Working Paper 94-69, 1994.

By Kent Arnold

This paper was written with a small grant from the Conflict Resolution Consortium, University of Colorado. Funding for the Consortium and its Small Grants Program was provided by the William and Flora Hewlett Foundation. The statements and ideas presented in this paper are those of the author and do not necessarily represent the views of the Conflict Resolution Consortium, the University of Colorado, or the William and Flora Hewlett Foundation. For more information, contact the Conflict Resolution Consortium, Campus Box 327, University of Colorado, Boulder, Colorado 80309-0327. Phone: (303) 492-1635, e-mail: crc@cubldr.colorado.edu.

Copyright (C) 1994. Kent Arnold. Do not reprint without permission.

Funded by the William and Flora Hewlett Foundation and the University of Colorado, the Conflict Resolution Consortium is a coordinated program of research, education and application on three of the University's four campuses. The program unites researchers, educators, and practitioners from many fields for the purposes of theory-building, testing, and application in the field of conflict resolution. Current focus areas include international conflict; environmental and natural resource conflict; urban, rural, and inter-jurisdictional conflicts; and the evaluation of alternative dispute resolution mechanisms.


The Conflict Resolution Consortium working paper series includes a variety of papers written by our members as a part of their research. Usually these papers are in preliminary draft stage and are being prepared for eventual publication in professional journals or books. Other papers record discussions from Conflict Resolution Consortium seminars and plenary presentations.

The purpose of the working paper series is to generate a dialogue about the work presented. Readers are encouraged to respond to the papers either by contacting the author directly or by contacting the Consortium office.


This thesis provides a theoretical and practical analysis of dispute systems design as an approach to conflict management.

The research is based in South Africa. During its transition to majority-rule, that country dealt with a number of systems of chronic, destructive conflict. This paper addresses two of these systems, hostel community relations and the black taxi industry.

The thesis first provides a social theoretical analysis of dispute systems design. It uses Collins' (1981) work on interaction ritual chains to explain how the dispute systems design process can modify social structures by implementing dispute resolution roles and procedures designed for specific types of disputes occurring in the conflict system. After this theoretical presentation, the dispute systems design process is applied to taxi conflict and hostel community conflict. The main system of analysis for taxi conflict is in the Western Cape taxi industry. Hostel community conflict in the greater Johannesburg area serves as the foundation for the study of this conflict system.


A project of this size usually requires assistance from a number of people. This effort was aided by quite a few folks, both here in Boulder and in South Africa. Much of the research was made possible by my advisor, Paul Wehr. Paul secured a grant from the Hewlett Foundation that permitted me to return to South Africa in August, 1993. While the financial assistance was considerable, it was Paul's rich knowledge and experience in peacemaking that added the greatest value to this study.

I am also indebted to Oto Bartos and Kirk Williams for their assistance on this project. Oto introduced me to social theory.

I have found his comments to be especially helpful in tying together a series of theoretical threads in this work. Kirk was instrumental in the development of the framework for this research. His ability to conceptualize and ground research projects has taught me much about assumptions I make when entering a system of study.

In addition to my contacts at the university, there were two other people who were very involved in this study. Chris Moore and Mary Margaret Golten of CDR Associates provided me with an opportunity to work with them as an international intern. This internship created the framework for the thesis. Their work on dispute systems design focused me on this emerging field of conflict resolution and management. Traveling with them in South Africa enabled me to learn a great deal about this topic and their contacts provided me with an extensive research base.

In addition to those people I interviewed, there were quite a few others in South Africa that impacted this study. IDASA's Ivor Jenkins is my strongest link to South Africa. Ivor was the person who invited me to come work with Koinonia back in 1990.

He has provided me with a great deal of insight on situations in his home country. I am also grateful to Ron Kraybill, Jean Jeannot, Pearl Mxenge, Isabel Bosch, Pieter Venter, Michael Barry and Judy Merkel who have hosted me on various occasions and given me their interpretations of events that occur in this amazing country.

Finally, I am indebted to the support and encouragement of my circle of friends here in Boulder. I know times arose when all of us were awaiting the finish of this project. Lu and Laura were especially important during this time. Thanks for your patience. I guess we get to switch roles now.



It was the third day of rioting in the Northern Areas in the coastal city of Port Elizabeth. Just 8 months earlier, State President F.W. de Klerk had announced the unbanning of political organizations and preliminary plans to dismantle various pieces of apartheid legislation. The euphoria of de Klerk's announcement faded into the anomie of a society not quite sure how to define its new mission. The 1990 riots in Port Elizabeth erupted from a rent and school protest in the so-called coloured area of the city. I was traveling in the car of a well-known Anglican priest to a meeting at a church near the heart of the rioting. As we drove past South African Defence Force armored personnel carriers and picked our way through burned-out barricades in the narrow streets, the crowd closed in around the car. When it seemed inevitable that our automobile would be set alight, the priest opened his door, stood and spoke to the those who surrounded us.

For people who grow up in an anabaptist religious setting, there are numerous opportunities to hear the stories of peacemakers. Having roots in both the Mennonite Church and the Church of the Brethren, created fertile soil for me to learn more about the field of conflict and peace. It was a fellow Mennonite who first taught me how to mediate conflict and it was my Brethren background that provided me with the contacts I needed to make my original journey to South Africa in 1990.

I agreed to spend a year assisting a friend who was running a Christian reconciliation program called Koinonia, Southern Africa. Koinonia attempts to bring together people from all religious and political persuasions to bridge the social and physical divides created by apartheid structures. I worked in various capacities for this organization, from assisting regional coordinators to managing the operations of the national office in Johannesburg. This allowed me to travel to most provinces throughout the country, observing regional political, economic, and racial/ethnic nuances. It also permitted me to experience many types of conflict first-hand. I visited violence-torn townships near Johannesburg and in Natal. I worked with foundations in the rural Eastern Cape facilitating discussions about economic and social development projects for black communities. I met with church leaders who were struggling with emerging issues of fostering participatory democracy and promoting black leadership in non-racial structures. I lived with political leaders who helped me distinguish and contextualize what, at times, seemed subtle differences in party politics.

In 1990, most organizations challenging the government struggled to redirect their mission. It was a rich time to observe organizations and a society in flux. I had numerous opportunities to work with groups of people engaged in the creative process of negotiating structure within the void borne from this societal transformation. The setting provided me with invaluable lessons about conflict which, in turn, yielded interesting tales of peace.

Peacemaking and the transformation of South Africa into a democratic state formed two-thirds of the basis for this thesis. The other third came from both my international internship with CDR Associates and my previous professional career as a systems programmer in the field of computer science. I am intrigued with concepts that bridge academic disciplines. While assisting Chris Moore and Mary Margaret Golten in preparing for a series of seminars on conflict management and resolution in South Africa, I was introduced to dispute systems design. The idea of designing systems for handling disputes was new to me, but the process of systems analysis was a well-worn path. It was gratifying to find a way to apply what I learned in the field of computer science to the resolution of conflict. The ensuing study allowed me to meld interests in peacemaking, systems analysis and South Africa into this applied-theoretical examination of conflict.

The purpose of this thesis is threefold. The first is to present a historical analysis of conflict in the migrant labor hostels and in the black taxi industry. This will be done by describing both of these systems of conflict, showing how the focal participants are waging the conflict and then presenting various attempts at creating peace. The second objective of this study is to prescribe a dispute resolution system which is designed to reduce violence and destructive forms of waging conflict. The final purpose is to contribute to the body of conflict management theory. I will do this by tracing the societal implications of dispute systems design back to the Collins' (1981) analysis of interaction rituals and Simmel's (1955) and Coser's (1956; 1967) writings on conflict's effect on groups.

Following this introduction is a brief overview of the social, political and economic history giving rise to conflicts in the taxi and migrant labor industries. This section provides the background needed to understand the selected conflict systems. Following this overview is a chapter which describes the process of designing dispute systems. This section also contains a theoretical analysis of dispute systems design. In Chapter 5 I will apply the theory of dispute systems design to two specific conflicts. One conflict system involves the taxi industry of the Western Cape, while the other looks at hostel community violence on the East Rand. The conclusion ties together dispute systems design theory and praxis, showing how each informs its counterpart. This final section also addresses some of the limitations of this approach.

Dispute systems design seeks to implement a wide range of decision-making, conflict management and conflict resolution procedures in order to empower people in communities and institutions to handle chronic, destructive disputes. This thesis is about how to do this and why South African society should hold out hope that it can work. When violence occurs in that country, power-based methods for restoring order are often used. By not considering the long-term effects on the community, coercing conformity to norms defining acceptable means to wage conflict is tempting; however, this is a costly practice.

Dispute systems designers look for ways to encourage relationship building as a way of sustaining peace. In South Africa, there are a number of notable achievements in doing this. A few provided the toehold that peace needed to start the transformation of power-based systems of destructive conflict.

As the priest spoke, I stared out the car window watching the retreating crowd. The account he gave them allowed us to proceed. In a conflict that claimed more than 60 lives and left hundreds wounded, I wondered aloud why the crowd let us go. The priest said that maybe they still had "respect for the cloth." This encounter was a small slice of a conflict that ran for five days. It is probably only remembered by those of us inside the automobile. Nonetheless, it had a significant impact on how order was restored because that priest was one of the primary players in the ongoing negotiations. In violent forms of conflict it is easy to discount these micro-interactions as impotent. But sometimes courageous acts create momentum which dispute systems designers can exploit to nurture the peace process.



I stood on the floor of Kwazekhele Stadium squinting up at the Port Elizabeth sky. Overhead a couple of South African Defence Force (SADF) helicopters monitored the now-legal activities of the people below. As SADF pilots buzzed the packed stadium every few minutes, speakers' voices were drowned-out by derisive whistles. The once banned colors of the ANC, South African Communist Party and Pan-Africanist Congress were now unabashedly displayed amongst the more than 17,000 supporters who were commemorating the 14th anniversary of the June 16 student uprising in Soweto. Only four months earlier on February 2,

1990, State President F.W. De Klerk announced the National Party's intentions of moving away from its racially-based social engineering policies known as apartheid. With the unbanning of all political parties and the release of Nelson Mandela from prison, came a feeling of euphoria to the millions inside and outside South Africa who waged the struggle for democracy.

On this June 16 standing amongst thousands of toyi-toyiing protesters, I listened to ululating voices sing and chant. As their feet rhythmically pounded the bare stadium ground, a cloud of red-orange dust rose auspiciously with their hopes of freedom and justice. This ritual, acted out in the classic Durkheimian sense, was charging the emotional batteries of black people at similar events throughout the country. It stood in stark contrast to the gathering I observed just two weeks before at the Voortrekker Monument. On that day, nearly 100,000 white conservatives met on a hilltop outside Pretoria to solemnly proclaim the start of the Afrikaner nation's third fight for freedom.

Now three years later, recollections of those rallies ran through my mind as I sat in the back seat of a white Opel Kadet zipping through a township northeast of Johannesburg. From 1990 to this September day in 1993, Tembisa had seen its share of the violence engulfing many townships on the East Rand. According to the Tembisa Local Peace Committee members, things were now relatively quiet here. But that was not the case in places like Thokoza and Kathlehong south of Johannesburg. The group I was traveling with was returning from the launch of the latest WitsVaal Regional Peace Committee project which was placing two armored personnel carriers in these townships to monitor violence. Few weeks pass without news of bloodshed in the townships in the Vaal Triangle. In a scene played out too many times in this region, we drove down the road leading out of Tembisa, passing a helmeted soldier standing over a black man lying in a pool of blood on the side of the road.

The high levels of violence experienced in South Africa today have strong ties to the development of the apartheid system by the National Party. After their surprise victory over Jan Smuts in the 1948 general election, the National Party (NP) set about proscribing its policy of separate development. The Population Registration Act of 1950 was one of the first pieces of apartheid legislation. It forced a rigid classification of all people along racial lines. This act, along with the Mixed Marriages Act and Group Areas Act, formed the pillars of apartheid (Omond, 1986). Ensuing pieces of legislation would attempt to extend governmental control of social, political and economic relations for people of all races in South Africa. From 1948 to 1990, the National Party would dominate the South African political scene, never relinquishing the state presidency.

During this time, pass laws, forced removals of black communities, bannings of individuals and political parties, and detentions without trial brought the government international infamy.

While reasons for the present-day violence can easily be linked to the hegemonic grip of the National Party, "racial separateness" was strongly rooted in South African society before the NP's victory in 1948. Prior to the latter half of the nineteenth century, the inward movement of Dutch and British settlers north and east from the Cape Province created the most significant imposition of European culture on Southern Africa; however, when diamonds were discovered in the Orange Free State in 1866 and gold near modern-day Johannesburg twenty years later, the emerging mining industry would sweep this pastoral society into the industrial age (Sparks, 1990). The demand for plentiful labor pulled black people from rural settings into urban areas near the gold and diamond mines, but the 1913 Natives Land Act prevented them from purchasing property in these growing cities. This Land Act set aside just ten percent of the country's land in regions designated as native reserves for purchase by Blacks (Omond, 1986). By not permitting blacks to own property in the newly developing urban settings, South Africa started down the path of apartheid.

Much of today's violence in the townships is rooted in the social trauma borne out of the development of the gold and diamond mines. While the mining industry grew, drought, plague and sub-division of family property reduced the economic viability of farming in the Afrikaner community, forcing many to compete with blacks for jobs in the mines in the rapidly expanding cities. In The Mind of South Africa, Allister Sparks described this event:

Neither the Afrikaner nor the black tribesman was equipped to cope with life in the city the alien city where the Englishman was baas... The impact on both was immense. It hardened and changed the whole focus of Afrikaner nationalism, switching its reference point from conflict with the English to conflict with the blacks. And it ripped apart the fabric of black tribal life, turning a landless peasantry into an urban proletariat...(1990:121)

Paradoxically, the mines created a need for a large, experienced labor force near the cities, but property rights prohibited blacks from owning land in these areas. To resolve this problem, mining companies developed a migrant labor system. The hostels built to house these migrant workers are at the heart of some of the worst violence in contemporary South Africa.


Migrant workers came from all over southern Africa to work the mines. Contracts to work as a migrant laborer were temporary, usually in 11 month stints. Initially, the workers were housed in single-sex units and in some cases these compounds were closed, prohibiting the workers from leaving the area for the duration of their contracts. This was done to control discipline and to prevent theft from the mines (Turrell, 1987). Mine operators' interests in efficient production and profit led them to create a cheap, docile, experienced labor supply through the migrant system.

Mining companies made a number of attempts to stabilize their work force. They sought to promote permanent residence on the mines through the creation of family housing. Mine operators also tried to develop a portion of their work force into professional miners, encouraging long term employment with a specific company. These attempts did not succeed for the following reasons:

These factors combined to create an environment where large numbers of men live in over-crowded, single-sex dwellings that they consider only temporary homes.

Internal conflict in the hostels has occurred since the beginning of the century. At that time, hostel units were often divided amongst the different ethnic groups residing in them. Originally it was thought that factional fighting was due to the workers' ethnic loyalty. It was not until the late 1970's that ethnic division in the hostels was abandoned. This decade was a time of great unrest throughout the country. A study of mine conflict of that time period cites three causes of violence: faction fights, wages, working and living conditions, and changes in the employment status of foreign workers (Crush, et. al., 1991). By 1976 the violence had abated, leaving 132 mineworkers dead and approximately 500 injured (Lipton, 1980). A government commission of inquiry into the violence recommended that the mining industry upgrade the hostels, create family units, and abolish ethnic division of workers. While many of these requests were attempted, they had marginal success (Wentzel, 1993). There was some reduction of the internal violence in the hostels but this changed dramatically in 1990.

