JMTE 6 (3) pp.

323–336 Intellect Limited 2013

Journal of Music, Technology & Education

Volume 6 Number 3
© 2013 Intellect Ltd Article. English language. doi: 10.1386/jmte.6.3.323_1

Adrian Moore, Dave Moore, Stephen Pearse and

Adam Stansbie
The University of Sheffield

Tracking production
strategies: Identifying
compositional methods in
electroacoustic music

Abstract Keywords
In recent years, electroacoustic theorists have drawn attention to the experiences and compositional methods
interpretations of listeners; well-known examples include the writings of Stéphane production strategies
Roy, Luke Windsor, Christiane Ten Hoopen and Denis Smalley. During the same electroacoustic
period, relatively little has been said about the methods and techniques employed acousmatic
by composers during the creation of their works. This article starts by considering Sound Composer
some of the reasons why methods have been overlooked within the existing body of software
literature. It goes on to discuss the existing software tool developed at the University SCMake
of Sheffield, and explains how this software documents the process chain and affords musical analysis
an interaction with the compositional process. The software is then contextualized tracking methods
within a proposed research project, which aims to further develop the software tools,
and deliver a series of complete compositional traces by a number of composers,
offering these for compilation and comparison. It finishes by proposing ways in
which compositional methods might be identified and analysed.

323
Adrian Moore | Dave Moore …

Introduction
The compositional methods employed within the electroacoustic tradition are
characterized by extreme heterogeneity. This is, at least in part, because the
term electroacoustic refers to a diverse body of art-music genres that devel-
oped in Europe, Japan and the Americas in the 1950s (Emmerson and Smalley
2001: 1); the convergence of these seemingly diverse genres, and the eclectic
mix of compositional methods and techniques found within them, has had
a profound effect on contemporary compositional practice within the elec-
troacoustic tradition. For example, the concrete methods pioneered by Pierre
Schaeffer and the Groupe de Recherches Musicales (GRM) are often cited as a
major influence on composers creating contemporary acousmatic music, whilst
the serial methods employed by composers working in the Westdeutscher
Rundfunk (WDR) in Cologne have clearly informed composers that are
currently involved in algorithmic and evolutionary composition. Despite the
ostensible differences that hold between these well-documented forms of
music-making, they remain (at least at the time of writing) part of the broader
electroacoustic tradition.
The mix of compositional methods found within the electroacoustic tradi-
tion is further influenced by a diverse range of contemporary (sometimes
‘mainstream’) compositional genres; these add yet further complexity to an
already diverse network of influences, leading Trevor Wishart to highlight
some of the manifold links that hold between various different contemporary
forms of art-making:

We have reached an era in which the electroacoustic arts have become

normalised. The sophisticated large-scale structuring of sound in which
I’m interested takes its place in a continuum of possibilities from musical
‘Kunst’, through popular culture to pure amateur messing about with
sound. […] In this context where there is no longer any absolute divide
between ‘high art’, popular culture and reportage, either professional
or amateur, it is possible for electroacoustic projects to grow out of very
local situations not even connected with the arts world […].
(2008: 137–38)

Wishart goes on to suggest that modernist tendencies, so clearly associated

with the early days of electroacoustic composition, have virtually disappeared;
contemporary compositional practice draws influence from an ever-increasing
body of genres and traditions and, by extension, the compositional methods
found within them.
The recent proliferation of affordable and accessible sound technologies
has further diversified compositional methods in the electroacoustic tradi-
tion; contemporary composers have a bewildering array of hardware and soft-
ware tools at their disposal and it seems reasonable to suggest that such tools
have a radical impact upon their compositional methods and techniques. In
some cases, tools facilitate the use of particular methods, perhaps enabling
composers to realize creative goals that were previously beyond their reach.
In other cases, tools may inspire the development of new methods, enabling
composers to further develop their compositional practice in the light of tech-
nological advances. Unfortunately, the term electroacoustic implies that sound
technologies have been employed during the compositional process, but
tells us little about the resulting sound-world or the distinctive idioms made

