Edward Campbell, Peter O'Hagan - Pierre Boulez Studies-Cambridge University Press (2016) PDF

Pierre Boulez Studies
Pierre Boulez is considered one of the most important composers of the twentieth and
early twenty-first centuries. This collection explores his influence, legacy, reception and
works, shedding new light on Boulez’s music and its historical and cultural contexts.
In two sections that focus firstly on the context of the 1940s and 1950s and secondly on
the development of the composer’s style, the contributors address recurring themes
such as Boulez’s approach to the serial principle and the related issues of form and
large-scale structure. Featuring excerpts from Boulez’s correspondence with a range of
his contemporaries here published for the first time, the book illuminates both Boulez’s
relationship with them and his thinking concerning the challenges which confronted
both him and other leading figures of the European avant-garde. In a third and final
section, three chapters examine Boulez’s relationship with audiences in the UK, and the
development of the appreciation of his music.
edward campbell is senior lecturer in music at the University of Aberdeen and a

co-director of the University’s Centre for Modern Thought. He specialises in
contemporary European art music and aesthetics. He is the author of the books
Boulez, Music and Philosophy (Cambridge, 2010) and Music after Deleuze (2013).
peter o’hagan is a pianist specialising in the performance of contemporary music.
He has performed Boulez’s piano works at festivals in the UK and abroad, and in 2008
gave the London première of the unpublished three-movement version of the Third
Sonata at Wigmore Hall with the composer’s authorisation. He has recently completed
a monograph, Pierre Boulez and the Piano.
Pierre Boulez Studies
edited by
Edward Campbell
and
Peter O’Hagan
University Printing House, Cambridge CB2 8BS, United Kingdom
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107062658
© Cambridge University Press 2016
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2016
Printed in the United Kingdom by TJ International Ltd. Padstow, Cornwall
A catalogue record for this publication is available from the British Library
Library of Congress Cataloguing in Publication data
Campbell, Edward, 1958– | O’Hagan, Peter.
Pierre Boulez studies / edited by Edward Campbell and Peter O’Hagan.
Cambridge : Cambridge University Press, 2016. | Includes bibliographical references
and index.
LCCN 2016021875 | ISBN 9781107062658
LCSH: Boulez, Pierre, 1925–2016 – Criticism and interpretation. | Music – 20th
century – History and criticism.
LCC ML410.B773 P576 2016 | DDC 780.92–dc23
LC record available at https://lccn.loc.gov/2016021875
ISBN 978-1-107-06265-8 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Contents
List of Figures page [vii]

List of Contributors [ix]
Preface [xi]
Acknowledgements [xiii]
List of Abbreviations [xv]
Part I The Context of the Late 1940s and 1950s [1]

1 Pierre Boulez: Composer, Traveller, Correspondent
Edward Campbell [3]
2 Traces of an Apprenticeship: Pierre Boulez’s Sonatine
(1946/1949) Susanne Gärtner [25]
3 Schoenberg vive Jessica Payette [56]
Part II The Evolution of a Style [91]
4 ‘A score neither begins nor ends; at most it pretends to’:
Fragmentary Reflections on the Boulezian ‘non finito’
Robert Piencikowski [93]
5 Serial Organisation and Beyond: Cross-Relations of Determinants
in Le Marteau sans maître and the Dynamic Pitch-Algorithm
of ‘Constellation’ Pascal Decroupet [108]
6 ‘DU FOND D’UN NAUFRAGE’: The Quarter-tone Compositions
of Pierre Boulez Werner Strinz [139]
7 ‘Alea’ and the Concept of the ‘Work in Progress’
Peter O’Hagan [171]
8 Casting New Light on Boulezian Serialism: Unpredictability and Free
Choice in the Composition of Pli selon pli – portrait de
Mallarmé Erling E. Guldbrandsen [193]
9 Serial Processes, Agency and Improvisation Joseph Salem [221]
10 Listening to Doubles in Stereo Jonathan Goldman [246]
11 Composing an Improvisation at the Beginning of the 1970s
Paolo Dal Molin [270]
v
vi Contents
Part III Reception Studies [301]

12 Pierre Boulez in London: the William Glock Years Peter O’Hagan [303]
13 Tartan from Baden-Baden: Boulez at the 1965 Edinburgh International
Festival Edward Campbell [327]
14 Pierre Boulez and the Suspension of Narrative Arnold Whittall [354]
Bibliography [373]
Index [388]
Figures
1.1 Itinerary for the Renaud-Barrault tour to South America in

1950. page [6]
2.1 Pencil draft, Musée de la musique, Paris, p. 1 (detail). Photo Matthias
Abherve © Cité de la musique. [32]
2.2 Pencil draft, Musée de la musique, Paris, p. 5 (detail). Photo Matthias
Abherve © Cité de la musique. [32]
3.1 Textural quality and registral span of instrumental interludes in
‘Complainte du lézard amoureux’. [69]
6.1 Distribution of pitch cells in ‘Post-Scriptum’. [145]
6.2 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e. [153]
6.3 Deployment of the family of series belonging to ∫A16. [154]
6.4 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e; partial
transcription. [154]
6.5 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1f; partial
6.6 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e; partial
10.1 Sketch for seating plan of Doubles, transcribed and translated. Used with the
kind permission of the Paul Sacher Stiftung. [260]
10.2 Final seating plan of Figures – Doubles – Prismes, transcribed and translated,
with geometry of woodwinds and brass instrumental groups indicated.
Pierre Boulez Figures – Doubles – Prismes|für Orchester © Copyright 1964 by
Universal Edition (London) Ltd., London/UE 13994. [261]
10.3 Serial derivation of six segments (accords lents) of slow theme (thème lent).
10.4 Instrumental groups used for each constituent chord of the thème lent.
10.5 Transcription of Boulez’s sketch for spatialisation and superstructural
procedure in Doubles. Used with the kind permission of the Paul Sacher
Stiftung. [265]
vii
viii List of Figures
10.6 Reduction of first seven bars of first occurrence of fast theme in

Figures – Doubles – Prismes (rehearsal 3) and the instrumental groups
used for each of its constituent chords; pitches and durations only. Pierre
Boulez Figures – Doubles – Prismes|für Orchester © Copyright 1964 by
13.1 Concerts involving Boulez at the 1965 Edinburgh International
Festival. [332]
Preface
A composer studies volume in the Cambridge University Press series is

intended to be a compendium of scholarly contributions to the field of
study in question, and it is in this spirit that the current project has been
designed. Its fourteen chapters, written by twelve authors, many of whom
have already produced distinguished work in the field of Boulez studies,
comprise a series of substantial essays on a number of aspects of the
compositional and theoretical work of composer Pierre Boulez. The
contributors are all scholars in the field, a number having written Ph.D.
dissertations and monographs on the composer as well as producing author-
itative performing and facsimile editions of his compositions. Several others
are in the process of completing significant individual projects for publica-
tion, and all have produced studies of the composer in publications of
international stature.
The list of contributors brings together for the first time within
a single volume scholars working in the field of Boulez studies in
Belgium, Canada, France, Italy, Norway, Switzerland, the United
Kingdom and the United States. It also reflects a range of approaches
to the composer’s work. While several scholars focus primarily on what
we can learn from sketches, correspondence and other archive material,
held primarily in the Paul Sacher Stiftung in Basel, others bypass this
material and work directly from published scores, recordings and
writings. The editors have attempted to provide a balance between
these two approaches which are, broadly speaking, represented by
those contributors working on mainland Europe and those working in
the UK/North America. In doing so, our aim has been to ensure that the
results of a range of very recent European Boulez scholarship, which
has hitherto been inaccessible to an anglophone readership, is made
available for the first time within a single volume alongside new work
from UK and North American scholars.
In line with other Cambridge Composer Study volumes, the current
collection of essays makes no attempt to cover all of Boulez’s activities.
Indeed, the format is a fairly simple one, comprising fourteen chapters
grouped together under the headings ‘The Context of the Late 1940s and
1950s’, ‘The Evolution of a Style’ and ‘Reception Studies’. All of the
chapters are substantial and original contributions to Boulez scholarship
and approach the composer’s work from a variety of angles including
xi
xii Preface
study of his correspondence, his compositional practice, the key phases in

his compositional development, the mix of theory and practice that is
manifest in many of his most important works, the changing reception of
his work and assessment of his place in the development of twentieth-/
twenty-first-century music. While most chapters do not deal with a single
musical work, a great number but by no means all of Boulez’s composi-
tions are discussed to some extent in the course of the volume, with
prominent place being given to works from across his career, from the
early Sonatine to the middle-period Pli selon pli and late works such as . . .
explosante-fixe . . . There is also discussion of some of Boulez’s composi-
tional projects that, while not resulting in finished work, nevertheless
offer significant insight into the development of his compositional tech-
nique, and several authors approach Boulez’s practice of reworking
scores, in some cases multiple times. A number of chapters challenge
some of the commonplaces of Boulez reception, for example that his
Sonatine and First Sonata for piano were first produced in the format
we have them today and that he is a systems-obsessed composer for whom
freedom, choice, preference and the irrational are all alien concepts. Many
of the chapters are mutually illuminating and the authors are at times
engaged in a quasi- or virtual dialogue as they discuss the same works,
concepts or approaches in their own individual ways.
While volumes of essays on Boulez have appeared in French and German
language publications in recent years, the last single volume to contain
a number of significant essays in English was the book Pierre Boulez:
A Symposium (Eulenburg, 1986). This new, Pierre Boulez Studies volume
offers much more recent work from a completely different set of scholars and
stems from a growth in Boulez scholarship since the early 1990s, the results
of which are now being made available for the first time to the interested
reader as well as to the subject specialist.
Edward Campbell
Peter O’Hagan
Acknowledgements
We are especially grateful to Vicky Cooper, former music editor at

Cambridge University Press, who accepted the proposal for a volume of
Pierre Boulez studies and encouraged us in its production with enthusiasm
and patience. Thanks to Kate Brett, the current music editor at Cambridge
who has enabled us to complete the volume, and to Fleur Jones who has
helped us on many points of detail. All three have provided invaluable help,
responding to the text at each stage of its formation. We are also very
grateful to Sarah Starkey, content manager at CUP, who oversaw the pro-
duction of the book, to Andrew Dawes our copy-editor and to Velmurugan
Inbasigamoni and his team who typeset the book. They helped in innumer-
able ways to improve the volume. Thanks also to the two anonymous readers
who accepted the proposal and to the Cambridge University Press syndicate
who approved the contract.
We are most grateful to Pierre Boulez, who was generous in responding to
a number of enquiries at different points in the project and kindly
gave permission for us to cite from his correspondence and various other
archival sources. Grateful thanks also to the members of his Secretariat, in
particular Klaus-Peter Altekruse who generously facilitated these exchanges.
In addition, we wish to acknowledge our gratitude to: the staff of the Paul
Sacher Stiftung Basel, Director Felix Meyer, and most especially Robert
Piencikowski, Michèle Noirjean-Linder, Evelyne Diendorf and Johanna
Blask; the staff of the Bibliothèque nationale in Paris, especially Elizabeth
Giuliani, Laurence Decobert and Marie-Gabrielle Soret; the staff of the
British Library; the staff of the National Library of Scotland; the staff of the
BBC Written Archives Centre, especially Jeff Walden; the staff of the
Internationales Musikinstitut Darmstadt, especially Claudia Mayer-Hasse;
the staff of the Stockhausen Stiftung, especially Suzanne Stephens and
Maria Luckas; the staff of the Getty Research Institute, Los Angeles,
especially Virginia Mokslaveskas. Thanks are due also to all who responded
to enquiries we made of them: Gilbert Amy, Cécile Auzolle, John Carewe,
Frauke Jurgensen, Gillian Leach and the Edinburgh International Festival.
Thanks are due to the Carnegie Trust for a grant which made a number of
research trips possible.
Published English translations of texts have been used where available,
unless otherwise noted. All other translations, for example of correspon-
dence and texts, are by the authors, unless otherwise noted.
xiii
part i
The Context of the Late 1940s and 1950s

Edward Campbell
Pierre Boulez was, particularly in his early years, a great letter-writer and
a frequent correspondent. Among the many letters that have been preserved
and are now available for study in the Paul Sacher Foundation in Basel
and elsewhere, are some extended correspondences that provide invaluable
insight into his development as a composer, conductor and theorist. Perhaps
surprisingly, the only correspondence published to date in English is his
well-known exchange with John Cage, a series of fifty letters and other
documents dating, for the most part, from between 1949 and 1954.1 His
correspondence with musicologist and ethnomusicologist André Schaeffner
was published in France in 1998, covering their communications between
1954 and 1970.2 In the main, this constitutes the totality of Boulez’s corre-
spondence currently in the public domain, its limited scope giving no real
indication of the extent of his activity.
Among the correspondence held at the Paul Sacher Foundation, Boulez’s
lengthy exchange with Karlheinz Stockhausen is one of the richest, with
almost 200 items. It is unfortunate that this correspondence has not yet been
published, as it contains many points of great interest as the two young
composers exchange ideas, discuss the state of their compositions and
comment on the work and ideas of the other.3 One reason for the lack
of publication is the incompleteness of the correspondence as it stands.
As Robert Piencikowski has pointed out, there is a serious gap in our
knowledge of the letters Boulez ‘received between 1954 and 1959, the date
of his relocation to Germany’.4 The whereabouts of these letters is unknown
and, following information received from Boulez, Piencikowski relates that
‘the letters were mislaid’ during the composer’s change of domicile from
Paris to Baden-Baden in January 1959.
For those wishing to follow the trail of Boulez the letter-writer, also of
great interest is the correspondence between him and the Belgian composer
Henri Pousseur, probably the most prolific letter-writer of the post-war
1 3
Boulez and Cage, Correspondance et The original copies of Boulez’s letters to
documents. Stockhausen are held in the Stockhausen
2
Boulez and Schaeffner, Correspondance. Foundation in Kürten.
4
Piencikowski, ‘. . . iacta est’, p. 42.
3
4 Edward Campbell
generation. There are correspondences with Edgard Varèse from 1952

to 1965, with Igor Stravinsky between 1956 and 1966 and an extended
correspondence with poet René Char, mostly between 1948 and 1957.
Beyond the Sacher Foundation, Boulez’s letters to André Souris are held in
Brussels, while those to Sir William Glock are housed in the British Library.
One of the largest collections and, to my mind, one of the most important, is
Boulez’s correspondence with Pierre Souvtchinsky. Over 100 of Boulez’s
letters to Souvtchinsky dating from between 1947 and 1984 are held in the
Bibliothèque nationale in Paris while a smaller number of Souvtchinsky’s
letters are held in Basel.5
Three Tours of South America with the Compagnie

Renaud-Barrault
Acknowledging the importance of Boulez’s correspondence, this chapter
focuses on the three tours of South America the composer undertook with
the Renaud-Barrault theatre company in 1950, 1954 and 1956.6 While
a number of scholars have covered aspects of these journeys, no study to
date has made them the principal focus of attention. In the current chapter,
while information is gathered from a range of sources, from contempora-
neous documents and from later writings, Boulez’s letters are of the
greatest importance: the letters to Cage and Souvtchinsky in 1950, to
Cage, Souvtchinsky and Stockhausen in 1954 and to Stockhausen and
Souvtchinsky in 1956. Placing these letters side by side enables us to form
a vivid picture of the composer’s emerging preoccupations and concerns
during three finite and discrete moments of his early career.
Arriving in Paris in the autumn of 1943 to become a musician, Boulez
studied counterpoint with Andrée Vaurabourg, the wife of Arthur
Honneger, until 1945. He was a member of Olivier Messiaen’s harmony
class in 1944–5, graduating with a first prize in harmony. He attended classes
with René Leibowitz in 1945–6. Having done with formal study, and needing
to find paid employment, he was offered the position of musical director
with the newly formed Renaud-Barrault theatre company, a position he
held between 1946 and 1956, and in which his job consisted of conducting
theatre scores by composers such as Auric, Honneger, Milhaud, Poulenc
and Sauguet.7
5 6
The frequency of Boulez’s letters to First tour, 24 April to 28 July 1950; second
Souvtchinsky tails off at the end of the tour, 23 April to 16 August 1954; third tour,
1960s and there are only three written from 11 April to 23 June 1956 (dates recorded
communications from Boulez from 1972 in Bassetto, ‘Orient-accident?’, p. 103).
7
onwards. Jameux, Pierre Boulez, 1991, pp. 17–18, 82.
Not only did the Compagnie Renaud-Barrault perform in Paris, it also

undertook a great number of international tours. In the aftermath of
World War II, the Comédie-Française, the Théâtre National Populaire
of Jean Vilar and the Compagnie Renaud-Barrault were encouraged to
undertake a number of international tours as part of a cultural foreign
policy that used artistic productions, and in particular theatrical tours, to
propagate French culture and language. In a post-war and later Cold War
context, it seems that the most prestigious places to disseminate French
culture were the United States, the Soviet Union and Latin America.8
While the Compagnie Renaud-Barrault made visits to North America in
1952 and 1957 with Boulez as musical director, the main focus of this
chapter concerns the three tours of South America they made together in
1950, 1954 and 1956.
The First Tour of South America

Undertaken before the easy availability of long-distance air travel, the first
tour of South America commenced with the lengthy voyage across the
Atlantic. Setting sail on Le Florida from Marseilles, the company spent
fourteen days on board ship before arriving in Rio de Janeiro. A group of
thirty in all, including Boulez, travelled, taking with them eleven plays which
would be the basis of nine discrete theatrical programmes.9 In this first tour
of South America, which ran from 28 April to 28 July 1950, the company
visited Brazil, Uruguay and Argentina. The full itinerary can be seen in
Figure 1.1.
In a letter to Cage, sent from Rio de Janeiro sometime in May, Boulez
describes Brazil as ‘amazingly beautiful’ before going on to say: ‘in the midst of
all these distractions, I have a hard battle to safeguard my own work and not
submit to complete dispersal’.10 He regrets that he has ‘not met any young
musicians’, since the company has been ‘mainly moving in society circles’.
In his next letter, sent from São Paolo sometime in June, he tells Cage: ‘we may
have to extend our travels to Chile (we would cross the Andes cordillera by
plane, which would be wonderful!!) and thus extend the length of the tour by
a fortnight’.11 This possibility did not in fact come to fruition. He reflects: ‘it is
very nice to empty oneself completely and live in the midst of a fog of words
which one can’t understand – and that brings unexpected meetings’.12 Things,
8 11
See Falcon, Théâtres en voyage. Nattiez, Boulez–Cage Correspondence,
9
Barrault, Memories for Tomorrow, p. 62.
12
pp. 212–13. Ibid., p. 64.
10
Nattiez, Boulez–Cage Correspondence,
p. 58. Letter from Boulez to Cage, May 1950.
6 Edward Campbell
Itineraire
28 Avril: Départ Gare de Lyon
29 Avril: Embarquement à Marseille à bord du ‘FLORIDA’ des Transports Maritimes
14 Mai: Arrivée à Rio de Janeiro
17 Mai: Début à Rio au Théâtre Municipal
4 Juin: Fin de la saison à Rio
4.06. Malborough s’en va-t-en Guerre
On purge Bébé
5 Juin: Voyage Rio – Saô-Paulo
7 au 17 Juin: Saison à Saô-Paulo
18.06. Hamlet
20.06. Baptiste
18 Juin: Embarquement à Santos pour Montevideo à bord du ‘CAMPANA’ des Transports Maritimes
21 Juin: Arrivée à Montevideo
22 au 28 Juin: Saison à Montevideo au Théâtre SOLIS
29 Juin: Départ de Montevideo pour Buenos-Aires
30 Juin: Arrivée à Buenos-Aires
30 Juin au 17 Juillet: Saison à Buenos-Aires au Théâtre ODEON
12.07. Partage de midi
18 au 28 Juillet: soit prolongation en République Argentine soit saison à Santiago du Chili au Théâtre
Municipal
Le 29 Juillet embarquement à Buenos-Aires à bord du ‘FLORIDA’ des Transports Maritimes
Le 15 Août arrivée à Marseille
Fig. 1.1 Itinerary for the Renaud-Barrault tour to South America in 195013
however, were not going so well with composition and he adds: ‘with this
atmosphere of travelling, work has slowed right up. I am mainly orchestrating
old things. A task, after all, which requires less concentration than composition
proper. Nevertheless, I am not moving an inch away from my Mallarmé!’
By this, he is referring to the projected and later abandoned setting of Un coup
de dés, which he mentions elsewhere in the correspondence with Cage during
this period.14
Much of the discussion in the Boulez–Cage correspondence from the time
of the 1950 tour concerns Boulez’s hopes that he can accept an invitation
from Cage to visit the United States immediately after the tour, to participate
in a conference in Vermont. This additional trip was not however to be
realised, for the lack of a visa.15 Despite Cage securing a grant for Boulez, and
making strenuous efforts to contact him by letter and telephone, Boulez’s
next letter to him was written from the boat on the way back to Paris.16
Explaining his failed efforts to secure a visa in Buenos Aires, he tells Cage:
13 15
I am grateful to Peter O’Hagan for this See Cage’s letters to Boulez, 21 June 1950,
itinerary. Christina Richter-Ibáñez provides June 1950 (undated), 2 July 1950,
more exact dates for the stay in Argentina 26 July 1950; also Boulez’s letter to Cage from
which suggest that the itinerary as shown in late June/early July 1950 and Cage’s letter to
Figure 1.1 was not final. The company Souvtchinsky from after 18 July 1950.
arrived in Buenos Aires on 28 June and gave Nattiez, Boulez–Cage Correspondence,
the closing performance on 25 July. See pp. 65–70.
16
Richter-Ibáñez, Mauricio Kagels Buenos Undated. Estimated date July
Aires, p. 94. or August 1950. Nattiez, Boulez–Cage
14
Ibid., p. 62. Correspondence, pp. 71–2.
‘I am going straight back to Paris to work. I have a month or two ahead of me

to work quietly in the Rue Beautreillis. That will be an antivoyage!’
We have three letters from Boulez to Souvtchinsky dating from this
period, two of which were sent from Buenos Aires at the end of the tour.
In the first, he apologises for a silence of more than two months and tells his
friend that throughout the South American trip he has ignored all invitations
to embassies and consulates where his ‘presence [was] not indispensable’.17
He was not terribly interested in the musicians he met in Brazil who, while
very kind, were too interested in folklore and Brazilian rhythms. He
continues:
Some very beautiful promenades for Rio is a magnificent city in a setting that is no
less extraordinary. São Paolo is a city where you cannot take a step without running
into a skyscraper being built – of Montevideo a privileged city [‘privilégiée’] . . . the
impression is of a provincial city, very pleasant, very conventional and as static as
possible. But for São Paolo and Montevideo, these are only fleeting impressions for
with work, we have scarcely the time to visit.18
Consonant with his earlier letter to Cage, he stresses that whenever he sets
himself to complete a task for himself, he finds that he cannot get back to it
for two or three weeks. He has managed to complete a few pages of
orchestration and has been working on the structure of the Coup de dés,
but other than that ‘these continual changes of hotel room are not favourable
to withdrawing oneself completely’.19
He finishes the letter noting that he had seen a Brazilian macumba,
something which Barrault also remembered in his account of the tour.
Boulez writes:
some impressive hysterical states, but the rites and cults that are addressed to God, to
the devil, to the phallus or to the virgin, are always ineffectual rites and cults for their
own ends; I am more and more convinced that Artaud was on completely the wrong
track and that the Coup de dés contains the true magic, which leaves no room, even
for hysteria, hysteria being one of the most passive states, despite the paradox that
implies.20
To understand this statement it is important to recall that in the article

‘Propositions’ [‘Proposals’] from 1948, Boulez had called for a music that is
‘collective hysteria and magic, violently modern – along the lines of Antonin
17
Undated. Estimated date, after 28 June/ Barrault notes in his account of the 1950 tour
early July 1950 (BNF: NLA 393 (6) 40–2). See that ‘with guides, we went deep into the forest
Steinegger, Pierre Boulez et le théâtre, to witness macumbas’, and he mentions
pp. 45–6. arriving back ‘at 5am from an excursion in
18
Undated. Estimated date, after 28 June/ the forest. Our breastbones were still
early July 1950 (BNF: NLA 393 (6) 40–2). resounding from the tom-toms that had
19
Ibid.. beaten at us all through the night.’ (Barrault,
20
Undated. Estimated date, after 28 June/ Memories for Tomorrow, p. 214.)
early July 1950 (BNF: NLA 393 (6) 40–2).
8 Edward Campbell
Artaud’.21 Further, in 1958, in ‘Son et verbe’ [‘Sound and Word’], he

identified the organisation of delerium, again specifically related to Artaud,
as an imperative for ‘effective art’.22
In the second letter to Souvtchinsky from Buenos Aires, Boulez notes the
intensity of the company’s work on tour with ‘8 to 9 performances per week
not counting rehearsals’.23 He mentions having found ‘some interesting
people who direct a critical review, on painting, sculpture, architecture,
which is well made, well presented and inexpensive’. One of the young
people he met in the Argentine capital was the composer Mauricio Kagel,
who was involved as an extra, in the role of a rabbi in Kafka’s The Trial and as
a soldier at the end of the performance of Hamlet.24 Noting that Boulez
played the piano and celeste, operated the tape and conducted the music for
the performance, Kagel recalls that he already knew Boulez by name and that
they became friends. They would meet again in 1954.
While on tour, Boulez missed a performance in Paris of his Second Sonata
for piano and the first performance of Le Soleil des eaux on 18 June.25 In the
letter of 22 July, he thanks Souvtchinsky for the work he had done in
preparation for the première. He asks for news concerning ‘the rehearsals,
if the orchestra sounded well’, and writes: ‘send me your criticisms of the
work (above all on orchestral sonority). Were the three voices suitable? What
did our friends say about it?’26
Boulez is more open in stating his immediate aims to Souvtchinsky and, as
the tour draws to a close, the only thing he desires is ‘to get back as soon as
possible and to work on the Coup de dés’. He continues: ‘I’m thinking of
spending the end of August and the month of September in peace and quiet
in Paris, working in a very intense way. Ideas are coming to me in a very
precise way and I don’t want to let go of such capricious and delicate fruits.’
On the prospective trip to North America, he reacts with ‘Damn the
conference and damn the discovery of the New World. If I haven’t had
a response [in]27 three days, I’m giving up on it.’ In addition he notes: ‘I don’t
regret the voyage. I simply regret having seen almost nothing of the countries
where we played – in any case, nothing very authentic.’ In the third letter to
Souvtchinksy, he announces that they are taking their leave of South
America on the evening of 28 [July] and he expects they will be making
stopovers in Casablanca and Cadiz, noting that he is looking forward to
seeing Spain.28
21 25
Boulez, Stocktakings, p. 54. Le Soleil des eaux was conducted by Roger
22
Ibid., p. 43. Désormière and the soloist was Irène Jacob.
23 26
Undated. Postmark dated 22 July 1950 Undated. Postmark dated 22 July 1950
(BNF: NLA 393 (3) f19). (BNF: NLA 393 (3) f19).
24 27
Richter-Ibáñez, Mauricio Kagels Buenos Word is illegible in the text.
28
Aires, p. 98. Undated. Estimated date, late July 1950
(BNF: NLA 393 (6) f5).
The Second Tour of South America

A second tour of South America was undertaken from 23 April to
16 August 1954. Where Souvtchinsky’s continued place in Boulez’s life is
attested to by the seven letters he received during this time, we only have two
letters addressed to Cage, but five to Stockhausen. Boulez wrote to
Stockhausen around 22 April, the eve of the voyage, lamenting the great
amount of work he had to do at the time,29 and he also summarised this
workload succinctly in a later letter to Cage:
if you knew the work I have had this year! Arranging the four Petit Marigny concerts
was no small task. For I did absolutely everything from arranging the programmes to
hiring the instruments (not to mention such things as contacting artists or taking
care of lodgings). . . I don’t mind telling you that I am not keen to lose all my time as
I have done this year. Practically speaking, I have been able to do absolutely nothing
from December to April. At the end of April we went on tour. You can easily
imagine this season’s disastrous history as far as my work goes.30
Concerts apart, Boulez was also editing two journals, a volume of the
Cahiers de la Compagnie Madeleine Renaud-Jean-Louis Barrault and the sole
published number of the journal Domaine Musical.31 With all of this activity,
it is no surprise that he writes to Stockhausen on the eve of the tour:
I have been able to do practically nothing for myself; which makes me more than
nervous at the moment. And I have no great pleasure with the prospect of this
voyage to South America. Given that I’ve satisfied my curiosity a propos these
countries, I would prefer now to have some peace and quiet, to work – I hope to
pinch as much time as possible from the performances to finish the work for
Donaueschingen which has not moved forward since February.32
The work he is referring to is Le Marteau sans maître, and references to it are

threaded throughout the letters from this trip. He tells Stockhausen: ‘during
the voyage, I’m going to work very seriously on Le Marteau sans maître,
which [conductor Hans] Rosbaud must give on 16 October in
Donaueschingen’. He also informs Stockhausen that he is ‘taking some
extracts from the concert, on tape to present them in South America in
universities and various cultural organisations – as one says – with a little
preparatory lecture’. He complains, ‘I’m royally fed up with having practi-
cally no time to write’, adding with a barb, ‘Let’s hope that Brazil will inspire
me to the heights of Villa-Lobos and Milhaud.’ The mention of Milhaud
29 31
Undated. Date estimated by Robert A number of references to the Domaine
Piencikowski, 22 April 1954 (Paul Sacher Musical review are found in the letters to
Stiftung, Sammlung Pierre Boulez, hereinafter Souvtchinsky, Stockhausen and Cage from
‘PSS’). the time of this tour.
30 32
Undated. Nattiez, Boulez–Cage Undated. Date estimated by Robert
Correspondence, pp. 147–8. Piencikowski, 22 April 1954 (PSS).
10 Edward Campbell
refers not only to the time Milhaud spent in Brazil as secretary to Paul
Claudel,33 but also to the fact that Claudel’s play Christophe Colomb was one
of the works being performed, with music by Milhaud.
Most importantly in this letter, Boulez gives Stockhausen dates and
destinations for the tour.
5 May – 24 May Teatro Municipal Rio de Janeiro (Brazil)

24 May – 4 June Teatro Santana São Paolo (Brazil)
7 June – 24 June Teatro Solis Montevideo (Uruguay)
24 June – 14 July Teatro Odeon Esmeralda 367 Buenos Aires (Argentina)
15 July – 27 July Teatro Municipal Santiago (Chile)
The first extant letter from the tour was sent to Souvtchinsky from Rio de
Janeiro around 14 May,34 and Boulez notes that the tour commenced with
Molière’s comedy Amphitryon with music by Poulenc. In an undated letter
from São Paolo he writes of beginning rehearsals for Claudel’s Christophe
Colomb.35 Both letters are very much concerned with the review Domaine
Musical.
In a five-page letter written on 3 June from the Hotel Nogaro,
Montevideo, Boulez informs Souvtchinsky of the company’s arrival there
two days previously. Describing the city as ‘ugly as can be imagined [laide au
possible] and in the middle of winter. Rain, dead leaves, cold and tutti frutti!’,
he reflects nevertheless:
Fortunately, I’m going to be able to work a little more and a little better in this city.
For, in São Paolo, the rehearsals for Colomb were frightful, on account of the
incompetence of the choir they provided us with. Never having sung in French, and
not being professional musicians [‘spécialement musiciens’]. Some average
instrumentalists – except for one or two; some deplorable.36
Turning later in the letter to the question of composition, he writes:

on the subject of Le Marteau, I’m getting seriously back down to it. In São Paolo,
with the boring work with these minimally gifted choirs, with conferences and
people to right and to left, in an ultra-noisy hotel, I have not had the leisure to devote
myself very much to “fruitful meditations”! Here, in this little provincial city
[Montevideo], everything is perfectly tranquil.37
33 36
Paul Claudel had been French Ambassador Letter dated 3 June [1954] (BNF: NLA 393
to Brazil in 1917–18. (6) 10–14). The stay in Montevideo is
34
Undated. Postmark dated 14 May 1954 described as ‘boring’ in a later letter to Cage
(BNF: NLA 393 (6) 2–3). Boulez refers to (Nattiez, Boulez–Cage Correspondence,
a letter he sent to Souvtchinsky from Dakar p. 149).
37
but which has not survived. Letter dated 3 June [1954] (BNF: NLA 393
35
Undated. Estimated date, between 24 May (6) 10–14).
and 4 June (BNF: NLA 393 (6) f.6, f.7, f.8).
On a more positive note, he tells Souvtchinsky that he has given two

lectures in the School of Music in São Paolo titled ‘The Antecedents of
Music Today’ and ‘Recent Aspects of Musical Sensibility’.38 In a later
letter, to Stockhausen, he reveals that this was through the auspices of
the soprano Gabrielle Dumaine who was spending six months as professor
at ‘L’Escola libre de São Paolo’ [sic].39 He enthuses that he has found in São
Paolo ‘a really very interesting milieu; curious to know everything that is
new and au courant – if not through concerts, then at least with records of
all of the current scores’.40 This is equally the case with the painters and
poets he has met there and he notes: ‘I’m keeping in touch with them. For
they are going to publish a review “LYNX”, and have asked me to write
some texts, or at least translate the most recent ones I’ve written.’ He tells
of a ‘long discussion’ he had one Sunday afternoon in São Paolo with these
artists and intellectuals on Pound, Joyce and Cummings, recalling that
‘they spoke to me of Mallarmé’s Coup de dés as the greatest poem in the
French language, a level which had never been attained afterwards by any
French poet’.41
Describing the city as ‘the most passionate milieu in Brazil’, he reports that
his new friends gave him ‘the address of a related group in Buenos Aires’.42
Two of those whom Boulez met, the concrete or Noigrandes poets and
brothers Haroldo and Augusto de Campos, have written of their encounter.
Haroldo de Campos recalls that they met Boulez in São Paolo in 1954 at the
home of the painter Valdemar Cordeiro, ‘where we all enthusiastically
discussed Webern and Mallarmé’.43 Reflecting on the encounter in 1981,
de Campos remembers that Boulez was shown some early examples of
concrete poetry,44 Augusto de Campos’s Poetamenos which are printed
in various typographical arrangements, employ multiple colours and are
capable of multiple readings. Looking back, Haroldo de Campos sees
a relationship between the score of Boulez’s Third Sonata from 1957 and
his brother’s Poetamenos, both using ‘different colors to distinguish certain
alternative routes’. Interestingly, the poems, which date from 1953, were
stimulated by Webern’s use of Klangfarbenmelodie and, beyond these works,
38 42
Ibid. The lectures are titled ‘Antécédentes Undated. Date estimated by Robert
de la Musique Actuelle’ and ‘Aspects récents Piencikowski, 9 June 1954 (PSS).
43
de la Sensibilité Musicale’ [sic]. H. de Campos, Novas, pp. 171–2. Haroldo
39
Undated. Written on headed notepaper de Campos (1929–2003).
44
from Hotel Florida, São Paolo, Brazil. Haroldo de Campos describes concrete
The envelope has a Uruguayan stamp. Date poetry as ‘a new poetics, national and uni-
estimated by Robert Piencikowski, versal. A Planetarium of “signs in rotation”,
9 June 1954, Montevideo (PSS). whose point-events were called (like topo-
40
Letter dated 3 June [1954] (BNF: NLA 393 graphic indexes) Mallarmé, Joyce,
(6) 10–14). Apollinaire, Pound and Cummings, or
41
Ibid. Oswald de Andrade, Joao Cabral de Melo
Neto [et al.].’ H. de Campos, Novas, p. 171.
12 Edward Campbell
Augusto de Campos’s production, more generally, favoured a form of writ-

ing that fuses music and language, and he continued to work with
Webernian ideas throughout his career.45 This early addition of colour to
essentially Mallarméan typography is certainly noteworthy in the light of
Boulez’s use of similar means in his ‘Constellation’ (‘Constellation-Miroir’),
the centrepiece of the Third Sonata.
Beyond poetry, in a number of texts produced in the early 1950s by the
concrete poets of São Paolo, there are frequent references to sound, most
particularly to the new music of composers including Boulez and
Stockhausen. In the ‘Pilot Plan for Concrete Poetry’ (1958), the manifesto
of the concrete movement,46 amongst all their other references and sources
‘Webern and his followers: Boulez and Stockhausen; concrete and electronic
music’ are identified unambiguously as influences.47 While, unlike Boulez,
Augusto de Campos mixed ‘Viennese dodecaphonic theory and Brazilian
bossa nova swing’,48 it is not difficult to see the attraction for Boulez in these
Brazilian poets and artists who shared his enthusiasm for Webern and
Mallarmé. Indeed, Augusto de Campos’s book Música de invenção includes
a defence of Boulez from 1957 as well as his translation of Boulez’s ‘Homage
à Webern’.49
Returning to Boulez’s letter to Souvtchinsky of 3 June, we note in passing
a significant moment in Boulez’s growing self-awareness. Given that he had
not yet started to conduct beyond the theatre ensemble for the Renaud-
Barrault company, he now shares with Souvtchinsky that Jean-Louis
Barrault,
speaking to me about this and that, said to me that I should take up orchestral
conducting, that it could be useful to me; that Deso [Roger Désormière]50 needs to
be replaced – I admit that being part of a theatrical company and playing music for
the theatre no longer has much interest for me, and I now feel myself capable of
doing better. But is it worth the effort? We must speak about it again.51
Boulez wrote his first letter of the trip to Stockhausen around 9 June in
Montevideo,52 where the company had arrived two days previously.
In addition to many points made in the letters to Souvtchinsky, he discusses
what he describes as the ‘epistolary hermeticism’ of Stockhausen’s recent
45 50
See Bessa, ‘Sound as Subject’, pp. 219–36. Roger Désormière (1898–1963) conducted
46
Ibid., p. 220. The ‘Pilot Plan for Concrete the première of Boulez’s Le Soleil des eaux in
Poetry’ was written by Augusto de Campos, 1950. Suffering from a thrombosis, he
Haroldo de Campos and Décio Pignatari and stopped conducting in 1952.
51
first published in São Paolo in 1958 (H. de Undated. Postmark dated 3 June [1954]
Campos, Novas, pp. 217–19). (BNF: NLA 393 (6) 10–14).
47 52
H. de Campos, Novas, p. 218. Undated. Date estimated by Robert
48
Bessa, ‘Sound as Subject’, p. 222. Piencikowski, 9 June 1954 (PSS).
49
A. de Campos, Música de invenção.
communications on the subject of electronic music.53 Nevertheless, with

a September visit to Stockhausen and the electronic music studio in Cologne
in prospect, Boulez warns his friend that he can expect a ‘deluge of questions’
and that he ‘will not leave Cologne without a blinding light (ah! ah!) in [his]
brain’. He admits that he has only agreed to go to Darmstadt in the face of
insistent requests and because Le Visage nuptial will be performed there.
Despite having agreed to participate in the composition seminar, he tells
Stockhausen ‘there won’t be any course’ and ‘I will be happy to listen to what
the other two [Henze and Maderna] will say’. He also discusses certain
aspects of the third piece of his first book of Structures.54
In a letter from around 28 June and written on headed notepaper from the
Claridge Hotel, Buenos Aires, Boulez tells Souvtchinsky that he has been late
in replying to his last letter on account of the rehearsals for Christophe
Colomb. He remarks: ‘some very nice choirs; but amateurs who didn’t
know solfège! Nor French apparently.’55 Barrault, remembering the pre-
parations for two scheduled performances of Christophe Colomb on 14 July,
noted: ‘the work was hard. I can still hear Boulez yelling in time with the beat:
“Et merde, et merde, et mille fois merde!”’56
On the subject of his own composition, he relates: ‘I have been able to
work a little on “Le Marteau”. I’m preparing two pieces for Romanova which
she will have in around three weeks, I hope, if all goes well.’ Also, he has been
informed by the publisher Heugel that some of the instrumental parts for the
third piece of Le Visage nuptial are missing, a serious problem given the
imminent scheduling of the work for Darmstadt.57
In a letter to Cage, also from Buenos Aires, he discusses his work on Le
Marteau sans maître, explaining:
I am trying to go ever further and deeper, and also to widen my outlook. With the
two a cappella choral pieces I wrote last year, it is one of the works that has given me
the most trouble. I am trying to rid myself of my thumbprints and taboos; I am
trying to have an ever more complex vision – less visible and more worked out in
depth – I am trying to expand the series, and expand the serial principle to the
maximum of its possibilities.58
The choral work to which he is referring, Oubli signal lapidé (1952), though
unpublished, was performed once on 3 October 1952 in Cologne by the
53 57
The first number of the journal Die Reihe, Undated. Postmark dated 28 June 1954
edited by Herbert Eimert and Stockhausen, (BNF: NLA 393 (6) 17–19). This is the second
was published in 1955 and was devoted to of the three versions of the piece (1946–7,
electronic music. 1951–2 and 1985–9).
54 58
Undated. Date estimated by Robert Undated. Sent from the Claridge Hotel,
Piencikowski, 9 June 1954 (PSS). Buenos Aires. Boulez tells Cage he is in
55
Undated. Postmark dated 28 June 1954 Buenos Aires and will be there until 14 July
(BNF: NLA 393 (6) 17–19). (Nattiez, Boulez–Cage Correspondence,
56
Barrault, Memories for Tomorrow, p. 218. pp. 149–50).
14 Edward Campbell
Ensemble Marcel Couraud. Its continued importance stems principally from

the fact that it is the first work in which Boulez developed the principle of
pitch multiplication, that in Le Marteau sans maître and later works becomes
a significant way forward in the generation of pitch materials and, as he
himself says, in ridding himself of his thumbprints.
In a letter from July 1954, written on headed notepaper from the Hotel
Crillon in Santiago de Chile,59 Boulez tells Souvtchinsky that he has had
a letter from Heinrich Strobel of Südwestrundfunk (SWR) informing him
that
‘Le Marteau sans maître’ is in danger!! Because of the guitar. He wrote to me that all
of the guitarists he asked have refused! There are two reasons:
1. The part has not been copied according to the usual notation for guitarists:
an octave above how it sounds. . . that’s easily fixed. I need to redo a copy
quickly. I can’t understand Universal-Ed. not having thought of that,
without me needing to specify it. I’m afraid they may have copied the flute
in G without transposing it, or the xylophone. That would be almost all of
the material needing to be redone.
2. They claim they cannot play certain extremely high notes: there are only three
or four of them in the entire score; and they can correct them in any case.
I always wanted to ask advice of Ida Presti,60 and you know well that you like
me have never managed to arrange a meeting for that purpose. If there are
some instrumental checks to be made, I couldn’t ask for better than to work
with the instrumentalists.
[. . .] if the German guitarist persists in his refusal, I must find someone in Paris
who is willing to study the part, who can be available for Südwestfunk from the
morning of 8th October to the evening of the 16th. Incidentally, I said to him that if
I had found no-one fifteen days after my return, I would give up.61
Giving the name of an amateur guitarist of the name Aubin, Boulez notes
that, if this guitarist is able to participate in the première of Le Marteau, the
part will be available from 8 August.
Beyond these problems, Boulez disloses that he has still not completed
Commentaire III from the ‘Bourreaux de solitude’ cycle in the work, adding:
the two final sung pieces are not even begun. I want to do one thing: As I no longer
have any interest in going to Darmstadt (and, in addition to that, I’ve learned that
Leibowitz is going), I’m going to withdraw and spend around ten quiet days in Paris
to finish Le ‘Marteau’ before the end of August. It will still be more than a month
before the rehearsals.62
59 60
Undated. Estimated date between 15 and Ida Presti (1924–67) was a French classical
28 July 1954, sent from Hotel Crillon, guitarist.
61
Santiago de Chile (BNF: NLA 393 (6) 21–3). Undated. Estimated date between 15 and
Boulez tells Stockhausen in an undated letter 28 July 1954 (BNF: NLA 393 (6) 21–3).
62
(from around 4 August 1954, sent from Ibid.
Bahia) that he had been in Santiago from 15
to 28 July.
He adds:
I must tell you that I’m awaiting the end of the tour impatiently in order to work
freely! I’ve had more than enough of starting again in each town with the same
idiotic and mind-numbing work on this crappy Christophe Colomb. But I’m
rambling on for I already said that to you in my last letter. Nevertheless, this loss of
time, at my age, is beginning to obsess me to a point.63
In a second letter to Souvtchinsky, also from Santiago de Chile, and

presumably written only a few days later, Boulez writes:
As for Le Marteau sans maître, it’s not moving forward. . . and for a good reason.
The work has been intense in Buenos Aires. Ch. Colomb in particular. Many
performances, rehearsals – some conferences, some editing to do in view of these
conferences. In short nothing nothing nothing! I’ve merely sent my instrumental
piece to Schlee;. . . I’m driven round the bend [‘Je suis damné moi-même’].
I envisage doing two of them fairly soon. For at last this Colombien [sic] nightmare is
finishing and I’m going to be able to work again.64
Towards the end of the letter, he writes: ‘I would like to have three months in
front of me – just for work; I would finally make some progress’; and, ‘I think
I’ve had enough of the theatre and of the time I’m losing on it. I must find
something else. We’ll speak about it when I get back. But what?’65
In a letter to Stockhausen with postmark Santiago, Chile, 20 July 1954,
Boulez relates the same difficulties with the choirs and with finding time for
composition on Le Marteau. He tells also of having met a group of young
people in Buenos Aires who were ‘very keen on everything new’. While he
has told them all about recent musical research, there is no great enthusiasm
in his tone.66 He also met Mauricio Kagel for the second time, and the young
Argentine composer who was participating as an extra in the performance of
Christophe Colomb took the opportunity to show him ‘his Variations and
parts of the incomplete first version of what was to become the String
Sextet’.67 According to Kagel, it was Boulez who convinced him to leave
Argentina and go to Europe and, having failed to win a scholarship to study
musique concrète at the Club d’Essai in Paris, he took Boulez’s advice to go to
the electronic studio of the WDR (Westdeutscher Rundfunk) in Cologne,
a decision that was realised thanks to the receipt of a Deutscher
Akademischer Austauschdienst scholarship. Boulez’s recommendation of
Cologne is particularly interesting in the light of his own intended visit
63 67
Ibid. Heile, The Music of Mauricio Kagel,
64
Ibid., 25. pp. 14–15; ‘There Will Always Be Questions
65
Ibid. Enough: Mauricio Kagel in conversation with
66
Undated. Postmark dated 20 July 1954. Max Nyffeler’, www.beckmesser.de/neue_
Written on headed notepaper from Hotel musik/kagel/int-e.html, accessed
Crillon, Santiago de Chile (PSS). 15 March 2016; Trubert, ‘Les “franges aux
limites indécises”’.
16 Edward Campbell
there in September 1954 to work with Stockhausen on some pieces of

electronic music.
Christina Richter-Ibáñez provides a great deal of detail on Boulez’s activities
in Buenos Aires in 1954. She reports that while the company failed to summon
the enthusiastic response of the previous trip, Boulez received greater attention
in 1954 on account of his appearance at Darmstadt and his essays, most
especially ‘Schoenberg is dead’. An interview with the title ‘Pierre Boulez,
el músico más moderno de Francia, habla para “Buenos Aires Musical”’ was
published in the Buenos Aires Musical on 15 July. In the course of the inter-
view, Boulez discussed the importance of performance and participation in
public discourse for the young composer in search of an aesthetic. He spoke of
Webern, dodecaphony and serialism, of his break with Leibowitz and his
respect for the Cologne electronic music studio, while explaining aspects of
the work of Stockhausen, Messiaen, Cage, Fano and Nono.68
Kagel notes that ‘at every free moment, Boulez ran to his room in the
Hotel Claridge, to work on the last numbers of “Marteau sans maître”’.
Despite his need to work at composition, he nevertheless took time to meet
composers including some members of the ANM (Agrupación Nueva
Musica). Richter-Ibáñez tells us that Kagel met separately with Boulez,
showing him ‘the finished score of the “Quarteto mixto”’ as well as ‘the
completed “String Sextet” and some numbers from “Música para la torre”’.69
Boulez also gave a lecture at the Galerie Krayd, where he spoke on the
background of current music and introduced the audience to some of
the most recent developments, with the help of the sound recordings from
the Petit Marigny. Kagel relates that Boulez spoke in French ‘about Cologne
and the newest developments in electronic music’ and Juan de Prat Gay,
a member of the ANM, penned an article at the end of the year for the
journal Sur on Boulez’s contemporaneity. Despite reprising a certain num-
ber of polemical positions, Boulez focused primarily on new conceptions of
rhythm, showing the young composers something of his most recent work
on ‘L’Artisan [sic] furieux’.70 Indeed, de Prat Gay notes that ‘those of us who
followed Boulez closely during his stay in Buenos Aires, assisted in some way
in the creation and analysis of his work, which he himself undertook during
the two classes he taught, with generosity and apostolic enthusiasm, to
a group of young Argentinians’.
In a letter to Stockhausen, sent from Bahia most likely around 4 August,
Boulez states in definitive terms that he has decided not to go to Darmstadt
since Le Visage nuptial will not not be performed there, because he wishes ‘to
finish the first part of Le Marteau’ and because he wishes to avoid meeting
68 69
Richter-Ibáñez, Mauricio Kagels Buenos Cited in Richter-Ibáñez, ibid., p. 159.
70
Aires, p. 158. Ibid., p. 160.
Leibowitz.71 Informing Stockhausen that he had been in Santiago from 15 to

28 July, he notes ‘a great curiosity in this country from people who are
usually questioning [‘douteux’] – for musique concrète [and] electronic
music’. Having returned to Buenos Aires from Santiago, the company then
travelled to Bahia and Recife for a ‘well-deserved rest’ before their eventual
arrival in Marseilles on 16 August.
As in 1950, Boulez and the company witnessed a Candomblé ceremony,
a form of macumba, during their stay at the Brazilian port of Salvador de
Bahia.72 Peter O’Hagan has studied the impact that the Candomblé had on
Barrault’s conception of L’Orestie,73 a production for which Boulez unu-
sually not only conducted but also composed the music. Drawing attention
to ‘the percussion writing . . . in Boulez’s score’, O’Hagan notes certain
similarities between Boulez’s rhythmic patterns and transcriptions of
Candomblé music in the appendix to Gisèle Binon’s study of the genre,
and he suggests that ‘it is almost as though much of the percussion writing is
a Boulezian stylisation of his experiences of these ceremonies in Brazil, in the
company of Jean-Louis Barrault’.74
In his final letter to Stockhausen of the tour, sent from Dakar around 9–11
August, but written on the boat the day before their arrival there,75 Boulez
announces: ‘I am already back at work on Le “Marteau sans maître”.’
Recalling the recent plane journey to Bahia and Recife, he says: ‘we feasted
our eyes on exotic landscapes’, and ‘I’ve brought back a haul of “exotic”
instruments: wooden bells, double bells made of iron [‘cloches doubles en
fer’], Indian flute, little Indian guitar, frame drum, bells [‘grelots’], Jew’s harp
[‘birimbao’] (a very curious instrument from Bahia, but of African origin)’.
He continues: ‘after that, it’s sad, we had to leave . . . and in driving tropical
rain’.
In a letter to Souvtchinsky from Dakar, Boulez writes: ‘We had eight days
in Bahia and Recife that were absolutely astonishing [. . .] the north of Brazil
is absolutely captivating. These eight days have been the well-earned reward
[‘recompense de bien’] for the Colombian [‘colombiennes’] trials and
tribulations! . . . On the boat, I went back to Le Marteau.’76 He continues:
I don’t know if I told you, but finally I’m not going to Darmstadt. It annoys me and is
no longer of any interest . . . [I’m] staying in Paris from 16 August to 28 August, right
71 74
Undated. Date estimated by Robert Ibid., p. 46.
75
Piencikowski, 4 August 1954, Bahia (PSS). Undated. Sent from Dakar, Senegal with
72
A detailed account of Barrault’s impres- the postmark 9–10, 11 August 1954 (PSS).
sions of the Candomblé and its importance Boulez tells Stockhausen that they will be
for the production of L’Orestie is given in arriving in Dakar the following day.
76
Barrault, Theatre of Jean-Louis Barrault, Undated. Sent from Dakar, Senegal with
pp. 65–9. the postmark 11 August 1954 (BNF: NLA 393
73
O’Hagan, ‘Pierre Boulez and the Project of (6) 26–7).
L’Orestie’, pp. 38–40.
18 Edward Campbell
up to the moment of leaving for Cologne. I’ve informed Steinecke today. . . And I’ll
have some peace and quiet after so many months wandering.77
On his return from South America, Boulez did not go to Darmstadt; he

continued working on Le Marteau sans maître and visited Stockhausen in
Cologne in September 1954. While it is beyond the scope of the present study
to trace the gestation of Le Marteau any further, we can note that the
Donaueschingen première of the work, set for 16 October under Hans
Rosbaud, was cancelled since, as Boulez puts it in a letter from 20 September:
‘sad news: it’s impossible to play le “Marteau” in Donaueschingen. The guitarist
has run off [‘s’est defilé’]. It’s impossible to find a suitable [one].’78 The première
of the work was finally given in Baden-Baden on 18 June 1955 with performers
from the Südwestfunk Orchestra of Baden-Baden, conducted by Hans Rosbaud,
and with mezzo-soprano Sybilla Plate. An extended account of the occasion is
given by Boulez in letters to Souvtchinsky79 and Stockhausen.80 The first French
performances of Le Marteau took place on 21 and 22 March 1956, with soloists
from the Domaine Musical and the mezzo-soprano Marie-Thérèse Cahn.
Boulez was the conductor, making his debut in a concert programme that
also included works by Webern, Nono and Stockhausen.81
The Third Tour of South America

The third tour of South America took place from April to June 1956. Boulez
wrote to Stockhausen before his departure, in a letter postmarked 1 April
that sums up some of the tasks in hand at the time.82 It is only nine days or so
after the concert in Paris in which Boulez first conducted Le Marteau, and he
ventures the opinion that it went well. Both performances have been
recorded from which a montage is to be formed. There is the second volume
of the journal Die Reihe to think of, a volume dedicated to Webern, for which
Boulez is due to write an article on the Second Cantata. On the subject of
composition, he is ‘working on something for piano with “formants” which
replace thematicism. And the form is in constant evolution with undefined
zones between homogeneous zones. I will play it for the first time in
Darmstadt this year.’ He is referring to the Third Sonata for piano.
77 81
Ibid. The programme was as follows: Webern:
78
Boulez to Souvtchinsky. Postmark dated Symphony; Nono: Incontri; Stockhausen:
20 September 1954 (BNF: NLA 393 (6) 30). Kontra-Punkte; Webern: Two Songs op. 8,
79
Undated. Sent from ‘Osterreich’ after the Four Songs op. 13; and Boulez: Le Marteau
première of Le Marteau on 18 June 1955, sans maître.
82
possibly in July (BNF). Undated. Postmarked Paris, Gare St
80
Undated. Postmark dated 11 July 1955 Lazare, 1 April 1956 (PSS).
(PSS).
Boulez next wrote to Stockhausen around 17 April from Mexico,83 by

which time the montage recording from the concert had been made. As with
the 1954 tour, he includes dates and locations as follows:
Up to 2 May Teatro Belles Artes – Mexico D.F. (Mexico)

7 May – 11 May Teatro Municipal – Lima (Peru)84
16 May – 19 May Teatro Nacional Sucre – Quito (Ecuador)
22 May – 25 May Teatro Colon – Bogota (Colombia)
29 May – 23 June Teatro Municipal – Caracas (Venezuela)85
As Barrault notes, this time they travelled by plane, ‘the Air France Super-
Constellation called Le Parisien Spécial’, departing from Orly and travelling
on to Mexico by way of New York.86 The itinerary of Mexico, Peru, Ecuador,
Columbia and Venezuela was a new one, and Barrault tells us that ‘in the
course of the journey [he] extended the adventure to take in the Caribbean.
Return via Puerto Rico, on the good ship Antilles.’
In two postcards to Stockhausen from Lima, around 7–11 May,87 Boulez,
referring to his Third Sonata for piano, notes:
I have begun to work again on my new work for piano. But I’m interrupted
constantly. I’ve seen some wonderful things in Mexico, where, when I was free,
I travelled a lot.
In Peru, where I’ve just arrived, I’m going to spend four days in Cuzco to see the
Inca civilisation a little more closely. While regretting not being able to go as far as
Lake Titicaca and to Tichuanaco.
That consoles me when playing insipid and unimportant music. Culturally
speaking, only Venezuela is of some interest.88
In a further postcard to Stockhausen from Peru, from around the same time,
he writes:
‘The Archeology’ makes up here for the lack of music. I have never seen so many
things in such a short time. Coming here, the plane journey is magnificent. One
towers over the valleys of the Andes. I’ll have to tell you about it when I get back.
I’m staying here for another three days, to see some marvels, and suitably far from
the theatre and from its tiresome incidental noises.89
83 87
Undated. Postmarked Mexico, D.F., Two undated postcards, sent from Lima,
17 April 1956 (PSS). A previous letter sent to Peru. Date estimated by Robert
Stockhausen during the voyage has not Piencikowski, between 7 and 11 May 1956
survived. (PSS).
84 88
According to Barrault, the company stayed Undated. Date estimated by Robert
for nine days in Lima. See Barrault, Memories Piencikowski, between 7 and 11 May 1956
for Tomorrow, p. 227. (PSS).
85 89
Undated. Postmark dated 17 April 1956 Undated postcard sent from Cuzco, Peru.
(PSS). Date estimated by Robert Piencikowski,
86
Barrault, Memories for Tomorrow, p. 225. between 7 and 11 May 1956 (PSS). See
Barrault, Memories for Tomorrow, p. 228.
20 Edward Campbell
As before, Boulez continued to correspond with Souvtchinsky. Writing

from Lima, he confesses ‘this journey is weighing on me a little’ and, as in
his letter to Stockhausen, he notes that he has made touristic visits to
certain archaeological sites, adding for Souvtchinsky his need ‘to find
something else’ to do since his current mode of employment is ‘absolutely
uninteresting’ to him.90 The letter concludes: ‘after archaeology, I’m get-
ting back to composition. I still have to see Machupichu. Long live archae-
ology! It’s better than the theatre.’ In a postcard featuring the image of an
Inca ruin, evidently sent around the same time, Boulez, as in 1954, reveals
to Souvtchinsky what he is really thinking and feeling about life in the
theatre and his future direction:
Have a look at this card. It will show you the rhythm of my breathing when I’m
alone! I’m here for two days completely alone, face to face with that. I’m ventilating
myself for the future [‘Je m’oxygène’]. And all of the old connections are going to fall.
I’ve practically decided to no longer continue this dreadful profession as a purveyor
of theatrical noises [‘fonction théâtral à bruits’]. Sanctuary in these places has
dispelled my edginess but strengthened my resolutions. We will speak about it in
Paris.91
In the letter from Lima, Boulez praises Souvtchinsky’s article on Le

Marteau sans maître from the May edition of the Nouvelle Revue
Française, writing ‘I’m very pleased that something important has been
said about this work’,92 and on the question of the forthcoming release of
the partial recording of the work, he has decided that, for reasons of space, it
should include:
‘Pièce 2 Commentaire I au Bourreaux de solitude
Pièce 3 L’artisanat furieux
Pièce 6 Bourreaux de solitude
Pièce 9 Bel édifice et les pressentiments, double’.
He also adds that ‘if it’s too long, remove piece 2’, which is exactly what
occurred since the recording featured only pieces three, six and nine. A letter
sent from Quito in Ecuador around 16 May has Boulez ask Souvtchinsky to
forward him the proofs for the soon-to-be-published score of Le Marteau,
either to Bogota before 25 May or after that to Caracas, since Universal
Edition have mistakenly sent the proofs to Paris.93
90 92
Undated letter. Written on headed note- Undated. Estimated date, between 7 and
paper from ‘Gran Hotel Bolivar’. Estimated 11 May 1956 (BNF: NLA 393 (7) 12–14).
93
date (from information given to Undated. Written on headed notepaper
Stockhausen), between 7 and 11 May 1956 from ‘Hotel Humboldt, Quito – Ecuador’.
(BNF: NLA 393 (7) 12–14). Postmark dated 16 May 1956 (BNF: NLA 393
91
Undated. Estimated date, between 7 and (7) 17).
11 May 1956 (BNF: NLA 393 (7) 16).
A wonderful letter to Souvtchinsky from Caracas gives a rare insight into

Boulez’s growing self-discovery as a conductor,94 on the occasion of his first
experience of conducting a full orchestra, the Venezuelan Symphony
Orchestra at the Caracas Municipal Theatre on 16 June 1956.95 According
to Jean Vermeil, the Cuban writer Alejo Carpentier, who was a friend of
Barrault, instigated the occasion, having previously failed to bring Boulez to
Caracas in 1954 for the first Grand Festival of Latin American Music.
As Boulez later told Joan Peyser, ‘to conduct so far from home is not
dangerous’.96
In the letter, Boulez tells Souvtchinsky he has not written for a while
because he was preparing the concert and a lecture he was giving at the
piano.97 He judges that the concert ‘went well’ while acknowledging that he
had to change the programme on account of the unavailability of certain
scores, Schoenberg’s Five Orchestral Pieces op. 16, his op. 31 Variations and
an unnamed piece by Bartók. The programme consequently was as follows:
Debussy: Jeux; Prokofiev: Symphony no. 1 ‘Classical’; Stravinsky:
Symphonies of Wind Instruments; Debussy: Ibéria.
Noting that it was the orchestra’s first performance of Debussy’s Jeux,
Boulez writes that those who heard it were very impressed and that he
achieved a certain flexibility. Despite his ambivalence towards Prokofiev’s
‘Classical’ Symphony, he enjoyed playing it ‘with irony and whipping it up
a bit’. The success of Stravinsky’s Symphonies was mixed. The brass per-
formed well in their breathing, they captured the hieratic nature of the
chords and the rhythmic passages worked well. The two dialogues featuring
the flute and second clarinet, however, were ‘troublesome’. He judges that
Debussy’s Ibéria is ‘relatively easy’ except for the second movement,
continuing:
I conducted practically by heart, despite learning Ibéria and the Prokofiev at the
last minute. The orchestra was very pleased with me, the public also – small in
number but welcoming – I think I played in a serious way; and I think, perhaps, I’ll
come back next year for the Schoenberg and Bartók and also Stravinsky (The Rite or
The Nightingale).
As for my own impressions? To be honest, I feared the test of the large orchestra.
I feel relieved of a big question mark. I can conduct the large orchestra without
difficulty. I questioned conducting ‘Jeux’ in particular; I’ve done it and I pulled it off
without doing any harm. I’m now thinking of accepting the engagements that come
94 96
Undated. Postmark dated ‘1956’ (BNF: Vermeil, Conversations with Boulez,
NLA 393 (7) 3–6). The date ‘23.06.56[?]’ is pp. 142–3; Peyser, ibid., p. 133; Barrault,
appended at the top of the letter. Estimated Memories for Tomorrow, p. 232.
97
date, between 17 and 26 June 1956 [E. C.]. Undated. Estimated date, between 17 and
95
Peyser, Boulez: Composer, Conductor, 26 June 1956 (BNF: NLA 393 (7) 3–6).
Enigma, pp. 132–3. Griffiths in Vermeil,
Conversations with Boulez, pp. 179–80.
22 Edward Campbell
along, I have the impression that I’ll be able to rise to the challenge. And I like that.
To be able to form the sonorities of the large orchestra, and to reconstruct a work in
this space, is exciting (though the repertoire isn’t). After the last concert at the
Marigny and this one, I have more confidence in myself for conducting. I believe
I know how to communicate with the orchestra.98
In addition to the concert, Boulez reports giving a lecture-concert at

which the attendance wasn’t too bad and he notes the presence of
a number of young people. He played Webern’s Variations for Piano op.
27, three of Stockhausen’s piano pieces including one which, as he tells
Souvtchinsky, he had to ‘re-transcribe completely from a rhythmic point
of view for it was impossible to think!’, and finally the second and third
movements of his own Second Sonata for piano. He admits, ‘I worked on the
piano every afternoon so as not to undermine the performances; in fact the
works haven’t suffered greatly.’
Boulez announces that he is leaving Caracas on 26 June, to spend four
days in Martinique as a tourist before returning to Paris on 1 July.99 Beyond
touristic pursuits, the final letter of the trip, sent to Stockhausen from
Martinique around 30 June, shows him to be fully focused on his future
projects. He will be in Darmstadt from 9 July, keen to hear Stockhausen’s
new electronic piece, presumably Gesang der Jünglinge, which was to be
performed there on 19 July.100 Given that Stockhausen will now have heard
Boulez’s Domaine Musical recording of Kontrapunkte and, thinking ahead
to concert programmes for the following year, Boulez suggests that he per-
form and record Stockhausen’s quintet [Zeitmasse (1955–56)] the
following year in Paris.101 But it is in the previous letter to Souvtchinsky
that he probably sums up his feelings best. He writes that he hopes ‘to return
from Darmstadt immediately and to remain in Paris for three months
without moving, to be able to work to full capacity. I have had enough
holidays with this voyage.’102
Conclusion
In the course of an interview in 2013, Peter O’Hagan asked Boulez about ‘any
musical influence’ he was conscious of during his travels in South
America.103 While the composer does not differentiate between the three
98 101
Ibid. Undated. Postmarked Martinique,
99
Ibid. After Caracas, Barrault reports that 30 June 1956 (PSS).
102
he had managed to arrange performances in Undated. Estimated date, between 17 and
Guadeloupe and Martinique. See Barrault, 26 June 1956 (BNF: NLA 393 (7) 3–6).
103
Memories for Tomorrow, p. 235. The interview was conducted on
100
Stockhausen completed Gesang der 17 June 2013.
Jünglinge in 1955–6.
tours, his response is perhaps the most interesting information we have on

how the experience may have influenced his compositional work. He
responds:
In Rio di Janeiro there was no influence, because the popular music was so trite that
you could hear it is a kind of divertimento. You couldn’t take that seriously – it was
pleasant, you could see people dancing – that’s normal life, everyday life. The only
thing that impressed me was in Chile, the music of the peasants, because we went
there in a car with a man who was a journalist, and he brought us to the country for
one day. I liked the songs and the sound of the harp – that was so unique – and also
the high register on the piccolo/flute with air [blows air]. The sound between the
harp, flute and piccolo – that was really something. I took it in Pli selon pli, where
there are four piccolos and three harps: this sonority comes from Peru directly.
I think it was completely unknown, so there was no danger of it being imitated
immediately, so I used this sonority.
But otherwise, the only thing I used was the percussion of the music of the
candomblé, because it was mainly percussion, very impressive, and all the ceremony
was very impressive, when you saw people who were very heavy, turning like mad,
stopped, because the man who was in charge of the ceremony stopped, and was
suddenly normal: there was a kind of exhalation, and after that – finish. You saw all
the children going through the ceremony, not understanding quite a lot, but they
were not disturbed by it. The candomblé was the thing that was most impressive,
a mixture of sound: the excitement of the percussion, and then when there was
a calm moment, it was always with voice – the contrast between percussion/voice,
like psalms. The most fascinating thing – we met with Barrault, who was asking
about the sources, and he told us that now one can reconstitute the same regard
geographically, because all the ceremonies of the candomblé don’t use the same
language, and the language is forgotten. That’s like for most people, when they
attend a ceremony in a Latin church, when they have the prayers in Latin, it doesn’t
mean anything to them because they don’t understand Latin. Similarly there, they
don’t understand the language of the ceremony, and through that you can make
progress in geography to see that the ceremony was from this part of Africa to this
part of America, and so on and so forth. And so something which was not at all
scientific can lead to scientific discovery.104
After the 1956 tour, no longer bound by his engagement with Barrault,
Boulez was involved in a series of projects. He continued to compose, to lead
the Domaine Musical concerts as conductor, and to write essays setting out
his evolving musical position. The letters from the tours of South America
show a young Boulez, conflicted between his own compositional work and
his duties as musical director, rehearsing and conducting music he didn’t
value. In the course of the letters, we discover something about the circum-
stances in which some key compositions developed, the aborted Coup de dés
project in 1950, Le Marteau sans maître in 1954 and the ongoing work on the
104
Cited in O’Hagan, Pierre Boulez and the
Piano. I am grateful to Peter O’Hagan for
sharing this information with me.
24 Edward Campbell
Third Sonata for piano in 1956. We learn something of the competing

demands on his time as by 1954 he not only juggles composition, musical
direction and the writing of articles, but also the challenge of setting up what
would become the Domaine Musical as well as preparing two edited volumes
for publication. By 1956, we see the beginnings of Boulez the conductor with
the Paris performance of Le Marteau and the Caracas orchestral perfor-
mance. The fact of having spent a significant amount of time in South
America and of having experienced so many cultures at close hand, is of
great interest in its own right, and there is still work to be done here in
showing more precisely how the music of Boulez, far from being the product
of some technocratic force, is irrigated with ethnic musical traditions from
around the world and the sounds he first encountered with the Compagnie
Renaud-Barrault.
2 Traces of an Apprenticeship: Pierre Boulez’s
Sonatine (1946/1949)
Susanne Gärtner
Pierre Boulez’s Sonatine requires no introduction: performed throughout

the world, it has come to be recognized as a standard work for flute and
piano of the twentieth century. Furthermore, written in January and
February 1946 towards the end of his studies, the Sonatine also marks a
key moment in the development of Boulez’s musical language, being the
first major work in which the twenty-year-old attempted to synthesise the
diverse influences of his apprenticeship. The Sonatine is, so to speak,
Boulez’s ‘Opus 1’.
Interest in the structure of the piece has been considerable since its
publication. However, analysis has proved problematic for three principal
reasons:
1. The Sonatine was not published until 1954 and had its official première at
the ‘Internationale Ferienkurse für Neue Musik’ in Darmstadt in 1956, at
a time when Boulez was already known as a dodecaphonic and serial
composer. Consequently, commentaries and studies focused mainly on
pitch analyses with only partial success; apart from passages which
seemed to illustrate Boulez’s brilliant and direct reaction to the style of
Anton Webern, large parts of the piece could not be explained easily in
terms of twelve-note theory.1
2. Boulez commented repeatedly on the Sonatine. Interviewed by Antoine
Goléa in 1958 he called it his first attempt to combine classical twelve-
note technique with independent rhythmic structures such as Olivier
Messiaen had demonstrated.2 Boulez also mentioned Arnold
Schoenberg’s first Chamber Symphony as a formal model. Apart from
1
For the problems of a dodecaphonic comprehensive dodecaphonic analysis see
approach see Bradshaw and Bennett, ‘In Chang, ‘Boulez’s Sonatine’.
2
Search of Boulez’, 11–12; for a response and a See Boulez in Goléa, Rencontres, pp. 38–9:
first partial analysis see Baron, ‘An Analysis’; ‘This Sonatine is my first stage on the path of
for emphasizing Boulez’s brillant handling of serial composition such as I understand it . . .
the newly acquired technique see Goléa, You could say that it is in this Sonatine that I
Rencontres avec Pierre Boulez, pp. 37 and 45; tried for the first time to articulate independent
see Griffiths, Boulez, p. 10; and Jameux, Pierre rhythmic structures, the possibilities of which
Boulez (trans.), pp. 22 and 227; for the Messiaen had revealed, in the context of
reaction to the style of Webern see Bennett, classical serial structures.’
‘The Early Works’, p. 57; for the attempt at a
25
26 Susanne Gärtner
form, however, he had tried to avoid any stylistic influence. It was in this
context that he later spoke of his approach to models as ‘une espèce de
dissociation chimique’.3 In his Collège de France lectures 1978–88 he
discussed the Sonatine in relation to athematicism and the virtual theme.
The piece is here described as an initial effort to confront thematicism
and athematicism alternating between different states.4 Taken as whole,
Boulez’s commentaries, whilst illuminating, are rather selective and tend
to retrospective systematisations.
3. The fact that the Sonatine had been revised before its publication was
generally disregarded by scholars. Questioned on this topic by Goléa,
Boulez had contributed to the misunderstanding, declaring that he had
changed only ten bars.5 However, in the past few years several manu-
scripts of the original version have reappeared and it became evident that
in April 1949 a revision was made which was far more comprehensive
than previously believed.6 In fact, Boulez changed the metrical notation,
and modified or rewrote more than a third of the piece. In tracing the
development of Boulez’s musical language, we have therefore to consult
the Sonatine not only in its published version, but also in the early version
from 1946, and it is precisely the comparison of both versions that
demonstrates the synthesis as well as the conflict between the diverse
formative elements.7
Genesis and Publication

Joan Peyser and Dominique Jameux assert that the Sonatine was commis-
sioned by Jean-Pierre Rampal, who never played it, because it was
too extreme for his taste.8 Rampal however assigns the initiative to
3 5
See Boulez, Conversations with Célestin See Boulez in Goléa, Rencontres, p. 38: ‘Je
Deliège, p. 28: ‘I have always had a tendency sais que cela aussi a été dit . . . Mais c’est
to separate the formal context very clearly faux: j’ai corrigé, en tout et pour tout, dix
from the ideas themselves, although I know mesures!’ In a private letter, Boulez in 1963
full well that in composition style is inti- insisted again that the fundamental text
mately bound up with form; I conduct a sort remained unchanged, but mentioned a
of chemical dissociation to help me to seize concentration of different elements and a
and retain what interests me and to drop stylistic purification; see Mellott, ‘A
what does not.’ Survey’, App. H, p. 361.
4 6
See Boulez, Leçons de musique, esp. p. 296: On the discovery of the early version see
‘Thus one can summarise in broad terms the Gärtner, ‘Pierre Boulez’ “Sonatine”’.
7
employment of thematicism in this Sonatine What follows here is an updated summary
in four ways: general thematicism, themati- of my book Werkstatt-Spuren.
8
cism exclusively confined to a cell, athemati- Peyser, Boulez: Composer, Conductor,
cism based on the neutrality of the constituent Enigma, p. 37 and Jameux, Pierre Boulez,
elements and the strength of the surrounding p. 227.
context, and precompositional athematicism.’
27 Pierre Boulez’s Sonatine (1946/1949)
Boulez.9 Although mainly specialising in Baroque music, Rampal had

won first prize in the flute competition at the Paris Conservatoire
in 1944 with his performance of André Jolivet’s Chant de Linos.
He also performed Jolivet’s Five Incantations and was one of the
musicians who, with Leibowitz, undertook the clandestine recording
of Schoenberg’s Wind Quintet op. 26, which was broadcast directly
after the liberation of Paris on 25 August 1944.10 Leibowitz had dedi-
cated his Sonata for flute and piano op. 12 to Rampal in September of
the same year,11 and it was therefore natural for Boulez to get in touch
with a fellow student appreciated by Jolivet (and probably Messiaen) as
well as by Leibowitz.
In January 1946, when Boulez started to write the Sonatine, his appren-
ticeship was already in its final stage.12 He had completed the advanced
harmony class of Olivier Messiaen with a first prize in June 1945. Assigned
in October to a course in counterpoint under Simone Plé-Caussade whose
way of teaching he couldn’t bear, he skipped the class and as a result,
according to Boulez’s subsequent account, was expelled from the
Conservatoire.13 Nevertheless his private studies continued. From April
1944, he had benefited from weekly counterpoint lessons with Andrée
Vaurabourg, the wife of Arthur Honegger, and in addition, she and her
husband offered guidance in respect of Boulez’s compositions.14 He still
attended Messiaen’s private analysis courses, which had aroused his desire
to compose and confronted him with works such as Stravinsky’s Rite of
Spring and Debussy’s Pelléas et Mélisande as well as with Messiaen’s own
compositions. With increasing enthusiasm, Boulez dedicated himself to
ethnomusicological studies at Paris’s Musée Guimet and Musée de
l’Homme. He also developed his proficiency as a pianist, and above all
he regularly met with Leibowitz to study dodecaphonic works of
Schoenberg, Berg and Webern. A letter, which Boulez wrote while work-
ing on the Sonatine, illustrates the impact of Leibowitz’s lessons and
thoughts at that time:
9
See Rampal, Music, my Love, p. 117: Boulez: Composer, Conductor, Enigma, p. 32
‘Boulez wanted to write a piece for the flute, and Jameux, Pierre Boulez, p. 28.
14
and I was honored to be asked to play it. I had See Boulez’s letter to Vaurabourg, written
met Boulez at the Paris Conservatory.’ in his home town Montbrison on 22
10
See Rampal, ibid., p. 46. September 1945: ‘I am going therefore to
11
See Monod (ed.), René Leibowitz, p. 23. return seriously to work from the end of this
12
On Boulez’s apprenticeship see Jameux, week in order to try to present you with
Pierre Boulez, pp. 3–19 and Gärtner, some magnificent counterpoints! I have also
Werkstatt-Spuren, pp. 17–127. composed a little . . . I will show it to you as
13
See Boulez, ‘Entretien avec Sylvie de soon as I come to Place Vintimille. – I would
Nussac’, pp. 4–5 (Paul Sacher Stiftung, hope also that Mr Honegger might give me
Sammlung Pierre Boulez, hereinafter ‘PSS, his opinion.’ (PSS, Sammlung Pierre
Sammlung Pierre Boulez’.) See also Peyser, Boulez.)
28 Susanne Gärtner
Dear Master,
Please excuse this hasty scrawl. But I wished to reply without undue
delay to your letter which I received the other day. I will bring you your
Concerto on Tuesday 5 around three o’clock in the afternoon. I thank you
for having lent it to me. It has taught me a lot . . . I can’t tell you how much
your articles in Temps Modernes have fascinated me. It is the first lucid
analysis that I have read, and I never anticipated the self-evident clarity with
which you are able to present the evolution of music since the Middle Ages.
At last, something which is not empirically based!
As for me, I am in the middle of composing a Sonatine for flute
and piano, in which I have worked particularly on structure and counter-
point. I have not yet completely finished it. It lacks a final development
section. I would like to be able to show it to you shortly.
I hope perhaps to see you on Tuesday – otherwise I will leave your
Concerto with the concierge.15
By the end of 1945 Boulez had already written several works,16 and in his
second year at the Conservatoire, 1944/45, he had produced five piano
compositions of considerable proportions.17 Probably the earliest is a
Nocturne, which took the style of Gabriel Fauré as a starting point, expand-
ing the harmonic vocabulary and showing a rhythmic ‘inquiétude’ inherited
from Stravinsky and Messiaen. Three further compositions, Prélude, Toccata
et Scherzo, in the form of a triptych, refer to the counterpoint lessons with
Vaurabourg and to the music of Honegger whilst paying homage to Johann
Sebastian Bach in their use of the BACH cipher.18 Certain tone rows in the
15
Undated letter, probably written on 2 finie. Il manque un développement final. Je
February 1946 (PSS, Sammlung René voudrais pouvoir vous la montrer d’ici peu.
Leibowitz). Je pense peut-être vous voir mardi. – Sinon
je remettrai votre Concerto chez le concierge.
Cher Maître, 16
On Boulez’s works prior to the Sonatine see
Excusez ce griffonage à la hâte. Mais Bennett’s important essay ‘The Early Works’;
je veux répondre sans trop tarder à la lettre see also the observations of O’Hagan, ‘Pierre
que j’ai reçue l’autre jour. Je vous porterai Boulez: “Sonate, que me veux-tu?”’, pp. 4–24;
votre Concerto mardi 5 vers 3h l’après- see Nemecek, Untersuchungen zum frühen
midi. Je vous remercie de l’avoir prêté. Il Klavierschaffen; O’Hagan, ‘Pierre Boulez and
m’a beaucoup appris . . . Je ne pourrais pas the Foundation of IRCAM’, 303–7; and
vous dire comme vos articles dans Temps Gärtner, Werkstatt-Spuren, pp. 28–9, 51–68,
Modernes m’ont passionné. C’est la 86–123.
17
première analyse lucide que je lis et je ne See Boulez, ‘Frühwerke (unveröffentlicht)’
m’étais jamais douté de l’évidence avec la (PSS, Sammlung Pierre Boulez, Mappe A,
quelle vous décrivez l’évolution de la musi- Dossier 3).
18
que depuis le moyen âge. Enfin, quelque The triptych as a whole is reminiscent of
chose qui n’est pas empirique! Honegger’s Prélude, arioso et fughette sur le
Quant à moi, je suis en train de nom de BACH (1933); the beginning of
composer une Sonatine pour flûte et piano, Boulez’s Prélude refers furthermore to
où j’ai travaillé surtout l’architecture et le Honegger’s Prélude from Trois Pièces pour
contrepoint. Je ne l’ai pas encore tout à fait piano (1919).
Scherzo point to a first acquaintance with twelve-note music and suggest that
Boulez’s lessons with Leibowitz had already commenced. In the next work,
Thème et variations pour la main gauche, which Boulez completed in June
1945, the reference to Schoenberg’s Variations op. 31 becomes apparent.
The variations as a whole reflect a range of different influences and ways
of writing, most strikingly the resemblance to André Jolivet’s ‘style
incantatoire’.
During the summer holidays of the same year, he completed the Trois
Psalmodies. Slightly contrasting in their textures, all three of these piano
pieces are closely connected with the musical languages of Messiaen and
Jolivet. Directly afterwards, he composed two movements of a Quartet for
Ondes Martenot, which are stylistically close to the Psalmodies. Here,
improvisatory and homophonic sections alternate, some of them using
twelve-note rows, others introducing melodies in quarter-tone tuning. In
November 1945, after revising and extending Psalmodie 3 in favour of a
more complex contrapuntal texture, Boulez wrote his Douze Notations.
These miniatures for piano are almost an inventory of compositional possi-
bilities: although each of the twelve pieces contains twelve bars and uses the
same twelve-note series, their sequence is characterized by contrasts. Apart
from dodecaphonic pieces, others remain indebted to Jolivet’s repetitive
‘style incantatoire’, the rhythmical techniques taken over from Messiaen
are omnipresent, reminiscences of works by Bartók and Debussy are to be
heard, and Boulez even experiments with an ethnic style.19 The fragmented
textures and the precision of articulation and dynamics suggest the incipient
impact of Anton Webern, but still the influence of Schoenberg is predomi-
nant. What looks like an obvious response to the style of Webern points
also to Schoenberg’s piano style, which captivated Boulez at that time.20
The Notations were completed towards the end of December 1945 and
subsequently he orchestrated eleven of the pieces employing an imaginative
use of the orchestra. With the exception of the piano version of Notations,
which Boulez released for publication in 1985, all works prior to the Sonatine
remained unpublished.
Boulez finished the Sonatine on 8 February 1946 and dedicated it to
Rampal.21 Rampal states that disagreements about the choice of pianist
and the impractical notation were reasons for the failure of the
collaboration.22 At the end of the same year the Belgian composer,
19
No. 8, originally entitled ‘Afrique’, is (1945)’ and ‘“Douze Notations” von Pierre
dominated by a percussive pattern which Boulez’.
20
might have been inspired by the African See Boulez, Conversations, pp. 29–30.
21
lamellaphone sanza. On the Notations and See below for a description of the
their different references see, besides the manuscripts.
22
above-mentioned commentaries, See Rampal, Music, my Love, pp. 117–8:
Hirsbrunner, ‘Pierre Boulez: Notations ‘He wanted me to play the work with a
30 Susanne Gärtner
conductor, musicologist and writer, André Souris, offered Boulez the chance
to have one of his works performed within his concert series ‘Aspects de la
musique d’aujourd’hui’ in Brussels. Boulez was delighted and let Souris
make the choice.23 Out of four suggested pieces, the latter decided in favour
of the Sonatine. A lively correspondence ensued, in which Boulez requested
Souris to encourage a bold and shocking interpretation.24
Together with five other first performances of works by stylistically
quite divergent composers including André Jolivet and Frank Martin,
the early version of the Sonatine had its première at the Palais des Beaux
Arts in Brussels on 28 February 1947.25 The performers were Herlin Van
Boterdael, flautist at the Théâtre de la Monnaie, and the pianist Marcelle
Mercenier. Boulez himself couldn’t attend the concert in person, but
subsequently he learned with pleasure from Souris that the Sonatine had
provoked vociferous protests.26 In April the two musicians met in Brussels
and a lifelong friendship arose. It was again André Souris, this time as
editor, who in 1948 enabled the unknown young composer to enter the
music debates of his time with the articles ‘Propositions’ and ‘Incidences
actuelles de Berg’.
After the Brussels première there seem to have been no further perfor-
mances of the Sonatine. When John Cage came to Paris for an extended
visit in 1949, he recommended Boulez to the publishers Philippe Heugel
and Amphion, and it was probably in this context that, in April 1949,
Boulez revised the Sonatine as well as the First Sonata. In late summer of
the same year through the renewed mediation of Cage, deals were finally
closed to publish everything Boulez wanted published.27 Heugel released
favorite pianist of his rather than with my Perhaps this upset him, because time passed
partner, Robert Veyron-Lacroix, because he and I heard nothing further. I must admit
didn’t think Robert’s style was right for the that the piece slipped my mind, too.’
23
music. For his part, Robert wasn’t all that Letter to Souris, 9 December 1946 (AML
keen on Boulez’s music, either. As far as No. 5436/35).
24
notation was concerned, the music was Letter from 31 January 1947 (AML No.
extremely difficult to decipher – and I’m a 5436/38).
25
good sight-reader. There were no measure See Wangermée, André Souris et le com-
bars or any other helpful signs. I could sense plexe d’Orphée, pp. 257–8.
26
that the work had a strong emotional appeal, See Boulez’s undated letter to Souris (AML
but with an extremely heavy concert sche- No. 5436/32): ‘I was gratified to learn from
dule, the idea of spending hours, perhaps your letter that my Sonatine had provoked a
even days, picking my way through a difficult stir. Besides, I saw A. Jolivet a few days
modern work was sapping my spirit. afterwards. He took me to task for the total
Cautiously I asked if he could put in a few absence of “a memorable tune” in this
measure bars. “Play it as it’s written. The Sonatine . . . No comment.’ See also Souris’s
unity of the rhythm is counted at the beat,” he souvenirs of the performance in Souris, La
replied, ever true to his principles. But with- lyre à double tranchant, pp. 181–2. See
out measure bars, the rests are difficult to Deliège in Boulez, Conversations, p. 28.
27
follow and it is really a nuisance to play. I sent On Cage’s role as intermediary, see Peyser,
the music back, and again asked Boulez if he Boulez: Composer, Conductor, Enigma,
could make me a cleaner – and clearer – copy. pp. 60–1.
the Second Sonata in 1950; in 1951 the First Sonata followed published
by Amphion, but it was not until 1954 that the Sonatine appeared.28 On
15 July 1956, more than ten years after the genesis of the piece, the
revised, published version had its official première at the ‘Internationale
Ferienkurse für Neue Musik’ in Darmstadt, the performers being
Severino Gazzelloni and David Tudor.
The Documents
Since the early version’s reappearance, several further documents relating to
the Sonatine have surfaced. A full list of available sources is set out in
Table 2.1.
The oldest extant manuscript source of the early version is a complete
pencil draft, which Boulez later dedicated to Roger Désormière and which
is now found in the Musée de la musique in Paris. It is dated ‘le 8 février’,
already includes metronome markings, but contains virtually no indica-
tions for dynamics and articulation. Several erasures suggest that this was
the earliest working draft of the Sonatine. So far there are no available
sketches from 1946 apart from some analytical notes on this draft (see
Figures 2.1 and 2.2).
Table 2.1: Sources of the Sonatine
Early version (1946)

Pencil draft Musée de la musique, Paris
Fair copy Bibliothèque nationale, Paris, Dept. de la musique, Louvois, Ms. 21612
Separate flute part Archives Jean-Pierre Rampal, Paris
Photocopy in Paul Sacher Stiftung, Sammlung Pierre Boulez, Mappe A,
Dossier 4a
Score copy (1947) Private archives, Brussels
Published version (1949)
Pencil entries Bibliothèque nationale, Paris, Dept. de la musique, Louvois, Ms. 21612
Draft sketch Paul Sacher Stiftung, Sammlung Pierre Boulez, Mappe A, Dossier 4a
Single sketch Paul Sacher Stiftung, Sammlung Pierre Boulez, Mappe A, Dossier 3h
Fair copy with corrections Paul Sacher Stiftung, Allgemeine Sammlung, Fonds Isabelle Berthou
Print (1954) Éditions Amphion, Paris
Revised edition (forthcoming) Éditions Durand, Paris (ed. S. G.)
28
There are no reasons known. Enquiries at
the publishers failed to elicit any information.
32 Susanne Gärtner
Fig. 2.1 Pencil draft, Musée de la musique, Paris, p. 1 (detail)
Fig. 2.2 Pencil draft, Musée de la musique, Paris, p. 5 (detail)

Located in the Bibliothèque nationale in Paris is the fair copy of the early
version, by means of which Boulez revised the piece. The original version is
in ink, pencil entries mark the corrections and at the bottom of the last page
the respective datings ‘Février 46’ and ‘Avril 49’ are to be found. The manu-
script had evidently been in the hands of pianist Yvette Grimaud for some
time, as is shown by her signature on the front page. On the reverse of this
page is found the original dedication to Jean-Pierre Rampal, subsequently
crossed out by Boulez. Towards the end of the 1950s, he gave the document
to Pierre Souvtchinsky and added a new dedication.
A separate flute part of the early version is found in the archives of Jean-
Pierre Rampal, a photocopy of which is available at the Paul Sacher
Stiftung in Basel; the corresponding piano score is missing. From the
Brussels performance, the copy used by the flautist Herlin Van Boterdael
has surfaced, written in an unknown hand and privately owned. As the
correspondence between Boulez and André Souris reveals, two copies of
the Sonatine were made in Brussels in January 1947. When required to
submit the manuscript to Souris, Boulez requested the return of the piano
score from Rampal.29 However as Rampal failed to locate it, Boulez’s own
fair copy was used, the one which he had given to Yvette Grimaud.30 The
Brussels copy from which Marcelle Mercenier played may still lie in her
private archives, which are not yet accessible. The Paul Sacher Stiftung holds
a partial draft and a single sketch, which Boulez noted during the revision in
April 1949. There is also a fair copy of the revised version, which he
entrusted to Amphion adding last-minute corrections.
The Form
Within a continuous composition of about twelve minutes’ duration, the
Sonatine comprises a slow introduction and four distinct movements: a
Rapide, a slow movement, a Scherzo and another Rapide which functions
as a recapitulation (see Table 2.2).31
29 31
See Boulez’s letters to Souris 21 December The formal disposition remained
1946 and 16 January 1947 (AML Nos. 5436/ unchanged by the revision only the propor-
36 and 33). tions vary slightly in the two versions. The
30
See Boulez’s letter to Souris 31 December different numbering of the bars is mainly due
1946 (AML No. 5436/34): ‘Excuse me if this to the fact that Boulez changed the bar layout
Sonata [sic!] has not reached you sooner. But of the whole piece. The originally wide-
the flautist Rampal must have lost one of my stretched bars were shortened and at the
copies because he cannot find it again. Thus same time he added an extra line with
only Yvette Grimaud’s copy remains. You rhythmical signs bundling the semiquavers
can keep it as long as you need it.’ into dyads and triads.
34 Susanne Gärtner
Table 2.2: Sonatine – formal disposition
Published version Early version
Très librement – Lent bb. 1–31 Introduction bb. 1–25

I Rapide bb. 32–96 Exposition bb. 26–74
bb. 32–52 Theme bb. 26–37
bb. 53–79 Development bb. 38–63
bb. 80–96 Transition bb. 64–74
II Très modéré, presque lent bb. 97–150 Slow movement bb. 75–116
bb. 97–115 1. Section bb. 75–93
bb. 116–140 2. Section bb. 94–110
bb. 141–150 3. Section/Transition bb. 111–116
III Tempo scherzando bb. 151–341 Scherzo bb. 117–218
bb. 151–194 Tempo scherzando I bb. 117–144
bb. 195–221 Interlude bb. 145–166
bb. 222–295 Tempo scherzando II bb. 167–202
bb. 296–341 Transition/Rhythmic Canon bb. 203–218
IV Tempo rapide bb. 342–510 Recapitulation bb. 219–332
bb. 342–361 Theme bb. 219–230
bb. 362–378 Development bb. 231–241
bb. 379–495 Final Development bb. 242–318
bb. 496–510 Coda bb. 319–332
In the combination of the four movements of a sonata with first-move-

ment sonata form, the Sonatine follows the structure of Schoenberg’s first
Chamber Symphony op. 9, as Boulez has pointed out. He did not mention
though that Leibowitz also praised Schoenberg’s piece for its economical
structure, using it as a formal model for compositions including his own
Chamber Concerto op. 10. On 5 December 1945, this work had been per-
formed under Leibowitz’s direction in a concert at the Paris Conservatoire,
together with Schoenberg’s Chamber Symphony, Webern’s Symphony op. 21
and Schoenberg’s Herzgewächse op. 20, with Boulez at the reed organ.32 As
his above-cited letter reveals he afterwards studied Leibowitz’s op. 10 while
working on the Sonatine.
Whereas Leibowitz had copied the form of Schoenberg’s op. 9 includ-
ing even metronome markings and stylistic references to the extent of
literal quotations, Boulez kept his distance when adopting the model. He
interchanged the order of Scherzo and slow movement and did without an
32
See Boulez in Goléa, Rencontres, pp. 27–8;
see also Boulez, ‘Entretien avec Sylvie de
Nussac’, p. 8.
extensive development in the middle. Instead, short transitions link the

four movements.33 In conversation with Goléa, Boulez mentioned oppo-
sitional kinds of developments in the Sonatine, some still being of a
thematic character, others already approaching athematicism.34 In his
Collège de France lectures, he later spoke of athematic transitions which
contrast with the thematically shaped movements.35
While adopting the form from Schoenberg, Boulez had tried to avoid any
stylistic influence. Ambitious to create something new, he increased the
economy and ambiguity of form by reducing the musical material to a
minimum.36 In fact, the material of the Sonatine consists only of a twelve-
note row as initial basic material, a twelve-note theme, a scherzo motive,
which figures as a sort of adversary, and some material for the accompani-
ment, namely arpeggios and chords derived from the row, as well as a bass
cluster.
The Series
The series of the Sonatine divides into two hexachords and is characterised
by semitones at the beginning and end, enclosing two groups of four notes,
each composed of a tritone, a fourth/fifth, and a major third: the two
segments are linked together by another semitone. A further tritone (F-
natural – B-natural) frames the series as a whole.
Ex. 2.1a Pierre Boulez, Sonatine, series
33 35
For a detailed comparison between See Boulez, Leçons de musique, p. 293: ‘In
Schoenberg’s Chamber Symphony and the order to link these four movements to one
Sonatine see Gärtner, Werkstatt-Spuren, pp. another, it was necessary to have a certain
223–30. number of transitions which would not have
34
See Boulez in Goléa, Rencontres, p. 38: ‘One a precise thematic profile, but which would,
finds there . . . developments made up of on the contrary contrast by their “vague”
straightforward motives clearly derived from character with the sharply delineated profile
the series, but retaining a thematic character; of the movements. In this way, I contrasted
these developments contrast with other thematicism with athematicism.’ On the
developments resulting from the combina- ideas of athematicism and virtual theme in
tion of the series and rhythmic cells – in other the Sonatine see also Campbell, Boulez, Music
words, a beginning of athematicism.’ and Philosophy, pp. 160–9.
36
See Boulez, Conversations, p. 27.
36 Susanne Gärtner
This series clearly relates to that of Webern’s Symphony op. 21, which had
left a lasting impression on Boulez in the concert on 5 December 1945.37
Amongst his dodecaphonic studies housed at the Paul Sacher Foundation
there is a serial table for op. 21 in Boulez’s hand where he had marked the
symmetry of Webern’s row, with the semitones in the middle, as well as the
surrounding groups of four notes:38
Ex. 2.1b Anton Webern, Symphonie op. 21, series; analysis by Boulez
Leibowitz had already constructed twelve-note rows modelled on

Webern’s Symphony, namely for his Vier Klavierstücke op. 8 (1942–3) and
Wind Quintet op. 11 (1944).39 In comparing the series it is evident that
Boulez also copied the structure of op. 21, adopting the tritone as an
encompassing interval, but inverting the position of semitones and four-
note groups. Instead of a strict symmetry, the row of the Sonatine reveals
Boulez’s preference for what he describes as an ‘équilibre dissymétrique’.40
Interestingly, we find a few dodecaphonic annotations in the early pencil
draft of the Sonatine, which is held in the Musée de la musique in Paris and
which only surfaced in 2013 (See Figures 2.1 and 2.2). Boulez denoted the row
forms there by Roman and Arabic numerals with additional directional arrows
on top as Leibowitz used to do.41 These indications are sufficient to enable us to
reconstruct the serial table he worked with. Roman numerals denote prime
forms, Arabic numerals the inversions of the series (see Table 2.3).42
37
See Boulez in Goléa, Rencontres, p. 28 and in Leibowitz, Introduction à la musique de
Boulez, ‘Entretien avec Sylvie de Nussac’, p. 8. douze sons, p. 337.
38 42
See Boulez, ‘Zwölftonstudien’ (PSS, It remains doubtful whether Boulez under-
Sammlung Pierre Boulez, Mappe A, stood form I which opens the piece as prime or
Dossier 2c). inversion. Leibowitz usually denoted primes by
39
See Gärtner, Werkstatt-Spuren, pp. 166–8. Roman and inversions by Arabic numerals, but
40
See Boulez, ‘Stravinsky Remains’, in Boulez, rebellious student as he was, marked
Stocktakings, p. 84: ‘There is, I think, no the row forms of Webern’s op. 21 the exact
further need to emphasize the symmetries of opposite, primes by Arabic and inversions by
asymmetry or the asymmetries of symmetry, Roman numerals. For compatibility my
which are an essential feature of the rhythmic nomenclature follows Baron, who interpreted
architecture in The Rite.’ See also p. 100: ‘The the first appearance of the row as prime. In my
reader will I think observe, without my book Werkstatt-Spuren I used as prime
describing them further, the effects of (‘Grundreihe’ G1) the form on F which opens
balanced asymmetry which control these the theme of the Sonatine and which relates
various groups.’ directly to Webern’s theme in op. 21 – see
41
See for instance Leibowitz’s serial table for Example 2.1a. See likewise Jameux, Pierre
Schoenberg’s Variations for Orchestra op. 31 Boulez, p. 289 and Chang, ‘Boulez’s Sonatine’,
p. 90.
Table 2.3: Sonatine – serial table

38 Susanne Gärtner
Table 2.4: Sonatine – row families
Semitones Row Forms
Row-family I: C–B/F–F♯ P(C), R(F♯), P(F♯), R(C), I(F), RI(B), I(B), RI(F)
Row-family II: C♯–C/F♯–G P(C♯), R(G), P(G), R(C♯), I(F♯), RI(C), I(C), RI(F♯)
Row-family III: D–C♯/G–A♭ P(D), R(A♭), P(A♭), R(D), I(G), RI(C♯), I(C♯), RI(G)
Row-family IV: E♭–D/A♭–A P(E♭), R(A), P(A), R(E♭), I(A♭), RI(D), I(D), RI(A♭)
Row-family V: E–E♭/A–B♭ P(E), R(B♭), P(B♭), R(E), I(A), RI(E♭), I(E♭), RI(A)
Row-family VI: F–E/B♭–B P(F), R(B), P(B), R(F), I(B♭), RI(E), I(E), RI(B♭)
In Webern’s Symphony op. 21 the symmetry of the row reduces the

possible row forms from 48 to 24. The partially symmetrical structure of
the Sonatine’s row does not lead to series forms which can be superimposed,
but to transpositions which resemble each other and can be grouped into six
row families, each containing eight row forms with the same framing
semitones. For example, transpositions I (Prime on C), VII (Prime on F♯),
6 (Inversion on F) and 12 (Inversion on B) and their corresponding retro-
grade forms, share the same framing semitones C–B/F–F♯, and form row-
family I (see Table 2.4).43
The Movements and their Revision

