Francesco Giomi, Damiano Meacci,

and Kilian Schwoon

Live Electronics in
Luciano Berios Music

Luciano Berio has made signi cant contributions

to the development of electronic music, from the
Studio di Fonologia Musicale of the RAI in Milan,
to IRCAM in Paris, to the Italian Centro Tempo
Reale, founded in Florence in 1987 with the objective of creating a structure in which to investigate
the possibilities of real-time interaction between
live performance and programmed digital systems.
Tempo Reale is today a center of musical production, research, and education where many major
Italian composers, in addition to Berio himself, create their works and where important works from
the electroacoustic repertoire are performed.
In this article, we focus on three compositions by
Berio that are not only particularly signi cant in
themselves but also represent an important testimonial to the production activity of the center in
recent years: Outis, the azione musicale composed
in 1996 for La Scala theater and then performed
again in 1999; Ofan`m (19881997) for female
voice, two childrens choirs, two instrumental
groups, and live electronics; and nally, Altra voce
(1999), a chamber music composition for mezzosoprano, ute, and live electronics. The electronic
parts of these compositions were totally revised beginning in 1999, and various updated musical
technological contributions have been presented in
their recent performances. They are in any case
works that, ranging from musical theater to chamber music, clearly exemplify Berios approach to
composing for live electronics in different contexts.
This article discusses the problems involved in
this approach within the context of the musical
technological solutions designed and implemented
by Tempo Reale for the live performance of the
three works. The techniques adopted are rst described in relation to their use in music; they are
then viewed within their proper context in a more
detailed description of the three speci c works.
The performances in question are the most recent

ones for each composition, namely Outis in Milan

(September 1999, La Scala Theater) and Paris (November 1999, Theatre du Chatelet), Ofan`m in
Milan (October 2000, Sala Verdi), and Altra voce
in New York (March 2001, Carnegie Hall), Tokyo
(October 2002, Kajimoto Music), and Rome (March
2003, Renzo Pianos Auditorium).

Centro Tempo Reale

Villa StrozziVia Pisana 77
50143 Florence, Italy
{fg, dm, kilian}@centrotemporeale.it

Computer Music Journal, 27:2, pp. 3046, Summer 2003

q 2003 Massachusetts Institute of Technology.

30

Technology Serving Music

One of the major features of Berios work is undoubtedly his search for continuous mobility in
music. The possibilities of physical movement of
sound offered the composer by new technologies
are many, involving, for instance, the trajectories
followed by sound events through space, continuous modulation on harmonic and dynamic levels,
and various types of proliferations of sound layers.
But what truly interests Berio are not these situations in themselves, but rather the relationships
that are established between such physicalacoustic
sound mobility and the effective mobility of the
musical thought.
Another fundamental aspect in the use of technology is what Berio himself de nes as adaptability of the musical thought to different spaces and
listening situations. Computer technologies and
those of sound diffusion allow the composer to inhabit new, unconventional acoustic spaces, as well
as to open up and render exible spaces that are
typically closed (such as traditional theaters) and
often strongly linked to standard modes of music
presentation.
Analytical Listening: Principles of Ampli cation
and Sound Spatialization
The simple ampli cation of a sound should be
viewed, according to Berio, in the light of its musical consequences:
Our listening is today conditioned by topquality recording. A recording of this type alComputer Music Journal

ways has an analytic nature and therefore my

ideal is that of creating a type of acoustics and
of sound that is typical of a great recording studio. (Berio 1999b)
This consideration implies a series of consequences for the entire electroacoustic chain, from
microphone techniques to sound diffusion systems.
These decisions must, however, take account not
only the maximum possible quality of the sounds
reproduced through the loudspeakers but also the
interaction between these and the natural sounds
of instruments and voices:
Technologies interest me when they become
an extension of human work, of the performance, the sound, the voice, but not as an end
in themselves, not as effects. In fact, I have
no interest in searching for new sounds. A new
sound is created when there is a new musical
idea to produce it. What interests me is extending the possibilities of instruments, of the
voice, but in an organic manner, which is not
in con ict with the sources of sound. (Berio
1999b)
This concept differs profoundly, for example,
from the traditional ampli cation of a rock-and-roll
concert. In fact, here it serves not so much to increase the loudness of the sound, but rather to
modulate the natural acoustics of a space. This
calls for great homogeneity between the natural
and the ampli ed soundhomogeneity that can be
achieved only by analyzing the acoustics of each
individual space and adapting accordingly all of the
parameters important to the spatial perception of
the sound: the number of sound loudspeakers used
for ampli cation and sound spatialization, their positions, equalization, delaying systemseven slight
reverberations to avoid excessive superimposition
of the more directional sound emanating from the
loudspeakers over the natural one.
One of Berios special interests is the potential
for mobility in space of the ampli ed sound. Already in his works without electronics there exist
many examples in which the positioning of the performers in space is used as a compositional parameter. It might almost be said that this is a constant

