The Social Conscience of Generative Art
After the advent of the mainframe in the 1940s, it took less than two decades for computers to reach beyond the confines of military R&D to become a widely used platform for artistic experimentation. In the late 1950s, corporate research institutes like Bell Labs and the RAND Corporation began to look into the machine’s ability to create art and music, and engineers initially orchestrated the making of these works. In 1963—two years before anything called “computer art” appeared in galleries—the trade journal Computers and Automation held its first “Computer Art Contest.” The jury awarded first prize to Splatter Diagram, a computer-generated fractal graphic of a shattered camera lens created by the US Army Ballistic Research Laboratories. In the initial years of computer art, then, it was impossible to uncouple even the most experimental uses of computer technologies from their association with what the historian of science Peter Galison has called “the Manichean sciences” of militarized command and control.¹
Computers were first developed to execute the complex calculations that were increasingly part of modern warfare: ballistic trajectories, cryptography, speculative modeling to virtually test the atomic bomb. Creating and maintaining the then immense and unwieldy machines required sprawling conglomerates with both public and private funding—RAND and the US Army Ballistic Research Laboratories being two of the more salient examples. Entanglements of corporations, the military, and universities made computers possible and were instrumental in determining their future uses. It was not incidental that the “glitches” that happened in ballistic laboratories came to be called “computer art,” nor that companies like RAND initiated artistic programs. These organizations had incentives to foster an expansive research program. They recognized the possibility that the innovative, unprecedented results would, in some unforeseeable but immensely profitable way, be wrangled to ensure a competitive edge in one of the many arenas where computers were beginning to take hold.
Importantly, computers continued to make their way into art and culture not in spite of their entanglement with the military-corporate research complex, but because of it. The artworks, exhibitions, actions, and texts that comprise the early history of generative art were meant not only to integrate computers into artmaking, but also to reimagine the political agency of artists and artworks alike. Generative art, in other words, was tied to a generative understanding of art’s political role.
This is most apparent in the writing and work of two of the most influential figures in the history of generative art: German cybernetician and philosopher Max Bense, who sought to quantify aesthetic experience and computerize the creation and evaluation of art, and the German-born, UK-based artist and activist Gustav Metzger, whose notion of auto-destructive art was intended to dismantle the myth that technology was rational or neutral, and expose the destructiveness at its very core. Despite their differences, both envisioned art as a laboratory for imagining how computers could defy their militaristic origins and be structured according to alternative logics—the eruptive spontaneity of the creative act, for example, or the mutually beneficial, horizontal processes of communication and collaborative action. They each imagined generative art as a means of catalyzing a more engaged public community and a more transparent, participatory, and democratic world.
Of the two, Bense was the less likely candidate for art theorist or inspiration for activists. He studied philosophy, physics, mathematics, and geology at the University of Bonn in the 1930s and, after the war, joined the philosophy faculty at the University of Stüttgart. Bense began writing about aesthetics in the 1950s, with the concerted aim of calculating and quantifying this seemingly elusive (or at least, ineffable) subject. His conception of art and aesthetics drew on a number of precedents, namely mathematician George David Birkhoff’s 1933 notion of aesthetic measurement, which defined aesthetics mathematically as a ratio between order and complexity. Also significant were mathematician Claude Shannon’s information theory—which statistically defines information in relation to the amount of noise a signal encounters as it traverses a channel—and Norbert Wiener’s cybernetics, the study of communication within complex systems, both of which contributed to the founding of modern computers and, not unrelatedly, reconceived humans as information-processing machines.
Synthesizing these sources, Bense envisioned art as a communications network in which aesthetic information flowed between artist, artwork, and audience. He defined aesthetic information as the unexpected element within this field: aesthetic signals transmit information insofar as they are unpredictable. His formula expands on Birkhoff’s, calculating the relationship between complexity and redundancy—between regular and irregular rhythm, for example, or uniform and jagged shapes or spacing of lines. Bense’s ratio also takes into account the relationship between the number of possible variations within a particular medium or machine (tones in music, spectrum of colors, etc.) and those actualized in the work. For Bense, a successful artwork pushes the boundaries of communication and expands experience, but does not go so far as to exceed the viewer’s ability to process the work. This balance between creativity and convention, Bense claims, is art’s raison d’être. His understanding resonates with definitions of the avant-garde as breaking with tradition, but here, rupture can be automatically produced, statistically measured, and mathematically defined.
