A. Michael Noll and the Birth of Computer Art

A. Michael Noll and the Birth of Computer Art

Few of the pioneers who shaped the intersection of art and technology have left as palpable an influence as A. Michael Noll. At a moment when computers were largely seen as instruments of calculation, Noll interrogated the computer’s capacity for creativity, asking how the machines might contribute to a broad array of artistic disciplines.

From 1962 to 1968, he pursued this question through a combination of artistic experiments, public demonstrations, and detailed technical papers, which together made one of the most compelling cases for using the computer in the arts. His work has been featured in many of the defining exhibitions and canonical histories of computer art, and his oeuvre has exerted tremendous influence on the arts, extending to prominent figures such as Nam June Paik and Salvador Dalí.

Noll’s passion for the arts formed early. As a child he loved classical music, and while in high school he worked part-time as a salesman at Hudson Radio, a stereo equipment store.¹ Noll took the job as an opportunity to learn about the technical principles of audio reproduction and innovations in the field. By the late 1950s, Noll had published several technical articles of his own in Audiocraft magazine.²

Bell Labs, 1962

After graduating from Newark College of Engineering³ in 1961, Noll joined the technical staff at Bell Telephone Laboratories, Inc., in Murray Hill, New Jersey, where he studied human responses to telephone audio distortions while pursuing his master’s degree in engineering at New York University.

During the summer of 1962, he worked under Manfred R. Schroeder in the acoustics research department at Bell Labs, where he gained access to the IBM 7090 mainframe and Stromberg-Carlson SC-4020 microfilm plotter. He used the computer to prove Schroeder’s theory that the pitch of human speech can be accurately determined using cepstrum analysis, a method of identifying periodic structures in a sound’s frequency spectrum.⁴ This research was a crucial step toward the creation of natural-sounding synthetic speech. Noll continued to make significant innovations in the field of cepstrum analysis, ultimately joining the department full-time. In 1967, Bell Labs applied his research to the American space program, analyzing the final communications of Apollo 1.⁵

In the 1960s, engineers and researchers at Bell Labs typically shared time on one central computer, dividing its use across departments. While processing the massive amounts of data needed to prove the cepstrum method, Noll found himself working alongside Elwyn Berlekamp, who was completing an internship under John Larry Kelly Jr. in the Information Coding and Programming Department. After accidentally executing a programming error that reduced graphical precision to abstract chaos, Berlekamp jokingly proclaimed the tangle of lines as “computer art.”⁶

“Patterns by 7090”

Aware that his contemporaries had used the computer to create musical compositions, Noll asserted that the same machine could be used to make visual art as well.

Noll started working on cepstrum analysis during the day while programming computer-generated artworks by night. In this manner, he produced an exploratory body of artworks and selected eight examples for a series he titled “Computer Art by 7090” (though corporate management insisted on changing it to “Patterns by 7090” due to concerns about potential backlash from the arts community and complaints from AT&T).⁷ One of the earliest instances of computer-generated art, “Patterns by 7090” was published as a Bell Labs technical memorandum on August 28, 1962.⁸

Gaussian-Quadratic and “Vertical-Horizontal”

As one of the few artist-programmers to emerge in the 1960s, Noll positioned the computer as his closest collaborator. He defined each artwork’s structure through programmed rules, allowing the computer to calculate the exact arrangement of lines and dots within the mathematical and statistical limits he coded. While appearing random to the passive observer, the generated numbers were derived from defined values and a standard deviation. Noll maintained control over the field of probability without explicitly defining each point.

Gaussian-Quadratic (1962–64), one of his earliest titled works, is a symphony of artistic agency, calculated randomness, and mechanical intervention. Vertically guided by quadratic order and horizontally guided by Gaussian randomness, a single line ascends through the composition in increasingly large, unpredictable steps. Upon exceeding the SC-4020’s upper limits, it abruptly returns to the bottom, creating an unintentional resemblance to Picasso’s Ma Jolie (1911–12). Noll first executed the work in 1962 and distributed copies to guests and colleagues, often embellishing the composition by hand to add color.⁹

Expanding on this interplay of artistic agency and calculated randomness established in Gaussian-Quadratic, Noll’s first titled series replaced mechanical intervention with programmed structure. Across the three works in “Vertical-Horizontal” (1964), Noll interrogated the results of variation within a fixed algorithmic system. The program selected a new random position along one axis at a time, alternating between horizontal and vertical to produce a continuous right-angled line whose density, proportion, and directional emphasis shifted with Noll’s adjustments to segment count and coordinate range. Of the three works in the series, Vertical-Horizontal No. 3 is the best known, having appeared in many exhibitions and publications.

