Peacebuilding in Divided Societies. A theoretical framework and

some findings from Northern Ireland and Cyprus

Abstract

Against a background of a growing number of internal conflicts and the failure of traditional means of conflict resolution to achieve a durable peace in divided societies, this paper presents a two-track approach to conflict resolution. One track is represented by peacemaking, defined as attempts at finding a settlement to the issues in conflict at the political elite level. The other track, peacebuilding, refers to contact initiatives at the grassroots level aiming at the improvement of intergroup relations. After a discussion of role of grassroots peacebuilding in a peace process some preliminary findings on the relationship between peacemaking and peacebuilding in two divided societies, Northern Ireland and Cyprus, are presented.

Introduction

In the post-cold war era international war is becoming an increasingly rare phenomenon. The conflict data project at the Department of Peace and Conflict Research, Uppsala University, Sweden, monitors on-going armed conflicts on a yearly basis. Of 101 armed conflicts registered for the period 1989­1996 only six were conflicts between states (Sollenberg, 1997). Consequently it is the problem of internal conflict that today is attracting the attention of scholars, diplomats and policy makers. The majority of these internal conflicts are ethnic conflicts, that is they involve at least one party identified according to ethnic identity lines.

The impact of internal conflicts on the states in which they occur is dramatic. Along with the disintegration of multiethnic federations such as the Soviet Union and former Yugoslavia, internal conflict has led to the complete breakdown of state authority in several recent cases, among them Liberia, Sierra Leone, Zaire and Somalia, a situation captured by new concepts such as state failure (Helman, 1992/93) or collapsed states (Zartman, 1995a).

In Lebanon, South Africa, Sri Lanka, Israel, Cyprus and Northern Ireland intergroup conflict and disagreement on fundamental issues, notably governance and state legitimacy, have produced deeply divided societies, that is societies where intergroup relations are characterised by separation and segregation.

Internal conflicts differ in a number of important aspects from international conflicts which makes the challenge of ending conflict even more difficult. For example, in situations of interstate conflicts former enemies can retreat to their respective side of a border after having signed a peace agreement. They need not unless they wish to do so engage in further contact and interaction with each other. In most situations of internal conflicts, however, former enemies will have to continue to co-exist on the territory they share also after the conflict has formally ended. Typically these conflicts are protracted. Periods of violence are followed by periods of tranquillity in a recurrent pattern. The continuing threat of intergroup violence prevents the development of trust.

One of the foremost challenges of future co-existence involves addressing the effects of violence on intergroup relations. As neighbours turn on neighbours internal conflict tears at the very fabric of society. As can be witnessed in former Yugoslavia, and notably in Bosnia and Herzegovina, demographic separation is a regular feature of internal conflict. And although segregation might be a rational response to the fact that one's neighbour has come to represent a threat, its legacy presents challenges to the post-conflict reconstruction of society.

Empirical research has shown that internal conflicts are less likely than international conflicts to end through a negotiated settlement: about two-thirds of international but only one-third of internal wars ended by negotiation. The remaining two-thirds ended either in the surrender or elimination of one of the parties involved in the conflict (Pillar, 1983). Furthermore, in the event that a peace agreement is successfully negotiated, the track-record for peace agreements gives further ground for pessimism. Not only are they difficult to achieve; negotiated settlements also often break down. Licklider (1995) found that only one third of the negotiated settlements of 'identity civil wars' ­ that is, struggles between identity groups ­ that last for five years 'stick'.

It is against this background that scholars have begun to explore new approaches to conflict resolution. New concepts such as 'interactive conflict resolution' (Kelman, 1990; Fisher, 1997), 'conflict transformation' (Rupesinghe, 1995) and 'peacebuilding' (Fisher, 1993; Lederach, 1997) have entered the field of conflict resolution theory. What these approaches have in common is a concern for relationship issues and for the wider social context in which conflict occurs.

Increasingly the argument is being made that the top-down approaches of mediation and negotiation, that is, the arena of official diplomacy, need to be supplemented with bottom-up approaches, or grassroots peacebuilding initiatives (Diamond, 1991; Chufrin et al., 1993; Fisher, 1993; Rupesinghe, 1995; Lederach, 1997). The argument is based on the realisation that peace cannot be created at the negotiating table, what a peace agreement can achieve ­ at best ­ is to create the political conditions for peace to develop.

