Seeking Out the Spaces Between: - Recherche en IA spécialisée au MIT

Seeking Out the Spaces Between:
Using Improvisation in Collaborative
Composition with Interactive
Technologie

a b s t r a c t

This article presents findings

Sarah Nicolls

In seeking to create responsive “performance en-

vironments” at the piano, I explore live, performative control
of electronics to create better connections for both performer
(providing the same level of interpretive freedom as with a
“pure” instrumental performance) and audience (communi-
cating clearly to them). I have been lucky to witness first-hand
many live interactive performances and to work with various
empathetic composers/performers in flexible working envi-
ronments. Collaborating with experienced technologists and
musicians, I have witnessed time and again what, for me, is a
fundamental truth in interactive instrumental performance:
As a living, spontaneous form it must be nurtured and in-
formed by the performer’s physicality and imagination as
much as by the creativity or knowledge of the composer and/
or technologist.

Specifically in the case of sensors, their dependence on the
detail of each person’s body and reactions is so refined as to
necessitate, I would argue, an entirely collaborative approach
and therefore one that involves at least directed improvisa-
tion and, more likely, fairly extensive improvised exploration.
The fundamentally personal and intimate nature of sensor
readings—the amount of tension created by each performer,
the shape of the ancillary gestures or the level of emotional
involvement (especially relevant when using galvanic skin
response or EEG)—makes creating pieces with sensors ex-
tremely difficult for a composer to do in isolation. Improvisa-
tion therefore provides a way for performer and composer to
generate a common musical and gestural language.

Related to these issues is the fact that the technical and
notational parameters in interactive music are not yet (et
may never be) standardized, thereby creating a very real and
practical need for improvisation to figure at least somewhere
in the process.

from experiments into piano
and live electronics undertaken
by the author since early 2007.
The use of improvisation has
infused every step of the
process—both as a methodol-
ogy to obtain meaningful results
using interactive technology and
as a way to generate and char-
acterize a collaborative musical
space with composers. Le
technology used has included
pre-built MIDI interfaces such as
the PianoBar, actuators such as
miniature DC motors and sensor
interfaces including iCube and
the Wii controller. Collaborators
have included researchers at the
Centre for Digital Music (QMUL),
Richard Barrett, Pierre Alexandre
Tremblay and Atau Tanaka.

Context
Many practitioners in the field of
live performance with electronics
make their own interfaces or in-
struments with which they impro-
vise; this is readily demonstrated
by communities such as New In-
terfaces for Musical Expression
(NIME), inspired by leading figures
such as Nicolas Collins and Michel
Waisvisz. From what is now a hugely
broad field, performances I have
witnessed recently that seem most relevant, either through
their use of physical drama or a particular technology, include
Chikashi Miyama (“Angry Sparrow”), in his dazzling, virtuosic
and humorous performance on a self-made interface [1], et
Derek Holzer [2], whose optical discs are attached to spinning
motors and thrust under an overhead projector for instant,
rough-and-ready multimedia effect.

Relevant sensor performances include Atau Tanaka (Sen-
sors_Sonics_Sights) (SSS) [3] and Benjamin Knapp [4]; les deux
use the BioMuse sensor system, which was invented by Knapp
and Hugh Lusted [5]. Tanaka and Knapp present a fascinating
contraste, as they use the same system to quite opposite ends:
Tanaka is a highly gestural, physically active and expressive
performer, while Knapp performs seated and—using sensors
including EEG and galvanic skin response—plays with emo-
tional readings, generating music from a quite inward control
of his internal self. At MIT, Elena Jessop developed a beauti-
fully intuitive glove [6], which enables her, Par exemple, à
grab notes seemingly from her mouth and lengthen them by
pulling away from the face smoothly.

The individuality of each of these performers only strength-
ens the case that improvisation is not only a way of generating
music but also the key to inventing and learning a host of new
instruments, interfaces or systems of interaction.