factor in his catalogue, considering works such as

Allelujah II (19561958), Sinfonia (19681969),
Coro (1974), and Formazioni (19851987). The musical use of space can serve very different functions.
The modi cation of the listening perspective and
the determination of a multiplicity of sound levels
in continuous transformation are among its most
important aspects. This is exempli ed by the third
group of violins in Sinfonia, placed behind the orchestra, and the two harps in Formazioni positioned at the two far ends of the stage, as close as
possible to the audience. A spatial organization of
this type leads to greater acoustic transparency,
whereas in other cases it is possible to suggest a
different mode of perception, capable of rendering
listening ambiguous and orienting it toward a
greater fusion of timbres. This happens in Coro, for
example, where a group of 40 singers and 40 instrumentalists (not counting the keyboard and percussion instruments) is divided into 40 pairs, each
composed of a voice and an instrument in the same
register, thus contributing to the creation of a single quasi-hybrid ensemble of voices and instruments. As a result, the organization of space in
music results at times in a greater acoustic transparence, at times in a more tight-knit perceptive
unitycharacteristics that even without electronic
means confer a more analytic perspective on listening.
The effect which in the orchestral writing is
achieved through static dispositions can acquire a
new dimension of mobility through electronics: by
arti cially moving different sound structures closer
or further away in space, it is possible to continuously modulate the degree of af nity between
sound layers. This principle holds true for the interaction between natural sound and ampli ed
sound as well as between different electronic layers, but can also be emphasized through suitably
differentiated systems of reverberation serving to
vary the degree of perception of each sound source
by the listener. The trajectories of the movements can also become signi cant, as is already the
case in Berios compositions without electronics,
where at times he creates virtual movements of
sound. Formazioni is, for example, dense in situations where identical notes or similar gures pass
Giomi, Meacci, and Schwoon

31

between groups that are similar in timbre, but

separated in space, thus creating the illusion of a
shifting of the sound. But it is expressly in works
such as Ofan`m that the sound trajectories, created
arti cially, can acquire further signi cance, as will
be seen in the nal part of this article.
In the works featuring electronics, many parameters of the sound spatialization or transformation
systems are not established in detail by the composer. However, the sound locations are xed in
relatively precise manner in an electronic score
thanks to a notation that de nes sequences of con gurations of loudspeakers. (This term electronic
score will be used repeatedly in this article: in addition to the normal, traditional score, Berio constructs an annexed document containing a general
description of the points where electronics intervene and provide cues, including also the spatialization of sounds.) In this regard, the traditional
score is enriched by a series of pointers that indicate the starting and ending points of the cues and
of the groups of instruments progressively involved
(see Figure 1). Accordingly, the electronic score is
not a score in the traditional sense, but rather it
constitutes a list, in symbolic form, of the general
instructions to be carried out for each individual
action of electronics.
In the diagram in Figure 2, an example of sound
spatialization can be seen. Each box represents a
certain loudspeaker, while the numbers indicate
the order in which each con guration appears for
the rst time. To each diagram corresponds a de nition of holding times tp and movement times tm
(i.e., of the passage from one con guration to another). The sequence proceeds the rst time in the
order in which it is written, after which various
types of permutation may succeed each other in
the electronic score. The sequences may stop at the
last con guration, or begin again from the start
(loop option).
For these three works in particular, but also in
general for all of the latest works realized by
Tempo Reale, a general-purpose system of sound
spatialization called Smov (Sound Movement) has
been developed that, while retaining basically the
same structure from one piece to another, allows
the realization of compositions where the arrangement and number of loudspeakers for spatialization
differs substantially. The system can be applied for

a variable number of sound sources to be moved. It

consists of a series of engines which operate on the
amplitude of the audio signal, redirecting it toward
a variable number of outputs according to a series
of musical models of movement. In the works examined here, predominant use is made of sequences (path between sound loudspeakers selected
with determined times of stasis and of movement)
and some types of random algorithms (random
times and localizations, weighted randomness on
the localizations, permutation of localizations and
time). The reverberation parameters associated
with each sound source (natural or transformed) are
controlled separately by the module for general
management of events described in the following
sections.

32

Computer Music Journal

Rami cations and Accumulations:

The Use of Harmonizers
The creation of a homogeneous path between
acoustic sources on the one hand (voices and instruments) and electroacoustic sources on the other
(live electronics) is undoubtedly one of the main
concerns of Berios recent composing with electronics. In this sense, his decisions are oriented toward
processing algorithms that do not radically transform the sound but are akin to the procedures of
vocal and instrumental composing. This is certainly
true of the use of harmonizers, algorithms that
transpose the pitch of the input signal a certain interval, thus adding to the input one or more transposed signals. Berio is well aware of the possible
hazards of this type of processing, consisting primarily in the rigidity of a xed transposition, at times
resulting in parallelisms that may easily become
annoying. To render the perceptive effect of harmonizers more mobile in time, he often uses asynchronous time sequences of intervals, describing
them in detail in the electronic score (see Figure 3).
Fixing the tempo in such a precise manner may
seem like excessive control, but in reality, owing to
the freer performance times of the interpreters, the
result is never mathematical. Rather, it consists
of a certain indetermination in the effective harmonic results that break away from the algorithm
in time. To give even greater range to the use of

Figure 1. Excerpt from the

score of Outis. (Copyright
1996 by Ricordi-BMG. Used
with permission.) The electronic processing refers to
woods (cues 911) and
brasses (cues 10 et seq.).

Giomi, Meacci, and Schwoon

33

Figure 2. Excerpt from the

manuscript electronic
score of Altra voce containing a description of the
sound spatialization (cue
24). (Copyright 1999 by
Universal Edition A.G.
Used with permission.)

