Previous Page Table of Contents Next Page

Land conflict management in South Africa: lessons learned from a land rights approach - D. Bosch

Independent land and labour dispute mediator and arbitrator

The various people involved in land mediation in South Africa have learned many lessons from the experience of the past nine years of land mediation. Unfortunately, very few have been recorded and analysed to date. These lessons can help to improve the use of mediation locally, but could also assist others in planning and defining the appropriate use of land mediation in their countries.

Despite the fact that there are some difficulties regarding the use of land mediation, the lessons described in this article have contributed to the improved use of mediation in various areas in South Africa. Similar processes have since been built into legislation on landlord - tenant disputes in urban housing and in environmental legislation.

La gestion des conflits fonciers en Afrique du Sud: leçons tirées d'une approche fondée sur les droits fonciers

Les diverses personnes ayant participé à la médiation foncière en Afrique du Sud ont tiré de l'expérience des neuf dernières années dans ce domaine de nombreuses leçons qui, malheureusement, n'ont guère été enregistrées ou analysées à ce jour. Or, ces leçons pourraient être mises à profit non seulement pour améliorer l'utilisation de la médiation au niveau local, mais aussi pour aider d'autres personnes à planifier et à définir l'utilisation appropriée de la médiation foncière dans leur pays.

Même si certaines difficultés ont été constatées, les leçons décrites dans l'article ont contribué à promouvoir la médiation dans diverses régions de l'Afrique du Sud. Des processus analogues ont depuis été intégrés dans la législation relative aux différends entre propriétaires et locataires de logements urbains et dans la législation relative à l'environnement.

Gestión de los conflictos por la tierra en Sudáfrica: enseñanzas de un enfoque basado en los derechos

Los diversos grupos de personas que han recurrido a la mediación agraria en Sudáfrica han extraído numerosas enseñanzas de la experiencia acumulada en los últimos nueve años. Desgraciadamente, muy pocas de estas lecciones se han podido registrar y analizar hasta la fecha. Dichas enseñanzas pueden ayudar a mejorar el uso de la mediación a nivel local pero también podrían resultar útiles a otros interesados para planificar y definir la utilización apropiada de la mediación agraria en sus países.

A pesar de que existen algunas dificultades con respecto al uso de la mediación agraria, las enseñanzas que se describen en este artículo han contribuido al mejorar su utilización en diversas zonas de Sudáfrica. Desde entonces, se han incorporado procesos similares en la legislación relativa a las controversias entre propietarios e inquilinos de viviendas urbanas, así como en las leyes sobre el medio ambiente.


Land history

Two-and-a-half centuries of conquest and settlement by European colonists deprived Africans of most of their land in South Africa. Economic and legal instruments were used in the late nineteenth century, and most of the twentieth, to exclude African farmers from increasingly lucrative urban markets. At the same time, the authorities started to make it first difficult, then impossible, for Africans to use land outside reserves in the more remote areas that had been set aside for them.

Since the mid-1950s, when the African National Congress (ANC) led a process of adopting a visionary document for South Africa called the Freedom Charter, the ANC had put nationalization forward as the mechanism necessary to redress decades of dispossession and destruction of black property and economic rights. This was the ANC's policy when it, and many other organizations, fought from 1960 until the early 1990s to dismantle apartheid through armed struggle, encouragement of economic sanctions and civil disobedience.

The ANC leadership did not abandon its ideas of nationalization until 1992, two years before it was elected the majority party in South Africa's first democratic election. Fears of nationalization had caused widespread concern among white farmers, business people and foreign governments. Returning exiles also opposed nationalization and expropriation because of negative experiences elsewhere, e.g. Mozambique and the Soviet Union.

Four years of constitutional negotiations resulted in an interim constitution in 1994 that heralded the start of a democratic South Africa. The "final" constitution adopted in 1996 retained the earlier negotiated "property clause". Section 25 reads that "no law may permit arbitrary deprivation of property". While expropriation is allowed "in the public interest" and this term is defined to include land reform policy, the constitution inevitably had the effect that the parameters of land reform were negotiated. Nonetheless, politicians have tended subsequently to blame the property clause for the continuing racially skewed landownership and their inability to alter it substantially.

Land laws

Land law in South Africa addresses the following aspects: land claims (often called "restitution"), land tenure and redistribution.

The main statute dealing with restitution is the Restitution Act of 1994. The Restitution Act provides for the restitution of rights in land in respect of which persons or communities were dispossessed under, or for the purposes of furthering, the object of racially based discriminatory legislation after 1913. It established the Land Claims Commission with a remit to investigate and mediate land claims, and a Land Claims Court, which may hear land claims disputes.

