“The reason for separating the story as a different level of meaning from the narratives that express it is the fact that the same story may be expressed in many different narratives, either within the same medium or across different media” [my emphasis]

the implication that a unitary composition stage might proceed a distributed cross-media publication stage is an interesting one.

In his GDC 2005 talk “Interactive Narratives Revisited: Ten Years of Research”, Ernest W. Adams reviews his description of “three problems for interactive storytellers” from GDC 1995 - internal consistency (choice v characterization), narrative flow (choice v pacing), and amnesia (player v character knowledge). At that time Adams rejected interactive narrative as an incoherent buzzword, calling instead for game developers to create “playgrounds for the mind.” Now Adams surveys the emergence of narrative elements over the past ten years in game genres including rail shooters, action-adventure hybrids, and strategy games with hero units. Revisiting the problems, he suggests replacing amnesiac characters with carefully crafted introductions, and recognizes the partially-constrained hero as sufficient for internal consistency given that “players don’t want to violate a character’s essential nature anyway.” Still, Adams sees these strategies along with the general “resurgence of linearity” as sidestepping key challenges in interactive narrative, and he blames lack of progress on both unproductive critical vocabulary (talk of conflict or dramatic tension ignores that gameplay tension allows repetition and randomness) and infighting in academia (ludology v narratology turf wars and postmodern politics). Criticizing over-reliance on Joseph Campbell or on the Aristotelian three-act structure, Adams also deplores trivial themes and emotions in games, and responds with ten quality guidelines whose bottom line is “interactivity is not an excuse for bad writing.” On the more positive side, he covers Marc LeBlanc on emergent narrative (including caveats on quality) and MMOG storyworlds (including caveats on lack of roleplaying). He also credits productive academic researchers including Janet Murray, Henry Jenkins, Joseph Bates, and in particular Craig Lindley’s paper on structural plot generation based on Proppian analysis. Adams concludes by listing research projects including the Erasmatron, Zoesis, Extempo, and Façade, describing his own vision of the future is an “AI dungeon master.”

Reaction:

I enjoyed Adams’ talk and its focus on the utility of theories from a developer’s perspective. It was interesting to read Adams on amnesia as a narrative device, since this has interested me throughout my own work on IF (although I never considered it a problem so much as a solution). I was also unfamiliar with Extempo… and Lindley’s paper, which contains several interesting resonances with our Benchmark Fiction initiatives in describing how “the same story may be expressed in many different narratives, either within the same medium or across different media.”

I like the attitude Adams takes towards utilitarian structuralism, and I generally enjoy his taste in sources and his enthusiasm for R&D projects (e.g. Façade). Given that he began by disclaiming that there was an explosion of research that he just hadn’t read, however, his “Academia Earns a B-minus” evaluation was a bit strange. My only other nitpick was with odd slips that seem to lack historical perspective - talking about 1995 comments that “anticipate” emergent narrative, or dating Propp’s work to the 1960s instead of the 1920s. If we want to talk seriously about what needs doing, it helps to be clear about what has already been done. Still, overall, Adams’ talk delivers a great summary of the field and the time from a development perspective, and is well worth reading.

Mark, are you familiar with Doug Church’s Gamasutra article on “Formal Abstract Design Tools”? Adams cites it discussing sports and narrative, and it reminded me of an earlier conversation of ours.