‘‘Anecdotal evidence’’ has become a central point of contention in two recent controversies over science and technology (We follow Nelkin (1992, 10) in referring to our cases as controversies over science and technology.) in the United Kingdom and a contact point between individuals, expert institutions, and policy decisions. We argue that the term is central to the management of the boundary between experts and nonexperts, with consequences for ideas of public engagement and participation. This article reports on two separate pieces of qualitative social research into recent UK public risk controversies with the aim of unfolding the processes by which anecdotal evidence comes to be defined. We do not define anecdotal evidence as an epistemic category that exists prior to the public controversies themselves; rather, we aim to show how the term is constructed and contested by expert and nonexpert actors. We find that anecdotal evidence comes to be accepted (albeit in different ways) by the main actors as an epistemic category, yet that it is multidimensional, open to interpretation as subjective reports, as an indicator of expert ignorance, as a source of novel hypotheses and as a set of political claims for recognition and inclusion. We conclude that the flexibility of anecdotal evidence at the boundary between science and its publics can offer opportunities for participation and engagement, as well as exclusion and alienation.