Playing on Words

a Guide to Luciano Berio's Sinfonia

 ROYAL MUSICAL ASSOCIATION MONOGRAPHS

General editor: David Fallows

This series is supported by funds made available to the Royal Musical

Association from the estate of Thurston Dart, former King Edward
Professor of Music in the University of London. The editorial board is
the Publications Committee of the Association.

No.1: Playing on Words: a Guide to Luciano Berids Sinfonia (1985)

by David Osmond-Smith
No. 2: The Oratorio in Venice (1986) by Denis and Elsie Arnold
No. 3: Music for Treviso Cathedral in the Late Sixteenth Century: a
Reconstruction of the Lost Manuscripts 29 and 30 (1987) by
Bonnie J. Blackburn
No. 4: The Breath of the Symphonist: Shostakovich's Tenth (1988) by
David Fanning
 ROYAL MUSICAL ASSOCIATION
MONOGRAPHS

Playing
on Words
a Guide to Luciano Berio's
Sinfonia

DAVID OSMOND-SMITH
Lecturer in Music, University of Sussex
First published 1985 by Ashgate Publishing

Published 2016 by Routledge

2 Park Square, Milton Park, Abingdon, Oxon OXI4 4RN
711 Third Avenue, New York, NY 10017, USA

Routledge is an imprint ofthe Taylor & Francis Group, an informa business

© The Royal Musical Association 1985

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any
form or by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.

Notice:
Product or corporate names may be trademarks or registered trademarks, and are used only
for identification and explanation without intent to infringe.

First published in 1985; unaltered reprint 1989

British Library Cataloguing in Publication Data

Osmond-Smith, David
Playing on words: a guide to Luciano
Berio's Sinionia. - (Royal Musical
Association monographs; no. 1)
1. Berio, Luciano. Sinfonia
1. TItle II. Series
785.1 '1'0924 MU10.8491

ISBN: 9780947854003 (hbk)

This book is typeset by Goodfellow & Egan., Cambridge

The music examples were produced by Paul Courtenay
Copies may be obtained from:

Brian Jordan (agent) or RMA Secretary (members)

12 Green Street
Cambridge
CB23JU

Transferred to Digital Printing 2014

 Contents

Preface vii

1 Sinfonia and its Precursors 1

Work in progress 4
An overview of Sinfonia 5

2 Mythologiques 8
The sources of Berio's text 9
Isolated words and their function 12
Phonetic materials 13
The poetics of fragmentation 13
Coherence as structure 15
Musical structure 15

3 '0 King' 21
Structural properties of the pitch set 22
The pitch set in macrocosm 23
Climax and epilogue 25
The rhythmic set 26
Upbeat patterns 27
The growth of rhythmic stability 28
Rhythmic commentary in Sinfonia 30
Parallel processes in rhythm and pitch 30
Selective resonance 31
Independent melodic lines 32
Pitches outside the set 33
Harmonic commentary upon the chamber
version in Sinfonia 34

v
 The text and its structural potential 34
Permutation and troping 37

4 'In ruhig f1iessender Bewegung' 39

From sermon to scherzo 40
Texts for a sermon 41
Obliteration as form 43
Reduction and distortion 46
Quotation 47
Chromatic clusters 50
Semantic associations of Mahler's scherzo 53
Sources for the text 54
Interactions between text and music 56
An inventory of interrelations 57

5 Synthesis and Dissolution 72

The fourth movement 72
Rhythm and texture 73
The text 74
The fifth movement 74
Before '0 King' 77
Materials from '0 King' 78
Superposed materials 79
The final monody 81
Text and music 86
Firevs. Water 87
M.l vs. M.124 88
'La vie breve' 88

6 Epilogue 90

Bibliography 92

Index 94

vi
 Preface

This study is indebted to several people. First, and most obviously,

to Luciano Berio - not only because he provided the subject matter,
but because he suggested to me many years ago that Sintonia would
repay close attention. He read a version of the resultant monograph
several drafts back, and I must thank him for some all-too pertinent
criticisms. I must also thank Joan Peyser for permission to reprint
materials from an essay on Berio and Levi-Strauss that appeared in
The Musical Quarterly, and David Fallows for the benign editorial
harassment without which this study would have made more
arduous reading than it now does. None of these beneficent critics
(d. p.ll and 88) is responsible for the lacunae which I wish other
analysts the joy of discovering.
I base my observations throughout on the printed score of Sintonia
issued in 1972 (Universal Edition, no. 13783). This replaced a photo-
copy of the manuscript that was made available in 1%9 (no.13f783).
It corrected a number of errors in the manuscript version (though
alas, not all) and also allowed Berio to rethink various matters of
detail. In two instances where this reworking has rather wider for-
mal implications the differences between the two versions have been
noted; otherwise, I have confined myself to the 1972 score, regarded
by the composer as definitive.
The aim of the study is not to give a complete deScription, but
rather to pursue through the work all musically pertinent processes
- to avoid, in other words, the tautology of reasserting in words
what will be self-evident to the score-reader, while seeking to
deploy analytical propositions where they are most useful to the
musician: in suggesting a wider formal context within which
individual details make sense.

vii
 1
Sinfonia and its
Precursors

It is no accident that often I instinctively find myself working on different

projects at the same time, amusing myself with one and sweating blood
over another. But it is the tendency towards reunion that justifies the
tendency to many-sidedness.
Luciano Berio l

Because Berio has found the simultaneous pursuit of two or three

musical projects so congenial. his large-scale compositions have
tended to take on a special role in relation to the rest of his work.
They have provided the framework within which the exhilaration of
these centrifugal enthusiasms could be countered by an act of
synthesis. Such 'encyclopaedic' works recur throughout Berio's
career: AlIez-Hop (1959), Epifanie (1961), Sinfonia (1968), Opera (1970)
and Coro (1975--6) are all to a greater or lesser extent examples of this
tendency. But whereas AlIez-Hop and Epifanie unite their disparate
materials simply by placing them side by side, Sinfonia and Coro go
beyond this, compelling different musical concerns to interact in the
moment of composition itself. This is one of the primary reasons
why they occupy so central a position within Berio's reuvre. 2
In Sinfonia Berio synthesised several of his most long-standing
concerns, making them literally 'sound together' as the title
indicates. Throughout the sixties he had experimented with the
resources of the human voice - not merely the range of articulation
and tone that it could offer, but also the ways in which it could
enhance the musical impact of language by emphasising, or indeed
isolating its phonetic components. At first he had concentrated on
the solo voice - in Circles (1%0), Epifanie (1961) and above all
Sequenza III (1965--6) - while exploring in a fairly straightforward way
the juxtaposition of singing and speaking groups in Passaggio (1962)

1 Berio 1985, p.lOO.

2 It is also, no doubl, one of the reasons why Sinton", in particular has attracted a
good deal of commentary. The studies by Altmann and Hicks are referred 10 in Ihe
lexl, bUI Ihe reader should also nole Budde 1972, Dressen 1982, Flynn 1975, Jahnke
1973, Krieger and Stroh 1971, Ravizza 1974 and Stoianova 1974.

2
Playing on Words

and l.i!borintus II (1965). But in Sinfonia he drew together these

experiences in order to integrate them with an older, and in some
ways more complex project: the search for satisfactory ways of ex-
tending the resources and traditions. of the symphony orchestra. By
using eight vocalists, the Swingle Singers, whose experience of jazz
singing had accustomed them to close microphone work, he was
able to blend their amplified voices into the orchestral textures - a
deliberate reversal of the traditional polarity between 'choir' and
'orchestra' .
Sinfonia marked Berio's return to orchestral writing after a six-year
gap (broken by Chemins I for harp and orchestra of 1965, which
made only sparing use of its full orchestral resources). From the
mid-fifties, when Berio first began to attract international attention,
through to the last of the three orchestral Quaderni, all gathered
together in Epifanie, his remarkable grasp of instrumental texture
and colour had been widely acknowledged. This represented some-
thing of a personal triumph, for the wartime environment of his
youth had been almost devoid of orchestral music. 3 It was only after
he had begun his studies at the Milan Conservatorio in 1946, and
particularly his composition studies with G. F. Ghedini (a composer
also noted for his fine grasp of instrumentation), that he began a
rapid and voracious assimilation of the European orchestral tradi-
tion. By 1954, when he produced Nones, his first major work for
large orchestra, he was already in full command of the fragmented
timbre-counterpoint typical of his more radical contemporaries.
But a more individual approach to the orchestra was quickly to
follow, and it was one conditioned by Berio's work with electronic
media in the Studio di Fonologia that he and Bruno Maderna had
set up in Milan (1955). For here he was able to explore the super-
position of layers of sonic material, each with its own complex
internal structure. This affinity for a counterpoint of textures quickly
transferred itself to his orchestral writing, where it was complemen-
ted by an expanded harmonic vocabulary that sought to go beyond
the abstemious post-Webernian chiaroscuro of second- and seventh-
based chords to a richer mix. AUe/ujah (1955-6) was his first attempt
to come to grips with these possibilities, organising an orchestra
minus upper strings into five groups - two of brass and percussion,
one of pitched percussion and harps, and two of mixed woodwind
and low strings - and using these to superpose layers of material.
Dissatisfied with the results of this experiment, he reworked the
piece as Alle/ujah II (1956-8), adding upper strings and a wider range
of percussion to his resources, and working with more complex
groupings of instruments, each separated from the other round the
concert hall. Although this work allows for dialogue between rela-
tively complex timbre-combinations (e.g. b.396-429, or the final
pages), it is the superposition of distinct 'choirs', each with its
characteristic timbre and texture, that most faithfully reflects Berio's
1 d. Berio 1985, p.45.

2
 Sinfonia and its Precursors

electronic experiences (e.g. b.39-{iO and b.208-267, amongst others)

and provides a foretaste of the approach that he was to bring to a
more conventionally disposed orchestra in subsequent work. It was
with the Quademi - three sets of orchestral pieces subsequently
interspersed with a cycle of vocal settings to form Epifanie - that this
approach came to fruition. There Berio grouped his resources in a
way that he was to take up with only minor modifications in Sinfonia
six years later. The intervening Chemins 1 uses slighter resources but
analogous principles.
In Epifanie there are four of each woodwind and brass family, save
for six horns and a single tenor tuba. In Sinfonia grouping by four is
all-pervasive (a reflection of the eight voices and of the consequent
tendency towards four- and eight-part harmony), except that an alto
and a tenor saxophone complement three upper double reeds and
three bassoons respectively. Chemins 1, grouping its instruments by
threes, extends this 'choral' principle to keyboards with piano,
celesta and harpsichord, used to articulate a rapid heterophony that
is the direct ancestor of that played in the first movement of Sinfonia
by an analogous group of piano, harp, electric harpsichord and elec-
tronic organ. Epifanie deploys three percussion groups separated as
widely as possible from each other, but with several instruments in
common (tam-tams, caisses claires, almglocken) to allow for dia-
logue. Sinfonia does the same, with tam-tams, snare drums and
bongos in all three groups, but absorbs glockenspiel, marimbaphone
and vibraphone into the groups rather than allocating them to
separate players, as in Epifanie. In all three works there are eight
instruments in each string section (save for Epifanie's double basses);
but as well as the two normal violin sections a third is seated in a
row at the back of the orchestra, where they can provide a sustained
background over which to project other layers.
Clearly, the presence of such carefully balanced strata of homoge-
neous timbre implies a marked interdependency between harmonic
process, timbre and texture. A specific harmonic process may be
marked out by a highly integrated selection of timbre and texture, as
in Epifanie A where flutter-tongue flutes and muted trumpets (a
favourite combination) explore a series of fixed pitch-fields at the
opening and meander upward through the chromatic gamut at the
end; or the start of Epifanie B where three separate layers - an
ostinato-saturated major third from harps, pitched percussion and
muted strings, whose complex inner life sets it off against quiet
clarinet and trombone chords and violent lower string pizzicati -
pursue partly independent existences. Conversely, a single har-
monic 'object' may be passed from choir to choir, as in the opening
section of Epifanie C and in the first and fourth movements of
Sinfonia. The more complex harmonic aggregates that often occur in
Berio's orchestral music likewise depend intimately on the presence
of instrumental choirs. By the use of different dynamic contours for
different groups the internal structure of such aggregates can be

3
Playing on Words

revealed, as in Epifanie E and the third movement of Sinfonia; and

the extraordinary internal variety that can be achieved within
that characteristic gesture, a dense chord used as a sharp, staccato
attack, by subtle variations in interval distribution and more or less
integrated timbre layers is apparent throughout Epifanie - notably in
the latter half of C, and the proliferating stutter attacks of D - and in
the third and fifth movements of Sinfonia. The technique is seen at
its most concentrated in Still (1973), an orchestral study devoted
entirely to soft, staccato aggregates.
Adding an extra 'choir' of singing and speaking voices seems no
more than a logical extension of this approach. In practice, the
integration of amplified voices and orchestra has sometimes proved
difficult to achieve in the concert hall, but even so it was the
most cogent solution that Berio had yet found to a familiar problem.
He had presumably attempted to combine chorus and orchestra in
Nones, first conceived as an oratorio setting W. H. Auden's poem of
that name but then converted into a purely orchestral piece (1954).
When he returned to the problem, working with Edoardo Sanguineti
on the theatre piece Passaggio (1962), he chose to place a singing
chorus in the pit with the orchestra and to distribute a speaking
chorus around the auditorium. But his complex harmony did not
benefit from the sort of projection that a conventional chorus must
employ in such circumstances (the problem is a notoriously general
one in twentieth-century vocal works). So when he and Sanguineti
revived their collaboration in a work for smaller forces, Laborintus II
(1965), he again separated speakers and singers, but specified that
the latter should be used to microphone-singing, and have vibrato-
less, non-operatic voices. It was this solution that he adapted to a
larger scale when using the Swingle Singers in Sinfonia. Without
their experience of amplified ensemble work, the synthesis that
he achieved between orchestral sound and the vocal resources
characteristic of his solo vocal works would have been a less inviting
proposition.
Work in progress
If any single external influence is to be looked for in Sinfonia, and
indeed in much else of Berio's work during the sixties, it is that not
of another musician, but of James Joyce. That this should be so in
relation to Berio's handling of texts is relatively unsurprising: the
matter will be discussed in chapter two. But a more general prin-
ciple, in part reflecting Joyce's own praxis over the last twenty years
of his life, governs the major instrumental works produced during
this period: the Sequenzas, Chemins and Sinfollia. In producing the
various pieces of 'work in progress' that finally emerged as Finnegans
Wake, Joyce elaborated on a demotic Anglo-Irish basis an extra-
ordinary tissue of multilingual puns, adding layer upon layer of
associations. The texts often underwent several revisions, usually in
the direction of greater complexity, sometimes obliterating the

4
 Sillfollia and its hecursors

original proposition in the process. Certain texts were thus pub-

lished in several successive versions before the Wake finally
appeared. The process could, in theory, have gone on and on:
Finnegans Wake is by nature intransigently unfinishable.
Berio began to explore a parallel approach with his Chemins I for
harp and orchestra of 1965. He had produced his Sequenza II for harp
in the previous year and, wishing to elaborate its material further,
decided to add extra instrumental layers on top of it. The process
was entirely compatible with his general approach to the orchestra,
save that now one layer was complete before the others were added.
Several more Sequenzas were produced in rapid succession in the
mid-sixties, but it was not until Sequenza VI for viola (1967) that
Berio once again sensed the potential for elaboration, producing in
Chemins II for viola and chamber ensemble (1967) a work whose
density of texture frequently obliterates the contours of the original
solo line. The process was compounded in 1968, when Chemins III
added an extra orchestral layer to Chemins II; and it was taken to a
logical conclusion when the original solo line was entirely absorbed
in an orchestral reworking: Chemins IIb. 4 It is these works that
provide the context for Sinfonia, for here, too, pre-existent musical
texts are subjected to elaboration. But to understand how these pro-
cesses help to shape the work it is necessary to review its large-scale
structure, which is as much conditioned by Berio's relish for creating
a unity out of the most improbable diversity as it is by a search for
organic proliferation.
An overview of Sinfonia
The deliberate combination of heterogeneous materials is common
to several of Berio's works of the later sixties. Questo vuol dire che
(1968), close in spirit to the 'happening' so prevalent at the time,
displays this preoccupation in its most freewheeling form: there is
no score, and Berio provides only a tape and certain other materials
to be coordinated as the individual performance is prepared. Opera
(1970) unites disparate dramatic schemes - a mental hospital, the
sinking of the Titanic, a resetting of part of Striggio's libretto for
Monteverdi's L'Orfeo - all linked by the underlying theme of death.
It also brings together, as the title punningly suggests, several
autonomous works and makes them function within a dramatic
framework. But so far as the synthesis of disparate elements is con-
cerned, Sinfonia is the most rigorous and technically adventurous of
these works. For where Opera can look to its dramatic framework to
help hold the musical components together, Sinfonia achieves
synthesis by purely musical means.
It takes three disparate projects: (a) a setting of fragments from
Claude Levi-Strauss's Le cru et Ie cuit in which verbal sense, at best
only partly perceptible, gives way to a gradual build-up of nervous,
4 For a fuller discussion of both Joyce's and Berio's procedures, see Osmond-
Smith 1983.

5
Playing on Words

rapid melody; (b) a reworking of the commemorative chamber piece

o King (1967) in which, on the basis of a cyclic pitch sequence that
generates its own harmonic environment, disjointed phonemes
gradually connect to form the name of Martin Luther King; and (c) a
commentary on the scherzo from Mahler's Second Symphony that
superposes upon the original 'text' other materials of differing
harmonic densities, some derived from the orchestral repertoire,
others invented by Berio himself. On the surface these have little to
do with each other (apart from certain semantic congruences to be
discussed below), and the gong strokes that mark the passage from
movement to movement seem to be their only common feature. Yet
all three reflect a common concern: the recuperation of a vocabulary
of third-based chords.
This was no new preoccupation for Berio, but it was one that was
moving to an increasingly central position in his work, as is clear
from such direct antecedents as Sequenza IV for piano (1966) and
Laborintus II. In the first movement of Sinjonia, third-based chords
provide no more than a static background to the detailed exploration
of vocal and instrumental texture; but in the second movement a
harmonic alternation between two whole-tone harmonic areas -
each most characteristically defined by a major third plus a tone
- lies at the core of a complex and constantly shifting harmonic
network. Inevitably, the third movement brings a related set of
harmonic possibilities into play when it superposes denser layers
of commentary upon a diatonic original. At first, Berio had con-
sidered rounding off this sequence with a quiet epilogue - the
present fourth movement - which would complement both the
harmonic vocabulary of the second movement and the play with
fixed harmonic 'objects' of the first. It was thus that it was first
performed and recorded. s But these experiences convinced him of
the need for a genuine synthesis between those disparate projects -
an idea to which he had already given thought, though he had not
fully worked it out - and in the early months of 1969 he produced
the fifth movement which reworks, concurrently, materials from all
the previous movements, thereby adding a further layer of meaning
to the title of the work.
A work for voices also implies a semantic content and structure.
Although within each of the first three movements this is fairly
sharply defined, the relations between the movements depend upon
two generalized themes: water and death. The fragments that Berio
5 The recording, CBS 610791606259. with the Swingle Sing.rs and the New York
Philharmonic Orchestra conducted by the composer, was the only one readilv
available until a complete performance was issued in 1985 with Swingle " and the
Orchestre National de France conducted by Pierre Boulez. The earlier performance is
an essential document, particularly because the vivacity and raw edge uf the Swingle
Singers gave to the work an excitement that later and more professionally polished
ensembles could not easily match. But at the same time it has propagated a lopsided
and slightly inconsequential view of the work owing to the absence of the final
movement.

6
 Sillfollia and its Precursors

takes from Levi-Strauss's Le crll et Ie cuit to provide the text for the
first and last movements derive from myths recounting the origins
of water. From them, Berio extracts a single image, the 'heros tue',
with which to close the first movement and to usher in his tribute to
the assassinated Martin Luther King. Both themes coexist in the
background to the third movement, for the scherzo of Mahler's
Second Symphony started life as a song about St Anthony of Padua
(another preacher, in this context a rather futile one) delivering a
sermon to the fishes. Berio responds with a number of aqueous
quotations: among them are Debussy's La Mer, 'Farben' from
Schoenberg's Op.16, and a single moment at the very heart of the
work where the two themes meet in the drowning scene from
Wozzeck. But in its symphonic version, Mahler's movement became
part of a programme of death and resurrection, even if the detailed
significance of this movement as a moment of 'disgust for all being
and becoming'6 is not particularly germane to Berio's project, save
as a reflection of the text. This is extracted from Samuel Beckett's The
Ullllarnable and also evokes death of a sort, for its narrator is trapped
in a Dantesque limbo, attempting to talk himself into oblivion. The
fourth movement returns to the image of spilt blood with its 'rose de
sang', and the final movement, though initially supplying some of
the missing pieces in the jigsaw puzzle of myth fragments set up by
the first movement, ultimately turns back to the overriding theme of
'la vie breve' and human mortality.

• d. p.54.

7
 2
Mythologiques

When Berio chose to use the 'overture' to the 'Sirens' chapter from
Joyce's Ulysses as the starting-point for his Thema: Omaggio a Joyce
(1958), he brought into creative focus two ways of operating upon a
pre-existent text that were to be central to much of his vocal music
over the next decade. The first of these operations was provided by
Joyce himself, for he had built the opening section of the 'Sirens'
chapter by extracting fragments from the ensuing narration (of Mr
Bloom's late lunch in the Ormond bar to a background of singing
around the pub piano) and building from them a stream of images,
vivid in their isolation, that interact to create not only implicit
meanings quite divorced from their original context but also a word-
music of their own.
In Thema Berio took the process a stage further, allowing that
word-music its autonomy by breaking down the text-fragments into
their phonetic components, and using the articulatory relationships
between them as the basis for a musical structure. In this instance he
realized the piece electronically, using recordings of the text made
by Cathy Berberian;! but he was soon using symbols from the
phonetic alphabetl in his vocal scores as a means of continuing the
investigation - briefly in Circles, but far more richly and thoroughly
in Sequenza III for voice. He thus had at his disposal two disparate
principles for organizing phonetic materials: sequences extracted
from, and 'standing for' the original order of the text; and sequences
derived from simple games played out within the matrix of articula-
tory positions provided by the phonetic alphabet. The tension
between these two kept his vocal music balanced on the borderline
between sound and sense. That tension was to find one of its most
elegant resolutions in a Killg (1967), incorporated as the second
movement of Silrfa/ria. A more technical discussion of how Berio
handles phonetic materials may therefore be deferred to chapter
three, where that movement is analysed.
But although phonetic materials, rather less rigorously handled,
I See Berio 1959.
2 International Phonetic Associati{ln 1949.

8
 Mythologiques

also make an important contribution to the surface texture of

Sin[onia's opening movement, it is Joyce's own technique of verbal
fragmentation that provides the essential background to the piece.
Berio had already returned to the example of the 'Sirens' chapter
when collaborating with Sanguineti on Laborintus II, written to
celebrate the 700th anniversary of Dante's birth. Sanguineti's poetic
style - in part inspired by the experiments of Pound and Eliot -
frequently juxtaposed discontinuous fragments, sometimes in
different languages, and thus lent itself readily to the incorporation
of elements from Dante's texts. As in Joyce, these fragments, often
chosen to represent salient features of the original text, developed
a strikingly individual resonance in isolation and combined to
generate new and sometimes unexpected meanings. 3
This technique, only modestly developed in Laborintus ll, became
absolutely central to the first movement of Sin[onia, for here Berio
was at work on a text that would not immediately suggest itself as a
candidate for musical setting: Claude Levi-Strauss's anthropological
study of South American Indian mythology, Le cru et Ie cuit. 4 Like
Joyce, Levi-Strauss was passionately devoted to music, though he
invoked it not merely as an ingredient of language's poetic impact
but as an analytical tool. In Le cru et Ie cuit, the first volume in a
series of four entitled Mythologiques, he used the structures of
Western classical music as metaphors for the transformational
relations that he was seeking to demonstrate between the myths of
different tribal groups. 5 His book therefore offers a series of myth
narrations transcribed into a vivid and lucid French, interspersed
with analytical commentary that extracts salient images in order to
show their transformational relations to other myths from different
sources. Berio's work in looking for isolated images that will 'stand
for' a complex whole is thus already half-done by Levi-Strauss's
mode of analysis.

The sources of Derio's text

The mythical universe that Levi-Strauss sets out to explore gravitates
towards no central point. Consequently, the myth that provides him
with his starting-point is arbitrarily chosen (see Le cru et Ie cuit, p.lO).
It is a Bororo myth concerning the origin of rain (and featuring as
3 Examples of the process are analysed in Osmond·Smith 1981, pp.234-7.
Sanguineti's text for L1ho,i"!u, II is reprinted in Sanguineti 1983.
• All references in the text are to the original French edition, Levi-Strauss 1964. The
reader of the English translation (1970) will locate almost all page references by
subtracting 6 from the original reference - except in the introduction to the work,
where 8 should be subtracted.
, As Levi-Strauss explains in Nattiez 1973, p.5, this in turn was the fruit of a long-
.tanding frustration at not being able to compose music. By book four of
MythologUjues - L'homme nu - the procedure had clarified itself into the hypothesis that
music and myth have played complementary roles in the history of humanity, and
that as the functions of myth fade away in post-Renaissance Europe, so their role is
taken over by music.

9
Playing on Words

part of its complex story the theft of a triad of musical instruments, a

penance that the hero has to undergo for having raped his mother),
thereafter labelled M.l. Levi-Strauss then takes a further Bororo
myth concerning the creation of rivers and lakes, M.2, and demon-
strates that it is a transformation of M.l. This opposition between
'eau celeste' and 'eau terrestre' is central to Levi-Strauss's argument,
for in the opening section of chapter four he is able to demonstrate
that a Sherente myth concerning the origin of earthly water, M.124,
is a transformation of M.1, and a Kayapo myth concerning the origin
of heavenly water, M.125, is a transformation of M.2. Thus M.124
and M.125 reverse the transformational relationship between M.1
and M.2 - a situation summarized by Levi-Strauss (p.217) in the
diagram reproduced as example 1, where the Sherente and Kayapo
myths are resumed under the more generic Ge grouping.

