This page has been archived by the Computer Arts Society.

Applying Nelson Goodman’s theories to Computer Art

<<Previous Up Next>>

William J. Mitchell considers digital images to be “problematic” since they lack negatives or any other unique instance. “An image file may be copied endlessly, and the copy is distinguishable from the original only by its date since there is no loss of quality.” He also points out that digital images synthesised from geometric data have no obvious original due to the range of possible renderings of the same data. “Is the rendering procedure really the original”? [1]

One solution he considers is the idea of idea of “one-stage” and “two-stage” arts, proposed by art theorist Nelson Goodman. Sketching or taking a Polaroid is one stage, for instance, and production of music (composition plus performance) is two-stage: an act of composition followed by a performance. Digital imagery, argues Mitchell, is two-stage: it is first coded then displayed. He also points out that “in a two-stage process, the work is often divided among different individuals”, i.e. that different people work on different stages of the process. It is still the same piece of art, however.

This relates to Goodman’s distinction between autographic and allographic systems. A musical score, which is written in musical notation and thus comprehensible to performers, is allographic, whereas a painting, which is produced as a unique image rather than a symbolic system, is autographic. In Goodman’s opinion, not only can the score be copied exactly, but each instance of it may be regarded as “genuine”. The authenticity rests not in its material form, so long as it is accurate to the composer’s work, but in the more intangible aspects of its performance.[2]

Goodman proposed that “allographic” artforms were amenable to replication because they consisted of sequences of symbols (like musical scores or text) rather than arrangements of concrete structures executed in a one-off painting by an artist. However, in the case of painting which has no such notation system:

none of the pictorial properties – none of the properties the picture has as such- is distinguished as constitutive; no such feature can be dismissed as contingent, and no deviation as insignificant. [Goodman][3]

This means that any painting made by another hand than that of the purported artist is a forgery. Mitchell then goes on to discuss something he touched on before: the fact that autographic or analogue works cannot be reproduced without noise or degradation, whereas allographic or digital works can be copied exactly. Of course, the appropriate software is required to produce and view these additional copies. For instance, the layers of a Photoshop image may only be retained as separate layers – and thus amenable to alterations – in a program supporting the Photoshop file format.

As Timothy Binkley explains, whereas analogue media carry image and sound through transcription, i.e. an impact on the medium that corresponds to the visual or auditory form of the original event, digital media work by conversion, turning the original information into “formal relationships in abstract structures.”[4] Thus the computer creates a visual display from a set of instructions, though the viewer always approaches it as an image with a set of inherent assumptions about its existence on or within a flat surface.

Perhaps this is only problematic for those instances of Computer Art that emulate static pictures. On the other hand, Computer artworks that take full advantage of the dynamic and interactive nature of the computer cannot be bound by the conventions of painting, even in their presentation. A work that is subject to user modification is cannot have an “original” because it has no fixed original state.

As with recorded music, the visual form of the code can be captured on film or paper and relayed in purely analogue formats, but in doing so loses its digital basis, even if at a later stage it is re-scanned and thus converted back into digital form. It could be argued, therefore, that the digital code from which the piece of Computer Art is constructed is its original, and it is this that governs the picture’s authenticity. Thus even direct copies may be considered “authentic”, provided they are made at the level of the code. Goodman considers that “all accurate copies, even if forgeries of [the] original manuscript, are equally genuine instances of the score. Performances may vary in correctness and quality and even in authenticity of a more esoteric kind; but all correct performances are equally genuine instances of the work.”[5]

It could be argued, therefore, that the digital code from which the piece of Computer Art is constructed is its original, and this governs the picture’s authenticity. Thus even direct copies may be considered “authentic”, provided they are made at the level of the code.

The implication for digital art made of discrete data is interesting. Though a visual art, it has neither the unique instance of a painting nor the inherent reproducibility of film. Rather, its raw state consists of a code by which its visual form may be reconstructed anew each time it is displayed. This digital form can be copied exactly without any intervening loss of quality and shares with the musical manuscript the necessity of “performance” to release the art in a form that can be appreciated by its viewers.

The lack of an identifiable original creates obvious problems for the reception and distribution of Computer Art through existing channels. However, it is a problem that etching and printmaking also faced previously, though even there an original plate could be identified. As Martin Kemp says:

The uniqueness argument is also important – though the great print-makers, Durer above all, transcended it.  […] If the procedures are hidden in the machine and the digital code allows for perfect replicability (in theory at least), there’s a double problem.[6]

In 1987, Rodney Chang identified the lack of an original, and the potential for replication, piracy and disk corruption, as major obstacles in the uptake of the computer in art. He also noted the problem of transient digital media and the possible deterioration of storage formats.[7]

When Andy Warhol was questioned whether his Computer artwork would somehow threaten the value of his original art, he replied that his computer pieces were “sketches”; they were more in the nature of drawings than his large prints. He also believed that in time they come to be seen as “artworks in their own right”.[8]

It would seem, then, that even if there is no physical original in Computer Art, there is an original act that should be identified and recognised. As the artist progresses from their initial idea to its final, or most recent, realisation, they build up a sequence of actions that taken in their entirety constitute a non-repeatable act of which the piece of art is the ultimate expression. In other words, even if the code is transferable, the act and its understanding is not necessarily transferred. Perhaps this goes some way towards defining the “original’s” status.

[1] William J. Mitchell, The Reconfigured Eye, pp49-50.

[2] Ref to Goodman

[3] Ref to Goodman

[4] “Refiguring Culture” Timothy Binkley, Future Visions: New Technologies of the Screen, eds. Philip Hayward and Tana Wollen (London: British Film Institute, 1993): 90-122. Also at http:// find ref

[5] Gooodman, ibid, p113

[6] Email to Catherine McIntyre, November 2001

[7] Chang, Rodney “Computer Images as “Fine Art”: Some problems in the field/medium”, 1987. Web ref?

[8] Warhol interview: AmigaWorld Jan/Feb 1986, pp19-20