Procedural Thriller

by Misha Lepetic

"The edge of the unnavigable,
the region of no information.
~ Pynchon, Bleeding Edge

ALast month I ended with a question: If art is partly about eliciting a diversity of reactions that come from a shared experience of a single object, gesture or construct, then how do the potential meanings of art change when reproduction is made deliberately impossible? As we'll see, recent advances in software allow for custom (or 'procedural') generation of worlds and narratives that are not only unique to a single individual, but will also never be repeated, even for that person. Nevertheless, this approach is not entirely without precedent: precursors can be found, as always, in the work of artists going back at least as far as the 1950s.

For me, the example that immediately comes to mind is John Cage's 'Imaginary Landscape No. 4'. An early experiment in removing the author from the piece, Cage's score is for 12 radios. Each radio is operated by two performers, one charged with turning the frequency dial and the other attending to volume and timbre. The score provides instructions for duration and frequency, and the overall effect intends to liberate listeners from the tyranny of the composer's intention. The Guardian's Robert Worby gives a sense of what the 1951 premier of the piece might have sounded like:

What the audience heard was the gentle crackle and hiss of radio static as the players glided between stations. Occasionally there was a burst of speech, a snatch of music, the reassuring flurry of violins playing a sweet, late-night melody. The audience giggled, coughed, and applauded wildly when a recognisable fragment of Mozart blasted out.

This last bit is interesting: on the one hand, giggling may imply delight when a surprising moment or juxtaposition occurs. But on the other, a sense of congratulation (or perhaps relief) when the performers stumble across 'real music'. But there is no one to congratulate – neither the composer nor the performers could follow the score and game the system to create this moment. It's also clear that the reactions of the audience constitute a further part of the piece itself. Cage created the space where chance drove the performance, and this opened the possibility for more sound events to further accentuate the uniqeness of that particular event. In effect, the audience itself takes up the role of performer.

Conceptualists such as Sol LeWitt also pursued the instruction-based generation of art, most notably with his 'wall art' explorations. Like Cage, one of the reasons behind his commitment to this approach was the notion that unique artworks could be created not only at any time or place, but by anyone. When LeWitt visited the students at the School of the Museum of Fine Arts in Boston in 1971, he traded in giving a lecture to the students for creating a piece with them, the entirety of which is represented with these instructions: "On a wall surface, any continuous stretch of wall, using a hard pencil, place fifty points at random. The points should be evenly distributed over the area of the wall. All of the points should be connected by straight lines." If you knew how to follow simple instructions, you could make art: it was as easy as connecting the dots.

LeWitt's legacy has carried forward, too. Last year, Fridman Gallery in New York hosted choreographer and dancer Abigail Levine in a 5-day ‘performance' that entailed her drawing the 3,744 lines that constitute LeWitt's 'Wall Drawing #56' (1970). What was once the province of curators and art assistants – making the work out of sight of the public in anticipation of showing it only in its finished state – was now rendered open and accessible to all. To further emphasize the process of making, 'sound designer Dave Ruder captured the sound of each line, the pencil's mark amplified via contact microphones and played back into the room. As the drawing was performed, a sonic archive accumulated around it in a multi-channel sound installation in speakers spread throughout the space.' In other words, always find a way to make more out of what is already being made.


BLet's take stock of what has been jiggled loose by these approaches to making art. We no longer need a piece to carry a closely defined intention. Nor is there such a clear line between who is making the art, and who is witnessing the act of its being made. Finally, anyone can make this art, as long as they have the correct instructions, whether it is a fairly simple musical score or an even simpler paragraph of prose. In all cases it's still art (I suppose because the artist says so) but the artist's participation has been deliberately and dramatically curtailed, although not completely removed.

Today we might well substitute 'algorithms' for 'instructions'. Decades of accelerating improvements in hardware and software have created a fertile ground in which we can now plant and cultivate procedurally generated schemes that were unimaginable to the people working on technology's bleeding edge even thirty years ago. As a result, our machines are now the ones creating the space where our stories happen.

When it was announced a few years ago, 'No Man's Sky' promised to complete a radical refiguring of what it meant to play a video game. The vast majority of games are goal-based: find the treasure, save the princess, destroy your enemy. These belong to the category of 'finite games': like chess or Monopoly, the point of the game is to win, at which point the game is over. But as software developed and game designers were able to create more immersive scenarios, the number of games that allowed participants to go 'off-script' increased. So while games like 'Assassin's Creed' and 'Skyrim' still have larger narratives that involve success or failure, there is enough detail built into the gameplay that players could wander off and do quite a lot of things that were totally irrelevant to the task at hand.

In fact, certain players sought out these non-spaces. Writing in Real Life, Robert Minto describes what happened when he realized he could abdicate his character's responsibilities at a crucial moment in an old game called 'Redguard'. The spaces he discovered in this manner became an obsession for him:

After I'd played through the story several times, I started experimenting with other ways to use the game. If you simply jumped off the ship where the final boss battle took place, you found yourself in a wireframe version of the game world where you could pass through walls…This part of the game — the illicit, post-story part — is better than anything that might have preceded it in the name of story. In a world empty of fate, gone slack without a narrative, my character, alone and aimless, has a life for the first time.

'No Man's Sky' takes this trend to its final conclusion. There is no way to 'win' the game, but that doesn't mean that the game lacks a point. It's an 'infinite game', one whose only purpose is to simply keep you playing. The longer you play, the better you get at it, and the more enjoyment you derive from it. Sound familiar? Bernard Suits, in his 1978 meditation The Grasshopper: Games, Life and Utopia, posits the 'lusory attitude', or the suspension of disbelief required to enter into a state of gameplay. We submit to certain rules, promise not to cheat or color outside the lines, in order to access the pleasures that the game offers us. But infinite games partially suspend this suspension. If you can do anything, what would you do?