Figure 3. Harmonizer
sequence in an excerpt
from the manuscript electronic score of Outis (cue
4). (Copyright 1996 by Ricordi-BMG. Used with permission.) In this score, in

(2)

this technique, Berio also requests continuous

modulation of the relationship between the original
signal and the transformed signal. As a general
principle, their contributions should be practically
the same (as with two orchestral groups, for example). However, the relationship between the levels
should never sound mechanical but constantly in
motion.
The harmonizer is used mainly in two types of
musical situations, the rst linked to the concept
of heterophony, the second to that of sound
masses. In the case of heterophony, a melodic line
appears simultaneously with other lines derived
from it. This principle can be easily applied by superimposing on a line played live its transformations from the harmonizer. The example appearing
in Figure 4, taken from Altra voce, shows the part
of the mezzo-soprano line with one of the possible
results of its electronic transformation.
In other situations, characterized instead by the
presence of sound masses, the harmonizer serves to
augment the vertical density of structures already
very rich harmonically. This is the case, for example, of Section V of Ofan`m: in the harmonic excerpt shown in Figure 5, the results of the
transpositions are not explicitly indicated, but it is
instead possible to see the temporal development
of the harmonic shifting in relation to the chords of
the rst of the two instrumental groups.

addition to the holding

time tp , Berio species the
movement time tm , which,
in the case of time values
other than zero, refers to a
gradual passage between
two transpositions (glis-

sando). In the scores of

Ofan`m and Altra Voce,
this element is omitted
and is considered always
equal to zero.

(3)

In the relationship between instrumental sound

and transformed sound, delay and sampling are of

crucial importance. The so-called real-time

freeze processes are one of the most distinctive
features of the newest versions of Altra voce and
Ofan`m. This involves the real-time recording of
selective fragments of the performance in process
and their reappearance in loops at subsequent
times shown in the score. In some cases, the duration of the recorded fragment is established independent of a particular performance, while in
others it is deduced from the sampled event (as, for
example, in every case in Altra voce).
Usually, the freeze operation concerns the continuous reproduction of individual notes, individual chords, or notes and chords with rhythmic
articulation or internal melodic pro le. This situation results in the perception of prolonged, interwoven, dynamically variable textures of sound that
function as structures of continuity in the economy of the compositions. In other cases, the freeze
may operate instead on the reproduction of longer
fragments articulated rhythmically and melodically
(entire musical phrases, for instance). This use of
sampling, which plays on the strati cation of
memory, resembles the use made by Berio himself
of delay algorithms with relatively long delay times

34

Computer Music Journal

Memory and Continuity:

Delay and Sampling Algorithms

Figure 4. A fragment of the

vocal part of Altra voce.
(Copyright 1999 by Universal Edition A.G. Used
with permission.) The line
at the bottom shows one

of the possible results that

can be obtained by applying real-time transposition
through the harmonizer algorithms.

Figure 5. A fragment of the

part of the rst orchestral
group of Ofan`m (Copyright 1988, 1997 by Universal Edition A.G. Used

with permission.) with indication of the transposition intervals (without indication of the overall
result).

(4)

(5)

(up to 8 sec at some moments in Outis). This also

reiterates the concept of the dialogue and of proliferation in time of the sound events in question,
characteristics common to all of the works analyzed here.
Concerning other sampling systems, in Ofan`m
and Outis we nd among the performers a sampler
(keyboard in the score) that is used with instrumental or vocal timbres to augment the potential
of the originals in both harmonic aspects and
rhythmic articulation (e.g., clusters of large size,
particular glissandi, rhythmic gures). In both of
the compositions, the sampled sounds never stand
out alone from those of the orchestra but blend
with the rest of the live sounds and serve the function of both sound enrichment and orchestral extension.
Algorithms and Musical Events: System
Management and Technological Solutions
The algorithms employed by Berio are not in themselves complex but belong to the classical repertoire of live electronics. The possible problems of
implementation regard on the one hand a series of
micro-level details (e.g., selection of the individual

parameters of the algorithms, times and modes of

their input and output, etc.) and on the other the
global management of the score. Regarding the latter aspect, some of Berios scores often call for, at
least in two of the cases described, large orchestras
with electronics that at various times operate on
different instrumental and/or vocal groups. These
scores also call for very large ampli cation systems
for both the sheer number of instruments involved
and the number of diffusion groups. In addition, the
high degree of strati cation of the electronics (i.e.,
the number of individual operations in parallel),
which evolves continuously during the playing of
the pieces, contributes still further to the general
complexity, making it necessary to devise very
highly evolved solutions for managing the plurality
of events.
Over the years, Tempo Reale has developed a
system based on a number of fundamental characteristics: versatility and adaptability to different
musical contexts, simplicity and rapidity in bringing about changes in events (changes that become
necessary in regard to compositional requisites and
potentially at any stage in production), stability
of the entire system, and overall sound quality. In
the general construction of the system, a primary
goal was that of attaining the maximum degree of
Giomi, Meacci, and Schwoon

35

Figure 6. General logical

diagram of the hardware
and software system used
by Tempo Reale for live
electronics. Additional

systems refers to MIDI

keyboards, sensors, fader
controllers, display interfaces, and so on.