This paper comes out of an on-going research project at the Department of Peace and Conflict Research, Uppsala University, Sweden. The focus of the study is the interrelationship between top level peacemaking processes and grassroots peacebuilding in three divided societies (Northern Ireland, Cyprus, Israel/Palestine). While the project seeks to analyse the relationship between top-down and bottom-up approaches as a two-way-relationship, that is, the impact of one on the other, we will in this paper limit ourselves to two questions: What is the role of grassroots peacebuilding in a peace process? How can peacebuilding facilitate attempts at finding a negotiated settlement?

With references to two of the cases studied in the project, Northern Ireland and Cyprus, we will also look at the impact of the development of the top level peacemaking process on grassroots peacebuilding.

A peace process: three elements, two tracks

The theoretical framework within which we will analyse peacebuilding departs from an understanding of a situation of conflict as involving three interrelated elements: a substantive, a behavioural and a relational element.

The substantive element of a conflict situation refers to the issues on which there exists an actual or perceived incompatibility, that is, the fulfilment of the goals of one of the parties prevents, or is believed to prevent, the fulfilment of the goals of the other party. Conflict behaviour, the behavioural element, is the behaviour (violent or non-violent) in which parties engage in order to achieve their goals in relation to these issues. However, as parties usually respond to conflict by withdrawing from contact and interaction with the other side (avoidance), conflict behaviour should also be taken to include the absence of such behaviour one would normally expect in a peaceful situation. The relational element, finally, are the negative attitudes, perceptions and evaluations the parties hold towards each other and which effects their relationship. These subjective elements are crucial in determining the parties' interaction and hence the development of the conflict.

It is suggested here that a peace process involves a) finding a settlement to the substantive issues in conflict; b) ending violent conflict behaviour and restoring non-violent intergroup interaction, and c) transforming intergroup relations in a peaceful direction. Furthermore, we suggest that c) constitute a central element for the development of a durable peace in divided societies. For reasons to be discussed below, as long as intergroup relations remain antagonistic and dominated by mistrust and fear, attempts at finding a settlement to the issues in conflict will be hampered. Also, in the event of the political leaders of the communities arriving at a settlement to the conflict (or if a settlement is imposed from the outside) in the absence of transformation of intergroup relations the situation is likely to remain unstable. We therefore propose that neglect of the relational element of a conflict situation, that is, failure to transform intergroup relations in a peaceful direction, helps explaining the cyclical pattern of many internal conflict referred to earlier.

Finding a settlement to the issues of conflict and ending violence are the objectives of what will be referred to here as a peacemaking process. Through this process, the conflict shifts from violence to politics. In order to settle intergroup conflicts a political system in which the parties to the conflict feel that they have a stake must be created. The peacemaking process takes the form of dialogue and negotiations between the parties' political leaders. To that extent it is an elite process, involving only a limited number of people directly.

Peacebuilding is here taken to mean a more comprehensive process whose objective is to transform the context of conflict and lead to the establishment of a new relationship between the parties. Unlike peacemaking, peacebuilding is therefore an inclusive process, involving not only the leadership of the conflicting parties but also their constituencies. Thus while peacemaking primarily is a political process, peacebuilding can be described as a process of social reconstruction aiming at building peaceful intergroup relationships.

In the project the empirical investigation is guided by the following working definitions of the two concepts: peacemaking refers to attempts at finding a settlement to the issues in conflict at the political elite level and peacebuilding refers to contact initiatives at the grassroots level aiming at the improvement of intergroup relations.

A public peace process: the case for citizen involvement

How can peacebuilding facilitate attempts at finding a negotiated settlement? For analytical purposes a peace process can be divided into three distinct phases: the pre-negotiation phase, the negotiation phase and the post-agreement or implementation phase. While peacebuilding has a role to play in all these phases, we suggest that its role is particularly important in facilitating the onset of negotiations and in preparing the ground for a settlement.

Groups with a history of violent conflict tend to hold dehumanised images of the other side. These images are embedded in the collective belief system of respective groups. By bringing members of opposing groups together and facilitating direct interaction, peacebuilding initiatives can help to undermine negative beliefs or stereotypes about 'the other'. Image change that occurs as a result of face-to-face contact can help to remove one important obstacle for negotiation, namely the refusal to recognise 'the other' as someone worthwhile to negotiate with. As Kriesberg (1989) reminds us, in order for fundamental transformation to occur it is not enough that this image change occurs only at a leadership level. Major segments of the population must shift to support moves toward accommodation. However, mass media research has shown that 15­20 per cent of a population represents a critical mass for the change of collective belief systems (Montville, 1993). By initiating dialogue and debate, creating a 'lobby for peace' and giving support to those who oppose extremes on both sides, peacebuilding can create an environment conducive for political negotiations.

