L'Afrique du Sud dispose d'un éventail de régimes fonciers, à la fois formels et informels. La gestion des terres dans les zones «formelles» est extrêmement développée, et est fondée sur un registre central d'actes notariés et de systèmes de documentation foncière perfectionnés. Cependant, la plus grande partie de la population habitant dans ces zones demeure invisible aux yeux du système formel du fait qu'elle occupe des aires informelles au sein des zones urbaines ou du fait qu'elle réside sur des fermes commerciales qui appartiennent à d'autres. Dans les zones «informelles» ou à régime de propriété communal, réservées par le passé à des occupants noirs, la gestion des terres qui remonte à l'ère de l'apartheid s'est effondrée et une gamme de solutions impromptues l'a remplacée. Il reste qu'un système de gestion des terres unifié en Afrique du Sud doit s'élaborer. Dans l'ensemble, les départements gouvernementaux aux échelons national et provincial ainsi que le secteur privé peuvent accéder aisément aux données foncières formelles, mais une grande partie de l'administration locale dispose de connaissances limitées eu égard aux sources et doit lutter afin d'obtenir les données foncières les plus élémentaires pour les zones relevant de leur juridiction. Les données sur les zones à régime de propriété communal sont en général disséminées et le plus souvent de qualité médiocre.
Sudáfrica dispone de una serie de sistemas de titularidad territorial, tanto oficiales como no oficiales. La administración territorial en las zonas «oficiales» está muy desarrollada, y se basa en un registro central de escrituras y en complejos sistemas de información catastral. No obstante, la mayor parte de la población residente en estas zonas se mantiene invisible para el sistema oficial, ya que ocupa asentamientos no oficiales en zonas urbanas o réside en granjas comerciales propiedad de terceros. En las zonas de titularidad comunal o «no oficial», antiguamente reservadas a la población negra, la administración de las tierras implantada en la época del apartheid está a punto de derrumbarse sin que se hayan encontrado soluciones alternativas. En Sudáfrica todavía no existe un sistema unificado de administración territorial. Los departamentos gubernamentales nacionales y provinciales y el sector privado acceden fácilmente a los datos de titularidad oficiales, pero la mayor parte de los funcionarios del gobierno local tiene un escaso conocimiento de las fuentes disponibles y tropieza con dificultades para obtener los datos de titularidad básicos de sus respectivas zonas de jurisdicción. Los datos relativos a las zonas comunales están muy desperdigados y son de poca calidad.
E. Lahiff
Edward Lahiff, Programme for Land and Agrarian Studies,
University of the Western Cape, Cape Town, South Africa
South Africa has a range of land tenure Systems, both formal and informal. Land administration in the “formal” areas is highly developed, and is based on a central deeds registry and sophisticated cadastral information Systems. Much of the population residing within these areas, however, is invisible to the formal System, because it occupies informal settlements in urban areas or resides on commercial farms owned by others. In the areas of “informal” or communal tenure, formerly reserved for black occupation, apartheid-era land administration is in a state of collapse and a range of ad hoc solutions have taken its place. A unified land administration System for South Africa has yet to emerge. National and provincial government departments and the private sector generally have good access to formal tenure data, but much of local government has limited knowledge of the sources available and struggles to obtain the most basic tenure data for their areas of jurisdiction. Data on communal areas is widely scattered and generally of poor quality.
INTRODUCTION AND OVERVIEW
Land tenure, and land tenure data, in South Africa are a highly contested terrain, marked by an extreme duality. As a result of deliberate policies under colonialism and apartheid in the nineteenth and twentieth centuries, South Africa today is characterized by two broad tenure Systems -the System of private property based on formal (mainly freehold) title, and the System of communal tenure based largely on informal rights and customary practices. While there is considerable variation and even overlap within and between these Systems, and certain elements that do not fit neatly into either, this duality is, nonetheless, central to an understanding of land tenure issues in the country.
The System of private property based on freehold title - what can be loosely described as the formal System - applies to the great majority of land in South Africa, including both urban and rural areas. The technical and administrative Systems associated with it are comparable to those in the most developed countries of the world. State services, such as the deeds registry and the surveyors-general, are supported by private-sector professionals, including property lawyers, valuers, conveyancers, surveyors, town and regional planners and estate agents, as well as by a sophisticated body of law. A thriving land market exists across the residential, agricultural and commercial sectors, supported by a range of financial institutions providing mortgage bonds and other services. This formal System of property ownership and administration in turn underpins most of the economic activity of the country.
Despite the advanced nature of the formal property System, it continues to exclude a large proportion of the population, as a direct result of centuries of colonial and apartheid practices that aimed to segregate the population along racial lines and deny or limit the property rights of the majority black population. Prior to the advent of democracy in 1994, black people generally lived under one or another inferior form of property rights - as rental tenants in urban areas, as tenants-at-will on commercial farms, as subjects of tribal chiefs in rural communal areas or as squatters on land belonging either to the state or private owners. Only a tiny minority of the black population were able to acquire land rights comparable to the formal rights enjoyed by most of the white population.
