Saturday, February 26, 2011

Emergent Stories

When you play a game, whether tabletop or electronic, how do you think of it? Do you think of it as being merely a collection markers, or do you factor in the actual world they're meant to represent? Is a Space Marine just numbers, or is it an abstraction of an armored warrior from the far future? In essence, are you playing for the rules, or for the setting? When you tell stories of your games, do you think of them in terms of rolls and rules, or in terms of events? When it comes down to it, why are you playing a game that has aesthetics and background and lore if not for your attachment to the concept of an existing universe?

Compare your thoughts about a wargame to a roleplaying game, or even a strategy RPG. Does the focus on individual characters change anything, even though the mechanics themselves are not that different? The main difference between an RPG and a wargame comes down to nothing more than "filling in the gaps" - i.e. going outside the rule system with regards to things like interaction, background, and personality. So why are wargames different?

Wargames represent an under-represented entrant in the concept of an emergent story. By this, I mean that the rules in the average wargame represent things in-universe, and those events unfold based on a combination of strategy and luck. Therefore, in the same way that a role-playing game can serve as a fair and unbiased moderator of events, wargames can serve as a larger-scale unit's story. There are a few games like Necromunda and Mordheim that specifically deal with an ongoing campaign with veterancy and improvement, for example. It's the same concept as an RPG, but on a different scale.

Here is the basic rule: a wargame tells the story of the characters and units currently involved in a given match. The scale of most wargame settings, whether historical or fantastic, means that the number of units involved is generally negligible on the larger scale. It's perfectly believable to have a skirmish between a company of units be part of a far larger ongoing war. More importantly, though, a wargame has a set list of characters (all the units involved) and events unfold based on a logical ruleset that represents what's supposed to be happening in-universe.

This is all a little abstract, so let's throw out a concrete example. This character on the left is Erasmus Tycho, a Space Marine from Warhammer 40,000. Tycho was, in essence, an ascended generic. He originally came from a battle report in an old issue of White Dwarf, where the Blood Angels player named his captain "Tycho". During that battle, he was hit by an Ork's psychic blast and was "killed" in game terms. This was rationalized as the character being grievously disfigured, and his character changed because of it.

These events can be split into two groups: the rule-based events, and the story-based events. The rules provide the skeleton: there is a captain, he was "killed" by an Ork Weirdboy. The story provides the meat: The captain was given a name, the "death" was reflected in a way that furthered his development. This is what emergent story can do. It is a way to provide avenues of imagination that are tempered and moderated by an existing structure of logic and luck, so that the story is "fair", rather than "i got you nuh-uh yeah huh".

What about this is different from an RPG? The larger scale would suggest that there is less room for individual characterization - yet, I do not think this is a bad thing. A complicated character is not necessarily a good one; as long as the character's personality traits and motivations are expressed, it ought to be good enough. It actually doesn't take that much to let players empathize with their soldiers, because imagination can fill in the details that the framework of basic traits provide. This is what leads people to write after-action reports: the fact that the gameplay provides a framework for a narrative. It's relatively simple to ascribe a name and a few traits to a generic officer, but it is the way they distinguish themselves during the battle that gives them real character. This is the kind of thinking that leads people to create their own Chapters or Regiments in WH40k, or their own color schemes in any other game. They're your soldiers.

This isn't limited to tabletop wargames, either. Games like X-COM are dependent almost entirely on the story emerging from the gameplay, because other than the background that is the story. Yet, almost everyone who's played classic X-COM has stories about some character or mission that was intrinsically interesting as an event, without the context of dialogue or plot. Dwarf Fortress is a goldmine for these sorts of things, because weird things happen all the time and the player naturally seeks to explain them in the context of the universe.

In Boatmurdered, for example, the fortress was randomly situated next to an aggressive herd of elephants. That's a random detail in game terms. In-universe, though, it's a whole story in itself. It is the player who fills in those details. A weird character trait or event is not just "a programming thing", it is internalized as something that exists in-universe, and hence the humor comes from subverted expectations and bizarre behavior. It's not unusual in a meta-sense, because that's just numbers, but it's unusual in an in-universe sense, because it's something that wouldn't make sense in real life.

Naturally, there are entire pages devoted to this sort of thing. Unlike most "CMoA" pages, Dwarf Fortress is populated entirely by emergent gameplay events. These are things that happened according to the rules, and were impressive based on in-universe expectations. For other series, CMoAs are generally scripted events or cutscenes. Dwarf Fortress and X-COM don't have that - all the events are based on rules, and the whim of chance determines whether they're a success or a failure.

So let's go over what we've got thus far:
- Rules can provide a moderated, balanced way to determine events.
- Therefore, rules can provide the skeleton or frame of a larger story.
- The human imagination is capable of filling in details and finding meaning in events.
- Therefore, imagination provides the specifics of that story.

