Group Cohesion and Peace Processes

Print Friendly, PDF & Email
Image courtesy of Cristian Santinon/Flickr. (CC BY-ND 2.0)

This article was originally published by the United States Institute of Peace (USIP) on 12 September 2017.


  • Weak cohesion within nonstate armed groups can—and has often threatened to—under­mine negotiated transitions away from conflict.
  • Cohesion is measured along two axes: vertical (degree of command and control over cadres) and horizontal (degree of unity among leaders).
  • Challenges are typically related to negotiating partners who have little credibility, nego­tiating positions that are either unclear or incoherent, factions within groups that oppose the peace process, and splintering within groups.


Weak cohesion within nonstate armed groups (NSAGs) has often threatened to undermine negoti­ated transitions from conflict.[1] This can have an impact at any time—when parties are deciding on whether to join a process, during negotiation of peace agreements, and into implementation.

Cohesion can generally be measured along two axes: vertical (command and control over cadres) and horizontal (unity among leaders). Vertical cohesion is weak when leaders cannot control their fighters, and strong when they can. Horizontal cohesion is weak when leadership includes competing and disjointed factions, and strong when leaders have consensus over goals and are coordinated in action. Weak cohesion manifests in various combinations along these axes and is often a blend of the two.[2]

Group cohesion shifts over time and can be affected by a wide range of factors. These include internal debates about participation in a peace process, differing priorities, access to criminal networks that allow cadres to profit independently, diverse (ethnic, tribal, ideological, or religious) affiliations, territorial expansion, and counterinsurgency campaigns.[3]

How Does Group Cohesion Affect Peace Processes?

Challenges to peace processes related to group cohesion range from delays in negotiating or implementing accords to undermining the viability of the entire process.

  • Weak negotiating partners and positions

Weak horizontal cohesion among leaders can prevent NSAGs from presenting coherent nego­tiating teams or positions. Weak vertical cohesion can make it difficult for negotiators to convince mediators and other parties that they will be able to impose the terms of an agreement on their rank and file. Both open the door to a peace process stalling or breaking down.

In Senegal, competing leadership claims within the separatist Movement of Democratic Forces of Casamance (MFDC) have long complicated negotiations with the government. In 2006, multiple individuals claimed to be the secretary general of the MFDC though none could credibly claim to represent MFDC combatants as a whole.[4] Internal contestation among political leaders, in addition to many of those leaders’ weak links with combatants, has prevented the group from being able to convince Dakar it can implement what it promised at the negotiating table.[5]

In Uganda, the 2006–08 Juba talks between the Lord’s Resistance Army (LRA) and the government also suffered from a lack of credible LRA negotiating partners. Members of LRA peace delegations, many of whom had questionable ties to LRA leader Joseph Kony, prioritized different personal grievances. This lengthened the talks, strained the agenda, and made it difficult for mediators to know whether the delegates could deliver on what they negotiated. By stressing the need to prepare papers, meet deadlines, and participate in multiple mediation initiatives, the Juba process may have shrunk the space for internal LRA reflection and thus exacerbated these problems.[6]

In Burma, leaders of the Karen National Union have between 2012 and 2017 been accused of negotiating and in some cases signing agreements without the approval of the organization’s political decision-making bodies. Personal rather than institutional decision making has exacer­bated factional rivalries within the leadership and threatened the organization’s withdrawal from negotiations associated with the country’s ceasefire and national dialogue processes.

In Colombia, however, the Revolutionary Armed Forces of Colombia (FARC) have managed to maintain sufficient horizontal and vertical cohesion despite years of counterinsurgency efforts designed to undermine its unity. This cohesion enabled the FARC to offer Bogotá a coherent negotiating position for new talks in 2012. The FARC maintained internal cohesion throughout the talks by “rotating its negotiating team, giving leaders from all major combat units the opportunity to learn about the process and shape its outcomes.”[7] The inability of the smaller and more decen­tralized National Liberation Army to forge robust internal consensus on the terms of disarmament, meanwhile, contributed to the failure of past talks with the state.[8]

  • Objecting factions

Weak groups may include factions that mobilize against rapprochement. Objecting factions may seek to delay or limit a group’s engagement in negotiations or to openly denounce decisions by other elements of the group. Some objecting factions may be marginal. Others may be veto players, able to spoil the peace process. Leaders who control large numbers of combatants, enjoy popular support, and have access to substantial weapons or money are more likely to be veto players than those who do not.[9]