Thus in the pre-negotiating phase grassroots peacebuilding can help to what has been referred to as the 'entry problem' of negotiations, the fear of political leaders to enter into negotiations with adversary for fear of criticism from those they represent (Kelman, 1995).

While there is a role for peacebuilding in sustaining the political process also when negotiations have been initiated and thus preparing the ground for a political settlement, the real test for peacebuilding comes when a peace agreement has been signed. The successful implementation of a peace agreement will depend on to what extent what Lederach (1997) refers to as a 'peace constituency', has developed, and ­ one should add ­ to what extent the outcome of negotiations reflects what this constituency is prepared to accept.

Peacebuilding in earlier phases can also facilitate implementation of a peace agreement more directly. Through their involvement in peacebuilding initiatives there will be a substantial number of people who for years have been meeting with individuals from the other group. As a result they have acquired intimate knowledge about their culture, fears, views, etc. In the event of a political settlement, these people will be able perform a vital bridge building role between the communities.

To summarise, it is argued here that peacebuilding can play a role in all three phases of a peace process. Peacebuilding can help to prepare the ground for a political agreement, sustain a process of negotiation as well as facilitate the implementation of a peace agreement.

Peacemaking and peacebuilding in Northern Ireland and Cyprus

After this theoretical introduction we will now turn to an examination of the impact of the peacemaking process on grassroots peacebuilding in two cases, Northern Ireland and Cyprus. To set the context for the analysis, we will begin with a very brief presentation of the current state of the peacemaking process in the two settings.

In June 1996, after repeated failures to find a political settlement to the conflict in Northern Ireland after the outbreak of renewed violence in 1969, an inclusive process of negotiation was initiated. All the major parties to the conflict, including the British and the Irish governments, participated in the negotiations. The process culminated in April 1998 with the signing of a peace agreement. Among the signatories were representatives of the largest paramilitary organisations on both sides. The agreement was subsequently endorsed by the people in Northern Ireland and the Republic of Ireland in two separate referendums. Elections to a new regional assembly in Northern Ireland were held in June that year.

The development of the peacemaking process in Cyprus stands in sharp contrast to the development in Northern Ireland. Since the partition of the island in 1974 the geographic separation of the two communities have been almost complete. Despite the absence of intergroup violence, efforts to find a political settlement have been virtually fruitless. The parties' positions have moved further apart.

The latest attempt to find a political settlement through the UN-sponsored negotiations broke down in July 1997 after the European Union's decision to include the Republic of Cyprus in the next phase of the enlargement process. After the Luxembourg summit meeting in December 1997, when the accession talks with the Republic of Cyprus were given a go-ahead at the same time as Turkey was put on hold as a candidate country, the Turkish Cypriots' leaders broke off all contacts. The current Turkish Cypriot position is that they will only re-enter negotiations on a state-to-state basis, that is, if the Turkish Republic of Northern Cyprus is internationally recognised ­ a demand that is unacceptable not only for the Greek Cypriots but also for the international community. Hence while peacemaking in Northern Ireland have made considerable progress recently, the situation in Cyprus has deteriorated.

How has peacebuilding been influenced by the development of the peacemaking process? Here we will seek to answer that question on the basis of interviews with representatives from grassroots organisations in Northern Ireland. In addition, we will present some preliminary findings from the on-going field work in Cyprus.

Ending violence was identified earlier as one element of a peace process. Threat of violence can be assumed to be one major factor in preventing people from becoming involved in intergroup contact initiatives. In addition to the psychological obstacles involved in reaching out to 'the other side' there is a fear of reprisals. Those who do so risk exposing themselves to criticism from their own community and sometimes even to physical danger.

That fear of violence is a primary factor in preventing people from involving themselves in intergroup contact work is borne out by the effect of the ceasefires. The end to political violence was reported to have had a 'mushrooming effect' on the amount of such interaction. More people were willing to become involved.