Under apartheid, the System of administration of black communal areas was legally and administratively well defined, albeit racist, oppressive and manifestly inferior to that prevailing in the white freehold areas. With the transition to democracy (beginning around 1990), the communal System entered a state of general collapse and administrative and legal chaos. Key institutions such as the South African Development Trust and the former “homeland” (or Bantustan) governments were abolished, and not ail of their functions were taken over by other government departments. Departments of agriculture in the nine provinces of South Africa have undergone major rationalization, and no longer carry out many of the land administration functions that they did in the past. Magistrates' offices have, officially, ceased to have any role in land administration. New institutions, notably elected local government (municipalities), have been created, with considerable powers in areas such as planning and economic development, but without clarity in their role in either communal areas or on commercial farms, where most of the rural population resides. Unelected tribal authorities continue to operate within the communal areas but without any clear legal guidelines or official supervision.
With the repeal of much (but not ail) apartheid-era legislation on communal land, and the collapse of many of the institutions responsible for land administration, there has emerged a legal and administrative vacuum, greatly exacerbated by a lack of clear state policy on the reform of communal tenure. The government is constitutionally obligated to provide tenure security to ail citizens, but it has taken ten years to pass a law - the Communal Land Rights Act of 2004 - that would address issues of land ownership and administration in the communal areas. The law was not expected to be implemented before the end of 2005 and, in the meantime, interim measures and assorted regulations from the apartheid era were deployed in an ad hoc manner. The dominant characteristic of the land tenure System remains legal and administrative chaos.
This article, which is based on a review commissioned by FAO, approaches the question of land tenure and land tenure data from the perspective of policy-makers and officials responsible for land administration and for developmental issues. It does not deal in any depth with the day-to-day management of tenure data but instead focuses on broader questions of the availability, reliability and general utility of tenure data to policy-makers and implementers, and the implications for both land administration and broader development objectives. It is based on a review of available literature on land tenure matters in South Africa (including Web-based sources), supplemented by interviews with selected individuals with knowledge of tenure data and land administration issues. Research was carried out in South Africa in January and February 2005.
HISTORICAL BACKGROUND AND ORIGINS OF A DUALISTIC LAND SYSTEM
The arrival of Dutch, and later British, colonialists at the Cape in the mid-seventeenth century unleashed a process of systematic dispossession of indigenous people and destruction of African societies that continued up to the dawn of the democratic era. As the European settlement at the Cape grew, traders, hunters and farmers gradually advanced inland, greatly disrupting the lives of the Khoisan and later Xhosa and other people with whom they came in contact (Thompson, 1995). As the settlement expanded, settlers began to enforce military and political control over the people they encountered and the land they occupied. Dispossessed of most of their land, black people were drawn into a variety of subservient positions on farms and in towns, and later in mines and factories.
Freehold tenure - based on the System, then prevailing in the Netherlands - became established at the Cape, particularly for residential and arable land, from as early as the 1670s, along with a System of loan farms in the outlying areas (Carey Miller, 2000). This System of Roman-Dutch law was not displaced by English common law during the period of English colonial rule, and continues to define the law of property in South Africa up to the present. Under British rule at the Cape (from the early nineteenth century), loan farms in the outlying areas were gradually converted to perpetual quit-rent, and ail new land grants were made in quit-rent title. Farms were formally surveyed and registered in the deeds office. A similar System prevailed in the nineteenth century Boer republics of the Transvaal and the Orange Free State, where initial land grants to settlers were followed by formal surveying and registration of title. Carey Miller (2000) argues that formal titling was largely reserved for whites throughout the eighteenth and nineteenth centuries, and was practically unavailable to people of colour.
With the discovery of vast deposits of diamonds and gold in the late nineteenth century, strenuous efforts were made by settlers and the colonial powers to coerce black people to work in the mines for extremely low wages (Crush, Jeeves and Yudelman, 1991). A central part of this strategy was to curtail drastically black people's access to land, thereby forcing them into wage employment. Under the Natives Land Act of 1913 and the Native Trust and Land Act of 1936, just 13 percent of the land in the country was set aside for the black occupation, and black people were prohibited from owning land in the remaining 87 percent of “white South Africa” (Bundy, 1979; Wolpe, 1972). On the white-owned farms, where the majority of the black population actually lived, a range of measures were implemented to abolish the Systems of sharecropping and labour tenancy, under which black tenants had been able to gain access to plots for independent production (Keegan, 1986). Black tenant farmers were thus transformed into agricultural labour ers, subjected to harsh labour conditions and denied rights to land. Within the reserves - the 13 percent set aside for black occupation - residents were generally only permitted to occupy, rather than to own, the land, and relied heavily on migrant labour for their survival
Under the System of apartheid, from 1948 on, these reserves (later know as homelands or Bantustans) became synonymous with extreme poverty, underdevelopment and misrule (Lodge, 1983). In urban areas, the notorious Pass Laws and the Group Areas Acts led to the destruction of vibrant, multiracial communities such as District Six and Sophiatown, and the forced removal of millions of people to the Bantustans (Platzky and Walker, 1985). Church mission stations, which gave black people access to land within designated white areas, along with the few remaining black-owned farms, were systematically eliminated by the government of the day in pursuit of its racist policies.
