Tracing a Community of Practice:A Database of Early African American Race Film Marika Cifor (bio), Hanna Girma (bio), William Lam (bio), Shanya Norman (bio), Miriam Posner (bio), Karla Contreras (bio), and Aya Grace Yoshioka (bio) The definition of "race film," as many scholars have noted, is notoriously elusive. Any proposed criterion—an all-black cast, for example, or a black director, or an exclusively black audience—ultimately will fail as a singular defining factor, because, in reality, the race film industry was multifaceted, contentious, and ever changing. Filmmakers and audiences defined race film not by pronouncing criteria but by actively creating and negotiating the community. Historians of the race film industry have dealt with the difficulty of defining race film by discussing the problem at length, acknowledging that many films sit uneasily within any classification system. We discuss in this article the methodological, historical, and epistemological questions we faced as a research team in building a comprehensive database of the early African American race film industry, 1905–30. The database seeks to reconstruct a detailed history of silent films made for and by African Americans in the early twentieth century. Because of widespread racial discrimination, silent-era productions that fit the description of "race films"—that is, films designed for African American audiences—could be shown only in certain theaters or at certain times or in nontheatrical settings and were exhibited to racially segregated audiences. For these and other reasons, most of these race films received scant attention in the mainstream media. The actual film reels were not preserved in any systematic way, so few of these early race films have survived.1 Though scholarship on early race film thrives,2 until our project, a filmography of these early silent films had not been publicly available. Moreover, various spelling and orthographical errors have propagated through scholarship on early race film, so the creation of our database was an opportunity to verify the original spellings, dates, and—as far as possible—casts and production companies of these films. Race, as Stuart Hall and many others have shown, is a category that is both exceedingly fluid and unforgivingly rigid:3 fluid, because race functions in different ways in different places and times, and rigid, because, once applied, the label of a race ramifies through a person's life in infinite ways. The example of filmmaking in the first half of the twentieth century is no exception. Early-twentieth-century filmmakers experienced and articulated race in myriad ways, but the stark categories of black and white nevertheless had an undeniable role in shaping the film industry. This article demonstrates how a data-driven project should hold both realities in tension: the truth, on one hand, of people's varied and disparate lived experiences and the equal truth, on the other, of the real political, social, and legal implications of the way a person is classified. A database, no matter how well conceived, tolerates relatively little ambiguity. Therefore the media scholar who wishes to encode race must make and adhere to difficult, carefully considered decisions about how to classify people's lives. This project team, however, did not have this luxury. Because we were setting out to build the first publicly available, downloadable database of the race film industry, we had to decide if any given film was, in stark binary terms, simply in or out of our data set. The team discusses here how we dealt with this daunting challenge, linking our efforts to the broader question of how moving image data might be created and used with nuance and rigor by scholars, archivists, librarians, and students. In addition to the complexities of the subject matter itself, we faced a number of practical questions in our creation of this digital humanities (DH) project. We begin by discussing the details of the project and its specifications. In the next section, we turn to the difficulties posed by defining and classifying race films. Through film studies scholarship, we develop a definition of race film for the project and consider how that definition shaped our project. Finally, we turn to the many practical concerns we faced, including the selection of an [End Page 101] appropriate...