordera

The ballet de cour, and in particular the Balet comique de la Royne, a


piece which has been described as the ancestor of the genre by current his-
toriography, is one of the most extensively studied episodes in the history
of dance.24 The reasons for such particular interest are manifold: above all,
the qualitative and quantitative wealth of sources which document it has
allowed the reconstruction of various aspects and of the salient episodes
in its development. In addition, its evolution coincides with the political
apogee of the French court, an object of many studies because of the impor-
tant place which the Sun King and his entourage occupy in French history,
also serving as a model to lesser courts in Germany and other European
countries. Finally, the ballet de cour represents a key moment in the histo-
riographical project, which claims that the noble origins of classical ballet
legitimate it as the highest form of dance. This legitimisation operates at the
same time in aesthetic-literary and political terms. The “mythical” origins
are made to go back, in fact, to the century in which the literary codi-
fication of theatrical genres took place and when the will of enlightened
and powerful sovereigns used this form of performance not only as enter-
tainment and as an end in itself but also for their political and personal
self-promotion.
The ballet de cour enjoyed great success beyond the confines of the
Parisian court, in particular in the form of the English masque and in Pied-
mont at the Savoy court, where it developed a particular form thanks to the
activity of cavaliere Filippo d’Aglié, the younger son of a distinguished fam-
ily, who became secret councillor to Christine, daughter of the French King
Henry IV and Maria de’ Medici, when she married the Duke of Piedmont
Vittorio Amedeo I. Our knowledge of the ducal repertory is derived from
a rich collection of sources. There are reports of historians and chroniclers,
libretti, engravings and paintings but, above all, several codices conserved
in the Royal Library and the National Library in Turin. These volumes were
written in the hand of the ducal secretary Tommaso Borgonio who, in addi-
tion to the texts, reproduced stage scenes and costumes in priceless images.
Altogether, the collection represents an incomparable corpus of documen-
tation by comparison to what remains on the Parisian ballet de cour.
As in Paris, the ballets were interpreted by court nobility but the grotesque
characters required the employment of professional dancers and actors, fre-
quently people from the commedia dell’arte, who are, however, not named
in the documents. D’Aglié took on the composition of the ballets as far
as the themes, the development, the characters, the stage design and the
music were concerned. The structure of the entries, the thematic choices
and the procedures for conveying meaning were similar to those in Paris
but with a preference for demonstrative stage effects and surprises in cos-
tumes and accessories. Usually, after an initial entry, which presented the
29 Ballet de cour

theme, the remaining entries followed in a crescendo of ever more spec-


tacular effects. For example, Il tabacco (1650), whose theme was suggested
by the recent legalisation of the tobacco leaf, displayed the natives of the
island of Tobago, who praised the virtues of the tobacco. Successive scenes
presented smokers, ancient and modern, from all the nations. Il Gridelino of
1653 celebrated the gris-de-lin (literally grey linen) in a series of entrées; this
grey was the favourite colour of Madame Christine and symbol of “constant
and persevering love”. After the God of Love has chosen gris-de-lin among
all the colours presented to him by Iris, gardeners alternate in presenting
flowers and plants of that colour; weavers then present this colour in cloth
and so on through a series of entrées on the set theme – grey linen. Tak-
ing account of what happened outside the French court and following the
parallel trail to Savoy and England helps to reposition correctly the bal-
let de cour in the historiographical panorama of Europe in the eighteenth
century.
Balet comique de la Royne or Allegorie de Circé, was staged at the court
of Henry III, Valois-Angoulême King of France, for the marriage of his
favourite Monsieur Anne d’Arcques, Duke of Joyeuse, to the sister of the
queen, Mademoiselle de Vaudemont, Marguerite de Lorraine. It was per-
formed in Paris on 15 October 1581. We are accustomed to consider it as the
very first ballet de cour in history, which it was not: we have much evidence
that before this work other performances of the same type had been staged
in France and Italy, but they were not called “ballet”.25 The historic relevance
given to the Balet comique de la Royne is closely linked to the importance
of the event which it celebrated and hence the fact that it motivated the
production of a rich documentation, which has survived to our time: most
importantly the libretto, which was written by Balthasar de Beaujoyeux26
and published roughly four months after the performance; secondly, the
music, then the printed illustrations, which portray the staging, costumes,
the fixed and movable scenic elements, the plan of the hall in which the
production was staged and finally official and unofficial chronicles of the
occasion. The libretto through which the Balet comique de la Royne is trans-
mitted to us is a unique example of its kind. The great majority of ballets de
cour lack such documentation. The rich documentation of the promotional
apparatus of the Balet comique de la Royne forces itself on the attention of
the historian who has to keep alert to make a correct exegesis of the material,
peeling off the thick layers of rhetoric and bombast. The libretto provides
exaggerated data, which makes it necessary to check every fact. As an example
of this exaggeration, we are told that the number of spectators was between
9,000 and 10,000, while we know in fact that the hall could only accommo-
date 500.27 We have therefore to read the sources carefully, and the weaving
in of other sources can give more accurate dimensions of this type of show.
30 Marina Nordera

For example, analysis of the contribution of the various arts in the Balet
comique refutes the declaration of Beaujoyeux that he had given the central
role to dance: “I’ay toutefois donné le premier tiltre & honneur à la dance, &
le second à la substance, que i’ay inscrite Comique” (I have always given the
first role and honour to dance and the second to substance, which I inscribed
in Comique). Beaujoyeux actually used the word balet for only two chore-
ographic scenes that the piece contained. He described them as “n’estant
à la verité que des meslanges geometriques de plusieurs personnes dansan
ensemble sous une diverse harmonie de plusieurs instruments” (being, in
truth, no more than the geometrical groupings of people dancing together,
accompanied by the varied harmony of several instruments).28 Sparti has
remarked justly that “this oversimplified definition of ‘balet’ hardly sounds
like the words of a choreographer, even one writing for a ‘non-specialised’
audience. One would think that Beaujoyeux as choreographer would have
wanted to impress, explicitly, the spectators-readers with the importance of
his dances by alerting them to their complexity, significance and worth.”29
It is legitimate here to ask why that theatrical production should have been
called ballet in the first place. To pose such a question opens more general
issues about the corpus of work which we habitually call by that name.
Indeed, the fact that successive studies have from time to time assigned
diverse chronological limits to the ballet de cour shows how definitions of
the genre can fluctuate. Lacroix presented the Ballets et mascarades de cour
that were staged between 1581 and 1652; Prunières, who intended to illumi-
nate the Italian origins of the genre, studied the phase preceding the “clas-
sical” and “French” epoch of Benserade and Lully (see Fig. 4); McGowan’s
treatment begins with 1581, which she considers as the date of the “first
effort that constructs a ballet that realises all the aspirations of the theoreti-
cians” and the period for her ends in 1643, which corresponds with the death
of Richelieu and Louis XIII and marks the introduction of gusto italiano by
Mazarin. Mark Franko broadens the thematic and chronological span of his
research, which extends from 1573 (the first known work of Beaujoyeux,
the Ballet des ambassadeurs polonnais) to 1670 and the ultimate comédie
ballet, Le Malade imaginaire. In this way, Franko overthrows the conven-
tional definition of ballet de cour, which is both improper and invented,
and proposes a much wider idea of ballet, which would comprehend the
proper, true ballet de cour, the ballet burlesque and the comédie ballet.30
The historiographical approach proposed by Mark Franko abandons the
assumption of a long-standing critical tradition that presupposes a continu-
ity between ballet de cour and classical ballet and argues that the seventeenth-
century composite form had more in common with the performance art
of the twentieth century than with the evolution of the ballerina in the
tutu.31
31 Ballet de cour

This brief historiographical excursion shows how the ballet de cour, by


its multiform heterogeneity, by its capacity to renew itself and by the strate-
gies adopted to plant itself in its time, naturally lends itself to successive
exploration and leads us to pose questions of much greater scope about the
relationship between sources, history and historiographical work.

Translation Jonathan Steinberg


Translation from Old French Kristin Stromberg Childers
and Roger Chartier
3 English masques
ba r ba r a r ave l h o f e r

English masques were allegorical entertainments with dance and music, cos-
tumes, songs and speeches, and festive scenery. As a protean phenomenon,
which flourished in the sixteenth and seventeenth centuries, masques might
assume the character of a low-key countryside event, a civic festivity, a high-
profile state occasion, a university romp or a jollification at the Inns of Court
in London. Plays of the early modern period often treated their audiences
with a masque en miniature: silent dancing highlighted pivotal moments in
the action and added to the suspense.1 Shakespeare’s The Tempest includes a
wedding masque whose “graceful dance” of nymphs and reapers “in coun-
try footing”2 is dangerously interrupted by invaders. In Restoration operas,
masques showcased magical characters and dazzled spectators with spec-
tacular ballet and scene transformations.3 Masques fused English traditions
with foreign performance practices. With their abundance of danced pan-
tomime, they represent an important precursor to John Weaver’s balletic
drama in the eighteenth century. This chapter will provide a brief history
of this multifarious genre, and explore in greater depth the impact of con-
tinental balletic forms on masques performed in London during the early
seventeenth century.
Masque-like mummeries had been popular since at least the early Tudor
period. Henry VIII is said to have introduced disguisings in Italian style to
the English court. In 1512, the king and his courtiers celebrated epiphany

disguised, after the maner of Italie, called a maske, a thyng not seen afore in
Englande, thei were appareled in garmentes long and brode, wrought all
with gold, with visers and cappes of gold, & after the banket doen, these
Makers came in, with sixe gentlemen disguised in silke bearyng staffe
torches, and desired the ladies to daunce, some were content, and some that
knewe the fashion of it refused, because it was not a thyng commonly seen.
And after thei daunced and commoned together, as the fashion of the
Maskes is, thei toke their leaue and departed, and so did the Quene, and all
the ladies.4

Audience participation remained a crucial element in “masks”. As the


Milanese ambassador reported about nightly revels in 1514, the king was
dancing “in his shirt and without shoes” with the ladies, and leaping “like
[32] a stag”.5
33 English masques

Elizabeth I inherited the passion for Italian entertainment culture. In her


youth, the French ambassador De Maisse recounted that the queen “danced
very well, and composed measures and music, and had played them her-
self and danced them . . . without doubt she is a mistress of the art, having
learnt in the Italian manner to dance high . . . they called her ‘the Floren-
tine’.”6 De Maisse not only alluded to Elizabeth’s well-known enthusiasm for
Italian dancing (she also employed at least one Italian professional, Jasper
Gaffoyne) – he even suggested that she composed music and choreogra-
phies (one meaning of “measure” at the time).7 Entertainments in honour
of Elizabeth sought to please her with artistic feats by foreign professionals.
At Kenilworth in 1575, an eyewitness was bowled over by an Italian tumbler
who showed off his bravura technique:

feats of agilitee, in goings, turnings, tumblings, castings, hops, iumps, leaps,


skips, springs, gambauds, soomersauts, caprettyez & flyghts: forward,
backward, sydewyze, doownward, vpward, and with such wyndyngs gyrings
& circumflexions: al so lightly and wyth such eazyness, az by me in feaw
woords it iz not expressibl [sic] by pen or speech.8

The tumbler’s athletic prowess recalls the manifold leaps and capers
described in Italian dance manuals of the period (see Chapter 1). Doubtless it
pleased a queen who herself took pleasure in “dancing high”. Skilful jumping
seems to have remained a benchmark for fine dancers in the years following
Elizabeth’s reign. In a reception in 1604, Prince Henry Stuart danced before
a Spanish visitor. As the guest observed, he performed “with great ease, yet
dignified restraint” and mastered “some capriols”.9 It has been suggested that
Henry acquired these skills from perusing Cesare Negri’s Nuove inventioni
di balli (1604), and that both this work and Fabritio Caroso’s famous Nobiltà
di dame (1600) were available at court during Henry’s lifetime – a hypothesis
worth considering.10 Given the eminence of Italian festival culture at the
time, dancing styles as practised in Florence, Rome or Milan must have had
an impact on the early Jacobean court. Prince Henry repeatedly asked for
Florentine festival books.11
After the accession of the Stuarts in 1603, entertainment culture devel-
oped a new professional dimension. Under James I and Anna of Denmark,
and Charles I and Henrietta Maria, lavish masques were performed at White-
hall between 1604 and 1640. These so-called “court masques” celebrated the
monarchy before a specially invited audience of diplomats, courtiers and
citizens.12 Given their excellent documentation (stage and costume designs,
texts and bills survive), court masques prove particularly attractive for dance
and performance research.
The best poets of the day composed the songs and speeches for masques,
among them Samuel Daniel, George Chapman, Thomas Carew and William
34 Barbara Ravelhofer

Davenant. The artistic partnership of the poet Ben Jonson and the architect
Inigo Jones – who was responsible for stage and costume designs – led to
such legendary productions as The Masque of Blackness (1605), The Masque
of Queens (1609) and Oberon (1611).
While masques stunned the senses by the combined impact of words,
music and visual effects, it is fair to say that no other form of theatre in
early modern England placed greater emphasis on dancing. As the poet
Thomas Middleton remarked, masques consisted in “one houres words, the
rest in Songs & Dances”.13 Much of the action evolved within a specially
constructed dancing area which separated the stage from the auditorium;
court records mention a wooden floor covered with green fabric to muffle
the noise.14 There, participants exerted themselves for long hours. As the
courtier Thomas Roby reported about one of Henrietta Maria’s dancing
nights, “ye Mask. . . on Sunday last begann at Tw[e]lve at night and was
not done till almost fyve”.15 Not surprisingly, balletic expertise was in great
demand. Masques thus encouraged the rise of the dance professional in
early modern England.
Professional dancers were highly mobile and contributed to the cul-
tural interchange between different countries. English artists ventured to
the continent. In 1608, William Pedel, a pantomimist skilled in “dauncing
& vaulting”, travelled to Leyden where he exhibited “beautiful and chaste per-
formances with his body, without using any words”.16 Caleb Hasset served
the Landgrave of Hesse-Kassel in the 1630s, and George Bentley worked
in Dresden in the 1650s.17 Continental artists and dancers settled in early
Stuart London. The Tuscan designer Constantino de’ Servi provided set-
tings for Thomas Campion’s The Somerset Masque (1613). The renowned
violinist and choreographer Jacques Cordier (also known as Bocan) served
both the French and English courts, where he was involved in Campion’s
Lords’ Masque (1613) and probably Carew’s Coelum Britannicum (1634) (see
Figs. 5 and 6).
Structurally, court masques often fell into three parts with distinct kinds
of dancing. They started with a mimic display of the forces of chaos and mis-
rule; then royal and aristocratic performers restored order by their formal
ballet; and finally audience and performers joined for dances and refresh-
ments. The first part (called “antemasque” by Daniel and Middleton, “anti-
masque” by Jonson, and also known as “antic masque” or “entry/entrée”)
abounded in grotesque, amusing roles (such as nymphs, satyrs or drunk-
ards), which were commonly taken by professional actors and dancers either
in group or solo performances. Although the movements must often have
seemed spontaneous or erratic to the onlookers, they were tightly chore-
ographed, which can be deduced from the carefully cued changes of rhythm
in surviving music and the references to choreographers in masque texts.
35 English masques

Figure 5 Cover page, Coelum Britannicum 1634.

Thus in The Masque of Queens (1609), male dancers erupted onto the stage
in the guise of witches:

with a strange and sudden music, they fell into a magical dance, full of
preposterous change and gesticulation, but most applying to their property
[i.e. habits], who, at their meetings do all things contrary to the custom of
men, dancing back to back, hip to hip, their hands joined, and making their
circles backward to the left hand, with strange fantastic motions of their
heads and bodies. All which were excellently imitated by the maker of the
dance, Master Jerome Herne, whose right it is here to be named.18
36 Barbara Ravelhofer

Figure 6 List of masquers, Coelum Britannicum 1634.

James I loved antimasques and often asked for an encore. In Francis


Beaumont’s The Masque of the Inner Temple and Gray’s Inn (1613), dancing
statues, servants, baboons and other droll figures won “perpetual laughter
and applause”:

the dancers, or rather actors, expressed every one their part so naturally and
aptly, as when a man’s eye was caught with the one, and then passed on to
the other, he could not satisfy himself which did best. It pleased his Majesty
to call for it again at the end, as he did likewise for the first anti-masque, but
one of the Statues by that time was undressed.19
37 English masques

With the years, these “antic” performances increased in number, partly


thanks to popular success, partly a consequence of French tastes at Whitehall.
French ballet de cour favoured serial entrées and allowed courtiers to perform
as “low” characters. Most English court masques maintained a distinction
between “professional” and “aristocratic” dancers in the casting of roles.20
Yet events on a smaller scale, especially those performed in private residences
in the countryside, were less rigid in their divisions of rank. Thus, the
francophile Duke of Buckingham danced as a pickpocketing gipsy in 1621.21
With the arrival of Henrietta Maria, a French princess, at the English court,
continental customs became even more fashionable. Henrietta Maria was
an accomplished dancer and singer who had performed in ballets since
her childhood. French kings mingled with the bourgeoisie in public ballet
events22 – unthinkable for Charles I, a monarch obsessed with protocol.
Henrietta Maria, however, felt at ease in venues outside the court. Invited
to a performance of The Triumph of Peace (1634) in the Merchant Taylors’
Hall, the queen was carried away when she danced before the citizens: “The
Queene dancing at the Lord Maiors, strained her foote & was like to haue
taken much hurt.”23 The anecdote is emblematic for the increasing strain
on both royal effort and civic goodwill. Salmacida Spolia (1640) – the last
court masque staged before the Civil War – anxiously exhorted spectators
to preserve peace under a benevolent Stuart monarchy. The event boasted
an inflated number of twenty antimasques, which indicates how desperate
the king and queen were by then to entertain their audience.
The theatrical part of a masquing night culminated in the main masque
(or “masque proper”). Here an ensemble of splendid aristocratic performers
engaged in silent dancing. As in the antimasque, the movements were spe-
cially choreographed; but now they emphasised formality in a grand ballet.
Masquers traced geometric patterns or alphabetic letters, which was com-
mon practice in Italian theatre repertoire of the period and the grand ballet
in French ballet de cour. Typically, The Masque of Beauty (1608) saw nymphs
performing “squares and rounds” and “a most curious dance full of excellent
device and change”, which ended “in the figure of a diamond”.24 Geometric
circles and squares projected to all sides of an auditorium.25 In masques they
would have been equally visible from any of the seats which surrounded the
dancing space on three sides. Choreographies with such figures were clearly
intended to please as many viewers as possible and thus challenge the cur-
rent view that formal dances were intended for the ruler’s eye only. Yet
the association of early Stuart courtly repertoire with absolutism cannot be
dismissed. In a politically motivated inversion of masquing conventions,
Milton’s A Masque Presented at Ludlow Castle [Comus] (1634) elevated
country dances – usually the domain of the antimasque or the revels –
to the “masque proper” section. The upright simplicity of these dances
38 Barbara Ravelhofer

(“without duck or nod”) contrasted with the “ill-managed merriment” of


Comus and his “glistering”, beast-like courtiers, caricatures of Cavalier rev-
ellers. By privileging local festive traditions, Milton offered a poignant cri-
tique of what he perceived as a centralised court culture of excess under
Charles and Henrietta Maria.26
While antimasque characters were always impersonated by men, the
main masque was open to women: Queen Anne and her ladies gave dar-
ing (in the view of some English courtiers, too daring) performances as
black daughters of the River Niger or Amazonian queens. Henrietta Maria
pioneered speaking and singing roles for women, as they had long been prac-
tised on the continent. Her court masque Tempe Restored (1632) featured
the first female professional singers in English theatre history.
In the final part of a masque, the “revels”, select members of the audi-
ence joined the masquers for yet more dancing.27 The revels drew upon
a pre-existing pool of social dances. Contemporary accounts mention, for
example, country dances, branles, or “measures”.28 The latter may, in this
case, allude to a repertoire taught in English dancing schools since Eliza-
bethan times.29 The choice of dances and participants would have been pre-
established; even so, revels or revellers enjoyed themselves. Queen Anne’s
masquing nights reveal a sense of merry abandon:

for galliards and corantoes they [Anne and her ladies] went by discretion,
and the young prince [probably Henry] was tossed from hand to hand like a
tennis ball. The Lady Bedford and Lady Susan [de Vere] took out the two
ambassadors and they bestirred themselves very lively, especially the
Spaniard, for his Spanish galliard shewed himself a lusty old reveller. The
goddesses they danced with did their parts, and the rest were nothing
behindhand when it came to their turns; but of all for good grace and good
footmanship Pallas [Anne] bore the bell away.30

Revels repertoire was probably easier to perform than the preceding


theatrical parts and thus facilitated audience participation, as shown by the
Spanish guest’s vigorous contribution. Dancing immersed a heterogeneous
audience in the total experience of a masquing night. It rendered the event
more inclusive, creating a “corporate and festive identity”.31
Unfortunately all choreographies for the great court masques are lost,
and their nature must be inferred from circumstantial evidence, such as
eyewitness accounts of masque texts. Two dance sources are directly con-
nected with masque performers: François de Lauze’s Apologie de la danse
(1623) and Barthélemy de Montagut’s Louange de la danse (before 1623),
both dedicated to the Duke of Buckingham, a famous dancer and spon-
sor of masques. But they describe social, not theatrical dance repertoire.
Their new emphasis on gliding steps and pliés suggests an affinity to French
39 English masques

danse noble.32 There is no evidence that de Lauze ever worked in England;


in later years he danced in French ballet de cour. Montagut, on the other
hand, taught Buckingham and orchestrated entertainments for him; he later
became Charles’s dance instructor, performed in at least two court masques,
and was a well-remunerated member of Henrietta Maria’s household. In
seventeenth-century England, reading and learning by doing offered par-
allel access routes to dance repertoire. The lawyer Justinian Pagitt noted,
“write the marks for the stepps in every daunce under the notes of the tune,
as the words are in songs”.33 A few manuscripts containing social dances
may indicate the dissemination of repertoire by manuscript circulation; a
collection of eleven country dances entitled Chorea, for instance, contains
errors as they typically occur in the process of copying another document.34
Persuasive evidence for learning dances from a written source derives from
John Playford’s country dance collection The English Dancing Master (1651,
subsequently expanded, eighteen editions until 1728). Yet, as far as early Stu-
art masques are concerned, the absence of any choreographies is striking,
especially given the wealth of surviving masque music.
One exceptional source shows how a masque dance could have been
fixed to paper. In about 1650, the merchant clerk Robert Bargrave penned
an entertainment for the wedding of the English ambassador’s daughter
at Istanbul. His diary records the masque poetry next to music scores
and verbalised choreographies. This unassuming document may well con-
tain the first surviving theatre choreographies in English drama. Bargrave
uses one-letter abbreviations for performers, and very basic floor patterns
such as squares and triangles. His notes represent memoirs rather than an
instruction for dancing – in the case of one dance, Bargrave writes “figure
forgotten”.35
The choreographies consist of four seasonal dances (a popular topic in
English masques) and an antic dance. The seasonal dances are relatively
simple; their terminology is reminiscent of that used for contemporary
country dances, as in this typical passage:

Chace one another forward, then backward & so fall thus:


a[utumn] w[inter]
s[ummer?] s[pring?]
2:d time :So: & Sp: face W :&: A: then Au: & So: separat from W: & Sp: then
back & fall into a ranke :2.d part Hay till each comes into his Place: 2.d time
salute each other, & the Company, & go off.-36

The final “antic” dance, a pantomime, involves four “men”, two “boys” and
one “woman” (probably played by a man, given the graphic nature of the
performance). Accompanied by a “ranting” tune, the performers tumble
40 Barbara Ravelhofer

over each other, fight a mock battle, and beat the “woman”. Then the action
proceeds to a mock marriage, and finally the “woman” gives birth, onstage,
to two boys:

First part; Men bring in W. great Bellied, hiding two naked boyes under her
coats :2:d time :B. doune on all fower, A kneeling behind him (to resemble a
Chaire) C. & D. make her sitt downe :2.d part She sets her selfe in travell, &
they officiat :2.d time, the two boyes runn out from under her: – then the :4:
daunce round the W. who nurses the boyes at each breast, & so runn all In:-
(p. 98)

Bargrave’s simple devices are the work of an enthusiastic amateur, not a


professional but they show an imaginative sense of humour. It is a pity that
they were never performed, as the wedding was cancelled.
Back in England, James Shirley’s Cupid and Death (1653) pessimisti-
cally declared about the gentry that “their dancing days” were “done”.37 Yet
masques continued to be performed: Henry Purcell’s Dido and Aeneas (two
versions, pub. 1698, 1700) is variously called “masque” and “opera”, and
John Dryden composed a Secular Masque as late as 1700. Charles II briefly
considered reviving sumptuous court masques as they had been practised
before the Civil War. John Crowne’s Calisto (1675), the magnificent result of
such deliberations, borrowed its five-act structure from the French comédie-
ballet. Calisto was an Anglo-French choreographic co-production. Josias
Priest, later responsible for Dido and Aeneas, collaborated with eight French
professionals, including Mr St-André, a member of the Académie Royale de
Danse at Paris. Preparations extended over six months, involving a cast of
ninety-one (with Princess Mary in the leading role).38 In contrast to Caro-
line spectacle, the spoken parts were taken by courtiers, while most dances
were performed by professionals.39 Had theatrical repertoire become too
difficult for courtiers? It is tempting to think of Michel de Pure, a critic
writing about Paris theatre in the 1660s. For de Pure, the “stupidity of most
great lords and persons of quality” affected the choreographies in opera
productions. These noble amateurs were

not capable of anything, and constrain in this manner the neatest and most
famous masters to restrain the force of their steps, the speed of movement,
and other graceful elements of the dance, so as not to cripple an entry by the
discrepancy in step and action [. . . the pas de balet] must be more expressive
and vigorous than that of common dancing, as applied in the balls and those
ordinary and domestic dances which the women proudly boast of doing as
well as the men . . . it needs to have something faster and more vivacious.40
41 English masques

Yet in London, the cast succeeded so splendidly that the poet Crowne apolo-
gised for daring to put their achievement into words. A self-effacing preface
to Calisto states that such efforts must be eclipsed by the actual performance:
Reader, If you were ever a Spectator of this following Entertainment, when
it was Represented in its Glory, you will come (if you come at all) with very
dull Appetite, to this cold, lean Carkass of it [i.e. Crowne’s printed account].
The Dancing, Singing, Musick, which were all in the highest Perfection, the
most graceful Action, incomparable Beauty, and rich and splendid Habit of
the Princesses . . . must needs have afforded you a delight so extraordinary,
that this will appear very insipid.41

Calisto afforded a short-lived epiphany of royal splendour. Fostering the


comédie-ballet, Charles II emulated Louis XIV in France, the absolutist sun
king since his legendary appearance in the Ballet royal de la nuit (1653).42
Harking back to Caroline court masques he sought to capture once more
Stuart magnificence. Charles I and Henrietta Maria made universal “motion
cease and time stand still” when they traced their noble figures on the
dancefloor: “good is here so perfect, as no worth / Is left for after-ages
to bring forth”, gushed the lyrics of Coelum Britannicum (1634).43 Calisto
belonged to an “after-age”. The gods of the Restoration stage granted at
most, as Crowne’s modest libretto knew, “the small dominion of a Star”.44
4 The baroque body
m a r k f r a n ko

Baroque dance developed from the late sixteenth-century and early


seventeenth-century court ballet, ballet de cour, and survived into
eighteenth-century opera ballet until it was eventually displaced by the dra-
matic innovations of Marie Sallé and Jean-Georges Noverre.1 Although
it disappeared entirely from European stages with the emergence of the
nineteenth-century romantic ballet, the baroque made a return in the twen-
tieth century through a series of “baroquisms” in modern dance and ballet,
as well as through historical reconstructions of a scholarly and theatri-
cal nature. There is evidence of a serious attempt to reconstruct baroque
dance as early as 1910 in Germany.2 The creation of original or specula-
tive baroque movement languages, however, was more prevalent until mid-
century. Modern dancer Alexander Sacharoff, for example, choreographed
and performed solos such as Au temps du grand siècle/Pavane royale in 1919,
and Kurt Jooss choreographed Pavane on the Death of an Infanta in 1929.3
Oskar Schlemmer was influenced by early seventeenth-century burlesque
ballet costume design in his experimental Triadic Ballet (1922). Bronislava
Nijinska choreographed Les Fâcheux for the Ballets de Monte Carlo in 1924
and danced the male lead herself.4 Martha Graham choreographed Impe-
rial Gesture in 1935 and José Limón choreographed The Moor’s Pavane,
based on Othello with a Henry Purcell score, in 1949. In all of these cases,
something understood as the marker of period style was incorporated into
a twentieth-century concept of dance modernism.
The baroque could easily seem the natural domain of classical bal-
let even if choreographers had no specific knowledge of historical dance;
for example, the Stuttgart Ballet’s Sleeping Beauty in the 1970s skilfully
used the contrast between seventeenth- and nineteenth-century ballet cos-
tume to convey the two worlds of the narrative. George Balanchine’s Agon
(1957) was inspired by François De Lauze’s 1623 dance treatise Apologie de
la danse; I have seen performances of Agon where the early seventeenth-
century influence is indicated explicitly by period bows performed in inter-
vals to the main dances. But, for the most part, the baroque body is
not visible in Agon, although doubtless still present by virtue of musical
reference.5
Baroque dance can also seem the domain of modern dance in that it
[42] preceded and was radically different from classical ballet. Unlike classical
43 The baroque body

ballet, the legs are never lifted high in the air, and the foot is often flexed at
the ankle rather than pointed. The arms do not trace airy, expansive port
de bras above the head but contained half circles in front of the chest and
not higher than shoulder level. Supple hand rotations starting at the wrist
ornament these circular arm movements. Jumps do not soar through the air
but are more like hops. In general, one could say that baroque dance blurs
the distinction between classical ballet and modern dance while remaining
a historical style in its own right. As the polarity between classical and
modern softened by the latter half of the twentieth century, baroque dance
became more visible as a viable contemporary performance option. No
matter what technical skills the dancer starts with, baroque dance technique
is not easily mastered. Being highly cerebral, musically complex and often
counter-intuitive, it demands special study.
Although full-length seventeenth-century ballets are rarely, if ever, inte-
grated into the repertories of major ballet companies and modern dance
companies present only historical reconstructions of twentieth-century
choreographers, by the mid- to late twentieth century baroque dance tech-
nique became the domain of dancers specialising in period style. Such spe-
cialisation was possible because of dances surviving in the period notation
of Raoul Auger Feuillet and Pierre Rameau.6 In the United States, Wendy
Hilton analysed the notation and translated it into performative terms.7
In 1976 Catherine Turocy and Ann Jacoby formed the New York Baroque
Dance Company, which staged full-scale productions of the eighteenth cen-
tury and brought baroque dance to a worldwide audience (see Fig. 7). In
the 1980s, French dance scholar Francine Lancelot also cracked the code of
Feuillet notation. Her dance company Ris et danceries, founded in 1980, also
toured internationally.8 Lancelot taught this material to modern dancers and
ballet dancers who incorporated aspects of it into contemporary choreogra-
phy. Former company member Christine Bayle founded L’Eclat des Muses
in 1983, and Beatrice Massin’s Fêtes Galantes was founded in 1993. Both
companies, based in Paris, as well as the New York Baroque Dance Company,
still perform today, as does Marie-Geneviève Massé’s Compagnie L’Eventail.
The specificities and peculiarities of baroque dance technique gained
greater public exposure when Lancelot staged her baroque-derived chore-
ography, such as Quelques pas graves de Baptiste (1985), for the Ballet de
l’Opéra de Paris. Similar exposure for the idea of reinvention occurred when
Mikhail Baryshnikov danced Leonid Jacobson’s Vestris (1969) or when Beat-
rice Massin choreographed for the film Le Roi danse (1999). Other dancers
currently working with baroque reconstruction and reinvention on the con-
cert stage include Linda Tomko, Sarah Edgar and Patricia Beaman. While
reconstruction attempts faithfully to reconstitute the original choreography,
reinvention takes liberties with the historical sources.9
44 Mark Franko

Figure 7 Catherine Turocy, Artistic Director, The New York Baroque Dance Company.

Another important aspect of twentieth-century choreographic creativity


is research into early modern performance practices.10 Baroque gesture is a
key area of such research; the technique of baroque gesture is not identical to
the arm and hand movements of baroque dance, nor is it the pantomime of
nineteenth-century ballet, but rather a form of dramatic action that bridges
theatre, dance and opera and has links to rhetorical delivery. The schol-
arly presentations of baroque dance reconstruction at conferences such
as the Congress on Research in Dance and the Society of Dance History
Scholars frequently focus on variants in notated dances. Such presenta-
tions are in their way extremely erudite and aimed at specialists. In some
cases, the knowledge required to stage responsible reconstructions makes
45 The baroque body

Figure 8 Mark Franko in Le Marbre tremble.

demonstrations appear to be a strictly academic exercise. Reinvention, on the


other hand, treats historical materials – technique, music, costume and sce-
narios – as springboards for the development of new work that can speak to
a contemporary public but is also anti-modernist. Klaus Abromeit’s L’autre
pas company in Berlin has experimented with the juxtaposition of baroque
and everyday movement. The commitments of pedestrian movement of the
late 1950s and 1960s are here transformed into a historicisation or histori-
cal relativisation of contemporary experience. Works by François Raffinot
such as Caprice (1988) bring baroque technique into contexts that are wholly
reinvented, thus implying that the creative premises of another era apply to
contemporary experience. Mark Franko’s Le Marbre tremble (1986) for his
company Novantiqua set in motion the caryatids of Pierre Puget through
choreographic interpretation and photographic projection (see Fig. 8). The
46 Mark Franko

idea behind the piece was that of a baroque slave figure – Puget’s models
were galley slaves – exploring physical suffering by applying baroque visual
art to dance. Recently members of Catherine Turocy’s New York Baroque
Dance Company presented experimental works (“post-modern baroque”)
based on the baroque technique and theatrical conventions contrasted with
their contemporary analogues. Choreographic reinvention is the outcome
of the authenticity debates that also took place in musicological circles
during the 1980s.11 According to these debates, despite rigorous efforts to
determine what was “historically correct”, no one interpretation could ulti-
mately be authoritative. Unlike the modernist adoption of the baroque, these
works are post-modern in that they reflect critically their own means and
ends.
Scholars who have conducted important research into the aesthetics
and politics of baroque dance include Françoise Christout, Margaret M.
McGowan, Rudolf zur Lippe and Mark Franko.12 In the background of
these and other studies lies the link between court ballet aesthetics, ritu-
als of state and early modern subjectivity. Court ballets were created for
specific state occasions and were thus imbued with allegory, diplomacy
and intrigue. Research has shown how baroque dance enables a reflection
on spectacle and power, the early modern performance of gender and the
relation of self-fashioning to political absolutism. Giovanni Careri has given
special attention to the relation of the numerous seventeenth-century ballets
based on Tasso and the development of modern subjectivity and affectiv-
ity.13 These topics, like choreographic reinvention, create a bridge between
seventeenth-century ballet and contemporary cultural criticism. Spectacle
during the reign of Louis XIV, for example, was an extension of his political
power; thus the notion of the spectacular in itself continues to suggest power
relations. In some ways, vanguard scholarship has been more influential on
baroquist reinvention than on academic reconstruction. William Forsythe’s
Artifact (1984), largely inspired by baroque dance, was among other things
a conscious attempt to create a hyperbolically spectacular ballet through
choreographic geometries.14 At the same time, the spoken word in this
piece was used to explore the theme of the fold, which clearly suggests and
refers to Gilles Deleuze’s15 influential theory of the baroque based on an
analysis of the writings of German philosopher Gottfried Wilhelm Leibnitz.
The early modern engineering of state power as a spectacle – an idea whose
relevance to contemporary life has been demonstrated by Guy Debord’s
Society of the Spectacle16 – has also been the subject of important research
on representation and power by Louis Marin, Stephen Orgel and Jean-Marie
Apostolidès.17 The antipathy for expression in post-structuralist theory and
a concomitant interest in surface finds its analogue in the baroque’s rejec-
tion of psychological depth. From the perspective of aesthetics, a renewal of
47 The baroque body

interest in allegory in the 1980s also brought baroque aesthetics to the fore
in a number of areas.18
An interesting focus of recent scholarship to emerge from this trend
is the dancing king’s body itself: the cross-dressed king and the king in
the role of the hermaphrodite extends thought on the performativity of
power to the realm of sexuality, gender and dance in the early modern
setting.19 How can the king, who represents patriarchal authority in the
body politic, represent sexual ambiguity on stage? Again, one can note
the contemporary resonance of baroque dance, not just because it sug-
gests pastiche or the post-modern recycling of historical styles, but also
because it pre-dates psychological motivation in the performance of uncon-
ventional sexualities and transgression, thus bypassing bourgeois morality
while remaining fundamentally anti-modernist.20 Mark Morris’s Dido and
Aeneas (1989), performed to the music of Henry Purcell’s opera, is a good
example of modern dance riding the wave of the baroque music revival
while evoking the story of love and death in the context of the HIV/AIDS
pandemic.21 Morris himself played the cross-dressed roles of Dido and the
Witch. While the choreography does not reflect early modern performance
practices, it is a baroquist work. It is unusual in that the most interest-
ing choreographic passages are performed to the recitatif that was usually
reserved for a more rhetorical presentation, while the set pieces tradition-
ally reserved for dancing are markedly less interesting. That is, the dancing
invades the domain of gesture and Morris uses deaf-mute sign language as his
methodology.
The desire to find an alternative to the narrative embedded in psycho-
logical motivation that has characterised much modern dance and contem-
porary ballet since the 1940s, as well as an alternative to post-modernist
formalist abstraction since the 1960s, has transformed our view of baroque
and baroquist work from a specialised historical exercise to a provocative
theatrical experiment. The non-modern became appealing and fascinating
and we might call it the post-modern archaic. This is a different gesture
from that of assimilating tradition in the interest of overcoming subjec-
tivism as, for instance, in the works of T.S. Eliot. However, the baroque
revival also complicates facile notions of the post-modern by rethinking
modernism through the lens of the early modern. In this sense, the return
of the baroque in the twentieth century could be considered a modernist
project.
German artist Oskar Schlemmer’s interest in dance developed out of
his enthusiasm for the tradition of classical ballet and his belief that
baroque costume design counteracted two trends of early twentieth-century
dance modernism: nudity and the fluttery veils of the ballerina. Schlem-
mer sought formal innovation in choreographic movement based on a
48 Mark Franko

re-conceptualisation of the human form that led him to the idea of the
Kunstfigur, the art-figure or artificial figure. This idea was largely influ-
enced by early seventeenth-century French costume drawings. Court ballet,
for Schlemmer, was “free of constraints and thus predestined to furnish
time and again the starting point for a theatrical Renaissance”.22 Schlem-
mer was perhaps the most influential artist to assimilate baroque stylistics
to the modernist project of aesthetic innovation. Interestingly, he is now
considered a progenitor of post-modern performance art. Thus, one can
say that the baroque is a cultural space of cohabitation between modernism
and the post-modern. All this would not be possible without the substan-
tial preliminary research on court ballet prior to the twentieth century.
In the mid-nineteenth century Paul Lacroix and Victor Fournel collected
and published many court ballet libretti, which contain accounts of the
action of the song lyrics (récits).23 The music for much of this song reper-
tory survives in the Philidor collection,24 but the dance music, for the most
part, does not. The baroque music revival can be traced back to the Bach
revival of the early and mid-nineteenth century; the most important event
was surely the performance of the St Matthew Passion by Johann Sebastian
Bach in the Singakademie Berlin on 11 March 1829 under the direction of
Felix Mendelssohn-Bartholdy. The Bach revival may have awakened a sim-
ilar curiosity of bibliographers and theatre historians for the nineteenth-
century research into court ballet texts.25 But the cultural production of the
late Renaissance and early seventeenth century, which is often thought of
as pre-baroque but no longer Renaissance, remained largely hidden. Musi-
cologists had long neglected the music of this earlier period.26 French court
ballet was also characterised by an extravagant costume culture from the
workshops of Daniel Rabel in the early seventeenth century and Jean Berain
in the later seventeenth century.27 The interdisciplinary nature of perfor-
mances and their elaborate production values conspire to make court bal-
let a challenging form to revive in the contemporary dance world. This is
perhaps why the work of reconstruction began with the dance technique
itself.
Along with a vogue for historical reconstruction from other periods of
dance history and the revival of baroque music in the 1980s through an
intense concentration on period instrument performance, the performance
of baroque dance was internationally well received during that decade. The
apogee of the baroque revival is said to be the lavish Arts Florissants pro-
duction of Lully’s Atys (1987), perhaps because the economic resources
expended on this production evoked the riches available to court perfor-
mance in the baroque period itself. It would be hard to duplicate such
production values in further work. (In reality there was not a lot of baroque
dancing in the production; the high point was Jean-Christophe Paré’s solo
49 The baroque body

as Morpheus.) Possibly for this reason, the baroque music revival has con-
tinued to flourish because it does not depend on production values that are
so onerous to sustain. If dance, on the other hand, is to progress beyond
what it has already demonstrated about vocabulary and performance style
to work as persuasively with the reconstruction and/or reinvention of the
entire stage space, enormous resources will have to be dedicated to it. As
scarce resources are currently devoted to work that is most obviously new
and innovative, this is a difficult proposition to sustain.
Nevertheless, baroque dance, art and music catalysed experimental
choreography during the 1980s, the decade of twentieth-century dance that
should be called “baroque”. Baroquist aesthetics and its offshoots resonated
with the post-modern recycling of historical styles. Yet, it was really not
about random sampling or an equalisation of choreographic languages, but
instead a way to reflect allegorically on movement and identity. But, coming
after the dance boom of the 1960s and 1970s, interest in baroque dance
also reflected a greater sophistication about the history of dance itself in
the Western tradition. This was due in part to reconstructions of historical
modern dance starting with the work of Isadora Duncan that began to take
place in the early 1970s. Although there is always the possibility that it can
be read as anti-democratic royalist nostalgia it also has the effect of reveal-
ing an unusual arsenal of theatrical movements, costume and make-up. It
has a sociological connection to the financial excess and irresponsibility of
the Reagan era, a time when even dance believed itself to be a big busi-
ness. The powers of defamiliarisation in baroque dance rival its historical
connotations. The idea of the baroque as a brand of post-modernism, an
anti-modernist gesture without a classical model or an extension of mod-
ernism through the absorption of the past in the generation of new forms,
was prevalent in the visual arts as well as in dance of the 1980s. Connec-
tions of the baroque to anti-normative or non-classical aesthetics made it
a cultural rendez-vous of 1980s choreographic culture throughout Europe
and America. One of its appealing effects was an assumed artificiality that
counteracted high seriousness without falling into the other extreme of
indifference.
The vogue for the baroque did not extend much beyond the 1980s and
early 1990s, however, and allusions to this style are now less frequent on
the ballet and modern dance stage. As French choreographers came of age
during the 1990s, baroque or baroquist choreography may have become
perceived in France as less relevant to the contemporary focus as a sine qua
non of choreographic creativity. Once the interest in the baroque was char-
acterised as antiquarian, its pertinence decreased and its venues dwin-
dled. Also, with the funding crisis of the early 1990s, which particu-
larly affected dance, resources on both sides of the Atlantic have been
50 Mark Franko

consecrated to preservation over performance. Paradoxically, the empha-


sis on heritage preservation in the dance world as a result of decreased
funding has served to displace the role of historical reconstruction in live
performance. International dance festivals seem less eager to programme
dance reconstructions of any period. It is also likely that the baroque project
as a whole missed its true mission as an anti-normative response both to
classical ballet and to high modernism. As such, both reconstruction and
reinvention remain unfinished projects. The baroque body is by definition a
phantasm.
pa rt i i

The eighteenth century: revolutions in


technique and spirit
5 Choreography and narrative: the ballet d’action
of the eighteenth century
dorion weickmann

The ballet d’action, a narrative ballet, was an invention of the eighteenth


century. It replaced the pompous grands ballets of baroque absolutism that
had evolved out of the Italian Renaissance intermedii. Both court enter-
tainments, grands ballets and intermedii or divertissements, had primarily
represented and glorified the sovereign; the ballet d’action, on the other
hand, was supposed to tell stories that followed their own narrative logic
and lay beyond princely power fantasies.
The emergence of the ballet d’action relied on three factors: first, enlight-
enment ideas had spread to dance theory. Hence it became possible to intro-
duce the notion that dance could and should be independent from the other
arts. Secondly, the academic ballet of the eighteenth century strove towards
a technical refinement that aristocratic amateurs practising dance could no
longer fulfil. Thirdly, theatre as a cultural institution underwent a process
of professionalisation in which the performer and the observer began to be
separated from each other.
Critical contributions for a ballet reform came from France and the
French danse d’école. Paris, the Mecca of the art of dance, housed the
Académie Royale de la Danse, founded by Louis XIV in 1661, the insti-
tution that had overseen dance and had created a style that reached far
beyond Paris or France into Europe. Students of the Académie had been
celebrated as virtuosi. Increasingly, though, the empty pompousness of
the absolutist court ballets and the rococo divertissements aroused more
and more criticism. Louis de Cahusac, librettist, Secretary of the Comte de
Clermont and also contributor to the Encyclopédie, was one of those critics
who initiated a change. In 1754 he declared the grand ballet a means simply
to achieve hollow effects and wrote that “Everything that is without action
is unworthy of the theatre; all of it becomes an ornament without taste
and without warmth”.1 Choreographer Jean-Georges Noverre, with whom
the invention of the ballet d’action is still associated today, aimed at some-
thing similar. In 1760 he published his fifteen Lettres sur la danse, et sur les
ballets2 and introduced, at least theoretically, a new understanding of the-
atrical dance: the ballet de cour, the courtly ballet and the divertissements,
were to be replaced by a dramaturgically conceived ballet with action and
[53] concept.
54 Dorion Weickmann

Despite Noverre’s intervention in dance history through his famous let-


ters and his claim to have been the first to do so, the course of ballet had
changed much earlier, at least half a century before Noverre, in England.
The English dancing master John Weaver had published his Essay towards a
History of Dancing in 17123 in which he assessed the status quo of the dance
very critically. Dancing, Weaver emphasised, had turned into a “ridiculous,
unskilful movement”4 and sunk to a low level. The prototype of empti-
ness of form and content, Weaver remarked sarcastically, could be studied
nowhere better than in Paris at the Opéra. One of the soloists there, Jean
Ballon, “pretended to nothing more than a graceful Motion, with strong and
nimble Risings, and [by] casting his Body into several [perhaps] agreeable
Postures: But for expressing anything in Nature but modulated Motion,
it was never in his Head: The Imitation of the Manners and Passions of
Mankind he never knew anything of, nor even therefore pretended to shew
us.”5 The French, according to Weaver, were always making “a confus’d
chaos of steps . . . which they indifferently apply’d, with any Design, to all
Characters”.6
Weaver suggested a revival of the ancient pantomime as counter-model
to the French boring mess. In his opinion imitation of passionate emotions
and actions through gesture without words was to be much preferred to the
stilted and overblown spectacles of the danse d’école. For Weaver, the real
weakness of contemporary dance lay in the demand for sensational caprioles.
Instead of learning “what is natural, fit, or proper”7 dancers were taught
merely to please. He recognised that “Stage Dancing was at first design’d
for Imitation; to explain Things conceiv’d in the Mind, by the gestures
and Motions of the Body, and plainly and intelligibly representing Actions,
Manners and Passions; so that the Spectator might perfectly understand the
Performer by these Motions, tho’ he say no word”.8
As ideal there emerged a mime-style dance language that orientated itself
along ancient examples, managed to exist without words and offered a real-
istic expression of human passions. Weaver tried to realise this ideal in his
own dances on stage, for instance in The Loves of Mars and Venus; A Dra-
matick entertainment of Dancing, Attempted in Imitation of the Pantomimes
of the Ancient Greeks and Romans,9 conceived for the Drury Lane Theatre
in London. Yet despite all his good intentions Weaver did not manage to
work without accommodating public taste. He could not omit declaiming
or singing; he could not forgo extensive comments explaining his intentions
in programme notes. His audience, he knew only too well, was not prepared
to follow his “dance event” that was made out of one consistent action.
On the other side of the channel the audience was even less willing to
accept a new view on and new interpretation of dance. The grand ballet that
still dominated France had reflected the autocratic make-up of the French
55 The ballet d’action of the eighteenth century

state on several levels: in the self-disciplined and deferential attitude of the


dancing courtiers, in the geometrical design that always led to the king as
the centre, and on the discursive level of the content of the story told. Gods
and ancient heroes, forces of nature and mythical creatures lived in this
cosmos, its focus always the ruler. The grand ballet, composed of dance,
declamation, song and illumination, impressed with its anti-Aristotelian
qualities, with colourful changes from entrées to divertissements, with its
intoxicating colours and shapes. The dramaturgy followed ceremonial for-
mulas; only occasionally would these formulas contain an exposition, a
development and a finale, as, for instance, in the Balet comique de la Royne
of 1581. The grand ballet completely lacked “individual expressivity”.10 The
individual dancer remained nothing but a tiny wheel in a gigantic machinery
that aimed at praising the king.
It was this state of affairs that French ballet theorists of the eighteenth
century criticised; they began to imagine a different practice of ballet. They
intended to supply the dance with an unmistakable and autonomous poetics
instead of understanding it as a hybrid form of poetry, music and painting.
Against the absolutist demand of servitude the French writers defined form,
function and content of ballet in a new way and harmonised all these new
components. Instead of presenting mere virtuosity and conceptual empti-
ness, dance was supposed to translate human emotions and affects. Instead
of simply confirming power relationships, dance was now to show “la belle
nature” itself, that is, the beauty of the human condition, of human tem-
peraments and characteristics. Dance literature from the mid-eighteenth
century onwards followed this motif of a new sensibility.
The problem of action, of the plot, had occupied a central position after
1715, during the Regency. The mythological, allegorical and heroic themes
of the absolutist period never disappeared completely during the eighteenth
century and even survived into the post-revolutionary phase in France. At
the high point of absolutism every figure pointed to the monarch’s power. His
right to represent the state was manifest by being transfigured – personally
or symbolically – into a god or legendary hero. Though the figures of the
Hellenic myths still populated the stages, their bodies told different stories.
Monarchical glory and grandeur disappeared and were replaced by human
conflicts, sentiments and souls.

Most theorists who thought about the future of ballet did not consider bod-
ily codification or dance notation any more, as Raoul Auger Feuillet had
still done in 1700 in his Chorégraphie ou l’art de décrire la danse.11 They
turned their attention to historical and dramaturgical problems. This ten-
dency became obvious in Louis de Cahusac’s three volumes called La Danse
ancienne et moderne ou Traité historique de la danse that was printed in
56 Dorion Weickmann

Paris in 1754. Cahusac judged the stage practices of his time very critically.
In his opinion, the poverty of action of the court ballets and divertisse-
ments destroyed the aesthetics as well as the function of theatrical dance:
“L’Opinion commune est que la Danse doit se réduire à un developpment
des belles proportions du corps, à une grande précision dans l’exécution des
airs, à beaucoup de grace dans le déployment des bras, à une légerté extrême
dans la formation des pas.”12 (Common opinion assumes that dance can be
reduced to the development of beautiful physical proportions, great preci-
sion in execution of airs, graceful deployment of the arms and an extreme
lightness in making steps.) Yet that was not enough in itself because it would
often only mean mechanical repetition.
Cahusac demanded that ballet abandon empty entertainment pieces and
replace them with stories in which the protagonists truly express their spir-
itual situations. Cahusac therefore pleaded for a “danse en action”, a dance
in action that substituted the art of storytelling that would reach the heart
and soul of the audience for pure bodily technique. “La Danse en action a
sur la Danse simple la supériorité qu’a un beau tableau d’historique sur des
écoupures de fleurs. Un arrangement méchanique fait tout le mérite de la
seconde. Le génie ordonne, distribue, compose la premiere.”13 (Dance in
action is as superior to the simple dance as a beautiful historic tableau is
to a bunch of cut flowers. A mechanical arrangement represents the entire
value of the second but genius orders, distributes and composes the first.)
Cahusac not only demanded that absolute spectacle be replaced by narra-
tive composition but also suggested a dramaturgical structure for all ballets:
they were to follow an Aristotelian three-part organisation in which the
exposition of the dramatic conflict was followed by its development and its
dissolution. Cahusac hoped that clarification of the subject in turn would
lead to a revival of ballet.
Jean-Georges Noverre in his Lettres sur la danse, et sur les ballets wrote
down a series of revolutionary views which, unfortunately, he could not
always carry out in practice. After he had received his education with Louis
Dupré, one of the most famous dancers of his time, Noverre made his debut
at the Paris Opéra-Comique in 1743. Later he joined the company as its ballet
master and staged his first full-length works there. His Les Fêtes chinoises of
1754 received ovations. Because Noverre could not immediately launch his
career at the bigger rival, the Paris Opéra, in 1760 he went as ballet master
to Stuttgart. As one of the favourite courtiers of the Wurttemberg Duke
Carl Eugene, Noverre celebrated one triumph after the other. Nevertheless,
he moved to the Viennese Royal Opera in 1767 where he remained in the
position of ballet master until 1774. During the years in Vienna Noverre
engaged in a violent controversy with his colleague Gasparo Angiolini, in
which both sharpened their views on meaning, form and intention of the
57 The ballet d’action of the eighteenth century

art of ballet. In 1775, at last, Noverre received the long-awaited position


at the Paris Opéra but he failed there as a result of intrigue and envy. In
1781 he capitulated and settled in London where he spent his last years as
a choreographer at the King’s Theatre. Noverre returned to France in 1796
and died in October 1810.
In spite of his professional reverses, his Letters on Dance were republished
several times in amended form. He argued that ballet had to go beyond enter-
tainment. Most choreographers neglected the beauty of naive and graceful
sentiments and presented instead symmetrical dances, mechanical repeti-
tion and spectacles without content or message. Noverre’s first rule stated
that the choreographer had to break the custom of symmetry and evenness.
Noverre used arguments of the Encyclopedists to substantiate his claims:
nature knew no symmetrical movements and human beings had differ-
ent, not identical temperaments. Therefore the theatre should make these
unmistakable differences clearly visible in every scene, in every drawing of
a character. As real ballet was nothing but the imitation of “belle nature” –
a view Noverre and many of his contemporaries firmly believed in – it had
to resist the temptation of dull symmetry.
Noverre articulated the spirit of a new age: the symmetrical design that
centred on the monarch had outlived its time. The Enlightenment with
Rousseau’s demand “back to nature” had changed civilised societies’ under-
standing of values; it had pervaded notions of right and wrong and left
its mark on political and aesthetic paradigms; it had also influenced dance
theory. For Cahusac or Noverre, representation no longer meant the glo-
rification of the king but the imitation and illustration of human passions
and sentiments. The audience wished to see not the godly ruler but itself; it
wanted to confront its own feelings, desires and needs.
The moral mission of ballet shifted to the foreground with the ability to
move the observer. All dramaturgical means had to submit to this demand.
The classical division of acts and scenes should be mirrored in the texture
of its individual parts – every particular scene within an act should have
an exposition, a middle and an end. The structure of the piece as a whole
should be mirrored in all its parts. The corps de ballet had to come out of
the shadow that the soloists had cast on it and shed its merely decorative
function. All members should be able to turn into acting individuals, though
they should never interfere with the soloists.
Noverre asserted that any situation could be turned into dance. For
instance, Racine, Corneille, Voltaire or Crébillon could be adapted to make
a ballet. In principle “il faut qu’un Maitre de Ballets connoisse les beautés
& les imperfections de la nature. Cette étude le determinera toujours a en
faire un bon choix: ces peintures d’ailleurs pouvant être tour-à-tour his-
torique, poétiques, critiques, allegoriques & morales, il ne peut se dispenser
58 Dorion Weickmann

Figure 9 Jason et Medée by Jean-Georges Noverre, music by Jean-Joseph Rodolphe 1781. Gaëtan
Vestris as Jason.

de prendre des modèles dans tous les rangs, dans tout les états, dans tout les
conditions.”14 (It is necessary that the ballet master understand the beauties
and imperfections of nature. That study will always help him to make a good
choice of them: these paintings in addition can be in turn historical, poet-
ical, critical, allegorical and moral. He cannot avoid taking some models
from all ranks, all estates and all conditions.)
Noverre also tried hard to get rid of the overloaded costumes and wigs the
dancers were forced to wear. Everything that hindered natural appearance,
covered up the figure of the dancer or concealed movement impeded the
newly developing art form. He demanded that all the actors on stage had
to be free, free to move. The face, above all, had to be laid bare and made
visible because it formed the centre of the silent scene, it conveyed emotions
to the audience (see Fig. 9).
Expressivity demanded the trained human body. To portray “nature”
required rigorous exercises and physical control. Like Proteus, the dancer
was supposed to slip into any figure and express any sentiment that the plot of
the ballet demanded. The body turned into a tool, a finely tuned instrument.
In Noverre’s theory the body characterised the genre as well as the drama-
tis personae. Previous authors like Weaver had touched upon the problem,
59 The ballet d’action of the eighteenth century

yet it was Noverre who systematically defined the body in relation to plot
and action. He initially distinguished between three areas of dance: firstly,
the serious, heroic and noble – danse noble or sérieux, secondly, character
dance – danse demi-caractère and thirdly, comic dance – danse comique.
Physical traits classified the dancer’s place within the dance genres: height
and looks determined whether a dancer was a beautiful hero, a tender
lover or a comic fool. Such interweaving of genre and bodily qualities
pointed to the identity of content and form which became one of the
most important criteria of the neoclassical theory of art to which Noverre
belonged.
The heart of his reform Noverre described in the following way: “il faut
s’écarter avec grace des règles étroites de l’Ecole, pour suivre les impressions
de la nature & donner à la Danse l’âme, & l’Action qu’elle doit avoir pour
intéresser”15 (It is necessary to put aside in a gracious manner the narrow
academic rules in order to follow the impressions of nature and give to
dance the soul and the action which it must have to be interesting). Noverre
hated the mechanical perfection of many contemporary dancers; in his
view they were unable to perform dance in action. Action was the magic
word. So what about plot? “L’Action en matière de Danse est l’Art de faire
l’expression vraie de nos movements, de nos gestes & nos passions dans
l’âme des Spectateurs. L’Action n’est donc autre chose que la Pantomime.”16
(Action with respect to dance is the art of evoking a true expression of
our movements, our gestures and our passions in the soul of the spectator.
Action is therefore nothing but pantomime.) Hence dance and mime melted
into one dramatic plot that would be conveyed through gestures, steps and
figures. The difference between this and the model of antiquity consisted in
the elimination of language and singing. In Noverre’s ballet the body alone
reigned – without words, without arias or explanatory slogans. In practice,
Noverre never got rid of the symmetrical figurations on the stage. His Fêtes
chinoises, first performed at the Opéra-Comique in 1754, ran through a set
of variations on Far Eastern themes and La Toilette de Vénus of 1757 never
had a consistent plot though it was called a ballet d’action. Noverre had a
weakness for elaborate sets and costumes that were hardly less opulent than
those absolutist ones he accused of excess.
This is why Noverre clashed so violently with Gasparo Angiolini,
Noverre’s predecessor at the Viennese Opera; the controversy between
both was carried out directly and personally, in memoranda, programme
notes or public correspondence. Angiolini insisted on a clean structure
based on Aristotelian unity of place, time and plot. He objected to the fact
that Noverre’s ballets did not achieve a dramaturgical climax and instead
contained scenes superfluous to the action. For Angiolini drama with its
strict rules provided the real standard for dance. Therefore, he sought
60 Dorion Weickmann

inspiration and subject matter in works of contemporary playwrights, which


he in turn adapted to the ballet stage. Noverre, on the other hand, at least
when staging his own ballets, believed in following the principle of ut pictura,
the painterly effect that the work should attain. He flatly refused to follow
any pre-set rules just as much as he rejected the idea of notating dance. In
his view such practice would inevitably lead to stagnation. The application
of notation, he predicted, would invite simple, standardised reproductions
and replications of once original works, in addition to which, only profes-
sional notators would be able to decode complicated sign systems. Angiolini
envisioned the opposite: written signs would stabilise the ephemeral art of
dance by turning it into a very specific tradition and thus make it part of a
greater cultural heritage.
Angiolini and Noverre never agreed on the place of music. Angiolini,
who also composed music and advised other choreographers and ballet
masters to do the same, treated music and dance as two separate components
that the dancer had to unite within his own body. Noverre disagreed and
declared that a musical score would pre-set and manipulate the actions and
movements of the performer. The protagonist was supposed to translate
music into gesture, make its meaning transparent.
Angiolini represented a moderate modernisation in the development of
dance; he intended to hold on to tradition as a secure guideline. He energet-
ically demanded a consistent dramaturgy that would structure dance and
movement according to spoken theatre. Noverre wanted freedom – freedom
to invent and freedom to display sensibilities. For him Angiolini symbolised
narrow-minded dogmatism and a stubborn adherence to principles; in his
view Angiolini missed the essence of the new art: true copying of nature.
Angiolini, though, had a point when he accused Noverre of proclaiming
himself the genius-inventor of the ballet d’action. That he certainly was not.
But in true romantic fashion Noverre swept away all such considerations.
Angiolini declared that his teacher Franz Hilverding had already fused dance
and mime elements in order to make them convey a dramatic action. Angi-
olini, his protégé, had carried on from there. With his Don Juan of 1761
and his Semiramide of 1765 he had staged an action ballet, a dramatic entity
conveyed solely through dance and movements. The dancer Marie Sallé, one
of the visionaries of her day, had done away with spoken and sung language
in a Pygmalion performance in 1733 in London and thus had, in practice,
staged a ballet d’action.
Noverre would have none of that and refused to acknowledge a gradual
evolution of ‘his’ new genre.
Even without Noverre’s prodding, the age of heavy wigs and ornate cos-
tumes was coming to an end. Then, in 1772, Pierre Gardel refused to appear
on stage in full dress and decorative wig in the entrée of the opera-ballet
61 The ballet d’action of the eighteenth century

Figure 10 Dancers from the Australian Ballet as hens in La Fille mal gardée, 1970s.

Castor et Pollux. The audience greeted his bare head with thunderous
applause – the dam had broken. The new trend produced its first real
triumph in La Fille mal gardée, first performed on the eve of the French
Revolution in 1789 in Bordeaux. Libretto and choreography had been writ-
ten and developed by Jean Bercher (called Dauberval), who had been a pupil
of Noverre but had parted acrimoniously from his teacher.
Dauberval told a love story, in a rural setting: Lisa, daughter of a female
landowner, is to be married to the clumsy Alain, also the child of wealthy
peasants. But the girl loves Colin, a poor farmer. In order to get her beloved
she has to battle with her cunning and unscrupulous mother, but eventually
succeeds and wins the man of her own choice. La Fille mal gardée, still in the
repertories of our ballet companies today, was the first ballet to unite all of
Noverre’s requirements: a dramatically consistent narrative with character
portraits and a clear subject, a piece that was totally different from court
affectation and allegorical plots full of myths and gods. Dance alone told the
story. What Weaver, Hilverding, Angiolini and Noverre had begun, Dauber-
val completed – a narrative ballet that really deserved its name – a ballet
d’action (see Fig. 10). Apart from the feeling of saturation experienced by
the theatre-going public with exalted absolutist hymns of praise, there were
other dynamic relationships working in the eighteenth century to produce
the desire for reform.
In addition, ballet technique evolved during the eighteenth cen-
tury beyond the amateur’s ability to execute it on stage. The absolutist
grand ballet could recruit its performers from talented courtiers; their
62 Dorion Weickmann

education included dancing as well as fencing, riding or playing music. The


foundation of the Académie de la Danse as a certified institution separated
stage and stalls, performer and observer. The Académie professionalised
dance by regulating dance training and promoted professional dancers to
replace courtiers. In 1713 the Ecole de danse, part of the academy, had no
other object than to produce professional dancers.
The eighteenth century also generated the phenomenon of the étoiles,
the star performers at the Paris opera, who vied for the admiration of the
audience. Marie Sallé, Anne Cupis de Camargo, Marie Madeleine Guimard,
Jean Pierre Aumer, Pierre and Maximilien Gardel, Gaëtan Vestris and his
son Auguste enchanted the public and received unprecedented ovations.
Their brilliant performances made ballet an art of technical perfection. At
the same time the Paris opera house employed a corps de ballet, a group of
well-prepared yet badly paid professional dancers, men and women from
the poor quarters of Paris. As technique improved, the world back stage
expanded. That world turned into a complex artistic cosmos in which the
rise and fall, the glamour and wealth of the soloists was intricately inter-
woven with the misery of the poverty-stricken members of the ensemble.
Professionalisation of dance also meant that social dance and stage dance, up
to the early eighteenth century inextricably linked to one another, gradually
moved apart. The routine of accepted dance forms such as musette, tam-
bourine, chaconne or passepied or other popular dances with a set of elaborate
steps or movement sequences were replaced. The eighteenth-century books
describing body and dance techniques show just how much over a relatively
short period of time the movement and step vocabulary changed, advanced
and increased. Raoul Auger Feuillet had been one of the first dance masters
who attempted to write a grammar of dance and to notate dance by fixing
positions of feet and legs as well as by designing spatial patterns. Positions
of arms, later called port de bras, were more or less completely absent. But
by 1725 Pierre Rameau’s Maı̂tre à danser, published in Paris by Jean Villete,
offered a supplement to the sequences Feuillet had compiled and now also
included arm movements and new combinations with turns, pirouettes. In
1770 a dance master called Malpied,17 who had set up shop near the Paris
opera, presented a complete codification of arm positions. Just as the feet
had been placed in five agreed positions (positions that had turned into
part of an authoritative technique since the seventeenth century), so the
arms and hands of the dancer now also had to be used according to a set of
rules. The neoclassical demand to harmonise content and appearance had
taken hold of the entire body. It had to be shaped and moulded according
to a powerful superior will towards form that fixed the beginning and end
point of every movement. The danse d’école drove form and technique to
new heights. But such excellence and precision in execution called for strict
63 The ballet d’action of the eighteenth century

self-control; it was far beyond the means of a non-professional to be able to


fulfil such technical demands.
Authors of scholarly books now focused on training future dancers and
not on pleasing amateurs. Social dance hence was only dealt with in passing.
The last dance notation of the eighteenth century was written by Gennaro
Magri who had made a name for himself at the opera houses in Naples,
Vienna and Venice. His Trattato di ballo from 1779 clearly demonstrates
how complicated the movement sequences had become over the last eighty
years. Since the beginning of the century, since Feuillet’s Chorégraphie,
the repertory of steps, figures, turns and jumps had been augmented by
about one third.18 The professionalisation and academisation of dance had
made progress and began to pay off. Quantitative as well as qualitative
changes emerged: movement patterns had become complicated, combina-
tions diverse, difficult passages were woven into one another rather than set
apart; the obligatory arm movements demanded skilful coordination. Sev-
eral steps and pirouettes, often composed of challenging variations, could
be accomplished only by professionals. All this brought about the division
between stage and social dancing.
The departure of stage from social dance had another cause: the organ-
isational and public structure of the opera house had undergone dynamic
development. Though this phenomenon still needs to be studied far more
extensively, several contemporary sources indicate that the aristocracy was
increasingly losing its privileges. Boxes and stalls were being sold to the
emerging bourgeoisie, not given away to favoured courtiers. That devel-
opment had begun very gradually in the era of Jean-Baptiste Lully, the
composer who had been opera director under Louis XIV. Lully moved the
opera from its location at court to a separate building, the Palais Royal,
and opened it to the interested public.19 In the first half of the eighteenth
century the opera house was already frequented by the aspiring bourgeoisie
who mixed with the king and his court. The French bourgeoisie, divided
into representatives of trading and banking, factory owners, non-aristocratic
notables and professionals such as doctors, engineers and lawyers, climbed
the social ladder, whereas the disenfranchised, impoverished or disowned
aristocrats from the royal court increasingly lost their once influential posi-
tions. But the landowner who also kept a little artisan workshop in the city,
the speculator, the journalist, the doctor, the notary, the barrister or solicitor
all cultivated aristocratic habits. They went to the opera in order to amuse
themselves and copied those customs that they had or should have criticised
as aristocratic evils, above all sloth. Instead of acting they often only re-acted
or re-enacted. Their ambiguous judgement of opera, and even more so of
ballet, reflected a tendency to preserve a “bourgeois Versailles”, but with-
out absolutist commitments. On the one hand, the bourgeois audience saw
64 Dorion Weickmann

Figure 11 ‘Modern grace, – or – the operatical finale to the ballet of Alonzo e Caro’ (Rose Didelot;
Charles Louis Didelot; Madame Parisot) by James Gillray, 1796.

theatre as the embodiment of moral decay and sin, as the perpetuation of


those gallant ways now repudiated. On the other the theatres, the opera
houses became places of merrymaking, temples of amusement where the
factory owners and merchants forgot about their money troubles.
The growing demand for amusement had its dark side also in that it
forced the artists to provide enjoyment. If the enlightened bourgeois wanted
to see light-hearted pieces, then the theatre or opera house had to offer that
illusion of graceful distraction; if the bourgeois wanted it together with a
little historical instruction and moral edification, then again, the theatre
had to oblige. Only one thing was out of the question: nothing was allowed
to resemble the grands ballets. The answer the choreographers had at hand
was the ballet d’action. It seemed an ideal, flexible vehicle that made it
possible to find endlessly new variants of human experience. What Jean-
Georges Noverre had foreseen and proclaimed in the mid-eighteenth cen-
tury was never forgotten: “Rien n’interesse si fort l’humanité que l’humanité
même”20 (Nothing interests humanity as strongly as humanity itself). This
sentence contains the secret motto of the ballet d’action, it summarises its
ideological credo and comprises its aesthetic programme (see Fig. 11).

Translation Marion Kant


6 The rise of ballet technique and training: the
professionalisation of an art form
sandra noll hammond

On any given morning of the year, if you were to ask a ballet dancer, “What
are you going to do today?” the answer most probably would be, “First, I’ll
take class.” “Taking class” means the daily regimen of formalised exercises to
refine, strengthen, maintain, and prepare the dancer’s body for performance.
This is the leçon or lesson – a process based on a codified, although ever-
evolving, academic theatrical dance technique, done under the supervision
of a ballet instructor. This chapter will discuss early ballet technique and
training, with particular focus on developments in the eighteenth century,
when the codification, the instruction, the academies and the performing
companies – the professionalisation of ballet – became well established,
setting the tone for decades to come and influencing the art into our own
day.
At the outset, it is important to acknowledge that the historical traces
of the development of ballet technique and training, as well as of ballet
repertoire, are relatively rare. Unlike its sister arts, music and drama, bal-
let did not develop a comprehensive and universally accepted way to leave
written or notated records capable of reflecting the complexities of its tech-
nique and choreographies, although in the early eighteenth century there
was one valiant attempt at notation. Prior to the early nineteenth century,
there were no detailed accounts of systematised training practices for pro-
fessional dancers, although there are many tangential sources about train-
ing exercises from earlier periods and many later sources for corroborative
material.
The complexities of ballet’s highly codified technique required a method-
ical and formal approach to the training of dancers; dancers were not self-
taught; they studied with master teachers who had acquired certain cre-
dentials and attained a necessary level of competence. Gradually, this led to
the formation of schools or academies of dance. Thus, a common defini-
tion of ballet is “A form of Western academic theatrical dance based on the
technique known as danse d’école (the classical school)”.1 Although certain
individuals have had important influences on the development of ballet
technique and training, and certainly on stylistic and choreographic trends,
any one dancer’s contributions must be seen as part of a continuum in this
[65]
66 Sandra Noll Hammond

highly institutionalised art form. An important early institution to legitimise


dance in the Western world, “as a discipline with both artistic and scholarly
lineage – its establishment as an art rather than a guild-regulated craft”, was
the Académie Royale de Danse, founded in Paris in 1661.2 The young King
Louis XIV, already an avid patron of ballet productions and a skilled per-
former in court ballets, signed the letters of patent for the founding of the
academy. The documents themselves were the result of efforts by thirteen
young dancing masters “to explore new directions of dance technique, to
go beyond the mastery of their fathers and other dancing masters”.3 These
mid-seventeenth-century dancers were requesting autonomy from the min-
strels’ guild, which, from its founding in 1321, had “supervised the training
of dancers throughout France, granting them the mastership that allowed
them to teach”.4 This tradition had ensured that those who taught dancing
were perforce also musicians, and the violin, typically played by the danc-
ing master himself, came to be the instrument used to accompany dance
classes.
Having been granted independence from the musician’s guild, the new
academicians now could instruct classes of aspiring dance teachers, approve
or reject new choreographies, and confer teaching credentials on their stu-
dents – all this without having to pay the customary fees to the musicians’
guild or relying on its approval.5 The new academicians did not, however,
sever the close association between music and dancing; dance masters still
were expected to play a violin to accompany their classes, and many also
continued to compose and to arrange dance music. The dancer and dance
master Pierre Beauchamps composed dance tunes as well as creating chore-
ographies for productions with both the composer Jean-Baptiste Lully and
the playwright Molière. Beauchamps, who counted Louis XIV among his
private pupils, became the second director, in 1680, of the Académie Royale
de Danse. Beauchamps moved the academy from its headquarters in the
Louvre to his own home, where free dance classes were given each Thurs-
day for professional dancers as well as for noble amateurs.6 By the articles
outlining the formation of the academy, the academicians could hold their
meetings at a place of their choosing and rented at their own expense.
Beauchamps probably offered his house in an effort to be economically effi-
cient; thereafter the academy met at the house of the subsequent heads of
the academy.
The stated purposes of the academy, “to restore the art of dancing to
its original perfection and to improve it as much as possible”,7 seem to
have been taken to heart by Beauchamps, for, as acknowledged by other
dance masters and writers, Beauchamps is credited with the codification of
the five basic positions of the feet (all ballet steps, movements, and poses
relate in some way to one or more of these positions) and for the regulation
67 The rise of ballet technique and training

Figure 12 The five positions of the feet, as depicted by Feuillet in his Chorégraphie, 1700.

of the movements and positions of the arms.8 These are the rules that have
formed the foundation of ballet technique (Fig. 12).
Around 1680, Beauchamps also began to develop a system of dance
notation, but it was another choreographer, Raoul Auger Feuillet, who, in
1700, was the first to succeed in publishing a revolutionary new system
of recording dance by means of abstract symbols.9 Chorégraphie, as the
Feuillet system was called, could indicate positions, steps and the seven
basic movements of the dance – plié (bend), élevé (rise), sauté (jump),
cabriole (beat of the legs while jumping), tombé (fall), glissé (glide or slide)
and tourné (turn). The notation symbols, which could be augmented with
additional signs to indicate more complex movements, were aligned along
68 Sandra Noll Hammond

Figure 13 An example of the opening phrase of a theatrical duet, as notated by Feuillet in his
Recueil de Dances, 1700. The woman is on the right side, the man on the left.

the choreographic figure or pattern of each dance. The accompanying music


was written at the top of each page of the dance notation, and a small
line drawn across the dance pattern indicated each corresponding musical
measure (see Fig. 13).
A collection of social dances and a collection of theatrical dances were
published with the first edition of Chorégraphie.10 Feuillet’s notation system
proved to be popular, and for the next twenty-two years annual publications
69 The rise of ballet technique and training

of notated dances were eagerly sought by those wishing to study the latest
choreographies from Paris.
By the early eighteenth century, France, and specifically Paris, was the
acknowledged leader in the art of both social and theatrical dance in Europe.
Throughout the century, study with a French dancing master, either in
France or with one of the many teachers who had emigrated to other Euro-
pean courts or to English and American cities, was considered essential
for an aspiring dancer. Ballet terminology was, and remains today, in the
French language. This did not mean that there was only one style of the-
atrical dance in eighteenth-century France or elsewhere. Leading soloists
typically were assigned to one of the three, sometimes four, stylistic genres:
the serious (or heroic, noble) style, the demi-caractère (or gallant) style, the
comic and/or grotesque style. These genres called for different body types;
the serious/heroic style was best suited for a tall, statuesque dancer, whereas
the comic or grotesque categories required a dancer with a more muscu-
lar, compact build. More importantly, the different genres were associated
with the portrayal of different types of characters in a ballet. This in turn
determined the types of costumes worn as well as the choice of certain move-
ments and the manner in which steps and body positions (“attitudes”) were
performed. A dancer’s training, although steeped in the basic fundamentals
of ballet technique, would require emphasis not only on the technical but
also on the stylistic needs of the genre appropriate to the individual. Indi-
vidual training, studying privately with a dancing master, was widespread,
but early in the century an official school was established for the training of
professional dancers.

Dance training
In 1713, the Paris Opéra gave formal recognition to its dance constituents
by establishing a permanent troupe of twenty dancers, ten women and ten
men,11 and by establishing a school of dance for the purpose of “train-
ing performers suitable to replace those who are found lacking”.12 Classes
were to be free of charge, but limited to those already employed in the
ranks of the opera dancers in order to perfect their technique. By all
contemporary accounts, dance training continued to be largely at private
studios and must have begun at a very early age. In 1707 Johann Pasch
advises that lessons for a would-be dancing master should start at the
age of six and consist of at least three to four hours of daily practice.13
Prodigious child performers were commonplace in the eighteenth century.
The dancer and choreographer Marie Sallé, for instance, made her first
recorded appearance at about ten in London. Her partner was her older
brother, aged twelve. Dance instruction was a necessity, not only for children
70 Sandra Noll Hammond

and youth, but also for seasoned performers who wished to enhance their
technique.
Dancers often came from families who were associated with the per-
forming arts; they trained with their parents or with relatives who were
professional dancers. François Ballon, a Parisian dancing master, was the
father of the celebrated Opéra dancer, teacher and choreographer Claude
Ballon, who in 1719 became head of the Académie Royale de Danse. Sallé,
the daughter of an actor-tumbler, was a member of a large family of enter-
tainers at the Parisian fairs. It was with these itinerant artists that she had
her first training. Then, according to her London publicity, she and her
brother had become “scholars of M. Ballon (that would be Claude), lately
arrived from the Opéra at Paris”.14 These are examples of the familial dance
lineage so prevalent in the eighteenth century, and they underscore the very
personal handing down of ballet technique from one generation to the next
that has always been and continues to be a feature of ballet training.
Information on the dance studio where François Ballon taught, found
among contracts and other legal documents of professional dancers,
describes it as containing two large mirrors and a small one, two arm-
chairs, a gilded wooden plaque, a small crystal chandelier and four bro-
caded benches.15 Dance masters from the period mention the importance
of mirrors for checking one’s form when practising alone: “I would advise
you to stand before a glass and move your arms as I have directed, and if you
have any taste you will perceive your faults, and by consequence mend.”16
Mirrors remain a feature of ballet classrooms. Importantly, a main feature
of Ballon’s studio was empty space, a wooden floor on which his students
could practise, first learning the fundamentals of ballet and then eventually
studying the steps of virtuosity and the nuances of individual expression so
necessary for career advancement and public acclaim.

Ballet technique: basic principles


Even more fundamental to ballet technique than the five positions of the
feet was and is the vertical, balanced stance of the dancer. In the words of
dance master Pierre Rameau in 1725, “The head must be upright, without
being stiff, the shoulders falling back, which extends the breath, and gives
greater grace to the body . . . the waist steady, the legs extended, and the
feet turned outwards”17 – a posture that reflected the elegant bearing of
the nobility. Basic to that elegant and graceful bearing was the turnout or
outward rotation of the legs from the hip joint. This “handsome carriage of
the leg”,18 was displayed on the ballroom floor as well as on the fencing field,
and it was incorporated into the training of dancers for the stage, where it
71 The rise of ballet technique and training

was found to be a more efficient and visibly legible way for dancers to move
quickly while facing an audience seated “out front”.
“In order to dance well, Sir, nothing is so important as the turning out-
wards of the thigh,” wrote ballet master Jean-Georges Noverre,19 and, as
stipulated by earlier authors, such as Giambattista Dufort in 1728, both legs
should be rotated to the same degree so that the toes are equally turned
out.20 To be turned out “tout à fait”,21 or absolutely, was seemingly con-
trary to nature, but could be accomplished by regular practice of specified
exercises.

The structure of the lesson


Typically, the first exercises, then as now, were the pliés, the bending and
then straightening of the knees while standing in the basic positions of the
feet. Rameau noted the fundamental importance of these movements, even
for the amateur: “dancing is no more than to know how to sink and rise
properly”.22 The rise also could be onto the balls of the feet (sur les pointes
was the usual phrase) in order to make the insteps both strong and flexible.
These and other “premiers Exercices”,23 could be practised while holding
onto the hands of the dancing master, then later by lightly holding onto the
back of a chair and then finally without support.24 Each exercise had a
particular purpose. “In order to be well turned outwards [in the legs]”,
Noverre, in 1760, advocated the “moderate but regular exercise” of ronds
de jambes en dehors and en dedans, and grands battements tendus “working
from the hip”.25 The ronds de jambe are circling motions of the leg; the large
battements, or beats, are straight (tendu), forceful lifts of the leg, followed by
a firm closing to fifth position. To the “height of the hip” eventually became
typical for the execution of grands battements, although virtuoso dancers
such as the grotteschi would routinely execute higher lifts of the leg. In 1779
Gennaro Magri, a skilled performer in the grotesque category, suggested
that the large battements go “at least as high as the shoulder”, but he also
advised the student “to keep the supporting leg very steady”. Otherwise, he
warned, careless practice could result in injuries, and he gave some vivid
examples.26
Small beats and circles of the lower leg were important to practise,
because they frequently were used to embellish steps, jumps and turns.
Students were advised to start these exercises slowly and then gradually to
increase the speed. Many repetitions were important. In discussing a par-
ticular battement exercise, Magri said to repeat the beating action as many
times as possible and with the greatest possible speed in order that the move-
ments may be used with ease when dancing. Besides, he argued, the quicker
the action, the more beautiful.27
72 Sandra Noll Hammond

It was important first to practise these preliminary exercises by hold-


ing onto a chair or rail and finally without any support. Then the lesson
continued with a series of movements that emphasised balance, control and
harmony of design. By the early nineteenth century, these exercises had coa-
lesced into well-defined combinations of movements designated by names
such as the tems de courante, the coupés, the attitudes, the grands fouettés,
and so on.28 The names as well as the movements themselves were derived
from those used in the training of dancers in the eighteenth century, and
they in turn had originated as steps or step sequences used in dances. As
classroom exercises emphasising aplomb, these combinations became quite
lengthy, progressing from simple to complex, adding turns and higher lifts
of the leg, as well as increases in speed.
Training included the practice of a variety of turns (pirouettes), steps
(pas), and combinations (enchaı̂nements) linking several different steps.
Some steps could be executed close to the ground (the temps terre à terre);
others required high elevation (the temps de vigueur) and many were embel-
lished with beats and quick ronds de jambe. Thus far the lesson embodied
what Noverre termed “the mechanism of the dance”. By that he meant
the “steps, the ease and brilliancy of their combination, equilibrium, sta-
bility, speed, lightness, precision, the opposition of the arms with the
legs”.29

The making of a professional dancer


Ultimately, however, the dancer, to be successful, needed to cultivate a vari-
ety of expression in his movements. In order for this to happen, Noverre
wrote, it was necessary for the dancing master “to arrange entrées” in which
the student “would have many passions to represent”.30 Moreover, individ-
uality was important. Giovanni-Andrea Gallini, a dancer, choreographer
and theatre impresario contemporary with Noverre, wrote in 1762 that “a
dancer, like a writer, should have a stile of his own, an original stile”.31 Echo-
ing this sentiment many years later, G. Léopold Adice, a dancer and teacher
at the Paris Opéra, recalled ballet classes during his youth in the early 1800s.
To finish the class, he said, students would perform sequences of intricate
steps

designated by the name of entrée de ronds de jambe, entrée de fouettés, etc.,


and from them each dancer chose the one that seemed fitting to his
inclinations. He gave particular attention to the task of perfecting it in order
thus to create for himself a kind of dance and execution of it that were
uniquely his.32
73 The rise of ballet technique and training

Magri, too, stressed the importance of inventing “new, surprising and


pleasing things” in order to “make an excellent ballerino”. He also revealed
that in the chaconne “for the most part, all the solos are danced impromptu”,
meaning the dancer invented or improvised his solo.33 This was no small
assignment, inasmuch as the chaconne was one of the most intricate and
lengthy of the dance forms.
Individuality and invention were prized in performers; it is not surpris-
ing that Noverre would declare “undoubtedly, one of the essential points in
a ballet is variety”.34 He deplored the excessive use of “caprioles, entrechats,
and over-complicated steps” at the expense of expression and sensitive inter-
pretation of the character and style of the ballet. This, then, was the challenge
facing dancers and ballet masters in the eighteenth century – how to embrace
both virtuosity and expression.
There were individuals who succeeded in early attempts at this fusion and
in crossing the boundaries associated with the customary genres. One was
the versatile English dancer Hester Santlow, who was also an accomplished
actress in both tragic and comic roles. Her repertoire as a leading dancer
at Drury Lane Theatre included serious minuets and passacailles as well
as Harlequin dances and diverse characters in the new type of pantomime
ballets by John Weaver. Seven of Santlow’s vast repertory of dances survive
in Feuillet notation.
Italian dancers were especially fond of and skilled at swift movements
that marked the virtuoso dancer – multiple beats while jumping (“capers”)
and high elevation both in jumping and in leg extensions. The clearest
account of this style from a dancing master comes from Magri in Naples
in 1779. Early in his treatise Magri wrote about how “we moderns” differ
from “the ancients”. One of those ways was by putting “more steps into one
bar” of music. Where three battements had been done, now “up to eight or
ten battements are repeated”.35 But even earlier in the century movements
requiring great virtuosity were well known, as cited in German texts by
dancing masters Louis Bonin (1712) and Gottfried Taubert (1717) who
described cabrioles (beats of one leg against the other) where the feet were
raised to the height of the hips.36 And, interestingly, Bonin was discussing
the “serious dance”, where apparently all was not terre à terre.
French dancers were noted and admired for their grace and elegance
in the serious or noble style, especially at the Paris Opéra, but even there,
and certainly at other French theatres, dancers in the demi-caractère and
comic genres gained accolades for their virtuosity. Antoine-Bonaventure
Pitrot, a dancer and ballet master, worked in London, Warsaw, Dresden,
St Petersburg and in many Italian venues as well as in Paris at the Opéra and
the Comédie Italienne, where he also directed diverse productions – serious,
74 Sandra Noll Hammond

pastorale, comic. Magri acknowledged Pitrot’s strength and virtuosity as


“incomparable” and admired his superb “aplomb”, describing how Pitrot
balanced “on the tip of his big toe” so that his entire leg from thigh to foot
was in “one perpendicular line”.37 This, of course, is the stance of a ballerina
today on pointe.
Although there seem to be no accounts of female dancers clearly dancing
on full point in the eighteenth century, there are illustrations and descrip-
tions from 1801 and soon after depicting female dancers on full point.
Such accomplishments, in the un-reinforced shoes of the time, would have
required considerable prior training and practice. In the eighteenth cen-
tury, dancers certainly were rising higher and higher onto the tips of their
feet, as indicated by contemporary accounts of Giovanna Baccelli “stand-
ing on the toe of her foot” and “alighting, standing and pirouetting on
the toe”.38 In any case, eighteenth-century danseuses clearly were capable
of great virtuosity. One vivid example is Marie Camargo, whose brilliant
twenty-five-year career, beginning in Rouen and then as a star of the Paris
Opéra Ballet, had significant influence on eighteenth-century Western the-
atrical dance. Noverre recalled Camargo’s dancing as “quick, light, full of
gaiety and brilliancy. She could perform with extreme facility jetés battus,
the royale, cleanly cut entrechats.”39 Steps such as these usually had been
the prerogative of male dancers, but Camargo challenged these and other
norms, paving the way for other female dancers. Not long after her debut at
the Opéra in 1726, Camargo grabbed her opportunity for achieving noto-
riety and acclaim. When Dumoulin, a leading danseur, failed to appear for
his solo entrance, Camargo quickly took his place, “improvising the part of
the absentee dancer, under showers of applause”.40
Undoubtedly the century’s most accomplished technician, its greatest
virtuoso, the dancer whose performing range encompassed aspects of all
three genres – serious, demi-caractère and comic – was Auguste Vestris. A
child prodigy, who received his early training from his father, the illustrious
noble dancer Gaëtan Vestris, twelve-year-old Auguste first appeared on stage
at the Paris Opéra in 1772 and soon after was admitted there as a pupil.
Four years later he was a soloist. He was especially noted for his prodigious
elevation, rapid pirouettes, the lightness and precision of his cabrioles, and the
multiplicity of his entrechats. These and other tours de force were performed
with an ease and grace and artistry hitherto unseen. As he matured, Vestris
developed “a dramatic talent that seemed to grow with every new rôle he
undertook in the ballet-pantomimes that were becoming so popular”.41
Vestris not only succeeded in embracing both virtuosity and expression, he
also “mixed together elements from all the styles . . . to the point where a
new and unique composite style of dance came into being”.42 His career as
a dancer lasted some forty years, but his influence continued long after as
75 The rise of ballet technique and training

one of the leading teachers of the Paris Opéra, where he helped “prepare
dancers for the demands of the new style”.43
By the end of the eighteenth century, the dance form that had begun to
flourish in the court of Louis XIV could boast of a theatrical dance technique
with a language, a step vocabulary and rules of movement for the arms, legs
and torso in a system of training that was international in the Western
world. A great many of its leading dancers, instructors and choreographers
were truly international in their careers, going from theatre to theatre, city
to city and country to country. Ironically, at the Opéra in Paris, where so
much influence had emanated, its School of Dance, by the 1770s, still “did
not yet play a very prominent part in training soloists, or sujets, as they
were called”.44 Most of them apparently studied privately. But, as ballet
became a more important component of productions at the Opéra, and as
the popularity grew for the newer dance-pantomimes or ballets d’action, the
number of Opéra dancers was increased and new performance standards
were demanded for even the lower ranks, the choeur de la danse (later called
the corps de ballet). These changes required reorganisation of the school
so that it could supply well-trained dancers at all levels for the company.
In 1780, Jacques-François Deshayes, a dancer and later ballet master at the
Comédie-Française, was appointed director of the school, a position he
held until his death in 1798. Under Deshayes’s able leadership the school
began to set a higher standard for ballet training. At the time, two full-time
teachers were the norm at the Opéra; a number of private teachers were
“attached to the Opéra but not on its regular payroll”.45 The 1799 regulations
state that classes would be daily, taken alternately with the two masters, “as
the fundamentals are unchanging and uniform”. From 1805 until around
1827, the regulations stipulate that only three regular classes per week were
required, the boys studying on Mondays, Wednesdays and Fridays while the
girls were taught on Tuesdays, Thursdays and Saturdays. The regulations of
1807 mention two levels of classes, première or elementary and supérieure
or advanced. Upon graduation, usually at the age of sixteen, students would
continue to study privately, at their own expense, in order to perfect their
technique.
Vestris and other teachers, notably Jean-François Coulon, instilled more
rigorous classes, attracting talented dancers of the next generation, such as
August Bournonville and Jules Perrot, both outstanding dancers and later
important choreographers; Geneviève Gosselin, an early exponent of point
work; M. Albert (the professional name of François Decombe), considered
the last of the great noble dancers; and Fanny Elssler and Marie Taglioni,
who were to become stars of the romantic ballet. Elsewhere, other important
schools were established, guided by innovative teachers, many of whom had
received their training in Paris. Salvatore Taglioni and Louis Henry, both
76 Sandra Noll Hammond

Figure 14 From Carlo Blasis, The Code of Terpsichore, 1828, showing that the raised foot is in a
“position of a pirouette on the instep”. The dancer is not quite on a full pointe, and the arms are
carried very low.

students of Coulon, founded the school of ballet at the Teatro San Carlo
in Naples in 1812. The next year an academy of dancing was established in
Milan at La Scala in order to provide a corps de ballet for its opera. Carlo
Blasis and his wife, Annunziata Rammaccini, became the school’s directors
in 1837. In his Notes Upon Dancing, Historical and Practical, Blasis discusses
the staff, admission requirements, school regulations, dress code and the
two “spacious rooms” for lessons at the academy, where the floors were
“laid with a descent [raked], like the stage itself, to accustom the pupils to
dance upon an inclined plane”,46 which was the reality in most theatres at the
77 The rise of ballet technique and training

time. As we know from his other books, the Blasis system of instruction was
rigorous indeed. But, as Blasis was quick to admit in his first publication
in 1820, his instructions emanated “from the schools of leading masters
who have contributed immensely to the progress and beauty of modern
dancing”.47 He, of course, was referring to the tradition of training that had
begun over a century before and had been handed down to him.
In summing up the goals of ballet training, Blasis lists “ease, freedom,
lightness, life, vigour and the power of leaping (sbalzo)”, all the while being
“ever mindful that the strictest harmony should exist between the music and
every motion”. He reaffirms the eighteenth-century notion that a dancer’s
“attitudes, movements, gestures and positions” should emulate the ideals
found in painting and sculpture (see Fig. 14). And, finally, in a clear echo
of Noverre’s sentiments, Blasis writes that “dancers both male and female,
should execute their dances with appropriate feeling and expression, adapted
to the part they perform; their action ought not only to satisfy the eye, but
it should also say something to the heart and the imagination – it should be
the poetry of motion”.48
7 The making of history: John Weaver and
the Enlightenment
tim blanning

Weaver’s life
Weaver was born on 21 July 1673 at Shrewsbury in the county of Shropshire,
where his father (also called John) was a dancing master. The family moved
to Oxford shortly after John junior’s birth. It is probable that he was sent back
to Shrewsbury at about the age of ten to attend the eponymous school. He
followed in his father’s footsteps as a dancing master in the town until 1700
when he moved to London to work as a dancer in the theatres of the capital.
Two years later he staged The Tavern Bilkers at Drury Lane Theatre, claiming
later that it was “the first Entertainment that appeared on the English Stage,
where the Representation and Story was carried on by Dancing, Action and
Motion only”.1 He also worked closely with Queen Anne’s dancing master,
known simply as Mr Isaac, in notating the dances devised by the latter for
court occasions. He enjoyed a mutually supportive relationship with the
other major metropolitan dancing master of the period, Thomas Caverley.
Important for his career as a writer was his friendship with Sir Richard Steele
the Irish-born journalist and playwright who with Joseph Addison edited
The Spectator (1711–14). Steele commissioned several articles on dancing for
his periodical from Weaver, gave helpful publicity to his history of dancing
(1712) and may also have had a hand in getting the book published in
the first place. In 1707 Weaver had returned to Shrewsbury, where his wife
Catherine, whom he had married in 1696, died in childbirth in 1712. After
marrying a fifteen-year-old girl, Susanna, in 1716, he moved back to London,
again finding employment at Drury Lane Theatre, where Steele had become
manager in 1714. Although his best-known ballet, The Loves of Mars and
Venus, was given a good reception when premiered in 1717, he was never
able to establish himself permanently in the capital’s theatrical world. As
the anonymous author of The Dancing-Master. A Satyr jeered:

W– –r has Understanding, Parts and Sense,


And knows right well to gather up the Pence
...
But him, as others, Stars malignant rule.
Tho’ always rubbing off, still runs a Score,
[78] Tho’ always getting, he is always poor.
79 John Weaver and the Enlightenment

By 1720 he was back in Shrewsbury, earning a living as a dancing master,


although still making the occasional foray to London. He retired formally
from the stage in 1721 to devote himself to the theory and science of his art.
However, it appears that economic necessity forced him to tread the boards
again in 1728 at the ripe old age of fifty-five as a comic dancer. Even a second
retirement the following year did not prove to be final, as he returned to
Drury Lane for another production of The Judgment of Paris, A Dramatic
Entertainment in Dancing and Singing, After the Manner of the Ancient Greeks
and Romans in 1733. He spent the rest of his exceptionally long life back in his
home town, concentrating on the running of his boarding school. According
to the distinguished musicologist Charles Burney, who was educated at
Shrewsbury School in the 1730s, Weaver’s establishment enjoyed a great
reputation. With the help of the beautiful Mrs Weaver he continued to run
it until the end of his days.

