Leonardo Thinks


Historical Opinion by by Frieder Nake

Frieder Nake looks at how information and the computer have opened up the discourse on aesthetics, reality, and the nature of art.
Art in the Time of the Artificial

In February 1965 when the first exhibit of computer art opened at the study gallery at the University of Stuttgart, Germany, showing a small selection of plotter drawings by Georg Nees [1], uneasy feelings emerged among the audience. Was this supposed to be art, if it was true that a computer had produced the works? To what extent was the computer the creator? To a considerable degree, the audience consisted of artists from the Stuttgart area. Max Bense, the philosopher who had arranged the show [2], felt obliged to cool down the disturbance by stressing a distinguishing line between art as a purely human activity and art produced by computer. The latter he called artificial art (künstliche Kunst); thus, he coined a new concept.

It is obviously unnecessary to stress the fact that art is human-made, i.e. artificial. The word “art” itself carries this connotation. When Bense used the combination of art as a purely human activity and art produced by computer, he implied that there was a very real possibility of having something like an artificiality of a second order. Supposedly, he was implicitly referring to artificial intelligence.

Now, art has always been artificial, and it is ridiculous to assume anything different. Nowhere in nature do we encounter works of art, and it is only after art had become an important facet of human culture that we discovered aesthetics in nature. We discover it by projection and interpretation only. And yet, compared to art created by the machine, traditional art appears to be natural. In this view, we take what we do without the help of a machine as natural, thus opening up a domain of artificiality for the machine. Artificial art then becomes a synonym for machine art, i.e. art that is created through a process of the partial delegation of human activities to a machine. There is nothing terribly shocking about artists delegating some of their activities to someone else, even if delegated to a mechanical or electronic machine. Indeed, many artists have been clever managers in distributing labor to their employees. When software takes the place of the artist, or of his or her helpers, however, the question of who or what is the creator shows up: the artificiality of art.

Bense very much liked the idea of generating aesthetic objects with the aid of computers because this could prove the point of information aesthetics. Information aesthetics was the heroic attempt by Bense and Abraham A. Moles [3] to use Shannon’s and Weaver’s concept of information [4] as the guiding principle for an analysis of aesthetic processes, both analytic and generative. Although some exciting insight into the nature of aesthetic processes was gained this way, the attempt failed miserably. Nothing really remains today of their theory that would arouse any interest for other than historical reasons.

The failure of information aesthetics is due to its most fascinating starting point: the radical idea of an aesthetics of the object. All subjectivism was to be banned from aesthetics: the focus was to be on measure rather than value judgment, number rather than feeling, mathematics rather than psychology. An aesthetics of the object was supposed to produce methods of measuring the object such that a quantitative feature vector would replace the aesthetic object in any matter of value judgment.

Information aesthetics failed when it became clear that information was no objective measure, but rather a subjective construct. The constructivist notion of information as an emerging quality when systems adapt to their environment turned information aesthetics into an extreme case of European scientific imperialism.

But one concept remains that was central to information aesthetics: the concept of the aesthetic object as a sign, i.e. as a semiotic entity. This great assumption has gained tremendously in importance. Information aesthetics was turned into a semiotic aesthetics, borrowing the term and concept of “sign” from Charles S. Peirce, the well-known American philosopher. [5] Bense was one of the very first in Europe to understand the great importance of Peirce’s work for any communicative processes-and art is a process of communication.

A semiotically grounded aesthetics not only opens the discourse of postmodernism, it also links parts of aesthetics to informatics, which, in my opinion, turns out to be a technical semiotics or semiotic engineering. The concept of sign is central to informatics, to aesthetics and to postmodernism. Postmodernism is the times of enhanced artificiality.

Any formalistic approach to aesthetics is capable of addressing only the lower levels of aesthetics. In particular, if computers are to play a role, only computable aspects of aesthetics may be addressed. Treatment of any real process by computer presupposes three reductionistic steps: a semiotic transformation of things to signs, a syntactic transformation of signs to “representamens” (Peirce’s concept [6]) and an algorithmic transformation of representamens to computable structures.

