Politics in the shadow of the gun: revisiting the literature on ‘Rebel-to-Party Transformations’ through the case of Burundi

Abstract

This article provides a critical review of ‘rebel-to-party transformation’ scholarship. It shows how three flawed assumptions have underpinned much of the literature: (1) an ideal-typical differentiation between rebel group and political party as distinct by their use or rejection of violence; (2) the analysis of armed conflict as breakdown of ‘normal’ politics, and the study of ‘rebel-to-party conversions’ as a gradual, natural shift from violence back to politics; (3) a failure to integrate the study of rebel legacies into an examination of broader authoritarian legacies. These assumptions have clouded our understanding of politico-military organizations in conflict-torn societies, which combine social protest, armed rebellion, political violence, and party politics throughout their history. Drawing on the ‘no peace, no war’ and ‘armed politics’ paradigms, this article revisits these assumptions through the case of Burundi.

Acknowledgements

First and foremost, I want to thank all the Burundians that I have met during my fieldwork, who have helped me in discovering this beautiful country and its complex history. Second, I would like to thank the editors of this special issue, Johanna Söderström and Gyda Sindre, for their tremendous work in moving this project forward. I want to extend my gratitude to my PhD supervisor, Marie-Joëlle Zahar, for her candid and critical advice. Special recognition goes to Cara E. Jones, for her support and inspiration. I also want to thank René Lemarchand, Christophe Chowanietz, Jeremy Speight, Sule Tomkinson, and two anonymous reviewers for their insightful comments on previous versions of this article. Last but not least, I want to thank the Pierre Elliot Trudeau Foundation for its ideal and financial support.

Notes

1. The term ‘post-conflict’ is ambiguous. It usually refers to the formal end of hostilities stipulated in a peace agreement. However, peace accords do not necessarily coincide with the end of political violence. I therefore prefer to employ the term ‘post-accord’ (Zahar 2006).

2. PALIPEHUTU-FNL had to drop its ethnicist name to comply with the multi-ethnic power-sharing agreement stipulated in Burundi’s post-accord Constitution.

3. In Mozambique, the United Nations set up a donor trust fund with a contribution of $17 million to assist RENAMO to transform into a political party.

4. I do not posit that the waves are mutually exclusive. Instead, I use the wave analogy as a heuristic tool to summarize the evolution of the scholarship.

5. The UCDP/PRIO Armed Conflict Data Set (1946–2015) identifies the years 1965, 1991–1992, 1994–1999, 2003–2006 and 2008 as minor armed conflict (defined as resulting in 25 to 999 battle-related deaths) and 2000–2002 as major armed conflict (defined as experiencing at least 1000 battle-related deaths per year). This cursory overview illustrates that Burundi has experienced recurrent armed conflict during and beyond its ‘official’ civil war (1993–2008).

6. According to most sources, Burundi’s population comprises three ethnicities: 85 per cent Hutu, 14 per cent Tutsi, and 1 per cent Twa (Chrétien and Prunier 2003).

7. At the height of the war, it is estimated that FNL counted around 5,000 and FDD 16,000 fighters.

8. Various negotiation attempts culminated in the Arusha Peace and Reconciliation Agreement signed in 2000. Arusha stipulates an ethnic quota system for the security forces (granting equitable representation to Hutu and Tutsi) and political institutions (giving the Tutsi minority 40 per cent of posts in government and National Assembly, as well as 50 per cent in Senate, whereas the Hutu minority occupies 60 per cent and 50 per cent, respectively). Neither PALIPEHUTU-FNL nor CNDD-FDD ended up signing Arusha, and only accepted it later: CNDD-FDD in 2003 and PALIPEHUTU-FNL in 2008.

9. Third mandate contenders comprise a variety of political stakeholders, including opposition parties, high-ranking members of the ruling party and civil society. They refer to a general two-term limit included in the Arusha Accord. The presidential camp points to an ambiguity in the Constitution, which sets a two-term limit for directly elected presidents. Given that Nkurunziza was elected during the first post-Arusha elections in 2005 by Parliament, the presidential camp argues that he was legally allowed to run one more time.

10. (Appendix 1) for summary timeline.

11. Examples include the massacre of about 160 Banyamulenge in the Gatumba refugee camp in August 2004.

12. Initially, Imbonerakure were mostly civilian members. During the war, they became a key recruitment base for FDD fighters.

13. Informants highlight that there are two different groups of Imbonerakure: first, ordinary party youth members; second, paramilitary members. Paramilitary members are primarily recruited from former FDD, but also FNL members. It is extremely difficult to know how many of the thousands of Imbonerakure have been armed and trained. The most credible estimates range in the hundreds (Interviews with CNDD-FDD members, Bujumbura, summer 2014 and 2015).

14. For an excellent analysis on this process see (Alfieri 2016).

15. FNL received 14.15 per cent as compared to 63.03 per cent for CNDD-FDD.

16. FNL-Rwasa militants participated in the elections under a coalition umbrella because the Ministry of Interior does not currently legally recognize his FNL branch.

17. Travaux communautaires oblige all Burundians (older than 18 years) to participate in a half-day mandatory community service on Saturday mornings.

18. It would be interesting to better flesh out these commonalities and differences in future research, including how youth party members perceive and negotiate political manipulation.