Research on gender and language has typically been underpinned by the premise that women and men are essentially different,and that these differences are evident in how they talk.In this paper,I introduce—and argue in favour of—the conversation analytic(CA)approach to gender as an alternative to a gender-differences paradigm.I describe what I take to be the three main challenges that conversation analysts have levied against such an approach,which I summarise as:a question of evidence,a question of relevance,and a question of existence(or ontology).I then describe,and critically evaluate,the classic solution to these questions that conversation analysts have proposed:a focus on‘participant orientations’to gender.I suggest that—in addition to correcting some of the mistaken analyses of interactional phenomena evident in gender-differences research—CA can inspire new ways of thinking about old research questions.I also propose that the question of what counts as an orientation to gender and,relatedly,how best to analyse the taken-for-granted social world,provide particularly fruitful areas for future research.I conclude by calling on researchers and students of gender to avoid using the assumption of essential difference as the starting point for academic endeavour.