You are on page 1of 189

The contributions in this volume reflect the efforts of musicology to

Héctor J. Pérez (ed) • OPERA AND VIDEO

understand a hybrid area with a fascinating evolution. They aim to
address the relationship between opera and audiovisual technology
from its origins to today by offering the results of a balanced critical and
innovative approach. The reader interested in opera, aesthetics, narrati-
ve or transmediality will find concrete approaches devoted to an unex-
plored diversity of aspects with an impact on the narrative conditions
in which we watch opera on screen. The variety of perspectives shows
how original methodological approaches are able to design a new map
of the main transmedial problems of opera in TV, DVD and even in pho-
nography. The book offers not only isolated theoretical contributions
but seeks a connection of them with significant practice oriented ap-
proaches coming from the fields of video direction and composition.

Héctor J. Pérez is Associate Professor of Audiovisual Communication

and Aesthetics and a member of the Technology and Information Re-
search Team, CALSI, at the Universidad Politécnica de Valencia. Among
his main publications on Opera are “Shakespeare jenseits des Dramas”
(1998); El Nacimiento de la tragedia. Un ensayo sobre la metafísica del artis-
ta en el joven Nietzsche (2001); “Opera Narratives: From Mythology to
Audiovisual Aesthetics” (2006); Expression in the Performing Arts (ed with
Inmaculada Álvarez and Francisca Pérez-Carreño, 2010); “Una estética
audiovisual de Electra” (2010).

Héctor J. Pérez (ed)


Peter Lang

ISBN 978-3-0343-0542-6

Peter Lang
The contributions in this volume reflect the efforts of musicology to

Héctor J. Pérez (ed) • OPERA AND VIDEO

understand a hybrid area with a fascinating evolution. They aim to
address the relationship between opera and audiovisual technology
from its origins to today by offering the results of a balanced critical and
innovative approach. The reader interested in opera, aesthetics, narrati-
ve or transmediality will find concrete approaches devoted to an unex-
plored diversity of aspects with an impact on the narrative conditions
in which we watch opera on screen. The variety of perspectives shows
how original methodological approaches are able to design a new map
of the main transmedial problems of opera in TV, DVD and even in pho-
nography. The book offers not only isolated theoretical contributions
but seeks a connection of them with significant practice oriented ap-
proaches coming from the fields of video direction and composition.

Héctor J. Pérez is Associate Professor of Audiovisual Communication

and Aesthetics and a member of the Technology and Information Re-
search Team, CALSI, at the Universidad Politécnica de Valencia. Among
his main publications on Opera are “Shakespeare jenseits des Dramas”
(1998); El Nacimiento de la tragedia. Un ensayo sobre la metafísica del artis-
ta en el joven Nietzsche (2001); “Opera Narratives: From Mythology to
Audiovisual Aesthetics” (2006); Expression in the Performing Arts (ed with
Inmaculada Álvarez and Francisca Pérez-Carreño, 2010); “Una estética
audiovisual de Electra” (2010).

Héctor J. Pérez (ed)


Peter Lang

Peter Lang
Opera and video
Héctor J. Pérez (ed)



Peter Lang
Bern · Berlin · Bruxelles · Frank fur t am Main · New York · Ox ford · Wien
Bibliographic information published by die Deutsche Nationalbibliothek
Die Deutsche Nationalbibliothek lists this publication in the Deutsche National-
bibliografie ; detailed bibliographic data is available on the Internet
at ‹ ›.

British Library Cataloguing-in-Publication Data : A catalogue record for this book

is available from The British Library, Great Britain

Library of Congress Cataloging-in-Publication Data

Opera and video: technology and spectatorship / [edited by] Héctor Julio Pérez. – 1st ed.
p. cm.
ISBN 978-3-0343-0542-6
1. Opera. 2. Video recording. 3. Motion pictures. 4. Television. I. Julio Pérez, Héctor
ML1700.O647 2012

Cover illustration : © J. Dugo, GRAMMA (J. M. Sánchez-Verdú),

Berlin (Zeitgenössiche Oper Berlin und Deutsche Oper Berlin, 2006)

Cover design : Didier Studer, Peter Lang AG

ISBN 978-3-0343-0542-6 E‐ISBN 978‐3‐0351‐0347‐2

© Peter Lang AG, International Academic Publishers, Bern 2012

Hochfeldstrasse 32, CH-3012 Bern, Switzerland,

All rights reserved.

All parts of this publication are protected by copyright.
Any utilisation outside the strict limits of the copyright law, without the permission
of the publisher, is forbidden and liable to prosecution.
This applies in particular to reproductions, translations, microfilming, and storage
and processing in electronic retrieval systems.

Printed in Switzerland
Table of content

Introduction 7

The Fairy Tale of Bel Canto: Walt Disney, Theodor Adorno,
Kurt Weill Play the Gramophone 13

Opera on Italian Television: The First Thirty Years, 1954-1984 45

“Temps Spatialisé”: Opera Relays and the Sense of Temporality 71

Overtures on Screen 91

Playback Problems when Filming Opera for the Screen:
Two Case Studies 115

The Opera Director’s Voice: DVD ‘Extras’
and the Question of Authority 129

Special Contributions
A Conversation with Pietro D’Agostino,
Video Director at the Gran Teatre del Liceu 153

Composing in New Synaesthetic and Interdisciplinary Spaces:
Libro de las estancias (Book of Abodes) as a Musical,
Architectural and Visual Installation Proposal 159

Authors 185


Since the nineties, audiovisual technology has been a threat to the

mystique surrounding live opera as an aesthetic experience that is
unique to the genre. Many spectators have been watching opera on
screen for years and, in most cases, this experience is not exclusive
but may actually encourage attendance at live opera. In late 2011, with
the economic crisis at its most destructive peak, very few theatres
have managed to keep their budgets intact. In Spain, the country
whence I write, there has been a considerable decline in public fund-
ing for most of the country’s theatres, as in many other places. How-
ever, the fragile economic context does not seem to have paralyzed
certain technological practices that have become widespread over the
last decade in opera theatres worldwide. Audiovisual productions of
main titles have not decreased significantly. It seems that the crisis is
not seriously affecting technology projects of major opera theatres.
What reason lies behind this?
I think we may well be looking at a second trajectory in the im-
pact of digital technology on opera. The experiences that have trans-
formed opera fans into digital consumers bring us closer to the para-
digm of cultural convergence. A spectator used to watching opera on
TV is now fast approaching the stage of watching it on a computer,
perhaps on a big screen at home, or on a tablet PC equipped with ex-
cellent headphones. Convergence has knocked on opera’s door, and
the answer can be experienced in what is on offer at several theatres.
In 2011, the Teatro Real in Madrid launched a digital box, which al-
lows us to watch live or to record a significant number of their pro-
ductions. Over the last two years, the MET has increased several ser-
vices based on the use of the Internet as a distribution platform. The
best-known is MET HD, which distributes live productions in high
digital quality to many cinemas around the world. The Met Player
8 Héctor J. Perez

service is also a remarkable initiative, by which means the institution

makes available to Internet users around the world (with very few
geographical restrictions) its private catalogue of recorded produc-
tions, many with a choice of subtitles and at truly affordable prices.
With the development of convergence, live experience at opera
theatre increases possible alternatives. Convergence is not about the
adoption of one type of device but rather the inclusion of many types
of experiences within a device, which today can be a smart TV, game
console, computer, smart phone or tablet. That convergence should
give rise to the possibility of a new experience of digital content is one
of the chief attractions of technological development. Watching opera
on screen no longer means being in the living-room; it can take place
on a train, in a park, at an airport, or while in a hospital waiting room.
These are all opportunities resulting from technological developments,
and it seems that the initial barrier that identified a type of art with a
kind of experience in very specific conditions has definitely been bro-
ken. Theatres themselves not only broadcast live opera in very suit-
able spaces, such as cinemas, but also in squares and parks, and even
after hours. In many cases, the latter type of experiences consists of
promotional acts by theatres trying to expand their influence and to
bring new spectators to the house. There is no doubt that the break-
down of barriers implied by the second wave of technology conver-
gence causes not only the variety of watching experiences to prolifer-
ate but also the scope of potential recipients. Not only do those who
attend in parks and squares do so for free, but those who now enjoy
opera streamed via Internet do so at a reasonable price. Unlike other
major cultural areas, where convergence is blocked by certain preju-
dices, above all in Europe, opera seems to be making good use of this
opportunity in positive ways.
But the relationship between opera and communication technolo-
gies is much older than the recent developments we have raised. Some
important contributions in this volume are concerned with phenomena
that require a historical perspective on the relationship between spec-
tators and technologies. It all reflects the efforts of musicology to un-
derstand a hybrid area whose main attraction is that it offers numerous
aspects that have yet to be explored as objects of study. This volume
aims to address this challenge by offering a balance between cultural
Introduction 9

and aesthetic issues that have emerged in the history of the relation-
ship between opera and audiovisual technology from its origins to
Gabriela Cruz proposes in “The Fairy Tale of Bel Canto: Walt
Disney, Theodor Adorno, Kurt Weill Play the Gramophone” a per-
spective of a significant continuity between the beginning of the pho-
nograph and the digital age. Both historical contexts are dominated by
technologies producing aesthetic qualities with the aim of exceeding
the qualities of reality. This is the basis of her extraordinarily fruitful
analysis of Cinderella’s song, “Oh Sing Sweet Nightingale”, from
Walt Disney’s eponymous animated feature (1950). Cruz’s main hy-
pothesis on that song as a new form of aural plenitude is contrasted
with a critical reading of the phonographic golden age of opera. She
does it through a dialogue with the most relevant thoughts on that
topic by Theodor Adorno, such as the contemplation of song as acous-
tic enigma. Cruz offers the case of the phonogram-scene “Tango
Angèle” from Kurt Weill’s Der Zar lässt sich photographieren (1928)
as a second phase of her approach. What does the gramophone deliver
to opera and to its stage? The question addressed requires several
pages for showing the difficulties of the scene as a theatrical/critical
proof of Adorno’s phonographic theory and several more to explore
the complexities of a Benjaminian allegory of song.
The second contribution to the relationship between opera and
technology also adopts an historical perspective in “Opera on Italian
television: The first thirty years, 1954-1984”. However, instead of a
traditional view of the story of Opera in Italian Television, Senici
focuses on some especially-revealing moments. He departs from a
fragment of a letter by Lucchino Visconti, where he shows his amaz-
ing ability to understand the problems of television. These comments
provide the opportunity to convert historical perspective into a spe-
cific analysis on a transmedial problem: “the gap between represented
time and performance time”. But this question leads to the core of the
problem Senici thematizes, the different modalities of tensions be-
tween opera and television. The first one considers the conflict be-
tween realism and theatricality in television opera. Senici’s diagnosis
shows the existence of a problem emerging in several aspects of the
analyzed broadcasts, such as the prevailing types of shot and camera
10 Héctor J. Perez

movements. The second one, regarding the first live opera broadcast
in colour in Italy, which took place on 7th December 1976 with
Verdi’s Otello, also reveals a tension between the aesthetics of televi-
sion and theatre although this was at a time that represented an impor-
tant evolution in the language of television compared to the fifties.
The last one, the broadcast on 25th August 1984 of the fourth per-
formance of Il viaggio a Reims in Luca Ronconi’s stage direction,
shows a television spectacularization of theatrical reality quite close to
the main reference of the TV spectacles of the eighties, Prince Charles
and Diana’s wedding. Senici invites comparison of the results of those
analyses with a general assessment of the subsequent developments
TV has experienced, characterized today by its evolution within cul-
tural convergence.
In this volume the reader will also find musicological research for
a better understanding of aspects relevant to the aesthetic experience.
How does the filming of an opera change our perception of temporal-
ity? Delphine Vincent’s “Temps spatialisé: opera relays and the sense
of temporality” is devoted to one of the most general issues related to
the change in the way we watch opera as film. She describes the lim-
ited status of the linear progressive temporality of opera, compared
with the richness of temporal modalities of cinema. This is an interest-
ing path to connect some important intermedial questions. For exam-
ple, opera uses cinematic effects but, when it is itself shot, it is no
longer able to account for them properly. Thus, the core of the contri-
bution is an analysis of the temporal structure of the alternation be-
tween kinetic movements (scena, tempo d’attacco, tempo di mezzo)
and static movements (adagio, cabaletta) in the Italian romantic oper-
atic conventions. This is an unconventional approach because the nar-
rative properties of musical structures usually remain ignored, though
they are always implying different kinds of conditions for the final
narrative results of filmed opera. Vincent’s découpages of Giuseppe
Verdi's passage lead us to evaluate the ways the visual narrative as-
pects are imposed on the musical structures, dictating a problematic
new perceptual field of temporality.
Is the overture part of the opera? What happens when we sit in
front of a screen during the overture? These two questions are indica-
tive of the orientation Gaia Varon raises in “Screen Overtures”. She
Introduction 11

analyzes different possible cases, such as the use of the overture as

title music, films that show the complete performance of the overture,
others in which the sound approaches the role of music in silent films
and the existence of a case in which the effectiveness of the relation-
ship between music and image has special value for the rest of audio-
visual production. Varon’s perspective also shows that the subtlety of
the relationship between music and image can be diagnosed as early
as the overture, even deeply enough to analyze the results of syn-
chrony between music and credits. Thus, all the pages devoted to
Jean-Pierre Ponnelle’s productions show the overture as an authentic
example of excellence reached in coherence with the whole narrative
of the work. In other cases, as in the study of Paul Czinner’s produc-
tion, the value of this section may be oriented to emotional factors
external to the opera, creating an aura around the legendary figure of
W. Furtwängler. Varon concludes her contribution with a systematic
proposal to express the most widespread forms of the narrative rela-
tionship between music and image in overture: Background music,
Film music, Programme music and (Absolute) Symphonic music per-
Jaume Radigales addresses one of the core dichotomies of the re-
lationship between film and opera. The problem of synchronization
goes beyond immediate aesthetic effect. It serves as a sign for detect-
ing whether cinema is going to be close to opera or vice-versa. As we
know, opera production following film criteria, with strict dubbing
and singers replaced by actors, is an extreme case and is not always
convincing. The reverse hypothesis, in which opera shares some char-
acteristics with documentary, does not seem to be a fruitful one. Radi-
gales takes into account these extremes and goes deeper into two op-
eras with almost classic status in audiovisual format, Joseph Losey’s
Don Giovanni and Hans Jürgen Syberberg’s Parsifal. Each case
shows a different alternative formal proposal that proves how flexibil-
ity in the relationship between artistic languages becomes a way of
enhancing artistic production. Radigales’s approach is very interest-
ing, not only because it offers a relevant perspective for film opera but
also because it is useful for discussion of most popular formats, such
as the filming of live opera performances.
12 Héctor J. Perez

Áine Sheil’s essay “The opera director’s voice: DVD ‘extras’ and
the question of authority” focuses on one type of bonus feature in-
cluded in many opera DVD ‘extras’: interviews with, and commentar-
ies, by stage directors. She explores how these can provide valuable
perspectives on the interpretative strategies behind the productions in
question, as well as how they may inevitably guide reception and in-
terpretation of the main features. The case studies are Calixto Bieito’s
commentary on his production of Wozzeck at the Gran Teatre del
Liceu, Barcelona, and an interview with Peter Brook on his production
of Don Giovanni at the Aix-en-Provence Festival. Bieito’s commen-
tary on his own provocative style of direction is analyzed critically in
his dependence on the so-called ‘work concept’. Brook’s fifteen-
minute commentary reveals not only paradoxical aspects, as in
Bieito's case, but also the way he subverts normal opera practice with
his authority. Sheil’s perspectives follow different paths (Bordieu’s
statements on cultural production, Philip Auslander on economic im-
plications, and R. Taruskin and others on the question of authorship)
before suggesting that the commentaries ultimately serve to stabilize
and fix the traditional values of the artistic field of opera.
The book concludes with two special contributions, the first one
is a transcription of a conversation with Pietro d’Agostino during the
seminar at the outset of several of the papers collected here.
D’Agostino is video director at the Gran Teatre del Liceu and was so
kind as to share a few days of reflection with us and offer his profes-
sional experience in an open dialogue with several of the speakers at
the seminar. Finally, we thought it would also be of great interest to
include reflections on his own creation by the Spanish composer José
María Sánchez Verdú. He has written an excellent contribution ex-
plaining some creative keys in his Libro de las estancias, which has
been a complementary perspective to our discussion as it involves a
unique approach to technology, more determined by poetic and cul-
tural decisions than by prevailing fashions or trends. His proposal,
through the richness and coherence of his artwork, may reflect the
open and critical perspective on the relationship between technology
and art that this book would wish to promote.
The Fairy Tale of Bel Canto: Walt Disney, Theodor
Adorno, Kurt Weill Play the Gramophon


This essay explores bel canto (re)formed by phonography and ad-

dresses lyrical apotheosis as an effect of the materialities of recording,
transmission, and reproduction. Bel canto is understood here as a
form of lyrical beauty somewhat unhinged from operatic history, as an
ideal of song and of singing implicated with a modern poetics of won-
der. Fin-de-siècle inventors and fabulists first conceived of the prom-
ise of reproduction – the historical and technical domain of recorded
sound – as one of enchantment, formed in magical intercourse with
the inhuman. They thus inaugurated an influential line of discourse
about recorded song, one echoed most recently in Christopher Morris’
discussion of lyrical song in the digital age as a hybrid form, an ex-
pressive moment forged in the encounter with the radical otherness of
technical mediation, grounded in the pleasures of “dispersion, distri-
bution, and blurred boundaries” or, as Morris puts it, “of transmis-

1 An early version of this essay was presented at the International Workshop on

Opera and Video, Universidad Politécnica de Valencia/Instituto Valenciano de
la Música, held on the 22–23 March 2010. I am grateful to Héctor Perez Lopez,
the conference organizer, and to all the conference participants for the generosi-
ty of their reactions to the initial paper. Thanks also to Roger Parker, Dana Goo-
ley, and Alessandra Campana, who read the later version of the essay, asked im-
portant questions, and made crucial suggestions that greatly improved the final
version published here.
2 The topic is addressed by Friedrich A. Kittler in Gramophone, Film, Typewriter,
trans. Geoffrey Winthrop-Young and Michael Wutz (Stanford, CA: Stanford
University Press,1999), 21–114; and Felicia Miller Frank, The Mechanical
Song: Women, Voice, and the Artificial in Nineteenth-Century French Narrative
(Stanford, CA: Stanford University Press, 1995), 143–71.
14 Gabriela Cruz

sion.” Morris’ statement is part of a new critical argument for opera
that positions the genre beyond the premise of liveness, accepting
technological re-mediation as essential to lyrical fruition today. While
recognizing the importance of Morris’ view, my purpose here is both
less celebratory and essentially retrospective: to reflect on the move-
ments of technical invention and critical injunction that have impelled
the historical dominance of transmission in lyrical culture, contem-
plating the wish-images of phonographic utopia in parallel with the
cost they have entailed.
Song understood as a form of transmission belongs to the history
of creative commerce between people and things, a domain that brings
the notion of musical agency under new scrutiny. Writing exclusively
from a technosonic perspective, and addressing the mass media phe-
nomenon of the posthumous duet, Jason Stanyet and Benjamin Piekut
have recently re-defined agency as a diffuse form of collaboration,
which “transpires along differential axes of access, emplacement,
privilege, capacity, and responsibility.” They inscribe song in a his-
tory of human accommodation to technology, one that shuns old
anxieties about the self-bounded nature of the human subject. Their
statement elicits, of course, a re-examination of established philoso-
phical precepts on voice and song as expressive of an essential subjec-
tivity, a topic I pursue below. I call attention to the technical and sci-
entific history that sustains current understandings of the singing
voice, noting that modern discourse on the subjective powers of song
is largely grounded on habits of listening and sonic pleasures forged
within the sensorial experience of modern media. Lyrical song in
transmission characteristically devolves a grandiose dream horizon.
Operatic song, crystallized in the age of reproduction as a restricted
economy of beautiful singing drawn from a canonical repertory of
past greatness, obsessively recorded and staged, trades on illusions of

3 Christopher Morris, “Digital Diva: Opera on Video”, Opera Quarterly 26/1

(2010), 96–119: 114–15.
4 Liveness designates not just the condition of live performance but also its cul-
tural status. The word is coined by Phillip Auslander in Liveness: Performance
in a Mediatized Culture (London and New York: Routledge, 1999.
5 Jason Stanyek and Benjamin Piekut, “Deadness Technologies of the Intermun-
dane“, The Drama Review 54/1 (2010), 14–38: 33.
The Fairy Tale of Belcanto 15

dematerialization – of timelessness, genesis, perfection and plenitude.

Even the operatic voice, an object that remains powerfully symbolic
of the experience of immediacy – of being “in touch” – is commonly
encircled by critical discourses that celebrate various forms of disper-
sal, flight, and removal.
Below, I explore the dream-force of song, as it emerges from pho-
nographic practice. I consider it in light of Benjamin’s “dialectical
image,” contemplating the utopian projection and the sedimented past
that constitute it in two apparently irreconcilable scenes of lyrical
transmission and reproduction. One scene, issuing from the technical
apparatus, is Cinderella’s song, “Oh Sing Sweet Nightingale,” from
Walt Disney’s eponymous animated feature (1950), offered as a mod-
ern dream-image of bel canto. The other, conceived for the live stage
of opera, is the phonogram-scene “Tango Angèle” from Kurt Weill
Der Zar läß sich photographieren (1928). It presents lyrical beauty as
an object arising from the collaboration between a singer and a
gramophone. Together, the two scenes chart a significant territory of
cultural production (popular and conservative vs. erudite and avant-
garde), of media and genre (film/reproduced art vs. opera/performative
art), of musical style (tonal vs. atonal grammars), and of taste. Yet,
important affinities bind these two scenes of bel canto. Both imagine
song as a form of intercourse between a singer and a machine. In both,
lyrical beauty appears in the guise of a wish, a lyrical trace exceeding
the singularities of the phonographic operation. Both offer this trace as
an element that registers a captivating surplus to be experienced in a
standstill. The trace intersects past and present, petrified recording and
transitory being. Finally, and so far unremarked, but crucial to my
argument below, these chosen instances call for a form of retrospec-
tive contemplation that brings into a new aesthetic and critical focus
the relevance of Baroque magic to modern experience. Thus, Cinder-
ella absorbs the fairy-tale motif of the magical shoe, and Der Zar
brings to presence a form of seraphic divinity. The two figures are
offered in film and in opera as portals to enchantment, bringing into
modern consciousness a long-repressed memory of an otherness made
to sound beneath and in-between known and familiar objects. This
otherness is the surfeit – the valuable node of experience lost to the
rationalities of technosonic effectiveness – that this essay aims to sal-
16 Gabriela Cruz

vage and restore to a new critical understanding of song in transmis-


The Perfect Fit

Cinderella’s shoe recalls sartorial pleasures. More broadly, it evokes

the notion of the perfect fit, a utopian accord between human desire
and matter that is central to baroque magic and remains so to our me-
dia age. In the fairy tale collected by Giambattista Basile and pub-
lished posthumously in his Pentamerone (1634–1636), we are told
that “as soon as ever [the shoe] approaches Zezolla’s foot, it darts on
to it of its own accord, the way iron flies to the magnet.” In Perrault’s
better-known version included in Ma Mère L’Oÿe of 1697, a glass
slipper melds to the foot like wax. Later, in German lands, the won-
ders performed by the shoe take on a gruesome patina. In Grimm’s
Aschenputtel, the shoe retains a mineral inflexibility and hardness, but
is now made of gold. Each of the two stepsisters fits it to her foot by
means of ghastly self-mutilation – the elder cuts off her toe and the
younger cuts off her heel. Only Aschenputtel fills the shoe without
violence. The topos of the shoe is that of manifest power, ignored
altogether by nineteenth-century theatrical versions of the tale, includ-
ing Charles G. Étienne’s and Nicolo Isouard’s Cendrillon (1810) and
Jacopo Ferretti’s and Gioachino Rossini’s La Cenerentola (1817). In
earlier versions of the tale, however, Cinderella’s slipper, an object of
fashion, acts like a magnet seeking and finding its complementary
charge in moral beauty. It darts about, and attaches itself and melds to
only one foot, rejecting all others. Its miraculous behavior demon-
strates the magic of the perfect fit, emphasizing the characteristics of

6 Alan Dundes, Cinderella: A Casebook (Madison, WI: The University of Wiscon-

sin Press, 1988), 10.
7 Paolo Fabbri, “Librettos and Librettists”, in The Cambridge Companion to
Rossini, ed. Emanuele Senici (Cambridge, UK: Cambridge University Press,
2004), 51–67: 59.
The Fairy Tale of Belcanto 17

adherence, envelopment, and repetition. All this fabulous surplus, lost

to nineteenth-century audiences, is what Walt Disney’s film version of
the tale restores to our modernity, along with a new American vision
of female virtue.
The perfect fit is the pleasure celebrated in Disney’s Cinderella.
Disney restores Perrault’s shoe, producing a glass slipper unmistak-
able in its late-1940s chic and highlighted by playful manipulation –
in the film, Cinderella’s elegant foot keeps slipping out of all foot-
wear. Yet, as a way of distinguishing goodness, the shoe remains a
comic after-thought. Disney added a musical scene to the fairy tale,
introducing song as a new site of moral and aesthetic judgement. The
moment is fashioned by means of contrasting scenes. A shot shows
the music room of Cinderella’s household. The camera settles on the
two stepsisters and their mother occupied in music-making. It travels
towards them. One sister sings; the other plays the flute, accompanied
by the mother at the piano. The two girls, drawn as hapless creatures,
produce a horrid performance of the song, “Oh Sing Sweet Nightin-
gale,” witnessed by Lucifer the cat, the resident listener. The song is
delivered in a nasal singing voice with flat intonation, wrong notes
conveyed by a tinny flute sound and a depth-less voice characteristic
of music heard at a relative distance. Lucifer listens a bit and then,
repelled by the performance, retreats from the room. The cat closes
the door behind him and shrugs off the bad performance.
As the door closes, there is a split second of silence, followed by a
rendition of the same song by Cinderella. Poor Cinderella, too humble
for music lessons and accompanying instruments, sings alone as she
scrubs the floors of the mansion’s palatial entrance, one floor down.
But her rendition is given an extraordinary musical and acoustic emi-
nence. Her performance, as noted by Billboard magazine in December

8 On Cinderella and American morality, see Naomi Wood, “Domesticating Dreams

in Walt Disney’s Cinderella”, The Lion and the Unicorn 20/ 1 (1996), 25–49.
9 The reasons behind the controversy were financial. The complexity of the song
scene made it expensive to produce. See Susan Ohmer, “‘That Rags to Riches
Stuff’: Disney’s Cinderella and the Cultural Space of Animation”, Film History
5/ 2 (1993), 239–49: 224.
10 The scene is avalaible here:
18 Gabriela Cruz

of 1949, is musically flawless. She is also accompanied by a full
orchestra on the soundtrack, delivering a superior version of the song.
This is a point registered within the film narrative by Lucifer who,
listening to her singing voice, moves quickly downstairs and towards
her, and stands in rapt attendance. In the overdetermined logic of the
scene, Cinderella’s musicianship is impeccable and mobilized in sup-
port of another form of perfection. Her gorgeous singing, the richness
of the musical means supporting her song, and the sense of acoustic
proximity produced by the music, write off a bad object from the past:
the poor musical rendition of the less well-endowed stepsisters deliv-
ered in an indifferent live performance.
The music parlor occupied by the stepsisters showcases both poor
musicianship and poor audioship, emphasizing the relative conditions
of listening in proximity and from afar. Here the imaginary camera’s
movement of approach and departure is sutured to acoustic gradations
of closeness and distance. In contrast, the following scene of Cinder-
ella’s correct singing underscores a voice conceived as a perfect audio
object, the ideal fit to the acoustic channel she occupies. This studio
voice – a re-mastered acoustic object with no equivalent in nature – is
designed for perfect acuity, endowed with sonic depth, made intimate
to the ear from no matter what distance, and made clearly present. The
contrasting scenes underwrite the idea of transmission as a channel
superior to liveness, one that emancipates voice from the restrictions
of nature. In this sense, Cinderella’s singing magic stands in a peculiar
relationship to the culture of bel canto. While the musical text of “Oh
Sing Sweet Nightingale” mobilizes powerful operatic clichés regard-
ing the purity, beauty, and the bird-like nature of female voices, offer-
ing the voice of Cinderella as an uncomplicated manifesto for the
wholesome virtues of mimesis in music, the aesthetic purpose of the
number is other, namely, to deploy nature as a form of camouflage.
Birdsong, used conventionally in opera as a simile for the natural
voice, is brought into the animated feature as a decoy, deployed in
praise of transmission.
Cinderella’s singing voice is reproduced sound, and hence an ob-
ject of the kind upon which the operatic stage has long frowned. In

11 Billboard, 31 December 1949, 13.

The Fairy Tale of Belcanto 19

this sense, the celluloid scene may well be understood as a modern

counter-argument, a rejection of the ingrained suspicion regarding
technological mediation which for so long has afflicted lyrical prac-
tices on stage. In Paris in 1831, the infernal souls of Meyerbeer’s
Robert le Diable sang through megaphones. In Harry Kupfer’s 1987
production of Gluck’s Orfeo and Euridice, songs reproduced on stage
with a tape recorder and television serve as imposing signifiers of
absence and irrecoverable life. More recently, at the Lisbon São Car-
los Katharina Thalbach’s production of Die Fledermaus (2010) recast
Prince Orlofsky as a decadent vampire, his supernatural status con-
veyed by a voice electronically amplified. Today, these and similar
operatic instances of recoiling before sonorous plasticity are pretend
gestures amid the rapid mediatization of the genre. They sustain the
fantasy of acoustic exceptionality, helping listeners imagine opera as
singularly untouched by the onslaught of technics and therefore still a
domain of pure immediacy. However, the historical point made by the
singing voice of Cinderella was that already by 1950 the natural voice
was an object of the past. Cinderella’s song stands as a moment of
consolidation, in the vernacular of film, of a modern understanding of
the singing voice mostly experienced as an object of transmission.
“Oh Sing Sweet Nightingale” emphasizes the pleasures of simula-
tion, of singing re-conceived as an effect of itself. Performed by Cin-
derella, the song opens the path towards new sensorial environments,
technically manufactured, which have since become a hallmark of the
Disney “live” experience. The character’s hyper-real vocality, a fore-
runner of the audio-animatronic voyages of Disneyland described by
Umberto Eco, produces the work of sensorial envelopment and re-
wiring. The singing voice, plastically altered for enhanced evenness
and intimacy, re-shapes the auditory field as an experience of prox-
imity and plenitude. This re-mastering develops out of a history of

12 The point is made by Melina Esse in her “Don’t Look Now: Opera, Liveness,
and the Televisual”, Opera Quarterly 26/ 1 (2010), 81–95: 86–87.
13 Jean Baudrillard, Simulacra and Simulation (Ann Arbor, MI: University of
Michigan Press, 1994), 22. On modern practices of audio manipulation in the
recording of opera, see Morris, “Digital Diva”, 113. For a discussion of Disney’s
audio-animatronics, see Umberto Eco, Travels in Hyperreality (New York, NY:
Harvest, 1986), 43–48.
20 Gabriela Cruz

technical involvement with acoustic emission and hearing that I ad-

dress below. For now, I would simply like to note the foundational
force of this sort of transmitted song. Voice, re-engineered as desir-
able blanket of sound and for decades a part of our televisual quotid-
ian, sustains, and in turn has found critical legitimacy in the fantasy of
the maternal voice and in recent psychoanalytical readings of the
voice that re-inscribe the technological effect in a new natural matrix.
Guy Rosolato’s description of the maternal voice as an “envelope
of sound” that “surrounds, sustains, and cherishes the child,” applies
to the aural awareness proper to infantile containment as much as it
does the engulfing acoustics of the “right” singing voice in Disney’s
fairy tale. Both instances of audio envelopment introduce an hesita-
tion in the perception of voice, imagined both as projection and as
introjection. Cinderella’s singing voice, which was recorded in a stu-
dio and re-mastered to deliver acoustic evenness and an experience of
intimate proximity, offers a concrete instance of the fantasy of the
maternal voice. It installs vocality in the ether of transmission and
proffers it as a new form of aural plenitude, a figure critically dislo-
cated from bodily materiality. In this instance of Cinderella’s singing,
as in many other moments of classical cinema, the visual and auditory
fields are not synchronized or, more subtly, they are not synchronized
absolutely – voice is offered along different points of an unhinged
relationship to the represented body, its fictional material source.
These variations in the process of audio-visual correspondence, a
technical feature that Disney’s animated film exploits for aesthetic
purposes, turn on the culturally pedigreed notion of disembodiment,
transforming it from the preternatural event enshrined in the Romantic
imagination to an everyday occurrence. Unsurprisingly, then, the dis-
embodied voice, hovering over representation and the classical matrix
of subjectivity, is commemorated in recent critical work on film and

14 Guy Rosolato, “La voix: entre corps et langage”, Revue française de psychana-
lyse 38/ 1 (1974), 75–94: 81.
15 The importance of unhinged synchrony to the fabric of film is addressed by
Carolyn Abbate in “Speaking and Singing: What is Real?” (Paper delivered at
the Opera Seminar, The Humanities Council at Harvard University, 2007).
The Fairy Tale of Belcanto 21

on opera. In opera studies, it feeds a potent concern for the structur-
ing role of pre-technological fantasies of transcendence in the canon
and an ongoing debate about the sensorial qualities of the operatic
experience today, an inquiry that has also mobilized the intellectual
resources of psychoanalytic theory. It is the transmitted voice of
technological high fidelity, rather the experience of song performed
live or captured in the imperfect thinness of the early zinc disc, that
animates most effectively recent musicological interests in opera as a
historical culture of subjectivity.

Under the Spell of the Engineer

Philip Auslander observed that emerging media tend towards the con-
ventions and pieties of established ones. So it is with Cinderella,
whose song purposefully suppresses liveness in favor of a mode of
aural proximity developed for telephony in the 1920s and soon after
absorbed into the experience of radio and cinema. In this sense, Cin-
derella’s musical delivery was not meant to produce a statement about
the future; it simply gave aural form to the old turn-of-the-century
dream of transmission, repeating what was by 1950 already a long

16 My main sources here are Kaja Silverman’s The Acoustic Mirror: The Female
Voice in Psychoanalysis and Film (Bloomington, IN: Indiana University Press,
1988); and Michel Chion, Audio-Vision: Sound on Screen, trans. Claudia Gorb-
man (New York, NY: Columbia University Press, 1994).
17 On opera’s pre-technological fantasies, see Carolyn Abbate, In Search of Opera
(Princeton, NJ: Princeton University Press, 2001), 145–60. On presence effects,
see Hans Ulrich Gumbrecht, The Production of Presence: What Meaning Can-
not Convey (Stanford, CA: Stanford University Press, 2004), 17–20; Mladen
Dolar, “The Object Voice”, in Gaze and Voice as Love Objects, ed. Renata Sa-
lecl and Slavoj Zižek (Durham, NC, and London: Duke University Press, 1996),
7–31; and Michelle Duncan, “The Operatic Scandal of the Singing Voice:
Voice, Presence, Performativity”, Cambridge Opera Journal 16/3 (2004), 283–
306: 289.
18 Phillip Auslander, Liveness: Performance in a Mediatized Culture (London,
UK: Routledge, 1999), 22.
22 Gabriela Cruz

established technical standard of broadcasting and reproduction. The

film was the brainchild of old-timers at Disney studios, a planned
celebration of the pre-war triumphs of Snow White (1937), and Fanta-
sia (1940). It addressed sound as part of this retrospective, carefully
registering and putting to cinematic use gains in the quality of sound
inscription and reproduction achieved during the inter-war years. The
strange condition of Cinderella’s voice, intimately present to our ear
as the animated body is seen at a distance downstairs, fulfilled an old
aspiration that physicist Harvey Fletcher, one-time director of the
Acoustics Division of Bell Laboratories, described as “reproducing
speech so that one’s interlocutor seemed one meter away” and im-
proving “telephony to a point where, from sound alone, the listener
would be unable to tell whether or not speech was coming through a
Beginning in the 1920s, the project at Bell Labs mobilized sig-
nificant scientific resources, and after 1931, it involved a notable mu-
sical collaboration with Leopold Stokowski and the Philadelphia Or-
chestra, where “sound with depth” came into being. Sound with depth
was the result of a new multi-track recording approach, involving an
innovative placement of microphones close to individual instruments,
combined with the development of the bi-aural headset. Stokowski
recognized immediately the “sense of space, of direction and consid-
erably more definition” afforded by the new technology. The dream
of telephony, of molding sound frequency and intensity in transmis-
sion and reproduction so as to produce the impression of physical
immediacy, seemed to have been achieved, and was soon mobilized
for radio, as it would later be for the cinema and TV. On April 13,
1933, a concert given by the Philadelphia Orchestra under the baton of
Alexander Smallens, with Stokowski sitting at the controls in a box at
the rear of the hall, produced a first operatic demonstration of the new
technology. The event made headlines in The New York Times:

19 Robert E. McGinn, “Stokowski and the Bell Telephone Laboratories: Collabora-

tion in the Development of High-Fidelity Sound Reproduction”, Technology and
Culture 24/1(1983), 38–75: 40.
20 Ibid., 49.
The Fairy Tale of Belcanto 23


strated by Dr. Stokowski/Pours Rich Music from Empty Stage. FULL Range is
covered / A.T.&T. Test at Philadelphia Is Hailed as Opening Way to New Music

Waldemar Kaempffert, the science editor of the Times, enthused:

The lights of the auditorium were dimmed, the curtain rose and from an empty
stage came the strains of Wagner, played as never before, so far as dynamic
range and volume are concerned. Although the Philadelphia Orchestra was in
the ballroom of the Academy, it seemed to the audience in front of the empty
stage that violinists, double-bass players and wind instruments were seated in
their familiar places.

For Kaempffert, the importance of the experiment was not only that it
demonstrated that transmission of music was now possible, but that
the conditions under which music was transmitted actually improved
upon the sound of live performance. Under the sub-heading “Sto-
kowski Controls Switches,” Kaempffert explained how this was done:

In back of the house sat Dr. Stokowski before two dials and three switches. With
these he controlled the volume and quality of the music that came from three
concealed loudspeakers on the stage, connected with three microphones in the
ballroom, where the orchestra was playing.
The musicians in the audience freely admitted that they had never heard effects
like those obtained by Dr. Stokowski. Pianissimi that conductors hear only in
their mind’s ear and never succeed in getting and a piling up of volume in tu-
multuous crescendos that are beyond the power of orchestral performers, gave a
new aspect to Wagner.

Under the article’s final sub-heading, “Spooky Experiments Made,”

Kaempffert even ventured a bit of music criticism:

With this equipment, Dr. Stokowski showed what could be done with an excerpt
from Götterdämmerung in which Agnes Davis sang the part of Brünnhilde. The
climaxes were overwhelming to the audience. Seated at his controls, Dr. Sto-

21 “New Tone Quality By Wire Achieved”, New York Times, 13 April 1933, His-
torical Newspapers, Proquest, accessed 15 December 2010 <http://hn.bigchalk.
24 Gabriela Cruz

kowski superimposed his interpretation on that of the invisible orchestra under

Mr. Smallens’ direction.
The voice of the singer always soared above the instruments. With his hand on
the switches, Dr. Stokowski achieved depth where he wanted it, so that even
without amplification the orchestra sounded as if it had more than the foundation
supplied by the usual double basses and tubas.
Musicians present were impressed with the effects. In the building up of rushing
crescendos, it seemed as if an army instead of 100 men were playing and as if
the invisible Miss Davis were as huge as the Statue of Liberty.

