Built-in Safeguards and the Implementation of Civil War Peace Accords

ABSTRACT

This article contributes to analyses of peace agreement implementation by focusing on the role of built-in safeguards as procedural mechanisms within peace agreements. Recent empirical studies suggest that negotiated peace agreements are a frequent mode of armed conflict termination and implementation of those agreements to be the primary predictor of enduring peace and the quality of that peace. However, in many instances implementation takes years or even decades and is subject to breakdown before achieving the key implementation objectives. In this article, we identify and theorize three key safeguards in peace accords: transitional power-sharing, dispute resolution, and verification mechanisms. We argue that these safeguards lead to higher implementation of provisions negotiated in peace agreements by addressing mutual suspicion and by facilitating more constructive working relationships among former rivals. To test our arguments, we analyze implementation of comprehensive peace agreements negotiated between 1989 and 2012 from the Peace Accords Matrix Implementation Data (PAM_ID). We find that the built-in safeguards that we have identified significantly explain variations in levels of peace agreement implementation.

KEYWORDS:

• Built-in safeguards
• civil war peace agreement
• levels of implementation
• peace accords matrix

Notes

¹ DeRouen et al.’s study (2010) examines implementation of agreements in Northern Ireland, Aceh, Burundi, Mali, and Somalia.

² Transitional powersharing mechanisms in this article should be distinguished from long-term powersharing plans that aim to develop a new polity recognizing and reflecting the different interests and perspectives of various social groups. Instead, the transitional powersharing in this study denotes the establishment of a national unity or powersharing government representing the main parties involved in the negotiation process and ceases to exist once postconflict elections are held.

³ The effectiveness of powersharing government has been questioned in some studies (see Mehler 2009; Sriram and Zahar 2009; Roeder and Rothchild 2005; Tull and Mehler 2005). These studies suggest that powersharing stifles the development of democratic institutions and is not likely to stop the cycles of violence as those who are in power would like to stay in power and those outside would engage in violence to gain access to power.

⁴ The PAM project provides texts of peace agreements and provides annualized qualitative narratives as well as implementation levels on an annualized basis (Joshi et al. 2015). Table A1 in the appendix presents a list of provisions. Also see http://peaceaccords.nd.edu/sites/default/files/PAM_ID%20CODEBOOK%20V.!.5%2029July2015.pdf.