Riley and Mondrian

Over the course of 1964, Noll’s exploration of two-dimensional imagery gained new depth. Seeking to engage with Op art, and inspired by Gaussian-Quadratic’s unintentional resemblance to Picasso’s work, Noll created Ninety Parallel Sinusoids with Linearly Increasing Period (1964), a deliberate attempt to re-create the visual effects of Bridget Riley’s Current (1964).

Where Current was defined by a dense field of undulating hand-painted lines, Noll’s version showed the computer’s ability to approximate the aesthetic using the repetition of the smooth curves of a sine wave. The work ultimately reached Riley, but she overlooked Noll’s attempts to preserve her authorship, instead seeing the computer as a direct competitor to her practice.¹⁰

While Riley’s Current aligned with a definite mathematical structure, Noll’s Computer Composition with Lines (1964) reinterpreted an artwork whose structure was almost entirely obscured: Piet Mondrian’s Composition with Lines (1917). Recognizing that Mondrian’s composition had been guided by an unseen visual order, Noll identified the patterns in the work, devised an algorithm to emulate them, then refined the parameters until the algorithm not only approximated Mondrian’s composition, but also created variations.¹¹

Where Ninety Parallel Sinusoids with Linearly Increasing Period proved the computer could re-create the visual effects of an artwork, Computer Composition with Lines went further, testing whether a computer-generated composition could rival the human original. Around 1964, Noll surveyed 100 employees at Bell Labs, showing them both the original Mondrian and his computer variation. He asked which they preferred, and which they thought was made with a computer. Fifty-nine percent chose Noll’s work as their favorite, with 72 percent misidentifying Mondrian’s as the computer-generated art.¹²

Around 1966, Noll revisited the algorithm he had written for Computer Composition with Lines to see whether individuals with artistic backgrounds preferred a particular level of randomness, and if their aesthetic judgment was shared. In a study of 124 undergraduate students, including 39 with artistic training and 85 without, he found preferences to be entirely individual, independent of artistic training.¹³

Three Dimensions and Dalí

In 1964, Noll extended his inquiry into three, then ultimately four, dimensions. Influenced by a stereoscopic viewer he used as a child, Noll developed a method for producing stereoscopic computer graphics, one of the earliest three-dimensional experiments of the era.

Salvador Dalí encountered this aspect of Noll’s work while reading Jasia Reichardt’s 1971 book Cybernetics, Art and Ideas and the catalogue for her 1968 exhibition “Cybernetic Serendipity.” A decade later, he would declare:

Dalí had been attempting to innovate in dimensional expression within his practice for decades, hand-embellishing stereoscopic viewing cards as early as 1950. He made his first three-dimensional works in 1965, using stereoscopic photography as a geometric base, but was unable to achieve a true sense of depth until discovering Noll’s research. He used Noll’s Three-Dimensional Picture Pair of 85 Mutually Perpendicular Connected Line Segments (1964) and the computer graphics of William A. Fetter in a collage demonstrating the technique.

Applying the method to painting, the Surrealist appropriated Noll’s Three-Dimensional Picture Pair of a Random Structure Consisting of 50 Lines Falling at Random Within a Cube (1964) as the central component within one of his first three-dimensional paintings, The Sleeping Smoker (1973). Dalí used this technique to execute a number of stereoscopic paintings over a two-decade period, ultimately pioneering a new form of painting. His biographer Robert Descharnes described the process:

Virtual Sculpture and Computer Ballet

Expanding the computer stereography method into animation in 1965, Noll produced two computer-generated studies of virtual sculpture.

Kinetic Sculpture (1965) sequentially connected 40 randomly distributed points within a virtual cube to form a continuous 39-segment structure in three-dimensional space—an early example of what we now call “virtual objects.”

