Ruminations on Generative Art

Stochastic process, emergent form

2009-03-05T07:02:00.000-08:00

Here is my favorite definition of stochastic (which is one of my favorite word-tools). The definition is from Caldwell, Robert (2001): Excellence in Singing: Multilevel Learning and Multilevel Teaching. Redmond, WA: Caldwell Publishing.

"A stochastic process has three basic parts:

a random component -- a set of unpredictable elements
a nonrandom component -- a mechanism, sometimes called a filter or bias, that selects from the random elements
a unique result from the nonrandom filter interacting with the random elements

An example: Madelyn, one of our students, related her son's purchase of his first car -- a blue Subaru station wagon -- to stochastic processes. After her son bought the car, Madelyne spontaneously began noticing blue Subaru station wagons everywhere. They seemed to just pop out of the stream of cars whizzing by on the freeway. Notice the stochastic process: her attention (the filter, resulting from her son's purchase) sifted through the cars rolling down the freeway (the random elements) to produce the result (her distinguishing blue Subaru station wagons)."

I like the stochastic tool for a bunch of reasons. First, it captures a complex combination of two normally-separate concepts (to wit 'random' and 'nonrandom') without hyphenation! (This kind of brings me back full-circle to one of my firsts posts on this blog, back in January, called Hyphenated Epistemologies, which was kind of a rant about hyphenated-terms.)

I also like stochastic because I think it provides a useful tool to build further understanding of generative art as we've come to experience it this term. We have in some instances used 'rules' as nonrandom filters, and in other instances the nonrandom filtration has come from our own interests and personalities. Whether as rules, personality features, system parameters, or initial starting-points, nonrandom filtration has been sticky; it has been a consistent, perhaps even insistent, part of everything we have seen and done I would say.

Randomness has also been sticky (and sometimes even stinky :) I mean we had bags of rotting meat hanging from a bridge and in the hall, y'all! We also had a paint-ball mechanism, a tube and a quarter, a mechanical bug, a bunch of film stock hanging from a reel, a Freddy Kruger windchime-set, water, water, everywhere, a burger eating festival, etc. No matter what we did in terms of rules or other forms of nonrandomness, randomness made its presence known. It reared its ugly head. It headed its ugly rear. It hung around, it seeped in under the crack in the door, it wafted in the wind, it got our attention, oh yes it did.

But, even as compelling as the above two were and are, its the third piece of the stochastic trio that is my topic here. Caldwell describes it as the "unique result from the nonrandom filter interacting with the random elements."

So, what have all of our ruminations, actions, discussions, preoccupations, deconstructions, howlings-at-the-moon, etc. added up to? What has been the unique result? Has anything (in chaos-theory terms) emerged? Is there a strange attractor in the house? I think so.

It seems to me a unique result has, indeed, emerged. In this regard, the day that exploded like a seedpod in the bright hot sun was the one wherein we divided into groups (based on text, sound, video, and image), dispersed, worked (on our own and together), reconvened, and ultimately represented what we did (as a 'monsta-mashup' but also in synapses and other representational forms). By engaging in the stochastic process that was and is this class we arrived at a unique form of generative art that moved through distinct phases of 1) recapitulating the avant garde tradition, 2) exploring present variants of gen-art, and finally 3) producing a unique form resulting from the interactions of a particular nonrandom filter and particular random elements.

It also occurs to me that our filter re-constituted, re-discovered and possibly re-energized and updated the art of the production --a form or genre that is at the very heart of our beloved institution. Could this be possible? Have we discovered the ground beneath our feet?

Once upon a time, way back when, when I was an undergraduate at U.C. Santa Barbara, I recall standing out on the funky patio of my apartment building in the ultra-funky student ghetto called Isla Vista, at the end of the day, with a roomate and a friend of his, grilling chicken on the barbie and chatting. (I have a feeling you can relate to such scenes and conversations.)

Someone looked up, I guess, noticed the moon or something, and made a comment about outer space. I took no notice of this, but then my roomate started a rather long-ish ramble that boiled down to the idea that it had always surprised him to hear people talk about outer space as if it were 'out there' when in fact the earth itself was in outer space, boogying along in it at quite a rapid clip thank you very much, etc., and so forth. This comment was sticky for me (obviously since I am writing about it many years later, but also because it was I think deeply and beautifully insightful).