Sarah Nicolls (artist, educator), Centre for Contemporary Music Practice, School of Arts,
Brunel University, Kingston Lane, Uxbridge, UB8 3PH, ROYAUME-UNI. E-mail: . Web site: .

Voir et for supplemental files
related to this article.

Co-author on Case Study 1: Richard Barrett, Wilhelm-Stolze-Strasse 30, 10249 Berlin,
Allemagne. E-mail: .

Co-authors on Case Study 4: Samer Abdallah, Kurt Jacobson, Andrew Robertson, Adam
Stark and Nick Bryan-Kinns, Centre for Digital Music, Queen Mary, University of London
(QMUL), Mile End Road, London E1 4NS, ROYAUME-UNI. E-mail: c/o . Web: .

PhysiCality
The consideration of physicality is one of the key aspects in
creating instrumental performances designed to give control
of the electronics to the performer. Several texts affirm the
importance of the physicality inherently learnt and absorbed
as part of instrumental study. John Richards’s article “Lost
and Found” [7] has an array of excellent quotations, the most
succinct of which is Bob Ostertag’s Human Bodies, Computer

LEONARDO MUSIC JOURNAL, Vol. 20, pp. 47–55, 2010 47

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

Music: “An intelligence and creativity is
actually written into the artist’s muscle
and bones and blood and skin and hair”
[8]. O’Modhrain and Essl make a simi-

larly poetic, yet astute, indiquer: “Implicit
in the experienced musician’s under-
standing of the relationship between ac-
tion and sound, between performance

(un)

(b)

gesture and musical phrase, is a more or
less complex internal representation of
the dynamics of an instrument” [9]. Ce
notion that there is an imprint of the
performer’s instrument within the per-
former’s body is vital when considering
why improvisation might be necessary in
generating the language and parameters
for performing interactive instrumental
musique.

To improvise means at one level to fol-
low one’s instinctive urges, the internal
reactions and responses that could be
referred to as pre-analysis in the perform-
er’s own cognitive process. If this is then
paired with a physical internal awareness
of the capabilities of one’s setup or in-
strument, then the responses will logi-
cally be faster and more innate, intuitive
and highly responsive than if one or an-
other is non-instinctive. If the performer
can intuitively know the edges of physi-
cal possibility for the sensors—where the
highest and lowest readings are found
for example—then the manipulation of
these will be managed most deftly.

Instrumentalists have finely tuned sys-
tems of tactile or physical feedback (les deux
external, when touching keys etc., et
internal—knowing when or how to re-
lax when playing fast, Par exemple) et
in working with interactive technology,
muscle memory gets built up in a similar
chemin. Also crucial to this discussion: Quand
creating new composed pieces with an in-
teractive setup, to have the performer
improvise with the technology means to
unlock this inner physical language, à
find both what is possible and natural
and also what is unnatural, or outside of
the natural body language: “the spaces
between pianism,” in my case. This then al-
lows for the fundamental aesthetic judg-

figue. 1. Examples of tremblay’s score. (© Pierre alexandre tremblay)

(c)

48 Nicolls, Seeking Out the Spaces Between

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

6 Independent Audio Outputs

piano

computer

2 Audio Feed (1 omni microphone + 1 magnetic pickup)

2 Data Feed (1 control pedal + 1 MIDI note stream)

figue. 2. tremblay’s schema. (© Pierre alexandre tremblay)

public

ment of whether to make the interactive
control in addition to, or part of, the in-
strumental playing.

How the use of physicality may change
the original gesture also needs consider-
ation. The writings of Wanderley and
Cadoz [10] on this topic are well known;
their discussion is furthered by Wander-
ley and Miranda’s extensive 2006 study of
new instruments:

found that I would focus on playing the
sensors, thereby turning the previously
nearly subconscious movement into a
material action. As a solo performer is
only one body, one mind, these cycles
of complexity and confusion perhaps
begin to disrupt the artistic spontaneity
and intuitive physical sense, potentiellement
undermining the original meaning of
the gesture.