Ex. 1

80roro Ge

eau celeste (M.l) (M.12S) eau celeste

eau terrestre (M.2) (M.124) eau terrestre

It is this structure that provides Berio with his framework. Setting

aside M.1 for the moment, he starts his movement with the opening
and closing paragraphs of M.124, which are heard up to C, and
finishes the vocal section of the movement (H3--10)6 with phrases
from a diagram in which Levi-Strauss analyses the transformational
relationship between M.2 and M.125 (,arbre resorbee', etc: p.215).
The fundamental opposition of example 1, 'eau celeste/terrestre',
is also used as a background for much of the previous material,
appearing at F9 and running through to H. Above it, a single voice
announces the opposition 'pluie douce de la saison seche'/'pluie
orageuse de la saison des pluies'. This comes from another of Levi-
Strauss's diagrams (p.221) in which he summates his conclusions
about the distinction made by the Bororo between different sorts of
rain, having first demonstrated that the Bororo myth of 'pluie
douce', M.127, is a transformation of the Kayapo myth of 'pluie
orageuse', M.l25 .
• All references to the score are by rehearsal letter plus bar number.

10
 Mythologiques

Berio has thus provided himself with two sets of materials con-
cerned with water. In between them he puts a brief section deriving
from an opposing element: fire. Immediately after his discussion of
'pluie douce' and 'pluie orageuse', Levi-Strauss recalls a myth, M.9,
which is linked to the myths of example I by the fact that each
features three important objects or elements (the three musical
instruments of M.l being one such triad). In M.124 three types of
alimentary detritus (which are therefore 'anti-nourritures') are used
by three beneficent animals in turn, in order to hide the hero from a
pursuing crocodile. Corresponding to these in M.9 is another series
of three inedible objects - rock, hard wood and rotten wood - that
cry out to the hero on his journey. Once again, Berio takes his
materials from a diagram (p.161) analysing these and other triadic
relations. The 'appel bruyant' of E2 onward is that of rock and
hard wood to which the hero must respond; the 'doux appel' of the
same section is that of rotten wood to which the hero mistakenly
responds, thereby bringing death to mankind. In turn, M.9 is linked
to M.127 and M.12S, for Levi-Strauss posits a correspondence be-
tween 'pluie douce', 'pluie orageuse' and 'rivieres et lacs' (Le.
earthly waters such as those of M.2 and M.124) and three forms of
food that are the opposites of the 'anti-foods' of M.9 (p.161).
Ex. 2

1-84

1-84 M1N

earlhlywatE'f
WATER
1-84 ';lnll-foods' conceal hero
heavenly water 1-84
(MI27)
1-84 'anh-foods' cry out to hero
FIRE
E2-F3 1-84
'antl-Ioo'h'
1-84
Gft'ilfthly w ..ter Gee.ulhly walt'r
food,
1-84 1-84
FI+-Gt.
Ge heavenly water M 127 1-84 MIlS Bororo ear:thly water
soft fait' stormy ram

G9-H7 (MI27)
(M9)
1-84
earthly fire'

1-84

hrilVE'nly fir~

H3-1 MI2SvsM2.

Berio has thus drawn from I.e cru e/ Ie cuit a closed universe of
interrelated myths, summarized in example 2. The simple Water-
Fire-Water sequence that they create is slightly complicated by two
brief insertions, here shown in parentheses: materials from M.127
('pluie douce' at E) preface M.9; and the triad of 'anti-nourritures' -
'bois pourri', 'bois dur', 'roc' - from M.9 is introduced fleetingly

11
Playing on Words

in the midst of the second Water section at H. Both are there as

vehicles for Joycean semantic games, to be discussed below on p.14.
But the second insertion also finds Berio amusing himself by ex-
tending Levi-Strauss's own game: M.9 concerns the beneficent
'feu terrestre', M.12S speaks of the destructive 'feu celeste' - the
thunderbolt which kills many people. Although Levi-Strauss ac-
knowledges the opposition in theory (p.217), he does not expand
upon it in relation to these two myths. A similar extension of Levi-
Strauss's approach can be found in Berio's treatment of M.124, the
only extended myth narration in the movement. The narrator starts
his story at b.6, impeded by a curious semantic stutter that hints at
fragmentation to come, but is almost immediately interrupted by a
taboo-imposing orchestra just as he is about to recount the group
rape of a mother by her sons. Once the orchestral tumult has
subsided, Berio jumps straight to the last paragraph of the myth -
unexplored by Levi-Strauss - which functions as a speakable version
of the unspeakable first paragraph: the sons now plunge themselves
into the sea and are, as a result, not vilified but purified. Levi-
Strauss's comment, 'ce my the nous retiendra longtemps', follows -
a remark not immediately true so far as Sinfonia is concerned, for we
move straightaway to other material, but an apt enough prophecy of
the last movement. 7
Isolated words and their function
The inchoate background out of which these fragments of narration
and commentary emerge is built from isolated words and phonetic
materials. The latter are in large measure free from any obligation to
the structure of Levi-Strauss's argument (though they all derive
from his text) and will be discussed below. Many of the isolated
words are also simply echoes of the text. But two of these words,
'pluie' and 'eau', acquire a more independent existence, and they
are joined by three others, 'sang', 'feu' and 'vie', that are not other-
wise present in the fragments that Berio has chosen to use. With the
exception of 'sang'. all of these represent one of the basic themes of
the myths discussed above, and they are deployed as background
up to the point where the relevant myth assumes the foreground.
This is most evident in the case of M.9, whose materials appear just
after E. M.9 accounts not only for the origin of fire but also for the
brevity of human life. 'Feu' is much in evidence from the opening
and continues up to E where it disappears. Likewise 'vie' is strik-
ingly introduced between D and E never to appear again. In accor-
dance with the same principle 'sang', not an important theme but
nevertheless a striking image in M.2 and M.12S, appears frequently
throughout the movement until H, where it disappears and the
myths in question take over. 'Eau' and 'pluie' are more pervasive,
as might be expected, being used consecutively to accompany the
7 For a more detailed analysis of the relations between the text of this movement
and Levi·Strauss·s book, see Osmond·Smith 1981.

12
 Mythologiques

opening narration, but disappearing during the Fire section. 'Eau'

then reappears at FS, ushering in the water myths that follow.
Certain of these words are emphasized as rapidly alternating
pairs. Further delving into Levi-Strauss's text reveals almost all of
these as distant echoes of what were originally narrative elements.
Granted that it is only a passing image in M.2 and M.12S, 'sang'
seems to assume an unduly insistent role in the movement, albeit
one that underlies the death theme running through the whole of
Sinfonia. But it appears in opposition to 'feu' both during the
opening narration and at 0, and if one takes the two words as re-
spectively standing for 'Ie cru' and 'Ie cui!' their function is quickly
clarified: as a proper preface to the whole movement before A, and
as an introduction to M.9 materials at E, in which myth the jaguar
initiates a young Indian into the delights of cooked meat (previously
unknown to mankind). Similarly, 'sang' reappears in alternation
with 'eau' at FS, ushering in M.12S where the action is precipitated
by the hero's failure to wash the blood of a hunted animal from his
hands.
Phonetic materials
These isolated words also provide the phonetic materials that form a
backdrop to the first Water section. It is mainly the vowels that are
used - [0) from 'eau', [e) from 'feu', [i) from 'pluie' and 'vie', [a)
from 'sang'S - thus creating a suitably fluid background. But apart
from the [v) of 'vie', all the sounding consonants are also repre-
sented. From E, the phonetic materials become less independent,
echoing the verbal fragments of the Fire section, after which they
disappear altogether. The Fire section also introduces random
mutterings that mirror the rapid wind oscillations. No more than a
detail here, they will reappear in the fourth movement as an impor-
tant cadential feature. Throughout these first two sections two other
vocal sounds are used: -; indicating an undifferentiated vowel
sound, and + indicating singing with closed lips.
The poetics of fragmentation
It will be clear by now that there is a deliberate gap created by Berio
between the (mytho)logic of his source and the semantic impact of
the work either as score or as performance - a gap which not even
the most expert listener can bridge. Jean-Jacques Nattiez records a
conversation with Levi-Strauss as follows: 'si la Sinfonia de Berio
utilise des passages de Le cru et Ie cuit, noire auteur [i.e. Levi-
Strauss) a I'impression que son texte a ete ramasse par hasard, qu'i1
n'est pour rien dans l'ceuvre',9 This impression is hardly surprising,
however inaccurate in point of fact. It is possible for Berio to take a
musical, time-based structure such as the Mahler scherzo of the
, A phonetician would have added a nasal -: perhaps the [a) instead derives from
the narration, where it constantly recurs: 'avail', 'adultes', 'appelait', 'Asare', etc.
• Natliez 1973, p.6.

13
Playing on Words

third movement and to fragment it in a manner at times equally

ruthless, without the knowledgeable listener becoming entirely
disoriented. But Levi-Strauss has created a complex conceptual
structure which, although it must perforce be expounded sequen-
tially, is fundamentally synchronic (and necessitates constant cross-
reference in the text to make itself understood). This castle in the air
cannot survive fragmentation; nor indeed does Berio intend it to.
Instead, he follows Joyce's example, isolating individual images in
order to give fun resonance to their poetic impact. Levi-Strauss may
properly say of his book that 'ce livre sur les mythes est-ii, 11 sa
fa"on, un my the' (p.l4), but it is a quite different sort of myth, one
that is the fruit of an attempt to 'reduire des donnees apparement
arbitraires 11 un ordre' (p.lS). By fragmenting that order into a
further set of 'donnees apparement arbitraires', Berio proposes to
release the poetic potential of Levi-Strauss's language, to restore
ambiguity where coherence had supervened. (And the present
attempt at exegesis offers a further swing of the pendulum.)
Isolated phrases can give rise, here, as in Joyce's 'Sirens' chapter,
to new semantic affinities. It is this possibility which prompts Berio
to insert a water myth fragment at the start of the Fire section, and a
fire myth fragment in the second Water section - in both instances
creating a fleeting combination that is the very opposite of their
relationship in Levi-Strauss's text. Thus when, at E, the 'pluie
douce' of M,127 is juxtaposed with the 'doux appel' of M.9, the
listener might well suppose the latter to be a poetic attribute of the
former. In fact, as we have already seen, the two are in direct
opposition: the 'doux appel' emanates from the 'bois pourri' which
is the 'anti-nourritures' opposite of the 'plants cultives' with which
the 'pluie douce' is associated; and when the 'bois pourri' is finally
named at H, it is associated with the 'arbre resorbe sous l'eau' of
M.2. Again, the listener might well suppose that the latter consisted
of the former. In fact, the' arbre resorbe sous I' eau', far from rotting,
is instead the bringer of life-giving water to the vegetable realm,
whilst the 'bois pourri' is the bringer of early death to human
beings.
Berio also poeticizes the text by emphasising its musical values:
simple phonetic oppositions, assonance and rhyme. Most of these
are perfectly obvious, and need no analytical commentary; but it is
worth underlining the overall sequence. The narrations of the first
Water section are complicated by repeating certain words or phrases
several times, creating a semantic stutter that galvanizes the whole
of the vocal section into nervous life. These repetitions underline
front-back oppositions between the final vowels of successive frag-
ments thus treated: 'j) y avait' (front) versus 'plusieurs' (back),
'aujourd'hui' (front) versus 'vers Ie fin' (mid-mouth), etc. After
vowel oppositions, the Fire section from E to F focuses on vowel
identity through the recurrent Ii] of 'vie' and 'pluie'. Vowel identity
gives way to consonant identity in the second Water section (F8-H7)

14
 Mythologiques

with its recurrent 's's and 'r's, and both identities are finally sub-
sumed in the recurrent syllable 'tu' at the close of this section
('tuant', 'tue', 'rituelle': d. H8-11).
Coherence as structure
Berio compounds the listener's estrangement from the structural
relations of Levi-Strauss's text by presenting different fragments
Simultaneously, forcing him to grasp at momentarily comprehen-
sible gestures within the general melee. This fascination with
working at the very limits of coherence is confined to a verbal level
in the first movement, but it is to proliferate into other, strictly
musical levels in the third and fifth movements. Even so, different
levels of comprehensibility are available to Berio as an element of
musical form, and he makes full use of them. The crucial factors
here are the semantic 'completeness' of each piece of material, the
number of superposed verbal events, and the various modes of
articulation from singing to whispering. Bringing these together,
three clear-cut sections emerge between the start and I, where the
voices break off. In the first, stretching to E, two passages of con-
tinuous myth narration for solo voice are flanked by passages con-
sisting of phonemes and isolated words. Although stylised by the
use of fixed intonation and internal repetition, and although
interrupted by occasional outbursts from other voices and eventually
plagued by a manic echo, these narrations constitute the most lucid
moment in the movement, and stand out clearly against multifarious
background murmurings. The odd isolated word emerges from the
phonetic background in the surrounding episodes; but here atten-
tion is focused on vowel timbre and rhythmic texture. From E
through to the fourth bar of F a new semantic texture is developed
in which, with continuous narration having disappeared and with
phonetic materials taking a subsidiary role, a polyphony between
sung fragments is established. But with long, melismatic vowels and
two or three different lines superposed, the ear must strain to catch
at any meaning. The final section however returns to the compara-
tive clarity of a single speaking voice versus the other seven in a
part-whispered, part-spoken unison, although by now the material
consists entirely of fragments and isolated words. Clearly, these
three types of semantic texture correspond directly with the alter-
nating themes of Water-Fire-Water concealed beneath their surface.
Musical structure
The deliberately parsimonious musical materials deployed during
the vocal part of the movement simply underline this tripartite
division. They also establish a harmonic language that is to per-
meate much of the score, a language based upon accumulations of
thirds. The first Water section is underpinned by an alternation,
three times repeated, between two such chords. The first of these is
an eight-part chord that establishes a model for much of the har-

15
Playing on Words

monic writing to follow with its wider spadng at the bottom (here,
typically, a fifth) underpinning a chain of thirds (diminished seventh
and fifth chords superposed); the second is a four-part chord. They
are set out in white note-heads in example 3. The return of chord 1
at A and C introduces another of Berio's harmonic preoccupations.
For the surrounding orchestral parts now flesh it out to chromatic
saturation, expanding the third-chain and shifting the bare fifth at
the root of the chord down a fifth to produce the additions notated
in black note-heads in example 3.10 Since two notes, c and eb, are
common to chords 1 and 2, it follows that two others are missing.
These two notes - f and gU - are used to initiate additions to chord 2
at its first recurrence: the gU at the bass of a downward extension of
the chord at A9, the f initiating a chord of all the other missing
pitches save bb four bars later (d. ex. 3). When chord 2 returns for
the third time, an added bass line begins with f a& (C5), while the
high flute d prepares for brief saturation at Cll. This simple
prindple of a basic harmonic layer being surrounded by other layers
derived from its chromatic complement will recur, in more sophisti-
cated form, throughout the rest of the work.
Ex. 3

A A9 A13~B2
04 E3 E4

A 2 3 4

Such harmonic simplicity allows for a rich play of textures. They

are of no great complexity, for the constant superpositions all derive
from quavers and triplet quavers; but again they provide simple
examples of devices that will constantly recur. Berio's delight in
seeking out correspondences between vocal and instrumental sound
is immediately apparent when the full orchestra enters at A. Rapid
crescendos and decrescendos within the chord establish an analogy
between, on the one hand, voices playing with vowels and, on the
other hand, an octet of trumpets, hom and trombones each playing
with a different mute. The subsequent resolution onto chord 2 from
A4, and its parallel passage from C4, both allow Berio to demon-
strate how his wind and brass sections resolve into four-part choirs.
The introduction of the upper range of complementary pitches at
A13 provides a first example of another constantly used device
10 Readers of the earlier facsimile score (1969) should note thot Berio originally
wrote this chord with a g# in the bass and a g~ at the top of the treble stave, thus
emphaSiSing the lOgiC of superposed diminished chords. He then revised the score
before definitive publication in 1972 so as to conserve the characteristic fifth at the
base of the chord, and to contrast with the gU-based addition to chord 2 (d. A9).

16
 Mythologiques

when a sforzando line - here just two notes in the trumpet - leaves
behind its own harmonic echo (in flutes 1 and 2); and at B the
sopranos profit from this new harmonic field to initiate a further
characteristic gesture, a nervous staccato line permutating a few
fixed pitches.
At 04 Berio closes the first Water section with a hybrid chord, a
cross between the extended forms of chords 1 and 2.11 A transitional
passage (EI-3) of shifting harmonies, setting the juxtaposition of fire
and water materials discussed above (p.ll), stabilizes momentarily
on a diminished seventh chord over an alien at., emphasising once
more chains of minor thirds. It then gives way to a new, stable
chord of two superposed minor thirds, chord 3 in example 3, upon
which the remainder of the Fire section is based. Its characteristic
texture is no longer one of repeated attacks, as in the first section,
but rather of oscillation - uneven in the voice parts, consistently
rapid in clarinet and saxophone. Against these, piano and harpsi-
chord execute ornamental flurries, loosely focused upon the notes of
the chord. Both of these features are to assume a more significant
role once the voice parts have exhausted their materials.
At E8 the first soprano leaps to a high g, directly anticipating her
analogous gesture at the climax of the second movement (d. II, E9).
Here, however, she opens up the register that will provide the har-
monic basis for the second Water section, while at the same time
placing it 'beyond the reach' of the other voices, who from now on
will speak. That harmonic basis is quickly established: a gb-bb
cluster, chord 4, saturated save for the g which introduced it. It
completes a simple harmonic process, easily seen by comparing
chords 1 to 4 in example 3. Chords 1 and 3 are governed by minor
thirds, chords 2 and 4 by major thirds. Chords 2, 3 and 4 each
conserve one third relationship from chord 1 - with chord 4 trans-
posing the lower major third in chord 1 up an octave, and the
borrowed third from chord 1 providing the only minor third in
chord 2.
The inner articulations of chord 4 recall those of the first Water
section, as do the changes of timbre as it is passed from one high
wind choir to the next. But this is no longer the only focus of
interest. For what were no more than ornamental flurries in the Fire
section have now consolidated themselves into a nervous, widely
leaping melodic line articulated by the strings, with heterophonic
commentary from woodwinds and occasional brass. Initially this is
based upon a rotating pitch sequence, which becomes increasingly
fragmented and dispersed as a manic orchestral unison establishes
itself. The eighteen-note sequence is set out in example 4 in its most
typical order (based entirely upon the string line up to 13, though
the wind elaborations are in the main quite easily derived from it).
As might be expected, the pitches employed are the chromatic
11 Its dual nature would have been even clearer in the earlier version: see note 10

above.

17
Playing on Words

Ex.4

10
10

H
2
10

4
10

10 10
10

10
10

3
10
10

10

10 10

complement of chord 4 - save for cD, which is to find a different

function. They are deployed so as to form a pattern of internal
permutation and substitution within its two halves (indicated by~)
and a retrograde structure pivoting around the final e of the first half
(indicated byro). The retrograde demands a g as the penultimate
pitch class of the first half; but if it were also to be retained as a
component of the last group of four, too radical a dominance would
be achieved by d and g, each occurring four times where the others
occur twice. Accordingly, Berio replaces it with an lb, thus establish-
ing a recurrent d-e. coupling which is to act as an idee fixe through-
out this section, even when the original pitch set is lost. The various
realizations of the sequence are set out below, running through to
the point where at best only permutated two- or three-note frag-
ments are discernible within the continuous hemidemisemiquaver
movement. The first four rotations of this sequence - two complete,
two fragmentary - plus their heterophonic commentaries amalga-

18
 Mythologiques

mate with the spoken texts and the polyrhythmic cluster articula-
tions to form a complex, layered texture that is completed by an
independent bass line in which the c#/d~ absent from both cluster
and melodic line is prominent. At first (F4) a nervous staccato bass
permutating d~, d and eb (and later adding g and c) underlines
the setting of the 'eau celeste/terrestre' opposition and the M.127
materials - developing further the idiom introduced by the sopranos
at B. But as the text moves on to the interpolated fire myth materials
('bois pourri/dur', etc.) the woodwind heterophony plunges down-
ward to establish a low c# that initiates an independent line. A
further layer is established a bar later, at H, where the final eb of
statement two of the pitch cycle generates an independent, gradu-
ally quickening oscillation between eb and db in horns and trom-
bones. What is here but a single element, helping to prepare the
forthcoming climaxes at 14 and 19, will reappear in the fourth
movement as the central harmonic focus.
All of these layers except one are to be swept quickly away.
Voices dissolve into noise at 12 and disappear. Two bars before,
string glissandi have expanded chord 4 to a saturated fifth, e-b,
which erupts into a first climax at 14. It is answered immediately by
a c from the horns; keyboards reply with the opening d and e~ of the
pitch cycle, plus a high db; and, thus prompted, the cluster expands
into a vast agglomeration at 19 that completely saturates the chro-
matic gamut from the piccolo's a bin a/tissimo down to the cellos' db,
and then proceeds by steps of two through to six semi tones to the
bass tuba's f. Such monster aggregates are not to be heard again
until the central section of the third movement, where they become
a major feature. Here, however, the chord rapidly dissolves, leaVing
only harp, piano and harpsichord who have taken over from strings
as the main bearers of the melodic line.
This trio at first articulates a melodic line still bound by the pitch
cycle (d. ex, 4, cycle 5). But by J5 it has begun to permutate pitch
order; and although a recognizable relationship is regained from
J8-lO, it is cut short by a c-f# interruption and, retaking the start of
the cycle, freewheels off into an autonomous melodic line. Although
this continues to gravitate around the pitch cycle's d and eb, it
employs all pitch classes except a. Its oscillating patterns create a
phantasmagoric virtual counterpoint of tone and semitone steps not
unlike that which Berio had explored extenSively in Sequenza VI for
viola of the previous year, a similarity underlined from K on when
upper strings realise the implicit harmonies as tremolando chords.
The melodic line gathers both momentum and weight, being scored
first as a unison, then in octaves. After a hom interruption at J20 the
thickening process continues through parallel fifths (K4), triads (K7)
and dominant sevenths (L) into discordant contrary motion (L4) and
thereafter saturation.
The purpose of excluding a from the melodic line becomes clear as
this accumulative process reaches its first climax at L. The tutti is

19
Playing on Words

answered by a solo piano with a spread chord whose top note is a,

and whose other notes, d b g~ f, form the diminished seventh chord
not represented in chord I. Three times the piano answers the full
orchestra with this chord before resolving into chord I. The piano
broods further on this relationship in L5--6; and then, as the voices
re-enter with chord lover a saturated orchestral diminuendo, it
launches forth in an independent rhapsodic vein that is to appear
once more at the start of the last movement. The orchestral explo-
sions at last exhaust themselves, leaving a solitary flute at L11
which, as the voices resolve chord 1 into chord 2, completes the
chromatic field with f and a., plus a low f ~ echoing that of the piano.
Over chord 2 a final orchestral flurry leaves chord 1 in the strings.
This then evaporates, and the three gongs close the movement as
they had opened it.

20
 3
'0 King'

Berio produced the chamber version of 0 King in 1967 in response to

a commission from the Aeolian Players and as a tribute to the
memory of Martin Luther King, whose name, slowly constituting
itself from a series of phonetic fragments, furnishes the verbal
material for the piece. It is scored for mezzo-soprano, flute and
clarinet, violin and cello, and piano; and it revolves around a cycli-
cally repeated pitch set (ex. 5, below) which eventually breaks
through to a climax and fades away. The pitch set is presented by
the voice and by the piano which also (a) emphasises certain notes
by sforzandi and (b) occasionally embroiders the set with the pitches
excluded from it. Both pedals of the piano are permanently down,
so as to provide a continuous and general resonance. This is fined
down into selective resonances by the other four instruments which
sustain or echo particular notes of the set, with the flute occasionally
introducing independent melodic fragments.
When Berio decided to incorporate this piece into Sinfonia in the
following year, he chose to maintain the original materials more or
less intact, but added extra instruments that would be able to
provide a subtle elaboration of each of the functions mentioned
above. Thus vibraphone and, after a while, harp join the piano in
following through the complete pitch set (though the harp also
reinforces the flute's independent melodic fragments, and con-
tinually explores the piano's extraneous pitches). The selective
resonances are enriched by the addition of three more female voices
and by the expansion of the functions of the original violin and cello
lines, now distributed among violins C, violas and cellos. (The
interplay between violins C, at the back of the orchestra, and the
violas and cellos at the front introduces a sense of aural perspective
quite absent from the intimacy of the chamber version.) The lower
strings also provide resonance for the lower extraneous pitches, as
do the tenors and basses. Sforzandi on the piano are reinforced by
saxophone and horn ,- though the latter also aids the resonance
process. Thus apart from harp and horn, which have multiple (and
complementary) roles, specific groups have specific harmonic

21
Playing on Words

functions. The full orchestral complement is reserved to underline

the climax and final bars.

Structural properties of the pitch set

The pitch set consists of seven pitches, f a~ a b~ b~ d d, each used
three times. It is divided into three sections by the consistent
coupling of f and a. The first section uses only four pitches, the
second all seven, and the third all seven plus repetitions of those
pitches omitted from the first section. The properties of the resultant
set are analysed in examples 5, 6 and 7. Recurrent pitch groupings
are shown in example 5 byrl, with retrograded groupings indicated
by \ J . The two three-note repetitions are particularly significant,
since they establish the complementary whole-tone nuclei fa band
d b~ a~. These act as centres of gravity within the resonance system,
encouraging the ear to perceive the set as oscillating between two
whole-tone areas.