It strikes me as applicable to us in the present, for example. It's possible that by exploring the outer-edges of generative aesthetics we have re-discovered, re-defined, re-energized, and re-awakened dormant possibilities of great beauty within the production as an artform. Might we further distill our principles, hatch a new action, secure Watson Hall or some other campus space as a venue, and do something else? We could.

Certainly we could apply the principles we arrived at, the filtration system we constructed, to individually produced artworks in any medium, but part of my point here is that the strange attractor in our specific case, the unique fingerprint of generativity we created, reflects something like the structure of performance and/or the production. Dean has been calling attention to the performative aspects of our work for at least a few weeks.

In general, though, to again abstract from what we created for possible use in other contexts, the filter we constructed looks something like this: 1) Establish a starting point or initial condition by gathering disparate datapoints of a particular media-type or types (i.e., the monsta-mashup day was, arguably, keyed by gathering text-media datapoints simultaneously with gatherings of image, audio, and video datapoints. The final project is being keyed by audio-media datapoints in a kind of solo performance). 2) Gather parallel datasets that represent the other main media-types (if this was not done simultaneously in the start-up phase). 3) Assemble media-specific datasets into proto-products. 4) Combine proto-products into an overall production.

Any such production will of course reflect its process: it will include and/or frame further interaction of random elements and nonrandom filtration. (Hmm. Filtration. Maybe that's why the watery/liquid theme?) Okay, rambling now, time to ramble on.

Genealogy of Generative Art

2009-02-15T17:42:00.000-08:00

Generativity and creativity are for me synonymous, and both terms ultimately resolve to entering into dialogue with the world. I tend to associate generative, creative, dialogic methodology with Leonardo, who according to legend looked closely at things like water-stained plaster walls and related sorts of phenomena to see what kinds of forms or images might exist-in and/or emerge-from them. In the terms I'm setting up here, he entered into dialogue with such phenomena, and these instances constitute creative acts that might (or might not) have been redoubled or iterated by him into artworks (or passages thereof) at later points in time.

In drawings and preparatory works Leonardo would apparently create his own equivalents of water-stained walls by building up rich and varied loams of texture and pattern as a matrix or ground from which an image or figure would emerge. (The Leonardo drawing below illustrates this.) One way or another, it seems to me Leonardo established basic conditions of dialogue --between stained walls and emergent forms, between textured and patterned grounds and emergent figures.

For me this entering into dialogue is the essence of generativity or creativity, and it is independent of materials or profession. Leonardo illustrates this by working, as he did, with a variety of materials and for a variety of purposes (scientific and technological as well as artistic). But children illustrate it just as well when they enter into dialogue with and see shapes in clouds. Verbal conversations at any age routinely generate emergent figures such as insights or understandings. In terms of the GenArt course we've established a ground of industrial materials and social interactions by entering into dialogue with these materials and formations, and out of this ground have emerged forms or figures. Different materials. Different purposes. Different grounds. Same process.

I tend to call the process entering into dialogue but it could just as well be called immersion. If you click on the thumbnail picture of text below you will see an enlarged, readable version of text by an author who is comparing the aesthetic views of a particular philosopher with Leonardo's 'brainstorming' drawings in terms of "...an immersion into the creative mix" .

When I was immersed in the activity of making paintings I would build up many layers of paint and line by repeatedly turning a canvas 90 degrees, adding layers of color and shape, and articulating possibly-emergent forms or figures with line. Eventually I began to experience the same dynamic in different contexts (dialogic writing is one such context for me, teaching is another).

So for me the creative or generative process involves entering into dialogic or immersive relations that, in turn, establish creative ground.

The Role of Random Language

2009-02-15T16:27:00.001-08:00

I think the most surprising aspect of the recently-enacted GenArt MonstaMashUp piece was the role played by the random spoken-text.

It struck me that the spoken-text functioned sort of like a reassuring set of instructions, something like "It's okay to not understand exactly what is going on in this movie, just enjoy the mix of images, sounds, and intersections." In other words, the random text seemed important to me as a way to dial-in a particular reception of the piece, without violating the piece's integrity.

Further, the constancy of the spoken-text made its reassuring effect almost trance-like, sort of like the effect of those messages that get repeatedly broadcast in airports --except in the GenArt piece the message was much more reassuring to me than being warned against leaving my baggage unattended. (Although, if we switch contexts to the psychological, I guess we all have psychological baggage that warrants constant reminding to not leave unattended.)

I'm not sure what else to make of the reassuring effect the random spoken-text had on me, other than I might think of using random language in other mashup pieces to try to induce in others the same sort of cognitive relaxation I experienced in the Monsta piece. I imagine the effect of the spoken-text may point to the primacy of language among human sign systems, but I don't really know enough about linguistics or semiotics to even hazard a guess.