LAND ADMINISTRATION
From one perspective, South Africa has a highly developed and efficient System of land administration. Over much of the country, land markets function effectively, property rights are recognized and protected, environmental regulations are enforced and land use functions within a sophisticated planning framework. Another perspective, however, suggests that fundamental issues of land management in the country remain unresolved, with particular challenges posed by the extension of the dominant “formal” System to the “informal” areas of communal land in the former homelands.
For Kingwill, there are “continuing misfits between the dominant land management frameworks in developing countries like South Africa, and the marginalized informal Systems that serve the poor in both urban and rural landscapes” (Kingwill, 2004, p. 1). This lack of a unified System of land administration is highlighted in reports on land administration in the Eastern Cape by the Public Service Commission, which found no shared understanding on how the land administration function in the province should be performed, “multiple and parallel Systems” of land administration, and a “vacuum” in the regulatory framework (Public Service Commission, 2000, 1.1; and 2003, p. x).
Similar concerns are raised by Cross, who highlights risks of overly complex land administration Systems and lack of clarity about responsibilities and procedures, especially with regard to the provision of housing for the urban and rural poor in South Africa:
Under the complicated arrangement of delegated land powers that prevails, actual tenure control on the ground is sometimes weak and administrators who are frustrated, unsure of their authority and sometimes inclined to push the limits are handling land matters at many different levels. In this grey area, fly-by-night developers and pirate land allocators in tribal and informal areas have been establishing room to operate. Their activities detract from tenure security for an unknown number of disadvantaged landholders with weak tenure security or rights, which are difficult to document.
(Cross, 2002, p. 199)
Current policy reforms are based largely on the assumption that the existing formal System can, with only minor adjustment, meet the requirements of less formal and communal Systems, an assumption that is challenged by many authorities in the field. Carey Miller (2000) argues that the historic importance of registration is continued in the reform era, as government seeks to replace lesser, permit-based rights with rights of ownership. Kingwill argues that official thinking on reform of land policy since 1994 has been based on the erroneous assumption that the formal System can be expanded to incorporate the informal, without itself undergoing radical change: “Emerging land policies aim to ‘suck’ the informal Systems into the formal Systems. There is a need for more rigorous analysis of why these policies continue to fail.” (Kingwill, 2004, p. 1).
FORMAL SYSTEM OF LAND TITLING IN SOUTH AFRICA
The formal land titling and registration System in South Africa is based on the deeds registries, as regulated by the Deeds Registries Act of 1937. The System is based on the transfer of ownership of land by public registration, rather than by private conveyance (Carey Miller, 2000). While the common law ownership of land includes the ownership of ail fixed improvements erected on the land, South African law also recognizes separate ownership of buildings or parts of a building. Such ownership is regulated by the Sectional Titles Act of 1986.
Both the Deeds Registries Act and the Sectional Titles Act of 1986 provide that deeds and supporting documents shall be prepared and lodged in the deeds registries by a conveyance or public notary. Under the South African System, title to land and other real rights are not guaranteed by law. Rather, security of title is the result of the respective responsibilities carried by professional land surveyors (under the authority of a surveyor-general), the deeds registries established throughout South Africa and an independent attorneys' profession. The System of registration that operates in South Africa has been typically described as highly efficient and reliable (Pienaar, 2000).
The deeds registration System provides four formal functions - registration of deeds, maintenance of a public land register, preservation of registration records and provision of registration information to the public. A total of 1.2 million transactions of all sorts were carried out in all deeds registries in the financial year 2003–2004, including 399 138 transfers, 251 104 bonds related to freehold titles and over 20 000 leasehold transactions (Department of Land Affairs, 2004).
A range of additional services essential to the functioning of the land registry System are performed by the national Department of Land Affairs, of which the most important are survey and mapping, and cadastral surveys, as regulated by the Land Survey Act of 1997. As with the deeds registries, the System of cadastral surveying in South Africa is generally held in high regard. Information pertaining to the Cadastral Information System has recently been incorporated into a computerized map of ail seven million registered land parcels and ail administrative boundaries in the country. Other surveyed real rights, such as servitudes and leases, are also included in this map. According to the office of the Chief Surveyor-General, this map is in great demand and is used for a wide variety of spatial planning purposes, in the fields of housing, health, education, water, electricity and post and telecommunications.
The periodic agricultural census reports on the number of farming units in both the commercial sector and family farming, but does not provide specific detail on land tenure. Aggregate data are provided on the nature of the land-holding entity (e.g. individual, private company or government enterprise), but not on the actual form of land tenure (National Department of Agriculture, 2002). Elsewhere, the census differentiates between workers and proprietors/tenants employed on farms, but this latter category is not disaggregated and again refers more to the position of individuals within the farming enterprise rather than to the nature of the land tenure.
Public institutions generally have good access to tenure data related to privately owned land, particularly with regard to boundaries, extent, ownership, ownership history, mortgages and leases. This information is available via the deeds registry, but also from a variety of data Systems maintained by government departments and a range of private-sector service providers that interact closely with the offices of the Chief Registrar of Deeds and the Chief Surveyor-General's Office.
While the formal System described here has many strengths, and serves many of the planning and developmental needs of the country, it also has severe limitations, especially when it comes to meeting the needs of the majority of the poor and marginalized. The first major limitation is that the formal System does not extend much beyond the 87 percent of territory that formerly constituted “white” South Africa. Insofar as formal tenure data exist for the former homelands, they tend to refer only to nominal ownership of substantial parcels (e.g. the outer boundary of land held in trust by the state on behalf of communities) and not to individual land holdings or, as in the case of much quit-rent title in the Eastern Cape, have not been maintained and are therefore long out of date.
