The South African constitution, agreed and entrenched in 1996, promised that security of tenure would be enhanced for all landholders.
Section 25 (6) and (9) of the Constitution state:
25 (6) A person or community whose tenure of land is Iegally insecure as a result of past racially discriminatory laws or practices is entitled, to the extent provided by an Act of Parliament, either to tenure which is legally secure or to comparable redress.
25(9) Parliament must enact the legislation referred to in subsection (6).
Following the transition to democracy, there were high expectations that all South Africans would have reasonable access to secure tenure through new laws and institutions. While the government passed some relevant legislation, it was not backed up by effective institutional and administrative arrangements. The legislation has only been sporadically applied, often as result of court orders, but this has not translated into the envisaged transformation of land rights. Unregistered rights are inadequately protected, managed and regulated. These landholders are vulnerable to encroachments by private corporations, state authorities and local accumulators who are able to bypass the limited legislative protection offered to individual family rights. The state has thus failed to give full effect to the Constitutional injunctions for land tenure security for all South Africans.
During the 1990s, legislation was passed extending some protection to those who occupied land with diverse forms of potentially vulnerable off-register tenure rights. The aim was to provide a legal defence for them against arbitrary dispossession. The legislation did not clarify their rights in a positive way but it began to provide a rights-based legal framework that could satisfy Constitutional provisions.
The approach was to enact a series of laws that applied to defined categories of tenure contexts and relationships. Collectively the measures were designed to cover the country 'wall to wall'. Thus vulnerable rights on farms, in former communal and customary areas and in informal settlements are covered by the Land Reform (Labour Tenants) Act (LTA), Act No 3 of 1996; the Interim Protection of Informal Land Rights Act (IPILRA), Act 31 of 1996; the Extension of Security of Tenure Act (ESTA), Act No 62 of 1997; the Prevention of Illegal Eviction from and Unlawful Occupation of Land Act (PIE), Act No 19 of 1998. The Communal Property Association Act, Act No 28 of 1996, was regarded as new group ownership vehicle for restituted or redistributed land.
However, property law is premised on a rigid hierarchy of rights which privileges individual, titled ownership registered at the Deeds Registry. These rights are backed up by rigorous systems of land administration. The various pieces of legislation enacted in a bid to strengthen security of tenure of off-register rights have no comparable governance framework for effective administration thereof. (I will use the term off-register to cover all other rights that are not registered, although they are not all the same.) This undermines implementation of existing laws and renders land holders with off-register land rights, an estimated 60 per cent of the population, vulnerable to dispossession. Their position has been exacerbated by the dismantling of apartheid era land administration systems without putting functional alternatives in their place.
Table 1: South Africans holding land or dwellings outside the formal property system in 2011
Location Number of people Percentage of SA population (total of 51,8 million)
Communal areas 17 million 32,8
Farm workers and dwellers 2 million 3,9
Informal settlements 3,3 million 6,3
Backyard shacks 1,9 million 3,8
Inner city buildings 200 000 0,38
RDP houses¾no titles 5 million 9,6
RDP houses¾titles inaccurate/outdated 1,5 million 3,0
Total 30,72 million 59,7
Source: Hornby, D., Royston, L. Kingwill, R., & Cousins, B. 2017
The Constitution anticipated that rights in communal, customary and so-called informal contexts would evolve into stronger forms of legal recognition beyond mere protection in line with rights of citizenship. This would not only prevent arbitrary dispossession, but facilitate investment and set up accountable administrative processes. The Communal Land Rights Act of 2004 attempted to move beyond the defensive or protective legislation noted above in order to specify these forms of landholding in a more positive way. However, the approach adopted by the government entangled land rights with local-level authority structures in communal areas, thereby elevating the role of traditional authorities and undermining the autonomy of individual families. This Act was not implemented and the Constitutional Court decided in 2010 to invalidate the Act, but on procedural rather than substantial grounds.
Moreover, there is still no overarching, national land administration framework for such landholdings. Most land in the former homelands was held through a system of PTOs (Permission to Occupy certificates) established through a series of proclamations from the early twentieth century in different parts of the country, and largely retained by the homeland administrations during the apartheid era (1948-1994). Some areas where PTOs were not introduced remained under various forms of customary tenure, also regulated by proclamation. In the Eastern Cape, roughly 100,000 landholdings were allocated under quitrent. Although different practices were followed in the various provinces, PTOs were not generally and formally issued following the transition to democracy because they were seen as an inadequate and unequal right to land.
