These relationships are kept in a delicate balance between clarity and obscurity. Are visible behaviours triggered and controlled by sound, or do their behaviours create sound? Are we listening to their actions, or are we watching them respond to the sounds of their environment? This projected universe operates on its own laws; familiar, but somehow elusive. These laws appear to be based in sound, or to emit sound: objects turn; the projected world turns… while sound events strike, scurry past, or resonate…

The projected visual space is a field of strange architecture, environment, and multiple horizons, peopled by strange, somehow familiar organic entities. We are presented with a 'field', which seems to be a somewhat contradictory landscape – both architectural, evoking buildings and skyscrapers, composed of grids, lines, and planes, and simultaneously reminiscent of natural landscapes of forests, woods, and mountains. This is a somewhat curious and compelling duality – a seeming contradiction in our world, between the natural and the artificial, yet somehow consistent and coherent in Nieminen's projected universe. This landscape is then peopled with a range of objects and figures – vaguely organic, strange but somehow familiar forms, which evoke a range of natural objects and shapes from our world, without triggering direct equivalencies. These subject-figures are associated with a range of behaviours, again inspiring real-world associations without making direct references: objects rotate or spin in place, or spiral upwards, for example.

Further, this is a dual universe: two screens present two seemingly distinct worlds – or, are they simply two unique views of the same world? In fact, both worlds are present in each screen, but with one dominating, while the other can be seen as a dark mirror image, hovering below as a kind of ghostly reflection. This pairing is then reversed in the second projection.

The soundscape of the piece is built from a collection of single, isolated sound elements, organised into four main categories: reverberated pitched impulses; reverberated unpitched impulses; resonating feedback; and short, gestural sounds (culled from a range of sources). There is also a fifth category, consisting of short composed phrases and gestures, constructed from multiple materials; these appear more sporadically than the other categories, and are the only sound category not linked with the visuals – they are a solely sonic element of the work. Only a single sound from each of these five categories is heard at a time, meaning that a maximum of five simultaneous sounds is heard at any given moment. A key component of compositional control is thus exerted through control of the overall density, by using varying delay times in the playback from each category within the MaxMSP patch.

In practice, control of the system is, in fact, unidirectional: motion and movement of visuals are controlled by data taken from analysis of the sound outputs. However, this is far from clear when experiencing the work; in fact, our oft-noted 'visual bias' serves to obscure or even contradict this control relationship, with many visitors assuming the reverse – that the visuals are triggering or controlling the sound. This is also an intriguing question, beyond the issue of control: do we perceive the sound as causing their behaviour, or are their behaviours creating sound? Are we listening to their actions, or are we watching them respond to the sounds of their environment? Or do we accept the dual nature of a single event?

Much of the work of composing the overall experience of the installation went into tuning this central relationship. Primary concerns and controls included:
– Which sounds should be associated with which objects?
– What kind of movement should be generated by a given sound?
– Which parameters of a given sound should be used to drive movement?
– What should the range of the control parameter be, and what should the range of the resultant motion be?
– What should the overall sound density be? What should the overall density of visual change and motion be? And how can these two be enacted together?
– Which agents and motions in the visuals should be independent of any sounding output?

The balance of these elements is subtle, and critical. It has a primary impact on the reception and perception of the work, and, mishandled, can cause a complete rupture between the visual and sonic elements.

The attempt to discover the secrets of this relationship forms an essential part of the experience for most visitors. However, the rules of interaction between the visuals and the soundworld, as coded into the system, are only part of this experience. It is not simply a matter of uncovering the riddle of these coded interactions, but rather these preconceived relationships act as a kernel, around which a whole, living system grows up and bursts forth. Unforeseen, imagined, or surprising relationships and interactions are allowed to build up and evolve in the visitor's mind and perception. Critical here is the use of materials conducive to such evolution and interaction, and the balancing of controlled relationships with space for fortuitous and evocative simultaneities. If relationships are too direct or too obvious, then the visitor quickly resolves their quest, and potentially grows bored or moves on; more importantly, this results in a weak and potentially unconvincing system. Of course, at the other end of the spectrum, if the relationships are sufficiently obscured, connections are lost, the work devolves to a seemingly random series of sonic and visual events, and again the sense of a system is lost.

Thus there are two key tools which are used to manage these qualities of the overall system. One is to limit a given relationship's activeness: each control parameter only triggers its associated action or movement a certain percentage of the time – generally a significant percentage, but with enough discontinuity to keep the visitor curious and engaged. The other is to keep some visual agents out of the system and independent, triggered and engaged solely according to their own processes, without being affected by sound.

It is extremely relevant to note that the visitor will nevertheless tend to assign relationships to these independent visual elements. Particularly remarkable is that this also occurs for those who coded the relationships – Merja and myself – and who are therefore fully aware that no such 'official' relationship exists. As Simon Emmerson has pointed out, "what sounds causal is effectively causal"; as a result, the composition of the work through these relationships is as much involved with implied relationships and causalities, as with actual or literal relationships and causalities. Leaving space for such implied causalities to occur is thus a critical step in this process. Furthermore, these implied causalities are of a slightly different quality than their more literal, pre-programmed counterparts; they are more flexible, more adaptive to the circumstances and qualities of a particular moment. They thus have the capacity to seem more subtle, more magical – and, somehow, more 'real'.

Some other aspects of the sonic/visual relationship are worth noting. One is the (fully predictable) dramatic difference between either of these presented singly, versus their combined impact. In particular, the visuals, seen without the sound, have a somewhat unsettling effect – strangely ghostly, somehow immaterial. It seems ironic, somehow, that they require sound – a particularly disembodied sense – in order to become somehow tangible and present. Also critical to this experience is the way each of these aspects of the work creates space. Without sound, the projected space begins only at the screen, and extends into the distance, away from the visitor, from there; the visitor is therefore outside the environment, looking in. The sound, on the other hand, is projected within the room; the visitor is therefore situated within this environment. Taken in combination, the sound elements tend to fuse with associated objects or movements in the visual world; as a result, in situating the visitor within the projected sound environment, the sonic layer of the work effectively situates the visitor within the total environment of the work.

The technical requirements for this installation are somewhat flexible, dependent on the potential size and dimensions of the installation space; the technical requirements can thus be either shrunk or expanded, in accordance with the space. In its first presentation, we used: - two 4000 ansilumen video projectors with VGA-inputs, with ceiling mounts and a sufficient amount of VGA-cable. These can be reduced to one projector or expanded to several, depending on the exhibition space. - an equivalent number of screens, white walls, or other projection surfaces. - ten loudspeakers and a subwoofer, plus an audio mixer with an equivalent number of channels, and the associated cabling. This is particularly dependent on the exhibition space; four loudspeakers could potentially suffice for a single, smaller room, with more required for larger or more varied spaces. - Ideally, we would need a Mac Pro, with two graphics cards (ATI radion HD5770). (If necessary, we could bring the graphics cards and install them ourselves.) - An audio interface, ideally with 16 channels of output (presumably 8 channels analog, plus 8 channels optical output to an additional digital-to-analog converter). If necessary, we could potentially bring these oursevles. As mentioned, these requirements are potentially negotiable and size-dependent. Please feel free to contact us with questions or for further information. This installation requires a suitable space for visual and sound projection. Ideally this space would be dark, and with a limited amount of outside noise, although of course we can manage in less-than-ideal circumstances if necessary. Size is flexible. In the first presentation of this installation, the exhibition space had two main chambers/areas, and both were used; the installation can thus be expanded to two chambers, or multiple chambers.