The instrumental gesture . . . is ap-
plied to a concrete (matériel) objet
with which there is physical interac-
tion; specifi c (physical) phenomena are
produced during a physical interaction
whose forms and dynamics can be mas-
tered by the subject. These phenomena
may become the support for communi-
cational messages and/or be the basis for
the production of a material action [11].

The consideration of how adding a
sensor to a pianist’s arm may affect both
the pianist’s and the audience’s relation-
ships to the original semiotic function
of the gesture was one of the main ques-
tions resulting from work on Case Study
3. Although not a central issue for this
article, I briefl y illustrate the problem I
found here. Imagine the pianist lifting
the arm away from the keyboard, par-
haps signifying a breath between musical
phrases. When using this gesture to gen-
erate data and, à son tour, process sound, je

Case study 1
Richard Barrett’s Adrift (2007) était
commissioned as part of my fi rst Arts
and Humanities Research Council
(AHRC)–funded project in 2007 [12],
which sought to increase the repertoire
for piano and live electronics. I had al-
ready performed Barrett’s Lost, and this
became the foundation for Adrift [13].
Essentially, Adrift amplifi ed Lost into a
semi-improvised duet for Barrett and me
(with Barrett playing his keyboard system
using STEIM’S LiSa program). He began
the compositional process by recording
my performance of Lost and chopping it
into upwards of 70 sections. These were
reordered and gradually shifted in pitch,
the fi rst ones very slightly, increasing as
the piece progressed. The degree of
other processes (fi ltering, short delays
and feedback) also generally increased,

so that the result gradually diverged in
pitch and timbre from the original.

Our pre-made parts (my part: par-
forming the Lost score, and Barrett’s
part: playing the new recorded version)
were now a basis for Adrift, a consistent
continuation (into the real time of per-
formance) of the compositional process
that gives Lost its particular structure:
taking basic material and interpolating
more and more inserts into it until the
original material becomes almost liter-
ally lost in its own extrapolations, distor-
tion, refl ections, etc.. Either performer
in Adrift could interrupt her/his given
part at any time and interpolate an im-
provised passage before continuing from
the same point where he/she left off (comme
if using a pause button).

What was fascinating was how, having
ingested Lost through hours of prac-
tice, I found myself quite naturally and
subconsciously improvising in Barrett’s
compositional language. Helped by hav-
ing rehearsed in close proximity over sev-
eral sessions and having witnessed several
performances by Barrett in his groups
FURT and fORCH [14], I effectively
internalized the physical language that
accompanies his music. In the live per-
formance [15], Barrett sat at the other
end of the piano, facing me, in the posi-
tion of a second pianist in a two-piano

Nicolls, Seeking Out the Spaces Between 49

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

travail, and the amplification was local (un
stereo pair was placed at either end of
the piano). Ainsi, both of us performed
toward each other, creating an intimate
mirror image.

Barrett’s experience of improvisation
within the compositional process has
constantly evolved; he wrote the follow-
ing to me after our collaboration:

The more I attempt to define what im-
provisation is, the more it seems to slip
through the fingers; si, Par exemple, it
is defined as those aspects of musical
creation which are spontaneous or un-
planned, you run into difficulties. So in-
stead, I prefer to think of “composition”
as defining the act of bringing music
into being, and “improvisation” as one
element among various means by which
that might be brought about. Ainsi, it
isn’t really a matter of bringing “impro-
visation” and “composition” together,
which at first I thought it was: it is more
a question of realizing that they aren’t
really two different things [16].

It is interesting to note that as the pia-
nist, I was able to move from studying
Barrett’s compositional language to im-
provising something comparable, alors que
as composer Barrett used improvisation
to seek out his compositional language.
This cyclical process informed the proj-
ect and helped to create a fine balance
between freedom and a stylized, con-
sistent musical language. Bien que le
interpreter was relatively free, the com-
poser’s voice in fact infused all aspects
of the music.