Existing statutes dealing with tenure include the Land Reform (Labour Tenants) Act and the Extension of Security of Tenure Act. The latter statute intends to achieve secure tenure for rural people living on land owned by others in three ways: (i) protection of occupiers against unfair evictions while regulating the circumstances under which they may be evicted; (ii) protection of other tenure rights of occupiers; and (iii) facilitation of the provision of off-farm settlement of farm workers.

The 1998 Transformation of Certain Rural Areas Act set in place a mechanism, operating on a relatively large scale, whereby issues of land tenure security can be achieved through a detailed consultation and mediation process in areas of the Western and Northern Cape.

A Communal Land Rights Bill,[24] aimed at addressing security of tenure primarily in ex-homeland areas where a large portion of rural people live, has been in the process of drafting and negotiation for the past five years. It has been haunted by controversy from left and right. The Bill aims, among others, to allow communities currently living on communal land to obtain land title and management land tenure in terms of the run of community rules. Concern has been expressed that the Bill will "exacerbate social dislocation, poverty and inequality" and that the process of land transfer, in the way proposed, would take "hundreds of years to implement" and would become "bogged down in border disputes and power struggles".

Redistribution of land is achieved mainly by using the Provision of Certain Land for Settlement Act 126 of 1993. The emphasis is currently on small farmer enterprises involving 1 - 20 small farmers, with a preference for lower numbers. The effect of this is that fewer people have benefited recently from the redistribution programme than before, because much more money is spent per beneficiary.

Apart from the low budget and concerns about the appropriateness of the focus of the present land reform programme, another serious shortcoming is the lack of effective aftercare, i.e. after land has been transferred to beneficiaries.

Progress in land reform

Despite a promising start, only modest progress has been made with land reform in the last nine years. This is a reflection of the low priority land reform has received in South Africa, and also of the many difficulties faced in achieving targets.

In 1994, when the first democratic government was elected, whites owned about 71 percent of agricultural land (not including state land or other land in urban areas) despite being only 11 percent of the population.[25] Since then, land reform has reduced white landholding by only 1.75 percent. The slow progress in delivering land reform is due at least in part to the inadequate resources allocated to land reform. This is remarkable in view of the land reform targets that the ANC had set for itself. The budget of the Department of Land Affairs (DLA) has always been small compared with other programmes, representing only 0.4 percent of the national budget. Compared with the task, the budget seems inadequate.

Land conflicts and development of land mediation

The land tenure and related conflicts can be classified as follows:

Most citizens had been stripped of land rights before 1994. For this reason, the process of formulating the above statutes focused on identifying appropriate new land rights, establishing their content and their appropriate level of priority. This was of key importance, but may also have reflected an overemphasis on land rights as against addressing people's needs regarding and interests in obtaining access to land.

The new democratic government sought to avoid counterdispossession of white landowners, partly because of the destabilizing effect this may have had. A process of negotiating land disputes and land reform was favoured.

The strong emphasis on land rights meant that much time had to be spent in assessing and identifying people's land rights. Because of the complexities of land occupation and earlier dispossession, a very complex set of land laws was developed.

These laws recognized conflicting rights and provided for the mediation of these conflicting rights. This led to the relatively extensive experience of rights-based land mediation processes in the South Africa, to which we will return shortly.

Another reason for the rights-focus in land policy is the South African's distrust of the state, especially at the time that the Constitution was negotiated. Linked to this was the belief that the ordinary courts were not equipped to deal with the complexity of land matters. Therefore, the Land Claims Commission (originally the Commission on the Restitution of Land Rights) was established, with the main task of investigating land claims and facilitating their resolution with an emphasis on negotiation and mediation.


Government institutions

The key national government department responsible for land reform is the DLA. The Land Claims Commission (above) has made extensive use of independent land mediators.

Most provincial governments also have departments responsible for aspects of land affairs, although this is a matter in which provinces are subordinate to the national government. Some provincial governments still play a key role in some land matters, such as those involving illegal occupation of provincial land or the eviction of farm workers, and they have sometimes engaged in mediations on these issues.

The National Land Reform Mediation and Arbitration Panel

In 1995, the DLA designed and constituted the National Land Reform Mediation and Arbitration Panel (NLRMAP, or "the land panel"). This formed the key element of a strategy to manage the conflict that the proposed land reform programme was expected to generate. The terms of reference for the NLRMAP were to "establish a national panel of mediators... trained and accredited ... as a resource in preventing and resolving land disputes. ... Interventions should aim to promote consensus, facilitate fair community participation and ensure efficient use of financial and human resources." Objectives included the provision of arbitration and balloting services on behalf of the DLA, and the enhancement of the capacity of departmental officials and panel members to manage and resolve land reform conflicts effectively.