324
 Tracking production strategies

possible by such technology (Emmerson and Smalley 2001: 2) and even less
about the methods that have been employed during the act of composition.
For some, the diversity mentioned above renders that the term electroacous-
tic ‘a meaningless catch-all’ (Chion 1982, cited in Dhomont 1995: 1), analogous
to applying the term acoustic music to define the entire instrumental repertoire
(Dhomont 1995: 1). For others, diversity distinguishes the electroacoustic tradi-
tion, ensuring ongoing potential for originality and creativity (Stansbie 2012).
Either way, it remains difficult to generalize about the kinds of compositional
methods that may be employed within the electroacoustic tradition, and this
may help to explain why electroacoustic theorists have tended to focus upon
reception strategies – the experience and interpretations of listeners – as opposed
to production strategies – the methods and techniques employed by composers
during the creation of their works (Stewart 2007). Some of the most famous
examples of texts dealing with reception strategies include: Stéphane Roy’s
theory of functional analysis (2000); Luke Windsor’s perceptual approach to
electroacoustic music (1995); Christiane Ten Hoopen’s thesis on perceptions
of sound and human presence (1997); Denis Smalley’s extensive descriptive
terminology, as found in his papers on spectromorphology and space-form
(1986, 1997. 2007). In all of these cases, the focus remains upon the perceptual
responses of listeners as opposed to the methods of production found within
the electroacoustic tradition.
This article focuses upon production strategies. Section 2.0 focuses upon
some of the existing ways in which such strategies may be identified and
discussed, before highlighting the need for a means of identifying and tracking
them. Section 3.0 introduces the Sound Composers software (SCMake
and SCMix); these tools, developed by Dr Dave Moore at the University of
Sheffield, were not originally designed to track production strategies but
their recent use has opened up possibilities for doing so, thus throwing light
on the methods employed by electroacoustic composers. In Section 4.0, the
software is contextualized within a research project that delivers a series of
complete compositional traces by a number of composers for compilation and
comparison. The article finishes by highlighting some of the various ways
in which compositional methods might be discerned from these exemplars
and how beneficiaries may further use the Sound Composers Software as an
analytical resource.

The search for production strategies

The problem with identifying compositional methods in electroacoustic
music does merely relate to their extreme heterogeneity (described in the
introduction); it also relates to the medium employed in the creation of such
music and the means of dissemination. Most works of electroacoustic music
are composed using a fixed medium (such as a CD, DVD or computer hard-
drive) and are presented using loudspeakers; both the medium and the loud-
speaker conceal the methods employed by the composer. We shall briefly
consider this point.
In some musical traditions, composers work with musical scores; these
may be studied aside from the sounding content of the musical work, and
in some (but certainly not all) cases compositional methods may be identi-
fied. Electroacoustic composers rarely provide musical scores and, as a result,
this option is rarely available; unless they are writing for instruments, electro-
acoustic composers issue a fixed medium rather than a musical score. There

325
Adrian Moore | Dave Moore …

1. For more information is, of course, a set of data associated with any fixed medium, and some have
on the differences
between gestures,
referred to such data as a form of score. For example, Dale Perkins notes that
sources and causes see: electroacoustic composers either enter data (directly, often in numerical form)
‘Spectromorphology: or create data (indirectly, by manipulating computer programmes that subse-
Explaining sound
shapes’ (Smalley 1997). quently produce data) in order to realize certain compositional goals (2007);
he goes on to describe such data using the term data score. Perkins recog-
nizes that electroacoustic composers rarely think in terms of score production.
However, he believes that data scores underpin their creative acts and are
therefore fundamental to the production of acousmatic music: ‘Interestingly,
whether the composer […] is conscious or not of the score(s) s/he has gener-
ated, the concept of score remains highly important to the production of a
work. […] behind the graphic interface data is generated which the software
application and computer responds to’ (Perkins 2007: 31–32). Perkins goes
on to suggest that: ‘the data score instructs the software instrument how to
perform sounds’ (Perkins 2007: 31).
The notion of a data score provides a useful way of thinking about the rela-
tions that hold between pieces of software and resulting sound materials.
However, it does not necessarily help in the search for compositional methods;
there are three reasons for this: (1) the data that one enters into a particular
piece of software do not necessarily convey any readable information about the
resulting sounds that will be produced; there is no learnt code or language that
would enable one to examine a data score to gain an insight into the sound-
world of the related composition (2) the data that is entered is not the only
variable; the output depends upon the sounds that are being processed and,
unless these sounds are known, the data score will provide limited informa-
tion about the methods employed (3) the electroacoustic composer will often
use multiple processes, and multiple pieces of software or hardware, to create a
single piece of music; if one ascribes to the notion of the data score, one must
accept that every electroacoustic work could have a potentially vast number of
data scores. Clearly, therefore, compositional methods in the electroacoustic
tradition cannot be determined by considering a musical score, or a data score.
With the above in mind, one may wonder whether compositional methods
may be identified through the act of listening. Unfortunately, this is rarely the
case; works of electroacoustic music are typically presented over loudspeakers,
which invariably hide the methods employed by composers; loudspeakers
necessarily separate listeners from the kinds of sound-producing gestures,
sources and causes1 encountered in other forms of music-making and, further
to this, many electroacoustic composers (particularly those in the acousmatic
tradition) intentionally explore such ambiguities within their compositions.
Although these ambiguities may constitute an essential aspect of the language
of electroacoustic music, they make it difficult to judge what a composer
has actually done to produce particular sound materials. For example, it is
often difficult to determine whether a given sound has resulted from an
act of synthesis or recording; this basic example serves to highlight the fact
that listening, in the absence of further information (programme notes, for
example), does not necessarily provide any meaningful insight into aspects
of the compositional process. Further to this, some electroacoustic composers
are extremely wary of technological listening – a term invented by Smalley to
describe something that: ‘occurs when a listener “perceives” the technology
or technique behind not the music rather than the music itself, perhaps to
such an extent that true musical meaning is blocked’ (1997: 109). Smalley goes
on to say:

326
 Tracking production strategies

Many methods and devices easily impose their own spectromorpholog- 2. Denis Smalley
invented the term
ical2 character and clichés on the music. Ideally the technology should spectromorphology to
be transparent, or at least the music needs to be composed in such a refer to the spectral
way that the qualities of its invention override any tendency to listen content of a sound
and the way that
primarily in a technological manner. such content shapes,
(Smalley 1997: 109) or morphs, over time
(Smalley 1986, 1997).

Smalley’s argument is perhaps contentious. However, many electroacoustic

composers do try to conceal the technologies employed during the act of
composition in order to counter technological listening. In such cases, they
often disguise their compositional methods, meaning that the quest to uncover
such methods is unlikely to succeed solely through an act of listening.
An alternative approach may be to ask composers about their working
methods, perhaps by conducting a spoken interview or written question-
naire. Barring the obvious limitations associated with composer availability
and access, this approach would enable one to consider specific pieces (and
methods) in the light of comments made by the composer. There is, of course,
a precedent for this approach. For example, in a recent interview, Smalley
provided the following answer to a question about his method of working:

When I started out I followed the basic French, musique concrète

method, as taught at the GRM, and in principle this has stayed with me.
First discover and then record your source sounds; sort and catalogue
them using pertinent criteria (I keep card indexes), which may be spec-
tromorphological or refer to source bonded qualities; experiment with
transformations to create families; along the way try out combinations
and sequences through mixing, to see if relationships are going to work;
constantly assess whether there is sufficient variety and contrast in the
nature of the sounds and in the ways they are developing. The gradually
the piece emerges: form grows out of materials.
(Gayou 2010: 15)

In this short statement, Smalley aligns his compositional methods with those
developed and explained by Schaeffer and, in doing so, he communicates
the main thrust of his compositional approach with relative ease. Despite
this, there are obvious limitations; Smalley’s response is relatively brief, it
lacks specific details and he does not provide concrete examples, and as a
result leaves many potential questions unanswered. For example: What does
‘discover’ mean in this context? What are the implications of recording in this
context? What does the process of sorting and cataloguing entail and what
is relevant in terms of the compositional process? What does experimenta-
tion entail? and so on. This is often the case when considering responses that
are provided during a spoken interview; composers resort to familiar terms,
phrases and sentiments, which, given the need to be brief, often fail to provide
a full and detailed account of the manifold and complex methods that will
have been employed.
An alternative approach would be for composers to become involved in
the process of documenting their own compositional methods; they could
keep a log or diary that highlights the various stages in the completion of
particular works or, alternatively, provide an audio commentary associ-
ated with each piece. Once again, a precedent has been set. For example,
Schaeffer’s A la Recherche d’une Musique Concrète/In Search of a Concrete Music

