I’m pondering a bit more about the relationship between games and narratives. I’m prompted to write about my pondering due to the Koster post here.
Koster’s argument is that a narrative can serve a game’s feedback function, but only, basically, once. His conclusion is that narratives are not a very good — not a very sustainable — game mechanic. I agree with him, mostly. I’ve written previously about how narratives and games are a bit like oil and water in several respects. (But I’m broadening my views more recently.)
And, though the Koster post makes sense as written, it’s based on a very simple model of both games and narratives.
Koster’s narrative model is focused on the sequencing of events.
And his game model is a concoction of stimulus-response mechanisms and pattern-matching. This model presents itself as more about the pattern-matching than the stimulus-response mechanisms, but we know it’s all the same because there is no clear indication in the model as to why pattern-matching is fun: pattern-matching is fun, this model says, because it is — and “fun is the process of discovering areas in a possibility space” (from Theory of Fun).
This is unsatisfying in that pattern-matching is not universally fun. Certainly, some pattern-matching is more fun than some other. And, in fact, some pattern-matching is no fun at all. Pattern-matching folk like Koster admit this, but what they admit about it is this: Some patterns are too difficult to match, and others are too easy; those patterns are no fun. The patterns that are the most fun to match are the ones in the middle of being too difficult and too easy.
That makes sense, as far as it goes. Indeed, the patterns that are most fun to match are sort of like Goldilock’s bed: a bed with dimensions and complexities similar to the dimensions and complexities of Goldilocks. So, likewise, the patterns that are the most fun to match are those patterns with dimensions and complexities most similar to the dimensions and complexities of pattern-matchers. But there’s a problem: The patterns that are really the most fun to match are not simply those patterns with dimensions and complexities most similar to the dimensions and complexities of pattern-matchers, but patterns with dimensions and complexities most similar to the dimensions and complexities of the human condition of pattern-matchers. These latter sorts of patterns might qualify as patterns, but they also qualify as something else: beauty and art.
Stimulus-response and pattern-matching models don’t say too much about these latter and special sorts of patterns: the beauty and art sort. Or, when they do say something, they say something like this: Games can’t deliver these sorts of patterns. Koster says something like this: “[G]ame systems,” he says, “have a very limited emotional palette.”
Some people — some of the same ones — would then further say that games are inferior to those things that can deliver the beauty and art sort of patterns: things like films, novels, and narratives. (Roger Ebert, for instance, has said something very similar to this.)
And then there are some other people entirely — Marie-Laure Ryan in Avatars of Story, for instance — who say that both narratives and games can deliver beauty and art. Games and narratives may even be able to deliver these sorts of patterns simultaneously and together — a claim based on a very different model of games and narratives than is Koster’s claim. In Ryan’s view, narrative is less critically about the sequencing of events than it is about the construction of a narrative world, including both world building and world manipulation (cf Herman’s Basic Elements of Narrative).
All told, this is a more engaging claim than is Koster’s in that if a narrative provides feedback in a game through world building and world manipulation, then you no longer have to worry about the diminishing feedback of a repetitive narrative sequence; you can start to benefit from the more interesting and compounding feedback of a recursive narrative function.
Unfortunately, because this realization is really a very good one, it has led many narratologists, including Ryan, astray. Because narrative (or, more circumspectly, narrativity) can be a game mechanic, say the narratologists, game and narrative are compatible. They are sympatico.
But no. They are not.