In Burma, some leaders of the Kachin Independence Organization remain highly opposed to compromise in current negotiations with the government and army. Significantly, these are not marginal players: their reluctance to accept the state’s terms is shared by the majority of Kachin communities, who largely distrust the sincerity of the government and army’s overtures.[10]

In Colombia, although the FARC maintained enough cohesion to reach a peace accord with the government in 2016, some are concerned that internal dissent could undermine implementation. One of the FARC’s guerrilla fronts announced in 2016 that it would not participate in the peace process. The FARC responded by effectively expelling it. At the same time, the state made it clear that it would continue to wage counterinsurgency against any dissident FARC fronts.[11] Since the agreement was signed, various additional elements from other fronts have stopped taking orders from FARC leadership and refused to lay down their arms.[12]

  • Splintering

When divisions within groups become irreconcilable, nonstate actors risk splintering during a peace process.[13] Disagreements may arise over whether to participate in negotiations, the content of negotiations, or during implementation. Fragmentation can make achieving sustainable peace accords more difficult. The more veto players involved, the less likely it is that a negotiated agree­ment will satisfy them all.[14] Splintering can also generate spoilers that undermine the sustainability of peace.[15]

In Burma, the Palaung State Liberation Organization splintered in 2005 over whether to accept the army’s ceasefire demands, giving rise to the Ta’ang National Liberation Army, one of four non­state armed groups still in conflict with the army. This and numerous other examples of splintering in Burma have contributed to a significant fragmentation problem, rendering common agreement between all parties in the nationwide ceasefire process highly difficult.

Mediators themselves can inadvertently encourage splintering. In Darfur, for instance, the African Union’s (AU) recognition of splinter groups “encouraged divisions because factions wanted to participate in the negotiations as independent bodies and bargain for their own interests.” The AU switched tactics in 2006, refusing to recognize a Sudan Liberation Movement/Army splinter group so as not to drive further fragmentation.[16]

Conclusion: Navigating Group Cohesion

Mediators faced with the challenge of weak cohesion within a NSAG should consider the following interrelated strategies.

  • Conduct dynamic analyses to assess cohesion

Group cohesion is dynamic, shifting as the context (including the peace process) empowers or disempowers leaders, encourages convergence or divergence in interests, or enhances or under­mines the institutional fibers of command and control. Mediation teams should analyze cohesion along both vertical and horizontal axes throughout the process, identify objecting factions within groups, and consider whether such factions could emerge as veto players.

  • Provide parties with breathing space to build internal cohesion

Mediators should build enough flexibility into a peace process that NSAG leaders can periodi­cally step back to resolve internal disputes to establish support for cohesive negotiating positions. Mediators may also consider creating time and space for delegates to return home to build sup­port among both their rank and file and their broader constituency. Periods of internal reflection and joint decision making can be time-consuming. But when enough opportunities are allowed for, mediators have mitigated risks that political leaders, military brigades, or constituents might later object to or undermine negotiations.

  • Verify institutional positions

When horizontal cohesion is weak, mediators may need to verify that NSAG delegates have an institutional mandate to make decisions. In Burma, for instance, mediators sought to encour­age public statements from the entire organization, rather than from individuals, which helped demonstrate broad endorsement of policies and positions during negotiations. Learning from the Colombian case, mediators may also want to encourage groups to rotate their negotiating teams to both maintain and demonstrate buy-in to the process across the leadership.

  • Empower bridging figures

Bridging figures have both the respect of leaders within an armed group and the ability to com­mand them. When such individuals are amenable to negotiations, mediators should seek to ensure that the process empowers rather than undermines or isolates them. Turkey demonstrates their importance. For years, Abdullah Ocalan—the imprisoned leader of the Kurdistan Workers Party and the only person who can bridge its various centers of power—was the chief Kurdish negotia­tor in the peace process. Since 2016, however, Turkey has refused to allow him contact with the outside world via intermediaries, something a key insider has suggested is an “invitation to war.”[17]

  • Accommodate a broad range of interests

Mediators can support internal cohesion in NSAGs by investigating and respecting the needs of the entire organization—horizontally and vertically—including potential veto players who might not be present during negotiations. Burma’s 2015 Nationwide Ceasefire Agreement, for instance, established interim arrangements that legitimized the security and administrative functions of NSAG brigades. These stipulations have the potential to provide a negotiation dividend for the semi-independent units of NSAG signatories, who might otherwise feel marginalized.