Theoretical works and choreographies


Although Weaver was a prolific writer, he has three main claims to pos-
terity’s attention in the literary department. The first was his translation of
Raoul Auger Feuillet’s Chorégraphie ou l’art de décrire la danse par caractères,
figures et signes démonstratifs (Paris, 1700), which he published in 1706 with
the title Orchesography or the Art of Dancing, by Characters and Demonstra-
tive Figures, By which any Person, who understands Dancing, may of himself
easily learn all manner of Dances, and dedicated to Mr Isaac. As the leading
modern authority on Weaver, Richard Ralph, observes, Weaver took great
pains to find appropriate English equivalents for the technical terms, con-
sulting widely among his fellow dancing masters, many of whom returned
the compliment by subscribing to the finished product. In the same year he
translated a second, shorter work by Feuillet entitled A Small Treatise of Time
and Cadence in Dancing, Reduc’d to an Easy and Exact Method, Shewing how
Steps, and their Movements, agree with the Notes, and Division of Notes, in
each Measure, and thus became “the first of his profession to make a system
of rules relating to time in dancing available in English”.2 The most impor-
tant work of his own was An Essay Towards an History of Dancing, In which
the whole Art and its Various Excellencies are in some Measure Explain’d, con-
taining the several sorts of Dancing, Antique and Modern, Serious, Scenical,
Grotesque, etc. with the Use of it as an Exercise, Qualification, Diversion etc.,
which appeared in 1712 dedicated to Thomas Caverley, whose school he
hailed as “a Nursery of Virtue and Good-Breeding”.3 For all too long, he
argued, exponents of dancing had been content to communicate the secrets
of their art by word of mouth, with the result that “an undeserved Contempt
80 Tim Blanning

has been cast unwarily on the Art, as Low and Mechanick”.4 It was his mission
to demonstrate that “it is an Art both Noble and Useful, and not unworthy
the Encouragement of all Lovers of Elegance and Decorum; without which
Gentlemen and Ladies, are but half accomplish’d”.5 Consequently, much of
the book reads more like a manifesto than a history, with a good part of the
volume devoted to showing how “beneficial and delightful” dancing is and
to defending it against the charges of the Fathers of the Church that it was
an “Effeminate, Lascivious, Amorous, Lust-exciting and dangerous Incen-
diary of Lust”, and an Occasion of and Preparative to “Whoring, Adultery,
Wantonness, and all Effeminate Lewdness”.6 The History of the Mimes and
Pantomimes, with an Historical Account of several performers in Dancing,
living in the Time of the Roman Emperors of 1728 is essentially just a reprint
of chapter 6 of An Essay Towards an History of Dancing, with a number of
amendments and additions. Thirdly, in 1721 Weaver published a treatise
entitled Anatomical and Mechanical Lectures upon Dancing, wherein Rules
and Institutions for that Art are laid down and demonstrated, acclaimed as
“the first scientific treatise to relate the kinetic workings of the human body
to a technical analysis of classical ballet positions and steps”.7

Intellectual and artistic influence of Weaver


Weaver’s translation of Feuillet’s book was greatly superior to other English
versions, not least on account of its prolific illustrations. The standardised
system of dance notation allowed dancing masters across the length and
breadth of the country to be confident that they were following faithfully
the latest metropolitan fashions. In The Art of Dancing. A Poem, published
in 1729, Soame Jenyns lamented that dancing had for long been “unfix’d
and free” and so each master taught differently, with the result that every
new dance died with its creator. Now, however, the new system of notation
meant that:

Hence o’er the World this pleasing Art shall spread,


And ev’ry Dance in ev’ry Clime be read;
By distant Masters shall each Step be seen,
Though Mountains rise, and Oceans roar between.
Hence with her Sister-Arts shall Dancing claim
An equal Right to Universal Fame
And Isaac’s Rigadoon shall last as long
As Raphael’s Painting, or as Virgil’s Song.8

The previous year John Essex had written in the preface to The Dancing-
Master that the profession was obliged to Weaver “for the Many Proofs of
81 John Weaver and the Enlightenment

his Knowledge, that are so many Helps to our Art, which in Reality he has
rather made a Science”.9 Weaver’s influence long outlasted even his long
life. In 1802 Charles Burney wrote of Orchesography, “we remember it in
general use even in the country, among the professors of the Art”.10 Also of
enduring influence was his history of dancing, even if subsequent authors
did not always acknowledge the source of their information and insights.
As Richard Ralph dryly comments, An Essay Towards an History of Danc-
ing has been used as “a source of silent borrowings”11 by English writers
on dancing until the present day. Less durable were Weaver’s productions
for the stage. Next to nothing is known about the trail-blazing The Tavern
Bilkers apart from its attempt to reintegrate mime and dancing into the
story and focus on its presentation through motion. It might have had a
poor public reception at the time but dance scholars today see this piece as
one of the earliest and most important expressions of a new ballet type, the
ballet d’action, which was to transform the ballet stage for several centuries
to come.12 He did publish detailed accounts of The Loves of Mars and Venus
(1717), Orpheus and Eurydice (1718), Perseus and Andromeda (1728) and
Judgment of Paris (1733) but they found few if any imitators. London audi-
ences seem to have preferred the coarser fare offered by John Rich at the rival
Lincoln’s Inn Fields Theatre, including parodies of Weaver’s serious ballets.
Significantly, although his last work The Judgment of Paris was claimed to
be “a d r a m at i c e n t e rta i n m e n t of dancing after the Manner of the
ancient Pantomimes”, it had a libretto by William Congreve13 set to music
by Seedo (the German expatriate Sidow).

Weaver’s significance in historical context


Weaver epitomises the cultural advantages and disadvantages of London,
which during his lifetime (and for long thereafter) was the greatest metropo-
lis in the world in terms of size, wealth, commercialism and variety. Its
population had more than trebled in the course of the seventeenth century,
reaching c.675,000 by 1700 and still rising at an accelerating pace to pass
the three quarters of a million mark by the time Weaver died. This was at
a time when only nine other European cities had populations in excess of
100,000. Moreover, the city’s stout resistance to Stuart attempts to estab-
lish an authoritarian regime ensured that it was also exceptionally liberal,
de facto if not de iure, rivalled in this regard only by Amsterdam. Conse-
quently, it also boasted the largest and most developed “public sphere”, that
space in which private individuals meet to exchange information, ideas
and criticism and thus to form a whole greater than the sum of their
parts. An early sign of its cultural implications was the emergence of public
82 Tim Blanning

concerts towards the end of the seventeenth century. If a public concert can
be defined as a musical performance, at which there is a clear distinction
between performers and audience and to which the anonymous public is
admitted on payment of an entrance fee, then the first took place in London
in 1672 when The London Gazette printed an advertisement for “Musick
performed by excellent Masters . . . at Mr. John Banister’s house (now called
the Musick School) over against the George Tavern in Whyte Fryers”.14
By the time John Weaver reached London in the first year of the new cen-
tury, there was a thriving concert scene in several different locations, from
inns to theatres.
It was just this sort of commercial opportunity that took him to the
capital in the first place. As Roger North complained, “mercenary teachers,
chiefly ‘forreiners’ had discovered ‘the Grand Secret’, that the English would
follow Musick & drop their pence freely, of which some advantage hath bin
since made”.15 Mutatis mutandis, the same applied to the closely related art of
dancing. The regularity with which Parliament convened after the “Glorious
Revolution” of 1688 meant that the country’s landed elites now gathered
in the capital each year and so a proper season developed. A prominent
feature of the elite’s social gatherings was, of course, dancing and so the
demand for dancing masters increased correspondingly. If their status was
low, their influence on their pupils was believed to be great. The Marquess
of Halifax observed caustically in Advice to a Daughter that a coquette would
adopt virtue “if she had her Dancing Master’s Word that it was practis’d at
Court”.16
As this comment suggests, Weaver’s career also demonstrated that
London’s public sphere was not so developed that the court did not mat-
ter. St James’s was not Versailles, but it still set the tone for polite society.
Mr Isaac enjoyed his acknowledged supremacy in his profession because he
was dancing master by appointment to the court. He published the dances
he devised for official occasions, so that they could then be copied at balls
in the capital and provinces. In 1706, for example, he published A Col-
lection of Ball-Dances perform’d at Court: viz. The Richmond, the Rondeau,
the Rigadoon, the Favorite, the Spanheim, and the Britannia, all notated
by Weaver. The latter also notated a dance created the following year by
Isaac called The Union to mark the union of the kingdoms of England and
Scotland. London may not have been a residential city in the manner of
Versailles or Vienna, whose economies were shaped by and depended on
the presence of the court, but its tradesmen were well aware of the advan-
tages it conferred. So they petitioned the Vice-Chamberlain to encourage
the queen to give “dancings and balls at Court on her birth night and other
public occasions . . . for a Ball at Court . . . is . . . the cause of hundreds of balls
among the quality in the City and all over England”.17 As this demonstrates,
83 John Weaver and the Enlightenment

it is misleading to postulate too sharp a contrast between the culture of the


public sphere and the culture of the court.
The court may have set the tone for the rest of the kingdom, but its
sponsorship of dancing was confined to balls. This was in stark contrast to
contemporary France, where Louis XIV not only appeared as a dancer in
ballets (for the first time at the tender age of fourteen), but lavished a great
deal of time and money on the art. Most crucially, he gave it an institutional
basis in the Académie Royale de Danse, founded in 1661 and arguably the
world’s first ballet school. With Molière and Beauchamps as choreographers
and Lully in charge of the music, Louis’s patronage secured for France a
hegemonic position in the world of dance, which was to endure for the best
part of two centuries. Yet Weaver does not even mention Louis XIV in his
Essay Towards an History of Dancing, despite devoting a substantial chapter
(number VII) to “Modern Dancing”. He grudgingly concedes “It must be
allowed that the French excel in this kind of dancing”18 (i.e. ballet) but
otherwise he only mentions them to damn with faint praise or to condemn
outright. Even the repertoire of the best French dancers “who have been
seen with so much Applause, and follow’d with so great an Infatuation”19
consists of nothing more than motion, figure and measure. The most famous
of them, Claude Balon, confined himself to trying to move gracefully, leap
nimbly and assume what he thought to be agreeable postures, “but for
expressing any thing in Nature but modulated Motion, it was never in his
head”.20
In Weaver’s eyes, it was the English who had achieved most among the
moderns: “The Dancing so much esteem’d among us, and so necessary a
Qualification for Gentlemen and Ladies, whether taught privately or pub-
lickly, I shall call common Dancing, and in which the English do not only
excel the Ancients, but also all Europe, in the Beauty of their Address, the
Gentleness and Agreeableness of their Carriage, and a certain Elegancy in
every Part.” This was thanks to the fact that “there are not better Mas-
ters for instructing Scholars in a genteel Movement and Address, than the
English”.21 He singled out Isaac and Caverley for special praise, modestly
neglecting to mention his own services to the art. English supremacy in this
all-important social dancing was no accident but derived from the coun-
try’s especially fortunate political and social culture. In “this free nation of
England”,22 there was no great divide between the classes and consequently
none of that uncouthness to be found in less favoured nations. In Mimes
and Pantomimes too, the English had shown themselves to be superior to
the degenerate Italians by getting closer to the Roman originals than any
other national group. Such complacency was the stock-in-trade of many
if not most English contemporaries, whose sense of superiority was based
securely on what seemed to them to be the mutually supportive tripod of
84 Tim Blanning

liberty, prosperity and Protestantism. No true culture could flourish in the


barren soil of despotic, impoverished and Papist France. It was the English
who had invented country dancing which – mutated into contre danses –
now dominated dancing on the continent. If this seems absurdly insular,
it should be borne in mind that Weaver’s long life coincided with three
episodes of the Second Hundred Years War between England and France:
the Nine Years War (1689–97), the War of the Spanish Succession (1702–13)
and the Seven Years War (1756–63). Isaac’s Court Dance for 1705 was chris-
tened Marlborough to commemorate the eponymous duke’s crushing defeat
of the French the previous year at Blenheim and that for 1706 was Britannia.
Although he was happy to make his reputation by translating Feuillet’s trea-
tise, Weaver had only contempt for “our French Pretenders to Dancing”,23
as he dubbed them. It is one of the clichés of historical writing that the
Age of Reason was also an age of cosmopolitanism; in reality, the nation-
alism of Weaver and Isaac was much more representative of contemporary
attitudes.
In the battle of the time between the Ancients and the Moderns, which
began when Charles Perrault read his poem The Age of Louis the Great to the
Académie française in 1687, Weaver should probably be numbered among
the Ancients. He relied heavily on Lucian’s Greek dialogue Peri Orcheseos
when writing his history of dancing, taking it at face value and clearly
not sharing some modern commentators’ doubts about Lucian’s sincerity.
What Weaver admired most about the classical world’s attitude to dancing
was its seriousness: “It began in Religion, and was, in the politer Times
of the Greeks and the Romans, the necessary Qualification of a Hero.”24
Anticipating none other than the German composer Richard Wagner, he
was especially attracted by the Greek combination of dancing with music
and poetry to form a total work of art. He concluded: “It must indeed
be granted, that our Modern Dancing, in several Particulars of Beauty,
falls infinitely short of that of the Greeks and Romans, if we may believe
Eye-witnesses of its Perfection, and admirable Effects.”25 Together with his
remarks about the degeneracy of so much of the contemporary scene, this
verdict seems to place Weaver unequivocally among the Ancients. However,
he immediately added a qualifier: “Yet this must be said, that as to Dancing in
its Fundamentals and Expediency, Modern Dancing is of equal Desert, as will
appear hereafter.”26 In fact, what followed did not show that at all, only that
what Weaver, Isaac, Caverley and the other London dancing masters were
achieving was of a high order of distinction. It is reasonable to compromise
by claiming Weaver as an early exponent of neoclassicism in the definition
offered by Charles Rosen in his analysis The Classical Style, namely: “I have
used ‘neoclassicism’ in a narrow sense of a return to the assumed simplicity
of Nature through the imitation of the ancients.”27 Weaver’s ambitions for
85 John Weaver and the Enlightenment

dancing corresponded very closely to the more celebrated demand made for
art by his contemporary, Anthony Ashley Cooper, Third Earl of Shaftesbury,
namely that it should be “chaste, severe, just and accurate”.28 Like Weaver,
Shaftesbury believed that the Glorious Revolution had ushered in an era of
gentlemanly rule and gentlemanly culture and it was this which both men
sought to foster. A free state, they held, was both supported by and helped
to create a true culture: “the high Spirit of Tragedy can ill subsist where the
Spirit of Liberty is wanting”.29 Those last words stem from Shaftesbury but
they might just as well have come from Weaver, with the word “dancing”
substituted for “tragedy”.
If, as is often the case, Voltaire is thought to be the personification of the
European Enlightenment, then Weaver must be denied enlightened status.
It was Voltaire, after all, who believed that the cultural history of Europe
could be divided into peaks and troughs, the former being Ancient Greece,
Augustan Rome, Renaissance Italy and Louis XIV’s France. In a more general
sense, however, he does merit inclusion, not least because of his insistence
on the need for standardisation, uniformity and the “accuracy” demanded
by Shaftesbury. The great problem in the past, Weaver maintained, was
that the practitioners of dancing had passed their art from one to another
and from one generation to another orally (“like the Druids”).30 The result
had been analogous to Chinese Whispers: distortion, misrepresentation,
fragmentation, decadence. What was needed was a clear, rational, and, above
all, universal system of notation. In A Small Treatise of Time and Cadence in
Dancing, Reduc’d to an Easy and Exact Method, Shewing how Steps, and their
Movements, agree with the Notes, and Division of Notes, in each Measure,
published in 1706, Weaver stressed: “The best and only Method of finding
the Cadence or Time of each Step, is to know its just Value in the same
manner as the Notes of Musick; after which all the Steps in a Measure, are
to be reduc’d to an equal Time or Length, as the Notes in the Measure of the
Tune.”31 His translation of Feuillet’s system of notation was a massive step
in the desired direction of uniformity.
Weaver also appreciated that anarchy in the dancing world had had a
detrimental effect on the status of those who taught and practised the art.
When priests were its leaders, he observed, it had enjoyed high prestige as
part of a total work of art that was also worship. But when it fell into the hands
of men who understood just the steps and taught it simply as a means of social
recreation, it slumped to become nothing more than an “amusing Trifle”.32
Indeed, the status of the dancing master at the turn of the eighteenth century
appears to have been low, the stereotype condemned as both effeminate
and lascivious. It is not known whether Weaver attended a performance of
William Wycherley’s contemporary comedy The Gentleman Dancing Master,
but he would not have enjoyed the experience. This intensely Francophobe
86 Tim Blanning

play (which also features a renegade Englishman “Mr. Paris, or Monsieur de


Paris, a vain coxcomb, and rich city heir, newly returned from France, and
mightily affected with the French language and fashions”)33 is centred on
the attempt by a true gentleman to gain access to his beloved Hippolita by
pretending to be her dancing master. As she says: “A dancing-school in half
an hour will furnish you with terms of the art.” But the father is not deceived:
“Nay, nay, dancing-masters look like gentlemen enough, sister: but he’s no
dancing master, by drawing a sword so briskly. Those tripping outsides of
gentlemen are like gentlemen enough in everything but in drawing a sword;
and since he is a gentleman, he shall die by mine . . . Your dancing masters
and barbers are such finical, smooth-tongued, tattling fellows; and if you
set ’em once a-talking, they’ll ne’er a-done, no more than when you set ‘em
a-fiddling: indeed, all that deal with fiddles are given to impertinency.”34 Yet
Weaver believed himself to be a gentleman and not the least of his numerous
achievements was to raise the station of his profession by his demonstration
that it was possible to be both a dancing master and an intellectual, thus
anticipating Sir Joshua Reynolds’s celebrated advice to painters later in the
century.35
8 Jean-Georges Noverre: dance and reform
j u d i t h c h a z i n - b e n na h u m

Noverre’s life
Jean-Georges Noverre was born in Paris on 27 April 1727 and died in Saint-
Germain-en-Laye on 19 October 1810. He had an abrupt and demanding
temperament, made many enemies and stirred up a variety of fierce artistic
controversies during his long career. More than any other choreographer
before him, he turned ballet into an independent art form and means of
expression. Diderot called him “le génie”, the one who would save dance,
and Voltaire named him “Prométhée de la danse”.
Noverre’s mother, Marie Anne de la Grange, is thought to have been born
in Lausanne, although P. J. S. Richardson suggested that she was born in
Picardy. His father, Jean Louys, was a Swiss soldier. Like Rousseau, Noverre’s
Swiss origins created suspicion about his loyalty to France, although he
always considered himself a Frenchman. His father expected him to be a
soldier, but the boy insisted on dancing and eventually followed his own
instincts. Still a child, he went to Paris to study with M. Marcel and then
with the famous Louis Dupré. His critical eye quickly led him to question
why the arms had such a circumscribed use, and to wonder why masks,
high heels, panniers and over-blown wigs were popular in ballets as they
constrained all movements of the head, arms, legs and face.
Noverre’s first dancing probably occurred at Jean Monnet’s Opéra-
Comique at the Foire Saint Laurent on 8 June 1743 in Favart’s vaudeville
Le Coq du village. There, in that free environment, he detected other ways
of using gesture and movement. He learned much by observing the Italian
comedians with their emphasis on physical and improvisational theatre. At
the age of fifteen he appeared at the court at Fontainbleau. With his “copains”,
Campanini and Lany as well as his teacher Dupré, Noverre was recruited to
dance in Berlin before the young Frederick II and his brother Prince Henry
of Prussia. It was the first of his many trips outside Paris. Still in his teens,
he performed in Dresden, Strasburg and again at the Opéra-Comique. At
the age of twenty, Noverre became ballet master in Marseilles and created
what became a well-known piece, Les Fêtes chinoises. There it is thought
that he met his wife, the actress Marie-Louise Sauveur. In 1750 he became
Principal Dancer in Lyons and partnered the celebrated Marie Camargo. In
addition, he made his first ballet pantomime, Le Jugement de Paris. He then
[87]
88 Judith Chazin-Bennahum

moved from Lyons to Strasburg in 1754 for one year, and then returned to
Paris and the Opéra-Comique. There he created ballets which organised the
elements of a production into beautifully arranged pictures. From 1755 to
1756 he and his company went to London to work with the famous David
Garrick at the Drury Lane Theatre. Like Samuel Johnson, the writer, with
whom he had gone to school in Lichfield, “Davy” Garrick was a poor boy
making it in the new commercial market for theatre and books in London,
then one of the richest and least censored cities in the world. He performed
Shakespeare with notable brilliance and gained a reputation for eloquent
and expressive gestures in his performances and for the consummate taste
of all his theatrical productions. Noverre, who received handsome fees from
Garrick, was able take his new wife and sister with him.
The London period introduced Noverre to different conceptions of the-
atrical style. Garrick belonged with Rousseau and others in the middle of the
eighteenth century to the school of “natural” theatre. In 1752 the arrival of
an Italian opera company in Paris, which performed opera buffa by Pergolesi,
Alessandro Scarlatti and other composers, triggered a fierce debate between
proponents and opponents of that new natural style, which emphasised
melody over harmony. Jean-Jacques Rousseau made his name in a famous
attack on Jean-Philippe Rameau, the most prominent court composer at
the time. This guerre des buffons (the war of the opera buffa) raised the
same issues that had begun to occupy Noverre. Like Rousseau, he rejected
artificiality, the use of masks and statuesque formalities and thought that
art had to achieve “truth”. In order for theatre to be true, pantomime must
be used, which aroused the intelligence of the viewers by engaging them
in the emotions of those on stage and thus presenting human dramas to
be considered. While working in Drury Lane, Noverre became acquainted
with Eva Weigel, Garrick’s Austrian wife. She had studied ballet with the
innovative Franz Hilverding in Vienna, who also experimented with what
would become known as the ballet d’action. As early as 1740 Hilverding
produced such dance dramas for the Viennese court as Jean Racine’s Bri-
tannicus, Crébillon’s Idoméneo and Voltaire’s Alzira. These tragedies had
proper plots and portrayed human emotions. Noverre adopted that model.
The true ballet was not defined by the steps but the sentiments and nuances
that were enacted by the dancer to create the real dramatic effect.
In spite of his progressive views, he ran into political trouble in London.
In 1756 a war between Britain and France broke out, which was part of a
worldwide conflict on the continent that was known as the Seven Years War
but in North America as “The French and Indian War”. The English audience
rioted at the performance of Noverre’s Fêtes chinoises; they were incensed by
the way the French had stirred up Indian tribes to massacre English settlers.
Both the French, whose colonial economy rested less on settlement than
89 Jean-Georges Noverre: dance and reform

trade in forest products, and the Indians, who found English pioneers cutting
homesteads from their forests, had a common enemy. Noverre had to leave
London in 1757. He wanted to work at the Paris Opéra, but quickly realised
that he would be swimming against the prevailing currents of courtly dance.
He returned to the provinces and to Lyons where he created new and very
different ballets based on his newly found expressive movement. He worked
with the seductive and distinguished Mlle Guimard in an anacreontic ballet,
Les Caprices de Galathée. He garbed the fauns in tiger skins, shoes that
looked like tree bark and leg stockings the colour of carnation for the forest
inhabitants. The nymphs’ material was speckled with flowers and greenery.
All of these features associated Noverre with the new naturalist attitude
towards costume and put him in the mainstream of the culture of the French
Enlightenment. One of his enduring successes centred on his work La Mort
d’Ajax, a tragic ballet that permitted him to portray strong emotions that
touched and moved his audiences.
Noverre always had a knack for sensing which way the artistic wind was
blowing. The year 1760 was ideal for publishing his Lettres sur la danse, et
sur les ballets and he managed to find both a French and a German publisher
respectively in Lyons and in Stuttgart. His ideas on dancing and performance
fitted the new culture in Germany based on the so-called Affektenlehre, the
doctrine of emotions or “sensibility”. Lyons had been good to him but an
invitation to serve at the court of the Duke Carl Eugene of Wurttemberg
could not be resisted. Carl Eugene, who ruled his principality for more than
forty years, may have been the most extravagant, luxurious and irresponsible
of the substantial German princes, a group notorious for conspicuous con-
sumption. He built six huge palaces on the model of Versailles on the taxes of
his 600,000 groaning subjects. In Stuttgart, Noverre had a 350-seat theatre
for his ballets, the services of the distinguished scene designer Servandoni,
the costume designer Louis-René Boquet and composers Nicolò Jommelli,
and Jean Joseph Rudolph (Rodolphe) who composed his most famous ballet
Jason et Médée in 1763. Since the duke never cared how expensive his artists
and architects were, Noverre staged ballets on the grandest scale imaginable.
He created at least twenty new works and built his reputation. As the author
of a serious treatise on choreography and ballet, he had the ideal setting to
put his ideas into practice. He attracted remarkable dancers such as Gaëtan
Vestris (“the god of dance”) who played Jason, Anna Heinl who is credited
with having invented the double pirouette, Le Picq and the extraordinary
Dauberval who went on to create the famous La Fille mal gardée (1789) in
Bordeaux.
When Noverre launched Jason et Médée in Stuttgart, he borrowed ideas
explored in Vienna by Hilverding, Christoph Willibald Gluck and the libret-
tist Raniero di Calzibigi. They too were seeking a deeper definition of, and
90 Judith Chazin-Bennahum

credibility for, ballet and opera. Noverre knew that where the ideas of the
French Enlightenment had spread, important patrons emerged; in such
areas his doctrine of natural dance, his work on gesture and expression
would find a favourable reception. And patronage was the name of the
game in the eighteenth century. For after all, Noverre was like Samuel John-
son and David Garrick, a poor boy without class. Johnson could write in
his great dictionary of 1755 that “a patron is a wretch who supports with
insolence and is paid in flattery”,1 but he could get away with it because
London had a huge commercial printing and publishing business and his
dictionary easily found a market. Johnson never needed a patron. Neither
did David Garrick with his commercial Theatre Royal in Drury Lane. But
nobody could buy tickets to Duke Carl Eugene’s theatre. Noverre worked
on the continent where royalty still controlled the arts. Hence Noverre had
to choose his patrons with extreme care. Though he wanted to return to
the Drury Lane in London, the British victory in the French and Indian
war had not yet calmed anti-French sentiments. He decided to try his luck
in Vienna, the capital of the Holy Roman Empire, the most important city
of the German-speaking world, and the place where the young Emperor
Joseph II, to the irritation of his mother, Maria Theresa, the reigning queen
(a woman could not hold the imperial title), made serious attempts to
impose the culture of the Enlightenment on the huge, polyglot realm. As
always, Noverre had guessed shrewdly. He found royal patronage and man-
aged to stage nearly thirty-eight new ballets. He also revived earlier works
and created dances for operas that were great successes, such as Gluck’s Pâris
et Hélène, Orphée et Eurydice and Alceste. Noverre knew what historians tend
to forget: in the Ancien Régime no artist could succeed without very wealthy
protectors, and in Vienna he secured the highest possible patroness, the
Queen Maria Theresa herself. He became maı̂tre de danse for the twelve-
year-old Marie-Antoinette, the queen’s favourite daughter, and remained
her close friend and confidant until she was guillotined during the French
Revolution in 1793. He knew he only had to wait until she came of age to
do as he liked. With the composer Franz Aspelmayer he created Iphigénie
et Tauride (1772), Apelle et Campaspe (1773) and with Joseph Starzer Adèle
de Ponthieu (1773). With Horaces et des Curiaces (1774) Noverre deployed a
frightening and moving scene of ancient Roman history, a combat between
brothers and also a grand military parade with extraordinary pomp and
splendour. Unfortunately, his use of a theme of the great French playwright
Corneille went too far. It must have annoyed the very conservative Maria
Theresa, who presumably ceased to protect him. In any case, the produc-
tion cost him his job. In 1774 he decided to move to the Teatro Regio
Ducale in Milan, which was the capital of Austrian Lombardy and by far
the richest Italian city, the sort of place that could afford a Noverre. Here
91 Jean-Georges Noverre: dance and reform

unfortunately his ambition collided with reality. The Milanese did not take
kindly to Noverre, and many reviews of his ballets cast aspersions on him,
criticising the fact that his works were slow moving, “With long stretches of
pure mime, and boring”.2 He grew despondent at his lack of success. Italians
were used to spectacular and exciting grotesque dancing that included jumps
and virtuosic moves and, besides, Italian national sentiment had begun to
stir. As in Germany in the 1770s, so in Italy, young, influential intellectuals
no longer looked to France for their artistic models. After only two years, he
returned to Vienna to manage their Kärntnertor Theatre. Finally, in 1776,
after a thirty-three-year journey throughout Western Europe, his long-term
investment in royal patronage paid a big dividend. In 1770 his former pupil
Marie Antoinette had married the Dauphin Louis, grandson of the King
of France, Louis XV and who thus had the prospect of becoming queen
of the richest, most powerful and most populous state in Europe. Paris
alone had more inhabitants than the entire Duchy of Wurttemberg. In 1774
King Louis XV died, and Marie-Antoinette became Queen of France. His
esteemed friend Marie Antoinette had not forgotten her dancing master.
She ordered his return and secured his appointment at the Paris Opéra.
His ambition had now been fulfilled. Noverre, poor, provincial, without
class or breeding, had managed his patronage so adroitly that now he could
rule the world of dance. But the loyalty and support of royalty can shift.
Other geniuses had access and the fate of a low-born subject mattered little
especially to the frivolous and casual queen. Noverre got into trouble with
Mozart, who had composed Les Petits Riens, a bagatelle performed at the
Paris Opéra in 1778. Noverre constructed a slight and charming theme to
Mozart’s music: the famed Vestris pursued with a net L’Amour (Mlle Allard)
disguised as a bird. Mlle Guimard captured the bird and, in a happy ending
the bird escaped. Dauberval played a character role in the piece. Apparently
Mozart was disappointed with the ballet’s production and that was seri-
ous since Mozart already enjoyed the reputation of musical genius. Noverre
with his usual arrogance must have assumed that with the queen as patron
nobody could touch him. But Dauberval and the ambitious Maximilien
Gardel enlisted a cabal against Noverre. They called him egotistical, overly
serious and, what is more, they did not like his pantomime pieces. Mlle
Guimard also rebelled against his authority and poisoned many prominent
people against him. He was unseated and, in 1779, wrote a seventeen-page
memorandum describing the difficulties of his reign. His felicitous effect
on the Paris ballet world was perpetuated by his later staging of his tragic
ballet Jason et Médée (1780) that created a sensation, as did his ideas on cos-
tume and his thoughts on ballet pantomime, which were dutifully carried
through by Pierre Gardel, who inherited the throne and the running of the
Paris Opéra in 1787.
92 Judith Chazin-Bennahum

Noverre returned to his beloved London and the King’s Theatre where
he had a brilliant season in 1781, and staged many of his successful ballets
from Stuttgart and Vienna. He also returned to Lyons from 1782 until 1787
to recreate many of his old works. There was a poetic justice in this happier
golden era of his later years. Again from 1787 to 1788 he redid older works
in London, many of which did not conform to his loftier, youthful aesthetic
pronouncements, but were much appreciated by London audiences.
The French Revolution broke out in the summer of 1789 and Noverrre,
in spite of his lowly birth, had depended on royal patronage for much of his
career and saw earlier than most that France would be an uncomfortable
place for royalists. He hid for a while in the remote area of Triel in France,
and managed to get to London, where he spent two years from 1791 to
1793. When the “Terror” and the executions of “enemies of the people”
finally ended in July 1794 with the fall of Robespierre and the Jacobins,
the fanatics of the revolutionary era gave way to the less puritan and less
devoutly republican regime of the Directory.
In 1795 Noverre could safely return to his home in Saint-Germain-en-
Laye, in front of the Château, near the Hôtel de Noailles where Mozart
had been received. Noverre was nearly destitute because of his association
with the old regime and had lost his funds from the Opéra. He returned to
visit its studios and stages where young talents were nurtured. He attended
rehearsals where he gave advice to young dancers, “and deplores that no
male dancers are being formed, with the result that the danseuses now
exceed the danseurs in numbers”.3 He republished in 1803 his Lettres in
St Petersburg, amended with the benefit of many years of disappointment
and experience. His ballets, however, continued to be revived on the major
European stages. He died in Saint-Germain-en-Laye on 19 October 1810 at
the age of eighty-three.

The intellectual and aesthetic world of the young Noverre


The art of dance paralleled other developments in theatre and performance
in the eighteenth century. Its reliance on carefully patterned steps and an
extremely stylised approach to movement, gesture and emotion associated
with the absolutist style of the seventeenth century came under criticism.
“The real meaning of dance movement was lost in the flash of twinkling
feet, in the curve of voluptuous arms, in the coquettish empty glance.”4
A new emphasis on the expressionistic possibilities of plot and movement
enlarged the emotional opportunities for dancers, especially if a ballet had
its own beginning, middle and end, and was not merely a divertissement in
an opera. Several aestheticians in the early eighteenth century had already
93 Jean-Georges Noverre: dance and reform

expressed interest in a modern form of dance. P. J. Burette (1719) and


Abbé du Bos (1719) cited Plato and Aristotle as well as the contemporary
Italian actors as advocates of gesture with meaning. Pantomime became
an important topic since pantomimes in antiquity were of high repute. In
addition, the idea of a ballet with a plot or the ballet d’action also took root.
In 1741 Rémond de Saint Mard objected to the uniformity and passionless
quality of the contemporary ballet. Five years later, Charles Batteux saw the
possibility of dance as an independent art form and wrote copiously about
its lack of variety and interest. Louis de Cahusac, a librettist, also wrote
eloquently about ballet. He was responsible for several articles on dance in
the Encyclopédie and in 1752 agreed with Batteux about ballet’s enormous
unfulfilled potential. “Sur nos théâtres nous avons des pieds excellents, des
jambes brillantes, des bras admirables. Quel dommage que l’art de la danse
nous manque.”5 (In our theatres we have excellent foot work, brilliant leg
work and admirable arms. What a shame that the art of dance is missing!)
But the most celebrated heroes of this discussion were the philosophers
of the Enlightenment. Denis Diderot, who edited the Encyclopédie, wrote
extensively on many of the arts including painting and theatre. He also took
notice of the ballet and in his Entretiens sur le fils naturel (1758), emphasised
the goal of imitating nature as had the Ancients. He felt that ballet needed a
genius, a writer who would bring down to earth those magical and enchanted
regions.6
Jean Jacques Rousseau in La Nouvelle Héloı̈se (1761) included a brief
exposition of ballet’s decadence in his diatribe. He noted that dances were
thoughtlessly thrown onto the stage without any dramatic reason. “The
priests dance, the soldiers dance, the gods dance, the devils dance; people
dance into their graves and everyone dances no matter for what reason.
Dance continually interrupts the plot or finds itself on stage for no reason,
really imitating nothing.”7

Noverre’s dispute with Gasparo Angiolini


Franz Hilverding’s dance dramas profoundly affected Italian ballet through
the activities of his pupil Gasparo Angiolini and most dance historians
believe that pantomimic dancing was well established in Italy in the 1740s.
Written scenarios or libretti were necessary accompaniments to perfor-
mances of ballet pantomimes, since the gestures and plots were not always
clear to the audiences. The ballet’s scenes and plots needed a structure, laid
out on paper, just as the plot and the scenes of a play were written. These
well-written libretti became the occasion for a passionate conflict between
Angiolini and Noverre. In his Lettres, Noverre had urged the use of the
94 Judith Chazin-Bennahum

libretti and the importance of a tight and moving plot without subplots
and silly scenes, but instead libretti founded on logic and simplicity. Not
only was Angiolini infuriated for his mentor Hilverding’s sake by Noverre’s
claims to have invented the ballet pantomime, but he was appalled that
Noverre had to depend upon a written crutch for the success of his ballets.
Using his aviso or preface to his ballet Le Festin de Pierre, ou Don Juan (1761)
and Sémiramide (1765) as forums for a debate, Angiolini quarrelled with
Noverre’s statements that ballets must be developed like plays. After all, plays
lasted as long as three hours, while a ballet would not normally endure for
more than one and a half hours. In 1773, Angiolini published a pamphlet,
in which he criticised Noverre’s use of overly long and detailed libretti. One
year later, in the aviso to Thésée en Crète (Vienna 1774) Angiolini argued
that programmes were an abuse. “If the ballets were unintelligible without
the printed scenarios, the movements of the ballet should clarify the plot.”8
The bitter dialogue continued; but history tells us who won this debate. It
was Noverre, as fairly lengthy and descriptive ballet programmes continued
to be used throughout the nineteenth and twentieth centuries.

The letters on dancing and ballet


Noverre’s greatest claim to influence in the history of ballet rests on his
famous Lettres. Shortly after their publication, Noverre sent Voltaire a copy
of the book and must have been delighted by Voltaire’s response in October
1763: “The title of your book may only indicate dance, but you shed a
great deal of light on all the arts. Your style is as elegant as your ballets
have imagination.” After expressing his sorrow at being unable to see one
of Noverre’s ballets, Voltaire added: “I think that your talent will be much
appreciated in England, since they love the truths of nature. But where will
you find the actors capable of executing your ideas? You are a Prometheus and
you must create men and then animate them.”9 Six months later Voltaire
reaffirmed his high opinion of Noverre: “I find that all you do is full of
poetry; painters and poets will argue with one another as to whether you
are a painter or a poet.”10 This praise from Europe’s most prominent thinker
encouraged Noverre to pursue his grand programme.
Like the Encyclopedists, Noverre was never content to make observations
on a level of abstract criticism. He expected his ideas to yield practical results
and from 1757 began writing a discourse that was inspired by practical
experience. His own ballets, particularly Fêtes chinoises, presented in Lyons
in 1751, Paris in 1754 and London in 1755, had already demonstrated his
desire to go beyond traditional entrées and create an integrated cohesive
presentation.
95 Jean-Georges Noverre: dance and reform

The writings of the philosophers confirmed his own thoughts. The first
edition of the Lettres of 1760 criticised contemporary ballet and enthusiasti-
cally proposed reforms. Later he added letters to defend both his principles
and his innovations in technique. In the final editions, those of St Peters-
burg, 1803, and Paris, 1807, he evaluated his previous attitudes, explaining
why he had changed or modified certain ideas and expressing displeasure
with current trends, some of which directly contradicted his basic tenets.
Noverre dedicated the first publication of his Lettres to the Duke of Wurt-
temberg who, as we have seen, allowed him to spread his wings and practise
his newly found artistic principles. Following prevailing eighteenth-century
aesthetics, Noverre insisted on truth, simplicity, harmony and the imitation
of nature. He usually linked the word nature with “belle” and maintained
that the artist must make choices in order to correct the defects of nature
and embellish his subject sufficiently to arrive at a noble or picturesque
effect. The artist must avoid servile reproduction, rejecting details that are
confusing, useless or that delay action: the art was to disguise the effort.11
The letters discuss their subjects in a question and answer format; they
address an imaginary intelligent, genteel and educated reader. Though most
letters begin with a specific problem, they quickly move to general observa-
tions and to the demands for reform that Noverre made. Noverre regularly
attacked colleagues, ballet masters and dancers alike, which made him many
enemies. But the polemics always lead back to his grand and unified vision
of a new ballet aesthetic, which he explains over and over again. Noverre
accepted the theory that the art of pantomime had its origins in antiquity
and found this a convenient justification for his own innovations. Later he
discovered that pantomime was not as effective as he had hoped, and he
looked to the future of ballet by championing the dancing of his contem-
poraries. They offered more variety, more sophistication and truth in their
simulation of natural sentiments.
Like Diderot, Noverre refused to limit his canvas to mythological and
heroic subjects. He advocated unlimited choice and urged maı̂tres de ballets
to be alert to subjects taken from daily life in the country and the city. They
were not to preach but never to forget the moral purpose of art. “How many
varied pictures will he not find amongst working men and artisans! Each
has different attitudes born of the positions and movements which his work
imposes. This stance, this deportment, this way of moving, always common
to his trade and always pleasing, must be suggested by the choreographer”
(Letters, p. 73). Indeed Noverre put peasants on stage in Annette et Lubin
and in Adelaı̈de ou la Bergère des Alpes taken from Marmontel.
In spite of his claims to represent real life, he really preferred the depth
and sublimity of tragedy. He showed in his grand historical works that
tragedy could work in a ballet (Letters, p. 85). The term historical here
96 Judith Chazin-Bennahum

embraced historical, mythological and religious themes. He criticised the use


of arbitrary, abstract, exaggerated allegory (Letters, p. 143), but accepted that
it could be instructive as well as entertaining when used with circumspection
to depict human virtue (Letters, pp. 279-83).
Noverre understood that the audience might be dazzled by virtuosic
dancing, but he believed that ballet could not achieve its full potential if
brilliant mechanical execution lacked warmth and feeling. He proposed
a reorientation in purpose: dancers needed to understand that ballet was
a composition of passions and that their souls spoke through their eyes
(Letters, p. 94). For this reason dancers should sacrifice complicated steps,
which detract from the natural. He advised them to learn the art of pan-
tomime in order to hide the effort of dance and to evoke the passions being
represented (Letters, p. 35).
Noverre assumed that facial expressions could be more expressive than
words because words, which have to be arranged in sequences over time,
can never have an immediate effect. Since the language of pantomime came
from the soul of the performer, it went directly to the heart of the viewer
(Letters, p. 140). He had watched Garrick as the best possible representative
of this technique. Noverre rejected the use of masks, because they destroyed
the true proportions between the head and the body, and prevented the
communication of sensitive facial expressions. He compared wearing masks
to giving a performance from behind an un-raised stage curtain (Letters,
pp. 20–4).
Noverre associated ballet with all the arts, but especially with painting,
either in detail of execution, or in overall effect: “A ballet is a painting,
the scene is the canvas, the mechanical movements of the dancers are the
colours, their facial expressions are the paintbrush. Thus the whole ensem-
ble, the vivacity of each scene, the musical voice, the decoration and the
costume comprise the subtle colorations – and finally the choreographer
is the painter” (Letters, p. 88). Of all the arts, he saw dance as the highest
expression because of its dynamic portrayal of a series of ideas or emotions
in time. However, he expressed regret that dance was ephemeral and that
time, energy and talent expended for a moment of beauty would disappear
so quickly (Letters, p. 324).
Noverre’s methodology for the training of dancers paved the way for
many future treatises, especially Carlo Blasis’s ground-breaking 1820 work
An Elementary Treatise upon the Theory and Practice of the Art of Dancing.
Noverre understood that human anatomy was paramount. In Letters 11 and
12 Noverre analysed why knowledge of the muscles, bones and nerves of
the body was essential to the dancing master. How could a teacher ask the
student to turn out at the knees if the knee was incapable of turning out?
It could only bend and straighten. Each student had particular difficulties
97 Jean-Georges Noverre: dance and reform

with the ballet technique and the teacher had therefore to take them into
consideration when giving corrections.
Noverre, never a modest man, announced that he single-handedly
intended to transform the entire art of dance. He spoke about pulling the
dance out of its lethargy and languor. He alone had the courage to struggle
against deeply engrained prejudices and habits. If something went wrong, he
quickly found someone else to blame: he was either successful or misunder-
stood and therefore less successful. But his conclusions in 1803 swept away
all uncertainties: “The glory of my art, my age and my numerous brilliant
successes, permit me to state that I have achieved a revolution in dancing, as
striking and as lasting as that achieved by Gluck in the realm of music. The
successes even by imitators today are the greatest testimony to the value of
the principles which I have laid down in my work” (Letters, p. 2).
9 The French Revolution and its spectacles
i n g e bax m a n n

Revolutionary festivities and cultural disruption


During the French Revolution of 1789 a deep change in the representation of
social communication and interaction took place. The spectrum of the new
sociability reached from refashioning the national costume that replaced
the old dress codes of the guild and social hierarchies to the republican,
informal way of addressing each other, from imitating Roman slave haircuts
to rituals of fraternisation, from frugal banquets in the open to name-giving
ceremonies of newborn babies (Brutus was one of the preferred choices),
from the revolutionary catechism that spread ancient Roman models of
virtue to the introduction of a new, republican calendar. All these measures
were intended to mark a break not only with the Ancien Régime before 1789
but with history as it had been known.
The French Revolution understood itself as a “regeneration”, a return to a
social order that was close to nature. A stable political order was the goal and
certainly not a permanent revolution. But the Revolution, particularly after
the execution of the king, had become a threatening crisis. To the horror of
many politicians, post-revolutionary France proved extremely mobile and
unsettled. The movement cultures and body performances of this time can
be seen as an attempt to contain the “too much” of social movement by
controlling the individual body within the mass. Hence the painter David
and the choreographer Gardel saw themselves confronted with the problem
of convincing passionate crowds to move within ordered forms, to make a
moving, yet manageable, organised, yet innocuous, collective body out of an
unruly mass that threatened to explode at any moment. The revolutionary
fête was to construct the citoyen, the citizen of the new republic. The move-
ment culture of the Revolution strictly rejected the court ballets of Louis
XVI, their “seductive culture” and elitist claims. In their place something
new was to be practised on and off stage. Inside and outside the theatre the
actors sang republican hymns together with the audience.
The Fête de la Fédération (Festival of Federation) began on 14 July
1790 when actors, musicians, singers and dancers from several Parisian
theatres gathered in Notre Dame cathedral and imitated the fall of the
Bastille, the signal that had begun the Revolution. The next day a pro-
cession marched through Paris; in the evening the people danced in front
[98]
99 The French Revolution and its spectacles

Figure 15 Festival of Federation, 14 July 1790; copper engraving by Paul Jakob Laminit, no date.

of the Bastille; the whole square had been converted into a dance space
(Fig. 15).
The revolutionary fêtes endeavoured to transfer enlightenment ideals
into political practice. The construction of a “natural body” with “natu-
ral movements” and as part of a “natural way of life” belonged to such an
endeavour. So-called “codes of naturalness” were to initiate the restructur-
ing of French civil society. Old, aristocratic norms were to be banished.
Instead, a new net of social responsibilities was to anchor “natural” human
sentiments as moral behaviour and make unequivocally clear the universal
and unchangeable truths of the Enlightenment and the Revolution. The
restructuring of movement culture therefore must be understood as part of
a fundamental change of social codes and moral sentiments.
The French Revolution put the people on stage. At court festivities the
people stood outside as observers, but revolutionary festivities invited the
people to join in. They could act out new roles in folk dances such as
the Carmagnole, a round dance, in “banquets fraternels” (fraternal ban-
quets), or in processions and orchestrated eruptions of emotions like, for
instance, the “cris d’allegresse” (lighthearted cries) or the “pas joyeux”
(jumps of joy). Everyday life during the French Revolution had to be aes-
theticised, because reality had not yet caught up with the ideal. New forms
of life had to be practised in the pedagogic sense of the word. The the-
atre and the revolutionary festivals were a means of instruction, aimed at
100 Inge Baxmann

forming a republican sensibility. On 30 September 1792 the Opéra in Paris


performed the Offrande à la Liberté, a lyrical divertissement or interlude that
reproduced the events of the Revolution as an opera. It was a tremendous
success. The music had been composed by François-Joseph Gossec and the
scenario had been written by Pierre Gardel. The Marseillaise, the revolution-
ary national anthem, was sung on stage. Groups of women, children and
soldiers marched towards a “Temple of Freedom”, where they worshipped
the goddess of freedom, danced in her honour and promised to use their
weapons in her defence. The festivities expressed an “image dynamique de
rassemblement” (dynamic image of gatherings) in dances, ritual activities
and collective gestures.

The new movement ideal: the “natural body” opposed to


the “artful body”
Dances like the Carmagnole deliberately used the ecstatic energies of the
masses. In the Carmagnole the dancers formed enormous circles. While
singing they slowly turned, stamped their feet and accelerated their move-
ments during the refrain. Pierre Gardel’s choreography for La Réunion du
10 âout (The reunion of 10 August), performed on 5 April 1794 also con-
tained market scenes and village people. This was a complete novelty. In
addition, Gardel integrated everyday movements into the dances and let
the people appear in their real clothes.1 Contemporary commentators felt
threatened by such popular dances and accused them of being ecstatically
disordered and chaotic. Such rejection is easily explained. An audience in
the 1790s could easily recall the display of disciplined bodies and aristo-
cratic movement styles of court ballet. Dances like the Carmagnole were
provocations, particularly if they were put on stage. They created a counter-
model to court style: ecstatic chaos versus geometric movements. The court
culture of the Ancien Régime was based on a particular power model that
relied on perfect interaction of the performance of political representation
and forms of sociability, together expressed in the ideal of the “honnête
homme”. This ideal incorporated the concept of the “artful body”, which
demanded constant practice; the nuanced gestures, movements and com-
plex step sequences had to be inscribed on the body. The “natural grace-
fulness” of the aristocracy had to appear effortless. Only by perfect staging
could the aristocratic elite advance its claim to leadership. The fencing and
riding or dancing exercises that occupied the elite of the Ancien Régime were
thus much more than simply a pleasant way to pass time. For this reason, the
court ballets of Louis XIV have been described as “seductive”.2 A collective
sensibility emerged in the process of forging the “honnête” ideal of the body.
101 The French Revolution and its spectacles

Only someone who could master the techniques of the body would have a
chance to survive the power struggles within the elite. Manuals and rituals
of court behaviour as well as court dances were directed towards one goal:
control of the affects and sensibilities that guaranteed promotion at court.
Gracefulness in dance and aptness on the battlefield formed this collec-
tive body. The danse noble and the “artful body” presented an image of noble
perfection. The ballets drew symmetrical, linear and elegant lines around
an invisible central axis, the symbolic body of the king. The “artful body”
expressed a refined sociability that justified a claim to leadership. The nation
was symbolically represented through the body of the king and the artfully
arranged bodies of his courtiers.
The enlightened counter-model established the “natural body” as an
expression of universal human feelings. It was free of masquerade, affec-
tation, disguise or learned rules of behaviour. “Nature” versus “artifice”
became a fundamental dichotomy of enlightened thinking. Within the
enlightened model “natural” gestures and “natural” bodies opposed the
“artificial” movement style of the Ancien Régime. Long before the French
Revolution, these gradual transitions had prepared the emergence of new
codes and cultures in all the arts from hierarchical, guild and class-orientated
social distinctions to a broader communication among social groups. In the
theatre the melodrama, the “drame bourgeois”, and in the fine arts the
discovery of antiquity in neoclassical paintings marked stages in the tran-
sition. Artists such as Jacques-Louis David or the genre painter Greuze, or
in dance the ballet d’action of Jean-Georges Noverre point to the evolu-
tion of a new model of social interaction and communication.3 The ballet
d’action expressed “sentiments”. The goal of dance, according to Noverre,
was to express natural feelings through movement and touch the soul of
the observer. Instead of the court ballet with its elegant movement patterns
around a central spatial axis, the audience now could follow the performance
of natural human passions and conflicts.
Thus elements of the new revolutionary art had emerged before 1789.
After the outbreak of the Revolution, Maximilien and Pierre Gardel devel-
oped a movement aesthetic and a choreographical concept derived from
Noverre’s dance theory and Diderot’s theory of drama. Gardel used them
for the revolutionary festivals and the choreographies at the opera. In both
genres movement emerged as an expression of natural human feelings. He
adopted the ballet d’action as a formal model to organise dramatic conflict
and eliminated elaborate movement techniques. Gesture and style came
close to mime. Costumes changed: corsets, heavy wigs and weighty brocade
disappeared and were replaced by lighter costumes. Dancers wore tunic-
like frocks made of light material and sandals or slim shoes, designed after
Roman models. Another model with revolutionary connotation was the
102 Inge Baxmann

everyday-dress worn by the Sansculottes, the radical, working-class Jacobins.


Their distinctive costume, the pantalon (long trousers) replaced the culotte
(silk breeches) worn by the upper classes. New clothes made new movements
in both senses of the word visible. Higher jumps, dancing on demi-point
and new dramatic expression appeared on stage as did the equivalent figures
represented in the new dress.

Antiquity as utopian past


It proved much simpler to abolish all the Ancien Régime’s signs of social
distinction in dress, codes of behaviour and communication styles, than to
put something new in its place. The term “nature”, from the beginning of
the Enlightenment, served as a collective concept for very different ideas of
change. The concept indicated that “return to nature” was a call to arms
against the old regime. It pointed to revolutionary social utopias. After 1789
all this could be traced in the self-representations of the new order, from
collective symbols to republican festivities, from movement cultures to dress
and everyday rituals.
The “civil republic” needed new norms of behaviour that rested on uni-
versal moral values. Since the Enlightenment had turned antiquity into a
utopian past, the revolutionaries hoped to draw new patterns from it. The
social changes constructed by the Enlightenment were a “regeneration”, a
return to an earlier stage of social equality and freedom. Jacques-Louis David
as a painter and Pierre Gardel as a choreographer had both been represen-
tatives of the neoclassical style even before the French Revolution; both had
therefore the same aesthetic ideal. Pierre Gardel’s choreographies, though
following the neoclassical tradition embraced by the new order, were com-
patible with certain aspects of the Ancien Régime, which in its own way
had a vested interest in classicism. Court ballets, after all, also acted out
antique myths and stories. Gardel’s ballets were called Le Jugement de Paris
or Psyche or Télémaque, whereas his revolutionary productions (including
those repeated on the theatrical stage) combined neoclassical style with
political content, as for instance in his L’Offrande à Liberté. If we want to
look for something new in Gardel’s ballets, we must look not merely at the
integration of classical antiquity into ballets.
The Enlightenment had generally interpreted antiquity in a special way.
Johann Joachim Winckelmann’s History of Ancient Art had been translated
into French in 1766. In it he described Greek antiquity as model for a demo-
cratic society. Based on the surviving artefacts, he pictured a culture that
encouraged human development. Antiquity as unifying cultural model, typ-
ified in neoclassicism, seemed to offer a consensus for a new cultural policy
103 The French Revolution and its spectacles

that the elite of the Revolution needed. From an enlightened perspective,


antiquity realised the unity of human nature and society. Ancient heroes of
the Roman republic could be converted into incarnations of republican state
virtues. But this involved a fundamental shift in the perception of antiquity:
it was to serve not as cultural legitimisation of royal claims to power, but
instead as a model for an egalitarian society.
Jacques-Louis David in his Oath of the Horatii of 1785 had already inter-
preted antiquity with enlightened concepts as a backdrop. This picture,
more than any other, moulded the self-conception and self-representation
of the French revolutionaries. In a celebration organised by David and Robe-
spierre, participants re-enacted the oath. The picture functioned as a kind of
emblematic summary of imperturbable loyalty to the new state, a reference
which contemporaries at the time understood without any knowledge of
the complex history of ancient Roman reality. The emotional fracture of
the family is another important theme in the painting, and with it gender
dualities and gender ethics. The scene stems, as mentioned, from Roman
history; it showed the battle between Rome and Alba, which represented
different, rival concepts of the state. The Horatii swore to fight the Curatii
even though they were related by marriage. The moral strength of the men
(the father who holds the sword and the three sons who swear the oath)
manifested itself in gesture and body stance. The firm gaze, the energetic
way of holding their heads, the tense leg muscles and the outstretched arms
mark the men in the centre of the picture, while the group of women and
children sit on the right-hand side. Helpless and resigned, the mother of the
Horatii and her two daughters Sabina and Camilla have their eyes half closed
and pose in a weak, effete way. Within such republican display of values and
ideals femininity was reduced to an object and mirror of manly virtue and
will to power. The painting was re-enacted in revolutionary celebrations as
well as on stage.
The constitution of 1792 demanded that every active citizen swear the
“serment civique” (the civil oath) to the constitution and its core idea, the
rights of man. The oaths were also part of festive rituals; men would swear
that they were willing to defend the republic whereas women swore that
they would only marry a republican.

Fraternal space and the natural order of society


Revolutionary notions regarding space are reflected in many contempo-
rary debates; there was a great need to define the new order spatially
and at the same time justify it as based on natural laws. Hence the pre-
ferred topographical forms were circles, globes and spheres, all considered
104 Inge Baxmann

natural, harmonious and tested in ancient times. These spatial constellations


would ensure the opening up of a previously restricted space. The open-
ing up of space, making it public, signified the attempt to abolish social
barriers.
The sphere represented an egalitarian model of power; in one such model,
a design for a tombstone honouring Isaac Newton, Etienne-Louis Boullée
argued in 1784: “Tout les points de son surface sont également distans de
son centre . . . Il [le corps sphérique] réunit l’exacte symétrie, la régularité
la plus parfaite, la variété la plus grande; il a le plus grand développement,
sa forme est la plus simple, sa figure est dessinée par le contours le plus
agréable”4 (All points of its surface are equally distant from its centre . . . It
[the spherical body] combines the most exact symmetry, the most perfect
regularity with the greatest variety; it is the greatest development, its shape is
most simple, its form is designed according to the most agreeable contours).
Hence it was no coincidence that the Abbé Sieyès used just such an image
to describe the social contract in his treatise What is the third estate?: “Je me
figure la loi au centre d’un globe immense; tout les citoyens sans exception
sont à la même distance sur la circonférance et n’y occupant que des place
égales; tout dependent également de la loi, tous lui offrent leur liberté et leur
propriété à protéger; et c’est ce que j’appelle les droit communs de citoyens,
par où ils se ressemblent tous” (I picture the law as being in the centre of
a huge globe; all citizens, without exception, stand equidistant from it on
the surface and occupy equal positions there; all are equally dependent on
the law, all present it with their liberty and their property to be protected;
and this is what I call the common rights of citizens, the rights in respect of
which they are all alike).5
This image expressed the problem of planning social relationships among
equals without introducing anything that could be associated with anarchy,
violence or chaos. Rounded shapes when applied to demarcate a public
space, like the amphitheatre, made possible the presence of the faceless
masses with their tendency towards chaos as citizens, individuals in a non-
hierarchical order, but held in check by the architecture of republican space.
In such “fraternal spaces” the citizen could envisage himself in a newly uni-
fied society that was based on natural principles and he could explain his
relationship to other individuals as one within the “grande famille humaine”
(the great human family). The view was to be open so the gaze could roam
freely; the ideal example would show a society that seamlessly merged into
external nature – landscape, mountains, forests, rivers – and at the same
time reflected it. The representation of social order as an analogy with
the harmonic order of the cosmos would displace history and make rules
unequivocally universal and eternal. Conversely, the eternal laws of nature
would certify everlasting social principles. Hence circular spatial forms
105 The French Revolution and its spectacles

confirmed a natural, cosmic order, which would channel anxieties about


the future in the right direction.
Contemporary iconography tells us that the caesura created by the Rev-
olution meant a new beginning for humankind but one mixed with fear
of the future, of violence and anarchy. The revolutionaries had crossed an
important boundary: they had abolished the necessity of hierarchy and had
suggested equality. But that raised the question of how it would be possible
to maintain order if, at the same time, those in power questioned hierarchi-
cal order and eradicated difference. Hence revolutionary visions of equality
themselves came dangerously close to chaos and disorder because they could
only be achieved through violent change. This deep-seated anxiety found its
representation through the paradigm of femininity. The Contrast, a litho-
graph, shows “British Liberty” as a seated woman wearing a Bonnet Rouge
and classical robes, in her left hand the Magna Carta and in her right hand
scales. At her feet sits a lion, in the background we see a ship, the symbol
of trade. Opposite her, on the torso of a decapitated figure, stands “French
Liberty”, in a threatening posture and with her face distorted by aggression.
In her left hand she holds a spear which has pierced through a decapitated
head, dripping with blood, in her right hand – ready to fight – a sword. Her
face and her arms look mannish; on her Gorgon’s head snakes grow instead
of hair and the belt around her torn frock is also a snake. In the background
hanged bodies, wearing aristocratic garments, dangle from the gallows and
clouds of smoke rise. The concept of equality appears as a contradiction of
natural order and as a perversion of known gender characteristics. This ver-
sion of the Revolution was not only that of foreign counter-revolutionaries
but also resonated in the collective anxieties of those in France scared of
“confusion”. The republic was founded on an act of violence. From the
point of view of the revolutionaries it was justified, yet they too feared fur-
ther acts of violence. This fear has to be recalled if we wish to understand
the pedantic attention to detail in staging revolutionary celebrations. In all
directives on celebration the emphasis lay on the danger of “confusion”. If
participants did not keep to their assigned places or social roles the worst
had to be feared. The “bad” citizen and the counter-revolutionary would
immediately take advantage of such slippage. Movement patterns in the
new public spaces, in pageants and ritual festivities followed the model of
“natural” order that relied on “natural” differences and necessitated clearly
demarcated spaces and movement orders. The inherent conflict between
the ecstatic, “chaotic” moment of general fraternisation, which every cele-
bration demanded, and the urge to represent the Revolution as structured
and ordered was not easily reconciled. The tension between mobility and
order proved to be fundamental for revolutionary politics. As one of the
politicians pointed out: “Il nous a fallu être révolutionnaire pour fonder la
106 Inge Baxmann

Révolution, mais pour la conserver il faut cesser de l’être”6 (We have had to
be revolutionary to establish the Revolution, but to conserve it, we have to
cease to be revolutionary).
Gardel directed several choric dance dramas that interpreted episodes
from the revolution in a patriotic spirit, for instance, La Réunion du Dix
Août at the Paris Opéra, where the Fête de la Régénération (the festival of
regeneration) of 10 August 1793 had to be contained within a theatrical and
orderly dramatic action. The phases of the celebrations were copied onto the
stage; the Revolution appeared as liberation of the poor and innocent from
slavery. One of the last festivities was called the Fête de la Raison (festival
of reason), held on 10 November 1793; it coerced the not very cooperative
singers and dancers of the Opéra into participating.7

The contemporary aesthetic debate in the fine arts, particularly in painting,


and dramatic theatre established the cultural codes that the organisers of
revolutionary festivities needed. In their presentations and performances
they searched for the typical and constant in human nature, for instance,
for the expression of passion in all sorts of situations. The organisers focused
on the universal and the unchangeable in thought, feeling and taste. The
goal of aesthetic representation could not be found in the empirical, the
real example, but in the idealised form of beautiful nature. The festivals
had a double task; they were supposed to form the interior while represent-
ing the ideal exterior. A successful celebration could prove itself as a useful
contribution towards a republican education. The aesthetic conception of
self-representation in revolutionary celebrations corresponded to contem-
porary knowledge of human perception. And that suggested that the soul
of the viewer or participant could be touched above all through the eye,
through images. Hence J. B. Gence conceived of the festival parade as a
“Tableau mouvant”. Every group within the procession was to express one
particular idea and therefore should be noticed as a separate, isolated event.
“It would be ideal neither to mix nor to disperse the artefacts, neither to
isolate nor to muddle ideas and perceptions of every group.” In addition it
would also be good if the movement of the pageant were to be “uniform and
rather slow so the eye has time to catch the story line and learn the lessons
and keep them well in mind”.8
During the Fête de l’Etre Suprême (Festival of the Supreme Being) in
Nancy, one wagon was decorated as “Char du Bon Ménage” (cart of the good
marriage); it showed a mother bent over the cot and a father teaching the
child on his knees to read. A third child embraced both parents and a fourth
crowned them with laurel. In this “tableau vivant de la morale et du patri-
otisme” the principles of Greuze’s moralistic painting as well as the literary
theory of the Enlightenment guided the conception. Greuze, as Diderot had
107 The French Revolution and its spectacles

in his theory of drama, considered the tableau an emphatic climax of a scene


that immediately became contained. The unifying principle of the tableau
made it possible to move away from history towards the anthropological
constants of human nature and to make visible the ethics of that nature. In
comments on the festive events one can often find comparisons with paint-
ings. One critic compared the Fête de la Fédération (1790) with a “tableau
d’ordonnance romaine”. The participants were mostly characterised accord-
ing to natural biology; they were separated by gender and age. Figures from
the ancient republican repertoire of state representation were added, such as
the “magistrat”. The simplification of social relationships through stylisation
in the tableau made social interaction direct, transparent and graspable as a
new social policy. “Familialism”, the principle of seeing society naturally as
a family, brought together the masses into manageable blocs. In enlightened
philosophy and state legal doctrine, the family was universally accepted as
a kind of Ur– or original – model of human interaction. The metaphor of
family hence could be used as the model for social relationships as such,
though their interpretation shifted during the time of the Revolution. In its
first phase, aristocrats and the wealthy remained part of the family if they
could be purified through education. A characteristic feature of celebra-
tory representation excluded the acknowledgement of social antagonisms;
as taboos they were banned from the social family. The harmonious order
of society included a strict spatial division among its individuals according
to their “nature”: men and women occupied separate spaces. The Fête de
l’Etre Suprême, organised by David and Gardel on the Champ de la Réunion,
choreographed the bloc of marching men and the elderly descending from
the right of the hill whereas women, mothers and young girls were grouped
on the left hand side. “Battalions of the Young Men” encircled the hills
and the “national representatives” were placed on the top of the hill. Each
group wore banderoles on their sleeves with mottos that summarised the
ideals they embodied. Their spatial position emphasised their distinct social
place and divided the groups from one another9 (Fig. 16.) Commissioners,
responsible for keeping order, oversaw the pageants so that there would be
no danger of “confusion”. “Each occupies his post. Each holds a green twig,
flower or corn ear in his hand. There were as many commissioners as groups
to control movement. The bad citizens and the immoral are recognisable in
that they try to introduce disarray and turmoil into this honourable event.”10
The costumes and props also denoted roles: women wore white frocks
cut to match antique patterns and braided hair (associated with the ancient
republics and chastity), whereas the men and boys carried swords as a symbol
of readiness to fight.
Not only were the modes of behaviour and movements of the specific
variants of human nature applied to individual groups of participants in
108 Inge Baxmann

Figure 16 The Festival of the Supreme Being, 8 June 1794 at the Champs de Mars in Paris, painting
by (Pierre) Antoine Demachy, no date.

the festivals but also the different codes of expressions of emotion found in
painting and theatre were used (for example from Charles Le Brun’s Méthode
pour apprendre à dessiner les passions of 1702). All these expressions were
distinguished by age and gender. Tears of deep emotion, “cris d’allegresse”
(shrieks of joy) as well “pas joyeux” belonged to the repertoire of ges-
tures and movements, which made externally visible the appropriately felt
emotions.
The French Revolution imposed a clear differentiation of male and
female and turned them into decisive patterns of order. Gender roles were
pushed to the top of the agenda because they could be used to combat the
“crisis of difference”, a crisis that was concerned with the invalidation of
traditional hierarchies and the implicit fear of loss of order and ensuing
chaos.

A new-found equilibrium
Movement threatened the unstable balance between nature and history.
Revolutionary celebrations confirmed exactly those values that constructed
their own social utopias and new behaviour had to be practised in order to
introduce a new, acceptable French reality. “Equilibrium” instead of mobility
was advocated as the future social ideal. One hymn that was sung in the
celebrations asserted: “May a free people emerge out of your midst. May
you love the new people, love one another – that is the happy pillar of your
equilibrium.”11
109 The French Revolution and its spectacles

Because the Revolution destroyed the old equilibrium, organisers of rev-


olutionary festivities attempted to create a new one. Movements, minutely
arranged and defined in gestures and grand rituals, “performed” social
order. Through repetition, order was to inscribe itself in the physical and
psychological body. Revolutionary spectacle turned from nature to order.
When the egalitarian-democratic myth of state self-representation broke
down during the 9th Thermidor (27 July 1794) and Robespierre and the
Jacobin leadership were sent to the guillotine, society after the “Great Terror”
needed to find a new stability. Many of the old conflicts and antagonisms
emerged. The harmonious vision of society was unmasked as a fiction. Rev-
olutionary celebrations degenerated into military parades and mass enter-
tainments. During the Directory (1795–9) a clear division of state and society
was proclaimed and the absolute ruler as symbol of representation reintro-
duced. The established forms and scenarios of state representation that the
earlier periods of the French Revolution had developed were still used but
gradually stripped of their utopian potential. The people, the masses, were
once again observers at the spectacles.

Translation by Marion Kant


pa rt i i i

Romantic ballet: ballet is a woman


10 Romantic ballet in France: 1830–1850
s a r a h dav i e s c o rd ova

Premiere of Giselle ou les Wilis


Monday 2 8 June 18 4 1
Ballet-Pantomime in 2 acts at the Académie Royale de Musique

The first chords of Rossini’s music announcing the beginning of the third act
of the opera Moı̈se can be heard faintly by the spectators who, uninterested in
the 1827 opera chestnut, still wander about the Opéra’s foyer as its prominent
clock strikes eight. They have at least another hour before the curtain rises to
reveal the long-awaited and much talked about new ballet, Giselle. Others –
men, dressed in black tailcoats and top hats, asserting their privileged status
derived from their associations with the worlds of business, of politics and
the intelligentsia – have paid for the entitlement of entering the foyer de la
danse to chat with the dancers. Aglow with the excitement of opening night,
the dancers warm up and mark out their steps before the foyer’s full-length
mirrors. Or, in their dressing-rooms, they look to their make-up, jewels,
last-minute fitting of their villagers’ costumes; many lay out their calf-length
white dresses of layered gauze and crowns of flowers for the ballet’s second
act peopled with wilis, those enchanting dancing ghosts of maidens who,
having died before their wedding day, lure each passing man into a dance
that only ceases with his death from exhaustion and their diurnal fade-out.
As is customary at the Paris Opéra, the evening will last three to four hours,
almost until midnight.
Despite recurring postponements of the premiere due to the leading
dancer Carlotta Grisi’s slow recovery from her accident in May, then due to
the tumour François Antoine Habeneck, the conductor, was fighting, and to
safety issues facing the machinists and stage hands, frequent press releases
succeeded in keeping interest in this momentous production alive. Since
the Opéra could not bank on a memorable midsummer night, the on-stage
dress rehearsal on Saturday 26 June had been closed to the press and on
Sunday, further rehearsals had kept dancers, stage hands and musicians at
the Opéra for most of the day in order to ensure Monday’s commercial and
artistic success.
After the performance, before setting off home or going on to the Café
anglais, or to Tortoni for an ice-cream, bedazzled, and some tear-stained,
[113] patrons of the Opéra linger in the foyer, under the covered arches, on the
114 Sarah Davies Cordova

steps, where they await their coachman, or a hackney carriage. They chatter
enthusiastically about the new ballet.
— Why did Giselle die? From a broken heart? Why did Hilarion denounce
Loys and reveal him to be duke Albert?
— Didn’t you read the libretto? Weren’t you paying attention? You’ll have
to wait for the newspaper reviews. They always recount the story in great
detail.
— Giselle’s mother, Berthe was so concerned! And Myrtha, queen of the
wilis, what determination! Why do the wilis not kill Albert?
Punctuating discussions of the plot, exclamatory praise for Ciceri’s decor
abounds, especially about the contrasts between the first act in a sun-
drenched autumnal German wine-producing valley and the second on the
grassy bank of a northern European pond in a dark forest with its vaporous
moon-like lighting, and the wilis’ shadowy figures with diaphanous wings
before the final sunrise.
“Carlotta Grisi” falls from everyone’s lips. The ballerina from Italy con-
joins in her incarnation of Giselle both the everyday young girl and the
otherworldly evanescent feminine figure. Her character’s tragic transfor-
mation from disobedient, self-centred dancing villager madly in love with
Loys into generous, self-sacrificing, yet still insubordinate wili caught the
collective imagination. Her exuberant yet lilting execution of the waltzes
in the first act contrasted with her ensuing dance of death, which in
turn was echoed by Loys/Albert’s exhaustion as the wilis slipped back
into the tufts of flowers and tall grasses at daybreak. Impudent and dar-
ing, Grisi remained delicate and chaste all the while standing up for the
love of her life and her passion for dance. She performed the mime and
movement needs of the role with precision and clarity without over-
emphasising them. The hybridity of her performance combined pan-
tomime to portray the stereotypical naiveté of the peasant and her inten-
sive single-mindedness as a wili with sharp, precise seemingly effortless
pointe work and flowing port-de-bras arms and torso for the demanding
adagio work during the demure yet defiant second-act solos and pas de deux
(see Fig. 17).
Jules Perrot’s emotive choreography and attention to Giselle and Albert’s
partnering pleased many. Their duets delineated both their separate des-
tinies and the clear demarcation between male and female ballet vocabu-
laries. Lucien Petipa, whose pantomime matched Grisi’s with its graceful
despair and touching passion, partnered her expressively and garnered his
share of praise for his role as duc Albert, initially disguised as Loys, her
fellow peasant. Appealing to the spectators’ melodramatic verve, the final
guilt-free reintegration of Albert into his own social milieu also played out
the recognisable romantic hero’s plot.
115 Romantic ballet in France: 1830–1850

Figure 17 Carlotta Grisi as Giselle, Paris 1841 from Les Beautés de l’opéra, ou Chefs-d’oeuvre
lyriques, illustrés par les premiers artistes de Paris et de Londres sous la direction de Giraldon . . . (Paris:
Soulié, 1845).

The second act lingered on in the audience’s mind’s eye since the
wilis, shimmering in white gauze during their nightly outing, recalled the
ballet des nonnes of the convent scene in Robert le diable (1831) with the
white-clad dancing lapsed nuns, the woods in La Sylphide (1831) peo-
pled with flying ethereal, translucent sylphs, and the underwater scene
in La Fille du Danube (1836) with its female water-sprites, the undines.
Giselle reiterated these successful antecedents with its troupe of alluring
wispy wilis, hauntingly illuminated by the ghostly, other-worldly effects
of the developing lighting technology.1 As the latest incarnation of these
two-act ballets, which would become synonymous with the “ballet blanc”
(the white ballet or the ballet in white) as one constitutive element of the
romantic ballet, the inspiration for this box-office success derived from the
German author Heinrich Heine’s De l’Allemagne, and Victor Hugo’s poem
Fantômes.
Wilis referred to those girls “who loved to dance too much”, and who, it
was imputed, died “on their wedding’s eve, at the height of unconsummated
sexual arousal”.2 This deathly notion associated with dancing too much
holds for much of the nineteenth century as can be seen in Edmond de
Goncourt’s 1885 novel Chérie. For some, loving to dance too much connoted
sexual promiscuity, while dying from dancing signified orgasm as in “la
petite mort”. The French poet, novelist and journalist Théophile Gautier, a
fervent admirer of Victor Hugo and partisan of romanticism, first worked
116 Sarah Davies Cordova

these sources into a ballet, which the prolific librettist Jules-Henri Vernoy
de Saint-Georges then translated into a ballet libretto, one palatable to the
Opéra’s management. Indeed since the 1820s, popular male literati were
often called upon to write the libretti as against an earlier practice according
to which the dancing master (who would later be known as choreographer)
both devised the story and the dancing. Some, such as Filippo Taglioni,
however, took exception to the new practice. During the 1830s,3 he often
occupied both positions as he worked to ensure that his daughter Marie
Taglioni had the stylistic vehicle to sustain her supremacy on the European
stage.
The alterations Saint Georges brought to Gautier’s manuscript included
a major rewriting and resetting of the first act in a realistic yet seemingly
anachronistic village, as well as the determination of two key (melodra-
matic) elements in romantic stories, the marriage plot4 and the love tri-
angle,5 whereby Giselle is pursued by Hilarion and by Albert disguised
as Loys, who in turn is already engaged to Bathilde, the daughter of a
prince. Adolphe Adam, one of the regular composers of the Opéra, arranged
the expressive music over a two-month period, so that it followed ref-
erentially and mimetically the characters’ actions, gestures and mimed
(and therefore silent) dialogues. Jean Coralli, who admitted that Jules Per-
rot had contributed significantly to Giselle’s pas, was credited with the
overall choreography wherein each clearly defined character performed
steps and mime sequences which adhered quite literally to the scripted
plot.
The audience pouring out of the Salle Lepelletier of the Académie
royale de musique’s theatre, otherwise known as the Opéra, on 28 June
1841 experienced a very different performance from a comparable night
at the beginning of the twenty-first century at the Opéra Garnier, in Paris.
Priorities and expectations have altered considerably the shape, content,
form and melodrama of ballet-pantomimes or what have since been cat-
egorised as romantic ballets. Yet the theatre managements in both epochs
reach for similar goals. As is borne out by their respective media resources,
they seek to amaze and fascinate. Over the course of the nineteenth cen-
tury, posters became part of the urban landscape and the Paris Opéra
utilised the medium to its advantage for advance sales, while the news-
papers both announced and reviewed theatrical events as well as devel-
oped gossip columns. Special volumes printed on fine paper about the
theatrical world and its members, which often included revised versions or
reprints of press reviews and a series of illustrations – portraits of dancers,
actresses, and opera divas both in costume and in street-wear as well as
drawings of key scenes from the popular ballets and operas – furthered
117 Romantic ballet in France: 1830–1850

the theatre’s publicity campaigns and the individuals’ renown. Much of


what we can now understand about performances in the 1830s and 1840s
relies upon these descriptive sources and in some cases on rehearsal notes
(répétiteurs), which circulated abroad to such cities as Brussels, London,
St Petersburg, and among the provincial theatres where these ballets were
staged.6
Clearly, the evening’s entertainment aroused delight and appealed to the
spectators’ senses. As at most performances during the July Monarchy, a
cross-section of Parisians attended, including members of the government
who often cut short their political discussions in order not to miss the ballet.
The curtain went up somewhere between seven and eight and finally came
down well into the night since, as in the case of Giselle’s premiere, the design
of the programme invariably included a cross-genre double bill and several
intermissions. When the full-length work being performed was a ballet, the
evening generally opened with a one-act opera, or a favourite act from a full-
length opera, with or without a danced scene which lasted about an hour.
The reverse occurred when a full-length opera played. Such a programme
catered to all tastes and reminds twenty-first century audiences that the two
genres – ballet and opera – cohabitated closely under the same theatrical
roof, with ballet as at least the equal, if not the predominant, partner of
opera. Indeed dancers often performed in the course of a single evening
in both a ballet and an opera, in which the danced sections and especially
the title roles in such often repeated operas as La Muette de Portici (Auber,
1828), Le Dieu et la bayadère (Auber, 1830) and La Tentation (Halévy, 1832)
were as demanding as in the ballets (see Fig. 18).
Apart from the patrons with their own boxes who came and went as they
pleased during the evening and often only watched one act – they actually
complained when the main ballet act came too early and prevented them
from enjoying a leisurely dinner – most spectators witnessed two genres of
spectacles. Both required similar skills to decipher the codes at play in the
plots’ staging. Romantic ballet’s storylines incorporated complicated twists
with intricate mimed passages, which proved difficult to understand without
the help of the published libretti. These often translated the mimed sections
into verbalised dialogue, and during the July Monarchy, some members of
the press corps quipped that people read more than they watched at the
Opéra. Mime generally played as large a part as the dancing in the romantic
ballets for it advanced the characters’ emotional state and developed the
intricacies of the plot’s evolution. Its affective intensity was echoed by the
accompanying music, which was shaped by the libretto’s necessities and
expressed by analogy the location(s) and the period of the story. Together
they sentimentalised dramatically the story’s circumstances.
118 Sarah Davies Cordova

Figure 18 Mademoiselle Marquet in Le Dieu et la Bayadère, opera-ballet 1830.

Giselle perhaps best epitomises the ballet blanc of romantic ballet. Its
enduring popularity stems in part from the way it uses dance as a medium
for exploring the interconnectedness of human interaction. Indeed the
dancing seemed naturally induced: the logic of the plot itself called for
dancing to celebrate in Act I the harvest, and in Act II the arrival of the
new wili. Grisi as Giselle conjugated two character types and styles into
one, since she shifted from earth- to spirit-bound as she moved from
the first to the second act. “Christian ethereality” and “pagan earthliness”,
qualitative characteristics which Gautier applied to Grisi’s two antecedent
rivals, Marie Taglioni and Fanny Elssler respectively, fleshed out an oppo-
sition which highlighted the two ballerinas’ contrasting dancing styles.
Giselle combined the distinctive traits taken from their signature roles:
Taglioni as eponymous sylph in La Sylphide (1831) and Elssler as Florinde
in Le Diable boiteux (1836); or from the sylphide (Taglioni) and Effie
(Lise Noblet), the two opposite (aerial and earthly) characters in La Syl-
phide itself. Grisi embodied and rearranged the two to figure as Christian
119 Romantic ballet in France: 1830–1850

earthliness in Act I with her terre-à-terre dancing and as pagan ethereality


in Act II with her seemingly aerial pointe work and diaphanous fleeting-
ness. As both, she upheld a social role, which cohered with the ballerina’s
own life and created the balletic realisation of the nineteenth-century poetic
muse.
Giselle, like so many romantic ballets developed between 1830 and 1850,7
can be read allegorically. As the numerous exegeses divulge, and as its vari-
ous adaptations make manifest, the ambivalence of Giselle’s messages layers
its appeal.8 During its first run, its legibility depended upon its spatial and
temporal structures, which referred both to the artists involved in its cre-
ation and to their own socio-historical epoch: the July Monarchy. While
the fury of the wilis’ abhorrence of men contrasted with the gentility of
Giselle’s resolve to save Albert from Queen Myrtha’s revenge, Berthe’s moth-
erly protective determination prefigured her daughter’s forgiving gesture,
which reunited the earth-bound, non-dancing Bathilde with the divided
forsworn Loys/Albert. This array of single women – the cohort of wilis and
their queen, mother, daughter, fiancée – participated in the period’s dis-
course about women in the public and private spheres: their assigned roles,
their contestation about choices, the acknowledgement of their availability,
together with society’s indecision about the role of the Church within post-
revolutionary France.9 Giselle’s themes of impossible love and of love that
kills demonstrate that the union of the protagonists is unrealisable because
of the ideological codes at work during the period. As saviour, Giselle exem-
plified perfectability by enacting the feminine values of a patriarchal society;
as covert witch, she troubled the prevailing social mores. Sexual desire was
performed under the cover of modernity’s aesthetic. The ballet brilliantly
repressed the erotics of the body’s staging even as it participated in the
commodification of the object of desire: the dancer herself and romantic
ballet.

Romantic ballet changed the way Europeans, in particular, danced and the
way they looked at dance. The requirements of its technique ensconced the
separation of social and theatrical dance so characteristic now of Western
societies. The paradigm shift, which occurred as ballet abandoned its use
of spoken language, resulted in romantic ballet’s exclusive reliance on non-
verbal movements. Carefully coordinated musical illustration, iconic and
objective props such as a portrait or a royal accoutrement, occasional writ-
ten signposts indicating location, a character’s gravestone, and constructed
machine-dependent effects which created the illusion of flying, all comple-
mented the bodies’ pantomime and movement to convey ballet’s stories.
Even as it evoked the social within its own performance, romantic ballet set
120 Sarah Davies Cordova

Figure 19 Cut-out paper dolls, depicting Marie Taglioni and Fanny Elssler in contemporary
fashionable frocks, 1830s.

itself apart from the traditions of French courtly dancing and demarcated
social dancing from theatrical dance, the social event from the spectacle.
As the French romantic movement shifted from its aristocratic lean-
ings to address the social problems of the workings classes, ballet asserted,
through its disciplined and decorous corporeal expressiveness, the harmo-
nious disjunction of its modes of moving. It incorporated the danse d’école
as formalised by the dancing masters serving at the Académie Royale de
Musique since its inception by Louis XIV in 1661, social dancing (from the
court and high society balls), hand and face pantomime and national dances
(folk dancing), which added local colour, often well known to the audience,
and enabled a distancing of the action from a French setting. Romantic
ballet’s choreographic richness emphasised the language of line, extension
and verticality and a belief in the universality of the language of dance
and mime. The radical new look of pointe work and adagio, of effortlessly
graceful feminine bodies dressed in tutus disguised the increasingly strenu-
ous technical virtuosity and extensive training for these performances. The
silent solos of the practitioners, or the pas de deux, pas de trois or pas de qua-
tre wove aesthetic and unified figures in space as their movements patterned
their interconnectedness. The responsive corps de ballet’s deployment over
the entire stage around the individual performances revealed, allegorically,
French notions of social hierarchy and power relations at home and abroad.
It underscored as well the hierarchical divisions of the company and ballet’s
gendering (see Figs. 19 and 20).
121 Romantic ballet in France: 1830–1850

Figure 20 Cut-out paper dolls, depicting Marie Taglioni and Fanny Elssler in contemporary
fashionable frocks, 1830s.

Star appeal also played its part in romantic ballet’s allure and accen-
tuated the commodification of the female performer, as the directors of
the Opéra sought to attract paying audiences by determining a ballerina’s
future with her Paris debut. Although many were foreigners, a good number
of the artistes were products of the Académie’s own school. Such depend-
able French dancers as the Dumilâtre and Noblet sisters, Pauline Leroux and
Pauline Montessu established their reputation through hard work and often
decried the unfair compensation which the foreign dancers like the Elssler
sisters, and later Fanny Cerrito, reaped. By the century’s second decade,
pointe work, gendered as a feminine technique, appeared ubiquitous for
the stars while masculine and feminine roles and styles no longer mirrored
each other and became sharply gendered. The specific male and female roles
undermined masculine dominance of the ballet stage. The ensuing intro-
duction of the female travesty dancer in lead roles as well as in the corps de
ballet produced a series of new role possibilities for the female dancers.10 This
trans-sexuality made ballet more and more feminine, an association which
reached its apogee during the second half of the nineteenth century. Despite
this feminisation of Western theatrical dance, the Opéra still employed a
large number of men to dance and teach, to administer the institution and
to hold the creative positions. The male stars, such as Mazilier, Barrez, Elie,
Simon, Coralli, Petipa, who increasingly supported their partners during
the pas de deux, tended to be known for their personal strengths – elevation
and beaten footwork, character roles, bravoura, pantomime. They served
122 Sarah Davies Cordova

long terms and could extend their careers as dancing masters – teaching and
in some cases choreographing – options which were not readily accessible
to women.11
By 1841, the Opéra, as state-sponsored or royal theatre or private enter-
prise, had accustomed its audiences to magnificent, stirring danced specta-
cles that served to sustain Paris’s reputation as artistic capital of Europe. The
fabulous shows evolved in dialogue with the current fashions and artistic
trends as well as changes in knowledge bases about scientific enquiry and the
new sciences, together with the codification of gender conventions which
surrounded the emergence of the bourgeoisie’s ideologies of gender and
sexuality. As the dancers’ bodies told stories against an ever more stunning
representation of other historical moments and locations, the plots, which
echoed their novelistic counterparts as well as the popular dramas of the
time, outlined private affairs of the heart.
The ballets’ titles, like their literary novelistic counterparts, often bore
the (first) name of the female lead character, like la Sylphide, Giselle, la Péri,
Paquita, while their subtitles, ‘ballet-pantomime’ or ‘ballet-féerie’, alluded
to the hybrid nature of their mimed and danced form. Ballets frequently
drew their inspiration from recognisable stories such as popular fairy tales
(Cinderella, 1823; Sleeping Beauty, 1829); comic opera successes (Nina ou la
folle par amour, 1823); or literary best-sellers (Manon Lescaut, 1830, adapted
from Abbé Prévost’s novel); La Sylphide, 1832 (based on Charles Nodier’s
short story Trilby ou le lutin d’Argail); La Tempête ou l’ı̂le des génies, 1834
(loosely inspired by Shakespeare’s Tempest); Le Diable amoureux, 1840 (with
key scenes from Cazotte’s text). These blueprints travelled back and forth
across the English Channel and circulated among the various theatrical gen-
res, even though critics frequently denounced the librettists who borrowed
effects and plot twists from the popular vaudevilles, opéras-comiques and
boulevard theatre blockbusters as in the case of Le Diable à quatre (1845),
and exerted pressure on the dancing masters to collaborate closely with their
literary peers and composers.
The productions tended to fall into one of four, albeit overlapping, cat-
egories. The successful nuns’ ballet in the opera Robert le diable determined
the emergence of the ballet blanc as one constitutive aspect of romantic
ballet. Those like Giselle, La Sylphide, La Fille du Danube generally incor-
porated an act which takes place in the world of the living and a second
one which draws a male hero into a supernatural world – that of the ballet
blanc – which he does not control. The second group of romantic ballets
coheres due to the presence of a Mephistophelian or devilish character who
allows the male hero to experiment with relationship choices, as in Le Diable
boiteux, 1836, Le Diable amoureux, 1840, or La Fille de marbre, 1847. The
third includes ballets like La Péri (1843) and Ozaı̈ (1847) that expend their
123 Romantic ballet in France: 1830–1850

energy in ‘exotic’ settings (the Orient, an island, the Americas) and whose
plots generally question French values by offering a political alternative,
where, for example, women govern or determine their life choices, as in La
Révolte au sérail (also known as La Révolte des femmes), 1833; Brézilia ou
la tribu des femmes, 1835; la Volière ou les oiseaux de Boccace, 1838; Nisida
ou les Amazones des Açores, 1848. The fourth grouping borrows most from
melodramas with plots, which tend to portray a young woman – orphaned,
exchanged and/or kidnapped, caught in another class or in a different cul-
ture’s caste system – who is eventually restored to her birth rights: La Gypsy,
1839; La Jolie Fille de Gand, 1842; Paquita, 1846.
In allying itself with popular fiction, romantic ballet drew upon the fan-
tastic, the exotic and the gothic. The ballets referenced the dislocation of
the Ancien Régime’s principles and the new revolutionary institutions. They
represented allegorically questions of nationalism and the erosion of the
sacred as seen through the lens of marriage troubles.12 The stories which
the ballets conveyed used the fantastic to mirror human violence, betrayal,
persecution, exclusion and marginalisation. In embracing a range of con-
tradictions, the dancers embodied the romantic generation’s afflictions. The
ballets adopted the literary topos of forgiving the wayward male character
who strays from the righteous path in order to take him back into the fold.
At the same time they found a place for the female protagonist which did
not infringe on male prerogatives and preserved her worth as an ideal. As
sylph, wili, undine or péri – all archetypal forms – she pushed her plight
to the limits of death or transformation. Whereas the ballets d’action used
traditional, re-creational archetypes such as Pygmalion sculpting stone into
living woman,13 the romantic ballets offered a new individualism with a pen-
chant for Narcissus’s story. Framed by the proscenium arch, which served
as a giant mirror, the spectators watched with fascination as the dancers
performed for them a biography: their character’s emotional turmoil and
impassioned identity quest.
Not all romantic ballets ended with the tragic death of the ethereal “other”
woman. In the modernising post-revolutionary French society, the female
characters danced through a kind of cultural manipulation. They showed
the fragility of their power in such a society, and exposed the male character’s
retrograde traditions.14 Indeed the female characters neither championed a
new order in gender relations nor in society, but rather seemed to indicate
instead that the dancers, as they skimmed the ground in glissades, pas de
bourrée, jetés and assemblés in a brilliant new technique, hinted at their
renewed subordination in the Napoleonic civil code. Additionally, the public
display of barely clothed female bodies created anxieties about women in
the public and domestic or private spheres of society and complicated the
social position of the female performers.
124 Sarah Davies Cordova

Figure 21 The Dream of a Ballerina. Léon F. Comerre, late nineteenth century.

As the house lights dimmed, spectators turned away from each other’s
appearance and looked to the stage where the performers figured as super-
natural, exotic or domestic female icons in fantasy scenarios and settings,
which evoked feminine private space. As an available yet unattainable char-
acter, repeatedly flitting just out of the male protagonist’s reach, the dancer
titillated the ogling males’ desires. As the object of the gaze, whether por-
traying a fantastic, Spanish, Oriental, Amerindian being, or one of African
descent, the female corporeal display created a voyeuristic sexualised and
racialised image. The Opéra’s directors used this impression and the dancers
themselves to lure male spectators into subscribing to the theatre’s season
and paying for direct access to the foyer de la danse. This practice was par-
ticularly associated with Doctor Véron, who agreed to run the Opéra as a
private enterprise for the first time in 1831. Thus the fashioning of the danced
events contributed to the commodification of the ballerina and erased the
fine line, which the artifice of the stage drew between desire and its actu-
alisation. Increasingly, over the course of the second half of the nineteenth
125 Romantic ballet in France: 1830–1850

century, the spectacle of the dancers encoded femininity to signify eroticism


and sexuality.
The commodification, to which the press reviews only added with their
segmentation of the female dancing body into arrow-like legs, gracefully
rounded arms, wasp-like waist and marble-white complexion, simultane-
ously translated her uncanny physical presence into a living incarnation of
modernity’s fleetingness. Whether otherworldly, supernatural, or down-to-
earth, dressed as woman or young man, the female ballet dancer conferred
upon romantic ballet an aura, which intrigued and yet masked and belied the
financial and physical hardships most dancers endured as rats de l’Opéra15 or
members of the corps de ballet. In spite of the phallocentric foundation and
patriarchal establishment of the Parisian Opéra, romantic ballet affected the
history of bodies. Affirming roles for female dancers that carried the spot-
light allowed women to envisage dancing as a career choice, and to perform
in a repertoire of narrative ballets that address questions about sexual and
gender identity, and socio-political configurations, as well as supporting
France’s national agenda to be Europe’s cultural leader. The mirror, which
such ballets as Giselle offered, reflected for some the narcissistic messages
they wished for; for others, romantic ballet partnered illusion with realism
to provoke the excitement of the subversive concept of a feminine art form
(see Fig. 21).
11 Deadly sylphs and decent mermaids: the women
in the Danish romantic world of
August Bournonville
anne middelboe christensen

Is the Sylphide just a dream? Does she only exist as a supernatural being,
split between dream and reality, in the thoughts of a romantic hero? Or is
she a woman of flirtatious flesh and boiling blood?
There she is with her fragile wings, sitting right next to James in his
armchair, shining in her pale whiteness and pointing a symbolic finger
under her chin. She is so close that he could feel her fairy breath if he only
woke up from his wedding nap in the armchair. When he actually does
open his longing eyes, she immediately begins her unspoiled and natural
dance of joy – only to vanish up the chimney without any warning the next
minute. . .
As the spectator watches the Sylphide flying around in her beloved Scot-
tish forest, she seems most of all to be an unreachable fairy-tale creature,
collecting butterflies and looking at birds nests, just for the fun of it. As she
loses her wings and her eyes go blind, with her hands creeping down her
trembling arms in catastrophic fear, she is no longer a supernatural creature.
At this moment, she is transformed into a mortal creature, conquered by
gravity, dying in front of the man whom she loves above anything else in
the world.
Silence. Tears. Curtain.

Nothing old-fashioned
Danish-French choreographer August Bournonville “borrowed” this dual-
istic story for his own La Sylphide from Filippo Taglioni’s ballet at the Paris
Opéra in 1832. Nevertheless, it is Bournonville’s version that has stayed in
the repertoire of romantic ballets – first danced in 1836 by the Royal Danish
Ballet at the Royal Theatre in Copenhagen and danced ever since by the same
company. Over the years, it has been staged by Bournonville-lovers such as
Valborg Borchsenius together with Harald Lander in 1939, Hans Brenaa in
1967, Henning Kronstam in 1988, Dinna Bjørn in 1997, and lately Nikolaj
Hübbe in 2003.
[126]
127 Deadly sylphs and decent mermaids

Figure 22 The Royal Danish Ballet, Season 2003/4. August Bournonville, La Sylphide. Gudrun
Bojesen, Thomas Lund.

Even though August Bournonville created around fifty different ballets


in his forty-seven years as ballet master between 1830 and 1877,1 La Sylphide
has survived as the ultimate favourite of both the audience and the dancers.
It is one of his very few ballets with a true tragic ending – and it is the only
one to deal so clearly with the bourgeois split of the new European citizen
in the 1830s: the split between the compulsory marriage of convenience and
the bohemian love of the heart (see Figs. 22 and 23).
The fascinating La Sylphide flew into the international ballet repertoire as
well. In 1953, Harald Lander staged it for Grand Ballet du Marquis de Cuevas,
in 1960 Elsa Marianne von Rosen reinterpretated it for the Scandinavian
Ballet in Sweden – to be followed by Peter Schaufuss’s dream version for
London Festival Ballet in 1979, Flemming Flindt’s interpretation for Dallas
Ballet in 1984, and Frank Andersen’s staging for Chinese National Ballet,
Inoue Ballet and the Royal Swedish Ballet in 1999.
In Copenhagen, the characters of the Sylphide and James are like real-life
figures, familiar to both the dancers and the spectators, very much in the
same way as Hamlet and Ophelia are to actors and theatre-goers. They are
considered a paradoxical and enticing mix: they become family members,
they turn into reflections of human desire, or they hide symbolic meaning
that performers and onlookers have to decipher.
“The Sylphide is innocent because she loves somebody”, Royal Danish
principal Rose Gad explained in 2002.2 “I didn’t know whether I was James
128 Anne Middelboe Christensen

Figure 23 The Royal Danish Ballet, Season 2003/4. August Bournonville, La Sylphide, rehearsal.

or Alexander – whether I was standing on stage or was out somewhere


flying”, principal Alexander Kølpin recalled from his first, wild experience
dancing James in 1987. As Nikolaj Hübbe, former principal with the Royal
Danish Ballet, since 1992 with New York City Ballet, added: “Some people
say: ‘La Sylphide is just something old.’ But a young man, burning from
affection and passion and love and lust, that is something archetypical.
James is pumped by adrenaline and testosterone and dream and ideals –
there is nothing old-fashioned about that.”
The dualism of Bournonville seems as real as it can get. Dancers of the
Royal Danish Ballet incessantly discuss the figures in rehearsal or in the
canteen. However, if an outsider asks if James really dies from the powerful
revenge of the witch Madge – or if he just falls to the ground in exhaustion –
the answer has a condescending tone to it: “He dies of a broken heart, of
course . . . ” No question, falling in love with a sylph is deadly.