Paths to Interminacy

2009-01-18T09:46:00.000-08:00

The re-awakening of interest in analog processes via robotics is fascinating to me for a particular reason. Computer programming is, I would say, notoriously or even fiendishly deterministic and inflexible. Yet (as the chaos theory video clip and the Brian Eno / Will Wright video demonstrate) programming can be used to produce indeterminacy/generativity if the programmer keeps the objects and interaction-rules simple and provides a way for the interactivity to build on itself (i.e., via simple iteration/repetition).

The robotics example is the same in principle. Each bug features a collection of simple intra-bug elements (in other words the actions it can perform --e.g., walking, perhaps simple-sensing, perhaps turning etc.). A bug's mobility provides the iteration: it just keeps trying to do whatever it is capable of doing.

Setting multiple bugs let loose in an environment ramps the complexity up (because the sum of types of interactions multiplies) but in a self-similar way the principle is the same: now what you have is a collection of simple bugs, each of which is simple unto itself, interacting simply with an environment that now includes each other (many more possible sensing opportunities, etc.).

In the computational venue or context, the critters and the environment (for example the game environment in the Wright/Eno video) are simply virtual. They are created with language/code processes, tools, and materials. The bugs are created with electronic processes, tools, and materials and move in the f2f environment.

In the biological setting, chemistry anyone? I am out of my league or knowledge-zone here, but talking about simple elements in combination and interaction producing complex results sounds a bit like the periodic table of the elements/atoms, molecules, and chemical reactions.

Can generativity be understood as learning in the key of life?

Media and Mediatedness

2009-01-15T07:40:00.000-08:00

The topics of non-dualism and distributed intelligence connect to media and mediatedness. For example to think about human nature in non-dualistic terms is to think about the various ways that human beings are mediated. Being mediated simply means needing externals to survive. For example we are (as social critters) mediated by others. We need others to survive. We are (as biological critters) mediated by the natural world. We need it to survive (we have to eat and drink of the natural world regularly --or else!). We are also technologically mediated. Toolmaking / tool-using is key to our survival. Mediated bears an interesting relationship to media in this regard. When we worry about the role of "the media" (in politics for example, or in socialization of young people) we are in effect recognizing that we are semi-permeable and 'mediated' by what we see and read, and by whatever channels it comes through. We are what we eat in more ways than one.

With regard to human beings being mediated by tools or technologies, several writers come to mind. Bruce Mazlish is a historian who writes about the central and very intimate role played by toolmaking and tool-using in the development of human beings over the long haul of our history. He cites physical-anthropologists who posit the idea that the upright carriage of human beings may have evolved in direct relationship to tool use (e.g., an upright walker has both hands free to work a toolset, etc.). The Mazlish book that talks about this is called The Fourth Discontinuity: The Co-Evolution of Humans and Machines http://yalepress.yale.edu/yupbooks/book.asp?isbn=9780300065121

Marshall McLuhan similarly posited that all technologies simply extend the human being (clothing extends the skin, hammmering devices extend the hand, computers extend the nervous system, etc.). McLuhan's classic text is Understanding Media: The Extensions of Man. http://en.wikipedia.org/wiki/Understanding_Media

Donna Haraway is a contemporary writer who made 'cyborg theory' famous. (A cyborg is a critter that is a hybrid of human and machine.) Her argument, similar to Mazlish's, is that humans are pretty much cyborgs in their 'natural' state. Her touchstone essay is called A Cyborg Manifesto: Science, Technology, and Socialist-Feminism in the Late Twentieth Century http://www.stanford.edu/dept/HPS/Haraway/CyborgManifesto.html

Sherry Turkle is a writer and researcher who has studied how kids who hang out around computers sometimes fashion conceptual tools of their own about what is 'alive.' The short version is that they create categories that defy the binaries 'alive' and 'not alive' --things can be alive in surprising ways, as they see it. So in some ways Sherry Turkle's work brings a practical and very concrete dimension to the theoretical work of Haraway and others: namely, she documents that kids internalize technologies into the inner workings of their thought processes. One of Turkle's best-known books is Life on the Screen: Identity in the Age of the Internet http://www.amazon.com/Life-Screen-Identity-Age-Internet/dp/0684833484

Finally, James Zull is a writer and researcher who writes about the physical dimensions of thought as the literal building of synaptic connections. Thought is material, according to contemporary cognitive science, and this allows us to extrapolate and consider certain thought-formations (aka ideas) as conceptual tools. His touchstone book is called The Art of Changing the Brain http://books.google.com/books?id=mby-SNRiglgC&dq=james+zull+the+art+of+changing+the+brain&printsec=frontcover&source=bn&hl=en&sa=X&oi=book_result&resnum=4&ct=result.