327
Adrian Moore | Dave Moore …

3. The First Journal ([1958] 2013) offers an extremely detailed account of his various compo-
of Concrete Music
(Schaeffer 2013: 3–66)
sitional experiments in the form of a diary. In this case, the reader gains a
is filled with examples surprisingly frank account of the various successes and failures that Schaeffer
of dead ends and encountered in his attempts  to elaborate a new music for the twentieth
problems that Scheffer
encountered during century, including his thoughts about musical scores, musical instruments,
the development recorded sound materials, various technologies, text-based concepts, contem-
of concrete music, porary practitioners and traditions, amongst others. The methods outlined in
particularly those
associated with the this legendary text have clearly informed numerous aspects of contemporary
composition of Etude music-making  – particularly in the electroacoustic tradition – and one may
aux Chemins de Fer.
wonder whether the detail and clarity of Schaeffer’s documentation of the
4. Git was designed by compositional process is partly responsible for its enduring value to the elec-
Linus Torvalds for the
development of the
troacoustic community.
Linux Kernel. It is one of Schaeffer is not the only person to provide an account of his compositional
a number of concurrent methods. For example, in Sound Composition (Wishart 2012), Wishart provides
version control systems
enabling simultaneous an extremely detailed and clear account, adopting a style and approach that
development and links with his previous writings whilst an accompanying CD provides sound
revision history. examples abstracted from his various compositions. The accompanying sound
examples are extremely valuable, since listeners have relatively few opportu-
nities to engage with elements of a given work that have been abstracted from
the finished piece. Even so, Wishart’s discussion focuses almost exclusively
upon finished pieces and, as a result, upon those successful, tried-and-tested
methods that resulted in materials in his pieces; this marks a radical departure
from the approach adopted by Schaeffer, who documented various composi-
tional ‘dead ends’ and thus highlighted perceived failing in his own methods.3
Both are clearly valuable. However, they are also remarkably rare; there are
few other examples that offer such a candid account of the compositional
methods and techniques employed during the composition process.
The highly personalized accounts described above provide valuable
insights into the compositional methods of particular composers. However,
they tell us relatively little about methods found within the electroacoustic
tradition at large; it is difficult to know whether certain methods have been
pioneered by individual composers or whether they are used more generally by
members of the electroacoustic compositional community. A more substantial
set of documented methods may serve to clarify the situation. However,
unless composers, who may understandably feel that they do not have time to
engage in writing the kind of detailed documentation or that it would disrupt
and even hinder the compositional process, start to provide such documenta-
tion the picture will remain extremely fragmented. If compositional methods
are to be identified and understood, then we need an alternative approach.

Sound Composers Software

This section introduces the Sound Composers Software, SCMake, before
considering how such software may be used to identify, track and under-
stand compositional methods. A broader research project, in which the Sound
Composers Software is central, is introduced in Section 4.0.
SCMake was designed by Dr Dave Moore at the University of Sheffield in
order to explore the various ways in which sequences of audio transformations
may be tracked using version control systems (VCS). Version control systems
(such as git4) manage and track changes that have been made to individual
documents; they look for changes that have been made to a given document,
snap in and out of a tree of development and replace or merge difference

328
 Tracking production strategies

between the new file and the old file. This makes them extremely efficient at 5. When using SCMake,
transformations are
tracking changes to text documents. However, VCS are much less efficient scripted, meaning that
when handling audio files; changes to such files often produce substantial any command line
differences between the new file and the old file. For example, if reverberation audio environment
can be used. For
is added to an audio file, every stored sample is likely to change; this typically example, entire CSound
means that the new file must be stored in its entirety, and thus one cannot scores can be used as
simply replace or merge differences as one might when working with text transforms.

files. With this in mind, SCMake was developed; incorporating the SoX toolkit
and basic Csound files, SCMake tracks changes made to audio files but only
stores a process identifier and an instruction set (a command with variables
and breakpoint files) rather than a collection of newly produced audio files.5
These instructions enable one to recreate any transformation, or set of trans-
formations, using only the source sound(s) and a data set.
SCMake is entirely non-destructive; transformations are recreated,
producing an output file that a database registers and monitors. In order to
guarantee accuracy in recreation, SHA-1 hashes are used to check that the
recreated files matched the hash of the original transformation. The data-
base at the heart of each set of transformations contains a complete history
of each action, marks all changes and stores this information in Extensible
Markup Language (xml). Within any given database, each file’s dependencies
and transformation history are stored so that any sound can be recreated in a
single command. As a result, the system has the ability to track and recreate
the entirety of the sound generation and transformation process. In order to
clarify and document this tracking process, SCMake comes with a collection

Figure 1: Three key transformation categories.

