Jason Salavon: Golem (100,000 abstract paintings)
Jason Salavon’s work has been popping up all over the place lately. This week Infosthetics posted about his Playboy Centerfold piece, with several other blogs picking up the link. At the Generator.x conference Golan Levin showed his Form Study #1 as an example of how generative work could tap into richer conceptual dimensions.
Salavon’s project is a kind of anti-visualization. He hints at profound hidden meaning, but ultimately obscures or ridicules it. This is particularly true of his series of what he calls amalgamations, which includes the Playboy Centerfolds, 76 Blowjobs, 100 Special Moments etc. In these and works like Everything, All at Once the strategy of data averaging is an ironic device, ultimately reducing signal to noise. Not coincidentally, the resulting images tends to be pleasing to the eye, composed of pastel colors and soft shapes. In this way Salavon succeeds at creating visually interesting abstract images, while imbuing them with a suggestive content.
As attractive as the amalgamations are, Salavon’s Golem is a better comment on generative art. Golem is a series of 100 000 abstract paintings, created by software designed by Savalon to “relentlessly generate an infinite variety of such paintings”. This is a conceptual piece disguising itself as a visual work. Salavon claims that Golem “might be said to pass a Turing Test for abstract painting”, but in reality it is an ironic (if not nihilistic) comment on generic abstract painting and the use of software to create infinite variations.
Golem cuts to the heart of an issue meticulously avoided by most generative artists: What is the value of a single image produced by a process that generates infinite series? What constitutes the art object, the singular output or the process as a whole? For artists that are trying to operate in a commercial art reality (as Salavon is), these are dangerous questions, potentially undermining the value of software-based work. Artists like C.E.B. Reas have circumvented this problem by selling software works as unique one-off objects with no editions, or as single large-scale prints presenting snapshots of the work. The difference is that Reas is genuinely interested in the single image, whereas Salavonseems intent on demonstrating the meaningless of the same.