Turns

(Originally published on 31/8/18 at JRPGsaredead.fyi)

The recent resurfacing of the ‘is JRPG a genre?’ debate on twitter didn’t really break any new ground (it totally isn’t, by the way), but I did finally decide to sit down and respond to one common structure in the discussion. When we point out that the word ‘Japanese’ refers to a nationality and thus using it in a marketing role like a genre name is suspect at best and outright appropriative at worst, we often get responses along the lines of ‘but you have to admit there are some common features, like random encounters, turn-based combat… etc. etc. etc.’

I’m going to focus on just one of those common examples (some of which are much more common than others), but the general thrust of my argument applies to each of them in turn. There is something iconic about turn-based combat and JRPGs, in the sense that games culture takes the presence of a turn-based system as one of the key signifiers that a JRPG is being played. In reality, though, ‘turn-based combat’ refers to a very loose grouping of mechanical styles that contribute a wide range of different affects (feels) to the games they appear in.

What I want to do here is highlight some differences that exist between different kinds of ‘turn-based combat’ systems. I’m not trying to develop a rigorous taxonomy – I no longer believe this to be a useful project – and I want to avoid sweeping claims about the effects produced by different systems. Some of that is inevitable, though; one cannot talk about difference without pricking the common neurological tendency to classify.

Breaking up the concept of ‘turn-based combat’ is particularly important because it helps us get over a stumbling block that emerges when we try to move from a strictly genre-based account of JRPGs to a more genealogic model like the history we’ve been trying to develop at DGC. This is the tendency of the western JRPG press to treat consistency across generations of JRPGs as a product of development inertia rather than the work of skilled designers who understand their own history – a tendency that reinforces narratives of the Japanese industry as backward and obsolete rather than a distinctive and legitimate cultural industry in its own right.

One final note before we get to some examples: the temptation to see ‘turn-based combat’ as one of the defining characteristics of JRPGs overlooks a long history of Japanese-made RPGs and games with clear RPG backbones that are not turn-based. These run from Secret of Mana and the early Zelda games through Kingdom Hearts to Dark Souls. Arguably at least some metroidvanias fall into this category. These games absolutely belong in a history of Japanese RPG development, both in terms of their influences and the games they have influenced, but they aren’t my main focus here.

A good place to start is with a difference that is often overlooked, but nevertheless has profound consequences for the kind of decision-making that players are invited to do in engaging with a game. This is between combat systems that divide turns between the player and the computer, and systems that divide turns among characters and other agents in the environment. Do all the player’s characters act before any of the computer-controlled enemies? Does the player input all their commands for a round in one go but have them play out intermingled with computer-determined actions? Or does each individual unit have its own timer, so that a player’s input is always tied to one specific character at a time?

For a straightforward example of the difference this can make to play, compare Disgaea to Final Fantasy Tactics. In Disgaea, there is a player phase, in which the player’s units can act, once each, in any order, followed by an enemy phase in which the player’s units sit completely passive. A key part of play is arranging complex sequences of actions that build combos, move enemies around the field, and exploit environmental effects (like the geo puzzles that dot most areas). This is only possible because – with the exception of counterattacks – the player’s turn cannot be interrupted by enemy responses.

In Final Fantasy Tactics, on the other hand, each character acts in turn, independently of the others. A fast character might get twice as many turns as a slow one, and the order in which character and enemy turns come up depends entirely on their stats. Here there are no dramatic combos and any grand plan may be interrupted by the enemy moving around.

This brings up a second kind of difference, too, which is the extent to which action order is random. In Disgaea, the player has almost total control over the order in which their units’ actions happen, and while they cannot control the order in which enemies act, they can nevertheless be sure that enemy actions won’t spill out of the enemy phase. In a game like FF Tactics, some prediction of the turn order is possible, but only a few steps into the future.

In the early Final Fantasy games, combat proceeds by rounds in which each unit in the fight gets to act once, in an order determined at random but influenced by each unit’s stats. Players enter all commands at the start of the round and then must wait to see what order the actions happen in; any planned sequence of actions (for example, one character casting a spell to increase another character’s damage output before the second character launches an attack) may thus fall apart on execution if the virtual dice rolls go against it. How capricious a game’s turn order is makes a huge difference to its difficulty and its mood, and is something that each game’s design team must give substantial consideration to for balance reasons if nothing else.

