We present a novel method which can be used to show that contemporary party systems may originate much further back than is usually assumed or might be expected - in reality many centuries. Using data on Republic of Ireland and Northern Ireland, jurisdictions with party systems that differ significantly, one of which poses significant challenges to the universality of many political science theories, we show how using surnames as markers can confirm the obvious explanation for the Northern Ireland party system and then propose a novel explanation for the Irish party system. We suggest that surnames could be an objective way of studying migration patterns and ethnic heritage which may be important in explaining party systems. © The Author(s) 2012.