Listeners can only apprehend music if they discover a perceptual affinity with its materials and structure. Such an affinity depends on the partnership between composer and listener mediated by aural perception.... The primacy of perception is unassailable since without it musical experience does not exist. [Smalley, 1986]

The significance of the electroacoustic medium is that, like cinema, it is a medium of illusion. Composers have available as raw material any sound (recorded, synthesised, processed...) and therefore any illusion, so the basic carriers of musical meaning are not confined to simple parameters such as pitch and duration, but may be ambiguous and difficult to identify. Structures may instead rely purely on layers of association or recognition, or a host of other parameters, easily comprehended, but not usually associated with music. We are dealing with the articulation of illusory objects in virtual space and such illusory spaces and objects may be "real", surreal or abstract.

As Denis Smalley reminds us, the listener needs to know what aspect of the musical material defines structure. Whatever this might be, in electroacoustic music, as with any other music, it is the function of performance to communicate that structure in a particular space at a particular time. I would like to propose that the "gentle art" of sound diffusion should be seen as a necessity rather than an option in the public presentation of electroacoustic music, as it is only through careful performance that such "carriers of meaning" can be effectively communicated.

Jeff Pressing suggests [Pressing, 1994]

"..we know that one of the reasons that listeners show limited liking for contemporary art music is their inability to code it either on the basis of simple pitch or rhythmic structures, and hence assign it meaning"

As the meaning in electroacoustic music often lies in other parameters this is hardly surprising. He goes on to suggest, however that:

" we are making computer music "unphysical" in failing to tap basic cognitive motor designs used in human communication...Mainstream computer music seems to harbour a resentment towards things dance-like or motoric....[which] seems to stem from intellectual positions that de-emphasise the importance of the body."

I am not sure what music Pressing has in mind in his criticism as this is simply not true of much electroacoustic music, and I think it is key to our understanding of the music that we realise that it very often does "tap basic cognitive motor designs" in simple and direct ways. However, if this is not in the rhythmic sense that Pressing seems to be suggesting, what is the physical nature of the experience of electroacoustic music?

I would like to suggest that it is our listening experience of the physical, sounding world which is key to our understanding of both the illusory objects and virtual space of electroacoustic music, and if the music appears to be "unphysical" it is often the fault of poor performance.

Trevor Wishart observes that when we hear a sound we infer something about what physical action caused it - we hear the difference between something being plucked, scraped, struck and blown. And furthermore, we instinctively try to infer such actions/causes even for sounds we do not recognise, even those which may be completely synthetic, or "unreal". And whilst we can choose to ignore our everyday experience and suspend the inference of "reality", we do so "in the knowledge that this background exists." [Wishart, 1985]

Denis Smalley, too, notes:

Not only do we listen to the music, but we also decode the human activity behind the spectromorphologies through which we automatically gain a wealth of psycho-physical information.... [Smalley, unpublished]

He calls this property

...source bonding...: the natural tendency to relate sounds to supposed sources and causes, and to relate sounds to each other because they appear to have shared or associated origins.

and proposes that these relationships hold for all musics. Abstract, instrumental music is no exception and it is not only the physical world which is significant to our understanding:

The listener’s experience of listening to instruments is a cultural conditioning process based on years of (unconscious) audio-visual training.

A knowledge of sounding gesture is .... culturally very strongly embedded. This cannot be ignored and denied when we come to electroacoustic music. It is particularly important for acousmatic music where the sources and causes of sound-making become remote or detached from known, directly experienced physical gesture and sounding sources.

[Smalley, unpublished]

But it is not only the perceived source object and cause which are significant

... in mixed work [works for tape with instruments] the perceived behavioural relationships between the visible, gesture-bearing performer and the surrounding acousmatic context will be crucial to the work’s understanding.... Spatial perception is inexorably bound up with spectromorphological content.....[and] Space, heard through spectromorphology, becomes a new type of "source" bonding. [Smalley, unpublished]

And Wishart uses the term "landscape" to describe the "virtual acoustic space" which is the imagined source of the sounds we hear, and describes four types of landscape based on the combinations of real and unreal, objects and spaces.