The introduction Très librement – Lent (bb. 1–31) gives the impression of
an improvisation.44 Slowly, the row emerges from the bass with a chord in
the piano and an ascending figure in the flute. This opening gesture
resembles in detail the beginning of Leibowitz’s Sonata for flute and
piano op. 12 (1944). While Leibowitz immediately closes his introduction
Boulez lets the material grow. Interrupted by rests, wide-stretched chords
alternate with arpeggios and figures whose rhythmical outlines foresha-
dow the theme.
The dodecaphonic structure of the beginning is easy to comprehend.
Boulez skilfully linked forms from row-family I (C–B/F–F♯) through
their shared semitone groups. Concurrently, in the published version the
rhythmic figures vary in a supple way, which can be seen most noticeably in
43
On further details concerning the row Spuren, pp. 168–70 and 236–9; see also
families and the similarities between the Chang, ‘Boulez’s Sonatine’, pp. 104–5.
44
transpositions see Gärtner, Werkstatt- If not mentioned otherwise bar numbers
refer to the published version.
Ex. 2.2 Sonatine, early version, bb. 1–7 with dodecaphonic analysis by Boulez
the five-note grouping as semiquaver-quintuplet (flute b. 2), five demisemi-

quavers (flute b. 3), five semiquavers (flute b. 5), quaver-quintuplet (flute b.
8) as well as five quavers (piano bb. 7–8). In the early version of the Sonatine
this rhythmic finesse was not as yet fully developed: there, semiquaver
quintuplets prevailed (see Example 2.2).
In April 1949, when revising the piece, Boulez refined the rhythmic
figuration of the introduction as a whole. He also fragmented the melodic
lines of the flute introducing registral displacements.45 Furthermore, he
discarded nine bars of the original introduction. They included a melody
in the flute with repetitions and appoggiaturas reminiscent of André Jolivet’s
‘style incantatoire’ (see Example 2.3 esp. bb. 16–22).
45
For a detailed comparison between the see the catalogue of all corrections in Gärtner,
early and published versions of the Sonatine Werkstatt-Spuren, pp. 373–9.
40 Susanne Gärtner
Ex. 2.3 Sonatine, early version, bb. 14–25
In the published version, three newly composed bars were added

instead (bb. 23–5). They develop the preceding figures expanding the
intervallic range and increasing the rhythmic complexity. At the same
time, the repetitive gestures in the piano part at the end of the introduction
were reduced.
The Rapide first movement of the Sonatine (bb. 32–96) starts with the
exposition of the theme. As Boulez never described it in detail, there
have been different interpretations as to its length, ranging from flute part
bb. 33–5, to bb. 33–40, 33–44 and 33–52. The early version leaves no doubt
Ex. 2.4 Sonatine, theme (bb. 33–52)
that the broadest definition is the right one. In the fair copy, by means of
which Boulez revised the piece, the corresponding bars are crossed out with
the remark ‘refaire l’harmonisation de ce thème’ and they reappear on the
draft sketch with a changed accompaniment.46
In his Collège de France lectures, Boulez spoke of the Sonatine’s theme as
still being traditional.47 Transferring the terminology he used while analys-
ing the thematic content of Stravinsky’s Rite of Spring, we can divide the
theme of the Sonatine into two periods, each consisting of an antecedent and
a consequent, and a short conclusion (see Example 2.4).48
The pitches of the theme derive from four row forms. While the first
period presents I(F) (Inversion on F) as a whole, the second period links I(G)
(Inversion on G) and P(D) (Prime on D) through their common boundary
dyads D–C♯, and the final phrase concludes RI(D) (Retrograde Inversion on
D). The intervals are subject to octave displacement and most of the notes
have to be played marcato and fortissimo. Thus the theme of the Sonatine is
in stark contrast to the elegant cantilenas of the traditional flute repertory.
When the first period reappears later, in bb. 217–21, the intensity is further
increased through triple forte, parallel ninths and sevenths, accompanied by
46
See the fair copy (BNF Ms. 21612), p. 1 and of the word: that is to say, a sequence of
the draft sketch (PSS, Sammlung Pierre figures and propositions which form a
Boulez, Mappe A, Dossier 4a). coherent whole, and of which the individual
47
See Boulez, Leçons de musique, p. 293: ‘In elements can be deduced.’
48
this instance above all, the series served to For Boulez’s definitions see ‘Stravinsky
generate a theme, in the most classical sense Remains’, in Stocktakings, pp. 57–9 and 64–7.
42 Susanne Gärtner
the indications ‘percuté, résonné’. The sound world evoked is made explicit
by the indication in the early version: ‘percuté, résonné comme un game-
lang’. Rhythmically the theme shows the ‘inquiétude’ inherited from
Messiaen. Rhythmic cells group small units into dyads and triads. The
resulting irregular pulse is furthermore combined with irrational values
(triplets and a quintuplet) just as André Jolivet had used them copiously in
his Five Incantations for flute.
In the published score, the piano part of the exposition (bb. 32–52) has a
fragmented texture which surrounds the theme. Different row forms are
artfully connected with the ‘melody’, and their rhythms adopt and develop
the rhythmic cells of the theme. This passage has been cited to illustrate
Boulez’s early twelve-note mastery, following and even surpassing Webern,
but such commentaries fail to take account of the fact that the complex
dodecaphonic structures were added only during the revision. Originally the
theme was accompanied by single chords, as well as repetitions of a minor-
third motive in the bass and a rapid fortissimo twelve-note figure (see
Example 2.5).49
The rapid figure resembles the accompaniment at the beginning of
‘Répétition planétaire’ from Messiaen’s cycle Harawi (1945), extending
as far as identical pitch combinations.50 While Messiaen repeated his
figure several times without variation, Boulez repeated and shortened it.
The pitches of the twelve-note figure follow row form I(D), but its
repetitive presentation can hardly be called dodecaphonic in the strict
sense.
The exposition is followed by a development (bb. 53–79). Apart
from sporadic interjections, only thematic material remained in the
published version. In the bass of the piano part, the rhythmic cells of
the antecedent, the head-motive x1 and the iambic cell x2, are varied and
linked with different row forms, mainly from row-family V (E–E♭/A–B♭),
thus approaching and receding from the theme. All this material is also
to be found in the early version, but in a quite different context (see
Example 2.6).
Here the thematic cells were bound into a three-part counterpoint with
repeated interjections in the flute derived from the accompanying twelve-
note figure, and with a chromatic ostinato consisting of the three lowest
notes on the piano A–B♭–B forming a pulsating foundation.51 Messiaen had
used this drum-like cluster in Vingt Regards sur l’Enfant-Jésus (1944) as well
as in Harawi (1945), and Boulez had already adopted it for Thème et
49 50
See also the only remains of the original See ‘Répétition planétaire’, bb. 1–5.
51
accompaniment in the published score, bb. Due to the frequency of occurrence the
32–3. cluster was symbolized by a stemmed x.
Ex. 2.5 Sonatine, exposition of the theme, early version, bb. 26–37
44 Susanne Gärtner
Ex. 2.6 Sonatine, early version, bb. 38–48 (equivalent to print bb. 53–7)
variations pour la main gauche, Psalmodie 3 plus Notations 2 and 9.52 In the
revision of the development he banished the cluster completely and largely
eliminated the interjections.53 Repetitions of thematic material were also
deleted, the remains superimposed and linear presentations of the series
distributed among both hands of the piano.
Ascending from the bass and with a crescendo, the development ends in a
transition (bb. 80–96), a kind of codetta of quite different character. Now it is
only the iambic cell x2 which is interlocked and rhythmically varied in the
high register. The row forms modulate from row-family VI (F–E/B♭–B) to
row-family II (C♯–C/F♯–G) and row-family III (D–C♯/G–A♭). This transi-
tion remained unchanged by the revision:54 here we encounter the first of
the athematic developments which Boulez mentioned in conversation with
Goléa, and Jameux’s pointer to Webern’s op. 27 is important.55 For
Leibowitz, Webern’s Variations for Piano represented ‘not only the
52 54
On Boulez’s use of the chromatic ostinato Only some repetitions of the opening dyad
in the early version of the First Sonata see B–B♭ doubled by octaves were crossed out.
55
O’Hagan, ‘Pierre Boulez: “Sonate, que me See Jameux, Pierre Boulez, p. 231. On the
veux-tu?”’, p. 28. similarity between this transition and the first
53
The only trace of clusters remaining in the movement of Webern’s Variations see also
whole piece is to be found in print bb. 258–9. Campbell, Boulez, Music and Philosophy, pp.
166–8.
culmination of its composer’s work, but also the culmination of our musical
language’; according to him they were a fundamental contribution to athe-
matic composition, because within this work ‘everything is variation, or, to
put it another way, everything is theme’.56 Leibowitz had modelled his
Sonata op. 12 closely on op. 27, and it thus suggested to Boulez a means of
approaching athematicism by following the example of Webern’s
Variations.57
Accompanied by a relaxation of tempo (bb. 93–6), the antecedent of the
theme reappears and a glissando leads to the Très modéré, presque lent
second movement (bb. 97–150). In strong contrast to the forte dynamic of
the preceding passage, the mood is now subdued, the dynamic piano, and the
pulse more flexible, almost improvisatory in character. Nonetheless, three
sections can be distinguished.
The first section (bb. 97–115) is based in the piano on the ostinato trill
G–A♭ which thickens in b. 107 into a tremolo G/A♭–A/B♭. Expanding
arpeggio figures erupt, evoking with the indication ‘scintillant’ the piano
style of Claude Debussy, their pitches derived from P(C). In bb. 99 and
106 a chord is interjected which was already heard in b. 6 of the introduc-
tion. It is built out of the centre notes of I(F), with the opening dyad F–F♯
articulated separately in the left hand of the piano (bb. 104–5, 111, 115).
As a whole, the piano texture here resembles the accompaniment of the
theme in the early version. The flute adds legato figures which are only
partially deducible from the series; their rhythms are vaguely reminiscent
of the theme. In bb. 105–6 the motive which will dominate the Scherzo is
first heard.
The second section (bb. 116–40) is again introduced by a glissando.
Now different trills sound in piano and flute, sometimes interlinked by
arpeggios. Out of the trills, the antecedent of the theme in various trans-
positions, and the opening flute figure of bb. 98–100, a three-part counter-
point develops. With the trills, tremolos and glissandos, elements of
Messiaen’s and Jolivet’s vocabulary join the material of the Sonatine,
evoking the ethnic ‘style incantatoire’ of the Psalmodies. This time how-
ever, the trills are bound to the dodecaphonic principle, formed from
the parallel row forms I(G) and I(A♭), and acting as the backbone of
both sections. A transition passage follows, marked ‘Peu à peu scherzando’
56
See Leibowitz, Schoenberg and his School, Leibowitz’s lessons shortly before Boulez
pp. 240–1. started to compose the Sonatine. Besides his
57
For a comparison between Webern’s op. 27 own copy of op. 27 there exists a copy by his
and Leibowitz’s op. 12 see Gärtner, fellow student Maurice Le Roux who – as a
‘Komposition als klingende Analyse’, 323–9. letter to Leibowitz suggests – only joined the
Boulez’s dodecaphonic studies are not dated, group at the beginning of 1946. See PSS,
but it is most probable that Webern’s Sammlung Pierre Boulez and Sammlung
Variations op. 27 were discussed in René Leibowitz.
46 Susanne Gärtner
Ex. 2.7 Sonatine, scherzo motive
(bb. 141–50), where the motive of the Scherzo gradually emerges (see
Example 2.7).58
The scherzo motive consists of seven semiquavers grouped into two dyads
and an iambic triad in the middle, which is introduced by an appoggiatura.
The rhythmic cells of the motive are combined with varying pitches, but a
basic intervallic shape pervades the Sonatine in the background. It can be
heard for the first time in bb. 153–4 in the bass of the piano and starts with
the minor third C–E♭, followed by the iambic group in the centre covering a
fourth with a tritone appoggiatura, before another tritone leads to the note
repetition at the end. The minor third is not one of the intervals of the series,
therefore the scherzo motive can only loosely be related to a particular row
form.59 With its minor third, the head of the motive sets itself clearly apart
from the thematic material. As a kind of motto, the third-motive C–E♭ acts
as an adversary throughout the whole piece. It was hidden already in the
first two bass notes of bb. 1–2 and 4, it is heard at the end of the introduc-
tion at b. 30 and it formed part of the original accompaniment of the
theme.
The rhythmic shape, as well as similarities in the melodic line, in articula-
tions and dynamics, suggest that the scherzo motive was inspired by a
birdsong motive from Messiaen’s Vingt Regards (see Example 2.8).
While Messiaen repeats the bird motive twice without changes, Boulez
starts to vary his motive right from the beginning of the Scherzo, without
having definitely established it before. In the first Tempo scherzando (bb.
151–94) the motive crosses the whole pitch range, and is mirrored both
vertically and horizontally, and fragmented. The opening minor third, the
central iambic cell and the final repeated note form shortened motives.
These motivic splinters are developed in a three-voice polyphonic texture,
until from b. 185 only the iambic cell remains. An interlude (bb. 195–221)
58 59
There have been only minor changes in the Baron, ‘An Analysis’, 94, sees it as a per-
slow movement. Boulez in 1949 refined the mutation of six successive notes from P(B).
rhythmic shapes of the flute figures and One could as well see a permutation of six
crossed out some repetitions, mainly of the successive notes from I(F).
isolated dyad F–F♯ in the piano.
Ex. 2.8 Olivier Messiaen, ‘Regard des hauteurs’, bb. 2–3 also bb. 57–8
follows, recalling the preceding movements, and culminating in a quotation

of the theme’s first period in piercing minor ninth intervals in the piano part,
marked sempre fff.
The second Tempo scherzando (bb. 222–95) originally had the indication
‘avec beaucoup de fantaisie’60 in marked contrast to the rigorous construc-
tion of this section, which leaves the interpreters virtually no freedom at all.
In two episodes (bb. 222–52 and 259–84) the motivic splinters of the first
Scherzando are now bound into fixed rhythms, and in each section three
different rhythms are mirrored and exchanged in a complex three-part
counterpoint.61
The Scherzo ends with a transition (bb. 296–341) which Boulez had
already discussed in ‘Propositions’, describing it there as an athematic
passage, a rhythmic canon without the support of characteristic contra-
puntal cells.62 This transition is based exclusively on the iambic cell. As the
iamb is a component both of the theme and the scherzo motive the reference
is ambiguous in isolation. Besides, Boulez used the iambic cell in rational as
well as in irrational values and in both directions. Out of this primary
material he produced – similar to the procedure in the second Tempo
scherzando – three different rhythms which were superimposed in a three-
part counterpoint. Twelve superimpositions occur in succession in a com-
plex, quasi-symmetrical construction with voice exchanges.63
Figure 2.2 shows Boulez’s own analysis of the superimpositions of the
rhythms in the early pencil draft. His dodecaphonic numbering of the first
rhythmic combination illustrates furthermore the quite anarchic way in
which the rhythms were later combined with pitches. Row forms RI(F),
60 62
See early version, b. 167. See Boulez, ‘Proposals’, in Stocktakings, pp.
61
It was Bennett in ‘The Early Works’, 60–1, 51–3.
63
who first noticed the underlying rhythms. For a detailed analysis of the rhythmic
For a comprehensive analysis of the second canon and a comparison with rhythmic
Tempo scherzando see Gärtner, Werkstatt- canons of Messiaen see Gärtner, Werkstatt-
Spuren, pp. 207–10, 285–8 and 368–9; see Spuren, pp. 260–70; see also Chang, ‘Boulez’s
also Chang, ‘Boulez’s Sonatine’, pp. 172–7. Sonatine’, pp. 138–41.
48 Susanne Gärtner
R(C) and I(G) jump through the voices, and most of the notes are inter-
preted twice.64 The results this time are dodecaphonic splinters: ‘en
éclaboussures’ was the indication he chose for the transition as a whole.
Elaborately constructed, the Scherzo was only lightly revised in 1949.
Boulez differentiated the dynamic indications and increased the tempo of
the rhythmic canon to ‘subitement tempo rapide’. In contrast, the revision of
the last movement Tempo rapide (bb. 342–510) was extensive. It starts with
the recapitulation of the theme, this time inverted in the bass of the piano
(bb. 342–61). As in the exposition, the original accompaniment consisted of
single chords and repetitions of the shortened twelve-note figure, now
coupled with high chromatic interjections in the flute similar to those in
Jolivet’s Five Incantations and Chant de Linos. This alternated with the
return of the cluster repetitions now transferred into the top register of the
piano (see Example 2.9).
In the published version, only the opening arpeggio figure remained with
its coupled interjection in the flute. All the rest of the accompaniment was
newly composed and correlates in inversion and transposition by a semitone
with the piano part of the exposition.
There follows a development (bb. 362–78), shorter than the correspond-
ing one in the first movement and in both versions only involving thematic
material. The head-motive x1 and the quintuplet y1 are bound into a loose
texture, while the iambic cell x2, which characterised the transitions, is now
omitted. The fourth movement culminates in a long final development, Très
rapide (bb. 379–495), mentioned by Boulez in his letter to Leibowitz. Like
dramatis personae in the finale of an opera, all the material of the Sonatine
assembles for an intoxicating closing scene.
The not-yet-treated thematic cells z1 and z2 are the first to appear in the
bass register of the piano. In b. 386 the scherzo motive joins them, building
growing successions of semiquavers and repeating the third motto now with
a tone repetition, thus provoking an irregular pulse. In the printed score the
beginning of the final development is contrapuntally tightly woven, whereas
in the early version repetitive sequences followed each other. The thematic
cells were doubled by octaves and the cluster again played a prominent role
(see Example 2.10).
Starting from b. 417, twelve-note semiquaver lines emerge, first singly, in
the right hand of the piano (bb. 417–27), in the flute (bb. 430–40) and in the
left hand (bb. 440–9), finally culminating in up to three parallel twelve-note
64
RI(F) starts in the flute (12, 11, 10), jumps left hand (9), right hand (8), left hand (7),
to the left hand of the piano (7, 9, 8), back to right hand (6), left hand (5), flute (4, 3). I(G):
the flute (6), then to the right hand of the left hand (1, 2), right hand (3), flute (4), right
piano (5, 4), to the flute (2, 3) and the left hand (5, 7, 6, 8), left hand (9).
hand again (1). R(C): right hand (12, 11, 10),
Ex. 2.9 Sonatine, recapitulation of the theme, early version, bb. 219–30
50 Susanne Gärtner
Ex. 2.9 (cont.)
lines in increased tempo (bb. 463–89). To form them Boulez used a proce-
dure he had already tried out in the second movement of the Quartet for
Ondes Martenot as well as in Notation 6. Different row forms are linked
together by common notes, while the row sections continually grow or
contract.65
Like entries in a fugue, the presentations of the twelve-note lines are
separated by short episodes, where the third-motive sequences are com-
bined with irregular chord repetitions reminiscent of Stravinsky’s Rite of
Spring.66 Chord repetitions in the manner of Stravinsky had already been
65 66
See for instance the first twelve-note line, See for instance bb. 453–7 and 461–3.
bb. 417–27: P(E), 1–12; I(A), 1–10; P(B), 1–9;
I(C♯), 1–8; P(E), 1–7; P(A♭), 1–5; I(E), 1–4.
Ex. 2.10 Sonatine, final development, early version, bb. 242–56 (equates to
print bb. 379–89)
an ingredient of Boulez’s style. We find them in the early Nocturne, the

Prélude and the Scherzo. In the finale of the Sonatine they originally
accumulated, doubtless with the intention of achieving a spectacular
end, and Boulez in 1946 obviously bore the ‘Sacrificial Dance’ in mind.
Three years later, the chord repetitions were drastically reduced (see
Example 2.11).
52 Susanne Gärtner
Ex. 2.11 Sonatine, clear copy (1949), last corrections (equates to print bb. 425–8)
The fourth movement closes with a Coda (bb. 496–510). The twelve-note
figure of the theme’s original accompaniment leads to extended trills, to
which varied fragments of the theme are cited in the tempo of the slow
movement. In the published version, the arpeggio figure is heard four times
with retrograde forms in both hands so that each repetition is different.67 In
the early version, the figure was repeated seven times literally with the
indication ‘incisif’, the entry of the flute F–F♯ was repeated also, with varied
rhythms, and the thematic fragments in the piano were doubled by octaves.
In bb. 503–6 the first period of the theme is cited for the last time in
piercing minor ninths, but without its tone repetition at the end. Then in the
lowest bass register the scherzo motive builds up, also without its repetition.
‘Très rapide’, ‘très brusque’ and ‘brutal’ the repetitions follow in a final
eruption, combined with an interjection in the flute, leading to the notor-
iously demanding high F. As a counterbalance, the piano overhangs sforza-
tissimo with a minor ninth. (In the early version, the Sonatine had ended
with a cluster.)
Dissociation and Synthesis

The comparison of both versions of the Sonatine explains the difficulties of
analytical approaches based exclusively on the printed score. Boulez in 1949
changed and eliminated so much, that the basis for what remains in the final
67
See print, bb. 496–502.
version is often difficult or even impossible to retrace. While the slow

movement and the Scherzo were only slightly modified, the revision of the
introduction was more extensive. The first and the last movements are in
parts hardly recognisable. The theme in particular, which in the early
version was presented as an accompanied melody, was now concealed in
a dense piano texture and is difficult to distinguish, especially in the
recapitulation.68
The formal structure remained unchanged by the revision. As Boulez had
mentioned in his letter to Leibowitz, with regard to architecture and coun-
terpoint the piece had been clearly conceived from the outset. The Sonatine
follows the form of Schoenberg’s Chamber Symphony in a three-voice
counterpoint with variable texture. The strictly constructed parts of the
Tempo scherzando are opposed by the improvisatory introduction; between
these poles of ‘écriture obligé’ and ‘écriture libre’ there are contrapuntal
passages with a looser texture, parts which follow the model of melody and
accompaniment, and also sections with homophonic sequences. Each move-
ment shows an increasing contrapuntal condensation towards its end with
the culmination of four-part passages in the final development. The contra-
puntal framework of the piece and its skilful realisation remind one of the
importance of the teaching of Andrée Vaurabourg-Honegger, which coin-
cided with Leibowitz’s demand for a reactivation of polyphony following the
example of Schoenberg.
The musical material which Boulez used was reduced to a minimum. The
dodecaphonic theme with its traditional structure relates to Schoenberg and
exhibits at the same time the rhythmic ‘inquiétude’ inherited from Messiaen.
The scherzo motive was obviously inspired by a birdsong motive of
Messiaen, and the influence of the latter is again apparent in the material
for the accompaniment: the rapid twelve-note arpeggio figure, the cluster
and even the chords. Their structure, with a seventh or ninth in the bass
below a combination of tritone and perfect fourth is reminiscent of chords in
Psalmodie 2 and of Messiaen’s seventh mode and his chord in fourths.69 But
in contrast to the Psalmodies, the different ingredients now share the twelve-
note row as common denominator. The arpeggio figure and the most
prominent chords are directly derived from the row, whilst the scherzo
motive can be interpreted as a permutation of the notes of one of the row’s
hexachords.
The series of the Sonatine clearly relates to Webern’s Symphony op. 21.
With its semitones, fourths and tritones it guaranteed an ‘anarchic’
68
Besides the discussion about its real exis- and 232 classified the recapitulation in the
tence and the precise length of the theme, it is fourth movement as a long cadenza.
69
significant that Jameux, Pierre Boulez, pp. 228 On this relationship see in detail Gärtner,
Werkstatt-Spuren, pp. 92–3, 185–6, 319–24.
54 Susanne Gärtner
harmony, whilst its partially symmetrical structure, with framing semitones

and a semitone in the middle, provided abundant possibilities to link
different row forms. Boulez used this treatment extensively; the six row
families, which subdivide the forty-eight transpositions, are treated like a
harmonic reservoir. In the early version, no other individual dodecaphonic
procedures are to be found. The theme does not follow Webern, deriving
directly from the structure of the series; its rhythmic cells build five-note
groups in opposition to the symmetries of the row. The original accompani-
ment can hardly be called dodecaphonic; the development which follows
presented the thematic material horizontally in an ‘ultra-thematic’
Schoenbergian style and was marked by repeated interjections and cluster
repetitions. The slow movement plays with fragments of the theme and
elements of Messiaen’s and Jolivet’s vocabulary, loosely held together by a
dodecaphonic backbone. In the Scherzo, only the joins between sections are
derived from the series, whilst the structure of the main parts is dictated by
rhythmic and contrapuntal ideas. Passages in strict dodecaphony are only to
be found at the beginning of the introduction and in the closing transition of
the first movement. Consulting the early version of the Sonatine, it becomes
apparent that twelve-note technique at the beginning of 1946 was just one
ingredient in a network of references. We see the attempt of the twenty-year-
old Boulez to integrate dodecaphony into a musical language which had
already started to develop idiosyncrasies under the influences of Messiaen
and Vaurabourg.
Within the contrapuntal framework, rhythm, following the techniques
of Messiaen, and pitch, organised in accordance with twelve-note proce-
dures, now competed for dominance. Boulez utilised this rivalry to
approach the problem of thematicism and athematicism. Just as themati-
cism was to be understood as a distinctive fixed combination of both
rhythm and pitch, so athematicism could be achieved through a separa-
tion and an independent development of either. In the ‘precompositional’
introduction, row forms are serially organised, while their rhythmical
outlines change in an improvisatory way. In the athematic Codetta of
the first Rapide movement, only the iambic cell of the theme is used to
shape the twelve-note texture. In the rhythmic canon, the development
of the iamb becomes fully independent of the pitches, which were subse-
quently grafted onto the rhythmic cells.
On the whole, the Sonatine in its early version presents itself as a quite
heterogeneous piece. Elements of Messiaen’s and Jolivet’s styles predomi-
nate, with their obsessive rhythmic patterns and ethnic connotations. It was
mainly here that the revision of 1949 concentrated. Now the decision was
made in favour of the Viennese composers and their principles of variation
and non-repetition. Virtually all repetitions were deleted, and the few
remaining ones modified. As the newly composed accompaniment demon-

strates, Boulez’s priorities had now shifted towards a sophisticated dodeca-
phonic technique. The Sonatine therefore had to be cleansed of stylistic
elements of his French predecessors as redundant interjections, and most
prominently, the cluster. Besides stylistic purification, Boulez used the revi-
sion for rhythmic and contrapuntal refinements. The variability of the
rhythmic figures was increased, the pulsations ruptured and concealed.
The contrapuntal texture was made even more dense, octave doubling
eliminated and sequences interlaced. At the same time, the compass of
both instruments was enlarged and horizontal lines were fragmented by
octave displacements pointing towards the development of a personal con-
trapuntal style.
The revision resulted in a compression and amelioration. The comparison
of both versions illustrates Boulez’s considerable development as a composer
between 1946 and 1949. Through the newly composed passages the piece
became even more stylistically diverse. But it is precisely its heterogeneity
that makes the Sonatine so explosive.
Written in English in collaboration with Peter O’Hagan.

3 Schoenberg vive
Jessica Payette
The pronouncement that ‘Schoenberg is Dead’ has persisted as an enticing

mantra for Pierre Boulez’s radical break with tradition, but composers,
musicians and intellectuals alike often forget the entirety of the phrase that
he references: ‘Le Roi est mort, vive le Roi!’ (The king is dead, long live the
king!).1 In the text of the essay that precedes Boulez’s provocative allusion to
this customary French heralding of a new monarch, he cites the music of
both Anton Webern and Arnold Schoenberg as crucial to his compositional
maturation and legacy, intimating that he is the heir apparent to the Second
Viennese School. While Boulez certainly makes a plea for nuanced study of
Webern’s music in this obituary, the allusion to the royal adage upholds
Schoenberg as the originator of this lineage and, as such, denotes that
Schoenberg’s death fulfilled a transfer of sovereignty from Schoenberg to
Boulez.
From 1952, the year of publication of Boulez’s commemorative essay,
onward to the present, both analytical and historical accounts gauging
Webern’s influence on Boulez2 and his colleagues have continuously
displayed evolving depth and sophistication while scholarly considera-
tion of Schoenberg’s influence on Boulez’s compositional practice or
worldview has been almost exclusively confined to discussion of the
twelve-note method.3 This is surprising because Boulez’s most laudatory
remarks about Schoenberg’s output reveal his gratitude to Schoenberg the
intrepid modernist. What is now commonly designated as Schoenberg’s
‘free atonal’ output, works composed from circa 1908 to 1915, is unques-
tionably most captivating to Boulez, as he denotes in the obituary:
‘Tonality is effectively suspended in the Three Piano Pieces of Op. 11.
Thereafter the experiments become increasingly intense until they
1
Boulez, Stocktakings, p. 214. The essay first The Musical Language of Pierre Boulez,
appeared in the February 1952 issue of pp. 41–52.
3
The Score, just six months after Schoenberg’s Even articles directly addressing this sphere
death. of influence are scant; see, for example,
2
This course of study begins with the second Ashby, ‘Schoenberg, Boulez, and Twelve-
issue of Die Reihe (1955), dedicated to Tone Composition as “Ideal Type”’,
Webern on the tenth anniversary of his pp. 585–625.
death, and runs through Goldman,
56
57 Schoenberg vive
culminate in the resounding triumph of Pierrot lunaire.’4 Later, in his

1961 encyclopedia entry for Schoenberg, Boulez expresses the same
appreciation, again emphasising the superiority of the works that precede
Schoenberg’s shift to serialism:
Through these works, all of them outstanding, we can, I think sum up Schoenberg’s
creative personality: it is in this world that is neither tonal nor yet serial that he
shows his most brilliant gifts and his greatest vitality; the force of renewal locked up
in his language comes out much more in these works than in the later compositions
where he adopted the serial principle.5
Throughout his long career Boulez routinely stresses the dialectical para-
meters of his own music as ‘strongly organised, but free’.6 He is clearly
driven by an ongoing pursuit to discover new sonic apparatuses and
compositional premises that orient the listener in a dynamic auditory
space that fosters interplay between the distinct and murky perception of
structural constants and spontaneous zones. This conceptual framework
resonates strongly with Theodor Adorno’s well-known endorsement of
Schoenberg’s free atonal masterworks as models of musique informelle, as
conveyed in his 1961 lecture at Darmstadt. Adorno considers ‘informal
music’ a historically evolving phenomenon and predicts that the next
works in this vein will emerge as ‘a-serial music’ generated from a fusion
of ‘unrestricted freedom’ with the ‘rationalised character of postwar
musical material’.7 Edward Campbell alerts us to several texts in which
Adorno chides Boulez and his Darmstadt colleagues for downplaying
Schoenberg’s influence and enjoins them ‘to account for their failure to
acknowledge fully their patrimony’.8 In typical hyperbolic fashion,
Adorno overstates the severity of this negligence, and this perhaps partly
accounts for his hesitancy to carefully investigate the Schoenberg–Boulez
lineage because there is no dispute: Adorno and Boulez clearly agree that
Schoenberg’s free atonal works were seminal achievements. For example,
Adorno praises Erwartung as a work that spawns detachment from
the foregrounding of motivic and thematic repetition because it ‘inte-
grates partial complexes of relative autonomy into a relationship which
manifests itself cogently through its characters and their reactions to each
other’.9 Around the same time, Boulez arrives at precisely the same
conclusion and describes Erwartung as ‘invention in a perpetual state of
becoming, and freed from all predetermined formal frameworks’.10
4 7
Boulez, Stocktakings, p. 210. Adorno, Quasi una Fantasia, p. 275.
5 8
Ibid., p. 281. Campbell, Boulez, Music and Philosophy,
6
Boulez, ‘Pierre Boulez on His Works: p. 85.
9
Interview by Wolfgang Schaufler’, Universal Ibid., p. 294.
10
Edition Interviews, 14 May 2012. Boulez, Stocktakings, p. 283.
58 Jessica Payette
As Adorno follows Boulez’s career and compositional output he continu-

ally returns to Boulez’s explanation of métier as the ‘challenge to go
beyond’ the artisanal: innovation cannot result from adopting a stance
of compositional objectivity and the resultant positing of a tabula rasa,
but must ideally exhibit some projection of subjectivity and break with
tradition in a compelling manner.11
Prior to his acquaintance with Adorno, Boulez’s post-war studies with
Olivier Messiaen and René Leibowitz in 1944–6 led to the disappointing
realisation that the output of both Stravinsky and Schoenberg fell
creatively flat after the ‘brilliant firework display’ of their early years as
they were ‘haunted by history with a capital H’.12 Boulez and his peers
were disappointed that this generation reverted to conventional formal
structures in the 1920s, yielding an uninspiring neo-classicism and down-
playing the sui generis compositional techniques and forms that had
emerged from the intriguing dramatic scenarios that initially fuelled the
‘firework display’. In his short period of formal study with his elders in
France, Boulez’s maturation as a composer was influenced by competing
advocates who directed him to musical developments occurring
subsequent to the emergence of free atonality in 1908 and to those that
appeared in the aftermath of the First World War. Messiaen’s exaltation
of Stravinsky’s early rhythmic experimentation in the Russian ballets,
especially Le Sacre du printemps, was seminal in Boulez’s formative
years, and Francis Poulenc notes that ‘The Messiaenistes are “very against
Stravinsky’s last period”. For them, the music of Igor stops with Le
Sacre.’13 Boulez’s period of study with René Leibowitz was less fulfilling
due to Leibowitz’s internalisation of Schoenberg’s twelve-note technique
as a phenomenological reduction14 in which the determination of the
initial series is privileged as the chief compositional act, which propagates
all other musical parameters. Leibowitz described the morphological and
aural properties of compositions as rooted in the series and refused to
present them as worthy of independent consideration because ‘the musical
elements and figures invented by the composer are linked a priori to the
11
Campbell, Boulez, Music and Philosophy, ‘The 1945 Stravinsky Debates’, p. 119. Sprout
pp. 93–6. Campbell provides an insightful investigates how the academic activities of
discussion of the aesthetic values that are Boulez and Serge Nigg were affected by the
invested in the term ‘métier’, and demon- reintroduction of modern ‘degenerate’ music
strates that Adorno espouses Boulez’s posi- to post-war Paris.
14
tion on the role of the artist primarily to Campbell, Boulez, Music and Philosophy,
further explicate the philosophical and p. 21. Campbell discusses Leibowitz’s appro-
sociological ramifications. priation of Husserl’s ideas, which is reflected
12
Boulez, Orientations, pp. 358–9. in statements such as Schoenberg having
13
Letter from Francis Poulenc to Darius placed ‘the musical world “between
Milhaud 27 March 1945, cited in Sprout, parentheses”’.
59 Schoenberg vive
series’.15 Boulez was not timid in his frustration with Leibowitz’s didactic
promulgation of Schoenberg’s serial system and his outspokenness on
the matter is often emphasised in the scholarly literature in which his
pejorative comments are quoted.16 Leibowitz wholly neglected the impor-
tance of the stylistic innovations that first appeared in Schoenberg’s
expressionist works, especially his novel ideas about crafting omnidirec-
tional axes of counterpoint and the possibilities for endowing timbre, or
tone colour, with large-scale structural ramifications.17 Leibowitz’s por-
trayal of Schoenberg as a mechanistic composer led Boulez to question the
viability of integral serialism as a compositional approach and to identify
in his own writings features of the Second Viennese School’s music that
Leibowitz neglected. Boulez’s prolific writings that deliberately minimise
discussion of pitch organisation in favour of addressing this repertoire’s
vanguard compositional premises and unique sonic properties lend
insight into the qualities that he deems significant. In a short essay entitled
‘Kandinsky and Schoenberg’, Boulez compares his physiological response
to viewing Kandinsky’s Munich collection – ‘in which this liberating
force explodes with a youthfulness and an audacity that penetrate the
depths of my being’ – to experiencing ‘the sumptuous, dazzling quality of
Schoenberg’s Erwartung and Die glückliche Hand’.18 Similarly, in ‘Bach’s
Moment’ he cites Schoenberg’s treatment of sonic components (intona-
tion, timbre and register) as pivotal to the expansion of morphological
possibilities: ‘Schoenberg’s work, in direct contrast to Bach’s, goes in
search of a new constitution of the sound world; and it seems to me that
this is its main and unique virtue: an important discovery if ever there was
one, in the history of musical morphology.’19
15 17
Leibowitz, Introduction à la musique de Hinton, ‘The Emancipation of
douze sons, p. 103. Kapp, ‘Shades of the Dissonance’, p. 574. Hinton highlights the
Double’s Original’, p. 14: ‘The failure of the parallels that Schoenberg draws to visual art
serial movement marks the failure of an in his discussion of harmonic density and
a priori, abstract conception to attain Klangfarbenmelodie at the end of the
realization. This must have been what Harmonielehre (Theory of Harmony), as the
Leibowitz – who had made available the view- composer argues that ‘primitive’ chords, like
point of historical consequence – meant when triads, are akin to flat paintings without any
he called serialism an “ultra-consequential” sense of depth, or perspective. Hinton
theory.’ reinforces that Schoenberg’s conclusion in
16
Boulez tells Joan Peyser: ‘I found a new this text conveys his belief that looser har-
voice in Webern, one that Leibowitz could monic strictures and Klangfarbenmelodie will
not possibly understand because he could see enable music to engage with a greater range
no further than the numbers in a tone row’ of psychological conditions, and these new
(Peyser, To Boulez and Beyond, p. 134). He techniques will ultimately ‘bring us closer to
comments to Antoine Goléa that ‘Leibowitz, that which is projected to us in dreams’
for serial music, was the worst academicism; (Schoenberg, Theory of Harmony, p. 424).
18
he was much more dangerous for serial music Boulez, Orientations, p. 345.
19
than tonal academicism had ever been for Boulez, Stocktakings, p. 7. In the essays in
tonal music’ (Goléa, Rencontres avec Pierre the Stocktakings and Orientations antholo-
Boulez, p. 46). gies pitch organisation is only examined in
60 Jessica Payette
Theorists Joseph Straus and Michael Cherlin address the morphology of

Schoenberg’s free atonal compositions and propose that they display
a unique handling of the (un-)alignment of melody and harmony. Their
analyses demonstrate that ‘deviations from inversional symmetry’ and ‘time
shards’ are common early Schoenbergian compositional practices, which
will be examined later with regard to Boulez’s works. Straus studies pitch
symmetry to show that ‘the atonal music of Schoenberg and Webern often
narrates the establishment, disruption, and reestablishment of a normative
symmetry’.20 Cherlin suggests that time shards are a kind of emphatic
slippage, a recollection of common-practice-era phraseology and metric
structures in which passages in Schoenberg’s music revert to ‘the use of
a steady pulse-stream, set in contrast to its immediate musical environment,
to express a sense of altered, “uncanny” time’.21
Boulez similarly regards Schoenberg’s works prior to the 1920s as the
rare products of a composer who is gifted with the opportunity to encapsu-
late the complexity of his world in music: ‘It is only very seldom that the
composer finds himself in the presence of a world that he has glimpsed,
like Schoenberg, in a single flash of heightened awareness, a world he then
has to bring into actual existence.’22 From Boulez’s various comments
on Schoenberg’s works it is apparent that his appreciation of the elder
composer’s rendering of musical morphology is of great importance because
he believes that it endows the musical work with a multidimensionality
that fosters the entwinement and variation of spatial, textural and
temporal levels.23 In 1958 Boulez describes op.11 no. 3 as Schoenberg’s
first composition that amalgamates ‘violent contrasts’ and ‘a form in
constant evolution’.24 These features continue to be the focus of Boulez’s
comments on Schoenberg’s innovation, and serve as the basis for many of his
generalised comments on expressive purpose and morphology: for example,
he remarks that ‘Music is perhaps the least dissociable of all expressive
media, in the sense that it is its actual morphology, before all else, which
detail in ‘Possibly . . . ’ and ‘The System own musical persona: ‘Boulez’s thought is
Exposed’. based on a retrospective, critical, scathing
20
Straus, Extraordinary Measures, p. 73. view of the key aesthetic moments of
21
Cherlin, Schoenberg’s Musical Imagination, twentieth-century music, which he takes as
p. 173. Both theorists point out that these his lead for his own path. The path has but
techniques also appear in Webern’s compo- one purpose: to uncover a unity in style and
sitions, but they credit Schoenberg as the first in writing that is decisively free from any
to demonstrate creative application of these scent of the past.’
24
techniques, which were subsequently Goléa, Rencontres, p. 22. ‘La troisième
embraced by his students. pièce accuse des contrastes violents dans un
22
Boulez, Orientations, p. 77. mouvement agité . . . On peut voir dans ce
23
Nattiez, The Battle of Chronos and récitatif très libre un des premiers essais de
Orpheus, p. 73. Nattiez addresses the larger Schoenberg, et des plus concluants, de créer
role that Boulez’s assessment of earlier une forme en constante évolution.’
composers plays in the development of his
61 Schoenberg vive
expresses the emotional development of the artist.’25 He also asserts that the
major task for new music is the creation of a ‘morphology that is in constant
evolution’ such that ‘formal criteria based on the repetition of material are
no longer applicable’.26 In 1974 Boulez still views Schoenberg’s break with
tonality as ‘an explosion as much in form – the method of composition – as
in actual language’ because ‘dimensions are fused and interchanged; the
conception flouts order and finds renewal in the extreme tension and effort
of instantaneous invention’.27 The weight of Schoenberg’s influence on
Boulez is perhaps perceived most powerfully in his vocal works as he further
extends Schoenberg’s destabilisation of the reciter, or in some cases, the
implied protagonist. In order to amplify the ‘internal violence’ contained
particularly within René Char’s surrealist poetry, but also more generally
within iconic French symbolist poetry, Boulez makes audible a distinction
between singing poetry and singing ‘a poetic proposition’28 modelled on
Schoenberg’s morphological properties and his commitment to his ‘choice
of “subjects”’, which reveal ‘his profound preoccupations as a creative
artist’.29
Both composers accomplish this destabilisation of identity in part by
increasing the complexity of musical interludes, a crucial compositional
attribute that separates their music from Webern’s: such passages often
defy the conventional function of ‘transitional material’ as they upend the
continuity of the foregoing material or produce intertextuality through the
coalescence of formerly independent strands. The literary theorisation of
intertextuality, and its application to musical works, continues to build on
Julia Kristeva’s reading of Mikhail Bakhtin: ‘any text is constructed as
a mosaic of quotations; any text is the absorption and transformation of
another’.30 Boulez’s long-standing analogy comparing musical works to
labyrinths embraces this idea and denotes that his compositional activity is
essentially a long-term exploration of musical intertextuality: ‘The most
tempting situation is to create a labyrinth from another labyrinth, to super-
impose one’s own labyrinth onto the labyrinth of the composer, rather than
the futile attempt to reconstitute the composer’s own process. To create,
from the uncertain image one has, one’s own process.’31 We shall now
25
Boulez, Stocktakings, p. 139. d’un autre labyrinthe, de superposer son
26
Boulez, Orientations, p. 144. propre labyrinthe à celui du compositeur:
27
Ibid., p. 327. non pas essayer en vain de reconstituer sa
28
Boulez, Conversations with Célestin démarche, mais créer, à partir de l’image
Deliège, p. 43; Orientations, p. 342. incertaine qu’on en peut avoir, une autre
29
Boulez, Orientations, p. 328. démarche.’ Translation in Goldman,
30
Kristeva, Desire in Language, p. 66; Klein, ‘Understanding Pierre Boulez’s Anthèmes’,
Intertextuality in Western Art Music, p. 16. See also Di Pietro, Dialogues with
pp. 11–12. Boulez, pp. 7–8.
31
Boulez, Jalons, p. 37. ‘La situation la plus
séduisante est de créer un labyrinthe à partir
62 Jessica Payette
discover how Boulez builds on Schoenberg’s precedents in relation to the

parameters discussed above.
Spatial and Textural Omnidirectionality

As Boulez’s access to scores increased and performances of new music
became more plentiful in the 1950s, his study of the kinds of contra-
puntal practices found in the music of the Second Viennese School led
to the development of a concept of ‘diagonal’ polyphony. While
Boulez’s esteem for Webern’s ‘creation of a new dimension, which
one might label diagonal’,32 is discussed at length in the scholarly
literature, he actually regards Wagner as the progenitor of this compo-
sitional technique.
With Wagner a point was reached at which two ideas are on the brink of
amalgamating and producing an overall phenomenon in which the vertical and
the horizontal are projected on to each other. In this way we find tonal functions
increasingly undermined by the individual power of the intervals; and it was from
this point that the style of first Schoenberg, and then Berg and Webern,
developed.33
Gilles Deleuze and Félix Guattari stress that Boulez’s interest in rendering
an array of different spatial configurations stems from his interaction with
diverse historical precedents: ‘When Boulez casts himself in the role of
historian of music, he does so in order to show how a great musician, in
a very different manner in each case, invents a kind of diagonal running
between the harmonic vertical and the melodic horizon. And in each case it
is a different diagonal, a different technique, a creation.’34 Although Boulez is
both impressed by and dismissive of Schoenberg’s ‘contrapuntal
constructivism’,35 he views its usage in Pierrot lunaire as facilitating an
ideal fluctuation between strictness and freedom: ‘Technically it is much
less “learned” than people have liked to imagine . . . Anyone who studies the
score closely cannot fail to be struck by the logical basis of the various
musical deductions, and also by the freedom and ease with which
Schoenberg manipulates that logic.’36
Schoenberg’s intricate formulation of radically modern counterpoint is
perhaps best understood by reviewing his questioning of Ernst Kurth’s
32 34
Boulez, Stocktakings, p. 297. He defines the Deleuze and Guattari, A Thousand
diagonal as ‘a kind of distribution of points, Plateaus, p. 327.
35
blocks, or figures, not so much in the sound- Boulez, Stocktakings, p. 198.
36
plane as in the sound-space’. Boulez, Orientations, p. 336.
33
Boulez, Orientations, p. 255.
63 Schoenberg vive
term ‘linear counterpoint’ in his 1931 essay ‘Linear Counterpoint: Linear

Polyphony’. As Daniel Grimley explains, Kurth’s Grundlagen des linearen
Kontrapunkts (Foundations of Linear Counterpoint, 1917) suggests that
the tendency for melodic dissonance is held in check by harmonic pro-
gressions that gravitate toward consonance and thus a dynamic equili-
brium governs the formation of music, implying that unbalance in the
form of abundant dissonance should not occur.37 Schoenberg views this
postulate as contradictory to the presumption that linearity constitutes
a horizontal flow and contends that ‘parts ought to be independent of each
other even in their harmonic relationship’.38 In fact, already in the
Harmonielehre he asserts that this equilibrium will be disturbed because
‘chords are formed merely as accidents of the voice leading, and they have
no structural significance since responsibility for the harmony is borne by
the melodic lines’.39 Schoenberg’s contrapuntal trademark that frequently
brings these ‘accidents’ to fruition is the pre-emptive initiation of a new
phrase in a primary voice (Hauptstimme) before the phrase in the former
predominant voice has ended, creating a kind of updated stretto proce-
dure. This is groundbreaking in its execution, but it is a traditional
compositional technique that Schoenberg deploys with confidence
because of his fundamental conviction in developing variation, which
assures that all material shares the same nascent origins in
a composition’s opening bars and thus can be manipulated such that the
arrival of new entrances need not necessarily feature transparent melodic
repetition or coincide with metric stresses.
Although Schoenberg’s general reliance on this stretto tactic is discern-
ible throughout the atonal output, the way that he pairs it with other
distinct rhythmic, registral and timbral schema is quite diverse and often
integral to the composition’s overriding aims, as Joseph Straus emphasises
when identifying a striking example of how Schoenberg uses pitch orga-
nisation to mimic the poem’s theme of corporeal deterioration in ‘Valse de
Chopin’, the fifth movement of Pierrot lunaire. Straus observes that the
distortion of inversional symmetry actually mimics the sense of non-
equilibrium that accompanies chronic illness as it recurs in the text’s
refrain (‘Wie ein blasser Tropfen Bluts/Färbt die Lippen einer Kranken’;
‘As a pale [tubercular] drop of blood stains the lips of a sick man’),40
37 40
Grimley, Carl Nielsen and the Idea of English translations of Hartleben’s trans-
Modernism, p. 218. lations of Giraud vary greatly as they are
38
Schoenberg, Style and Idea, p. 291. twice removed from the original poetry and
39
Schoenberg, Theory of Harmony, p. 312. many liberties are taken with translation,
For an insightful summary of this exchange idioms and syntax; for example in published
see Dudeque, Music Theory and Analysis in translations ‘einer Kranken’ is translated
the Writings of Arnold Schoenberg, pp. 55–7. variously as ‘a sick man’, ‘a consumptive’ and
‘an invalid’.
64 Jessica Payette
Ex. 3.1a Schoenberg, ‘Valse de Chopin’, Pierrot lunaire op. 21, bb. 1–9. D♯ disrupts the
symmetry around A♮
announced in the opening bars when the occurrence of a D♯ in the piano

disrupts the sense of symmetry around A♮ (Example 3.1a). This tritone
opposition intensifies throughout the movement, especially when the
clarinet in A is suddenly exchanged for bass clarinet, and it enters
65 Schoenberg vive
piercingly trilling on D♯ in bb. 32–3.41 However, Straus does not comment

on the pronounced alteration of rhythmic alignment that occurs at the
beginning of this sentimental and subjectively focused unit of text (b. 30,
‘Melancholy, sombre waltz, never leave my senses, cling to my thoughts
like a pale drop of blood!’): the vocal part suddenly relaxes into triplet
subdivisions of the beat, the bass clarinet’s trills are placed on beat three
for the first time in the movement and the extended tertian chords of the
previous phrase subside into amorphous harmonic clusters (Example 3.1b). It is
these pile-ups or conglomerations of compositional alterations in Pierrot
lunaire that typically correspond to junctures in the poetic terrain where
a shift from abstract imagery to interior subjectivity occurs.
Similarly, the establishment of gestural configurations, and their dis-
tribution and accumulation through the registral space, are fundamental
to Boulez’s compositional aims. In an essay that convincingly proposes
that parallels can be drawn between the spatial dynamism of Jackson
Pollock’s Allover painting and post-war music, David Gable observes that
Boulez and Stockhausen were the first composers to truly actualise
Schoenberg’s description of musical space as omnidirectional, as envi-
sioned in his essay ‘Composition with Twelve Tones’.42 Schoenberg is
optimistic that the imaginative configuration of the pitches comprising
a row will result in a liberation of aural perception similar to Emanuel
Swedenborg’s vision of heaven, as described in Balzac’s Seraphita: ‘there
is no absolute down, no right or left, forward or backward’.43 While
Gable’s supporting examples only address Boulez’s and Stockhausen’s
sostenuto styles in works that he classifies as ‘seamless and unmeasured
continuums’, he illustrates that this state is achieved primarily by the
composers’ careful dispersion of harmonic tension throughout the entire
registral field in which ‘the intervallic structure is projected through the
available space’.44
In the first movement of Le Soleil des eaux, ‘Complainte du lézard
amoureux’, a setting of a poem sung from the perspective of a lizard, Boulez
cleverly juxtaposes the text’s manifold gestures of swooping motion versus
the lizard’s solidified stasis. The lizard is perched on the wall and gazing
upward, admiring a lovely goldfinch, sensitive to the motion of flying as
birds, owls and butterflies circle around, while, more alarmingly, threatening
bullets fired by men dart about. The opening presentations of the prime row
enact this contrast between airy buoyance and terra firma: initially in b. 1
a pair of descending major sevenths punctuate the end of the row when
introduced by the flutes and these four pitches (G♮–A♭–C♮(B♯)–C♯) are
41 43
Straus, Extraordinary Measures, p. 75. Schoenberg, Style and Idea, p. 223.
42 44
Gable, ‘Words for the Surface’, p. 263. Gable, ‘Words for the Surface’, p. 261.
66 Jessica Payette
Ex. 3.1b Alteration of rhythm and meter, bb. 27–35

67 Schoenberg vive
Ex. 3.2a Boulez, Le Soleil des eaux, ‘Complainte du lézard amoureux’, bb. 1–4
Prime row with large descents doubled in the strings
emphatically doubled by the first violins.45 A new row statement then begins
in the oboe, now completed by English horn and bass clarinet, and again
features a plummeting descent in the English horn, a diminished octave/
major seventh (C♮–C♯), approached with a grace note that is doubled in the
xylophone and violas (Example 3.2a).
45
This is an analysis of the 1965 published revisions improve the clarity: ‘The music is not
score. A comparison to the score published in changed at all, but as for the orchestration,
1959 suggests that the revision of the opening now you can play it!’ (Gable, ‘Ramifying
bars served to set the rows in high relief in the Connections’, p. 110).
woodwinds. Boulez mentions that his
68 Jessica Payette
Ex. 3.2b ‘Complainte du lézard amoureux’, bb. 10–13

Spatial compression of the opening row
In b. 10 the voice of the lizard enters reiterating the prime row as a

compressed spatial entity; here the span from G♮ to C♯ is converted into an
ascending augmented fourth in b. 12 (Example 3.2b).
This establishment of intervallic motives that project outward to the
listener from a neutral background due to their spatial and timbral qualities
is Schoenbergian to the core, and we will continue to see that both compo-
sers’ stylistic inclinations are to create unifying musical condensations of
the poetic imagery. Thus synoptic text painting is generally an essential
component of the Grundgestalt (basic shape). Boulez’s decision to utilise
vocal monody in this movement may stem from a desire to illustrate a small
creature’s longing for the ability to access the vast environmental expanse
and fly through space. The high percentage of vocal cadential gestures
comprised of exposed tritones, and major and minor sevenths and ninths
suggest that Boulez emphasises wide intervallic distances to reinforce the
lizard’s wish for liberation from her adhesion to the earth.46
The sparse opening bars of ‘Complainte du lézard amoureux’ are the
nearest the piece comes to a soloistic instrumental passage and, in addi-
tion to the unfolding of the intervallic material, Boulez also introduces the
arguably predominant textural attribute: the interruption of luminescent
polyphony by aggressive sound blocks that puncture the tranquil hori-
zontal flow. The unrelenting textural juxtapositions seem to be a musical
replication of the structural design of Char’s poem, and also articulate Le
Soleil’s overriding theme of the ill effects of human intrusion in nature
and resultant environmental damage. In ‘Complainte’ Char composes
chains of descriptive imagery that break off into incitements to action
(especially forceful when interjected in the imperative mood). Figure 3.1
illustrates the ongoing modification of registral expanse and textural
46
Since the vocal delivery is clauses that are followed by a quaver rest or
quasi-improvisatory, cadences are accorded more. Based on these criteria, there are
varying degrees of emphasis, and the weight twenty-four vocal cadences in the movement:
of some are even left to a performer’s discre- ten are tritones, sevenths (one spelled as
tion. This statistical tabulation includes vocal a diminished octave) or ninths; thirteen are
cadences in which the second pitch is marked unisons, seconds, thirds or fourths, leaving
with a fermata, those that occur at the end of only one cadence (an augmented fifth) in the
sentences, and those that occur at the end of middle of the intervallic spectrum.
69 Schoenberg vive
Fig. 3.1 Textural quality and registral span of instrumental interludes in

‘Complainte du lézard amoureux’
variance in the five instrumental sections and highlights the intensifica-

tion of contrast that arises in the third and fourth interludes. The third
interlude consists strictly of diagonal polyphony as four semiquavers
introduced by the bass clarinet at rehearsal figure 6 are dispersed through
the orchestral forces (C–B♭–A♭(G♯)–A♮ [again, a major 9th descent plus
a diminished octave/major 7th descent]). The fourth interlude features
Stravinskyian blocks in mixed metre that subside into a harsh trill
one bar after rehearsal figure 9. Finally, the concluding interlude per-
forms a dramatic synthesis of omnidirectionality: it begins with elasticity
at rehearsal figure 11 (staggered entries of unequal groupings of two to
six quavers expand and contract through the registral space) and culmi-
nates with the reallocation of the full aggregate into vertical blocks in the
strings at rehearsal figure 13 (b. 87), just before the last vocal statement.
The agitation of the prevailing tone colour and metric groupings is
comparable to the Schoenberg example above and anticipates the dramatic
disclosure of the speaker’s identity as non-human (‘Who better than
70 Jessica Payette
Ex. 3.3 Full aggregate in strings, ‘Complainte du lézard amoureux’, bb. 87–8
a lizard in love can tell the secrets of the earth?’) with a sudden shift to a torrent
of aggressively articulated quavers supported by punctuating chords in the
strings at b. 87 (Example 3.3).
Temporal Modes
Schoenberg was one of the first composers to compellingly convey unstable
or shifting perspectives in large-scale vocal works. In his essay ‘New Music:
71 Schoenberg vive
My Music’, written around 1930, he reveals that his control of temporal

pacing was central to the dramatic effect of Erwartung and Die glückliche
Hand. He writes:
The common denominator of the two works is something like this: In Erwartung the
aim is to represent in slow motion everything that occurs during a single second of
maximum spiritual excitement, stretching it out to half an hour, whereas in Die
glückliche Hand a major drama is compressed into about 20 minutes, as if
photographed with a time-exposure.47
Schoenberg’s representation of psychically driven musical time that aims to

supersede clock time often intentionally leaves the listener ungrounded by
retreating from any regular scheme for the placement of structural sign-
posts and relying heavily on points of motivic, textural or instrumenta-
tional convergence for formal delineation. Schoenberg’s large-scale
Expressionistic vocal works – Das Buch der hängenden Gärten (1908–9),
Erwartung (1909), Pierrot lunaire (1912) and Die glückliche Hand
(1910–13) – explore two contrasting possibilities for the musical construc-
tion of identity: replications of an individual’s psyche and amalgams of
fragmentary points of view.
In his companion pieces Schoenberg’s achievement is the creation of
musically convincing representations of phenomenological or ‘human’
time. Erwartung exposes a distressed individual’s subjective, unfiltered
thought processes whereas Die glückliche Hand does the same in an
expanded, and thus selectively filtered, fashion. Boulez’s fascination with
‘discontinuous time’ leads him to poetry that annuls Schoenberg’s reliance
on a tangible protagonist, but both create text-based compositions in
which ‘the temporal dimension of the poem bears no comparison with
the chronometric time of the music’.48 Michael Cherlin draws on the work
of Henri Bergson and Paul Ricoeur when explaining that many of
the Second Viennese School’s free atonal vocal works avoid metric solidity
and a sense of regular rhythmic periodicity to frustrate our perception of
the pacing from the outset. Listeners respond to the steady pulse-stream of
‘time shards’ because they stand ‘in sharp contrast to the more elastic
sense of subjective time that had become the norm’ for Schoenberg and his
disciples.49
Yet, as described earlier and as Jonathan Dunsby highlights, in Pierrot
lunaire (and Das Buch der hängenden Gärten) Schoenberg also introduces
fragmentary glimpses of subjectivity to establish conflicted identity, or
47
Schoenberg, Style and Idea, p. 105. remaining in airtight containers’. Boulez,
48
Boulez, Stocktakings, p. 19. Boulez calls for Orientations, p. 341.
49
‘a concept of discontinuous time made up of Cherlin, Schoenberg’s Musical Imagination,
structures which interlock instead of p. 174.
72 Jessica Payette
a plurality of perspectives, through the combination of multiple poetic

voices, obscuration of the gender of the primary speaker and vague connec-
tions between the individual movements.
Whereas in Erwartung the protagonist, the main character, is a mysterious singer,

with no name, and hardly any identification, in Pierrot there is no protagonist at all,
no lucid relationship between the focus of attention, the woman reciter, and the
focus of textual attention, Pierrot himself.50
In ‘Speaking, Playing, Singing’ Boulez reports that Pierrot’s ‘art of con-

trasts’ has convinced him to regard the work as much more than a clinical
portrait of the psychological pathologies and angst of the fin de siècle. He
praises Pierrot’s enigmatic quality, noting that ‘the ambiguities that it con-
tains and Schoenberg’s bold ideas about the relation between words and
music represent an inexhaustible wellspring for the future’.51 Schoenberg
designs a carefully calculated fluctuation in narrative perspectives and
instrumentation to indicate that Pierrot’s ‘inner world is as illusory as his
whole existence’, an aspect which George Perle suggests is reflected in the
increasingly prismatic nature of the music as the work progresses.52 In his
investigation of Pierrot lunaire’s number symbolism Colin Sterne catalogues
how the numbers 3 and 7 dominate the numerological processes with regard
to pitch and duration, but he does not address the additive integer scheme
(1+2+3+4+5+6 = 21) that may also have been devised as a premeditated
aspect of the composition, especially in the manifestation of disjunct poetic
vantage points that traverse a diverse spectrum of implied voices: women’s
fondness for Pierrot; male lusting for feminine archetypes; the neutral
narration of Pierrot’s journey; Pierrot’s own voice.53 In Schoenberg’s com-
pilation of twenty-one poems, selected from a total of fifty in the complete
collection, six rondeaux are concerned with Pierrot’s actions (Nos. 10, 11, 15,
16, 18, 19), five feature purely descriptive imagery (Nos. 1, 5, 8, 14, 20),
four of the poems impart sentiments of feminine adulation (Nos. 4, 6, 7, 12),
three are in the first person (Nos. 2, 9 and 21), an omniscient narrator
appears in two poems and describes what Pierrot is thinking or feeling
(Nos. 3 and 13) and only one rondeau speaks of an anonymous admirer’s
love for Pierrot (No. 17).
50 52
Dunsby, Pierrot lunaire, p. 35. Perle, The Right Notes, pp. 34 and 36.
51
Boulez, Orientations, pp. 330 and 337. ‘An inspection of the distribution of the
Boulez recognises that Pierrot is ‘not mono- various ensembles throughout the work as
valent’, but he nonetheless views it as akin to a whole shows that the texture tends to get
a song cycle in which ‘one piece follows fuller and the instrumental variety richer as
another without any change in direction’. He the work progresses.’
53
views the multiperspectivity that I’m exam- Sterne, ‘Pythagoras and Pierrot:
ining here as ‘oblique references’ to people An Approach to Schoenberg’s use of
and ideas associated with Pierrot’s cabaret Numerology’, p. 513.
noir milieu.
73 Schoenberg vive
Boulez ruminates on ‘the alliance of music and poetry’ through

a consideration of how musical temporality differs from ‘the single, exact
datum’ of reading a poem. He believes that there are ‘two times’, or experi-
ential modes of perception, that the composer regulates when setting text to
music: ‘the poem as action is directly “taken over” by the music’ whereas ‘the
poem as reflection may be submitted to a kind of fragmentation or distortion
from its original form, may indeed even absent itself from the music, in
which it persists in the form of appended commentary’.54 In Le Marteau sans
maître Boulez inflates a concise poetic text into an expanse of music as a way
to unearth potential contiguities between incongruent modes of temporal
progression. In ‘Sound and Word’ (1958) he speaks again about his typical
approach to text setting: ‘a whole web of relationships will make itself
felt, including, among others, the affective relationships, but also the
entire mechanism of the poem, from its pure sound to its intelligible
organisation’.55 Le Marteau certainly exemplifies this strategy, yet its com-
plexity, which Boris de Schloezer cites as due in part to its rejection of
Webern’s ‘perfectly assimilated’ style in favour of greater elusiveness,
makes the work’s premises difficult for the listener to grasp on first hearing,
as was noted by Schloezer after Marteau’s Paris première: ‘Le Marteau sans
maître obliges us, yet again, to pose the general question of the relationship
of music and words; one asks oneself, indeed, what the role of Char’s text
is in the work’.56
Lev Koblyakov has already documented Marteau’s intricate serial pro-
cesses in great detail and he discovers that Boulez subjects five derived series
(achieved through rotation and transposition of the internal groupings of the
prime form [the proportion row 24213]) to frequency multiplication, which
generates five harmonic domains.57 In reviewing Pascal Decroupet’s facsi-
mile edition, Peter O’Hagan also reminds us that the proportion row (24213)
of the basic series, and its dispersion into vertical sonorities in the
‘Bourreaux’ vocal movement (VI), correspond to the number of syllables
in the poem’s individual words (i.e. there is one four-syllable word and one
four-note bloc sonore that correlate to the four-note group in the series);
furthermore, deviation from strict serial execution is often undertaken to
connote the poetic imagery.58
54
Boulez, Orientations, p. 196. analyses of the score, the relationship of
55
Boulez, Stocktakings, p. 40. music to poetry in Le Marteau remains
56
Schloezer, ‘À propos des concerts du a comparatively neglected field’ (‘From
Domaine Musical’, La Nouvelle Revue Sketch to Score’, p. 634).
57
Française 4/41, 930–2. The review is rep- Koblyakov, Boulez: A World of Harmony,
rinted in Decroupet, Le Marteau sans maître: pp. 3–5.
58
Facsimile of the Draft Score, p. 79. In 2007 O’Hagan, ‘From Sketch to Score’, p. 636.
Peter O’Hagan redirects attention to this With regard to localised diversion from the
matter: ‘To this day, despite several technical series for the ‘Bourreaux’ cycle, O’Hagan
74 Jessica Payette
Char’s austere poetry compelled Boulez to envision Le Marteau ‘as an

intellectual drama prompted by reading the poem and the echoes that it
creates in a world that is, properly speaking, interior’.59 The intellectual
drama is made more dynamic by Boulez’s addition of movement titles that
are not derived from poetry, but instead allude to enduring literary genres
(narrative, rhetoric and exegesis).60 Thus Le Marteau’s musical discourse
employs serial processes, variation of instrumentation and the aural discern-
ment of temporal relationships (the fluctuation of tempo within movements
and proportional ratios between movements) to accentuate the poetic
devices of a poem and/or the literary conventions of the cyclic genre.
The transference of these literary modes to music appears to influence the
temporal proportions and defining aural markers of the piece. The simplest
proportion (2:1) is applied to the movements of the linear ‘Artisanat’ cycle,
which proceeds according to a specified sequence of before, during and
after.61 Simply tallying the number of bars in each movement indicates
that movement I (‘Avant “L’artisanat furieux”’, 96 bars) contains twice as
many bars as movements III (‘L’artisanat furieux’, 48 bars) and VII (‘Après
“L’artisanat furieux”’, 48 bars). However, only movements I and VII share
the same ‘rapide’ tempo; the fast speed coupled with the temporal reduction
makes the period of ‘after’ seem incredibly aphoristic to the listener, as it
is the briefest section in this linear cyclic sequence and the entire piece.
The proportional dimensions of the ‘Bourreaux’ cycle, based on a tabulation
of the basic unit of a semiquaver, are also fairly simple, coming extremely
close to producing the more equally balanced ratio of 9½ (II): 7 (IV) : 8 (VI) :
11 (VIII), but Boulez creates a different trajectory for this cycle by lengthen-
ing the movements as the chronometric time of the piece progresses. Again,
the variation in tempo and instrumentation influences the listener’s experi-
ential journey: in movements VI and VIII the exclusion of entire sections
played in a fast tempo creates the sense of a uniform passage of time.
Conversely, the tempo changes of movements II and IV are also coupled
with idiosyncratic timbres (a kind of delicate accompanimental lattice with
pizzicato in the viola in movement II and vibrating overtones in movement
IV) that the listener will recall when confronting the more serious
writes, ‘Boulez modifies the serial parameters limits of Western tradition’. Boulez’s inter-
in order to shape a four-note cell – A♮, B♮, C♮, mixture of literary genres evokes a similar
E♭ – which contains within it the basic spirit of humanism, encouraging the listener
intervallic content of the cycle as a whole: to reflect on the longevity of genres that
a musical image of the pendulum’ (p. 640). shaped artistic and civic values in both
59
Boulez, Orientations, p. 339. Western and non-Western civilisations.
60 61
Salem, ‘Boulez Revised’, p. 165. Salem Koblyakov, Boulez: A World of Harmony,
offers an eloquent synopsis of Le Marteau’s p. 115. Koblyakov observes: ‘On the large
instrumentation: ‘the sounds are not meant plane, however, the proportions can be
to reference other cultures or places, but to simple.’
transport the composition itself beyond the
75 Schoenberg vive
hermeneutical challenge of determining how these commentaries may

enrich the poem’s meaning for the listener, a topic that is explored further
towards the end of this chapter.62
Boulez’s rejection of common-practice era approaches to phrasing, rhythm
and metre results in a wholly revisionist conception of periodicity (if we
accept the consensus view that Boulez does not abandon periodicity
altogether).63 His notion of smooth and striated time as ‘capable of reciprocal
interaction, since time cannot be only smooth or only striated’,64 is partially
an outgrowth of Schoenberg’s approach, as David Gable recognises
when proposing that ‘Boulez seems to have glimpsed one solution in
Schoenberg’s expressionist works. In an original conception extrapolated
from Schoenberg’s prose rhythms, Boulez for a time attempted to
abolish any sense of rhythmic regularity either in the large or in the smallest
details.’65 However, also analogous to Schoenbergian expressionism,
continuous streams of homogeneous sonorities and interlocking vocal–
instrumental and instrumental–instrumental units influence our perception
of the pacing and poetic meaning. These present opportunities for listeners to
create and recall memories within the confines of the piece, lending insight
into the work’s large-scale expressive trajectory. In Le Marteau these points
of impact arise through motivic consolidation or reconfiguration and have an
effect on the listener that can be likened to that of time shards, thus they often
emerge within zones that Boulez defines as ‘non-homogenous time’, in which
‘striated and smooth time can be alternated or superposed upon one
another’.66 A comparison of passages from ‘Bel édifice et les pressentiments’
(V) and ‘Bel édifice et les pressentiments – double’ (IX) demonstrates
62
The common denominator of 114 yields which had run contrary to the a priori
a proportion of 7:11 as there are 798 semi- postulate of non-repetition, had been tossed
quavers in IV and 1,254 semiquavers in VIII. out the door, only to return through the
Interestingly, the second (1,078 semiquavers) window. Remember the syllabic articulation
and sixth (907 semiquavers) movements each of the form in the instrumental movements
fall five semiquavers short of yielding that of Boulez’s Le Marteau sans maître’
precise ratio. (Schnittke, A Schnittke Reader, p. 148).
63
Decroupet, ‘Rhythms–Durations– Wayne Wentzel points out that the
Rhythmic Cells–Groups’, Unfolding Time, prominent musical characteristics in the
p. 82. This question of the need to divest ‘L’artisanat furieux’ cycle (irregular
contemporary music of periodicity emerges groupings, strings of similar durational
in Boulez’s correspondence with Henri values, and frequent grace notes) are
Pousseur around 1952. Decroupet markedly different from those in Boulez’s
summarises Pousseur’s stance, noting that he integral serial works, and designates the style
considered ‘every kind of regularity and as ‘typical of improvisation and common to
periodicity as a submission under the former “fantasia” movements in earlier, traditional
laws of tonality’. Boulez’s reintroduction of music’ (Wentzel, ‘Dynamic and Attack
periodicity into Le Marteau has been Associations in Marteau’, p. 163).
64
addressed often. In his essay on ‘Static Form’ Boulez, Orientations, p. 87.
65
Alfred Schnittke observes that subsequent to Gable, ‘Boulez’s Two Cultures’, p. 435.
66
the ‘academic’ examples of serialism Campbell, Boulez, Music and Philosophy,
(Kreuzspiel and Structures I) ‘periodicity, p. 235.
76 Jessica Payette
how Boulez adopts aspects of Schoenberg’s compositional practice to articu-

late structural properties and produce ‘uncanny effects’. The music’s artistic
ingenuity is predicated ultimately on its conveyance of experiential variables
culled from the superimposition of a number of musical parameters:
the listener’s perception of morphology is largely established through the
detection of points in the composition that teeter on the brink of familiarity
and by gauging how they relate to other sections.
The first rendition of ‘Bel édifice’ (V) is characterised by the predomi-
nance of melismatic vocal phrases which, combined with recurrent gestures
like slurred ‘sighing’ descents, stinging guitar attacks and an abundance of
trills, produce a consistent and memorable soundscape. The music’s empha-
sis on undulating contours and intermittent consolidation into homophonic
blocks grounds the listener and seems to portray perfectly the poem’s focus
on grotesque motion and viscerality. Compared to the first version of ‘Bel
édifice’, the ninth movement begins with a sense of despondency and
compression, the voice enters immediately, now syllabic and stripped of its
former sinuousness, and the capricious instrumental trills no longer fill
empty spaces. In both versions, Boulez’s conspicuous disruption of Char’s
stanzaic structure severs the poem’s central antithesis of ‘child’ and ‘man’
with the placement of a musical interlude between the couplet that com-
prises the second stanza. In fact, this couplet constitutes Marteau’s centre-
piece (the middle of the work’s mid-point), and the affect of the setting of the
first line, ‘Enfant la jetée – promenade sauvage’, is comparable to
a Schoenbergian time shard. In the fifth movement the shift to a slower
tempo at b. 63 and precise homophonic synchronisation in the instrumental
parts from bb. 63–6 invoke both a steady pulse stream and a consolidating
gesture reminiscent of concordant passages in common-practice era cham-
ber music,67 heard very rarely in the piece (Example 3.4a). When this line of
text returns in the ninth movement (b. 15) it is sung in an entirely different
fashion with stark vocal Sprechstimme, but Boulez clearly achieves
a manifestation of elapsed time that triggers a sense of déjà vu remarkably
similar to the examples Cherlin locates in Schoenberg’s free atonal works.
The viola plucks out the melodic vocal line from the earlier version: the
faintness and timbre of this musical quotation lends it a spectral quality.
The instrumental synchronisation (bb. 17–20) now follows the text with the
xylophone substituting for the flute (Example 3.4b).
67
‘Almost all later eighteenth-century chamber music of the time; and all
instrumental music can be understood as instrumental genres can be understood as
having conversational aspects; a heightened metaphors for social relations’ (Sutcliffe,
awareness of texture, as implied by the ‘Haydn, Mozart and their Contemporaries’,
imperative of “equality”, surely marks all p. 186).
77 Schoenberg vive
Ex. 3.4a Homophonic synchronisation in ‘Bel édifice et les pressenti-

ments’ (V), bb. 62–6
In both movements (V and IX), the instrumental partition placed

between the lines that begin with the words ‘child’ (enfant) and ‘man’
(homme) builds anticipation for the most strikingly alliterative line of the
entire piece in which the words ‘l’illusion imitée’ (‘the illusion imitated’)
emerge as vocally hollow (in the fifth movement the word ‘imitée’ is sung
sans timbre and the second iteration features dark intonations morphing
into Sprechstimme). The distance between child and man is portrayed by
a musical edifice, and the musical representation of ‘imitated illusion’ is
conveyed in the instrumental interludes first through a sprightly poly-
phonic panoply of slurred gestures (movement V, Example 3.5a), all
initiated by grace notes, which prefigures the voice’s imitation of these
instrumental trademarks in the ninth movement. This occurs in b. 24 of
the ninth movement (Example 3.5b), as the voice performs the gesture
and momentarily coalesces into the ensemble, quickly breaking away to
dramatically repeat the slurred gesture when commencing the word
‘l’illusion’.
Of course, Marteau’s conclusion, when the voice becomes fully com-
mensurate with the ensemble, suggests that the text itself is the illusion,
challenging the listener to ascertain if, and how, it is present when it is not
present.
Recent scholarship which identifies unusual musical syntax and non-
linear temporal episodes as musical evocations of literary and psychoanaly-
tical concepts like the ‘uncanny’ has elicited a heightened awareness of the
importance of musical allusions and illusions in the works of the Second
78 Jessica Payette
Ex. 3.4b Homophonic synchronisation in ‘Bel édifice et les pressentiments – double’ (IX),
bb. 15–21
79 Schoenberg vive
Ex. 3.5a The instrumental passage following the alliterative text ‘l’illusion
imitée’ in ‘Bel édifice et les pressentiments’ (V), bb. 72–77
Viennese School and their successors.68 With regard to Le Marteau, Boulez

states that ‘actual reminiscences and “virtual” relationships’ are crucial to
hearing ‘the cycles overlap in such a way that the course of the work becomes
increasingly complicated’.69 Yet further scrutiny of how Boulez’s works
embrace this legacy of Austro-Germanic modernism will open new avenues
to interpretive discovery.
Grundgestalt: Musical Interludes and Multi-perspectival

Morphology
As Jean-Jacques Nattiez intimates in his introduction to Orientations, many
of Boulez’s writings invoke the problematic notion of Grundgestalt,
68
For further discussion on psychoanalytic pp. 58–82 (Ch. 3: ‘Music and the Birth of
principles in relation to compositional pro- Psychoanalysis: Anton Webern’s Opus 6,
cedures in Second Viennese School reper- no. 4’).
69
toire, see Pedneault–Deslauriers, ‘Pierrot L.’, Boulez, Orientations, p. 342.
pp. 601–45, and Schwarz, Listening Awry,
80 Jessica Payette
Ex. 3.5b The alliterative text ‘l’illusion imitée’ in ‘Bel édifice et les pressentiments –
double’ (IX) displays the voice adopting instrumental gestures from Example 3.5a,
bb. 23–30
U
U
U
81 Schoenberg vive
a premise that is developed by Schoenberg in the period subsequent to his

self-professed exhaustion of free atonality in 1915. By coining this term,
Schoenberg affirms that numerous shapes, or compositional ideas, can
interact with the basic formal framework, as Célestin Deliège reinforces
when observing that for both Schoenberg and Boulez ‘it is the quality of
the internal substance which gives life to form and not the opposite’.70
Although, beyond the scope of this discussion, Boulez’s suggestion that he
views the entirety of his output as one large interconnected composition
attests to his reverence for the eloquence and viability of the
Grundgestalt principle, and is possibly even elided with Char’s own concep-
tion of his poetic oeuvre as an archipelago.71
Perhaps the most important, and under-researched, element that aligns
the music of Schoenberg and Boulez, and distinguishes it from that of
Webern, is a penchant for composing purely instrumental music as
a crucial means to project morphological formation. Webern’s musical
treatment of dark, psychological subject matter is certainly too important
to go unmentioned here, and perhaps makes its most sustained impact in the
Trakl songs op. 14, composed between 1915 and 1921. Anne Shreffler views
these songs as conveying ‘inward loneliness’ in contrast to Schoenberg’s
‘public loneliness’.72 As such, Webern adopts an entirely different musical
approach to represent a strain of expressionism that privileges pictorial
imagery as a manifestation of contemplative discontinuity. This is sharply,
and perhaps intentionally, differentiated from Schoenberg’s tendency to
emphasise stream of consciousness pacing and hyperexpressive individua-
listic reactions to disturbing thoughts or images, even when setting strict
poetic forms to music, like Giraud’s rondeaux. Shreffler describes Webern’s
approach as anomalous in relation to iconic German song cycles: ‘Eighty
years after Schumann’s Dichterliebe, there are no extensive introductions,
interludes, or postludes. The instrumental parts complement the voice;
though they articulate independent contrapuntal lines, they do not function
on their own.’73 As his valuing of compression escalates during this period,
Webern dispenses with instrumental interludes, essentialities that burgeon
from Schoenberg’s and Boulez’s fixation on ‘the verbal text as a kernel, the
centre around which the music crystallises’.74
70
Deliège, ‘On Form as Actually a metaphorical passage between poems and
Experienced’, p. 106. poetic collections. See the introduction of
71
Nattiez, Orientations, p. 23. Nattiez quotes Robert Baker’s 2012 translation of Char’s Le
from Par volonté et par hasard: ‘The different parole en archipel (The Word as Archipelago,
works that I write are basically no more than 1962) for further discussion.
72
different facets of a single central work, with Shreffler, Webern and the Lyric Impulse,
a single central concept.’ Char encourages p. 5.
73
approaching his work as an archipelago in Ibid., p. 44.
74
order for readers to embark on Boulez, Orientations, p. 341.
82 Jessica Payette
Christopher Morris demonstrates that in post-Wagnerian operatic

contexts musical interludes enrich the content and verbal delivery of
the libretto through the mimesis of concrete action and the expression
of metaphysical states. Schoenberg clearly engages with this develop-
ment in different ways: for example, by his refusal of vocal character-
isation for the Woman and Gentleman in Die glückliche Hand (they are
characterised exclusively through instrumental music and pantomime),
and the creation of passages like the instrumental transition between
the third and fourth scenes in Erwartung (bb. 114–24) where the
stacking of multiple tremolos and ostinati produces a whirling effect
and functions as a musical indicator of the Woman’s internal palpita-
tions. In his examination of the influence of instrumental interludes on
spectatorial response Morris draws on film theory, revisiting Christian
Metz’s claim that cinema has the ability to subvert the spectator’s
usual demarcations of ‘external’ and ‘internal’.75 While dichotomies
are perhaps too frequently applied to the music of Schoenberg and
Boulez, this is a productive dichotomy for best assessing the listener’s
perception of their vocal scores as multi-perspectival artworks in which
internal morphology defies the score’s external structural divisions.
In their vocal works Schoenberg and Boulez use musical interludes to
establish large-scale external interconnections (with motivic, metric,
timbral or gestural recurrences) that aim to project internal ‘poetic
propositions’ with instrumental music that replicates and aggrandises
aspects of the poetry’s structural makeup and essence. As such, the
ontological foundations of the musical work are predicated on display-
ing something even more engrossing than a convincing portrayal of the
text’s sentiments. For Erwartung, a work the composer himself
described to Ferruccio Busoni as ‘something entirely new’, this musical
response to the imaginative features of Marie Pappenheim’s text has
been described as ‘seismographic’ by Adorno, a comment on which
Cherlin elaborates, characterising the piece as ‘a crazed network of
fragments – some loosely connected, others clearly disruptive’.76
In the works discussed here, the external conventions of teleological
music are not eradicated (Marteau’s coda in movement IX definitely
evinces this), but they are subdued because the Grundgestalt construct
enables the music to simultaneously manifest the poetry’s broader
worldview.
75
Morris, Reading Opera Between the Lines, The Atonal Music of Arnold Schoenberg, p. 95.
pp. 15–16. Cherlin, Schoenberg’s Musical Imagination,
76
Letter from Arnold Schoenberg to p. 205.
Ferruccio Busoni (1909), quoted in Simms,
83 Schoenberg vive
As Char’s poetry garnered global acclaim critics recognised as its most

significant traits a forging of continual beginnings and an eclectic depiction
of habitat and habitation, vividly described by the philosopher Reiner
Schürmann:
Duration is the most captivating of all dwellings, the narrowest of all prisons. Its
dismissal is forcefully urged. Char cherishes the dawns. The early morning, the
moment without precedent, makes the world rise anew, immaculate. The moment of
waking is much more than the discovery of things familiar. It makes the world begin,
absolutely.77
A brief consideration of Boulez’s realisation of Char’s poetic proposi-

tion of time as habitat – his creation of a space ‘where music does not
distend time but can be grafted onto it’78 – begins to reveal how
Boulez’s implementation of the Grundgestalt principle is congruent
with Schoenberg’s. With respect to morphology in general, the
‘Bourreaux’ cycle is most radical due to its three instrumental commen-
taries, two of which precede the text setting. The exclusion of the
first person point of view from ‘Bourreaux’ makes it tangibly different
from the two other poems, yet it amalgamates the impending violence
of ‘L’Artisanat’ and the emphasis on walking and imitation in ‘Bel
édifice’. As many analysts have noted, the ‘Bourreaux’ cycle, comprising
the majority of the entire piece with four separate movements, is set
strongly apart from the other two cycles by its instrumentation,
which features unpitched percussion.79 This is the ‘Bourreaux’ cycle’s
provocative external attribute: it immediately entices with the prospect
of steady pulsation. Yet ironically the tangible foregrounding of
synchronicity is generated through the accentuation of stalled momen-
tum and declining inertia: an overriding Grundgestalt concept that
makes the commentaries compelling, and convincingly demonstrates
that the non-verbal medium of instrumental music magnifies Char’s
poetic priorities.
77
Schürmann, ‘Situating René Char’, p. 517. ‘The cycle Bourreaux is clearly pulsatory . . .
78
Boulez, Conversations with Deliège, p. 44. The specific serial technique of deriving the
79
Boulez notes that in the first part the pitch-structures resulted in a 12x12 table,
percussion appears in only the ‘Bourreaux where horizontally and vertically, each box
de solitude’ cycle ‘in which it marks the indicated a specific density, varying irregu-
time . . . The percussion thus plays larly from 0 to 4. Empty boxes were realised
a complementary part, filling with indeter- through interventions of the untuned per-
minate pitches the void left by determinate cussion instruments, so that the lack is only
pitches – a kind of architectural time game’ relative to pitch but not to rhythm’
(Boulez, Orientations, p. 340). Pascal (Decroupet, ‘Rhythms–Durations–
Decroupet explains in more detail: Rhythmic Cells–Groups’, p. 81).
84 Jessica Payette
One way of ‘reading’ the text setting of ‘Bourreaux’ (VI) is to regard

the motion of the pendulum’s oscillation as pervading the musical
surface. The simulation of the changing amplitude of its swing is
enunciated through a series of carefully positioned stopping points,
many of which are prolonged with fermatas of differing lengths.
The entrances of the maracas (bb. 6, 13, 21, 23, 25 etc.) generally
mark the continuous recommencement of this pattern and the shortest
amplitude is a period consisting of only eighteen semiquavers, set
to the text ‘Le marcheur’ (‘the walker’, bb. 21–2), the poem’s only
reference to a person. In the commentaries the impression of an
episodic habitat is also imparted through incessant stopping and
restarting as breaks and pauses infiltrate the micro- and macro-
structures. The basic formal layout of all four movements clearly
explores the perceptual ramifications of nested internal subsections
of varying lengths and tempos, demarcated by silences, within tripar-
tite forms. The absence of fermatas from the very first section of
the second movement and its clearly designated tempo changes
from ‘Lent’ to ‘Rapide’ to ‘Plus lent que le Tempo I’ announce the
basic tripartite structure. On a localised level, the flute and voice
melodies in the ‘Bourreaux’ cycle are generally characterised by
fragmentation as separation between individual attacks is the norm –
rests, breath marks or extreme registral distance occur between
pitches. For example, in the opening section of movement VIII
(bb. 1–50) the flute plays 134 pitches (including grace notes) and
only 86 fall consecutively (not separated by rests or breaths) within
an octave span of each other; similarly, there are a desolate 37 pitches
in the vocal part in the sixth movement and ten distinct gaps within
this delivery.
At times Boulez’s treatment of the Grundgestalt topos of regular
and irregular patterns of distance and separation brings forth
a mimetic rendering of oscillating waves. This is captured most directly
by the frequent repetition of an echo effect in which the progression
through the series is delayed as the same pitches ripple through various
octaves and instruments.80 The echo effect is very audible in movement
VI at bb. 17–20 (rippling of D♮ and A♮) and bb. 50–7 (rippling of
C–B♭–G♯–D–C♯) (Examples 3.6a and 3.6b).
80
Winick remarks, ‘I would suggest that the considers only bb. 44–9 (Winick,
numerous pitch repetitions in movement ‘Symmetry and Pitch-Duration
VI are subtle examples of text painting on Associations in Boulez’s Le Marteau sans
the part of the composer.’ He supports this maître’, p. 285).
hypothesis by a brief examination that
85 Schoenberg vive
Ex. 3.6a Undulation of repeated pitches in ‘Bourreaux de solitude’ (VI),

bb. 17–20
Undulating homophonic gestures or accompanimental patterns are

introduced at the beginning of each commentary, and pitch echoing
also defines many of the overtly contrapuntal passages, such as the
instance when the viola shifts from pizzicato to arco at b. 48
of movement IV. A micropolyphonic chain is initiated in which
nearly every pitch is immediately echoed by another instrument
(Example 3.7).
The regulation of duration in this cycle also creates another layer of
chaotic periodic behaviour. A systematic process that is derived from
integral serialism, pitch-duration associations (PDAs) are formed
by taking the pitch of the smallest rhythmic denominator (e.g. PDA on
C = beginning with a semiquaver on C) and adding a semiquaver to
each ascending semitone, then repeating the series on each pitch.
The first bar of ‘Bourreaux de solitude’ (VI) establishes a PDA
that begins on D (D = 1 semiquaver, E♭= 2 semiquavers . . . C♯ in the
vibraphone = 12 semiquavers) (Example 3.8a). Theorists have shown that
PDAs are employed quite rigorously in the text setting, but none under-
score the significance of the disintegration of the PDA system at the end
88 Jessica Payette
Ex. 3.8a Establishment of pitch-duration association (PDA) in ‘Bourreaux de solitude’

(VI), bb. 1–2
suggests, then it seems to reiterate that a state of controlled regularity is

neither attainable, nor desired.82
Michel Foucault designates Boulez as ‘the strictest and most creative
heir of the Vienna School’ and thus affirms that his friend’s exploration of
concepts introduced by Schoenberg and Berg are as consequential as those
of Webern.83 Beginning with Verklärte Nacht, still a work steeped in
Wagnerian aura and compositional techniques, Schoenberg’s practice of
modelling instrumental music on the deeper artistic and philosophical
axioms that emanate from poetic material eventually led to an enlarged
82
Koblyakov, Boulez: A World of Harmony, (mm. 126–138) seems to create the coda of
p. 52. Koblyakov does not elaborate on his movement 8’.
83
observation ‘that this last section of the form Foucault, Aesthetics, Method, and
Epistemology, p. 236.
89 Schoenberg vive
Ex. 3.8b Retraction of the PDA in ‘Commentaire III de “Bourreaux de

solitude”’ (VIII), bb. 132–8
function of instrumental music in twentieth-century vocal music, particu-

larly for the portrayal of abstract and interior strata of subjectivity. Boulez
commends Schoenberg’s ingenuity in this realm and further interrogates
many of his compositional breakthroughs and perspectives in his own
vocal works.
part ii
The Evolution of a Style

4 ‘A score neither begins nor ends; at most
it pretends to’:1 Fragmentary Reflections
on the Boulezian ‘non finito’
Robert Piencikowski
Following the paraphrase of Mallarmé in its title, this study is concerned

with certain paradoxical aspects of the Boulezian ‘non finito’. These
aspects are paradoxical since they do not follow the traditional orthodoxy
of previous commentaries, which often paraphrase the composer’s
commentary on his own creativity. The first signs of Boulez’s dissatisfac-
tion in relation to the completion of his compositions are found early in
his production. I am not speaking here of his youthful indiscretions, those
early works that were withdrawn from the catalogue on account of their
conceptual immaturity: there is nothing unique in the retrospective
judgements he brought to bear on those pieces composed in his appren-
ticeship years, something one can find equally in a number of other
composers. What is more striking is a habit of revision that would seem
to have arisen out of an obsessional concern for perfection, that has
intensified over the years and which has taken the notion of rewriting to
the extreme limits of composition. Hence the growing number of new
works that were announced and long expected only to be cancelled or
postponed at the last moment. In this sense, if one were to acknowledge
that his working habits changed according to circumstances on account of
the numerous additional responsibilities he had been prevailed upon to
assume, must one conclude from this that he made a virtue out of necessity
and that the interruption became a providential pretext freeing him from
having to complete certain works immediately or inopportunely? Or
might it be the case that this process, of which we perceive only the
exterior manifestations, responds to more profound motivations,
a growing awareness bound to the evolution and modification of his
concept of composition?
Before embarking upon this journey, we recall briefly, and hopefully
without pedantry, the fundamental ambiguity of the word ‘fragment’ – from
1
Scherer, Livre. This chapter is based on the Louvre invite Pierre Boulez. Œuvre:
text for a lecture that was given in the audi- Fragment, 6 November 2008 to 9 February
torium of the Louvre in Paris, 2009.
8 November 2008, during the exhibition Le
93
94 Robert Piencikowski
the Latin frangere : to break. From this we have the sense of a piece, debris,
ruins, remains, ashes, out-takes, wreckage; but also at the same time
rough outline, sketch: it links the untouched to the scrap, what is scarcely
begun and what remains after destruction – in opposition to completion,
perfection, completeness. To paraphrase Debussy, ‘the notion of the frag-
ment goes back to antiquity’: thus the fragments of Empedocles or those
of Heraclitus – which would lead us to René Char, whose poems from
Le Marteau sans maître would resemble as many fragments, scraps of
consciousness snatched out of the night. His forename: Re-né [re-born],
invites us to reflect on the re-naissance [re-birth] – as already suggested by
Chateaubriand, so dear to Berlioz. Closer to Boulez, we call to mind inter
alia the parcelled design of Michel Butor’s Mobile,2 Yves Bonnefoy’s
eulogy to the ‘unfinishable’,3 and Roland Barthes’ celebrated Fragments
d’un discours amoureux.4
We will spare the reader a prolonged foray into the domain of painting or
more generally of the visual arts; avoiding pouring a supplementary drop
onto the existing ocean of commentary and the plethora of exegesis, I will
recall only the genre of the non finito from the Renaissance, in which
Donatello and Michaelangelo stand in opposition to Leonardo da Vinci.
We can see this moreover in the catalogue for the exhibition in the context of
which this reflection was originally given.5
Musical history is no less rich in precedents. Without going back to the
Flood, and citing only one of the most celebrated examples from the
romantic period: Schumann’s Fantasy for piano in C major op. 17, com-
posed 1836–8, of which the autograph manuscript is subtitled Obolus auf
Beethovens Monument. Ruinen, Tropheen, Palmen. Grosse Sonate für das
Pianoforte (‘Ruins, Trophies, Palms [or: Laurels]. Grand Sonata for piano-
forte’). The struggle against the ephemeral – the melancholic contemplation
of ruins.6 And closer to the end of the twentieth century, we must not forget
three of the most celebrated fragments – Harrison Birtwistle’s . . . agm . . .,7
Luigi Nono’s Fragmente – Stille, an Diotima8 and György Kurtág’s Kafka-
Fragmente op. 249 – which demonstrate the permanent interest this form
holds for musicians and vindicate its incompleteness.
2
Butor, Mobile, 1962. by the John Alldis Choir and the Ensemble
3
‘L’imperfection est la cime’, trans. as Intercontemporain conducted by Pierre
‘Imperfection is the Summit’ (Bonnefoy, New Boulez.
8
and Selected Poems, pp. 38–9). Composed in 1979–80 and first performed
4
Barthes, Fragments, 1977. on 2 June 1980 as part of the 30th
5
See n. 1. Beethovenfest in Bonn, by the LaSalle
6
Rosen, Romantic Generation, 1995. Quartet.
7 9
Composed in 1978–9, subtitled ‘The Fayum Composed in 1985–7 and first performed
fragments of Sappho’, with translations by on 25 April 1987 at the Wittener Tage für
Tony Harrison, and first performed on Neue Kammermusik, by Adrienne Csengery
9 April 1979 at the Théâtre de la Ville, Paris (soprano) and András Keller (violin).
95 Fragmentary Reflections on the Boulezian ‘non finito’
Pierre Boulez did not await the work of his colleagues to make his own
contribution to the genre: something that is manifest in his penchant for
ellipses in his titles ‘Eventuellement . . .’,10 ‘ . . . auprès et au loin’,11 . . .
explosante-fixe . . .12 – with Stockhausen following in his footsteps with ‘ . . .
wie die Zeit vergeht . . . ’.13 He expressed his views on the subject in his last
course of lectures at the Collège de France, ‘L’œuvre: tout ou fragment’
(‘The work: whole or fragment’).14 To the point where, like mischievous
eccentrics, one ends up seeing fragments everywhere: all is fragment, the
fragment is all. Its legibility is especially problematic since when dealing
with form, the author often uses these terms as synonyms for the whole and
for detail. We might then ask why he did not give his course the title ‘Form:
whole or detail?’ He summarised this for us masterfully during the lecture he
delivered here two days ago, concluding with this quotation from René Char:
‘only traces stir our dreaming’,15 which I will inscribe conversely as an
epigraph to highlight what follows, inviting the reader, if not to dream, to at
least retrace retrospectively the imprints left by the composer on a journey he
set out upon almost three-quarters of a century ago.
*
To attempt to respond to all these questions, and for reasons of convenience,
I will begin first of all by dividing a significant part of this evolution
into three stages:16 (1) 1946–52; (2) 1953–65; (3) 1966 and beyond –
acknowledging that this arbitrary structure, adopted for the sake of
simplicity and clarity, cannot be applied rigidly to what in reality is much
more subtle and shifting. I propose only to dwell and to focus our attention
momentarily on three decisive points from his self-reflection, which led
him much later to question himself a posteriori on the particular topic
which is here considered. Taking account also of context, influences and
10
Boulez, ‘Eventuellement . . .’, trans. as font rêver’, trans. as ‘A poet should leave
‘Possibly . . .’ (Boulez, Stocktakings, traces of his passage, not proofs. Only traces
pp. 11–40). stir our dreaming’ (Char, The Word as
11
Boulez, ‘ . . . auprès et au loin’, trans. as ‘. . . Archipelago, p. 119). See Boulez, ‘Fragment:
Near and Far’ (Boulez, Stocktakings, entre l’inachevé et le fini’, lecture delivered in
pp. 141–57). Paris on 6 November 2008 (see n. 1), Boulez.
12
First versions composed from 1971 to Œuvre: Fragment, pp. 9–16.
16
1973 – see Dal Molin, ‘Introduction à la It did not occur to me to apply to Boulez
famille d’œuvres “. . . explosante-fixe . . .”’. the model proposed by Wilhelm de Lenz in
13
Stockhausen, ‘. . . wie die Zeit vergeht . . .’, his influential book Beethoven et ses trois
trans. as ‘. . . How Time Passes . . .’ (Die Reihe styles. As will be seen, it is less a case of
3 (1959), 10–40). defining three more or less stylistically dis-
14
Boulez, Leçons, 2005. The lectures were tinct periods than of focusing attention on
delivered in October 1994 and from February three experiences selected on account of their
to April 1995. exemplary and emblematic character, and the
15
‘Un poète doit laisser des traces de son compelling nature of their emergence.
passage, non des preuves. Seules les traces
socio-historic conditions, I hope to be in a position to outline the composer’s

portrait, while keeping in mind the inevitable degree of personal projection
which makes its way nolens volens into its composition.
I. The First Symptoms of ‘Revisionitis’

As we have noted already, we are setting to one side the youthful works
that were withdrawn from the catalogue, those works that we have
never had access to and which we know solely from various direct
sources emanating from the composer17 or which we have been able
to get to know as a consequence of their availability in the Boulez
archive of the Paul Sacher Foundation.18 Nevertheless certain pieces
have resurfaced dramatically, such as the Douze Notations pour piano
(1945), which were taken up again and orchestrated from 1978, over
thirty years after their composition. Furthermore, some have been
composed in parallel with others, which have breached the wall of
silence19 in the course of benefiting from the fragments recovered
from the pieces that were withdrawn.20 But from his first published
pieces, this has taken place only after they have been subjected to
substantial modifications, like in the cases of the Sonatine for flute
and piano and the First Sonata for piano, both of which were revised
carefully before the acceptable version was released.21 If the Second
Sonata – with the exception of some details – remains close to intact
from its first edition,22 the same cannot be said for Le Visage nuptial, Le
Soleil des eaux and the String Quartet, all of which were transformed in
the course of the following decades – to the extent that one hesitates in
considering each new version that is proposed as definitive: at most one
could speak of successive distinct versions, each one presenting
a particular state of the musical material, without the successor invali-
dating the previous one or vice versa. Yet the composer continues to
17
Goléa, Rencontres; Bennett, ‘The Early Boulez (hereinafter ‘PSS’): Piencikowski and
Works’, in Glock (ed.), Symposium, Noirjean, Sammlung Pierre Boulez.
19
pp. 41–84: the article was written at least ten ‘Je franchis le mur du silence’, Nattiez,
years beforehand, as is clear from a letter Boulez–Cage Correspondence.
20
from Glock to Boulez dated 21 May 1975 Quartet for Ondes Martenot, Sonata for
(Paul Sacher Stiftung, Sammlung Pierre two pianos, in which the movement titled
Boulez). That is to say, Bennett will have had Passacaglia forms the basis for ‘La Sorgue’ in
access to the composer’s personal archives Le Soleil des eaux (PSS).
21
long before they arrived at the Paul Sacher Gärtner, Werkstatt-Spuren.
22
Stiftung (1986). Tissier, ‘Variations-Rondeau’.
18
The list can be found in the inventory for
the Paul Sacher Stiftung, Sammlung Pierre
exercise a categorical right of veto over those interpreters who

would have the temerity to revisit previous versions, albeit with the
historic reconstruction of original forms in mind. Then follows
the (strictly speaking) experimental period, with the two versions of
the Polyphonies which, despite the composer’s first intentions, have
remained in an abandoned state since the première of Polyphonie X,23
as well as the two musique concrète studies.
We come now to those compositions that have become milestones –
which are noteworthy, by dint of their position in his catalogue of works,
their historic dimension, indeed their public success, as fixed points in
his output, a two-faced Janus: the first book of Structures for two pianos
(1951–2) and Le Marteau sans maître (1952–5). If the first work appeared
to be the outcome of previous research, the second marks an impulse
towards a new compositional beginning. But there again, one must qualify
this: for not only was Structures I conceived as the first part of a large-scale
project,24 of which the Deuxième livre (1956/1961) was the only new stage
that was completed25 before the project was abandoned definitively, but Le
Marteau, which seems to be completely folded in upon itself, proceeds
from material recovered from diverse sources,26 and in turn provides
material for scores that apparently have no direct link to it: ‘Don’ and
‘Tombeau’ from Pli selon pli, as well as cummings ist der dichter, an
unexpected later flowering from Oubli signal lapidé.
We mention, without dwelling too long on it, the very particular case
of a forgotten composition, the Symphonie concertante (1947) that was
lost by Boulez in 1954 and which, despite this, or, paradoxically, for
precisely this reason, continued to exercise a subterranean force on future
compositions.
23 24
Contrary to the composer’s remarks, The composer described Structure Ia ret-
Polyphonie X was not withdrawn immedi- rospectively as ‘a scrap of possibilities among
ately after its première at the Donaueschinger an eternity of other possible combinations’
Musiktage für Neue Tonkunst (SWR (‘un lambeau de possibilités au milieu d’une
Sinfonie-Orchester, conducted by Hans éternité d’autres combinaisons éventuelles’):
Rosbaud, 6 October 1951). As can be Boulez, ‘Nécessité d’une orientation
observed through consulting the documen- esthétique’, Points de repère I, p. 568. These
tation held at the PSS, the composition had at were originally Pierre Boulez’s Horatio
least three further performances: Los Angeles Appleton Lamb Lectures at Harvard
(6 October 1952, conducted by Robert Craft), University, ‘The Necessity of an Aesthetic
Naples (11 May 1953, conducted by Bruno Orientation’, delivered on 9–11 April and
Maderna) and Barcelona (27 January 1954, 7–9 May 1963.
25
conducted by Jacques Bodmer). A project to Some sketches for a Troisième livre are to
revise and complete the piece had been be found at the end of the file for Structures
envisaged and the score was announced, in I and II (PSS).
26
the first French edition of Penser la musique [Trois Essais pour percussions], Oubli
aujourd’hui, 1964, as being ‘in print’ to be signal lapidé (PSS).
published by Editions Heugel.
II. The Development of a Formal Idea,

and its Compositional Consequences
From this point onwards, one sees in Boulez the crystallisation of the
conflict between two antagonistic traditions: the Germanic and Franco-
Russian conceptions of formal development.27 His compositions from this
period represent an attempt to find a solution to the tension between the
principle (prestige) of Durchkomponieren (Schoenberg, Berg, Webern)
and that (audacity) of development through alternative interruptions
(Debussy: Jeux, 1913; Stravinsky: Symphonies of Wind Instruments,
1920).28 Not that this fusion had no known precedents, be they Berg
(Lyric Suite, 1925–6) or Bartók (Music for Strings, Percussion, and
Celesta, 1936). Once again, the division, which this time is binary, must
not be understood in too schematic a manner since the overlapping of the
two principles is more subtle.29
Moreover, Boulez’s perspective benefited from historic socio-economic
conditions that proved to be particularly favourable to formal experimenta-
tion. From 1949 to 1960, thanks to the Marshall Plan,30 the West German
radio stations were the recipients of important subsidies, which fell like
manna from heaven on the radio symphony orchestras – hence the great
flexibility in rehearsal time they were able to offer. The Südwestfunk was able
to launch its activities from 1946,31 just as the Donaueschingen festival could
do from 1950.32 Budgetary restrictions only commenced with the beginning
of television (1961), and continued with the economic recession of the
1970s. When Boulez assumed responsibility for two important symphony
orchestras (the BBC Symphony Orchestra, 1971–5; the New York
Philharmonic, 1971–6), he was subjected to the full force of trade union
27
Gable, ‘Boulez’s Two Cultures’, 1990. constitute one of the most important periods,
28
Prestige: the authority of the Austro- in the same way as the resurgence of dode-
Germanic tradition as considered from Paris; caphony in the aftermath of World War II.
audacity: the daring counterpart imagined by Boulez’s repeated public declarations, stating
Debussy and Stravinsky to compete with the his claim of taking over first of all the
Austro-Germanic models (again as perceived influence of the Germanic musical traditions,
in Paris – and St Petersburg). in opposition to the alleged lack of an
29
After almost two centuries of relentless equivalent French musical tradition of the
struggle against the influence of Italian same standard, is just another manifestation
music, music in France was the site for of this ancient conflictual relationship –
permanent conflicts regarding the increasing deliberately intended to exasperate the
influence of Germanic music since the death well-known chauvinism of conservative
of Beethoven (1827), above all following the French musicians.
30
conflict of 1870: the ‘inferiority complex’, Signed 3 April 1948 following the inter-
which grew among the French towards the vention of the USSR in Czechoslovakia and
Germans, resulted in them defining accepted by West Germany in 1949: Hogan,
themselves for or against the influence of Marshall Plan.
31
music emanating from beyond the Rhine. Heyen and Kahlenberg, Südwestfunk.
32
The disputes revolving around Wagner Häusler, Donaueschingen.
restrictions concerning the timing of rehearsals – particularly in the

United States – and he was consequently inclined to revise certain scores:
Le Visage nuptial, for example, where he abandoned the refinement of the
quarter-tone writing that was originally employed in two of its movements;
or the rewriting of ‘Don’ and ‘Improvisation III sur Mallarmé’ from the cycle
Pli selon pli, where those sections that gave a degree of initiative to the
performers were fixed ‘definitively’.
In retrospect, we can consider a number of different types of revisions and
stages of abandonment, according to the various ways in which the early
versions of works have been modified:
– ‘raw material’ work: L’Orestie (recovered material)33
– works in expansion: Third Sonata, Pli selon pli
– interrupted works: Figures – Doubles – Prismes, Structures II
– abandoned works: Strophes
– fixed works: Le Marteau sans maître
This articulatory principle on two alternative planes is familiar in the theatre,
designated as a technical convention for private conversation (an aside).
Several citations taken from Boulez’s writings and correspondence show him
grappling with the difficulties encountered in the course of realising this
project – first of all during the composition of Le Marteau sans maître
(1952–5):
I believe that one must accept increasingly that not everything is determined
and it would be more satisfying for the mind – less essentialist – not to create
a hierarchy before commencing, but to discover this hierarchy as we go along
with the work. I believe that this is not yet the case. But late Debussy is there to
show us the way. A ‘work’34 perpetually ‘in progress’35 (dear Joyce). Thus one
would be led to compose without sketches, which would be very pleasant!!
The sketches would be made in the course of the work and not before. I intend
to integrate that into the variation principles (generative principles) which
would themselves be submitted to a vertical and horizontal serial universe.
Consequently this would not be a question of muddled variations, which would
be too related to the old working methods, but rather part of a construction
where the materials are renewed, reappear and always combine in different
ways. The form would no longer be envisaged in time such as an organising,
globally perceptible hierarchy (: assisting memory, by habit); but the form
would only be perceptible in the continuity of its unfolding. I believe that it is
an extremely important problem (see Debussy’s étude in fourths and Jeux).
It would in any case be a plausible synthesis and very fascinating for the
language as such with the very structure of this language deferred to a superior
level.36
33 36
O’Hagan, ‘L’Orestie’. Letter to Stockhausen (No. 23; c.
34
Original in English. 26 April 1953), Paul Sacher Stiftung,
35
Ibid. Sammlung Karlheinz Stockhausen
I wish only to propose for now a musical work in which this division into
homogeneous movements would be abandoned in favour of a non-homogeneous
distribution of developments. Let us claim for music the right to parentheses and
italics . . . a concept of discontinuous time made up of structures which interlock
instead of remaining in airtight compartments; and finally a sort of development
where the closed circuit is not the only possible answer. / Let us hope for a music that
is not this series of compartments to be visited willy-nilly, one after the other. Let us
try to think of it as a domain in which, in some sense, one can choose one’s own
direction.37
It is a situation which seems close to being resolved in the composition of

the Third Sonata, which responds to the experiments of his North American
challengers (John Cage, Morton Feldman, Christian Wolff, Earle Brown)
and Stockhausen’s Klavierstück XI. Through his writings and correspon-
dence, we can follow the development of the idea progressively, at first
confused and emerging gradually from the mist – as the title of Pli selon
pli will soon suggest:
I’m working at the moment, and it’s difficult, to succeed with this new form of
work that I have a glimmer of and which is difficult to grasp. It’s always very
difficult. For me, the devil is hidden in a labyrinth. He is giving me the
runaround38 but nothing to do with Ariadne. I’m sometimes afraid of producing
something that is completely unbalanced, insufficiently constructed, or more
exactly too damaged!39
Here I think a reasonable objection could be raised: does not such a form run
the great risk of sectionalisation? Are we not liable to fall into one of the errors the
most damaging to composition in the true sense, the error of simply juxtaposing
self-contained ‘sections’? This is a reasonable procedure only so long as one is not
actually thinking of the overall form, but simply developing from hand to mouth.
To offset such compositional renunciation, one must have recourse to a new
concept of development which would be essentially discontinuous, but in a way
that is both foreseeable and foreseen; hence the need for ‘formants’ of a work, and
for that ‘phrasing’ which is so indispensable to the relating of heterogeneous
structures.40
As I say in the article, a dangerous and open sectionalisation is the worst enemy of
form, at least so it seems to me. What I’m looking for in the domain of a non-closed
(photocopies). Boulez and Stockhausen first one some trouble’, or ‘giving a headache’,
met in Paris in March 1952, during here used as a link between the devil and the
Stockhausen’s attendance at the Messiaen following allusion to Ariadne’s thread.
39
classes at the Conservatoire (8 January to Letter to Stockhausen (No. 119; Paris,
30 June 1952): Kurz, Stockhausen. August 1957), Paul Sacher Stiftung, Sammlung
37
Boulez, ‘Recherches maintenant’, trans. as Karlheinz Stockhausen (photocopies) – my
‘Current Investigations’ (Boulez, italics.
40
Stocktakings, pp. 15–19) – my italics. Boulez, ‘Alea’, lecture delivered in
38
‘Il me donne quelque fil à retordre’: Darmstadt by Heinz Klaus Metzger in his
literally, ‘He is giving me some thread to own German translation, 27 July 1957,
rewind’ (or ‘some wire to be twisted’), from translated as ‘Alea’ (Boulez, Stocktakings,
untranslatable French expression ‘donner du pp. 26–38).
fil à retordre’, meaning approximately ‘giving
form, is rather a state of suspension in time, as one says of a body that is in suspension
for example in a liquid. To get to this state of perfect dilution, is more difficult than to
leave the form at one’s disposal.41
The Sonata is threatening to go on endlessly [. . .] and it’s ending up more a Book
[Livre] for piano than a sonata. The problem with its form is awful. I’m trying out
some mobile frameworks against fixed frameworks. Work has never been so
hard [. . .] it will be unperformable in complete form. At least such as I still imagine
it, unless an entire evening is dedicated to it [. . .] The damned thing is that I always
conceive these serial novels ! I’ve never finished on time and I always find myself
dragged much further on than I foresaw.42
Everything is justified with extremely guided chance.43 I’m working intensely on
all of that at the moment trying to find a structurally aleatoric form, that can be
combined with fixed forms. I believe I’m on the right track. But how many destroyed
efforts had to be gone through beforehand . . . I have never moved in such
quicksand. I’m reminded a great deal of an image in a film by Calder (Museum of
Modern Art) where at a given moment the mobiles were illuminated only in black
light, and one saw every now and then the structure of the mobile through coloured
shards. One saw the mobile through its instantaneous reflections; in another
sequence, one saw the mobiles ‘fixed’ for a few seconds, that is to say that their
movement became a succession of ‘fixations’. This will explain clearly to you (or
fairly clearly . . .) what I’d like to achieve . . . Without forgetting Mallarmé (but
willingly forgetting the unfortunate and nice but dim Cage!).44
I sent you two or three days ago this absolutely extraordinary book on Mallarmé’s
‘Livre’. I found it on the way home from Berlin and I’ve been absolutely stunned and
overwhelmed by its conclusions which corroborate exactly everything I was in the
midst of researching in the 3rd Sonata. / It’s all there. Unbelievable! And he
conceived it in 1890! That has given me a push forward. It’s a miraculous encounter.
And I’m even reworking my formants, for I’m haunted in particular by the idea of
thickness. The form of this Sonata is taking shape. It will have a length I no longer
dare to predict . . . [this is followed by a technical description of the five formants]
I believe that it is a work which can become something! If I get to the end of it [. . .]
I now feel the strength of combining Joyce and Mallarmé, which is my dearest
desire [. . .] Read this book. Read this book. We are going to overturn all concepts of
form and snatch something extraordinarily important. / Dear Karlheinz, I have been
in a hurry to share this epiphany with you. Now that we have a sufficiently solid and
quite a broad basic technique, we must now work madly on poetics. In the form
41
Letter to Pousseur (No. 60; end Stiftung, Sammlung Karlheinz Stockhausen
of July 1957), Paul Sacher Stiftung, (photocopies) – my italics.
43
Sammlung Henri Pousseur – my italics. ‘Alea’ (see n. 37).
44
The formula ‘en disponibilité’ (in availability) Letter to Stockhausen, No. 122,
unconsciously anticipates Earle Brown’s 27 September 1957, after the première of the
Available Forms (1961). Boulez and Pousseur Third Sonata in Darmstadt on
first met in Paris in June 1951, during the 25 September 1957, and before its reprise in
symposium Décade La musique et le cœur, Berlin on 28 September 1957. Paul Sacher
organised by Boris de Schloezer at the Stiftung, Sammlung Karlheinz Stockhausen
Abbaye de Royaumont (5–15 June 1951): (photocopies) – my italics. Boulez and Cage
Wangermée, Souris. first met in Paris in May 1949, during Cage’s
42
Letter to Stockhausen (No. 120; sojourn in Europe (30 March to
beginning of September 1957), Paul Sacher 30 September 1949): Boulez, Cage,
Correspondance.
I envisage for this Sonata, I have (1) guided chance (2) a chosen labyrinth (3) a break
in time (4) assumed structure (5) a cycle enclosed by initials [sigles] but open
through the possibility of its renewal – one still needs therefore a principle of identity
between the first initial and the final initial. / The work therefore arises perpetually
from itself. Creation, which once begun no longer comes to an END. Much work
remains to be done, but this thought sustains me – the wind arose. At last!45
Are you immersed in innovation? I am; but as if I am in a burrow. I set up my
labyrinths and my bubbles. Do you not find that these works with multiple
bifurcations recall the construction of the burrow, by our dear Kafka? It is a sickness
as old as the world and the tower of Babel (truth to tell, it was on high and above,
while Kafka was looking underground!).46
The result of this realisation is manifest in the arch forms of the