Hostels display many features of total institutions with the residents as inmates, cut off from the surrounding communities. Living conditions, which vary from compound to compound, are deplorable in most units. In areas where there has been a high degree of violence between the hostel and the neighboring community, the compounds are sometimes surrounded with razor-wire creating a coercive reminder of which group the migrant belongs to.

The bed is the total claim to space a worker can make in a hostel. Even then, this bed is in a room with 12 to 16 other men and may be shared with a man working a different shift. On the Reef, an individual's basic space consists of a concrete bunk whose top is removable, acting as a trunk for the worker's personal belongings. If the beds are steel or if the concrete bunk has no trunk space, the migrant's clothing is stowed on walls or under the bed. In some cases there are metal lockers, but these are usually not lockable. There are few provisions made for securing workers private belongings. Typically there is one communal washroom and lavatory with little or no hot water and no arrangements made for privacy. Facilities within the hostels are frequently in a state of disrepair. Workers are sometimes found using water from toilet tanks when taps are not functioning. While conditions in some parts of the host township might be comparable, the combination of the lack of working facilities, absence of privacy, and intense alienation of migrant laborers make hostels volatile environments (Minnaar, 1993).

Conflict between hostels and the surrounding communities has a long history. Hostel workers were historically viewed as country bumpkins by their sophisticated urban counterparts.

Their physical and social segregation served to reinforce these stereotypes (Wentzel, 1993). Since the unbanning of the ANC in 1990, violence has flared in a number of regions, most notably on the East Rand. In this region hostel community conflict usually takes a political form. Many hostel residents on the Reef are of Zulu descent. People in the community associate these workers with the Inkatha Freedom Party regardless of whether or not they actually support that party. The squatter settlements around the hostels are seen as ANC strongholds because they are mostly populated with Xhosa-speaking people. These perceptions exacerbate the tensions between the communities and the hostels (Minnaar, 1993).

Industrialization, which fueled urban growth, attracted a flourishing black population on the edges of its cities. This prompted the government to enforce a number of regulations controlling the influx of black people into these developing areas. By 1958, each African male had to carry a pass book to prove that he was able to reside legally in a specific area. For an African to live legally in white urban areas, he had to have "Section 10 rights." Section 10 of the Urban Areas Consolidation Act of 1945 generally meant that the person either had worked in the given area for more than 10 years or had a permit from a labor bureau to work in that area.

For this burgeoning black population, there emerged a number of formal and informal markets, one of the more important being transportation. Public transportation demands arose to handle the daily commute to work and for migrant laborers who needed to return to their respective homelands periodically to visit family. The failure of the government to provide an adequate public transport system combined with various public transport boycotts in the 1980's to create a thriving market for taxi operators (Goldstone, 1993; Minnaar and Torres, 1993). This market grew dramatically throughout the 1980's by providing a cheaper alternative to public transportation. By 1993 there were between 60,000 and 70,000 minibus taxis operating throughout the country garnering approximately 45% of the daily commuter market, nearly double the market share of any transportation alternative in South Africa (Pretorius,1993).

With the taxi industry originally being one of the few legitimate entrepreneurial opportunities for blacks, competition in this market grew intense. Minnaar (1993) argues that the industry became politicized as it matured. Revenue from the taxi transportation system was an important source of black empowerment and some of the generated income was used to fund black political activities during the 1980's. When the African National Congress (ANC) and a number of other political organizations were unbanned in 1990, the power struggle for political influence heightened the need for rival taxi groups to become politically aligned.

The idea that political rivalry and affiliation play a role in taxi violence is countered by the Goldstone (1993) Commission of Inquiry on this issue. The Goldstone report points more to economic and administrative problems as the primary causes of taxi violence. Nonetheless, this conflict seems at times to wear a political mask, as evidenced by the July 1993 massacre in Germiston, outside Johannesburg. A minibus was stopped and passengers were separated into ANC and Pan-Africanist Congress (PAC) and Zulu-speaking groups. Gunmen then executed the Zulu passengers, under the presumption that they were Inkatha Freedom Party (IFP) supporters (Minnaar and Torres,1993). One facet of this complex conflict seems to manifest itself in a distinctly political nature; however, there seem to be other issues more fundamental to taxi disputes.

Some of the more significant non-political problems have arisen from the administration of permit allocations for operating taxis. With the Urban Areas Consolidation Act controlling where people could live, two types of taxi permits developed over time. First, daily commuter needs provided a market for taxis operating within a certain radius, often in townships or homelands around major metropolitan areas where jobs were plentiful. The second need was to transport migrant workers to and from their respective homelands on a weekly, or less frequent basis. Radius and route permits evolved, controlling access to these different markets.

In municipalities, the granting of permits was officially controlled. In early years permits were tightly regulated, but since the start of deregulation of the industry in the late 1980's the number of permits has skyrocketed creating a glutted transport market (Minnaar,1993). In squatter settlements which grew near metropolitan areas, a different problem emerged. There were no municipal structures to govern the allocation of taxi permits in these areas, so rival taxi organizations fought for control of these new commuter markets. Even with the new markets, the demand for taxi service was far outstripped by the supply generated from the tremendous increase in permit allocations.

The resulting increased competition combined with the costs of entering and remaining in the market to drop profit margins for taxi operators. Obtaining a permit, a certificate of fitness for the mini-bus, income forfeited during the lengthy application process, and hefty taxi association dues (in some areas the cost is nearly R5000 or approximately $1500) all worked against the legitimate taxi operator (Goldstone, 1993). These costs created a number of problems in the industry.

First, "pirate" taxis operating outside the regulation of the industry now account for almost 50% of the taxis operating in an area. Second, taxi drivers, in an effort to keep their businesses profitable, started overloading minibuses and engaging in dangerous driving practices. Third, rival taxis and operators and passengers of other modes of public transportation (e.g. buses and trains) became victims of intimidation campaigns (Goldstone, 1993). Fourth, traffic police were accused of accepting bribes and harassing taxis during peak hours by stopping them to examine their permits or to perform roadworthiness inspections (Minnaar and Torres,1993).

These factors combined with market forces to create a situation where taxi businesses quickly become marginally profitable. Making monthly payments on mini-buses and dealing with increased insurance, fuel and maintenance costs, created a situation where operators were operating at a loss (Weekly Mail, September 10-16, 1993). The mass action campaign launched by taxi organizations throughout the country during the September 1993 petrol price increase helped raise awareness about the crisis situation of the taxi industry.

Since the unbanning of political parties in 1990, conflict in the taxi and migrant labor industries has escalated dramatically. A number of conflict management agencies have worked to help parties involved in the conflict resolve their differences. For example, some local peace committees of the National Peace Accord (see Appendix B for a discussion of the structure and purpose of the National Peace Accord) have played a significant role in these two types of conflict. A few projects, truces, and agreements have had success in keeping, making, or building peace. Still, there remains a great deal of violence amongst taxi associations and between migrant labor hostels and their surrounding communities. These systems of conflict are complex and, at times, interrelated. To effectively reduce levels of violence in the system, one must implement a series of conflict waging procedures that are geared to mitigate specific types of arising problems. I will present the theory undergirding this dispute systems design approach and then apply it to problems experienced in the black taxi industry and with migrant labor hostels.



Simmel (1955) saw conflict as an ubiquitous feature of all social systems. He noted how associative and disassociative processes blend to develop social relationships within systemic contexts. He attempted to resolve the tension between the dualistic elements of unity and discord by moving the level of abstraction from individual relationships to the group, showing how discord's negative nature in micro-interactions can change to a positive role for the social system. In addition, he recast unity as the ultimate wholeness of the group covering "the total group-synthesis of persons, energies, and forms" (1955:17). In order to reconcile this tension between unity and discord for both the individual and the group, society develops behaviors, norms, procedures, and institutions governing the engagement of conflict.

People use these behaviors, norms, procedures, and institutions in an attempt to resolve conflict in either powerbased, rights-based, or interest-based ways (Ury, Brett and Goldberg, 1988). Power-based methods determine which party is stronger. Rights-based procedures rely on independent standards to determine how to mete out justice. Interest-based methods attempt to reconcile needs, desires, and fears of the parties to the dispute. Comparing the advantages and disadvantages of each option, Ury, et.al. (1988), developed the following proposition:

...in general, reconciling interests is less costly than determining who is right, which in turn is less costly than determining who is more powerful. This proposition does not mean that focusing on interests is invariably better than focusing on rights and power, but simply means that it tends to result in lower transaction costs, greater satisfaction with outcomes, less strain on the relationship, and less recurrence of disputes. (1988:15)

As a system's conflict practices grow more power-based and rights-based oriented, parties' engagement of disputes tends to become more costly. In a distressed system, people will expend excessive amounts of time, money, energy, and emotion in dealing with conflict situations. Over time, the system as a whole will have greater dissatisfaction with solutions to problems and the strain on relationships will weaken the integrative energy that conflict provides.

As the intensity of conflict increases, its integrative value for the social whole is compromised. Instead of conflict creating cohesion in the system, the parties to the dispute develop more clear-cut boundaries. This creates mistrust, polarization, and animosity which impede the effect of interestbased conflict resolution techniques. The inability of the system to provide effective and adequate avenues for grievance redress will result in people or parties questioning the system's legitimacy (Coser, 1956).

When protracted conflict has attenuated the cohesion of the system, there are usually multiple problems to address requiring a variety of solutions. Tactical conflict resolution techniques such as truces or peace monitoring may have little or no impact on the conflict because they do not focus on its root causes.

The repeated disregard of cease fires arranged by the European Community in the former Yugoslavia is evidence of this problem with tactical solutions. Alternatively, a high degree of conflict escalation may prevent the initiation of strategic resolution options like creating democratic decision making bodies or building economic development projects because parties mistrust one another greatly.

In systems experiencing costly levels of conflict, a comprehensive approach is needed to identify the contents of the various types of disputes, why they recur, how they impact the system, how they are handled, and why certain conflict management procedures are used and others abandoned. Dispute systems design (DSD) incorporates these elements of conflict analysis and attempts to resolve the tension between unity and discord for the individual and the group. This process is fostered through a systematic approach to decision making and conflict management which focuses on empowering individuals or groups of people in communities and institutions. Empowerment is engendered through the creation of roles, procedures, and institutions designed to assist decision making and address chronic forms of conflict.

Moore, et. al. (1992) describe 6 aspects of the DSD process:

1. Identifying types and causes of recurring issues and conflicts.

2. Developing and institutionalizing a range of decision making, conflict management, and dispute resolution procedures which parties can use to make decisions, lower the occurrence of destructive forms of conflict, and assist parties in the resolution of differences.

3. Matching issues and disputes with the appropriate decision making, conflict management or resolution procedures.

4. Implementing efficient operations and administrative procedures to manage and track the newly created dispute handling and decision making systems.

5. Designing, developing and implementing information programs which promote the new dispute systems and educate people about how the range of decision making and dispute handling processes can assist them in reaching settlements and resolving conflict.

6. Recruiting and training people to provide services needed by these newly created systems.

This systematic approach to decision making and conflict engagement has applicability in small spheres of conflict, like communities and organizations, and larger societal spheres such as regional or national institutions. Systems development enables the designer to integrate local, regional, and national peace initiatives into a comprehensive framework for the transformation of arenas of intractable conflict. While conflict contents within a targeted sphere of intervetion may remain the same, newly created decision making and conflict handling mechanisms and procedures reconstruct the forms these conflicts take. This empowers people by reducing costly aspects of conflict while fostering the process of recreating norms governing the constructive engagement of conflict within the system.

Key elements of systems theory

The dispute systems designer must make a number of assumptions about the systems under examination. The following properties of systems theory are fundamental to this study of dispute systems design:

1. The systems under analysis are open systems.

2. These systems are cybernetic in that they use negative feedback to regulate themselves and positive feedback to increase their capacity for goal attainment and adaptability.

3. The systems seek to preserve themselves.

4. They require a constant input of energy.

5. Identifiable units within a system influence one another.

6. Changing an input to a subsystem will have ramifications throughout the system (Turner,1991).

It is the task of the dispute systems designer to determine how the above elements come together to create the system under observation. The designer looks for inputs into the system and seeks to understand how changes to inputs affect the system. To accomplish this the designer must search for identifiable units and note how these units impact other units in the system. By modeling these direct, indirect, and feedback influences, it is possible to make assumptions about the order of the systemic whole. Knowledge of these principles of systems theory is necessary to diagnosing, designing, and implementing a conflict resolution and decision making system.

Principles of dispute systems design

There are a number of key principles the designer must incorporate into any analysis of dispute systems. Moore, et.al. (1992) identify the following:

1. The dispute systems designer should view allorganizations and communities as systems

2. The dispute systems design should emphasizeintegrative and interest-based approaches to dispute resolution. This means that the DSD should attempt to put dispute mechanisms in place that seek to maximize the degree to which all parties' interests are met while ensuring that fundamental rights ofmembers of the system are protected.

3. When interest-based decision making procedures do notwork or are not appropriate, the system should develop rights-based procedures that are flexible and low-cost in terms of system resource and relationship damage. When interest-based approaches break down,often power struggles (e.g. strikes) or highly adversarial alternatives (e.g. litigation) are resorted to by the parties involved in the dispute. A systems approach attempts to find lower cost rights-based approaches (such as arbitration) in order to maintain relations and reduce the strain on institutional resources.

4. DSD should work to ensure that issues are resolved by the affected parties. This is typically done by cooperative problem solving and decision making. Empowering the parties to solve their own problems creates an integrative energy which reinforces acceptability of the newly designed mechanisms for handling conflict or decision making. While this process of empowerment is vital to the parties, they must also be protected from negative and futile attempts at resolving problems beyond their abilities. In these cases, the system needs to provide avenues for third party assistance to prevent conflict intractability.

5. When parties cannot solve their own problems in an interest-based manner, recommended procedures should focus on creating opportunities for parties to engage in collaborative processes. The dialectic nature of conflict may create possibilities for disputants to return to interest-based options. An effective DSD will ensure flexibility by permitting parties to move away from adversarial alternatives when the efficacity of integrative options re-emerge.

6. A community's or organization's cultural views towards things like relationships, cooperation,competition, communication, time, and space must be reflected in both the process and goals of thedesign.

7. While understanding that conflicts are an important source of energy for the system, it should be noted that prevention is more powerful than intervention.The goal of DSD is not the suppression of conflict but the appropriate expression and engagement ofconflict. It is possible to anticipate many forms of conflict and put mechanisms in place to diffuse potential conflict situations by dealing with issues prior to escalation of the dispute. This is not done to inhibit conflict, but rather to govern the form it takes.

8. DSD should seek to provide multiple options for addressing conflict. There is no single procedure appropriate for all conflicts in a system.Cooperative problem solving and decision making, peacekeeping, peacemaking and peacebuilding all require different types of procedures. Some conflicts should be immediately referred to third parties while other conflicts can be managed or resolved by the affected parties. DSD must incorporate a process of evaluating each conflict, determining what issues to address and which mechanism or mechanisms to use.