Case study 2
Pierre Alexandre Tremblay’s Un clou,
son marteau et le béton (2008) [17] illus-
trates the use of improvisation in gener-
ating material, finding a common and
genuinely cumulative language between
composer and performer, and creating
a long piece of music (autour 22 min-
utes), which is rigorously composed,
yet with only approximately 36 bars of
music written on a stave. I had heard
La Rage, Tremblay’s 50-minute suite
for free-jazz drummer and electronics,
showing Tremblay’s ability to frame a
multi-dimensional performer/machine
interaction, combining composition and
improvisation. Tremblay and I used im-
provisation from the outset, improvising
together at first to get to know each other
as musicians, with Tremblay on laptop
and bass guitar. We then began the piece
using bare-boned notation that I impro-
vised upon to test musical gestures and
specific real-time processing. Tremblay
also asked me to improvise freely within
some settings: over a fixed electronic
part, or within some real-time process-
ing that I would subvert with my own
musical inputs. We recorded the results
to use as triggers for the next session.
Tremblay created notation (mixtures of
text, conventional and guided improvi-
sation) to show how he would interpret
what I had played, which allowed me to
feed back on what was communicated
to me, building up the most efficient or

relevant language for the piece (figue. 1).
This process enabled us to understand
each other’s perceptive interpretations
of symbol, word and sound.

We had several sessions like this, grad-
ually obtaining passages that we could
easily re-create through notational short-
hand and that could also be understood
by someone else coming fresh to the
score. Sections of the piece used system-
atic processes, thereby limiting the need
for notation; one example is a section
in which the computer and I built up
an intensifying call and response, avec
the computer taking my notes and re-
ordering and speeding them up, et moi
then imitating its rhythmic profile with
new pitches. Other sections used a bias
from the background to direct the im-
provisation subliminally: par exemple, un
relatively free section with a given fixed
electronic part allows imposition of a tar-
get on an improviser without explicitly
giving musical instructions.

Tremblay then wrote the piece, en utilisant
either audio control signals (certain au-
dible pitches or the creation or absence
of sound) or direct inputs (we used the
PianoBar—a MIDI device placed over
the keys of the piano, reading the pitch
and velocity of each key when played)
(figue. 2). Because the piece used different
input messages (for example, depress-
ing keys or making vocal noises) to the
patch at different points in the piece, it
felt highly responsive in performance: Il

figue. 3. atau tanaka wearing the EMG sensors in the positions we used. (Photo © sarah Nicolls)

50 Nicolls, Seeking Out the Spaces Between

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

figue. 4. Practicing for the collaboration with centre for Digital Music. (Photo © sarah Nicolls)

constantly shifted my attention, et le
process thus felt akin to performing with
other live musicians. En effet, the result-
ing piece I found to be such a detailed
web of interactivity, with many subtle
changes of technological response, que
it felt very much as if I were improvis-
ing with Tremblay himself: The “perfor-
mance environment” engaged me in a
living, breathing way. What fascinates me
is how strict yet supple the piece is, et
I conclude that using improvisation as a
methodology enabled this commonality
to thrive.

Case study 3
Atau Tanaka’s Suspensions [18] shows
how improvisation was used to generate
the grammatical or theatrical language
for an interactive system. We used one
EMG sensor (reading electrical currents
created by muscle contraction) on each
arm (on the forearm extensor muscles)
(figue. 3) and a double-axis accelerometer
on the right wrist. We set out to create

a piece that would combine my natural
gestural/physical/emotional approach
to the piano—including improvisation
in the final performance—with Tanaka’s
compositional language and detailed
and practical knowledge of the sensors.
Although we used his physical perfor-
mance language as a basis for my learn-
ing to “play” the sensors (c'est à dire. his gestural
shapes and tricks to create the right read-
ings), we allowed room, using improvisa-
tion, for my own performative language
to develop.

