No abstract

Proponents of democratization often claim that liberal institutions have a palliative effect on the level of conflict within societies. Critics, however, suggest that the instruments of democracy, especially elections, can spark political violence, particularly in weakly institutionalized settings. Using the newly available Social Conflict in Africa Database (SCAD), we examine the relationship between executive elections and social conflict in Africa for the period 1990-2009. We also assess the conditions which make elections more or less violent. We examine elections in (1) countries faced with armed conflict, (2) post-conflict settings, (3) elections in autocracies, and (4) in relatively poor countries. We also look at characteristics of elections themselves, including the margin of victory, the presence of observers, and allegations of vote fraud. Results show that while elections can sometimes spark violence, free and fair elections in genuinely democratic contexts are much less conflict prone, while illiberal elections are especially problematic. We do not find that current or recent armed conflict on a country's territory makes elections more violent.

Since the end of the Cold War, walls, fences, and fortifications have been constructed on interstate borders at a rapid rate. It remains unclear, however, whether these fortifications provide effective security. We explore whether border fortifications provide security against the international spread of violent militancy. Although barriers can reduce the likelihood that militant activity diffuses across international borders, their effectiveness is conditional upon the roughness of the terrain on which they are built and the level of infrastructure development in their proximity. Barriers require intensive manpower to monitor and patrol, and so conditions like rough terrain and poor infrastructure render security activity more difficult. However, rebels and other militants prefer to operate in such difficult areas, ultimately reducing the effectiveness of barriers in containing the international spread of violent militancy. Analyses on newly collated data on interstate border fortifications within a global sample of contiguous-state directed-dyad-years show that border fortifications are only effective in limiting the diffusion of militancy in contexts in which states can plausibly monitor and police their borders. This paper has significant implications for the academic literatures on national security and intrastate conflict, and it also speaks to the broader policy debate over border walls and fences.

No abstract

Much research disputes the president's ability to lead public opinion and shows media to have influenced public opinion concerning the war in Iraq. We argue that although news tone is likely to have affected public support for the war, presidential rhetoric could be influential for two reasons. First, heightened presidential attention to the war increases the public's accessibility to the president's perspective on the war. Second, a survey question that cues the respondent to consider the president explicitly in their evaluation of the Iraq war is likely to encourage responsiveness to presidential rhetoric. To assess these arguments, we simultaneously examine the impact that presidential tone and media tone have on public support for the war in Iraq by analyzing an original dataset of presidential speeches, news coverage, and public support for the war and the president's handling of it from 2002 to 2008. Our findings reveal that although media tone drives public support for the war in Iraq, presidential tone influences the public's view of President Bush's handling of it.