60 JUDITH BEYER to a pre-existing common sense (see Schuster 2003). Instead, interviewers, translators and asylum applicants are ‘doing procedure’ (Scheffer et al 2010: 8), thereby ‘literalizing’ what will later be scrutinised ‘for specific purposes of the legal establishment’ (2010: 7). In participating in this unequal encounter, all parties assume a ‘particular interactional status as “collector[s] of facts” or “evidence-maker[s]”’ (2010: 91). What could the role of a legal anthropologist be in such a setting? Anthropological Knowledge and Legal Categories Anthropologists sometimes find it difficult to use their knowledge in court settings, where the court imposes formal and strict parameters. My COI reports are written in the form of answers to specific questions that the claimants’ lawyers want answered. As in the hearing situation between the state representative and the applicant, I am therefore also involved in the preparation of documents that are used by the claimants’ lawyers to defend their case, but which I have to write for the court, not the side seeking my support. In my role as expert, I am therefore not party to the dispute but obliged to remain neutral. While colleagues already professionally active in this field before me helped me to acquire knowledge about asylum procedure, the genre of report writing and my putative role in it, I have tried to broaden this scope for interpretation and to identify gaps in the procedure within which I believe I can act effectively as an anthropologist. In addition to gaining insight into documents that I would otherwise not be able to access and that I evaluate from an academic perspective as I have done here, what fascinates me most about this expert role – and where I believe the greatest promise lies – is the opportunity to reflect their own categorising practices back to the lawyers and judges. This perspective refers less to factual knowledge about certain regions or ethnic groups or the plausibility of an individual’s biographical ‘story’, and more to anthropologically informed statements about the effectiveness and applicability of (legal) categories, especially when it comes to the question of the interpretation of these categories in the alleged ‘home state’ of the applicant. While caseworkers are bound to assess credibility according to indicators laid out by The European Union Agency for Asylum (2023), in my reports I am exclusively allowed to make a direct statement on the plausibility of an applicant’s account. However, COI experts are regularly asked what might happen if the asylum seeker were to be sent back to their presumptive home country. As states are obliged by international treaty not to expel asylum seekers to countries which they consider unsafe, where they could face structural discrimination, imprisonment, torture or even death, I emphasise in my reports how likely I consider it to be that, under the given uncertain (knowledge) circumstances, this person will be classified as Rohingya or as a member of another ethnic minority in the country of return, effectively sidelining the question of who they truly ‘are’. In most of those cases where I have argued that there is a ‘high probability’ that this will occur, it seems to have been convincing enough to overturn a decision by the Home Office or even a First-tier Tribunal. However, essentialised understandings of ethnicity (or ‘race’) continue to prevail in judges’ decisions. Even if I base my argumentation exclusively on the constructivist