In Stokowski’s hands, the new technology played a symphonic game

of asynchrony, decoupling sound from its material source. The con-
ductor at the control knobs in 1933 announced a revolution in grand
opera, radically reconfiguring the relation between voice and body,
and calling for a new relationship between performing labor and
sound. The conductor himself declared the problem of the prima
donna finally resolved. “Can we change that lady? She might change
herself if she would exercise, if she would eat less [...] but it is really
not going to be necessary. Electricity will change the lady.”
Stokowski was not given the opportunity to pursue his adjust-
ments further. His collaboration with Bell Laboratories soon came to
an end. Disney, however, took to the idea. The animation studios were
quick to make the multi-track system of enhanced stereophony (re-
named “Fantasound”) their acoustic signature in 1940 with Fantasia.
In Cinderella, the animators addressed Stokowski’s “problem of the
singing lady.” The studios dedicated special attention to the creation
of the main character, a task they accomplished in three phases: a film
was produced of Helena Standley acting the role of Cinderella; the
animated image of the character was drawn frame-by-frame from film
footage, and sound engineers then grafted the voice of Ilene Woods
unto the image.
Cinderella’s singing in “Oh Sing Sweet Nightingale” acknowl-
edged its own unreality in covert and overt ways, in its broadcasting
quality described above, but also in a purposeful abandonment of mi-

22 Cited in “Stokowski and the Bell Telephone Laboratories”, 59.

23 Leopold Stokowski, “New Horizons in Music”, Journal of the Acoustical Socie-
ty of America 4/ 1 (1932), 11–19: 12, cited in McGinn, “Stokowski and the Bell
Telephone Laboratories”, 60.
The Fairy Tale of Belcanto 25

mesis for a new form of spellbinding effect and for self-conscious

play with sonic inscription. At the end of the scene, Cinderella sings
in polyphony with herself, encoding the old notion of the cadenza in a
new multi-track format. Her final melodies, ostensibly born out of
themselves, present the listener with the possibility of music delivered
from the limitations of nature. The techniques of transmission open
new possibilities for listening, one of which is to experience the quirks
of live performance – the possibilities of interruption, memory lapses,
and error – as aesthetic effect. The cadenza playfully addresses this
possibility. Here, the refrain is sung in varied repetition, alternating
between slightly longer and shorter versions of the same text, each
contained in a floating soap bubble and harmonically superimposed
upon another. Incompleteness and interruption, traumatic possibilities
in live performance, are artfully recast as a controlled play of desire.
To this, the singing soap bubbles add an even more revolutionary pos-
sibility. They transpose the special relationship between foot and shoe
enshrined in the fairy-tale to that of song and the electronic circuit –
magic here becomes a fully mediatized domain. The multiplication of
vocal parts in “Oh Sing Sweet Nightingale,” surprisingly unnoticed in
the critical literature, dramatizes the pleasures of trying over and over
the perfect shoe. Predictably, Disney’s lore on this point stressed com-
fort. Ilene Woods described later how, during her recording of the
song, “Disney listened with his eyes closed” and then said,

As I listened to you sing, I got a picture in my mind. Cinderella is scrubbing the

floor. As she sings, a soap bubble sings in two-part harmony with Cinderella,
then another soap bubble rises and we have three-part harmony. And another
soap bubble and Cinderella becomes a quartet, and eventually a choir. I see all
of these images in floating bubbles, and I hear your sweet voice, repeated again
and again, and it all blends so beautifully.

The voice that blends so beautifully is the same that audiophiles in the
1960s briefly celebrated as “audiogenic,” a sound bearing on the
physiology of the listening subject. Meanwhile, Cinderella’s voice,
comfortably installed in the ether of transmission, looks ahead to vo-

24 Pat Williams and James Denney, How To Be Like Walt: Capturing the Disney
Magic Every Day of Your Life (Deerfield Beach, Fl.: HCI, 2004), 63.
26 Gabriela Cruz

cal apparitions in present video-recordings of opera, and to the new

discipline of re-mediation these apparitions have imposed on the per-
formance culture of opera. Christopher Morris, inspired by Bruno
Latour, has written that operatic song re-conceived in the digital age
as a practice of re-mediation, exists as an experience of human hybrid-
ity. Re-mediated bel canto recovers something of the baroque notion
of art as a mode of alien encounter, configured in unexpected acts of
intimacy binding unequal properties: foot and shoe, singer and ma-
chine. This notion of song as an art of binding with an indifferent
otherness – an artifact, divine will – separates the lyrical voice from
an enlightened belief in the expressive purity of singing and in song as
symptom of human nature. Cinderella’s paradigm, promoting blend-
ing, repetition, multiplication, and dispersal is mostly relevant for
what it occludes: the instances of life excluded from the process of
physical and cultural re-mediation and an early history of sound re-
cording populated by unacceptable stepsisters.

Fantasizing the Ugly Stepsister

Let us revisit Theodor Adorno’s controversial and much discussed

comment from 1927 that

male voices can be reproduced better than female voices. The female voice eas-
ily sounds shrill, but not because the gramophone is incapable of conveying high
tones, as is demonstrated by its adequate reproduction of the flute. Rather, in or-
der to become unfettered, the female voice requires the physical appearance of
the body that carries it. But it is just this body that the gramophone eliminates,
thereby giving every female voice a sound that is needy and incomplete. Only
there where the body itself resonates, where the self to which the gramophone

25 For an approach to the topic, see the collection of essays edited by Melina Esse
published in Opera Quarterly 26/ 1 (2010). On the phenomenon of re-mediation
of opera, see Morris, “Digital Diva”, 114–15.
26 Ibid., 115.
The Fairy Tale of Belcanto 27

refers is identical with its sound, only there does the gramophone have its le-
gitimate realm of validity: thus Caruso’s uncontested dominance.27

Empirical observation sustains Adorno’s revolutionary claim that the

phonograph provides a new scheme of identity formation, propping up
a new Ur-voice which the modern listener echoes. Barbara Eng notes
the pioneering character of Adorno’s critical statement. She remarks
that Adorno’s discovery of voice as an object of primordial affect and
his brief outline of the new psycho-social schema of phonography,
already anticipates the fundamentals for the later psychoanalytical
fantasy of Voice as acoustic mirror, the notion of the maternal voice
elaborated by Rosolato and Silverman in the feminine. Within the
scheme, Adorno’s insistence on an exclusive mechanism of “mascu-
line identification” raises pertinent questions also considered by Eng.
She suspects that the philosopher’s argument follows from prejudice,
an anachronistic belief in the feminine as a gender of immanence,
incompatible with the condition of disembodiment. The young
Adorno certainly seems to write on the basis of automatic reflex when
he annotates the lesser quality of the recorded female voice. But his
explanation, while registering the psycho-social discomforts enunci-
ated by Eng, also responds to sensorial experience in a concrete man-
ner. The sound of the early acoustic gramophone, which is the focus
of the philosopher’s attention, reproduced only a limited spectrum of
frequencies, lacking especially in the higher acoustic range. Acoustic
recording captured a frequency range of approximately 200 to 2,400
Hertz (about three octaves), enlarged by electric recording in 1925 to
approximately 6,000 Hertz (about an octave extension). These upper
limits for recorded frequencies meant that the full spectrum of high
lyrical notes was not registered, let alone reproduced by early phono-

27 Theodor W. Adorno, “The Curves of the Needle”, trans. Thomas Levin, October
55/ 1 (1990), 48–55: 54. The essay is discussed in Barbara Eng, “Adorno and
the Sirens”, in Embodied Voices: Representing Female Vocality in Western Cul-
ture, ed. Leslie Dunn and Nancy A. Jones (Cambridge, UK: Cambridge Univer-
sity Press, 1997), 120–38: 128; and in Simon Frith, Performing Rites: On the
Value of Popular Music (Cambridge, MA: Harvard University Press, 1998),
28 Eng, “Adorno and the Sirens”, 126.
28 Gabriela Cruz

graphs. An understanding of the dimensions of acoustic loss in sound

recording was only reached in 1930–1931, when Bell Lab scientists,
collaborating with the Philadelphia Academy of Music, set the mod-
ern upper limit for acoustic filtering at 16,000 Hertz. Thus, acoustic
and early electric recording eliminated a spectrum of frequencies,
those that the philosopher re-writes as the missing body of a woman.
Under the physical constraints of early phonography, male voices
recorded best, not because of an inherent ontological correspondence
between instrumental reason as embodied in the technical device and
the male vocal organ, but because the acoustic spectrum produced by
the tenor and bass happened to fit snugly in the acoustic spectrum of
contemporary recording. In other words, the shoe determines the best
Caruso, approvingly installed by the philosopher in the shellac
grooves, inscribes bel canto in futuristic fantasy. The recorded singer
is a curious and unexpected figuration of the new “man-machine” that
resonates with bolder modernist statements of the time. Italian futurist
Filippo Marinetti’s definition of the ideal universe as that “which re-
mains devoid of women, consisting only of man and machine” [in
war], comes to mind, as does Soviet filmmaker Dziga Vertov’s poetic
defense of film as an act of peace-making “between man and ma-
chines.” In a vein similar to that of Marinetti and Vertov, Adorno
seems to speak here of conciliation, or of a desirable accommodation
with the inhuman, imagined at the expense of the emoting body (i.e.,
woman). The very image of “the curves of a needle,” symbolic of a
modern matrix forged by technics, speaks to a utopian exclusion of
maternal biology, as of nature tout court. In Adorno’s “curves,” the
suggestive line distilling the female body activates a longer tradition
of fantasizing female inhumanity as sublime form. More prosaically,
the image also alludes to the ways in which phonography put pressure

29 Dziga Vertov, “Kino-Eye” (1922), 11, cited in Seth Feldman, “‘Peace between
Man and Machine’: Dziga Vertov’s The Man with a Movie Camera”, in Barry
Keith Grant and Jeannette Sloniowski, eds. Documenting the Documentary:
Close Readings of Documentary Film and Video (Detroit, Mich.:Wayne State
University Press, 1998), 42.
30 Miller, Mechanical Song, 190–93.
The Fairy Tale of Belcanto 29

on the established female professionals of bel canto during the first

decades of the twentieth-century, casting an aura of unease around
women’s bodies. This unease had as much to do with the inability of
early recording to register without distortion the entire spectrum of
frequencies contained in a high voice, as with the cumbersome nature
of acoustic recording, seen as especially inimical to female singers. It
was one thing for Caruso to sing with his mouth firmly placed on the
recording funnel, and for Fred Gaisberg, the sound engineer of the
Gramophone Company who produced the recordings of all major
operatic stars in the early years, to instruct him on how much distance
he should put between himself and the device in moments of critical
pitch-level and loudness. It was quite another thing for the same sound
engineer to request similar immobility and control of movement from
Melba, Patti, or Tetrazzini, forcing an obedience to technical requi-
sites on their celebrated habits of performance.
Gaisberg wrote liberally about his discomfort at recording female
singers. Here is an excerpt from his 1942 memoirs:

One of the tensest moments I have ever experienced in any recording studio was
when, at City Road, Melba had just completed a charming record of “Caro
nome.” As the operator lifted the wax from the turntable, his hands trembled so
much that he let the disc fall and it rolled along the floor on its edge. Everybody
was aghast, silently watching its progress and wondering on which side it would
come to a rest. If on its face, the wax would be ruined. Luckily, it fell on its
back. Then Melba’s pent-up feelings were let loose in a tirade in which she told
the poor operator in the plainest terms just what she thought of his clumsiness.

And here is another example indicative of female trouble. Adelina

Patti’s 1906 recording of Marguerite’s cabaletta “Ah, je ris” from
Gounod’s Faust. The singer recorded the piece, but at the very end
she murmured, ruining the matrix. A second take had to be produced,
one that notoriously traded freshness for a musically paralyzing con-

31 Beyond the female voice, the topic of what kind of sound belonged in the elec-
tric medium was of intense interest to early twentieth-century composers, sound
engineers, filmmakers, and critics. See Douglas Khan, Noise, Water, Meat. A
History of Sound in the Arts (Cambridge, MA: MIT, 1999), 123–57.
32 Frederick W. Gaisberg, Music Goes Round (New York, NY: MacMillan, 1942),
30 Gabriela Cruz

trol. The murmur that eliminated the first recorded effort follows the
last high B natural, sung full voice, fortissimo, reached from the F
sharp below. The high note, inscribed with distortion in the shellac
surface, is one of those lyrical pearls that recorded badly back then,
requiring the singer to step away from the recording cone. As far as
the recording scene can be reconstructed from the inscription on the
disc, Patti forgot to pull back, Gaisberg nudged her away, and she did
not like it.
Gaisberg did not include Luisa Tetrazzini, his friend of many
years, in his personal litany of recording woes. He preferred to re-
member her as a force of nature, triumphing among Argentinean mag-
nates or during the odd publicity stunt, at the large cabinet factory of
His Master’s Voice at Hayes in England, singing “to the working girls
during their dinner hour.” But it was the Italian diva, a long-time
friend of Caruso, with whom she had shared home remedies for the
larynx and the authorship of a slim but popular manual titled The Art
of Singing Naturally, who came closest to playing the role of lyrical
stepsister to Caruso’s Cinderella for posterity.
A short newsreel from 1932, when sound was still relatively new
to film, returns us to the notion of the perfect fit. In the film, the 61-
year-old Luisa Tetrazzini, long retired, and voiceless for all operatic
purposes, sits by a new electric Victor Victrola to listen to the singing
voice of the deceased Enrico. Sitting to her left is Fred Gaisberg. Diva
and engineer are seen concentrating on an electrically remastered copy
of Caruso’s rendition of Flotow’s “M’apparì,”recorded by Gaisberg
probably in 1906 in the old Victor Talking Machine Company studios
in Camden, New Jersey. The old diva seems delighted, and in a staged
moment of the extraordinary; she even joins in the singing, vocalizing
the final, beloved high notes of Lyonel’s lament “Tu la pace mi
rapisti, di dolor io morirò. Ah, di dolor morirò!” in calculated admira-

33 Adelina Patti, CD Symposium 1324, Vol. 14, Tracks 5–6. For an account of the
engineer’s stay in Craig-Y-Nos Castle in Wales, see Gaisberg, Music Goes
Round, 114–16.
34 Gaisberg, Music Goes Round, 109.
35 Luisa Tetrazzini, The Art of Singing: Golden Voice of Singing (DVD.NVC Arts,
1999), Track 8.
The Fairy Tale of Belcanto 31

tion for the sound that the electric gramophone casts as lyrical perfec-
tion. Conceived both as cultural piece and as an advertisement for the
new acoustic quality afforded by recent electric technology of repro-
duction, the little film also does the work of conciliation with the ma-
chine. First, through its repertory. “M’apparì” is not just any song
once recorded by the great Caruso. It is first and foremost an exercise
in lyrical emission and control, one that Caruso and colleagues (Ben-
jamino Gigli is a documented case) were known to use as a diagnostic
of their own vocal health, testing their production of beautiful tone,
phrasing, and color. The song imagined as diagnostic tool addressed
the human voice as technical apparatus. The newsreel alluded to this
mechanical understanding of the voice sedimented in modern lyrical
practice. But the film also identified good and bad objects of phono-
graphy. It absorbed the exemplary voice of Caruso into the machine,
and ejected the diva, outed as caricature, an old woman (as old as the
recorded Patti) who would not measure up to the sound inscription.
Here, Tetrazzini performs the role of the stepsister, a certainly inad-
vertent albeit curious lyrical precedent for Disney’s later scene. Fi-
nally, the film advocates a new condition for opera, not as live event
but as inscribed sound. The figure of repetition celebrates the inscrip-
tion. Still, this is an awkward scene, one that eludes the inherent life-
lessness of the sound decoded out of the grooves of the gramophone.
The recorded singing voice of Caruso is naturalized, staged as a live
event, protected by a constellation of natural pieties – living witnesses
to its authenticity, the warmth of friendship, and the depth of living
memory – which deflect the listener’s attention from what bel canto
has become in modernity: a dead art.

36 Dorothy Caruso, Enrico Caruso – His Life and Death (Read Books, 2007), 266.
Alfred Tomatis, Roberta Prada trans. The Ear and The Voice (Lanham, Md.:
Scarecrow Press, 2005), 31.
32 Gabriela Cruz

Herbaria of Artificial Life

Adorno was most attentive to this quality of deadness. In “The Form

of the Phonograph Record” (1934), he called recordings “herbaria of
artificial life” and added:

There is no doubt that, as music is removed by the phonograph record from the
realm of live production and from the imperative of artistic activity and becomes
petrified, it absorbs into itself, in this process of petrification, the very life that
would otherwise vanish. The dead art rescues the ephemeral and the perishing
art as the only one alive.

The notion of mineral petrification returns us to fairy-tale magic: the

shoe of Zezolla darting incomprehensibly to the foot like iron to a
magnet in the Pentamerone, and the glass slipper that envelops Cen-
drillon’s living foot in Perrault’s version. But what had been a stroke
of magic, a moment detached from everyday possibility, presents it-
self to Adorno as part of a progressive quotidian. Thomas Levin has
explored the philosopher’s argument pressing for a contemplation of
song as acoustic enigma. The acoustic traces played on the gramo-
phone are a form of writing that is not readily intelligible; rather,
“they are indexical and enigmatic” like a hieroglyph, at once natural,
immediate, esoteric, and inaccessible, requiring decoding.
Bel canto, read in this manner, emerges as a different art, an alle-
gory of singing, song writing itself, deriving its truth from an inherent
deadness. The notion contains a provocation for opera production in
the age of video, an implicit call for the abandonment of counterfeit
liveness in light of a poetics of the inaccessible and the inscrutable.
Adorno, who in 1927 had pointedly noted the conservative function of
records as “ideologies,” i.e., objects that register the desire of the
bourgeois subject for self-recognition in the acoustic mirror, moved
on to a substantially different critical position in his “The Form of the

37 Theodor Adorno, “The Form of the Phonograph Record”, trans. Thomas Levin
October 55/1 (1990), 56–61: 59.
38 Thomas Levin, “For the Record: Adorno on Music in the Age of Its Technologi-
cal Reproducibility”, October 55/ 1 (1990), 23–47: 37.
The Fairy Tale of Belcanto 33

Phonograph Record” of 1934, one more adventurous and more accept-

ing of the new culture of sound reproduction. Finally, and thus far
unremarked, his contemplation of the enigmatic quality of recording
absorbs into critical discourse another equally compelling statement
on the subject, which was composed by Kurt Weill in 1927, first
staged in 1928, and which I address below.

The Gramophone on Stage

Der Zar läßt sich photographieren (1928), Weill’s one-act opera

buffa, with libretto by Georg Kaiser, offers an elaborate meditation on
bel canto in light of mechanical reproduction. The plot has two ma-
chines at its center: a camera and a gramophone. The Czar, mere
history by 1928, finds himself in the Parisian studio of photographer
Angèle, to have his portrait taken. But Angèle has been kidnapped by
terrorists and replaced by a false photographer who is intent on shoot-
ing the Czar dead by pressing the trigger of a pistol rigged in the eye
of the camera. The comedy of errors takes place around the altered
machine: attempts at seduction (of False-Angèle by the Czar) and at
murder (of the Czar by False-Angèle) mutually frustrate one another.
Finally, a telephone call announces that the police have unraveled the
murder plot and that officers will soon appear. Angèle places a record
on the gramophone, pretending to assent to the Czar’s designs on her,
while in fact she prepares her escape.
Weill’s opera buffa has long been criticized for its narrow con-
temporary relevance. Adorno wrote in 1928 that it was a work of the
“here and now,” not of the future. He recognized in the work a com-
mitment to “actuality,” that which Weill also described as an essential

39 For a discussion of the nexus between popular music, sound media, and music
aesthetics in Weill’s work, see Alexander Rehding, “On the Record”, Cam-
bridge Opera Journal 18/1 (2006), 59–82.
40 Theodor Adorno, Musikalischen Schriften VI, Gesammelte Schriften XIX (Frank-
furt-Main, Germany: Suhrkamp, 1997), 135.
34 Gabriela Cruz

element of contemporary opera. In light of the composer’s well-
advertised concern for the present-day, the substitution of the love
duet by a scene sung exclusively to the phonogram “Tango Angèle”
has been imagined as a new form of death by technics, lyrical fulfill-
ment downgraded to a mode of musical consumption. Recent writing
on the scene generally forecloses critical consideration of the potential
of technology for opera. Yet, the question of what the gramophone
may deliver to opera and to the stage is central to Weill’s composition.
Exactly one year after Ernst Krenek first put a gramophone on the
stage of Leipzig’s Stadtstheater, for the premiere of Jonny spielt auf,
Weill did the same in the Neues Theater. In anticipation of a critical
reaction, he wrote in the Leipziger Bühnenblätter (1927/1928):

Finally, I believe that I could only achieve this inward intensification, which I
envisioned for the escape scene, through a complete change of color. Thus, I ar-
rived at the inclusion of the gramophone scene, in which I gave a plot-driven
meaning to a mechanical instrument and to dance music. For this Tango Angèle
(as I called it) I could now save the sax and jazz sounds. After careful gramo-
phone studies, I did a specialized instrumentation for the gramophone and re-
corded this dance piece with Lindström A. G.

Weill noted that this gramophone was no mere ornament; it performed

a dramaturgical function, harnessing the exact chronometry of the

41 Richard Taruskin, “The Golden Age of Kitsch”, in The Danger of Music and
Other Anti-Utopian Essays (Berkeley, CA: University of California Press,
2009), 241–60: 255. Weill’s statement that the genre should strive to serve as a
“concave or convex mirror, showing life with the same magnitude or smallness
with which it appears in reality” cements the point. Kurt Weill, Gesammelte
Schriften, ed. Kim Kowalke (Berlin, Germany: Kurt Weill Foundation, 1990),
42 See Mario Mercado, “Kurt Weill and the Tango“, in “Neue Welt”/”Dritte
Welt”: Interkulturelle Beziehungen Deutschlands zu Lateinamerika und der Ka-
ribik, ed. Sigrid Bauschinger and Susan L. Cocalis (Tübingen, Germany:
Francke Verlag, 1994), 97–107: 104–105; Guido Heldt, “Austria and Germany,
1918–1960”, in Mervyn Cooke, The Cambridge Companion to Twentieth-
Century Opera (Cambridge, UK: Cambridge University Press, 2005), 146–64:
157, and Rehding, “On the Record”, 71–72.
43 Rehding, “On the Record”, 78.
44 Weill, Gesammelte Schriften, 51.
The Fairy Tale of Belcanto 35

reproduced disc and the precise kinetic quality of jazz to announce

and reflect the motion of False-Angèle’s flight. Thus, he combined
machine and character. But unlike more prevalent fantasies of the
post-war cultural industry, where man-made objects are often hu-
manlike, contaminated by life and feeling (Disney’s mirror in Snow
White or the teapot in Beauty and the Beast), Weill stressed a parallel
track to the anthropomorphized machine, asking us to wonder about
the enigma of a commonality between people and the industrial ob-
ject. The phonogram, he argued, was above all a means towards “in-
ward intensification.” How does this work? In the theater, rather than
in the listening experience provided by disc or DVD, the sudden si-
lencing of the orchestra, and the playing of any gramophone made in
1928 with its limited amplification capacity, is perceived as an odd
shift. Such stunts, first deployed by Krenek and Weill, are now com-
mon on stage – Kupfer’s Orfeo is one of many examples – and ex-
pertly employed as provocations of feeling, opening the listener to
critical uncertainty or, in other words, to contemporaneity on stage.

The Angel of Song

The recorded tango invites critical attention but, I would argue, not of
the kind paid to the exotic, commercial, and popular object meant to
enthrall the famously fickle Weimar Republic. “Tango Angèle” re-
fers back to operatic history and addresses the one fundamental
branch of philosophical reflection routinely declared irrelevant to Zeit-
oper: metaphysics. The playing on stage of a recorded song instead of
a live action “love duet” charts an alternative to Romantic form and its
modes of symbolization. Whereas the sentimental lyrical duet had
delivered a formula of isomorphic correspondences, uniting appear-

45 Mario Mercado, “Kurt Weill and the Tango”, 104–105; Jürgen Arndt, “Tango
und Technik: Kurt Weills Rezeption des Amerikanismus der Weimarer Repub-
lik”, in Werner Keil (ed.) Musik der Zwanziger Jahre (Hindesheim: Georg Oms,
1996), 42–58; Rehding, “On the Record”, 72–73.
36 Gabriela Cruz

ance and essence, love and redemption, the recorded tango exposes
the theater audience to mechanical automatism and to simple musical
kinetics offered as a new means of symbolic encryption. Here, tran-
scendence is etched so as to be decoded. It is an idea written first in
the plot and the title of the phonogram. Photographer Angèle, Terror-
ist Angèle (False-Angèle), Tango Angèle are by order of appearance
forged echoes of each other, naming an earlier, now absent, image that
the Gramophone Company first substituted for the classical icon of
divine song: the recording angel, the company’s brand retired from the
market in 1900. All Angèles look back at the first industrial knockoff
of lyrical power, drawn and printed on the labels that once graced the
shellac of discs recorded by Gaisberg, Caruso’s and Patti’s included.
While ruins, the Angèles remain objects of attraction, fashionably
accented in French, at once absurd forgeries of lyrical flight, desirable
consumer goods, and reminders of a lost horizon of otherworldliness.
How do these phony angels, abstract and suspiciously material,
sustain a trajectory of inward intensification? Precisely this way:
False-Angèle places the disc on the machine. The audience and char-
acters on stage hear a tango to which no one dances. Instead, False-
Angèle sings with and to the recording. The moment appears to pro-
vide a theatrical demonstration of phonographic theory, offering what
is at first sight a musical elaboration of Adorno’s 1927 statement that:

What the gramophone listener actually wants to hear is himself, and the artist
merely offers him a substitute for the sounding image of his own person, which
he would like to safeguard as a possession. The only reason that he accords the
record such value is because he himself could also be just as well preserved.
Most of the time, records are virtual photographs of their owners, flattering pho-
tographs – ideologies.

Appearances deceive, however. The mirror image commonly found at

the core of the modern imagination of sound reproduction – think of
Cinderella’s song – is not featured in the musical relationship of the
two Angèles. In the program notes for the opera, Weill called attention

46 This point is made by Rehding, “On the Record”, 62.

47 Adorno, “Curves of the Needle”, 54.
The Fairy Tale of Belcanto 37

to the Other in the electrified gramophone. Researching for “inward
intensification,” he settled not on the voice double, but on the saxo-
phone and on jazz, writing a “specialized instrumentation” for the
reproduction device now elevated to musical instrument. Thus, he
followed a compositional path of de-naturalization, pointedly discard-
ing the trope of the anthropomorphized machine. The saxophone,
patented in France by Adophe Sax in 1846, and by the 1920s an in-
strument of impeccable industrial lineage and popular proclivities,
colors the singing line delivered by the gramophone with further in-
human hues. Angèle on stage sings live music, while the dead trace of
recorded Angèle is decoded as sound. This is a scene of song that does
not stage the collaboration of living and dead voices in Adornian
terms, as a process of subordination that repeats the mythic relation-
ship of Narcissus and Echo. Instead, the two Angèle’s are offered as
complementary figures in the operatic scene, but remain ontologically
and acoustically distinct, one living and embodied, the other inanimate
and mechanically produced and reproduced. The intimacy forged be-
tween the two restores a confidence in the permeable, elevated here
over a pedigreed belief in natural boundaries and unspoiled essences.
The hybrid nature of the relationship posited on stage, offered in the
collaboration between the singing body and the machine, evokes
magic, the perfect fit of slipper to living foot in the baroque fairy tale.
And the scene mobilizes still another important baroque thread:
the angelic. The angels on stage, weirdly alive and dead, uncertainly
genuine and false, evoke the famously unreliable nature of seraphic
creatures, long suspected of deceiving form: in appearance human, in
substance air, and of another vocality. Saint Augustine’s thesis that
“God does not speak to the angels in the way that we speak to each
other, or to God, or to the angels” and that “when we grasp something
of this kind of speech with our inward ears, we ourselves become like

48 Weill’s interest in mechanical music, including the gramophone and radio, was
professionally motivated and finds striking parallels in the attention other Ger-
man composers dedicated to sound media at the time. See Christopher Hailey,
“Rethinking Sound: Music and Radio in Weimar Germany”, in Music and Per-
formance during the Weimar Republic, ed. Bryan R. Gilliam (Cambridge, UK:
Cambridge University Press, 1994), 13–36: 33.
38 Gabriela Cruz

angels,” is finally, if obliquely, at play in Weill’s consideration of the

new cultural figure of reproduction. In Der Zar, phonogram and
singer replay for modernity the ancient attraction to the material ci-
pher, articulating musically an old concern for the hieroglyph.
This is a scene of baroque reason. Simply put, the modern preoc-
cupation with the ancient Egyptian symbol, radically inscrutable, goes
back to the seventeenth-century. Athanasius Kircher, Jesuit priest,
polymath, and inaugural Egyptologist, thought hieroglyphs a reminder
of a sacred truth “whose force derived its own impenetrability to the
eyes of the profane.” In Obeliscus Pampilius (1650), he stressed the
allegorical nature of its symbolical operation:

a nota significativa of mysteries, that is to say, that it is the nature of the symbol
to lead our minds by means of certain similarities, to the understanding of things
vastly different from the things that are offered to our external senses, and
whose property is to appear hidden under the veil of an obscure expression.

Kircher’s description of the hieroglyph as a sign with a double sense,

outwardly an object material and obscure and inwardly a property of
illumination, is relevant to our phonographic scene. This is, of course,
not because Weill or Kaiser cultivated a special interest in the arcane
scholarship of the seventeenth-century Jesuit, or meant to bring opera
closer to the hermetic world of Egyptology. More simply, their han-
dling of the scene of phonography in light of a poetics of allegory –
my object of discussion below – brings us back to Walter Benjamin’s
philosophical salvaging of allegory from the forgotten depths of ba-
roque drama and to his defense of its operation as profoundly illumi-
nating for the modern experience, an experience to which both opera
and phonography belong.

49 Saint Augustine, The City of God against the Pagans, trans. R. W. Dyson (Cam-
bridge, UK: Cambridge University Press, 1998), 705.
50 Umberto Eco, The Search for the Perfect Language, trans. James Fentress (Ox-
ford, UK:Willey-Blackwell, 1997), 154.
51 Athanasius Kircher, Obeliscus Pampilius, II, 5, 114–20, cited in Eco, The
Search for the Perfect Language, 154.
The Fairy Tale of Belcanto 39


Like a woman of bad repute – the modern body of the female prosti-
tute excavated by Baudelaire and held up to critical scrutiny by Ben-
jamin – “Tango Angèle” consistently attracts the wrong kind of atten-
tion. On disc, it is an object that calls for consideration of style and
artifice in music. It is outwardly a piece of fastidious up-to-dateness,
its newness signalled in fashionable dissonance and in the persistent
anacruses delivered in the staccato-sforzando characteristic of re-
corded early jazz. The piece may strike the listener more like a pose
than an idea, a figure of sound that petrifies expression, offering mind-
less movement in its substitution. Thus, Alex Rehding has called at-
tention to its frivolous spirit, which he sees as a reflection of the mood
of the Republic notoriously attracted by the surfaces of the popular
and the alien, remastered in commodified experience.
And yet, within and around the music that spirals out of the pho-
nogram, intimations of “things vastly different” emerge. Even before
the machine is set to work, a stage indication reads, “The whole
gramophone scene is to be sung very sweetly and softly,” very much
unlike a danced tango. After the initial cadential chords, establishing
E as the tonal center of the piece, the dance unfolds with propulsive
certitude. Yet, a shadow of another musical register settles progres-
sively in the ear. The accompanying chordal texture of the tango in-
cludes fragments of chromatic descent. It first unfolds a descending
line spanning a major third (C#-C-B-A#-A) underneath the Czar’s
opening interrogation, “Music, at this moment?” Then it continues
with a similar downward figure (E-D#-D-C#-C) written into the musi-
cal accompaniment to Angèle’s answer, “It is the Tango Angèle.”
Finally, the chromatic unfolding of the minor third (D-C#-C-B) un-
derscores False-Angèle’s physical movement to close the door. The
tango, which outwardly instructs us to dance, whispers of something
other, which the musicological ear recognizes as part of a distant hori-
zon of experience, alien to modern dance. The chromatic fragments
are ruins, incomplete and enigmatic, that bring to aural presence a lost

52 See Ex 1,
40 Gabriela Cruz

object. This is a puzzle easy to solve, requiring the simple re-ordering

of what was heard. The solution to the puzzle delivers a familiar ob-
ject: the chromatic descending tetrachord, emblem of the baroque
lament, here filling in the interval of a perfect fifth, instead of the ex-
pected perfect fourth. (E-D#-D-C#-C/ E-D#-D-C#-C/C#-C-B-A#-A).
The beginning segment of the phonographic tango underpins mu-
sically a circumstantial exchange between the Czar and Angèle and
the closing of a stage door which symbolically isolates the scene from
the outside world. The music does double work. It is written in a new
idiom of cool detachment, opaque to feeling and outwardly addressed
to the music market, but it is secretly grounded in a timeless emblem
of emoting beauty. More than a simple inscription, it is a nota signifi-
cativa, sound externally prosaic and inwardly inspired, a frozen juxta-
position of irreconcilable horizons, conceived after the allegorical
fashion that Walter Benjamin reclaimed for the density of modern
experience. Thus, this tango surveys modernity and simultaneously
intimates its cost. It names the tradition of beauty ruined in the wake
of modernity’s fury, a tradition that the composer cannot restore but
offers for contemplation in the incompleteness of its vestiges, the
skeletal notes that pour out of the gramophone along with the noise of
the present. The descending chromatic figure registers a constellation
of losses: of song, voice, and grievance. This music, superficially
mindless and automated, invites us to listen in melancholy absorption
to present cacophony in search of a vanished past of lyrical introspec-
tion and catharsis, preserved in vestiges.
Angèle locks the door and a lyrical outline begins to emerge, first
carried by a very short lyrical gesture in the violin solo, then extended
in a two-part harmonization for saxophone duet (first saxophone: E-E-
F-G-E-D) on the recording. The little melody coalesces around the
idea of a singing voice, a sound marked as from elsewhere, born out
of melodic invention after a baroque fashion, mono-thematic in de-
sign, and centered on the figure of the sigh, the chief emblem of lam-
entation described by the falling second. On stage, the dialogue con-
tinues matter-of-factly:

53 See Ex. 2

The Fairy Tale of Belcanto 41

Czar: What are you doing?

False-Angèle: I’ve locked the door.
Czar: But why?
False-Angèle: If the police should burst in and surprise us while we’re embrac-
Czar: At last you’re willing to surrender! […] But no conditions. No

[Der Tzar – Was tuen Sie?/ F. Angèle – Ich schloß die Tür/ Der Tzar – Warum?/F.
Angèle: Die Polizei soll uns nicht uberraschen, wenn wir uns lieben./ Der Tzar –
Gewährung noch im letzten Augenblick? […] Bedingungslos und ohne Photogra-

False-Angèle, poised between the utter banality of dramatic inter-

course and the lyrical pull of the gramophone, behaves in a double
sense. She answers the Czar but re-orients her musical expression
towards her recorded homonym. She is no common Cinderella, bound
to mirror the voice in the disc. Instead, she is singly attracted to the
enigma of the inscription, concentrating solely on the sound of a lyri-
cal sigh, which she echoes, complements, and amplifies in live per-
formance. How does this happen? She reacts to the descending second
decoded by the machine by essaying an incipient gesture of melodic
flight (A-D-E-F, sung to “I’ve locked the door”). Her melodic line
extends the lamenting phrase of the first saxophone and sets the char-
acter/singer in a path of pure music delivered from the semiotic con-
straints of verbal signification. Then she echoes and amplifies the
“sigh” ever more dramatically, from “wir uns lieben,” “selig” to “Nur
Liebe!” and quietly again in “Geliebter”.
Singing coalesces in a statement of love, a dramatic expedient
seemingly addressed to the Czar, but actually meant for and sung to
the machine. The amorous expression records a fatal attraction to the
acoustic enigma decoded by the gramophone and presented in the
theater and to the singer as both immediate and inaccessible. Voice
touches petrified song in “Nur liebe! Nur liebe! [Only love! Only
love!].” It acquires a shadow, as if found in the process of melding
with the acoustic traces of the Dobbri saxophones that issue from the
grooves. The display of attraction is not towards music, but towards

54 See Ex. 3,

42 Gabriela Cruz

the traces of music retained in the reproduced sound of the wind in-
struments – ghostly fragments of voice that allude to a lost historical
object, but that also index the indistinct quality of the high lyrical
voice in early acoustic recordings. The trace becomes evanescent,
capturing a moving fragility even in death. It transposes to the gramo-
phone a hallmark of photographic technique: the purposeful exploita-
tion of fading of contrast and form in light of an undecided elegance
and distortion.
It is particularly tempting at this point of the performance to hear
in the duet between False-Angèle and recorded Angèle a final instance
of the perfect fit, pairing live singing and fossilized song. On stage, as
always, the challenge rests with the soprano. Will she measure up to
the recording? Will she turn out to be a Cinderella, or are we in for a
show of mutilation? This is a scene of high stakes for the singer, asked
to search for love not in herself, but in an unknown quantity mechani-
cally inscribed in shellac, and to make sense of it. Never mind the
suggestion of modern enchantment – the implication that shellac, the
costly exotic material, might be magically animated into life just like
Cinderella’s shoe. Phonography, as Adorno well knew in 1934, prom-
ises to enliven the world of things only on the condition that we face
up to its deadness.

