Build It to Understand It: Ludology Meets Narratology in Game Design Space

Mateas and Stern's paper argues that building experimental games can advance game studies beyond the analysis of existing titles. Their central thesis is that construction yields theoretical and design insights that purely analytical methods cannot reach. They ground this argument in the stalled ludology versus narratology debate, specifically the question of whether high player agency can coexist with strong narrative. The authors contend that ludologist skepticism partly reflects the absence of successful agency-rich narratives, and they use their interactive drama Façade as a test case.

The paper's account of Façade combines design description with preliminary evaluation. The system recasts story interactions as abstract social games, including affinity, hot-button, and therapy games, driven by natural-language discourse acts and organized through a layered architecture and drama manager that coordinate multiple narrative fronts. The evaluation finds that responsive local agency came at the cost of global agency: only 27 beats were authored, a limit attributed to the substantial time required for authoring. Communicating complex social game state to players without numeric displays also emerged as a major unresolved challenge.

Drawing on Rittel and Weber's concept of "wicked problems," the authors argue that game design space can only be meaningfully explored through the construction of experimental games, since analysis of existing titles cannot anticipate new regions of possibility. On this basis they conclude that Façade's playable results make ludologist assumptions about the incompatibility of narrative and agency appear premature. The scope is a single system and its design lineage, situated within AI-driven interactive narrative and game design theory, with engagement with scholars including Frasca, Murray, Juul, and Church.