When spoilers succeed…the results are catastrophic.
– Stephen Stedman[i]

Peace processes are very often lengthy and difficult, many cease-fires negotiated to end civil wars often result in a return to violence, sometimes worse than before. This essay will examine the role of those actors who ‘actively seek to hinder, delay, or undermine conflict settlement’[ii] for a range of reasons and through a variety of methods. This essay takes Stephen Stedman’s term ‘spoiler’ as a means of classifying these actors.[1]

Stedman is the father of the spoiler debate, he gave birth to the term ‘spoiler’; identifying them as ‘leaders and parties who…use violence to undermine attempts to achieve [peace.]’[iii] This definition has since been subject to much criticism for its narrow definition and its pejorative connotations however.

In examining the role that spoilers play in peace processes we must first acknowledge that a spoiler is not a constant notion that acts the same and for the same reasons in every situation where peace is being negotiated. Spoilers play different roles based on a number of factors. To examine the roles they play, this essay will explore the differences in the location, variation and strategies of spoilers.

This essay takes as a key text, Stephen Stedman’s work, ‘Spoiler Problems in Peace Processes’, as a means to understand the original ideas about ‘spoilers’ and as a focus of critique of the established notion of what spoilers are and what they do. It will draw equally heavily on works critical of Stedman; principally those by Marie-Joëlle Zahar and Feargal Cochrane. By using a similar model of examining spoiler activity as Stedman, but by applying independent and more contemporary thinking I will examine more rigorously the roles that spoilers play and the effects they have had on a number of recent peace processes. In this examination I will identify concerns about the over simplification in traditional debates on spoilers and the disputed discourse used to quantify resistances to peace.

Location of the Spoiler

Despite the debate about the classification of spoiler activity it has been widely accepted that the location of the spoiler can help answer the question of why spoilers ‘spoil’[iv][v][vi] which can enrich the examination of the role they play. Stedman identifies that there are two locations of spoilers: inside and outside. Inside spoilers are those parties or actors who are included in the peace negotiations and who ‘show a willingness’ to reach a peaceful solution but who later ‘fail to fulfil key obligations to [an] agreement’.[vii] Whereas outside spoilers are those actors or parties who are excluded, for whatever reason, actively or unwittingly from the negotiations for peace.

Actors[2] who are on the inside of the peace talks are likely to continue to support the discussion or the implementation of a negotiated settlement up to the point that it ceases to benefit them or their cause; thus the actions of inside spoilers are defined by the fact that they must conceal any opposition up to the point of the actual spoiling, lest they be revealed as a potential spoiler. Accordingly the actions of inside spoilers are characterised by stealth; once an inside spoiler is party to a peace agreement it must demonstrate, however falsely, that its intentions are to continue to adhere to the terms of the settlement. Stedman notes that actors are likely to do this whilst maintaining an ‘offensive military capability’.[viii] Should the peace process show signs of disadvantaging an actor, or indeed advantaging its rival, the actor may begin to spoil. Cochrane argues, though he does not use Stedman’s terminology, that because peaceful (at least initially) resisters are not likely to ‘be visible and articulate their demands …towards the political process’ they can be more damaging to long term peace.[ix]

An example of an internal spoiler is the União Nacional para a Independência Total de Angola^[3](UNITA) who, in May 1991, signed the Bicesse peace agreement only to return to war a year later. The Bicesse Accords aimed to put an end to years of civil war which ravaged Angola and paved the way for Angola’s first democratic elections in September of 1992; elections that international observers declared as being ‘free and fair’.^[x] The outcome of the elections did not serve UNITA well, they lost a majority vote to the Movimento Popular de Libertação de Angola^[4] (MPLA) and consequently Jonas Savimbi, the leader of UNITA at the time, rejected the election results and resumed violent fighting. This is a clear case of an inside spoiler, participating in negotiations, signing an agreement then when the negotiated peace poses a threat to its survival and legitimacy (in this case a lost election) it spoils and returns to violence. This does not, however, take into account other factors which may have influenced, and diminished the tactical reasons for spoiling, such as the personality of Savimbi, a man known for his ‘caudillo-like’[xi] persona. Though this does serve as a good demonstration of the role inside spoilers can play and the harm they can cause through, amongst other things, the power of deceit.

