Niklas Karlén 
Published online: 22 Oct 2020

Policymakers sometimes argue that material assistance to rebels involved in a civil war can create a ‘ripe moment’ that is favorable for negotiations. Ripeness theory provides support for this idea. However, this notion has never been systematically assessed. This article evaluates this claim by using global data on negotiations in all intrastate armed conflicts from 1975 to 2009. Contrary to popular belief, the article demonstrates that external state support to rebel groups does not increase the prospect of negotiations. Instead, the results suggest that external support is likely to reduce the likelihood of negotiations between the warring parties, especially if the state sponsor is a great power. The study contributes to our understanding of civil war processes by demonstrating that military assistance hinders rather than promotes the onset of negotiations and by questioning the utility of ripeness theory as the most suitable framework for understanding this phenomenon.

Previous articleNext article

“Ultimately a combination of diplomacy and pressure will be needed to bring about a political transition. Military pressure particularly may be necessary given President Assad’s reluctance to negotiate seriously” 1

– John Kerry, US Secretary of State, 2013-2017

“Absent a credible challenge to their militarized control of Nicaragua, the Sandinistas have no incentive to negotiate a lasting political solution to the conflict in Central America. The resistance can provide such a challenge – if we help” 2

– George P. Shultz, US Secretary of State, 1982-1989

Policymakers sometimes argue that military assistance to rebel groups can put pressure on a government and push it to the negotiation table. In 2016, more than fifty U.S. State Department officials signed an internal memo which advocated for an increase in military pressure on Syria’s President Bashar al-Assad in order to get him to the negotiation table. 3 Similarly, France and the United Kingdom claimed in EU deliberations that lifting the arms embargo against the Syrian opposition would pressure the regime to negotiate. 4 In the 1980s, when providing support to the rebel group UNITA, U.S.’ main goal was “to bring enough pressure on the Angolan government to enter into negotiations and for all the parties to find a political solution” 5 . Relatedly, U.S. President Ronald Reagan said in a public speech that support to the Contra movement in Nicaragua had compelled “the Communist Sandinistas to the negotiating table and forced them to negotiate seriously”. 6 More recently, Russian President Vladimir Putin said that support to separatists in Eastern Ukraine would continue as long as the Ukrainian government did not engage in substantive negotiations 7 while Iran’s Supreme Leader Ayatollah Ali Khameini has pledged support to Yemen’s Houthi movement while calling for dialogue between the warring parties. 8 Some policymakers seem to believe that providing material assistance to the rebel side involved in a civil war can create a ‘ripe moment’ that is favorable for negotiations and that it thus offer an attractive path to a negotiated settlement or a political transition. However, this belief has never been systematically assessed. Does state support to rebel groups engaged in civil wars increase the likelihood of negotiations?

In this article, I test whether this belief holds up to scrutiny. By anchoring this claim in ripeness theory, I hypothesize about why parties are likely to engage in negotiations and suggest under what conditions external military pressure is most likely to increase the government’s perception of a hurting stalemate. External military pressure is defined as tactical assistance from a foreign government to an armed opposition movement engaged in civil war, which increases that group’s capacity as a challenger to the incumbent regime. 9 The type of assistance provided can range from weapons, money, logistics and air support to safe havens, military training, troops and shared intelligence. Negotiations are defined as talks between the government and a rebel group that concern conflict-related issues.

It is imperative to understand what conditions are favorable in bringing the warring parties to the negotiation table as this is often seen as the crucial first step toward achieving other policy objectives such as a political settlement or a political transition. Moreover, it is also important to understand when civil war negotiations occur as they frequently serve other strategic purposes. For example, warring parties often use negotiations instrumentally to rearm and reorganize 10 and the initiation of talks legitimizes the rebel group 11 . Talks also increase the likelihood of rebel group fragmentation 12 , foster temporary reductions in violence 13 and constitute an important transformation of the form of contestation 14 . Taken together, this highlights the need to understand why negotiations take place.

