[eu-gene] There must be no generative, procedural or computational art

[Philip Galanter]

On Jan 4, 2012, at 3:54 AM, alex wrote:

> Hi Philip,
> 
> I am talking about artistic/significant intent, not simple variations
> of blobs in a line of potato prints.

Generative artist create pieces that conform to their artistic intent all the time. See Harold Cohen and many many others.

> 
> I think the confusion instead comes from your use of the word
> 'computer'.  Following rules to produce output requires a computer
> (this is a tautology), therefore generative art is a subset of
> computer art.  

Who says generative art has to be limited to rule-based systems?  Are all generative systems rule based? From a computability point of view I suppose it is arguable that any system can be *simulated* or *modeled* computationally (except, of course, they can't...see chaotic systems...), and to that extent they are rule-based.

But really, isn't it a stretch to say...

Oh, Warhol's oxidation (aka piss) paintings?  Sure, that's computation... [1]
Oh, John Cage's I Ching music compositions? Sure, that's computation... [2]
Oh, Hans Haacke's Condensation Cube? Sure, that's computation... [3]
Oh, Daro Montag's use of microbes to create images? Sure, that's computation... [4]
Oh, the Vasulka's use of analog video feedback? Sure, that's computation... [5]


[1] http://www.warholstars.org/aw76p.html
[2] http://en.wikipedia.org/wiki/Music_of_Changes
[3] http://slought.org/content/11208/
[4] http://www.microbialart.com/galleries/daro-montag/
[5] http://www.vasulka.org/Videomasters/MA_index.html

I could go on like this for some time. How are these *not* generative art? Or more to the point, doesn't it seem like a better use of language to call these "generative art" than "computation" or "computer art?"

In fact after this I'm going to finish a piece that uses a thin layer of self-organizing chemicals developed at A&M's Chemical Engineering Dept. to generate ever changing patterns of color.  Are you saying my choices are either (1) to think of it as computation or (2) to not call it generative art?

I think you have it backwards.  It's computer art that is a subset of generative art. (Or more precisely, it is generative computer art that is a subset of generative art).

> Any objections to this should take account of the fact
> that before electronic computers, we humans did computing in our
> heads, on paper, with pebbles, etc and we still do.  I think we can
> consider generative art in more precise terms by acknowledging this
> fact.  I think this is preferable to opening up the definition wider
> and wider until it includes everything.

The danger of widening the definition of generative art to the point where it contains all art is something I've explicitly addressed from the outset. I've made sure that there is a bright line between generative and non-generative art. If the artist doesn't cede control to an external autonomous system, then it's not generative art. On this basis in my "What is Generative Art" paper from way back when I deem Pollock's work to not be generative (his is a practiced skill, not randomization), while finding some of Haacke and much of LeWitt generative.


> 
> To my reading, your definition clearly separates the artist from the
> end result, by intermediate computation.  Your definition does not
> include the notion of computation itself as the artwork.  I'll have to
> read your papers on the subject with an open mind, but if that isn't
> what you meant I don't think it's a very good definition.

Some very careful parsing is called for here. 

* If the element in question is literally "the notion of computation" then I would classify it as perhaps conceptual art, not generative art. 
* If the element in question is a display of code it's more like an installation or an art print.
* If the element in question is unseen code, i.e. the notion of code, then it is more like conceptual art.
* If the element in question an unplugged computer it probably isn't generative art, it's more like sculpture or an installation. 
* If the element in question is an executing computer but there is no tangible result apparent then it's also more like an installation. 
* If the element in question is an executing computer that produces some tangible result it probably is generative art.

> 
> But the way we think about 'computation' in computer science is
> changing, away from the strict mathematical definition of halting
> Turing machines, to interactive processes embedded in human culture.
> I think the way we think about making art with computers is changing
> accordingly, not as definining some rules to produce some end results,
> but as manipulating and interacting with processes as part of a larger
> system.

There is a standing problem in the philosophy of mathematics...are mathematical objects 2nd class entities that are mental abstractions of reality, or are mathematical objects ontologically prior to, i.e. "more real than," the everyday physical objects that seem to comply with mathematical laws?

This is a similar problem. There is a tendency by some to want to see the universe as being the ultimate computer. I'm sure more than a hundred years ago some wanted to see the universe as being the ultimate steam engine. My own take on this is that the universe comes first, and our ideas and abstractions come second.