9. All dispute systems require monitoring. Dispute systems are usually implemented in stages. A process of step-wise refinement is necessary to ensure that procedures and structures put in place are meeting the needs they were designed to address. Since dispute systems are cybernetic, they will use feedback to change over time requiring continual evaluation and maintenance of the effectiveness of the design.

Dispute systems design involves more than technical assistance in developing a conflict management program such as the creation of grievance procedures or a mediation service. While the designer may provide this type of assistance, it is only one aspect of the total process. To produce a functioning system, the designer must determine what changes need to be made, develop a support-base for these changes, handle resistance to change, and motivate and train people to use the new procedures. The goal of the DSD process is to develop a system which attempts to reconcile interests amongst the parties when possible (see figure 1). In situations where interest-based procedures cannot or should not be used, the DSD process should attempt to find lower cost methods to determining who is right or who is more powerful (Ury, et.al., 1988).

Figure 1. Moving from a distressed to an effective dispute resolution system (Ury, et.al., 1988:19).

The theoretical framework for dispute systems design

This section will merge Collins' (1981) work on microfoundations of social structure with Moore's (et.al., 1992) and Ury's (et.al., 1988) theory on dispute systems design. The purpose of this section is to show how designers of dispute systems can use Collins' concept of interaction rituals stretching across space, time, and number of people to create or recreate social structures within systems of conflict. This section will address the following questions:

Interaction rituals and the creation of sacred objects in dispute systems design

In Collins' analysis, an interaction ritual is a conversation which has the following properties:

These properties are diagrammed in figure 2.

Group members who believe in these sacred objects are considered "good." These individuals defend the group against those who either do not "believe" or attack the symbols which provide group cohesion. The objects are sacred to the group because they are the means used by the group to unify itself. Successful interaction rituals produce varying degrees of solidarity which, in turn, create and recreate social order (Collins, 1981).

Figure 2. Properties of an interaction ritual (Turner, 1991:233).

One of the first tasks of the dispute systems designer is to convene a change team which represents all the stakeholders in the system of conflict. This group starts the creation or identification of sacred objects for the system. A system's view towards relationships, cooperation, competition, and communication gets acted out in the process of designing a dispute system. Allowing the change team to construct what is sacred about relationships, communication, etc. ensures that interests from all of the parties represented are included. For example after years of authoritarian rule in many South African institutions, there is a current trend towards developing participatory democracy. "Participatory democracy" has become a sacred object for certain groups. It is defended fiercely from attack. The designer must allow the change team to create and model group symbols in an effort to inculcate their ideals in the group's consciousness. This is necessary because these change team members will promote the sacred objects to their constituencies.

In addition to these value-oriented ideals, there are other symbols that may become sacred to the change team. If representatives of certain groups are only present due to a truce or because some agreement on a code of conduct was made, this truce or code is sacred to the proceedings. Any violation is then reason to abandon the negotiations due to the desecration of the symbol of "good-faith." Change team members bring to the dispute systems design certain elements that are sacred to their constituencies. The evolution of the change team requires using or transforming these objects into something that the whole group agrees is sacred. Successful interaction rituals for the group depend on commitment to these objects. This generates moral solidarity which provides the momentum needed to keep the group together as it moves through each stage of the design.

If sacred objects work to maintain group cohesion, it seems important to know what will happen if there is a breach in the group. Collins (1981) argues that the world is too complicated for individuals to evaluate courses of action cognitively in a conversation. So in most exchanges, people operate under a tacit understanding that there is a common definition of the given situation. When this sense is disrupted, perhaps through some profane act in the interaction ritual, conflict becomes manifest. Lederach (1988:39) writes: "[c]onflict situations are those unique episodes when we explicitly recognize the existence of multiple realities and negotiate the creation of a common meaning."

One of the things that disputants bring with them as they prepare to "negotiate the creation of a common meaning," is their definition of the situation. As noted previously, in distressed systems of conflict disputants tend to choose power-based or rights-based dispute resolution practices too frequently. This turns conflict situations into a competition amongst the disputants (Ury, et.al., 1988). Part of a disputant's definition of the situation is that one either wins or loses in a conflict encounter. This ingrains the perception that this is a hostile environment because winning the encounter becomes a sacred group object.

By focusing on the development of interest-based means of dispute resolution, dispute systems design attempts to reconstruct the definition of a conflict situation as hostile by creating dispute encounters that are friendly. This is done by controlling the environment within which the disputants work to reconcile each party's interests. By altering the definition of the situation from "hostile/win-lose" to "friendly/win-win," DSD gives disputants the opportunity to engage in a safe, successful conflict micro-encounter. Successful interactions then become sacred, through both the agreement reached and the skillful manipulation of new dispute resolution techniques (e.g. listening/reframing skills, option generation skills, etc.). The perception that conflict situations are hostile starts to erode with ensuing successful interactions.

How negative micro-encounters reproduce themselves

Theory on the speech exchange system of ordinary conversation gives insight into what takes place during conflict interactions and how these conflict situations reproduce themselves. In the micro-encounter, oppositional utterances are used to accuse the other disputant. Oppositional utterances require a paired reply from the second participant in the form of a preferred response such as a denial, or a dispreferred response such as an account, an admission of guilt, or delay in response (Pomerantz, 1984:64).

Typically actors in encounters operate on a principle of preference for agreement. In disputes, accusations work conversely because preference for agreement with the accuser would indicate an admission of guilt, causing a loss of face. In these cases, the lack of delay in the preferred response, denial, inhibits either participant from repairing the encounter. In addition, accusations often foster counter-accusations which puts the accuser in a position of either responding with a denial or losing face. The preferred response of dispute conversations creates a structure which reproduces itself over time (Whalen, Zimmerman, and Whalen, 1988:353-354). Once a dispute has commenced, the potential interactional disincentives (i.e. some form of self-degradation) for breaking the cycle of oppositional utterances, become the motivation for continuation of the conflict interaction (Garcia, 1991:822).

In the South African context, this idea of oppositional utterances is often played out at the highest levels of negotiation between the government and the ANC. For example after the Boipatong Massacre in 1992, the ANC accused the government of harboring white extremists in its security forces who were directly or indirectly responsible for the attack. The government's paired response could have taken the form of an agreement (and subsequent loss of face) or a denial which would force the ANC to substantiate its claims. By denying its involvement, the government kept the conflict alive.

What broke the escalatory spiral of this conflict was an agreement to have an international police expert evaluate the South African Police's (SAP) handling of the situation. While the report did not charge the SAP with complicity in the attack, it did condemn their abilities and lack of preparedness. This created a loss of face for the government; however, it occurred long after the massacre, softening the blow to government prestige. The decision to bring in a neutral third-party to investigate became a sacred object for both the ANC and the government because it met important interests for the two parties involved. For the ANC, it provided confirmation of its complaints of police ineptitude and a retrospective account which helped absolve them of responsibility for the intractability of township violence. For the government, the investigation bought them time for damage-control and it mitigated the sanctions leveled against them both domestically and internationally. By using the structure of an international investigation and agreeing to its findings, both parties were able to demonstrate their reverence for this sacred object, making the conflict interaction successful. The emotional energy produced through this success increased trust between the parties. This exercise became one of the planks bridging the gap between the National Party and the ANC.

If the parties refuse to identify reference points of mutual solidarity, they will get trapped in the chain of oppositional utterances. In the framework of a dispute interaction that continues, there is a potential for escalation. When this occurs, a number of dynamics emerge:

When conflict escalates, disputing parties move too frequently to trying to determine who is right or who is more powerful. This increases transaction costs, relationship costs, and dispute recurrence, while decreasing satisfaction with dispute outcomes (Ury, et.al., 1988).

An individual's response to intractable conflict will depend on that person's position in the organization or network. For actors who take subordinate positions in a dispute, unsuccessful microrituals will create a deficit in their emotional energy which will lead to less confidence in further interactions (Collins, 1981:1002). In more egalitarian exchanges as the individual's investment in the conflict mounts, a situation is created where incurring additional loss of emotional energy through some act of self-degradation such as "backing-down" becomes less attractive as an alternative. Those individuals who are able to exert their authority in these conflict interactions will experience an increase in emotional energy. In these cases there is a motivation to seek more domination in the exchange or abandon the relationship (Collins, 1981:1007). Regardless of whether the actors are in subordinate, equal, or dominant positions in the system, there is a possibility that the dispute will become protracted if interest-based resolution options are rejected and costly resolution techniques are used.

How dispute systems design reduces costly conflict forms

In chronic, destructive conflict, the goal of creating a common reality is often surrendered, preventing the interactants from demonstrating their solidarity. If interaction rituals are unsuccessful, the participants experience a deficit in emotional energy. This, in turn, affects future interactions through its negative impact on the individuals in terms of self-confidence and social warmth (Collins, 1981:1003). When conflict is protracted, the interactants may find decision making or dispute resolution techniques ineffectual and frustrating. This often occurs when parties attempt to use the wrong procedure for a certain type of conflict. For instance if the parties are engaged in peacemaking exercises, such as mediation, but hostilities have not ceased, it may prove difficult to keep everyone at the table. There is no single method for handling all types of conflict. The dispute systems designer must determine which procedures are most appropriate and under what circumstances they should be used (Moore, et.al., 1992). By matching methods to dispute types, the designer increases the opportunity for successful negotiation of the ritual exchange. Successful interactions increase emotional energy which enhances future exchanges.

Skilled manipulation of group conflict management and resolution procedures upholds these techniques as sacred, generates emotional energy for the participants, and helps create a common definition of the situation. The moral solidarity thereby produced creates momentum for peace by instilling confidence in the parties' abilities to resolve their differences. As these chains of conflict interactions are carried over space, time and number of people, party-lines develop amongst those that share a common definition of the conflict. Creating inter-group solidarity mitigates the divisive nature of conflict escalation.

At times when parties cannot or should not use interes tbased procedures, the design should prescribe low-cost alternatives. For example, arbitration and popular justice (at least its most recent South African form) are generally less costly to relationships than is litigation. If power-based methods are appropriate, the system design could promote the use of the ANC's rich history of non-violent direct action campaigns. By getting agreement on acceptable means of displaying power, it becomes a sacred object to the parties involved. Violation implies lack of respect for the group.

How dispute systems design creates or recreates macro- structures in system of chronic, destructive conflict

If the DSD puts in place procedures, roles and institutions which group members can use to make decisions and manage or resolve conflict successfully, these procedures, roles and institutions become the nexuses of newly emerging or re-emerging social structures within the system network. The increased emotional energy from successful exchanges leads to greater confidence in future interactions, amplifying the legitimacy of the components of the dispute systems design (Collins, 1981). As the legitimacy of an avenue of redress increases, it is likely that people will increasingly seek that alternative for a given situation, making the structure more and more macro.

In a rapidly escalating conflict, intra-group solidarity is high with little room for dissent (Coser, 1956). Groups which instill a laager mentality in their members attempt to detach themselves from the surrounding social structures. The need to reintegrate polarized parties is one of the main functions provided by the procedures, roles and institutions developed by the dispute systems design.

One can look at how the DSD process creates or recreates social structures by extending Simmel's (1955) analysis of how the intersection of group affiliations defines the individual. As disputants meet at the structures created to help them resolve their differences, they bring with them their group's "web of affiliations." If the ensuing interaction provides a boost in emotional energy and fosters a link between the individuals involved, there is the potential for using these relational bonds to start eroding the walls of intra-group solidarity. If the individual is the incarnation of one's web of group affiliations, then the group can be thought of in a similar manner. The group takes form out of its linkages to certain groups and its detachment from others. The procedures, roles and institutions promoted by DSD help to recreate linkages amongst groups by providing points of articulation designed to empower the parties to converse about conflict issues by redefining conflict situations as beneficial to the parties involved. As the frequency of using these points of articulation increases, they become imbedded in the group's web of affiliations.

An example of this comes from the Western Cape Taxi War which will be discussed in greater detail in Chapter 5. Two taxi associations were engaged in a series of destructive interactions lasting a few years. Eventually a third-party was able to intervene successfully and create a forum which allowed these associations to start to converse with one another and other stakeholders in the conflict. Forum interactions amongst the parties present enabled them to negotiate successfully a series of group symbols, one being the creation of a unified taxi body. This new macrostructure resulted from micro-encounters in the forum. It is one example of how the DSD process can create new group affiliations which modify macrostructures within the system.

From the microritual level to the macrostructural level, dispute systems design influences the system of conflict. At the micro level, DSD prescribes roles, procedures, and mechanisms which enable people to make decisions or manage and resolve destructive forms of conflict. At the macro level, DSD alters societal structures by providing connection points which empower groups to resolve or manage their differences. If successful, these nexuses work to reintegrate groups by fostering crosscutting ties to groups and individuals seen to be the enemy.

This is done by focusing on less costly forms of decision making and conflict resolution which hold the relationship as sacred rather than who is right or who is more powerful. It is through this process that dispute systems design uses the interaction ritual to re-establish group solidarity and membership in systems that have been torn apart by chronic conflict.



Entering a system with the complexity and vastness of present-day South Africa is a daunting task. The seeds of my entrance were sown during a thirteen month period, starting mid1990, that I spent in the country working with Koinonia, Southern Africa. My activities with Koinonia enabled me to meet a wide variety of religious, political, and academic leaders. This time provided a fertile environment for understanding the richness of South African society and it laid the foundation for my gaining entry into various systems of conflict as a dispute systems design researcher. When I returned to South Africa in August 1993, I had numerous contacts which permitted me to meet with academicians, historians, conflict management professionals, NGO members, and politicians.

My research in South Africa was greatly enhanced by a dispute systems design training project funded by the United States Institute of Peace. It was developed jointly by CDR Associates in Boulder and the African Centre for the Constructive Resolution of Disputes (ACCORD) in Durban. This project included five training sessions designed for conflict management professionals, local and regional peace committee members of the National Peace Accord, political leaders, and academicians. I assisted these sessions as part of my international internship with CDR. My attendance and participation in four of these trainings permitted me to meet with and interview conflict monitors and mediators from Natal, Transvaal, Northern Transvaal, Eastern Transvaal, Western Transvaal, Eastern Cape and Western Cape.

I used formal and informal ethnographic interviews to collect data, operating in the context of observer-as-participant (Babbie,1992). The participants in each of the seminars knew that I was researching conflict and conflict management efforts in South Africa. I used informal interviews to determine the following broad categories:

Once I identified these general categories, I formally interviewed the people who I felt were the most involved in the monitoring or intervention of the disputes (see interview schedule in Appendix A). I collected data from these formal interviews by either taping them or taking field notes during the conversation. The overall goal of both the formal and informal interviewing was to define the different systems of conflict identified by the people either observing or participating in them.

In addition to interviewing participants in the dispute systems design seminars, I formally and informally interviewed a number of others engaged in resolving or studying conflict. Outside the seminars, I met with historians, sociologists, psychologists, political scientists, politicians, religious leaders, economic development specialists, conflict management professionals, members of the National Peace Accord structure, members of NGO's, teachers, and United Nations monitors. For this thesis, I formally interviewed 33 people. Interviewing multiple practitioners in each of the regions I visited, enabled me to cross-check stories, develop a more detailed analysis of the dispute system, and listen to different racial/ethnic perspectives on the meanings attached to various disputes.