After generating some initial motivic
and textural musical ideas through im-
provisation, I further improvised upon
these while wearing the sensors to see
what kind of data they would generate.
In two or three initial sessions together,
I took the sensors and began to under-
stand what they did, by simply making
gestures and watching their resultant
data in a patch. This soon became too
limited, so we made a frame patch with
which I could practice using sonic feed-
back. From this point, I practiced purely

with the sensors—using my arms in mid-
air to find the thresholds and to find ges-
tures that produced the right amount of
muscle tension and to begin to memorize
where and how I needed to be to create
useful signals—internalizing the “instru-
ment” of the sensors.

Returning to the piano was difficult af-
ter this extended period of learning the
sensors. Space here is not sufficient to go
into more detail about the process, mais
after giving a work-in-progress showing
to a theater-trained audience I worked
to restore the relationship of sensor use
to pianistic gesture. The tangent of per-
forming without the piano in a theatrical
setting however did give useful insights
into meanings attached to gestures and
also raised questions about whether or
not it was desirable to reveal the tech-
nology [19] and how much I wanted
the audience to understand the connec-
tions between gesture and sound. Encore,
Tanaka and Knapp serve as useful ends
of the spectrum in this case, with Knapp
wearing his sensors hidden beneath a suit

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

Nicolls, Seeking Out the Spaces Between 51

figue. 5. schema for case study 4 resulting piece. (© adam stark)

jacket while Tanaka wears his much more
openly, almost reveling in them.

For me, the single most successful mo-
ment in the proceedings took place when
improvising with a sampled chord “in my
arm” (c'est à dire. an EMG sensor on my arm was
mapped to a sample). I sat at the piano
and when approaching the keyboard
increased the tension in my arm. I be-
gan triggering the very, very beginnings
of the sound—much like the sounds
of breathing or bowing before a note
speaks on a string or wind instrument—
cependant, instead of then allowing the
actual pitched sound to come out of
the computer, I instead played the same
chord on the piano. This moment en-
capsulated for me a genuinely new
approach to the piano—one that could
only be enabled by this technology—
creating a fascinating and intimate
space in which to explore the relation-
ships of the sensors to the pianistic
performance.

This scenario might never have been
realized without improvisation, as it was
the combination of Tanaka and his as-
sistant’s ideas and computer expertise
and my own pianistic approach (again,

referring back to earlier descriptions of
the internalized imprint of the piano in-
side my own body) that gave life to this
idea. Ce, Je pense, is the nub of why im-
provisation is such a useful tool: It allows
the performer to be responsive to the
moment, to the environment and to the ac-
cidents and discoveries that we intuitively
find.

Case study 4
In collaborating with the Centre for
Digital Music [20] I sought to create an
interactive instrumental performance,
with flexible performer-computer inter-
action that would produce live generative
computer algorithms and give the player
both significant control and room to be
surprised by a computer’s responses. Nous
hoped to answer some of the challenges
discussed as far back as 1973 by Cornock
and Edmonds [21], creating a circular
performative feedback loop to make the
relationship between algorithms and the
physicality of the performance seamless
and meaningful. We wanted the per-
former to provide input to generative
algorithms—responding to and modify-

ing them in real-time—and simultane-
ously make an engaging spectacle for the
audience.

Methodology and
collaborative Process
For this project, I invented PianoLab: un
research space that would allow for real-
temps, genuinely collaborative, evolution-
ary research. Surtout, it placed a piano
at the heart of the research environment
and provided room to build electronics
and house several computer stations.
As it was the first PianoLab project, its
methods of research and implementa-
tion were developed as we worked, à
solve the balance between what was
practical or possible artistically and
technically.

The team of four technologists and I
worked for an intense week of iterative
prototyping. Working in the same room,
with ongoing experimentation as part of
the development process, we designed,
implemented and tested the technolo-
gies with active feedback from me. Im-
provisation formed the bedrock for our
recherche, as the key focus was always
how the technology could be used to

52 Nicolls, Seeking Out the Spaces Between

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

Use of Performance
as a research Method
Prior to the final performance, we held
two pilot performances to further de-
velop the piece, the first of which took
place at the end of the first week of col-
laboration. Limited by the then-current
state of the technology, this performance
was incredibly informative for us, provid-
ing instant feedback.