329
Adrian Moore | Dave Moore …

6. The Free Sound Object of ‘housekeeping’ functions, which enable one to clean a project by removing
Mixer (FSOM) can be
accessed here: https://
extraneous files, rebuild all files on demand and draw graphs showing the full
github.com/spearse/ project history.
fsom. Initial research identified three key transformation categories: one-to-one,
7. ‘Compose With Sounds’ divergent and convergent transformations, as shown in Figure 2. One-to-one
is an open source, transformations represent an audio process that works on a given audio file
cross-compatible
digital audio to produce a singular output. Convergent transformations, such as mixing
workstation developed or convolution, are instances whereby several input files are used to create a
under an European resultant file. Divergent transformations represent processes whereby several
Commission FP7
project led by Professor output files are created.
Leigh Landy at De In each of these three key transformation categories, processes can be split
Montfort University.
The software was
into several singular transformations. For example, divergent processes may
mainly created at the be remapped into multiple one-to-one versions in order to produce single
University of Sheffield. output files.
Information about the
Compose With Sounds Free Sound Object Mixer6 (FSOM) is a purpose-built C++ library that
project and software was written in conjunction with SCMake and Compose with Sounds7 at
may be found here: the University of Sheffield by Dr Dave Moore, David Devaney and Stephen
www.cws.dmu.ac.uk.
Pearse. The purpose of FSOM was to allow audio to be sequenced in an envi-
ronment whereby tracks and audio busses were not required. In this context,
all processing and transformations apply on a region by region basis. SCMix
is proposed as a front-end graphical program (DAW) to be built on SCMake
and FSOM technology. SCMix would exist both as a graphical composition
tool allowing sequencing data, such as timings, amplitude levels and automa-
tion data, to be stored in the SCMake format whilst remaining transparent to
the composer. In doing so, SCMix would log the sequencing and transforma-
tion history in an SCMake database, thus enabling every aspect of the compo-
sitional process to be tracked.
During the original SCMake experiment, the test composer (Adrian
Moore) was able to study his own compositional processes and methods by
exploring the various SCMake housekeeping functions; by treating composi-
tions as a complex data set, SCMake offers access to the all of the various
transformations undertaken by the composer, producing a ‘trace’ from which
composers may observe traits of process and selection at both micro and
macro levels. For example, one may study the development of a single sound
by observing and auditioning all of the various stages of its transformation
from start to finish, encountering all the various dead ends along the way. In
addition to the focus upon individual sounds, one may observe or recreate
specific composites and sub-mixes that may, once again, have been used or
discarded during the act of composition. In doing so, one may build up a clear
picture of the compositional process and retrace (in full) all of the composer’s
various actions. At this stage, one may begin to build up a detailed picture of
the compositional methods employed during the creation of a work. In the
following example (Figure 2), we can see how three source sounds have been
developed throughout a compositional process; we can see a range of one-to-
one, convergent and divergent processes and, even in this relatively simple
example, the complex trace of the composer’s actions.
By retracing the actions of a composer, their various methods and techniques
may be identified, examined and discussed. However, SCMake offers further
possibilities; process chains may be compared across entire works, potentially
highlighting more generalized approaches to the development of sound mate-
rials. For example, one can see whether a composer always approaches sound
sources by using the same set of processes, or whether their choice of process

