The Duration of Peace Following Civil War: A Study of the Importance of Security Guarantees and Institutional Arrangements

Led by Åshild Falch
Jan 2005 - May 2006

MA project at the University of Bergen, Comparative Politics
Supervisor: Michael Alvarez CSCW Supervisor: Scott Gates

When the belligerents cease firing at each other and a civil war is settled, what determines whether peace lasts or war resumes? What factors influence the probable durability of peace following settled civil wars, and what, if anything, can be done by the international community or the belligerents themselves to improve the chances of durable peace?
 
Not surprisingly, the cases in which the settlement of a civil war has contributed to a solution to the underlying conflict are exceptions rather than the norm. Although the enemies cease firing at each other and sign a peace accord, little or nothing by the conflict situation is usually changed, and the factors that triggered the former outburst of civil war are most likely still latent. Many post-civil war societies do thus continue to be haunted by the unresolved and war-related problems several years after the civil war was settled. Fear and mistrust about the adversary’s intentions often prevail, and credible commitments to the peace agreement become difficult. If the fear and mistrust developed during the war continue to be the prevailing feelings among the adversaries after the war has been settled, a wholehearted commitment to peace is unattainable. The most fertile way of improving the likelihood of peace is thus to make changes so that the security concerns of the contending parties will be addressed.
 
In the short run this may be done by international actors giving guarantees that each of the groups will be protected, that the terms of the agreement will be complied with, and that the promises formulated in the settlement will be kept. In post-civil war societies, such as in Cambodia, El Salvador, Mozambique, Namibia and Nicaragua, third parties have helped in ending military hostilities, defusing tensions, and creating stable security environments. Third-party actors do presumably hold a potential for reducing the security concerns of the adversaries after a civil war, and might hence be an effective conflict management tool, which could improve the prospects for stable and durable peace. In this thesis the focus is on the four following hypotheses about international assistance and durability of peace:
  • Post-civil war situations in which the international community intervenes to provide security guarantees to both/all adversaries are more likely to foster an enduring peace than situations where such intervention is absent.
  • The probability of lasting peace is greater in post-civil war societies in which the UN plays a central role than in such societies where the UN is totally absent.
  • Peace is more likely to prove stable in cases where the international community has an extensive mandate.
  • Post-conflict situations where substantial financial assistance is available are more likely to produce a durable peace than those where such assistance is absent.
For the peace to last the long run, on the other hand, much more involvement from actors within the post-civil war society is required. To improve the likelihood of durable peace, the post-civil war society most likely will need to opt for reconstruction of the central authority and development of institutions that may address the security concerns of the contending parties. The role of the state in the period of post-war reconstruction is hence believed to be crucial for the long run stability of peace. States can devise a number of different mechanisms to try to make peace easier to maintain. In this thesis, hypotheses about the effects of four such mechanisms are tested:
  • Peace is more likely to prove stable after extensively institutionalized settlements of the civil war than after settlements with no formal agreements.
  • Decentralization of power through territorial autonomy reduces the tension and contributes positively to the likelihood of lasting peace.
  • Provisions of executive power-sharing arrangements in the settlement of the civil war will increase the likelihood that parties will commit to an enduring peace.
  • Peace is likely to prove more stable if the civil war settlement provides rules for arms control, rather than if no such rules are provided.
The units of analysis in the project are states in which civil wars were fought in the post-World War II era. Intrastate conflicts that erupted between 1945 and 2000 were classified as civil wars if they met the criteria applied by Melvin Small and J. David Singer in the Correlates of War (COW) project. A total of 106 intrastate conflicts between 1945 and 2000 met the criteria of the COW definition, and this is hence the number of units that will be analyzed. The thesis has a quantitative approach, and makes use of a Cox proportional hazard model.

Members

Projects

Related pages