Third Sonata and Pli selon pli – which are like mirror images of one
another, in that both altarpieces take shape and unfold alternatively
around a central axis.
III. Where Compositional Technique, Form

and Compositional Act Meet
This development leads us gradually to the middle of the 1960s, a decisive
and crucial moment where circumstances drove Boulez to take on five
jobs simultaneously: in line with the demands made by his career, he
was effectively at once, by turn and simultaneously a composer, essayist
(theoretician, critic, polemicist), performer, teacher and organiser.
The compositions from this period can moreover be placed in five
categories, according to their different degrees of completion:
– sketched works: Marges47
– interrupted works: Éclat/Multiples
– initiated works: Domaines, . . . explosante-fixe . . .
– revised works: cummings ist der dichter
– fixed works (completed ?): Rituel, Messagesquisse
It is precisely at this moment while mentioning the compositional project
Éclat/Multiples that he glimpsed a kind of ‘compositional state’ within which
he starts to move himself:
45 46
Letter to Stockhausen, No. 123, beginning Letter to Stockhausen (No. 142, beginning
of October 1957, after the reprise of the Third of July 1959), à propos ‘Improvisation III sur
Sonata in Berlin on 28 September 1957. Paul Mallarmé’. Paul Sacher Stiftung, Sammlung
Sacher Stiftung, Sammlung Karlheinz Karlheinz Stockhausen (photocopies).
47
Stockhausen (photocopies). This obsession On this unpublished composition, and its
with re-naissance (re-birth) is found subterranean impact on Boulez’s output, see
throughout all of Boulez’s work. Bassetto, ‘Marginalia’, in Leleu and
Decroupet, Boulez, pp. 255–88.
A propos a different notion of temporality, of cyclical works which would seem to

have no beginning or end – one listens to music for a very long time in India or
Japan, one enters, one leaves, one listens, one gives up in a time span very different
from our own – I can say also that on the creative plane, I live in a kind of plasma
which allows me to move around gliding backwards and forwards. I remain within
the same thing and I irradiate in several directions at once. I now have supple
material which allows me to shift about in time and these re-creations. That’s how
I have made several versions of Pli selon pli and I’m thinking of expanding Éclat.48
This led him to an awareness of the autonomy of the sketch, which is

manifest in the choice of the title: Messagesquisse (1976), a portmanteau
word, whose models can be found either in Finnegans Wake or in Through
the Looking-Glass. As a result of this, the categories which are defined
provisionally continue in the form of the following works:
– initiated works: Répons, Dérive 1, Dérive 2, Dérive 3
– revised works: Notations, cummings ist der dichter, Dialogue de l’ombre
double, Anthèmes
– fixed works (completed?): sur Incises
The case of Notations would in this sense be emblematic: like the famous
ouroboros, a dragon which bites its own tail, the work folds back on
itself, the long-disregarded juvenile miniatures being transformed into
showpieces. But even in the case of the revisions for Pli selon pli, notably
those affecting the form and performance plans for ‘Don’, ‘Improvisation
III’ and ‘Tombeau’, the composer took care to specify in the form of
a warning to the reader ‘a fixed version’, implying that the current realisa-
tion presents only one of many possibilities for successfully completing
the work’s practical notation from the original model. Its fundamental
formal opening has been duly settled, its circumstantial closure however
remains provisional.
I note in passing: consultation of the sketches of composers such as
Stravinsky and Webern among others, as well as the acquisition of his own
archive by the Paul Sacher Foundation (1986), thus making it immediately
accessible for musicological research, will certainly have contributed in
precipitating Boulez’s reflection on the question of incompletion. Be that
48
Interview by Martine Cadieu (1924–2008) Boulez conducted a concert on 30 September
in London, 29 October 1966, intended for with the BBC Symphony Orchestra (works by
release on the occasion of the concert- Debussy, Webern, Stravinsky and
discussion ‘Orient – Occident’, organised by Volkonsky): the event was disrupted by
UNESCO in Paris for 2 November 1966, with demonstrations, following the publication of
Jack Bornoff, Yehudi Menuhin and André his interview with German journalists
Schaeffner in attendance. Published as Schmidt and Hohmeyer, recorded in
‘Musique traditionnelles – un paradis perdu?’ Bayreuth during the summer of 1967, pro-
This issue appeared shortly before the vocatively entitled ‘Sprengt die Opernhäuser
Berliner Festwochen (from 24 September to in die Luft!’, trans. as ‘Opera Houses? – Blow
11 October 1967), in the course of which them Up!’
as it may, once again he will not have waited for Paul Sacher’s offer
before composing a piece with as unintentionally premonitory a title as
Messagesquisse (1976), in which the title itself embodies the notion of
a rough outline: as if he was unconsciously sensing the role Sacher would
soon play in the preservation of his own manuscripts.
*
Thus, Boulez managed to adapt his compositional technique according to
the circumstances prevailing with regard to the resources at his disposal.
In a way parallel to exterior influences (musical, literary, pictorial, sculptural,
architectural, scientific), his output reflected the post-war economic curve:
following the period of need came one of economic expansion – with, as
a corollary, the generosity of the German radio stations (permitting periods
of prolonged rehearsal time for formal and acoustical experimentation) –
followed in turn by the economic recession, inversely marked by a form of
pragmatism linked to union constraints. To simplify, we can say that his
early utopian idealism was in time replaced by the pragmatic realism of his
maturity. His reflection on the fragment demonstrates his becoming con-
scious of the double movement of his professional experience: between
interior necessity (the work as labyrinth, formal relativity) on the one
hand, and exterior constraints (limitations on rehearsal time, professional
interruptions) on the other.
The direction he embarked upon at the start of his career, which
was provisionally frozen as he approached the period of structural
contraction (1949–52), was precipitated around the time of formal
pressure (1953–65), becoming a state of affairs after 1966, a period of
expansion. As a result, we recognise that what we think we perceive within
a work is in fact only the provisional emergence of a work in gestation –
islands within an archipelago of which one sees only their summits
emerging from the surface. Just as a glass may well appear to be half
empty or half full, depending on one’s turn of mind, the work reveals
itself simultaneously as subdivided into – as well as composed of – parts,
movements, pieces or cuts – in other words: fragments.
Following on from this review, we can now draw attention to some
particular cases, by way of conclusion:
– the cadential gesture can manifest itself in the shape of pre-composed
codas, the composer reserving the freedom to insert later developments
after their composition (Répons, Dérive 2);
– the physical nature of the object can embody the very principle of the
composition: for example the score of Trope, in which the spiral binding
of the notebook reflects the circular structure of the piece; or again
‘Constellation’, which was initially planned, but again left unrealised,
before unfolding itself like the folds of a fan, the reverse side presenting
the inverted form of Miroir;
– the arrangement of the instrumental apparatus makes the mobility of the
form apparent in the performance space: electro-acoustic spiral (Poésie
pour pouvoir), concentric circles (Domaines, Rituel, Répons) – highlighted
in the actual physical space of the concert hall; hence the repeated requests
for architects to design mobile concert halls, also in order to avoid wasting
time between different stage settings;
– revision can affect the work just as much in the expansion/contraction
of form (duration) as in its instrumental ensemble (density) –
sometimes at the cost of dislocating the polyphony of antiphonies:
‘Tombeau’, Figures – Doubles – Prismes: dense polyphonies made way
for loosely spaced antiphonies. ‘Tombeau’ was originally conceived for
instrumental groups that were positioned on the four sides of the hall –
responding to the layout of the loudspeakers in Gesang der Jünglinge
and the arrangement of the ensembles in Gruppen and Carré – with
their rotational effects around the centrally placed piano; hence the
disappointing, flattened result of what originally constituted a relief.49
Not to mention the problems with synchronising the groups, whose
entries were originally conceived as independent; nor the compromise
solutions adopted within the cycle Pli selon pli. Boulez attempted to
resolve this dilemma in each of his later compositional projects: Éclat
(1965: a slimmer ensemble with concertante piano, solo antiphonies);
Domaines (1968: medium-sized ensembles, clarinet soloist, alternative
obligato antiphonies); Rituel (1974–5: medium-sized ensembles,
heterophonies controlled by supplementary conductors, alternating
group antiphonies) – up to Répons and the successive revisions of
‘ . . . explosante-fixe . . . ’. Répons ’s ‘subsidiary’ pieces (Messagesquisse,
Dérive 1, Dialogue de l’ombre double, Anthèmes, Incises, etc.) may be
considered moreover as ‘out-takes’, satellites that have become
centrifugal meteors having escaped the central gravitation of their
mother-planets. In this way, the prophecy heralded in ‘Current
Investigations’ was fulfilled;50
– the consecutive compositions pursue the goal of solving the problems
raised by the previous ones, as a result of which the very concept of
completion is fundamentally incompatible with this vision of intermittent
composition.
49 50
Boulez, ‘Tombeau’. ‘Utopias? Let us realize them . . . now it is
time to smash some of our worn-out habits.’
(See n. 34.)
To understand the specific nature of Boulez’s compositional evolution, we

need to revise the categories of our musicological vocabulary, reconsidering
it in order to adapt it to the flux of his un-interrupted creation – I would be
tempted to write: in-interruptible, interruption marking not a break from
composition, but a composition that is extended in a subterranean way in
order for it to rise up again in the manner of the ‘fontaines du Vaucluse’, so
dear to Petrarch and René Char: remember ‘La Sorgue’ in Le Soleil des
eaux.51 In this way, one must call into question the notion of a score52 in
order to restore its etymological sense of sharing53 – and combine it with that
of partitioning54 – echoing the idea of re-verberation (in both the acoustical
and optical sense): to share out55 would be at the same time the beginning of
a new departure.56 All of which brings us back to the reflections that were
sketched at the threshold of this journey, à propos re-naissance:57 ‘perpetual
alternation’.58
The performance of the work in concert then becomes the ritual celebra-
tion of the work accomplishing its re-birth at the very moment of its
re-emergence from the silence. It has been observed that in the course of
Boulez’s evolution, the still ‘coarse’ sonority of his youthful works gave way
increasingly to the ‘smoother’ sonorities of his mature works, the harshness
having been progressively eroded with the passage of time. This change
comes at a price. Observing the composer’s work through the study of the
sketches to which he has given free access over the last thirty years, one
observes the following recurrent phenomenon: a piece tends to become
internally disproportionate as the composer goes ever further in the work
of composition. Hence the tendency to revise the initial bars with the
intention of rebalancing them with what follows. But this resumption
takes place at a moment of stylistic growth which offers few connections
with the style with which he began: hence the perpetual reorganisation – will
I chance a play on words on ‘Pierre’s re-revision’?59 A kind of infernal
vicious circle, ultimately saved by what he designated finally as ‘spiral
form’. This resembles strangely the desperate burden of the protagonist in
51 54
‘I could decipher the city through its In French: répartition.
55
fountains, and I can try at any moment to In French: répartir.
56
decipher myself in them’: Boulez, ‘Les In French: re-partir.
57
fontaines de Paris’, unpublished text for Baby, ‘One must give birth to oneself anew
a documentary in the series Passion, broad- every morning’ (‘La naissance à soi-même
cast on TF1 in September 1983 (PSS). doit s’accomplir chaque matin’).
52 58
In French: partition. ‘L’alternance se perpétue’: beginning of the
53
In French: partage – see Claudel, Partage motto placed by Boulez as an introduction to
de midi, 1905; first performance by the Rituel in memoriam Bruno Maderna, 1975.
59
Compagnie Madeleine Renaud-Jean-Louis In French: le remaniement de Pierre – thus
Barrault at the Théâtre Marigny, Paris, the pun on le reniement de Pierre (The Denial
16 December 1948, with incidental music by of Peter in the New Testament).
Arthur Honegger, conducted and performed
by Boulez. See Steinegger, ‘Boulez et Claudel’.
Fritz Lang’s Metropolis, who is condemned to switch off the light bulbs on

a gigantic clock while they are flashing, and which speed up to the point of
transforming him into a kind of malfunctioning spinning-top.60 It is there-
fore by dint of revising his preambles that the composer put himself at the
service of the work, ready to cross the threshold anticipating its initial
upbeat: preceding the anacrusis that is absent from every composition.
Parodying an author who is poles apart from Boulez’s universe, will I yield
to the temptation of the sin of disrespect, to conclude by risking this final
jest: when one is preparing to listen to a piece by Boulez, the silence preceding it
is already his?61
PS: There was insufficient space for me to trace the avatars of the much
discussed Symphonie concertante for piano and orchestra, composed in
1947, and lost during a stay in Cologne in September 1954, a short time
before its première in the context of the Musica Viva concerts in Munich on
22 April 1955. In summarising Boulez’s passage from project to project,
we can say that the idea of a composition for piano and orchestra was
split in two giving rise to two parallel compositions: the Third Sonata for
piano (1955–7/1963); and Doubles (1957–8), which ultimately became
Figures – Doubles – Prismes (1963/1965–8).
Translated from French by Edward Campbell in collaboration with the

author.
60
Fritz Lang (1890–1976), Metropolis, after is still his’ (‘Ô privilège du génie! Lorsqu’on
the novel of the same title by Thea von vient d’entendre un morceau de Mozart, le
Harbou (1888–1954) (Berlin, Universum- silence qui lui succède est encore de lui’):
Film AG, 1927); see Jacobsen and Sudendorf, Guitry, Réflexions. The then commonly used
Metropolis. phrase ‘morceau de Mozart’ suggests that the
61
‘O privilege of genius! When one listens to music was consumed slice by slice.
a piece by Mozart, the silence which follows it
5 Serial Organisation and Beyond: Cross-Relations
of Determinants in Le Marteau sans maître and
the Dynamic Pitch-Algorithm of ‘Constellation’
Pascal Decroupet
Introduction
What does a musicologist expect to find, when he consults archival material
such as that of Pierre Boulez made available by the Paul Sacher Foundation?
As far as I am concerned, the answer is basically a better understanding of
the music, including contextual as well as internal information – that is to
say, the way that this music has been composed. Knowing how
a composition has been constructed is essential for an aesthetic understand-
ing of it, as has been demonstrated by numerous analytical studies of
Boulez’s music over the last thirty years since these sources have become
accessible. Schoenberg’s familiar remark that ‘to know how it is made does
not yet say what it is’ has been taken out of context, becoming for too long
a time an alibi for a disinclination to come to terms with the source material.
For the music of numerous composers active after 1951, no more inap-
propriate slogan has ever hidden the path to its creation.
As music analysts, we all suffer from the ‘enigma’ syndrome: for music
which is renowned as being ‘difficult’ to analyse if not ‘unanalysable’ our aim
is, to put it crudely, ‘to crack the system’. From this point of view, music
analysis shares the basic conditions of cryptography, even if the implications
are of less potential significance. The self-reflection on the task of music
analysis can be more precisely focused if we adopt a few concepts from the
basic vocabulary used by cryptographers, since this makes clearer both what
we are looking for and the results we can hope to achieve. To explain the
deciphering of secret messages, cryptographers distinguish between the
‘algorithm’ (the encrypting device) and the ‘key’ (the specific way of employ-
ing this device). But what is to be called an algorithm in the case of Boulez’s
music, and what a key? Are these notions stable or do they change in the
course of the compositional process? Are they constant over a whole piece or
movement, or do they undergo transformations during it? What are the
consequences when, in a given utilisation of a specific technique, that which
functions as a key subsequently becomes part of the algorithm on a higher,
more complex level of the same technique?
108
109 Serial Organisation and Beyond
I. Preconditions and Active Choices

When in 1948 Boulez sketched the basis for his generalisation of serial
techniques to sound determinants other than pitch organisation, his orien-
tation was clearly a European one: considering the different strands of
musical modernity in the first half of the twentieth century (the two major
contributions being so-called ‘free atonal’, and subsequently dodecaphonic,
pitch organisation in Schoenberg’s Viennese School on the one hand, and on
the other Stravinsky’s rhythmic cells of variable length), Boulez defines the
next goal in the development of Western music as being the elaboration of
a synthesis of these former achievements. Since he adheres to the positive
implications of serial pitch organisation, in 1948, his preference is thus for
‘atonal rhythms’. His openness to the other acoustic dimensions of musical
sound was strongly accelerated when John Cage, staying in Paris for a few
months from the spring of 1949 onwards, shared with him the concerns that
guided him during his compositions over the previous decade, especially
those for percussion ensembles and prepared piano. Behind Cage, there is
the entire ultramodern tradition of the US avant-garde of the twenties and
thirties, including among others Henry Cowell with his theoretical work
New Musical Resources and his reflections concerning the interrelations
between music and dance that culminated in the formulation of the ‘elastic
form’ (a method to ensure both ‘pure musical’ coherence and adaptability to
an evolving dance).1 Furthermore, the development of electronic instru-
ments from these decades onwards contributed to a more scientifically based
approach to sound phenomena somehow ‘hanging in the air’.
A first attempt at a synthesis which took into account an important
number of these diverse currents in contemporary music was achieved by
Olivier Messiaen in the compositions that immediately followed Cage’s
presentation to his class at the Paris Conservatoire, namely Cantéyodjayâ
and Mode de valeurs et d’intensités.2 In pursuit of an even more compre-
hensive synthesis (since in Messiaen’s pieces the main orientation remained
modal and did not explore the potential of an overall serial construction),
Boulez reconsidered from a serial perspective Messiaen’s pre-compositional
organisation of the various determinants. His first decision (besides the
systematic expansion of Messiaen’s scales to twelve steps in each dimension)
consisted of disconnecting the dimensions (which is not the case with
Messiaen’s fixed ‘sound objects’) in order to submit them to independent
serial organisation. The well-known first practical result of this reflection is
1 2
Cowell, New Musical Resources; Essential Hill and Simeone, Messiaen, pp. 178 and
Cowell: Selected Writings on Music, 190–1.
pp. 218–34.
110 Pascal Decroupet
Structure Ia. The experience with the two Études de musique concrète,
realised in November 1951 and March 1952 respectively, led Boulez to
further innovations in Structures Ic and Ib, as well as in the following
work, Le Marteau sans maître.
Another decisive impact of Cage (and indirectly Cowell’s idea of an elastic
form) on the young Boulez concerns a specific relation between the different
levels of structuring in a musical composition, namely the relation between
material and form, which was defined by Cage through his concept of
square-root or micro-macrocosmic form, regulating both the inner propor-
tions of a group of bars and its multiplication on a higher level to determine
the subdivision of the formal proportions in a similar way. By adopting
a selective approach to the treatment of the range of musical parameters,
Boulez developed a more flexible technique with regard to clearly articulated
formal processes. That is to say, rather than working with a statistically equal
distribution of all the available ranges of possibility within the different
parameters, at any given point in a work, this allowed him to realise
global structures which were a consequence of qualitative differences in his
treatment of the various musical parameters. Furthermore, he soon trans-
cended the four basic acoustic dimensions of pitch, duration, dynamics
and articulation by integrating such variables as average speed sensation
or registral concentrations. Such selections ‘colour’ the different elements
of a form and give them individual characteristics, certain specificities of
the material thus prevailing at particular moments and consequently
dominating the surface for a limited time. To adopt Boulez’s terminology
since the mid-fifties, he articulates forms in terms of related ‘formants’:
thus formal construction relies on variable hierarchies rather than what
has erroneously been characterised as the ‘ahierarchical’ tendency in
post-tonal music.
The major shift Boulez achieved at the beginning of the fifties was the
development of a set of serial techniques transcending the former linear
thinking inherent in the Viennese conception of the series – consisting, at its
lowest level, in perceiving linear considerations as being only one possibility
within a more general system. Whereas in his compositions up to the Livre
pour quatuor, Boulez had regularly to intervene in the system’s determinants
or outputs to achieve results with which he was aesthetically satisfied (that is
to say, he had to take some freedom with these determinants by means of
permutations, re-orderings and other changes), from Structure Ib on he
developed his basic determinants onto a more global level in such a way
that during the process of composition, he would allow himself sufficient
flexibility not to be forced to contradict or suspend the system in order to
achieve aesthetically acceptable results, since these basic rules are in a certain
sense ‘incomplete’. From that moment on, freedom was in the system itself.
Up to the Third Sonata for piano, Boulez constructs his music basically
bottom up, that is to say the fundamental materials concern the basic sonic
elements, and the criteria for expansion into the formal domain are derived
from specificities within these materials. This does not prevent the existence
of general formal ideas or surface characteristics imagined separately, but the
entire technical elaboration would consist in joining these extremes in such
ways that reciprocity between the levels would be the most important goal to
achieve. This has as one consequence that Boulez’s ‘algorithms’ are essen-
tially complex, since he develops ways to integrate into his serial mechanics
strategies that are not only simple expansions of the most elementary
determinants through principles of self-similarity (which was the reference
method for structuring in Cage’s square-root or micro-macrocosmic form).
Furthermore, the places where such ‘external’ elements are woven into the
system change from situation to situation, being either a consequence of
an ‘out of time’ setting to coordinate different ways of organising serial
hierarchies into a higher-level synthesis, or else dependent on very local
decisions that came to Boulez’s mind in the precise moment when he was
engaged in the creative realisation of a work. Compositional decisions are
thus an integral part of the serial mechanics, but instead of reducing the so-
called ‘pre-compositional’ organisation to a few material predispositions,
which the composer would afterwards use with complete freedom without
having to refer to supplementary rules, Boulez on the contrary increases the
number of levels in his algorithms, connecting back every new level to
specific aspects of the point of departure. This is especially evident with
the derivation of pitch material for ‘Constellation’, the central movement of
the Third Sonata for piano, where a single dynamic algorithm connects the
sparse pointillistic points structures and the thickest blocs sections tending
towards total chromaticism with blocs sonores containing up to 9, 10 or even
11 different pitch-classes.
As will be clarified by the two examples that will be considered in some
detail in the following pages, ‘sound shaping’ characteristics operate at
different levels and at different stages of the project’s evolution. While in
Le Marteau Boulez considers his material at first as being in some respect
‘amorphous’, so that all its formal characteristics will emerge through
specific local treatments, in ‘Constellation’ virtually all elements of the
individual components – pitch, duration, dynamics, extending even to
registers and the organisation of harmonic resonances through different
possibilities of pedalling – are sketched prior to any elaboration. Thus
sound shaping plays a central role in composition within serial determi-
nants, and morphological identities and differentiations contribute to the
final result as form-building resources, realising specific aspects of the
range of networks.
II. Le Marteau sans maître: Cross-Relations between

the Cycles
As is well known, Le Marteau sans maître is structured according to three
interwoven and complementary cycles, and in his article ‘. . . auprès et
au loin’ (‘. . . Near and Far’), published at the very beginning of 1954 when
he was still planning for the composition to be premièred at the
Donaueschingen Festival the following autumn, Boulez explains a set of
four different compositional techniques directly linked to this score. Each
of these techniques relates to a specific algorithm (a derivation technique
to create hierarchies within the material) chosen in order to elaborate
a unique basic series: besides the tables of materials generated by a first
application of the algorithm, there exist among Boulez’s sketches docu-
ments attesting further derivation processes which are irreducible to the
former settings and definitions. By these means, Boulez generates specific
local materials resulting from more elaborated levels of the algorithm,
levels that did not need to be defined prior to the actual compositional
exploitation of these materials. For instance, even if we know about chord-
multiplication, the principle (i.e. the algorithm) in itself is not sufficient to
explain the specificities of the ‘L’Artisanat furieux’ cycle. In his analytical
studies realised during the 1970s, Lev Koblyakov achieved something that
did not seem possible at that time: he reconstructed essential parts of
Boulez’s compositional mechanics, so that he was able to explain the
production of the specific materials. Thus he was able to demonstrate
the division of the basic twelve-note series into groups of variable density
as well as their harmonic combinations. Furthermore, he analysed the
various ways in which Boulez utilised his material, including his reading of
the pitch charts in such a way as to arrive at distinctive results (for
example, horizontal or diagonal readings), which in turn influenced the
creative stage of the compositional process.
Nevertheless, from a purely analytical perspective, one advantage of Le
Marteau sans maître is the fact that, with the exception of movement VIII,
‘Commentaire III de “Bourreaux de solitude”’, the algorithm to produce the
pitch material is in itself stable. To consider again the case of the ‘L’Artisanat
furieux’ cycle: once the five tables for each of the two variants of the series
(original and inversion) have been derived, they constitute the reservoir of
pitch-class sets to be used in the composition. Later on, when considering
the basic materials for the three cycles, Boulez might treat these materials
differently, eventually adding supplementary levels (i.e. new ‘keys’), but
without producing new basic materials that would be the starting point of
further developments. That is the reason why the ‘cross-determinants’
between specific aspects of the different cycles do not result in new
algorithms but only in the unpredictable addition of new levels to the

existing set of algorithms.
To clarify this, it is necessary first to summarise the different methods
which Boulez uses to shape each cycle, so that the resulting formal inter-
weaving becomes evident on various levels. (The idea of ‘shaping’ structures
elaborated from the essential dimensions of pitch and duration by means of
dynamics, registers, timbre combinations and so forth becomes theoretically
explicit only a few years later in the article ‘Alea’, where Boulez coins
the phrase ‘enveloping phenomena’ to characterise these incidental
dimensions.)3 This exposition shows that at the initial stage, Boulez has
chosen specific polarities to arrive at clear characterisation of his material.
First, there is the instrumentation, since the percussion instruments with
indefinite pitch are only used in the cycle ‘Bourreaux de solitude’ (with the
exception of the gongs and tamtams which appear in the work’s final move-
ment precisely to signal the interweaving with inserts from ‘L’Artisanat
furieux’). Secondly, a specific type of musical time, with a perceptible
(or at least intended) pulsation in the cycle ‘Bourreaux de solitude’ (in the
Char poem, the measuring of time is a prominent feature), whereas in the
other cycles time conforms more to the notion of ‘smooth’ time. Thirdly, the
vocal style contrasts a clearly syllabic declamation in ‘Bourreaux de solitude’,
where the sections of the poem are separated by instrumental passages, with
the continuous melismatic vocal style of ‘L’Artisanat furieux’. In a second
stage, however, Boulez considered different kinds of ‘mediations’ between
such fundamental characteristics to achieve a higher level of formal
integration. This is essentially the position in the cycle ‘Bel édifice et les
pressentiments’ within the whole network of fundamental characteristics.
Thus, the universe of aesthetic negotiation between opposite types does
not evolve along a straight line between black and white (even if it is
possible to imagine a great number of grey tones on such a scale) but
according to an expandable web of triangular relations. The issue now is
whether or not there are ‘necessary links’ between all these different
characteristics, and how the composer’s choices have an impact on the
listeners’ understanding of the coherence of the whole as well as of the
structural aspect of the formal articulation.
In ‘. . . auprès et au loin’, Boulez presents at the core of the article four
different techniques with which to deduce (in Boulez on Music Today he
would have said ‘produce’) serial material within a framework conditioned
by the notions of ‘field of action’ and ‘punctual encounter’ which he defines
as follows: ‘A field admits the possibility of free will operating within limits
wide enough not to be inhibiting; the punctual encounter, on the other hand,
3
Boulez, Stocktakings, p. 34.
is the only conceivable solution at any given moment.’4 Technique 1

consists in the transposition of a twelve-note series according to its own
order of notes: this technique is the very basis of Boulez’s serial technique
and was used as such from the serial tables for Polyphonie X and Structures
I onwards. In the Le Marteau sketches, this way of deriving a serial square
was used as partial determinant only for the cycle ‘Bel édifice et les
pressentiments’. This technique is evidently ‘punctual’. Technique 2
concerns the chord multiplication as used in the cycle ‘L’Artisanat fur-
ieux’, and thus introduces the notion of ‘field of action’ since the order of
succession among the components of one bloc sonore is not part of the
rules but the very place for the composer to exert his aesthetically deter-
mined choices. Historically it is interesting to stress once more the
parallel Boulez underlines between this technique and, on the one hand,
the most ‘sound-oriented’ achievements in Webern’s music, specifically
the Second Cantata, and on the other, the recent ‘preoccupations of
electronic music,’5 thus approaching the point where an element is, so
to speak, absorbed by a higher-order entity. This is a tendency that will
lead to different conceptions of statistical determinants as they begin to
assume prominence in the writings of Stockhausen from the end of 1954
onwards. Technique 3 is again punctual, but coordinates the connection
between pitch and duration in a new way. In fact, ‘the interval . . . tied to
a duration’6 simply means that if chromatic scales in the domains of pitch
and duration are connected as are those in the basic series, the intervallic
difference will be counted in chromatic steps and expressed in parallel
form (or according to the principle of inversion) in both domains. (This
represents a significant improvement in comparison with the tables for
Structures, where the figures in the number tables simply labelled the
notes of the basic series according to their order of appearance from 1 to
12, but did not translate the row’s inner hierarchy in terms of intervals.)
Compared with Technique 1, here the intervallic hierarchy of the row is
no longer applied as a complete ‘function’ (a term which Boulez uses in its
mathematical sense, beginning with the article ‘Eventuellement . . .’), but
to create the individual movements of each isolated sound on the chart,
thus generating with each row form new sound ‘encounters’. Thus, chan-
ging the level of application of the row’s hierarchy leads to the generation
of constantly renewing pitch material instead of the simple transpositions
of the series in Technique 1. This technique is used in the cycle ‘Bourreaux
de solitude’. Technique 4 is again about fields since the strict organisation
now concerns the registral distribution of the sounds without any
4 6
Boulez, Stocktakings, p. 151. Ibid., p. 153.
5
Ibid., p. 152.
specification of their order of succession within a given time-span: the

composer is thus completely free in his action to formulate specific
textures, which, since this technique is the second determinant for the
cycle ‘Bel édifice et les pressentiments’, allows a variety of mediations
between the ‘punctual’ and the densest blocs sonores. The information
that can be deduced from the currently available sketches for Le Marteau
sans maître largely confirms Boulez’s theoretical writings, at least as far as
the elementary level of the compositional process is concerned, and his
exposition of the four techniques relates directly to those found in the
three core movements of the cycles.
In the various developments and commentaries which Boulez added to
build each of the cycles, cross-relations between the determinants enabled
the composer to increase the complexity of the network of relationships
between the ‘disparate’ aspects of the work and to prepare for their direct
encounter in the final movement, ‘Bel édifice et les pressentiments – double’.
As a consequence, the cross-relations operate to the greatest extent using
Techniques 2 and 3, that is to say, in the cycles ‘L’Artisanat furieux’ and
‘Bourreaux de solitude’. Even if Technique 3 has earlier been defined as
‘punctual’ (with regard to the interaction between the organisation in the
domains of pitch and duration), the results of the derivation process high-
light an intersection with those of Technique 2 since in a row form contained
in the tables to ‘Bourreaux’, the vertical density of the sound-events shares
a common characteristic with the blocs sonores obtained through multi-
plication. This intersection consists in the variability of the density between
the individual cells, as shown in Example 5.1a.
In the ‘Bourreaux’ tables, there are cells devoid of pitch (this will require
a particular treatment in the composition since the columns of the tables
remain connected with durational chromaticism), others with a single pitch
(a ‘punctual’ event, so to speak), and at the other extreme, cells of two to four
simultaneous sounds, each with a different duration (a perhaps not intended
parallel with the acoustic notion of ‘formant’). In the ‘L’Artisanat’ tables, the
blocs vary in density from one to four components for the non-multiplied
groupings and up to ten components for the densest ones. All these sono-
rities contain harmonic characteristics with the potential to be realised in
various ways: since these sonorities are unordered pitch-class sets, it is the
registral distribution of their component pitches and/or their specific order-
ing in time that will focus the attention on certain inherent qualities rather
than others. This similarity in the basic materials of the two cycles explains
the parallel multiplication of row forms in the first two commentaries to
‘Bourreaux de solitude’: whilst the introduction of this new feature in the
middle of ‘Commentaire I’ is achieved in a straightforward way, enhancing
the harmonic characteristics by means of parallel intervals within given
Ex. 5.1a Series tables for the ‘Bourreaux de solitude’ cycle
time-spans, towards its end all the serial strands are melodically interwoven.
At b. 54 (top of page 12 of the published score), Boulez composes
a development based on the pair of row forms V and IX: the logic of this
choice is that they share common elements (interval classes 2 and 3) whilst
their differences are manifested through other harmonic characteristics
Table 5.1: Distribution of row forms in Le Marteau sans maître, ‘Commentaire I de Bourreaux
de solitude’
bar 54 57 59 end 65 69
V (2) mult.ic. 3 V mult.ic. 3 V (4×2 synchr.) V (2×2 asynchr.)

mult.ic. 2 mult.ic.5
IX (2) mult.ic. 3 IX mult.ic. 3 IX (4×2 synchr.) IX (2×2 asynchr.)
mult.ic.2 mult.ic.1
harmonically harmonically harmonically harmonically harmonically
homogeneous homogeneous homogeneous homogeneous homogeneous
series series series series series
heterogeneous heterogeneous homogeneous homogeneous heterogeneous
synchronicity synchronicity synchronicity synchronicity asynchronicity
(interval class 5 in series V, and interval class 1 in series IX). The overall
design, as already analysed by Koblyakov and Ulrich Mosch,7 reveals a basic
structure which is summarised in Table 5.1. This shows the distribution of
row forms with indications concerning pitch multiplication by means of
transposition, and the treatment of the homogeneity or heterogeneity in the
domain of harmony and series as well as synchronicity between the poly-
phonic layers.
Boulez thus works within a combinatorial frame linking harmonic enrich-
ment and relative serial diversity, adding in the final section a separation of
the serial layers through polyphonic asynchronicity. From b. 74 on, a new
pair of row forms, VII and XII, are introduced, and the qualitative difference
with the former section is that these row forms both contain sonorities of
three sounds. Nevertheless, this specificity of the row forms does not lead to
a significantly higher density of the actual figures and textures, and com-
plexes of three simultaneous sounds displaying the characteristic harmonic
groupings are even rather rare (the chord in the viola at b. 74 is one such
example). Bars 103–4 constitute a kind of transition since the new serially
defined structure begins there but with a mode of presentation that is
a continuation of the previous sections, whilst the actual interweaving of
the row forms only begins with b. 105. Row form III and its transpositions
appear in Example 5.1b; in Example 5.1c the score is transcribed in
a structural analysis which demonstrates the relationship between the
three aggregates in the row form and the three layers of grouped
7
Since Koblyakov (see n. 3) did not know identifies the transposition levels by purely
Boulez’s sketches, the naming of the row quantitative means (differences with the
forms as well as the underlying combinatorial pitch class in the original row form; the same
principle are not central to his analysis; procedure has been adopted here). Mosch,
Mosch knows Boulez’s serial tables and ‘Disziplin und Indisziplin’, pp. 39–66.
Ex. 5.1b ‘Commentaire I de Bourreaux de solitude’, bb. 105–7, transpositions of series III
based on the multiple pitch sonorities contained in the row form itself
transpositions in the score (sounds are indicated only by their onset). It is

certainly remarkable that in this highly dense texture, a chord consisting of
three pitches, which was responsible for one strain of multiplication, also
appears as a vertical coordination between the polyphonic layers (see the
crosses in Example 5.1c): this situation is thus perfectly consistent with the
observations relating to the previous sub-section.
The next level of similarities between the cycles ‘L’Artisanat furieux’
and ‘Bourreaux de solitude’ concerns the treatment of the ‘enveloping
phenomena’. Although in each of the cycles the generation of material is
distinctive in terms of technique, the principles are analogous, and the end
result is that they are complementary in terms of the composition as
a whole. As revealed in sketch studies of Le Marteau in various publica-
tions by Mosch,8 for the commentaries of ‘Bourreaux’, Boulez used dif-
ferent types of filtering applied to the parameter of durations (from the
initial twelve chromatic values to a filtered range of six, four, three, two or
one). Even a quick look at the score of movement II confirms that similar
selections also occur in the dynamics. That is to say that Boulez, whilst
maintaining the overall ‘punctual’ approach that is characteristic of this
cycle, concentrates on selected aspects of his material, a strategy that is
8
Mosch, ‘Disziplin und Indisziplin’ (see n. 9);
Musikalisches Hören serieller Musik.
Ex. 5.1c ‘Commentaire I de Bourreaux de solitude’, bb. 105–7, analytical reconstruction
parallel to the chord multiplication in ‘L’Artisanat furieux’ and to the

multiplication of transpositions of row forms in the ‘Bourreaux’ cycle.
As shown by the combined sketch transcription in the facsimile edition of
the sources to Le Marteau, in the developments of ‘L’Artisanat’ (especially
the opening movement),9 the unity of different groups of blocs sonores
manifests itself by means of other sonic characteristics, namely a common
unique intensity, a variable range of register (pitch ambit) defined only
quantitatively (the precise position inside the overall range is not part of
the pre-compositional organisation) and a reduced duration range (duration
ambit). It is this variation in durations which governs the fluctuation
between dominating short values and more sustained notes, resulting in an
aural experience rather reminiscent of a wave form, with constant fluctua-
tion in the sense of tempo.
Considered in detail, and taking into account the first version of the
opening movement for two instruments, vibraphone and guitar, which
was still included in the first ink copy dating from 1954, the evidence
is conclusive that the relation between the two instrumental strands is
based on the principle of complementarity of inversions with regard to
two determinants (length of the groupings; correlation between the two
variable ranges – register and durations), while the relation between the
pitch range and the dynamic level is constant between the two parts and
follows a ‘chromatic parallelism’ (greatest range combined with maximum
intensity). Comparison between the two lines shows that the density of the
groupings is inverted (from 5 to 1 in the vibraphone; from 1 to 5 in the
guitar). Incidentally, neither of these instruments has a range of five octaves,
so that the feeling of a particularly large pitch range could only be realised
through a strategic combination of both instruments at a given moment.
In the upper line of the system (for the vibraphone), the longest group (five
cells) with the widest range (five octaves) and the ff intensity focuses on the
shortest value (the introduction of triplets is a variation which takes the form
Ex. 5.2 Frame for the beginning of ‘Avant l’Artisanat furieux’, showing the interaction of the
various musical parameters in the original version for vibraphone (upper half of Ex. 5.2) and
guitar (lower half) (min. = minimum; oct. = octaves)
9
Boulez, Le Marteau sans maître. Facsimilé,
ed. Decroupet, p. 86.
Ex. 5.3 Transcription of the beginning of the first movement of Le Marteau sans maître, ‘Avant l’Artisanat furieux’,
original version for two instruments
of a rhythmic diminution rather than a structural device), while the smallest

register (one octave) over three cells is heard pp but with a greater range of
durations; it is noticeable that since everything runs very fast in this first
movement, to make the differences between the durational ranges audible,
Boulez privileged the longest durations within a given ambit as much as
possible, a fact which explains the immediate repetitions of values in bb. 7–8
(crotchets) and 9–10 (dotted quavers). By contrast, in the lower part (for the
guitar), the first group consists of only one cell with two points of attack
within a range from semiquaver to crotchet over a span of five octaves in ff.
These determinants could simply be realised approximately: whilst
the jump through the pitch range is the largest one in the whole first
section of the movement, the variability of the durational range could not
be realised, and Boulez chose two identical values at the centre of the
range (two quavers), thus leaving the articulation of the long durations
(dotted quaver and crotchet) in this section to the solo vibraphone. (See
Examples 5.2 and 5.3.)
III. ‘Constellation’ from the Third Sonata for Piano: One

Unique, Dynamically Evolving Algorithm for the
Complete Derivation of Pitches
As is known through various publications concerning the Third Sonata for
piano, including Peter O’Hagan’s article on the unpublished ‘Antiphonie’,10
Boulez reconsidered his basic materials at different stages of the composi-
tional process. This is partly due to the fact that this movement had been
realised in preliminary form for the performances Boulez gave himself in the
years 1957–8, was further developed by the time of his performance at the
Darmstadt summer courses in 1959, and was subject to further extension
during his years at Harvard in the first half of the sixties (this latter version
then being abandoned by Boulez).
The problem is somewhat different in the case of ‘Constellation’, the
central movement of the sonata, which reached its final shape prior to the
first performances in 1957, and was printed in 1963. Comparing the printed
score and its specific performance directions with the versions given in 1957
and 1958 by the composer himself at the piano, it becomes clear that this
movement was improved in its practical aspects, since at the time of the first
performance Boulez had not decided upon restrictions in the connection of
the different sequences. Evidently he did not follow a strict prearranged plan,
10
O’Hagan, ‘“Antiphonie”: une analyse du
processus de composition’, pp. 109–31.
as a consequence of which he forgot or repeated one or other sequence

during his performances – as is evident from the available recordings of
these early performances. The printed score adds arrows of different shapes
to the sequences so that a version would (or even could) no longer result
from partly aleatoric choices at the moment of its performance but would
need careful preparation with regard to the sequencing of the proposed text
fragments.
What is so specific to this movement is the fact that its formal shape is in
a certain way independent of any realisation. Basically, there is an alternation
of two types of texture: on the one hand, sparsely composed polyphonic
structures of interwoven lines in the points sections; on the other, the
eruptive blocs sections. Whilst the points sections remain more or less
constant in shape throughout their three appearances, the blocs sections
show a clear, directed evolution. Indeed, Blocs I is evidently denser than Blocs
II, the dynamic ambit is larger, as is its global tessitura. The form of this
movement is thus a dynamic overall process (to put it the simplest possible
way: either an intensification or a reduction around a fixed centre) with
variants during the performances limited solely to the inner structuring
within each section.
The question now is twofold: whether there exists a strong relation
between this outer formal shape and the sonic material itself (especially
the pitch material since the difference in density is principally expressed
on this level), and if the material for such a dynamic process can be
produced by means of a single algorithm. All previous published attempts
at analysis of this problem work well for Blocs II,11 and what is evident
from these analyses is the fact that the algorithm is far from being stable.
However, in the case of Blocs I, similar analytical procedures have so far
failed to reconstruct a combinatorial system, leading my predecessors to
the assumption that the only possible solution seemed to be the likelihood
that at this point Boulez had integrated supplementary choices into his
system. The decision to proceed by a process of selection rather than by
automatism seemed to be perfectly consistent with the open-form layout
of the piece, as well as resonating with Boulez’s stringent rejection of any
kind of system in his writings since the mid-fifties. It is worth reminding
ourselves that this is the point in the score where Boulez chose to leave
behind the aggregates of Blocs II with their average density of 4.5 in favour
of more complex chord combinations with an average density of eight –
thus including blocs sonores with nine, ten or even eleven pitches.
11
Peter O’Hagan gave me access to the cor- part of the dissertation by Rosângela Pereira
responding chapter of his dissertation ‘Pierre de Tugny has been published as the article
Boulez: “Sonate, que me veux-tu?”’, ‘Au commencement était l’esquisse’,
University of Surrey 1998; the corresponding pp. 41–61.
For many years, in common with my colleagues, I imagined that at this

point Boulez simply wanted to arrive at such dense structures in Blocs I,
and that the aesthetic goal was much more important than the means of
achieving it. However, my assumption proved to be incorrect, and the
newly discovered solution shows how the previous misunderstanding of
the serial processes arose. First, certain hints contained in the sketches,
since they were unique occurrences in Boulez’s sketches of the fifties, were
simply not interpreted with an awareness of their true significance in the
development of an analytical model: indeed, for those who have studied
sketches of integral serial compositions of the European avant-garde of
the fifties, it is rather typical not to be aware of the precise meaning of
various signs such as crosses and strokes. Secondly, on the basis of the
situation regarding the analytical literature on Boulez’s compositional
procedures in the 1990s (and this has not been changed by more
recent mathematically based theoretical contributions), the horizon of
knowledge limited the understanding of compositional mechanisms
which would need such a quantity of supplementary parameters and
such intricate derivation processes.
The present analytical reconstruction of the pitch production consists of
one single algorithm. This explanation for the pitch derivation is complete
with regard to the material used in ‘Constellation’, including all the blocs
sections. The verification of the end product of this process results in only
three inconsistencies with Boulez’s tables, that is to say that if one considers
the hundreds if not more than one thousand sounds contained in the blocs
sketches, these deviations (of which two are very easily accounted for) are
totally insignificant.
Reconstructing an algorithm is a question of sequence order: to arrive at
plausible results, the different parts of the mechanism need to be put
together in the right order. But how to know which order is the right one?
How to know which element of the mechanism is, at a certain level, part of
the algorithm or the specific key to be applied to it? In the case of
‘Constellation’, the chronology of the derivations attested in the sketches is
of crucial help, especially since in the blocs sections the algorithm evidently is
dynamic.
The first rules are identical to those employed in the Points sections:
1. In the series tables, the row forms obtained through transpositions are
arranged into groups of 1–2–3 row forms, each grouping being present
twice. The order for the prime form of the series is 1–3–3–2–1–2, whilst
for the inverted form the number sequence is reversed, as in Example 5.4.
These groups of series are identified in the sketch by Greek letters, α–ζ for
the prime forms, and η–μ for the inversions.
2. The series themselves are segmented according to the same principle.
Ex. 5.4 ‘Constellation’, transcription of the pitch table with identification

of the transpositions, their grouping and their segmentation into cells
3. The cells obtained through segmentation are spatially arranged into

charts, as shown in Examples 5.5–5.8.
The next rules are valid for all blocs sections, beginning with the specific
developments for Blocs II.
Ex. 5.5 Series form A within Blocs II: the structural cells are multiplied by the retrograde
cells of the form itself. The origin of the multiplying factors is thus internal (T = tenu =
sustained; S = sec = extremely short; H = harmonique = over an artificial resonance created
by silently depressed keys; p = with pedal)
structural cell Aa Ab Ac Ad Ae Af
internal multiplication factor Af #2 Ae #1 Ad #2 Ac #3 Ab #3 Aa #1
type of multiplication unique / unique unique unique /
4. The structural cells are ‘thickened’ through the principle of chord multi-
plication with the multiplication factors being derived from the retrograde
form of the series. The cells used as multiplication factors are ordered sets
with regard to their inner order of succession (i.e. the strict order in
accordance with the retrograde) (Example 5.5).
5. When different series are grouped together, the rule might allow for
increased options since the multiplication factors could be taken from the
other row forms in the grouping, from the structural row form itself or from
all of them (Example 5.6). Thus the origin of the multiplying factors is either:
– internal
– external
– a combination of internal and external
6. Since the grouping can include three different row forms, and since
Boulez might combine internal and external origin of the multiplying
factors, the number of multiplication factors is variable and either:
– simple multiplication
– compound multiplication
7. When it came to the further elaboration of the already enriched series J,
Boulez first sketched a derivation similar to row form A. But this solution
appears in the sketches to have been crossed out and replaced by another
solution that is externally characterised by denser results. Did Boulez at
that time already have an overall formal concept of progression in the
blocs that made him decide to expand the set of rules? We may never
Ex. 5.6 Row forms L–K within Blocs II: the multiplication factors are
external, thus taken respectively from the retrograde of the other row form
in the grouping
structural cell within L La Lb Lc Ld Le Lf

external multiplication factor from K Kf #2 Ke #1 Kd #1 Kc #3 Kb #2 Ka #3
type of multiplication unique / / unique unique unique
structural cell within K Ka Kb Kc Kd Ke Kf
external multiplication factor from L Lf #3 Le #2 Ld #1 Lc #1 Lb #3 La #2
type of multiplication unique unique / / unique unique
know, but what is certain is that this ‘changing the rules’ is not a process
of shifting from one strategy to another, but an increased complexity of
results through multiple applications of the same processes: a kind of
‘resonating feedback’. This strategy, which has achieved a certain celeb-
rity in its fractal variety known as ‘self-similarity’ (in which a pattern is
subject to constant unvarying repetition at different structural levels), is
one of Boulez’s basic assumptions since his generalisation of serial
Ex. 5.7 Row form J within Blocs II: multiple multiplication by cells from the row form
itself in retrograde
structural cell Ja Jb Jc Jd Je Jf
internal multiplication factor Jf #2 Je #1 Jd #2 Jc #3 Jb #3 Ja #1
type of multiplication multiple / multiple multiple multiple /
principles at the very beginning of the 1950s. This leads to a modification of

rule 4: the multiplication factors are no longer applied according to the
specific melodic hierarchy of the original retrograde form of the series, but
are treated as ‘unordered sets’ – in other words, Boulez allows himself the
option of altering the original series order. As a result, a notion of mobility
is introduced into the derivation process. Furthermore, this rule is flexible
in its application to one and the same series, certain cells being multiplied
in fixed retrograde order, others according to the variable set principle
(Example 5.7). This rule thus allows two alternative possibilities:
– single multiplication
– multiple multiplication
The last rules are specific for Blocs I. How to obtain blocks of densities of
up to eleven sounds without leaving the (already enlarged) algorithm? What
is needed to go beyond the results obtained so far is the diversification of the
multiplication factors. At that stage small crosses surrounding the number
rows appear in Boulez’s sketches: strangely enough, they do not relate
specifically to this supplementary stage of the derivation process but reveal
themselves to be precisely the tool to select the cells which, in relation to
internal and external multiplication factors, will be treated through multiple
multiplication (in Example 5.9, the numbers in the column for parameter 7
indicate the selection process).
8. To each ‘principal’ multiplication factor is added a ‘complementary’
factor chosen from the same row form on the basis of identity of density.
Ex. 5.8a Row form K within Blocs I: use of external and internal
multiplication factors (either single or multiple), augmented partially by
additional complements; for cells Kc and Ke, all pitch classes produced by
complements are redundant; it is impossible to decide whether this is part
of the mechanics or not; since for Kf, both complements, Le and Ke,
produce pitch classes that are not contained in Boulez’s sketch, they have
to be considered as suppressed, and thus rule 9 must be considered as
‘partial’ in the global algorithm
structural cell Ka Kb Kc Kd Ke Kf
external multiplication factor from L Lf #2 Le #3 Ld #1 Lc #1 Lb #2 La #3
type of multiplication single multiple / / multiple single
internal multiplication factor from K Kf #3 Ke #2 Kd #3 Kc #1 Kb #1 Ka #2
type of multiplication single multiple multiple / / single
external complement from L Lb #2 La #3 / / Lf #2 [Le #3]
type of multiplication single multiple / / ??? suppr.
internal complement from K Kd #3 Ka #2 Kf #3 / / [K2 #2]
type of multiplication single multiple? ??? / / suppr.
Indeed, there are always two sets of the same density within a row form:
one is chosen through the application of rule 4 (the retrograde), and
the second of the same density is added as ‘complement’. This diversifi-
cation of the multiplication factors is variable with regard to one and the
same row form, certain cells being complemented, others not. This rule
allows three different solutions:
– absent complements
– partial complements
– total complements
9. Since the last rule (8) was introduced in the context of groupings of three
row forms, the complementary cells are at first chosen from among the
external row forms; at a second level, internal complements belonging to
the structural row form are added to the process. In this latter case, the
application of complement is always unique (rule 4, without reference to
the flexibility introduced by rule 7).
Ex. 5.8a (cont.)
Ex. 5.8b Row form D within Blocs I: the details concerning cell Db provide
an answer which was left unresolved with Kb (whether an internal com-
plement can be multiple or remains necessarily single): since the pitch class
generated by single application of the complement Df (F♮) is part of
Boulez’s sketch while the A♮ resulting from multiple application is not, the
consequence is that complements admit only single multiplications
structural cell Da Db Dc Dd De Df
external multiplication factor from C Cf #3 Ce #1 Cd #3 Cc #1 Cb #2 Ca #2
type of multiplication mult. / mult. / mult. mult.
external complement from C Cd #3 / Cf #3 / Ca #2 Cb #2
type of multiplication mult. / mult. / mult. mult.
external multiplication factor from B Bf #1 Be #3 Bd #1 Bc #2 Bb #2 Ba #3
type of multiplication / mult. / mult. mult. mult.
external complement from B / Ba #3 / Bb #2 Bc #2 Be #3
type of multiplication / mult. / mult. mult. mult.
internal multiplication factor from D Df #2 De #2 Dd #3 Dc #1 Db #3 Da #1
type of multiplication single single single / single /
internal complement from D De #2 Df #2 Db #3 / Dd #3 /
type of multiplication single single single / single /
As is evident from these last examples, there are numerous redundan-

cies generated by compound and multiple multiplications as well as by
complements: this is certainly a feature that obscures the understanding
of the procedure, but even if certain stages of the algorithm might seem
‘useless’ since they do not produce new pitch classes, they are an integral
part of it, since it is only in its complete assembly that the mechanics can
work. Furthermore, it is only as such that the overall strategy can be
understood.
The last element of the present demonstration shows the resulting algo-
rithm with its different parameters and the different positions chosen within
each parameter for each row form.
Up to this point, the demonstration has concentrated only on the
‘mechanical’ aspect of Boulez’s method in producing his material. It is
Ex. 5.9 Blocs sections of ‘Constellation’; complete algorithm for the pitch derivation
5 origin 6 number of 7 number of 8 external 9 internal compl.

cells applications compl.
Parameters i e i+e s c u m a p t
row form
BLOCS 3
A mélange i s u a
BLOCS 2
A i s u a
L e s u a
K e s u a
J i s m a
I i+e c u a
H i+e c m a
G i+e c u a
F i+e c u a
E i+e c u a
D e c m a
C e c m a
B e c m a
BLOCS 1
L i+e c m a
K i+e c 4u 4m p
J i s 2u 2m a X
I e s 2u 2m t
H e s 2u 2m p
G e c 5u 3m t
F e c 4u 4m t
E e c 5u 3m t
D e c m t X
C e c m t X
B e c m t X
clearly of interest to study the varying outcomes of this process with

regard to their specific sound results and to ask how Boulez used these
qualities in the next stage of the compositional process. This needs some
contextualisation.
In his article ‘Alea’ (1957), Boulez introduced a distinction between
‘structural’ and ‘enveloping’ aspects within the musical material. If the
inclusion of all the musical parameters within integral serialism at the
beginning of the nineteen-fifties consisted of a presumed equality among
them, Boulez now clarified the levels by considering pitch and duration
as structural, whilst timbre, tempo and dynamics are categorised as envel-
oping qualities, linked to the local clarification of the structural basis
(in fact, the compositional approach from Structure Ia to Le Marteau

already demonstrates a quite different reality as compared to the rigorous
theories of serialism developed mainly on the basis of the composer’s
writings).
Before any detailed analysis, it is necessary to establish the criteria
defining the context within which the discourse will unfold. From the
beginning, the pitch material was associated with so-called durational
chromaticism, that is to say values increasing from one to twelve, thus
analogous to the equal steps of a tempered scale.12 Consequently, in the
sketches for the pitch cells, their structural durations are already included.
According to the basic rules, we know that the grouping principle is
applied on different levels: notes into cells; graphic groupings of cells
into three columns for each row form; grouping of rows into higher
sets; redistribution of these sets to generate the six sections distinguished
in the score by their specific names: Mélange (with its mixture of points
and blocs), Points 1 to 3, Blocs I and II. For the actual elaboration of the
specific material tables, as a first step, the row forms were combined
in accordance with graphic criteria, distributed either horizontally or
vertically. In a second step, Boulez adopted specific ways of reading his
charts, not necessarily reflecting the former hierarchies contained in the
material. Such a strategy already existed in Le Marteau, especially for the
‘L’Artisanat furieux’ cycle, where the horizontal reading of the blocs
sonores in movement III reveals at the very surface of the composition
the harmonic qualities shared by the different blocks derived by the
multiplication with a common factor, while in movements I and VII the
diagonal readings through the charts create further, more distant
similarities.13 In ‘Constellation’, and especially in the blocs sections,
I suspect that Boulez introduced a supplementary stage into his working
out of the score by drawing conclusions from the qualitative output of his
pitch algorithm. Indeed, for the row forms E, F and G for Blocs II,
presented as a vertical set in the sketches, the algorithm led to three
identical charts, the only difference among them consisting in the level
of transposition. I do not think this was intended by Boulez but was
a merely aleatoric result due to the specific redundancies that are
a logical part of multiple multiplications. Nevertheless this output shows
something that is revealing of the aesthetic intention behind the algorithm
12
Since Boulez began working on his Third summer 1957) had no relevance for this
Sonata for piano as early as summer 1955, the composition.
13
critique of durational chromaticism in early Koblyakov, A World of Harmony; Mosch,
European serialism as contained in Karlheinz Musikalisches Hören serieller Musik;
Stockhausen’s article ‘. . . How Time Decroupet and Leleu, ‘Penser sensiblement la
Passes . . .’ (first published in Die Reihe 3, musique’, pp. 177–215.
with its dynamic transformations: the accumulation of combinatorial steps

will at a given point lead to the negation of qualitative differences instead
of enhancing them. What is so interesting with the algorithm in
‘Constellation’ is the fact that this step was reached with the penultimate
stage of the derivation process, so that the combinatorial principles newly
introduced for the last step had to invert that tendency to produce again
distinctive results. The other consequence Boulez drew from this inter-
mediate result was to adopt a reading through the charts that neutralised
the results of the derivation so that, in the actual composition, the three
identical tables would not occur together within one compound sequence.
That is why the reading of the charts, initially intended to be vertical
(since that corresponds to the order of the original series) conforms in fact
to a horizontal combination. The concrete realisation within the so-called
star section on page d of the score shows further combinatorial decisions.
First of all, within the three row forms, H is isolated (ascending diagonal
from left to right; the cells of H always appear in homogeneous groupings,
the central section of the star consisting of the single cell in the central
column of H) while C and F are treated either homogeneously (central
horizontal line on both sides of the middle point) or through homoge-
neous successions (descending diagonal from left to right). Furthermore,
as is evident from the score, the different sequences within this star section
are highly differentiated according to the layout of their registers, homo-
geneous sequences remaining in one registral position while successions of
parts of different row forms within the same sequence are characterised by
a register change (upper-left and lower-right sequence). For each row
form, Boulez treats the columns of his serial tables according to
a unifying timbre, every row form containing one common element
with one of the others and one specific timbre (the total range of timbres
using all the possibilities that had been sketched).
Considering ‘Constellation’ as a whole, a list of different combinatorial
possibilities can be elaborated (not appearing in the sketches), which is
presented here according to a principle of increasing complexity:
1. homogeneous presentation: a sequence is built only on cells from a single
row form;
2. homogeneous succession: within a sequence, different fields exist referring
to different row forms while within these fields the cells from one row
form are presented homogeneously; beginning and ending refer to dif-
ferent row forms;
3. homogeneous alternation: the fields within a sequence alternate according
to homogeneous groups taken from each row form;
4. homogeneous intercalation: components pertaining to one cell of one row
form punctuate the cells of one or more other row forms;
5. tiling (‘tuilage’): two different row forms share a common element which
is used to effect the transition between the cells of one row form to
another;
6. interlacing: the cells of different row forms are interwoven without any
vertical separation between the different cell components.
Concerning the timbre itself, Boulez sketched two different levels that would
have a direct impact on the aural experience. On a sketch entitled ‘Attaques
et Corps’, two specific sections relate to pitch itself, or to its behaviour during
its duration. Boulez distinguishes three categories – tenu = sustained; sec =
extremely short; harmonique = over an artificial resonance created by
silently depressed keys – to be played with or without pedal, making
a total of six possible modes of attack. Combined with specific treatments
during the actual duration, the behaviour of individual sounds is further
refined. For the three basic morphologies, Boulez sketched:
T = tenu normal; normally held sound for its full value.
S = sec sur le début/de la valeur; short attack at the beginning of the total
note value, the rest of the duration until the onset of the next being
replaced by silence (or stress at the end of the total value, as if the sound
is considered as ‘mechanically’ retrograded – an idea analogous to the
techniques employed in sound transformation in early electronic
music).
H = dans la résonance/suppression des harmoniques/suivant les formants;
transformation of the resonating sound by suppressions of specific
harmonics according to the formants, i.e. the modification of the
components of the artificial resonance. This treatment becomes even
more complex when combined with the use of the pedal, the transfor-
mations being submitted to the analytical treatment of each individual
component within the allotted time value. These timbral determinants
are transcribed onto the pitch charts, either attached to the individual
cells of some row forms (as in Example 5.5), or to complete columns of
others (Examples 5.6 and 5.7). By this means, Boulez guarantees that
within row combinations, each element will be characterised in terms
of a specific treatment of sound.
Finally, supplementary characteristics can be observed which are the
result of the actual placing (mise en place) of the music in time. Nothing
concerning this aspect has been explicitly formulated in the sketches. These
characteristics can be arranged according to a kind of network linking the
extremes of an isolated single sound and dense groups of blocks. The two
means of transformation of a single sound involve either the horizontal or
the vertical density of the sequence of sounds. The horizontal transformation
of a sound will result either in a figure, that is, a group of measured
durations, or in a quick group of short notes (notated as groups of
appoggiaturas or as very small measured subdivisions of a value).

The vertical transformation leads to blocs or aggregates which will be treated
in various ways: as a staccato block followed by a silence, as a staccato block
with resonance, or else as a block with a specific duration. The combination
of both density transformations generates groups of blocks in quick succes-
sion, scored as appoggiaturas, measured durations or combinations of both.
The attacks themselves vary according to a variable ‘resolution time’ or
‘stretching’ of the ‘transitional’ components of a sound: in its simplest
version, the attack is sharp, all components sounding strictly simultaneously;
a first spreading out can be realised through arpeggiations, in one single
direction for a whole block or with separated hands; further zooming leads
to a quick succession of partial blocks (perhaps the formants Boulez had in
mind in his sketches); and finally, the maximum stretching out of the
components results in a quick group of sounds of low vertical density,
normally single sounds.
Let us consider one last example. At the bottom of page e, the two initial
sequences of Blocs I are extended on page f into a shorter sequence, Très lent,
in the extreme top register of the instrument. These three sequences com-
bine the multiplied row forms J–K–L, grouped in the sketches as J and L+K.
In the score, the three row forms are combined in various ways which
demonstrate their hierarchical differences: the final Très lent sequence is
the only one to employ a single series, in the form of a homogeneous
presentation of series J. Here, the three central cells of Example 5.7 (marked
Tenu in the sketch) form the pitch content. In the two preceding sequences,
beginning on page e, the remaining cells of series J interact respectively with
the complete series L (upper system) or K (lower system). The combinatorial
strategy in those two sequences differs qualitatively, since the unfolding of
the cell groups from L is interrupted twice by single notes which originate in
the divided cell Ja (marked Harmonic pedal in the sketch), whilst the
complete statement of series K is framed by the two remaining cells of series
J (marked Harmonic in the sketch). For the upper sequence, even within its
uniform register, the differences in treatment between the two series are
sufficient to clarify the structural subtext through a distorted symmetry
including a strategy close to Schoenberg’s concept of developing variation.
The opening part of the section consists of three segments: La–b–c/staccato
chords (Boulez’s sec) are followed by the first single note, D-flat, from Ja with
a short attack una corda, and with pedal depressed after the attack.
The chords formed from group L resume with Ld–e/pedal sounds over an
artificial resonance (sostenuto pedal) presenting the sound itself for a certain
duration and then shifting to the resonance for the sound’s end (silences
superimposed with notes in brackets according to the score’s notation), the
three notes of cell Le entering progressively over the artificial resonance.
After this sequence the second single note, B-natural, from Ja is interjected,
with the same staccato attack as the previous single note and with pedal
depressed after the attack. The final two chords (Lf) consist of a synthesis of
the two former modes of presentation for the L cells, that is to say a staccato
chord followed by a longer chord with resonance (even if the treatment of
the resonance is this time regulated through variable pedalling). For the
lower sequence, the framing cells taken from J present distinct character-
istics: isolated sounds separated by silences at the beginning and a short
chord to finish, both groups being unified in terms of sonority through the
use of the una corda pedal. Row form K contrasts with the framing cells by
a sudden shift from lowest to highest register, the inner articulation being
guaranteed through the different treatments of the quality of resonance:
pedal for cells Ka and Kb, chord over artificial resonance for cell Kc, short
sounds (according to the sketches) for the cells Kd, Ke and Kf. In the score,
these short sounds appear nevertheless as ‘acoustically divided up’ in attack
from grace notes to principal note, the overall tendency being one of
progressive reduction to single principal notes towards the end of this cell
group. The Très lent sequence with the central cells from series J on page
f resonates with the lower sequence through its placing in the highest register
as well as the division into grace notes and principal notes, this distinction
being determined not by serial means but by giving this sequence its unique
‘envelope’.
Conclusion
Confronted with such analytical evidence as that presented here, there is
some urgency to reconsider more than a purely technical understanding of
one or the other compositional tools within Boulez’s craftsmanship: to know
‘how it was made’ opens the gate to the aesthetic and stylistic background of
his music. Musicological knowledge has definitively crossed the rubicon of
elementary recapitulations concerning punctual interactions between sonic
dimensions within integral serialism. During the last decades, within a first
stage of investigation into Boulez’s music characterised by an increasingly
systematic analysis of his creative processes, it was indispensable to under-
stand and describe in some detail the combinatorial features of his music,
especially of the 1950s. The selected examples studied in the present chapter
show the enormous gap between a historical plot that has crystallised too
soon around a limited number of anecdotes (‘neo-serialism’ as continuation
of the Viennese dodecaphonic tradition, Cold War, Cage in Darmstadt, and
so on) and the musical realities encapsulated within the masterworks of this
aesthetic. We have to admit humbly that we are still at the beginning of
a long path to a better understanding of post-war European serial music and

how this stage in Boulez’s career was essential to his musical thinking and
compositional output beyond a few pretended ‘youthful errors’. None of the
supplementary levels of construction within Boulez’s technique is
a ‘transgression’ of serialism but its very essence. The discovery of such
unexpected stratifications necessarily has far-reaching consequences for our
understanding of Boulez’s compositional evolution, but also stakes a claim
for a profound revision of the general discourse concerning post-war serial-
ism on the basis of an ever-deepening knowledge of the music itself.
Written in English in collaboration with Peter O’Hagan.

6 ‘DU FOND D’UN NAUFRAGE’:1 The Quarter-tone
Compositions of Pierre Boulez
Werner Strinz
The exploration of the pays fertile beyond the realm of tempered sounds
is – to adopt one of his figures of speech – at once present and absent in the
creative work of Pierre Boulez. It is present from his first experiences in
composition after arriving in Paris, extending to the researches in sound
with the technical resources of IRCAM. But the progression of works
which incorporate quarter-tones shows many gaps and digressions.
The first use of quarter-tones in the youthful Quatuor pour Ondes
Martenot is soon transformed in parts of a Sonate pour deux pianos, also
unpublished, whilst the extremely ambitious project Polyphonies pour
49 instruments did not progress beyond a fragmentary state of elaboration.
Only Le Visage nuptial retained his attention over the long term; he
subjected the work to a revision between 1948 and 1950, but in a third
version (1988–9) abandoned the quarter-tone writing. After these early
projects there follows a period in which microtonal composition is absent
in the compositions, but amply developed in theory. The variability of the
concept of space is addressed in the extension of the serial principle to
encompass the parameters of the acoustic space itself. Fundamental ideas
with regard to the organisation of sonorous realms are developed at length
in ‘Possibly . . .’.2 They are differentiated in Boulez on Music Today
by means of the binary opposition applied to specific metrical types,
interval of partition, and the modulatory division of space, without
going as far as citing the quarter-tones of the ‘Improvisation III sur
Mallarmé’.3 Despite the absence of a coherent genealogy of creative
realisations, and probably because the subject is susceptible of leading,
more than any other, to this ‘unknown’ so dear to the composer, the
question of ‘metric’ organisation of sonorous space is addressed with
striking insistence, when Boulez states that ‘the time has obviously come
to explore variable spaces, spaces of mobile definition capable of evolving
1 3
The quotation is from Mallarmé’s poem, Un Boulez, Boulez on Music Today, pp. 83ff; on
coup de dés. binary oppositions cf. Campbell, Boulez,
2
Boulez, Stocktakings, pp. 117ff. Music and Philosophy, p. 59.
139
140 Werner Strinz
(by mutation or progressive transformation) during the course of

a work’.4
We will be following this trajectory, sometimes evident, at other times
latent, in the quarter-tone compositions from the youthful works up to and
including Pli selon pli.
I. The First Experiences: the Quatuor pour Ondes

Martenot and Le Visage nuptial
Boulez’s interest in the use of microtones came about as a result of a number
of favourable circumstances which accompanied the evolution of his
musical development following his arrival in Paris in September 1943.
In Olivier Messiaen, he encountered a composer and teacher whose interest
in extra-European cultures and creative experiences were at the centre of
a circle of personalities likely to stimulate the imagination of the young
composer. Messiaen’s interest in the quarter-tone compositions of Ivan
Wyschnegradsky is not only testified by an enthusiastic review of the
Festival de musique de quarts de ton dedicated to his music,5 but also by
his own tentative use of quarter-tones in Deux Monodies en quarts de ton
(1938) for Ondes Martenot. It is likely that along with Wyschnegradsky,
other members of the Parisian Russian émigré circle known to Messiaen
included individuals who later had important links with Boulez: Boris de
Schloezer and Pierre Souvtchinsky.6 Guy Bernard-Delapierre, who had
made his home available for Messiaen’s private courses in analysis, served
in the role of concert manager on behalf of the circle of devotees of quarter-
tone composition: four students recommended by Messiaen, including
Pierre Boulez, participated as pianists in a concert of works by
Wyschnegradsky presented by Bernard-Delapierre.7 Finally, that fashion-
able instrument of the 1930s, the Ondes Martenot, made it possible to
experiment with micro intervals; in 1946, Boulez acquired a sufficient degree
of mastery in the instrument to be recommended by Arthur Honegger to
perform his incidental music for the inaugural production of Hamlet by the
Compagnie Madeleine Renaud-Jean-Louis Barrault.8
4 7
Boulez, ibid., p. 84. Wyschnegradsky, Libération du son, p. 264;
5
See Wyschnegradsky, Libération du son, concert programme, p. 263: Concert de
p. 262; Messiaen’s review mentions the Musique à Quarts de Ton, Cosmos, poème
‘penetrating charm of the harmonic aggregates symphonique for four pianos, Linnite,
and the absolute precision [‘netteté’] of the mimodrame en un acte, Cinq variations sur la
intervals’; quoted with reference to note UT for two pianos, Prélude et fugue for
Wyschnegradsky, ‘L’ultrachromatisme et les two pianos, Premier Fragment symphonique
espaces non-octaviants’, p. 140. for four pianos.
6 8
See Campbell, Boulez, Music and See Steinegger, Pierre Boulez et le théâtre,
Philosophy, pp. 59 and 222. p. 23.
141 The Quarter-tone Compositions of Pierre Boulez
Composed between September 1945 and March 1946, the Quatuor pour
Ondes Martenot9 represents at the same time the coming together of recent
influences and a state of passage in rapid evolution, month by month. Even if
the technical and aesthetic concerns of this quartet were quickly left behind,
the manner in which quarter-tones were introduced remained a fundamental
principle up to Pli selon pli. The compositional process is not based on
a specifically quarter-tone ‘genetic code’, but quarter-tones will appear as
subordinated to twelve-note structures. We will see later that even in the
case of the application of serial principles to quarter-tones in the first version
of the Polyphonies project, the series remain linked to the twelve-note series
both in terms of their generation and internal structure.
The subordination of quarter-tones to semitonal serial structures can be
illustrated by an example taken from the second movement of the Quatuor
pour Ondes Martenot (Example 6.1a). In the formal context of this frag-
ment, episodes of two and four voices alternate, the latter reinforcing the
contrast between held sounds and melodic figures.10 Here, the instruments
are grouped in twos, exploiting the characteristic playing modes and
sonorities of the Ondes.
The twelve-note series (Example 6.1b) is more than a simple means of
obtaining a chromatic texture:11 in this example, its internal structure is
reflected in its melodic deployment (the first six notes of the series) and its
vertical coincidences (the chromatic cluster at the centre). The quarter-tones
grouped around B-natural and G-sharp, the second and third notes of the
series, appear as an ornamental extension of the initial figure.
After Boulez’s first creative experiences in Paris, and nourished by its
stimulating intellectual environment, there followed a period of critical
reaction, fuelled by successive encounters which roused in his imagination
resonances with the world of theatre and literature. In considering the
concentration and originality of the writing shown in the quarter-tone
movements of Le Visage nuptial,12 it seems that the poetry of René Char,
with its pursuit of ‘the inexpressible’, neither by means of obscurity nor
9
Two scores are accessible at the Paul Sacher (MS 26163, data.bnf.fr), dedicated ‘À Pierre
Stiftung, Sammlung Pierre Boulez Souvtchinsky. Le vrai visage P.B.’.
(hereinafter ‘PSS’): (1) Pencil draft, Mappe A, The movements are dated as follows:
Dossier 3f, 1, dated: 1. mouv. : 1.09.1945; 2. ‘Conduite’: 26 octobre 46; ‘Gravité:
mouv. : 21.09.1945; 3. mouv. : 8.03.1946; (2) L’Emmuré’: 3 novembre 46; ‘Le Visage
Fair copy with dedication to Ginette and nuptial’: 22 novembre 46; ‘Evadné’: 27
Maurice Martenot, Mappe A, Dossier 3f, 2. novembre [46]; ‘Post-Scriptum’: 30 novembre
10
On Boulez’s use of sustained notes, see [46] (photocopy, PSS, Mappe A, Dossier 4d,
Nemecek, Untersuchungen zum frühen 1c, 1); (2) Fair copy, 32 p. (PSS, Mappe A,
Klavierschaffen von Pierre Boulez, p. 117. Dossier 4d, 1c, 2); (3) Fair copy, 36 p. (PSS,
11
See Bennett, ‘The Early Works’, p. 47. Mappe A, Dossier 4d, 1c, 3); (4) Fair copy
12
Four manuscript scores are accessible: (1) with indications for the third version, 38 p.
Fair copy, Bibliothèque nationale de France (PSS, Mappe A, Dossier 4d, 1c, 4).
Ex. 6.1a Quatuor pour Ondes Martenot, second movement, bb. 132–41 (PSS, Sammlung Pierre Boulez, Mappe A, Dossier 3f, 2)
Ex. 6.1b Quatuor pour Ondes Martenot, series I II (PSS, Sammlung Pierre
Boulez, Mappe A, Dossier 3c, 1)
Ex. 6.2a Le Visage nuptial, series P9 (PSS, Sammlung Pierre Boulez,

Mappe A, Dossier 4d, 1b)
exaltation, but by the brilliance of a language in a state of extreme internal

tension, would have especially contributed to the individuality of Boulez’s
musical thinking. In Le Visage nuptial, Boulez has recourse neither to the
possible solution of the transfer of the serial principle to quarter-tones, nor
to their ornamental function as seen in the Quatuor pour Ondes Martenot,
but he integrates quarter-tones in the generation of intervallic structures
derived from the basic series of the work. In ‘Post-Scriptum’, the derivation
takes as its point of departure the division of the basic series (Example 6.2a)
into segments of three, five and four notes respectively, as indicated in the
annotations. The first and last segments form a cell of seven sounds which
serve as a marker for the quarter-tone displacements. The choice of dis-
placed sounds, the positioning and the degree of displacement guarantee the
distributive variability of ‘specific’ intervals for the quarter-tonal region
(multiple odd numbers of the quarter-tone unit) and ‘non-specific’ (multiple
pairs). The result is a quarter-tone cell (Example 6.2b, cell a) from which
three other cells are drawn by the conventional means of transposition,
retrograde and inversion, the individual identities of which are assured by
the variability of the quarter-tone displacements and completed by the
addition of sounds which veil the connection of cells in pairs, a–b and c–d.
These four cells are deployed in a reservoir of forms transposed and inverted,
such that the coordination of the intervals of transposition and of the means
of presentation produce a symmetrical relationship around the note
A-natural; only the cells designated for the second verse relate to a centre
of symmetry which is at the extreme of three-quarters of a tone from
A-natural. The cells, of which the number of sounds is less than the number
of syllables of twelve-syllable verses, are associated with four rhythmic
figures distributed between the voice and the two Ondes Martenot, and
linked to the ordering of the melodic cells.13
13
See the analysis of rhythmic structures in a complete transcription of the manuscript
‘Post-Scriptum’ by Gerald Bennett, (4) of the first version.
‘The Early Works’, p. 66; the article contains
Ex. 6.2b Le Visage nuptial, ‘Post-Scriptum’, quarter-tone cells (analytical rewriting)
The first strophe of the poem is connected with the last two by the
order of the cells, which are folded back on themselves, mirroring the
retrograde sense of the final strophe. They thus enclose the second
strophe of Char’s poem, which names the place of isolation – ‘the
desert’ – towards which the casting away evoked in the poem’s opening
line eventually leads:
voice: O.M. 1: O.M. 2:
Écartez-vous de moi qui patiente sans bouche; a }a1 b1R
À vos pieds je suis né, mais vous m’avez perdu; b
Mes feux ont trop précisé leur royaume; c }c1R d1
Mon trésor a coulé contre votre billot. d
timp. picc.: a (1–2)
Le désert comme asile au seul tison suave a2 }a3 b3R
Jamais ne m’a nommé, jamais ne m’a rendu. b2
cymbales susp.
Écartez-vous de moi qui patiente sans bouche: aI }b5R a1I
Le trèfle de la passion est de fer dans ma main. b4
Dans la stupeur de l’air où s’ouvrent mes allées, d2R d3R c3R
Le temps émondera peu à peu mon visage, c2 }b5R a1IR
Comme un cheval sans fin dans un labour aigri. b4R
[postlude:] a (1–6) bR (1–5) + a (7)
timp. picc.: a (1–2)
Tam-tam
Fig. 6.1 Distribution of pitch cells in ‘Post-Scriptum’
In the second movement, ‘Gravité: L’Emmuré’, the semitonal and quarter-

tonal worlds are used in the composition in different ways:
(1) the relation between the musical form and the form of the poem is
coordinated by the basic series (Example 6.3), the notes of which are
distributed to the eight stanzas in the form of sustained pedal notes, the
quatrains being divided into groups of couplets by two such pedals.
The movement ends on a fermata, with the last note, E-natural, fading
away in a diminuendo jusqu’au silence.
Ex. 6.3 Le Visage nuptial, ‘Gravité: L’Emmuré’, pedal notes (analytical rewriting)
The spatial distribution of pedal notes seems to be governed by the

displacement of neighbouring chromatic notes, but also by the intention of
emphasising the global formal relationships – for example, the descent in the
five strophes of the first seven-note cell towards the low C-natural.
Ex. 6.4 Le Visage nuptial, ‘Gravité: L’Emmuré’, stanza 3, series and quarter-tone derivations (analytical rewriting)
(2) The voice lines and those of its alter ego, the Ondes Martenot, feature
quarter-tone structures developed from the basic series by single and
double displacement of individual sounds in the direction of their
neighbouring quarter-tones. Each strophe contains a two-voice canon,
as shown in Boulez’s Example 3 from the article ‘Proposals’,14 where the
opening verse of the original version is quoted as an example of irregular
rhythmic canons, with the voices linked by means of inversion and
retrograde inversion. The melodic shapes of the two voices are similarly
entwined, and Example 6.4 demonstrates the relationship of the voice
and the Ondes Martenot to the series in the third strophe. Here, the
evocation of the extreme conflict between desire and the absence of the
desired one concentrates the image of the woman developed in the first
part of the poem:
Ô toi, la monotone absente,

La fileuse de salpêtre,
Derrière des épaisseurs fixes
Une échelle sans âge déploie ton voile!
The pedal note D-natural enclosing the first two lines is at the same time the
point of departure for the deployment in inversion of the intervallic content
of the voice line and the two Ondes Martenot. This is based on two series
related by common notes, symmetrically arranged, the central series [P4]
straddling the two sections of the strophe.
The choice of quarter-tone displacements involves, as in ‘Post-Scriptum’,
the neighbouring chromatics contained in the twelve-note series; worth
noting is the fact that the neighbouring chromatic notes E-flat and
C-sharp around the central pivot D-natural are particularly developed in
this respect.
II. The Revision of Le Visage nuptial and the

Composition of Polyphonies pour 49 instruments
The realisation of these two projects dates from a period of extreme creative
concentration – the completion of the Second Sonata for piano and the
composition of Livre pour quatuor (June 1948 to July 1949) – but also
a calling into question of stylistic and technical issues, of which the visible
trace is found in other revisions (in November 1948, Le Soleil des eaux,
in April 1949 the Sonatine). If revisions by a process of deletion involve the
elimination of the most obvious Balinese/Javanese elements such as the
14
148 Werner Strinz
clusters in the extreme bass register so dear to Jolivet and Messiaen,15

other stylistic revisions are motivated by the increasing attention
accorded to the qualitative control of harmony, refining the rather
mechanical nature of the mobility of spatial/temporal registers
expounded in ‘Proposals’.16 A revealing indicator of the stylistic deci-
sions taken by the composer is given by the criteria for the placing of the
minor second. From the Livre pour quatuor onwards, and very noticeably
in both Polyphonie X17 and Structures, Premier livre, the immediate
juxtaposition of neighbouring chromatic notes is largely avoided, even
absent, clusters of three semitones usually being separated by the group-
ing of major second and major seventh.18
These criteria relate particularly to the revision of the last movement of Le
Visage nuptial. In the first version, ‘Post-Scriptum’ is linked to the second
movement not only by the presence of quarter-tones, but also by an inver-
sion of the relationship between voice and the two Ondes Martenot.
In contrast to the intertwining of the voice and its instrumental partner,
supported by pedal notes on the second Ondes Martenot – a texture which
dominates ‘Gravité’, the final movement – ‘Post-Scriptum’ is characterised
by significant oppositions of space and register. The two Ondes Martenot
find themselves at a distance from the voice, in the low register in the first
verse, in the extreme high register in the second verse, whilst the third and
final verse presents them at first reunited in the middle register, then far
apart in a disjointed relationship, low–medium–high. In the revision, these
spatial orientations are conserved by the reassignment of the original Ondes
Martenot parts to corresponding registers of the string family. But the
extreme polarisation of the spatial disposition and the linearity of the writing
found in the original version are largely abandoned. The melodic substance
of the entrées fuguées19 of the Ondes Martenot is transformed into harmonic
fields by prolongation of the durations. The rhythms of the original, char-
acterised by a disassociation between metre and internal beats resulting from
the combinations of irrational values, are conserved in order to maintain the
15 18
Robert Nemecek shows numerous This reduction in the role of the
examples of stylistic affinities between Boulez minor second as a direct interval recalls
and Jolivet in the youthful works a similar evolution in the music of Anton
(Untersuchungen zum frühen Klavierwerk von Webern from the first atonal works to the
Pierre Boulez, pp. 49–64 and 121); concerning twelve-note compositions. See Hanson,
the revision of the Sonatine for flute and ‘Webern’s Chromatic Organisation’, p.135;
piano, see Gärtner, Werkstatt-Spuren: Die on the qualitative dimension of harmony, see
Sonatine von Pierre Boulez, p. 195. Piencikowski, ‘Nature morte avec Guitare’,
16
See Boulez, Stocktakings, p. 50. pp. 66–81.
17 19
Polyphonie X pour 18 instruments (1950–1) Sketch of ideas for ‘Post-Scriptum’ (PSS,
is not a revision but a reworking of the Mappe A, Dossier 4d, 1a).
structural basis of the the project of
Polyphonies.
timing of successive entries in the harmonic texture. A comparison of

the original and revised versions indicates indubitably the distancing
of neighbouring semitones and quarter-tones as a prime cause of the
considerable extension of the range of harmonic fields. But in the revised
version of the ‘postlude’ the general stylistic decision to avoid chromatic
neighbourhoods is transgressed. In the two original Ondes Martenot parts
Boulez annotates the registral displacements with single or double
octave shifts, the directions of which are indicated by the placing of the
signs above or below the corresponding notes (Example 6.5a). In the
revised score, this extended range forms a harmonic field of sustained
and trilled notes, which is maintained until the end of the movement
(Example 6.5b-1). The intensity of the acoustic dynamism imparted by the
trills is abruptly increased by the entrance of a second twelve-note chord
(Example 6.5b-2), superimposed onto the existing harmonic field.
This second chord, consisting of the remaining twelve complementary
quarter-tones, placed in direct juxtaposition to the first chord, completes
the harmonic structure. These quarter-tonal links are sketched in the
manuscript in the form of a scale, summarising the complementary
relationship between the two chords (Example 6.5c).
The extraordinary ending of ‘Post-Scriptum’ testifies not only to Boulez’s
embracing of direct quarter-tone relations, but also to an opening up of
compositional possibilities emphasising the plasticity of the sound and occu-
pying a complementary relation to the generation of structural materials.
Certainly, the end of ‘Post-Scriptum’ is far from the only example of
such comprehensive exploration of the expressive potential of quarter-
tones in Le Visage nuptial. We return to the example of the third verse of
the second movement, ‘Gravité: L’Emmuré’, and to the observation of
neighbouring quarter-tones around the axis note D5. In both versions, the
entry of the voice on the emphatic exclamation ‘O toi, la monotone
absente’ is accompanied by a quarter-tone cluster consisting of the notes
C-three-quarter-sharp, D-natural, and E-three-quarter-flat. In the revised
version the close position is retained, and indeed the addition of trills
thickens the texture, thus emphasising the effect of the cluster.20
The experiences of the complementary relations of quarter-tonal and
semitonal spaces in the Quatuor pour Ondes Martenot and in Le Visage nuptial
culminate in a synthesis of structural and harmonic aspects of composition
in the large-scale project of Polyphonies. The fundamental principles of
this work with regard to its structural organisation are fully described in
20
Pierre Boulez, Le Visage nuptial, ‘Gravité:
L’Emmuré’, Heugel, Paris, 1959 H.31.702,
b. 18.
Ex. 6.5a Le Visage nuptial, ‘Post-Scriptum’ (PSS, Sammlung Pierre Boulez, Mappe A, Dossier 4d, 1c, 1)
Ex. 6.5b Le Visage nuptial, ‘Post-Scriptum’, vertical quarter-tone disposi-

tions (PSS, Sammlung Pierre Boulez, Mappe A, Dossier 4d, 1c, 1)
Ex. 6.5c Le Visage nuptial, ‘Post-Scriptum’, quarter-tone relations (PSS,

Sammlung Pierre Boulez, Mappe A, Dossier 4d, 1c, 1)
a letter to John Cage, and partly published in ‘Possibly . . .’.21 Conceived as

a collection of single movements the number and order of which could be
chosen for execution, the instrumental disposition and the structural orga-
nisation of the concept are based on the number seven; Bernard Saby
commented on the seven instrumental families each with seven instruments
respectively in terms of a ‘musique de grande chambre’.22 The structural
foundation of the pitch material is not exclusively based on quarter-tones,
but on the interrelationship of quarter-tonal and semitonal dimensions
(Example 6.6):
(1) by means of the division of the basic quarter-tone series (A) into two
series of twelve notes consisting of the semitonal (a) and the transposed
and retrograded quarter-tonal (α) parts of the series;
(2) by the projection of the series a and α into the quarter-tonal realm,
either by diminution (a semitone = a quarter-tone, series Bb and Cc), or
by augmentation (a semitone = five quarter-tones, series Dd and Ee);
(3) by means of the reconfiguration of six quarter-tones of the series in the
semitonal mode, precisely in the segment of six semitones not occupied
by these series, by means of the augmentation of their intervallic pro-
portions (series β, γ, δ, ε);
(4) by the inversion of (1): two quarter-tone series (F and Φ) are interwoven
into a series (fφ) containing the notes of the semitonal series, and its
21 22
Boulez and Cage, Correspondance et Nattiez, ibid., p. 80. Manuscript score,
Documents, pp. 154–60; Nattiez, Boulez–Cage uncompleted (PSS, Sammlung Pierre Boulez,
Correspondence, pp. 80–90; see also Mappe C, Dossier 1h).
Stocktakings, pp. 121–8.
Ex. 6.6 Polyphonies, series (PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1b)
quarter-tone transposition, both related according to the order of quar-

ter-tones and semitones in A.
The fusion and separation of quarter-tonal and semitonal elements con-
tinues in the further elaboration of the serial material. In the sketches we find
tables with exhaustive combinations of varying ordering and numbering for
the fourteen series, and the following disposition is sketched for the first two
movements:
‘1°/ Pour l’ensemble complet (Ouverture-fugue) séries A, a et α
2°/ Pour 4 Violons, 4 Altos, 4 Violoncelles, 2 C. Basses séries A’23
Of this ambitious project, only the structural organisation and the formal
layout of the first movement are completed, the three-part formal design of
which comprises an exposition, a central development section and
a conclusion. The only allusion to the originally proposed Ouverture-
fugue model, with its contrast between homophonic and linear writing, is
found in the vertical and horizontal textural dispositions of this first
Polyphonie. The alternation of these textures would result in the super-
imposition as well as the separate unfolding of quarter-tonal and semi-
tonal series, thus creating fluctuating harmonic fields.
Vertical-horizontal: the formal conception of the first Polyphonie pro-
jected the horizontal deployment in three sections of development, of the
serial material presented vertically in three sections of the ‘exposition’;24 we
will illustrate this relationship by means of the first segment of this section.
Six quarter-tonal and semitonal serial forms are assigned to the six families
of instruments at fixed pitches (Figure 6.2).
woodwind I ∫aIv
woodwind II ∫αIv
brass ∫α8
percussion I ∫a12
strings I ∫A16
strings II ∫AIXIV
percussion II –
Fig. 6.2 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e
They are linked by means of ‘pivot notes’,25 i.e. a common note in two
series at one of four pivot-note positions, the first note, two notes in the
centre, and the last note of a series. The borrowing from mathematics of the
symbol of the integral (∫) signifies that the individual series are to be treated
by the same principle of pivot notes described above. See Figure 6.3 – ∫A16
(strings I).
23
PSS, Mappe C, Dossier 1h. regarding the form of the Polyphonies, p. 135;
24
See Kovács, Wege zum musikalischen reproduction of the serial tables, p. 116.
25
Strukturalismus; reproduction of the sketch Boulez, Stocktakings, p. 116.
154 Werner Strinz
∫A16:
positions: 1 12 // 13 24
notes-pivots: A B F C-half-flat
positions: 12 1 // 24 13
series: A22 A7 A23 A12
positions: 13 13 // 1 1
positions: 24 24 // 12 12
Fig. 6.3 Deployment of the family of series belonging to ∫A16
The families of series produced by the six integral (∫) series are linked in
the respective instrumental layers. As a consequence of their superposition,
the first formal section of the exposition consists of a harmonic field char-
acterised by the presence of the totality of quarter-tones carried by the strings.
Horizontal-vertical: these same six series are disposed lineally in order to
form the serial ‘superstructure’26 of the first part of the development, in which
each serial form will be responsible for the six layers of a structural segment.
If the series are indicated in terms of integrated series, their contents will be
distributed individually to each of the six layers; in the opposite instance,
a single serial form will be assigned collectively to these layers (Figure 6.4).
∫A16 – ∫AIXIV – ∫ aIv – aRIv – ARIXIV – AR16 – aR12 – αR8 – αRIIV – ∫αIv – ∫α8 – ∫a12
Fig. 6.4 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e; partial transcription
In the first structural segment of the first section of the development

corresponding to the series ∫A16, we thus find once more the series belonging
to this integral family, assigned in the exposition to the strings I, and now
distributed individually over the six layers (Figure 6.5).
woodwind I ←
A7 minus 1/4 de ton = silences
woodwind II ←
A22 id.
brass ←
A5 id.
percussion I ←
A20 id.
strings I A23 in augmentation
strings II A4
percussion II –
Fig. 6.5 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1f;
partial transcription
Since Boulez excludes quarter-tonal for the winds, the designated series of
family A will be reduced to their semitonal content, with quarter-tones
26
replaced by silences in the places where the relationship of serial material to

rhythmic cells (Example 6.7, rhythmic cells indicated by hooks) makes this
substitution possible.
The process of coordinating and elaborating the seven structural threads
is governed by a ‘principal group’ which serves as an axis, expanded in time.
Around this are placed a series of interjections by the other instrumental
groups, forming a constellation in delimited fields of time. The role of the
‘principal group’ at the beginning of this first development section is
assigned to the strings I, as a result of its association with the integral series
∫A16, the series responsible for this segment; subsequent groups are placed in
the order indicated in Figure 6.6:
main group: strings I:

strings II – woodwind II – percussion I – woodwind I – brass – percussion II //
brass – woodwind II – percussion II – strings II – woodwind I – percussion I
Fig. 6.6 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e; partial transcription
The quantitative properties of the composition’s temporal dimensions –

individual durations and distances of entries of the instrumental groups – do
not form part of the structural predetermination. However, the element of
duration is taken into account in the compositional process in the form of
seven basic rhythmic cells – aptly named ‘rhythmic possibilities’27 by the
composer – which are not considered as individual forms or motifs, but as
elementary types of rhythmic proportions and of value grouping, whose
literal realisation depends on the application of seven procedures of varia-
tion. Thus, the structural organisation concerns the transformation of basic
cell types, and the temporal placing of the groups is subject to the shaping of
the various musical components.
The realisation of this first development section reveals the impact of
specifically quarter-tone harmonies (Example 6.7). The segment begins
with a quarter-tone harmonic field essentially deployed by the second
group of strings. Immediate quarter-tone relations in the same tessitura
are largely avoided by systematic octave shifting (Example 6.8–1, register
disposition of the first harmonic field), except for the fleeting clash of the
sustained D-half-sharp (strings I, violin 2) and the E-flat pizzicato (strings
II, viola 2). As a consequence of the augmented rhythmic values and
the reduced vertical density of the principal group (strings II), the number
of quarter-tones weaved into the consecutive interjections of the
other groups diminishes to the farthest point of disequilibrium –
a single quarter-tone in the semitonal field of the brass group. This
27
Nattiez, Boulez–Cage Correspondence, p. 101.
Ex. 6.7 PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1f; partial transcription
Fl. picc
Fl. en
sol
Htb.
Saxo sopr.
Cl. sib
Cl. basse
Fag.
Fl.
Htb.
C.A.
Cl. picc.
Saxo alto
Fag.
C.Fag.
Tr. ré
Tr. sib
2 Cors
3° Cor
Trb.
Tuba
1° Piano
2° Piano
Ex. 6.7 (cont.)
158 Werner Strinz
gradual diminution in the number of quarter-tones is accompanied by an

increasing prominence given to the quarter-tone interval itself: direct
quarter-tonal neighbourhoods (Example 6.8–2, register dispositions of
the second harmonic field) are exposed in low and high registers, and
are accompanied by complementary quarter-tone relations spaced at the
octave (B-half-flat 2 – Double Bass 1; B-natural 2 – Harp; B-flat 1 –
Contrabassoon/Piano 2 // F-three-quarters-sharp 6 – Violin1; G-natural
6 – Harp/Alto Flute; F-sharp 5 – Harp).
This process reaches a climax in a prolonged quarter-tonal clash in the
principal group (strings I), coinciding with the entry of the second percus-
sion group of indeterminate pitch. Unfortunately, the score only contains
indications of modes of attack, which does not allow one to estimate if this
encounter is developed by the beats that it can produce (assuming that both
notes are allowed to resonate with the same intensity).28 But even the
coordination of the entries in itself provides an important measure of the
coming together of acoustic and harmonic dimensions as a consequence of
the admission of quarter-tones.
When Heinrich Strobel, director of the music department of the
Südwestfunk, commissioned from him an orchestral composition for the
Donaueschinger Musiktage für zeitgenössische Tonkunst, Boulez returned
to the abandoned Polyphonie project, reducing the instrumental forces and
‘translating’ the quarter-tonal serial material into twelve-note series. This
allowed him to re-utilise the serial superstructures of the first, unrealised part
of the Polyphonies for the three movements of Polyphonie X pour 18
instruments.29 However, the formal realisation of this second attempt is
independent from the initial exposition/development conception.
The abandonment of the quarter-tonal and semitonal space variability of
Polyphonies in its original version marks a temporary end to the specific
compositional explorations of microtonal possibilities. Nonetheless,
Boulez’s intention to include acoustic space itself in the category of musical
material evolves in the following decade, as shown in a reciprocal adjustment
of the concepts of temperament and harmony, far outstripping the quarter-
tone experiences of the past, and equally the tools of realisation of the period.
In ‘Possibly . . .’ the considerations are still dominated by the intention of
transferring the serial principle onto the creation of spatial divisions.
The series, free from the constraints of semitonal temperament, creates its
own habitable space by means of transpositions on itself at the interior of
a band of frequencies other than the octave, acting as a modulo in the case
28
The documents relating to Polyphonies in which would allow for the reconstruction of
the Paul Sacher Stiftung, Sammlung Pierre the details of intensities.
29
Boulez do not contain more of the sketches Première at Donaueschingen,
6 October 1951.
Ex. 6.8 Polyphonies, register dispositions (analytical rewriting)
160 Werner Strinz
where these transpositions overstep its limits. The limiting upper and lower
frequencies of such a band create an ‘interval of definition’ for the division of
musical space into registral zones, whose transposition factor corresponds to
the ratio of these frequencies. The confrontation of this idea with the
modulatory acoustic and perceptive function of the octave appears in an
exchange of letters between Boulez and Henri Pousseur, the latter seeing
Boulez’s concept as called into question by the strength of the identity of the
octave:
The octave seems to me one of the rare but ineluctable fundamentals [données]
that we have still to take account of. You envisage it like a residue of tonal
language, but it is necessary to make of it a means of articulation typically serial.
Besides, if one wishes to apply your principle, one is going to give place
constantly to quite uncontrollable short circuits of octaves. Sincerely, pardon my
presumption, but I ask myself if your reflections on this subject have not been too
abstract.30
His objection is related to the experience of his microtonal composition for

three pianos, Prospection, where the six tones of the octave are tuned in
sixths of a tone over a range spanning six octaves.31 Boulez’s response sheds
light on the difference between a ‘modulated’, non-tempered space, not
based on the octave, and a microtonal tuning within the field of an octave:
If you are troubled by the employment of an interval other than an octave as

the interval of definition (and not of transposition) of a series (have I made
a mistake? I do not have my article in front of me), it is because you conceive
the space of definition as divided into equal parts such as 1/4 or 1/6 tone. It is
thus evident, in enlarging the registers, you will always have octaves . . . you
forget that one can divide this interval of definition in non-multiple intervals
of the same interval, and consequently your chance of finding octaves in
the changes of register – or doubled frequencies – would be infinitely
small, smaller than the chances of having octaves in a realisation with an
octave-based series.32
In ‘At the Edge of Fertile Land’, these two possibilities – the microtonal
isometric division of space and its irregular partition, produced by the serial
organisation inside a modular interval of definition different from the octave –
are neatly differentiated with the aid of metaphors borrowed from the
geometry of ‘plane’ and ‘curved’ surfaces.33 The precise distinction between
the notions of ‘interval of definition’, ‘interval of division’, of the identity of
the octave and of temperament, also sheds light on the differences between
30 32
Letter from Henri Pousseur to Pierre Letter from Pierre Boulez to Henri
Boulez, 29 December 1952 (PSS, Sammlung Pousseur, early January 1953 (PSS,
Henri Pousseur). Sammlung Henri Pousseur).
31 33
See Decroupet, ‘. . . wie die Redaktion zur Boulez, Stocktakings, pp. 162–3.
Systemprämisse wurde . . .’, pp. 8–10.
non-octave space divisions as conceived by Boulez and Wyschnegradsky.