The Independent Mediation Service of South Africa (IMSSA), a mediation and arbitration body with extensive experience in managing labour and community disputes, was appointed as the serviceproviding agency following a public tender process.

During the five-year period that the NLRMAP existed, the IMSSA handled 225 interventions (Fraser, 2001). The average cost per intervention was R15 000 (about US$1 875 at an exchange rate of US$1 = R8), although much of this amount was obviously spent on longer and more complicated mediations. Mediation of most disputes cost less than R5 000 (US$625) per intervention, which is reasonable in view of the particular expertise required - provided that cases for mediation have been selected with care and due regard to cost. The service operated from 1996 to November 2000, when the IMSSA closed suddenly as a result of financial problems related primarily to its labour dispute resolution services.

The DLA located the NLRMAP within its offices, and commissioned a review of its work. The review was conducted by a development consultancy which concluded that, although the NLRMAP had met its basic commitments, it had "outlived its usefulness and needs to be replaced by an approach that integrates conflict management and prevention into the core project cycle." (Fraser, op. cit., p. 20). The DLA accepted its recommendations and disestablished the NLRMAP in 2001.

No new integrated or comprehensive service has been established in the two years since the NLRMAP closed. In 2002, the DLA indicated that it believed that such a service was still necessary, and started a process of consultation and analysis in to determine how such a service should be structured. It is unclear when proposals in this regard will be formulated and another service established.

Non-governmental organizations

Non-governmental organizations (NGOs) in South Africa play a very important role in land reform and land reform mediation. They are involved in four main capacities: (i) providing legal advice and representation regarding land disputes with others, such as landowners, and within communities; (ii) facilitating community processes and providing training and advice; (iii) mediating disputes with other interested parties, including landowners, often informally; and (iv) engaging in broader policy formulation, including lobbying government.

The Legal Resources Centre (LRC) is a key NGO focusing on providing legal advice to the vulnerable and marginalized, including the poor, homeless, and landless people and communities. There are various other NGOs offering legal services who assist in similar ways. For example, Lawyers for Human Rights and the Centre for Rural Legal Studies have specific programmes providing advice and assistance regarding land and eviction disputes of farm workers. Universitybased "law clinics" and institutes, such as the Centre for Applied Legal Studies at the University of the Witwatersrand also assist or represent communities in land matters and assist in the development of land policy. The National Land Committee assists poor rural black people across eight provinces to access land rights and development resources. It serves as a network with eight affiliated land rights organizations and often works in close association with the LRC.


As explained above, disputes regarding a wide range of land rights have been mediated in South Africa. This section discusses two case studies in order to illustrate some of the lessons that can be learned from the South African land mediation experience.

The case studies relate to the mediation of land claims. Both focus on the issue of restitution of land claims to conservation land (of which there are many in South Africa) and involve successful settlement of claims resulting in the establishment of contractual national parks. The case studies raise pertinent questions about the mediation of land disputes and how these issues link to local and national issues.

The Makuleke land claim


The Kruger National Park was established in 1926 based on exclusionist principles: the area was fenced off, people were removed forcibly, and benefits went primarily to whites. This was common to conservation throughout South Africa and resulted in a relationship of hatred and suspicion between black communities and the primarily white conservation authorities. In 1969, the Makuleke community was removed forcibly from its ancestral lands in the Pafuri area (between the Limpopo, Mutale and Luvuvhu rivers) and surrounding state-owned land when the Kruger Park was extended northwards. In terms of apartheid policies, the Tsongaspeaking portion of the community was relocated to Ntlhaveni, in the then Gazankulu homeland, and Venda- speaking members to the then Venda homeland.

The Makuleke land claim was lodged in terms of the Restitution of Land Rights Act in December 1995. The claimed Pafuri area is an environmental hotspot from a biodiversity viewpoint and, in the initial stages, opposition was forthcoming from conservation circles with some saying that "if the Makuleke claim is upheld in respect of the land within the park, all conservation areas will be under threat" (Makuleke and Steenkamp, 1998).

After two years of intensive and complex negotiations a settlement agreement was reached and the Land Claims Court ordered the restoration of the Makuleke community's ancestral land, subject to various conditions aimed at ensuring that both the land's conservation status and the community's rights are protected. Under the terms of the agreement, a contractual park between the community and the South African National Parks (SANParks) was established for 50 years on 22 734 ha of pristine conservation land in the Pafuri area. The members of the community agreed to remain in Ntlhaveni, where they currently live, but aim to benefit from restitution of their land through ecotourism development.