330
 Figure 2: An example of a project history using three source-files.

331
Tracking production strategies
Adrian Moore | Dave Moore …

differs depending on the sonic character of the source; this would enable one
to link specific processes with specific sorts of sound materials and thus iden-
tify common approaches that a composer adopts during the compositional act.
Further to this, SCMake offers an ideal opportunity for an even broader explo-
ration of methods; rather than tracking individual composers, one may compare
decision chains across several works and several composers and, in doing so,
identify common approaches and, crucially, common methods (should they
emerge). This would be most effective if a number of composers worked
with the same set of source sounds and processes, so that direct comparisons
between the resulting data sets could be compared and contrasted. In doing
so, one may discover striking similarities (or, indeed, differences) between: the
selection of sound materials; the selection of particular transformations; the
order in which transformations are used; the amount of processing that takes
place with each chosen sound; the amount of processing that takes place before
a mixing stage begins (if there is a defined ‘mixing’ stage); the various ways in
which submixes have been formed; the number of submixes that are created
before the larger forms begin to be defined; and so on. At this stage, one begins
to see the potential of SCMake; the compositional ‘traces’, or data sets, left by
composers using the software are ripe for analysis and may reveal a wealth of
information about the methods and techniques employed by composers.
At the time of writing, there are certain limitations with the Sound
Composers Software and, as a result, its ability to track various different kinds
of compositional methods. For example, consider the following two methods
for the development of sounds: (1) The use of command lines with break-
point files (semi-structured, predictable and repeatable) and (2) Free play or
‘performance in the studio’. Both methods offer a composer the opportunity
to audition and develop sound materials during the act of composition, yet
free play gives rise to enormous proliferation of data, which are, for the time
being, too complex to track and recreate. As a result, SCMake supports and
tracks method 1. but does not currently support method 2. Command-line
processes are much more suited to SCMake because they offer a series of
discrete steps that are much easier to track, but this invariably means that
the software may only provide meaningful results if the chosen composers
use trackable methods. Along similar lines, although SCMix would track the
mixing process, mixing also blurs the boundary between the discrete and the
continuous and is, as a result, hard to track.
In addition to the various technical limitations discussed above, the data
produced by the Sound Composers Project will only provide a snapshot, or
trace, of technical procedures undertaken by the composer; the composer’s
thought processes, aesthetic intentions and decisions and thoughts about the
compositional process itself will remain elusive. Even so, assessment of the
data trace may enable an analyst to draw reasonably accurate conclusions as
to the intentions of composers. For example, at a local level it is possible to
determine how a technique has been used by assessing the degree to which
certain (prominent) parameters have been used. For example, mining ‘Q’,
‘centre frequency’ and ‘bandwidth’ in filter usage would indicate whether the
filter was being used more as a mastering tool (relatively large bandwidth)
than to shape frequency content (progressively higher Q). Understanding
where filter Q changed over a work’s progress might indicate where pitch
(centres) became important. Reverberation parameters such as width and
decay time will indicate (in conjunction with the position of the sound in the
resultant mix) whether reverberation is being used in a vertical structure (as

332
 Tracking production strategies

part of a layering process) or horizontal structure (a decayed ending). Effects

identified as having mild or strong influence on their sounds will indicate the
‘choice chain’ of the composer and potentially say a great deal about how
a composer hears their sounds in relation to the effects palette in front of
them. In short, the ability to mine the system to search for process chains
that concretely relate to a starting sound and a finishing sound (no matter
how fragmented this process may be) implies that sustained usage across
multiple composers could illuminate a more robust aesthetic context. Clearly,
the Sound Composers Project sets out to deliver data that could potentially be
useful for such identification.

The Sound Composers Project

This section considers the future of the Sound Composers Project and
introduces a research project that has the following five objectives:

1. Create a self-contained composition environment that tracks the compo-

sitional process
2. Create and publish a series of compositions and their associated traces
3. Evaluate and analyse compositional methods in relation to the series of
compositions and traces
4. Produce and disseminate an account of compositional methods in
electroacoustic music
5. Make all of the research software publicly available.

The project will run for a period of two years. During the first year, further
software development will take place. A team of two composers (Adrian
Moore and Adam Stansbie) and students at the University of Sheffield will
produce musical works using the software and, feeding back to two engineers
(Dave Moore and Stephen Pearse), make suggestions for development of both
the toolset and the mixer. It is hoped that the composer’s toolset will ulti-
mately comprise the vast majority of time and frequency domain tools, granu-
lation, alongside graphical editing and mixing with real-time in-line effects
sends. During the same period, a number of analytical tools will be provided
to enable future users to search, filter and translate the compositional trace
into a variety of visualizations. This may include documenting processes used
over the complete work or identifying sounds where certain effects have been
used. Completed works will exist in a format that can be rebuilt from source
material, potentially affording further opportunities for intervention and inter-
pretation, especially for the purpose of analysis.
During the second year, a series of commissioned composition experi-
ments will take place; a range of composers will be invited to create musical
works using the evolved software. All of the composers will be limited to the
same selection of processing tools and the same selection of sound materials;
by minimizing variables, one will able to make direct comparisons once the
various pieces have been completed. Upon the completion of their works, we
will be left with the original set of sound sources and a chain of events docu-
menting the compositional methods employed. This will be made available to
the electroacoustic music community, who will have access to the composers’
music, the traces (for recompilation) and the software all supported by an
interactive website. We hope that this resource would encourage further anal-
ysis of the works. Ultimately, the aim of the Sound Composers Project is to