CIn the case of 'No Man's Sky', the gaming engine creates a fittingly enormous canvas for this sort of infinite play. It is also the (virtual) manifestation of Douglas Adams's definition of space. Writing in The Hithchiker's Guide To The Galaxy, he pokes fun at our inability to imagine cosmic scale: "Space is big. You just won't believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it's a long way down the road to the chemist's, but that's just peanuts to space." To be specific, players can fly their spaceship to something like 18 quintillion worlds.

There are consequences to this enormity, not least of which is the vastly reduced chances of meeting another player. This may not seem like a big deal at first – the worlds are rather exquisitely rendered, and come complete with unique flora and fauna swaying and galloping about. At the same time, all these worlds seem to settle into a sort of consistency over time, an exotic homogeneity that eventually fails to enchant or surprise. So while the game is for all practical purposes infinite, it is nevertheless insufficient.

It's understandable then that players rapidly settled on a key demand for the game's further development: they wanted to meet fellow travelers. In the game's first iteration, there were in fact instances where players found themselves on the same world, but were unable to see one another – a reminder that the boundaries of software are still quite real. A major update to the game now allows players to see one another, but not much else, and the complaints continue. People still want to meet one another, make things and build communities, or maybe just shoot each other if that's more fun. Pesky humans; but whoever said playing the Creator would be easy?


Perhaps it turns out that people don't respond well to unconditional freedom after all. As Minto notes,

The freedom game designers seem to want for me and the freedom I want to seize from them are radically different. One freedom concerns choice. Game designers wager that the more they pack into their worlds, and the larger those worlds become, the more a game's simulation will approximate the freedom of real life. The other freedom concerns autonomy. By exhausting a game's content, I wager that I'll find the pleasurably desolate state in which, whatever I do, I'll be doing it on my own account and for its own sake. The first freedom requires a world built to accommodate it, a sumptuous palette of sanctioned choices; the second freedom depends upon a world it can defy.

The implication here is that people crave drama, tension and even conflict; they must measure themselves against the world, or bear witness to those who seek to do the same. In any event, there must be a narrative that helps to structure the world around us, especially as it might exist in opposition to ourselves. This narrative can be repetitive, but with the right amount of variation its deployment is wholly satisfactory. Fairy tales function in this way, as do procedural thrillers like television's 'Law & Order'. It is the narrative arc that counts. In this regard, the machines may have a better chance of amusing us to death with 'Sheldon County' than with ‘No Man's Sky'.

The brainchild of AI doctoral student James Ryan, 'Sheldon County' is still in development. However, it's still worth discussing, as even the initial results are weirdly compelling. So far as The Verge describes it, the characters and events in Sheldon County are seeded by a random number, completely generated by AI, and are unique to each listener. The transcript is read into audio by a text-to-speech algorithm. In the interest of full disclosure, we are told at the beginning of the first episode that this is "a simulated American county that inhabits your phone. Your county is #1,515,459,035. This is your very own Sheldon County…No one has beheld it before, and no one will behold it again…In this sense, the events recounted in this series are actual, true events. This is non-fiction."

DThis introduction has enough cheek that it was clearly not written by an AI. But the following story has a deeply strange, Beckett-like quality. Our protagonist seems literally drawn out of the darkness: "He was a nothing man, but somehow he had become something, someone. Outside of this place there is a blankness and in that void he was himself a void." But traits begin to accrue ("He values eloquence, sacrifice… tradition"), if not a history ("Although this man has no past, he will have a future"), and then, a name: Jonathan Patience.

Frankly, this sounds pretty good, for the same reason that we are always helpless before the prospect of being told another story. What tests will Jonathan Patience face in his struggle to, hmm, I don't know what? But we know this much: he values eloquence, sacrifice and tradition, and he's heading for a future. Something is bound to happen. I'd certainly sign up for my own private Sheldon County, if only I knew where to do so. My qualms, however, are not about the quality of the narrative, but rather the consequences of what seems to be an endless and intensely private experience, one that is nearly unshareable. As Ryan's own website puts it, "In the days of binge watching and binge listening, [Sheldon County is] an infinite podcast." Bingeing is for amateurs; real media consumers reach for infinity.

If we return to the reframing that artists like John Cage and Sol LeWitt incited with their own procedural experiments, we can see that there is a particular concern for meaning that threads its way throughout. It may be tenuous and a bit improbable, like a well-played round of Jenga, but these artistic practices hold constant the need for people to participate in the creation and appreciation of those practices. In the case of 'No Man's Sky' there is nothing to push against, and in the case of 'Sheldon County' there is no way to introduce the content into the greater social, where explication, interpretation, critique and argument serve to enrich and re-contextualize the initial experience. In neither case is there much 'making' going on, this being handled by the machines, to varying degrees of satisfaction.

Perhaps most limiting, the very nature of these ventures precludes the possibility of the 'real world' intruding on its consumers (even 'Law & Order' had its "ripped from the headlines" shtick). I recall an anecdote told to us by our college music theory professor. In 1963 a student ensemble was performing Cage's 'Imaginary Landscape No. 4', but instead of the usual chance-driven cacophany of voices, music and noise, they came to realize that all the radios were playing the same broadcast, which was relaying news of the assassination of President Kennedy that afternoon. Cage wrote, "I had a goal, that of erasing all will and the very idea of success," but the universe conspired to turn the performance into an incandescent bolt of meaning, searing enough that it was worth recounting to a new generation, more than 25 years later. After everything, there still seems much to be said for the collective experience of art.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone