You can see the pieces of complexity in the way artists talk about their process. Jasper Johns wrote to himself in his sketchbook: “Take an object, do something to it, do something else to it, repeat” (MoMA 2016). Head down a recursive, noisy path—one of many—and see where it takes you. Francis Bacon (painter) called himself “a medium for accident and chance” (Sylvester 1975, 158) and described the painting process as “really a continuous question of the fight between accident and criticism” (Sylvester 1975, 140)—random combinations and negations.
We can think of the different isms over the decades as levels of dynamic stability. A particular aesthetic takes hold and remains stable until some new style takes over—a yellow daisy pops up. The new ism builds off the old ism—harmonics again.
The artistic process is a way to put forward a prediction—to see if something will work or not, to ask a question. Wilson, in talking about aesthetics and evolution, says “The painting is not an answer. It is a question” (Wilson 2012, 6), which feels right to me.
I created the images between each section with various AI programs using prompts based on complexity and evolution. I took pieces from different images and combined them, cutting parts out, refilling them, adding elements of my own, and repeated the process until the prompts resulted in something I liked. As the various AI programs have improved, it’s become much harder to get anything interesting out of them; they feel averaged down—there seems to be less noise and more predictability. Maybe the earlier models were noisier (had higher temperatures), or maybe they made more distant connections. Now it feels like everything is being averaged, and the average of the Internet is not that interesting.
The paintings in (A1) begin with a similar process, but then I use an archival print of an image on canvas as the starting point and paint over parts of it. At some point, they work for me—I can’t explain why. Good paintings have what we might call aesthetic closure, and that closure is probably different for each of us. Sometimes they work; more often they don’t.
It seems entirely possible to me that aesthetics fits into the architecture of complexity, as outlined. And in doing so, we’d add one more thing to what Nagel was after.
“Let us set beauty as equal to such a measure of life. Then it could follow that beauty also is associated with lower entropy … perhaps we have always known this” (Lovelock 2016, 135).