Opera as/in the Scene of Modernity

Kurt Weill conceived “Tango Angèle” as an allegory of song, a musi-

cal contemplation of the distance separating the new age of mechani-
cal reproduction from that of pre-industrial bel canto. Thus, he carved
out a new space for opera in modern experience, one radically differ-
ent from the one increasingly assigned to it by his contemporaries. At
a time (the 1920s) when opera houses were becoming increasingly
museological in outlook, mindful of the preservation of a specific reper-
tory of lyrical treasures, a repertory catalogued by authoring genius,
The Fairy Tale of Belcanto 43

style and national affiliation, Weill offered opera as a state of “dream-

ing-while-awake,” a form of lyrical consciousness touched by surreal-
ism. He carved out for lyrical song a new space of productive encoun-
ter, found halfway between body and symbol. His tango is offered as a
scene of irreconcilable tension, an inscription of a trivial and oppressive
present interrupted by a found object, a ruin that in the mode of a dream
liberates into the present a vital energy and feeling long forgotten.
The gramophone placed on stage inscribes within the material
scene of opera the circular, unprogressive logic that is the hidden face
of modern progress. That which Esther Leslei, following Benjamin,
has described as “a wheel without beginning and end, whose fateful
destructive/productive dynamic could ultimately only be ripped apart
– ruined again.” Leslei’s observation confirms a present-day sense of
the gramophone, so unlike that of 1928. This machine threatens no
one today. It has become another ruin of musicality belonging to the
“gay twenties,” one into which scholars of the Weimar Republic regu-
larly peer in half-distracted curiosity. In 1928, however, the machine
assigned to play “Tango Angèle” allowed for a double exposure to
lyrical song, envisaging musical composition as a form of historical
insight. The scene laid out a progressive narrative of history, of opera
commodified and altered by the endless and totalizing pull of the new
– of democratized taste, of advanced dissonance, of stylistic progress,
of a new utopian stake in fidelity. Simultaneously, it invited listeners
to face up to what the totalizing narrative of the new occluded. For the
tango also outlined what progress had rejected and cast aside: the lyri-
cal lament, a song of a time and a place now vanished but for a few
petrified traces left among the rubble of history. Of such traces,
Adorno explained, following Benjamin’s theory of history with in-
delible rigor, that they “absorb into [themselves], in the process of
petrification, the very life that would otherwise vanish.”
Weill faces up to the task of imagining the life contained in the
petrified trace with philosophical courage. He imagines a new purpose
for composition as that of freeing the remaining vitality caught in the
trace. What does it take to animate the trace? The intervention of a

55 Esther Leslie, Walter Benjamin (London, UK: Reaktion Books, 2007), 196.
56 Adorno, “The Form of the Phonograph Record”, 59.
44 Gabriela Cruz

dream, Angèle turned into an angel, a being both estranged and myste-
rious, in appearance mundane and of this world, but in essence of
elsewhere. The dream reveals the essence of Angèle: a seraphic crea-
ture whom we know capable of divine speech, but whose voice, ac-
cording to Augustinian wisdom, sounds out only to those blessed with
a special inward sense.
“Tango Angèle” is offered as a piece of historical clairvoyance. It
restores to opera, boxed in the panoptic discourse of historical pro-
gress, a sense of its pastness, expressed in moving fragility and an
awareness of loss. Clairvoyance in this sense is the outcome of crea-
tive labor – the labor of the singer who re-voices song after the en-
crypted object, which she decodes and of which she becomes a source.
The work of re-sounding acoustic ruins is a special and enigmatic one.
It retains a quality of wonderment and an oracular dimension, even if
it emphasizes modern exactness, a new form of musical philological
absorption, and specially, as Angèle and Cinderella’s songs insist, the
melancholic discipline of singing to and against oneself. The golden
age for this form of hieratic lyricism may seem now past – the heyday
of Stokowski’s electric lady were the post-World War II years, with
Maria Callas’s diet as an cherished epicenter – but only because our
experience of opera has become increasingly re-envisioned by digital
re-mediation. Video imposes an orientation towards visuality, in light
of which music and re-mediated performance have accrued a new
illusory liveness, or truth. Thus, we commonly lose sight of the acous-
tic cipher at the center of our own auditory experience and forget that
the critical statement which bel canto (re)formed by phonography
produces for our age is not about the all-encompassing powers of song
– the illusion that music inscribed in transmission reconnects us to the
world and to a past that becomes transparent in the machine, as sug-
gested by Cinderella’s song – but about the cultivation of an empathic
and generous musicality open to the contemplation of untimely pas-
sion. A musicality predicated not solely on the pleasures of immedi-
acy, but on the love of leisurely cogitation about song, of pondering
the singularities of the past that survive in the material distortion of an
etching. It is this precise point that the awkward, uncomfortable and
nearly forgotten protest of the elderly Patti performing the Jewel song
for posterity illuminates.
Opera on Italian television: The First Thirty Years,


There has been a lot of opera on Italian television over the past half
century. Yet, this substantial presence has not received much scholarly
attention, including in Italy. What is more, the few publications on
this phenomenon are all by historians and critics of television or mass
media in general rather than musicologists. It seems high time to re-
flect on this topic in some depth, then, and to do so from a musico-
logical point of view. There are several good reasons for this, above
and beyond the sheer number of opera-related programmes on Italian
television. First, the relevance of both opera and television for Italian
culture of the last half century is undisputed by the many who have
studied either one or the other; it makes sense, therefore, to bring them
together. What is more, from a theoretical point of view, discussing
the relationship between opera and television opens up new and chal-
lenging perspectives for the study of opera on video. Finally, observ-
ing this phenomenon with musicological eyes may help not only to
reveal unexplored ways in which image and music interact, but also to

1 See Giovanni Buttafava and Aldo Grasso, La camera lirica. Storia e tendenze
della diffusione dell’opera lirica attraverso la televisione (Milan: Amici della
Scala, 1986), 13–32; Giorgio Simonelli, ‘Evoluzione storica del teleteatro’, in
Sipario! Storia e modelli del teatro televisivo in Italia, ed. Gianfranco Bettetini
(Rome: RAI, 1989), 73–100, especially 94–98; Luisella Bolla and Flaminia
Cardini, Macchina sonora. La musica nella televisione italiana (Rome: Rai-ERI,
1997), 158–163, 198–201; Claudia Polo, Immaginari verdiani. Opera, media e
industria culturale nell’Italia del XX secolo (Rome and Milan: Accademia Na-
zionale di Santa Cecilia – Ricordi, 2004), 138–161. For some further references,
see footnote 6 of Emanuele Senici, ‘Il video d’opera “dal vivo”: testualizzazione
e liveness nell’era digitale’, Il Saggiatore Musicale 16 (2009), 273–312.
46 Emanuele Senici

investigate how the discourse of opera was influenced by that of tele-

vision, rather than the other way around, as has been the case thus far.
This study entails significant risks precisely for a musicologist,
however. The first among them is of unduly overestimating the impor-
tance of opera for Italian television. It is better, then, to make it clear
from the start that, although opera did give a very important contribu-
tion to the beginnings of Italian television, after those first crucial years
its role was reduced to that of a comprimario – like an old aunt whose
generosity helped us find our place in the world and whom we must
make a show of remembering for her birthday and at Christmas, but
whom we can more or less remorselessly forget for the rest of the year.
If opera did not count for much on Italian television after the
1950s, however, Italian television had a massive impact on opera, an
impact still strongly felt. For one thing, television appropriated a con-
siderable share of spectators who, before its generalized spread in the
1960s, used to attend opera performances (or film screenings), and
since then have tended to stay at home in front of a television set. This
is a theme for sociologists and mediologists rather than musicologists,
and therefore I will not address it here. There are other ways in which
television has had a strong influence on opera, however, especially
regarding its cultural and aesthetic aspects. It is on this kind of influ-
ence that I will focus my attention, not least because an opera scholar
is probably better equipped to discuss it than a television one.
In light of these considerations, it seems unwise to attempt a tra-
ditional history of opera on Italian television, treating opera almost as
if it were the protagonist of a realist play, with a coherent dramatic
profile and a linear psychological evolution. An approach better suited
to the nature of the object might consist in focusing on a few espe-
cially revealing moments. Among the several possible ones, I have
chosen three that appear particularly instructive, associated with
events that took place on precise dates: they function as specific and
localized points of view from which to observe significantly longer
periods. Before delving into them, however, some general contextu-
alization is in order.
Italian television literally began with opera. Starting from the
very first day of official television broadcasting, 3 January 1954, and
continuing until the 1970s, the tune that signalled the start of trans-
Opera on Italian Television 47

missions each day was an instrumental arrangement of the closing

ensemble of Rossini’s William Tell, ‘Tutto cangia, il ciel si abbella’
(‘Everything changes, the sky becomes beautiful’) in the version of
the libretto current in Italy at that time. This music accompanied im-
ages of swirling clouds that, in Cormac Newark’s description, ‘gradu-
ally cleared to reveal a transmitter giving off dynamic-looking pulses
of light’. In the opera, as the stage directions indicate, ‘the clouds
slowly disappear and the sky clears up’, and ‘everything looks sweet
and new’ to the ‘uncertain, wandering gaze’ of the characters. Televi-
sion used Rossini’s music, then, to introduce itself as a means of
enlightenment for the minds of Italians.
This would prove an auspicious beginning for opera, since mate-
rials, moments and figures related to this art form have appeared on
Italian television in widely different guises but probably in higher
doses than in any other country, with the possible exception of Austria
and Germany (it would be obviously very hard to come up with firmly
documented figures). Among the most memorable – for reasons that it
would be too long to explain here – I would include: a participant in a
game show, Lascia o raddoppia, being asked a question on which
Verdi’s opera includes the contrabassoon in its orchestra (1955); a
popular music group, Quartetto Cetra, singing a song about going to
La Scala, ‘In un vecchio palco della Scala’, in the context of a general
entertainment programme called Serata di gala (1960); soprano Anna
Moffo hosting an opera-themed show, Bel canto, during which she
interviews colleagues and sings with them (1962-1963); mezzo-
soprano Cecilia Bartoli launching her career with a remarkably accom-
plished performance of Cenerentola’s rondò from Rossini’s eponymous
opera within a Saturday-night entertainment show, Fantastico, hugely
popular in 1980s Italy (1985); Puccini, a very successful multi-
episode drama (1973); Verdi, another, even more successful such pro-
gramme (1982); the first concert of the Three Tenors, broadcast live
from the Baths of Caracalla, Rome, in July 1990 during the football
World Cup; the funeral of Luciano Pavarotti, relayed live from

2 Cormac Newark, ‘Guillaume Tell’, in The Cambridge Companion to Rossini,

ed. Emanuele Senici (Cambridge: Cambridge University Press, 2004), 175–185:
48 Emanuele Senici

Modena cathedral on 8 September 2007; and the special episode of

Che tempo che fa, a programme of interviews with famous people,
entirely dedicated to the opening of the La Scala season on 7 Decem-
ber 2009 with Bizet’s Carmen, with conductor Daniel Barenboim and
director Emma Dante among the guests. I will not discuss the rele-
vance of these ‘shards’ of opera for Italian television any further,
since in order to do these theme justice I would need an entire book. I
will only add that a study of the relationship between opera, television
and Italian society of the last half century would probably find its
most rewarding sources among these materials. In the present chapter
I will focus instead on the broadcasting of operas in their entirety, an
object better suited to musicological scrutiny.

23 April 1954

As it had already been the case in other countries, the first broadcast
of an entire opera took place on Italian television soon after the begin-
ning of official transmissions. On 23 April 1954 Rossini’s Il barbiere
di Siviglia was relayed live from the RAI studios in Milan. For the
three following seasons, until the summer of 1957, RAI, the Italian
state broadcasting corporation, televised an entire opera almost on a
monthly basis, always live, and almost always from the Milan studios.
After this initial surge of interest, opera transmissions from television
studios became progressively less frequent, until they all but disap-
peared after 1960, only partially replaced by live relays from some of

3 For a short history of opera on television, see Lionel Salter, ‘Television’, in The
New Grove Dictionary of Opera, ed. Stanley Sadie (London: Macmillan, 1992),
Oxford Music Online, <> (viewed 31 March
2009); Jennifer Barnes, ‘Opera and Musical Theatre Relays and Recordings’, a
section of the multi-authored entry ‘Television’, in The New Grove Dictionary
of Music and Musicians, ed. Stanley Sadie and John Tyrrell, 2nd ed. (London:
Macmillan, 2001), Oxford Music Online, <http://www.>
(viewed 31 March 2009), with further bibliography.
Opera on Italian Television 49

the most important Italian theatres (see a complete listing of studio

relays between 1954 and 1970 in the appendix).
Some of the reasons for the initial prominence of opera on televi-
sion, including the Italian one, can be meaningfully related to those
for which opera had acted as godmother to cinema a half-century ear-
lier, namely, the search for cultural prestige on one hand, and narra-
tive and dramatic models on the other – even if in the latter case opera
was just a sub-genre of theatre in general, with spoken plays receiving
the lion’s share of attention by fledgling television. More specific to
the Italian case was, I would suggest, an attempt to ‘re-make’ Italians
after the disaster of World War II through an artistic form that was
considered somehow ‘genetically’ Italian – an artistic form which,
after having already contributed to ‘making’ Italians after the unifica-
tion of the country in 1861, had reached the zenith of its popularity in
the first half of the twentieth century. Furthermore, I would add a de-
sire to emulate opera films, which had met with considerable success
in Italy during the decade immediately following the war. Historians
of Italian television have investigated in some depth the theatricality
of the new medium, testified by the weekly transmission of a play,
beginning on that fateful 3 January 1954 with Goldoni’s L’osteria
della posta. In general, the conclusions reached by these historians on
the relationship between theatre and television are also valid for op-
era. Here it may be more interesting to focus on the differences, or in
any case on the features proper to the operatic genre.
For a useful starting point we may turn to the opinions of an ex-
ceptional television viewer: theatre and film director Luchino Vi-
sconti. Informed by Giovanni Battista Meneghini, Maria Callas’s hus-

4 See Cristina Bragaglia and Fernaldo Di Giammatteo, Italia 1900–1900. L’opera

al cinema (Florence: La Nuova Italia, 1990), cap. 5, ‘Follie per l’opera’.
5 See, among other contributions, Valentina Valentini, Teatro in immagine, 2
vols. (Rome: Bulzoni, 1987); Sipario!; Sipario! 2. Sinergie videoteatrali e rifon-
dazione drammaturgica, ed. Annamaria Cascetta (Rome: RAI-Nuova ERI, 1991;
Sipario! Volume terzo. Teatro e televisione: modelli europei a confronto, ed.
Adriano Bellotto and Luigi Bellotto (Rome: RAI-ERI, 1996); Giorgio Tabanelli,
Il teatro in televisione, 2 vols. (Rome: RAI-ERI, 2002–2003). For an emblema-
tic case, see Eduardo. L’arte del teatro in televisione, ed. Antonella Ottai (Ro-
me: RAI-ERI, 2000).
50 Emanuele Senici

band, that RAI was trying to convince the soprano to take part in a
televised Traviata in the autumn of 1954, Visconti answered with the
following words:

I am flabbergasted by a piece of news you give me, that of television! How can
Maria let herself be seduced by such a dangerous and absurd project? [...] Have
you ever seen a theatre show on television? And an opera show at that? My
God! My God! In my opinion it is the ugliest, most unpleasant and anti-artistic
show you could ever see today. First of all, the present technical conditions of
the TV medium: horrible vision, terrible image definition, just a flat, soulless,
feeble grey. And let’s not mention the execution! Execution for which these
pseudo-directors, desperate for boldness and originality, confuse theatre with
cinema, opera with documentary. Just imagine the mess! And, what is worse, in
order to obtain an annoying interpretive liveliness (which they consider inspired)
they move the cameras in such a way as to induce seasickness. Moreover, do
you know that they do the opera in playback? That is to say, the opera has al-
ready been sung and recorded, and then the singers (I imagine Maria!) re-do the
whole opera (the staging) just moving their mouth! These things are for second-
raters, not for a true artist. And Maria would anticipate in such a barbarous and
compromising way her Traviata, new and important, letting herself be tele-seen
(what a nice word!!) badly photographed in close ups (which would need to be
thought through, carefully lit, well studied) and forced to be Violetta two or
three days in advance, and then doing like a fish in a bowl? [...] The television
medium is still so rudimentary. At least wait for colour, and for television to be
done by artists, and for these shows to be past the trial stage, as they are at pre-
sent! [...] Didn’t you see The Barber of Seville on TV? What a shame. You
could have got an idea! It gave me stomach trouble.

There is no doubt that Visconti’s polemic verve was partly fanned by

the fear of seeing Callas’s debut in Traviata taken away from his pro-
duction for La Scala, planned for the following May. In any case, the
director’s outburst highlights a few crucial issues of the relationship
between opera and television in its initial phase, issues well worth
considering more closely.
The first concerns the complex production process, which began
with the recording of the audio track. This was then played on loud-
speakers in the television studio while the singers acted their roles
miming the act of singing in front of the cameras, which relayed the

6 Letter dated 19 June1954, in Giovanni Battista Meneghini, Maria Callas mia

moglie, ed. Renzo Allegri (Milan: Rusconi, 1981), 180–182.
Opera on Italian Television 51

image live. The sound of the relay did not come from the studio, of
course, but directly from the recording. In terms of coordination be-
tween audio and video, the results of this process were very mixed
indeed, much more so than for the opera films of the time, which of
course were not relayed live. Moreover, in opera films singer and
actor were often not the same – the case of Sophia Loren as Aida
(with the voice of Renata Tebaldi) in Clemente Fracassi’s 1953 film is
well known, but a more convincing example would be Gina Lollo-
brigida as a very sexy Nedda (with the voice of Onelia Fineschi) in
Mario Costa’s 1948 Pagliacci. While this split between singer and
actor may have lowered expectations when it came to lip-synching
coordination, the situation was different on television, where, as far as
I know, the singer was always also the actor. Finally, even in its in-
fancy television was already the medium of the close-up, and its aes-
thetic already one of intimacy: therefore, the lack of lip-synching co-
ordination comes across much more strongly on it than in opera films,
in which the spectrum of shot types is much wider. All this did not
matter to spoken theatre, where both sound and image were relayed
live, and where, therefore, the aesthetic of television – which has been
often called ‘realist’, a definition that can be accepted as valid for the
purpose of the present chapter, even if I would want to question it in
other contexts – could be abided to much more easily.
The difficulties encountered when this aesthetic confronted opera
emerge on other levels as well. Confined in the narrow space of a
television studio, and therefore deprived of the spatial freedom of ci-
nema, and further limited by the liveness of the broadcast, directors
who either had already demonstrated their remarkable abilities in
other theatrical and television genres, or would do so later on, such as
Franco Enriquez, Mario Lanfranchi, Alessandro Brissoni and Anton
Giulio Majano, had a hard time confronting one of the most peculiar
features of opera, that is, the gap between represented time and per-
formance time. In the words of musicologist Carl Dahlhaus, this dis-
tinction can be partially mapped onto the difference between ‘recita-
tive, in which musical-formal time usually is approximately
commensurate with real time’ and ‘the passage of time in closed num-
bers’, which ‘is almost always irregular and in some sense rhapsodic’.
An important consequence of this difference is that ‘the alternation
52 Emanuele Senici

between flowing and halting action and speech in opera leads to the
dissociation of time into a formal passage of time manifested by the
duration of the performance and a passage of time within the action,
which the members of the audience must deduce from the progress of
events on the basis of everyday experience’. This dissociation is a
crucial feature of Italian opera of the long nineteenth century (albeit
less so for works from the late nineteenth and early twentieth century),
that is to say, the backbone of opera broadcasts on Italian television in
the 1950s. This means, though, that the majority of televised works
were far removed from the television aesthetic of realism, according
to which, among other things, represented time should be as close as
possible to performance time. Such divergence is felt most acutely
during arias that set an interior monologue, the most difficult hurdle
for television (and also for cinema, which, however, has more re-
sources to overcome it successfully). It seems relevant, then, to ob-
serve how these musical numbers were staged and shot in the 1950s.
A typical example is Violetta’s aria at the end of Act 1 of La
traviata, ‘Ah fors’è lui che l’anima’, in the broadcast for which RAI
had wanted Callas but in the end got Rosanna Carteri instead, relayed
in December 1954 with Enriquez’s direction. The number is struc-
tured in four movements, as follows:

Scena ‘È strano!... è strano… in core’

Cantabile ‘Ah fors’è lui che l’anima’
Tempo di mezzo ‘Follie!... follie!... delirio vano è questo!...’
Cabaletta ‘Sempre libera degg’io’

The cabaletta is repeated, as usual: during the bridge between its ex-
position and the repetition Alfredo’s voice is heard coming from the
wings (from ‘underneath the balcony’, according to the score). Scena
and tempo di mezzo are considered ‘kinetic’ movements, in which re-
presented time is fairly close to performance time. Cantabile and
cabaletta, on the contrary, are ‘static’, since here represented time

7 Carl Dahlhaus, ‘The Dramaturgy of Italian Opera’, in Opera in Theory and

Practice, Image and Myth, ed. Lorenzo Bianconi and Giorgio Pestelli (Chicago:
The University of Chicago Press, 2003), 73–150: 108–109.
Opera on Italian Television 53

slows down until it almost grinds to a halt, moving further away from
performance time. A brief analysis of the relationship between images
and music in this aria as staged and shot by Enriquez involves paying
attention to the placing of the cuts, to the different types of shots, and
to the movements of the cameras.
Cuts are very few, especially in light of what happens in most
other numbers; there are only four: at the beginning of the cantabile,
at the beginning of the tempo di mezzo, at the word ‘vortici’ within the
tempo di mezzo, and at the beginning of the repetition of the cabaletta.
The main criterion followed by the découpage is that of musical form,
then: only once do other parameters prevail, both musical – the arrival
of coloratura on ‘vortici’ – and visual – at ‘vortici’ Violetta moves
from a room to the next one; naturally a character’s actions might
have been – and most likely were – planned with the music in mind.
Moreover, the level at which segmentation operates is macro-formal:
the basic segment is a movement within a number. A technical con-
sideration might have contributed to this choice: more frequent cuts
could have been risky during a live broadcast, not to mention hard to
perform with such a small number of cameras, which, judging from
the final product, were no more than three. Whatever the reasons, the
end result is that the rhythm of the action feels rather relaxed.
As for shots, the first thing to notice is the stark differentiation
between the cantabile, where the close-up and extreme close-up rule,
and the cabaletta, where, on the contrary, full and medium shots are
the norm. Equally striking is the contrast between the limited camera
movements during the cantabile (for a good part of which the camera
does not move at all, staying glued to Violetta’s face instead), and the
frequent zooming and tracking shots of the cabaletta. Obviously, both
shots and camera movements are closely tied to Violetta’s actions: for
most of the cantabile she remains still, sitting on a sofa, while during
the cabaletta she runs from one room to the next, opens and closes
doors, stands up, sits down again, then gets back up, and so on. These
actions were seemingly chosen to match the emotions conveyed by
text and music in the two movements – in a few words, ‘Am I per-

8 As of January 2011, the tempo di mezzo and the cabaletta are available on You-
Tube: <>.
54 Emanuele Senici

chance falling in love?’ in the cantabile, ‘No, I can’t fall in love, I can
only go on with my life of endless partying’ in the cabaletta. The di-
rector chose to express these different emotions through different ac-
tions. Such actions, however, would not automatically require specific
shots. The shots chosen by Enriquez were favoured, I would suggest,
in order to generate visual movement where movement is not obvi-
ously contemplated in the opera, that is, in the ‘static’ formal seg-
ments. Treating the two expositions of the cabaletta as two tracking
shots, as Enriquez does, enhances the sensation of incessant move-
ment, which would not seem so incessant with more frequent cuts. At
the same time, filling the screen with Violetta’s face during the can-
tabile guarantees movement through the mobility of Carteri’s face,
since – whatever we may think of her acting gifts – she must move at
least her mouth in order to mime the act of singing. But the close-up is
totally foreign to the visual aesthetic of opera as performed in the
The issue of close-ups in opera videos is often mentioned but sel-
dom discussed in any depth. The close-up is a particular television
need, since it is not only the most effective visual choice for the small
screen (especially the really small screens with low-quality image
definition of 1950s television sets), but also one of the crucial means
through which television promotes its realist aesthetic and achieves
that sense of intimacy that has always been considered a defining
characteristic of the medium. It is therefore in the context of television
that the issue of close-ups in opera videos becomes burning. Here I
would like to use it as another point of view from which to investigate
the relationship between television aesthetic and opera aesthetic in the
RAI studio productions of the 1950s.
The question that the operatic close-up asks in this context is
whether we should think of it as ‘realistic’ or ‘anti-realistic’, ‘natural’
or ‘artificial’. As I have just mentioned, there is no doubt that its effect
is squarely anti-realistic from the point of view of opera as performed
in the theatre; there is also no doubt that opera as performed in the
theatre was the yardstick by which televised opera was evaluated in

9 In the words of British critic Tom Sutcliffe, ‘opera is an art without close-ups’;
Sutcliffe, Believing in Opera (London: Faber, 1996), 205.
Opera on Italian Television 55

the first years of the new medium, despite the popularity of opera
films – Visconti’s letter is clear proof of this tendency. The attitude
toward the close-up in Violetta’s aria, as in the whole of Enriquez’s
Traviata, however, belongs unquestionably to television: the close-up
is the characterizing feature of the new medium, the one that allows it
to show reality from ‘up close’ (as the English language tellingly has
it), therefore making it somehow ‘truer’ and ‘more real’, and as such
this device is put to service with no hesitation, and perhaps without
thinking too hard about its role within traditional operatic aesthetic.
What is more, it seems obvious that the need for close-ups had a
significant impact on the choice of the main interpreters. They might
not have always been the best ones available from a vocal and musical
point of view, but, for one thing, they were all more or less of the
same age of the characters they embodied, which is notoriously sel-
dom the case in the theatre. Carteri, for example, was twenty-four
years old at the time of her Violetta; her resplendent, noble beauty
received the assiduous attentions of Enriquez’s cameras not only in
the 1954 Traviata, but also as Desdemona in Otello four years later
(meanwhile, in 1956 she had been Alice Ford in Herbert Graf’s Fal-
staff). Any worry that she was very young, perhaps too young for such
a complex and difficult role as Violetta, was evidently sidelined; in
the end her interpretation was entirely respectable if not particularly
interesting. Tenor Franco Corelli, who appeared as Canio in Pa-
gliacci, Cavaradossi in Tosca, Don José in Carmen and Calaf in
Turandot, was a great singer and a very good-looking man – as a critic
put it, ‘a mixture between Victor Mature (but better looking) and
Rock Hudson (but less irresistible)’. His obvious limitations as an
actor were clearly not considered a hindrance.
These considerations might suggest that the tension between tele-
vision realism on one side and theatrical artificiality on the other –
speaking from the point of view of television, of course – was system-
atically resolved in favour of the former. Taking into account other
elements, however, rather complicates the picture. Let us look at the
broadcasts of two early-nineteenth-century opere buffe, Rossini’s

10 Pier Maria Paoletti, Quella sera alla Scala (Milan: Rusconi, 1983), 17.
56 Emanuele Senici

L’Italiana in Algeri in Mario Lanfranchi’s interpretation and Doni-

zetti’s Don Pasquale with the direction of Alessandro Brissoni.
While Lanfranchi does not make any concessions to the theatre,
Brissoni does so quite explicitly, at least in the frame of the action.
Lanfranchi uses the long overture to L’Italiana as the soundtrack to a
succession of short sketches introducing ‘Algiers’ and the court of its
bey. During the overture to Don Pasquale, which is almost as long,
Brissoni shows the RAI Milan Orchestra playing in a television studio
under Alberto Erede’s baton – the orchestra is definitely playing, but
what is heard may be a pre-recorded performance, at least judging
from a few moments of imperfect synchronization between image and
sound. On the coda of the overture a dissolve brings on the drawing of
a theatre curtain, on which a moment later the words ‘Atto Primo’
appear; drawing and words dissolve in turn to an ornate iron gate,
which is then opened by two extras on the orchestral introduction to
the opera’s initial number: behind it the interior of Pasquale’s home
comes into view. At the end of the opera the reverse takes place: the
gate is closed, the words ‘Fine dell’opera’ materialize, and the final
shot is the curtain drawing. The attempt to locate the broadcast with a
theatrical space, or, perhaps better, to hint at some crucial aspects of
the theatrical experience, would seem unquestionable.
I use the conditional because, during Don Pasquale’s overture,
the four main characters fade in and then out in turn, each doing
something emblematic for a few seconds: Ernesto sighs while gazing
at a small portrait, presumably Norina’s; Norina looks like she is
thinking something up which amuses her; Malatesta arrives, kisses her
hand and bursts into laughter; and Don Pasquale ties the belt of his
dressing gown and then shakes his pocket watch trying to figure out
whether it works. These apparitions, which decidedly belong to the
television aesthetic, complicate Brissoni’s apparent attempt to recu-
perate the theatrical dimension to the genre of the studio broadcast, a
dimension that in the same years other directors seemed intent on
minimizing instead – especially Lanfranchi, as testified by his
Madama Butterfly and La sonnambula among other titles, but also
Enriquez, whose Rigoletto and Otello are especially noteworthy, the

11 Unfortunately I could not get hold of that initial Barbiere.

Opera on Italian Television 57

latter probably influenced by Orson Welles’s filmed version of Shake-

speare’s tragedy, which had been released in Europe in 1952. In the
context of this genre, and specifically in light of the appearance of the
main characters during the overture, the drawing of a curtain, the gate,
and perhaps even orchestra and conductor wearing tails, give an im-
pression of artificiality, of gauche awkwardness even, while the ele-
ments belonging exclusively to television seem somehow more ‘natu-
ral’, more ‘real’. I should add that Brissoni is more cautious than
Enriquez with close-ups, and that our notions of what is ‘natural’ and
‘real’ on television are based on half a century of constant interaction
with the medium, which of course was not the case in 1955, the year
of Brissoni’s Don Pasquale.
These observations suggest that the RAI studio broadcasts of the
1950s were characterized by a profound tension between a traditional
operatic aesthetic and an incipient television one, as I have already
suggested. This tension emerges most evidently in some aspects of
these broadcasts, such as the prevailing types of shots and of camera
movements, that is to say, those aspects where the distance between
theatrical and television experience was widest. The initial interaction
between opera and television seems to have been less fraught in
spheres for which, all things considered, the dimension of the per-
formance in the theatre does not count as much, such as that of décou-
page. If shots and camera movements were often chosen with an al-
ready developed specific television aesthetic in mind, their rhythm
could respond to, or at least keep an eye on, a musical-formal articula-
tion that is in its turn a way to structure the drama. I should make it
clear that I am not saying that the editing of these broadcasts is some-
how ‘neutral’, since I do not believe neutrality is a meaningful cate-
gory in this matter. All I want to point out is that that this editing often
seems to follow the macro-formal structure of the works being relayed
– presumably intuitively, since musicological research into what is
usually called ‘la solita forma’ was still to come. In other words, the
encounter between opera and television in 1950s Italy highlighted
space as potentially more problematic than time, leading to an inter-
estingly wide array of solutions. What did the following decades bring
about in this respect? Did anything new emerge?
58 Emanuele Senici

7 December 1976

The live broadcast of the opening night of the 1976-1977 season at La

Scala, which took place on 7 December 1976 with Verdi’s Otello, was
unanimously hailed at the time as a turning point in the relationship
between opera and Italian television, and has continued to enjoy this
reputation ever since. The contemporary discourse of this event em-
phasized explicitly political, even revolutionary terms: eventually the
privilege of few would give way to the participation of many; eventu-
ally the most important date on the social calendar of the Italian high-
est classes would become a celebration for the whole of society. His-
torians of Italian television such as Giorgio Simonelli, Giovanni
Buttafava and Aldo Grasso, Luisella Bolla and Flaminia Cardini have
commented on the excitement of the ‘media event’ built around this
broadcast, which was saluted as the beginning of a new era for opera
on television. Despite the increasingly difficult climate for live relays
from theatres over the following years, in the context of the begin-
nings of commercial television in Italy (a private station called Tel-
emilano, owned by a certain Silvio Berlusconi, started broadcasting
precisely in 1976), there is no doubt that, for about a decade, live
transmissions from La Scala, especially of season opening nights,
became key moments of the television season, and an unmissable
rendezvous for Italian opera lovers – very much including the present
writer’s younger self. Observing this Otello from the point of view of
opera, two issues emerge with particular force, one historiographical,
the other aesthetic.
Reading both what was written back then on the 1976 Otello live
broadcast, and what has been published since, one would be justified
in believing that this was the first opera ever broadcast on Italian televi-
sion live from a theatre. The fact that many such broadcasts preceded
this Otello is sometimes mentioned in passing, more often completely
ignored. This is especially surprising given that we are not talking about
second-rate performances, but, for example, of the Visconti-Callas-
Bernstein Sonnambula from La Scala in 1955 (only the first act), or the

12 See the references in footnote 1 above.

Opera on Italian Television 59

Tebaldi-Dominguez-Corelli-Bastianini-Christoff-Capecchi Forza del

destino from Naples’ Teatro San Carlo in 1958, or the Gencer-
Simionato-Limarilli-Guelfi-Serafin Aida from Verona’s Arena in
1963, or another Aida also from the Arena three years later, this time
with Gencer-Cossotto-Bergonzi-Colzani-Giaiotti, which, according to
Bolla and Cardini, was watched by a record-high number of viewers.
I can well believe it: these are nights that, at least on paper, are decid-
edly more important in the history of operatic performance than the
majority of the studio broadcasts of the 1950s. Why have they been all
but forgotten, then?
In the first place, some of these broadcasts are lost forever, since
RAI did not think it was worth te trouble to capture them on film
through a complex machine called ‘kinescope’, the only means of
preserving a television transmission until the early 1960s, when RAI
eventually adopted videotape. The logic behind this choice was pre-
sumably that these were not RAI productions, but ‘only’ relays of
performances put on by others, and therefore they were not worth
keeping – so we must sadly kiss goodbye to a video of Callas at the
zenith of her career in Visconti’s historic Sonnambula. Other broad-
casts perhaps still exist in the RAI archives, but they have never been
either re-broadcast or made available to the home-video market, and
therefore have never been textualized, falling below the radar of criti-
cal and scholarly attention. Others still, however, have circulated in
various formats, among them the Neapolitan Forza and the second
Veronese Aida, and constitute potentially rich sources for the study of
operatic visuality (as well as aurality, of course) in the central decades
of the twentieth century, sources that have not yet received the atten-
tion they deserve.
This collective amnesia may also be due to the fact that the 1976
Otello was the first opera broadcast in colour, and therefore it could be
promoted as a more faithful representation of the theatrical perform-
ance than the preceding black-and-white ones. In the light of this hy-
pothesis, the second issue highlighted by this Otello emerges with
special force. This broadcast approaches what happens onstage so
deferentially that, paradoxically, it denies the viewer meaningful ac-

13 See Bolla and Cardini, Macchina sonora, 244.

60 Emanuele Senici

cess to it, not least in the context of preceding live broadcasts. Simo-
nelli is right in arguing that, on that 7 December 1976, ‘the television
medium, concerned exclusively with its informative and documentary
role (interviews, declarations, and so on), almost devoured by a kind
of “I-was-there” fever, ended up neglecting precisely the central as-
pect, that is, the opera and its mise en scène, as incapable of reproduc-
ing faithfully the central elements of theatrical communication (even
more so than for spoken theatre), as heedless of elaborating original
forms of translation.’ Since the informative and documentary role is
now lost to us (interviews and declarations are not included in the
video that was made commercially available later), and leaving aside
the pious fantasy of a ‘faithful’ reproduction of theatrical communica-
tion, we are left with a spatially amorphous and rhythmically flaccid
object, stuffed full with total shots of the stage that are hard to read on
the small screen, and enfeebled by weak editing, which cannot decide
whether to follow the visual dimension or the textual-musical one,
leaving the viewer in an uncomfortable state of rhythmic uncertainty.
It would be unfair not to mention the technical challenges that
this broadcast had to face, especially regarding lighting, which was
generally too dim for the video cameras of the time. A propos of light-
ing, it may be worth mentioning that Giorgio Strehler’s refusal to alter
the lighting of his production of Don Giovanni was given as the offi-
cial reason for RAI’s decision not to relay live the opening night of
the 1987-88 La Scala season – to date, the last such relay is the pre-
ceding year’s Nabucco. In any case, and to conclude on this Otello, it
seems clear that the tension between theatrical and television aesthetic
that had characterized the live studio productions of the 1950s re-

14 I have discussed the relationship between live opera performances and their
videos at length in Senici, ‘Il video d’opera “dal vivo”‘, and ‘Porn Style? Space
and Time in Live Opera Videos’, Opera Quarterly 26 (2010), 63–80.
15 Simonelli, ‘Evoluzione storica del teleteatro’, 97.
16 The video director was Franco Zeffirelli, who also directed the theatrical pro-
duction. It is not surprising that a scenographer by training such as Zeffirelli –
and many have insinuated that he remained a scenographer at heart even when
he was entrusted with the overall responsibility for a production – would insist
on so many full shots of the stage as the only ones that showed the complete pic-
ture, so to speak.
Opera on Italian Television 61

turned on 7 December 1976, but in rather different terms. By the

1970s television had fully elaborated its own original language, which
was profoundly imbricated in Italian society, culture and visuality;
unfortunately for opera, this language had very little in common with
the aesthetic orientations and the practical demands of theatrical per-
formances, especially operatic ones. A provisional bridge between
these two opposite sides was built around what I would call the ideol-
ogy of liveness, that ‘“I-was-there” fever’, as Simonelli calls it, which
for a moment sidelined the fact that for half of the opera viewers saw
a brown mush – if they were lucky and had a colour television set,
otherwise the mush was greyish. And it was this same ideology of
liveness that would characterize the next case study, albeit rather dif-
ferently, as we shall see.

25 August 1984

On 25 August 1984, the fourth performance of Rossini’s Il viaggio a

Reims in Luca Ronconi’s mise en scène took place in Pesaro as part of
that year’s Rossini Opera Festival. This production fully deserves
pride of place in a history of the relationship between opera and tele-
vision (even without the restriction to Italy), since Ronconi staged the
opera as a sort of television ‘media event’, probably a first in the his-
tory of operatic production. A few video cameras placed prominently
on the stage filmed the action taking place there, while others scat-
tered outside the theatre, in the streets and the piazzas of Pesaro, cap-
tured a performance of the coronation procession of Charles X as
King of France (the event for which the opera was commissioned and
which it celebrated in 1825). The images caught by both sets of cam-
eras were alternatively relayed live on a screen placed upstage, until,
in the final scene of the opera, the king and his retinue arrived in the
auditorium, ‘reality’ eventually merging with its live representation on
62 Emanuele Senici

Ronconi has declared that ‘the use of technology in Il viaggio a

Reims was self-consciously arrogant. It was a memory, a kind of quo-
tation of a real event already seen on television: the coronation cere-
mony of Elizabeth II of England’. Ronconi’s reference to this cere-
mony is puzzling, since the coronation of Elizabeth II, which took
place in 1953, is the foundational moment of television in Britain, not
Italy, where in 1953 regular transmissions did not yet exist; and, any-
how, in 1984 the average operagoer had no memory or awareness of
it, at least in Italy. At that time the immediate television reference
would have been not a coronation, but another royal event, a wedding,
that between Charles and Diana of course, broadcast only three years
earlier to an estimated 750 million viewers. The royal weeding of
1981 was without doubt the television event of the 1980s with the
widest resonance, both in Italy and worldwide.
According to Ronconi himself, the first impulse behind his televi-
sion spectacularization of theatrical reality was due to the fact that this
very reality was going to be audio- and videorecorded: ‘Since televi-
sion relay and audiorecording were part of the commission, I thought I
would locate the performance in the concert hall [Pesaro’s Auditorium
Pedrotti], which is the most appropriate space for recording, so that
the recording itself would become part of the show. Hence the micro-
phones in full view, hence the cameramen and sound technicians in-
tervening into the action, so that it was impossible to distinguish be-
tween the real technicians, those of Fonit Cetra or Deutsche
Grammophon, from the fake ones, the lifeguards of Pesaro’s beaches
hired as extras, wearing white and pink coats’.
Ronconi’s production, then, reflected on its own televisualization
and on its own position within the media landscape, focusing specifi-
cally on the ideology of liveness, on television’s cannibalistic attitude
toward reality, thanks mainly to the real-time projection on the on-
stage screen of what was being captured outside the theatre. When

17 ‘Luca Ronconi’ [interview], in Tabanelli, Il teatro in televisione, vol. 1, 263–

310: 305.
18 ‘Luca Ronconi: tre opere d’occasione’, in Luca Ronconi, Inventare l’opera.
L’Orfeo, Il viaggio a Reims, Aida: tre opere d’occasione alla Scala (Milan: U-
bulibri, 1986), 13–29: 15.
Opera on Italian Television 63

Charles X and his retinue entered the auditorium, it was not so much
the reality of the stage and that of the street eventually coming to-
gether; it was rather television appropriating the whole of reality and
turning it into ‘reality’, in a representational and mediatic short-circuit
which, in light of television’s development in the following decades, it
is tempting to call prophetic.
I have entitled this section ‘25 August 1984’ because it was on
this day that RAI videorecorded the fourth and final performance of
this Viaggio (for later broadcasting). According to Buttafava and
Grasso, ‘the relay of Pesaro’s Il viaggio a Reims was an unusual in-
stance of television at the second degree’. After watching the video
of that broadcast, I would rather say that this might have been televi-
sion at the second degree, but in the end it was so only insomuch as
the mere act of filming Ronconi’s production inescapably meant put-
ting television in front of television. This is because the video direc-
tion behaved as if television had nothing to do with the mise en scène.
The interaction between characters on one side and onstage cameras
and monitors on the other – for example, Trombonok-Enzo Dara invit-
ing a cameraman to zoom in on the Countess-Lella Cuberli going mad
over her hat – is generally shot from afar, as if these cameras and
monitors were inert objects rather than machines recording and relay-
ing what is happening on stage at that precise moment. The vertigi-
nous mise en abîme of the Countess’s hat between stage and screen
during her cabaletta, for example, which provides a particularly
strong visual equivalent for her crazy coloratura, ends up being at best
domesticated, and at worst neutralized, by alternating total shots of
stage with medium shots of Cuberli, almost totally ignoring the on-
stage screen.
Ronconi would lambast the video direction of the revival of his
Viaggio at the Vienna State Opera four years later, which resulted,
according to him, in a product ‘conceived only with the singers’ fans
in mind’ rather than intended as ‘truthful documentation of the
show’. I wonder what Ronconi had in mind when he spoke of ‘truth-
ful documentation’. What I can say is that the broadcast of the Pesaro

19 Buttafava and Grasso, La camera lirica, 46.

20 ‘Luca Ronconi’, 301.
64 Emanuele Senici

Viaggio seems completely oblivious to the fact that the production

takes television as its subject. If Ronconi’s mise en scène is riotously
postmodern, intent on unmasking theatrical and television tricks, its
broadcast is rather dolefully modernist, indifferent toward – and per-
haps even a bit annoyed by – the production’s play with media.
Should we conclude, then, that, when it comes to opera, television can
become the subject of a theatrical production, but not of a television
This is certainly also the case of another broadcast that, for en-
tirely different but no less compelling reasons, could have been taken
as an invitation to do television about television, and that I could have
selected as another very revealing case study, entitled ‘11-12 July
1992’. I am referring to the famous Tosca in the Settings and at the
Times of Tosca, the grandest operatic instantiation of the ideology of
liveness and of television ‘realism’. Despite other attempts to pro-
mote this aesthetic, such as Traviata à Paris (3-4 June 2000) and Ri-
goletto in Mantua (4-5 September 2010), the past two decades have
been characterized instead by the broadcasting of videos of theatrical
productions, in most cases assembled by editing together materials
taken from a few recordings of different performances.
Since the 1980s, however, these ‘live’ videos are conceived pri-
marily not for television broadcast, but for the home video market.
This is but one symptom of the ongoing fragmentation of the video-
media landscape, which started out with videocassettes in the 1980s
and has sped up exponentially after the so-called digital revolution.
Specifically concerning opera, the last decade has seen a previously
unthinkable explosion of DVDs of ‘live’ performances. Conversely,
television has progressively lost the dominant position in the realm of
video media which it enjoyed between the mid-1950s and the mid-
1980s, when even most of the (not many) opera films made at the time
were never distributed to cinemas, but went straight to television. If
the arrival of DVD had perhaps the strongest impact on this progres-
sive corrosion of television’s power, more recently the possibility of

21 On this Tosca, see Jonathan White, ‘Opera, Politics and Television: Bel Canto
by Satellite’, in A Night in at the Opera, ed. Jeremy Tambling (London: John
Libbey – The Arts Council of England, 1994), 267–294.
Opera on Italian Television 65

downloading videos of entire operas from the internet (more or less

legally), as well as the appearance of such internet sites as YouTube,
have further reconfigured the relationship between opera and televi-
sion, often relegating the latter to a mere machine: the television set
has become just one of several possible screens on which to watch a
digital file. At the same time, satellite and then digital television have
altered the landscape of television networks, encouraging a differen-
tiation of channels between generalist and thematic ones: a complex
artefact and niche product such as opera has naturally found in the-
matic channels a comfortable habitat. These are very recent and there-
fore very complex scenarios, and doing justice to them would need at
least twice the space available here. I will leave them aside, therefore,
and will try instead to draw some concluding remarks from the three
cases discussed above.