Maybe the most conspicuous difference is between those turn-based systems that are ‘active’ and those that are ‘static’ – roughly between games in which, if the player stops making inputs, the player characters will eventually die and those in which the game will wait indefinitely for input. Whether or not you can put the controller down safely determines how a game fits into your life, how quickly you can switch attention to something else in your environment. One of the most maddening things about Eternal Sonata is that different segments of the combat sequence have different rules for this. Enemy actions are not safe since the player must be alert for the opportunity to block or counter. Player character actions are safe at the start of the game, but as the combat system evolves they become bound by a timer that eventually completely stops waiting for player input.

A difference that opens up particularly productive questions is between games that feel turn-based but aren’t and those that are turn-based in some strictly code-centric sense but disguise the fact. Final Fantasy XIII frequently gets described as a ‘return’ to turn-based combat after the real-time action of Final Fantasy XII, because in FFXIII, a character must pause for several seconds between chains of action – it feels like waiting for another turn to come up. Meanwhile, characters stand idle for much shorter periods in Final Fantasy XII, and can move around under direct player control while they wait (the relationship between spatiality and temporality hinted at here deserves much more detailed development, but it will have to wait for another time).

Under the hood, the opposite is true. In FFXII (except in the recent Zodiac Age remaster), each character’s actions, once fully charged, are placed on a first-in-first-out stack called the ‘effect queue’, from which they are then discharged in strict order (it’s actually a bit more complex than this – there are multiple stacks for different kinds of actions – but the point stands). When an ability with a long animation is being discharged from the queue, combat grinds to a halt while it plays out. There are some extraordinary speedrun techniques involving ‘jamming’ the effect queue to prevent particularly large and flashy enemy abilities being added to it.

In FFXIII, there is no queueing. Player-character actions go off in the middle of enemy attacks and vice versa. Indeed, high-level play in FFXIII requires carefully timing these intersections, for example to ensure that a character is in the middle of a jump at the moment an enemy uses a ground-based attack.

Is it reasonable, or even meaningful, to call a system turn-based if turns can happen at the same time? Which matters more for counting a system as turn-based, the mechanical rigidity of ordering in FFXII or the wait-your-turn feeling of FFXIII? What’s really interesting about this question is that, however you answer it, you implicate a variety of western games too.

The BioWare/Black Isle AD&D games – Baldur’s Gate, Icewind Dale etc. – are turn-based in just the way that FFXII is. There are strong similarities between the 90s X-Com games and Final Fantasy Tactics or Tactics Ogre, and I gather also between the 2012 X-Com remake and Valkyria Chronicles. That’s not to say that these games influenced each other directly in either direction, just that strict, mechanical turn-basis appears on both sides of the Pacific.

The more experiential kind of turn-taking found in FFXIII is actually pretty widespread in western games. Pong has it, for crying out loud. So does any cover shooter – sure, you can poke your head up out of sequence, but good luck with that.

 
Again, the point of all this is not to propose a new taxonomy but to resist the idea that there’s a single, stable systemic pattern distinctively associated with Japanese RPGs. When a Japanese studio decides to make an RPG, its designers must choose among all these differences, picking which design heritages to draw on and combine and which to leave aside. Saying that JRPGs have turn-based combat (and by implication that western games don’t) is a gross oversimplification.

Why do you care if it's realistic?

(Originally published on 20/12/17 at JRPGsaredead.fyi)

By far the most common kind of complaint about JRPGs is that they are 'unrealistic' – most commonly in their combat systems or character designs. This holds true despite the fact that all JRPGs are fantasy stories of one kind or another (I know of no exceptions). In fact, given that nearly every western game is also fantastical, it's weird that we care about 'realism' at all in games.

Anyone who's ever been told that people of colour in The Witcher would be unrealistic but demons and ghosts are fine will be unsurprised at the suggestion that there's something ideological in our use of 'realism'. Trace the concept carefully enough, though, and this is revealed as the tip of a very large iceberg. The valorisation of 'realism' goes alongside many of the most toxic features of games culture.

First we should get clear about exactly what 'realism' means in games. There are two things that seem to be important here. The first is obvious: the idea that game worlds should simulate relevant features of the real world. The second takes a little more explaining but concerns authorship and the relationship a game establishes (or doesn't) between its players and creators.

 

No video game has ever accurately simulated anything more complex than very basic physics. The computational resources have never been available – the kind of high-detail simulations sometimes used in scientific research frequently use orders of magnitude more computing power than a PC or home console can provide. For the most part, game 'simulations' are complex systems of smoke and mirrors that aim at the appearance of simulation without its substance.

