Not all properties of artifacts are under the control of the people creating them. Some properties are planned by the designer and controlled in the process of production, but there is a rest of “natural” properties. In a previous article, I have called this the “residuum”. In that article, I was referring to the incompleteness of our descriptions of things, but this incompleteness also holds for the production of artifacts.
In most parts of technology, people try to make the residuum as small as possible. Each item produced should come out like the other. The residual, natural properties of things can mostly be found in the microscopic domain. If you take a piece of metal, for example, cut it and put it under a microscope, you may see small grains, called crystallites. The exact shape and orientation of these grains is, to a large extent, a result of random processes, although some statistical properties of these textures can be controlled for. On the macroscopic level, however, one object, e.g. a car, will look like the other. Only through processes of wear and tear, of rusting and cracking, the microscopic properties are gradually amplified and the residuum slowly creeps in from below, bit by bit destroying the designed structure and turning the object back into something natural (see The Philosophy of Cars).
In creating works of art, however, artists may allow for the residuum of the artifact to claim a larger share of the object’s properties right from the beginning. The artist may develop processes in which random events and the resulting random structures are not restricted to the micro-world of features smaller than what can be seen by the unaided eye, but where they appear on macroscopic level. The artist is exploiting the fact that not all properties of what he makes are controlled and controllable.
Prime examples of this are the abstract paintings of Gerhard Richter, as shown in the video clip from the documentary “Gerhard Richter Painting”. Richter has developed techniques that lead to a mix of controlled and uncontrolled properties of his paintings. He is first applying paint to the canvas (or another surface, like aluminum) and this is a process that is controlled to a large extend, although the exact distribution and thickness of paint probably also has elements of randomness. He then uses a large metal squeegee blade to scratch some of the paint off or change its distribution. The result is partially controlled, resulting in a pattern of parallel stripes, and partially random.
You can hear the randomness. The process creates not a harmonious sound, like it would be created when you strike a chord. Such a sound consists of a mix of regular tones in harmonious relations. Here, however, you have a noise, a random mix of frequencies produced by random micro events of scratching or – in the example where the squeegee is glued to the surface of the drying paint – the noises of tearing.
The process has aspects under the artist’s control, like the place where the tool is applied and the directions of its movements, perhaps also the pressure exerted. The choice and initial distribution of colors is another controlled parameter. Parameters like the degree to which the paint has dried when a tool is applied probably influence the outcome as well, so the time when tools are applied is another parameter under the artist’s control. However, in large parts the processes are random.
The resulting patterns show a mix of order and disorder. The process produces stripes, scratches or patterns of holes exposing underlying layers of paint. These structures show some correlation in their positions and shapes as a result of the movement of the tools and the application of paint, but the distribution and shape of the structures is also largely random, especially on smaller scales (see also On the Aesthetics of Scale). The repetition of such processes with several layers of paint results in complex patterns and textures.
A side effect of this is that the creation of these objects cannot be repeated, each one is unique. Even if you would repeat the process as exactly as possible, even if you would do it by machine, the result would probably never be the same. In terms of physics, this might be due to non-linear processes, i.e. processes where small differences in initial conditions can result in large differences in the outcome, thus amplifying microscopic differences to macroscopic levels.
The resulting structures are partially ordered, enabling our perception to find regularities, and partially disordered, providing anomalies to the order somebody looking at the object has identified so far. More correlations and patterns can be identified and the anomalies can be integrated into the perceived patterns and structures. However, the high complexity of the structures ensures that the perceptive system will not come to rest quickly. The picture provides enough complexity to avoid boredom and enough order to discover partial order again and again.
As I have described in a previous article (see On Beauty), I suppose that the experience of beauty is an experience of success of finding order in the perceptual data. It therefore requires correlations between parts of the perceived object. This is provided by the order-creating aspects of the artistic processes. On the other hand, too much order would lead to boredom since no new experiences of order-discovery are possible once all the order present has been identified. Too much disorder, on the other hand, would frustrate and confuse our perception. The “beauty zone” is between these two poles. Processes like the ones developed by Richter and some other artists provide a way to create artifacts just in this area of beauty, fascinating and mesmerizing…