From production to ‘documentation’

I would like to begin my conclusions with a historiographical reflec-

tion. Until about a decade ago ‘opera on television’ was generally
taken to mean ‘opera on video’ tout court, and the reverse was also
true. I believe that this was the case because, as I have suggested,
when it came to opera television completely dominated the realm of
video media between the 1950s and the 1980s. In light of the radical
revolution in this field over the past two decades, this overlap no
longer seems justified: now it is important to distinguish between dif-
ferent types of videos, different media through which they are dis-
seminated, and different screens on which they are seen, which of
course entail different modes of fruition. Why, then, have I not done

22 For a standard example, see Marcia Citron, Opera on Screen (New Haven: Yale
University Press, 2000).
23 For a discussion of these levels of differentiation, see Senici, ‘Il video d’opera
“dal vivo”‘. For other recent contributions that are particularly alert to these is-
sues from the point of view of television, see Mauro Calcagno, ‘Performing the
66 Emanuele Senici

so in the main body of the present text, mixing instead observations

strictly related to television with others concerned with video in gen-
eral? My only excuse is that I have conducted a kind of analysis with
period instruments, or, better, with period eyes, the eyes of those who
watched opera on Italian television between the 1950s and the 1980s,
for the vast majority of whom this was the only way to watch opera on
screen – and in many cases opera tout court. This is perhaps the main
reason why I have chosen to focus on three salient moments located
between 1954 and 1984: then, to quote the finale of William Tell,
‘everything changes’ – even if the sky does not necessarily ‘become
This much clarified, the first basic conclusion to be drawn seems
to be that the history of opera on Italian television during these thirty
years draws a trajectory from production to reproduction. If in the
beginning Italian television presented itself as a medium that could
develop an original form of opera, over the following decades relays
of theatrical performances became the dominant type of broadcast,
with ‘documentation’ as their avowed goal – a goal that had a strong
and in my view rather negative impact on their style. The reasons for
this trajectory from production to ‘documentation’ are to be found in
the history of television rather than in anything to do with opera, and
therefore it is the job of television historians to study them.
I would like to suggest, however, that opera’s radical alterity vis-
à-vis television and the wide gulf between operatic and television
aesthetics should be included among them. In the case of broadcasts
from theatres, the consequences of this alterity could be expediently
imputed to the constraints of theatrical space – hence ‘excusing’ tele-
vision, which could hide behind the presumed ‘need’ to ‘document the
event’ in its original space. Studio productions required instead an
assumption of aesthetic responsibility, or at least a reflection on the
peculiarities of opera, including its dramatic structures, its unique
handling of time, and the acting skills and physical appearance of its

Self’, Opera Quarterly 24 (2008), 247–274, especially 264–270; Melina Esse,

‘Don’t Look Now: Opera, Liveness, and the Televisual’, and Christopher Mor-
ris, ‘Digital Diva: Opera on Video’, both in Opera Quarterly 26 (2010), 81–95
and 96–119 respectively.
Opera on Italian Television 67

interpreters, as we have seen. Visconti thought that these production

were ‘ugly, unpleasant and anti-artistic’; whatever we may think of
them now, they constituted the first meeting between opera and televi-
sion in Italy, and therefore worked as an experimental space for the
very peculiar kind of encounter between moving image and sound
engendered by opera on television.
There is no doubt that this encounter had a very powerful impact
on opera, bringing about new modes of production and reception of
the genre. This impact was felt most strongly precisely by those as-
pects of opera most distant from the television aesthetic, such as its
handling of dramatic time and the characteristics of its interpreters.
What is more, it should be remembered that, between the 1950s and
the 1980s, television was the only means through which the majority
of Italians had access to the visual dimension of opera. Television
created an imaginary repertory of productions unthinkable before its
arrival, a repertory that cannot but have had an impact on the expecta-
tions of the public of opera, whether this public saw opera on televi-
sion or in the theatre. It would perhaps be more accurate to say espe-
cially in the theatre, however, since after the 1950s the majority of
broadcasts has been ‘live’, or in any case from theatrical spaces. As I
have mentioned above, these broadcasts have been presented and dis-
cussed in terms of ‘documentation’ of theatrical performances. It is all
the more logical, then, that the expectations of those who went to live
opera performances were influenced, whether consciously or not, by
those broadcasts. Moreover, although I have used the past tense thus
far, I would suggest that, despite the profound changes in the field of
video media over the last two decades, much of what I have said about
the period between the 1950s and the 1980s is still true.
Finally, I would argue that, paradoxically, the radical alterity of
opera vis-à-vis television has made television’s impact on opera all
the stronger: higher the tension, greater the consequences of its resolu-
tion in favour of one of the poles. At the same time, however, pre-
cisely the high level of this tension has turned opera into a kind of
moment of truth for television, a moment when its claims to documen-
tation have been unmasked, or at least put to the hardest test – more so
than in the case of spoken theatre or non-theatrical music, I would say.
If we can say with no hesitation, then, that television has made us
68 Emanuele Senici

watch opera differently, both on television and in the theatre, I believe

we can also say that, at the same time, opera has invited us to watch
television differently, or at least with a particularly sceptical eye to-
ward some of its aesthetic, cultural and ideological claims.
Opera on Italian Television 69


Studio productions broadcast by RAI, 1954-1970

All productions were broadcast from the Milan studios unless otherwise indicated:
* = broadcast from the Turin studios
** = broadcast from the Rome studios

Date Title Composer Conductor Director

23-04-1954 Il barbiere di Siviglia Rossini Giulini Enriquez
26-09-1954 I pagliacci Leoncavallo Simonetto Enriquez
23-10-1954 L’elisir d’amore Donizetti Rossi Brissoni
27-11-1954 La bohème Puccini Sanzogno Enriquez
22-12-1954 La traviata Verdi Sanzogno Enriquez
29-01-1955 Gianni Schicchi Puccini Votto Majano
26-02-1955 Rigoletto Verdi Sanzogno Enriquez
26-03-1954 Adriana Lecouvreur Cilea Simonetto Enriquez
21-05-1955 Don Pasquale Donizetti Erede Brissoni
09-06-1955 Ciottolino* Ferrari- Ferrari- Brignole
Trecate Trecate
18-06-1955 Manon Massenet Simonetto Blasi
16-07-1955 La vedova scaltra Wolf-Ferrari Sanzogno Pavolini
24-09-1955 Tosca Puccini Votto Blasi
15-10-1955 Andrea Chénier Giordano Questa Landi
22-11-1955 La Cenerentola Rossini Gavazzeni Colosimo
24-12-1955 Amahl e gli ospiti notturni Menotti Scaglia Bolchi
24-01-1956 Madama Butterfly Puccini De Fabritiis Lanfranchi
14-02-1956 Il matrimonio segreto Cimarosa Caracciolo Enriquez
13-03-1956 La fanciulla del West Puccini Simonetto Lanfranchi
11-04-1956 Fedora Giordano Bartoletti Lanfranchi
09-05-1956 Falstaff Verdi Serafin Graf
13-06-1956 Carmen Bizet Sanzogno Enriquez
11-07-1956 Cavalleria rusticana Mascagni Basile Lanfranchi
12-09-1956 Le nozze di Figaro Mozart Sanzogno Graf
17-10-1956 Manon Lescaut Puccini Questa Wallmann
21-11-1956 Un ballo in maschera Verdi Sanzogno Enriquez
19-12-1956 La sonnambula Bellini Bartoletti Lanfranchi
23-01-1957 Il tabarro Puccini De Fabritiis Brissoni
20-02-1957 Mavra Stravinskij Gracis Pavlova

24 Data taken from Giorgio Gualerzi and Carlo Marinelli Roscioni, 50 anni di
opera lirica alla Rai, 1931–1980 (Turin: ERI, 1981), 128–198.
70 Emanuele Senici

24-04-1957 La medium Menotti Rescigno Wallmann

29-04-1957 Il trovatore Verdi Previtali Fino
10-07-1957 L’italiana in Algeri Rossini Sanzogno Lanfranchi
18-09-1957 I due timidi Rota Gracis Molinari
25-12-1957 Hänsel e Gretel Humperdinck Sanzogno Cottafavi
25-02-1958 Il turco in Italia Rossini Sanzogno Enriquez
30-09-1958 Otello Verdi Serafin Enriquez
23-12-1958 Turandot Puccini Previtali Lanfranchi
19-05-1959 Le campane Rossellini Pedrotti Wallmann
02-02-1960 Francesca da Rimini* Zandonai Basile Lanfranchi
15-03-1960 Tosca* Puccini Vernizzi Lanfranchi
26-04-1960 Don Giovanni Mozart Molinari- Vaccari
11-12-1960 La figlia del reggimento Donizetti Mannino Lanfranchi
12-02-1961 Battono alla porta** Malipiero Sanzogno Bolchi
31-12-1962 Il cappello di paglia di Rota Sanzogno Lanfranchi
24-12-1963 Canto di Natale* Liviabella Boncompagni Macchi
11-02-1964 Il linguaggio dei fiori Rossellini Bellugi Wallmann
12-01-1965 Faust a Manhattan** Nascimbene Ferrara Bolchi
01-06-1968 La fantarca* Vlad Sanzogno Cottafavi
“Temps spatialisé”:
Opera Relays and the Sense of Temporality1


Temporality implies the sense of running time. The time of the diege-
sis, lived or not in succession, raises narratological questions as dif-
ferent conventions are used across diverse genres and in the media
about how to tell a story. Are the ways in which temporality is em-
ployed in the genres of opera and film the same? And how does the
filming of an opera change our perception of temporality?

I. Opera, cinema and temporality

The spectator in an opera house always sees a story moving forward.

Therefore it is possible to miss some events (for example, what Tosca
accomplishes after Scarpia’s murder and before rejoining Mario in
Sant’Angelo Castle), nonetheless the spectator always watches a story
progressing on the timeline (even in operas with a structure based on
disjunct tableaux as in Debussy’s Pélleas et Mélisande). Even Wag-
ner, when he imagines the antecedents to Siegfried, decides to show
them to the spectator (in only two new libretti, Das Rheingold and Die
Walküre). However this does not mean that opera is unable to play
with the past.

1 The discussions during the International Workshop on Opera and Video in

Valencia (22–23 March 2010) organised by Héctor Julio Pérez-López contribute
to this article and I would thank the participants for it. Moreover, I am grateful
to Luca Zoppelli for his inspiring comments. Eventually, I thank Sarah Lambert
for her “British reading” of this essay.
72 Delphine Vincent

Opera is a musical spectacle and the music itself may indicate re-
lationships between events, and therefore temporal relations. This is
evident when composers employ recurrent motives; their relations and
their eventual transformation create links and temporal indications (as
the gold Leitmotif which is changed after Alberich steals it). The tem-
poral phenomenon is particularly strong when composers imagine a
musical motive for a past moment not shown on stage, as with Philipp
II’s monologue in Verdi’s Don Carlos (act IV): The king remembers
the arrival of his betrothed (“Je la revois encore”) on a march motive,
depicting a past event. We, the audience, are not able to see it and can
only imagine with him when he is experiencing this sort of musical
flashback. Otherwise some twentieth century composers mistreat
narrative codes. The most evident example of a new way of “tempo-
ral” thinking is found in Zimmermann’s Die Soldaten, which presents
some superimposed stages and epochs. It remains, however, an excep-
tion, linked to Zimmermann’s understanding of life and reading of
Lenz’s play.
Cinema, however, tells stories in a rather different way. Nowa-
days (and for a long time) it has been impossible to see a film being
presented only by a linear narration; it always includes some
flashback or, even if more rarely, a flashforward. The spectator may
even, on occasions, see the same story in the same film presenting
different points of view (cf. Zhang Yimou’s Hero, 2002).
Obviously opera relays cannot superimpose such cinematic
strategies to linear operatic stories. Moreover, the staging is a funda-
mental component of operatic performances. In an opera house, the
spectator watches a musical story; every part of the staging means
something to him. Opera relays must select some part of this staging,
but opera relays, like cinema, are based on shot breakdown and mise
en scène on variations of movements (as in theatre plays). The result
is sometimes artificial, because on the one hand opera relays do not
respect the theatrical vision, and on the other they are not able to use
all the cinematic strategies: therefore ending up being a sort of hybrid.

2 See Luca Zoppelli, L’opera come racconto: modi narrativi nel teatro musicale
dell’Ottocento (Venezia: Marsilio, 1994), 120–122.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 73

II. Cinematographic processes in staging an opera

This is even more complicated because some stage directors use

cinematic processes for their staging. In some mise en scène we find
cinematic processes such as close ups, focusing or extensive use of the
backstage as in a in/off screen. This tendency is found in a few con-
temporary spectacles, especially in those of Robert Carsen. For exam-
ple, he manages a “close up” of Scarpia’s entrance in his staging of
Tosca (Zürich, Opernhaus, 2009). In Sant’Andrea Church (quite sim-
ply staged: an inclined plane with an imposing column on the left, red
chairs and a red curtain at the far end of the stage), the Sacristan and
the children are making too much noise, Scarpia enters and repri-
mands them (“Un tal baccano in chiesa! / Bel rispetto!”). Scarpia’s
arrival is underlined by his musical motive, moreover an impressive
one, leaving no doubt that he is the most powerful man in the story
(singing con grande autorità). The stage direction indicates that the
Sacristan and the children “saltellano e ridono sgangheratamente
quando Scarpia appare improvvisamente dalla porticina. Alla vista di
Scarpia tutti si arrestano allibiti come per incanto”. In Carsen’s mise
en scène Scarpia appears on the pedestal of the column (at a height of
at least 2.5 metres). It therefore comes as quite a surprise, because it
was highly improbable that somebody would be there. However it
becomes impossible to not be aware of his presence, as from the very
moment he appears he is the only one illuminated; all the stage is in
darkness, therefore the spectator has to look at him. I understand this
staging as a transposition of the close up’s technic on stage.

3 For example, we find an attempt of reconstructing a travelling shot during the

love duet in Olivier Py’s staging of Tristan und Isolde, Geneva, 2005.
74 Delphine Vincent

III. Filming cinematographic processes in opera staging

Unfortunately there is not a recording of Carsen’s Tosca in existence

so I am not able to compare the effect on the spectator in the theatre
and in front of an opera relay. But the same process is found in his
staging of Tchaikovsky’s Eugene Onegin. I have not had the good
fortune to see it at a theatre, but I can imagine that the sensation would
be the same. But how is it possible for the video director to film this
close up without losing the effect accomplished for the theatre specta-
During the ball in the third act, Onegin meets Tatyana, who is
now a princess. Her husband, Prince Gremin, introduces them and
immediately Tatyana is flustered, but tries to conceal it. The couple
leaves and the stage gets darker. Onegin is standing in front of the
armchairs that demarcate the dance floor. The spotlight illuminates
him when he realises that he is in love (just after the first occurrence
of “lyubov”, “love”) in his arioso “Uvï, somnen’ya net” (“Alas, there
is no doubt”). This light generates a projected shadow on the wall at
the far end of the stage. Behind him the guests are still there, but
Eugene is isolated (by the chairs, but above all by the lighting). The
audience is forced to focus its attention on him, as in a cinemato-
graphic close up. When the dance starts again (a Schottische), the
stage is instantly full with light, as if we were going back to a long
shot. But the film is unable to render this impression, because by only
using an extreme long shot it becomes boring to watch and a close up
is impossible because of the loss of the general view which is essential
in understanding Carsen’s idea. The solution is a compromise: the
arioso begins with an extreme long shot, then we find a plan améri-
cain becoming a medium shot (with a zoom) and for the Schottische
we are back to an extreme long shot.

4 At the time of the writing of this article, the DVD did not exist.
5 Pyotr Ilyich Tchaikovsky, Eugene Onegin, Dmitri Hvorostovsky (Onegin),
Valery Gergiev (conductor), Robert Carsen (stage director), Brian Large (video
director), New York Metropolitan Opera, 2007.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 75

This example emphasises passages of technics between the two

media (as close ups) but mainly the difficulty that a video director has
when filming a cinematographic process put in a staging. This there-
fore shows us that the video director cannot use his complete filmic
arsenal, but has to bear witness to a live performance.

IV. Creating temporal effects with the film

After insisting on the differences between the two media, I will try to
determine how opera relays may change perception of temporality.
The easiest response is to look at videos integrating strictly cinemato-
graphic effects (such as a flashback or flashforward). It is quite a
marked intervention of the video director, as it radically changes tem-
The most common effect, used in a few – but increasing – num-
ber of recordings, is the flashback. It changes temporality in different
ways, according to where it is positioned in the music. For example,
we find a flashback at the end of Tristan und Isolde in Oliver Py’s
staging (Geneva, 2005) in the TV broadcast. Since the music of
Isolde’s Liebestod rearranges blocks of music from the love duet (act
II), the video director chooses to highlight this connection between the
moments made by Wagner. In this case the flashback does not change
the temporal and narrative relationship between the two moments, but
explains it through the images. On the one hand it is used in a peda-
gogical way to underline the musical rearrangement, and on the other
hand it also seems to be a way of allowing an internal focalisation:
Isolde is lying on Tristan’s corpse, dying of her love for him, trying to
transfigure it and to join him.

6 It is quite an interesting case: the flashback occurs in the TV version (seen on

Mezzo) but is suppressed in the DVD release. Richard Wagner, Tristan und
Isolde, Armin Jordan (conductor), Olivier Py (stage director), Andy Sommer
(video director), Geneva, 2005.
76 Delphine Vincent

In other cases, the flashback seems logical (even if it is implied

only by the lyrics and not by the music ), because it doubles a charac-
ter’s narration. The scene where Donna Anna tells Don Ottavio about
the time she was attacked (“Don Ottavio, son morta!” in Mozart’s
Don Giovanni) is a perfect example. During it a super-imposition
starts, which supplants her present image and becomes a flashback.
The TV spectator is able to see Don Giovanni’s attempt at rape and
his Commandatore’s murder again. The past images materialise sud-
denly because of the violence of the recollection (starting with “un
uom che al primo istante” and ending with Ottavio’s “ohimè res-
piro!”). But it creates a problem regarding narration, Anna was at-
tacked by Giovanni and her father came to rescue her, responding to
her scream. At this point, she leaves the room and does not see her
father dying (discovering the corpse only when she comes back with
Ottavio). The video director puts the flashback there in order to indi-
cate to the TV spectator that the rape attempt leads to the Commanda-
tore’s murder. It is not Anna’s memories but an interpretation and also
a way to animate the long recitativo accompagnato (not easy listening
for a profane). It also gives more weight to the account, because of the
violence of the images.
The flashback is perhaps a radical example, which could lead to a
redefinition of temporality. But it is not this type of strictly cinematic
effect that dominates opera relays.

V. Emotion and temporality

Narrative codes and their temporal implications are also meant to gen-
erate emotion. The discourse on emotions is a complex one and I do

7 Film often uses flashbacks as a way to enter a character’s recollections.

8 Wolfgang Amadeus Mozart, Don Giovanni, Alexandra Deshorties (Donna
Anna), Mark Padmore (Don Ottavio), Peter Mattei (Don Giovanni), Gudjon Os-
karsson (Commandatore), Daniel Harding (conductor), Peter Brook (stage direc-
tor), Vincent Bataillon (video director), Aix-en-Provence, 2002.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 77

not have the pretension to cover it. But it is not difficult to remark that
mental structures lead to diverse musical responses (cf. musical form).
For example the Baroque Age was fond of da capo arias, which often
seem strange nowadays for an uneducated audience. It appears to be a
redundancy: why do we have to hear a part of the thought twice (the A
part of the da capo form)? It is possible to find some responses in the
vision of emotion prevalent at this time, but that is not my aim. What I
want to underline is the fact that musical forms imply a temporal vi-
sion of emotion. The Baroque tendency to freeze the affects in an aria
and to make the story go forward during the recitativi has nothing in
common with the narration dominant in films. It leads to some issues
as Brian Large’s comment reveals:

Furthermore, the formalized musical patterns can affect the structure of the vis-
ual pacing; in the 1989 videotaped version of Vivaldi’s Orlando furioso from
the San Francisco Opera, the da capo arias often involved three shots: one for
the initial A section, a second for the B section and a return to the initial shot for
the reprise of the A section.

In reality it does not correspond to the manner in which he films the

arias in this production, but this remark sanctions the fact that Large
was limited in his filming choices and that he needed to furnish an
explanation for it. I will try to find an answer to this problem with a
study on the effect of emotional temporality in the filming. For this
purpose, I will use Italian Romantic repertoire, which uses a conven-
tional form (solita forma) for arias and ensembles. A duet conceived
in such a form presents the following structure: a scena, a tempo
d’attacco, an adagio, a tempo di mezzo and a cabaletta. It is composed
of an alternation between kinetic movements (scena, tempo d’attacco,
tempo di mezzo) and static movements (adagio, cabaletta). The scena

9 In Brian Large, “Filming, videotaping.” In The New Grove Dictionary of Opera,

edited by Stanley Sadie. Grove Music Online. Oxford Music Online, http://www. (accessed Oc-
tober 11, 2010).
10 Antonio Vivaldi, Orlando furioso, Randall Behr (conductor), Pier Luigi Pizzi
(stage director), Brian Large (video director), San Francisco Opera, 1989.
11 See Harold S. Powers, “La solita forma” and “The Uses of Convention”, Acta
Musicologica, 59/1. (1987): 65–90.
78 Delphine Vincent

has a character of preparation and presents a free versification (mixing

settenario and endecasillabo as in the eighteenth century recitative).
The tempo d’attacco points out the beginning of the closed number,
but still has the function of moving the action forward (a huge differ-
ence with the strict eighteenth-century delimitation of recitatives and
static closed number). The differences in the tempo d’attacco are the
regular versification and the well-ordered dialogue (with parallel stan-
zas and closely-argued exchanges). After the characters’ confronta-
tion, an adagio follows, presenting a character’s emotional or concep-
tual reaction (singing a parte, which allows the superposition of
voices). This movement is a static one, with a lyric feature. The versi-
fication is regular (generally the metre changes between the move-
ments). During the tempo di mezzo the dialogue starts again in a regu-
lar versification. It leads to the cabaletta (with a reprise), a static
movement with an energetic character and often sung a parte. Here
again the versification is regular.
Within the framework of this essay, what interests me is the tem-
poral structure involved in this musical form. It is an alternation of
moments where the action moves forward (actions pieces) and mo-
ments where the characters are reflecting about a situation (set pieces).
It is completely unknown in a film’s narrative, so how are the video
directors able to deal with it?
In this regard, I will analyse the duet in Simon Boccanegra (act
I) when the Doge recognises Amelia as his lost daughter, which is a
standard solita forma. By comparing the editing of set pieces with
action pieces, I have observed a rapid shot’s alternation in the latter
(especially in the tempo di mezzo) and statism in the former. The ten-
dency to follow dialogues with a rapid editing is not a huge surprise.
Video directors often declare their aim to have a musical editing, but it
is no more than a logocentric one, as in filmed theatrical plays. I must
emphasise that the majority of shots are still and are edited by direct
cuts. The rapid editing of the action pieces gives us an impression of

12 Giuseppe Verdi, Simon Boccanegra, Cristina Gallardo-Domâs (Maria Boccane-

gra), Thomas Hampson (Simon Boccanegra), Chor und Orchester der Wiener
Staatsoper, Daniele Gatti (conductor), Peter Stein (stage director), Anton Reitzen-
stein (video director), Wiener Staatsoper, 2002.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 79

angle/reverse angle editing (mostly because the actor on screen is

generally the one singing and the cuts are corresponding to a phrase
beginning as in narrative cinema). The closely-argued exchanges in
the music allow this editing. In both action and set pieces we find
reaction shots (notably for an illustration of the lyrics). Nevertheless it
seems that set pieces do not allow for a deployment of cinematic
ways, on the contrary it seems easier to deal with action pieces.
The remarks concerning the editing should lead to an extreme dif-
ferentiation between the temporality forms in the image. Looking
superficially at the cabaletta (“Figlia! a tal nome io palpito”), it seems
to be the case, but we have to reconsider this opinion after a deeper
look at it. It is quite true that we find a greatly reduced number of
shots (also due to the staging, they are sitting together). It begins with
an plan américain of Amelia and the Doge. It changes only when the
stage direction indicates ‘Si abbracciano, ed Amelia parte accompag-
nata dal padre’ during an orchestral transition, and a long shot show-
ing the Doge kissing Amelia’s forehead. After Amelia has sung a last
“O Padre!”, we see a close-up of their arms, underlying the fact that
she hands the locket back and their emotion is shown by a series of
shots (a Doge’s close-up, a Amelia’s medium shot a Doge’s medium
close-up for his last “Figlia!” leading to a Paolo’s long shot as an an-
nouncement of his entrance and his question to the Doge).
The number of shots is by far inferior to an action piece, but in
this set piece, I have noticed that the editing accelerates at the end,
when there is hardly anything left to sing. At this point, the editing is
closer to that of an action piece. In this case the insistence is put on
the purely descriptive aspect. It is quite impossible not to look at those
shot’s changes because it is a constituent of classic narrative cinema
(if you lose some shots you are not able to understand the story). At
this very moment, the eyes are working faster than the ears, but it
appears to be mostly on orchestral transition. Is it the only case where
the editing logic takes liberties with the musical temporalities?
It seems almost logical to look for an answer in the rather uncon-
ventional duet between Violetta and Giorgio Germont (act II) in
Verdi’s La traviata. It is also a highly emotional moment because
Germont tries to convince Violetta to renounce his son, Alfredo, with
a perfidious argumentation. This duet is built on the solita forma, but
80 Delphine Vincent

it may be difficult to hear it the first time, because Verdi plays with it.
In fact, in order to accentuate Giorgio’s psychological blackmail,
Verdi proposes a very long tempo d’attacco, which presents a series
of brief sections linking one to the other (linking more the movement
with a theatre drama than an opera). As Powers writes:

Though Violetta has relatively few words to sing in the settenario stanzas, com-
pared to Germont, this part of the duet is as much musico-dramatic dialogue as
if their lines were equal in number; it is as kinetic dramatically as it is ongoing
verbally. Germont is applying three specious persuasions one after another, and
each one evokes a brief but dramatically significant punctuating response from
Violetta: first “Ah più non dite[...]”, then “Gran Dio!”, and at last, with “È
vero!”, she is vanquished; after her a parte contemplative reflection in “Così
alla misera” she adresses Germont in submission with “Dite alla giovine”, and
the first static movement in the duet is under way.

This very long tempo d’attacco may only be resolved in the adagio
when Violetta is convinced that she has to leave Alfredo (“Dite alla
giovine”). After it the solita forma runs very conventionally towards
her tempo di mezzo (“imponete”) and her cabaletta (“Morrò! la mia
memoria”), which expresses infinite pathos:

The convention of repeating the cabaletta after a “ritornello” is converted to a

poignant gesture of infinite pathos: the final “Addio” quatrain of new text is
sung through once, beginning in b flat minor fortissimo as though it were going
to be a “ritornello”; the character of the music changes midway, and Violetta
sings a reprise of her second cabaletta quatrain “Conosca il sacrifizio”, sung
now adagio, almost unaccompanied, and broken off without its last line.

For the study of how a video director films this duet I have chosen
Jürgen Flimm’s staging for Zürich Opernhaus (2005), mostly be-
cause I had the opportunity to see it at the theatre and it will allow me

13 Powers 1987, 80.

14 Powers 1987, 78. For a complete analysis of this duet, see Table III p. 79.
15 Giuseppe Verdi, La traviata, Eva Mei (Violetta Valéry), Thomas Hampson
(Giorgio Germont), Irène Friedli (Annina), Giuseppe (Noel Vasquez), Chorus
and Orchestra of the Zurich Opera House, Franz Welser-Möst (conductor), Jür-
gen Flimm (stage director), Felix Breisach (video director), Zürich, 2005.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 81

to draw comparisons. The following table presents a recapitulation of

the shots (for a complete description of the shooting, see Appendix I).

Scena Tempo Adagio Tempo di Cabaletta

(3:37) d’attacco (4:50) mezzo (3:32)
(6:50) (1:11)
ELS 2 3 1 – 2
LS 5 12 3 3 4
PA 13 5 7 2 11
MS 4 11 3 7 5
MC 1 14 8 – 2
Total 25 45 22 12 24

It appears that the frequency of cuts is higher in action pieces than in

set pieces. But if we look at the exceptions in the movements we can
find, for example, a very long shot at the tempo d’attacco’s beginning,
however this is not a surprise because Germont’s melody is indicated
cantabile. This is an indication that designates not only the adagio
movement but also any sustained and flowing vocal line. The first two
quatrains focus on Germont and only the third one proposes an alter-
native editing. The aim is to focus on the cantabile quality of the sing-
ing. On the contrary for Violetta’s stanzas, which present no structural
differences with Germont’s one (“Pura siccome un angelo” and “Non
sapete quale affetto”), we find a different editing, notably a more rapid
one. It corresponds naturally to the fact that her music is jerkier, and
seemingly out-of-breath. In this case, what is important is the expres-
sive difference between the two stanzas. It means that the image’s
narrativisation gives the text a statute that it does not possess.
On the contrary nearly half of the shots contained in the set pieces
are edited at the end of them, when text repetitions occur (as in the a
due section in the adagio) or coda (as at the end of the cabaletta,
where a rapid shot’s exchange allow to show the emotion of the pro-
Generally it seems that the video director tries to distinguish be-
tween action and set pieces, but if a narrative detail of importance or a
section with more dialogue does appear in a set piece, he chooses to
abandon the musical logic in order to follow a narrative one. If this
82 Delphine Vincent

sort of editing is not faithful to the musical concern, we have to de-

termine how it corresponds – or not – to perception in a theatre.
This is likely to be the very problem that the video director tries
to avoid in Simon Boccanegra’s adagio movement. It presents quite
an interesting shooting (for an extensive description, see Appendix II):
a few shots prepare the a due section, when the video director want to
avoid going back and forth between the protagonists. For this reason,
he proposes a few shots of Amelia singing with the Doge blurred in
the background and then two superimpositions prepare the TV specta-
tor for a sort of inverted pan and scan on the a due section. This op-
eration consists of cutting the middle of the stage, reuniting his ex-
tremities and allowing the audience to see father and daughter
together, even if they are actually very distant on stage. But this time,
the problem occurs in the first part of the adagio.
In set pieces the important factor is another – psychological –
temporality implying that audition takes the precedence over vision.
The operatic temporality differences imply an interior time which
opera relays find difficult to treat. The issue is probably linked to
Bergsons’s ‘temps vécu’.

Quand nous écoutons une mélodie, nous avons la plus pure impression de suc-
cession que nous puissions avoir, – une impression aussi éloignée que possible
de celle de la simultanéité –, et pourtant c’est la continuité même de la mélodie
et l’impossibilité de la décomposer qui font sur nous cette impression. Si nous la
découpons en notes distinctes, en autant d’«avant» et d’«après» qu’il nous plaît,
c’est que nous y mêlons des images spatiales et que nous imprégnons la succes-
sion de simultanéité: dans l’espace, et dans l’espace seulement, il y a distinction
nette de parties extérieures les unes aux autres. Je reconnais d’ailleurs que c’est
dans le temps spatialisé que nous nous plaçons d’ordinaire.

In preceding video examples, it is quite clear that this is not the case.
At the opera, the spectator must deal with a hierarchical variation be-
tween visual and auditory aspects in both action and set pieces. On the
contrary, opera relays do not respect this change, always presenting
the same relationship between vision and audition, giving priority to

16 Henri Bergson, La pensée et le mouvant (Paris: Quadrige/PUF, 2009), 166.

Quotation extracted from a conference, La perception du changement, at Oxford
University in 1911.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 83

the visual aspect. In a theatre set pieces do not seem static because the
spectator has started to use another mode of perception, a non-visual
one. By giving a visually still form, opera relay contributes to make it
concrete, to reify it, in opposition with the interior temporality implied
by music. Eventually it seems that video exacerbates the opera’s tem-
porality and therefore renders it even more problematic. The impossi-
bility of making the auditory aspect dominant over the visual aspect
during the set pieces suppresses one of the fundamental roles of this
alternation between set and action pieces. The image, even if editing is
less rapid, remains prevalent.
It seems that we have to accept that music and film are media
with different conceptions of temporality, form and narration. It means
that the dream of a perfect appropriateness between musical and vis-
ual forms is only a fantasy, because the types of media are not compa-
rable. Moreover in opera relays the video director is appointed to de-
liver a relay of a live-performance, in a conservational logic. Nothing
may be cut from the spectacle, contrary to opera on film (cf. the caba-
letta’s eradication in Zeffirelli’s films), which renders it more “cine-
Looking, with the solite forme’s analysis, for a visual equivalent
to the musical form and therefore temporality, I remarked that if the
articulations seem to be respected, it is not a guarantee of an equiva-
lence of perception (even if the film presents itself as enslaved to op-
era). Music, thanks to its succession and architectural simultaneity,
gives a psychological dimension to the chronometric time. Opera re-
lays give a visual succession to the opera and therefore places it in the
temps spatialisé.
84 Delphine Vincent

Appendix I

The Scena e Duetto begins with a few orchestral chords and an extreme long shot
showing all the stage (quite a conventional start, giving the spectator the opportunity
to see the set). It presents this shot breakdown.