The second work, Noll’s Computer Sculpture Combining Random Scattering of Lines About Specified, but Never Drawn, Trend Lines (1965), was a virtual reconstruction of Richard Lippold’s Orpheus and Apollo (1962), a geometric sculpture in Lincoln Center’s Philharmonic Hall. The work modeled the sculpture through randomized lines distributed around specified, but never drawn, trend lines. It captured the attention of conductor Hermann Scherchen¹⁶ during his visit to Bell Labs. Noll’s early computer-generated films prompted a thorough discussion of the medium with the conductor at the reception following Scherchen’s performance there that same evening.

Music continued to be a source of inspiration for Noll. While attending a performance of Stravinsky’s ballet Apollo, choreographed by George Balanchine, Noll conceived Computer Ballet (1965). Powered by the lab’s newer IBM 7094 and visualized with the Stromberg-Carlson SC-4020,¹⁷ the program was designed to let choreographers visualize dancers’ movements without the need for actual dancers or rehearsals. While a few hours passed between programming a sequence and seeing it plotted, the method could radically accelerate the choreographer’s creative process. Presented in various theoretical configurations in the years thereafter, it offered a poignant glimpse into the medium’s future.

The Howard Wise Show

Thanks to his hand-embellished prints of Gaussian-Quadratic and the many technical memos he published, Noll’s artwork became well known at Bell Labs. While other scientists, such as Béla Julesz, had been producing designs by computer as early as 1959,¹⁸ the resulting images were intended only as scientific visualizations.¹⁹ Noll’s works were the earliest intentional instances of visual art made by computer at Bell Labs, and some of the first globally.

One of Julesz’s computer-generated scientific images was printed on the cover of the February 1965 issue of Scientific American, attracting the attention of art dealer Howard Wise. His New York gallery had a reputation as one of the city’s most cutting-edge, thanks to past presentations of kinetic sculpture and Op art. Wise instantly recognized the artistic merit of computer-generated imagery. Wise invited Julesz to exhibit his scientific images as fine art, and when Julesz mentioned Noll, Wise included him in the show. Noll, in contrast, had seen his computational outputs as art from their inception.²⁰

“Computer-Generated Pictures” marked a pivotal moment for the institutional validation of computer art. The exhibition of works by Noll and Julesz was presented in New York City at Howard Wise Gallery, April 6–24, 1965. Wise didn’t just provide a venue for the show. He took care to elevate the work, mounting it on Masonite or between backlit plexiglass.²¹ He persistently defended the show, going so far as to threaten litigation against AT&T, the parent company of Bell Labs, when it tried to cancel the exhibition.

AT&T didn’t want to be in the business of producing art and culture, so the company’s legal team encouraged Noll to copyright his work in his own name. Noll took their suggestion, filing Gaussian-Quadratic for copyright protection, effectively transferring ownership from Bell Labs to himself.²²

“I filled in the copyright form and posted it to the Copyright Office, describing the work as computer art. The Office replied that a copyright could not be registered for a work done by a machine. I then described that a program by a human controlled the machine, and that the program combined a mathematical equation with elements of randomness. The Office then replied that a work produced by randomness could not be registered. I then explained that a program calculated the ‘randomness,’ and the numbers produced only appeared ‘random’ to humans—algorithms did all. The Copyright Office was then finally satisfied, and the copyright was registered.”

– A. Michael Noll, 2023²³

Paik at Bell Labs

While not heavily publicized, Noll and Julesz’s exhibition had a tremendous influence on the arts, capturing the attention of Nam June Paik, who in a letter to Noll dated November 30, 1966, wrote: “I have been following your work with great interest since the Howard Wise show in 1965.”²⁴

Noll and Max Mathews, the Bell Labs engineer widely regarded as the father of computer music, visited Paik’s Canal Street studio. Mathews then invited the video art pioneer to create computer artworks on the IBM 7094, with Noll responsible for Paik.²⁵ Working at Bell Labs, Paik created a series of artworks in 1967 and 1968.²⁶ While Noll spent only a day or two instructing Paik on use of the computer, the time left a considerable influence on the artist. Writing in 1968, Paik reflected on two core motifs from Noll’s oeuvre, randomness and repetition, and concluded, “I personally feel that high level applications like Noll’s … means the augmentation of new knowledge, and creation of new art…”²⁷

While many pioneers of digital art sought to establish the computer’s place within the broader canon of art, few exerted an influence as far-reaching as Noll. Through his artworks, demonstrations, and writings, he made the computer’s artistic possibilities legible to artists working beyond the emerging field of computer art. He shaped the thinking of some of the era’s most important artists. Working against the stubborn pretense that “engineers will never understand art,” Noll produced a body of computer-generated work that has become among the most institutionally recognized in digital art history, appearing in nearly every major survey of the field since its inception.

