Mechanical Dances

IN-2: Completeness and clarity


Let’s start with the assumption of completeness and clarity. (In what follows, I am going to use the terms “narrative” and “story” more or less interchangeably—I will discuss the nuances in a later entry.)

It seems obvious and uncontroversial to say that interactive narratives like all satisfying stories must have a beginning, middle, and end. The question is what that means—in particular what kind of an ending must a story have. There is always the issue of cultural difference. Do different (historical or contemporary) cultures have different notions of what makes a story complete?
But we don’t have to look that far. Even within European and American culture in the past 100 years, there have been many attempts to call into question the notion of a complete and satisfying narrative.  Practically every literary, dramatic and cinematic avant-garde in the 20th century has offered alternatives to the rounded, closed narrative form. The futurists and Dadaists presented plays and performances with incomplete or absurd narrative lines. Most experimental filmmakers from the surrealists of the 1920s  (e.g. Man Ray) to the structural filmmakers of mid-20th century (e.g. Frampton and Brakhage)  either mocked or ignored the narrative techniques of mainstream cinema. Video art of the 1970s and 1980s seldom aimed to create coherent narrative lines. Then there are all the literature experiments from the futurist poets to the lettrists to stream of consciousness authors to postmodern writers who in one way or other subvert closed narrative forms. Bertholt Brecht, whose own plays were quite accessible and in some cases even popular, adopted a style that worked against what he called “Aristotelian drama.”

The interactive narrative community ignores this history. There are works that might be said to be on the fringes of the IN community that do break the rules of completeness and clarity—for example, some Interactive Fiction games such as Adam Cadre’s Photopia and the work of the early hypertext movement (Joyce, Moulthrop, etc.). These experiments are not models for the work that is reported at conferences such as ICIDS, however.

The Interactive Narrative project has as its goal to proceduralize narrative: to create an algorithm that can author narrative texts.  The reader/player helps to shape the narrative by inserting herself into the procedural loop—in other words, she becomes part of the algorithm. The interactive narrative produced in this way is expected to have the dramatic arc and coherent plot of a Hollywood “three-act” film.