What I learnt about Digital / Generative Art from Grayson Perry

The other day I went to Grayson Perry’s “Who are you” exhibition at the National Portrait Gallery. At this point I think he — along with his alterego Claire — is one of my favourite living artists. I do love the guy. In fact he’s so on the ball that it’s almost depressing.

The accompanying Channel 4 program “Who are you” is essential viewing as context for the exhibition. In fact I think it’s essential viewing in general for any person, especially any artist, especially any ‘digital’ / ‘generative’ artist. Even though Perry never mentions anything to do with ‘digital’ or ‘generative’ art, I think there are many things he does and says which are incredibly important and relevant to us.

(Side-note: I really hope this series gets good ratings, and Ch4 — and perhaps even other channels — decide to do more of them, opening up a whole new genre of TV program. Even a new financial model for artists and commissioning art. Wouldn't it be amazing if we had more of these and less of the X-Factor crap.)

Grayson Perry is also the first visual artist to deliver the Reith lectures, and they are definitely worth listening to (article, direct link). Again I think it’s especially relevant to anyone involved in any way with digital or generative art.

I personally find it very refreshing to see a champion of craft within the contemporary art world. “I don’t want to see something I could think up in the bath and phone in”, she expresses (for this is Claire, Grayson’s alter ego talking) in this short interview on art vs craft.

I find that this is something lacking, not only in the contemporary art world, but in ‘digital’ as well in a way. Obviously digital art involves an incredible amount of craft. As an artist working with custom software as my primary tool, I'm very much aware of this, and can appreciate the insane amount of craft going into a piece of software. However there is often an essence of something lacking in the output. A lot of time invested yes, but perhaps not always where it’s needed.

I'm not going to invoke the “But where’s the artists hand in digital art?” argument. That’s mostly an ignorant point-of-view coming from those who fail to understand the relationship between the artist and the software/hardware/generative system. The hand of the artist is of course very much in the work. If you can look at a piece of digital / generative work and say “that’s a so-and-so” — which is quite often the case — then the hand is clearly there. However, sometimes it may not be very obvious in the traditional aesthetics (colours, composition etc.) of the work, but instead in the movement, or in the behaviour, the concept, the visualization, the sonification, the translation from input to output. To paraphrase (and completely re-contextualize) Sam Harris, to an untrained eye this might be as subtle and unnoticeable as the difference between a malbec and a merlot is to me. But to someone who has spent considerable time sampling many different flavours, the brain allocates larger real estate to the different palettes. So I dismiss the “lack of artists hand” argument as mainly the short-comings of an ignoramus to digital technologies.

However it does refer to some valid issues, especially in the generative / procedural domain.

In the ‘generative’ world we often boast about how much detail we can have. We can fill trillions of pixels and hours of frames, with unique little images, algorithmically derived patterns, shapes and movements, never repeating, always different and unique. Something that could never be done by hand, or even envisaged by a human. We often say this with pride, as a unique selling point of generative, procedural, computational art. But what benefit does that really bring? Is it truly unique detail that we generate? They may be aesthetically unique, but isn't that a bit superficial if it’s all semantically or statistically homogeneous. All pine trees are different, aesthetically; but semantically and statistically they’re all the same, they’re all pine trees! And humans are very good at differentiating that. Seeing a thousand different looking pine trees just isn't that interesting, even if they are aesthetically slightly different.

Looking at Grayson Perry’s work at the National Portrait Gallery really drove this message home for me. All the works are so incredibly intricate and detailed. Every square cm has something really unique, something semantically different. Small little details from the imagination of a human, thinking of its relevance to the over-arching concept, and placed specifically in that location. Every square cm has had thought put into it. A small little detail which on its own may not mean much, but in context of the bigger piece fits into the story like a jigsaw puzzle. So as a viewer I end up spending a lot of time in front of each work, literally scanning the entire surface. Even scanning the same areas but in different directions, and unravelling a story. Speaking to other people about it, to see what they saw, how they connected the details.

Unfortunately algorithms are just not smart enough to create this kind of rich content yet. I don’t think it’s an inherent property of algorithmic art, quite the contrary, I think it’s a fascinating challenge. Instead I think that algorithmic artists just aren't trying to do this. Perhaps many of us are disillusioned by the possibilities that procedural approaches bring. That we can fill giant canvases, or hours of video with patterns that are never-repeating and unique on a superficial level. Or perhaps many of us just don’t care, we like creating forests of different and unique pine-trees, instead of a story taking place in the forest.

I'm not trying to take a moral high ground on this, in that one of these approaches is better than the other. I think it genuinely is an oversight in our field. I do think this is starting to change though.

But the most interesting point I find regarding Grayson Perry is something he said in one of his interviews or lectures — which unfortunately I can’t locate right now so I’ll have to paraphrase. He said something along the lines of “I'm heralded as this poster boy of hand-made crafts, because I make pots, and tapestries. But often I’ll just use Photoshop and a wacom pad to make my designs. I then send them off to be digitally woven on a giant computer controlled loom.

I find this to be a rather important critique on digital arts. Because the digital component of our work is often the foremost component. If it did have a stronger component, like say Perry’s tapestries, then it would be called just ‘art’ — even though it was made by a giant computer controlled loom.

Maybe this is why most ‘digital’ art is called ‘digital’ art and not just ‘art’.