Outside the electroacoustic music community I often encounter the opinion that electroacoustic music should be experimental and surprise the listener with novelty or provocation, the art-form criticised if it builds on existing tradition or repertoire: a curious experience compared to that of instrumental composition where tradition, techniques and languages are familiar to most protagonists, and where surprise or novelty is but one of many aspects determining the 'quality' of the musical work and its performance. Within the electroacoustic music community, composers are free to explore the compositional process for the sake of the music itself.
An analysis of this situation is the subject for elsewhere. Here I present the Barely project, which was motivated by a bi-level aesthetic reflecting my schizophrenic experience of being freelance electroacoustic composer operating outside the education and research network yet with a background, interest and continual contact with this network. At one level Barely drew on specific aspects of the acousmatic language connected to principles of perception, sound identity, environmental listening and spatial interactive contexts. On other level Barely attempted to explore some ideas less evident in the acousmatic tradition or in current trends in sound-art. These ideas included investigating a compositional language based on listening thresholds, how external social frameworks may influence the work's intrinsic structure, and specifically to Barely_part-1 how architectural collaboration influenced the music.
Barely is a dynamic project that has changed definition since its first conception in 2006. At the current time Barely is a paradigm of composition, sound-art and multi-media addressing the relation of the composed work to perceptual and social frameworks, and how this external connection may subsequently find its way into the intrinsic structure of the work as a self-contained entity. It specifically reflects on the increasing complexity of our everyday sound-world, the redundancy of noise in art and society and hopes to encourage detailed listening. In concrete terms Barely presents a highly detailed, just perceptual layer above the "experienced threshold" of our senses. In an anechoic room, the experienced threshold is that of the threshold of hearing. At a busy train station the experienced threshold is the sound to which we consciously listen to as information-gatherers, rather than as passive receivers hearing the sound but with no intention to listen or interpret meaning. The background sound is regarded as information, however redundant it may be. Barely-sound is an added layer of information and structure. By developing a compositional 'sound etching' technique, the ear and the memory, instead of being presented with a silent chasm is encouraged to explore the continuation into this background. Barely is manifest in a number of forms spanning acousmatic composition, acoustic instrumental performance, spatial sound-architecture and sound installation. Three parts have to date been realised: Barely_part1: installation; Barely_part3: concert acousmatic work ; Barely_part4: concert work for cello and live amplification.
Barely_part-1 addresses compositional structure and the experienced threshold in terms of space, sound metaphor, acoustics, perception and the social framework within in a public-space installation where architectural aspects interact with the temporal domain. It aims to create a "sound-sanctuary", enticing the individual into deep concentration and a sensual experience by offering great detail and structural complexity at a level that is only just perceptual in terms of volume, identity and structure, yet drawing on sound archetypes latent in our everyday experience - a detailed information network hovering above and connecting into the existing background.
The work involves a large 'barely visible' spatial installation created in collaboration with experimental designer and architect Birger Sevaldson. From a conceptual angle, there are many ways to problematise Barely in terms of visual space. The design solution involved hung lengths of clear, colourless plastic foil, UV light and UV reflective paint, sightlines and spatial conjunction. The design acts as a spatial modulator, enhancing what is already present in the immediate (local) and greater (diffuse areas) of the installation space. In terms of sound, a 16-channel composition is spatialised over 12 high quality near-field loudspeakers and 40 transparent miniature loudspeaker elements. The composed material is 70-minutes in duration and looped continuously. The opening of the installation involves a three-hour performance where spatialised layers of information are mixed in real-time in response to the audiences' interaction (staying time, physical motion, audience created sound, changes in room acoustics due to public presence). Figures 1a, 1b, 1c, 1d, 1e.
2. Visual-spatial architectural layer
20 clear, colourless 1-mm thick and 1-meter wide polycarbonate foils were hung from a ceiling grid as shown in figure 2. On each foil were painted lines with UV reflective paint. This patterning (figure 1b) was designed in the following way: Three points of interest from the real space were used as anchor points in a 3-D model. Lines of perspective were drawn into the 3_D model stretching from these anchor points towards different viewing locations defining sightlines within the real space. The sightlines were numerous, representing real viewing points and also points elevated or located in impossible viewing positions (over 800 sightlines in total). Within the model, foils were aligned along sightlines so that when viewed from any of the three anchor points they would be seen from the edge and effectively disappear. In the model the site lines were projected onto the foils along both the x- and y-axis (the 3-D perspective was in other words flattened to 2-D surfaces). Life size templates then allowed accurate painting of the real lines. The anchor points were chosen such that in the real space a visitor would be unable to stand at the exact point. This, combined with the great number of slight-lines created a complex patterning that clearly related to the installation space, yet defied accurate quantification.