automation possible, but without excessively constraining the freedom of the performers, both instrumental and electronic.
The system has a central management core with
a series of collateral elements whose interaction
and integration ensure easy management of a wide
range of musical and control events. The logical
scheme of the control structure (see Figure 6) is
conceptually simple, which guarantees its adaptability to numerous situations. The central node of
the system consists of an audio matrix including
one or more mixers within which converge all audio signals to be processed or to be ampli ed. In
Altra voce, the work that calls for the fewest performers ( ute and voice) of those discussed here, a
single digital mixer is used, while Outis (a musical
theater composition involving, in addition to chorus and an orchestra, numerous soloists), requires
two digital mixing boards, an analog mixer of large
size for the orchestral premix, and an additional
small analog mixer for the vocal group. The structure of the audio matrix can thus vary substantially, but the central core always consists of one
or more digital mixers entirely controlled by MIDI.
Thanks to this feature, it is possible to remotely
manage the quantity of each individual signal to be
sent to the signal processors and to the diffusion or

spatialization system, thus obtaining the greatest

possible degree of exibility in signal routing.
For signal processing operations that are particularly computationally demanding, dedicated commercial hardware systems have been selected that
share with the entire system the characteristics of
programmability and reliability in the majority of
operational situations. These devices have been
used basically for the reproduction of recorded samples and for three types of algorithmsharmonization, delay, and reverberation, as well as for
different combinations of the three. In all cases, the
various parameters are controlled by MIDI. For
other operations, a different approach has been
used: that of programming workstations in the
Max/MSP environment. This is the case with the
reproduction of playlists of long duration, of
freeze algorithms, and of the sound spatialization
system. The management module (Cue Manager),
which controls all of the elements of the system, is
also programmed in Max/MSP. Through internal
messages, MIDI, and Ethernet, it controls all of the
other subsystems, opening and closing the channels
of the matrix, sending sequences of intervals to the
harmonizers, actuating the spatialization engines,
controlling the sampling systems, generating monitoring signals, and so on.

36

Computer Music Journal

The Cue Manager is based on a structured sequence of events in which each block represents a
precise point in the score (cue) along with all of the
electronics events associated with it. During a live
performance, the list is scrolled manually by means
of a computer keyboard or through a MIDI peripheral. (In Ofan`m, for example, the keyboard player
in the orchestra controls the Cue Manager.) To the
possibility of sequentially scrolling the list of
events, the system adds that of jumping directly to
any point in the score, instantaneously recalculating the lines of the electronics enabled at that
point, and repositioning itself at the musical situation requested. This characteristic is highly useful
in the testing stage and during rehearsals with the
instrumentalists.
The structured list of events is normally complex
and thus dif cult to write and to edit. A specially
designed relational database (Meeg, or Max Electronic Events Generator) is used expressly for the
purpose of facilitating writing and subsequent editing. This database can be automatically generated
after the formalization of the electronics commands. Through interaction between Meeg and the
Cue Manager, it is possible to recreate the entire
con guration of the system and the ensemble of instructions speci cally used in each composition
without having to act directly on the database
code. Thanks to this approach, Meeg, like most of
the system, can be used in a broad range of situations without requiring modi cations in its code,
and it allows rapid changes in the sequence of
events to satisfy new musical or technical requisites.
The problem of stability is particularly important
in musical systems having a high degree of integration between software and hardware devices and
has been the subject of special attention. Usually,
the weak point lies in the computer sub-systems,
which may be subject to complications of various
kinds not always easily predictable. To lessen the
possibilities of a system crash, redundant subsystems that function as backup devices have been integrated into the structure. These subsystems,
proceeding synchronously with the main ones, can
be used instantly in case of problems during a performance.

Inside the Music

Although many of the musical concepts in the
three works are in some way correlated, and the
guidelines for designing their electronics comply
with a unifying logic, it seems useful to describe in
greater detail the relationship between music and
technologies within the context of the three compositions which, as a whole, offer a panorama of
different aspects of making music, from chamber
music to musical theater. In this article, the three
works are examined in chronological order based
on the rst drafting of the scores. Actually, the
evolutionary cycle of the latest revisions of the
electronic parts should follow a different order
(OutisOfan`mAltra Voce). This indicates how
Berio constantly reconsiders each performance with
electronics; in this sense, each new version is inevitably in uenced by previous experience and
work.
Inhabiting Space: Ofan`m
Ofan`m for two childrens choirs, two instrumental
groups, a female voice, and live electronics can be
considered a work in progress. The rst version
was presented in Italy, at Prato, in 1988, and it has
since undergone various processes of rewriting of
not only the electronics but also the entire score
(Bernardini 1995). In the latest versions, the vocal
and instrumental parts have undergone no changes,
but Berio already speci ed the open nature of the
electronics part in these terms:
Musical thought today must be able to interact
with the new technologies and to adapt itself
creatively to every kind of space, exploring its
virtualities and reshaping acoustically. The image of music as sound architecture is no longer
a mere metaphor: it represents a concrete possibility, realizable in all its aspects. It is, of
course, a mobile and exible architecture, capable of adaptation to different situations and
environments. Therefore, the acoustic strategy
of Ofan`m has to be modi ed with each new
performance and consequently several aspects
of the entire work are recomposed. (Berio 1992)
Giomi, Meacci, and Schwoon

37

Figure 7. Con guration of

the loudspeakers in
Ofan`m (top view).