The formal interviews had different goals depending on the role of the interviewee. If the person was a conflict resolution professional or a member of the Peace Accord structure, the discussion focussed on identifying the types of conflict and waging techniques employed by the parties and intervention agencies. Members of academic institutions were questioned about how they interpreted the conflict from their respective discipline. Learning more about the history, social or political analysis of a conflict is helpful in a number of stages of the dispute systems design process. Foreign monitors provided objectivity from the perspective of "the stranger" (Simmel, 1950), permitting me to check stories and interpretations of conflict events against information gleaned from people who potentially had a vested interest in the outcome. NGO members were interviewed to reframe regional issues into the national picture. These people were helpful in identifying national implications of various forms of conflict and their waging tactics and strategies. Interviews with politicians, economists, and religious leaders were used to generate options around ways to gain entry into systems of conflict and to discuss options for getting conflicting parties to start the process of dialog. Each of the types of people interviewed added another level of detail to the overall picture.

I chose this method of collecting data because it is similar to the method a dispute systems designer would use when diagnosing a conflict system. The designer would attempt to acquire detailed information by interviewing a number of people within the system to determine: types of disputes; who the formal and informal leaders are; ways in which the disputes are being handled successfully and unsuccessfully. In this stage of diagnosis, I used purposive sampling in an effort to gain as comprehensive a picture of the system as time allowed. While surveying a larger audience generally yields more data, I abandoned this approach after arriving in the country. Many of the organizations I planned to study were dealing with violent crises on a daily basis. I was concerned that answering a survey would be deemed a low priority task. For this reason, I chose face-to-face interviews because I felt the participants would find this method most efficient. In addition, interviews allowed flexibility in determining which types of conflict (e.g. taxi, hostel, political expression, academic, etc.) I would pursue with ensuing interviewees. This helped me narrow the focus of data collection by selecting the types of conflict that were of greatest interest to the respondents and would provide the most data for analysis.

Types of conflict identified for research

At the first dispute systems design seminar in Pretoria, I had the opportunity to speak with Peace Committee members from the Transvaal, Northern Transvaal, Eastern Transvaal and Western Transvaal. While each of these regions identified conflicts specific to their socio-political situation, there were a number of common themes in the types of conflict waged in the various regions.

Each region was dealing with conflict over political expression. While in the Transvaal, this most often occurred between Inkatha and ANC supporters, in other regions the conflict manifested itself between the ANC and the white right-wing organization, the Afrikaanse Weerstandsbeweging (AWB). Peace Committee members in each region also spoke about intervening in conflict in educational institutions. At the primary and secondary school levels, lack of resources and inadequate teaching were described as the heart of the conflict in black schools. At the university level, the issues changed to questions about educational standards for both entrance requirements and progress towards a degree. There were other common types of conflict that Peace Committees are dealing with, including police community relations, implementation of community development projects and conflicts between township communities and their municipal bodies. During the process of interviewing intervenors from these regions, two other types of conflict emerged as likely targets for this study.

Taxi conflict had been addressed in each of the regions, making it easy to develop a rich understanding of the common issues to these types of disputes. After the implementation of the fuel price increase, media accounts of the economic situation of taxi owners added more detail to the analysis of this conflict system. The most significant factor in choosing taxi conflict came from the amount of documentation on the Western Cape Taxi War and access to a number of people who intervened in this conflict. Many characteristics of this system would match an ideal type for taxi conflict. Interviews with people who were involved in this conflict provided historical background, identified conflict issues, detailed intervention attempts and suggested possible dispute resolution roles and procedures for the various types of recurring disputes. The richness of this conflict and the attempts at intervention made taxi conflict an ideal foundation for this research.

Hostel community conflict is the second conflict type chosen for this study. While this type of conflict was not discussed much by Peace Committee members outside the greater Johannesburg area, it was a main topic for Peace Committee members on the Reef. In addition to it being a significant issue for these people, there was a considerable amount of research done by the Goldstone Commission on this problem.

Hostel community conflict on the Reef was chosen because I had access to a conflict resolution practitioner who had knowledge of dispute systems design and he was intervening in these conflicts. This person was able to provide a detailed information on the conflict issues, how to gain access to the system and how to motivate participation amongst the parties. His practical knowledge of these conflict systems provided me with an analysis of why certain dispute resolution procedures would work and others would not. Combining this information with data collected from interviews with Local Peace Committee members who were intervening in hostel community conflict created a framework for how to access and initiate peacekeeping, peacemaking and peacebuilding efforts in this type of conflict system.

Limitations of the research model

This data collection model has a number of limitations. Adler and Adler (1987:26-27) describe three problems inherent in participant observation field research: interpretation as an outsider, reflexivity, and the search for first-order interpretations. Each of these has implications for this study. As an outsider, especially an Anglo, North American, there were both problems with trust and problems with multiple interpretations of gestures, talk, and meanings of objects. For example when I asked for permission to tape an interview with a researcher of popular justice systems, he assented to the request while acknowledging that it reminded him of prior state interrogations. Early in this interview, it became evident that recording would be impossible. Whenever the conversation moved into specifics about community justice projects he was involved in starting or studying, he requested that the recorder be turned off. His prior experience with white people recording discussions about his work created a context within which the gestures, talk, and objects fostered a sense of mistrust. In a country with the tremendous racial and ethnic diversity of South Africa, opportunities for misinterpreting social communication abound.

The second problem of reflexivity occurs because social facts are constructed by interpretations of what occurred in a given situation. Without being a full group member, it is impossible to constitute the social world in the same manner as members of the group under observations (Adler and Adler, 1987:27). This problem is complicated in my research by the limited amount of time I spent observing and interviewing people in the systems of conflict. In addition to people I interviewed, it would have been helpful to speak to members of the disputing parties. Time, levels of violence, and contacts usually prevented this from occurring.

The third problem of dealing with the development of firstorder interpretations of events is related to the first two. Schutz (1962:6) writes:

The constructs used by social scientists are, so to speak, constructs of the second degree, namely constructs of the constructs made by the actors on the social scene, whose behavior the scientist observes and tries to explain in accordance with the procedural rules of his science.

My interpretations of events come, not only second-hand, but sometimes months or years after the event has occurred. This lapse of time allows the event to be constructed and reconstructed based upon unfolding knowledge about the system under study. For example, in taxi and hostel conflicts, the Goldstone Commission recently lent credibility to long-standing allegations of police misconduct. This impacted the reconstruction of prior conflicts that may have had more to do with political affiliation than economics. By interviewing people who intervened or studied the conflict from an academic perspective, I was constructing a social system from their construction of the system. This put me at least one step away from the disputing parties' interpretation of the events.

This dispute systems design study is limited by still other factors. The amount of time spent observing the system, the number of people, and the breadth of their roles in the system would all necessarily be increased to develop a true dispute systems design. However, since this research was not tied to a particular dispute system I focussed on collecting data from several situations throughout South Africa to compare disputetypes across regions. By raising the level of abstraction on a specific dispute-type, there is a greater chance that peace committees or conflict management professionals will find applicability in their system of conflict. For instance, in conflicts involving taxi associations, political affiliation may play a significant role in the Transvaal and Natal; however, in the Western or Eastern Cape designers could probably exclude this issue. In fact, as the ensuing dispute systems design for the Western Cape Taxi War shows, the lack of political conflict between taxi organizations may provide an avenue to pursue unifying groups so that their collective power is used to solve issues to the dispute. By typifying conflict issues and intervention techniques, designers have a tool to compare and contrast the system they are analyzing; however, this approach potentially excludes aspects of a specific conflict that played a more important role than was estimated.



In this section, I will apply the dispute systems design process to specific conflicts. To facilitate this, I will present an example of a conflict occurring between communities and migrant labor hostels and a conflict involving the black taxi industry. The case study of hostel community conflict takes place on the Reef outside Johannesburg. The taxi conflict case study is from the Cape Town metropolitan area. After discussing these situations, I will explain how each step of the dispute systems design process relates to these conflicts.

The Western Cape Taxi War

The roots of the Western Cape taxi conflict are in the influx control laws governing where people could legally reside. The first black taxi organization to legally operate in Langa, Gugletu and Nyanga outside Cape Town was called LAGUNYA. Over time, "pirate" taxis operating without permits developed a market by providing service between the townships and downtown Cape Town. These pirates were harassed and prosecuted by the police but were generally tolerated by LAGUNYA. With the abolition of influx control in 1986, the Local Road Transportation Board (LRTB) deregulated the permit process.

Deregulation allowed pirate operators, who formed the Western Cape Black Taxi Association (WEBTA) in 1987, to acquire "route" permits and legally service the township-to-city center demand. The legalization of WEBTA increased tension because it was seen as a threat to LAGUNYA for a number of reasons. First, the increase in number of permits allocated increased competition in a tight market. Second, the LRTB was given instructions to favor WEBTA applications in order to speed up the process of legalizing taxis. Third, at times the radius and route permits overlapped, creating confusion about who could service certain areas. Fourth, upon legalization, WEBTA operators started acquiring the more lucrative route permits. LAGUYNA operators, who were stuck with radius permits handling travel within the townships, saw this as unfair, especially in light of the fact that they had historically been the legal taxi association (Collin, 1992; Goldstone, 1992, 1993).

Toward the end of 1990, violence between the two taxi associations erupted over access to a taxi rank in central Cape Town. Tensions continued to ignite paradoxically over access to taxi ranks, definition of routes and allocation of permits for the next two years. By 1993, over 70 lives were lost, damage to vehicles topped R4 million (over $1.2 million) and property damage ran more than R25 million (approximately $7.4 million). In addition to this destruction, 90 to 100 taxis were repossessed monthly at the height of a consumer boycott of both taxi associations in 1991 (Goldstone, 1993).

This conflict saw a number of organizations involved as intermediaries. Political parties, including ANC, Pan-Africanist Congress and Azanian People's Organisation, along with the Congress of South African Trade Unions, Black Sash and the Cape Town Peace Committee all assisted in the various attempts to intervene between WEBTA and LAGUNYA (Goldstone, 1993). Matters were relatively calm until February/March 1991 when serious violence arose in the township of Khayelitsha. The community imposed a boycott of both associations in response to the violence. A newly formed organization called the Taxi Crisis Coordinating Committee (TCCC) managed to get LAGUYNA and WEBTA to sign a 10 point plan in April 1991 which detailed a code of conduct, agreement to unify both associations, and a common rate structure, but this plan was short lived as WEBTA decided to void the agreement.

The conflict escalated in July 1991 after a WEBTA member was killed and WEBTA retaliated. By mid-July, mediators had gotten both parties to agree to a plan for allocation of taxi ranks, routes and loan rescheduling for taxi payments. This plan was to be administered by the city council instead of the Department of Transport which was seen to be partial. However, when WEBTA attempted to return to Khayelitsha the agreement quickly fell apart. WEBTA vehicles were attacked and it responded in kind.

By September 1991, intimidation by both parties escalated out of control. The community was drawn in on the side of LAGUNYA. WEBTA, which was still unable to operate in the township, started attacking the people of Khayelitsha. The township was turned into a virtual war zone with arson, murder and intimidation taking place on a large-scale (Goldstone, 1993). In October a peace conference which included government, business, religious, academic, service and community leaders was called by Archbishop Desmond Tutu and a number of other high-profile individuals. While this meeting only fostered a one-day respite from the violence, it did create the Cape Town Peace Committee (CTPC) from the leaders present.

Using the National Peace Accord mandate, the CTPC continued to try to obtain a negotiated settlement. When this failed, the South African Defense Force was brought in to restore order. The CTPC continued to pursue other options for resolving this conflict. They asked the city council to re-register all of the taxis in an attempt to exert some control over this intractable issue. By year end 1991, the violence had subsided due to the presence of the South African Defence Force, the formation of the South African Police's Taxi Unit and the return of many taxi operators to Transkei for the holiday season (Goldstone, 1993).

At the beginning of 1992, the South African Black Taxi Association was invited to attempt to mediate the conflict. A complicated mediation process culminated in a written agreement dealing a new division of routes between the associations on January 31, 1992; however, on the next day violence flared again (Collin, 1993). The police closed a taxi rank at the center of the new violence and the ANC made another call for boycotting the taxis.

By March 1992, the Peace Committee was able to reconvene the parties once again, establishing the Cape Organisation of Democratic Taxi Associations (CODETA). CODETA's mission was to create a constitution and code of practice for taxi operators. While this process is still going on, the taxi associations did agree to create a unified body designed to serve the interests of both associations. Violence has erupted from time to time but the Goldstone Commission (1993) has found that none of it can be attributed to the conflict between former LAGUNYA and WEBTA operators. Before addressing a specific dispute systems design for this conflict, I will present background on hostel conflict on the East Rand.

The Reef Township War

After the February 1990 unbanning of political organizations, the ANC, Pan-Africanist Congress, and AZAPO started to organize their constituencies. Throughout the 1970's and 1980's, one main black organization, Inkatha, had been able to legally organize its members as a Zulu cultural association. Now Inkatha also cast itself as a national political party. Tension on the Reef grew as these emerging political parties struggled for influence amongst the country's black majority.

The Inkatha Freedom Party (IFP) and the ANC-aligned United Democratic Front had been conducting a low-intensity civil war in Natal and kwaZulu throughout the 1980's. Now this regional tension extended its grip to the Johannesburg area. While the actual start of the "Reef Township War" is difficult to pinpoint, on July 22, 1990 Inkatha held a political rally near Johannesburg. Its supporters were harassed as they traveled to the rally. Upon their return, they attacked and killed 27 people with traditional weapons. Hostel residents who did not attend the rally refused to let the rally-goers re-enter the hostel. Tensions between hostel dwellers and the community escalated over the next few weeks with attacks and counter-attacks spreading throughout a number of townships on the East Rand. The general perception was that this was an ethnic/political conflict. The hostels were dominated with Zulus assumed to be Inkatha supporters while the surrounding squatter settlements were dominated by Xhosas assumed to be ANC-aligned. Violence escalated with ANC-aligned township youths fighting IFP-aligned hostel residents (Minnaar, 1993).

Simmel's (1955) analysis of the intensity and violence of conflict suggests that this conflict would see a high degree of polarization and suppression of dissent within the respective parties. That is indeed what occurred. Over time, hostel residents drove non-IFP supporters out of the hostels and forced some residents to participate in attacks. In the communities, IFP-supporting squatters were forced from their homes and into "refugee-hostels" which housed entire families. There were reports of ANC members shutting down taxi operations in the townships to ensure that community residents would be home to repel attacks from the hostels. The intensity of the division between hostel and community was so great that hostels started to become mini-communities, housing shops, churches, and schools. At times, hostel residents were barricaded into their compounds under siege-like conditions (Minnaar, 1993).

The war escalated in the latter part of 1990 with the proliferation of firearms. Township residents started to organize in Self-Defense Units while hostels grew more militarized. Accusations of biased behavior by the police or their lack of interest in preventing violence came from both sides. The ANC alleged that there was a "third-force" involved in the violence, implying that white right-wingers in the government were involved in fomenting the violence for political gain.[2] In highly unstable areas, razor-wire was used to encircle problematic hostels, adding to the community's negative stereotype of hostel residents.

Throughout 1991 and 1992 the violence continued. Even in Alexandra where hostels and the community had a record of cooperation and peaceful co-existence, violence flared. Both sides accused the other of forcing people out of their accommodations and moving their supporters into the freed space. The South African Police came under attack for not bringing to book the people involved in the violence. The conflict came to a head on June 17, 1992 at Boipatong. Thirty-nine squatters were massacred by residents from a nearby hostel. The ANC would at this point break off high-level talks with the government about political transition charging the absence of police response permitted events like the Boipatong Massacre to occur. The ANC felt that such events undermined their credibility and showed the government's lack of good-faith at the negotiating table.