3. Combinations of miniature and main loudspeakers
Six high-quality near-field loudspeakers were located around the sides of the space 1 meter above ground. Six more were arranged on the floor in two triangles, angled upwards 45 degrees. 40 small loudspeaker elements (with a frequency response from 300-5000 KHz) were glued to the installation's plastic layers at a variable height between one and two meters above ground level. The foils effected the sound-field in two ways: they created clear first-reflections from closer main loudspeakers and reduced the propagation of sound outside the frontal zone of the miniature loudspeaker elements. Both effects were desirable in that they enhanced image clarity in an otherwise highly reverberant space.
Barely_part-1 needed to address a public installation space where background sound was continuous and which most people considered as noise. Noise can be discussed in sounding terms of acoustics, signal processing, perception, 'noisy' sound metaphors and more general statements of artistic or expressive intent. Noise can also be discussed in structural or organisational terms of sociological noise, information overload and redundancy. The Barely approach attempts to elevate itself from the status of noise in both sounding an organisational terms. In general this approach meant each detail would be considered and labored over regardless of whether it was immediately appealing, to explore diversity, commonality and encourage detailed listening, particularly in terms of an oscillation between abstract and referential reflections on everyday sounding experience.
4a. The sounds, their reference and the public space
Sound projected over loudspeakers may lead to acousmatic abstraction, particularly when heard by an unassuming listener in a public space. However, by working along a continuum from a direct causal basis to abstract reference (in other words showing a contrast of extremes) listeners found that the sound and its behavior regained contact to everyday experience. Direct causalities explored indicators of human activity. Such material included everyday sounds recorded over a period of time inside the installation space (Kanonhallen - an 80 x 14 x 12 meter glass and concrete industrial space in Oslo): the activity of a janitor, two onsite workers, sounds intruding from a neighboring construction site and from a kindergarten, motorbike couriers, a distant propeller plane and the acoustic of this unusual space activated by these sounds. Other direct causalities explored sounds recorded in the studio: intimately recorded non-linguistic vocal expressions designed to create a sense of human presence but without language or narrative. Abstract reference was explored mainly via transformation techniques. For example vocalisation fragments created articulation, texture and gesture; a vocal consonant, when turned into a texture becomes a vague noise wash, yet when presented as a single articulation connects directly to the primal expression of the original sound. Other transformations created new, semi-abstract metaphors that found meaning when connected to the greater installation space: watery textures, tearing and rasping energy, machines and miniturisations of industrial construction-sites.
4b. Recontextualising techniques from the acousmatic language
In the majority of sound-art and also in the acousmatic reperoire, attacks are normally high volume, clear and precise articulations, enhancing 'attackness' in both intrinsic (self-referential terms) and extrinsic terms (referring outside itself). Attacks are used to inject excitement, energy and musically appropriate articulation into the music. Such an approach also dampens our hearing sensitivity to information in the attack's aftermath. Barely-part_1 does not ignore the extrinsic nature of an attack but presents it with a 'naturally' softened intrinsic content: attacks and articulations in the environment are captured with the microphones some distance away, where the volume difference between gesture and field, and between attack and aftermath is reduced. Harsher frequencies arrive at the microphone with a lower energy. Sounds recorded within a spatial acoustic are transposed downwards to expand their sense of size and cavity, rather than sense of attack and articulation. Thus extrinsic attack references remain but without the destructive effect on our auditory perception.
Noise and temporal perception
Many sounds regarded as noise contain interesting intrinsic detail beyond the range of our normal temporal perception. This detail is drawn into a perceptual range by processing techniques appropriate to the sound's identity. For example, a machine sound will be looped enhancing its mechanicalness; a noisy squeaky wheel will be treated with spectral modulation that enhances the sense of friction; a downwards transposition enlarges temporal qualities of the sound that may be regarded as noise simply because detail is too fleeting to grasp at normal speed.
In softening the noise from the environmental-industrial soundscape, micro-scale variations in temporal and spectral information are also lost. This detail is re-explored in delicate close-microphone studio recordings. Sometimes these materials verge on the aggressive in the sense of a microscopic insect world but are never intended to cross the threshold into noise or alienation on a human scale. The recording technique eliminates spatial information such that these materials function particularly well over the miniature loudspeakers - which become tiny sound objects defining their own space.