What these writers have in common is a perception of the materials of electroacoustic music based on patterns of human perception which extends beyond traditional musical parametrisation, allowing an overview across the range of possibilities and suggesting a model of how these diverse materials can generate musical meaning. Most significantly they suggest that we instinctively try to make sense of a (new) sound by attributing a source object with an associated action or cause, that we interpret its perceived location and spatial behaviour, and that or responses are based on our instincts and real-world experience.

So how does this recognition of our subconscious understanding of this material help us? In what ways do imaginary objects in virtual space articulate musical structure?

John Young suggests that our instinctive response along a "reality-abstraction continuum" becomes a powerful tool for structural interpretation of a work. In addition to the implied source,

"abstraction" is a measure of the psychological distance between a sound which displays a source-cause ambiguity and a surmised source-cause model. [Young, 1994]

He goes on to propose "juxtaposition and mediation" as structural tools:

As soon as sounds are articulated in a tangible three-dimensional spatial field, an important aspect of environmental reality has [to] be analogised...... Although [a] sound itself ...[may not be]... specifically from a particular environmental or cultural source, it [may] nevertheless serve to define a "realistic" acoustic space and behave as though it were a physical entity... [Young, 1994]

So it is in perceiving the distances (physical, psychological etc., analogous, say, to harmonic distance in tonal music) that we infer structure.

Wishart, too, discusses the creative possibilities of using "landscapes" for making structure.

The acoustic space which we represent need not be real and we may in fact play with the listener’s perception of landscape. This aspect of sonic architecture was not an aspect of the traditional craft of the musician.....It is therefore easy to dismiss it by linking it with the somewhat cruder and culturally circumscribed procedures of associationism (programme music) and mimicry which exist as a somewhat marginal aspect of ... Western art music. This, however, would be foolish... Not only does the control and composition of landscape open up large new areas of artistic exploration and expression, in the sphere of electroacoustic music, [but] it will enter into the listener’s perception of a work regardless of the composer’s indifference to it. [Wishart, 1985]

This last sentence is critically important. What Wishart is suggesting is that the nature of the (electroacoustic) medium fashions our listening response whether the composer likes it or not.

As we noted above sounds heard over loudspeakers are essentially illusory. They all have perceived spatial location and behaviour in relation to a listener; some are particularly clear whereas others, at the opposite extreme have "nowhere in particular" as location and "stasis" or "drift" as behaviour. The listener is aware of his/her position in space relative to a sound or environment. Our response, however, goes beyond simple recognition of apparent location and behaviour of a sounding object and our location in relation to it. Other psychological, sociological and aesthetic responses are triggered. Wishart identifies the principal, or most obvious spatial opposition as

‘in front’ and ‘behind’... Sounds coming from behind....tend to be more stressful, mysterious or frightening... This separation of ‘in front’ and ‘behind’ also has a social dimension for most higher animals." [Wishart, 1985]

Similarly, sounds at extreme left and right positions suggest dialogue or opposition. And the distinction between "present" and "distant" suggest levels of engagement from participation to observation as does the distinction between a sense of envelopment and a perceived single point source.

Speed of movement and regularity and shape of motion, too, are important factors in the characterisation of space. Fast motion expresses energy; slow movement may imply passivity; regular motion suggests something mechanical; angular, irregular motion may convey hesitancy or uncertainty.

To sum up, there are a number of ways of expressing structural ‘distance’ or separation: perceived physical distance, distance of recognition, transformation between different objects/sounds, distance between observation and participation, and psychological or social distance. We also know that characteristics of spatial behaviour play a crucial part in our interpretation of sound morphologies, contribute to our recognition of materials and give significance to unfamiliar materials.