Although there has been extensive research about the outcome of peace processes, only some large-N studies have been devoted to understanding the onset of negotiations in civil wars. 15 None of these studies focus specifically on how external military pressure from outside states can lead to negotiations. 16 In fact, most of the literature on civil war negotiations has been limited to study external pressure in the context of third-party mediation – as in when a third party actively facilitates talks. 17 Although this literature has offered several important insights about when warring parties are likely to accept mediation and under what conditions such efforts might produce a political settlement, these studies only look at a small subset of all negotiations that take place during a civil war. States may seek to actively influence the prospect of negotiations without themselves extending an offer to mediate. Negotiations are an integral part of conflict processes both with and without outside mediators. However, we still have a limited understanding as to why warring parties decide to negotiate. 18

The main contribution of this article is that it systematically evaluates a popular conception in policy circles about the effectiveness of military aid to opposition movements engaged in civil wars. In doing so, it assesses the usefulness of ripeness theory in helping us explain the initiation of negotiations. The notion that military assistance pressures governments to negotiate has not been assessed empirically with global data, yet policymakers commonly consider it to be a viable tool to force governments to the negotiation table. This is particularly important to explore further given that if policymakers are wrong, it might be that resources are wasted and that intended actions have no effect. Furthermore, given the prevalence of ripeness theory, it is worthwhile evaluating if this is the most suitable framework in trying to understand civil war negotiations. The article also adds to the expanding research on the effects of external support on conflict processes – which so far has been mostly concerned with conflict duration 19 or termination 20 while overlooking central conflict dynamics such as the initiation of talks.

The results suggest that the effectiveness of external military pressure in getting governments to engage in negotiations is at best dubious. On average, external support to rebel groups does not lead to negotiations between the warring parties. This finding brings into question a widespread belief in policy circles about the usefulness of such actions in getting parties to the negotiation table. Results indicate that direct military intervention, sustained pressure over time, or pressure from a great power do not make it more likely that governments engaged in civil wars will alter their initial calculation and agree to talks. On the contrary, the article offers support for the notion that military assistance may in fact act as a barrier to negotiations as the statistical results show a negative correlation between external support and the onset of negotiations. This negative effect is further accentuated if assistance is provided by a great power. Theoretically, the findings call into question the logic of ripeness theory in helping us understand the initiation of civil war negotiations.

The article proceeds as follows. The next section elaborates on why external states employ military pressure, highlighting that support to rebels in civil wars is a common practice. I then explain why we would expect that the government is most often the party that needs to be persuaded for negotiations to take place. The third section shows that there is theoretical backing in the literature for policymakers’ intuition since external aid to the rebels can pressure the government into negotiations by increasing the costs and thus the perception of a hurting stalemate. In relation to this I discuss why sustained pressure, direct military intervention and support from great powers should be especially important. This is followed by a description of the research design. I then turn to an empirical analysis of the onset of negotiations in all intrastate armed conflicts from 1975-2009. The article ends with some concluding remarks and a discussion of how its findings could inform future policy.

Why Do External States Pressure Governments?

The idea that outside states can influence the cost-benefit calculus of the warring parties is not new. External military pressure can essentially be seen as a form of coercive diplomacy. Coercive diplomacy is an attempt to get a target to change its behavior through either the threat to use force or the actual use of limited force. 21 Coercive diplomacy entails a larger repertoire of tactics that could be used to advance a state’s foreign policy goals in relation to another state, but this article focuses specifically on evaluating the impact of military assistance to rebels. Coercion is most likely to succeed when the coercer reduces the adversary’s chances of achieving an outright military victory. This type of denial strategy prevents the adversary from obtaining the anticipated benefits of conflict and makes it more favorable to concessions. 22 There is a belief among decision-makers that such a denial strategy often is useful.