In considering the latter’s ‘structures of fundamental level and of diverse
densities’, it is clear that the common base of an equal temperament at
a twelfth of a tone – the ‘interval of division’ – causes numerous ‘short
circuits’ of octaves, if the presented scales are transposed by a factor corre-
sponding to their respective ‘intervals of definition’, varying within the limits
of a major seventh and a minor ninth.34
In Boulez on Music Today, the dimensions of musical space and the
quality of musical time are associated by the opposed criteria of ‘striated’
and ‘smooth’ surfaces.35 Being based on either regular or irregular modular
divisions, ‘plane’ and ‘curved’ spaces are subsumed under the category of
‘striated’ space in contrast to the fundamentally non-metric disposition of
‘smooth’ space.
These considerations concerning musical space mark an enlargement of
the conventional concept of harmony – in which equal temperament
serves as a neutral background canvas – to an integral dimension of
sound organisation with converging concepts of harmony and tempera-
ment. This integrating perspective has its origins in the relation between
the serial generation of materials and their potential being realised in the
interplay of complex sounds and complexes of sounds. The procedure of
chord multiplication represents the mediating element between serial
organisation and the acoustic dimension, assuring on the one hand the
structural coherence of the processes, and on the other allowing for the
fixing of the density of a given complex, as shown in the varied use of the
technique in ‘Constellation’, the middle formant of the Third Sonata.36
The increasing attention shown to the malleability of space, not by the
organisation of its metric division, but by the acoustic dynamism of the
sounds which occupy it is testified equally by more extreme examples of
contraction and spatial expansion, such as (i) clusters, conceived as
bands of saturated frequencies – structurally delimited as in ‘Glose’;
(ii) fields of resonance, as in ‘Constellation’; and (iii) the remarkable
reappearance of the minor second as a direct interval, avoided in the
earlier evolution of Boulez’s harmonic language on account of its resi-
dual appoggiatura effect, and now admitted as sound complex or
complex sound.
34 35
Wyschnegradsky, L’ultrachromatisme et les Boulez, Boulez on Music Today, pp. 85
espaces non octaviants, pp. 104–5; for and 93.
36
Boulez’s critical stance regarding the concept See O’Hagan, ‘Pierre Boulez: “Sonate, que
of ‘continuum’, see ‘At the Edge of Fertile me veux-tu?”’, pp. 171–3.
Land’ (Stocktakings, p. 163).
162 Werner Strinz
III. Pli selon pli, ‘Improvisation III sur Mallarmé’

The return of quarter-tones in ‘Improvisation III’ is part of this adjustment
of the focus of composition towards the internal dynamism of the sound
material – its states of aggregation, to borrow a term from physics (and
with a sideways glance at Edgard Varèse). The function of quarter-tones is
particularly noticeable when they become the ‘interval of definition’ for
‘imperceptible’ glissandi which, being far from conventional portamenti,
raise the speed of beats to the status of compositional elements. These
glissandi appear in one of the threads in the complex textures of
‘Improvisation III’, generated by the ‘reduction of a polyphony’ into
a single line, a principle described in detail in Boulez on Music Today,
Examples 54–6, which is taken from the opening of ‘Improvisation III’.37
The basic material of these threads is derived from a fragment (Example 6.9)
of the theatre music for Aeschylus’s trilogy L’Orestie, composed by Boulez
for the production by the Renaud-Barrault Theatre Company in 1955,
and from which four cells provide the basis for the derivation of rhythmic
and harmonic material in the passages in ‘Improvisation III’ designated
‘sectionnements multiples, polyvalents’.38
With their long, sustained notes, these threads define three large-scale
structural sections of the work (α – β – γ). In the first version of the
composition, Boulez realised the potential for formal mobility within each
section. These threads produced by the ‘reduction of a polyphony’ of four
homophonies are not intended to be executed complete, but the interpreters
can choose one of four alternative sub-sections, the beginnings and endings
of which correspond to the beginnings and endings of the four original
homophonies. In the manuscript, these sub-sections are distinguished from
one another by indications in four colours, red, green, blue and black; this
notation is applied equally to the ulterior layers which are added to the
principal one, and of which the elements must be played or omitted in line
with their respective colours and in accordance with the choice of sub-
section played. Example 56 from Boulez on Music Today shows that the
four original homophonies can maintain a degree of differentiation in their
reduced state due to the relatively close or spaced layout of their blocs
37
Boulez, Boulez on Music Today, pp. 135–7. ‘sectionnements’ is listed by Raphaël Brunner
38
PSS, Sammlung Pierre Boulez, Mappe G, in ‘L’“Improvisation III sur Mallarmé” de
Dossier 3e, 4a. The global concept and the Pierre Boulez: éléments pour une mise en
relationship between the mobile form of the perspective’, Dissonanz 50, November 1996,
first version, and fixed form of the second pp. 4–14 . See also Brunner’s ‘Entre style
version of ‘Improvisation III’ are described musical et signification musicale: la
by Erling E. Guldbrandsen in Tradisjon og stylisation. A partir des quatrièmes pièces de
tradisjonsbrudd: En studie i Pierre Boulez’s Pli selon pli de Pierre Boulez et de Sept Haïkaï
‘Pli selon pli – portrait de Mallarmé’, p. 203; d’Olivier Messiaen’, in Applied semiotics 1/1
the table of blocs sonores utilised in the (1996), pp. 89–117.
Ex. 6.9 Pierre Boulez, L’Orestie, ‘Entrée Agamemnon’ (PSS, Sammlung Pierre Boulez, Mappe G, Dossier 1c)
164 Werner Strinz
sonores, the defining intervals of which are based on seconds and fourths.
The four sub-sections are also associated with specific instrumental and
vocal contours, presented respectively in four ‘Indicatifs’ preceding the
entire section (in section α: ‘Indicatif I’: harps; ‘Indicatif II’: voice;
‘Indicatif III’: guitar and mandolin; ‘Indicatif IV’: xylophones).
It is precisely inside of the minor and major seconds present in the sustained
notes that the ‘imperceptible’ glissandi based on quarter-tones make their
appearance. In sections α and β, they generally appear in the form of
a gradual shift away from the pitch of one of the sustained clusters. However,
in the third section, γ, the treatment becomes more refined. Here the density
and speed of the glissandi are subject to a temporal organisation based on
rhythmic values derived from the four duration cells associated with the
‘sectioning’ technique. In the following example, the glissandi divide and
enlarge the sustained chromatic cluster G-sharp-A-natural (Example 6.10a),
their movement being organised by a rhythmic module of twelve values
(Example 6.10b).
Ex. 6.10a PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3e, 4m, partial transcription
Ex. 6.10b PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3e, 4j, partial transcription39
The appearance of this quarter-tonal organisation of glissandi inside of

one single chromatic cluster takes part in an augmenting intensity of sound
in the third formal section γ, produced not by the conventional ways of
increasing intensity and sound mass, but by the possibilities of generating an
intrinsic dynamism of sound matter itself, or, borrowing once more from
physics, an increase of its ‘molecular’ movement.40
39
‘Improvisation III sur Mallarmé’, rehearsal no. 37 (the values are increased
Universal Edition, London, 1975 (= P1), fourfold). ‘P1’ and ‘P2’ hereinafter refer to
pp. 38–9; Pli selon pli (Portrait de Mallarmé), version 1 and version 2 respectively of the
No. 4 ‘Improvisation III’, Universal Edition, work.
40
London, 1982, UE 19521 (= P2), p. 64, ‘Improvisation III’, P1, p. 43; P2, pp. 73–4.
As for individual intervals, quarter-tones manifest themselves according

to two options: (1) they result, as in the first quarter-tone works, from the
deviation from semitonal structures; (2) quarter-tones are used as a form of
textural thickening when grafted around one or several notes of standard
pitch.
(1) The function of unity of displacement may be observed in the layers of
sonority assigned to the three harps, superimposed on a succession of pedal
notes in section α, and conceived according to the organisational principle of
bulles de temps.41 These layers consist of six consecutive complexes containing
a variable number of elements; we will concentrate on the second complex
comprising seven bulles, where the differentiation of the notation by the
colours red and green indicates the alternative modes of execution of the
route chosen: ‘Indicatif I = Harps, play: red’;‘Indicatif II = Voice, play: red and
green’.42 The basic material for the pitches similarly derives from the music
for L’Orestie and consists of a reservoir of seven blocs sonores of variable
density (Example 6.11b–1) assigned individually to each of the seven bulles
of the complex. The quarter-tone shifts are evidently not applied to these
blocs in accordance with an organisational principle (Example 6.11b–2),
but seem to be orientated towards the sounds of one of the two pedal notes
of section α onto which this complex is superimposed (Example 6.11a); we
may note in passing the enrichment of the third bloc by means of its
multiplication with the inversion of the fourth bloc:
Ex. 6.11a PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3c, 4m
The quarter-tonal material thus obtained is presented in registral dis-

placement, as announced by the harp – ‘Indicatif’ (Example 6.11b–3). Its
distance from the microtonal glissandi within the sustained notes leads us
again to consider the question of the acoustic quality of quarter-tone
harmony. Compared with the use of quarter-tones within adjacent inter-
vals, the effect of such clashes diminishes in the context of more extended
intervals. But the aural impression depends also on the nature of the
41
See Boulez, Boulez on Music Today, p. 58; different means. Going from total chance in
the principle for the organisation of the bulles the order to multiple alternatives. Use of
de temps is described in the sketches for series of multiple chords . . . less and less . . .
‘Improvisation III’ (PSS, Sammlung Pierre a single series of chords.’
42
Boulez, Mappe G, Dossier 3e, 3a): ‘Following ‘Improvisation III’, P1, p. 3.
a sign from the conductor, or by some quite
166 Werner Strinz
Ex. 6.11b PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3c, 3b, partial transcription
(1) and analytical rewriting (2, 3)
intervals concerned; quarter-tones in the context of fifths and fourths –

intervals sensitive to acoustic distortion – produce an ‘out-of-tune’ effect
distinct from that in other contexts. Thus, the available acoustic possibi-
lities of quarter-tones provide for a range of aural effects much more
differentiated than the ‘falseness’ of quarter-tonal or microtonal harmo-
nies mentioned by Boulez.43 In the individual elements of the harp bulles,
we observe fourths and fifths in close or extended positions and being
in quarter-tone relationship to the pedal notes. As these pedal notes
are submitted to quarter-tonal deviations produced by the glissandi
mentioned above, the acoustic context of the harp interventions is not
stable. In the first version of ‘Improvisation III’, the choice of alternatives
for the ordering of the seven harp interjections allows them to be heard in
a mobile context – that is to say, the effect of the quarter-tones alters
according to the order chosen. Thus depending on the chosen order of
execution, the perfect fifth E-half-flat/B-half-flat of the second interjection
can be heard as ‘in tune’ in terms of its relationship to the D-three-
quarters-sharp glissando position of the corresponding pedal note
(Example 6.11a).
The displacement into the quarter-tonal sphere of semitonal structures is
the starting point for the generation of some of the vocal phrases. Applied in
a more systematic manner than in the bulles de temps, these lines are laid out
in superimposed form, indicating alternative possibilities of execution,
designated by Boulez with the term échiquier.44 We may observe the intro-
duction of quarter-tones in the initial ‘Indicatif’ of the voice, one of four such
signals placed before the variable sections, and indicating the available
choices of the sub-section to be performed. Two blocs of four notes
(Example 6.12–1) are transformed to a linear cell of seven notes around
the common note C-natural. This centre is the fixed point for the contrac-
tion of the intervals of the cell by bisection, producing quarter-tones when
their distances correspond to odd multiples of semitones (Example 6.12–2).
The result of this shrinkage is inserted in the groups of three notes to the
right and left of this centre (Example 6.12–3). To obtain further derivations,
the notes at the extremes of the initial cell replace C-natural as the centre of
reference. These derivations are completed by the inversion of the initial cell
around its centre C-natural as point of departure (Example 6.12–4).
In the original version of ‘Improvisation III’, the first vocal entrance –
the vocal ‘Indicatif’– consists of a wordless melisma, resulting from
a variable trajectory across semitonal and quarter-tonal ‘échiquier’ fields.
The pitch content of the second ‘échiquier’ field is derived directly from
43 44
See ‘At the Edge of Fertile Land’, Literally, ‘chessboard’.
Stocktakings, pp. 161–4.
168 Werner Strinz
Ex. 6.12 PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3e, 1e, partial transcription
(1, 3, 4) and analytical rewriting (2)
Example 6.12–4, beginning with the right-hand column, read downwards.

The second half of the phrase is formed from the segments of the left-hand
column, and the use of forward and retrograde form for each of the eight
cells is indicated by the arrows.45 The voice ‘Indicatif’ presents the potential
of linear quarter-tonal tension in restricted ambit in a manner as concen-
trated as the spatially separated quarter-tones in the harps ‘Indicatif’.
(2) Introduced in the first harp ‘Indicatif” to the section β, the quarter-
tones appear in the three harp parts in the form of displaced notes which are
grafted onto the equivalent notes of the tempered scale, an effect which
brings to mind Cage’s use of objects fixed to the strings of a prepared piano.
We will observe this condensing of the sound material more closely in
a structural fragment of section β which corresponds to one of the pedal
45
‘Improvisation III’, P1, p. 1; P2, rehearsal
number 4.
Ex. 6.13a PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3e, 4n
Ex. 6.13b PSS, Sammlung Pierre Boulez, Mappe G, Dossier 3e, 4n46
complexes of the ‘multiple sections’ (Example 6.13a). The harp parts are
sketched without the quarter-tone intervals added (Example 6.13b–1). They
only appear in the definitive score, disposed around the axes D-natural 5 –
B-flat 5 – A-natural 6 (Example 6.13b–2); these axes, denser in texture as
a result of the addition of quarter-tones, mark the junctions of register with
those of the flute parts. The addition of a chromatic cluster A-natural 2 –
B-flat 2 – B-natural 2 to the pedal notes reinforces the thickening of the
texture and complements the quarter-tonal layering in the upper tessitura.
The presence of quarter-tones, and particularly the acoustic consequences
of their placing, must be seen as more than simply an evocation of ‘exoti-
cism’ – a Balinese component which complements the other extra-European
references in the third ‘Improvisation’, with its evocation of Peruvian harps,
the Nô elements in the voice, the sounds of the Japanese Shō in the flute
parts.47 For a listener familiar with their divergent cultural significances –
folklorist, theatrical, ritual – the composition presents a complex aesthetic
perspective in which the subliminal presence of these references intertwines
with the extraordinary sounds emanating from the quarter-tone harmony.
It is tempting to see these transitory elements of stylisation in Boulez’s
compositions in relation to one of the central poetic investigations of
Stéphane Mallarmé – the redefinition of the verse as a ‘total word’: ‘Le vers
qui de plusieurs vocables refait un mot total, neuf, étranger à la langue et
46 47
‘Improvisation III’, P1, p. 28; P2, p. 44, Bassetto, ‘Orient-accident? Pli selon pli, ou
rehearsal number 28. l’“eurexcentrisme” selon Boulez’, p. 40.
170 Werner Strinz
comme incantatoire.’48 Like the words in Mallarmé’s sonnet, the harmonic

and stylistic elements of ‘Improvisation III’ are related in an ‘internal
mirroring’, transcending questions of cultural signification and stylisation.
A particular deepening of the ‘Portrait de Mallarmé’ in ‘Improvisation III’ is
confirmed by the very choice of the sonnet ‘À la nue accablante tu’. Besides its
narrative context – the tableau of a ‘furious and dark sea’49 – this sonnet is
closely related to Un coup de dés by the common, significant images of the
shipwreck (‘naufrage’) and the Siren.50 ‘À la nue accablante tu’ shares its
relationship to Un coup de dés with the ‘Sonnet en X’, whose ‘inanité sonore’
and ‘croisée au nord’ (crossing to the North) interferes with the ‘septuor’ (the
instrumental constellation as well as the astronomic constellation of the Great
Bear).51 A possible relation of the ‘Sonnet en X’ to the early Polyphonie project,
in both the role of the number seven and the presence of the letter X in the title
of its second version, lacks direct testimony. But in the same letter to John
Cage in which Boulez expounds the fundamentals of the Polyphonies, he
outlines his intention to extend a ‘microcosm’ of microtonal divisions as far
as a 1/24-tone in the context of a project based on Un coup de dés.52
One of Boulez’s most advanced creative goals, the twofold exploration of
musical space and of harmonic possibilities beyond the limits of
a generalised temperament, seems to be associated with the composer’s
gravitation to Un coup de dés in the 1950s. But unlike the Mallarméan
context of his investigations of distinctively new notions of work, form
and unity in the Third Sonata for piano, the constellations53 of composi-
tional techniques and of heterogeneous musical references in ‘Improvisation
III’ arise from a creative ‘navigation’ of the musical matter itself. To this
extent, the first quarter-tonal experiences of his early works are inextricably
linked to the synthesis of electro-acoustic and instrumental possibilities in
Répons, . . . explosante-fixe . . . and Anthèmes 2.
Translated from French by Peter O’Hagan in collaboration with the author.
48
Mallarmé, Œuvres complètes, vol. ii, Mallarmé, jusqu’à en faire une œuvre longue
p. 1,452. qui me liquidera Mallarmé pour qque temps
49
Butor, ‘Mallarmé selon Boulez’, p. 108. jusqu’à ce que je reprenne goût au Coup de
50
See Marchal, Lecture de Mallarmé, dés – qui sera une synthèse totale’ (PSS,
pp. 251–5. Sammlung Karlheinz Stockhausen).
51 53
Ibid., pp. 165–89. The folders containing the sketches for the
52
Letter to John Cage, 30 December 1950, six compositional techniques of
Nattiez, Boulez–Cage Correspondence, p. 88. ‘Improvisation III’ are marked with a symbol
In a letter to Karlheinz Stockhausen similar to the sign of the scorpion (♏), the
from December 1959, Pierre Boulez dominant constellation of the southern
emphasises the aspect of synthesis in regard hemisphere (PSS).
to Un coup de dés: ‘J’ai agrandi le plan de mon
Peter O’Hagan
That issues of chance and indeterminacy in the music of the avant-garde

were to become among the most contentious of the 1950s is scarcely
apparent until after the midpoint of the decade. Indeed one of the principal
sources concerning the interaction of the prominent personalities of the
period, the early correspondence between John Cage and Pierre Boulez, is
remarkable for its tone of good humour and mutual empathy, with no sign of
the abrupt parting of the ways that would shortly occur. Thus a letter from
Boulez describes the beginning of work on his Polyphonie X in the following
way: ‘Really it will be a collection of 14 or 21 polyphonies (maybe more),
I don’t know yet, very long in duration. But one will be able to select what
one likes.’1 After a technical discussion of the serialisation of quarter-tones
and the generation of rhythmic cells, the letter continues:
Above all, I would like to get rid of the notion of the musical work made to be given
in a concert, with a fixed number of movements. Instead, this is a book of music with
the dimensions of a book of poems (like the grouping of your Sonatas or the Book of
Music for Two Pianos).2
Elsewhere, the letter is shot through with references which were likely to
appeal to the experimental, pioneering spirit of Cage – for example, the
image of a graphic formula to represent microtones, and the prospect of
constructing a specially tuned instrument for use in the abortive Coup de dés
project. Cage’s reply contains some account of the composition of his
Concerto for Prepared Piano and Chamber Orchestra, expressed in terms
with which Boulez was likely to empathise – the use of a bridge to produce
microtones on the piano, and reference to charts of a ‘thematic nature’
producing an ‘athematic’ result.3 Even in the following description of tossing
of coins to determine the various parameters, there are passages which were
likely to reassure Boulez: ‘At this point my primary concern becomes: how to
become mobile in my thought, rather than immobile always. And then I saw
1 3
Nattiez, Boulez–Cage Correspondence, p. 80. Ibid., p. 93.
2
Ibid., p. 86.
171
172 Peter O’Hagan
one day that there was no incompatibility between mobility & immobility
and life contains both.’4 Boulez’s initial response was enthusiastic: ‘I must
write you a long letter soon on the subject of your last letter. I found it
incredibly interesting. We are at the same stage of research.’5 The promised
long letter, which followed in December 1951 after a gap of some five
months, whilst acknowledging Boulez’s interest in the opposition between
mobility and immobility and the hope of arranging a performance of Music
of Changes in Paris, sounds a first note of divergence between the two:
‘The only thing, forgive me, which I am unhappy with, is the method of
absolute chance. On the contrary, I believe that chance must be extremely
controlled.’6
Nonetheless, despite his growing reservations about Cage’s adoption
of chance procedures, Boulez continued to seek opportunities for
the performance of his colleague’s music in Europe. During the
following year, having made the acquaintance of Karlheinz Stockhausen
in Paris, he subsequently facilitated contact between Cage and
Stockhausen regarding a broadcast of Cage’s music by Cologne Radio
in November/December 1952. It was at this time that realisation of the
incompatibility of chance procedures with his own rigorous approach
to deriving all the elements of the composition from the properties
of the series led to a gradual parting of the ways, as is evident by
the time of the Renaud-Barrault Company’s tour to Canada and the
USA in the autumn of 1952.7 During the Company’s season in
New York, Boulez was able to stay at Cage’s apartment, and it was at
this time that he met a number of eminent contemporaries including
Stravinsky. However, his response to a letter from Pierre Souvtchinsky,
enquiring about his and Cage’s support for a proposed conference and
festival to be organised by Nicholas Nabokov in Rome the following year,
is revealing. The reply, sent during the time of his stay at Cage’s apart-
ment, recounts news of a meeting with Varèse, and a forthcoming concert
of musique concrète organised by Virgil Thomson in which his own recent
Deux études would be performed, but is noteworthy also for its brutal
dismissal of Cage: ‘In any case, Cage counts for nothing in the story, and
would absolutely be nothing.’8
It is to other sources that we must now turn for evidence of the evolution
of his style and compositional philosophy during the subsequent years.
The correspondence with Stockhausen provides details of the development
of both composers during the period, and is an invaluable record of their
4 7
Ibid., p. 94. The Company left Paris on 6 October 1952,
5
Ibid., p. 97. and arrived back there on 2 January 1953.
6 8
Ibid., p. 112. Letter sent after 14 November 1952 (BNF).
exchange of ideas, and indeed the convergence of much of their thinking

despite inevitable disagreements. In addition, the letters provide detailed
information with regard to progress on compositional projects, including
Boulez’s visit to Cologne in September 1954 to work in the electronic
studio. It was during this visit that the Symphonie concertante, a work of
Boulez for piano and orchestra dating from 1947, was evidently mislaid,
and subsequently presumed lost. As early as the following month, Boulez
mentions in a letter to Stockhausen a new work for piano and orchestra
to replace the lost score, and during the course of it he gives some details
of the new piece, which would include ‘structures not closed, musical
parentheses, musical italics’.9 In view of the subsequently contentious
atmosphere surrounding the question of indeterminacy, the unavailability
of Stockhausen’s letters to Boulez during this time is especially regrettable,
as his response to such ideas would help shed further light on the genesis
of the works of both composers.
However some idea of the trajectory of Stockhausen’s thinking can be
gathered from another, less visited source, his correspondence with David
Tudor. A letter dating from November 1955 penned in Stockhausen’s
idiosyncratic English contains information about the early stages of the
composition of Gruppen, before continuing: ‘Just in the last months
I begin to feel how I can go a new way of composition including all
consequences which I have proposed when you have been here: statistical
parties [sic] in connection with fixed parties.’10 The language is strikingly
similar to that used by Boulez in a letter to Stockhausen sent at the beginning
of April 1956, concerning the composition of a proposed Third Sonata: ‘the
form is in constant evolution with imprecise zones between homogeneous
zones’.11 Although the Sonata was not ready in time for a first performance
at the 1956 Darmstädter Ferienkurse, Boulez and Stockhausen were in
attendance, as was David Tudor, whose programmes on his first visit to
Darmstadt included works by both composers.12 Stockhausen’s subsequent
account of a discussion with Boulez during this time is worth quoting in full,
especially in view of its claims concerning the chronology of events at that
seminal point in the development of both composers. The context was
evidently a performance by Tudor of Cage’s Music of Changes, followed by
a seminar devoted to the piece:
I told Boulez about Klavierstück XI, which I had written shortly before. At first he
was astonished then he became angry and abusive: he could not understand such
9 12
Stockhausen Stiftung. Among his other duties, Tudor performed
10
The Getty Research Institute, Los Angeles Stockhausen’s Klavierstücke V–VIII on
(980039). 18 July, and took part in a performance of
11
Stockhausen Stiftung. Boulez’s flute Sonatine with Severino
Gazzelloni three days earlier.
174 Peter O’Hagan
nonsense. All this time, Tudor was laughing slyly. I was afraid of fixing everything
exactly in the notation, and wanted to brush off responsibility . . . Then more than
a year passed before Boulez sent me the first sketches of the five formants of his
Third Sonata.13
It is revealing to compare this anecdotal version of events with other

sources, specifically Stockhausen’s correspondence with Tudor during
1956. Stockhausen had sent him scores of Klavierstücke V–VIII
during November/December 1955, and in a letter of 28 February 1956,
expressed the hope that Tudor would play them at Darmstadt during the
1956 Ferienkurse, adding, ‘and in the summer I shall write some new piano
pieces (9, 10 and transformation of 6)’.14 Tudor’s presence at Darmstadt
that year having been confirmed, Stockhausen again wrote to him on
13 April 1956, inviting him to Cologne to record Klavierstücke I–VIII on
either 9 or 10 July, immediately prior to the beginning of the Darmstadt
Ferienkurse.15 There is no further reference at this time in the correspon-
dence to the composition of more Klavierstücke, and indeed Stockhausen
had been occupied during the earlier part of that year with the expansion of
Zeitmasse which was scheduled to receive its première in the new definitive
form at Darmstadt on 15 July.16 Shortly after the end of the Ferienkurse,
Stockhausen invited Tudor to write an article on the music of Cage for
a forthcoming number of the journal Die Reihe, explaining that ‘we can do
something for John’s contact in Europe, and an article by you is the best one.
I think at something like you did in your last lecture about Music of
Changes.’17 The letter continues: ‘After Darmstadt I became a bit ill. I am
copying the Zeitmasse (– two weeks). Then I shall write down the piano piece
(it is growing in my head, more and better).’ The new piece is mentioned in
further correspondence in the course of the autumn, during which period
Stockhausen planned three recitals for Tudor in Cologne during the last
week of November. However, whilst Klavierstücke I–VIII were to be
included in further projected concerts in Vienna, Milan and London, there
is no information about a new piece except for the reference in a letter of
22 October 1956: ‘My new piece cannot be finished until November.’18
Finally, Stockhausen was able to confirm in a letter of 23 February 1957
that Klavierstück XI was now printed, and the proofs awaiting correction,
pending Tudor’s comments.19 The letter goes on to express pleasure at the
13 17
Quoted in Kurtz, p. 87. Letter sent after 22 July and before
14
The Getty Research Institute, Los Angeles 18 August. The Getty Research Institute, Los
(980039). Angeles (980039).
15 18
Ibid. The Getty Research Institute, Los Angeles
16
According to Kurtz, the performance had (980039).
19
to be cancelled at the last minute. See Kurtz, Ibid.
p. 86.
news that Tudor had been re-engaged for the 1957 Darmstadt Ferienkurse,
and the hope that he will be able to perform the new piece there.20
Judging from the correspondence with Tudor, it is certainly the case that
Stockhausen had by this time become increasingly sympathetic towards the
music of Cage and his circle, and that his attitudes were shifting. It may well
be that he had conceived the indeterminate form of Klavierstück XI by the
time of the encounter with Boulez at Darmstadt. However, Stockhausen’s
account of the genesis of the piece in the letters to Tudor casts considerable
doubt on his later claim that it had been completed prior to the 1956
Ferienkurse. Nonetheless, he would certainly have been well aware that
Boulez was likely to find even the conception of indeterminacy highly
provocative, and would make little intellectual distinction between it and
the chance processes of Cage’s Music of Changes. A sense of betrayal must
have sharpened the divide in any confrontation, since although
Stockhausen’s letters to Boulez from this period are missing, it is clear
from the following passage in one of Boulez’s letters that Cage had pre-
viously been the subject of some less than flattering exchanges between the
two younger composers:
I have very much appreciated your reflections on the puberty of Cage. For
myself, I can say that I have passed only a minimum amount of time with him, and
that I have not distressed myself by going to hear his no.2.372.899, or something
similar. I like Tudor very much, but with Cage, I can tolerate less and less a certain
glandular atrophy. All this bazaar, as you say.21
The second part of Stockhausen’s account of a confrontation with Boulez at

Darmstadt in 1956 likewise merits further scrutiny. The reference to Boulez
sending him sketches for the five formants of the Third Sonata ‘more than
a year’ later seems likely to refer to a letter sent by Boulez on his return to
Paris in the immediate aftermath of his first performances of the work, in
Darmstadt on 25 September 1957, and Berlin three days later. In fact, this is
the last in a series of four surviving letters which Boulez wrote to
Stockhausen over a period of less than two months following the end of
the 1957 Darmstadt Ferienkurse, all of them containing information on
progress with the Third Sonata. It is evident from the first of these letters
that Boulez had been on the receiving end of what could well have been
a coordinated series of positive reports on the success of the 1957
Ferienkurse following his own abrupt withdrawal at less than a month’s
notice: ‘I am happy that Darmstadt has gone well. I have had reports from
20
Tudor gave the first performances of that summer, and his place was taken by Paul
Klavierstück XI at Carl Fischer Hall, Jacobs, who likewise gave two performances
New York, on 22 April 1957, playing two of the work on 28 July 1957.
21
versions of the piece. Because of illness, he Letter postmarked 18 November 1954
had to cancel his engagement at Darmstadt (Stockhausen Stiftung).
176 Peter O’Hagan
Schlee and Pousseur. Bravo! But I regret nothing.’22 The letter is remarkable
not least for its conciliatory tone, perhaps partly a consequence of his
uncertainty at Stockhausen’s reaction to his seminar paper, ‘Alea’, delivered
at Darmstadt on 24 July in Boulez’s absence in a German translation by
Heinz-Klaus Metzger, just four days before the European première of
Klavierstück XI. In fact, some of the language of the opening onslaught on
chance procedures in ‘Alea’ is anticipated in the correspondence with
Stockhausen, and could well have been reinforced in private conversations
between the two. It was clearly Boulez’s intention that the attack would be
interpreted by Stockhausen as one on Cage and his disciples rather than on
Stockhausen himself, and he goes on during the course of the letter to
acknowledge receipt of scores of Zeitmasse and Klavierstück XI, adding of
the latter, ‘I am in the process of studying it. It is interesting for me to compare
the solutions which you have found to those I am in the process of employing.’
In the final letter of this period, following the first performances of the
Third Sonata, Boulez gives an outline of the structure of each of the
movements of the still incomplete work, including details of the use of
formal mobility. After enthusing about his discovery of the newly pub-
lished Le ‘Livre’ de Mallarmé by Scherer,23 and drawing parallels with the
open form of the new Sonata, he touchingly goes on to make what
amounts to a plea to Stockhausen’s feelings of comradeship: ‘Dear
Karlheinz, I am anxious to make you a part of this epiphany. Now that
we more or less have a technique sufficiently solid and broad, it is neces-
sary for us to work like mad on the poetic.’24 The poetic impulse at the
heart of Boulez’s thinking had evidently caused some friction at an earlier
period in their correspondence, and at times Boulez had adopted what
must have struck Stockhausen as an irritatingly didactic tone, as in a letter
written towards the end of 1954:
I think that the great innovation that music needs is this pulverisation of unitary
time. In effect, you call that the French spirit. But I believe that in saying that, you
pass completely by the important problem. For God’s sake, reread Joyce and the
Coup de dés of Mallarmé; and you will understand exactly what I want to say.25
22 24
Undated letter to Stockhausen [after Undated letter to Stockhausen [beginning
7 August 1957] (Stockhausen Stiftung). of October 1957] (Stockhausen Stiftung).
25
The original French Mais je ne regrette rien is Undated letter to Stockhausen [end
an ironic reference to the popular song, written of December 1954] (Stockhausen Stiftung):
the previous year and subsequently made ‘Je crois que la grande nouveauté dont
popular in Edith Piaf’s recording. The final a besoin la musique est cette pulvérisation de
verse is as follows: No, nothing of nothing/No, temps unitaire. Vous appelez cela l’esprit
I don’t regret anything/Because my life, because français de la suite. Mais je crois qu’en disant
my joys/Today that starts with you! cela, vous passez complètement à côté du
23
Jacques Scherer, Le ‘Livre’ de Mallarmé problème important. Pour Dieu, relisez Joyce
(Paris: Gallimard, 1957). et de Coup de dés de Mallarmé; et vous verrez
exactement de quoi je veux parler . . . ’.
It could well be the case that the adoption of grace-note interpolations,

later additions to the structural bones of Stockhausen’s second set of
Klavierstücke, could owe something to this onslaught on the limitations of
unitary time. Nonetheless, Boulez’s persistence in attempting to draw
Stockhausen into the world of his own literary inspirations, as well as
indicating a continuing refusal to accept that the contemporary composer
he most esteemed was developing in different directions from his own, also
illustrate the extent to which his own thinking continued to be shaped by the
urge to develop musical syntax along lines paralleling those in poetry.
A mindset which conceived musical coherence in syntactic terms would
inevitably come into conflict with any relinquishing of control over the basic
structural elements. As he expressed it in ‘Alea’: ‘You see what it comes back
to? Always a refusal to choose.’26
*
So far, the chronology of events seems clear, and Stockhausen’s statement
implying that Boulez’s Third Sonata was written well over a year after
Klavierstück XI is certainly an exaggeration. Leaving this aside, the
question of dating remains a complex one, partly because of differences
in the working methods of the two composers, the slow and painstaking
evolution of Boulez’s compositions contrasting with the more fluent and
confident trajectory of Stockhausen’s works. Judging from his letter to
Tudor of 28 February 1956, Stockhausen’s intention during the coming
months was to complete the cycle of six pieces, Klavierstücke V–X, by
composing two new pieces – i.e. IX and X – and completing a revised
version of VI. The evidence suggests that the composition of a new piece
introducing indeterminacy was a separate project, conceived subse-
quently, and completed by the end of 1956, thus leaving the group of
six pieces incomplete for the time being. In the meantime, Boulez had
been working intermittently on a Third Sonata over a period dating back
to the autumn of 1955.27 Independent confirmation of this is provided in
a series of anniversary tributes to Heinrich Strobel to mark the tenth
anniversary of his work with Südwestfunk, Baden-Baden in 1955. There
is some irony in the fact that Stockhausen’s contribution consisted of
a tiny piece for contralto and wind trio, which became in expanded and
revised form the first version of Zeitmasse. Boulez’s equally brief offering
consisted of a piano piece ‘from a work in progress’, the untitled fragment
in question being none other than the unfinished formant ‘Séquence’,
26 27
‘Voyez-vous où l’on en revient? Toujours à A Third Sonata is first mentioned in a letter
un refus de choix.’ Boulez, Stocktakings, p. 28. to Stockhausen written in October 1955
(Stockhausen Stiftung).
178 Peter O’Hagan
the fifth movement of the Third Sonata, in the form in which it was
performed by Boulez in 1957, and dated 9 October 1955. Among
other early drafts for the Sonata is one relating to the fourth formant
‘Strophe’, consisting of a verbal sketch, with the jotting, ‘At the end of
a Strophe, define the Tempo? [sic] and the dynamic of the following.’
This idea closely resembles the procedures in Klavierstück XI, where the
tempo, dynamics and register of the succeeding randomly chosen section
are determined by the indications at the end of the previous section.
Since the dating of sketches cannot be established with any certainty,
it would be rash indeed to confer the accolade of primacy on any
individual, and indeed the central issue remains that of indeterminacy as
defined by each individual composer according to his creative needs: the
nature of the works themselves rather than questions of chronological
precedence.
If one purpose of ‘Alea’ was a carefully worded attack on Cage and his
disciples, as distinct from Stockhausen, the second half of the paper shifts
the argument onto new territory. As Martin Iddon remarks in a most
perceptive account of the context in which Boulez’s ‘Alea’ was delivered by
Metzger: ‘Though it was probably unclear to listeners at the time, what
Boulez was tacitly moving into the frame of discussion was his own Third
Piano Sonata.’28 There is every reason to believe that when he delivered
the text to Metzger for translation at the end of May 1957, Boulez assumed
that he would have time to complete the work prior to the rescheduled
première on 18 July as part of the programme at the Ferienkurse.
Therefore, many of the arguments in the second part of ‘Alea’ are designed
to prepare the ground for a performance of the Third Sonata – in essence,
they form an introduction and an attempt to make a philosophical
distinction between his own approach to formal mobility and that of his
contemporaries:
However, the obsession with the Can replacing the Must is not simply due to
feebleness of compositional resource [i.e. Cage], or the desire to draw the
subjectivity of the player or the listener into the work, and thus force him
continually to make instant choices [i.e. Stockhausen, Klavierstück XI]. One
could find other obvious reasons which are equally justifiable. First as regards
the structure of the work, the rejection of a pre-established structure, the
legitimate wish to construct a kind of labyrinth with a number of paths; on the
other hand, the desire to create a self-renewing kind of mobile complexity,
specifically characteristic of music that is played and interpreted in contrast to the
self-renewing complexity of the machine.29
28 29
Iddon, New Music at Darmstadt, Boulez, Stocktakings, p. 29.
pp. 184–93.
At the première of the Third Sonata on 25 September, although Boulez

performed all five formants, in duration some twenty-five minutes’ worth
of music, only two of the formants had reached their final form. For the
other three – ‘Antiphonie’, ‘Strophe’ and ‘Séquence’ – what Boulez played
consisted of a series of three miniatures lasting under five minutes in total.
Although there were plans as outlined in the letter to Stockhausen to
expand each of them, introducing various degrees of performer choice,
these remained unrealised in 1957, and the pieces were performed using
fixed notation. In the case of the subsequently published formants, ‘Trope’
and ‘Constellation’, both were in essence complete at the time of the
première, and both included the elements of performer choice described
in ‘Alea’. With regard to the central formant, ‘Constellation’, the choice
was between different routes through the score, although each section had
to be performed at some point: hence the allusion elsewhere in ‘Alea’ to
Mallarmé’s Un coup de dés, with its novel approach to syntax and typo-
graphical layout. In ‘Trope’, the choice involved passages in parenthesis in
two of the four sections, with the option (in theory at least) of omitting
some or all of them. This distinction between what are essentially two
different manifestations of choice procedure is noted in ‘Alea’, ironically
placed in parenthesis: ‘(Remember that this choice is not necessarily
a process of selection, but it may be limited to a variable freedom of
execution).’30 Since the compositional procedures remained rigorously
serial, Boulez was at pains to emphasise his continuing control over the
structure as a whole:
[In order] to escape from the complete loss of any global sense of form, as well as to
avoid falling into a kind of improvisation with no other imperative than free will . . .
one must have recourse to a new concept of development which would be essentially
discontinuous, but in a way that is both foreseeable and foreseen; hence the need for
‘formants’ of a work.31
*
In view of the protracted timescale over which the Third Sonata evolved, it
would be of relevance to establish a detailed chronology for the composi-
tion of the work, especially in the context of Stockhausen’s assertions
regarding chronology. Despite the existence of first drafts dating back to
the autumn of 1955, the successive cancellations of the scheduled
première – first at the 1956 Darmstadt Ferienkurse, then an abortive
rescheduling in the autumn, followed by Boulez’s withdrawal from the
‘definitive’ date of 18 July 1957 – suggest that much of the composition
30 31
Ibid., p. 36. Ibid., p. 33.
180 Peter O’Hagan
must have taken place during 1957, with only a proportion of the work
completed during the period between July and September. Unfortunately,
there are remarkably few clues provided by the available correspondence,
although earlier in the same letter to Stockhausen in which Boulez acknowl-
edges receipt of the score of Klavierstück XI, there is a brief account of work
on the Sonata, with the comment, ‘For me, the devil is hiding himself in
a labyrinth’ – a likely reference to continuing last-minute work on
‘Constellation’, with its various pathways and options for performer choice.
Little evidence concerning chronology is provided by the surviving
sketches, although the comparatively more straightforward serial techniques
used in ‘Trope’ suggest that this formant is likely to predate ‘Constellation’.
The successive stages of sketching for ‘Trope’ have certain elements
in common with the procedures in Stockhausen’s Klavierstücke V–VIII,
especially in the composition of a set of ‘skeletons’ (squelettes) for each of
the four sections, which are then elaborated using ‘fields’ (champs) of related
series. The sketches for the Third Sonata include a structural outline for
this second formant in which the relationship between squelette and champs
is established for each of the four sections (labelled α β γ and δ), and even at
this stage there is mention of the order of sections being ad libitum. In two of
the sections, β and γ, the squelette and champs are to be separated from one
another, but crucially, there is no indication as yet that these sections are to
include elements of performer choice. Indeed the sketches for each of the
four sections proceed along very similar lines up to the drafting of a score in
which the familiar titles of the sections, respectively ‘Texte’, ‘Parenthèse’,
‘Commentaire’ and ‘Glose’, replace the Greek letters. In contrast to the
comparatively straightforward opening section, ‘Texte’, the other three
involved a considerable amount of redrafting, even in the case of the squel-
ette itself. An early sketch for ‘Parenthèse’ (Example 7.1) shows in outline
the inverted crab canon formed by the interlocking series, and although the
other musical parameters are not as yet fully defined, the points of separation
at which the champs was to be inserted are clearly marked by vertical arrows.
Even the pitch content of the champs is indicated, with the related series
tentatively identified by letters placed above the stave.32 The midpoint of the
palindrome is marked by a fermata, and the placing of the insertions is
symmetrical, with those of the second half a mirror image of the first part.
It was only at the next stage of the sketching process that some adjustment was
made to this balanced structure, so that it becomes slightly ‘tilted’ in the second
half of the section:
32
For a detailed analysis of the pitch content
of ‘Trope’, see O’Hagan, ‘Pierre Boulez:
“Sonate, que me veux-tu?”’
Ex. 7.1 Third Sonata: sketch for ‘Parenthèse’
Sketches for the champs appear alongside those for the framing
squelette, but without any indication as yet that they are to play
a subsidiary role. It was not until the final drafts that the familiar marks
of parenthesis appear, clearly separating the interpolations from the crab
canon, and only with the publication of the score and its preface was the
relationship between the two formally defined: ‘In two sections – “Parenthèse”
and “Commentaire” – there are both compulsory and optional structures:
these last, in smaller type – enclosed between brackets – can be played or
omitted independently of each other.’33
A study of the practical operation of these procedures is somewhat
circumscribed by the fact that early performances are confined to those
by the composer himself – no doubt a reflection of the still fragmentary
state of the work in 1957–8, when the majority of Boulez’s performances
took place. Noting in passing the pianistic virtuosity displayed by the
composer, a remarkable feature of these performances is their similarity
of approach to the question of performer choice. The première itself was
followed by a second performance in the same concert, and the occasion
was therefore an ideal opportunity for Boulez to demonstrate the range
of possibilities available to the executant. Perhaps he was mindful of the
example set by David Tudor in his performances of Stockhausen’s
Klavierstück XI when, encouraged by the composer, he gave repeat
performances exhibiting considerable diversity. A comparison of
33
UE 13292 (1961), preface.
182 Peter O’Hagan
Boulez’s two performances of the Third Sonata, recorded at a specially

arranged concert at the Kranichsteiner Musikinstitut, Darmstadt on
25 September 1957, is highly revealing.34 He was evidently intent on
demonstrating the unique circular structure of the work, and as
a consequence, the order of the five formants is reversed in the second
performance:
Table 7.1
1957 (1) Strophe Séquence Constellation Antiphonie Trope

1957 (2) Trope Antiphonie Constellation-Miroir Séquence Strophe
The principle of complementary positioning of formants is observed, with

‘Constellation’ placed at the centre, and pairs of formants revolving round it,
‘Séquence’ balanced by ‘Antiphonie’, and ‘Strophe’ and ‘Trope’ framing the
work – the latter already in complete form. The largest formant,
‘Constellation’, is played in both its forward and mirror versions, with the
six sections being reversed in order the second time – a choice still denied the
modern performer, with only the mirror version so far having been
published.
Most intriguing is the operation of choice on a more local level, even
though the range of possibilities is restricted by the incomplete state
of three of the five formants. Despite this limitation, Boulez allows
himself to reverse the order of the first two of the five sections of
‘Antiphonie’ in the second performance – a procedure which would
subsequently be forbidden in the revised form of the formant, where
alternative versions of the sections are joined as a group of two followed
by a group of three, the order within sections being fixed. By the time
of Boulez’s visit to Darmstadt two years later, this alternative version of
‘Antiphonie’ as described in the lecture ‘Sonate, que me veux-tu?’35
was complete, and Boulez’s performance after his delivery of the
lecture consisted of the original version of the first two sections, and
the new version of the three final sections. Unfortunately ‘Antiphonie’
has remained unpublished, but performances of this version of the
formant were given by Leonard Stein in the first American concerts,
and subsequently by Marcelle Mercenier, before it was entirely
withdrawn.
Most interesting of all in Boulez’s performances is the situation
concerning the circular form of ‘Trope’, with the choice of beginning
34 35
Archiv IMD. The lecture is reprinted in Boulez, Points de
repère I: Imaginer, pp. 431–44.
with any of the four sections of the spiral, and the additional option of
omitting passages in parenthesis. With regard to the first level of
choice, Boulez’s four available recorded performances show consider-
able variety:
Table 7.2
1957 (1) Glose Texte Parenthèse Commentaire

1957 (2) Parenthèse Commentaire Glose Texte
Cologne 1958 Glose Texte Parenthèse Commentaire
Darmstadt 1959 Texte Parenthèse Glose Commentaire
Since ‘Trope’ has remained the only complete formant apart from
‘Constellation’, the position of which was fixed, the tendency was to
conclude performances of the Sonata with ‘Trope’, the only exception
being the second Darmstadt performance, and it was therefore natural
that Boulez would arrange the sections so as to conclude the work as
a whole with the musically climactic ‘Commentaire’. The position
regarding the optional sections in ‘Parenthèse’ and ‘Commentaire’ how-
ever is revealing, since there is not a single instance in any of the four
recordings of Boulez omitting a section in parenthesis. This is the
practice followed by the first generation of performers, including in
the first commercial recording made by Charles Rosen under Boulez’s
supervision. Some more recent performers have taken the composer at
his word, and chosen to omit sections of ‘Parenthèse’, which raises the
question of the status of these parenthetical sections and their function
in the movement – put more directly, is there a rationale for making
these decisions, an authentic case for disregarding the composer’s own
practice as a performer?
The extreme solution to the question of performer choice in
‘Parenthèse’ is simply to omit all the optional passages, which although
permissible within the parameters set by the composer, violates the spirit
of the work since it destroys the proportions between the four sections of
‘Trope’. And yet . . . to omit any of the five parentheses within this
palindrome is to risk unbalancing a structure which is already delicately
poised. Just how delicately poised is illustrated by the exact centre of the
section, the sustained D-natural, which acts like a fulcrum supporting
the two axes rotating back and forth from the opening and closing
G-sharp. Such tritonal relationships dominate the Third Sonata as
a whole, and indeed ‘Parenthèse’ is almost a study for the much more
ambitious scale of ‘Constellation’, the still heart of the work, where the
two axes are presented as giant reflectors of each other, with one of the
two versions chosen by the performer. In ‘Parenthèse’, the sustained
184 Peter O’Hagan
D-natural is a background presence in the central parenthesis which

itself has a strong palindromic element, with matching cells fanning
outwards from the axis. This point is marked by an explosive eight-part
arpeggiated chord, which is immediately repeated in transposed form,
furtif, before dissolving to a ppp dynamic level – the moment of the
most extreme contrast of mood and dynamic in the section. The more
such relationships and compositional subtleties are understood, the
more ‘Parenthèse’ is seen in relation to the intended overall structure
of the Sonata, the more likely is one to follow the composer’s lead in
performing the parenthetical sections in their entirety.
In terms of performer choice, ‘Parenthèse’ presents a conundrum
which does not recur elsewhere in the work to anything like the same
extent. The most musically complex section of the four comprising
‘Trope’ is ‘Commentaire’, which shares with ‘Parenthèse’ the appear-
ance of optional passages in brackets. Whilst these play a similar
role in providing commentary on and elaboration of the original
squelette, the division between it and the surrounding champs is not
as clear-cut as in ‘Parenthèse’, since the compulsory sections them-
selves incorporate commentaries on the squelette. The overall structure
is therefore a more open one with compulsory passages more extended
and developed than is the case with ‘Parenthèse’. Thus an acceptance
of the invitation to omit one or two of the shorter bracketed passages
can be decided on without detriment to the overall structure of the
section.
The other published formant, ‘Constellation’, raises issues of perfor-
mer choice of a different order from those in ‘Trope’. It is worth
remembering that in its published form at least, the movement must
be considered unfinished, with the original intention to publish the two
versions on either side of a large folding sheet never having been
realised. I have argued elsewhere that the disastrous saga of the pub-
lication of ‘Constellation’ may well have contributed to the composer’s
abandonment of the work.36 Be that as it may, as the publication stands,
the performer is deprived of the fundamental choice of performing the
six sections in forward order beginning with Points 1 (‘Constellation’),
or with the order of sections reversed (‘Constellation-Miroir’): rather
perversely, as already noted, only the latter order has been published,
and that in a format which makes the original intention of spontaneous
choice impossible without considerable adjustment to the layout of the
published score. Boulez’s intention of providing the performer with
36
O’Hagan, ‘Pierre Boulez: “Sonate, que me
veux-tu?”’, pp. 253–5.
what amounts to a musical mirror – an invitation to reveal the work’s

structure, and incidentally to reflect the pianist’s own personality by
interaction on a local level with the choices provided – is rooted in the
formant’s design. In some respects this movement has a claim to be
considered the most perfect in its integration of the series and the
overall form of any work by Boulez, consisting as it does of
a statement of all forty-eight possible forms of the series, with virtually
no repetition. Furthermore, the six sections comprising ‘Constellation’
are paralleled by the sixfold division of the series itself, and this division
is based on a principle of rotation so that the segmentation of the series
is constantly varied – a microcosm of the mirror structure of the
formant as a whole. The three Points are confined to linear statements
of the inverted form of the series, whilst the two alternating blocs
sections employ transpositions of the prime form, but with the applica-
tion of the technique of chord multiplication to enhance the resonance;
therefore a circular design, which by its nature retains its identity
irrespective of whether performed in forward or reverse form, hence
the decision to present the formant in each of its two forms at the
première.
In Boulez’s performances on that autumn evening in Darmstadt, the
choice of order within the six sections of ‘Constellation’ and
‘Constellation-Miroir’ shows some variation, but his performances,
whilst remarkable in their range of resonances, demonstrate some of the
difficulties imposed by the format of the score, even in its manuscript
form as used by Boulez, consisting of a large folding sheet with the two
versions of the movement written on reverse sides. Put in straightforward
terms, it is extremely difficult for the performer to keep track of the order
of sections if the decisions are made, as it were, on the spur of the
moment. In Blocs II, for example, there are no less than twenty fragments
of lengths ranging from a whole line to a single chord, and the possibility
of omitting one or more is an ever present danger – as revealed on
occasion in these performances. The issue is an important one, and in
two subsequent recorded performances Boulez adopted a more uniform
approach to the ordering of sections, suggesting that a route through
the various alternatives was planned in advance.37 The question of
choice within the formant has been exacerbated for subsequent perfor-
mers not only by the unwieldy format of the published score of
37
A performance at Cologne took place on Darmstadt was given on 30 August 1959, and
24 March 1958 in a concert which included preceded by the lecture ‘Sonate, que me
the première of Stockhausen’s Gruppen. veux-tu?’.
Boulez’s subsequent performance at
186 Peter O’Hagan
‘Constellation-Miroir’, but also by the fact that Boulez added consider-

ably to the range of directional arrows at the proof stage prior to
publication. Bearing this in mind, courageous indeed would be the
performer who would leave the operation of choice entirely to the
occasion of the performance. Where more than one performance by
the same artist can be compared, as for instance those by Rosen and
Bernhard Wambach, it is highly likely to be the case that the ordering of
the fragments of ‘Constellation’ are very similar, and indeed in some
cases identical, suggesting a high degree of preparation and also an
acknowledgement of the fact that despite the range of directional
arrows, the possibilities remain tightly controlled by the composer.
It is indeed a paradox that a work which started life as the composer’s
most ambitious attempt to realise his vision of an integration of musical
form with a Mallarméan aesthetic, should have become, through
the circumstances of its publication and the incompleteness of the
other formants, something of an isolated experiment in the realm of
performer choice.
*
Throughout this period, another ‘work in progress’ remained in the
background. The year prior to the first performance of the Third
Sonata, Boulez had completed the first of a projected three pieces which
would comprise a second volume of Structures, the first performance of
which took place as part of the Donaueschinger Musiktage on
21 October 1956. The work would certainly have been in his thoughts
during 1957, since he gave several further performances with Yvonne
Loriod of all four extant pieces of Structures during the earlier part of
that year, the last of them only some four months before the delivery
in July of ‘Alea’ at Darmstadt. Although it was not until 1961 that an
additional piece was finished, Boulez devoted a section of ‘Alea’ to the
incorporation of mobile structures in writing for ensembles, citing the
combination of two pianos. The language is so specific that it might
almost be a description of aspects of the formal structure of Structures,
Deuxième livre, Chapitre II, an indication that however last-minute the
detailed working out of the composition was to be, the essential features
of a mobile structure had been germinating over several years: ‘When
there are two instrumentalists – perhaps two pianos – the problem is
hardly any greater, given the supplementary adoption of signals and
common reference points.’38 He goes on to describe the temporal rela-
tionships possible between the two instruments, a concept he was
38
subsequently to develop three years later in the series of Darmstadt

lectures ‘Penser la musique aujourd’hui’: ‘If a fixed tempo on one
piano is combined with a variable one, an accelerando for example,
on the other, it will suffice to mark the points of synchronisation – of
departure and arrival – of the two structures, whose co-ordinates will
have been plotted by the composer with as much precision as he
needed at the time.’39 The description presages the passage at the
conclusion of the cycle of canons in the first half of the movement,
which is accompanied by insertions of groups of grace-note figurations
in free time.
In the second part of Chapitre II, the principle of the troped insertion
is taken onto a new level, with three sets of interpolations, played by
Piano 1, folded within the slowly evolving series of arpeggiated chords
in Piano 2, and the whole coordinated by an exchange of signals
between the two pianists. Two extended pièces mark the outer limits
of the insertions, enclosing six textes, arranged in two groups of three,
with four encarts at the centre of the section. Each pièce is to all intents
and purposes through composed, although both have a bipartite struc-
ture with the choice of playing straight through or pausing at the
central fermata and allowing Piano 2 to continue its series of arpeg-
giated chords, whilst awaiting a signal to resume playing the second
half of the pièce. Although the six textes are modest in proportion
compared to the pièces, they incorporate an element of choice in the
placing of the two groups, commencing with either textes 1–3 or alter-
natively 4–6 – but with no variation in order within the group. More
striking are the notational innovations of the textes, which have an
appearance akin to the neumes of medieval music. The music of
Machaut and Dufay had exercised a fascination for Boulez throughout
the previous decade, and one of his programming innovations in the
Domaine Musical series was the juxtaposition of masterpieces of med-
ieval music with first performances of music by his contemporaries.
Here, the introduction of notations reminiscent of neumes results from
Boulez’s decision to replace conventional rhythmic notation with signs
which are ‘intentionally imprecise’, and which can be varied in the
approximate proportion 3/2. These are listed in the preface to the textes
(see Table 7.3).
Equally imprecise is the notation of attacks and dynamics, each con-
fined to four levels and with some freedom in their application. Whilst
the extension of choice to these musical parameters was to have been
a feature of two of the incomplete formants of the Third Sonata,
39
Ibid.
188 Peter O’Hagan
Table 7.3
‘Strophe’ and ‘Séquence’, work on which dated back some six years
before the completion of Chapitre II of Structures, there is a sense that
Boulez was under some pressure to respond to the radical notational
innovations of his contemporaries in the intervening period. His per-
formance of the Third Sonata at Darmstadt in 1959 occurred only five
days after the première of Stockhausen’s percussion piece Zyklus at the
opening concert on 25 August, and during the same week Stockhausen
presented his lecture series ‘Musik und Graphik’ in which he discussed
the graphic notation in his new piece as well as Cage’s Concert for Piano
and Orchestra. Needless to say, Boulez’s response would be on his own
terms, and in the lecture ‘Sonate, que me veux-tu?’, which prefaced the
performance of the Third Sonata, he renewed his opposition to nota-
tional novelty for its own sake: ‘Altering the physical appearance of
a work without any real interior necessity to justify changing the impact
of the score on the eye could so easily result in amusing, decorative
“calligrams”, fashionable gimmicks in fact.’40 Above all, in the case of
certain (unnamed) experiments, ‘there is no feeling that the desire to
alter the exterior form corresponds to any interior, structural
remodelling’.41 Returning to Structures, Chapitre II, the most significant
loosening of control in the six textes occurs in those notated on two or
three systems, in which the performer is invited to alternate or super-
impose the material between the fermatas. At first sight it is difficult to
reconcile this freedom with the need for an ‘interior necessity’, but
closer examination reveals that the material for each system consists of
the same basic cells, albeit in transposed form and varied sequence.
Hence the freedom is more apparent than real: not strictly speaking an
illusion of choice, but one circumscribed by a rigorously controlled
framework.
40 41
Boulez, Orientations, p. 147. Ibid.
The material for all three of the groups of insertions in the movement –
pièces, textes and encarts – is drawn from eight bars of material borrowed
from the first piece (Chapitre I) of Structures, Deuxième livre. This
passage (bb. 170–7) is one of two short sections in that movement
where Boulez reverts to the linear writing of the three pieces of
Structures, Premier livre (1951–2), and its use here, albeit in heavily
disguised and elaborated form, helps to maintain a unity of style between
two pieces separated in composition by half a decade. The material is first
sketched in retrograde form, before being split into three sets of seven
cells, one set for each of the three groups of insertions. In the case of the
four encarts, the retrograde form of bb. 172–3 is used as the basis for the
pitch material. The example below (Example 7.2) shows the excerpt in its
original form (A), then the first sketch of the retrograded material (B),
followed by an intermediate sketch which breaks it into linear form (C).
The final stage involves the elaboration of each of the cells by combining
them in vertically enriched form in encart 1. The result is shown in the
final line of the example, in which the seven cells are laid out in sequence
at the beginning of the section. Whilst the process of pitch enrichment
generates considerable textural complexity, the beam groupings remain
the same as in the previous sketch:
Ex. 7.2 Structures, Deuxième livre, Chapitre II: the derivation of encart 1
190 Peter O’Hagan
The second half of encart 1 is based on the same seven cells, but permutat-
ing their order. Boulez is able to generate a remarkable quantity of material
from these tiny fragments borrowed from Chapitre I, and encart 2 takes as its
starting point the inversion of the cells of Example 7.2. This unity helps to
explain his decision to loosen the formal structure to a greater extent than in
the other interpolated groups, extending the principle of choice to allow the
inclusion or omission of whole sections. If this might suggest a return to the
design of ‘Trope’, with its parenthetical passages incorporated within two of
the four sections, the options for the four encarts are more systematically
planned, and more restricted than the potentially structurally disruptive
degree of license in ‘Trope’. The choice is to play a minimum of two encarts,
either 1 or 3, and 2 or 4, with the additional possibilities of playing all four
and playing three, thus omitting one. Encarts 1 and 3 are complementary,
sharing the same musical gestures, and comparatively fixed in formal

terms. In the case of the even-numbered encarts, 2 and 4, a much more
mobile structure allows for eight possible orderings of the eight sections,
exhaustively listed in the performance instructions. The scheme of eight
cells for dynamics and attacks is fixed, hence applied irrespective of the
chosen order of sections. Tempo fluctuations are encouraged by the
marking for each of these sections: D’une extrême souplesse; très capri-
cieuse et fantastique, and are indicated by the use of undulating lines, or
for more irregular changes, directional arrows, which recall the typogra-
phy of Stockhausen’s Klavierstück X. In the context of his writing for
piano, the encarts are at the limits of Boulez’s incorporation of indeter-
minacy, yet even here, there are broad structural connections between the
four pieces, and the use of inversion and retrograde applied to whole
sections maintains the serial principle whilst conceding a degree of local
freedom to the performers.
A question arises at this point as to the sheer overload of notational
information in scores such as this. If a legitimate criticism of the notation
of the first totally serial works of the early 1950s may be the impossibility
of realising in performance all the detail (which at times, especially in the
relationship between dynamics and attack, can verge on the contradic-
tory), a parallel problem can arise with performer choice. Put another
way, if the overall structure of a section has been defined by the composer,
the performer can be overburdened by the sheer quantity of alternative
options in the detail. Encarts 2 and 4 represent a particular problem in this
regard: on what basis is a performer to choose between the eight possible
orderings of these sections – and just as importantly, when? This is a point
addressed by Pousseur in the preface to his two-piano work Mobile,
completed in 1959 midway between the two chapters of Structures,
Deuxième livre, and dedicated to Boulez: ‘The order of the mobile
sections can be planned and prepared beforehand, or chosen during the
moment itself, or even left entirely to chance.’ In the performing direc-
tions for his own work, Boulez is silent on this issue, although he would
presumably have concurred with the statement, at least until the final
phrase, where Pousseur is clearly attempting to triangulate between the
extremes of chance and choice. The issue of preparation nonetheless
presents a considerable practical problem in the performance of
Chapitre II, and again the dilemma posed by ‘Constellation’ confronts
the performer: whether to ‘play safe’, determining the choices in advance,
or maintain a degree of spontaneity at the risk of omitting material.
Unlike the Third Sonata, where there are several recorded performances
from which to gain an insight into Boulez’s approach to the question of
indeterminacy, there are scarcely any known occasions on which Boulez
192 Peter O’Hagan
performed Structures, Deuxième livre, Chapitre II after the première,

although he did participate in a performance (with Karl Kohn) at the
University of California Monday Evening Concerts in March during his
three-month sojourn in the USA in 1963 – a by now rare appearance as
a pianist.42 In any case, Boulez evidently took the somewhat less demand-
ing Piano 2 part on these occasions, and virtually all the decisions in terms
of the timing and placing of the three groups of insertions are made by
Piano 1.
In many ways, the final completed chapter of Structures marks
a watershed in Boulez’s development. Although the succeeding ensemble
work Éclat (1965) features an important solo piano part, Chapitre II is the
final published piece of this period in which the piano is exclusively used
as a solo instrument. Moreover, there was a gap of some thirty years
before another work for solo piano appeared, during which time his
compositional concerns had shifted, and many of the issues addressed
in ‘Alea’ had taken on a historical perspective. The importance of alea-
toricism as a compositional principle faded during the course of the
decade of the 1960s, and with the various revisions of Pli selon pli which
eventually eliminated virtually all the elements of indeterminacy from the
score, Structures, Deuxième livre, Chapitre II remains one of the very
few completed works from this period in which mobile elements have
survived. Given the seemingly abandoned, incomplete form of the
Third Sonata, it remains one of the most perennially fascinating, if least
performed, of any of Boulez’s works. In its integration of indeterminacy
with serial principles it is at the same time a challenge to his contempor-
aries and a highly imaginative and creative solution to the concerns
articulated in ‘Alea’.
42
Boulez gave a performance of the work on Loriod. Plans for a subsequent performance
2 September 1965 in a concert at the at a BBC concert in the Royal Festival Hall
Edinburgh International Festival (see were eventually abandoned.
Chapter 13) in which he partnered Yvonne
8 Casting New Light on Boulezian Serialism:
Unpredictability and Free Choice in the Composition
of Pli selon pli – portrait de Mallarmé
Erling E. Guldbrandsen
Les idées musicales n’existent

que réalisées dans l’écriture. En
elles-mêmes, elles ne sont rien.
Pierre Boulez
The familiar picture of European post-war modernism of the 1950s as

a rationalist aesthetic of ‘strictly logical composition’ is certainly changing,
but, regrettably, rather slowly. From the 1950s onwards, mainstream struc-
tural analyses of Pierre Boulez’s music seem to have taken for granted
a traditional representation of serial thinking that emphasises a striving for
airtight structural unity and rational compositional control. This chapter
takes a critical look at such notions, based on analyses of his largest work, Pli
selon pli.
To be sure, many scholars have detected traits of compositional ‘freedom’
in Boulez, but these have been too easily misconstrued as deviations from,
rather than constitutive parts of, the serial procedures themselves. Even
those who have unravelled unforeseen complexities in Boulez’s technical
methods have seemed unwilling to rethink their tacit aesthetic presupposi-
tions of structural coherence and control.1 Likewise, early critiques of serial-
ism from figures as various as Xenakis, Ligeti and Ruwet all seem to derive
from the common misunderstanding that the aim of serial composition is to
communicate the ‘serial content’ itself across the aesthetic gulf from its
desktop construction to the ears of the listener.2 Moreover, the rationalist
picture of serialism entails a certain historiographical construction through
which post-war modernism is construed as a fundamental rupture with the
classical-romantic tradition.3 This historical construction is just as mislead-
ing as the analytical one; indeed, the two are intertwined. The tide started to
turn after 1986, when the collections of sketches and manuscripts at the Paul
1 3
See, for example, Koblyakov, A World of See, for instance, Danuser, Neues
Harmony. Handbuch, pp. 303–7; Morgan, Twentieth-
2
Xenakis, ‘Crise de la musique sérielle’; Century Music, p. 334; and Taruskin, Music
Ligeti, ‘Entscheidung und Automatik’; in the Late Twentieth Century, pp. 27–50.
Ruwet, ‘Contradictions’.
193
194 Erling E. Guldbrandsen
Sacher Foundation were made accessible to researchers.4 Still, with valu-

able exceptions in recent decades,5 the overwhelming amount of literature
on Boulez and serialism has too long relied upon such concepts as
unity, coherence, consistency, order, strictness, discipline, logic, necessity,
formalism, constructivism, scientism and rationalist compositional
control. I refer to this rather one-sided picture as the ‘unity and control
model’ of serialism.6
This chapter concerns itself with the compositional techniques in Pli
selon pli – portrait de Mallarmé.7 Boulez created and developed this five-
movement work for soprano and orchestra – which consists of three
middle movements, ‘Improvisations I–III sur Mallarmé’, framed by the
outer movements ‘Don’ and ‘Tombeau’ – between 1957 and 1962, with
later revisions. Compared to the textbook version of what modernist,
serialist composition in the 1950s and 1960s was all about, the musical
structures in Pli selon pli appear to have been generated in both unexpected
and peculiar ways. The multiple serial techniques, that is, appear to generate
a structural ‘raw material’ that requires further articulation by Boulez the
practical musician, who often departs freely from the procedures that
generated the structures in the first place. Also, in the final phrasing and
articulation of those structures into his stylistically characteristic musical
textures, he appears to make, again, free aesthetic choices following his own
musical taste and judgement. Indeed, Boulez sometimes works in a manner
that evokes Beethoven’s famous sketchbooks much more than the pinched
formulae of rationalist serial composition.
First published in 1992,8 my analyses of compositional procedures in Pli
selon pli later lent themselves to a wider interpretive context, one that
encompassed Boulez’s particularly complex relationship to the Western
musical tradition and his acute readings of Mallarmé’s literary poetics and
the sonnets used in Pli selon pli.9 Also, these analytical findings further
informed my discussions of musical phrasing and articulation, aesthetics
and listening, and I applied them to the complex formal processes of ‘Don’
4
See Sammlung Pierre Boulez, Paul Sacher example, Ligeti, ‘Entscheidung und
Stiftung, Basel (hereinafter ‘PSS’). Automatik’; Ruwet, ‘Contradictions’;
5
See Piencikowski, ‘Nature morte’, as well as Danuser, Neues Handbuch; Morgan,
his several other studies of music by Boulez. Twentieth-Century Music; Born,
See also, for instance, Albèra, Entretiens et Rationalizing Culture; and Griffiths,
études; Leleu and Decroupet, Techniques ‘Serialism’.
7
d’écriture; and Goldman, Musical Language Pli selon pli – portrait de Mallarmé for
of Boulez. soprano and orchestra (1957–62; 1982–3;
6
Variations upon the ‘unity and control’ 1989–90), Universal Edition.
8
model of serialism appear in countless music Guldbrandsen, ‘Serialismen i nytt lys’.
9
history books, articles and textbooks on Guldbrandsen, Tradisjon og
serialism and on Boulez’s music; see, for tradisjonsbrudd.
195 Unpredictability and Free Choice in Pli selon pli
and of ‘Improvisation III’ in particular.10 In a conversation with Boulez at

the Cité de la musique in Paris in 1996, I presented my seemingly icono-
clastic readings of serialism to him and was met with enthusiastic confirma-
tion, and in 2011 the complete exchange was published by Cambridge
University Press in four successive issues of Tempo.11
More than forty years after the invention of serialism, Georgina Born
still proclaims the following (in her 1995 study on Boulez and IRCAM
entitled Rationalizing Culture): ‘We can discern, then, in this period
a process of growing legitimation of serialism to which the character
of the discourse – rationalist, determinist, theoreticist, formalist, scien-
tistic, concerned with high technology – was particularly well suited.’12
Methodologically speaking, this characterisation of serialism seems to
have been seamlessly intertwined with the imperatives of hardcore struc-
tural analysis in ‘formalist’ musicology from the 1950s to the 1980s on
both sides of the Atlantic.13 A similar inclination also seems to permeate
readings of Boulez’s theoretical articles, from the Relevés d’apprenti to
Leçons de musique and beyond.
It is true that Boulez’s own rhetorical strategies as theorist and polemi-
cist have themselves contributed to the ‘rationalist’ optics that have
governed our picture of Boulez the composer. Several statements on his
part, particularly in his early writings (mainly collected in Relevés
d’apprenti and Penser la musique aujourd’hui), could point in such
a direction. Nevertheless, there is tension here, since many of his texts
are distinctly ambiguous in this regard. He repeatedly mentions the
powerful and various aesthetic influences on his compositional thinking
from literature, poetry, visual arts, architecture and non-European music,
and this is not merely a nod to some vague artistic inspiration: Boulez
studied reconstructions of the presumed ‘grammar’ of artistic language in
Schoenberg, Webern and Berg, Stravinsky, Debussy, Cézanne, Kandinsky
and Klee, Joyce, Michaux, Char, Cummings and – not least – Mallarmé,
whose poems he read with enthusiasm from 1948 onwards, and whose
poetics in Le Livre de Mallarmé became seminal to Boulez’s rethinking of
musical language, ‘mobility’ and form.14
In addition to his abiding interests in the arts, Boulez describes, from the
early 1950s, an irreducibly unpredictable dimension to his generative and
technical procedures – one which directly challenges any insistence upon
10 12
Guldbrandsen, Tradisjon og tradisjons- Born, Rationalization, p. 54.
13
brudd, pp. 507–88. See Ian Bent’s description in Bent,
11
Guldbrandsen, ‘Pierre Boulez in Interview, Analysis, and Joseph Kerman’s diagnosis of
1996’. For our discussion of serial techniques, Western musicology and the hegemonic
see esp. Tempo 65/256 (April 2011), ‘Part II: position of structural analysis in Musicology.
14
Serialism Revisited’, pp. 18–24. Published by Jacques Scherer in 1957.
exclusively serialist coherence and control. Also, since the early 1960s, the
interplay between Boulez’s compositional work and his musical practice as
an orchestral conductor became increasingly apparent. Later I take up his
conducting of Wagner, Debussy, Berg and especially Mahler in the 1960s
and 1970s in relation to his revisions to the formal process in ‘Improvisation
III sur Mallarmé’ from 1959 to 1982 and beyond.15
Into the Labyrinths of Serial Generation

In this chapter I will concentrate on ‘Improvisations I and II sur Mallarmé’
alone, in an attempt to illustrate the nature of compositional processes
that are at work on a much broader scale in Boulez’s music as a whole.
The two ‘Improvisations’ were composed as separate pieces in June/July
and December 1957 and later came to constitute the second and third
movements of Pli selon pli. I will begin by looking at the serial generation
of the vocal line in ‘Improvisation I’: ‘Le vierge, le vivace et le bel
aujourd’hui’.16 This is a comparatively brief movement of seventy-eight
bars and is approximately six minutes in duration; the vocal line, in addition,
remains unchanged in the 1962 version of this piece for soprano with a larger
chamber ensemble.
The form of the movement appears to be quite simple, with clear seg-
mentations in direct accord with the sonnet form of Mallarmé’s text.
The soprano performs the fourteen sonnet verses chronologically, and
when the sonnet ends, so does the work. Between the four strophes of the
poem, in the ‘blank’ areas on the page (les blancs, according to Mallarmé),
are three instrumental interludes and a postlude. Example 8.1 shows the first
verse of the soprano part.17
The erratic agility of the vocal line can be seen in its contour, which
features frequent leaps between different fixed pitch areas. It reflects an
Ex. 8.1 ‘Improvisation I’, soprano, verse 1
15 17
Guldbrandsen, ‘Modernist Composer’. Ibid., p. 1.
16
Boulez, ‘Improvisation I’.
almost consistent syllabic phrasing, and the word grouping is largely true to
the syntax. Throughout the movement, the differentiated rhythms and
dynamics partly support and partly counter the phrasing, as does the con-
tinuous alternation between grace notes and principal notes.
At this first stage, for the sake of clarity, I will focus on the generation of
the pitch structure of the vocal line, methodically setting aside aspects
such as text, phrasing, rhythmic figures, instrumental textures and timbre,
musical form and so on in order to trace the extent of compositional ‘unity
and control’ in the organisation of pitch. My readings of Mallarmé’s ‘swan
sonnet’ and its relation to the music are presented elsewhere.18 In the
context of this chapter, I shall instead escort the reader through the stages
of my structural-analytical process. From beginning to end, then, what
follows is the total pitch material of the soprano in ‘Improvisation I’
(Example 8.2):
Ex. 8.2 ‘Improvisation I’, complete vocal line, pitches
18
See Guldbrandsen, Tradisjon og tradis-
jonsbrudd, pp. 251–380.
In Example 8.2, the fourteen lines of the sonnet text by Mallarmé corre-
spond to the sections marked from 1 to 14. When one listens to this vocal
line, it is hard to ignore the large intervallic leaps, the many note repetitions
and the slowly rotating registers. Each line sees the soprano move around in
a self-defined static room of fixed pitches – a kind of vertical harmonic
spectrum that is traversed by a leaping line that, needless to say, does not
evoke thematic twelve-tone thinking à la Schoenberg. Boulez’s writings in
the early and mid-1950s verified instead his express concern not with the
thematic principles of the row but with its generative function. The main
impression of this vocal line, in turn, is one of unusual freedom and
flexibility – there is an almost improvisatory air to the progression that
could just as well be the result of direct composition with no predetermining
system or calculated principle.
Still, might we detect any stricter organisational logic in this pitch field?
Over the course of the movement, the soprano passes through a series of
slowly shifting harmonies or spectra, each consisting of approximately eight
pitches. This preliminary observation, at least, accords with the remarks of
other analysts. Bradshaw talks of seven-note chords that gradually fade into
one another; Deliège comments on harmonies with variable numbers of
notes, mostly with seven or eight pitches.19 However, it is difficult to recreate
any clear correspondence between these harmonic spectra and the sonnet
verses: the shifting of spectra occurs independently of the lines in the poem.
Both Bradshaw and Deliège also propose a structural correspondence
between the harmonic material in the soprano and the instrumental parts.
However, they fail to mention that there is a retrograde logic to the complete
pitch field. Regardless of octave register, the end of the pitch field turns out to
be the equivalent of the beginning in retrograde form (Example 8.2).
The final section, with verses 14, 13, 12, 11 and 10, is a fairly accurate
retrograde presentation of the opening section, with verses 1, 2, 3, 4, 5, 6
and half of 7 – up to and including the first F# of this verse (by the term
‘verse’ I refer to a ‘line’ in the poem). Verses 9, 8 and the rest of 7 are not part
of this retrograde pattern. Given that verses 10–14 derive from the begin-
ning, then, verses 1–9 actually contain the bulk of the content. So how were
these first nine verses generated?
Significantly, the sonnet verses, with their beginning and ending points,
escape the retrograde logic of the movement as a whole.20 This may suggest
19
See Bradshaw, ‘Instrumental and Vocal’, the whole of verse 2 and the opening of verse
and Deliège, ‘The Convergence of Two Poetic 3. Verse 12 takes the rest of verse 3 and the
Systems’, pp. 99–125. opening of verse 4. Verse 11 uses the rest of
20
That is, verse 14 is a retrograde of the verse 4, the whole of verse 5 and the opening
opening of verse 1, but not the whole verse. of verse 6. And verse 10 is a retrograde of the
Verse 13 is a retrograde of the rest of verse 1, better part of verse 6 and half of verse 7.
that Boulez organised pitch before forming any relationship between the
verses and the text. In the same way, the application of the retrograde
principle obviously anticipates the seven- or eight-note spectra, which are
therefore themselves not constitutive or generative but instead a result of
some other generative logic. To explore what that logic might be, we must
turn to the composer’s sketches.
The sketches for Pli selon pli comprise some 750 pages (137: 56–803).21
In addition, among a thousand sketches in Boulez’s almost microscopic
handwriting, are several hundred pages relating to other works that may
be of relevance. The Pli selon pli sketches fall into four broad categories: idea
sketches (verbal notes, graphic figures and formal plans); material-
generation sketches (series, tables, pitch diagrams, rhythmic diagrams, mul-
tiplications and other generative operations); score drafts; and final copies.
Most of the sketches are concerned with the serial generation of material.
Surprisingly, on the first pages of the sketches for ‘Improvisation I’, the
vocal line appears in its fully completed form (137: 248ff). Its generation,
then, cannot be found in the sketches directly pertaining to this piece, so
we must look elsewhere. In the same year, 1957, Boulez wrote a large-scale
work for flute solo, Strophes, which apparently remained incomplete and
was later withdrawn from his catalogue of works. About fifty pages of
sketches relate to this work (137: 1–55). Not far into them, we find a page
containing three large fields of pitches (137: 10). These fields, together
comprising almost five hundred notes, appear to constitute an outline of
the pitch material used in the composition of Strophes. This is confirmed
upon comparing this material with the finished flute score. Of greatest
interest to us now, however, are the remarks in the margin, entered with
a different pen and probably at a later point in time. Here, we find the
words 1ère Impr., 3ème Impr. and 2ème Impr., together with a number of
other comments. The pitch fields are labelled as bc, ac and ab respectively,
in the following order:
‘Impr. I’– bc
‘Impr. III’– ac
‘Impr. II’– ab
Next to the bc field, under the wording 1ère Impr., is the expression Idéogr.
total. It would appear, then, that this is some kind of ‘total ideogram’ for the
first Mallarmé improvisation, which proves to be the case: the first pitch field
21
PSS, film 137, pp. 56–803. In the text and number (i.e. the number of each picture on
notes, the sketches are referenced with the the microfilm) following the colon.
film number before the colon and the page
from Strophes, henceforth called bc, is in fact virtually identical to the vocal
part in ‘Improvisation I’, albeit notated in the flute register, with a larger
tessitura and, on occasion, in an extremely high register.
The original sketches on paper (in contrast to the black and white
microfilm) show the additional markings in different colours (red, blue
and green), while the notes themselves are written in pencil. A small
table in the margin containing the same characters reveals that they
refer to the sonnet form of the poem ‘Le vierge, le vivace et le bel
aujourd’hui’. In the table, the stanzas are numbered from I to IV.
The lines are numbered from alpha to delta (in the sonnet’s quatrains),
and from alpha to gamma (in the tercets). The Roman numerals,
together with the Greek lettering, are also noted on the Strophes manu-
script, apparently to indicate how each individual verse extends over
the pitch field. The sonnet text has therefore been matched with the
unfolding pitch field of the early flute sketch at a later point in time.
The other two pitch fields, ab and ac, are related to, but do not directly
conform to, the vocal parts in ‘Improvisations II and III’.
Notably, field bc (137: 10) is divided into nine sections that correspond to
the phrase divisions in Strophes. Example 8.3 features my transcription of
pitch field bc from Strophes.
Ex. 8.3 Pitch field bc from Strophes

At first, the soprano progresses through the entire bc field from begin-
ning to end, which on the whole contains the pitches used for verses 1–9
of ‘Improvisation I’, inclusive of grace notes and principal notes.
The voice then jumps to the middle of section 7, to the point marked
with an arrow by the composer (included in Example 8.3), then moves in
retrograde from there back to the beginning, to produce the material for
verses 10–14.22
The vocal part is not always entirely true to the bc field. Notes are added
here and there, perhaps to furnish the syllables in the text, or for other
reasons. The bc field shows nothing more than abstract pitches, irrespec-
tive of register, duration or contrast between grace notes and principal
notes. There is no reference to dynamics, phrasing or arrangement of the
text, not to mention the musical interpretation of the poem’s words
(swan, winter, white light, being trapped in the ice, and so on). Any
such interpretation of these further contextual roles for pitches and
intervals can find no purchase in the actual generative procedures.
To reiterate, the entire vocal part of ‘Improvisation I’ is contained in
these flute sketches, in the form of pitch field bc, and the vocal material of
‘Improvisations II and III’ is related to fields ab and ac. The further
analytical question, then, involves how the pitch fields bc, ac and ab
came into being.
Towards the end of the final draft of the flute piece (137: 12–18) we
encounter the three pitch fields bc, ab and ac again, all in their entirety
but now in retrograde (137: 15–17). In addition, they are now split
up by way of a series of interjected, melismatic, virtuosic flute figures
which are not found in the pitch fields (137: 10) or in the vocal part.
As mentioned, the bc field encompasses nine sections – a division that
has apparently been carried out according to the phrasing in the flute
piece, not the vocal part. Nevertheless, these nine sections later become
decisive for the direction and retrograding of the pitch field of the vocal
part. The sections occur in this order: 12345–6789–54321. As it turns
out, our earlier enquiry into the use of retrograde and the rigidity of its
implementation in the vocal part was far too conscientious. Here, the
composer simply transferred arbitrary divisions between sections that
happened to be in the flute piece, irrelevant to the vocal part, and then
22
The following remark is found on a page of clear: numbers 1–9 are not sonnet verses but
sketches from a completely different context; the nine sections into which the flute pitch
it provides general plans and outlines for Pli field is divided, and the layout shows the
selon pli as a whole (137: 48): ‘1ème Impr.– progression of the soprano part through
bc–1 à 9, puis 5 à 1’; and on a new line: ‘12345 these flute piece sections.
6789 54321’. These notes have now become
continued to build upon them during the structuring of the material for
the soprano.
The flute piece in its entirety is divided into sixteen large formal
sections that have the following designations in the score draft:
1a–2a–3a–1b–1c–2b–3b–1d–1e–2c–3c–1f–1g–2d–3d–1h. They are named
according to the following pattern (exposing a local, systematic logic that is
typical of Boulez):
1a–2a–3a–1b
1c–2b–3b–1d
1e–2c–3c–1f
1g–2d–3d–1h
These sixteen sections can be distinguished by their varying characters and
tempi in the flute piece. The sections also grow gradually longer and more
complex. Pitch fields bc, ab and ac are all used in the final section 1h of the
flute work, which is decidedly the longest and most complex of all sixteen.
The first, short sections (1a, 2a, 3a and so on) are straightforward and look to
have a classic dodecaphonic structure; they differ greatly in style and struc-
ture from section 1h, the bc pitch field and the ensuing vocal part. If we can
recreate the generative process of these sixteen sections, then, we may
uncover exactly how the pitch fields were formed.
Unfortunately, the remainder of the sketches for Strophes fail to indi-
cate how the sections were created; in fact, the structures are not generated
in the Strophes sketches at all. However, the names ‘Strophes 1a’, ‘2a’, ‘3a’,
‘1b’, and so on also feature in the manuscripts to another work by Boulez
from this period: his theatre music for the drama L’Orestie.23 The drama
was put on at the Théâtre Marigny by the Renaud-Barrault theatre com-
pany in 1955, where Boulez was musical director at the time, to effusive
critical acclaim.24 Boulez composed this music hurriedly and later dis-
avowed it. Still, it has since provided a reservoir of material for later works
(Strophes, ‘Improvisations I–III’ and ‘Don’ (piano version, 1960), among
others).
The surviving sketches for L’Orestie are quite thorough, albeit
incomplete.25 In contrast to the sketches discussed thus far, they do contain
complete twelve-tone rows, and we are thereby – perhaps – closer to
unearthing the structural origins of an extensive well of musical material.
The theatre music can be traced back to the following row table in one of the
first pages of the Orestie sketches: a standard row matrix containing eleven
twelve-tone series, labelled alphabetically from A to K (Example 8.4):26
23 26
See O’Hagan, ‘Project of L’Orestie’. 136: 759. Similar connections can also be
24
See Goléa, Rencontres, pp. 19–20. found in Éclat: see Piencikowski, ‘Assez lent,
25
PSS, 136: 717–1,027. suspendu’.
Ex. 8.4 Row table from L’Orestie
This table is generated from the first twelve-note series (A); a certain rotation
technique produces the other rows. Each new row is moved one position to
the left and transposed up a minor second (as indicated by the bubbles in
Example 8.4). The new concluding note is (if necessary) changed so that the
interval between the first and the last note in the series is always a semitone.
This intervention means that one or more notes in each series is subject to
repetition, and repetitions accumulate further down in the matrix.
Consequently, each series has twelve positions and nine, ten, eleven or
twelve different pitches, and the technique thus generates a gradual trans-
formation of the row structure throughout the table.
I would like to emphasise that this technique is a relatively rigid
mechanism which does not allow the end result to be anticipated before
the process has actually been carried out. Nor, interestingly, does it
guarantee a unified result; instead, it produces a displacement and dis-
tribution of fixed differences between series. The emerging material
always varies in relationship to the inductive series. Mallarmé’s notion
of mobility (mobilité), much discussed by Boulez, is pertinent here, as is
Derrida’s dissémination and neologism différance, which is understood as
both displacement and difference and thereby offers a striking metaphor
for the workings of Boulez’s serial modes of writing.27 The interplay
between similarity and difference, rigidity and unpredictability, in the
actual production of the row matrix, is very telling. Through it, we see
how Boulez’s suspension of simple notions of unity and control play their
part in his contrivance of serial techniques.
So, what might be the connection between this original twelve-tone row and
section 1h of Strophes, and also the bc, ab and ac fields from the flute piece?
Only certain elements in short passages of section 1h are recognisable as having
originated from the row matrix, and the long, cleaved sequences of section 1h,
though related to pitch fields bc, ab and ac, cannot be directly reproduced by
means of the row matrix. Instead, we arrive at this connection via the unex-
pected transformation of a certain section of other material from elsewhere in
L’Orestie.
‘ . . . of practically canonic writing’

Next to the row matrix A–K, on the same page of sketches (136: 759), there is
a complicated set of numbers and letters that perhaps indicates how the
material from a certain section of the theatre music is to be used. We can
deduce from the instrument codes in the table (for flute, English horn and
harp, plus vibraphone and xylophone) and the abbreviations OR and EL that
this section is a lengthy dialogue between Orestes and Elektra in the theatre
piece. The table-like presentation of letters A–K could be the key to how the
rows are used in the dialogue. The plan is also laced with cryptic hints about
27
See Derrida, L’écriture, and La would appear that Boulez and Derrida were
dissémination. Derrida’s reading of Mallarmé subject to similar kinds of inspiration. See
in La dissémination (‘La double séance’) was Guldbrandsen, ‘Boulez och Mallarmé’. See
published fifteen years after Boulez’s com- also Campbell, Boulez, Music and Philosophy.
position of ‘Improvisation I and II’, though it
the separation and commingling of the note groups in the series (séparés
or mélangés), together with a number of arrows.
In the relevant part of the score for L’Orestie, during the aforementioned
dialogue, an inscription (added in a different pen) gives the names of all
fifteen initial formal sections of Strophes, from sections 1a to 3d.
The different sections of the music set to the dialogue must in one way or
another correspond to the various sections of the ensuing flute work. But the
final section, Strophes 1h, which is the most interesting with regard to
‘Improvisation I’, is not specified in this way. At the place in question in
the L’Orestie score, there is instead a lament by Elektra.
This section, featuring the flute, English horn and harp playing in three-
part polyphony, was later outlined and crossed out with a heavy black
pen – singled out, but then seemingly discarded. However, in the margin
there is a collection of symbols, most probably entered later, consisting of
the letters a, b, c and A, B, C in various combinations, displayed with
connecting circles or bubbles, arrows and crossing lines. Interestingly,
these exact symbols can be found on a sketch page in the material for Pli
selon pli relating to the work’s main plan (138: 48). Thus the outlined,
short section of ‘Elektra’s lament’ is apparently included in Pli selon pli in
some form or other. We appear to be on the right track. Moreover, the
crossed-out flute part in L’Orestie is labelled A, the English horn part B,
and the harp part C, suggesting a connection between the letters in the
margin and the pitch fields that are now named BC, AB and AC. However,
none of these parts corresponds whatsoever to the pitch fields or vocal
parts in ‘Improvisations I–III’! The plot thickens: we are as lost as Elektra
when she says, ‘But where can we find the words, the words that mean
something?’28
During this long analytical process, as I sought previously undetected
clues, I frequently returned to Boulez’s theoretical texts. In Boulez on Music
Today I had long pondered the following passage, not least because it
accompanies a music example showing the first verse of ‘Improvisation I’.
Boulez writes: ‘Starting from an extreme rigidity of conception – of practi-
cally canonic writing – a suppleness of realization is reached which can easily
be mistaken for a flexible improvisation.’29 This is a cryptic passage, which
took me a long time to unravel. In short, it turns out that Boulez took one
page from the finished score of L’Orestie, chose the three parts for flute,
English horn and harp and submitted them to a completely new procedure
of transformation. Observe them in Example 8.5.
28 29
‘Mais où trouver des mots, des paroles qui Boulez on Music Today, pp. 137–8 (Penser
vaillent?’ (sketches for L’Orestie). la musique, p. 160).
Ex. 8.5 Instrumental parts from L’Orestie superposed over ‘retrograde canon’
Detailed analysis ultimately confirms that (part of) the soprano voice in
‘Improvisations I–III’ is actually generated from the three instrumental parts
in ‘Elektra’s lament’. This is done in the following astonishing – not to say
eccentric – fashion:30 to begin with, the voices (flute, English horn and harp,
or A, B and C) are in retrograde. They are then combined, two by two, in
a form of retrograde ‘canon’ that generates three new systems (ac, ab and bc).
Matters are further complicated by the fact that it is the endings of the
durations, not the beginnings, that dictate the order. (As a consequence,
notes with longer durations may end up after notes with shorter durations
on the timeline although their onset may in fact be earlier.) Boulez discusses
this very idea at an earlier point in Boulez on Music Today,31 in a description
30 31
See Boulez’s reference to the significance of See Examples 55a and b, p. 136 (Penser la
la folie utile (‘useful madness’) in the musique, p. 158). These two examples are
compositional process: Boulez, Stocktakings, actually extracted from his later generation of
p. 30 (Relevés, p. 46). instrumental material for the movement
of a ‘false homophony, which is really the reduction of a polyphony’.32 In the

present case, in contrast to the main notes, the grace notes do not extend
over time and are thereby themselves subject to a new set of circumstances –
note endings. This combinational logic therefore introduces interchanges in
note order. In addition, the composer has taken the liberty of shuffling the
order of certain notes and freely adding occasional note repetitions due to
other considerations.
Example 8.5 features my reproduction of the retrograde ‘canon’; the
three instrument parts from the specified location in L’Orestie are here
presented together. The top system shows the three voices (flute, English
horn and harp), with durations notated proportionally. The middle sys-
tem shows the actual durations, with their endings. The lower system
shows the resulting combinations of the voices, backwards and two by
two, resulting in the pitch fields bc, ab and ac. Now, from this procedure,
three new sets of pitches emerge that roughly correspond to the three
pitch fields in reverse order.
Again, we see that the bc field is used with remarkable flexibility to mark
out the vocal line of ‘Improvisation I’. With further additional changes and
manipulations, the ab and ac fields are partly used for the vocal melodies in
‘Improvisations II and III’.
In Example 8.6, in the system labeled bc, one can see the ‘raw pitches’ for
the beginning of the vocal line of ‘Improvisation I’, ‘Le vierge, le vivace et le
bel aujourd’hui’, read from right to left (D – C – A flat – G – E flat – B – A flat,
etc.). When we compare this to the finished shape of verse 1 (see
Example 8.1), we can see how Boulez handles these pitches freely and adds
occasional note repetitions at will.
Surprisingly, we must concede that, in all three movements, the vocal
part (which consists, again, of pre-generated pitch relationships) is quite
arbitrarily set to the sonnet texts. The pitches generated by the ‘retrograde
canon’ mechanism are not produced with the particular words in mind
but arrived at through isolated and independent procedures, using mate-
rial taken from completely different contexts. The pitch structure is
generated for its own sake and can have nothing to do with the poem’s
syntax, rhyme patterns, phonemes, sonnet structure or word meanings.
In his later conversations with Deliège, Boulez tellingly describes the
text–music relation in Pli selon pli as both a ‘complete osmosis’ and
a ‘complete transformation’ of the text.33 Nevertheless, Deliège and
32
‘Improvisation III’ (appearing in the sketch Boulez on Music Today, p. 137 (Penser la
folder called ‘sectionnements polyvalents’, musique, p. 159).
33
later to be notated in durations multiplied by Boulez, Par volonté, p. 124.
four in the score). Boulez does not identify
the origins of the examples in his book.
Ex. 8.6 Retrograde ‘canon’, ending
a number of other analysts proceeded to interpret the melodic lines and

interval patterns in the soprano part according to poetic associations
within the ‘swan sonnet’.34 As we have seen, however, the pitch structures
were actually produced in a manner that in effect excluded the possibility
of meaningful relations in this sense at the generative level.
Still, the process of aesthetic signification of the music does not end here.
In fact, these findings open up a different type of question: what is the
relevance of these generative techniques to the interpretation of the musical
artwork? Is the experience of musical, stylistic, poetic or aesthetic meaning
affected by the arbitrariness of generative techniques? To play on
Schoenberg’s old distinction between ‘how it was made’ and ‘what it is’ (in
34
See Deliège, ‘Convergence’ and Bradshaw,
‘Instrumental and Vocal’.
dodecaphonic music), how might we bridge the ‘how’ and the ‘what’ (and, by
extension, the ‘why’)?35
First of all, it is certainly the case that further technical transforma-
tions and revisions concerning phrasing and stylistic surface did take
place as the vocal part was inscribed into the instrumental contexts of
the ensuing musical scores. The result was then incorporated into the
large-scale work Pli selon pli. The function of the vocal part in
‘Improvisation I’ was itself revised when the work was developed for
a larger chamber orchestra (1962) – whereby harmonic aspects were
meticulously mutated (that is, expanded through his technique of multi-
plication of chords) for the orchestra (see sketches 137: 248ff).
In ‘Improvisation II’, the vocal part was reinforced by a kaleidoscopic
flourishing of the instrumental textures in that movement’s vast number
of brief formal sections with different tempi and different modes of
singing. This movement reuses material that was originally generated
for ‘Séquence’, the last formant of his Troisième Sonate (1955–7).
In ‘Improvisation III’, the rather modest central material, taken from
many different sources, was transformed into a fantastically rich move-
ment of more than twenty minutes (1982). Parts of the three
‘Improvisations’ are also quoted in ‘Don’, which was subject to revision
until 1989. The process of transformation, then, continues.
‘The unpredictable’ becoming Necessary

What can be said about these conclusions? What remains of the rationalist,
‘unity and control’ model of serialism? It is indeed a long journey from the
initial twelve-tone series in the row matrix for L’Orestie, via the three
instrumental parts, the ‘retrograde canon’, the regrouping of pitches into
the bc, ab and ac fields, their new sectioning in Strophes, to their new
transformations in the vocal lines of the three ‘Improvisations’. From start
to finish, we follow a long chain of generative stages with drastic methodo-
logical leaps and disjunctions between them. The function of all of these
techniques is to generate ‘raw structures’ that can later be freely moulded,
discarded or musically articulated. The logic of generation is carried through
almost to the point of absurdity, and the alleged origin is literally twisted into
oblivion.36 Any notion, then, that these different techniques are applied to
35 36
See Boulez’s statement in his conversation This point was emphatically supported by
with Philippe Albèra: ‘C’est pourquoi je Boulez in our conversation in 1996. See
pratique l’analyse sceptique. Car je sais que je Guldbrandsen, ‘Pierre Boulez in Interview,
ne découvrirai pas le pourquoi.’ Albèra, 1996’, Tempo 65/256 (April 2011), ‘Part II:
Entretien et études, p. 10. Serialism Revisited’, p. 23.
guarantee a structural unity or coherence from beginning to end would seem

strangely misplaced or irrelevant, indeed untenable.
But what about the control aspect of serialism? From the outset, there is
no structural control over the final result, because there is no possible means
to fathom its textures when starting a compositional process that is this
complex. In fact, the structural result is radically and fundamentally unfore-
seeable; the possibilities of the opening material appear to arise, in effect, in
the very process of writing.
Is this something that Boulez only admits later in his life, or had he
actually communicated the structural unpredictability of his intricate
methods of generation early on? As a matter of fact, he underlined the
unforeseeable aspect of his serial composition as early as 1952. In his
article ‘Possibly . . .’ (‘Eventuellement . . .’), he writes: ‘From the prescrip-
tions we have been examining in detail arises the unforeseen.’37 In the
article ‘. . . Near and Far’ (‘. . . auprès et au loin’) from 1954, he writes: ‘Let
me recall our original determination to consider the series, not as an ultra-
theme, permanently tied to pitch, but as a generative function of all aspects
of the work.’38 And in ‘Alea’, from 1957, he states: ‘In my experience it is
impossible to foresee all the meanders and virtualities in the material with
which one starts.’39 These remarks draw fresh attention to particular blind
spots in his traditional analytical and historiographical reception.
If Boulez has been, at best, ambivalent in his rhetoric, widespread
structural analyses of his music have been, at worst, methodologically one-
eyed. To be sure, Boulez’s many sources of inspiration – from literature
and the arts, aesthetics, listening to earlier composers and rehearsing and
conducting their scores together with musicians and ensembles – have
been frequently though loosely mentioned, but their concrete impact has
rarely been positioned right at the heart of his compositional method.
Loyal to the hegemony of ‘formalist’ structural analysis, and fettered
by a restricted picture of serialism, itself branded with the notions of
‘rationalism’, ‘coherence’ and ‘control’, too many scholars have been
unable to discern what was already present in Boulez’s writings and
music. Pascal Decroupet elegantly crosses these borders in his discussion
from 2003 of Boulez’s musical thinking around 1960.40 (As mentioned,
there are other exceptions too; see footnote 5.) Boulez distinguishes
between the production of his material and its mise-en-place (see Boulez
on Music Today), and the perception of the latter, that is, of musical form,
cannot be reduced to criteria of serial production.
37 39
Boulez, Stocktakings, p. 133; Relevés, p. 174: Boulez, Stocktakings, p. 29; Relevés, p. 45.
40
‘À partir des données que nous avons Decroupet, ‘Comment Boulez pense’,
étudiées en détail, l’imprévisible surgit.’ pp. 49–58.
38
Boulez, Stocktakings, p. 149; Relevés, p. 193.
In summary, in the generative processes of the vocal line in

‘Improvisation I’, we do not find continuous structural necessity and
logical unity but rather an entirely new generative technique that has
been more or less capriciously reinvented at each additional stage. This
lack of unity actually goes for the instrumental textures as well, which have
generically nothing to do with the vocal line.
The first and third interlude, and the second and fourth (the fourth one
being the piece’s postlude) in ‘Improvisation I’, are reorchestrations of
Boulez’s small piano pieces, Douze Notations, nos. 5 and 9 respectively, from
1945.41 These pieces were written in a conventional dodecaphonic style and
were absent from his work catalogue at the time when he reused them in
‘Improvisation I’ in 1957, an observation noted by Hirsbrunner as early as
1986.42 Far from generating a structurally unified piece, Boulez presents
us with a musical montage, though one that has been moulded into
a stylistically coherent result. And in fact, he has applied similar procedures
a number of other times as well. In 1957, when these operations were actually
carried out, the revelation of such processes might very well have produced
considerable astonishment, not to say a minor scandal, on the part of the
self-proclaimed ‘scientific’ attitude to strict composition in Darmstadt and
certain milieus in Paris. Be that as it may, Boulez kept quiet. Forty years later,
however, when questioned on these issues in 1996, he laconically replied:
‘Well, I also did this kind of montage in other pieces . . . I organize this as
a jigsaw puzzle, and afterwards you cannot imagine that the various parts
were composed in totally different ways.’43
This brings us to a brief discussion, reflecting on music analysis and the
kind of unity that scholars demand from it, equipped as they are with
methods of structural analysis established in the mid-twentieth century.
As Joseph Kerman famously proclaims in his methodological rethinking
of analysis published in 1980, too much effort has been invested in
demonstrating ‘organic unity’ at a structural, compositional level. He
finds, instead, that a musically meaningful unity is something that
emerges in the play among those structures on the musical surface –
play that must be sought in the interaction between structural facts
and aesthetic experience.44 This entails that the structural facts that are
established through analysis must be made relevant at the aesthetic levels
of musicological interpretation, performance and listening (in a form of
analysis that Kerman terms ‘criticism’).
41 43
Boulez, Douze Notations pour Piano, 1945 Guldbrandsen, ‘Pierre Boulez in Interview,
(Universal Edition, 1985). 1996’, Tempo 65/256 (April 2011), ‘Part II:
42
Hirsbrunner, ‘Notations’. Serialism Revisited’, p. 24.
44
Kerman, ‘How We Got’.
To construct (as a composer) or reconstruct (as an analyst) a generative,

serial or structural coherence among the technical elements of a score does
not in itself guarantee a musically unified – or musically satisfying – result.
And conversely, a structurally heterogeneous material does not at all
exclude the possibility of experiencing a musically coherent and mean-
ingful result. Clearly, there is no automatic or causal connection between
these different levels. The work’s ultimate musical unity is something that
emerges in the final articulation of the musical surfaces, where aesthetic
judgement and musical listening become seminal factors. This is precisely
what comes across in the case of Boulez. He is first and foremost a practical
musician, and from the raw structures he takes what he wants and moulds
it according to his own musical taste.
Musical Phrasing and Articulation

At this point in my analytical research in the early 1990s, having discovered
the structural disjunctions in some of Boulez’s procedures, I began working
on questions of phrasing and form, particularly in the huge movement
‘Improvisation III’, and I studied with interest the revisions that Boulez
made to the movement between 1959 and 1983. In Pli selon pli as a whole,
a vast multitude of transformation techniques are developed and put into
play (jeu), in the sense discussed above. This dimension of play is no less
characteristic of Boulez’s serialism than are the initial ideas of logical unity
and rational control.45 The final articulation of the material is what turns
the raw structures into real music, or into what Stravinsky, according to
anecdote – and with Boulez in mind – once characterised as ‘Webern music
sounding like Debussy’.
This brings us to the following question: once the ‘raw material’ of serial
structures has been generated, how does Boulez actually proceed in his
further articulation of the music? How does he work on the musical
surface, including phrasing and articulation, rhythm and allure, textures
and timbres, musical gestures, formal processes, stylistic choices and text
setting? On the whole, his compositional sketches tell us very little about this,
but there is at least one instance where they are particularly revealing. In the
vocal line of ‘Improvisation II’, ‘Une dentelle s’abolit’, the opening verse
appears in the following way (Example 8.7):46
45
See, for instance, Boulez on Music Today, structures in relation to each other’ [original
p. 99: ‘we must return to the play of serial italics]. (Penser la musique, p. 113.)
46
Boulez, ‘Improvisation II’.
Ex. 8.7 Verse 1, soprano, ‘Une dentelle s’abolit’
Ex. 8.8a Main notes, verse 1
Ex. 8.8b Grace notes, verse 1, taken from pitch field ab
This vocal phrase actually consists of two separate structures that have
been combined, as happens on numerous occasions in Boulez’s creative
process. The main notes (see Example 8.8a) were written separately,47
and the grace notes (see Example 8.8b) are taken from the note field ab
in the sketches for Strophes. (To do so, he extracts note groups from the
thirteen sections that were already defined in the ab pitch field, taking
them in the following order: 1–13, 2–12, 3–11, and so on, until the
pitch field from Strophes is exhausted. The first eight groups are
shown in Example 8.8b, and I have labelled them with lower-case
letters from a–h.)48
Then Boulez sets out to fuse these two structures into one musical line
with phrasing and text and so forth. This time the sketches, strikingly,
reveal a compositional process of trial and error, which is where my initial
parallel to Beethoven’s sketchbooks arises. In the following musical
examples, which I originally copied by hand from Boulez’s sketches to
47
See 137: 297 and 298, at the top of the groups of grace notes interspersed in the
pages. soprano part in ‘Improvisation II’.
48
The groupings of pitches in the aforemen-
tioned note field ab correspond directly to the
‘Improvisation II’,49 we find no less than eleven different attempts to finish

just the first musical line and set it to the sonnet verse ‘Une dentelle
s’abolit’. On the same large page, there are also several attempts to write
out the subsequent verses of that piece. I shall here concentrate solely on
verse 1; his first attempt is presented as Example 8.9, which can be
compared to the final result in Example 8.7.
Ex. 8.9 Verse 1, soprano, first attempt
Here, the beginning is almost complete, whereas the middle part and
the ending are unfinished. A small part of the text is also scribbled in.
Then Boulez makes three more attempts to rephrase the beginning (see
versions 2, 3 and 4, in Example 8.10). Interestingly, the two-note group
I have labelled [x] cannot be found in the ab pitch field:50
Ex. 8.10 Verse 1, soprano, second, third and fourth attempts
It is obvious from these examples that Boulez is working on musical phrasing

and articulation not relating to ‘serial’ techniques. He appears to have
exerted a great amount of effort in shaping his opening phrase. On his
fifth attempt (Example 8.11), the grace note [a] has been displaced and the
[x] group has been inverted.
49
137: 298. discussion. I have here indicated a certain
50
The order of his different attempts, scat- amount of his crossed-out notes, but not all
tered on the huge sketch page, can surely be of them, since some of his scratching has
problematised. Also, the mere deciphering of been nearly impossible for me to decipher.
Boulez’s handwriting is always a matter of
Ex. 8.11 Verse 1, soprano, fifth attempt
The next attempt is quite interesting: here he attempts to rethink the melodic
contour of the opening (Example 8.12):
Ex. 8.12 Verse 1, soprano, sixth attempt
Next, the low F-sharp in the opening is retained and then crossed out, and
the ending of the verse is lifted to a very high register (Example 8.13):
Ex. 8.13 Verse 1, soprano, seventh attempt
Preoccupied, as it were, with the C-sharp from the ab field and the main, final
A in a high register, he struggles with the phrase ending before making a new
attempt using a different approach. Here, the metrical notation is rethought
completely, the distinction between main notes and grace notes is tempora-
rily blurred and the ending is left suspended (Example 8.14):
Ex. 8.14 Verse 1, soprano, eighth attempt