Lack of clear rights to off-register land creates many problems including insecurity of tenure as well as difficulties in planning and development of services such as water and sanitation, roads and public utilities. Landholders can seldom borrow against their properties and there is uncertainty over rights of succession and inheritance. Lack of security of tenure is an international development challenge and the majority of people globally, and especially in sub-Saharan Africa, do not have formally documented property rights. In South Africa, the issue is most acute in the former homelands, where 16-18 million people still live, and in the informal settlements on the urban peripheries. Insecurity affects people who are already amongst the poorest in South Africa.
In the debates around upgrading of land tenure during the transition, many felt that titles were the best form of political redress to overcome racially discriminatory laws that restricted black ownership. Along with international development priorities at the time, it was argued that title would bring access to credit, using land as security, and facilitate transactions and a land market that would benefit those previously excluded on racial grounds (Barry & Kingwill 2017).
The Upgrading of Land Tenure Rights Act (ULTRA) passed in 1991 and amended in 1996 and 1998 was initially seen as an important vehicle for land policy but it envisages upgrading to full title and requires survey of sites up to a standard required by the Deeds Office. While it has been implemented in well-established townships, such as Mdantsane where there has been a legacy of planning and clear site demarcation, but it nevertheless took land professionals and officials over ten years to convert the previous less formal system there to full title. In informal settlements and communal areas, this process is virtually impossible to implement.
In theory, quitrent holdings, which are already surveyed, can also be upgraded but the government has not been proactive in this regard and it is up to the holders of quitrent to convert them to title. Families that hold quitrents have to take the initiative themselves, agree to the updating of records, the nomination of an individual title-holder and pay legal costs to register the title. The Deeds Office in King William's Town, which holds the records of the former Ciskeian quitrents, notes that upgrading has been limited. In the limited number of Eastern Cape villages where land has been held by private title since the nineteenth century, the titles are generally out of date, largely because title holders prefer the land to be held by, and transmitted to, the family as a whole.
The preference for titling in policy has been carried forward in the policies for subsidised or RDP housing. It is here that policy assumptions about the superiority of titled and registered ownership are most strongly articulated, and at the same time most contra-indicated by evidence that government is struggling to deliver titles or to keep them up to date. A large percentage of title deeds that have been issued in the subsidised housing sector have already become outdated due to off-register transactions, local rent seeking and the expense of following formally recognised procedures (Barry & Kingwill 2017). Government agencies have fallen drastically behind in the issuing of titles. Moreover, informal settlements with no recognised tenure systems continue to burgeon.
Not only does it seem that private title is unsuitable for some contexts in South Africa but it is also difficult to register land rights adequately in the areas of customary tenure. In former PTO and customary systems, rights are diffuse. Generally, families have clear rights over a residential plot with an adjoining garden. Many of these have boundaries that are locally recognised and some are fenced but they are not surveyed or formally recorded. Beyond the residential plots, the picture is more fluid and can depend on local customary arrangements. Families may have rights over arable plots, but they are not generally fenced and they may have to give up exclusive access for parts of the year when livestock are allowed to use them as pastures. Most of the land in communal areas is grazing land, to which people have a collective right of access and use.
The title would have to distinguish between different rights over different areas of land and in certain respects could not give exclusive rights. Moreover, land held in title is clearly the property of one named individual or named joint owners. While PTOs and quitrent did name individuals, other family members are generally seen to have rights in these and in customary systems, and the named individual is not seen as the owner. Land registered in title in the name of CPA's and Trusts on behalf of groups have failed to resolve these problems. These have been the major vehicle for transfers of land under the Restitution Act and, until recently, of other land redistributed through government schemes. The Deeds Office does not have a role in relation to the governance and trusteeship of the committees of such groups. It is particularly troubling that in some cases Traditional Authorities see themselves as 'owners' with registered title to surveyed parcels of land rather than as custodians for collective ownership. The rights of families are not registered in the Deeds Office and hence become vulnerable to intermediaries, whether committees, Trusts or Traditional Authorities
Recent evidence is provided by the public hearings of the High Level Panels in 2017, which heard a wide range of testimony on the insecurity of rights in former communal areas. In the Eastern Cape a witnesses spoke of the complete breakdown in land administration and allocation systems contributing to arbitrary changes in land use, often to the detriment of women's rights in land. In Mpumalanga it was observed that
"... people continue to suffer because the land is sold... Communal land belongs to the people, it is not tribal land. It is common here in Mpumalanga where I live that traditional leaders sell land to foreigners".