For the programmers the perfor-
mance provided the opportunity to as-
sess the usability of the technology in
the context of a unified musical piece,
observing how well the necessary tran-
sitions between elements of the system

worked in the pressure of a real-time en-
virement, and whether this integration
of technology with musical performance
succeeded aesthetically. I made the fol-
lowing crucial discovery: If the technol-
ogy is already complex, then the actual
musical substance can be more direct or
simplified without lowering the overall
complexity of the artistry.

This question of where complexity
resides is also a powerful design point:
If one desires the gross result to be bal-
anced, then cases where more complex-
ity occurs at the interface level should
be balanced by less complexity in the
detailed physical control: c'est, quelques-

figue. 6. First prototype of Nicolls’s piano. (Photo © sarah Nicolls)

allow the performer to contribute to live
algorithms, manipulate the feedback and
create an engaging spectacle.

compositional Process
In an early brainstorming session we
came up with the idea of using a hat to
house a Bluetooth triple-axis acceler-
ometer (iCube Gforce 3D-3 v1.1). Ce
would give us an easily identifiable and
highly performative input mechanism.
I then asked the technologists to show
me their current work, to get an idea of
what might be possible in the time we
had (environ 2 months to the final
performance, avec 3 weeks allotted col-
laborative working time).

The piece evolved as a section-by-sec-
tion improvisation; I would practice with
pre-built systems or patches and suggest
ideas that could be created immediately.
The use of improvisation was a vital
mechanism for me to understand what
the pre-existing software systems did and
how they might interact with the live,
acoustic piano sound. Having decided
upon the hat as a major input device, je
also wore it while exploring different pia-
nistic textures and simultaneously seek-
ing out a gestural language with my head.
Our final output was a 20-minute
work for grand piano, electronic sound
and mechanical devices created using a
MIDI controller and pedal, nine DC mo-
tors and the top hat (figue. 4) (the per-
formance can be viewed on-line [22]).
The piano was placed in the middle of
a quadraphonic speaker system, en utilisant
contact microphones to avoid feedback.

technology Notes
The sensors in the hat triggered an algo-
rithm, which we developed to map the tilt
of the accelerometer to a 2D parameter
space with x and y mapped to the pitch
and temporal dispersion parameters of
a granular synthesis effect (figue. 5). Pour
the looping patch, analysis of the spectral
range of the live piano audio was used
as an onset and offset detector to trigger
suitable start and end points for the loop.
These loops were continuously stored as
indices of an audio buffer with a memory
de 1 minute. New onset events triggered
the playing of previously recorded loops
in a stochastic manner. The system was
designed so that the performer would be
able to fix the loop being used so that
it could provide a repetitive background
for further improvisation. The rhythmic
patterns used by the motors and piano
samples were generated using Markov
chains organized by varying degrees
of predictability, selectable by the per-
former using a MIDI controller.

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

Nicolls, Seeking Out the Spaces Between 53

thing with a lot of buttons or faders that
need to be used frequently might not be
best paired with detailed muscle tension
control. De la même manière, if balance is sought
between predictability and unpredict-
ability, then placing features with results
that cannot be predicted in certain “ar-
eas” of one’s instrument can create useful
creative springboards.

