Electronic Feedback Systems

Words by Barry G. Nichols (ECM:323)

Video stills from phase installation

“Causality arises from a key concept of systems theory — the idea of feedback. A heater thermostat is an ideal example of a negative feedback system — a system which regulates itself and maintains a stable condition. Once feedback was defined and explored by information theorists and systems analysts in the 1940s and early ’50s, scientists began to look for examples and analogues of it. With rising excitement they began to find similar stability-protection systems in every field from physiology to politics. Negative feedback appeared to be at work all around us, causing things to maintain their equilibrium or stability.

“By the early ’60s, critics like Professor Magoro Maruyama began to note that too much attention was being paid to stability and not enough to change. What was needed, he argued, was more research on ‘positive feedback’ — processes that do not suppress change, but amplify it, do not maintain stability, but challenge it, sometimes overwhelming it. Positive feedback can take a small deviation or ‘kick’ in a system and magnify it into a giant structure- threatening shudder. Positive feedback could illuminate causation in many previously puzzling processes” (Alvin Toffler, The Third Wave).

This article largely documents ECM:323’s experiments with electronic feedback over four years with a variety of equipment and represents our personal viewpoints and practices as opposed to a technically precise analysis of feedback processes.

The impetus behind our art installation work was born from the desire to examine critical relationships between sound and visual form, and how one might influence the other. This area of research emerged from the premise that physical, environmental, ambient and even cosmic sound might provide access to the interpretation of objective, tangible processes which occur throughout the natural world.

Initial experimentation with a series of classic ’Chladni’ vibration pattern experiments proved highly successful. Around 1787, Chladni was developing a previously little visited area of physics which eventually became the science of ‘acoustics’. Using a metal plate scattered with fine particulate matter, he experimented with a violin bow drawn across the edge of the plate, creating subtle resonances. These resonances formed beautifully descriptive wave patterns in the material which settled upon the key ‘nodal points’ of attraction and repulsion. Chladni’s experimentation yielded some 52 documented nodal patterns, now famously known as ’Chladni figures’. Our own experiments evolved empirically over a period of time, and we presented the results in finished form at The Museum Of Installation in Deptford under the heading, ‘Test Sites’. The Test Sites series was to become the framework around which we developed and refined our working methods. By projecting precisely calibrated sine waves upwards through a membrane containing water, we were able to generate incredibly hypnotic fluid geometries — radiating hexagonal patterns known as ‘Macrons’. (Test Sites: Trace TS02) With relatively modest apparatus, and a rudimentary knowledge of physics it became possible for us to manifest sound as visual information.

Relatively recent breakthroughs in fractal geometry and the chaos sciences were revealing previously undiscovered truths about the nature of matter and pattern formation. It became obvious that there was a distinct cognitive connection between these sciences, and what we were trying to achieve artistically. Our discovery of a definitive paper by James Crutchfield (Centre for Non-Linear Studies, Los Alamos, California) rather grandly entitled ‘Space-Time Dynamics in Video Feedback’ provided the necessary link that we had been looking for. In his paper, Crutchfield uniquely and concisely links electronically generated video feedback patterns with an array of chaotic and fractal phenomena, and uses this method to analyse and illustrate chaotic behaviour. It became essential for us to explore the possibilities inspired by Crutchfield’s paper using electronically generated audio and visual feedback signals. The option of using electronically generated feedback was a deliberate inversion of our previous works which dealt with more organic phenomena, for one simple and incontrovertible reason; electricity has largely been identified with the technology that is driven by it, and as human beings, we tend to overlook (or perhaps de-emphasise) the fact that we too are created and driven via electrical impulses. We are perhaps, largely unaware of electricity’s pervasive influence throughout the natural world, and that we, and indeed all life depends upon electricity and electrical impulses. With this in mind, the perplexing philosophical question — ‘Where does life come from?’ is suddenly brought into sharp relief — it emerges from itself. Evolutionary biologist Brian Goodwin explains: ‘The organism is the cause and effect of itself, its own intrinsic order and organisation. Natural selection isn’t the cause of organisms. Genes don’t cause organisms. Organisms are self-causing agencies.’

Life is thus an auto-conspired form. It emerges to transcend itself, a snake, swallowing its own tail, like the mythical symbol Ourobouros, which to many has come to represent the concept of feedback. Ironically, feedback systems are relatively simple to create — a portion of the output signal is fed back into the input of the same, or preceding stage of the loop. Students of cybernetics (the science of control systems) will recognise this as circular causality. As simple as a feedback loop is, it can be stitched together in endless combinations, and forever layered until it forms an unimaginable edifice of complexity and intricacy of subgoals, some of which cross their own paths — A triggers В, В triggers C, and C triggers A. In outright paradox A is both cause and effect. Hackers know this phenomenon as a recursive circuit — nevertheless, whatever the riddle is called, it totally undermines classical theories of everything. If something can be its own cause and effect, then rationality is up for grabs.

My introduction to working with feedback was via electronic composer Peter Hodgkinson (of T:un[k] Systems and Mechos), with whom I was collaborating. He introduced me to a unique feature of the Behringer Eurorack 802 mixing desk, which generates feedback when two effects are thrown together. The Behringer manual dryly states: ‘when using Aux Send I as a second pre fader effects send, DO not engage S16. The connection from Aux Return to Aux Send I would be illogical, and could cause feedback.’

