Scholars frequently recount the ups and downs, the purportedly embarrassing – although always heroically turned to account – mishaps of research. However, acknowledging that one has rethought an analysis in the absence of new data makes explicit that social science is an interpretive project, and as such is rarely discussed in print. In this article I break that taboo, analyzing how I began to doubt my claim that a global shop floor was organized around an ungendered shop-floor subject. I then detail the more contextually sensitive reading of my fieldnotes that allowed me to grasp the fundamental masculinization of the shop floor in question. In the process, I theorize the aspects of gendered structure that enabled the error at the outset. Thus, the discussion reconstructs the life history of an argument – tracing the shifting development of analysis in a particular ethnographic case. In so doing, it follows epistemic problems back to their ontological roots, looking at how the tricky, obdurate situatedness of meanings – gendered and ungendered alike – requires an ongoing analysis of context in interpreting even our most minute and focused observations.