On the other hand, this very process of reductions opens up the field of aesthetic semioses for new algorithmic works and, thus, for a new kind of aesthetic experience. The field of algorithmic semioses is still to be explored aesthetically, on both the analytic and generative levels. An aesthetics of algorithmic semioses is more likely to produce interesting results for sequences of objects than for individual objects. Its genuine realm is the small difference between two pictures, and thus, it is more appropriate to the animated film sequence than to the single great painting on the wall.

In an oversimplification of computer art, we may identify two transformations occupying the artist. The first type of transformation takes the world as it is given and produces an aesthetic sign by abstraction. The second type of transformation takes the world as it is imagined and produces an aesthetic sign by concretization. The first transformation takes our bodily experience of moving in time and space as its starting point. The second transformation starts out from our experience of dreaming and thinking. The first type, the abstract one, is exemplified by Harold Cohen, the Briton living in the West of America. [7] The second type, the concrete one, is exemplified by Manfred Mohr, the German living in the East of America. [8]

Virtuality is not the opposite of reality. It is part of reality! Virtual reality is the semiotic domain of reality. Actual reality is the corporeal domain of reality. We now encounter signs in the state of algorithmic semioses. This is a new aspect of art and of the sciences as well. It is a fascinating-yet grossly overrated-aspect. It causes rather stupid speculation about the self-determination of the machine. It is wise to remain relaxed. We tend to interpret the world by projecting our currently most beloved artifact onto it, and then we interpret the artifacts by projecting ourselves onto them. This turns things and relations upside down. I prefer to identify signs on the computer as signs of a new type: calculated and calculating. Signs on the computer themselves become sources and sinks of signs. This makes us wonder what they really are-agents?

[1] Georg Nees, a mathematician and engineer, was at the time of the exhibition working for Siemens. He was the first to write a doctoral thesis on computer art, which he submitted to the philosophy department of the University of Stuttgart. It was published as a book (in German): Georg Nees, Generative Computergraphik (Berlin, Munich: Siemens, 1969).

[2] Max Bense was one of the founders of information aesthetics, i.e. the attempt to apply information theory to the analysis of aesthetic objects. The art exhibition of February 1965 is documented in a small booklet (in German): Max Bense, Georg Nees: Projekte Generativer Aesthetik (Stuttgart: Rot 19., 1965).

[3] Abraham A. Moles, Information Theory and Esthetic Perception (Urbana, IL: Univ. of Illinois Press, 1968; originally published in French, 1958).

[4] Claude E. Shannon and Warren Weaver, The Mathematical Theory of Communication (Urbana, IL: Univ. of Illinois Press, 1964; originally published in 1949). The famous quantitative concept of information as measured in bits is outlined in this seminal work.

[5] Peirce defined the sign as a triadic relation. A sign, according to this theory, is constituted by a representamen, an object, and an interpretant. The representamen is the tangible, corporeal ingredient; the object is that which is designated; the interpretant is the interpretation by the recipient of the sign. See The Collected Papers of Charles Sanders Peirce, Charles Hartshorne and Paul Weiss, eds., 6 Vols. (Cambridge, MA: Harvard Univ. Press).

[6] See [5].

[7] For information on Cohen and his work, see Pamela McCorduck, Aaron’s Code: Meta-Art, Artificial Intelligence, and the Work of Harold Cohen (New York: Freeman, 1990).

[8] Manfred Mohr’s work has been exhibited and cataloged all over the world during the past 20 years.
Frieder Nake is a professor of Graphic Data Processing and Interactive Systems at the Department of Computer Science, University of Bremen, Germany.
ISSN No: 1071-4391
Author: Frieder Nake, Leonardo Editorial Advisor, E-mail: nake@informatik.uni-bremen.de
Originally published in: Leonardo Vol. 31, No. 3 (1998), pp. 163-164
Print: ISSN 0024-094X, Online: ISSN 1530-9282,
DOI: http://www.jstor.org/stable/1576563.
Leonardo is a registered trademark of the ISAST.