Danish romanticism
What we know of Bournonville’s work fits perfectly into most of the exist-
ing definitions of romanticism. Let us use four central elements to analyse
romanticism in ballet: the supernaturalism, the diabolic, the exotic, and the
fantasies of birth right.3 All these elements can be found in Bournonville’s
ballets.
129 Deadly sylphs and decent mermaids

Supernaturalism is most clearly embodied in Bournonville’s flying


female creature in white, La Sylphide. It is typical that the ballet carries
the heroine’s name, not that of James. This already points to a focus on the
female, even though later interpretations tried to shift that focus more onto
the figure of James (for instance in the version by Peter Schaufuss).
The appearance of a diabolic male figure who uses his power to take
over an innocent woman also fascinated Bournonville. In The Kermesse in
Bruges from 1851, the diabolic appears in the three gifts for three brothers,
given by an old alchemist. Their magic makes the women lose their hearts
and almost their bodies, in that piece at least in an innocent way. Carelis
plays his magic fiddle so that Eleonore cannot stop her legs from dancing.
She, of course, has no control over herself and does not understand what
is happening to her body and why it reacts to his music in such a powerful
way. However, this diabolic power is initially used as a joke; when it is used
on his beloved Eleonore it seems a light-hearted jest – but later, that same
power actually saves them from the death penalty.
The fancy for exotic settings and folkoristic dances dominate
Bournonville’s works. Abdallah from 1855, still in today’s repertoire, takes
place in Basra in Iraq; it shows off veils and harem dancing and sheikhs
in violent conflicts. All of them are integrated into a story about Christian
values of generosity, forgiveness and love.
Finally, there is the romantic fantasy about exchanged babies – the
changelings who have been stolen and removed from their parents and
their environment but who end up being granted their legitimate birth
rights after all. This made-up dilemma is central to Bournonville’s A Folk
Tale from 1854 where the blond girl from the hills, Hilda, is discovered as
the true heiress to the manor, replacing the red-haired Birthe, Hilda’s troll
changeling, who is forced to go back to the tough troll life.
Whereas the productions in Copenhagen have carried on the heritage
from one generation of dancers to the next, the international stagings, which
lack the more general cultural as well as specific dance traditions, had to
invent the complete performance from scratch. If it were only for the steps,
the pas, this would not have been very difficult since the choreographies do
not contain advanced point work or complicated lifts. Nevertheless, foreign
dancers often seem surprised by the hidden difficulties of the Bournonville
steps that tend to look so easy.4 A particular “Bournonville style ” was
recognised and internationally accepted after the first Bournonville Festival
in 1979 and the scholarly discussions that took place at the same time. In
the ballets, the style is most clearly demonstrated in The Conservatory from
1849. He allows no bravura, the arms are kept low, and the calves beat the
air endlessly. At the same time the choreography never leaves a moment
for the dancer to breathe in any corner or take any kind of rest. By dancing
130 Anne Middelboe Christensen

backwards upstage, the dancer has to continue conquering the space without
breaking the soft flow of the movements – and without interrupting the
tilted Bournonville “épaulement” where eyes focus downward, shoulders
and head are directed towards the working foot – and has to question the
earth-bound balance of the body as far as possible without disturbing the
grace of the lines. Harmony is the ideal of every movement.

Subtle dualism
Bournonville’s portrayal of dualism – the real and the unreal – at the peak of
the romantic era in the 1830s and the 1840s should be considered very subtle.
Admittedly, the definition of a truly romantic oeuvre might only apply to La
Sylphide from 1836. Bournonville’s later, so-called romantic ballets always
seem to diverge from a strict romantic dualism. These ballets jump into a
handy, happy ending instead of exploring the many deaths of the heroines.
Nevertheless, their first acts often show the fascinations and temptations
which distract a romantic, male hero from pursuing his proper, bourgeois
plan of life. Bournonville simply never takes the romantic dualisms to their
extremes. Sooner or later the hero realises what he is actually about to risk
if he follows his improvidence and physical weakness – and his common
sense wins.
Vilhelm in Far from Denmark from 1860 is such a challenged romantic
male. Anchoring off the coast of South America, he has completely fallen
for Rosita, his sensual South American hostess, who also charms all the
other sailors on board his Danish frigate. The moment that he sees Rosita
throwing his engagement ring into the water, he reacts with his Bournonville,
bourgeois instincts: he throws himself into the water to rescue his golden
ring – and to save his engagement to the faithful blonde, back home in
Denmark (see Fig. 24).
Similarly, Edouard in The King’s Volunteers on Amager from 1871 proves
himself a reliable bourgeois male. At a Shrovetide party in a village outside
Copenhagen, Edouard realises that the masked beauty he has been flirting
with at the party is no one but his own wife. Embarrassed in front of every-
body, he begs her for forgiveness. The male craving for women is depicted
as something inevitable, just as the male apology is confirmed as a regularly
emerging monologue within every marriage.

Naked secrets
Bournonville himself knew those promiscuous, masculine feelings quite
well himself. As a young dancer at the Paris Opéra in the 1820s, he had
131 Deadly sylphs and decent mermaids

Figure 24 The Royal Danish Ballet, Season 2004/5. August Bournonville, Far from Denmark. Mads
Blangstrup, Marie-Pierre Greve.

a love affair with a French dancer, Louise Simon. This led to one of his
most well-kept secrets: in 1829, an illegitimate daughter, Louise Antoinette
Simon, was born.
At this time, Bournonville was already engaged to the Swedish Helena
Fredrika Håkonsson, whom he married the following year and to whom
he remained married for the rest of his life. In his marriage, he had four
daughters (born 1831, 1832, 1835 and 1840) and one son (born 1846) –
besides an orphaned teenage girl whom he adopted in 1851. Obviously, he
kept his child in Paris as a secret as its revelation would have created a serious
threat both to his marriage and to his position at the Royal Theatre. What
is less known, though, is that he continued to send money to the child and
even visited her in Paris in the guise of her “Nordic uncle”.5
Even though Bournonville did everything to support his reputation as a
respectable family father and husband, he fancied several of his young female
dancers. The most famous case is probably his affection for the young dancer
Lucile Grahn whom he personally taught and coached – and allowed to
mature – before her debut in 1834. His bad luck was that she did not return
his feelings and, increasingly angry, rejected his advice. She was his first
Sylphide but their relationship soured and became a problem impossible
for the dancer to resolve. In 1839, she left for Paris and fame. In 1845, she
danced herself in Jules Perrot’s legendary Pas de quatre in London with, or
more accurately, against Marie Taglioni, Carlotta Grisi and Fanny Cerrito –
thus becoming the only Danish dancer who managed to leap to international
fame before Toni Lander. As an example of a true, masculine seducer of his
132 Anne Middelboe Christensen

time, Bournonville had a paradoxical urge to improve the status of his female
dancers. He truly wanted to lift the reputation of the Danish female corps
de ballet from the French courtesan status to the position of respectable
citizens. Accordingly, his hope was to make it desirable for members of the
respectable middle class to send their young daughters to the ballet school
at the Royal Theatre because they believed in artistic ambition, not because
they were forced to do so out of poverty. He also fought for higher salaries
and social security for the dancers, and in 1869 founded a special pension
fund. According to Bournonville, the artists of the ballet should be granted
the same rights as all other respectable citizens of Copenhagen.
However much he believed in social equality, he enjoyed being compared
to his famous fellow artists of the time: the sculptor Bertel Thorvaldsen, the
composer J. P. E. Hartmann, the writer Adam Oehlenschläger and the story-
teller Hans Christian Andersen. All of them also shared his fascination with
the exotic. In spite of his professed egalitarianism he relished royal marks of
favour. In the four portraits that survive, he wears his medals with pride.6 On
the other hand he tended to take his relationship with the king for granted
and in an age of monarchical absolutism that could be risky. During a perfor-
mance of Toreadoren on 14 March 1841 a group of spectators began shouting
whenever Bournonville danced and tried to hiss him off the stage. From the
footlights, he had the impudence to turn to Christian VIII in his royal box
and to ask for advice. His Majesty was not amused, ordered Bournonville off
the stage and subsequently charged him with lese-majesty. He was punished
first with house arrest and then with a six-month exile abroad without pay.
The artistic result of this exile was the ballet Napoli of 1842.
In the so-called Danish Golden Age, women still had to face moral out-
rage if they wished to abandon their roles as housewives and mothers. The
urge to become a professional artist and not just to remain an educated
amateur had a high price. Female writers and painters were often guarded
by male professors who excluded them from respected artistic societies.
Actresses were among the first to achieve public respect. They were seen as
mediums, conveying poetry. Artists who used their own bodies, as ballet
dancers did, seemed very far away from bourgeois decency and acceptance
in Copenhagen, at least until Bournonville began his fight for “justice”.
He treated certain delicate matters with effective initiatives: for example,
he insisted upon having the skirts of the ballerinas sewn together into wide
pantalons, in order to prevent male spectators from getting more than what
was considered just a glimpse of skin.
Nakedness, even the possibility of just a glimpse of flesh, posed a seri-
ous threat to Christian Danish society. In the 1830s Danish Christianity
worshipped the soul as superior and condescended to the body as inferior.
The body on show and on stage was a constant, provocative reminder of its
133 Deadly sylphs and decent mermaids

inherent sexual connotations. The arts were considered that sphere where
the body constantly slipped through the closely guarded moral borders. In
Copenhagen, the Academy of Fine Arts employed its first nude, female model
in 1833; the fine arts were, contrary to ballet, after all, a very respectable art
form.7 The Danish Golden Age never really came to terms with the problem
of the body, its nudity and its sexuality. There remained an underlying ten-
sion and fear. It needed the emergence of a radical modernism to allow open
discussion of the place of women, marriage, freedom of thought, atheism,
urban life, industrialisation and so on. In Denmark, modernism made its
early mark around 1870; the literary critic Georg Brandes, one of the mod-
ernist protagonists, still attracts contemporary intellectuals with his appeal
for cultural radicalism.8

Sweaty words
Bournonville attacked the bourgeois approach to the body with words
instead – the safest way a cultivated soul in the Danish Golden Age could
chose. He sensed that the ordinary bourgeois theatre-goer in little Copen-
hagen would probably never understand the fleshiness and physicality of
muscular efforts and rehearsing in the ballet studios and the very un-
aesthetic drops of sweat, which such training produced. Words were neces-
sary and less offensive. Through his combative prose Bournonville made a
name for himself within artistic and social circles. As a twenty-three-year-
old in 1828 he made his debut as a writer with the booklet Nytaarsgave
for Dandseyndere (A New Year’s Gift for Dance Lovers). He soon gained an
audience with his ability to express his choreographic ideals. Among his
well-written books, the most popular have been his essayistic memoires My
Theatre Life from 1848.9 In his memoirs, Bournonville stated his “Chore-
ographic credo”10 which is still frequently quoted by his successors, both
teachers and dancers. Several sayings are especially popular, such as: “The
beautiful always retains the freshness of novelty, while the astonishing soon
grows tiresome.” The following is also: “The Art of Mime encompasses all
the feelings of the soul. The Dance, on the other hand, is essentially an
expression of joy, a desire to follow the rhythms of the music.”
Furthermore, his credo has achieved a natural position of greatness and
splendour in Bournonville discussions, mainly since the first Bournonville
Festival in Copenhagen in 1979 when the Royal Danish Ballet celebrated
the centenary of Bournonville’s death. It was the first time that the specific
Danish ballet tradition was seen by a wider circle of international ballet crit-
ics and historians. It was also around this time that his definition of “grace”
from the “Credo” became the central explanation of the Bournonville
134 Anne Middelboe Christensen

secret:
The Dance can, with the aid of music, rise to the heights of poetry. On the
other hand, through an excess of gymnastics, it can also degenerate into
buffoonery. So-called “difficult” feats can be executed by countless adepts,
but the appearance of ease is achieved only by the chosen few. The height of
the artistic skill is to know how to conceal the mechanical effort and strain
beneath harmonious calm.

Diving dualism
The most joyful and harmonious ballet by Bournonville is probably Napoli
from 1842. That sounds puzzling because in it romantic dualism between
the real world and the dangers of the supernatural world turn into a question
of life and death. The story is as follows: Gennaro, a fisherman, has been out
on his boat together with his fiancée, Teresina, but a storm has made them
fall overboard, and only Gennaro is saved. Teresina has disappeared. In his
search for Teresina in a scene that takes place in the Blue Grotto somewhere
in the sea outside Naples, Gennaro not only confronts the possible loss of
his girlfriend but also the nature of his own sexuality, aroused by tempting
mermaids. So does Teresina; she is saved by the king of the ocean, Golfo. In
his Blue Grotto, he conveniently transforms her into a naiad, a mermaid, and
strategically erases everything in her memory connected to her former life
and her human lover Gennaro. Though mermaids are usually lesser divine
pagan creatures, here a lesser mortal is granted the transformation in order to
make her part of the sea god’s submarine harem. The amnesia is supposed to
induce feelings of physical attraction. Golfo almost succeeds in his seduction.
We can thus justify calling Bournonville a romantic because he extends
the typical and subtle romantic dualism to the female character of Napoli
and not just to the male. Gennaro delays his search for Teresina because
he meets some irresistible naiads, glittering with seaweed in their long, wet
hair. Teresina too confronts her sexuality as she flirts with the muscular king
of the sea. At the last moment Gennaro disrupts the scene; he finally has
found the way into the secret cave – only to discover that Teresina does not
recognise him. Desperately, he points to the medallion of the Holy Madonna
and the miracle works: Teresina recalls her past and because she can now
remember her Christian faith she is restored to her old human form in her
old dress.11 Teresina is transformed back from a nautical creature into a
human woman in love with a man, just as earthly as herself. To the heart-
warming story are added the technical effects and stunts so typical of the
romantic theatre (see Fig. 25).
The tale of Teresina and Gennaro has received many interpretations;
fundamentally it is about a girl developing into a woman, rendered in the
135 Deadly sylphs and decent mermaids

Figure 25 The Royal Danish Ballet, Season 2004/5. August Bournonville, scene from Napoli,
Act iii.

way Marius Petipa and Pyotr Tchaikovsky pursued in their ballets some
fifty years later. The story about the teenage Aurora in The Sleeping Beauty,
who has to choose a man to marry, or the little girl Clara in The Nutcracker,
revolve around that same exciting and terrifying moment when they realise
that they are about to grow up. They dream about mice or cats, large,
enchanted animalistic forces fighting not only their noble princes but above
all enticing them into a sensuous world.
Teresina, the naughty mermaid and seductress, flirts more in her
thoughts than with her legs. Bournonville manages to provide her with
the necessary bourgeois strength: She does not give in to her instincts. Her
bodily desires are controlled and then defeated by her religious faith: her
body behaves itself and stays decent, thanks to the power of the Madonna.
Her proper reputation is saved; Teresina is in the end a well-behaved
mermaid.
This second act has challenged later choreographers. If they want to
convert their inspirations into new versions of the ballet, they usually
begin where Bournonville’s second act left off: with the unanswered and
unexplored dilemmas that occur when one dives into the subconscious.
Bournonville could not have gone any further without seriously endanger-
ing his own and his society’s sense of propriety, which demanded that a
story always end with an intact couple. Psychologically focused artists such
as Bournonville pioneers Elsa Marianne von Rosen and Allan Fridericia
in the 1970s went further and challenged the notion of intactness. Dinna
136 Anne Middelboe Christensen

Bjørn and Frank Andersen developed this idea even further in several of
their stagings of Napoli, particularly in their latest version of the ballet for
the Finnish National Ballet in 2005. More recently, modern dance forms of
Napoli interpretations by Tim Rushton in his psychological, abstract style
and by Thomas Lund and Johan Holten in their underwater video fusion12
show the abiding fascination of the work.
The Danish Golden Age more or less seriously flirting with the ideas
of romanticism is most interestingly portrayed in A Folk Tale from 1854.
For once, Bournonville chose to bring a Danish folk tale into a ballet and
focused on the superstition of peasants in the country. However, it is the
young nobleman Ove who feels the longing for something more than a
future in his manor house. He stays in the woods after a picnic with his
angry fiancée Birthe and there experiences the powers of the subterranean
creatures in the Danish summer night: A hill opens in front of his eyes and
trolls are sitting there underground, hammering away at their forges. In
their midst is Hilda, a beautiful, light creature, who carries a magic, golden
cup, and who looks like a princess with a secret. In the end the power of the
trolls is broken and Ove celebrates his wedding to Hilda by dancing their
special Wedding Waltz around the maypole in the light midsummer night.
That same waltz by Niels W. Gade is played at every Danish wedding even
today.
A Folk Tale takes Bournonville’s romanticism into a Danish reality where
dreams come true. As such it is the complete opposite of the nightmare of the
exotic supernaturalism of La Sylphide. Hilda proves that born virtue always
defeats supernatural forces. And that true happiness only can be achieved in
the real world. In a way, Hilda can be seen as a “happy end” transformation
of the Sylphide, or as the tamed bourgeois version of the same dream of
Bournonville.
Bournonville’s ballets and their dualistic figures have been saved over
time by the loyalty and enthusiasm of the dancers and the directors of
the Royal Danish Ballet. Since the first Bournonville Festival in 1979, ini-
tiated by ballet director Henning Kronstam and assistant ballet director
Kirsten Ralov, the Bournonville heritage has been treated as the interna-
tional treasure it truly is. For the third Bournonville Festival at the Royal
Theatre in 2005, just as for the second Bournonville Festival in 1992, the
still-existing Bournonville repertoire reappeared, consisting of La Sylphide
(1836), Napoli (1842), The Conservatory (1849), Kermesse in Bruges (1851),
A Folk Tale (1854), Abdallah (1855),13 La Ventana (1856), Far from Den-
mark (1860) and The King’s Volunteers on Amager (1871) – and the pas de
deux from The Flower Festival in Genzano (1858). Furthermore, the festival
programme offered the two cheerful pas de deuxs Polka Militaire (1842) and
Jockey Dance (from From Siberia to Moscow 1876).14
137 Deadly sylphs and decent mermaids

Lately, the company has had severe problems in maintaining its


Bournonville tradition. On the one hand there is the influence of other
dancing styles and on the other a massive intake of foreign dancers without
the specific Bournonville style in their bodies. The fact that the company
had five directors in eight years spanning 1994 to 2002, each with differing
approaches to Bournonville, made it hard to perform the repertoire.15 It
seems that the third Bournonville Festival in 2005 has inspired the Royal
Danish Ballet once again to take pride in the Bournonville tradition. It has
also proved that the ongoing survival of this tradition depends upon the
dancers’ enthusiasm about one work above all: La Sylphide, the ballet with
the love story that any dancer in the company would do almost anything to
dance. The survival of Bournonville’s ballet depends more on its peculiar
erotic and social meanings: however adorable the troll queen Hilda might
seem, it is the Sylphide who creates the ultimate attraction and who comes
closest to Bournonville’s concept of a truly romantic female, a dream cre-
ation of male imagination about feminine eroticism. As principal dancer
and director Nikolaj Hübbe says about his dedication to La Sylphide in Ulrik
Wivel’s film I Love You (2005):16 “I am part of that story.” Apparently, the
sylph is not just a dream.
12 The orchestra as translator: French
nineteenth-century ballet
marian e. smith

The Paris Opéra stood as perhaps the most influential house for ballet in
nineteenth-century Europe until the ascendance of the Imperial Russian
Theatres in the 1880s. Its stage attracted the talents of the great choreogra-
phers and dancers of Europe, and its audiences witnessed some of the most
important phenomena in nineteenth-century ballet: the rivalry of Vestris
and Duport in the first decade; the birth of the romantic style in the cloister
scene of Robert le diable in 1831; the controversial failure (accompanied by
the derisive whistles of the balletomaniac Jockey Club) of Richard Wagner’s
Tannhäuser in 1861. At the Opéra, audiences not only insisted that ballets be
included within opera, but exalted independent ballets, sometimes prais-
ing them more highly than the operas with which they shared the stage.1
The house’s directors, with their generous budgets and top-notch teams of
choreographers, composers, designers and librettists, oversaw the creation
of new ballets at a steady pace – including such staples of today’s repertoire
as the other-worldly Giselle (1841), the heroic Le Corsaire (1856) and the
comic Coppélia (1870) – and by popular demand exported many of them
to other houses in Milan, Copenhagen, Moscow and Philadelphia, to name
a few.
From century’s beginning to century’s end, custom at the Opéra dictated
that for each new ballet a new musical score be produced. These scores did
much more than provide background music; they helped constitute the bal-
let itself. Indeed, because they were tailor-made for each ballet, these scores
stand as the best eyewitnesses to the century of stage action in question
here. In this short chapter I shall offer an overview of this music, including
its subdivision into two distinct types – dance music (including specialised
music composed for “national” or “character” dance), and pantomime or
dramatic music – followed by sketches of some of the most important com-
posers of the century.

The ballet composer’s explanatory mission


Perhaps the most striking feature of ballet music in the early decades of the
[138] century, at least to the latter-day observer, is its frequent use of borrowed
139 French nineteenth-century ballet

music. Some ballets were taken over from comic operas – for instance,
Mozart’s Le nozze di Figaro – and these operas’ original aria melodies were
deployed in the corresponding scenes in the danced version. But even in
operas with newly contrived plots and characters, composers typically made
free use of pre-existing music, weaving it together with their own, not only
to please the ear but to help the audience follow the action. The critic Castil-
Blaze extolled both advantages as he bristled at the lack of borrowed music
in Alfred le grand (1822) and complained, a bit prematurely, of the decline
of musical borrowing by ballet composers:

Un ballet étoit un concert plein d’intérêt, où tous les genres de musique se
réunissoient pour plaire au moyen d’une séduisante variété . . . D’ailleurs
. . . les airs d’opéra, même, après avoir perdu leurs paroles, conservent une
expression mémorative bien précieuse pour expliquer les énigmes du
langage mimique, tandis qu’une musique neuve et sans originalité ne frappe
point assez l’imagination pour que son expression soit sentie.2

(A ballet was an interesting concert, where all the genres of music came
together to please the audience with seductive variety . . . Moreover . . . opera
airs, even after losing their words, preserved an expressive memory of them,
very precious for explaining the enigmas of the language of mimic. I would
not [underestimate] the power of known melodies, and of the clarity they
bring to the silent dialogues of pantomime.)

The relative novelty of wordless ballet in the early nineteenth century


helps explain the need for musical interpretations of these “enigmas”. In
1800, only twenty-odd years had passed since the advent of the Noverre-
style ballet d’action, the narrative ballet. But at the Opéra, a tradition-bound
house, such wordless narrative works, independent from the operas with
which they had long been intertwined, still struck some stodgy observers as
new-fangled as late as the 1830s and 1840s. Critics expressed in no uncer-
tain terms the high expectations they held of ballet composers, revealing
an assumption largely foreign to today’s ballet aesthetic: that music must
explain the story.

La musique des ballets a son caractère particulier; elle sera plus accentuée,
plus parlante, plus expressive que la musique d’opéra, car elle n’est pas
destinée seulement à accompagner et à rehausser les paroles du poète, mais
à être elle-même le poème tout entire.3

(Ballet music has a particular character; it is more accented, more parlante


[i.e. communicative], more expressive than opera music, because it is
destined not only to accompany and enhance the librettists’ words, but to be
the entire libretto itself.)
140 Marian E. Smith

En général, ce n’est pas de la musique qu’on demande à un compositeur de


ballet-pantomime; c’est un orchestre qui soit la traduction, le commentaire
du texte qu’on aurait pu ne pas saisir.4

(Generally, one does not ask for music from a ballet-pantomime composer,
but for an orchestra that is the translation, the commentary of the text,
which would not otherwise be understandable.)

si la musique n’exprime pas les sentimens des acteurs, qu’y a-t-il pour les
exprimer? Les grands mouvements de bras sont une pauvre langue.5

(if music doesn’t express the feelings of players, what else is there to express
them? Arm movements are a poor language.)

[La] musique . . . a mission d’expliquer ou de traduire [les scènes].6

(The music has a mission to explain or translate [the scenes].)

Le compositeur est presque chargé de raconter l’action.7

(The composer is virtually charged with telling the action.)

par leur caractère, par leur expression et par leur style, les mélodies peuvent
and doivent compléter la signification du geste et du jeu de la physionomie.8

(By their character, their expression and their style, the melodies [in ballet
music] can and must complete the meaning of the gestures and the play of
the physiognomy.)

Most of the devices composers employed to fulfil this mission might strike
us today as particularly explicit. One of these is the air parlant, used well
into the 1840s, an excerpt of texted music from opera or popular song,
sometimes only a phrase or less, which could make particular words pop into
the spectator’s mind.9 (In Jean-Madeleine Schneitzhoeffer’s La Tempête, for
example, the orchestra plays “Je sens mon coeur qui bat, qui bat” from the
moment in Grétry’s Richard Coeur de Lion when Lea begins to fall in love.)10
Music often provided the sound of the human voice too. In some cases,
a solo instrument played a recitative as performers on stage mimed. (A fine
example may be heard today in the trombone “voice” of a street entertainer
in August Bournonville’s Napoli, first produced in Copenhagen in 1842.) In
others, the rhythm of the music matched that of key phrases of text written
in the libretto (for instance, “écoutez-moi” (listen to me) in a key scene in
La Somnambule, 1827). In still others, composers used syncopation, dotted
rhythms, a wide tessitura or a phrase-ending uptilt to imitate the sound of
spoken French.11 The instrumental voice became more subtle by the 1840s
and 1850s but continued to be heard in ballet scores – for instance, the trilling
violins and flutes that give us the sound of Paquita’s rivals’ derisive laughter
141 French nineteenth-century ballet

in Le Diable boiteux (Gide, 1836); the syncopated clarinet and oboe melody
that allow us to hear Berthe’s admonition of the village girls to stop waltzing
in Giselle (Adam, 1841); the high woodwind solo at the beginning of Coppélia
(Delibes, 1870) for Swanilda’s opening mime monologue. Thus, even when
ballet music lost the blatantly vocal quality of the parlante music composed
earlier in the century, it carried forward the tradition of acknowledging
ballet characters’ implicit vocality, and helped to deepen the performers’
portrayals.
Recurring musical motifs evoking concepts, characters and emotional
states also appeared in several Parisian ballet scores, including Manon
Lescaut (Halévy, 1830), Giselle and Coppélia, taking their place among the
constellation of other devices that helped the audience follow the action,
and at the same time allowing the audience to hear for themselves the
transformation of characters and emotions as the plot unfolded. Perhaps
because Richard Wagner’s anti-Terpsichorian stance is well known, little
attention has been paid to the likely influence of Parisian ballet’s recurring
motifs upon the German master, who used the device in part to the same
didactic end. Moreover, unlike the operatic and non-theatrical precursors of
the Wagnerian leitmotif, recurring themes in French ballet were composed
in an atmosphere in which audiences had made plain their desire for the
intelligibility that could be furnished by the music.

Dramatic or pantomime music


It was largely in the dramatic or pantomime scenes, as opposed to the dance
scenes, that the musical cues mentioned above are to be found. Though
ballet composers were expected to write dance music too, and plenty of it,
the dramatic segments constituted a vital component of the ballet and could
take up a large portion of stage minutes – nearly half in some cases. They
shrank in number and length as the century wore on and danced segments
grew longer, yet remained a vital feature of ballet past 1900, for the simple
reason that character and story remained crucial elements in ballet.
The music required for dramatic scenes served a wide variety of func-
tions: it could express a character’s feelings or reflect personality or ethnicity,
set a mood, establish locale, and imitate ambient sounds such as a village
band, a human voice or a thunderstorm. Composers deftly switched from
type to type as the need arose. Consider, for example, the opening min-
utes of Giselle. At curtain-up, Adolphe Adam provides rustic music (with a
drone in the lower instruments) to help situate the scene in a tiny Silesian
village. The gamekeeper Hilarion soon appears, his furtive melody (played
by lower strings) reflecting his mission to uncover information about his
142 Marian E. Smith

rival Albrecht. As he mimes his love for Giselle and points towards her
cottage, the music becomes sweet and soft, but turns harsh for a moment as
he points at Albrecht’s dwelling. (These stage directions are indicated in the
score.) Shortly thereafter Albrecht arrives, in the midst of a heated argument
with his squire Wilfride, which is well brought out by tempestuous music.
The ins and outs of their conversation are deftly followed by Adam, who
provides warm, gentle music for Albrecht’s declaration of love for Giselle
and anxious, louder music for Wilfride’s objections to his master’s scheme to
wear a disguise and woo the innocent peasant girl. Finally Wilfride accedes
to his master’s wishes and leaves him alone, and as the scene ends peacefully
Adam matches Albrecht’s words of relief: “Il est parti” (he has gone). Shortly
thereafter Albrecht knocks on Giselle’s door – pizzicato strings allow us to
hear it – and Giselle makes her first appearance to the strains of a gay waltz,
which helps express both her national character and her love of dance.

Dance music
Music for dancing was composed in a style more familiar to us today than
that of the dramatic music, in part because of the enduring popularity of
dance numbers often excerpted from such Parisian scores as Adam’s Giselle
(1841) and Le Corsaire (1856)12 and Delibes’s Coppélia (1870), and from later
Russian scores carrying this style forward, including Ludwig Minkus’s Don
Quixote (1869) and La Bayadère (1877), Tchaikovsky’s Swan Lake (1877)
and Alexander Glazunov’s Raymonda (1898). Indeed, when one thinks of
nineteenth-century ballet music, one thinks of this style, and its familiar
features make it instantly recognisable as such. These include regularity of
phrasing and simplicity of melody, harmony and texture (two-voice textures
are common, and often, for solo dances, composers wrote simply a solo for
an obbligato melodic instrument and a pizzicato accompaniment), rep-
etition by the full orchestra of a melody first stated by solo instrument,
punctuation of sparsely textured passages with occasional tutti chords,
extended tutti cadences increasing in volume and sometimes quickening
in tempo towards the end, and a heavy reliance upon repetition, both of
melodies and of catchy rhythmic patterns. Often, dances are preceded by
an anticipatory-sounding introduction dwelling on the dominant, music
often originally choreographed for a couple engaging in a “shall we dance?”/
“no – well, yes” dialogue.

The divertissement
Dance music found its most concentrated use in the divertissements that
were featured in ballets and operas when the action was temporarily halted.
143 French nineteenth-century ballet

Such dancing was always carefully accounted for dramatically with such
rationales as village festivals, masked balls, entertainments for royalty, cele-
brations of weddings and military victories. Moreover – this point is impor-
tant for comprehending the popularity and accessibility of the ballet diver-
tissement – a strong affinity existed between ballroom and stage. First, the
overall musical structure of the ballroom dance set and the staged diver-
tissement was roughly the same: a long series of short numbers in varying
keys and metres, which could make for a pleasing variety of mood (a sweet
and sentimental adagio, for example, was likely to be followed by a brisk
and vigorous allegro). So too did the melody types, phrase structure and
harmonic language of the dance music for the stage match those of ball-
room music. It was also customary to end both bals and divertissements
with a rousing galop, a fast dance in 2/4 with a strong back beat, the musical
forerunner to the cancan.13 Composers liked to end the grand pas de deux
with a galop as well, a custom extended later in Russia in Don Quixote and
Swan Lake.
Not only music but choreography easily glided back and forth between
stage and ballroom. So receptive was the Paris Opéra to social dance fashion
that dances new to the city’s ballrooms in the 1830s and 1840s – for instance,
the polka and galop – were virtually guaranteed to appear on the Opéra’s
stage.14 By the same token, the Opéra’s ballet stars sometimes ignited social
dance crazes with their spectacular, crowd-pleasing performances of new
social dances. And staged ball scenes at the Opéra looked so familiar and
inviting that spectators occasionally tried to join in the dancing onstage, a
reaction that confirms the close connection between social and theatrical
dance. This familiarity was further fostered by the use of dance music from
ballets and operas by orchestra leaders in ballrooms, and the prevalence of
piano arrangements of such music for playing at home.

National or “character” dance


Among the favourite dances of stage and ballroom were so-called “national”
dances, or “character dances” as they are more commonly called today.
Though character dance had found a place on the Opéra’s stage since well
before the period under consideration here, Fanny Elssler’s famous cachucha
(danced in a scene from Le Diable boiteux, 1836) sparked an enthusiasm for
the genre that carried through for decades, and choreographers commonly
created both character dances appropriate to the setting of the work – like a
jig in Scotland [sic] – and “foreign” dances as well, usually with the rationale
that entertainers representing distant lands were dancing for the pleasure of
onlookers. (This rationale was highly believable, for such entertainments did
indeed take place as the revellers rested during real-life balls.) Composers,
144 Marian E. Smith

for their part, called upon a vocabulary of musical types, which like the
steps and gestures used by choreographers, and the props, costumes and
landscapes used by designers, were meant to convey an “authentic” sense of
place. Germany was often signified by waltzes, France by minuets, Spain, by
boleros, the Middle East by exotic-sounding minor-key oboe solos, Poland
by mazurkas and krakowiaks, Sicily by tarantellas, Hungary by the csárdás
and so forth. Occasionally, when generic “Others” from faraway lands were
to be depicted, composers turned to the triangle, bass drum and cymbals that
had signified Turks in the eighteenth-century West. Composers also made it
a point to write music generally appropriate to the geographical setting and
local character; Adolphe Adam, for example, was praised for demonstrating
the “grace, the suavity and the vaporous poetry of the Germanic deities
that inspired the composer”;15 Delibes’s Coppélia featured rustic village
music, mazurkas and a stirring csárdás, helping bring life to the Galician
(Austro-Hungarian) setting; Messager’s Deux pigeons imitated the gypsy
violin, befitting the presence of gypsy characters. For “character-variety”
divertissements, composers offered music to match each ethnic character as
they appeared one by one. Even the wilis in Giselle – a few of them at least –
had national identities, musically identifiable in their ball scene. National
dance remained a popular feature of dance scenes, both in opera and ballet,
and its continuing presence in the current repertoire (Swan Lake and The
Nutcracker, for instance), serves as a reminder of its living quality for much of
the nineteenth century, and of the crucial role that music played in bringing
it to life.

Critical comments in the nineteenth century


By what standards did nineteenth-century critics assess ballet music? In the
first few decades, as noted above, they often praised it for its sheer help-
fulness in making the plot understood. Throughout the century they com-
mended lively, graceful, varied, tasteful, inventive, danceable (“dansante”)
scores, expecting appropriateness to the setting and story. Melodies were to
be accessible, plentiful, pretty and fresh, and orchestration interesting and
colourful.
Occasionally, too, critics disclosed their assumptions about ballet music’s
place within the larger scheme of things. Most agreed it should be lighter
in tone, more transparent, less complex, less ponderous than opera, and
“in a more facile, more relaxed style”, as one critic put it in 1863.16 Any
score too weighty or too rich in intrinsic interest was liable to be deemed
operatic or symphonic.17 In 1844 one critic of Eucharis wrote: “[Delde-
vez] would be more at ease with an opera libretto than a ballet . . . his
145 French nineteenth-century ballet

talent is serious, elevated, instead of light and coquettish.”18 Another wrote


of Le Diable amoureux in 1840, “[Reber’s] talent is better fitted to the
symphony” – perhaps because it was, in the assessment of a more favourable
observer, worth hearing even with one’s back turned to the stage.19 Lightness
could be taken too far, however: “one mustn’t lose all sense of propriety, nor
forget that we are in the house where Guillaume Tell and Les Huguenots are
performed. It’s a good idea to keep a distance between the Paris Opéra and
the neighbourhood tavern” – a distance one critic believed was breached by
Cesare Pugni in Diavolina (1863).20
In the 1880s, the debate raging in France over Wagnerianism seeped into
critics’ assessments of ballet music’s proper tone. A reviewer of Widor’s La
Korrigane (1880), for example, found in ballet a synthesis of symphony and
drama – a quality ascribed by pro-Wagnerians to their hero:

[ballet music is] a sort of free symphony in which [the young masters of the
modern school] can let their imaginations roam freely. The melody follows
the movement of the ballerina and the expression of the gesture; one
thought is scarcely expressed before another one comes up to replace it.
This art is both improvised and carefully planned (soigné) at the same
time – a special art in which symphony merges with drama.21

This balance between the “improvised and soigné” struck one critic as lack-
ing in Edouard Lalo’s Namouna (1882). Though admired by the young
Debussy, among others, this score was thought to be too studied, and, more-
over, its harmonic language too tangled – qualities found in Wagner’s music
by his detractors. Lalo wrote good symphonies and operas, observed a critic,
but Namouna suffered from

le raffinement excessif de l’idée musicale et d’inutiles, sinon regrettables,


dissonances qui viennent troubler le cours et le mouvement d’une partition
de ballet nullement appelée à prêcher les dogmes de la musique dite
nouvelle. L’art de moduler à l’infini, avec plus ou moins de transition, a pris
en ces derniers temps des proportions intolérables, – a ce point que l’on ne
sait le plus souvent le ton dans lequel on se trouve. Par suite, des accords se
précipitent, s’entre-choquent d’une façon peu harmonieuse; qu’on nous
serve ce regal qui emport la bouche, puisque c’est la mode aujourd’hui,
mais du moins, comme disait Rossini, qu’on y mette un peu de sucre.22

(excessive refinement of musical ideas, and the useless – if not deplorable –


dissonances which disturb the flow and sparkle of a ballet score. Ballet
music is not at all appropriate for preaching the dogmas of the so-called
new music. The art of modulating infinitely, with or without benefit of
transition, has taken on intolerable proportions lately, to the point where
you often don’t even know what key you’re in. As a result, chords dash into
each other in a most inharmonious fashion. Serving up a feast so hot it takes
146 Marian E. Smith

the roof off your mouth — that’s the style these days. But at least, as Rossini
said, you can add a little sugar.)

The very notion of defining ballet music as a separate genre with partic-
ular characteristics – serious, “soigné”, improvisatory, light, “dansante”, or
otherwise – was discarded by the originators of “new ballet” in the early
twentieth century. As Michel Fokine famously proclaimed in 1914:
In contradistinction to the older ballet [the new ballet] does not demand
“ballet music” of the composer as an accompaniment to dancing; it accepts
music of every kind, provided only that it is good and expressive . . . It does
not impose any specific “ballet” conditions on the composer . . . but gives
complete liberty to [the composer’s] creative powers.23

Thus did the ballet world acquire the great commissioned scores of Stravin-
sky (including Le Sacre du printemps, 1913, composed with much liberty
indeed), Ravel, Prokofiev, Copland, Khachaturian, and a host of others, as
well as the habit of making choreographies to music never intended for
dance.
Yet, while it is true that ballet benefited when traditional restrictions
on composers were lifted, and an inexhaustible supply of non-dance music
opened up to choreography – overdue developments as the young moderns
of the early twentieth century saw it – some of the music composed for
the nineteenth-century Parisian ballet remains as vital a part of the dance
repertory today as ever. Aside from the continuing currency of such scores
as Giselle and Coppélia, without which these ballets would likely not have
survived into this century,others have been retrieved from the archives and
used to great effect with new choreographies, including Donizetti’s dance
music for Dom Sebastien (Balanchine’s Donizetti Variations), Verdi’s for Les
Vêpres siciliennes (MacMillan’s The Four Seasons),24 Delibes’s for La Source
(Balanchine’s short ballet of the same name, and a pas de deux by John
Cranko) and Ferdinand Hérold’s full-length La Fille mal gardée, brought to
Frederick Ashton’s attention by the dance historian Ivor Guest, arranged by
John Lanchbery and used in the Royal Ballet’s highly successful production
of that ballet.25 Indeed, the great store of French ballet music remaining
untouched in the archives could be profitably mined even more, for the
necessary qualities of danceability and narrative vigour originally invested
in the best of these scores remain as vibrant today as they were when the
music was composed.

Composers
By way of concluding this study, I offer sketches of Adolphe Adam and
Léo Delibes, the two most admired ballet composers in nineteenth-century
France, followed by brief discussions of a few others of note.
147 French nineteenth-century ballet

Adolphe Adam
As a youth, Adam paid the Opéra-Comique’s triangle player 40 sous for
the privilege of taking his place for one performance. This enthusiasm
for the theatre continued to manifest itself throughout his career, as he
became one of most successful composers in Parisian history of comic
opera and ballet, as well as founder of the Opéra-National in 1847 (forced
by the political disturbances of 1848 to close its doors), Professor of
Composition at the Conservatoire and writer of newspaper critiques as
lively and open-hearted as his music. Most of his fourteen ballets scores (ten
composed for Parisian premieres) were met with the unanimous appro-
bation of critics, and when La France musicale declared in 1847 that the
Opéra had “chosen the master of masters in this field to compose the
music for the new ballet”, there was no doubt about whom they meant.26
Adam’s ballets are notable for their danceability, apt characterisations,
atmosphere and moment-to-moment adherence to the action. A partic-
ularly masterful example is the Act I finale of Giselle (structured like a
grand opera finale), with its numerous characters, climactic confrontation,
rapid mood changes, and the mad scene featuring Giselle’s delusional rec-
ollections and death. Less known but equally effective is the opening of
La Jolie Fille de Gand, Act II, in which the title character wakes up in a
Venetian palace, takes a dancing lesson, meets her supercilious rival and
quarrels with her, and then sets forth for the ball after hearing the sounds
of carnival music. In both cases, Adam expertly interweaves strains of allur-
ing dance music with parlante and descriptive music, focusing the audi-
ence’s attention on the moods or actions of now this character’s and now
that one’s, seamlessly sweeping the audience through changes in mode,
tempo, metre, dynamics and instrumentation, making use of every gra-
dation of volume from silence to full orchestra at fortissimo. Adam was
also adept at keeping the audience’s attention on the stage instead of the
orchestra pit. For instance, he sometimes repeats a melody several times
in mime scenes, as silent-film pianists did many decades later, but with
small adjustments when needed, adapting to the particulars of the con-
versation and the moods of its interlocutors, as his annotations demon-
strate. Like today’s best film scores, Adam’s ballets were never intended in
their entirety for concert performance, but to serve as a vital partner in the
drama.
A key to Adam’s success was his sheer enjoyment of ballet composition;
he fondly recalled his collaboration on Giselle with Carlotta Grisi and Jules
Perrot “in my salon” and spiritedly rebutted critics’ suggestions that he save
his best ideas for opera: “Nothing pleases me more than the task for which
the inspiration comes from observing the feet of dancers . . . [Critics] scold
me for using the springtime of my life in the production of [ballets] but so
be it. This work is my muse and my life.”27
148 Marian E. Smith

Léo Delibes
Coppélia (1870), though Delibes’s first ballet commission as sole composer
(he shared compositional duties with Ludwig Minkus on La Source, 1866),
shows his rare talent and sure-handedness in depicting stage action and
character. Noël Goodwin has even averred that Delibes “could be called the
first impressionist composer, for he shared similar principles with his con-
temporaries in pictorial art: he made colour and rhythm the most important
elements in his compositions”.28 Indeed, Delibes’s command over these two
elements ensures the effectiveness of his characterisations; each major fig-
ure in this ballet (including the life-sized doll of the ballet’s title) is strongly
invested with musical character – Dr Coppelius’s rhythm, for instance, is
off-kilter and his sonorities thin and peculiar (foretelling Drosselmeyer in
Nutcracker); Franz’s music sounds eager and a bit impetuous. Of further note
is the charming music of the character-variety scene (in which Swanilda,
impersonating the inanimate Coppélia, comes to life, dances a bolero and
a jig). The finale, a dedication of the new village clock, wittily and sweetly
demonstrates Delibes’s wide range in affect and style, with its lovely Waltz of
the Hours and dances for each of the occasions for which the new clock will
chime: dawn, prayers, work, village weddings, war (is this a light-hearted
tribute to Beethoven’s “Battle” Symphony and “Rage Over a Lost Penny”?)
and peace.
Delibes demonstrates equal virtuosity in his only other full-length bal-
let, Sylvia (1876), though it is entirely different in spirit, since the scenario
(based on the sixteenth-century pastoral drama Aminta of Torquato Tasso)
aimed to be loftier and more sedate than the broadly comic Coppélia. The
score, with its rich and full orchestration “reveals the hand of a master sym-
phonist”, wrote one critic, finding that “the picturesque choice of themes, the
expressive variety of melodies, the attractive improvisation of harmonies,
and the highly-coloured orchestration make this ballet, to my mind, an
exquisite work, perhaps too refined and too delicate for the glare of the
footlights”.29
Others found the music too obscure (as did some of Carmen’s first lis-
teners), though today we would scarcely concur; its harmonic language
is indeed more complicated than that of earlier ballet scores in Paris,
and Diana’s nymphs sound downright Walkurian. Yet, Delibes resisted the
Germanic influence embraced by some of his contemporaries (for instance,
César Franck) and made his loyalties to the French spirit explicit:

Je professe pour Wagner une admiration sans borne. C’est un génie


admirable . . . Mais j’éstime qu’en musique, comme en tout autre art, chaque
nation doit conserver son génie personnel, que les musicians français
doivent conserver leur tempérament propre, en lieu de s’efforcer à des
149 French nineteenth-century ballet

imitations stériles. Pour ma part, je suis reconnaissant à Wagner des


émotions très vives que sa musique m’a fait ressenti, des enthousiasmes
qu’elle a soulevés en moi. Mais si, come auditeur, j’ai voué au maı̂tre
allemand une profound admiration, je me refuse, comme producteur, à
l’imiter.30

(My esteem for Wagner is boundless – he is a genius well worthy of


admiration . . . But I reckon that in music, as in all the other arts, each nation
must keep its own particular spirit; that French musicians must conserve
their own temperament instead of trying to make sterile imitations. For my
part, I am grateful to Wagner for the vivid emotions his music brought out
in me, and for the rapturous feelings it raised. But if as a listener I admire
the German master, I refuse, as a composer, to imitate him.)

Sylvia is widely considered the finest ballet score before Swan Lake (the first
version of which appeared only a year later); Tchaikovsky himself declared
Sylvia the better of the two. Upon his death, Delibes was deemed “incon-
testably without rival” as a ballet composer, and George Balanchine, in the
next century, called him “one of the three great musicians of the dance”,
along with Tchaikovsky and Stravinsky.31

Other composers
Among the Opéra’s many other notable musicians who produced ballet
scores during the nineteenth century are the following (NB work lists in this
section are not necessarily complete):
Pierre Gardel, the great ballet master whose career at the Opéra spanned
the years of the turn of the century, including the tumultuous years of the
Revolution. A fine violinist and dancer (who both danced the menuet and
played the violin in La Dansomanie, 1800), Gardel owned copies of Haydn’s
music, and is likely to have arranged, or helped arrange, ballet scores for
some of his choreographies. (Le Jugement de Paris, 1793, Une demi-heure de
Caprice, 1804, Le Retour des Lys, 1814).
Ferdinand Hérold, whose early death left unfulfilled his ambition to com-
pose grand opera. He composed and arranged witty and vibrant ballet scores
in the late 1820s (including La Fille mal gardée, 1828), and composed comic
operas so popular that his name, along with Mozart’s and Rossini’s, was
inscribed in gold on the curtain at La Scala.32 (La Somnambule, 1827, La
Belle au bois dormant, 1829.)
Fromental Halévy, a major composer of French opera whose sole bal-
let Manon Lescaut (1830) is admired for its careful incorporation of
150 Marian E. Smith

eighteenth-century melodies (including Rameau) to evoke the period set-


ting at the Opéra in the eighteenth century.
Jean-Madeleine Schneitzhoeffer, tympanist and composer of La Sylphide,
1832, striking for its evocations of Scotland in the divertissements, its
dreaminess and its use of “airs parlants”. This work remains a popular staple
of the Opéra repertoire today. (August Bournonville, reportedly because
of the high price of the Schneitzhoeffer score, commissioned music from
Løvenskjold for his production of this ballet, first performed in Denmark
in 1836 and still widely performed today.) (Le Séducteur du village, 1818,
Zémire et Azor, l824, Mars et Vénus, l826, collab. L’Orgie, 1831, La Tempête,
l834.)
Frederich Burgmüller, composer of over 100 piano pieces, as well as the
music for the “peasant pas de deux” in Giselle (including his waltz “Souvenirs
de Ratisbonne”), interpolated shortly before the premiere of that ballet. (La
Péri,1843, collab. Lady Henriette,1844.)
Edouardo Deldevez, violinist and conductor, and composer of Paquita,
1846, of which he recalled that Mazilier waited to choreograph the ballet
“until after he had heard the music and meditated for a long time upon it;
until he had profited from the scenic intentions of the musician”.33 (collab.
Lady Henriette, 1844, Eucharis, 1844, collab. Vert-Vert, 1851.)
André Messager, an ardent fan of Wagner, admired by Saint-Saëns and
Fauré, and composer of comic operas and of Les deux pigeons (1886, re-
choreographed by Frederick Ashton in 1961). (Fleur d’oranger, 1878, Les
Vins de France, Mignons et villains, 1879, all for the Folies-Bergère.)
Ambroise Thomas, best known for his operas Mignon (1866) and Hamlet
(1868) (collab. Le Gipsy, 1839, Betty, 1846), and composer of the ballet La
Tempête (1889).
Charles Marie Widor, the great organist at St-Sulpice in Paris for sixty-
four years, best known for his organ symphonies. His theatrical debut,
the ballet La Korrigane (1880) (deemed a “master stroke” by La Revue et
gazette musicale)34 includes a syncopated mazurka clog dance, a gigue in
the Breton style, and a number for humming chorus and typophone (a
keyboard percussion instrument invented by Victor Mustel, and forerun-
ner to the celesta made famous twelve years later by Tchaikovsky in The
Nutcracker).35
13 Russian ballet in the age of Petipa
ly n n g a r a f o l a

On 24 May 1847 Marius Petipa, a young French-born dancer and ballet


master, landed in St Petersburg.1 He was not the first dance artist to brave
the long journey to Russia and the rigours of a Russian winter, nor even
the only Petipa; only five months later, his own father signed a contract to
teach the senior classes at the Imperial Ballet School.2 Like so many other
danseurs, Petipa fils was drawn to the “Venice of the North” because of
decreasing opportunities for male dancers in the West and the unusually
generous terms of an imperial contract, in his case, 10,000 francs a year and
“half a benefit” for the position of premier danseur.3 He accepted the offer
with alacrity, little imagining that he would remain in Russia until his death
in 1910, marry twice there (both times to Russian ballerinas), raise a family
and rule the Imperial Ballet from 1869, when he became chief ballet master,
to his retirement in 1903.
Petipa’s long stewardship of the company had an incalculable effect on
Russian ballet. He presided over the shift from romanticism to what is usu-
ally termed ballet “classicism”, laid the foundation of the modern Russian
school by marrying the new Italian bravura technique to its more lyrical
French counterpart and helped transform an art dominated by foreign-
ers and identified with the West into a Russian national expression. Petipa
choreographed scores of ballets and innumerable dances, codifying their
structure while expanding the lexicon of their movements, and created sev-
eral generations of distinguished dancers. His works that survive, even in
much altered form, La Bayadère (1877), Giselle (which he revived and signif-
icantly revised in 1884), Swan Lake (which he choreographed in 1895 with
his assistant Lev Ivanov), Raymonda (1898), above all The Sleeping Beauty
(1890), are the classics of an art with a performance tradition that goes back
no further than the romantic period. Even if the designation of these works
as “classics” came in the twentieth century and reflected a desire to estab-
lish a canon in the face of modernism, the fact remains that both in Russia
and the West this canon was overwhelmingly identified with Petipa. Indeed,
what we call “Russian ballet” in terms of repertory and style is virtually
synonymous with Petipa, his colleagues and descendants.
The Imperial Ballet, renamed the Kirov Ballet during the Soviet period
and the Maryinsky Ballet after the fall of the Soviet Union, is one of the
[151] world’s oldest companies. It celebrated its 200th anniversary in 1983, but
152 Lynn Garafola

its origins date back to the 1730s and the court entertainments organised
by Empress Anna Ivanovna. Although Russia was rich in folk performance
traditions, ballet was an imported art nestled in the lap of the court. Ballet
masters came from abroad (usually from France and Italy) and brought
to Russia the dance fashions of the West. The first dance master of note
was Jean-Baptiste Landé, who petitioned the empress for permission to
establish a school to train dancers for court performances; the training was
to take three years. In 1738 the first ballet school in Russia opened its doors,
first in Landé’s house, then in one of the wings of the imperial palace. The
twenty-four pupils were children of palace servants.4 The imperial court
was generous to foreigners, especially artists, paying and treating them far
better than the Russians who worked for them and whom they sometimes
despised. In the decades following the establishment of Landé’s school,
numerous ballet masters made their way to St Petersburg. Sometimes they
brought dancers with them, but by the 1760s, when the ballet d’action arrived
in Russia with Franz Hilverding and Gaspero Angiolini, Petersburg-trained
dancers could hold their own with foreign performers.
By then, the Imperial Theatres had come into existence. This state system,
founded by decree of Catherine the Great in 1756, gave official status to the
imperial opera, ballet and drama troupes, as well as the theatre school,
and instituted pensions for artists.5 The system was fully subsidised, with
funding from the Ministry of the Court; it survived even the upheavals of the
1917 Russian Revolution. Suitably renamed, the former Imperial Theatres
were nationalised and supported in full by the new Soviet state. When the
socialist state collapsed in the 1990s, the post-Soviet government continued
to pay for them. The state system had become so deeply entrenched that it
outlasted any single form of government.
In 1783 the St Petersburg ballet moved into its first real home, the Bolshoi
(not to be confused with the Moscow theatre of the same name) or Kamenny
(“stone”) Theatre, a huge, neoclassical building seating 2,000, opposite the
site where the Maryinsky Theatre now stands. In 1885 the Imperial Ballet
moved across the street to the Maryinsky, which remains the company’s
home today. Thus it was at the Bolshoi Theatre that most of the com-
pany’s nineteenth-century history played out. Here, romantic sylphs took
flight and designers from abroad created marvels of baroque perspective or
shipwrecks on the high seas, as in Andreas Roller’s famous design for Le
Corsaire (1858).6 There were scene-painting shops, costume shops and the
finest stage machinery that money could buy. Dance personalities of the first
rank found their way to St Petersburg, spending, in some cases, years with
the company. Filippo Taglioni revived most of his ballets for the company,
including La Sylphide,7 which starred his daughter, Marie Taglioni, the most
celebrated of romantic ballerinas. Balletomanes, a term coined in Russia for
153 Russian ballet in the age of Petipa

fanatical enthusiasts of ballet, crowded her performances and even, it is said,


sipped champagne from her ballet slipper. “Taglioni is the synonym of Air!”
enthused the novelist Nikolay Gogol. “Nothing more ethereal had existed
heretofore on the stage.”8 Théophile Gautier, perhaps the foremost French
dance critic of the time, visited St Petersburg in the 1850s. In Voyage en
Russie (1858) he had nothing but praise for the Petersburg company: “The
Russians are great connoisseurs . . . and the dancer who has withstood the
marksmanship of their opera-glasses must be very confident of herself.”9
This was the Russia to which Petipa had come.
Marius Ivanovich Petipa, to give him his full Russian name, was born in
Marseilles in 1818. Like so many nineteenth-century dancers, he came from
a theatrical family. His father, Jean-Antoine Petipa, was a dancer and ballet
master, his mother an actress who gave up the stage to raise a family.10 His
brother Lucien danced the role of Albrecht in the first Giselle (1841) and
later became a ballet master at the Paris Opéra. Jean-Antoine was his son’s
first dance teacher. “At seven”, Marius wrote in his memoirs many years later,
“I started instruction in the art of dancing in the class of my father, who
broke many bows on my hands in order to acquaint me with the mysteries of
choreography.”11 He made his stage debut in Brussels at the age of nine in a
ballet staged by his father and as a student danced in several of Jean-Antoine’s
original works. He got his first job in Nantes where he choreographed his
first ballets, toured the United States with a small company organised by
his father,12 then settled in Paris to study with Auguste Vestris, one of the
great teachers of the French school and a strict custodian of the academic
tradition. Nevertheless, a place in the Paris sun eluded him. In 1847 he set
out for St Petersburg and, except for vacations and the occasional sabbatical,
never left.
During his first fifteen years in Russia, Petipa honed the craft that
would distinguish his mature productions. Jules Perrot, the greatest of the
romantic-era choreographers who spent more than a decade in St Peters-
burg, exerted a major influence over Petipa. Petipa danced in several of
Perrot’s ballets and also served as his assistant, at times even as his col-
laborator, rehearsing the “great pas stratégique” from Catarina and staging
Giselle (1850) for Carlotta Grisi’s Russian debut under the watchful eye of
his mentor. A second influence on Petipa during these formative years as
a choreographer was Arthur Saint-Léon, who followed Perrot to Russia in
1859 and held the post of ballet master for eleven successive seasons. A vio-
linist of the Paganini school and sometime composer, Saint-Léon was an
enormously musical choreographer; he had a facility for arranging dances,
a fertile imagination and a gift for showing his dancers at their best.
In 1862 Petipa produced his first great success, The Daughter of Pharaoh.
He had been working on it for nearly two years, commissioning a libretto
154 Lynn Garafola

Figure 26 Gala performance at Peterhof, 11 July 1851, from Geirot’s Opisanie Petergofa, 1868
(Carlotta Grisi and Jules Perrot in The Naiade and the Fisherman).

from Vernoy de Saint-Georges, who had written the “book” for Giselle and
other ballets and a score from Cesare Pugni, a prolific ballet composer
in the employ of the Imperial Theatres. A “grand ballet” in three acts and
nine scenes, with prologue and epilogue, The Daughter of Pharaoh was set in
Egypt. Gautier’s short story “The Romance of the Mummy” (1857) inspired
the plot, but the excitement generated by Egypt’s fabulous monuments and
the building of the Suez Canal gave the subject a topical interest.
The ballet was monumental in scale and a melodramatic tour de force.
It was also a huge success, and Petipa was promoted to the rank of second
ballet master. In 1869 came another success, Don Quixote, a ballet that
has enjoyed continuous theatrical life. That year, promoted to chief ballet
master, Petipa became sole master of Russia’s choreographic revels. In his
hands rested the fate of more than a half-dozen imperial enterprises: two
theatres (the Bolshoi and the Maryinsky), a company, a school and several
smaller court theatres (the Hermitage, Peterhof, Tsarskoe Selo, Krasnoe Selo
and Kamennyi Ostrov) (see Fig. 26).13
Like the romantic ballets that fed his imagination, Petipa’s works told
stories. Many were romances touched with old-fashioned melodrama, with
heroines who died of broken hearts or suffered in some way before wedding
their rightful mates. Like their romantic predecessors, Petipa’s ballets were
always about women; more often than not, they took their titles from the
heroine – hence, The Daughter of Pharaoh (1862), La Bayadère (1877), The
Sleeping Beauty (1890), Raymonda (1898) and many others. Petipa revelled
155 Russian ballet in the age of Petipa

in his ballerinas, and in work after work, generation after generation, he dis-
played their growing prowess as technicians and their personalities as artists.
Although the romantic ballet had already identified the ballerina with
the feminine mystique, emphasising her elusiveness and ethereality, Petipa
added technical brilliance to the formula. In the post-romantic era, ballet
technique developed rapidly, above all in Italy. By the 1860s the blocked shoe
had come into use with its greater support; dancers began to run, jump,
hop and perform multiple turns on pointe. Skirts rose, revealing greater
expanses of leg and the prowess of “steel” toes. Not everyone welcomed the
new virtuosity. To many the modern ballerina seemed akin to a machine, a
miracle of industrial precision, nowhere more so than in the multiple fouetté
turns introduced into Russia by visiting Italian ballerinas in the late 1880s
and 1890s.
The emphasis on drill, as essential to the handling of Petipa’s stage masses
as to the manipulation of armies of “ballet girls” in the spectacular ballets
popular in Western Europe and the United States, was another aspect of
this mechanisation. Standing backstage at the Théâtre de la Gaı̂té in 1873,
an American observer compared the production of an opéra féerie, with
its spectacle effects and army of dancers, to “the manoeuvring of the great
machine, which night after night worked so smoothly and so beautifully
before the public”.14 Finally, unlike the sylphs and other romantic emblems
of eternal girlhood, the ballerina of the closing decades of the nineteenth
century was a full-bodied woman, wasp-waisted, corseted and glamorous,
fully cognisant, when she stepped out on stage, of her star power. Utterly
different from the chaste, “Christian” image of ballerinas like Marie Taglioni,
she dominated the stage by her presence, not by the premonition of her
absence. With flesh, sexuality and power as her weapons, she was all too
real, anything but a metaphor for the spirit.
Petipa jettisoned other romantic conventions as well. Most romantic
ballets (or “ballet-pantomimes”, as they were usually called) consisted of
two acts and were typically performed on a bill with opera. Petipa’s ballets,
by contrast, were usually three or four acts, full-evening entertainments.
His ballets may have told stories like their romantic predecessors, but they
embedded those stories in a vast array of dances and transformation scenes
that enhanced the spectacular aspect of his work, while undermining its
narrative power. Bigger meant better, so revising meant adding rather than
trimming material. Instead of offering a simple contrast with the preceding
realistic act, as in La Sylphide or Giselle, Petipa’s ballet blanc was a jewel in
an ever more elaborate setting.
In 1874 August Bournonville visited St Petersburg and saw several
of Petipa’s ballets, including The Daughter of Pharaoh and Don Quixote.
Although he admired the “richly imaginative arrangement of the settings
156 Lynn Garafola

and transformations”, the corps of more than two hundred and the “superb
talent that displayed itself especially among the female members”, the Danish
ballet master was also shocked by what he saw:

I sought in vain to discover plot, dramatic interest, logical consistency, or


anything which might remotely resemble sanity. And even if I were
fortunate enough to come upon a trace of it in Petipa’s Don Quixote, the
impression was immediately effaced by an unending and monotonous host
of feats of bravura, all of which were rewarded with salvos of applause and
curtain calls.15

Bournonville also found appalling the “lascivious tendency” that pervaded


“the whole ballet movement”, except for the Slavic national dances. He hated
the women’s “excessively short skirts” and the “regulation caleçons de bain”,
swimming trunks, that were “an established fact for the danseurs”.16
However, at this very moment, Petipa was creating what a later generation
would regard as a defining moment in the genesis of Russian classicism –
the “Kingdom of the Shades” in La Bayadère. Produced in 1877 to a score
by Ludwig Minkus, the ballet marked Petipa’s maturity as a choreographer,
with the “Shades scene” regarded as both a masterpiece of classical style and
one of his greatest dances. Here, in a story bathed in the romantic exoticism
of The Daughter of Pharaoh, was Petipa’s vision of classical heaven – forty-
eight moonlit women in white descending one by one from the Himalayas
and winding slowly forward, all the while performing a simple arabesque
phrase with the hypnotic deliberateness of ritual.

Although the Imperial Ballet was largely insulated from outside events, the
1880s were a time of change. In 1881 Prince Ivan Vsevolozhsky became
the new director of the Imperial Theatres. In contrast to his predeces-
sors, Vsevolozhsky was cultured, well-educated and polished, a European
in speech and manners.17 Petipa adored him and the seventeen years that
followed his appointment were Petipa’s happiest and most productive. As
conceived by Vsevolozhsky, the Maryinsky Theatre, where the Imperial
Ballet began to perform in 1885, was to be the temple of an art outside
time, unsullied by ugly nationalism, untouched by change except in “safe”
or ideologically neutral matters such as staging, technique and personnel.
Vsevolozhsky was above all a courtier and under his rule the Imperial Ballet
may have reached its apogee. But it did so at the price of cultural isola-
tion; at no time did the company find itself so divorced from the society
around it or from the artistic trends – nationalism, realism, symbolism –
that transformed the Russian cultural landscape between the 1870s and early
1900s. Rather than a fully national art, ballet in nineteenth-century Russia
remained an appendage of the court.
157 Russian ballet in the age of Petipa

In 1882 the monopoly of the Imperial Theatres was abolished, and for
the first time ballet audiences in cities like St Petersburg and Moscow could
see companies and entertainments from abroad. People flocked to the new
establishments, although critics disagreed over the merits of the offerings,
which included operettas, variety programmes and ballets-féeries. In marked
contrast to the “grand ballet” as it had developed in Russia, the ballet-féerie
depended on effects so spectacular as to overwhelm both the choreography
and the story; at the same time it introduced the public to a new generation
of virtuoso Italian ballerinas. However much Vsevolozhsky might disdain
the new trends, he could hardly deny their appeal, and in several works of
the mid- and late 1880s, he borrowed judiciously and selectively from the
new genre.
Gradually, his vision of the Imperial Ballet took shape, which, as Roland
John Wiley suggests, united “the dance-intensive grand ballet long favoured
in Petersburg and the extraordinary mise-en-scène of the Franco-Italian
féerie . . . Vsevolozhsky wanted to match the West European fashion for
grandiose staging, then better it with elegant choreography and sophisti-
cated music, which the West European model lacked.”18 He abolished the
post of official ballet composer, thus partly abandoning the “specialist”
tradition of nineteenth-century ballet music and sought “to develop new
sources of ballet scores, principally from Russian composers”.19
It was Vsevolozhsky who initiated the collaboration between Petipa and
Tchaikovsky that became The Sleeping Beauty. “I have thought of writing a
libretto based on Perrault’s story ‘La Belle au Bois Dormant’,” he wrote to
the composer in 1888. “I want to do the mise-en-scène in the style of Louis
XIV. Here one can let one’s musical fantasy play – and create melodies in
the spirit of Lully, Bach, Rameau and so on. If the idea is to your liking,
why couldn’t you undertake the composition of the music?”20 Tchaikovsky
allowed himself to be tempted. “The subject is so poetic”, he wrote to his
patron Nadezhda von Meck, “so gratifying for music, that I was captivated
by it.”21 Two years later the ballet was a reality. Petipa was fully aware of
the importance of the collaboration. For the first time he was working with
a major composer from outside the ballet “specialist” tradition, probably
Russia’s most distinguished musical artist. But Petipa did not significantly
alter his usual working method. He wrote out a detailed plan for the ballet,
breaking it down by section and for each section, describing the action,
the quality of the music he wanted, the time signature, length (in terms of
bars) and even suggestions for orchestration.22 Tchaikovsky, for his part,
responded with inventiveness to Petipa’s call for expressive effects and almost
always complied with his instructions regarding metre, tempo or scoring.
The result was a score that followed the conventions of nineteenth-century
ballet, even as it invested them with unusually high artistic quality.
158 Lynn Garafola

The Sleeping Beauty premiered in 1890 and despite initial criticism of


its weak libretto, féerie-style lavishness and Gallic source and setting, was
soon hailed as a masterpiece.23 The chief reason, wrote Alexandre Benois,
was the music, which inspired Petipa to “a height of perfection hitherto
unsurpassed by him. It is enough to recall the variations of the fairies in
the prologue, the grand pas de deux in the third scene and, the greatest
masterpiece of all, the dance of the Blue Bird and the enchanted Princess. But
what innumerable other gems of choreographic art are scattered by Petipa
throughout.”24
A grand spectacle, four hours long, with a prologue, three acts and an
apotheosis, Beauty was a summation of the conventions elaborated over the
course of the century. Like all of Petipa’s ballets, it drew on a multiplicity
of movement idioms. There were long mime scenes, dances in national
dance style, ballroom dances, as well as classical dancing in the vision scene,
divertissements and a grand pas d’action.
It is impossible today even to imagine the human density on the imperial
stage of the late nineteenth century. In 1903–4 the Imperial Ballet consisted
of 122 female and 92 male dancers, plus ballet masters and régisseurs.25
In addition, there were the children and advanced students of the ballet
school, regularly used in productions. Dozens of them appeared in The
Sleeping Beauty, as Cupids, pages, “young girls” with lutes, Tom Thumb and
his brothers; in the Garland Waltz alone there were twenty-four children.
Finally, there were the drama students and members of the imperial drama
troupe, sometimes pressed into action as supers; in a pinch soldiers from
one of the elite guards units stationed in St Petersburg could be called upon.
The imperial stage teemed with life.
It also mirrored the hierarchy that governed all aspects of imperial life.
Ranks determined the minimum or maximum number of dancers who
could appear in a group. Coryphées could dance in groups of no more
than eight; second soloists in groups of no more than four; first dancers in
groups of no more than two. The ballerina danced alone. Thus hierarchy
was built into the very substance of the choreography. It also went to the
heart of the ballet’s social “message”. Sleeping Beauty, like Swan Lake (1895)
and Raymonda (1898), invokes a monarchical ideal of the well-ordered
polity ruled by its rightful king. Hence, the obsession in all three ballets
with marriage as source of continuity in an absolutist state and the risk of
choosing an inappropriate mate or (in the case of Aurora) no mate at all.
The chaos produced by Carabosse in Sleeping Beauty, Odile in Swan Lake
and Abderrakhman in Raymonda suggests that what is really at stake in these
ballets is the idea of autocracy itself. To be sure none of these ballets (or any
other ballets produced during Vsevolozhsky’s directorship) takes place in
Russia. But in their court settings and in the storyline that identifies them
159 Russian ballet in the age of Petipa

with outmoded forms of government, they celebrated a political vision that


pointed directly to the Romanovs.
Sleeping Beauty and its successors, especially Swan Lake and Raymonda,
offered a veritable encyclopedia of the codified forms constituting Petipa’s
“classicism”. Among these was the grand pas de deux, a virtuoso dance for
the ballerina and her partner that marked the climactic end of an act. Petipa
did not invent the pas de deux, but he significantly transformed it, codifying
it as a multi-part number that opened with a supported adagio, continued
with one or two pairs of variations and ended in a triumphant coda. Just as
Petipa had codified the pas de deux, so he did the same with the virtuoso
variation. It was in three parts, with repeats, so that what was done first
on one side was repeated on the other. Unlike the transitional ballets of
the late 1880s, Sleeping Beauty did not simply display the new technical
developments associated with the Italian school. Rather, for the first time,
Petipa succeeded in fusing them with the elegance of the French school, the
basis of traditional Russian training. This synthesis laid the foundation of
the modern Russian school.
Petipa also brought a very high degree of perfection to what is sometimes
called the grand pas or grand pas d’action. This was the climax of a ballet, a
dance in classical style for a ballerina, premier danseur, soloists and corps de
ballet that generalised the main idea of the ballet.26 Although the number of
characters might vary, a grand pas typically included the ballet’s protagonists
along with soloists, coryphées and corps members. The grand pas had a high
degree of formal unity but little narrative function, the reason it could easily
be extracted from the surrounding ballet and performed independently.
The “Jardin Animé” in Le Corsaire is one such dance; another the grand pas
of Paquita; still another the grand pas classique (also called the grand pas
Hongrois) in Act iii of Raymonda.
In the flush of excitement that followed the success of The Sleeping Beauty,
Tchaikovsky agreed to do another ballet, The Nutcracker. Amazingly, given
its longevity and protean identity, the ballet was ill-starred. Tchaikovsky
disliked the libretto, based on a story by E. T. A. Hoffmann, and after the
ballet went into rehearsal, Petipa fell ill and turned the choreography over
to his assistant ballet master, Lev Ivanov. When the ballet finally opened in
1892, it met with sharp criticism. “Nutcracker can in no event be called a
ballet,” wrote one critic.27
Three years later, Swan Lake, the most iconic of nineteenth-century bal-
lets, came to life at the Maryinsky. Petipa had choreographed Act I and
most of Act III, the ball scene, including the “Black Swan” pas de deux for
Odette’s look-alike, Odile: the two roles were danced by the same ballerina.
Here, writes Krasovskaya, Petipa “brilliantly [set] off Ivanov’s Odette, with
her elegiac arabesques, against Odile, the bird of prey, with her resilient and
160 Lynn Garafola

Figure 27 Swan Lake, St Petersburg 1910, choreography by Lev Ivanov and Marius Petipa, music
by Pyotr Ilyich Tchaikovsky.

commanding attitudes. His skill triumphed in the fouetté” – a sequence of


thirty-two of those highly virtuosic turns – “which was no longer a techni-
cal stunt but the culmination in the depiction of cunning temptation: the
swift repetition of the dancer’s spins put the finishing touches to Odile’s
character.”28 Although Ivanov choreographed the lakeside dances of the
last act, Petipa sketched out at least some of the action. He identified the
dancers of various numbers and pondered the colour scheme of the cos-
tumes, deciding on black and white to underscore the theme of moral guilt
and the presentiment of death.29 The intonation of sorrow was intensified
in the waltz, when the black swans cut through the lines of white swans
and climaxed at the end of the act, when first Odette, then Siegfried died by
their own hands, thereby breaking the spell. With the lovers united in death,
Rothbart, the “evil genie”, as he was called in the libretto, fell dead. In the
apotheosis the lovers appeared in the clouds, seated on enormous swans,30
giving the ballet a happy, if banal, ending (see Figs. 27 and 28).
Apart from Giselle, which Petipa revived and significantly revised in
1884, what survives of his work comes from the 1890s or turn-of-the-century
recensions of earlier works such as La Bayadère, Le Corsaire and Don Quixote
(although how much of Petipa remains in Aleksandr Gorsky’s 1902 restaging
is a matter of debate). Part of the reason these works have survived is because
they were choreographed or revised towards the end of Petipa’s career, thus
representing his last word, so to speak, on an art that he had spent the
previous fifty years perfecting. They also survive because they comprised
161 Russian ballet in the age of Petipa

the stock of works belonging to the “old ballet”, as Petipa’s work began to
be called around 1910, that dancers of the 1910s and 1920s brought with
them to the West and that the post-revolutionary generation in Russia made
the cornerstone of the Soviet, “academic” repertory. Finally, the ballets of
the 1890s have survived because of their high musical quality. This was
especially true in the West, where Diaghilev’s revolution in ballet music
made composers like Pugni and Minkus sound hopelessly old-fashioned.
It was not until the early 1960s, when the Kirov began to tour outside the
Soviet bloc, that the “Kingdom of the Shades” was introduced to Western
audiences. Le Corsaire and Don Quixote came later.
Although Russian ballet of the 1890s revealed an even greater Western
focus than in previous decades, the number of foreigners in the Peters-
burg company was actually declining. Visiting ballerinas came and went,
but with the departure of Pierina Legnani, the first Petersburg Odette–
Odile, they pretty much left for good. Now, led by Mathilde Kshesinska, the
company’s prima ballerina assoluta, an extraordinary generation of Russian
talent emerged from the Imperial Ballet School. The dancers themselves
differed from their nineteenth-century predecessors in several important
ways. Not only had the social class of the dancers risen, but so had salaries.
Although these were not especially generous, they were far more than the
pittance of the 1870s, when members of the corps earned between 174 and
240 roubles a year. (Yekaterina Vazem, the highest paid member of the com-
pany, received 1,143 roubles in salary plus 25 roubles per performance.)31
Conditions at the school had also improved and memoirists like Tamara
Karsavina describe the ample meals and generous allotment of clothing that
the children received, regardless of means. A growing number of dancers,
including Mikhail Fokine and Adolph Bolm, came from merchant families,
a sure sign that the social status of the dancer was on the rise. Many dancers
were now the children, siblings, or spouses of dancers, members of clans that
formed an increasingly privileged caste within the Russian theatrical world.
Some, like the Kshesinskys, were quite well off. Others, like the Karsavins,
lived in “reduced circumstances” on the father’s small pension and meagre
salary as a part-time teacher.32 Nevertheless, conditions at the end of the
nineteenth century were far better than at mid-century, when poverty was
well-nigh universal.33
Finally, the training had significantly changed during Petipa’s years in
Russia. Students who attended the Imperial Theatre School in the 1830s
and 1840s studied acting, singing as well as dancing and were then placed
according to their talents. By the 1890s ballet training had become highly
specialised (although students continued to study music as well as aca-
demic subjects).34 The result was a company with a very high degree of
professionalism, which compared favourably with standards elsewhere in
162 Lynn Garafola

Figure 28 Scene from Swan Lake, Bolshoi Ballet Moscow, 1959.

Europe, as became evident once the ban on private theatrical enterprise was
lifted.
Despite the social changes, liaisons remained a fact of life. Some “protec-
tors” eventually married the women who had been their lovers: this was the
case of Agrippina Vaganova, the legendary Soviet-era pedagogue and her
husband, Andrey Pomerantsev,35 as well as Anna Pavlova and Victor Dandré.
Kshesinska reserved her favours for Romanovs – first, the future Nicholas
II, then his cousins, Grand Duke Sergey and Grand Duke André, whom
she married years later in emigration. She collected houses as well as jewels
and was not averse to throwing around her power. When Vsevolozhsky’s
successor (and nephew), Prince Serge Volkonsky, fined her for an unautho-
rised change of costume in the ballet Camargo (1872), she went straight
to her former lover, Nicholas II, who requested that the fine be annulled.
In Volkonsky’s audience with Nicholas, he tried to explain “the impossible
conditions . . . produced by interference with my dispositions, owing to the
exceptional position in which one dancer was placed and her precedence
before all the others”.36 Volkonsky ended by requesting permission to resign.
This incident, which took place in 1901, underscored the very close
relationship between the Romanovs and the ballet. Members of the impe-
rial family celebrated birthdays and name-days at the ballet; they attended
school performances and distributed chocolates to the children; they made
friends with the dancers, weighed in on performances, used their influence
to promote their favourites and sometimes found mistresses. “How happy
163 Russian ballet in the age of Petipa

we felt at the thought of being allowed to dance in the presence of the Impe-
rial Family!” exclaimed Kshesinska, recalling the school performance when
Alexander III singled her out for praise.37 Far more than any of the other
arts, ballet in Russia was a reflection of what Richard Taruskin has called,
“the last surviving eighteenth-century (hierarchical, aristocratic) society in
Europe”.38
Although Petipa remained very much a Frenchman, it was under his
require that the Imperial Ballet began the slow transformation from an
art of the court, understood in the broadest sense, to an art of the nation.
By the 1890s nationalism was a recurring theme in ballet criticism, both
with regard to genres like the féerie and foreign ballerinas who did not find
favour with Russian audiences. Ironically, this identity was only fully realised
after the triumph of Diaghilev’s Ballets Russes in Paris beginning in 1909.
Even though the Ballets Russes never performed in Russia and exemplified
the “new ballet”, it created for Russian ballet, old and new, a place in the
European cultural imaginary and the international cultural marketplace.
By then the age of Petipa was long over and Russian ballet had ceased
to be a purely Russian phenomenon. Now, through the diaspora of Russian
dancers in the 1910s and 1920s, his last generation of dancers extended his
influence far beyond the Imperial Theatres where Petipa had spent his most
productive years. In an art where oblivion is the fate of all but the tiniest
minority of works, his ballets, virtually unknown outside Russia during his
lifetime, have become the “classics” of a common heritage, while his school,
updated and transformed, continues to be its lingua franca.
14 Opening the door to a fairy-tale world:
Tchaikovsky’s ballet music
t h é r è s e h u r l ey

Ballet is the most innocent, the most moral of all the arts. If that is not so, then why do they always
bring children to the ballet?1 ( p. i . tch a i kovs k y )

Tchaikovsky’s belief in the purity of ballet as well as its link to childhood lies
at the core of his intent in and approach to composing ballet music. Accord-
ing to friend and music critic Herman Laroche, the composer welcomed
writing for the ballet because, “in that magical world, it was pure fairy tale
expressed by pantomime and dance”.2 Laroche added, “Tchaikovsky could
not stand realism in ballet.”3 Tchaikovsky reserved ballet as the ideal genre
for complete submersion into a child’s fantasy world and so his Swan Lake,
The Sleeping Beauty and The Nutcracker are all fantastic tales that draw the
audience into a magical world of swan maidens, sorcerers, fairies, mouse
kings, princes and princesses. In this chapter, I will show how Tchaikovsky’s
music for these ballets is the key that unlocks the door to a fairy-tale world.
For pantomime and dance, the visual tools used to convey these stories
cannot alone evoke the fantasy. It is Tchaikovsky’s music, especially in his
masterful choice of instruments and ways of deploying them, that transports
his audience to a world of fantasy and magic.
Before delving into Tchaikovsky’s ballet music, one first should recognise
the role of a ballet composer in the nineteenth century, which was to high-
light the actions on stage. The pantomimes required music that expressed
the feelings of the characters, moment to moment, but the dance music
often was simpler and somewhat repetitive in both melody and rhythm,
allowing the audience’s attention to be drawn to the dancers’ steps rather
than to complexities in the music. Composers were expected to meet the
requests of the choreographers – usually known as “ballet masters” – and to
compose music that helped express the action and accompany the dancing
in suitable fashion.

Swan Lake
Swan Lake: A Ballet in Four Acts (composed August 1875–10 April 1876)
[164] was premiered in Moscow on 20 February 1877 to less than enthusiastic
165 Tchaikovsky’s ballet music

reviews, but the 1895 revival of the ballet by the choreographers Marius
Petipa and Lev Ivanov, at the Maryinsky Theatre in St Petersburg, found
much greater success. (Alas, Tchaikovsky did not live to see it.) For this later
version – upon which today’s productions are based – Tchaikovsky’s brother
Modest reworked the libretto and the Maryinsky conductor and composer
Riccardo Drigo reorchestrated the score, adding to it three of Tchaikovsky’s
piano pieces from Op. 72, L’Espiègle, Valse Bluette and Un poco di Chopin,
which he orchestrated as well.
Little evidence survives of the collaboration of Swan Lake’s choreogra-
pher Julius Reisinger and the composer; we cannot know how Tchaikovsky
might have responded to any specific requests Reisinger made. We do know
that the composer took a keen interest in more than just the musical aspects
of the ballet’s first production. He was very specific about special effects, as
machinist Karl Valts recalls:

Peter Ilyich gave special attention to the final act. In the storm scene, when
the lake overflows its banks and floods the entire stage, a real whirlwind was
built at Tchaikovsky’s insistence. Branches and twigs of trees were broken,
fell into the water, and were carried away by the waves. After the storm, for
the apotheosis, dawn came and the landscape was illuminated by the first
rays of the rising sun at the curtain.4

His music brings out the drama of the storm scene, so much so that Laroche
criticised the orchestration:

For what do the trumpets, trombones and kettledrums roar when in the
depth of an empty stage flies a band of swans? This moment demands soft
peaceful sounds . . . I think . . . Tchaikovsky, as on many earlier occasions,
was simply carried away by his peculiar weakness for loud sounds.5

Another critic remarked that “Tchaikovsky displayed an astonishing knowl-


edge of instrumentation – a characteristic that is being admitted by friends
and enemies alike. He gave new evidence of his ability to make masterful
use of orchestral forces.”6
Clearly, Tchaikovsky had a vision of the way in which his ballet was
to come to life – not just the music, but also the visual and sound effects
on stage combined. And he was more interested in conveying the tumult
and despair felt in the final scene than the immediate picture of swans in
flight. When he chose to depict the swans, he did so very effectively. Let us
consider the case of the famous Swan Theme, which is heard for the first time
at the end of Act i with the appearance of the swans at night. Tchaikovsky
creates a sense of mystery and magic by simulating the rippling waves of
the lake with B minor arpeggios on the harp over the shimmering tremolo
of the strings. In the second bar, the oboe announces the theme, which
166 Thérèse Hurley

bears a striking resemblance to Lohengrin’s warning to Elsa in Wagner’s


opera.7 (Fully acquainted with the score of Lohengrin, Tchaikovsky most
likely chose to echo the tragedy that was to come in the musical warning
to Siegfried.) When the Swan Theme recurs at the opening of Act ii, the
tempo is accelerated, adding a sense of urgency and making the oboe sound
as a cry for help. Moreover, instead of modulating to major and adding a
new theme (as he does in Act i), Tchaikovsky continues to develop the Swan
Theme until the orchestra ends in fff (very loud) and finally fades with the
B minor tremolo in the upper strings – the other-worldly tone is thus set
for the rest for the act.
The famous love pas de deux of Act ii provides not only another example
of Tchaikovsky’s deftness as an orchestrator, but also his ability to narrate
the story (albeit subtly) through his music. The opening, virtuosic harp
cadenza incorporates unresolved half-diminished seventh chords clearly
meant to evoke the feeling of desire between Odette and Prince Siegfried.
By giving this music to the harp, Tchaikovsky offers a sense of fantasy and
paints a picture of what is (passion) and what is to be (magic and sor-
cery). The harp cadenza segues into a simple, chordal accompaniment for
the solo violin’s statement of the first theme in G major. The oboes and
clarinets soon disrupt the heart-wrenching melody with a repetitive semi-
quaver rhythmic pattern (bars 34–42). This unusual rhythm causes a sense
of anxiety as it chromatically ascends, modulating to E major, and arrives
at the violin’s playful second theme. When the first theme returns, the cello
(representing Siegfried) plays the melody while the violin (Odette) sings a
plaintive counterpoint.8 After this moving and beautiful declaration of love,
the two lovers and their fates are altered forever. The effect of this change
is heard in the reprise of the Swan Waltz (Variation vi) when Tchaikovsky
interpolates the repetitive, palpitating rhythm (this time as quavers) and
confirms through his music that now everything (even the Swan Waltz) is
different.
In Act iii, Siegfried makes the fatal mistake of believing that the evil Odile
is his beloved (though his friend Benno sees no resemblance at all between
the two). How does Tchaikovsky fool Siegfried into this blunder? The answer
is found in his instrumentation. For when Odile appears unexpectedly at the
ball, she dances to an oboe solo accompanied by harp. Siegfried last heard
this combination when he saw his Odette, so the return of this duo, he
believes, must mean the return of the Swan Princess. In addition to creating
the illusion that Odile is Odette, Tchaikovsky also leaves a musical clue that
Odile is an impostor: we only have to look at the key signature to learn that
the dance is not in one of Odette’s keys, but rather in the sorcerer Rothbart’s
key of F minor – clearly Siegfried is doomed. His devastating mistake is
revealed when he vows he will marry Odile.9
167 Tchaikovsky’s ballet music

By the few musical examples above, we can glean that Tchaikovsky was
very conscientious in his use of instruments to simulate nature (harp arpeg-
gios as waves), affect the plot (instrumentation used to deceive Siegfried)
and add a touch of magic (harp cadenza and accompaniment in the love
pas de deux). Yet the composer himself was not satisfied with his instru-
mentation for Swan Lake and according to Drigo “had intended to take up
the matter, but he never managed to do this”.10 We can only speculate what
changes Tchaikovsky himself would have made had he lived long enough to
collaborate on the Petipa/Ivanov production.

Sleeping Beauty
Eleven years after the premiere of Swan Lake, Tchaikovsky received a request
from Ivan Alexandrovich Vsevolozhsky, the Director of the Imperial The-
atres of St Petersburg, to compose a new ballet:
I conceived the idea of writing a libretto on La Belle au bois dormant after
Perrault’s tale. I want to do the mise-en-scène in the style of Louis xiv. Here
the musical imagination can be carried away, and melodies composed in the
spirit of Lully, Bach, Rameau, etc., etc. In the last act indispensably
necessary is a quadrille of all Perrault’s tales.11

Tchaikovsky accepted enthusiastically:


I very much want to tell you straight away that I’m charmed, delighted
beyond all description. It suits me perfectly, and I would ask for nothing
more than to set it to music. It would be impossible to make a better stage
arrangement of the elements in this delicious subject.12

As he worked on the new ballet, the composer described his labour in a


letter to his patroness, Nadezhda von Meck: “I worked on its instrumenta-
tion with special love and care and invented some completely new orches-
tral combinations which, I hope, will be very beautiful and interesting.”13
Indeed, Sleeping Beauty, upon its premiere on 2 January 1890 at the Maryin-
sky Theatre, was deemed by Herman Laroche “one of our theatre’s pearls”
and a musical triumph for Tchaikovsky as well.14

The choreographer Marius Petipa wrote out instructions for Tchaikovsky for
Sleeping Beauty, and they make fascinating reading, for they show us how the
composer responded musically to specific – and sometimes very narrow –
requests. Petipa indicated expressive musical effects, metres, types and
lengths of each dance in bars and, occasionally, he even requested instru-
ments, as in “Variation d’Aurore” in Act i: “3/4 pizzicato for violins, cellos,
and harp. (Excuse me for expressing myself so oddly.) And then lute [sic]
168 Thérèse Hurley

and violin.” For the evil Fairy Carabosse’s appearance in the Prologue, for
instance, Petipa requested “music of a fantastic character”; and Tchaikovsky
responded with the menacing Carabosse Theme, in which horns and wood-
winds strongly accent the offbeats as the clarinet sings out the chromatic
melody, sometimes in duet with the bassoon. These darker woodwinds
return to represent the mysterious and vengeful Carabosse throughout the
ballet.
In contrast, for the Lilac Fairy’s softening of the evil curse of Cara-
bosse, Petipa suggested “tender and rather derisive music”.15 In response,
Tchaikovsky presented the lullaby-like Lilac Fairy Theme, introduced by an
ascending E major scale glissando on harp and performed by the oboe. Like-
wise, this theme reappears in the first and second acts. In fact, Tchaikovsky
included both of these themes in his Introduction to the ballet, so the audi-
ence is made aware immediately of the contrasting forces in the story. The
Carabosse Theme evokes danger and mystery and the Lilac Fairy Theme
bodes tranquillity and magic; this sets up the opposing forces that provide
the drama and suspense of the entire ballet.
Note further the close connection of the harp to the Lilac Fairy. From
the beginning, the harp is used to represent magic and the mitigating of
Carabosse’s curse of certain death on the Princess Aurore to a mere hundred
years of sleep. Throughout the Prologue and first two acts, the harp provides
glissandos, arpeggios, lushly rolled chords and even harmonics (which make
for a shimmering, celestial effect). However, the instrument is silent during
the final act. What is the reasoning? First, let us recall that the final act is
a set of dances at the wedding celebration of Princess Aurore and Prince
Désiré. By this point, all fantastical activity has ceased. The Lilac Fairy is
no longer needed and neither is her magic. Second, we encounter a new
sound emerging from the orchestra pit: a piano. Just as the Lilac Fairy has
given way to the Gold, Silver, Sapphire and Diamond Fairies, so has her
supernatural instrument, the harp, been replaced by the earthly sound of
the piano. The piano works especially well, for instance, for the Silver Fairy,
for whom Petipa had specified that “the sound of coins must be heard”. The
clarity and percussive quality of the piano lends itself well to this music.
Also, by engaging the flute to help project the melody over the campanelli
(orchestra bells) and pizzicato string accompaniment, Tchaikovsky adds a
sparkle to the piano’s timbre.
In addition to expressive musical sound effects, Petipa also informed
Tchaikovsky of the specific metre in which he was to compose. Although
Tchaikovsky followed these suggestions, the dancers seemed to have dif-
ficulty with some of his metres and rhythms. Riccardo Drigo, conductor
of the first production, mentioned that their “internal resistance . . . was
great, as each variation went beyond the usual formulas to which the ear
169 Tchaikovsky’s ballet music

was accustomed”.16 It seems that the composer was continuing to push the
boundaries of what was considered to be typical ballet music at the time.
However, Petipa’s prescribed metres were not always in simple duple or
triple. For example, in the Dance of the Sapphire Fairy of Act iii, Petipa wrote,
“Sapphire – five points, music in quintuple time”. Therefore, Tchaikovsky
composed the dance in 5/4 – an unusual metre to be sure, but one that the
choreographer had ordered.
Petipa also specified the types of dances such as the mazurka in Act II
and the grand polonaise at the beginning of Act iii. In addition to these
requests, he described the character of the dances performed in Act ii. For
the Dance of the Duchesses, he merely wrote: “They are noble and proud.”
For the Dance of the Baronesses: “They are arrogant and conceited.” To
compose a minuet for the duchesses and a gavotte for the baronesses was
entirely Tchaikovsky’s decision. Since this act occurred a hundred years
after the time of Louis xiv, it made perfect sense for Tchaikovsky to have
composed the music in a Mozartian fashion. This classical style evokes the
air of aristocracy while alluding to the change in time period.
As we have seen so far, Tchaikovsky did attempt to follow Petipa’s requests
for expressive effects, metre and dance. But the composer often exceeded
the length requested by the choreographer. For the Act i waltz in Sleeping
Beauty instead of the 16 bars of introduction and 150 bars of waltz Petipa had
requested, Tchaikovsky supplied 36 and 261 bars respectively. He surpassed
the length suggested for the introduction and polonaise at the beginning
of Act iii as well. One wonders how Petipa and Tchaikovsky handled dis-
crepancies such as this. Fortunately, we have a hint from Alexander Shirayev
(grandson of the composer Cesare Pugni and dancer with the Petersburg
Imperial Ballet), who remarked that Tchaikovsky, like Glazunov, was will-
ing to meet Petipa halfway regarding alterations to the score, but that the
choreographer “was hesitant to make demands of them as he had of [minor
composers] Pugni and Minkus, who at his wish reworked their composi-
tions straightaway at rehearsal. Petipa had therefore to work quite hard on
Sleeping Beauty. This he confessed to me.”17 From this recollection, we can
surmise that Tchaikovsky was greatly respected by Petipa. The respect must
have been mutual for when asked to collaborate again – this time for The
Nutcracker – Tchaikovsky agreed.

The Nutcracker
Almost two years after finishing The Sleeping Beauty, Tchaikovsky wrote his
final ballet, The Nutcracker: Ballet in 2 Acts (composed February 1891–
4 April 1892). Although somewhat sceptical of the subject matter, the
170 Thérèse Hurley

composer set out to work on the ballet, which would become a worldwide
Christmas favourite. Originally, the ballet was to have the same collaborators
as The Sleeping Beauty, but Lev Ivanov replaced Petipa when the great ballet
master fell ill. The ballet premiered at the Maryinsky Theatre on 6 Decem-
ber 1892 to mixed reviews, in part because the plot never seemed resolved
and the heroine, Clara, never returned from Confiturembourg, the Palace
of Sweets. Many critics complained that the story was pointless and were
surprised to hear that Tchaikovsky would even consider working on such a
ballet.18 In any case, his efforts proved fruitful indeed, and his score surely
accounts in great part for the continuing popularity of this work.
As with The Sleeping Beauty, Tchaikovsky received instructions from
Petipa for The Nutcracker and responded with character-related instrumen-
tation, musical sound effects and memorable melodies. He also borrowed
music from other sources and worked them into the score. As in his earlier
ballets, he adhered to an overall tonal plan and established a basic frame-
work from which to compose music that draws the audience into a magical
world.
For Councillor Drosselmayer’s (later changed into Drosselmeier) first
appearance, Petipa asked for “very serious, rather sinister and even droll
music”. At first, the Drosselmayer Theme is introduced by viola with trom-
bone and tuba accompaniment, but then the clarinet and bassoon perform
the mischievous melody with cello and bass accompaniment. We heard
these lower woodwinds before: during Carabosse’s Theme in The Sleeping
Beauty. We can assume that Tchaikovsky wished to evoke the same mystery
and darkness. This darkness returns at midnight when an eerie glow shrouds
the Nutcracker. Although the melody is altered, the instruments of clarinet
and bassoon remain the same and so does the sentiment of uneasiness and
fear.
The battle scene between the Nutcracker and Mouse King (Act i, no.
7), which culminates in the transformation of the Nutcracker doll to an
enchanting Prince, is the most action-filled part of the entire drama. Already
a seasoned battle composer with the 1812 Overture behind him, Tchaikovsky
was certainly well equipped to simulate a “hail of grapeshot, volley of guns,
piercing cries”. The lower strings often play a galloping rhythmic motif with
snare drums, woodwinds and brass sounding military calls. And at the point
for which Petipa called for the Nutcracker to summon his men “To arms!”
Tchaikovsky wrote a trumpet call to verbalise this order musically – there
is no mistaking this as a military command in the midst of the belligerent
action on stage. The choreographer was also specific in his requests for
the post-ballet scene: “The mice are victorious and devour the gingerbread
soldiers. 8 bars after the 48 b[ars] of battle, in order that the mice’s teeth
can be heard chewing on the gingerbread.” Tchaikovsky responded with a
171 Tchaikovsky’s ballet music

repetitive descending scale in the strings to depict the ravenous rodents.