In the midst of the accusations and retaliations of mid-1992 arose a peace initiative that mitigated some of the tension between hostels and communities. In August, a Zulu from the Jeppe hostel by the name of Jacob Dlomo drafted a peace letter, and at tremendous personal risk, delivered it to the rival Selby hostel in Johannesburg. By year-end, Reverend Mvume Dandala was facilitating a series of meetings between these groups. At the beginning of 1993, IMSSA became involved in this "Hostel Project," which now involved more than 30 hostels and their communities. This project brought together bitter rivals by working with the hostels, civic groups, youth organizations, and churches. Rallies, sporting events, and prayer meetings were held to attempt to reintegrate these enemies. The participants worked hard to create an environment of tolerant, participatory democracy. Each local structure was formed by the parties involved in the conflict, allowing them to control the processes by which peacekeeping, making and building occur.

The Hostel Project was cause for great hope on the Reef; however, the de-escalation of violence ended dramatically in April, 1993 when Chris Hani, leader of the ANC Youth League, was assassinated by a white right-winger. Violence throughout the country soared. On the Reef, deeply entrenched feelings of mistrust crushed arising peace efforts. Sporadic acts of violence and retaliation became a weekly occurrence. Previous allegations from both sides re-emerged. Rumors of violent acts created episodes of retaliation for events that may or may not have happened. At this writing, criminal gangs continue to profit from displacing people and reselling their homes and personal property. Some hostels remain under siege by ANCaligned groups. In certain areas of some townships barricades were erected, stopping any transportation through them. Self Defence Unit's patrol the townships with communities taking responsibility for justice and safety since the police and defense force seem unable or unwilling to intervene (Minnaar, 1993b). Nonetheless, the Hostel Project continues with participating bodies taking very small steps in an effort to rebuild the trust that was being fostered in early 1993.

In systems of conflict like this one, or in the taxi industry, many approaches to end the violence have been tried. A few have had success but in most cases the fundamental issues to the conflict were left unresolved. It is tempting to conclude that in these violent, intractable situations power-based methods of restoring order are the only viable options; however, an effective dispute systems designer would argue the shortsightedness of this idea. What is needed is a comprehensive plan for peace that addresses practical needs at the grassroots and symbolic needs at party leadership levels. Dispute systems design focuses on interest-based decision making, peacekeeping, peacemaking, and peacebuilding, but also incorporates rightsbased and power-based methods of engaging conflict.

Dispute Systems Design: Preparation and Implementation

Dispute systems design looks to develop systematic solutions to chronic conflict by creating mechanisms which enable the parties to regulate or resolve recurring disputes and find ways to engage in mutually beneficial forms of decision making. The designer should analyze the following aspects to determine the readiness of a system of conflict for this type of intervention:

Moore (Ibid.) describes seven stages of dispute systems design:

1. Gaining entry into the system.

2. Forming a change team and intervention plan.

3. Conflict system diagnosis.

4. Design process.

5. Developing support for the new system.

6. Implementation of design.

7. Operating the new system.

A systems approach fosters a comprehensive method to decision making and conflict management which is tailorable to each region or community and it permits the designer to create interfaces to national, regional and grassroots peacekeeping, peacemaking and peacebuilding initiatives.

Gaining entry into the system

During the first stage of dispute systems design, the intervenor must perform the following:

In this phase, the building of trust is initiated on two levels. First the requesting parties must start to trust the ability of the intervenor. Secondly, the designer must encourage client trust in the efficacy of the process of dispute systems design.

In the systems of conflict under study, there is usually a high level of mistrust between disputants due to the use of power-based conflict waging techniques. Attempts to convene parties publicly have often been unsuccessful as evidenced by the Western Cape Taxi War. In polarized conflict, political posturing can impede progress in negotiations because the leadership of each party is beholden to the sacred objects of their group, not the emerging sacred objects of the negotiations. Coaching parties to make a good-faith gesture to the other may be interpreted as a profane act by group members who are not "at the table."

In the taxi war, the action which broke the impasse was a request by one of the parties for a confidential meeting with two mediators involved in the public proceedings. These mediators conducted a series of clandestine interviews with the focal participants to develop a picture of the conflict and condition parties to commit to the process. By allowing the parties to covertly exchange good-faith gestures, these successful interaction rituals started creating symbols of common group membership. The secret shuttle negotiations fostered the sacredness of trust while eliminating the need for parties to perform public damage-control which may have torn their fragile relationship.

In most hostel conflicts, there is a history of intervention which provides structures that the designer can use to gain entry. Much of the intervention preceding dispute systems design work is either police/military presence or monitoring of the violence. Using these structures, as well as any local peace committee or Hostel Project committee, allows the designer access to people who are knowledgeable about the conflict and can help them define the scope of the intervention. These people are also in position to enlist commitment from the parties to the dispute systems design process.

Forming a change team and intervention plan

The second stage of dispute systems design deals with the creation of a change team and establishing the goals, tasks, and schedule of the intervention process (Moore, et.al., 1992). In South Africa, conflict management experts cite this as one of the most difficult phases because of the dynamic and volatile nature of conflict. During this stage of entering the system of conflict, the dispute systems designer should note who the primary parties, formal and informal leaders, and stakeholders are.

During the creation of the change team, it is important to include people from different organizations and levels within organizations to develop legitimacy throughout the breadth and depth of the system of conflict. Because the democratic process in South Africa has been subverted in many ways, formal leaders do not always have legitimate authority. Therefore, it is important to insure that all identified parties agree that they are adequately represented. The task of creating legitimacy is started with the selection of change team members. In order to continue this task, it is necessary to define change team roles, communication procedures and guidelines, and determine the decision making process.

Within the micro-interactions of the change team come the sacred symbols of this emerging group (Collins, 1981). For this reason, the change team must model the values, roles, and procedures it is designing for the conflict system. This process embeds these arising group symbols into the implementation of the design, showing disputing parties how to identify common-ground and perform rituals which uphold the sacredness of this commonground. For example, if participatory democracy is cited as a common goal, the change team needs to model this concept for two reasons. It gives them experience using the skills (e.g. group facilitation) necessary to create a democratic body. This, in turn, helps them disseminate these skills and their importance to members of the conflict system.

Once these procedural issues are agreed upon, the change team must focus on how to uncover what needs to be known about the conflict and who can provide that information. The intractability of taxi conflicts makes this an especially difficult task. Convening the appropriate stakeholders, formal and informal leaders at the start of the process is complicated by the fact that at the root of taxi conflicts, there are interests which involve many groups including taxi operators, the police, communities, local politicians and government administrators. Therefore the creation of any forum for discussing potential problems and solutions must deal with the issue of how to identify and include people or groups who have a large enough stake in the conflict that they can promote the successful implementation of any plan for regulation or resolution. Provisions for including arising leaders and stakeholders throughout the intervention and subsequent implementation of any conflict regulating mechanisms are a necessary part of this phase. While change team continuity is vital to the process, the designer must prepare for the addition or removal of participants.

To develop legitimacy in the process, the change team should include all segments of the communities and organizations involved in the conflict. For example, taxi association leaders may feel that they speak for both owners and drivers; however, there may be reasons for enlisting representatives from each subgroup within the taxi association so that issues important to drivers are not bartered away in favor of issues more important to owners. Likewise, communities may see the conflict from differing perspectives. To assume that town council members, political, religious or business leaders can represent all those interests may be problematic. The inclusion of informal community leaders that are directly affected by violence at a specific taxi rank or are personally familiar with problems encountered during the daily commute may provide the process with critical information which fosters support of the dispute systems design process at the grassroots. Since the change team is a source of legitimacy for the process, its development of roles, guidelines, and decision making procedures should reflect the intent of participatory democracy.

In hostel community conflicts, building support throughout the system for any process is paramount. While the conflict is usually reduced to tension between the ANC and Inkatha, there are other groups involved and there are subpopulations within those groups that will have varying interests. Within the ANC, representation by its local political leaders, the Youth League, and leaders running the Self-Defence Units would be a necessary start. For Inkatha, its local political leaders and the hostel indunas (these are people in traditional leadership positions) should be included. Other political parties like the PAC and AZAPO would have an interest in how the violence is resolved. In addition to these parties, a number of other groups have vested interests. Representation from the hostel owners and business community is necessary, especially from the transportation industry where taxi drivers from politically aligned and nonaligned taxi associations should have a voice. Incorporating representatives from the mining companies would ensure that their interest in holding onto a continuing, experienced labor force is factored into any agreement. Leaders from the religious community continue to play a significant role and must remain in the process. Some Local Peace Committees members have a credible reputation which would increase the legitimacy of an intervention.

Conflict system diagnosis

The diagnosis stage starts with determining who has information about the conflict and who is best suited to collect that information. The groups identified in the first two stages are included in the diagnosis process. Identifying who to talk to is a problem similar to determining who is part of the change team. In this case, the designer should find out who the key decision makers are in institutions within the system and who are the informal leaders. This phase of the process should identify:

The following tables show the typical issues involved in taxi and hostel conflict. They also outline the types of waging techniques people are using. The designer can use these tables to compare and contrast typical taxi or hostel conflicts with the system under study. This would shorten the learning-curve of the change team in determining what types of approaches work and which do not. Table 1 outlines the typical disputes occurring in a system experiencing taxi conflict. The information for this table was taken from the Goldstone Commission's inquiry into taxi violence, events from the Western Cape Taxi War, the August 1993 Johannesburg Taxi Indaba and numerous discussions with local and regional peace committee members who were involved in resolving various types of taxi conflict in regions throughout the country.

By categorizing the different types of disputes, such as unfairness of permit distribution, intimidation at taxi ranks, or allegations of police interference, the designer can start to look for appropriate ways of handling these issues. This information may point to reasons why truces or codes of conduct are not working. For example, if political affiliation does play a role in the conflict but local political leaders are not involved in the decision-making process, then accountability along party lines is easily compromised. Including these leaders in the process will increase the legitimacy of the truce or conduct code.

Table 1. Taxi Industry Conflict.

Table 2 addresses hostel community conflict. It delineates the typical issues and ways that people are waging the conflict.

Table 2. Hostel Community Conflict

Using these tables, the change team can start identifying the needs of the system. For example, perhaps a conflict occurs where either hostel residents or the community are cut off from work, school, or food and a taxi war breaks out in another part of the township. These events may seem unrelated; however, using systems analysis, one may find that one taxi association switched its routes to work around the "no-go" area. This may have put them in direct competition with another association in a tight transportation market, precipitating the conflict. Or a longrunning dispute over allocation of funding for upgrading municipal services in the township and hostels may clarify why peacebuilding efforts that do not require proper financial accountability are failing.

By seeking to understand how elements of the conflict system impact one another, the change team is able to develop a comprehensive picture of the system of conflict. The success of the dispute systems design hinges on how well the change team can capture how the system's elements impact one another.

Design process

This phase is a collaborative effort involving the groups identified in prior stages of analysis. During the design process, each dispute type is studied to determine which conflict management procedures are best suited to handle it and create the desired outcome (Moore, et.al., 1992). The options that system designers must consider range from unassisted procedures where the parties collaborate to resolve their differences, to assisted procedures designed for advocacy, peacekeeping, peacemaking and peacebuilding objectives. The designer must be familiar with the objectives of each method of conflict resolution or regulation in order to select the most appropriate option or set of options.

Unassisted procedures, such as collaborative problem solving, where parties work together to mutually beneficial solutions, and negotiation, are best used when hostilities are low, power relatively equal and parties have a high degree of trust or easily cooperate with one another. A second form of conflict management is needed in systems where there are institutionalized procedures to handle requests, grievances, etc. and these procedures are either complicated or disempowering to the client (Moore, 1993). In this situation, advocacy groups, such as The Black Sash or Lawyers for Human Rights, perform a necessary role in the system of conflict.

Other assisted conflict management techniques run along the continuum of peacekeeping, peacemaking and peacebuilding activities (see table 3). The goals of peacekeeping procedures are to separate the parties and end hostilities. Monitors, observers, rally marshals and rumor control mechanisms would fall in the category of peacekeeping. Peacemaking exercises are designed to allow parties to manage or reconcile their differences through collaboration, negotiation or mutually agreed upon decision making procedures or bodies. Peacemaking roles would include message carrier, advisor, facilitator, mediator, judge or arbitrator. Peacebuilding focuses on strategic rebuilding of relationships, reintegration of social institutions and addressing economic problems. Peacebuilding procedures include economic development projects, forums for restoring relationships amongst community groups or education programs geared to address issues like democratic decision making or political tolerance (Moore, 1993).

Taxi conflict design

Due to the complexity of issues in taxi conflict, a single dispute type may have multiple conflict management options. This creates a need to evaluate disputes as they arise to ensure that the most appropriate management procedure is used for that dispute. In addition to this evaluation process, the designer must also find ways to integrate dispute procedures with other traditional and legal-rational conflict mechanisms operating within the system or having an interface to the system of conflict.

Since it is likely that the level of violence is high and that there is little trust left between conflicting parties, the intervenor may have to first focus attention on peacekeeping. Fostering mutual agreement on a code of conduct and procedures for monitoring the code builds trust and creates momentum which the change team can use to implement the entire plan. Without sustaining momentum, the code of conduct will likely be crushed under the pressure of perceived economic injustice. If the code is successful and monitoring activities are working, the designer should incorporate these mechanisms into the design structure to address the ongoing desire to prevent future violence to taxi operators, their vehicles, their customers and private or public property.

Even if the peacekeeping effort is highly successful, the likelihood of future violations of codes of conduct is great. In order to prepare for this, a peacemaking mechanism is needed. One possibility is to use the popular justice concept to handle taxi disputes. South African taxi courts, similar to the community courts operating in places like Alexandra or the rural Eastern Cape, could mediate disputes over violations of codes of conduct or complaints brought forth by customers. Codification of criteria for acceptable taxi court cases could be combined with procedures for how to hand off cases to the legal-rational justice system that are deemed inappropriate (e.g. those in which acts of violence occurred). The creation of a taxi court could act as a mechanism for resolving conflict which will reinforce traditional concepts of collective justice (Nina, 1993). This, in turn, reintegrates the sacred nature of community interrelatedness. That is, the agreement reached in the Taxi Court becomes a sacred object, representing the wisdom of the collective conscience. A person who violates the court's decision would be seen as attacking the community, forcing it to defend itself (Collins, 1981).

There are a number of other possibilities for managing daily disputes which could disrupt the peace process. At the Taxi Indaba, the idea of putting unbiased queue marshals at taxi ranks was presented. These marshals would insure that parking spaces and passengers were allocated equitably as taxis entered the ranks. Another possibility would be to use rumor control mechanisms like Peaceline in the greater Johannesburg area to quell unsubstantiated rumors or provide timely response to genuine problems by alerting the appropriate authorities. A third peacekeeping idea was developed in the Western Cape Taxi War. Teams of trained conflict managers were put on call. They acted in both peacemaking and peacekeeping capacities by responding to flashpoints to either monitor or intervene in the conflict. These are the types of activities that Local Peace Committees are mandated to perform. Providing trained mediators to intervene in crisis situations increases the likelihood that disputants can successfully negotiate an interest-based conflict interaction. If they are successful, the boost in emotional energy will encourage them to attempt similar procedures in the future (Collins, 1981). Otherwise, they may conclude that powerbased methods are easier and more effective without understanding the long term costs of using these alternatives.