In a concerted effort, he then drafts a complete version of the verse with the
full text as his ninth attempt (Example 8.15), only to erase most of it in the
end. He also inserts the numbers ‘1 et 13’ at the beginning, referring to the
groups he is extracting from the Strophes field:
Ex. 8.15 Verse 1, soprano, ninth attempt
Next, he focuses solely on the ending, which still ascends to a very high
register (Example 8.16):
Ex. 8.16 Verse 1, soprano, tenth attempt
Finally, with a ‘Copernican turn’, Boulez manages to rearrange the pattern to

achieve a more acceptable ending. The final gesture is lowered to a more
suitable register while retaining the rising third on the final syllable, ‘-lit’
(eleventh attempt, Example 8.17):
Ex. 8.17 Verse 1, soprano, eleventh attempt
Combining this ending (Example 8.17) with the opening from Example 8.13
and cleaning up some of the details, he finally completes the first soprano
verse.
A more extensive presentation of the material from Boulez’s sketches
would validate these claims to a greater degree, but the implications of my
findings should be clear: the composer is working on musical phrasing and
articulation not according to serial systems, in the end, but in the interests of
a satisfying musical result. The following question remains: upon what
criteria does he rely?
Free Aesthetic Choice

Obviously, Boulez’s aesthetic choices are not formalised in any kind of
technical system, they are largely a result of his judgment – his conception
of style, embedded in his musical taste. In the case of Beethoven’s sketches,
Maynard Solomon writes: ‘With Beethoven, not only is there no prospec-
tive inevitability, there may even be no inevitability after the fact. His
sketches and autographs may well be a series of rough maps to the multi-
plicity of universes he glimpsed, to a plurality of possibilities.’51 Thus, the
dictates of structural coherence or teleological necessity over the entire
compositional process are not evident in the study of Beethoven’s
sketches. Likewise, in Pli selon pli, what seems to be at stake is Boulez
the musician’s spontaneous feel for what may work well for a soprano in
this specific context.52 In a conversation with Boulez in 1996 he was asked:
‘What are the criteria for your choices when you work like that? [On the
vocal line of ‘Improvisation II’.] You are beyond the realm of serial
procedures here?’ Boulez answered: ‘Oh yes, that’s completely beyond
seriality. The criterion is simply the meaning of the line. The phrasing.
The relationship between the intervals. What they mean musically. In this
case the range of the voice. And – that’s it.’53
As we have seen from the generative process, his unpredictable serial
structures might just as easily have turned out differently, and, conse-
quently, several ‘final solutions’ [sic] are conceivable in each case.
The phrases – or let us call them the final musical ideas – clearly transgress
the limitations of the generative system that produced them. To be sure,
the terms ‘system’ and ‘idea’ are Boulez’s own. In his 1986 article
‘The System and the Idea’ (‘Le système et l’idée’),54 he writes that the
system of generative procedures is nothing more than a crutch – an
incentive for the imagination to get started.55 By this account, he draws
upon serialist writing for the raw material of structural objects, then
selects his music from these objects in the next stage of composition.
‘I choose, therefore I am’, he writes, adapting Descartes’s famous ‘Je
51
Solomon, ‘Beethoven’s Ninth’, p. 292. Interview, 1996’, Tempo 65/257 (July 2011),
52
Not everybody agrees that there are ‘Part III: Mallarmé, Musical Form and
dimensions of Boulez’s music that are not Articulation’, p. 18.
53
regulated by the system. See Losada, Guldbrandsen, ibid.
54
‘Complex Multiplication’. Even though one First printed in InHarmoniques 1,
cannot rule out the theoretical possibility that pp. 62–104; reprinted in Boulez, Jalons (pour
somebody may find other evidence in the une décennie), pp. 316–90; and again in
future, nobody has this far been able to prove Leçons de musique, pp. 339–420.
55
that Boulez follows systematic rules ‘Cela revient à considérer le système
throughout all his compositional choices. comme une aide, une béquille, un excitant
What is more, he explicitly states the opposite pour l’imagination.’ Boulez, Jalons, p. 378;
himself. See Guldbrandsen: ‘Pierre Boulez in Leçons de musique, p. 407.
pense, donc je suis’. And what does he choose? ‘I choose’, says Boulez,
‘what I judge to be good, beautiful, necessary’.56
Until recently, these free, aesthetic choices in Boulez’s compositional
practice seem largely to have been misrepresented in the main body of
structural analyses of his work. Again, this general analytical bias has con-
tributed to the historiographical conviction that European post-war mod-
ernism was somehow breaking with the Western, classical-romantic
tradition by inventing a fundamentally new way of composing – or con-
structing – music. Yet free, aesthetic choices are clearly part of what makes
this music work.
The dimension of phrasing and articulation, and the subsequent pro-
duction of a more elastic formal continuity, also point towards Boulez’s
increasing affinity with composers such as Berg, Debussy and Wagner in
his compositional practice. This point applies especially to the telling
revisions of ‘Improvisation III’ up to 1983 and beyond.57 In addition,
this affinity resonates with his increasingly important practical experience
as an orchestral conductor – not least of the great Austro-German
repertoire from Wagner to Mahler to Berg – during his conducting career
from the 1960s onwards.58 Musical ideas do not spring directly from the
composer’s imagination but are produced through writing, whether
structurally or freely. Then they are moulded by the composer via his
search for what is ‘good, beautiful, necessary’.
In his text ‘Le système et l’idée’, Boulez writes that the development of the
work is nothing but a struggle between the system and the idea: ‘The system
and the idea reflect one another as they seesaw between the finite and the
infinite.’59 This notion is not so different from the category of the ‘work’
emerging in early German Romantic aesthetics around 1800.60 This
work does not reveal itself at first glance: it is constituted through repeated
readings, performances and listenings, in ever newer versions. As long as no
single interpretation can capture it as such, the musical work, ontologically
speaking, achieves a kind of virtual existence.61
This dimension of ‘generative writing’ in Boulez’s serialism sheds impor-
tant light on his affinity to Mallarmé’s poetics of literary writing, or écriture.
The deeper kinship between the German Romantic idea of absolute Musik
56 58
Boulez, Jalons, p. 378; Leçons de See more on this in Guldbrandsen,
musique, p. 407: ‘Je choisis, donc je suis; ‘Modernist Composer’.
59
je n’ai inventé le système que pour me ‘Le système et l’idée se renvoient l’un à
fournir un certain type de matériau, à l’autre dans un jeu de bascule entre fini et
moi d’éliminer ou de gauchir ensuite, en infini.’ Boulez, Jalons, p. 379; Leçons de
fonction de ce que je juge bon, beau, musique, p. 408.
60
nécessaire.’ See Goehr, ‘Philosophy of Music’.
57 61
See Guldbrandsen, ‘Playing with See Dahlhaus, Idee der absoluten Musik,
Transformations’. pp. 140ff.
and Mallarmé’s poésie pure (later to be revised by Paul Valéry as poésie

absolue) has been pointed out by many, but it is rarely discussed in greater
technical detail in the growing literature on Mallarmé and Boulez. Mary
Breatnach provides some enlightening readings at an aesthetic level, but she
does not at all set out to connect her readings to hands-on analyses of
Boulez’s compositional writing.62 Hence, the deeper and more concrete
connections between Boulez and Mallarmé’s modes of writing at the level
of poetics remain underserved.
In a certain sense, Mallarmé considered the poem less a fixed result
than a strategy for reading. According to the literary textual theory
(théorie du texte) of Roland Barthes, and especially Derrida (who was
deeply influenced by Mallarmé’s poetics), reading a poem amounts to
rewriting it. This evokes a parallel in the act of interpreting a score by
playing it. Applied to serial composition, there might be a further shift in
perspective from regarding the work as a fixed result to regarding it as
a performative procedure – for playing, for interpretation, and for further
compositional writing.
In Boulez’s case, this dimension of unpredictability is already inherent
in the interplay of serial procedures in the early stages of the composi-
tional process, as I have traced in the generation of the vocal line in
‘Improvisation I’. Boulez’s approach privileges the anonymity of the
author’s voice – as Mallarmé famously put it, ‘giving away the initiative
to the words’, or, in Boulez’s case, to the productive play of the serial
procedures.63 This search for ‘anonymity’ of the author’s voice is in fact
discussed by Boulez around the time he composed Pli selon pli.64 This
goes for the stage of serial generation. But then, in the second stage of the
compositional process, as I have illustrated in my example from
‘Improvisation II’, the composer intervenes and articulates the final
textures according to his musical taste and aesthetic judgement.
Through this specific musical practice, the very categories of ‘work’
and ‘art’ are thrown open in the wake of some radical, and largely
unacknowledged, consequences of the tradition of early German
Romantic aesthetics. In order to grasp the particular ambivalence and
strangeness of Boulez’s serial compositional practice, it may well prove
beneficial to integrate into our musicological understanding certain
aesthetic insights from French literary textual theory, as well as
62
Breatnach, Boulez and Mallarmé. from Mallarmé’s letter to Verlaine of
63
‘L’œuvre pure implique la disparition 16 November 1885, later named
élocutoire du poëte, qui cède l’initiative aux Autobiographie: ‘[le Livre] qui, je crois, sera
mots.’ Mallarmé, Œuvres complètes, p. 366. anonyme, le Texte y parlant de lui-même et
64
See Boulez, ‘Sonate, que me veux-tu?’. sans voix d’auteur’. Mallarmé, Œuvres
Without giving the reference, Boulez quotes complètes, p. 663.
Mallarmé’s poetics of writing, in tandem with thorough technical analyses

of the musical works. In its own peculiar way, Boulez’s post-war moder-
nist music turns out to confirm certain central categories of the Western
musical tradition at the same time that it transcends them, paving the way
for new compositional productivity and musical experiences.
9 Serial Processes, Agency and Improvisation1
Joseph Salem
Recent sketch studies of the music of Pierre Boulez are remarkable for
a number of reasons.2 First and foremost, this body of scholarship provides
a guide for understanding Boulez’s compositional processes by charting
the progression of his sketches from his earliest outlines through to his
published compositions. Second, scholars are contributing to our apprecia-
tion of the interdependence of Boulez’s many works by identifying the
proliferation of musical ties among his compositions, from the use of
common organisational sketches in early works through to the reuse of
motivic or thematic material in later ones. Finally, these studies pave the
way for future investigations, when scholars may at last move beyond the
compulsion to identify serial processes to address broader questions regard-
ing Boulez’s musical intuitions and creative process.
However, as scholarship on Boulez continues to proliferate, it is worth
re-evaluating the goals of manuscript studies of his works in particular,
especially as they relate to the development of a hermeneutics to guide the
interpretation and poetics of Boulez’s music above and beyond its mere
explication.3 Of special consideration are the following questions: what do
sketch studies tell us about the musical affect of Boulez’s works? And how do
sketches help us convert a cerebral understanding of Boulez’s music into an
emotional one? Perhaps one of the most remarkable aspects of Boulez’s
1
I would like to thank Paolo Dal Molin for ‘Moments doubles, figurés en prisme’;
his private discussions about certain topics O’Hagan, ‘Pierre Boulez: “Sonate, que me
that surface in this essay. I would also like to veux-tu?”’; Piencikowski, ‘“Assez lent, sus-
thank Peter O’Hagan, Robert Piencikowski pendu, comme imprévisible”’; and Tissier,
and Ian Quinn for their comments on earlier ‘Variations-Rondeau de Pierre Boulez’.
3
versions of this research. The author also Other recent publications, including
graciously acknowledges the financial and Goldman, The Musical Language of Pierre
institutional support of the Fulbright orga- Boulez and Campbell, Boulez, Music and
nisation and the Paul Sacher Stiftung, both of Philosophy, precede my call in their efforts to
which supported the research on which this explicate the differences between descriptive
chapter is based. and hermeneutic analyses of Boulez’s works;
2
Representative examples include, to name however, while both formulate new models
but a few: Dal Molin, ‘Introduction à la for interpreting the meaning of his works,
famille d’œuvres “. . . explosante-fixe . . .”’; neither is entirely exempt from the con-
Decroupet, Le Marteau sans maître: structive criticism I provide below.
Facsimile of the Draft Score; Decroupet,
221
222 Joseph Salem
rigorous, serial approach is that virtually every note can be accounted for
using some type of serial organisation, according to some concrete sketch
progression. Yet, merely identifying the progressions from one sketch to
another is not sufficient for explaining the underlying musical connections,
even if it is necessary for understanding the gestation of any given work.
Furthermore, concretising the relationship between sketches mistakes caus-
ality – the notion that one sketch comes from or out of another – for agency,
or the idea that every serial process and sketch relationship is intentionally
designed, crafted and applied by Boulez. In short, what is missing from
current sketch studies is not a willingness to engage with Boulez’s composi-
tional techniques, but rather a clear sense of how causality, agency and
improvisation contribute to his creative process.
It is easy enough, when charting the progression from one sketch to
another, to suggest causal links that imply every ‘next’ sketch is the natural
and inevitable result of a previous or ongoing serial process. However, this
should not be a foregone conclusion. In fact, one of the most remarkable
aspects of Boulez’s compositional process is the spontaneity of his adapta-
tions, which continually remind us that each and every sketch has the ability
to redirect the composition. This impression invites the conclusion that very
little is sacred in Boulez’s serial process; the fact that he often continually
adjusts his method to control the final trajectory of the work, often in the
liminal space between consecutive sketches, confirms this judgement.4 Yet,
because his agency is exercised ‘off the page’ as it were, the evidence of
Boulez’s most spontaneous, creative impulses is overshadowed by the towers
of concrete, empirical evidence provided by the sketches themselves. It is
precisely such impulses that cause Boulez’s works to deviate from their
original designs with the espousal or eschewal of greater or fewer dynamic
changes, rhythmic patterns and thematic repetitions (or variations). Such
details are often lost to all but the most critical and observant scholars of his
manuscripts. Thus, despite the tangible, observable evidence provided by his
intricate sketches, Boulez’s compositional process should be advertised as
a remarkably flexible one, especially in regards to the development of his
phraseology and formal architecture in his formative years and his thematic
and melodic developments in later ones.
It is all the more troubling, then, that these simple facts are consistently
trumped by a shared – even dominant – musicological assumption: Boulez’s
compositions are organic, his compositional methods and aesthetics,
4
A tremendous number of examples exist, of same work which greatly reduce its form and
course, as this is a primary aspect of Boulez’s phraseology (e.g. Structures, Deuxième livre,
compositional process. The most obvious Chapitre II; Éclat/Multiples; Figures –
ones occur between large-scale outlines of Doubles – Prismes).
a given work and consequent sketches of the
223 Serial Processes, Agency and Improvisation
organicist.5 Yet, what is perhaps even more limiting than this claim is the
confusion over just what the term ‘organicist’ means when applied to Boulez.
His thoughtful use of the word – long after the polemics of his early
writings – is still laced with a naive ambivalence that belies its rich etymology
in the musicological discourse. In response to the question, ‘your actual
process of composition is your actual process of life?’, Boulez responds, ‘Yes,
like an organism, in which the cells develop in an organic way.’6 Here, the
term and description correspond to the earliest, biological sources of organi-
cism in music and are wedded to the aesthetic ideal of a seed or germ
growing into a living, breathing compositional structure.
Later, in the same interview, Boulez critiques Stockhausen’s ‘formula
idea’:
I would say more than just intuition, organic development is a consequence of

intuition, which is a combination of intuition and order. This kind of dialectic
between a real form and a preconceived idea, when you are developing your musical
ideas with the material, brings us back to what I called before ‘accepting the
accident’. You make room for this accident within the development of the work like
a mutation – that I call organic.7
Suddenly, the concept of organicism is revised. Gone is the sense of

genetic growth from seed to plant graced by the mystical halo of a prime
mover or intelligent creator; in its place, Boulez stresses a process depen-
dent on mutation and adaptation, accident and creative development.
In effect, this is a Darwinian approach to genetics, a perspective that
emphasises generational change, not intelligent design. It is an approach
that puts the power of the idea and its realisation wholly in the hands of the
composer.8
Or so Boulez would have us believe. Comparing Stockhausen’s ‘formula
idea’ to a cupboard that has to be filled, Boulez unfairly implies that the
former’s massive creations (at the time, his operas) may parallel the use of
5 7
While the terms ‘organic’ and ‘organicist’ Ibid., p. 37. Earlier in the text, Boulez also
may not appear frequently in the literature on associates the accidental with the organic:
Boulez, remnants of this aesthetic outlook ‘I need, or work with, a lot of accidents, but
appear in various guises throughout his within a structure that has an overall trajec-
essays. This is particularly true regarding the tory – and that, for me, is the definition of
use of ‘system’ as a (perhaps unintended) what is organic.’ Ibid., p. 25; quoted in
substitute for ‘nature’ as the underlying Whittall, ‘“Unbounded Visions”’, p. 77,
organising force that imbues Boulez’s works which fuelled my own return to this topic.
8
with inherent, but ill-defined, value. See For more on Boulez’s interest in cognitive
Goldman, The Musical Language of Pierre and neurological speculations on the mind
Boulez, and Losada, ‘Complex and creativity of the composer (and not
Multiplication, Structure, and Process’, for without mention of Darwin), see Boulez,
representative examples. Changeux and Manoury, Les neurones
6
Di Pietro, Dialogues with Boulez, p. 36. enchantés. I thank Edward Campbell for
introducing me to this reference.
224 Joseph Salem
primitive and misunderstood Formenlehren by young composers instructed

to merely fill out preconceived formal schemes. In his own words:
For example, you build the cupboard first and then you put the content inside. For
me it’s the opposite: you create the content and you make the cupboard because of
the content and not on the contrary; so that’s a different approach. The more
I thought about it, the less I was ready to conceive of a form (just like that – all by
itself). For me, I discover the form progressively as I go on. So again, it’s an organic
process.9
It seems, then, that Boulez’s use of the word ‘organic’ inverts the traditional
values of the term in German musical aesthetics. Whereas earlier musicol-
ogists prized organic works for their inherent continuity, their motivic
saturation, and, above all, their singular, flawless elaboration as a reflection
of nature, Boulez values an organism for its ability to change, to mutate, to
develop, to grow new appendages, to defy preconceived plans and struc-
tures – in short, a behavioural philosophy that promotes the effects of
‘nurture’ at great expense to ‘nature’. Even more, Boulez mixes the two
together: on the one hand, his organisms are brought to life and animated
by his role as a composer, but on the other, they have a life of their own,
guiding his hand as he watches them develop.
One can assume Boulez was entirely aware of these caveats. Note, for
example, how he creates exactly the same dichotomy between his own brand
of organicism and that of Schoenberg:
I cannot say like Schoenberg that the whole work is in my head like a vision and then
all that is left is to just write it down. I don’t believe that. Again, this theocratic
vision – ‘God created the light –here is the light; God created water and air – here is
water and air’ – [t]his is a God-like view of creativity which I don’t believe at all, since
I myself believe very much in accidents.10
The references succinctly package the aesthetics of traditional organicism.

Schoenberg, the preconception of the autonomous work, the magical, causal,
unstoppable and inexplicable growth from idea to material – all of these
circumscribe a historicised notion that is, for Boulez, dead. In response,
Boulez takes this ‘God-like view’ of creativity and removes the divine
influence, the external agent, and replaces it with ‘accidents’. But these are
not normal accidents – they are Darwin’s accidents, where mutations
become advantages, and where form follows function.
There is an obvious slippage here. Are we to accept Boulez’s updating of
the term ‘organic’ as a natural evolution of the term to complement obvious
innovations in science? Or should we criticise his use of the term as we have
9
Di Pietro, Dialogues with Boulez, p. 35. clarify the meaning of the original; no words
10
Ibid., p. 38. I have altered the original have been changed.
punctuation and added internal quotes to
the uses of it by past theorists, pointing out the facile relationship between
musical causality and the creative process in both models? Either way, the
term ‘organic’ remains problematic, as it blurs the distinction between
privileged musical relationships and haphazard ones without recourse to
a well-defined system of musical principles (or aesthetics). In historical
models, the principles were provided by God and nature, only to become
outmoded as contexts changed and musical systems developed; in
Boulez’s model, they are defined by him, couched in abstract systems
and virtual relationships. As a result, the fundamental differences between
Boulez’s organicism and that of Schoenberg are much less significant
than their similar dependence on a problematic aesthetics based on an
ill-defined value system. We should therefore remain suspicious of using
Boulez’s organicist aesthetics to evaluate his works, choosing instead to
dig deeper and discover the more idiosyncratic aspects of his creative
process and the most significant contributions he has made to music as
a composer – both consequences of context as much as composition. It is
in these areas – his creative process and stylistic contributions, respec-
tively – that we may find the criteria for judging and evaluating his works,
rather than in some tautological aesthetics based on outdated notions of
musical autonomy, motivic saturation, or even formal coherence provided
by his sketches alone.
The irony is that Boulez’s first serial works were often overly dependent
on early formal outlines – this notwithstanding his criticism of
Stockhausen for the same offence. Of course, his creative process has
changed over his long career and has, for the most part, become increas-
ingly close to the generative processes he outlines above. Nonetheless,
during his formative years, Boulez’s conception of musical form was
a central focus in his adoption of serialism, and one that caused him as
many problems as it offered him solutions. Furthermore, it is in these
years that the heart of Boulez’s modern creative process, so dependent on
self-borrowing and additive formal developments, found its musical
inspiration and poetic justification.
In the remainder of this chapter, I discuss a few features of Boulez’s
formative serial works. I focus on the relationship between his early pre-
ference for formal outlines and rigid serial processes and his increasingly
strong desire to make room for more spontaneous or creative compositional
decisions. Key to this transition was Boulez’s slow adoption of self-
borrowing, first through basic revisions, and then through wholesale trans-
position. The terms agency, serial process and improvisation are key to
my discussion, as each suggests a different degree of creative planning
and freedom for the composer: serial processes execute predetermined
operations, improvisation implies the manipulation of pre-existent material,
226 Joseph Salem
and agency suggests the direct intervention of the composer at any or all
stages of the creative process. Ultimately, I conclude that Boulez slowly
developed the creative process he came to describe as ‘organic’ by repla-
cing a reliance on modular, formal schemata with more flexible or additive
forms that require increased intervention (or agency) in the compositional
process.
Integral Serialism and Modular Outlines

Boulez’s turn towards integral serialism began years before the celebrated
première of Structures Ia. Sketches for Livre pour quatuor and Polyphonie
X reveal serial approaches to rhythm and pitch that predate the more
methodical approach taken in the first Structures book, just as Boulez’s
first essays on the topic – including ‘Proposals’ (‘Propositions’, 1948) and
‘Possibly . . .’ (‘Éventuellement . . .’, 1952) – were written before Boulez and
Olivier Messiaen premièred Structures Ia on 4 May 1952.11 The significance
of dating Boulez’s turn to integral serialism stems from the intention with
which it was made: Boulez was openly self-conscious about his changing
compositional techniques, and the effects these changes had on his composi-
tions were significant and lasting. Put another way, when Boulez adopted
integral serialism, he also committed to reinventing the technical (or, in his
words, morphological) basis of his compositional process.
In consequence, Boulez developed new techniques for aligning dispa-
rate rows of durations, pitches, dynamics and articulations at each stage of
composition. Several of these techniques are outlined in his early writings,
in which Boulez (after Messiaen) discussed how one might design dode-
caphonic rows for all sorts of musical parameters. However, it is only in
Boulez’s sketches that one can observe the changing ways in which he
methodically combined the various rows together in his compositions.
It is here that the first fissures in Boulez’s serial method crackle with life,
providing clues as to how his compositional process will develop in later
years.
In particular, Boulez developed a penchant for two types of sketches to
associate rhythm, pitch and one or more other serialised parameters
throughout the compositional process. Elsewhere, I have described these
sketch types as modular-symbolic outlines (m-s outlines) and modular-
temporal tables (m-t tables), although it is important to remember that
11
In fact, ‘Éventuellement . . .’ is mentioned letter no. 28, p. 167, and Piencikowski,
in a letter to Cage as early as May 1951. See ‘“Printemps: Sacre: Strawinsky” (1951–52)’,
Boulez/Cage, Correspondance et documents, p. 91.
Ex. 9.1 A transcription of an early-stage m-s outline for Polyphonie

X (PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1f)
these are my terms and not those of Boulez.12 These outlines situate various
symbols (used to represent dodecaphonic rows of pitch or rhythm material)
along one or more temporal grids or tables. In the case of m-s outlines, the
grids are quite abstract, using symbols to pair multiple series together in
a linear fashion, but without the specificity of individual notes or rhythms.
Thus, in m-s outlines, one can sense the general progression of all the row
materials together, but usually at the rate of twelve (or more) notes per
symbol. M-t tables are more precise, zooming in on smaller portions of
a work to provide a close-up view of individual row combinations. In these
tables, each column usually refers to a specific pitch and rhythm pairing, as
well as a schematic phraseology (often defined using double barlines) within
which the rows will be situated.
Examples 9.1 and 9.2 represent two such grids transcribed from
the manuscripts for Polyphonie X (1951). Here, the outlines remain
quite abstract, avoiding one-to-one correspondences between individual
notes, or even entire rows, and relying on a plethora of additional row
tables for deciphering the symbols themselves. Nonetheless, the
early m-s outlines effectively condense the entire work into a short, sym-
bolic outline, while the later m-t tables for the work expand these symbols
to provide virtually all of the pitch, rhythmic and orchestral information
required for the first pencil draft.
It is easy, in such cases, to work backward from a finished composition,
placing each and every note within its cell on the m-t table, and each row of
the m-t table among the symbols of the m-s outline. Working in the opposite
direction is more mysterious. While the symbols of the m-s outline appear to
account for those of the m-t table (where each row is merely elaborated or
expanded), the reality is that each stage of the compositional processes
remains dependent on a sequence of decisions. This is particularly true of
the transition from the m-t table to the score, where decisions about the
inflection of each row (its shape and register) and the distribution of rows
12
See esp. Salem, ‘Boulez Revised’.
Ex. 9.2 A transcription of a late-stage m-t table for the revised version of Polyphonie X (PSS, Sammlung Pierre Boulez, Mappe C, Dossier 1e)
Ex. 9.3 A transcription of a late-stage m-t table for Structures Ia, bars
65–72 (PSS, Sammlung Pierre Boulez, Mappe D, Dossier 1a, 2)
among the instrumental ensemble provide ample opportunities for Boulez to

inflect the ‘serialised’ musical materials with his musical intuitions.
In Structures Ia (1952), the organisational tables are even more precise.
Here, one can literally see the correspondence of each note, rhythm,
dynamic marking and articulation mark in the m-t table, this despite the
lack of any musical staves or rhythmic stems (see Example 9.3). Perhaps even
more astonishing, all of the material in these tables directly corresponds to
the m-s outlines that precede it – the expansion is eerily automated.13
In contrast, the earliest outlines appear quite vague, using Greek symbols
to represent entire rows of material in a very compact space; it is only after
these rows are expanded to represent individual pitches that the analyst can
see the degree of predetermination as it extends all the way back to the
original schemas, leaving the composer with seemingly little to do besides
follow his own instructions.
But things are not what they seem. For example, Polyphonie X was heavily
and substantially revised before it was ever performed. While one can easily cite
basic differences between the two versions of the score, including the reduction
of instruments from forty-nine to eighteen and the removal of quarter-tones, it
is however much harder to convey the extraordinary effort Boulez put into
these changes.14 Indeed, while both versions of Polyphonie X share a number of
13
Some would relate this to the tone of Ligeti – Xenakis – Boulez’ and ‘Structure, que
György Ligeti’s famous analysis of Structures me veux-tu?’ for further rebuttals of Ligeti’s
Ia in Die Reihe. However, as I point out argument.
14
below, Ligeti’s analysis, while virtuosic, See Strinz, Chapter 6 in this volume, in
misses the broader significance of this addition to his Variations sur l’inquiétude
movement within Structures, Premier livre as rythmique and ‘Quelques observations sur
a whole. See also Piencikowski, ‘Inscriptions: des “objets retrouvés”’.
230 Joseph Salem
basic formal features, Boulez completely reconstructed the sketch

progression from the m-s outlines to the m-t tables in order to convert
the twenty-four pitch rows of the original into the dodecaphonic ones in
the revised version. Again, the new sketches allow the analyst to trace the
composition in reverse, from the score backward, with relative certainty.
However, they also convincingly demonstrate the number and type of
decisions Boulez made during each stage of the compositional process,
including not only the design of each type of outline or schema, but also
the more spontaneous decisions required to retrofit the original design
according to the new, strictly dodecaphonic rows.15 In effect, it becomes
clear that, when revising the work, Boulez was more dependent on certain
serial processes and less dependent on other ones.
Structures, Premier livre as a whole provides an even better microcosm
of Boulez’s compositional development over the following decade. While
the mechanics of Structures Ia and its corresponding ascetic clarity were
retrospectively described as ‘absurd’ by Boulez,16 Structures Ib and Ic
feature substantial changes to their m-s outlines and m-t tables that enrich
them considerably. These changes are particularly revealing of Boulez’s
later behaviours as a composer. For example, Structures Ic uses the same
basic progression of materials as Structures Ia, but Boulez begins to
spontaneously elaborate the symbols of the m-s outline midway through
the compositional process. Example 9.4 provides a glimpse of these
changes: even without a thorough explanation, one can see how the
various substitutions in the left margin (Notes→[become]→Intensités)
change which rows are used for what parameter, and how the various
new annotations (including the majority of sub- and super-script letters,
numbers and arrows) amend the source and character of most rows. Many
of these changes result in readings of the dodecaphonic matrix that defy its
original purpose: diagonal readings of the matrix, for example, feature
a number of pitch duplications, effectively thwarting the concept of
chromatic saturation or maximal variation within each row. As far as
I can tell, all of these changes and substitutions were spontaneous
decisions, occurring between one iteration of the m-s outline and its
elaboration as an m-t table.17 They represent lasting evidence of
a transient phenomenon: Boulez inserted himself as an active participant
15 16
In fact, Boulez appears to have carefully re- See Boulez, Conversations with Célestin
calibrated some aspects of the original orga- Deliège, pp. 55–7.
17
nisation when he revised the work, while This is not to imply that Boulez did not
merely re-using others. I discuss some create a new ‘process’ for deriving and
aspects of the organisational process behind applying these changes, but to emphasise that
this work in ‘Boulez Revised’; see also Strinz, the changes themselves are not documented
Variations, ibid. or outlined prior to their appearance in the
new, edited m-s outline.
Ex. 9.4 A transcription of an altered m-s outline for Structures Ic (PSS,

Sammlung Pierre Boulez, Mappe D, Dossier 1a, 2)
among the automated processes of his new serial method, but only between
the serial processes themselves, such that little to no concrete evidence of
his actual thought process exists.
Structures Ib goes even further. One of the major innovations in this last-
composed movement of the set is Boulez’s shift from rhythmic rows based
on flags or durations (where a given number represents the value of each
rhythmic stem), to rhythmic series based on the number of stems, where the
duration of a stem is changed at will. This change, among others, completely
disrupts the most basic feature of Structures Ia; that is, the direct, one-to-one
correspondence between each musical parameter (i.e. for every row of
pitches there is exactly one row of rhythms, articulations and dynamics).18
The ramifications are many: the incongruity among the various series
required Boulez to constantly recalibrate the relationship between the musi-
cal parameters of pitch, duration, articulation and dynamics, introducing
a whole list of techniques to synchronise, say, twenty-one rhythmic rows
with just ten rows for pitch. These changes provided Boulez with one of his
18
In truth, some pitch rows are accompanied correspondence, in that for every one row of
by a single value for rhythm, articulation or pitches, there is either one row or one value of
dynamics. However, these instances still durations, articulations and/or dynamics.
follow the concept of one-to-one
232 Joseph Salem
first opportunities to work with non-dodecaphonic series of values, while

also creating many additional layers of discrete serial operations. It is
between these layers that Boulez finds his home, redefining the ‘rules of
the game’ by twisting, truncating and expanding the values of his various
rows using fractions, rotations and other techniques to create more compel-
ling gestures on the musical surface. The effects are immediately noticeable:
the explosive surges and cascading streams found throughout Structures Ib
are far closer to the sumptuous textures of Boulez’s earlier and later works
for piano than those of Structures Ia, revealing, as it were, the invisible hand
of the composer.
What these early works provide, then, is a guide for where to locate
Boulez’s creative process in later works. Among some of his most auto-
mated, rigid compositional structures – here called m-s outlines
and m-t tables for their unique, schematic representation of pitch, rhythm
and time – various liminal spaces become a silent, even abstract play-
ground of compositional activity. Tracing the works in reverse – from the
score backward – accepts these changes as inevitable consequences of the
compositional process. But studying the progression of sketches as under-
determined layers in an expanding, evolving universe reveals the silent
effect of genetic mutation in the spaces between the sketches. Where,
for example, is the source for Boulez’s changes to the m-s outlines for
Structures Ic? What instigated his switch from durations to stems for
calculating the rhythms in Structures Ib? These, among a host of other
subtle changes,19 reveal that the causal relationships between Boulez’s
sketches are suspect, masking the catalysts of change and development
that subtly challenge the so-called automated unfurling of the m-s
outlines, m-t tables and the final composition of each work.
Self-Borrowing and the Roots of Improvisation

It is no secret that Boulez has frequently revised his works.20 Even when
revisions predate the publication of a work, they usually include certain types
of alterations which clearly differentiate the process of ‘revision’ from that of
19
Ironically, the most obvious changes that specific instances of revision throughout
Boulez makes to his works during the process Boulez’s oeuvre than any other source.
of composition are related to large-scale Throughout his massive and ambitious
form: many of his works fail to ever fulfil the project, Tissier takes pains not only to
original scope of his preliminary outlines. summarise general trends in Boulez’s use of
20
While my own recent research reveals revision, but to actually document (bar by
a few particulars related to revision, Tissier, bar) the type and degree of specific revisions
‘Variations-Rondeau de Pierre Boulez’ is far in a large number of Boulez’s works.
more comprehensive and descriptive of
the original design, organisation and execution of a piece. For this reason, it
is useful to differentiate the many ways Boulez uses revision to aid his
compositional process. In our first category of revision, changes are
subsumed into our primary understanding of a composition. Such is the
case with the early sonatas: while specialists concern themselves with the
early manuscripts of these compositions, performers and listeners gen-
erally assume the revisions were just minor improvements designed to
bring the work closer to the composer’s original intentions.21
In our second category, the revisions are significant enough to require
a separate opus to allow both versions of a work to continue to coexist.
This is the case with the first . . . explosante-fixe . . . (1972), the
later Mémoriale (. . . explosante-fixe . . . Originel) (1985), and the still
later . . . explosante-fixe . . . (1993).22 Finally, a third category contains
works that are hardly equated with traditional ‘revision’ at all, even while
they share specific ideas – or entire passages – with one another. This last
category aligns most closely with Boulez’s own idea of ‘organic’ composi-
tion, where a core body of musical material forms a trunk, while addi-
tional musical works extend outward as explorations of new, derivative
possibilities.23 Examples include the use of the ‘Originel’ series in . . .
explosante-fixe . . ., Rituel and Anthèmes 1 and 2, and the ‘Sacher’ series
in Messagesquisse, Répons and Dérive 1 and 2.24
The differences among these types of revision are significant for under-
standing Boulez’s compositional process and corresponding aesthetics.
A related conclusion is that grouping these varied uses of revision as
undifferentiated ‘works in progress’ (according to his own parlance) does
a disservice to the variety of creative impulses felt and used by the composer
in his works.25 This makes untangling the use of revision in Boulez’s
compositions a necessary component of any hermeneutics for analysing
his works, even when the revisions themselves are a hidden or integral part
of the original composition (and not edits of finished works). In point of fact,
the greatest differences among Boulez’s various types of revision reveal
21 23
I do not mean to suggest that performers of This is particularly true of Boulez’s
these works are unaware of the significance of description of his compositional process in
Boulez’s revisions, but rather that these same a recent DVD recording of Éclat (Pierre
performers consistently choose to perform Boulez: Éclat (Idéale Audience International,
the most recent version of such works, 2006)), wherein he describes it in relation to
regardless of their research into Boulez’s the growth of a tree (around thirty-four
compositional process. For an additional minutes into film 1).
24
approach to treating Boulez’s process of For a recent, approachable, and excellent
revision, see Gärtner, Chapter 2 in this discussion of how Boulez reuses such series
volume. in multiple works, see Goldman, The Musical
22
See Dal Molin, Chapter 11 in this volume, Language of Pierre Boulez.
25
as well as his ‘Introduction à la famille See also Piencikowski, Chapter 4 in this
d’œuvres “. . . explosante-fixe . . .”’. volume.
234 Joseph Salem
significant changes to the source and elaboration of his creative impulses

over time – changes that should affect where and how we identify the
emotive features of each work. Consider, for example, that Boulez’s earlier
works often featured borrowings or revisions that were quite limited to
a single series or a composed score; it was only with the borrowed blocs
sonores that bind Oubli signal lapidé to Le Marteau sans maître that
he created an entirely new composition that was deeply indebted to the
organisational sketches of the prior work. Around 1955 this type of
borrowing became more overt, with the Symphonie mécanique and
L’Orestie sharing bits of organisational material in a much more haphazard
way, a sign both of the creative catalyst that the blocs sonores from the
Symphonie – later known as the Séquence matrix for the Third Sonata for
piano – provided for Boulez, and the intense time pressures that affected the
composition of both works. Just a few years later, Boulez switched to sharing
massive amounts of thematic material between his works, transcribing whole
swathes of notes to create Le Crépuscule de Yang Koueï-Fei (1957), Strophes
for flute (1957) and the first two ‘Improvisations sur Mallarmé’ from Pli
selon pli (1957–8).26 Admittedly, the borrowings remained discreet for quite
some time: Boulez rarely reused thematic material that had been performed
more than once or published in any form, and his continued reuse of
organisational sketches such as row matrices and blocs sonores was virtually
untraceable without recourse to his manuscripts. Nonetheless, these devel-
opments, largely strewn across a single decade of continuous productivity,
show just how integral revision was for stimulating Boulez’s musical
intuitions during his formative years. They also illustrate how profoundly
different the implications of terms such as ‘self-borrowing’, ‘transcription’,
‘intertextual borrowing’ and ‘revision’ can be when constructing
a hermeneutics for Boulez’s works.
In fact, it was only in later years that Boulez’s reliance on sharing material
between works became publicly acknowledged; or, to put it another way, it
was years before members of the broader musicological community could
use this knowledge to interpret his musical works. At first, it was primarily
Boulez’s recycled titles and dropped hints in interviews that led to his own
nondescript comments about the meaningful musical relationships that
enrich his creative process.27 Slowly, families of compositions emerged
(such as the bonds between the many versions of . . . explosante-fixe . . .
26 27
Several commentators, including myself, This is particularly true after the publica-
have described these borrowings in greater or tion of Boulez’s conversations with Célestin
lesser detail. See esp. Guldbrandsen, Deliège (1975 in French, 1976 in English).
Tradisjon og tradisjonsbrudd and ‘New Light In these interviews, a number of specific
on Pierre Boulez and Postwar Modernism’, relationships between pieces are made quite
Tissier, ‘Variations-Rondeau de Pierre explicit by the composer (and are thoroughly
Boulez’ and Salem, ‘Boulez Revised’. indexed by the publisher).
and Rituel on the one hand, and Messagesquisse, Répons, Dérive, Incises and
sur Incises on the other). Still later, the opening of the Sammlung Pierre
Boulez at the Paul Sacher Stiftung in Basel, Switzerland finally encouraged an
increasing number of scholars to clarify the complex combinations of
organisational and notational borrowing found among the manuscripts for
these families of works.28 These later uses of self-borrowing and revision
undoubtedly built on the practices of the 1950s, but the clarity and efficiency
of new compositional techniques suggest that still further evolutions shaped
Boulez’s compositional process in these years. To this day, many of these
relationships remain unexplained or cryptically buried in his sketches,
despite their essential role in his creative process and, no doubt, in our de
facto understanding of his musical works.29
Thus, the broad categories of ‘revision’ suggested above become all the
more relevant for outlining the various developmental stages in his crea-
tive process. First, Boulez’s earliest, most basic revisions reveal his unusual
talent for refining precomposed material. Second, his reuse of organisa-
tional and – only later – thematic material extend the idea of revision
backward, deeper into the compositional process. This stage is clarified by
Examples 9.5a and b, which help to illustrate just how much Boulez has
relied on borrowed figures as the starting point for new gestures or
harmonies, shifting his attention as a composer from the abstract realms
of serial organisation (so prominently showcased in Polyphonie X and
Structures, Premier livre) to the contours, dynamics and orchestration of
actual notes. However, it is important to emphasise that Boulez probably
resorted to composing in this way to reap maximal rewards from his
compositional efforts by reusing material that had little to no future in
its original context, resurrecting it instead in newer, more promising
works. Third and finally, it was only later that Boulez incorporated these
techniques into his ‘normal’ creative process when, as pressing deadlines
abated and he began promoting the concept of ‘works in progress’, his
pieces begin to gestate over many years. As families of compositions
emerge, not only did Boulez begin to openly acknowledge the borrowings,
but his sketches appear more organised, and the borrowings more
consistently handled. Thus, in his most mature compositions, the reliance
28 29
The efforts of the PSS staff should be This is particularly true of some of the most
especially praised in this regard. esoteric relationships between Strophes,
The Sammlung Pierre Boulez is itself an ‘Don’ and ‘Improvisation sur Mallarmé III’.
extraordinary resource for tracking the While some of these relationships have been
cross-pollinations among Boulez’s many articulated by a number of scholars, sketch
works. There is little doubt that the clarity evidence suggests there are still more
and robustness of the catalogue has strongly connections to be made, particularly as
contributed to the nature and direction of regards thematic borrowings between
recent research on the composer. L’Orestie and the ‘Improvisation’.
Ex. 9.5a A transcription from Strophes for flute (strophe 1h, cycle 1) used as the basis for ‘Don’ for piano, below (PSS, Sammlung Pierre Boulez,
Mappe G, Dossier 2c)
Copyright © 2016. Cambridge University Press. All rights reserved.
Ex. 9.5b A transcription from ‘Don’ for piano, based on the Strophe excerpt above (PSS, Sammlung Pierre Boulez, Mappe G, Dossier
4b, 6)
238 Joseph Salem
on ‘revision’ shifted from heightening the efficiency of his compositional

process to enhancing its musical efficacy.
This narrative is roughly supported by the aesthetic revelations Boulez
experienced during the 1950s. Alongside his adoption of integral serialism,
Boulez’s passion for art and literature was fostered by a host of musicologists
and composers.30 Perhaps first and foremost on this list – at least in regard to
literature – was John Cage. While Boulez had already developed a frequent
correspondence with René Char (due in large part to his desire to set Char’s
works in his earliest years as a composer), his letters to Cage are some of the
most intimate and exploratory ever written by the composer, and they often
reveal Boulez’s first impressions of important literary influences such as
Joyce, E. E. Cummings and Mallarmé. In later years, he discusses similar
influences with Stockhausen, with the letters from the mid-to-late 1950s
chronicling several epiphanies regarding the work of Mallarmé and Rougier
in particular.31 All of these sources reveal Boulez’s ongoing fascination with
modernist literature, including a particular emphasis in later years on
labyrinthine narrative constructions (no doubt indebted to Kafka’s
burrows).
While it is possible to cite these sources to suggest various aesthetic
reasons for Boulez’s increasing use of revision, it is also important to
remember that his practices were greatly affected by much more practical
concerns: the lack of time and energy to compose, the need for pre-
constructed organisational materials and – last but certainly not least –
the creative catalyst pre-notated music provided for his écriture.
The congestion of activities during the 1950s supports this perspective
time and time again: Boulez struggled to finish Marteau amongst the
distractions of world tours and the burgeoning Domaine Musical; he
raced to finish L’Orestie between the varied deadlines and rehearsal plans
set by Jean-Louis Barrault; he was pressed to fulfil a commission for
a movie soundtrack (Symphonie mécanique), and much more. Later
years introduced new professional pressures, among them the challenge
of Stockhausen’s growing reputation and success (Gesang der Jünglinge,
Gruppen), Cage’s growing influence at Darmstadt, and Boulez’s own
success with the Mallarmé ‘Improvisations’ – and all of this before the
commencement of his professional conducting career and the exponential
growth of his political and social involvement in countless international
30
Campbell, Boulez, Music and Philosophy Mallarmé and Stacey, Boulez and the Modern
provides an excellent summary of Boulez’s Concept, among many others.
31
intellectual cohorts at this time. Past scholars This is particularly true of letters from
have also articulated the nature and degree of Boulez to Stockhausen dated end
Boulez’s interest in modernist art and litera- of August 1956, and from beginning
ture, including Breatnach, Boulez and of October 1957.
music organisations and festivals. In effect, Boulez’s works not only

quickly outgrew the earlier aesthetic of integral serialism, but also actively
sought to redefine musical gesture and timbre amongst the rapidly grow-
ing, caste-crushing anarchy provided by graphic notation, theatrical sta-
ging, electro-acoustic experimentation and chance procedures in the late
1950s.
My point is that revision facilitated a new emphasis on lyricism precisely
when Boulez needed it most, to combat what he saw (at the time) as
a proliferation of groundless compositional exploits. It is now impossible
to doubt the literalness of notated borrowings between L’Orestie, Strophes
and ‘Don’ for piano; likewise, several scholars have elaborated upon the
close ties between the notated musical ‘sonnets’ originally designed for use
in Strophes but used at length in all three of the ‘Improvisations sur
Mallarmé’.32 These latter examples are particularly demonstrative of
Boulez’s new focus on lyricism: the borrowings, which vary between literal
transcriptions of exact notes in ‘Improvisation I’ to highly elaborate recom-
positions of similar material in ‘Improvisation III’, are recorded in a number
of drafting sketches. Such elaborate reworkings of material show just how
obsessed Boulez could be regarding the shape and contour of a melodic line
(see Example 9.6).33 In effect, malleable, precomposed material supplants the
original series or predetermined row as the counterpoint to his creative
impulse. As with his earliest revisions, Boulez performs his best not when
creating new material from scratch, but when revising something that he has
lived with as actual music.
In sum, works such as the ‘Improvisations’ remind us that the difficulty of
interpreting Boulez’s music stems – ironically – from the lack of a consistent,
reliable compositional system or organisational process. In fact, Boulez’s
dependence on revision and his treatment of precomposed material in his
later works serve as constant reminders of the attention he brings to the
surface of his compositions, regardless of the organisational schematics
underlying the music. While earlier works used modular outlines and tables
to control the combination of multiple dodecaphonic rows, Boulez’s
ultimate goal was to eliminate this type of pre-planning entirely. To be
sure, his later compositions still feature early outlines, and his organisational
sketches remain highly schematic. Over time, however, the basis for such
outlines lies not in abstract proportions, numbers or rows, but in precom-
posed bodies of music or a single, undefined series or set of blocs sonores.
32
The relationship between these works has and my own work (‘Boulez Revised’) builds
been discussed by other scholars. on this.
33
Guldbrandsen, Tradisjon og tradisjonsbrudd See, in addition to the above,
and ‘New Light on Pierre Boulez and Postwar Guldbrandsen, Chapter 8 in this volume.
Modernism’ offer representative treatments,
Ex. 9.6 A transcription of Boulez’s melodic sketches for ‘Improvisation III’, relating to p. 4 of the 1959 score (PSS, Sammlung Pierre Boulez,
Mappe G, Dossier 3e, 5c). The sketch at the top is read backwards (in retrograde) to produce the rhythmically elaborated line below, which is
subsequently revised
This shift to using notated elements as the basis of his compositional method
corresponds to the loosening of limits – structural, formal, musical – in his
works. It is no surprise, then, that his later works feature more internal
expansion and more additive forms: with the diminishing importance of
rigid, predetermined formal structures based on serial vectors or matrices,
Boulez became free to multiply and expand his more successful musical ideas
right on the musical surface, crafting his compositions more and more
frequently with his ear while satisfying his obsessions with visual refinement
and narrative abundance.34
Revision as a Reflection of Changing Priorities

There is little reason to deny that, as Boulez’s conducting responsibilities
proliferated during the 1960s and ’70s, his approach to composing
adapted to his new, increasingly saturated schedule.35 It is also relevant
that Boulez wrote less about music during these years, probably
in response to his growing international stature as well as his rapidly
diminishing free time. Together, the effects of these changes have caused
some scholars to suggest an ebbing in Boulez’s interest in composition at
this time, not least because, after the extremely productive 1950s, rela-
tively few new works were successfully completed in the following decade.
In hindsight, however, it seems more appropriate to highlight these years
as the natural continuation of the late 1950s, when small streams of ‘works
in progress’ shifted Boulez’s attention away from completing commissions
and competing with Stockhausen and Cage towards the gradual develop-
ment, revision and expansion of a series of works independent of the
pressures of publication and performance. This latter perspective helps to
explain how – despite the lack of completed, published works – the
quantity and quality of Boulez’s output during the 1960s and ’70s is
quite extraordinary in breadth and depth.
Contrast, for example, his attempt to quickly compose Strophes in 1957
versus the long, ongoing adaptation of Domaines for clarinet into a work for
small ensemble (1961–8). Both works were meant to explore the scenario of
a mobile soloist who performs with an interactive ensemble of select instru-
ments. Yet, while a virtually complete fair copy of Strophes does exist, no
other parts were ever completed, and the work was never performed (nor
was the commission fulfilled) – this despite what appears to be a brilliant,
34 35
My conclusion is meant to resonate with This is one topic where Peyser’s treatment
the hermeneutic applications of ‘virtual’ remains undramatised. See Peyser, Boulez:
themes and development proposed in Composer, Conductor, Enigma, pp. 191–3.
Campbell, Boulez, Music and Philosophy.
242 Joseph Salem
virtuosic and lyrical work of the highest quality for solo flute. In contrast,
Domaines began as a humble work for solo clarinet before its slow
expansion – over the next eight years – into multiple iterations of an
ensemble piece. In later years, Domaines also became the basis for a new
composition, Dialogue de l’ombre double (1985).36 It would be wrong to
suggest that these two bodies of work feature entirely different composi-
tional processes: both are heavily based on borrowed material, and the
conceptual design and musical goals of each work are remarkably similar,
if physically unique. The difference, then, lies in Boulez’s new, more relaxed
approach to expanding and developing his works, an outlook which post-
dated Strophes but enveloped Domaines. As a result, Boulez spent nearly
a decade on Domaines, continually expanding and improving the work as it
gained in popularity.
A similar story describes Doubles, which went from an incomplete
movement for orchestra in 1957–8 to a larger, but still ‘incomplete’ work
in progress marked by multiple revisions (1964, 1968). Éclat also builds on
works from the 1950s, only to expand – again, in the later sixties – to
include the well-known Multiples and the much less-known (and never
performed) expansion of Multiples.37 Like Strophes, the latter revision
represents a polished, completely performable fair copy that has been
relegated to the Sacher vault and which is, remarkably, considerably longer
than the entire recorded (but later withdrawn) Éclat/Multiples pair.
Finally, ‘Tombeau’ was started in the late 1950s, but continued to expand
at the dawn of the next decade. As we know, the same is true for
‘Improvisation III’ and ‘Don’, although new editions of these works did
not appear until the 1980s. The list is significant: in truth, virtually all of
Boulez’s compositions from the 1960s – including Domaines – develop out
of earlier works, either through various forms of self-borrowing or, more
consistently, through ongoing revision and expansion.38
All of the above compositions illustrate a metamorphosis of Boulez’s
compositional process. These works undoubtedly look and sound differ-
ent than his earlier works in significant ways. Yet, all of the above works
borrow either organisational material, notated material, or both, from
previous works in precisely the same ways as L’Orestie, Strophes and the
‘Improvisations’ exchanged material in the 1950s. In this sense, they
remain tied to Boulez’s earlier compositions and methods, incorporating
36
For a rather detailed account of this imprévisible”’ and Edwards, ‘Éclat/Multiples
compositional family, see Tissier, et le problème de la forme musicale’.
38
‘Variations-Rondeau de Pierre Boulez’. Again, for a broad summary of tactics
37
For more on the development and (including a close reading of Domaines in this
expansion of Éclat and Multiples, see regard), see Tissier, ‘Variations-Rondeau de
Piencikowski, ‘“Assez lent, suspendu, comme Pierre Boulez’.
‘revision’ in the broadest sense ever more deeply into the compositional
process.
Nonetheless, the works of the 1960s also add an important new dimension
to Boulez’s approach to composition. Much like his earliest practices in
the sonatas and cantatas, new musical changes begin to more closely align
with how we usually define ‘revision’. Instrumentations are adjusted,
new sections are added, ‘open’ compositional structures or choices are either
wholly incorporated and ‘fixed’ or entirely eliminated, and so on (see
Examples 9.7a and b). Such changes serve to refocus our attention on the
contradictions inherent in Boulez’s modernist aesthetic: they flow so freely
from one another as to obfuscate the complex causality between his
organicist metaphors and his actual compositional process. Boulez
would have us believe these works were all meant to lead to one another,
creating a network of ideas that build upon and develop each other as
unique and individual manifestations of a shared musical potential.
However, in considering the actual compositional process behind these
works – with special emphasis on their dependence upon fragments from
the previous decade – it becomes clear that these compositions are not all
created equal. Instead, what binds them is not a network of potential, but
a compositional drive to individually reshape, reorganise, rework . . .
Ex. 9.7a A short passage from the percussion parts of ‘Don’ for orchestra,
published in 1967
Ex. 9.7b A revision to Example 9.7a, published in 1989. Note the significant expansion of pitch material, all of which is based on the harmonic
structure of the original
revise . . . his stuttering utterances into more expressive, more compelling

musical works. In effect, the variable in any Boulezian hermeneutics is not
the organisational structure of his works, but the mutability of these
structures in service of his developing musical ear.
This last point is not meant to immediately question the quality of
Boulez’s earlier or later compositions. Instead, I mean to push us again
towards the source of musical expression in his works. As more and more
attention is placed on the gradual revision of surface figuration, it
becomes clear that capturing the musically relevant details in Boulez’s
style is as much (or more) about gleaning the musical significance of such
comparisons as the ‘Don’ of 1962 and the one of 1989, as it is about tracing
the generation of each and every note in his mature compositions.
Instead, with the introduction of the word ‘improvisation’ in the titles
of Pli selon pli, Boulez seems to be guiding us away from deep structures
and towards how the surface brilliance of his music – its extraordinary
timbres, harmonies and angular but mesmerising instrumental lyricism –
articulates the same variations in form, figure and gesture so admired in
the work of great improvisatory artists. Should we be so ashamed to
‘demote’ Boulez’s music? Should we not loosen the gravitational pull of
the deep structures that hold his works together against the threat of serial
entropy in order to preserve a semblance of the autonomous, organic
work? I think so. For Boulez’s great compositional nuances, the most
expressive moments in his works, prove time and time again to show up in
the spaces between, the changes between, his otherwise autonomous serial
processes. It is there, in the countless adaptations of shape, contour and
character, that Boulez comes into his own, outpacing his contemporaries
to find sounds so unique as to forever change the resonances of twentieth-
century music.
10 Listening to Doubles in Stereo
Jonathan Goldman
The 1958 orchestral work Doubles, an unfinished fragment that would

later be integrated into Figures – Doubles – Prismes, was premièred in
successive versions in 1964 and 1968.1 Its most obvious feature, and the
one latched onto by its first critics, is its use of an unusual seating plan, in
which the orchestral choirs are divided into several groups and scattered
across the stage.2 Characteristically for Boulez, it also bears certain family
resemblances to other works in his catalogue: the pitch and rhythmic
structures are derived from tables elaborated for the Third Sonata
for piano;3 as a work that exploits space as a compositional parameter,
involving sound sources set in a unique geometric formation, Doubles has
close parallels with the work that followed on its heels as a result of
a commission from the German Southwest Radio (SWF) in 1958, Poésie
pour pouvoir, for three orchestral groups and five tape tracks. In turn,
the spatialised plan of Doubles binds it to the circular plan of the six
instrumental ensembles in Domaines (1961–8), to the circular layout of
the orchestral Rituel (1975) and even to Boulez’s major work from the
1980s, Répons. Although Boulez himself tends to regard his works as
synchronic fragments of an unfinished whole in the manner of Mallarmé’s
1
After the 16 March 1958 première of latest version of Figures – Doubles – Prismes,
Doubles, a second version (now titled minus several fast interpolations that were
Figures – Doubles – Prismes) was premièred added later. The latest (although still
on 10 January 1964 by the SWF Orchestra in theoretically unfinished) version of
Basel, and later performed on Figures – Doubles – Prismes can be heard in
11 March 1965 in Cleveland. A third ver- recordings conducted by Boulez (Erato
sion, performed on 7 March 1968 by the 45494, 1990), David Robertson (Montaigne
Hague Residentie Orkest in Utrecht, has 782163, 2003) and Bruno Maderna
been released on Darmstadt Aural (Stradivarius 10028, 2013).
2
Documents (NEOS 11060, Box 1, CD 3). As witnessed in the notorious title of
Allen Edwards’s investigations into the Clarendon’s review of the première, ‘La polka
sketch material of the work reveal that des chaises’; Gavoty, ‘Pierre Boulez ou La
Boulez had planned Figures – Doubles – polka des chaises’, p. 18.
3
Prismes to extend to thirty minutes, Edwards, ‘Boulez’s Doubles and Figures
conceived in three parts (stages), a plan that Doubles Prismes: A Preliminary Study’, p. 6,
was never realised. The version heard in n. 3.
1958 corresponds to the first section of the
246
‘Livre’,4 it is instructive to also regard Doubles diachronically, i.e. as owing

its existence in part to the specific characteristics of the historical era into
which it was launched. One of the important technological developments
of this era as far as music is concerned took place in the domain of
recording and sound reproduction, specifically the commercial introduc-
tion of stereo long-playing records that led to the mass distribution of
stereo sound technology into homes throughout the Western world.
A historiographical question immediately arises: even if Doubles partakes
in the tradition of antiphonal music from Gabrieli or Berlioz, to what
extent is an allusion to the technology of stereophony inscribed into the
fabric of this work? Answering this unusual if apparently anodyne ques-
tion requires an evaluation of Boulez’s constantly evolving discourse on
this work, as well as an analysis of the extent to which the experience of
two-channel stereophony lay at the horizon of expectations of contem-
porary listeners of Doubles. Moreover, it must be ascertained whether the
score contains gestures which might evoke in a listener a kind of
‘unplugged’ stereophony in this work that marked Boulez’s ascension
into the elegant high society of Parisian orchestral concerts.
I. Audio Technology in 1958

Doubles was commissioned in 1957 by Igor Markevitch and Georges Auric,
who requested an orchestral work from the 31-year-old composer to be
performed at a concert of the venerable Société de concerts Lamoureux. This
first properly orchestral work by Boulez was much anticipated by critics and
the concert-going public, since he had by that time gained a considerable
reputation as the foremost young avant-garde composer in France through
various networks, most notably the success of the Domaine Musical
concerts, which in 1958 were already into their fifth season. Moreover, the
success of the Vega LP recording of Le Marteau sans maître, released in 1956,
further contributed to his notoriety, not to mention Boulez’s many articles
published in non-musical journals such as the Nouvelle Revue Française
starting in 1954.5
In the same years, ‘high fidelity’ technology was increasingly finding its
way into middle-class homes in Europe and the Americas. Stereophony had
been demonstrated by Bell Labs at the Chicago World Fair as early as 1933,
4
In an interview with the author on a single work for orchestra and electronic
9 June 2014, Boulez went so far as to consider sounds.
5
Répons a later version of Poésie pour pouvoir, Véga C30A67; taken from Griffiths, ‘Boulez
as if he conceived his oeuvre as containing Discography’, prepared for the now defunct
Andante website; Boulez, Stocktakings, p. xv.
248 Jonathan Goldman
and the experience of multi-channel stereo had already been accessible in

movie theatres since the 1940s, notably through ‘Cinemascope’ and
‘Cinerama’ technology, which used two, three or more loudspeakers to
create the illusion of acoustic depth as well as to localise sounds in the
auditory plane. The most famous example of this use of multi-channel
stereo in the realm of classical music was in the 1940 film Fantasia, whose
score, conducted by the technophile Leopold Stokowski, was recorded in
multi-tracked stereophony using a pioneering technology that was
dubbed ‘Fantasound’. Stereo technology had also been adapted for home
use beginning in 1953, when two-track stereo was available to audio
enthusiasts through the medium of magnetic tape, and, beginning in
1958, on stereo LPs played with needles able to read two distinct sound
tracks within a single groove of a phonograph disk.6 In the same year,
some FM radio stations began transmitting signals in two-channel stereo.7
From that point forward, stereo sound would be distributed on a massive
scale, prompting the New York Times to announce in a 1959 headline
that ‘Plants quicken tempo to meet stereophonic sales crescendo’.8
Some musicians of a commercial bent took to producing recordings that
highlighted the characteristics of these new stereo systems, one of the
most infamous (and successful) being the big band leader Enoch Light,
whose 1959 stereo LP Persuasive Percussion featured a ‘musical line [that]
shuttled back and forth from one speaker to the other, like the ball in
a ping-pong game’,9 through an immoderate use of the left-right
stereo field coupled with musical arrangements (of familiar poppy jazz
numbers like ‘Brazil’) designed to have musical figures wander to distant
quarters of the big band over the course of a song. Although a modernist
response to new sound technology would not be expected to adopt sonic
strategies typical of Persuasive Percussion (i.e. sonic ping-pong, an effect
that as we shall see, Boulez explicitly claims to avoid), the familiarity of
typical listeners of avant-garde works with this new cutting-edge audio
technology may well have informed their perception and comprehension
of them.
Of course, many avant-garde creations exploit the possibilities of new
audio technology, Pierre Schaeffer’s use of sounds recorded on cylinders or
grooved disks in his earliest musique concrète experiments being one of the
6 7
The system for developing single-groove ‘WFUV To Program Stereophonic FM:
two-channel stereo disks was developed by Fordham Radio Station Will Begin
the Westrex company, and demonstrated at Broadcasts Oct. 1 – New Technique Planned’,
the annual meeting of the Audio Engineering New York Times, 22 September 1958, p. 52.
8
Society in New York in October 1957. Zipser, ‘Plants quicken tempo to meet
The major record labels began producing stereophonic sales crescendo’, p. F1.
9
stereo disks in 1958 (Gelatt, The Fabulous Roland Gelatt, The Fabulous Phonograph,
Phonograph, p. 316). p. 318.
most famous examples. These works demonstrate a ‘phonograph effect’,

a term Mark Katz coined to designate ‘any change in musical behavior –
whether listening, performing, or composing – that has arisen in response
to sound-recording technology’.10 The first example Katz offers of
a phonograph effect is the temporal structure of Stravinsky’s Serenade
in A (1925) for piano, since, following a commission from the Brunswick
Gramophone Company, the Russian composer planned for the work’s
four movements to fit onto the sides of two 78 rpm shellac disks.11
Other avant-garde creations may reference audio technology in their
conception even without actually making use of any technological means.
The ‘phonograph effect’ might still be discerned in such cases, since Katz
defines this effect broadly as ‘any observable manifestation of recording’s
influence’.12 It might be reasonable then to suppose that works for multiple
orchestras or orchestras divided into spatialised ensembles, the most famous
example in the 1950s being Karlheinz Stockhausen’s Gruppen (1955–7),
were in part composed in response to new advances in multi-channel
sound dissemination, even if these works employ no microphones, loud-
speakers or other electronic equipment. Similarly, it is instructive to measure
the extent to which stereo audio technology influenced the conception as
well as the reception of Boulez’s Doubles.
II. Boulez on Stereo

The claim that stereo technology could have had an impact on Boulez’s
musical thought becomes plausible when one considers that early in the
1950s he had produced tape studies in Pierre Schaeffer’s studio that
employed a multiple loudspeaker setup. Moreover, Boulez refers to stereo
occasionally in his writings throughout the 1950s and 1960s, although these
passages need to be read carefully since the meaning of the word ‘stereo’ only
stabilized to mean two-channel sound sometime in the 1960s, it having
conveyed a more general sense of spatialised multi-channel sound before
that time.13 In Penser la musique aujourd’hui, Boulez’s 1963 monograph
based on lectures delivered at the Darmstadt Summer Courses in 1960, he
cites the use of stereo sound in the cinema: ‘Stereophonic sound in cinemas
also dates from the same period (Cinerama in New York, late 1952) as do the
10
Katz, Capturing Sound, p. 2. pour Piano came to be written.’ Stravinsky,
11
‘In America I had arranged with An Autobiography, pp. 123–4; cited in Katz,
a gramophone firm to make records of some Capturing Sound, p. 2.
12
of my music. This suggested the idea that Katz, ibid.
13
I should compose something whose length Valiquet, ‘The Spatialisation of
should be determined by the capacity of the Stereophony’, p. 404.
record. And that is how my Sérénade en LA
Son et lumière displays.’14 The typescript Boulez prepared before publica-

tion of his monograph reveals that the reference to son et lumière shows
was added at a later stage of revision, Boulez thus revealing himself to
be eager to position his own research with respect to the most recent mass-
market applications of multi-channel sound.15 He goes so far as to
acknowledge that for a time anyway, progress in avant-garde music and
advances in popular cinema technology were being made at a comparable
pace: ‘Commercial and industrial applications appear to be approximately
in step with more disinterested research.’16 Even if Boulez aligns himself
in this passage with ‘more disinterested research’, he still places his work
in a field which also contains commercial stereophonic ventures like
cinema and son et lumière.
Five years before the 1960 Darmstadt lectures that would become Penser
la musique aujourd’hui, Boulez praised in the pages of Die Reihe the potential
of electronic music to create a ‘multi-dimensional space’ that could ‘be
happily expressed through the real multiple dimensions of stereophonic
space’.17 He nevertheless goes on to issue a telling caveat about the rash
musical use of stereophony:
Thus the arrangement in space becomes a structural necessity and represents

considerably more than an appropriate setting for a more or less spectacular
exhibition – though the very idea of this stereophony is enveloped in such a mist of
confusion, owing to its continual vulgarisation in the cinema and in all kinds of ‘son
et lumière’, that the best intentions are discouraged by the incidental experiences of
similar appliances.18
It is clear from this and other passages that Boulez was wary not only of the
‘anecdotal’ use of stereophony in the cinema but also of its unsubtle use in
musical works. In Penser la musique aujourd’hui, he deplores the ‘dreadful
14
‘La diffusion stéréophonique dans les salles par une réelle multiplicité de dimensions
de cinéma date d’ailleurs de la même époque dans l’espace stéréophonique’ (Boulez, ‘À la
(CINERAMA à NEW YORK, fin 1952) limite du pays fertile’ (1955), Relevés d’ap-
(Boulez, Penser la musique aujourd’hui, prenti, p. 210; Stocktakings, p. 163).
18
pp. 72–3; Boulez on Music Today, p. 66 Boulez, ‘At the Limit of Fertile Ground’,
[translation revised by the author]). p. 21; French original: ‘[L]a répartition
15
On p. 22 of the typescript, Boulez adds the spatiale n’est pas alors une mise en scène en
sentence ‘C’est également le temps où com- vue d’effets plus ou moins spectaculaires,
mencent à prendre leur essor les spectacles mais devient une nécessité structurelle.
SON et Lumière’ (‘it is also the time in which Toutefois, cette notion de stéréophonie, si
sound and light shows become in vogue’). vulgarisée par le cinéma, ou diverses formes
Fonds Boulez, Université de Montréal. de parades son-lumière, a été absorbée par
16
‘Les applications industrielles et commer- ces prétextes voyants, si bien que la confusion
ciales, on le voit, vont à peu près de pair avec règne en ce domaine, et que les meilleures
les recherches plus désintéressées’ (p. 22 of intentions sont découragées par les
typescript). incidences anecdotiques de pareilles utilisa-
17
‘Une espace multidimensionnelle’ qui tions’ (‘À la limite du pays fertile’ in Relevés
‘pourrait d’ailleurs heureusement s’exprimer d’apprentis, p. 207).
and regular epidemics’ of certain fleeting aesthetic trends, among others ‘the
stereophonic year’.19 Boulez is opposed to the superficial or simplistic use of
stereophony in musical compositions, judging that ‘the summary use of
stereophony borders on the delights of Cinerama, that is to say, it relies on
a cheap and anecdotal idea of space’.20 The non-anecdotal use of stereo-
phony that Boulez does endorse, differs in several respects from its simplistic
counterpart:
[S]patial distribution seems to me to merit a type of writing just as refined as the

other sorts of distribution already encountered. It ought not only to distribute
spaced-out ensembles according to simple geometric figures, which after all always
turn out to be contained in a circle or an ellipse: equally – and in fact even more so –
it must order the micro-structure of these ensembles. While speed of displacement
has always been stressed above all, little attention, amounting almost to total neglect,
has been paid to the properties of statistically distributed objects linked in a circuit,
or of mobile objects.21
Boulez arrives at the conclusion that ‘the real interest in distribution lies in
the creation of “Brownian movements” within a mass, or volume of sound,
so to speak’.22 By referring to ‘Brownian motion’ – that is, the random
motion of particles – he seems to be advocating a use of space that defies
systemisation, whether in the form of sound trajectories describing geo-
metric shapes or some permutational system: space appears to be in this
passage a dimension that Boulez is unwilling to quantify or parametrise, and
therefore a source of unpredictability; at the same time, the passage displays
his habitual binary thinking, in which the interest of spatialised musical
objects resides in the opposition that can be established between mobile and
static objects.23
19
‘. . . épidémies redoutables et régulières’ de On a mis l’accent surtout sur la vitesse de
‘l’année stéréophonique’ (Boulez, Penser la déplacement, on n’a pas assez prêté attention,
musique aujourd’hui, p. 17; Boulez on Music on a totalement négligé même, la qualité des
Today, p. 21). objets répartis statiquement, reliés entre eux
20
‘Qui emploie sommairement la par un parcours, ou encore des objets
stéréophonie, rejoint les délices du Cinérama; mobiles’ (Boulez, Penser la musique
c’est dire, qu’on se réfère à une idée assez peu aujourd’hui, p. 73; Boulez on Music Today,
relevée de l’espace anecdotique’ (Boulez, p. 67).
22
Penser la musique aujourd’hui, pp. 18–19; ‘Il me semble que le véritable intérêt de la
Boulez on Music Today, p. 22). répartition réside en la création de “mouve-
21
‘[L]a répartition spatiale me paraît mériter ments browniens” dans une masse, dans un
une écriture aussi raffinée que les autres types volume sonore, si je puis m’exprimer ainsi’
de répartition déjà rencontrée. Elle ne doit (Boulez, Penser la musique aujourd’hui, p. 75;
pas seulement distribuer des ensembles Boulez on Music Today, p. 67).
23
éloignés suivant des figures géométriques On binary thinking in Boulez’s thought,
simples, lesquelles arrivent toujours, en fin de noted by Nattiez and Deliège, see Goldman,
compte, à s’inscrire dans un cercle ou une The Musical Language of Pierre Boulez,
ellipse : elle doit aussi, et plus encore, dispo- pp. 63–4 and Campbell, Boulez, Music and
ser la micro-structure de ces ensembles. Philosophy, pp. 37–67.
III. Boulez’s Discourse on Doubles

As far as the specific case of stereophonic associations in Doubles is concerned,
the title itself seems as if it might well refer to the doubling of sound sources
that one associates with domestic stereophony, even though Boulez draws
attention in at least one programme note (that of the US première in 1965) to
other associations of the word: ‘the eighteenth-century word for variation’ and
‘the German word Doppelgänger, which means a human double’.24 It is
instructive to track Boulez’s discourse on this work in the form of programme
notes to concerts presenting new versions of the work, in order to see to what
extent stereo technology is invoked. In the note that was included in the
programme to the 1958 première, reproduced here in full, Boulez makes
explicit reference to stereophony as the poetic source of the work:
First of all, the title: Doubles. This short work will doubtlessly be included later in
a suite of pieces for orchestra composed in due course. Here, I used the standard
orchestra, by which I mean the number of performers, or thereabouts, that form
such an association. But I took the liberty to have them change their position.
In effect, the arrangement of the orchestra on stage always follows, with a few
variants, the type established in the 19th century that was itself inherited in large part
from the preceding century. Although I have added to the orchestral forces, I have
not considered the acoustic problems that musical écriture posed, and
I accommodated myself to the three screens of timbre that constitute the ‘classical’
orchestra.
Composition [écriture], in our time, calls the physics of the orchestra into
question. No one will contradict me when I state that when timbres follow each
other in rapid succession, they should not be excruciatingly stuck to each other
through a distance-obstacle; no one will contradict me when I claim that the ear in
our time requires stereophony in its desire for clarity and movement.
This ‘demonstration’ through an arrangement of the instruments of the orchestra
is truly required by the musical composition [écriture], therefore, it originates in the
poetics of new works. This is one of the characteristics of this score which I wished to
emphasise.25
24
Programme note to Figures – Doubles – La disposition de l’orchestre sur une scène, en
Prismes, concert of Cleveland Symphony effet, suit toujours, avec quelques variantes, le
Orchestra, Pierre Boulez, conductor, 11–13 type fixé au XIXe siècle qui, lui-même, était
March 1965; archive of Cleveland Orchestra. hérité en grande partie du siècle précédent. Si
I wish to thank Deborah Hefling, archivist of l’on a augmenté les effectifs, on n’a guère
the orchestra, for her help in locating this songé aux problèmes acoustiques que posait
programme. la transformation de l’écriture musicale, et
25
My translation. French original: ‘Tout l’on s’était accommodé de ces trois écrans de
d’abord, le titre : Doubles. Cette courte œuvre timbre que constitue l’orchestre “classique”.
s’insèrera sans doute plus tard dans une suite L’écriture, de nos jours, met en cause la
de pièces d’orchestre écrites en son prolon- physique de l’orchestre. Nul ne me contredira
gement. Ici, j’ai utilisé l’orchestre normal, je si j’affirme que lorsque des timbres se
veux dire le nombre de titulaires, ou à peu succèdent rapidement, ils ne doivent pas être
près, que comporte une association. Mais j’ai péniblement accrochés les uns aux autres par
pris la liberté de leur faire changer de place. delà une distance-obstacle; nul ne me
An optimistic attitude towards stereo as an invention that had allowed for

the conquest of space may have rubbed off on the Boulez who authored the
programme note to Doubles only a few months before the release of the first
commercially distributed stereo LPs. This attitude of creative excitement was
ultimately (and inevitably) followed by the banalisation of stereo, a product
of its ubiquity, culminating in a later era in which musicians adopted an
attitude towards two-channel stereo that verged on contempt, regarding it
either as a questionable domestic makeshift for the richer spatial possibilities
offered by multi-speaker setups or as a vulgar sound effect invented with the
sole aim of selling new sound systems to gullible audiophiles. The latter
opinion was powerfully and sardonically voiced by Boulez’s one-time tea-
cher René Leibowitz in Le compositeur et son double (1971):
So it is too with a more recent category of discophiles, the stereophony enthusiasts.

They are proud of their phonographic equipment that sometimes costs
a considerable amount and they ceaselessly set about improving it through the
addition of all sorts of little technical supplements that can themselves be extremely
costly. These enthusiasts take a particular pleasure in demonstrating the quality of
their equipment to their friends and acquaintances, often without any musical
consideration as such. They are proud of their equipment and the records that they
buy reconstitute what they call the ‘real sound space’, in that one is effectively able to
hear the first violins mostly in the left speaker and the cellos and trumpets in the
right. They delight in doing all sorts of experiments of this kind – encouraged as
a matter of fact by the record manufacturers themselves who seek to train and to
satisfy this new clientele in this way – in which the so-called ‘ping-pong’ effect plays
a preponderant role. There is nothing more ridiculous, in truth, than certain of these
meetings among stereo enthusiasts.26
contredira encore si j’affirme que l’oreille, de adjonctions techniques qui finissent, elles
nos jours, exige la stéréophonie dans un désir aussi, par coûter fort cher. Ces amateurs
d’évidence et de mouvement. prennent un plaisir tout spécial à démontrer
Cette “manifestation” par une disposition la valeur de leurs appareils à leurs amis et
entre les instruments de l’orchestre est bien connaissances, et cela aussi, bien souvent, en
réellement exigée par l’écriture de la musi- dehors de toute considération musicale pro-
que, donc, à l’origine, par la poétique des prement dite. Ils sont fiers du fait que leur
oeuvres nouvelles. Telle est une des appareil et les disques qu’ils achètent restitu-
caractéristiques de cette partition sur laquelle ent plus ou moins fidèlement ce qu’ils
je voulais attirer l’attention’ (Boulez, appellent “l’espace sonore réel”, de ce qu’on
‘Quelques mots sur ma nouvelle partition’, arrive effectivement à entendre les premiers
insert to the Carnets Lamoureux, violons surtout dans le haut-parleur situé à
16 March 1958, in the Paul Sacher Stiftung, gauche et les violoncelles et trompettes dans
Sammlung Pierre Boulez; reprinted in Goléa, celui de droite. On s’amuse à faire toutes
Rencontres avec Pierre Boulez, p. 247; my sortes d’expériences de cette sorte –
translation and emphasis). encouragées d’ailleurs par les fabricants de
26
‘Il en va de même d’une catégorie de dis- disques eux-mêmes qui cherchent à former
cophiles plus récente, les amateurs de ainsi et à satisfaire une nouvelle clientèle – où
stéréophonie. Ils sont fiers de leur installation ce qu’on a appelé l’effet de “ping-pong” joue
phonographique dont le prix atteint parfois un rôle prépondérant. Rien de plus ridicule,
des chiffres fort élevés et qu’ils n’ont de cesse en vérité, que certaines de ces réunions
de perfectionner par toutes sortes de petites
This suspicion of the dubious charms of stereo ‘experiments’ that Leibowitz

voices (deriding the facile ‘ping-pong’ effect) was shared not only by many
composers, but also by recording engineers, becoming the default aesthetic
among composers, sound engineers and discriminating listeners in the later
1960s. An anecdote recounted by the US composer Roger Reynolds of an
incident that occurred in 1968 is revealing in this regard:
The recording industry includes persons of diverse persuasions about what

constitutes a ‘natural’ or an ‘ideal’ recorded sound. An anecdote about an experience
that instructed me a few years ago: I had written a work for chamber ensemble
[Quick are the Mouths of Earth (1964–5), the recording of which was released in
196927] and, utilizing spatially separated instrumental groups, incorporated patterns
of spatially defined ‘motives’ into the composition. This fact was clearly spelled out
in the printed score’s introductory notes. Appearing, by chance, at the final editing
sessions for a stereo recording destined for commercial release, I was perplexed to
note a sense of dimension which included depth and resonance but lacked left-right
differentiation. I was informed that the motival structure that was intrinsic to the
work’s architecture had been, in effect, suppressed, because the recording engineer
disliked the ‘ping-pong effect’.28
This generalised suspicion of left-right stereophonic effects may go some

way towards explaining why, seven years later, in the programme note for
the 1965 US première of Figures – Doubles – Prismes in Cleveland, Boulez
displays a greater degree of circumspection with respect to the work’s
relationship to stereophony. The central portion of this programme note is
in the form of an interview:
The sheer visual appearance of the orchestra in this work, for example, presents
a new departure. As I pointed out in 1958 ‘the ear of our time demands stereophonic
listening in its desire for clarity and movement’. The orchestra here is symmetrically
grouped; there is a kind of soloistic ensemble in the middle of two larger groups.
The woodwinds are subdivided in three groups, the brass in four, and the strings in
five. The harps, the xylophone, vibraphone and celesta, the tympani and the
percussion are placed individually between these main groups. There is no aiming
for spectacular effects of the ‘tennis’ or ‘ping-pong’ type; rather a structural
disposition of the orchestra which allows what in physics is called ‘Brownian
movements’, here represented by musical elements in motion. By their ‘geography’,
the instrumental groups participate in the musical form itself . . .
Figures refers to simple elements, primarily and very sharply characterized by
such means as dynamism, violence, softness, slowness, and so forth. These elements
can be purely harmonic, or more rhythmically oriented, or something purely
melodic. They are not themes in the conventional way, but ‘states’ of musical being.
d’amateurs de stéréophonie’ (Leibowitz, Le 1969). I wish to thank Roger Reynolds for

compositeur et son double, p. 514). giving me this information (personal com-
27
‘Spectrum: New American Music’, Vol. I, munication, 9 July 2014).
28
Contemporary Chamber Ensemble, Arthur Reynolds, ‘Thoughts on What a Record
Weisberg, conductor (Nonesuch, H-71219, Records’, p. 31.
Doubles has two meanings: the first is that of the eighteenth-century word for
variation, the second is related to the German word Doppelgänger, which means
a human double, as it is so often described in Romantic literature. Thus, in the
process of development, each figure may have its double, which is related only to it
and to no other. As in my Marteau sans maître, a variation may occasionally
precede the figure, so that the element of time is not successive but moves along my
concept of the labyrinth. Similarly, one finds in the modern novel the methods of
the ‘flashback’ or, conversely, that of anticipation by going through the future to
the past.
Prismes occur when the figures (or their doubles) refract themselves one through
the other. And in this case, one figure becomes the prism, and the other is refracted
through it. By this process is obtained the maximum of complexity, and the effect
will be akin to the kaleidoscope.29
Although Boulez does use the word ‘stereophony’, he distances himself from
it by quoting his earlier text (from the programme note to the 1958
première). Rather than claiming stereophony as his own, he sets it within
quotation marks, as if to imply that although it is familiar to the point of
being commonplace in 1965, it represented a forward-looking outlook in
1958. Moreover, as soon as he uses the word ‘stereophony’, he hastens to add
the proviso that no straightforward left-right panning effects (those all-too-
familiar ‘ping-pong’ effects) of the kind associated with stereophonic
demonstration disks and trendy ‘stereophonic year’ pieces are to be found
in his work. Boulez appears to be careful in 1965 to manage listeners’
expectations in a way that he hadn’t considered necessary in 1958. He
seems to be adopting a strategy of pre-empting the kinds of objections that
were frequently advanced by critics of the first performance of Doubles in
1958, as will be discussed in the next section, when they argued that the
purported stereophonic effects could not be perceived.
Predictably, in the programme notes to the performances of the third
version of Figures – Doubles – Prismes by the Hague Residentie Orkest in
1968, the author (Hans Citroen) makes no allusion whatsoever to either the
unusual seating plan of the work or to its purported ‘stereophony’, although
it is not known whether this omission was his own choice or the result of
a consultation with the composer.30
Boulez’s apparent caginess regarding the association between the seating
plan and the stereo technology from which it was inspired persists in his
1975 published interviews with Célestin Deliège. While introducing
29
From programme notes to US première under Boulez’s direction (except in
with the Cleveland Symphony on Darmstadt, where he was replaced by Bruno
11 March 1965. I wish to thank Deborah Maderna due to illness). Citroen’s
Hefling for sharing this document with me. programme notes were obtained from the
30
In March 1968, the Residentie Orkest per- archive of the Residentie Orkest. I wish to
formed Figures – Doubles – Prismes in Paris, thank Jorien Veenhoven for retrieving this
Darmstadt, Utrecht, Schevingen and Liège document for me.
a question about the work, Deliège explains that he sees one of the work’s
‘two major preoccupations’ to be ‘orchestral stereophony’.31 Without
directly contradicting his interlocutor, Boulez omits any reference to stereo-
phony in his response, explaining simply that:
In 1958, when I decided to compose this work, I thought about modifying this
structure by separating the individual groups while leaving them a certain
autonomy, and doing so in such a way that the woodwind in particular would be
split up among different groups, and the same with the brass . . . When you hear the
work live, the sonorities are extremely homgenous [sic] yet at the same time
scattered, so that it is not a homogeneity of neighbouring groups but a homogeneity
of fusion. To that extent this new geography of the orchestra has been a success.32
The theme of stereophony seems to be gradually purged from Boulez’s

discourse on the work after 1958, perhaps as a reaction against those
trendy works that make the superficial use of stereophony that Boulez
alludes to in Penser la musique aujourd’hui. The change in attitude sig-
nalled by Boulez’s aligning himself with stereo in 1958 and then down-
playing his relationship to it in 1964 mirrors changes in the aesthetics of
recording during this period that have been recently discussed in studies
by the musicologists Jochen Stolla and Martin Kaltenecker. These scholars
have studied the way in which a ‘positivist’ or ‘realist’ orientation in the
1950s yields to a later aesthetic paradigm that exploits the audio possibi-
lities of stereophony to define an enhanced audio space without necessa-
rily claiming fidelity to any real performance (an orientation that
Kaltenecker terms ‘idealist’, best exemplified by the ‘stagings’ of stereo
that one finds in Solti’s Decca recording of Wagner’s Ring, produced by
John Culshaw). This later paradigm is itself followed by a banalisation of
stereo technology that is the result of its sheer familiarity.33 Boulez’s
apparent downplaying of the stereophonic implications of his work
might be a result of the fact that by the middle of the 1960s, stereo effects
had become familiar to the point of being trite. His avoidance of the theme
of stereo in his programme notes to Figures – Doubles – Prismes might also
be rooted in a discursive strategy of negotiation in the artistic field.
31
‘Deux préoccupations majeures’, également . . . Vous l’avez entendue en direct,
‘stéréophonie de l’orchestre’ (Boulez, Par c’était une sonorité extrêmement homogène,
volonté et par hasard, p. 130; Conversations mais dispersée, donnant non pas une
with Célestin Deliège, p. 99). homogénéité de groupes voisins, mais une
32
‘En 1958, quand j’ai eu l’intention de faire homogénéité de fusion. En ce sens, cette
cette œuvre, j’ai pensé à modifier cette géographie de l’orchestre est réussie’ (Boulez,
structure en séparant les groupes individuel- Par volonté et par hasard, p. 131;
lement tout en leur laissant une certaine Conversations with Célestin Deliège, p. 100).
33
autonomie, et en faisant en sorte que les bois Stolla, Abbild und Autonomie, p. 78;
en particulier se répartissent dans les Kaltenecker, ‘Trois perspectives sur l’image
différents groupes, et les cuivres sonore’, p. 3.
In a study of ‘the discursive construction of a wall of aesthetic difference

between multi-channel work inside and outside the electroacoustic studio’,
Patrick Valiquet has shown how with respect to electroacoustic music, ‘an
ideology of aesthetic isolation supported the otherwise contradictory work of
appropriating the tools of commercial broadcasting and recording’.34 It may
well be that Boulez, in first aligning Doubles with innovative stereophony and
then downplaying its affinity with commercial or unartistic stereophonic
enterprises, is also in the business of constructing such an aesthetic wall
around Doubles, a relatively straightforward task given that Doubles does
not itself employ electronic means.
Curiously, however, decades later, and long after the era of Boulez’s major
aesthetic battles, stereophony reappears in his discourse on Figures –
Doubles – Prismes. In an interview with Cécile Gilly in 2002, Boulez notes
with respect to the orchestral seating plan of the work that ‘This [orchestral]
arrangement is related to the idea – new at the time – of stereophony, since
the timbre is distributed across the stage in a potentially mobile fashion while
the classical orchestra is founded on fixed domains of timbres.’35 Long after
stereophony had ceased to be either a source of wonder or of contempt,
Boulez apparently finds this observation to be nearly self-evident, and offers
no further comment on it in his response.
IV. Listening for Stereophony in Doubles

If the links between stereophony and Doubles fade in and out of Boulez’s
discourse on the work and its successor, they were obvious to all sensitive
observers at the time of its first performance. One of these was none other
than Igor Stravinsky, who, on 15 March 1959, a year after the Doubles
performance, published an article on the subject of stereo LPs in the
New York Times in which he includes a passing reference to Doubles.
Although Stravinsky had not attended the première of Doubles, he had
received a letter from attendee Pierre Souvtchinsky that reported on the
concert and included a wittily annotated programme.36 In his article,
Stravinsky considers the new stereo technology, at first claiming to be
wary of the pretentions of verisimilitude made by stereo equipment
manufacturers, observing that:
34
Valiquet, ‘The Spatialisation of étant fondé sur des domaines fixes de timbres’
Stereophony’, p. 415. (Boulez, L’écriture du geste, p. 114).
35 36
‘[C]e dispositif est en relation avec l’idée, Letter from Pierre Souvtchinsky to Igor
neuve à l’époque, de stéréophonie, le timbre Stravinsky, 3 April 1958. Sammlung Igor
étant réparti sur scène d’une façon Stravinsky, Paul Sacher Stiftung, microfilm
éventuellement mobile et l’orchestre classique 103.1, film 1353.
We do not hear live performances ‘stereophonically’, therefore, and stereo, instead

of giving us ‘the best seat in the house’ gives us, in fact, a kind of omnipresent seat
not found in any house.
He concludes however with a squarely optimistic prediction with respect to

stereo technology, expressing a desire to hear antiphonic works in stereo,
whether by Mozart, Berlioz or Ives. Finally, he considers the case of certain
recent works that integrate the stereo principle into their conception:
Stereophony has already influenced composed music. At the most superficial level
this amounts to an exploitation of the stereo effect (the stereo fault, rather) by
‘building’ stereo ‘in’, creating distance and separation by reseating the orchestra, etc.
(When I listen to this sort of music I find myself looking in the direction of the
sound, as one does in Cinerama; therefore ‘direction’ seems to me as useful a word as
‘distance’ to describe this effect.) Examples of this kind of music are Stockhausen’s
‘Gruppen’ and Boulez’s ‘Doubles’. But a more profound influence of stereo will
come when composers see that they have to construct an independently interesting
‘middle dimension’.37
It is remarkable – and retrospectively predictable – that Stravinsky

hears Doubles in this manner. He was well aware that technologies of
sound reproduction (stereophony) could influence the conception of
a work, having himself succumbed to the same kind of phonograph effect
thirty-five years earlier when he composed the Serenade in A for piano.
Since Stravinsky’s familiarity with the stereophonic intentions of Doubles
stemmed only from the programme note, rather than from first-hand
familiarity with the work, he was not in a position to pass judgement on
the perceptibility of the stereo effect/fault in Doubles. Critics of the first
performance of Doubles were, however, able to weigh in on the success of
Boulez’s stereophonic effort. In the pages of Paris’s newspapers and music
journals, most of which included reviews of the Doubles performance, critics
were divided as to the perceptibility of the announced stereophonic effects.
Writing in Le Monde, René Dumesnil confessed that ‘I did not experience
any particular sensation of “stereophony” announced in the programme.’38
Marc Pincherle, writing in Les Nouvelles littéraires, assumes that the unusual
seating plan was used in order to ‘obtain a stereophony adequate to the
economy of the work (the horns in front, in two groups facing each other, the
double basses in three groups, two on the sides and the other facing the
audience, etc.)’, but admits that ‘I only grasped the exterior, the play of
37
Stravinsky, ‘New Sound as Stravinsky not inconceivable that the article was written
Hears it: Stravinsky and Sound’. This article by or co-written with Robert Craft.
38
may have been conceived as a promotional ‘Je n’ai aucunement éprouvé une sensation
tool, since it was published only two months particulière de “stéréophonie” annoncée par
after the recording sessions that Stravinsky le programme’ (Dumesnil, ‘Pierre Boulez –
made for the stereo LP of his Threni. It is also Peter Frankl – Georges Prêtre aux concerts
Lamoureux’).
sonorities’.39 Jean Hamon, in Combat, notes Boulez’s reference in the pro-

gramme notes to stereophony, but admits that he could not perceive it;
instead, he writes that ‘for us, it amounted to hearing a sort of short poem in
two parts’,40 a comment that must have caused him embarrassment when
irate readers reminded him that, far from being in two parts, the work had
simply been performed twice in succession at its première, a fact he claimed
to have been aware of in a later article.41 Similarly, after the UK première of
Figures – Doubles – Prismes in 1964, Hans Ulrich Lehmann observed in the
pages of Tempo that:
The doubly symmetrical lay-out of the six groups . . . permits all sorts of
stereophonic effects. These, however, were clearly perceptible only to listeners in the
front rows of the hall, for which failing the small size of the stage may well take the
blame.42
The stereophonic reading of Figures – Doubles – Prismes continues among

critics following successive performances of the work and its newer
versions. After describing the seating plan of Doubles, Marcel
Schneider, in an article on musical life in France that was published in
an Italian journal in 1959 and in which he reiterated ideas he had
expressed in a concert review published in Combat,43 sees the stereo-
phony in Doubles as a means of representation of the disembodied
Doppelgänger of the title:
In order to create a sort of stereophony, the composers then had the idea to place the
instruments in a particular way . . . Not that this is a new idea, but Boulez used it in
a felicitous way, creating with it a disquieting and mysterious atmosphere that
justifies the title Doubles, corresponding to the Doppelgänger of Schubert and
Schumann.44
39
‘. . . obtenir une stéréophonie adéquate à in maniera particolare . . . Non è del resto
l’économie de l’œuvre (les cors au premier un’idea nuova, ma Boulez l’ha saputa
plan, en deux groupes se faisant face, les impiegare con un felice risultato, e ha in tal
contrebasses en trois groupes, deux latéraux, modo creato quell’atmosfera inquietante
l’autre face au public, etc.)’; ‘je n’en ai saisi e misteriosa che giustificava il titolo Doubles,
que l’extérieur, le jeu des sonorités’ corrispondenti ai Doppelgänger di Schubert
(Pincherle, ‘La musique’, p. 10). et di Schumann . . . Ci stanno di fronte due
40
‘. . . il s’agissait pour nous d’entendre une elementi circolari: ora il tema è fisso mentre
sorte de très court poème en deux parties’ i contrappunti che passano da uno strumento
(Hamon, ‘Propos sur Doubles de Pierre all’altro gli circolano intorno, ora sono
Boulez’, p. 2). i contrappunti che restano fermi mentre il
41
Hamon, ‘Toujours à propos de Doubles de tema, ripreso successivamente da tutti
Pierre Boulez’, p. 2. i diversi timbri strumentali, compie una
42
Lehmann, ‘First Performances: Boulez’s metamorfosi sia sonora sia visiva, dato che lo
Figures Doubles Prismes’, p. 34. si vede vagare come un “double” come un
43
Schneider, ‘La musique à Paris’, p. 3. fantasma, per tutta l’orchestra’ (Schneider,
44
‘Per creare una sorta di stereofonia l’autore ‘La vita musicale all’estero: Francia’,
allora ha avuto l’idea di disporre gli strumenti pp. 173–4).
Other critics described the first performance with no mention of stereo-

phony at all, not considering it essential to a discussion of the work.45
V. Stereophony in the Score

Beyond the sedimented layers of discourse on the work, in what way is
stereo inscribed in the score of Doubles? Are stereo effects perceivable to
the ear or visible to the eye that views the stage or the score? The seating
plan is instructive in this respect. Boulez divides the orchestra into three
groups of winds, four groups of brass and five groups of strings.
In addition, percussion instruments and three harps are laid out across
the stage. The sketches Boulez made for the seating plans of the work (one
of which is shown in Figure 10.1) show how fundamental the symmetrical
Fig. 10.1 Sketch for seating plan of Doubles, transcribed and translated46
45
Cf. for example Bourgeois, ‘La révolution Figures – Doubles – Prismes (1957–1958); 3)
de Pierre Boulez a fait long feu à Lamoureux’, Orchesterpläne. Microfilm 0584–0356, 0357
p. 9. and 0362.
46
From Paul Sacher Stiftung, Sammlung
Pierre Boulez, Mappe H, Dossier 3a, 3
Fig. 10.2 Final seating plan of Figures – Doubles – Prismes, transcribed

and translated, with geometry of woodwinds and brass instrumental
groups indicated
plan was to the work. Boulez was committed to constructing a seating

plan with a double axis of symmetry – front-to-back and side-to-side.
Figure 10.1 shows how the stage is divided into four quadrants marked
‘avant-droite’ (front-right), ‘avant-gauche’ (front-left), ‘arrière-droite’
(rear-right) and ‘arrière-gauche’ (rear-left). The instruments are laid
out symmetrically across the left–right axis (a group of fourteen
strings, one harp and one brass group can be found on either side) with
approximate symmetry along the front–back axis (the latter group is set in
opposition to a group composed of twelve strings and one group of brass
instruments, with no harp), while the three groups of winds line the
central front–back axis. The seating plan Boulez ultimately opted for
in the published version of Figures – Doubles – Prismes retains much of
this symmetrical plan, but further separates distinct timbral groups;
for example, the strings and brass are now separated in space. This
final seating plan, in which the woodwind groups form a triangle inscribed
within a rectangle formed by the four brass groups (Figure 10.2), was
later characterised by Boulez as ‘a single orchestra with groups placed
symmetrically: the woodwinds in the centre, the brass at the edges and
the strings in the interstitial space, with the percussion in the back’.47 With
this spatial canvas laid out, Boulez was able in his sketches to tag musical
ideas with one of four labels: ‘GaV’ (i.e. ‘gauche avant’ or front-left), GaR
(‘gauche arrière’ or rear-left), DaV (‘droite avant’ or front-right) and DaR
(‘droite arrière’ or rear-right).
With regards to the way in which musical objects are projected in
space, only two studies to date have been published on Doubles or its
successor Figures – Doubles – Prismes, the first a 1993 journal article by
Allen Edwards and the other a 2006 book chapter by Pascal
Decroupet.48 These two pioneering and complementary articles do not
emphasise the spatial aspects of the work, instead focusing on the sketch
material – the matrices used to derive pitches and rhythms.
In discussing the spatial arrangement, Decroupet does nevertheless
allude to ‘a form of “stereophony’’’ – a word he places in quotation
marks – ‘in the form of serialized Brownian motion’. He also provides
a short description (p. 145) of three types of stereophonic sound motion
that can be found in the score, the first in which a chord is held by
a single group, another in which a chord is passed from one group to
another, and a third, intermediate case. These three examples are taken
from the first pages of the score, in the deployment of the figure that
Boulez refers to in his sketches as the ‘thème lent’. This slow theme is
set in opposition in Doubles to a fast theme in a kind of antiphony
familiar from later works by Boulez, for example Rituel (1975).49
The slow theme, like its fast counterpart, is derived from serial tables
elaborated for the Third Sonata for piano, using the familiar procedure
of the partition of a twelve-note theme followed by pitch-class set
multiplication and (sometimes) transposition.50 This slow theme is itself
made up of six slow chords (‘accords lents’) – lettered from A–F in
Figure 10.3 – each composed of 1–3 sound blocks. Figure 10.3 shows the
serial derivation of the slow theme and its segmentation into six por-
tions (with one exception – the penultimate chord containing an
E where the table indicates the presence of a B). The first stave of
47 48
‘C’est un seul orchestre avec des groupes Edwards, ‘Boulez’s “Doubles” and “Figures
qui sont disposés symétriquement : les bois Doubles Prismes’’’; Decroupet, ‘Moments
sont au centre, les cuivres à la périphérie et les doubles, figurés en prisme’.
49
cordes dans l’espace interstitiel, la percussion See Edwards, ibid., p. 7, schema.
50
étant placée tout au fond’ (Boulez, L’écriture See Decroupet, ‘Moments doubles, figurés
du geste, p. 114). en prisme’; and, more generally, Losada,
‘Isography and Structure in the Music of
Boulez’ on pc-set multiplication.
Fig. 10.3 Serial derivation of six segments (accords lents) of slow theme
(thème lent)
Figure 10.3 contains the partitioned original series, the second the
product of the multiplication of this series by the second to last
chord. This theme first appears in interrupted fashion (rehearsal figures
0–2, 8–10 and 11 in the most recent Figures – Doubles – Prismes
score51) throughout Doubles, with very long durations, creating a kind
of broad harmonic canvas. In order to describe the spatial movement of
this theme, Figure 10.4 indicates which instrumental groups play
the notes of each of the chords, whose location is then plotted on the
stage.
The instrumental groups shown in Figure 10.4 are far from the only
ones sounding in these passages. The other groups play figures derived
from three compositional procedures that Boulez developed and
which Edwards terms ‘superstructures’ and Decroupet ‘structures envel-
oppantes’. Boulez names these three procedures ‘canons d’intensité’
51
Universal Edition, 1964.
Fig. 10.4 Instrumental groups used for each constituent chord of the thème lent
[Slow chord] [Superstructural procedure] Stéréo

A canons d’intensité fixe
B mélismes (récitatif) fixe/tournant
C accords complémentaires (+ mélismes tournant
d’accords)
F accords complémentaires (+ mélismes fixe/tournant
d’accords)
E canons d’intensité + Echo F fixe
D mélismes (récitatif) + Echo F E tournant/tournant
Fig. 10.5 Transcription of Boulez’s sketch for spatialisation and superstructural

procedure in Doubles52
(which present a series of chords using a serialised arrangement of

dynamic markings), ‘accords complémentaires’ (that derive new chords
from the notes found outside a frequency band derived from the voicings
of the chords of the slow theme) and ‘mélismes (récitatif)’ (‘linear figura-
tions constructed upon a segment of the slow theme’,53 and each is the
source of new sonorities, figures, arpeggios and pitch collections in other
instrumental families, providing foreground interest to the distended
slow theme in the background.54 Figure 10.5 is the transcription of
a sketch that shows how Boulez planned to assign one structure envel-
oppante to each chord of the slow theme.
In contrast to the spatially mobile figures of the slow theme shown in
Figure 10.4, in the fast theme Boulez seemingly relies more on the differ-
entiation afforded by different instrumental groups than on that created by
spatialisation per se; in general, either all three of the woodwind groups are
employed in a chord or none of them is, either all four brass groups or none
of them, while string ensembles, though more variable, are chosen to sound
from stage left, right and centre simultaneously (e.g. group 1 – left; group 2 –
right; and group 5 – centre – see Figure 10.6). Sound is in general spaced
evenly across the stage and only the timbre changes starkly from chord to
chord. Whereas Boulez uses spatial coordinates to differentiate figures in the
52
0584–0389: Mappe H, Dossier 3b, 1b, words echo procedures he himself had put
Figures – Doubles – Prismes (Orch; 1963; 2. into practice in Figures – Doubles – Prismes:
Fassung). ‘Il me semble que l’on a à faire à des “fonds”
53
Edwards, ‘Boulez’s “Doubles” and “Figures travaillés d’une façon extrêmement raffinée et
Doubles Prismes”’, p. 10. que par moment on voudrait des “figures”
54
Boulez’s approach to the orchestral incrustées sur ces fonds pour leur donner une
foreground and background in Figures – signification’ (‘It seems to me that it has an
Doubles – Prismes is nicely captured in advice extremely finely wrought “background” and
he offers Karlheinz Stockhausen in a letter at times one would hope for “figures” set
dated 15 July 1966. Boulez expressed against this background that would give them
criticism about the latest version of the meaning’) (from the Pierre Boulez Archive,
latter’s composition Punkte, which he was University of Montreal).
considering conducting in Helsinki. His
Fig. 10.6 Reduction55 of first seven bars of first occurrence of fast theme in Figures –
Doubles – Prismes (rehearsal 3) and the instrumental groups used for each of its consti-
tuent chords; pitches and durations only
slow theme, he uses timbral coordinates to differentiate the figures of the fast
theme with which it alternates.
VI. Concluding Remarks

With the spatialisation of the chords of Doubles, does Boulez attain the
stereophonic ‘delights of cinerama’ that he decries in others? Probably
not, since he enriches the orchestral texture with the three superstructural
procedures that result in a great variety of sounds heard elsewhere on the
orchestral chessboard: these superstructures obscure the spatial movement
of the theme illustrated in Figure 10.5, precluding much of the perceptibility
of spatial sound movement, let alone ‘ping-pong effects’. As a result, the
possibility of hearing the chords of the thème lent moving through space is
55
Transcribed from Edwards, ‘Boulez’s
“Doubles” and “Figures Doubles Prismes’’’,
p. 10.
diminished by the presence of other sounds in other instrumental groups.

In the end, it is unlikely that a typical listener would hear a ‘stereophonic
effect’ in the sense of musical figures consistently moving through space,
although here and there, such effects can be perceived (e.g. the first sonority
of the thème lent, that travels from stage right to stage left).
Boulez’s 1985 recording of Figures – Doubles – Prismes is instructive in
this regard.56 It is probable that Boulez was partially responsible for the
sound design of this recording, which was engineered by John Rushby-
Smith. To what extent does this recording employ the stereo field of the
recording to translate the spatial aspects of this work? How is a work that
integrates stereophony into its conception remapped into stereophonic
space in the recording? It is safe to say that the recording strives to offer
a listening experience similar to one enjoyed by a concert-goer – a ‘realist’
rather than an ‘illusionist’ goal. As a result, since the instrumental groups
are seated in close proximity on the stage, audible traces of left-right
movement are only subtly rendered in the stereo mix. Certainly, the stereo
field is not used in this recording to accentuate spatial aspects of the music,
for example by setting different ensembles at a greater distance than that
afforded by a concert stage, let alone through the use of studio-engineered
panning effects. This is not to say that the left-right movements are, as
with Roger Reynolds’ experience, downplayed or effaced entirely, but they
are not electronically accentuated in the way that they were in Boulez’s
1998 recording of Répons.57 But of course, the comparison is unfair, since
the latter work features soloists placed in a circular plan around the
audience, each at a considerable distance from each other; as a result,
the concert-goer’s experience would involve more pronounced spatial
effects than those of Figures – Doubles – Prismes, and the (‘realistic’)
recording, consequently, also displays a correspondingly greater use of
the stereo field.
Boulez would later become aware of the imperceptible nature of his sonic
renderings of ‘Brownian motion’. In a 2014 interview, he wondered about
the viability of his earlier approach to space in this work and others of the
period, in which he strove to transpose structures developed for the organi-
sation of pitch onto the spatial dimension.58 After Doubles, he would go on
to compose works employing spatialised ensembles, like Domaines, Rituel
and Répons, in which the spatial plan is obvious visually and the individual
56 57
Pierre Boulez conducts the BBC Symphony Ensemble Intercontemporain (Pierre
Orchestra, Erato 2292 45494–2, 4509 Boulez conducting), Deutsche Grammophon
98495–2, 8573 84248–2; reissued on disc 5 of 0289 457 6052 0 GH.
58
the box set of Boulez’s complete works, Interview by the author with Pierre Boulez
Deutsche Grammophon 0289 480 6828 9. on 9 June 2014 in Baden-Baden.
ensembles are widely spaced, making them audibly localisable in space.

