In Analogy to Painting

Vera Molnar’s Computer Paintings

by Aline Guillermet

The Hungarian-born French painter Vera Molnar pioneered the use of the computer as a creative medium beginning in the late 1960s. This article by Aline Guillermet explores how Molnar’s computer-generated works used programming as a means to reflect on the autographicity of the handmade trace in drawing and painting.

The essay begins:

In 1948, Claude Shannon and Norbert Wiener each published a work that would durably shape the concept of information for the digital age. Both mathematicians defined information in terms of entropy—a term borrowed from physics, which describes the disorganization or unpredictability of a system. While they differed in their interpretation of the term, both agreed to define information as a probability function wholly independent from material conditions. As Wiener famously stated in his book Cybernetics: “Information is information, not matter or energy.” The distinction Shannon and Wiener inaugurated between information and its context would have lasting consequences far beyond the narrow realm of communication engineering. Writing fifty years later, the literary critic N. Katherine Hayles summarized: “The time was ripe for theories that reified information into a free-floating, decontextualized, quantifiable entity that could serve as the master key unlocking secrets of life and death.”

The definition of information as disembodied pattern, in turn, led to a series of misconceptions that still inform contemporary conceptions of digital media. By contrast, Hayles argues that “for information to exist, it must always be instantiated in a medium. . . . Conceiving of information as a thing separate from the medium instantiating it is a prior imaginary act that constructs a holistic phenomenon as an information/matter duality.” In this article, I aim to extend Hayles’s analysis to the history of art by focusing on a series of computer-generated works produced by the Hungarian-born French artist Vera Molnar. A classically trained painter who started working in French computer laboratories in 1968, Molnar opened a space of mediation between the computational realm of information processing and the material practice of painting; in so doing, she directly challenged the duality between information and materiality that Hayles condemned.

Information needs to undergo a certain amount of analogizing before humans can experience it, a task that today is routinely, and more or less invisibly, performed by interfaces. Information, therefore, not only needs to be “instantiated” in a medium in the general sense, as Hayles contends, but also relies on specific material conditions in order to be perceptually experienced and cognitively processed. This holds important consequences for the visual arts. Well before the advent of the user-friendly interfaces that we know today, early technologies of data visualization and inscription enabled such an experience. Originally developed for the military during World War II, the electronic visualization and plotting of data (on a cathode ray tube [CRT] screen and on paper, respectively) was refined throughout the 1960s for the benefit of the booming postwar industry. It is in this context that a few computer scientists, working in research institutions that encouraged collaboration between engineers and artists, exploited the creative possibilities of the machines at their disposal.

As early as 1962, the engineer A. Michael Noll produced a series of “computer-produced patterns” at Bell Telephone Laboratories, using the newly acquired Stromberg-Carlson 4020 microfilm printer (also known as the “microfilm plotter”). From the outset, Noll situated his “patterns”—black and white plotted line drawings generated by connecting a series of points with straight lines—at the periphery of artistic creation in order to avoid “an unintentional debate at this time on whether the computer-produced designs are truly art or not.” However, the relation between these computer-produced works and modern visual culture was far from unambiguous. While Noll did not set out to create “art,” several of his productions appropriated an aesthetic derived from the canon of modern painting, whether intentionally or not: for instance, one of his first patterns, Gaussian-Quadratic (1962), “reminded [him]” of Pablo Picasso’s cubist painting Ma Jolie of 1911–12. Two years later, Noll also produced a series of computer-generated simulations of Piet Mondrian’s Composition in Line (second state; 1916–17)—a work to which I shall return in reference to Molnar’s practice.