Second, the formal System does not capture the detail of landholding and land use in situations where large numbers of people are not the owners of the land they occupy. This applies to millions of people residing in so-called informal settlements in and around urban centres, as well as millions of farm dwellers residing on commercial farms. Urban informal settlements may be found on land registered in the name of private individuals or corporations, or on state-owned land and, in certain areas, such as greater Durban and built-up areas within the former homelands, they may be on communal land (Royston, 2002). In Soweto (the largest township in South Africa), where freehold title was introduced in the early 1990s, more than 15 percent of recent transfers have not been registered - a pattern repeated throughout the country (Fourie and van Gysen, 1996).
The commercial farming zone accounts for approximately 65 percent of the territory of South Africa and is home to an estimated three million farm dwellers, of whom less than one-third are employed on the farms (Hall, 2004). Under apartheid, farm workers (and farm dwellers more generally) were, by law, tied into a highly subservient relationship with the white landowners, severely restricting their rights to change jobs or move off a farm, or to organize for better working or living conditions. Landowners generally provided rudimentary services for their farm dwellers, often with the help of subsidies from the state, although in many cases farm dwellers built their own houses. In the early decades of the twentieth century, many farm dwellers were provided with land to grow their own food crops or graze livestock, while others entered into a variety of tenancy arrangements, including cash tenancy, sharecropping and labour tenancy. Over the course of the twentieth century, most farm dwellers were deprived of access to agricultural land, leaving them with only basic accommodation and possibly a small garden plot. Farm dwellers are protected by law from arbitrary eviction, but evictions of farm dwellers remain common, and little has been done by the state to promote long-term security of tenure for farm dwellers on the farms they occupy (Hall, 2003).
The content of farm dwellers' rights - how long they have lived on a farm, what terms have been agreed with the landowner, the boundaries of land allocated for their use, and other customs and practices established over generations - is typically known only to them and to the landowner. Where written agreements exist between owners and dwellers, these typically relate only to matters of employment, with a general reference to housing or other benefits. Land rights per se axe almost never recorded in writing. The great inequality between farm owners and farm dwellers - both in terms of the enforceability of their land rights and their wider socio-economic status - determines the views and responses of outsiders, including government agencies. The national Department of Land Affairs and provincial departments of agriculture maintain no registers of farm dwellers or their rights, and typically have no information on their living conditions or even the pattern of evictions in their areas of operation. Local government, which is responsible for the provision of basic services to ail citizens, generally treats private commercial farms as no-go areas and excludes farm dwellers from many of its activities. In this context, the general absence of land tenure data with respect to farm dwellers is both a symptom and a contributing factor to the ongoing marginalization of this socially vulnerable but numerically large social group.
A third category of land where the formal tenure System tends to break down is state-owned land. It is estimated that the state (across ail spheres of government and state-owned companies) owns approximately 24 million hectares, or 19.8 percent of the national territory (as at 31 December 2002) (Department of Land Affairs, 2003). Just over half of the total is controlled by the Department of Land Affairs, with the rest falling under a range of state agencies, including the Department of Public Works, national and provincial parks, the Department of Water Affairs and Forestry and the defense force. Sizeable areas of state land are held in trust on behalf of various communities, which occupy it under various forms of communal tenure. While information on specific portions of state land is generally available via the deeds registry, most is held in the name of either the Minister of Land Affairs or the Minister of Public Works, with little or no detail on the actual use of such land (e.g. for a school, hospital or military barracks). Needless to say, details of informal occupiers of state land are generally not recorded in any source. Further problems arise from the fact that much state land in the former homelands remains registered in the names of old homeland governments, ten years after their abolition. Repeated efforts have been made, by the Departments of Public Works and Land Affairs, to compile a detailed and complete inventory of state properties, with mixed results. Within the Department of Land Affairs, this function was carried out by two directorates, Public Land Inventory and State Land Management, which were merged in 2002 into the Directorate: Public Land Support Services.
Provincial and local government officials frequently express frustration about the lack of detailed information about state land in their jurisdictions and about the lengthy processes involved in acquiring state land for specific purposes (a process much complicated by the lack of certainty regarding responsibilities of various stakeholders with respect to state land). Some municipalities do not even have information available on land that is under their direct ownership (Hall, Isaacs and Saruchera, 2004). There is a need for better access to information on state land, particularly at the level of municipalities, to allow for more effective spatial planning and developmental activities.
COMMUNAL TENURE
The black rural areas of South Africa are marked by a wide variety of tenure forms, shaped by varying degrees of interaction among communitarian Systems of indigenous origin, modem Systems of individual ownership and state-imposed Systems designed to maintain social and economic control (Lahiff, 2001). This results in very different forms of tenure, which can diverge widely from what their nominal forms suggest and from the regulations laid down by the state (Cross, 1991).