In contrast to those inside peace negotiations, outside spoilers may have reasons to fear peace altogether. Their exclusion from the political dialogue means that they are denied a platform from which to air their grievances, and their lack of presence can mean that post-settlement representation would be unlikely. Outside actors can often become outside spoilers if the prospect of negotiated peace (from which they are excluded) threatens their legitimacy, power or  ideology of continued strife.[xii] Take for example the Palestinian Hamas[5], a political movement with the belief in a holy war (jihad) against the state of Israel at its core[xiii], if a peace agreement were to be negotiated, Hamas would be sure to suffer from a lack of legitimacy; as Zahar rightly affirms ‘peace with Israel would undermine the very bases of [Hamas’] existence.’[xiv] Stedman’s observation of spoilers informs us that outside spoilers are likely to use violence, and particularly favoured tactics include the ‘assassination of moderates’ and ‘massacres that coincide with…progress in reaching a…settlement’[xv]; such tactics were used in Northern Ireland by dissident Republican ‘splinter groups’[xvi] attempting to undermine the Good Friday Agreement; the bloodiest incident being the 1998 Omagh bombing. This

Spoilers do not always benefit from a monopoly of the spoiler behaviour; Stedman’s spoiler discourse does not exclude the possibility that more than one spoiler with different, even opposing, aims can exist within the same locus of conflict. A situation in which more than one spoiler is acting causes complex difficulties for those wishing to contain and manage the spoiler activity. As Stedman acknowledges, actions taken to manage one spoiler may, unconsciously, have strengthening effects on the other(s). One such example was seen in Rwanda where pressure from outside bodies, principally the UN, on President Habyarimana to adhere to the Arusha peace agreement only served to embolden the other spoiler in the region, the Coalition pour la Défense de la République[6] (CDR). The UN threatened a withdrawal of its peacekeeping forces if Habyarimana didn’t agree to implement the Accord. The CDR, a group comprised of former members of Habyarimana’s government, saw the threats made by the UN as an opportunity to attack and undermine the peace agreements and began conspiring to kill Belgian peacekeepers to force a UN withdrawal. Thus, attempting to resolve the problem of Habyarimana served to embolden the spoiler activity of the CDR. [xvii][xviii] This is an example of how, not only there can there be multiple spoilers in a process, but also that the location of the multiple spoilers is of significance. I would argue that the significance here lies in the fact that the UN would have been unlikely to enter seriously into diplomatic (in that a threat was given and force wasn’t the coercive devise) discussions with the CDR, essentially an extremist militia group, who were not party to the ongoing peace negations (nor a signatory to the Arusha Accords.) Internal parties have the attention of the international mediators (and as such they are most likely to suffer the public demands and pressures), thus outside spoilers have the advantage of being able to exploit any pressure applied to those inside parties to either undermine their opponents or, as seen in this example, to nudge a would-be ally in their direction.

Stedman’s ideas of inside and outside spoilers do not explicitly recognise that spoilers are not universally violent; informing us of the violent tactics used by outside spoilers (massacres, assassinations etc.) but neglecting the mention the importance or indeed the presence of peaceful resisters. The focus of the case studies he uses to demonstrate the importance of location in the spoiler debate, focus on those non-state or individual spoilers who use violent force as a means of resistance; perhaps because there are simply more instances of violent resistance, but whatever the reason, much of Stedman’s theory serves (implicitly at least) to render spoilers as violent actors with little or no legitimacy. This over simplified definition of a spoiler has been the locus of much of the criticisms of Stedman’s work; this, along with the issue of violent and non violent resistance is addressed later in the essay.

