Following the December 2004 Indian Ocean tsunami, speculations arose in policy circles as to how this would influence the ongoing conflicts in Sri Lanka and Aceh. Both conflicts had seen decades of violent fighting between separatist rebel groups and government forces with thousands of casualties on all sides. In Aceh, the Free Aceh Movement had been fighting an ethnicnationalist guerilla war for independence from Indonesia since 1976. Sri Lanka has been involved in a similar conflict since 1983, pitting the Tamil Tigers against the Sinhalese dominated government and army in a guerilla struggle for an independent Tamil state. Both conflicts harbored deep mistrust between government and rebel forces – yet both areas were also devastatingly impacted by the 2004 tsunami, leaving tens and hundreds of thousands dead, and many more displaced.

In its immediate aftermath, optimism soared that the disastrous impact could be a blessing in disguise, an opportunity for conflicting parties to unite and foster reconciliation (Stokke, 2005). Surprisingly, the two conflicts developed quite differently in the months following the tsunami. In Aceh, peace negotiations quickly initiated and, in August 2005, a peace deal was signed between the government and GAM, resulting in the withdrawal of security forces and disarmament and demobilization of rebel fighters (ICG, 2005a). In Sri Lanka, however, initial cooperation between government forces and Tamil Tigers after the tsunami soon gave way to a breakdown of the preexisting ceasefire, and, at the time of the Aceh peace deal, the Sri Lankan conflict was resuming a level of hostilities not seen in three years (ICG, 2006).

This paper addresses the puzzle of why two seemingly similar conflicts affected by the same disaster evolved in two opposite directions. In doing so, it places itself within the more general literature on the political fallout of disasters. Two main arguments are generally presented in this literature. First, that disasters can foster political change (Birkland, 1998; Prater & Lindell, 2000) by creating a window of opportunity for peace (Kelman, 2003), or, less fortunately so, by aggravating existing political unrest (Drury & Olson, 1998). Second, that the political results from disaster largely reflect pre-disaster trends and contexts (Hoffman & Oliver-Smith, 2002; Lindell & Prater, 2003). In other words, post-disaster outcomes will tend to represent an acceleration or amplification of already existing dynamics.