Many of the problems taxi operators experience are rooted in economic issues. Engendering real peace means taking on this difficult problem. In some disputes, intervenors have worked with banks to rescheduled loans. While this is a very innovative solution, it still does not address the fundamental issue of economic viability in the taxi service industry. Without tackling this, taxi operators will continue to overload their minibuses, engage in unsafe driving practices, fight over more profitable routes and attempt to enter markets that cannot support the current supply of taxis. Both the August 1993 Taxi Indaba in Johannesburg and the 1993 Goldstone report on taxi violence underscored the need to train taxi operators in basic business management. Goldstone found that few taxi operators had the business knowledge necessary to analyze potential markets. There is little understanding of how healthy competition can sustain a market and keep it profitable. When competition is encountered, the response is usually to eliminate it. Business education of taxi operators would help promote realistic expectations of business opportunity in this tight market.

In addition to business training forums, the Taxi Indaba participants put forth an idea for a Taxi Bank. The Taxi Bank would offer reduced interest rates on loans, address problems of discrimination on loan qualification and work with the state to provide subsidies for taxi operation (Minnaar and Torres, 1993). Taxi Banks and business education forums are two possible peacebuilding mechanisms which would address a number of the identified economic issues.

Allegations of police harassment and Local Road Transportation Board (LRTB) corruption create an aura of unfair access to the taxi market. It has also delegitimized the authority of the police and LRTB (Goldstone, 1993). As with disputes in other spheres of conflict, relationships with the South African Police are a source of deeply entrenched conflict. While communities are starting to address the issue of rebuilding police community relations, this will take time. In taxi conflicts, there are a few things that could be done to encourage this relationship. If the LRTB would make permit allocation records public, a fact-finding capability would emerge. This would allow independent bodies to analyze allocation records to determine if certain taxi associations were given greater access to markets. In addition, there are allegations that police members own taxis and that their harassment campaign is designed to hold up competing taxis during peak hours of operation. This allegation of a conflict of interest would either be confirmed or proved unfounded if the permit process were made public.

Since the credibility of the LRTB is often an issue to the conflict, there are ways to address this problem. During the Taxi Indaba, a number of ideas were generated about including local input on route planning, taxi rank facility upgrades and general safety issues. Convincing the LRTB that it is in their interests to open the planning process to all parties affected, may be a worthy investment. Public forums would provide a mechanism for gaining input from the many sectors of the community which are affected by this industry. If forums are successful in negotiating conflict, they become a ritual venue where the community draws together to reify its sacred symbols and recharge its emotional energy (Durkheim, 1965; Collins, 1981). They also provide a place where disputants must publicly prove their commitment to or rejection of these sacred objects of conflict engagement. Finally, it shows that the microinteractions of decision-making and conflict handling are growing over space, number of people, and perhaps even time, modifying the macrostructures of the conflict system (Collins, 1981).

Finally, as the Western Cape Taxi War showed, the creation of a single taxi association or governing body should be examined. Settlements to this conflict fell apart repeatedly until CODETA was created. The common interests of taxi associations are readily apparent. If the taxi conflict is politicized preventing unification, the change team may be able to link this conflict to broader issues in the system of conflict, such as political tolerance. By doing this, political channels become avenues for ensuring commitment by taxi organizations to the peace process. Whether or not it is possible to unite the associations under one body, creating an ongoing forum for discussion between associations may provide a means for promoting trust. By making the meetings periodic, parties can address issues on a timely basis rather than letting conflict fester until an incident precipitates confrontation. In addition, the forum could be used to prepare taxi operators to respond to issues that affect the industry, such as changes in legislation. The creation of a joint taxi association forum can be used to develop a track record of concerted effort between competing taxi organizations.

At the Taxi Indaba, the creation of a national taxi association was recommended. While the logistics of creating a single body seem daunting, the idea has a great deal of merit. For taxi operators, the increased lobbying power could be used to acquire state subsidies or influence future legislation which may affect the taxi industry. The potential for reigning in violence associated with this industry would benefit both the taxi operators and the government. Additionally by institutionalizing the format and rules of conflict engagement, the government would benefit from negotiating with a unified body, as opposed to a number of parties with disparate agendas and tactics (Coser, 1956). This would potentially enhance resolution efforts between the taxi operators and the state.

Designing mechanisms for controlling or resolving conflict requires that the systems designer know the types of disputes occurring in the system and what types of outcomes people in the system of conflict will find desirable. Once these two things are clarified, the designer must determine how conflict management procedures can be woven together to produce the desired outcomes from types of conflict occurring in the system.

Hostel conflict design

Since mid-1990, hostel conflict on the East Rand has grown less tractable. The conflict is permeated with deep-seated mistrust and negative stereotyping of opposing parties. The resultant violence has polarized parties, leading them to force conformity to party goals and behaviors within their own ranks. There is a long history of violent retribution for attacks.

While violence may not occur on a daily basis, intimidation does. Some people call for power-based means of peacekeeping; others see a need to rebuild relationships severed or never fostered in less destructive times. The start of a number of peacemaking efforts at the grassroots shows that there is a contingent of people who are frustrated enough with the present situation to take the risk needed to start dialog between the parties.

Suggestions for designing a dispute handling system for hostel conflict need to start with an approach to promote and keep peace. In areas of townships where barricades inhibit movement in the streets, there is a need to intercede to end the violence. The question of using military force is complicated by a number of factors. First, the use of military force does not address one of the primary issues of relationship building amongst parties. Second, the police and defense forces are usually seen as biased participants by either or both sides.

This may escalate tension in the area, creating a situation where the intervention force ends up a party to the conflict. Third, hierarchical military decision making styles and the need to maintain military efficiency may marginalize civic, religious, and humanitarian agencies who are better equipped to build democratic structures and long-term peace (Lederach, 1993a). While military intervention may provide tactical relief, it is unclear that a coerced reduction in violence will create the conciliatory seedbed needed to foster peacemaking and peacebuilding efforts.

An interest-based alternative to military intervention would focus on the relationship issues which are fundamental to this conflict. The approach of building a "peace constituency" attempts to empower people within the conflict to use culturally relevant conflict handling techniques to encourage a long-term commitment to peace (Lederach, 1992a). Within hostel conflict, there is a significant number of people who are weary of the violence. Developing a peace constituency involves creating space in the conflict for people, from leadership to grassroots, who are no longer interested in waging the struggle violently. It would involve the establishment of a social movement designed to reclaim popular power from warmongers and criminal elements who benefit financially or politically from the status quo. Polarization of the conflict forces people to take sides, regardless of their political affiliation. The theory that most participants cannot identify their party's political goals and are more tied to economic issues would suggest that by reducing hostilities, the change team could organize party alliances around other issues such as economic interests (Shaw, 1993). The goal of the initial creation of this peace constituency would be to sustain codes of conduct, rumor control mechanisms, and forums for continued dialog amongst the parties. The development of a peace constituency becomes the infrastructure for longer-term peacemaking and peacebuilding efforts.

The Independent Mediation Service of South Africa's (IMSSA) Peaceline has demonstrated the importance of a rumor control mechanism on the Reef. On the day of Chris Hani's funeral in April, 1993, Peaceline fielded a number of phone calls about rumored attacks carried out by both sides. During this tense time, the rumor control center was able to prevent escalation of conflict by providing accurate, timely information to neighborhoods concerned about their well being. While Peaceline has had notable success, this center attempts to monitor the entire Transvaal Province. There remains a local need for rumor control centers, especially since there is limited access to telephones and broadcast media for township residents. One suggestion is to have Peaceline use Local Peace Committees to set up and operate centers in or near areas of violence. This would allow quicker response to the hostels and communities. By encouraging participation from people in affected neighborhoods, local rumor control centers could recruit more people to the peace constituency and start the process of empowering indigenous populations to resolve their own conflicts while promoting relationship-building across the ethnic/political divide. For each center, establishment of successful ritual exchanges amongst volunteers from different parties would help them evolve conversations that grow into personal and long term interactions (Collins, 1975). These cross-cutting social bonds mollify the escalatory nature of current conflict.

The high degree of criminal activity combined with the ineffectiveness of the police create a need for a mechanism to reestablish security in neighborhoods and hostels. The ANC's Self-defense units (SDU's) have had success in this role. SDU's patrol the community with teams generally armed with traditional weapons (e.g. spears, pangas, etc.) but follow a code of ethics. While ANC-aligned, the SDU's do not operate under the auspices of the ANC, although many commanders are former cadres of the ANC's military wing Umkhonto weSizwe. In addition to patrolling the neighborhoods, they also provide a grievance mechanism by checking in with residents to see what problems have occurred recently. These informal checks have uncovered accusations of police misconduct. The SDU's seem to have legitimacy and status in the communities (Weekly Mail, December 17, 1993). The 1993 Goldstone Report on hostel conflict recommends that SDU's would have a place in a peaceful transition if they were depoliticized and turned into night-watch organizations (Minnaar, 1993).

Once the above peacekeeping measures start to reduce hostilities, peacemaking and peacebuilding efforts become viable. Local Peace Committees (LPC's) are already functioning in most townships on the Reef, but much of their activity is geared to monitoring the violence or acting in some role as a third party observer. Since LPC members represent all parties involved in the conflict and many members are trained intervenors, they are ideal resources for mediation, facilitation, and training. Perhaps the most important role that LPC and Regional Peace Committee members can play is in training people in conflict management techniques. Capacity building, or indigenous empowerment through skills needed by organizations or communities, is often cited as a primary goal. As LPC members gain experience in their intervention activities, they should continually look to transfer this knowledge to the people on the ground. This could be done by developing "partnering programs" or recruiting and training neighborhood leaders to run popular justice centers, rumor control centers, or mediation/facilitation programs. Providing this service at the grassroots empowers neighborhoods to accept responsibility for their own peace. It reasserts the importance of successful day-to-day microinteractions within the neighborhood rather than relying solely on interactions of high-level political leaders to establish peace.

There are other types of third party peacemaking procedures operating with a large degree of success. IMSSA's effectiveness in mediation and arbitration of a number of industrial labor disputes suggests that their services may have a role to play in conflict over the relationship between migrant workers and mining companies. IMSSA also has training resources and experience in initiating public forums which could be used in hostel community conflict. Another peacemaking possibility deals with using popular justice systems that have arisen throughout the country. While much maligned in the 1980's for their violent, retributive forms of meting out justice, new People's Courts are operating well in Alexandra by merging western-styled mediation techniques with traditional values around communication and authority. The community is encouraged to participate in order to construct a "collective truth" (Nina, 1993). Current popular justice systems, which are neighborhood oriented and focus on developing solutions to problems, are moving away from their infamous role as punisher.

There is much work to do in establishing peace among rivals. The success of any hostel peace project should be judged on its long term ability to rebuild positive relationships and structures in society. A number of peacebuilding exercises would not only serve this purpose, but also benefit the community. In post-World War II Europe, work camps were used to bring youth from countries who previously fought one another to rebuild areas devastated by the war. This started the peacebuilding process by bringing former enemies together to create positive relationships, while providing a service to the communities involved. This type of project could greatly benefit the townships in South Africa. Rates of unemployment throughout the country are approaching 50%. In the townships, they are even higher. Combining this labor force with the need for housing, upgrading of the hostels, and community services (e.g. schools, parks and recreation facilities, electricity, water, etc.) would provide an excellent opportunity for creating work camps to bring enemies together to build up their community's infrastructure.

It would serve at least three purposes. First, the laborer would gain employment and, perhaps, new marketable skills. Second, by assisting the upgrading of the workers' hostels and the surrounding community, it would give the migrant a greater stake in the community in which he lives. Finally, this type of effort could exploit successful micro-encounters of these workers to create group membership symbols and increase their emotional energy. As relationships become more personal and long-term, they enable the peacebuilding process to take root (Collins, 1988; 1975).

Another recommendation for continuing peacebuilding activities deals with the problems experienced by hostel and townships residents when interacting with the police. The Institute for a Democratic Alternative in South Africa (IDASA) has started facilitating conferences designed to reconcile relationships between the community and the police. Historically police and defense force presence in the township was not used to provide safety for the residents but was used to harass activists and enforce apartheid laws. IDASA's ability to convene talks which include political enemies would be important in broaching this volatile issue. It would allow community and hostel members to express their frustration with the current lack of security. And as communities set up justice projects or self-defense units, it would allow the police assist in their planning and implementation, permitting each side to define how they will work with the other.

At South African Council of Churches (SACC) head office in Johannesburg, a number of ministers have expressed frustration over the intractability of hostel conflict. In the townships themselves ministers talk about their hope to re-integrate hostels into the communities within which they are found. The church continues to play a large role in this transition to majority rule. The hostel project was born from an SACC initiative. Peacebuilding is well-suited to the church because of its credibility in the townships and its mission of developing sustained relationships amongst people. Using Christian values and symbols would provide a strong basis for constructing common ground amongst a "peace constituency" in the township.

The Goldstone report on hostels acknowledges the diversity of hostel conflict in geographical regions throughout South Africa. This section focused only on the violence occurring on the East Rand. Other regions have their own manifestations of these disputes. Even within the East Rand hostel conflict can vary considerably from township to township. Still, there is much to learn from each attempt to resolve some aspect of this violence. The effective dispute systems design will not only use successful approaches, but seek to understand why unsuccessful efforts failed. It may have little to do with the specific method of peacekeeping, making or building and more to do with how and when the procedure was implemented. Resurfacing ideas can be as valuable as creating new ones, especially when the stakes are so high.

Developing support for the new system

Once the design is created, the next step works at gaining a commitment from the affected parties. Before formalizing the design, each party should have an opportunity to review and comment on the new dispute handling system. During this process, the intervenor should identify formal and informal leaders who capable of either promoting or denigrating the new system within their scope of authority. Plans to use the promoters or persuade the detractors should be developed to aid the campaign to acquire formal acceptance of the new system. Once this is accomplished, the process of education begins. All the parties affected by the dispute handling changes should be briefed on how the system works and what the benefits of the new procedures are (Moore, et.al., 1992).

Implementation of design

Implementation of the new procedures is training intensive. People responsible for system operation must be taught to diagnose arising disputes and refer them to the most appropriate dispute handling mechanisms. People providing services to the new system must have their roles defined and know how to refer clients back to the system operators when their services do not match the needs of the requestor. Clients must learn how to access the system and, perhaps, unlearn former responses to conflict (Moore, et.al., 1992). An extensive grassroots education plan is necessary to develop support for using the new dispute systems. For people involved on the frontline (e.g. monitors, people operating rumor control stations), a thorough knowledge of how to access the dispute mechanisms is important. For both hostel and taxi conflict, local peace committees could serve as the foundation for this type of effort. The LPC's could partner schools, churches and NGO's to act as agents of dissemination about how the new system works, as well as educate members of these organizations to provide the required services. One last aspect of implementation is putting procedures in place to monitor and record cases as they move through the new system. The tracking of cases is an important source of feedback for the dispute system. This information is used in the ongoing stage of operating the new dispute system.