I also made discoveries about gestural
control of algorithms while playing, dans
particular related to the advantages of
the sensors placed in the hat: They af-
forded physically communicative control
without impinging on or affecting the
piano playing itself and thus were eas-
ily isolated from the other elements of
the music at any point. Feedback from
the audience was also immensely valu-
able and was acted on in the following
performance: The sensors used in the
first performance were placed in an in-
conspicuous hat, and we noticed that
several audience members did not make
the connection between the gestures and
the sound manipulations, as they seemed
intrinsic to the performance. Par conséquent,
the sensors were placed in a top hat for
the next performance, which the audi-
ence found very straightforward.

reflections from
the technologist(s)
The experience of working toward spe-
cific artistic goals, as opposed to scientific
ones, was both a novel and a rewarding
experience. With the focus on work-
able real-time implementations, alors que
demanding the technology produce a
subjectively interesting aesthetic, the pro-
cess led to extended discussion and large
output from all involved. In comparison
to work developed in a laboratory, the in-
stant feedback from the pianist allowed
quick identification of creative dead ends
and forced the focus upon the most in-
teresting ideas, with little misunderstand-
ing. Development “onsite,” with the
ability to immediately test ideas, led to
the elimination of erratic technological
behavior and a convergence toward the
technologies of the final piece.

reflections from the Performer
In developing interactive performance,
the complexity of the potential relation-
ships between gesture and sound is much
better expressed in real-time demonstra-
tion than in remote conversations or,
worse, written debate. For choosing the
correct input signal, guiding it through a
relevant process and producing a mean-
ingful output, I found the laboratory
method to be extremely liberating and
time-saving.

54 Nicolls, Seeking Out the Spaces Between

It was interesting to note the differ-
ent approaches of artist and scientist in
this context—the desire for immediate
visible or tangible elements with which
to play, measured against a detailed,
thorough and quite time-consuming
process of making the technology do
what we wanted. Michael Zbyszyn´ ski
[23] made reference to these paral-
lel demands when discussing a similar
project with Frances-Marie Uitti at the
Center for New Music and Audio Tech-
nologies and the need for technologists
to work much faster and on-the-fly than
would be appropriate for securing tech-
nology for an actual performance. Dans
reality, the balance between time in the
laboratory and the time apart worked
best for technological and musical
development.

Some points I had discovered previ-
ously were reinforced—for example,
that improvising with algorithmic sys-
tems can stimulate greater artistic free-
dom or range. Other discoveries were
new; one of the most important of these
was about the actual language of mu-
sic in interactive music. It is a potential
equation: If the understanding of the in-
teractivity in the audience’s perception
lends greater weight to the performative
communication, then demonstrative
and simply made gestures or musical
material can carry the message most
effectively.

Dans l'ensemble, the project served to high-
light the need for and benefit of this
kind of collaborative lab-based work,
especially when dealing with interactive
technologie.

Current develoPments
At the time of writing I have begun ex-
perimenting with a purely live sampling
scénario, where I can grab the sounds I
am currently playing by reaching into a
particular point in the air above the key-
board and then manipulate these with
different gestures. This research is be-
ing undertaken with Nick Gillian at the
Sonic Arts Research Centre, Belfast, us-
ing a Polhemus magnetic tracking device
[24]. Setting up PianoLab as a perma-
nent space is a long-term goal, et plus
le suivant 3 years we will also be develop-
ing further prototypes of the new piano
(figue. 6) [25]. To complete the circle,
this itself was the result of improvisation:
during the first PianoLab, I dismantled a
piano, hanging the soundboard from the
ceiling; while it hung there, I began to
imagine re-attaching a keyboard to it, à
create a new spatial relationship between
keyboard and strings.

acknowledgments

The HCI:ROYAUME-UNI 2008 collaboration was made pos-
sible by a (concernant)Actor3 Artist in Residence Commis-
sion, sponsored by the Centre for Digital Music,
Queen Mary, University of London, and produced
by BigDog Interactive Ltd. My initial research into
interactivity was funded by the Arts and Humanities
Research Council; the development of PianoLab was
enabled by a Brunel Research Innovation and En-
terprise Fund. I am grateful for further PianoLabs,
including those at CNMAT, Newcastle University
and the University of Cincinnati, and for input from
Pierre Alexandre Tremblay for Case Study 2.

references and Notes

1. Chikashi Miyama (Angry Sparrow) and Ben Knapp
(with Eric Lyon, Gascia Ouzounian), NIME 2009,
Concert 2, Friday, 5 Juin 2009. Voir (accessed 28 Décembre 2009).