Even in this attenuated sense, some things are easier to simulate than others. Computers are number-crunchers; things that are easier to quantify are easier to computerise. The physical properties of medium-sized material objects (at least, those without complex internal structures) are probably easiest. Prices, which are already quantification, are also easy to handle.

RPGs are often taken as simulations, in this reduced sense, of particular people; the reduction of human beings to numbers. This is a bad way of looking at them, generally, but in looking at what RPGs can and can't plausibly be thought to simulate, it's possible to see hints of the underlying ideology. Anyone who's ever tried to play a tabletop RPG with 'social combat' – the reduction of conversation to dice-rolling – can attest to how far short it falls of modelling human interaction. Video games rarely even try on this front, except for the odd dialogue option locked behind a stat or RNG requirement.

The dream of video game simulation, I think, is the same as the simulation dreams of futurists like Nick Bostrom: that there is some algorithm, and a sufficiently powerful computer to run it, that will generate whole virtual worlds from first principles, lacking perhaps the fine structure that we can never perceive in the real world anyway but nevertheless perfectly convincing to human perception. Such virtual worlds would be no different from the real world, except in respect of how they are constituted – by silicon and electrical charge rather than whatever the building blocks of the real world turn out to be.

I am sceptical that an algorithm of this kind exists, and even more so that a computer could be built to run it. But even if I am proved wrong, it will not be for a long time yet – the simulations we will make and play in for many years yet will be reductive approximations with the rough edges very carefully hidden from view (think for a moment about how angrily gamers react when the rough edges aren't well-enough hidden, especially when these moments are perceived as 'glitches').

In a way, the second important feature of realism builds on this concealment. It's the lie of completeness, the effort to hide all the ways in which a game is a software artifact rather than a comprehensive world. Most importantly, though (and in a sense most 'realistically'), it's the hiding of the role of designers and craftspeople in the creation of game-worlds. The real world doesn't show any unambiguous evidence of design, so a game world can't be 'realistic' if it does.

This is where some of the excitement over procedural generation comes from. It's not just that algorithmically-led games offer inexhaustible permutations of level and encounter design for a finite cost. Just as the real world emerges from the wilderness of basic physics, so procedurally-generated spaces emerge from the mathematical wilderness of the algorithm anew with each instantiaion.

There is, of course, a long tradition of other art forms incorporating randomness in composition ('aleatoricism'). I see this as almost diametrically opposed to the mainstream attitude to procgen in video games. In other artforms, procedural generation has aimed at spreading out our concept of artistic value; in games, the many instantiations of a procedural algorithm are supposed to converge on some particular value – fun, or 'replayability' (notice how this term frames the experience as a repeat despite its claim to newness or variety).

And the appearance of emergence, the placing of a mathematical wilderness between developers and players, also spreads a thin veneer of neutrality over in-game events. Gamer and designer both can point to the algorithm and say 'I didn't want this specifically, the computer did it'. This is the technological analogue of economic meritocracy – people born rich handing off all responsibility for the inequalities they benefit from to 'the market' or the lottery of birth.

All of which clearly travels with the poisonous devaluing of developer labour that pervades the game industry. If a game developer's only task is to produce an algorithm that replicates reality, there's no reason for anything creative or personal to intrude. No specific person needs to do the work, provided someone does, so one developer is much as good as another and there's no reason to care if one is fired, burned out, or excluded by hostile office culture. One could almost say that developers themselves are treated as algorithms – 'coffee in, code out', as the mug has it.

I try to avoid calling this 'realism' in my own work, though I've drawn this definition from how that word is used in games. The real – whatever it may be – is not algorithmic, nor is it ever impersonal. Instead, I prefer to call this ideology 'naturalism'. It is the idea that the order that produces game worlds cannot be moralised or challenged – in just the way that we are encouraged to take the material world. It's just something we find ourselves in, we have to accept it...

...by which, of course, the speaker always means 'we' have to accept the part of it they want you to accept.

This goes some way to explaining why these two features of realism are prized so highly in games culture, but we can dig deeper yet. We value simulation because the more detailed a simulation is, the closer its objects approach to the kind of reality we ascribe to 'ordinary', 'real' material objects; we value these because that's how we've been encultured under capitalism. The quality or character of the experience produced by an object doesn't matter except insofar as it reflects its thing-ness.