A joint management board (JMB), consisting of members of SANParks and the community, is responsible for managing the land. SANParks has been contracted as an agent by the Makuleke to conduct day-today conservation management for an initial period. The Makuleke have full rights to develop the land for ecotourism ventures for their own financial benefit.

The mediation process and lessons learned

The mediation process ran from December 1996, involving as many as 13 parties, until a written agreement was signed in May 1998. The mediators facilitated this process until final agreement was reached in December 1998.

The mediation included joint sessions with the key parties, and later with all parties, individual sessions with each of the key parties, community hearings and rights enquiries. Many lessons were learned from the process.

The first relates to the fact that, for mediation to be an appropriate process, there must at least be the possibility of a settlement between the disputing parties. At the beginning of the Makuleke mediation, a dispute arose that could not be settled. Chief Mhinga claimed that the community fell under his authority and that, in traditional law, the community must act through him as chief when making the land claim. For this reason, the chief lodged a claim to the same land and on the same basis in parallel with the Makuleke community itself. The members of the community denied that Mhinga was their chief, maintaining that they had their own chief who had been stripped unfairly of his chieftainship owing to apartheidrelated actions against groups that opposed the government. After an initial period of deadlock, the Regional Land Claims Commissioner rejected Mhinga's claim because the Restitution Act empowers a community to claim, not a chief. This allowed the detailed mediation with the key parties to start in earnest.

The second lesson is that all the important parties must participate. Otherwise, any eventual settlement could be undermined or challenged by one not participating. However, what if a distinct but small section of the originally displaced community refuses to participate? A small group of Venda-speaking families were removed from the Pafuri to the Venda homeland, and were located about 50 km from where the Makuleke community was located. There they fell under the authority of chief Mutale. This chief refused to allow members of his community to participate directly in any settlement discussions. He insisted that he must represent the community. The Commissioner again refused on the basis of the Mhinga ruling.

Despite many attempts to convince the members of the Mutale community to participate, and the passing of many months, they declined. Their nonparticipation could have prevented the conclusion of an agreement that took 18 months to negotiate. With the concurrence of the Land Claims Court, a mechanism was therefore designed to give the right to become members of the landholding entity to people who can show that they or their ancestors were removed forcibly from the Pafuri area around 1969. This allowed the finalization of the settlement agreement and the eventual transfer of the restored land in 1999, 30 years after the Makuleke community had been dispossessed.

The Khomani San and Mier land claims


In 1995, descendants of various San families, who later decided to call themselves the Khomani San, lodged a land claim to an area in the northwest of Northern Cape Province. This claim was not for landownership but for use rights to more than 4000 km2 of land (calculated using an internationally applied reduction formula of 4:1 for hunting-gathering territory - see Chennels, 1998) in the area now known as the Kgalagadi Transfrontier Park and the Mier Municipality, which their ancestors had used in a nomadic way.

Their land claim is unique in South Africa because the San people are acknowledged as one of the first peoples of South Africa, having lived in southern Africa for more than 20 000 years. In the early 1990s, they and their language were thought by many to be extinct. The lodging of the claim brought together 300 initial land claimants. This number is expected to expand to about 1 000 as the verification process of people claiming to form part of this community is finalized. The San people were originally not one community with a joint structure of governance. They were descendants of various San clans or family groups, which historically had only occasional contact with each other. During the twentieth century, the San were "scattered all over South Africa in search of refuge ... Their centuries old culture, one of the oldest known to mankind, was gradually disintegrated."[26] The descendants of the erstwhile occupiers became a disparate group of people, some now living hundreds of kilometres away from one another.

The San claim overlapped with a claim of another community in the area, the Mier community. The Mier community came to live in the Northern Cape from about 1865 when members of the "bastervolk" fled British rule in the Cape Colony, thereby displacing many of the San. The Mier community claimed areas within the Kgalagadi Transfrontier Park from which they were displaced when a nature reserve was first established in 1931.

The settlement had to be negotiated between four main parties: the San community, the Mier community, SANParks and the Land Claims Commission. The negotiations ran for four years. A settlement framework was concluded in 1999, and a detailed agreement settling all San and Mier claims was reached in 2002.

The 1999 agreement resulted in the transfer of about 37 000 ha of land to the San community and 42 000 ha to the Mier community, each area to be used and occupied by the respective communities under rules to be established by each community. In accordance with the terms of the final 2002 agreement, the South African Government transferred the ownership of 28 000 ha of land in the Kgalagadi Transfrontier Park, called the San Heritage Land, to the San community. The government also transferred ownership of 30 000 ha of park land, called the Mier Heritage Land, to the Mier community.


