This paper uses a novel computational method to spatially join 1921 Polish demographic census data to contemporary Polish administrative boundaries. I analyse patterns in the geographical variation of the 1921 Jewish community, finding that support for PiS in the 2015 and 2019 elections was concentrated in formerly Jewish districts. This relationship was strongest in rural areas. I argue that this stems from higher levels of out-group intolerance in these districts, brought on by cognitive dissonance driven value-adjustments in response to inter-ethnic competition. I will present my novel data collection strategy, outlining how it can be used in further research, and the challenges of causal identification in latent historical processes.