For Clara’s throwing of her slipper at the Mouse King, Petipa requested
“2 bars for the piercing cry and 6 for the whistling of the mice, which
disappear”. Then follows the transformation of the Nutcracker into the
Prince, for which Petipa instructed “One or two chords”. The instructions
seem rather minimal for the most triumphant scene of the entire production,
but Tchaikovsky cleverly created a transition from the evil battle scene to the
beauty and magic of Confiturembourg on the basis of these few suggestions.
The oboe, clarinet and horns blare out a tritone for the King’s cry in the
suggested two bars. The clarinets and bassoons are called upon once more
to add a feeling of dark magic as the mice scurry away in the suggested six
bars. In the third bar of the scurrying mice, the cellos and basses pluck a
pizzicato heartbeat. The trombones and tuba enter to sound the death knell
of the Mouse King. Suddenly, we witness the transformation as the upper
strings play a swelling ascending scale leaping to a high C and then quietly
descending over the heartbeat motif. The Nutcracker is no longer a doll, for
we hear his heartbeat and the scale sounds as the breath of life, bestowed
upon the Prince. Although Petipa asked for a meagre two chords for this
transformation, Tchaikovsky exceeded expectations again by creating a full-
fledged musical metamorphosis.
With all signs and sounds of evil and mischief dispelled, the next scene
opens with the rippling arpeggios of two harps. As in both earlier ballets, the
composer continues to use the harp to signify magic. What was Clara’s house
at night has become a fir forest in winter. Gnomes with torches are paying
homage to the Prince and Clara. All semblances of reality and evil disappear
and the characters as well as the audience enter a childlike fairy-tale world.
In this fantasy-land, the harp – aside from evoking the magic of the place – is
used to create a whirlwind within the Waltz of the Snowflakes. (Petipa here
noted that a “strong burst of wind breaks up the snowball and the dancers
spin around”.) By writing the harp glissandos for two parts in contrasting
directions, Tchaikovsky achieves this blustery effect, demonstrating both
his expertise in composing idiomatically for the instrument and his creativ-
ity in finding new ways to simulate non-musical sounds. Ever fascinated
with new effects and the possibilities of the orchestra, Tchaikovsky made
use of a newly invented instrument in The Nutcracker: the celesta. When he
discovered it in Paris, he was so charmed by the instrument and its bell-like
tones that he immediately asked his publisher to order one.19 For the ballet,
Petipa had requested that the Dance of the Sugar Plum Fairy should sound
“as if drops of water shooting out of fountains are heard”. Combining the
delicacy of the harp with the percussive clarity of the piano, the celesta
beautifully depicts staccato sprays of water; Tchaikovsky had found his
fountain.
172 Thérèse Hurley

In addition to the celesta’s famous solo, the instrument is heard often


throughout the ballet, specifically in Confiturembourg and usually with the
harps. In fact, Tchaikovsky (not content with just one new sound in this
ballet) incorporated the quivering tones of a specific flute technique called
flutter-tonguing (frulato) – something rarely used at the time. Together,
these three instruments simulate the undulating currents of the river of rose
oil, which flows through the kingdom of Confiturembourg. Therefore, just as
the lower woodwinds evoked darkness and the brass and percussion led us
into battle, the combination of celesta, harps and flutes create the sparkling
sounds of the wintry fairy- tale land.
In addition to the usual suggestions of metre and dance style, Petipa
asked Tchaikovsky to incorporate some previously composed music into
the score, specifically the French song, “Bon voyage, Monsieur Dumolet”,
during the Christmas party in Act i, /no. 3 (bars 61–118). Tchaikovsky also
included a German dance called the Grossvater for the party (Act i, no. 5, bars
187–219) – appropriate, since the Christmas party takes place in Germany.
In Act ii, Tchaikovsky used two popular French songs for the character
dance of Mother Gigogne and her children as well as a Georgian lullaby for
the Arabian Dance.20 The inclusion of two merry French songs for Mother
Gigogne seems a plausible choice. She is French and the songs, very playful
and sprightly, fulfil Petipa’s descriptions of dancing buffoons. But why did
Tchaikovsky choose a Georgian lullaby for a sultry Arabian dance? Petipa’s
instructions call for an “Oriental Dance. From 24 to 32 bars of charming
and voluptuous music” and he even specified the “kingdom of Yemen”.
Tchaikovsky’s selection compels us to glance at the geographic location of
Georgia as well as its history. Georgia (which shares borders with Turkey,
Armenia and Azerbaijan) was part of Imperial Russia at this time, but had
also endured many other invasions and occupations. The region had been
part of the Persian and Ottoman empires in the sixteenth century and had
been invaded by Arabs in the seventh century. With Georgia’s longstanding
relationship to the Middle East in mind, the melody choice begins to become
clear. To Tchaikovsky, this melody must have been quite evocative of the
Arabic musical genre and therefore an ideal choice. He accentuated the
connection with his orchestration and his melodic ornamentation. We hear
the violins softly playing the lullaby theme over a repetitive G and later
a G–D drone in the lower strings. This string arrangement establishes a
folk-like quality, but the woodwinds (specifically the oboe and cor anglais)
decidedly add the timbre of Middle Eastern wind instruments. By this time
in nineteenth-century ballet, an oboe solo in a minor key was considered a
standard choice for a character (also called national) dance of Arabia. The
subtle rattle of a tambourine punctuates the phrases and adds a finishing
touch of exoticism to the music.
173 Tchaikovsky’s ballet music

As noted above, Tchaikovsky tended to use simple melodies and fre-


quent repetition for passages of virtuosic dance, so as to avoid diverting the
audience’s attention from the dancing to the music. In The Nutcracker he
chose a simple repetitive melody to create the illusion of a growing Christmas
tree in Act i, no. 6. Petipa instructed that “the Christmas tree becomes huge.
48 bars of fantastic music with a grandiose crescendo”. At first, the horns pro-
vide the harmonic accompaniment beneath the melody of the first violins
while the harp evokes images of the tree’s growth. All this begins pianissimo
and gradually increases in volume until the grand cymbals crash. Then the
violas and cellos quietly introduce the melody in canon with the violins.
This time, the woodwinds have joined the horns in harmonic accompani-
ment. The trombones and tubas soon play the melody with the cellos as the
orchestra crescendoes once again. Another thunderous crash of the cymbals
is heard followed by the violas and cellos playing the melody piano this time.
The trumpets, flutes and clarinets take their turns at the short phrase while
increasing to ffff and arriving at the triumphant metamorphosis of the
tree. Although the audience can see the tree magically grow on stage during
the performance, we can hear the fantastic change take place even without
any visual aids. As Tchaikovsky’s short ascending phrase gives the Christmas
tree its height, the harp arpeggios add breadth as they cascade up and down
the instrument. Simply by developing a short phrase, Tchaikovsky was able
to create one of the most effective illusions in nineteenth-century theatrical
music. In The Nutcracker, Tchaikovsky once more received instructions and
fulfilled them with a keen ear for musically depicting the action on stage. He
follows the choreographer’s requests in most areas, but, as before, exceeds
them in others – namely the number of bars. The Arabian Dance was meant
to be 24–32 bars. Tchaikovsky wrote 102. The Final Waltz and Apotheosis of
Act ii was to be 128 and 16–24 bars and ended up at 239 and 55 respectively.
Yet, it does not seem that his excessive productivity had any negative effect
on the music or the ballet. On the contrary, it was surely his score that saved
the ballet from being forgotten and elevated it to the status it enjoys today
as a Christmas standard for so many ballet companies.

In Swan Lake, The Sleeping Beauty and The Nutcracker, Tchaikovsky, pro-
pelled by his belief in the innocence of ballet and perhaps his own happy rec-
ollection and longing for the early days of childhood, followed the requests
of ballet masters (notably Petipa) and composed music for swan maidens,
fairies and mice in all sorts of situations such as enchanted lakes, weddings
and even battles. Like any ballet specialist composer, he wrote expressive
music for the pantomimes and danceable music for the dances, but in his
music we find something more than a mere fulfilment of choreographic
demands. Tchaikovsky possessed a genius for melody (whether original or
174 Thérèse Hurley

borrowed) – the Swan Theme, the Arabian Dance, the Dance of the Sugar
Plum Fairy, to name only a few. But he also displayed an acute sensitivity to
orchestral colour and the ability to evoke a magical sound world through
his skilful approach to instrumentation and orchestration. Whether it is the
swanlike cry of the oboe, the rushing waves of harp arpeggios or the fountain
sprays of the celesta, it is through these means that the music springs forth
with a vitality that sparks the imagination and opens the door to a fairy tale
world.
15 The romantic ballet and its critics:
dance goes public
lu c i a ru p re c h t

The extraordinarily successful phenomenon of the romantic ballet repre-


sents a period of renewal of theatre dance but also a symbiosis between the
performed and the written, between dancer and critic. Romantic ballet is an
aesthetic movement both embodied and discursive. Ballet in the 1830s and
1840s cannot be considered without taking into account its written testi-
monies, which described a new and sensational physical technique, sugges-
tive stage technology and an elaborate dramatic style. A cult of the romantic
ballerina grew up that soon reached the higher spheres of myth-making.
While there had always been admiration for stellar dancers, the ‘star system’
came into its own in the nineteenth century. The new writing on dance fol-
lowed an era of aesthetic redefinition and fits perfectly into Habermas’s the-
ory of the emergence of Öffentlichkeit or the “public sphere”. The commer-
cialisation of opera performance brought in its wake a demand for consumer
information and led to a flood of journalistic and fictional writings that
grounded ballet firmly in the rapidly developing field of publicity.1 Audi-
ences expanded and diversified. They extended to those who did not have to
be present at a performance at all, to the “liseuses de feuilleton” and to those
who enjoyed being able to observe the dancers through the eyes of a critic
who might even allow glimpses into the secret spaces behind the stage, the
green rooms of Europe’s theatres to which only the lucky few were admitted.
If we think of criticism as a medium that makes art and aesthetic concepts
public property, nineteenth-century ballet criticism distinguishes itself at
first sight by its overtly voyeuristic attitude: it made the female performers
public property, raising their attractiveness by an elaborate “hierarchy of
proximity” as Marian Smith called it. It created an image of the opera
house as a place where gaining access to dancers’ bodies – whether visual
or otherwise – was not just a privilege of the rich and influential. Through
the voices of male critics, the ballerina gained a prominent place in the
feuilleton, but as object not subject.
Since voyeurism among dance critics will hardly surprise readers, it may
be more useful to look at a cross-section of a range of views of well-known
and also lesser-known critics. Their writings display the specific character-
istics of nineteenth-century dance journalism, situated between aesthetic
[175] theory, advertisement and purple prose.
176 Lucia Ruprecht

The Paris Opéra ballet was at the heart of a movement that spread out via
Milan, London, Copenhagen, Vienna and Berlin. French criticism set the
tone by the sheer quantity, as well as the quality of the reviews. These reviews
report on various aspects of the performances, and reveal the aesthetic taste
or even political opinions of the individual critic. They discuss the economic
strategies of the enterprise which the Paris Opéra had become in the 1830s.
Louis Véron, its director from 1831 to 1835, certainly understood the impor-
tance of publicity. He elevated Marie Taglioni to the rank of principal dancer.
He started a tradition that would lead over the course of the next two decades
to an ever increasing level of concentration on the ballerina as celebrity.2
Véron also raised the status of the critics. They were allowed to enter the
inner circle at the Rue Lepeletier in exchange for their services as passionate
promoters of the silent art. In spite of their privileged status, few had any real
understanding or even appreciation of the choreography and the dancing.
The nineteenth-century feuilleton indulged instead in lengthy descriptions
of plots, costumes and scenery. Yet it did provide a continuous record, a ver-
bal acknowledgement of an art form which at that time had almost no other
way to leave traces of its existence. The professional critic wrote a new genre
of reflection on dance. From the philosophical writings of classical antiquity
and the Renaissance to eighteenth-century works by philosophers such as
Jean-Baptiste Du Bos, Charles Batteux, Adam Smith, William Hogarth or
Denis Diderot, dance had been included in aesthetic treatises, even if in
rather marginal ways. More importantly, the choreographers and dancing
masters themselves had articulated their critical positions in their manuals.
Fabricio Caroso’s Nobiltà di dame (1600–5), François de Lauze’s Apologie
de la danse (1623), John Weaver’s Anatomical and Mechanical Lectures upon
Dancing (1721), Jean-Georges Noverre’s Lettres sur la danse, et sur les ballets
(1760) and Gasparo Angiolini’s writings on the eighteenth-century action
ballet – to name only a few – all formulated aesthetic programmes, while
at the same time dealing with theoretical or pedagogical aspects of dance.
The ballet master Carlo Blasis continued this tradition well into the nine-
teenth century. Yet most of his publications (among them An Elementary
Treatise upon the Theory and Practice of the Art of Dancing, 1820, and The
Code of Terpsichore, 1830) were influential mainly because they codified
the classical technique and constantly refined the essentially unchanged
standards of ballet as we still know it today. However, for a critic ‘it was
unheard of . . . to participate in a ballet class’;3 the aesthetic judgement of
theatre dance fell therefore almost exclusively into the hands of spectators
who may have been professional writers, but were amateurs in the art of
dancing.
The early generation of critics like Castil-Blaze of the Journal des débats,
from 1815 to 1832 considered the leading Parisian ballet critic of the time,
177 The romantic ballet and its critics

knew their limits. For Castil-Blaze and his fellow reviewers, the onset of an
aesthetic of pure dance that evoked abstract concepts rather than expressing
a plot certainly posed a problem. His criticism still follows the lines of
Noverre’s dogmas, opposing the craft of meaningless mechanical virtuosity
to the art of meaningful mimic expression. The extended mime scenes,
which often dominated the action on stage more than the dancing, were
criticised or ridiculed. Berlioz’s comical desperation in his review of La
Chatte métamorphosée en femme for the Revue et gazette musicale de Paris
(22 October 1837) suggests that the writing of a scenario was an easy task
compared to its analysis:

Writing a programme for a ballet is not difficult in itself; it is much harder to


write a critique of the plot, especially when one is unfortunate enough not to
have understood anything of the pantomime and does not want to consult
the libretto. I find myself in precisely that situation here. Innumerable
experiences have shown me that the mimic art was a closed book for me.

Berlioz’s account testifies to the fact that the reception of ballet was still in
thrall to an aesthetic that had been formed in an age in which theatre dance
was inseparable from opera, and had to match up to the latter’s unique
combination of music, words and mime. Thus the composer-critic cannot
avoid expressing his frustration at what he perceives to be a deficiency of
the performance, rather than of his expectations: “I have never been able to
persuade myself that my mimic sense was so obtuse that I couldn’t discover
the . . . fundamental idea, the basic idea, the i d e a , at the first performance
of a work of this nature.”4
The new generation of critics displayed such a degree of confidence or
sophistication in taste, style and judgement that they in turn became stars
of the ballet scene. Outstanding among these celebrities of criticism was the
writer Théophile Gautier, the most widely known chronicler of the romantic
ballet. He wrote his first proper review of a ballet as dramatic critic of La
Presse in 1837, and continued to monitor dance in Paris until 1871. His career
thus encompassed the heyday of the romantic ballet, as well as its decline
and short-lived renaissance in the shape of Arthur Saint-Léon’s Coppélia.
Gautier’s credentials as a dance writer were his enthusiasm – not least for the
beauty of the ballerinas – his contacts in the profession, aesthetic insights
into the art that grew with his experience and his talent for vivid prose. His
virtuoso writing matched the virtuosity of the performances, and outlived
the ephemeral glory of the feuilleton. The editors of the six volumes of the
Histoire de l’art dramatique en France depuis vingt-cinq ans, a collection of
Gautier’s newspaper articles, elevated his criticism to a position of equality
with the spectacle on stage; indeed they suggested that the review could
enhance the experience of the performance for the retrospective reader.5
178 Lucia Ruprecht

Although Gautier remained relatively ignorant of choreography, he also


produced various ballet scenarios from 1838 onwards. No less than six
libretti were actually staged. The most famous of them was Giselle (1841),
based on Heinrich Heine’s writings on the popular myth of female spirits
of revenge known as the wilis. Gautier created it in collaboration with the
librettist Vernoy de Saint-Georges and the choreographer Jean Coralli. He
thus knew what he was talking about when in his review of La Gipsy in 1839
he complained: “It is not easy to write for the legs. There can be no proud
bombastic tirades, no fine verse, no poetical clichés, no words for effect, no
puns, no declamations against the nobles, nothing but one situation after
another.”6 The writer laments the gap between his two preferred media, the
word and the body. Reading this danced literature was equally difficult; the
journalist who sought clear dramatic content and a decipherable message
in a ballet could not find it. It was as hard to write about ‘the legs’ as it was
to write for them – yet the possibilities of glamorous literary invention were
almost limitless (see Fig. 29).
Jules Janin, the other most prominent critic similar to Gautier of the
romantic ballet, abandoned the more content-oriented judgements of a
Castil-Blaze. They expressed a new aesthetic of theatre dance. Janin suc-
ceeded Castil-Blaze at the Journal des débats. Some of his reviews, for instance
that of Nathalie (9 November 1832), echo the mocking edge of Berlioz’s
account given above. Janin intersperses his text with italicised passages from
the ballet’s libretto, comparing them either to the incongruous activities on
stage or ridiculing their general absurdity. Yet his critique of La Révolte au
sérail (6 December 1833) shifted the emphasis from preoccupation with the
tension between words and movement: “Some people claim that an idea
was necessary in order to make a ballet; these people are wrong: for making
a ballet, one only needs dancers, to be precise female dancers, since we have
expelled the male dancers . . . The good Mr Taglioni is the living proof . . . Mr
Taglioni has never any idea.” The critic continues that the choreographer
possessed something that surpasses ideas, for “he has his daughter”, the
incomparable Marie Taglioni, first among the gallery of female stars of the
romantic ballet. His remark “Elle danse, tout est dit” (She danses, that says
it all)7 disarmingly proves the insignificance of such secondary aspects as
“his dancing” or storylines. Over the course of Janin’s critiques it turns into
a genuine aesthetic appreciation of theatre dance as sufficiently expressive
in and of itself.
Much has been written about the disregard for the male dancer during
the romantic era, a disregard that is not necessarily reflected in the actual
casting lists, but finds its strongest articulation in the views of male critics.
Janin is the most outspoken among them, frequently spicing up his eulogies
of the female stars with utterances like “Under no circumstances do I grant
179 The romantic ballet and its critics

Figure 29 Caricature by Cham from Le Charivari, 1 August, 1858. “Monsieur Théophile Gautier
himself demonstrating to Madame Ferraris various steps of his ballet which she has only to
interpret for the public.”

a man the right to dance in public”.8 Only the most impressive male dancers
could challenge such an attitude. Was this suppression of the male dancer
the result of a conceited sense of omnipotence and possessiveness on the
part of the spectator? After all, Gautier declared that “the true husband of
an actress . . . is the public”.9
The vogue for danser en travestie engaged the audience in the complex
sexual subtext of the romantic ballet. Cleverly, through its playful under-
statement, travesty emphasised the femininity of the ballerinas. Even more
than Gautier’s, Janin’s reviews are rich in invectives against those who occu-
pied space on stage that could have been more pleasingly filled by the
Taglionis, Elsslers or Grisis. Yet the writing of the two balletomanes has
180 Lucia Ruprecht

more to offer. Their articles reveal a clear sense of the renewal of dance
that leaves the pre-romantic aesthetic behind. Echoing, at first sight, Castil-
Blaze’s tirade against mere mechanical virtuosity, Janin also directs his scorn
against the poses and tricks, the vanities and the “Un, deux, trois!” of the
traditional genre of the danse noble. Rather than opposing it to the powers
of pantomime he celebrates “the revolution in dancing” (24 August 1832),
which had been introduced by Marie Taglioni, “this young girl who has cre-
ated modern dance” (22 July 1832). In a description of the choreography,
which is rare in such detail, he praises the naturalness of her comportment,
the suppleness of her arms and legs, her ability to simply walk on stage,
and the absence of forced pirouettes and entrechats. He talks of an overall
quality of her performance which he calls the absence of “dance” in her
dancing (24 August 1832), a judgement which does not lack irony, since she
was the first one to excel in the most explicitly artificial aspect of the clas-
sical technique, the dancing on the pointe. At the same time he documents
Taglioni’s extraordinary mastery of difficulties, which made the technical
side of dancing disappear.
Janin and Gautier introduced an approach to ballet criticism which
united formalist sensibility with the emphatic language of poetry and an
outspoken appreciation of physical beauty. Apart from the mime scenes,
movement was freed from its bondage to a plot. It became available for
metaphorical readings that consolidated the much-praised abstract quali-
ties of romantic ballet, epitomised in the ballet blanc: lightness, evasiveness,
transparency and purity, titillatingly embodied by sensual dancers. Tech-
nique, dramatic framework and viewing situation were woven together into
a maze of artful eroticism. The ballerina reflected, through her profession
and through the specific characteristics of her dancing, the simultaneity of
erotic availability and elusiveness which was at the heart of the roles she
performed. The poetry of the dance begged to be matched by the poetry of
the feuilleton. The following review of Carlotta Grisi dancing Giselle, writ-
ten for The Musical World of 12 June 1847 by a critic called ‘D.’ who clearly
adopts the French style, illustrates this to great effect:

Nothing can be lovelier, nothing more ethereal, nothing more deeply


imbued with poesy than her Giselle. She looks a thing of air that would melt
away at your approach. It is impossible to imagine you could touch her, so
transparent, so fragile is her appearance. She floats along the stage. Scarcely
can you perceive the movements of her pretty feet. You might almost fancy
her an animated lily of the valley, but that she is endowed with the power of
locomotion. She shrinks instinctively at the approach of her mortal lover,
feeling him near even when her face is turned away from him. Catch her he
never can, though she makes not the least apparent effort to avoid him. It is
as though she were a magnet with repulsive, instead of attractive power.
181 The romantic ballet and its critics

And yet how entirely attractive and how entirely unrepulsive! How playful
and seemingly full of life and spirits! How prettily she coquets with her
lover, fleeing as he approaches and pursuing as he flies – but never coming
into actual contact with him; like hope long deferred, always seeming to be
realised but never really compassed. It is the very essence of imaginative art,
and proves Carlotta what we have many times pronounced her, the true
Poetess of the Dance.

Artists like Carlotta Grisi and Fanny Elssler complemented abstract and
sensual beauty with an expressiveness of mime that spoke directly to the
critics, as becomes evident in Janin’s review of Giselle on 30 June 1841.
He writes after quoting a few of Giselle’s “lines”: “That’s how she speaks,
and I can assure you that she speaks even better than that. Her gestures are
ideas, her dance is speech.” When an anonymous critic of the Revue et gazette
musicale de Paris states that “the gaze of the viewer, seized by the deepest pain,
rests on the pale and devastated face” of Elssler performing the role in 1843,
he not only acknowledges the performer’s talent for acting, but also indicates
a mode of spectatorship that engages the emotions and senses before the
intellect. Gautier draws a clear line between situations of intellectual and
sensual involvement by claiming that ballet was “directed only to the eyes”.10
Thus Janin writes with regard to the rational inaccessibility of the ballet La
Tentation,

the way in which you are able to judge all this is to surrender to the
impression of the moment, to be nonchalant enough not to scrutinise your
pleasure, to let yourself be guided, without resistance, by the painter, the
musician, the danseuse, to whichever place they lead you, in short, to allow
yourself to be happy: that’s the secret of ballet’ (27 June 1832).

This surrender simultaneously empowered the critic, since the non-


discursive allusiveness of the productions granted him the opportunity to
create his own spectacle on the blank page.
How to describe dance would become a major problem at the turn of
the twentieth century, but the critics of the romantic era were less troubled
by the problem of representation; they playfully evoke the ineffable, the
“grâce inexprimable”,11 and tease out the unsayable, the erotic charge of
the bodily display. While being important mediators of the new aesthetic
of dancing and seeing, they also took the liberty of producing their prose
on the occasion of, rather than on dance. They hid their amateurism under
a flamboyant style and anecdotal digressions. They mused; they described
physical features and revealed indiscretions. The explosion of journalism
in the first decades of the nineteenth century led to anxieties about a new
“self-interested instrumentality” of language as opposed to the truthfulness
of high literature.12 Short-lived, consumer-oriented, amusing but superficial
182 Lucia Ruprecht

and a potential battlefield of personal animosities, the feuilleton introduced


a new written discourse whose characteristics echoed common prejudices
about the entertainments which it depicted. Journalistic writing displayed a
form of overstated showiness that was perceived in gendered terms as an
unsettling feminisation of the masculine bourgeois domain of the word.
Louis Gentil, author of the manuscript “Les Cancans de l’Opéra”, failed
journalist himself, holder of an administrative post at the Opéra and most
sarcastic observer of the goings-on behind its wings, focused his scorn on
Janin. Gentil’s attack on the critic indeed recalls the latter’s disdain of the
effeminate male dancer: “He is provocative, mischievous, capricious, pretty,
coquettish, sulky, playful, amorous without energy, quarrelsome without
humour, irritable without anger, someone who speaks up without force,
mumbles rather than murmurs, and chirps without singing . . . Fly-catchers
of Paris, dilettantes of the big city, voilà Janin!”13 In a less spicy manner,
a Paris correspondent for the Musical World called ‘S.S.’ suggests that the
writer got as carried away by the flow of his own prose as by the ballerinas
whose praise he composed:

All the world here is full of your Pas de Quatre, about which the London
papers are so profusely eulogistic. The metaphrastic Jules Janin has
ventured an article upon the subject, which is more remarkable for its
verbosity than its truth. In apostrophising the four queens of the dance, he
says Taglioni depended upon her laurels, Carlotta Grisi upon her beauty,
Cerito [sic!] upon her freshness, and Lucile Grahn upon her talent, for
maintaining their positions before a British public. Nothing can be more
absurd. Janin dubs Cerito “The forbidden fruit” – “The unknown” –
because, forsooth, she has never appeared before the Parisians. That
admirable artist need not repine at this, since she has won her laurels from
the severest public in Europe. (31 July 1845)

The press coverage of the Pas de quatre (1845), showcasing four of the
most famous ballerinas of the day, not only documents the fact that crit-
icism itself had developed its own star system, complete with glamorous
hierarchies and resentments. It also represents a pinnacle in the cultivation
of the female star. The popularity of the piece, still fuelled by the fame of
outstandingly talented dancers, reveals the new dynamic of stardom that
relied increasingly on media spectacle, rather than on artistic abilities. In
La Fanfarlo (1847), Charles Baudelaire draws a highly entertaining and sar-
castic picture of a self-appointed critic who embarks on the journalistic
profession with the single goal of attracting the erotic interest of an infa-
mously seductive celebrity of the Opéra. He wins her love by bombarding
her with insulting reviews that only add to her fame:
183 The romantic ballet and its critics

From now on Fanfarlo was mauled every week at the tail-end of an


important periodical. It was impossible to assert or suggest that she had an
ugly leg or ankle or knee, as her skirts concealed them and every lorgnette
would have denounced such blasphemy. But she was accused of being rough
and common and tasteless, and doing her best to inflict German or Spanish
mannerisms on the French theatre, with her castanets and spurs and high
heels – apart from which, she drank like a trooper, was too fond of lapdogs
and the caretaker’s daughter, and suchlike dirty linen of domestic life which
are the staple food and titbits of our dimmer newspapers.14

Baudelaire might well have had in mind the danseuse terrible Lola
Montès, whose impulsiveness, extravagant life-style and illustrious affairs
made up for her lack of dancing skills and gained her the attention of jour-
nalists over several years. Gautier, too much of an admirer of good dancing to
yield entirely to Montès’s aura, left ambivalent accounts of her performances.
His colleague Pier-Angelo Fiorentino of Le Corsaire-Satan was among the
many victims of the Irish-Creole conqueror and fervently defended her. On
8 March 1845, he reports on Montès’s controversial appearance at Porte-
Saint-Martin, where she had been engaged for a programme of popular
national dances after having been expelled from the Opéra for allegedly
throwing one of her shoes at the audience on the evening of her debut.
Fiorentino writes: “From the moment in which the sound of the castanets
indicated the first bars of the cachucha, an avalanche of bouquets landed on
the stage and plastered the space in such a manner that the young dancer
could hardly make a step.” The cult of the dancer had literally taken over her
performances. Whether we see the notoriety of Lola Montès as one of the
indicators of the decline of the romantic ballet, which had become fixated
on its stars, or as an entertaining exception, it shows to what extent acts of
devotion and of discourse were involved in the aesthetic project. It seems
a fitting end to these glimpses of the world of nineteenth-century criticism
to evoke the image of a dance that is as much disfigured as it is enhanced
by the flowers of the journalists’ words. The vanities, the imagination and
the discernment, the calculation and the enthusiasm of the reviews give
evidence of a spectacle that developed alongside the main spectacles on
stage; the writings bear witness to the fact that the romantic ballet was a
multi-authored phenomenon.
16 The soul of the shoe
m a r i o n ka n t

Ask any young woman on her way to a performance of Giselle or Sleeping


Beauty what most clearly symbolises ballet and she will probably answer –
the skirt and the pointe shoe. She will not quote sentences from the story
and may recall only a few names of the characters. But after the performance
she will remember the ballet costume of the female dancer. If she ever had
ballet lessons she will reminisce about her first pointe shoes; she might still
have them in the attic. Why this cult of the costume? Has ballet no message?
Is it merely a flighty art form of beautiful lines, of flowing skirts and satin
shoes?
The history of these two items of dress tell us exactly the opposite. The
skirt and the point shoe represented a complete change in the nature of the
ballet as an art form. They have not always been there. When they were
introduced in the 1830s, roughly 180 years ago, they initiated a revolution
in artistic values and a fundamental shift in the attitude towards women in
public life.
How and why the tutu and the slipper achieved this pride of place in
ballet will be explained in what follows. Less clear is why much ballet today
still uses a dress code frozen in time.
This new apparel, which ballet lovers know so well, was an inextri-
cable part of the evolution of romanticism in the arts; during the nine-
teenth century romanticism gave dance its particular and enduring look.
When Théophile Gautier described Marie Taglioni in 1834 he wrote about
her delicate appearance: “She floats like a spirit in a transparent mist of
white muslin with which she loves to surround herself, and she resem-
bles a contented soul scarcely bending the petals of celestial flowers with
the tips of her rosy feet.” And then he continued with a description of
Fanny Elssler, the other great ballerina of the day: “She recalls the muse
Terpsichore with her tambourine and her tunic slit to reveal her thigh and
caught up with clasps of gold.”1 Ten years later he defined romanticism
in ballet through the figure of the seemingly weightless, fairy-like female
performer:
Mlle Taglioni dances the Sylphide: That says it all. This ballet opened the
door for a whole new era in choreography, and through it romanticism
entered the realm of Terpsichore. After La Sylphide, Les Filets de Vulcain
[184] and Flore et Zephyr were no longer possible. The Opéra was given over to
185 The soul of the shoe

gnomes, ondines, salamanders, elfs, nixes, wilis, peris, all those strange,
mysterious creatures who lend themselves so wonderfully to the fantasies of
the ballet master. The twelve mansions of marble and gold of the Olympians
were relegated to the dust of the scenery store, and artists were
commissioned to produce only romantic forests and valleys lit by that pretty
German moon of Heinrich Heine’s ballads. Pink tights remained pink, for
there can be no choreography without pink tights, and all that was changed
was the satin ballet slipper for the Greek cothurna. This new style brought
in its wake a great abuse of white gauze, tulle and tarlatan, and colours that
dissolved into mist by means of transparent skirts. White became almost the
only colour.2

Gautier’s evocation of romanticism in dance focused on the female dancer


in a mist of white, with dainty, rosy feet that scarcely bent a petal. (In printed
illustrations of the early 1830s the dancers indeed were often barefooted.)
By the time Gautier wrote the second description in 1844 these distinct
elements had become the norm: the female dancer in white gauze, tulle and
tarlatan dissolving into transparency with pink tights and satin slippers.
All traces of classicism with its allegorical figures had given way to a stage
covered by mysterious romantic creatures.
The “transparent mist” of tulle conquered the stage as the standard
costume of nineteenth century ballet, sometimes white, sometimes more
colourful, the nuances in colour helping to define female characters. But
white, as Gautier remarked, became the prevalent colour and eventually lent
its name to a particular genre in ballet: the ballet blanc, or white ballet. In it
a group of female dancers (mostly in the second act) would all be dressed
in the same white skirts, they would look alike, they would move alike,
supernatural beings, neither alive nor dead. The world that they evoked lay
beyond life. White became the uniform colour and the colour of the ballet
uniform.
Romanticism marked a real break in the staging as well as the costuming
of ballet. Though many of the elements that made up ballet emerged
gradually, the whole came across as new and it challenged tradition. For
Jean-Georges Noverre3 in the middle and late eighteenth century, the cos-
tume was but one part of the dance drama. It was one means among others –
story, design and steps – to express human emotion. Noverre thought a
reform of costume was necessary because he desired it to support rather
than hinder movement. The drama of the ballet d’action determined the
shape of the costumes. Nor could Carlo Blasis in his treatise on dance of 1820
imagine a dance costume detached from the ballet and its movement. His
descriptions and analyses focused on movement, technique, the acquisition
of muscular strength and flexibility. He assumed that costume followed
function. For Blasis, ballet was defined through the study of the legs, the
186 Marion Kant

arms or the entire body, with shoes and dress an afterthought. The illus-
trations in his treatise show us today what his students in 1820 wore: white
garments imitative of ancient Greek drapery, with a loose, sleeveless blouse
and a free flowing skirt. This costume was complemented by soft-looking,
snug shoes that allowed for easy and flexible stretches and extensions of
the feet.4
Blasis’s illustrations of his dancers reflect two aspects of the costume:
practicality of dress for the dancer (male as well as female) and closeness to
conventional fashion of the time. The craze for ancient Greek and Roman
culture during the French Revolution had left its mark in a fashion that ide-
alised tunic-frocks. From Napoleon’s seizure of power in 1799 to the 1820s,
high-waist “Empire” dresses became the rage, with the elegant Empress
Josephine Bonaparte symbolising the age and many women competing for
the highest waist. They too accented the body without imprisoning it. From
the 1830s female fashion focused more on the tight corseted bodice and
narrow waist with much fuller skirts. The relatively free and unrestricted
movement increasingly became difficult. Ballet dancers also pressed their
bodies into corsets (first reinforced with fishbone, later with steel wire) to
acquire the “wasp” waist.
Romanticism embodied a paradox, in dance as well as in the other art
forms. What today is regarded as the romantic movement’s “other-worldly
charms” and fairy-tale escape from reality depended on the real world’s
industrial progress and economic development. A few examples will illus-
trate the case: in 1832 Theobald Boehm, a Bavarian court musician, designed
a new type of flute in his own factory, which used interlinked rod-axles to
transmit the motion of fingers to remote tone holes. In other words, he
devised a system of metallic levers and rings by which the opening or clos-
ing of the tone hole could be controlled. The new instrument, just like other
new brass instruments invented around the same time, created a larger
sound, corrected and maintained the pitch and, through its greater variety
of instrumental volume and tone colour, made orchestras more expres-
sive (and much louder). This improvement depended on the rapid advance
of machine tools and metallurgy.5 Stage lighting also shifted significantly.
In the first half of the nineteenth century gas lighting was widely used.
The limelight, an early type of a spotlight, relied on the chemical process
of combining gases, oxygen and hydrogen with lime. Gas was particularly
dangerous for dancers whose gauze dresses frequently caught fire, often
burning the dancer, the entire stage or even the opera building. Fire preven-
tion and protection became vital in theatres. With the invention of the first
incandescent light bulb by Edison in 1879 the application of safer electric
lighting and particular mood setting could begin; unheard-of effects – the
gradual dimming of the chandeliers and the darkened auditorium – could
187 The soul of the shoe

be introduced and the audience would be enchanted and lured into a magical
sphere.
The romantic ballet costume, too, was a product of industrialisation.
The intensive use of muslin, tarlatan, gauze and tulle – all cotton products –
coincided with the growth of cotton manufacturing in Europe, above all in
England and France. Between the 1770s and the 1840s Britain increased its
cotton production tenfold. The technological innovations that introduced
industrialised processes into cotton production increased the availability
of the material: mass production initiated mass use. It also forced female
workers into the factory, turning them and their children into cheap labour.
In fact, industrialisation in the textile industry depended on the allocation
of female labour to factories. Cheap cotton made possible the special design
of costumes for every new production instead of recycling from previous
ballets, as had been the custom. It also introduced imitation and replication
of costumes: the costume of the soloist could be replicated in the costumes
of the corps. The great availability of material also facilitated the intro-
duction of costume detail from ballets into wider fashion and vice versa.
Fashion at rapid speed offered a new design of women’s wear for every
season.
The other two materials widely used in the making of clothing and
household material – silk and wool – lost out to cotton. Wool was not yet
as suited to fit into industrialised processing: spinning, bleaching, dying,
weaving, cutting and so on. Lingerie could now be made of the cheaper
cotton instead of silk, a relevant factor, above all, for the making of corsets,
which were deemed an indispensable part of female fashion on stage and
off in the nineteenth century.
The ballet costume up to 1830 had distinguished itself not by its complete
difference from ordinary dress but by its aping and appropriation of soci-
ety’s fashions. During the previous centuries costume on stage was either
randomly chosen (the dancer took what was available, however irrelevant
the costume might have been in relation to the role performed, especially if
she had to pay for it herself) or, more importantly, indicated the social class
of the wearer and her position within an aristocratic hierarchy of order. The
romantic costume shifted the focus from social or historic markers to gen-
der indicators because the stories focused on gender battles. New costume
and shift in storytelling happened at the same time and were closely linked
and both indicate a reconceptualisation of dance.
The romantic ballet costume, that mass of white cotton products, was
different from fashionable dress in the length of the skirts. The bare arms
or the low-cut neckline exposing the bust were not unusual in female fash-
ion – but exposing the ankles, the calves and or even the thighs definitely
was. No decent woman would have allowed her legs to be seen, neither bare
188 Marion Kant

nor stockinged: yet the dancers on stage revealed what was not supposed
to be visible in public. From the beginning, the romantic ballet costume
emphasised gender as problematic and sexualised the female characters,
and through them, the dancer herself. Let us look at one example, how
the tension between being and appearance was created and manipulated:
in 1831 in Paris Marie Taglioni exploded into attention in Robert le diable,
a grand opera by Giacomo Meyerbeer. The dancer played a nun who had
fallen for the devil and lost her faith.6 Taglioni and the other dancers entered
in costumes that copied nuns’ habits; when they discarded their nuns’ robes,
white skirts became visible. Within a couple of years those infamous skirts
would symbolise romanticism. Taglioni’s abbess and her nuns were not the
first to wear white on stage. But their costumes transformed the meaning of
the scene through the connotations of white and thus made a deep impres-
sion on contemporary audiences. The shrewd use of this colour created
a frightening uncertainty – satanic nuns whose white costume projected
innocence. It called into question the romanticised ideal of woman. The
colour defined the state of femaleness by confusing and reversing the moral
categories of good or evil. These costumes broke another tradition; they had
no equivalent in the world outside the theatre. The ballet costume was not
“historic” but completely new: no social class, guild or profession wore such
garb. Instead, the distance between the clothes worn on and off stage cleared
the way for a new kind of appearance with a radically new symbolic content.
It turned women on stage into an abstract category: the woman, the female
per se, with a certain ethical value system attached. The plots of romantic
ballets of the 1830s and 1840s involved murderous sylphs, rebellious female
slaves and amazons fighting patriarchal society. They portrayed women as
good or evil, caring or tempting, strong or weak but whatever their roles,
they were primarily women, women opposite men.
The colour white had undergone an important philosophical and aes-
thetic transformation after Isaac Newton broke down sunlight into separate
rays in 1672 and realised that colours were “Qualifications of Light . . . but
Original and connate properties”. They could be simple or compound and
they could be put on a wheel to show their physical relationships. But what-
ever was done, light was nothing but a physical phenomenon; it could be
refracted, analysed and measured. With this recognition Newton destroyed
the mystique behind light and colours. As he argued in his Opticks: A Treatise
of the Reflections, Refractions, Inflections & Colours of Light, practical exper-
iments could prove that white light was the additive result of the spectrum
of colours and black nothing but their absence.
After the shock of Newton’s reasoned analysis the German writer and
poet Johann Wolfgang von Goethe in the last years of the eighteenth
and the early years of the nineteenth century provided colour with a
189 The soul of the shoe

metaphysical dimension. For him, colours were expressions of energy and


they could affect the psyche and the soul. His colour theory, less popular
with physicists, enchanted and fascinated contemporary philosophers and
artists like Arthur Schopenhauer, architects Otto Runge and Gottfried Sem-
per and had a long-lasting impact on the humanities in the twentieth cen-
tury. Colour had meaning again; colour was not simply an appearance of the
physical world but a phenomenological problem that had to do with human
perception and thus could stimulate, even harm, psychological moods, dis-
positions and sentiments. They could make people feel light, animated and
happy or sad, agitated and sombre. One of Goethe’s three fundamental
emotional colour pairs was white–black, the ur-phenomenon representing
the ur-polarity of light/darkness out of which all other colours evolved. In
Zur Farbenlehre (Colour Theory) of 1810 Goethe wrote that colours were
“deeds of light, actions and sufferings”. Thus, during the early nineteenth
century the meaning of the colour white changed once more; now it stood
for innocence, virginity, chastity, clarity and purity.
By 1831 the evil nuns dressed in white created tension between the exter-
nal, beautiful façade and the internal corruption of character. Hence they
doubly deceived the audience: the dancer-nuns dressed in white signalled
innocence and purity but acted its opposite. Finally, beneath the white nun’s
habit emerged the transparent white habit of the dancer. If the stage lanterns
cast light from a particular angle her body appeared no longer covered at
all but exposed as what it really was: a dangerous sexual challenge. Without
the light shining through the fine cotton she could again act as the inno-
cent, reliable lover. Within an instant the dancer shifted from one extreme
to the other. Which part of the ambiguous costume/message was to be
trusted?
The sylphs, wilis, the shadows in La Bayadère, the ghosts and spirits,
the sylphs and immortal maidens, swans and butterflies of the nineteenth
century ballet, were all dressed in white or very light translucent colours.
They were fragile and tender creatures and at the same time destructive
forces. Nowhere does this become more clear than in the second act of
Giselle, which finally established the look and symbolism of the ballet blanc.
The wilis, those “dead bacchantes”,7 arrive in uniform white costume – tight-
fitting bodice and wide flowing skirt – instead of the faded colours of their
former existence as Gautier had originally suggested.8 They form battalions
and execute their dance movements in military-style precision. The white
gauze skirts and flowing hems soften the frighteningly martial image of the
disciplined troop of wilis, led by Myrtha, their queen. They constitute a
beautiful but threatening force out to destroy all men who cross their path.
They are vengeful beasts that bring death; yet they wear the colour white,
the colour of innocence and purity.
190 Marion Kant

The costume of the ballerina was developed to help tell the hair-raising
and horrific stories of women’s lives and their transformation from alive to
dead (or somewhere in between). The situations in which the heroines of
ballet narratives found themselves differed slightly from one another. The
geographical or time settings varied: some took place in a fantasy or fairy-tale
world, others in a recognisable reality. But they were united by one common
feature: the stories tested women in situations that questioned their very
existence. These questions always revolved around the fundamental issues
that had occupied citizens of European societies after the French Revolution:
what if women are given equal rights? What might we to do with these
emancipated women? What then happens to the familiar, traditional gender
divisions? What happens with love relationships? The essential questions in
these stories were so similar that the standard costume only needed small
amendments.
Throughout the nineteenth century we can trace the woman in white as a
romantic trope, in novels, paintings, operas, poems, but above all in ballets.
At the end of the century Edgar Degas painted the dancers many times.
He depicted them exhausted, happy, exhilarated, disengaged, training and
practising or slumped over on the floor – they wear the light skirts that were
introduced several decades before, now even shorter, cut just above the knee.
They wear them with colourful sashes or ribbons, flowers, embroideries
or other small ornaments and attributes – ballet added to the standard
ballet costume whose absence by then would have been unthinkable. If the
choreographies have not survived and the stories have been relegated with
their music to the compartment of less important works of theatre history,
the ballet costume invented in the same period still lives on. It is still the
dream of many a young girl to fit into that pretty white tutu9 and wear those
beautiful satin slippers with their toughened caps that are so cruel to the
feet.
The ballet shoe forms just as important and interesting a part of the
costume as the skirt. The transformation that ballet underwent thanks to
the employment of a special shoe is breathtaking. In fact, the development
of the ballet shoe from a soft, satin slipper to a device made according to the
latest technology with steel-hard cap is nothing short of a revolution in style,
means and equipment. We cannot date the invention of the pointe shoe, but
we can follow its changes that lead to the form and shape we know today. It
began with the choreographer Charles Didelot, who used flying machines
in several of his ballets at the turn of the eighteenth and early nineteenth
centuries to sweep his ballerinas across the stages of Paris, London, Milan
and St Petersburg. The performers were attached to wires to haul them from
one corner to another while their feet flashed across the ground. From the
early nineteenth century on, lithographs and etchings show female dancers
191 The soul of the shoe

Figure 30 Fanny Elssler’s pointe shoes.

hovering in the air. Their tiny feet, bare or clad, rest on petals or translucent
insect wings. They barely touch the earth if at all. Dancers such as Amalia
Brugnoli and Elisa Vaque-Moulin reaped applause for moving on the tips
of their toes before Taglioni or Elssler became famous for feet sheathed in
soft satin. This footwear was supple and where the toes touched the satin
they were reinforced with stitches. The soles were made of flexible leather.
Gautier noted:

The sole, which is very hollowed out in the centre does not reach the top of
the foot, but ends squarely, leaving about two finger-breadths of material
projecting. The purpose of this is to enable the dancer to perform pointe
work by giving a sort of jointed point of support, but as the whole weight of
the body is borne on this part of the shoe, which would inevitably break, the
dancer has to strengthen it by darning, almost as old-clothes menders do,
the heels of stockings to make them last. The inside of the shoe is lined with
strong canvas and at the very end, a strip of leather and cardboard, the
thickness of which depends on the lightness of the wearer. The rest of the
shoe is chevronned on the outside by a network of ribbons firmly sewn on,
and there is also stitching on the quarter, which is adjusted by means of a
little tag of ribbon in the Andalusian manner.10 (Fig. 30)

Those brief moments on the tip of the toe appeared so breathtaking and
amazing because technique beforehand had placed attention on other
aspects of dance. The technique that would allow a dancer to rise smoothly
onto point and linger – a moment that has come to define ballet as a genre –
had not yet developed. In Taglioni’s and Elssler’s generation, the arms had to
support the elevating body. (Cartoons of the time depict ballerinas flailing
their arms around, adding a sense of instability and hysteria as inherent in
women.) Gradually ballerinas learned to rise effortlessly or jump onto point
and remain on their toes. The shoes in the 1830s and 1840s did not yet have
192 Marion Kant

any of the hard reinforcements, characteristic of shoes made later in the


century. The development of the pointe technique in ballet thus depended
on the interaction between the materiality of the shoe, its constant updat-
ing (in the same way the technology of the flute was continually bettered)
and the revision of training methods. A. E. Théleur was probably the first
teacher to develop a technique for the strengthening of the foot. In his Let-
ters on dancing of 1832, the title an intentional reference to Noverre’s book,
Théleur came up with several exercises that resulted in a robust foot. He also
described the way in which dancers should raise or lower their bodies by
analysing what happened with the body weight in relation to ankles, heels
and toes, though he was still concerned with training more than perfor-
mance.
If we look at the printed illustrations of the time we see dancers with
anomalous anatomies: long swan necks, small waists, extremely long, thin
arms and miniature feet. It is remarkable that the entire dance style and
technique depended on a body part that was more and more hidden away and
distorted in images of ballerinas. The newly emerging technique depended
on the strength and durability of the feet; the tip of the foot had to carry
the entire body and make possible that technique. But the feet in popular
printed illustrations cannot carry the dancer, they cannot sustain a jump,
they cannot run the body across the stage. Under no circumstances can
those tiny feet make the body spin in pirouettes and hold it, even briefly,
on pointe to develop an arabesque. Why was the foot – the vital part of an
entire aesthetic and technical approach to the body – so misrepresented?
Because that new technique showed too much flesh? Was it to ease the shock
and make the women on stage as inoffensive as possible? Or to pretend the
provocation was not happening? Or because it was improper and tasteless
to associate the feet of delicate dancers with those flat and clumsy organs
of proletarian women who scuffled around in wooden clogs? Whatever the
reason might have been, the result was astonishing. On the one hand, the
printed illustrations demonstrate that the technique executed by women
was considered an amazing accomplishment; they also suggest that it was
an extreme problem. The visible aspect of ballet’s technical revolution was
made to disappear in illustrations of its performers, diminishing their feet
to the point of non-existence.
Many critics of ballet from the 1830s on objected fervently to these
new choreographic developments and proclaimed that the art form was in
decline: the ballerinas were no doubt brilliant technicians, but their tech-
nique was nothing move than mindless acrobatics. (That seemed true even
when the technique was engineered by male choreographers in the back-
ground though female choreographers were replacing their male colleagues
and most famous ballerinas choreographed their own scenes.) Ballet had
193 The soul of the shoe

been hijacked by women who wanted to show off. Men no longer had signif-
icant roles to play, so it seemed, and had become superfluous. Other critics
like Gautier or Jules Janin saw the new ballerinas as proof that men were nei-
ther beautiful nor useful. They argued that a feminised dance was the only
tolerable one. But “hommes de lettres” like Gautier also believed in female
emancipation, championing the beliefs of Count Saint-Simon, who fought
for women’s suffrage. The opponents replied with fury: women had no
intelligent minds of their own. Philosophers could reason and science could
prove that women were less intelligent than men. Female technique and its
technology (costume and shoes) had to be treated with scorn. Through the
influence of these hostile critics romantic ballet was declared to have had
a short life span and after 1848 was not worth remembering. Then an odd
reversal in historiography and criticism took place: in the late nineteenth
century the Imperial Ballet in Russia was celebrated as a revival of ballet. That
same romantic technique and costume, the tutu (now shorter)11 and the
ballet shoe (now more supportive), suddenly emerged like a phoenix from
the ashes. For a man, the choreographer Marius Petipa, had clearly taken
control again and there was nothing to fear. Since after mid-century the
French romantic ballet, it was said, “perished”, its Russian revival required
a new name: general usage replaced romantic with classical, a confusion of
terminology and concepts that has never been resolved.
From the 1860s on and certainly by the late 1870s the pointe shoe was
no longer a slender leathery slipper but an instrument with which ballerina
Pierina Legnani could turn thirty-two fouettés pirouettes in Swan Lake. It
had a harder box to sustain the ballerina on point. The front of the shoe,
the box, had been reinforced with layers of glued paper or burlap, allowing
the dancer to stand on it for a much longer time. She had been trained
in countless classes to jump onto pointe or rise gradually,12 whatever the
choreographer demanded. The shank, also inserted to support the dancer,
was made of one or a split piece of leather or again glued paper (in the
twentieth century of plastic). That is in principle what the shoe looks like
today. Miniscule nails tacked the textile part of the shoe to the sole and the
box (today, glue is preferred). Both were so hard, stiff and inflexible that
the ballerina had to “break in” the shoe before she could dance in it. That
literally meant damaging it: bending and making it more flexible by using a
hammer or simply closing the door on the shoe. Of course this was a tricky
moment as the shoe had to retain some stiffness; if it was rendered too soft
it would be useless. Until recently, leading ballerinas would use up several
pairs of pointe shoes during one performance. Improvements focused on
making the shank and the box more durable, on increasing their flexibility
and reliability. There was less need for a brutal “breaking in” process and
the average shoe now supported the foot much better.
194 Marion Kant

All of this technological progress had thus far never questioned the pro-
duction process but merely enhanced features that were already there. This
might change with the latest, truly innovative approach (only three years old
and just patented): it treats the pointe shoe as a prosthetic tool. The approach
to making a pointe shoe is slightly modified by looking at it as a “missing
limb”. It becomes more than an aid (that the traditional pointe shoe is)
by providing something the foot cannot naturally do. It enhances technical
possibilities by rethinking weight distribution, shifting weight away from the
tip of the toe. The shoe is cast as a socket: the foot of the dancer is wrapped
in plaster-impregnated gauze over which an elastic sock is pulled to hold
the gauze tight to the foot. Over gauze and sock a plastic bag is placed to
keep the layers clean. The entire foot-gauze-sock-plastic-bag-contraption
is positioned in a pail, which is filled with sand, which provides a semi-
weight-bearing environment while the splinting material is drying. The
result: a negative impression of the foot in action made of flexible material.
From the cast a flexible inner liner is custom-made, together with an outer
shell, which can be standardised. The outer shell is also reinforced with a
compact block and a tongue, which make point dancing possible by shifting
the weight of the body away from the front of the foot and distributing
it evenly upwards. In addition, the polyurethane and polypropylene mate-
rials diminish the impact of jumps – just like a lightning rod they direct
the force away. Finally, inner liner and outer shell can be beautified with
any colour satin. The new point shoe looks like its traditional cousin but is
much healthier and lasts considerably longer (see Fig. 31). The drawback,
of course, is the high costs of the individualised production process.13
By the twentieth century ballet costume and point shoes had been pub-
licly accepted as symbols of the art as a whole. They carried all the repre-
sentational implications for the female body. Around 1900, a new ideal of
femininity led to an attack on ballet in the name of “nature” directed at
its costume. The skirt and shoes became literally the critical points in the
emergence of the so-called “modern” or free dance. Modern dancers around
1900 began articulating themselves and their revolt by discarding the cos-
tume. Out went corsets, pointe shoes and tutus, and in exchange came bare
bodies and bare feet. Isadora Duncan declared that a free woman would have
to free her body from traditional clothes. The body could be exposed in a
“natural” way as long as the movements themselves embodied ideas. Nudity
was more desirable than half-clothed bodies – indeed, it was more natural
and less sexualised than transparent materials that suggestively veiled ballet
dancers. Duncan’s designs for her stage costumes harked back to the tunics
of ancient Greece and Rome and the new loose garments were implicitly
associated with Plato’s, Aristotle’s and Socrates’ ancient ideals of freedom,
195 The soul of the shoe

Figure 31 Pointe shoes 1950–2005 from the Soviet Union, Bulgaria, Germany, Great Britain and
the United States.

republicanism and democracy. That, of course, had been exactly the same
ideal with which the French Revolution had introduced its own tunics as a
dress code. But neither Duncan nor any of her contemporaries recognised
the irony that lay in the appropriation of exactly those notions that had once
made ballet modern. For the modern rebels, ballet was outdated; it was no
longer what it had been for Gautier (a critical analysis of an oppressive soci-
ety) but instead misunderstood as a proponent of patriarchal dominance.
The costume, the tutu and the pointe shoes, were tools to stifle women and,
accordingly, any dancer who tolerated such dress made herself part of the
repressive system. Even today advocates of modern or post-modern dance
caricature ballet because of its costume and draw inspiration from the satire
that the misapplication of toe shoes and frilly skirts offer. A modern dancer
might not have been able to overturn the principles but at least she could
change the exterior. By rejecting the look, she hoped to affect the reality.
Whether the reformers of dance successfully introduced twentieth century
modernity into ballet or whether ballet would have reformed itself on its
own remains open to interpretation. But from the 1910s and 1920s onwards
dancers in leotards or body stockings as gender-equalising, androgynous
costumes increasingly populated half the stage, while traditional ballets and
ballerinas occupied the other half. In the twenty-first century, the great bal-
lets of the past still draw huge crowds to the opera houses of the world
in spite of (or because of) their now stylised costumes and strict body
movements.
196 Marion Kant

Three questions arise at this point, none easy to answer:

1 How do we understand the romantic ballets today?


2 How do we see the costumes and the shoes?
3 Why have the romantic (turned classical) tutus and pointe shoes remained symbolic
of the ballet as such?

The continuities in dress are remarkable. From the 1830s to the present,
frocks with their tight bodices and wide fluffy skirts, the little sylph or
butterfly wings, the satin shoes that hide the hardened pointe, the garlands
of flowers, the parted hair, firmly combed back into a bun, have appeared
over and over again. All those ingredients returned in every ballet and they
were merged into the standard costume. There is a very clear, uninterrupted
line between the Sylphide of 1832 and the sylphs in Chopiniana of 1908. Does
that matter today? Does anybody bother to consider the meaning of these
well-known costumes? Do mothers and grandmothers know what they are
doing when they take their daughters to see a romantic ballet? Probably
not. What once scandalised audiences in the Parisian opera has become
quaint, unchanging and, above all, safe. Modern audiences marvel at the
same movements of today’s great ballerinas as the audiences of Gautier’s
time cheered Taglioni and Elssler. The magic of movement, plot and music,
all well known and unthreatening, still draw comfortable crowds.
Do these audiences of a modern performance of Giselle know that they are
about to confront an existential conflict about the place of women in society?
Do they remember that romantic ballet used to ask many uncomfortable
questions and challenge moral values?
Even if they wanted to do so, explanations and analyses which might take
them behind the packaging are hard to find. It may well be that contemporary
choreographers and audiences have made themselves accomplices of those
forces, which undermined the emancipation of women in the 1830s and
still do so today. Nevertheless, romantic ballet is not about fairy tales. It is
about the very rough reality that women face. And the costumes once were
able to tell that story of conflict.
What today’s audience sees and demands to see is the ballerina in a
costume that has been disconnected from its own history. Romantic ballet
was once radical; the contrast between the weak and dependent female of
the narratives and the strong female executing technique reflected the social
status of women. The treatment of women able to create and perform their
own technology echoes the discomfort and anxiety about women taking
control and demanding equality, whether as suffragettes or factory workers
in the cotton mills. Romantic ballet has been idealised and reduced to soppy
stories that are seldom told in historical context. Romantic ballet is not
197 The soul of the shoe

merely about unhappy love affairs but about gender relations on the most
profound level. What the audience gets with the tutu and the pointe shoe
is a whole ideology, packed into the libretto and the music as much as the
costume as an entity. That ideology did indeed evolve around the female
performer and the ballerina became its carrier; she made it visible through
the things she wore.
pa rt i v

The twentieth century: tradition


becomes modern
17 The ballet avant-garde I: the Ballets Suédois and
its modernist concept
e r i k n ä s lu n d

Serge Diaghilev’s Ballets Russes arrived in Paris in 1909 and immediately


turned ballet into a high fashion in the Western hemisphere. But in 1920 a
rival introduced itself to Paris: the Ballets Suédois. This company dared to
perform in the Théâtre des Champs-Elysées, home ground of the Russians
before the First World War. The newcomers quickly established themselves
in the French capital and challenged Diaghilev’s concept of avant-garde
ballet. They took modernism to a new level and defined avant-garde not
through the brilliance of ballet stars or spectacular extravaganza but through
a concept that extended beyond all known theatrical conventions. Through-
out its brief and hectic existence from 1920 to 1925, the Ballets Suédois were
constantly criticised for being not Swedish enough as well as too Swedish,
and above all for not presenting real ballet. In many ways the critics were
right. The company was more international than Swedish in its character
and instead of reinforcing the balletic tradition it searched for new paths in
dance. The company’s name pointed to its Swedish founders and its most
prominent dancers. Its existence had come out of the art collection of the
Swedish aristocrat Rolf de Maré, who had assembled significant modernist
works by European, and particularly French, artists during the 1910s. He
himself was the favourite grandchild of another great art collector in Sweden,
the Countess Wilhelmina von Hallwyl. Around 1900 she was considered one
of the wealthiest people in her country. Her grandson inherited her fascina-
tion for art and for collecting it. Both grandparents provided Rolf de Maré
with the financial independence to pursue his exquisite but expensive inter-
ests. This financial independence also allowed him to feel secure enough to
live his life as openly gay.
A turning point in his life was the encounter in 1912 with the Swedish
painter Nils Dardel who was exactly the same age. He had settled in Paris in
1910 to study with Matisse and quickly made friends with leading painters,
poets, composers and art dealers. Dardel was to cultivate his new friend’s
taste in art and he guided Maré towards modernism in general and cubism
in particular.
Dardel began to buy works of art, paid for by his friend; the collection
included established French masters such as Monet, Bonnard and Seurat.
[201]
202 Erik Näslund

Above all he turned to the artists who were experimenting with new shapes
and forms and called themselves cubists, to Braque, Picasso and Léger.
The high society to which Rolf de Maré belonged sneered and laughed at
the acquisitions. He soon understood that there would be no place in his
home country for his newest and boldest initiative – the founding of a
dance company based on novel artistic concepts. It was the encounter in
1918 with the young dancer of the Royal Swedish Ballet Jean Börlin that
spurred on this idea. Börlin had been inspired by and later studied with the
famous Russian choreographer Michel Fokine, who fell out of Diaghilev’s
favour and choreographed his last work for the Ballets Russes in 1914.
Fokine had visited Stockholm in 1913 and 1914 and had stirred up the dull,
conventional repertory at the Swedish Royal Opera House. Now Börlin also
began to choreograph; but because Stockholm was conservative he worked
on a smaller, more careful scale, and invented pieces for himself or himself
with a partner and organised tours to the Swedish provinces.
Fokine’s tremendous success in Stockholm had inspired a project to
establish a company with Swedish dancers and a repertory based on Swedish
themes; he was to become artistic director, but the outbreak of the First
World War in August 1914 made these initial plans obsolete. Fokine left
Russia for good in the spring of 1918, after the effects of the Bolshevist
Revolution were making themselves felt; he first came to Sweden and then
settled in Denmark. After de Maré had met Fokine through Börlin the pre-
vious plan for a modern Swedish ballet company abroad must have been
transplanted into his mind. By then the two men had no intention of involv-
ing the Russian master any more; instead the project mutated into providing
support and opportunities for the budding Swedish choreographer.
At the same time, the development of the Ballets Suédois also was to be
directly linked to Rolf de Maré’s art collection. The inspiration for ballets
would have to come from painters and paintings. Several of the artists
represented in de Maré’s collection – Pierre Bonnard, Nils Dardel, Fernand
Léger – would also be engaged as stage designers. The paintings could be
turned into a life form, into a new dance genre. “I wish to transfer something
of the beauty found in these paintings into dance,” as de Maré once explained
it.1 He soon confessed that it was his desire to see his El Greco paintings
turned into dance motion because they had made him think of founding a
ballet company.2 The violent power of an El Greco painting or the sense of
moving shapes in works by Léger such as his L’Escalier or Le Soldat, were to
be transferred onto the living body and into movement. The repertory of
the Ballets Suédois thus very concretely reflected de Maré’s art collection.
The El Greco paintings were developed into the piece El Greco, which was
neither a ballet nor a dance drama but a visualised poetic vision of the works
of the Toledo master; the expressionism of the Greek-Spanish painter was
203 The Ballets Suédois and its modernist concept

rediscovered during this period and mirrored in a dance theatre piece called
La Maison de fous (The Madhouse). Rolf de Maré’s large Léger collection
found its natural continuation in the two Léger ballets Skating Rink and La
Création du monde. Pierre Bonnard, one of several representatives of French
impressionism in the de Maré collection, was commissioned to design Jeux.
But Picasso, whose work de Maré also collected and who was a friend, could
never be convinced to work for the Swedish company. Most likely he was
bound by contracts he had received from Diaghilev.
In a metaphorical sense one can say that de Maré not only bought indi-
vidual works of art or commissioned new ballets but made an entire modern
art movement for himself. With the dance company he created his own liv-
ing art gallery. His wealth made him independent from box-office success;
he was his own patron and did not bother about any whims but his own.
In 1924 the painter Francis Picabia – one of de Maré’s many
collaborators – described the impresario’s unique contribution to mod-
ern art by stating that he had been of much greater importance than even
he himself thought. Maré, Picabia said, made it possible for an entire cos-
mopolitan generation in Paris to work with a purpose, to express itself
freely and not have to give in to paralysing worries about the demands of a
capricious audience.3
Yet the opening programme of the Ballets Suédois on 23 October 1920
was devised to please with a new version of Nijinsky’s and Debussy’s Jeux,
the picturesque Spanish scenes Iberia, designed by the much-loved painter
Steinlen, a light-hearted evocation of Swedish folklore entitled La Nuit de
Saint-Jean and the exotic Derviches, designed in the style of an Indian minia-
ture; this was indeed a safe way to programme and perform in the legendary
footsteps of Fokine and the Ballets Russes. It was supposed to establish the
company as a legitimate artistic enterprise. It did exactly that and in addi-
tion it made clear that the works of contemporary painters and composers
would be closely related to a choreographed spectacle.
If we compare this fairly conventional opening with the very last new
programme that the company presented in Paris on 4 December 1924 we
realise that the Ballets Suédois underwent a remarkable aesthetic develop-
ment: from a relatively conventional focus that defined choreography in the
traditional way the company moved to the presentation of stage art that had
disobeyed and dissolved all rules and accepted criteria.
When Relâche by Picabia and Satie was premiered on that last evening,
together with the Léger works Skating Rink and La Création du monde, the
Ballets Suédois had become much more than the name indicated: out of a
ballet company had emerged the concept of multi art and total theatre.
The revolutionary development the Ballets Suédois gradually underwent
was only possible with its international circle of collaborators, all of whom
204 Erik Näslund

had been enticed to collaborate after the first autumn season in Paris. Again
it was de Maré’s close friend Nils Dardel who acted as an intermediary. It was
to a great extent his close circle of friends, many of whom he introduced to
the company, who came to inspire the Ballets Suédois and reflect publicly on
the different art movements of the 1920s. It must be stressed that the ideas
and concepts tried out by the Ballets Suédois did not exist in a vacuum. Sim-
ilar impulses could be found among the Dadaists, the Russian constructivist
avant-garde or the Bauhaus group. The many artists working within and
around the Ballets Suédois functioned as agents and connectors. They were
tempted by the stage as it offered an unique possibility to try out ideas on a
larger scale, which they otherwise lacked. And the ballet gained innovative
partners. Let us, for instance, take a painter like Fernand Léger, whose whole
oeuvre strove towards a large, spatially minded art. Throughout his life he
was a mural painter in soul and heart, although he rarely had access to that
format. Considering this passion, one understands why he quickly became
involved with the activities of the Ballets Suédois. Another friend of Dardel
was Jean Cocteau who became an ardent propagandist for the Swedes and
who generated new contacts. His attraction to the newly established com-
pany was different from Léger’s. Cocteau’s cooperation with Diaghilev in
Parade in 1917 had disappointed the poet. Cocteau had imagined a fusion of
dance and poetry; the Russian impresario had destroyed this vision by elim-
inating Cocteau’s text. But the possibility of combining two artistic media
was the reason why Cocteau had wanted to collaborate in the first place.
Therefore he wanted to create his new piece, Les Mariés de la Tour Eiffel,
with the Swedes instead of Diaghilev. Here was a young, un-established and
unspoilt company, and one in which an authority like Diaghilev would not
have the last word. Cocteau realised that he was himself the authority and
that he could leave his mark on a production. Les Mariés, premiered in June
1921, was permeated with Cocteau’s artistic taste and personality. The anti-
naturalistic and ironical style, which Cocteau already had tried out in some
previous works, pointed towards the theatre of the absurd; his multidisci-
plinary approach was reflected in a mixture of dance, spoken drama, music
and visual arts, which created a new genre. Although Börlin was responsi-
ble for the choreography, Irène Lagut the décor, Jean Hugo the costumes
and the masks and the young composing group labelled Les Six wrote the
music, there is hardly a detail in the piece which does not carry Cocteau’s
“taste”. This way of sharing the work became typical of the openness of
the Ballets Suédois. Not only did the company venture further into the
realm of experimental dance and theatrical spectacle, it also surrendered
control of production aspects to the visual artists it employed. The involve-
ment of Cocteau, Léger and Picabia in their respective productions extended
far beyond ordinary stage designing and included and then merged concept
205 The Ballets Suédois and its modernist concept

with visual expression and invention of movement gestures. Börlin seems


to have possessed an almost chameleon-like gift in adjusting to the different
concepts and styles, to the different “isms”, which the various collaborators
represented.
At the time it had become fashionable to cross the borders between
the arts as Cocteau did in Les Mariés. In music, composers moved briskly
between different genres; popular musical forms were adopted and adapted
by “serious” composers and in general artists were looking for popular art
forms as circus, music-hall, bal musettes etc. It was a reaction to established
bourgeois academic art and nineteenth-century romanticism. The use of
quasi popular forms of expression was often intended as provocation, as
was the case with Erik Satie and the group Les Six, all of whom collaborated
with the Ballets Suédois. The mixture of high and low art and anti-romantic
and anti-naturalistic attitudes was typical of Paris in the 1920s and also
left its mark on the repertory of the Ballets Suédois. The company was
often only too pleased to provoke and annoy its audience. The polemical
approach reached its climax with the final production in December 1924 of
Picabia’s and Satie’s Relâche, which in its anti-theatricality, its simultaneity
and disintegration of all forms pointed to the happenings of the 1950s and
1960s.
This kind of simultaneity – where disparate as well as interrelated actions
and expressions took place at the same time but independent of each other,
with which futurists and Dadaists had experimented – was one of the imme-
diate results of a concept that let various artistic forms work together or
confront each other, with idiosyncratic effect. The search for a synthesis
of movement, dramatic action, poetry, music and painting was even more
evident as an artistic goal in the Ballets Suédois than it had been with its
rival, the Ballets Russes. Jean Börlin often expressed his dream of creating
a harmonious fusion of all these elements, forming a whole just like in a
painting. This fusion was sometimes labelled “plastic theatre”; ballet in its
traditional meaning it certainly was not. Hence the discussions around the
Ballets Suédois were often very lively. The advocates of the classical ballet
sharpened their pens, whereas those who were more sensitive to the com-
pany’s ambitions tried to find new ways of appreciating the productions:
“Ballet? No. Play? No. Tragedy? No. More of a secret wedding or ambigu-
ous cross between the antique tragedy and the New Year’s revue, between
the ancient chorus and the numbers in a music hall,” as Cocteau himself
formulated the problem regarding Les Mariés.4
Plastic theatre, rhythmic pantomime, animated pictures, paintings with
gestures, sculptures in space, plastic poems . . . the attempts to find terms
that would do justice to the Ballets Suédois’s fusion between dance and
pictorial art were many. “Ballet becomes more and more expressive,” Börlin
206 Erik Näslund

explained in an interview. “Some years ago it was only dance. Now it not only
interprets the modern life, but also epochs in the past and the characteristics
of different countries and peoples.”5
It could be argued that it was with the Ballets Suédois that urban, modern
life entered into the field of dance. Börlin and de Maré wanted something
more than pure dance and technical skill, they wanted to give expression
to a thought, an idea. Often outer action had to be disregarded in order
to interpret the depth of the human soul. What they strove to do, de Maré
explained in an interview, was to tell a story of real life by dancing and
mimicry much as the actors in a play do with words and gestures. “Every-
thing should convey a meaning, every movement, every phrasing in music,
every scenic picture . . . all the ingredients should collaborate to create the
same expressiveness as spoken theatre . . . It is a new and I think fascinating
adventure into the realm of art,” de Maré said. “If only we could find a
new word for ‘ballet’ we would be better satisfied. The word ‘ballet’ is not
comprehensive enough for the work which we are doing.”6
The attempt to go beyond “ballet” made the Ballets Suédois create a
genre that was often strongly symbolic and expressionistic. The full impact
of expressionism came to its fore in works like El Greco, La Maison de fous and
Skating Rink. Maison de fous (1920), set in a madhouse, where the dancers
as its patients expressed grotesquely exaggerated emotions, represented an
important step in the evolution of modern ballet. Dance was no longer
limited to expressions of romantic longings or exotic beauty; it was also
capable of finding means to express the horrific aspects of contemporary
society. Skating Rink (1922) introduced an urban as well as modern aspect to
ballet, presenting a cross-section of human society, including workers and
fashionable types, engaged in a popular pastime. The circular arena stood for
the panorama of life, which was in itself the material of art. Fernand Léger’s
decor with its geometric shapes and figurative elements provided a pictorial
equivalent to the complex pattern of motions created by the dancers onstage.
Léger’s design resulted in the first completely abstract ballet design to be
realised; he followed earlier experiments by Appia, Craig, Malevich, Depero
and others. There was nothing decorative or virtuosic about the movement
design either. Börlin had developed possibilities of plastic expression in the
skating rink and had furnished them with angular gestures reflecting their
frustrated inner selves and placed them in relationship to the geometrical
patterns of the background (see Fig. 32).
No wonder, perhaps, that some critics reproached the Ballets Suédois
for so poorly representing “pure” dance. Some critics complained that the
dancers were too often squeezed into the straitjackets of futuristic and
grotesque costumes. But de Maré and Börlin defined the word dance in
a different way: against traditions and against the advocates of classical
207 The Ballets Suédois and its modernist concept

Figure 32 Skating Rink, Les Ballets Suédois, 1931.

ballet who quickly turned against them. Börlin’s teacher Michel Fokine
found the entire concept wanting: “No dance!” he summarised laconically.7
Others took offence at what they saw as a Scandinavian, or rather Germanic,
bad habit: the urge to want to say “one million words” in every gesture, in
every pose.8 But there were critics who were fascinated by the new “Swedish”
way to convey “dramatic feeling” to an audience. “Their productions are
something between a ballet, a mimed play, and a series of tableaux vivants,”
explained an enthusiastic critic in The Spectator after the company’s London
debut in December 1920. An ordinary play, the reviewer said, conveyed itself
by actions and words. Here the music had been substituted for the spoken
words and rhythmic gesturing for the actions. It differed widely from the
successful Russian style in that the characters often did not dance at all; it
differed from mime play in that the music formed an essential yet invis-
ible part. Indeed, the connection between movement and music – things
seen and things heard – appeared perfect. The players – this critic could
hardly call them dancers – all seemed to swim on the broad stream of the
music.9
Dance was only one part of the total image that the Ballets Suédois strove
to create and often it was not a matter of dance in the traditional meaning.
“In classical dancing”, Börlin once explained, “there is no correspondence
between the movements of the legs and the movements of the upper part of
the body. . . In searching for a remedy, it has been natural to try to transform
dance into pantomime.”10 The Swedish choreographer sought an original
208 Erik Näslund

and modern way of moving that stood in close relation to the contemporary
pictorial art. In his new idiom, Börlin incorporated impulses from classical
technique (especially Fokine), modern dance forms and pantomime. Such
synthesis was guided by the experiments of visual arts. What he had in mind
were moving images, set in motion by the many layers of a painting. In no
better way could this aim have been demonstrated than in La Creátion du
monde (1923), based on an African legend of how the world was created.
Léger was once more able to use his ideas of an art work based on geometrical
motifs in movement. Darius Milhaud’s jazz-infused music surrounded the
action like an atmospheric sound design. Music and movement existed
independently of each other and yet together, in the same kind of universe
and based on the same principles that Merce Cunningham and John Cage
began to explore and adopt in the 1940s in the United States. The dancers
were considered objects and became an integral part of the stage setting.
They disappeared in and behind costume and design elements, which were
set in constant movement. Thus the image of a painting was crafted, which
constantly changed before the eyes of the audience. Léger’s and Börlin’s
work pointed towards the form of total movement theatre later taken up
by, among others, the American artist Alwin Nikolais: the dancer was and
is only one part of a visual play with abstract forms, patterns, colours and
sounds. La Création du monde presented itself as multi art, mixed media
or movement in art work decades before these concepts had acquired the
names that are now familiar to us. The dancers had been dehumanised
and transformed into sculptures crawling or walking on stilts. This kind of
“dance” had not been imagined before. As the company motto once went:
we are the dancers who dare not dance!
La Création du monde was also the first jazz ballet, followed in the same
programme by Within the Quota, for which Cole Porter composed his only
“serious” score. Modernism was clearly linking popular musical forms and
jazz, which became highly fashionable in Paris after the First World War, with
“serious” music. Although there had been touches of an American sound
beforehand in both Parade (1917) and Le Boeuf sur le toit (1920), Within the
Quota was really the first ballet with an all-American theme. This satire on
the American way of life reflected the contemporary interest in cinema and
a cinematic quality, flashes of images, spread through the ballet (see Fig. 33).
The modernity of the Ballets Suédois consisted in reflection upon contem-
porary life and art. In a more symbolic form a similar inner contemplation
on existential themes was achieved in another work, L’Homme et son désir
(1921), based on a poem by Paul Claudel and with a provocative musical
score by Darius Milhaud. It was a plastic drama focused on “a man in his
nakedness”, who was born in the Brazilian forest, a setting that provided
means to explore animalistic or atavistic instincts. The striking stage design
209 The Ballets Suédois and its modernist concept

Figure 33 Within the Quota, Les Ballets Suédois, 1931.

by Audrey Parr consisted of four constructivist levels representing the dif-


ferent planes of symbolic action, which all took place simultaneously. Once
more, this was neither ballet nor dance, but a gestured poem, in which
the plastic expression of the poetic image correlated to Börlin’s ideas of
translating painted images and words into dynamic movements.
This urge to create a new idiom, a new way to express movement, can
be compared to the development of modern dance in the US and in Europe
during the 1960s and 1970s, when once more the boundaries of what con-
stitutes dance were being investigated – this time by post-modernists. The
Ballets Suédois can hence be regarded as one of the most important prede-
cessors of experiments in modern dance later in the twentieth century. In
terminology and perspective on dance traditions, techniques and content
we are, for instance, not that far away from the 1965 manifesto of the radical
American choreographer Yvonne Rainer, in which she said NO to most of
the conventional notions of theatre and dance. She proclaimed her refusal
in connection with her own performance of Parts of some sextets (1965).
It is therefore telling that the Ballets Suédois found perhaps its staunchest
support in Central Europe where it presented its works during extensive
tours. The idea of dance outside the ballet tradition that had gained strength
after Isadora Duncan’s appearances in European countries around the turn
of the century had made possible a completely new genre: modern dance.
This was no longer ballet but decidedly different in all of the most important
aesthetic aspects. Modern dance per se made it its goal to work in opposition
210 Erik Näslund

to and outside the theatrical tradition of ballet; this new dance genre pursued
the search for new and unconventional forms to express movement. “These
Swedes explore with their living material [the dancer], what earlier only
painters on their canvases and modern musicians with their pianos and
orchestras have done,” exclaimed Hans Siemsen in Die Weltbühne. “They
create expressionistic, cubistic yes even dadaistic ballets . . . You can say that
these Swedes have a different perception of the notion of dance than what
you are used to and what tradition prescribes. They not only dance dances,
they dance whole dramas.”11
The earthliness, in other words the contemporaneity and sensitivity to
social reality of the 1920s that often characterised Börlin’s movements, began
to appear in the choreographies of the German Ausdruckstanz. The dramatic
impetus and the reinvented mimetic gestures of Rudolf von Laban’s dance
dramas for his Kammertanz Theater in Hamburg from 1923 onwards and
the dramatic-pantomimic expressions of Kurt Jooss, who also tried to fuse
classical and modern idioms, certainly can be compared to the achievements
of the Ballets Suédois. The Börlin style is even more evident in the early
works of Ninette de Valois. She not only recreated some of the Ballets Suédois
repertory, but her theatrical style, the psychologically or socially meaningful
gesture, was very close to what Börlin tried to achieve. Job, for instance,
created in 1931 and based on William Blake’s famous visionary illustrations
for the Book of Job, would not have been possible without Börlin’s El Greco.
In a characteristic Börlin manner de Valois described Job not as a ballet but
as a serious dramatic production based on theatrical artistic conventions in
general. The Hungarian Aurel von Milloss was another choreographer of the
same generation as de Valois who openly conceded that he was influenced
by Börlin. He too re-created a number of works from the Ballets Suédois
repertory; above all it was the development of his dramatic dance-mime
style that owed a lot to Börlin.
It is always difficult to assess exactly the influence of one movement on
another, of the Ballets Suédois on later modern dance and ballet styles and
concepts. But we can certainly compare the concepts of the Ballets Suédois
to those of the emerging modern dance movements. And there we do see
interesting cross-overs, continuations and developments that owe their core
ideas to the earlier Ballets Suédois. It is in reality more difficult to distinguish
and recognise exactly the stylistic impact and influence; can one speak of real
influences or do we have to settle merely on similarities in aesthetics instead?
Although many critics and their opinions split into decisive groups, they all
unanimously praised the way in which the Ballets Suédois supported French
art in general and music in particular. It appeared almost embarrassing
that a Swedish company supported and spread French music around the
world. By its example the Ballets Suédois also put pressure on Diaghilev and
211 The Ballets Suédois and its modernist concept

during the 1920s the legendary impresario felt obliged to turn away from
his Russian heritage and employ French painters and composers instead.
The fresh vision of the Ballets Suédois also challenged Diaghilev in another
way: modernism had to be rethought and Diaghilev was forced to compete
if he wanted to retain his avant-garde reputation.
Most of Jean Börlin’s ballets represented a foray into the unknown, and
thus demanded a new style. Unfortunately he never had the time to bring
his ideas to maturity; he died in 1930, only thirty-seven years old. He was the
company’s principal dancer and producer as well as its only choreographer;
he was also the manager and responsible for all the rehearsing and teaching.
He created twenty-four new works in five years. It was an inhuman task
and eventually he cracked under the strain. But the innovations of the
Ballets Suédois were of considerable historical importance and many of the
experimentations of later decades were foreshadowed or even forestalled by
its modernist concepts.
18 The ballet avant-garde II: the ‘new’ Russian and
Soviet dance in the twentieth century
tim scholl

In 1908, a collection of articles on contemporary Russian theatre appeared


in St Petersburg. Modestly titled Theatre. A Book on the New Theatre, the
volume featured contributions by the painter Alexandre Benois, theatre
director Vsevolod Meyerhold, future Commissar of Enlightenment Ana-
toly Lunacharsky, the symbolist poets Andrey Bely and Valery Bryusov, and
novelist Fyodor Sologub. The diversity of this group suggests the signifi-
cance of Russian theatre in St Petersburg at the turn of the century and the
breadth of the quest for new forms in the arts in Russia in the early years of
the twentieth century. The writers mostly advocated the latest movement in
Russian theatre, shaped as it was by a fascination with emerging symbolist
tendencies that sought to correct, or at least to dethrone, the naturalism of
Konstantin Stanislavsky’s Moscow Art Theatre, though Stanislavsky’s inno-
vations were still relatively new.
One year after the theatre volume appeared, Sergey Diaghilev presented
Russian dancers in five ballets in his Saisons russes in Paris. The fame and
notoriety of this “new” dance from Russia would soon eclipse the discussion
of new theatre – and outlast that earlier phenomenon. Nonetheless, Russia’s
new ballet owed much to the experimentation of new theatre. The new ballet
emerged alongside it, and, like the new theatre, new dance was simpler to
define by what it was not. However variously writers conceived of the ‘new’
ballet, one thing was clear: Marius Petipa and the large repertory he created
for the Russian Imperial Ballet represented the old.
Petipa’s 1898 production of Raymonda was the master choreographer’s
last “grand” ballet. He created a series of smaller-scaled works for the Her-
mitage (court) Theatre in 1900 and 1902; the production of his last ballet,
The Magic Mirror met with unprecedented failure in 1903. Ironically, the
reasons routinely cited for the fiasco could serve as the template for the
innovations of the new ballet: the “symphonic” score, the sets by Aleksandr
Golovin, a leading easel painter, and the curious provenance of the libretto,
concocted from the unlikely pairing of the Brothers Grimm and Russia’s
great romantic poet, Aleksandr Pushkin. A mere six years after the failure
of Petipa’s Magic Mirror, Sergey Diaghilev’s Ballets Russes stormed Paris
with ballets set to “concert” music, with sets and costumes designed by
[212]
213 Russian and Soviet dance in the twentieth century

fashionable painters, and libreti drawn from a variety of sources, including


the poetry of Stéphane Mallarmé.
The ballets and choreographers eventually termed “new” were a loose
collection of ballets and ballet innovators who worked mostly at the fringes of
a ballet establishment centred in St Petersburg’s Maryinsky Theatre, Petipa’s
laboratory since 1847. With ballet education and production the exclusive
domain of the Imperial Theatre system, the new ballet had to emerge, with
difficulty, from the old. And although foreign ballerinas, mostly from Italy,
regularly received contracts to dance on the imperial stages, visiting ballet
troupes (and choreography from beyond the Russian empire) only began
to visit Russia once the Imperial Theatres’ monopoly was relaxed in 1882.
However Russian dance writers were none too impressed with the quality
of the dancing or the choreography they saw when troupes such as Luigi
Manzotti’s staged their productions on the summer stages of suburban
amusement parks.
The most stunning blow to the old ballet was delivered on the illustri-
ous Maryinsky stage in St Petersburg, when Aleksandr Gorsky’s production
of Don Quixote (1900, Moscow), was brought in to replace Petipa’s 1869
staging. Aleksandr Gorsky, a former dancer with the Petersburg troupe,
established his reputation as a ballet master by staging Petipa works in
Moscow (including The Sleeping Beauty, from dance notations, in 1899).
Gorsky moved from these fundamentally faithful restagings of Petipa’s bal-
lets to full-scale revisions of his works in later years. As was typically the
case in “new” ballets, Gorsky’s attempts at revision were mostly attempts to
make the old ballets more logical and dramatically viable. Gorsky was influ-
enced by Konstantin Stanislavsky’s work at the Moscow Art Theatre, which
was then enjoying its artistic peak, staging premieres of Anton Chekhov’s
plays. From Stanislavsky, Gorsky learned the importance of the unity of
the production as a whole, as well as the value of its details. In Gorsky’s
productions, dancers were encouraged to analyse their characters’ moti-
vations, decors were painted in a more realistic manner and costumes
were increasingly designed for individuals rather than for groups. Most
importantly, Gorsky focused his directorial attentions on establishing a
clear line of action in his so-called choreo-dramas, revealing a clear debt to
Stanislavsky.
When Gorsky’s version of the Petipa classic Don Quixote arrived in
St Petersburg, local critics were shocked at the asymmetry of Gorsky’s chore-
ography, his attempts to integrate the group dances into the dramatic fabric
of the work and to rid the ballet of these conventionalised divertissements.
The decors, painted by Konstantin Korovin and Aleksandr Golovin (who
would design Petipa’s Magic Mirror the following year) represented another
departure from the work of Petipa’s academically trained designers. Their
214 Tim Scholl

works did not meet with general approval; they were deemed “decadent”
for their lack of perspective and for the predominance of mottled colours.
Gorsky staged Giselle four times in Moscow (1901, 1907, 1918, and 1922).
The evolution of this work in Gorsky’s stagings offers a snapshot of the
trajectory of his evolving approach to the classics, and to his evolution as
a choreographer. His first staging remained faithful to Petipa’s version and
used old decors. Gorsky’s 1907 staging of the work was for Vera Karalli, the
dramatically gifted dancer who had graduated from the ballet school one
year earlier. This staging of the ballet updated the action to the Directoire
period and assigned individualised tasks to the crowd. Karalli clearly stepped
outside the bounds of traditional interpretations of the classic role: she was
criticised for laughing loudly in the mad scene. In the second act, the wilis,
dressed in nightgowns, behaved more as seductresses than spirits from the
underworld.
Gorsky’s quite radical notions of ballet dramaturgy suited the new politi-
cal and cultural climate that followed the decisive October 1917 Revolution –
for a time. His attempts to democratise ballet institutions – as well as ballets –
won him enemies at the Bolshoi, particularly among established virtuoso
dancers who were replaced by a new generation of “dancing actors”. Gorsky’s
favourite ballerinas were dramatically gifted but technically weak, and this
preference for acting showed in the dances Gorsky created. Gorsky’s 1918
version of the ballet was criticised as overly cinematic, as was the acting
of the character dancers who played the lead. By 1922, Gorsky advised his
ballerina not to dance on pointe, but to jump like a young goat, to really go
mad, and die with her legs apart.1 In Gorsky’s hands, in successive stagings,
Giselle became a mimed melodrama.
Gorsky played a central role in ballet reform in Russia in the early years
of the twentieth century, but the radical nature of his later experiments
made his innovations ultimately unworkable, and his productions were
quickly replaced by more traditional treatments of the classic ballets Gorsky
reconfigured. By the time of his death in 1924, little remained of the repertory
Gorsky created for the Bolshoi Theatre.
Despite Gorsky’s pioneering efforts in the creation of the new ballet, the
body of work he created was little known beyond Moscow. Paradoxically,
some of the most famous of Russia’s new ballets were not seen in Russia
until the wave of new ballet experimentation was over. Michel Fokine’s
Schéhérazade, Firebird and Le Spectre de la rose, essential to the early success
of the Diaghilev ballet and exemplars of Russia’s new ballet, did not become
part of the Soviet repertory. Fokine’s ballets, set mostly to concert music
and therefore much shorter than the nineteenth-century’s three- to five-act
ballets, were deemed ‘choreographic miniatures’ in the Soviet Union, where
works of epic length and scale were preferred.
215 Russian and Soviet dance in the twentieth century

Figure 34 Anna Pavlova, studio photograph.