When I hear about story in games, though, it's almost never this kind of thing. People tend to prefer dialogue and voice-acting (things that cannot be easily replicated), and they prefer what could be called "complex" plots. There's always the assumption that games should be books or movies, rather than developing naturally from what they are.

The thing that I feel is often overlooked is that small events lead to larger context. Dwarf Fortress works because you're doing everything, and hence things like political disputes and resource issues occur in a fairly natural manner. In contrast, what people expect out of games is something along the lines of an unrelated political issue with a brief period for "gameplay". Is that really a game's story, or is it just a story that's stapled to gameplay?

The idiosyncrasies of human interaction make natural dialogue difficult, at least when it's made to look realistic. There's a few things that can be done, though. Context is an important aspect of things, reflecting different emotional states and situations. For example, in Company of Heroes, units would change their voices to reflect whether or not the unit was in combat, and the status of the unit. This can be heard in this collection. Notice how believable the reactions seem: it seems like a perfectly rational response to whatever event is occurring, and there's a sense that the tank commander is worried when he's in danger, and grateful when he escapes it. Compare that to the siege tank from Starcraft, which has a single measly "I'm in trouble" quote and otherwise displays no sense of changing emotion or fear. There's no sense that the unit is in any danger, and thus it's not believable.

Dialogue also exists in the sense that it's an environmental indicator. Aircraft chatter is a great way to represent this: it may sound like pointless gibberish if you don't know what they're saying, but everything said on an aircraft radio has some meaning. Ace Combat made good use of brevity codes to help indicate what was going on. It relied on an "[x] [y], [z]" concept, where [x] is the squadron designation, [y] is the plane number, and [z] is the brevity code. For example, "Red 4, Fox 2" - i.e. Red squadron's #4 plane has launched an air-to-air IR missile.

One game that used this pretty well was Freelancer. The thing about Freelancer is that there's always some context to use, whether it's a destination planet ("Headed for [x]") or a targeted vehicle ("Targeting [y] [z]"). This meant that phrases could be assembled based on existing nouns and verbs. Silent Hunter did that sort of thing too, although relying more heavily on stated numbers ("Depth [x], bearing [y]"). These aren't going to result in the kind of performances that are notable on their own, but it is a way for dialogue to be utilitarian and purposeful with regards to a logically-developed message.

The main problem with regards to dialogue and speech comes from nuances. It's one thing to set up a pre-programmed voice with a bunch of different variables, or a text system that's able to create sentences out of "building blocks", but creating natural-sounding voices, with distinct pitch, word usage, and characteristics, is beyond the reach of current technology. It is not yet possible to wholly synthesize a voice, and while permutations can be applied to a sample, they don't give the same range as a wholly different voice.

However, I don't feel voices are necessary for the most part. I appreciate them when they come up, and they're very good at conveying emotion (as well as gameplay cues in an audio format), but when it comes to story and dialogue, a book is hardly worse than a movie. When there's no voice given, the player can use their imagination to, again, fill in the gaps. They can create their own concept of the character's voice based on their established traits and personality. It's not a perfect solution, but the innovation of the human mind should not be understated or ignored, either.

When it comes down to it, what is a plot? Plot is comprised of events and characters. We've discussed both of those. What separates a "real" plot from an "assembled" plot? It's really going to end up being down to the details. Let's look at an example video game plot, in this case Warhammer 40k: Dawn of War 2, compared to what can be accomplished with a tabletop and some imagination.

DoW2's campaign is a combination of mundane patrols (hold this area) and occasional story-advancing battles. One strand of this plot is the fate of Davian Thule, the player's character in Dawn of War: Dark Crusade, who is now the commander of the forces that the player is part of. Early on in the game, Thule is injured by a Tyranid's poisons. Later in the game he is revived as a dreadnought. Does this sound familiar? On some level, it's essentially the same as Erasmus Tycho's story - killed in-game, brought back through a plot abstraction.

The difference between these two things is that DoW2's events are scripted and must happen. It is something that is not left to chance, and there is no opportunity for a major character to be permanently injured or killed in the same way. The reason for this is twofold. Firstly, Davian's role as a static part of the plot is necessary to introduce the Dreadnought frame later in the game, when it's more balanced. Secondly, if a character was easily taken down, gameplay would quickly grind to a halt (as there are only a few characters in the game with the kind of voice acting and character modeling that makes a character unique). In essence, they needed Thule to die and return, and they needed the other characters to not die because of the resources that had been sunk into them. This is a video game "plot": things that need to happen to build up complexity that cannot be supported as a logically constructed concept.