Counterintuitively, given Bense’s fixation on “information,” none of this amounts to new knowledge or meaning in any common sense of the term. In Bense’s view, aesthetic information measures the formal novelty of a work. His theory wrests aesthetics from the contingency of subjective associations and the whimsy of taste, making it a methodical investigation of the possibilities and limits of creativity within a system. The subjectivity and ineffability usually associated with these kinds of experiences are eradicated, subsumed by mathematical formulas. Bense’s theory of generative art assumes not only that aesthetic experience can be measured but that it can also be rationally and predictably produced.
The culmination of Bense’s theory was a short 1965 essay called “The Projects of Generative Aesthetics,” which presented a set of schemas for evaluating aesthetic states alongside a set of formulas for stimulating it through “a methodical combination of planning and chance”—irregular or randomized elements combined with ordered ones. Production took precedent over reception; it was really a theory of generative art, rather than aesthetics. The text was written to accompany an exhibition of Georg Nees’s computer-generated graphics at the University of Stüttgart gallery, which Bense had founded in 1958. Nees was an engineer who in 1964 began experimenting with what he called “statistical graphics” at his corporate job, where he had access to computers and a programmable drawing machine, the Zuse Graphomat Z64. He used a simple algorithm to instruct the machine, for example, to distribute eight dots inside a square and connect them with a closed line. The results were simple geometric line drawings such as 8-corner (1964), a gridded arrangement of small glyphs comprising jagged, crisscrossing lines. The work illustrated not only the creativity of algorithms, but also the rational basis for aesthetically balanced forms, presumed to be beautiful. Everything about these works could be connected to some mathematical principle. As Bense put it, “the aim of generative aesthetics is the artificial production of probabilities, differing from the norm using theorems and programs.”²
The political impetus of generative aesthetics might be easily lost in this sea of equations, formulas, and technical language, if not for Bense’s own insistence on the urgency of this approach. Bense argued that rationality is humanity’s first defense against fascism. He was opposed to National Socialism, and during the war worked in Berlin as an engineer and physicist at a high-frequency engineering laboratory that made transmitters for the Navy. (According to Bense’s wife, Elisabeth Walther, the lab also secretly produced equipment for Admiral Wilhelm Canaris, a covert resister to the Nazi regime.)³ In the postwar period, Bense became a vocal critic of what he perceived as the persistence of mythical, irrational currents in German culture. Fascism might have been over in Germany, but its epistemological seeds could be found everywhere. Bense’s theory of generative aesthetics works to remove subjectivity from art and aesthetic judgment and imbue both with the transparency and clarity of science—striving, in his words, to “transform the metaphysical discipline into a technological one.”4 Underlying all the technical language of information, complexity, redundancy, and signs was an attempt to salvage meaning in an increasingly dispersed and confused information age, while at the same time expressing an anti-totalitarian skepticism about whether such shared meaning was possible or desirable. Bense’s generative aesthetics—as well as the experimental practices that it inspired in Nees and other German engineer-artists like Frieder Nake—was a sustained inquiry into the possibilities for a more transparent, participatory process of collective communication.
While the aesthetic Theory Bense developed in Stüttgart stressed the coherence of the artwork-as-network, Metzger’s auto-destructive art sought to accelerate and amplify the anti-systematic nature of the creative act. Like Bense, Metzger came to computers through a circuitous and interdisciplinary route. Metzger was born in Nuremberg, Germany, in 1926, but left in 1939 as a Jewish refugee during the Kindertransport to England. He worked as an apprentice cabinetmaker in a furniture factory in Leeds from 1941 to 1942, and lived for a brief but influential period in Trotskyist and anarchist commune in Bristol. After the war, Metzger studied art at Borough Polytechnic in London with David Bomberg, whose paintings of the 1940s testified to the catastrophic impact of technological warfare that he witnessed during the Blitz. Following Bomberg, Metzger’s artistic practice was integrated with his activism, and centered on a concern with the devastation wrought by technology; throughout the 1960s, Metzger was an active participant in the antinuclear movement. But unlike Bomberg, who expressed ideas about technology through the medium of painting, Metzger—propelled by his political concerns—experimented with new media, integrating destruction as a productive element in the artistic process.
Metzger’s first manifesto of auto-destructive art was written in 1959 and printed on a broadsheet to accompany the exhibition “Cardboards, selected and arranged by G. Metzger,” held at a coffeehouse run by his artist friend, Brian Robins, in central London. Discarded remnants of a TV-set carton were displayed without any modification, highlighting the material debris produced by this exciting new technology—a seemingly democratizing innovation—as evidence of the inherent wastefulness of consumer capitalism. “Auto-destructive art is primarily a form of public art for industrial societies,” Metzger’s manifesto began. Comprising only eight sentences, the text asserts that auto-destructive art is an art of process; it doesn’t matter what material is used or who makes it (“The artist may collaborate with scientists, engineers”).5 What is crucial to auto-destructive art is that disintegration, ephemerality, and, most important, destruction all figure centrally in the work.
