Thursday 22 September 2011

Consider the Unicorn

The late-night drunk has a new song in his repertoire. I heard it not too long ago, when I was still a student, returning home from a club at 2am. Three others, crossing the road ahead of me, began without warning to belt out Erasure’s camptastic sugar-bomb: ALWAYS! I WANT TO BE WITH YOU! MAKE BELIEVE WITH YOU! We all know who to blame. It’s that bastard unicorn.

How did a tiny two-button flash game end up on the streets of Brighton? Some time ago Rod Humble – EA’s Captain of Sims by day and spare-time indie meddler by night – wrote about the ancient Egyptian game of Senet, whose intricate boards and pieces have been excavated and studied but whose animating principles are lost forever. Yet any future archaeologists who dig up Robot Unicorn Attack will have the opposite problem: they’ll have the code, but they still won’t understand why anyone gave a shit. Although you can peel back its skin and sweep away the sparkles to focus on the underlying rules, trying to pin down its appeal to any mechanical facet seems fruitless.

I only say this because I haven’t seen anyone do it yet. One reviewer at threw up his hands: “I dunno why, but it’s surprisingly fun!” Gamezebo.cotm noted that it “really involves little more than smashing stars and jumping from platform to platform”, judged it “far from the terms of gameplay”, nevertheless wholly recommending it. And Touch Arcade summed up the minor consensus:
“This is a game that is so ridiculous, you can't help but to love it. Between the silly soundtrack, rainbows everywhere, and both dolphins and sparkles popping up all over the screen when you're doing well, it's hard to not smile while playing Robot Unicorn Attack...”
An earlier preview on the same website is (I eventually discovered) the actual source of the claim so gaily quoted on Adult Swim’s website: that the game “creates an experience that is quite honestly too awesome for words”. As the second most over-used word on the internet after ‘epic’, ‘awesome’ is quickly becoming an empty signifier, a perfect stand-in for the game’s absent mechanical appeal. It may, as one comment reads, “give a whole new meaning to the term ‘awesome’,” but the term illuminates nothing about the game - except its mystery.

The question of the game’s appeal is a test case for game critics, specifically for pure ludology – a formalist approach to videogames that demands they be studied not as narrative artefacts but as systems of rules – as games, rather than any offshoot of film or literature. In a gesture made by Raph Koster and Jesper Juul as well as Humble, ludology points out the asymmetric relationship between aesthetics and mechanics: subtract from any game the system of rules which govern interaction with it, and you’re left with a bunch of art assets existing in no meaningful relation to each other. But remove those ‘representational’ components – replace all models and sprites with the red and blue squares from Humble’s The Marriage, swap the sound files for white noise – and you still have a game.

Accordingly, in its stronger forms the ludologist perspective tends to regard aesthetic and narrative elements as incidental to the study of the game, almost accidentally attached to the rules (which are the fun part), and therefore disregard them. In that aforementioned article, Humble wrote that “a game needs nothing else apart from its rules to succeed as a work of art. It can certainly benefit from other elements but it doesn’t need them.” Then there's Soren Johnson's brilliant ‘Theme Is Not Meaning’, which treats the zombie setting of Left 4 Dead (for example) as “simply...a plausible backdrop” for its team dynamics.

Such ‘hard’ or pure ludology has been an enormous benefit to the analysis of videogames, but its logic doesn’t always hold true. Robot Unicorn Attack (of all things) highlights the risks inherent when we disregard a game’s representational elements in order to separate from their chaff the grains of ludological insight. Critics who completely dismiss the aesthetic are throwing away a great big chunk of their toolbox, and blinding themselves to whole octaves of the spectrum. None of RUA’s reviewers seemed willing or able define the game’s (vast) appeal in terms of its (crude) gameplay.

Now, I’m not trying to cast myself as the one who points out the emperor’s penis. Those reviews I quoted earlier aren’t missing anything that I’m revealing. Everybody knows that the emperor is naked and his dick is flapping about in our faces - hell, that’s why we like him. In other words, pure ludology is inadequate to deal with such games, because they are primarily remarkable and generally loved not for their formal systems – the rules of the game – but for the supposedly irrelevant representational elements. They “create an experience” which does not spring only from their programming.