If musical structure, then, relies so heavily on spatial factors, it becomes obvious that careful public presentation is essential to preserve musical meaning in an acousmatic work. For even if the composer considers space and spatialisation very carefully in the compositional process, experience shows that as soon as a piece is played in another room, many aspects of the sound are subject to the acoustic qualities of that room and the many varied positions of the members of the audience. Spatial details become unclear, the effect of dynamic range is reduced by reverberation, background noise and other factors and subtle timbral relationships can be severely affected.

These problems have been tackled by composers in many ways, but one can usefully divide composers into two groups: those who wish to standardise the listening experience or neutralise the listening environment, and those who seek to make use of the characteristics of every new space and extend the creative act into performance.

For those in the first group, the objective would seem to be a much greater degree of control over spatial location and behaviour. To achieve this usually involves multi-track formats in order to translate particular 3-D spatial movement into the concert hall, ambisonic encoding/decoding which can compensate for the many problems encountered with making even simple four channel presentation work with any degree of success, or automated spatialisation. However, these technically complex solutions make a piece much less portable; they may preclude some venues; control of spatial location does not necessarily solve the problems of making real in a performance space the spatiomorphology of sound; and the very specificity might actually emphasise problems of, for example, audience orientation. Nor do some of these solutions allow for the different dynamic shaping needed for different spaces and contexts. The extra control afforded by more complex systems may, paradoxically, create a situation of diminished control.

However, even if we could overcome these difficulties, let us not forget that the translation of a piece of music into a listening space is not only a set of technical problems, but also an opportunity to interpret the virtual space in the actual acoustic by a performer and for a particular public.

For those in the second group, material (in whatever format) should imply shapes and spatialisation which can be optimised in performance for a wide range of venues. Simple stereo encoding is already rich and multi-track formats offer yet more possibilities. Of course, an invitation to a performer to alter carefully composed details has its dangers, but even for pieces from the first group, a degree of sensitive diffusion is often necessary.

Both the grammatical details and the psychological messages of spatial apprehension are unstable because they depend not only on space as composed, but on the relationship between the composed space and the space(s) in which listening takes place....

...Both the personal listening space and the diffused listening space are open to widespread abuse which undermines spatial perception. [Smalley, unpublished]

So, how should we respond to this in performance? In simple, practical terms, what should we do?

Firstly let us define the parameters we wish to translate in to a

performance venue: loudness (i.e. perceived volume); spatial dimensions (width, depth, height); spatial location; motion/spatial behaviour; distance & presence (related to a perceived notion of observation versus engagement); reverberation (of the performance space) and even timbre (where the performance space has problematic acoustic anomalies).

In the absence of a complex technical solution, assuming the most common case (a work in stereo with a clear stereo image to be conveyed and a fixed orientation), a simple solution offers a number of possibilities for a wide repertoire. Multi-speaker diffusion can adapt to both the limitations and creative potential of a range of possible venues: it doesn’t require the availability of particular loudspeaker positions or audience seating arrangement; it doesn’t require a particular number of matching loudspeakers; it offers a number of solutions to each challenge which allow for the development of different performance styles. It also allows anyone with even a modest mixing desk and only a few loudspeakers to begin to explore the "relationship between the composed space and the space(s) in which listening takes place....".

What follows is not intended to be an exhaustive discussion of performance practice. Neither, for reasons which should be clear from the

foregoing, is it intended to be technical. Rather, I would like to make some general suggestions and observations based on experience of concerts in many spaces over a variety of performance loudspeaker arrays, small and large. Description is no replacement for experience so this should be seen only as one possible starting point for a practical exploration of performance. I trust that other articles in this issue of the Journal will expand on these ideas, offer alternatives and cover in more depth areas which here are only mentioned.

Let us start with two loudspeakers in an imaginary concert hall (though this assumption is already contentious): where should they be placed? If they are put at the extreme left and right of a typical stage in front of an audience, there will usually be a significant "hole in the middle" of the stereo image for much of the audience. If we move the speakers in from the sides to compensate for this, some listeners will be situated outside the stereo image. Whilst the listener might be able to perceive the movement in the stereo stage from this vantage point, the level of engagement compared to a listener within the stereo stage will be completely different. And as we have already noted, engagement can be an important structuring parameter, in contrast to most instrumental music where we are usually mere observers of action on a stage.

Staying with our single pair of loudspeakers, distance from the front of stage to the front row of audience may be many times shorter than the distance to the back row, again resulting in a difference in sense of presence. If we can lift the loudspeakers high enough we may be able to minimise the relative distance between sound source and front and back rows, but everything will then appear to come from above head height which would hardly be desirable. And is the front row of audience below or above stage level, and is the auditorium itself flat or sloping? All of these factors will significantly alter subtle though critical details of space and perspective for large parts of an audience, and therefore change (or destroy) musical meaning!

We can begin to solve these problems with multiple pairs of loudspeakers, each replicating the stereo signal, each independently controlled.

Given two pairs of loudspeakers, a useful array would be one wide-of-stage pair, with another pair towards mid-stage. In combination these two pairs could provide a single image for most of the audience across the sound stage with no "hole"; used independently one can emphasise width or focus.

With additional loudspeakers one can emphasise distance with a pair up-stage as far as possible. Perhaps only as a fourth pair would one use loudspeakers behind the audience, since (unless specified by the composer) this is not a dimension used in stereo, and as noted above, has a particular social/psychological significance (not to be used lightly!). Other speaker placements would depend more on the particular space. If possible one would use height, perhaps using overhead-front, -side and -rear speakers, and, possibly something with a less definable overhead location. Additionally one could use other spatially vague placements in contrast to the ones mentioned above, perhaps to give a sense of envelopment or added presence.

Obviously, performance complexity increases the more speakers one uses, and different techniques are needed for different locations and different musics, but it should quickly become apparent that many of the aspects of spatial location and behaviour which we noted above can be articulated with carefully placed and mixed pairs of loudspeakers. Sound becomes an almost palpable plastic object in the concert environment and the sense of space implied so strongly in the medium takes on a real three-dimensional quality appropriate to the venue.

But even with a single pair of loudspeakers the performer’s role is critical. Careful preparation of each piece is necessary to judge appropriate levels for a work, and in a programme of several works, performance of each work needs to take into account of all of the others. Adequate rehearsal time in the venue with the loudspeaker configuration set up is indispensable and in addition to skill with the particular diffusion ‘instrument’, the performer needs to have a clear understanding of each piece s/he is to perform in order to realise a successful interpretation.

....the fragile art of sound diffusion. In a medium which relies on the observation and discrimination of qualitative differences, where spectral criteria are so much the product of sound quality, this final act becomes the most crucial of all. [Smalley, 1986]

This article first appeared in the Computer Music Journal, 19:4 Winter 1995 and was rewritten in 1997 for the Journal of Electroacoustic Music.


Smalley D. (1986)Spectro-morphology and Structuring Processes in: S. Emmerson (Ed), The Language of Electroacoustic Music (London, Macmillan)

Pressing J. (1994) Novelty, Progress and Research Method in Computer Music Composition,Proceedings of the 1994 ICMC (Aarhus, Denmark, DIEM)

Wishart T. (1985)On Sonic Art (York, Imagineering Press)

Smalley D. (unpublished) Spectromorphology: explaining sound shapes

Young J. (1994)The Extended Environment,Proceedings of the 1994 ICMC (Aarhus, Denmark, DIEM)

Alistair MacDonald is a composer and performer of electroacoustic music. Currently the Director of the Electroacoustic Studios at the Royal Scottish Academy of Music & Drama, he is a member of Invisible Arts and a Director of Sonic Arts Network. He has worked extensively with the electroacoustic ensemble BEAST and for Sonic Arts Network, producing and diffusing concerts throughout Europe, and leading workshops in schools colleges and arts centres.

No part of the article may be reproduced or transmitted in any form or by any means, without prior permission of the individual authors.


Home | About SAN | Education | Sounds | Research | Membership | Links | Shop
Copyright 2004 Sonic Arts Network