Fokine began his choreographic career in 1905 with Acis and Galatea, a
stylised Greek ballet for his students. His best-known and most-performed
work, The Swan, was created for Anna Pavlova two years later (see Fig. 34). In
these and subsequent works, the inspiration of Isadora Duncan is evident.
Duncan began her first tour of Russia in 1904, and the self-taught, free-
form dances she created furnished a ready model for Fokine and for others.
Her dances were produced independently of state-supported academies and
theatre bureaucracies. Duncan danced to concert music, without special
sets, and with minimal costumes that revealed a freer body than tights and
tutus allowed. Fokine responded to Duncan with a series of retrospective
stylisations, yet The Swan, despite its conventional costuming and steps, was
216 Tim Scholl

Fokine’s most significant contribution to the new ballet repertory. Its nearly
naturalistic focus on the moment of death directed the new ballet’s priorities
towards expression, the watchword of so many modern dance innovators
of Fokine’s day.
Fokine drafted a manifesto of the new ballet that first appeared in The
Times of London in July 1914, one month after the choreographers’s last
work for Diaghilev premiered. Fokine called for the ballet to abandon its
usual conventions, including those of steps and costuming in favour of new
forms better suited to the time and settings of individual ballets. Fokine
believed that the ballet should also abandon the divertissement as a diversion
from the action of the dance, and that dance and pantomime must be
combined to express the idea of the ballet as a whole. Finally, the dance
should unite with other art forms; the new ballet should function as a union
of the arts, and dance should cease to be subordinate to music and the visual
arts.
Fokine’s Chopiniana (1907, the 1909 version for Diaghilev is known
in the West as Les Sylphides), like his Swan from the same year, stylises
the dance of the nineteenth century. These two works provide a fair rep-
resentation of the principles of the new ballet that Fokine would later
draft. They feature integrated dances, not divertissements, and relate events
(relationships, death) that are readily understandable without recourse
to pantomime. Although Gorsky had staged a similar work years earlier
(Valse fantaisie, 1901), Chopiniana is generally regarded as the first plot-less
ballet.
Fokine worked with the painter and set designer Alexandre Benois on
the production of Pavillon d’Armide (1907). The first ballet shown by the
Diaghilev ballet in Paris, Pavillon could serve as the template for the Bal-
lets Russes and the works Diaghilev would produce. The ballet’s opulent
visuals were faithful to the rococo period. The dance was not; yet the har-
monious blend of dance, drama, decor, and music captured the attention
of the European public and established the choreographer’s early fame.
The ballets Fokine created for the Diaghilev ballet – Schéhérazade (1910),
Firebird (1910) and Le Spectre de la rose (1911) – follow the choreogra-
pher’s principles to varying degrees, though Petrushka (1911) is arguably
the choreographer’s most accomplished work. Fokine employed a variety of
dance styles to create the world of the pre-Lenten urban Russian fairground.
The divertissements for nurses, coachmen and others remain mostly in uni-
son, as in the old ballet, but blend seamlessly into the fabric of the work.
Fokine arrived at creative movement solutions to delineate his characters,
utilising movements and gestures for his stars which were usually performed
by character dancers. The Moor’s splayed and Petrushka’s turned-in posi-
tions built on grotesqueries from the Petipa ballet, although in Petrushka,
217 Russian and Soviet dance in the twentieth century

Figure 35 Vaclav Nijinsky in the title role of Petrushka, 1911, choreography


by Mikhail Fokine.

these movements were given to soloists who were mostly deprived of vir-
tuoso movement (see Fig. 35). Fokine responded in kind to the innovative
character of Stravinsky’s groundbreaking score, answering the simultaneous
sounding of two melodies with two different dances performed at the same
218 Tim Scholl

time, although Gorsky had already attracted attention with this device in
his 1900 production of Don Quixote.
In a lifetime of making dances, Fokine never regained the success he
achieved in his early works, or the fame he gained in his work for Diaghilev.
A careful look at Fokine and his innovations reveals enormous debts to
his predecessors: Petipa, Gorsky and Duncan. A clever packager of other
choreographer’s ideas, Fokine found an ideal outlet for their dissemination
while he was part of the Diaghilev enterprise, an enormous travelling pro-
duction company that put the best Russian and European designers, com-
posers and dancers at the service of the choreographer. After breaking with
Diaghilev, Fokine choreographed in Russia, Europe and North America,
mostly restaging the hits of his early career. His early innovations, so central
to the new ballet, soon became commonplaces of twentieth-century dance,
while the Wagnerian hope for the total art work, the harmonious unification
of dance with music and painting, required a constellation of collaborators
and resources impossible to assemble outside Diaghilev’s orbit. Nonetheless,
many of the works Fokine created for the Ballets Russes have survived – a
remarkable achievement (given the life expectancy of ballets from the early
twentieth century) and a testament to Fokine’s ability to translate important
new cultural trends in Russia into ballets.
The decisive October Revolution of 1917 dramatically shifted the land-
scape of dance in Russia. For some time after the Revolution, the future of the
Imperial (then State) Theatre system remained in doubt. Lenin’s position on
theatres and culture was ambivalent. The new Soviet leader believed vaguely
in the need to preserve culture, but it was Anatoly Lunacharsky, the Com-
missar of Enlightenment (or education) who campaigned most vigorously
on behalf of the theatres. Eventually, the State Theatres received a life-saving
appellation: they were called “academic”, to convey their status as educa-
tional tools. Government committees scrutinised their repertories, singling
out supposed counter-revolutionary works and recommending revisions of
others. The libretto of the ballet Sleeping Beauty was reworked as The Sunny
Commune, for example, though the new version was never produced.
Meanwhile, lively debates on the future of dance in the new Soviet repub-
lic appeared in a variety of theatre and culture journals. These discussions
focused on appropriate themes for contemporary dances and on the content
of the new choreographic spectacles. The radical left recommended that the
ballet vocabulary be jettisoned in favour of vernacular movement, acrobat-
ics and folk dance. Others noted that many of the alternative movement
idioms suggested (especially those from the West, such as Isadora Duncan
and ‘machine’ dances) were as alien to Russia as the European court dances
that flourished in Russia’s theatres for two centuries. Until the crackdown
on independent arts groups in the early 1930s, when the state gradually took
219 Russian and Soviet dance in the twentieth century

control of all arts production in the Soviet Union, a number of experimen-


tal dance groups and choreographers flourished. Kasyan Goleizovsky and
Fyodor Lopukhov are the best known of these vanguard choreographers,
and well represent the range of the experimentation in Russian dance in the
1920s.
A dancer in the Bolshoi Theatre at the time of the Revolution, Goleizovsky
had already opened his own school and was much in demand as a chore-
ographer in Moscow’s private theatres and cabarets. In the year following
the Revolution, Goleizovsky took charge of the Bolshoi’s theatre school, but
left the company months later to choreograph full time. A harsh critic of
the routine and the increasingly archaic repertories of the State Theatres,
Goleizovsky nonetheless saw the professional ballet theatres as the sole
repositories of skilled, well-trained dancers in Russia. Goleizovsky based
his ever-expanding movement idiom on classical technique and required
trained dancers, but nonetheless believed that all movement was legitimate
and could be used to invigorate classical technique. Goleizovsky’s dancers
might perform somersaults, daring lifts, or lie on the stage, but the choreog-
rapher’s diverse movement idiom relied on a foundation in ballet technique.
A new attitude towards the visual elements of the ballet production marked
Goleizovsky’s new dance theatre and brought the ballet fully into the artistic
vanguard of 1920s Russia. The costumes, however, lent the enterprise a hint
of scandal. Goleizovsky believed in the nude body as both an aesthetic and
a moral ideal, and although his dancers never appeared completely nude,
they often appeared in minimal costumes. Like other dance reformers in
the twentieth century, including George Balanchine, Goleizovsky preferred
minimal costumes as a better way to reveal the body’s movement. And like
many theatre directors in Russia in the 1920s, Goleizovsky arranged these
bodies on constructivist stage sets; dancers and dances were arranged on
multiple planes.
Goleizovsky’s best-known work, The Legend of Joseph the Beautiful,
staged in 1925 for the Bolshoi’s Experimental Theatre, made extensive use
of stage platforms, stairs and constructions. Boris Erdman’s costumes were
asymmetrical and eccentric, updating the dress of Ancient Egypt for 1920s
flappers. Goleizovsky’s choreography avoided the archeological stylisations
of Fokine and Gorsky, instead incorporating motifs from a variety of folk and
historical dance traditions. Yet despite the fusion of dance styles Goleizovsky
used in the ballet, contemporary commentators noted an unusual degree of
coherence in the dances.
Work on The Whirlwind (1927) led to Goleizovsky’s resignation from
the Bolshoi Theatre, though he continued to contribute occasional works
to the theatre until 1964. The choreographer’s withdrawal from the Bolshoi
anticipated the conservatism of the “academic” theatres in the 1930s and
220 Tim Scholl

beyond, a time when Goleizovsky retreated to music halls and cabarets, and
arranged dances for films.
If the work of Gorsky and Goleizovsky ultimately proved too radical
and eccentric for the Bolshoi Theatre, St Petersburg’s Maryinsky (called the
State Academic Theatre of Opera and Ballet until it received the appellation
“Kirov” in 1935) followed an even more conservative path to a Soviet approx-
imation of modernity. The Revolution made for a very difficult situation
within the former Maryinsky, not least because so many of the company’s
former stars (Nijinsky, Pavlova, Karsavina, to name a few) had established
themselves in the West with the Diaghilev troupe and chose to remain
there. Nikolay Sergeyev, the troupe’s arch-conservative régisseur, left Russia
in 1918 with the dance notations that recorded the bulk of the ballet reper-
tory; Fokine left St Petersburg the same year. A series of male dancers staged
ballets and worked as régisseurs for the company during the chaotic period
following the Revolution until Fyodor Lopukhov was appointed director in
1922.
The first directive of Lopukhov’s new administration amounted to a
purge. A statement announced that Petipa’s ballet would form the basis of
the troupe’s repertory and that special efforts would be made to cleanse them
of the accretions of recent years (the work of other régisseurs from the time
of Petipa’s retirement in 1903 and death in 1910). Lopukhov began a process
that continues to dominate discussions concerning the Maryinsky and its
performance practices to the present day. His determination to return to
a more pure or authentic version of Petipa’s ballets inaugurated a quest as
impossible as stepping twice into the same stream.
The complexity of Lopukhov’s undertaking is demonstrated by a small,
but telling moment in Russian ballet history in which Lopukhov played
a leading role. In 1972, Lopukhov admitted that in 1914, he had chore-
ographed the most commonly performed variation for the Lilac Fairy
in Petipa’s Sleeping Beauty. Lopukhov maintained that the variation was
attributed to Petipa in order to escape the scrutiny of the régisseur, Niko-
lay Sergeyev. In his 1972 account of this history, Lopukhov nonetheless
maintains that his variation functions as a kind of quintessence of the role,
thus justifying his questionable maintenance of the Petipa legacy and con-
gratulating himself for the deception. This pattern of “improving” Petipa
continued throughout the Soviet era.
However questionable the authenticity of Lopukhov’s revisions of Petipa,
the decision to preserve the legacy proved more fruitful than the Moscow
tendency to create increasingly eccentric versions of the nineteenth-century
repertory. Throughout the Soviet period, the Petersburg/Leningrad ballet
remained a repository (if an imperfect one) of the nineteenth-century reper-
tory, whose productions were copied and reproduced for ballet companies
221 Russian and Soviet dance in the twentieth century

around the Soviet empire and the world. And despite Lopukhov’s reputation
as a conservative, intent on preserving the classical legacy, the ballet master
and choreographer was also interested in new forms.
His best-known work had only one performance; the dance-symphony
Magnificence of the Universe was set to Beethoven’s Fourth Symphony
and featured dancers from George Balanchine’s Young Ballet in its cast.
Lopukhov’s fascination with the symphony reflected a generalised anxiety
over the ballet’s place in some imagined hierarchy of the arts in the young
Soviet republic. At a time when the score of Sleeping Beauty was singled out
as the sole, musically acceptable score suitable for performance in the Soviet
“academic” theatres, Lopukhov sensibly scrambled for the higher ground of
an indisputable genre. Beethoven was admired in the Soviet Union in this
period, both by conservatives and by radicals. The latter judged him close
to the spirit of the French Revolution. The symphony quickly became an
idée fixe for Soviet dance. Russian writers still use the term ‘symphonic’ to
connote choreographic sophistication.
Lopukhov’s choreography to the Beethoven score was perhaps less
remarkable than the programme notes he wrote for the performance. With
sections of the ballet titled The Conception of Light, and Life in Death and
Death in Life, Lopukhov’s vision for the new ballet wed pretension to naivety.
The ballet proved an unpalatable concoction that, regrettably, suggested
much of the future direction of Soviet dance. An uneasy step into the world
of abstraction, Lopukhov’s ballet retained narrative as an organising prin-
ciple. At a time when flirtations with abstraction would be denounced as
formalist experiments (the most damning denigration in Soviet arts criti-
cism), Soviet choreographers intent on exploring the plot-less potential of
dance were careful to disguise these “deviations” with an overlay of plot.
The theoretical foundations for Lopukhov’s 1923 ballet may be found
in a written work that appeared two years after the ballet, although he
had begun it much earlier. In Paths of a Ballet-Master, Lopukhov outlined
his notions of an ideal relationship between dance and music. Essentially,
Lopukhov’s tract calls for a unity between the two forms, though many of
the specifics strike the modern reader as naive. Lopukhov’s insistence on
correspondence between the two forms included such particulars as the
suggestion that minor keys be reflected in en dedans movement and major
keys mirrored by movement en dehors.
Like Goleizovsky, Lopukhov worked only intermittently in the Soviet
academic theatres after the 1920s. In both cases, the two men’s notions of
the future of dance proved too radical for an arts bureaucracy that came to
favour slow evolution over new theories and “revolutionary” change. The
future of Soviet dance lay with less progressive ballet masters who were
willing to parrot formulaic approaches to art as handed down by party
222 Tim Scholl

tribunals. It is not surprising, then, that the next important wave in Soviet
ballet production had no identifiable author.
Much as Lopukhov, and others, sought to “symphonise” the ballet, chore-
ographers in the 1930s and 1940s turned to another unshakeable genre from
yet another art form as a basis for new ballets. The Stalin-era adoration of
epic forms resulted in the Soviet ballet’s new enthusiasm for adaptations of
novels and the plays of literature’s Beethoven: William Shakespeare. Rostislav
Zakharov’s The Fountain of Bakhchisarai (1934) and Leonid Lavrovsky’s
Romeo and Juliet (1940), both based on literary monuments, functioned
as exemplars of the new wave in Soviet choreography, the drambalet. A
contraction of “drama” and “ballet”, the drambalet was meant to fuse the
two seamlessly in a marriage of gesture and movement that avoided the
nineteenth-century’s division of pantomime and dancing. With time, it
became clear that these dances privileged storytelling and pantomime over
movement, and that dance as such took a second place to narrative con-
veyed in highly conventionalised gestures. The close-ups in the film version
(1954) of Lavrovsky’s Romeo and Juliet make it more accessible than the
staged ballet and point out the genre’s greatest deficiency: with so much
of the story conveyed by silent-film gesturing, the live version of the ballet
suffers by comparison.
The burst of creativity, experimentation and theorising that charac-
terised Soviet arts in the 1920s was largely absent by the 1930s, when the
state sought, and mostly achieved, control of avenues of creative expression.
The ballet proved especially malleable, since dance activity was centred in
the large theatres of large cities and the Western system of independent
choreographers leading small troupes of dancers had not taken root in Rus-
sia. Despite a fervent period of activity in the dramatic theatre and the ballet
theatre, the new Russian theatre and dance that captured the imagination
of practitioners and writers in the first years of the twentieth century failed
to blossom in Russia and the Soviet Union. The 1917 Revolution had dras-
tically changed conditions in the Russian theatres; the emigrations of artists
immediately thereafter left a creative vacuum impossible to fill in the lean
and hungry years of civil war and cultural revolution that followed.
Russia’s new dance, like new theatre, had a greater impact in the West,
where experimentation and artistic collaboration were prized long after
both became problematic in the Soviet Union. Abstraction and formalism,
dangerous concepts for Soviet choreographers, became the rule in ballets
created by Russian émigré dance-makers, especially George Balanchine. Bal-
anchine’s revolution in the ballet certainly drew upon his experience in 1920s
Russia: his incorporation of a plentitude of dance idioms and styles echoed
Goleizovsky’s catholic approach to choreography; Balanchine’s thorough
investigation of the relationship of choreography to music revealed a debt
223 Russian and Soviet dance in the twentieth century

to Lopukhov. In the hands of Russian émigrés, ballet became a prominent


feature of the European and North American cultural landscape in the twen-
tieth century. The dance that these émigrés created was no longer identified
with Russian new ballet, yet it grew from the revolt against the old bal-
let begun by Gorsky, Fokine, and others in the first years of the twentieth
century.
19 George Balanchine
m at i l d e bu t ka s

It’s like watching light pass through a prism. The music passes through him, and in the same
natural yet marvellous way that a prism refracts light, he refracts music into dance.”
( m a rt h a g r a h a m ) 1

Balanchine has been likened to Mozart and Shakespeare for his univer-
sal appeal and prolific creative output and to Picasso and Matisse for his
contributions to twentieth-century art. His friend and collaborator Igor
Stravinsky, not famous for complimenting his peers, declared that “The
world is full of pretty good concert pianists, but a choreographer such as
Balanchine is, after all, the rarest of beings”.2 In the years since Balanchine’s
death in 1983 his ballets have remained a staple of New York City Ballet
programmes each season and are widely performed by other companies
as well, many of them led by former Balanchine dancers who work hard
to recreate “Mr B’s” ballets as faithfully as possible. It is rare for a chore-
ographer to have this many works still in performance (to say nothing of
the continuing influence of his school of technique – a streamlined ver-
sion of classical ballet notable for its speed, energy, clarity, restylings of
some of the original dance positions and emphasis on music as the heart of
ballet).
What was so extraordinary about Balanchine as a working artist? First,
he approached choreography as a craft akin to cooking or carpentry. He
worked easily with what was available, adjusting his choreography to a
dancer’s strengths and was never fazed by the notion that a ballet might
disappear forever after its performance. Moreover, Balanchine made ballets
quickly and was, as a result, prolific, creating hundreds of ballets over the
course of his life. He needed, most of all, to be in the studio working directly
with his dancers and he found joy in the process of, as he liked to say,
“assembling”.
Second, Balanchine’s working style inspired great loyalty in his dancers.
He was, by many (albeit not all) accounts, easy to work with. In addition, he
was a hands-on choreographer, intensely detail-oriented. His hand touched
everything – from the sewing of costumes to the programme notes to the
position of a ballerina’s tiara – and he taught company class frequently until
late in his life. These finer details, though perhaps unnoticed by the public,
made the New York City Ballet distinct and the high quality of its dancers
so consistent.
[224]
225 George Balanchine

Third, his choreography was particularly responsive to music. This is


widely acknowledged, but less often brought up in the literature than
“Mr B’s” personal life, his place in dance history, his technique, dancers,
company and his choreographies. Balanchine’s musical training and native
abilities were unusual for a choreographer – he was a pianist who studied at
the conservatory in St Petersburg, composed, made his own piano reduc-
tions of ballet scores, analysed these scores and even conducted. His deep
interest in music led him to create choreographies that were, as Terry Tea-
chout put it (describing his first experience of watching Concerto Barocco),
“sound made visible, written in the air like fireworks glittering in the night
sky”.3 Charles Joseph wrote, along the same lines, that Balanchine, “a musi-
cally astute choreographer, possessed the uncanny gift of clarifying what my
ears heard through what my eyes saw”.4 However, as Edwin Denby pointed
out, “the so-called imitations of music by the dancers, far from being literal,
have a grace at once sophisticated and ingenious. The musical play and the
play of dance figures, between them, create bit by bit a subtle strength”5
Balanchine’s choreography does not simply visualise or imitate the music,
but rather, the musical structure determines the movement in more com-
plex ways. A few scholars, led by the pioneering work of “choreo-musical”
analyst Stephanie Jordan, have shed light on this much celebrated musical
aspect of Balanchine’s choreography.
That most Balanchine ballets are plotless provides further complexity;
it also affected his choices and uses of music. Though his ballets are rich
in drama, often that between a man and a woman, the story is left for the
viewer to interpret (or not). Balanchine felt overt plots were unnecessary
and that words were not needed to describe or discuss his ballets. “He always
said that seeing [his ballets] was enough: words about them, how he came to
make them, or about his life were of little use,” wrote Francis Mason.6 Since
Balanchine so often did not rely on plot, he could focus purely on combining
bodies in motion with sound, without undermining the importance of
either. Important to his approach was to keep choreography from interfering
with or obscuring the music.7 He claimed to subdue his dances, submitting
them to the music, taking a “less is more” attitude.8 Indeed, the music itself
provided drama and he underlined this with choreography. His approach
to music and drama led to an amazing versatility in styles of both music
and choreography, from Gershwin, Rodgers and Hart to Bach and Webern
and from classic choreographies for prima ballerinas to dances for circus
elephants.
The balance achieved in his work brings clarity to both the music and
the dance in unprecedented ways. Stravinsky said that Balanchine’s chore-
ography “exposed relationships of which I had not been aware in the same
226 Matilde Butkas

way. Seeing it, therefore, was like touring a building for which I had drawn
the plans but never completely explored the result.”9

Origins: Russia and Europe


Georgi Melitonovich Balanchivadze was born in St Petersburg in 1904.
His Georgian ancestry, the cosmopolitan and European influence of
St Petersburg and a musical family all constitute important parts of Balan-
chine’s heritage. His mother was his first piano teacher, his father a composer
nicknamed “the Georgian Glinka”. Though the details of accounts vary, Bal-
anchinvadze ended up as a ballet student in the Imperial Theatre School of
St Petersburg with his sister, instead of the Imperial Naval Academy of St
Petersburg as his mother had wanted. At first he was miserable and lonely,
more interested in music and religion classes.
One night, when he performed as a cupid in Petipa’s Sleeping Beauty at
the Maryinsky Theatre, he fell in love with ballet. Not long after this, in 1917,
the Russian Revolution broke out, changing everything. The circumstances
were dire as well as bizarre. Dances were performed in the freezing cold as
interludes during political debates on the stage of the Maryinsky, which only
a few years before had displayed the lush, extravagant performances for the
tsar. During these deprived post-revolution years, Balanchivadze contin-
ued to pursue music and played the piano often at school. Yuri Slonimsky
recalled him playing pieces by Chopin and Russian contemporaries Scri-
abin, Medtner and Rachmaninoff.10 Balanchivadze graduated in 1921 and
became a member of the ballet company but also enrolled in the conser-
vatory of music, not sure yet in which direction he would travel: music or
dance.
His first choreographies as a teenager were scandalous enough to raise
eyebrows even though the early 1920s atmosphere was ripe for young artists’
indulgence and experimentalism. At this time Balanchivadze was influenced
by Fokine’s “plotless” Chopiniana (Les Sylphides), Goleizovsky’s barefoot
and bare-skinned ballets, Lopukhov’s avante-garde works such as Dance
Symphony, and he even saw Isadora Duncan perform, though he was not
favourably impressed. In 1923, Young Ballet, a new performing group Bal-
anchine helped form, staged his choreography to Blok’s poetry chanted by
a chorus. The Maryinsky faculty prohibited the dancers from taking part in
further Young Ballet programmes.
In 1924 Balanchivadze left Russia on a tour with a troupe of dancers
including his wife, Tamara Geva, and Alexandra Danilova. Balanchine mar-
ried three more times and made no attempt to disguise the fact that his
inspiration was from women; he would later assert “ballet is woman”. He
would remain friendly with his former wives and worked professionally with
227 George Balanchine

them, creating ballets for them throughout his career. Years later, Danilova
remembered taking one class with him at the School of American Ballet
in which “there at the barre were all the wives and ex-wives: Geva, Zorina,
Tallchief, Tanaquil LeClerq and myself ”.11
Surprised at having plentiful food for once, the Russian troupe ate their
way across the Baltic Sea. They also decided not to return to Russia. They
were eventually contacted by Diaghilev and auditioned for his famous Bal-
lets Russes. Looking back we see this as a decisive moment; Diaghilev’s
Ballets Russes was at the cutting edge, providing an ideal atmosphere for
Balanchine to continue developing choreographic skills and ideas. And so,
Georgi Balanchivadze became George Balanchine at Diaghilev’s request,
the company’s newest choreographer, and began creating works for the
Ballets Russes and for the Opéra de Monte Carlo. Balanchine learned
to appreciate painting and sculpture and also discovered his talent for
cooking.

Early works
Apollo (Stravinsky, 1928),12 a work that Balanchine considered a turning
point in his creative life. Balanchine realised the value of economy in ideas
and “that, like tones in music and shades in painting, gestures have certain
family relations which, as groups, impose their own laws”.13 Diaghilev recog-
nised Petipa-like qualities and Balanchine’s ability to create classical ballets.
But Balanchine became fascinated with jazz dance just as much as with clas-
sicism. In fact, jazz movements, though not necessarily new to Balanchine,
entered his vocabulary during this time, probably through the influence of
the Parisian star Josephine Baker, as Beth Genné has pointed out.14 There
is evidence that Balanchine not only choreographed for, danced with and
admired Baker, but most importantly that the two traded material, with
Baker learning to dance en pointe and Balanchine absorbing her jazz style.
This yielded a hybrid, “a jazz infused classicism that integrates the angled
arms akimbo, the crooked (cats paw) wrists with every finger displayed with
the upright torso, precise placement and line lengthening point work of the
danse d’école”.15 The meeting of these two young dancers during the 1920s
and 1930s played a crucial part in Balanchine’s future artistic development
and his challenge of American race taboos. The jazz-influenced neoclas-
sicism would characterise many of Balanchine’s American works such as
The Four Temperaments and Agon. Baker was probably the first “Balanchine
ballerina”; she represented the American ideal with the much celebrated
long legs, grace and athleticism, qualities that Balanchine would cherish in
dancers such as Suzanne Farrell.16
228 Matilde Butkas

Also during this period Balanchine began a lifelong friendship with


Stravinsky; many of Balanchine’s great neoclassic ballets were made to
Stravinsky’s music.17 The choreographer and composer had much in com-
mon, from their shared love of pre-revolutionary St Petersburg to their
artistic philosophy and aesthetic. Each of the two also felt a deep sympa-
thy for the other’s art. “If a dancer’s heart beat within Stravinsky, that of
a musician invigorated Balanchine,” wrote Charles Joseph in his study of
their relationship.18
Following Diaghilev’s death, Balanchine battled with tuberculosis and
after recovery, found work with Cochran’s Revues, the Royal Danish Ballet,
Sir Oswald’s variety shows in England, the new Ballets Russes de Monte
Carlo and then its competition, Les Ballets 1933. Many works from this
period disappeared like the butterflies in Balanchine’s famous metaphor:
“[A ballet] is wonderful, it is now. It is like a butterfly. I always say butterflies
of yesterday don’t exist.”19 Balanchine was twenty-nine, dancing despite an
injured knee, choreographing for whomever needed him and uncertain of
his future. It was as this point of restlessness that Lincoln Kirstein saw a
performance of his work in London and promptly proposed that he should
become part of Kirstein’s vision for a distinctly American ballet company,
school, repertoire and audience. He arrived in New York in October 1933
(see Fig. 36).

America: Broadway and New Ballet


It took some time before Balanchine found resources and stability. But he
loved his adopted country and its culture and found much to influence his
art, including Broadway, hoe-downs, Hollywood, Fred Astaire and Ginger
Rogers, Western movies (which he watched avidly), jazz, modern dance
and composers such as Ives, Sousa and Gershwin. On 1 January, 1934, the
School of American Ballet officially opened and in June of that year its stu-
dents performed Balanchine’s first American ballet: Serenade. Serenade, to
Tchaikovsky’s Serenade in C for string orchestra (1880)20 begins with a rit-
ual that not only evokes the elemental movements of ballet class, but records
how Balanchine started from scratch, teaching his American students a basi-
cally unfamiliar art. Years later, tears would spring to Martha Graham’s eyes
when she saw the first section where the entire corps suddenly turns their
feet out into first position, a striking and concise movement amidst a flow-
ing and lyrical passage of dance and music.21 Such precision would later be
identified as a trademark of the Balanchine style.
Serenade illustrates an important aspect of Balanchine’s choreographic
process: working with what is available and whatever circumstances arise.
229 George Balanchine

Figure 36 Pennsylvania Ballet Principal Dancer Riolama Lorenzo with Soloist Philip Colucci in
The Prodigal Son, choreography by George Balanchine.

Each day of rehearsal, a different number of dancers were present and the day
he put together the opening scene, the odd number of seventeen appeared,
and so Balanchine used a bow-tie formation rather than standard parallel
lines. One day a dancer rushed in late, another day a dancer fell; he would
add these moments and mishaps to the choreography.
The structure of this ballet, like so many of Balanchine’s works, derives
not from a plot but from the music. Of Serenade, Balanchine tells us that,
“making a ballet is the choreographer’s way of showing how he understands
a piece of music, not in words, not in narrative form (unless he has in mind a
particular story), but in dancing.”22 Upon closer examination we see another
example of Balanchine’s classicism. Both Balanchine and Tchaikovsky use
a limited set of themes, which are then explored and restated and brought
back in variation. The dance reflects the music but not by merely imitating
230 Matilde Butkas

the musical events (high elevation equalling high pitch, fast movement
equalling fast musical tempos, etc.). Balanchine and Mason wrote that in
Serenade, “the only story is the music’s story, a serenade, a dance, if you like,
in the light of the moon”.23 Balanchine also alluded to fragments of a story,
thought by many to be autobiographical: “It’s like fate . . . Each man going
through the world with his destiny on his back. He meets a woman – he
cares for her – but destiny has other plans.”24 Narrative drive in Serenade
may be debated,but the music and choreography are dramatic, especially
when one considers the myriad changes made in Serenade’s “evolution” –
changes which take place still with each new dancer, stager, conductor and
orchestra. Serenade, whether it is about pure dance, music, fate, a woman,
a process or inevitable loneliness, is enduringly popular, and still widely
performed today.

Broadway and Hollywood


Balanchine soon found many venues for his talents. He felt just as com-
fortable choreographing for Broadway and Hollywood as he did for the
ballet stage. He knew how to entertain and knew what was popular, sexy
and funny. He worked easily with Rodgers and Hart and spoke fondly of
Larry Hart teaching him English. Balanchine could sing Hart’s “My Funny
Valentine” in flawless English without a trace of Russian accent.25 Balan-
chine brought the term “choreography” to the programmes of Broadway
musicals, beginning with the hit On Your Toes in 1936. The time was just
right. People were beginning to appreciate the dance in musicals and the
choreography grew more sophisticated. That year he also choreographed
for the Ziegfeld Follies of 1936, which included the stars Fanny Brice, Bob
Hope and Josephine Baker, who hoped to make a triumphant return to her
native country. The critics sharply disapproved of his casting Baker in a
romantic lead role dancing with four white male suitors; the racist response
was devastating.26
Also in 1937, he tried to get the American Ballet (the original Kirstein–
Balanchine company) back on its feet with a Stravinsky Festival featuring
Apollo, Le Baiser de la fée (The Fairy’s Kiss) and the newly commissioned
ballet about a poker game: Jeu de cartes (Card Game),“a ballet in three
deals”.

Ballet Society
Around 1940, “after all kinds of expenses”, Balanchine said that he “still had
five hundred dollars left. I thought, what should I do with the money? buy
231 George Balanchine

some extraordinary cigarette case or something? And then I decided, ah! I’ll
ask Hindemith to write something for me.”27 This is how The Four Temper-
aments for piano and strings came into existence, a work cherished today
even by musicians unfamiliar with its choreography. It had its premiere
in 1946 for a new, post-war Kirstein–Balanchine venture: Ballet Society, a
subscription-only company (even for the press), with the purpose of per-
forming newly commissioned or unfamiliar works. The American public
was ready for such a company, for the mid-1940s was a time when, as a
1944 article title put it, “The Ballet Comes into its Own”.28 Balanchine was
still working for Broadway and Hollywood and his contributions to Song of
Norway, a musical on the life of Grieg, was seen as a “considerable part of
the production’s total assets”.29
Two years later, the Ballet Society became the New York City Ballet at
City Center. This would be Balanchine’s (Mr B’s) home company until his
death.
The Four Temperaments became part of the new company’s repertory.
Subtitled “A Dance Ballet without Plot”, it is a theme and four variations
based on medieval cosmology: I. Melancholic, II. Sanguinic, III. Phlegmatic
and IV. Choleric.30 Dancing evolves out of the music, “it is the musical
pulse, which provides rhythmic drive . . . , [the] pulse drive is celebrated too
in movement style, most clearly with change of weight and in the work of legs
and feet”.31 The “pulse drive”, “change of weight” and “foot and leg work”
form the basis of Balanchine’s modernist vocabulary. In addition to the pair-
ing of distinct musical themes with dance motifs and dancers, Balanchine
subtly reformulated the strong correlations between the music and move-
ment, between “family relations” and the economy of ideas, with which
he had been concerned in earlier years. The Four Temperaments became
known as one of the “black and white” ballets. In his incessant reshaping of
choreography and costume, Balanchine soon after the premiere discarded
the original costumes, which he thought were an impractical assortment
of wraps, tubes and bandages. (“Balanchine was reportedly snipping away
pieces of the costumes right up to curtain time.”32 ) He made the dancers
wear rehearsal clothes, simple black leotards and tights, and then set them
against a solid-coloured backdrop. Thus, Balanchine shifted the focus from
spectacle to the dance itself and especially to the unadorned body and its nat-
ural line, its flexibility, agility, strength and submission to the force of gravity.
Gravity, the pull that ballet traditionally attempts to defy, is the force above
all explored in The Four Temperaments, especially in the Melancholic varia-
tion where the male soloist succumbs to it, thwarted by the sinister marching
of the corps women en pointe to the entrance of a prominent piano theme.
Many of the lifts and supports clearly show the effort involved, and at times
do not even accomplish a “heightening” of the dancer. The male soloist in
232 Matilde Butkas

Melancholic exits with his torso arched backward as if being invisibly pulled.
Next to the problem of gravity, Balanchine explored physical and spatial
limitations. In addition to the restricted vocabulary of select “motifs”, the
dancers are often obstructing each other through entanglements of arms
as in Phlegmatic or the holding of limbs of the female soloist in Choleric.33
(These examples are now seen as Balanchine trademarks.) The dancers arch
their backs forward and backward as if by pushed by an unseen force. Tor-
sos are malleable and twisted, legs and hips are turned in at times, arms
are angular and jagged and supporting legs are bent, even when on pointe.
It often seems as though a dancer is trying to escape her body but is lim-
ited by unseen gravitational forces or the snares set up by other dancers.
This ballet, though somewhat modified for the screen, was Balanchine’s
first choice for the PBS broadcast Dance in America in 1977. Balanchine
had long been fascinated with the possibilities of filming dance and taking
camera angles into account. One Hollywood anecdote tells of Balanchine,
in 1937, constantly moving the exasperated director and others around the
studio to demonstrate how differently the dance would look from different
positions.34
In the years to come the company grew in strength; students who
had been with the school from the start became its principal members.
It was a slow battle for financial security with much hard work and
sacrifice.Balanchine’s Firebird (1949) to Stravinsky’s newly arranged suite
from the original score received overwhelming success and took the pub-
lic by surprise. Maria Tallchief ’s demonically difficult solo caused gasps of
awe. The Firebird and other successful pieces formed the repertoire for the
extensive touring period in the 1950s.
In February 1954, Balanchine and the New York City Ballet presented
The Nutcracker, a ballet that would reformulate the meaning of Christmas
in America. From 1954 onwards, the New York City Ballet would not miss a
single year of this holiday favourite and from year to year the Christmas tree
grew. The ballet seemed to stand in odd juxtaposition to other, increasingly
abstract choreographies and indulged in all those elements that seemed to
be missing from many Balanchine choreographies, namely mime, costume,
narrative.

1957 and Beyond


Disaster struck in 1956, as Tanaquil LeClerq, married to Balanchine and one
of his leading dancers, fell ill with polio. Balanchine, devastated, set his work
aside completely to care for her until November 1957. When he returned
to the studio to choreograph, four ballets, completely different from
233 George Balanchine

anything he had ever done, spilled out in under three months: Square Dance
(Vivaldi/Corelli), Gounod Symphony, Stars and Stripes (Sousa/arr. Kay) and
Agon (Stravinsky). There was something for everyone is this season but it
was Agon that received the most attention. Was it the nature of Balanchine’s
close collaboration with Stravinsky (he had proposed the idea of the ballet
and Stravinsky took it up) or the daring and purposeful partnering of Diana
Adams and African-American dancer Arthur Mitchell that astounded the
audience once more? Duchamp compared Agon to Le Sacre du printemps
and Arthur Mitchell wrote that it “set the standard for neoclassical dance
today”. The ballet represented “a perfect union between two great artists”35 –
composer and choreographer and female and male dancer.
Agon showed all the characteristics of previous ballets: the usual hard
work, the craftsmanship, the eye for the strengths of particular dancers and
the incorporation of personalities and banal events into the work. Arthur
Mitchell echoed most dancers’ sentiments, when he characterised Balan-
chine as “a master at knowing what looked good on a dancer’s body and
what movement could extend a dancer’s body”.36 Mitchell recalled how, on
a day when Balanchine’s knee was bothering him, “he added a limp to the
choreography. The ‘toe, flat,’ ‘toe, flat’ step that the boys do came out of that
day”.37 Balanchine had exact timings in minutes and seconds in mind and
even had stick figure drawings for the number of male and female dancers
to be used for each part. Agon, Greek for “contest”, best represents the ideal
relationship between music and choreography as Balanchine and Stravinsky
saw it: “a struggle between music and choreography to bring about harmony
and synthesis . . . a strenuous working together of two independent forces”.38
Above all, rhythm as the constant from which movement derives is organ-
ised in such a way that “the most varied and complex inter-relationships
between what we see and what we hear” occur.39 Did Balanchine make the
score visible? If so, then the visualising techniques used by Balanchine could
be a step per note, mimicking musical highs and lows, or dancing the “look”
of the score or its specific elements (such as the “tone row”). A major Bal-
anchine trait is showing the underlying metre (including notes and rests
or silences), such as in the opening of Agon, which serves to “[clarify] an
ambiguous sense of musical pulse and metre”.40 Again, counterpoint (dance
in opposition or tension with the musical score) is especially prominent.
There are also places where the dancers anticipate or echo what is heard.
Balanchine never strictly visualises what can be heard (sometimes called
“mickey-mousing”) rather, moments of visualisation pass quickly into the
shifting flow, as in the short Bransle Gay female solo. Throughout, one
hears the steady pattern on the castanets while the musical metre shifts. The
dancer only lines up with the rhythm in an obvious way when the castanets
are heard alone: at the opening, the ending and twice in the middle.
234 Matilde Butkas

The focal point of the work is the pas de deux, which departs from an easily
countable rhythmic pulse. The dancers must know the music extremely well
and react while they “play along”. Simply reading about these moments only
serves to whet one’s appetite for the actual performance, but it is this way of
working with music that awed dancers and critics. One does not necessarily
have to have music or dance training to appreciate or notice these things.
In fact, Balanchine would say in his typical offhand way that, if people liked
looking, that was fine, they would probably come again. If not, they could
always close their eyes and listen to the wonderful music.

Balanchine was consistently creative throughout his life, pausing only infre-
quently, and the quality of his work rarely suffered. There were, of course,
times of stress, political, cultural or personal. The emotionally intense tour
of the Soviet Union in 1962 coincided with the Cuban missile crisis. During
the challenging move to the Lincoln Center and New York State Theater in
1964, Balanchine discovered during the building process that the orchestra
pit was half the size he expected. There was tension in the company and
Balanchine faced problems in his relationships. But he was now accepted
as the foremost American choreographer of ballet. He had finally found a
home and stability. Soon he even received a salary – though he gave it all
to his newly rehired secretary, Barbara Horgan (without her knowing). The
Ballet started spending the summers in Saratoga Springs in upstate New
York. There were many excellent dancers to work with. Balanchine began
to make his ballets available to other companies and to produce them for
film.
Of the many great ballets during this period only a few can be
mentioned: Episodes (1959) – the famous collaboration with Martha
Graham to Webern’s complete orchestral works; A Midsummer Night’s
Dream (1962) – the first full-length original story-ballet to his own selec-
tions from Mendelssohn’s music, complete with mime and a pas de deux
between Titania and the perfectly choreographed Ass; Movements for Piano
and Orchestra (1963) – the ballet that Stravinsky “heard with his eyes”,
which also marked the beginning of the “Suzanne Farrell years”; Don Quixote
(1965) – commissioned from Nabokov, in which Balanchine, autobiograph-
ically, danced the role of Don with the young Farrell as Dulcinea. There were
Jewels (1967) to music by Fauré, Stravinsky and Tchaikovsky (see Fig. 37),
Who Cares? (1970) to music by Gershwin and Mozartiana (rev. 1981) to
music by Tchaikovsky (hailed as his last great ballet). There were festivals
for Stravinsky, Ravel and Tchaikovsky. Even on his deathbed he thought
about choreography. Maria Tallchief remembers her last visit: “music was
playing. Balanchine was tapping the fingers of both hands against each
other.” When she asked what he was doing, he replied, “You see, I’m making
235 George Balanchine

Figure 37 Suzanne Farrell and Peter Martins in Jewels – Diamonds, choreography by George
Balanchine.

steps.”41 He died on 30 April, 1983. That night Lincoln Kirstein announced


to the audience, “I don’t have to tell you that Mr. B. is with Mozart and
Tchaikovsky and Stravinsky.”42

Balanchine’s legacy
In his will, Balanchine left his ballets to his friends and dancers. He thought
that in fifty years no one would care; he was wrong. The efforts of the George
236 Matilde Butkas

Balanchine Foundation and the George Balanchine Trust seek to ensure that
the ballets are carefully preserved; many of his ballets have been filmed, and
former students on video tapes demonstrate his technique. Many of his
dancers have been interviewed, have written memoirs recounting his work-
ing practices and describing his technique. Balanchine has been turned into
an ongoing new research project. If anything, Balanchine’s legacy is not only
secure but growing. Such policy goes against Balanchine’s philosophy – that
ballets are ephemeral, like butterflies and that they should be updated with-
out concern for the “authenticity” of the first version. But his inspiration
was too great, his ballets too beautiful and too compelling for his disciples,
and the greater audience, to relinquish. Can his effect on the dance world
be measured? Balanchine would say that “probably dance would stop if we
didn’t have Stravinsky”43 but it is certain that without Balanchine, ballet
would not be what it is today. It is not only his technique and his moderni-
sation of classical ballet but his open-hearted and eclectic approach to the
art that so altered it, and that continues to attract audiences to it. Moreover,
his particular love for his adopted country, his commitment to create a truly
American ballet company and to celebrate something American exerted an
enormous influence on the ballet of the twentieth century.
20 Balanchine and the deconstruction of classicism
j u l i e t b e l l ow

I would say quite frankly that I detest the spirit of these parodies, and that the perpetual
sniggering of the choreographer Mr Balanchine seems to me, at times, a sign of powerlessness. He
uses and abuses classical dance by literally putting it to torture; with a type of sadistic satisfaction,
he . . . forces the leader of the Muses to play the clown.1

With this scathing critique André Levinson, the most prominent defender of
the nineteenth-century balletic canon, assessed George Balanchine’s Le Bal
(The Ball, 1929). As anyone familiar with Balanchine’s career knows, this is
a surprisingly harsh treatment of the choreographer who has come to epit-
omise the resuscitation of classical ballet in the twentieth century. Scholars
invariably place Balanchine within an “apostolic succession” of ballet mas-
ters “extending back in time through Petipa, and Didelot before him, to
Noverre and Lully”, the founding fathers of the genre.2 Often, his irrever-
ent attitude towards classicism in his early work is dismissed as lacking the
mature, pure-dance approach of his later years. Balanchine himself set the
tone of this interpretation by allowing more unorthodox works, including
Le Bal, to fall out of repertory – and by altering the choreography of such
early ballets as Apollon musagète (Apollo, Leader of the Muses, 1928) in order
to stress their continuity with the classical tradition.
Yet Balanchine’s early ballets, and the outraged responses of knowledge-
able observers like Levinson, deserve to be taken seriously. Audiences of the
1920s saw in Balanchine’s work an aggressive, even violent anti-classicism:
an eagerness to “abuse” ballet by splicing it with modern and popular dance,
vaudeville and gymnastic routines. This pastiche of different types of move-
ment – which to late twentieth-century eyes seems a harmonious synthesis –
struck its original viewers as a deliberately disjunctive, chaotic jumble of
clashing styles. Placing the danse d’école in unfamiliar contexts, Balanchine’s
early choreography constituted a “composite, hybrid, inorganic spectacle”
akin to the mismatching assemblage of architectural fragments featured in
Giorgio de Chirico’s set and costume designs for Le Bal (see Fig. 38).3
Balanchine’s fusion of classical ballet with other dance forms did not
just break the compositional rules codified by Marius Petipa in the nine-
teenth century. As several critics recognised, these works also deconstructed
classicism’s technique and structure, exposing its underlying conventions.
Balanchine thus pitted himself directly against an emerging definition of
classicism articulated most eloquently in Levinson’s writings of the 1920s.
[237] Levinson conceived of classical ballet as a pure and autonomous art form
238 Juliet Bellow

Figure 38 Giorgio de Chirico, costume for a male guest in Le Bal, 1929.

based on universal principles of corporeal symmetry and grace.4 “The clas-


sical dance is not a label,” he insisted, but “a determined and conscious
servitude to spirit and body accepted with a view to a highly elevated aes-
thetic result. It is not a commodity, because it can only be realised through
disinterested sacrifice.”5 At the beginning of his career, Balanchine refused
classicism this status as a timeless, eternal order by treating it as simply one
amongst a panoply of disparate styles of movement.

Balanchine came of age as a choreographer in the wake of Marius Petipa’s


death in 1910 and the October Revolution of 1917 – events that gave
239 Balanchine and the deconstruction of classicism

rise to a contentious debate about the relevance of classical ballet within


the Russian dance community. While some choreographers and critics
denounced classicism as bourgeois, decadent and obsolete, others viewed
it as the foundation for a new Soviet style of dance. Under the auspices of
Anatoly Lunacharsky’s State Commissariat for Enlightenment, the Bolshoi
and Petrograd (formerly Maryinsky) theatres continued to stage Petipa’s
most famous works, including La Bayadère, The Nutcracker, Swan Lake and
Sleeping Beauty. Surprisingly, even such radical choreographers as Fyodor
Lopukhov and Kasyan Goleizovsky worked from within the classical tra-
dition by tweaking its basic positions, components and sequences. After
Balanchine left the Soviet Union in 1924, he witnessed similar experiments
by choreographers in Serge Diaghilev’s Ballets Russes troupe. Critics noted
despairingly the use of classical dance in Léonide Massine’s Pulcinella (1920)
“to create character, just as the czarda, the bolero and the jig [functioned
in] . . . Coppélia” and the “geometrical dessication of its vocabulary” in Bro-
nislava Nijinska’s Les Noces (The Wedding, 1923).6 Several reviews claimed
that rather than enriching classical ballet, these works impoverished the
genre by making spectators feel as if they were attending a lowly circus or
variety show.
Balanchine’s work of the 1920s featured many of these choreographers’
most infamous distortions of the danse d’école. His ballets broke with the
classical ideal of bodily symmetry and balance, employing diagonal lines
and uneven groupings, high leg extensions, extreme backbends, deep splits
and overhead lifts. He invented unconventional forms of partnering and
support in the pas de deux: in La Pastorale (1926), the ballerina executed
développés over her partner’s head, performed a promenade turn in which
the male dancer pivoted her by supporting her calf, and assumed the “clip-
on” pose, with her leg hitched to the danseur’s torso. Le Bal’s pas de deux
likewise included a lift in which the ballerina, facing her partner, performed
a plié on pointe, then jumped onto his chest, arching back.7
As this acrobatic sequence suggests, Balanchine juxtaposed classical or
quasi-classical elements with movements perceived to be foreign to the
danse d’école. He appropriated steps from folk and character dances, such
as the jig, and from contemporary social dances including the foxtrot and
cakewalk. Flexions of the torso in his works alluded to the modern dance
vocabulary of Isadora Duncan. Intending to liberate the dancer’s midsec-
tion, Duncan designed such movements in opposition to what she saw as the
artificiality of classical ballet, symbolised by the constraints that the corseted
tutu imposed upon the dancer’s body. Balanchine also incorporated non-
dance movements into his early choreography. Militaristic marches and
music-hall gymnastics appeared in these ballets, as did sequences reminis-
cent of Vsevolod Meyerhold’s biomechanical exercises: a system of theatrical
240 Juliet Bellow

training in which repeated motions allowed actors to transform their bod-


ies into rationalised, controlled machines. In employing such an array of
movement styles, Balanchine not only flouted the compositional rules that
isolated classical steps from character dances and mime, but he also delib-
erately brought together movements that were aesthetically distinct and
fundamentally incompatible in their underlying philosophy.
Moreover, abrupt transitions from one movement to the next exagger-
ated their inherent dissonance: the solo variation for Le Bal’s ballerina, for
example, incorporated constant alternations of pointe and heel, as well as
jumps from fifth position to second in the air to fifth again.8 The middle
of this leap, with legs spread wide, transformed the dance into an acro-
batic routine that contrasted with its classical preparation and landing.
Many viewers objected to such alterations of classical technique, accusing
Balanchine of treating the dancers “like a docile and malleable material”
and forcing them to “maneuver about jerkingly” with “angular movements
[and] starchy gestures”.9 Crude and awkward, this choreography seemed
an assault on ballet’s fundamental principles – a “deformation . . . a draw-
ing and quartering of the pure and graceful forms of the danse d’école”,
according to Levinson.10
While most critical commentary focused on his innovative (and pur-
portedly torturous) choreography, Balanchine’s dissection of classicism
extended beyond the level of technique. His works of the 1920s also exper-
imented with the traditional structure and subject matter of nineteenth-
century ballets. He adapted standard plots involving love between human
male and otherworldly female characters, which traditionally gave rise to
a confrontation of the “real” world (dominated by national and character
dances) with a dream or fantasy world (in which classical dance is privi-
leged). La Chatte (The Cat, 1927), for example, was based on one of Aesop’s
fables in which a young man falls in love with a cat, and persuades the
goddess Aphrodite to give her human form. When she chases a mouse, the
woman turns back into a cat, and the young man dies of a broken heart.11 La
Pastorale translated this cross-species romance into modern-day characters:
a movie starlet and a telegraph boy who meet on the set of a film.
That Balanchine used these flimsy scenarios to parody the standard bal-
letic format is borne out in The Triumph of Neptune (1926), which multi-
plied the two acts of romantic-era productions into ten tableaux alternating
between real and fictional worlds.12 The ballet’s narrative of a foreign cor-
respondent who decides to travel to Fairyland for a story also allowed Bal-
anchine to play with the conventional separation of character from classical
dancing. Just after completing her solo variation for a classical pas de deux
with Serge Lifar, Alexandra Danilova “ran quickly into the wings and put
trousers on under [an] opera- or knee-length, tutu, ran back out onstage,
241 Balanchine and the deconstruction of classicism

and danced the hornpipe”.13 Cheekily combining the classical tutu with a
more “realistic” costume typical of character dance, Balanchine referenced
Petipa’s strict rules of balletic composition in a “mélange of Taglionesque
flight and fairground flashiness” that Levinson took to be a “deliberate par-
ody of classical ballet”.14
Nowhere is Balanchine’s deconstruction of classicism more evident,
however, than in his pas de deux, the “traditional classic ballet centerpiece”,
as he later termed it.15 As noted earlier, Balanchine invented unusual forms
of partnering, many of which exaggerated the gendered conventions of the
pas de deux.16 The female dancer was spun, twisted, held, lifted, dragged –
in short, manipulated like an inanimate object. In Le Fils prodigue (Prodigal
Son, 1929), still in repertory at the New York City Ballet, the character of the
Prodigal Son holds the Siren by her armpits and lowers her down into a deep
split, then lifts her to standing. At another point in the pas de deux, the Siren
arches around the Prodigal Son’s body, forming a belt around his torso; he
holds her at the groin and waist and lowers her to the floor. A moment in Le
Bal similarly featured the character of the Young Man straddling his partner,
pinning her in place as she lay on the floor. Moving beyond the established
vocabulary of partnering and support, such sequences emphasised the male
dancer’s control of the ballerina’s movements.
Yet the female characters in Balanchine’s early works also remained elu-
sive sources of fear and desire: his heroes’ fascination with their partners
translated into a kind of passivity. In both La Pastorale and Le Bal, the male
lead pursues the female dancer, allowing his movements to be dictated by
hers – and dies when his quest fails. This lack of male agency was magni-
fied by Balanchine’s predilection for tall ballerinas. His male dancers often
were the same height as, or shorter than, their female partners, thus reduc-
ing the association of masculinity with physical domination. Balanchine
heightened this reversal by stressing the female dancer’s weight, at times
making her seem an almost unbearable burden (see Fig. 39). Supporting
the ballerina from literally debased positions, his male dancers knelt, sat,
and reclined, as when the Telegraph Boy in Le Pastorale turned the Star in
arabesque from the ground, holding her by the knee.17 Likewise, in the first
encounter between the Prodigal Son and the Siren, the male dancer sud-
denly is thrust beneath his partner’s legs; she straddles the nape of his neck
and slides down his back. Drawing out the complexity of the classical pas de
deux, Balanchine’s dancers oscillated between activity and passivity, agency
and objecthood.
Balanchine’s Apollon musagète is described as the choreographer’s first
truly neoclassical work – a pure-dance ballet blanc that marked the end of
his parodic experiments with the danse d’école.18 However, the ballet also
can be read as a profusion of distinct classicisms and neoclassicisms that
242 Juliet Bellow

Figure 39 Alice Nikitina, Felia Doubrovska, Lubov Tchernicheva, Serge Lifar in Apollon musagète
(Apollo, Leader of the Muses), 1928, choreography by George Balanchine, music by Igor Stravinsky.
First performance 12 June 1928, Théâtre Sarah-Bernhardt, Paris. Ballets Russes de Serge
Diaghilev.

refuse to cohere. Juxtaposing the gods of antiquity, the seventeenth-century


ballet d’action, and the nineteenth-century Petipa tradition – all forms that
have been described as “classical” – Apollon musagète exposed the tensions
underlying the concept of classicism. The ballet subtly demonstrates that
the meaning of this slippery term has changed over time, and carries dif-
ferent connotations when applied to visual art, music and dance. A close
reading of Apollon musagète reveals the ways in which Balanchine and his
collaborators strategically heightened the confrontation of these various
“classicisms”.
Apollon musagète’s characters and story come from classical Greek
mythology: the god Apollo in his role as master of the Muses, represented
by Calliope (muse of poetry), Polyhymnia (muse of mime) and Terpsi-
chore (muse of dance). As the souvenir programme for the Ballets Russes’
243 Balanchine and the deconstruction of classicism

1928–9 Paris season noted, Apollon musagète began with a short prologue
“representing Apollo’s birth” that Balanchine later removed from the
ballet.19 Leto, Apollo’s mother, kneels to the ground and gives birth to the
god; two goddesses arrive, swaddling him in white clothes and a golden belt,
present him with nectar and ambrosia, and carry him off to Olympus. The
second tableau centres around a “Judgment of Paris” scenario, with each
muse demonstrating her art for Apollo in a solo variation. Apollo chooses
Terpsichore as his favorite muse and they dance a pas de deux, followed
by a coda with all four dancers. “These allegorical scenes end”, according
to the programme notes, “with an apotheosis in which Apollo leads the
muses, Terpsichore at the front, to Parnassus, which will henceforth be their
residence”.
In addition to its explicit invocation of the gods and demi-gods of clas-
sical antiquity, the ballet implicitly alluded to the seventeenth-century neo-
classical court ballets of Louis XIV and their subsequent development into
the eighteenth-century ballets d’action. The Sun King often used the figure
of Apollo as his symbol and alter ego, playing the role in several ballets,
including Le Ballet de la nuit and Les Noces de Pelée et Thétis, which also
featured his Muses. Balanchine’s inclusion of sun imagery, with the Muses’
extended legs creating a burst of rays behind Apollo, further cemented Apol-
lon musagète’s references to seventeenth- and eighteenth-century neoclassi-
cism. So did André Bauchant’s set, with its incarnation of Mount Parnassus
and Apollo’s chariot onstage. One critic noted that “nothing is lacking in
[the ballet’s] wilful positioning in bygone days, not even the flying mecha-
nism hitched to cardboard horses, as in an opera-ballet by [Jean-Philippe]
Rameau”.20
This reference to one of the French Royal Academy’s most famous
eighteenth-century composers was just: with its emphasis on order, clar-
ity and simplicity, Stravinsky’s score for Apollon musagète deliberately
recalled that era’s musical classicism. Moving away from the constant
shifts in tempo, rhythm and tonality that characterised Stravinsky’s ear-
lier works for the Ballets Russes, Apollon musagète established a structure
of theme and variations that remained stable throughout. The critic for
the Nouvelle revue française termed this score a “notary’s contract”, arguing
that “in abandoning Bach and Handel”, Stravinsky “appeared to want to
revive the tradition of eighteenth-century French ballet, and even recalled
Lully”, Louis XIV’s chief composer.21 Apollon musagète thus conveyed to
its first listeners a self-conscious flatness or “intentional frost”, as critic
Louis Laloy put it, produced by Stravinsky’s pastiche of eighteenth-century
classicism.22
Balanchine, of course, added Petipa’s classicism to the mix, complete with
the knee-length tutus worn by the three Muses. As Levinson noted in his
244 Juliet Bellow

review, Apollon musagète “submits itself to the form, always common in the
classical ballet, of theme and variations”, with traditional solo and ensemble
dances and a pas de deux.23 Yet, as in his other works of the period, Balanchine
added “angular attitudes”, “cut and dried movements”, and “vague acrobatic
impulses” abutting the graceful steps of the danse d’école.24 For example,
Terpsichore begins her solo variation with a series of classical steps: coupés,
grands jetés, and soutenu turns. This traditional variation quickly morphs
into unfamiliar territory as the dancer twice repeats a sequence consisting
of two consecutive relevés in arabesque à la seconde with her arms overhead
in fifth position. She then lowers herself down from her pointe and ambles
around in a flat-footed circle; arranging her arms in a diagonal line, with
hands flexed at the wrist, she breaks with classical rules of alignment in her
upper body. This sequence is accompanied by a syncopated musical phrase
that emphasises the disjunctiveness of the two different forms of movement.
Later in the same variation, Terpsichore assumes a position straight out of
La Sylphide, with her front knee in plié, her back leg angled back grace-
fully. She then pivots to face her back leg, which she raises onto pointe and
bends deeply, her knee extending beyond her planted foot to form a hard
angle.
The pas de deux between Apollo and Terpsichore also juxtaposed
classical and unclassical movements. The dance begins with Apollo seated
on the floor. Terpsichore takes his hand and performs a développé, then
strides over his shoulder and turns in arabesque, allowing her extended leg
to pass over his head, and takes a seat on his raised knees. An equally unusual
sequence occurs later when Terpsichore stands in relevé in arabesque, sup-
ported by Apollo, who holds her waist in the traditional manner. He then
grabs her inner thigh and lifts Terpsichore overhead; with her torso draped
over his back, her legs remain in a split, as if performing a leap upside down.
Apollo lowers her down and both dancers lean forward, lunging deeply.
Together, they repeat the sequence in Terpsichore’s variation in which she
moves from grand battement to flat-footed circle, this time with Apollo’s
support. Later, in the famous end to this pas de deux, Apollo kneels on the
ground as Terpsichore performs a grand battement développé behind him.
He lifts her onto the back of his neck and she balances there. Both partners
sweep their arms forward and back, as if swimming. Apollo places Terpsi-
chore back down on the ground as he rises up, one foot forward and slanting
his body diagonally; she rests her body on his as they arch their backs in
unison.
While admitting in his review of Apollo musagète that “some [of Bal-
anchine’s] groups are ingeniously constructed”, Levinson seconded Louis
Laloy’s assertion that the ballet conveyed an air of “intentional frost”. Bal-
anchine’s aim, he declared, was “that of ‘refrigerating’ classical ballet” in a
245 Balanchine and the deconstruction of classicism

manner that “borders on parody”.25 Balanchine preserved the danse d’école,


Levinson seems to say, but not the hierarchy within which it occupied the
premier rank. To put it in Levinson’s terms, Balanchine’s choreography of
the 1920s treated classicism as a “label” or “commodity” – as an artefact or
set of arbitrary conventions rather than a universal principle of movement
transcending time and place.
In a 1971 essay entitled “The Greatest Master of them All”, Balanchine
claimed that “the movements that Petipa used are essentially the same move-
ments we know today”. While on occasion “I myself turned those steps upside
down”, he wrote, “the language of classic dance is unchanging, universal and
eternal”.26 When did Balanchine come around to Levinson’s point of view?
What precipitated this change of heart? Was it simply a maturation that
moved him away from the rebelliousness of his youth, as most scholars
maintain? Did it emerge from his relationship with Lincoln Kirstein, or
reflect the turn towards formalism in the United States after the Second
World War? Or was Balanchine responding to the tastes of an audience for
ballet that he had helped to create? Answers to these questions are specula-
tive, stymied in large measure by Balanchine’s constant rewriting of his own
history – rendering himself, in retrospect, into an apostle of the classical
tradition.
21 The Nutcracker: a cultural icon
j e n n i f e r fi s h e r

In 1892, when The Nutcracker premiered at the Maryinsky Theatre in


St Petersburg, it received mixed reviews. Like other Imperial Russian bal-
lets, it featured flashes of brilliance from a talented cast, but it had a rather
ordinary plot about a girl who receives a magical Christmas gift. Even the
Tchaikovsky score failed to win unequivocal approval because it was more
symphonic than the music that usually accompanied a ballet at the time.
No one suspected The Nutcracker would eventually become the most pop-
ular and most often performed ballet in the world. This transformation
from the least respected of the three Tchaikovsky ballets (the other two are
Swan Lake and The Sleeping Beauty) into the most visible and lucrative of
all classical ballets, is a story of virtual immigration. There are no borders
at which ballets must clear customs and The Nutcracker never actually left
its homeland, as other immigrants do; nonetheless, like all successful trans-
plants, it thrived in a new location by adapting in creative and unexpected
ways. Through all the changes, the ballet has retained much of its “genetic
material” – the Tchaikovsky score, a version of the original libretto, some
ideas and steps from the original Lev Ivanov choreography and the aura of
its distinguished ballet heritage.
The Russian pedigree of The Nutcracker was a strong factor in recom-
mending the ballet to North Americans. St Petersburg and Moscow, like
Paris and Copenhagen before them, were ballet capitals long before rep-
utable schools and companies were established in the United States and
Canada. Imperial Theatre director Ivan Vsevolozhsky first had the idea to
make a ballet based on E.T.A. Hoffmann’s 1816 short story, “The Nutcracker
and the Mouse King”. Simplifying the plot considerably, he wrote the libretto,
with contributions from master choreographer Marius Petipa. Although the
ballet has undergone countless changes since it first appeared, elements of
the following story tend to reappear: at a Christmas party, the mysterious
Herr Drosselmeier gives a nutcracker doll to Clara, also called Marie in
some productions, following Hoffmann’s story. Dressed as a soldier, this
doll represents the tradition of nutcracker carving developed by miners in
the Erzgebirge Mountains of Germany as a way of lampooning figures of
authority. After the guests leave, Clara falls asleep near her nutcracker and,
when a battle between mice and tin soldiers breaks out in her living room,
[246]
247 The Nutcracker: a cultural icon

discovers the doll has come to life. She helps her Nutcracker Prince con-
quer the wicked Mouse King, then accompanies her hero to a land of snow,
where the first significant classical dancing takes place in the form of waltzing
snowflakes. In the second act, Clara and her Nutcracker Prince arrive in the
Land of the Sweets, where plot development halts, while a series of dances
are performed in their honour. These include fancifully imagined ballet ver-
sions of national dances, and those performed by swirling bits of candy, a
few storybook characters, and waltzing flowers. The reigning monarch, the
Sugar Plum Fairy, partnered by her cavalier, appears for her pas de deux near
the end of the ballet. Then, Clara says goodbye to her fantasy land.
Tchaikovsky had been commissioned to write the score, but he had
doubts about whether the story would provide as much scope as that of The
Sleeping Beauty. Still, he went to work, following the meticulous written
instructions from Petipa as to what kind of music each scene required. In
letters, Tchaikovsky worried about the quality of his music, but it would
be hard to separate his disparaging remarks from the emotional turmoil of
his life at the time. In the spring before the ballet premiered, he received
ovations after conducting some selections from The Nutcracker as a concert
suite. Petipa may also have had doubts about the ballet’s potential, and
before rehearsals began, he pleaded illness and turned the project over to
his assistant, Lev Ivanov. There was no dearth of talent to help Ivanov with
Nutcracker chores. His dancers vied with visiting Italian ballerinas to amaze
audiences with their athleticism and charm and together with the resources
of copious Maryinsky designers, seamstresses and set builders, a certain
amount of excellence was practically guaranteed.
On opening night in January 1892 The Nutcracker evidently had a few
problems. The critics praised only parts of the score and a few group dances
and individual performances. They condemned other aspects of the ballet
in such definitive terms that those comments are perhaps best remembered:
“For the artistic fate of our ballet it is yet one more step downwards”, one
critic lamented.1 The plot was deemed unlikely and imbalanced, the battle
scene called chaotic, and the use of children in major roles considered a
mistake. Nor did it help that Antonietta Dell’Era, the first Sugar Plum Fairy,
appeared very late in the evening, near midnight, because Tchaikovsky’s
opera Iolanthe premiered on the same programme, before the ballet.
It took a few decades for Tchaikovsky’s ballet scores to gain acceptance as
eminently “danceable” masterworks, but the composer died in 1893, never
knowing the lasting impact his ballets would have. The Nutcracker was slower
than most to gain popularity and then, not in its homeland. Although the
Russians did not abandon The Nutcracker, neither did they embrace it very
firmly. It remained in the repertory of the Maryinsky (known as the Kirov
later) and other Russian and Soviet ballet companies, occasionally enjoying
248 Jennifer Fisher

a revival or appearing in a new version, but its reputation languished. Often,


only the second act appeared on a mixed programme, or it was relegated to
school performances and dismissed as a children’s ballet, not serious enough
to match the emotionally sophisticated score.
In the second half of the twentieth century, The Nutcracker came into its
own. It all began when Russian artists started to travel across Europe in the
decades just before and after the First World War, taking various versions of
their classical repertoire with them. Bits of Nutcracker music showed up out
of context in works performed by Diaghilev’s Ballets Russes in Europe, and
Anna Pavlova’s company toured the world with a version of the snow scene
choreographed by Ivan Clustine. This one-act ballet, called “Snowflakes”,
featured the former Maryinsky ballerina in a pas de deux and may have
started today’s tradition of having a “Snow Queen” in that scene (there was
none in Ivanov’s original).2 The Russian Revolution and two world wars
made returning to Russia difficult for many dancers and choreographers,
so they ended up touring and settling abroad. Memories of The Nutcracker
travelled with them, and in London the full ballet was staged in 1934 by the
Vic-Wells Ballet, with the help of Russian émigré Nikolay Sergeyev.3
It was not until The Nutcracker reached the United States in the 1940s
that it took root in a more significant way. The culturally young country
welcomed a ballet with a dual personality, on one hand featuring serious
music and classical vocabulary, and on the other, an atmosphere of child-
like fun and fantasy. In 1940, Alexandra Federova made a condensed version
that criss-crossed the continent for a few decades with the Ballet Russe de
Monte Carlo, one of the companies formed in the wake of Diaghilev’s death
in 1929.
Another introduction to The Nutcracker came in the unlikely form of
the 1940 Walt Disney film Fantasia, which began with a segment featuring
animated figures “dancing” to selections from the ballet’s second act. It
may well have introduced many people who would never go to a ballet
performance to the graceful, rhythmic aesthetics of ballet. Disney animators
used dancers as models, and made a group of ordinary thistles perform a
respectable Russian character dance. The rest of the figures often move like
a corps de ballet: fairies skating on frozen ponds, flowers spinning down
a stream or fish swirling their fanned tails and fins. If the “high” arts that
came from Europe and Russia seemed remote and elitist to many Americans,
cartoons did not; Fantasia, in many ways, brought ballet “down to earth”
for them.
The first full-length American Nutcracker came about in San Francisco in
1944, through a series of coincidences. Willem Christensen, artistic director
of the San Francisco Ballet, had begun to do a ballet at Christmas the year
before, because the city’s War Memorial Auditorium happened to be free.
249 The Nutcracker: a cultural icon

The next year, local Russian immigrants convinced him to try a Nutcracker.
He had never seen the full ballet and therefore asked for advice from George
Balanchine and Alexandra Danilova, ex-Maryinsky dancers by then based
in New York. Danilova dutifully tried to recall the original choreography,
but Balanchine advised Christensen to make his own version.4 It was a
course Balanchine himself took in 1954, when his New York City Ballet
needed a classical challenge, as well as an evening-length crowd pleaser.
The choreographer was well suited to both tasks, he had been schooled in
Russian imperial style, which he energised by incorporating modernistic
streamlining and jazzy elements from African-American cultural forms.5
He seemed to have a special affection for The Nutcracker, had danced in it
at the Maryinsky while still a student. In interviews, he remembered old-
fashioned Christmases (though he thought of Russian Easter as a much
more festive time), and he added touches to his Nutcracker that made it
particularly appealing in his new homeland.6 Even with a relatively small
budget at the time, he insisted on having a tree that expanded impressively
to Tchaikovsky’s soaring crescendos, snow that fell so quickly that dancers
traced paths in it across the stage, and flying reindeer to whisk Clara away.
Balanchine’s commitment to training young dancers dovetailed nicely
with the audience’s interest in seeing their children onstage. In his party
scene, young dancers break out of the formal, starchy mode seen in other
versions, playing leapfrog, bursting with emotion, and urging the comically
scary Drosselmeier to perform magic. Balanchine created a picture-postcard
Christmas, with Marie’s mother coming to check on her daughter after she
has fallen asleep in the living room. To a sweetly serene violin solo, she crosses
the stage holding a candle, then covers her sleeping daughter solicitously
with a shawl and closes a window against draught.7
With all the nostalgic touches and special effects, there is still enough
complex, technical dancing to keep aficionados happy, so that Balanchine
seems to have succeeded where other choreographers failed: he trusted the
music, his choreography and the holiday season to provide enough reso-
nance for both a good ballet and a new holiday tradition. The “immigrant
Nutcracker” was indeed fortunate to be adopted and adapted by a “country-
man” who was already established in the United States. Because of New York
City’s centrality in the world of American ballet and Balanchine’s influen-
tial artistic leadership, his production had a major impact on the “annual
Nutcracker” trend. Not only did other choreographers start thinking about
the financial benefits from return ticket buyers, but dancers who left the
company staged a Nutcracker wherever they settled. City Ballet sometimes
took its production on tour (even in the summer at first) and used local chil-
dren, further planting Nutcracker seeds in various communities. In 1957 and
250 Jennifer Fisher

1958, an even larger audience saw the ballet during Christmas day television
broadcasts, with Balanchine himself as Drosselmeier the second year.
Starting in the 1960s, scores of North American ballet companies
mounted their own Nutcrackers, with newly created choreographies or with
influences from elsewhere, until it seemed there was a production anywhere
that had a ballet school and a Tchaikovsky recording. An influential 1985
video of a Royal Ballet production, with choreography credited to Ivanov
and Peter Wright, was perhaps as widely imitated as Balanchine at first.
Perceived to have strong connections to the first St Petersburg production,
it benefited from the incomplete Sergeyev notation, as well as research con-
tributed by musicologist Roland John Wiley, who has studied the original
Nutcracker extensively. But it was not long before choreographers started
to go their own way. Although most of them created a picturesque version
of Christmas traditions inherited from Germany and Victorian England,
some choreographers opted to shift its location to represent their own com-
munities more directly. In cities with a significant immigrant population
of Germans or Scots, Clara’s home was located in the German section of
an American town, or became the scene of Scottish country dancing in
kilts. A Mexican piñata and cowboys found their way into productions in
the American south-west; hula was added in Hawaii and hockey in Canada.
More than one urban Nutcracker has reflected inner-city realities, and satiric
versions have featured political commentary that outshines the dancing. By
the 1990s, there even was a Barbie film version, starring the popular doll
as the most plastic of Sugar Plum Fairies. In 2001 English National Ballet
struck a deal with toymaker Mattel who sponsored the six-week run of The
Nutcracker around Christmas at London’s Coliseum with $85,000, in order
to cover the deficit the ballet had incurred.8
Perhaps because of its ubiquitous nature, choreographers have felt com-
pelled to put a unique stamp on The Nutcracker. Both John Cranko and John
Neumeier, American choreographers who made careers in Germany, turned
the ballet’s Christmas gathering into a birthday party for Clara in their ver-
sions, in 1966 and 1971 respectively. Graeme Murphy’s 1992 adaptation
for the Australian Ballet makes Clara a Russian ballerina who emigrates
to Australia. In England, Matthew Bourne set his first act in an orphan-
age straight out of Dickens’s Oliver Twist. Two of the more notable depar-
tures in the 1990s were Mark Morris’s The Hard Nut, and Donald Byrd’s
Harlem Nutcracker. Morris set the ballet in the 1960s, adding an op-art set,
cross-dressing and much irreverence, as well as his own brand of modern
dance musicality. Byrd changed the location to current-day Harlem, where
Clara is a grandmother who has a flashback to her youthful days during the
Harlem Renaissance. Her “Land of the Sweets” becomes “Club Sweets” and
swings to the rhythms of Duke Ellington’s arrangement of the score.9 Byrd’s
251 The Nutcracker: a cultural icon

Figure 40 The American department store chain Lord & Taylor announces its presentation of The
Nutcracker in the 1974 catalogue.

Nutcracker, like other versions with African-American influences, changed


Euro-American customs and names to reflect another cultural reality; it
featured a gospel choir in the first act, and a jazzy “Sugar Rum Cherry” in
the second. In 2000 Maurice Béjart used his own life to create a Nutcracker
for the Théâtre Musical de Paris-Châtelet. He replaced the mother and Clara
characters with his own mother and himself as a young boy growing up in
Marseilles, also adding the character of Petipa to represent Russian ballet
tradition.
In many American locations, the impulse to “customise” the ballet
has reached epic proportions: There have been tap dancing Nutcrackers,
Nutcrackers on ice, dance-along versions (bring your own tutu and tiara),
and even a short-lived production in Southern California that tried to tell the
story with bharata natyam. Every ballet company that makes an adaptation
252 Jennifer Fisher

of The Nutcracker seeks an ingenious way to keep audiences interested. Some


convince local celebrities or politicians to appear in walk-on roles. Or they
surround the performance with related events – a “Nutcracker Fair”, a “Tea
Party with Clara”, or a Nutty Nutcracker, a performance in which dancers
switch roles or send up their characters in some way. They also entertain
audiences with ever-expanding lobby boutiques that sell nutcracker dolls
of all sizes, as well as Nutcracker themed T-shirts, socks, scarves, foodstuffs,
placemats, finger puppets, soap and stationery (see Fig. 40).
Along with the wide popularity of The Nutcracker has come a certain
amount of scepticism about its artistic merit. Can a ballet featuring such
childlike glee and spin-off products be taken all that seriously? Aficionados
and critics have tended to look down on it as too accessible, too lopsided
in plot (all story in the first act, all dancing in the second) and too full of
children. But these elements, echoing critical condemnation back in Russia,
only seemed to recommend the ballet in North America. Accessibility was
a good thing for newcomers to ballet, helping to educate and build audi-
ences; and the story’s quick switch from the everyday world to fantasy did
not seem so unbelievable; it resembled the great American rags-to-riches
dream. Another discourse that grew up around many Nutcrackers, especially
those by regional and amateur ballet companies, was that of “community
togetherness”. Many productions rely on volunteers to mount the ballet,
often involving whole families, onstage and backstage. For them, and even
for audiences who identify strongly with their local professional company,
the yearly Nutcracker has evolved into a secular holiday ritual that you cel-
ebrate with family, real or imagined.
Edwin Denby was one of the first American dance writers to provide a
meaningful subtext for the ballet. In the 1940s and 1950s he used a psycho-
logical framework and a light tone in deciding why the simple structure of
the ballet works. At the seemingly ordinary Christmas party, he said, with
its grounded dancing and fractious squabbling, we can imagine the social
pressures that make Clara’s childhood fraught. Then, the quick transition
into fantasy provides a release for her, and for the audience, because friendly
characters provide stability and warmth with their graceful, clear dancing.
It is really a dream about Christmas, Denby proposed, “since it succeeds in
turning envy and pain into a lovely invention and social harmony”.10 For
Denby, and perhaps for many people in the Nutcracker’s new home, classical
dancing spoke a language of hope and reassurance, and the ballet’s focus
on children enabled audiences to recall youthful dreams and excitement.
Denby’s emphasis on underlying themes and the way the dancing itself con-
veys moods make perceived plot deficiencies less problematic. Look beyond
the narrative, he advised. The ballet is not confusing or meaningless to a
viewer who can link the real and fantasy worlds well enough in his or her
253 The Nutcracker: a cultural icon

mind. Nor is the second act flawed for stalling the story to offer a suite of
dances, he said. It allows the viewer to savour each dance individually, then
proceed, satisfied, to the next.11
When lauding The Nutcracker as a perceived site of Christmas ideals, it is
also necessary to note the potential pitfalls of the ballet’s so-called “national
dances”. Appearing in the second act, they are called “Hot Chocolate”,
“Tea”, “Coffee”, and “Trepak”, but are more commonly known as “Spanish”,
“Chinese”, “Arabian”, and “Russian”. These dances are meant to represent
Clara’s encounter with an innocently exotic treat, but they have in the past
also reflected negative stereotypes of the “Other”, inherited from nineteenth-
century ballet. Not surprisingly, the least contentious of them is usually
“Russian”, which often is staged as an athletic character dance, and other
times is a dance of candy canes (Balanchine followed Ivanov in the latter
choice). The “Spanish” dance also poses few problems, because it tends to
resemble some form of dancing actually done in Spain, except that it is more
“balleticised”. But “Chinese” and “Arabian” rarely resemble any dance done
in Asia or the Middle East; instead, some versions can feature distorted,
insulting stereotypes.
Though it is hard to generalise, and impossible to consider every spe-
cific production, there are certain trends when it comes to staging these
dances.12 They can be divided into two basic categories: those that are more
or less respectful and naively combine elements of Asian or Middle East-
ern dance with ballet; and those that are oblivious to potential ethnic slurs,
reproducing iniquitous images of lounging villains, obsequious servants
and loose women. The two key discourses for understanding the differ-
ence between these “insensitive and insulting” and “naive but respectful”
versions are “Orientalism” and “artistic hybridisation”. Orientalism, as first
defined by Edward Said, refers to the ways in which Western colonialists
(specifically the British, French and Americans) stereotyped the Middle
East as “a place of romance, exotic beings, haunting memories and land-
scapes, remarkable experiences”.13 Whether or not ballet audiences who
see The Nutcracker are familiar with this definition, they may be absorb-
ing its negative reverberations when “Arabian dance” soloists portray one-
dimensional sensualists, or Chinese dancers bob and scurry in clown-like
fashion.
On the other hand, some versions may simply combine classical lines
with fanciful angles and costumes without suggesting anything more nega-
tive than an incomplete ethnographic understanding. This brings to mind
historian John M. MacKenzie’s proposal that a kind of benign artistic bor-
rowing has always gone on in the world of the arts, and that choreographers
may have a genuine admiration for the cultures in question.14 In other
words, a balleticised “Arabian” or “Chinese” dance can be an artistic hybrid,
254 Jennifer Fisher

Figure 41 Pennsylvania Ballet Company Member Heidi Cruz in The Nutcracker, 2005,
choreography by George Balanchine.

not made “to reinforce colonial power or superiority, but to enable artistic
innovation”.
Some companies have experimented with practitioners from the coun-
tries in question; others have eliminated the dances or changed their char-
acter to something less specific. Whatever the version, there has been little
productive discussion in the ballet world about how the “Chinese” and
“Arabian” dances might have evolved historically and how they might be
staged today. With The Nutcracker’s long history of adjusting to new people
and places, it may be the right classical ballet to continue experimentation
with new strategies in eliminating potentially injurious stereotypes in old
story ballets. Often fitting into the parameters of any classical ballet, it has
also become a nexus point for attitudes, opinions, allegiances and renova-
tions to the classical ballet world itself. Because The Nutcracker appears so
often and is critiqued in so many situations, it can stimulate questioning
of the usually conservative ballet world. How much of ballet’s ethnocen-
tric past will survive? How much of it do we want? How can the glorious
255 The Nutcracker: a cultural icon

nineteenth-century ballets evolve and yet retain core values that are worthy?
(See Fig. 41.)
Although revenues have sometimes dropped off in economic hard times,
the annual Nutcracker tradition still grows. Far from being a children’s ballet
that got out of hand in a country that did not know any better (a common
notion among Nutcracker detractors), it has become an annual phenomenon
that gathers resonance in many ways in many locations. Increasingly, ballet
companies outside North America also make December Nutcracker season.
For many, it is still a Russian ballet, and critics may lament the distance
between current incarnations and the original. But for today’s Nutcracker,
the discourse of “authenticity” regarding St Petersburg origins seems less
relevant than the ways in which the ballet has become meaningful in many
new places. In that case, the idea of repeating The Nutcracker each year does
not seem a bore or an accident; it seems like a powerful tradition.
22 From Swan Lake to Red Girl’s Regiment :
ballet’s sinicisation
z h e n g ya n g w e n

Chinese dance is a riot of identities – court, folk, ethnic and dynastic. It has a
glorious history and is a topic on its own.1 It draws inspiration from martial
arts and “the art of sex” as the quintessential man of letters and leisure Li
Yu wrote: “When people teach girls to sing and dance, they do not really
teach them how to sing and dance but how to be sensual. If you want her
body to be so, then you must have her dance.”2 Several legendary sensual
dancers rose to be Senior Consorts and Empresses during the Han (206 bc–
ad 220) and Tang (618–960) dynasties. Yang Yuhuan, the most beautiful
woman in Chinese history, was a sensual dancer. She captivated the Tang
emperor Xianzong (685–762), became his favourite consort and came to
shape Chinese history.3 The splendid history of court dance, like Chinese
dance itself, waits to be researched. Chinese dance absorbed its properties
and values from many cultures, especially Central Asia, during the Han
and Tang dynasties. This pattern of assimilation continued as the Mongols
conquered China in 1279 and the Manchus in 1644. It developed rapidly
after the Opium War (1839–42) when China began to have direct intercourse
with the world beyond greater Asia. This exposure not only dramatically
changed China’s history but also the very fibre of Chinese culture. Dance in
the form of ballet is a great example of such cultural change.
I cannot discuss the introduction and naturalisation of ballet without
situating it in the landscape of late Qing, Republican and communist China.
Art in general and ballet in particular had to develop within the larger polit-
ical and socio-cultural framework. Many believe that the West viewed the
East as exotic and stereotyped its culture;hence the impact of Western art
and thinking has been labelled cultural imperialism. The study of China
itself falls into four intellectual frameworks: impact-response, tradition-
modernity, imperialism and China-centred. Chinese ballet would fall under
any but the fourth paradigm. The story itself is, however, more complicated
than that. Ballet, like opium and communism, was foreign to China. Impor-
tant in its introduction and naturalisation, similar to the stories of opium
and communism, were agents and mechanisms of transmission.
Mainland historians credit the introduction of ballet to Yu Ronglin,
daughter of late Qing diplomat Yu Geng and sister of author Yu Deling.4
[256]
257 From Swan Lake to Red Girl’s Regiment

Born in 1882, Ronglin followed her father to Japan where he served as


China’s ambassador in 1893. She was exposed to and began to learn ballet.
The family moved to Paris in the late 1890s and Ronglin took classes from the
famous Isadora Duncan. They returned to Beijing in 1903; she and her sister
became the Empress Dowager Cixi’s ladies-in-waiting. Ronglin performed
ballet for Cixi and the court at large. She also began to use ballet techniques
to choreograph Chinese dance; this can be considered to be the beginning
of the naturalisation of ballet in China. One of her enduring works was
Dance of the Lotus Fairy Maiden, inspired by the Zhongyuan Festival where
lotus-shaped lanterns were set floating on water on the fifteenth day of the
seventh month of the Chinese lunar calendar. The old Buddha, as Dowager
Cixi was reverently addressed, took a liking to Ronglin’s dance and to the two
sisters.5 Imperial exposure and endorsement helped promote ballet among
the late Qing elite and upper classes; it had set the tone for the twentieth
century.
Late Qing (1842–1911) was a stirring time as imperial China disinte-
grated and something new was in the making. The Nationalist Revolu-
tion in 1911 ushered in the May-Fourth or New Culture movement where
Chinese intellectuals criticised, if not abandoned, China’s heritage in their
effort to modernise the ancient culture. Yangwu or “foreign dance”, as early
newspapers and pictorials labelled it, did not mean ballet exclusively but
included it.6 Along with other non-Chinese art forms, it was making its
way into treaty port cities like Shanghai where early performances took
place. Russian ballerina Anna Pavlova brought The Dying Swan to Shang-
hai in 1922, while a whole array of foreign dance companies and dancers,
Denishawn and Irma Duncan for example, gave classic and modern dance
performances between 1925 and 1928. They attracted the serious atten-
tion of native artists, both traditional and modern. Denishawn fascinated
the master of Peking opera, Mei Lanfang, whereas Duncan attracted left-
wing writers like Tian Han.7 This was significant as indigenous artists and
progressive men of letters welcomed and began to explore Western art and
dance. China’s cultural elite appreciated, if not completely accepted, ballet in
the 1920s. Soon, first-generation Chinese ballet dancers and choreographers
emerged.
Most important among them was Wu Xiaobang, the founding father
of Chinese ballet. Wu attended an American school in Shanghai in the
1910s and loved modern dance and ballet.8 He went to study in Japan in
1929, came back to found a dance school in Shanghai and gave China’s
first public modern dance-ballet performance in 1935. Wu choreographed
and performed before and during the Japanese war; his works have become
classic repertoire for ballet students in China. Another ballerina who was
instrumental in the spread of ballet was Trinidad-born and London-trained
258 Zheng Yangwen

Dai Ailian or “Mother of Chinese dance”. She studied at the Mary Wigman
and Jooss-Leeder schools and with masters like Rudolf Laban.9 Like many
patriotic overseas Chinese, Dai returned to her ancestral land in 1940 or
during the height of the Japanese war where Madam Sun Yat-sen welcomed
her. Dai, Wu and Sheng Jie were the dynamic trio in the wartime capital
Chongqing. They choreographed patriotic dances and helped mobilise the
war against Japan in an extremely difficult time. Patriotism was injected
into a performing art; this had set the precedent for the communist era. The
trio threw their lot in with the communists at the end of the war. Dai and
Wu became the Chair and Vice-chair of the National Dancers Association
in 1949. The Beijing Academy of Dance was founded in 1954; Dai became
its first Master and Dean. Between 1957 and 1960, the Academy mounted
La Fille mal gardée, Swan Lake, Le Corsaire and Giselle with the help of such
Russian artists as O. A. Yealina.
The 1950s gave birth to a new generation of outstanding ballerinas.
Dance theatre in general, ballet in particular, flourished into the mid-1960s
when the country was in a mood of revolutionary optimism and originality.
Chinese ballet classics, Hongse Nianzijun or Red Girl’s Regiment for example,
emerged. It was choreographed and first performed by the Ballet Troupe
of the Central Song and Dance Ensemble in Beijing in 1964. It tells an old
story, set in the 1920s Hainan Island, Guangdong province. Slave girl Wu
Qionghua was captured every time she ran away from Landlord Nan Batian.
She was caught again, beaten badly and left to die in the coconut forest. Hong
Changqing, Party Secretary of the Red Girl’s Regiment who pretended to be
a rich overseas merchant, was on assignment in close proximity. He saved
Wu and led her to the revolutionary base where she joined the Red Girl’s
Regiment. After much struggle, Wu killed Landlord Nan and rose to become
the Regiment’s Party Secretary when Hong died.
Another, similar classic hit was White Haired Girl. It was choreographed
and performed by the Shanghai Academy of Dance in 1965. It is also a tale
of oppression and class struggle. It seemed ironic that Chinese drama and
class struggle were told in a quintessentially imperialistic and bourgeois
artistic form. It certainly enlightens us about the making of the communist
performing arts and, more importantly, ballet’s naturalisation in China.
Red Girl’s Regiment serves as an excellent example of ballet’s “sinicisa-
tion”. It used the language of ballet, a Western European vocabulary foreign
to the Chinese dance dictionary, to tell a Chinese story in its indigenous
customs and setting. It introduced ballet techniques into Chinese dance
and enriched its dance theatre. First of all, the use of the unique ballet shoes
and poses signalled a new style of dance even though the costumes and
stage settings were still Chinese. Poses specific to ballet were used to portray
Chinese characters whose circumstances were familiar to the Chinese audi-
259 From Swan Lake to Red Girl’s Regiment

Figure 42 Escape Scene from the ballet Red Girl’s Regiment.

ence. Some ballet poses proved better suited to portray narrative moments
than others. A great example is Slave Girl Wu’s escape from Landlord Nan’s
estate. The forty-five degree mid-air pose with her two fists tightened in
the air highlighted her hatred for the Landlord and her determination to
escape (see Fig. 42). This was not a dance vocabulary which the choreog-
rapher could find in the dictionary of Chinese dance because oppression
and class struggle had not been the focus of traditional dance theatre. The
choreographer used ballet techniques to emphasise class struggle, drama-
tise the escape and build a strong character; this laid the foundation for the
climax.
The following act, when Wu changed into the Red Army uniform and
joined the Communist Party, was electrifying (see Fig. 43). To the classic
ballet pose, the choreographer simply added a gesture every communist
understood and millions wished to act out – holding the right fist high above
the head to swear one’s allegiance to the Party. It is a brilliant combination of
260 Zheng Yangwen

Figure 43 Shooting Scene from the ballet Red Girl’s Regiment.

ballet and communist politics. Even more electrifying was the drilling scene.
The all-female regiment with rifles in their hands and short hair in Mao
caps and puttees conveyed the essence of “Red Girls”. This was breathtaking
because it was something the audience could easily comprehend even in
its foreign pose and gear. Traditional Chinese drama had featured female
fighters but none with rifles and short hair in Mao caps, the emblem of
liberation and revolution. Many such scenes in Red Girl’s Regiment and
White Haired Girl signalled the arrival of a different era with its own artistic
language.
If innovation was key in ballet’s sinicisation, even more important was
the integration of traditional Chinese dance techniques that choreogra-
phers skilfully incorporated: the performance would have to be indigenous
because the story took place in the ethnic Li minority village, hence the vil-
lage scene, where young men and women gathered to dance, provided that
native flavour. In essence, it is similar to the Chinese dance in Swan Lake,
Nutcracker and other ethnic dances in Western ballets and operas. Chinese
martial art techniques and drama poses were embedded in the battle scenes
when the “Red Girls” fought their enemies. Such integrating approaches
made it easier for the ordinary Chinese to understand and accept ballet. The
communist regime was injecting revolutionary ideas into the construction
of a new China; their platform was to advance and uphold the interests of
the proletariat. The regime and this unprecedented era therefore demanded
new performing art styles. The introduction of new techniques was wel-
come and opportune; it brought new life to the Chinese dance theatre. It
inspired and facilitated creativity in the process of character-building and
storytelling. It enriched the vocabulary of Chinese dance and enhanced its
narrative power. The process also transformed, if not modernised, Chinese
dance.
Two years after Red Girl’s Regiment, the Culture Revolution (1966–76)
threw the country into chaos. Politically, it was a power struggle between
261 From Swan Lake to Red Girl’s Regiment

Mao Zedong, the ideologue, and his colleagues who were more pragmatic
in their approach to the construction of a new China. The “cultural” aspect
of the revolution involved Mao, his extreme supporters, his wife Jiang Qing
among them, in a campaign to destroy old elements of Chinese culture and
society in order to build a communist utopia in his lifetime. Mao believed
strongly in the superstructure theory of Karl Marx whereas Jiang Qing led
the “gang of four” to define and dictate what communist art and culture
should be. Performing art classics like Red Girl’s Regiment were turned into
what was called “revolutionary model plays”. Jiang, originally an actress, had
a personal interest in pushing forward the revolution in culture and refash-
ioning the arts for the proletariat. She intervened in the making of “model
plays”, that is the remaking of classics like Red Girl’s Regiment. She will
remain infamous for her destruction of Chinese art and culture. However,
the Culture Revolution itself had undoubtedly contributed to the spread
and indigenisation of ballet in China.
Dance was the best device because it could be reduced to a body language
that was so basic that even a child would understand it. The “revolutionary
model play” was directed towards the proletariat; it had to be simplified so
that every worker and peasant could understand it. In the frenzy to popu-
larise the “model plays”, hundreds of Red Girl’s Regiments were produced
around the country as the whole nation was thrown into a fervour of making
revolutionary art. The “model play” politicised ballet and it spread the idea
of the proletarian consciousness and its class struggle through ballet lan-
guage. Many peasants knew about ballet; some could even name the “eight
big revolutionary model plays”. The Culture Revolution on the one hand
disfigured the original art and its sinicised form; on the other hand it helped
to spread ballet among the masses. Red Girl’s Regiment served more than
just its artistic purposes. Like other artistic forms in the communist era, it
helped popularise Marxist ideology and promote its key platform – class
struggle – advance feminism and strengthen communist rule. It also bred a
generation of outstanding dancers, some of whom have won the recognition
of the international ballet community as we can see from Li Cunxin, one of
the ballerinas handpicked by Jiang Qing’s followers. The star, whose fame
depended on the Cultural Revolution, defected while performing with the
Houston Ballet in 1979 and continued her career on the global stage.10
The sinicisation of ballet had much to do with patriotism and politics.
In the late Qing, it was limited to the court and elite. In the Republican
era, it was embedded in the New Culture movement and in the war against
Japan. Under the communist regime, it was used to popularise an ideol-
ogy and deepen authoritarian rule. This government-sponsored political
cultural extravaganza was most effective when it came to the indigenisa-
tion of ballet. It became a household name overnight during the Culture
262 Zheng Yangwen

Revolution. It is now part of a common Chinese vocabulary and artistic


heritage. The sinicisation of Buddhism and opium consumption took a
few hundred years, but the naturalisation of ballet, alongside communism,
took only a few decades. The involvement of a political regime is key to the
quick indigenisation of a foreign art and ideology. China was ready for a
new political thought with which the communists could build a new state;
it was ready for a new performing art genre with which the regime could
educate and re-educate its new/old citizens. China provided the best soil
for both ballet and communism to grow; it gave birth to the most bril-
liant dancers and choreographers. It enriched the language of dance and
vocabulary of ballet. The post-Mao era has seen a revival of Chinese arts
and culture. Chinese dance theatre in general and ballet in particular will
undoubtedly continue to draw inspiration from other cultures in this age of
globalisation.
23 Giselle in a Cuban accent
l e s t e r to m e

In New York, on 2nd November, 1943 Alicia Alonso, a Cuban ballerina,


danced the title role of Giselle replacing a sick Alicia Markova in a perfor-
mance of Ballet Theatre.
In 1943, a Cuban ballet dancer was a rarity, even more if she starred in a
work considered the epitome of the European Romantic tradition. Though
Cuba was immensely popular in the United States in the 1940s, it was by
no means associated with ballet. Since the early 1920s, Hollywood and
Broadway had made Americans familiar with images of Cuba as an exotic
destination offering exuberant women, hot rhythms, beaches, casinos and
endless nights of romance and alcohol. I’ll See You in Cuba (1920), The
Cuban Love Song (1931), Girl from Havana (1940) and Week-End in Havana
(1941) were among the era’s numerous films and musicals with Cuba as
their subject or background. Cuban songs topped American charts while
rumbas, congas, cha-cha-chas and mambos took centre stage in dance halls
from New York to Los Angeles.
Rising to fame as a ballet dancer and becoming one of the most cele-
brated Giselles of the twentieth century, Alonso undermined Americans’
stereotypical images of Cuba and Cubans. She also challenged the assump-
tion that ballet belonged to Europeans. Asked about the reactions that her
nationality elicited at the beginning of her career, she said: “A lot of people
were astonished. I danced the more classical works, like Giselle and Swan
Lake. In my first visits to England and France, the audience could not believe
that a Latina represented classicism. In England, a critic even asked me how
I had the courage to dance Giselle. Regardless of their disbelief, I achieved
success”.1
Alonso’s nationality, always brought to attention by critics, was a prob-
lematic factor in the early years of her career. Impresarios suggested that she
drop her Spanish name and adopt a Russian or British one for the sake of
commercial success. She refused. “I firmly believed that achieving success
under my name would help to debunk false and discriminatory concepts
about Latinos’ possibilities in ballet”.2 Alonso’s national identity was beyond
her name. It emerged through her dancing: “When I performed, something
different was always noticed in my dance. We Latinos have specific features,”
she said in an interview.3 In 1945, the New York Herald Tribune’s critic Edwin
[263] Denby emphasised her ethnic origin: “Alonso is a delightfully young and a
264 Lester Tome

Figure 44 Alicia Alonso in Act i of Giselle.

very Latin Giselle, quick, clear, direct in her relation to her lover. She is pas-
sionate rather than sensuous. She is brilliant in allegro, not so convincing in
sustained grace . . . She has little patience for those slow-motion, vaporous
effects that we Northerners find so touching”.4
Denby, who focused on Alonso’s Latin foreignness in contrast to an audi-
ence of “Northerners”, classified her according to prevalent clichés of the
time. Brilliance of movement, quickness and passion were seen as char-
acteristics that represented the common superficial American views on
Cuban dance stirred by the rumba and conga crazes. Grace, slow motion
and vaporous effects, on the other hand, were contrary to the stereotype.
Hence Denby had to see them as “Northern” features inaccessible to the
dancer from the South.
A decade later, Alonso had achieved artistic maturity; she was receiving
widespread critical acclaim. By 1955, she had been dancing Giselle for twelve
years and had perfected her interpretation of the title role. In the New York
Herald Tribune, Walter Terry described her as “one of the great Giselles
of the time” with an “impeccable classicism”. Yet her national or ethnic
identity remained a significant factor in how critics judged her work. In the
context of Alonso’s undisputable stardom, her geographical origin was now
perceived as enhancing her rendering of Giselle. Terry pointed out “a Latin
warmth which (. . . ) suffuses the entire ballet with a rich range of dramatic
colors”.5
Departing from the opposition between Latin and Northern sensibilities
that Denby had seen, Terry connected Alonso to both Cuba and Europe,
265 Giselle in a Cuban accent

Figure 45 Alicia Alonso in Act ii of Giselle.

pointing out “Latin warmth” and “impeccable classicism”. Most critics


seemed surprised at Alonso’s mastery of classicism, which, after all, was
essentially European with its French, Danish, Russian and British roots.
Here lies the paradox: Alonso’s dancing was a perfect embodiment of
European classicism and yet it was infused with Cuban accents. From her
own point of view, this was not necessarily a contradiction. Cubans perceive
their national identity not in opposition to but in dialogue with the European
culture brought by colonisers and settlers who mixed with the native and
African population of the country. The island takes pride on its links to
Europe. But this does not make the paradox of Alonso’s dancing less striking.
How could something be Cuban and European at the same time? How
266 Lester Tome

could Alonso’s classicism be faithful to its European origins if her dance


had a Cuban tone? How could her Cubanness be authentic if it was tackled
through a European canon?
The key to Alonso’s excellence as a classicist was her discipline, her deter-
mination, and her understanding of the history of the works that she danced.
After first seeing Giselle in a performance by Markova and Anton Dolin, she
fell in love with the ballet and decided that she too would dance it someday.
Even before the opportunity appeared, she began to prepare for it:

I read all the books. I tracked the ballerinas that had performed the part:
how were they, what had been said about them, how were their costumes,
makeup, hairdos and headdresses . . . In 1943, Markova was going to dance
Giselle during the season of Ballet Theatre but she became sick. Nobody
wanted to replace her. When impresario Sol Hurok asked me if I would dare
to do it, I said yes. By then, I had learned not only the part for the title
character but also the choreography for the rest of the soloists and the corps
de ballet. I knew the ballet by heart. I felt completely identified with Giselle,
to the point of having planned even the movements for the fingers and
eyelashes. [After the debut,] I continued my study of the piece, deepening
into its style, defining the profile of each character, questioning the
transitions between scenes . . . I loved studying the lithographs of the
romantic ballerinas from the nineteenth century. I looked at every detail,
paying attention to their way of using every part of the body to resemble a
volatile creature. Looking at their poses and the position of their torsos,
arms and heads, I tried to imagine how these women danced. It was like
putting a puzzle together.6

It was this dedicated study of the background of Giselle together with her
very own interpretation that represented the curious mixture of cultures.
Some of the most influential critics of the twentieth century, such as Walter
Terry or Arnold Haskell, praised her as the most outstanding Giselle of
her generation. In a review that captured the tone of many critics’ opinion
on Alonso, Olivier Merlin wrote for Le Monde in 1966: “In the past, Olga
Spessivtseva, Alicia Markova, Galina Ulanova; today, Margot Fonteyn and
Ivette Chauviré: these are the only great, moving Giselles. Nevertheless,
Alonso, for some mystery, keeps her rank as the leader in this constellation.”7
In a personal letter from the same year, Anton Dolin, the famous Albrecht
who had danced with most of the great Giselles of the era, wrote to Alonso:
“[Before] the intelligence of your interpretation of the ballet Giselle, I lay
myself again at your feet.”8 The paradox now manifested itself through the
tension between ”intelligence” and ”passion”. Dolin’s acknowledgement of
Alonso’s intelligence evidences how she challenged the stereotypical view
that identified Latinos with uncontrolled passion and a lack of rationality.
Alonso’s performances of Giselle were both impassioned and intelligent.
267 Giselle in a Cuban accent

Again and again during Alonso’s career as a ballerina her detailed knowl-
edge of the romantic style was highlighted, together with the blurring
between her person and the character she danced, her psychological insight
into the role, her intense emotional expressiveness, the simplicity and lack
of affectation of her acting, her refined musicality in phrasing, her virtu-
osic technique and the subordination of herself to the stylistic and dramatic
demands of the ballet.
Her own choreographic production of the ballet, premiered in Cuba
in 1948 by the Alicia Alonso Ballet and perfected over the years, achieved
the same level of recognition. In 1972, the Paris Opéra, where Giselle had
been born in 1841, discarded its ossified version of the piece and adopted
Alonso’s. Her production, applauded for bringing the romantic style of the
original back to life, was also incorporated into the repertoire of the Vienna
State Opera, the San Carlo Theatre in Naples, and the Colón Theatre in
Buenos Aires.
But there remained the difficulty for dancer as well as critic of pinpointing
the bodily manifestations of her national identity. There was the inadequacy
of words to describe dance in the first place. The greater problem consisted in
identifying what was essentially Cuban in her dancing. Asked for a definition,
Alonso replied: “Cubanness exists. Who doubts it? But, at the same time,
who is able to define it? Cubanness is not a single thing but many; it is
something alive and difficult to capture.”9 Critics Arnold Haskell and Pedro
Simón, who believed that a Cuban School in ballet exists, faced the same
problem. Haskell acknowledged that he was not able to describe what was
particular about it: “I can recognize it, but it is almost impossible to explain
the movement in words”.10 In a similar way, Simón explained that “a School
is not a phenomenon as defined as a recipe that one can put on paper”.11
Vagueness predominated even when particular Cuban or Latin inflec-
tions were pointed out in Alonso’s dancing. For instance, Terry indicated the
“Latin warmth” that brought a rich range of dramatic colours to Alonso’s
style. But he offered no further explanations on its specific Latin- or Cuban-
ness. In an interview, Alonso suggested that femininity was another Cuban
feature in her dancing. However, it was not clear how she was more femi-
nine than other ballerinas or how her femininity differed from theirs. She
also observed that “phrasing, the mode of joining the steps, is among the
characteristics of Cuban dancers. It is mentioned that we follow the melody
instead of the rhythm.”12 But are melody and rhythm separate in the chore-
ography of Giselle, and is it possible for a dancer to follow one instead of
the other?
In spite of the vagueness and ambiguity of the Cuban elements in her
style, Alonso was conscious that her national identity made her dancing
unique. She nurtured those characteristics and tried to turn them into a
268 Lester Tome

national style: “In those early days I wasn’t exactly clear about the style we
would later develop . . . But I did know that I wanted things to look a certain
way . . . When the Latin, Cuban qualities appeared, they just seemed to come
out, as if the richness of the earth were coming into my art. This was not
calculated. Home and the richness of its heritage were always there. I took
it, just instinctively, and used it”.13
The paradox of Alonso’s simultaneous embodiment of Cuban sensibility
and a European aesthetics turned into an attitude that attempted to be truth-
ful to the factors in a relationship that had hitherto been seen as irreconcil-
able. Alonso achieved complete identification with the European romantic
style and, at the same time, channelled her national identity through her
dance; she was truthful to both a European and a Cuban legacy. Both became
equal instead of Cuban culture subordinate to European tradition.
The artist’s double identity created a curious dynamic between Eurocen-
trism and nationalism that affected how she saw herself and how she was
perceived by the international dance community. Alonso, who initially was
subject to cliché views on Latinos and expelled from the European canon,
became interested herself in developing a national style. In this endeavour,
triumphing in the European repertory was indispensable to attaining recog-
nition for herself as a Latin ballet dancer as well as for her Cuban-inflected
classical dancing. Her repertory included pieces by Cuban choreographers
and twentieth-century Western creators such as Fokine, Balanchine, Rob-
bins and Tudor. Yet, nineteenth-century European works such as Giselle
remained her letter of presentation and she took steps to secure her status as
a legendary performer of the piece. She made sure that popular and academic
publications alike commemorated her in her career as Giselle and assured
her a definite place in history. In a self-celebratory note, the introduction to
a book issued by the company in 1988 for her forty-fifth anniversary in the
role states: “The eternity of the role of Giselle and the work itself . . . owes
a fundamental breath to Alicia Alonso, a resurrection that points towards
the future with the certainty of permanence . . . For future generations, the
pictures of Alicia Alonso in Giselle will be equivalent . . . to the drawings of
Marie Taglioni in La Sylphide, Fanny Elssler in The Gipsy, and Carlotta Grisi
in La Peri.”14 Further reinforcing her status as mythical Giselle, the National
Ballet of Cuba frequently performed gala evenings dedicated to the romantic
ballet that emphasised Alonso’s authority as a performer of that repertory
and her success in passing down its style to successive generations of Cuban
dancers. One of those Romantic Galas “in tribute to Giselle-Alonso” was
offered in Havana on 2 November 1993, as part the celebrations for the bal-
lerina’s fiftieth anniversary in the role. It included landmark romantic works
such as Jules Perrot’s Grand pas de quatre (restaged by Alonso) and frag-
ments from August Bournonville’s Le Sylphide, Napoli and Flower Festival
in Genzano, as well as Arthur Saint-Léon’s La Vivandiére.15
269 Giselle in a Cuban accent

Alonso followed her nationalist agenda persistently; she pursued it at


the same time as she positioned herself as the keeper of a European legacy.
In 1948, she founded the Ballet Alicia Alonso in Havana and, two years
later, she opened the National School of Ballet Alicia Alonso. Her goal was
not only to solidify ballet in Cuba but also to develop a Cuban hallmark in
classical dance. Nationalism had been a strong force in the development of
Cuban culture during the 1930s, 1940s and 1950s, permeating the realms
of philosophy, history, music, literature, drama, musical theatre and the
visual arts. Zarzuela, the Spanish equivalent to opera and operetta, experi-
enced a boom and acquired a defined national profile in the works of Ernesto
Lecuona and Gonzalo Roig during the 1920s and the 1930s. Simultaneously,
the trend known as afrocubanismo drove the attention of classical composers
such as Amadeo Roldán and Alejandro Garcı́a Caturla to the African Cuban
cultural legacy. In 1928, Jorge Mañach published his Indagación del choteo,
an essay on Cuban national identity. In these years, Fernando Ortiz intro-
duced anthropological research as a tool for studying Cuban culture. His
essay “Los factores humanos de la cubanidad” appeared in 1939, a year after
Medardo Vitier published the first history of ideas in Cuba (Las ideas en
Cuba). In 1940, the University of Havana organised the first retrospective
exhibition of Cuban art, covering three centuries of painting. In that same
year, another crucial book by Ortiz examined the formation of Cuban iden-
tity, Contrapunteo cubano del tabaco y el azúcar, while Martha de Castro
published her Contribución al estudio de la arquitectura cubana, an analysis
of Cuban architecture. During the 1940s, Cuban-ness became a conscious
element in the work of national painters such as Wilfredo Lam, Amelia
Peláez, René Portocarrero and Mariano Rodrı́guez. In 1944, a group of
writers led by Vitier, José Lezama Lima and Virgilio Pin̂era founded the
literary magazine Orı́genes, concerned with expressing a national sensibility
in poetry. Two years later, Carpentier presented his pioneer study La música
en Cuba. Also, over the 1940s and 1950s, five histories of Cuban literature
were published, authored by Salvador Bueno, Max Henrı́quez Ureña and
three other scholars.
This national context informed Alonso’s attitudes and evolution as an
artist as well as the development of her company. In the early 1950s, the group
was already staging experimental works by Cuban choreographer Alberto
Alonso, “using Cuban music and our folklore, utilising Cuban elements,
enriching our ballet technique, artistry and way of dancing”, according to
the ballerina.16 Sociedad Pro-Arte Musical,17 umbrella organisation for the
company and the school at the time, preferred Alonso and her company
to perform the nineteenth-century European classics that satisfied the taste
of Havana’s bourgeois audiences. Pro-Arte did not regard the innovative
works by Cuban choreographers with sympathy. The organisation censored
and prohibited some of these pieces, but Alonso pushed them forward and
270 Lester Tome

negotiated a deal with Pro-Arte: she would dance in these new productions
as a guarantee of their quality. In those early years of Cuban ballet, the
credentials and money that Alonso needed to nurture a national style could
be obtained only if she continued to dance the European classical repertory
both inside and outside Cuba. Her nationalist agenda depended on the
Eurocentrism associated to audience’s preferences and box-office decisions.
After Fidel Castro took political and economic control of the country
in 1959, the company became the National Ballet of Cuba. The dancer,
at the height of her career, chose to remain on the socialist island. The
communist government never prevented Alonso from travelling abroad.
But Cuba’s isolation and the tense diplomatic relationship with the United
States had a negative impact on Alonso’s international career. For this loyalty
towards Cuba she received state support for her truly Cuban ballet. Her
company was recognised as a national institution, officially affiliated to the
Ministry of Culture; ballet schools across the country were opened. Alonso
and her collaborators distilled a Cuban methodology for the teaching of
ballet to ensure that the students in these schools learned to dance within
the consistent style that she was developing. Another goal of the policy of
enforcing a Cuban methodology was to safeguard the national elements
of the style by preventing members of her company from copying foreign
dancers. According to Alonso, “dancers from other Schools move in their
own way and it’s beautiful that they do so. But let them dance in a way that fits
their character and their culture. We should dance in our own differentiated
style. Only the most qualified, knowledgeable and mature ballet teachers,
those who truly know the Cuban School and are able to discuss and arrive at
the right conclusions on these matters should decide, which foreign elements
can be assimilated. Otherwise, we start to copy things that are not related
to us and our School becomes blurred”.18
The evolution of Cuban ballet was determined, in part, by the political
reality of Cuba. Alonso was aware of the impact that communism had on
the island and paid special attention to align her nationalist goals with the
philosophy of Castro’s Revolution. In 1978, she pointed out: “We are living
fabulous events in our homeland, where new elements, life styles and ways
of expression are constantly emerging. But we cannot lose our identity or
national roots. That’s why the enrichment of the [Cuban] School should
be done in an orderly and careful way.” Alonso’s objectives fitted within
communist cultural policies that encouraged a nationalist type of art. In
1993 Fidel Castro wrote a letter to mark the forty-fifth anniversary of the
National Ballet of Cuba. He addressed Alonso: “Your great achievement has
been to know how to take advantage of the potential that the Revolution
has offered the people to develop all the art forms and dance in particular,
an art that is part of the essence of Cubans. The Revolution and the Cuban
people are proud of the National Ballet of Cuba.”19
271 Giselle in a Cuban accent

Alonso and her company became cultural ambassadors for the Revolu-
tion, spreading the gospel of communist ideology through the arts during
tours abroad and at home. The pride that the government took in Cuban
ballet was not exempt from Eurocentrism. Dance might be “part of the
essence of Cubans”, as Castro stated, but ballet was a form that Cubans had
adopted from Europe and North America. Such pride contained implicit
satisfaction that the state-supported company triumphed within an art that
had belonged to Europe and, most recently, North America. Communist
Cuba’s achievements in the arts and in education compared well to the
developed world and formed an important part of the official rhetoric that
made Europe once more a point of reference.
The interaction between nationalist and Eurocentric interests also
shaped the repertory of the National Ballet of Cuba. Fiesta negra, Floras,
El solar, El rı́o y el bosque, Sóngoro cosongo are some of the numerous pieces
by Cuban choreographers that explore national legends, dances and tra-
ditions. During international tours though, the ensemble usually presents
European battle horses such as Giselle, Swan Lake, Coppélia and Don Quixote,
polished performances of the classics that situate the dancers on the same
level as the finest ballet companies of Europe and North America.20

In conclusion, the career of Alicia Alonso and the emergence of Cuban bal-
let have been shaped by a dialectic interaction between a strong interest in
nurturing Cuban-ness in classical dance on the one hand, and Eurocentric
perspectives inherent to the art form on the other. Although the creation of
a Cuban school of ballet demanded the reaffirmation of a national identity
on Alonso’s part, it did not presuppose that the ballerina and her com-
pany distanced themselves from the European balletic tradition. Quite the
contrary; Alonso conceived Cuban-ness in ballet as a continuation of a
European dance legacy. In the artist’s view, the Cuban tone in her dance and
her embodiment of a European aesthetics were equally valid. That interplay
between national and European components has become complex, with the
emphasis shifting from one area to the other, depending on the opinions of
the critics, the demands of the presenters, the preferences of the audiences,
the stereotypes and expectations in diverse political scenarios and the tactics
that Alonso displayed to secure recognition for herself and her troupe.
24 European ballet in the age of ideologies
m a r i o n ka n t

Between 1789 and 1989 Europe and then the world chased the rapidly
receding hope that human society could be remade. The struggle to renew
the social order reached its zenith in the years before 1945 in modernism,
socialism, Freudianism, expressionism, Bolshevism, Fascism, Nazism, the
New Deal, the Mexican Revolution, Peronism and Gandhian nationalism.
After 1945, the world froze into two blocks in which both the West and
the East feared and suppressed experimentation, hunted and imprisoned
radicals and stamped out the danger represented by the “youth movements”,
the last and least powerful of the great attempts to transform human nature.
The collapse of the Soviet Union and its satellite states ended the French
revolutionary age and its attendant “isms”. Since 1991 Western capitalism
has become the universal form of economic enterprise but with unexpected
consequences. The two most populous states of the world, India and China,
have turned out to be very good at capitalism and now threaten the economic
futures of the United States and the European Union. Globalisation has also
called forth the genie of resistance from the ancient Arab bottle – Wahabi
Islam has emerged to reject every aspect of the great European experiment,
especially the emancipation of women, in the name of an ancient theocratic
puritanism.
Ballet has been at the centre of these developments both as protagonist
and as victim, because – that is the central argument that emerges from the
preceding chapters – ballet like all art reflects the social, economic, polit-
ical, legal, institutional and intellectual environment in which its creators
work. Yet the history of ballet raises a new issue, a possibility, which I shall
consider at the end of the chapter: the effect of new technology on the the-
atre and on the public sphere. The year that the Soviet Union disappeared
was the year that Dr Tim Berners-Lee inaugurated the World Wide Web: a
revolution so profound that we all reel from its impact. In the decade and
a half since that “giant step for mankind” the instant transmission, storage,
reproduction and consumption of culture has been so fundamentally trans-
formed that nobody, not the great movie moguls, the newspaper barons, the
record companies, the publishers, the museums and universities, theatres
and opera houses, know whether they can survive. When millions of people
contribute to their own “Wiki” encyclopedia, what happens to the Britan-
[272] nica? When I-Pods owned by millions store more music than any library
273 European ballet in the age of ideologies

in the world before 1991, why go to concerts? Where will ballet be in this
new environment if theatrical culture crumbles under the explosion of new
forms of technology, or, as Walter Benjamin famously put it, what happens
to “art in an age of technical reproduction”?
Ballet was thus inevitably affected by the huge changes in culture and
society after 1900. They disrupted the continuous dialogue in which ballet
had found itself: the heated, aggressive, emotional, often insulting, debates
that involved the artists who produced it and the public that received it
shifted between 1890 and 1914. Ballet had now also to deal with the emer-
gence of a rival, a self-proclaimed “modern dance” evolving around a new
sense of self. Dance and ballet were forced to take sides.