That is not to say the experience has no ludic component whatsoever. After all, you can fly. Since every candy-coloured rocket dash can be chained with a jump, and every jump with another dash, it’s possible to stay almost perpetually aloft by alternating between the game’s (only) two controls. There is a limit, but it resets with the slightest brush of hoof against ground. The impression is of near-infinite flight, and since dashing increases your speed, it is only achieved by racing even further on – one clear mechanical factor in the feeling of momentum and freedom that is the game’s fist-pumping raison d’etre. Another is the continual forward movement, which commits the player to an ongoing plummet that’s both desirable and inevitable. Said lack of agency rejuvenates the age-old Mario mechanic whereby the character jumps higher the longer a button is pressed. Here, speed is applied: as long as you hold the button, you’ll stay up on a forward arc – as if by the power of will! The power of love! – but sink when you let go (helpful for avoiding the all-too-common underhangs). So in fact, the mechanics make no small contribution to what has been identified as ‘awesome’ – but not enough to be the whole story.

Instead, awesome is concentrated in those supposedly irrelevant representational elements. These elements accompany each ludic action, but they are far from incidental, for without them the game would be nothing. The joy, the music, the dolphins, the stars, brighter than suns - ! In bad kids’ movies (and Doctor Who) there’s always a little gleaming sparkly sound to signify that something magical has happened. Here, every jump makes that sound. It’s like a parody of a coin-collecting platformer (or Lego Star Wars), where swooshes and bells that normally signify significant progress are a constant chorus. It tickles your lizard-brain and few players ever want it to stop. Meanwhile, the simple backgrounds are drawn in gentle curves so that, at speed, they appear to roll seamlessly past.

Most essential is that song. It’s well chosen for its specifics: a baseline canter which resembles synthetic hoofbeats, a melody which blasts ever upwards, and the cherishable possibility that dolphins might coincide with the bit where the guy sings “jump into the ocean”. But it’s important that there is a song at all. Playing constantly in the background, absorbing all the glitz and sparkle into a kind of aesthetic totality, its ceaseless onward movement is synonymous with that of the player, and together they rush towards a series of repeated climaxes. That is, of course, unless you fail. There’s nothing worse than a joyful high-score run where you die just before the chorus comes in. So, inciting a natural desire that the crescendos of the song should correspond to those of your play, you start again, attempting to align the one with the other. Lasting long enough to reach the exultant pop finale, which repeats the chorus several times back to back, is its own reward once you’re caught up in the sparkles. The danger to our future archaeologists when they crack open the tomb won’t be an ancient curse or grumpy mummy but a terrifying memetic virus whose hosts lapse periodically into triumphant glossolalia.

From a ludological perspective the song is invisible; it could be swapped for one of Gavin Bryars’ half-hour singing tramploops and there would be no mechanical difference. And yet here we are, screaming it out on the streets of Brighton. The game’s effect on its players simply cannot be divorced from its narrative, representational aspects – and they are the secret of its appeal, however short-lived. I suspect this is also partly true of Canabalt. That game, though better designed and paced than RUA, remains endlessly frustrating, demanding decisions whose outcome can’t be predicted. It is praised as often for its minimalist aesthetics as anything else: the ‘feeling’ of movement, shaking the screen and populating the background with giant silhouettes in parallax smoke.

This is not to say that the most fruitful critique will explain RUA’s allegory about trying to keep up a gay relationship in the modern world, punching through the barriers of prejudice, avoiding the underhangs of infidelity. Nor will it figure out the ‘plot’ of Canabalt, definitively identifying those silhouettes. Instead it is only to remember that ‘abstract and formal systems’ are never directly accessed. Modding aside, players are never in unmediated contact with the rules of the game, only groping their contours in a system of action and reaction that is ultimately delivered to them through sensory means. Complex machinery may exist behind every rule, but it must be flattened onto a screen before it is intelligible. The game itself does not even take place there, but in the mind: without the player to give it breath, it is dead, inert, and cannot exist. This life-giving requires continual acts of perception; we are inverse gorgons whose gaze turns stone into flesh.