In Europe, too, computer graphics originated within a scientific context. At the University of Stuttgart, the first works of computer art were produced between 1963 and 1964 by the mathematics students Frieder Nake and Georg Nees using the Zuse Graphomat Z64 plotter. Yet these early works were already embedded in a broader intellectual environment that sought to rethink the production and reception of art in the age of the computer. As early as 1964, the inventor of the Graphomat Z64, Konrad Zuse, had foreseen that his device—while primarily designed for technical purposes—could be put to artistic use. Moreover, Nake and Nees were closely associated with the philosopher Max Bense, whose information aesthetics aimed at developing a mathematical framework for the perception and creation of art. For these pioneers of computer graphics, modern art and visual culture functioned as a point of reference rather than as an example to emulate. However, as the context I have sketched indicates, their endeavors existed in tension with, rather than outside of, the artistic sphere.

By the late 1960s, a small number of artists had started experimenting with computers. For these artists, computer graphics did not exist in contradiction to modern art; rather, they envisioned using the machine to further develop features usually associated with human-made creation, such as inventiveness and organicity. Foregrounding the importance of materiality at the point when the algorithm is not only traced but also drawn and painted became a key strategy to legitimize the computer as an artistic tool. At the time, however, such works were largely dismissed by critics of traditional art, who failed to recognize any aesthetic specificity to the artistic appropriation of the computer as medium. Today, computer art is no longer an object of condemnation; yet it has remained on the periphery of the artistic canon and is generally classified by museum institutions under the category of print media. While media history and media archeology have done much to clarify the technological context of emerging computer art, such approaches fail to account for the way computer art has related to fine art. Among those artists who saw in the computer a new means to expand the possibilities of painting, Molnar best brought into productive discussion the so-called immateriality of the algorithm and the materiality of the computer-generated trace.

Molnar studied painting at the Budapest College of Fine Arts between 1942 and 1947 and moved to Paris in 1947. Between 1947 and 1960, she collaborated with her husband, François Molnar, an academic researcher in experimental psychology at the CNRS (Centre national de la recherche scientifique), the French national center for scientific research, on artistic productions they saw as “scientific experiments.” Having gained access to a computer in 1968, Molnar relied on the new discipline of information aesthetics, developed independently by the French physicist and philosopher Abraham A. Moles and the German philosopher Max Bense during the 1960s, as a theoretical framework for her creative practice. As a result, the balance between randomness and redundancy—crucial to information theory and its application to the visual arts—became central to the dialogue between classical painting and computer programming that she inaugurated. The manipulation of random parameters, in particular, enabled the production of “autographic” effects, such as trembling and hesitation, suggesting that the plotted line could imitate key characteristics of the handmade trace.

In this article I argue that Molnar’s computer-generated works reflect upon painting as a practice, a historical tradition, and an aesthetic experience. This claim, in turn, invites a reassessment of computer art in relation to the materiality of painting. The elements that constitute computer art as a medium—the coding process, the computer, the screen, the plotting table—all engage with materiality in ways that are fundamentally different from drawing and painting. Molnar’s production, however, shows that as soon as code becomes materialized, it is perceived in relation precisely to these existing instances of materially instantiated visual culture. My argument is structured around three series of works realized between 1973 and 1988. The first section deals with Molnar’s appropriation of information aesthetics in two series of plotter-drawn variations, Hommage à Barbaud (Tribute to Barbaud; 1974) and Computer-rosace (Computer-rose; 1975), in which Molnar experiments with Bense’s concept of an “aesthetic state.” In the second section, I investigate the different ways in which the materiality of computer-generated paintings comes to the fore, first in (Dés)ordres ((Dis)orders), a plotter drawing in color ink on white paper (1974), and second in a group of acrylic paintings begun in 1973 entitled Computer icône (Computer icon). In the last section, I focus on a single series of works produced during the 1980s, entitled Lettres de ma mère (My mother’s letters), to suggest that coding, in simulating autographic qualities, may evoke the materiality of painting. Continue reading …

is a Junior Research Fellow at King’s College, University of Cambridge. Her research interests include postwar German art, the impact of technology on painting, and digital art. She is currently completing a book on Gerhard Richter.