But it could be that the elusive use of spatial effects in Doubles, so remarked
upon by contemporary critics, can be viewed in terms of the conflicting goals
of modernist composers. This elusiveness would come as no surprise to
listeners of later versions of the work who had read Penser la musique
aujourd’hui, since, after dismissing the Venetian antiphonal composers as
‘the most decorative of composers’, Boulez states, in what might be read as
a ‘reductio ad Xenakum’, that ‘the abuse of such glissandi of space seems to
me to be founded in an aesthetic as summary as the immoderate use
of clusters, glissandi or other white noise’.59 It would nevertheless be
tempting to explain the simultaneous presence and absence of stereo-
phony in Doubles as an example of ‘practical denial’ that Pierre Bourdieu,
inspired by psychoanalysis, invokes in order to understand behaviours
that ‘can only do what they do by pretending not to be doing them:
defying ordinary logic, these double practices lend themselves to
two opposite readings, both equally false’.60 Avant-garde composers
aligned with technological progress always fear that their use of any
given technology (or, as in Doubles, its metaphorical evocation) will be
understood as a simple ‘effect’ or ‘gimmick’ that will soon be rendered
obsolete along with the machinery it exploits or evokes. Karlheinz
Stockhausen displays this uneasiness in a particularly clear way when
speaking about his seminal work Gruppen (1955–7), a work which, like
Doubles, employs spatialised instrumental groups. Gruppen is in a sense
the ‘double’ of Doubles, and, although composed earlier, was premièred
only eight days after the Doubles performance in Paris, with Boulez
serving as one of the three conductors.61 Commenting on his famous
multi-orchestra piece, Stockhausen wishes to discourage interpretations
of his work that invoke electronic ‘effects’:
As far as the orchestral sound of the groups as such is concerned, it was obtained
from the typical facilities of the selected instruments, and it is wrong to speak of
a ‘translation of electronic sounds into orchestral ones’ as has so often been done in
the last few years: the ‘Gruppen’ were written for a particular orchestra, and their
orchestral sound is the result of particular laws of a functional application of this
instrumentation.62
59
‘L’abus de tels glissandi d’espace me paraît comme si elles ne le faisaient pas : défiant la
relever d’une esthétique aussi sommaire que logique ordinaire, ces pratiques doubles
l’emploi immodéré de clusters, glissandi et prêtent à deux lectures opposées’ (Bourdieu,
autres bruits blancs . . .’ (Boulez, Penser la ‘La production de la croyance’, p. 4).
61
musique aujourd’hui, p. 73; Boulez on Music Gruppen was premièred on 25 March 1958
Today, p. 66). in Cologne by the Cologne Radio Symphony
60
‘. . . la classe des pratiques . . . qui, fonc- Orchestra with Stockhausen, Boulez and
tionnant comme des dénégations pratiques, Bruno Maderna conducting.
62
ne peuvent faire ce qu’elles font qu’en faisant Stockhausen, ‘Music and Space’, p. 70.
It may be that composers like Stockhausen and Boulez fall more or less
consciously into a sort of double musical discourse in which they refer to
technology while at the same time purporting to avoid such references in the
name of the presumed autonomy of the musical work. Boulez announces the
use of ‘stereophony’ in the programme note to Doubles, but then offers little
that corresponds for listeners of the day to the experience they might have
had of this new technology, of what Stravinsky describes as ‘an exploitation
of the stereo effect (the stereo fault, rather) by “building” stereo “in”’.63
Boulez shows himself to be deeply ambivalent with respect to the idea of
stereophony: he does not want the work to be perceived simply as ‘Boulez’s
stereophonic piece’, just as Gruppen might well be thought of as
Stockhausen’s stereo endeavour. At the same time, as an avant-garde artist
invested in the aesthetic consequences of technological innovation, Boulez
is eager to incorporate the lessons of new sound technologies into his
works, both those used exclusively in specialised electronic music studios
in Cologne, Baden-Baden or Paris, and those that have passed over into
the domestic or mass-entertainment markets. In conclusion, is there
a phonograph effect in Doubles or not? Was Doubles composed in stereo
or was it listened to in stereo? Perhaps, like the first stereo LPs, Doubles can
be played on mono or stereo equipment without causing damage to your
needle.64
63
Stravinsky, ‘New Sound as Stravinsky figures of this chapter, and Pierre-Arnaud Le
Hears it’, p. 14. Guérinel for help in finding concert reviews
64
I wish to thank Julie Delisle for her of Doubles.
tremendous help in the preparation of the
11 Composing an Improvisation at the Beginning
of the 1970s
Paolo Dal Molin
To Robert Piencikowski
Two months after the death of Igor Stravinsky in April 1971, on the occasion
of a concert at St John’s Smith Square in London, Pierre Boulez met David
Drew, who had recently been named director of the journal Tempo, which at
that time was produced by Stravinsky’s publisher Boosey & Hawkes. On the
following day Drew addressed a letter to Boulez, inviting him to write
a canon in memory of the Russian composer:
It would mean a great deal to all concerned if you would consent to compose
for TEMPO a ‘Canon in Memory of Igor Stravinsky’. It would be reproduced
from your autograph and published – together with memorial canons by
composers of various nationalities – in a special supplement to
the September 1971 issue of TEMPO. The canon may be either vocal or
instrumental; but, if instrumental, it should be either for string quartet (or
instruments therefrom), as in Stravinsky’s ‘Double Canon’, or for flute, clarinet
and harp, as in his ‘Epitaphium’.1
If an anecdote reported by Joan Peyser is to be believed, Boulez began

composition of the new piece a few weeks later, inspired by an improvisation
he had heard in a Scottish castle:
I began to think about the work in August 1971, soon after receiving the
commission. That month I visited a castle in Scotland that had once belonged
to the Duchess of Argyll. The woman who invited my sister and me was an
Austrian who lived in France, and she had with her a son who was not very
oriented; he did not know what he was to do with his life. Since then he
committed suicide. The young man played the flute as an amateur and he
improvised in this empty eighteenth-century castle. It was quite impressive.
I had the idea then of the work beginning with a flute solo. The notes were to
provide the basic idea, a kind of ground music. Then I wrote a simple text
[Boulez followed two pages of musical notes with six pages of verbal
1
I greatly appreciate the generosity of Pierre Letter from David Drew to Pierre Boulez of
Boulez, Universal Edition (Vienna) and the 8 June 1971 (Paul Sacher Stiftung, Sammlung
Paul Sacher Stiftung in kindly authorising the Pierre Boulez, hereinafter ‘PSS’).
publication of extracts from Pierre Boulez’s
manuscripts.
270
instructions] indicating how to make it more complex. My student, Heinz

Holliger, realized the piece but his realization was not sufficient. I wanted one
much more refined from this matrix. When I did it myself, I did it in a very
complex way, with a definite form, far too complex to improvise. I have
rewritten some parts twice, others three times.2
The manuscript was not completed until mid-November.3 Between

times, as Drew had anticipated, number 97 of Tempo – entitled Igor
Stravinsky 17 June 1882 – 6 April 1971 – was published, accompanied
by a musical supplement In Memoriam Igor Fedorovich Stravinsky:
Canons and Epitaphs. Set 1. Boulez’s homage appeared subsequently
in the following number of the review as part of a second supplement
In memoriam Stravinsky: Canons & Epitaphs. Set 2, with ‘contributions
from America and France, and . . . the Requiescat by Elisabeth
Lutyens’.
Boulez’s contribution was presented in the form of a ‘score’ followed
by six pages of instructions (with translation on the opposite page by
Susan Bradshaw), intended for improvisation or the composition of
a piece of instrumental music that composers and performers could
realise by themselves.4 Indeed, beyond Boulez’s own realisations of the
material, a number of other musicians have produced versions, notably
Heinz Holliger, Jürg Wyttenbach, Klaus Stichweh, Karl and Margaret
Kohn, Johannes Schöllhorn and the Ensemble Avantgarde. The title, . . .
explosante-fixe . . ., is taken from André Breton, and, as Boulez revealed,
it ‘remains in my memory completely isolated from its context, but
coincides precisely with the idea of the piece’.5 Its re-emergence as the
title for Boulez’s tribute may even have been prompted by Drew when
he wrote to Boulez, ‘the canonic idea can be exploded, boxed in, or
ignored as you wish’.6
2
Peyser, Boulez, p. 238. Some sketches are number of flutes and clarinets prescribed for
written on notepaper headed ‘Inveraray Stravinsky’s Epitaphium (letter to David
Castle’ (cf. PSS, Mappe J, Dossier 1a, 1). Drew of 16 November 1971, PSS).
3 5
Cf. the letter from Pierre Boulez to David Programme note for the world première
Drew of 16 November 1971, which accom- of . . . explosante-fixe . . . (1972–4). The title
panied Boulez’s submission (PSS). provoked interest immediately among critics,
4
The instructions were organised in three interpreters and commentators.
parts: ‘Structure, Order’ (page [1]), The opposition of the two component parts
‘Interpretation, Play’ (pages [1]–[3]) and of the title ‘explosante-fixe’ has been related
‘Presentation, Instrumental Choice’ (page equally to the genesis of the piece in 1972–4,
[4]), to which a note concerning the ‘total to the conception of its ‘components’ (size
number of possible transpositions’ was of ensemble, form, etc.), to its electronic
attached (page [5]). A never realised proposal transformation, or again to its later
for the instrumentation was added to it (page elaborations.
6
[6]), with ‘suggested instrumentation’ for Letter from David Drew to Pierre Boulez of
seven players, ‘inadvertently’ doubling the 23 August 1971 (PSS).
272 Paolo Dal Molin
‘ . . . un texte à faire proliférer . . . ’

Written on a single page of manuscript paper, the score of . . . explosante-
fixe . . . (1971) is in reality musical material (a musical matrix) divided up
into seven ‘sequences’: an ‘Originel’ surrounded by six ‘Transitoires’
numbered from II to VII.7 Each sequence consists of several objects
distributed, according to their composition, within a different number
of staves, from the single stave A of ‘Originel’ to the seven staves
ABCDEFG of ‘Transitoire VII’. Objects are of three types, which Boulez
designates as follows: ‘notes encadrées’, ‘notes encerclées’ and ‘figures’, and
which Susan Bradshaw translates respectively as ‘notes enclosed in boxes’,
‘notes enclosed in circles’ and ‘figures’. They are referred to in this chapter
as boxed groups of notes (or boxed objects), encircled notes/groups of
notes (or encircled objects) and figures. According to the composer’s
draft,8 the sequences were generated in numerical order, following the
diagonals shown in Table 11.1. Diagonals are orders of staves whose
objects share different, remarkable common properties (see below).
The first of them, labelled ORa . . . g, is the series of final staves of each
sequence, starting from ‘Originel’, that is ‘Originel’ A (ORa), ‘Transitoire
II’ B (IIb), ‘Transitoire III’ C (IIIc), and so on up to ‘Transitoire VII’
G (VIIg); diagonal IIa . . . f is formed from the penultimate stave of the
sequences (except for ‘Originel’ which consists of only one stave), that is:
‘Transitoire II’ A, ‘Transitoire III’ B, ‘Transitoire IV’ C, and so on up to
‘Transitoire VII’ F; etc.
‘Notes encadrées’
The seven boxed groups of notes in . . . explosante-fixe . . . (1971) are
composed of three elements following a typically Boulezian, stylised
profile of ‘anacrusis-accent-inflexion’.9 A principal note (indicated by
a black head extended by a horizontal line) is preceded by a one-note
7
Boulez has designated it in a number of Dal Molin, ‘Introduction à la famille
ways over the years: ‘Kern’ (‘nucleus’), d’œuvres “. . . explosante-fixe . . .”’,
‘Kernstück’, ‘Modell’ (Häusler, ‘Gespräch mit pp. 251–6.
8
Pierre Boulez’, p. 28); ‘texte à faire proliférer’, PSS, Mappe J, Dossier 1a, 2.
9
that is ‘a text to serve as a basis for prolifera- ‘Anacrusis’, ‘accent’ and ‘inflexion’ or
tion’ (Boulez, Par volonté et par hasard, ‘decay’ correspond to the French ‘anacrouse’,
p. 136; Boulez, Conversations with Célestin ‘accent’ and ‘désinence’ employed by Vincent
Deliège, p. 104); ‘matrix’ (cited by Peyser, d’Indy and then used by Olivier Messiaen,
Boulez, p. 238), etc. A transcription of the Boulez and others with different purposes.
sequences with critical notes is proposed in
Table 11.1: . . . explosante-fixe . . . (1971). Diagonals: constituent orders of staves from ‘Originel’ and ‘Transitoires’ II–VII
→ Sequences
‘Originel’ ‘Transitoires’
OR II III IV V VI VII Diagonals’ labels
↓ Staves ⌧
a a a a a a a
↘ ↘
⌧
b b b b b b VIIa . . . a
↘ ↘
⌧
c c c c c VIa . . . b
↘ ↘
⌧
d d d d Va . . . c
↘ ↘
⌧
e e e IVa . . . d
↘ ↘
⌧
f f IIIa . . . e
↘ ↘
⌧
g IIa . . . f
↘
ORa . . . g = ORa, IIb,
IIIc, . . . and
VIIg
274 Paolo Dal Molin
‘anacrusis’ (a quaver grace note linked to the principal note which follows
it) and concluded by an ‘inflexion’ of from one to seven notes which are
written as unmeasured quavers or demi-semiquavers and marked either
staccato or legato. For any given box, the pitch of the ‘anacrusis’ corre-
sponds to that of the principal note in the preceding box – hence the
absence of an ‘anacrusis’ in ‘Originel’; the pitches of the ‘inflexion’ are
taken from the principal notes in both its own box and in those of the
previous sequence. As Boulez’s instructions prescribe, these seven boxed
groups should last for between two and thirty-four durational units,
conforming to the numbers of the Fibonacci series, and each one should
be characterised by a combination of specific modes of sound production
(to which the encircled letter ‘n’ refers), evolving from flutter-tonguing
(or tremolo) to regularly or irregularly repeated notes, according to seven
different possible pathways.
‘Notes encerclées’
The encircled objects are formed from a principal note (indicated by
a black head followed by a long horizontal line) to which one or several
secondary notes can be adjoined: a different type of secondary note
corresponds to each diagonal and their number increases progressively
from one stave to another. For a given encircled group of notes, the
pitches follow on, in principle, from those of the encircled objects
that precede it within the same diagonal. While the composer gives no
indication as to their arrangement, it seems clear nevertheless that the
secondary notes can be permutated and that they can be played together,
before or after the principal note or even interrupt its prolongation. In the
matrix, one possible layout for these objects is given by the composer, but
only as an example. As for duration, Boulez requests that the global value
within a sequence should vary from two to eight units according to the
arithmetic progression 2–3–4 . . . 8, but he does not establish selection
principles for sequences with less than seven encircled objects. Finally he
connects to ‘Transitoires’ II–V, III–IV and VI–VII three alternative pairs
of indications relative to the sound of the principal note, which will be
stable or modulated.
‘Figures’
The figures are composed of from one to seven notes. They are written in
terms of three rhythmic values – demi-semiquaver, semiquaver and dotted
semiquaver – which can be multiplied by 2/3, 3/2 or by 2, and are

marked staccato, legato or tenuto; the latter two modes of sound
production can be combined with flutter-tonguing or tremolo, depend-
ing on the instrument used. In the A staves, where all the notes in the
figures are written staccato, one note is also accented: its pitch corre-
sponds to that of the encircled note placed on the same stave. However,
in the realisation, the constitutive parameters of any figure (pitches,
durations, modes of attack) can be separated. As with the encircled
groups of notes, only one possible form is given in the score, while
from the point of view of its conception, each figure is a more general
non-ordered object. The density and the number of figures vary
from one stave to another according to three concurrent principles:
(1) The number of notes per figure is, in principle, constant within
a stave and for all the staves belonging to the same diagonal. It amounts
to two notes in the diagonal ORa . . . g, and increases each time by one
unit up to seven notes (VIa . . . b), reducing to one in VIIa . . . a.10
(2) In terms of the number of figures per stave within a given diagonal,
this begins with one and increases, with each new stave, according to
the succession 1 3 4 5 7 10 13.11 (3) The one-note figures are comple-
mentary objects, an aspect which appears evident when the pitch con-
tent of the A staves is considered. As Boulez indicates in the
instructions, ‘each of these figures, especially those comprising 4, 5, 6,
7 notes, may undergo, ad libitum, modifications of its speed, its basic
dynamics, and its technique of sound production’, for which he himself
provides some suggestions.
‘Basic dynamics’
For each sequence, the dynamics placed to the left of the respective
staves form a series beginning with pppp and progressing up to f:
pppp ppp pp p mp mf f. The same series is found, inverted, in the
dynamics placed to the right of the same staves. Thus, the dynamic
range within the matrix increases from pppp (‘Originel’) to pppp–f
(‘Transitoire VII’), in parallel with the increasing number of staves.
Furthermore, within a given sequence, the dynamics and density of the
10 11
The A staves of the four first sequences Exception is made however for staves
(‘Originel’ to ‘Transitoire IV’) stray from this B from ‘Transitoires’ IV and VI, since these
principle in that they present equally one or contain two figures instead of three.
several figures made up of a single note.
276 Paolo Dal Molin
figures are therefore connected: all of the figures having the same
density will have the same dynamic; to the progressive increase of the
first corresponds a progressive augmentation or diminution of
the second. In the instructions for the realisation of the matrix, these
nuances are called ‘basic dynamics’ (‘dynamiques de base’), since they
can be modified freely with the application of a crescendo and/or
decrescendo.
Pitches
The matrix uses a twelve-pitch tessitura unfolding the twelve-tone
collection within less than two octaves, from C♮4 to B♭5 (see
Example 11.1).12 The pitches of the encircled notes and of the figures
belonging to the seven diagonals are included in seven sub-collections
(one per diagonal) of the general tessitura. Each sub-collection corre-
sponds to seven transpositions of the heptachord P = {D♮ E♭ E♮ G♮ A♭
A♮ B♭}. Within the upper stave of each sequence (except for VIa),
the pitches of the encircled notes and of the figures reduce to six
elements of the associated transposition of P (TnP), the seventh ele-
ment being reserved for the principal note of the boxed group within
the same stave. The seven referential heptachords TnP enter and accu-
mulate in the sequences according to the series P = ‹E♭ G♮ E♮ B♭ A♮ D♮
A♭› (see Tables 11.2 and 11.3). However, the pitch collection assigned
to VIa . . . b, corresponding to T11P, is modified with the pitch E♮5 in
place of E♭4, and other exceptions entail the addition of pitches G♮4
and E♮5 in some figures belonging to diagonals IIa . . . f and IIIa . . . e
respectively (see ‘Transitoires V’, ‘VI’ and ‘VII’ in the published
matrix).
Ex. 11.1 . . . explosante-fixe . . . (1971). General tessitura
12
In the matrix published by Tempo the contain an E♭5, in place of E♭4, which Boulez
encircled groups of notes from ‘Transitoire has corrected afterwards.
VI’, stave B, and ‘Transitoire VII’, stave C,
Table 11.2: . . . explosante-fixe . . . (1971). Transposition of the

referential heptachord P associated with the seven diagonals and
number of notes per figure within each diagonal
Transpositions of P ↓P Number of notes per figure
Diagonals
ORa . . . g T0P E♭ 2
IIa . . . f T4P G♮ 3
IIIa . . . e T1P E♮ 4
IVa . . . d T7P B♭ 5
Va . . . c T6P A♮ 6
VIa . . . b T11P* D♮ 7
VIIa . . . a T5P A♭ 1
Table 11.3: . . . explosante-fixe . . . (1971). Transpositions of the

referential heptachord P associated with the constituent staves of the
seven sequences and the number of notes per figure within each stave
Sequences Transpositions of P Number of notes per figure
staves
‘Originel’
a T0P 2, 1
‘Transitoire II’
a T4P 3, 1
b T0P 2
‘Transitoire III’
a T1P 4, 1
b T4P 3
c T0P 2
‘Transitoire IV’
a T7P 5, 1
b T1P 4
c T4P 3
d T0P 2
... ... ...
‘Transitoire VII’
a T5P 1
b T11P* 7
c T6P 6
d T7P 5
e T1P 4
f T4P 3
g T0P 2
278 Paolo Dal Molin
Each group of notes from the matrix deploys a subset of the corresponding
referential collection without repetitions, except for the seven boxed notes and
the four encircled notes of the diagonal IVa . . . d. Thus, the density (i.e. the
cardinality) of a given object and that of the deployed pitch subset coincide,
and increase from one stave to another in parallel with the accumulation of
TnP. Each two-note figure in the matrix (ORa . . . g) corresponds in fact to
a two-note subset from P (T0P), each three-note figure (IIa . . . f) to a three-
note subset of T4P, each four-note figure (IIIa . . . e) to a four-note subset of
T1P, and so on. For the same sequence, it follows therefore that not only the
basic dynamic, and the density of the objects, but also their referential collec-
tions vary simultaneously from one stave to another.
In other words, from the point of view of the generation of the matrix, the
figures within a given diagonal are obtained through the selection of
a constant number of different elements from the referential collection
TnP which are associated with it. Within the same diagonal, the encircled
objects of notes are given a progressively greater number of elements from
the same collection, taken in the order of the corresponding serial form
TnP (see Tables 11.4 and 11.5). Similarly, the boxed groups of the A staves
from ‘Originel’ to ‘Transitoire VII’ progressively spell out the basic hepta-
chord P in the order of P (see Table 11.6).
The composer’s first realisation: . . . explosante-fixe . . .

(1972–4)
Boulez evidently designed . . . explosante-fixe . . . (1971) envisioning a piece
for seven instrumental groups with a monodic instrumental part embedded
within it. It is noteworthy that some of the work’s features appear previously
in one of his projects from the 1960s, Marges: the reinterpretation of the
canonic idea, the de/re-synchronisation of interdependent instrumental
parts and the layout of the matrix.13 When the composer received the
proposal from Drew, he was already commissioned to write two instrumen-
tal pieces: one for the Chamber Music Society of the Lincoln Centre14 and
another, prior to that, for the California Institute of the Arts, the latter
requiring ‘a small combination: between 8/10 musicians – maybe with
some electronic devices (amplifiers, filters, and speaking system)’.15
13
Concerning Marges, see Bassetto, presumably written in 1969 (PSS). Cf. also
‘Marginalia, ou l’opéra-fantôme de Boulez’. the letters exchanged, in the middle of the
14
See the letters exchanged by Pierre Boulez following year, between Boulez and Paco
and Charles Wadsworth between 1970 and Lagerstrom, professor of applied mathe-
1972 (PSS). matics at the California Institute of
15
Undated draft of a letter from Pierre Technology (PSS). The concert should have
Boulez to Madame Alexander Whittle, taken place at the Beckman Auditorium at
Table 11.4: . . . explosante-fixe . . . (1971). Encircled notes/groups of

notes from diagonal IVa . . .d: transcription and generation from the
entries of T7P (entries in bold within the table correspond to the
principal notes)
B♭ D♮ B♮ F♮ E♮ A♮ E♭ T7P
staves
IVa F♮
Vb B♭ F♮ E♮
VIc B♭ F♮ E♮ A♮
VIId B♭ F♮ E♮ A♮ E♭
Table 11.5: . . . explosante-fixe . . . (1971). Annotated transpositions of

P following the generation of the encircled notes/groups of notes
Transpositions of P
Diagonals
ORa . . . g E♭ G♮ E♮ B♭ A♮ D♮ A♭ T0P
IIa . . . f G♮ B♮ G# D♮ C# F# C♮ T4P
IIIa . . . e E♮ (G#) F♮ B♮ B♭ E♭ A♮ T1P
IVa . . . d B♭ (D♮) (B♮) F♮ E♮ A♮ E♭ T7P
Va . . . c A♮ (C#) (B♭) (E♮) E♭ G# D♮ T6P
VIa . . . b D♮ (F#) (E♭) (A♮) (A♭) C# G♮ T11P
VIIa . . . a (A♭) (C) (A♮) (E♭) (D♮) (G♮) D♭ T5P
Caltech, with the support of the Coleman

Chamber Music Association.
280 Paolo Dal Molin
Table 11.6: . . . explosante-fixe . . . (1971). Boxed groups of notes:

transcription and generation from P
Principal
‘Anacruses’ notes ‘Inflexions’
Sequences
ORa E♭ {E♭}
IIa E♭ G♮ {E♭ G♮}
IIIa G♮ E♮ {E♭ G♮ E♮}
IVa E♮ B♭ {E♭ G♮ E♮ B♭}
Va B♭ A♮ {E♭ G♮ E♮ B♭ A♮}
VIa A♮ D♮ {E♭ G♮ E♮ B♭ A♮ D♮}
VIIa D♮ A♭ {E♭ G♮ E♮ B♭ A♮ D♮ A♭}
This vision was realised shortly afterwards in . . . explosante-fixe . . .

(1972–4) for chamber group and real-time electronics.16 The work was
first premièred in New York on 5 January 1973, after the provisional
version for flute, trumpet and clarinet in A was performed in London
on 17 June 1972, together with all the other homages to Stravinsky that
had appeared in Tempo. Subsequent to the New York performance,
Boulez continued to develop the piece in collaboration with the
Experimentalstudio in Freiburg and it was regularly programmed up
to October 1974. A commercial recording was scheduled by CBS and,
while it was reported on a number of occasions, it was never ultimately
made.17 After 1976, every subsequent project for the expansion of the
16
Several correspondences confirm that studios in Europe and the United States
during the summer of 1972 Boulez under- (PSS).
17
took a series of visits to electronic music See the correspondence exchanged by
Boulez, his secretary Astrid Schirmer and
composition in this form was first of all postponed, then abandoned.

It features in the Universal Edition catalogue as a ‘withdrawn version’
of the homonymous piece from 1991–3,18 which is based on the most
advanced version of the flute part written twenty years before.
Instrumental Parts
For the instrumentation of . . . explosante-fixe . . . (1972–4) the composer
persevered with the solution combining the maximum number of instru-
mental groups (seven) with the minimum number of instruments per
group (one instrument), a single group of two instruments being the
exception. The instruments in the ensemble are flute, viola, trumpet,
cello, clarinet in A, violin and the coupling of vibraphone and harp.
A monodic part comprising at least seven sequences is assigned to each
instrument, which realises ‘Originel’ and six ‘Transitoires’, starting out
from the matrix. While ‘Originel’ is the final sequence in each instrumental
part, the order of the ‘Transitoires’ varies from one instrument to another,
following the instructions for . . . explosante-fixe . . . (1971). From one to
three ‘Emprunts’ (‘borrowings’) extracted from ‘Originel’ can be inserted
into the succession of ‘Transitoires’ within each instrumental part, in line
with the composer’s instructions, and in Boulez’s own arrangement these
feature as anticipations of the end of the piece, where ‘Originel’ is heard
complete (see Tables 11.7 and 11.8).
The homonymous sequences for the eight instrumental parts – for
example all of the ‘Transitoires VII’ (i.e. for each instrument) as well as
all of the ‘Originels’ – are related indirectly in that they elaborate
different transposed selections of the same basic sequence from the
matrix. While the complete exposition of the material is entrusted to
the flute, the other instruments play all the material which is notated
on from one to six staves. In this way vibraphone and harp play from
staves ABCDEF,19 the viola from staves ABCDE, the cello from ABCD,
the clarinet from ABC, the violin from AB and the trumpet from
A. For each instrumental part, excepting that for the trumpet,
the selection corresponds to the constitutive staves of the first
‘Transitoire’, e.g. the violin begins with ‘Transitoire II’ and realises
only staves A and B from each sequence. In a similar way, the clarinet
19
Paul Myers between 28 August 1973 and The parts for vibraphone and harp differ
14 May 1974 (PSS). only in the order of their constitutive
18
See for instance the version published sequences: the text for these sequences is
in March 2003, labelled KAT UE60869-99. itself identical and issues from the flute part
from June 1972.
282 Paolo Dal Molin
Table 11.7: . . . explosante-fixe . . . (1972–4), instrumental parts. ‘Structures’ –

i.e. arrangements of ‘Transitoires’, ‘Emprunts’ and ‘Originel’ – selections
of staves for each sequence and ranges of tessituras (with fundamental notes
[‘note originelle’] in bold)
Stave
‘Structures’ selections Tessituras
Trumpet iv emp1 v ii emp2 iii vi emp3 vii or a G♮3–B♭3–F♮5

Violin ii v vii emp vi iii iv or ab B♮4–D♮5–A♮
Clarinet iii v emp1 iv ii vii emp2 vi or abc C#3–E♮3–B♮4
Cello iv emp1 v ii emp2 iii vi emp3 vii or abcd F♮2–A♭2–E♭4
Viola v iii vi iv ii vii or abcde F#3–A♮3–E♮5
Vibraphone vi iv emp v vii ii iii or abcdef C♮4–E♭4–B♭5
Harp vi emp1 vii ii iv emp2 v iii or " "
Flute vii v emp1 iv vi iii emp2 ii or abcdefg C♮4–E♭4–B♭5
begins with ‘Transitoire III’ and realises only staves A, B and C from
each sequence, and so on. In addition to this, each instrument plays
the basic material from the selected staves in a given transposition of
the matrix tessitura. The factors of transposition are drawn from the
realisation of P in the general tessitura and its inversion around E♭4 as
the instructions prescribe. Table 11.7 shows the selections and trans-
positions used for each instrumental part, and indicates the range of
the tessitura and the resulting transposition of the ‘original note’ (‘note
originelle’) E♭.
Form and Textures

The formal plan of the piece in twenty sections (see Table 11.8)20 results
from the correlation between the structures of the instrumental parts
and two other factors. The first is the desynchronisation of the instru-
mental entries in the plan, a factor which is determined in turn by the
following principles: (1) the parts enter and accumulate progressively in
number; (2) those that begin with the same ‘Transitoire’ (that is to say,
trumpet and cello, and vibraphone and harp) enter together; (3) the
order for the entries of the parts follows from the structure of the flute
part which opens the piece with a solo sequence (‘Transitoire VII’).
20
The table is deduced from Boulez’s own are published in Dal Molin, ‘Introduction à la
‘Performance’ plan and the available famille d’œuvres “. . . explosante-fixe . . .”’,
recordings of the piece. A facsimile of the pp. 325–6 and 268–78, respectively.
former and a list of references for the latter
The second factor is the distribution of sequences and ‘Emprunts’, for

which Boulez provides specific criteria: first of all within a given section
of the form an instrument plays a single sequence or a single ‘Emprunt’;
next, the ‘Transitoires’ and the borrowings from ‘Originel’ (‘Emprunts’)
are distributed in a homogeneous manner antiphonally; finally, the work
concludes with a section in which all of the instruments perform
‘Originel’, the form in sum echoing that of Stravinsky’s Symphonies of
Wind Instruments.21
In the first half of the formal schema, the number of instrumental
parts combined within each ‘Transitoire’ section (see Table 11.8, section
numbers 1, 2, 4, 6, 8 and 10) increases from one to eight (1, 2, 4, 6, 7, 8).
In the second half, it diminishes symmetrically, according to the
retrograde of the same progression (8, 7, 6, 4, 2, 1). On the other
hand, within each ‘Emprunts’ section, the number of instrumental
parts varies irregularly from one to four parts, ‘so therefore the
crescendo and the decrescendo will be interrupted by a zigzag move-
ment of “Emprunts”, [and] the form is consequently very irregular, as
the simple description suggests’.22 As for the immediate succession of
two sections of ‘Transitoires’, an arrangement that occurs uniquely at
the beginning and end of the piece (sections 1–2 and 18–19), this
proceeds from the structure of the flute and violin parts: the flute part
beginning in effect in the same manner in which the violin part ends,
namely with two ‘Transitoires’ that are linked without the intervention
of an ‘Emprunt’.
Depending on the number of parts involved at any given moment, the
writing for the instrumental groups gives rise, syntactically, to either
a heterophony of monodies or a simple monody (solo sections), but
where every monody is nevertheless capable of being transformed by
the electronics into homophony or heterophony. For each section with
more than one instrument, the encounters take place ‘with a degree . . . of
controlled chance’23 that is inversely proportional to the number of
instruments involved. Finally, the concrete musical form, in other
words the form of the work as it is performed, results from the segmenta-
tion and arrangement of the section’s constituent sequences (one per
instrument) that are established in advance of the performance by the
musicians.
21 22
For some recent commentaries on this type Haüsler, ‘Gespräch mit Pierre Boulez’,
of form see Campbell, Boulez, Music and p. 30.
23
Philosophy, pp. 207–9 and Goldman, Boulez, Par volonté et par hasard,
The Musical Language of Pierre Boulez, pp. 137–8; Boulez, Conversations with
pp. 70–7. Célestin Deliège, p. 105.
Table 11.8: . . . explosante-fixe . . . (1972–4). Formal schema: arrangement of the sequences (‘Transitoires’, ‘Emprunts’ and ‘Originel’) from the
instrumental parts
Sections 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Flute vii v emp iv vi iii emp ii or

Viola v iii vi iv ii vii or
Trumpet iv emp v ii emp iii vi emp vii or
Cello iv emp v ii emp iii vi emp vii or
Vibraphone vi iv emp v vii ii iii or
Harp vi emp vii ii iv emp v iii or
Clarinet iii v emp iv ii vii emp vi or
Violin ii v vii emp vi iii iv or
Development of the Instrumental Parts. The Flute,

1972–3
Between the end of 1972 and the summer of 1974, Boulez reworked six of the
eight instrumental parts up to three times, revising the musical text of the
different sequences without changing the order within each part. As the parts
were developed and the programmes for the conversion of sound grew in
number, the composer worked on perfecting the aleatoric organisation of
the parts within each section.24 Drawing on sources conserved in the Paul
Sacher Stiftung, at Universal Edition and in other archives, four versions of
the flute part exist, three versions of the parts for viola and for cello
respectively, two versions of the parts for trumpet, clarinet and violin, and
a single version of the parts for vibraphone and harp (see Table 11.9).25
On the basis of the instrumental parts, and referring to the chronology of the
different performances of the work, as well as to available recordings, we can
distinguish the following variants for . . . explosante-fixe . . . (1972–4),
namely two versions of the piece (labelled a and b in Table 11.9), two stages
of the second version (b.1 and b.2), and several arrangements of the latter
(b.21, 2, 3), all of them provisional.
In the course of thirteen months, the flute part evolved from the almost
literal realisation of the matrix (June 1972) to a musical text that is char-
acterised by a profusion of arabesques, formed from basic figures and
interspersed with polarised clauses that draw on the boxed and encircled
groups of notes, and which sustain or repeat one or more of the polar notes.
From the time of the first reworking, Boulez developed most of the objects by
means of measured or non-measured supplementary notes, only rarely
modifying their succession. As their layout proliferates, their pitch structures
24
See documents conserved in the library of which the parts for players were printed.
the former Experimentalstudio der Heinrich In most cases, these transcriptions were rea-
Strobel Stiftung des Südwestfunks in lised by Krystyna Reeder. The existence of
Freiburg (now Experimentalstudio für such transparencies is confirmed by a letter
akustische Kunst), the recordings of the from Reeder to Eva Smirzitz of 15 May 1981
performances from 1972 to 1974 (listed in (conserved in the archives of the publishing
Dal Molin, ‘Introduction à la famille house in Vienna); their precise location
d’œuvres “. . . explosante-fixe . . .”’, remains, however, unknown. The first ver-
pp. 123–36) and Haller, Das sion of the flute part and of the clarinet part
Experimentalstudio, vol. ii, pp. 55–64. date from before 16 June 1972 (letter from
25
Boulez’s fair copies for the different David Drew to Pierre Boulez, 22 June 1972,
versions of the instrumental parts are con- PSS). The terminus ante quem of mid-
served at the PSS, with the exception of the November 1972 is inferred from Haller, Das
parts for vibraphone and harp, and Experimentalstudio, vol. ii, p. 55. Completion
the second version of the flute part dates from December 1972 onwards are
(December 1972). As Boulez completed the noted in pencil in Boulez’s fair copies (PSS),
parts, Universal Edition produced transcrip- probably by Astrid Schirmer. All these dates
tions (most often on transparent paper) from are corroborated by other documents.
286 Paolo Dal Molin
Table 11.9: . . . explosante-fixe . . . (1972–4). Versions of the instrumental parts and stages of
the whole work
Instrumental parts
Date of achievement Versions
stages 1972 before 17 June Fl1 Tpt1 Cl1

before mid-November Vla1 Vlc1 Vib-Hp Vln1
a December Fl2 " " " " " "
b.1 1973 April Fl3 Vla2 " Vlc2 " " "
.21 July Fl4 Vla3 " Vlc3 " " "
2 1974 March (Cl), April (Vln) " " " " " Cl2 Vln2
3 August " " Tpt2 " " " "
(which are always included in the basic tessitura shown in Example 11.1 up
to the text from April 1973) stabilise the starting objects or draw specific
melodic and harmonic consequences from them. Simultaneously, their
other qualities are specified and thus each object is more sculpted in favour
of the clarification of the musical morphology and syntax of the instrumental
part. Finally, in the version from July 1973, the composer intervenes even
with regard to the general tessitura and establishes metronome markings.
This was the basis for the later developments of 1985 and 1991–3, as will be
seen in the conclusion to this chapter.
‘Originel’
The ‘Originel’ of December 1972 develops the six objects from the previous
version of June 1972 (see Example 11.2a) in the form of six musical sen-
tences, which proceed differently, depending on whether it is a question of
the five cells (‹A♮ B♭›, A♭, E♮, G♮ and D♮) or of the boxed note on E♭.
Example 11.2b shows the first development of the one-note figure G♮.
The sentence which is generated deploys seven new figures, opening
with a principal figure marked plus lent, comprising a principal note
followed by an ‘inflexion’ – a sort of main clause within the sentence,
and a ‘signal’ in Boulez compositional practice. The other six figures are
composed of a principal note preceded by an ‘anacrusis’ and are to be
performed in the general tempo Très lent established in the first version
from June 1972, but now characterised as extrêmement souple et léger,
rubato. The seven pitches of the principal notes reduce to the funda-
mental heptachord transposed by T4, that is P anchored on E♭ trans-
posed on G♮ itself (see Example 11.2b, lower stave); the pitches of the
Ex. 11.2 . . . explosante-fixe . . . (1972–4), flute part: ‘Originel’, develop-

ment of the basic figure G♮. © Copyright 1972 and 1973 by Universal
Edition (London) Ltd. with kind permission
(a) Version from June 1972. Transcription from the autograph fair copy
(Paul Sacher Stiftung, Sammlung Pierre Boulez)
(b) Version from December 1972: first development of the basic one-note
figure G♮ from the previous version (a). Annotated excerpt of transcription
from the copyist’s manuscript (Paul Sacher Stiftung, Sammlung Pierre
Boulez)
(c) Version from April 1973: second development of the basic figure G♮
from the previous version (b). Excerpt from the autograph fair copy (Paul
Sacher Stiftung, Sammlung Pierre Boulez) with the fragment concerned
highlighted
grace notes belong to P or to T4P, depending on whether it is a case of

the principal figure or of the six other figures. Consequently, in
‘Originel’ from December 1972, the five sentences issuing from the
basic cells ‹A♮ B♭›, A♭, E♮, G♮ and D♮ deployed in the previous version
of the sequence, realise the transpositions of P by ‹T6 T7›, T5, T1, T4 and
288 Paolo Dal Molin
T11 respectively, i.e. the translations on ‹A♮ B♭›, A♭, E♮, G♮ and D♮
themselves of P anchored on E♭. At the same time, their principal
figures (i.e. those marked plus lent) unfold the basic P to the point of
its complete deployment in the last sentence, at the very end of the
sequence.
As for the boxed note, the sustained E♭ is now embellished with inter-
spersed groups of grace notes; the pitches are taken from the general
tessitura and, in the second half of the sentence, are reduced to subsets of
the six transpositions of P that we have already mentioned.
This text is reworked in detail in the version from April 1973 (See
Example 11.2c). Each sentence of the version from December 1972, such
as the one shown in Example 11.2b, is greatly amplified, notably
through the internal proliferation of the ‘anacruses’. Furthermore,
their unfolding is now marked by the alternation of four different
expression markings within the basic general tempo which, as set out
in the previous version of the piece, are always Très lent: Stable;
Irrégulier, vacillant; Régulièrement modulé; Souple. To implement this
reworking, the composer reinterprets, in line with the serial principle
that was developed in the early 1950s, the figures generated from the six
transpositions of the fundamental heptachord P following the order in
Table 11.10. Within each sentence, the figure that is derived from P (see
Table 11.11, first column), previously marked plus lent, and which
functions as main clause or signal, now has the expression mark
Souple and the notes are flutter-tongued and trilled. The principal
note takes the value of a dotted crotchet (the version from July 1973
will specify quaver at metronome marking 72) and is played ppp sou-
tenu; the ‘inflexion’ is notated in quavers (with quaver at metronome
marking 84–92 in the July 1973 score), più ppp soutenu.26 A significant
difference is noticeable here in relation to the previous version: in place
of the gestural sequence principal note – caesura – ‘inflexion’, the
complete figure is now isolated from those that surround it. At the
same time, the internal momentum (accelerando) dissipates and is
thereby superseded by a kind of incantatory stasis.
The other clauses, in so far as they result in April 1973 from the revision
of the corresponding ones from December 1972, are divided into three
families according to their principal notes. These in effect possess the
same rhythmic value, the same mode of attack and the same dynamic
marking depending on whether they are derived, respectively, from
the second to the fourth columns of Table 11.11 (quaver, tenuto), from
26
In Tables 11.11 and 11.13, the two entries clauses which are exchanged in the realisa-
followed by an asterisk correspond to figures/ tion of the respective shapes and envelopes.
Table 11.10: . . . explosante-fixe . . . (1972–4), flute part

(1973): ‘Originel’. Partitioning of P, with partial orders,
and its transpositions (notes from basic figures in bold)
T0 E♭ {E♮ G♮ A♭} {A♮ B♭} D♮

T1 E♮ {F♮ A♭ A♮} {B♭ B♮} E♭
T4 G♮ {G# B♮ C♮} {C# D♮} F#
T5 A♭ {A♮ C♮ C#} {D♮ E♭} G♮
T6 A♮ {B♭ C# D♮} {E♭ E♮} A♭
T7 B♭ {B♮ D♮ E♭} {E♮ F♮} A♮
T11 D♮ {E♭ F# G♮} {A♭ A♮} C#
Table 11.11: . . . explosante-fixe . . . (1972–4), flute part (1973): ‘Originel’.

Transpositions of P (partitioned) following their arrangement in the sequence,
with the shapes of the principal notes and inflexions from the figures corre-
sponding to the entries (notes from basic figures in bold, ‘original note’ E♭
boxed)
A {B C# D} {E E } A
B {B D E} {E F } A
A {A C C#} {D E} G
E {F A A} {B B} E
G {G# B C} {C# D } F#
E
D {E F# G} {A A } C#
290 Paolo Dal Molin
Table 11.12: . . . explosante-fixe . . . (1972–4), flute part (1973): ‘Originel’. Envelopes for
clauses marked Stable, Régulièrement modulé and Irrégulier, vacillant
Expression marking for

the T° and metronomes Anacruses: layout Note values Mode of attack
Stable one group of x notes semiquavers legato

Ú = 132 (♪= 66) (ordinary or triplet,
quintuplet, even
septuplet)
Régulièrement modulé x groups demi-semiquavers portato or
♪ = 92/72 The number of notes per group staccato
varies progressively, according
to two (possibly combined)
principles: the regular augmenta-
tion and diminution of a note, from
1 to y notes and from y to 1 notes
Irrégulier, vacillant x groups demi-semiquavers legato,
Ú = 120–132 (♪ = 60–66) The number of notes per group is (triplet or quintuplet) portato or
constant (3 or 5) staccato
the fifth to the sixth columns (dotted quaver, flatterzunge) or from the
seventh column (crotchet, trilled). The expression marks for the clauses
from the same family, as well as the properties of the ‘anacruses’ (internal
constitution, rhythmic value and mode of attack), are unified as the
clauses progress according to the rotation of three envelopes (see
Tables 11.12 and 11.13).
The values of variables x and y which are responsible for the content of
the ‘anacruses’, as can be seen from Table 11.12, depend on the corre-
sponding ‘anacrusis’ from the previous version from December 1972, and
more precisely on the number n of its constitutive notes. While y remains
equal to n throughout the sequence, the value of x increases progressively
from n (in the first sentence of the sequence, deriving from the basic
cell ‹A♮ B♭›) to 2n (in the following sentences, deriving from the basic cells
A♭, E♮, G♮), then to 3n (in the last sentence, from D♮).27 This augmenta-
tion of the coefficient from 1 to 3 clarifies the formal articulation of the
sequence.
27
The numerous exceptions to this rule are the second complex (Régulièrement modulé).
explained in different ways (according to While the augmentation and diminution of
whether it is a question of the first, the second the densities of the groups take place one
or the third complex), and end up confirming after another within the same ‘anacrusis’, the
the rule itself. Those few that can be under- two central groups fuse and lose a note.
stood without entering into detailed exami- Hence, the succession 1, 2, 5, 2, 1, for exam-
nation are those that can be seen in the ple, results from a contraction of 1, 2, 3 and 3,
internal constitution of the ‘anacruses’ of 2, 1.
Table 11.13: . . . explosante-fixe . . . (1972–4), flute part (1973): ‘Originel’. Transpositions of

P (partitioned) following their arrangement in the sequence, with the placing of envelopes
Stable, Régulièrement modulé and Irrégulier, vacillant through rotation (notes from
basic figures in bold, ‘original note’ E♭ boxed)
T6 A♮ {B♭ C# D♮} {E♭ E♮} A♭

Souple Stable Régulièrement modulé Irrégulier, vacillant
T7 B♭ {B♮ D♮ E♭} {E♮ F♮} A♮
Souple Irrégulier, vacillant Stable Régulièrement modulé
T5 A♭ {A♮ C♮ C#} {D♮ E♭} G♮
Souple Régulièrement modulé Irrégulier, vacillant Stable
T1 E♮ {F♮ A♭ A♮} {B♭ B♮} E♭
Souple Irrégulier, vacillant Stable Régulièrement modulé
T4 G♮ {G# B♮* C♮} {C# D♮} F#*
Souple Stable Régulièrement modulé Irrégulier, vacillant
T0 E♭
Libre
T11 D♮ {E♭ F# G♮} {A♭ A♮} C#
Souple Régulièrement modulé Irrégulier, vacillant Stable
Meanwhile, the pitches that are added to the reworked ‘anacruses’ result,
for the most part, from a type of proliferation that the composer calls
‘morphous’ (‘morphe’),28 since they derive from the interval array of the
‘anacrusis’ with which they begin, through the application of a number of
different operations.29
‘Transitoires’
As for the ‘transitoires’: the resistance they offer to the type of reduction
undertaken in the analysis of ‘Originel’ or, on the contrary, their con-
formity to it (as is the case, for example, with the one and three-note
figures in ‘Transitoire II’) indicates the different way in which Boulez
reworked them from the first revision. Moreover, the basic givens are
very different if one considers that ‘Originel’ was at first very pared
down (as with ‘Transitoire II’) and that the redevelopment was intended
to make a longer final sequence for it. The matrix continues nevertheless
to branch out in the definition of the enveloping categories, which are
specified increasingly from one version of the text to the other (see
Table 11.14). While the version from December 1972 represents an
28 29
See Boulez, ‘Le système et l’idée’, pp. 98–9 See Dal Molin, ‘Mémoriale de Pierre
(and the pages immediately preceding). Boulez’, 499–504.
292 Paolo Dal Molin
Table 11.14: . . . explosante-fixe . . . (1972–4), flute part, ‘Transitoire VII’. Evolution of the
envelopes for the encircled notes and figures in the succeeding versions
Encircled notes
Matrix June 1972 April 1973 July 1973
Basic T° and metronome Transpositions

Staves dynamics Expression markings for the T° markings (pitch space)
A ppp Flexible, balancé Très Rapide (♩ = 158) 0

B p Répétitif, obstiné Rapide (♩ = 76) 6
C mp Bref Assez Rapide (♩ = 72/84) −3
D mf Source/Échos en ricochets Modéré (♩ = 80↓72) 10
E f Balancé, variable Assez Rapide (♩ = 84/92) 3
F ff Répétitif variable Rapide (♩ = 104) 8
G fff Soutenu/Interrompu Très Rapide (♩ = 158) 14
Figure
Matrix June 1972 April 1973 July 1973
Basic T° and metronome Transpositions

Staves dynamics Expression markings for the T° markings (pitch space)
A ppp Régulièrement modulé Très Rapide (♩ = 108/92) 0

B p Strict, absolument (aussi vite que possible) Rapide (♩ = 76) 6
C mp Abruptement irrégulier Assez Rapide (♪ = 144) −3
D mf Exact, mais flexible Modéré (♪ = 120–144) 10
E f Régulier, légèrement modulé Assez Rapide (♪ = 168) 3
F ff Rythmiquement rigide Rapide (♪ = 192, 176 flatterzunge) 8
G fff Souple, oscillant Très Rapide (♩ = 120 legato, 112 14
staccato, 108 flatterzunge)
intermediate stage, the version that was completed in April 1973 con-
nects different complexes of features with the figures and encircled notes
of the same ‘Transitoire’, following the staves of the matrix from which
they originate (. . . explosante-fixe . . ., 1971). The figures from the same
stave are in this way homogenous with regard to the basic dynamic, in
the flow of the tempo, as well as in the two most prominent character-
istics which it is easier to hear than to tabulate: the type of contour and
the modes of sound production. Finally, in July 1973, the objects are
made more individual by means of specific metronome markings and
transpositions, the latter implying a radical change and amplification of

the harmonic fields. In this way, the speed of the figures deriving, for
example, from the G staves of the matrix (Souple, oscillant) varies in
each ‘Transitoire’ according to the mode of attack (the crotchet is 120
for the legato, 112 for the staccato and 108 for the flatterzunge), and
their pitches are inscribed in seven different transpositions of the basic
tessitura (one transposition per sequence).
However, if one observes from this point of view all of the ‘Transitoires’ in
decreasing order from numbers VII to II, one notices the flexibility of
Boulez’s selection and adjustment of the complexes shown in Table 11.14.
Indeed, the exceptions in the layout and in the character of the figures that
are deployed are so numerous and significant on the musical plane that we
must accept for the moment that the system gives way to more refined local
strategies.
In comparison with the development of ‘Originel’, the most noteworthy
differences concern the rewriting of the objects. Despite the principles that
were established in 1972 and that were developed in the ‘Originel’
from April 1973, their layout (i.e. the number of notes and the pitches)
follows much less strict rules than those described above: in place of
engendering their proliferation through some kind of systemic necessity,
Boulez capitalises freely here on the stylistic potential and properties of
the starting material. Such, at least, is the impression, given the number of
variables in play. Two extracts from ‘Transitoire VII’ serve to illustrate the
point.30 Example 11.3 shows the beginning of the sequence – that is, the
opening of the whole work – which realises the encircled note D♭ from
stave VIIa, five two-note figures from VIIg, etc. (see the matrix and
Example 11.3a).31 Starting with the version from December 1972 (see
Example 11.3b), the prolongation of the pitch D♭ (ppp) is first of all cut
midway by a group of five little notes (staccato, poco accelerando and
crescendo) that are derived from the same pitch collection associated with
VIIa, that is the referential heptachord P transposed by T5 (E♭→A♭). Each
of the five figures which follow increases by the addition of a grace note
‘anacrusis’ preceding either the two basic notes or just the second of
them. The pitches C♮, D♭ and F#, which are not included in the
referential collection associated with the stave VIIg (that is P itself),
belong to its inversion around A♭ (/D♮) corresponding to the encircled
30 31
The flute part that was completed In the examples which follow, the principal
in June 1972 is reproduced in full in Dal errors made by the copyist are amended
Molin, ‘Introduction à la famille d’œuvres within brackets.
“. . . explosante-fixe . . .”’, pp. 263–4.
294 Paolo Dal Molin
Ex. 11.3 . . . explosante-fixe . . . (1972–4), flute part: ‘Transitoire VII’ (opening).

Development of the first six basic objects: encircled note D♭ and five two-note figures.
© Copyright 1972 and 1973 by Universal Edition (London) Ltd. with kind permission
(a) Version from June 1972. Excerpt of transcription from the autograph fair copy (Paul
Sacher Stiftung, Sammlung Pierre Boulez)
(b) Version from December 1972: first development of the six opening objects from the
previous version (a). Excerpt from the copyist’s manuscript (Paul Sacher Stiftung,
Sammlung Pierre Boulez) with the fragment concerned highlighted
(c) Version from April 1973: second development of the six opening objects from the
previous version (b). Excerpt from the autograph fair copy (Paul Sacher Stiftung,
Sammlung Pierre Boulez) with the fragment concerned highlighted
note within VIIg itself. While it seems to govern the proliferation of

figures issuing from VIIg, this principle does not apply in the following
‘Transitoires’, where the figures from the diagonal ORa . . . g are also
augmented with complementary notes (F♮ and B♮), as is already
confirmed in ‘Transitoire V’. However, the revision of the inaugural
group (Flexible, balancé) in the text from April 1973 results also
from its application (see Example 11.3c). Here Boulez compresses
two layers in a monodic line: the basic D♭, which is prolonged by
the last note of the group, and the development of the five-note cell
where each of the former grace notes (now demi-semiquavers, staccato
and marcato) is linked to a different sustained note, in a new order.
In such a reconfigured unfolding of the basic encircled note centred on
D♭, F♮ and B♮ expand the basic space of T5P by way of its inversion
around D♭ (/G♮) itself. The subsequent clauses in the fragment
reshape the figures from the previous version: their durational values
are now completely measured; the basic principal notes are assigned
a characteristic marking (marcato); the other notes are no longer
simply ‘anacrusis’ notes but form an outline characterised by their
repetition around some principal notes. The harmonic properties that
they manifested already in the text from December 1972 are now more
pronounced, and the efficacy of this reworked opening of the piece lies
in its arrangement of distinct harmonic colours that are evident from
the first gesture.
Later in the sequence (see Example 11.4), to give another example,
a fragment links a figure of five notes, one of seven and one of six, that
derive respectively from the staves D, B and C of the basic ‘Transitoire VII’
(see Example 11.4a). In the first of these figures, the ‘anacruses’ which are
added in anticipation of the principal notes (see Example 11.4b, T° extr.t
flexible) set out the five notes of a pentachord from T7P (the referential
transposition of VIId), spelled ‘progressively’: F♮; F♮ A♮; F♮ A♮ B♭; F♮ A♮ B♭
B♮ E♮ (repeated). In the second figure (T° strict), each of the seven principal
notes (marcato) is followed by two notes written in a triplet, the pitches of
which can be reduced to the basic collection associated with VIIb. Finally, in
the third figure (T° irrég.), new tritones are grafted onto the basic
notes, independently of whether or not they are present in the referential
T6P. The text from April 1973 (see Example 11.4c) defines, refines or later
develops the configurations of these three clauses,32 before they are trans-
posed and assigned metronome markings in the following version.
32
Two, three or four triplets of semiquavers well as of the others, it is interesting to study
are added in the second figure as a result of the only page of draft known today, which is
the durations of the fundamental notes. reproduced in Boulez, ‘Le texte et son pré-
Concerning the proliferation of this object, as texte’, [143].
Conclusions
The foregoing analysis of the withdrawn versions of . . . explosante-fixe . . .
could lead to a number of general conclusions. In terms of Boulez studies
it can be related to the analysis of other works as well as to Boulez’s
statements in his writings, interviews and lectures, with a view to tracing
aspects of continuity and discontinuity in his musical and theoretical
production. For example, . . . explosante-fixe . . . (1971), with its idiosyn-
cratic, multiparametric and virtual objects (boxed notes, encircled notes
and figures), grouped together in filterable complexes (‘Originel’ and
‘Transitoires’ of the matrix) which, in turn, can be arranged in multiple
possible assemblages, provides a striking example of Boulez’s thinking,
especially in the light of what he distils from his compositional practice in
certain essays from the 1950s and in Penser la musique aujourd’hui.33
Moreover, the matrix and the composer’s instructions most surely
demonstrate the overarching exercise of compositional power that is
necessary, according to Boulez, in the face of the introduction of chance
into musical works. In this regard, the early ‘versions’ of . . . explosante-
fixe . . . provide new musical responses to the problems raised in certain
previous works, from the Third Sonata up to Domaines for clarinet and six
instrumental groups, problems which are discussed in a number of well-
known texts such as ‘Alea’ and ‘Construire une improvisation’ up to the
conversations Par volonté et par hasard and beyond.34
While acknowledging this, beyond the musicological domain it is
necessary to consider some questions pertaining to the poetics of the
original project, the compositional consequences of the piece elabo-
rated in 1972–4 and Boulez’s own a posteriori discourse. It is known
that, after the 1970s, the composer no longer reworked the entire piece
but only some sequences from one or another instrumental part, in
a series of independent pieces: . . . explosante-fixe . . . (1982–7) for
vibraphone and live electronics; Mémoriale (. . . explosante-fixe . . .
Originel) (1985) for flute and eight instruments; . . . explosante-
fixe . . . (1991–3) for MIDI-flute, ensemble and live electronics (stem-
ming from ‘Transitoires VII’ and ‘Transitoire V’ for flute and
from Mémoriale); and Anthèmes 1 (1991–2) for violin.35 At the same
time, he explained his abandonment of the initial project as he had set
it out in the 1970s: not only were the technical means inadequate for
33 35
See the seminal overview by Decroupet, These pieces re-elaborate respectively:
‘Comment Boulez pense sa musique’, certain sequences from the vibraphone part;
pp. 49–57. ‘Originel’ for flute (text from July 1973);
34
See Campbell, Boulez, Music and ‘Transitoires VII’ and ‘V’ for flute (July 1973)
Philosophy, pp. 193–218 (‘Expanding the and Mémoriale; ‘Originel’ for violin
virtual’). (April 1974).
298 Paolo Dal Molin
the realisation of his ambitions, but ‘the writing for each of the seven
instruments [sic] gradually became too complex, the result got close to
being indistinguishable’. This was all the more the case, as soon as
two or more extremely developed textures were transformed by the
electronics. ‘That is why’, Boulez adds, ‘I have now dissociated these
components’.36
The conversations with Célestin Deliège contain two revealing
passages on precisely this point. Based on interviews recorded
in August 1972 and August 1974, they are particularly valuable in that
they are contemporaneous with the composition of the withdrawn
versions of . . . explosante-fixe . . . and contain some important references
to the work. The first is found at the beginning of the second chapter.
In relation to his ‘probably innate feeling’ for ‘the proliferation of materi-
als’, Boulez affirmed: ‘the tendency to proliferation has its dangers
because it can lead one towards the same type of density, in other words
a density that is extreme at every moment’.37 Since this is not a recent
discovery for him but rather something he had already experimented with
in the 1940s, for example in the Quatuor à cordes, it would be nothing less
than naive to believe that he had continued to elaborate . . . explosante-
fixe . . . for several months in the 1970s to the point of gratuitous and
excessive complexity, as he suggested in the 1990s. In the spring of 1976,
the composer still seemed to have ambitions for this work, for developing
rather than reducing it, and he regretted not having had the time or the
means to realise his idea. It is clear however that even at this point when
the practical development of the work lacked the necessary technological
means, Boulez did not repudiate the supposed excessive complexity of the
instrumental parts, and we have to acknowledge that . . . explosante-
fixe . . . (1991–3) which was realised at IRCAM and which constitutes
the work as it is now performed, is far from having completed the original
project. As he stated to Hans Oesch:
We are now planning a close collaboration between the Freiburg

Experimentalstudio and IRCAM, to realise . . . explosante-fixe . . . in the next year
[1977] as I envisage it. The piece is so designed – as is part of Rituel –, that
instruments within a particular time frame can be completely independent. And
this is precisely what has not yet been realised in Freiburg. In the meantime, there
are just four channels for all eight musicians. I need a separate channel for each
instrument, connecting to a control centre. Figuratively speaking: I need
a switchboard, through which eight conversations can be conducted
36 37
Boulez, ‘Le texte et son pré-texte’, p. 144. Boulez, Par volonté et par hasard,
pp. 14–15; Conversations with Célestin
Deliège, p. 15.
simultaneously. I will now produce a much richer version for each part. I would
like to begin with one instrument and then see, how far I can go in the combination
with one. Eight simultaneous ‘orchestras’ are naturally impossible! There is also an
optimum number. What is more, the transformation of sound in space should take
place in particular directions. One cannot achieve this in my sense with the
halaphon. I need a punctual distribution of sounds in space. That this is very
costly, I know well.38
Returning to Boulez’s conversations with Deliège, the second passage from

Par volonté et par hasard that is of particular importance for the current
discussion concerns some remarks the composer made on improvisation, as
he conceives it.39 Boulez enunciates here the relation between rule, gesture
and form in the improvisational practices of different cultures and epochs,
concluding with some negative remarks on recent practices in Western art
music, aiming in particular at ‘intuitive music’, notably Stockhausen’s Aus
den sieben Tagen (1968).40 In two fragments that are situated at the extremes
of the passage in question, Boulez states the following:
Improvisation, and especially improvisation in groups where there is a degree of

sympathy between the individual members, always follows the same curve of
invention: excitement – relaxation – excitement – relaxation. In so-called
primitive societies a similar situation exists in religious ceremonies whose
relatively simple form involves a building up of psychological tension followed by
relaxation. There is a whipping-up of collective excitement and when the uproar
reaches its peak there comes the need to release the tension, and a period of
relaxation follows.
...
At present improvisation is a sequence of negations. If a lot of things happen in
register A, for the next few minutes A will be avoided and we shall have B; then after
B has had its outing, it will disappear and we shall have C instead. It is the opposite of
what happens in composition, where one combines elements A, B and C sometimes
in an extremely complex way. But mixing is excluded by this type of instantaneous
improvisatory creation, as a result neither of aesthetic nor of any other deliberate
policy but simply of inadequate memory – because the mind is incapable of mixing
certain elements.41
Would it be too much to find in these statements an allusion to . . .

explosante-fixe . . ., a work which the composer was at that very moment
in the process of composing? To find some reference there to what he
38
Oesch, ‘Interview mit Pierre Boulez’, européenne?’ (1980, 1984, republished in
p. 296. Boulez, Regards sur autrui, pp. 590–604).
39 40
See Boulez, ‘Le système et l’idée’, p. 95 See Boulez, Leçons de musique, p. 666.
41
(corrected and developed in Boulez, Jalons, Boulez, Par volonté et par hasard, pp. 150
p. 377, republished in Boulez, Leçons de and 152; Boulez, Conversations with Célestin
musique, p. 405); and Boulez, ‘Existe-t-il un Deliège, pp. 114–15.
conflit entre la pensée européenne et non
300 Paolo Dal Molin
had been researching and notably avoiding in this piece, first of all in
conceiving (1971) and then in pursuing (1972–4) the fulfilment of a finely
composed, collective ritual?
Translated from French by Edward Campbell in collaboration with the

author.
part iii
Reception Studies
12 Pierre Boulez in London: the William Glock Years
Peter O’Hagan
We have as yet heard nothing by Boulez in this country. When we do, shall we find
him entertaining? And if entertaining, will that be sufficient?1
From a contemporary perspective, it is surprising that a musician who

came to such prominence as an influence on London’s musical life
during the second part of the previous century first attracted attention
in the UK not as composer and conductor but as a writer. The context of
the veteran critic Eric Blom’s remarks was his perusal of the journal
Contrepoints, which included one of Pierre Boulez’s first articles,
‘Trajectoires’, a piece which Blom praised as being ‘brilliant and
penetrating’.2 The tone of benevolent paternalism which characterised
Blom’s review was somewhat less in evidence in the outrage which
greeted Boulez’s first article in an English journal, the notorious
‘Schoenberg is Dead’, which initially appeared in William Glock’s
recently established periodical, The Score.3 According to Glock’s subse-
quent account, the decision to publish it cost him ‘some precious
friendships’,4 and it may well have been the polemical tone of this article
which contributed to the stereotype of Boulez as being an enfant terrible
of contemporary European music, an impression reinforced by colourful
accounts of the first performances of his works in Paris and
Donaueschingen during the early 1950s. Nonetheless, the article marks
the beginning of a collaboration and friendship which lasted until
Glock’s death in 2000, an association which was crucial in the trajectory
of the careers of both men, and the results of which were to have lasting
effects on London’s musical life over the succeeding decades.
It was in between the appearance of these two articles that, almost
unnoticed, Boulez made his first professional visit to London.
Passengers on the overnight train which left the Gare du Nord, Paris on
the evening of Friday 21 September 1951 included the members of the
Renaud-Barrault Theatre Company, headed by the husband and wife
team of Madeleine Renaud and Jean-Louis Barrault. They were bound
1 3
The Observer, 2 April 1950. The Score, 6 February 1952, pp. 18–22.
2 4
Contrepoints 6 (1949), pp. 122–42. Glock, Notes in Advance, p. 58.
303
304 Peter O’Hagan
for London, for a three-week season of French plays at the now defunct St
James Theatre at the invitation of Laurence Olivier and Vivien Leigh, and
the group included the Company’s Musical Director, Pierre Boulez. From
1946 onwards, he had been engaged by the newly formed Company, work
which had originally involved him in playing the Ondes Martenot for the
inaugural production, a performance of Hamlet in a new version by
André Gide, with incidental music by Arthur Honegger. During the
subsequent years, Boulez’s role expanded to include the preparation and
conducting of small ensembles as part of the productions, and he accom-
panied the Company on extended tours, including one to South America
of over three months from the end of April to the middle of August 1950
(the first of three such tours). Barrault took an almost paternal interest in
his protégé during these years, as evidenced by the indulgent tone of the
biographical notice carried in the programme for the London season:
‘When one is twenty-five years old and senses oneself consumed by
a fervent longing to express oneself, it is a good omen to be intransigent
and not to make allowances, assuming the prerequisite of having been
gifted with rare intelligence. In this case all expectations are permissible.
That is why we esteem Pierre Boulez.’5 Nonetheless, the young composer
must have made the journey to London with some reluctance, as the terms
of his contract with the Company obliged him to miss the première of his
first orchestral work, Polyphonie X, which had been commissioned for the
1951 Donaueschinger Musiktage on the initiative of Heinrich Strobel.
One imagines the composer in his room at the Strand Palace Hotel,
anxiously awaiting reports from his friend Pierre Souvtchinsky on the
performance of the work under Hans Rosbaud on 6 October, and his relief
on receiving Souvtchinsky’s telegram informing him of the success of the
première. In fact, Boulez’s duties with the Company seem to have been
relatively light during the two-and-a-half weeks of the tour, since the only
productions to feature incidental music were the double bill of Molière
plays, Amphitryon and Les Fourberies de Scapin with music by Poulenc
and Henri Sauguet respectively, and Baptiste, the mime sequence from
the film Les Enfants du Paradis by Jacques Prévert, presented in the
Company’s London season as a ‘Mime in six tableaux’, with music by
Kosma. Boulez and Francis Chagrin6 are listed in the programme as
5
Add MS 80677, Olivier Archive, vol. cmxii, remembered as a composer of film music,
British Library. including that for the episodes ‘The Dalek
6
Francis Chagrin (1905–72), composer and Invasion of Earth’ from the popular Dr Who
conductor. Born in Bucharest, Chagrin TV series. In 1951 he founded his own
studied in Paris with Paul Dukas and Nadia chamber ensemble, which gave numerous
Boulanger at the École Normale before broadcasts over the next two decades, and it
settling in England in 1936. Whilst his music seems highly likely that Chagrin would have
has fallen from the repertoire, he is provided the ensemble for the
Musical Directors for the production, but no details are forthcoming

concerning any instrumental forces employed. The mime scene was
chosen as the final item on the first night of the season, 25 September,
a glittering occasion, which was concluded by a black-tie reception given
by Laurence and Lady Olivier at Claridges Hotel in which guests were
invited to meet members of the Company. The same production rounded
off what was evidently a very successful tour with two performances on
Saturday 13 October, the day before the Company’s return to Paris.
It would be some five years after the season at St James before Boulez
would be heard again in London, and in the meantime, his music was slow
to travel to the UK with only occasional performances of his works.
The parochialism of British musical life at the time is remarkable,
evidenced by a review in The Times of the 1952 ISCM Festival in
Salzburg, during the course of which the anonymous correspondent
observed that ‘the most enigmatic work of the Festival, Le Soleil des
eaux, a setting of two poems for three soloists and orchestra, was the
work of Pierre Boulez, a twenty-seven year old Parisian composer’. Whilst
praising the work’s ‘individuality of imagination’, the writer concluded:
‘To sum up: the Festival did not bring any new masterpieces to light. There
was much music of a technically high standard, but nothing to equal the
flowering of talent among contemporary composers in England.
Continental composers seem to be in a cul-de-sac which is the aftermath
of the upheaval caused by Schoenberg and the twelve-note school.’7 It was
in this climate that, thanks to the advocacy of William Glock, acting in
his capacity as chairman of the music section of the Institute of
Contemporary Arts,8 Yvonne Loriod was invited to give a recital at
Wigmore Hall in a programme which included music by Debussy and
Messiaen, as well as the first two movements from Boulez’s Second Sonata.
The concert took place on 24 January 1956, and the decision to
programme the sonata in incomplete form was a subject of correspon-
dence between Mme Loriod and Glock, who vainly attempted, rather late
in the day, to persuade the soloist to perform the complete work.9
The critical response to the piece was one of bafflement, summed up by
The Times reviewer:
Unattentive ears would be wrong in dismissing the piece as thoughtless rubbish.

The really disconcerting thing is that the aural and visual implications of the music
Renaud-Barrault Company’s season at St Music Festival in 1955, where he attended the

James Theatre. final rehearsal and first performance of Le
7
The Times, 16 July 1952. Marteau sans maître.
8 9
Glock held the position from 1954–8. In this MS Mus 954, William Glock Collection,
capacity he was one of the jury members at vol. xii, ff. 69–83, British Library.
the International Society for Contemporary
306 Peter O’Hagan
are quite distinct; either eye or ear can find cause for respect, but the impression
derived from watching, and listening simultaneously was most bewildering and
much more disagreeable.10
In November of that year Boulez was back in London, again with the
Renaud-Barrault Company, and as well as works which had already been
extensively toured, such as Claudel’s Christophe Colomb with incidental
music by Milhaud, the company brought with it a new production of
a play by Georges Neveux, Le Chien du jardinier. Among the credits listed
in the programme is the amusing listing, ‘Music on classic Spanish themes
arranged by Pierre Boulez’.11 Taking advantage of his presence in the UK,
Glock invited Boulez to give a talk at the International Music Association
entitled ‘New Orientations in Contemporary Music’. Despite the still nega-
tive reception of his music in London, Boulez’s reputation as an important
figure in the contemporary musical scene was evidently spreading and
in December, during the final week of the Renaud-Barrault tour, he partici-
pated in a short radio interview in French, recorded at Bush House, London
and broadcast on Saturday 8 December 1956 by the BBC World Service as
part of the French Weekly Magazine programme. Within months of this
broadcast, Boulez returned to London for his most important appearance so
far in the UK, a concert at Wigmore Hall on Tuesday 19 March 1957, which
took the form of a two-piano recital with Yvonne Loriod, again given under
the aegis of the Institute of Contemporary Arts. The programme was iden-
tical to that which Boulez and Loriod had performed in a number of
European cities over the preceding months, consisting of the three pieces
comprising Structures, Premier livre followed by the recently completed
Chapitre I of a second projected volume of three complementary pieces,
and with Debussy’s En blanc et noir completing the programme. Although
the concert was apparently attended by a number of distinguished British
musicians, the uncomprehending response to Boulez’s music was typified by
an anonymous notice which appeared in The Times, at the end of which the
reviewer summed up his thoughts with the comment: ‘Nevertheless, the
artistic, as opposed to the intellectual, value of the experiment seemed akin
to that provided by two naughty children mauling the keyboard with their
arms and fists.’12 The day before the Wigmore Hall concert, Boulez and
Loriod visited the BBC studios at Maida Vale, where, on the morning of
18 March 1957 they recorded the three pieces of Structures, Premier livre.
The recording was transmitted that same evening by the BBC Third
Programme, and was announced as being the first broadcast performance
10 12
The Times, 30 January 1956. The Times, 23 March 1957.
11
Add MS 80677, Olivier Archive, vol. cmxii,
British Library.
of the work in the UK. The occasion was notable also for the highly erudite
and informative introductions to each of the pieces, and it seems likely that
these were the work of William Glock, even though the voice of the anon-
ymous announcer cannot be identified with certainty.13
Meanwhile, Glock had more ambitious plans for Boulez’s future activities
in London, and it was evidently on his initiative, acting through the ICA, that
the next proposed engagement of Boulez was to take the form of a concert
in May 1957 to be mounted jointly with the BBC Third Programme. On this
occasion, Boulez was to be formally engaged as a conductor for the first
time in the UK, directing the ‘Marigny Players’ in a programme of works
consisting of:
Canti per Tredeci Luigi Nono

Concerto op. 24 Anton Webern
Zeitmasse Karlheinz Stockhausen
Le Marteau sans maître Pierre Boulez
There were at the time various obstacles in place regarding the engage-
ment of foreign musicians, and the letter of the BBC’s Music Booking
Manager to the Ministry of Labour justifying the proposal strikes a quaint
note, describing the programme as consisting of ‘four ultra-modern works
(unknown in this country, as far as we are aware)’ and concluding, ‘You
will, I feel sure, readily agree that it would be impracticable to try to mount
these works with English players recruited “ad hoc” and trust, therefore,
that you will not have any difficulty in issuing the requested labour
permit.’14 The documents preserved in the BBC Written Archives show
that as with the Concerts du Domaine Musical in Paris, Boulez was
personally involved in numerous practical details with regard to labour
permits and contracts, even including financial matters (it is revealing to
note in passing that all the musicians including Boulez himself were
contracted at the same modest fee). The concert took place before an
invited audience in the Concert Hall of Broadcasting House on the eve-
ning of Monday 6 May 1957, and the ensemble included several musicians
familiar from Boulez’s first recording of Le Marteau, including the
soprano soloist Marie-Thérèse Cahn, and the guitarist Anton Stingl.
Whilst the playing of the ensemble was praised, as on previous occasions
13
Sound Archive, C236/284, British Library. Structures, Deuxième livre, Chapitre I at the
Glock’s recollection that three of the pieces of Wigmore concert the following day. See
Structures were performed is correct with Notes in Advance, p. 93.
14
regard to the broadcast, but omits mention of Norman Caroll to Mr H. W. Clark,
a fourth piece, the London première of 22 March 1957.
308 Peter O’Hagan
the critical response to Boulez’s music was mixed, the composer

Humphrey Searle writing in The Musical Times that:
Le Marteau sans maître . . . proved to be an interesting study in sonorities and

rhythmical devices; but in the end one’s ear tired of the tone colour, and the
introduction of three tam tams in the last of the work’s nine movements came as
a welcome contrast. Here again the framework was too big for the musical thought:
but Boulez again showed himself to be an imaginative composer as well as, on this
occasion, a skilful conductor of his own ensemble.15
Although the aftermath of the concert was marred by a misunderstand-

ing over fees for the recording rights, interest in Boulez’s work was mount-
ing in the UK, and in September of the same year the broadcaster
Humphrey Burton made a proposal for a programme about the place
of music in the productions of the Renaud-Barrault Company, and speci-
fically about the contribution of Boulez. The project was tantalisingly
unrealised, as was a later request to Boulez to conduct a chamber music
programme to include Le Marteau sans maître at the Victoria and Albert
Museum on 19 November 1959. In view of Boulez’s existing conducting
commitments at that time, which included two concerts at the
Donaueschinger Musiktage the previous month,16 it was unsurprising
that he declined the invitation, but in the meantime, he was being actively
promoted to the BBC’s new Controller of Music by the London office of his
publisher, Universal Edition: ‘It seems likely that Boulez will be asked to
conduct more in the future than has so far been the case, and we wonder
whether the BBC might be interested in inviting him to conduct a concert.
His astonishing fluency and concentration were, perhaps, the outstanding
impression of Donaueschingen this year. We hope you will find this
suggestion interesting.’17
Glock evidently did not act on the proposal immediately, but
in February 1961 he wrote directly to Boulez on behalf of his colleague,
The Observer critic Peter Heyworth, who was seeking to obtain an interview
with the composer. During the course of the letter, he expressed regret
that he would be unable to hear Pli selon pli in Paris the following month,
but noted that Heyworth would attend. The latter’s detailed review of the
concert is a landmark in the critical reception of Boulez’s music in the UK,
showing a gradual shift away from the incomprehension of a majority of the
15
The Musical Times, May 1957. included the first performance of the original
16
The concerts took place on 17 and version of ‘Tombeau’. Boulez took over
18 October 1959, featuring respectively the direction (shared with Luciano Berio) of
Ensemble du Domaine Musical and the the second concert as a consequence of the
Südwestfunkorchester, and the first concert, illness of Hans Rosbaud.
17
dedicated to the memory of Prince Max Egon Letter dated 26 October 1959, BBC Written
zu Fürstenberg, patron of the Festival, Archives.
reviews of the previous decade. The work was performed on this occasion in
the provisional version, with the opening movement ‘Don’ cast in the form
of a piano solo with vocal interjections, and although Heyworth expressed
some reservations about a perceived lack of textural variety in the
‘Improvisations’, the conclusion is unequivocal: ‘there is at work here
a mind at once imaginative, powerful and individual, and one that may
well be writing a new chapter in the history of music’.18 Within days of the
appearance of Heyworth’s review, Glock contacted Boulez on 29 March 1961
offering him a series of engagements with the BBC Symphony Orchestra
over a short period either at the end of that year or early in the following one.
The residency would include a Royal Festival Hall concert as well as studio
concerts, and Glock suggested a programme to include Le Marteau sans
maître as well as works by Webern and Debussy for the Festival Hall date.19
Boulez again declined the offer, pleading that he was already over-
committed during that period,20 and it was not until December of the
same year in response to a renewed invitation that he wrote to the BBC
confirming that he would be available from 21 February to 4 March 1964 to
work with the orchestra.
The principal concerts of this residency would include two programmes at
the Royal Festival Hall, the first of these on 26 February 1964 under the
auspices of the Royal Philharmonic Society, the second on 4 March 1964 to
be promoted by the BBC. Glock wrote to Boulez on 12 March 1963, suggest-
ing the following programme for the second concert:
Adagio and Fugue in C minor Mozart

Six Orchestral Pieces op. 6 Webern
Le Soleil des eaux Boulez
Brandenburg Concerto No. 5 Bach
Images Debussy
Throughout the spring of 1963, Boulez was in residence at Harvard

University as Horatio Appleton Lamb Visiting Lecturer in Music, and
lines of communication were not always reliable. Boulez’s eventual reply,
dated 20 April 1963,21 accepted the proposal for the first part of the concert,
but made the not unreasonable point that the Brandenburg Concerto did not
sit well before the closing Debussy work. He therefore proposed either the
Ricercar from The Musical Offering in Webern’s arrangement, or else a work
by Stravinsky, mentioning the Symphonies of Wind Instruments and Quatre
Etudes pour orchestre as possibilities, and it was quickly agreed that both the
18 20
The Observer, 19 March 1961. Undated letter, BBC Written Archives.
19 21
BBC Written Archives. BBC Written Archives.
310 Peter O’Hagan
Webern arrangement and Symphonies of Wind Instruments would be

included. Negotiations for the Royal Philharmonic Society programme
were more protracted, as the pianist Clifford Curzon had already been
engaged to play a classical concerto (Beethoven’s Fourth), and the rest of
the programme had to accommodate this work. In response to a proposal
from Eric Warr, Assistant Head of Music Programmes, that the programme
include a work not previously performed in England, Boulez replied from
Harvard at some length, and his response is worth quoting for the light it
sheds both on his grasp of practicalities and on his approach to programme
building, consistent with his Domaine Musical programmes in their alert-
ness to the possibilities of juxtaposing contrasting styles. After proposing
works by Varèse and Berio for possible inclusion, Boulez expanded on the
various options in words which must have struck a chord with Glock, given
his grounding in the classical tradition and his empathy with the works of
the Second Viennese School:
Apart from these works, I can propose, equally in relation to the ‘modernism’ or
the ‘classicism’ of the programme two types of work: in the first instance, the
Symphony No. 5 by Schubert, or the Symphony No. 104 by Haydn; in the second
one, the Variations op. 30 of Webern and the Pieces op. 16 by Schoenberg. If for
Le Soleil des eaux, in fact, which takes place a week later, you invite Helga
Pilarczyk to come, it would be an economy to utilise her not only for my piece,
which is very short, but to ask her to sing either the Berg or the Schoenberg;
obviously this would be two soloists in one programme, but as Pilarczyk will be in
the programme the following week, it would not be financially inconvenient.
The only difficulty concerns the Songs op. 22 by Schoenberg; these are rarely
performed and at the same time are of great interest; but the reason for their
infrequent performance is a result of the forces required, which, for a radio
station like the BBC, would not be insurmountable.22
In the event, Boulez’s radical suggestions were toned down, and the eventual
programme for the Royal Philharmonic Society concert on 26 February 1964 –
in effect Boulez’s début concert as an orchestral conductor in London – was as
follows:23
Symphony No. 104 Haydn

Variations for Orchestra op. 30 Webern
Piano Concerto No. 4 Beethoven
Jeux Debussy
22
Letter dated 13 May 1963, BBC Written down to Worthing with the BBC orchestra
Archives. and gave a concert in the Assembly Hall on
23
It may be noted in passing that these Sunday 23 February 1964 in which Vladimir
concerts were not quite Boulez’s first Ashkenazy was soloist in a performance of
appearance as an orchestral conductor in the Chopin’s First Piano Concerto.
UK, since the previous weekend he travelled
The discernible shift in attitudes towards Boulez’s music, evident earlier in

the decade, was now reinforced with universal acclaim for his conducting, in
particular of the Webern and Debussy works. It was again Peter Heyworth
who best summarised the response: ‘Until last week his reputation here was
that of a rather fierce and far-out composer. But the concert he gave on
Wednesday with the BBC orchestra was sufficient to establish him as one of
the most stimulating young conductors to have emerged for many years.’24
In between the two Festival Hall concerts, a studio concert in Maida Vale was
organised and broadcast live on 29 February 1964, the programme including
both the Haydn and Debussy works performed three days earlier, with the
addition of Schoenberg’s Lieder op. 22 in accordance with the wishes
expressed in Boulez’s letter from Harvard, quoted above. The second
Festival Hall concert took place on Wednesday 4 March 1964, preceded by
three days of intensive rehearsal, and was notable as being the first occasion
that Boulez had appeared in the UK as an orchestral conductor of one of his
own works – the cantata, Le Soleil des eaux, which was advertised as receiving
its first performance in this country. Again the programme featured works
by Debussy and Webern, the two composers on whom Boulez’s early
reputation as a conductor was founded:
Adagio and Fugue in C minor Mozart

Six Pieces for Orchestra op. 6 Webern
Symphonies of Wind Instruments Stravinsky

Ricerar (The Musical Offering) Bach-Webern
Images Debussy
It would be unsurprising if the acclaim accorded to Boulez’s conducting

début in London was a factor in the softening of attitudes towards his music,
even in those parts of the press which had previously been unreceptive.
An anonymous reviewer writing in The Times was guarded in response to
the performance of Le Soleil des eaux, but praised the first movement as
foreshadowing the sound world of Le Marteau sans maître and the Mallarmé
‘Improvisations’.25 Meanwhile, Glock acted decisively in seeking to secure
Boulez’s services for further periods with the BBC Symphony Orchestra, and
preliminary discussions evidently took place on the evening of the concert
concerning a projected tour to the USA the following year, and a possible
series of concerts in the UK during the subsequent three seasons. A letter to
Boulez’s then agent G. De Koos, written two days later, puts the series of
24 25
The Observer, 1 March 1964. The Times, 5 March 1964.
312 Peter O’Hagan
propositions in more formal terms, the letter concluding with the prophetic
words:
You will gather that his visit just now was a great event for the BBC orchestra and
for London music; and I wanted without delay to propose to him and to you
a further series of visits, before his diary becomes filled. The idea of a long-term
association between Mr. Boulez and the orchestra is one that fills me with
enthusiasm, and I hope that you will find that the dates I have given are all of
them practicable.26
It seems that the enthusiasm was a shared one, Boulez writing to Glock
shortly afterwards: ‘May I tell you again how much I enjoyed my London
visit: it was one of the most marvellous experiences I have ever had.’27
In the event, Boulez’s next appearances with the orchestra were on its
American tour in the spring of 1965, during which he conducted
a concentrated series of five concerts including two in New York on 1 and
7 May 1965. This tour included the orchestral piece Doubles, featured in
three concerts, including a broadcast performance from the final New York
concert. Strictly speaking, the work performed on this occasion was the
preliminary version of Figures – Doubles – Prismes, identical to that used
for Boulez’s contemporary recording of the work with the Hague
Philharmonic Orchestra,28 and including the first of three projected violon
chinois episodes not found in the original 1958 version of Doubles. Bearing in
mind that this was only the second time that the BBC orchestra had
performed one of Boulez’s works, and judging from the recording of the
broadcast, the remarkably assured performance of one of Boulez’s most
complex scores was testimony to the growing rapport between orchestra and
conductor.29 Within a few months Boulez made his début at the BBC Proms
in a concert with the orchestra on 7 September 1965, an event which took
place just a few days after a visit to the Edinburgh Festival where he
conducted the Hamburg Radio Orchestra. The programme in London was
as follows:
Quatre Etudes Stravinsky

Three Fragments from Wozzeck Berg
Symphonies of Wind Instruments Stravinsky
Images Debussy
Six Pieces for Orchestra op. 6 Webern
26 28
Letter dated 6 March 1964, BBC Written This recording is transcribed in ‘Darmstadt
Archives. Aural Documents’, Box 1, CD 3 (NEOS
27
Undated letter, probable date late spring Music GmbH, 2010).
29
1964, BBC Written Archives. Sound Archive, M5228W, British Library.
Reviews of the concert were very positive, and even Boulez’s own work,
given in its 1958 version for three solo singers, choir and orchestra, seemed
to glow in the general acclamation, typified by The Times reviewer, who
wrote of Le Soleil des eaux: ‘The web of sound is far too complex for the mind
to unravel all at once, but the ear is constantly held captive by kaleidoscopic
textures.’30 Indicative of the mutual respect which had developed between
Boulez and the orchestra was a memo to the players, written on the day after
the concert:
Before leaving London, I would like to express to you my deep gratitude for last
night’s concert.
I enjoyed not only the way you performed, but also – and not least – the effort you
made to rehearse in such a short time a programme which was not really easy!
Thank you for the music and for the friendship.
With my warmest regards, PB
It was at this time that details of Boulez’s work with the orchestra during the
following year were agreed. After protracted negotiations, he was eventually
committed to two periods of residency during the months of March
and May 1966, each of approximately two weeks. As well as containing the
by now standard repertoire of Boulez at this period, including both Webern
Cantatas and Le Martyre by Debussy, the programmes were characterised by
the increasing prominence being given to Boulez’s own music, featuring
the British première of Éclat, and an ingeniously planned tribute to
Mallarmé, in which Boulez’s Pli selon pli would be prefaced by Mallarmé
settings by Debussy and Ravel. This latter programme was to be the climax
of Boulez’s first residency in March, which also included concerts in
Nottingham and Leicester, prior to four days of intensive rehearsal for the
first performance by a British orchestra of Pli selon pli at the Royal Festival
Hall on 16 March 1966. For the performance, Boulez brought with him to
London an array of percussion instruments including cowbells, plate bells
and tubular bells, but in the event the occasion was something of an anticli-
max, since within the allotted schedule it was only possible to rehearse and
perform the first three of the five movements. Nonetheless, critical opinion
was positive, best exemplified by the influential Andrew Porter, who wrote in
the following terms: ‘Pli selon pli is one of the most important compositions
of our day I have no doubt at all. Intuition, faith say so; and I think reason
may be able to find reasons. The future of music is with him – not just with
Pli selon pli, but with a whole, varied complex of compositions which have
their roots in the past, in the fundamentals of musical experience, which
consolidate the present, and point to the future.’31 By the beginning of May,
Boulez was back in London for a second residency, which included the
30 31
The Times, 8 September 1965. Financial Times, 18 March 1966.
314 Peter O’Hagan
première of Éclat, in a recording made on 7 May 1966 in Maida Vale Studio 1

before an invited audience, preceded by a three-hour rehearsal on each of the
previous days. A packed schedule over a ten-day period concluded with
a concert at the Royal Festival Hall on 1 May, the programme consisting of
Webern’s final three works, Debussy’s Nocturnes and Stravinsky’s Le Sacre
du Printemps.
That Boulez was rapidly moving from the typecast of the previous
decade as an enfant terrible to a position at the centre of London’s musical
life is demonstrated by the fact that he was engaged to conduct no less than
three Prom concerts in a four-day period commencing on 30 August 1966
and followed by concerts the following evening and on 2 September. These
engagements came hard on the heels of the last of a series of performances
of Parsifal at Bayreuth, little more than a week beforehand, Boulez’s
first visit to the Festival, and an experience which would certainly have
taken a toll on his stamina. Now he was committed to three demanding
programmes, each of which would include a major work by Debussy –
respectively, La Mer, Nocturnes and Images – with the second evening
ending with a televised broadcast of Le Sacre du Printemps. Although the
repertoire for the programmes had been rehearsed and performed by
Boulez and the orchestra elsewhere during the previous eighteen months
or so, the schedule remained a daunting one, not least because of the
prospect of the first public performance in London of Éclat in the final
Prom. The recording of this performance leaves a mixed impression, not
least because of its tentative character – partly attributable to its duration
which, at over thirteen-and-a-half minutes, is some four minutes longer
than that of the 1983 commercial recording of the work by the Ensemble
Intercontemporain under the composer’s direction.32 It may well have
been the case that shortage of rehearsal time, combined with Boulez’s
novel deployment of the instrumental soloists in which the timing of
entries was dependent on cues from the conductor, may have resulted in
a certain inhibition in some of the performers, unaccustomed to the
operation of choice in an ensemble work. Be that as it may, critical
reactions to the series of Proms were fulsome in their praise of Boulez
the conductor, but somewhat guarded in their response to Éclat.
Peter Heyworth had attended the studio concert in May, and his reaction
to the new piece was equivocal, compared to his previously expressed
enthusiasm for Boulez’s music: ‘Once again, the ear is assailed by a collage
of sound effects, some imaginative and instantly arresting, others less
so . . . But that these sounds are not haphazardly put together was very
32
Reissued in ‘Boulez: Oeuvres Complètes’
(Deutsche Grammophon DGG4806828).
evident at a rehearsal I attended. Even where the players are allowed some
element of choice, Boulez knows exactly what those choices are, down to
the last inflexion.’33 Andrew Porter, writing after the Prom performance
also addressed the issue of choice in the work, observing that: ‘The web of
sound is attractive in timbre. The general feel of each episode comes
across, if not the logic or “sense” of the music. It is a work hard to imagine
without the composer-conductor himself in command, holding the reins
of nine interacting freedoms.’34
By this time plans were well underway for the following season with the
BBC orchestra, which would include a six-concert tour to Eastern Europe
in January 1967. Meanwhile, a major event in the autumn calendar of the
Royal Festival Hall, a Beethoven cycle to be directed by Otto Klemperer,
was plunged into jeopardy following an accident sustained by the octo-
genarian conductor. At short notice, Boulez stepped in to conduct two of
the scheduled concerts with the New Philharmonia Orchestra – his
first engagements with a London orchestra other than the BBC SO.
The first concert, on 27 September 1966, included the Second and Fifth
Symphonies, and he returned to direct the final concerts of the series, two
performances of the Choral Symphony on 30 October 1966 and
1 November 1966, praised by The Times as ‘not sensational, but right’.35
The success of these concerts would lead to further engagements with the
Philharmonia, but Boulez’s prime commitment remained with the BBC
during this period, a relationship strengthened further by the success of
the Eastern European tour, during the course of which he directed the
orchestra in a series of six concerts in Prague, Leningrad and Moscow.
Boulez’s subsequent letter to Glock expressed his thanks for the opportu-
nity to participate in the tour, ‘which, especially in Russia, was very
exciting for me’36 and received a response in the form of a letter from
the BBC’s Board of Governors, and signed by the Director of Sound
Broadcasting, which concluded: ‘The Board wishes you to know of its
very great appreciation of your key contribution to this highly successful
enterprise.’37
The impression of a meeting of musical minds between Boulez and Glock
is strengthened by the programmes for the 1967 season, which included
four concerts at the Festival Hall during Boulez’s residency in March.
The concept of themed programmes was further developed, with a mixed
programme of Webern, Schoenberg and Bartók on 8 March followed by
evenings devoted to Stravinsky (15 March) and Berg (20 March). Although
33 37
The Observer, 15 May 1966. Letter dated 10 February 1967, signed by
34
Financial Times, 3 September 1966. Frank Gillard, Director of Sound
35
The Times, 31 October 1966. Broadcasting.
36
Letter dated 6 February 1967.
316 Peter O’Hagan
the final concert on 29 March included a performance of Éclat, William

