When analysing social and political beliefs we often do so from their relationship to the truth. We evaluate the claims made by figures like Donald Trump based on whether or not they're accurate, and in doing so assume that relation to the truth is central to their power. Hidden within these practices is the assumption that the way people relate to the world around them is empirical in nature. That in disproving them, we strip them of some credence or believability. However, if previous US elections are anything to go by, this simply isn't true. This talk proposes the idea of a degenerating myth; the narrative epistemological counterpart to Lakatos's degenerating research program. Following the line of Bruno Latour and Mary Midgley, I argue that much of the way humans understand the world is narrative in nature, rather empirical or rational. Consequently, when analysing these narratives we should assess them according to their own function, rather than equating them to the function of scientific research programs, which as I argue, we often do. To account for this difference in epistemological function, I propose a criteria for identifying degenerating myths, an evaluative framework not reliant on truth or external accuracy.