How, then, could representation be necessarily incidental? The spread of low-poly models for games like TF2 and Quake 3 shows how closely it is linked to manipulation of the rules: competitive players like to eliminate all distractions, paring back the details, practicing their own ludology. Monopoly, after all, is difficult for the colour blind. Those coloured squares from The Marriage may be abstract, but they are just asaesthetic as the pretty art of , and no less ‘representational’ with regards to the game-rules. Blow’s work in turn offers two perfect examples: World 2 (the first) contains an abstruse puzzle where the player must rearrange parts of a painting so that they can jump on part of its depiction to reach an elusive pick-up. Doing so is based on your ability to recognise the likeness between a painted table in a piece of art, and a painted gantry in the art of the game; progress hinges on aesthetics.

Later, in the first chapter of World 4, we encounter two infuriating locks, walling in a puzzle piece. One will open as expected, but the other will break your key and force you to quit and restart. Presumably it is obvious in the level editor that these are very different entities, but on screen, to the player, they look identical, and there is no way to tell which one is which until they are tested. As far as the game goes, the confusion only exists in a purely aesthetic, non-fundamental realm – but for the player, that’s enough to render the line between frustration and satisfaction completely arbitrary.

Naturally these are also arguments for the primacy of mechanics, which break through the apparently unproblematic image to slap the player in the face; they are what ‘really matter’. Indeed, to speak of aesthetics as ‘affecting’ the player’s relationship with the game-rules is to presuppose the precedence of those rules. Nevertheless, the player’s experience and her capacity to respond to it have both been changed by faulty representation. Perhaps the upshot is merely that aesthetics are only relevant insofar as they affect or distort our experience of the ruleset. But this would be to imagine that a ruleset can ever be neutral or uncontaminated in the first place, that it can exist - as a game - without representation. As every coder knows, no program ever survives contact with the playtesters. For them, ludic and aesthetic content are experienced simultaneously, neither prior to the other.

Formalist approaches tend to be essential to their fields, offering methods which quickly become indispensible. Game criticism is no different. We are all ludologists nowadays – and thank fuck for that. But story and aesthetics shouldn’t be considered off-limits for analysis and it’s unsafe to dismiss them as irrelevant. While rules can be productively examined in isolation, they must ultimately be studied as experienced by the player. We have to be sensitive to the channels through which a game is accessed - the mechanisms of perception by which players must apprehend the systems the play. Doing so, we can practice a romantic criticism, a criticism of experience, which means recognising that the game takes place behind the player’s eyes, identifying what happens there, and articulating the specific formal features which provoke that response. Unicorns seem as good a subject as any.


  1. Great post! I completely agree about the role of music in Robot Unicorn Attack and Canabalt. In fact, I used those two as examples in a presentation back in May (at a university in the American Midwest—you weren't there, by any chance??), swapping the soundtracks to see how it changed the “feeling” of play. The composer for Canabalt discussed this a bit in a Gamasutra piece.

  2. Thank you! I have a huge respect for the scholars of the University of the American Midwest; that sounds very interestng. What music did you use? Ode to Joy? Living on a Prayer? Benny Hill? There is, in fact, a heavy metal version of the game ( last desperate charge through Hell itself to game.html ), which despite not being much 'heavier' than Evanescence does make the whole thing feel like a desperate charge through the jaws of Hell itself to rescue a doomed lover. For some reason I find it much easier than the vanilla version. Perhaps, subconsciously, I just can't bring myself to believe in rainbows and dolphins.

  3. Obviously I 'believe in dolphins' (as opposed to believing in fairies), but you know what I mean.

  4. "Later, in the first chapter of World 4, we encounter two infuriating locks, walling in a puzzle piece. One will open as expected, but the other will break your key and force you to quit and restart."
    Actually these both act the same as all the other locks in Braid. In that world time moves forward as you go right, backwards as you go left. So this is what happens if you try to go through the lock to the right of the puzzle piece:
    1. You hit the lock. It opens, and the key is permanently destroyed (since it is green and glowing).
    2. You try to move left. This makes time go backwards, resetting the lock.

    1. Oh, that is fucking fiendish. Well spotted, Anon. And incidentally nice to realise people are going back and reading this old stuff still.

      It's an interesting observation because it means that there is an actual chance someone could realise in advance what would happen and avoid the problem. This certainly has implications for my example and for the meaning of that moment in Blow's game (although I think the example stands as a hypothetical at least!)