Volume and engagement
We have all experienced how low volume sound of sufficient detail encourages calmness and engagement: silence yields concentration; the sound of an unamplified acoustic instrument in a large space sends a hush across an audience attentive to every detail; talking to a child in a quiet and calm way will often communicate more information than when shouting. Likewise, within a silent scene, tiny sounds normally ignored can evoke irritation and annoyance - a ticking clock, a noisy breath, the clicking of a computer keyboard - whether for good or bad we actively engage with the sound-world when low volume sensitises our hearing. Barely-part_1 develops this phenomenon in approaches to both the listener and to the compositional process. Tiny details that the ear would normally ignore or be unable to detect are under focus. The tail of the sound - the last trace of gesture normally lost in the background - is saved. Temporal and spectral processing techniques bring to the fore residual detail in the sounds' aftermath. Information normally lost in a background sound-field is elevated to the experienced threshold. Within the space, voiced and other sounds with clear identity could be of a lower volume and remain clear - the ear latching onto that which is identifiable yet out of place. Identifiable sounds of similar type to those in the natural background were naturally less clear. Sounds that were non-continuous (micro-sound durations of 1 second - 1 ms spaced with silence) were also heard clearly at lower volumes due to the ear detecting the discontinuity when set against a continuous background sound-field. Likewise, unnatural changes in morphology drew the ear to a sound without increasing its volume.
On entering a sound installation in an urban context an attentive listening tactic is rarely the first approach the public will take. In most cases their hearing and listening has been dampened by the noise of the environment from which they have come. How to encourage physical calmness, curiosity and concentration are therefore prime objectives. A visitor's own footsteps, possibly talking, coughing or breathing may be the most prominent sounds heard. Local 'pockets' of sound combined with the installation's architectural definition are designed to evoke a sense of arrival and individuality, encouraging a listener to stay and be enveloped by the materials of the work. A change in the sound elsewhere, or a period of redundancy in the current location is designed to encourage the visitor to move and explore a new orientation (rather than leave!).
5. Compositional structure
5a. The interaction of space with volume and frequency
At a constant volume, that which a listener hears is dependent on acoustical and psychological factors. Acoustical factors include room acoustics, frequency content and timbral variation in the sound and the distance the listener is from the loudspeakers . Structure therefore exists across a range of spatial distances. The distribution of installation elements is designed with these factors in mind. In terms of space, the volume was set such that for up to four meters from a loudspeaker sound was clearly defined. With distances greater than four meters the listener is required to physically move to investigate the sound-field. Therefore in any one location a listener would hear a main layer of sound and increasingly remote background layers. In terms of frequency and texture, sounds that play simultaneously in close proximity are designed to occupy different frequency bands or if overlapping in frequency are of different temporal variation. In addition to the intrinsic volume calibration, the 12 full-range loudspeakers projected sound of different spatial metaphor: sounds that are by character 'distant', are placed on the remote six loudspeakers, while sound spatially closer in identity are placed on the central triangular arrangements (see figure 2). The triangles create two zones where sound 'mobiles' ranging from 10 seconds to a few minutes unfold. In occupying these locations a listener would in addition hear a distant layer of sound from the remote 6 loudspeakers (these zones were particularly popular with always someone standing or sitting in the centre). Four channels of audio were distributed over the 40 miniature loudspeakers. The short range of these loudspeakers, combined with the reflective and barrier characteristcs of the plastic foils meant that most listeners were unaware that only four different source channels were being used. As mentioned earlier, the characteristics of the small loudspeaker elements determine which sounds were played on these channels. Their frequency response and directionality made spatial illusion in the source appear false and inappropriate. (A spatial illusion is where a sound-object appears to contain a recorded or simulated spatial definition, connected in some way to room acoustics, object sizes, object relationships or object interaction). In contrast, 'dry' sound without any sense of space was more successful. The small loudspeaker elements then become sounding objects in their own right. The combination of listener-loudspeaker proximity, virtual image reflections from the plastic foils, main and miniature loudspeakers created a complex yet 'transparent' three dimensional sound-filed for the visitor to explore, where at no single point in the space could you hear all that was going on. If intrinsic noise can be regarded is a chaotic saturation of time, frequency and space, Barely-part_1 structures time, frequency and space to prevent layers of information collapsing into a state of noise. The physical installation further encouraged the need to dwell and explore. The installation was open for the general public to enter and leave as they wished. Many locals who would never had entered an electroacoustic concert or sound gallery stayed for lengthy periods were footsteps were hushed and even whispering was saved for afterwards (it was in fact the acoustic orchestral performers, who were for some days sharing the same space, that talked and created social sounds outside their rehearsal time, not realising that the installation could have been of interest).
5b. Temporal organisation and structure
In practice the compositional structuring of Barely-part_1 approached time and material as an electroacoustic development of Stockhausen's 'moment form'. (A 'moment' is a formal unit recognizable by its personal and unmistakable character, where certain characteristics remain constant for a period of time and this period is not necessarily predetermined. 'Moment form' is where the degree of change in the form is controlled in the compositional process and in the passing of time, rather than by the material of the 'moments' themselves). This approach allows the materials from Barely-part-1 to be re-assembled to function in other spaces where a pre-production phase is used to create a site-specific version. To compose the 'moments' a sound classification process following a traditional approach was used. Sounds were grouped into categories via the perceptual qualitative evaluation of their most prominent characteristic. This evaluation was carried out some months after the individual sounds were composed in an attempt to avoid evaluations based on memories of the original sources rather than on that present in the sound. All evaluation and composition was carried out at low volume over a variety of loudspeakers, including those that would be used in the installation. Only at the final mastering stage was the volume increased to normal monitoring level for production quality control. Figure 3 shows an overview of this process. The categories in stages1- and Ð2 were derived from the material. The material in these categories was then developed to satisfy criteria in stage-3. Likewise to the previous categories a perceptual evaluation was used. (However, it would also have been interesting to investigate automatic assessment via computer analysis of parameters such as variations in pitch, amplitude and centroid). Stage-3 criteria were decided independently and chosen to explore 'moments' that would maintain clarity when layered in a mix.
6. Installing the work and the performance
Just as in an acousmatic diffusion performance, Barely_part-1 needed calibration for variations in room acoustics with an audience present. Unlike an acousmatic concert the audience were in motion. The sound needed to emerge above the existing sound environment for where ever the listener was positioned - a difficult but not impossible task if building on an acousmatic diffusion performance experience. Fore-mostly the volume at different frequencies required calibration such that it was not too loud or too quiet for any length of time in any one location.
7. Gentle Sediment (Barely_part-3)
Gentle Sediment is the Barely acousmatic concert work. It is designed such that Barely ideas can be performed to function at the experienced threshold under different concert situations. The work is performed from five stereo layers. Each layer contains material that would be heard differently under diverse concert conditions: one layer containing bass and lower sustained surges, one carrying sustained treble sounds, one with sounds suggesting large environments, one with vocal materials and one with abstract short fragments. The layers may each be diffused over many loudspeakers. The purpose of their separation in the playback source is to achieve balance and clarification, rather than for specific spatialisation. For some listeners the work offers a minimalistic and meditative experience. For others its low volume and shimmer of details may evoke the opposite experience of active listening and heightened awareness.
Gentle Sediment drew on the sound material from Barely_part-1. Apart from the work needing to be of a fixed duration (somewhere between 9 and 14 minutes), no specific structure was defined at the outset. The task was to find a compositional method that would result in more precise indicative relationships than those created in the moment-form approach, and for the work to function for a stationary listener in a concert situation. Barely_part-1 offered a vast amount of material yeilding an already 'composed' network of information. With faith that some of these relationships would re-converge regardless of how the materials were organised, I began by layering vast amounts of material. The selection of material was based on an approximation of that which a listener would hear if spending 15 minutes exploring the installation (most visitors entered the installation from the same point and after a few meters explored a greater diversity of paths, staying for duration ranging between 5 minutes and 40 minutes). In this way the initial stages of Gentle Sediment were derived from the response of the public to Barely_part-1.
At one point 42 stereo sound files were simultaneously playing in the first stage of the mix. In such a mix, hearing individual sounds is impossible. The problem was compounded by the fact that the installation consisted of a 16-channel source while Gentle Sediment was to be a stereo work (despite the five layers of the final format). Here began a sound etching technique. First, sounds that had no chance of surviving in a dense mix in terms of volume, frequency, texture or identity were removed. After this initial 'thinning' I attempted to balance the material in each track such that some aspect of all materials could be heard. This too was a tricky process as the total volume often spiralled upwards as each sound was tested to see if a slight change in volume would have a marked difference on its function. After some time certain materials appeared consistently clear above the rest and began to articulate shape and form in what was initially a constant block of sound, and in turn pointing to further etching in the mix. In hindsight we can look at the structure as a combination of a compression of space-time criteria perceived by a visitor to the installation and 'interactive immergence' derived from perceptual-compositional assessment in the studio.