Role-Playing Game Theory

Role-playing gamers have theories of role-play. Actors have theories of role-play too, of course, but players of role-playing games (RPGs) do not consider these to be appropriate to their particular circumstances. Instead, they’ve developed their own theories, the origins of which lie in the discussions of role-playing that occur on the Usenet group rec.games.frp.advocacy. Most of this work was done in the context of human-moderated games, whether face-to-face or live action; nevertheless, much of it can be applied to virtual worlds (and to single-player computer role-playing games, come to that).

There are two key aspects to the theories: mode and stance.

Mode is the style of play that players adopt, and concerns the reason that they participate in role-playing games. The usual answer they give—“to have fun”—is the same that players of virtual worlds reply. The RPG categorization of the answers, however, is different to the one I derived for virtual world player types. In fact, there are two main approaches: The GDS and GNS models, named after the initial letters of their categories (Gamist, Dramatist, Simulationist, and Gamism, Narrativism, Simulationism). GDS came first, and is described in the rec.games.frp.advocacy FAQ ¹. GNS is a refinement by Ron Edwards ².

I’ll outline GNS here:

• Game mode. This concerns competition between players, rather than their characters. Players create victory/loss conditions for themselves that are reflected in their playing strategies.

• Narrative mode. Here, the idea is the creation of a story with a recognizable theme. Characters are protagonists, and the referee provides the narrative cloth from which the players cut the story. The GDS model has drama mode here instead.

• Simulation mode. This style of play focuses on exploration—of the character, setting, situation, atmosphere, and/or game mechanics. Internal logic and consistency of experience are regarded as paramount.

These playing styles are specific to individuals. The same game may have players in different modes, and conflicts will therefore occasionally arise (for example, game mode players might hire a half-orc tracker when their characters are all elves, whereas simulation mode players might object that their characters “wouldn’t do that”).

From a virtual world designer’s point of view, it’s interesting to compare these modes to the player types model (RPG theorists have done the reverse, too). There are some definite similarities: Game mode means achievers; simulation mode means explorers. Narrative mode could fit either socializers or killers, but serves neither very well. Why is this?

The root of the disparity is that in a virtual world, different characters do different things at the same time; in a human-moderated role-playing set-up, they don’t. Human beings can’t listen to instructions in parallel and react to each instantly; computers can, so they facilitate the person-to-person action and interaction that killers and socializers seek. Conversely, computers cannot adapt to unexpected actions to keep a series of planned events on track; humans can, therefore they can sustain a narrative.

In terms of mode, RPG theory has some relevance to virtual worlds; where it doesn’t apply, its limitations are clearly defined.

The second characteristic of role-playing theory, stance, concerns how individuals decide what to do in response to a role-playing event. There are four basic stances, although only the first three typically apply:

• Actor stance. Players decide what their characters do based solely on the knowledge that their characters have. I know there’s a dragon behind the door because I saw the referee’s notes, but my character doesn’t: My character goes through the door.

• Author stance. Players use their own priorities to determine what their character does, then retroactively ³ provide a rationale for their character to act that way. If we found the dragon this early, the gaming session would end too soon; the fact that my character is greedy enables me to send him back to look at some other doors he walked past, “before I get too far away.”

• Director stance. The player determines their character’s actions with no reference to that character’s knowledge or abilities. My character pulls out their secret, dragon-slaying sword and goes through the door.

• Audience stance. Events unfold, my character merely watches. This doesn’t apply in a role-playing situation: If I’m called on by the referee to act, my character must act—even if that means explicitly “doing nothing.” It can apply in virtual worlds in certain stylized situations (one of which is described in the section on drama theory that follows).

The director stance is rare among players of RPGs. It may be more common in virtual worlds where players have build privileges, but it’s impossible in those where they don’t. The important stances are the first two, which equate to in-character (IC) and out-of-character (OOC) action. Edwards suggests that stances may be linked to the concept of immersion, too, which seems reasonable.

I’ve mentioned GDS and GNS here, but they aren’t the only models in RPG Theory. Ever since the appearance of GDS, people have been creating their own player type models, almost invariably working to a threefold template. Brian Gleichman argues ⁴ that this archetype has become so ingrained that people don’t consider there may be better ways of doing things. It occurs to me that there may be a similar problem in virtual worlds following on from my own four-fold player types model.