(This kind of materialism – the valuing of stuff over humanity – takes root under capitalism because stuff is easy to quantify, surveil and control. Stuff, including human flesh, can be subjected to physical force; predictive models can be formulated about the future behaviours of stuff. Again, Bostrom's simulations are an echo of this.)

Explaining the valorisation of naturalism is more complex. After all, if games are art – as so many different segments of games culture assert so vehemently, with varying degrees of sincerity and understanding – perhaps the most obvious feature of art is that it is in some sense the product of intentional action. That's not to say that authorship is essential to or necessarily an important part of the theory of art; only that what delineates such-and-such an artwork as itself is the scope of the labour invested in its creation.

In one sense, then, there's a conflict between the concealment of the labour in game development and the claim that games are works of art. In practical terms, the apparent contradiction can be resolved by recognising that most of the people who value naturalism are declaring games to be art either disingenuously or without any substantive understanding of what it means to say something is art. Capitalist labour theory does more to explain the emphasis on naturalism than any of the surface claims games culture makes around the topic.

I need to be a bit careful in how I approach this area. This is material I studied at an English university, where it was presentated as entirely natural and right, and it's only relatively recently that I've got my head far enough out of that culture to see the problems with it. Hopefully I can describe this theory without implicitly endorsing it; let me state for the record that I certainly don't intend to endorse it.

The theory I want to talk about is often attributed to John Locke, though there's an argument that it took on a meaning Locke never intended with the advent of the industrial revolution. For Locke, property is a natural right (there's that word again – remember that 'natural' generally indicates something that you're supposed to accept without question)

The argument goes something like this: we all have a fundamental right to life, and we all, as a fundamental part of being human, own our own bodies. To live, we need various forms of sustenance, and to sustain ourselves, we need to put these forms of sustenance into our bodies in such a way that they become part of our bodies – and thus part of our property. So there must be a point at which we take or claim ownership of the stuff we eat, drink, and otherwise depend on to live.

For Locke, the turning point is the application of our own labour to the raw materials of sustenance – the archetype of this being growing your own crop to eat it. A pack of seeds, a patch of soil and a shovel full of fertiliser don't make a particularly edible or nutritious meal; a farmer labours to combine these into a crop of vegetables that can be eaten. Locke holds that the farmer's labour has become the edibility of the crop – that, in a sense, the farmer has actually combined part of theirself with the seeds and soil. By the principle of self-ownership, then, the farmer owns the crop even before they eat it.

There are a lot of complexities on top of this. Locke proposes a couple of limits on our right to 'appropriate' raw material from nature in this way: that we have to leave enough for others and that we shouldn't cultivate more than we need, to avoid wastage. However, he also introduces the possibility of trade, and from that the idea of currency, which can be accumulated indefinitely without transgressing either limitation. This is how the theory makes the jump to justifying large-scale modern capitalism.

The theoretical stuff may sound fanciful, but it's not the logic of the argument that's the problem so much as the principle of self-ownership. This was an obvious way for a 17^th-century philosopher to talk about bodily autonomy as a right within the context of a fundamentally feudal society, but that doesn't make it a good way to do so. That we have a right of some sort to bodily autonomy should be self-evident, but it's not clear what the theory gains by turning this into a property over our own bodies (unless, of course, you're specifically trying to justify capitalism).

Much of the more complex later stages of the theory aren't of interest right now. It's the basic idea of labour-mixing that I think is relevant. A crucial part of Locke's view is that, without the consent of other human beings, we can only appropriate raw materials from nature. I can't harvest your crop without your permission, but I can take unclaimed land and raise my own crop there (it's probably important to note that, according to Locke, in doing so I come to own the land as well, since in working the soil I inevitably increase its productivity).

There's another precondition, too, which is that Locke, a Christian, believed the Earth to have been given to humanity by God specifically for the purpose of sustaining us. We had not just a right but a god-given imperative to cultivate it. Locke's scornful and racist remarks about 'primitive' cultures not doing enough cultivating to count as having taken ownership of the land they live on certainly emerge in this context, and served to justify the disposession and displacement of native peoples who opposed European colonisation.

What does this have to do with video games? Well, there's no 'unclaimed nature' in the real world anymore – all (or at least almost all) land is legally the domain of some institution or individual. If video game environments are analogous to the real world thanks to simulation, though, they offer us new, unclaimed, natural worlds with each new instantiation.