In Metzger’s subsequent manifestos, the message is increasingly political and the metaphor of destruction more compounded. The second, written in 1960, is explicitly anti-capitalist and anti-war:
Auto-destructive art re-enacts the obsession with destruction, the pummeling to which individuals and masses are subjected. Auto-destructive art demonstrates man’s power to accelerate disintegrative processes of nature and to order them. Auto-destructive art mirrors the compulsive perfectionism of arms manufacture.6
The third manifesto (1961) ends even more assertively: “Auto-destructive art is an attack on capitalist values and the drive to nuclear annihilation.”7
The second manifesto was meant to accompany Metzger’s Model for an Auto-destructive Monument (1960), three steel towers that would gradually disintegrate over a period of ten years. Since he was unable to secure funding, the monument was never built. Instead, he organized his first public lecture/demonstration, employing a strategy he would repeat several times: stretching nylon over a sheet of glass and using brushes and a spray gun to soak the fabric with hydrochloric acid, so that the nylon disintegrated. Metzger wore a gas mask during the performance, which gave it a sense of urgency and anticipated apocalypse.
Later incarnations incorporated new media, and thus, degrees of mediation. In a 1963 lecture at the Bartlett School of Architecture, London, the artist used a projector, covering slides with nylon and acid to display the process of disintegration on a screen. In the 1965 Chemical Revolution in Art, Metzger replaced the nylon/acid combination with heated liquid crystals—a tactic he would later use to create luminous stage backgrounds for bands such as Cream and the Who.8 Whether performed or projected, Metzger’s works function as apt illustrations of how to render destructive processes creative and humane. Consistent with his largely discursive output, Metzger organized the “Destruction in Art Symposium” in London in 1966, a two-day conference that drew an international array of artists, including Yoko Ono and Vienna Actionist Hermann Nitsch, and evidenced the substantial overlaps among body art, performance, and new media practice at this time, especially a shared concern with the intertwined histories of violence and rapid technological change.
Alongside these lectures, performances, and discussions, and increasingly in his manifestos, Metzger integrated the computer as a privileged collaborator in the creation of auto-destructive art.9 From 1965 to 1969, he conceived and modeled various versions of Five Screens with Computer, a computerized, degenerating public monument meant to occupy the courtyard of an apartment complex.10 Five walls, each thirty feet high, forty feet across, and two feet deep, would be spaced twenty-five feet apart. Each wall would be composed of 10,000 pieces of steel, plastic, or glass, which would be randomly ejected over the course of ten years. The timing of these projectiles would be programmed by a computer, factoring in variables such as weather and the revolution of the earth. Metzger imagined that a barrier made of glass or compressed air could adequately keep people out of the danger zone as the elements crash to the ground. These sporadic projectiles and the gradual disintegration of these public structures would constantly remind residents that destruction and entropy ultimately govern their material world. As Metzger put it in 1965, “architecture can be used to undermine values, to protest, subvert and destroy.”11
As this summary may already suggest, there is a profound ambiguity at the basis of Metzger’s auto-destructive art. Is the work meant to demonstrate the destructive logic at the heart of our technologically mediated world? Or does auto-destructive art render the negative into a positive, harnessing disintegration as a creative principle? It seems almost impossible for destruction to be both, but this is exactly what Metzger posits. The worst manifestations of destructiveness—war, waste, environmental catastrophe, and the capitalist system behind them—all need to be laid bare and eliminated. But destruction, at the metaphorical level of the artwork or social critique, can be wielded to create a productive opening. This positive effect of destruction is evident in Five Screens. As Metzger explained in 1971, “Since there is no viable power structure or ruling ideology, the work can serve as a testing-ground for the forging of new concepts of social reality.”12 Here, the disintegration of the work prompts a collective reimagining of possibilities.
The social significance of Five Screens therefore derives from its interrelation with the surrounding community, especially in the presumed analogies between artificial and actual life. Because of its changing nature, Metzger believed that Five Screens would give people a touchstone for understanding, communicating, and relating their own personal development. People would remember births and deaths in terms of where the sculpture was in its process. They would mark important events by increasing the number of projectiles—like a permanently integrated and adaptable fireworks display. Not volatility but rather variability is the work’s key principle: Five Screens, as an adaptable, interactive system, would serve as a concretization of the collective yet mutable dimension of the spaces and events that the residents share.
Usually, discussions of Metzger’s work foreground his critique of computers, with the technology’s inextricability from wartime destruction and militarism assumed to be the central message of his auto-destructive art. But looking at his writings and works in which computers were most centrally featured brings forth another dimension of his idea, one that comes closer to Bense’s notion of generative art. Metzger’s values of spontaneity and randomness seem to contrast starkly with Bense’s prioritization of programming and transparency, but both are resolutely concerned with communication. Each sees the computerized creation of artworks as a concretization of some kind of more ethical collective life.
For Bense, this manifests in a circuit of communication based on universally affective signals and signs; for Metzger, in a shared social space that can connect individual and collective experiences, events, and temporalities. For both, the ethics of their envisioned systems stems from their transparency, dynamism, and ability to be understood by everyone. Both Bense and Metzger produced elaborate theoretical treatises—be it manifestos or multiple philosophical volumes. Clear communication is central to what they thought computers could give us, and essential to whatever future collective life they imagined such technologies might support.
Bense and Metzger understood that technology was neither neutral nor unilateral. At the same time, they recognized that it has the capacity to actively support and sustain certain kinds of life, action, and existence. At a time when it once again seems impossible to parse the negative from the positive effects of computerization, and the creative from the destructive consequences of the social networks that they support and deny, it is instructive to look back on figures who imagined other futures. Following the generative politics of Bense and Metzger’s visions would mean probing the constitutive power of technology to establish collective experiences and working to create, differently.
1 Peter Galison, “The Ontology of the Enemy: Norbert Wiener and the Cybernetic Vision,” Critical Inquiry, vol. 21, no. 1, Autumn, 1994, pp. 228–266.
2 Max Bense, “The Projects of Generative Aesthetics,” in Jasia Reichardt, ed., Cybernetics, Art, and Ideas, London, Studio Vista, 1971. Originally published in Bense’s journal Rot 19, in 1965. For a more detailed account of Bense and “the Stüttgart School” in English, see Christoph Klütsch, “Information Aesthetics and the Stuttgart School,” in Hannah Higgins and Douglas Kahn, eds., Mainframe Experimentalism: Early Computing and the Foundations of the Digital Arts, Berkeley, University of California Press, 2012, pp. 65–89.
3 Elisabeth Walther, “Max Bense und die Kybernetik,” in 10 Jahre Computer Art Faszination, ed. Gerhard Dotzler, Frankfurt, Dot Verlag, 1999, p. 360. The founder and director of the Berlin laboratory, Hans Hollmann, moved to the United States after the war to work for NASA. In 1949 he sent Bense a copy of Wiener’s writing on cybernetics.
4 Bense, quoted in Claus Pias, “Hollerith ‘Feathered Crystal’: Art, Science, and Computing in the Era of Cybernetics,” tran. by Peter Krapp, Grey Room 29, Winter 2008, p. 120.
5 Gustav Metzger, “Auto-Destructive Art,” (1959), in Kristine Stiles and Peter Selz, eds., Theories and Documents of Contemporary Art: A Sourcebook of Artists’ Writings, Berkeley, University of California Press, 1996, p. 401.
6 Gustav Metzger, “Manifesto Auto-Destructive Art,” (1960), in Stiles and Selz, eds., Theories and Documents of Contemporary Art, pp. 401–02.
7 Gustav Metzger, “Auto-Destructive Art, Machine Art, Auto-Creative Art,” (1961), in Stiles and Selz, eds. Theories and Documents of Contemporary Art, pp. 402.
8 Metzger had lectured at London’s Ealing School of Art in 1962, possibly inspiring in part the explosive stage antics of then student Pete Townshend, soon to become famous as lead guitarist of the Who, a band known for ending their performances by destroying their instruments.
9 The second manifesto specifies: “The immediate objective is the creation, with the aid of computers, of works of art whose movements are programmed and include ‘self-regulation.’” Gustav Metzger, “Manifesto Auto-Destructive Art,” 1960, in Stiles and Selz, Theories and Documents, p. 402.
10 Metzger writes that Five Screens “should be sited as a central concourse between three very large blocks of flats, housing more than ten thousand people, preferably in the country.” Gustav Metzger, untitled paper on theme number three for the “Computers and Visual Research” symposium, Zagreb, 1969, Bit International 7, 1971, p. 31.
11 Gustav Metzger, Auto-Destructive Art: Metzger at AA, London, Bedford Press, 2015, p. 22. This is a transcript of a lecture given at the Architectural Association in London, February 24, 1965. This is important, because the meaning of the work requires a consistent community, and thus could not be sustained in either a public park, for example, or a private home.
12 Gustav Metzger, “Sculpture with Power” in Higgins and Kahn, eds., Mainframe Experimentalism, p. 217.
This article appears under the title “The Social Conscience of Generative Art” in the January 2020 issue, pp. 50–57.