Sound spatialization assumes a crucially important role in characterizing the formal sections musically, while ampli cation is entrusted to a
speci c stereophonic system (see Figure 7, groups
LR). The eight groups of loudspeakers employed
for movement in space form a circle around the audience (see Figure 7, groups 18), thus reinforcing
the initial concept of the wheel. Alternation between the two levels of the text (Ezekiel and the
Song of Songs) takes place formally by blocks with

very strong contrasts. The fragments of Ezekiel are

linked to very rapid movements of striking physical impact, while the fragments of the Song of
Songs are often distinguished by xed sound locations. Table 1 illustrates the formal scheme, including very simple verbal descriptions of the
spatialization, and gives some idea of the dramaturgical relationship between the various sections of
the piece. The table also shows how at certain moments the spatialization anticipates textual and
musical elements of the next section (IU II, VIII U IX)
or reiterates and develops characteristics from the
previous section (IIIK IV, VK VI).
To investigate the function of electronics in
Ofan`m, we may consider some details of the composition. The score begins with a continuous proliferation of elements. The whole originates from a
single note sung by the soloists in the chorus,
which is sampled and immediately reproduced in
loop. At the moment in which the recorded version
becomes perceptible, the soloists begin a melodic
path that arises from this central note, while subsequently other rami cations of the clarinet and
the sampling of the childrens choir are added. In
this fragment, marked by a strong timbral af nity
between the instrumental and vocal elements, the
electronic ampli cation contributes to a very amalgamated acoustical result. The process of sound
opening, both harmonic and timbral, is accompanied by spatial projections that slowly oscillate between the front loudspeakers, creating a sense of
inner motion along the vast stereo front. At a following stage, onto this sound texture are superimposed more lively gurations of the two clarinets
as well as rototoms, diffused with zig-zag motions around the circle of loudspeakersa moment
of great surprise which seems a preparation for the
explosion of sound at the beginning of the second
part, a sort of presentiment of Ezekiels vision of
the hurricane.
Another section of Ofan`m of great importance
for the electronics part is the clarinet solo (Section
IV). Throughout the preceding fragment, taken
from the Song of Songs, the rst clarinet emerges
strongly with gurations that are almost arpeggiations, very sweet and played pianissimo, but that
remain well integrated into the overall harmonic
structure. From this series of gurations emerges

38

Computer Music Journal

In his program notes, Berio also explains the use

of the texts:
The text of Ofan`m (in Hebrew both wheels
and modes) alternates fragments from the
book of Ezekiel with verses from the Song of
Songs. The dramatic vision of Ezekiel (Chapter
1)the most personal and apocalyptic of all
prophetsstands in strong opposition to the
earthly sensuality of the verses from the Song
of Songs (Chapters 4 and 5). The phantasmagoric apparitions of Ezekiels vision whirls
around in perpetual motion against a burning
sky: the poetic images from the Song of Songs
dwell longingly on the face and body of the beloved. (Berio 2000)

Table 1. Formal diagram of Ofan`m and of the relevant distribution of the text with respect to the
individual sections. The third column contains a verbal description of the sound movements.
I.

EZEKIEL (1:28) And I heard a voice of one that spake.

frontal calm movements; rototoms with fast

zig-zag movements through full space

II.

EZEKIEL (1:410) And I looked, and, behold, a whirlwind

came out of the north, a great cloud, and a re infolding
itself . . . four living creatures . . . and every one had four
faces, and every one had four wings . . . and the sole of
their feet was like the sole of a calfs foot . . . they four
had the face of a man, and the face of a lion . . . the face
of an ox . . . the face of an eagle.

fast movements of various kinds (full space)

III.

SONG OF SOLOMON (4:16) Awake, O north wind; and

come, thou south; blow upon my garden, that the spices
thereof may ow out. Let my beloved come into his garden, and eat his pleasant fruits.

xed diffusion (frontal)

IV.

CLARINET SOLO
EZEKIEL (1:28) . . . A voice . . .

fast movements (softly blowing )

V.

EZEKIEL (1:1421) And the living creatures ran and returned as the appearance of a ash of lightning . . . behold one wheel upon the earth by the living creatures . . .
and when the living creatures went, the wheels went by
them: and when the living creatures were lifted up from
the earth, the wheels were lifted up . . . and when those
stood, these stood.

fast movements (full space)

VI.

INSTRUMENTAL TUTTIROTATING

fast movements

VII.

SONG OF SOLOMON (5:34) I have put off my coat;

how shall I put it on? I have washed my feet; how shall I
de le them? My beloved put in his hand by the hold of
the door, and my bowels were moved for him.

xed diffusion (frontal)

VIII.

TROMBONE SOLO

fast movements on the rst four loudspeakers

IX.

EZEKIEL (1:2224) And the likeness of the rmament

upon the heads of the living creature was as the colour of
the terrible crystal . . . and . . . I heard the noise of their
wings, like the noise of great waters, as the voice of the
Almighty . . . as the voice of an host.

fast movements (full space)

X.

SONG OF SOLOMON (4:19) Behold, thou art fair, my

love; behold, thou art fair; thou hast doves eyes . . . thy
teeth are like a ock of sheep, which came up from the
washing . . . thy lips are like a thread of scarlet; and thy
speech is comely: thy temples are like a piece of pomegranate . . . thy neck is like the tower of David . . . thy
two breasts are like two young roes . . . how fair is thy
love . . . thou hast ravished my heart, my sister, my
spouse.

xed diffusion (frontal)

XI.

EZEKIEL (1:2628) And above the rmament . . . was the

likeness of a throne . . . and the appearance of a man
above upon it . . . as the appearance of re . . . and . . .
brightness round about . . . as the appearance of the bow
that is in the cloud in the day of rain . . . A voice . . .

fast movements (full space)

XII.