Major differences are observed within communal areas, based on factors such as the history of acquisition, population pressure and proximity to urban areas. Land under the jurisdiction of tribal authorities, for example, may comprise old “reserves” (pre-1913 lands), tribal farms (lands purchased by tribes before 1936), private lands (lands purchased by individuals or groups prior to 1936) and “trust land” (lands purchased by the South African Native Trust/South African Development Trust after 1936). Important differences in tenure arrangements are also to be found among lands used for different purposes, such as residential land, arable fields, grazing land (typically used also for harvest of natural materials such as firewood, thatching grass and medicinal plants) and business sites.
The majority of land in ail the former homelands is registered in the name of the Government of South Africa, or some proxy of the government, such as one of the former homeland governments, the Ingonyama Trust (in KwaZulu-Natal) or the South African Development Trust. This can, technically, be referred to as state-owned land, albeit of a special kind, because much of it has been officially granted for the exclusive use of particular tribes or other groups. Considerable areas of state land, however, have not been formally allocated to any group, and can properly be referred to as state land, although much of it has been informally occupied by communities. In the former Coloured Rural Reserves, concentrated in the Northern and Western Cape, land is under the control of local municipalities, but effectively remains state (or public) land.
Other categories of rural land occupied by black communities include land purchased (or transferred by the state in terms of a deed of grant) and held in freehold by individuals, by groups of individuals, by corporate entities (such as tribes) and by church bodies (Haines and Cross, 1988). This can, technically, be referred to as privately owned land, whether individually or collectively owned, in that the state has no legal standing as owner. Nonetheless, much of the “private” nature of such land has been compromised by failure to transfer land properly between generations, influx of non-owners and, in some cases, abandonment of the land by church bodies. An important intermediate category, that combines elements of state and private ownership, is land purchased by tribes or other groups but registered in the name of the government or one of its proxies. This special arrangement arose as a consequence of the 1913 Natives Land Act, which prohibited blacks from acquiring land in their own name outside the scheduled areas. Such land cannot be described as state land because the state did not purchase the land, but acted as an intermediary on behalf of what were effectively private purchasers. It can, therefore, be described as a limited form of private ownership, where owners were denied full rights of private ownership. Finally, in the Eastern Cape, especially in the former Ciskei, sizeable areas are held in quit-rent, a limited form of private ownership. Whereas quit-rent would usually be seen as a relatively secure form of individual ownership, in the black areas it has been compromised by repeated interference and regulation by the state and by social pressures that have imposed a degree of communalism on much quitrent land (Haines and Cross, 1988; Kingwill, 1993).
The formal ownership of much land in black areas is thus both confused and contested, the result of efforts by a colonial state, over a century or more, to limit and control the rights of black people. While certain decisions about land allocation and adjudication of land rights take place within local institutions - that is, internal to the System of communal land - a range of land administration and land management functions are carried out by institutions (local, provincial and national government agencies) that straddle formally titled areas and communal land. Dealing with this duality is an enormous challenge for the institutions involved, given the maximally divergent practices, quality of information and even laws that apply in each.
In a study of the Eastern Cape, the South African Public Service Commission (2003) described the challenge facing administrators in the following terms:
… the administration of land in the [Eastern Cape] is not in terms of a uniform legal framework. There are special legislative measures for “surveyed/unsurveyed” areas, urban/rural areas, communal and freehold-held land and so on. Even through the 1936 LandAct itself has been repealed, none of its retrogressive regulations have been done away with. The total impact of this is confusion, frustration and uncertainty at the level of administrators, who are under great pressure to deliver in terms of demands of the new dispensation albeit within an archaic legal framework. This kind of contradiction simply cannot co-exist with efficient and effective service delivery efforts.
(Public Service Commission, 2003, 1.3)
Needless to say, this confusion has direct, negative consequences for land administration:
This situation has led to poor land and service delivery in the former homeland rural areas and to the prevalence of informal and sometimes grossly unaccountable actions in the rural areas of the former homelands, since land administration in those areas is no longer contained within any organisational framework in the provincial government or any functioning legal framework.
(Public Service Commission, 2003, p. xi).
Such a lack of legal and administrative clarity not only leads to inefficiency and creates an obstacle to development, but it also provides opportunities for local elites to advance their own position at the expense of others (see Claasens, 2001).
The mix of formal and informal tenure types, generally viewed by officials as unequal, has a direct effect on where development takes place and for whose benefit. In many municipal areas, development is concentrated in freehold areas or on state land, to the detriment of areas under communal tenure. Nevertheless, a certain amount of development does take place on communal land, but usually requires extended negotiation with local stakeholders and ad hoc solutions. In the Eastern Cape, for example, extensive use has been made of the Interim Protection of Informal Land Rights Act to allow for the transfer of portions of land from communal use to various government departments for a variety of purposes. This process is usually managed by the district offices of the Department of Land Affairs. Elsewhere in the country, another law, the Upgrading of Land Tenure Rights Act, has been used intermittently, mainly by individuals wishing to obtain freehold rights in communal areas prior to establishing a shop or other business, but also for housing schemes (Claasens, 2001; Lahiff and Aphane, 2000). This results in a pattern whereby isolated parcels of land are surveyed, registered and held in freehold, surrounded by large areas of unsurveyed land held communally. Data on such parcels are generally available within local municipalities and other institutions, but data on surrounding areas are generally absent.