An integral part of the U.S. strategy in Syria has been to arm rebels to put pressure on the regime to enable a diplomatic solution of the conflict and to give moderate opposition movements a greater bargaining position at the negotiation table. 23 In a similar manner, U.S. President Reagan authorized the provision of support to the rebel group UNITA in the Angolan Civil War in the 1980s as part of a larger strategy aimed at getting the government of Angola (MPLA) to realize that negotiations and a political settlement would be the only way forward. 24 Similarly, the decision by U.S. President Bush to arm rebels in Cambodia was said to “strengthen the hand” of the resistance and increase the chances for a political settlement. 25 More recent developments in Ukraine, where Russia seems to be stepping up its support to secessionists, also appear to be part of an effort to strengthen the armed opposition’s bargaining position and to pressure the regime into negotiations. 26

Foreign involvement in other states’ civil wars has long been common. 27 San-Akca estimates that almost 60 % of all rebel groups active since WW2 were supported by at least one external state. 28 Previous research has identified a range of motives behind external state support to rebels. States have diverse strategic interests such as to weaken a perceived rival, to gain access to economic resources, to increase their sphere of influence, to export a certain ideology or even to ensure political survival at home. 29 Support can furthermore be underpinned by humanitarian concerns or kinship ties. 30 We currently lack global data on reasons for state support to rebel groups and the fact that state sponsors can have multiple goals that change over time further complicates any attempt to collect such data. 31 Whatever the underlying motives, biased support to the opposition strengthens the capabilities of the rebel side and makes it a more attractive bargaining partner. 32 What is key is that external support is perceived as increasing the costs of war for governments in civil war, thereby negatively affecting their cost-benefit calculus. The next section explains how the warring parties differ in their views on negotiations, and why we should expect that it is generally the government that needs to be forced to the negotiation table.

Arm Wrestling with the Government?

Civil war belligerents vary widely in their willingness to negotiate. While some warring parties rarely or never negotiate, others engage in frequent talks. Parties are unlikely to be willing to negotiate as long as they believe that they could achieve a better outcome by military means. If one party thinks that they can prevail in the conflict without compromises, they are likely to refuse to engage in negotiations since this usually involves concessions. Both parties need to conclude that the conflict is ripe for resolution for talks to ensue. Zartman has argued that negotiations become more likely once a civil war has reached a ‘mutually hurting stalemate’. 33 This is a condition in which neither side is able to prevail militarily but both parties are paying a significant cost for continued fighting. Combatants face a cost-benefit calculation about the relative merits of fighting versus talking. As one side’s expectation of winning decreases, continuing to fight is less useful, and that side’s willingness to negotiate should increase. Negotiations take place when both parties lose faith in their chances of winning and see an opportunity for cutting losses and instead achieving their objectives through talks. 34

The decision to begin negotiations entails both costs and benefits to the warring parties. However, rebels should, on average, be more willing to negotiate than governments. 35 The main reasons being the power asymmetry inherent in internal conflicts and that talks could significantly increase a group’s legitimacy. Many rebel groups crave legitimacy, so even when talks involve no direct concessions just recognizing the insurgents as worthy interlocutors could be seen as a potential victory. 36 Recognizing a rebel group by agreeing to talks can increase their legitimacy both domestically as well as abroad. Since rebels are most often the weaker party, they should be less confident in a decisive military victory and hence more prone to accept concessions. For governments on the other hand it is much more costly to initiate negotiation efforts as talks with insurgents involve several risks. These risks range from political embarrassment to strengthening the group’s capacity and potentially encouraging other challengers.

The most common objection to negotiations is that talks formally recognize the rebels as worthy spokespersons. Most of the time, governments would refuse to grant the insurgents legitimacy as a bargaining partner. 37 If negotiations are associated with a ceasefire the rebels could use the talks to rearm and reorganize, thus strengthening their capabilities. This would both increase their bargaining power and improve their chances of military success if the talks break down. Furthermore, negotiations may lead to the belief that groups that use violent methods will be rewarded. Talks can signal that the government has no resolve and is likely to make concessions. 38 This could open up for additional challengers. Another potential risk for the government, particularly in democracies, is that political opponents can use the negotiations to discredit the incumbent government and claim that it is weak on “terrorists”. 39 Because of these reasons, I assume that in most cases, the government is the party that needs to alter perspective in order for negotiations to take place.