The ideologies of the body and the mind


Ballet, once it had become an art performed by professionals, depended
on the opera house and the theatre. As court ballet it had been a theatrical
performance for several centuries but, with the growth of the public sphere
in the eighteenth century, it needed professional performers and audiences,
critics, costume and stage designers, composers, orchestras with conductors,
stage hands, cleaners, ushers, choreographers, directors, managers, finan-
cial and press officers and so on. It was now performed in special buildings
that were designed to bring together people who bought tickets and then
entered that architectural space. Future dancers went to special ballet schools
to follow a career: after an admissions process that tested talent and physical
suitability, young students, often from the age of six years old, would be
trained for years and years so that they could carry out the ballet technique
and embody the characters of the narratives. Writers invented stories, com-
posers wrote music, choreographers devised movements that would convey
the stories. Ballet as part of theatre was a concerted effort of a large staff of
people, all gathered to create the performance event evening after evening
(or matinee after matinee performance). The theatre was (and is) a hier-
archical institution that very much reflected the social order into which it
was born, out of which it grew and within which it remained. The hierarchy
shifted but was always characterised by allocating people to certain places,
making them work towards the performance. That produced a particular
division of labour, with certain dependencies and the need for equilibrium
among the competing departments: power structures, financial viabilities,
economic incentives, aesthetic ideas played their part to keep the theatre
going, to stabilise the balance or tip it towards the dangerous and existential
edge. Ballet cannot be taken out of the theatrical landscape and European
civilisation is unthinkable without its theatrical culture.
274 Marion Kant

But at the end of the nineteenth century something fundamental hap-


pened: a diffuse rebellion against bourgeois society bubbled out of the new
mass urbanised society. Irrational and anti-rational ideas – the cult of vio-
lence, crowd psychology, anti-semitism as a mass movement, nudism, vege-
tarianism – spread across the traditional cultural landscape. Drama, opera,
ballet – that entire theatrical culture – was questioned and its mere sur-
vival could no longer be taken for granted. The questions came from within
by those who knew how theatre worked and who suggested reform and
adaptation to the changing world. Such changes, in a sense, had always
been an important part of the constant evolution of theatre itself and were
reflected in new aesthetic movements and new technologies (for instance
photography and film in the early twentieth century). More importantly
though, attacks came from outside the theatrical sphere and its proponents
demanded radical transformations, even abolition of the entire institution
of “theatre” or “opera house” or “ballet”.
One of the first to articulate these concerns in the dance sphere was
Isadora Duncan, an American who came to Europe to find the meaning of
life. Her concept of dance was completely different from ballet. Dance had
to be made an expression of modern life. For her, ballet as an old-fashioned,
antiquated and outmoded art could not fulfil the demands of the new era,
it could not express modernity. “Modern” and “expression” became the
catchwords of the time and programmatic for the future. Duncan’s theory
of dance sought to interpret the German philosopher Friedrich Nietzsche’s
writings. Through him she hoped to enter into the mysteries of artistic
creation and human existence. After she had been introduced to his work
in 1903 in Germany, she proclaimed his book The Birth of Tragedy out of the
Spirit of Music of 1871 her “bible”. With it she found a guide that would enable
her to express her female humanness in movement. It would equip her to
go back to the original sources of artistic inspiration with which she could
restore the art of movement to its former glory and place. Dance had once
been magnificent, in antiquity, and it now needed Nietzsche’s conception
of tragedy to resuscitate it by incorporating the primal principles of the
Apollonian and Dionysian in humanity. Those powerful forces of creation
supplied Duncan with the means to translate idea, inspiration and feeling
into movement and make it visible. Dance without the backbone of an
idea, a philosophy, was dead, worthless. Yet thought could only stand at
the beginning of the creative process because motion was driven then by
emotion and thinking had to be fused into feeling. Her physical preparation
could not have been more different from that of a ballet dancer:

For hours I would stand quite still, my two hands folded between my
breasts, covering the solar plexus . . . I was seeking, and finally discovered the
275 European ballet in the age of ideologies

central spring of all movement, the creator of motor power, the unity from
which all diversities of movements are born. I . . . sought the source of the
spiritual expression to flow into the channels of the body filling it with
vibrating light . . . 1

She danced as though she were in trance, without technical drill. Ballet
was decadent because it followed mindless repetition and had particularly
disastrous effects on the female frame. Ballet training harmed limbs and
tore muscles, it bloodied the feet with the cruel and unnecessary pointe
shoes, the corset imprisoned the nature of femininity; it took away the nat-
ural beauty born out of music and the ebb and tide of water or the gentle
swaying of branches in the wind or sand blown over the shores, in short,
the flow of nature. Her dance, infused with Nietzschean philosophy, would
also resurrect the female body by liberating it from all previous social con-
straints. Out of her dance would materialise the beautifully contemporary
and essentially natural embodiment of womanhood. Her fervent feminism
took up the “woman question” of the Victorian age but gave it a different
perspective and outlook. Duncan did not need the theatre but she cleverly
used it if available. For her, any space was good enough for dance, which
meant that Duncan performed anywhere – in parks, tea rooms, salons, on
the stage of theatres or opera houses.
Isadora Duncan soon turned into the icon of modernity and exerted the
greatest influence on dance and on its perception in the twentieth century.
From choreographer Mikhail Fokine to dancers Anna Pavlova and Vaclav
Nijinsky – hardly any artist remained untouched by her passionate appeal
to renew the arts, particularly dance, and many changed their aesthetics and
performance styles, inspired by Duncan.
Following Isadora Duncan’s forceful impact, Rudolf von Laban became
the defining authority within the movement that labelled itself modern
and avant-garde. In looking for new principles and a new understanding
of dance, Laban, a Hungarian officer who moved to Germany after the
First World War, formulated, devised and organised a theory and dance
practice which concentrated on expression of the inner soul of the human
being. Hence this movement called itself and was recognised as Ausdruck-
stanz or expressive dance. Very much like Isadora Duncan, Laban searched
for the source that had made an Ur-Tanz, an original, untainted dance
before civilisation and history had left its nasty traces and contaminated it.
And like Duncan, Laban and his followers were convinced that they had to
look outside European high culture and far beyond ballet. Laban, as many
other modernists, mistrusted Western civilisation. Ballet clearly epitomised
Western culture and therefore needed to be eliminated together with its
entire history. He admired Duncan for her intuition and originality but
276 Marion Kant

particularly for what he deemed a non-rationalistic approach to dance, for


the courage to find the “soul” through dance. As he wrote:

At a time when science, and especially psychology, endeavoured to abolish


radically any notion of a “soul”, this dancer had the courage to demonstrate
successfully that there exists in the flow of man’s movement some ordering
principle which cannot be explained in the usual rationalistic manner.2

The essential principles of his modern dance can be found in a new theory
of space, followed by a new theory of the performer. His space redefined
performers and audiences; Laban made a dance space outside the theatrical
realm and restructured dance as performance. In fact he rejected the notion
of “performing” and replaced it with “experiencing”. Thus he reformulated
the performer’s attitude to originality and self-expression and addressed
the preparation and training for dance events so that they no longer had
to rely on talent or technique but on belief. Modern dancers followed their
convictions; anyone could dance who shared a deep faith in movement as
source of humanity. Laban also “wrote dance” by devising a notation as a
language of movement and, finally, he reorganised the dance networks.
Germany together with the United States provided the environments
for modernist experiments. In both countries the battle between ballet and
modern dance was decided in favour of modernism though the fight was
less violent and antagonistic in America. In both countries national iden-
tities were being altered not through ballet but modern dance. Only after
Balanchine’s arrival in America did attitudes towards ballet alter, gradually
but substantially.
One primary concern of the theatre and performing arts avant-garde
had been to change the nature of the performer–audience relationship by
examining the theatrical space. Reforms in spatial arrangements of the-
atre/performance space developed in parallel: Laban rethought dance at
the same time as Bertolt Brecht developed the principles of epic theatre,
Vsevolod Meyerhold advanced a new movement theory called biomechan-
ics and Walter Gropius and Erwin Piscator discussed their project of “total
theatre”. Theatre turned into a weapon to to realise social changes and bring
about the utopian future; art and artists were supposed to accept social
responsibilities and make their art an active social presence.
Laban, also troubled by the state of urban civilisation at that time,
attempted to create a new religious order that followed a cosmological idea
of wholeness and oneness of mind, soul and body. He founded dance com-
munities in which his dancers underwent conversion experiences before
connecting to the cosmic universe. Dance, a community affair, functioned
as a ritual, which created the community, held it together and brought har-
mony back into individual human lives and thus to the community. Dance
277 European ballet in the age of ideologies

was healing, soothing, a way of life and no longer merely a paid profession.
He intended to go beyond the mechanical, meticulous movement struc-
ture of ballet and instead make and celebrate the whole and unified person.
Laban’s movement system and his system of spatial and movement analysis
attempted to unify life with art in a way which had not been envisaged by
anyone in the dance world. Though he built on knowledge created by bal-
let masters such as Pierre Beauchamps and Raoul Feuillet so as to invent a
new system to write dance according to his new rules, he surpassed every
convention on which ballet rested. He had to deny ballet its past in order
to prevent its future and instead secure that of his modern dance. Coex-
istence was not an option. He could not accept that the balletic system
had developed through the centuries according to claims and necessities
of the society to which it belonged: from representation of worldly power
in the aristocratic courts, particularly of northern Italy, in the fifteenth
century to the bourgeois theatre dance of the romantic era, first in France
then throughout Europe, presenting social conflicts and telling stories of a
contemporary world to a paying audience. To acknowledge that the stage,
the theatrical space, had changed considerably over the centuries would
mean to admit the possibility of further change and adaptation of ballet to
the modern world.
The ballet of the nineteenth and twentieth centuries belonged to exactly
that society that many modernists intended to overthrow. Laban’s relation-
ship to the theatre therefore had to be contradictory: he felt drawn to it for
the possibility of creating another imaginary world; yet he disliked, even
loathed, the “lies” and the “falseness”, the pretentiousness and artificiality,
of the theatre. He imagined a performance space “true” to values, which
had been destroyed by modernisation, urbanisation and technology. Ide-
ally, dance would not be a performance in the common sense, there would
be no paying audience because no commodity would be bought. Instead,
he needed a space for devotion and transfiguration, which also severed itself
from social connotations; it had to be accountable to the community not the
society. Laban’s image of space broke totally with traditional dance and the-
atrical spaces; in fact it was a break with the inheritance of Western culture. It
disconnected dance from its history and replaced it with an imagined archaic
tradition. At the end of the twentieth century his complete, self-reliant sys-
tem once more became attractive to choreographers of post-modern dance
as well as ballet who also had lost confidence in civilisation and Western
culture.
The modern or avant-garde movements in dance represent an interesting
paradox: they planned an ideal world in opposition to the existing modern
world yet needed the modern world they detested to make their ideals. In
the years before the First World War ballet was blamed together with other
278 Marion Kant

cultural phenomena for a world falling apart: ballet had run its course,
very soon it would be replaced by one of the modernist movements. Most
modernists could not or would not apprehend that ballet had more survival
strength than anticipated and that it had not outlived its time but lived with
it – though in a different way from the modernist projects. Renewal within
ballet might have been painful for those involved in it, but it took place. The
Ballets Russes or the Ballets Suédois were examples of the successful reform
of the art form, not despite of the century-old ballet tradition or because
of its weakness but because of its vigour.3 But not even at the end of the
twentieth century was this view commonly shared in the dance world. It is
still a contentious issue.

The new ideologies


The modernist concepts of the body and the mind grew out of the discontent
with civilisation. They were an expression of the desire to solve the conflicts
brought about by a world in rapid transition. Economic crises, unemploy-
ment, women’s liberations, secularisation, urbanisation turned the world
upside down – where was this society going? What was going to happen
next? The loss of securities, the loss of the certainty that there was a future,
the collapse of value systems, the disappearance of social frameworks, the
fading of economic stability led to new political movements with new ide-
ologies. They counteracted fear and ambiguity with enticing projections of a
future world. For most of these radical movements the French Revolution of
1789 still provided a model for social change: that of a revolution combined
with the intention to generate a new man who would inhabit the revolu-
tionised world.4 Dance, just as the other arts, played an important role in
those projects. It became part of the activist ideologies, it was a propaganda
device, a medium to convey the message and forge an alliance between the
believers of the new movements and their prospective members. Above all,
it was the practice field to discover and test suitable embodiments of the
new ideal. On this level too, the struggle between ballet and modern dance
continued, as modern dancers argued that their movement concepts were
more suited than ballet to produce future man, that they were better able
to rule the masses out of which the new man was to emerge. Once more,
ballet’s history and tradition were considered proof of its elitism and back-
wardness. On the other hand, ballet’s history and tradition lent legitimacy to
the new movements. Hence confused messages surfaced. The odd mixture
of preserving and destroying traditional movement concepts become clear
in Italian futurism. It was one of the first organised assaults on traditional
aesthetic concepts in the twentieth century. Filippo Tommaso Marinetti
279 European ballet in the age of ideologies

established futurism by issuing a founding manifesto in 1909. There was


hardly an art form, artistic phenomenon or aesthetic idea that was not
examined, denounced and reinvented with a better, futuristic outlook, from
painting, to film, to theatre, to noise in general, to collage, to sensuality and
sexuality. In another passionate manifesto Marinetti celebrated war as the
grand hygienic cleansing of the decaying old world. Dance, of course, could
not be left out. In July 1917 Marinetti surprised his followers with a man-
ifesto of futurist dance in which he did not propound a modern dance
movement above ballet. Instead he managed to fuse all sorts of national or
folkloric dances with ballet:

Once the glorious Italian ballet had died and been buried, there began in
Europe a stylisation of savage dances, an eleganticisation of exotic dances
and a modernisation of ancient ones. Parisian red pepper + helmet + shield
+ lance + ecstasy before idols that no longer mean a thing + undulations of
the thighs of Montmartre = a passé anachronistic eroticism for
foreigners . . .
From an artistic point of view the ballet russe [sic] organised by Diaghilev
is extremely interesting. It modernises Russian popular dancing in a
marvellous fusion of music and dance, the one penetrated by the other, and
thus gives to the spectator an original and perfect expression of the essential
force of the race. With Nijinsky there appeared for the first time pure
geometry in dance liberated from mimicry and without sexual excitement.
We have here the apotheosis of musculature.5

Isadora Duncan as an over-emotional, childish woman did not appeal to


Marinetti, she was of little interest to manly futurism. He ended his mani-
festo with several ballet libretti which rejoiced in shrapnel, aeroplanes and
machine guns: to visualise Vaclav Nijinsky not as Faun or Spectre of the
Rose but as bullet shooting across stage. Futurism set its hopes on Benito
Mussolini, the leader of the Fascist movement, anticipating that the Duce
would march towards the “futuristic future”. In a similar way German Mod-
ern Dance, forged by Rudolf von Laban, aligned itself with racist concepts of
the body and collaborated with Nazism. Adolf Hitler would create the folk
community and exalt Laban’s religious dancing cult. Thus many dancers
could shift their movement theories, whether modern or ballet, towards the
right, but they could also move towards the left. After the Russian Bolshevik
Revolution in 1917 a surge in radically politicised ideas took hold of the
dance world. Isadora Duncan had a short-lived dalliance with Bolshevism
for which she was never forgiven in her native United States but remembered
as the “red hussy”. Her notion of dance as education made a deep impression
on Russians even though, in the long run, ballet became the representative
dance culture of the communist state. Why did ballet win over modern dance
in the Soviet Union, why preservation instead of continuous exploration?
280 Marion Kant

Because after the revolutionary mastermind Vladimir Ilyich Lenin’s death in


1924 the conservative, regressive and oppressive flank of the Bolshevik party
won. After ambiguous statements by state bureaucrats, ballet advanced to
state art, the art representing Soviet Russia. Here, its strong traditional past
and glamour, its imperial grandeur reinforced the link to an intact Russian
past that the rulers of the regime needed and that in turn would make itself
felt on the European scene. Ballet legitimised the regime that paid and sus-
tained it on a historical and political level. If ballet from the mid nineteenth
century on had failed in Western Europe, which dance scholars seemed to
agree on, then the Russian Imperial ballet after 1917 was one example of how
Russian artists could revitalise European ballet, and Europe on the whole.
Ballet might just be part of that proletarian world revolution and thus not
by its nature anti-revolutionary but worth incorporating into the revolu-
tion. It would now set out to tell the well-established fairy tales together
with the revolutionary stories of the people of Soviet Russia, the heroes of
the kolkhoz – the peasants, the heroes of the factories – the proletariat, the
heroes of the Red Army who protected the united workers and peasants.
The immediate nineteenth century past also bestowed its narrative and dra-
maturgical structure on many ballets of the Soviet era. Realistic storytelling
won over dream-like abstract, symphonic ballets. It took several decades
after the Second World War before any experiments in abstraction occurred
again. The ideological and aesthetic ideal was called “socialist realism”. It
grew out of political demands that the Stalinist state placed on artists and
relied on the belief that they, the artists, had an obligation to fulfil towards
the masses, the people. In 1934, at the First All-Union Congress of Soviet
Writers, “socialist realism” was declared standard for all arts: they had to
glorify the political and social achievements of communism.
Soviet bureaucracies left the institutional structure of opera and bal-
let intact. In fact, even tsarist obligations to artists were taken seriously
and payments of pensions to former ballerinas who had performed at
the court in St Petersburg were still made into the 1940s and later.
Ballet schooling continued in the traditional way, the hierarchical order
was upheld with great shining stars, slightly less great and twinkling stars
and the corps de ballet. Like their predecessors, the Taglionis, Elsslers, Gri-
sis or Grahns, the ballerinas of the Soviet age became national idols: Galina
Ulanova, Maya Plisetskaya, Marina Semenova, Natalya Dudinskaya, Olga
Lepeshinskaya . . . The technical standard and rigidity were also maintained;
Agrippina Vaganova who, after her dance career, began teaching at the
Leningrad State Choreographic School (formerly the Imperial Ballet School)
in 1921 modernised and at the same time canonised the training system and
methods.6 Both, school as well as training principles, bear her name today
(see Figs. 46 and 47).
281 European ballet in the age of ideologies

Figure 46 Galina Ulanova in Act ii of Giselle, Bolshoi Ballet Moscow, 1954.

The Russian Revolution spread ballet back to the West. In the years after
the Revolution of 1917 dozens and dozens of choreographers, ballerinas and
their entourages left for the West. In Berlin and Paris, the largest Russian
émigré communities in Europe came together and ballet schools and studios
teaching classical Russian style with Russian teachers multiplied though they
did not carry the proletarian dogmas with them. European ballet culture
experienced a tremendously vitalising force. In Germany, with a relatively
weak ballet tradition but strong theatrical network, Russian ballet seriously
threatened the modernist claim to power. In France and Great Britain it
282 Marion Kant

Figure 47 Maya Plisetskaya as Odette in Swan Lake, Bolshoi Ballet Moscow, 1966.

rejuvenated and infused with new life a respected cultural heritage. Two
companies, above all others, broadcast ballet across the globe: the Ballets
Russes de Monte Carlo in its twofold manifestations.7 Both ensembles were
born with the intention to exploit the fame of Diaghilev’s Ballets Russes. Both
lived off the eminence of Russian ballet and Russian ballerinas. The waves
of refugees fed the ranks of the two companies from the 1930s onwards;
adoration and notoriety accompanied them but also helped business and
reinforced name recognition. Scandal, though, was founded on real original-
ity and imagination: Léonide Massine explored abstract formal structures
in his ballet symphonies and George Balanchine began his experiments with
ballet’s form and structure. Russian ballet as “classical” ballet entered the
minds of the audiences from London and Paris to New York and Buenos
Aires as a fusion of the old – French and the new – Russian.
European and American ballet, invigorated, carried on, more or less
pragmatically in the municipal or court theatres, in the smaller opera houses
and in the many provincial companies, neither making itself part of the
283 European ballet in the age of ideologies

political right nor the left. It found a powerful defence against any ideological
claim and the attack of any political movement: l’art pour l’art, art for art’s
sake. This slogan constituted the most successful creed or ideology claimed
by artists to protect themselves as well as their art from the onslaught of
political programmes and demands. It was (and is) ideology masquerading
as anti-ideology. Théophile Gautier, the man who had invented Giselle and
had written many other ballet libretti, had picked up the motto and used it
to rescue art (and ballet) from the claws of utilitarian politicians in 1834.
L’art pour l’art declared then, as it does now, that art is only an aesthetic
means unto itself, that it serves no ulterior purpose. “Only those things that
are altogether useless can be truly beautiful; anything that is useful is ugly,
for it is the expression of some need, and the needs of man are base and
disgusting, as his nature is weak and poor.”8 Hence the most useful thing
for Gautier, and the least artful, was the toilet bowl. (Marcel Duchamp in
his Dada phase answered in kind by exhibiting a urinal – Fountain – in
1917 as ultimate art object.) With the idea of l’art pour l’art, artists could
declare their independence and resist attempts at political take-over. It also
implied that the artist was somehow removed from society, alienated to such
a degree that art need not enter into a direct and immediate relationship
or dialogue with social forces. In the age of ideologies, art for art’s sake
proved itself as the most powerful shield for ballet; any philosophies that
did not support the aesthetic integrity of the work of art were – and are –
considered unacceptable. The separation of technique from narrative had
been undertaken in the nineteenth century and conveniently continued. A
plié, an attitude, an arabesque, a développé, a battement or any other element
of the ballet method was beautiful in itself; it meant nothing and hence could
not be appropriated by ideologies. If there existed any ideological trace it
was embedded in a story or narrative. Those, though, were exchangeable
and negligible. The contradiction between abstract movement principles
and concrete narrative could not be revoked, but it could be neutralised.
Ballet had been classified as mindless physical activity by the advocates
of modern dance; ballet dancers – by repeating the same steps in every
story and by using a universalised technique – paradoxically confirmed the
case. Ballet’s presumed intellectual emptiness was tied to l’art pour l’art’s
intention of keeping the movement component as pure as possible and
making it valuable in itself; the lines and shapes that the body achieves
within a balletic movement sequence constituted then – and now – ultimate
beauty. The arch of the foot preparing to rise on pointe had to be more
important than the arch of the foot rising on pointe as a means to convey
a story, be it in Sleeping Beauty, Romeo and Juliet or The Red Poppy. Thus
it has always been possible to disconnect the technical aspect of ballet from
the content of the story, which the technique is supposed to help tell. An
284 Marion Kant

example of the peculiar dynamics of ballet, l’art pour l’art, self-supporting


technique and the endeavour to instil ballet with a new spirituality can
be found in Oskar Schlemmer’s Triadic Ballet of 1922. When Schlemmer
conceived the idea, he was working at the Bauhaus, the college that tried to
unite and unify the arts, to explore new ways of manufacturing and crafting
in the age of mechanisation. With Schlemmer (as with Laban’s student Kurt
Joos who in 1932 choreographed The Green Table and founded the Ballets
Jooss), a careful cross-fertilisation between ballet and modern dance began.
Schlemmer studied dance history and deliberately called his stage work
“ballet” though the ballet world either ignored or laughed at his strange
experiments. He imagined a ballet that found a new meaningful subject
through technique as abstraction and condensation of human movement.
Ballet was taken to the extreme limit of coherence by replacing the human
body with costumes representing elementary geometrical shapes, which
led to intensely focused applications of movement principles: circles, lines,
ellipses, diagonals, colours all emanated from primary human gestures.
Non-ideology, l’art pour l’art, made a new and acceptable belief-system. The
reduction of movement to essential factors and the constant repetition of the
Holy Trinity as symbol of a new spirituality, Schlemmer predicted, would
lead to a fresh awareness of religiosity. But Europe was moving towards the
next disastrous war. Dancers made good soldiers, for any army, German
or Russian, Italian or French or English. Dancers entertained soldiers of
every army, whether the Wehrmacht, the Red Army or the Allied Forces.
Dancers joined the resistance in France or danced for the Nazi occupiers,
they went underground in Italy, they fought in the eastern forests with the
partisans or on the eastern front with the SS, they entered ammunition
factories in Germany or supported the war effort in Britain. They were
thrown into prison and taken to concentration camps if they were Jews,
or communists, or socialists or gay. And they danced in the camps and in
the prisons, secretly using dance to survive and subvert the systems. They
entered the ever growing army of refugees, which took them to every part of
the world. Wherever they went, they carried with them their dance, modern
or traditional, opened up schools, taught and performed.

The ideology and politics of the state


In 1945, at the end of the Second World War Europe was destroyed and
though its theatres and opera houses were largely bombed, burnt and razed,
its performance culture was among the first utterances to come alive again.
If dance had provided hope during the war then dance was the art form that
sparked new energy and belief in the future in peace time. Germany, now a
285 European ballet in the age of ideologies

smoking ruin, had to endure the dance culture of the allied forces – ballet.
Modern dance, tainted by its association with Nazism, had to retreat for a
time. Ballet now was imported from the victorious United States and Soviet
Union. The United States brought their own very young ballet and theatrical
dance culture, to which they had been introduced through European soloists
and their accompanying groups in the nineteenth century and through
artists who emigrated in the twentieth century.
With the beginning of the Cold War, that battle between the capitalist
West, led by the Americans and the communist East, led by the Soviets,
Europe entered the next period. The American Marshall Plan, designed
to restore European industry and provide dollars for imports of new US
machinery and technology, in the long run also paid for and fed the-
atres. Growing economic wealth was measured not only in higher pro-
duction or increasing wages but in the number and quality of orchestras,
operas and ballets. Liberalism could be distinguished by its manifold artistic
approaches and styles. Ballet companies were proud to engage foreigners
because ballet was international, a shield against the destructive nationalist
forces: Stuttgart from 1961 had John Cranko from South Africa, Hamburg
since 1973 has John Neumeier, born in Milwaukee, Frankfurt since 1984
has William Forsythe, born in New York City in 1949, who had danced
under Cranko in Stuttgart. Glen Tetley, born in Cleveland was invited to
the Nederlands Dans Theatre in 1962; Jiřı́ Kylián, born in Czechoslovakia
in 1947, trained at London’s Royal Ballet School and encouraged by John
Cranko to choreograph, became artistic director of the Nederlands Dans
Theatre in 1975. In Western Europe, internationalism and international
exchange, openness to experiments and personal styles, were advanced,
and in the face of socialist realism, abstraction and aesthetic independence,
l’art pour l’art was ever more promoted. But internationalisation meant
Americanisation.
In the East, internationalism meant Sovietisation. The Russian school,
above all the Vaganova method, prevailed and Russian dancers came to teach
in the new state ballet schools in Berlin, Prague, Warsaw, Sofia, Budapest.
The Soviet system superimposed socialist realism, “the truthful, historically
concrete representation of reality in its revolutionary development . . . linked
[to] the task of ideological transformation and education of workers in
the spirit of socialism”, onto all the states in the Eastern bloc with ballets
like The Flames of Paris, The Fountain of Bakhchisarai, Spartacus, Romeo
and Juliet.
In the West as well as the East, ballet offered itself as a cultural export arti-
cle. Imperialism (here understood in the context of the Western and Eastern
empires) once more spread ballet and made it a global asset with an ideo-
logical dimension, just as in the nineteenth century. The West believed in
286 Marion Kant

liberalism and democracy, the East in socialism and dictatorship of the pro-
letariat, and both sides promoted the ballets that best represented these ide-
ologies. After decades of isolation Russian ballet companies made startling
appearance, on the stages of the Western world in the early 1950s, with a
huge impact, from Paris to New York. Long queues formed and the audi-
ences wept with Julia – Ulanova and rejoiced with Kitry – Plisetskaya. The
Russians and the West utterly misunderstood each other in this process of
exchange. For the East their triumphant dancers promoted socialism with
a human face. These tours took place at the height of the Cold War, a
war fought with every means except nuclear weapons. The Soviet leaders
assumed that the triumphs of Russian ballet would establish a dialogue but
at the same time convince Western audiences of the superiority of their sys-
tem. An unfortunate side effect of the international tours lay in the defection
of dancers like Natalia Makarova, Rudolf Nureyev or Mikhail Baryshnikov.
To the East, these were traitors who abandoned their ideals, to the West
they were heroes who had jumped into freedom. Their physical rejection
of socialism was always turned into a political coup: when ballet dancers
left, they more clearly than anyone else seemed to prove the physical disin-
tegration of the Soviet system; to provide proof also of the correctness of
Western propaganda that the East was intolerable. The east counter-argued
that the West corrupted the minds and the souls of the weak and therefore
borders needed to be fortified. For the West, l’art pour l’art, the dominant
and apolitical ideology of post-war modernism, ensured that the Russian
political message could not remotely attract Western audiences. The great
Russian ballerinas were symbols of pure art not of the workers’ and peasants’
state.
Though the focus on ballet was very much determined by political fac-
tors, and interest concentrated on those dancers, choreographers and com-
panies who would best satisfy political ambition, ballet flourished in the West
as representation of l’art pour l’art. In France and Great Britain the steady
and enduring ballet cultures were never endangered species. Both countries
have always produced choreographers who challenged and broadened the
system from within. The visionary Maurice Béjart came to attention with
his philosophical ballet Symphonie pour un homme seul in 1955. No aspect
of life has since been too trivial for him to investigate, no culture not inter-
esting enough to compare or work into Western ballet, no political event
not worth using as reference (see Fig. 48). Ever since his first success he fused
ideas and styles; used Nô Theatre or Indian drama, existential philosophy
or Zen Buddhism to enhance ballet. In 1960 he founded the Ballet de XXe
siècle with a school attached. Béjart’s biography reminds us of an impor-
tant aspect of ballet: its interconnectedness throughout the world. Béjart
had danced with Roland Petit (who cultivated dramatic, realistic ballet and
287 European ballet in the age of ideologies

Figure 48 Maina Gielgud in Forme et ligne (Squeaky Door) by Maurice Béjart. Australian Ballet,
1974.

explored twentieth-century literary sources) and Birgit Cullberg (the force


behind Swedish modern dance and ballet and author of a specific danced
drama) before making himself independent. The centres of schooling and
performance in Paris, London, Milan, Moscow and so on functioned as
central network points from which talents emerged and spread – in the
nineteenth just as much as in the twentieth centuries.
In Great Britain, the confrontation between ballet and modern dance
can also be traced though it seemed to have taken less acrimonious forms.
The oldest modern dance company was founded by Marie Rambert in 1926,
whereas Dame Ninette de Valois had trained and danced with Enrico Cec-
chetti and Serge Diaghilev and from 1928 built Sadlers Well’s Ballet, later the
Royal Ballet. With these two companies the foundations were laid for artists
like Frederic Ashton, Anthony Tudor, Kenneth MacMillan and ballerinas
288 Marion Kant

Figure 49 Felia Doubrovska as Bride in Les Noces, choreography by Bronislava Nijinska.

like Alicia Markova or Margot Fonteyn to employ their skills and nurture
the narrative tradition and dramaturgical structure of nineteenth century
ballet. Famed for its wit and amusing quality, British ballet for a long time
struck a balance between the old and the new (though it remains to be seen
what happens next at The Royal Ballet).9
Ballet in Europe always had to cope with its past and learn to modernise.
It managed to preserve and at the same time depart from its own tradi-
tion, the shadow of the past looming over it, as burden or inspiration (see
Fig. 49). Its survival rested on maintaining the balance between invigorat-
ing explorations and reverence for the well-worn repertoire. That is just as
true today: ballet has to transform itself constantly, otherwise it will lose its
vitality and turn into a stale, museum-like artefact, smelling of moth balls.
289 European ballet in the age of ideologies

Figure 50 Three Atmospheric Studies by William Forsythe, the Forsythe Company 2005.

This is an ongoing challenge. Contemporary interpretations of nineteenth


century favourites often split audiences and professionals alike: Mats Ek’s
vision of Giselle, Mark Morris’s Hard Nut or Maurice Béjart’s autobiograph-
ical Nutcracker do what art is supposed to do, question the canon and push
the boundaries. They keep the art form alive as much as those who go far
beyond the known repertoire: choreographers like William Forsythe, who
deconstruct human movement through the categories and methods that
Laban had once invented, to infuse his ideal community with core values.
Forsythe seeks that same kinetic energy and essence that Laban studied and
contemplated. Movement as defining category of humanity therefore pro-
vides the structures with which to experiment and which to manipulate (see
Fig. 50).
Ballet’s further survival does not rest on the debates around how much
invigoration it needs or receives from adventurous dance artists nor on its
relationship to modern dance, a movement genre in its own right. Neither is
it threatened by internationalisation; just as great violinists or cellists come
from China, Japan or Korea, so the traditional ballets schools from Buenos
Aires to Shanghai continue its structures, institutions and cultural legacies.
It depends on the effect that globalised capitalism will have on the place and
the fabric of the art form.
Ballet is not a universal art form, but a Western cultural articulation. It
spread with European trade, commerce and cultural exchange. It depends
still on the public sphere and human interaction but both of which may
now be disintegrating in the twenty-first century. Ballet has up to the
present day retained certain characteristics that guarantee constancy: the
stage, the schools, the company structures and hierarchies, the international
290 Marion Kant

competitions. Ballet technique – its technology – has not yet substantially


changed in this globalised era. It is still a training method supposed to
shape dancers’ bodies for the performance. Training at the barre, with a
ballet mistress or master remains a purely technical preparation for dance
events that follow the most diverse performance principles and incorporate
movement vocabulary that is no longer limited to ballet technique. The
school, the training, the traditions emerged and survived because the insti-
tution “theatre” survived. The great ballet companies depend on the state
and its subsidies everywhere except America, where they rest on private
philanthropy. Whatever the source of funding, the ballet system requires its
institutional casing and framework. In the age of globalised communication,
the cell phone, the I-Pod, the Blackberry, all of which isolate and individ-
ualise human interaction with culture understood in its widest sense, can
a collective enterprise, born in the first period of the commodification of
art in the public sphere of the early nineteenth century, survive when other
institutions of that period – the daily newspaper, for example – seem to be
in terminal decline? Is there a public sphere out there any more? When great
media moguls worry about the fragmentation of the public into ever smaller
coagulations of special interests, when hundreds of specialised TV channels
confuse the viewer, when the traditional weekly and monthly journals born
in the eighteenth-century market for ideas and coffee houses for a single
public give way to blogs and pods and other electronic avenues to private
consumption of culture, what will happen to ballet?
With the loss of the common public sphere to which ballet was and is
linked, the transition is open to an art form that is no longer ballet. This
transition could be the end of the art form in that it makes superfluous
and eventually abolishes the institution. If ballet technique, taken out of its
context, is nothing more than an article that is used by anyone who can buy
it for whatever reasons, then the art form begins to give way to the relentless
onslaught of consumerism. Will it now, finally, succumb to the ultimate
commodification of culture and the arts, unimagined and unimaginable
to the modernists of the previous centuries? And will ballet then leave the
public sphere – the theatre, which is consistently being undermined and
destroyed – and appear occasionally as reflection of its former self and
of what it can never be again, once it has been turned into a corporate
ornament?
Notes

1 The early dance manuals and the structure 9 See chapter 4 in Jennifer Nevile, The Eloquent
of ballet: a basis for Italian, French and Body: Dance and Humanist Culture in
English ballet Fifteenth-Century Italy (Bloomington: Indiana
1 Guglielmo Ebreo da Pesaro, Guilielmi Hebraei University Press, 2004), for a detailed discussion
pisauriensis de pratica seu arte tripudii vulgare of the four misure and how the dance masters
opusculum incipit, 1463, Paris, Bibliothèque married the practical concerns of the dancer to
Nationale, MS fonds it. 973, fo. 19r. (The the intellectual ideas about the nature of the
translation is the author’s.) cosmos.
2 For a detailed discussion of improvisation and 10 None of the terms is defined precisely by
ornamentation in the balli and bassadanze, see Domenico, Guglielmo or Cornazano, but it is
Jennifer Nevile, “Disorder in Order: clear all refer to subtle movements of the dancer’s
Improvisation in Italian Choreographed Dances body performed as part of the execution of each
of the Fifteenth and Sixteenth Centuries”, in step. The meaning of campeggiare is not entirely
Timothy J. McGee (ed.), Improvisation in the clear, although the word seems to refer to a
Arts of the Middle Ages and Renaissance horizontal shading movement of the body above
(Kalamazoo, Mich.: Medieval Institute the foot which makes the step. This horizontal
Publications, 2003), pp. 145–69. movement is contrasted with ondeggiare, a
3 See Anne Daye, “Skill and Invention in the vertical movement like the waves of the sea, with
Renaissance Ballroom”, Historical Dance 2/6 a slow rising of the dancer’s body followed by a
(1988–91), pp. 12–15, for further details on quicker lowering of the body. (For further
these virtuosic steps. discussion of campeggiare see Mark Franko, The
4 For a detailed discussion of improvisation and Dancing Body in Renaissance Choreography (c.
ornamentation in sixteenth-century Italian 1416–1589) (Birmingham, Ala.: Summa
dances, see G. Yvonne Kendall, “Ornamentation Publications, 1986), pp. 59–61.) In his treatise
and Improvisation in Sixteenth-Century Dance”, Domenico likens maniera to the movement of a
in McGee, Improvisation in the Arts of the Middle gondola that in its passage across the sea rises
Ages and Renaissance, pp. 170–90. slowly and falls quickly (Domenico da Piacenza,
5 Timothy J. McGee, “Dancing Masters and the De arte saltandj & choreas ducendj De la arte di
Medici Court in the Fifteenth Century”, Studi ballare et danzare, Paris, Bibliothèque Nationale,
musicali 17/2 (1988), p. 205. MS fonds it. 972, fo. 1v). Cornazano also defines
6 Francis Ames-Lewis, The Intellectual Life of the maniera as a rising and falling movement of the
Early Renaissance Artist (New Haven: Yale body, encompassing both ondeggiare and
University Press, 2000), p. 74. campeggiare (Antonio Cornazano, Libro dell’arte
7 Keith Polk, German Instrumental Music of the del danzare, Rome, Biblioteca Apostolica
Late Middle Ages: Players, Patrons and Vaticana, Codex Capponiano, 203, fo. 3v).
Performance Practice (Cambridge: Cambridge Guglielmo describes aiere as “an act of airy
University Press, 1992), p. 11. presence and elevated movement, with one’s
8 The anonymous fifteenth-century French own person showing with agility a sweet and
basse danse treatises were far closer to a gentle rising movement in the dance” (fo. 10r).
collection of choreographies, with very little 11 The term fantasmata referred to the way a
additional information being included. Dance step should be phrased. Domenico (fo. 2r)
records from England in this period are scarce. describes how at the end of every step the dancer
For information on English sources see David should freeze briefly for a fraction of a second
Fallows, “The Gresley Dance Collection, c. 1500”, before commencing the next step, and this
RMA Research Chronicle 29 (1996), pp. 1–20; should all be done with so little effort that the
and Jennifer Nevile, “Dance in Early Tudor dancer appears to be like a falcon taking wing.
England: An Italian Connection?”, Early Music For further discussion of fantasmata see Mark
26/2 (1998), pp. 230–44. The sixteenth-century Franko, “The Notion of ‘Fantasmata’ in
French and Italian dance manuals follow the Fifteenth-Century Italian Dance Treatises”,
fifteenth-century Italian model. Dance Research Annual 16 (1987), pp. 68–86.

[291]
292 Notes to pages 14–21

12 Nicholas Clulee, John Dee’s Natural 24 Thomas Elyot, The Boke Named the
Philosophy Between Science and Religion Governour (London, 1531), repr. ed. S. E.
(London: Routledge, 1988), p. 77. Lehmberg (London: Dent, 1962), pp. 79–80.
13 Marsilio Ficino, Three Books on Life, critical
edn and trans. with intro. and notes by Carol V.
Kaske and John R. Clark (Binghamton, N.Y.: 2 Ballet de cour
Medieval and Renaissance Texts and Studies, 1 Mark Franko, Dance as Text: Ideology of the
1989), pp. 331–3 and 363. Baroque Body (Cambridge: Cambridge
14 See Françoise Carter, “Celestial Dance: A University Press, 1993), p. 1.
Search for Perfection”, Dance Research 5/2 2 Paul Lacroix, Ballets et mascarades de cour sous
(1987), pp. 3–17, for further discussion on Henri IV et Louis XIII (de 1581 à 1652) (Geneva:
divine dance from Plato to the seventeenth J. Gay et fils, 1868–70; repr. Geneva: Slatkine,
century. 1968).
15 The translation is Margaret M. McGowan’s 3 In addition to Lacroix cited above, for a
from her book, Ideal Forms in the Age of Ronsard, review of ballets and masquerades with
(Berkeley: University of California Press, 1985), indications of the sources which document them
p. 224, from Ronsard’s sonnet “Le soir see appendix in Margaret M. McGowan, L’Art du
qu’Amour vous fist en las salle descendre”. ballet de cour en France 1581–1643 (Paris: Centre
16 See Thomas M. Greene, “Labyrinth Dances National de la Recherche Scientifique, 1965,
in the French and English Renaissance”, repr. 1978), pp. 251–309; for the later phase see
Renaissance Quarterly 54/4.2 (2001), Philippe Hourcade, Mascarades et ballets du
pp. 1403–66. Similarly, the English court Grand Siècle, 1643–1715 (Paris: Desjonquières et
masques, on one level, can be viewed as “one Centre National de la Danse 2002), pp. 253–336.
vast moving talisman with emblematic figures in 4 The success of some ballets led to a series of
diverse colours moving amongst incantatory repeats as was the case with the Ballet de la
scenes designed to draw down influences on the douairière de Billebahaut, staged first in 1626 at
British court” (Vaughan Hart, Art and Magic in the Louvre and a few days later by popular
the Courts of the Stuarts (London: Routledge, demand at the Hôtel de Ville.
1994), p. 187). 5 See, for example, the Ballet de la prosperité des
17 Stockholm, Kungliga Biblioteket, Cod. Holm armes de la France (1641), ordered by Richelieu
S 253. to celebrate recent French military successes for
18 For further information on this manuscript which he considered himself to be responsible.
and on other surviving notated dance figures, 6 McGowan established a periodisation for the
see Jennifer Nevile, “Dance Patterns of the Early ballet de cour that focuses on the themes
Seventeenth Century: The Stockholm presented: allegorical and political from 1581 to
Manuscript and Le Ballet de Monseigneur de 1610, melodramatic from 1610 to 1620 and
Vendosme”, Dance Research 18/2 (2000), burlesque until 1636. Cf. McGowan, L’Art du
pp. 186–203; and Mark Franko, “Writing ballet de cour.
Dancing, 1573”, in Ann Dils and Ann Cooper 7 Susan Foster, Reading Dancing: Bodies and
Albright (eds.), Moving History / Dancing Subjects in Contemporary American Dance
Cultures. A Dance History Reader (Middleton, (Berkeley: University of California Press), p. 101.
Conn.: Wesleyan University Press, 2001), 8 François De Lauze, Apologie de la danse et
pp. 191–201. parfaite méthode de l’enseigner tant aux Cavaliers
19 Charles Nicholl, The Chemical Theatre qu’aux dames (n.p., 1623), pp. 35–6. McGowan
(London: Routledge & Kegan Paul, 1980), p. 97. (L’Art du ballet de cour, p. 33) concludes on
20 Lyndy Abraham, Marvel and Alchemy that basis that “the ballet had already evolved
(Aldershot: Scolar Press, 1990), p. 26. into a genre in which all amateurs would be
21 Very similar sentiments were expressed at the excluded”.
end of the sixteenth century by Caroso, who 9 Michel de Pure, Idée des spectacles anciens et
repeats this belief that movements of the body nouveaux (Paris 1668; facs. edn Geneva:
represent inward emotional states(Fabritio Minkoff, 1972), p. 215.
Caroso, Nobiltà di dame (Venice, 1600; facs. edn, 10 Jerôme de la Gorce, “Un aspetto del mestiere
Bologna: Forni, 1980), p. 1). teatrale di Torelli: la riutilizzazione delle
22 Guglielmo, De pratica seu arte, fos. 19r–19v scenografie dell’ ‘Andromède’ per il ‘Ballet de la
(the translation is the author’s). nuit’ ”, in Francesco Milesi (ed.), Giacomo
23 Prudence represented the knowledge of Torelli. L’invenzione scenica nell’Europa barocca
things men and women ought to desire and of (Fano: Fondazione Cassa di Risparmio di Fano,
things they ought to avoid. 2000), pp. 235–41.
293 Notes to pages 21–29

11 Marie Françoise Christout, Le Ballet de cour 19 Marie-Thérèse Bouquet-Boyer (ed.), Les


au XVII˚ siècle (Geneva: Minkoff, 1987), p. 76. Noces de Pélée et de Thétis. Venise, 1639–Paris,
This volume presents a rich iconografic 1654, Actes du colloque international de
documentation. By the same author, Le Ballet de Chambéry et de Turin, 3–7 novembre 1999
cour de Louis XIV. 1643–1672 (Paris: A. et J. (Bern, New York, Oxford and Vienna: Peter
Picard, 1967; new edition Paris: Centre national Lang, 2001).
de la danse and Picard, 2005). 20 Franko, Dance as Text, p. 21.
12 Christout, Le Ballet de cour au XVII◦ siécle, 21 Ibid.
p. 148. 22 De Pure, Idée des spectacles, p. 249.
13 Henry Prunières in his Le Ballet de cour en 23 Menestrier, Des ballets anciens et modernes,
France avant Benserade et Lully (Paris: Henri p. 154.
Laurens, 1914; New York and London: Johnson 24 See Bibliography for more information.
Reprint, 1970) argue for the Italian origins, with 25 For analysis of the Italian antecedents, see
James R. Anthony, French Baroque Music from Sparti, “Dance and Historiography”. It is worth
Beaujoyeulx to Rameau (New York and London: remembering that in a sort of obsession with
Norton, 1974; rev. edn 1981); and Barbara research on origins, Italian historiography has
Sparti, “Dance and Historiography. Le Balet for a long time identified Begonzio Botta as the
Comique de la Royne: an Italian Perspective”, in inventor of the ballet on the basis of an equivocal
Ann Buckley and Cynthis Cyrus (eds.), story due to a distorted reading of a
Festschrift for Ingrid Brainard, forthcoming (I fifteenth-century chronicle which refers to the
thank Barbara Sparti for putting at my disposal wedding of Galeazzo Sforza and Isabella of
two diverse versions of the unedited manuscript Aragon in 1489. In reality, Botta, who was a
of her essay). wealthy landowner, financed and occupied
14 The masquerade consists of a parade or a himself with the organisation of that event,
stage action, for the most part improvised, and which took place in his residence in Tortona.
executed by characters who are dressed up or There is no evidence that he was author of the
wearing masks. The dramatic content was stage actions of a mythological kind, which
limited and rather basic, resting on allegory and included dance or mimed parts specially
myth. Some parts could have been recited and prepared for the occasion. See Eugenia Casini
others accompanied by instruments or voices. Ropa, “Il banchetto di Bergonzio Botta per le
The main purpose was to create a spectacular nozze di Isabella d’Aragona e Gian Galeazzo
impression by using lavish costumes and Sforza nel 1489: quando la storiografia si
accessories. Intermedii that were placed between sostituisce alla storia”, in Myriam Chiabò and
the courses of a banquet or the acts of a comedy Federico Doglio (eds.), Spettacoli conviviali
were normally composed of recited texts and dall’antichità alle corti italiane del ’400 (Viterbo:
instrumental or vocal music and various types of Tipolitografia Agnescotti, 1983), pp. 291–306.
bodily moves such as dance, pantomime, The origin of the historiographical error is
acrobatics or fighting scenes. The theoreticians probably to be found in Ménestrier, Des
of the sixteenth century saw in its theatrical représentations, p. 157.
form the functions of the ancient Greek chorus: 26 He has passed into history as the first
temporal and technical transitions between acts, choreographer of the first ballet. It could be the
a pause for the actors, time to change the case that he never composed a single dance.
scenery, time for the spectators to relax from the Balthasar de Beaujoyeulx, born Baltazarini o
dramatic tension of the main work. Baldassarre (born before 1535 and died about
15 McGowan, L’Art du ballet de cour, p. 7. 1587) was of Italian origin but not precisely
16 De Pure, Idée des spectacles; Claude-François identified. He went to France in the entourage of
Ménestrier, Des représentations en musique the Marechal de Brissac, governor of Piedmont,
anciennes et modernes (Paris: René Guignard, at the request of Catherine de’ Medici. He was an
1681; repr. Geneva: Minkoff, 1992 ); excellent violinist and integrated himself rapidly
Claude-François Ménestrier, Des ballets anciens at court, frenchifying his name. He rose rapidly
et modernes selon les règles du théâtre (Paris: from simple valet de chambre to become an
René Guignard, 1682; repr. Geneva: Minkoff, officier. Some scholars think that the real author
1972); M. de Saint-Hubert, La Manière de of the Balet comique de la Royne was the court
composer et de faire réussir les ballets (Paris: poet Agrippa d’Aubigny. Carol and Lander
Targa, 1641; repr. Geneva: Minkoff, 1993). MacClintock (eds.), Le Balet comique de la Royne
17 De Pure, Idée des spectacles, p. 214. 1581 (New York: American Institute of
18 From the libretto for the ballet cited in Musicology, Musicological Studies and
McGowan, L’Art du ballet de cour, p. 152. Documents 25, 1971), p. 12.
294 Notes to pages 29–37

27 Cf. Diane L. Woodruff, “The ’Balet It remains speculation whether (a) these dance
Comique’ in the Petit Bourbon: A practical books were already available in the early Stuart
View”, Proceedings of the Society of Dance History period, and (b) whether readers really used them
Scholars (Riverside: University of California, as dance manuals rather than bibliophile
1986), p. 123. objects. Late seventeenth-century book lists in
28 Translation in Selma Jeanne Cohen, the Royal Collection are, for example, MSS
“Balthasar de Beaujoyeulx, Ballet Comique de la Royal App. 73 and 86. Barbara Ravelhofer, The
Reine Paris, 1582”, in Dance as a Theatre Art: Early Stuart Masque: Dance, Costume, and Music
Source Readings in Dance History from 1581 to (Oxford: Oxford University Press, 2006), ch. 2.
the Present (New York: Harper & Row, 1974), 11 Roy Strong, Henry Prince of Wales: The Lost
p. 19. Renaissance (London: Thames & Hudson, 1986),
29 See Sparti, “Dance and Historiography”. pp. 92, 95.
30 Franko, Dance as Text, p. 2. 12 They have attracted a massive amount of
31 Ibid., p. 1. criticism, for instance: Stephen Orgel’s The
Jonsonian Masque (1967, repr. New York:
3 English masques Columbia University Press, 1981, with new
1 Inga-Stina Ewbank, “‘These Pretty Devices’: A introduction) and The Illusion of Power
Study of Masques in Plays”, in A Book of Masques, (Berkeley: University of California Press, 1975);
ed. T. J. B. Spencer and S. Wells (Cambridge: D. Bevington and P. Holbrook (eds.), The Politics
Cambridge University Press, 1967), pp. 405–48. of the Stuart Court Masque (Cambridge:
2 William Shakespeare, The Tempest, ed. David Cambridge University Press, 1998); Kevin
Lindley (Cambridge: Cambridge University Sharpe, Criticism and Compliment: The Politics
Press, 2002), iv. i. 138 and stage direction. of Literature in the England of Charles I
3 Henry Purcell’s Operas: The Complete Texts, ed. (Cambridge: Cambridge University Press, 1987);
Michael Burden (Oxford: Oxford University numerous articles by Martin Butler.
Press, 2000), pp. 6, 9–10. 13 Thomas Middleton, A Courtly Masque: The
4 Edward Hall, The Union of the Two Noble and Device Called The World Tost at Tennis (London:
Illustre Famelies of Lancaster [and] Yorke G. Purslowe, 1620), sig. Br .
(London: R. Grafton, 1550), third year of 14 For instance, in Oberon.
Henry’s reign, sig. ciiiir . See also Enid Welsford, 15 Huntington Library, MS HA 10543. Letter to
The Court Masque: A Study in the Relationship the Earl of Huntingdon, date estimated 1627 by
between Poetry and the Revels (Cambridge: cataloguer and 1633 by John Yoklavich in “The
Cambridge University Press, 1927), pp. 130–5. Seven-Thousand-Pound Pastoral”, Huntington
5 John Stevens, Music and Poetry in the Early Library Quarterly 28 (1964), pp. 83–7. I am
Tudor Court (Cambridge: Cambridge University indebted to Eva Griffith for a transcription of
Press, corr. repr., 1979), pp. 244–5. this document.
6 André Hurault, Sieur de Maisse, A 16 Edwin Nungezer, A Dictionary of Actors
Journal . . . Anno Domini 1597, ed. G. B. Harrison (1929; repr. New York: Greenwood, 1968), p.
and R. A. Jones (London: Nonsuch, 1931), p. 95. 270.
7 B. Ravelhofer, “Dancing at the Court of 17 Walter Salmen, Der Tanzmeister: Geschichte
Queen Elizabeth”, in Christa Jansohn (ed.), und Profile eines Berufes vom 14. bis zum 19.
Queen Elizabeth I: Past and Present (Münster: Jahrhundert (Hildesheim: Olms, 1997), pp.
LIT, 2004), pp. 101–15; p. 104. Robert Mullally, 49–52.
“Measure as a Choreographic Term in the Stuart 18 Ben Jonson, The Masque of Queens, in Court
Masque”, Dance Research 16/1 (1998), pp. 67–73. Masques: Jacobean and Cardline Entertainments,
8 A Letter. . . of the Entertainment. . . at 1605–1640, ed. David Lindley (Oxford: Oxford
Killingwoorth Castl (London: s.n., 1575), p. 24. University Press, 1995), pp. 43–4, ll. 318–24. A
Traditionally ascribed to Robert Laneham but topic explored in Anne Daye’s work.
apparently produced by the scholar William 19 Francis Beaumont, The Masque of the Inner
Patten. See Benjamin Griffin, “The Breaking of Temple and Gray’s Inn, in A Book of Masques, ed.
the Giants: Historical Drama in Coventry and Spencer and Wells, p. 139, ll. 241–8.
London”, ELR 29/1 (1999), pp. 3–21. 20 An exception is Luminalia (1638), where
9 Cited from Judy Smith and Ian Gatiss, “What aristocrats and professionals performed together
Did Prince Henry Do with His Feet on Sunday in an entry. Interestingly, this masque was
19 August 1604?”, Early Music 14/2 (1986), commissioned by Henrietta Maria. On
pp. 198–207; p. 199; my translation. Anglo-French relations see also Peter Walls,
10 Gatiss and Smith point out that early British Music in the English Courtley Masque, 1604–1604
Library catalogues mention Italian dance books. (Oxford: Clarendon Press, 1996), ch. 6; and
295 Notes to pages 37–42

Marie-Claude Canova-Green, La plagiarised from an early version of Apologie, ed.


Politique-spectacle au grand siècle: les rapports B. Ravelhofer (Cambridge: RTM, 2000).
franco-anglais (Paris: Papers on French 33 BL MS Harl. 1026, fo. 7r , c.1633–1635 (date
Seventeenth Century Literature, 1993). according to HMC/NRA database, February
21 Martin Butler, “‘We Are One Mans All’: 2003).
Jonson’s The Gipsies Metamorphosed”, Yearbook 34 For instance, BL MSS Add. 41996, Lansd.
of English Studies 21 (1991), pp. 253–73. For 1115; Oxford Bodl. MSS Douce 280, Rawl. D
masques away from Whitehall see also James 864, Rawl. poet. 108; London Royal College of
Knowles, “The ‘Running Masque’ Recovered: A Music, MS 1119. Also BL MS Sloane 3858 –
Masque for the Marquess of Buckingham (c. “Chorea”.
1619–20)”, English Manuscript Studies 8 (2000), 35 Oxford, Bodl. MS Rawl. C 799. All citations
pp. 79–135; Timothy Raylor, The Essex House from the excellent The Travel Diary of Robert
Masque of 1621 (Pittsburgh: Duquesne Bargrave, ed. Michael Brennan (London:
University Press, 2000). Haykluyt Society, 1999), pp. 88–99; p. 96.
22 Margaret M. McGowan, “Ballets for the 36 Bargrave, Diary, p. 97. Bargrave’s
Bourgeois”, Dance Research 19/2 (2001), idiosyncratic vocabulary includes more
pp. 106–26. theatrical terminology than Playford’s first
23 William Whiteway’s diary, February 1634, in edition. In this it is reminiscent of another mid-
C. E. McGee, “‘Strangest Consequence from to late seventeenth-century country dance
Remotest Cause’: The Second Performance of source, the “Lovelace” or “Pattricke
The Triumph of Peace”, MRDE 5 (1991), Manuscript”. On the latter, see Carol Marsh,
pp. 309–42; p. 320. “The Lovelace Manuscript: A Preliminary
24 Orgel and Strong, Inigo Jones, p. 95, ll. 264, Study”, in Uwe Schlottermüller and Maria
272–3. Richter (eds.) Morgenröte des Barock: Tanz im 17.
25 Jennifer Nevile’s “Dance and the Garden: Jahrhundert (Freiburg: fa-gisis, 2004),
Moving and Static Choreography in Renaissance pp. 81–90, and a full transcription by Carol
Europe”, Renaissance Quarterly 52/3 (1999), Marsh and John Ward, Harvard Library Bulletin
pp. 805–36; p. 819. (forthcoming).
26 John Milton, Complete Shorter Poems, ed. 37 A Book of Masques, ed. Spencer and Wells,
John Carey (London: Longman, 2nd edn, 1997), p. 382, l. 43.
p. 185, stage direction, p. 189, l. 171, p. 229, l. 38 For cast lists and bills see Eleanore Boswell,
959. The Restoration Court Stage (1929; London:
27 See Anne Daye, “‘Youthful Revels, Masks, Allen & Unwin, 1969). Andrew R. Walkling,
and Courtly Sights’: An Introductory Study of “Masque and Politics at the Restoration Court:
the Revels Within the Stuart Masque”, Historical John Crowne’s Calisto”, Early Music 24/1 (1996),
Dance 3/4 (1996), pp. 5–22. pp. 27–62.
28 Jean E. Knowlton, “Some Dances of the 39 Peter Holman, Four and Twenty Fiddlers: The
Stuart Masque Identified and Analyzed”, 2 vols., Violin at the English Court, 1540–1690 (Oxford:
Ph.D. diss., Indiana University, 1966, vol. i, Clarendon Press, 1993), p. 367.
ch. 2. 40 Michel de Pure, Idée des spectacles anciens et
29 John Ward, “Newly Devis’d Measures for nouveaux (Paris 1668; facs. edn Geneva:
Jacobean Masques”, Acta Musicologica 60/2 Minkoff, 1972), pp. 248–9, my translation.
(1988), pp. 111–42, and “Apropos ‘The olde 41 John Crown, Calisto (London: Th.
Measures’”, Records of Early English Drama 18/2 Newcomb, 1675), “To the Reader”, sig. ar .
(1993), pp. 2–21. 42 Walkling, “Masque and Politics at the
30 Dudley Carleton on The Vision of the Twelve Restoration Court”, p. 51.
Goddesses (1604). Dudley Carleton to John 43 Thomas Carew, Coelum Britannicum, in
Chamberlain, ed. Maurice Lee (New Court Masques, p. 192, ll. 1023–5.
Brunswick, N.J.: Rutgers University Press, 1972), 44 Calisto, Act V, p. 79. – I would like to
p. 56. thank Lisa Vargo for her comments on this
31 Marshall McLuhan, “Roles, Masks, and chapter.
Performances”, New Literary History 2/3 (1971),
pp. 517–31; p. 518.
32 François de Lauze, Apologie de la danse 4 The baroque body
(n.p., 1623; facs. edn Geneva: Minkoff, 1977), 1 See Susan Leigh Foster, Dance and
and J. Wildeblood’s edition of the same (London: Narrative: Ballet’s Staging of Story and Desire
Muller, 1952). Barthélemy de Montagut, (Bloomington: Indiana University Press,
Louange de la danse, a manuscript treatise 1996).
296 Notes to pages 42–47

2 Alfred Heuss, “Eine Vorführung the hands, one the arms and the last to the
altfranzösischer Tänze”, Gesellschaft (1910), year ensemble.
2, part 12, pp. 386–9. 11 See “Repeatability, Reconstruction and
3 See Patrizia Veroli, “The Mirror and the Beyond”, in Theatre Journal 41/1 (March 1989),
Hieroglyph: Alexander Sacharoff and Dance pp. 56–74. (“Reproduction, reconstruction et
Modernism”, in Frank-Manuel Peter and Rainer par-delà”, Degrés 63 (Fall 1990), pp. 1–18). For
Stamm (eds.), Die Sacharoffs. Two Dancers the musical context of this debate, see Richard
within the Blaue Reiter Circle (Cologne: Taruskin, Text and Act: Essays on Music and
Wienand Verlag, 2002), pp. 169–217. Performance (New York: Oxford University
4 Georges Detaille and Gérard Mulys, Les Ballets Press, 1995).
de Monte-Carlo, 1911–1944 (Paris: Arc-en-ciel, 12 Marie Françoise Christout, Le Ballet de cour
1954), pp. 104–7. Tim Scholl argues for “a de Louis XIV, 1643–1762: Mises en scène (Paris:
reasonably authentic homage to the court of A. et J. Picard, 1967); Margaret M. McGowan,
Louis XIV” in Petipa’s Sleeping Beauty (1890). L’Art du ballet de cour en France 1581–1643
The retrospectivism of the Russian ballet spilled (Paris: CNRS, 1978); Rudolf zur Lippe,
over into Diaghilev’s repertoire. See Tim Scholl, Naturbeherrschung am Menschen (Frankfurt/M:
From Petipa to Balanchine: Classical Revival and Syndikat Reprise, 1979); Mark Franko, Dance as
the Modernization of Ballet (London and New Text: Ideologies of the Baroque Body (Cambridge:
York: Routledge, 1994). Cambridge University Press, 1993), also
5 See Charles M. Joseph, “The Making of Agon”, translated into French as La Danse comme texte:
in Lynn Garafola with Eric Foner (eds.), Dance ideologies du corps baroque (Paris: Editions
for a City: Fifty Years of the New York City Ballet Kargo, 2005).
(New York: Columbia University Press, 1999), 13 Giovanni Careri, Gestes d’amour et de guerre.
pp. 99–118. La Jérusalem délivrée, images et affects
6 See Raoul Auger Feuillet, Chorégraphie ou l’art (XVIe–XVIIIe siècle) (Paris: Editions EHESS,
de décrire la dance (Paris: Feuillet & Bruncl, 2005).
1700); and Pierre Rameau, Le Maı̂tre à danser 14 William Forsythe discussed baroque dance
(Paris: Jean Villette, 1725). For an interesting with German dance scholar Rudolf zur Lippe
discussions of Feuillet notation, see Jean Noel before creating Artifact (personal
Laurenti, “Feuillet’s Thinking”, Laurence Louppe communication, October 2000).
(ed.), in Traces of Dance: Drawing and Notations 15 Gilles Deleuze, The Fold: Leibniz and the
of Choreographers (Paris: Editions Dis Voir, Baroque, English trans. and foreword Tom
n.d.). For a discussion of alternative Conley (Minneapolis: University of Minnesota
seventeenth-century notational systems, see Press, 1992).
Rebecca Harris-Warrick and Carol G. Marsh, 16 Guy Debord, Society of the Spectacle, English
Musical Theatre at the Court of Louis XIV: Le trans. Donald Nicholson-Smith (New York:
Mariage de la Grosse Cathos (Cambridge: Zone Books, 1994).
Cambridge University Press, 1994). 17 Louis Marin, Portrait of the King, trans.
7 See Wendy Hilton, Dance of Court and Martha M. Houle (Minneapolis: University of
Theater: The French Noble Style, 1690–1725 Minnesota Press, 1988); Stephen Orgel, The
(Princeton: Princeton Book Company Illusion of Power: Political Theater in the English
Publishers, 1981). Also important for the Renaissance (Berkeley and Los Angeles:
interpretation and production of baroque dance University of California Press, 1975); Jean-Marie
were Melusine Wood, Belinda Quirey and Apostolidès, Le Roi-machine. Spectacle et
Shirley Wynne. politique au temps de Louis XIV (Paris: Editions
8 Despite her involvement with and significance de minuit, 1981).
to performance, Francine Lancelot was 18 For an overview of these affinities, see Omar
primarily involved in a research project, the Calabrese, Neo-Baroque: A Sign of the Times,
capstone of which was the publication of La trans. Charles Lambert (Princeton: Princeton
Belle Danse: catalogue raisonné fait en l’an 1995 University Press, 1992).
(Paris: Van Dieren, 1996). 19 Mark Franko, “Majestic Drag: Monarchical
9 See Linda Tomko, “Reconstruction”, in Performativity and the King’s Body Theatrical”,
International Encyclopedia of Dance (New York: in Drama Review 47/2 (T178) (Summer 2003),
Oxford University Press, 1998). pp. 71–87. See also, Mark Franko, “Figural
10 Dene Barnett, “The Performance Inversions of Louis XIV’s Dancing Body”, in
Practice of Acting: The Eighteenth Century”, in Mark Franko and Annette Richards (eds.),
Theatre Research International (1977). Three Acting on the Past: Historical Performance Across
articles appeared in this series; one devoted to the Disciplines (Hanover, N.H.: Wesleyan
297 Notes to pages 47–68

University Press University Press of New 17 Malpied, Traité sur l’art de la danse (Paris:
England, 2000), pp. 35–51. Bouı̈n, 1770), p. 84.
20 See Louise Fradenburg and Carla Freccero 18 Cf. Gennaro Magri, Theoretical and Practical
(eds.) Premodern Sexualities (New York and Treatise on Dancing, trans Mary Skeaping
London: Routledge, 1996). (London: Dance Books, 1988), chs. 13–58;
21 Morris spoke of the relation of his ballet to originally Trattato teorico-prattico di ballo
the AIDS crisis of the 1980s at a talk he gave at (Naples: Vicenzo Orsino, 1779).
Barnard College, New York City, on 11 October 19 Cf. George Touchard-Lafosse, Chroniques
2004. secrètes et galantes de l’Opéra, 4 vols., (Paris:
22 Oskar Schlemmer, The Letters and Diaries of Schneider, 1846); Emile Campardon, L’Académie
Oskar Schlemmer (Middletown, Conn.: Royale de Musique au XVIII siècle (Paris:
Wesleyan University Press, 1972), p. 196. Berger-Levrault & cie, 1884); François Henri
23 Paul Lacroix, Ballets et mascarades de cour Joseph Castil-Blaze, L’Academie Impériale de
sous Henri IV et Louis XIII (de 1581 à 1652) Musique. Histoire littéraire, musicale,
(1868–70; repr. Geneva: Slatkine, 1968); and chorégraphique, pittoresque, morale, critique et
Victor Fournel, Les Contemporains de Molière, galante de ce théâtre de 1645 à 1855, 2 vols.
receuil de comédies rares ou peu connues, jouées de (Paris: Castil-Blaze, 1855).
1650 à 1680 (Paris: 1866). 20 Noverre, Lettres, p. 53.
24 Collection housed in the Bibliothèque
Nationale de France in Paris. 6 The rise of ballet technique and training:
25 See Harry Haskell, The Early Music the professionalisation of an art form
Revival: A History (London: Thames & Hudson, 1 Debra Craine and Judith Mackrell, Oxford
1988). Dictionary of Dance (Oxford: Oxford University
26 For a corrective to the neglect of this Press, 2000), p. 40.
repertory, see Kate Van Orden, Music, Discipline, 2 Rose A. Pruiksma, “Generational Conflict and
and Arms in Early Modern France (Chicago: the Foundation of the Académie Royale de
University of Chicago Press, 2005). Danse: A Reexamination”, Dance Chronicle 26/2
27 See Margaret M. McGowan, The Court Ballet (2003), p. 169.
of Louis XIII: A Collection of Working Designs for 3 Ibid., p. 182. Pruiksma postulates that the
Costumes 1615–33 (London: Victoria and Albert original membership of the academy may have
Museum, n.d.). included a woman, as the list includes a Moliere
La Jeune, who might well be Marie Blanche
5 Choreography and narrative: the ballet Mollier, daughter of a court musician and
d’action of the eighteenth century dancer, Louis Mollier.
1 Louis de Cahusac, La Danse ancienne et 4 Régine Astier, “Académie Royale de Danse”, in
moderne ou Traité historique de la Danse (Paris: Selma Jeanne Cohen et al. (eds.), International
Jean Neaulme, 1754), vol. iii, p. 118. Encyclopedia of Dance (New York and Oxford:
2 Jean-Georges Noverre, Lettres sur la danse, et Oxford University Press, 1998), vol. i, p. 3.
sur les ballets (Lyons: Delaroche, 1760). 5 Astier, “Académie Royale de Danse”. See
3 John Weaver, Essay towards a History of also, Régine Astier, “In Search of L’Académie
Dancing (London: J.Tonson 1712). Royale de Danse”, York Dance Review 7 (1978),
4 Ibid., p. 159. pp. 2–14.
5 Ibid., p. 137. 6 Régine Astier, “Pierre Beauchamps”, in
6 Ibid., p. 167. International Encyclopedia of Dance, vol. i,
7 Ibid., p. 159. p. 397.
8 Ibid., p. 160. 7 Astier, “Académie Royale de Danse”, p. 3.
9 John Weaver, The Love of Mars and Venus 8 See Pierre Rameau, Le Maı̂tre à danser (Paris:
(London: W. Mears, J. Browne, 1717), p. 1. Jean Villette, 1725), p. 9. A contemporary
10 Mark Franko, Dance as Text. Ideologies of the English translation was published by John Essex
Baroque Body (Cambridge: Cambridge as The Dancing-Master (London: J. Essex and J.
University Press 1993), p. 33. Brotherton, 1728) and will be cited hereafter in
11 Feuillet, Chorégraphie ou L’art de décrire la translated material from Rameau.
danse (Paris: Feuillet & Brunel 1700). 9 Raoul Auger Feuillet, Chorégraphie ou l’art de
12 Cahusac, Danse, vol. iii, p. 125. décrire la dance (Paris: Feuillet & Brunel, 1700).
13 Ibid., p. 139. 10 They included fifteen theatrical dances
14 Noverre, Lettres, p. 84. composed by Feuillet and nine ballroom dances
15 Ibid., p. 262. composed by Guillaume-Louis Pécour, a leading
16 Ibid., p. 262. ballet master of the time.
298 Notes to pages 69–80

11 Ivor Guest, Le Ballet de l’Opéra de Paris 33 Magri/Skeaping, Theoretical and Practical


(Paris: Théâtre National de Treatise, pp. 48–9, 143.
l’Opéra/Flammarion, 1976), p. 19. 34 Noverre/Beaumont, Letters on Dancing, p. 18.
12 Quoted in Ivor Guest, The Ballet of the 35 Magri/Skeaping, Theoretical and Practical
Enlightenment (London: Dance Books 1996), Treatise, p. 61.
p. 23. 36 Cited in Edmund Fairfax, The Styles of
13 Johann Pasch, Beschreibung wahrer Eighteenth-Century Ballet (Lanham, Md.:
Tantz-Kunst (Frankfurt: Wolffgang Michahelles Scarecrow Press, 2003), p. 19.
and Johann Adolph 1707), p. 369. I am grateful 37 Magri/Skeaping, Theoretical and Practical
to Edmund Fairfax for this citation. Treatise, p. 128.
14 Cited in Régine Astier, “Marie Sallé”, in 38 Cited in Guest, The Ballet of the
International Encyclopedia of Dance, vol. v, Enlightenment, pp. 205-6.
p. 503. 39 Quoted in Régine Astier, “Marie Camargo”,
15 Régine Astier, “La Vie quotidienne des in International Encyclopedia of Dance, vol. ii,
danseurs sous l’Ancién Régime”, Les Gôuts p. 27.
reunis, 3rd series/1 (1982), p. 35. 40 Ibid., p. 28.
16 Rameau/Essex, The Dancing Master, p. 125. 41 Ivor Guest, Ballet under Napoleon (Alton:
17 Ibid., p. 2. Dance Books, 2002), p. 9.
18 Joan Wildeblood, The Polite World (London: 42 Fairfax, Styles of Eighteenth-Century Ballet,
Davis-Poynter, 1973), p. 94. p. 276.
19 Jean-Georges Noverre, Lettres sur la danse, 43 John Chapman, “Auguste Vestris”, in
sur les ballets et les arts (St Petersburg 1803), International Dictionary of Ballet, ed. Martha
transl. Cyril W. Beaumont as Letters on Dancing Bremser (London: St James Press, 1993), vol. ii,
and on Ballets (New York: Dance Horizons p. 1485.
1968), p. 117. 44 Guest, The Ballet of the Enlightenment,
20 Giambattista Dufort, Trattato del ballo nobile p. 24.
(Naples: Felice Mosca 1728), pp. 4–5. 45 John V. Chapman, “The Paris Opera Ballet
21 C. Sol, Méthode très facile et fort nécessaire, School, 1798–1827”, Dance Chronicle 12/2
pour montrer à la jeunesse de l’un et l’autre sexe la (1989), pp. 196–220.
manière de bien dancer (La Haye: l’Auteur 1725), 46 Carlo Blasis, Notes Upon Dancing, Historical
p. 52. and Practical, transl. R. Barton (London: M.
22 Rameau/Essex, The Dancing Master, p. 11. Delaporte, 1847), pp. 56–61.
23 Sol, Méthode, p. 18. 47 Carlo Blasis, Traité élémentaire, théorique et
24 Gennaro Magri, Trattato teorico-prattico di pratique de l’art de la danse (Milan: Joseph Beati
ballo (Naples: Vicenzo Orsino 1779), vol. ii, et Antoine Tenenti, 1820), trans. Mary Stewart
pp. 10–11. Hereafter, translated material from Evans as An Elementary Treatise upon the Theory
Magri will be from the English translation by and Practice of the Art of Dancing (New York:
Mary Skeaping, Theoretical and Practical Treatise Dover 1968), p. 5.
on Dancing (London: Dance Books, 1988). 48 Blasis, Notes, p. 62.
25 Noverre/Beaumont, Letters on Dancing,
p. 119.
26 Magri/Skeaping, Theoretical and Practical 7 The making of history: John Weaver and
Treatise, p. 75. the Enlightenment
27 Ibid., p. 74. 1 Quoted in Richard Ralph, The Life and Works
28 Carlo Blasis, The Code of Terpsichore, trans. of John Weaver. An Account of his Life, Writings
R. Barton (London: printed for James Bullock and Theatrical Productions, with an Annotated
1828), p. 102. Reprint of his Complete Publications (London:
29 Noverre/Beaumont, Letters on Dancing, p. 19. Dance Books 1985), p. 50.
30 Ibid., p. 91. 2 Richard Ralph, The Life and Works of John
31 Giovanni-Andrea Gallini, A Treatise on the Weaver, Ibid., p. 108.
Art of Dancing (London: The author, 1762), 3 John Weaver, An Essay Towards an History of
p. 236. Dancing (London: Printed for Jack Tonson,
32 G. Léopold Adice, Théorie de la gymnastique 1712), reproduced in Ralph, The Life and Works
de la danse théâtrale (Paris: Chais 1859), p. 80. A of John Weaver, p. 395.
portion of his book dealing with the ballet class 4 Ralph, The Life and Works of John Weaver
is translated by Leonore Loft in Selma Jeanne p. 405.
Cohen, Dance as a Theatre Art (New York: 5 Ibid.
Harper & Row, 1974), p. 76. 6 Ibid. p. 469.
299 Notes to pages 80–100

7 Carol Lee, Ballet in Western Culture: A History 31 Weaver, Essay Towards the History of
of its Origins and Evolution (New York and Dancing, p. 403.
London: Routledge, 2002), p. 368. 32 Weaver, A Small Treatise of time and Cadence
8 Poems by Soame Jenyns, containing Art of in Dancing, reproduced in Ralph,The Life and
dancing, To Lord Lovelace, Essay on virtue, Works of John Weaver, p. 365.
Written in Locke, Epitaph on Doctor Johnson; to 33 William Wycherly, The Gentleman Dancing
which is prefixed a sketch of the author’s life Master, www3.shropshire-
(Manchester 1797). Canto 2, p. 8 cc.gov.uk/etexts/E000294.htm.
9 John Essex, The dancing-master: or, The art of 34 Ibid., Act ii, scene ii.
dancing explained. Wherein the manner of 35 Joshua Reynolds, Seven Discourses on Art
performing all steps in ball dancing is made easy (delivered between 1769 and 1776),
by a new and familiar method. In two parts . . . www.authorama.com/book/seven-discourses-
The whole containing sixty figures drawn from the on-art.html.
life, and curiously engraved on copper plates. Done
from the French of Monsieur Rameau (London: 8 Jean-Georges Noverre: dance and reform
Printed and sold by him, and J. Brotherton, 1 Samuel Johnson, A Dictionary of the English
1728), p. x. Language (London, 1755).
10 Quoted in Ralph, The Life and Works of John 2 Cf. Horst Koegler, “The Northern Heirs of
Weaver, p. 107; Charles Burney, Choregraphy, Noverre”, Dance and Dancers (London, 1987),
The Cyclopaedia, ed. Abraham Rees (1819), vol. p. 24.
vii. 3 Deryck Lynham,The Chevalier Noverre: Father
11 Ralph, The Life and Works of John Weaver, of Modern Ballet. (London: Dance Books, 1972),
p. 142. p. 117.
12 Cf. ibid., p. 49. 4 Lillian Moore, “Noverre, First of the
13 Considered one of the greatest British Moderns”, Dance Magazine 9(1952), p. 44.
playwrights of comedy. 5 Artur Michel, “Le Ballet d’action avant
14 Quoted in Hugh Arthur Scott, ”London’s Noverre”, Archives Internationale de Dance, pt. 2
earliest public concerts”, Musical Quarterly 22 (Octobre 1935), p. 116.
(1936), p. 454. 6 Cf. Denis Diderot,Oeuvres complètes de
15 Roger North, Memoirs of Musick being some Diderot, 20 vols., ed. J. Assézat and M. Tourneux
Historio-criticall Collections of that Subject (Paris: Garnier frères, 1875–7), vol. vii, p. 157.
(1728), ed. Edward F. Rimbault (London, 1846), 7 Cf. Jean-Jacques Rousseau, Julie ou la Nouvelle
p. 111. Héloise,Oeuvres complètes (Geneva: Editions
16 George Savile Marquis of Halifax, Advice to a Gallimard, Bibliothéque de la Pléiade, 1964),
Daughter, Chiefly with Regard to Religion vol. ii, pp. 287–9.
(Aberdeen: printed for and by Francis Douglass 8 International Encyclopaedia of Dance (New
and William Murray, 1688; 7th edn 1701), York: Oxford University Press, 1998), p. 174.
p. 141. 9 Voltaire (François-Marie Arouet),
17 Petition to the Vice-Chamberlain. Correspondance, Oeuvres completés, 50 vols., ed.
18 Weaver, Essay Towards the History of L. Moland (Paris: Garnier, 1877–85), vol. liii,
Dancing, p. 658. p. 76.
19 Ibid., p. 612. 10 Voltaire,Correspondance, vol. liv, pp. 18–85.
20 Ibid., p. 614. 11 Noverre, Lettres sur La danse (Paris: Editions
21 Ibid., p. 666. Lieutier, 1952), 92. This edition is based on that
22 Ibid., p. 436. of St Petersburg, 1807). Subsequent references in
23 Ibid. the text are to the English translation, Letters on
24 Ibid. Dancing and on Ballets, trans. Cyril Beaumont
25 Ibid. (New York: Dance Horizons, 1968).
26 Ibid.
27 Charles Rosen, The Classical Style: Haydn, 9 The French Revolution and its spectacles
Mozart, Beethoven (New York: W.W. Norton, 1 See Ivor Guest, Ballet under Napoleon (Alton:
1972; 1997), p. 172 n. 1. Dance Books, 2002), p. 20: “Aristocratic
28 Anthony Ashley Cooper, Third Earl of appropriation of the court ballet’s sensuous
Shaftesbury, Soliloquy: Or, Advice to an author techniques likely reinforced the overall image of
(London: Printed for John Morphew, 1710), royal authority, but it also established a
p. 65. fascination with bodily display and attraction
29 Ibid., p. 135. that would catalyze the art forms associated with
30 Ibid. ‘aristocracy’ in the following decades.”
300 Notes to pages 100–22

2 Cf. Sarah Cohen, Art, Dance and the Body in supporting bourgeois and patriarchal marriage
French Culture of the Ancien Régime. practices.
(Cambridge: Cambridge University Press, 2000). 5 René Girard’s concept of mimetic desire, or
3 Cf. Inge Baxmann, Die Feste der Französischen triangular desire, captures aptly the romantic
Revolution. Inszenierung von Gesellschaft als love element of the ballets’ plots. He argues that
Natur (Weinheim/Basel: Beltz, 1989). the desire, which a subject, a man, has for an
4 Boullée, cited in Fritz Wagner, Isaac Newton object (a woman, for example) has more to do
im Zwielicht zwischen Mythos und Forschung. with the prestige associated with the person who
(Freiburg/München: Verlag Karl Alber, 1976), possesses or is about to desire the same object
p. 127. than with the object’s intrinsic worth.
5 Abbé Emmanuel Joseph Sieyès, Qu’est-ce que Simultaneously the person is not passive. He too
le tiers état? (Paris: Flammarion, 1989), is invested in the object’s worth and thus seeks to
pp. 173–4. What is the Third Estate? Trans. M. awaken desire in the subject. As they copy one
Blondel, ed. S. E. Finer (London and Dunmow: another, the triangulation of the mimetic desire
Pall Mall Press, 1963), p. 162. turns them into rivals. See in particular his
6 Jacques Grenier, Opinion sur la question de Deceit, Desire, and the Novel: Self and Other in
savoir si l’on doit supprimer de la formule du Literary Structure, trans. Yvonne Freccero
serment civique les mots de haine à l’anarchie. (Baltimore: Johns Hopkins Press, 1965).
Paris, an vii, cited in Mona Ozouf, La Fête 6 For the circulation of répétiteurs, see Marian
Révolutionnaire (Paris: Gallimard 1976), p. 339. Smith, “The Earliest Giselle? A Preliminary
7 See Guest, Ballet under Napoleon, pp. 17–18. Report on a St Petersburg Manuscript”, Dance
8 S. Thomas, Nancy avant et après 1830, cited in Chronicle: Studies in Dance and the Related Arts
Ozouf, La Fête Révolutionnaire, p. 135. 23/1 (2000), pp. 29–48.
9 Plan de la Fête de l’Etre Suprême, qui sera 7 More exact dates would be 1831 to 1847,
célébrée à Tours, le 20 Prairial en exécution du marked by the premiere of the opera Robert le
Décret du 18 Floréal, l’an second de la diable with its “Ballet des nonnes” and finishing
République, une et indivisible. Bibliolegue with Ozaı̈ (26 April) and La Fille de marbre (20
Nationale Paris, p. 8/9. October).
10 Ibid., p. 6. 8 The various transformations Giselle has
11 Receuil de chants philosophiques, civiques et undergone in its history has pulled it towards
moraux, à l’usage des Fêtes Nationales et more class- or race-conscious readings. Others
Décadaires. Paris An vii, p. 58. have looked at Giselle’s madness in terms of
drug addiction.
10 Romantic ballet in France: 1830–1850 9 I am indebted to Marion Kant’s essay
1 Gas lighting to illuminate the stage was “Giselle – la jolie morte”, which traces and
introduced in 1822 while the dimming of house contextualises through a reading of Giselle the
lights occurred in 1831. transformations the wili has undergone, and
2 Susan Leigh Foster, “The Ballerina’s Phallic Christianity’s manipulation of dancing, in Musik
Pointe”, in Susan Leigh Foster (ed.), und Gesellschaft, ed. Verband der Komponisten
Corporealities: Dancing Knowledge, Culture and und Musikwissenschaftler der DDR (Berlin:
Power (London: Routledge, 1996), p. 5. For the Henschelverlag), no. 3 (1988).
tie with sexuality, see Felicia McCarren, Dance 10 For travesty dancing, see Lynn Garafola,
Pathologies: Performance, Poetics, Medicine “The Travesty Dancer in Nineteenth Century
(Stanford, Calif.: Stanford University Press, Ballet”, Dance Research Journal 17/2 (Fall–Spring
1998). 1985–6), pp. 35–40. Examples of ballets with
3 Later in the late 1840s and 1850s, Arthur dancers en travesti include: Le Diable boiteux, Le
Saint-Léon returned to Filippo Taglioni’s Diable amoureux, and Paquita with its squadron
practice and devised the plot, the choreography of hussars.
and often partnered the leading dancer, 11 Few women worked as teachers or
sometimes even playing the telling tune on his choreographers at the Paris Opéra until the
violin on stage. 1860s. Thérèse Elssler stands out for her
4 See Sally Banes’s section “The Marriage plot” choreographic work on her sister Fanny Elssler.
in her introduction and the chapter “The Her ballet, La Volière ou les oiseaux de Boccace
Romantic Ballet: La Sylphide, Giselle, Coppélia”, (1838), set in the Caribbean on a plantation,
in Dancing Women: Female Bodies on Stage from which all men have been banished,
(London and New York: Routledge, 1998), configured gender differently and asserted
pp. 5–7 and 12–42, where she argues against women’s emotional growth. Fanny Elssler (or
reading the ballets as simple depictions her sister) choreographed a ballet entitled La
301 Notes to pages 123–36

