This paper presents the results of a qualitative micro-study of a 3-minute spoken interaction between a research participant and a researcher. The talk in the interaction concerns the past of the contemporary Polish town of Oswiecim, internationally better known as Auschwitz. Borrowing methods and concepts from linguistic ethnography and interactional sociology, the article demonstrates that people know different narratives about the same past event and are able to move between those narratives when the interactional context requires them to. The combination of micro-discourse analysis with ethnographic detail provides an insight in the flexibility of the remembering self in interpersonal interaction. It highlights the potential to reconceptualise aggregated forms of memory, be they cultural or collective, as less stable as they were hitherto assumed to be. The 'frames' (Irwin-Zarecka 1994) or 'schemata' (Erll 2011, 2014) that people employ to attribute meaning to the past are multiple, everchanging and constantly being re-actualised in everyday interactions. The findings and methodological framework presented in this paper also engage in a dialogue with some fundamental critiques on the field of memory studies. These include, among others, the need to connect the micro, meso and macro, the individual with the social (Kansteiner 2010, Gensburger 2016, Keightley, Pickering, Bisht 2019) and the urge to actively develop and think through methods in memory studies research (Kansteiner 2002, Roediger and Wertsch 2008, Keightley and Pickering 2013).