That peacebuilding attracted more people was thus one immediate effect of the ceasefires. Another effect was that those who already were involved in peacebuilding work found that their importance increased. They were doing the work that more people now saw the importance of, they had the experience that was now in demand. As one interviewee put it:

When there is war, the battle on the streets and so on, we appeared to be of no significance because the man with the gun has the significance and the power.

After the ceasefires those who had the experience of peacebuilding became more significant actors.

When the IRA resumed their violent campaign after a seventeen months' ceasefire the new peacebuilding momentum that had developed after the declaration of the ceasefires suffered a backlash. A realisation that what one interviewee referred to as a 'lack of war-process' had to develop into a real peace process involving a sufficient number of people from both sides provided a learning experience and initiated discussion on how to involve more people and to reach out to those who represented a threat of the return to violence.

However, we should bear in mind that the extension of the invitation to political negotiations also to representatives of the paramilitary groups in Northern Ireland, and the subsequent ceasefires of the main paramilitary organisations, did not mean that all violence stopped. Incidences of sectarian street violence and punishment attacks carried out by the paramilitary organisations have increased. On both sides there are also paramilitary groups opposed to the peace process who continue their campaigns of violence. These groups have also attracted dissidents from the major groups bound by the ceasefire.

Although the ceasefires can be said to have created space for peacebuilding, the continuation of violence and the influence of paramilitary groups thus remained a major obstacle to peacebuilding despite the progress of the peacemaking process.

What about the attempts to find a political settlement? How did they influence peacebuilding? Here we will focus on two aspects of that process; the effects of the initiation of political talks and the effects of a political settlement. As the latter had not yet been achieved at the time of the interviews, what will be discussed are the likely effects a settlement would have on peacebuilding. One finding is that absence of political progress also prevents progress in the peacebuilding field. In the absence of talks at the leadership level it becomes difficult to motivate grassroots to become involved in peacebuilding. On the other hand, as illustrated by this quote, when politicians talk, it is seen as a sanction of contacts also at the grassroots level:

When they are not talking politically it is harder to get people on the ground to talk. They simply take their leadership from their political leaders. Once you get David Trimble [leader of the largest unionist party] speaking with Gerry Adams [the Sinn Fein leader], there is something shifted there that means that the ordinary people are saying that if David Trimble is doing it then...

What impact would a political settlement have on the grassroots work at community level? A recurrent theme was that it would mean that community organisations would be able to have a pro-active rather than a re-active approach to their work. Rather than reacting to emergencies and crisis they would be able to focus on what they would see as the main issues in improving intergroup relations.

In the absence of a political settlement, the situation remains volatile. As Northern Ireland experienced during the summers of 1995 and 1996, and to a lesser extent in 1997, in a very short period of time confrontations can polarise the whole community. During this period peacebuilding virtually came to a standstill, and efforts had to be focused on crisis management.

In addition to facilitating a more strategic approach, some voiced a belief that the stability that a settlement would bring would make it easier to motivate people to become involved in peacebuilding initiatives. It would also help attract funding for such initiatives, as there would be some certainty that efforts invested would pay off and not be 'blown up' by paramilitaries.

While a political settlement is thus believed to facilitate peacebuilding work for the reasons stated above there were also a realisation that it would not solve all problems. Sectarianism, segregation, and the personal traumas are some of the legacies of more than 25 years of violent conflict that would still be there. However, a settlement would create a situation where these problems could be tackled more effectively.

The case of Northern Ireland gives some indications of the impact of progress in the peacemaking process on peacebuilding, but as we have seen, peacebuilding faces several obstacles also when progress is being made in the peacemaking process.

We now turn to Cyprus where the situation is one of political stalemate. At the same time the immediate threat of intergroup violence is absent. However, unlike Northern Ireland in Cyprus peacebuilding initiatives have to overcome physical barriers to intergroup contact. The context of peacebuilding here is one of extreme segregation. Along with the occasional visit to either side or trip abroad most intergroup meetings take place within the UN controlled buffer zone in Nicosia. Intergroup peacebuilding meetings thus literally take place on the interface between two ethnically homogeneous entities separated by a border that is more impermeable than most international borders today. Furthermore, Turkish Cypriots can only attend intergroup meetings within Cyprus itself with permission from their authorities.