That frees gamers, in some sense, to appropriate new property from their virtual activities. Of course, virtual objects can't be literally mixed with a person's body for sustenance, but real human activity (button-pressing, and decisions about which buttons to press) can be mixed with virtual objects to produce new phenomena, or at least new instantiations of designed phenomena – this is what happens whenever a player input affects whatever's on screen or coming out of the speakers.

(It's definitely true that player activity contributes to the production of new and valuable experiences during play, but experiences – being non-material – aren't often highly-valued under capitalism. They're too ephemeral, too hard to measure and control.)

This leaves two problems. Firstly, because these virtual natures are infinitely reinstantiable, anyone can possess them – there's nothing special about possessing what anyone can have. Secondly, button-pressing, and making decisions about which buttons to press, generally doesn't rank very highly on the arbitrary scale of labour value that our culture has built.

'Difficulty' – a word that, like 'realism', turns out to have a special, and pretty incoherent, meaning in video games – is the crutch used to remedy this. 'Difficult' work is supposed to be more valuable, which is why we pay bankers millions and nurses barely a living wage (to say nothing of the pittance we pay to people who can, by mind-blowing dexterity and stamina, keep a supermarket checkout flowing through lunchtime rush). Similarly, if a thing is 'difficult' to do, that is supposed to entail that few people have done it; if a thing is difficult to appropriate, then few people must own it, restoring its scarcity.

I hardly need to make explicit the relationships to games culture here, but also there are too many apposite examples to list. It's every 'hard' difficulty mode that describes itself as 'the way the game was meant to be played' and every gamer who says 'well, you haven't really beaten it until you've beaten it on hard'. It's also every single use of the phrase 'dumbing down', and every complaint about 'girl gamers', 'casual gamers' and other dogwhistles for non-cishet-white-male gamers.

(There's a reciprocal relationship here, too, just as in conventional capitalism. Activities that are associated with less-privileged groups get devalued by that association, so it's entirely consistent that forms of game interaction associated with less-privileged groups are automatically assumed to be less 'difficult' – and thus a less legitimate path to owning games-culture real estate).

But most of the games celebrated as 'hard' are actually finely-tuned to be completable. Think how narrow the tolerances are on a Super Meat Boy level. Or how carefully the challenge of Demon's Souls rises to each boss, how there's always just enough room that, if you're really good, you can dodge and not die. It takes extraordinary testing to produce that kind of precision in experience.

Difficulty as it features in life in the real world is a very different beast. Not everyone starts with the same basic resources. Misfortune can ruin a life in seconds without warning or recourse to justice. Games that come close to this pattern (and few if any really deliver it, since the costs are so carefully constrained within the virtual) tend to be regarded as 'unfair' and shoddily-made (interestingly, a lot of older, less well-known-in-the-west Japanese RPGs fall into this category).

If we still lived against the backdrop of Locke's theological context, with our appropriations justified because God ordained them for us, this might not be a problem. But the context has changed, perhaps vaguely thanks to Darwin. The idea that ownership is a paternalistic grant from a benevolent deity smacks of 'participation trophies' and – again – 'dumbing down'.

So just as capitalism has come to function without the idea that the world was created by God for our benefit, and even depend on its absence, so games culture needs to deny that its worlds are created specifically to be claimed.

All this, of course, is a barely-plausible web of half-baked ideological positions. There are contradictions if you look much closer than I have here. And it would be wrong to say that the culture as a lived and enacted thing was predetermined by this structure of ideas; ideologies are always constituted by the human behaviours they promote. But a chair is constituted by the ass-holding-up behaviours of its legs and seat, and it's still true that the chair holds your ass up.

JRPGs – specifically the Square Enix/later Final Fantasy kind – have suffered more than most under this ideology not only because they make little pretense of simulation. This particular strand, which has come to stand for the whole Japanese RPG tradition in the west by accident of marketing and sales more than any defining aesthetic or formal feature, also unapologetically lifts and guides the player, arranging the environments and encounters carefully around them. There's no better example of this than Final Fantasy XIII, and we know how that was received.

There's nothing neutral or simple about the way video games use 'realism', nor about the complaint that such-and-such a game isn't 'realistic'. This terminology is grounded both in an immediate material/industrial context and in a philosophical history that predates video games by several centuries. The baggage that we bring in when we talk about realism is complex, dangerous and already contributing to wide-ranging harm.