Notes

1. A. Michael Noll, an oral history conducted in 2016 by Mary Ann Hellrigel and Michael Geselowitz, IEEE History Center, Piscataway, NJ, USA.
2. Noll, A. Michael. “Stylus Brush.” Audiocraft, vol. 2, no. 11, Nov. 1957, p. 39.
3. Noll, A. Michael. “Classical Music Prelude to Engineering.” Engineering and Technology History Wiki, 28 Sept. 2022.
4. Noll, A. Michael. Memories: A Personal History of Bell Telephone Laboratories. Quello Center, Michigan State University, 6 Aug. 2015.
5. Noll, A. Michael. “Apollo 1 Disaster Tape Analysis by Bell Telephone Laboratories, Incorporated.” Engineering and Technology History Wiki, 11 Sept. 2015.
6. Noll, A. Michael. “Computer Art at Bell Labs.” Predella, no. 30, 4 Dec. 2011.
7. Noll, A. Michael. “Early Digital Art at Bell Telephone Laboratories, Inc.” Engineering and Technology History Wiki, 2 Feb. 2014.
8. Noll, A. Michael. “Patterns by 7090.” Bell Telephone Laboratories, Inc., 28 Aug. 1962. Technical memorandum MM-62-1234-14.
9. Noll, A. Michael. “The Story of Gaussian-Quadratic.” 28 Mar. 2026. Unpublished manuscript. Spalter Digital Library.
10. Riley, Bridget. “Angeblich abstrakt.” Interview by Robert Kudielka, translated by Robert Kudielka and Inga Rensch, Parkett, no. 61, 2001, pp. 30–39.
11. Noll, A. Michael. “Human or Machine: A Subjective Comparison of Piet Mondrian’s Composition with Lines (1917) and a Computer-Generated Picture.” The Psychological Record, vol. 16, 1966, pp. 1–10.
12. Noll, “Human or Machine.”
13. Noll, A. Michael. “The Effects of Artistic Training on Aesthetic Preferences for Pseudo-Random Computer-Generated Patterns.” The Psychological Record, vol. 22, 1972, pp. 449–62.
14. Quintà, Alfons. “Dalí: ‘Tengo tantas ideas que me vienen a la cabeza...’” El País, 27 Aug. 1981.
15. Descharnes, Robert. Salvador Dali. Translated by Eleanor R. Morse, Harry N. Abrams, 1976, p. 124.
16. Noll, “Classical Music Prelude to Engineering.”
17. Noll, A. Michael. “Computer-Generated Three-Dimensional Movies.” Computers and Automation, vol. 14, no. 11, Nov. 1965, pp. 20–23.
18. Julesz, Bela. “Binocular Depth Perception of Computer-Generated Patterns.” The Bell System Technical Journal, vol. 39, 1960, pp. 1125–62.
19. Noll, “Early Digital Art at Bell Telephone Laboratories, Inc.”
20. Noll, A. Michael. “The Howard Wise Gallery Show of Computer-Generated Pictures (1965): A 50th-Anniversary Memoir.” Leonardo, vol. 49, no. 3, 2016, pp. 232–39.
21. Noll, “The Howard Wise Gallery Show of Computer-Generated Pictures (1965).”
22. Noll, A. Michael. “Computer Art Copyright.” Engineering and Technology History Wiki, 26 Aug. 2023.
23. Noll, “Computer Art Copyright.”
24. Noll, “The Howard Wise Gallery Show of Computer-Generated Pictures (1965).”
25. Noll, A. Michael. “Flashback: The Story Behind Nam June Paik’s ‘Digital Experiment’ at Bell Labs.” Asia Society, 30 Oct. 2014.
26. Paik, Nam June. “A-Day Project (Excerpt).” Nam June Paik: Videa ’n’ Videology, 1959–1973, edited by Judson Rosebush, Everson Museum of Art, 1974, unpaginated.
27. Yalkut, Jud. “Art and Technology of Nam June Paik.” Arts Magazine, vol. 42, no. 6, Apr. 1968, pp. 50–51.