There are three distinct areas regarding which lessons can be learned from this process: handling of competing claims to the same land; designing joint institutions that allow for different approaches; and resolving difficulties within one community regarding allocation of rights and management of community assets.

The issue of handling competing claims to the same land arose in a context where there were concerns that one or other of the parties may not always be able to engage, because of lack of capacity or differences in priorities. The process of mediation was complicated by the fact that there were competing claims to the same land. The following can be learned from this:

The need to design joint institutions that allow for different approaches arose in relation to managing adjacent areas of land. All three main parties to the agreement, SANParks and the Mier and San communities, own or manage land adjacent to one another.

A JMB was created to manage the overlapping issues, but allowing for differences in approach. The following lessons could be learned from the design crafted by the parties and their representatives:

The third area concerns difficulties within one community regarding the allocation of rights and management of community assets. Since the transfer of the first land to the San community in 1999, allegations have been made that: farm infrastructure has collapsed; the community has no motorized transport and virtually no livestock; much of the game on the land has been sold or poached, and the remaining game is dying of thirst because water pumps are broken; leading community members have occupied farmhouses earmarked for community tourism initiatives; and some others have occupied other parts of the land through self-help. Possible ways of avoiding such problems in the future are:


Many lessons have been learned from land mediation. This section focuses on such lessons regarding the mediation of land rights in the South African context. Because of limited space, lessons regarding the general practice of mediation are not included unless a specific land rightsrelated issue is involved.

When should mediation be used?

Mediation should be used only for disputes where it may potentially deliver a settlement or contribute to settlement. Even where mediation cannot deliver a settlement on all aspects of a dispute, if it can contribute to settlement of some of the elements of the dispute, it may be an appropriate process.

There must be a dispute

Where there is no dispute, then true mediation is not required. A criticism often raised against land mediation practice in South Africa is that matters have been referred to mediation where there is no dispute, e.g. the negotiations on the setting up of a land redistribution project. While discussion or negotiations may be held, proper mediation will be an inappropriate process unless a definite dispute that threatens an agreement has arisen.

Types of disputes that can be mediated

Practically any dispute can be mediated. The necessary conditions for mediation include:

Experience shows that court-ordered mediation is often successful, even where one party is only a reluctant participant. By convening pretrial meetings (often by teleconference), Land Claims Court judges have been able to bring parties to consent to mediation even where the parties refused mediation before the matter came to court. However, a reluctant mediation party may frustrate a mediation process, for example by failing to agree to a mediation date.

Who should mediate?

A professional land mediator could conduct the mediation. Alternatively the mediation could be conducted: by NGOs offering legal services or a focus on land reform; informally by a resource person in the area concerned, such as a priest; or by staff of a land reform agency (although some parties may object because of their involvement with policy aims).

For effective land mediation on rights issues, the following is usually required:

The importance of selecting an appropriate mediator with knowledge of land law and policy, experience of land issues, and an understanding of the difficulties that implementation of an agreement is likely to face cannot be overemphasized. Much of the criticism against land mediation in South Africa can be sourced back to mediators who may be good at mediation but who fall short in some of these areas.

NGOs often work for long periods in a given area or community. This gives them a very good understanding of local issues and the practicalities of different solutions. Thus, they can play an invaluable role in mediating and facilitating settlement of disputes on issues within a given community, such as tenure reform and disputes regarding control of assets or allocation of resources.

Who should be involved in the mediation?

All the parties to the dispute

It stands to reason that all parties to a dispute should be invited to participate in the mediation. If a key party does not participate, the dispute may arise again and an eventual settlement agreement may not be implementable.

The government land reform agency

Land reform agencies initially expected mediators to ensure achievement of the government's land reform objectives. This is inappropriate because, for effective mediation, parties need to have trust in the mediator as an independent person.

For these reasons, mediation should not be seen by the relevant land reform agencies as a way of dealing with their lack of capacity. The agencies must have the necessary capacity to engage actively in the mediation. The absence of agencies during negotiations has caused such serious difficulties in implementing agreements that some mediators have refused to mediate unless a land reform agency is represented.

However, there are times when the government officials participating do not have the necessary negotiating skills, information or mandate. While it is not the mediator's primary task to ensure adherence to the government's policies, a settlement agreement reached without proper government input may not be implementable.

Other institutions responsible for implementation

Those institutions, government and otherwise, who will be responsible for implementing or monitoring parts of any settlement agreements, should participate in, or be consulted in detail about, the mediation. They can then ensure that settlement agreements can be implemented and that the necessary resources in this regard can be obtained or allocated.