Variation of Spoilers

Though this essay does not seek to place undue importance on the ideas of Stedman, he did identify a number of variations in what he termed the ‘type of spoiler’, and though more contemporary readings of his work have informed us of alternative ways of classifying spoilers, Stedman’s classifications provide a worthy means of examining the roles played by different spoilers (even if such classifications are not considered exhaustively nuanced.) Stedman suggested that spoilers could be classified into three types; limited, greedy and total. I will take each of these and examine how the role of a spoiler differs according to its classification or ‘type’. The first type of spoiler Stedman describes is the limited spoiler. The limited spoiler has, as its name would suggest, limited goals; Stedman identifies examples of limited goals as ‘recognition and redress of grievance [and]…a share of power.’ Whilst these types of spoilers may have limited goals, it must be acknowledged that their commitment to these goals may (and often are) not be so limited. Limited spoilers offer some prospect for a negotiated settlement, providing their limited goals can be addressed and negotiated

One example of a limited spoiler with a strong and perhaps ‘unlimited’ commitment to their goal is the Resistência Nacional Moçambicana[7] (RENAMO) of Mozambique in 1992. RENAMO’s goals were limited, they sought political recognition and respect as a political organisation not just a militant force. Their desire to achieve these goals however were not so limited; RENAMO were known as the ‘Khmer Rouge of Africa’[xix] and committed a great deal of violent acts against civilian populations are were notorious for their use and exploitation of child soldiers. RENAMO subsequently retracted its spoiling activity and participated in democratic elections.

Total spoilers are on the opposite side of the spectrum; they have total goals and have complete unwavering commitment to those goals, seeing the world in ‘all-or-nothing terms’.[xx] Often proponents of radical thought and lead by individuals who are surrounded by ‘sycopaths. (sic.)’[xxi] Total spoilers offer little hope for a negotiated settlement, because if their all-or-nothing worldview. Stedman gives as an example of a total spoiler the Khmer Rouge (KR) in Cambodia. The KR’s signing of the Paris Peace Accords in 1991 was seen by some, including those in the UN force – UNTAC, as being tactical. The KR was viewed as a total spoiler because it was perceived to have spoiled the peace process in order to secure it own total goals. More recently it has been recognised by Sorpong Peou that the KR was in fact not a total spoiler and that UNTAC misread the intentions and concerns of the KR; concluding that the KR spoiled primarily because of genuine security concerns about their security, post-implementation of peace agreement, arising from its enforced disarmament. He goes to inform that UNTAC’s misreading of the KR’s intentions and the deep rooted culture of distrust in Cambodia exacerbated the spoiler behaviour.[xxii] The KR therefore can perhaps best be described as a greedy spoiler, a terminology dealt with next.

Perhaps a more apt case study might be that of Euskadi Ta Askatasuna[8] (ETA), who have consistently demanded the creation of an independent Basque state in the north east of Spain/south west France. There have been a number of examples of ETA ‘coming to the table’ only to later spoil through the use of violence (which has largely been non-lethal of late), despite a number of peace negotiations, spanning decades and several successive Spanish governments, ETA’s aims have remained immutable. The most recent example of ETA’s expression of interest in negotiating a settlement was seen in 2006 when it declared a ‘permanent ceasefire’; but ETA’s detonation of a bomb in Madrid’s Barajas airport during the peace talks, and after the declaration of the ceasefire, led to the breakdown of talks and the resumption of violence by ETA. Previous peace talks had broken down because ETA (perhaps understandably) had concerns with demands for it to disarm prior to negotiations.[xxiii] But nevertheless ETA’s goal has remained constant, and at the time of writing, ETA has refused to accept negotiation on the issue of an independent Basque state; resorting once more to the use of violence, most recently in Majorca in August 2009.

A clear example of a total spoiler having unchanging aims and unwavering commitment, and the potentially devastating consequences this can have, is the actions of the CDR in Rwanda, discussed earlier, who carried out genocide in the pursuit of its aims.

Stedman’s notion of total spoilers has come under criticism from Zahar and Darby who note that the definition of total spoilers as unchangeable is problematic and they express doubt over the existence of total spoilers, explaining that many spoilers deemed to be ‘total’ have eventually signed refreshed agreements or ceased violence and other spoiler activity[xxiv][xxv]; but as Stedman notes, labelling parties as ‘totalistic’ is often done for tactical reasons to de-legitimise and undermine their legitimacy, and is not always a true analysis of the prospect for a settlement.[xxvi]