V. Alfredo? (PA V.)
A. Per Parigi or or partiva. (PA A.)
V. E tornerà? (PA V.)
A. Pria che tramonti il giorno. (PA A.)
Dirvel m’impose.
V. È strano! (LS V.; on the repetition of the
initial chords, J. enters)
J. Per voi.
V. Sta ben. In breve (PA V./J.)
giungerà un uom d’affari. Entri all’istante.
Ah, ah! Scopriva Flora il mio ritiro, (ELS)
e m’invita a danzar per questa sera! (PA V./A.)
Invan m’aspetterà.
J. È qui un signore. (J. comes back in the PA)
V. Sarà lui che attendo (on the orchestral transition, G.
enters in the PA)
G. Madamigella Valery?
V. Son io.
G. D’Alfredo il padre in me vedete.
V. Voi?
G. Sì, dell’incauto che a ruina corre,
ammaliato da voi.
V. Donna son io, signore, ed in mia casa;

17 I removed the stage directions and use the following abbreviations: V. for Violetta;
G. for Germont; A. for Annina; J. for Joseph-Giuseppe. The shots are described as
following: “In the extreme long shot [ELS], the human figure is barely visible.
This is the framing for landscapes. [...] In the long shot [LS], figures are more
prominent, but the background still dominates. The so-called plan américain
(“American shot”) [PA] [...] the human figure is framed from the knees up. [...]
The medium shot [MS] frames the human body from the waist up. The medium
close-up [MC] frames the body from the chest up.” In David Bordwell and Kristin
Thompson, Film Art: An Introduction (New York: McGraw-Hill, 2004), 262.
18 Without indication, the shot begins with the in bold syllable.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 85

ch’io vi lasci assentite,

più per voi, che per me.
G. (Quai modi!) Pure...
V. Tratto in error voi foste! (PA V.)
G. De’ suoi beni (MS G.)
egli dono vuol farvi.
V. Non l’osò finora, (PA V.)
rifiuterei. (PA G.)
G. Pur tanto lusso...
V. A tutti (PA V.)
è mistero quest’atto, a voi nol sia. (on the orchestral transition, V.
moves forward G., in the middle of
it PA G. and V. enters in shot, then
LS V./G.)
G. Ciel! Che discopro! D’ogni vostro avere
or volete spogliarvi?
Ah, il passato, perché, perché v’accusa?
V. Più non esiste. Or amo Alfredo, e Dio (PA V.)
lo cancellò col pentimento mio!
G. Nobili sensi invero! (PA G.)
V. Oh, come dolce (LS V./G.)
mi suona il vostro accento!
G. Ed a tai sensi
un sacrifizio chieggo
V. Ah, no! Tacete! (MC V.)
Terribil cosa chiedereste certo!
Il previdi... v’attesi... Era felice (MS V. as following shot of previ-
ous one; LS G.)
G. D’Alfredo il padre (MS G.)
la sorte, l’avvenir domanda or qui
de’ suoi due figli!
V. Di due figli? (LS V.)
G. Sì. (MS G.)
Tempo d’attacco
G. Pura siccome un angelo
Iddio mi die’ una figlia;
se Alfredo nega riedere
in seno alla famiglia,
l’amato e amante giovine,
cui sposa andar dovea,
or si ricusa al vincolo
che lieti ne rendeva.
86 Delphine Vincent

Deh non mutate in triboli (LS V.; those two lines are repeated
twice, on the second occurrence PA
le rose dell’amor.
A’ prieghi miei resistere (MS G.)
non voglia il vostro cor. (during the orchestral punctuation
LS V.)
V. Ah! comprendo. Dovrò per alcun tempo
da Alfredo allontanarmi... Doloroso
fora per me... pur..
G. Non è ciò che chiedo. (MS G.)
V. Cielo! Che più cercate? Offersi assai! (LS V.)
G. Pur non basta.
V. Volete che per sempre
a lui rinunzi?
G. È d’uopo. (MS G.)
V. Ah no! Giammai! No, mai! (LS V.; MS G.)
Non sapete quale affetto (LS V.)
vivo, immenso m’arda in petto?
Che né amici, né parenti
io non conto tra’ viventi,
e che Alfredo m’ha giurato (PA G.)
che in lui tutto troverò? (MS V.; on the orchestral punctua-
tion PA G.)
Non sapete che colpita (MS V.)
d’atro morbo è la mia vita,
che già presso il finne vedo?
Ch’io mi separi da Alfredo!
Ah, il supplizio è sì spietato
che a morir preferirò! (on the last text repetition MS G. on
ah! – b flat’’ –, coming at the end of
the note MS V.; then on ah! – a’’ –
ELS V./G.)
G. È grave il sacrifizio; (MS G.)
ma pur, tranquilla uditemi. (ELS V./G.)
Bella voi siete e giovine. (PA V./G.)
Col tempo...
V. Ah, più non dite...
V’intendo. M’è impossible... (MC V.)
Lui solo amar vogl’io! (PA V./G.)
G. Sia pure, ma volubile
sovente è l’uom.
“Temps spatialisé”: Opera Relays and the Sense of Temporality 87

V. Gran Dio! (LS V. following shot of the previ-

ous one)
G. Un dì, quando le veneri
il tempo avrà fugate,
fia presto il tedio a sorgere. (MC V.; MC G.)
per voi non avran balsamo (ELS V./G.)
i più soavi affetti,
poiché dal ciel non furono (MC G.)
tai nodi benedetti.
V. È vero! (LS V.)
G. Ah dunque sperdasi (MC G.)
tal sogno seduttore!
V. È vero! (LS V.)
G. Siate di mia famiglia (MC G.)
L’angiol consolatore!
Violetta, deh pensateci, (LS V.)
ne siete in tempo ancor. (LS G.)
È Dio che ispira, o giovine, (LS V.; on the second repetition MC
tai detti a un genitor.
V. (Così alla misera, ch’è un dì caduta, (MC V.)
di più risorgere speranza è muta! (MC G.; MC V.)
Se pur benefico le indulga Iddio, (LS V.)
l’uomo implacabile per lei sarà).
G. Siate di mia famiglia, ecc. (during the repetition a due, MC G.,
then MC V., on her cadenza MS G.
on implacabile; MC V. on per; MS
G. when he starts again with siate)
V. Ah! Dite alla giovine sì bella e pura (MC V.)
ch’avvi una vittima della sventura,
cui resta un unico raggio di bene,
che a lei il sacrifica e che morrà. (on the text repetition MS G., then
MC V.)
G. Piangi, o misera... Supremo, il veggo. (ELS V./G.; on piangi’s repetition
LS V., then MS G.)
è il sacrifizio, ch’ora ti chieggo...
Sento nell’anima già le tue pene; (PA V.)
coraggio, e il nobil tuo cor vincerà! (MC G.)
88 Delphine Vincent

Then repetition a due,19 I use Violetta’s stanza to indicate the shooting and Germont’s
stanza for the coda.
V. Dite alla giovine sì bella e pura (PA V.; MC G.)
ch’avvi una vittima della sventura, (PA V.; MC G.)
cui resta un unico raggio di bene, (PA V.; MS G.)
che a lei il sacrifica e che morrà. (PA V.)
G. Piangi, piangi, piangi, o misera! (LS V./G.)
Coraggio e il nobile cor vincerà
Piangi, piangi, piangi, o misera! (MS V./G.; PA V.)
Coraggio e il nobile cor vincerà (MC G.; LS V./G.)
Coraggio e il nobile cor vincerà (MC G. ; PA V.)
Ah sì, il nobile cor vincerà (MC G.)
Tempo di mezzo
V. Imponete (PA V)
G. Non amarlo ditegli. (MS G.)
V. Nol crederà. (LS V.)
G. Partite! (MS G.)
V. Seguirammi. (LS V.)
Allor... (MS G.)
V. Qual figlia m’abbracciate, forte (PA V.)
così sarò. Tra breve e vi fia reso, (MS V./G.; on vi V. goes out the
ma afflitto oltre ogni dire. A suo conforto (LS V./G.).
di colà volerete.
G. Che pensate? (MS G.)
V. Sapendol, v’opporreste al pensier mio. (MS V.)
G. Generosa! E per voi che far poss’io? (MS G.)
V. Morrò! La mia memoria (MC V.; on Morrò’s repetition MS
V., following shot on from the pre-
vious one)
non fia ch’ei maledica,
se le mie pene orribili

19 “Dite alla giovine”, is indeed a “cantabile” in the formal sense that he seems to
intend, that is, a lyric set piece: first each character sings a stanza alone, which
takes up 28 measures; of the remaining 28 measures, sung entirely a 2, the first
16 comprise a complete repetition of the words and music of Violetta’s stanza,
accompanied throughout by Germont, who sings every half-line of his own
stanza except the first (not in their original order everywhere); that opening half-
line finally becomes the head-motive of the coda, which thus begins as though
his stanza too were going to be repeated, but then takes a different turn and con-
cludes with a typical cadenza comune.” (Powers 1987, 80)
“Temps spatialisé”: Opera Relays and the Sense of Temporality 89

vi sia chi almen gli dica.

G. No, generosa, vivere, (MS G.)
e lieta voi dovrete.
Mercé di queste lagrime
dal cielo un giorno avrete.
V. Conosca il sacrifizio (MS V.)
ch’io consumai d’amore,
che sarà suo fin l’ultimo
sospiro del mio cor.
G. Premiato il sacrifizio (PA G.)
sarà del vostro amore,
d’un’opra così nobile
sarete fiera allor.
For the repetition a due, I use Violetta’s stanza
V. Conosca il sacrifizio (MS V.)
ch’io consumai d’amore,
che sarà suo fin l’ultimo (LS V./G.)
sospiro del mio cor.
che sarà suo fin l’ultimo (MS V.)
sospir ... (LS V./G.)
For the following text repetition, the same shot is conserved.
V. Qui giunge alcun: partite. (MC V./G, V. goes out the shot at
the end of her line)
G. Oh, grato v’è il cor mio!
V. Partite... (PA V.; LS G.)
Non ci vedrem più forse. (PA V.)
V./G.Siate felice, (PA G.)
V. Addio! (PA V.)
G. Addio! (PA G.; PA V.)
V. Conosca il sacrifizio
che consumai d’amore, (ELS V./G.)
che sarà suo fin l’ultimo (PA V.)
V. Addio!
G. Addio! (LS G.)
V./G.Felice siate, addio! (PA V.; PA G.; PA V.; ELS V./G.)
90 Delphine Vincent

Appendix II

During the first four orchestral measures a series of shots is edited20: MC A.; PA D.;
MC A.; LS A./D. (continuing with the beginning of the singing).

A. Orfanella il tetto umile

M’accogliea d’una meschina, (MC A.)
Dove presso alla marina
Sorge Pisa... (MC D.)
D. In Pisa tu?
A. Grave d’anni quella pia (MC A.)
Era solo a me sostegno ;
io provai del ciel lo sdegno, ah! (C A., with D. fuzzy in the back-
ground; MC A.)
Involata ella mi fu. (during rests PA D.)
Colla tremula sua mano (MC A.)
Pinta effigie mi porgea,
Le sembianze esser dicea (MC A., with D. in the back-
ground )
Della madre ignota a me. (MS A.; during rests MS A. – different
angle – with D. in the background)
Mi baciò, mi benedisse,
Levò al ciel, pregando, i rai... (on pregando D. goes out of the
quante volte la chiamai
l’eco sol risposta die’. (first a superimposition and then
MC D.)
D. (Ah! se la speme, o ciel clemente,
Ch’or sorride all’alma mia,
Fosse sogno!... estinto io sia
Della larva al disparir!)
A. Come tetro a me dolente (first a superimposition and then a
PA A.)
S’appressava l’avvenir! (here an “inverted pan scan” begins
with PA A. + C D. during the a due
section; at the end D. disappears (on
disparir!) leaving A. alone in a MS.
On adagio’s last measure, ELS

20 I use the following abbreviations: A. for Amelia; D. for Doge.

21 This shot is obtained thanks to a change in the camera angle.
Overtures on Screen


The overture of an opera – by which I mean any instrumental intro-

duction – always has a somewhat ambiguous status: Is it part of the
opera? Is it music meant for an attentive hearing, should we listen to it
with a contemplative, aesthetic attitude? Or is its main function to call
for attention and introduce something that has not started yet?
When we sit in an opera house, we know full well what to expect:
the conductor appears and lifts his baton; it is time to be silent and
listen. The opera has begun. Music is in the foreground. We do not
have to watch, although in some theatres and in certain seats we may
be able and willing to watch the orchestra playing.
But what happens when we sit in front of a screen? Should our
mental behaviour as we listen to an overture correspond to the formal
dress required for the theatre or the concert hall? Or is it legitimate to
listen to it informally attired, maybe chatting with the person sitting
next to us, as we normally do in cinemas during the film opening titles?
A video production, whether on film or as a television broadcast
or a DVD, by definition implies watching – the existence of some-
thing to be watched – at all times. Normally, when we sit in front of a
screen in a cinema or at home, and something begins, the first things
we see are the opening credits, and most often there is music accom-
panying them. In our current mental habit we know that we may dis-
regard the sequence, or at least its soundtrack. The overture is the first
portion of the opera and has no binding visual dimension of its own:
in a video production its ambiguity is thus increased by its position.
During the overture, something will appear on screen and some-
body, usually the film director, has to decide what this will be. He has

1 Although not statutory, watching is nevertheless implied in stage productions

where the director invents an action to take place on stage during the overture.
92 Gaia Varon

to make choices as to where to put the film credits, with what music
and what image. He or she decides whether the overture will be a brief
symphonic concert before the opera, or whether the duration of the
overture will be used for something else – information about the pro-
duction or the opera, or maybe a visual narrative. Such decisions will
affect the spectators’ perception of the overture – and to a certain ex-
tent their perception of what will follow. Whether implicitly or explic-
itly, the director’s decisions convey a specific reading, a specific idea
of the opera, and thus call for some form of attention of the spectator.
This chapter will focus on different video versions of the over-
tures of a few Italian operas of the eighteenth and nineteenth century,
mainly by Mozart, Rossini and Verdi, offering a provisional catalogue
of possibilities. I will focus mainly on the overture – or any other in-
strumental introduction to an opera – as essentially a hybrid portion of
space and time, or as the threshold between two worlds: before the
overture starts, we are in our everyday space and time; at the end of
the overture we are ready to enter the space and time of the fictitious
world the opera offers us. In between, there is a time, precisely set by
the composer, during which the action on stage has not yet started, but
something has begun.
This investigation will examine excerpts from three sample
groups: a selection of Italian opera films produced immediately after
World War II; a selection of studio productions by one director, Jean-
Pierre Ponnelle; and some examples of what I will call the ‘symphonic
concert version’ of the overture. The video productions include opera
films, television studio productions and live recordings, mostly in
their remediated versions on VHS or DVD. This discussion is not
concerned with any sort of ‘philological’ or ‘historically informed’
approach: the aim is not to understand the original destination and
concept of the video productions – which would certainly be relevant
in other perspectives – but rather to consider video productions as they
are available today through DVDs and the internet.

2 In this chapter I will not touch on issues concerning the change of the material
form of a film and thus of its cultural form and meaning. These issues have been
rigorously discussed in J.D. Bolter and R. Grusin, Remediation: Understanding
New Media (Cambridge, MA: MIT Press, 1999).
Overtures on Screen 93

Let us start with some films produced in Italy after the end of World
War II. These are not the first complete opera films, but here we have
a flourishing production of opera films within the relatively short time
span of a few years. It probably answered a need, standing in for
theatre performances that were made difficult, if not impossible, by
the damage caused by the war. These films offer an interesting sample
of the possible presentations of the overture on screen, and we will
now examine some of these in detail.
A very well-known film director at the time, Carmine Gallone di-
rected Rigoletto in 1946. The black and white film is conceived as a
fictitious documentary shooting of a theatre production. The first
notes of the opening Prelude of Rigoletto are heard while we are pre-
sented for one second with a black screen, and then the image of the
first page of the score, with the front title of the opera itself and the
moving shadow of the conductor’s hand. Then, while the credits
scroll, we see a sequence of images of the conductor, of small groups
of instruments (the trumpets, the horns, the timpani) and of smaller
and larger sections of the orchestra. The sequence is well-timed with
the music and each change of image gives the impression of being
generated by the music. At the end of the Prelude the camera shows
the curtain opening while the first act begins. This way of filming an
overture, showing the conductor and the orchestra – the symphonic
concert version – is nowadays the most widespread approach, espe-
cially for broadcasts and live recordings of staged operas, but it is
unusual among the Italian opera films of the late 1940s.

3 On the flourishing of opera films in Italian cinema in the 1940s and 1950s see
G.P. Brunetta, Cent’anni di cinema italiano, vol. 2, Dal 1945 ai giorni nostri
(Bari: Laterza, 19982), 132–5; the chapter ‘Follie per l’opera’ in C. Bragaglia
and F. Di Giammatteo, Italia 1900–1990: L’opera al cinema, (Florence: La
Nuova Italia, 1990) (no page numbers); G. Casadio, Opera e cinema: la musica
lirica nel cinema italiano dall’avvento del sonoro ad oggi, Musica Cinema Im-
magine Teatro No. 15 (Ravenna: Longo Editore, 1995), 9–48; Cinéma et Opéra,
“L’Avant-scène opéra”, Mai 1987 (“Opéra” series No. 98, “Cinéma” series No.
360), 14–5 and 67–8.
4 E. Del Monaco, ‘Gallone Carmine’, in Enciclopedia del Cinema, ed. E. Siciliano
(Roma: Istituto della Enciclopedia Italiana, 2003), vol. 2, 690–1. A. Farassino,
‘La parole e il suono: il cinema-opera di Carmine Gallone’, in Non solo Scipio-
ne: Il cinema di Carmine Gallone, ed. P. Iaccio (Napoli: Liguori, 2003), 27–38.
94 Gaia Varon

We are given an effective counterexample by Mario Costa’s Il

barbiere di Siviglia, dated 1946 (but actually filmed in 1945). With
the first chord of the Andante maestoso with which the overture be-
gins, the logo of the film studio appears, with its name, ‘Tespi’, being
shown right afterwards. The title of the opera comes onto the screen
two bars later, just before the second chord, and it introduces the
graphics that are used all the way through the opening titles: a some-
what stylized picture frame whose shape is reminiscent of a curtain.
Within this frame the names of the performers appear, one after the
other, while we listen to the entire Andante maestoso.
The overture is used as an ordinary soundtrack over film credits,
basically a background music that may be quite independent from
what is to come. The idea of Rossini’s music as a generic background
music is reinforced by the fact that we actually have less than half of
the overture. Quite astoundingly, after the last chord of the Andante
maestoso, there is no Allegro con brio; instead we are abruptly thrust
into the beginning of the Introduction. Astonishing as it may sound
today, the cutting of most of the overture was certainly not a main
problem for Costa and his contemporary audience. Most of the films
connected with opera at that time were adaptations rather than com-

5 S. Bassetti, ‘Costa Mario’ in Enciclopedia del Cinema, ed. E. Siciliano (Roma:

Istituto della Enciclopedia Italiana, 2003), vol. 2, 158–9.
6 Or it may convey some information about what is to come. In Cinema’s Illu-
sions, Opera’s Allure. The Operatic Impulse in Film (New York, London: Con-
tinuum, 2002), David Schroeder writes: ‘The title itself generally does not tell us
if the film will be comic, serious… Usually before getting very far into the title
music that question will be answered’ (p. 85). And later, while comparing the
respective function of the overture in opera and title music in film: ‘Music
served a similar function for cinema from the earliest stages of the medium’ (p.
7 The Andante ends in B major, the dominant key leading back to E major, the
tonic in which the following Allegro begins and ends. The Introduction then
starts in G major. Without the Allegro, the transition in Costa’s film is somewhat
awkward, from B Major to G major.
Overtures on Screen 95

plete operas on screen and fidelity to the text was not a primary issue
even in theatre performances.
Costa’s decision to use the overture as title music, while credits
scroll on screen within a graphic format and without any specific film-
ing or shooting related to the opera, is not exceptional; it is in fact
very common, being obviously also the simplest solution and the clos-
est to cinema’s conventions. Nor is the discrepancy between the time
needed for the film titles and the duration of the opera’s overture un-
usual. While in a symphonic concert version, as in Gallone’s Rigo-
letto, the end of the film titles allows (finally, one may say) us a view
of the conductor and the musicians at work, Costa’s graphics could
certainly not be kept static or empty on screen while the music was
playing. Instead, Costa chose to cut part of the music, and the clear
division of the overture of Il barbiere di Siviglia’s into two clear-cut
sections made it an easy solution, although a questionable one on an
aesthetic perspective. Another common solution is to divide the
visuals during the overture into two parts; one part for the credits, and
a second part where something else is shown. We are given an inter-
esting example in Cenerentola, a film that Fernando Cerchio produced
in 1948.
The film begins quite ordinarily with the overture accompanying
the film titles scrolling on a black screen. The sequence takes a little
more than two minutes and then, eight bars before the end of the
Maestoso, there is a title card that says, in essence: ‘Fairy tales are
evocative and Rossini brought the brilliant and merry charm of his
music to the old fairy tale of Cinderella. Although the story in Ros-
sini’s opera differs from the traditional tale, it is nevertheless good old
Cinderella reminding us to keep our faith in the triumph of goodness’.

8 Cuts are frequent in the rest of Costa’s film of Il barbiere di Siviglia, even
within well-known aria and duets, like the Count’s cavatina ‘Ecco ridente in
cielo’ or Don Bartolo’s aria ‘A un dottor della mia sorte’.
9 ‘Suggestivo è il racconto di una bella fiaba. Il genio di Gioacchino Rossini ha
dato alla vicenda di Cenerentola – dalla fuliggine della cucina ascesa ai fastigi
del trono – lo scintillio malizioso e festoso della sua musica incantevole. La vi-
cenda musicata da Rossini è un po’ diversa da quella che si narrava, ma è pur
sempre la dolce, la buona Cenerentola a ricordare che è bene credere nel trionfo
della bontà.’ Cenerentola ossia La bontà in trionfo is the full title of the opera,
96 Gaia Varon

The title card is kept for forty-two seconds, until the end of the Maes-
toso, then – in perfect coincidence with the first chord of the Allegro
vivace – it switches to a new and briefer card: ‘Prince Ramiro is look-
ing for a bride. But it is not easy to find one as good as she is beauti-
ful. The wizard Alidoro comes to his aid…’ A film sequence then
begins, which continues for the whole Allegro vivace of the overture,
showing the brief action that must take place between what the title
card has introduced and what the first scene of the opera will show, in
other words a short antefactum to the First Act: Alidoro is travelling
on a horse-drawn carriage; he examines an enchanted book, which
tells him that goodness, grace and beauty can be found at the old cas-
tle of Montefiascone. Alidoro then abandons his rich dress for a peas-
ant’s attire, sends a carrier pigeon off and moves towards the old cas-
tle. The camera follows the pigeon’s flight until we see a magnificent
palace, which we enter and pass through, until we meet Prince
Ramiro, who is having breakfast in a splendid room. The pigeon ar-
rives with the message for the prince: in Montefiascone he will find
three sisters; if he will go there disguised as a poor man he will be
able to tell where pure innocence and goodness lie. The prince starts
changing, messengers leave from the castle; during the last bars of the
overture we see the disguised Alidoro approaching the castle of Mon-
The sequence is cleverly constructed, with a fine exploitation of
musical elements of the Allegro vivace, whose connections with the
visuals make it function effectively as film music. The hopping qual-
ity of the first theme, for instance, fits in convincingly with the trot-
ting horses at the beginning of the sequence, and when – less than
three minutes later – the theme returns, it brings a touch of irony to the
servants’ march through the palace with the breakfast tray. The cres-
cendo well serves the purpose of animating the two parallel moments
where a decision takes place, the first one when Alidoro changes his
dress and moves to the castle, the second one when Ramiro, having

whose libretto by Jacopo Ferretti tell in its own terms Charles Perrault’s fairy
10 ‘Il principe Ramiro cerca una sposa, ma non è facile trovarne una che sia tanto
buona quanto bella… Il mago Alidoro viene in suo aiuto…’
Overtures on Screen 97

read the message brought by the pigeon, starts undressing and giving
instructions. The most effective, and at the same time gracefully
amusing moment is when the second entry of the second theme,
played first by the flute and then by the piccolo, is used almost
diegetically: while Dandini is serving breakfast to the prince he turns
around, as if having heard something, just as the strings play their
pizzicatos; as if following his look, the film cuts to a close-up of the
pigeon while the brief motif is played by the flute. The whistling qual-
ity of this brief musical motif, enhanced by the timbre of the flute,
makes it easy, although not necessarily conscious, to perceive it as the
‘sound of the pigeon’.
Cerchio seems to have made good use of both the details and the
form of the overture. His approach effectively exploits the clear two-
part structure of the overture: the first and somewhat more preparatory
is used for the film titles, while to the second one Cerchio entrusts the
task of introducing the opera.
There is not space here to list in detail the different solutions
adopted in other contemporary Italian opera films by the same direc-
tors, or by others such as Piero Ballerini or Cesare Barlacchi, but a
brief comparison allows us to sketch a few common principles. First,
as in Costa’s Il barbiere di Siviglia, the integrity of the music is not a
priority. To cite only some examples, Ballerini’s film of Lucia di
Lammermoor cuts out nineteen bars of the Larghetto; while Barlacchi
simply skips the opening Larghetto of the overture of Donizetti’s La
Favorita and begins his film with the following Allegretto, of which
he uses the first forty-seven bars, then cuts to bar 70, keeping in an-
other fifty-one bars and omitting the rest. Second, with the exception
of Gallone’s Rigoletto, the overture is not used to introduce the idea of
the musical performance, although at the very beginning of Costa’s
film of Donizetti’s L’elisir d’amore the sound of an orchestra tuning
up is heard, while we see the logo of the production company and the
title of the opera, and then the image of a detail of the score is used as
the background for the following credits to scroll in front of. Third, a
title card or even a voiceover is used quite frequently during the over-
ture or right at the end of it.
Apart from Gallone’s Rigoletto, which is an exception that we
will return to, in most Italian opera films of the 1940s and 1950s the
98 Gaia Varon

overture is used as part of a typical film opening sequence, always

with credits and sometimes with visual or verbal inserts giving some
information about the plot of the opera. What kind of listening is im-
plied by such visual elements that go with the overture? Music is
mostly used as a soundtrack, a film’s title music. Although it is in the
acoustic foreground, the fact that the visuals give us something else –
something that is not yet part of the fiction on screen and conveys
information that may be interesting but is not essential to the enjoy-
ment of the fiction – pushes the music to the background of our over-
all perception. In Cerchio’s Cenerentola the second part of the over-
ture functions much as music would in a silent film – an accompani-
ment with only a distant connection with the film sequence – and here
again it does not demand of us the kind of attentive listening that is
usually implied in a theatre or in a concert hall.

My second sample includes some video productions by Jean-Pierre

Ponnelle. Most of them date to thirty to forty years later than the
Italian opera films examined above and are television studio produc-
tions. The difference in production and destination between cinema
and television, although relevant in other perspectives, does not
change the main problem a director has to face with the overture: what
is to appear on screen while we hear the music? In his screen produc-
tions Ponnelle generally uses the overture to tell the spectator some-
thing either about the opera itself or about the specific music and stage
production. He often uses visual elements or the kind of narrative we
saw in Cerchio’s Cenerentola.

11 Ponnelle’s work has been studied extensively by Marcia Citron; see for instance
her ‘Subjectivity in the Opera Films of Jean-Pierre Ponnelle’, Journal of Musi-
cology, 22: 2 (2005), 203–40; a revised version is in M. Citron, When Opera
meets Film (Cambridge: Cambridge University Press, 2010), 97–135.
12 In his 1981 television production of Rossini’s Cenerentola, conducted by Clau-
dio Abbado with Milan’s Teatro alla Scala chorus and orchestra, the overture is
used to introduce the location of the production: it opens with a view of La Scala
from outside and it finishes with the image of the closed curtain, having ex-
plored the foyer and the hall (the recording was released on DVD in 2005 by
Deutsche Grammophon with catalogue number 04400734096).
Overtures on Screen 99

This approach is used in Così fan tutte (1988). The filming of
the overture for Così fan tutte is clearly articulated in two parts (not
corresponding to a parallel articulation of the music). The music be-
gins on its own, while we are presented a black screen. Music is thus
unequivocally in the foreground and we are allowed, or even per-
suaded into an attitude of aesthetic, contemplative listening. Very
slowly, the image of the closed curtain fades in, onto which credits
start scrolling; since we are actively listening, each visual element
appears to be clearly timed with the music, it somehow gives the im-
pression of being generated by the music, or at least to respond to it
naturally and gracefully. Although the overture is here used for cred-
its, the careful timing of the images prevents the soundtrack effect,
and the music is not pushed in the background of our experience. The
filming of the first part of the overture is thus well conceived to in-
duce or support attentive listening. Once the credits are finished, a
new film sequence begins: the curtain rises, and very gradually we
discover the space where the dramatic action will begin. The sequence
is not narrating a separate antefactum, nor is it explicitly or implicitly
anticipating something that will take place during the opera; it literally
‘introduces’ us to the space and time of the opera. What we see, at
first from a distance and then gradually closer and closer, is an action
with no specific sound, actually a pantomime: the characters are talk-
ing, and we may well feel that the first number of the opera com-
mences in perfect continuity. The introductory function of the over-
ture is literally made visual.
The most interesting example of a visual narrative during the in-
strumental introduction to an opera is given by Ponnelle’s well-known

13 W.A. Mozart, Così fan tutte, Wiener Philharmoniker, cond. N. Harnoncourt,

staged, directed and designed by Jean-Pierre Ponnelle, Unitel, 1988 (DVD re-
leased 2006, Deutsche Grammophon, 04400734237)
14 The very first credit, ‘Così fan tutte’, enters precisely on the forte at bar 12,
when the orchestra plays the second part of what can be considered the motto of
the opera, i.e. eight bars (bars 8–15 of the overture) corresponding to the words
‘Così fan tutte’, as sung in Act 2, No. 30, once by Don Alfonso alone, piano,
and immediately repeated by him, Ferrando and Guglielmo, forte (bars 20–26).
100 Gaia Varon

film of Verdi’s Rigoletto (1983). Film credits run over a somewhat
faded picture of Parma’s Teatro Farnese; the theatre is empty and
there is no sound whatsoever for twenty-five seconds. When eventu-
ally the credits for the orchestra (the Wiener Philharmoniker) start to
fade out, music begins, very softly, trumpet and trombone, as if nam-
ing the players had elicited their sound. As they play their second
note, a C again, we are told who is conducting these sounds from a
music stand we do not see: the name of the conductor (Riccardo
Chailly) appears. On the first chord (bar 2) we are presented with the
very last credit, that of the film director, but very soon the credit dis-
appears and a character comes in – he literally enters the framing. It is
now clear that the image behind the credits, which could have been a
picture, a still frame, is in fact the space of the action. The careful
timing brings life to the space, and the space guarantees the continuity
between the time of silence and that of music.
The visual narrative that follows is precisely built on a shrewd
segmentation of the music: first there are three bars for the connection
between credits and the beginning of the visual narrative; then three
more bars – the answering half-sentence – in which nothing happens,
we are facing, and getting familiar with, the character that has entered.
In the following two bars, a shortened version of the initial half-
sentence, we have the first camera movement, a fast zoom out that
gives us the full figure of the character, his space – the space in which
he moves (until now, we could only see a somewhat blurred back-
ground) – and a prop on the floor, which the character picks up during
the following two bars. At bar 11, when trumpet and trombone once
more play their repeated C, now with a sustaining tremolo of the
lower strings, the character begins to walk forwards; the camera
moves backwards, at first in pace with him and then a little faster, thus
opening up a wider portion of the space and revealing a female body
lying on the floor. It is exactly on the bursting of the fortissimo chord
(bar 15) that the character seems to discover the body and lets the prop
drop. The again repeated C (bar 16) gives him the time to approach

15 G. Verdi, Rigoletto, Wiener Philharmoniker, cond. R. Chailly, directed by Jean-

Pierre Ponnelle, Unitel, 1983 (DVD released 2006, Deutsche Grammophon,
Overtures on Screen 101

the body and kneel beside it, and during the next fortissimo chord (bar
17) he touches it; C again while he tries to move it and another fortis-
simo chord (bars 18 and 19 respectively) while he takes it into his
arms. Now the violins, and the higher woodwind with them, play a
‘crying’ melody and the strength with which character clasps the body
in his arms seems to reduce with the descending melodic line and he
delicately caresses the woman’s face with diminishing energy while
the orchestra moves from the initial fortissimo to diminuendo and
finally pianissimo. After a brief silence (prescribed by the score, the
rest at the end of bar 24), the trumpet plays the repeated C once more
and now the character looks up to his right, as if he heard something,
and during the repetition of the motif by the trombones (bars 27–28)
he turns to the left, while his facial expression changes to suspicion, as
if he perceives a threat. When the roll of the timpani, pianissimo,
starts (bar 29), the character bends again over the dead body he is still
loosely holding and looks lovingly at it; then, as the music moves into
its final crescendo and the camera accordingly zooms in, he lifts his
head, looks straight at the spectator (at the camera) with an increas-
ingly worried expression until, at the end of the long fortissimo chord
of bars 33–34, there is a sudden cut to a view of a river under a stormy
sky with a city skyline in the background, and this is kept until the end
of the chord. During the rest at bar 34, we see the character again, now
with a truly terrified expression and holding the dead body closer to
his chest; on the final chord (actually a unison C from the whole or-
chestra), the landscape view reappears on screen and immediately a
painted curtain drops.
It is a fine and effective visual construction evoked by the music.
There are no cuts until the last chord and every shot and camera
movement is carefully synchronized with the music. Such close rela-
tion between the two elements is something Jean-Pierre Ponnelle was
steadily aiming for, as he himself stated:

Ich glaube, die Kameratechnik, die Film- und Fernsehentechnik erlauben eine
zusätzliche musikalische Komponente, eine Interpretationskomponente, um Mu-
siktheater zu realisieren, zum Beispiel durch Schnitt, Travelling mat[t] der Ka-
mera, Optik, Slow-motion… Der Schnitt kann Rhythmus, Travelling mat[t] Cre-
scendo sein. Die absolute, autogene Technik der Kamera gehört zur musikali-
schen Sprache, zum musikalischen Vokabular. Es reizt mich sehr zu interpre-
102 Gaia Varon

tieren, Oper oder Musikalische Stücke mit einem technischen Vokabular anzu-
reichern, mehr als es auf der Bühne möglich ist.16

Such filming of an overture does not bring in the idea of the musical
performance; it actually absorbs the overture in the sequence of the
opera: the spectator is not induced to an attitude of pure listening; he
or she is immediately swallowed up by the dramatic action. The brief
sequence during the overture is not prescribed by the libretto, but, one
might say, it is implied by the score. The Prelude is in fact based on
what is usually referred to as the ‘curse motif’ of Rigoletto and is thus
dominated by a sense of threat, anguish and ineluctable fate. What
we see in Ponnelle’s film is actually Rigoletto facing the fatal out-
come of his own fatherhood. The spectator already familiar with the
plot will certainly identify the main character of the sequence as Rigo-
letto and the apparently lifeless body as Gilda, but in fact, it is not that
simple: Ponnelle has the same singer, Ingvar Wixell, play the roles of
both Rigoletto and Monterone; the body that Wixell hugs in the Prel-
ude is that of Edita Gruberova, who will sing the role of Gilda, but at
the beginning of the Introduction Wixell brings the seemingly lifeless
body right into the feast at the Duke’s palace and the viewer now in-
fers that it is the body of Monterone’s daughter. The sequence actu-

16 ‘I think that the camera equipment, the film and television technology allow an
additional musical component, an interpretative component to produce music
theatre, through editing, camera travelling, camera optics, slow-motion [...] Edit-
ing can become the rhythm, a travelling a crescendo. The absolute, autogenous
technique of the camera belongs to the musical language, the music vocabulary.
I find it very exciting to interpret opera or music pieces by means of a technical
vocabulary, to enrich them more than it is possible on the stage.’ ‘Bemerkungen
zur Opernregie im Fernsehen. Gespräch zwischen Jean-Pierre Ponnelle und Dr.
Heinz Oepen anläßlich der Fernsehinszenierung der Hochzeit des Figaro’, Mu-
sik im ZDF (1976/77), 14–17: 14, quoted in Kii Ming Lo, ‘Der Opernfilm als
Erweiterung der Bühne: Versuch einer Theorie an Hand von Jean-Pierre Ponnel-
les Rigoletto’, in Das Musiktheatre in den audiovisuellen Medien: Vorträge und
Gespräch des Salzburger Symposions 1999, ed. Peter Csobadi, Gernot Gruber,
et al. (Anif/Salzburg: Müller-Speiser, 2001), 266.
17 See for instance D. Kimbell, Verdi in the Age of Italian Romanticism (Cam-
bridge: Cambridge University Press, 1981), 439.
18 In other words, Ponnelle exploits and pushes forward the correspondence be-
tween the two father/daughter couples. On this, see a detailed analysis in Kii-
Overtures on Screen 103

ally narrates events prior to the opera, but it is extremely interesting

that the brief sequence works well and effectively whether the body
belongs to Gilda or to Monterone’s daughter, that is, as a premonition
– as the knowledgeable viewer may be at first led to believe – or as a
prior event, as it later proves to be. It works in fact because the music
itself embodies the link between the two events. At the same time, the
brief sequence creates symmetry: it frames the opera with an introduc-
tory scene that will be complemented by the final one. It might be the
case, though, that while clarifying the premonition, in a way the scene
weakens it; by making it dramatically explicit, the scene weakens the
sense that there is something fatally imminent and yet still unpredict-
able that the pure listening of the Prelude conveys. In spite of this
quibble, the effectiveness of the close relationship between music and
image that Ponnelle achieved in the filmed Prelude of Rigoletto is
certainly unsurpassed in his other work.

As a last sample, I would like to focus on some examples of what I

call the symphonic concert version of the overture; that is, films that
show the complete performance of the overture. As we have seen, this
is not a very common choice for opera films or studio television pro-
ductions, but it is a frequent option when filming theatre productions
of opera. In this type of audiovisual production, the director normally
has or wants to convey to the screen audience the sense of the live
event, for which showing the hall, the audience and the players is an
essential approach. As in the other types of production, there is still
the need to place the credits. The director must thus find a satisfactory

Ming Lo, ‘Der Opernfilm als Erweiterung der Bühne: Versuch einer Theorie an
Hand von Jean-Pierre Ponnelles Rigoletto’ in Das Musiktheater in den audiovi-
suellen Medien: Vorträge und Gespräch des Salzburger Symposions 1999, ed.
Peter Csobadi, Gernot Gruber, et al. (Anif/Salzburg, Müller-Speiser, 2001),
19 It is however a premonition: Rigoletto would not show such affection for the
dead young woman, and such terror, were he not foreseeing his own destiny.
20 Obviously, this was made possible by the particular musical construction of the
Prelude itself and by its specific musical connection with what follows in the
104 Gaia Varon

combination of the three elements: credits, live event and the perform-
ance of the overture.
When filming the performance, although the positions of the mu-
sicians and the cameras are quite different, the filming techniques and
strategies are the same as those used to film symphonic concerts. To-
day the performance of a symphonic work is filmed most often along
what I would call the ‘standardized approach’ – an approach that aims
at a good balance between all the available ingredients: images of the
conductor, of the orchestra, and some details of individual instru-
ments. The details have a very clear, almost didactic function; they
bring the spectator’s attention to a specific sound element by showing
its source. The shots of the conductor and those of groups of musi-
cians may sometimes fulfil a merely informative task, at other times
they may have an emotional charge, and very often carry out both
functions at the same time. This approach may be seen as a way of
inviting, or perhaps imposing, an attitude of attentive listening. For-
mally dressed musicians, we might say, bring with them the contem-
plative habit. This is generally true for any production of this kind, but
once we start to examine some of them more closely, we are con-
fronted by a variety of situations. I will focus on three examples, all
recordings of staged performances of Mozart’s Don Giovanni.
The La Scala 1987 film production of Don Giovanni, conducted
by Riccardo Muti with Giorgio Strehler’s stage production and Carlo
Battistoni’s TV direction, is a good example of the standardized ap-

21 I describe and theorize the standardized approach in my PhD dissertation, pre-

sently in progress (‘Tecnica, stile e ideologia nella musica sinfonica in video: la
Quinta Sinfonia di Beethoven’, University of Bologna). See also my ‘Symphon-
ic Music on Screen. Beethoven’s Fifth Symphony’, in Proceedings of the Inter-
national Conference on The Embodiment of Authority: Perspectives on perfor-
mances, Helsinki: Sibelius Academy, September 2010 (forthcoming).
22 In all three, the filming strategies aim at conveying the sense of the event, al-
though only the first is actually a live recording. In this chapter I will not touch
upon such issues as what exactly is a ‘live recording’ and what kind of problems
are connected with it. The general issue of liveness is thoroughly discussed in
Philip Auslander’s well-known book, Liveness Performance in a Mediatized
Culture (New York, London: Routledge, 20082). For a discussion of liveness in
mediated opera see also E. Senici, ‘Il video d’opera “dal vivo”: testualizzazione
e liveness nell’era digitale’, Il saggiatore musicale, 16 (2009), 273–312.
Overtures on Screen 105

proach as applied to operatic overture. The film opens with a long
sequence devoted to the credits, made of two different images; at first,
we see a long shot of the La Scala auditorium, with the curtain in the
background, the lights still on, the orchestra softly tuning up and re-
hearsing individually, and the audience quietly finding their seats. The
image then changes to a medium shot of the curtain, now a still frame,
which is kept for almost two minutes, until eventually we are brought
back to the filming of the hall. During the whole sequence, we always
hear the buzzing sound of the hall and the orchestra, while the credits
appear. Finally, the lights go down, the conductor comes in and the
opera begins. Through the opening Andante of the overture the focus
is mostly on the conductor: the first shot, on Muti alone, is kept for
about forty-five seconds; then, as if cross fading, a shot of Muti and
the orchestra appears, but in fact it never quite obliterates the bigger
Muti of the first shot: the superimposed images stay on screen for
more than one minute, a very long time by television standards. We
leave it only when the Allegro starts and we are then offered a more
descriptive approach: shots of small groups of instruments alternating
with shots of the conductor and the orchestra. The alignment of the
different functions is linear, simple and clear. The long sequence be-
fore the beginning of the music fulfils the task of conveying the sense
of the live event while giving the relevant information; once the music
starts, the filming of the Andante pivots mainly on the emotion of the
live event, while introducing the mechanics of the music; with the
Allegro the focus on the mechanics narrows and we are offered a film-
ing of the musical performance with a good balance between emotion
and information.
In the 1987 film of Don Giovanni’s, conducted by Herbert von
Karajan – who also holds, together with the director, the responsibility
for the TV realization – the sense of a live event, of its space and its

23 W.A. Mozart, Don Giovanni, Orchestra del Teatro alla Scala, cond. R. Muti,
prod. G. Strehler, dir. C. Battistoni, Rai Trade, 1987 (DVD released 2002, Elleu
Multimedia, no catalogue number).
106 Gaia Varon

audience, is partly conveyed by the sound. We first see a view of
Salzburg, over which the very first titles appear; the image then
switches to a picture of a dark wall, which is in fact, as one finds out
when it opens up during the last bars of the overture, the ‘curtain’, and
this image is kept for most of the time while the credits scroll and
while we hear an alleged ‘real’ sound – a diegetic off sound – of the
audience taking their seats and waiting for the performance to start.
This sound – added during the editing process – is a genuine sound-
track; it gives a realistic sonorous embodiment of the time before the
start of the performance without showing it. When the last credit, the
name of Herbert von Karajan, appears on screen, we hear the ap-
plause, and only after that do we also see the audience, very briefly,
before the conductor is shown and the music begins. Here again, as in
the La Scala production, the filming of the musical performance
seems to pivot on the emotion, starting with a long shot of the conduc-
tor. Although there are more details on individual instruments, the
filming never becomes merely ‘informative’, visually translating an
implicit analysis of the score: the pivot seems to be an organic, con-
tinuous link between the music – Music, actually, with a capital M,
indicating ‘great music’ –, the conductor and the orchestra. Each shot
is enveloped in the adjacent ones and is itself enveloping the others;
the choices of lighting and shooting effects make each image smooth;
the whole filming is fluent, as if there were no cuts between shots. Far
from aiming at inviting the audience to a somewhat ‘analytical’ un-
derstanding of the mechanics of the music, the Karajan film seems to
impose an absorbing, even immersive listening attitude.
In a 1954 film of Don Giovanni, also shot in Salzburg and con-
ducted by Wilhelm Furtwängler, the invitation to wear the formal
attire of contemplative listening is explicit. The film opens with a
full screen image of a drawing that soon rotates slightly, revealing a

24 W.A. Mozart, Don Giovanni, Wiener Philharmoniker, cond. H. V. Karajan, prod.

M. Hampe, dir. C. Viller and H. V. Karajan (artistic supervision), ORF/ZDF/
Telemondial, 1987 (DVD released 2008, Sony Classical 88697296049).
25 W.A. Mozart, Don Giovanni, Wiener Philharmoniker, cond. W. Furtwängler,
produced and directed by Paul Czinner, Salzburg Festival 1954, Unitel, 1955
(DVD released 2001, Deutsche Grammophon, 073 019).
Overtures on Screen 107

framing: the image is in fact the cover of a book that is opened by a

barely visible hand, and all the credits will appear in individual pages
of the book as they are being turned one after the other by the same
hand. No sound whatsoever. The last page has a text: ‘Both the idea
which underlies this film, and the methods by which it has been car-
ried out, originated with Paul Czinner. He elaborated and applied the
technique by means of which outstanding stage performances can be
caught and preserved, for the enjoyment of wider audiences today and
as a record for posterity’. The film’s director, Paul Czinner, explicitly
draws the spectator’s attention to the ‘outstanding’ quality of the per-
formance featured in the film, and in so doing requires from the spec-
tator an aware and attentive listening and viewing. The book page
fades out and the camera follows the conductor as he enters from the
side of the pit and walks to the music stand, very briefly bends to the
audience – so that we see his face on screen – and then turns around to
the orchestra and gives the cue. During the first chord we see the back
of the conductor; in the background two players in the pit are barely
detectable; then the film cuts to a shot of the conductor’s face and
During the whole overture we will see nothing but these two
shots – and mostly the one showing Furtwängler’s back. No details,
no orchestra, no camera movements. In a sense, and with the excep-
tion of the few shots showing the conductor from the front, throughout
the overture Czinner generally gives the film spectator the view that a
spectator sitting in the hall would have. It is not a necessity, it is a
choice. It is as if Czinner tried to bring within the medium itself what
Walter Benjamin called the ‘aura’. The written text preceding the
performance is an essential part of this strategy: its aim is not so much
to give information about the performance, but to summon the specta-
tor’s attention, inducing a respectful, reverent attitude. The audience –
which is never seen nor heard – is not brought nearer the aesthetic
object: the relation between the conductor – that is, the music – and
the audience is hierarchical. In doing this, Czinner creates a specifi-
cally mediatic ‘aura’. As a result of this approach this film is the least
‘informative’: not only there is no use of filmic devices to enhance
musical details, but we do not even get an idea of how many musi-
cians are in the orchestra or how they are seated.
108 Gaia Varon

It would obviously be relevant to investigate whether there is consis-

tency between each of the different choices described above for the
filming of an opera overture – informative, emotional, analytical-
aesthetic or ‘auratic’-aesthetic – and the filming of the rest of the op-
era. In other words, whether the overture is – always or at times –
perceived as a picture frame, an audiovisual frame that can be de-
signed in close relation to the subject of the artwork it frames, to its
the style, to its specific location, or that can be fairly independent. One
would also need to consider, of course, the opera’s specific style: the
relation between the Prelude and the rest of the opera in Rigoletto is
very different from that of Il barbiere di Siviglia and its overture. Here
I can only try to summarize the results of my above analysis by group-
ing them into four categories of how images and music are related in
opera overtures in screen productions: background music, film music,
programme music, symphonic music performance.
Background music: Costa’s Il Barbiere di Siviglia. The specific
combination of sounds and images does not bring the spectator’s at-
tention to the music; on the contrary, the sequence is explicitly an
informative one, with information coming from the screen to the eye,
similar to what happens during the opening credits in many films and
television programmes. Obviously, it is implicit that the music that
accompanies the sequence is part of the opera we are watching, but
the specific audiovisual approach in combination with widespread
conventions and habits not only fails to enforce this unity, but some-
how underlines a difference, a separateness, a ‘non-belonging’ of the
overture to the musical and dramatic action which is about to start. In
other words, it invites an absent-minded listening.
Film music: Cerchio’s Cenerentola. It is the most interesting and
most creative approach from a film perspective. The director chooses
to use this threshold, this in-between time, to add something, a spe-
cific combination of information about the opera and a filmic inven-
tion, but without a close relation to the musical construction. The mu-
sic thus accompanies the action, generating a specific relation between
images and music. In some individual fragments the relation is closer
and the music seems to function as if conceived originally as film
music – as music illustrating or enhancing what appears on the screen.
In other words, the music often functions as if the visual had been
Overtures on Screen 109

conceived independently and the music designed to cooperate with it.