Mediatedness and non-duality imply a world in which everthing (intelligence, creativity, the self itself) is distributed. If we take this seriously, do we need to rethink and retool our artistic tools (including our conceptual tools, our processes, and our techniques). I guess 'Generative Art' is a tool for exploring this possibility.

Hyphenated Epistemologies

2009-01-10T16:39:00.000-08:00

In this post I venture into tinkering with the Western cultural machine.

Western culture seems to me to be evenly divided between dualistic and hyphenated (non-dualistic) epistemologies. On one side of the coin-of-the-cultural-realm are intellectual and spiritual epistemologies: both of these ways of knowing seem invested in maintaining dualistic structures (mind/body, immaterial/material, etc.). The more Eastern (or non-dualistic or hyphenated) side of the Western coin features the aesthetic and scientific epistemologies. Both of these ways of knowing combine and intertwine physical and mental, material and immaterial, in various yet identifiable ways.

The Western cultural machine thus seems to me to strike a balance: two dualistic ways of knowing, two non-dualistic ways of knowing. Is there a pecking order? I think so, but I'll get to that part later. In part I explored the pecking order just a bit in my earlier post on 'english' as a toolset, in which I argued that the language itself tends to preclude non-dualism from gaining anything like an upper hand or primary role in the epistemological barnyard. But I'll touch on a slightly different strand of the same idea later in this post.

For now, I'll start with the idea that there may be a certain kind of 'doubling' or intensifying of non-dualism that occurs when art hooks up with science (via technology). This possibly creates a kind of overbalance, too much for the dualistic side of Western culture to bear without triggering push-backs. Labels are one way to push back: non-dualist art sometimes gets called avant garde, which is a kind of compliment but also a kind of locked box.

Hyphenated terms in the intellectual realm may similarly threaten to upset the cultural balance by introducing non-dualism into traditionally dualistic epistemological terrain. Maybe this is why such terms are forced, when they do emerge, to bear the tell-tale sign-of-the-hyphen. Hyphenated names in marriages may raise similar suspicions! Now that I think about this, teaching digital media courses in a liberal-arts arena sometimes gets me to feeling like a hyphenated term myself.

What does the above have to do with generative art? It seems to me what we are exploring is an incredibly provocative and evocative form. Particularly now that computers have entered the picture, it is possible to elide the otherwise obvious mechanical side of industrial-era technology and make tech-infused art which evokes traditional transcendent resonances and intellectual appeal --thus satisfying the needs of the dualist side of the Western cultural machine while also furthering the advance, so to speak, of non-dualism. Because of this, we may be in the unique position of participating in the encoding of so-called 'new-media' into the cultural mainstream. In some ways we may be recapitulating history by starting with generative sketches that employ, for the most part, highly visible industrial-era media (and, going back even further in history, to nature-powered media). (Which reminds me to make a note that I might want to discuss creating a passel of solar-powered mini-sculptures that could be planted, so to speak, strategically around campus).

There are organizations popping up that are dedicated to an overall convergence of epistemologies. This is interesting to me because it represents an attempt to include all epistemes in a merged mode, not a balanced mode --which highlights the founding metaphor of 'balance' in our cultural machine; 'merge' is not the main mode, in other words. One such organization is called HASTAC (the acronym stands for Humanities, Art, Science, and Technology Advanced Collaboratory. The only epistemology HASTAC leaves out is the spiritual one (the organization captures scientific, aesthetic, and intellectual epistemologies only). The grand-daddy of electric media studies, Marshall McLuhan, included them all in his analysis, synthesis, prognostications, and creative ramblings.

A link to the HASTAC website. A link to an article in wired magazine (in reference to the spiritual) about the influence of Teilhard de Chardin on some prominent media-convergence folks including McLuhan. A link to an article on technoshamanism (I haven't read this article closely but it looks like it could be mined for some of its links and resources).