What influences the government’s willingness to negotiate? A perceived turning point is likely needed in order for the government to rethink and reevaluate its current strategy. Previous research has emphasized how timing, costs, external shocks and third party mediators can increase the perception of a hurting stalemate and help foster a ‘ripe moment’. 40 External pressure by third parties has however primarily been explored in the context of mediation. 41 But can external states produce ripe moments that would lead to talks by providing material aid to the rebel side? The next section anchors the conception among policymakers that military assistance to rebels can foster negotiations in ripeness theory.

Pushing for Peace? External Military Pressure and Ripe Moments

Internal conflicts are marked by a strong power asymmetry. In civil wars, the capacity of the rebels rarely matches that of the government they seek to challenge. Governments start from a position of relative strength and rebels from a position of relative weakness. Governments already have legitimacy, allies, armies and access to resources at the outset. Rebels need to fight in order to obtain all of these. This asymmetry rarely produces a stalemate that would be conducive to negotiations. In the absence of power equality, the softer notion of a perceived stalemate as a no-win situation for both sides is often the best that can be produced. 42 Parties often decide to negotiate when they perceive the distribution of power between them as moving towards equality. 43 Escalations and attempted escalations could help produce turning points. Potential escalations can also serve as warnings that reinforce stalemate and foster negotiations.

Several scholars have noted that negotiations during civil wars are most likely to open up when conditions shift toward parity. 44 In order to get closer to power parity one of two things need to happen: (1) the rebel group needs to be strengthened or (2) the capacity of the government side needs to be reduced. External military pressure is likely to affect both. Material aid to the rebel side decreases the gap in capabilities between the government and the rebels by increasing their capacity and as an indirect consequence this weakens the government side. External resources such as weapons, military advisors and money are all likely to increase the organizational capacity of the rebels. Some scholars have shown that outside support to rebel groups also make them increasingly like to win the war. 45 Assistance provided by an external state decreases the chances of an immediate government victory and increases the government’s perception of a hurting stalemate and hence its willingness to negotiate.

When external military pressure is introduced, the government’s assessment of the time necessary to win increases. The involvement of a foreign state on the side of the rebels decreases the probability of a government victory and increases the costs of conflict. This should, ceteris paribus, make negotiations a more attractive option for the government. However, at the same time as external military pressure increases the government’s willingness to engage in talks it might also decrease the willingness of the rebel side to negotiate. The reasons for this is that external assistance can strengthen the capacity of the group and thus make the insurgents believe that a military victory might be more feasible than before. 46 This effect should however be mitigated by the relative lack of capacity that rebel groups have at the outset and the fact that rebels usually have more to gain from talks than the government. Sufficiently strong rebel groups might even be reluctant to accept support, as they know that it comes at the expense of decreased autonomy and meddling by the state sponsor. 47 External patrons are likely to have significant leverage over insurgents. State sponsors of rebel groups are thus not only able to pressure governments into agreements. Similarly, they should have considerable leverage over the groups they are supporting. If an external state sponsor is sincere about pressuring the government to negotiate it is likely to put pressure on the group as well (although not necessarily by military means). A state sponsor can influence the rebels by indicating that material support would be withdrawn in the future if they do not agree to negotiations. The rebels, which anticipate the harm caused by a potential loss of an outside ally, are thus likely to negotiate in order to keep the flow of external resources. China basically forced the Khmer Rouge in Cambodia to engage in negotiations with the government based on a threat to otherwise withdraw its support for the group. 48

It should be noted that a hurting stalemate is a matter of perception. Although objective indicators are imperfect predictors of ripeness this does not mean that an objective indicator – such as outside military assistance – cannot alter the perception of the warring parties. Governments abandon conflict when it becomes overly costly and when the alternative track is more promising and comparatively cheaper. This leads to the first hypothesis:

Hypothesis 1: External military pressure increases the likelihood of negotiations

States vary in their ability to induce costs on foreign governments. It is reasonable to assume that external pressure varies depending on (1) the capabilities of the state supporter, (2) the application of sustained pressure over time, and (3) the type of pressure used. Hence, I stipulate three additional hypotheses intended to capture these dynamics.