EZEKIEL (19:1013) Thy mother is like a vine in thy

blood, planted by the waters: she was fruitful and full of
branches by reason of many waters. And she had strong
rods . . . and her stature was exalted among the thick
branches, and she appeared in her height with the multitude of her branches. But she was plucked up in fury, she
was cast down to the ground, and the east wind dried up
her fruit; her strong rods were broken and withered; the
re consumed them. And now she is planted in the wilderness, in a dry and thirsty ground.

calm movements (full space, gradually moving

away from the female voice, then returning to
her)

Giomi, Meacci, and Schwoon

39

Figure 8. Processing algorithm for the clarinet solo

in Ofan`m (cue 17). The
two delays have the same
delay time (3 sec). The

output of delay A (with

feedback) is opened and
closed alternatively, while
that of delay B remains
constantly open.

ery of acoustic space. Regarding the electronics,

only the physical arrangement of the sound loudspeakers still alludes to the concept of wheel.
Electronics for Musical Theater: Outis
the solo passage, accompanied only by a few
punctuations of the two choirs, the percussion,
and the sampler. The entire section is treated with
a single algorithm of live electronics, a combination of spatialization, harmonization, and delay
with variable feedback (see Figure 8). The sequences of spatialization, harmonization, and the
amount of feedback in the delay are all looped and
superimposed in an irregular manner, both among
each other and with the original sound. The result
is a kind of dialogue of the clarinet with its electronic double. The score oscillates between zones
dense in arpeggiation and other more rari ed ones
(see Figure 9), continuously shifting attention between the original clarinet sound and its electronic
transformation.
Lastly, consider Section VI, a fragment without
voices that follows the central movement of the
text of Ezekiel, the vision of the wheels. This section begins with a great crescendo of the tremolos
reinforced again by electronics that proceed with irregular zig-zag spatialization and with harmonization sequences that are different for the two
orchestral groups. The very particular orchestral
writing (combinations of whirring trills, very fast
repetitions, and arpeggios), the use of slide whistles, the rhythmically asynchronous organization,
and the almost circular melodic pro le of the individual parts create, as a perceptive result, a single
mass of sound, overwhelming, turbulent, and almost in rotation. Through the various electronic
shifts (in pitch, time, and space) it becomes even
more agitated and teeming with life.
At points such as these, an important change has
been made in the latest version of Ofan`m presented in Milan. Whereas the former versions were
always distinguished by real rotations in the spatialization, these have now been abandoned in favor of much more irregular movements that confer
a somewhat probabilistic nature on the sense of
spatialization, conveying a striking sense of discov-

Outis is one of Berios two more recent works of

musical theater, along with Cronaca del Luogo,
which inaugurated the Salzburg Festival of 1999. It
is a complex work lasting nearly two hours with
texts by Berio himself and by the renowned Italian
Hellenist Dario Del Corno, whose title is evocative
of Homers Odyssey and of Ulysses answer to
Polyphemus: Outis emoi gonoma or My name
is no one (Del Corno 1999). However, the work
does not follow the narrative path of the Odyssey
but is instead distinguished by a division into ve
cycles within which emerge a plurality of histories
and plots, referring to the most widely differing authors, from Homer himself to Catullus, from Auden to Brecht, to Joyce, Melville, Sanguineti, and
Celan. The multiplicity and variability of both text
and scenebut above all of the musicare to be
considered the true narrative constants of Outis,
characteristics that make it possible to perceive
different gures and episodes in the same light or
to grasp the sense of a single element in constantly
changing lights and musical perspectives (Berio
1999a). This pertains to all elements of the musical
path, whose amplitude and continuous transformation bear witness to their complexity, suggested
also by the relatively large number of performers:
19 soloists (13 singers, 1 actor, 5 instrumentalists
on the scene), a vocal group of 8 singers, chorus, orchestra, and live electronics.
Performed for the rst time at La Scala in Milan
in 1996, directed by Graham Vick, Outis was presented again in 1999 rst at La Scala and then at
the Theatre du Chatelet in Paris, where it was directed by Yannis Kokkos. For these two performances, Berio designed a new electronics part more
suitable to both solving problems typical of electronics for the musical theater and the identi cation of the particular musical and sound features of
the work itself. The basic premise for electronics in
Outis is an attempt to free the sound from the undesirable tube-like effect deriving from the posi-

40

Computer Music Journal

Figure 9. Clarinet solo excerpt from the score of

Ofan`m. (Copyright 1988,
1997 by Universal Edition
A.G. Used with permis-

sion.) The electronics begin pianissimo after letter

K (cue 17) with the algorithm shown in Figure 8.