Salamandrine, which was first produced at Dance Books, 1992). Besides that, the Royal
Covent Garden on 22 May 1847 (see the Danish Ballet has published The Bournonville
Bibliothèque de l’Opéra’s libretto (Liv m. 149)), Schools – the DVD, coached by Frank Andersen,
some 113 years after Marie Sallé’s staging and Anne Marie Vessel Schlüter, Eva Kloborg,
performance of Pygmalion. Marie Taglioni Flemming Ryberg and Dinna Bjørn. Director:
would return later to teach and she Ulrik Wivel, 2005.
choreographed and produced the ballet Le 5 According to the Danish ballet critic Erik
Papillon at the Paris Opéra in 1860. Other female Aschengreen, Bournonville’s illegitimate child
teachers, such as Madame Dominique, had always been known about within a small
developed a loyal following of students during circle of people, but only in 1997 did it become
the latter part of the century. public knowledge when Danish ballet historian
12 For romanticism’s use of the fantastic and Knud Arne Jürgensen included the daughter in
the question of marriage see Scott M. Sprenger, his article on “The Ballet Tradition” (1997), also
“Figures du fantastique: la logique du mariage in his Bournonville biography for the catalogue
raté chez Gautier et chez Zola”, Bulletin de la for the exhibition Europæeren Bournonville
Société Théophile Gautier 21 (1999), (Bournonville the European), vol. i, p. 20 at the
pp. 191–207; esp. p. 192. Royal Library, Copenhagen, 2000.
13 Susan Leigh Foster elaborates the recurrence 6 The most famous painting is that by the
of this story in Choreography and Narrative: Danish painter Carl Bloch from 1876 where
Ballet’s Staging of Story and Desire (Bloomington the seventy-one-year-old Bournonville is
and Indianapolis: Indiana University Press, portrayed with five different medals. He
1996). received the royal honour of being knighted
14 For a similar argument regarding the Ridder af Dannebrogordenen 1848. Cf. Dansk
novelistic characters, see Jann Matlock, “Novels Biografisk Leksikon (Copenhagen: J. H. Schultz,
of Testimony and the ‘Invention’ of the Modern 1979).
French Novel”, Cambridge Companion to the 7 Den nøgne Guldalder (The Nude Golden Age),
French Novel from 1800 to the Present exhibition catalogue by Annette Johansen,
(Cambridge: Cambridge University Press, 1997), Emma Salling, Marianne Saabye (Copenhagen:
pp. 10-35; p. 33. The Hirschsprung Collection 1994), p. 164 (with
15 The origin of this appellation is disputed, but English summary and translation).
Théophile Gautier used it consistently to refer to 8 Women were not seriously invited to share
the dancers of the Paris Opéra who performed these intellectual discussions in the circles
walk-on roles or as part of the corps de ballet around Brandes and they were not accepted as
(“Le Rat”, Les Français peints par eux-mêmes: students at the University of Copenhagen until
encyclopédie morale du dix-neuvième siècle 1875.
(Paris: L. Curmer, 1841–2), vol. iii, pp. 249–56). 9 Mit Theaterliv was published in five volumes:
Honoré de Balzac also uses the term. vol. i in 1847 (and 1848), vol. ii in 1865 and the
others in 1877–8 (Copenhagen: C. A. Reitzel).
11 Deadly sylphs and decent mermaids: the Niels Birger Wamberg created a new edition,
women in the Danish romantic world of linguistically revised, for the centenary
August Bournonville celebration of Bournonville’s death: Mit Teaterliv
1 Bournonville was ballet master at the Royal (Copenhagen: Thaning & Appel, 1979). On the
Danish Ballet 1830–77, except for his time as same occasion, Patricia McAndrew translated
ballet master in Vienna, Austria 1855–6 and as this gigantic work into English, with a foreword
director in Stockholm, Sweden 1861–4. by ballet dancer Erik Bruhn and an introduction
2 Hvor danser Den Kgl. Ballet hen? pp. 52 and by ballet critic Svend Kragh-Jacobsen: My
55. www.schoenbergske.dk. Theatre Life (London: A. & C. Black, 1979).
3 Cf. Chapter 10 by Sarah Davis Cordova 10 “Choreografisk Troesbekjendelse”
“Romantic ballet in France: 1830–1850”. (Choreographic Credo) was published in his My
4 Examples of the Bournonville style and its Theatre Life, vol. ii, ch. 1. It was reprinted in Erik
basic exercises can be seen on the video Aschengreen’s Ballettens Digter. 3 Bournonville
Bournonville Ballet Technique: Fifty essays (Copenhagen: Rhodos, 1977) (Danish
Enchaı̂nements annotated by Hans Beck only). Translations taken from Patricia
(London: Dance Books 1993), featuring Rose McAndrew, see n. 9.
Gad and Johan Kobborg, directed by Vivi Flindt 11 In Napoli, the change of dress for Teresina
and published along with the book Bournonville takes place so quickly that the surprised sigh of
Ballet Technique: Fifty Enchaı̂nements by Vivi the audience has become part of the
Flindt and Knud Arne Jürgensen (London: transformation. The secret consists of two
302 Notes to pages 136–43

dresses on top of each other – and a decisive, with decor by her majesty, Queen Margrethe II)
masculine hand from underneath. and Napoli several times since 1992 (with
12 Elsa Marianne von Rosen and Allan Henning Kronstam, Dinna Bjørn and Anne
Fridericia: Napoli, Gothenburg Ballet, 1971. The Marie Vessel Schlüter).
same production was later staged for the Kirov 16 Jeg Dig Elsker (I love you). An interpretation
Ballet in St Petersburg and the Royal Swedish of La Sylphide with ballet director Nikolaj
Ballet in Stockholm. Tim Rushton, Napoli – den Hübbe, directed by Ulrik Wivel. Dancers: Mads
nye by (Napoli – The New Town), New Danish Blangstrup, Gudrun Bojesen, Lis Jeppesen.
Dance Theatre, Copenhagen, 2003. DVD, 25 minutes. The Royal Theatre and the
www.nddt.dk. Thomas Lund and Johan Holten, Danish Film Institute 2005.
En anden akt (Another Act), Copenhagen
International Ballet, Bellevue Teatret, 12 The orchestra as translator: French
Copenhagen, 2004. www.sommerballet.dk. nineteenth-century ballet
Dinna Bjørn and Frank Andersen, Napoli, 1 Operas of four acts or longer nearly
stagings for the Royal Danish Ballet in 1992, always included a ballet; shorter operas (or
1998, 2005 and for the Finnish National Ballet in shortened versions of longer operas) were
2005. performed on the same evening as independent
13 In Bournonville’s time, Abdallah was ballets.
performed between 1855 and 1858. Then it was 2 Le Journal des débats, 28 September 1822. The
practically forgotten until ballet director Bruce score for Alfred le grand, first performed 18
Marks and his wife, Toni Lander, bought September 1822, was by Gallenberg and
Bournonville’s handwritten libretto at an Dugazon.
auction in New York in 1970. In 1985, they 3 Auguste Baron, Lettres et entretiens sur
reconstructed the ballet, together with Danish la danse (Paris: Dondey-Dupré, 1824),
Bournonville teacher Flemming Ryberg. At that p. 296.
time, Toni Lander had accepted the position as 4 Le Moniteur universel, 21 September 1827.
ballet director of the Royal Danish Ballet and she 5 La Siècle, 23 September 1836.
had planned to bring Abdallah to Copenhagen 6 La Sylphide, 26 September 1840.
with her. Tragically, Toni Lander died of a 7 Le Constitutionnel, 11 August 1845.
sudden illness. In spite of this, Bruce Marks 8 Gustave Chouquet, Histoire de la musique
and Flemming Ryberg together with dramatique en France (Paris: Librairie Firmin
Bournonville coach Sorella Englundand Didot Frères, 1873), p. 170.
managed to stage the ballet for the Royal 9 See Stephanie Jordan, “The Role of the Ballet
Danish Ballet in 1986: Abdallah had come back Composer at the Paris Opéra: 1820–1850”,
home. Dance Chronicle, 4/4 (1982), pp. 374-88; and
14 At the gala for the third Bournonville Festival Ivor Guest, The Romantic Ballet in Paris
in 2005, the Jockey Dance was danced as a unisex (London: Dance Books, 1980), p. 10.
duet by two principal ballerinas en travestie, 10 Over a half-century later, Tchaikovsky wove
Gudrun Bojesen and Gitte Lindstrøm. the royalist tune “Vive Henri Quatre” into the
15 Artistic directors of the Royal Danish Ballet final scene of Sleeping Beauty (1890) to help
since the first Bournonville Festival: Henning conclude that ballet “with great pomp and
Kronstam 1978–85, Frank Andersen 1985–94, ostentation”, and a royalist flair, recalling the
Peter Schaufuss 1994–5, Johnny Eliasen Parisian tradition of using it to suggest just those
(temporary appointment) 1995–7, Maina qualities.
Gielgud 1997–9, Aage Thordal-Christensen 11 Such “vocal” passages were frequently
1999–2002, Frank Andersen 2002–8. Frank indicated in rehearsal scores, and the text
Andersen (born 1953), was a Danish principal “spoken” by the characters written below the
dancer and Bournonville instructor. He was staff.
artistic director at the Royal Danish Ballet 12 To Le Corsaire, 1856, a new divertissement
1985–94. He then was ballet director for the was added by Delibes in 1857. The ballet was
Royal Swedish Ballet 1995–9 and artistic adviser restaged by Petipa at the St Petersburg
for the Chinese National Ballet, and judge at Maryinsky Theatre, with additional music by
international dancer competitions, before he Drigo, Minkus and Pugni. The music for the
returned to his position as artistic director of the famous pas de deux so often performed without
Royal Danish Ballet in 2002. Abroad, he has the rest of the ballet is by Drigo.
staged Napoli and La Sylphide. For the Royal 13 Galops may be found, to name only a few
Danish Ballet, he has staged A Folk Tale in 1991 examples from the stage of the Paris Opéra, in
(together with Anne Marie Vessel Schlüter and the ballets La Jolie Fille de Gand, Le Diable à
303 Notes to pages 143–52

quatre, La Fille de marbre, Diavolina, La Encyclopedia of Dance (New York and


Korrigane, and the operas Dom Sébastien, Oxford: Oxford University Press, 1998), vol. ii,
Gustave III and Le Prophète. p. 368.
14 The polka, which was according to Guest 29 L’Opinion, trans. Cyril Beaumont, in
(The Romantic Ballet in Paris, p. 229) “all the Complete Book of Ballets (London: Putnam,
rage in the ballrooms and public balls” in the 1937; repr. 1956), p. 495.
early 1840s, is said to have made its first 30 Le Telegraphe, 3 June 1885.
appearance on the stage of the Opéra in the 31 “Le XIX Siècle” (in Delibes dossier d’artiste,
ballet Lady Henriette (first performed 21 Bibliothèque de l’Opéra); George Balanchine
February 1844). and Francis Mason, Balanchine’s Complete
15 La France musicale, 17 August 1845. Stories of the Great Ballets (Garden City, N.Y.:
16 Le Nord, 13 July 1863. Doubleday, 1977), p. 608.
17 Such levity proved a valuable 32 Jouvin, Hérold, p. 194.
counterbalance to the serious and even 33 Edouard Deldevez, Mes Mémoires (Paris:
violent tone of many operas with which ballet Marchessou fils, 1890), p. 34.
shared the stage. See Smith, Ballet and Opera, 34 5 December 1880.
pp. 59–96. 35 See Ivor Guest, Le Ballet de l’Opéra de
18 La France musicale, 11 August 1844. Paris (Paris: Flammarion, 2001), pp. 297–314,
19 La Revue et gazette musicale, 27 September for a list of the Opéra’s ballets and their
1840. composers, 1776–2000.
20 Le Nord, 13 July 1863.
21 La Revue et gazette musicale, 5 December 13 Russian ballet in the age of Petipa
1880. 1 Russian Ballet Master: The Memoirs of Marius
22 Le Menéstrel, 12 March 1882. This music Petipa, ed. Lillian Moore, trans. Helen Whittaker
fared well in the next century, in Serge Lifar’s (London: A. & C. Black, 1958), p. 23.
Suite en blanc (1943), revived as Noir et blanc, 2 Vera M. Krasovskaya, “Petipa, Marius”,
and as a concert suite. International Encyclopedia of Dance, ed. Selma
23 Letter to The Times of London of 6 July 1914, Jean Cohen and Dance Perspectives
reprinted in Roger Copeland and Marshall Foundation (Oxford: Oxford University Press,
Cohen (eds.), What is Dance? Readings in Theory 1998, 2005) (hereafter IED), p. 150.
and Criticism (Oxford: Oxford University Press, Jean-Antoine Petipa died in St Petersburg in
1983), p. 260. 1855, at which point Marius took over his
24 This choreography also uses ballet music position, which he continued to occupy until
from Verdi’s Jérusalem and Don Carlos. 1863, when at Petipa’s request, the director of
25 Music of Delibes’s La Source and Naila was the Imperial Theatres appointed Christian
used for Balanchine’s La Source (1968); parts of Johansson to succeed him.
the Sylvia score have also been choreographed 3 Petipa, Memoirs, p. 22. A “benefit” refers to a
by, among others, Balanchine and Ashton. The performance of which all or part of the proceeds
many other choreographies to Delibes’s ballet were earmarked for a particular artist or group
music include Léo Staats’s Soir de fête (1921), (such as the corps de ballet).
using Le Corsaire and La Source, which has been 4 Natalia Roslavleva, Era of the Russian Ballet,
performed over 250 times. foreword Ninette de Valois (London: Gollancz,
26 1 August 1847. His ballets include La Chatte 1966), p. 22.
blanche (Paris, 1830), Faust (London, 1833), La 5 Ibid., p. 32.
Fille du Danube (Paris, 1836), Les Mohicans 6 This is the date of the Russian as opposed to
(Paris, 1837), L’Ecumeur de mer (St Petersburg, the French premiere, which took place at the
1840), Die Hamadryaden (Berlin, 1840), Giselle Paris Opéra in 1856. The 1858 St Petersburg
(Paris, 1841), La Jolie Fille de Gand (Paris, 1842), production was by Jules Perrot. Petipa restaged
Le Diable à quatre (1845), The Marble Maiden the ballet five years later, with significant
(London, 1845), Griseldis ou les cinq senses revisions in 1868 (when he choreographed the
(Paris,1848), La Filleule des fées (Paris, 1849), “Jardin Animé”), 1885 and 1899. See “Works by
Orfa (Paris, 1852), and Le Corsaire (Paris, 1856). Marius Petipa in Russia”, in The Diaries of
27 Remarks made by Adam while working on Marius Petipa, ed., trans. and intro. Lynn
La Filleule des fées, quoted in Benoı̂t Jouvin, Garafola, Studies in Dance History, 3/1 (Spring
Hérold, sa vie et ses oeuvres (Paris: Heugel, 1868), 1992), pp. 81–2.
p. 148. 7 La Sylphide premiered at the Paris Opéra in
28 “Léo Delibes”, in Selma Jeanne Cohen and 1832. Three years later, Antoine Titus staged the
Elizabeth Aldrich (eds.), International ballet after Taglioni at the Bolshoi Theatre,
304 Notes to pages 153–63

St Petersburg. The ballet was restaged by Taglioni 25 The Yearbooks of the Imperial Theatres
in 1837. Petipa restaged the work, again after (Yezhegodniki Imperatorskikh Teatrov (St
Taglioni but with additional music by Riccardo Petersburg/Petrograd, 1890–1915) listed all the
Drigo, in 1892. “Sil’fida”, Balet Entsiklopediia dancers and artistic personnel in the employ of
(hereafter Soviet Ballet Encyclopedia) (Moscow: the St Petersburg and Moscow companies.
Sovetskaia Entsiklopediia, 1981); Cyril W. 26 Krasovskaya, “Petipa”, pp. 153–4.
Beaumont, Complete Book of Ballets (London: 27 Quoted in Wiley, Tchaikovsky’s Ballets,
Putnam, 1937), p. 103. p. 221.
8 Quoted in Roslavleva, Era of the Russian 28 Vera M. Krasovskaya, “Ivanov, Lev”, IED,
Ballet, p. 59. pp. 565-6.
9 Quoted ibid., pp. 66–7. 29 Wiley, Ivanov, p. 174.
10 Krasovskaya, “Petipa”, p. 149. 30 Ibid., p. 176.
11 Petipa, Memoirs, p. 2. The “bow” is a 31 Sergei Khudekov, “The Petersburg Ballet
reference to the small violin or “pochette” During the Production of The Little
that ballet teachers played as accompaniment in Humpbacked Horse (Recollections)”, in Wiley, A
class. Century of Russian Ballet, pp. 266–7.
12 For an account of this unsuccessful tour, see 32 Tamara Karsavina, “Platon Karsavin”,
Lillian Moore, “The Petipa Family in Europe Dancing Times (October 1964), p. 12.
and America”, Dance Index, 1/5 (May 1942), 33 “Before graduation . . . many students never
pp. 76–8. imagined the poverty in which their parents
13 Yury Slonimsky, “Marius Petipa”, trans. lived”, wrote the dancer Anna Natarova in her
Anatole Chujoy, Dance Index, 6/5–6 (May–June recollections of theatre life in the 1840s.
1947), p. 106. “However badly they fed us at school, it was
14 “A Glance Behind the Scenes”, Appleton’s worse at home” (“From the ‘Recollections of the
Journal of Literature, Science and Art, 1 April Artiste A. P. Natarova”, in Wiley, A Century of
1876, p. 433. According to the author, some fifty Russian Ballet, p. 160).
dancers appeared in the divertissement. 34 For a description of training in the 1830s and
15 August Bournonville, My Theatre Life, trans. 1840s, see “Recollections of T. A. Stukolkin”,
Patricia N. McAndrew, intro. Svend Kragh- ibid., pp. 108-9. For training in the late
Jacobsen (Middletown, Conn.: Wesleyan nineteenth and early twentieth centuries, see
University Press, 1979), p. 581. Tamara Karsavina, Theatre Street: The
16 Ibid., pp. 581–2. Reminiscences of Tamara Karsavina, foreword
17 Slonimsky, “Marius Petipa”, p. 115. J. M. Barrie (London: Heinemann, 1930), ch. 7
18 Roland John Wiley, The Life and Ballets of (“The Pupil”); Princess Romanovsky-Krassinsky
Lev Ivanov (Oxford: Clarendon Press, 1997), [Kshesinska], Dancing in Petersburg: The
p. 106. Memoirs of Kschessinska, trans. Arnold Haskell
19 Ibid., p. 93. For abolishing the post of official (London: Gollancz, 1960), ch. 3 (“The Imperial
ballet composer, see A Century of Russian Ballet: Ballet School”); Michel Fokine, Memoirs of a
Documents and Accounts, 1810–1910, ed. and Ballet Master, trans. Vitale Fokine, ed. Anatole
trans. Roland John Wiley (Oxford: Clarendon Chujoy (Boston: Little, Brown, 1961), ch. 2
Press, 1990), p. 350. (“Life in School”); and Bronislava Nijinska,
20 Quoted in Vera Krasovskaya, “Marius Early Memoirs, ed. and trans. Irina Nijinska and
Petipa and ‘The Sleeping Beauty’”, trans. Cynthia Jean Rawlinson (New York: Holt, Rinehart &
Read, Dance Perspectives 49 (Spring 1972), p. 21. Winston, 1981), chs. 11–17.
21 Ibid. 35 Vera Krasovskaya, Vaganova: A Dance
22 Wiley reproduces Petipa’s instructions to Journey from Petersburg to Leningrad,
Tchaikovsky and his “plan” for the ballet in trans. Vera M. Siegel, intro. Lynn Garafola
Tchaikovsky’s Ballets: Swan Lake, Sleeping Beauty, (Gainesville: University Press of Florida, 2005),
Nutcracker (Oxford: Clarendon Press, 1985), p. 51.
Appendix D, pp. 354–70. 36 Prince Serge Wolkonsky, My Reminiscences,
23 For criticism of the ballet, see Tim Scholl, trans. A. E. Chamot (London: Hutchinson,
Sleeping Beauty: A Legend in Progress (New n.d.), p. 107. For Kshesinska’s account of the
Haven: Yale University Press, 2004), ch. 1 incident, see Kschessinska [Kshesinska],
(“Genre Trouble”). Dancing in Petersburg, pp. 81–2.
24 Alexandre Benois, Reminiscences of the 37 Ksehessinska, Dancing in Petersburg, p. 27.
Russian Ballet, trans. Mary Britnieva 38 Richard Taruskin, The Oxford History of
(London: Putnam, 1941), Reminiscences, Western Music, vol. iv (New York: Oxford
p. 128. University Press, 2005), p. 142.
305 Notes to pages 164–81

14 Opening the door to a fairy-tale world: the use of a children’s tale for the ballet, remarks
Tchaikovsky’s ballet music on the composer’s skill at writing music of both
1 Solomon Volkov, Balanchine’s Tchaikovsky: a serious, melancholic nature as well as happy,
Interviews with George Balanchine, trans. lively material, and touches on the Russianness
Antonina Bouis (New York: Simon & Schuster, of his music.
1985), p. 144. From letter to Taneyev (Taneev). 15 Wiley, Tchaikovsky’s Ballets, p. 361: “Musique
2 Volkov, Balanchine, p. 153. Herman Laroche douce et un peu ricaneuse”. This request seems
recalls Tchaikovsky’s turn from opera to ballet. rather confusing. How can the Lilac Fairy be
3 Volkov, Balanchine, p. 153. sweet and derisive at the same time? Perhaps
4 Karl Fedorovich Valts, Shest’desyat pyat’ let v Tchaikovsky thought similarly, since the end
teatre (Sixty-five Years in the Theatre) result is a theme that is très douce and not
(Leningrad 1928), p. 108. Translation in Roland ricaneuse at all.
John Wiley, Tchaikovsky’s Ballets: Swan Lake, 16 Wiley, Tchaikovsky’s Ballets, p. 161.
Sleeping Beauty, Nutcracker (Oxford: Oxford 17 Ibid., p. 161.
University Press, 1985), p. 56. 18 Brown, Tchaikovsky, p. 339.
5 German Avgustovich Larosh (Laroche), 19 Wiley, Tchaikovsky’s Ballets, p. 228. Auguste
Izbrannye stat’i (Collected Articles), 5 vols. Mustel was the inventor of the celesta.
(Leningrad, 1974), vol. i, pp. ii, 99. Translated in: 20 Wiley, Tchaikovsky’s Ballets, p. 234.
Wiley, Tchaikovsky’s Ballets, pp. 52–3. Composer Mikhail Ippolitov-Ivanov had
6 “Sovremennye izvestiya” (25 February 1877), notated the Georgian lullaby and sent it to
p. 3. Translated in Wiley, Tchaikovsky’s Ballets, Tchaikovsky.
p. 52.
7 As Wiley points out, St Petersburg newspapers
commented on this similarity: Novosti i 15 The romantic ballet and its critics: dance
birzhevaya gazeta, 17 January 1895, p. 3; Novoe goes public
vremya 17 January 1895, p. 3. Tchaikovsky 1 See also Jeremy Noble’s article on Opera
earlier used the Swan Theme for a children’s play criticism in Grove Music Online. Among the
at his sister Alexandra Davydov’s estate in 1870. fictional accounts of life at the Opéra are Albéric
Cf. Wiley, Tchaikovsky’s Ballets, p. 37. Second, Les Petits Mystères de l’Opéra (Paris: G.
8 Wiley, Tchaikovsky’s Ballets, p. 85. Kugelman/Bernard Latte 1844); and Albert
Tchaikovsky originally composed this duet for Smith, The Natural History of the Ballet Girl
Undine and her mortal lover, who must die for (London: D. Bogue 1847).
having betrayed her, in his opera Undine (1869). 2 See Ivor Guest, The Romantic Ballet in Paris
In Swan Lake, the soprano and baritone parts are (London: Dance Books, 1980), p. 109.
played by violin and cello respectively. 3 Ivor Guest, “Introduction”, in Théophile
9 Wiley, Tchaikovsky’s Ballets, pp. 80–8. Wiley Gautier, Gautier on Dance, ed. Ivor Guest
provides a detailed analysis of the use of keys (London: Dance Books 1986), pp. xix–xxvi;
throughout the ballet and how they relate to the p. xxii.
characters. 4 Quoted in Marian Smith, “About the House”,
10 Ibid., p. 244. in Roger Parker and Mary Ann Smart (eds.),
11 Ibid., p. 104. Ivan Alexandrovich Reading Critics Reading: Opera and Ballet
Vsevolozhsky was Director of Imperial Theatres Criticism in France from the Revolution to 1848
1881–99. He had eliminated the post of a (Oxford: Oxford University Press, 2001),
resident ballet composer, making it easier to give p. 225.
commissions to composers such as Tchaikovsky. 5 Théophile Gautier, Histoire de l’art dramatique
12 Tchaikovsky’s response was written in en France depuis vingt-cinq ans, 6 vols. (Leipzig:
Moscow on 22 August 1888 – more than three Edition Hetzel, 1858–9) (Geneva: Slatkine
months after Vsevolozhsky’s initial enquiry (my Reprints 1968), vol. i, p. i.
trans.). Wiley speculates that the director’s 6 Gautier, Gautier on Dance, p. 58.
original letter may have been lost (Wiley, 7 Jules Janin, Journal des débats, 6 December
Tchaikovsky’s Ballets, p. 104). 1833.
13 Wiley, Tchaikovsky’s Ballets, p. 125. 8 Janin, Journal des débats, 27 June 1832.
14 Laroche. Izbranniye stat’i vol. ii (Leningrad, 9 Gautier, Histoire de l’art dramatique, vol. i,
1975), pp. 140–3, trans. David Brown, p. 127.
Tchaikovsky Remembered. (London: Faber & 10 Gautier, Gautier on Dance, p. 6.
Faber, 1993), p. 190. In addition to praising 11 Anonymous critic of the Revue et gazette
Tchaikovsky’s achievement withSleeping Beauty, musicale de Paris, 18 August 1839, on Fanny
Laroche’s review for theMoscow Gazette defends Elssler dancing the cachucha.
306 Notes to pages 181–225

12 Susan Bernstein, Virtuosity of the Vaganova style emphasised forcefulness and


Nineteenth Century: Performing Music and strength.
Language in Heine, Liszt, and Baudelaire 13 Dr Janice Bruckner, anthropologist, formerly
(Stanford, Calif.: Stanford University Press, Director of Research for the department of
1998), p. 11. physical therapy at Philadelphia’s Jefferson
13 Louis Gentil, in Jean-Louis Tamvaco (ed.), University, now at Widener University
Les Cancans de l’Opéra, 2 vols. (Paris : CNRS, Pennsylvania, is the inventor of this pointe shoe.
2000), vol. i, p. 338. She developed it together with Prosthetic
14 Charles Baudelaire, La Fanfarlo, in The Orthotic Solutions International. Cf. Marion
Poems in Prose, vol. ii, ed. and trans. Francis Kant, “The New Ballet Shoe”, in Ballettanz
Scarfe (London: Anvil Press, 1989), pp. 214–63; (Berlin: Friedrich Verlag, 2004), p. 74.
pp. 245–7.
17 Ballet avant-garde I: the Ballets Suédois
16 The soul of the shoe and its modernist concept
1 Théophile Gautier, in: Gautier on Dance, 1 Berlingske Tidende (Copenhagen), 12
selected, trans. and annotated Ivor Guest September 1922.
(London: Dance Books, 1986), p. 15. 2 Pall Mall Gazette, 21 October 1922.
2 Ibid., p. 140. 3 Francis Picabia, “Rolf de Maré”, La Danse
3 Cf. Judith Chazin-Bennahum on (November–December 1924), unpaginated.
Jean-Georges Noverre in Chapter 2. 4 Jean Cocteau, La Danse (June 1921),
4 Carlo Blasis, Traité élémentaire, théorique, et unpaginated.
pratique de l’art de la danse was published in 5 The Minneapolis Tribune, 22 May 1930.
1820. See Fig. 14. 6 New York Telegraph, 17 November 1923.
5 Cf. New Grove Dictionary of Music and 7 Dagens Nyheter (Stockholm), 26 May 1925.
Musicians, ed. Stanley Sadie and John Tyrrell, 8 Reynaldo Hahn in Excelsior, 10 November
2nd edn (London: Macmillan, 2001). Entry 1920.
Theobald Boehm. 9 The Spectator, 25 December 1920.
6 The scene was later immortalised in Edgar 10 Midi (Brussels), 3 June 1921.
Degas’s painting The Ballet Scene from 11 Die Weltbühne, 27 April 1922.
Meyerbeer’s Opera, “Robert le Diable”, 1876.
7 Heinrich Heine, Elementargeister (Berlin,
Paris: Säkularausgabe Akademieverlag and Edn. 18 Ballet avant-garde II: The ‘New’ Ballet –
du CNRS 1979), vol. ix, p. 95. Trans. Russian and Soviet dance in the twentieth
Marion Kant. century
8 First draft of libretto for Giselle. Cf. Théophile 1 Elizabeth Souritz, Soviet Choreographers in the
Gautier, “Théatres, Mystère, Comédies et 1920s, trans. Lynn Visson, ed. Sally Banes
Ballets”, in OEuvres complètes (Geneva: Slatkine (Durham, N.C.: Duke University Press, 1990),
Reprints, 1978), vol. viii, p. 267. p. 140.
9 Tutu, “childish alteration of cucu, dim. of cul –
rump or buttock”, term used since around 1910. 19 George Balanchine
Oxford English Dictionary (Oxford: Clarendon 1 Michael Walsh, “The Joy of Pure Movement”,
Press, New York: Oxford University Press, 1991), Time Magazine, 9 May, 1983.
entry “tutu”. Cf. the comprehensive overview of 2 Quoted in Bernard Taper, Balanchine: A
the evolution of the tutu in Martine Kahane et Biography (New York: Times Books, 1984), p. 51.
al., Le tutu (Paris: Flammarion, Les petits guides 3 Terry Teachout, All in the Dances: A Brief Life
de l’Opéra, 2000). of George Balanchine. (Orlando, Fla: Harcourt,
10 Théophile Gautier, Peau de Tigre. (Paris, 2004), p. 3.
1866), pp. 335–6, quoted in Judith 4 Charles Joseph, Stravinsky and Balanchine: A
Chazin-Bennahum, The Lure of Perfection, Journey of Invention (New Haven: Yale University
(London and New York: Routledge, 2005), Press, 2002), p. xi.
p. 194. 5 Edwin Denby, Dancers, Buildings and People in
11 When the tutu was shortened even further the Streets (New York: Horizon Press, 1965),
and nothing remained but the tulle sticking out p. 84.
at the hip, the tutu required a metal hoop to 6 Francis Mason, I Remember Balanchine (New
keep it in shape. York: Doubleday, 1991), p. vii.
12 This is one of the distinctions between the 7 See Stephanie Jordan, Moving Music:
Russian and the American school. George Dialogues with Music in Twentieth-Century Ballet
Balanchine preferred a gentle rise, whereas the (London: Dance Books, 2000), pp. 123 and 87.
307 Notes to pages 225–38

8 Balanchine picked music for its dansante 26 See Genné, “Glorifying the American
quality; he held steadfast opinions about what Woman”.
kind of music was danceable. 27 Solomon Volkov, Balanchine’s Tchaikovsky:
9 Quoted in Joseph, Stravinsky and Balanchine, Interviews with George Balanchine (New York:
p. 295. Originally from Igor Stravinsky’s Themes Sunon & Schuster, 1985), pp. 147–8.
and Conclusions (repr. Berkeley: University of 28 Anon., “Ballet Comes into its Own”, Musical
California Press, 1982). On the dispute over the America, 64 (April 1944), p. 16. The author cites
authenticity of this well-known quotation, see successful American-themed ballets such as
Joseph, Stravinsky and Balanchine, p. 411, n. 35. Rodeo, Billy the Kid, Filling Station and Fancy
Stravinsky is referring to Movements for Piano Free (by Jerome Robbins – soon Balanchine’s
and Orchestra. associate artistic director of NYCB).
10 For a detailed account from this time period, 29 Ronald Eyer, “‘Song of Norway’ Scores Hit”
see Yuri Slonimsky’s memoir in Mason, I (Review), Musical Times 64 (September 1944),
Remember Balanchine, pp. 19–78. p. 15.
11 Mason, I Remember Balanchine, p. 7. 30 Balanchine, New Complete Stories, p. 171.
Danilova and Balanchine were never legally 31 Jordan, Moving Music, pp. 112–13.
married but were seen as such within their circle 32 Teachout, All in the Dances, p. 87.
during the time they were together. 33 See Tim Scholl, From Petipa to Balanchine
12 Apollo was originally entitled Apollon (London: Routledge, 1994).
musagète. For its various titles, see Harvey 34 See Taper, Balanchine, pp. 187–8.
Simmonds (ed.), Choreography by George Balanchine’s choreography was to Gershwin’s
Balanchine: A Catalogue of Works (New York: American in Paris, which was dropped by
Eakins Press Foundation, 1983), p. 86. Adolph director Samuel Goldwyn due to his “too
Bolm was the first to choreograph to this score; experimental camera techniques” (Robert
his version made its debut 27 April 1928, and Gottlieb, George Balanchine, p. 92).
Balanchine’s followed June 12 of the same year. 35 Costas, Balanchine: Celebrating a Life in
13 Taper, Balanchine, p. 100. Dance, pp. 23–5.
14 See Beth Genné, “‘Glorifying the American 36 Ibid.
Woman’: Josephine Baker and George 37 Ibid.
Balanchine”, Discourses in Dance, 3/1 (2006), 38 Jordan, Moving Music, pp. 122–3.
pp. 29–65. 39 Jordan, Music Dances (video recording).
15 Ibid., p. 35. 40 Jordan, Moving Music, p. 114.
16 See ibid. 41 Quoted in Gottlieb, George Balanchine,
17 Apollo was not composed for Balanchine; p. 181.
Balanchine only commissioned four scores from 42 Ibid., p. 182.
the composer. 43 Richard Buckle in collaboration with John
18 Joseph, Stravinsky and Balanchine, p. 14. Taras, George Balanchine: Ballet Master (New
19 Robert Gottlieb, George Balanchine: The York: Random House, 1988) p. 269.
Ballet Maker (New York: Harper Collins, 2004),
p. 206. (From a reprint of an interview 20 Balanchine and the deconstruction of
published in Life magazine, 11 June, 1965.) classicism
20 See Simmonds, Choreography by Balanchine, 1 André Levinson, “Le Deuxième Spectacle des
p. 117. Listed as “Serenade in C for string Ballets Russes: Le Bal”, Comoedia (30 May 1929),
orchestra, Op. 48, 1880, first three movements; clipping in the collection of the Bibliothèque
arranged and reorchestrated by George Antheil”. Nationale, Archives Rondel (hereafter
The fourth movement (Tema Russo) was added abbreviated AR). All translations are mine
in 1940 and inserted before the Elegy. unless otherwise indicated.
21 See Costas (ed.), Balanchine: Celebrating a 2 Bernard Taper, Balanchine: A Biography
Life in Dance (Windsor, Conn.: Tide-mark Press, (Berkeley: University of California Press, 1996),
2003), p. 149. p. 292.
22 Balanchine, Balanchine’s New Complete 3 André Levinson, “Grandeur et décadence des
Stories of the Great Ballets, (ed.) Francis Mason ‘ballets russes’”, in La Danse d’aujourd’hui
(New York: Doubleday, 1968), p. 363. (Paris: Editions Duchartre et Van Buggenhoudt,
23 Balanchine, Balanchine’s New Complete 1929), pp. 9–10.
Stories, p. 365. 4 André Levinson, “Some Commonplaces on
24 Taper, Balanchine, p. 160. the Dance” (1922), in Joan Acocella and Lynn
25 Gottlieb, George Balanchine, pp. 167–8. A Garafola, (eds.), André Levinson on Dance:
memory of Barbara Horgan. Writings from Paris in the Twenties (Hanover,
308 Notes to pages 238–53

N. H.: University Press of New 25 André Levinson, “Apollon-Musagète”.


England/Wesleyan University Press, 1991), p. 30. 26 George Balanchine, “Vyshe vsyekh masterov”
5 “Grandeur”, pp. 20–1. (“The Greatest Master of them All”), in Marius
6 Ibid., p. 17; D. Sordet, Review of the Ballets Petipa: materialy, vospominaniia, stat’i (Marius
Russes (27 June 1928), AR. Petipa: Documents, Reminiscences, Essays), ed. A.
7 Alexandra Danilova, Choura: The Memoirs of Nekhendzi (Leningrad: Leningrad State Theatre
Alexandra Danilova (New York: Alfred A. Knopf, Museum, 1971), pp. 277–82. Trans. Irina
1986), pp. 102–3. See also M. Hunt, “The Klyagin.
Prodigal Son’s Russian Roots: Avant-Garde and
Icons”, Dance Chronicle 5/1 (1982), p. 33. 21 The Nutcracker: a cultural icon
8 Danilova, Choura, pp. 102–3. 1 Quoted in Roland John Wiley, The Life and
9 Levinson, “Le Deuxième Spectacle”; L. Laloy, Ballets of Lev Ivanov: Choreographer of The
“Théâtre Sarah-Bernhardt (Ballets Diaghilev): Nutcracker and “Swan Lake” (Oxford: Clarendon
Le Bal”, L’Ere nouvelle (30 May 1929), AR; W. Press, 1997), p. 140. Wiley offers many details of
George, “L’art à la scène: En marge des Ballets the 1892 Nutcracker premiere.
Russes”, Scène (5 June 1929), AR. 2 Jack Anderson, The Nutcracker Ballet (New
10 André Levinson, Les Visages de la danse York: Gallery Books, 1979), p. 82. Anderson was
(Paris: Bernard Grasset, 1933); in Acocella and the first to chronicle the many versions of the
Garafola, Levinson on Danse, caption ballet and suggest it had become an American
accompanying plate 11. tradition.
11 André Levinson, “La Chorégraphie et 3 Ibid., pp. 92–6. Anderson spells the Russian’s
l’interprétation”, Comoedia (29 May 1927), AR. name “Sergueeff ”.
12 Ibid. 4 Cobbett Steinberg, San Francisco Ballet: The
13 Danilova Choura, p. 87. First Fifty Years (San Francisco: The San
14 Levinson, “La Chorégraphie”. Francisco Ballet Association, 1983), pp. 63–4.
15 George Balanchine, Balanchine’s New 5 Discussion of the ways in which Balanchine
Complete Stories of the Great Ballets, ed. Francis incorporated Africanist influences in his version
Mason (Garden City, N.Y.: Doubleday, 1954), of neoclassical ballet can be found in Brenda
p. 10. Dixon Gottschild’s Digging the Africanist
16 On Balanchine and gender, see the analysis of Presence in American Performance: Dance and
Agon (1957) in Sally Banes, Dancing Women: Other Contexts. (Westport, Conn.: Greenwood
Female Bodies on Stage (London: Routledge, Press, 1996).
1998) as well as Ann Daly, “The Balanchine 6 Solomon Volkov, Balanchine’s Tchaikovsky:
Woman: Of Hummingbirds and Channel Interviews with George Balanchine. (New York:
Swimmers”, Drama Review 31/1 (Spring 1987), Simon & Schuster, 1985), pp. 192–3.
pp. 9–21. 7 This violin solo, which does not appear in the
17 Richard Buckle, Diaghilev (New York: original Nutcracker score, was cut from the
Atheneum, 1979), p. 470. composer’s Sleeping Beauty score (it is
18 See, for example, Tim Scholl, From Petipa to sometimes restored in current versions).
Balanchine: Classical Revival and the Tchaikovsky had used its melody again in The
Modernization of Ballet (London: Routledge, Nutcracker, so Balanchine’s bringing the two
1994), p. 97. together with this interpolation made musical
19 Souvenir programme from the December sense (Volkov, Balanchine’s Tchaikovsky, p. 188).
1928–January 1929 season at the Théâtre de 8 Cf. Tania Branigan. “Barbie sponsors
l’Opéra, Paris, AR. Nutcracker ballet”, Guardian, Monday, 10
20 P.-B. Gheusi, “Les Ballets Russes”, Figaro (19 September 2001.
June 1928), AR. These links are discussed in D. 9 Billy Strayhorn collaborated with Ellington on
Harris, “Balanchine: Working with Stravinsky”, the original adaptation, which only used
Ballet Review 10/2 (Summer 1982), p. 20. excerpts of the ballet. For the Bird Nutcracker,
21 B. de Schoelzer, “Chronique Musicale: Les composer and musical director David Berger
Ballets Russes de 1928”, Nouvelle revue française added to the score.
(1 July 1928), AR. 10 Edwin Denby, Dance Writings, ed. Robert
22 Louis Laloy, “‘Apollon musagète,’ ballet de Cornfield and William MacKay (New York:
Stravinsky”, Ere nouvelle (17 June 1928), AR. Knopf, 1986), pp. 272–5.
23 André Levinson, “Apollon-Musagète”, 11 Ibid., pp. 272–3.
Candide (21 June 1928), AR. 12 These dances are discussed at more length in
24 André Levinson, “Une escale des ‘Ballets Jennifer Fisher, “Arabian Coffee in the Land of
Russes’”, Candide (3 January 1929), AR. the Sweets”, Dance Research Journal 35/2 and
309 Notes to pages 253–68

36/1 (Winter 2003 and Summer 2004), 10 Lin Cunxin, Mao’s Last Dancer: A Memoir
pp. 146–63. (New York: G. P. Putnam’s Sons, 2003).
13 Edward Said, Orientalism (London:
Routledge & Kegan Paul, 1978), p. 1. 23 Giselle in a Cuban accent
14 John M. MacKenzie, Orientalism: History, 1 See A. Alonso, “Bailar ha sido mi vida”,
Theory and the Arts (Manchester and New York: interview by Lester Tome, in El Mercurio, 19
Manchester University Press, 1995), p. 211. November, 2000 (Santiago, Chile), p. E24.
2 See A. Alonso, “Primeros recuerdos, primeros
22 From Swan Lake to Red Girl’s Regiment: pasos en la danza”, originally issued in T.
ballet’s sinicisation Gutiérrez, Alicia Alonso prima ballerina assoluta,
1 Marcel Granet, Danses et legends de la Chine imagen de una plenitud (testimonios y recuerdos
ancienne, 2 vols. (Paris: Presses Universitaires de de la artista) (Barcelona: Edición Salvat, 1981),
France, 1959). See also E. Yuan, Zhongguo repr. in A. Alonso, Diálogos con la danza, ed. P.
Wudao Yixianglun (Beijing: Wenhua Yishu, Simón (La Habana: Letras Cubanas, 1986), p. 13.
1995). 3 See Alonso, “Bailar ha sido mi vida”, p. E24.
2 Li Yu, Xianqing Ouji, 16 vols. (Shanghai: 4 See E. Denby, “Youth and Old Giselle”, in New
Shanghai Guji, 1995), vol. vii, pp. 593–4. York Herald Tribune, 24 October, 1945, repr. In
3 Their love story and history has been the E. Denby, Looking at the Dance (New York:
subject of many writers, poets and historians. Horizon Press, 1968), p. 167.
4 Wang Kefen and Long Yingpei, Zhongguo 5 See W. Terry, “The Ballet Theatre”, in New
Jinxiandai Dangdai Wudao Fazhangshi (Beijing: York Herald Tribune, 15 April, 1955, repr. in W.
Renmin Yingyu Chubanshe, 1999), pp. 40–3; Terry, I Was There: Selected Dance Reviews and
Wang Ningning, Jiang Dong and Du Xiaoqing, Articles, 1936–1976 (New York: Dekker, 1978),
Zhongguo Wudaoshi (Beijing: Wenhua Yishu, p. 295.
1998), pp. 69–71. See also Ou Jian-ping, “From 6 See A. Alonso, “Bailar ha sido mi vida”, p. E24.
‘Beasts’ to ‘Flowers’: Modern Dance in China”, in 7 Olivier Merlin’s review (Le Monde, 1966) is
Ruth and John Solomon (eds.), East Meets West quoted in the booklet XLV Aniversario de Alicia
in Dance: Voices in the Cross-Cultural Dialogue Alonso en el personaje de Giselle (Havana: Gran
(Chur, Switzerland: Harwood Academic Teatro de la Habana, 1988), no pages.
Publishers, 1995), pp. 29–35. 8 A facsimile of Anton Dolin’s letter, written in
5 Ronglin continued to promote ballet during Montreal on 24 and 26 June 1967, is reproduced
the nationalist and communist eras. She died in in XLV Aniversario de Alicia Alonso en el
Beijing in 1973. She left a memoir Qinggong Suiji personaje de Giselle, no pages.
or “The Miscellaneous of the Qing Court”. 9 See A. Alonso, “El arte no tiene patria, pero el
6 The illustrious Dianshizhai Huabao is one of artista sı́”, transcript of an interview by Raúl
them. Rivero, in Alonso, Diálogos con la danza, p. 177.
7 Mei Lanfang was the most famous Peking 10 Arnold Haskell is quoted by Pedro Simón in
opera singer from the late Qing through the A. Alonso and P. Simón, “Fuentes y antecedentes
communist era. He was the founder of a new de la escuela cubana de ballet” (Sources and
style of singing. He held numerous positions antecedents of the Cuban School of ballet),
during the communist era. Tian Han was a transcript of a lecture offered by Simón and
famous writer and poet. He wrote more than Alonso at the offices of the Union of Cuban
one hundred plays, many songs/poems, and the Journalist on 10 November 1978, in Alonso,
lyric for the National Anthem. Diálogos con la danza, pp. 44–5.
8 Wang Kefen and Long Yingpei, Zhongguo 11 See Alonso and Simón, “Fuentes y
Jinxiandai Dangdai Wudao Fazhangshi (Beijing: antecedentes de la escuela cubana de ballet”,
Renmin Yingyu Chubanshe, 1999), pp. 64–72; p. 25.
Wang Ningning, Jiang Dong and Du Xiaoqing, 12 See Alonso, “Bailar ha sido mi vida”, p. E24.
Zhongguo Wudaoshi (Beijing: Wenhua Yishu, 13 Alonso is quoted by Walter Terry in W. Terry,
1998), pp. 83–7. Ou Jian-ping, “From ‘Beasts’ to Alicia and her Ballet Nacional de Cuba (Garden
‘Flowers’: Modern Dance in China”, pp. 29–35. City, N.Y.: Anchor Books, 1981), pp. 34–5.
9 Wang Kefen and Long Yingpei, Zhongguo 14 See the prologue to XLV Aniversario de
Jinxiandai Dangdai Wudao Fazhangshi (Beijing: Alicia Alonso en el personaje de Giselle, no pages.
Renmin Yingyu Chubanshe, 1999), pp.72–80; The author of the prologue is unknown. Its
Wang Ningning, Jiang Dong and Du Xiaoqing, content was probably sanctioned by Alonso as
Zhongguo Wudaoshi (Beijing: Wenhua Yishu, director of the National Ballet of Cuba and the
1998), pp. 88–92. Ou Jian-ping, “From ‘Beasts’ Havana Grand Theatre, which issued the
to ‘Flowers’: Modern Dance in China”, pp. 29–35. publication.
310 Notes to pages 268–88

15 The programme is reproduced in F. Rey century” and chapter 17 by Erik Näslund, “The
Alfonso and P. Simón, Alicia Alonso: Órbita de Ballets Suédois and its modernist concept” in
una Leyenda (Madrid: SGAE, 1996), pp. 190–1. this volume.
16 See Alonso and Simón, “Fuentes y 4 Cf. Chapter 9 by Inge Baxmann “The
antecedentes de la escuela cubana de ballet”, French Revolution and its spectacles” in this
p. 35. volume.
17 Sociedad Pro-Arte Musical, founded in 1919 5 F. T. Marinetti, Teoria e invenzione futurista,
in Havana. Alonso received her first ballet classes ed. Luciano De Maria (Milan: Arnaldo
in Pro-Arte’s studios. Later in her career, the Mondadori Editore, 1983), pp. 144–52. trans.
Ballet Alicia Alonso appeared at the Auditorium Jonathan Steinberg.
Theatre as part of Pro-Arte’s series. 6 Cf. Agrippina Vaganova, Basic Principles of
18 See Alonso and Simón, “Fuentes y Classical Ballet, trans. Anatak Chujoy (1953;
antecedentes de la escuela cubana de ballet”, New York: Dover Publications 1969); the book is
p. 55. Italics added. still in print and widely used.
19 See Alfonso and Simón, Alicia Alonso: Órbita 7 The Ballets Russes de Monte Carlo were
de una Leyenda, pp. 135–4. founded in 1932, as a fusion of the Ballets de
20 In 1978 and 1979, the company danced l’Opéra de Monte Carlo and the Ballet de
Giselle, Coppélia, Grand pas de quatre and Les l’Opéra Russe à Paris, with Colonel de Basil as
Sylphides. The programme for these two initial director and René Blum as artistic director. The
tours also included pieces by Cuban Ballets de Monte Carlo, founded by René Blum
choreographers, but their choreography did not at Monte Carlo in 1936 were the result of a row
receive positive criticism. For later visits the between him and Colonel de Basil. Léonide
group therefore focused on its European Massine took over as artistic director in 1938
repertory. Recent tours have featured the and the company’s name was changed to Ballet
company in a production of Cinderella by Russe de Monte Carlo. In 1985 Les Ballets de
Cuban-French choreographer Pedro Consuegra Monte Carlo saw a brief revival. The Ballet Russe
(1998), Giselle (1999), excerpts from Don de Monte Carlo, a descendant of René Blum’s
Quixote, Nutcracker, Sleeping Beauty, Giselle, Ballets de Monte Carlo opened in 1938 and
Coppélia and Swan Lake (2001), Don Quixote lasted into the 1950s.
(2003) and Swan Lake, Azari Plissetski’s Canto 8 Théophile Gautier, Mademoiselle Maupin
Vital and Antonio Gades’s Blood Wedding (Paris: 1834), p. 22.
(2004). 9 According to ballet critic Clement Crisp, the
future is bleak: “We have stopped being a nation
24 European ballet in the age of ideologies of ballet choreographers, we have stopped being
1 Isadora Duncan, My Life (New York: H. a nation of interesting ballet-dancers, and we
Liveright, 1928), p. 32. have stopped being a nation who go to ballet to
2 Rudolf von Laban, Modern Educational Dance enjoy ourselves.” Interview with Ismene Brown,
(London: Macdonald & Evans 1948), p. 5 Dance Magazine (December 2001).
3 Cf. Chapter 18 by Tim Scholl, “The ‘New’ http://www.dance.co.uk/magazines/yr 01/
ballet: Russian and Soviet dance in the twentieth dec01/ismene b int clement c.htm
Bibliography and further reading

Part I
Dance treatises
Arbeau, Thoinot, Orchésographie. Langres, 1588. Facsimile edition, Geneva:
Minkoff, 1972, English trans. Mary Steward Evans, 1948. Reprinted with
introduction and notes by Julia Sutton, New York: Dover Publications, 1967.
Arena, Antonius, Ad suos compagniones studiantes. c.1529. English trans. in John
Guthrie and Marino Zorzi, “Rules of Dancing: Antonius Arena”, Dance Research
4/2 (1986).
Brussels, Bibliothèque Royale 9085. Le Manuscrit dit des Basses Danses de la
Bibliothèque de Bourgogne. Facsimile edition Graz: Akademische Druck-u.
Verlagsanstalt, 1988. Transcript and edition in David Wilson and Véronique
Daniels, “The Basse Dance Handbook”, forthcoming.
Caroso, Fabritio, Il Ballarino. Venice, 1581. Facs. edn New York: Broude Brothers,
1967.
—, Nobiltà di Dame. Venice, 1600. Facs. edn Bologna: Forni, 1980. English: ed and
trans. Julia Sutton, Courtly Dance of the Renaissance: A New Translation and
Edition of the Nobiltà di Dame 1600. Fabritio Caroso. New York: Dover, 1995.
Cornazano, Antonio, Libro dell’arte del danzare. Rome, Biblioteca Apostolica
Vaticana, Codex Capponiano, 203. English: trans. Madeleine Inglehearn and
Peggy Forsyth, The Book on the Art of Dancing: Antonio Cornazano. London:
Dance Books, 1981.
Domenio da Piacenza, De arte saltandj & choreas ducendj De la arte di ballare et
danzare. Paris, Bibliothèque Nationale, MS fonds it. 972. English trans. of
choreographies David Wilson, 101 Italian Dances c. 1450–c. 1550: A Critical
Translation. Cambridge: Early Dance Circle, 1999.
Gresley of Drakelow papers. Derbyshire Record Office, D77 box 38, pp. 51–79. A
transcription of this material is found in David Fallows, “The Gresley Dance
Collection, c . 1500”, RMA Research Chronicle 29 (1996).
Guglielmo Ebreo da Pesaro, Guilielmi Hebraei pisauriensis de pratica seu arte tripudii
vulgare opusculum incipit. 1463. Paris, Bibliothèque Nationale, MS fonds it. 973.
English: ed. and trans. Barbara Sparti, De pratica seu arte tripudii: On the
Practice or Art of Dancing. Oxford: Clarendon Press, 1993.
Negri, Cesare, Le Gratie d’Amore. Milan, 1602. Facs. edn New York: Broude
Brothers, 1969. English: G. Yvonne Kendall, “Le Gratie d’Amore 1602 by Cesare
Negri: Translation and Commentary”. D.M.A. Stanford University, 1985.
Santucci, Ercole, Mastro da ballo, 1614. Facs. edn, Hildesheim: Olms, 2004, with
intro. by Barbara Sparti.
S’ensuit l’art et instruction de bien dancer. Paris, Michel Toulouze, n.d. Facs. edn,
Geneva: Minkoff, 1985.

[311]
312 Bibliography and further reading

Renaissance
Arcangeli, Alessandro, Davide o Salomè? Il dibattito europeo sulla danza nella prima
età moderna.Rome: Viella, 2000.
Brainard, Ingrid, The Art of Courtly Dancing in the Early Renaissance. West Newton,
Mass.: Private Pub. 1989.
Carter, Françoise, “Number Symbolism and Renaissance Choreography”. Dance
Research 10/1 (1992), pp. 21–39.
Fermor, Sharon, “Studies in the Depiction of the Moving Figure in Italian
Renaissance Art, Art Criticism, and Dance Theory”. Ph.D. diss., Warburg
Institute, University of London, 1990.
—, “Movement and Gender in Sixteenth-Century Italian Painting”. In Kathleen
Adler and Marcia Pointon (eds.), The Body Imaged: The Human Form and Visual
Cultures since the Renaissance. Cambridge: Cambridge University Press, 1993,
pp. 129–45.
Heartz, Daniel, “The Basse Danse: Its Evolution circa 1450 to 1550”. Annales
musicologiques 6 (1958–63), pp. 287–340.
Jones, Pamela, “Spectacle in Milan: Cesare Negri’s Torch Dances”. Early Music 14/2
(1986), pp. 182–96.
Kendall, G. Yvonne, “Dance, Music and Theatre in Late Cinquecento Milan”. Early
Music 32/1 (2004), pp. 75–95.
McGinnis, Katherine Tucker, “Moving in High Circles: Courts, Dance, and Dancing
Masters in Italy in the Long Sixteenth Century”. Ph.D. diss., University of North
Carolina, 2001.
Marrocco, W. Thomas, Inventory of 15th-Century Bassedanze, Balli, and Balletti.
New York: CORD, 1981.
Nevile, Jennifer, “Cavalieri’s Theatrical Ballo: ‘O che nuovo miracolo’: A
Reconstruction”. Dance Chronicle 21/3 (1998), pp. 353–88.
—, “Cavalieri’s Theatrical Ballo and the Social Dances of Caroso and Negri”. Dance
Chronicle 22/1 (1999), pp. 119–33.
—, “Dance and the Garden: Moving and Static Choreography in Renaissance
Europe”. Renaissance Quarterly 52/3 (1999), pp. 805–36.
—, The Eloquent Body: Dance and Humanist Culture in Fifteenth-Century Italy.
Bloomington: Indiana University Press, 2004.
Padovan, Maurizio (ed.), Guglielmo Ebreo da Pesaro e la danza nelle corti italiane del
xv secolo. Pisa: Pacini, 1990.
Sparti, Barbara, “Rôti Bouilli: Take Two ‘el gioioso fiorito’ ”. Studi Musicali 24/2
(1995), pp. 231–61.
—, “Would You Like to Dance This Frottola? Choreographic Concordances in Two
Early Sixteenth-Century Tuscan Sources”. Musica disciplina 50 (1996),
pp. 135–65.
Sutton, Julia, “Musical Forms and Dance Forms in the Dance Manuals of Sixteenth
Century Italy: Plato and the Varieties of Variation”. In The Marriage of Music
and Dance. Papers from a conference at the Guildhall School of Music and
Drama, London 1991. Cambridge: National Early Music Association,
1992.
313 Bibliography and further reading

Ballet de cour and baroque


Anthony, James R., French Baroque Music from Beaujoyeulx to Rameau. New York
and London: Norton, 1974; rev. edn 1981.
Apostolidès, Jean-Marie, Le Roi-machine: Spectacle et politique au temps de Louis
XIV. Paris: Editions de minuit, 1981.
Auld, Louis E., “Social Diversity in the Ballet de Cour: Le Château de Bicëtre”. In A.
Maynor Hardee (ed.), Theater and Society in French Literature. Columbia, S.C.,
1988.
Bouquet-Boyer, Marie-Thérèse (ed.), Les Noces de Pélée et de Thétis. Venise,
1639 – Paris, 1654, Actes du colloque international de Chambéry et de Turin,
3–7 novembre 1999, Bern, New York, Oxford, and Vienna: Peter Lang,
2001.
Buch, David J., Dance Music from the Ballets de Cour, 1575–1651: Historical
Commentary, Source Study, and Transcriptions from the Philidor Manuscripts.
New York: Stuyvesant, 1994.
Christout, Marie Françoise, Le Ballet de cour au XVII˚ siècle. Geneva: Minkoff, 1987.
—, Le Ballet de cour de Louis XIV. 1643–1672. Paris: A. & J. Picard, 1967; new edn
Paris: Centre national de la danse and Picard, 2005.
Coeyman, Barbara, “Theatres for Opera and Ballet during the Reigns of Louis XIV
and Louis XV”. Early Music (18 February 1990), pp. 22–37.
Debord, Guy, Society of the Spectacle, trans. Donald Nicholson-Smith. New York:
Zone Books, 1994.
Feuillet, Raoul Auger, Chorégraphie ou l’art de décrire la danse, par caractères, figures
et signes démonstratifs. Paris: Feuillet & Brunel, 1700.
Franko, Mark , Dance as Text. Ideologies of the Baroque Body. Cambridge:
Cambridge University Press, 1993.
Gorce, Jerôme de la,“Un aspetto del mestiere teatrale di Torelli: la riutilizzazione
delle scenografie dell’ ‘Andromède’ per il ‘Ballet de la nuit”. In Francesco Milesi
(ed.), Giacomo Torelli. L’invenzione scenica nell’Europa barocca. Fano:
Fondazione Cassa di Risparmio di Fano, 2000.
Harris-Warrick, Rebecca, and Marsh, Carol G., Musical Theatre at the Court of Louis
XIV: Le Mariage de la Grosse Cathos. Cambridge: Cambridge University Press,
1994.
Haskell, Harry, The Early Music Revival: A History. London: Thames & Hudson,
1988.
Hilton, Wendy, Dance of Court and Theater: The French Noble Style, 1690–1725.
Princeton: Princeton Book Co., 1981.
Hourcade, Philippe, Mascarades et ballets du Grand Siècle, 1643-1715. Paris:
Desjonquières et Centre National de la Danse, 2002.
Lacroix, Paul, Ballet, et mascarades de cour sous Henri IV et Louis XIII (de 1581 à
1652). Geneva: J. Gay & fils, 1868–70, repr. Geneva: Slatkine, 1968.
Laurenti, Jean Noel, “Feuillet’s Thinking”. In Traces of Dance. Drawing and Notations
of Choreographers, ed. Laurence Louppe. Paris: Editions Dis Voir, n.d.
De Lauze, François, Apologie de la danse et parfaite méthode de l’enseigner tant aux
Cavaliers qu’aux dames. n.p., 1623.
314 Bibliography and further reading

Lippe, Rudolf zur, Naturbeherrschung am Menschen. Frankfurt/M: Syndikat Reprise,


1979.
McGowan, Margaret M., L’Art du ballet de cour en France 1581–1643. Paris: Centre
National de la Recherche Scientifique, 1965, repr. 1978.
Ménestrier, Claude-François, Des représentations en musique anciennes et modernes.
Paris: René Guignard, 1687; repr. Geneva: Minkoff, 1992.
—, Des ballets anciens et modernes selon les règles du théâtre. Paris: René Guignard,
1682; repr. Geneva: Minkoff, 1972.
Orgel, Stephen, The Illusion of Power. Political Theater in the English Renaissance.
Berkeley and Los Angeles: University of California Press, 1975.
Prunières, Henry, Le Ballet de cour en France avant Benserade et Lully. Paris: Henri
Laurens, 1914; New York and London: Johnson Reprint, 1970.
Pure, Michel de, Idée des spectacles anciens et nouveaux. Paris 1668; facs. edn
Geneva: Minkoff, 1972.
Rameau, Pierre, Le Maı̂tre à danser. Paris: Jean Villette, 1725. English: The
Dancing-Master, trans. John Essex. London: J. Essex & J. Brotherton, 1728.
Rice, Paul F., The Performing Arts at Fontainebleau from Louis XIV to Louis XVI.
Rochester: University of Rochester Press, 1989.
Saint-Hubert, M. de, La Manière de composer et de faire réussir les ballets. Paris:
Targa, 1641; repr. Geneva: Minkoff, 1993.
Schlemmer, Oskar, The Letters and Diaries of Oskar Schlemmer. Middletown, Conn.:
Wesleyan University Press, 1972.
Schwartz, Judith L., and Schlundt, Christena L., French Court Dance and Dance
Music: A Guide to Primary Source Writings, 1643–1789. Stuyvesant, N.Y.:
Pendragon Press, 1987.
Sparti, Barbara, “Dance and Historiography. Le Balet Comique de la Royne: An
Italian Perspective”, in Ann Buckley and Cynthis Cyrus (eds.), Festschrift for
Ingrid Brainard (forthcoming).
Taruskin, Richard, Text and Act: Essays on Music and Performance. New York:
Oxford University Press, 1995.
Tomko, Linda, “Reconstruction”, in International Encyclopedia of Dance. New York:
Oxford University Press, 1998.
Vallière, Louis César de la Baume le Blanc, duc de la, Ballets, opéra, et autres
ouvrages lyriques, par ordre chronologique depuis leur origine. Paris, 1760.

For an exhaustive bibliography on ballet de cour cf. Bibliographie de l’air de cour


compiled by Georgie Durosoir for the Centre de Musique Baroque de Versailles and
available online at http://www.cmbv.com/images/banq/cp/cp014.pdf http://
www.cmbv.com/images/banq/cp/cp014.pdf  C Georgie Durosoir, CMBV, dépôt

légal: novembre 2003.

On the Balet comique de la Royne


“Balthasar de Beaujoyeulx, Ballet Comique de la Reine Paris, 1582”. In Selma
Jeanne Cohen (ed.), Dance as a Theatre Art: Source Readings in Dance History
from 1581 to the Present. New York: Harper & Row, 1974, pp. 19–31.
Bonniffet, Pierre, ”Esquisses du ballet humaniste 1572–1581”. In Le Ballet aux XVIe
et XVIIe siècles en France et à la cour de Savoie. Geneva: Editions Slatkine,
Cahiers de l’I.R.H.M.E.S., 1992, pp. 16–49.
315 Bibliography and further reading

Dellaborra, Mariateresa (ed.), “Une Invention Moderne”: Baldassarre da Belgioioso e


il “Balet Comique de la Royne”. Lucca: Libreria Musicale Italiana, 1999.
Loret, Jean, La Muze historique 1650–1665. 4 vols., ed. C.-L. Livet. Paris, 1857–1878.
MacClintock, Carol and Lander (eds.), Le Balet Comique de la Royne 1581. New
York: American Institute of Musicology, Musicological Studies and Documents
25, 1971.
McGowan, Margaret (ed.), Le Balet Comique by Balthazar de Beaujoyeulx.
Binghamton, N.Y.: Center for Medieval and Early Renaissance Studies, 1982.
Prunières, Henry, Le Ballet de cour en France avant Benserade et Lully. Paris: Henri
Laurens, 1914 ; New York and London: Johnson Reprint, 1970.
Ravelhofer, Barbara (ed.), Terpsichore 1450-1900, Proceedings of the International
Dance Conference. Ghent, Belgium: The Institute for Historical Dance Practice,
2000.
Woodruff, Diane L., “The ‘Balet Comique’ in the Petit Bourbon: A Practical View”,
Proceedings of the Society of Dance History Scholars. Riverside: University of
California Press, 1986.
Yates, Frances A., The French Academies of the 16th Century. London: The Warburg
Institute, 1947; Nendeln, Liechtenstein: Kraus Reprint, 1968.

English masques
Sources
Ashbee, A., and Lasocki, D., et al., A Biographical Dictionary of English Court
Musicians, 1485–1714, 2 vols. Aldershot: Ashgate, 1998.
Court Masques: Jacobean and Caroline Entertainments, 1605–1640, ed. David
Lindley. Oxford: Oxford University Press, 1995.
Keller, Robert, database, The English Dancing Master: An Illustrated Compendium,
1651–1728, at www.izaak.unh.edu/nhltmd/indexes/dancingmaster/.
McGee, C. E., and Meagher, John, checklists on Tudor and Stuart entertainments,
in Research Opportunities in Renaissance Drama, since 1982.
McManus, Clare, Women on the Renaissance Stage: Anna of Denmark and Female
Masquing in the Stuart Court 1590–1619. Manchester: Manchester University
Press, 2002.
Records of English Court Music, ed. A. Ashbee, 9 vols. Snodland, Aldershot:
Ashbee/Scolar Press, 1986–96.
Tomlinson, Sophie, Women on Stage in Stuart Drama. Cambridge: Cambridge
University Press, 2005.

General
Britland, Karen, Drama at the Courts of Queen Henrietta Maria. Cambridge:
Cambridge University Press, 2006.
Daye, Anne, “Torchbearers in the English Masque”, Early Music 26/2 (1998).
Holman, Peter, Four and Twenty Fiddlers: The Violin at the English Court, 1540–1690.
Oxford: Clarendon Press, 1993.
Howard, Skiles, The Politics of Courtly Dancing in Early Modern England. Amherst:
University of Massachusetts Press, 1998.
Inigo Jones: The Theatre of the Stuart Court, ed. S. Orgel and R. Strong, 2 vols.
London: Sotheby Parke Bernet, 1973.
316 Bibliography and further reading

Peacock, John, The Stage Designs of Inigo Jones: The European Context. Cambridge:
Cambridge University Press, 1995.
Ravelhofer, Barbara, The Early Stuart Masque: Dance, Costume, and Music. Oxford:
Oxford University Press, 2006.
Walls, Peter, Music in the English Courtly Masque, 1604–1640. Oxford: Clarendon
Press, 1996.

Part II
Adice, G. Léopold, Théorie de la gymnastique de la danse théâtrale. Paris: Chais, 1859.
Angiolini, Gasparo, Dissertation sur les Ballets Pantomimes des anciens. Vienne: Jean
Th. de Trattnern,1765.
Astier, Régine, “In Search of L’Académie Royale de Danse”. York Dance Review 7
(1978).
—, “La Vie quotidienne des danseurs sous l’Ancien Régime”. Les Gôuts reunis, 3rd
series (1982).
Baxmann, Inge, Die Feste der Französischen Revolution. Inszenierung von Gesellschaft
als Natur. Weinheim/Basel: Beltz 1989.
Blasis, Carlo, The Code of Terpsichore, trans. R. Barton. London: James Bullock,
1828.
—, Notes Upon Dancing, Historical and Practical, trans. R. Barton. London: M.
Delaporte, 1847.
—, Traité élémentaire, théorique et pratique de l’art de la danse. Milan: Joseph Beati et
Antoine Tenenti, 1820. English: An Elementary Treatise upon the Theory and
Practice of the Art of Dancing, trans. Mary Stewart Evans. New York: Dover,
1968.
Bonin, Louis, Die neueste Art zur galanten und theatralischen Tantz-Kunst. Frankfurt:
Joh. Christoff Lochner, 1712.
Beaumont, Cyril W., Three French Dancers of the 18th Century: Camargo, Sallé,
Guimard. London: C. W. Beaumont, 1934.
Cahusac, Louis de, La Danse ancienne et moderne ou Traité historique de la Danse. 3
vols. Paris: Jean Neaulme, 1754.
Campardon, Emile, L’Académie Royale de Musique au XVIIIe siècle. Paris:
Berger-Levrault & cie, 1884.
Carones, Laura, “Noverre and Angiolini: Polemical Letters”. In Dance Research 5/I,
New York: MacMillan, 1987, pp. 42ff.
Castil-Blaze, François Henri Joseph, L’Académie Impériale de Musique. Histoire
littéraire, musicale, chorégraphique, pittoresque, morale, critique, politique et
galante de ce théâtre de 1645 à 1855. 2 vols. Paris: Castil-Blaze, 1855.
Censer, Jack Richard and Hunt, Lynn, Liberty, Equality, Fraternity: Exploring the
French Revolution. University Park, Pa.: Pennsylvania State University Press,
c.2001.
Chants Révolutionnaires et patriotiques à l’usage des fêtes nationales et décadaires,
Paris 1790.
Chapman, John V., “Auguste Vestris and the Expansion of Technique”. Dance
Research Journal 19/1 (1987).
—, “The Paris Opera Ballet School, 1798–1827”. Dance Chronicle 12/2 (1989).
317 Bibliography and further reading

—, “August Vestris”. International Dictionary of Ballet. vol. ii, 1993.


Chazin-Bennahum, Judith, “Cahusac, Diderot, and Noverre: Three Revolutionary
French Writers on Eighteenth-Century Dance”. Theatre Journal (May 1983),
pp. 169–78.
—, Dance in the Shadow of the Guillotine. Carbondale and Edwardsville: Southern
Illinois University Press, 1988.
—, The Lure of Perfection: Fashion and Ballet 1780–1830. New York: Routledge,
2004.
Chéruzel, Maurice, Jean-Georges Noverre: Levain de la Danse Moderne. Conférence
avec Projections. Paris: Saint-Germain-en-Laye, 1991.
Cohen, Sarah, Art, Dance and the Body in French Culture of the Ancien Régime.
Cambridge: Cambridge University Press, 2000.
Craine, Debra, and Mackrell, Judith (eds.), Oxford Dictionary of Dance. Oxford:
Oxford University Press, 2000.
Didelot, Charles Louis, Three King’s Theatre Ballets, 1796–1801. London,
1796–1801; intro. Roland John Wiley, London: Stainer & Bell, 1994.
Diderot, Denis, Troisième Entretien sur Le Fils Naturel, Oeuvres Complètes, ed. J.
Assézat and M. Tourneux. Paris: Garnier, 1875–7.
—, Rameau’s Nephew, trans. L. W. Tancock. Harmondsworth, Midd: Penguin,
1966.
Dufort, Giambattista, Trattato del ballo nobile. Naples: Felice Mosca, 1728.
Editions d’Histoire Sociale (ed.), Jean-Jacques Rousseau dans la Révolution Francaise.
Paris, 1977.
Fairfax, Edmund, The Styles of Eighteenth-Century Ballet. Lanham, Md.: Scarecrow
Press, 2003.
Feuillet, Raoul Auger, Chorégraphie ou l’art de décrire la danse, par caractères, figures
et signes démonstratifs. Paris: Feuillet & Brunel, 1700.
Franko, Mark, Dance as Text. Ideologies of the Baroque Body. Cambridge: Cambridge
University Press, 1993.
Gallini, Giovanni-Andrea, A Treatise on the Art of Dancing. London: The author,
1762.
Gravelot et Cochin, Iconologie ou Traité de la science des allégories à l’usage des
artistes. vol. i, Paris, 1791.
Guest, Ivor F., “Jason and Medea: A Noverre Ballet Reconstructed”. Dancing Times.
London (May 1992).
—, The Ballet of the Enlightenment: The Establishment of the Ballet d’Action in France,
1770–1793. London: Dance Books, 1996.
—, Ballet under Napoleon. Alton: Dance Books, 2002.
Habermas, Jürgen, The Structural Transformation of the Public Sphere: An Inquiry
into a Category of Bourgeois Society, trans. Thomas Burger, with Frederick
Lawrence. Cambridge, Mass.: MIT Press, 1989.
Halifax, George Savile, Marquess of, Advice to a Daughter, chiefly with Regard to
Religion. Aberdeen: printed for and by Francis Douglass and William Murray in
1688.
Hammond, Sandra Noll, “Searching for the Sylph: Documentation of Early
Developments in Pointe Technique”. Dance Research Journal 19/2 (1987–8).
318 Bibliography and further reading

Hammond, Sandra N., and Hammond, Phillip E., “Technique and Autonomy in
the Development of Art: A Case Study in Ballet”. Dance Research Journal 21/2
(1989).
Harris-Warrick, Rebecca and Brown, Bruce Alan (eds.), The Grotesque Dancer on
the Eighteenth-Century Stage: Gennaro Magri and his World. Madison: University
of Wisconsin Press, 2004.
Hunt, Lynn Avery, Politics, Culture, and Class in the French Revolution. Berkeley :
University of California Press, 1984.
Isaac, Mr., A Collection of Ball-Dances Perform’d at Court: viz. The Richmond, the
Rondeau, the Rigadoon, the Favorite, the Spanheim, and the Britannia.1706.
Jenyns, Soame, Poems by Soame Jenyns, containing Art of dancing, To Lord Lovelace,
Essay on virtue, Written in Locke, Epitaph on Doctor Johnson; to which is prefixed a
sketch of the author’s life. Manchester, 1797.
Kirstein, Lincoln, Movement and Metaphor: Four Centuries of Ballet. New York:
Praeger, 1970.
Koegler, Horst, “The Northern Heirs of Noverre”. Dance and Dancers (London)
(February 1987).
Krüger, Manfred, Jean-Georges Noverre und das Ballet d’Action. Emsdetten: Lechte,
1963.
La Caffinière, Alex and Kanter, Katherine, “Why is Noverre not in the Curriculum?”
Dance Now (London) 3/1 (Spring 1994).
Le Brun, Charles, Méthode pour apprendre à dessiner les passions, Amsterdam,
1702.
Lynham, Deryck, The Chevalier Noverre: Father of Modern Ballet. London, Sylvan
Press, 1950; London: Dance Books, 1972.
McLeave, Sarah, “Marie Sallé: A Letter to the Duchess of Richmond”. Dance
Research 17/1 (Summer 1999), pp. 22–46.
Magri, Gennaro, Trattato teorico-prattico di ballo. Naples: Vicenzo Orsino, 1779.
English: Theoretical and Practical Treatise on Dancing, trans. Mary Skeaping.
London: Dance Books, 1988.
Malpied, Traité sur l’art de la danse. Paris: Bouı̈n, 1770.
Michel, Artur, “Le Ballet d’action avant Noverre”. Archives Internationales de la
Danse pt. I vol. 2 (April 1935), pt. 2 vol. 3/4 (October 1935).
Mori-Sugaya, Tatsuko, “Jean Georges Noverre’s view on Dance Music”. Society of
Dance History Scholars Proceedings. Conference 2002: Philadelphia, Pa.
Moore, Lillian, “Noverre, First of the Moderns”. Dance Magazine (September 1952).
North, Roger, Memoirs of Musick being some historio-criticall collections of that
subject (1728), ed. Edward F. Rimbault. London, 1846.
Noverre, Jean-Georges, Lettres sur la danse, et sur les ballets. Lyons: Delaroche, 1760.
—, Lettres sur la danse et les arts imitateurs. Paris: Éditions Lieutier, 1952.
—, Lettres sur la danse, sur les ballets et les arts, revised and enlarged edition of 1760
Lettres. St Petersburg, 1803. English: Letters on Dancing and on Ballets, trans.
Cyril Beaumont. New York: Dance Horizons, 1968.
Ozouf, Mona, La Fête Révolutionnaire. 1789–1799. Paris: Gallimard, 1976.
Pasch, Johann, Beschreibung wahrer Tanz-Kunst. Frankfurt: Wolffgang Michahelles
& Johann Adolph, 1707.
319 Bibliography and further reading

Plan de Fête à l’Etre Suprême, qui sera célébré à Tours, le 20 Prairial en exécution du
Décret du 18 Floréal, l’an second de la République, une et indivisible. Bibliothèque
National, Paris.
Pruiksma, Rose A., “Generational Conflict and the Foundation of the Académie
Royale de Danse: A Reexamination”. Dance Chronicle 26/2 (2003).
Ralph, Richard, The Life and Works of John Weaver. An Account of his Life, Writings
and Theatrical Productions, with an Annotated Reprint of his Complete
Publications. London: Dance Books, 1985.
Rameau, Pierre, Le Maı̂tre à danser. Paris: Jean Villette, 1725. English: The
Dancing-Master, trans. John Essex. London: J. Essex & J. Brotherton, 1728.
Reynolds, Sir Joshua, Seven Discourses on Art (delivered between 1769 and 1776),
www.authorama.com/book/seven-discourses-on-art.html
Rice, Tamara, “The Evolution of Ballet”. Dancing Times (London) December 1936).
Richardson, P. J. S., “Noverre’s Parentage”. Dancing Times (London) (November
1948).
—, “The First Noverre”. The Dancing Times London February 1960.
Rosen, Charles, The Classical Style: Haydn, Mozart, Beethoven. New York, W. W.
Norton, 1972; 1997.
Rousseau, Jean-Jacques. Julie ou la Nouvelle Héloı̈se. Oeuvres Complètes. Geneva:
Gallimard, Bibliothèque de la Pléiade, 1964, vol. ii.
Roy, Sanjoy, “Wrestling Noverre”. Dance Now (London) 3/4 (Winter 1994).
Sieyès, Emmanuel Joseph, Qu’est-ce que le tiers état? Paris: Flammarion, 1988
(1788).
Sol, C., Méthode très facile et fort nécessaire, pour montrer à la jeunesse de l’un et
l’autre sexe la manière de bien dancer. La Haye: l’Auteur, 1725.
Strässner, Matthias, Tanzmeister und Dichter. Literaturgeschichte(n) im Umkreis von
Jean Georges Noverre. Berlin: Henschel, 1994.
Taubert, Gottfried, Rechtschaffener Tantzmeister. Leipzig: Friedrich Lanckischens
Erben, 1717.
Touchard-Lafosse, Georges, Chroniques secrètes et galantes de l’Opéra. 4 vols. Paris:
Schneider, 1846.
Voltaire (François-Marie Arouet). Oeuvres Complètes. 50 vols., ed. L. Moland. Paris:
Garnier, 1877–85.
Weaver, John, An Essay towards an History of Dancing. London: J. Tonson, 1712.
—, The Loves of Mars and Venus; A Dramatick Entertainment of Dancing, Attempted
in Imitation of the Pantomimes of the Ancient Greeks and Romans. London:
W. Mears, J. Browne, 1717.
Wildeblood, Joan, The Polite World. London: Davis-Poynter, 1973.
Winter, Marian Hannah, The Pre-Romantic Ballet. London: Pitman, 1974.
Wycherly, William, The Gentleman Dancing Master, www3.shropshire-cc.gov.uk/
etexts/E000294.htm.

Part III
Aschengreen, Erik, “The Beautiful Danger: Facets of Romantic Ballet”, trans.
Patricia N. McAndrew. Dance Perspectives 58 (Summer 1974), pp. 2–52.
320 Bibliography and further reading

—, Der går dans. Copenhagen: Gyldendal, 1998 (in Danish but with all important
dates and names of the Royal Danish Ballet 1948–88).
Balanchine, George, Balanchine’s Complete Stories of the Great Ballets, ed. Francis
Mason. Garden City, N.Y. Doubleday, 1977.
Banes, Sally, Dancing Women: Female Bodies on Stage. London and New York:
Routledge, 1998.
Baron, Auguste, Lettres et entretiens sur la danse. Paris: Donde-Dupré, 1824.
Barringer, Janice, and Schlesinger, Sarah, The Point Book. Shoes, Training and
Technique. Hightstown N.J.: Princeton Book Co., 2004.
Beaumont, Cyril W., Complete Book of Ballets: A Guide to the Principal Ballets of the
Nineteenth and Twentieth Centuries. London: Putnam, 1937, reprinted 1956;
and Supplement to Complete Book of Ballet. London: C. W. Beaumont, 1942.
Beaumont, Cyril W., The Ballet Called Giselle. London: C. W. Beaumont, 1945.
Beaumont, Cyril W., and Idzikowski, Stanislas, A Manual of the Theory and Practice
of Classical Theatrical Dancing. Mineola, N.Y.: Dover Publications, 1975–7.
Beaumont, Cyril W., and Sitwell, Sacheverell, The Romantic Ballet in Lithographs of
the Time. London: Faber & Faber, 1938.
Benois, Alexandre, Reminiscences of the Russian Ballet, trans. Mary Britnieva.
London: Putnam, 1941.
Bernstein, Susan, Virtuosity of the Nineteenth Century: Performing Music and
Language in Heine, Liszt, and Baudelaire. Stanford, Calif.: Stanford University
Press, 1998.
Binney, Edwin, III, Glories of the Romantic Ballet. London: Dance Books, 1985.
Bournonville, August, My Theatre Life, trans. Patricia N. McAndrew. London:
A. & C. Black; Middletown, Conn.: Weslayan University Press, 1979.
Brown, David, Tchaikovsky Remembered. London: Faber & Faber, 1993.
Bruhn, Erik, and Moore, Lillian, Bournonville and Ballet Technique: Studies and
Comments on August Bournonville’s Etudes Choregraphiques. Hightstown, N.J.:
Princeton Book Co., 1961, 2005.
Burt, Ramsay, The Male Dancer. London: Routledge, 1995.
Castil-Blaze, François-Henry-Joseph, Dictionnaire de musique moderne. 2 vols.,
Paris: Magasin de Musique de la Lyre Moderne, 1821.
—, De l’opéra en France. Paris: Chez l’auteur, 1826.
—, La Danse et les ballets depuis Bacchus jusqu’à Mademoiselle Taglioni. Paris:
Castil-Blaze, 1832.
Chouquet, Gustave, Histoire de la musique dramatique en France. Paris: Librairie
Firmin Didot Frères, 1873.
Clarke, Mary, and Crisp, Clement, Ballet, An Illustrated History. New York: Universe
Books, 1973.
Clarke, Mary and Crisp, Clement, Design for Ballet. New York: Hawthorn Books,
1978.
Clarke, Mary, and Crisp, Clement, Ballerina: The Art of Women in Classical Ballet.
London: BBC Books, 1987.
Christensen, Anne Middelboe, Hvor danser Den Kgl. Ballet hen? Interview med
danserne om Bournonville-traditionen (Diversions of the Royal Danish Ballet)
(English essay). Copenhagen: Schønberg, 2002.
321 Bibliography and further reading

Cohen, Selma Jeanne, Dance as Theatre Art: Source Readings in Dance History from
1581 to the Present. New York: Dodd & Mead, 1992.
Cordova, Sarah Davies, Paris Dances. Textual Choreographies in the Nineteenth
Century French Novel. San Francisco: International Scholars Press, 1999.
Dahlhaus, Carl, Nineteenth Century Music, trans. J. Bradford. Berkeley: University
of California Press, 1989.
Daly, Ann, “Classical Ballet: A Discourse of Difference”, Women and Performance 3/2
(1987–8), pp. 57–66.
Delarue, Allison (ed.), Fanny Elssler in America. Brooklyn: Dance Horizons, 1976.
Deldevez, Edouard, Mes Mémoires. Le Puy: Marchessou Fils, 1890.
Dent, Edward, The Rise of Romantic Opera, ed. Winton Dean. Cambridge:
Cambridge University Press, 1976.
Deshayes, André, Idées générales sur l’Académie royale de Musique. Paris: Mongie,
1822.
Dithmer, Monna (ed.), Of Another World. Dancing between Dream and Reality.
Museum Tusculanum Press, Copenhagen, 2002.
Doolittle, Lisa and Flynn, Anne (eds.), Dancing Bodies, Living Histories: New
Writings about Dance and Culture. Banff, Canada: The Banff Centre Press, 1999.
Fétis, François-Joseph, Biographie universelle des musiciens. 8 vols., Paris:
Firmin-Didot & cie, 1875–83.
Fischer, Carlos, Les Costumes de l’Opéra. Paris: Librairie de France, 1931.
Foster, Susan Leigh, Choreography and Narrative: Ballet’s Staging of Story and Desire.
Bloomington and Indianapolis: Indiana University Press, 1996.
—, “The ballerina’s phallic pointe”, Corporealities: Dancing Knowledge, Culture and
Power, ed. Susan Leigh Foster. London: Routledge, 1996, pp. 1–24.
Fulcher, Jane F., The Nation’s Image: French Grand Opera as Politics and Political Art.
Cambridge: Cambridge University Press, 1987.
Garafola, Lynn (ed.), Rethinking the Sylph: New Perspectives on the Romantic Ballet.
Hanover, N. H., and London: Wesleyan University Press, 1997.
—, “The Travesty Dancer in Nineteenth Century Ballet”, Dance Research Journal
17/2 (Fall–Spring 1985–6), pp. 35–40.
Gautier, Théophile, Histoire de l’art dramatique en France depuis vingt-cinq ans.
6 vols. Leipzig: Edition Hetzel, 1858–59; Geneva: Slatkine Reprints, 1968.
—, Théatres, Mystères, Comedies et Ballets. OEuvres complètes, Geneva: Slatkine
Reprints, 1978.
—, Gautier on Dance, ed. and trans. Ivor Guest. London: Dance Books, 1986.
Gentil, Louis, Les Cancans de l’Opéra: le journal d’une habilleuse 1836–1848. Ed.
Jean-Louis Tomvaco. Paris: CNRS Editions, 2000.
Gerhard, Anselm, The Urbanization of Opera: Music Theater in Paris in the
Nineteenth Century. Chicago: University of Chicago Press, 1998.
Girard, René, Deceit, Desire, and the Novel: Self and Other in Literary Structure,
trans. Yvonne Freccero. Baltimore: Johns Hopkins Press, 1965.
Guest, Ivor F., Fanny Cerrito: The Life of a Romantic Ballerina. London, Phoenix
House, 1956.
—, Victorian Ballet-Girl: The Tragic Story of Clara Webster. London: A. & C. Black,
1957.
322 Bibliography and further reading

—, Adeline Genée, a Lifetime of Ballet under Six Reigns; Based on the Personal
Reminiscences of Dame Adeline Genée-Isitt. London: A. & C. Black, 1958.
—, The Romantic Ballet in Paris. Middletown, Conn: Wesleyan University Press,
1966; London: Dance Books, 1980.
—, Fanny Elssler. Middletown, Conn.: Wesleyan University Press, 1970.
—, The Romantic Ballet in England: Its Development, Fulfilment, and Decline.
Middletown, Conn.: Wesleyan University Press, 1972.
—, The Ballet of the Second Empire. London: Pitman and Middletown: Wesleyan
University Press, 1974.
—, Le Ballet de l’Opéra de Paris. Paris: Flammarion, 1976.
—, The Divine Virginia: A Biography of Virginia Zucchi. New York: M. Dekker, 1977.
—, Jules Perrot, Master of the Romantic Ballet. New York: Dance Horizons, 1984.
—, Ballet in Leicester Square: The Alhambra and the Empire, 1860–1915. London:
Dance Books and Princeton, N.J.: Princeton Book Co., 1992.
Hallar, Marianne, and Scavenius, Alette (eds.), Bournonvilleana. Copenhagen:
Rhodos, 1992.
Hanna, Judith Lynne, Dance, Sex and Gender: Signs of Identity, Dominance, Defiance,
and Desire. Chicago: University of Chicago Press, 1988.
Horosko, Marian, “If the Shoe Fits”. Dance Magazine 60 (April 1986), pp. 80–1; 60
(May 1986), pp. 98–9.
Hutchinson, Ann, Fanny Elssler’s Cachucha. Intro. Ivor F. Guest. New York: Theatre
Arts Books, 1981.
Join-Diéterle, Catherine, Les Décors de scéne de l’Opéra de Paris à l’époque
romantique. Paris: Picard, 1988.
Jowitt, Deborah, Time and the Dancing Image. Los Angeles: University of California
Press, 1988.
Jürgensen, Knud Arne, The Ballet Tradition: The First Fifty Years, 1829–1879. Vols.
i–ii. London: Dance Books, 1997.
Kahane, Martine, et al., Le tutu. Paris: Flammarion (Les petits guides de l’Opéra),
2000.
Karsavina, Tamara, Theatre Street: The Reminiscences of Tamara Karsavina. London:
Heinemann, 1930.
Krasovskaya, Vera, “Marius Petipa and ‘The Sleeping Beauty’”, trans. Cynthia Read,
Dance Perspectives 49 (Spring 1972).
—, Vaganova: A Dance Journey from Petersburg to Leningrad, trans. Vera M. Siegel,
intro. Lynn Garafola. Gainesville: University Press of Florida, 2005.
Kschessinska, Mathilde (Kshesinska, Princess Romanovsky-Krassineky), Dancing in
Petersburg: The Memoirs of Kschessinska, trans. Arnold Haskell. London:
Gollancz, 1960.
Lawson, Joan, A History of Ballet and its Makers. London: Dance Books, 1923.
—, A Ballet-Maker’s Handbook: Sources, Vocabulary, Styles. London: A. & C. Black
and New York: Theatre Arts Books/Routledge, 1991.
Levinson, André, Marie Taglioni, trans. Cyril W. Beaumont. London: Dance Books,
1977.
—, Ballet Old and New, trans. from the Russian Susan Cook Summer. New York:
Dance Horizons, 1982.
323 Bibliography and further reading

—, Levinson on Dance. Writings from Paris in the Twenties. Ed. Joan Acoccella and
Lynn Garafola. Hanover N.H.: Wesleyan University Press, 1991.
Lifar, Serge, Ballet, Traditional to Modern, trans. Cyril W. Beaumont. London:
Putnam, 1938.
—, Giselle: Apothéose du ballet romantique. Paris: Albin Michel, 1942.
—, La Musique par la danse, de Lulli à Prokofiev. Paris: R. Laffont, 1955.
Lynham, Derek, Ballet Then and Now. A History of Ballet in Europe. London: Sylvan
Press, 1947.
McCarren, Felicia, Dance Pathologies: Performance, Poetics, Medicine. Stanford,
Calif.: Stanford University Press, 1998.
Meglin, Joellen A., “Representations and Realities: Analyzing Gender Symbols in
the Romantic Ballet”, Ed.D. Diss. Temple University Philadelphia, 1995.
Mellor, Anne K. (ed.), Romanticism and Feminism. Bloomington: Indiana
University Press, 1988.
Migel, Parmenia, The Ballerinas. From the Court of Louis XIV to Pavlova. New York:
MacMillan, 1972.
Nettl, Paul, The Story of Dance Music. New York: Philosophical Library, 1947.
Nørlyng, Ole og Urup, Henning (ed.), Bournonville. Tradition. Rekonstruktion.
(English summaries). Copenhagen: C.A. Reitzels Forlag, 1989.
Petipa, Marius, Russian Ballet Master: The Memoirs of Marius Petipa. Ed. Lillian
Moore, trans. Helen Whittaker. London: A. & C. Black, 1958.
—,The Diaries of Marius Petipa. Ed., trans. and intro. Lynn Garafola. Studies in
Dance History 3/1 (Spring 1992) Society of Dance History Scholar.
Pougin, Arthur, Adolphe Adam: sa vie, sa carrière, ses mémoires artistiques. Paris:
Charpentier, 1877.
Reading Critics Reading: Opera and Ballet Criticism in France from the Revolution to
1848, ed. Roger Parker and Mary Ann Smart. Oxford: Oxford University Press,
2001.
Robin-Challan, Louise, Danse et danseuses: l’envers du décor (1830–1850). Thèse
sociologie. Université de Paris VII, 1983.
Roslavleva, Natalia, Era of the Russian Ballet. Foreword Ninette de Valois. London:
Gollancz, 1966.
Saint-Léon, Arthur, La Sténochorégraphie: Ou, art d’écrire promptement la danse.
1852.
—, Letters from a Ballet-Master: The Correspondence of Arthur Saint-Léon. Ed. Ivor F.
Guest. New York: Dance Horizons, 1981.
Second, Albéric, Les Petits Mystères de l’Opéra. Paris: Kugelman 1844.
Slonimsky, Yuri, “Marius Petipa”, trans. Anatole Chujoy, Dance Index 6/5–6
(May–June 1947).
Smith, Albert, The Natural History of the Ballet Girl. London: D. Bogue, 1847.
Smith, Marian, Ballet and Opera in the Age of Giselle. Princeton and Oxford:
Princeton University Press, 2000.
—, “The Earliest Giselle? A Preliminary Report on a St Petersburg Manuscript”.
Dance Chronicle: Studies in Dance and the Related Arts 23/1 (2000), pp. 29–48.
Solomon-Godeau, Abigail, “The Other Side of Venus: The Visual Economy of
Feminine Display”, in Victoria de Grazia and Ellen Furlough (eds.), The Sex of
324 Bibliography and further reading

Things: Gender and Consumption in Historical Perspective. Berkeley: University


of California Press 1996, pp. 113–150.
Sorell, Walter, Looking back in Wonder: Diary of a Dance Critic. New York: Columbia
University Press, 1986.
—, Dance in its Time. New York: Columbia University Press 1986.
Souritz, Elizabeth, “Carlo Blasis in Russia (1861–1864)”. Studies in Dance History
4/2 (Fall 1993).
Spilken, Terry, The Dancers Foot Book: A Complete Guide to Foot Care. Hightstown
N.J.: Princeton Book Co., 1990.
Stepanov, Vladimir Ivanovich, Alphabet of Movements of the Human Body: A Study
in Recording the Movements of the Human Body by Means of Musical Signs.
Originally published in French 1892, New York: Princeton Book Publ. Dance
Horizons, 1969.
Taruskin, Richard, The Oxford History of Western Music, New York: Oxford
University Press, 2005.
Théleur, E. A., Letters on Dancing. Intro. Sandra N. Hammond. Society of Dance
History Scholars: Pennington, N.J., 1990.
Terry, Walter, On pointe! The Story of Dancing and Dancers on Toe. New York: Dodd
& Mead, 1962.
Verdy, Violette, Giselle: A Role for a Lifetime. New York: M. Dekker, 1977.
Véron, Louis-Désiré, Mémoires d’un bourgeois de Paris. Paris: Librairie nouvelle,
1856.
Wiley, Roland John, Tchaikovsky’s Ballets : Swan Lake, Sleeping Beauty, Nutcracker.
Oxford: Clarendon Press and New York: Oxford University Press, 1985.
—, (ed. and trans.), A Century of Russian Ballet: Documents and Eyewitness Accounts
1810–1910. Oxford: Clarenden Press, 1990.
—, The Life and Ballets of Lev Ivanov: Choreographer of The Nutcracker and Swan
Lake. Oxford: Clarendon Press and New York: Oxford University Press,
1997.
Wolkonsky, Prince Serge, My Reminiscences, trans. A. E. Chamot. London:
Hutchinson, n.d.