For example, where a land reform project will require agricultural extension services, the extension office in the area should be involved or at least consulted. The extension office should be able to contribute to an assessment of the viability of the project or different models for structuring it so that it will be more viable. Without such involvement, the extension service may not dedicate sufficient resources to supporting the project.

Other factors influencing the efficacy of land mediation

The efficacy of land mediation can be assessed in terms of factors such as the speed of the mediation process, or the cost per case or per beneficiary. While they are relevant, a key factor in the final analysis is whether the settlement agreement will be implemented and work in practice.

Much of this is out of the hands of the mediator. For example, an effective settlement may require support services from various government departments that they may not deliver well or at all. A small farming business established through the mediation will often require extension and mentoring.

Analysts and land reform agencies should be realistic about what mediation can deliver. Transfer of ownership, often the result of mediation, is certainly an important achievement. However, it is not sufficient to ensure that development will happen and that communities will benefit. Mediators must be mindful of this and try to facilitate a settlement agreement that takes account realistically of what actual and real support will be provided. However, land reform agencies carry a major responsibility to facilitate the provision of such support. Without it, land reform and land reform mediation are unlikely to make the contribution that they could potentially make.

Sufficient resources for land mediation

Few resources have been available for land reform in general. While land mediation in the context of land rights disputes has played a significant role in major land disputes, fewer than 300 disputes were mediated formally under the auspices of the NLRMAP. Few resources have been available for land mediations since the disestablishment of the NLRMAP. However, the largest gap in available resources has been for post-settlement facilitation, training, backup and mentoring services. This gap reduces the likelihood of a successful settlement, and also makes it more difficult for mediators to craft, with the help of the parties, a settlement that is sustainable.

Community representation and stakeholder involvement

At appropriate stages of the mediation process, different groups of stakeholders may need to participate. In other words, attempts to keep all stakeholders involved all the time often to lead to serious inefficiencies. Processes are expensive and must be focused. One can achieve this by facilitating the clustering or separation of issues, and ensuring that key stakeholders relevant to the given issue participate.

Moreover, where there are no elected community representatives, it may be appropriate for the mediator to facilitate the election of community representatives or a steering committee.

Where a community is represented by unelected people, e.g. a hereditary chief, the mediator should ideally facilitate the election of such a broader steering committee, of which the chief may be a member. Traditional structures may favour an elite within the community vis-à-vis other sections of the community. This is an area of much dispute in South Africa, because of conflict between traditional community structures and demands within communities for more democratic forms of government.

Nonetheless, there are usually good reasons for policy and mediation practice to ensure broader participation and representation. Where a dispute is settled to the satisfaction of an unrepresentative decision-making structure in the community, but not to the satisfaction of the people affected directly by the dispute, it is unlikely that the settlement will be effective in the longer run.

Even where representatives are elected, they may act as gatekeeper, keeping certain parts of the community (e.g. the poor) away. They may allow only certain members of the community to register as claimants. Settlement agreements should ideally address issues such as training and planning to avoid circumstances ripe for new conflict.

Addressing unbalanced power relations between stakeholders

Where unbalanced power relations between stakeholders are not addressed, it is likely that settlements will favour the more powerful. This will often undermine the efficacy of settlement agreements. As Pienaar of the LRC states, these inequalities can be addressed through: "issues from choice of language to choice of venue, to time for meetings, to suggesting quotas for representation, to proposals that secure rights in the name of a person (a single mother), instead of a vague notion such as a household."

Communities understanding their rights and obligations

Where people understand their rights and obligations, land mediation is more efficient and chances of a sustainable settlement are greater. This often requires community training sessions as part of, or associate with, mediation. In land mediation, such training must focus on the community's rights related to what it wants to do with the land.

Thus, once a community claiming or demanding land has clarified what it wants to do with the land, it would be clearer what kinds of rights education would be appropriate. General and detailed rights education unrelated to the issues important to the community is often wasteful and contributes little to the resolution of land disputes. Communities will often lose interest if the education is too technical or detailed.

Rights education should be provided to the correct people. Some people are more interested in, or inclined to focus on, detail. Such people should be identified and provided with more detailed training, as is appropriate regarding the community's needs and desires. The power relations within a community should always be borne in mind. Steps must be taken to ensure that the rights education is spread so as to include the disempowered sections of the community.

