This page has been archived by the Computer Arts Society.

The position of the ‘original’ in Computer Art

<<Previous Up Next>>

Computer graphics have an “ideal” form that is only approximated by our display techniques, which render the mathematical code in a graphical form that we can comprehend. In its raw form, Computer Art consists of lines of code that provide the instructions for the image to be drawn on screen or printed. In this, it comes very close to what the art theorist Nelson Goodman termed an “allographic” artform: one that is can be easily rendered into notation and has to be performed to produce its artistic effect. In its transcribed form it only possesses the potential for art, which is released during performance.

It has been claimed that images on the computer can be copied exactly. This is, however, an oversimplification. Rather, the creation and display of images is a complex process in which the computer responds to instructions from the human artist. The computer then renders the digital result in a visible form which the artist observes and absorbs into further artistic activity, deciding whether to modify these results or leave them as they are. In the process, sections of physical reality may be incorporated into the digital image through scanning, and the finished piece may enter physical reality through printing and other processes.

Each display format has its own limitations, and these restrictions foist a number of compromises on the images we see. An analogy may be drawn with the piano rolls that were used in player pianos. The quality of reproduction did not depend on the execution, since the mechanical piano simply executes the instructions on the piano roll. Instead, sound quality depended on the physical qualities of the piano itself, which rendered the holes in the punchcards of the piano roll into music. Of course, the player piano was a mechanical precursor to the digital computer, and the American composer Conlan Nancarrow wrote music specifically for it, which was unplayable by a human pianist.

The digital image thus has two states: the potential image encoded in the software and the rendering shown on screen. Yet the objects it displays have definite descriptions within its own space, and obey whatever constraints have been programmed into the software. Moreover, they may be experienced with more senses than the purely visual – recent developments in haptic technology show they can now be “felt” and “moved”, with their weight and resistance determined by the parameters of the program.

The purely visual aspects of the image may be captured by cameras or on video, but these fail to convey its movement or interactivity, or its subtle tactile aspects. More importantly, the camera can only deal with the visual form; it cannot copy the digital substructure of the image. Thus a Computer artwork ceases to be digital when it is photographed, no matter what subjectively “digital” artefacts are left in the visual trace. Since a work of Computer Art includes both its visual manifestation and the dynamic operation of the art, no still picture of it can be confused with the work itself. Instead, it is a record.

The material realisation of Computer Art raises interesting questions about the “native form” of computer graphics, and whether they can become physical and yet retain some digital “quality”. Timothy Binkley sees the on-screen realisation of digital images as

arbitrary in a way media are not […] Since there is no object that can receive the blessing of conventional priority, no particular interface has any claim to being the “true” realization of the virtual image. [1]

For this reason, the status of any stage of the digital image as the “original” is hard to establish. To Binkley, this is yet another reason to disqualify the computer as a medium, because physical media by their very nature determine what constitutes the “original” of a given image.

Yet the code may also produce a physical work that exists independently of the computer. This is obviously not digital, yet has passed through a “digitised” phase. Scanners can incorporate pieces of material reality into the computer world; printers make it physical and painting machines enable the computer to inscribe on paper (as opposed to simply laying down a pattern of ink dots on a page). However, the source of this image was a set of instructions that, in principle, might have been copied directly from the artist’s computer. However, there was no sense of these being the “original”.

Even in 1971, Herbert Franke realised that Computer Art’s digital structure implied that pieces of Computer Art might be copied exactly, thereby ending the primacy of the original in the art world.[2] The only element which came close to being an “original” was the code from which the art was generated. Of course, there are salient issues of hardware compatibility and software portability, but in essence this is true. The “ideal form” of the computer image also means that its physical reification is not an inherent property of the data.

Insofar as the art consists of precise instructions describing an image, it could be said that possessing the code, as opposed to the result, is equivalent in a sense to the original of the artwork, since from this multiple copies can be produced.  This data is the “concrete” form of most Computer Art. In this context, Roman Verostko considers architectural plans, musical scores and even choreographic notation to be the precursors of algorithmic images, since they each specify a form which is executed in a certain medium:

Given sufficient detail and a sufficiently robust language, any procedure for executing a task can be translated into a computer compatible instruction (algorithm). A computer, connected to appropriate machines, can execute instructions for playing music, drawing a form, or displaying a figure moving in space.[3]

However, this lack of an original can be offset somewhat by seeing the creative process in a different light. If one understands the computer “original” not to be merely the code, but the sequence of artistic decisions that led up to the creation of the first image, then this would change the terms of the argument. In fact, “Computer Art” might be linked to a specific performance, an instance of the code being used to generate an image. Ultimately, digital art could be distributed like music or perhaps even patented by individual artists rather than depending on galleries selling original one-off works.

An analogy may be drawn with musical scores and a musical performance. The notes on the sheet are instructions to the musicians, thus their interpretation of this code and their technical ability are the principal means of rendering the music audible. The relation of the computer code to its display medium is much more exact, however; questions of individual interpretation do not arise, but technical ability is paramount.

Also, as long as the image exists in its digital form, it can only be viewed through the display attached to the computer. The computer is a necessary adjunct to the digital image, which may be one of the limiting factors in the acceptance of Computer Art – at least, as long as the physical form of the computer remains cumbersome.

[1] “The Quickening of Galatea: Virtual Creation without Tools or Media” by Timothy Binkley

[2] H.W Franke, Computer Graphics, Computer Art (London, 1971)

[3] “ALGORITHMIC ART – Composing the Score for Visual Art” by Roman Verostko, 1999, get web ref