Now let's compare this, not to the tabletop, but to another Warhammer 40,000 game: Chaos Gate. Chaos Gate was essentially the 40k version of X-COM: you've got a bunch of Space Marines with names, veterancy, and gear. Chaos Gate was far less "plot-intensive", because it was focused around the completion of missions rather than direct dialogue and character development. However, the player's ability to fill in those gaps came through again for a lot of players. I remember reading about someone who was playing through (as a Let's Play) and had a very popular character who distinguished himself repeatedly in combat. Unfortunately, this character was felled - an ignominious end, but a logical one. Later, however, the character was brought back indirectly in the form of a Dreadnought, to wild applause.

The difference between this story and the story of Davian Thule is slight, except for the fact that it's much more "by the rules". Davian Thule happened because it was always going to happen - this happened because that's what happened in the game. They're essentially the same plot, but one of them requires player abstraction and imagination, and the other has a more direct audio-visual connection based on distinct graphics and voice acting.

In essence, what I'm saying is this: the difference between "a complex, but pre-generated plot" and "a simple, but logically constructed plot" is going to come down to how much weight the player's imagination is going to have to pull. When you're spoon-fed characters with hundreds of lines, they make a more distinct impression, but your imagination doesn't have to do anything. All the work is being done for you, and while that's not necessarily bad, it's kind of a misuse of resources. There's no sense of player involvement, they're just watching an incredibly long movie that they occasionally get to interact with. They're not your characters, they're just characters.

Design & Construction
When artists design characters or armor, they tend to not worry as much about how believable it is. We've established this pretty well in the past - the armor only works because the game says it works, not because it makes sense in-universe. This is where it becomes necessary to differentiate two different types of design: "premade" and "logical".

A "premade" design is something that exists as a set inventory item with an abstracted connection to it: this is a sword, it has five attack. Obviously this sort of abstraction is necessary for most things, but it also leaves a void of consistency. It's not that the intrinsic physical properties, shape, and material of the thing give it that stat - it's just "five attack". Armor can be the same way; no matter the coverage or sensibility, it's "six defense".

A "logical" design, on the other hand, is assembled according to rules and logic, whether it be physics or chemistry or whatever. In real life, swords are used not because of an abstract concept of "attack", but because their construction and shape grants them advantages in certain kinds of combat. This is based on their weight, their sharpness, the way they can be swung or thrust, and so on. In essence, it possesses physical properties based on underlying principles and that is what makes it useful.

The clearest example of this that I can think of can be found in Dwarf Fortress, a game where it's possible to build a working computer (based on water flow and other internal principles) but not a gun (because small objects are premade). One of these things is on a large enough scale to be affected, and the other is "an inventory item". There's rules for the former that can be twisted to the player's logical advantage, but the latter is pre-generated.

Now, think of armor. Armor in real life has a bunch of physical properties that I've discussed in the past - coverage, thickness, weight, and so on. In a game, those things are generally abstracted, so coverage and thickness come together to form a vague armor level that's meant to connect to what it represents. That's just a number, though, so it's easy to slap on an inaccurate number. Compare, though, how Mount & Blade handles armor based on locational damage. It's still abstracted, but less so: in M&B, attacks are handled by swinging in a given direction or thrusting. An attack that connects deals damage to the body part that was hit. There's very little abstraction there except regarding the damage taken.

Why is that important? Because the rules of the game logically connect to the rules of in-universe reality. A character would wear a helmet for the same reason that the player would give them one: to protect their head. There are still minor differences in the decision-making process, of course, but the logic is still basically the same: "wear a helmet so that if someone hits your head you don't die". It's natural logic that's fully explainable in-universe. It's an aspect of believability that makes game decisions share the same logic as character decisions.

In essence, things work the way they do for a reason. If you recreate the physics and logic behind them, then they can be assembled in a sensible way. If not, then people are just going to have to be stuck with premade concepts. The former allows for some exercising of creativity, the latter serves as an easy-to-make template that nonetheless lacks a lot of deeper properties.

The point I'm trying to make with this article is that story can come from anywhere, but having a story that emerges as the logical end result of the player's actions ought to be more immersive than having a story that the player simply shuffles along. I've also pointed out the ways that such a story would be inferior based on the limitations of technology, and yet things like After-Action Reports and Let's Plays suggest that the human mind is capable of filling in the blanks even if statements are not specifically made. If the writers wanted to be more specific they could do what FFXI did and have a few different personalities with malleable lines (and I'm sure there have been other games that did this too).

Basically, there's a lot of ways that a seemingly normal or low-story game can develop a story. A story like that rewards the player for making choices in the context of the gameplay itself, and thus should be developed. The average video game story, at the very most, allows the player to pick which linear path they trod down.

So, To Sum Up:
1) Gameplay in a game exists conceptually in-universe: it is meant to be a battle or event that's taking place for all the characters involved with it.
2) Therefore, there's no reason to assume that things like personality and characterization cannot be attributed to those characters the same way they are in an RPG.
3) Allowing the events of the game to provide a framework for a story allows for more exercising of imagination and creativity than simply being told what their lines are and what their voices sound like.

No comments:

Post a Comment