These circumstances puts peacebuilding in a precarious situation and makes it difficult for peacebuilding to develop a momentum of its own. The lack of progress in the peacemaking process has had one very immediate effect on peacebuilding. After the Turkish Cypriots leaders broke off the official contacts with the Greek Cypriots in response to the Luxembourg decision referred to earlier, they also introduced a ban on all unofficial contacts between Greek and Turkish Cypriots. Since then permissions to attend intergroup meetings in the buffer zone have been withheld. Those who have attended meetings with Greek Cypriots abroad have faced public criticism.

In both parts of Cyprus there is a fear that peacebuilding initiatives work to undermine the official position in the conflict, and thus that it would weaken one's side in the peacemaking process.

The Turkish Cypriot official position in the peacemaking process is that the current separation must be maintained until the asymmetry between the two communities have been remedied. Turkish Cypriots who are involved in peacebuilding represent a threat to this position. Firstly, they are in general favourably disposed to the establishment of a federation with the Greek Cypriots, that is, they support what is the official position of 'the enemy'. By demonstrating that dialogue is possible, and that there could be mutual benefits for Greek Cypriots and Turkish Cypriots in working together, peacebuilding initiatives are seen by the authorities in the north as undermining their official position in the negotiation. They fear that improved intergroup relations would be a step towards a federation. From their perspective, people involved in peacebuilding therefore represent a potential 'fifth column'.

On the Greek Cypriot side there is less of outright opposition from the authorities. One reason is that peacebuilding can be said to be more compatible with the Greek Cypriot position which favours integration over separation. Nor do the authorities try to control the peacebuilding process by imposing restrictions or requesting permissions. However, those involved in peacebuilding in the south face criticism from vocal nationalists for diverting attention away from what they consider the real problem, namely the Turkish occupation in the north of the island. Emphasising the intergroup elements of the conflict is seen to dilute the position that the problem is an international one of foreign intervention.

Furthermore, the Greek Cypriot authorities (and public opinion) are opposed to everything that could be seen as conferring legitimacy on the current situation and recognition of the Turkish Cypriot political administration.

Thus, we can conclude that the effects of the developments of the peacemaking process on peacebuilding in Cyprus are immediate and direct. The constraints that prevent peacebuilding from developing are also the same that hamper political agreement. Finally, it is interesting to note the importance that the authorities on both sides seem to attach to the political implications of peacebuilding. For them grassroots initiatives clearly represent a political potential that can have an influence on the peacemaking process.

Conclusions and directions for future research

What then can we say so far about the role of peacebuilding initiatives in a peace process? In this paper we have presented theoretical arguments as to why grassroots peacebuilding has a role to play in a peace process as a complement to peacemaking. We have also presented some findings on the interrelationship between these two processes.

What the analysis peacebuilding in Northern Ireland and Cyprus have revealed so far is how the development, or lack of development, of the peacemaking process impacts on the peacebuilding process. We have found support for the contention that at least there exists a one way relationship between the two processes: peacemaking does impact on peacebuilding. In Northern Ireland, peacebuilding fluctuates between action and re-action depending on the political situation. The ceasefires and the start of negotiations had an immediate effect on peoples' willingness to become involved in peacebuilding but did also create new challenges. In Cyprus, the absence of political progress puts obstacles in the way of the development also of peacebuilding.

We have also identified some variables that should be explored further: residual intergroup violence, physical barriers to intergroup contact, definition of the problem, official positions, and leadership dialogue.

Ideally we would now like to proceed with an analysis of the inverse relationship, that is, the impact of peacebuilding on peacemaking. Given the differences in nature between the two processes this linkage is much harder to establish. With peacebuilding being a comprehensive process, it is near impossible to isolate its effects on peacemaking, which would be desirable from a methodological point of view. The method we have chosen is to conduct semi-structures interviews with both peacebuilders and peacemakers. The latter category consists of people who, depending on the situation in the specific conflict, either have been, are or are likely to be involved in political negotiations as representatives of the respective groups. This will at least enable us to understand how the actors in these two processes themselves perceive of this interrelationship. Data from these interviews will primarily serve as illustrations to an argument that will have to be largely theoretical.References

Adekanye, J. 'Bayo 1998. Disarming Ethnic Guerrillas, Power-Sharing and Transition to Democracy in Africa. Ethiopia, Mozambique, South Africa and Uganda as Comparative Cases. International Peace Research Institute, Oslo.

Chufrin, Gennady I. & Harold H. Saunders, 1993. "A Public Peace Process", Negotiation Journal, pp. 155­177.

de Silva, K. M. & S. W. R. de A. Samarasinghe, ed. 1993. Peace Accord and Ethnic Conflict. London: Pinter Publishers.