First, scholars of international relations have long highlighted the unique role and responsibilities of great powers in the international system. 49 Great powers possess extensive military capabilities and a substantial recognition by other states. Their capacity to project force abroad combined with a wide-ranging foreign policy agenda set them apart from other states. 50 Great powers are because of this uniquely situated to raise governments’ perceptions of costs.

Great powers differ from other states in the type and range of support they can offer as they are able to provide resources that are far superior and in higher demand than those offered by other states. Technological advancement allows for the provision of high-quality resources such as advanced weaponry, satellite intelligence information and logistical resources such as airlift capacity. Moreover, forces are generally highly trained and well equipped if the support effort encompasses military personnel. Today the great powers account for roughly two thirds of all military spending in the world. 51 This suggests that great powers are potentially more able than other states to inflict large costs on the government in a civil war.

The involvement of a great power on the side of the armed opposition is symbolically important as well. The psychological effect of knowing that you face someone much stronger than yourself should not be underestimated. When the conflict in Georgia erupted again in 2008 Russia quickly responded by supporting the secessionists in South Ossetia. Russian troops and tanks with air support were present in the country within the next couple of days. 52 Facing such a superior adversary, the government of Georgia realized quickly that they had no chance of winning and thus no other option than to initiate talks with the separatists. A ceasefire agreement was signed just two weeks after the conflict had restarted. 53

Hypothesis 2a: Great power pressure increases the likelihood of negotiations

Second, regardless of the power capabilities of the supporter, sustained military assistance over time is likely to increase pressure on the target government. Even when conflict is characterized by gradual intensification, it is very likely that at some point “more and more” eventually – if not unequivocally – becomes “different”, and introduces a qualitative change. 54 Sustained support to rebels implies a stronger support commitment by state sponsors and signals that the pressure is not likely to go away. As such, governments will eventually be forced to give in to negotiations.

Hypothesis 2b: Sustained military pressure increases the likelihood of negotiations

Third, the type of external pressure may be important. The most intrusive form of support should be full-scale military intervention in which one or more state supporters actively commit troops to fight alongside the rebels. Such a strong show of force is likely to increase pressure on the target government more than the indirect provision of support.

Hypothesis 2c: Direct military intervention increases the likelihood of negotiations

Research Design

I utilize yearly data on all intrastate armed conflicts from 1975-2009 in order to test whether the expectation that external military assistance from outside states to rebels is associated with the onset of negotiations. I follow UCDP’s definition of intrastate armed conflict as “a contested incompatibility that concerns government and/or territory where the use of armed force between two parties, of which one is the government of a state, results in at least 25 battle-related deaths in one calendar year”. 55 Conflicts identified as coups by either Thyne 56 or Cunningham et al. 57 are excluded from the sample with the reasoning that coups follow a different logic and are usually over within a few days. 58 The time period is determined by data availability. My unit-of-analysis is conflict-dyad year and the dataset includes 1711 conflict-dyad years and negotiations between the warring parties commenced in 14 % of these. Because of the structure of my data and the binary nature of the outcome variable I use binary time series cross sectional (BTSCS) logit models. In order to account for temporal dependence I follow Carter & Signorino (2010) and incorporate a variable of time since the last round of negotiations along with its squared and cubed term. 59 To account for spatial dependence I use robust standard errors clustered on conflict dyad. 60