Ex. 5
i

22
~

22

Ex.6
'F group' 'D group'

Ex.7

22
 '0 King'

The alternations between them are set out in example 6: the first
discrete, the second overlapping. Each switch from one whole-tone
area to the other is heralded by a semitone. The two semitones in
section ii of the set (see ex. 5) focus attention on a further three-unit
repetition, this time not of pitch but of interval content as f a a b is
followed by /Jb d c#. This conjunction stands at the centre of a
fourteen-pitch sequence richly structured by mutual pitch and
interval relations, which is followed by a relatively 'loose' seven-
pitch sequence, as in example 7.
We have thus at any rate three different ways of subdividing the
twenty-one units of the pitch set according to different criteria:
4+7+ 10 (the triple return of fa); 6+3+5+2+5 (whole-tone areas);
and 14+7 (pitch and interval repetitions). It is precisely because of
its multi-faceted nature that the resultant set is capable of yielding so
much musical substance when differently inflected at each repetition
by the rhythmic set (see pp.26ff).
The pitch set in macrocosm
The pitch set is also divided up by a series of sforzandi which,
during the course of four and a half cyclic repetitions, map onto it a
macrocosmic version of itself. The pattern thus created is set out in
example Sa. The end of each section of the macro-set is indicated by
a vertical line, the end of the first cycle of that set by a double
vertical line. Repetitions of individual pitches in the macro-set are
placed in parentheses, interruptions (including the sforzando con-
figuration in cycle 0) in square brackets. Since cycles two to five
begin at rehearsal letters B-E, I shall use these for reference, quali-
fying them with the numbers i-iii to indicate the three sections of
the set.)
A simple displacement process generates the macro-structure.
The two cycles start together on f, and the macro-set immediately
dislocates itself by waiting for the a and b of Aii, the c# from Aiii,
and so forth. Naturally the two cycles will once again coincide at Bii,
and it is an obvious logical possibility simply to repeat the initial
dislocatory gesture whenever this occurs. This would create the
hypothetical model shown in example 8b. As will be seen, by the
start of the seventh cycle one has created a closed system which
returns not to its starting-point, but to the start of the second cycle.
While logically impeccable, this structure is musically somewhat
over-extended, and provides for the contrast of groups of sforzandi
only as and where the repetitions shown in example 5 permit them.
By allowing the macro-set to run parallel with the normal set for
several notes at a time in Bii and Ciii, Berio abbreviates this hypo-
thetical model by two cycles, and at the same time provides a richer
contrast between isolated sforzandi and groups of sforzandi.
1 I label the first cycle A even though rehearsal letter A in the scores occurs at the
position corresponding to the third note of Aii in this and subsequent diagrams.
Rehearsal letter E in the Sintoni. score occurs one bar before the start of the fifth cycle.

23
 Ex.8

iii iii iii

iii

iii

Ex.8
 '0 King'

But these parallels are themselves subject to rigorous logical

a.
restraints. The first coincidence of macro-set and normal set occurs
on the of Bii. It is a matter of indifference whether the parallelism
is broken after the subsequent b. or d, since in either case the next
coincidence will be in Biii; the choice is therefore made on the basis
of rhythmic considerations, the d of Bii falling on a downbeat. To
extend the parallelism to either the c# or the b would only modify
the sforzando distribution pattern until the g # of Ciii, and would
make the start of the third macro-group coincide with the start of C,
thus creating a simplistic situation in which one section of the
macro-set would correspond to each repetition of the basic set. To
extend it any further would create an implausible imbalance in the
sforzando distribution pattern, without providing any novel exit
from its internallogic. 2
A similar checkmate characterizes the second parallel group which
starts on the d of Ciii, ushered in by the first break in the macro-set:
the sforzando f at C3 which by now constitutes an insistent and
independent pedaL To break the coincidence after either c# or g#

situation of Dii. To extend it to b. or a.

would merely generate an extra set-cycle before arriving at the
would again produce Dii;
and on grounds of rhythm and excessive length Berio discards both
alternatives. The extension to the final b. would produce a different
distribution pattern up to the crucial and unavoidable c# of Diii, but
is clearly too long; the extension to f would close the cycle.
This closure is, in any case, inescapable. The d of Diii equals the
c # of Aiii. If the pattern is not immediately broken, the cycle may
continue indefinitely. Accordingly, on the first b. of Diii, the
sforzandi leap to g, gt d and b., and then focus on the second bl>,
marking it out as the note which, when it next recurs, will usher in
the disruption of the normal set cycle and provoke a climax.
Climax and epilogue
The second b. of Diii also introduces a new feature: rapid crescendi
and diminuendi; and these continue as the fifth cycle starts at E,
confusing the clear-cut relation between normal structure and
macro-structure. Indeed, the distinction between the two threatens
to dissolve entirely when, in Eii, they reach the parallel to the first
coincidence at Bii, discussed above. This is avoided only because
neither ever completes Eii. Instead, after a long-held crescendo on
the b. of Eii, the pitch set is disrupted and leaps to a high g, after
which the melodic line falls away to a close. The resultant structure
is shown in example 9, where the slightly more complex pattern
used for the Sinfonia version is indicated by the additions in paren-
theses. Clearly, the melodic line after the climax consists of a syn-
thesis of the retrogrades of the first four notes of ii, and the first five
notes of iii. By adding a b. in the Sinfonia version to the alien a ~
2 The' and a of Biii would likewise lead to a further coincidence on the g# of Ciii;
the subsequent b. d and c ~ would all generate line 5 of example 8b.

25
Playing on Words

Ex. 9

iii

iii

iii existent

iii existent

already existent in the chamber version Berio creates an analogous

structure before the climax, integrating the pitches missing from the
retrograde version of iii to be heard subsequently with the near-
complete version of ii discussed above_ Thus the missing Eiii none-
theless maintains a substantial presence.
At the end of the chamber setting, the voice part falls away to do,
thus ending with an inversion of the omnipresent /-il relationship,
and at the same time recalling the punctum do/ens at which the set
was shattered in Eii. In the Sin/onia version, the voices continue this

a descending major scale down to the first bass, with the e.

process downward to form an eight-part chord (the interval order of

second bass balancing by inversion the major ninth between / and

of the

the climactic high g).

The rhythmic set
The twenty-one unit pitch set is combined with a twenty unit
rhythmic set, so that with each repetition of the cycle the pitch set's
internal relationships are inflected differently by the rhythmic set's
displaced position. This is shown in example lOa, where the rhyth-
mic values assigned to each note of the pitch set are set out as
multiples of a quaver, with the variants introduced into the Sin/onia
version noted below each line of the complete chamber version_
Clearly, the rhythmic set is not a fixed entity like the pitch set, but
gradually evolves. This is made immediately clear by the final b. of
Aiii which, instead of starting the second statement of the rhythmic

a.
set with a further 8, elongates it to 17, which in turn becomes 7
when the third statement starts on the of Biii.
Such elongations apart - and their function will be discussed in
due course - the rhythmic set is composed of eight different units

26
 '0 King'

Ex.toto
Ex.
'F group' °0 group'
'0 ''F
F group' '0 group'
8 8 iii

8
A 8 7 6 f€ 6 7 3 2 ,
1 4 3 5 2 8 4 2 7 ,
1 6 3 U 17
'7
3 8
B 7 6 6 6 6 3 2 1 4 4 4 4 4 4 2 7 1 6 3 I1 7 7
6 6
cC 6 6 6 8 2 3 1 4 3 5 5 8 4 2 7 1 1 311 7
311 7 7

0
D 6 6 8 3 2 , 4 3
3 3l\ ,lI 6
3l\,,, 4 2 5 ,
1 6 311 1,
311 7 7 12
12
'2 18
'8
0
D 0
D 0
D 0 6
1D 33 3 1 ,9l!
,sl! III
,~

0
D
A
A

0B
D

cD0
0o
D

'F group' dominance; 'D group' dominance as determined

determincd by upbeat patttms
patterns

ranging from one to eight quavers, and deployed so as to avoid

repeated units within the sequence. This ensures that such aurally
evident pitch repetitions as the tab common to sections i and iii of
the pitch set are always articulated in rhythmically contrasting ways.
The structural properties of the rhythmic set are instead dictated by
three partly related devices: (a) extreme durational contrasts,
producing short 'upbeats' to long 'downbeats'; (b) sequences of
durations that can be heard as implying metric rather than additive
organisation (in which upbeat patterns play an important role); and
(c) sequences of similar durations such as the 8:7:6:6:6:7 group
which distinguishes the first whole-tone grouping in Ai and Aii
from the second (articulated by a sharply contrasting 3:2:1).
Upbeat patterns
The oscillation between whole-tone groups that is so fundamental a
feature of the pitch set is intimately bound up with the deployment
of short-long durational patterns because of the implied accent that
the longer note carries. To aid discussion, I shall refer to the two
groups as f groups and d groups, reflecting the inversional relation-
ship (f a b vs. d bb ab) that exists between them, as in example 6.
The first upbeat relationship of the rhythmic set is provided by the

27
Playing on Words

1:4 relationship that first occurs in Aii.3 Section ii of the pitch set has
a d-group in the middle, flanked by {-groups; and in statement A the
downbeat provided by this relationship falls on the first note of the
second {-group. But as the 1:4 relationship shifts back one place in
statements Bii, Cii and Dii, the downbeat now falls within the d-
based group. Thus the d-group in Aii, although distinguished by
shorter durations, is nonetheless subservient to the surrounding f-
group, whereas in the other three statements it contains a rhythmic
accent and therefore resists that dominance.
By contrast, the next upbeat relationship, the 2:8 of Aiii, under-
goes no such modification of function, remaining within the f-based
group spanning sections ii and iii of the pitch set throughout its
three displacements (the last of which presents it in non-propor-
tional diminution).
The only ambiguous area within the pitch set from the point of
view of whole-tone oppositions is the overlap between the two
groups provided by the pitch sequence b d c ~ g # in iii; and the two
remaining upbeat relations in the rhythmic series, provided by the
rhythmic sequence 2:7:1:6, are used to resolve this ambiguity of
harmonic orientation one way or the other. In statement A, the
2:7 relatiOOl gives the c ~ an upbeat function, thus ensuring the
dominance of the d-group from the first d of Aiii, through to the end
of A, and reinforcing it with the 1:6 relation that follows. But in
statement B, the c # acquires a downbeat function, maintaining the
dominance of the {-group up to that note, after which the 1:6 re-
lation vigorously asserts the dominance of the d-group. Statement C
is likewise clearly polarized, with both downbeats asserting the
dominance of the d-group so that the dominance of the f-group can
extend only to the b of Ciii, as in statement A. Naturally, state-
ment 0 will revert to the pattern of statement B; only this time the
polarities are exaggerated with 1:6 converting to 1:12 in the Sin{onia
version, and with a new upbeat pattern being introduced by
converting 7:6 of cycles A-C into 1:6 in order to give an equally
emphatic downbeat to the d-group. These relationships are sum-
marized in example 1Ob.
The growth of rhythmic stability
The upbeat patterns also serve to articulate areas of comparative
metric coherence within an additive context. The first of these, the
3:2:1:4 in section ii of the pitch set; may be scanned as a 6/8 pattern
or indeed as a syncopated 3/4, which is how Berio notates it. The 6/8
interpretation is reinforced in statements A and D by articulating the
last dotted crotchet of the previous duration as a vowel change,
while in statement B the 3/4 interpretation is somewhat reinforced
by the preceding crotchet-based durations. The implication of 3/4
, The 'downbeat' character of the 4 is further amplified by the implied 6/8 of the
3:2:1 sequence, discussed in the next section,
4 In this contf"xt the 4 is relevant as an attack rather than a duration.

28
 '0 King'

created by the permutation of this pattern in statement C will be

discussed below.
The second metrically coherent sequence, the 2:8:4:2:7:1:6:3, is
rather more complex s After a crotchet upbeat, it may be scanned
(and is notated) as an alternating pattern of 414 and 3/4. In the
chamber version this pattern is maintained throughout the first
three statements, but has its durational relationships (though not its
upbeat patterns) modified in statement 0, with metric consequences
again to be discussed below.
As was noted above, the whole of the first [-group in A is unified
by a group of similar durations, 8:7:6:6:6:7. This sequence is how-
ever subjected to immediate transformation, so that by its third
repetition in statements B-C it has become 7:6:6:6:6:8. A crotchet
additive basis now predominates over the original quaver basis; and
dotted minims have now become more emphatically normative. The
effect is compounded in C by the 2:3:1:4 permutation of the 3:2:1:4
sequence discussed above which follows, leading in turn to the
crotchet-based second sequence, with only a minor additive disloca-
tion in between. The ambiguities of statement A (additive quaver
base vs. 6/8 metric base vs. metric alternation of 3/4 and 4/4), already
under threat in statement B, have by now been ironed out into a
norm of three crotchets, occasionally contradicted by groups of four
or two. This final contradiction is to be eliminated in statements 0
and E, but not before a flurry of disruption as the 6/8 pattern
reasserts itself, followed by a contradiction of the quaver additive
basis in the 3%:1 V3 relationship. The second metric sequence now
follows, so modified as to scan entirely in 3/4 and, after a single bar
of 214 (3:1), the music settles permanently into this metre until the
precipitation of the climax.
An inordinately extended b b at the end of statement A had
ushered in the dislocation between pitch and rhythmic sets. A
similarly extended b I> at the end of statement 0 now signals the
collapse of the fixed relationship that has since then been main-
tained between them. The rhythmic series shifts not one place but
three, relative to the pitch series, so that Ei is now articulated by the
first metric sequence (3:2:1:6). This sequence is recapitulated for Eli
in the same permutated version as Cii (2:3:1), thereby (a) breaking
the rhythmic series, and (b) literally repeating an association of pitch
and rhythm for the first time in the piece. 6 The music freezes once
more onto bb, and breaks through to the climax - a 30-quaver long
high g?
The epilogue of the chamber version uses the rhythmic sequence

5 Again. the final 3 is relevant only as an attack.

• With the exception of the dotted minims that will inevitably start both statements
C and D.
7 Its five dotted minims thus usurp the well-established pattern of four dotted
minims in the rhythmic series; but in the 5infonill version this also becomes five
dotted minims in its last cycle.

29
Playing on Words

4:6/ 7:2:41 7:6:18, here grouped to show possible derivations from the
rhythmic set: 4:6 is a summary retrograde of the first metrically
coherent sequence, 3:2:1:4;8 7:2:4 is a retrograde of part of the
second metric sequence; and 7:6:18 is a summary version of the
familiar 7:6:6:6:6 sequence. Inevitably, though, this interpretation is
coloured by knowledge of comparable retrograde fragments in the
pitch line at this point.
Rhythmic commentary in Sinioni"
In the cyclic repetitions leading up to the climax, only six units
within the rhythmic set are modified in the 5infonia version (see
ex. lOa). But inasmuch as they are designed to alter the patterns of
pitch-rhythm interaction discussed above, they consistute a first
level of commentary. The first two groups of alterations are both
designed to slow down the levelling out of rhythmic ambiguities
discussed above. In Alii, the substitution of 3 for 4 ensures that the
second metric group will not achieve coherence until statement B,
and that the point of maximum harmonic ambiguity in the pitch set
is matched by equal rhythmic ambiguity. However, the substitution
of 8 for 6 asserts the presence of 414 amid these additive dislocations.
This will proliferate into an alternation of 3/4 and 414 in statements B
and C, before finally resolving into a predominant 3/4 in D. The
substitution of 4:6 for 6:8 in statement B mitigates the increasing
emphatic chain of 6s, and at the same time provides a summary
retrograde of the 3:2:1:4 that follows, thereby reinforcing, by more
extensive preparation, the upbeat pattern in that sequence, with its
important harmonic implications discussed above. The other group
of modifications, in Diii, is exclusively concerned with emphasising
pre-established harmonic polarities: the 12 on the c # firmly resolving
any suggestion of ambiguity in the pitch set, and the final b ~ being
raised to 18 by way of compensation.
The epilogue of the 5infonia version offers a substantially altered
sequence - 4:6:8:1V3:Yl3:8:7:6. The 4:6:8 could be construed as an
extension of the same sequence in the chamber version;9 the
Il!J:Yl3 retrogrades the configuration from Dii which, in its
expanded form at Eii, ushers in the climax; and 8:7:6 simply
recapitulates the opening sequence.
Parallel processes in rhythm and pitch
Thus within the two fixed systems of this piece analogous processes
operate. Both derive their dynamism from internal oppositions
either between levels of structure (pitch set vs. macro-set, quaver vs.
crotchet additive unit), or between different types of organisation
(additive vs. metric, triple vs. quadruple). Both are obliged to break
through to a climax once these oppositions are played out, lest they
should lapse into a monolithic and lOgically unending gyre of
• A relationship clarified in the Sintoni. version: see below_
• Save that in the Sbl/oni. version the third unit should no longer be 8 but 6_

30
 '0 King'

continuous sforzando statements of the pitch set in 3/4. In each case

the moment at which oppositions become reconciled is instantly
countered by a flouting of the system (the pitch macro-structure in
Diii, the rhythmic structure in Ei) and a build-up towards a climax.
Once the climax is spent, predominantly retrograded fragments
from both series provide an epilogue.
Selective resonance
The interactions of pitch and rhythmic sets produce a single, highly
structured melodic line which is subjected to comparatively little
modification as it is transferred from the chamber version to the
Sinfonia version. But within each note is heard a series of echo
attacks from the sustaining instruments and, in Sinfonia, the other
high voices. Sometimes these anticipate the pitches of the main line;
almost invariably they hold each note of that line over into the next
- either to create a momentary harmonic blur, or else to sustain
that note through one or several subsequent notes. The aggre-
gate rhythm of these attacks lO does not appear to be subject to a
system, II but the harmonic process that they create is the major
generator of large-scale form in both versions. It consists, quite
simply, in a gradual increase of harmonic density achieved by a
progress from momentary blurs as the general norm to sustained
notes as the general norm. The process is more rapid and more
marked in the Sinfonia version (see below, p.34).
In the early stages of this process, and before the harmonic field
has been complicated by the frequent incursion of alien pitches, the
result of selective resonance is to focus the ear on the alternation
between whole-tone groupings discussed above. However, on
occasion Berio enriches his resources by holding the outer note from
each group over into the opposite group. This gesture is particularly
characteristic of statement B of the pitch cycle in the chamber ver-
sion - though the use of an f pedal is extended through much of the
subsequent material in Sinfonia - and in each case it produces a
fleeting tonal reference. An f pedal sounded below the a ~ b ~ d group
evokes the second inversion of a dominant seventh chord on b~, as
10 They are complicated further by internal rhythms within the main vocal line as it
changes vowel or introduces a jaw tremolo within the note. The vocal line also
deploys acciaccaturas, some of which effect temporary permutations of the pitch sel.
11 Serio himself thinks that he can recall some sort of systematic starting point for
their organisation, so others may well care to correct me. In the chamber version
there are, however, remnants of a system governing the number of notes from the
twenty-one pitch set played by each of the four sustaining instruments. In statement
A the flute plays eighteen, clarinet, violin and cello fourteen. In statement C all play
fourteen save the cello, which plays eleven. In statement D all play eleven save the
cello, which plays nine. This modest game with a number series serves to systematize
the in(-reased number of pitches held over to consolidate the central melody's
harmonic implications, as a result of which each instrument can play fewer notes of
the series. But it is quite lost sight of in statement B (flute fifteen, clarinet twelve,
violin sixteen, rello eleven), and in any case disappears within the more complex
circumstances of Sinfonia.

31
Playing on Words

in the final notes of Bill, reproduced in example lla in rough pro-

portional notation. Similarly, a d sustained above the opposite group
of I a b d gives a momentary hint of d minor - an impression re-
inforced when I is sustained throughout. Example llb outlines an
instance of this: Bi and the first three notes of Bti. In both these
examples the total harmonic impression is a slightly more complex
experience, due to the embroidery of extraneous pitches supplied by
the piano.
Ex. 11
a b

Independent melodic lines

All the features so far discussed either fall within, or are generated
by the central melodic line. But from time to time the flute, sup-
ported by the harp, introduces its own melodic fragments based on
the pitch repertoire set out in example 12. Clearly, this incorporates
the d-based whole-tone group, plus a complementary c and an alien
d that mirrors the c#-4 coupling in the pitch set. As might be
expected, the flute entries therefore tend to underline the appear-
ance of the d-group within the set - at first as a heterophonic
ornament to the vocal line, but thereafter as an independent voice.
Ex. 12

In statements A to C, it always appears in conjunction with the

d-group of section ii of the set; in statement A it also occurs both at
the start and the end of the d-group in section iii - while in state-
ment B it occurs only at the start, and in statement C only at the
end. In this last instance similar materials are heard for the first time
on the clarinet; and the flute in echoing it spills over into alien
territory: the I-group at the start of statement D. The flute now takes
over from the piano part the high g that is to dominate the music in-
creasingly in preparation for the climax. It pursues an independent
path until Diii when, abandoning its established pitch repertoire, it
follows the vocal line - ti10ugh breaking free from it to insist upon
the high g to which the voke, too, must eventually come.
As the flute begins to herald one feature of the climax at D, the
violas (cello in the chamber version) hint at another. Picking up the
flute's low c they establish another independent line which rises by
degrees to I, where it blends into the vocal line. No more is heard of
this until the long b~ that precedes the climax, when all the free
parts rise in a flurry of similar lines beneath the flute's held g,
saturating the chromatic space from the gft below.

32
 '0 King'

Pitches outside the set

In the chamber version of 0 King, the five notes excluded from the
set are assigned to the piano, save for the c incorporated into the
flute's pitch repertoire. Four of them, eb, g, c and e, are disposed
symmetrically around the gamut of the pitch set, with the j~ estab-
lishing a further downward extension, as shown on the left-hand
side of example 13. They remain in these positions throughout state-
ments A and B, but from statement C onwards begin to attain
greater mobility, as shown by the black note-heads in example 13.
However, a new octave placement for a pitch class does not pre-

statement C we find an alternation between e.

clude the recurrence of previous placements, so that throughout
and f ~ in both
extreme registers. Octave placements of eb, e and j~ become even
more fluid in statement D, though the f# in parentheses occurs only
in the SinjQnia version. However, c and g remain fixed throughout -
the latter in preparation for the climax, of which it is the focus.
Ex. 13

Ci en Oi Dii Diii G

Although initially employed as whole-tone complements to the

pitch set's two harmonic areas (so that c and e in Aii usher in the
first d-group, eb the second f-group, etc.), these pitches soon com-
bine freely, forming a network of extraneous pitches around the
harmonic core of the pitch set. But in Sinjonia Berio assigns a more
complex role to the lower pitches, initially c, e and f#, by having
them sung or played by sustaining instruments, so that they interact
with the harmonic projections of the pitch set. Since the conse-
quences of this depend in part upon modifications in the resonance
system, the two are best considered together in the next section.
Immediately after the climax, as all voices converge upon the high
g, a new independent bass line emerges whose pitches - c b. ab
repeated, and then cU b a repeated - underline the opposition
between whole-tone areas that has lain at the heart of the piece
(resolving it, by interaction with the vocal line, in favour of the j
whole-tone scale). Above this, the chamber version rearranges the
alien pitches in a series of chords, shown on the right-hand side of
example 13, that finally settle into an 'inversion' of the original
chord around its central c. In the Sinfonia version these are thickened
into twelve-note chords reminiscent of those of the first movement
in interval distribution, and hinting at the massive, saturated blocks
that are to play so formidable a part in the third movement.

33
Playing on Words

Harmonic commentary upon the chamber version in Sinlani.

The greater harmonic density of the Sinfonia version of 0 King has
already been noted. This is in part due to the more persistent use of
pedals. Thus, whereas in the cham!'er version an f pedal does not
become an important factor until Biii, and retires into relative in-
significance from Ciii on, in Sinfonia it is a persistent presence from
the start of statement B through to the climax. Furthermore, in both
Dii and Diii it is expanded to produce a recurrent harmonic core,
f ab bl>, around which the dense and constantly shifting harmony
can revolve. Since the earlier whole-tone and quasi-tonal impli-
cations are by now completely obliterated, these recurrent pedals act
as an important stabilising influence.
But from the start the Sinfonia version has a richer harmonic tex-
ture due to the presence of sustained pitches from outside the set.
The recurring presence of c and e in the low strings and basses is a
crucial factor in this respect. Their first two appearances, in Aii and
Aiii, merely give added resonance to the original piano part. But
when they are introduced beneath the sustained f and d that initiate
both Bi and Bii, they abolish the bare simplicity of the chamber
version, as they do when added, along with a low f#, to the sus-
tained f and c # of Bii.
This corruption of simple consonance continues throughout state-
ment C. Thus the a of Ci whose consonance with a sustained f was
originally troubled only by a quiet resonance of e and f# from the
piano now has that resonance amplified; and a sustained eb from
double basses ensures that overtones will cloud the clarity of the f-d
and f-{l relationships that start Bii. A tritone relationship is likewise
invaded with the final ab of Ciii ... originally sounding only with a
sustained d, but now with the familiar c e and f# to cloud matters in
the bass. The pitches c and e return in the latter part of Diii to
further densen a texture already far richer than the original because
of the sustained f, a I> and bb discussed above.
By thickening out the harmonies in this way, Berio establishes a
greater measure of harmonic identity between '0 King', and the
first, fourth and fifth movements of Sinfonia ... particularly the fourth,
where chords of alternating thirds and seconds over wider-spaced
bass notes are a fundamental feature. (Compare, for instance, Aii in
'0 King' with the final chords of the first two progressions in the
fourth mowment.) But he also establishes an important harmonic
progression from the deliberately simple, third-based chords of the
first movement, through the more sophisticated harmonic flux
around a single melodic line of the second movement to the ex-
tremely rich harmonic vocabulary of the third movement.
The text and its structural potential
Although phonetic materials derived from the text were used in the
first movement to help maintain a precarious balance between
sound and sense, only the simplest of structural relations were

34
 '0 King'

established between them (the front-back opposition of 'tktktktk' as

the voices disappear, for instance). But here the gradual constitution
of '0 Martin Luther King' from its phonetic constituents demands a
more analytical approach. Berio is much helped in this by the
structured nature of the International Phonetic Alphabet, which
offers a system based upon the changing positions of the mobile
speech organs. The very elegance and concision of this 'postural'
typology, which was synthesized from the pre-war work of Daniel
Jones and other members of the International Phonetic Association,
have made it an object of some controversy amongst phoneticians
since its publication in 1949. 12 But it is precisely these qualities that
make the system useful and stimulating for a musician. For instance,
the alphabet distinguishes a set of 'cardinal vowels' governed by
two factors: where in the mouth the vowel is sounded (Le. front vs.
back), and whether the mouth is opened or partly closed. These are
combined to produce the matrix set out in example 14.13 Similar
oppositions can be used to classify consonants, though here three
factors must be considered: (a) front vs. back (upon which the I.P.A.
bases its own classification to the exclusion of other factors I4);
(b) total stoppage of the airflow, followed by release (plosives) vs.
unhindered airflow (nasals and laterals), with the vibrating rolls and
fricatives occupying a mid-point; (c) voiced vs. unvoiced.
Ex. 14 Front
BackBack
[u) Back
[u)
[u)
[u)
[u) [u)
[u) [u)
[u) [u)
Front Back
If we now examine the phonetic constituents of '0 Martin Luther
King' in the light of these distinctions, their structural potential
becomes obvious. The vowels (0) [a) [i) [u) (3)15 [i) can be arranged
so as to form a circular sequence around the mouth from front
closed, through front open and back closed - [i) (3) [a) (0) [u) [i) -
and it is thus that they are first presented. This anticlockwise motion
around example 14 is then maintained by permutating the sequence
into two triangular sequences: (0) [i) (3) and [a) [u) [i). The purpose
of this begins to become clear in the next sequence, where a clock-
12 See, for instance, John Laver's critique in Lyons ed. 1970, pp.S!;"7.
13 Reproduced from International Phonetic Association 1949, p.6. The peculiar
shape of this figure is due to the maximal front-back opposition obtained with a
closed mouth.
14 See International Phonetic Association 1949, p.lO.
15 Berio's 131 equals the lEI of the International Phonetic Association and represents
the vowel of Ither/.