tioning of the orchestra in the theater pit, an effect

which undoubtedly con icts with the ideal of analytical listening. What nds application here is the
general concept of an orchestral sonority, which,
thanks to ampli cation and proper equalization,
can approach the character of a perfect studio recording. In this case, however, the sound diffusion
must be such as to prohibit any perception of the
presence of loudspeakers or of evident electronic
transformations; the processing of the sound is
something that appears but which the audience
should not have time to identify and recognize:
In this workwhere there is not, in effect, any
real live electronics, but there is a great respect for the musical acoustic substance of the
work and also for space itselfthe technologies tend to prolong certain aspects, to develop
them interiorly in a way that might almost be
called devious and concealed. (Berio 1999b)
In this sense, the difference between this concept
and the electronics of Ofan`m, where the presence
of loudspeakers constitutes an element of great dramatic impact, is striking.
In the case of Outis, ampli cation and diffusion
systems adapted to the scenic space of La Scala and
of the Theatre du Chatelet were used. The frontal
arrangement of the loudspeakers provides four
groups of loudspeakers placed in the stage boxes, to
which is added a central cluster serving to close the
stereo front and for diffusion of the voices (see Fig-

ure 10). The four side groups are also used to

achieve sound diffusion with horizontal, vertical,
and diagonal movements. Furthermore, a group of
loudspeakers is placed inside the chandelier located
at the center of the ceiling over the stalls and is
used to play particular recorded sequences of childrens voices. As a whole, this is a diffusion system
that, on the one hand, reiterates structural features
of the work, while on the other interacts with and
is grafted onto the most classic and dif cult scenario, that of the traditional theater space. Berio
himself clearly explains this dual function of the
diffusion techniques:
An acoustical dimension is created which no
longer corresponds to that of the orchestra pit.
There are loudspeakers concealed at the sides
of the stage and above the stalls, of which the
audience will be unaware, but which will serve
to enlarge the sound perspective. The central
interest of these sophisticated systems occurs
when they are adapted to the musical work.
There can be no indifference toward this type
of technology in relation to the work represented[;] it must enter into it and be redesigned
in accordance with the nature of the work.
(Berio 1999b)
Although the computer music environment uses
conceptually simple algorithms, it is their relationships with the musical score and with the orchestral sound that determine their functioning. The
Giomi, Meacci, and Schwoon

41

Figure 10. Con guration of

the loudspeakers in Outis
(front view).

spatialization and the harmonizing as well as, the

microphone choice and placement, thus contribute
to a process of extending the orchestral sound and
projecting it into a dimension of continuous mobility and transparency:

All of the instruments in the orchestra, the choir,

and the instrumental soloists are given individual
microphones, whereas the voices all sing without
ampli cation with the exception of the speaker and

the vocal group. Microphoning has the objective

of capturing the signals both for sound reinforcement and for the live electronics algorithms. The
electronics includes the spatialization and a series
of delay and harmonization algorithms that are applied at various times to different orchestral groups.
In addition, recorded sequences (playlists) containing both recited phrases and real sounds of various
kinds (animal sounds, a ships foghorn, birds, and
so on) are reproduced.
The setup also includes a sampler placed in the
orchestra pit with which a keyboard player executes sound events with mainly instrumental or
choral timbres. During the second, fourth, and fth
cycles, the work uses a motion detection system
(realized already for the rst version of Outis by
the Musical Informatics Laboratory of DIST in
Genoa). The system is based on two footboards
placed at stage level that, when activated by a
movement of the foot, effect reproduction of percussion samples.

42

Computer Music Journal

I have always tried to renew the instrumental

sound, the one produced by known and accepted instruments. I have tried to transform it
to make listening more analytical, with several
layers interacting with each other. . . . There
are articulations which are harmonically very
complex and the microphone can, in effect,
contribute to better listening. Basically, it is
used as a microscope, for the enlargement of
minimal aspects, acoustic and musical, of the
work. (Berio 1999b)

Figure 11. Con guration of

the loudspeakers in Altra
voce (front view).

Many aspects thus make Outis a complex and

important work among the recent musical achievements of Berio: the fragmentation of the texts, the
many ways of singing them, the concealed but intense use of live electronics, and the adaptability of
technologies to the work and to the scenic space.
In this sense, the electronics function to enrich a
sound organism that, albeit moving with relatively
traditional characteristics, is enhanced by continuous mobility and variability over time, with the result of constantly drawing the listeners attention
back to the musical dimension of the work, making sound, more than the other parameters of musical theatre, the true protagonist of the long and
complex dramaturgical process (Giomi 2000).
The Live Electronics da Camera of Altra voce
Altra voce for contralto ute, mezzo-soprano, and
live electronics is an example of Berios use of electronic techniques within a chamber music context.
This composition, however, is derived from a work
of musical theater:
In one episode, namely Il Campo (The Field),
from my azione musicale[,] Cronaca del luogo,

there is a virtual love duet. Two voices and

several instruments fall in love and follow
one another in a constantly renewing
relationship. . . . As we all know, in true polyphony each voice contributes to the whole,
yet retains its own identity, if not complete autonomy. In Altra voce, I have liberated one
voice (mezzo-soprano) and one instrument (alto
ute) from the whole and developed their respective autonomies and harmonic premises
by, among other means, using live electronics.
(Berio 2001)
The choir, the orchestra, and the loudspeakers of
Cronaca del Luogo are placed behind the stage at
various heights, forming a kind of wall of sound.
In Altra voce, this idea of a wall is translated into a
con guration with two diverging diagonal lines of
loudspeakers that, adapting themselves to the performance space, are positioned at the farthest possible distances (see Figure 11, groups 16). To the
diagonals are then added two groups of loudspeakers positioned at the performerss feet for sound reinforcement (see Figure 11, groups LR).
In this composition, the electronics play a crucial
structural role and are characterized by three types
Giomi, Meacci, and Schwoon

43

Figure 12. Formal diagram

of electronics in Altra
voce. For sampling, an index and an approximate
duration of the fragment
recorded and reproduced

in loop are indicated (for

example: #1 ||:49 :||). For
the spatialization, the
loudspeakers are denoted
by Roman numerals. For
the harmonizers, the inter-

vals used are shown (e.g.,

0, 2# , 2` $ with the sequences which work on
permutations of these intervals).