The third type of spoiler identified by Stedman is the greedy spoiler; greedy in the sense that it changes its goals and commitment thereto depending on the costs and risks of making such changes. Greedy spoilers are located between their limited and total counterparts on the spoiler spectrum. They exploit any advantage they may have to expand their goals and commitment, such as did UNITA in Angola when Savimbi  took advantage of US strategies of inducement over force, and its seemingly constant efforts to appease UNITA (e.g. by pressuring the MPLA to enter into a power sharing agreement with UNITA.) The US strategy of inducement served to strengthen Savimbi who avoided further peace talks whilst continuing to amass and deploy his army, knowing there was no threat of force from either the US or the UN. Savimbi’s goals seemed to have expanded as a result of the low cost options for expansion, though there is no consensus on whether Savimbi desired total control over Angola from the outset or not; what is clear however is that the US strategy towards UNITA lead to Savimbi strengthening his commitment, and his ability, to achieve those goals (pre existing or otherwise) of total power.[xxvii][xxviii] Stedman notes that there can be hope for the incorporation of greedy spoilers into negotiations so long as their initial limited goals can be met and the risks and costs of spoiling remain high, thus preventing the expansion of their goals.

These classifications have attracted much criticisms due to their simplification of complex issues and because of their focus on negative and violent spoilers. Cochrane criticises the whole spoiler discourse, expressing concerns about the over-simplification of issues and notes that Stedman’s work focus too heavily on the ‘simplified binary discourse’ of who is and who is not a spoiler. Furthermore, Cochrane argues that the pejorative connotations that come with a word like ‘spoiler’ unjustly, but automatically, label any resistor to a certain peace process as an ‘extremist who prefers violence to peace’[xxix], and Newman and Richmond further express concerns that in the post 9/11 world spoilers could be labelled as ‘terrorists.’[xxx] As the next section briefly covers, Cochrane argues that a better way of classifying spoilers or resisters (though classification does in some way contradict Cochrane’s view of the spoiler debate), and as such assessing the roles they play, is those who are ‘violent’ and those who are ‘peaceful.’

Strategies of Spoilers

As noted earlier, the spoiler discourse has not been accepted by everyone. Cochrane has presented problems with the label ‘spoiler’, and refers instead to ‘resisters’. Cochrane makes one key point in his book ‘Ending Wars’; that those who engage in war do so mainly for political purposes which are often complex and dependant on context; and importantly they are not, as they are often purported to be, irrational actors.[xxxi] To this end, Cochrane’s work allows us to linearise, somewhat, the spoiler debate to include everyone; it allows us to widen the definition of spoiler to all parties, explaining all parties are spoilers in some respect if we consider each party as equal and rational actors. Having noted that the spoiler problem is wide to be fully explained by Stedman’s definitions of limited, greedy and total spoilers, Cochrane adopts an analysis focused on the strategies of spoilers and how these interlink with their motives.

One of Cochrane’s main criticisms of the mainstream spoiler theories is their focus on the agency of non-state actors,  he argues that such emphasis on the greed of non state actors has ‘certain rebel-centric characteristics’ that, he argues, echoes the sentiments expressed by Collier that greed more than any other factor motivates civil war.[xxxii] The intention therefore of widening the debate on spoiler is to include debates about the structural causes of violence.

In his descriptions of the role of spoilers, Cochrane focuses on violent and peaceful resisters. He notes that violent spoilers do not act mindlessly and very much aware of their actions and the consequences thereof. It would be of little benefit to discuss the impact of violent spoilers here as almost all case studies presented up to this point have been of violent resistance; instead I will address the notion proposed by Cochrane that violent resisters can have the opposite effect of prolonging violence by bringing reluctant warring parties back to the table, ‘for fear of the alternative.’[xxxiii] An example of such an event took place in South Africa, where fear of catastrophic violence, which had been ‘sampled’ previously, forced the creation of the National Peace Accord – which facilitated the transition of democracy in South Africa.