  • Navigate objecting factions

When objecting factions within a group are identified as potential veto players, interviews revealed, mediators may consider engaging their leaders early and disproportionately. In Burma, mediators have used confidence-building strategies to reduce trust deficits, such as commitments to issues of personal concern to the factions’ leaders (such as intervening to help secure prisoner release), or supererogatory administrative assistance (such as organizing visas and arranging travel). When mediators identify objecting factions as lower risk, nonveto players, they might want to consider how and when to isolate them from the process.18

  • Engage splinter groups sparingly

To discourage splintering, mediators may consider shutting out of the process any nonveto splinter groups that do emerge, as the AU began to do in Darfur. This would signal that objecting factions will not be able to participate in the process if they break away from their mother party.


  1. Weak cohesion is not limited to nonstate actors, but state cohesion is beyond the scope of this report.
  2. See Paul Staniland, Networks of Rebellion (Ithaca, NY: Cornell University Press, 2014).
  3. On criminal networks, Sebastian von Einsiedel, “Civil War Trends and the Changing Nature of Armed Conflict,” Occasional Paper no. 10, UNU-CPR, March 2017; on territorial expansion, Staniland, Networks of Rebellion.
  4. Kim Mahling Clark, “Final Report: Support to the Casamance Peace Process” (Washington, DC: USAID, December 2009), 3.
  5. Paul Collier and Nicholas Sambanis, Understanding Civil War: Africa (Washington, DC: World Bank, 2005), 280.
  6. Dylan Hendrickson and Kennedy Tumutegyereize, “Dealing with Complexity in Peace Negotia­tions: Reflections on the Lord’s Resistance Army and the Juba Talks” (Conciliation Resources, January 2012), 19.
  7. International Crisis Group (ICG), “On Thinner Ice: The Final Phase of Colombia’s Peace Talks,” Briefing no. 32 (Brussels: ICG, July 2, 2015).
  8. ICG, “Left in the Cold? The ELN and Colombia’s Peace Talks,” Briefing no. 51 (Brussels: ICG, Febru­ary 26, 2014).
  9. David Cunningham, “Who Should Be at the Table?,” Penn State Journal of Law and International Affairs 2, no. 1 (2013): 38–47.
  10. David Brenner, “Authority and Contestation Inside Rebel Groups: Insights from Myanmar,” Politi­cal Violence at a Glance [blog], June 21, 2017, /authority-and-contestation-inside-rebel-groups-insights-from-myanmar/.
  11. ICG, “Colombia’s Final Steps to the End of War,” Report no. 58 (Brussels: ICG, September 7, 2016).
  12. Jeremy McDermott, “FARC Unity Shatters in Colombia,” Insight Crime, January 12, 2017, www
  13. Victor Asal, “Why Split?,” Journal of Conflict Resolution 56, no. 1 (2011): 94–117.
  14. David E. Cunningham, “Veto Players and Civil War Duration,” American Journal of Political Science 50, no. 4 (2006): 875–92.
  15. See Stephen Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (1997): 5–53; Peter Rudloff and Michael G. Findley, “The Downstream Effects of Combatant Fragmentation on Civil War Recurrence,” Journal of Peace Research 53, no. 1 (2016): 19–32.
  16. Nuredin Netabay, “The Darfur Peace Process: Understanding the Obstacles to Success,” Beyond Intractability, May 2009.
  17. Gulsen Solaker, “Ocalan’s isolation an ‘invitation to war’ in Turkey: Pro-Kurdish MP,” Haaretz, Janu­ary 7, 2016,
  18. David Cunningham suggests two rules in deciding whom to include (veto players) and exclude (nonveto players) in peace processes (“Who Should Be at the Table?”).

About the Authors

Cale Salih is a research officer at the United Nations University’s Centre for Policy Research.

Stephen Gray is a strategic adviser for Hope International Development Agency, and director of Adapt Research and Consulting.

For more information on issues and events that shape our world, please visit our CSS Security Watch Series or browse our Publications.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.