Monday, 27 July 2015


In my work on Tales of Vesperia particularly and JRPGs in general, I’ve been poking at a lot of virtual worlds, looking for the places where they start to come apart and the meaning that can be gleaned therefrom. In trying to develop a theoretical framework for this, I spent some time going through the etymology of the word ‘world’[1], and hopefully what I found is at least worth this post.

‘World’ comes from a Germanic root, a combination of ‘were’, meaning ‘man’ (as in ‘werewolf’) and ‘old’ for (if this isn’t obvious) things relating to age. The OED gives an ‘originally literal’ meaning of ‘age of man’[2]. What interests me about this is the temporal component.

I think it would be fair to say that we generally use ‘world’ as a geographic or ontological descriptor; ‘the world’ is either a place, or a(n in-some-way-maximal) set of objects. That’s not universally true, as we might speak of ‘the Jurassic world’ or ‘the modern world’, or say of some era of history that ‘it was a different world back then’, but I think it fits how the word is used today (I’d be very interested to hear from anyone who disagrees). We tend to use ‘age’ or ‘era’ to speak of periods of time.

And I think this is worth remembering when discussing ‘world’ in videogames, because it’s the temporal dimension of game worlds that tends to be the more complicated. Games compress distance, and even chop it up a bit, but as I’ve argued before, game spaces are generally straightforwardly spatial.

Time in game worlds, though, is rather more complicated. Partly out of technical limitations and partly out of courtesy to players, games have to convey the passage of time in mostly-abstract ways. Even the most literal driving/racing game, with a millisecond-precise lap timer in the top corner of the screen, will run its career mode, car tuning and car select menus in abstract time (thank heavens).

One of the coolest and most effective single experiences I’ve had in a game was Queers in Love at the End of the World by Anna Anthropy (go play it quickly and come back). This seems, on the face of it, to be a game with very literal time – a timer at the top of the screen counting down from 10 in seconds – but it relies on an abstraction to make that time meaningful. Specifically, the duration of your actions is abstracted to the time it takes to read, understand and navigate the hyperlinked options.

Without wanting to dig too deeply, this does complicate things. Time in Queers in Love passes differently depending on your reading speed. I tend to have time for about two choices; I feel like I can usually do more than ten seconds’ worth of actions in the game’s ten literal seconds. A slower reader, someone reading in their second language or with a condition like dyslexia, may find time flowing rather faster for their in-game avatar than I do.

The kind of games I spend most of my time playing – JRPGs and other epic narrative games – have to compress time much more severely. Sequences of events that by rights should take years must be fitted into forty hours or less. There are lots of ways to do this, all of which can be clunky in some contexts and entirely graceful in others.

It’s often in handling time that JRPGs get most abstract, or furthest from literalism. Plot developments that are tied to player progress can seem preposterous coincidences; taking time to complete sidequests after reaching the final save point before the final boss can drain all urgency from the impending apocalypse.

I’m not going to go into these topics in more detail this time, because each is worth at least a post on its own. But I do think it’s important to look at how games represent time, how we should interpret those representations, and how virtual experiences may conflict with our relationships to non-virtual time – and the non-virtual world.

[1] This is a Thing Philosophers Do, and I don’t necessarily mean to advocate for definitions based on etymology over use, but it’s often helpful when trying to develop a lexicon to get a sense of what other linguistic roots may be relevant.

[2] It should be noted – though it’s by no means my place to do more than note – that ‘were’ means ‘man’ in the sense of ‘male person’, not as a generic term for human beings. I leave it to more astute feminist scholars to examine the fact that ‘world’ is by root a gendered term.

Thursday, 9 July 2015

The Only 'Real' That Matters

You know that feeling of disbelief when you beat a difficult final boss and instead of another phase the ending cutscene starts to play? The sunrise-slow shift from breathless relief to triumph?

Has a plot twist ever punched you right in the gut (looking at you, Aeris)? Has an environment ever sent chills down your spine[1]? Ever find yourself wanting to chew your controller (don’t inspect my old Gamecube pad for toothmarks, please), or pitch your TV out the window?

These are the kind of effects that often get forgotten in the popular debate about whether games can have ‘real effects’. The focus tends to be on whether games are turning ‘our children’ into (a) uncontrollably violent beasts or (b) posthuman supergeniuses. But of course games can have real effects at the personal level, and the other concerns are best left to psychologists with some grasp of proper investigative methodology.