35
Playing on Words

wise triangle [0] [a] til is juxtaposed with an anticlockwise triangle

[3] [u] Ii]. The first of these presents the vowels as in the text; and
accordingly Berio produces a fourth sequence which permutates the
second triangle so as to produce the complete correct order. These
games are summarized in example 15. 16
Ex. 15
i3aoui o;3aui oai3ui oaiu3;

The correct order of vowels having been established, Berio starts

to add consonants. These too are arranged according to their pho-
netic characteristics. Of the six consonants used structurally, three
are voiced,17 and three unvoiced. Of the three voiced consonants
two are nasal (the velar [a) that ends 'King' and the bilabial [m)) and
one lateral (the alveolar [I]). Of the unvoiced consonants two are
plosive (the velar [k) and the dental It)) and one fricative (the dental
[9) representing the ith/ of 'Luther'). The two voiced nasals and the
two unvoiced plosives each establish a front-back opposition,
though it is only the former, between opening and closing conso-
nants, that is put to structural use.
The gradual addition of consonants proceeds from the vowel-like
voiced consonants to the more disruptive, unvoiced ones. Statement
A of the pitch series is taken up entirely with the ordering of
vowels. Statement B introduces the nasal opposites [a) and [m], and
these are joined in statement C by [I]. From statement 0, however,
the two versions part company. The chamber version continues with
vowels and voiced consonants, witholding the first unvoiced conso-
nant, [k), until just before E. It then adds [9] at the climax, and
allows [t) only a single appearance in the fuJI version of the text
sung in the epilogue. Both voiced and unvoiced consonant pairs are
thus introduced in the order from back to front. In Sinjot/in, on the
other hand, it is ttl that is introduced first, and it appears soon after
the start of statement D. [9) is introduced in the middle of 0, and
[k] at the same place as in the chamber version. The order of intro-
duction of the unvoiced consonants is thus reversed.
16 Games with vowel rotation, normany triangular, can also be found in Sequenza
Ill. They bring into play a set of 'secondary cardinal vowels' for \vhich a diagram
analogous to example 14 can be found on p.6 of International Phonetic Association
1949. Comparison between those diagrams and the first sy:ih.~m of p.2 in the score will
reveal a fairly complex example (provided that the vowels from word fragments are
phonetically transcribed).
17 A fourth voiced consonant. the nasal alveolar InJ. occurs in the text; but it is too
close to [r)1 to be used distinctively in Serio's structural games, It therefore appears
only when h. introduces the (t) with which it is associated. It is only in the Sintonia
version that a distinction is made between (mal and 'Mar' after D, thereby
acknowledging the sounded Irl of American pronunciation.

36
 '0 King'

Permutation and troping

As far as the chamber version is concerned, the pattern that results
from this gradual accumulation of phonetic constituents is a rela-
tively simple one, varied only by the alternation between adjacent
vowels or syllables of the text. It is set out in example 16, where
each element that cannot be accommodated on the same line starts a
new one. IS What first strikes the eye is Berio's opening sequence of
permutations which are as coherent from the point of view of text
order as they are from that of vowel rotation. Initially the vowel
chain forms a pattern of one step back (noted by II' ), three steps
forward. 19 The next group of six then forms two chains of two steps
forward (noted by-). The third group of six combines~and II' to
permutate its second half.
Once the correct vowel sequence has been established, the per-
mutations that follow are dictated by two considerations. In the first
place there would appear to be the relics, only partly obliterated, of
a simple system determining the number of vowels or syllables to be
sung within each statement of the pitch set. In statement A there are
twenty-seven, in statement B eighteen, in statement C nine/a
in statement 0 seventeen (presumably eighteen in the original
groundplan), and from E to the climax nine. This would give an
underlying scheme of 9x9 (and indeed, if one includes the other-
wise mysterious [01 of statement C, the overall number is still
adhered to). In addition, the permutations serve to build towards
the introduction of new consonants in all cases save Ither/. As might
be expected, Berio selects for his two principal alternations the
strongest opposites contained within the normal order of the text:
the front-back closed 'i/u', and the even stronger back open/front
closed 'ali'.
The thinning-out of verbal events in the middle of the chamber
version produced extended melismas; and these offered Berio the
opportunity for elaboration when he came to write the Sinfonia
version. He responded with phonetic tropes. Initially these reflect
the systematic permutations of the opening, so that after the
complete text sequence at the start of statement C (complicated only
by the insertion of an extra [uD phonemes jump backwards by twos
- '[~llu [al' - and then by ones - '[~I [31Iu'. But they become rather
more maverick by statement 0 - as well they might, for during this
statement the other seven voices, which up to now have echoed the
first soprano's materials faithfully, become infected by the general
move towards proliferation, and proceed to trope her line. The
resultant phonetic melee produces a richer, more disordered verbal
texture that parallels the steady increase in harmonic denSity.

" This method of exposition is borrowed from Ruwet 1972, pp.100ff.

19 Provided, that is, that 'the ambiguous (i]s are resolved as shown.
20 If the (0] straddling statements Band C is counted as part of B, and the (,,] that
occurs after the next (0] is taken simply as an inflection of that vowel.

37
 /Tl
~
~
'"
....
Ex. 16

.~
bI)
'" 0
c
0 0 0 0
a
0 0 0 0 0 0 0 0 0 0 0 0 0
::
I"l "C
...'"
to to '" '" c s::
Mar

s:: 3 3 3 3 3to 3 3 e:
~ ..3 ~ to
~ '" '" '" '" "'
c c c c c c c c c c c c c c c c c
tin

5·
5
'"
~
r r ;: c c c c c c c c c r
Lu

c c C C C C c

c ~
c c c c c c
ther

~ ~ c c c c ~
~ ~

7':
~ T. 7': .:> .:> .:> .:> c c c c 5
King

2:' ;:; O<l

:l 5
O<l O<l O<l

/Tl 0
 4
'In ruhig fliessender
Bewegung'

The second movement of Sintonia, like the Chemins, had been the
result of reopening a set of creative questions that were temporarily
closed. But in the third movement Berio undertook a more daring
and problematic project: that of building fresh layers of material not
out of the residue of his own past compositional decisions, but
around a work by another composer. He had long admired and
studied Mahler's music, finding in its vivid but ironic eclecticism a
congenial example for his own work; and his choice accordingly fell
on the scherzo from Mahler's Second Symphony.' But its diatonic
language posed a complex technical problem. For if Berio had
sought to generate layers of commentary from the Mahler text itself
he would have had to subject it to extensive transformation. If
instead he had relied entirely upon his own harmonic vocabulary
the gap between text and commentary would have been too great.
So he opted for materials that establish a wide harmonic range -
many of them quotations from other composers' work. Thus against
Mahler's predominant diatonicism are set the more sumptuous
harmonies of Ravel, Strauss and Debussy, the atonality of the
second Viennese school, and massive, chromatically saturated
orchestral clusters. 2 Merely to superpose these leaving the original
intact, as in the previous movement, would clearly make for a
cloying density of texture - as well as demanding gargantuan forces.
Berio therefore blocked out ever greater amounts of the Original
material, at first so as to provide room for the various commentary
materials, but later as an autonomous process that leaves only a
skeleton of Mahlerian fragments. It is this incremental obliteration
that provides the large-scale shape of the movement.

1 Previously Berio had considered using the second movement of the symphony
and even the final three movements of Beethoven's Op.131 as vehicles for a similar
exercise. See Berio 1985, pp.107-8.
2 The actual choice of materials was in part a matter of circumstance. Berio wrote
the movement while on holiday in Sicily, and therefore relied upon the few scores
that he had with him, those that happened to be available from Catania public library,
and his own memory in order to establish a suitable range.

39
Playing on Words

From sennon to scherzo

The movement from Mahler's Second Symphony chosen by Berio as
a vehicle for this experiment was in several respects a peculiarly
suitable starting-point. In the first place it, too, was the product of
elaboration from a pre-established musical text. Although Mahler's
song Des Antonius von Padua Fischpredigt and its expansion into the
scherzo of the Second Symphony both reached completion at
roughly the same time in the summer of 1893, 3 it is clear from
Mahler's comments to Natalie Bauer-Lechner concerning the genesis
of the scherzo that it must have been preceded by a more or less
complete draft of the song. 4 The song in question sets an ironic text
from the Des Knaben Wunderhorn, a collection of 'old German songs',
compiled by Achim von Arnim and Clemens Brentano and pub-
lished in two volumes of 1806 and 1808 with a dedication to Goethe.
It narrates how, for lack of a congregation, 5t Anthony goes to
preach to the fishes who are listed - each with their more or less
venial characteristics - as they rise to the surface to listen in ad-
miration. But, after enjoying a moment's edification, each returns
refreshed to its favourite indulgence. The association between this
song-text and the music of the scherzo gives rise to a number of
semantic games within Berio's movement, to be discussed below
(pp.53ff). But for the moment our concern must be with the musical
form that Mahler moulds around it. In the song this is a simple
ternary structure, with a first section consistently focused upon c
minor, answered by c major, a middle section that similarly never
moves far from the ambit of f major, and finally a reworking of the
first section, incorporating somewhat more devious tonal sallies as
the fish disperse to their sinful pursuits.
The elaboration of so concise a structure into an orchestral scherzo
involves considerable expansion. The correlation is direct until the
end of the middle section of the song, 5 which forms a first 'trio'
within this orchestral context. Materials from the final section of the
song are incorporated into the second reprise6 and at the end,7 but
otherwise the orchestral version introduces new materials or re-
works previously stated ones. Thus the first trio in f is followed by a
first reprise of the scherzo in the tonic, allowing a slightly richer
range of temporary modulations (b. 149--89). There then follows a
substantial new section - to be referred to, slightly incongruously in
view of its emphatic character, as Trio II (b. 190-347). This drama-
tizes the rather bland contrast previously established between c and
, See Mitchell 1975. pp.137-8.
4 Indeed, Mahler's observation that 'Without knowing at first where it's leading.
you find yourself pushed further and further beyond the bounds of the original form.
whose potentialities lay hidden within it like the plant within the seed' (Bauer-
Lechner 1980, p.32) would, but for its implications of diachronic expansion, provide
an excellent summary of Berio's approach to musical commentary.
5 b.2-133 of the song - b.6-\37 of the scherzo.
• b.14()"'58 of the song. transposed up a fourth - b.387-406 of the scherzo.
7 b.171-<md of the song - b.55Hnd of the scherzo.

40
 'In ruhig flies sender Bewegung'

{by brusque upward steps from c major through d major to e major.

A return to c major leads back into a second reprise of the scherzo
materials (b.34S--406). This is followed by a combined reprise of
materials from Trio I (b.407--40) and Trio" (b.441-543), with the
latter moving to a massive climax represented by a chord of b. minor
over a c pedal (b.465) whose programmatic significance will be
discussed on p.54. There follows the final reprise of the scherzo
materials, blending back into the final bars of the song, as noted
above. s
It should however be emphasised that these relationships are
confined purely to matters of pitch and rhythm. Since the first
orchestral draft of the scherzo was completed only eight days after
the piano score of the song,9 and since the 'meticulous' instrumen-
tal indications of the latter were fleshed out into an orchestral score
of the song three weeks later,1O there is clearly no justification for
regarding the integrated, post-Wagnerian orchestral palette of the
scherzo as being a commentary or elaboration upon the simpler,
'choral' use of chamber orchestra resources in the song. On the
contrary, the two versions would appear to offer a remarkable ex-
ample of closely related melodic, harmonic and rhythmic materials
projected, more or less concurrently, into two contrasting sound-
worlds. 11
Texts for a sermon
But Mahler's scherzo provides a further and rather more curious
precedent for Berio's experiment. The idea of using materials from
other composers' works as a means of blocking out the Mahler text
is complemented by Mahler's own use - conscious or unconscious-
of alien materials in his Fischpredigt. These analogies are set out in
example 17. The parallel between b.44-52 of the Mahler and the trio
from Bruckner's Fourth Symphony is fairly straightforward, save
that Mahler does not employ Bruckner's surprise modulation for the
last two bars of the passage and he resolves the dominant har-
monies of the third bar onto tonic harmony in the fourth. However,
the similarity to Beethoven's trio does not end merely with the
thematic material,12 for the overall structure of the central section
of the Fischpredigt (and therefore of Trio I of the scherzo) also shows
close affinities with Beethoven's model. In both cases the eight-
bar phrase quoted is repeated once, with some textural variation,
followed by two two-bar phrases that develop their respective
• For a a more thorough discussion of relations between song and scherzo, see
Tibbe 1971, pp.5G-58.
• On 16th and 8th July 1893 respectively; see Mitchell 1975, pp.137-8.
10 Finished on 1st August 1893; see Mitchell, Joe cil.
11 A simple but extended example may be had by comparing bAG-59 of the song
with b.44-63 of the scherzo; the interested reader will find many other more detailed
contrasts elsewhere in the two scores.
12 Though this, too, is reworked by altering the final modulation and by converting
Beethoven's dominant seventh in the third bar to a subdominant.

41
Playing on Words

Ex. l'

Bnacbtr~ Fourth SJlDPlIoDy, Trio. aod Mahler b.....·s:

a A
r;:
0

;;
a a
;;
0

BcctbDvm, Violin Sonata Op. 96. Trio.1Dd Mah1er b.IQ4-12

;;
~

S..humaoo. Das isr ,i" Fl8een Jmd GG~m, Dichtcrliebe. No. 9. JiDal ban. and Mahler final ban
.

;;
~

;;
~
Jl

;;

;;
;;
a
.!!

opening gestures to provide some tonal and harmonic variety,

followed by a reprise of the original phrase which is, however,
extended - in the Beethoven by extensive canonic repetitions based
on the first half of the phrase, in the Mahler by developing a repe-
tition of the second half of the phrase. The allusion to the final
gesture from Schumann's 'Das ist ein Floten und Geigen' at the end
of both song and scherzo is straightforward from a musical point of
view. Its possible programmatic significance will be discussed below
on p.54. 13
13 The Schumann quotation is noted and discussed in Tibbe 1971. pp.5S-9, where
she, too, underlines its congruence with the programme of the symphonic scherzo.
The Bruckner quotation is noted in Altmann 1977, pp.I9-20, though without
discussion of its significance. Neither of these authors discusses the Beethoven
quotation.

42
 'In ruhig fliessender Bewegung'

Whether or not these should be regarded as conscious quotations

is difficult to determine. Despite Mahler's notorious delight in
incongruous stylistic juxtapositions, there would appear to be little
evidence in the current literature for the deliberate use of quotation
elsewhere in his works. And his allegiance to a largely diatonic
idiom inevitably encouraged parallels with the earlier nineteenth-
century repertoire - though parallels as substantial as those dis-
cussed above can hardly be attributed to chance. Furthermore,
Mahler was not unduly reticent in commenting on his own work,
and while he does indeed acknowledge the influence of popular
music from his Bohemian childhood in the Fischpredigt,14 he makes
no direct acknowledgement of these other influences.
On the other hand, since Mahler was well acquainted with all
three works in question it seems unlikely that the parallels should
have entirely escaped him. Indeed, there is some flimsy but tempt-
ing evidence to support the view that they were entirely deliberate.
In another discussion of the Fischpredigt with Natalie Bauer-Lechner,
Mahler is reported as saying: 'St Anthony preaches to the fishes;
his words are immediately translated into their thoroughly tipsy-
sounding language (in the clarinet), .15 The clarinet entry in question
could be either that at b.16 of the song (b.20 of the scherzo) or that
following the Bruckner quotation at b.48 of the song (b.52 of the
scherzo). If the latter is indeed the case, then we have a delightfully
wicked parallel between St Anthony and the similarly naive but
devout Anton Bruckner. Further hints at hidden meaning are prof-
fered as Mahler describes the recalcitrant fishes returning to their
indulgences: 'Not one of them is one iota the wiser for it, even
though the Saint has performed for them! But only a few people will
understand my satire on mankind: 16 Could that satire include the
deliberate use of 'the tootling of the Bohemian pipers,17 as a counter-
subject to Beethoven? And might not a parable on that permanent
Mahlerian obsession, the philistinism of his bourgeois public, well
end with the final bars of a song whose last two lines run 'da-
zwischen schluchzen und stiihnen die Iieblichen EngeIein,?18
Obliteration as form
Mahler's groundplan infuses the balanced, sectional design of a
scherzo and two trios with a measure of dynamic unity by allowing
the second of the trios such comparatively weighty materials that,
when recapitulated along with those of Trio I, they carry the move-
ment through to a unifying dramatic climax. Berio imposes upon
this succession of discrete sections a single process of incremental
obliteration that has already achieved fairly drastic form by Trio II -
14 Bauer-Lechner 1980, p.33.
15 Bauer-Lechner 1980, p.32.
16 Bauer-Lechner 1980, p.33.
17 Ibid.
18 •Amidst all this, the lovely angels sob and sigh.'

43
Playing on Words

so that thereafter the continued presence of the Mahler is principally

established by the sudden re-emergence of such major rhetorical
gestures as the transitions between sections, the climax and the final
bars. This process is summarized in example 18, which distinguishes
five different levels of 'presence' of the original text (whose formal
outlines are summarized beneath):
(i) conservation of the melodic and harmonic substance of Mah-
ler's scherzo;
(ii) conservation of single - usually melodic - lines only;
(iii) conservation of fragments of Mahler's text, which appear in
the same position relative to the metric framework that they occu-
pied in the original. When that is not the case, as between Rand S,
U and V, or Wand X, this is shown by a line oscillating between
this level and level (v);
(iv) obliteration of the Mahler text while retaining its metric
framework, so that when it does reappear, the same number of bars
have elapsed as would have passed in the original;
(v) obliteration of text and metric framework. Although this
occurs frequently from P onwards, the addition and subtraction of
bars balance each other to a reasonable degree - with the result that
Berio's movement, from the point at which it engages with the
Mahler scherzo (b.7) to the end, lasts only eighteen bars more than
its model.
(0) indicates distortions imposed upon the Mahler without inter-
rupting its continuity, to be discussed below on p.47.
Above the graph are summarized the major features used to block
out Mahler's original: chromatically saturated clusters (Ci), chro-
matic scales (Sc), quotation (Qu) and reduction of the Mahler to a
few skeletal features (R) - when this process is used autonomously,
and not merely to make space for one of the other three. Quotations
are noted in this outline only where they completely block out
Mahler's text: the many instances of dialogue between quotations
and Mahler will be discussed below (p.48), and analysed instance by
instance on pp.57ff. Although before H superposed clusters and
scales perform a similar function of dialogue with Mahler's text,
they have nonetheless been noted in brackets so as to show their
localised occurrence.
On the whole, the application of these devices is regulated by
individual features of the Mahler text rather than by some auto-
nomous, self-regulating system. But it will be clear from example 18
that their distribution also helps to shape the form of the movement.
Apart from a brief appearance at T as the quotation from Wozzeck
escalates in density, the massive orchestral clusters are lIsed ex-
clusively to block out Mahler's upward tonal steps in Trio II and
in the reprise of the same material at Y. Reduction gradually in-
creases in importance, so that from the trio reprise onwards its
gaunt textures establish a norm that contrasts sharply with the
hectic abundance of earlier sections. Chromatic scales are at best

44
 A r n F G H K K M N 0 Q R
Ex. 18
Blocking Agents (el) (Sc) QI,.I(Scl Qu I(Sc) Qu Ct el Qu Qu R Qu

v Metric Dist. +n +22

ov Metre

iii Fragments

Melody
" (DJ

" Mel . & Harm . (D ID)-

Mahler Tn(ll RE'pri se I Trio II

Tonal Structure < e (D (D e D (DJ

~
T T V W Y AA BB ee DO FF

Blocki ng Agents Qu el Qu W CI Qu W Qu R
s-
a
::r
-9 +2 +1 - 15 - 10 dQ.
; Metric Dist.
i:!l
;v Metre

Iii Fragments

Iii Melody

Iii Mel. &: Harm.

MahleJ Reprise 11 Trio reprise Repnse !II

Ir
Tonal StructurE Iii C C C ],
Playing on Words

only a comparatively minor resource, but their disappearance after

the first reprise, recurring only as an integral part of the Wozzeck
quotation at S, serves as a modest structural marker.
Although Berio has imposed a linear if vacillating process of ob-
literation upon an originally sectional conception, he does respect
the rough contours of Mahler's dramatic structure. He thus con-
serves - indeed amplifies - the first part of Trio II as a high-point of
tension, preparing it with a lengthy quotation from Stravinsky's Le
Sacre du printemps before I, and allowing its energy to reverberate
through to the combined Ravel-Strauss quotation at O. Although a
comparatively contained dramatic crisis is inserted at T, prepared by
the Wozzeck quotation, Berio otherwise follows Mahler in restraining
his forces until the reprise of the Trio II material. Here, however,
metric distortion begins to play an important role in reworking the
dramatic structure, for whereas the outburst at Y should, according
to the Mahlerian model, lead straight into the climax (d. Mahler,
b.44lff), Berio inserts a sudden quiet flashback to previously heard
materials from La Mer. Similarly after the climax, where Mahler
allows his nightmare vision to continue with quiet but undimmed
energy for a considerable period, Berio cuts the overall length to
compensate for his earlier insertion, and collapses the music im-
mediately into a bleak, static texture whose suppressed tensions are
only occasionally allowed to erupt.
Reduction and distortion
Reduction of Mahler's text to a few skeletal features is only slowly
allowed to emerge as an autonomous feature. In a passage such as
E12-F9 the snatches of Mahler have no perceptual priority over the
various other quotation fragments and chromatic scales. By 17-K
and L4-M, although fragments from other sources are still the
dominant feature, the paring down of Mahler's texture has begun to
assert itself more prominently. Particularly characteristic is the
retaining of flute and piccolo high cs as a fixed point of reference
against which a variety of fragments are projected. At Q the process
finally moves to the foreground, isolating the oscillating trumpets of
Mahler b.320 from their context, and using them to set in relief frag-
ments of Mahlerian melody. The device is expanded upon at V6,
where an ostinato is built up from a combination of Mahler b.433-4
(horns) and b.407 (cello) along with an oscillating woodwind figure
that is a legacy from previous pages, all serving as a background to
distorted fragments of Mahler and Hindemith. At CC (Mahler b.481)
Berio amplifies upon his treatment of Mahler at 17, retaining the
upper woodwind pedal but resisting the opportunity that Mahler
offers for a further set of oscillations. In the final example at FF, it
is only the grelots' ostinato that provides a fixed point of reference
for the melodic fragments - though that too disintegrates as the
movement closes.
Just as Mahler's text is constantly subject to some measure of re-

46
 'In ruhig flies sender Bewegung'

duction, so too his instrumentation is frequently modified. In part

this is merely a further consequence of the need to make room for
other events, but it is also used to break up the even flow of Mah-
ler's lines. This may be achieved by the subtle changes of timbre and
acoustic direction that come from tossing a melody from one string
group to another as at A23,19 or by etching in fragments of a line in
another part as at N (Mahler b.272), where the original second violin
countermelody is tossed from instrument to instrument through an
enormously enriched string texture, doubled fragmentarily by saxo-
phone and voices. Only occasionally does a radical rescoring come
to the forefront of aural attention, but L4-M (Mahler b.235-57) is
a remarkably sustained example of the device. Details of this and
the many other minor examples will be found in the commentary
(pp.57ff).
There is one other, very much more localized transformation to
which Mahler's text is subjected: that of melodic distortion. Three
different processes are used. The first is the expansion of a melodic
line into a heterophony by simultaneous rhythmic distortion. This
appears briefly at D4, but is more expansively treated from H
onwards, where its scope is widened by instructing the singers to
imitate - presumably in delayed and distorted form - given instru-
mental lines. In contrast to the blurry conservation of melodic
shapes ensured by this device, the wide-ranging octave displace-
ment that briefly follows upon its first appearance at D9 offers a
capital opportunity for disguise. This is exploited to the full by piano
and solo violin from Won. The developed use of heterophony at H
also explores a mode of distortion peculiar to the keyboards: that of
apparently random octave doublings. Here they are mainly confined
to the right hands, while the left juggle with the bass part from the
Mahler abandoned by the cellos. But the device recurs in purer form
in the reorchestrated passage at L, with each keyboard elaborating
upon one melodic line.
Quotation
The harmonic gamut established by the Mahler in its original and
distorted forms is substantially amplified by the overlaid quota-
tions. The majority of these come from works that were composed
within the three decades following the appearance of the Second
Symphony and cloud Mahler's fundamentally diatonic idiom with
increasingly ambiguous harmonic propositions. They are also pre-
dominantly drawn from composers who were, like Mahler, virtuoso
orchestrators: Debussy, Ravel, Strauss, Stravinsky and others. Thus
on one level this movement constitutes Berio's hom mage to past
masters of an art that he had himself so assiduously cultivated in the
195Os. There are backward looks to Berlioz - Mahler's great precur-
sor in the art of orchestration - and to the Germanic tradition of
'9 Mahler himself uses this device here and there, so that at FFS Berio is merely
expanding upon his example.