333
Adrian Moore | Dave Moore …

illuminate methods within the electroacoustic community and thus provide

greater clarity as to the practice at large.

Conclusion
This article started by considering the ostensible impenetrability of composi-
tional methods employed by composers of electroacoustic music. It went on
to introduce the Sound Composers Project, and the software that has been
built at the University of Sheffield (particularly SCMake and FSOM). SCMake
has shown that it is possible to track key transformation categories and repro-
duce the resulting audio from the original source files. Early experiments with
a single composer have allowed us to produce a basic analysis of a composi-
tional process and suggested further possibilities for examining methods in
the future. The final section outlined a project that is in the early stages of
development and proposed a series of experiments researching compositional
methods that will be underway in the near future.

References
Chion, M. (1982), La musique électroacoustique, Paris: PUF.
—— (1983), Guide des objets sonores: Pierre Schaffer et la recherche musicale,
Edititions Buchet/Chastel.
—— (2009), Guide to Sound Objects (trans. J. Dack and C. North), EARS:
Leicester, http://www.ears.dmu.ac.uk/IMG/pdf/Chion-guide/. Accessed
6 June 2012.
Dhomont, F. (1995), ‘Rappels acousmatiques’/‘Acousmatic update’, Contact!,
8: 2, pp. 1–4.
Emmerson, S. and Smalley, D. (2001), ‘Electro-acoustic music’, Grove Music,
http://www.grovemusic.com/index.html. Accessed 18 May 2007.
Gayou, E. (2010), ‘Interview with Denis Smalley’, in E. Gayou (ed.), Denis Smalley,
Portraits Polychromes, Paris: Institut National de l’Audiovisuel, pp. 7−18.
Perkins, D. (2007), ‘Portfolio of Original Musical Compositions and Written
Commentary’, Ph.D., Dissertation, Leeds: The University of Leeds.
Roy, S. (2000), ‘L’analyse de la Musique Acousmatique: bilan et propositions’,
Ph.D. dissertation, Montreal: Université de Montréal.
Schaeffer, P. ([1958] 2013), A la Recherche d’une Musique Concrète/In Search
of a Concrete Music (trans. C. North and J. Dack), Berkeley: University of
California Press.
—— ([1977] 1966), Traité des object musicaux: essai interdisciplines, 2nd ed.,
Paris: Seuil.
Smalley, D. (1986), ‘Spectro-morphology and structuring processes’, in
S. Emmerson (ed.), The Language of Electroacoustic Music, Basingstoke:
Macmillan Press, pp. 61−93.
—— (1997), ‘Spectromorphology: Explaining sound shapes’, Organised Sound,
2: 2, pp. 107–26.
—— (2007), ‘Space-form and the acousmatic image’, Organised Sound, 12: 1,
pp. 35−58.
Stansbie, A. (2012), ‘Between plasticity and performance: An introduction to
acousmatic art’, Digital Art Criticism: Sonic Arts, Taiwan: Foundation of
Digital Arts, pp. 132–144.
Stewart, I. (2007), ‘Functional analysis and electroacoustic composition:
Theory, extensions and implications’, Ph.D. dissertation, London: City
University.

334
 Tracking production strategies

Ten Hoopen, C. (1997), Perceptions of Sound. Source, Cause and Human Presence
in Electroacoustic Music, Amsterdam: University of Amsterdam.
Windsor, L. (1995), ‘A perceptual approach to the description and analysis of
acousmatic music’, doctoral thesis, London: City University.
Wishart, T. (1994), Audible Design, York: Orpheus the Pantomime.
Wishart, T. (2008) “Globally Speaking” Organised Sound, 13: 2, pp.137–140
—— (2012), Sound Composition, York: Orpheus the Pantomime.

