I want to thank Steven Taylor for his thoughtful comments on the volume Daniel Goldstein and I edited. In this response I will address two key points he raises that help to highlight some of the key differences in our approaches.

While Taylor recognizes that Daniel Goldstein and I have shown that “there are multiple types and sources of violence” operating in Latin America, he suggests that this is the “definition“of “failed democratization.” Taylor's argument reflects what Goldstein and I acknowledge are paradigmatic dissimilarities between our approach to understanding conflict in Latin America and those of the dominant democratic institutionalist approach to the region. While institutional scholarship offers substantial insights into the problems of violence, we argue that these approaches are excessively narrow in focusing on the limitations of policies and the inability of state institutions to live up to expectations, rather than on the multiple ways that both state and nonstate armed actors engage with one another to promote state policies, win elections, and create collaborative, though often violent, micro-level orders. Our approach recognizes deviations from democratic norms, but we have also sought to open up new spaces for thinking about these regimes from the more concrete perspective of how actors in these places develop and sustain persistent systems of governance involving multiple armed actors.

Taylor also points out what he sees as a gap between some of the case studies in the volume that examine policy failure and limited state capacity and arguments made by Goldstein and I that existing violence does not reflect state failure. These differences are more perceived than real. Resisting the language of state failure, as we do in the volume, is consistent with our acknowledgment, and that of our chapter contributors, of a complex terrain of, at times, limited state capacity, the reality good and bad policies, and more affirmative choices by state and nonstate armed actors to collaborate. Such collaboration leads to complex systems of political order better understood through the interactions among multiple armed actors that generate varied governance systems and structures of political subjectivity than by looking at the distance between democratic norms and the conditions of ongoing violence that characterize many places in Latin America. The two chapters that Taylor suggests differ from some of the arguments made in the introduction and conclusion, and elaborate ways of thinking about these orders in their own language and particular contexts. Indeed, in the volume's conclusion (pp. 244–45, 249–50, and 258–63), I explain in some detail the connections between these two chapters and our larger arguments.

The framework Goldstein and I develop has important analytical and policy implications. A deeper understanding of the connections between state and nonstate actors and the types of orders maintained by nonstate actors in the midst of the elective regimes that characterize the region today will offer scholars the opportunity to contemplate policy solutions that go beyond shoring up failed state institutions. In many cases, these strategies fail precisely because they do not acknowledge the links between state officials and armed organizations or the delicate networks and localized orders maintained by these armed groups. As a result these policies can contribute to more violence and more robust connections between state and nonstate armed actors. (See, for example, José Miguel Cruz, “Central American Maras: From Youth Street Gangs to Transnational Protection Rackets,” Global Crime 11 [November, 2010]: 392). Developing cogent and responsive policies involves understanding the limitations of state power as well as its projection through illicit engagements between state actors and social actors, understanding how to reorient those structures of power, and empowering civic actors seeking to control conflict amid diverse systems of often violent informal control.