47
Playing on Words

Brahms, Beethoven and Bach, as well as forward to Webern and the

Darmstadt school.
The constant fluctuations in harmonic density that result from
using this range of reference do not obey any underlying large-
scale process of the sort that the vacillating but progressive oblitera-
tion of the Mahler produces. Instead, the distribution of these
materials is primarily determined by local details within the Mahler
text - and later, once that text has begun to disappear, by inter-
relations between the quotations themselves. However, a series of
groupings by genre is discernible as the movement progresses:
(i) The first is a group of solos from violin concertos, starting at
A19 with the Hindemith Kammermusik NrA and following this at CI0
with solos from the Berg and Brahms concertos.
(ii) A series of quotations from symphonic scherzos is anticipated
at C2 (the point where Mahler's scherzo itself begins to quote the
scherzo of Bruckner's Fourth Symphony) with a quote from Berlioz's
Symphonie fantastique. A while later, at E26, Berio extracts from
Ravel's La Valse a quotation - conscious or otherwise - from the
scherzo of Beethoven's Ninth Symphony, which is immediately
followed by a further quote from the Berlioz and one from the
scherzo of Mahler's Ninth Symphony.
(iii) Throughout this section there is also a wider dominance of
ballet scores, starting with Ravel's La Valse (from 010) and Daphnis et
Chloe (022), and progressing, after a return to La Valse at 017 to the
'Danse de la terre' from Stravinsky's La Sacre du printemps (from HIS
onwards) and Agon (16 and L8). La Valse then returns (N13 onwards)
as ballet gives way to opera.
(iv) The opera in question is Der Rosenkavalier, mingled with frag-
ments of La Valse after 0, and returning at P8. This is followed after
a quiet interval by Wozzcck at s.
(iv) Finally, the Darmstadt trinity of Webern, Boulez and Stock-
hausen are presented in close association from EE.
The links that are established between Mahler and the superposed
or interpolated fragments - and indeed between the fragments
themselves - are almost all founded upon one or more of the follow-
ing features:
(i) common pitches between counterpointed lines;
(ii) the sharing of a common harmonic basis, although in several
instances a melodic line is estranged from its original harmonic
context in order to combine plaUSibly with another whose harmonic
basis is preserved;
(iii) common melodic shapes.
Since numerous instances of all three processes will be examined
in the detailed commentary below, two examples will suffice to
illustrate the various possibilities: the quotations from Hindemith's
Kammermusik and from Berlioz's Symphonie faPltastique. The Hinde-

48
 'In ruhig fliessender Bewegung'

mith excerpt illustrates a frequent problem within this movement -

that of marrying heterogeneous styles. The passage in question con-
tains several fleeting tonal allusions, but these are far too brief to
allow any common harmonic movement. Berio therefore treats it as
a partly independent line which, although it contains melodic
shapes in common with the Mahler, normally uses only pitch con-
junctions as the means for establishing a relationship.
Examples of both occur when the solo line from the Hindemith is
first introduced at All. The initial fragment contains shapes that

the text at A13 relies upon the common f and a.

also occur at the beginning and end of the Mahler phrase, whereas
to establish a
relationship, as does that at Al6 with its common c~. All three are
shown in example 19a. When the full violin solo emerges at A19,
it pursues an independent path until the two converge upon a
roughly compatible counterpoint just before B. Similar principles of
partial independence govern the rest of the Hindemith quotations,
the Berg quotation at CIO, and several others.
The idee fixe from Berlioz's Symphonie fantastique provides clearer
illustration of the other two modes of relationship. It is first in-
troduced at C to provide an augmented version, transposed up a
fourth, of Mahler's answer to Bruckner: see example 19b. But when
Ex. 19
b A4 A 11 Hindt:mith

Ex. 19

A16

b
Berlioz C2

Mahler

b E12

49
Playing on Words

it returns at E12 it instead provides a compatible counterpoint to

Trio I, albeit at the cost of its own harmonic identity, as in example
19c. (An even more striking example of original harmonic identity
negated occurs when the trumpet solo from LA Valse is set against
that of the Mahler at P.)
Similar principles are used to link together chains of quotations
once the Mahler text has gone into temporary eclipse. Two of the
more striking instances are the use of common upbeat patterns to
consummate the garish union of LA Valse and Der Rosenkavalier
between N20 and P14, and the analogy established between Bach's
first Brandenburg Concerto and Schoenberg's Op.16 no.3 from Q19
to R22.
Chromatic dusters
At one extreme of the harmonic continuum explOited in this move-
ment stands chromatic saturation, both partial (as in the use of small
clusters and scales) and total. The clusters appear in three different
sections. They are used from the start until just after D to 'muddy'
the tonal clarity of the Mahler but never to obliterate the outlines,
even when the texture becomes comparatively thick, as at C. Then
they are left in abeyance until Trio II, where they come into their
own, entirely obliterating each of the major tonal shifts (from c to d
and from d to e) that characterize that section. From N on, they once
again disappear, save for a brief moment during the second reprise,
when another cluster formation crowns the quotation from Wozzeck
at T. But, appropriately, when the Trio II materials are recapitulated
at Y the cluster that first appeared at K returns. Other chords -
usually not saturated clusters - are used as rapid punctuating agents
from the beginning to B, at R, and in the Boulez quote at EE. In the
main autonomous events, ungoverned by any englobing process,
these will be discussed here only when they contribute to the pro-
cesses governing the sustained 'blocks'.
A summary of these materials appears in example 20 where black
note-heads indicate punctuating chords. It is a general feature of
these chords that while the clusters in the upper registers pursue
some sort of sequential logic - albeit often rudimentary - the non-
saturated, lower portions of each chord are a good deal more idio-
syncratic and only occasionally present themselves as parts of a
coherent process.
The initial group of cluster materials derives in part from external
sources. Thus the initial attack superimposes two clusters: the lower
saturating the gamut of the initial chord from the opening quotation
of Schoenberg's Op.16 no.4, and the upper covering the same
gamut minus a semitone with the major and minor scales of c - the
key of the Mahler. The saturated b--f # in the strings is similarly
derived from the quoted opening of Mahler's Fourth Symphony,
and appropriately subsides as the 'correct' Mahler movement is
established in its stead, landing on the filled c-g fifth. The next

50
 'In ruhig fliessender Bewegung'

Ex. 20
A Cl1 B C Cl1 DS-8

...
... K K19 L M M4-10 T T5 v. Z2

...
cluster attack reinforces this process by establishing the saturated
outlines of the second inversion of the tonic triad. The attack itself
also transposes the upper opening cluster attack down an octave,
extends it downwards by a tone, and saturates it - thus establishing
the e~ in the top stave as a fixed upper limit for this opening section.
At B the c-e~ cluster passes from strings to wind, thereby ushering
in the other main structural feature of these clusters apart from their
gamut. For while within string clusters integration of timbre is not a
particularly variable factor, clusters for wind or, as later, full
orchestra allow for contrast between the classical integration of dis-
parate timbres at one extreme and the juxtaposition of contrasting
timbre layers at the other. This first section limits itself to integrated
clusters, within which adjacent pitches are preferentially assigned to
instruments of different families - the only exception here being the
juxtaposition of clarinets 2 and 3. (The attack clusters at the start and
after A both allow for slightly less stringent integration, but their
sheer brevity makes it less than likely that the ear will distinguish
the difference.) Inevitably, certain limitations upon integrated
timbre-texture may be imposed by the ranges of the instruments
themselves, so that the answering cluster that establishes itself in
the extreme bass has to use a layer of low brass.

51
Playing on Words

At C the upper cluster extends downwards to b, while the lower

diversifies into the more widespread chordal writing that is to be
generally characteristic of the lower register from now on. The b in
turn briefly reinstates the b-f~ cluster from the opening (used here as
a sort of harmonic interpretation of the solo cadenza from the Berg
Violin Concerto that follows). The full orchestral cluster returns
briefly at C16, contracting its gamut by a semitone and offering a

briefly towards e.major after D, a non-saturated e.-b.

different but equally integrated orchestration. As the Mahler moves
cluster is
produced by flutes and trumpets, fulfilling the same function as its
predecessors; but thereafter the device remains in abeyance until its
full flowering in Trio II.
Whereas all the examples so far considered have simply been con-
tributions to a multifarious texture, the massive blocks introduced
from K onwards obliterate all else - and consequently have to
sustain musical interest from their own internal resources. The first
cluster, at K. establishes c as a fixed upper limit, and a massive
saturated gamut, broken only by two leaps of a tone. The colonising
of such high registers necessitates the use of a layer of high violins
immediately below the piccolo, but otherwise the texture is rigor-
ously integrated. This time, however, the integration is emphasised,
and aurally explored, by assi§ning to each of five timbre groupings
(flutes plus clarinets; 'oboes'2 and trumpets; horns; 'bassoons' and
trombones; strings) differing dynamic curves so that each constantly
eclipses the other (a device that had already been thoroughly ex-
plored in fpifanie f).21 The voices all start to oscillate around their
point of departure; and when they regain stability at K19, not only
have the tone gaps shifted position - the d.-b gap having descended
two octaves - but flute, string, and oboe layers, plus an extensive
layer of strings mixed with clarinets, have begun to emerge. When
the cluster is reiterated two bars later, at L, the transformation is
complete. Not only has the gamut of the cluster retracted, but its
upper voices are now systematically organised into timbre layers:
flutes; clarinets; oboes; strings; trumpets and strings, etc.
Much the same layering pattern is maintained when the next
large-scale cluster is unleashed at M, quickly growing from its
initially restricted formation, emphasising the b dominant of Mah-
ler's more or less annihilated e major, to a reasonably faithful semi-
tone-downward transposition of the cluster at L, varied only in its
lower reaches.
The brief re-emergence of another cluster at T, c,apping the pro-
liferating chromatics of the Berg quotation, shifts the upper note
down a further semitone, and slightly contracts the cluster's gamut.
But while maintaining the layering principle it shuffles the order,
20 Quotation marks here denote a whole timbre grouping - in this case oboes, cor
anglais and alto saxophone,
21 Though the suggestion in Hicks 1981-2 that these are 'quotations' from that piece
is incorrect.

52
 'In ruhig f1iessender Bewegung'

which now sounds from the top as: flutes; violins; clarinets; trum-
pets; 'oboes', etc.
Y recapitulates the cluster at K, although slightly modifying the
arrangement of pitches below middle c. It resolves at Z2 onto a
twelve-note chord whose upper nine pitches progressively contract
the intervals between them from major thirds to semitones - scored
first for full orchestra, and then for wind and strings in dynamic
opposition.

Semantic associations of Mahler's scherzo

So far, commentary has been discussed as a purely formal device in
the tradition of the Chemins. But with vocal materials and quotations
from theatrical or programmatic works all superposed upon a
musical 'text' that itself has a wealth of connotations attached to it,
Berio is able to accomplish a parallel semantic proliferation and, as a
result, to bring this movement closer than any other that he has
written to the labyrinthine 'stratification' of Joyce's Finnegans Wake.
Since Mahler's scherzo was originally associated with a song-text,
and he subsequently provided two partly discrepant programmatic
deScriptions for it, the range of images that may be brought into
play is considerable. It is therefore worth separating the various
strands:
Song text
(i) Imagery: the primary image that Berio borrows from the story
of St Anthony preaching to the fishes is that of water, which unites
his frequent quotations from Debussy's La Mer with a series of frag-
ments that occur between Rand Y. These are: Schoenberg's Op.16
no.3 which, though entitled 'Farben', was described by the com-
poser as being inspired by a lake at dawn (Rll); the drowning scene
from Berg's Wozzeck (5); the second Water section from Berio's own
first movement (U); and the 'Scene by the Brook' from Beethoven's
Sixth Symphony (X). The underlying presence of this image here as
in the first and last movements is an essential unifying feature of the
work as a whole.
(ii) Allegory: Mahler's 'satire upon mankind' centres upon the
proposition that although art may move and uplift its audience it
will not change the way they behave. A more contemporary view of
this theme is introduced by Berio towards the end of the movement,
at AA. Up to this point, the first tenor has delivered monologues
from Beckett's The Unnamable (discussed below, p.5S). But as the
climax gathers force, Berio breaks free of this text to expand upon a
quotation from his own article 'Meditation on a Twelve-tone Horse'
(Berio 1968) written in the same year as Sinfonia - 'all this can't st~
the wars, can't make the young older or lower the price of bread',
etc. - ending with an absurdly fragile declaration of faith in art,
22 Itself in tum an echo of Brecht's Das Badener LLhrsljjck; see Serio 1985, p.167.

53
Playing on Words

instantly countered by Beckett's 'there must be something else.

Otherwise it would be quite hopeless. But it is quite hopeless', etc.

Programmes
At different times, Mahler proposed two partly related images to
convey the emotional import of his scherzo. The first was conserved
by Natalie Bauer-Lechner after a conversation with Mahler. She
records his words as follows: 23
If, at a distance, you watch a dance through a window, without being able
to hear the music, then the turning and twisting movement of the couples
seems senseless, because you are not catching the rhythm that is the key to
it all. You must imagine that to one who has lost his identity and his
happiness the world looks like this - distorted and crazy, as if reflected in a
concave mirror. The Scherzo ends with the appaUing shriek of this tortured
soul.
The image of the meaningless, maniacal dance seems to have stimu-
lated musical associations for Mahler himself, since the song from
Dichterliebe, 'Das ist ein Floten und Geigen', whose final bars end
the movement, is itself an alienated description of a wedding ball
observed by the bride's abandoned lover. 24 Berio enlarges upon this
image, introducing quotations from the second movement ('Le Bal')
of Berlioz'S Symphonie fantastique, whose programme describes a
parallel situation, and from Ravel's LA Valse, whose Viennese
lyricism grows increasingly hysterical as the work progresses.
The second, rather more generic image that Mahler suggested for
this movement was formulated in a programme for the whole work
that Mahler sketched in Berlin in 1901. The first three movements
are all described as retrospects upon the life of a dead hero. After
the struggles of the first movement, and the temporary idyll of thi!
second, in the third 'the spirit of unbelief, of presumption has taken
possession of him, he beholds the tumult of appearances, and
together with the child's pure understanding he loses the firm foot-
ing that love alone affords; he despairs of himself and of God. The
world and life become for him a disorderly apparition; disgust for all
being and becoming lays hold of him with an iron grip and drives
him to cry out in desperation,.25 Apart from the evident correlation
between this programme and Berio's version of the climax discussed
above, this text evokes no musical illustration. Instead, Berio
chooses a remarkably apt literary parallel in the form of Beckett's The
Unnamable to provide the verbal materials for the movement. It is to
these that we must now tum.
Sources for the text
Beckett's L'Innommable was published in 1952, the last volume of a
2J Bauer-Lechner 1980, pp.43-4.
,. d. Tibbe 1971, pp.5S-9.
2S Mitchell 1975, p.I83.

54
 'In ruhig f1iessender Bewegung'

trilogy whose other two parts were Molloy (1950) and Malone meurt
(1951). Beckett then produced his own translation into English,
which appeared in 1958. Berio presumably chose to work with this
latter version because he was writing Sinfonia for New York and
would otherise have had the complex semantic games of the first,
third and fifth movements all played out in French.
The trilogy accomplishes a gradual dissolution of traditional
narration and character such that by the third book both are under
constant threat. Thus The Unnamable presents the monologue of a
word-spinning narrator placed in a limbo not unreminiscent of
Dante's Divine Comedy, and assailed by voices that seemingly
attempt to foist upon him some recognizable character - whether
that of Mahood, imprisoned up to his neck in an urn outside a
restaurant near the shambles, or the near-insensate Worm, who
resists all but the most negative characterization. While the narrator
attempts to talk himself into extinction, he is continually propelled
back to more or less substantive identity by the fragments of per-
sonality with which he believes that his torturers tempt him.26
By adopting this text, Berio is able to invert the 'heros tue' theme
of the first and second movements. Hero becomes anti-hero. Death
deplored becomes death desired. Although within The Un namable
there are initially areas of concentration upon the personas of Ma-
hood and Worm, the constantly self-renewing and self-confounding
monologue eschews large-scale narrative structure with such de-
termination as to make any resume by salient fragments, in the
manner of the first movement, meaningless. 27 Accordingly, Berio
lifts from the text any fragments that suit his purpose. While some
of these are simply striking, autonomous images, others are used to
comment upon processes at work within the movement and upon
the situation within the concert hall itself. Both of these devices will
be examined below.
Although Beckett provides the substantial majority of the text,
other materials proliferate from it. As well as passing references to
Joyce's Ulysses and a quotation from Valery, Berio incorporates titles
and written indications from various of the scores involved, solfege,
student slogans from the disturbances of the previous spring in
Paris, plus the self-quotation noted on p.53. The full range of Berio's
vocal resources is employed to articulate these materials - though
most of the texts are spoken, leaving solfege or phonemes for the
sung sections. Berio also seizes the opportunity to elaborate a multi-
voiced version of the flux of dramatic mood already explored in
Sequenza III for voice. He therefore follows the same method as in
26 For further discussion of this text, and its relation to some of the musical
materials, see Hicks 1981-2.
17 Save that materials from the first Mahood episode, where he spirals inward
towards a tower containing his family, are consistently avoided; and that the final and
initial materials from the second Mahood episode that shows him imprisoned in an
urn are grouped together, between X and Y.

55
Playing on Words

the previous work, placing different adjectives over the various

vocal fragments. 28
Interactions between text and music
Although most of the interplay between words and music concerns
matters of detail, certain features are used to underline the partly
submerged contours of Mahler's scherzo. Thus, as the Mahler estab-
lishes itself after A, materials from the opening sentences of
Beckett's text are heard, of which the phrases 'where now?', 'when
now?' and particularly 'keep going' are subsequently used to herald
each reprise of the scherzo. 'Keep going' also marks the start of
Trio I, and the fact that a structural milestone has been reached is
further emphasised by the slightly modified Beckett extract that
follows: 'Yes, I feel the moment has come for us to look back, if we
can, and take our bearings, if we are to go on'. This passage also
initiates a large-scale association between the Mahlerian trios and
the use of a solo speaker - a device that is amplified on the semantic
level by mirroring Beckett's discourse on 'the show' from p.99 of The
Un namable, first hinted at in fragments during Trio I (F3-GlO), but
explored at length during Trio II (L5-R19), with Berio's own reflec-
tions upon 'the show' during the trio reprise (AA2-CC16).
The sheer multiplicity of local interactions between text and music
is such that proper justice can be done to them only in an inventory
set out as below. But they can be categorized initially according to
which of the three areas of commentary discussed above they
reflect:
(i) The progressive blocking out of the Mahler scherzo. Thus
when, at V6, the reprise of Trio I is pared down to a static, obses-
sively reiterated texture, it is prefaced by Beckett's 'I am here so
little'. And Berio's final surgery upon Mahler's score, omitting ten-
bars at BBS, follows immediately upon the remark 'for the unexpec-
ted is always upon us'.
(ii) Quoted materials: the most richly explOited category. While
some of the references are wittily direct - the moment where Stra-
vinsky's 'Danse de la terre' from Le Sacre du printemps suddenly gives
way to Agon is coupled with Beckett's 'the earth would have to
quake, it isn't earth' (15), and the reprise at AA of some of Berio's
earlier quotations from Debussy's La mer provokes 'La mer, la mer,
toujours recommencee' from Valery's Le cimetiere marin - others
display a Joycean taste for the recondite. Thus Hindemith's Kammer-
musik Nr.4, dismissed with some scorn after C, is ushered in at AIS
with the apparently innocuous 'nothing more restful than cnamber
music'. Only upon consulting Beckett's text do we discover that the
original, at p.lOS, reads 'nothing more restful than arithmetic'.
Again, there is hidden irony behind the rewritten Beckett used to
28 This counterpointing of dramatic mood lS a device of long standing in Berio's
work, first appearing in Visage (1%1) and Passaggio (1%2). It is further developed in
A-Ronne (1974).

56
 'In ruhig fliessender Bewegung'

accompany the 'Scene by the Brook' from Beethoven's Sixth Sym-

phony at X, 'he shall never hear again hear the lowing cattle, the
rush of the stream', since in the original (p.62) this passage de-
scribed not pastoral bliss but the shambles. It will be recalled that
such deflections of a text's original meaning by placing it in a new
context were also characteristic of certain portions of the first
movement (see pp.14ff).
(iii) Berio's own superposed materials. Some comments are
straightforward, such as the first alto's complaint about the insis-
tently repeated high g at C, or the remarks upon increasingly
insistent chromatic scales at E. But again there are other more
recondite references. When, in the immediately preceding passage
at D25, the first tenor throws in the phrase 'counting the seconds',
keyboards and strings immediately begin to echo each other, as do
the two flutes recapitulating the g ostinato. Consulting p.78 of
Beckett's text, one discovers that the passage from which these three
words are taken proposes that, unlike Beckett's Worm, a man would
sound out the limits of his confines by crying out and waiting for the
echo to return.

An inventory of interrelations
Although the prospect may be a daunting one to any but the more
determined reader, the interaction of all the different processes
discussed above can be accounted for properly only in a bar-by-bar
study of Berio's score. Granted the exuberant variety of relations
that proliferate throughout the movement, both intra-musical and
between music and text, it would be foolish to claim that the survey
which follows is in any way exhaustive. But it does nonetheless seek
to record all significant interactions between materials from different
sources, though indicating only in outline those large-scale pro-
cesses that have already been discussed.
Rehearsal letters have been used to divide the account, with bar
numbers counted from each letter as in the rest of this text. All
references to the Mahler scherzo are given as Ma. followed by the
bar number. All references to other scores follow the sequence:
composer, title, movement (if the work is so divided), rehearsal
number (in italics), bar number (counted from the previous re-
hearsal number when these are present, or otherwise from the
beginning). All references to Beckett's The Un namable are to the
Calder & Boyars edition of 1975, and are given as Beck. followed by
the page number. All orchestral instruments are referred to by the
Italian abbreviations found in the score, with Str standing for
strings, and WW for woodwind.

1 Tr &< Tbn: Schoenberg, Fant Orchesterstiicke, Op.16 no.4 ('Peripetie'), 2;

Str, then WW & Cor play the accompanying string chord.
2 (i) Voices take over the brass chord, announcing the title of the Schoen-
berg; Vni B, Vc & Cb take over the accompanying chord.

57
Playing on Words

(ii) FI 1 &: 2, Grelots: Mahler, Fourth Symphony, 1st movement, 1, while

Vni C chromatically fill its bare fifth.
(iii) Vni A: Debussy, La Mer, III, 54 5 (d. the accompanying d~ triad,
inexplicitly present in T, B &: Cb 2 of Berio's accompaniment). This is the
first water reference in the movement and the source of a number of
subsequent quotations.
4 (i) WW, Vni B, Vc, Cb &: Harp: Debussy, La Mer, II, 1. The bare fifth and
flattened sixth transpose those of the Mahler Fourth Symphony opening
up a tone, thus establishing positions on either side of the goal: the c, g
and a ~ of the opening of Mahler's scherzo.
(ii) T 1: direction from the score of Mahler's Fourth Symphony.
5 S 1 announces title of Debussy.
6 (i) Vni B: Mahler, Fourth Symphony, 3. A filled fourth anticipates the
g-c of the next few bars.
(ii) B 1 announces the accompanying direction in the score.
(iii) Pf develops the harp gesture from Debussy,
7 (i) The Mahler scherzo finally enters (ct. Ma.l0).
(ii) S 2 and A 2 name both Mahler symphonies.

A
2 Cl 2 acciaccaturas echo those of Mahler's Fourth Symphony.
2 (i) Pf and Harp underline the harmonic relationship between Debussy
and Mahler.
(ii) Voices confuse the issue as to which movement is to be played.
3 B 1 announces the direction at Ma.1.
55&: A begin solfege analysis of fragments of Mahler melody - a device
that frequently recurs,
8 (i) Sxf a: a warped anticipation of Ma.19.
(ii) B 1: Beck.7.
10 T 2: Beck.7.
11 (i) FI &: Ott: Schoenberg, Op.16 no.4, 3, down one tone (though this
figure appears elsewhere at pitch in the Schoenberg without the piccolo
note).
(ii) Vni B: Hindemith, Kammermusik Nr.4 (Violinkonzert), Op.36, no.3, 5th
movement, 1. To be used extensively. Although its opening three notes
transpose Mahler's opening, here it is the g-f#-gld-c~-d relationship that
is emphasised.
13 Sxf a: Hindemith, op. cit., 2: f and a~ in common with Ma.
15 T 2: Beck.l0S rewritten with reference to Hindemith. Original: 'Nothing
more restful than arithmetic'.
16 Hindemith, op. cit., 3.
18 (i) S 2 comments on Mahler's scoring.
(ii) B 1: Beck.7.
19 Vn solo: Hindemith starts again in its true medium.
20 FI 1 &: Ob 1: Debussy, La Mer, III, 54 7. Accompanying Vn harmonic
starts to thicken into a cluster just as its identity is affirmed.
22 Berio begins to rescore individual instrumental lines of the Mahler in a
more fragmented fashion (ct. Ma.31). Thus Mahler's Vn 2 line is passed
from Sxf a, Sxf t and A 1, the re-entering melody in the next bar is
divided among Sir and 5, etc.

58
 'In ruhig f1iessender Bewegung'

B
2 (i) The g-<,~ gamut of Mahler's fragmented Vn 2 line is now transposed
and saturated to form a WW cluster that clouds both the initial c minor
and the ensuing e ~ major.
(ii) Vn solo: Hindemith, op. cit., L4.
2 Tbn 1 starts an analogous bass cluster.
3 Vn solo leaps to Hindemith, op. cit., P8.
5 (i) Mahler by now entirely transferred to vocal solfege plus pizzo lower
strings.
(ii) Vn solo: Hindemith, op. cit., Q4.
10 Tr continue motion of Hindemith on monotone g.

C
4 (i) Cemb carries on momentum of Hindemith, soon joined by Pf (its see-
saw motion not unreminiscent of Hindemith from op. cit., 0 onward).
(ii) High cluster extends downwards (see ex.20).
(iii) FI & Ob 1: adapted from Berlioz, Symphonie fantaslique, 2nd
movement, 120, imitating the Mahler at C2 in augmentation.
2 T 1 & B 2 dismiss the Hindemith.
4 (i) S 2: Beck.n.
(ii) A 1 reacts to ostinato g.
6 T 1: clearly a joycean reference, though its source has not yet been
traced.
8 (i) S 1: Beck.14.
(ii) Vni C re-establish the I>-f~ cluster: d. opening bars.
9 (i) A 1: Beck.l1.
(ii) B 1: Beck.I2.
10 (i) T 1: origin unknown.
(ii) Vn solo: Berg, Violin Concerto, 2nd movement, 5, sharing common b
and bO with Mahler (a).
13 (i) B 1: transforms the Beckett quote of C9 into a comment on the Vn
solo.
(ii) Vn solo jumps to Berg, op. cil., 1st movement, 169, while amplifying
Mahler's move towards a minor climax.
14 (i) Vni B develop Mahler into tremolo.
(ii) T I: Beck. 78.
15 Voices solfege Berg.
16 (i) Vn solo switches to an equally rhapsodic gesture from Brahms, Violin
Concerto, 2nd movement, 48.
17 A 2 underlines this transition.
20 Fg & Cfg elaborate upon Ravel, La Valse, 26 4-6, the first of many quotes
from this work. It is identical with the third bar of the main Mahler
theme (d. 03, Vni A), and recurs frequently during the next few pages.
22 T 1 identifies the Ravel, subtitled a 'Paeme choregraphique', but runs on
into Beck. 73 to create an apparent description of the 'paeme' out of a
passage that originally alluded to the 'torturers'.