Suggested citation
Moore, A., Moore, D., Pearse, S. and Stansbie, A. (2013), ‘Tracking production
strategies: Identifying compositional methods in electroacoustic music’,
Journal of Music, Technology & Education 6: 3, pp.  323–336, doi:  10.1386/
jmte.6.3.323_1

Contributor details
Adrian Moore is a composer of electroacoustic music. He mainly composes
music for fixed formats (CD, DVD), music intended for ‘sound diffusion’ over
multiple loudspeaker systems. He occasionally writes for instruments, often
with a live processing element using Max-MSP and custom-built software. He
directs the University of Sheffield Sound Studios (USSS) where researchers
and composers collaborate on new musical projects. Adrian Moore’s research
interests are focused towards the development of the acousmatic tradition
in electroacoustic music, the performance of electroacoustic music, signal
processing and human-computer interaction in music. His music has been
commissioned by the Groupe de Recherches Musicales (GRM), the Institute
International de Musique Electroacoustique de Bourges (IMEB) and the Arts
Council of England. A significant proportion of his music is available on
three discs, ‘Traces’, ‘Rêve de l’aube’ and ‘Contrechamps’ on the Empreintes
DIGITALes label. For more information, visit: www.adrianmoore.co.uk.
Contact: Sheffield University, Music Department, 206 Jessop Building, 34
Leavygreave Road, Sheffield S3 7RD, UK.
E-mail: [email protected]

Dave Moore is a teaching associate and studio technician at the Department of

Music, University of Sheffield. His research software for large array sound diffu-
sion has featured in concerts around the world. He teaches Music Technology
and Computer Music Programming at undergraduate and postgraduate levels
and manages the infrastructure of University of Sheffield Sound Studios. He
has extensive experience in the field of music software development and is
the principle software developer in a number of active open source projects.
He is a specialist in open source music technology tools including Pure Data,
CSound, Blue, Python and C++.
Contact: Sheffield University, Music Department, 206 Jessop Building, 34
Leavygreave Road, Sheffield S3 7RD, UK.
E-mail: [email protected]

Stephen Pearse is a composer and audio software engineer nearing the

completion of a Ph.D. under Dr A. Moore and Dr D. Moore at the University
of Sheffield in the United Kingdom. His compositions have been played
across the United Kingdom, Europe and more recently at the New York City

335
Adrian Moore | Dave Moore …

Electroacoustic Music Festival (2013). His primary research interests consist

of acousmatic composition, compositional interfaces, audio software design
and live coding. Pearse recently completed work as a software engineer for De
Montfort University in Leicester (UK) under Professor Leigh Landy engineer-
ing ‘Compose with Sounds’, an open source, cross-compatible digital audio
workstation. His recent research has taken the form of a scriptable, multi-
threaded artificial intelligence and synthesis system that uses image streams
as control data. He regularly lectures on a variety of Music Technology-related
courses at the University of Sheffield.
Contact: Postgraduate office, Sheffield University, Music Department, 206
Jessop Building, 34 Leavygreave Road, Sheffield S3 7RD, UK.
E-mail: [email protected]

Adam Stansbie is known for his electroacoustic/acousmatic compositions,

which have been widely performed (throughout Europe, Asia, North and South
America and Australasia), published (Elektramusic, Musique et Recherche,
Taukey and Sargasso) and prized (IMEB, Musiques et Recherches, Destellos
Foundation). Alongside his creative work, Adam has written extensively on
the presentation and performance of acousmatic music and he is currently
interested in the various ontological/phenomenological paradoxes that the
acousmatic tradition seems to produce. He has taught at a number of higher
education institutions and is currently Lecturer in Music Technology at The
University of Sheffield. For more information, visit: www.adamstansbie.com.
Contact: Sheffield University, Music Department, 206 Jessop Building, 34
Leavygreave Road, Sheffield S3 7RD, UK.
E-mail: [email protected]

Adrian Moore, Dave Moore, Stephen Pearse and Adam Stansbie have asserted
their right under the Copyright, Designs and Patents Act, 1988, to be identified
as the authors of this work in the format that was submitted to Intellect Ltd.

336
Copyright of Journal of Music, Technology & Education is the property of Intellect Ltd. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.