Mann, in a review of the third concert, under the headline ‘Boulez con-
ducts tour of Berg Museum’, struck a rare discordant note with regards to
programme planning, arguing in favour of a more radical approach:
‘While the B.B.C are lucky enough to have engaged him for their orchestra
(also experienced in modern idioms) and can grant him the time for
rehearsal that he needs, it would be more fruitful to use the opportunity
to let him bring forward recent works in which he believes than to
encourage him to duplicate programmes of classics that other conductors
can and do tackle in London.’38 As it happened, Boulez was already in
discussions with Glock about programmes for two Prom concerts he was
to direct later that year, the first of which included a recent work by Andrei
Volkonsky,39 Les plaintes de Chtchaza, which was performed on a specially
constructed platform in the middle of the arena, in order to maximise the
effect of its small ensemble in the vast spaces of the Albert Hall.
More ambitious still was the programme of the second concert, planned
around the first London performance of Stockhausen’s Gruppen for three
orchestras. It had been intended that Stockhausen would join Boulez and
Edward Downes as one of the three conductors, but in the event
Stockhausen withdrew his support for the project on the grounds that
the proposed rehearsal schedule was not in accordance with his written
stipulation,40 and he was replaced by Michel Tabachnik. It transpired that
Stockhausen’s concerns proved unjustified to a degree since the perfor-
mance won considerable acclaim. In a letter of thanks to Boulez, written
the day after this concert, Glock began: ‘I just want to thank you again for
those three outstanding concerts,41 and the untiring work you put into the
Stockhausen’; he noted the highly satisfactory attendance figures for
the Gruppen Prom, before continuing: ‘a remarkable state of affairs
which is naturally chiefly due to a certain Pierre Boulez, but also I think
to the fact that there now seems to be a lively response at the Proms to
any programme of contemporary music which has bold outlines and
a substantial content’.42 Boulez’s reply included the comment: ‘I like
very much to work with you – that is easy and makes fun! (and
sense . . .).’43
38 40
The Times, 21 March 1967. Stockhausen to Glock, 17 June 1967, MS
39
Andrei Volkonsky (1933–2008). Russian Mus 956, William Glock Collection, British
composer whose adoption of serial techni- Library.
41
ques in the mid-1950s led to his music being Between the two London concerts, Boulez
banned from performance in the USSR. Les conducted the orchestra in a concert at the
plaintes de Chtchaza, for soprano and Edinburgh Festival.
42
instrumental ensemble, was completed in BBC Written Archives.
43
1961. Glock, Notes in Advance, p. 138.
At least as important to Glock as Boulez’s conducting activities was his

recognition of the primacy of his creative work, and hence a desire to
encourage and promote Boulez’s music. The programmes for his concerts
at the 1968 Proms season included a performance of Le Marteau sans
maître, but Glock was evidently anxious to commission new music from
Boulez. With regard to Éclat, discussions had already taken place the
previous year concerning an extended version of the work for performance
as early as the orchestra’s concert at the Berlin Festival that autumn.
Although Boulez indicated early in the year that the new version would
not be ready in time, Glock evidently persisted, and a letter to Boulez’s
London agent, Howard Hartog, dated 19 May 1967 gives an indication of
his determination to secure the work’s première: ‘I forgot to ask you on
Wednesday [17 May] about Éclat, which Pierre Boulez had promised
to conduct for the first time in the complete version on 23 June 1969.
I hope I may still take it for granted that we will have the first performance
of this work – now, presumably in 1970.’ Meanwhile, plans were being
prepared for the 1969 season which would include a Festival Hall concert
in February, entirely devoted to Boulez’s music – the second book of
Structures and Pli selon pli, of which a commercial recording was planned
for CBS during May of the same year. By now, there were certain parallels
developing between Boulez’s relationship with the BBC Symphony
Orchestra and his work with SWR beginning in 1958, a period which had
seen the first performances successively of Poésie pour pouvoir, Pli selon pli
and Figures – Doubles – Prismes. Specifically, there are echoes of the friend-
ship with Heinrich Strobel in Boulez’s identification of William Glock as
someone who shared his musical ideals, and with whom he could establish
a working relationship founded on mutual trust.
Perhaps the fact that the next two premières of Boulez’s music in London
were to be with organisations other than the BBC increased Glock’s sense
of urgency with regard to securing his future services, and specifically
the première by the BBC Symphony Orchestra of a new version of Éclat.
A concert given by Boulez with the Philharmonia Orchestra on
1 December 1968 included the first performance of Livre pour cordes, an
arrangement for string orchestra of movements from Livre pour quatuor
(1948–9). In the event, one of the two movements was withdrawn from the
Royal Festival Hall concert because of lack of rehearsal time, and the two
completed movements were not heard until a week later at a concert in
Brighton. Doubts about the direction of Boulez’s energies which surfaced in
the reception of Éclat were more pronounced in the critical response to
the new string piece, Peter Heyworth writing: ‘Boulez has confessed
his frustration with the inability of string quartets to measure up to the
formidable difficulties of Livre pour quatuor, and it is understandable that he
318 Peter O’Hagan
should have wished to recast a neglected work in a form in which he can

himself assure adequate performances. But when he goes on to invoke
Webern’s Opus 5, which similarly started life as a string quartet, and was
subsequently transcribed for string orchestra by Webern himself, as
a happy precedent for what he has undertaken here, I begin to wonder if
he is not allowing his experience as a conductor to get the better of his
musical judgement.’44 A London Sinfonietta concert the following month
(21 January 1969) included the British première of Domaines in its
original solo clarinet version, complete with the spectacle of the instru-
mentalist (Alan Hacker) moving round the platform between six music
stands. The complementary nature of the visual element in the work
prompted a discursive article from William Mann, during the course of
which he speculated on the possibility of there being ‘another version of
Domaines in which there are groups of instrumentalists posted around
each music-stand [sic], commenting on what the clarinettist plays
there’.45 The reviewer was evidently unaware that Boulez himself had
directed the première of an ensemble version of Domaines less than
a month previously,46 and the mobile elements must have struck
a somewhat incongruous note in the solo clarinet version performed in
London.
Meanwhile, a few months prior to the performances by the New
Philharmonia and the London Sinfonietta, a set of circumstances arose
which enabled Glock to take decisive action, the catalyst being the immi-
nent departure of the BBC Symphony Orchestra’s Chief Conductor, Colin
Davis, to succeed Georg Solti at the Royal Opera House. Glock may well
have been emboldened by the success of three concerts which Boulez gave
with the Orchestra on successive Wednesdays in February 1968, the last of
which, on 21 February, included a repeat performance of Gruppen to
a packed Royal Festival Hall audience. Seizing the moment, according
to Glock’s subsequent account,47 he met with Boulez in Holland on
21 May 1968 and offered him the post of Chief Conductor of the BBC
Symphony Orchestra from September 1971, an appointment which was
formally announced at the beginning of 1969, just two days after the first
London performance of Domaines. It was Boulez’s expressed wish to
reform musical life in London, and a draft contract prepared at this
time stipulated a period of five months a year working with the orchestra,
to include a total of forty concerts per season over a three-year period,
44 46
The Observer, 8 December 1968. On 20 December 1968. The performers
45
The Times, 24 January 1969. were Walter Boeykens (clarinet) with the
Orchestre symphonique de la Radio Belge.
47
Glock, Notes in Advance, p. 139.
with the option of further extensions by mutual agreement. Even at this

stage, Glock was anxious to safeguard Boulez’s creative work, and mindful
of his increasing conducting commitments, not least with other London
orchestras including by this time the LSO, sought to limit his conducting
activities in the UK exclusively to the BBC.48 As further proof of the BBC’s
commitment to promoting Boulez’s activity as a composer, arrangements
were by now in place for the commercial recording of Pli selon pli
following the first complete performance of the work in London at
a RFH concert scheduled for 7 May 1969.49
There was no inkling of any contractual problems with the BBC when
Boulez departed for an American tour in early 1969.50 A letter sent by
Boulez from Boston was partly taken up by detailed discussion of the
programme for the Festival Hall concerts in 1970 which would include
Glock’s proposed programme juxtaposing Éclat with Berlioz’s Symphonie
Fantastique. It is only in the last paragraph when Boulez commented on
the pervasive malaise which characterised American concert life that one
might, with hindsight, anticipate what was to unfold. Even so, the letter
concluded with an affirmation of solidarity: ‘From all these contacts,
I conclude that the time is ripe for deep and significant reforms; and we
are in the right time to give, with the BBC, of what can be a model of a new
conception.’51 All of these plans were thrown into disarray by an offer
from the New York Philharmonic to become its new Principal Conductor,
after a series of concerts with the orchestra the following month. Various
accounts of this sequence of events have been written,52 but the practical
consequences were formulated at a meeting in London on 29 May 1969
between Boulez, Glock, Hartog and Carlos Moseley, Director of the
New York Philharmonic. A memo from Glock to the then Director of
BBC Radio, Ian Trethowan, the following day summarised the agreed
changes to Boulez’s work with the BBC in order to accommodate his
acceptance of the post with the New York Philharmonic: the commitment
to the BBC would be reduced to four months and thirty-two concerts
a year; and ‘these two appointments in New York and London will be all
the conducting he will undertake’.53 Glock’s continuing unease at the
situation, expressed in an aside to a BBC colleague the same day, ‘Let’s
48 51
BBC Written Archives. Boulez to Glock, postmarked
49
The recording was issued by Columbia 11 February 1969, MS Mus 953, William
(no. 72770) in 1969. Glock Collection, British Library.
50 52
Boulez was in the USA from 12 January to See Peyser, Boulez: Composer, Conductor,
20 April 1969. His appearances in New York Enigma, pp. 180–6; Heyworth (in Glock (ed.),
were preceded by engagements in Los Pierre Boulez: a Symposium), pp. 35–6; and
Angeles and Boston, and followed by con- Glock, Notes in Advance, pp. 139–40.
53
certs in Cleveland. BBC Written Archives.
320 Peter O’Hagan
hope he doesn’t kill himself’,54 could only have been increased by the
subsequent determination by Boulez to continue his association with the
Cleveland Orchestra in addition to his commitments in London and
New York.
Glock must have felt that his concerns were all too well founded when
a scheduled performance of a new version of Éclat at a Festival Hall
concert on 15 April 1970 had to be postponed, but eventually his persis-
tence was rewarded, and the world première of Éclat/Multiples took place
at the Royal Festival Hall on 21 October 1970. The performance was
preceded by an intensive series of rehearsals over the previous two
weeks, but even so, the circumstances were evidently far from ideal, and
the announcer for the live BBC broadcast of the première acknowledged
that the performance was of an incomplete work, consisting of parts I and
II of a larger-scale work of unspecified length. Edward Greenfield writing
after hearing a repeat performance in Paris on 9 November 1970 offered
further insight into the situation prior to the London performance:
‘Report has it that when Pierre Boulez conducted his latest work, Éclat/
Multiples in London less than three weeks ago, the ink was barely dry on
many of the pages.’55 According to Greenfield, some further revisions and
expansions occurred prior to the Paris performance, which he compared
favourably to the London première: ‘this time there was no mistaking the
energy of [sic] inspiration’. It is worth observing in this context that Éclat/
Multiples in its expanded form consisted of almost fifteen additional
minutes of music, occupying over one hundred pages of score –
a considerable quantity of new material. Any sense of exasperation at
the performance of an incomplete work might have been tempered by
a realisation that the gradual evolution of works was characteristic of
Boulez’s working methods, extending back to Le Marteau san maître,
the first performance of which was of an incomplete version. To be sure
a recording of the Royal Festival Hall performance56 leaves a somewhat
mixed impression, but despite a certain lack of continuity between the
sections and (in the broadcast) some balance issues, the overall effect is
a coherent one, and judging by the enthusiastic audience response, the
occasion as a whole was highly successful.
By now, plans were in place for Boulez’s forthcoming role as Chief
Conductor, with a schedule of his availability over the next three seasons.
In an interview in The Times, the month after the compromise agreement
54 56
Memo to Martin Turnell, 30 May 1969, Sound Archive, C1398/0572, British
BBC Written Archives. Library.
55
The Guardian, 10 November 1970.
reached in London in May 1969, Boulez expressed his aspirations in the

following way:
Although there are still two years to go before I take over, my ambition is to build an
entirely new repertoire; a repertoire in which the classics and contemporary works
are of equal status and importance. This work will go on in a parallel sense in both
cities, New York and London.
What we need today is a new public which is not so conservative and is more
receptive to modern works. Regeneration of the concert-going public, if you like,
and there will be educational concerts for young people. Both my contracts with the
two orchestras run until 1974, and I would hope in this time to build up what I would
like to call a ‘model’, a repertoire which reflects my own personality.57
According to Glock’s later account, one of the practical considerations in

this process of regeneration was to increase the range of available concert
venues, with the recently refurbished St John’s Smith Square and the Round
House designated for concerts of choral and contemporary music
respectively.58 A first product of this policy was a late-night BBC Prom
concert at the Round House on 6 September 1971, a date which fell within
the first week of Boulez’s appointment as Chief Conductor. Ligeti’s
Aventures and Nouvelles Aventures formed the final items of the concert,
but a first British performance of Messiaen’s Sept Haïkaï had to be cancelled
because the programme would have overrun. Peter Heyworth was scathing
about the occasion and in particular the quality of the remainder of the
programme: ‘The Round House provides exactly the informal and intimate
setting a certain sort of contemporary music calls for, and the basic idea of
using it for a late-night Prom last Monday was a good one. But in the event
the evening misfired badly, simply because too much of the music was
simply not good enough.’59 Undeterred, Boulez and Glock went ahead
with a scheduled series of four concerts at the Round House the following
season, arranged to coincide with Boulez’s residencies with the BBC
Symphony Orchestra in January and May. Each of the four late-night
concerts included a specially commissioned work by a British composer,
culminating in the first performance of Blind Man’s Buff by Peter Maxwell
Davies at the final concert on 29 May. With regard to the format of the series,
Stanley Sadie made the point, ‘As regular listeners to Radio 3 will know, it
has been a patchy series. The experiments ought to continue, but
more thought might profitably be given to ways of enlivening both the
programmes themselves and the following question time: surely there is
something wrong with a discussion on new music when not only are no
tempers lost but no one even seems to care deeply about what is happening
57 59
The Times, 13 June 1969. The Observer, 12 September 1971.
58
Glock, Notes in Advance, p. 141.
322 Peter O’Hagan
and why. It was all disappointingly good natured.’60 The concerts of new
music were designed to complement the more mainstream repertoire, which
in this first season consisted of a retrospective of the music of Haydn and
Stravinsky – programming consistent with Boulez’s stated aim of juxta-
posing classical and contemporary works. His three concerts at the
Royal Festival Hall during the season included a highly demanding all-
Stravinsky evening on 26 January 1972, and this was preceded by a Haydn
programme at St John’s on 10 January. Each of these concerts was
followed by a programme of new music at the Round House (on 17 and
31 January), and the period of these BBC engagements coincided with
a series of performances of Debussy’s Pelléas et Mélisande at the Royal
Opera House – a truly phenomenal workload. In addition, as Chief
Conductor Boulez was heavily involved in the 1972 Proms season,
appearing no less than seven times, in programmes ranging from
a performance of Beethoven’s Missa Solemnis on the opening night,
21 July, to a Round House concert on 7 August which included a repeat
performance of Davies’s Blind Man’s Buff, as well as a first hearing at
the Proms of his own cummings ist der dichter. If not all of these projects
were equally successful, there was no doubting Boulez’s commitment to
changing fundamentally London’s concert life: as he put it in an interview
with Peter Heyworth at the beginning of this first season, ‘What I want to
do is create models of concert life in two cities – London and New York.
After that, anyone can do it.’61
In the meantime, only a few months into Boulez’s contract with the
BBC, change was already afoot with the announcement in January 1972
that Glock would be relinquishing his post as Controller of Music on
30 November that year, although he would retain responsibility for the
Prom concerts in the following season. Boulez greeted the news with
dismay, the more especially since plans were already in place for
a continuation of the Round House series in the 1972–3 season, as well
as a major retrospective centred on the music of Berlioz and Schumann:
‘You can be sure that it is without any pleasure at all that I see that you are
leaving the BBC . . . Why are you sixty-three?’62 After a protracted series
of negotiations, it was announced in August 1973 that Boulez’s contract
would be extended for one year until 1975, although already there was
a sense that he was winding down his involvement in a project so closely
dependent on his relationship with Glock. ‘In the coming season he
intends to continue his Round House series. These will now consist of
60 62
The Times, 31 May 1972. Boulez to Glock, 25 February 1972, MS
61
The Observer, 3 October 1971. Mus 948, William Glock Collection, vol vi, ff.
138–216, British Library.
one new piece, one novelty repeated from a previous series and a British
composer introducing a modern classic . . . His own speaking contribu-
tions will be more confined than in the past.’63 Only two months after this
announcement came the confirmation that Boulez would be leaving the
post of Chief Conductor in August 1975 to prepare for his new role as
Director of IRCAM, scheduled to open the following year.
In attempting to assess Boulez’s impact on London’s concert life over
the decade, one is drawn to the inescapable conclusion that his period
as Chief Conductor was rather less satisfying than the preceding years,
when his revelatory performances of Debussy and the Second Viennese
School in particular decisively shifted reception of their music. Glock
subsequently acknowledged that Boulez had not achieved the funda-
mental change which he had sought in London’s concert life, adding:
‘He says that if he had been a conductor and nothing else, he might
have persevered for longer and succeeded. On the other hand, he would
surely not have been fired with a conception never before imagined in
London’s concert-giving.’64 An easily forgotten aspect of this period of
Boulez’s work in London is the accompanying transformation in his
reputation as a composer – somewhat ironic in view of the frequently
voiced complaint that his creative work was being sacrificed with the
expansion of his conducting commitments. In fact, compared to the
1950s, when his work was rarely heard in the UK, London now became
a focal point for first performances of his music. As we have seen, Livre
pour cordes and Éclat/Multiples had their premières in London, and on
17 June 1972, . . . explosante-fixe . . . in a preliminary version for flute,
clarinet and trumpet was performed as part of a London Sinfonietta
concert at St John’s Smith Square.65 A new, greatly extended version,
for eight instruments and electronics received its UK première on
17 August 1973 at the Proms played by members of the BBC
Symphony Orchestra with Boulez conducting. If it leaves an overall
impression of a project not yet completely realised, the recording of the
performance66 is a fascinating study in the work’s evolution, and
a testament to Boulez’s ongoing mission to integrate electronic trans-
formation of sound – a goal shortly to be achieved at IRCAM. Boulez’s
final season with the BBC Symphony Orchestra before his return to
Paris was marked by the world première of Rituel in memoriam
Maderna, enthusiastically hailed by Peter Heyworth as ‘music that,
63
The Times, 13 August 1973. a concise account of the work’s genesis up to
64
Glock, Notes in Advance, p. 142. that point.
65 66
See Bradshaw, ‘. . . explosante-fixe . . .’, Sound Archive, 1CDR00 18338, British
Tempo (September 1973, pp. 58–9), for Library.
324 Peter O’Hagan
once heard, stamps itself indelibly on the memory’.67 As Heyworth

noted in the same article, during the ten years previous to Rituel
Boulez had written a considerable quantity of music despite his con-
ducting commitments, and it is arguably the case that his creative work
(inextricably bound up with his work at the BBC) rather than his
conducting activities represents his most enduring legacy to London’s
concert life during this period.
Although Boulez continued to appear with the BBC Symphony
Orchestra in the aftermath of Glock’s retirement, and the ending of
his period as Chief Conductor, inevitably there was a reduction of the
commitment on both sides. Nonetheless, Boulez continued as a regular
guest at the Proms, and featured in BBC concerts at the Festival Hall in
the subsequent seasons. Furthermore, he continued to enjoy warm
personal relations with Glock, returning to conduct a seventieth birth-
day tribute to the former Controller, which featured a performance of Le
Marteau sans maître. The event took place in the Concert Hall of
Broadcasting House on 3 May 1978, just a few days before Boulez
departed with the orchestra on a short European tour, to include per-
formances of his Rituel. By now, Glock had moved on to fresh pastures
as Musical Director of the Bath Festival from 1975, and at the end of his
tenure there, Boulez was invited to write a short piece to mark the
occasion. Although his contribution was too late to be included in the
celebratory concert in Bath, the score duly arrived with a dedication,
‘pour William Glock, Bath, 8.6.84’. In an accompanying letter, Boulez
explained the title: ‘Why Dérive? Because it is a deviation of a sort on
some chords and rhythms of Répons.’68 It is rather apt that the piece is
an offcut of the work dedicated to Heinrich Strobel, who as we have
observed, played a role at an earlier stage in Boulez’s conducting career
not dissimilar to that of Glock in a later era. The work received its first
performance on 31 January 1985 as part of a London Sinfonietta concert
at St John’s Smith Square, conducted by Oliver Knussen, and was
described by Peter Heyworth as ‘an exquisite little miniature’, and
with the added comment, ‘There is not a bar that fails to bring surprises,
yet the music’s movement is never wayward.’69
Boulez and Glock continued to meet and correspond right up to the time
of Glock’s death in 2000, and Dérive was chosen as the opening work in
a memorial concert, broadcast live from St John’s on 18 December 2000.
The unseen presence behind both Répons and Dérive is Paul Sacher, and the
67
The Observer, 6 April 1975. Glock Collection, vol vi, ff. 138–216, British
68
Boulez to Glock, undated letter Library.
69
[early June 1984], MS Mus 948, William The Observer, 3 February 1985.
six-note cipher derived from the letters of the Swiss conductor’s name is
common to both works. Sacher was in turn the dedicatee of Boulez’s most
extended work of the following decade, sur Incises, and at the time of
Sacher’s death in May 1999 Boulez was working on an extension of the
original piano solo work, Incises. A letter to Glock sent at the beginning of
the new millennium mentions slow progress with the piano piece, unsur-
prising in view of the fact that he was about to embark on a European tour
with the LSO, which included five concerts at London’s Barbican Hall over
a five-week period beginning on 26 January 2000. In the middle of
a demanding schedule, Boulez returned to Paris, from where he wrote for
a final time, complaining of the time pressure he was under, but promising to
call following his return to the UK. The letter has a poignant tone, not least
because Boulez clearly still identified Glock as one of the few he was able to
confide in concerning his own music:
You are right. The end of sur Incises is derived from Les Noces. I don’t know what
tape you have. But in Edinburgh (August 18) the chords were bare, like in
Stravinsky. Since then, I have interspersed reminiscences of previous material,
which are more convincing as a solution for the ending.
I have reworked on [sic] some electronic counterpoint to the violin in Anthèmes 2,
before it is recorded. And it is more satisfying than before, more continuity, more
interferences between the various ideas: which is always a preoccupation, almost an
obsession with me: unity and diversification.70
Within a few months, Glock was dead.71 An obituary, published in

The Times the day afterwards, summed up his work at the BBC, a period
which defined his career:
. . . it was in Pierre Boulez that he found the man who most closely shared his ideals.
He had Boulez conduct the orchestra for the first time in 1964, and Boulez’s
appointment as chief conductor in 1971 was the culminating achievement of his
BBC career, as well as the symbol of his legacy to the Corporation.72
For Boulez, the loss in little over a year of two figures so crucial to him over
a period extending back virtually half a century would undoubtedly have had
a great impact, not least because Sacher and Glock were in their different
ways inextricably associated with what is arguably his greatest creative
achievement – the works of the SACHER cycle. Work on the solo piano
piece seems to have come to an abrupt halt in the aftermath of this loss, with
a considerable quantity of detailed sketches remaining unrealised in the
published version of Incises. The sombre, almost funereal coda of the work
in its final version (2001) stands in stark contrast to the dazzling brightness
70 71
Boulez to Glock, 15 February [2000], MS Glock died on 28 June 2000.
72
Mus 948, William Glock Collection, vol vi, ff. The Times, 29 June 2000.
138–216, British Library.
326 Peter O’Hagan
of the ending of sur Incises, with its evocation of the close of Les Noces.
The note of utter desolation casts a shadow retrospectively over the work, in
a manner which recalls the final movement of Le Visage nuptial, ‘Post-
Scriptum’. It is the purest speculation, but the tolling bells in the lower
bass register in the final bars of Incises have a valedictory character: an
epitaph to the memory of both men, as well as marking the end of an era
in Boulez’s creative life.
13 Tartan from Baden-Baden: Boulez at the 1965
Edinburgh International Festival
Edward Campbell
While the history of Pierre Boulez’s involvement with musical life in

England, more particularly in London, is fairly well known, his activities in
Scotland – most specifically at the Edinburgh International Festival – are
perhaps less celebrated. Boulez made a number of important visits to the
Scottish capital, from his first trip in 1948 to his last performance there in
2004. Noting only some of the most outstanding highlights, his appearance
in 1948 as the Ondes Martenot-playing musical director of the Renaud-
Barrault theatre company was his first performance anywhere in the United
Kingdom; Edinburgh staged the UK première of Pli selon pli in 1965, as was
also the case with the 1994 performance of . . . explosante-fixe . . .. More
significantly again, the Edinburgh performance of sur Incises at the 1998
festival was the world première of the work. The festival has in addition been
the occasion for two full-scale retrospectives of his work, in 1965 and 1998; it
has been a platform on which he has staged generous surveys of twentieth-
century musical modernism, and while there he has given festival lectures
and participated in public conversations, interviews and press conferences.
While the trajectory of his long-time relationship with the festival is a rich
topic in itself, the current chapter is restricted in scope to the 1965 festival
and the six concerts in which he appeared as featured composer, conductor
and pianist – the first full-scale retrospective of his work anywhere in the
world.
By 1965, Boulez was widely known as a composer and theorist, as
a leading figure of European musical modernism, and he had
a developing reputation as an orchestral conductor of world class. He
had visited England in 1956 and 1957 and, as Sir William Glock notes, it
was thanks to Dartington that by 1960 his Le Marteau sans maître, the
Sonatine for flute and piano, and the Mallarmé ‘Improvisation’ ‘Une
dentelle s’abolit’ had had their first performances in England.1 He may
just as well have stated that these were the first performances of works by
1
Glock, Notes in Advance , p. 59. See
Chapter 12 in this volume.
327
328 Edward Campbell
Boulez anywhere in the United Kingdom. Having seen Boulez conduct

a Domaine Musical concert in 1957 and a performance of Wozzeck in
1963, Glock invited him to lead the BBC Symphony Orchestra as guest
conductor in a number of concerts in the spring of 1964,2 a move which
marks the beginning of a much more sustained relationship with the
United Kingdom. It is against this background that we can gauge the
significance of Boulez’s visit to Edinburgh in 1965.
According to Joan Peyser, Boulez’s decision ‘to replace Michel de Koos,
the manager he had inherited from Hans Rosbaud, with Howard
Hartog . . . was stimulated by Hartog’s efforts to bring the Südwestfunk
to Edinburgh for a full-scale Boulez festival’.3 While Hartog’s plan did not
succeed at first, it did so in 1965 when Edinburgh staged the most
extensive retrospective of Boulez’s work yet undertaken. This was not
only significant at the time, but it remained in 1976, the year of Peyser’s
account, the largest retrospective of Boulez’s works to have been staged
anywhere in the world.
The 1965 festival was the last under Lord Harewood’s4 direction
(1961–5), a period now acknowledged as something of a musical golden
age in which Edinburgh audiences were exposed to significant retrospec-
tives of the music of Schoenberg (1961), Shostakovich (1962), Bartók,
Berlioz, Britten (1963), Janáček (1964), Boulez and Tippett (1965).
As David Haworth notes, ‘the Festival was dull and in atrophy for several
years before Lord Harewood’s arrival’ and ‘he undoubtedly restored its
artistic integrity and gave it a proper sense of enterprise which had also
been missing’.5
Boulez Studies and Reception Theory

Beyond celebrating the fact of the 1965 retrospective in terms of the six
concerts, what was played, where and when, this chapter focuses also on the
wealth of critical coverage of the event in a range of newspapers and journals
in the United Kingdom and beyond, at once important indicators of its
perceived value and significant records of the responses of a privileged group
of listeners and observers. Working largely from these sources, the chapter
operates within the territory of reception theory or, what we might call
tentatively, listener response theory.
2 5
Glock, Notes in Advance, p. 134. Haworth, ‘Edinburgh Festival: What Comes
3
Peyser, Boulez: Composer, Conductor, Next?’, Socialist Commentary
Enigma, p. 172. (November 1965, p. 35).
4
George Henry Hubert Lascelles (1923–
2011), 7th Earl of Harewood.
329 Boulez at the 1965 Edinburgh International Festival
Within Boulez scholarship, a great deal of work has been undertaken on

the study of the composer’s preparatory sketches and working methods
whereby he passes from initial idea to completed score, what Jean-Jacques
Nattiez terms the ‘poietic level’ of analysis.6 In a similar way, there are
a number of studies of Boulez’s completed scores, which operate on what
Nattiez more problematically terms a ‘neutral level’ of analysis. Much less
effort however has been expended on approaching Boulez’s compositions in
terms of Nattiez’s third way, the ‘esthesic level’, in which the reception of the
composition is the chief point of interest.
Nattiez’s esthesic level focuses on everything relating to the reception of
a work of art, a mode of analysis that corresponds in some respects to the
reception theory of Hans Robert Jauss, Wolfgang Iser and others in the late
1960s and 1970s, and which indicates a shift in focus from the study of
complete and final literary texts to texts as they are encountered by their
readers.7 Much of this theory that was formulated with reference to literature
holds equally well for music, and Mark Everist notes that studies of musical
reception, understood as audience response, both predated the rise of recep-
tion theory in the 1960s and 1970s and outlasted its relative decline in the
1980s.8 Carl Dahlhaus, following Walter Benjamin, refers to such studies as
concerning themselves with ‘the “after-life” of musical works’,9 and Everist
suggests with some justification that the terms ‘literature’ and ‘literary’, as
used by Jauss, can easily be replaced with ‘music’ and ‘musical’ to good
effect.10
While reception theory in music can embrace a wide range of activities
and sources including ‘performance history’, ‘critical reception’ (journal-
ism) and ‘scholarly or theoretical responses’,11 the current study is
restricted to journalistic responses only. To this degree, the published
reviews from the 1965 Edinburgh Festival are invaluable documents
providing a rich spectrum of interpretive responses to a body of work
that had never before been heard in such a concentrated way. Given that
not all of the compositions were available in recorded format or as
published scores, the retrospective allowed listeners to form judgements
about Boulez’s compositional development and to make comparative
evaluations that hitherto were impossible.
The articles and reviews document something of the critics’ pre-festival
expectations, and a number of substantial pieces were produced in anticipa-
tion. Viewed in the light of the later reviews, we can evaluate the extent to
6 8
Nattiez, Music and Discourse, pp. 11–13. Everist, ‘Reception Theories’, p. 381.
7 9
Hans Robert Jauss (1921–1997) and Ibid., p. 379.
10
Wolfgang Iser (1926–2007) are considered Ibid., p. 383.
11
the founders of the Constance School of Ibid., p. 379.
reception theory.
330 Edward Campbell
which expectations were fulfilled, confounded or frustrated. The critics

display an impressive range of strategies for making sense of works which
at the time undoubtedly seemed like music from another world. They are
remarkably inventive and imaginative in articulating micro-strategies for
listening, recounting what they experienced in the act of trying to make sense
of this music, which largely undermined traditional listening expectations.
Anticipating the 1965 Festival

The wealth of press coverage of Boulez’s six concerts at the 1965
Edinburgh International Festival demonstrates beyond doubt that they
generated a great deal of interest at home and abroad, receiving wide
coverage in English, German and French language newspapers and jour-
nals. Claude Samuel, in an article with the caption ‘400 journalists for
Marlène Dietrich and Pierre Boulez’, described it as ‘one of the most
fantastic festivals in the world’, noting that it attracted 400 journalists
and 90,000 visitors to the city.12 Without hope of summarising the totality
of news outlets covering Boulez’s activities, his concerts and media
engagements were written about in a small number of articles by French
journalists, a larger number of pieces in the German press, and there are
miscellaneous reports in the New Zealand, South African and US press.
All of the principal British broadsheet newspapers carried an impressive
number of articles with fairly detailed reports of each concert, sometimes
more than one in the same newspaper, as well as more evaluative articles.
The Guardian had both Gerald Larner and Neville Cardus in Edinburgh,
each with a very different take on events; The Daily Telegraph was repre-
sented by Peter Stadlen and Martin Cooper; there was an unidentified
critic from The Times, and The Sunday Times had Felix Aprahamian.
The events also received significant coverage in the Scottish press, with
Conrad Wilson writing for The Scotsman, and an unidentified critic, as
well as the partially identified ‘E.B.S.M’ and ‘T.M.’ writing for The Glasgow
Herald. Post-festival reflections were also produced by Philip Hope-
Wallace in the Guardian, Peter Heyworth in The Observer, Desmond
Shawe-Taylor in The Sunday Times and David Douglas in The Glasgow
Herald. The broadsheets apart, it is rather striking that several local
Scottish newspapers including the Perthshire Advertiser,13 The Inverness
Courier,14 The Northern Scot,15 the Fife Free Press,16 the Edinburgh
12 14
Samuel, ‘Un Ahurissant Festival’, Le The Inverness Courier, 29 January 1965.
15
Nouveau Candide, 20 and 26 September The Northern Scot, 2 January 1965.
16
1965. The Fife Free Press, 9 January 1965.
13
The Perthshire Advertiser, 2 January 1965.
Evening News,17 the Paisley Daily Express18 and the magazine Scottish
Field19 all drew attention to the presence of Boulez at the forthcoming
Festival.
For Neville Cardus, who covered the festival for the Guardian and
The Spectator, the 1965 programme seemed ‘a motley show, disparate in
content and atmosphere . . . a Festival more of appetite than of taste . . .
Haydn, Boulez, Tippett and Marlene Dietrich’,20 with altogether ‘less-known
names’ who now have a chance ‘to assert themselves’ ‘in the absence of
personalities of the stature of Walter, Furtwängler and Mitropoulos’.
In contrast, visual art critic David Irwin writing in The Burlington
Magazine in October 1964 was already looking enviously to the prospect
of Boulez conducting some of his own compositions while opining ‘if only an
art show could match up to such enterprise!’.21
In terms of how Boulez was presented at the 1965 festival, it is stated in
the programme booklets for all of his concerts that ‘as well as being one of
the most remarkable composers in Europe, springboard and fountainhead
of the avant-garde, Pierre Boulez is now regarded as one of the outstand-
ing conductors of his generation as well’, to which it is appended that ‘he is
also a mathematician’.22 Conrad Wilson of The Scotsman was looking
forward to Boulez’s visit but simultaneously rather conflicted about it.23
While noting that ‘Boulez is one of the few living composers whom
Stravinsky has deemed worthy of interest’, he adds that ‘he is also one of
the most difficult and, at times, daunting figures of modern music’.
Nevertheless, ‘even at their most bafflingly revolutionary, his works
compel attention’ and ‘one is aware of the powerful intensity of the
thought behind them, even when one fails to follow their argument’.
The unidentified critic for The Glasgow Herald predicts that the combina-
tion of Boulez and Messiaen ‘should be testing enough for even the most
avant-garde listener’,24 and there is clear ambivalence in the mind of this
reviewer, for whom ‘none of this is going to be easy’, yet ‘it could well be
not only interesting but exciting listening, and this is certainly the stuff
festivals should in part at least be made of’.
In all, Boulez was involved as conductor or pianist in six concerts in which
six of his own compositions for various instrumental forces were performed
(see Figure 13.1). As Susan Bradshaw noted, ‘because they span the twelve
17 22
Edinburgh Evening News, 14 August 1965. No author named. Edinburgh Festival
18
The Paisley Daily Express, 28 August 1965. Programme, 29 August 1965, p. 2.
19 23
Lindsay, Scottish Field, 1965. Wilson, ‘Music: Three Shapely Themes’,
20
Cardus, The Spectator, 2 September 1965, The Scotsman, Weekend Magazine,
p. 14. 21 August 1965, p. 5.
21 24
Irwin, ‘Edinburgh Festival’, The Burlington ‘Music at the Festival: Testing even the
Magazine, October 1964, p. 474. Avant-Garde’, The Glasgow Herald,
21 August 1965, p. 6.
332 Edward Campbell
Date Programme Performers Venue
28 August Beethoven: Symphony No. 2 Hamburg Radio Symphony Usher Hall

8.00 p.m. Messiaen: Oiseaux Exotiques Orchestra
Webern: Symphony op. 21 Pierre Boulez (conductor)
Debussy: La Mer Yvonne Loriod (piano)
29 August Boulez: Pli selon pli (UK première) Hamburg Radio Symphony Usher Hall
8.00 p.m. Orchestra
Pierre Boulez (conductor)
Halina Lukomska (soprano)
31 August Mozart: Sonata in A, K331 Yvonne Loriod (piano) Freemasons’ Hall
11.00 a.m. Messiaen: ‘La Rousserolle Effarvatte’
(from Catalogue d’Oiseaux)
Debussy: Études XI and VI
Boulez: Sonata for Piano No. 2
1 Sept Mozart: String Quartet in E♭, K428 Parrenin String Quartet Leith Town Hall
11.00 a.m. Boulez: Livre pour quatuor
Beethoven: String Quartet op. 18, no. 6
2 Sept Boulez: Sonatine for flute and piano Pierre Boulez (piano) Freemasons’ Hall
11.00 a.m. Boulez: Structures, Deuxième livre Yvonne Loriod (piano)
Debussy: Syrinx Severino Gazzelloni (flute)
Debussy: En blanc et noir
4 Sept Boulez: Le Marteau sans maî tre The New Music Ensemble Leith Town Hall
11.00 a.m. Schoenberg: Pierrot lunaire Pierre Boulez (conductor)
Jeanne Deroubaix (alto)
Fig. 13.1 Concerts involving Boulez at the 1965 Edinburgh International Festival
years 1946–58 . . . these six works act as a clear guide to the development of
the musical character of the composer – and, incidentally, to recent devel-
opments in writing for the keyboard’.25 All of this shows the unique impor-
tance of this retrospective in enabling listeners for the first time anywhere in
the world to form a global perspective on the totality of Boulez’s output to
date, a fact that was not lost on the critics.
Boulez as Conductor
The retrospective opened with an orchestral concert in which Boulez con-
ducted Beethoven’s Second Symphony, Messiaen’s Oiseaux Exotiques,
Webern’s Symphony op. 21 and Debussy’s La Mer. Neville Cardus had
already described the Hamburg Symphony Orchestra, after their previous
Edinburgh concert under conductor Hans Schmidt-Isserstedt, as ‘a solid,
thoroughly German band of instrumentalists, accustomed to making music
week in, week out, with no more show of fine feathers of virtuosity than
a housewife in Hamburg displays in her daily cooking’.26 His experience
25 26
Bradshaw, programme note for Yvonne Cardus, the Guardian, 28 August 1965,
Loriod’s recital, 31 August 1965. Concert p. 5.
programme, p. 7.
with Boulez at the helm was enough to alter his estimation and he declared
that Boulez’s
performance of Debussy’s “La Mer”, convinced me that he is certainly one of the

most authoritatively musical of living conductors. He combines virtuosity with
modesty. Without a single unnecessary gesture or movement, he controls the
orchestra to a man. The instrumentalists are obviously with him; he need not drive
them. He liberates music, doesn’t force or jockey the tempo, or over-accentuate the
phrases. Under his baton the Hamburg Orchestra was transformed from a good
homespun band of music-makers to a truly masterful and many coloured ensemble,
especially during the marvellously conceived and scored “La Mer”.27
While Cardus was less impressed with Boulez’s account of the Beethoven,
his criticisms of the Messiaen and Webern were more to do with a lack of
appreciation, and Boulez’s performances are described as ‘two expert expo-
sitions of the contents, technical and other’. It is interesting looking back to
a now less common mode of music criticism where the reviewer could
dismiss a work sneeringly, in this case the Messiaen piece, as ‘a composition
for want of a better descriptive term’.
For Martin Cooper, Oiseaux Exotiques ‘introduced that note of excite-
ment and controversy that had been lacking from the first week of the festival
but looks like dominating the second’, and he notes that Messiaen ‘received
an enthusiastic welcome’.28 While Boulez’s performance of the Webern is
complimented as ‘relaxed’, ‘persuasive’ and ‘even endearing’, Cooper, like
Cardus, found Boulez’s Beethoven ‘a characteristically objective interpreta-
tion’ with ‘impeccable orchestral tone and strong rhythmic impulse’ which
unfortunately ‘did not always quite save [it] from a certain facelessness’. Like
most of his colleagues, the critic for The Times judged that while the
Beethoven marked an inauspicious beginning, the Webern ‘has never before
sounded so clear and inevitable, so beautiful in texture and line’, and he had
never before ‘heard so much of the detail in Debussy’s La Mer . . . nor . . . so
much Debussyian atmosphere’.29 For Conrad Wilson, the Webern was
‘performed with exquisite lucidity’, Oiseaux Exotiques ‘proved harder to
swallow’, Beethoven’s Second Symphony ‘had a predictably clear-cut read-
ing, but less personality than one expected’ and La Mer ‘was fascinatingly
realised in an organic performance which . . . revealed some previously
uncharted currents’.30 For the critic from The Glasgow Herald the concert
was fun ‘on a big, exciting scale’ and it ‘brought the festival to a more
27
Cardus, ‘Hamburg Orchestra Concert’, the was first performed there on 10 March 1956
Guardian, 30 August 1965, p. 5. with Yvonne Loriod, its dedicatee, as soloist.
28 29
Cooper, ‘Audience fascinated by Messiaen ‘Conducting Other Men’s Music’,
bird music: Dazzling Counterpoint’, The Times, 30 August 1965, p. 4.
30
The Daily Telegraph, 30 August 1965, p. 9. Wilson, ‘Boulez’s Cornucopia of Sound
Boulez commissioned Oiseaux Exotiques for Effects: Visual as well as aural’, The Scotsman,
the Domaine Musical concerts in Paris and it 30 August 1965, p. 4.
334 Edward Campbell
controversial level’.31 The aural complexity of Oiseaux Exotiques is noted as

is Boulez’s clarity with the Webern and the Debussy, but perhaps more
surprisingly, this critic finds that the Beethoven performance was ‘as fine
Beethoven playing as one could wish to hear’. Only one other journalist,
Claire-Eliane Engel writing for Nouvelles Littéraires, seems to have agreed
that Boulez conducted Beethoven’s Second Symphony with ‘finesse’.32
The British Première of Pli selon pli

There was general unanimity among the critics in acknowledging Boulez’s
qualities as a conductor, but less in the appraisals of the concerts of his own
compositions, with nothing dividing critical opinion more than the UK
première of Pli selon pli which was broadcast live on the BBC Third
Programme. While this was the first performance of the work on British
soil, it had already been the subject of some interest and Barrie Gavin
together with Boulez made a television programme about ‘Improvisation
II sur Mallarmé’ (1966). Indeed, the work has a complex history and the
definitive version was completed only in 1989.
While the unnamed critic from The Times was the most enthusiastic of the
press pack, Neville Cardus was chief naysayer. For The Times critic it was
‘surely the concert of the year’, Boulez was ‘a paragon among interpreters’
and Pli selon pli was nothing short of ‘the most exciting piece of music
composed since – well, at the moment, under the spell of this red-hot
performance, one can only say since The Rite of Spring’.33 It is described as
‘a spectacular experience’. The ‘beautiful warmth and flexibility as well as
purity’ of the voice of soloist Halina Lukomska is praised fulsomely along
with the writing for two electric guitars and a particular moment at the end
of ‘Improvisation III’, described as one of ‘purest loveliness’. This critic
beheld not only a wonderfully sonorous experience but also a ‘spectacle’ as
interesting to the eye as to the ear; and its great length is judged not to have
been a barrier, it being ‘immediately communicative for all its uncompro-
mising language’. He finds in the work a form of physical drama that
‘includes elements of high sophistication as well as of unrestrained barbar-
ity’, this latter characteristic being nowhere more evident than in the closing
‘Tombeau’, one lengthy section of which he admits to ‘find[ing] exhaustingly
noisy’, yet this does not constitute grounds for doubt as to the music’s ‘sense
of form’ or ‘direction’. Again, acknowledging that in total it is ‘a huge
31 33
‘Music from France for Exotic Birds’, ‘Spectacular Concert Experience’,
The Glasgow Herald, 30 August 1965, p. 5. The Times, 30 August 1965, p. 4.
32
Engel, ‘Pierre Boulez Plébiscité’, Nouvelles
Littéraires, 30 September 1965.
unwieldy piece’, it is judged nevertheless to be ‘a masterpiece and to have

heard it complete is a privilege no cultured person will willingly have
forgone’.
Both Gerald Larner and Neville Cardus reviewed the concert for the
Guardian, the former judging that Pli selon pli is not to be treated ‘as if it
were a cantata or song cycle’ since like Mallarmé, Boulez ‘has set out to create
direct physical sensation, and he has succeeded’.34 It is ‘music reduced to its
basic element, sound’ and ‘its effect is immediate rather than formal’. Noting
the unusual range of instruments and their innovative layout with ‘strings
and harps on the left, keyboard instruments and wind on the right, and an
array of mostly indescribable percussion instruments extending across the
length of the platform behind them’, he suggests that
each of these percussion instruments was tuned to a different nervous reaction

rather than to tonality, and there were passages in the work, particularly in the
sensual second ‘improvisation’, when the reactions produced were very real and
disturbing. In the last movement, too, there was a process of frenzied rhythmic
intensification which was inescapably effective.35
For Larner the ‘only criticism’ that some might make is that it could be
considered ‘more applied neurology than it is music’ and he judges that
the work calls the listener to either accept or reject its aesthetic validity on
its own self-positing grounds. He concludes however that the work cannot
be reduced to the status of a laboratory experiment since it is the product
of Boulez’s imagination. The precision of Boulez’s orchestral direction is
yet another factor which further convinces him of the sureness of the
work, while he is at the same time unable to gauge the quality of the
performance definitively since he recalls certain moments ‘when things
seemed not quite right’. The performance of soloist Halina Lukomska
is praised for ‘its free-floating independence’, ‘linear flexibility’ and ‘into-
nation’, certain words emerging at times ‘as sensations’ and the voice
delivering moments of ‘extended line’ and ‘lyrical beauty which was
mostly absent elsewhere’.
Responding to the effusive review in The Times, Cardus, who had
attended the work’s rehearsal, judges that the claim that it ‘is to be
counted second in genius among the musical masterpieces composed
during the present century’ is an overstatement.36 Acknowledging that
he himself ‘could not relate the varied succession of aural phenomena to
music as [his] musical intelligence and senses recognise music’,37 he
34 36
Larner, ‘Boulez’s “Pli selon pli”’, the Cardus, The Spectator, 2 September 1965,
Guardian, 30 August 1965, p. 5. p. 14.
35 37
Ibid. Cardus, ‘Pli selon pli’, the Guardian,
2 September 1965, p. 6.
336 Edward Campbell
seems to endorse Larner’s recommendation that the work be received

as ‘direct physical sensation’, ‘music reduced to its basic element,
sound’. This leads Cardus to question whether Boulez has ‘made
recognisable music’, the starting point for an extended moment of intro-
spection in which he wonders ‘what in the way of natural ability to
respond to and understand music – actual and potential – do those
who find Boulez musically intelligible possess as listeners that is lacking
in my own capacity for musical responses?’ Undoubtedly venting his
frustration, he asks:
Where are the immediately identifying points or places in the score of ‘Pli selon pli,’
when it is made audible, which related to music as we have known and categorised
music these hundreds of years from, say Rameau to Schoenberg? I can discover no
continuous arrangements or succession of vibration, notes, instrumental sound-
data in ‘Pli selon pli’ which take the form of a theme or a melody. There was for me
no particularly perceptible variety of rhythm or of motion.
Willing to accept that Boulez has ‘invented or rather opened up, a new
tonal territory’, and acknowledging that it is ‘full of intriguing, fascinating
evocative noises’, Cardus nevertheless craves ‘a convincing verbal argu-
ment or demonstration revealing to an ordinarily experienced intelligence
exactly where and how “Pli selon pli” takes its place as a masterpiece of
organised music’. He admits to having left the rehearsal ‘a little chilled of
heart’.
A number of critics state that Pli selon pli is Boulez’s most important work,
an evaluation Conrad Wilson picks up from Howard Hartog’s essay in the
souvenir brochure.38 Wilson notes the difficulty of finding ‘a foothold on
which to cling’ given the seeming disparity between music and text, and his
initial judgement is that
it is hard to treat the work as more than a cornucopia of delicate sound effects, whose
impact is often just as much visual as it is aural, what with a whole row of
xylophones, enough flutes for an Orange march, a heaven of harps and the sight of
the batonless Boulez gesturing like a Boy Scout with an efficiency badge for
semaphore.39
Whatever the difficulties with comprehension, Wilson judges that it was

‘remarkably stylishly performed’ and that the audience, which was ‘quite
a large one in the circumstances, applauded with commendable enthusiasm’.
Maintaining his ambivalence, he is at once unsure ‘how many people . . .
would be eager to undergo a second performance’ while suggesting that
‘even if one does not care for the sometimes arid terrain the path traverses, or
for what may seem to be the final destination, one can hardly fail to have
38 39
Wilson, ‘Boulez’s Cornucopia’, p. 4. Ibid.
one’s perceptions aroused’. Pondering the enigma further in a second

article, Wilson acknowledges the difficulty of writing about such events
and the conflicting responsibilities facing the critic who is caught
perplexingly between ‘composer and audience’. Not wanting to be
‘caught napping by posterity’, how is the critic ‘to cope with an intellect
as obviously profound and sincere and yet as baffling as Boulez’s’?40
On this occasion, Wilson seeks the solidarity of fellow critic David
Cairns who has already ‘confessed himself unable to say anything
directly pertinent about the music [of Pli selon pli] – though he
believes, not without considerable misgivings, that it is perhaps
“something real and absolute and even conceivably waiting to be under-
stood”’. Wilson is still chewing over Pli selon pli in a third article
where he reflects on the work’s great length and Boulez’s interest in
non-European notions of time, yet another problematic point of
reception.41 By ‘the end of the first half of “Pli selon pli”, and of the
first movement of the second piano sonata’, he tells us, ‘I felt I had had
about as much as I could digest at my present very incomplete stage
of comprehension of his music’. All of this notwithstanding, the old
ambivalence remains and he makes clear that ‘to have left the hall at
either of these moments would have been a mistake, since both
works . . . gained in impressiveness as they proceeded’.
For T.M. in The Glasgow Herald, Pli selon pli is Boulez’s ‘most impor-
tant work to date’ and its ‘strength . . . seems to be in its most imaginative
fluctuation between numerous nicely controlled textures, defined by
typically exotic instrumentation’.42 This critic is more impressed by the
density of ‘Improvisation III’ and ‘Tombeau’ in comparison with ‘the
thinner and more fragmented textures of the first three movements’,
and ‘their more connected style, help to give the work a dramatic sense
of progress’. While conductor, soloist and orchestra are praised for the
performance, T.M. is unconvinced that serialism in its current state is
sufficient ‘to sustain a composition of this length’.
For Philip Hope-Wallace in the Christian Science Monitor, while the
performance had ‘gone with a bang’, ‘the hall was not overflowing with
listeners, and the reaction seemed largely bewildered or indifferent’.43
Having at first found it ‘the most inaccessible music one could
40 42
Wilson, ‘Boulez Forms Climax: Pianist’s T.M., ‘Truer understanding of serial music:
stunning feat’, The Scotsman, 1 September Dedicated and Inspired Performance by
1965, p. 6. Boulez’, The Glasgow Herald, 30 August
41
Wilson, ‘A Beautiful Example of Boulez’s 1965, p. 5.
43
Art: quartet play with warmth’, Hope-Wallace, ‘“Que Boulez-vous?”:
The Scotsman, 2 September 1965, p. 8. A bang at Edinburgh’, Christian Science
Monitor, 16 September 1965.
338 Edward Campbell
invent – so rhythmless, apparently’, he is irked by Boulez’s sparse and

occasional use of the vast orchestra, and the description of the work as
a portrait of Mallarmé rankled as he struggled to connect poetry and
music. Despite these difficulties he concludes that ‘for its sheer size and
audacity, but also for some other quality which manifests itself in the
fashionable word “reverberation”, the event was truly important’. He is
adamant that this was not ‘a hoax’ and he felt at the end ‘quite sure we were
in the presence of a work which may come for later generations, to stand
where “La Mer” or Stravinsky’s once so bewildering “Le Sacre du
Printemps” stands for us’.
For Peter Heyworth the performance was the ‘high point’ of the week,
and while acknowledging correctly that the work was not in its final form,
it nevertheless represented ‘a summa of [Boulez’s] achievements to date’.44
While he does not ‘understand’ Pli selon pli in any conventional sense, he
admits to enjoying perhaps more ‘superficial’ aspects such as the ‘graceful
vocal melisma of the improvisations’ and the ‘colours and textures’
produced by the huge orchestra. While grasping ‘the force of individual
sentences’, the work’s greater ‘logic’ escapes him, a deficiency which
results at times in a certain monotony that is nevertheless insufficient in
harming the ‘overwhelming impression’ or his fundamental conviction
that it is an ‘important work’.
Desmond Shawe-Taylor agrees that it was ‘the big event of the
week’,45 and he states in retrospect that he found it ‘far more approach-
able, certainly far less monotonous, than the piano sonata . . . mainly
because of the fascinating variety and complexity of the orchestral
apparatus with its gamelang-like [sic] array of percussion instruments;
and also because of the silver thread of pure, precise soprano tone
winding its way through the celestial kitchen’. Despite this, he acknowl-
edges ‘the threat of monotony’ which is only averted ‘by the extreme
delicacy and beauty of vocal and instrumental sound in the third
Improvisation’, but he fears that ‘it returns with a vengeance’ in
‘Tombeau’, ‘when xylophones, bells and gongs set up a loud and
seemingly endless jangle’. After the experience of this and the other
Boulez compositions, he finds that Tippett’s music sounded ‘gloriously
old-fashioned’.
For Peter Stadlen, Pli selon pli is Boulez’s ‘magnum opus’,46 and he
identifies several innovative characteristics of the work including the
44 46
Heyworth, ‘Bewildering Boulez’, Stadlen, ‘Boulez revives melody in his
The Observer, 5 September 1965, p. 24. own way’, The Daily Telegraph, 30 August
45
Shawe-Taylor, ‘Adrift in the 1965, p. 9.
tone-continuum’, The Sunday Times,
reactivation of melody in the form of a new ‘rhapsodic bel canto’, the

‘emancipation of the grace note’ which ‘often helps to counteract . . .
stagnation’, as well as great ‘chordal expanses and eruptions’.
In Stadlen’s view, ‘where all these elements work together notably in
“Une Dentelle s’Abolit” and “A la Nue Accablante Tu” the magic effect
is immediate and needs no effort on the part of the listener’. Furthermore,
‘if all sense of traditional logic has been suspended so is one’s belief in the
need for it’. ‘Tombeau’, he judges, was a little less convincing, with its
‘cataclysmic convulsions . . . suggestive of herds of dinosaurs stirring in
their sleep, they no doubt crushed the trees and undergrowth of the
primeval forest. But they could not shake my conviction that in music
large forms are unthinkable except as the biographies of distinct themes
and motifs.’
For the critic of the New Zealand Herald it was undoubtedly ‘the
most controversial music’ of the festival and the performance ‘made an
impact as shattering as many of its sounds’.47 For this reviewer,
‘whether or not what [Boulez] has written communicates anything
intelligible to the listener depends entirely on the receptivity of the
latter to advanced ideas’, and he judges that ‘most went away from the
concert quite baffled’.
Unlike contemporary performances, Pli selon pli was played in two
parts with a twenty-minute interval which allowed the retuning of the
instruments, that is for the ‘harps to be tuned down a quarter of a tone’.48
The critic from the New Zealand Herald notes that when he tapped his
pipe on a brass ashtray during the interval, ‘a fellow critic cynically
remarked that it sounded as if the performance had resumed. “All the
same”, I said, “it is interesting”. “Yes”, he replied, “and so is Euston station
any morning at 11 a.m.”’. Ultimately, the critic was happy to state that it
‘sounded a sincere essay in a new field’.49 Michael Tippett, another festival
composer that year, was disappointed by the static nature of the piece and
the lack of directional harmony, referring to it as ‘very motionless modern
music’.50 Larner noted in 1974 how Tippett responded to the static
aspects of Pli selon pli, stating that he ‘could never use this kind of thing
for expressive purposes unless it were part of a piece based upon sharp
contrasts’ and that this response stimulated in him the beginnings of
a new symphonic work. On a less serious note, a review article, syndicated
in several German newspapers in the first two weeks of September 1965,
identified the work as ‘Plui selon plui’ [sic].51
47 50
L.C.M.S, ‘Edinburgh welcomes stranger’, Bruce 1975, p. 67.
51
New Zealand Herald, 11 September 1965. ‘Edinburgh: NDR-Sinfonie-Orchester’,
48
Wilson, ‘Boulez’s Cornucopia’, p. 4. Westfälische Nachrichten, Münster,
49
L.C.M.S, ‘Edinburgh welcomes stranger’. 1 September 1965.
340 Edward Campbell
Boulez’s Second Sonata for Piano

Loriod’s solo piano recital at the Freemasons’ Hall on 31 August which
finished with Boulez’s Second Sonata was, for Gerald Larner, ‘Edinburgh’s
most memorable recital this year’ and, noting the ‘warm applause’, he was
persuaded ‘of the value of piano music which at other times had seemed
pianistically unrewarding and formally confusing’.52 Loriod demonstrated
how well Boulez and Messiaen ‘write for the piano’, her performance of
Messiaen’s ‘Reed Warbler’ is described as a ‘revelation’, and she is credited
with having articulated ‘at least the elements of the structure of the Second
Piano Sonata of Boulez without alienating the listener from the textural (not
emotional) argument of the work’. Shawe-Taylor, who acknowledges that he
has only been able ‘at stray moments – and then very likely for the wrong
reason’ to make ‘any consecutive sense’ out of Boulez’s music, identifies the
piano works as ‘the hardest to take, and of them the stiffest is the Second
Piano Sonata’.53 While praising Loriod’s performance, he describes the work
as proceeding ‘in a forbidding series of spits and spurts and eruptive tone-
clusters, like a fierce new bedroom tap in a bad hotel’. For pianist turned
critic Peter Stadlen, Boulez in his Second Sonata ‘still appears to aim at
traditional communication albeit through extremely intractable means, [so]
one feels thwarted by the thought of not understanding what he is saying’.
Furthermore, ‘such first hand information about the work’s structure and
poetics as one may have been favoured with tends to evaporate time and
again in the face of what is actually being heard’.54 While the critic from
The Times describes the sonata as ‘the most elliptical and bewildering music
known to me’, Loriod’s performance ‘though still rather short on dynamic
contrast’ is judged to be ‘absolutely convincing, suggesting a logic behind the
sustained outbursts of frenzied rage’.55
Rather surprisingly, for Conrad Wilson, who retains much of his earlier
ambivalence, the Second Sonata is ‘comparatively more approachable’ since
‘the listener can sense the forward progress and purpose of the music and
can even on one hearing comprehend something of its argument and
structure’.56 Despite this, the listener still has to contend with ‘a first move-
ment of expected density’ and a kind of ‘aural confusion’ it is presumed
relates to the ‘mathematical precision’ of the work’s origins. Consequently,
Wilson again finds himself unable to say more while remaining ‘inclined to
give the composer the benefit of the doubt (for no one, surely, would
52 54
Larner, ‘Edinburgh Concerts’, the Stadlen, ‘Messiaen’s lesson from
Guardian, 1 September 1965, p. 7. bird-song’, The Daily Telegraph, 1 September
53
Shawe-Taylor, ‘Adrift in the 1965, p. 16.
55
tone-continuum’, p. 36. ‘Mozart to Messiaen at Edinburgh’,
The Times, 2 September 1965, p. 6.
56
Wilson, ‘Boulez Forms Climax’, p. 6.
question the intellectual intensity and conviction that seems to lie behind
every note) and to remark hopefully that one will derive more nourishment
from the music next time around’. T.M. in The Glasgow Herald goes even
further, to the extent of stating that ‘the second piano sonata is a strong,
compelling work whose relatively connected language (it even has a clearly
defined scherzo and trio) and subtly balanced contrasts of mood make it one
of the composer’s most immediately accessible compositions’.57 Loriod’s
performance is described as ‘masterly’ and as exhibiting both ‘violence and
brutality’ as well as a ‘poetic quality’ for ‘the more relaxed and subdued
sections in the second movement, and to the haunting conclusion of the
sonata’.
The Livre pour quatuor

While Gerald Larner found the performance of Boulez’s Livre pour
quatuor by the Parrenin String Quartet ‘persuasive’, ‘convincing’ and
‘quite enchanting’, he was disappointed that only two of the work’s eight
sections (Ia and Ib) were heard, a legitimate option given the work’s
aesthetic kinship with Mallarmé’s mobile ‘Livre’.58 His misgivings not-
withstanding, Larner describes the sections heard in Edinburgh as
‘finely constructed Webernesque miniatures with fragmentary and pre-
cisely imagined textures which throw into relief brief but seductive
melodies’, and ‘the thematic relationship between the two movements’
and the ‘wide range of colour’ that was in evidence clearly left him
wanting to hear more.
Despite recognising Boulez as possibly ‘the most accomplished musi-
cian of his generation’, Stadlen, who was hearing the Livre for the first
time, and without the benefit of a score, was unsure of it since ‘the
intelligence that is felt to be at work’ in it ‘seems of a general kind rather
than specifically musical’.59 As we have encountered before, the critic was
unable to make a definitively negative judgement, acknowledging instead
that the work ‘is remarkable enough to sow doubts in the listener’s mind
whether he has not missed something – and what greater benefit is there
to be bestowed upon the student of modern music’.
Comparing the Livre very positively in relation to the Second Sonata, the
critic from The Times judged that it was ‘by far the more attractive, brilliant
57
T.M., ‘Pianist’s Expressive Playing’, booklet notes to ‘Pierre Boulez: Complete
The Glasgow Herald, 1 September 1965, p. 10. Works’, pp. 49–51.
58 59
Larner, ‘Edinburgh Concert’, the Stadlen, ‘Boulez work that sows doubt’,
Guardian, 2 September 1965, p. 7. The work’s The Daily Telegraph, 2 September 1965, p. 18.
complex gestation is summarised in Samuel,
342 Edward Campbell
and colourful in its explosions, widely ranging in its emotional survey, often
delicate and mysterious as the sonata almost never appears to be’.60 Conrad
Wilson felt that its Webernian concentration and brevity was ‘a more
profound experience than the 90-odd minutes of “Pli selon pli”’.61 Struck
by the ‘great conviction and warmth of feeling’ of the performance, he
described it as an ‘approachable and beautiful example of Boulez’s art’.
The mood, of course, was not to last.
The Chamber Works

Larner was less enthusiastic with regard to the chamber music concert on
2 September at the Freemasons’ Hall, with Boulez’s Sonatine and Structures,
Deuxième livre paired with Debussy’s Syrinx for solo flute and the two-piano
En blanc et noir, describing it as ‘perhaps the least rewarding’ of the
concerts.62 One of the main points of difficulty was undoubtedly Structures
and, by way of commentary, Larner summarises Boulez’s remarks on the
work given at a post-concert press conference.
‘chapters’ were written at different periods in his life and represent pages in
a ‘diary’ of his development. So those who detected a change of style in the two
chapters performed (composed in 1956 and 1961 respectively) were right, and it
was encouraging that the later one showed a greater interest in the blandishments
of sound than the forbidding earlier piece.63
Boulez’s disclosure that his model for the piece was the novel Kater
Murr by E. T. A. Hoffmann was helpful to Larner given that ‘the material
was shared between the two pianos in a way similar to the interleaving of
two separate stories in E. T. A. Hoffmann’s novel’. Of further interest is
Larner’s admission that he heard Debussy’s En Blanc et noir, which ended
the recital, in terms of Structures, finding in it an antecedent of the Boulez
piece. It seems nevertheless to have been more an intellectual experience
since Larner states that ‘these were observations from the outside: the
imagination and the emotions of at least one listener were rarely
involved’.64
Peter Branscombe was bewildered by the second book of Structures, and
while
60 63
‘Mozart to Messiaen at Edinburgh’, p. 6. Ibid.
61 64
Wilson, ‘A Beautiful Example’, p. 8. Ibid.
62
Larner, ‘Boulez at the Freemasons’ Hall,
Edinburgh’, the Guardian, 3 September 1965,
p. 9.
the sounds . . . had a fierce fascination all their own . . . the title of the work implies
a formal unity which I could not perceive. For long periods one of the players would
indulge in passionate (and highly pianistic) soliloquy, before waving in the other for
a seemingly unconnected solo or duet passage.65
For The Times critic, the piece ‘compares favourably with the finest
musical invention of Pli selon pli’.66 Stadlen contrasts the unprecedented
limitation of composer choice in the first book with the second, where
‘some of these [powers] are vested in the performers’.67 That Boulez and
Loriod were ‘elegantly signalling to each other [as] they carried on the act
of composition on the platform’ only added to his ‘familiar fears of not
understanding what the composer is saying, the further doubt whether it
is he who is saying it’. Despite this conundrum, it was nevertheless
a ‘supreme hour of audition’ and Stadlen goes so far as to posit
that ‘there are moments when Boulez, sculpting in clay as it were, does
create arrestingly profiled shapes whose sequence, moreover, carries that
semblance of meaning which alone will transform sounds into music’.
For Shawe-Taylor, much of the performance of Structures ‘falls
more agreeably on the ear’ than the Second Sonata, and he praises in
particular the second Chapitre with ‘all manner of delicate overlapping
sonorities and subtleties of keyboard coloration’ as well as its aleatoric
elements.68 While he professes to having begun ‘to have illusions of
comprehension . . . the honest listener’s self-confidence is sapped by the
realisation that in such music he would never notice handfuls of wrong
notes; they would have seemed no different’.
For Conrad Wilson Structures and the Sonatine marked the return
of ‘the tougher, denser, more inscrutable side of Boulez’s musical
personality’.69 Performance aspects of Structures are praised since
Loriod ‘coped throughout . . . with an admirably cool élan, sharing what
seemed a superhuman rapport with Boulez himself at the second piano’.
Despite this, Wilson, while wondering what they ‘are all about’ and
‘whether they can be called music at all’, judges that ‘their impact was
considerable – both literally, because the music is extremely forceful, and
in retrospect, because its uncompromising and unceasing complexity of
expression is something which one feels compelled, at present in
a somewhat masochistic way, to admire and remember’. While T.M. for
The Glasgow Herald acknowledged Structures as ‘ably performed’, they
65 68
Branscombe, ‘Festivals’, p. 874. Shawe-Taylor, ‘Adrift in the
66
‘20th Century Music to the Fore’, tone-continuum’, p. 36.
69
The Times, 6 September 1965, p. 5. Wilson, ‘Obscurer Side of Boulez’,
67
Stadlen, ‘Edinburgh Festival: Composition The Scotsman, 3 September 1965, p. 8.
on Platform: Pianists signal in Boulez’,
The Daily Telegraph, 3 September 1965, p. 18.
344 Edward Campbell
were nevertheless ‘the least compelling examples’ of Boulez’s work on

display.70 Despite having ‘exciting qualities’, exploiting for example
‘some fascinating stereophonic relationships between the two pianos’
and reaching out at times ‘to the very limits of technical possibility’, they
amount to ‘an arid discourse in pure serialism when presented in the one
instrumental colour’.
The most unfavourable response to Structures was unsurprisingly that of
Neville Cardus in a Spectator article titled ‘Edinburgh Music: Bouleversed’,
where he questioned ‘if anybody except the composer himself – and not
always the composer himself – could be sure that all the right wrong notes
were being struck’. Seemingly failing to understand both the meticulous
nature of Boulez’s compositional practice and the relative freedom afforded
in aleatoric works, he suggests that ‘Boulez probably doesn’t insist that all the
notes he has written down on paper, and for the time being decided on,
should be accurately sounded’.71
Reviews of the Sonatine, for a long time Boulez’s earliest acknowledged
work, were less negative, but still mixed. T.M. in The Glasgow Herald
described it as ‘an eminently attractive work with a discernible sense of
progress, and moments of lyricism’, and the performance is praised as
convincing and ‘quite enthralling’.72 David Douglas, writing in the same
newspaper, noted that the recital, which ‘might have been regarded as one
of the more difficult . . . was entirely sold out’.73 Conrad Wilson, in
a review which now seems severely overstated, judges that the Sonatine
‘pushed instrumental technique to its extremity, and about the only thing
Severino Gazzelloni was not required to do with his flute was to snap it in
two’.74 Stadlen’s remark that the performance ‘recalled the innocent days
of Boulez’s 12-note phase’ is much closer to the mark.75 Various critics
report the plight of pianist Margaret Kitchin who, it is claimed, having
rehearsed the work for thirteen hours the day before the performance, was
relaxing at midnight only to be visited by Boulez who had ‘just rewritten
a chunk’ of the score. Without any alternative, she ‘sat up late studying the
new section, and then put in an early appearance at the hall, where she
practiced until the last possible moment’.76 Since the composer made no
further changes to this score after its publication in 1954, it seems likely
that the official explanation was not the whole story.
70 73
T.M., ‘Boulez’s Arid Discourse in Douglas, ‘Recital an Answer to Carping
Serialism’, The Glasgow Herald, 3 September Critics’, The Glasgow Herald, 3 September
1965, p. 10. 1965, p. 10.
71 74
Cardus, ‘Edinburgh Music: Bouleversed’, Wilson, ‘Obscurer Side of Boulez’, p. 8.
75
The Spectator, 16 September 1965, p. 15. Stadlen, ‘Edinburgh Festival: Composition
72
T.M., ‘Boulez’s Arid Discourse’, p. 10. on Platform’, p. 18.
76
Wilson, ‘Maestro of modern serial music’,
The Scotsman, 3 September 1965, p. 8.
Branscombe judges that ‘it is not an easy work to grasp, but Boulez
shows a remarkable ability to gauge the ultimate of which the instruments
are capable’, and the performers are praised as having ‘seemed entirely
at home in this complex and beguiling piece’.77 Larner disagreed
completely,78 estimating that the Sonatine gave ‘the impression of being
unperformable’, and that Gazzelloni had difficulty keeping together with
Kitchin, who ‘played admirably’. Despite acknowledging ‘a few moments
of uncanny blending of flute and piano tone, listening to the performance
was an anxious experience’ and he finishes with a barb, suggesting that,
after the Boulez, Gazzelloni’s beautiful rendering of Debussy’s Syrinx
‘must have fallen on their ears like music’. In complete contrast, the critic
from The Times judges that Gazzelloni and Kitchin ‘gave a formidably
eloquent reading’.79
Le Marteau and Pierrot lunaire

In the final concert, Boulez conducted the New Music Ensemble in
performances of his own Le Marteau sans maître and Schoenberg’s
Pierrot lunaire. For Branscombe, Le Marteau was ‘reasonably familiar
and unfailingly fascinating’ and the performance was ‘lithe’ and ‘rapt’,
in contrast with ‘a poorly balanced and rather perfunctory rendering of
Pierrot lunaire’.80 Remembering that Le Marteau (1953–5) was barely
ten years old, Colin Mason noted in the concert programme booklet the
difficulty of René Char’s texts, the dearth of ‘illuminating or even
articulate commentators’ on the work and the fact that ‘no “rationale”
of Le Marteau has yet been put forward’. He directed attention to
Stravinsky, who stated in 1957:
It will be a considerable time before the value of Le Marteau sans maître is

recognized. Meanwhile I shall not explain my admiration for it but adapt
Gertrude Stein’s answer when asked why she liked Picasso’s paintings: ‘I like to look
at them’ – I like to listen to Boulez.81
Larner’s review displays for the most part the interests he had raised
previously in connection with Pli selon pli. The work’s idiosyncratic instru-
mentation is related to Char’s texts, which ‘if they mean anything mean so
much and contain so many implications that they require a more searching
musical treatment, and apparently a recherche group of instruments to
77 80
Branscombe, ‘Festivals’, pp. 873–4. Branscombe, ‘Festivals’, p. 873.
78 81
Larner, ‘Boulez at the Freemasons’ Hall’, Stravinsky and Craft, Conversations,
p. 9. pp. 127–8; cited in Colin Mason, programme
79
‘20th Century Music to the Fore’, p. 5. booklet, 4 September 1965, p. 2.
346 Edward Campbell
communicate them’.82 On this question of the comprehensibility of the text

and its relation to the music, as before he counsels the abandonment of
analysis of ‘how Boulez interprets the poetry’, and that the listener approach
the piece ‘simply as music, the words joining the texture melodically or even
percussively and adding their own special timbre’. Boulez ‘creates atmo-
spheres’ which produce ‘a unique sensation in music’ and Larner praises the
work’s ‘constant lyrical awareness’.
For Stadlen, who also aligns himself with Stravinsky, ‘the hedonist
element . . . rules supreme’ in Le Marteau,83 and the work, which was ‘always
stimulating if sometimes taxing or baffling’, nevertheless brought the Boulez
retrospective ‘to a beatific close’. Following Boulez, he notes that its
‘sensuous pleasure’ should not be reduced to exotic instrumentation since
the work’s ‘Western context’ is enough to diminish any ‘superficial Far
Eastern associations’. On a more flippant note, he relates the experience of
listening to ‘the more staggering attroupements of notes’ to trying ‘to keep up
with Boulez driving through the crowded streets of Darmstadt’, the former
exhibiting ‘tiny musical impulses [which] follow each other with unprece-
dented rapidity’.84
For Wilson, while Pierrot lunaire sounds ‘quite mild and romantic’ in
comparison with Le Marteau, the latter is ‘by far the most accessible
and, to ears not yet accustomed to his grumpier music, the most
engrossing of the Boulez works performed’ at the festival, thus demon-
strating that ‘the gap is by no means unbridgeable’.85 He praises it as ‘a
wonderfully delicate tissue of sound fascinating not only for its techni-
cal ingenuity but more important for its refined beauty as music’.
The performance was one of ‘such atmosphere and artistry, that one
was held spellbound . . . a concert to be grateful for’ and ‘surely . . . one
of the special memories of Edinburgh 1965’. For T.M., ‘the pointillism,
the exotic scoring . . . the economy, the higher pitch level of Boulez’s
language was more refreshing’, exhibiting ‘a newer, fresher language
which looks beyond central Europe and attempts to find for music
a broader, more universal cultural basis’.86 Praising the performance,
this reviewer recognises the work’s significance in stimulating reassess-
ment of both serialism and the ‘nature of musical expression’, and the
82 85
Larner, ‘Pierre Boulez at Leith Town Hall’, Wilson, ‘Weaving a Finer Fabric’,
the Guardian, 6 September 1965, p. 7. The Scotsman, 6 September 1965, p. 9.
83 86
Stadlen, ‘Exotic charm of Boulez “Le T.M., ‘Emphasising Boulez’s
Marteau”’, The Daily Telegraph, 6 September Individualism’, The Glasgow Herald,
1965, p. 14. 6 September 1965, p. 5.
84
Stadlen, ‘World of Music: Fragments for
our Time’, The Daily Telegraph,
experience is judged to have been ‘as fitting a conclusion to M. Boulez’s

activities at the Festival as one could have wished for’.
The critic from The Times did not share the enthusiasm of his colleagues,
judging that ‘neither performance was as finely judged and executed as one
hoped’.87 Jeanne Deroubaix was ‘standing too far away from the instrumen-
talists’ and ‘her vocalization seemed lightweight in style’. The performance of
Le Marteau ‘needed extra rehearsal’ and even Boulez’s ‘serene, matter of fact
style of direction’ failed to impress in comparison with the more rigorous
style of the ensemble’s habitual conductor John Carewe. Despite this, in
a second Times article, the work is described as containing ‘textures of
greater density but lighter pitch-compass, powerful emotion contained in
a loose strait-jacket’ but also, in the patois of the past, as intimating ‘sugges-
tions of a flute concerto, of a song-cycle, of a thriller (in the recapitulatory
developments of the last movement) – none of them allowed
confirmation’.88 The shortest and arguably most scathing response was
penned by Felix Aprahamian, who dismissed the performance of Le
Marteau curtly as ‘the usual approximation to the printed score’ while
praising ‘one of the most accurate and beautifully lyrical performances of
Schoenberg’s “Pierrot Lunaire”’.89
Post-Festival Summations and Responses

Branscombe noted grudgingly (and mistakenly) that the 1965 festival would
be remembered ‘as the year when Boulez first came to Edinburgh, though it
would be idle to pretend that his works gave the greatest pleasure’.90 Cardus
contended that while future festivals should not ‘overdo’ contributions from
selected composers, he did not ‘see why the “new” music should not be
presented unobtrusively, blended proportionately with familiar stuff’ – for
example ‘the extreme of Boulez and . . . the best of Schubert’.91 Ambivalent
to the last, Wilson judged that ‘none but Boulez’s most devoted disciples
could claim that the performances of his music in Edinburgh . . . were
a consistently rewarding experience – the piano “Structures” are surely
among the most inscrutable and agonising endurance tests in the history
of music this century – yet equally no one could attend these concerts
without feeling that, comprehend him or not, he is a composer whose gifts
are as deep and sincere as they are disturbing’.92
87 90
‘20th Century Music to the Fore’, p. 5. Branscombe, ‘Festivals’, p. 872.
88 91
‘Some Characteristics of Pierre Boulez’s Cardus, ‘Brahms, Britten, Tippett and
Work’, The Times, 10 September 1965, p. 13. Busoni at Edinburgh’, the Guardian,
89
‘Felix Aprahamian in Edinburgh’, 10 September 1965, p. 11.
92
The Sunday Times, 12 September 1965, p. 44. Wilson, ‘Weaving a Finer Fabric’, p. 9.
348 Edward Campbell
For a number of critics, the French visitors, especially Boulez, had

stolen the show. For Hope-Wallace ‘the light of real international festival
glamour has blazed’ with the appearance of Boulez, Messiaen and
Loriod,93 and ‘for Boulez, for the great Mahler Eighth occasion, for
Messiaen and Yvonne Loriod the trip would have been ten thousand times
worthwhile’.94 Describing Pli selon pli as ‘the experience of a lifetime’, he
notes having heard someone growling malevolently at the interval that it
was ‘Lord Harewood’s last defiance’. On a more personal level, he treas-
ures ‘the memory of seeing these three immensely intellectual artists
sitting making their first acquaintance with “sweets from the trolly”,
a British trifle, no less. What moués, what grudging comment: “c’est
mangeable”.’
E.B.S.M. from The Glasgow Herald noted Boulez’s ‘spontaneous
charm which belies all preconceived ideas of what an enfant terrible
of the musical avant garde should be like’.95 He reports that Boulez was
‘very pleased with the coverage given to his music at the Festival’ and
that ‘he was wearing a green and red tartan tie’ which he had bought, he
admitted, in Baden-Baden, not in Edinburgh. Ultimately, ‘however
elusive the appeal of Pierre Boulez’s music, his personal one is
immediate’.
In an evaluation titled ‘Bewildering Boulez’, Peter Heyworth agreed that
the 1965 festival was in some respects Lord Harewood’s finest hour. The fact
of having
brought this fountainhead of the avant-garde to Edinburgh and to have presented

an unprecedented range of his fascinating yet bewildering compositions was an
act of courage and perspicacity of the sort that is all too rare in our musical life
(viz. the enervating mists of the oncoming London concert season).96
For Heyworth, Boulez had demonstrated his credentials as a musician,

the quality of his thinking and the justness of his position within the
avant-garde. While acknowledging the excitement he experienced
repeatedly during the retrospective, he admits also to the limits of his
understanding. Shawe-Taylor also finds that Boulez is ‘respected’ and
‘baffling’ in equal measure,97 and he laments that he has not had time
during the visit ‘to expound his own music; to advise us what to look for,
what kinds of continuity, what types of formal structure, to expect’.
He reports that ‘among the medium-sized audiences that assembled in
93 95
Hope-Wallace, ‘“Que Boulez-vous?”’, E.B.S.M., ‘Festival Profile: Boulez’s
1965. Outlook on Modern Composing’,
94
Hope-Wallace, ‘Festival Postscript’, the The Glasgow Herald, 2 September 1965, p. 10.
96
Guardian, 13 September 1965, p. 7. Heyworth, ‘Bewildering Boulez’, p. 24.
97
Shawe-Taylor, ‘Adrift in the tone-
continuum’, p. 36.
the various halls there was never a bleat of dissent, not a boo, not an
ostentatious or even a furtive exit; only quiet attentiveness, decent
applause at the end, an occasional cheer from the back rows’. He never-
theless senses in this a loss of nerve on the part of audiences and critics,
who out of fear of being proved wrong by history are instead guilty of ‘a
faint whiff of hypocrisy, or at least of unreality’.
Stadlen, most likely drawing on texts by Boulez, Adorno and Lévi-
Strauss, concludes that the difficulty of Boulez’s music is due, at least in
part, to its density and irregularity and the challenges this poses to
perception.98 He suggests that it operates at the level of ‘local minutiae’
and ‘global form’ but that it is difficult to link these levels ‘given the
absence of the developmental middle layers which in traditional music
mediate between the two extreme dimensions’. While the ‘vocal melismas’
of the Mallarmé improvisations and the vocalising of Le Marteau are
‘beguiling’, both Le Marteau and Schoenberg’s Pierrot form ‘an incom-
plete art but perhaps the only possible one these days’.
For The Times, the retrospective was ‘important and beneficial for
everyone interested in music as a vital phenomenon of daily life’, and
Boulez’s ‘loudly applauded excellence as an orchestral conductor has
surely confirmed the ordinary concertgoer’s suspicion that Boulez must
be a real, even if difficult, composer’.99 The festival had succeeded in
establishing firm contact with Boulez even if his music ‘remains hard
going for most of us’. This critic goes on to make the idiosyncratic claim
that Boulez is engaged in forming ‘a valid high romantic language for the
late twentieth century’, as well as the rather bizarre statement that as he
‘develops towards greater freedom, more direct communication . . . he is
approaching a blend of Rossini and Beethoven – cantilena and epic
dynamism’. Beyond the narrowly musical sphere, art critic David Irwin
judged that ‘there was certainly in neither art nor in drama anything
comparable to the musical excitement of Boulez’, and ‘the Festival
authorities . . . showed great enterprise that was successfully ruthless in
the direction of the musical avant-garde’.100
As for the French press, the first of Claude Samuel’s two articles has
the headline ‘Boulez has not been booed in Edinburgh’,101 with the sub-
heading ‘now even his fiercest opponents no longer dare to declare their
hostility too openly’. For Samuel, the festival had produced ‘some decisive
98
Stadlen, ‘World of Music: Fragments for (‘C.R. Mackintosh and the Edinburgh
our Time’, p. 11. Festival’, p. 592).
99 101
‘Some Characteristics’, p. 13. Samuel, ‘Edimbourg n’a pas sifflé Boulez’,
100
Irwin, The Burlington Magazine, October Paris Presse, L’Intransigeant, 10 September
1965, p. 536. Irwin noted in 1968 that the 1965.
Boulez breakthrough was not followed up
350 Edward Campbell
days for the greatest glory of the French musical school’. A long-time
supporter and friend, he describes Boulez as France’s ‘best article of
export, who is red hot in Germany, is fought over in Scandinavia, who
fills the Japanese and the Americans with enthusiasm and who is coming
straight from causing a real sensation in South Africa’. As he states
unequivocally, the real significance of the event lies in the fact that
Edinburgh is the first and only city to have undertaken ‘a complete cycle
dedicated to Boulez as composer, pianist and conductor’ in which an
impressive number of his biggest pieces would be performed allowing
the listener to appreciate the diversity and general thrust of his work. For
Samuel this was particularly important since Boulez had just turned 40
and was in demand around the world as a conductor, his calendar being
fully booked until 1969.
Excluding the views of enthusiasts, Samuel judges that listeners and
journalists turned out ‘in great numbers’, treated Boulez’s works with
‘respectful’ attention and ‘observed an exemplary silence despite the
length of the scores’; there was neither ‘the slightest recrimination nor
the least catcall’. While recognising that the British critics ‘approved’
broadly while retaining notable reservations, he finds sufficient validation
to declare that Boulez is now ‘untouchable’ and that ‘in England [sic] as in
Germany, France or the United States, no “serious” [but] hostile critic
dares affirm his hostility openly’. More constructively, the simultaneous
presence of Messiaen, Loriod and Boulez was ‘the “Domaine musical”
family reunited in Scotland’, indeed ‘the most active, the most virulent, the
most celebrated of the new international music’.
In Samuel’s second article, Boulez, who had to respond to ‘tactless ques-
tions’ posed by British journalists, is described as ‘the shock personality of
the festival’ and, ‘despite certain [frissons] which agitate a public more
curious than informed, the shock was well-received’.102 An unnamed jour-
nalist in the Revue Française notes that ‘the Scottish public . . . welcomed
these very rich and sometimes disconcerting scores with enthusiasm’.103 For
Claire-Eliane Engel in Le Monde, the public ‘welcomed [Boulez’s] incon-
testably difficult works with sympathy; he was ‘the official French musician’
at the festival and the Hamburg orchestra performed Pli selon pli
brilliantly.104 The correspondent for La Libre Belgique105 of Brussels, how-
ever, was much less sympathetic, judging that with Structures Boulez and
Loriod ‘were conspiring against the musical security of the Edinburgh
bourgeosie’, while the audience reaction to Pli selon pli prompted him to
102 104
Samuel, ‘Un Ahurissant Festival’. Engel, ‘Théâtre et Musique à Edimbourg’,
103
‘Le Festival D’Edimbourg’, Revue 17 September 1965.
105
Française, October 1965, pp. 83–4. S.C., in La Libre Belgique, Brussels,
20 September 1965.
recall an unflattering ‘epigram written long ago [by Boileau] at the expense
of Corneille’, namely ‘Après Agésilas, hélas! Après Attila, holà!’. Boulez’s
festival activities were also reported widely in the German Press and, to take
only one example of many, for the Süddeutsche Zeitung Boulez was undoubt-
edly ‘the dominant personality’ at the festival.106
It would seem from the available correspondence that the New Music
Ensemble replaced the ensemble of the Domaine Musical,107 and it was
suggested as early as February 1965 that ‘M. Boulez will probably require
to rehearse with this Emsemble [sic] in London, prior to any rehearsals
which it may be possible to arrange in Edinburgh’. Given Boulez’s rehearsal
schedule in Hamburg immediately before the festival, this was evidently not
possible. After the festival, Boulez thanked the ensemble for its collabora-
tion, adding, ‘I regret that we have not had a little more time beforehand to
reach perfect agreement musically between your group and myself.’108
Despite this, he acknowledges ‘the pleasure I had collaborating with your
group, for the extreme attention and professional capacity I found there’.
The time spent with the Hamburg orchestra unsurprisingly led to a more
successful outcome. The management of the orchestra had written to Boulez
between the end of the rehearsals and the Edinburgh performances, expres-
sing the wish ‘that the concerts with you for the 1965 Edinburgh
International Festival will receive the response they are due’.109 Boulez
replied after the festival, asking them ‘to thank the orchestra for its perfect
collaboration’ in the Edinburgh concerts.110 Finally, Lord Harewood, at the
end of his tenure in Edinburgh, wrote to Boulez:
the week of presentation of your music during this year’s Festival gave me as much
satisfaction as anything for which I have been responsible during my five years in
this job. Listening to it, hearing it being talked about, understanding that the public
had a positive reaction to it – these things made this 1965 Festival really worth while
as far as I was concerned, and if you felt it has given you any satisfaction, I am all the
more pleased.111
He continued:
You worked so hard for us that I feel even you, for once, must have felt like a couple
of days holiday and I should really have written long ago to thank you for your
tremenduous contribution to our programme.
106 109
F. Thorn, ‘Eine Welt in Edinburgh’, Letter from Richard Fehrman, Rudolph
Süddeutsche Zeitung, Munich, 11 September Irmisch and Otto Gerhard to Pierre Boulez,
1965. 24 August 1965 (PSS).
107 110
Letter from Joyce Hinds to Dr G de Koos, Letter from Boulez to Richard Fehrman,
10 February [1965] (PSS). Rudolph Irmisch and Otto Gerhard,
108
Letter from Boulez to Paul Collins, 13 September 1965 (PSS).
111
13 September 1965 (PSS). Letter from Lord Harewood to Pierre
Boulez, 27 September 1965 (PSS).
352 Edward Campbell
Conclusion
The 1965 retrospective serves as a significant moment in the reception of
Boulez’s works. As the first anywhere in the world it allowed reviewers to
respond to the works not only individually but also in relation to one
another and to make judgements regarding the composer’s trajectory.
It also afforded the first live performance of Pli selon pli in the United
Kingdom.
Taken together, the reviews of the concerts present a rich range of
listener responses from unreserved acceptance to outright scepticism.
A number of critics endeavour to make sense of the works against existing
horizons of expectation, with some lamenting their inability to do so in
relation to more traditional exemplars. And there are reviewers who draw
attention to significant gaps in their comprehension of the works, at times
offering fairly idiosyncratic interpretations and suggesting individual and
imaginative strategies for listening and coping with the music’s perceived
strangeness. In this way, they are engaged in ‘constructing hypotheses’, in
‘mak[ing] implicit connections, fill[ing] in gaps, draw[ing] inferences
and test[ing] out hunches’.112 Their ‘pre-understandings’ along with
their ‘dim context[s] of beliefs and expectations’ are made explicit and
modified in the light of their auditory experiences. The encounter with
Boulez’s works clearly questioned their ‘customary codes and expecta-
tions’ and their ‘routine habits of perception’, violating and transgressing
these habitual ways of listening and suggesting new codes.113 A number of
reviewers, finding the conceptual apparatus in play for previous music
largely inapplicable, instead discuss the process of their encounters with
the performances, and while some responses seem to go beyond the kinds
of reactions more commonly experienced, many interesting and valuable,
and some surprising, orientations for future listening emerge in the
process.
The immediate experience of the concerts was not the only factor
shaping the reviews. The force of Boulez’s personality, his brilliance as
a conductor and his standing as a composer within the European avant-
garde clearly made it very difficult for some to take a strong stand against
what they were hearing even when they felt unable to account for their
experiences. It is interesting to note that this was the case as early as 1965
and that this moment in the reception of his work was already being
shaped and influenced by larger guiding forces. Stravinsky’s informal
support for Le Marteau sans maître is a case in point. Consequently, it
112 113
Eagleton, Literary Theory, p. 66. Adapted from Eagleton, ibid., pp. 67–8.
Quotation amended.
is difficult to gauge the sense of significance that some report in relation to

musical performances which by their own admission largely escaped
them.
After Edinburgh, a number of reviewers followed Boulez to London
where on 7 September he performed with the BBC Symphony Orchestra
in his first Proms concert in a programme of Berg, Debussy, Stravinsky
and Webern plus his own Le Soleil des eaux. For The Times critic ‘it was
a brilliant idea to lure him direct from Edinburgh to the Albert Hall . . .
before he deserts the concert platform for the next few months to with-
draw again into his composer’s shell’.114 Following the 1965 retrospective
Boulez returned to the Edinburgh International Festival in 1967, 1968,
1971, 1973, 1975, 1978 and 1984, but after that it was not until the
directorship of Sir Brian McMaster (1992–2006) that he returned with
the regularity of the earlier period, performing in 1994, 1995, 1998, 2000,
2001 and 2004. In contrast, Messiaen and Loriod refused to visit in 1966
for the British première of Messiaen’s Les Batteurs, when ‘the New
Philharmonia Orchestra was forbidden by the Musicians’ Union from
bringing over a French ensemble’.115 Despite such difficulties, the 1965
festival continued to be applauded as Lord Harewood’s ‘most
illustrious year as director’. It is remembered that Boulez and Loriod
‘were given star status’ and that ‘the entire festival was structured around
their presence’, and the claim is repeated that ‘in retrospect, “difficult”
though they sometimes seemed at the time, no events in the festival’s
first 20 years can truly be said to have glittered more brilliantly than
did these’.116
114 116
‘Rich Musical Sympathy by Mr. Boulez’, Obituary of Yvonne Loriod, The Herald,
The Times, 8 September 1965, p. 13. 22 May 2010.
115
Obituary of Yvonne Loriod, The Daily
Telegraph, 18 May 2010.
14 Pierre Boulez and the Suspension of Narrative
Arnold Whittall
Music is an art that has no ‘meaning’; hence the primary importance of

structures that are properly speaking linguistic, given the impossibility of the
musical vocabulary assuming a simply communicative function (Boulez,
Orientations, p. 32).
Boulez’s ability to provoke took many forms in interviews and essays,

from his transparently caustic comment in the 1960s that the German
wartime occupation ‘virtually brought high culture to France’1 to his
suggestion, early in the same decade, that ‘my present mode of thought
derives from my reflections on literature rather than on music’.2 In 1975
Boulez had noted that ‘Proust completely understood how Wagner
worked, never going back but always using the same motifs, the same
basic resources, in order to achieve a continuous development that is both
extremely concise and extremely free’.3 Later, a Proustian reference
underpinned what at face value is one of his most unguarded affirmations
of a classicising ethos: ‘I want to get rid of the idea of compartments in
a work . . . similar to Proust, where you find that the narration is
continuous.’4 Might that mean that Boulez saw his compositions as narra-
tions? If so, what would he have felt about the entry for ‘narrative’ in
The New Shorter Oxford English Dictionary? After word-favouring defini-
tions to do with ‘telling a story’ and asserting teleology by providing ‘an
account of a series of events, facts, etc. given in order and with the
establishment of connections between them’, there is a quotation from
Paul Griffiths’s Concise History of Modern Music: ‘Debussy’s music has
abandoned the narrative mode.’5
Before considering some possible connections between ideas about
narrative and Boulez’s compositions, this essay explores some of the stories
told about those compositions. In one of the earliest mentions of a Boulez
work in a British journal, American composer and Congress for Cultural
1 4
Peyser, Boulez: Composer, Conductor, Di Pietro, Dialogues with Boulez, p. 70.
5
Enigma, p. 25. Griffiths, A Concise History of Modern
2
Boulez, Orientations, p. 143. Music, p. 10.
3
Boulez, Conversations with Célestin Deliège,
pp. 52–3.
354
Freedom adviser Everett Helm wrote about the Darmstadt première of the
ten-year-old Flute Sonatine: ‘it is a fiendishly difficult work – but it makes
excellent sense. Driving, rhythmic passages give way just often enough to
something more relaxed to save it from the monotony that characterizes
much of the post-Webern music.’6 Helm hits on just that contrast
between moto perpetuo toccatas and ‘something more relaxed’ that is
often found in much later Boulez – in Messagesquisse and sur Incises, for
example. Writing shortly after Helm, and with a wider perspective but
a comparable concern for accessibility, David Drew observed that Le
Marteau sans maître marked ‘a notable retreat from the extreme position’
of Polyphonie X and Structures. Drew’s judgement was that Boulez ‘has
arrived at a point of crisis’ which made it inevitable that ‘he will be forced
to simplify his means of expression’. Here Drew seems to touch on the
possibility of what we might now define as Boulez’s retreat from an avant-
garde to a modernist – perhaps even modern-classic – aesthetic. As Drew
asked in 1957: ‘whether he will extricate himself from this crisis by
strengthening his ties with Debussy and early Schoenberg, or whether
he will be prepared to learn more from the clarity and humanity of
Webern remains an open question. In any event, there is already enough
evidence to suggest that Boulez may, in the future, produce work of the
first importance. There is certainly no French composer of today who
shows greater promise.’7
‘Retreat from the extreme’?

More recent commentators with longer perspectives than either Helm or
Drew often bring Boulez into the orbit of the musicological sub-genre
known as ‘Cold War studies’, which concerns itself more with music as
cultural, social practice than as autonomous and essentially self-
referential. Cold War studies also perform the useful scholarly task of
allowing neglected figures their moment in the historical sun – as, for
example, with Leslie A. Sprout’s discussions of the Messiaen student and
Boulez contemporary Serge Nigg.8 The attraction for some musicologists
of playing detective is understandable: it seems so much more interesting
to emerge from dusty archives with evidence to name and shame Nazi
sympathisers or fellow-travellers than to trace ever more sophisticated or
naive variants of serialism or neo-classicism in the music of the time.
6 8
Helm, ‘Darmstadt International Summer Sprout, The Musical Legacy of Wartime
School for New Music’, p. 490. France.
7
Drew, ‘Modern French Music’, p. 310.
356 Arnold Whittall
So far no unexpected skeletons have been found in the cupboard repre-

senting Boulez’s life between his arrival in Paris, aged 18, in 1943, and
the end of the war. Even that reported remark about the benefits of
German wartime occupation seems to have been part of a determined
attempt to wind up the gullible American journalist Joan Peyser, rather
than an unguarded – or bare-faced – expression of Fascist sympathies.
The young Boulez was notoriously abrasive, of course, and never sentimental
enough to allow any admiration he might have felt for the ways in which René
Leibowitz coped with life under German rule in Paris, or respect for his
pioneering spirit in performing and teaching twelve-tone music, to compen-
sate for what Boulez perceived as Leibowitz’s deeply flawed understanding of
Schoenberg’s actual significance. In this respect, you might infer, Boulez saw
little to choose between Leibowitz and Nadia Boulanger.
As Messiaen expressed it according to Peyser, ‘when he [Boulez] first
entered class [in 1944] he was very nice. But soon he became angry with
the whole world. He thought everything was wrong with music.’9 In this
way, the conviction began to emerge that Boulez evolved as a musician at
odds with established compositional techniques and aesthetic criteria:
and so the key to the Boulez narrative was not (modernist) accessibility,
but avant-garde esotericism. Ben Parsons is on strong ground in claiming
that, by the time Boulez came to compose Structures Ia in 1951–2, he
was not simply producing ‘an act of defiance in the face of pressure
to conform both artistically and ideologically’, as Mark Carroll and
others have argued. Boulez was not holding up ‘a mirror to his audience
to reflect their own ravaged world back to them’, as Schoenberg had
done – especially in A Survivor from Warsaw: rather, he ‘was intent on
smashing the codes and hierarchies of their world – the world whose
values so immanently threatened world peace once again – to clear space
for a new and brave vision of what it could be like’. Parsons quotes contem-
porary press comments (especially by Guy Dumur in the journal Combât) to
support his claim that ‘Structures Ia was produced in a city balanced precar-
iously between the liberation and the early Cold War, a Paris that heard
nascent serial music not as the neat disciplinary utopia of pitch-class set
analysis, but as a deeply contemporary and political reaction to these times’.
As Parsons tells it, ‘Structures Ia was heard as a rallying cry not only for
musical revolution but also for socio-political change.’10
9
Peyser, Boulez: Composer, Conductor, of Lutosławski’s ‘equation of moral integrity
Enigma, p. 31. with aesthetic autonomy’, and the argument
10
Parsons, ‘Sets and the City’, p. 63; and that ‘political and artistic detachment’ were
‘Arresting Boulez: Post-War Modernism in the ‘precondition for authentic social
Context’, p. 173. For related discussion of ‘the engagement’, see Jakelski, ‘Witold
complex status of abstraction during the Lutosławski and the Ethics of Abstraction’,
twilight years of the Cold War’ in the context pp. 169–202.
‘Smashing the codes and hierarchies’ of the world according to

Schoenberg and Leibowitz was a profoundly avant-garde enterprise, inher-
ently disrespectful of that ‘German high culture’ that had recently been
visited on Parisians and others during the years of occupation – and which
Boulez himself had imaginatively celebrated in the flute Sonatine and Second
Piano Sonata. After 1952, as he retreated from the ‘new and brave vision’ of
Structures Ia (and also from the expressionistic forcefulness he would have
found in writers like Artaud and Char), at least some of the music that
represented German high culture continued to feature prominently in his
life. He was prepared to spend many hours conducting and recording
Stockhausen, Webern, Berg, Schoenberg, Mahler, Wagner – and even
Bruckner. Yet for Boulez the composer the lure of the Germanic proved
less vital. Instead, he nurtured a Gallic core – a love of Mallarmé, Proust and
Debussy complemented by respect for Sartre, Deleuze and Foucault, among
others. And this helped to prompt the kind of exquisitely labyrinthine
dialogue with aesthetic and technical oppositions and interactions
expounded by Boulez’s more recent critical interpreters.
Narrating Negation
Musically, the Gallicly inflected modernism of Le Marteau and Pli selon pli
might have seemed to offer to the world at large the prospect of a non-
avant-garde modernism that responded to Debussy and Messiaen (as well
as to Webern and Cage) rather as Schoenberg had responded to Brahms
and Wagner: as something that laid usefully mainstream foundations for
future development. However, the immediate impact of the Cold War’s
cultural contexts seemed less salient in Western societies than the
new opportunities emerging from revived economic prosperity and tech-
nological advance; and after the 1960s Boulez the private composer
appeared to develop problems about how to respond to Boulez the more
public figure.
Recent scholarship has cast this process in terms of what Edward
Campbell characterises as ‘a fundamentally negational logic’: Campbell
suggests that ‘while negation was clearly a central element within Boulez’s
approach to composition, at least from 1946, it is only with the lecture
“The need for an aesthetic orientation” in 1963 that he provides
a sustained aesthetic reflection on this aspect of his practice. He scrutinises
the nature of the negation which has been undertaken in post-war music and
he questions whether or not creativity can begin with refusal, or whether
destruction is necessary before reconstruction can begin. He challenges the
success of such destruction and wonders if it has not been naive and
358 Arnold Whittall
presumptuous to build “on ruins or on a tabula rasa”.’11 In this way the

compositional stage is set for discourse and dialogue within
a terminological labyrinth bounded by such common conceptual opposi-
tions as avant-garde and modernist, structuralist and post-structuralist,
modern and post-modern. By the 1990s the availability of at least some of
the more complex results of Boulez’s compositional work at IRCAM
elaborated the terminological mix further – most spectacularly, perhaps,
in Jean-Jacques Nattiez’s aesthetic and technical analysis of Répons in
terms of its possible ‘classicism’.
Boulez worked on Répons from 1980 to 1984: and, in Nattiez’s words, he
gave Répons ‘the opportunity to rise to the status of a classic . . . It is not of
course “classical” in the stylistic sense of the term’: but its rejection of that
equalisation of importance between musical parameters which was
a result of the ‘dissolution of tonal coherence’ and the adoption of ‘total
serialism’ means that – in this sense, at least – Répons is ‘the major classic
work that one could have expected at the end of the twentieth century’.
According to Nattiez, ‘the composer of Répons does indeed seem to have
recovered his public because the work obeys some universal principles
that govern perception’. The different levels at which the music is heard
‘reflect the hierarchical organization of the work’.12 And this return to
hierarchy arguably facilitated another return. As both Campbell and
Jonathan Goldman have shown in their recent writings, that play of
thematicism and athematicism, identity and difference, that is to be
found in the flute Sonatine and the first two piano sonatas, is rediscovered
and refined in such post-Répons compositions as Anthèmes 1 and 2, Incises
and sur Incises.
For those who suspect that Nattiez is riskily unguarded in the way
he talks of hierarchies, Campbell has provided a useful refinement in
offering the alternative formal model of the rhizome: ‘in contrast with
the hierarchically structured branches found within [arborescent] tree
systems’, which have ‘hierarchical modes of communication and pre-
established paths, the rhizome is an a-centred, non-hierarchical, non-
signifying system’. Campbell argues that while Boulez’s work ‘comprises
a heterogeneous assemblage of materials, drawn from a variety of different
milieux . . . There is nothing of eclecticism in this approach, and Boulez
dissociates himself entirely from any heterogeneous synthesis of elements
which would amount to a superficial linking of disparate materials.’
As Campbell suggests, this is a kind of expression in which the possibility
of repetition ‘is no longer subject to identity and sameness, but rather to
11 12
Campbell, Boulez, Music and Philosophy, Nattiez, The Battle of Chronos and
pp. 38–42. Orpheus, p. 280.
difference and variation’: ‘it is a return of the same which is ever

different, and in which each return is a unique manifestation of the
virtual, which is inexhaustible in its possibility, and which has no
primary term’.13 This notion of the virtual, promoting a modernism
whose profound ambiguities go beyond anything classical, is also
a feature of the narratives about Incises and sur Incises by Goldman
and Tom Coult, considered later.
There is a relish for paradox in such thinking, an embrace of ambi-
guity rather than a simple falling into vagueness and imprecision, that
fits well with the character of much music which I would place within
the modernist mainstream. Hierarchic, yet not totally hierarchic (after
the model of Schenkerian diatony); classical, yet not traditionally
classical (and certainly not neo-classical): poststructuralist but not post-
modern. It was a sense of dissolving the ‘vegetal’ rhizome into liquid
form, while nevertheless responding to the siren-like call of Nattiez’s
claims for bestowing the accolade of ‘classic’ on Répons, which lay
behind my 2004 discussion that something worth terming ‘modern
classicism’ was a governing if not all-determining quality of Boulez’s
music as early as Pli selon pli.
One of Boulez’s most striking general formulations comes in the 1968
lecture ‘Where are we now?’ ‘What really interests me is a work that
contains a strong element of ambiguity and therefore permits a number
of different meanings and solutions. Profound ambiguity may be found in
a great classical work, though there it is limited by precise length and basic
structural data . . . On the other hand in today’s music and today’s means
of expression it is possible to investigate this ambiguity, giving the work
multiple meanings that the listener can discover for himself.’14 This might
seem to lead almost too neatly to that later Boulezian affirmation: ‘I need,
or work with, a lot of accidents, but within a structure that has an overall
trajectory – and that, for me, is the definition of what is organic.’15 Or,
as he put it slightly differently earlier on, with an even more overtly
verbal and literary focus on certain ‘classical’ virtues, the task was ‘to
reconstitute from the void all the morphological, syntactic, and rhetorical
qualities needed for an organic discourse to come into being’.16 I have
argued elsewhere that the Mallarméan view of music as ‘the totality of
relationships existing between everything’ might have encouraged Boulez
to promote ‘that modern-classical shift to subordinating disjunction to
combination which is . . . allusively anticipated in “Improvisation III”
13 15
Campbell, Boulez, Music and Philosophy, Di Pietro, Dialogues, p. 25.
16
pp. 143–7. Nattiez, The Battle of Chronos and
14
Boulez, Orientations, p. 462. Orpheus, p. 262.
360 Arnold Whittall
from Pli selon pli, and comes to fruition in . . . Répons, . . . explosante/

fixe . . ., and sur Incises’.17 And by so clearly linking his thoughts on
‘organic discourse’ to literary, verbal, poetic tropes and techniques,
Boulez might be thought to allow for the possibility that his musical
discourse could approach the characteristics of a narrative.
Words, Music, Narrative

Earlier I mentioned The New Shorter Oxford English Dictionary’s use of
Paul Griffiths’s declaration that ‘Debussy’s music has abandoned the
narrative mode’. In its original context, this comment comes within the
argument that Debussy ‘had little time for the thorough, continuous,
symphonic manner of the Austro-German tradition, the “logical” devel-
opment of ideas which gives music the effect of narrative’. In the Prélude à
l’après-midi d’un faune and after, Debussy’s music rejected ‘the coherent
linkage projected by the conscious mind; its evocative images and its
elliptical movements suggest more the sphere of free imagination, of
dream’.18 Griffiths wisely distinguishes between ‘the effect of narrative’
and the real thing, for as Nattiez has pointed out, ‘if music could, in itself,
constitute a narrative as language can constitute a narrative, then music
would speak directly to us, and the distinction between music and
language would disappear’. Citing Adorno’s playful claim that music
‘is a narrative that narrates nothing’, Nattiez nevertheless allows that
‘musical discourse’ not only has ‘semantic possibilities’ – as shown by
works ‘with explicitly literary titles’ – but a ‘syntactic dimension’ invol-
ving ‘techniques of continuity’: ‘musical discourse inscribes itself in time.
It is comprised of repetitions, recollections, preparations, expectations,
and resolutions.’ It follows that ‘music is not a narrative, but an incite-
ment to make a narrative, to comment, to analyze’.19
In classical music, with its organicising essence, it is difficult if not
impossible to avoid all sense of ‘the effect of narrative’: and even when
classicism yields to modernism’s fractures and discontinuities, this does
not necessarily mean that such an ‘effect’ – as both semantics and syntax –
is completely lost. As a composer intensely aware of his Debussian
inheritance, Boulez also prioritised ‘the sphere of free imagination’ over
‘the coherent linkage projected by the conscious mind’ – the kind of
dialogue or interaction described by Goldman as ‘a dialectical play of
17 19
Whittall, ‘Unbounded Visions’, p. 71. Nattiez, Music and Discourse, pp. 127–9.
18
Griffiths, A Concise History of Modern
Music, p. 10.
recognition and surprise’.20 In recent times, Boulez has sometimes

appeared to aspire to a reprioritising of traditional images of narrative
coherence, as in his comment, quoted earlier, about analogies with con-
nectedness in Proust. However, when the prospect of devising an opera
arose, he did not suggest a Proustian theme, rather the kind of absurdist
plot that resists plotting exemplified by Samuel Beckett’s En attendant
Godot.
Just as musical modernism has not proved so rigorously anti-classical as
to jettison all contact with tonal ways of structuring, so musicology has
responded to the post-classical fascination with post-tonal modes of
semantic characterisation and syntactic continuity in ways which explore
the possibility of keeping the effect of narrative in play, along with
hierarchic or rhizomatic modes of formation. As a result, ‘non-narrative’
and ‘anti-narrative’ must remain contested as terms describing certain
kinds of music; another troublesome term is present in the following
statement. ‘One might . . . speak of zero-degree narrativity when
a composer does not intend even ostensibly programmatic music to be
understood narratively: Déserts is one of Varèse’s few pieces with
a programmatic conception, but it is not narrative . . . In rejecting the
narrative approach Varèse said: “There will be no action. There will be no
story. There will be only images. Purely luminous phenomena.”’21
As Varèse saw it, Déserts sought ‘to erase or preclude narrative, whether by
working against a listener’s expectations or by finding ways to discourage
a listener from imposing a narrative reading’. But as Byron Almén and
Robert Hatten painstakingly enquire, ‘can any temporal medium like
music ever be devoid of narrative? Put another way, how might
a composer signal that his or her intention is to be non-narrative, even if
some kind of narrative interpretation is unavoidable?’22 The implication is
that simply saying so, as Varèse does, is not enough. If, however, Déserts can
be talked about in the same way as Arvo Pärt’s Spiegel im Spiegel, the
situation might change: ‘the teleology of Spiegel im Spiegel seems to deny
choice and therefore the possibility of narrative. If the end result is pre-
ordained by the process set up from the start, then there is no development
that is responsive to events, no significant change, no transvaluation. Despite
the illusion of motion and change, the effect is of a global topic –
a kaleidoscope image or a mystical ritual – rather than of a narrative
trajectory.’23
20
Goldman, The Musical Language of Pierre and Other Traditions in American Music,
Boulez, p. 184. p. 304.
21 22
Almén and Hatten, ‘Narrative Engagement Almén and Hatten, ‘Narrative Engagement
with Twentieth-Century Music’, p. 70. with Twentieth-Century Music’, pp. 60–1.
23
E. Varèse, ‘My Titles’, in Broyles, Mavericks Ibid., p. 81.
362 Arnold Whittall
‘Transvaluation’ is a term used by Almén and Hatten to identify

a necessary condition for narrative; a category to which a composition
must be shown to conform if the presence of narrative is to be pro-
posed. And this goes with the claim that a composition might create ‘a
lyrical or trance-like effect that denies teleology and, therefore, narra-
tive’, by following the principle of juxtaposing ‘the narrative mode’ with
‘the lyrical mode . . . which tends towards a spatial rather than
a temporal signification: what is aimed at is a mood, reflection, or
state of being rather than a significant temporal change. Narrative can
also be juxtaposed with a ritual mode, in which the performative
enactment of an established sequence is foregrounded, or with trance,
which attempts the transformation of ordinary consciousness onto
another level, as if to convert the temporal into the eternal. These
other modes of representation signify in their own right, yet they are
not in themselves essentially narrative.’24
Since I will be concerned with music by Boulez that arguably moves
from the trance-like to the teleological, the lyrical to the dramatic, the
constraints of the Almén/Hatten definition of narrative will come into
question, as will its usefulness in relation to any music that might be
termed ‘modernist’ in its aesthetic and technical orientation. Almén and
Hatten resist the argument that the absence, in Stravinsky’s Symphonies of
Wind Instruments, of ‘internal narrative logic’ amounts to a move away
from ‘causal narrative and logical argument to psychological association’.
If, as they inexorably conclude, ‘narrative conceptions of music can be
remarkably flexible and durable’,25 that flexibility might therefore allow
not just for the possibility of presence or absence, but of presence in
absence – as resistance to or suspension of narrative rather than simple
anti- or non-narrative – even if this ‘presence in suspension’ does not take
the form of those ‘transvaluations that guarantee narrative coherence’
according to the Almén/Hatten criteria.
Only Images
Among the myriad analyses of the ways in which French composers treat
interactions between words and music, Carlo Caballero’s comments on
Fauré and Mallarmé provide another possible context for the present narra-
tive, which I made use of in an essay first published in 2004. Fauré never
actually set any Mallarmé, but Caballero justifies bringing them together on
the grounds that Fauré’s ‘attempt to eliminate from his music time, places, or
24 25
Ibid., p. 63. Ibid., p. 82.
preordained images whose precision would allow it to signify manifest his

concern, independent of Mallarmé’s but parallel to it, to abolish the creator’s
elecutionary persona from the finished work’. Fauré’s aesthetic complements
the poet’s recognition that ‘if music seems capable of offering the contem-
plative modes of lyricism without the anecdotal presence of a speaking
author, it does so through a “language” which is not language, a distinctive
musical syntax’. Caballero claims that ‘what is most musical (and Fauréan)
about Mallarmé’s ideal is its preference for words – rather than rhetoric,
narrative, or description – as a vehicle for personal sensibility’, for although
‘Mallarmé’s poetics would not forgo all subjectivity’ they would ‘evade
mimesis, and particularly the elaborate naturalism of his contemporaries’.
Caballero therefore concludes that ‘whereas Fauré had the advantage of
being a musician who wishes to compose music, Mallarmé was a poet who
wished to do much more than write verse. He set himself the task of
rendering music back to its original domain, as it had been on Mount
Parnassus, by subsuming its most abstract expressive abilities in Poetry,
the supreme Music.’26
When, in 1963–4, Boulez declared that ‘in fact my present mode of
thought derives from my reflections on literature rather than music’, it
was Mallarmé who stood for the kind of literature he most valued: and
while Boulez dismissed ‘mimesis’ and ‘elaborate naturalism’ no less
vigorously than Mallarmé himself, he found in the poet’s sonnets
a discipline that opened up the prospect of a modernism whose
purity did as much to reinvent classicism as to reject it, and which
would ultimately lead him away from texts altogether. Retaining some
of Caballero’s terminology, my 2004 essay explored ‘Improvisation III’
from Pli selon pli, suggesting that ‘the absence of that narrative element
that the presence of an “elocutionary persona” might furnish
encourages critical interpretation to shift away from “meaning” and
back to form’.27
A little later, this ‘shift’ prompted the comment that ‘just as Mallarmé’s
“music” – his word-play – compensates for lack of “sense” and draws
coherence out of pervasive correspondences of pure sound, so Boulez’s
music compensates for its rejection of traditional harmonic processes
by emphasizing other kinds of relationships, notably the melodic hetero-
phony and homogeneous timbral interactions which promote flexible
consistency and fragile but sustained coherence, in both structure and
atmosphere’. My 2004 argument culminated with the declaration that
26 27
Caballero, Fauré and French Musical Whittall, ibid., p. 69.
Aesthetics, pp. 123–4, 253–6; Whittall,
‘“Unbounded Visions”’, 66–7.
364 Arnold Whittall
‘Mallarmé’s view of music as “the totality of relationships existing between

everything”’ prompted that ‘modern-classical shift to subordinating dis-
junction to combination which is, I believe, allusively anticipated in the
currently definitive version of “Improvisation III” and comes to fruition in
Boulez’s major works of the IRCAM years, Répons and . . . explosante-
fixe . . . as well as the later sur Incises’.28
Lost Presence
Reference to the ‘absence’ of a ‘narrative element’ in this 2004 text seems
to require the simple binary opposition between absence and presence,
between the setting of a non-narrative Mallarmé poem and the kind of
narrative-tracing verbal text with a sequence of events and actions (even
one without ‘elaborate naturalism’) that is to be found in a work like
Debussy’s opera Pelléas et Mélisande. That 2004 discussion keeps its
distance from the possibility of some intermediate category between
presence and absence such as Lawrence Kramer has since offered in
comments about Debussy’s Jeux. After surveying a range of remarks
about this work, including Boulez’s claim that it ‘marked the arrival of
a kind of musical form, which, renewing itself from moment to moment,
implies a similarly instantaneous mode of perception . . . The general
organization of the work is as changeable instant by instant as it is
homogeneous in development’, Kramer suggests that ‘such a “form” or
“organization” – the terms are vestiges of the very mentality that the
music abandons – entails the withholding or suspension of narrative,
which depends on significant repetition’.29 However, if the presence of
narrative depends entirely on ‘significant repetition’, and the absence of
narrative implies that no such repetitions can be shown to occur, does
‘suspended narrative’ indicate the presence of ‘significant’ subjects but the
absence of ‘significant’ events and actions?
This might well fit the scenario of Jeux. The setting – a tennis court – is
much more naturalistic than abstract, and the evenly balanced musical
dialogue between changeability and homogeneity does not exactly move
the work away from all associations with the presentation of a situation
about which the telling of a story seems perfectly possible. That story is
not told, and its nature is by no means obvious. But it is held in suspense
rather than eliminated altogether. After all, what is to prevent us from
thinking of the tennis-playing quartet as similar to the stressed-out pairs
28 29
Ibid., pp. 70–1. Kramer, ‘Narrative Nostalgia: Modern Art
Music off the Rails’, p. 167.
of lovers in Shakespeare’s A Midsummer Night’s Dream whose story

temporarily ends with a magical intrusion?
‘Changeability’ and ‘homogeneity’ might be mapped onto ‘modernism’
and ‘classicism’ to serve as aesthetic and technical markers for the Boulez
style in ‘Improvisation III’, while calling Mallarmé’s sonnet ‘a siren song’
brings a specific subject and a specific genre into the mix. Is a narrative
about the consequences of encountering a singing siren alluded to in ways
which make it plausible to speak of its suspension, rather than its presence
or absence? Is ‘Improvisation III’ also a siren song? If ‘there is a certain
degree of convergence between Mallarmé’s poetic tissue of phonemic
echoes and analogies and the balanced sequences of resonance in
Boulez’s play with heterophonic and homogeneous textures and
materials’30 then – as the comments quoted earlier about how poem
and composition both ‘compensate’ for comparable kinds of ‘absence’
suggest – the music supports the poem’s suspension of narrative, and
might therefore be felt to provide a metonymic version of that suspension
within its own materials.
It is not just a matter of locating ‘coherence’ in ‘temporal organization’,
so that ‘in Jeux the consistency of time replaces the consistency of narra-
tive’. Here Kramer seems on the verge of moving from suspension to
absence: ‘in its checkered, fluctuating, episodic movement, Jeux produces
what in classic Aristotelian terms would be precisely an anti-narrative
process’. But he then retrieves the suspension trope with the point that ‘as
each potential kernel of narrative disappears to be replaced by another
in this chain of chains, one hears narrativity continually looming and
dissolving away’. Ultimately, ‘the source of both the irony and the plea-
sure [in Debussy’s score] is a narrative that persists in the scenario of Jeux
by appearing there as an object, not a process. This object, however, is not
given but withheld, withdrawn, and the form of its withdrawal is, so to
speak, its retirement from its office. It is not a mere absence but . . . a lost
presence, something that persists while – persists by – remaining out of
reach.’31
Kramer’s persistent playing with words can easily turn counterproduc-
tive when its roots in aesthetic principle and compositional style are lost
sight of. But in any case the thematic and tonal features of Jeux seem to
have no parallels in ‘Improvisation III’, or in Pli selon pli as a whole.
Ironically, perhaps, the degree of convergence between such Debussian
features and Boulez’s style in his later works might have as much if
not more to do with his feeling for the productive tension between
Proustian continuity and Schoenbergian – post-Wagnerian – post-tonal
30 31
Whittall, ‘Unbounded Visions’, p. 70. Kramer, ‘Narrative Nostalgia’, pp. 168–71.
366 Arnold Whittall
methodology. Boulez might never have come to appreciate that