of processing: the live sampling of various fragments, spatialization, and the use of harmonizers.
From the formal diagram of the electronics part
(see Figure 12), it is immediately apparent that the
various kinds of processing do not take place one
after another but tend to overlap and interweave.
The rst sample, a single F for voice and ute, is
actually held from the third beat (see Figure 13) to
the last. Regarding the function of this continuous
note, Berio speaks of a sort of tonic, an alwayspresent element that gives a common perspective
to all other events. This F is accompanied at various times by other samples falling into two categories: fragments of held notes, which played in a
loop create continuous events (#1, #2, #3, and #6);
and entire musical phrases (#4, #5, #7, and #8),
which appear instead as repeated musical structures. This creates various reference elements that
could be described as attraction poles, compared
to the precise delimiting of the traditional tonic.
The musical discourse moves within the gap that
opens between these poles.
Following the formal scheme it can be seen how
the rst category of samples has an almost symmetrical distribution (#1-#2-#3-#3-#6-#2-#3-#2-#1),
while those with phrasing (#4, #5, and #7) mark the
beginning or the end of formal accumulations.
Sample #8 should be considered a special case, as it
consists of a structure recorded off-line, that is,
not during the performance, and enriches the nal

zone in which, in general, continuous rarifying of

the sound layers occurs.
The sound spatialization is characterized by a
process of expansion of the loudspeakers closest to
the performers (1 and 2) up to the ensemble of all of
the loudspeakers, reached at beat 49. From this moment on, various differentiations in the use of
space take place based on the type of movements
(speed and regularity) and on the use of pairs of
loudspeakers alternating with individual loudspeakers.
In addition to these ideas of movement, Altra
voce is undoubtedly an important example of how
Berio operates with electronics on degrees of af nity between different sonorities: initially, with the
long F held on the loudspeakers closest to the performers, the already intense fusion between the
timbre of the mezzo-soprano and the contralto
ute is reinforced still further, while subsequently,
with the accumulation of layers of samples (freeze)
and of harmonizers, the various signals are increasingly separated in space, thus rendering transparent
an ensemble of layers so homogeneous as to be otherwise almost impenetrable to the ear. Concerning
the harmonizers, the formal diagram demonstrates
the progressive augmentation of the intervals and
of the number of intervals per sequence. In beats
94109, the degree of parallelism of the harmonizers also increases, up to a maximum of three on the
ute and two on the voice. In the nal part of the

44

Computer Music Journal

Figure 13. A preliminary

version of the rst page of
the Altra voce score, realized by Paul Roberts and

the authors for Universal

Edition. (Copyright 1999
by Universal Edition A.G.
Used with permission.)

composition, the harmonizations progressively disappear in favor of the exclusive presence of spatialization alone.
The overall form can be described as two arches,
with the rst climax underlined predominantly by
progressive spatial expansion and the second,
thanks to the harmonic strati cation, achieved by
the augmenting of intervals, the length of the harmonizers transposition sequences, and their degree
of parallelism. The superimposition of freeze techniques on these elements creates for the listener a
texture that extends and develops throughout the
entire composition, a harmonic wall in ceaseless
evolution. All of these aspects of Altra voce may be
viewed within the light of the initial perspective:
the simple metaphor of falling in love leads to a
complex polyphony and to a surprising interaction

among the three dimensions of sound: vocal, instrumental, and electronic music, which, while retaining their individual autonomy, seem almost to
abandon themselves to each other, creating hybrid
situations of striking intensity.

Acknowledgments
Many people have contributed over the years to the
development of the computer music systems of
Tempo Reale in relation to the works described;
among these are Nicola Bernardini, Alvise Vidolin,
Thierry Coduys, Lamberto Coccioli, and Sylviane
Sapir. An important role for the success of the Italian performances has also been carried out by the
technicians of the BH-audio company who have

Giomi, Meacci, and Schwoon

45

Berio, L. 1992. Ofan`m. Concert program. Jerusalem:

Gala concert in honour of the dedication of the Supreme Court Building.

Berio, L. 1999a. Morfologia di un viaggio. Outis: Catalogue of the Opera. Milan: Teatro alla Scala.
Berio, L. 1999b. Interview for the television program Superquark. RAIRadioTelevisione Italiana.
Berio, L. 2000. Ofan`m. Luigi Nono e il suono elettronico. Milan: Milano Musica/Teatro alla Scala.
Berio, L. 2001. Altra voce. Carnegie Hall Stage Bill.
New York: Carnegie Hall.
Bernardini, N. 1995. Ofan`m di Luciano Berio. Lecture
Notes for Electronic Music. Padova: Conservatory of
Music.
Del Corno, D. 1999. Nessuno. Outis: Catalogue of the
Opera. Milan: Teatro alla Scala.
Giomi, F. 2000. Musical Technologies in Luciano Berios Outis. Proceedings of the XII Colloquium of
Musical Informatics. LAquila: Istituto Gramma/AIMI.

46

Computer Music Journal

provided audio services. Special thanks also to Paul

Roberts, transcriber of Luciano Berios scores, for
his invaluable aid with constant reference to the
most updated versions. The authors are grateful to
Universal Edition and to Ricordi-BMG for their permission to publish the excerpts from the scores.

References