Unlike more formal Systems of tenure, communal land rights do not depend on an objective, verifiable document or other artefact, such as a title deed. Communal land rights are often effectively personal rights, only loosely connected to specific parcels of land. Land parcels may be demarcated on the ground, but are not necessarily measured, and neither the exact location, nor the boundary, nor the extent of the plot may be recorded in any other form. For communal land, the outer boundaries and major subdivisions (e.g. tribal authority boundaries) are generally known to officials and others from paper-based maps and, in some cases, electronic Systems. The major weaknesses, however, are the lack of reliable data at the level of individual land parcels (i.e. residential and agricultural plots), lack of clarity (and often parallel process) in the allocation of rights and uncertainty over the status of land rights.
A variety of methods are used to record land rights in communal areas, although many are of dubious legal or practical value. The most common written record is the Permission to Occupy certificate issued under the Bantu Areas Land Regulations (Proclamation R188 of 1969), which drew its legal authority from the 1936 Native Trust and Land Act. Such permits usually indicate a right to a residential stand, as well as, possibly, the right to a specific amount of arable land and the right to keep a specific number of livestock. In many areas, however, the precise location of the land is not recorded, and land held under such permits is rarely surveyed at the level of the individual holding. A variety of other devices, both written and verbal, are also referred to as “permission to occupy” (PTO) and, like the R188 permits, effectively grant a lifetime, inheritable right to land in a particular area. PTOs and similar devices are generally seen as relatively secure forms of tenure, although their legal standing is not always clear. Their most obvious limitation lies in the area of transfer rights, as holders do not generally have the right to sell their interest in the property and permits are not accepted as collateral by commercial lenders.
Apart from records that were created at the time of allocation of plots, a range of measures have been employed to capture data (and even consolidate rights) retrospectively. Up to 1994, for example, agricultural officials sometimes surveyed arable plots, a process that captured certain information about the land users and could serve to formalize an otherwise informal land right. This was commonly done in the case of orchards, irrigation works or other “improvements”. Similarly, interventions by veterinary officials, typically concerned with control of disease or overgrazing, could result in the compilation of lists of livestock owners and the issuing of a variety of stock cards (grazing permits). In the absence of more systematic tenure information, data generated through a local agricultural survey or census of stock owners often served as the only available source of information on land rights and land usage, and a PTO certificate or a stock card, no matter how cursory the data, could become an important means for land users to assert their rights to either government officials or their neighbours.
The PTO System continues to characterize land administration in most of the communal areas, as a result of the continued reference to old PTO certificates as proof of land rights and, in many cases, the issuing of new and legally dubious PTO certificates by a range of actors. While older (i.e. pre-1994) PTOs may continue to have some legal validity (and are widely accepted by occupiers and officials alike as proof of land rights), newer documents issued by municipalities, agricultural officials or magistrate's offices would appear to have no legal basis, and are in some cases being issued in direct contravention of instructions from the relevant government department. According to the Public Service Commission (2000), officials of the magistrate's offices who lost their competence to perform certain tenure functions in the mid-1990s simply continue to perform what they regard as an essential service.
Land administration processes in communal areas are generally relatively informal and rooted in local customs, although heavily influenced by apartheid-era procedures and overlaid with a sometimes dubious officiousness. Outside formal townships, residential or arable plots are rarely formally surveyed, but may be marked with stones or fenced. Preference in land allocation is given to recognized members of respective communities, who may be required to pay a small fee to local leaders. In some cases, occupiers are issued with documentary proof, but such documents tend to be short on detail and of dubious legal status.
A recurring feature is the inferior position of women within virtually ail customary land Systems in South Africa. Despite the provisions of the South African Constitution and legislation on gender equality, women are routinely discriminated against in the allocation of land, in participation in communal meetings and in inheritance matters (Alcock and Hornby, 2004; Cross and Friedman, 1997; Small, 1997; Archer and Meer, 1997; Mann, 2000).
A minority of so-called communal land is held in private title, as freehold or quit-rent, which is officially surveyed and registered in the deeds office. However, a variety of factors serve to devalue such records. First, owing to racist laws, many de facto owners were recorded as beneficiaries of a variety of trusts, rather than as owners in their own right. Second, the failure to transfer title to heirs means that formal ownership of much private land is now far from clear. Third, as discussed above, the influx of population to black areas, much of it engineered by the state, has severely compromised the power of owners to exercise their rights, rendering many title deeds virtually worthless.
The principal findings of this examination of land administration and land tenure in South Africa are that, 11 years after the ending of apartheid, land administration remains fragmented and uneven, and that a sizable proportion (generally the poorest section) of the population remains outside the effective scope of the formal System. This has direct implications for the quality of tenure information and for the ability of public institutions to plan and implement developmental projects.
Over most of the national territory, the System of individual private property predominates, supported by an impressive array of state and private-sector services comparable to those found in the most developed countries. A central deeds registry and associated cadastral information service provide high-quality, detailed and up-to-date information in a variety of formats to owners, developers, planners and others, serving as the basis for a wide range of commercial and public administration activities.