Newspapers
Journal des débats
Revue et gazette musicale de Paris

Part IV
Acocella, Joan, Mark Morris. New York: Noonday Press, 1995.
Acosta, Carlos, Tocororo: A Cuban Tale. London: Oberon Books, 2004.
Alfonso Rey, F., and Simón, P., Alicia Alonso: Órbita de una Leyenda. Madrid: SGAE,
1996.
Alonso, A., Diálogos con la danza, ed. P. Simón. La Habana: Letras Cubanas, 1986.
Amberg, George, Ballet: The Emergence of an American Art. New York: Duell, Sloan
& Pearce, 1949.
Anderson, Jack, The Nutcracker Ballet. New York: Gallery Books, 1979.
325 Bibliography and further reading

—, The One and Only: The Ballet Russe de Monte Carlo. New York: Dance Horizons,
1981.
—, Ballet and Modern Dance: A Concise History. Hightstown, N.J.: Princeton Book
Co., 1986.
—, Art without Boundaries: The World of Modern Dance. Iowa City: University of
Iowa Press, 1997.
Anderson, Zoe, The Royal Ballet: 75 Years. London: Faber, 2006.
Baer, Nancy Van Norman (ed.), Theatre in Revolution: Russian Avant-garde Stage
Design, 1913–1935. New York: Thames & Hudson; San Francisco: Fine Arts
Museums of San Francisco, 1991.
—, (ed.), Paris Modern: The Swedish Ballet 1920–1925. San Francisco: Fine Arts
Museums of San Francisco, 1995.
Banes, Sally, Terpsichore in Sneakers: Post-Modern Dance. Boston: Houghton Mifflin,
1979.
Beaumont, Cyril W., The Monte Carlo Russian Ballet (Les Ballets russes du Col. W.
de Basil) London: C. W. Beaumont, 1934.
—, The Sadler’s Wells Ballet: A Detailed Account of Works in the Permanent Repertory,
with critical notes. London: C. W. Beaumont, 1947.
—, Ballets of Today: Being a Second Supplement to the Complete Book of Ballets.
London: Putnam, 1954.
—, Ballets Past and Present: Being a Third Supplement to the Complete Book of Ballets.
London: Putnam, 1954.
—, Michel Fokine and his Ballets. New York: Dance Horizons, 1981.
Beaumont, Cyril W., and Idzikowski, Stanislas, A Manual of the Theory and Practice
of Classical Theatrical Dancing: Cecchetti Method (1922). Rev. edn, London,
1940. Reprints, New York: Dover Publications, 1975 and 2003.
Béjart, Maurice, Béjart et la Danse de la symphonie pour un homme seul à la IX˚
Symphonie. Brussels: Verbeeck, 1966.
—, Le Ballet des mots. Paris: Les Belles Lettres: Archimbaud, 1994.
—, La Vie de qui? Paris: Flammarion, 1996.
Bogdanov-Berezovskii, V. M., Ulanova and the Development of the Soviet Ballet,
trans. Stephen Garry and Joan Lawson. London: 1952.
Bremser, Martha, Fifty Contemporary Choreographers. London: Routledge, 1999.
Brinson, Peter (ed.), Ulanova, Moiseyev, and Zakharov on Soviet Ballet, trans. E. Fox
and D. Fry. London: Society for Cultural Relations, 1954.
—, Background to European Ballet: A Notebook from its Archives. Foreword by Mary
Skeaping. Leyden: A. W. Sijthoff, 1966.
Buckle, Richard, Nijinsky. New York, Simon & Schuster, 1971.
—, Diaghilev. New York: Atheneum, 1979.
Buckle, Richard, in collaboration with John Taras, George Balanchine: Ballet Master.
New York: Random House, 1988.
Chauviré, Yvette, and Mannoni, Gérard, Yvette Chauviré – Autobiographie. Paris: Le
Quai 1997.
Choreography by George Balanchine: A Catalogue of Works, ed. Harvey Simmonds.
New York: Viking, 1984.
Chujoy, Anatole, The Symphonic Ballet. New York: Kamin, 1937.
326 Bibliography and further reading

Clarke, Mary, Dancers of Mercury. London, 1961.


Cohen, Selma Jeanne, Next Week, Swan Lake: Reflections on Dance and Dances.
Middletown, Conn.: Wesleyan University Press, 1982.
Copeland, Roger and Cohen, Marshall (eds.), What is Dance?: Readings in Theory
and Criticism. New York: Oxford University Press, 1983.
Coton, A. V., A Prejudice for Ballet. London: Methuen, 1938.
—, The New Ballet: Kurt Jooss and his Work. London: D. Dobson, 1946.
Crisp, Clement, Sainsbury, Anya, and Williams, Peter (eds.), Ballet Rambert: Fifty
Years and On. London: Ballet Rambert, 1981.
Daneman, Meredith, Margot Fonteyn: A Life. New York: Viking, 2004.
Danilova, Alexandra, Choura: The Memoirs of Alexandra Danilova. New York:
Knopf/ Random House, 1986.
Denby, Edwin, Looking at the Dance. New York: Horizon Press, 1968.
—, Dance Writings, ed. Robert Cornfield and William MacKay. New York: Alfred A.
Knopf, 1986, pp. 272–5, 445–50.
Duncan, Isadora, My Life. New York: H. Liveright, 1928.
Farrell, Suzanne, with Toni Bentley, Holding on to the Air: An Autobiography. New
York : Summit Books, 1990.
Fisher, Jennifer, Nutcracker Nation: How an Old World Ballet Became a Christmas
Tradition in the New World. New Haven and London: Yale, 2003.
Fokine, Michail, Memoirs of a Ballet Master, trans. Vitale Fokine, ed. Anatole
Chujoy. Boston: Little, Brown, 1961
Franchi, Cristina (ed.), Frederick Ashton. Founder Choreographer of the Royal Ballet.
London: Oberon Books, 2005.
—, (ed.), Margot Fonteyn. London: Oberon Books, 2005.
Frétard, Dominique and Tapie, Gilles (eds.), Invitation: Sylvie Guillem. London:
Oberon Books 2006.
Garafola, Lynn, Diaghilev’s Ballets Russes. New York: Oxford University Press, 1989.
Garcı́a-Márquez, Vicente, The Ballets Russes: Colonel de Basil’s Ballets Russes de
Monte Carlo, 1932–1952. New York: Alfred A. Knopf, 1990.
—, Massine. A Biography. New York: Alfred A. Knopf, 1995.
Garcı́a Victorica, Victoria, El original Ballet Russe en América latina. Buenos Aires:
n.p., 1948.
Garis, R., Following Balanchine. New Haven: Yale University Press, 1995.
Giselle, film with Alicia Alonso and the National Ballet of Cuba, directed by E.
Pineda Barnet (Havana: ICAIC, 1965).
Granet, Marcel, Danses et legends de la Chine ancienne. 2 vols. Paris: Presses
Universitaires de France, 1959.
Grau, André, and Jordan, Stephanie (eds.), Europe Dancing: Perspectives on Theatre
Dance and Cultural Identity. Eds. André Grau and Stephanie Jordan. London
and New York: Routledge, 2000.
Häger, Bengt, Ballets Suédois: The Swedish Ballet, trans. Ruth Sherman. New York:
H. N. Abrams, 1990.
Haskell, Arnold L., Balletomania. New York: Simon & Schuster: 1934.
—, Diaghileff: His Artistic and Private Life. New York: Simon & Schuster, 1935.
—, Dancing Round the World. New York: Dodge Comp., 1938.
327 Bibliography and further reading

—, The National Ballet: A History and a Manifesto. With an overture by Ninette de


Valois. London: A. & C. Black, 1943.
Jeanmaire, Zizi, and Mannoni, Gérard, Zizi Jeanmaire. Paris: Assouline, 2002.
Jordan, Stephanie, Moving Music : Dialogues with Music in Twentieth-Century Ballet
London: Dance Books, 2000.
Joseph, Charles M., Stravinsky and Balanchine: A Journey of Invention. New Haven :
Yale University Press, 2002.
Jowitt, Deborah, “Celebrating the Solstice with Candycanes”, in Dance Beat: Selected
Views and Reviews 1967–1976. New York and Basel: Marcel Dekker, 1977.
Kahane, Martine, Les Ballets russes de l’opéra. Paris: Hazan, Bibliothèque Nationale,
1992.
—, Nijinsky: 1889–1950. Catalogue établi par Martine Kahane. Paris: Editions de la
Réunion des musées nationaux, 2000.
Kavanagh, Julie, Secret Muses: The Life of Frederick Ashton. New York: Pantheon,
1996.
Kirstein, Lincoln., Dance: A Short History of Classic Theatrical Dancing. Westport,
Conn.: Greenwood Press 1935/1970.
—, Thirty Years: Lincoln Kirstein’s The New York City Ballet, expanded to include the
years 1973–1978, in celebration of the company’s thirtieth anniversary.
New York: Knopf, 1978.
—, Four Centuries of Ballet: Fifty Masterworks. New York: Dover, 1984.
Laban, Rudolf von, Modern Educational Dance. London: Macdonald & Evans, 1948.
—, A Life for Dance: Reminiscences, trans. and annotated Lisa Ullmann. London:
Macdonald & Evans, 1975.
Laws, Kenneth, Physics and the Art of Dance: Understanding Movement. Oxford and
New York: Oxford University Press, 2002.
Lieven, Peter, The Birth of Ballets-Russes, trans. Leonide Zarine. London: George
Allen & Unwin, 1936.
Lido, Serge, Lidova, Irène, and Béjart, Maurice, Ballet dans le monde: Formes
nouvelles. Paris: Vilo, 1977.
Lidova, Irène, Dix-sept visages de la danse française. Paris: Art et Industrie, 1953.
—, Roland Petit. Paris: Vilo, 1956.
Lifar, Serge, Serge Diaghilev, His Life, His Work, His Legend: An Intimate Biography.
New York: G. P. Putnam’s Sons, 1940.
—, The Three Graces: Anna Pavlova, Tamara Karsavina, Olga Spessivtzeva; the
Legends and the Truth, trans. Gerard Hopkins. London: Cassell, 1959.
Lopukhov, Fedor Vasilevich, Writings on Ballet and Music, ed. and intro. Stephanie
Jordan; trans. Dorinda Offord. Madison: University of Wisconsin Press, 2002.
Mannoni, Gérard (ed.), Roland Petit. Paris: Avant-scène, 1984.
Mannoni, Gérard, Roland Petit: Un chorégraphe et ses peintres. Paris: Hatier, 1990.
Mannoni, Gérard, and Masson, Colette, Maurice Béjart. Paris: Plume 1991.
Maré, Rolf de (ed.), Les Ballets Suédois dans l’art contemporain. Paris: Editions du
Trianon, 1931.
Massine, Léonide, My Life in Ballet. London: Macmillan, 1968.
Mazo, Joseph H., Prime Movers: The Makers of Modern Dance in America.
Hightstown, N.J.: Princeton Book Co., 2000.
328 Bibliography and further reading

de Mille, Agnes, America dances. New York : Macmillan, 1980.


—, Reprieve: A Memoir. Garden City, N.Y.: Doubleday, 1981.
Morris, Mark, Mark Morris’s l’Allegro, il pensoroso ed il moderato: a celebration, ed.
Jeffrey Escoffier and Matthew Lore. New York: Marlowe & Co., 2001.
Nemenschousky, Leon, A Day with Yvette Chauviré. London: Cassell, 1960.
Nijinska, Bronislava, Early Memoirs, ed. and trans. Irina Nijinska and Jean
Rawlinson New York: Holt, Rinehart & Winston, 1981.
Nijinsky, Romola. Nijinsky. New York: Simon & Schuster, 1934.
Nijinsky, Vaclav, The Diary of Vaslav Nijinsky, ed. Romola Nijinsky. Berkeley:
University of California Press, 1968 (c.1936).
Norton, Leslie, Leonide Massine and the 20th Century Ballet. Jefferson, N.C.:
McFarland & Co., 2004.
Perlmutter, Donna, Shadowplay: The Life of Anthony Tudor. Newark N.J.: Limelight
Editions, 1995.
Petit, Roland, and Mannonni, Gérard, Roland Petit, un choréographe et ses danseurs.
Paris: Plume 1992.
Plissetskaia, Maya M., I, Maya Plisetskaya, trans. Antonina W. Bouis, foreword by
Tim Scholl. New Haven: Yale University Press, 2001.
Prevots, Naima, Dance for Export: Cultural Diplomacy and the Cold War. Intro. Eric
Foner. Middletown, Conn.: Wesleyan University Press, 2001.
Propert, Walter Archibald, The Russian Ballet 1909–1920. London: John Lane The
Bodley Head, 1921.
—, The Russian Ballet 1921–1929. London: John Lane The Bodley Head, 1931.
Rambert. A Celebration, comp. Jane Pritchard, intro. Mary Clarke and Clement
Crisp. London: Rambert Dance Company, 1996.
Reynolds, Nancy and McCormick, Malcolm, No Fixed Points. New Haven: Yale
University Press, 2003.
Sachs, Curt. World History of the Dance, trans. Bessie Schonberg. New York: Norton,
1965. (Originally published in German in 1933.)
Schneider, Marcel, and Michel, Marcelle, Danse à Paris. Paris: Grasset, 1983.
Scholl, T., From Petipa to Balanchine: Classical Revival and the Modernization of
Ballet. London: Routledge, 1994.
Schorer, Suki, Balanchine Pointework. Studies in Dance History, no. 11. Madison:
Wisconsin University Press, 1995.
Siegel, Marcia, “Decomposing Sugar Plums and Robot Mice”. Ballet Review 19/1
(Spring 1991), pp. 58–62.
—, “Kingdom of the Sweet”, Hudson Review 50/2 (Summer 1997), pp. 255–67.
Solomon, Ruth and John (eds.), East Meets West in Dance: Voices in the
Cross-Cultural Dialogue. Chur, Switzerland: Harwood Academic Publishers,
1995.
Solway, Diane, Nureyev: His Life. New York: William Morrow, 1998.
Souritz, Elizabeth, Soviet Choreographers in the 1920s, trans. Lynn Visson. Durham,
N.C.: Duke University Press, 1990.
Svenska baletten: Les Ballets Suédois, 1920–1925. Stockholm, 1969, 1995. Exhibition
catalog, Dansmuseet.
329 Bibliography and further reading

Taper, Bernard, Balanchine. A Biography. With a New Epilogue. New York: Times
Books, 1984; and Berkeley: University of California Press, 1996.
Teachout, Terry, All in the Dances: A Brief Life of George Balanchine. Orlando:
Harcourt, 2004.
Terry, Walter, Ballet Guide: Background, Listings, Credits, and Descriptions of More
Than Five Hundred of the World’s Major Ballets. New York: Dodd, Mead, 1976.
—, I Was There: Selected Dance Reviews and Articles, 1936–1976. New York: Dekker,
1978.
—, Alicia and her Ballet Nacional de Cuba. Garden City, N.Y.: Anchor Books,
1981.
Vaganova, Agrippina, Basic Principles of Classical Ballet: Russian Ballet Technique,
trans. Anatole Chujoy, ed. Peggy van Praagh. New York: Dover Publications,
1953/69.
Vaughan, David, Frederick Ashton and his Ballets. London: A. & C. Black, 1977.
Volkov, Solomon. Balanchine’s Tchaikovsky: Interviews with George Balanchine. New
York: Simon and Schuster, 1985.
Walker, Katherine Sorley, Ninette de Valois: An Idealist without Illusions. London:
Hamish Hamilton, 1988.
Wiley, Roland John, “On Meaning in Nutcracker”. Dance Research 3/1 (autumn
1984), pp. 3–28.

Encyclopedias and dictionaries


Balet Entsiklopediia. See Soviet Ballet Encyclopedia.
Concise Oxford Companion to the Theatre, The, ed. Phyllis Hartnoll and Peter
Found. New York: Oxford University Press 1996/2003.
Concise Oxford Dictionary of Ballet, The, ed. Horst Koegler. Oxford, London, New
York: Oxford University Press, 1982.
Concise Oxford Dictionary of Opera, The, ed. John Warrack and Ewan West. London,
New York: Oxford University Press 1996/2003.
Dance Encyclopedia, The, ed. and comp. Anatole Chujoy and P.W. Manchester. New
York: Simon & Schuster, 1967.
Dictionnaire des danses de la Renaissance, ed. Henri Jarrie. Arles: Harmonia Mundi,
1970.
Dictionnaire de la musique en France aux XVlle et XVllle siècles, ed. Marcelle Benoit.
Paris: Fayard, 1992.
Encyclopedia of Dance and Ballet, The, ed. Mary Clarke and David Vaughan. New
York: Putnam, 1977.
Encyclopedia of Musical Theatre, The, ed. Kurt Ganzl. New York: Schirmer Books,
2001. 3 vols.
International Dictionary of Ballet, The, ed. Martha Bremser. Detroit: St James’s
Press, 1993.
International Encyclopedia of Dance, The, ed. Selma Jean Cohen and Dance
Perspectives Foundation. Oxford: Oxford University Press, 1998, 2005.
Musik in Geschichte und Gegenwart: Allgemeine Enzyklopädie der Musik. Kassel:
Bärenreiter, 1994.
330 Bibliography and further reading

New Grove Dictionary of Music and Musicians, The, ed. Stanley Sadie and John
Tyrrell. London: Macmillan, 2001. 29 vols.
New Harvard Dictionary of Music, The, ed. Don Michael Randel. Cambridge, Mass.:
Harvard University Press, 1986.
Oxford Dictionary of Dance, The, ed. Debra Craine and Judith Mackrell. New York:
Oxford University Press, 2000.
Oxford Dictionary of Music, The, ed. Michael Kennedy, associate ed. Joyce Bourne.
New York: Oxford University Press 1996/2003/4.
Oxford Encyclopedia of Theatre and Performance, ed. Dennis Kennedy. New York:
Oxford University Press, 2003–5.
Pipers Enzyklopädie des Musiktheaters: Oper, Operette, Musical, Ballett, ed. Carl
Dahlhaus and Sieghart Döhring. Munich: Piper, 1986–97.
Soviet Ballet Encyclopedia (Balet Entsiklopediia). Moscow: Sovetskaia Entsiklopediia,
1981.
Index of persons

Abromeit, Klaus 45 Bargrave, Robert (1628–61) 39, 40


Adam, Adolphe Charles (1803–56) 116, Barrez, Jean-Baptiste 121
141–2, 144, 146–7 Baryshnikov, Mikhail (b. 1948) 43, 286
Adams, Diana 233 Batteux, Charles (1713–60) 93, 176
Addison, Joseph (1672–1719) 78 Bauchant, André (1873–1958) 243
Adice, G. Léopold 72 Baudelaire, Charles (1821–1867) 182–3
Aesop 240 Bayle, Christine 43
Aglié, Filippo, d’ (1604–67) 28 Beaman, Patricia 43
Albert, François Decombe (1787–1865) 75 Beauchamps, Pierre (1631–1705) 66–7, 83,
Allard, Mlle 91 277
Alonso, Alicia (Alicia Ernestina de la Caridad Beaujoyeux, Balthasar de (Baltazarini o
del Cobre Martinez Hoya) (b. 1920) Baldassarre b. before 1535, d. c.1587)
263–71 29, 30
Alvarez, Tito (photo) 265 Beaumont, Francis 36
Andersen, Frank 127, 136 Bedford, Lady (Isabella, countess of Bedford
Andersen, Hans Christian (1805–75) 132 (1332–79) 38
Angiolini, Gasparo (1731–1803) 56, 59–61, Beethoven, Ludwig van (1770–1827) 148,
93–4, 152, 176 221–2
Anna of Denmark (1574–1619) 33 Béjart, Maurice (1927) 251, 286–7, 289
Anna Ivanovna, Empress of Russia Bely, Andrey (Boris Nikolaevich Bugaev)
(1693–1740) 152 (1880–1934) 212
Anne of Austria (1601–1666) 19, 25 Benois, Alexandre (1870–1960) 158, 212, 216
Anne, Queen of England (1665–1714) 38, 78 Benserade, Isaac (1612–91) 22, 30
Anne d’Arcques, Duke of Joyeuse (1561–87) Bentley, George (1828–95) 34
29 Berain, Jean (1638–1711) 48
Apostolidès, Jean-Marie 46 Berners-Lee, Tim 272
Appia, Adolphe (1862–1928) 206 Berlioz, Hector (1803–69) 177
Arbeau, Thoinot (Jehan Tabouret) (1520–95) Bertoluzzi, Francesco 58
15 Bjørn, Dinna 126, 136
Aristotle (384–322bc) 23, 93, 194 Blangstrup, Mads 131
Ashton, Frederick William Mallandaine Blasis, Annunziata (Annunziata Rammacini)
(1904–88) 146, 150, 287 76–7
Aspelmayer, Franz (1728–86) 90 Blasis, Carlo (1795–1878) 76–7, 96, 176,
Astaire, Fred (1899–1987) 228 185–6
Auber, Daniel-François-Esprit (1782–1871) Blok, Alexander (1880–1921) 226
117 Bocan (Jacques Cordier) 24–5, 34
Aumer, Jean Pierre (1774–1833) 62 Boehm, Theobald (1794–1881) 186
Börlin, Jean (1893–1930) 202–11
Baccelli, Giovanna (d. 1801) 74 Bojesen, Gudrun 127
Bach, Johann Sebastian (1685–1750) 48, 148, Bolm, Adolph Rudolphovich (1884–1954)
157, 167, 225, 243 161
Baı̈f, Jean Antoine de (1532–89) 15, 23 Bonaparte, Josephine (1763–1814) 186
Baker, Josephine (1906–75) 227, 230 Bonaparte, Napoleon (1769–1821) 186
Balanchine, George (Georgi Melitonovitch Bonin, Louis (18th c.) 73
Balanchivadze) (1904–83) 42, 146, 149, Bonnard, Pierre (1867–1947) 201, 202, 203
219, 221–2, 224–36, 237–45, 249–50, 253, Borchsenius, Valborg 126
268, 276, 282 Borgonio, Tommaso (1620–83) 28
Ballon (Balon), Claude (also Jean) Bos (Dubos), Jean-Baptiste (Abbé) du
(1671–1744) 54, 70, 83 (1670–1742) 93, 176
Ballon, François 70 Boullée, Etienne-Louis (1728–1799) 104
Banister, John (1630–79) 82 Boquet, Louis-René 89
[331]
332 Index of persons

Bourne, Matthew (b. 1960) 250 Colucci, Philip 229


Bournonville, August (1805–79) 75, 126–37, Comerre, Léon François (1850–1916) 124
140, 150, 155–6, 268 Congreve, William (1670–1729) 81
Brandes, Georg (1842–1927) 133 Cooper, Anthony Ashley, Third Earl of
Braque, Georges (1882–1963) 202 Shaftesbury (1671–1713) 85
Brecht, Bertolt (1898–1956) 276 Copland, Aaron (1900–90) 146
Brenaa, Hans 126 Coralli, Jean (1779–1854) 116, 121, 178
Brice, Fanny (1891–1951) 230 Cordier, Jacques 34
Brugnoli, Amalia (c.1810–?) 191 Corelli, Arcangelo (1653–1713) 203
Bryusov, Valery (1873–1924) 212 Cornazano, Antonio (1430–84) 13
Buckingham, Duke of 37, 38, 39 Corneille, Pierre (1606–84) 57, 90
Bueno, Salvador 267 Coulon, Jean-François (1764–1836) 75, 76
Burette, P. J. 93 Craig, Edward Gordon (1872–1966) 206
Burgmüller, Frederich (1804–74) 150 Cranko, John (1927–73) 146, 250, 285
Burney, Charles (1726–1814) 79, 81 Crébillon, Claude Prosper Jolyot de 57, 58
Bussus, Filippus 12 Crowne, John (1641–1712) 40, 41
Byrd, Donald 250 Cruz, Heidi 254
Cullberg, Birgit (1908–90) 287
Cage, John (1912–92) 208 Cunningham, Merce (b. 1919) 208
Cahusac, Louis de (1700–59) 53, 55–6, 93
Calzibigi (Calzabigi), Raniero di (1714–95) Dai Ailian (b. 1916) 258
89 Dandré, Victor (1870–1944) 162
Camargo, Marie Anne de Cupis (1710–70) Daniel, Samuel (1562/3–1619) 33, 34
62, 74, 87, 162 Danilova, Alexandra (1903–97) 226–7, 240,
Campanini, Mr 87 249
Campion, Thomas (1567–1620) 34 Dardel, Nils (1888–1943) 201, 204
Careri, Giovanni 46 Dauberval, Jean (Jean Bercher) (1742–1806)
Carew, Thomas (1594/5–1640) 33, 34 61, 89, 91
Carl Eugene, Duke of Wurttemberg (1732–97) Davenant, William (1606–68) 34
56, 89–90, 95 David, Jacques-Louis (1748–1825) 98, 101–3,
Caroso, Fabritio (c.1526–c.1605) 11, 15, 33, 107
176 Debord, Guy (1931) 46
Carpentier, Alejandro (1904–80) 269 Debussy, Claude (1862–1918) 145, 203
Castil-Blaze, François Henri Joseph Degas, Edgar (1834–1917) 190
(1784–1857) 39, 176–8, 180 Deldevez, Edoardo (Edouardo) (1817–97)
Castro, Fidel (1926) 270, 271 144, 150
Castro, Martha, de 269 Deleuze, Gilles (1925–95) 46
Caturla, Alejandro Garcı́a 269 Delibes, Léo (1836–91) 141–2, 144, 146,
Caverley, Thomas (c.1648–1745) 78–9, 83–4 148–9
Cazotte, Jacques (1719–92) 122 Dell’Era, Antonietta (c.1861–?) 247
Cecchetti, Enrico (1850–1928) 287 Demachy, Pierre Antoine (1723–1807) 108
Cerrito, Fanny (1817–1909) 121, 131 Denby, Edwin (1903–83) 225, 252, 263–4
Chapman, George 33 Denishawn (Ruth St Denis) (1879–1968) 257
Charles I (1625–49) 33, 37–9, 41 Depero, Fortunato (1892–1960) 206
Charles II (1630–85) 40, 41 De Maisse, French ambassador at Elizabeth I’s
Chauviré, Ivette (1917) 266 court 33
Chekhov, Anton (1860–1904) 213 Deshayes, André Jacques-François
Chirico, Giorgio, de (1888–1978) 238 (1777–1846) 75
Chopin, Frédéric (1810–49) 165, 226 Diaghilev, Serge (Sergei Diagilev) (1872–1929)
Christian VIII (1786–1848) 132 161, 163, 201–4, 210–11, 212, 214, 216,
Christensen, Willem 248–9 218, 225, 227–8, 239, 248, 279, 282, 287
Christine, Queen of France (daughter of Henry Didelot, Charles (1767–1837) 190, 237
IV and Maria de Medici) 28, 29 Didelot, Rose (Rose Paul) (d. 1803) 64
Christout, Marie Françoise (b. 1928) 22–3, 46 Diderot, Denis (1713–84) 87, 93, 95, 202, 206
Ciceri, Pierre (1782–1868 ) 114 Dolin, Anton (Sydney Francis Patrick
Cixi, Chinese Empress Dowager 257 Healey-Kay) (1904–83) 266
Clustine, Ivan (1862–1941) 248 Donizetti, Domenico Gaetano Maria
Cocteau, Jean (1889–1963) 204, 205 (1797–1848) 146
333 Index of persons

Doubrovska, Felia (1896–1981) 242, 288 Gaffoyne, Jasper 33


Drigo, Riccardo (1846–1930) 165, 167–8 Gallini, Giovanni-Andrea (1728–1805) 72
Dryden, John (1631–1700) 40 Gardel, Maximilien (1741–87) 62, 91, 98
Duchamp, Marcel (1887–1968) 233, 283 Gardel, Pierre (1758–1840) 60, 62, 91, 98,
Dudinskaya, Natalya (1912–2003) 280 100–1, 106–7, 149
Dufort, Giambattista (c.1680?) 71 Garrick, David (1717–79) 88–90, 96
Dumilâtre, Adele (1821–1909) 121 Gautier, (Pierre Jules) Théophile (1811–72)
Dumoulin, Mr (brothers, first half 18th c.) 74 115–16, 118, 153–4, 177–81, 183, 184–5,
Duncan, Irma (Irma Dorette Henriette 189, 191, 193, 195–6, 283
Ehrich-Grimme) (1897–1977) 257 Gence, J. B. 106
Duncan, Isadora (Dora Angela) (1877–1927) Gené, Beth 227
49, 194–5, 209, 215, 218, 226, 239, 257, Gentil, Louis 182
274–5, 279 Gershwin, George (1898–1937) 225, 228, 234
Duport, Pierre Landrin 138 Geva, Tamara (1908–97) 226–7
Dupré, Louis (1697–1774) 56, 87 Gide, Casimir (1804–68) 144
Dürer, Albrecht (1471–1528) 12 Gielgud, Maina (b. 1945) 287
Gillray, James (1757–1815) 64
Ebreo, Guglielmo (1420–84) 11, 13, 16–17 Glazunov, Alexander Konstantinovich
Edgar, Sarah 43 (1865–1936) 142, 169
Edison, Thomas Alva (1847–1931) 188 Glinka, Mikhail Ivanovich (1804–57) 226
Ek, Mats (1945) 289 Gluck, Christoph Willibald (1714–87) 89–90,
El Greco (Domenikos Theotokopoulos) 97
(1541–1614) 202, 206 Goethe, Johann Wolfgang von (1749–1832)
Eliot, T. S. (Thomas Stearns) (1888–1965) 47 188–9
Elizabeth I, Queen of England (1533–1603) Gogol, Nikolai Vasilievich (1809–52) 153
33 Goleizovsky, Kasyan (1892–1970) 219–22,
Ellington, Edward “Duke” (1899–1974) 250 226, 239
Elssler, Fanny (1810–84) 75, 118, 121, 143, Golovin, Aleksandr (1863–1930) 212–13
179, 181, 184, 191, 196, 268, 280 Goncourt, Edmond Louis Antoine Huot de
Elssler, Thérèse (Fanny’s elder sister) 179, 191 (1822–96) 115
Elyot, Thomas (c .1490–1546) 17 Goodwin, Woël 148
Erdman, Boris (1899–1960) 219 Gorsky, Alexander Alekseevich (Aleksandr)
Essex, John (d. 1744) 80 (1871–1924) 160, 213–14, 216, 218–20,
Este, Isabella, d’ (1474–1539) 12 223
Gossec, François-Joseph (1734–1829) 100
Farrell, Suzanne (b. 1945) 227, 234, 235 Gosselin, Geneviève (1791–1818) 75
Fauré, Gabriel (1845–1924) 150, 234 Gounod, Charles-François (1818–93) 233
Favart, Charles Simon (1710–92) 87 Graham, Martha (1894–1991) 42, 224, 228,
Federova, Alexandra (1884–1972) 248 234
Feuillet, Raoul Auger (1675–1710) 43, 55, Grahn, Lucile (1819–1907) 131, 182, 280
62–3, 67–8, 73, 79–80, 84–5, 277 Grange, Marie Anne de la 87
Ficino, Marsilio (1433–99) 14, 15 Grétry, André-Ernest-Modest (1741–1813)
Fiorentino, Pier-Angelo (1806–64) 183 140
Flindt, Flemming (b. 1936) 127 Greuze, Jean-Baptiste (1725–1805) 101, 106
Fokine, Mikhail Mikhailovich (Michel) Greve, Marie-Pierre 131
(1880–1942) 146, 161, 202–4, 207–8, Grimm, Jacob (1785–1863) and Wilhelm
214–20, 223, 226, 268, 275 (1786–1859) 212
Fonteyn, Margot (1919–91) 266, 288 Grisi, Carlotta (1819–99) 115, 118, 131, 147,
Forsythe, William (b. 1949) 46, 285, 289 153–4, 179–82, 268, 280
Fournel, Victor 48 Gropius, Walter (1883–1969) 276
Francini, Tommaso 21 Guest, Ivor (b. 1920) 146
Franck, César (1822–90) 148 Guimard, Marie-Madeleine (1741–87) 62,
Franko, Mark 26, 30, 45, 46 89, 91
Frederick II of Prussia (1712–86) 87
Fridericia, Allan 135 Habermas, Jürgen (b. 1929) 175
Håkonsson, Helena Fredrika (1809–95) 131
Gad, Rose (b. 1968) 127 Halévy, Jacques-François-Fromental-Elie
Gade, Niels W. (1817–90) 136 (1799–1862) 141, 149
334 Index of persons

Halifax, George Savile Marquess of (1633–95) Jonson, Ben[jamin] (1572–1637) 34


82 Jooss, Kurt (1901–79) 42, 210, 258, 284
Hallwyl, Countess Wilhelmina von 201 Jordan, Stephanie 225
Handel, George Frederick (1685–1759) 243 Joseph II, Holy Roman Emperor (1741–90)
Hartmann, J. P. E. 132 90
Haskell, Arnold (1903–80) 266–7 Joseph, Charles M. 225, 228
Hart, Lorenz (1895–1943) 225, 230
Hasset, Caleb 34 Karalli, Vera 214
Haydn, Joseph (1732–1809) 149 Karsavina, Tamara (1885–1978) 161, 220
Heine, Heinrich (1797–1856) 115, 178, 185 Khachaturian, Aram (1903–78) 146
Heinl, Anna 89 Kinney, Troy (1871–1938) and Margaret
Henrietta Maria, Queen Consort (1609–69) (1872–1950) 215
33–4, 37–8, 39, 41 Kirstein, Lincoln (1907–96) 228, 230–1, 235,
Henry II, King of France (1519–59) 19 245
Henry III, Valois-Angoulême King of France Kolnik, Paul, (photo) 229, 235, 254
(1551–89) 19, 29 Kølpin, Alexander 128
Henry IV, King of France (1553–1610) 19, Korovin, Konstantin 213
29 Krasovskaya, Vera M. 159
Henry VIII, King of England (1491–1547 r. Kronstam, Henning 126, 136
1509–47) 32 Kshesinska, Mathilde (1872–1971) 161–3
Henry, Louis (1784–1836) 75 Kylián, Jiřı́ (b. 1947) 285
Henry, Prince of Prussia (Prinz Friedrich
Heinrich Ludwig) (1726–1802) 87 Laban, Rudolf von (1879–1958) 210, 258,
Henry, Stuart Prince (1594–1612) 33 275–7, 279, 284, 289
Hérold, Louis Joseph Ferdinand (1791–1833) Lacroix, Paul (1806–84) 30, 48
146, 149 Lagut, Irène 204
Hesse-Kassel, Landgrave of (1720–85) 34 Lalo, Édouard (1823–92) 145
Hilverding, Franz Anton Christoph (1710–68) Laloy, Louis 243–4
60–1, 88–9, 93–4, 152 Lam, Wilfredo (1902–82) 269
Hindemith, Paul (1895–1963) 231 Laminit, Paul Jakob (1773–1831) 99
Hitler, Adolf (1889–1945) 259 Lancelot, Francine 43
Hoffmann, Ernst Theodor Amadeus Landé, Jean-Baptiste (d. 1748) 152
(1776–1822) 159, 246 Lander, Harald (1905–71) 126–7
Hogarth, William (1697–1764) 176 Lander, Toni (1931–85) 131
Hope, Bob (1903–2003) 230 Lany, Mr 87
Horgan, Barbara 234 Laroche (Larosh), Herman (German
Hugo, Jean 204 Avgustovich) (1845–1904) 164–5, 167
Hugo, Victor 115–16 Lauze, François de 20, 38–9, 42, 176
Hübbe, Nikolaj (b. 1967) 126, 128, 137 Lavrovsky, Leonid (1905–67) 222
Hurault, André, Sieur de Maisse 33 Le Brun, Charles (1619–90) 108
Hurok, Sol (1888–1974) 266 LeClerq, Tanaquil (1929–2000) 227, 232
Lecuona, Ernesto 269
Isaac, Mr 78–80, 82–4 Leeder, Sigurd (1902–81) 258
Ivanov, Lev (1834–1901) 151–3, 159–61, 165, Léger, Ferand (1881–1955) 202–4, 206, 208
167, 170, 246–8, 250, 253 Legnani, Pierina (1868–1930) 161, 193
Ives, Charles (1874–1954) 228 Leibnitz, Gottfried Wilhelm (1646–1716) 46
Lenin, Vladimir Ilyich (Ulyanov) (1870–1924)
Jacobson, Leonid (1904–75) 43 218, 280
Jacoby, Ann 43 Lepeshinskaya, Olga (b. 1916) 280
James I, King of Scotland, England and Ireland Leroux, Pauline 121
(1566–1625) 33, 36 Levinson, André (1887–1933) 237, 240–1,
Janin, Jules (1804–74) 115, 178–93 243–5
Jenyns, Soame (1704–87) 80 Li Cunxin 261
Jiang Dong 261 Li Yu (1611–79) 256
Jiang Qing 261 Lifar, Serge (1905–86) 240, 242
Johnson, Samuel (1709–84) 88, 90 Lima, José Lezama 269
Jommelli, Nicolò (1714–74) 90 Limón, José (1908–72) 42
Jones, Inigo (1573–1652) 34 Lippe, Rudolf zur 46
335 Index of persons

Lopukhov, Fyodor (1886–1973) 219–23, 226, Mendelssohn-Bartholdy, Felix (1809–47) 48,


239 234
Lorenzo, Riolama 229 Ménestrier, Claude-François (1631–1705)
Louis XIII (1601–43) 19, 25, 30 24, 27
Louis XIV (1638–1715) 19, 15, 41, 46, 53, 63, Merlin, Oliver 266
66, 75, 83, 100, 120, 157, 167, 169, 243 Messager, André (1853–1929) 144, 150
Louis XV (1710–74) 91 Meyerbeer, Giacomo (Jacob Meyer Beer)
Løvenskjold, Herman (1815–70) 150 (1791–1864) 188
Lucian of Samosata (c. ad 120–after 180) 84 Meyerhold, Vsevolod (1874–1940?) 212, 239,
Lully, Jean-Baptiste (1632–87) 25, 30, 48, 63, 276
66, 83, 157, 167, 237, 243 Middleton, Thomas (1580–1627) 34
Lunacharsky, Anatoly (1875–1933) 212, 218, Milhaud, Darius (1892–1974) 208
239 Milloss, Aurel von (1906–88) 210
Lund, Thomas 127, 136 Milton, John (1608–74) 37–8
Minkus, Léon (Ludwig) (1827–1917) 142,
MacKenzie, John M. 243 148, 156, 161, 169
MacMillan, Kenneth (1929–92) 146, 287 Mitchell, Arthur (b. 1934) 233
McGowan, Margaret 30, 46 Molière (Jean Baptiste Poquelin) (1622–73)
Magri, Gennaro (2nd half 18th c.) 63, 71–4 66, 83
Makarova, Natalia Romanovna (b. 1940) 286 Monet, Claude (1840–1926) 201
Malevich, Casimir (1878–1935) 206 Monnet, Jean 87
Mallarmé, Stéphane (1842–98) 213 Montagut, Barthélemy de 38–9
Malpied, N. 62 Montès, Lola (Eliza Rosanna Gilbert)
Mañach, Jorge (1898–1961) 269 (1821–61) 183
Manzotti, Luigi (1835–1905) 213 Montessu, Pauline (1805–77) 121
Mao Zedong (1893–1976) 260–2 Morris, Mark (b. 1956) 47, 250, 289
Marcel, M. 87 Mozart, Wolfgang Amadeus (1756–91) 91–2
de Maré, Rolf (1888–1964) 201–3, 206 Murphy, Graeme (b. 1950) 250
Marguerite de Lorraine (1615–72) 29 Mussolini, Benito (1883–1945) 279
Maria Theresa of Austria (1717–80) 90 Mustel, Auguste (1842–1919) 307
Marie Antoinette, Maria Antonia Josepha Mustel, Victor (1815–90) 150
Johanna von Habsburg-Lothringen
(1755–93) 91 Nabokov, Nicolas (1904–78) 234
Marmontel, Jean-François (1723–99) 95 Negri, Cesare (1535–1604) 11, 15, 33
Marin, Louis (b. 1931) 46 Neumeier, John (b. 1942) 250, 285
Marinetti, Filippo Tommaso (1876–1944) Newton, Sir Isaac (1642–1727) 104, 188
278, 279 Nietzsche, Friedrich (1844–1900) 274–5
Markova, Alicia (Lilian Alice Marks) Nijinska, Bronislava (1891–1972) 42, 239,
(1910–2004) 263, 266, 288 288
Marlborough, Duke of (John Churchill, 1st Nijinsky, Vaclav (Vaslav) (1889–1950) 203,
Duke of Marlborough) 84 217, 220, 275, 279
Marquet, Mlle 118 Nikitina, Alice (b. 1909) 242
Martins, Peter (b. 1946) 235 Nikolais, Alwin (1910–93) 208
Marx, Karl (1818–83) 261 Noblet, Lise (1801–52) 118, 121
Mason, Francis 225, 230 Nodier, Charles (1780–1844) 122
Massé, Marie-Geneviève 43 North, Roger (1653–1734) 82
Massin, Beatrice 43 Noverre, Jean-Georges (1727–1810) 42,
Massine, Léonide (1896–1979) 239, 282 53–65, 71–4, 77, 87–97, 101, 139, 176–7,
Matisse, Henri (1869–1954) 201, 224 185, 192, 237
Mazarin, Jules Cardinal (1602–61) 22, 30 Noverre, Jean Louys 87
Mazilier, Joseph (1797–1868) 121, 150 Nureyev, Rudolf (1938–93) 286
Meck, Nadezhda von (1831–94) 157, 167
Medici, Lorenzo de’ (1449–1942) 12 Oehlenschläger, Adam (1779–1850) 132
Medici, Catherine de’ (1519–89) 21 Orgel, Stephen 46
Medici, Maria de’ 28 Ortiz, Fernando (1881–1969) 269
Medtner, Nikolai Karlovich (1880–1951)
226 Pagitt, Justinian (1628–35) 39
Mei Lanfang (1894–1961) 257 Paré, Jean-Christophe (b. 1957) 48
336 Index of persons

Parisot, Rose 64 Rameau, Pierre (1674–1748) 43, 62, 70–1


Parr, Audrey 209 Raphael (Raffaello Santi or Raffaello Sanzio)
Pasch, Johann (early 18th c.) 69 (1483–1520) 80
Pavlova, Anna (1881–1931) 162, 215, 220, Ravel, Maurice (1875–1937) 146, 234
248, 257, 275 Reber, Napoléon-Henri 145
Pedel, William 34 Reisinger, Julius (1828–92) 165
Peláez, Amelia 269 Rémond de Saint Mard, Toussaint 93
Pergolesi, Giovanni Battista (1710–36) 88 Reynolds, Sir Joshua (1723–92) 86
Perrault, Charles (1628–1703) 84, 157, 167 Rich, John (1692–1761) 81
Perrot, Jules (1810–92) 75, 114, 116, 131, Richardson, P. J. S. 87
147, 153–4 Richelieu, Armand Jean du Plessis Cardinal
Petipa, Jean-Antoine (1787–1855) 153 (1585–1642) 22, 30
Petipa, Lucien (1815–98) 114 Robbins, Jerome (1918–98) 268
Petipa, Marius (Marius Ivanovich) Robespierre, Maximilien Marie Isidor de
(1818–1910) 121, 135, 151–63, 165, (1758–94) 92, 103, 109
167–80, 193, 212–13, 216, 218, 220, Roby, Thomas 34
226–7, 237–9, 241–3, 245–7, 251 Rodgers, Richard (1902–79) 225, 230
Petit, Roland (b. 1924) 286 Rodolphe, Jean-Joseph (1730–1812) 58, 89
Piacenza, Domenico da (c. 1400–76) 13–14 Rodrı́guez, Mariano (1912–90) 269
Picabia, Francis (François Marie Martinez Rogers, Ginger (1911–95) 228
Picabia) (1879–1953) 203–5 Roig, Gonzalo (1890–1970) 269
Picasso, Pablo (1881–1973) 202–3, 224 Roldán, Amadeo (1900–39) 269
Picq, Charles Le (1744–1806) 89 Roller, Andreas 152
Piñera, Virgilio (1912–79) 269 Romanov, Alexander III (1845–95) 159
Piscator, Erwin (1893–1966) 276 Romanov, Grand Duke André (1879–1956)
Pitrot, Antoine-Bonaventure (1744–70) 73–4 162
Plato (c. 427–347 bc) 15, 93, 194 Romanov, Nicholas II (1868–1918) 162
Playford, John (1623–86) 39 Romanov, Grand Duke Sergei (1857–1905)
Plisetskaya, Maya (b. 1925) 280, 282, 286 162
Pomerantsev, Andrei 162 Ronsard, Pierre de (1524–85) 15
Porter, Cole (1891–1964) 208 Rosen, Charles (b. 1927) 84
Portocarrero, René (b. 1912) 269 Rosen, Elsa Marianne von (b. 1924) 127, 135
Prévost, Antoine François (Abbé) (1697–1763) Rossini, Gioacchino (1792–1868) 113, 145–6,
122 149
Priest, Josias (d. 1735) 40 Rousseau, Jean-Jacques (1712–78) 57, 87–8,
Prokofiev, Sergei (Serge) (1891–1953) 146 93
Propert, Walter Archibald 288 Rudolph, Jean Joseph (Rodolphe) (1730–1812)
Prunières, Henry 30 58, 89
Puget, Pierre (1620–94) 45–6 Runge, Otto (1777–1810) 189
Pugni, Cesare (1802–70) 145, 154, 161, 169 Rushton, Tim (b. 1963) 136
Purcell, Henry (1658 or 1659–95) 40, 42, 47
Pure, Michel de (1620–80) 21–2, 24, 27, 40 Sacharoff, Alexander (1886–1963) 42
Pushkin, Aleksandr Sergeyevich(1799–1837) Said, Edward (1935–2003) 253
212 St-André, Mr 40
St Denis, Ruth (Denishawn 1879–1968) 257
Quinteuse, Guillermine La 22 Saint-Hubert, M. de 24
Saint-Léon, Arthur (1815–70) 153, 177, 268
Rabel, Daniel (1578–1637) 21–2, 48 Saint-Saëns, Camille (1835–1921) 150
Rachmaninoff, Sergei Vasilievich (1873–1943) Saint-Simon, Comte Henri de (1760–1825)
226 193
Racine, Jean (1639–99) 57, 88 Sallé, Marie (1707–56) 42, 60, 62, 69–70
Raffinot, François 45 Santlow, Hester (c.1690–1773) 73
Rainer, Yvonne (b. 1934) 209 Satie, Erik (1866–1925) 203, 205
Ralph, Richard 79, 81 Sauveur, Marie-Louise 87
Ralov, Kirsten (b. 1922) 136 Scarlatti, Alessandro (1660–1725) 88
Rambert, Marie (1888–1982) 287 Schaufuss, Peter (b. 1949) 127, 129
Rameau, Jean-Philippe (1683–1764) 88, 150, Schlemmer, Oskar (1888–1943) 42, 47–8,
157, 167, 243 284
337 Index of persons

Schneitzhoeffer, Jean-Madeleine (1785–1852) Taglioni, Salvatore (1789–1868) 75, 178


140, 150 Tallchief, Maria (1925) 227, 232, 234
Schopenhauer, Arthur (1788–1860) 189 Taruskin, Richard 163
Scriabin, Alexander Nikolayevich (1872–1915) Tasso, Torquato (1544–95) 46, 148
226 Taubert, Gottfried (c.1680) 73
Seedo (Sidow or Sydow) (1700–54) 81 Tchaikovsky, Peter Ilyich (Pyotr) (1840–93)
Semenova, Marina 280 135, 142, 149, 157, 159–60, 164–74,
Semper, Gottfried (1803–79) 189 228–9, 234–5, 246–7, 249–50
Sergeyev, Nikolay (1876–1951) 220, 248, Tchaikovsky, Modeste 165
250 Tchernicheva, Lubov (1890–1976) 242
Servandoni, Giovanni Girolamo (1695–1766) Teachout, Terry 225
89 Terry, Walter 264, 266–7
Servi, de Constantino 34 Tetley, Glen (b. 1926) 285
Seurat, Georges-Pierre (1859–91) 201 Théleur, A. E. 192
Sforza, Ippolita (1446–84) 12 Thomas, Ambroise (1811–96) 150
Sforza, Tristano (1422–77) 12 Thorvaldsen, Bertel (1770–1844) 132
Shakespeare, William (1564–1616) 32, 88, Tian Han (1898–1968) 257
122, 222, 224 Tomko, Linda 43
Shawn, Ted (Denishawn) (1891–1972) 257 Tudor, Anthony (William Cook) (1909–87)
Sheng Jie 258 268, 287
Shirayev, Alexander (1867–1941) 169 Turocy, Catherine 43–4, 46
Shirley, James (1596–1666) 40
Siemsen, Hans (1818–67) 210 Ulanova, Galina (1910–98) 266, 280–1,
Sieyès, Emmanuel Joseph (Abbé) (1748–1836) 286
104 Ureña, Max Henrı́quez (1885–1968) 269
Simon, François 121
Simon, Louise 131 Vaganova, Agrippina (1879–1951) 162, 280,
Simon, Louise Antoinette 131 285
Simón, Pedro (b. 1930) 267 Valois, Dame Ninette de (Edris Stannus)
Les Six: Georges Auric (1899–1983), Louis (1898–2001) 210, 287
Durey (1888–1979), Arthur Honegger, Valts, Karl Fedorovich (1880?) 165
Darius Milhaud, Francis Poulenc Vaque-Moulin, Elisa 191
(1899–1963) and Germaine Marcelle Vazem, Ekaterina (1848–1937) 161
Tailleferre (1892–1983) 204, 205 Verdi, Giuseppe (1813–1901) 146
Slonimsky, Yury (1902) 226 Vere, Lady Susan de (1587–?) 38
Smith, Adam (1723–90) 176 Vernoy de Saint-Georges, Jules-Henry
Socrates (470–399bc) 194 (1799–1875) 116, 153, 178
Sologub, Fyodor (1863–1927) 212 Véron, Louis (1798–1867) 124, 176
Sousa, John Philip (1854–1932) 228, 233 Vestris, Auguste (1760–1842) 62, 74–5, 138,
Sparti, Barbara 30 153
Spessivtseva, Olga (1895–1991) 266 Vestris, Gaëtan (1729–1808) 58, 62, 74, 89, 91
Stanislavsky, Konstantin (1863–1938) 212, Villiers, George, Duke of Buckingham
213 (1592–1628) 37–9
Starzer, Joseph 90 Virgil (70–19bc) 80
Steele, Sir Richard (1672–1729) 78 Vitier, Medardo (1886–1960) 269
Steinlen, Théophile Alexandre (1859–1923) Vittorio Amedeo I, Duke of Piedmont
203 (1587–1637) 28
Stravinsky, Igor (1882–1971) 146, 149, 217, Vivaldi, Antonio Lucio (1678–1741) 233
224–5, 227–8, 230, 232–5, 236, 242–3 Volkonsky, Serge (Prince Sergei Wolkonsky)
Sun Yat-sen (1866–1925) 258 162
Sun Yat-sen, Madame (Soong Ching Lin) Voltaire (François-Marie Arouet) (1694–1778)
(1893–1981) 258 57, 85, 87, 94
Vsevolozhsky, Ivan Alexandrovich
Taglioni, Marie (1804–84) 75, 116, 118, 120, (1835–1909) 156–8, 162, 167, 246
131, 152–3, 155, 176, 178–80, 182, 184,
188, 191, 196, 268, 280 Wagner, Richard (1813–83) 84, 138, 141,
Taglioni, Philippe (Filippo) (1777–1871) 145, 148–50, 166
116, 126, 152 Weaver, Catherine 78
338 Index of persons

Weaver, John (1673–1760) 32, 53, 58, 61, Wu Xiaobang (b. 1906) 257
73, 78–86 Wycherley, William (c.1641–1760) 85
Weaver, John (father) 78
Weaver, Susanna 78 Xianzong (685–762) 256
Webern, Anton von (1883–1945) 225, 234
Weigel, Eva 88 Yang Yuhuan 256
Widor, Charles-Marie (1844–1937) 145, Yealina, O. A. 258
150 Yu Deling 256
Wigman, Mary (1886–1973) 258 Yu Geng 256
Wiley, Roland John 157, 250 Yu Ronglin 256
Winckelmann, Johann Joachim (1717–68)
102 Zakharov, Rostislav Vladimirovich (1907–94)
Wivel, Ulrik 137 222
Wright, Peter (1926) 250 Zorina, Vera (1917–2003) 227
Index of ballets

Abdallah 129, 136 Conservatory, The 129, 136


Acis and Galatea 215 Coppélia 138, 141–2, 144, 146, 148, 177, 239,
Adelaide ou la bergère des Alpes 95 271
Adèle de Ponthieu 90 Corsaire, Le 138, 142, 152, 159–61, 183, 258
Agon 42, 227, 233 Création du monde, La 203, 208
Alfred le grand (Paris) 139 Cupid and Death 40
Alzira 88
Annette et Lubin 95 Dance of the Lotus Fairy Maiden 257
Apelle et Campaspe 90 Dance Symphony 226
Apollo (Apollon musagète) 227, 230, 237, Dansomanie, La 149
241–4 Daughter of Pharaoh, The 153, 154, 155, 157
Après midi d’un faune, L’ (The afternoon of a Derviches 203
faun) 279 deux pigeons, Les 144, 150
Artifact 46 Diable à quatre, Le 122
Atys 48 Diable amoureux, Le 122, 145
Au temps du grand siècle/Pavane Royale 42 Diable boiteux, Le 118, 122, 141, 143
Diavolina 145
Baiser de la fée, Le 230 Dieu et la bayadère, Le 117
Ball, Le 237–40, 243 Dido and Aeneas 40, 47
Ballarino, Il 11 Don Juan 60
Balet comique de la Royne (Allegorie of Circé) Don Quixote 142p, 154–6, 160–1, 213, 217,
28, 29, 30, 55 271
Ballet de la déliverance de Renaud 21 Donizetti Variations 146
Ballet de la douairière de Billebahaut 22, 24 Dying Swan, The 257
Ballet de la Nuit (Ballet royal de la Nuit) 25,
41, 243 EL Greco 202
Ballet de Monseigneur le Duc de Vandosme 26 El río y el bosque 271
Ballet des ambassadeurs polonais 30 El solar 271
Ballet des fées de la forêt de Saint-Germain 21 Episodes 234
Barriera 10 Eucharis 144, 150
Battaglia, La 10 Fâcheux, Les 42
Bayadère, La 117, 142, 151, 154, 156, 160, Far from Denmark 130, 136
189, 239 festin de Pierre, or Don Juan, Le 60, 94
Belle au Bois dormant, La 149, 157, 167 Fêtes Chinoises, Les 56, 59, 87–8, 94
Brézilia ou la tribu des femmes 123 Fête de l’Etre Suprême 106, 107
Boeuf sur le toit, Le 208 Fête de la Fédération 98, 107
Britannicus 88 Fête de la Raison 106
Fille du Danube, La 115, 122
Calisto 40, 41 Fille mal gardée, La 61, 89, 146, 149, 258
Camargo 162 Fille de marbre, La 122
Caprice 45 Fils prodigue, Le 241
Caprices de Galathée, Les 89 Firebird, The 214, 216, 232
Castor et Pollux 61 Flames of Paris, The 285
Catarina 153 Fleur d’oranger 150
Chatte, La 177, 240 Flore et Zéphire 184
Chopiniana 196, 216, 226 Flower Festival in Genzano, The 136, 268
Cinderella 122 Folk Tale, A 129, 136
Coelum Britannicum 34, 35, 41 Fountain of Bakhchisarai, The 222, 285
Comus (A Masque Presentd at Ludlow Castle) Four Seasons, The 146
37 Four Temperaments, The 227, 231
Concerto Barocco 225 From Siberia to Moscow 136
[339]
340 Index of ballets

Gelosia 10 Merçantia 9
Gratie d’Amore, Le 11 Midsummer Night’s Dream, A 234
Green Table, The 284 Mignons et villains 150
Giselle 113–19, 122, 125, 138, 141–2, 144, Moor’s Pavane, The 42
146–7, 150–3, 155, 160, 178, 180–1, 184, Mort d’Ajax, La 89
189, 196, 263–71, 283, 289 Movements for Piano and Orchestra 234
Gounod Symphony 233 Mozartiana 234
Gridelino, Il 29
Gypsy, La 123 Naiade and the Fisherman, The 154
Namouna 145
Homme et son désir, L’ 208 Napoli 132, 134, 136, 140, 268
Hongse Nianzijun (Red Girl’s Regiment) 256, Nathalie, ou la laitière Suisse 178
258, 260 Nina ou la folle par amour 122
Horaces et des Curiaces 90 Nisida ou les Amazones des Açores 123
Nobiltà di Dame 11, 33
Iberia 203 Noces, Les 239
Idoméneo 88 Nuit de Saint-Jean, La 203
Imperial Gesture 42 nozze di Peleo e di Teti, Le (Les Noces de Pelée et
Iphigénie et Tauride 90 Thétis) 25, 243
Nutcracker, The 135, 144, 148, 150, 159, 164,
Jason et Medée 89, 91 169–73, 232, 239, 246–55, 260, 289
Jeu de cartes 230
Jeux 203 Oberon 34
Jewels 234 Offrande à la Liberté, L’ 100, 102
Job 210 On Your Toes 230
Jolie Fille de Gand, La 123, 147 Orgie, L’ 150
Jugement de Paris, Le 87, 147, 243 Orpheus and Eurydice 81, 90
Judgment of Paris, A Dramatic Entertainment in Ozaı̈ 122
Dancing and Singing, After the Manner of
the Ancient Greeks and Romans, The 79, Paquita 122, 123, 140, 150, 159
81 Parade 204, 208
Parts of some sextets 209
Kermesse in Bruges, The 129, 136 Pas de Quatre 120, 131, 182, 268
King’s Volunteers on Amager, The 130, 136 Pastorale, La 239, 240, 241
Korrigane, La 145, 150 Pavane on the Death of an Infanta 42
Pavillon d’Armide 216
Lady Henriette, ou La Servante de Greenwich Péri, La 122, 123, 150
150 Perseus and Andromeda 81
Legend of Joseph the Beautiful, The 219 Petit Riens, Les 91
Lords’ Masque, The 34 Petrushka (Petroushka) 216
Loves of Mars and Venus, The 54, 78, 81 Psyche 102
Pulcinella 239
Magic Mirror, The 212, 213 Pygmalion 60
Malade imaginaire, Le 30
Magnificence of the Universe 221 Quelques pas graves de Baptiste 43
Maison de fous, La 203, 206
Manon Lescaut 141, 149 Raymonda 142, 151, 154, 158, 159, 212
Marbre tremble, Le 45 Red Poppy, The 283
Mariés de la Tour Eiffel, Les 204, 205 Red Girl’s Regiment (Hongse Nianzijun) 256,
Mars et Vénus ou Les Filets de Vulcain 150, 184 258, 260, 261
Masque Presentd at Ludlow Castle, A (Comus) Relâche 203, 205
37 Retour des Lys, Le 149
Masque of Beauty, The 37 Réunion du 10 âout, La 100, 106, 107
Masque of Blackness, The 34 Révolte au Sérail, La (La Révolte des femmes)
Masque of the Inner Temple and Gray’s Inn, The 123, 178
36 Robert le diable (opera with the ballet of the
Masque of Queens, The 34, 35 nuns) 45, 122, 138, 188
341 Index of ballets

Roi danse, Le 43 Tabacco, Il 29


Romeo and Juliet 222, 283, 285 Tavern Bilkers, The 78, 81
Télémaque 102
Sacre du Printemps, Le 146, 233 Tempe restored 38
Salmacida Spolia 37 Tempête, La 150
Schéhérazade 214, 216 Tempête ou l’ı̂le des génies, La 122, 140
Secular Masque, A 40 Tentation, La (opera ballet) 117, 181
Séducteur du village, Le 150 Thésée en Crète 94
Sémiramide 60, 94 Three Atmospheric Studies 289
Serenade 228–30 The Triumph of Neptune 240
Skating Rink 203, 206, 248 The Triumph of Peace 37
Sleeping Beauty, The 42, 122, 151, 154, Toilette de Vénus, La 59
157–159, 164, 167, 169, 173, 184, 213, Toreadoren 132
218, 220–1, 226, 239, 246–7, 249, 283 Torneo Amoroso 9
Sobria 9 Triadic ballet (Triadisches Ballett) 42,
Somerset Masque, The 34 284
Somnambule ou L’Arrivée d’un nouveau seigneur,
La 140, 149 Une demi-heure de Caprice 216
Song of Norway 231
Songoro cosongo 271 Valse fantaisie 216
Source, La 146, 148 Ventana, La 136
Spartacus 285 Vert-Vert 150
Spectre de la Rose, Le 214, 216, 279 Vestris 43
Square Dance 233 Vins de France, Les 150
Stars and Stripes 233 Vivandiére, La 268
Swan, The 215, 216, 257 Volière ou les oiseaux de Boccace, La 123
Swan Lake 142, 149, 153, 158–9, 164–7,
173–4, 193, 239, 246, 256, 258, 260, 263, Wedding Masque in The Tempest, A 32
271 Whirlwind, The 219
Sylvia ou La Nymphe de Diane 148, 149 White Haired Girl 258, 260
Sylphide, La 115, 118, 122, 126–30, 136–7, Who Cares? 234
150, 152, 155, 184, 196, 244, 268 Within the Quota 208
Sylphides, Les 216, 226
Symphonie pour un homme seul 286 Zémire et Azor 150
Subject index

absolutism 37, 46, 53, 55, 132 Autre pas, L’ 45


absolute ruler 109 avant-garde 201, 204, 211, 226, 275–7
absolutist period 55 Azerbaijan 172
abstraction 47, 221, 222, 280, 284, 285
Academy bal musette 205
Académie Royale de (la) Danse 40, 53, 61, balance (in dance) 70, 72, 74, 130, 239,
62, 66, 70, 76, 83, 123 244
Académie de Poésie et Musique 23, 243 ballerina
Academy of Fine Arts Copenhagen 133 Balanchine 227, 239, 240, 241
Académie Française 22, 84 beauty 177
Academy of dancing 76 (in) Britain 287
Beijing Academy of Dance 258 (in) cartoons 191
Imperial Naval Academy 226 celebrity 176
Shanghai Academy of Dance 258 Chinese 257, 258, 261
aesthetics 46–7, 49, 56, 95, 210, 248, 268, “Christian” 155
271, 275 commodification 119, 121, 124
anti-Aristotelian 55 costume 132, 190, 195, 196, 197, 224
Aristotelian 56, 59 critic 182, 184, 193
programmes 64, 176 Cuban 263–71
Affektenlehre 89 end of 19th c. 155
aire 14 eroticism 180
Albrecht (Giselle) 142, 153, 266 femininity 179, 267
alchemy 15p foreign 161, 163, 213
Alfred le Grand 139 image 192
allegory 19, 20, 21, 46, 49, 96 in tutu 30
Giselle as allegory 119 hierarchy 158, 159
Allemagne, De l’115 Italian 114, 155, 157, 213, 247
Allied Forces 284, 285 (as) machine 155
almain 9 (and) male dancer 241
American Ballet, The 227, 228, 230 modern/new 47, 121, 155, 193
Aminta 148 (and) music 146
Amsterdam 81 mystique 155
anatomy 96 on pointe 74, 193, 214
Ancien Régime 90, 98, 100–2, 123 Petipa’s 154
antimasque (antic) 34, 36–9 prima 161, 225
Aphrodite 240 romantic 118, 152, 175, 196, 266
aplomb 72 Russian 151, 248, 250, 257, 280, 281, 282,
Apollo 25, 242, 243, 244 286
Arabia 172 social role 119
aristocracy 17, 63, 100, 169 Soviet 214, 280
amateurs 40, 53, 66 technique 192
movement 100 virtuosity 159
norms 99 visiting in Russia 161
performers 34, 37 ballet
Armenia 172 d’action 53–64, 75, 81, 88, 93, 101, 123,
art pour l’art, L’ 283–6 139, 152, 159, 182, 242, 243
Arts Florrisants 48 anacreontic 39
assimilation (China) 256 blanc (white) 115, 118, 122, 155, 180, 185,
Ausdruckstanz 210, 275 189, 241
Australia 250 burlesque 30
Australian Ballet, The 250 classical see classical dance
[342]
343 Subject index

court (ballet and dance) 19–31, 37, 39, 42, Carmagnole 99, 100
53, 89 Carmen 148
exercise 65, 71–2, 79, 100, 192, 239 cascarda 9
féerie 21, 22, 157–8, 163 celebrity, see star system
grand 19, 20, 37, 53pp, 61, 64, 154, 212 chaconne 62, 73
lesson 65, 69, 71pp, 76, 106, 147, 184 Chérie 115
pantomime 24, 32, 39, 44, 54, 59, 60, 73–5, Chinese National Ballet 127
80, 87–8, 91, 93–6, 101, 113–14, 116–7, choreo-drama 213
119–22, 132, 138–42, 147, 155, 158, 164, Christianity
173, 177, 185–6, 206–8, 210, 214, 216, Christian values 129, 134
222, 232, 234, 240 Danish 132
purity of 164 and earthliness 118p
terminology 39, 69, 193, 209 and ethereality 118
tragic 89, 91 and image of dancer 155
Ballet de XXe siècle 286 Church Fathers
Ballet Society 230–2 attitude to dance 80
ballet theatre 219, 222, 263, 266 circus 205, 225, 239
balletomane 152, 180 citoyen, citizen 13, 33, 37, 98, 104, 105, 107,
Ballets, Les 1933 228 127, 132, 190, 262
Ballets de l’Opéra 43 Civil War 37, 40, 222
Ballets Russes 163, 201, 205, 212, 216, 218, Cleveland 285
227–8, 238, 242–3, 248, 278, 282 Cold War 285, 286
Ballets Russes de Monte-Carlo 228, 248, 282 Compagnie l’Eventail 43
Ballets Suédois 201–11, 278 cotton 189
ballo, bassedanze 9–13, 17 industry 187, 196
Bastille 98, 99 Clara, see Marie (The Nutcracker)
Battle Symphony 148 class (social) 13, 83, 90, 91, 101, 102, 120,
Bauhaus 204, 284 123, 132, 161, 187, 188, 257, 258, 261
benefits (see also salary) 249 classicism 102, 151, 156, 159, 185, 227, 229,
Berlin 45, 48, 87, 176, 281, 285 237–9, 263–6
bharata natyam 251 anti-classicism 237
black swan 159, 160 neo-classicism 84, 102, 227, 243
Blenheim 84 clothes, dress 9, 12, 14, 100, 102, 188, 194,
bolero 144, 148, 239 231, 243
Bolshevism 272, 280 Cochran’s Revues 228
Bolshoi Theatre (also Kammeny Theatre, Coliseum (London) 250
see theatre) comedy 85
bonnet rouge 105 comédié ballet 30, 40, 41
Book of Job 210 commedia dell’arte 28
Bordeaux 61, 89 commercialisation of opera 175
brando 9 commodification of dancer 119, 121, 124–5
branle 9, 20, 24–5, 38 communism 256, 262, 270, 280
Broadway 228, 230, 263 concert 82, 113, 147, 214, 215
Brussels 15, 117, 153 concert suite 247
Buenos Aires 267, 282, 289 Constitution of France 103
Congress on Research in Dance 44
cachucha 143, 183 constructivism, constructivist 204, 209,
Café anglais, Le 113 219
California 251 contre danse 84
cancan 143 Copenhagen 127, 129–30, 132–3, 138, 140,
Cancans de l’Opéra, Les 182 176, 246
Cakewalk 239 Coppelius, Dr (Coppelia) 148
campeggiare 14 corps de ballet 57, 62, 75, 76, 120–1, 125,
Canada 246, 250 132, 159, 248, 266, 280
canon (music) 173 Corsaire-Satan, Le 183
capriole 11, 20, 54, 67, 73, 74 corset 22, 101, 155, 186–7, 194, 239, 275
Carabosse (The Sleeping Beauty) 158, 168, coryphées (rank of ballet dancer) 158–9
170 cosmos, order of 14, 104
344 Subject index

costume 19, 21, 25, 28–9, 32–4, 42, 44, 47–9, manuals 9, 16, 33, 101, 176
58–61, 69, 89, 91, 96, 98, 101, 107, 113, modern 42–3, 47, 49, 136, 180, 195, 206–8,
116, 144, 152, 160, 162, 176, 184–6, 204, 216, 228, 239, 257, 273, 276, 278–9,
206, 208, 212–13, 215, 219, 224, 231–2, 283–7
237, 241, 253, 258, 266, 273, 284 national 120, 138, 143–4, 156, 158, 183,
courante 20, 38 240, 242, 253
court 9–11, 22, 28–9, 42, 46, 63, 69, 78, 99, natural 11, 54, 58, 88–90, 96, 99, 100–2,
218, 277 126, 194–5, 275
Chinese 256–7, 261 noble 17, 20, 39, 69, 73, 101, 180
Danish 132 notation 43, 55, 60, 63, 65, 67–8, 73, 80,
English 32–41, 78, 82–3 85, 213, 220, 250, 276–7
French 15p, 19–25, 28–9, 63, 75, 87, on pointe 71, 74, 114, 119–20, 121, 155,
100–1, 120 180, 192–3, 214, 272, 231–2, 239–40, 244
German 28, 89, 186 275, 283,
Imperial Russian 152–3, 158, 163, 212, 280 popular dances 62, 74, 100, 116, 142
Italian 11–2, 21 representation 57, 78, 122, 181, 194, 285
Savoy 28, 29 serious 59, 69, 73–4, 79
Viennese 88 simple 27, 56
court theatre 154, 212 social 20, 24, 27, 38–9, 62–3, 68, 143, 239
Hermitage, the 154, 212 spectacle 15, 20p, 40, 46, 54, 56–7, 98–100,
Kammennyi Ostrov 154 109, 117, 120, 122, 126, 155, 158, 177,
Krasnoe Selo 154 181–3, 203, 218, 231, 237
Peterhof 154 studio 69, 70, 92, 133, 224, 232, 281
Tsarskoe Selo 154 style 158, 172, 192, 216, 219
courtesan 132 symphony 221, 226
Cuba 263–71 theatrical 15, 38, 53, 56, 65, 68–9, 74–5,
czárdás, Csárdás 144, 239 119–20, 143, 285
Czechoslovakia 285 themes 10, 20, 23, 28, 55, 59, 96, 119, 141,
148, 168, 202, 208, 218, 229, 231, 252
Dada, Dadaists 204, 205, 283 Dance in America 232
Dallas Ballet 127 Danish Golden Age, the 132–3, 136
dance danse d’école 53–4, 62, 65, 120, 227, 237,
abstract 232, 280 239–40, 244–5
baroque 42–4, 46–9 Directory (Directoire) 92, 109, 214
character 59, 69, 143, 172, 239–41, 248, divertissement 23, 27, 53, 55–6, 92, 100,
253 142–3, 150, 158, 213, 216
class 12, 65, 66, 69–70, 72, 75, 176, 193, Dom Sebastien 146
224, 228, 257 drama 11, 23, 32, 39, 42, 59, 65, 88, 93, 101,
classical 19, 28, 30, 42–3, 47, 50, 65, 80, 106–7, 122, 145, 147–8, 152, 168, 170,
102, 151, 153, 156, 158, 169, 205–6, 207, 185, 202, 204, 208, 210, 213, 216, 222,
214, 224, 227, 236, 237–45, 246–7, 252, 225, 258, 260, 269, 274, 286–7
254, 259, 268–9, 271, 282 dramaturgy 55, 60, 214
comic 59, 69, 73–4, 147, 249 Drambalet 222
country 37–9, 84, 250 Dresden 34, 73, 87
court, courtly 9–11, 42, 46, 48, 53, 56, 66, Drosselmeyer, Drosselmeier (The Nutcracker)
84, 89, 98, 100–2, 218, 243, 256, 273 148, 170, 246, 249–50
drama 32, 59, 65, 101, 106–7, 116, 122–3, Drury Lane Theatre, see theatre
145, 147–8, 185, 202, 208, 210, 214, 216, Duchy of Wurttemberg 91
222
en travestie 179 Eclat des Muses, L’ 43
figured 25, 27 education 11–12, 17, 56, 62, 106–7, 213, 218,
folk dance 99, 120, 152, 218–19, 239 271, 279, 285
grotesque 69, 91 elevation 73–5, 121, 230
heroic 59, 69 elite 9, 12–13, 17–18, 82, 100, 103, 158, 257,
history 17, 28–9, 31–2, 48–9, 54, 78, 81–6, 261
94, 147, 225, 256, 272, 278, 284 emigration, immigration 162, 222, 246,
horizontal 25 249–50
injuries 71, 228 enchaı̂nements 72
345 Subject index

Encyclopédie 53, 93 gymnastics 134, 237, 239


English National Ballet 250 gypsy 144
Enlightenment 53, 57, 78–80, 85, 89–90, 93,
99, 102, 106, 212, 218, 239 Hall of the Petit Bourbon 25
entrée 24, 34, 37 Hamburg 210, 285
equilibrium 72, 108–9, 273 Harlem 250
Eucharis 144, 150 harmony 14–15, 25, 27, 30, 72, 77, 88, 95,
Eurocentrism 268, 270, 271 130, 142, 233, 252, 276
Expressionism 202, 206, 272 Havana, La 263, 268–9
Nouvelle Hélose, La 93
familialism and family 107 Hermitage, the (see court theatre)
Fanfarlo, La 182, 183 Hilarion 114, 116, 141
Fantasia 248 Himalayas, the 156
fantasmata 14 historiography 19, 28, 193
Fantômes 115 Histoire de l’art dramatique en France depuis
Faun (L’après midi d’un faune – Afternoon of a vingt-cinq ans 177
faun) 89, 279 Hollywood 228, 230, 263
féerie 21, 122, 155, 157–8, 163 Holy Roman Empire 90
fête 15, 16, 43, 56p, 87–8, 94, 98, 106–7 Homme de qualité 20, 27
Fêtes galantes 43 Huguenots, Les 145
feuilleton 175–6, 180, 182 humanists 14
figura (figures) 14–15, 26–7, 39, 41
Finnish National Ballet 136 ideology 197, 261–2, 271, 283–4, 286
Florence 12, 21, 33 identity 22, 38, 49, 59, 123, 125, 137, 159,
Fontainbleau 87 163, 264
Folk dance, folk performance 99, 120, 152, national 263, 265, 267–9, 270–1
218–19, 239 Imperial Ballet 151–2, 156–9, 161, 163, 169,
Foxtrot 239 193, 212, 280
Foyer de la danse 113, 124 Imperial ballet school 151, 161, 280
France Musicale, La 147 Imperial drama troupe 158
Frankfurt 285 Imperial Russian Theatres (see theatre)
Franz (Coppelia) 148 Imperial Theatres of St Petersburg (see theatre)
futurism, futurist 205, 278–9 imperialism 256, 285
impresario 72, 203–4, 211, 263, 266
galliard 9–11, 38 industrialisation 133, 187
gender 46–7, 103, 105, 107–8, 122–3, 125, Inns of Court 32
187–8, 190, 195, 197 Inoue Ballet 127
geometry (geometric design) 15, 15–17, 23, instrumentation 147, 165–7, 170, 174
37, 46, 55, 100, 206, 208, 239, 284, 279 intermedii – see divertissement
Georgia 172, 286 irony 20, 27, 180, 195
gesture 12, 14, 17, 23, 25, 42, 44, 47, 49, 54,
59–60, 77, 87–8, 90, 92–3, 100, 103, 108, Japan 257–8, 261, 289
116, 119, 140, 144–5, 181, 105–7, 210, jazz 208, 227–8, 249, 251
216, 222, 227, 240, 259, 284 jig 143, 148, 239
Giselle 114, 116–22, 142, 147, 180, 269–71 Jockey Club 138
Glorious Revolution 82, 85 Journal des débats 176, 178
god, gods, goddess 14, 25, 55, 61, 89, 93, 100, July Monarchy 117, 119
134, 240, 242–3 jumps 11, 14, 43, 63, 67, 71, 91, 99, 102, 108,
in honour of 17 194, 240
of love 29
of Restauration 38, 41 Kammeny Theatre (Bolshoi Theatre), see court
understanding nature of 14, 18 theatre
Grand Ballet du Marquis de Cuevas 127 Kärntnertor Theatre, see theatre
Greek drama 23 Kenilworth 33
Greek (Hellenic) myth 55 King’s Theatre, see theatre
gratia 14 Kirov Ballet 151, 161, 220, 247
guerre des buffons 88–9 kolkhoz 280
Guillaume Tell 145 krakoviak 144
346 Subject index

Kammenyi Ostorov, (see court theatre) Milwaukee 285


Krasnoe Selo, (see court theatre) mime, see pantomime
Ministry of the Court 152
Lausanne 87 Minstrel’s Guild 66
leitmotif 141 minuet 73, 144, 169
Leningrad 220 mirror (in dance studio) 70, 113
Leningrad State Choreographic School 280 misura (measure) 11, 13, 38
Leto 243 modernism 42, 47–8, 50, 151, 201, 208, 211,
Leyden 34 272, 276, 286
liaison (of dancers) 162 radical modernism 133
liberal arts 13 Moı̈se 113
liberalism 285, 286 monarchy 19, 33, 37, 117, 119
libretto 19, 26, 29, 93–4 Monde, Le 266
Lichfield 88 Moor, The (Petrushka) 216
Lilac Fairy, the (The Sleeping Beauty) 168, Moscow 138, 152, 157, 164, 213–14, 219–20,
220 246, 287
Lincoln Center 234 Muette de Portici, La 117
Lincoln’s Inn Field Theatre (see theatre) muse, muses 43, 119, 147, 184, 237, 242–3
Lohengrin 166 musette 62, 205
Lombardy 90 music
London 34, 39, 41, 60, 69p, 73, 78, 82, 84, 94, Balanchine and 225–6, 233
117, 131, 176, 182, 190, 207, 228, 248, baroque 46–9
257, 287 baroque revival 47, 49
Inn’s Court 32 Cuban 270
Noverre in 57, 88–90, 92 and dance/movement 11, 14, 17, 27, 32,
population 81 33, 60, 65–6, 77, 84, 157, 172, 208, 214,
public sphere 82 225, 230–2, 242, 248
Weaver in 78–81 Duncan and 275
London Festival Ballet 127 Marinetti and 279
Los Angeles 263 musical score 19, 21, 39, 42, 60, 138,
Louvre 66 140–1, 145–6, 154, 157, 165–7, 209, 212,
Lyons 87–9, 92, 94 217, 221, 243, 250
Nutcracker 169–73, 173, 246–8
machine (theatrical machinery) 21, 55, 119, and other arts 13, 23, 34, 55, 204–5, 216,
152, 155, 190, 218, 240, 279 218, 225, 227, 242
Magna Carta 105 Sleeping Beauty 167–9, 173
maniera 14 Swan Lake 164–7, 173
Manon Lescaut 122, 141, 149 music hall 205, 220
march 239 Musical World, The 180, 182
Marie or Clara (The Nutcracker) 135, 170–1, Myrtha (Giselle) 114, 119, 189
146–7, 249-52 mythology (mythological) 19–20, 22, 55,
Marseillaise 100 95–6, 242
Marseille 87, 153, 251
Marshall Plan 285 Nantes 153
Maryinsky Ballet 151, 248–9 Naples 63, 73, 75, 134, 267
Maryinsky Theatre (see theatre) narrative (see also plot) 19–21, 42, 47, 53–64,
mask 19–22, 24, 32, 34, 87–8, 96, 204 93–4, 125, 139, 190, 196, 221–2, 232, 240,
masque 15–16, 23, 32–41 253, 273, 280, 283
masquerade 23 National Ballet of Cuba, The 268, 270, 271
mazurka 144, 150, 169 National Dancers Association 258
melodrama 101, 116, 123, 154, 214 National School of Ballet Alicia Alonso, The
melody 88, 141–3, 145, 147, 164, 166, 168, 269
170, 172–3, 267 nationalism 84, 123, 157, 163, 268, 269, 272
memory 11–12, 134 nature 54–5, 57, 59, 71, 83, 98, 101, 103–6,
Merchant Taylor’s Hall 37 109, 194, 275
metaphor 16, 19, 107, 155, 228 “back to nature” 57–9
Middle East 144, 172, 253 belle 55–7, 95
Milan 12, 33, 76, 90, 138, 176, 190, 287 code of naturalness 99
347 Subject index

imitation of 60, 84, 93, 95 ordonnateur 21


natural body 99, 100, 101 Orientalism 253
natural human feelings 101 Ottoman Empire 172
laws of 105 Oxford 78
natural movement 99
natural way of life 99 Palais Royal 63
Nazism 272, 280, 285 pantomime 24, 32, 39, 44, 54, 59, 60, 74–5,
Nederlands Dans Theatre 285 80–1, 83, 87–8, 91, 93–5, 97, 113–15,
New York 43–4, 46, 128, 234, 249, 263, 283, 121–3, 138–40, 164–7, 179–80, 205,
285–6 207–8, 216, 222
New York Baroque Dance Company 43, Paris
46 audience in 196
New York City Ballet 225, 231–2, 249 ballet in Paris 28, 91, 141, 146, 216
New York Herald Tribune 264 ballet world 91
Nô Theatre 286 ballet critic 175–83
nobility 9, 12, 28, 70 cosmopolitan generation 203
Notre-Dame cathedral 98 opera, Opéra 56–9, 62, 69–70, 74, 89,
Novantiqua 45 91–2, 106, 113, 116, 125–6, 130, 138, 143,
nozze di Figaro, Le 139 145, 153, 176, 196, 276
nudity 47, 133, 195 poor quarters 62
nuns (Robert le diable) 115, 122, 188–9 refugees 281
Nuremberg 13 stages of 190
nymphs 32, 35, 37, 148 theatres 40, 98
1920s 205
Odette (Swan Lake) 159–61, 166, 282 Parnassus 243
Odile (Swan Lake) 159–61, 166 parody 240, 241, 245
Oliver Twist 250 pas 72, 116, 129
Olympus 243 grand pas d’action 158–9
ondeggiare 14 grand pas de deux 143, 158–9
opera 40, 43, 47, 64, 90, 92, 100, 116–17, pas de balet 24
138–41, 152, 155, 166, 175, 177, 190, 196, pas de bourrée 123
260, 269, 272–3, 280, 282, 285 pas classique 159
18th c. 42 pas de deux 114, 120–1, 136, 143, 146, 150,
buffa 88 159, 166–7, 234, 239–40, 241, 243–4,
comic 122, 139, 147, 150, 191 247–8
comique 122 pas de trois 120
féerie 21, 155 pas de quatre 121, 123, 182, 189, 268
grand 188 pas joueux 99, 108
house 63–4, 175, 186, 195, 284 passepied 62
Italian in Paris 88 pastiche 47, 237, 243
opera-ballet 60, 243 pavane 9, 11, 25, 42
Opéra de Monte-Carlo 227 peasant 16, 61, 95, 114, 136, 142, 150, 261,
Restoration 32 280, 286
Paris Opéra, Opéra Garnier 56–9, 62, 69, Peking 257
73–5, 89, 91–2, 106, 113, 116, 125–6, 130, Persian Empire 172
138, 145, 176, 196, 276 Peterhof (see also court theatre) 154
Paris Opéra-Comique 56, 59, 69, 87–9, Petrushka 216
147 Picardy 87
Peking 257 Piedmont 28
Royal Swedish 202 pirouette, see turns
State Academic Theatre of Opera and Ballet Philadelphia 138
220 philosophy 107, 228, 236, 240, 269–70, 275,
Vienna Royal 56, 59, 63, 267 286
Vienna State 267 Pléiade 23
opium 256, 262 plot 19–20, 55, 58–9, 61, 88, 92–4, 114,
orchestra 138–41, 166, 168, 172–3, 187, 210, 116–19, 122–3, 139, 141, 144, 154, 156,
228, 230, 234, 273, 285 167, 170, 176–7, 180, 188, 196, 221, 225,
orchestration 144, 148, 157, 165, 172, 174 229, 231, 240, 246–7, 252
348 Subject index

plotless 216, 221, 225–6 Romanovs, The (Russian Imperial family)


port de bras 159, 162
positions of feet 62, 66, 70–7, 216, 228, 240, romanticism 116, 128, 136, 151, 184–7, 188,
244 205
Porte-Saint-Martin Theatre, see theatre Rome 33, 85, 103, 194
post-modernism 49, 209 Rothbart (Swan Lake) 160, 166
post-structuralism 46 Rouen 74
Presse, La 177 Royal Ballet, The (London) 146, 250, 287,
programme notes 54, 94, 177, 224 288
proletariat 260, 261, 280, 286 Royal Ballet School, The (London) 285
propaganda 20, 278, 286 Royal Danish Ballet, The 126–8, 133, 136–7,
Protestantism 84 228
psychology 274, 276 Royal Swedish Ballet, The 127, 202
public 16, 20, 22p, 37, 43, 45, 62–3, 155, 155,
157, 179, 182, 188, 216, 224, 231–2, 273, Saint-Germain-en-Laye 87, 92
289, 290 Saint Matthew Passion 48
Öffentlichkeit, public sphere 81, 82, 83, St Petersburg 73, 92, 95, 117, 151–8, 161,
119, 123, 175, 272 165, 167, 169, 190, 212–13, 221
performance 12, 17, 82, 179, 257 Saisons Russes 212, 234
public space 9, 13, 104–5, salary (benefits) 151, 234
public taste 54 Salle du Louvre 25
saltarello 13, 14
“Rage Over a Lost Penny” 148 San Francisco 248
realism 125, 156, 164, 280, 285 San Francisco Ballet 248
reform 53, 59, 61, 87–97, 185, 195, 214, 219, Saratoga Springs 234
274, 276, 278 Scandinavian Ballet, The 127
Regency 55 School of American Ballet 227, 228
rehearsal 12, 92, 113, 128, 159, 169, 229, 231, secularism, secularisation 20, 278
247 sexuality 47, 121–2, 125, 133–4, 155, 258,
religion 84, 226 275, 279
Renaissance 13, 48, 53, 85, 176, 250 Shanghai 257–8, 289
Répétiteurs (rehearsal notes) 117 shoe 32, 74, 89, 101, 183
Republic 98, 105 blocked/pointe 155, 185–99
Civil 102 Shrewsbury 78, 79
French 98, 103, 105 Shropshire 78
Roman 103, 107 Siegfried (Swan Lake) 159–60, 166–7
Soviet 218, 221 Siren (The Prodigal Son) 241
Restoration 32, 41 social class of dancers 161, 187, 188
revels (also masque) 154 social contract 104
Revolution Society of Dance History Scholars 44
Chinese 257, 260, 262 Song of Norway 231
Cuban 270, 271 South Africa 285
Cultural 222, 257, 260, 261 Soviet bloc 161, 285
Culture 260, 261 space
French 61, 90, 92, 98–109, 149, 186, 190, architectural 273
195, 221, 272, 278 cultural 48
Glorious 82, 85 dancing/performance 9, 12, 37, 99, 120,
in dance 97, 161, 180, 184, 190, 222 130, 205, 275–8
Russian 152, 202, 214, 218–19, 220, 222, fraternal 103–6
226, 238, 248, 279–81 gendered 107
technical 192, 272 needed for lessons 76
Revue et gazette musicale de Paris 150, 177, 181 organisation of 26
rhythm 25, 34, 133, 140, 148, 164, 166, 168, private 124
233, 243, 250, 263, 267 public 9, 18, 104, 105
Richard Coeur de Lion 140 stage 49, 175, 179, 183
Ris et danceries 43 studio 70
Robert le diable 115, 122, 138, 188 theory of 276
349 Subject index

Spectator, The 78, 207 Bolshoi Theatre, Bolshoi Experimental


spectatorship 181 Theatre 152, 154, 214, 219, 220, 239
stage Coliseum 250
stage/costume design 19, 28, 33, 34, 42, 47, Colón 267
152, 187, 194, 203–4, 206, 208, 212, 216, Comédie Française 75
237 Comédie Italienne 73
flying machines 190 Court theatres 154, 282
stance of dancer 70, 74 Drury Lane 54, 73, 78–80, 88–90
star system 175, 182 Imperial Russian Theatres 138
Stockholm 202 Imperial Theatres of St Petersburg 152,
Strasbourg 87, 88 154, 156–7, 163, 167, 213
studio 69, 70, 92, 133, 224, 232, 281 Kärntnertor Theatre 91
Stuttgart 42, 56, 89, 92, 285 Kammertanz Theater 210
style 250 King’s Theatre 57, 92
absolutist 92 Lincoln Inn’s Field Theatre 81
Balanchine 228 Maryinsky Theatre 152, 154, 156, 159,
ballet 73, 210 165, 167, 170, 213, 220, 226, 239, 246–8
Chinese 258 Moscow Art Theatre 212, 213
dance 74, 158, 172, 192, 216, 219 Musical de Paris-Châtelet 251
French 180 New York State Theater 234
Indian miniature 203 Opéra Comique Faire St Laurent 87
Louis XIV 157, 167 “plastic” 205, 208
movement 100, 101, 231, 240 Porte-Saint-Martin 183
personal 285 Royal Theatre Copenhagen 126
romantic 267 San Carlo Theatre 76, 267
Russian 207, 281 Teatro Regio Ducale 90
Soviet 239 Théâtre des Champs Elysées 201
theatrical 69, 88, 210 Théâtre de la Gaı̂té 155
Sugar Plum Fairy (The Nutcracker) 171, Theatre Royal in Drury Lane (see Drury
174, 247 Lane)
Swanilda (Coppelia) 141, 148 theory
Swedish Ballet, see Ballets Suédois aesthetic 175
sylph 115, 118, 123, 126, 128, 137, 152, of art 59, 79, 95
155, 188–9, 196 baroque 46
sylphide 118, 122, 126–7, 131, 136–7, of dance 53, 57, 101, 274–5
186 drama 101, 107
symbol 14–16, 18, 21, 25, 29, 67, 102, enlightenment 106
105, 107, 109, 194, 243, 284, 286 Marxist 261
symbolism 156, 189 Öffentlichkeit 175
symmetry 21, 57, 104, 213, 238–9 post-structuralist 46
of space 276
tambourine 62 Times, The 216
tableau mouvant 106 Tobago 29
tableaux vivant 106, 107 Toledo 202
Tannhäuser 138 tonality 243
Tarantella 144 Tortoni 113
technique 20, 33, 43, 45–6, 48, 56, 61–2, total work of art 84, 85
65–77, 101, 119, 121, 123, 151, 155–6, tragedy 85, 95, 166, 205, 274
172, 175–6, 180, 185, 191–4, 196, 208–9, training, class 11, 12, 62–3, 65–70, 72, 74–5,
219, 224–5, 233, 236–7, 240, 257–61, 77, 96, 120, 133, 152, 159, 161, 190, 192,
267, 269, 273, 276, 283–4, 290 234, 240, 249, 275–6, 280, 290
Telegraph Boy (Le Pastorale) 240, 241 Triel 92
Tentation, La 117, 181 Trilby ou le lutin d’Argail 122
Tempest, The 32, 122 Trinidad 257
tempo 142, 147, 157, 166, 243 Trust, George Balanchine Trust 236
Terpsichore 184, 242, 243, 244 Tsarskoe Selo, see court theatre
theatre Turin 28
350 Subject index

Turkey 172 virtuosity 10, 20, 24, 55, 70, 73–4, 120, 148,
Turks 144 155, 164
turnout 70–1 voyeurism 124, 175
turns, pirouette 10, 11, 20, 38, 62–3, 71–2,
74, 89, 155, 160, 180, 192–3, 244 Wagnerianism 145
tutu 120, 184, 190, 193–5, 215, 239–40, 243, waltz 114, 136, 142, 144, 148, 150, 158, 160,
251 166, 169, 171, 173, 177, 180
Wehrmacht 284
urbanisation 277, 278 Weltbühne, Die 210
urban landscape 116, 216 Whitehall 33, 37
urban life 133, 206 Wilfride (Giselle) 142
urban society 274, 276 wili 113–14, 118–19, 123, 144, 178, 185, 189,
214
variation 10, 59, 63, 158–9, 166–9, 220, 229, woman question 275
231, 240, 243, 244 World War, First 201p, 208, 248, 275, 277
Venice 12, 63 World War, Second 245, 280, 284
Vêpres siciliennes, Les 146
Versailles 63, 82, 89 Young Ballet 221, 226
Vienna 56, 82, 88–90
Royal Opera 56, 59, 63, 267 Zen Buddhism 286
State Opera 267 Ziegfeld Follies 230
The Cambridge Companion to Music

Topics
The Cambridge Companion to Ballet
Edited by Marion Kant
The Cambridge Companion to Blues and Gospel Music
Edited by Allan Moore
The Cambridge Companion to the Concerto
Edited by Simon P. Keefe
The Cambridge Companion to Conducting
Edited by José Antonio Bowen
The Cambridge Companion to Grand Opera
Edited by David Charlton
The Cambridge Companion to Jazz
Edited by Mervyn Cooke and David Horn
The Cambridge Companion to the Lied
Edited by James Parsons
The Cambridge Companion to the Musical
Edited by William Everett and Paul Laird
The Cambridge Companion to the Orchestra
Edited by Colin Lawson
The Cambridge Companion to Pop and Rock
Edited by Simon Frith, Will Straw and John Street
The Cambridge Companion to the String Quartet
Edited by Robin Stowell
The Cambridge Companion to Twentieth-Century Opera
Edited by Mervyn Cooke

Composers
The Cambridge Companion to Bach
Edited by John Butt
The Cambridge Companion to Bartók
Edited by Amanda Bayley
The Cambridge Companion to Beethoven
Edited by Glenn Stanley
The Cambridge Companion to Berg
Edited by Anthony Pople
The Cambridge Companion to Berlioz
Edited by Peter Bloom
The Cambridge Companion to Brahms
Edited by Michael Musgrave
The Cambridge Companion to Benjamin Britten
Edited by Mervyn Cooke
352

The Cambridge Companion to Bruckner


Edited by John Williamson
The Cambridge Companion to John Cage
Edited by David Nicholls
The Cambridge Companion to Chopin
Edited by Jim Samson
The Cambridge Companion to Debussy
Edited by Simon Trezise
The Cambridge Companion to Elgar
Edited by Daniel M. Grimley and Julian Rushton
The Cambridge Companion to Handel
Edited by Donald Burrows
The Cambridge Companion to Haydn
Edited by Caryl Clark
The Cambridge Companion to Liszt
Edited by Kenneth Hamilton
The Cambridge Companion to Mendelssohn
Edited by Peter Mercer-Taylor
The Cambridge Companion to Mozart
Edited by Simon P. Keefe
The Cambridge Companion to Ravel
Edited by Deborah Mawer
The Cambridge Companion to Rossini
Edited by Emanuele Senici
The Cambridge Companion to Schubert
Edited by Christopher Gibbs
The Cambridge Companion to Schumann
Edited by Beate Perrey
The Cambridge Companion to Sibelius
Edited by Daniel M. Grimley
The Cambridge Companion to Verdi
Edited by Scott L. Balthazar

Instruments
The Cambridge Companion to Brass Instruments
Edited by Trevor Herbert and John Wallace
The Cambridge Companion to the Cello
Edited by Robin Stowell
The Cambridge Companion to the Clarinet
Edited by Colin Lawson
The Cambridge Companion to the Guitar
Edited by Victor Coelho
The Cambridge Companion to the Organ
Edited by Nicholas Thistlethwaite and Geoffrey Webber
The Cambridge Companion to the Piano
Edited by David Rowland
The Cambridge Companion to the Recorder
Edited by John Mansfield Thomson
The Cambridge Companion to the Saxophone
353

Edited by Richard Ingham


The Cambridge Companion to Singing
Edited by John Potter
The Cambridge Companion to the Violin
Edited by Robin Stowell

You might also like