It invites, to quote and distort the title of Claudia Gorbman’s well-
known book, an ‘unhearing’ attitude.
Programme music: Ponnelle’s Rigoletto. The director narrates
filmically something that he extracts from the music. In what we cate-
gorize as film music the director has more freedom, with the visual
narrative independent of the music itself, but in programme music the
relationship is if not strict, at least always very close, with the visual
narrative making visible something that is already present in the mu-
sic. Semantically, and not only chronologically, the music comes be-
fore the image on screen. As Ponnelle says:

Opernfilmregisseure müssen musikalisch sein. Ich verlange, daß sie mit der Mu-
sik, in der Musik, vielleicht auch gegen sie, aber in einer Relation zur Musik,
denken und agieren. In der Oper ist Musik die Essenz und keine Illustration.
Wer das respektiert, ist richtig am Platz.

We could say that this filming strategy induces a semantic or theatri-

cal listening.
(Absolute) Symphonic music performance: the three Don Gio-
vanni. No narrative is added; the film offers exactly what the score
prescribes and, with it, requires an attentive listening. Nevertheless,
through heir filming strategies directors may enforce specific nuances
and specific ways of that listening: analytical, emotional, immersive,
Obviously there are also several possible variants and combina-
tions among these categories, some of which we met in the examples
examined above, such as dividing the music of the overture in two (or
more) parts with different relations between the music and the image;
or the choice of Gallone’s Rigoletto, where the overture is filmed

26 C. Gorbman, Unheard Melodies: Narrative Film Music (Bloomington, Indian-

apolis: Indiana University Press, 1987).
27 ‘Opera directors must be musical. I demand that they think and act with the
music, in the music, perhaps even against it, but in relation to the music. In op-
era music is the essence, not illustration. Those who respect this, have it right.’
Imre Fabian, Imre Fabian im Gespräch mit Jean-Pierre Ponnelle (Zürich/
Schwäbisch Hall: Orell Füssli, 1983), 54, quoted in Lo, Der Opernfilm als Er-
weiterung der Bühne, 273.
110 Gaia Varon

along what we called the symphonic concert version but is also used
as the audiovisual background for credits (a frequent choice also in
live recordings).
Let us go back to our starting points: the overture as the threshold
between two worlds, between our everyday space and time and the
space and time of the opera’s imaginary world; the need for any video
production to provide something visual for the screen during the over-
ture; and the possible impact of the director’s choices on the spectator.
Can we say that one or another of these categories serves the purpose
better? A last brief comparison between the two Rigoletto films exam-
ined above and three live recordings of stage productions gives inter-
esting cues and may help to provide an answer.
In Gallone’s film, the Prelude is filmed, as we said, as a sym-
phonic concert version, with a skilful combination of information and
emotional shots and an effective use of camera movements. The visual
dimension seems to aim strictly at making the spectator focus on the
music: the first image we see is the score and right after, the conductor
from inside the pit. This ideal listening attitude is made difficult, how-
ever, by the overwhelming presence of the credits over the shots of the
performance. In Ponnelle’s version, almost every microelement of the
music – orchestration, chords, melodic profiles – triggers off a micro-
element of the action on screen, in a steady relationship that disre-
gards any possible formal segmentation of the piece and pivots both
on the surface – the series of microelements – of the music and on the
dramatic meaning of the musical motifs of the Prelude. Albeit in dif-
ferent ways, both Gallone’s and Ponnelle’s films recall the original
stage destination of Rigoletto, and they are both films in which the
director holds the responsibility for creating the scenic action and for

28 An extremely interesting case is the celebrated film of Mozart’s The Magic

Flute by Ingmar Bergman, showing most of the time a fictitiuos audience with
some inserts of a country landscape and the exterior of the theatre, where the fic-
titous stage production takes place. Formally it is constructed as a sequence aim-
ing at introducing a theatrical event, but the faces of the persons included in the
fictitiuos audience, their expressions, their origin tell another story, tell in fact a
whole reading of Mozart’s opera and its meaning. The filming choices serve
here, not unambiguously, two – or more – aims at the same time, a multiple
filming strategy, we could say.
Overtures on Screen 111

filming it. We may briefly compare these two films with three – out of
the many – live recordings of stage productions of Rigoletto: Metro-
politan Opera, 1977; Teatro alla Scala, 1994; Opéra National de Paris,
1996. In this last one, stage director Jérôme Savary invented an ac-
tion to be performed on stage during the Prelude and the TV director
has to show it. All three versions must as usual decide where to put
credits, and they all do it during the Prelude. In the Opéra National de
Paris production, the stage action conceived by the stage director is
carefully built on the music: with ghostly lighting, a palace in ruin is
populated with a few sleeping people; on the central crescendo of the
Prelude (bars 11–15), from an open door at the back of the stage a
man (Monterone) arrives, and during the three repeated fortissimo
chords he stands still; as soon as the violin motif starts (bar 19) he
begins to walk forward, until, when the violin motif is finished, he
sees a man and a woman (the Duke and presumably Monterone’s
daughter). He stands still again, during the rest in the music (bar 24),
and when the trumpets play the repeated C pianissimo, he drops his
head, as if defeated, and then starts walking slowly back to the door he
came from. As the last chord resonates he is no longer there.
The TV director, André Flédérick, cleverly follows the construc-
tion. A brief opening sequence features images and noise from the hall
and the first credits until the conductor comes in; as the Prelude starts,
a long shot of the motionless palace on stage is followed by a series of
closer shots of the sleeping people, while more credits appear, and
then we return to the long shot again, where now the man is visible in
the door way; on the fortissimo of bar 15 a closer shot shows the man
full length, standing still while the last credit appears. As he starts
moving the credits are finished and the spectator has a complete view
of the action on stage. In short, the Prelude sequence is clearly divided
into two parts, one for the credits, and the other for a stage action con-

29 G. Verdi, Rigoletto, Metropolitan Orchestra, cond. J. Levine, prod. J. Dexter,

dir. K. Browning, 1977 (DVD realeased 2005, Deutsche Grammophon 073 093–
9); G. Verdi, Rigoletto, cond. R. Muti, prod. G. Déflo, dir. P. Carmine, Teatro
alla Scala/Rai Radiotelevisione Italiana, 1994; G. Verdi, Rigoletto, cond. J. Con-
lon, prod. J. Savary, dir. A. Flédérick, Opéra National de Paris/Radio France,
112 Gaia Varon

ceived in a close relation to the music. Within the brief initial se-
quence, the combination of the three elements is very effective in
conveying the sense of the live event, giving the necessary informa-
tion and introducing the staged opera, skilfully exploiting the stage
director’s invention.
A similar division of the Prelude in two parts can be detected also
in the Teatro alla Scala TV production, directed by Patrizia Carmine:
bars 1–15 are used for credits, on a black screen and with the insert of
a moving puppet, some pictures of the singers and some questionable
graphic effects. On the fortissimo at bar 15, the filming of the per-
formance commences, at first with a shot of the conductor (Riccardo
Muti), then alternating with close-ups of individual instruments and
players; the credits are not finished, though, and they keep coming in
during the film of the overture performance, which is also interrupted
by a shot of the curtain, onto which the final credits appear, until the
performance is shown again for the last chords. The result is confus-
ing, the music not really exploited either for its purely musical value
or for what it could suggest dramatically, and we are not even granted
the possibility of an attentive viewing and listening of the music per-
Using the same, or even simpler means, Kirk Browning in his
film of the Metropolitan production obtains a far more effective result:
here again, a brief opening sequence features images and noises from
the hall until the conductor (James Levine) enters and gives the cue.
The performance then stays on screen for a few bars only, followed by
a series of stage pictures that provide the background for all the cred-
its, until at the very last bars we again see the conductor and the or-
chestra. In this version though, the filming of the pictures – using
careful camera movements to make them alive and expressive – and
the editing are clearly conceived on the music, with the music, and its

30 I am not concerned here with the legitimacy and effectiveness of stage direction;
it would of course be very relevant in other perspectives. On this, see the very
interesting debate in Il Saggiatore Musicale, started with Paolo Fabbri’s ‘Di ve-
dere e non vedere’, 14 (2007), 359–67, and continued with the contributions of
Gerardo Guccini, Lorenzo Bianconi and Luca Zoppelli, ‘Ancora sulla regìa
nell’opera lirica’, 17 (2010), 83–118.
Overtures on Screen 113

segmentation is often coincident and always consistent with Pon-

nelle’s and Gallone’s productions.

In the end, given that each filming strategy is part of a specific film,
with its ideal audience and its conventions for communicating, it is
not simply a case that one category or another can best fulfil the task
of solving the visual problem of the overture. The most relevant factor
seems instead to be the skilfulness of its realization: whether it pre-
sents credits, shots of musicians or a narrative, the visual is effective
when it is constructed in close relation with the music. It may be the
simple timing of individual images with sounds, or the exploitation of
the association of musical themes or motifs with the themes and
events of the opera; what seems to be crucial is that the rhythm and
the segmentation of the visuals fit in and collaborate with the musical
Playback Problems when Filming Opera for the
Screen: Two Case Studies


Film opera is a genre in itself. As such, however, it poses several

problems, since the “marriage” between opera and film is a difficult
and complex one. The image, montage and mise en scène in movies
have their own codes, while opera has others. Although the fusion of
the two is possible, it is not always satisfactory. Firstly, the division
into acts of most operas does not always fit in with the sense of conti-
nuity in movies, and the cross-cutting in movies is difficult to repro-
duce in opera performed on a stage. Moreover, there are problems
related to filming and performance: how can we film an aria without
resorting to the sequence shot? How can we train the singer to per-
form with the naturalness of a movie actor, leaving aside habits
learned on the grandiloquent operatic stage? And, most importantly,
how can we train a singer in the credibility of lip-synch?
The aim of this paper is to show that the latter was used particu-
larly successfully in two movies produced by Gaumont for Daniel
Toscan du Plantier’s project: Don Giovanni (Joseph Losey, 1979) and
Parsifal (Hans Jürgen Syberberg, 1982). In these movies, the use of
lip-synch and real sound manage to afford a naturalness to the artifi-
cial language of operatic performance.

Preliminary Ideas

Film and opera are two words that seem a little too close to one an-
other. Or perhaps a little too distant. Close because no one would re-
116 Jaume Radigales

fute the fact that only cinema has achieved what Wagner called Ge-
samtkunstwerk, in reference to “musical drama” (a euphemism for
opera), that is, the synthesis of diverse languages in one. And distant
because, since film is an art of masses, it can be seen as distant from
the elitist connotations that have often been associated with opera.
Moreover, this view does not consider the possible analogies between
the grammar and syntax of the two arts, with their musical and iconic
equivalents that could allow us to comprehend how film and opera are
condemned to understanding one another .
Whether too close or too distant, the truth is that the marriage be-
tween film and opera is an uneasy one, though nobody questions it.
The image, montage and mise en scène have specific codes in film
and others in opera. An assemblage is possible, though not always
satisfactory. The division into acts of some operas does not always
conform to the sense of continuity we expect of film; the cross-cutting
that has become a (bad) habit of ours since Griffith imposed it in his
epic – and incidentally, very operatic – films is difficult to apply to
opera; how can we film an aria without resorting to the sequence shot?
How can we train the singer to move in front of the camera as natu-
rally as a film actor and forget habits learned on the grandiloquent
operatic stage? And most importantly, how can we train an actor in
the credibility of lip-synching without compromising the exaggerated
effort of the mask used by singers in theatre for projecting their voice
in a medium of proximity like cinema?
If opera and film have not produced more than they have to date –
though this production is not really as meagre as it may appear at first
sight – it is not for wont of trying by the two artistic languages. On the
contrary, the difficulties and big budgets of the two artistic expres-
sions have prompted too many creative setbacks.
But there are more hurdles to overcome beyond formats and
budgets: opera is an unreality because its theatrical aspect is not the
spoken word, but song. And this lie has to be merged with another: the
moving image or the illusion of movement. Film is the paradigm of

1 Luca Zoppelli, L’opera come racconto. Modi narrativi nel teatro musicale
dell’Ottocento (Venezia: Saggi Marsilio, 1994), 99–100.
Playback Problems when Filming Opera for the Screen 117

illusion , created step by step: first came movement, then sound and
finally colour. Opera was already movement, sound and colour when
the Lumière brothers patented the cinematograph. Thus, the old illu-
sion in the framework of the new illusion meant that the latter could
not be verisimilar – until Bergman, Losey and Syberberg came along
and performed the miracle. A miracle that, sadly, has been imitated
but not repeated.
This paper will attempt to show how the difficulties of lip-
synching are resolved in two exemplary films from the point of view
of their production. Firstly, Joseph Losey’s pragmatism in Don Gio-
vanni and, secondly, the aesthetic approach of Hans Jürgen Syberberg
in Parsifal (1982). Both films were commissioned by Daniel Toscan
du Plantier, formerly one of the men in charge at Gaumont, a French
production company that commissioned filmed operas from a number
of established filmmakers in the 1980s. The two films studied here
rise to the challenge without difficulty.
Filmed opera, that is, filming opera on a set or on location with
the intention to dramatise an opera outside of its natural setting (the
stage of a theatre) poses many production problems. One of them,
already mentioned, is that of lip-synching. Sometimes, filmmakers
decide to film the singers themselves, who later dub themselves for
the soundtrack, while at other times they prefer to use actors, whose
acting is more credible than that of singers, but who are less skilled in
the techniques of singing. In all cases, an acousmatisation process
occurs, according to the theories of Michel Chion, who distinguishes
the opposition between the acousmatic and the visualised for a musi-
cal score that is, in principle, conceived from an obvious off-screen
setting: the soundtrack was recorded prior to the making of the film
and there are yet no expressive or emotional intentions linked to the
images, because what is written (and what is to be heard after the
splicing process of coupling image and music) works only on the basis
of a single text (the score), far from being materialised before the
cameras. Therefore, the artifice characteristic of opera (the least credi-

2 Jean-Paul Bourre, Opéra et cinéma (Paris: Artefact, 1987)

3 Michel Chion, La audiovisión. Introducción a un análisis conjunto de la imagen
y el sonido (Barcelona: Paidós, 1993), 74–81
118 Jaume Radigales

ble art of all) contributes to the overlooking of what in another genre

or cinematic form would be seen as inevitably artificial.
The essence of operatic language, what sets it apart it from other
stage languages, is the singing. It is therefore necessary to take into
account their morphological and aesthetic characteristics when taking
it to the big screen: “Operatic singing derives its force not simply
from the extravagance of the singing voice but rather from its pointing
to the limits of vocal expression and to meaninglessness” . It is this
same singing force that is a medium in itself when it is to be assumed
or “abducted” by cinema. An example of that effect are the scenes of
the film that Jean-Pierre Ponnelle made with Le nozze di Figaro
(1975), a TV movie in which many passages were shot without the
singers lip-synching when the action required conveying introspec-
tion. If we conventionally accept that an operatic aria is equivalent to
interior monologue in spoken theatre, it is logical to think that the
Countess of Almaviva sings “Dove sono” without moving her lips, so
what we see are flashback images that refer to the happy past she lived
with her husband, who is now unfaithful to her. Or that the Count,
played by Dietrich Fischer Dieskau, is not actually singing although
we hear his voice in many of the concertanti of the second act. This,
in turn, is also acousmatisation, and if “conventional” (i.e. spoken)
films – with or without the use of playback – admit the use of voice in
off, why should we not accept it in the field of operatic music in the
context of filmed opera?
Lip-synching, therefore, responds to an aesthetic consideration, a
choice, just as Pier Paolo Pasolini and Federico Fellini chose “faces”
which were later dubbed by other people’s “voices”. What lip-
synching aims at is not so much at making something credible, but at

4 Michal Grover-Friedlander, Michal, Vocal Apparitions. The attraction of Cin-

ema to Opera. (Princeton University Press, 2005), 20
5 In this regard, we are reminded of an opera of the Catalan composer Xavier
Montsalvatge (1912–2002) entitled precisely “Una voce in off”, in which the
relationship between a woman and her dead husband is told. The woman
repeatedly listens to her beloved husband’s voice recorded on tape. Somehow,
the “technical reproducibility” – using the term favoured by Benjamin – allows
the artifice of recording a song in a performance of an artificial art like opera
singing intended to be performed live, like all stage or musical art.
Playback Problems when Filming Opera for the Screen 119

establishing a synchronous process that helps the “musivisual” under-

standing of the whole.
There is, however, another issue to consider when addressing the
problem of lip-synching in cinema: naturalness. Obviously, films try
to portrait reality, but everyone knows that cinema doesn’t have to be
in narrative form or even plausible: no one will doubt by now that
Lynch’s iconoclasm allows us to see the artificiality of cinema. Why
must lip-synching always have to work in favour of verisimilitude? As
discussed below, the options chosen as examples in this article are
based on different concepts: naturalness (Losey) and artifice (Syber-
Verisimilitude is a problem because opera singing itself is quite
far from being believable. And, as noted above, when singers or actors
appear on the screen singing (or pretending to sing), it often produces
quite a few aesthetic jolts – a theme we have already mentioned but
which should be emphasized. First, opera singers are artists who have
been trained in a singing discipline that is extremely physical because
they learn to sing with the masking technique to produce a pitched
voice that is absolutely unnatural and therefore not believable. This
moves it completely away from any desire to make believable some-
thing that is not. On the other hand, if singers themselves play the
roles before the camera, their training as actors often leaves much to
be desired. Moreover, given that opera is an art designed especially
for large theatres and performing with large stage structures, singers
tend to deliver a grandiloquent, bombastic and exaggerated perform-
ance, which is an accepted convention in theatre, but can be inappro-
priate and even ridiculous when projected onto the big screen. Losey,
who we will talk about when discussing Don Giovanni, worked hard
during the filming of the movie to convince the players to avoid over-
acting, which is what they were used to in operatic stages.
Some filmmakers have used actors who lip-synched to a recorded
soundtrack. But there is another contradiction: accustomed to the
natural emission of the voice and often ignorant of the art of opera,
performers just move their lips and create an effect of synchronisation
between what you hear and what you see, but anatomically it is not

6 Alejandro Román, El lenguaje musivisual (Madrid: Visión Libros, 2008)

120 Jaume Radigales

credible, because the technique used does not match the effort of the
opera singing technique. Once again, therefore, verisimilitude is chal-
lenged by the use of close-up shots that allow viewers to see the ac-
tors’ faces in detail, which is something that does not happen in the
opera house – rarely are viewers allowed to be that close to singers. In
any case, film does not always capture the contrived nature of opera
particularly because of the use of lip-synching. Nevertheless, some
filmmakers have used it masterfully and two examples will serve to
exemplify this. In one case (Don Giovanni) because of skillful editing
and in another (Parsifal) because of purely discursive questions and
textual coherence.

The practical use of lip-synch in Joseph Losey’s

Don Giovanni

In 1978, the American director in exile in Europe, Joseph Losey, ac-

cepted Toscan du Plantier’s challenge to make a film about Mozart’s
opera Don Giovanni. Frantz Salieri’s script located the original action
of the opera in natural settings , taking as its reference the architecture
of the villas of the Italian cinquecento architect Andrea Palladio in
Venice, Vicenza (a city of northern Italy) and its surrounding area.
To complete the task, Losey had a team of renowned opera sing-
ers, which led to some initial hiccups during pre-production and film-
ing. The singers could only afford to give Losey a very small amount
of their time since their contractual theatre commitments meant that

7 The most immediate antecedent was Ingmar Bergman’s film version (commis-
sioned by Swedish television) of The Magic Flute (1975), filmed on a set that
recreated the stage and stalls of the theatre of Drottningholm Palace, near Stock-
8 Ruggero Raimondi (Don Giovanni), Kiri Te Kanawa (Donna Elvira), Edda
Moser (Donna Anna), John Macurdy (Commendatore), Kenneth Riegel (Don
Ottavio), José Van Dam (Leporello), Teresa Berganza (Zerlina) and Malcolm
King (Masetto), with the Paris Opera Orchestra and Chorus conducted by Lorin
Playback Problems when Filming Opera for the Screen 121

they could not spend more than three weeks filming together in north-
ern Italy. Thus, Losey, who arrived with his crew several weeks be-
fore the singers, tested with cameras and a group of extras the mise en
scène that would later be filmed.
However, sound/music was another major stumbling block.
Firstly, Losey demanded that certain sound effects be added to the
soundtrack – Mozart’s entire score – in post-production, such as the
sound of the sea and various objects that appear in the props. Sec-
ondly, in order to make what was heard more verisimilar, the parts
sung in recitativo secco were sung in full voice accompanied by the
harpsichord in the diverse spaces (indoors and outdoors) where the
actions take place in the film. The harpsichordist and musical consult-
ant Janine Reiss (from the Paris Opera) encountered more than a few
problems with the continuous alterations to the harpsichord tuning and
the uncomfortable conditions in which she had to play the instrument,
in places as disparate as a platform floating on water (when she was
not on a boat), or with the cold stiffening her fingers, especially dur-
ing the night scenes. Reiss also advised Losey and the singers on the
synchrony that demanded the miming of the songs. The singers did
not know the codes of cinematographic action (far removed from the
stereotypical operatic codes) and this led to a great deal of unease and
misunderstandings between them and Losey, particularly in the begin-
The American film-maker was not happy with the final sound be-
cause the soundtrack had been recorded in the Église du Liban in Paris
and had too much resonance, which was implausible for a medium
that often had to be played outdoors.
Losey chose to shoot most of the scenes in clever sequence shots
which nonetheless did not, strictly speaking, follow the conventions of
dependence with respect to music: the scenes or shots are not cut to
coincide with changes in time; instead, the shot is sometimes cut in
anticipation of a change in the recitativo in the sung part, playing with
a naturalness (read naturalism) that gives a true – though not verismo
– element to what is seen in relation to what is heard.
The singers act, as we said, miming the sung parts (i.e. with or-
chestral accompaniment), but the recitativo secco was performed live
122 Jaume Radigales

with the problems we saw earlier of tuning and panic among some
singers that they might catch a cold .
Losey approached the editing through the gentle fusion between
the recitativo secco (direct sound) and the music, with a wisdom prov-
ing that he was able to make a virtue out of necessity and that he knew
how to visually “read” the music of opera. A good example of this is
the end of the recitativo secco passage and the beginning of the duet
between Don Giovanni and Zerlina “Là ci darem la mano” in scene
nine of Act One: Losey films a sequence shot throughout the recita-
tive, sung live. But as the singers recorded the entire opera a year be-
fore filming , Ruggero Raimondi, who plays the title role, sang the
last word of the recitativo (sposeremo) linking it in a portamento and
uniting the concluding E of the recitativo and the A note of the first
word of the duet (Là), without interruption. During filming, the direct
sound ends slightly before the end of the recitativo and the lip-synch
begins at the penultimate bar of this section (ci sposeremo). But Losey
did not cut the shot during editing, keeping it until the second bar of
the duo per se (mano) . Indeed, this extremely interesting solution is
introduced in a duet constructed around two sections, a first in 2/4
time and a second in 6/8 time. Now, in the film, the second section
does not coincide with the change of shot because by the time An-
diam, andiam mio bene! is heard, the shot has already changed – two
bars earlier, at bar 48, coinciding with the crotchet rest before the An-
diamo! sung by Zerlina.
Losey’s solution, then, contributes to the narrative logic and sense
of continuity, which would have otherwise been altered by the obvi-
ous option of cutting the shot to coincide with the change in bar.
Despite these difficulties, Don Giovanni was one of Losey’s big-
gest movies and a box-office blockbuster. It was premiered to great
acclaim among audiences and critics in Europe, though snubbed by
some in the United States. This critical discrepancy may have been
due to Losey’s fundamental approach (always an uneasy director in
the eyes of his fellow Americans) and the fact that Don Giovanni

9 Many outdoor scenes were filmed early in the morning at low temperatures.
10 As the film soundtrack but also for its subsequent record distribution.
11 We are using the Dover Publications edition of the opera score (1974).
Playback Problems when Filming Opera for the Screen 123

opened up new avenues for solving the problems posed by opera in

film. New avenues that were not always fully understood.

Lip-synch as an aesthetic choice: Syberberg’s Parsifal

In 1982, Hans Jürgen Syberberg made Parsifal about the last opera of
Richard Wagner, a composer who had obsessively pursued the Ger-
man director, who has used him in more than one of his films .
Controversial, contentious, open to numerous interpretations and
with a provocative imagery, Parsifal reveals a Syberberg obsessed
with death and Nazi Germany. In other words: with the death of Ger-
many perpetrated by the criminal regime of Hitler and his butchering
henchmen. What Syberberg attempts in Parsifal, a cryptic and com-
plex film, is to generate a critical view of Wagner and Wagnerism
based on his work, a testimony and initiating work at the same time .
In practical terms, Parsifal is a low-budget production, so se-
quence shots were filmed suited to the reels that fit in the chassis of
the film cameras of the time. Moreover, filming had to be limited to
less than thirty days due to budget constraints. And that, in the case of
such a long movie, has its merits.
However, the artistic criteria of the production should invite
viewers to linearly follow the Wagnerian concept of endless melody
that only sequence shots can truly translate .

12 These include his films on King Ludwig II of Bavaria (Ludwig – Requiem Fur
einem jungfraulichen König, 1972) and Hitler (Hitler, ein Film aus Deutschland,
1977), in addition to the splendid Winifred Wagner und die Geschichte des
Hauses Wahnfried 1914–1975, an interview of almost four hours with Wagner’s
daughter-in-law and wife of his son Siegfried.
13 Let us not forget that Parsifal is, as the score and libretto read, a “festival play
for the consecration of the stage”. For a discussion of the film, see Syberberg
14 “A visual correspondence was sought with the flow, which can be described as
‘slow’, of the music, together with a parallel between the musical ‘leitmotifs’
and a certain optical ambivalence, shying away from shot-reverse shot as an
124 Jaume Radigales

Despite the difficulties of shooting a film of these characteristics

and the impossibility of distributing such a film, considering the
length of the opera, which is included in its entirety (just over four
hours), Syberberg proposes that Wagner (limited by his eminently
theatrical language) is an artist who can be perfectly understood
through film.
Therefore, the film proposes a speculative reading of the opera,
allowing us to see Parsifal as he should really be: a boy.
The German filmmaker used actors and occasionally singers (like
Robert Lloyd, who plays Gurnemanz) and the conductor Armin Jor-
dan, who plays the role of Amfortas with the voice of Wolfgang
Schöne. For the German director, “when an actor acts over music, the
effect is completely different to that of a singer lip-synching. If the
singer is filmed, the effect is documentary, while the actor is trans-
formed into a mask” In that sense, Syberberg’s choice finds a middle
way, because it is the conductor (who is neither a singer nor a conduc-
tor) who plays the knight of the Grail mortally wounded by Klingsor’s
spear which gives rise to a symbolic reading: Amfortas lives between
two worlds (that of the Grail and that of Klingsor), just as the conduc-
tor takes the helm of a complex ship, both musically and theatrically
In the case of the Parsifal character, miming can make credible
what the stage cannot. At the same time, however, the sound clashes
head on with the visual because we see a child moving his lips to a
musical background that gives us the voice of an adult singer to listen
to, a tenor in this case. We must acknowledge that Wagner’s choice is
in itself quite unlikely since he writes the leading role of the opera for
a tenor, that is, for an adult singer, and Parsifal is a boy, a teenager
who “grows” physically and spiritually throughout the drama. You
can even say that the vocal treatment changes, a type of metamorpho-
sis as the three acts advance, from the light simplicity of the first to
the distant heroism of the third, going through the pure lyricism of the

easy and clichéd formula”. Cfr. Álvaro del Amo, ‘El Parsifal de Syberberg y la
ópera de Wagner’, La música en el cine (Filmoteca Canaria, 1989), 67
15 Jean-Paul Bourre, Opéra et cinéma (Paris : Artefact, 1987), 118
Playback Problems when Filming Opera for the Screen 125

second, especially after Kundry’s kiss and the “Amfortas! Die

Wunde!” passage.
Syberberg, we insist, chooses a boy to portray Parsifal while we
hear the voice of tenor Reiner Goldberg. While the choice here is de-
batable, even more so is Parsifal’s splitting into a girl from Kundry’s
kiss scene in the second act, keeping the tenor voice for the shots in
which Parsifal indistinctly sings in the body of the boy and the girl
who appear on the screen. This is a feminisation of Parsifal from his
split self, perhaps alluding to the possibility of reading the painful
“self-analysis” of the character as a transfiguration, into Iseult . Based
on this ethical/aesthetic premise, the result of the split self is what
Syberberg proposes from the start: the androgyny that Wagner always
thought he saw in Christ (and thus in Parsifal, an eminently Chris-
tological image), which could constitute a cosmogonic and sexless
view of all religious practice .
This feminisation of Parsifal from the split self can also be under-
stood through Jungian approaches (Parsifal’s reluctance to be seduced
and hence become a woman, as Kundry) or Freudian readings, as Paul
Coates argues: the Freudian vision would allow us to observe a taste
for masochism as a result of Kundry’s castration of Parsifal through
the kiss . There is also the practical choice that Marcia J. Citron notes
in her study of opera in film: “This feature may have been intended by
Syberberg and explains why he chose amateurs for the roles. As nov-
ices ignorant of dramatic movement and gesture, they convey con-
vincingly the idea of the pure-fool-who-lacks-knowledge that is the

16 Slavoj Zizek, Opera’s second death (New York: Routledge, 2002), 51

17 According to Jean-Jacques Nattiez, “for the religion of the future, a sexless
Parsifal preaches the renunciation of desire in favour of the intermediary of a
work of art in which, at the last moment, all forms of sexual and racial distinc-
tion are abolished”. Cited in Serge Gut, “Parsifal, dramme païen ou dramme
chrétien?”, L’Avant Scène Opéra 213 (2003), 108–121: 119. Slavoj Zizek finds
analogies with Tristan und Isolde in this split self, in the light of the inherent
“negation” in the Liebestod of this opera, which can be read as an anticipation of
Parsifal’s conversion. Slavoj Zizek, Opera’s second death (New York: Rout-
ledge, 2002), 51
18 Paul Coates, The Gorgon’s Gaze. German Cinema, Expresionism, and the Im-
age of Horror (Cambridge University Press: 1991), 128
126 Jaume Radigales

essence of Parsifal. Appropriately this quality is more apparent in

Parsifal I, who depicts the Parsifal of the imaginary stage” . In any
event, Syberberg chose the split-self solution, at the risk of undermin-
ing credibility, based on the distancing or Verfremdungseffekt of
Brechtian theatre . Throughout the film, the director reflects on the
irrational nature of German romanticism, which Wagner set to music
and (again according to the theses of the director), produced Hitler,
defined by Syberberg as the bitter flower of that romanticism. Only
through distancing can we become aware of this.
Either way, Syberberg clearly defies the naturalism of film and
enters the realm of the illusory, challenging the credibility of lip-synch
through the splitting of a character whose singing has nothing to do
with what we are seeing. An aesthetic choice, then, that is justified by
the militant distancing of the German film-maker in relation to one of
the most complex scenes of an opera, Parsifal, made equally complex
by its ambivalence.


Lip-synching can be seen as a “lesser evil” in solving the problems of

filmed opera. Its credibility hinges on the exaggerated vocalisation of
the singers or the lack of facial movement when what we are hearing
does not reflect the frugality of the expression. Opera is an art of dis-
tance while cinema is an art of proximity, and close-ups can betray the
credibility of what we are hearing in relation to what we are seeing. In
their films on operas by Mozart and Wagner, Losey and Syberberg,
respectively, put forward two possible models, one based on a narra-
tive conception of editing and the other on an ethical/aesthetic criteria
relating to the essence of musical drama. Losey’s naturalism is one

19 Marcia Citron, Opera on Screen (New Haven & London: Tale University Press
New Haven & London, 2000), 153–154
20 Jeremy Tambling, Opera, Ideology and Film (New York: St. Martin Press,
1987), 194–211
Playback Problems when Filming Opera for the Screen 127

option, contrasting with the distancing effect of the illusory nature of

Syberberg’s film. All this serves only to introduce possible options for
successfully keeping opera in film as two artistic languages that mutu-
ally reinforce each other with certain achievements worthy of being
remembered and recalled.
The Opera Director’s Voice: DVD ‘Extras’ and the
Question of Authority


The first decade of the twenty-first century has seen not only an im-
pressive rise in opera available on DVD, but also increased attention
on the part of production companies to the bonus features often in-
cluded with new releases. The concept of the deleted scene, standard
in the case of film extras, is not a feature of opera DVDs, but inter-
views with directors, conductors and cast members are common. The
film critic Mark Kermode has questioned the value of interviews in-
cluded with DVD releases, remarking that ‘it is very rare that any
major film company promoting a recent product in which it has a
creative stake will allow anything other than the most fawning, ano-
dyne promotional material […] to appear on the DVD’. While the
same is arguably true in the case of opera DVDs, which also follow a
commercial imperative, this essay will argue that opera ‘extras’ re-
ward close attention. At the very least, the choices involved in the
inclusion of certain material can be highly revealing, demonstrating
not only the practicalities of opera DVD production, but also some of
the values surrounding opera that production companies cheerfully
This essay will focus in particular on one type of bonus feature:
interviews with and commentaries by stage directors. Undoubtedly,
these can provide valuable perspectives on the interpretative strategies
behind the productions in question, but they inevitably guide – and
perhaps even narrow – reception and interpretation of the main fea-
ture. It is an instructive process to watch several of these director in-
terviews back-to-back, because in this context, certain patterns are

1 Mark Kermode, ‘The Recalcitrant Interviewee’, Cinema Journal 47: 2 (2008)

135–141: 138.
130 Áine Sheil

liable to emerge. Those patterns will be explored throughout this es-

say, as will the questions of authority I believe lie behind them. Two
specific bonus features will serve as in-depth case studies: Calixto
Bieito’s commentary on his production of Wozzeck at the Gran Teatre
del Liceu, Barcelona (Opus Arte, OA 0985D: recorded 2006, released
on DVD 2007), and an interview with Peter Brook on his production
of Don Giovanni at the Aix-en-Provence Festival (Bel Air Classiques,
BAC 210: recorded 2002; released on DVD 2006). Interviews with
Francesca Zambello (Don Giovanni, OA 1009D, 2009), Peter Hall (La
Cenerentola, OA 0944D, 2006) and Nicholas Hytner (Così fan tutte,
OA 0970D, 2007) will also be examined. The self-denying aesthetic
(to borrow a phrase from American musicologist Richard Taruskin)
common to these interviews sits uneasily with popular ideas about
directorial self-indulgence, even in the case of Bieito’s production,
with its highly visible and controversial directorial interventions. As
this essay will demonstrate, 21 -century opera directors are as likely
as ever to trace their authority to the intentions of dead composers. As
a consequence, the opera DVD extra has the effect of upholding the
perceived sanctity of authorial intentions, even while it celebrates and
bestows a particular aura on the contemporary artists that are vital to
the DVD production industry.