And it’s not just that certain pieces of digital software can cause effects on real people. If you cried when Aeris died, or punched the air when you beat your first Bowser, it wasn’t the flipping of a bit somewhere inside the console, or a shifting pattern of electromagnetic radiation emitted by the screen that had that effect on you.

Okay, in a way it was, but describing the pattern of light or the behaviour of the silicon isn’t the best explanation of your response. You didn’t punch the air because of [obscure technical description of computer hardware]. You punched the air because YOU BEAT BOWSER YEAAHHHH.

And really, that should be all it takes to justify the serious study of games and gaming. We shouldn’t need esoteric philosophical arguments like this and this to defend what we do. In an ideal world, the capacity to cause real experiences would be the only thing that counted when deciding what’s worth taking an interest in[2].

It can be tempting to dismiss this position on the grounds that game events or objects are just fictions – that we can account for them in the same way that we account for Sherlock Holmes, or Batman, or Narnia. And you know, fair enough provided you’re going to take the study of Sherlock Holmes and Batman and Narnia seriously (pulp crime fiction, comics and children’s fantasy have all had their own fights for recognition in academia). But there is a bit of a difference.

A fiction is a kind of tacit agreement between audience and author. Very roughly, when you open a novel, you’re accepting that the author is going to tell you a bunch of lies (or at least, things that aren’t true), but that something in those lies will be worth your attention. You’re entertaining the lies in the hope or expectation that they will have some positive effect on you – make you feel good, teach you something new, guide you to self-reflection.

Some games are fictions, of course. But some don’t really involve straightforward fictive assertions at all. Look at a game like Geometry Dash, a game that consists almost entirely of level geometry. The geometry of a Dash level isn’t a lie about how things are in some other realm, it’s just there. If I tell you there’s a sequence of spikes at a certain point in the level, that’s true.

In this way, game objects are more like the notes of a piece of music. They may contribute to the telling of a story, or the communication of fictive claims, but they are real parts of the work, however transient or intangible. “There’s a mushroom in the second ‘?’ block of Super Mario Bros level 1-1.” is the same kind of statement as “The first notes of Scott Joplin’s The Entertainer are a D octave.”

This is why I think it’s useful to consider game events real; it’s not just that they’re causally effective (though in some ways that’s enough on its own), but that they are also intersubjectively consistent in a way that fictive events are not. If you claim that ‘Sherlock Holmes lived at 221B Baker St.’, I am not obliged to agree, since the statement is only accurate to Conan Doyle’s fiction, and not the Baker Street that you can walk along in London. We must agree a frame of reference before that statement is true. But ‘The Companion Cube appears in Test Chamber 17’ is true simpliciter.

There’s more to be said on this, particularly concerning the types of fictional statement (‘221B Baker Street’ is a fictional address in a real place, whereas somewhere like ‘Bag End, Hobbiton’ is a fictional address in a fictional place), but I do need to leave something for the academic paper I’m writing on this subject, so I’ll leave things here for now.

[1] I’m not able to offer any intelligible examples of this one because I scare easily and I’m not admitting just how terrified I was of the Great Deku Tree for a while in early 1999…

[2] For more on this terrible philosophy joke, I wrote a brief and slightly clunky introduction to philosophical idealism here.

Monday, 6 July 2015

Abstraction, Simulation and Narrative Time

This post grew out of a conversation that’s happened between several people over quite an extended period of time, so apologies if I have to send you on a bit of a reading list to catch you up.

It more or less started here with Devon Carter’s Critical Switch guest episode about the role of abstraction in JRPG mechanics (which is basically essential reading if you want to follow anything I write on this blog, so get to it if you haven’t already).

Then Austin C. Howe had some thoughts about JRPGs on generation 7 hardware, which I Storified, where he discussed how more powerful technology had revealed some of the ways in which the classic JRPG form struggles with ‘realism’.

That prompted LeeRoy Lewin to write this excellent piece about how the basic abstractions of JRPG combat – HP, MP and XP – are bad representations of anything ‘real’.

Finally, and possibly independently of all this, Vincent Kinian reviewed Yuuyami Doori Tankentai with a focus on how it actually does address the real successfully.

Got all that? The key point I want to focus on is LeeRoy’s, that the mechanical abstractions in JRPGs don’t directly represent the things we ordinarily take them to. HP are a poor representation of health, or even endurance, because health is not linear or one-dimensional. XP in particular, to quote LeeRoy, models “the purest meritocracy, the most awful gamification, the idea that labor can transfer 1:1 value.”

