This thesis is about how the concept of scene can be used to understand the relationship between interactions and discourses in various facets of human musical and social life. Through a case study of punk in Washington D.C. in the mid-1980s, it develops a new diagrammatic method for representing scenes, as well as a new approach to analyzing scenes in which diagrammatic and narrative representations are read together. Chapter One asks what the concept of scene can do, showing that to call something a scene is a productive act. The term scene originated in colloquial speech, where it is used to group and elide practices, people, and objects, describe the patterns of interaction and influence that characterize them, and discursively construct these actors and interactions as a whole. The chapter goes on to describe perspectives by three authors that have shaped the academic use of scene: participation and identification (Irwin), cultural production (Shank), and the global circulation of local music practices (Straw). Finally, the chapter relates the colloquial and academic conceptualizations of scene, focusing on the tension between interaction in scenes and the discursive construction of the scene as an axis of inquiry in contemporary scholarship. Chapter Two is an in-depth case study of the Washington D.C. punk scene at the time that it gave rise to two genres, hardcore punk and emo music. The goal of this chapter is to develop a new approach to representing and analyzing scenes, centered on the productive tension between the interactions that the term scene describes and the scene’s construction as a discourse. The chapter begins by analyzing D.C. punk as a model scene with respect to the scene perspectives of Irwin, Shank, and Straw. Then possible methods for representing the scene, as well as the limitations of existing approaches, are discussed. The remainder of the chapter centers the development and application of the scene diagram, a new approach to representing scenes. These diagrams are able to communicate the wealth of quantitative information that scenes generate simultaneously, including how musicians interact within and between bands. In particular, scene diagrams can show patterns of change among all bands in the scene at once, for example whether they formed or broke up in a given time period. Their construction also enables the diagrams to be read together with narrative representations of the scene, enabling a new analytic approach. The two scene diagrams in this thesis depict the D.C. punk scene in 1983-84, when the discourse about the scene suggests that the scene was disintegrating, and in 1985-86, when the discourse was one of rebirth. The picture of the scene that emerges through the diagrams does not correspond directly to the discourses. Instead, reading the diagrams together with a narrative shows that the discourses focus on a particular cohort: the creators of hardcore and the subset of this cohort that created emo music in summer 1985. This "Revolution Summer," as it is known, emerged from cultural conflict between different cohorts in the scene. This conflict helps explain how two genres were created in similar networks in a short time, as one cohort sought to distinguish itself from others. The case of Revolution Summer also suggests that such conflicts are one way that new scenes can emerge from other scenes. Overall, the diagrams make the interactions and theoretical tensions of scene newly visible, offering a mode of understanding that can be applied beyond music to many social phenomena.