Key to the spoiler debate is the ‘peaceful resister’ who, as mentioned earlier, can often be an inside spoiler who created complicated problems for the peace process by undermining negotiations from within. Cochrane notes that in example of such problematic yet peaceful resistance was Condoleezza Rice, former US Secretary of State, who, during the 2006 war in Lebanon, opposed a ceasefire and actively supported a continuation of war until Hezbollah had been defeated; Rice noted ‘a ceasefire would be a false promise’, calling the conflict a ‘birth pang of a new Middle East.’[xxxiv] Though I would dispute Rice’s role, and through her the US’s role, as a peaceful spoiler as the US ‘fast-tracked the delivery of precision guided bombs to Israel’[xxxv] thus, in the opinion of some merely constituting a third-party violent outside spoiler acting through the state of Israel, though this argument may be too political for such a debate.

Drawing a conclusion on such an issue is difficult; not least because the very concept of the issue is not universally understood. There are many factors which will influence the role of a spoiler, its location, its goals, and its capacity having been discussed in this essay. Additionally spoilers can act violently or peacefully and the outcome from either action is never truly predictable nor can it be said that peaceful spoiling results in fewer deaths (if such a metric is to be used to measure the severity of spoiling.) What can be effectively concluded is that spoilers are not driven by one universal goal nor do they act in universal methods with universal commitments; there is a categorical presence of complexity and difference in the contemporary spoiler debate which requires complex and differing solutions. In answering the brief of the essay to ‘examine the role played by spoilers’ one can conclude that roles differ and depend on numerous complexities; from peaceful to violent, from limited to total, and from inside to outside. Spoiler roles are affected by numerous factors and as the case studies in this essay have shown, roles played by spoilers cannot effectively be quantified or unified in any one theory.

[1] The use of the word ‘spoiler’ to describe these actors in this essay is not done in any way to place a greater importance on the work or classifications of Stedman – rather, it takes this term from the essay title. Recognition of the problematic nature of the term is discussed later in this essay.

[2] Spoilers cannot be termed spoilers ex ante. The capacity and willingness to spoil does not in itself make an actor a ‘spoiler’; thus, exploring the tactics and motivation of an inside spoiler before they have acted to spoil, is something we do through hindsight and this retrospection must be acknowledged lexically.

[3] National Union for the Total Independence of Angola

[4] Popular Movement for the Liberation of Angola

[5] Islamic Resistance Movement

[6] Coalition for the Defence of the Republic

[7] Mozambican National Resistance

[8] Basque Homeland and Freedom

[i] S. J. Stedman, ‘Spoiler Problems in Peace Processes’ in International Security, Vol. 22, No. 2 (1997) pp. 5

Location of the Spoiler

[ii] E. Newman & O. Richmond (eds.), ‘Challenges to Peacebuilding: Managing Spoilers During Conflict Resolution’ (United Nations University Press, 2006) pp. 1

[iii] S. J. Stedman, ‘Negotiation and Mediation in Internal Conflicts’, in M. E. Brown (ed.), The International Dimensions of Internal Conflict (Cambridge: MIT Press, 1996) pp. 369-71

[iv] S. J. Stedman, ‘Spoiler Problems in Peace Processes’ in International Security, Vol. 22, No. 2 (1997) pp. 8

[v] M-J. Zahar, ‘Reframing the Spoiler Debate in Peace Processes’, in J. Darby & R. MacGinty (eds.), Contemporary Peacemaking (second edition) (Basingstoke: Palgrave MacMillan, 2008) pp. 161

[vi] P. Wallensteen, ‘Understanding Conflict Resolution’ (London: SAGE, 2002) pp. 49

[vii] S. J. Stedman, ‘Spoiler problems in peace processes’, in International Security, Vol. 22, No. 2 (1997) pp. 8

[viii] Ibid.

[ix] F. Cochrane, ‘Ending Wars’ (Cambridge: Polity Press, 2008) pp. 117

[x] I. Tvedten, ‘The Angolan Debacle’, in Journal of Democracy, Vol. 4, No. 2 (1993) pp. 108

[xi] Ibid., pp. 117

[xii] M-J. Zahar, ‘Reframing the Spoiler Debate in Peace Processes’, in J. Darby & R. MacGinty (eds.) Contemporary Peacemaking (Second edition), (Basingstoke: Palgrave MacMillan, 2008) pp. 161

[xiii] S. Mishal & A. Sela, ‘The Palestinian Hamas: vision, violence, and coexistence’ (New York: Columbia University Press, 2006) pp. 13