The dependent variable – negotiations – is a binary measure of whether or not any negotiations between the warring parties started during a given year. Negotiations are defined as talks between the government and a rebel group that concern conflict-related issues. Issues discussed could concern matters such as autonomy, power sharing, ceasefires or the release of war prisoners amongst other things. Talks that are solely about initiating negotiations in the future – so called ‘talks about talks’ – are not considered to be negotiations. The focus is on the initiation of negotiations rather than their incidence as the claim I seek to evaluate is whether military pressure can force governments to the negotiation table, not if pressure is likely to keep the parties at the table once they are already there. I allow for the possibility of multiple talks within the same dyad but there needs to be a period of at least one year without negotiations before any new onset could occur. This means that no case exits the dataset just because the first negotiation onset has occurred. The main data comes from the variable ‘neg’ retrieved from the UCDP Conflict Database. 61 This data have then been supplemented with additional information from Kreutz (2012) for conflicts in Africa and by coding by the author for the other regions to reduce the amount of missing data. 62 Note that this operationalization of negotiations differs substantively from datasets on mediation such as that by DeRouen, Bercovitch and Pospiezna 63 or by Greig and Regan 64 since these only include talks that take place in the presence of a third party mediator. 65 Much of previous work on civil war negotiations have either been restricted to the mediation context or too focused on exploring negotiated settlements as a type of conflict termination rather than seeing the negotiation process as encompassing several distinct decisions. Findley has perhaps addressed this best by disaggregating negotiations into multiple stages. 66 This separation is important as the factors that bring the parties to the negotiation table may be different from those that explain why the parties reach or implement an agreement. In his article, effects differed depending on what stage of the process he was looking at. 67 This provides a strong rationale for focusing on the question that ripeness theory directly speaks to: the initiation of negotiations.

External military pressure is a binary variable that denotes whether any outside state provided support to the rebel group active in the conflict-dyad. The data come from the UCDP External Support Disaggregated/Supporter Dataset v.1-2011. 68 This dataset is used as it provides time-varying data on a yearly basis and encompasses a range of different types of support. Other datasets do either not offer time-varying data or are limited to military interventions (troop support). The subset of the data I use only includes support from states and is limited to confirmed instances of external support. Great power military pressure is a binary variable that indicates whether any support was provided by a great power in order to strengthen the rebel group. In order to determine if a great power delivered the support, I consider states that were identified as ‘major powers’ by the Correlates of War project during the investigated time frame. 69 This is then matched with the UCDP data. Seven states were considered great powers at some point in time 1975-2009: USA, Russia, U.K., France, China, Germany and Japan. 70 Most quantitative empirical studies have used this definition to identify great powers. 71 Sustained military pressure is a cumulative count of the number of years during which at least one state supporter provided external support to the rebels. Direct military intervention is a binary variable that captures whether there was a third party that directly intervened on the side of the rebels. This information is retrieved from the UCDP Dyadic Dataset v.1-2016, 1946-2015 from the variable denoting whether the rebel side received any active troop support by a foreign state. 72


A range of control variables is included in the statistical models. To begin with, I control for external support to the government side. The operationalization is equivalent to how support to the rebel side was measured. There are two rationales for including this. First, external support to the government side might reassure the government and increase the perception that victory might be forthcoming. This would naturally decrease the need for talks with the rebels. Second, depending on the relationship between the external supporter and the government in a civil war it is possible that the supporter could actively disfavor or favor negotiations. For instance, the U.S.-led coalition in Afghanistan long adhered to a policy that actively discouraged the Afghan government from negotiating with the Taliban. 73 On the other hand, government supporters can also encourage negotiations. The Soviet Union played an important part in persuading the government of Angola to engage in talks with the rebel movement UNITA in 1990. 74 Regardless of whether the government would feel more or less compelled to negotiate because of its supporters it is important to take this into account. This will be measured as a binary variable denoted government support and it captures whether there was at least one external state providing support to the government.

Parties might be more likely to initiate negotiations if they perceive the conflict as costly for reasons other than external pressure. Conflicts that have lasted longer might be more conducive to negotiations. 75 In order for negotiations to take place, rebels must survive their initial vulnerability. In the early stages of conflict, the government is likely to try to militarily defeat the challengers, but if the insurgents survive a certain period of fighting, a window for negotiations is likely to follow. 76 Duration is measured as a count of the number of years since the conflict-dyad first became active as defined by UCDP. Conflict intensity could also serve as an indicator for perceived costs, as an increase in fatalities can increase the perception of a mutually hurting stalemate. I measure this as a dummy variable that captures whether the conflict has more than a thousand battle-related deaths in a given year. This information comes from the UCDP Dyadic Dataset v.1-2016, 1946-2015.