In this paper, we reflect on the use of fictional source material and fictional formats in organization studies in order to explore issues of responsibility in the writing of research. We start by examining how research using fictional narrative methods has worked to radically destabilize distinctions between what is real and what is fictional. In relation to this, we ask the question: if a research account can be regarded as fiction, what are the implications of this insight for the responsibilities of authors? Opposing the view that using fiction necessarily leads to an 'anything goes' relativism, we argue that a recognition of the fictionality of research texts implies a heightened sense of researcher-author responsibility. We see our main contribution as extending existing discussions of reflexivity in research into a consideration of issues of ethics and responsibility as it relates to the textuality of research writing. To do so, we draw on Derrida's theorization of responsibility and undecidability as a way of problematizing and discussing the ethics of research in relation to its textuality. We argue that the explicit borrowing from fictional genres evinces the essentially 'written' and fictional status of research papers, and highlights the ethical dimensions associated with decisions related to representational strategies and authorial subjectivity.