D
4 (i) FI 2 tremolo begins a new duster.
(ii) Vni C begin a triplet heterophony around Mahler.

59
Playing on Words

5 Mahler is subjected to a variety of further heterophonic distortions:

(i) Ob & CI reduce it to a skeleton of repeated notes.
(ii) Vni B play one semiquaver late.
(iii) Vie add glissandi.
9 Vni propose a new distortion of Mahler through continuous jumps in
octave transposition.
10 Brass: the.rn rhythm first introduced in the Mahler at the previous
bar (in WW) instantly provokes an emphatic interruption in the same
rhythm from brass and percussion. Although this gesture derives from
Ravel, op. cit., 36-9 and 6~, and will appear as such at E23, its
harmony here is derived from the omnipresent bass figure first heard at
C20. Snare drum continues the rhythm as an ostinato.
12 (i) A: Ravel, op. cit., 32. This quote continues in A & Vni through
to D21, though truncated by one bar. The chromatic accompaniment
rapidly expands into an autonomous WW & keyboard 'cloud' (d. Stra-
vinsky, Le Sacre du printemps, 3~7, for a precursor of such textures). This
is the first of several elaborations upon chromatic materials.
(ii) T 1: Beck.12.
15 A 1 rephrases Beckett in ironic reference to chromatic proliferation.
22 FI!: Ravel, Daphnis et Chloe, 1764, down a semitone.
23 T 1 identifies the flute quote but, whether through misprint or design,
alters Daphnis's sex. According to Altmann 1977, p.32, the curious
description 'written in red' refers not to some dandyish predilection for
multi-coloured scores on Ravel's part but to the cover of an RCA record-
ing of the work.
24 Staccato wind chord based on scales of c minor underlines Ravel quote.
25 T 1: Beck. 78. In the original this refers to the possibility of exploring
one's confines by listening for echoes - a means of resolving the problem
alluded to by T 1 at C14. It immediately unleashes echo effects on flutes,
keyboards and strings.
27 FI 2 & Ott take up the obsessional g from Daphnis.
30 A 1 underlines the chromatic element in the Mahler.

E
CI 1 declares chromatic independence, and ends chasing the Vn solo at
E3.
2 Ob & Vn solo: Debussy, lA Mer, II ('Jeux de vagues'), 24 2: a further
exploration of the chromatic, as S 2 notes.
3 T 1: Beck. 104, used to comment on the formal situation as Mahler
dissolves into one portion of the 'Jeux de vagues', which then dissolves
into another. Berio acknowledges this latter by rewriting the end of the
quotation.
4 Vni & Vie join the Debussy.
6 (i) C i joins the Debussy.
(ii) Having reached Trio I (Ma.I03), Mahler's text goes underground,
asserting its existence only by occasional fragments.
7 (i) WW & Vc continue Debussy, while Vn solo, which should by now be
silent, fixates on the last eight notes of its solo.
(ii) S I misquotes Mahler.
9 (i) Debussy, op. cit., 19.
(ii) B 1 misquotes Mahler.
II (i) CI I & Ob I extend and develop Debussy motif.

60
 'In ruhig f1iessender Bewegung'

(ii) A 1 takes b from Debussy WW to revert to Mahler.

(iii) B 2: Beck.52.
(iv) Debussy cello figure leads back into:
12 (i) A similar figure from Mahler.
(ii) Fl 1, Ob 1, Vni A & B & Vie: Berlioz, Symphonie fantastique, 2nd
movement, 122.
16 (i) Vni C: harmonic from Debussy, op. cit., 545, returns.
(ii) End of Berlioz provokes:
17 (i) Ob 2, Fg 1 & Str: Ravel, La Valse, 18 7.
(ii) Sn drum directs attention back to r.T.I rhythm, in preparation for E23.
18 Further chromatic interference from keyboards.
19 T 1: Beck.52.
23 FI. Ob, Fg & Tr: Ravel, op. cit., 38 10, takes over from keyboard
chromatics, leading into Ravel's own quotation, involuntary or other-
wise, of the scherzo from Beethoven's Ninth Symphony at E26. The
Beckett quotation in T 1 reflects both this and the recurrent 'obsession
with the chromatic' that pervades this section.

F
1 T 1: Beck.97.
4 (i) Mahler in Vni B is compared with Berlioz, Symphonie fantastique, 2nd
movement, 154 (Vni C).
(ii) T 1: Beck.99 rewritten in 1960s jargon. We are to return to this dis-
course on 'the show' at length in Trio II.
(iii) FI 2: chromatic end of Ravel and Mahler converge upon:
5 Fl & Cl: Berlioz, op. cit., 156, rewritten to accompany the other Berlioz
quote in Vni C. It leads into:
8 (i) orchestra: Mahler, Ninth Symphony, 2nd movement, 20, thus
echoing the previous 'Ninth Symphony' interruption at E26.
(ii) B 1: Unknown source, but clearly provoked by Mahler's Ninth.
9 (i) Fl only sustain a rewritten version of Mahler's Ninth Symphony
altered to fit:
(ii) Str & Cor, which revert to the Mahler 'text'.

G
4 Tr: rhythm similar to several wind figures from Debussy's 'Jeux de
vagues', but identical with none.
S B 1: Beck.99.
6 Harp establishes a clf # polarity, with fifths in organum, immediately
taken up by the Tr. This polarity is to become a pivotal feature uniting
Stravinsky's 'Danse de la terre' at H15 and Debussy's 'Dialogue du vent
et de la mer' at J.
9 FI & Vni: a minor mystery. None of the commentaries on this movement
can suggest a source for this interpolation, nor can Berio recall it. A
source for the rhythmic combination might be the 2nd movement of
Mahler's Second Symphony (cf. for instance b.68), which Berio had at
one point planned to use as the basis for this movement. But there are
no corresponding pitch sequences.
11 Fragments of Mahler have been heard with greater insistence over the
previous pages. Here, with the first reprise, it takes over completely.
24 B 1: Beck.22, a possible reference to the incursion, as from the next bar,
of Berio's favoured device of heterophonic elaboration.

61
Playing on Words

H
55 (i) Vni A & B alternate with keyboards in delaying and 'flurrying'
Mahler'S Vn parts.
(ii) The texture is further thickened by voices being required to imitate
given instrumental lines.
(iii) Sxf start a final process of chromatic proliferation, reflecting Mah-
ler's own usage from H5.
2 T 1: Beck.99.
5 Vc join the chromatic quagmire.
11 Quickening chromatic motion prepares:
15 Stravinsky, Le Saere du printemps, 'Danse de la terre', 75 11. This starts
in Timp and Vni B &; C, but spreads immediately to the whole orchestra
save keyboards, who continue almost unheard with the two violin parts
from the Mahler. However, the version of the Stravinsky that ensues is
more bizzarely distorted than any previous manipulations to which
Mahler has been subjected. HI6-23 are eight bars, whereas the corres-
ponding Stravinsky 76 to 77 are nine; but each stratum of the texture
omits a bar at a different place. Thus Vni, Vie &; Vc amalgamate b.2 &; b.3
of the extract; the Timp part compresses b.3-5 into two bars of c//# oscil-
lation; WW and Cb omit b.6; Cor omit b.8; and Tr, omitting nothing, are
left one bar out at the end. The harp mediates between discrepant Vc
and Cb parts.
24 Str &; WW: one bar's reversion to Mahler.
25 Stravinsky, op. cit., 77 5.
26 Back to Mahler.
27 Back to Stravinsky, 78.
78.
2 (i) Fl: encapsulation of Mahler.
(ii) T 1: Beck.Sl, underlining the earth theme just before the 'Danse de la
terre' disappears.
4 Stravinsky, op. cit., final bar of 'Danse de la terre'.
5 (i) answered by brass chord from Mahler leading into Trio II.
(ii) Timp contrasts c1/# polarity of Stravinsky with the c1g of Mahler.
6 Str: Stravinsky, Agon, 'Double Pas-de-Quatre', 61.
7 (i) T 1 uses Beckett quote to underline passage from 'Danse de la terre'
to Agon.
(ii) The three Vc and Cb upbeats to Trio II lead to nothing save high
flute es.
(iii) Timp match this with the low c from Debussy, La Mer, III, 1.
8 The start of Trio II (Ma.I90), like that of Trio I, submerges into
fragments. These are in constant interaction with two other protagonists:
Stravinsky's Ago", and Debussy's 'Dialogue du vent et de la mer' from
La Mer.
(i) Ob, Fg &; upper Str: Stravinsky, 61.
(ii) Vc & Cb: Mahler contrasted with:
9 Debussy, 2.
10 (i) Mahler.
(ii) AI: Seck.35.
(iii) T 1 describes 'Double Pas-de-Quatre'.
11 Debussy, 4.
12 Ob, Fg & Str: Stravinsky, 62.
13 Lower Str: Debussy, 6.

62
 'In ruhig f1iessender Bewegung'

J
1 (i) CI, Db, Tr, Timp (d. H15) & Cb: Debussy, 43 1.
(ii) Other Str: Stravinsky, 73.
(iii) A 1: source unknown.
(iv): T 1: Beck.l01.
2 Low Str: Mahler.
3 Stravinsky, 76.
SLow Str: Mahler, using common g from Stravinsky.
6 & 7 (i) C 1, Fg, low Str: Debussy, 43 6.
(ii) T 1 repeats Beck.22.
7 Str (save Vni C): Mahler.
8 T 1: Beck.52, this time presumably referring to his repetition of Beck.22.
9 (i) Castagno Stravinsky, Agon, 'Bransle Gay', 310.
(ii)Str (save Vni C): Stravinsky, op. cit., 'Triple Pas-de-Quatre', 102, an
approximate transposition of the Mahler, to which they instantly revert.
10 (i) FI 3 & Ott resume high c of the opening to Trio II in response to the
Debussy c from low WW.
(ii) FI & Vni C: Stravinsky, Agon, 108, which rapidly spreads to the
whole ensemble.
14 Brass upbeat to first large tonal step in the Mahler (to d major)
encounters:
K
1 (i) massive cluster (see ex. 20 on p.51) that separates into four agglom-
erates, each with its own intensity pattern.
(ii) The only detail to break through is the independent timp line, in part
a continuation of Stra\insky' s 'Bransle Gay' rhythm.
S Str: Mahler fragment.
7 Cor: Mahler fragment.
11 AU parts start independent semitonal shifts around their original note.
Aggregate rhythms increase to repeated semiquavers. Dynamic contrasts
reduce to WW vs. Str.
19 The new aggregate thus established displaces the 'break' in the cluster
(d ,'b) by two octaves, and shifts bass notes by semitones. By now,
dynamics are unified.
20 Repeat of the brass upbeat to this section (not present in the Mahler
original).
L
1 Another huge cluster, this time with layered scoring.
2 Vc initiate a return to the Mahler,
4 (i) which, however, is radically reorchestrated: Vn 2 becomes FI; Db
becomes A 1; low Str becomes Bongos.
(ii) Ott, Db &: Org add high drone, harmonically complementing
Mahler.
(iii) Castagn revert to Stravinsky's 'Bransle Gay'.
5 (i) Str (plus Timp) interrupt periodically with their layer of the cluster
from Ll.
(ii) T 1 begins a long narration from Beck.99 on the nature of the enter-
tainment in hand. This provides some measure of unity as the Mahler
becomes increasingly fragmented or obliterated.
8 Another passage closely integrating materials from Agon, 'Triple Pas-de-
Quatre', and fragments of Mahler:

63
Playing on Words

(i) Tr & Vie: Stravinsky, 106, combined by both pitch and phrase-shape
with:
(ii) B 2: slightly warped version of Mahler.
9 (i) Tr make a unified phrase out of Stravinsky and Mahler.
(ii) Thn: Stravinsky, loco cit.
10 (i) Fg 1 & Vni A take over Stravinsky.
(ii) FI & B take on Mahler's Vn and Vc parts.
11 (i) FI & Tr: Stravinsky, 107.
(ii) A 1 takes on Mahler's CI part.
(iii) Vni B link Mahler with:
12 Stravinsky, loco cit.
13 (i) Thn & Tuba imitate their previous Stravinsky material in order to
underline a Mahlerian progression.
(ii) C1: Mahler.
14 (i) Fl & Vni: Stravinsky, 108.
(ii) Tr, Pf & Vc: Mahler.
15 Pf, Cemb & Cb continue Mahler, distorted by irregular octave doublings
from the two keyboards.
17 Tr combines the two upper lines of the Mahler, as does FI in the next
bar.
18 Cb pizzo d acts as catalyst for return of Stravinsky, 61.
20 Tr & Ott continue their previous gesture with an independent scale.
22 Ob, Fg & Str recapitulate Stravinsky, 6l.
23 C1: Mahler.
25 Mahlerian upbeat again leads to:

M
25 (d. Ma.257) a more layered duster, the gamut of whose upper group
contracts by a semi tone at either extremity the analogous group in the
cluster at K (see ex.20). This quickly fills out into a fuller cluster whose
upper components (down to the g of Tr 1) transpose the analogous
components of the duster at L down a semitone. Strings momentarily
penetrate with the Mahler.
3 Sxf t, Fg & low Str: a further Mahler fragment.
5 Layered dynamics are again used.
12 (i) FI: Mahler's Vn part (d. Ma.268) modified by leaping to positions
within:
(ii) a series of less dense wind aggregates using the rhythm of Ma.269.
13 Strings resume their chromatic shifting.
15 (i) Timp plays Mahler figure that should have occurred four bars
previously. This and the misplaced rhythm of M12 are the first hints of
the major metric dislocations to come.

N
25 Mahler returns in the Tr quartet and Harp, with Pf and Cemb playing
2nd Harp, and Sxf supplying Vn part until Str take over. Str, however,
descend slowly from the registers established by their MIS cluster,
providing a lush, widely spaced background that adds a patina of
Straussian indulgence to the direct, rather spare nostalgia of the Mahler
score. The Vn 2 line is thrown from part to part.
9 S 1 takes Mahler's Vn 1 part.
13 Fl superimposes Ravel, fA Valse, 32.

64
 'In ruhig fliessender Bewegung'

17 (i) FI jumps two bars in the Ravel.

(ii) Contrary motion chromatics (an exaggeration of the missing Ravel
accompaniment) in WW and Str blur:
(iii) Mahler, played by remaining wind.
(iv) Tamb displaces 'Bransle Gay' rhythm.
20 The start of an episode exploiting the common upbeat pattern of sections
from Ravel's La Valse and Act II of Strauss's Der Rosellkavalier that almost
entirely obliterates the Mahler.
(i) WW & Vc: Ravel, 78 2.
(ii) Timp combine rhythm of Mahler, low Str, with pitch of Ravel.
(iii) Similarly, Vni C develop Mahler's Vn figuration, at first following
Ravel's harmony.
22 A 1 joins Beckett narration.
23 S 1: Mahler fragments, passed to T 1 and back.
25 (i) Cor adapts Ravel's chromaticism to lead into:
(ii) Strauss, Der Rosenkavalier, II, 250, with a touch of parody added
through stuttering Cor and glissando Vc. The CI line adds an upbeat
flourish to the end of the previous section.
o
Ob counteracts Rosenkavalier with Ravel, 81 6, then jumps a bar as:
2 Rosenkavalier dissolves into its counterpart at Ravel, 82.
6 Another bar of Rosenkavalier, II, 254 10, with a counterpoint inserted by
Berio on Tr and middle Str, facilitates the jump to:
7 Ravel, 92 3.
12 Ott, FI 1 & CI picc enter over Ravel with Strauss, 253 5.
13 The rest of the orchestra join in, but immediately:
14 S 1 enters over Strauss with Mahler.
p
1 (i) Tr 2 & A 1 help re-establish Mahler, while Tr 1, in duet with Tr 2,
plays a transposed version of the solo from Ravel, 67 - which thus
acquires a new harmonic interpretation, centring on the dominant rather
than the submediant.
(ii) Fg 2 & Tuba: source unknown.
5 Fl: Vn 2 transposed (while Vni B instead anticipate next bar), leading to:
6 (i) retrograde of Mahler's scale, carried on into P7 by Vni C, and ending
on a high H that is the last relic of the Mahler text before its metric
underpinning is lost for the first time.
(ii) A 1 starts a vocal line (proceeding to 5 2 & 1) that combines the
Marschallin's moment of final renunciation at 293 of Rosenkavalier, Act III
(up a semitone) with Oktavian's vocal line at 2987 - by which time the
orchestra is playing the same passage.
7 Vni B & VIe introduce:
8 WW, Cor, Vni C & low Str: Strauss, op. cit., Act III, 298 5, again with
satirical touches. Vni A & Band Vc add passing notes. Vc at PIO
substitute an outrageous double sharp for Strauss's f#. Pf, Harp & VIe
render the texture extravagantly lush.
9 Fl & Ott maintain an independent line, taking over the Vni C f# that
echoes the missing Tr line of Ma.316.
14 Pf & Str: the final chord from the Rosenkavalier quote recalls Ma.317-18.
16 (i) Fl & Vni A encapsulate Ma.31S-19, and then join:
(ii) Vni Band Ob 1, which anticipate the Brahms quotation at Q.

65
Playing on Words

18 (i) Cfg, Tr & low Str take up Ma.320. Thus, since P = Ma.309, six bars
have been added - the first of a number of distortions of Mahler's metric
framework.
(ii) FI complements previous chromatics by a descent from its high f~.
19 T 2 relays instruction from Mahler's score.

Q
3 (i) FI & upper Str: over the e minor added sixth chord enters Brahms,
Fourth Symphony, 4th movement, 69.
(ii) T I inserts into the Beckett narration an acknowledgement of the
Brahms quote.
3 (i) Vie speed up Brahms's rate of flow, while FI 1 ornaments it.
(ii) Voices solfege Brahms.
(iii) Pf & Cemb take up their characteristic semiquaver movement,
developing the fixed anchoring pitch of the Brahms in a manner remi-
niscent of the first movement (d. K).
7 Sn drum, followed by Bongos, outline the Cb figure from the Mahler one
bar late.
8 Mahler's text begins to be treated with increasing freedom as the music
once again cuts free of its metric proportions:
(i) Ott plays an inversion of the main WW material from the Mahler,
answered by Ob, entering two bars early and a third higher.
(ii) T 1 quits his Beckett narration to comment on 'crossed colours',
referring both to the crossing instrumental lines and to Pousseur's
Couleurs croisees, of which a stylistic pastiche follows from Q14. He then
jumps to Beck.l03.
(iii) Cb develops Ma.328 too late.
11 Although in the Mahler the Tr choir stops its alternating chords here,
Berio continues to reiterate them obsessively, passing them to the oboes
in preparation for the Bach quote at Q19.
12 Another reference to Couleurs croisi!es, which is based upon transforma-
tions of the song 'We shall overcome'.
14 (i) T 1: however, it is not Henri Pousseur, but Beckett who says 'if this
noise' etc (Beck.92).
(ii) Harp, Pf & Cemb embark on a pastiche characteristic of the textures
and instrumentation of Couleurs croisees, though without actual quotation
from the score (pace Altmann 1977, p.3S, where an erroneous reference is
given).
16 Ob 1 sharpens the c of the Mahler chord, and Fg 1 enters on an a
producing a dominant seventh (plus minor ninth from FI in the next bar)
resolving onto the Bach quote in Q19.
17 T 1 resumes his narration from Beck. 99.
19 Alternately Fg & low Str, Ob, upper Str: Bach, first Brandenburg
Concerto, 2nd movement, last 4 bars, but with b~ for b, in the penulti-
mate bar.
R
5 Bach speeds up from units of four quavers to three quavers.
9 (i) Bach slows down again at cadence.
(ii) CI 1 picks out e an octave below that on to which Vni B must resolve
and:
10 is joined by Fg 2 g#, both preparing:

66
 'In ruhig fliessender Bewegung'

11 (i) Fl, Cl, Fg & Vie: Schoenberg, Fun! Orches/erstiicke, Op.16 no.3
(,Farben'), 2. As well as reflecting the alternating timbres of the Bach,
this quotation also recalls the alternating chords of the missing Mahler
text.
(ii) S I: Beck. 109, modified by the addition of 'in a lake full of colours' so
as to acknowledge Schoenberg's 'Farben' (colours) which in the 1949
revision had the title 'Sommermorgen an einem See'. The quotation also
anticipates the drowning scene from Wozzeck that begins at S, and
constitutes the first link in a chain of water references.
(iii) Vc begin a series of metrically dislocated fragments from Mahler -
here Ma.330 with the g sharpened to accord with the Schoenberg.
12 Vni C join in Ma.33l.
13 Vc switch back to Ma.330 as Vni C continue with Ma.332.
16 Vni C & Vc: Ma.334 (Vc again sharpen g).
18 Vni C & Vc: Ma.335.
19 Vni B: Ma.339.
20 (i) A 3, Vni A & B: Ma.342.
(ii) Cfg prepares a re-establishment of Mahler: cf. Ma.340.
21 (i) Schoenberg jumps one bar to the second half of 5.
(ii) S & A: Ma.343 without accidentals.
(iii) B: Ma.339 in the shape of Ma.333.
22 Vni C: Ma.338.
23 Mahler's text re-established (d. Ma.342) twenty-two bars late. Tr l's e~is
common to both Schoenberg and Mahler, and the long held low Str c
resolves, as in Mahler, onto d •.

s
2 The WW's chromatic descent in the previous bar evokes by way of
response not the 2nd scherzo reprise, but:
(i) Vc & Cb, plus Vie a bar later: Berg, Wozzeck, Act III, 284, the point at
which Wozzeck drowns, overheard by the Captain and the Doctor.
(ii) Fg and Guiro maintain the Mahler, which is once again subject to
fragmentation and encapsulation.
2 Wind & Vni combine Ma.349 and Ma.35l.
3 (i) S 1 carries on Ma.352.
(ii) A 1 comments on Wozzeck's murder of Marie.
(iii) B 1: Wozzeck's last word (though not with the intonation indicated
by Berg).
4 (i) WW play Ma.353-5 simultaneously, thus producing a duster to
match the Berg.
(ii) B 2 also comments on Wozzeck.
5 (i) A, with interjections from Tr and Vni C, superimpose a slowly
ascending major third cluster that is finally to establish itself, at TID, as
that of the second Water section from the first movement. But for the
moment it complements and thickens the texture of the Berg.
(ii) Fg encapsulate Ma.354-7.
(iii) S: Ma.355.
7 (i) A 1 initiates, several bars early, the conversation between the Cap-
tain and the Doctor as Wozzeck drowns.
(ii) T 1: Beck.89.
8 (i) Cl 1, joined subsequently by other WW: Ma.361-3.
(ii) B 1 carries on the Captain's initial remark.

67
Playing on Words

10 CI 1 & low WW start the next chromatic ascent from Berg.

13 The Mahler is temporarily re-established, nine bars early. Thus Ma.369
starts from Vie, adding Vni plus, in 516, Cor 1 playing CI 1 of the
Mahler, and Cfg playing Pg 1 of the Mahler.
16 B 1 switches to the German text of Wozzeck.
17 (i) S takes over Mahler line, with rhythmic distortion from A and Vni A.
(ii) Low Str start the next ascent from Wozzeck, but:
18 (i) this turns into a precipitous glissando, abetted by:
(ii) WW, who likewise abandon the Berg, speeding up the tempo of
their chromatic scales.
(iii) meanwhile with S & A Mahler is once again lost.
19 B 1 continues the Captain's words in English, while orchestra builds to:

T
2 Another briefly held large cluster whose upper limit shifts downwards
by a further semitone (d. K and M as well as ex.20).
2 (i) Again, the text of Wozzeck in English and German.
(ii) Meanwhile the cluster immediately begins to reduce to:
S the filled major third from the second Water section of the first move-
ment (at E).
10 The polyrhythmic texture of the Water section is restored.

u
3 A: source unknown.
4 Cl: Ma.396; Vni B: retrograde of same.
S Cb follow on with Ma.397, with which they continue, fixating on the d~
until the other parts, delayed by two bars of Berio, can join them.
7 PI 1, CI 1: Ma.397, upon which Cl pice provides heterophonic elabora-
tion.
8 (i) S take up the oscillation f...;!, to be sustained with melancholy persis-
tence during the coming bars.
(ii) Vni begin a slow descent from their g~-b~ duster towards an e-g
duster that heralds the return of the Mahler.
10 Cl 1 mixes PI 3 & Fl 2 parts of Mahler, harmonized by S oscillation,
12 as does PI 1.
13 5 oscillation generates upbeat to:
v
1 Mahler re-established (d. Ma.402), this time having gained two bars in
the process.
4 (il T 1: presumably Beckett, but untraced.
(ii) Sn drums adds a second ostinato, projecting the 'Bransle Gay'
rhythm into a four-quaver unit.
S WW joins in the e-f oscillation.
6 This should mark the start of the trio reprise, but Berio now embarks on
his most startling Mahlerian transformation yet. Taking as his cue the f
drone that pervades much of this section of the original, Berio freezes
fragments of Mahler's material into a doggedly persistent ostinato,
against which piano and sections of the solo violin part present two
sections of the Mahler grossly distorted by octave transposition and, at
times, rhythmic encapsulation. While the whole episode up to Y lasts 34
bars in both Mahler and Berio, the two portions used (Ma.408-23 and

68
 'In ruhig f1iessender Bewegung'

Ma.434--(0) are both displaced from their original positions within this
framework.
(i) An ostinato is built from Vc & Timp (Ma.407) plus Org & Cemb
(Ma.433).
(ii) Fg continues Mahler's chromatic scale, which ends on UI0 with the
Cb drone.
(iii) Meanwhile Vni C glissando through two octaves to establish
another high cluster on a-c.
9 T 1: Beck.81, an appropriate response to four concurrent ostinati.
11 Pf starts its distortion of Ma.408.
W
2 Solo Vn. joins in with an equally distorted version, whose b leads into:
2 a return of Hindemith, Kammermusik Nr.4 (Violinkonzert), Op.36 no.3,
P16. (It will be recalled that this work was previously the cause of some
exasperation - which its reappearance amidst this lunar landscape does
nothing to diminish).
3 (i) Vn solo: Hindemith restored to its original speed.
(ii) Pf: the pitches of this and the following two bars derive from the
WW figure that accompanies the Vn in Hindemith's Kammermusik, the
rhythm of which, at half speed, coincides with the ostinato into which
the 'Bransle Gay' has been transformed.
6 Pf and Vn solo each provide their own version of Ma.414, which the Pf
continues.
7 (i) Vn solo resumes Hindemith at the appropriate point.
(ii) The ostinati are progressively subtracted from this point, starting
with the e-f oscillation.
8 (i) T 1 switches to commentary on Beckett (with echoes of Beck. 107).
(ii) Vn solo continues down the scale started on the first beat of the bar
so as to lead into:
9 a similarly distorted version of the WW material from Ma.417 (at which
the Pf has also arrived).
12 Vn solo takes string line from Pf for one bar.
13 (i) Pf carries on with Ma.421-3.
(ii) Meanwhile VIe start Ma.434, and thereafter Str progressively sub-
stantiate the reference, through to Ma.340 (WI8).
14 (i) Vn solo takes up Hindemith at the point where it had broken off.
(ii) Vc and Timp ostinato ceases.
15 Timp accompany Ma.423 in Pf.
16 PI >uddenly switches to the rival Mahler passage in the Str (Ma.437).
17 Cor, Org & Cemb ostinato ceases, having failed to coordinate with
Ma.437 from which it originated.
19 (i) The last of the ostinati - on Sn drum - ceases.
(ii) e ~ implications of Mahler from Vni B, plus the oscillating solo Vn
figures usher in Beethoven, Sixth Symphony, 2nd movement ('Scene by
the Brook'), 69.
(iii) To this T I responds with an apparently apt Beckett quote, 'he shall
never again hear the lowing cattle' (Beck.62) - one of Berio's darker
jokes, since what is described in the original is a slaughter house.
X
4 Cor takes over Fg figures from Beethoven and holds it awaiting the
return of the Mahler.