Operating the new system

This final phase of dispute systems design deals with managing the daily operation of the new system. This stage uses information from case records to determine the effectiveness of certain dispute handling mechanisms, types of disputes the system is handling and information about the clients using the system (Moore, et.al., 1992). These data feed information back to the people managing the system. This feedback is used to modify the system to reduce destructive and structural conflict. Feedback also helps direct strategic plans for the system. Creation of a feedback loop permits the system to adapt to the conflict environment in which it resides, making the dispute system resistant to obsolescence.

The seven steps to dispute systems design presented above provide a comprehensive approach to managing chronic forms of conflict. These steps empower people within the system of conflict by institutionalizing procedures and roles to help them resolve or manage conflict. Dispute systems design attempts to use interest-based methods to encourage building relationships while addressing issues of conflict. If the designed procedures and roles are effective in helping the disputants negotiate successful interaction rituals, they are likely to continue using them. Shifting the emphasis on power-based to interest-based means of addressing conflict creates opportunities for macrostructural changes through the new conflict handling institutions and the relationships they are developing.



Dispute systems design attempts to institutionalize ways of handling various types of disputes within a system of conflict. By focusing on less costly forms of decision making and dispute resolution and management, DSD emphasizes reconciling all parties' needs, desires and fears. While choices made by the change team and designer will generally benefit majority interests in the system, an effective DSD will find ways to protect the rights and interests of minority groups.

In systems of protracted, destructive conflict, power-based and rights-based conflict engagement techniques are wielded to gain an advantage over other parties or to repress manifest conflict. In either case these techniques often sacrifice elements (e.g. transaction costs, satisfaction with dispute outcomes, relationships and dispute recurrence) which in the long term promote healthier aspects of conflict. Dispute systems design seeks to address this problem by encouraging parties to use interest-based, or less costly rightsbased and power-based, methods of dispute resolution (Ury, et.al., 1988). By de-escalating tension in the system, parties can reduce the level of internal group cohesion necessary to wage conflict. This provides a stabilizing force in the system by permitting members to reach across party boundaries to build relationships.

In distressed systems of conflict, symbolic or literal defeat of the other party becomes a sacred group object. If damaging the enemy through power-based or rights-based conflict techniques is most efficacious, these options will be pursued.

In micro-encounters, the success of a ritual exchange will be calculated by the tactical measure of "who wins." What is often lost in this process is the strategic value of the party's general well-being. As the number of conflict interactions increases over space, time and number of people, the resultant macrostructures mirror the victories and defeats of the parties engaged in the conflict.

At a point in time when the group realizes that the sacredness of defeating the enemy has negative consequences, space is created in the system for the emergence of alternative decision making and conflict management techniques. During the time that old group symbols are transformed into new ones, the dispute systems designer can have the greatest impact on the system. Through educating disputants about the total cost of continuing powerful waging practices and by explaining the value of interest-based methods, the designer can help parties determine what types of conflict behaviors, norms and procedures become sacred. The more parties to the conflict agree on these new, sacred waging practices, the greater the level of social integration in the system (Coser, 1956). One explanation for this is found in the analysis of micro-encounters.

Collins' (1981) theory on successful micro-rituals suggests that it is the resultant emotional energy which strings these interactions across space and time. In distressed conflict systems, the success of a micro-ritual is based on an interactant's ability to literally or symbolically beat the opponent. In this situation, there is one person who gains emotional energy and one who loses it promoting intra-group, rather than inter-group, solidarity. The dispute systems designer must transform the sacred nature of the competitive micro-interaction into something which builds inter-group solidarity. The reason interest-based procedures are best suited for this purpose is that in the micro-context, the primary goal is to meet one's interests, not defeat the other disputant. With the party's interests becoming the primary sacred object, two things occur. First, there is a greater satisfaction with the outcome because, amongst other things, damaging the other party is not necessarily in either disputant's best interests. Second, it is possible for opposing party members to engage in a successful micro-ritual which results in an emotional energy boost for both. When these types of inter-group interactions are strung across space, time and number of people, the emerging macrostructures reflect these relationship-oriented conflict management practices. In the former, escalated system of conflict, clear-cut party boundaries are replaced by more interdependent system units, decreasing the violence and intensity of the conflict (Coser, 1956).

Dispute systems designs for hostel and taxi conflict show how this can occur. In the Western Cape Taxi War, prior to the creation of CODETA, the single taxi association, WEBTA and LAGUNYA were engaged in a number of power-based struggles. What was lost in the battle to determine who was strongest was the reality that they wielded more power together than apart. Design processes which focused on uniting them through their common economic interests aided the transformation of the sacred object of "defeating the other party" to the new group symbol, CODETA. This symbol stood for a reduction of violence and greater economic and political power for both parties. This promotion of interdependence reduced destructive forms of conflict. In hostel conflict, the functional interdependence of parties could be linked to a number of issues (e.g. limiting of violence, community infrastructure, political power, etc.). The DSD for this conflict system suggests that initiatives which provide housing or utility services are one way to invest both parties in the peacebuilding process. The betterment of living conditions, which is precluded by violence in certain areas, becomes a possibility when both parties recognize the other's role in this joint interest.

The above examples show why DSD is better equipped to handle systems of destructive conflict than are single-threaded or nonintegrated peace efforts. Single-threaded attempts may wither before substantive peacemaking and peacebuilding options can take root. Non-integrated efforts lack the advantage of strategic coordination. The dispute systems design incorporates both tactical and strategic conflict waging alternatives to provide short-term ways to keep the peace while waiting for longer term approaches of making and building peace to germinate.

While this comprehensive approach provides an effective method for transforming systems of less tractable conflict, the institutionalized nature of dispute systems design has some inherent limitations. DSD atomizes conflict. With safety-valve mechanisms in place to handle typical disputes, there is less natural build-up of latent conflict. Reducing the effect of this energy-producer on the system lessens its impact on how frequently, quickly and radically the system is altered when tension builds to a breaking point.

Even though there is a feedback loop which seeks to modify the system to handle new forms of conflict, this adaptation is contravened by the process of growing bureaucratization of dispute handling institutions. Especially in South Africa, where unemployment is high and the political transition has created a tremendous power upheaval, people who provide dispute management services in a newly designed conflict system may resist changes dictated by feedback. If status or power in a structure is threatened, it is likely that service providers will create rules which modulate system adaptation to favor their interests. This growing rationalization makes the system less flexible. Over time, rigid processes for handling disputes run the risk of cracking apart either through disuse or lack of legitimacy.

Another problem for which the dispute systems design process is not well-suited concerns with extra-institutional forms of grievance redress. While DSD can determine how to handle extrainstitutional waging techniques, it cannot govern the forms that are popularly acceptable. As power bases form in the new system, groups may find themselves in positions where they will have to use empowerment techniques (e.g. collective action, non-violent direct action, terrorism, etc.) to modify the distribution of power. Attempts to co-opt an arising group into some institutionalized process while they are in the process of increasing their group's number may be seen as detrimental to their cause. The DSD process can be designed to let these conflict forms run their full course; however if the challengegroup successfully raises the question of the system's legitimacy, it is unlikely that allowing the system to adapt to the changing environment will be seen as an acceptable solution.

National political transition is a complicated and uncertain process. Since 1990, South Africa has used many of the standard practices (e.g. military intervention, high-level negotiations, education of the public) at keeping and sustaining peace. The National Peace Accord document is a remarkable example of how to create structures designed to foster peace from leadership to grassroots. Even though aspects of the Accord are maligned, there have been some notable successes. Nonetheless, many protracted, violent conflicts persist. As less costly intervention techniques are deemed unsuccessful, frustrated calls for more powerful methods of intervention arise. But as recent scenes from Somalia show, even military intervention does not provide an automatic solution. This is because the diversity of conflict types require a variety of techniques for resolving or managing them. Dispute systems design works at identifying the many disputes which comprise the system of conflict. DSD then seeks to find the most appropriate ways to handle these disputes. And along the way, it prepares the people within the system to learn or use skills which promote constructive forms of conflict.


Adler, Patricia and Adler Peter. Membership Roles in Field Research. Newbury Park, CA: Sage Publications, Inc., 1987.

Anstey, Mark. "Mediation in the South African Transition: A Critical Review of Developments, Problems and Potentials." Address. First European Conference on Peacemaking and Conflict Resolution. Antalya, Turkey, 1992a.

____________. "Some Comparisons Between Labour and Community Mediation." Unpublished paper, 1992b.

Augsburger, David. Conflict Mediation Across Cultures. Louisville, KY: Westminster/John Knox Press, 1992.

Babbie, Earl. Observing Ourselves. Belmont, CA: Wadsworth Inc., 1986.

____________. The Practice of Social Research. Belmont, CA: Wadsworth Inc., 1992.

Blake, R. and Mouton, J. Solving Costly Organizational Conflict. San Francisco: Jossey-Bass Publishers. 1984.

Blau, Peter. Exchange and Power in Social Life. New York, NY: John Wiley, 1964.

Blumer, Herbert. An Appraisal of Thomas and Znaniecki's "The Polish Peasant in Europe and America." New York: Social Science Research Council. 1939.

Burgess, Guy and Heidi Burgess. "Intractable Conflict and Constructive Confrontation." University of Colorado Conflict Resolution Consortium Working Paper #91-6, August 1991.

Burgess, R. and Bushell, D., Behavioral Sociology. New York: Columbia Press, 1969.

Burton, John. Conflict: Resolution and Provention. New York, NY: St. Martin's Press, 1990.

Cardenas, Leo. "Field notes: Structure and Operations of Rumor Control/Verification Centers." United States Department of Justice, Community Relations Service. Unpublished paper, 1993.

Collin, Susan. "Police Deviance in the Western Cape Taxi War." Centre for Intergroup Studies, Cape Town. August, 1992.

Collins, Randall. Conflict Sociology. New York, NY: Academic Press, Inc., 1975.

________________. On the Microfoundations of Macrosociology. American Journal of Sociology, 86 (March 1981): 984-1014.

________________. Theoretical Sociology. Orlando, FL: Harcourt Brace Jovanovich, Inc., 1988.

Cook, K. and Emerson, R., eds. Social Exchange Theory. Beverly Hills, California: Sage Publications, 1987.

Cook, K., Emerson, R., Gillmore, M., and Yamagishi, T., The Distribution of Power in Exchange Networks: Theory and Experimental Results. American Journal of Sociology, 89 (September 1983): 275-305.

Coser, Lewis. The Functions of Social Conflict. Glencoe, IL: The Free Press, 1956.

____________. Continuities In The Study Of Social Conflict. New York, NY: The Free Press, 1967.

____________. Masters of Sociological Thought. Ft. Worth, TX: Harcourt Brace Jovanovich, Inc., 1977.

Crush, J., Jeeves, A. and Yudelman, D. South Africa's labor empire. A history of black migrancy to the gold mines. San Francisco, CA: Westview Press, 1991.

Curle, Adam. In the Middle: Non-official Mediation in Violent Situations. New York, NY: St. Martin's Press, 1986.

Durkheim, Emile. 1965 [1915] Elementary Forms of Religious Life. New York, NY: Free Press.

Emerson, Richard. Power-Dependence Relations. American Sociological Review, 17 (February 1962): 31-41.

Emerson, Richard. Power-Dependence Relations: Two Experiments. Sociometry, 27 (September 1964): 282-298.

Emerson, Richard. 1972(a). Exchange theory, part I: A psychological basis for social exchange. In J.Berger,

M.Zelditch, and B.Anderson(eds.). Sociological Theories in Progress (Vol.2), pp.38-57. Boston:Houghton-Mifflin.

Emerson, Richard. 1972(b). Exchange theory, part II: Exchange relations and networks. In J.Berger, M.Zelditch, and B.Anderson(eds.). Sociological Theories in Progress (Vol.2), pp.58-87. Boston:Houghton-Mifflin.

Emerson, R. and Cook, K. Power, Equity and Commitment in Exchange Networks. American Sociological Review, 43 (October 1978): 721739.

Fisher, R. and Ury, W. Getting to Yes: Negotiating Agreement Without Giving In. New York, NY: Penguin Books, 1981.

Garcia, Angela. Dispute Resolution Without Disputing: How the Interactional Organization of Mediation Hearings Minimizes Argument. American Sociological Review, 56 (December 1991): 818835.

Garfinkel, Harold. Studies in Ethnomethodolgy. Englewood Cliffs, NJ: Prentice-Hall, 1967.

Goffman, Erving. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday and Company, Inc., 1959.

Goffman, Erving. Interaction Ritual. Garden City, NY: Doubleday and Company, Inc., 1967.

Homans, George. Social Behavior: Its Elementary Forms. New York, NY: Harcourt, Brace & World, Inc., 1961.

Kraybill, Ron. "The Illusion of Neutrality." Track Two, Centre for Intergroup Studies, Cape Town. November, 1992.

Kriesberg, Louis. The Sociology of Social Conflicts. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1973.

Kriesberg, L., Northrup, T. and Thorson, S. Intractable Conflicts And Their Transformation. Syracuse, NY: Syracuse University Press, 1989.

Kuper, Leo. Passive Resistance in South Africa. New Haven, CT: Yale University Press, 1971.

Lamb, David. The Africans. New York, NY: Random House, Inc. 1987.

Lederach, John Paul. "Conflict Transformation in Protracted Internal Conflicts: The Case for a Comprehensive Framework." Unpublished paper, 1992a.

___________________. "Beyond Prescription: New Lenses for Conflict Resolution Training across Cultures." Inter-racial and Cross-cultural Conflict Resolution Project Working Paper. Conrad Grebel College, Waterloo, Ontario. May 1992b.

___________________. "Pacifism in Contemporary Conflict: A Christian Perspective". United States Institute of Peace, July 1993a.

___________________. Director, International Conciliation Service of the Mennonite Central Committee. Interview. Middle East Report. March-April 1993b: 38-42.

___________________. "Tapping the Tradition and Sustaining Reconciliation in Somalia: The Relevance of Culture in Building a Comprehensive Process." NCPCR Conference, May 1993c.

___________________. "Of Nets, Nails and Problems: A Folk Vision of Conflict in Central America." PhD. dissertation, University of Colorado. 198?.

Lipton, M. "Men of two worlds. Migrant labour in South Africa." Optima, 29:72-201, 1980.

Lofland, J. and Lofland, L. Analyzing Social Settings. Belmont, CA: Wadsworth Inc., 1984.

Midgley, J.R. "Implementing the Peace Accord." Address. Fifth Conference on Negotiation and Mediation in Community and Political Conflict in South Africa." Port Elizabeth, South Africa, November 1992.

Minnaar, Anthony. "Taxi Wars: A Political or Economic Struggle?". Centre for Conflict Analysis, Human Sciences

Research Council, Pretoria, 1993.

________________. "Communities Under Siege." Centre for Conflict Analysis, Human Sciences Research Council, Pretoria, 1993a.

________________, editor. Communities in Isolation: Perspectives on Hostels in South Africa. Human Sciences Research Council, Pretoria. 1993b.

Minnaar, A. and Torres, S. "'Riding the Tiger!': Crisis Time in the Taxi Industry." Centre for Conflict Analysis, Human Sciences Research Council, Pretoria, 1993c.

Minnaar, A. Keith, T. and Payze, C. "SDUs or Comtsotsis:

Criminal Gangs and Political Violence. Centre for Conflict Analysis, Human Sciences Research Council, Pretoria, 1993d.

