Book excerpt: As in the cascading of water, violence and nonviolence can cascade down from commanding heights of power (as in waterfalls), up from powerless peripheries, and can undulate to spread horizontally (flowing from one space to another). As with containing water, conflict cannot be contained without asking crucial questions about which variables might cause it to cascade from the top-down, bottom up and from the middle-out. The book shows how violence cascades from state to state. Empirical research has shown that nations with a neighbor at war are more likely to have a civil war themselves (Sambanis 2001). More importantly in the analysis of this book, war cascades from hot spot to hot spot within and between states (Autesserre 2010, 2014). The key to understanding cascades of hot spots is in the interaction between local and macro cleavages and alliances (Kalyvas 2006). The analysis exposes the folly of asking single-level policy questions like do the benefits and costs of a regime change in Iraq justify an invasion? We must also ask what other violence might cascade from an invasion of Iraq? The cascades concept is widespread in the physical and biological sciences with cascades in geology, particle physics and the globalization of contagion. The past two decades has seen prominent and powerful applications of the cascades idea to the social sciences (Sunstein 1997; Gladwell 2000; Sikkink 2011). In his discussion of ethnic violence, James Rosenau (1990) stressed that the image of turbulence developed by mathematicians and physicists could provide an important basis for understanding the idea of bifurcation and related ideas of complexity, chaos, and turbulence in complex systems. He classified the bifurcated systems in contemporary world politics as the multicentric system and the statecentric system. Each of these affects the others in multiple ways, at multiple levels, and in ways that make events enormously hard to predict (Rosenau 1990, 2006). He replaced the idea of events with cascades to describe the event structures that 'gather momentum, stall, reverse course, and resume anew as their repercussions spread among whole systems and subsystems' (1990: 299). Through a detailed analysis of case studies in South Asia, that built on John Braithwaite's twenty-five year project Peacebuilding Compared, and coding of conflicts in different parts of the globe, we expand Rosenau's concept of global turbulence and images of cascades. In the cascades of violence in South Asia, we demonstrate how micro-events such as localized riots, land-grabbing, pervasive militarization and attempts to assassinate political leaders are linked to large scale macro-events of global politics. We argue in order to prevent future conflicts there is a need to understand the relationships between history, structures and agency; interest, values and politics; global and local factors and alliances.