A new system of land administration that gives strong rights to individual families who hold off-register land is clearly an urgent necessity in South Africa. Together with colleagues who have been involved in recent workshops on these issues, I believe that major legislation is required. In collaboration with Siyabulela Manona, we are conceiving of this new legislation as the Land Records Act (LRA)
The LRA would entail a statutory recording of all off-register rights. The proposed system would not only incorporate off-register land rights, but would include freehold titles and quitrent rights which have fallen out of step with the deeds registry system. The proposed land records system and the Deeds Registry should be closely co-ordinated, so that land parcels are visible to each, and there can be movement both ways. People should be able to upgrade to title if families agree. But many landholders are moving to less formal system by default - as indicated by informal selling of RDP houses without registering this with the Deeds Office. The Act should provide for an administrative system for all valid claims to rights that are currently 'off-register'. These will not be second class rights that are not comparable in gravity with title deeds.
The processes of clarifying rights through recordal would involve several interlocking processes. The first stage of enumeration and recordal of existing, unadjudicated rights could be done alongside a census. The second stage of adjudication of rights according to new forms of admissible evidence, including 'living law' and customary law norms, should culminate in recordal in a repository of certified rights, with the issuing of a certificate of a land right showing an adjudicated, legal land right that can stand up as evidence in a court of law.
The development of an effective system of adjudication will be a major element of the new law. New forms of admissible evidence that is compatible with the Bill or rights will have to be developed, and officials will have to be trained, and develop skills to apply them, in order to legally confirm rights and deal with, or refer, disputes and conflicts. Adjudication should apply sporadically to individual cases, or on a community-wide scale, a process we call 'land rights enquiries'.
Associated with the institutions dealing with land records, should be complementary institutions associated with enumeration and adjudication to provide recourse in the event of disputes or infringements, as follows:
(i) institutions for dispute resolution and arbitration to mediate enumerated rights that are contested or which are overlapping as a result of apartheid forced removals and sanctions against urbanisation;
(ii) an office to award new land rights for redress, restitution or redistribution; (iii) an independent public protector, which we will call a Land Ombud, to hold the executive and government officials to account in the event of infringement by the official processes landholders. An ombud office would have powers to investigate state departments, to instruct compliance, if necessary to issue subpoenas and enforce law horizontally across departments and vertically through the tiers of government. This official would also receive and investigate requests from individual members of the public and from organisations (Kingwill, Hornby, Royston and Cousins 2017).
(iv) the courts are already well functioning, however the new forms of admissible evidence should be constructed to hold up in courts of law in the event that land administration processes and procedures described above deadlock. While official (including local community-level) adjudication and dispute resolution processes are under way, however, disputes or contests should not initially be heard by courts of law.
(v) institutions that mirror the role of the Surveyor General for spatial measurement of land in terms of the Land Records Act. One central problem for a process of recordal of off-register land is the lack of detailed survey. This issue has received considerable international attention. In such contexts, which are common in developing countries, the International Federation of Surveyors advocates the recording of general boundaries, 'spaghetti' boundaries or single-points using GPS technology, as opposed to boundaries that must be mathematically surveyed within centimetre precision, thus breaking away from the tradition of spatial precision. For those outside the profession, and especially those who have limited literacy, the concept of visual boundaries is better understood. While land surveyors may argue that these systems should not exist side-by side, the proposed Land Records Act should accommodate both, allowing for progressive changes over time. These developments are increasingly feasible as a result of new digital technology at the disposal of land consultants, NGOs and professionals.
The proposals aim at an effective system that would develop strong rights for all off-register landholders in South Africa - who are in fact the majority of landholders. It protects rights at the level of families, either individually or within broader collectivities. It would be relatively inexpensive for the state to operate, since it is a re-engineering exercise rather than creating a new massive bureaucracy. Critically important, it will be relatively inexpensive for poor landholders. It could provide the basis for major new initiatives in planning and investment in the former homelands, and in informal settlements, where some of the poorest South Africans live.