Calixto Bieito’s Wozzeck, Gran Teatre del Liceu, Barcelona

Dubbed ‘the most offensive director working in Europe today’ by one

American journalist, Calixto Bieito is known and often criticized for
his transpositions of opera plots to violent contemporary dystopias.
Born in 1963, the Spanish director started his career in theatre rather
than opera; he remains active in the field of spoken theatre, including
in his role as the artistic director of Barcelona’s Teatre Romea. He is

2 Heather Mac Donald, ‘The Abduction of Opera’, City Journal Summer 2007
 (14 March
The Opera Director’s Voice 131

particularly noted for contemporary settings of Spanish and European

canonical plays, and for his dramaturgical reworkings of Shakespeare.
Positively disposed theatre critics and scholars have praised his pro-
ductions for their attention to detail, convincing contemporary stage
worlds, unflagging tempi and critical content. According to Maria

Bieito’s work shares profound parallels with that of a number of other radical di-
rectors, such as Planchon and Chéreau, who have re-viewed some of the seminal
works of the Western canon through the prism of a decaying society where
moral indignation is tempered by a spirit of disquieting enquiry.

Bieito began to work in opera in 1999, when he staged Carmen for a

Catalan arts festival and Il mondo della luna for Opera Zuid in Maas-
tricht. Despite his continued work in spoken theatre, he is now re-
garded predominantly as an opera director, and has worked inten-
sively in Germany, including at the Staatsoper Stuttgart, Komische
Oper Berlin, and Frankfurt Opera, as well as in Spain, the UK and
other European countries. Several of his productions have generated
particular controversy, including Un ballo in maschera (Gran Teatre
del Liceu, Barcelona, 2000; Royal Danish Opera, Copenhagen, 2001;
English National Opera, London, 2002), which opened with chorus
members sitting on a row of toilets, trousers down, and Die Ent-
führung aus dem Serail (Komische Oper Berlin, f. pr. 2004), which
was set in an ultra-violent brothel and contained graphic murder
scenes. Condemnation of Bieito’s Entführung aus dem Serail was
widespread in the German and international media, and unsurpris-
ingly, perhaps, it also spread to the academic sphere. In ‘A Season in
Berlin, or, Operatic Responsibility’ (2005), Michael P. Steinberg put
forward a typical argument that the production reduced the work, in
the process failing its composer:

Bieito’s production failed – and certainly failed Mozart, whom it was not con-
cerned to validate – by confusing political realism with literalness. No hidden
dimension of potential of the work emerged; rather, the work had to be eviscer-

3 Maria Delgado, ‘Calixto Bieito’, in Fifty Key Theatre Directors, ed. Shomit
Mitter and Maria Shevtsova (London, New York: Routledge, 2005), 262–6: 263.
132 Áine Sheil

ated along with some of the poor supernumeraries. No forgiveness, justice, or

enlightenment can exist in such a setting, so they didn’t.4

Bieito’s staging of Wozzeck has not gained the notoriety of his Ent-
führung aus dem Serail and is certainly not among his more infamous
productions, but it nonetheless bears many of the hallmarks of his
style. Set in an industrial world rather than the military setting of the
original play and opera, the designs constitute a massive tangle of
piping reminiscent both of human intestines and alienating, destruc-
tive technology. In the DVD commentary, Bieito explains that one of
his inspirations was Fritz Lang’s silent film Metropolis (1927), which
explores the dehumanizing effects of industrialization and boasts fu-
turistic cityscapes and industrial scenes that are still impressive more
than eighty years on. Bieito is well known for paying homage to clas-
sic films in his opera stagings, and in his commentary he also men-
tions the influence of another film on his staging, Terry Gilliam’s
futuristic fantasy, Brazil (1985).
In Wozzeck, dystopian images of technology and its consequences
abound: Wozzeck is an oppressed worker in what appears to be an oil
refinery, and his child is bald, apparently ill and in constant need of a
medical ventilator, probably because of the harmful surroundings.
Their home is a bare, harshly lit shipping container that offers no com-
fort apart from the forlorn plants that Wozzeck carefully tends. In the
scenes in which the doctor is present, corpses are piled up with aban-
don, and in the DVD recording, the camera lingers on the blood and
guts. The number of dead bodies and the doctor’s unhealthy attraction
towards them reinforces the idea of toxicity: this is a world in which
death is as common as life. The Drum Major is an aging, sleazy rock
star, whose social advantage over Marie (a refinery worker, like
Wozzeck) is signalled through his initial appearance on a raised

4 Michael P. Steinberg, ‘A Season in Berlin, or, Operatic Responsibility’, New

German Critique 95 (2005), 51–66: 61. For a more detailed and less negatively
disposed academic treatment of the production see Clemens Risi, ‘“Martern Al-
ler Arten”: Calixto Bieitos Suche nach der Wahrheit des Musiktheaters’, in
Realistisches Musiktheater. Walter Felsenstein: Geschichte, Erben, Gegenpo-
sitionen, ed. Werner Hintze, Clemens Risi and Robert Sollich (Berlin: Verlag
Theater der Zeit, 2008), 132–47.
The Opera Director’s Voice 133

walkway. During the final instrumental interlude in Act III, dozens of

naked actors advance slowly through the piping, as if to emphasize the
vulnerability of humankind in the face of crass industrialization.
Above them, a video projection shows an oil-contaminated bird flap-
ping its polluted wings desperately. If opera is all about excess, then
the sheer misery of this production is highly operatic. In the DVD
commentary, Bieito draws attention to his own style of direction, de-
scribing it as ‘direct, shocking and very provocative, in the best sense
of the word’.
The first shot of Bieito’s commentary shows the director’s eyes
only – a powerful gesture that has the effect of suggesting vision and
the capacity to look into and penetrate Berg’s work. Bieito begins his
commentary with Berg and what he describes as the composer’s inter-
est in human exploitation. According to the director, this concept of
human exploitation cannot be conveyed adequately with a contempo-
rary military setting, because most modern armies no longer use con-
scripts. He decided instead on a bleak industrial environment, which
would represent the exploitation of a harsh and unyielding capitalist
system and the ‘corruption of the human soul’. He is at pains to con-
nect his staging with the idea of an originating work, remarking that ‘I
thought, and still think, that I kept very close to the work’s original

The ‘work concept’

Bieito’s idea that Wozzeck has an original essence depends on the so-
called ‘work concept’, which has had a vexed history throughout the
humanities since the rise of poststructuralism and postmodernism.
Whether works can reach us as their authors intended, or whether the
word ‘work’ is, in fact, an appropriate label for something that may
not originally have been understood as such are issues that have led to
trenchant debate within musicology, a discipline that until recently
was overwhelmingly invested in written forms of ‘great works’. Often
134 Áine Sheil

the question of faithful interpretation is at stake, and as Taruskin has

reasonably pointed out, fidelity to an ‘objectified musical work-thing’
is impossible without a notion of the reified Werk. Critics are not
entirely agreed on when the idea of a musical work came into being,
but many link the work concept with romanticism. In her book The
Imaginary Museum of Musical Works: An Essay in the Philosophy of
Music (f.p. 1994), Lydia Goehr includes a telling quotation from the
German writer E.T.A. Hoffmann (1776–1822):

The genuine artist lives only for the work, which he understands as the com-
poser understood it and which he now performs. He does not make his personal-
ity count in any way. All his thoughts and actions are directed towards bringing
into being all the wonderful, enchanting pictures and impressions the composer
sealed in his work with magical power.

These words were written at a time when the idea of the interpretative
performer (as opposed to the performer-composer) was gaining
ground. According to musicologists such as John Butt and Bruce
Haynes, the division of music-making into creative and re-creative
roles led to the idea of Werktreue – fidelity to the work – as well as to
an ideal of transparent, selfless interpretation.

5 Richard Taruskin, Text and Act: Essays on Music and Performance (New York,
Oxford: Oxford University Press, 1995), 10.
6 ‘Johannes Kreislers, des Kapellmeisters, musikalische Leiden’, tr. as ‘Of Ka-
pellmeister Johannes Kreisler’s Musical Sorrows’ by R. Murray Schafer in
E.T.A. Hoffmann and Music (Toronto: University of Toronto Press, 1975).
Quoted in Lydia Goehr, The Imaginary Museum of Musical Works: An Essay in
the Philosophy of Music, rev. ed. (Oxford: Oxford University Press, 2007), 1.
7 John Butt, Playing with History: The Historical Approach to Musical Perfor-
mance (Cambridge, New York: Cambridge University Press, 2002); Bruce
Haynes, The End of Early Music: A Period Performer’s History of Music (Ox-
ford: Oxford University Press, 2007). For a more complex exploration of the
understanding of the interpretative performer in the early nineteenth century, see
Mary Hunter, ‘“To Play as if from the Soul of the Composer”: The Idea of the
Performer in Early Romantic Aesthetics’, Journal of the American Musicologi-
cal Society 58: 2 (2005), 357–398. Hunter argues that while the ideal of transpa-
rent interpretation did indeed exist, there was an alternative discourse surround-
ing performance, in which interpretative performance was associated with
genius and powers of creation equal to those of composers.
The Opera Director’s Voice 135

Bieito’s take on the work concept is flexible – he certainly does

not claim to erase his own personality or directorial style in a show of
Hoffmannesque servitude – but the concept nonetheless allows him to
present his staging as a mediation of Berg’s ideas. He suggests that the
essence of the opera not only exists, but that it goes beyond details
like time and place. His transposition makes sense of the work in the
here and now, but is just one possibility for accessing its timelessness.
As he remarks: ‘Replacing soldiers with[...] That’s a superficial detail.
Deep down it’s just the same. I’ve kept the essence.’ So on the one
hand, he invites his audience to consider the industrial setting as a
necessary mediation tactic that keeps intact Berg’s interest in human
exploitation, and on the other hand, he waves this transposition aside
as superficial. Below the surface, he implies, the timeless work is
honoured and conveyed. At no point does he entertain the possibility
that his production is the work, even temporarily in the theatre, or as
the viewer watches it on DVD. The emphasis is on serving Berg. An
odd thing emerges here: Bieito and his many critics all appear to ac-
cept the work concept, and in that sense they occupy one continuum.
But for Bieito, the work concept authorizes extreme interpretation,
while for his critics, the idea of the ‘work’ is the idea of conservation.
The German art historian Wolfgang Ullrich has written in illumi-
nating terms of the work concept and the relationship between ‘radi-
cal’ opera directors and those who criticize the productions of these
directors as self-indulgent and misguided. Directors would not con-
tinue to stage works of the canon, Ullrich argues, unless they believed
that the works in question are relevant to the present. They devise
ambitious interpretations precisely out of respect and deference, as
well as an Adornoesque belief in a work’s autonomous ability to out-
live the composer. Opponents of these directors believe that the time-
less essence of a work is static, Ullrich suggests, and do not demand
that the work be all things to all times. Their work concept has noth-
ing to do with protean sublimity and is therefore much more modest,
but it nonetheless serves as a basis to attack interpretative work that is
also heavily invested in the ‘work’.

8 Wolfgang Ullrich, ‘“Die Kunst ist Ausdruck ihrer Zeit”: Genese und Proble-
matik eines Topos der Kunsttheorie’, in Angst vor der Zerstörung: Der Meister
136 Áine Sheil

Bieito is a good example of an opera director who believes not

only in contemporary settings, but also in the enduring relevance of
the work, and in the possibility of determining that relevance through
studying written artefacts (the score). The London theatre critic Mi-
chael Billington, who interviewed Bieito during rehearsals for the
director’s Don Giovanni at English National Opera in 2001, notes that
Bieito had a copy of the score of Don Giovanni with him and made
‘constant’ reference to it during the interview. Billington suggests that

however controversial Bieito’s version [of the opera] may prove, no one can
deny that it’s based on close attention to the music and a clear-sighted view of
character. ‘If you follow the score’, says Bieito, ‘it tells you exactly where the
work changes from comedy to tragedy’.

Elsewhere in the interview, Bieito remarks: ‘I start with the score and
I listen to the music to find the soul of the piece.’ So while the direc-
tor does not lay claim to any acts of self-effacement, his emphasis on
establishing ‘the soul of the piece’ is curiously reminiscent of the
E.T.A. Hoffmann passage quoted above.
Bieito’s staging of Wozzeck could at certain points be said to ig-
nore Berg’s score: characters such as the Drum Major appear before
‘their’ music begins, and in that sense, the director muddies the exacti-
tude of Berg’s musical writing. (Indeed Berg’s opera is musically and
dramatically so tightly and symmetrically constructed that it is diffi-
cult to imagine an opera more deserving of the label ‘work’.) Of
course, Bieito’s interpretation of Wozzeck does not necessarily set out
to undermine the integrity of Berg’s score. The moments that deviate
from the composer’s stage instructions and his musical semiotics do,
however, illustrate a gap between the rhetoric of fidelity and the real-
ity of the staging process: Bieito can choose to blur the boundaries
between Berg’s scenes precisely because he is invested with interpre-

Künste zwischen Archiv und Erneuerung, ed. Robert Sollich, Clemens Risi,
Sebastian Reus and Stephan Jöris (Berlin: Verlag Theater der Zeit, 2008), 233–
9 Michael Billington, ‘Sex, Booze, Drugs and Mozart’, in The Guardian (30 May
2001), 14–15: 14.
10 Ibid.
The Opera Director’s Voice 137

tative authority. As Goehr has pointed out, the idea of Werktreue ‘de-
mands submission to the work just as it displaces authority onto those
(conductors and performers) who claim to be most submissive’.
Bieito may profess a fidelity to ‘the essence’ of Wozzeck, but in doing
so, he reserves the right to realize that essence according to his own
vision, which almost inevitably does not coincide entirely with the
score. Instead, what emerges – not least because of the testimony of
the DVD commentary – is paradoxical submission: an interpretative
process that involves a simultaneous exercise of deference and author-

Peter Brook’s Don Giovanni, Aix-en-Provence Festival

Unlike Bieito, Peter Brook is not primarily known for his work in
opera. Instead, he has been a prominent figure within spoken theatre
since the 1950s, and has worked on an eclectic mix of international
projects over the past half-century, including a lengthy and multi-
stylistic presentation of the Indian epic La Mahabharata (1985).
Brook’s very early career included a period as Director of Productions
at the Royal Opera House, London (1947–50), but this was evidently a
period of frustration, during which Brook clashed frequently with
Music Director Karl Rankl. Apart from two productions during the
1950s for the Metropolitan Opera, New York, the only other operas
that Brook directed before Don Giovanni were specially edited and
reduced versions of Carmen (1981) and Pelléas et Mélisande (1992).
In the case of La Tragédie de Carmen, Bizet’s score was stripped
down, and the re-ordered scenes were rearranged for small ensemble
by the composer Marius Constant. Brook explained his willingness to
re-write the opera as an attempt to reveal the austerity of Mérimée’s
novella Carmen, on which the opera is based, and as a means of shed-

11 Goehr, The Imaginary Museum of Musical Works, xli.

138 Áine Sheil

ding the trappings of nineteenth-century ‘big opera’. This flexibility
with the opera is characteristic of a director who does not believe in
rigid Werktreue: as Brook famously argues in The Empty Space (f.p.

when I hear a director speaking glibly of serving the author, of letting a play
speak for itself, my suspicions are aroused […] If you just let a play speak, it
may not make a sound. If what you want is for the play to be heard, then you
must conjure its sound from it.

Brook’s production of Don Giovanni for the Aix-en-Provence Festival

was first seen in 1998, during the fiftieth-anniversary year of the festi-
val. Superficially, it signalled a return to conventional opera produc-
tion for the director, but his involvement in the project came with
considerable strings attached. Everyone in the opera was expected to
sign on for a year, so that when the festival closed, the production
went on tour with the same cast, the same conductor and the same
orchestra. There were no star singers, and the conductor was the very
young Daniel Harding. In this way, Brook aimed to circumvent the
difficulties with opera institutions and their inadequate rehearsal times
that he had encountered half a century earlier at Covent Garden. He
notes in the DVD interview that ‘to feel […] you are forced by the
external conditions of opera world life to do worse work than you
would be doing in a normal theatre is a good reason for not doing
opera’. Although he concedes in the same interview that ‘opera has
[now] been freed from all the horrors of opera tradition’, he condemns
what he describes as ‘a new, even more horrific director’s tradition:
[…] jokes at all costs’. This development is lamentably reductive, he
feels, particularly in the case of Don Giovanni, which he describes as
‘a great work’ containing many different levels. Nevertheless, he
draws attention to Da Ponte’s label ‘dramma giocoso’ and describes

12 Margaret Croyden (ed.), Conversations With Peter Brook: 1970–2000 (New

York, London: Faber and Faber, 2003), 191.
13 Peter Brook, The Empty Space (London: Penguin Books, 1990), 43.
14 Musical direction of the first run in Aix-en-Provence in 1998 was shared be-
tween Claudio Abbado and Daniel Harding. Harding subsequently went on tour
with the production, and he conducted the performance in 2002 that was re-
leased on DVD.
The Opera Director’s Voice 139

this as the key to the opera, which he sees as a ‘joyous drama’ rather
than an epic tragedy.
In the production, Brook seeks to realize the different layers of
the opera above all through the interaction between the characters.
The setting is abstract: a shallow platform, open at the sides, serves as
an acting area, and when the cast members are not required in a scene
they can sometimes be seen sitting in the shadows just beyond this
area. Simple, brightly coloured items serve as tables, chairs, benches,
sticks, swords and gravestones; they resemble humble rehearsal props,
and are typical of the pared-back approach that Brook has adopted at
various points in his career. The cast members appear in contemporary
clothing of varying degrees of formality, with the aristocratic charac-
ters in evening dress that would not look out of place on a concert
platform (or in an opera auditorium). In this way, Brook focuses unre-
lentingly on the acting, which is as naturalist as the setting is abstract.
This mixture of naturalism and abstraction culminates in the ending,
which is shorn of all melodrama and representations of the afterlife.
The Commendatore appears as he did in life, and the only other sign
of the metaphysical is an abrupt change to a dark lighting scheme. But
after the other characters come in search of Don Giovanni, only to
find him dead/vanished, an innovation takes place: Brook brings both
Don Giovanni and the Commendatore back on stage, and the two dead
characters regard the living, unnoticed. In particular, Don Giovanni
goes from one character to another, looking at them searchingly and
with apparent tenderness, until he takes a seat for the final sextet. As
Donna Anna, Don Ottavio, Donna Elvira, Leporello, Zerlina and
Masetto sing about the fate of evil-doers, a tear is seen running down
his cheek.
In the fifteen-minute DVD interview, Brook speaks at length
about his characterization of Don Giovanni. One of his first arguments
is that Mozart identified with the character and felt very close to him.
‘Mozart himself wasn’t a little saint’, the director remarks. ‘He was
somebody bubbling over with excitement and also deeply in all the
different parts of life, including sex. This was natural to him.’ Brook
argues that ‘there isn’t one note in all [Don Giovanni’s] music that
criticizes the character’, and contends that the composer bestowed a
‘fundamental innocence’ and ‘excessive joie de vivre’ on the character
140 Áine Sheil

through his music. This emphatic statement that there is not one criti-
cal note in Don Giovanni’s music is curious. Many critics have re-
marked on the mercurial nature of the character and the difficulty of
understanding him, given the chameleon-like, non-reflective quality of
his music. Writing about the opening of the opera, Wye Jamison Al-
lanbrook argues that Don Giovanni

conceals himself in his music, adopting for his first utterances Donna Anna’s
vocal line, and never in the remainder of the trio […] originating any of its
rhythmic or melodic material […] Chameleonlike, he doesn’t even betray him-
self in speech, but borrows Donna Anna’s music and a combination of
Leporello’s and Donna Anna’s words.

Julian Rushton argues in a similar manner that Don Giovanni

has no self-reflective aria – he never sings about himself, as Mozart’s other cen-
tral characters do. We have no sense of what he is like when he is by himself.
He is presented always in action – the action, notoriously, of a seducer.

By this reasoning, the lack of criticism Brook perceives is due to a

lack of positive information and characterization in the music, and not
to identification with the character on the composer’s part. If, how-
ever, one accepts Brook’s basic point that there is nothing tortured
about Don Giovanni’s music, and that it does indeed suggest exces-
sive joie de vivre, then one also has to examine the music of the other
characters, and consider how it reflects on the character of Don Gio-
vanni. Certain arias of Donna Anna and Donna Elvira (‘Ah chi mi
dice mai’ or ‘Or sai chi l’onore’, for example) arguably convey pain,
and this pain could be said to reflect on Don Giovanni’s actions.
Brook argues with apparent conviction, however, that Don Gio-
vanni is pardoned, not by God, but by Mozart, through his music. So
although he says that he rejects the German Romantic tradition of
turning Don Giovanni into a tragedy, he nonetheless helps himself to
Romantic ideology by ascribing godlike powers to Mozart (the crea-

15 Wye Jamison Allanbrook, Rhythmic Gesture in Mozart: Le nozze di Figaro and

Don Giovanni (Chicago: University of Chicago Press, 1983), 208.
16 Julian Rushton, W.A. Mozart: Don Giovanni (Cambridge: Cambridge University
Press, 1981), 82.
The Opera Director’s Voice 141

tor) and his music. And in turn, Brook claims for himself power over
Mozart’s characters’ destinies. He notes later in the interview that the
music at the end ‘has something elegiac about it’ and for this reason,
his production brings back Don Giovanni after his death. Brook’s
stated view that Don Giovanni is ‘beyond judgement’ is thus made
manifest in the staging: there is no hell, rather the suggestion of what
the director describes as fundamental innocence. As Brook argues, the
character has touched all those around him, and they ‘retain something
not entirely negative from the experience each one had with Don Gio-
vanni’. That this is personal interpretation is hidden by the type of
‘paradoxical submission’ mentioned earlier: Brook claims to find in
the music alone Mozart’s understanding of his character, and, armed
with this knowledge, he aspires to peel back layers of corrupt per-
formance practice and reveal the work itself. In this way, the director
minimizes his own work and portrays himself as a transparent me-
dium for the composer’s genius, in spite of his professed wariness of
Werktreue noted above.
Brook’s apparent submission to Mozart’s greatness is, however,
paradoxical, because his staging of Don Giovanni inevitably involved
an exercise of considerable authority. Indeed, Brook enjoyed more
authority in this particular production process than most directors can
dream of: the lengthy rehearsal periods and commitments on which he
insisted may have been ‘almost unique in opera’, as he claims. In The
End of Early Music: A Period Performer’s History of Music (2007),
Bruce Haynes draws an interesting link between the custom of re-
hearsing and the advent in the late eighteenth century of the com-
poser-intention ideology. Even now, rehearsals are generally seen as
a means of doing justice to works and their creators, but they are also
a means of doing justice to a director’s or conductor’s vision. The
more tightly an ensemble works together, the better this reflects on
musical and stage directors. In conventional opera conditions, indi-
vidual singers can shine musically, since they often know their role
from dozens of other productions. But star singers rarely submit to
lengthy rehearsal periods or contribute to the cohesion of a cast, and

17 Haynes, The End of Early Music, 100.

142 Áine Sheil

are therefore of little interest to a director committed to working

gradually towards even ensemble results.
That Brook was able to subvert normal opera practice with his
production of Don Giovanni speaks of authority and the credit that
accompanies a long and illustrious career. One aspect of the produc-
tion remained typical of conventional opera, however: the musicians
and conductor performed from an orchestral pit, a practice Brook has
condemned elsewhere:

To me, in the twentieth century, there is something politically unacceptable in

the idea of musicians being stuck underground, musicians who are the co-
producers of a certain performance, who are equal artists, and yet apart from the
conductor; nothing is a more rigid totalitarian approach to theatre than the con-
ductor stuck up on a podium […] For me, the opera-house construction is a rigid
image of the nineteenth-century way of seeing the social system – the servants

Brook may well have brought his belief in co-production and equal
artistry to his production of Don Giovanni, but the DVD interview has
the effect of reinforcing the impression of his authority. The director’s
commentary on the production is the sole bonus feature on the disc:
there is no interview with Daniel Harding or any of the cast, and the
interviewer does not appear on camera. Instead, the questions are seen
as inter-titles, rather in the manner of a silent film, and Brook’s is the
only voice to be heard. In allying himself with Mozart’s intentions and
meanings, Brook underlines the idea of a single line of interpretation
and authority. In this interview, he performs the role of a prophet
within that line of succession.

18 Croyden (ed.), Conversations With Peter Brook, 204. The interview from which
this quotation is taken was conducted in Paris in 1983 in the wake of Brook’s
Tragédie de Carmen.
The Opera Director’s Voice 143

‘The Field of Cultural Production’

Drawing on the language and ideas of sociologist Pierre Bourdieu, it

could also be argued that Brook is allocated the role of ‘consecrated
artist’ in the DVD interview. The resemblance of his scarf to a liturgi-
cal vestment is apposite: it confirms the link with canon and consecra-
tion, and reminds viewers that Brook’s status as eminent theatre-
maker allows him in turn to consecrate the opera Don Giovanni. This
interview is a classic Bourdieu-style ‘position taking’: Brook con-
demns past interpretations of Don Giovanni and what he sees as a new
‘jokes at all costs’ directorial trend. He confirms his clout within the
field of cultural production with his successful challenge to conven-
tional opera practice. In other words, his refusal to work according to
the normal patterns of opera houses appears to turn complacent bour-
geois art into something far removed from potboilers and money-
making. As Bourdieu argues, the field of cultural production special-
izes in the disavowal of economics. It disguises its dependence on
competitive capitalism through an ideology of charisma bound up
with the idea of consecrated works, artists and positions:

The ‘charismatic’ ideology[...] is the ultimate basis of belief in the value of a

work of art and[...] therefore the basis of functioning of the field of production
and circulation of cultural commodities[...] It is this ideology which directs at-
tention to the apparent producer, the painter, writer or composer, in short, the
‘author’, suppressing the question of what authorizes the author, what creates
the authority with which authors authorize.

In ‘The Field of Cultural Production, or: The Economic World Re-

versed’, Bourdieu depicts the position taking of artists as a relational
and almost mechanical matter, and indeed it is unnecessary to view
Brook’s commentary as a conscious appropriation of authority in or-
der to ask what lies behind the style and content of the interview. The

19 Pierre Bourdieu, ‘The Production of Belief: Contribution to an Economy of

Symbolic Goods’, trans. Richard Nice, in Bourdieu, The Field of Cultural Pro-
duction. Essays on Art and Literature, ed. and introd. Randal Johnson (Cam-
bridge: Polity Press, 1993), 74–111: 76.
144 Áine Sheil

link between the consecration of composers and directors and underly-

ing (but camouflaged) market mechanisms does not depend on indi-
vidual intentionality, as Bourdieu has demonstrated.
Within opera, one way of minimizing the idea of economics and
competition is to stress the artistic pedigree of the work on offer. A
good example of the tending and cultivation of artistic reputations is
captured on a Royal Opera House recording of Don Giovanni released
in 2009 (OA1009D). In a bonus feature on the disc, the director Fran-
cesca Zambello describes the opera as ‘a masterpiece, a genius piece
of work’ and states that the score is ‘our bible[...] our roadmap’. Her
interviewer is none other than the Music Director of the Royal Opera
House, Antonio Pappano (who, as might be expected, is allowed to
appear on camera). Zambello remarks that when she works on a musi-
cal she can shape it and ask the composer and lyricist to make
changes. But in opera, the text appears to her as a pre-given – as a
sacred text, in fact. Authority therefore depends not on bending a liv-
ing composer’s will to her own, but on appearing to have privileged
access to historical authorial intentions. ‘Everyone out there thinks
they know Don Giovanni’, she says, but her job is to ‘find an es-
sence’. Zambello and Pappano talk about collaboration, and Zambello
says that in opera the conductor and stage director ‘have to be in har-
mony’. Her comment implicitly acknowledges the potential for power
struggles within the opera hierarchy – a potential that explains why
bonus features on opera DVDs generally feature conductors as well as
directors, and why conductors are often given the first word. But when
conductors and directors are in agreement on the genius of a dead
composer, consecration of the past tends to obscure claims to author-
ity in the present.
Displays of reverence, not just for dead composers and their mu-
sic, but more specifically for written sources, are common among
directors: as Peter Hall says on a Glyndebourne recording of La
Cenerentola (OA0944D), ‘you have to know the score backwards.
You have to know what comes next in every bar[...]’. In a bonus fea-
ture on a recording of Così fan tutte (OA0970D), Nicholas Hytner is
filmed in the library at Glyndebourne, an authorizing canon of works
represented physically in the background.
The Opera Director’s Voice 145

It is this type of visual detail that marks out DVD bonus features
from interviews in written form. In theory, a print article can contain
very similar sentiments to those conveyed on DVD: in interviews
published in book form and on the internet, the German director Peter
Konwitschny has, for example, depicted himself as a ‘mediator’ of
opera, and has remarked that his stagings ‘aim to return to the roots:
to get to the core of the pieces, through the jungle of interpretative
traditions, which in most cases, have distorted the pieces’. But when
one considers the semiotics of the Glyndebourne library setting, or the
effect of the camera zooming in on Bieito’s eyes, or the fact that in-
terviewers are usually relegated to an off-camera position, thus ex-
tending the familiar chain of hierarchy and submission, it becomes
clear that DVD technology brings powerful new dimensions to the
dissemination of directors’ voices. The DVD bonus feature also al-
lows for proximity between the director and his or her production.
There is no disjunction between stage and page, and the viewer can go
straight from production to commentary or vice versa. He or she can
relate a director’s comments to a production that the director worked
on, not a revival staged years later by a staff director.
At this point it is worth emphasizing that the impetus for director
interviews generally lies with DVD production companies, and not
with artists: the bonus feature was not invented by directors conspir-
ing to accumulate authority. Instead, DVD extras are created by pro-
ducers who are responsible for assembling a product. Very often those
who carry out the interviews are the DVD producers themselves, since
they are close to the production and know what the director is doing,
and their time does not represent additional expense for the production
company. But bonus features do involve increased expenditure, so
why do production companies invest in them? Opus Arte, one of the

20 See ‘“Ich betrachte mich als Vermittler”: Peter Konwitschny im Gespräch mit
Udo Bermbach über Richard Wagner und das Regietheater’, in Regietheater, ed.
Gerhard R. Koch, Wagnerspectrum 2 (2005), 177–97.
21 ‘Peter Konwitschny: I do not consider myself a representative of the Regietheater’,
interview with Per-Erik Skramstad and Mostly Opera http://www.wagneropera.
net/Interviews/Peter-Konwitschny-Interview-2009.htm (1 November 2010).
146 Áine Sheil

big players in the field, offered this explanation in the previous incar-
nation of its website:

The theme and content of each Opus Arte title is created as a crown of jewels,
which make the discs ‘cultural treasures’ with a limitless life. All our discs allow
the viewer/listener to explore a subject from several angles in an individual and
exciting way. Our philosophy is that experiencing our DVDs should satisfy the
eye, ear, mind and heart in equal measure.

This suggests that at a purely practical level, the bonus feature is a

means of creating brand identity and potentially increasing shelf-life.
In the process, the disc denies its functionality, and seeks the status of
timeless art for itself, as if the physical object could lay claim to the
same prestige as that of the art-work itself. Two cultural currents lie
behind this bid for artistic value, I believe. The first is the anxiety
about lost aura that Walter Benjamin explored in his influential essay
‘The Work of Art in the Age of Mechanical Reproduction’ (1936).
According to Benjamin, mechanical reproduction destroys the aura of
a work of art, and leads to a tremendous shattering of tradition. Now
that mechanical reproduction has removed restrictions on when and
where particular types of art are experienced, the aura of authenticity
is lost and cannot be recovered, he maintains. Benjamin’s essay was
first published in the 1930s, when anxiety about technology and its
effects on the arts was particularly acute, but his argument still reso-
nates today, and the fact remains that many opera fans regard opera on
DVD as a poor substitute for the ‘live’ experience in the opera house.
Some recent scholarship compellingly challenges this privileging of
the ‘live’ and emphemeral, but DVD production companies are in all
likelihood less interested in ontological debates than in consumer atti-
tudes and sales, and this is one possible reason why a company such
as Opus Arte would choose to promote its discs in terms of cultural
value and timelessness, generating in the process an ersatz aura for its

22  (17 March 2010).

23 See, for example: Christopher Morris, ‘Digital Diva: Opera on Video’, The
Opera Quarterly 26: 1 (2010), 96–119; and Philip Auslander, Liveness: Perfor-
mance in a Mediatized Culture, 2nd ed. (London, New York: Routledge, 2008).
The Opera Director’s Voice 147

The second cultural current that I believe lies behind Opus Arte’s
marketing copy is the disavowal of economics already touched on in
this essay. As Benjamin himself notes, ‘to an ever greater degree the
work of art reproduced becomes the work of art designed for repro-
ducibility’. Philip Auslander makes the economic implications of this
argument clear when he remarks: ‘By being recorded and becoming
mediatized, performance becomes an accumulable value’. In 2007,
Opus Arte was bought by the Royal Opera House, which must have
seen a monetary value in capturing its performances for the long-term.
But within public discourse, the commercial motivation behind the
move tends to get underplayed. In an interview published in 2009 in
The Times, Tony Hall, the Chief Executive of the Royal Opera House,
says that Opus Arte ‘makes enough to cover its costs, and a little bit
more’, and that the acquisition is not primarily about money: rather it
is about ‘extending the reach’ of the Royal Opera House, and allowing
the company to share its work with a wider public. This rhetoric of
accessibility allows the company to dwell on its charitable status, even
while it goes about the business of defending its pre-eminent position
within the cultural landscape of the UK and the international opera
field. As Bourdieu notes, ‘those in dominant positions operate essen-
tially defensive strategies, designed to perpetuate the status quo by
maintaining themselves and the principles on which their dominance
is based’. Increased market presence through mediatization is one
such defensive strategy, and opera on DVD is certainly a part of that
Within the field of opera DVD production, Opus Arte seeks a po-
sition of pre-eminence and distinctive brand identity through its gen-

24 Reproduced in Auslander, Liveness, 31.

25 Auslander, Liveness, 28.
26 Dan Sabbagh, ‘From the Ten O’Clock News to a night at the opera, Tony Hall is
taking it to the people’, in The Times (27 June 2009) http://business.times on- (1 November
27 Since 1962, the Royal Opera House has been a charity recognized under the UK
Charities Act. It is currently the largest recipient of public funding of any arts
institution in the UK (£28,294,806 in 2010/2011). http://www.artscouncil. (2 November 2010).
148 Áine Sheil

erous provision of bonus features. In other words, it positions its discs

as luxury items, precisely on account of the extra elements they in-
clude. These bonus features act as an enrichment and embellishment,
and in that sense they extend the authorship associated with the main
feature. As Stephen Donovan, Danuta Fjellestad and Rolf Lundén
explain in Authority Matters: Rethinking the Theory and Practice of
Authorship (2008), the word ‘author’ is connected not only with au-
thority and authorization, but also with ‘augmentation’ and giving
meaning to that which already exists. So in that sense, the DVD ex-
tras partake in and open up the process of authorship, challenging the
idea of the fixed work by providing additional viewing pathways for
the recipient. In another sense, though, they achieve quite the oppo-
site: when bonus features include directors talking about their work,
these commentaries channel interpretation and further the process of
direction itself. In addition to directing the cast, the director takes
responsibility for directing the viewer: a voice of authority guides the
audience and provides a way of understanding the recording, which in
itself represents a fixing or stabilizing act. The production becomes a
definitive text that aspires to the condition of a timeless work, and the
voice of the director provides a powerful sense of authenticity that
counters any undesired loss of aura associated with the effects of me-
chanical reproducibility.
In ‘Digital Diva: Opera on Video’ (2010), Christopher Morris ar-
gues that opera tends to be as fixed in the opera house as on DVD, and
that despite a common belief that every performance is unique, the
reality is that gestures and choreography rarely change from one per-
formance to the next. The complex apparatus involved in staging op-
era favours consistency, so that even in the opera house, productions
‘become works: no longer labor that is being done, but a trace of labor

28 Donovan, Fjellestad and Lundén’s explanation is as follows: ‘Today it is clear

that “auctor”, the Latin origin of “author”, is derived from the verb augere,
which means “to increase, augment, strengthen that which is already in exis-
tence”; in addition, it means to exalt, embellish, enrich.’ Stephen Donovan, Da-
nuta Fjellestad and Rolf Lundén, ‘Introduction: Author, Authorship, Authority
and Other Matters’ in Authority Matters: Rethinking the Theory and Practice of
Authorship, ed. Donovan, Fjellestad and Lundén (Amsterdam, New York: Ro-
dopi, 2008), 1–19: 2–3.
The Opera Director’s Voice 149

that has been done’. Morris attributes this lack of spontaneity to
Werktreue, ‘with its investment in the notion of serving and authenti-
cally realizing the operatic work bequeathed to us by notation and
authenticated by tradition’. This suggests that there is no need to
regard the DVD in opposition to performance in the opera house, and
that both form part of a continuum within opera practice that revolves
around texts, works and semi-permanent realizations of these in the
form of productions.
That the ideal of transparent or ‘faithful’ interpretation remains a
central feature within opera practice is borne out by DVD interviews
with directors: these provide evidence of prevalent values, not just
among directors themselves, but more importantly, within the wider
field of opera production. They also provide evidence that DVD pro-
duction companies perpetuate the concept of authority, whether this
involves the authority of the composer or the authority of the contem-
porary ‘charismatic’ director/conductor.
But why do the production companies perpetuate values that
speak more of the nineteenth century than the contemporary western
world? Do the production companies care about the work concept, the
imaginary museum of music, textuality and the authority of compos-
ers, directors and conductors? I think that they do, if only because the
field of cultural production still functions in these terms. As Goehr,
Taruskin and many others have shown, the classical music industry is
still heavily invested in the notion of the single originating creator and
the definable ‘work’. This is the basis on which classical music con-
tinues to be sold. Even though music-making on the whole, and opera
in particular, involves elaborate chains of collaboration and reception,
and even though poststructuralist criticism has long questioned the
stability of texts and idea of exclusive authorship, the need for author-
ity and authorization persists to this day within opera. DVD interviews
with directors add a funny twist to all this: the interpreter becomes
visible, but the ideal of invisible interpretation remains.
These interviews thus become quite subtle performances within
themselves: directors walk a fine line as they explain their work and

29 Morris, ‘Digital Diva’, 101.

30 Ibid., 102.
150 Áine Sheil

its contingencies, while allying themselves with perceived authorial

intentions. Whether or not this quest for authorization is sincere or
cynical, conscious or unconscious is, perhaps immaterial: more perti-
nent is the fact that opera practice still functions on the basis of le-
gitimacy and authority. Authority is traced back to the original creator
and is seen to descend from the composer, but is, in fact, more likely
to emanate from contemporary institutions and the complex social
interactions to which institutions have given rise over the past few
centuries. Opera directors are an important part of this complex chain;
they profit both materially and symbolically from their interpretation
of opera, and their belief in the work and the composer is not – and
can never be – disinterested. As disseminated on DVD extras, their
voices form a chorus that pays homage to dead geniuses and the con-
cept of fidelity through interpretation. These views, though remedi-
ated and re-presented with the help of powerful and suggestive televi-
sual effects, are not new: instead of serving the composer and
individual works, they ultimately serve to stabilize and fix the tradi-
tional values of an entire artistic field.
Special Contributions
A Conversation with Pietro D’Agostino,
Video Director at the Gran Teatre del Liceu

What is the process of preparing for filming?