Of course, we became fascinated with the possibilities yielded by this idiosyncratic feature, and Mr Behringer’s warning was too much of a temptation to resist! Peter had composed tracks utilising the feature and demoed them to me at his then London home studio. I quickly recognised that here was a way of using audible electronic feedback to emulate analogous natural systems, (as, of course, they are both governed by the same rules) and began experimenting with various effects through my own Behringer desk. My first successful recordings demonstrated moderate control of the equipment in order to generate some astonishing tonal feedback works, which were performed at the South London Gallery through a struggling PA system (ECM:323 — Test Site TS:02, August 1998 & CD release, Sound Factory — SFSLG 01).

As artists, we came to regard electronic feedback as an incredibly fascinating medium, and a rich source for our creativity, not least because it could be said to be a precursor to the recent development of software based ‘Generative Music’. Practising electronic composers will no doubt realise that feedback oscillators are a fundamental feature of all synthesised electronic sounds — without feedback, synthesisers would be almost useless. The creative potential resulting from the outgrowth of unprocessed feedback is almost limitless. From our own personal experiences it would appear that serendipitous feedback events arise from moments when the human has little or no control over the equipment from which it originates. Given that electronic feedback systems, whether audible or visual, draw from a vast array of variables, the potential to create new, coherent forms is, for us, tremendously exciting.

The compounded logic of the previously mentioned stacked loops which double back on themselves, becomes the source of the strange counterintuitive behaviours of complex circuits. Engineered under precisely controlled conditions, these electronic circuits normally perform dependably and reasonably, but then suddenly, and by their own drumbeat they veer off without notice. In industry, electrical engineers are paid to outfox the lateral causality inherent in all circuits. When pumped up to the density required for a robot, circuit strangeness becomes indelible. Reduced back to its simplest — a feedback cycle — circular causality becomes a fertile paradox.

When we encounter these patterns for the first time, (especially in video feedback) it is not unreasonable to treat them as our own offspring — we are intensely aware of the fact that these self-same variables are imposed upon the creation of biological forms, and it requires no great leap of the imagination to assume that in all forms of electronic feedback, we may be privy to the creation of the first generation of genuinely post-biological life. This may be straying a little from the strictest interpretation of the application of electronic feedback systems. However, it may be enlightening to project forward into the future potential of the medium. Although speculative literature concerning the development of Artificial Intelligence is replete with dystopian sci-fi scenarios where human life is superseded by machinic superiors, it might be more realistic to imagine a more harmonious evolution occurring.

We are now well into an era where musical software is capable of randomly feeding back and mutating. In fact, our increasingly complex computer systems are more than capable of creating their own surprises. In an imagined scenario not so many years from now, we can expect some startlingly original glitches to arise spontaneously from intelligent machinery. It will almost certainly be that musicians will be able to create algorithms which generate and harness audio forms, and enable them to evolve and propagate of their own volition, using the rules of positive feedback, combined with random mutation.

In his book Mind Children (Harvard University Press, 1988), MIT luminary Hans Moravec envisions a time when self-replicating digital life-forms in the form of viruses will be intelligent enough to plan and act in a deliberate, calculating, and creative manner in order to engineer their own survival. Not only will the data realm be populated by its equivalent of rats, sharks, and criminals, but also by flowers, trees and songbirds. Naturally, the musicians will embody their creative software with these viruses in order to ‘grow’ their own music. In fact, it is highly likely that musicians will have the ability to ‘farm’ and harvest their own audio viruses, giving rise to an expansive network of interactive musical distribution.

Our most recent audio visual installation ‘PHASE’ (Test Site TS:05) demonstrates our exploration into this possible future by combining a generative video feedback form with an audio feedback soundtrack. ‘PHASE’ encapsulates “a system in which the normally serial circuitry of audio visual apparatus is transformed into a closed, circular loop by the disarmingly simple expedient of re-orienting the camera to point directly at the monitor screen” (from an installation catalogue, Royal Society of British Sculptors 2001). The net result is beautiful, yet alien, and has taken audiences by surprise, especially when confronted with the fact that neither of the two elements were created via software, and that in fact, we literally ‘captured’ what we discovered, using minimal human intervention except when editing out the less aesthetically pleasing content. Our own description of the installation contends: “Feedback loops result in the generation of complex spatial and temporal patterns and structures analogous to those found throughout the natural world” — offering a possible model for relationships between circular, iterative systems and far more complex forms of natural behaviour.

This stage in our creativity hints at future excursions into the areas of Artificial Intelligence and information systems, currently being researched under the working title ‘Hemispherics’.

“Because positive feedback breaks stability and feeds on itself it helps to explain vicious cycles — and virtuous ones. When we put negative and positive feedback together and see how richly these two systems interplay in complex systems like the human brain or the economy of a country, startling insights emerge. Indeed, once we as a culture recognise that any truly complex system — whether a biological system, a city or an international political order — is likely to have within it both change amplifiers and change reducers, positive as well as negative feedback loops interacting with one another, we begin to glimpse a whole new level of complexity in the world with which we are dealing. Our understanding of causation is advanced.

“Yet another leap in understanding occurs when we further recognise that these change reducers and amplifiers are not necessarily built into complex systems from the start; they may be absent at first, then ‘grow’ into place, sometimes as a result of what amounts to chance. A stray event can thus trigger a fantastic chain of unexpected consequences. This tells us why change is so often hard to track and extrapolate. It is why a slow, steady process can suddenly convert into an explosive change, or vice versa. This in turn explains why similar starting conditions in any system can yield sharply dissimilar outcomes” (M. Mitchell Waldrop, ComplexityThe Emerging Science At The Edge Of Order And Chaos).