In formally established townships, processes are under way to convert leasehold rights (not ail of which are formally registered) to freehold through the issuing of deeds of grant. These areas are well on the way to individual freehold, and a number of innovative approaches are being tried with regard to surveying and titling. As with the upgrading of urban informal settlements, these township reforms can, with adjustments, be accommodated within the existing formal System. However, the limited resources of municipalities and other role-players often mean that such processes do not proceed at the pace, or with the degree of popular participation or awareness, that might be desired. Financial institutions have been widely criticized for failing to provide mortgages in many localities, because of negative perceptions of the creditworthiness of poor people and of the practical difficulties presented for repossession and resale of property. A related factor is the severely limited secondary market in many residential areas, and reports of government-subsidized houses changing hands for considerably less than the cost of construction. Where properties do change hands, it is often on the basis of informal transactions that are not registered in the deeds registry. This raises the possibility that, even where reforms have taken place and people and property are drawn into the formal System, they may revert to informality over time.
A different prospect faces those living on privately owned commercial farms and under communal tenure arrangements in rural areas (mainly in the former homelands). In the case of farm dwellers, their use and occupation rights, although protected by general law, are outweighed in practice by the well-established rights enjoyed by farm owners. Where the rights of these two parties clash, the rights of the owner virtually always prevail, leading to the eviction of the farm dwellers. These evictions are routinely upheld by the courts. Even in the relatively few cases where farm dwellers have taken court action to defend their rights under the law, this has, at best, led to the preservation of the status quo, rather than to any long-term security of tenure. The one option available under current policy - expropriation by the state and transfer of ownership to the occupier - has been widely ignored, no doubt because of the perceived extreme nature of an intervention that reverses the established order.
Land within the communal areas, with the exception of residential stands and some arable fields, has not been subdivided, surveyed or demarcated, so the cadastral basis for the incorporation of individual rights into the formal System does not exist. Moreover, even where residential or other parcels have been demarcated, these are not seen as the exclusive property of the occupiers. Nominal landholders may be required to share resources with other users, and may be limited by local rules as to how they use or dispose of the land.
Under both these circumstances, the poor quality of tenure data available to outsiders is directly related to the weak or uncertain nature of the underlying land rights. Similarly, the lack of progress with tenure reform in these areas can be related to objective difficulties in fitting such rights into the existing formal System on the one hand, and the failure to develop alternative forms of land administration appropriate to the prevailing conditions on the other. Although further tenure reforms are in the pipeline, the current situation clearly gives users with relatively weak rights little option but to find a way of securing their rights within the formal System, or go without.
In areas where freehold predominates, public bodies are generally capable of accessing good quality tenure data sufficient to carry out their duties with regard to land administration, environmental regulation and development projects, with the notable exceptions of informal settlements and farm dwellers. Although wide disparities are found between government agencies in terms of their understanding of tenure issues and awareness of existing tenure data Systems, it is clear that many are using advanced technology to access public tenure databases (such as the deeds registry) and to maintain their own geographical information Systems.
Two broad exceptions to this pattern were found. The first relates to state land, about which there appears to be considerable confusion, including difficulties in ascertaining which state body has responsibility for particular portions of land. Even where such information is available (and considerable progress has been made in recent years to make such information available through the public land audit), public bodies such as municipalities appear to experience difficulties in accessing state land for developmental purposes.
The second exception to this generally positive picture is municipalities in rural areas, especially in areas such as the former homelands where there is not a strong tradition of local government. Recent surveys have shown that many local municipalities lack basic information about tenure matters within their area of jurisdiction, are unaware of the advanced information services available to them from sources such as the Deeds Office and have a poor understanding of policy with regard to marginalized groups (such as farm dwellers and occupiers of communal land).
Land policy in South Africa is in a phase of rapid evolution and many areas are in considerable flux. On the basis of this study, the following recommendations for policy reform are offered:
New forms of registration should be developed for rights holders in communal areas and for farm dwellers. Whereas discussions are being conducted about the former, the latter has not yet found its way onto the policy agenda.
Interim measures should be developed for recording less formal rights in communal areas, based on existing PTO or similar records, prior to the upgrading of such rights. In addition, the roles of various stakeholders - particularly local municipalities and departments of agriculture - within such interim measures need to be clarified.
Policies should be developed to ensure adequate access to and understanding of deeds registry and cadastral information within local government, including the provision of training to officials and councillors in the use of tenure data.
Further reforms are required to ensure that reliable information on state land is available to ail stakeholders, but especially to local government. Improved processes for the release of state land for developmental purposes should also be considered.
There is a pressing need for the development of a comprehensive database on current and planned land reform projects, to be maintained by the Department of Land Affairs and made accessible to ail stakeholders.
REFERENCES
Alcock, R. & Hornby, D. 2004. Traditional land matters: a look into land administration in tribal areas in KwaZulu-Natal. Pietermaritzburg, South Africa, Legal Entity Assessment Project.
Archer, F. & Meer, S. 1997. Women, tenure and land reform: the case of Namaqualand's Reserves. In S. Meer, ed. Women, land and authority: perspectives from South Africa. Cape Town, David Philip and Oxford, UK, Oxfam.
Bundy, C. 1979. The rise and fall of the South African peasantry. London, Heinemann.
Carey Miller, D.L. with Pope, A. 2000. Land title in South Africa. Kenwyn, South Africa, Juta.