69
Playing on Words

6 T I: Beck.43, with 'more' misprinted for ·move'.

9 T I to 'no one': Beck.44. What follows is from an unknown source, but
refers to Vinko Globokar's Voie on a text by Mayakovsky.

y
1 The reprise of Trio II is allowed to start unimpaired, but:
3 immediately begins to be submerged under a rapidly growing cluster
(see ex. 20 on p.5I), beneath which the metric structure once again
becomes lost. Vni A derives from Ma.445.
4 The cluster is completed: it recapitulates that heard at K with minor
variations in the register just below middle c.
7 Mahler briefly re-emerges, with upper Str playing Ma.450 and low Str
Ma.449. Both unite on Ma.451 in the next bar.
12 The complete cluster is re-established.
Z
2 A shift to a new harmonic basis, fully established by Z3. The keyboard
punctuations are to survive as a unifying feature until almost the end of
the movement, thereby matching the earlier use of ostinati.
8 The same chord is now taken up by Str vs. WW, each following its own
dynamic pattern with a gradual accelerando in rates of change.

AA
2 Berio now starts to recapitulate and develop materials used at the start of
the movement in a parallel gesture to the missing Mahler:
(i) Opening chord plus Schoenberg, d. b.!. Whether by printer's error
or composer's design, Cor 3 & 4 now play bq and f~ and Vni A c# and a
(though the 1969 score has the expected notes from b.I).
(ii) As before, WW substitute for Cb.
2 (i) This is immediately followed by Debussy, La Mer, III.
(ii) T 1 now embarks on another monologue on 'the show', but this time
by Berio.
3 (i) A sf. a ~ from the Vie line of the Debussy links with the keyboards'
punctuating chord.
(ii) 52 uses Valery, Le cimetjere marin, line 4, to mark the perennial
presence of the Debussy.
6 a ~ now effects the transition to:
7 Debussy, La Mer, III, 59 10.
9 Sir elaborate on Debussy's figuration.
1lOver the preceding extract, Tr 1 enters with the cor anglais figure from
Debussy, La Mer, III, 18.
12 The rest of the orchestra answer with 18 2, save that Tr rl'tain a dotted
rhythm, running straight on into:
13 (i) Schoenberg, loc. cit., again in the brass while:
(ii) the rest of the orchestra take up Mahler at Ma.463, which thus
appears twenty-one bars late. This leads straight into:

BB
1 the climax of Mahler's movement, at which point his text is fully
restored.
9 5 1 adds alternate, compressed chromatic phrases in solfege and scat.

70
 'In ruhig f1iessender Bewegung'

CC
13 Mahler is again reduced to a series of fragments beneath sustained wind
chords, with Timp providing the most immediate continuity.
13 (i) Again, the Mahler is lost save for a string fragment at CClS.
(ii) Eight-part wind, with Tr & FI, Cor & CI doubling at the octave,
repeat three chords twice as FI I moves from d to g.
(iii) Harp harmonics echo Ma.493 and, in CCI7, Ma.497.
DO
I Harp projects a pitch line from previous WW chords 1 & 2 (linked by
their common d), followed two bars later by another linking the outer
two pitches of 3, the inner two of 2, and the outer two of 1.
(ii) T 1 reverts to a series of fragments from Beck.7.
3 Vni C: Ma.501 (which is thus one bar late).
5 Tr, joined by other wind, initiates a final quote from Der Rosenkavalier,
Act III, 295 7.
9 Full orchestra joins in Strauss.
10 T 1: Beck.23.
12 T 1 announces Boulez quote at EE.
13 Punctuating chord acts as upbeat to:
EE
1 Boulez, 'Don' (= 'present') from Pli selon pli, opening chord.
6 Over the remains of the Boulez chord, WW quartet introduce the choral
opening of Webern, II. Kantate, Op.31, 5th movement.
7 A quiet, aleatoric background is created, with Pf playing a passage from
another work in the programme, while A 1, followed by S, revive the
vocal imitation of instrumental parts first used at H.
9 Vn solo responds with its line from the Webern.
11 Sxf plays solo S part from the Webern.
14 Vn solo switches to Stockhausen, Gruppen, 22 3, joined by other Str in
the next bar.
17 A 1 & 2 sustain Harp c from Stockhausen, which is to become the
dominant of Mahler's f minor at FF.
FF
1 (i) Instead of Stockhausen's chord, Berio uses his own punctuating
chord for the last time.
(ii) Mahler returns (Ma.545), having lost 24 bars, and is tossed from
voice to voice.
8 T 1 refers to title of Mahler (Resurrection Symphony).
13 T 1: the title of a German song, best known for the variations Sweelinck
wrote upon it - though no reference to that work is intended here.
Instead, it serves to underline that this is not a resurrection symphony:
Berio's third movement, like all the others, ends with death.
GG
5 VIe shift by one semiquaver.
6 Hinted at by T I's previous remark, the Mahler jumps ten bars from
Ma.564 to Ma.S7S.
8 Vn A plays Ma.S76 over Ma.577 from rest.
12 The final bar freezes into ceremonial, with gongs and an appropriate
vote of thanks to the conductor.

71
 5
Synthesis and Dissolution

The fourth movement

For a moment it seems as if Berio's commentary upon Mahler is to
continue, for the fourth movement opens with the d. and with
which the vocalist enters at the start of the fourth movement of the
e.
Second Symphony. But there the similarity ends, for Berio, seeking
a measure of symmetry around his central movement, engages upon
a quiet, slow expansion from those two notes to four eight-part
vocal chords that become increasingly invaded and enriched by in-
strumental additions. The latter provide the opportunity for reviving
two features characteristic of the first movement: the exploration of a
fixed harmonic 'object' by multiple voices moving at different rates,
and the 'choral' differentiation of timbre groups.
The basic harmonic process is summarized in example 21. In each
instance, the top five voices move upwards through a seven-note
pitch sequence at differing rates, so that each is finally halted on a

Ex. 21
a
TI A2 Al S2 51

111 II ditto plus continuous and final

T2
Bl B2
F,

iv,

TI A2 Al 52 51 Tl A2 Al 52 51 'DVm

111 IV T2
T2 Bt B2
BI82

72
 Synthesis and Dissolution

different note of the sequence, and the bottom three voices move
downwards in similar fashion. Apart from their common starting-
point, all the upward pitch sequences also have in common the c
and d an octave above - preceded by an a in the first three cases,
and followed by f in the last two.
Most of the other features of pitch organisation reflect Berio's
concern with chromatic saturation. Thus in sections I and II the two
notes that do not form part of the basic pitch process are given
important subsidiary roles: the a b providing a temporary inflection
of the harmony in the bass, and the sustained g in the alto voice
finally resolving onto the missing f~. Although in I the two initial
notes are released as the voices progress towards their final chord,
from II onwards a number of superimposed rhythms from instru-
ments maintain a continuous oscillation between them.
This naturally underlines the move towards chromatic saturation
as each chord reaches completion; accordingly at the end of II the
bassoon re-enters with the transitory a b of a few bars before, and the
violins C and clarinet re-establish a g an octave above the one that
has just resolved onto an f~. This high g initiates an independent
melodic line, pursued well above the main textural aggregate but
complementing its pitch processes. It thus provides the two notes
missing from the chord built up in III: ab and b; and, having moved
to an extreme register, includes the g~ and fn missing from IV within
its line - although now contrabassoon and bass tuba provide an at
the other extremity, and violins in fact complete the twelve-note
a.
chord once the voices have come to rest.
Each chord has its characteristic spacing. I and II are coloured by
alternating tones and minor thirds, with fourth and diminished fifth
at the bottom. III consists of two diminished sevenths - one, more
widely spaced, stretching down from do and the other, in close
formation, moving up from eo - with the beginnings of the 'missing'
diminished seventh chord added above the next tone grouping of c
and d by the first and second sopranos, and completed by oboe and
clarinet. IV, though less logically constructed, retains the charac-
teristic spacing of larger intervals at the bottom (fifth and dimini-
shed fifth) with an aggregate based on thirds and seconds above
them. But the harmonic growth of this final chord is in any case
obscured by an upward moving chromatic cloud. There is thus a
parsimonious growth in harmonic texture, complementing the
gradual extension of harmonic gamut.
Rhythm and texture
The large-scale rhythmic framework is also extremely Simple. The
vocal parts of I and II are rhythmically identical, and both last
thirty-seven minims. III abbreviates this to twenty-five, and IV
compensates with forty-eight. But within this framework a gradual
shift in emphasis and proportion is at work: in I and II there is a
rough balance between the initial db/eb oscillation and the ensuing

73
Playing on Words

expansion, followed by a short final chord; III alters these pro-

portions, for it is now the oscillation and the final stasis that are
proportional, with a rapid expansion between them: IV (which is
nearly twice the length of Ill) mainlains similar proportions between
oscillation and expansion, but throws all its weight onto the long-
held final chord, passing it from choir to choir.
The instrumental oscillations of II and III, and the rising cluster of
IV each use eight superposed rhythms that subdivide a common
multiple (semibreve, breve and dotted breve respectively). As the
common multiples become larger, so the selected subdivisions
display more complex interrelations, producing a subtle growth in
density of the rhythmic texture.
Unlike the calculated labyrinths of the first, third and fifth move-
ments, the second and fourth movements, by virtue of their basic
simplicity, invite the ear to dwell consistently upon the exercise of a
specific form of sophistication: in the second movement that of a
constantly shifting palette of harmonic colour, in the fourth move-
ment a carefully controlled field of instrumental colour and texture.
Berio's choice of instruments is fastidious - a re-assertion of his own
sound-world after the parade of ancestral ghosts in the previous
movement. Its central d~/e~ pedal provides a characteristic combina-
tion: pitched percussion, tremolando string harmonics and flutter-
tongue flutes, joined from A2 by flutter-tongue trumpets, which
have resumed their assorted mutes from the first movement. Their
upward chromatic flurry at B4 is amplified by flutter-tongue muted
horns, a miniature recapitulation of the gesture that ended Epifanie A.
The text
Except in IV, the text consists of the phrase 'Rose de sang', phonetic
materials derived from this phrase, and interpolated 'random
sounds simulating very fast speech'. The words echo those of the
fourth movement of Mahler's symphony - '0 R6schen roth' - but at
the same time recall the gun-shot wounds that killed Martin Luther
King and the theme of blood running through the first movement.
IV reintroduces the 'appel bruyantldoux appel' opposition from the
first movement, thereby ushering in the recapitulation of much of
the verbal material from that movement in the finale; but it ends
with another statement of the 'Rose' text.
The fifth movement
The fifth movement completes a shift in technical focus initiated by
the second and third movements. The chamber version of 0 King
accreted around itself elaborations of its own material to form the
second movement; the third movement achieved a more elaborate
(and destructive) commentary on Mahler's scherzo by seeking out
relationships between that 'text' and other, heterogeneous materials.
It is this search for similarities and common elements which now
takes over as an autonomous principle. By means of it, Berio fuses

74
 Synthesis and Dissolution

together materials from all the previous movements into a new and
vitriolic synthesis. The gesture seems deeply indebted to the
nineteenth-century cult of organic completion. In practice, it offers
neither apotheosis nor resolution, but rather an explosion of raw
energy.
However, the search for structural analogies is also the underlying
principle of Levi-Strauss's Le cru et Ie cuit. Berio accordingly turns
back to the verbal materials of the first movement, and proceeds to
complete the processes initiated there. But again, what on a con-
ceptual level implies a sense of ending on a practical, auditory level
offers an open-ended stream of images only part-comprehensible at
best.
In the first movement it was the internal structure of Levi-Strauss's
text that determined the musical structure of the vocal sections.
Here, however, it is musical processes which have the upper hand

Ex. 22
II II III IV

ble e/b
E

bl
2=empty 7th
A 1
2= filled 5th b
B b Ib
F (=U)
e A

2
Ib

EI B

EI EI
2
b i Ie
I and2 ii
iii
EI D
E
G
E
B i E

FI I it
iii 0
W

75
Playing on Words

Ex. 22
G
C ii
ii
HI
iii
HI
D ii
HI
ii
iii
HI I
(2= empty 7th)
K
c
L
M

N
II
°1 c'

in determining large-scale form, with the text running in tandem. A

detailed examination of the text can therefore be deferred for the
moment. Although the musical processes are no longer articulated
around a single, central text, the formal outlines of the movement
are nevertheless determined by a series of dominant materials. At
the heart of the movement lies 0 Killg. It is preceded by the
alternating chords that opened the first movement, and is finally
overtaken by a hectic orchestral monody reminiscent of, and finally
merging into the one that closed the first movement. Around this
basic structure are woven other recapitulated materials. Materials
from the fourth movement alternate with those from the start of the
first, while above them are presented fragments from the other two
vocal sections from the first movement. Meanwhile a series of
fragments encapsulating the progress of the third movement appear
and dissolve back into the general melee. An outline of the way that
these materials are disposed is set out in example 22. Rehearsal
letters indicate the corresponding position in each movement;
b = beginning, e = end. In the first-movement column, 1 and 2
denote the opening chords (see ex.3); in the second-movement
column, i, ii and iii denote the three sections of its pitch set as
described earlier (see ex.5). In the more detailed examination of
these correspondences which follows, movements are referred to by
Roman numerals, so that III 85 means: the fifth bar of 8 in the third
movement. Material occurring before the first rehearsal letter of a
movement is indicated just by bar number, so III 5 denotes the fifth
bar of that movement.

76
 Synthesis and Dissolution

Before '0 King'

The movement opens with three superposed and partly related
layers. The piano's rhapsodic idiom derives from its solo at the end
of the first movement, but it now incorporates a series of fleeting
allusions. It starts from the final chord of IV, but immediately
follows it with elements of chord 1 from I (anticipating the return of
that chord as an explicit reference-point in V 13). From b.6 onwards
the two chords are made to interact. But I 1 also has an eb in common
with the d~/e~ oscillation of IV; the hint of this in V 1 will be fully
realized in V 12-13. The eb also summons up recollections of the ebld
alternation characteristic of I's instrumental section (d. I J8-Ll).
Echoes of that appear in V 6 and V 8. Above this, the flute expands
upon its solo from the last four bars of 1, embellishing it with the
oscillating thirds from the Fire section of that movement (I E4), to
which more explicit reference will be made at V 10. The voice mean-
while provides a new setting of the text from IV BI-C6, though
inverting its order so as to end with Levi-Strauss in preparation for
the main body of the movement. The close relation of the vocal line
to the flute part is made explicit in V 6--8; and it ends (V 8-9) with
a fleeting allusion to the monody which is to end the movement
(d. J5).
After this prelude, the voices enter at V 10 to initiate a more
substantive set of relations, summarized in example 23. They
employ the familiar db/e. oscillation from IV 1 which dissolves into
chord 1 from I in the strings at V 13, linked by their common eo.
Above them, the flute oscillations are now answered by clarinet and
saxophone - the original protagonists of I's Fire section. They do
not, however, play at their original pitches (d. I E4). Instead all
three now complement the pitch content of chord 1, achieving
chromatic saturation except for the b. This appears in the answering
tremolo from piano and harpsichord - a skeletal version of chord 2
(d. I 3). A further substitute for this chord is provided a few bars
later when chord 1, reinstated at Al and shot through with attacks
from wind choirs, implodes upon itself to form a saturated fifth, c-g,
at A6. Berio has thus sketched out the first two of three alternations
between chords 1 and 2 that make up the first Water section in I.
These chords now move into abeyance for a few bars before return-
ing to complete the reworking of that section.

Ex. 23
b.IO A B c o

A A A A A A A

77
Playing on Words

Having established its relation to chord 1, the d~/6 oscillation now

returns to demonstrate similar relations to two other chords (thus
mirroring its expansion into three different chords in IV). The first of
these also serves to underline its origins in the opening bars of
'Urlicht', the fourth movement of Mahler's Second Symphony.
During IV, the two notes had always avoided the resolution onto f
that completes the phrase in the Mahler.! Now, however, brass
sforzandi are employed to spell out the progress from d~ to f. 2 But
d~ e' f is also the retrograde of the first three notes of the chord
that ends II which the voices recapitulate, at B7, before resolving
onto a more mellifluous six-part chord incorporating the d-a basis of
chord 1. At B9 they return for the last time to the d.le. oscillation,
resolving into a complete version of chord 2 from I. This in turn
ushers in the final move from the expanded version of chord 1
(d. I C1) to chord 2 at 01-3.

e.
Superposed on this basic layer are other elements called back into
play by a process of association. The pitches d ~ and figure largely
in the staccato bass notes accompanying the second Water section at
I F4. Accordingly, the chord 4 cluster reappears, at B2-13, both in its
original form and as reinterpreted in III UI-5, thereby completing
the imploded recapitulation of materials from I. Similarly, the filled
fifth c-g at A6 is quietly answered by another, b-f~, from the elec-
tronic organ at B2. Not only does this complement the filled third,
g.-b. of the second Water section from I, here etched out by trum-
pets and piccolo clarinet, but it also reinstates the harmonic back-
ground to the start of III, from which fragments begin to be heard.
Thus at B the harp plays its initial quotation from La Mer, followed
by a snatch of Hindemith (B2), of Mahler's introduction (B4, cor
anglais), the upbeat to the main theme of the movement (B5,
strings), and the explosion that originally followed nine bars later
(B6, full orchestra plus clarinets). From then on until D, Hindemith
fragments constantly invade the texture, thereby preparing for the
important role that they are going to play later on in the movement.
From B to D the texture is also shot through with explosive
attacks from piccolo, brass and percussion that are to remain a
unifying feature of the movement through to the final section. At
this point (though not necessarily hereafter) the brass attacks serve
to underline pitch processes already discussed, while the piccolo
line keeps its own, non-systematic independence.
Materials from '0 King'
At D a reworking of the outburst from I C ushers in a restatement of
the principal line from II, which conserves the interaction of pitch
and rhythmic sets (the latter, due to the slower tempo, now counted
in semiquavers rather than quavers) through to the point where the

I Instead, it resolved onto f. gl and g respectively.

2 d. Altmann 1977, p.54.

78
 Synthesis and Ois;olution

principal voice abandons the articulation of that interaction (19, d. II

012). The four relevant statements of the pitch cycle are thus placed
as follows: A begins at 03, B at E8, C at G3 and 0 at H1O. The
more complex context in which this line is now placed entails a few
minor alterations, but apart from a moment of textural confusion in
which the a of section Aiii is blurred (011-12), the pitch set
remains intact throughout. Two minor glosses upon it are intro-
duced: the long-held final b. of section Aiii is temporarily deflected
to a bq (E7), and in statement B a g# is introduced to mediate
between the bs and c#s of sections Bi and Bii (Fl and FS). Similarly,
the rhythmic set maintains a semi quaver version of example lOa
in all but a few instances. The blurring of a in section Aiii provokes a
parallel dislocation in the rhythmic set with the aggregate 7 of f+a
being extended to 9. There are also long pauses on the g~ of section
Aiii (E2) and the f of section Ci (G3) as other materials are dealt with;
and the a of section Ciii is abbreviated from 4 to 2 to provide a more
powerful sequence (H4).
Of the rest of the structure of 0 King only vestigial fragments re-
main, and these are in any case transformed by their context. Thus
the interplay of attacks within each note takes on a quite new
character through the reappearance of flutter-tongue flutes and
trumpets, plus rolled [rls and rapid alternation between consonants
in the voices, periodically abetted by bongos. The hectic tone im-
parted by this use of rapid repetition increases with each cycle of the
pitch series. By section Biii (F2) rapidly tongued winds have joined
in. At the start of statement C (G2) the three per:cussionists abandon
bongos for snare drums and underline the brass as they crescendo
into, or fiercely attack each note of the pitch cycle. The sforzandos
which in 0 King mapped out a macrocosmic version of the pitch
cycle are now applied to every note - initially by piano and/or
saxophone, eventually by multiple brass. Although, as before,
anticipations and prolongations blur one note into the next, there is
no extended or sophisticated use of selective resonance: any such
exercise would merely render the already dense texture opaque. The
unselective sustaining of notes that occurs in statement C is a
reflection of other harmonic processes and is best discussed with
them. There are likewise clear traces of some of the extraneous
pitches; but these, too, are best discussed in relation to their new
context.
Superposed materials
The harmonic materials that accrete around the pitch cycle are
summarized in example 24, excluding the numerous fragments
recapitulated from the third movement, the sporadic outbursts of fff
attacks from piccolo, percussion and brass, and during statement 0
the introductory fragments from the ensuing monody - all of which
will be discussed below. Black note-heads indicate elements that
sound only briefly against the specified note of the pitch cycle. As in

79
Playing on Words

the second movement, there is a steady increase in harmonic

density; but here the progression is dramatized by sharper distinc-
tions between each statement. Thus while statement A uses chord 1
and 2 from the first movement, statement B introduces other mainly
eight-note chords whose spacing and intervallic content are similar
to those of chord 1 and to the three terminal chords from the fourth
movement. Although statement C increases the harmonic density of
the accompanying chords, its most striking feature is the introduc-
tion of sequentially built chromatic clusters, in preparation for
statement D where cluster attacks alternate registers in a manner
reminiscent of the start of the third movement. Meanwhile, the
pitch cycle complements the processes at work in the commentary,
building aggregates in Cii and being presented throughout state-
ment D by the piano in parallel clusters.
Much of this material is related to the pitch cycle by the familiar
principles of common pitches and chromatic complementarity. The

Ex. 24

i ii iii

D D

c
n

D D D

80
 Synthesis and Dissolution

process is clearly seen in cycle A, where the b common to the pitch

series and chord 2 is constantly underlined. Apart from that b, chord
2 contains three of the five pitches excluded from the pitch cycle,
which it almost disrupts when, at DID, it is used to link the bs of Ali
and Aiii. Beneath the latter, the chord is extended by a series of
thirds down to d, producing a cross with chord 1, and then drawn
up again. Chord 1 shares both c# and b.with the pitch cycle; it is
associated with the former in its first two appearances and with the
latter in its last appearance.
The twe, pitches excluded from the pitch cycle that are not part of
chord 2, I # and e, reappear along with c at their original octave
placement from '0 King' during statements Band C. In each case
their presence is emphasised by low brass and woodwind (at F3, G4
and H). Apart from those and two brief recurrences of chord 2, new
harmonic materials invade the last three statements of the cycle. At
first, they contain several common elements: a b. bass and a middle
c and e recur throughout statement B; and an ostinato c#ld alterna-
tion runs from Bii to the end of Biii. The gil #alternation set up there
serves as a further unifying factor until Ciii; but by this time har-
monic constants have given way to cluster attacks - some having
one extremity in common with the pitch cycle, others not.
The quotation fragments from the third movement continue
throughout this section. An outline of their progress can be seen
from example 22; but a more detailed concordance is provided in
example 25. One or two deviations apart, they summarize the
progress of the third movement from the first appearance of the
Hindemith Kammermusik at the start of the movement to its re-
appearance at W. The only point at which the sequence is broken is
just after the start of the B cycle (E8), where a transposed Debussy
fragment comments upon the introduction to Mahler's col legno
reprise of the main theme, and the more extensive solo violin
quotation reveals an affinity by inversion with the familiar fragment
from Agon. As in the third movement, the fragments sometimes
share notes in common with their 'text', and sometimes pursue their
own tangential path. Meanwhile the fff attacks from brass, piccolo
and percussion consolidate their previous function as markers of
significant structural events. They announce each section of each of
first three statements of the pitch cycle, except for Bii and Ciii - the
latter being marked by an attack chord from the brass. By statement
D explosions abound, making such markers undistinctive.
The final monody
At H7, as statement C comes to an end, the Hindemith fragment
played by clarinet and piccolo is answered by oboe and cor anglais
with the opening descent from an undulating pitch sequence that is
to generate the final monody. The fragment recurs at I, extended to
form a complete undulation; and throughout the rest of statement D
further fragments appear between the punctuating chords (at IS and

81
Playing on Words

Ex. 25

Movement V Movement III

BHarp 5
B2 Vc All Vni Band U (=Movement I, F3)

B6 A12
B70b&Fl A20Vn solo
BBff A20
B9--13 U- U4
C Ott & Cl pice 87
03Vnsolo ClO
040b B5 (approx.)
07 C15
080b C16

09 Cl pice C8

011 Fag C23

012Vni D9 (reworked)

Dl2FI 022

EVni Dl9
E2 Cl pice E19
E4 G9 (reworked)