[xiv] M-J. Zahar, ‘Reframing the Spoiler Debate in Peace Processes’, in J. Darby & R. MacGinty (eds.) Contemporary Peacemaking (Second edition), (Basingstoke: Palgrave MacMillan, 2008) pp. 161

[xv] S. J. Stedman, ‘Spoiler problems in peace processes’, in International Security, Vol. 22, No. 2 (1997) pp. 8 – 9

[xvi] F. Cochrane, ‘Ending Wars’ (Cambridge: Polity Press, 2008) pp. 116

[xvii] S. J. Stedman, ‘Spoiler problems in peace processes’, in International Security, Vol. 22, No. 2 (1997) pp. 9

[xviii] L. M. Howard, ‘UN Peacekeeping in Civil Wars’ (Cambridge: CUP, 2008) pp. 30 – 2

Variation of Spoilers

[xix] S. J. Stedman, ‘Spoiler problems in peace processes’, in International Security, Vol. 22, No. 2 (1997) pp. 6

[xx] Ibid., pp. 10

[xxi] S. J. Stedman, ‘Introduction’, in S. J. Stedman, D. Rothchild & E. M. Cousens (eds.), Ending Civil Wars – The Implementation of Peace Agreements (London: Lynne Rienner, 2002) pp. 13

[xxii] S. Peou, ‘Implementing Cambodia’s Peace Agreement’, in S. J. Stedman, D. Rothchild & E. M. Cousens (eds.), Ending Civil Wars – The Implementation of Peace Agreements (London: Lynne Rienner, 2002)

[xxiii] R. P. Clark, ‘Negotiations for Basque Self-Determination in Spain’, in I. W. Zartman (ed.), Elusive Peace – Negotiating an end to Civil Wars (Washington D.C.: The Brookings Institution, 1995) pp. 59-75

[xxiv] M-J. Zahar, ‘Reframing the Spoiler Debate in Peace Processes’, in J. Darby & R. MacGinty (eds.) Contemporary Peacemaking (Second edition), (Basingstoke: Palgrave MacMillan, 2008)

[xxv] J. Darby, ‘The Effects of Violence on Peace Processes’ (Washington D.C.: United States Institute of Peace, 2001)

[xxvi] S. J. Stedman, ‘Introduction’, in S. J. Stedman, D. Rothchild & E. M. Cousens (eds.), Ending Civil Wars – The Implementation of Peace Agreements (London: Lynne Rienner, 2002) pp. 13-14

[xxvii] S. J. Stedman, ‘Spoiler problems in peace processes’, in International Security, Vol. 22, No. 2 (1997) pp. 36-40

[xxviii] I. Tvedten, ‘The Angolan Debacle’, in Journal of Democracy, Vol. 4, No. 2 (1993) pp. 115-6

[xxix] F. Cochrane, ‘Ending Wars’ (Cambridge: Polity Press, 2008) pp. 109

[xxx] E. Newman & O. Richmond (eds.), ‘Challenges to Peacebuilding: Managing Spoilers During Conflict Resolution’ (United Nations University Press, 2006) pp. 9

Strategies of Spoilers

[xxxi] F. Cochrane, ‘Ending Wars’ (Cambridge: Polity Press, 2008)

[xxxii] Ibid., pp. 110

[xxxiii] Ibid., pp. 114

[xxxiv] C. Rice, ‘Special Briefing on Travel to the Middle East and Europe’, (July 2006)

[xxxv] F. Cochrane, ‘Ending Wars’ (Cambridge: Polity Press, 2008) pp. 118

—

Written by: Daniel Gray
Written at: Lancaster University
Written for: Feargal Cochrane
Date written: 2009

Further Reading on E-International Relations

• Local Peace Aspirations and International Perceptions of Peacebuilding in Somalia
• UN Intervention: Help or Hindrance in the Democratic Republic of Congo?
• Double Agency? On the Role of LTTE and FARC Female Fighters in War and Peace
• Do Coups d’État Influence Peace Negotiations During Civil War?
• Fragmentation, Back Channels, and Hurting Stalemates in the Oslo Accords
• Ripening Conflict in Civil Society Backchannels: The Malian Peace Process (1990–1997)