If MP and equivalents are innocent in this, it’s only because they represent something that makes no claim to reality – magical power. Because it’s entirely up to an author to invent a magic system, we can’t gainsay the device of MP, but we can point out that it’s a formulaic and often uninteresting way of limiting power. To be fair, sometimes it doesn’t need to be interesting and is just an unobtrusive balancing tool, but I feel like there’s often a lot that could be done here that isn’t.

The more complex, player-engaging systems built on top of these stats – materia, GF junctions, sphere grids and so on – may be a little more representationally rich, depending on how they are contextualised[1], but the statistical foundation is poor in this regard. So why are these numbers so ubiquitous?

Convention, and the legacy of tabletop roleplaying games, makes up one major part of it, of course. Things are Done This Way because That’s How They’ve Always Been Done (and in the case of at least HP, the problem isn’t unique to JRPGs or even RPGs in general). But these stats do serve a function, or at least are a crucial pillar of a system that serves a function, and that function, I would argue, can be meaningful and valuable.

What JRPG systems – the stuff we think of as ‘systems’, the numbers bits – do is control the delivery of a story. They may do other things as well, but one persistent function is to determine when the story moves forward. Part of this is just to space things out, so you’re not just playing a novel (and to meet the ever-louder consumer demand for ‘value for money’, as measured in hours of ‘content’), but there’s also a crucial contribution that this makes to JRPG stories.

Pacing is everything. JRPG stories tend towards the epic in style as well as length. They are stories of big deeds, world-shaking changes. Worlds do not change overnight. Many forms of storytelling work just fine over short periods – one can hardly object to the near-real-time duration of Twelve Angry Men, or the one-night framing of Die Hard – but the grander the scale of a story, the slower it needs to feel like it moves.

The games industry is already bankrupting itself and driving employees to exhaustion to deliver games of sufficient length and graphical quality to keep ‘the market’ placated. As graphical standards continue to rise, representing the passage of time non-literally is going to get more and more important for delivering long-form stories.

This is where RPG mechanics, and particularly their JRPG versions, can shine. HP and MP represent the costs of time; the wear and tear of travel, the hazards and labour it involves. XP represents the benefits of that time – as LeeRoy pointed out, often badly, but I think this approach to them at least suggests some ways to complicate them into something more meaningful.

What is represented by these stats – perhaps it would be better to say ‘encoded’, since ‘representation’ implies something more explicit – is not really anything about the software objects to which the variables belong. What is represented is the world, the time and space through which these objects move.

If an encounter, or sequence of encounters, leaves your party on their last legs, you’re given the sense of characters limping into town, propping each other up, maybe dragging a stretcher, bags emptied of supplies. Backtrack over the same ground later in the story, and the easy passage, the image it conjures of marching proudly through the same gates with head held high, represents the sheer length of time it takes to master what was once marginal.

Here is where that stuff from Austin and Vincent becomes important. Taking stats to be directly representative creates misleading indulgences about vast gains in personal power and importance over implausibly narrow stretches of time. Austin’s point that realism is about more than just visual literalism, and Vincent’s that the sense of the real comes from the mundane, the stuff that happens between the fantasies, suggest that the way to get real value from explicit stats is to look at what they imply.

If a magic spell allows you to do ten times as much damage to a target as the strongest physical attack, but costs ten times as much to use, what does that say about the scarcity of that spell, the rarity of this casting? If a character is dying of poison and you’re out of antidotes, can you make it to the next inn or item shop in time? Frequent easy encounters that wear you down slowly suggest a long journey; fewer more potent encounters suggest danger and pain but less time; a surfeit of brutal fights implies that the world must turn a little longer before you can make it through.

The actual games could do a lot more to draw attention to this side of reading them, but that doesn’t mean that existing games shouldn’t or can’t be read in this way. It’s an approach that I’ve found useful particularly when looking at games that focus their narratives on journeys – I'm a broken record, I know, but Final Fantasy XII and Tales of Vesperia are both excellent examples of this[2].

[1] Some are much worse, particularly ‘relationship’ systems that reduce social ties between characters to single score variables.

[2] There are a couple of sequences in Vesperia where mechanical representations of the passage of time are directly contrasted with what cutscenes have told you about the time-frame of events, which show both how effectively good stat usage can suggest the passage of time and how quickly bad stat usage can undercut it.