Taking the former Yugoslavia and some of its successor states as a case study, this paper explores the notions of ethnicity and mother tongue (with a side glance at religion) as used in recent population censuses. A special focus is on the sometimes important disparities between the ethnic and linguistic affiliations of the respondents, a phenomenon often ignored but possibly opening the door to manipulation of census returns in the service of specific interests. It is displayed that, while having distinct rubrics for the two features is in itself a helpful methodological tool, their mutual connection should be carefully explored rather than plainly assumed.