Abstract:
While content moderation began as a solution to online abuse, it has increasingly been framed as a source of abuse by a diverse coalition of users, civil society organisations, and politicians concerned with platform bias. The resulting crisis of legitimacy has motivated interest in more participatory forms of governance, yet such approaches are difficult to scale on platforms that lack bounded communities and designated tools to support collective governance. Within this context, we use a high-profile debate surrounding bias and racism in content moderation on YouTube to investigate how creators engage in meta-moderation, the participatory evaluation of moderation decisions and policies. We conceptualise the conversation that plays out across a network of videos and comments as aspirational platform governance, or the desire to influence content moderation without established channels or guarantees of success. Through a content analysis of 115 videos and associated online discourse, we identify overlapping and competing understandings of bias, with key fault lines around demographic categories of gender, race, and geography, as well as genres of production and channel size. We analyse how reaction videos navigate structural factors that inhibit discussions of platform practices and assess the functions of aspirational platform governance, including its counter-intuitive role in legitimising content moderation through the airing of complaints.