The Geography of Military Losses and This paper argues that taking both time and space into account can provide new insight on the multiple processes underlying opinion change in today’s complex information environment. We present a case study of the “proximate casualties” hypothesis (Gartner and Segura 2000; Gartner, Segura, and Wilkening 1997), the idea that popular support for American wars is undermined at the individual level more by the deaths of American personnel from nearby areas than by the deaths of those from far away. The novel contribution of this study is to model the dynamics of public opinion (1) over time and (2) across space while accounting for (3) the influence of competing sources of relevant information about war losses and (4) features of each respondent’s local information environment that might alter information costs and the nature of opinion change. The key to sorting out this tangle of competing influences at multiple levels lies in our ability to precisely model in both time and space the appearance of key facts presumed to undermine support for the president’s handling of the Iraq war: the date of death and hometown of every American military casualty that occurred from the start of the Iraq invasion through the 2004 general election. Our analysis yields a consistent finding: the influence of American deaths on public support for the president’s handling of the Iraq War is shaped less by exposure to local or national news coverage than by the social context in which the respondent is situated. Casualties matter, but the mechanism of influence appears to be social networks rather than direct exposure to news coverage that might carry information about war costs.