Schoenberg’s remarkably productive (and successful) balancing of mod-
ernism and classicism had the suspension (rather than just the extension,
or the rejection) of tonality at its core in his twelve-tone compositions.
Nevertheless, as analysis of one of his most extended late works, sur
Incises, suggests, his own balancing of contrasting impulses involved the
post-Debussian suspension of narrative – the distancing of a subject in the
absence of text – in ways as resourceful and provocative as any of
Schoenberg’s own suspensive routines.
Portrait de Sacher?
By calling Pli selon pli ‘portrait de Mallarmé’ Boulez left open the possi-
bility of regarding aspects of his music as equivalent to aspects of the
poetry. Rituel, in memoriam Bruno Maderna gives no comparable hints of
reflecting aspects of Maderna’s music – or personality – in Boulez’s work,
and as far as I know no-one has plausibly suggested any such reflections.
Otherwise, however, he kept the names of dedicatees out of his titles,
almost as if it were more important to present a portrait of himself as
provider of a suitably serious gift to a valued friend or colleague. Both the
works Boulez dedicated to Paul Sacher – Messagesquisse and sur Incises –
make much of basic contrasts between arioso, or fantasia, and moto
perpetuo, and both ‘embody’ Sacher by using the hexachord of pitch
and interval classes which Boulez derived from the Sacher surname.
But there is no suggestion that Sacher was alternatively whimsical and
vehemently volatile, fitting this pair of musical personae, and even if
this were so that would not ‘portray’ him much more distinctively
than millions of other similar characters. In 1984 Boulez had shown his
concern to distance the ‘subject’ of a dedication from the ‘subject’ of the
musical processes by basing Dérive, an 80th birthday tribute to William
Glock, on the same Sacher hexachord. On balance, then, it seems that sur
Incises is less a ‘portrait de Sacher’ than an embodiment of a celebratory
ritual whose sonic opulence and volubility parallel Sacher’s energy and
generosity of spirit.
That sur Incises is a ceremony celebrating a long life in the service
of music is perhaps suggested by the spectrum-rich resonances of the
instrumental sound with its embodiment of ‘nineness’ – three pianos,
three harps, and three percussion groups comprising two vibraphones,
marimba, glockenspiel, tubular bells, steel drums, crotales and cylindrical
drums (timbales). The music has a double source – the Sacher hexachord,
and the piano piece Incises whose first version Boulez produced for
a piano competition in 1994. (An expanded version, adding fourteen

pages to the ten-page original, and published in 2002, followed the
completion of sur Incises in 1998.) Apart from centring on ‘Cis’
(C-sharp), which provides the ‘0’ of the Sacher hexachord transposition
used as the music’s principal element, Incises suggests an incising or
cutting-in strategy which, given Boulez’s Proustian hostility to compart-
mentalisation, presumably aims to enhance continuity by increasing the
sense of magnetic attraction between distinct figures or types of gesture.
But rather than speak of anything that suggests a ‘cut and paste’ techni-
que, Boulez once described the most sharply profiled element in Incises as
a ‘gifle’, a slap in the face.32 If this gesture is generalised into the active,
aggressive opposite of more reflective, passive material, then the character
of Incises emerges as ‘a play of recognition and surprise’ – continuity and
discontinuity – that is highly specific in musical terms, much less so (at
least, unless Boulez had supplied a scenario or a suppressed text) in
‘dramatic’ terms. The prospect of creating a ‘proper’ narrative is therefore
held at bay.
Incises/Multiples?
Jonathan Goldman describes sur Incises as ‘one of Boulez’s most contin-
uous, through-composed pieces’,33 as if it might have been meant to
embody that Proustian quality, and also – possibly – to recollect
the kind of ‘classicism’ which, according to Jean-Jacques Nattiez, distin-
guishes Répons. This level of through-composed continuity is, naturally
enough, the result of the relationship spelt out in the title. The first part of
sur Incises elaborates the original piano piece in ways which are
less concerned with enhancing the drama of opposition or ‘incisive’
interaction between Incises’s two basic modes of expression than with
the multiple mirrorings and echoings that result from what Tom Coult
defines as ‘a kind of nine-headed compound instrument’ ensuring that
‘even antiphonal effects are ones of transition rather than opposition,
moving smoothly from one side to another’.34
Prioritising smoothness, Coult writes of the work’s ‘unique unity of
Boulez’s luxurious, lustrous side with the kind of hard-edged rhythmic
precision more typical of his earlier work’. But at the same time he does
not seek to deny the work-spanning role of that ‘polarity’ – deriving from
the piano piece – when ‘the lavish rhythmic incertitude and stasis of the
32 33
Goldman, The Musical Language of Pierre Ibid., p. 185.
34
Boulez, p. 176. Coult, ‘Pierre Boulez’s sur Incises’, p. 5.
368 Arnold Whittall
opening section are contrasted with the toccata’s regularity and rapidity’.
Coult quotes my own suggestion that this polarity ‘can perhaps be
thought of as Boulez’s version of the Nietzschean confrontation between
Dionysus and Apollo’. But Coult would, I imagine, also agree that there is
little aggressiveness or personal animosity evident in sur Incises’s textural
and gestural confrontations: and this is because ‘the “refraction” process –
the splitting up of elements to produce multiple differing forms of that
element – . . . affects the piece’s construction at every stage. Localised
gestures, harmonic fields, rhythmic figures, even the piano’s sound itself
are passed through filters, their make-up analysed and their distillate
viewed from multiple perspectives. It is the typical Boulezian hetero-
phony raised to a higher power – as well as heterophonic textures, we
get harmonies, rhythms and melodies presented as superimposed varia-
tion strata (a heterophony of heterophonies, perhaps).’35
Coult further underlines how Boulez’s concern for the kind of
‘comprehensibility’ that comes from connectedness and continuity
does not simply map itself onto a traditional, Schoenbergian, under-
standing of thematicism. With an echo of Goldman’s explanation
of ‘the virtual theme’, especially in Anthèmes, Coult argues that ‘it is
as if sur Incises refers at every level to something virtual, an idealized,
unseen gesture, phrase or harmony whose derivations exist even
when the original is gone. These absent idea(l)s form the basis of sur
Incises, as a short piano piece is placed in a hall of mirrors to produce
flickering images of sparkling beauty and cogent argument.’36 And
Coult usefully illustrates this quality in his detailed description of
how sur Incises’s first 18 pages (Figures 1 to 14) ‘compose out’ the
first page of Incises.
The piano piece’s initial, single ‘gifle’ is prolonged to provide
a heterophonic segment woven from ‘18 successive statements of the
gesture in its abstract, idealised form (an F preceded by a five-note
coloration that together form a SACHER hexachord). It is as if we are
viewing the object from multiple perspectives, no single one definitive
but together edging closer to the object’s “essence”.’ That ‘essence’
connotes maximum uniformity can be inferred from the fact that
‘when the gesture does appear in its original (Incises) voicing at the
start of the sixth bar (the first time all three pianos play the same thing)
it precipitates a statement of the ten-note downwards figuration’ which
followed the original voicing in Incises’s first bar. Yet this rapid, 32nd
note consequent to the slap-chord’s antecedent has been anticipated in
the ‘coloration’ to the successive Sacher chords provided from b. 1 by
35 36
Ibid., pp. 20–1. Ibid., p. 21.
Marimba 3, an effect that further erodes the ‘pure’ opposition found in

Incises, and strongly suggests that sur Incises involves not simply
prolonging the basic elements of the source but smoothing over the
differences between them.
Coult argues that ‘the fact that the music only proceeds once the
original gesture . . . in its original voicing is heard verbatim illustrates
a key facet of this section of sur Incises, that of “crystallisation”.
Frequently, a gesture from Incises, rather than being stated then devel-
oped, will seem to gradually emerge as a consequence of the music.’37
Coult then narrates how the remaining figures from Incises’s first page are
‘presented from multiple viewpoints, being allowed to emerge organically
and crystallise into the original form’ – what Coult – edging ever closer to
conventional analytical terminology – calls ‘motivic proliferation’. Only
later does he present the more general claim that ‘Boulez’s later music
never seems like a capitulation to regressive tendencies’, by way of ‘a
genuinely dialectical relationship between a modern language and the
structural principles of earlier music, without the simplistic juxtaposition
of objets trouvés or comforting neoclassicism’.38 So is the result
a synthesis – that ‘unique unity’ – that paradoxically permits the persis-
tence of a fundamental polarity within it? Or does talk of ‘unity’ tilt
the scales too strongly in the direction of classicism and away from
modernism? Is it possible that the resonant homogeneity of sur Incises’s
‘nine-headed compound instrument’ drives Boulez to reinforce the
irreconcilability of sober ritual on the one hand and exuberant hedonism
on the other? That the music’s acoustic reality remains productively at
odds with its formal design?
Music in Two Parts

It has become all too easy to characterise Boulez as a composer who,
having turned his back on his own early explorations of Germanic
expressionism, and having become no less disillusioned with both the
systematic and anti-systematic extremes of avant-garde experimentalism,
should have settled for the more accessible modernist middle-ground
excoriated by Heinz-Klaus Metzger in 1962, in a diatribe that might
have been designed to counter the line taken earlier by David Drew:
‘more than a decade ago, Boulez was the only clear representative of
musical progress and the only composer of any relevance’. But after the
abortive Polyphonie X of 1951, which Boulez himself withdrew, ‘the
37 38
Ibid., p. 7. Ibid., p. 19.
370 Arnold Whittall
regression of musical language was all-too apparent’. The fatal step with
Le Marteau, according to Metzger, was to try to reconcile the new music
with what the public would find acceptable, and Pli selon pli was, of all
things, a ‘masterwork’ of a kind that had lost all legitimacy since the early
twentieth-century time of Mahler. As Metzger sardonically concluded,
‘whatever Boulez took from his theoretical knowledge of Debussy is
artfully cashed in here to make a hit, worthy of the avant-garde.
The work [Pli selon pli] stands under the sign of a new suavity, smearing
a kind of sweet glaze over the ears of its listeners . . . The work has a bad
conscience.’39
For most of those qualified to judge today there is usually more than
enough multivalence and subtlety – or Mallarméan glassiness – in the
later Boulez to counter any hint of sonic candyfloss. But the most
important consequence of this ambiguity-enhancing, modern-
classicising tendency is the way it has steered his music away from the
dangerous rocks of that kind of Artaud-inspired expressionistic ferocity
that some of his earliest works acknowledged. In later Boulez there is
energetic exuberance in abundance, in all that moto perpetuo, toccata-
like writing; and this can be very effectively complemented by more
lyrical, poetically resonating materials. On the whole, however, the
Boulezian labyrinth – post-1960 – is not a place of fearful, anxiety-
ridden disorientation: and here I will cite David Metzer’s commentary
on Rituel.
Metzer approaches Rituel by way of the central position of the lament
genre in modernist aesthetic practice, and he uses Ligeti’s Horn Trio for his
first detailed analysis of the expressionistic power that lamenting topoi can
acquire in a post-tonal musical world. With Boulez, he notes the tendency to
write works that embody aspects of remembrance, but as ceremonial cele-
bration rather than grief-stricken sorrowing: and of Rituel Metzer says that
‘lament would seem to be one way to grasp the composition, so practised is it
in loss, memory, and the obedience paid to structure. Yet the work resists
designation as a lament, or any specific genre. Its resistance, though, makes
the lament all the more relevant, for the piece appears to be designed to
prevent a lament from forming.’
Metzer’s commentary is nothing if not multivalent. He claims that ‘Rituel
erects the architecture of what could be an impassioned lament’: as with
Ligeti’s trio, ‘the texture grows denser, the melodic lines sprawl, and energy
builds to a climax’. However, ‘the similarities end there. In Rituel the form
does not animate a rushing emotionality. . . . This is not to say that Rituel is
39
Iddon, New Music at Darmstadt,
pp. 301–2.
an inexpressive work. It has expressive qualities, but they remain difficult to

encapsulate. An austere mournfulness and a rigorous inevitability are two
impressions that come to mind. These qualities are achieved through an
elaborate structural scheme, which is elevated in this piece, lifted up to the
status of a ceremony.’40
This ceremony, Metzer might have added, depends totally on the calm,
controlling design of the master of ceremonies, the conductor out front.
And there might even be a connection here with the kind of thinking that
led Michael Tippett, in the early 1950s, to write that he considered ‘the
general classicizing tendency of our day less as evidence of a new classic
period than as a fresh endeavour . . . to contain and clarify inchoate
material. We must both submit to the overwhelming experience and
clarify it into a magical unity. In the event, sometimes Dionysus wins,
sometimes Apollo.’41 Just as the Tippett authority David Clarke can
suggest that Tippett’s images of the visionary signify ‘not an escape into
a different world, but a challenge to the existing one’,42 Boulez’s occa-
sional hints of the Dionysian after Pli selon pli could signify not an escape
from the Apollonian alternative, but a challenge to it, to be resisted – in
turn – in the true spirit of modern classicism. As with Boulez’s great
friend and colleague Elliott Carter, celebrated in Dérive 2, remembrance
and response to loss required stoicism and the offering of something that
might have a melancholic tinge but does the lost subject of remembrance
the honour of shunning hysterical expressions of grief. Other composers –
Ligeti, Kurtág, even Birtwistle come to mind – are less accepting of
what they might see as a dilution of the strongest human feeling, with
its attendant risks of leaving the late-modernist artwork as something
relatively inhuman, or even subhuman.
Maybe, however, Boulez’s modern classicism is in some ways even more
unsettling within the cultural practice of late modernism than the more
transparent expressionism of Ligeti or Birtwistle. Writing about sur Incises
a decade ago, I suggested that the balance of eloquence and exuberance in
that work might be regarded as a ‘late-century equivalent to Stravinsky’s
1920s “sacrifice to Apollo”’. I then suggested that ‘sur Incises avoids any hint
of pathos, of tragedy, or of the gently sorrowful spirit that steals into the final
vocal passage of Stravinsky’s Les Noces, or the chiming Postlude of Requiem
Canticles’.43 It is also in many ways a complement to Rituel – celebrating
a living rather than deceased colleague – in shunning ‘austere mournfulness’.
While there is nothing that matches the ‘Artaud-inspired expressionistic
40 42
Metzer, Musical Modernism at the Turn of Clarke, The Music and Thought of Michael
the Twenty-First Century, pp. 167–9. Tippett, p. 205.
41 43
Tippett, Tippett on Music, p. 208. Whittall, Exploring Twentieth-Century
Music, p. 197.
372 Arnold Whittall
ferocity’ of Le Visage nuptial, it is not unlike Répons in embracing what Coult

terms ‘moments of high drama’; and nothing is more dramatic in sur Incises
than the response made by Part 2 to the ‘éclat’ of Part 1.
Part 1 (to page 109) is not separated from Part 2, despite the blank page
in the original score. But it is preliminary, preparatory, in setting out the
basic contrast between Rituel-like sobriety and the exuberant toccata
music that then alternate, with the recurrences of the slower material
homing in on C-sharp (‘Cis’) centred chords (those used again at the end
of the revised Incises). In Part 2 the alternations and interactions are
between very fast material deriving from the initial toccata and freer,
more cadenza-like flights that are less ‘ritualised’ than Part 1’s slower,
quieter contrasts. They thereby bring a more dynamic quality to
exchanges that begin to suggest nothing less than a dialogue between
collective, disciplined élan and freer, more individual flamboyance.
The difference between the brittle restlessness of the textures at
Fig. 36 (p. 164) and the motoric force of the resumed Prestissimo at
Fig. 42 (p. 168) generates an unusually lyrical energy that becomes
irresistible by Fig. 52 (p. 184).
The whole point is that both qualities are positive, both necessary to the
life of the work, and the outcome is not the absorption of one by the other
but advance to a final contemplative phase against which the dramatic
action preceding can be measured. Nevertheless, it is time to suspend this
verbal narrative before it falls into the trap of turning the music into
a drama with characters who experience a logically determined sequence
of events. In the end, the suspended narrative of Jeux is a more direct
model for sur Incises than the ecstatically socialised rituals of The Rite of
Spring or Les Noces. Nevertheless, for a composer who claimed to be so
affected by literary, poetic associations, Boulez (after cummings ist der
dichter) remained unusually abstract, perhaps as the most efficient way of
ensuring that his musical narratives embodied those multiple ambiguities
to which his highly personal brand of modernism had such effective
recourse.
Bibliography
Adorno, Theodor W. Quasi Una Fantasia: Essays on Modern Music. New York:
Verso, 1992.
Albèra, Philippe (ed.). Pli selon pli de Pierre Boulez: Entretien et études. Geneva:
Contrechamps, 2003.
Albèra, Philippe. ‘L’instant et la durée: conception de la forme dans la musique de
Pierre Boulez.’ Dissonance 103 (2008), pp. 30–5.
Almén, Byron and Robert J. Hatten. ‘Narrative Engagement with Twentieth-Century
Music: Possibilities and Limits.’ In Michael L. Klein and Nicholas Reyland
(eds.), Music and Narrative since 1900. Bloomington & Indianapolis: Indiana
University Press, 2013, pp. 59–85.
Ashby, Arved. ‘Schoenberg, Boulez, and Twelve-Tone Composition as “Ideal Type”.’
Journal of the American Musicological Society 54/3 (2001), pp. 585–625.
Baron, Carol K. ‘An Analysis of the Pitch Organization in Boulez’s “Sonatine” for
Flute and Piano.’ Current Musicology 20 (1975), pp. 87–95.
Barrault, Jean-Louis. The Theatre of Jean-Louis Barrault. Trans. Joseph Chiari.
London: Barrie and Rockliff, 1961.
Memories for Tomorrow: The Memoirs of Jean-Louis Barrault. Trans. J. Griffin.
London: Thames and Hudson, 1974.
Barthes, Roland. Fragments d’un discours amoureux. Paris: Éditions du Seuil, 1977;
published in English as A Lover’s Discourse: Fragments. Trans. Richard Howard.
New York: Hill and Wang, 1979.
Bassetto, Luisa. ‘Orient-accident? Pli selon pli, ou l’ “eurexcentrisme” selon Boulez.’
In Philippe Albèra (ed.), Pli selon pli de Pierre Boulez: Entretien et études.
Geneva: Contrechamps, 2003, pp. 37–44.
‘Marginalia, ou l’opéra-fantôme de Pierre Boulez.’ In Jean-Louis Leleu and Pascal
Decroupet (eds.), Pierre Boulez: Techniques d’écriture et enjeux esthétiques. Geneva:
Contrechamps, 2006, pp. 255–98.
Bennett, Gerald. ‘The Early Works.’ In William Glock (ed.), Pierre Boulez:
A Symposium. London: Eulenburg, 1986, pp. 41–84.
Bent, Ian. Analysis. Basingstoke: Macmillan Press, 1987.
Bertin, Pierre. Carnet de Voyage. Collection des Cahiers de la Compagnie Madeleine
Renaud-Jean-Louis Barrault. Paris: Julliard, 1954.
Aux Empires du Soleil. 2ème Carnet de Voyage. Paris: Julliard, 1957.
Bessa, Antonio Sergio. ‘Sound as Subject: Augusto de Campos’s Poetamenos.’
In Marjorie Perloff and Craig Dworkin (eds.), The Sound of Poetry/
The Poetry of Sound. University of Chicago Press, 2009, pp. 219–36.
Blaukopf, Kurt (ed.). Technik, Wirtschaft und Ästhetik der Schallplatte: Symposion
auf der “hifi ’68 Düsseldorf”. Karlsruhe: Braun, 1970.
Bonnefoy, Yves. Hier régnant désert. Paris: Mercure de France, 1958; published in
English as Yesterday’s Desert Dominion, in John Naughton and Anthony
373
374 Bibliography
Rudolf (eds.), Yves Bonnefoy, New and Selected Poems. University of Chicago
Press, 1995.
Born, Georgina. Rationalizing Culture: IRCAM, Boulez, and the Institutionalization
of the Musical Avant-Garde. Berkeley: University of California Press, 1995.
Boulez, Pierre. ‘Sonate, que me veux-tu? (Troisième Sonate)’ [1960]. In Points de
repère. Paris: Christian Bourgois, 1981, pp. 163–74.
Penser la musique aujourd’hui. Geneva: Gonthier, 1964; published in English as
Boulez on Music Today. Trans. Susan Bradshaw and Richard Rodney Bennett.
London: Faber and Faber, 1971.
Relevés d’apprenti. Paule Thévenin (ed.). Paris: Éditions du Seuil, 1966; pub-
lished in English as Stocktakings from an Apprenticeship. Trans. Stephen
Walsh, with an introduction by Robert Piencikowski. Oxford: Clarendon
Press, 1991.
‘Musiques traditionnelles – un paradis perdu?’, interview with Martine Cadieu.
The World of Music, Quarterly Journal of the International Music Council
(UNESCO) in association with the International Institute for Comparative
Music Studies and Documentation, IX/2 (1967), pp. 3–10.
Par volonté et par hasard: Entretiens avec Célestin Deliège. Paris: Seuil, 1975;
published in English as Conversations with Célestin Deliège. London:
Eulenburg, 1976.
Orientations: Collected Writings. Jean-Jacques Nattiez (ed.), Martin Cooper (tr.).
London: Faber and Faber, 1986.
‘Le système et l’idée.’ InHarmoniques 1 (Paris: IRCAM, December 1986),
pp. 62–104.
Jalons (pour une décennie). Jean-Jacques Nattiez (ed.). Paris: Christian Bourgois, 1989.
‘Le texte et son pré-texte. Entretien avec Peter Szendy.’ Genesis. Revue internatio-
nale de critique génétique 4 (1993), pp. 135–48.
Points de repère I: Imaginer. New edition, Jean-Jacques Nattiez and Sophie Galaise
(eds.). Paris: Christian Bourgois, 1995.
L’écriture du geste: Entretiens avec Cécile Gilly sur la direction d’orchestre. Paris:
Christian Bourgois, 2002.
Leçons de musique (Points de repère III). Deux décennies d’enseignement au
Collège de France. Jean-Jacques Nattiez (ed.), introductions by Jean-
Jacques Nattiez and Jonathan Goldman, posthumous foreword by
Michel Foucault. Paris: Christian Bourgois, 2005.
Le Marteau sans maître. Facsimilé de l’épure et de la première mise au net de la
partition/Facsimile of the Draft Score and the First Fair Copy of the Full Score.
Pascal Decroupet (ed.). Mainz and New York: Schott, 2005.
Œuvre: Fragment. Essays by Pierre Boulez, Marcella Lista; interview of P. Boulez
with Henri Loyrette. Paris: Gallimard/Musée du Louvre, 2008.
‘Tombeau’, extract from Pli selon pli. Facsimiles of the Draft Score and the First Fair
Copy of the Full Score. Robert Piencikowski (ed.). Vienna, London and
New York: Universal Edition, 2010.
and John Cage. The Boulez–Cage Correspondence. Documents collected, edited
and introduced by Jean-Jacques Nattiez and translated and edited by
Robert Samuels. Cambridge University Press, 1993.
and André Schaeffner. Correspondance 1954–1970. R. Pereira de Tugny (ed.).
Paris: Fayard, 1998.
375 Bibliography
and John Cage. Correspondance et documents. Jean-Jacques Nattiez (ed.), new

edition revised by Robert Piencikowski. Mainz and New York: Schott Musik,
2002.
and Jean-Pierre Changeux, Philippe Manoury. Les neurones enchantés: Le cerveau
et la musique. Paris: O. Jacob, 2014.
Bourdieu, Pierre. ‘La production de la croyance.’ Actes de la Recherche en Sciences
Sociales 13/1 (1977), pp. 3–43.
Bourgeois, Jacques. ‘La révolution de Pierre Boulez a fait long feu à Lamoureux.’
Arts-Lettres-Spectacles (19–25 March 1958), p. 9.
Bradshaw, Susan. Concert programme note for Yvonne Loriod’s recital, 31 August
1965, p. 7.
‘. . . explosante-fixe . . .’, Tempo (September 1973), pp. 58–9.
‘The Instrumental and Vocal Music.’ In William Glock (ed.), Pierre Boulez:
A Symposium. London: Eulenburg, 1986, pp. 127–231.
and Richard Rodney Bennett. ‘In Search of Boulez.’ Music and Musicians 11
(1963), pp. 10–13, 14–18, 50.
Branscombe, Peter. ‘Festivals.’ The Musical Times 106/1473 (1965), 871–5.
Breatnach, Mary. Boulez and Mallarmé: A Study in Poetic Influence. Aldershot:
Scolar Press, 1996.
Bruce, George. Festival in the North: The Story of the Edinburgh Festival. London:
Robert Hale and Company, 1975.
Brunner, Raphaël. ‘Entre style musical et signification musicale: La stylisation.
A partir des quatrièmes pièces de Pli selon pli de Pierre Boulez et de Sept
Haïkaï d’Olivier Messiaen.’ Applied Semiotics 1/1 (1996), pp. 89–117.
‘L’Improvisation III sur Mallarmé de Pierre Boulez: éléments pour une mise en
perspective.’ Dissonance 50 (November 1996), pp. 4–14.
Butor, Michel. ‘Mallarmé selon Boulez.’ In Essais sur les modernes. Paris: Gallimard,
1960, pp. 95–109.
Mobile: Étude pour une représentation des États-Unis. Paris: Gallimard, 1962;
published in English as Mobile: Study for a Representation of the United
States. Trans. Richard Howard. New York: Simon & Schuster, 1963.
Caballero, Carlo. Fauré and French Musical Aesthetics. Cambridge University Press,
2001.
Campbell, Edward. Boulez, Music and Philosophy. Cambridge University Press, 2010.
Chang, Sangtae. ‘Boulez’s Sonatine and the Genesis of his Twelve-Tone Practice.’ Ph.
D. diss., University of North Texas, 1998.
Char, René. Œuvres complètes. Paris: Gallimard, Bibliothèque de la Pléiade, 1983.
La Parole en archipel. Paris: Gallimard, 1962; published in English as The Word as
Archipelago. Trans. Robert Baker. Richmond, CA: Omnidown Publishing,
2012.
Cherlin, Michael. Schoenberg’s Musical Imagination. Cambridge University Press,
2007.
Clarke, David. The Music and Thought of Michael Tippett: Modern Times and
Metaphysics. Cambridge University Press, 2001.
Coult, Tom. ‘Pierre Boulez’s sur Incises: Refraction, Crystallisation and the Absent
Ideal.’ Tempo 264 (April 2013), pp. 2–21.
Cowell, Henry. New Musical Resources. Annotated, with an accompanying essay by
David Nicholls. Cambridge University Press, 1930/1996.
376 Bibliography
Essential Cowell: Selected Writings on Music. Dick Higgins (ed.). New York:
McPherson, 2002.
Curry, Robert, David Gable and Robert Lewis Marshall (eds.). Variations on the
Canon: Essays on Music from Bach to Boulez in Honor of Charles Rosen on His
Eightieth Birthday. Rochester, NY: University of Rochester Press, 2008.
Dahlhaus, Carl. Die Idee der absoluten Musik. Kassel: Bärenreiter, 1978.
Dal Molin, Paolo. ‘Introduction à la famille d’œuvres “. . . explosante-fixe . . .” de
Pierre Boulez. Étude philologique.’ Ph.D. diss., Université de Nice – Sophia
Antipolis. AnnArbor, MI: UMI, © 2008.
‘Mémoriale de Pierre Boulez. Ce que les sources (ne) nous disent (pas).’ Revue de
Musicologie 95/2 (2009), pp. 475–523.
Danuser, Hermann (ed.). Neues Handbuch der Musikwissenschaft. Band 7. Laaber:
Laaber-Verlag, 1978.
de Campos, Augusto. Música de invenção. São Paulo: Editora Perspectiva, 1998.
de Campos, Haroldo. Novas: Selected Writings. Evanston: Northwestern University
Press, 2007.
Decroupet, Pascal. ‘. . . wie die Redaktion zur Systemprämisse wurde . . .’. In Heinz-
Klaus Metzger and Rainer Riehn (eds.), Henri Pousseur. Musik-Konzepte 69.
Munich: Edition Text & Kritik, 1990, pp. 3–14.
‘Comment Boulez pense sa musique au début des années soixante.’
In Philippe Albèra (ed.), Pli selon pli de Pierre Boulez: Entretien et études.
Geneva: Contrechamps, 2003, pp. 49–58.
(ed.). Le Marteau sans maître: Facsimilé de l’épure de la première mise au net de la
partition/Facsimile of the Draft Score and the First Fair Copy of the Full Score.
Mainz and New York: Schott, 2005.
‘Moments doubles, figurés en prisme.’ In Jean-Louis Leleu and Pascal Decroupet
(eds.), Pierre Boulez: Techniques d’écriture et enjeux esthétiques. Geneva:
‘Rhythms–Durations–Rhythmic Cells–Groups.’ In Mark Delaere and
Darla Crispin (eds.), Unfolding Time: Studies in Temporality in Twentieth-
Century Music. Leuven University Press, 2009.
and Jean-Louis Leleu. ‘“Penser sensiblement la musique”: production et description
du matériau harmonique dans le troisième mouvement du Marteau sans maître.’
In Jean-Louis Leleu and Pascal Decroupet (eds.), Pierre Boulez: Techniques
d’écriture et enjeux esthétiques. Geneva: Contrechamps, 2006, pp. 177–215.
Deleuze, Gilles and Félix Guattari. A Thousand Plateaus: Capitalism and Schizophrenia.
Trans. Brian Massumi. Minneapolis: University of Minnesota Press, 1987.
Deliège, Célestin. ‘The Convergence of Two Poetic Systems.’ In William Glock (ed.),
Pierre Boulez: A Symposium. London: Eulenburg, 1986, pp. 99–125.
‘On Form as Actually Experienced.’ In Stephen McAdams and Irène Deliège
(eds.), Music and the Cognitive Sciences. Contemporary Music Review, vol. 4.
New York: Harwood, 1989, pp. 101–15.
Derrida, Jacques. L’écriture et la différance. Paris: Seuil, 1967.
La dissémination. Paris: Seuil, 1972.
Di Pietro, Rocco. Dialogues with Boulez. Lanham and Folkstone: Scarecrow Press,
2001.
Drew, David. ‘Modern French Music.’ In Howard Hartog (ed.), European Music in
the Twentieth Century. Harmondsworth: Penguin Books, 1961, pp. 252–310.
377 Bibliography
Drummond, John. Tainted by Experience: A Life in the Arts. London: Faber and
Faber, 2000.
Dudeque, Norton. Music Theory and Analysis in the Writings of Arnold Schoenberg
(1874–1951). Aldershot: Ashgate, 2005.
Dunsby, Jonathan. Schoenberg: Pierrot Lunaire. Cambridge University Press, 1992.
Eagleton, Terry. Literary Theory: An Introduction, 2nd edn. Oxford: Blackwell
Publishers, 1996.
Edwards, Allen. ‘Unpublished Bouleziana at the Paul Sacher Foundation.’ Tempo
169 (1989), pp. 4–15.
‘Boulez’s Doubles and Figures Doubles Prismes: A Preliminary Study.’ Tempo 185
(1993), pp. 6–17.
‘Éclat/Multiples et le problème de la forme musicale dans les oeuvres sérielles de Pierre
Boulez.’ In Jean-Louis Leleu and Pascal Decroupet (eds.), Pierre Boulez: Techniques
Everist, Mark. ‘Reception Theories, Canonic Discourses, and Musical Value.’
In Nicholas Cook and Mark Everist (eds.), Rethinking Music. Oxford
University Press, 1999, pp. 378–402.
Falcon, Cécile. ‘Théâtres en voyage: les grandes tournées internationales de la
Comédie-Française, du Théâtre national populaire et de la Compagnie
Renaud-Barrault, 1945–1969.’ Ph.D. diss., Université de Montpellier 3, 2011.
www.biu-montpellier.fr/florabium/jsp/nnt.jsp?nnt=2011MON30081, accessed
9 March 2015.
Foucault, Michel. Aesthetics, Method, and Epistemology: Essential Works of Foucault,
1954–1984. James D. Faubion (ed.). New York: New Press, 1998.
Gable, David. ‘Ramifying Connections: An Interview with Pierre Boulez.’
The Journal of Musicology 4/1 (1985), pp. 105–13.
‘Boulez’s Two Cultures: the Post-War European Synthesis and Tradition.’ Journal
of the American Musicological Society 43/3 (1990), pp. 426–56.
‘Words for the Surface: Boulez, Stockhausen, and “Allover” Painting.’
In Robert Curry, David Gable and Robert Lewis Marshall (eds.), Variations
on the Canon: Essays on Music from Bach to Boulez in Honor of Charles Rosen on
His Eightieth Birthday. Rochester, NY: University of Rochester Press, 2008.
Gärtner, Susanne. ‘Komposition als klingende Analyse: René Leibowitz’ Sonate op.
12 in ihrer Beziehung zu Weberns Variationen op. 27.’ Schweizer Jahrbuch für
Musikwissenschaft 19 (1999), pp. 319–35.
‘Pierre Boulez’ “Sonatine für Flöte und Klavier” und ihre neu aufgetauchte
Frühfassung.’ Die Musikforschung 55 (2002), pp. 51–9.
Werkstatt-Spuren: Die Sonatine von Pierre Boulez. Eine Studie zu Lehrzeit und
Frühwerk. Bern: Peter Lang, 2008.
Gelatt, Roland. The Fabulous Phonograph, 1877–1977. New York: Macmillan, 1954/
1977.
Glock, William (ed.). Pierre Boulez: A Symposium. London: Eulenburg, 1986.
Notes in Advance: An Autobiography in Music. Oxford University Press, 1991.
Goehr, Lydia, et al. ‘Philosophy of Music.’ Grove Music Online. Oxford Music Online.
Oxford University Press. www.oxfordmusiconline.com/subscriber/article/
grove/music/52965pg3, accessed 3 March 2014
Goldman, Jonathan. ‘Understanding Pierre Boulez’s Anthèmes [1991]: “Creating
a Labyrinth out of Another Labyrinth”.’ Ph.D. diss., University of Montreal, 2001.
378 Bibliography
The Musical Language of Pierre Boulez: Writings and Compositions. Cambridge

University Press, 2011.
Goléa, Antoine. Rencontres avec Pierre Boulez. Paris: René Julliard, 1958, new edn
1982.
Griffiths, Paul. ‘Two Pianos.’ The Musical Times 114 (1973), p. 390.
Boulez. London: Oxford University Press, 1978.
A Concise History of Modern Music. London: Thames & Hudson, 1978.
Modern Music and After, 3rd edn. New York: Oxford University Press, 2010.
‘Serialism.’ Grove Music Online. Oxford Music Online. Oxford University Press. www
.oxfordmusiconline.com/subscriber/article/grove/music/25459, accessed 3 March
2014.
‘Boulez Discography’, prepared for the now defunct Andante website.
Grimley, Daniel M. Carl Nielsen and the Idea of Modernism. Woodbridge: Boydell
Press, 2010.
Guitry, Sacha. Toutes réflexions faites. Paris: Éditions de l’Élan, 1947; republished in
Cinquante ans d’occupations. Paris: Presses de la Cité, 1993.
Guldbrandsen, Erling E. ‘Serialismen i nytt lys? Det uforutsette hos Boulez.’
In Even Ruud (ed.), Nordisk musikkforskerkongress: Innlegg og referater. Oslo:
University of Oslo, 1992, pp. 276–92.
‘Boulez och Mallarmé, skrift och musikalitet (och Derrida är också med).’ Nutida
Musik 1/37 (1994), pp. 4–13.
Tradisjon og tradisjonsbrudd: En studie i Pierre Boulez’s ‘Pli selon pli – portrait de
Mallarmé’. Ph.D. diss., 1995. Oslo: Scandinavian University Press, 1997.
‘New Light on Pierre Boulez and Postwar Modernism: On the Composition of
“Improvisation I–III sur Mallarmé”.’ In Søren M. Sørensen (ed.), In the Plural:
Institutions, Pluralism, and Critical Self-Awareness in Contemporary Music.
Copenhagen: Department of Musicology, University of Copenhagen, 1997,
pp. 15–28.
‘Modernist Composer and Mahler Conductor: Changing Conceptions of
Performativity in Boulez.’ Studia Musicologica Norvegica 32 (2006), pp. 140–68.
‘Pierre Boulez in Interview, 1996’ (four interviews). Tempo 65/255–258 (2011).
‘Playing with Transformations: Pierre Boulez’s “Improvisation III sur Mallarmé”.’
In E. E. Guldbrandsen and Julian Johnson (eds.), Transformations of Musical
Modernism. Cambridge University Press, 2015.
Haller, Hans Peter. Das Experimentalstudio der Heinrich-Strobel-Stiftung des
Südwestfunks Freiburg 1971–1989. Die Erforschung der Elektronischen
Klangumformung und ihre Geschichte. Baden-Baden: Nomos, 1995, 2 Bde.
Hanson, Robert. ‘Webern’s Chromatic Organisation.’ Music Analysis 2/2 (July 1983),
pp. 135–49.
Häusler, Josef. ‘Gespräch mit Pierre Boulez.’ Donaueschinger Musiktage, vom 19–21
Oktober 1973 (programme notes), pp. 27–31.
Spiegel der Neuen Musik: Donaueschingen. Chronik – Tendenzen – Werkbespre-
chungen. Kassel: Bärenreiter, 1996.
Haworth, David. ‘Edinburgh Festival: What Comes Next?’ Socialist Commentary
(November 1965), pp. 34–5.
Heile, Björn. The Music of Mauricio Kagel. Farnham: Ashgate, 2006.
Helm, Everett. ‘Darmstadt International Summer School for New Music.’
The Musical Times 97/1363 (1956), pp. 490–1.
379 Bibliography
Heyen, Franz-Josef and Friedrich P. Kahlenberg (eds.). Südwestfunk. Vier Jahrzehnte

Rundfunk im Südwesten. Düsseldorf: Droste Verlag, 1986.
Hill, Peter and Nigel Simeone. Messiaen. Yale University Press, 2005.
Hinton, Stephen. ‘The Emancipation of Dissonance: Schoenberg’s Two Practices of
Composition.’ Music and Letters 91/4 (2010), pp. 568–79.
Hirsbrunner, Theo. ‘Pierre Boulez: Notations (1945).’ Melos 48 (1986), pp. 2–20.
‘“Douze Notations” von Pierre Boulez: Zum 65. Geburtstag des Komponisten am
26. März 1990.’ Musica 41 (1990), pp. 78–82.
Hogan, Michael J. The Marshall Plan: America, Britain and the Reconstruction of
Western Europe, 1947–1952. Cambridge University Press, 1987.
Holub, Robert. ‘Reception Theory: School of Constance.’ In Raman Selden (ed.),
The Cambridge History of Literary Criticism, vol. viii, From Formalism to
Poststructuralism. Cambridge University Press, 1995, pp. 319–46.
Iddon, Martin. New Music at Darmstadt: Nono, Stockhausen, Cage, and Boulez.
Cambridge University Press, 2013.
Jacobsen, Wolfgang and Werner Sudendorf (eds.). Metropolis: Ein filmisches
Laboratorium der modernen Architektur. Stuttgart and London: Edition Axel
Menges, 2000.
Jakelski, Lisa. ‘Witold Lutosławski and the Ethics of Abstraction.’ Twentieth-Century
Music 10/2 (September 2013), pp. 169–202.
Jameux, Dominique. Pierre Boulez. Trans. Susan Bradshaw. London: Faber and
Faber, 1991.
Kaltenecker, Martin. ‘Trois perspectives sur l’image sonore.’ In Pierre-
Henry Frangne and Hervé Lacombe (eds.), Musique et enregistrement. Presses
universitaires de Rennes, 2014, pp. 137–58.
Kapp, Reinhard. ‘Shades of the Double’s Original: René Leibowitz’s Dispute with
Boulez.’ Tempo 165 (1988), pp. 2–16.
Katz, Mark. Capturing Sound: How Technology has Changed Music, 2nd edn.
Berkeley, Los Angeles and London: University of California Press, 2010.
Kerman, Joseph. ‘How We Got into Analysis, and How to Get Out.’ Critical Inquiry
(Winter 1980), pp. 311–31.
Musicology. London: Fontana Press, 1985.
Klein, Michael Leslie. Intertextuality in Western Art Music. Bloomington: Indiana
Koblyakov, Lev. ‘Pierre Boulez: “Le marteau sans maître”: analysis of pitch structure.’
Zeitschrift für Musiktheorie 8/1 (1977), pp. 24–39.
Pierre Boulez: A World of Harmony. Chur and New York: Harwood Academic
Publishers, 1990.
Kovács, Inge. Wege zum musikalischen Strukturalismus: René Leibowitz, Pierre Boulez,
John Cage und die Webern-Rezeption in Paris um 1950. Schliengen: Argus, 2004.
Kramer, Lawrence. ‘Narrative Nostalgia: Modern Art Music off the Rails.’
In Michael L. Klein and Nicholas Reyland (eds.), Music and Narrative since
1900. Bloomington: Indiana University Press, 2013, pp. 163–85.
Kristeva, Julia. Desire in Language: A Semiotic Approach to Literature and Art.
New York: Columbia University Press, 1980.
Kurtz, Michael. Stockhausen: eine Biographie. Kassel: Bärenreiter, 1988; published in
English as Stockhausen: A Biography. Trans. R. Toop. London: Faber and Faber,
1992.
380 Bibliography
Lehmann, Hans Ulrich. ‘First performances: Boulez’s Figures Doubles Prismes.’

Tempo 68 (1964), pp. 34–5.
Leibowitz, René. Introduction à la musique de douze sons. Paris: L’Arche, 1949.
Schoenberg and his School: The Contemporary Stage of the Language of Music.
Trans. Dika Newlin. New York: Philosophical Library, 1949.
Le compositeur et son double. Essais sur l’interprétation musicale. Paris: Gallimard,
1971, exp. edn, 1986.
Leleu, Jean-Louis. ‘L’Écriture polyphonique dans le Livre pour quatuor de Pierre
Boulez: aspects techniques et esthétiques.’ In Jean-Louis Leleu and
Pascal Decroupet (eds), Pierre Boulez: Techniques d’écriture et enjeux
esthétiques. Geneva: Contrechamps, 2006, pp. 13–38.
Lenz, Wilhelm de. Beethoven et ses trois styles. Analyses des sonates de piano suivies de
l’essai d’un catalogue critique, chronologique et anecdotique de l’œuvre de
Beethoven. Paris: A. Lavinée, 1855.
Ligeti, György. ‘Entscheidung und Automatik in der Structure Ia.’ Die Reihe 4 (1958),
pp. 38–63; ‘Pierre Boulez: Decision and Automatism in Structure Ia.’ Die Reihe
4 (1960), pp. 36–62.
Losada, Catherine. ‘Isography and Structure in the Music of Boulez.’ Journal of
Mathematics and Music 2/3(2008): 135–55.
‘Complex Multiplication, Structure, and Process: Harmony and Form in Boulez’s
Structures II.’ Music Theory Spectrum 36/1 (2014), pp. 86–120.
Mallarmé, Stéphane. ‘Crise de vers.’ In Œuvres Complètes. Paris: Gallimard, 1945.
Œuvres complètes, vol. ii, nouvelle édition de Bertrand Marchal. Paris: Gallimard,
2003.
Marchal, Bertrand. Lecture de Mallarmé: Poésies – Igitur – Le coup de dés. Paris:
Corti, 1985.
McAdams, Stephen and Irène Deliège (eds.). Music and the Cognitive Sciences.
Contemporary Music Review, vol. 4. New York: Harwood, 1989.
Mellott, George K. ‘A Survey of Contemporary Flute Solo Literature with Analyses of
Representative Compositions.’ Ph.D. diss., State University of Iowa, 1964.
Metzer, David. Musical Modernism at the Turn of the Twenty-First Century.
Cambridge University Press, 2009.
Metzger, Heinz-Klaus and Rainer Riehn (eds.). Musik-Konzepte 89/90 Pierre Boulez.
Munich: Edition Text & Kritik, 1995.
Monod, Jacques-Louis (ed.). René Leibowitz 1913–1972: A Register of his Music and
Writings. New York: Mobart Music Publications, 1983.
Morgan, Robert P. Twentieth-Century Music. New York: W.W. Norton, 1991.
Morris, Christopher. Reading Opera between the Lines: Orchestral Interludes and
Cultural Meaning from Wagner to Berg. Cambridge University Press, 2007.
Mosch, Ulrich. ‘Disziplin und Indisziplin. Zum seriellen Komponieren im 2. Satz des
Marteau sans maître von Pierre Boulez.’ Musiktheorie 5/1 (1990), pp. 39–66.
Musikalisches Hören serieller Musik. Saarbrücken: Pfau Verlag, 2004.
Nattiez, Jean-Jacques. Music and Discourse: Towards a Semiology of Music. Trans.
Carolyn Abbate. Princeton University Press, 1990.
(ed.). The Boulez–Cage Correspondence. Documents collected, edited and intro-
duced by Nattiez and translated and edited by Robert Samuels. Cambridge
381 Bibliography
The Battle of Chronos and Orpheus: Essays in Applied Musical Semiology. Trans.
Jonathan Dunsby. Oxford University Press, 2004.
Nemecek, Robert. Untersuchungen zum frühen Klavierschaffen von Pierre Boulez.
Kassel: Bosse, 1998.
Nyffeler, Max. ‘There Will Always Be Questions Enough: Mauricio Kagel in conversa-
tion with Max Nyffeler.’ www.beckmesser.de/neue_musik/kagel/int-e.html,
accessed 11 March 2015.
Oesch, Hans. ‘Interview mit Pierre Boulez.’ Melos 4 (1976), pp. 293–6.
O’Hagan, Peter. ‘Pierre Boulez: “Sonate, que me veux-tu?”’ An Investigation of the
Manuscript Sources in Relation to the Third Sonata and the Issue of Performer
Choice.’ Ph.D. diss., University of Surrey, 1997.
‘“Antiphonie”: une analyse du processus de composition.’ In Jean-Louis Leleu and
Pascal Decroupet (eds.), Pierre Boulez: Techniques d’écriture et enjeux
esthétiques. Geneva: Contrechamps, 2006, pp. 109–31.
‘Pierre Boulez and the Foundation of IRCAM.’ In Richard Langham Smith
and Caroline Potter (eds.), French Music since Berlioz. Aldershot: Ashgate,
2006, pp. 303–30.
‘Pierre Boulez and the Project of L’Orestie.’ Tempo 61/241 (2007), pp. 34–52.
‘From Sketch to Score: A Facsimile Edition of Boulez’s “Le Marteau sans maître”.’
Music and Letters 88/4 (2007), pp. 632–44.
Pierre Boulez and the Piano. Farnham: Ashgate, 2016.
Parsons, Ben. ‘Sets and the City: Serial Analysis, Parisian Reception, and Pierre
Boulez’s Structures 1a.’ Current Musicology 76 (Fall 2003), pp. 53–79.
‘Arresting Boulez: Post-War Modernism in Context.’ Journal of the Royal Musical
Association 129/1 (2004), pp. 161–76.
Pedneault-Deslauriers, Julie. ‘Pierrot L.’ Journal of the American Musicological
Society 64/3 (2011), pp. 601–45.
Pereira de Tugny, Rosângela. ‘Au commencement était l’esquisse: à propos de
Constellation-Miroir de Pierre Boulez.’ Musurgia 6/1 (1999), pp. 41–61.
Perle, George. The Right Notes: Twenty-Three Selected Essays by George Perle on
Twentieth-Century Music. Stuyvesant, NY: Pendragon Press, 1995.
Peyser, Joan. Boulez: Composer, Conductor, Enigma. New York: Schirmer,
1976.
To Boulez and Beyond. Lanham, MD: Scarecrow Press, 2008.
Piencikowski, Robert. ‘Nature morte avec guitare.’ In Josef Häusler (ed.), Festschrift
Pierre Boulez. Vienna: Universal Edition, 1985, pp. 66–81.
‘“Assez lent, suspendu, comme imprévisible”: Einige Bemerkungen zu den
Vorarbeiten zu Pierre Boulez‚ Eclat.’ In Felix Meyer (ed.), Quellenstudien II:
Zwölf Komponisten des 20. Jahrhunderts. Winterthur: Amadeus, 1993,
pp. 97–116.
‘“Printemps: Sacre: Strawinsky” (1951–52).’ In Felix Meyer (ed.), Settling New
Scores: Music Manuscripts from the Paul Sacher Foundation. A Publication of
the Paul Sacher Foundation. Mainz and New York: Schott Musik International,
1998, pp. 91–3.
‘. . . iacta est.’ In Robert Piencikowski (ed.), Pierre Boulez, John Cage:
Correspondance et documents, nouvelle édition. Mainz: Schott, 2002.
382 Bibliography
‘Inscriptions: Ligeti – Xenakis – Boulez.’ In Jean-Louis Leleu and Pascal Decroupet

(eds.), Pierre Boulez: Techniques d’écriture et enjeux esthétiques. Geneva:
and Michèle Noirjean (eds.). Sammlung Pierre Boulez: Musikmanuskripte
(Inventare der Paul Sacher Stiftung, 3), 2nd revised edition. Mainz: Schott, 2008.
(ed.). ‘Tombeau’, extract from Pli selon pli. Facsimiles of the Draft Score and the
First Fair Copy of the Full Score. Vienna, London and New York: Universal
Edition, 2010.
‘Structure, que me veux-tu?’ In Nicolas Donin and Laurent Feneyrou (eds.),
Théories de la composition musicale au XXe siècle, vol. i. Lyon: Symétrie, 2013,
pp. 867–86.
Rampal, Jean-Pierre. Music, my Love: An Autobiography with Deborah Wise.
New York: Random House, 1989.
Reynolds, Roger, ‘Thoughts on What a Record Records.’ In E. Wiley Hitchcock (ed.),
The Phonograph and Musical Life: Proceedings of a Centennial Conference
7–10 December 1977. Institute for Studies in American Music. ISAM
Monographs No. 14, pp. 28–34.
Richter-Ibáñez, Christina. Mauricio Kagels Buenos Aires (1946–1957):
Kulturpolitik – Künstlernetzwerk – Kompositionen. Bielefeld: Transcript
Verlag, 2014.
Roriche, Mary. ‘Passing through the Screen: Pierre Boulez and Michel Foucault.’
Journal of Literary Studies 22/3–4 (2006), pp. 294–321.
Rosen, Charles. The Romantic Generation. Cambridge, MA: Harvard University
Press, 1995.
Ruwet, Nicholas. ‘Contradictions du langage sériel.’ Revue Belge de Musicologie 13
(1959), pp. 83–97.
Salem, Joseph. ‘Boulez Revised: Compositional Process as Aesthetic Critique in the
Composer’s Formative Works.’ Ph.D. diss., Yale University, 2014.
Samuel, Claude. Accompanying booklet to the CD box set ‘Pierre Boulez: Complete
Works.’ Deutsche Grammophon: 4806828, 2013.
Schaufler, Wolfgang. ‘Pierre Boulez Talks about His Music.’ [video file] Vienna:
Universal Edition, 2012. Accessed 17 March 2015.
Scherer, Jacques. Le Livre de Mallarmé. Paris: Gallimard, 1957.
Schmidt, Felix and Jürgen Hohmeyer. ‘Sprengt die Opernhäuser in die Luft!’
Interview with Pierre Boulez, Der Spiegel 40 (September 1967), pp. 166–7;
English translation (incomplete): ‘Opera Houses? – Blow them up!’ World
Theatre XVII/3–4 (1968 ), pp. 199–207.
Schneider, Marcel. ‘La vita musicale all’estero: Francia.’ Musica d’oggi 2 (1959),
pp. 172–4.
Schnittke, Alfred. A Schnittke Reader. Bloomington: Indiana University Press, 2002.
Schoenberg, Arnold. Theory of Harmony. Berkeley: University of California Press, 1983.
Style and Idea: Selected Writings of Arnold Schoenberg. Trans. Leonard Stein;
Leo Black (ed.) and Joseph Auner (foreword). Berkeley: University of
California Press, 2010.
Schürmann, Reiner. ‘Situating René Char: Hölderlin, Heidegger, Char and the
“There Is”.’ Boundary 2/4/2 (1976), pp. 513–34.
Schwarz, David. Listening Awry: Music and Alterity in German Culture. Minneapolis:
University of Minnesota Press, 2006.
383 Bibliography
Shreffler, Anne. Webern and the Lyric Impulse: Songs and Fragments on Poems of
Georg Trakl. New York: Oxford University Press, 1994.
Simms, Bryan. The Atonal Music of Arnold Schoenberg, 1908–1923. New York:
Oxford University Press, 2000.
Solomon, Maynard. ‘Beethoven’s Ninth Symphony: The Sense of an Ending.’ Critical
Inquiry 17/2 (Winter 1991), pp. 289–305.
Souris, André. La lyre à double tranchant: Écrits sur la musique et le surréalisme.
Présentés et commentés par Robert Wangermée. Liège: Mardaga, 2000.
Sprout, Leslie A. ‘The 1945 Stravinsky Debates: Nigg, Messiaen, and the Early Cold
War in France.’ The Journal of Musicology 26/1 (2009), pp. 85–131.
The Musical Legacy of Wartime France. Berkeley, Los Angeles and London:
University of California Press, 2013.
Stacey, Peter F. Boulez and the Modern Concept. Lincoln: University of Nebraska
Press, 1987.
Steinegger, Catherine. ‘Pierre Boulez et Paul Claudel.’ Bulletin de la Société Paul
Claudel 202 (2011), pp. 4–14.
Pierre Boulez et le théâtre: De la Compagnie Renaud-Barrault à Patrice Chéreau.
Wavre: Mardaga, 2012.
Sterne, Colin C. ‘Pythagoras and Pierrot: An Approach to Schoenberg’s Use of
Numerology in the Construction of “Pierrot Lunaire”.’ Perspectives of New
Music 21/1/2 (1982), pp. 506–34.
Stockhausen, Karlheinz. ‘Musique dans l’espace.’ Revue belge de musicologie 13/1–4
(1959), pp. 76–82.
‘. . . wie die Zeit vergeht . . .’ Die Reihe, Herbert Eimert and Karlheinz Stockhausen
(eds.), 3 (1957), pp. 13–42. Trans. Cornelius Cardew as ‘. . . How Time
Passes . . .’. Die Reihe 3 (1959), pp. 10–40.
‘Music and Space.’ Die Reihe 5 (1961), pp. 67–82.
Karlheinz Stockhausen bei den ‘Internationalen Ferienkursen für Neue Musik’ in
Darmstadt 1951–1996. Kürten: Stockhausen-Verlag, 2001.
Stolla, Joachim. Abbild und Autonomie. Zur Klangbildgestaltung bei Aufnahmen
klassischer Musik 1950–1994. Marburg: Tectum, 2004.
Stowell, Robin. The Cambridge Companion to the String Quartet. Cambridge
Straus, Joseph Nathan. Extraordinary Measures: Disability in Music. New York:
Oxford University Press, 2011.
Stravinsky, Igor. An Autobiography. New York: Norton, 1962.
and Robert Craft. Conversations with Igor Stravinsky. London: Faber and Faber, 1959.
and Robert Craft. Dialogues and a Diary. London: Faber and Faber, 1968.
Strinz, Werner. Variations sur l’inquiétude rythmique. Frankfurt am Main: Peter
Lang, 2003.
‘Quelques observations sur des “objets retrouvés” dans l’oeuvre de Pierre Boulez.’
In Jean-Louis Leleu and Pascal Decroupet (eds.), Pierre Boulez: Techniques
Sutcliffe, W. Dean. ‘Haydn, Mozart and their Contemporaries.’ In Robin Stowell
(ed.), The Cambridge Companion to the String Quartet. Cambridge University
Press, 2003.
Taruskin, Richard. Music in the Late Twentieth Century: The Oxford History of
Western Music, vol. v. Oxford University Press, 2010.
384 Bibliography
Tippett, Michael. Tippett on Music. Meirion Bowen (ed.). Oxford University Press,
1995.
Tissier, Brice. ‘Mutations esthétiques, mais continuité technique dans l’oeuvre de
Pierre Boulez.’ Ph.D. diss., Université de Montréal and Université de Paris-
Sorbonne, 2012.
‘Variations-Rondeau de Pierre Boulez: chronique d’une esquisse de la Deuxième
Sonate.’ Musurgia 19/2 (2012), pp. 33–51.
Trubert, Jean-François. ‘Les “franges aux limites indécises”: Mauricio Kagel et le
théâtre instrumental.’ In Loxias 20 Les paratextes: approches critiques. http://
revel.unice.fr/loxias/?id=2135, accessed 11 March 2015.
Valiquet, Patrick. ‘The Spatialisation of Stereophony: Taking Positions in Post-War
Electroacoustic Music.’ International Review of the Aesthetics and Sociology of
Music 43/2 (2012), pp. 403–21.
Varèse, Edgard. ‘My Titles.’ In Michael Broyles, Mavericks and Other Traditions in
American Music. New Haven: Yale University Press, 2004.
Vermeil, Jean. Conversations with Boulez: Thoughts on Conducting. Trans.
Camille Naish. Portland, OR: Amadeus Press, 1996.
Verrier, Mickael. ‘Un fragment chez Boulez.’ Mitteilungen der Paul Sacher Stiftung
21 (2008), pp. 38–43.
Wangermée, Robert. André Souris et le complexe d’Orphée: Entre surréalisme et
musique sérielle. Liège: Mardaga, 1995.
Wentzel, Wayne C. ‘Dynamic and Attack Associations in Boulez’s Le Marteau sans
maître.’ Perspectives of New Music 29/1 (1991), pp. 142–70.
Whittall, Arnold. Exploring Twentieth-Century Music. Cambridge University Press,
2003.
‘“Unbounded Visions”: Boulez, Mallarmé and Modern Classicism.’ Twentieth-
Century Music I/1 (March 2004), pp. 65–80.
Winick, Steven D. ‘Symmetry and Pitch-Duration Associations in Boulez’s Le
Marteau sans maître.’ Perspectives of New Music 24/2 (1986), pp. 280–321.
Wyschnegradsky, Ivan. ‘L’ultrachromatisme et les espaces non-octaviants.’ La Revue
Musicale 290–1 (1972), pp. 71–141.
Libération du son: Écrits 1916–1979. Textes réunis, présentés et annotés par
Pascale Criton. Lyon: Symétrie, 2013.
Xenakis, Iannis. ‘La crise de la musique sérielle.’ Gravesaner Blätter 1 (July 1955),
pp. 2–4.
Newspapers, Magazines and Programme Notes

Anon. ‘Music at the Festival: Testing even the Avant-Garde.’ Glasgow Herald,
21 August 1965, p. 6.
Edinburgh Festival Programme, programme note, 29 August 1965.
‘Conducting Other Men’s Music.’ Times, 30 August 1965, p. 4.
‘Music from France for Exotic Birds.’ Glasgow Herald, 30 August 1965, p. 5.
‘Spectacular Concert Experience.’ Times, 30 August 1965, p. 4.
‘Mozart to Messiaen at Edinburgh.’ Times, 2 September 1965, p. 6.
‘20th Century Music to the Fore.’ Times, 6 September 1965, p. 5.
‘Rich Musical Sympathy by Mr. Boulez.’ Times, 8 September 1965, p. 13.
385 Bibliography
‘Some Characteristics of Pierre Boulez’s Work.’ Times, 10 September 1965, p. 13.

‘Le Festival D’Edimbourg.’ Revue Française, Octobre 1965, pp. 83–4.
Obituary of Yvonne Loriod. Daily Telegraph, 18 May 2010.
Obituary of Yvonne Loriod. Herald, 22 May 2010.
‘Edinburgh: NDR-Sinfonie-Orchester.’ Westfälische Nachrichten, Münster,
1 September 1965.
Aprahamian, Felix. ‘Felix Aprahamian in Edinburgh.’ Sunday Times, 12 September
1965, p. 44.
Baby, Yvonne. ‘La naissance à soi-même doit s’accomplir chaque matin.’ Interview
with Pierre Boulez, Le Monde, 23 December 1976, p. 11.
Bradshaw, Susan. Edinburgh International Festival, programme note, 31 August
1965.
Cardus, Neville. Guardian review, 28 August 1965, p. 5.
‘Hamburg Orchestra Concert.’ Guardian, 30 August 1965, p. 5.
‘Pli selon pli.’ Guardian, 2 September 1965, p. 6.
The Spectator, 2 September 1965, p. 14.
‘Brahms, Britten, Tippett and Busoni at Edinburgh.’ Guardian, 10 September
1965, p. 11.
‘Edinburgh Music: Bouleversed.’ The Spectator, 16 September 1965, p. 15.
Cooper, Martin. ‘Audience fascinated by Messiaen bird music: Dazzling
Counterpoint.’ Daily Telegraph, 30 August 1965, p. 9.
Douglas, D. ‘Recital an Answer to Carping Critics.’ Glasgow Herald, 3 September
1965, p. 10.
Dumesnil, René. ‘Pierre Boulez – Peter Frankl – Georges Prêtre aux concerts
Lamoureux.’ Le Monde, 18 March 1958, p. 13.
E.B.S.M. ‘Festival Profile: Boulez’s Outlook on Modern Composing.’ Glasgow
Herald, 2 September 1965, p. 10.
Edinburgh Evening News, 14 August 1965.
Engel, Claire-Eliane. ‘Théâtre et Musique à Edimbourg.’ Le Monde, 17 September 1965.
‘Pierre Boulez Plébiscité.’ Nouvelles Littéraires, 30 September 1965.
Fife Free Press, 9 January 1965.
Gavoty, Bernard (alias ‘Clarendon’). ‘Pierre Boulez ou La polka des chaises.’ Le
Figaro, 18 March 1958, p. 18.
Hamon, Jean. ‘Propos sur Doubles de Pierre Boulez.’ Combat, 20 March 1958, p. 2.
‘Toujours à propos de Doubles de Pierre Boulez.’ Combat, 24 March 1958, p. 2.
Hartmann, P. ‘Edinburgh: NDR-Sinfonie-Orchester.’ Flensburger Tageblatt,
2 September 1965.
Heyworth, Peter. ‘Bewildering Boulez.’ Observer, 5 September 1965, p. 24.
Hope-Wallace, Phillip. ‘Festival Postscript.’ Guardian, 13 September 1965, p. 7.
‘“Que Boulez-vous?”: A bang at Edinburgh.’ Christian Science Monitor,
16 September 1965.
Inverness Courier, 29 January 1965.
Irwin, David. ‘Edinburgh Festival.’ Burlington Magazine 106/739, October 1964,
pp. 474–5, 477.
‘Edinburgh Festival.’ Burlington Magazine 107/751, October 1965, pp. 536,
538–41.
‘C.R. Mackintosh and the Edinburgh Festival.’ Burlington Magazine 110/787,
October 1968, pp. 588–92.
386 Bibliography
Larner, Gerald. ‘Boulez’s “Pli selon pli”.’ Guardian, 30 August 1965, p. 5.

‘Edinburgh Concerts.’ Guardian, 1 September 1965, p. 7.
‘Edinburgh Concert.’ Guardian, 2 September 1965, p. 7.
‘Boulez at the Freemasons’ Hall, Edinburgh.’ Guardian, 3 September 1965,
p. 9.
‘Pierre Boulez at Leith Town Hall.’ Guardian, 6 September 1965, p. 7.
L.C.M.S. ‘Edinburgh welcomes stranger.’ New Zealand Herald, 11 September 1965.
Lindsay, M. Scottish Field review, 1965.
Northern Scot, 2 January 1965.
Paisley Daily Express, 28 August 1965.
Perthshire Advertiser, 2 January 1965.
Pincherle, Marc. ‘La musique.’ Les Nouvelles Littéraires, 20 March 1958, p. 10.
Samuel, Claude. ‘Edimbourg n’a pas sifflé Boulez.’ Paris Presse, L’Intransigeant,
10 September 1965.
‘Un Ahurissant Festival.’Le Nouveau Candide, 20 and 26 September 1965.
S.C. La Libre Belgique review, 20 September 1965.
Schneider, Marcel. ‘La musique à Paris.’ Combat, 27 March 1958, p. 3.
Shawe-Taylor, Desmond. ‘Adrift in the tone-continuum.’ Sunday Times,
Stadlen, Peter. ‘Boulez revives melody in his own way.’ Daily Telegraph, 30 August
1965, p. 9.
‘Messiaen’s lesson from bird-song.’ Daily Telegraph, 1 September 1965, p. 16.
‘Boulez work that sows doubt.’ Daily Telegraph, 2 September 1965, p. 18.
‘Edinburgh Festival: Composition on Platform: Pianists signal in Boulez.’ Daily
Telegraph, 3 September 1965, p. 18.
‘Exotic charm of Boulez “Le Marteau”.’ Daily Telegraph, 6 September 1965,
p. 14.
‘World of Music: Fragments for our Time.’ Daily Telegraph, 18 September 1965,
p. 11.
Stravinsky, Igor. ‘New Sound as Stravinsky Hears it: Stravinsky and Sound.’
New York Times, 15 March 1959, pp. M1, M14.
Thorn, F. ‘Eine Welt in Edinburgh.’ Süddeutsche Zeitung, 11 September 1965.
T.M. ‘Truer understanding of serial music: Dedicated and Inspired Performance by
Boulez.’ Glasgow Herald, 30 August 1965, p. 5.
‘Pianist’s Expressive Playing.’ Glasgow Herald, 1 September 1965, p. 10.
‘Boulez’s Arid Discourse in Serialism.’ Glasgow Herald, 3 September
1965, p. 10.
‘Emphasising Boulez’s Individualism.’ Glasgow Herald, 6 September 1965, p. 5.
Wilson, Conrad. ‘Music: Three Shapely Themes.’ Scotsman, Weekend Magazine,
21 August 1965, p. 5.
‘Boulez’s Cornucopia of Sound Effects: Visual as well as aural.’ Scotsman,
30 August 1965, p. 4.
‘Boulez Forms Climax: Pianist’s stunning feat.’ Scotsman, 1 September 1965, p. 6.
‘A Beautiful Example of Boulez’s Art: quartet play with warmth.’ Scotsman,
‘Maestro of modern serial music.’ Scotsman, 3 September 1965, p. 8.
‘Obscurer Side of Boulez.’ Scotsman, 3 September 1965, p. 8.
‘Weaving a Finer Fabric.’ Scotsman, 6 September 1965, p. 9.
387 Bibliography
‘WFUV To Program Stereophonic FM: Fordham Radio Station Will Begin

Broadcasts Oct. 1 – New Technique Planned.’ New York Times, 22 September
1958, p. 52.
Zipser, Alfred R. ‘Plants quicken tempo to meet stereophonic sales crescendo:
Orders Booming In Stereo Field.’ New York Times, 6 September 1959, p. F1.
Other Sources
Various collections of the following institutions and copyright holders have
been consulted and are acknowledged in the text.
Archives et Musée de la Littérature, Brussels
BBC Written Archives Centre, Caversham
Bibliothèque nationale de France
The British Library, William Glock Collection
Brussels, Private archives
The Getty Research Institute, Los Angeles
Internationales Musikinstitut Darmstadt
National Library of Scotland
Paris, Musée de la musique
Paul Sacher Stiftung, Basel:
Sammlung Pierre Boulez
Sammlung René Leibowitz
Sammlung Henri Pousseur
Sammlung Karlheinz Stockhausen
Stockhausen Stiftung, Kürten
Index
Adorno, Theodor Wiesengrund, 57–8, 82, Bergson, Henri, 71

349, 360 Berio, Luciano, 310
musique informelle, 57 Berlioz, Hector, 94, 247, 258, 322, 328
Agrupación Nueva Musica (ANM), 16 Symphonie Fantastique, 319
Amphion, publisher, 30–1, 33 Bernard-Delapierre, Guy, 140
Aprahamian, Felix, 330, 347 Bibliothèque nationale, Paris, 4, 33
Artaud, Antonin, 7–8, 357, 370 Birtwistle, Harrison, 371
Auric, Georges, 4, 247 . . . agm . . ., 94
Bonnefoy, Yves, 94
Bach, Johann Sebastian, 28, 59 Boosey & Hawkes, publisher, 270
Brandenburg Concerto No. 5, 309 Boterdael, Herlin Van, 30, 33
Bakhtin, Mikhail, 61 Boulanger, Nadia, 356
Barrault, Jean-Louis, 12–13, 17, 19, 21, 23, Boulez, Pierre
238, 303–4, see Compagnie ‘A la limite du pays fertile’ [‘At the edge
Renaud-Barrault of Fertile Land’], 160
Barthes, Roland, 219 ‘Alea’, 113, 132, 171, 176–9, 186, 192,
Fragments d’un discours amoureux, 94 210, 297
Bartók, Béla, 21, 29, 315, 328 aleatoricism/controlled chance, 100–4,
Music for Strings, Percussion and 123, 133, 172, 186, 188, 190, 192,
Celesta, 98 285, 343–4
BBC Proms, 312, 314–17, 321–4, 353 and folklore, 7
BBC Symphony Orchestra, 98, 309, Anthèmes, 103, 105
311–25, 328 Anthèmes 1, 233, 297, 358, 368
BBC Third Programme/Radio 3, 306–7, Anthèmes 2, 170, 233, 325, 358
321, 334 ‘. . . auprès et au loin’ [‘. . . Near and
Beckett, Samuel Far’], 95, 112–13, 210
En attendant Godot, 361 Balinese/Javanese music, 147–8, 169
Beethoven, Ludwig van, 194, 213, 217, 315, blocs sections, 123–9, 131–3, 136, 185
349 blocs/blocs sonores, 73, 111, 114–15,
Missa Solemnis, op. 123, 322 120, 123, 133, 164–5, 167, 234, 239
Piano Concerto No. 4, op. 58, 310 Boulez on Music Today, 113, 139,
String Quartet op. 18, no. 6, 332 161–2, 187, 195, 205–6, 210,
Symphony No. 2, op. 36, 315, 332–4 249–50, 256, 268, 297
Symphony No. 5, op. 67, 315 bubbles, 102
Symphony No. 9, op. 125, 315 chord multiplication, 112, 114–15, 118,
Benjamin, Walter, 329 120, 126, 161, 185, 209
Berg, Alban, 27, 62, 88, 98, 195–6, 218, 310, Collège de France lectures, 26, 35, 41,
315, 353, 357 95
Lyric Suite, 98 conductor, as, 10, 12–13, 15, 18, 21–4,
Three Fragments from Wozzeck, 312 218, 268, 307–12, 314–22, 324,
Wozzeck, 328 328, 331–4, 337, 349, 352
388
389 Index
‘Construire une improvisation’, 297 labyrinth, 61, 100, 102, 104, 178, 180,
Coup de dés, 6–8, 23, 171 196, 238, 255
Crépuscule de Yang Koueï-Fei, Le, 234 ‘Le système et l’idée’ [‘The System and
cummings ist der dichter, 97, 102–3, the Idea’], 217–18
322, 372 Leçons de musique, 195
Dérive 1, 103, 105, 233, 235, 324, 366 Livre pour cordes, 317, 323
Dérive 2, 103–4, 233, 371 Livre pour quatuor, 96, 110, 147–8, 226,
Dérive 3, 103 298, 317, 332, 341–2
diagonal, 62, 69 Marges, 102, 278
Dialogue de l’ombre double, 103, 105, Marteau sans maître, Le, 9, 10, 13–18,
242 20, 23–4, 73–9, 82, 97, 99, 108,
dodecaphony/twelve-tone 110–22, 133, 234, 238, 247, 255,
composition, 25, 27, 29, 35–8, 42, 307–9, 311, 317, 320, 324, 327,
45, 47–8, 50, 52–5, 211, 239, 344 332, 345–7, 349, 352, 355, 357, 370
Domaines, 102, 105, 241–2, 246, 267, ‘L’artisanat furieux’, 16, 20, 74, 83,
297, 318 112–15, 118, 120–1, 133
Doubles, 107, 242, 246–7, 249, 252–3, ‘Bel édifice et les pressentiments’,
255, 257–60, 262–3, 265–9, 312 20, 75–80, 82–3, 113–15
Douze Notations pour piano, 29, 44, 50, ‘Bourreaux de solitude’, 14, 20,
96, 211 73–4, 83–9, 112–20
Éclat, 103, 105, 192, 242, 313–14, Mémoriale, 233, 297
316–17, 319–20 Messagesquisse, 102–5, 233, 235, 355,
Éclat/Multiples, 102, 242, 320, 323 366
electronic music, 16, 250 métier, 58
Encyclopédie Fasquelle de la musique, ‘Moment de Jean-Sébastien Bach’
57 [‘Bach’s Moment’], 59
ethnomusicology, 27, 29 music and poetry, 73, 81
Études de musique concrète, 97, 110, Nocturne, 28, 51
172, 249 Notations for orchestra, 96, 103
‘Éventuellement . . .’ [‘Possibly . . .’], 95, organicism, 233–6, 243, 245
114, 139, 151, 158, 210, 226 Orientations, 79
. . . explosante-fixe . . ., xii, 95, 102, 105, Oubli signal lapidé, 13, 97, 234
170, 233–4, 271–96, 323, 327, 360, Penser la musique aujourd’hui, see
364 Boulez on Music Today
Figures – Doubles – Prismes, 99, 105, Peruvian music, 169
107, 246, 254–7, 259, 261–3, pianist, as, 22, 122, 140, 179, 181–3,
266–7, 312, 317 185, 188, 191–2, 343
First Sonata for piano, xii, 30–1, 96, 358 pitch multiplication, 14, 73,
formants, 18, 100–1, 110, 135–6, 174–5, 117, 262
177–80, 182–3, 185–6, 188 Pli selon pli, xii, 23, 97, 99–100, 102–3,
Grundgestalt, 68, 79–84 105, 140–1, 192, 193–220, 245,
heterophony, 105, 283, 363, 365, 368 308, 313, 317, 319, 327, 332,
hierarchy, 99, 110–12, 114, 128 334–9, 342–3, 345, 348, 350, 352,
‘Homage à Webern’, 12 357, 359, 365–6, 370–1
improvisation, 29, 179, 205, 225, 245, ‘Don’, 97, 99, 103, 194, 202, 209,
299 237, 239, 242–3, 245, 309
‘Incidences actuelles de Berg’ ‘Improvisation I sur Mallarmé’,
[‘The Current Impact of Berg’], 30 196–202, 205–7, 209, 211, 219,
Incises, 105, 235, 325–6, 358, 366–9, 372 234, 239, 242
‘Kandinsky and Schoenberg’, 59 ‘Improvisation II sur Mallarmé’,
Japanese music, 103, 169 196, 200–2, 205–7, 209, 212, 214,
L’Orestie, 17, 99, 162–3, 165, 202–7, 217, 219, 234, 242, 327, 334, 335,
209, 234, 238–9, 242 339
390 Index
Boulez, Pierre (cont.) spiral, 105

‘Improvisation III sur Mallarmé’, Strophes, 99, 199–202, 204–5, 209, 213,
99, 103, 139, 162, 167–70, 195–6, 216, 234, 236, 239, 241–2
200–2, 205–7, 209, 212, 218, Structures, 186
239–40, 242, 334, 337–9, 359, Structures I, 13, 97, 114, 148, 189, 226,
363–5 230, 235, 306, 343, 355
‘Tombeau’, 97, 103, 105, 194, 242, Structures Ia, 110, 133, 226, 229–32,
334, 337–9 356, 357
Poésie pour pouvoir, 105, 246, 317 Structures Ib, 110, 230–2
points sections, 123–4, 133, 184–5 Structures Ic, 110, 230–2
points structures, 111 Structures II, 97, 99, 186–9, 191–2, 306,
Polyphonie X, 97, 114, 148, 158, 171, 317, 332, 342–4, 347, 350
226–9, 235, 304, 355, 369 sur Incises, 103, 235, 325–6, 327, 355,
Polyphonies, 97, 139, 141, 147, 149, 153, 358–60, 364, 366–9, 371–2
156, 158, 159, 170 Symphonie concertante for piano and
Prélude, Toccata et Scherzo, 28–9, 51 orchestra, 97, 107, 173
‘Propositions’ [‘Proposals’], 7, 30, 47, Symphonie mécanique, 234, 238
147–8, 226 thematicism/athematicism, 18, 26, 35,
quarter-tones, 29, 99, 139–41, 143–5, 41–50, 52, 54, 171, 198, 221–2,
147–9, 151, 153–5, 158, 160, 162, 234–5, 358, 368
164–5, 167–70, 171, 229 virtual theme, 26, 368
Quartet for Ondes Martenot, 29, 50, Thème et variations pour la main
139–43, 149 gauche, 29, 44
‘Recherches maintenant’ [‘Current Third Sonata for piano, 11, 18–19, 24,
Investigations’], 105 99–102, 107, 111, 122, 170,
Relevés d’apprenti [Stocktakings from 173–86, 188, 191–2, 246, 262, 297
an Apprenticeship], 195 ‘Antiphonie’, 122, 179, 182
Répons, 103–5, 170, 233, 235, 246, 267, ‘Constellation (Constellation-
324, 358–60, 364, 367, 372 Miroir)’, 12, 104–5, 108, 111,
Rituel, 102, 105, 233, 235, 246, 262, 267, 122–37, 161, 179–80, 182–6, 191
298, 323–4, 366, 370–2 ‘Séquence’, 177, 179, 182, 188, 209,
‘Schoenberg is Dead’, 16, 56, 303 234
Second Sonata for piano, 8, 22, 31, 96, ‘Strophe’, 178–9, 182, 188
147, 305, 332, 338, 340–1, 343, ‘Trope’, 104, 179–84, 190
357, 358 time
serialism, 13, 16, 25, 59, 73–4, 85, 88–9, bulles de temps, 165, 167, 168
99, 108–38, 139–70, 172, 179, 180, discontinuous time, 100
185, 191–2, 193–220, 221–45, Eastern conceptions of, 103
262–5, 358 smooth and striated/pulsed and non
Soleil des eaux, Le, 8, 96, 147, 305, pulsed, 75, 113, 161
309–13, 353 ‘Trajectoires’ [‘Trajectories’], 303
‘La Sorgue’, 106 Trois Psalmodies, 29, 44–5, 53
‘Complainte du lézard amoureux’, Visage nuptial, Le, 13, 16, 96, 99,
65, 67–70 139–41, 143, 147, 149, 372
‘Son et verbe’ [Sound and Word’], 8, 73 ‘Gravité: L’Emmuré’, 145–9
Sonate pour deux pianos, 139 ‘Post-scriptum’, 143–5, 147–51, 326
‘Sonate, que me veux-tu?’, 182, 188 ‘Where are we now?’, 359
Sonatine for flute and piano, xii, 25–55, Bourdieu, Pierre, 268
96, 327, 332, 342–5, 355, 357–8 Bradshaw, Susan, 198, 271–2, 331
space Brahms, Johannes, 357
multi-dimensional, 250 Breton, André, 271
smooth and striated, 161 Britten, Benjamin, 328
‘Speaking, Playing, Singing’, 72 Brown, Earle, 100
391 Index
Bruckner, Anton, 357 Concrete or Noigrandes poets, 11–12

Burton Humphrey, 308 ‘Pilot Plan for Concrete Poetry’, 12
Busoni, Ferruccio, 82 Contrepoints, 303
Butor, Michel Cordeiro, Valdemar, 11
Mobile, 94 Cowell, Henry, 109
elastic form, 110
Cage, John, 3–7, 9, 13, 16, 30, 100–1, New Musical Resources, 109
109–10, 137, 151, 168–70, 171–2, Culshaw, John, 256
174–6, 178, 238, 241, 357 Cummings, E.E., 11, 195, 238
Concert for Piano and Orchestra, 188 Curzon, Clifford, 310
Concerto for Prepared Piano and
Chamber Orchestra, 171 Dahlhaus, Carl, 329
Music of Changes, 172–5 Darmstadt, 13–14, 16–18, 22, 25, 31, 57,
prepared piano, 109, 168 122, 137, 173–6, 178–9, 182–3,
Sonatas and Interludes, 171 185–6, 188, 211, 238, 249–50, 346,
square root/micro-macrocosmic form, 355
110–11 Davies, Peter Maxwell
Cahn, Marie-Thérèse, 18, 307 Blind Man’s Buff, 321–2
Calder, Alexander, 101 Davis, Colin, 318
mobiles, 101 Debussy, Claude, 29, 45, 94, 99, 195–6, 212,
Campos, Augusto de, 11–12 218, 305, 309, 311, 313, 323, 353,
Música de invenção, 12 354–5, 357, 360, 370
Poetamenos, 11 En blanc et noir, 306, 332, 342
Campos, Haroldo de, 11 Études, 99, 332
Candomblé, 17, 23, see macumba Ibéria, 21
Carewe, John, 347 Images for orchestra, 309, 311, 314
Carpentier, Alejo, 21 Jeux, 21, 98–9, 310–11, 364–5, 372
Carroll, Lewis La Mer, 314, 332–4, 338
Through the Looking-Glass, and What Le Martyre de Saint Sébastien, 313
Alice Found There, 103 Nocturnes, 314
Carter, Elliott, 371 Pelléas et Mélisande, 27, 322, 364
Cézanne, Paul, 195 Prélude à l’Après-midi d’un faune, 360
Chagrin, Francis, 304 Syrinx, 332, 342, 345
chance, 171–2, 175–6, 191, 239, 297 Deleuze, Gilles, 357
Char, René, 4, 61, 73, 83, 95, 106, 141, 195, and Guattari, Félix, 62
238, 357 diagonal, 62
archipelago, 81 rhizome, 358–9, 361
Marteau sans maître, Le, 74, 76, 94, 113, Deliège, Célestin, 81, 198, 207, 255–6,
345 298–9
Soleil des eaux, Le, 68 Deroubaix, Jeanne, 332, 347
Visage nuptial, Le, 145, 147 Derrida, Jacques, 204, 219
Chicago World Fair, 247 Désormière, Roger, 12, 31
Claudel, Paul, 10 Die Reihe, 18, 174, 250
Christophe Colomb, 10, 13, Dietrich, Marlene, 330–1
15, 306 dodecaphony/twelve-tone composition,
Cleveland Orchestra, 320 12, 16, 25, 27, 29, 35, 56, 58, 109,
Cold War studies, 355 305, 356, 366
Comédie-Française, 5 Domaine Musical, 18, 22, 24, 238, 350–1
Compagnie Renaud-Barrault, 4, 5, 8, 12, Concerts du Domaine Musical, 23, 187,
16–17, 24, 140, 162, 172, 202, 247, 307, 310, 328
303–6, 308, 327 Domaine Musical (journal), 9–10
Cahiers de la Compagnie Madeleine Donaueschinger Musiktage, 9, 18, 98, 112,
Renaud – Jean-Louis Barrault, 9 158, 186, 303–4, 308
392 Index
Downes, Edward, 316 Holliger, Heinz, 271

Drew, David, 270–1, 278, 355, 369 Honegger, Arthur, 4, 27–8, 140, 304
Dufay, Guillaume, 187
Dumaine, Gabrielle, 11 indeterminacy, 171, 173, 175, 177–8,
191–2
Edinburgh International Festival, 312, 325, intertextuality, 61, 234
327–53 IRCAM (L’Institut de recherche et
electronic music, 12–13, 16–17, 114 coordination acoustique/
Ensemble Avantgarde, 271 musique), 139, 298, 323, 358, 364
Ensemble Intercontemporain, 314 Iser, Wolfgang, 329
Ensemble Marcel Couraud, 14 Ives, Charles, 258
Experimentalstudio, Freiburg, 280, 298
Janáček, Leoš, 328
Fano, Michel, 16 Jauss, Hans Robert, 329
Fantasia, film, 248 Jolivet, André, 27, 29–30, 45, 54, 148
Fauré, Gabriel, 28, 362–3 Chant de Linos, 27, 48
Feldman, Morton, 100 Five Incantations, 27, 42, 48
Fibonacci series, 274 style incantatoire, 29, 39, 45
Formenlehren, 224 Joyce, James, 11, 99, 101, 176, 195, 238
Foucault, Michel, 88, 357 Finnegans Wake, 103
fragment, the, 93–5, 104
Furtwängler, Wilhelm, 331 Kafka, Franz, 102, 238
The Trial, 8
Gavin, Barrie, 334 Kagel, Mauricio, 8, 15–16
Gazzelloni, Severino, 31, 332, 344–5 Música para la torre, 16
Gide, André String Sextet, 15–16
Hamlet, 304 Variaciones para cuarteto mixto, 16
Giraud, Albert, 81 Kandinsky, Wassily, 59, 195
Glock, William, 4, 303, 305–12, 315–25, Kerman, Joseph, 211
327–8, 366 Kitchin, Margaret, 344–5
Goléa, Antoine, 25–6, 35, 44 Klee, Paul, 195
Griffiths, Paul, 354, 360 Klemperer, Otto, 315
Grimaud, Yvette, 33 Knussen, Oliver, 324
Koblyakov, Lev, 73, 87, 112, 117
Hacker, Alan, 318 Kohn, Karl, 192, 271
Hague Philharmonic Orchestra, see Het Kohn, Margart, 271
Residentie Orkest Koos, G de, 311, 328
halaphon, 299 Kosma, Joseph, 304
Hamburg Radio Symphony Orchestra, Kristeva, Julia, 61
312, 332–3, 350–1 Kurtág, György, 371
Harewood, Lord (George Henry Hubert Kafka-Fragmente, op. 24, 94
Lascelles), 328, 348, 351, 353 Kurth, Ernst
Hartog, Howard, 317, 319, 328, 336 linear counterpoint, 62–3
Haydn, Joseph, 322, 331
Symphony No. 104, 310–11 lament, 370
Helm, Everett, 355 Lang, Fritz
Henze, Hans Werner, 13 Metropolis, 107
Het Residentie Orkest, 255, 312 Leibowitz, René, 4, 14, 16–17, 27–9, 34, 36,
Heugel, Philippe, publisher, 13, 30 44, 48, 53, 58–9, 254, 356–7
Heyworth, Peter, 308–9, 311, 314, 317, Chamber Concerto, op. 10, 34
321–4, 330, 338, 348 Le compositeur et son double, 253
Hoffmann, E. T. A. Sonata for flute and piano, op. 12, 27,
Kater Murr, 342 38, 45
393 Index
Vier Klavierstücke, op. 8, 36 Michaux, Henri, 195

Woodwind Quintet, op. 11, 36 Milhaud, Darius, 4, 9–10, 306
Leigh, Vivien, 304–5 Mitropoulos, Dimitri, 331
Lévi-Strauss, Claude, 349 Molière
Ligeti, György, 193, 371 Amphitryon, 10, 304
Aventures, 321 Les Fourberies de Scapin, 304
lament, 370 Moseley, Carlos, 319
Nouvelles Aventures, 321 Mozart, Wolfgang Amadeus, 258
Trio for violin, horn and piano, 370 Adagio and Fugue in C minor, 309
Light, Enoch, 248 Sonata in A, K331, 332
Persuasive Percussion, 248 String Quartet in E♭, K428, 332
London Sinfonietta, 318, 323–4 Musée de l’Homme, Paris, 27
London Symphony Orchestra, 319, 325 Musée de la musique, Paris, 31, 36
Loriod, Yvonne, 186, 305–6, 332, 340–1, Musée Guimet, Paris, 27
343, 348, 350, 353 Museum of Modern Art, New York, 101
Lukomska, Halina, 332, 334–5 musique concrète, 12, 15, 17, 172, 248
Lutyens, Elisabeth
Requiescat, 271 Nabokov, Nicholas, 172
narrative, 354, 356, 360–5
Machaut, Guillaume de, 187 Nattiez, Jean-Jacques, 79, 329, 358–60,
macumba, Brazilian, 7, 17, see also 367
Candomblé negation, 299, 357
Maderna, Bruno, 13, 366 neo-classicism, 58, 355, 369
Mahler, Gustav, 196, 218, 357, 370 Neveux, Georges
Mallarmé, Stéphane, 6, 11–12, 93, 101, 170, Le Chien du jardinier, 306
194–5, 204, 218–20, 238, 313, 335, New Music Ensemble, 332, 345, 351
338, 357, 359, 362–5 New Philharmonia Orchestra, 315, 318, 353
‘À la nue accablante tu’, 170 New York Philharmonic Orchestra, 98, 319
‘Le vierge, le vivace et le bel Nietzsche, Friedrich
aujourd’hui’, 196–8, 200 Apollo and Dionysus, 368, 371
Livre, 101, 176, 195, 247, 341 Nigg, Serge, 355
‘Sonnet en X’, 170 Nono, Luigi, 16, 18
‘Un coup de dés’, 6–7, 11, 170, 176, 179 Canti per Tredeci, 307
‘Une dentelle s’abolit’, 212–14 Fragmente – Stille, an Diotima, 94
Markevitch, Igor, 247 Nouvelle Revue Française, La, 20, 247
Martin, Frank, 30
McMaster, Brian, 353 Oesch, Hans, 298
Mercenier, Marcelle, 30, 33, 182 Olivier, Laurence, 304–5
Messiaen, Olivier, 4, 16, 25, 27–9, 42, 45, organicism, 223–6
53–4, 58, 109, 140, 148, 226, 305,
331, 333, 340, 348, 350, 353, 355–7 Pappenheim, Marie, 82
Cantéyodjayâ, 109 Pärt, Arvo
Harawi, 42 Spiegel im Spiegel, 361
‘La Rousserolle Effarvatte’ (from the Paul Sacher Foundation, xi, 3–4, 33, 36, 96,
Catalogue d’Oiseaux), 332, 340 103, 108, 194, 235, 242, 285
Mode de valeurs et d’intensités, 109 Perle, George, 72
modes of limited transposition, 53 Peyser, Joan, 21, 26, 270, 328, 356
Oiseaux Exotiques, 332–4 phenomenology, 58
Sept Haïkaï, 321 Philharmonia Orchestra, 317
Vingt Regards sur l’Enfant-Jésus, 42, Picasso, Pablo, 345
46–7 Piencikowski, Robert, 3
Metzger, Heinz-Klaus, 176, 178, Pilarczyk, Helga, 310
369–70 Plate, Sybilla, 18
394 Index
Plé-Caussade, Simone, 27 Das Buch der hängenden Gärten,

Pollock, Jackson op. 15, 71
‘Allover’ painting, 65 developing variation, 63, 136
Poulenc, Francis, 4, 10, 58, 304 Die glückliche Hand, op. 18, 59, 71, 82
Pound, Ezra, 11 Erwartung, op. 17, 57, 59, 71–2, 82
Pousseur, Henri, 3, 160, 176, 191 Five Orchestral Pieces, op. 16, 21, 310
Mobile, 191 Four Orchestral Songs, op. 22, 310, 311
Prospection, 161 free atonal works, 56–7, 60, 76, 81
Presti, Ida, 14 Grundgestalt, 68, 79, 81, 83
Prévert, Jacques Harmonielehre, 63
Les Enfants du Paradis, 304 Herzgewächse, op. 20, 34
Prokofiev, Sergei ‘New Music: My Music’, 70–1
Symphony No. 1 ‘Classical’, 21 Pierrot lunaire, op. 21, 57, 62, 65, 71–2,
Proust, Marcel, 354, 357, 361 332, 345–7, 349
‘Valse de Chopin’, 63–4
Rameau, Jean-Philippe, 336 Three Piano Pieces, op. 11, 56, 60
Rampal, Jean-Pierre, 26–7, 29, 33 Variations for Orchestra, op. 31, 21, 29
Ravel, Maurice, 313 Verklärte Nacht, op. 4, 88
reception theory, 328–9 Wind Quintet, op. 26, 27
Renaud, Madeleine, 303 Schöllhorn, Johannes, 271
Reynolds, Roger, 254, 267 Schubert, Franz, 259, 347
Quick are the Mouths of Earth, 254 Symphony No. 5, 310
Ricoeur, Paul, 71 Schumann, Robert, 259, 322
Rosbaud, Hans, 9, 18, 304, 328 Dichterliebe, op. 48, 81
Rosen, Charles, 183, 186 Fantasy for piano in C major, op. 17, 94
Rossini, Gioachino, 349 Schürmann, Reiner, 83
Rougier, Louis, 238 Searle, Humphrey, 308
Rushby-Smith, John, 267 serialism, 13, 16, 25, 57, 59, 73, 99, 124,
Ruwet, Nicolas, 193 138, 191, 193–5, 209–10, 219,
355–6
Saby, Bernard, 151 Shakespeare, William
Sacher, Paul, 104, 324–5, 366 A Midsummer Night’s Dream, 365
Sacher hexachord, 233, 325, 366–9 Hamlet, 8
Sadie, Stanley, 321 Shostakovich, Dmitri, 328
Samuel, Claude, 330, 349–50 Solomon, Maynard, 217
Sartre, Jean-Paul, 357 Solti, Georg, 256, 318
Sauguet, Henri, 4, 304 Souris, André, 4, 30, 33
Schaeffer, Pierre, 248–9 Souvtchinsky, Pierre, 4, 7–15, 17–18, 20–2,
musique concrète, 248 33, 140, 172, 257, 304
Schaeffner, André, 3 Sprechstimme, 76–7
Scherer, Jacques, 176, 195 Stadlen, Peter, 330, 338–41, 343–4, 346,
Schlee, Alfred, 15, 176 349
Schloezer, Boris de, 73, 140 Stein, Gertrude, 345
Schmidt-Isserstedt, Hans, 332 Stein, Leonard, 182
Schoenberg, Arnold, 21, 27, 29, 53–4, Steinecke, Wolfgang, 18
56–66, 69–72, 75–6, 81–2, stereophony, 247–60, 262, 266–9
88–9, 98, 108–9, 195, 198, 208, Stichweh, Klaus, 271
224–5, 305, 310, 315, 328, 336, Stingl, Anton, 307
355–7, 366 Stockhausen, Karlheinz, 3, 4, 9–13, 15–20,
A Survivor from Warsaw, op. 46, 356 22, 65, 101–2, 114, 172–80, 225,
Chamber Symphony No. 1, op. 9, 25, 238, 241, 268–9, 316, 357
34–5, 53 Aus den sieben Tagen, 299
‘Composition with Twelve Tones’, 65 Carré, 105, 269
395 Index
formula composition, 223 Trethowan, Ian, 319

Gesang der Jünglinge, 22, 105, 238 Tudor, David, 31, 173–5, 177, 181
Gruppen, 105, 173, 238, 249, 258,
268–9, 316, 318 Universal Edition, 14, 20, 281, 285, 308
Klavierstücke I-XI, 22, 100, 173–8,
180–1, 191 Valéry, Paul, 219
Kontrapunkte, 22 Varèse, Edgard, 4, 162, 172, 310, 361
‘Musik und Graphik’, 188 Déserts, 361
‘. . . wie die Zeit vergeht . . .’, 95 Vaurabourg, Andrée, 4, 27, 28, 53–4
Zeitmasse, 22, 174, 176–7, Venezuelan Symphony Orchestra, 21
307 Vermeil, Jean, 21
Zyklus, 188 Villa-Lobos, Heitor, 9
Stokowski, Leopold, 248 virtual, the, 79
Stravinsky, Igor, 4, 28, 58, 103, 109, 172, Volkonsky, Andrei
195, 212, 257–8, 269, 270–1, 280, Les plaintes de Chtchaza, 316
315, 322, 325, 331, 345–6, 352, 353,
371 Wagner, Richard, 62, 196, 218, 354, 357
Double Canon (‘Raoul Dufy in Parsifal, 314
Memoriam’) for string quartet, Ring, The, 256
270 Walter, Bruno, 331
Epitaphium for flute, clarinet and harp, Wambach, Bernhard, 186
270 Webern, Anton, 11–12, 16, 18, 25, 27, 29,
Les Noces, 325–6, 371–2 42, 54, 56, 60–2, 73, 81, 88, 98, 103,
Nightingale, The, 21 114, 195, 212, 309, 311, 314–15,
Quatre Études pour orchestre, 318, 353, 355, 357
309, 312 Cantatas
Requiem Canticles, 371 No. 1, op. 29, 313
Rite of Spring, The, 21, 27, 41, 50–1, 58, No. 2, op. 31, 18, 114, 313
314, 334, 338, 372 Concerto, op. 24, 307
Serenade in A for piano, 249, 258 Five Pieces for String Quartet, op. 5, 318
Symphonies of Wind Instruments, 21, Klangfarbenmelodie, 11
98, 283, 309–12, 362 Ricercar (The Musical Offering), Bach-
Strobel, Heinrich, 14, 158, 177, 304, 317, Webern, 309–10, 311
324 Sechs Lieder, op. 14, 81
Südwestrundfunk (SWR), 14, 98, 158, 177, Six Pieces for Orchestra, op. 6, 309, 311,
246, 317, 328 312
Orchestra, 18 Symphony, op. 21, 34, 36, 38, 53,
Swedenborg, Emanuel, 65 332–4
Balzac’s Seraphita, 65 Variations for Orchestra, op. 30, 310
Variations for Piano, op. 27, 22, 44, 45
Tabachnik, Michel, 316 Westdeutscher Rundfunk, 15
Temps Modernes, Les, 28 Wolff, Christian, 100
Théâtre National Populaire, 5 Wyschnegradsky, Ivan, 140, 162
thematicism/athematicism, 358 Deux Monodies en quarts de ton, 140
Thomson, Virgil, 172 Wyttenbach, Jürg, 271
Tippett, Michael, 328, 331,
338–9, 371 Xenakis, Iannis, 193

Edward Campbell, Peter O'Hagan - Pierre Boulez Studies-Cambridge University Press (2016) PDF

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Edward Campbell, Peter O'Hagan - Pierre Boulez Studies-Cambridge University Press (2016) PDF

Загружено:

Авторское право:

Доступные форматы

Pierre Boulez Studies

edward campbell is senior lecturer in music at the University of Aberdeen and a

Cambridge University Press is part of the University of Cambridge.

List of Figures page [vii]

Part I The Context of the Late 1940s and 1950s [1]

Part III Reception Studies [301]

1.1 Itinerary for the Renaud-Barrault tour to South America in

10.6 Reduction of ﬁrst seven bars of ﬁrst occurrence of fast theme in

A composer studies volume in the Cambridge University Press series is

study of his correspondence, his compositional practice, the key phases in

We are especially grateful to Vicky Cooper, former music editor at

The Context of the Late 1940s and 1950s

generation. There are correspondences with Edgard Varèse from 1952

Three Tours of South America with the Compagnie

Not only did the Compagnie Renaud-Barrault perform in Paris, it also

The First Tour of South America

‘I am going straight back to Paris to work. I have a month or two ahead of me

To understand this statement it is important to recall that in the article

Artaud’.21 Further, in 1958, in ‘Son et verbe’ [‘Sound and Word’], he

The Second Tour of South America

The work he is referring to is Le Marteau sans maître, and references to it are

5 May – 24 May Teatro Municipal Rio de Janeiro (Brazil)

Turning later in the letter to the question of composition, he writes:

On a more positive note, he tells Souvtchinsky that he has given two

Augusto de Campos’s production, more generally, favoured a form of writ-

communications on the subject of electronic music.53 Nevertheless, with

Ensemble Marcel Couraud. Its continued importance stems principally from

In a second letter to Souvtchinsky, also from Santiago de Chile, and

there in September 1954 to work with Stockhausen on some pieces of

Leibowitz.71 Informing Stockhausen that he had been in Santiago from 15 to

On his return from South America, Boulez did not go to Darmstadt; he

The Third Tour of South America

Boulez next wrote to Stockhausen around 17 April from Mexico,83 by

Up to 2 May Teatro Belles Artes – Mexico D.F. (Mexico)

As before, Boulez continued to correspond with Souvtchinsky. Writing

In the letter from Lima, Boulez praises Souvtchinsky’s article on Le

A wonderful letter to Souvtchinsky from Caracas gives a rare insight into

In addition to the concert, Boulez reports giving a lecture-concert at

tours, his response is perhaps the most interesting information we have on

Third Sonata for piano in 1956. We learn something of the competing

Pierre Boulez’s Sonatine requires no introduction: performed throughout

Genesis and Publication

Boulez.9 Although mainly specialising in Baroque music, Rampal had

Table 2.1: Sources of the Sonatine

Early version (1946)

Fig. 2.1 Pencil draft, Musée de la musique, Paris, p. 1 (detail)

Fig. 2.2 Pencil draft, Musée de la musique, Paris, p. 5 (detail)

Table 2.2: Sonatine – formal disposition

Published version Early version

Très librement – Lent bb. 1–31 Introduction bb. 1–25

In the combination of the four movements of a sonata with ﬁrst-move-

extensive development in the middle. Instead, short transitions link the

Ex. 2.1a Pierre Boulez, Sonatine, series

Leibowitz had already constructed twelve-note rows modelled on

Table 2.3: Sonatine – serial table

Table 2.4: Sonatine – row families

Semitones Row Forms

In Webern’s Symphony op. 21 the symmetry of the row reduces the

The Movements and their Revision

the five-note grouping as semiquaver-quintuplet (flute b. 2), five demisemi-

Ex. 2.3 Sonatine, early version, bb. 14–25

In the published version, three newly composed bars were added

Ex. 2.4 Sonatine, theme (bb. 33–52)