Claasens, A. 2001. It is not easy to challenge a chief: lessons from Rakgwadi. Research Report No. 9. Cape Town, Programme for Land and Agrarian Studies, University of the Western Cape.
Cross, C. 1991. Informal tenures against the State: landholding Systems in African rural areas. In M. de Klerk ed. A harvest of discontent: the land question in South Africa. Cape Town, Institute for Democracy in South Africa.
Cross, C. 2002. Why the urban poor cannot secure tenure: South African tenure policy under pressure. In A. Durand-Lasserve & L. Royston, eds. Holding their ground: secure land tenure for the urban poor in developing countries. London, Earthscan.
Cross, C. & Friedman, M. 1997. Women and tenure: marginality and the left-hand power. In S. Meer, ed. Women, land and authority: perspectives from South Africa. Cape Town, David Philip and Oxford, UK, Oxfam.
Crush, J., Jeeves, A. & Yudelman, D. 1991. South Africa's labour empire: a history of black migrancy to the gold mines. Oxford, UK, Westview Press.
Department of Land Affairs (South Africa). 2003, Extent of state land in the Republic of South Africa. Document supplied by the Department of Land Affairs, Directorate: Public Land Support Services, Pretoria.
Department of Land Affairs (South Africa). 2004. Deeds Registration Annual Report 2003/2004 (available at http://land.pwv.gov.za/land_ planning_info/deeds_registration.htm).
Fourie, C. & van Gysen, H. 1996. Land management and local level registries. South African Journal of Surveying and Mapping, 23, 6(142): 353–359 (also available at http://users. iafrica. com/a/au /augusart /online_vanguysen. html).
Haines, R. & Cross, C.R. 1988. An historical overview of land policy and tenure in South Africa's black areas. In C. Cross & R. Haines, eds. Towards freehold: options for land and development in South Africa's black rural areas. Cape Town, Juta.
Hall, R. 2003. Farm tenure. Evaluating Land and Agrarian Reform in South Africa series, No. 3. Cape Town, Programme for Land and Agrarian Studies, University of the Western Cape.
Hall, R. 2004. Land and agrarian reform in South Africa: a status report 2004. Research Report No. 20. Cape Town, Programme for Land and Agrarian Studies, University of the Western Cape.
Hall, R., Isaacs, M. & Saruchera, M. 2004. Land and agrarian reform in integrated development plans: case studies from selected district and local municipalities. Unpublished paper. Programme for Land and Agrarian Studies, University of the Western Cape. Cape Town.
Keegan, T.J. 1986. Rural transformations in industrialising South Africa: the southern Highveld to 1914. Johannesburg, South Africa, Ravan Press.
Kingwill, R. 1993. Quitrent tenure. Paper prepared for the National Land Committee, Johannesburg, South Africa.
Kingwill, R. 2004. Square pegs for round holes? Sortie questions relating to the emerging Land Management Frameworks in South Africa. Paper presented at the symposium on “Aligning development planning with communal tenure arrangements in the context of changing legislation in South Africa”. Durban, South Africa, 9 September 2004.
Lahiff, E. 2001. Communal land tenure in South Africa: a comparative analysis. Johannesburg, South Africa, National Land Committee.
Lahiff, E. & Aphane, J. 2000. Communal land tenure: a case study ofDikgale Tribal Area, Northern Province. Unpublished paper. Pietersburg, South Africa, Nkuzi Development Association.
Lodge, T. 1983. Black politics in South Africa since 1945. Harlow, UK, Longman.
Mann, M. 2000. Women's access to land in the former bantustans: constitutional conflict, customary law, democratisation and the role of the state. Land Reform and Agrarian Change in Southern Africa Occasional Paper No. 15. Cape Town, Programme for Land and Agrarian Studies, University of the Western Cape.
National Department of Agriculture. 2002. Census of commercial agriculture. Report No. 11-02-01. Pretoria.
Pienaar, G. 2000. The registration of fragmented use-rights as a development tool in rural areas. Paper presented to the conference on “Constitution and Law IV: Developments in the contemporary constitutional state”. Potchestroom University, South Africa, 2–3 November 2000 (available at http://www. kas.org.za/Publications/Seminar Reports/Constitution & LawIII/Constitution & LawIII.asp.)
Platzky, L. & Walker, C. 1985. The surplus people: forced removals in South Africa. For the Surplus People Project. Johannesburg, South Africa, Ravan Press.
Public Service Commission. 2000. Investigation into land administration, geographic information Systems and fleet management in the Eastern Cape. Pretoria.
Public Service Commission. 2003. Evaluation of land administration in the Eastern Cape. Pretoria.
Royston, L. 2002. Security of urban tenure in South Africa: overview of policy and practice. In A. Durand-Lasserve & L. Royston, eds. Holding their ground: secure land tenure for the urban poor in developing countries, pp. 165–181. London, Earthscan.
Small, J. 1997. Women's land rights: a case study from the Northern Transvaal. In S. Meer, ed. Women, land and authority: perspectives from South Africa. Cape Town, David Philip and Oxford, UK, Oxfam.
Thompson, L. 1995. A history of South Africa. New Haven, USA, Yale University Press.
Wolpe, H. 1972. Capitalism and cheap labour -power in South Africa: from segregation to apartheid. Economy and Society, 1(4): 425–56.