Molahlehi, Edwin. "Defusing Violence in South Africa: The Move to Establish Community Dispute Resolution Centers." Interview. Forum. National Institute for Dispute Resolution. Winter 1993. pp. 23-27.

Moore, Christopher. The Mediation Process. San Francisco, CA: Jossey-Bass Publishers, 1989.

__________________. "Implementing Peace Accords on the Ground." Track Two, Centre for Intergroup Studies, Cape Town. May, 1993.

__________________. "'Have Process, Will Travel' Reflections on Democratic Decision Making and Conflict Management Practice Abroad." Forum. National Institute for Dispute Resolution. Winter 1993. pp. 1-12.

Moore, C.W., Priscoli, J.D., Wildau, S.T., Smart, L.E. and Mayer, B.S., "Introduction to Dispute Systems Design". CDR Associates, Boulder, Colorado, 1992.

Nina, Daniel. "(Re)making Justice in South Africa: Popular Justice in Transition." Centre for Social and Developmental Studies Working Paper Number 8, University of Natal, Durban, 1993.

Omond, Roger. The Apartheid Handbook. Harmondsworth, Middlesex, England: Penguin Books Ltd., 1986.

Pomerantz, Anita. "Agreeing and Disagreeing with Assessments: Some Features of Preferred/Dispreferred Turn Shapes." In J. Atkinson and J. Heritage (eds.). Structures of Social Action: Studies in Conversation Analysis, pp. 57-101. Cambridge,

England: Cambridge University Press.

Pretorius, W. "An analysis of the current status and viability

of the industry." Address. Taxi Indaba. Johannesburg, August 1993.

Schutz, A. Collected Papers I: The Problem of Social Reality. (ed. by M. Natanson). The Hague: Martinus Nijhoff, 1962.

Seekings, Jeremy. "People's Courts and Popular Politics." South African Review 5. Ravan Press: Johannesburg. 1989.

Sharp, Gene. The Politics of Nonviolent Action. Boston, MA: Porter Sargent Publishers, 1973.

Shaw, Mark. "Crying Peace Where There Is None?" Centre for

Policy Studies Transition Series, Research Report Number 31. August, 1993.

Simmel, Georg. Conflict and The Web of Group-Affiliations. New York, NY: The Free Press, 1955.

_____________. The Sociology of Georg Simmel. (ed. by K. Wolff). New York, NY: The Free Press, 1950.

Sisk, Timothy. "South Africa's National Peace Accord." Peace and Change, 19 (January 1994): 50-70.

South Africa. Goldstone Commission of Inquiry. Interim report

on violence in the taxi industry. 10 June, 1992.

South Africa. Goldstone Commission of Inquiry. Second interim report on violence in the taxi industry. 2 July, 1992.

South Africa. Goldstone Commission of Inquiry. Third interim report on violence in the taxi and minibus industry. 4 December, 1992.

South Africa. Goldstone Commission of Inquiry. Fourth interim report on violence in the taxi and minibus industry. 23

February, 1993.

South Africa. Goldstone Commission of Inquiry. Fifth interim report on taxi violence. 26 July, 1993.

South Africa. Goldstone Commission of Inquiry. Final report of the Multi-national Panel regarding the curbing of public violence and intimidation during the forthcoming election. 11 August, 1993.

South Africa. Goldstone Commission of Inquiry. Report on the planning or instigation of acts of violence by members of the South African Police in the Vaal area . 27 October, 1992.

South Africa. Goldstone Commission of Inquiry. Report on the inquiry into the police response to and investigation of events in Boipatong on 17 June 1992. 20 July, 1992.

South Africa. Goldstone Commission of Inquiry. Final report of the Committee appointed to inquire into train violence. 6 May, 1993.

South Africa. National Peace Committee. National Peace Accord.

14 September, 1991.

South Africa. International Alert, United Kingdom. Mission to Evaluate the National Peace Accord and its Peace Structures. May 9, 1993.

Sparks, Allister. The Mind of South Africa. New York: Ballentine. 1990.

Thomas, W. and Znaniecki, F. The Polish Peasant in Europe and America. (ed. by Eli Zaretsky). Chicago, IL: University of Illinois Press. 1984.

Turner, Jonathan. The Structure of Sociological Theory. Belmont, CA: Wadsworth Publishing, 1991.

Turrell, R. Capital and labour on the Kimberly diamond fields, 1871-1890. Cambridge, England: Cambridge University Press,


Ury, W.L., Brett, J.M., and Goldberg, S.B., Getting Disputes Resolved: Designing Systems to Cut the Costs of Conflict. San Francisco, CA: Jossey-Bass Publishers, 1988.

van der Merwe, H. and Odendaal, A. "Mediating for a PostApartheid South Africa." Peace Research, 24 (February 1992): 2535.

Weber, Max. The Methodology of Social Sciences. Glencoe, IL:

Free Press, 1949.

Wehr, Paul. Conflict Regulation. Boulder, CO: Westview Press, Inc., 1979.

Wehr, P. and Lederach, J.P., "Mediating Conflict in Central America." Journal of Peace Research. vol. 28, no. 1, 1991: 8598.

Wentzel, Marie. "Historical origins of hostels in South Africa: Migrant labour and compounds." Communities in Isolation. Ed. Anthony Minnaar. Human Sciences Research Council, Pretoria,

1993: 1-9.

Whalen, J., Zimmerman, D. and Whalen, M. When Words Fail: A Single Case Analysis. Social Problems. 35: 335-362.


Beresford, David. "Gertrude Mzizi - Thokoza's living target."

The Weekly Mail. July 16 to 22, 1993: 10.

Davis, Gaye. "Mourners shot at in new Cape taxi violence." The Weekly Mail. July 9 to 15, 1993: 4.

Dube, Pamela. "Man with a mission: Peace in the hostels." The Weekly Mail. October 15 to 21, 1993: 12.

Forrest, Drew. "The strange past of the mysterious 'Q'." The Weekly Mail. March 25 to 30, 1994:

Gevisser, Mark. "The mothers who are warriors." The Weekly

Mail. July 16 to 22, 1993: 11.

Golding, Jacquie. "This house isn't a home." The Weekly Mail. October 29 to November 4, 1993: 11.

Koch, E. and Maurice, P. "Secret Koevoet base linked to Boipatong." The Weekly Mail. June 26 to July 2, 1992: 1-2.

________________. "Silent night with the shacks patrol." The Weekly Mail. December 17 to 22, 1993: 9.

Laufer, Stephen. "The lid's back on - until next time." The Weekly Mail. July 9 to 15, 1993: 7.

_______________. "What can be done to end the violence? Increasingly, all sides seem to be saying: bring in the troops." The Weekly Mail. August 6 to 12, 1993: 10-11.

_______________. "No official war, but people are dying on the East Rand." The Weekly Mail. September 24 to 30, 1993: 7.

Makhanya, Mondli. "Riding to the rescue of the taxi industry." The Weekly Mail. September 10 to 16, 1993.

Mvoko, Vuyo. "East Rand peace initiative collapses." The Weekly Mail. December 30, 1993 to January 4, 1994: 3.

Strober, Paul. "Was Boipatong an act of vengeance?" The Weekly Mail. June 26 to July 2, 1992: 3.

Stober, P. and Laufer, S. "Police general linked to gun

running." The Weekly Mail. March 18 to 24, 1994:

"Waddington memo led to SAP 'purge'." The Weekly Mail. August

28 to September 3, 1993: 1-2.

Waldman, Amy. "A look inside the shabby citadel of Kathlehong." The Weekly Mail. August 13 to 19, 1993: 10-11.


South Africa. map. The Atlas of Africa. New York, NY: The Free Press. 1973: 271.



Respondent 1. Natal Regional Peace Committee, Staff. Durban: August 31, 1994.

Respondent 2. Natal Regional Peace Committee, Member. Durban: August 31, 1994.

Respondent 3. Independent Projects Trust, Trainer. Durban: August 31, 1994.

Respondent 4. Rural Advice Centre, Transvaal Region. Johannesburg: September 9, 1994.

Respondent 5. Northern Transvaal Regional Peace Committee, Member. Pretoria: September 7, 1993.

Respondent 6. Western Transvaal Regional Peace Committee, Staff. Pretoria: September 7, 1993.

Respondent 7. Institute for a Democratic Alternative for South Africa, Northern Transvaal, Staff. Pretoria: September 8, 1993.

Respondent 8. Institute for a Democratic Alternative for South Africa, Transvaal, Staff. Pretoria: September 8, 1993.

Respondent 9. Institute for a Democratic Alternative for South Africa, Northern Transvaal, Staff. Pretoria: September 8, 1993.

Respondent 10. Human Sciences Research Council, research specialist. Pretoria: September 8, 1993.

Respondent 11. Institute for a Democratic Alternative for South Africa, Northern Transvaal, Staff. Pretoria: September 8, 1993.

Respondent 12. African Centre for the Constructive Resolution of Conflict, Staff. Pretoria: September 9, 1993.

Respondent 13. North East Rand Peace Committee, Member. Pretoria: September 9, 1993.

Respondent 14. Tembisa Local Peace Committee, Monitor. Johannesburg: September 9, 1993.

Respondent 15. Eastern Transvaal Regional Peace Committee, Member. Pretoria: September 10, 1993.

Respondent 16. Community Development Projects, Staff. Johannesburg: September 10, 1993.

Respondent 17. University of Port Elizabeth, South Africa Students Committee, Member. Port Elizabeth: September 16, 1993.

Respondent 18. Eastern Cape Peace Committee, Member. Port Elizabeth: September 16, 1993.

Respondent 19. Independent Mediation Service of South Africa, Staff. Port Elizabeth: September 16, 1993.

Respondent 20. St. Thomas High School, Teacher. Port Elizabeth: September 16, 1993.

Respondent 21. Eastern Cape Regional Peace Committee, Staff.n Port Elizabeth: September 17, 1993.

Respondent 22. University of Port Elizabeth, Professor of Political Science. Port Elizabeth: September 17, 1993.

Respondent 23. United Nations, Monitor. Port Elizabeth: September 19, 1993.

Respondent 24. United Nations, Monitor. Port Elizabeth: September 19, 1993.

Respondent 25. Western Cape Peace Committee, Member. Cape Town: September 21, 1993.

Respondent 26. University of Cape Town Institute of Criminology, Research Specialist. Cape Town: September 22, 1993.

Respondent 27. Centre for Intergroup Studies, Staff. Cape Town: September 22, 1993.

Respondent 28. Independent Mediation Service of South Africa, Staff. Johannesburg: September 24, 1993.

Respondent 29. Independent Mediation Service of South Africa, Staff. Johannesburg: September 24, 1993.

Respondent 30. Wits Vaal Peace Committee, Member. Johannesburg: September 28, 1993.

Respondent 31. African Law Review, Staff. Johannesburg: September 29, 1993.

Respondent 32. African Centre for Constructive Resolution of Conflict, Staff. Boulder: February 13, 1994.

Respondent 33. Iliff School of Theology, Southern African Historian. Boulder: March 14, 1994.




The ANC, Inkatha and the South African government have been involved in violent, destructive conflict for decades. Prior to the unbanning of political parties in 1990, the ANC and Inkatha were engaged in a low intensity civil war in Natal/kwaZulu. During 1990, violence erupted in a number of other provinces throughout the country. The major political players came together to create a plan to mitigate the violence and set political foes on the path to tolerance and reconciliation. This comprehensive peace plan was called the South African Peace Accord.

While its implementation is often maligned (International

Alert, 1993; Sisk, 1993; Shaw, 1993), this document stands as one of the most comprehensive national plans for fostering a peaceful governmental transition. The Peace Accord document sets out a structure which is tasked with encouraging groups to maintain a commitment to peaceful resolution of conflict. The Accord attempts to gain legitimacy by having its committees represent the people and the communities which they serve. All organizations created to perform duties defined by the Accord use a consensus model for making decisions (Midgley, 1992).

At the national level, the National Peace Committee (NPC) is tasked with monitoring and implementing the Accord as a whole and ensuring compliance with "The Code of Conduct for Political Parties and Organisations." The NPC is comprised of representatives from each of the primary parties, leaders from the religious and business communities, and representatives from many of the signatories parties to the Accord. This is the only body which can modify the constitution of the Accord. Also at

the national level is the National Peace Secretariat (NPS). The function of this body is to establish, coordinate and monitor the Regional Peace Committees (RPC's) and the Local Peace Committees (LPC's).

The RPC's generally operate at the provincial level. These regional committees are tasked with creating a body which represents the relevant political, religious, union, business, racial/ethnic, police and defense organizations in that region. The duties of the RPC's include advising the NPS on matters causing violence or intimidation, settling disputes causing violence or intimidation, guiding the LPC's, and monitoring peace agreements at the regional level.

The LPC's report to the RPC's (see figure 3 on page 102). Their tasks include creating trust and reconciliation at the grassroots level, cooperating with local authorities to combat intimidation and violence, settling disputes at the local level, gaining agreement upon the rules and conditions relating to marches and rallies, and promoting compliance with current and future peace accords. The Accord suggests that LPC's use mediation, arbitration or adjudication to resolve disputes at this level. It requires that disputes are handled simply and expeditiously and that parties to the dispute participate in the process. This creates an environment where LPC's are accountable to the communities they serve and the communities are responsible for waging peace in their particular location (Midgley, 1992).

Each of the regions has experienced varying degrees of success in implementing the RPC and LPC structures. In regions where there is large degree of political stability, such as the Eastern Cape, the RPC's and LPC's have shown their effectiveness. In regions where there is political instability, such as the Wits-Vaal (Johannesburg) area, these bodies have struggled to implement Accord measures. In fact, members of the peace committees have been intimidated and killed in volatile regions.

There are a number of problems that RPC's and LPC's are experiencing in their effort to stem the level of violence in their communities. At the grass roots, people are not always aware of the purpose of the peace committees. In black communities, some members of the white government now sit on the peace committees. This, combined with the fact that these committees are funded by Justice Department monies, creates a wariness of the organization. A second problem is providing enough experienced dispute resolvers at the local levels. Many committee members sit on numerous political, business, religious, and civic committees. For this reason, it is not feasible for them to focus solely on their Peace Accord tasks. This prevents expeditious response to flashpoints. One final problem deals

with the general lack of cohesion found within the structure of the Peace Accord. There is insufficient communication from the policy makers at the top of the Accord structure to the grassroots. This results in a lack of common vision and unity (International Alert, 1993).

Even with the problems experienced in implementing this concept, LPC's and RPC's have had a number of notable successes. They have played an important role in the taxi and hostel conflict researched for this thesis. In addition, they are a resource that an effective dispute systems designer would incorporate into any conflict resolution/management plan.

[1] This working paper is Arnold's Master's thesis. All ideas presented are those of the author and do not necessarily represent the views of the Consortium, the University, or Hewlett Foundation. For more information, contact the Conflict

Resolution Consortium, Campus Box 327, University of Colorado, Boulder, Colorado 80309-0327. Phone: (303) 492-1635, e-mail: crc@cubldr.colorado.edu.

Copyright 1994. Kent Arnold. Do not reprint without permission.

[2] The Goldstone Report released on March 19, 1994 linked highlevel officers of the South African Police (SAP) to an

SAP Inkatha gun-running scheme on the Reef and in the Natal Province. This plan was geared to produce political instability by instigating violence between Inkatha and the ANC (Weekly Mail 18-24 and 25-30, 1994).