The paper examines relations between natural hazards and social conditions in disaster, and problems of their integration in disaster management. This must be done against a background of ever-increasing numbers of disasters. The initiating roles and impacts of environmental hazards are acknowledged. However, expanding losses are not explained by increased geophysical risks. To the extent that scientific knowledge or engineering and planning skills are involved, the problems seem more one of (in)effective deployment than major deficiencies. Social analyses suggest the scope of today’s disasters follows primarily from greater concentrations of vulnerable people, exposed in dangerous situations, and lacking adequate protections. Firstly, the question of disaster causality is revisited as a problem of damage diagnostics. A basis is developed from the findings of formal disaster inquiries. Despite their limitations, well-conducted inquiries offer unusually comprehensive anatomies of the social and physical conditions of disasters. They demonstrate and trace out the interplay of environmental, societal, technological, and institutional components of emergencies. In the examples described, environmental hazards are investigated in great detail. Nevertheless, societal preconditions are shown to be more critical. Inadequacies in emergency preparedness, performance, and post-disaster response are highlighted, and for those most at risk. The conclusions present major challenges for the agent-specific view of disasters, and for disaster management preoccupied with natural forces, uncertainty, and emergency responses. Rather, a view of disaster causality emerges emphasizing avoidable failures of preventive, protective, and intervention measures. Evidence is cited to show this is increasingly relevant in so-called natural disasters lacking such inquiries. The discussion considers the relevance of a preventive and precautionary approach in this context. The histories of accident, disease, fire, and crime prevention support arguments for greater attention to context-specific environmental and societal aspects of risk. Aligning disaster management more closely with preventive priorities depends upon a much greater focus on people, places, and livelihoods most at risk, reversing the social processes that put them at risk. It requires listening to their voices and concerns, recognizing and bolstering their resilience. Much more can and should be done to disseminate the protections, from building regulations to insurance, that actually do save so many others in the disasters that happen. As such, the case for greater attention to issues of governance and social justice is strengthened.