Training initiatives often focus on rights without due regard to obligations. Where a community's obligations in terms of the law or agreements are underemphasized, it is more likely that the community will not comply, leading to future disputes and failure to achieve land reform objectives. Similarly, people elected to represent a community need clear rules of conduct and understanding of their obligations as representatives. Without them, disputes within the community are likely to arise.

Provision of sufficient information

Key elements of information should be made available to the mediator by the government land reform agency and the parties. Ideally, the agency appointing the mediator should compile a briefing document setting out the main facts and issues and listing contact people and location of documents. It is not necessary to provide all documents in the matter, as this may not assist the mediator and will be time-consuming and costly.

A clear assessment of the issues in dispute

Sometimes, the parties themselves do not identify key underlying issues that need to be addressed. The mediator should keep a keen eye out for such issues and, where identified, facilitate agreement on them.

However, some issues that may be identified as relevant to the land dispute are actually peripheral. Where they are intractable and not key to the settlement of the land dispute, it may be appropriate to seek to avoid such issues or to provide for a separate process to address them.

Otherwise, the resolution of the land dispute may often be delayed for very long periods by issues not related to it. Parties may often insist on the resolution of such peripheral issues. This is because mediation services are generally not available and they want to use them to facilitate agreement on issues that will otherwise be difficult for them to resolve. Moreover, they may want to use the power they have regarding the land dispute to settle unrelated disputes.

Grouping of cases

Where there are cases from the same geographic area with similar facts and legal rights, and the same or similar parties, it may be appropriate to group the cases and mediate them together. This has been done successfully, for example, in claims of labour tenants to land of one farmer in the district of Wakkerstroom, Mpumalanga.

Provision of services that would support mediation

Legal and strategic advice: Communities and other parties involved in mediation or contemplating their options regarding land disputes need legal and strategic advice. One party, often the intended land reform beneficiary, may be in favour of mediation, while the other is not. What is their next move? Do they take legal action? What are their chances of winning and how much will it cost them? Staff in the land reform agency may not always be the persons best placed to answer these questions.

Facilitation: Sometimes a wider facilitation service is required, where a mediator may be appointed to seek out interested parties not present, obtain documents, and advise parties. This is usually where the circumstances are very complex and where the land reform agency that requested the mediation does not have sufficient capacity.

Valuation: Effective procedures need to be in place in order to address the issue of the compensation or price that should be paid to the present landowner or occupant to release the required land. Settlements often include purchasing land. This in turn requires an assessment of the value of the land, referred to as valuation. Settlement of disputes is often delayed by long valuation processes. The price of land is often an important aspect to be negotiated in land redistribution and other legal claims to land or redistribution programmes. Where this is the case, the following mechanisms should be available:

Parameters for mediation and monitoring

Mediation can be unproductive and wasteful of resources where clear parameters are not set by the land reform agency paying for such services. Parameters could include setting aside a specified number of days for the process.

Initially, there was very little monitoring of mediation processes by the land reform agencies. This led to unproductive mediations that continued for long periods. The mediator and parties themselves should be aware of the cost of mediation. Where the parties themselves do not pay the mediator, as has most often been the case in land reform mediation in South Africa, the mediator or the parties should be required to motivate to the agency any extension of the period set aside for mediation. The real threat of withdrawing payment for mediation services, where applied in appropriate circumstances, often focuses the minds of the parties (and the mediator) and facilitates the earlier completion of mediation processes.

Dispute prevention

Settlement agreements must be designed with a view to preventing or managing disputes that may flow from the agreement.

The process of negotiation and concluding an agreement may also have the effect of reducing the likelihood of disputes, or increasing it. Fewer disputes are likely to arise when people have been properly consulted, mandates sought properly, and solutions reached collectively after thorough discussion to ensure that the strategies identified are both appropriate and will carry people's support. Moreover, those that do arise should be of a less fundamental nature.

Communities gaining access to land through land claims or redistribution projects, or whose tenure is made more secure through tenure reform, need effective mechanisms of allocating rights and preventing and managing disputes. One of the greatest failings of land mediation and land reform agencies has been the lack of attention to putting in place effective systems for ensuring the allocation of rights within beneficiary communities.

Mechanisms to prevent disputes should be pursued consciously. These could include:

Backup and mentoring

Land reform mediation usually aims to achieve settlements that contribute to a better life for the parties, especially the relevant communities. It has proved difficult to achieve this aim unless land reform beneficiaries and other parties to land disputes receive effective backup from land reform agencies or other identified institutions. A common assumption is that all that is required is the facilitation or mediation of a "good" settlement agreement. This is far from the truth.