I don’t have much time to prepare an opera recording. I begin thinking
of the shots with the first rehearsals on stage. Then, I only have two-
to-three weeks to prepare my story-board for the recording. They are
very full working-weeks: seven days a week, ten-to-fourteen hours a
day, but it is still a short time. We don’t have the chance, as other
production members do, to work with it from its conception. We al-
ways come behind the opera production. Basically, I go to rehearsals
to get to know all the main scene movements and secondary reactions.
Eventually, I know them by heart. Even though every performance is
different, it is incredible how much singers repeat themselves in the
details of their acting. This is probably due to the fact that music gives
them a very strong structure.
At this time, I begin recording something on video too. Since
Liceu can count on permanent video broadcasting facilities, I am
lucky enough to get to use several cameras and begin testing camera
positions and angles. As we all know, opera displays itself not only in
time, but also in space, which means that further actions are taking
place at the same time in different spaces. On the other hand, video
develops its structure only on a time-line (mainly). What is important,
then, is to make the right choices to show what is significant in order
to be faithful to the stage director’s intentions.
Even if, during rehearsals, I start planning cameras positions and
choose the lenses I’m going to use for each camera, I still have to wait
until the lighting design is finished to come up with the final configu-
ration (which is often no earlier than the dress rehearsal). The lighting
design is actually a big issue. We are going to have to change it since
it usually doesn’t work for camera. Even so, I don’t want to change it
drastically, I want to change it as little as I can, because I am con-
154 Pietro D’Agostino

scious of the fact that there is going to be an audience in the audito-

rium that will have come to see an opera, not a television set. So if one
camera angle will help to save a certain light cue planned by the light-
ing designer, I often prefer going with that instead of making changes.
Once I am done with that, I start writing down the actual story-
board. I write down all the sequences in a script, and I meet with the
musical assistant to put all the shots cues down on the score. Eventu-
ally we will have all the shots numbered on the score and on different
types of scripts (mine, the assistant director’s and one for each cam-
eraman). After the first recording, I do corrections. Both in order to
get a better second recording and in order to plan the second recording
in a different way for post-production reasons (If I want to record
different details, for example), in the case of productions for DVD.
Live broadcast always takes place on the second day of recording,
that way the first day works as a rehearsal for us. Before all this, be-
fore rehearsal, it is a time for gathering all the information on the pro-
duction I am going to film (pictures, video material, stage director’s
intentions) and to study it. In some cases I get to talk with the stage
directors, but not always and not necessarily.

What do you get out of these talks? Do you ask for changes to the
stage direction?
Of course, anything the director can tell me is welcome and can be
helpful. However, sometimes they just don’t care, or they might focus
on things that are not really relevant for my video. I might need some
changes to the stage direction, but I am conscious that I am not doing
the video direction of an opera. What I am doing is the “video ver-
sion” of a particular mise-en-scène. I certainly agree with what has
been said during this workshop; I totally believe that there is room for
authoriality in video direction of operas. At the same time, though, I
am conscious that Opera itself is a work of craft, too. So I think that
maybe what we are doing now belongs more to the world of craft than
to that of Art or authoriality. At the Liceu, we have a rehearsal room
named after Mestres Cabanes, a very good painter, who was the set
designer of the house in the mid 1900s, when opera settings were
mainly made of painted paper…
Pietro D’Agostino 155

It doesn’t mean that we will not develop to a different stage (just

as set design or stage direction did…) and I think that the great
amount of work we are doing to get live broadcasts polished - nice
and meaningful - is not going to become lost (like Mestres Cabanes’
paintings), though, right now, we are probably still just experimenting
in the kitchen with a primitive technique…
The opera houses themselves don’t really know what video is …
and the whole world around opera doesn’t really bother asking itself
what video is…
Movie theatres programme opera the same way they do live
sports or pop concerts…

Are you free of changing, or do you ever change the order of your
planned story-board while broadcasting live?
Yes, I am free of doing it, I might want to do it and happen to do so
during recordings, but I try to avoid it during live broadcasting since,
if I did so, the whole machine might not be able to work.

Is it conceivable that you computerized the cues on your story-board

(as it happens with modern lighting design) so that you would be free
to play with what is coming in without having to articulate it?
No, because live acting is always different, even slightly different, and
the tempos are different too. So the tempo of the video making is al-
ways changing. In fact, a great part of my job is to catch the unique
moment of the live performance and synchronize with it.

Do you always record all the camera signals? Can you count on all
the camera signals in postproduction?
No; due to budget reasons, besides the programme, I record just four
more signals, which I switch during the recording. For example, I
might want to record specific camera signals in one scene and some
different ones in another. Of course, while being recorded, the camera
signals must be good enough to be edited eventually, (not shaking, on
focus, etc.). That means that when I prepare the live broadcasting I
always have to keep postproduction in mind as well.
156 Pietro D’Agostino

Are you responsible for the chapters of a DVD? And subtitles?

No, it is the record label that does the DVD authoring.

I imagine you have to deal with a lot of cuts on the video track during
postproduction. Do you do the same with the audio track? Or do you
try to avoid editing the audio?
The idea is to try to keep live continuity as much as we can, but we
often have to deal with audio editing as well. Since the music duration
is the leading reference, what we do is to cut the audio first, and work
the video according to that audio edit. After we finish the video edit-
ing, we do the final audio mixing over it, so that the sound can match
the image. (Let’s say we have a close up; you will hear the voice
“closer”, while in a full shot we will hear “the presence of the ambi-
It might happen that a singer wants to supervise the editing, too…

Contractually, does any singer get the right of refusal?

Usually, at the Liceu, they sign the image rights contract when they
sign for the role.

What about HD?

It might sound strange, but what is interesting about opera on video is
that we are always on the cutting edge of technology development.
We have been doing HD for ten years now, and nowadays we are
among the first to be experimenting with 3D. As an example: last
season, at Liceu, we have been the first in this country to do a 3D live
broadcast following the DCI standards (the standards for digital cin-
ema). That is essentially because we might not satisfy a massive popu-
lar market, but we will maintain that market in the long run. We might
be recording a singer’s voice today that will still be demanded in the
distant future… so we must be ready for that…

What about robotic systems, do you use them?

Yes, the house has a robotic system (hot head cameras) created with
the intention of lowering the impact of camera presence in the audito-
Pietro D’Agostino 157

rium, so that the cameras and cameramen wouldn’t bother the audi-
ence. On the other hand, the presence of the audience is a great limita-
tion for us. We always have to shoot far from the stage and we cannot
play with focal length and depth of field as much as we would like, as
video makers. So whatever can help shooting (I’m thinking of robots
in otherwise positions unreachable…) is, of course, welcome.
As for now, we cannot always use a whole robotic system and we
have to put some cameras on tripods (due to lens weight or camera
position necessities).

DVD production is now several-years old. Do you think that this fact
has changed the way contemporary stage directors look at it? Do you
think they see it as an opportunity for their work?
I am sure that there are directors closer to video making than others.
Our imaginary is always developing, and all the different disciplines
of image-making influence each other, I guess. You often see direct
references to cinema in modern stage directions, for example.
For sure, I feel there are productions that are “easier” to recreate
on video than others. Of course, this has nothing to do with the quality
I might see in a specific production. There might be mise-en-scènes I
love that I think are not made for video, and others I hate that work
well on video and help me do my job (and the other way around). As
an example: a theatrical tool that’s always difficult to translate to
video is darkness. A dark stage is totally different from a black elec-
tronic monitor… at least, as far as we can achieve right now by alter-
ing the stage production for our recording purposes…

What if a director doesn’t stage the orchestral interlude, do you feel

that the only option is to show the pit?
By default, my camera goes to the orchestra. Because I think you are
always right if you do that, but I am not dogmatic on this and I hap-
pened to choose a different solution when I felt it was required. There
is certainly room for authoriality, here…
Composing in New Synaesthetic and
Interdisciplinary Spaces: Libro de las estancias
(Book of Abodes) as a Musical, Architectural and
Visual Installation Proposal


Libro de las estancias is a piece written in response to a joint commis-

sion from the Granada International Music and Dance Festival and the
Valencia Institute of Music . The work embodies a pilgrimage through
seven abodes, seven time-spaces that stake out an itinerary through
poetic-sonorous images ranging from the desert to writing (in remem-
brance of Edmon Jabès), and from rooms fraught by concepts such as
time or the labyrinth to others laden with symbolic materials linked to
the work’s content, such as lead, stone and alabaster. The work is a
poetic reflection on sound, space, light and the voice. It is a great pal-
impsest composed as a meditation on a part of Spain’s history that goes
beyond the merely sociological or political reality of the period ushered
in by the expulsion of the Moriscos (Moorish converts) after the 1609
decree. This religious and political controversy, with its multifaceted
development over a long period of rapprochements and misunderstand-
ings, is the context that partly nourishes this work. However, the scenic
and musical proposal and its own dramaturgy go beyond this context.

1 The work was premiered on July 9th, 2009 in the atrium of the Caja Granada
building as part of the aforementioned festival’s program. Performers were Car-
los Mena (Countertenor), Marcel Pérès (Arabic voice) and Isabel Puente (Pi-
ano), along with the Coro de la Generalitat de Valencia, the Orquesta Ciudad de
Granada and the Freiburg EXPERIMENTALSTUDIO des SWR (Joachim Haas,
Gregorio García Karman, and Sven Kestel), all of them under the direction of
José M. Sánchez-Verdú (conductor I) and Joan Cerveró (conductor II). The
score is published by Breitkopf & Härtel (Wiesbaden).
160 José M. Sánchez-Verdú

I. Two Gazes – Two Fictions

Libro de las estancias attempts to reflect on two gazes, two differing

views on man, art, and therefore the world, that have old roots in his-
tory and yet remain current. Spain, by way of Al-Andalus, has long
been an exceptional hybrid and meeting point for both ways of seeing
the world and living in it. I am speaking here of an “Arab gaze” and a
“Western gaze”. The former seems to arise from the desert and is fed
by particular ways of organizing space, light, sound and life itself,
tinged and enriched by Islam. The Western gaze, stemming from the
Greco-Roman heritage, is permeated via Christianity by an essentially
different way of observing and inhabiting the world. These are two
religions of the Book and both offer solutions of their own regarding
the confrontation between the scriptural and oral traditions. They are
two different ways of formulating architectural vision, of drawing the
horizon, of embellishing space and of making special use of images (a
conflicting endeavor in Islam due to its ban on figurative representa-
tion). Furthermore, they offer two different ways of confronting mat-
ter, texture, geometry, etc. in the creative process and as regards the
concepts of space and time. The repercussion of both views is notice-
able not only in literature and poetry, with their (often opposite)
themes and viewpoints, but also in the arts of calligraphy, painting,
decorative arts and architecture. The Muslim architect usually ad-
vances linking spaces together and superimposing them; the Christian
architect draws his master plan, sets it a priori, and expands space
adapting it to the plan. The Muslim besides fills all free space and
saturates it with texts and geometric ornaments – often united in the
art of Islamic calligraphy, with its semantic, ornamental and theologi-
cal implications. In this sense, the conceptual and artistic clashes be-
tween the Palace of Charles V and the adjacent Nazarite palaces in the
Alhambra where the former was embedded are paradigmatic, as ana-
lyzed by Gómez de Liaño in Los juegos del Sacromonte . These two

2 Gómez de Liaño, Los juegos del Sacromonte, Editora Nacional (Biblioteca de

Visionarios, heterodoxos y marginados), Madrid, 1975 (a new facsimile edition
was published by Editorial Universidad de Granada in 2005).
Composing in New Synaesthetic and Interdisciplinary Spaces 161

gazes determine the creation and development of musical matter, texts

and dramaturgy of space and light in Libro de las estancias.
As in Islamic poetry, the themes and texts in Libro de las estan-
cias are alluded to but not directly exhibited. The two gazes are linked
as well to two fictions that have shaped not only Spain’s fate but also
the paths of these two different worlds. The first fiction is the inven-
tion of the “lead books” of Sacromonte – including the famous
parchment of Torre Turpiana and the relics that began to surface at the
end of the 16 century in Catholic Granada and its nearby mount Val-
paraíso (today’s Sacromonte) . It is the fiction of a mythic-religious
past in a context where Islam and Christendom seemed to coexist, the
search for some kind of lifesaver on the part of certain Moriscos faced
with the actual threat of expulsion and diaspora; for example, Saint
Cyril, the first bishop of Granada, is presented as an Arabic author.
The second fiction has to do with the forging in Christian Spain of a
myth about the presence of Saint James in the land of Galicia. The
Apostle James arises as a figure supporting not only the Reconquista
but also as the very justification of the religious and political unity of
a newborn political system that eventually was to dominate and ex-
clude the other, whether Jewish or Muslim. The two mentioned gazes
sought support, protection and justification through fictions lending
depth and grounds for a clash that was not simply political and reli-
gious but also aesthetic and cultural.

3 Among the most recent studies on the topic, see Los plomos del Sacromonte.
Invención y tesoro (Eds. M. Barrios Aguilera y M. García-Arenal), Biblioteca de
Estudios Moriscos, Valencia, Granada, Zaragoza, 2006 and ¿La historia inven-
tada? Los libros plúmbeos y el legado sacromontano (Eds. M. Barrios Aguilera
y M. García-Arenal), Editorial Universidad de Granada, 2006.
162 José M. Sánchez-Verdú

II. The Work

Libro de las estancias is structured in seven movements or “abodes”,

besides two interludes, which define seven spaces, i.e., seven meet-
ings with symbols and writings of memory:

1 / VII Estancia del desierto (Abode of the Desert)

2 / VI Estancia del plomo (Abode of Lead)
Interludio I (Interlude I)
3 / V Estancia de la memoria (Abode of Memory)
4 / IV Estancia de la piedra (Abode of Stone)
5 / III Estancia del laberinto (Abode of the Labyrinth)
Interludio II (Interlude II)
6 / II Estancia del alabastro (Abode of Alabaster)
7 / I Estancia de la escritura (Abode of Writing)

Each of the parts proposes a different dramaturgy of sound, space and

color. The texts also offer a continuing alternation and dialogue be-
tween the two aforementioned gazes: a play of mirrors starting from
two vocal and instrumental groups and two contrasting soloists (a
countertenor and an Arabic voice) which interrelate via a piano and
are also expanded through the auraphon and live electronics. The Ara-
bic voice (Marcel Pérès) confronts the Western voice of a coutertenor
(Carlos Mena); in their spatial arrangement the two gazes/fictions
create opposing and symmetrical spaces. This also means that the
musical reading of materials from Abode I to Abode VII offers a
symmetrical reading that is reversed from VII to I (marked in the op-
posite reading with Arabic numbers from 7 to 1). They are two cross-
readings immersed in a play of mirrors and symmetry. The work may
be “read” in any one of two directions; yet at the same time, reading
one of them assumes an implicit reading of its opposite; such is the
play of mirrors and symmetry developed in Libro de las estancias.
That is why the work is organized in seven rooms or abodes numbered
with diverging yet superimposed Roman and Arabic numerals. Two
formal arrangements are offered to perform the work: following the
order of the Roman numbers for the seven rooms (from left to right) or,
viceversa, reading/performing the seven rooms from finish to start (or
Composing in New Synaesthetic and Interdisciplinary Spaces 163

right to left) following the Arabic numbers and the direction of Arabic
writing. Both interpretations are desired, and even the perceiving and
hearing of both “readings” is part of the work’s design: both gazes are
possible at once since they offer an open structure, issued directly from
the scriptural realm and its linear nature in two different systems. The
Arabic reading starts intentionally with “Abode of the Desert”; the
Western gaze or reading stems from “Abode of Writing”.

I – II – III – IV – V – VI – VII = 7 – 6 – 5 – 4 – 3 – 2 – 1

Abode of Abode Abode of Abode of Abode of Abode of Abode of
Writing of Alabaster the Stone Memory Lead the Desert
Memory Labyrinth

Abode of Abode of Abode of Abode of Abode of Abode Abode of
the Desert Lead Memory Stone the of Alabaster the World

The possibility of selecting a reading for each performance is deliber-

ate. However, the actual following of a reading implicitly entails the
reading of the opposite: hence the play of symmetry and mirrors de-
veloped by the musical material itself in the seven-space structure.
Each movement contains an intimate expression of the opposite read-
ing in its own musical matter: a synchronic cross-reading of both pos-
sible readings.
Each of the rooms or abodes possesses architectural, textual,
symbolic and light-related connections besides its own musical matter.
All these elements arise and interact based on these associations and
both the Arabic and the Western gaze deeply determine the work’s
scenic and musical development.
164 José M. Sánchez-Verdú

Fig. 1. Libro de las estancias. Page from the score. © Breitkopf & Härtel.
Composing in New Synaesthetic and Interdisciplinary Spaces 165

III. The Texts

The basic sources of the work are two manuscripts of extreme impor-
tance for each of the two gazes/fictions. One is the manuscript of
Torre Turpiana, seemingly discovered in 1588 when a tower of un-
known antiquity, located in the major mosque of Granada where the
cathedral now stands, collapsed. It is one of the first examples of the
Morisco fabrication, surely composed by people with intellectual
learning and a full command of both Castilian and Arabic . Alongside
with it are other texts collected in some of the “lead books” of the
time (The History of the Seal of Solomon, for example). The second
fundamental source is the manuscript of Codex Calixtinus, preserved
in the Cathedral of Santiago, which holds a large number of texts and
musical pieces about Saint James Apostle and the world of pilgrim-
ages to Santiago de Compostela.
An example of this intertextual use of historical sources is the
Arabic voice in the “Abode of Lead” (2/VI), which begins by present-
ing the six letters making up the angles in the star of Solomon – a
symbol of the Abbey of Sacromonte – as Choir I presents the text of
the “lead book” The History of the Seal of Solomon, now once again
kept in the Abbey of Sacromonte (where it was returned by the Vati-
can in the year 2000 after several centuries in Rome).
In the “Abode of Memory” (3/V), the countertenor’s voice in-
tones the Pange lingua, a piece from the winning side’s lore which for
a time was sung on January 2 on the festivities celebrating the taking
of Granada, in the ad hoc adaptation by Fray Hernando de Talavera.
Choir II is developed in turn from one of the emblematic pieces in
Codex Calixtinus, “Congaudeant Catholici”, which might be the old-
est three-voice piece preserved to date in Spain. A dialogue between
two texts, two manuscripts (“Abode of Writing” 7/I), a dialogue be-

4 Alonso del Castillo and Miguel de Luna rank among the foremost candidates as
authors of this great Morisco forgery. See M. García-Arenal and F. Rodríguez
Mediano: “Miguel de Luna, cristiano arábigo de Granada”, in ¿La historia in-
ventada? Los libros plúmbeos y el legado sacromontano, Editorial Universidad
de Granada, 2006, pp. 83 ff.
166 José M. Sánchez-Verdú

tween two scriptural forms representing the two mentioned gazes, is

the textual object offering meanings that, like the inscriptions in the
Alhambra, are not always discernible. Texts appear in their original
form: in Arabic (the Arabic alphabet in the “lead book” History of the
Seal of Solomon, an account of the key to interpret texts in the parch-
ment from Torre Turpiana, etc.) or in 16 century Castilian. I also use
fragments of so-called aljamía texts (written in Spanish with Arabic
characters), the interaction between two languages and two writing
systems that give rise to a fascinating and unique symbiosis.
Composing in New Synaesthetic and Interdisciplinary Spaces 167

Fig. 2. Libro de las estancias. Page from the Score. © Breitkopf & Härtel.
168 José M. Sánchez-Verdú

A portion of these many texts follows:

The History of the Seal of Solomon: lãm – alif – lãm – alif – mim – rã’

Beginning of the Transcription of the Parchment from Torre Turpiana:

La [h]edad de la luz ia comencad (The age of light has already begun

por el maestro i con la pasión by the master and with the Passion
rrod[e]mida con dolor del cuerp redeemed with bodily pain
o i los [p]rofectas pasados and the prophets of the past)

Text of Codex Calixtinus (Responsory):

Huic Jacobo […]

tristis est anima mea usque ad mortem (My soul is grief-stricken to death)

Pange lingua with the adapted text in the Officium of the Taking of
Granada by Fray Hernando de Talavera:

Pange lingua voce alta triumphi praeconium; laudes Deo semper canta, conditorium
qui, edomita Granata, bellis dedit somnium

(Sing, o tongue, out loud

the praise of victory.
Always praise God,
Creator of all things,
Who, once Granada was subdued,
Put wars to sleep.)

Text of “Congaudeant catholici” from Codex Calixtinus:

Congaudeant catholici, laelentur cives celici, die iste

(Let Catholics and heavenly beings rejoice today)

Composing in New Synaesthetic and Interdisciplinary Spaces 169

IV. A Multidisciplinary Stage: Synaesthesia and


The traditional concept of stage/audience is abolished in this work.

Implicit in the score is a spatial and architectural production that re-
quires the audience to be able to move around the whole space at will.
The audience is free and autonomous to see, to hear, to approach the
sound source or keep at a distance; it reacts to sound and light indi-
vidually, according to each one’s own free choice. The work is open
to a deliberate multiplicity of manners of perception: thus one single
person can have significantly diverse experiences of it.
The various sound sources, the real and virtual play with sound,
with its disparate spatial arrangements and emissions, its movement,
and the perception of architecture and the drama of colors, are all in-
tegral parts of the work. This architectural dimension and this multi-
dimensionality in the perception of sound, space, echoes, light… must
be boundless and open to the actual preference of each person attend-
ing the performance.
170 José M. Sánchez-Verdú

Fig. 3. Libro de las estancias. Rehearsal in Granada (2009).

Composing in New Synaesthetic and Interdisciplinary Spaces 171

One of the forms of listening arises from what I have called “intimate
singing”, a manner of using the voice along with electronics where one
cannot hear the real sound source that sings (and is visible), but rather its
resonance and aura in space (in a specific place) through the auraphon
(described below) or the piano. Besides this virtual spatial dimension,
there is a complex real or represented spatial dimension arising from the
specific arrangement of sound sources, as will be shown below.

Fig. 4. Libro de las estancias. Photo of the Audience.

The building where this first performance of Libro de las estancias

took place was designed by Alberto Campo Baeza, one of the fore-
most contemporary Spanish architects. His confrontation with matter,
and above all with light, grants all his work an unmistakable personal-
ity. The building was inaugurated in 2001 and is currently the head-
quarters of Caja Granada. Libro de las estancias, however, is designed
and conceived to be developed and presented in spaces where the
architectural idea of the cube, as an approximate shape, facilitates the
actual spatial arrangement of all sound sources as specified in the
score (cf. the last section in this article) while enabling the projection
of colors in that architectural space (with variable projection systems
172 José M. Sánchez-Verdú

depending on the specific space) . Therefore buildings such as han-
gars, churches, cathedrals, atria and other similar architectural forms –
which from an acoustic viewpoint add the benefit of a high reverbera-
tion – are locations where Libro de las estancias may be performed.

IV.1. The Auraphon

This installation-instrument, developed at the Freiburg EXPERI-

MENTALSTUDIO des SWR with Joachim Haas, springs from my
interest in creating an “aura”, a space of resonance where a certain
series of resonating instruments (eight gongs and tam-tams placed
around the audience in the case of Libro…) interacts with the instru-
ment players and the voices, sometimes independently, sometimes as
a true direct response with all of them. It is not played by anybody but
acts autonomously, controlled from the mixing table. It fulfills a first-
rate acoustic and visual role, akin to an interactive installation I found
compelling from the beginning. Only in Freiburg and thanks to
Joachim Haas and the Experimentalstudio could its technical aspects
and real potential be tested and established. I called it “auraphon” .
The auraphonist is the performer who, from a certain distance but
encompassing all space, follows the score and controls all of the aura-
phon’s levels.
Her acoustic and dramatic function plays a very significant role in
Libro de las estancias. The auraphon is composed of four tam-tams

5 In the Caja Granada building the technical development of colors was carried
out by placing color filters on the neon tubes behind the enormous alabaster
walls lining the atrium. Being translucent, alabaster allowed the light to shine
through, thereby illuminating the walls in the entire area.
6 The first project for an auraphon was linked to my opera AURA (2006–2009),
based on a story by Carlos Fuentes. It may be applied to other projects –some of
which have already been performed– where the auraphon’s elements are rede-
fined, adjusting them to new venues and to each work’s musical and spatial ap-
proach. Another later example was my piece Elogio del tránsito (2010), for
bass/contrabass saxophone, auraphon and orchestra. The essence of the aura-
phon, while rooted in the idea of the aura, may be expanded to new formulae
that can be redefined in their technical development, their instruments, and their
arrangement and perception in relation to sound and space.
Composing in New Synaesthetic and Interdisciplinary Spaces 173

and four gongs that surround all space and also envelop the audience:
listeners are invited to acoustically enter the work’s space in its
imaginary setting; they must perceive the different virtual spaces pro-
duced by the auraphon as well as follow the drama proper that this
installation unfolds in space and time. The auraphon’s instruments
may vibrate through two techniques: tenuto and resonance.

1) tenuto: the auraphon produces a tenuto-type sound – different in

each of its instruments – which is controlled and transmitted from
the mixing table. They are not played the traditional way, by
means of striking or rubbing, but sound on their own: therein re-
sides an important part of its halo of almost ghost-like mystery
arising from the inexplicable, from that type of “absence” and
from the aura ultimately created by the auraphon. Their visual
appearance is a special part of the scenography: they vibrate by
themselves, and in tenuto may acquire highly resonant dynamics
in fortissimo.
2) resonance: the auraphon is connected on several levels to certain
singers and instrument players through microphones, all of it un-
der control by the mixing table and the computer. Each instru-
ment creates a kind of concrete resonance, which is different in
the corresponding tam-tams and gongs. Thus the auraphon is
transformed into an aura or “echo”, into a resonance controlled by
its player. The auraphon thus articulates the various acoustic and
virtual spaces of the seven rooms (and interludes) in the work. In
fact, “Interlude I” is for auraphon solo, whereas “Interlude II” is
for piano and auraphon.

The various changes, developments and transitions of the auraphon,

both in the typologies called tenuto and resonance, are always deter-
mined in the score. One could speak of the auraphon as an installation
that projects and/or interacts both with singers and instrument players
as part of the scenography. This scenography becomes a sonorous part
of the work and its acoustic trajectory is conceived and established in
the compositional process itself: it is an integral and essential part of
the score. The aura acquires a philosophical dimension that was very
meaningful to me in my reflections on the musical and conceptual
174 José M. Sánchez-Verdú

development of the auraphon. In the sense given to this concept of

aura by Walter Benjamin it has been said that “die Reproduktions-
technologie vernichtet den Wert der Einzigartigkeit: es gibt keine Ori-
ginale mehr. Die Kunst wird ein Austellungsstück des Politischen” .
The dialogue between presence and absence arises as a theme for re-
flection about the possibility of reproduction of the work of art. From
the vanguard movements, the annihilation of meaning is at the heart of
this confrontation: “Die Präsenz, die man erwartet und die unaufhör-
lich aufgeschoben wird oder sich aus sich selbst heraus aufschiebt,
kann nur wie bei Beckett oder Derrida zur Enthöllung der Abwesen-
heit geraten” . Whereas with Kant the first dimensions of presence
were space and time, after the shift in the concept of presence we may
conclude that space and time are no more given but constructed and
realized. In the world of the performative, presence does not develop
in time and a space but creates its own time and space: “Das hic et
nunc, das sie ins Spiel bringt, hängt nicht mehr von der Aura ab” .
Walter Benjamin also writes on the “here and now” of the work
of art and its transformation and evolution through history . Benja-
min’s reflections on the aura as regards works of art in general, and
photography in particular , are present in the conception of Libro de
las estancias and in the auraphon as an artefactum meant to create this
aura on diverse levels. Benjamin thus defines the concept of “aura”:

als einmalige Erscheinung einer ferne, so nah sie sein mag.12

The relationship between original and copy, enormously important as

it is here both from a philosophical viewpoint and in the musical and
scenic approach of Libro de las estancias, was already a distinctive
topic in Benjamin’s musings at the beginning of the 20 century: “Die
technische Reproduzierbarkeit des Kunstwerkes emanzipiert dieses

7 Daniel Charles: Zeitspielräume. Performance Musik Ästhetik, pp. 75–76.

8 Ibid., p. 79.
9 Ibid., p. 79.
10 Walter Benjamin: Das Kunstwerk im Zeitalter seiner technischen Reproduzier-
barkeit, Frankfurt., p. 11
11 Ibid., p. 16.
12 Ibid., p. 15.
Composing in New Synaesthetic and Interdisciplinary Spaces 175

zum ersten Mal in der Weltgeschichte von seinem parasitären Dasein

13 14
am Ritual” : a “desecularized” ritual, in Benjamin’s own words . And
this is what Libro de las estancias develops: a ritual of cross-mirrors,
dramatic echoes, resonances between past, present and future in the
history of those two gazes, an exhibition of characters, texts, inven-
tions and clashes multiplied and crystallized on several levels at once.
Concepts like “original” and “copy” meet in the auraphon’s de-
velopment and in the use of space, resonance, echoes, mirrors and
diverse “unfoldings” offered by instrument players and singers. This
repetition, this multiplication in mirror images, plays with the specta-
tor’s imagination in the sense mentioned by Deleuze, “(…) Repetition
is, in its very essence, imaginary, since only imagination forms the
“moment” of vis repetitiva here (…) The imaginary repetition is not a
false repetition, which would supplement the absence of the true one;
the true repetition is imagination” . Repetition is constant in many
types of creation; it is not only essential in several world cultures but
also in the creative process of more recent Western artists (examples
are Claude Monet with his series on the Rouen Cathedral, Paul Klee,
Andy Warhol, and Pablo Palazuelo, among others). In drama, repeti-
tion acquires a special energy of great significance, even when a real
repetition is performed: what is repeated happens at a different mo-
ment than the original, and besides,

Es geht nicht um die Bedeutung des wiederholten Geschehens, sondern die Be-
deutung der wiederholten Wahrnehmung selbst. Tua res agitur: die Zeitästhetik
macht die Bühne zum Schauplatz einer Reflektion des Seh-Akts der Zuschauer.

Libro de las estancias aspires therefore to articulate an imaginary

scene where ritual acquires the maximum category of its expression,
and echoes, resonance and the aura contribute to recreate that gallery
of duplications that envelop the spectator in a poetic overlap of images
and endlessly superimposed worlds.

13 Ibid., p. 14
14 Ibid., p. 16–17.
15 Gilles Deleuze, Diferencia y repetición, Buenos Aires, 2002, p. 127 (Difference
and Repetition, Columbia University Press, New York, 1995).
16 Hans-Thies Lehmann: Postdramatisches Theater, p. 337.
176 José M. Sánchez-Verdú

IV.2. Live Electronics

The function of live electronics, both in the process of sound trans-

formation and in several forms of spatial arrangement, takes on a de-
finitive and essential role here. Whereas the work was conceived for
very particular spaces, such as the atrium of Caja Granada, I have
devised the application of several kinds of delays that partake of the
possibilities of highly resonating spaces and also of the two concepts
or gazes around which I developed the work and its instrumental, vo-
cal, and electronic transformation materials. Among these electronic
transformation processes are those I have called Hoch-Delay 1 and 2,
Mosaik-Delay and two different forms of Erosion-Delay. All of them
directly interact with the voices and instruments, establishing a first
level of listening and influencing the work’s dramatic development
and the sound materials in play. The spatial arrangement of loud-
speakers creates another network of sound sources that interacts with
the performers and also develops an independent sonorous and spatial
drama of its own.
Composing in New Synaesthetic and Interdisciplinary Spaces 177

Fig. 5. Electronic section in the score. (MAX-MSP).

178 José M. Sánchez-Verdú

IV.3. Light

The use of light and color, used in a streamlined and poetic fashion,
stands out as an integral element of this work. The dramatic plot set for
Libro de las estancias is directly linked to the use of color by the Mus-
lim architect in certain spaces such as the ceiling of the Palacio de Co-
mares in the Alhambra. The study of its restored original polychromy
has yielded symbolic and theological consequences . The dramatic
development of colors throughout Libro de las estancias stems from
this connection with the Muslim architect’s thought and interacts not
only with musical time and space and the various abodes in the work
but also with the auraphon and the musical material itself, in particular
through synaesthesia. The dramaturgy of colors unfolds in the work via
their diachronic appearance and compilation from the bottom up as we
move from one abode to the next. Colors and musical materials also
exhibit a synaesthetic link with musical matter. Thus in my own synaes-
thetic perception musical heights and textures are linked to colors (e.g.,
red is G, black is C, yellow is D and F, white is A, etc.). The drama-
turgy of colors is configured in the following outline and establishes a
structure in seven grounds (or floors) from the floor level to the upper
area, next to the ceiling. Some of these floors or stages break down in
turn into two horizontal rows dividing each floor in two horizontal strata.

17 On this topic, see D. Cabanelas Rodríguez El techo del salón de Comares en La

Alhambra. Decoración, Policromía, Simbolismo y Etimología, Patronato de La
Alhambra y Generalife, Granada, 1988 (especially p. 59 ff.).
Composing in New Synaesthetic and Interdisciplinary Spaces 179

HTLM Name RGB Hex Code RGB Decimal Code

FLOOR 7 Top row:

Bottom row:
Beige F575DC 245245220
FLOOR 6 Top row:
Gold FFDF00 2552150
Bottom row:
Yellow FFFF00 25525500
FLOOR 5 Both rows:
Dark orange FF8C00 2551400
FLOOR 4 Both rows:
Crimson DC143C 2202060
FLOOR 3 Both rows:
Red FF0000 25500
FLOOR 2 Top row:
Forest green 228B22 3413934
Bottom row:
Dark green 006400 110000
FLOOR 1 Black

Libro de las estancias is a journey through history and those two

gazes/fictions; it is a pilgrimage through (musical) matter, through
different texts, through space and light and its different hues. The
genealogical link of these particular colors with the Palacio de Co-
mares is one among many textual, sonorous and light-related connec-
tions established by this dense network of intertextual and interdisci-
plinary relations.
180 José M. Sánchez-Verdú

Fig. 6. Performance. ©Carlos Choin.

Composing in New Synaesthetic and Interdisciplinary Spaces 181

Fig. 7. Lights.
182 José M. Sánchez-Verdú

V. Appendix

Instrumental and Vocal Scoring

Group I
(Conductor I)

Wind instruments I (1 Horn, 1 Trumpet, 1 Trombone)
Choir I (12 voices: 3 Soprano, 3 Alto, 3 Tenor, 3 Bass)
Strings I (1 Violin, 1 Viola, 1 Cello, 1 Double Bass)


Group II
(Conductor II)

Arabic voice
Wind instruments I (1 Horn, 1 Trumpet, 1 Trombone)
Choir I (12 voices / 3 Soprano, 3 Alto, 3 Tenor, 3 Bass)
Strings I (1 Violin, 1 Viola, 1 Cello, 1 Double Bass)

Live electronic (MAX MSP) + Auraphon

Spatial Arrangement of Performers and Sound Sources

The cube is the basic shape inspiring several elements in the work’s structure.
The lighting design and the electronic spatial arrangement of sound are set up accord-
ing to the possibilities offered by this shape. The deployment of all musicians (singers
and instrument players) also responds to this shape and its possible symmetries, an-
gles, etc. Some performers, such as the wind instruments, are placed at the top, on two
of the cube’s upper faces.
Composing in New Synaesthetic and Interdisciplinary Spaces 183

Choir II + Strings II Arabic voice


Choir I +
Strings I Winds II


Winds I Piano

The spatial arrangement of loudspeakers (10 in number: 8 of them some two meters
above the floor and 2 further up, near the seventh floor of colors) and of the aura-
phon’s eight instruments (the odd-numbered are gongs, and the even-numbered are
tam-tams) is to be superimposed on this outline. A complex overlay of networks is
thus created covering and expanding all space in the performance and creating a
further level of virtual spaces.

L = Loudspeaker
A = auraphon (gong/tam-tam)

A1 L2 L3 A2



L9 (up)



L L7 A6 L10 (up) A5

Gabriela Cruz is Investigadora Auxiliar at the Centro de Estudos de

Sociologia e Estética Musical, Faculdade de Ciências Sociais e
Humanas, Universidade Nova de Lisboa, where she teaches music.
She has published on Meyerbeer, nineteenth-century opera, and music
in Portugal in The Opera Quarterly, Cambridge Opera Journal, Re-
vista Portuguesa de Musicologia, Current Musicology and other ven-
ues. She is currently leading a new research group studying musical
comedy in nineteenth-century theaters in Portugal and Brazil with
funding by the Fundação para a Ciência e Tecnologia, Portugal.

Pietro D’Agostino is video director of the opera productions and con-

certs season of the Gran Teatre del Liceu at Barcelona (Spain). He has
been videoartist of Studio Azzurro in Milan (Italy). He produced by
Rai Sat thirty videopoems and a documentary about contemporary art
for the Region of Lombardy. He directed and edited the short film "Il
piano dell'uomo sotto" with Lou Castel, produced by Lars Films in
collaboration with Videa. The film participated in the Torino Film
Festival (Italy), winning the Special Award of the Jury. It also won the
award of the Dipartimento dello Spettacolo (Ministry of Culture), has
been distributed by Istituto Luce and selected by A.I.A.C.E. (Associ-
azione Italiana Amici del Cinema d'Essai) between the Italian top ten
films of the year.

Héctor J. Pérez is Associate Professor of Audiovisual Communica-

tion and Aesthetics and a member of the Technology and Information
Research Team, CALSI, at the Universidad Politécnica de Valencia
(Spain). Among his main publications on Opera are “Shakespeare
jenseits des Dramas” (1998); El Nacimiento de la tragedia. Un ensayo
sobre la metafísica del artista en el joven Nietzsche (2001); “Opera
Narratives: From Mythology to Audiovisual Aesthetics” (2006), Ex-
186 Authors

pression in the Performing Arts (2010), “Una estética audiovisual de

Electra” (2010).

Jaume Radigales is PHD in Art History. Since 1994, he teaches Aes-

thetics, Music and Moving Image and Avant-garde and Movies at the
Facultat de Comunicació Blanquerna (Barcelona’s Universitat Ramon
Llull). He teaches also Dramaturgy and Music at the University of
Barcelona. He is the main researcher of the Music and Image Re-
search Group at Ramon Llull’s University. He has published some
books and a lot of articles about the relationship between music and
movies and hi is critic of some magazines, radio and newspaper of

José M. Sánchez-Verdú José M. Sánchez-Verdú is composer and

conductor. He has degrees in compositon, conducting, musicology and
also in Law (RCSM Madrid, Universidad Autónoma and Universidad
Complutense) and postgraduate composition studies in Frankturt
(Musikhochshchule). He has composed many music theater pieces,
orchestral and chamber music and different music projects for installa-
tions and interdisciplinaries fields (architecture, light design, etc.).
These pieces were presented until now in Berlin, Munich, Hamburg,
Stuttgart, Lucerne, Venise, Madrid, Buenos Aires, Cairo, etc. He was
"composer in residence" in different festivals in Germany, Austria,
Switzerland, Spain, Perú, etc. He was awarded with many prizes:
Siemens-Foundation (Munich), Junge Deutsche Philharmonie (Ger-
many), Bergische Biennale (Wuppertal), Irino (Tokyo), Premio Na-
cional (Spain), etc. He teaches Composition at the Robert-Schumann-
Hochschule (Düsseldorf) and at Conservatorio Superior de Música de
Aragón (Zaragoza).

Emanuele Senici is Professor of Music History at the University of

Rome La Sapienza. He has published books and articles on Italian
opera of the long nineteenth century, the theory and historiography of
opera (especially issues of gender and genre), and opera on video.
Between 2003 and 2008 he was co-editor of the Cambridge Opera
Journal with Mary Ann Smart.
Authors 187

Áine Sheil is a lecturer in music at the University of York, UK. She

previously worked in the music department at University College
Cork, Ireland, the drama department at Trinity College Dublin, and
the publications department of the Royal Opera House, London. She
has published on Wagner reception and staging, contemporary opera
practice and theatre-related arts policy, and is currently guest editing
the special issue ‘Digital Opera: New Means and New Meanings’ for
the International Journal of Performance Arts & Digital Media (Intel-

Gaia Varon Gaia Varon is completing a doctoral thesis on videos of

symphonic music at the University of Bologna. She has published
articles and book chapters on the topic of her thesis, as well as on
classical music recording style and technique, and music in avant-
garde short films. She lectures at the Libera Università IULM and the
Conservatorio Giuseppe Verdi in Milan, where she also works as both
author and presenter of music programmes for Rai Radio3.

Delphine Vincent studied musicology, history and aesthetics of cin-

ema, and history at the Universities of Lausanne, Geneva and Fri-
bourg. She received her Ph.D in musicology, «L’œil écoute». Musique
classique filmée: perception, reception, idéologie, at Fribourg Univer-
sity (Switzerland) in 2011. From 2005 to 2011, she was lecturer at
Fribourg University (Institute of Musicology). She is currently a re-
search fellow of Fonds National Suisse at Paris VIII University (since
2011). Her research interests are relations between music and visual,
film music, French music (1850-1950).