2 VIVIEN A.SCHMIDT better at explaining continuity than change.In all three frameworks for analysis, institutions serve primarily as constraints.RI focuses on rational actors who pursue their preferences following a 'logic of calculation'within political insti- tutions,defined as structures of incentives;HI details the development of political institutions,described as regularized patterns and routinized practices subject to a 'logic of path-dependence';and SI concentrates on social agents who act according to a 'logic of appropriateness'within political institutions,defined as socially constituted and culturally framed rules and norms.Because of their definitions of institutions,which they present largely as given,static,and con- straining,neo-institutionalist scholars in all three neo-institutionalisms have mainly explained change as coming from the outside,as the result of exogenous shocks.It is only relatively recently that they have sought to endogenize institu- tional change,by looking to explain the origins of or shifts in interest-based preferences,historical paths,or cultural frames.Although some such scholars have sought to do this while remaining within their older neo-institutionalism, others have turned to ideas and discourse.Just how many have done so in each of the three neo-institutionalisms varies considerably,however,with relatively few scholars working in the RI tradition,more in the HI tradition,and the most in the SI tradition.But importantly,all those who have come to take ideas and discourse seriously have broken with some of the fundamental presuppositions of their own institutionalist tradition at the same time that they have come to share enough in common to be identifiable as part of a fourth new institutionalism.And what they share includes not only an analytic framework but also a commitment to go beyond politics as usual'to explain the politics of change,whether this means the role of ideas in constituting political action,the power of persuasion in political debate, the centrality of deliberation for democratic legitimation,the(re)construction of political interests and values,or the dynamics of change in history and culture. I name this fourth new analytic framework 'discursive institutionalism'(DI)to call attention to the commonalities among the wide range of scholars who use ideas and discourse to explain political change (and continuity)in institutional context (see also Schmidt,2002:Ch.5,2006:Ch.5,2008).And I label all scholars who take ideas and discourse seriously as 'discursive institutionalists', whether they themselves use this term,prefer instead to classify themselves within one of the older institutionalist traditions,or have themselves come up with a different,more specific term to identify their particular take on ideas and/or discourse.Among these latter scholars,some focus primarily on the ideas side of the framework,calling their approach the 'ideational turn'(Blyth,2002),dis- cursive institutionalism (see Campbell and Pedersen,2001),ideational institu- tionalism (Hay,2001),and constructivist institutionalism(Hay,2006).Others emphasize the discourse side,whether calling it discourse analysis(Hajer,2003), the argumentative turn(Fischer,2003),or deliberative democracy(Dryzek,2000). Still others span the ideas/discourse divide,much like DI,whether through approaches focusing on the 'referentiel'(frame of reference)(Jobert,1989;Muller,better at explaining continuity than change. In all three frameworks for analysis, institutions serve primarily as constraints. RI focuses on rational actors who pursue their preferences following a ‘logic of calculation’ within political insti￾tutions, defined as structures of incentives; HI details the development of political institutions, described as regularized patterns and routinized practices subject to a ‘logic of path-dependence’; and SI concentrates on social agents who act according to a ‘logic of appropriateness’ within political institutions, defined as socially constituted and culturally framed rules and norms. Because of their definitions of institutions, which they present largely as given, static, and con￾straining, neo-institutionalist scholars in all three neo-institutionalisms have mainly explained change as coming from the outside, as the result of exogenous shocks. It is only relatively recently that they have sought to endogenize institu￾tional change, by looking to explain the origins of or shifts in interest-based preferences, historical paths, or cultural frames. Although some such scholars have sought to do this while remaining within their older neo-institutionalism, others have turned to ideas and discourse. Just how many have done so in each of the three neo-institutionalisms varies considerably, however, with relatively few scholars working in the RI tradition, more in the HI tradition, and the most in the SI tradition. But importantly, all those who have come to take ideas and discourse seriously have broken with some of the fundamental presuppositions of their own institutionalist tradition at the same time that they have come to share enough in common to be identifiable as part of a fourth new institutionalism. And what they share includes not only an analytic framework but also a commitment to go beyond ‘politics as usual’ to explain the politics of change, whether this means the role of ideas in constituting political action, the power of persuasion in political debate, the centrality of deliberation for democratic legitimation, the (re) construction of political interests and values, or the dynamics of change in history and culture. I name this fourth new analytic framework ‘discursive institutionalism’ (DI) to call attention to the commonalities among the wide range of scholars who use ideas and discourse to explain political change (and continuity) in institutional context (see also Schmidt, 2002: Ch. 5, 2006: Ch. 5, 2008). And I label all scholars who take ideas and discourse seriously as ‘discursive institutionalists’, whether they themselves use this term, prefer instead to classify themselves within one of the older institutionalist traditions, or have themselves come up with a different, more specific term to identify their particular take on ideas and/or discourse. Among these latter scholars, some focus primarily on the ideas side of the framework, calling their approach the ‘ideational turn’ (Blyth, 2002), dis￾cursive institutionalism (see Campbell and Pedersen, 2001), ideational institu￾tionalism (Hay, 2001), and constructivist institutionalism (Hay, 2006). Others emphasize the discourse side, whether calling it discourse analysis (Hajer, 2003), the argumentative turn (Fischer, 2003), or deliberative democracy (Dryzek, 2000). Still others span the ideas/discourse divide, much like DI, whether through approaches focusing on the ‘re´ferentiel’ (frame of reference) (Jobert, 1989; Muller, 2 VIVIEN A . SCHMIDT