As indicated above, mechanisms to allocate rights within communities need urgent attention in order to prevent future disputes. This does not happen automatically. It must be planned for, with community training and outside backup. Where land is to be transferred, communities will be given access to new land or an amended system of tenure will be put in place. The allocation of rights to members of the communities should preferably happen prior to such changes. Unless this is done, self-help within the group usually results in inequitable distribution.


The various people involved in land mediation in South Africa, from land mediators to NGOs to government officials to Land Claims Court judges, have learned many lessons from the experience of the past few years. Unfortunately, very few of these lessons have been recorded and analysed.

In the last decade, in the period leading up to and following the transition to democratic rule in South Africa, very complex matters have been negotiated and mediated. They range from the negotiation of a democratic constitution after the demise of apartheid to the widespread use of mediation in labour, community, land and environmental disputes.

The experience shows that mediation can help seemingly intractable opponents find common ground or common solutions. It also shows that successful mediation can have long-lasting positive effects on the relationship of the key protagonists. However, there are also many examples of mediation leading to unhappy compromises that generate further disputes and sometimes a situation that is worse than before the mediation started.

This article has discussed some of the many lessons learned in nine years' experience of land mediation. These lessons can help to improve the use of mediation locally, but they could also assist others in planning and defining the appropriate use of land mediation in their countries.

Despite the difficulties regarding the use of land mediation, the lessons learned here have also contributed to the use of mediation in other areas in South Africa. Similar processes have since been built into legislation on owner - tenant disputes in urban housing and in environmental legislation (DEAT, 2001). For example, the 1999 Rental Housing Act provides for each province to establish a Rental Housing Tribunal, whose functions include the mediation of disputes regarding "a practice unreasonably prejudicing the rights or interests of a tenant or a landlord" (in Gauteng Province, 76 percent of disputes lodged in 2001/02 were resolved by mediation).


The information and analysis in this paper was gleaned from a wide range of sources including those referenced in the footnotes, those listed in the Bibliography (below), and the personal experience of the author. A very useful resource was the case studies, notes and articles contained in NLRMAP newsletters, which were in turn based on input from people such as R. Alcock, E. Apelgren, E. Davison, M. Euijen, D. Hanekom (former Minister of Land Affairs), J. Margolius, S. Norfolk, S. Raubenheimer, J. Seremane (the first Chief Land Claims Commissioner), and H. Smith. Information was also sourced from interviews and discussions with a wide range of people including: present and past staff members of the Department of Land Affairs and the Land Claims Commission including M. Aliber, L. Archery, R. Clacey, E. Hirschfeld, M. Kenyon, S. Motswege, S. Ramakarane, L. Steyn, H. Toolo, C. Williams and T. Yates; A. Dodson (former judge of the Land Claims Court); land mediators including R. Alcock and E. Molahlehi; K. Pienaar of Legal Resources Centre; S. Powers and H. May of Surplus People Project; K. Makan of Lawyers for Human Rights' Security of Farm Workers Project; and J. Beaumont of the Department of Environmental Affairs and Tourism.


Bosch, D. 1999. The Makuleke land claim settlement agreement. Deneys Reitz Attorneys.

Bosch D. & Hirschfeld E. 2002. The!Ae!Hai Heritage Park bundle - including the agreement whereby the land claims of the Khomani San community and the Mier community are finalized and associated documents. Pretoria, Commission on the Restitution of Land Rights.

Chennels, R. 1998. The San-Kalahari restitution claim. In: Workshop on land claims on conservation land. Pretoria, IUCN South Africa Country Office.

Department of Environmental Affairs and Tourism (DEAT). 2001. Fair decision-making and conflict management - NEMA and provisions for dealing with conflict.

Fraser, D. 2001. A strategic review of the Department of Land Affairs' conflict management system. TFC Development Consultants.

Kepe, T. & Cousins, B. 2002. Radical land reform is key to sustainable rural development in South Africa. Programme for Land and Agrarian Studies. University of the Western Cape, South Africa, Plaas Publications. (also available at

Makuleke, L. & Steenkamp, C. 1998. The Makuleke experience. In: Workshop on land claims on conservation land. Pretoria, IUCN South Africa Country Office.

Reid, H. 2001 Contractual national parks and the Makuleke community. Hum. Ecol., 29(2): 135 - 155.

[24] Latest version available at
[25] The figures quoted in this section were collected by Lala Steyn, a land reform and development consultant. See also Kepe and Cousins (2002).
[26] Clause 3.6,!Ae!Hai Kalahari Heritage Park Agreement of May 2002.
[27] Clause 43.1 read with Clause 38 of the above agreement.
[28] Clause 43.1.3 to 43.1.8 of the above agreement.

Previous Page Top of Page Next Page