Is ethnic separatism the inevitable consequence of pursuing policies that allow for a reflowering of subnational ethnic identities, as in Quebec, or are there ways of having both a strong sense of attachment to one's own while still fostering loyalty to the larger state, as some variants of pluralist theory would have it? This is the central research question guiding our comparative study of the relationship between attachment to the individual ethnic group and loyalty to the larger country. Research on the relationship between strength of ethnic attachments and loyalty to the country as a whole impinges on the political wisdom of choosing public policies from affirmative action, to bilingual education to political autonomy for subregional groups. The comparative politics literature is fraught with assumptions about the nature of this relationship, but few studies have tried to empirically estimate it. Drawing on research by de la Garza et nl. (1996) and Sidanius et nl. (1997), we test pluralist, melting pot, and ethnic dominance models of ethnic attachment and overall levels of patriotism in the US and four other polyethnic states. Our data are derived from a 1995 ISSP National Identity Survey and the 1990-93 World Values Survey. We find mixed support for the alternative models when we replicate Sidanius and de la Garza and call for greater focus in cross-national surveys on assuring adequate samples of minority groups so that extant theories can be tested more fully.