Despite its prominence in comparative politics, historical institutionalism continues to be largely absent from the study of nationalism. Typically, when institutions are discussed in studies of nationalism they are described as instruments available for conflict management (O'Leary and McGarry 1994; Hechter 2000). The scholars that have recently brought the insights of historical institutionalism to the study of nationalism agree that institutions play a causal role in the construction and political mobilization of ethnic groups (Bertrand 2004; Lecours 2005). Once politicized, ethnic identities can have profound 'lock in' effects on a wide range of institutions such as the welfare state, party system, and constitution. The notion of path-dependency is employed to explain how institutions transform politics in a multinational state into nationalist politics, making institutional change in the opposite direction of the developmental pathway extremely difficult (Harty 2001). Change is, therefore, attributed to exogenous shocks or internal tensions that rupture institutions in a sudden and radical manner.These arguments are strong at explaining institutional stasis, but weak at explaining how nationalism also creates opportunities for subtle institutional change in the context of continuity. Institutional change often occurs in a gradual and piecemeal fashion, rather than wholesale rapid change as the exogenous shock model implies. I argue that there is a dynamic relationship between nationalism and institutional change that can only be understood with reference to the conceptual tools that Kathleen Thelen calls, "institutional layering and conversion" (Thelen 2004). The theoretical argument will be developed in the case of Québec. The survival of Québec's distinctive party system and welfare state is due to subtle transformations beneath the surface to bring the institutions in line with changing social and economic conditions. The argument has theoretical relevance beyond this single case, thus future research should focus on testing with appropriate comparative methods. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]