Is elite or mass polarization the driver of democratic backsliding? A number of recent papers in the political economy of backsliding have developed theoretical predictions about the effects of polarization in the electorate as well as elite polarization on the process of democratic backsliding. However, when tested, these predictions focus on the polarization of political elites, typically within the same country. Cross-national data is ill-suited for the purpose of testing these predictions as we have no common yardstick against which to measure what is a ``dangerous'' level of polarization or how to compare countries that use different electoral systems and, subsequently, have different party systems and democratic institutions. Yet scholars for the most part have avoided even describing the emergence of mass polarization outside of the US. We offer a longitudinal analysis of Polish public opinion data that anticipates the country's downturn into authoritarianism to examine if the polarization of the electorate preceded or followed the polarization of party elites. It focuses on a single, yet ultimately divisive issue---EU integration and national sovereignty. Specifically, it makes use of surveys conducted in regular intervals by CBOS, the Center for Public Opinion Research. These surveys offer evidence that polarization in attitudes to the EU followed, rather than preceded the polarization of parties on this issue over the last 20 years.