Generative Art

Dynamic Artistic Practice Systems

To produce “dynamic artistic practice” AARON would have to create AARONs. It would have to do what Harold Cohen does; set immediate and long term goals, create systems, evaluate results, reflect on its achievements and work this meta-knowledge into its goals and systems.

This sounds like the sort of goalpost-moving that has always plagued AI. If a computer can play chess then chess playing must not be a sign of intelligence… But “dynamic artistic practice” would be relatively simple to achieve given a system of axiomatic graphical microdomains and a Lisp-style function composition system.

The microdomain approach is the one that Douglas Hofstadter and FARG took: find a simple task that is representative of an aspect of creative behaviour and then analyze it in depth.

AARON, like any painter, just pushes pigments around in 2D. It's a historically sufficient aesthetic domain in which to create and evaluate constraints or axioms.

It may even generate historically novel techical practice, Margaret Boden's “h-creativity” from The Creative Mind. But this wouldn't be necessary to satisfy the requirements of “dynamic artistic production”, as Damien Hirst's appropriations show.

So I'm not making a ridiculous demand of an AARON-like system. This behaviour could be shown by a fairly simple system (like Copycat), although to be aesthetically as well as conceptually satisfying it would need to be more complex.

(From an answer to a comment by Yaxu below.)

Technorati Tags: ,

3 replies on “Dynamic Artistic Practice Systems”

Of course he doesn’t have to create a new painter, he could just modify himself to become a new painter, by developing his style. Cohen has decided to create a formal machine that embodies his style, and develop his style by modifying the machine, but perhaps that amounts to much the same thing.
I don’t find the distinction between H- and P-creativity useful. That a creative act is somehow of a different class just because it has been done before seems rather arbitrary.
Wiggins [1] finds clearer distinction from Boden’s work. “Exploratory creativity” is what AARON and draw-something do, they follow fixed rules exploring a creative space looking for things. They modify neither search method or search space. “T-transformational” creativity is where a system modifies its search method. “R-transformational creativity” is where it changes the definition of the creative space it’s working within.
I think these distinctions help make clear the kind of introspection and self-modification required to make an AARON which we would call creative.
I’m confused though about what a ‘goal’ is. It seems that often goals are decided upon after a creative act, rather than before it 🙂
[1] Sorry, paywalled –

Modifying oneself creates a new painter. 😉 I agree that a system could be self-modifying. I just personally find it easier to separate out the modifying and modified modules. I don’t think the difference is a major one.
Thank you for the Wiggins reference. I find Boden’s distinction useful when defending machine creativity; most critics want h-creativity but p-creativity is amazing enough.
As for goals you are absolutely right. I personally rarely know what I am doing with a project until after I have finished it. Project-management type goals are not part of the low-level mechanics of creativity. But some sort of goals are (mix the colours, apply the colours, keep going until it’s interesting), and long-term goals may be emergent properties of or calculated from more immediate goals.
AARON could be described as goal directed. It has a hierarchy of processes that work until it has done “enough”, for some self-determined value of enough. To achieve this it has to compose, draw, colour, from abstract symbols down to moment-to-moment drawing goals.
Harold Cohen talked about a self-modifying AARON in the 1980s. I’m sure he will have had good artistic reasons for not pursuing it. I think he found with the AARON that used abstract shapes that there were only so many interesting abstract shape schemas and that adding more didn’t make the system actually demonstrate more capabilities. So possibly a self-modifying AARON wouldn’t demonstrate anything more *artistically* interesting than a Harold Cohen-modified AARON.

Sorry Rob, I only just found your reply to mine.
All good points but I thought I’d suggest that the goal of creativity isn’t just to produce a finished work, but to change oneself in the process. Deciding whether something new you’ve made is good or not is in itself an act of self-modification. If you find a way to judge something new as either good or bad then perhaps you’ve made yourself better in the process.
According to Wiggins, if self-modifying AARON was stuck making abstract shape schemata, but eventually followed its rules to generate something that did not fit its idea of a valid shape but was still aesthetically valued, then it could transform itself to include the new shape in its rules defining all possible shapes.

Comments are closed.