Constitutions are foundational documents in modern democratic regimes. Because of their character, constitutions are intended to last, and in consolidated Western democracies, constitutions have typically remained stable over time. Yet this has not been the case in South America, especially during the wave of elected left-of-center presidents in the region beginning in the late 1990s and 2000s. Since Hugo Chávez was elected President of Venezuela in 1999, ten countries of South America have elected left-wing governments. In all countries, leftist presidents expressed intentions to radically replace their constitution – i.e., adopt new constitutions drafted by popularly elected constituent assemblies. However, only three of these governments rewrote their constitutions (Venezuela in 1999, Ecuador in 2008, and Bolivia in 2009). Why did some left-of-center governments in South America adopt new constitutions while others worked within their existing constitutional frameworks? This dissertation argues that the conjunction of intraparty and institutional constraints—tied to the characteristics of a leftist president’s elected party and the intensity of mass protests anteceding their election—explains the adoption of new constitutions. Accordingly, constitutions are replaced when leftist presidents lead outsider parties and are elected in contexts of strong social protests. This is because outsider presidents are neither constrained by their party in seeking radical constitutional rewrites nor face substantial institutional restrictions, given mass protests’ weakening of horizontal checks on the executive. In contrast, constitutional rewriting does not occur when leftist presidents belong to insider parties, because insider parties enforce higher intraparty constraints on their party leaders. Nor does it happen in the absence of mass protests due to institutional constraints on the executive’s power. To test this argument, this dissertation relies on a comparative study of Argentina, Ecuador, and Paraguay. Overall, this dissertation tackles a critical empirical puzzle in contemporary South America: in what conditions do governments undertake a successful re-founding of the constitutional order? This work advances our understanding of the processes of constitutional change in relatively new democracies, the role of mass protests and political parties in constitutional politics, and the implications of new constitutions for democratic citizenship.