Bottom line? My hunch is that the Western cultural machine is designed to preserve, above all else, transcendental dualism. I think the genius of cultural machines is that they scale up or down very easily, and morph freely to preserve their operational imperatives. For example if non-dualism emerges here and there within the cultural landscape, it will be naturally counterbalanced in myriad ways to preserve dualism's ultimate say and way. Culture seems sort of like an organic-machine, which is sort of like bio-tech, which is sort of like where this post started, exploring the whys, wherefors, hows, and trajectories of hyphenated terms.

Mashup Words

2009-01-07T08:12:00.000-08:00

Our conversation in class touched on the troubles we sometimes have using the toolset called 'english' because in many instances it lacks words that combine two or more concepts. We resort to hyphenations such as random-order in order to expand our toolset, and sometimes we spend a lot of time crafting these new tools via discussion. Dean mentioned Derrida's introduction of new words. Heidegger did quite a bit of this as well. He favored introducing Greek terms, which (since they were mostly unfamiliar) ended up being equivalent to making brand-new terms.

In one of my professional iterations I thought a lot about education, mostly in relation to media. One of the english tools I found myself tripping over regularly was 'instruction' --defined as 'the act of teaching' or something like that. The problem with this tool is that it leaves out the environment. Often, I found, real change in education would require changing the furniture, the lighting, the building --these kinds of things. Changing the instruction, though, seemed to grab most of the attention.

I'll try to give an example of what I mean: hard, uncomfortable chairs, arguably 'teach' students that heads matter more than butts, minds more than bodies. Arguably this is an important teaching in Western culture; so important that it is embedded in the environment of education, literally as a matter-of-fact. Once this important fact is established, once this lesson has been learned, it becomes possible to entertain high-level critique and theorization of the role of furniture in education but, unless one actually changes the chairs at some point, the original lesson continues to be delivered even as one speaks and theorizes about it. In effect this undermines the critique or theorization.

Yet --and here is the real problem-- focusing on the role of the environment in education can also lead to a lot of nonsense. i.e., one might create an open environment and then isolate groups of students and instruct them in a very closed way.

So, after many trips-and-falls over the english toolset in this regard ('instruction' on one side, 'infrastructure' on the other), I coined the term infrastruction to describe the way the environment actively interacts with instruction in a relational, conversational way.

I think Dean and I are acting as 'infrastructors' in this course. We are enacting the relationship of ideas to people, processes, tools, and environments. We are having some fairly usual professorial discussions with one another about readings and so forth, but we are doing this in the context of placing all of us --along with the media-lab, the hallway, computers (or not), etc.-- into a conversational framework. Infrastruction is one potentially helpful tool for building-out this framework.

In my view, current media dissolves boundaries simply by dispersing the material logic or liquid ontology of electricity. This affect is pervasive, yet even while new tools like infrastruction and mashup sprout up like neon signs (or, as Chuang Tzu would say, "like music from hollowness, like mushrooms from damp") many maintain that the technology itself --the infrastructure-- doesn't really matter. I think this is another case in point. Of course media doesn't matter when taken in isolation, but in relationship to other cultural and social formations (including education) it does matter.

In many ways, the toolset 'english' helps build and maintain a particular and familiar kind of illusion. We can look here. Or we can look there. But --by placing the here and there in close proximity-- the in-between or relational disappears as we look at one or the other opposing state in rapid oscillation.

Find at the top of this post a picture of this view of Western culture. If we are not careful we miss the innermost point of the illusion; namely, the picture is compelling even though it depicts a non-sequitur; faces and faces are completely unrelated. Yet we are still fascinated and fixed by the image. Imagine how well the illusion distracts us from relational thinking when the two oscillating elements are, even in a remote way, related.

Bad Boy Art

2009-01-06T10:11:00.000-08:00

I liked the silences as part of the conversation today. The 'so what' question I raised about the ultimate relevance of our work in gen-art is in part rooted in experience in the visual arts world. More than in other art-specialty areas (I think, anyway) simply tweaking the parameters of a form has qualified as an aesthetic act in the visual arts. In some ways this may qualify as a kind of reverse orthodoxy, in the sense that tweaking boundaries has, in some instances or strands, become a requirement. "Bad Boy Art" is, paradoxically, an identified genre.

I'm most interested in artists who seem to have escaped the gravitational pull of 'tweaking for tweaking's sake' (or 'art for art's sake'). A prime example of this for me is Bill Viola. Early on in his career he was among those who tweaked the parameters of the visual arts by doing video, but it seems to me he has always maintained a focus on having something to say. His content remains aesthetic (rather than intellectual, propositional, or didactic) but his work is content-driven nonetheless. I'd like to show an example in class.