Rambling thoughts on chance operations, mark-making, images, and human performance
This past Saturday I went to see the Casey Reas show, “There’s No Distance” at bitforms gallery in Manhattan (thanks for the tip, Ruby Childs, and the company, Dannie Wei, Jason Toy and Adnan Agha!). The show consisted (it ended Oct 16) of four pieces, a complementary pair of generative works on screens, a similar work projected to fill a wall (the three together part of the Still Life series), an older generative piece of simple black and white line forms and a single sheet of paper with instructions for producing some simple chance-based art works. If you don’t know Casey Reas and his work, he’s perhaps best known as the co-instigator/creator of Processing, an IDE and Java-based language for creative applications.
Reas describes his work as a performance being made in front of the viewer, where the production of the code is like the preparation of a theater piece and the running of the code like an ever-changing performance. It begs the question whether human presence is a requirement of performance. The performer in his works is not a human, but the processor, memory, and other hardware, ending in the display or projected image in the gallery space, the final interface with the audience. We are accustomed to other media such as film and video which are typically prepared in a way very similar to a theater or dance production but which most people would consider playback, rather than performance. The ever-changing nature of Reas’ work, and other generative art, distinguishes it from playback in that it will never play the same way. However, I’d have to say that my experience of viewing the art felt more like experiencing playback than witnessing a performance.
I’m not decided that human performers are an essential aspect of performance—I think they probably are not, but for the moment I’m not going to try to decide. I’m sympathetic to the idea that generative art of this kind could be a performance. The process of writing instructions with enough agents (hmm, that is sounding like a performant almost), interactions, and randomness—then watching as it runs and experiencing a predictably unpredictable output can be delightful and rewarding, the un- or not entirely predictable behavior marking a difference between author and agent. Reas’ generative pieces engaged my attention and were beautiful to watch in their unfolding. Because of their ever-changing nature, I could have watched with interest for quite some time.
It was actually thinking about the piece that was a single piece of paper on a wall, Yes No, which re-engaged me in this question of human performance in computational art. The sheet listed instructions from François Morellet’s Random Distribution of 40,000 Squares Using the Odd and Even Numbers of a Telephone Directory (1960). Someone carrying out the instructions would paint a square in a grid red or blue, after picking a random number, odd or even, from a phone book—red for one, blue for the other. Another written instruction directed a similar operation to make diagonal lines of one or the other angle from a chance binary choice with a coin flip. Reas added code for a simple Processing sketch to “perform” actions to generate 45 or 135 degree diagonal lines across squares in a grid.
At home I made an 8x4 grid of squares on paper and performed the randomization instructions with coin flips. It took about 5 minutes to generate just an 8x4 grid, in part because I had to chase my quarter across the room more than once! I found the process strangely satisfying. I was an automaton performing, very slowly, the random operations I usually assign to a computer processor. The mark-making on paper was perhaps the most satisfying—a partial release of my will to the whims of chance, at the same time using my will to conduct the process and draw a line as I chose on the paper. By using minimal decision making power in making a mark I noticed the pleasure of using a good pen on good paper, with no rush to get down an idea and no attachment to producing something “good.” I did feel like I was conducting a performance, but with only myself as an audience. The output on paper was secondary, though part of the satisfaction.
I “performed” this a second time, with a second grid of 8x4 squares, but made the grid in a light blue pencil and marks with the same thick black marker as before. I found this one much more satisfying, because this time the chance-determined marks took visual prominence over my hastily made grid. The marks take on the feel of pieces of some written character set. They seem to have some inscrutable meaning. I think this effect accentuates and elevates the part of hand-drawing in the mathematically determined outcome. It feels like a tiny sketch-piece more about the process of making marks on paper than about the randomness at all.
I’ve become interested in the physical process of drawing, making lines and marks on a surface—paper, or could be other surfaces, as well, but using the hand and a mark-making device. I’m inspired by cartoonist and teacher Lynda Barry and her thoughts on the importance of physical movement of the hand in generating images. I’ll try to do justice to her thoughts on images. In much of her work, she asks, “What is an image?” An image, as opposed to a picture. A picture is a representation of an image, but not the image itself. An image is more a thing of feeling and spirit and exists among but beyond words and pictures. Subjective things, images have a life of their own, and can haunt and evade a person like ghosts. An image can come from a picture to a person but is not contained by that picture. An image cannot be produced entirely from mental activity, but comes through the body as well, through the hand. Barry emphasizes the physical activity of the arm and hand in drawing, drawing out the image by the motion of the body. Letting go of mental control enough to let the image move out onto the paper.
What is the role of the human body in computational art? Different media will impose or suggest their own character on any kind of work. Paint will flow, ink will bleed in ways not always predictable and the physical process of working with them is more than mental control but an interaction between the mind of the artist, the muscles of the body, and properties of the tools and the medium. The chance effects of the imperfect interaction suggest entirely new images and ways of working.
OK, all of this may be a bit abstract, but I feel it is relevant to my experience performing the instructions to make my simple marks on grids. I think that the physical activity I did in drawing are what generated for me this image that my lines were some kind of character system. I think the act of actually flipping the coin brought up in me other images, associations between these simple instructions and concepts of play, risk, gambling, and “morality” that would not have come up had I not physically performed the flips. I liked the fact that my coin flipping seemed to not produce very good randomization — eight tails in a row—and made me feel more personally connected to the otherwise mechanical instructions! I actually felt a bit of—melancholy, I’ll call it—in not being able to stay up later at night flipping the coin more, making a larger grid, and staying with the process. It was a different feeling than other times I’ve wanted to stay up coding. A strange feeling to notice. Maybe I like being an automaton.
But what if the instructions were to randomize not between 0 and 1, but between 37 and 789, or 0 and 3,000,000? I could not as easily perform these operations for the randomization process (though it could be done). I definitely couldn’t perform these at the speed with which a computer processor provides randomization used in much computational, generative art. What if I were to try, or a group of people together were to try? I had the idea to replace the random() function in a program, or any program, with a group of performers. Besides being much slower, what else would change? How would the experience of the program or artwork using it, be different? What kind of images would come up for the human performers?
I also thought another kind of replacement for random()—a mechanical coin flipper and camera system that would report results to the program. Generating any kind of randomization from a larger set would require a lot of binary flips and would create a cacophony. Maybe the machines would require feeding or maintenance by human performers as well.
Touching back on the associations that came up for me between the coin flips and gambling, trading, morality, I thought it would be interesting to do this kind of performance of a high-speed financial trading system!
Or, can real time measurement of the human body itself be used to produced randomness? What kind of human intervention in that measurement of the body would be interesting, would provoke the greatest amount of image ghosts to come forward in the minds of the performers and audience?
Seems like an interesting area to keep exploring. We’ll see if I implement some projects from these ideas!
FOOTNOTE: A Python program (in process) to replace random.randrange() with instructions for human actors and computerized prompts.