2. Derek Holzer at noise=noise, Goldsmiths, Uni-
versity of London, 17 Février 2009, curated and
organized by Ryan Jordan.

3. SSS (Atau Tanaka, Cecile Babiole, Laurent Dail-
leau), stitched-up, sk-interfaces closing event, FACT
(Foundation for Art and Creative Technology), Liv-
erpool, 29 Mars 2008. Voir .

4. Miyama and Knapp [1].

5. Voir (accessed 28 Décembre 2009).

6. E. Jessop, “The Vocal Augmentation and Ma-
nipulation Prosthesis (VAMP): A Conducting-Based
Gestural Controller for Vocal Performance,” demon-
stration, NIME Pittsburgh, 2009.

7. J.. Richards, “Lost and Found: The Mincer,»
Leonardo Electronic Almanac 15, Nos. 11–12 (2008).
Available at .

8. Bob Ostertag, “Human Bodies, Computer Music,»
Leonardo Music Journal 12 (2002) p. 11.

9. G. Essl and S. O’Modhrain, “Enaction in the Con-
text of Musical Performance,” Interdisciplines virtual
workshop (by participants in Enactive interfaces Net-
travail) (2004) p. 1. Available at .

10. C. Cadoz and M. Wanderley, “Gesture—Music”
in M. Battier and M. Wanderley, éd., Trends in Ges-
tural Control of Music (Paris: Editions IRCAM, 2000)
pp. 71–93.

11. E. Miranda and M. Wanderley, New Digital Musical
Instruments: Control and Interaction beyond the Keyboard
(Wisconsin: A-R Editions, 2006) p. 10.

12. Arts and Humanities Research Council project,
May-–December 2007.

13. Richard Barrett, Adrift, PSI 09.10 CD. Recorded
live at The Warehouse, Londres, 29 Novembre
2007. Available at .

14. Performances: FURT, Red Rose pub, Finsbury
Parc, Londres, printemps 2007; and fORCH, Spitalfields
Festival, summer 2007. Voir .

15. Voir .

16. Richard Barrett, personal e-mail, Décembre 2009.

17. First performance, University of Huddersfield,
U.K., 6 Mars 2009.

18. First performance, Huddersfield Contemporary
Music Festival, Huddersfield, U.K., 21 Novembre
2009. Voir .

19. Whether to “reveal the magic and do the trick
anyway”—Augusto Corrieri summing up our pro-
cess for Soundwaves Festival 2007 .

Téléchargé depuis http://direct.mit.edu/lmj/article-pdf/doi/10.1162/LMJ_a_00012/1675178/lmj_a_00012.pdf by guest on 07 Septembre 2023

20. Centre for Digital Music at Queen Mary, Univer-
sity of London. Voir . Click on
Research/Centre for Digital Music.

21. S. Cornock and E. Edmonds, “The Creative Pro-
cess Where the Artist Is Amplified or Superseded by
the Computer,” Leonardo 6, Non. 1 (1973) pp. 11–16.

22. Voir

23. M.. Zbyszyn´ ski, “Augmenting the Cello” (NIME
Paris 2006).

25. Sunday Lunch Club series organized by Prototype
Theatre .

Michael Edwards I kill by proxy on sumtone records
.

Discography

Manuscript received 1 Janvier 2010.

The piano music of Niccolo Castiglioni on Metier .

Alexander’s Annexe Push Door To Exit on WARP
Records .

Sarah Nicolls is a pianist specializing in
contemporary music and live electronics,
regularly performing concerti with the Lon-
don Sinfonietta and featured on BBC Radio
3. Nicolls also plays in Alexander’s Annexe
(Warp Records). Nicolls is a Senior Lecturer in
Music at Brunel University. See also .