Doctoral Dissertation, The Australian National University, May 2024. http://hdl.handle.net/1885/317367

Non-decisive outcomes of conflict—instances where violence ends without a clear victory or formal settlement—are the most frequent and yet understudied scenarios in civil wars. The study of war termination and rationalist bargaining approaches typically relies on a 'Win, Lose, or Draw' framework. Yet this approach fails to capture the more complex processes of conflict ending without the clear symbols of a war’s termination. This omission limits the theoretical understanding of the bargaining process to reach non-decisive outcomes and narrows the perceived range of rational options available to conflict actors, particularly those seeking an off-ramp when neither victory nor a negotiated settlement is attainable. Addressing theoretical and empirical gaps, this article makes four contributions to bargaining and civil war outcomes literature: I) it conceptualizes non-decisive conflict outcomes with a typology; II) it refines rationalist bargaining approaches by accounting for unique commitment challenges in civil wars that explain why actors may avoid negotiations and fighting; III) it illustrates the variation of non-deceive outcomes through a case study on Myanmar’s armed conflicts; and IV) it advances theoretical explanations for previously overlooked empirical cases of intrastate conflict outcomes—covering long wars, mutually beneficial stalemates, and the transformation of conflict actors—helping scholars and policymakers more effectively recognize and study these alternative endpoints.

How does rebel fragmentation influence the outcomes of internal armed conflicts? This study examines ways multi-actor conflict shapes prospects for war termination, including peace agreements, rebel victory, government victory, and non-decisive outcomes. Two mechanisms of fragmentation are assessed—rebel splintering, when new rebel groups emerge from an existing organization, and overlapping conflict in which more than one original rebel group is actively fighting. This study uses random-effects multinomial logistic regression to analyze conflicts from 1946 to 2012 and the effects of fragmentation. Findings demonstrate that fragmentation shapes civil war termination by reducing the chances for decisive military victories and settlements and increasing the likelihood of non-decisive scenarios, which range from gradual termination through rebel disbandment, rebel mergers to create new groups, and low-activity or no violence through conflict management, with states and rebels entering into mutually beneficial stalemates. Splintering reduces the odds of peace agreements by 59 percent, while both splintering and overlapping conflict sharply decrease the chances of government victory, by about 73 percent and 65 percent, respectively. Overlapping conflict further reduces the likelihood of rebel victory by a factor of nearly 100 when compared to non-decisive outcomes. Peace settlements emerge as the second most common outcome—often through a series of partial agreements—representing a more constructive alternative to long-term conflict management, given the persistent risks of violence escalation where rebel groups have not disarmed or continued rebel proliferation when conflict causes remain unaddressed.