E6 GI5
E80b&C.i. E6
E9Str 16 and E2
F6 05 and 07
F7 W2
G2 W5

G3 W7

H3 W15

HIO [6

K6Vni B C3

82
 Synthesis and Dissolution

from 17 on - their derivation will be discussed below). Finally at J

the monody begins. Although modelled stylistically on its counter-
part in the first movement, it is based on a quite distinct pitch
process. Where the first movement recalled the virtual counter-
points, oscillations and tremolandi of Sequenza VI, this one estab-
lishes a 'wave-form' pitch set, covering the gamut of the treble
and bass staves - an idea that was to be further developed in
Bewegung (1971) and Eindrucke (1974).
It consists of three down-up undulations - here labelled a, band c
- which appear in three progressively longer forms set out as 1, 2
and 3 in example 26. In their various occurrences, each undergoes
minor modifications which are set out on separate staves below or
above the 'principal' version. 1 is first stated complete from JI-5. 2a
follows in bars J6 and J9, interspersed with other materials to be
discussed below; but the f c# of what should be the start of 2b at J12
instead short-circuits back to 1, followed by a full statement of 2
from JI4-16. It is directly followed by a first statement of 3 (J16-Ll);
and like the first statement of 2a at J6-9, it is a fragmented one. But
from the first entry of 3c at L the impetus is continuous: first a final
restatement of 2 (Ll-4), modified to bring its pitch content into
closer line with 3; and then two restatements of 3 (L5-9 and L9-M4)
that lead into a reworking of the final bars of the first movement,
discussed below.
No simple additive process leads from version 1 to version 3.
Comparing section b, and the lower extremity of a in each version, it
can be seen that although both 1 and 2 are contained in 3, 1 is not
contained in 2. In any case, it is 3 which displays the most striking
internal organisation. 3a combines a retrograded pitch sequence
with an inverted interval sequence - each accommodating the other
by interpolated notes. (Only two notes belong within neither pro-
cess: the bottom a which creates a retrograde inversion of the three-
note peak at the end of 3a and 3b; and the g of the descent, whose
insertion creates a c# g e eb complement to the subsequent e ~a f df.)
3b contents itself with a simpler rationale, transposing two pitch-
class sequences by an octave to create a satisfyingly varied retro-
grade. With 3c, internal cogency gives way to allusion: the descent is
the first movement's chord 1, with an added g# to spell out the
solitary major third, and a preceding g to complete the higher
'diminished seventh'. The ascent combines elements of chord 2 with
an allusion to the corresponding moment in 3a.
Version 2's simpler profile allows for further harmonic allusions,
albeit in a form that is not so explicit. The descent of 2b from g to the
Q.
low again contains elements of the first movement's chord 1, now
transposed down an augmented fourth, while the ascent from that
a~ to db [ I a b is derived from the opening chord of the fourth
movement, transposed down a minor third, and with first and
second alto parts replaced by a solitary f. The structural simplicity of
3b relative to 3a is explained if it is viewed as an elaboration of the

83
 £

'E
.~

~
§
ce

U)
Ex. 26
~

i!l
..
IJ9 IJ9
IJ9
J1·5

IJ9

J12-13

IJ9 IJ9
IJ9
J6 IJ9
J 14-16

IJ9

L1·4

K2-4
J 16-17

3a
L5 La
only only

3b

L1
only

3c
only

only
~
 Synthesis and Dissolution

harmonically allusive 2b. But 2a and 2c seem to be straightforward

reductions of the corresponding parts of 3. Version 1 is an even
more drastic reduction of 3 - so much so, that it can afford to create
its own internal logic, most notably in the intervallic retrograde of
la, part of which then serves to link the ascent of la to the descent
of lb. The f d. [ I c of Ib's ascent stands in an analogous re-
lationship to the g [ I f~ c of lc's descent; and a further two-
interval module unifies the descent of la when f c~ e is followed by
e. b [ I d.
The fragments that announce the onset of the monody are all
straightforwardly derived. As was noted above, those at H7 and I
derive from la, as does a further fragment at IS (clarinet). It is
followed at 17 (oboe) by the start of 2a, and at 18 by a composite
figure derived from the peak between Ib and lc, modified to
prepare for the fic# oscillation that starts the first full statement of 1.
As with its predecessor in the first movement, the momentum of
the monody is at first checked by a series of pauses - here filled with
vocal duets which recall the idiom of the soprano's opening solo
(and indeed begin and end with references to it) but employ pitch
materials from the monody. In the first hiatus at JS, the final c b
from 1 is simply answered by part of the initial descent from 3a,
passing from saxophone to first alto - a gesture already familiar from
b.8-9 of the movement. But at J7 the duetting begins: first T I and
B 1, both basing themselves on 2b; then A 1 mixing la with 2a trans-
posed up an octave, and T 2 initially retrograding 3a. The monody
gathers momentum, but is again interrupted at J17, where T 1 com-
bines la and 2a, while B 1 anticipates the continuation of 3a. Finally,
at K4, oboe and cor anglais play the two parts of 3a simultaneously,
thus emphasising the pitch retrogrades of example 26, and the
sopranos echo their final notes as background to a last Mahlerian
reminiscence. S 1 extends this into a variant upon her opening
phrase from this movement, whereupon the monody takes over
with uninterrupted impetus.
At first, the monody's harmonic clarity is clouded only by a back-
ground of glissando chords from the strings (whose internal struc-
tures recall - but do not reproduce - that of chord 4 from the first
movement, d. ex.3). Brass sforzandi, carried over from the final
statement of the '0 King' pitch set, underline individual notes -
though not to pick out any pitch sequence of autonomous signifi-
cance. However, from L onwards a process of thickening analogous
to that of the first movement monody begins: first octaves, then
fifths (L7, oboe), then thirds (L8, flute 1). At the same time, exten-
sive cluster-based chords similar to those underlying the latter two
statements of the '0 King' pitch set emerge: first in the background
at L3 (strings), and then dominating the texture from L6 to M,
where the configuration in brass and strings 'takes off' in parallel
motion to the monody. The monody itself begins to generate its own
harmonic backdrop from L9 on, where individual fifth-spaced in-

85
Playing on Words

strumental couples sustain their notes until the monody again rises
to catch them up into the general melee.
At the final semiquaver of M4, the monody flows into a rework-
ing of the final, climactic bars from the first movement (d. I L).
Within the climax itself, the pattern of pauses between outbursts is
rearranged, and two of the individual gestures are repeated. As a
result of the reworked pattern only a portion of the piano part is
retained. But the voices' final statement of chords 1 and 2 is sub-
stantially lengthened, with the oscillating flute answered as at the
start of the movement, though this time by oboe and piano. Along
with the pizzicato bass line of the last five bars, they provide the
complementary pitches to the long-held final chord, 2, which is
passed from one instrumental choir to another in a series of quiet
attacks. But to the end impetus is maintained by three snare drums
and high strings. As in the first movement, three gongs signal the
end.
Text and music
The relationship between text and music in this movement is, in
outline, simple. Fire and water materials alternate from b.5 up to the
final notes of statement A of the pitch cycle (ES). From after the
fermata at E2 up to the end of the '0 King' cycles at J, the voices
concentrate on materials from the four interrelated myths shown in
example 1: at first the M.21M.125 relationship, which dominates the
second statement of the cycle; and thereafter, from G, the relations
between M.124 and the 'missing' M.1, which occupy the third and
fourth cycle. As the final monody enters at J so, too, do new myth
materials that focus irrevocably on the underlying theme of the
whole work: 'la vie breve'.
Other materials are added to complicate this straightforward
pattern. There are occasional interjections derived from the Beckett
text of the third movement. But the most notable and idiosyncratic
of these additions is a series of four analytical observations taken
from U!vi-Strauss's text and rewritten by Berio so as to apply to
music. They are set out below:

Berio Levi-Strauss

Partiel ou provisoire,
.. Partiel et provisoire,
ce dernier commentaire I'ebauche de synthese OU nous
a mene la deuxieme partie
n'est pas convain~ant, n'est pas absolument convain~ante,
car iI laisse de cote car elle laisse de cote
d'importants aspects d'importants fragments
de nos themes. (BB-14) du my the de reference, (p.155)

86
 Synthesis and Dissolution

p
Mais pourtant les themes sont la, Pourtant les mythes sont la,
qui affirment la priorite qui affirment la priorite
de la discontinuite universelle de la discontinuite universelle
des themes sur la continuite des especes sur la continuite
de I' organisation interne interne du chromatisme
a chacuns. (D7-9, F2-G3) particulier a chacune. (p.330)

Avant de terminer
d'une fa,on provisoirement
" Avant d'en terminer
provisoirement
definitive (II Vienne on dit avec les codes sensoriels,
'definitivement provisoire')
il faudrait resoudre il est indispensable de resoudre
quelque contradiction (E2) une contradiction. (p.l68)

Partout, ailleurs, les themes

inversent la valeur
" inversent
Partout ailleurs, les codes sensoriels
regulierement la valeur
de leur termes de leurs termes,
selon qu'il s' agit selon qu'il s' agit
de retarder la mort de retarder la mort
ou d' assurer la resurrection (E2) ou d'assurer la resurrection.(p.171)

This curious procedure might lead one to suppose that, like cer-
tain of the Beckett quotations in the third movement, these are to be
used to comment on specific musical events. Some of them can be
interpreted in that way - notably IX , which is generically appropriate
within this context. Passage p was originally a discussion of how
different bird feathers are recognized as belonging to specific species
rather than being allotted a place in the colour spectrum; and in its
new version it might be taken as a comment on the various quota-
tions from the third movement, which listeners will identify accord-
ing to their source rather than in terms of common technical
features. But " and lJ, which appear simultaneously at E2, are
frankly indecipherable in musical terms. One is inclined to conclude
that if Berio was intent upon working at the limits of comprehensi-
bility, then these fragments of analytical mumbo-jumbo were placed
deliberately beyond it.
Fire vs. Water
The insertion of these materials clouds an otherwise simple process.
Materials representing the fire myth M.9 - in all cases but one the
familiar 'appel bruyantl doux appel' - alternate with the other myths
to which it is related by homologous triads (see above, p.ll). One of
these, Levi-Strauss's 'mythe de reference' M.l, was notably absent
from the first movement; and it is this which first alternates with
M.9, though always presented in tandem with its familiar trans-
formational opposite, M.124. They emerge from the phonetic under-
growth at b.13, as chord I from the first movement establishes itself;
but they manage only a few words before they are cut off. They are

87
Playing on Words

even further pruned when they reappear at 89, and thereafter

temporarily retire. Meanwhile the voices pursue an encapsulated
recapitulation of those materials in the first movement (I D-HS)
related by other homologous triads, though the 'eau celestel
terrestre' opposition that divided them is now absent. This runs
from 812 to E, and completes M.9's alternation with the various
water myths.
Berio now isolates a single opposition ('un fils prive de merel
nourriture') from Levi-Strauss's table analysing relations between
M.2 and M.12S,3 a selection from whose contents had been recited
at a gallop in the first movement; and he develops it as an extended
lyric setting. This starts at E4 and occupies the whole of the second
statement of the pitch set from the second movement.
M.l vs. M.124
From G to H the voices start a simultaneous narration of the
analogous openings of M.l and M.124: in M.I a single young man,
in M.124 the elder brothers of Asare, each rape their mother and are
punished by their father. Berio then proceeds to a simultaneous
comparison of other features from the two myths. In each, the hero
is protected from a potentially mortal encounter (in M.I with the
arrows of maleficent 'souls', in M.124 with crocodiles) by three
helpful birds or animals. Each episode is summarized in numbered
succession. Berio then enumerates other points of comparison:
fourth (11-2), that both heros hunt lizards (with disastrous results:
from one batch of lizards is born the pursuing crocodile (M.124),
while the other decomposes, attracting vultures who eat the hero's
buttocks); fifth (15-7), the unexpected results of immersion (Asare's
brothers, after bathing, are changed into stars, whereas the vengeful
father of M.l's hero, finally thrown into the lake by his son, is eaten
by fish, his lungs turning into aquatic plants); and sixth (110),
helpful relations (though since Asare's uncle is a skunk, the associa-
tion between what appears to be urination and what appears to be
phallic symbolism is yet another example of semantic proliferation
from isolated fragments).
'La vie breve'
As the pitch set comes to a halt, at 110, so too does the com-
parison between M.I and M.124. In the final monody the voices
move on to new materials but a long-familiar theme. At J the second
soprano begins a fresh narration, that of M.B7. This tells of how
cultivated plants, like the fire of M.9, must be paid for by 'Ia vie
breve'. She stops almost immediately to give way to another myth
whose origins have yet to be traced (it does not appear in Le cru et Ie
cuit nor in any other part of Mythologiques) but whose theme is
clearly also that of human mortality. Like the myth narration at the

3 Levi.Strauss 1964, p.215.

88
 Synthesis and Dissolution

start of the first movement, it is interrupted at the crucial moment,

just as we are about to learn what consolation the 'esprit special' is
to bring to mortal man. The voices throw themselves into a frenetic
exchange of mainly plosive consonants before resolving into a final
statement of the image underlying the whole work: 'peripetie':
'heros tW?'.

89
 6
Epilogue

Any score provokes two divergent aesthetic responses. The first,

more readily acknowledged by tradition, is that of perceiving
sounds structured in (and lost in) time. The second is the aesthetic
pleasure of exploring the score itself: a counterpoint between visual
pattern, knowledge of the rules of the game, and the sometimes
hallucinatory suggestions of the inner ear, played out within a time-
warp where the synchronic can become diachronic, and vice versa.
As far as contemporary composers are concerned, the divide
between these two aesthetics becomes most obvious in the examina-
tion of musical process and structure, for in many contemporary
works there is a wide discrepancy between aural impressions of
informality and the meticulous logics of the score. But although
there is more in the harmonic and rhythmic organisation of Sinfonia
than meets the ear, the gap here is not a wide one. Berio's choice of
musical language is deliberately simple and graphic, and it operates
within a circumscribed range of technical concerns. Yet the verbal
materials, with the exception of '0 King', not only mark out a wider
discrepancy between the listener's ear and the score-reader's eye,
but deliberately underline a further gap between those things that
become obvious from looking at the score and those that can be
made clear only by referring beyond it. This coupling of vivid,
directly comprehensible musical gestures and partly baffling verbal
fragments is entirely characteristic of Berio. It reverses the situatiGn
all too frequently found in contemporary music of a linear, coherent
text that holds together a rather loose agglomeration of musical
ideas. For Berio uses explicit musical processes to establish
continuity and security, while allowing words, half grasped, to
provoke a state of aesthetic risk where intuitive sense mayor may
not emerge amongst the isolated images of a semantic non-sequitur.
In thus transforming the playing upon words so dear to Joyce in
Finnegans Wake - making layer upon layer of superposed material
sound together around a musical or verbal core - Berio reactivates
one of the central experiences of the 'modernist' tradition. The
listener, troubled by the rich confusion of what he has heard, may

90
 Epilogue

well seek refuge in the score; but there he will discover a maze of
allusions to things beyond the score. The more avidly he seeks to
pin these down, the more clear will it become that there is no logical
end to his activities. But this moment of scholastic exasperation
(richly familiar to students of Joyce, or Pound, or Borges) serves to
underline the necessity of coming to grips with that initial confusion
in another, and complementary way - that of learning to be
receptive to the peculiarly vivid aesthetic impact of the half-
understood. Seen thus, the ellipses and allusions of the 'modernist'
tradition, and the verbal and musical superpositions of Berio's own
work offer not gratuitous mystification, but a survival kit against the
facile nihilism that so easily informs attempts to analyse a disjointed,
relativistic environment in which 'the unexpected is always upon
us'. The more determinedly a study such as this seeks to track down
allusions, to trace formal processes, the more clamorously does it
evoke the necessity for its complementary mode of perception. The
exposition of art as artefact is a necessary moment of demystification
- for by suggesting how it was put together, it challenges you to
consider what it would be like to do something similar - but it is
only useful if it can serve as a springboard back to the limits of the
comprehensible, where many things sound at once.

91
 Bibliography

Altmann, Peter. Sinfonia von Luciano Berio: eine analytische Studie. Universal
Edition, Vienna, 1977.
Bauer-Lechner, Natalie. Recollections of Gustav Mahler. Translation by Dika
Newlin of Erinnerungen an Gustav Mahler (Leipzig, 1923) with notes by
Peter Franklin. Faber Music with Faber & Faber, London, 1980.
Beckett, Samuel. The Unnamable. Translation, by the author, of L'innommable
(Paris, 1952). John Calder, London, 1958; new edition by Calder & Boyars,
London, 1975.
Berio, Luciano. 'Poesia e musica - un 'esperienza'. Incontri musicali: Quaderni
internazionali di musica contemporanea, iii (1959), 98ff. French translation
in Contrechamps, i: Luciano Berio (1983), 24-35.
- - . 'Meditation on a Twelve-tone Horse'. Christian Science Monitor, 15 July
1968. French translation in Contrechamps, i: Luciano Berio (1983), 4~50.
- - . Two Interviews. Translation by David Osmond-Smith of Intervista sulla
musica (Bari, 1981). Marion Boyars, London, 1985.
Budde, Elmar. 'Zum dritten Satz der Sinfonia von Luciano Berio' Die Musik
der sechziger Jahre: zwiilf Versuche, ed. Rudolf Stephan, 12S-44. Schott,
Mainz, 1972.
Dressen, Norbert. Sprache lind Musik bei Luciano Berio: Untersuchungen zu
seiner Vokalcompositionen. Gustav Bosse, Regensburg, 1982.
Flynn, George W. 'Listening to Berio's Music' The Musical Quarterly, Ixi
(1975), 388-421.
Hicks, Michael. 'Text, Music, and Meaning in Berio's Sinfonia, 3rd
Movement'. Perspectives of New Music, xx (1981-2), 199-224.
International Phonetic Association. The Principles of the International Phone'ic
Association. I.P.A., London, 1949.
Jahnke, Sabine. 'Materialien zu einer Unterrichtssequenz: Des Antonius von
Padua Fischpredigt bei Orff-Mahler-Berio'. Musik und Bildung, v (1973),
615-22.
Krieger, Georg, and Wolfgang Martin Stroh. 'Probleme der Collage in der
Musik, aufgezeigt am 3. Satz der Sinfonia von Luciano Berio' Musik und
Bildung, iii (1971), 229-35.
Levi-Strauss, Claude. Le cru et Ie cuit. Pion, Paris, 1964. English translation
by John and Doreen Weightman as The Raw and the Cooked. Jonathan
Cape, London, 1970.
Lyons, John (ed.). New Horiwns in Linguistics. Penguin, Harrnondsworth,
1970.

92
 Bibliography

Mitchell, Donald. Gustav Mahler: the Wunderhorn Years. Faber & Faber,
London, 1975.
Nattiez, Jean-Jacques. 'Rencontre avec Levi-Strauss'. Musique en jeu, xii
(1973), :>-9.
Osmond-Smith, David. 'From Myth to Music: Levi-Strauss's Mythologiques
and Berio's Sinfollia'. The Musical Quarterly, Ixvii (1981), 230-60.
- - . 'Joyce, Berio et I'art de I' explosition'. Contrechamps, i: Luciano Berio
(1983), 8:>-9.
Ravizza, Victor. 'Sinfonia fur acht Singstimmen und Orchester von Luciano
Berio'. Melos, xli (1974), 291-7.
Ruwet, Nicolas. Langage, mllsique, poesie. Editions du Seuil, Paris, 1972.
Sanguineti, Edoardo. 'Laborintus II'. Contrechamps, i: Luciano Berio (1983),
75-82.
Stoianova, Iwanka. 'Verbe et son, centre et absence'. Musique en jeu, xvi
(1974), 79-102.
Tibbe, Monika. Uher die Verwendung von Liedern und Liedelementen in
instrumentalen Symphoniesiitzen Gustav Mahlers. Emil Katzbichler, Munich,
1971.

93
 Index

amplified Yoices, 2, 4 Bruckner, Anton, 43; Symphony No.4, 41,

Anthony of Padua, St, 7, 40, 43, 53 48,49
Amim, Ludwig Achim von: Des Knaben
Wunderhorn,40 chord spacing, 73, 80
Auden, Wystan Hugh, 4 chromatic complementarity, 1~18, 20, 33,
73,81
Bach, johann Sebastian, 48, 66; chromatic proliferation. 32. 60--62, 73
Brandenburg Concerto No.1, 50, 66 clusters, 19,39,44, 5()"'53, 5S-9, 6:>-4, 6S-9,
Bauer-Lechner, Natalie, 40, 43, 54 70, 77, 80, 81
Beckett, Samuel: Molloy and Malone commentary procedures: harmonic, 21,
meurl, 55; The Unnamab/e, 7, 53-71, 39-53,74; rhythmic 30, 34
8(",,7 correspondence between voices and
Beethoven, Ludwig van: String Quartet instruments, 16, 47, 62
Op. 131,39 n.l; Symphony No.6, 53,
57,69; Symphony No.9, 48, 61; Violin Dante Alighier;, 7, 9, 55
Darmstadt School, 48
Sonata Op. 96, 41-3
Berberian, Cathy, 8 death imagery, 5, (""7, 12, 55, 7L 86, 8s-9
Debussy, Claude, 39, 47; La M,.... 7, 46, 53
Berg, Alban: Violin Concerto, 48, 52, 59;
Wozuck, 7, 44, 46, 48, 53, 67-8 56, 58, 6~3, 70, 78, 81
Berio, Luciano: A-Ronne, 56 n.28; displacement processes 23-5, 2~9
Alle/ujah, 2; Allelujah 11,2-3; Allez-Hop, distortion: melodic, 47, 60, 64. 67-9;
1; Bewegung, 83, Chemins, 4. 39, 53; metric, 4-1-0, 64, 6(",,71; rhythmic. 47,
Chemins I, 2, 3, 5; Chemins 11, lib and 11/, 60,62,68
5; Circles, 1,8; CorD, 1; Eplfanie, 1, 3-4; dynamic differentiation within chords,
Laborintu5 II, 2, 4, 6, 9; 'Meditation on a :>-4, 6:>-4, 70
Twelve-tone Horse', 53; Nones, 2, 4; 0
King, 6, 8, 21-37, 74; Opera, 1. 5; Eliot, Thomas Stearns, 9
Passaggio, 1, 4, 56 n.28; Sequenz"" 4;
Sequenz a II, 5; Sequenz" 11/, 1,8,55; fire imagery, 11-12, 86-8
Sequenz a lV, 6; Sequenza VI,S, 19,83; fragmentation of texts, S-9, 13-14,55
Still, 4; Them", Omaggio a Joyce, 8;
Quad"ni I-Ill, 2, 3; Questa vuol dire cite, Ghedini. Giorgio, 2
5; Visage, 56 n.28 Globokar, Vinko: \ '(1/1',70
Berlioz, Hector, 47; Syrnphonie Goethe, johann Wolfgang YOn, ~O
fantastique. 4S-50, 54, 59, &1
blood imagery, 13,74 harmony: third-based, 6, 15-17,20,73,
Bohemian pipers, 43 81; as proCE"ss. 17. 31-:!. 3-1.80; large-
Borges, Jorge Luis, 90 scale aggregates, .>-1.19, 5()"'53, 63-4;
Boulez, Pierre, 6 n.5, 48, Ph selon ph, 71 see also Chromatic and Clusters
Brahms, johannes, 48; Symphony No.4. Hindemith, Paul: KII1llma1Plusik Nr.-l. -Ih,
66; Violin Concerto, 48, 59 4S-9, 56, 5S-9, 69, 78, 81
Brecht. Bertolt: DDS Badenrr Lehrstuck, 53
n.22 [nternational Phonetic Association. 35
Brentano, Clemens; Des Knaben
Wunderhorn,40 jazz. 2

94
 Index
Jones, Daniel, 35 reduction, 39, 43-6
joyce, james, 4,55,59,90; Finnegans resurrection imagery, 7, 71
Wake, -1-5, 53, 90; Ulysses, 8,14,55
Sanguinet;, Edoardo, 4, 9
King, Martin Luther, 6, 7, 74 Schoenberg, Arnold: Fun! Orchesterstiicke
Op.1b, 7, 50, 53, 57--S, 67, 70
layering: of musical materials, 3-4, 5, Schumann, Robert: Dichterliebe, 41-3, 54
1S-19; of verbal materials, 04-5 Second Viennese School, 39
Levi-Strauss, Claude: Le au rt Ie (!lit, 5, 7, selective resonance, 21. 31-2
9-15,75,77,86-8; L'ho"'mc n", 9 n.5 sets: governing pitch, 17-19, 21-26, 76,
78,80,81-5; governing rhythm, 26-9,
Maderna, Bruno. 2 78-9
Mahler. Gustav, 39; Des Antollius von sforzando attacks. 4.17,23,78,79,81,85
Pad"a Fischpr<,digl, 7, 4(}-3; Symphony solfege. 58-9, 66
No.2, 6, 7,39--54. 56-71, 7~. 78. 81; Stockhausen, Karlheinz, 48; Gruppt'n, 71
Symphony NoA. 50, 58; Symphony Strauss, Richard, 39, 47; Der Rosenka1.lalier
No.9, 48. 61 46.48. 50. 65. 71
Mayakovsky, Vladimir, 70 Stravinsky, Igor, 47; Ago., 48, 56, 62--1,
Milan Conservatory, 2 68,81; Lr Saar du prinlemps. 46, 48, 56,
Monteverdi, Claudio; L'Orfeo 5 60.62
myth materials. see Levi-Strauss Striggio, Alessandro, 5
Studio di fonologia, Milan, 2
Nattiez, jean-jacques, 13 Sweelinck, Jan Pieterszoon: Variations
New York Philharmonic Orchestra, b n.5 'Mein junges Leben', 71
Swingle Singers, 2, 4, 6 n.5
octave displacement, 47, 60, 64, ~9 Swingle II, 6 n.5
orchestra (disposition of), 2-4
Orchestre National de France, 6 n.5 text, sources of. 9--11, 54-5, 74. 86-q
ostinato, 46,57, 59, ~9 texture, 3, 16, 74
third-based chords, see Harmony
permutation: of phonetic materials. 35-7; Titanic, sinking of, 5
of pitch materials, 17-19 troping (phonetic), 37
phonetic alphabet, 8,34
phonetic materials, 8, 13-15, 34-7, 74 Valery, Paul: Le crmetiere marin, 55, 56, 70
pitch <ydes/sets, see Sets virtual counterpoint, 19
Pound, Ezra, 9, 90 vocal resources, 1,55
Pousseur, Henri: CouleuTs crois/es, 66 vowel rotation, 35-6

quotation, 39, 41-4, 47-50 water imagery, 6-7, 9--10, 53, 58, 67--S,
86-8
Ravel. Maurice, 39, 47; Daphnis el Chloe, Webern, Anton, 48; 11. Kanlale, 71
48,60; La Valse, 46, 48, 50, 54, 59-61.
64-5

95