Diamond, Louise and John McDonald, 1991. Multi-Track Diplomacy. A Systems Guide and Analysis. West Hartford: Kumarian Press

Fisher, Ronald & Loraleigh Keashley, 1991. "The Potential Complementarity of Mediation and Consultation within a Contingency Model of Third Party Intervention", Journal of Peace Research, vol. 28, no. 1, pp. 29­42.

Fisher, Ronald J., 1993. "The Potential for Peacebuilding. Forging a Bridge from Peacekeeping to Peacemaking", Peace and Change, vol. 18, no. 3, pp. 247­266.

Fisher, Ronald J., 1997. Interactive Conflict Resolution. Syracuse: Syracuse University Press.

Galtung, Johan 1978 (1969). "Conflict as a Way of Life", in Johan Galtung ed. Essays in Peace Research, Vol. 3, Peace and Social Structures. Copenhagen: Christian Ejlers.

Hampson, Fen Osler, 1996. Nurturing Peace. Why Peace Settlements Succeed or Fail. Washington, D. C.: United States Institute of Peace.

Helman, Gerald B. and Steven R. Ratner, 1992/93. "Saving Failed States", Foreign Policy, , no. 89, pp. 3­20.

Holsti, Kalevi J., 1991. Peace and War: Armed Conflicts and International Order 1648­1989. Cambridge: Cambridge University Press.

Kelman, Herbert C. 1990. "Interactive Problem-Solving: A Social-Psychological Approach to Conflict Resolution", in John Burton and Frank Dukes ed. Conflict: Readings in Management and Resolution. New York: St Martin's Press.

Kelman, Herbert C., 1995. "Contributions of an Unofficial Conflict Resolution Effort to the Israeli-Palestinian Breakthrough", Negotiation Journal, vol. 11, pp. 19­27.

Kriesberg, Louis 1989. "Transforming Conflicts in the Middle East and Central Europe", in Louis Kriesberg, Terrel A. Northrup and Stuart J. Thorson ed. Intractable Conflicts and Their Transformation. Syracuse: Syracuse University Press.

Lederach, John Paul, 1997. Building Peace. Sustainable Reconciliation in Divided Societies. Washington, D. C.: United States Institute of Peace.

Licklider, Roy, 1995. "The Consequences of Negotiated Settlements in Civil Wars, 1945­1993", American Political Science Review, vol. 89, no. 3, pp. 681­690.

Lukic, Reneo and Allen Lynch, 1996. Europe From the Balkans to the Urals. The Disintegration of Yugoslavia and the Soviet Union. London: Oxford University Press.

Montville, Joseph V. 1993. "The Healing Function in Political Conflict Resolution", pp. 112­127 in Dennis J. D. Sandole & Hugo van der Merwe ed. Conflict Resolution Theory and Practice. Integration and Application. Manchester: Manchester University Press.

Nordquist, Kjell-Åke, 1992. Peace After War. On Conditions for Durable Inter-State Boundary Agreement. Uppsala: Department of Peace and Conflict Research, Uppsala University.

Pillar, Paul, 1983. Negotiating Peace: War Termination as a Bargaining Process. Princeton: Princeton University Press.

Randle, Robert F. 1973. The Origins of Peace. A Study of Peacemaking and the Structure of Peace Settlements. New York: The Free Press

Rupesinghe, Kumar, ed. 1995. Conflict Transformation. London: Macmillan Press Ltd.

Sollenberg, Margareta and Peter Wallensteen, 1997. States in Armed Conflict 1996. Uppsala: Department of Peace and Conflict Research, Uppsala University.

Stedman, Stephen John, 1990. Peacemaking in Civil War: International Mediation in Zimbabwe, 1974­1980. Boulder: Lynne Rienner.

Stedman, Stephen John 1996. "Negotiation and Mediation in Internal Conflict", pp. 341­376 in Michael E. Brown ed. The International Dimensions of Internal Conflict. Cambridge, Massachusetts: The MIT Press.

Zartman, William I., ed. 1995a. Collapsed States. The Disintegration and Restoration of Legitimate Authority. SAIS African Studies Library. Boulder: Lynne Rienner Publishers.

Zartman, William I., ed. 1995b. Elusive Peace. Negotiating an End to Civil Wars. Washington, D.C.: The Brookings Institution.