Land Reform in South Africa

Author(s): Lionel Cliffe

Source: Review of African Political Economy , Jun., 2000, Vol. 27, No. 84 (Jun., 2000),
pp. 273-286
Published by: Taylor & Francis, Ltd.

Stable URL: https://www.jstor.org/stable/4006600

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to
Review of African Political Economy

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
Review of African Political Economy No.84:273-286
? ROAPE Publications Ltd., 2000
ISSN 0305-6244

Land Reform in South Africa

Lionel Cliffe

The newly-elected South African government began in 1994 to make laws

and implement a programme for land reform. It consisted of three
dimensions: redistribution (transferring white-owned commercial farm land
to African users); restitution (settling claims for land lost under apartheid
measures by restoration of holdings or compensation); and land tenure
reform (to provide more secure access to land in the former bantustans).
Only a few restitution claims have been so far resolved. After much
rethinking a revised draft of a land tenure bill is to be presented to Parliament
in late 2000, but as one stated aim is to give 'land to tribes', it remains to be
seen whether it will bring increased democratisation, allowing for common
resource management, or will entrench 'decentralised despotism'.

This article concentrates on the most actively pursued dimension of land

reform: redistribution. Under the diverse influences of rights-based activism
of earlier years and of World Bank advice about a 'market-led' approach,
the government has set up mechanisms to help finance and facilitate
'community' initiatives to acquire land, to settle on it and, if possible, to
make productive use of it. What was advocated as a more rapid and less
bureaucratic approach than a government agency acquiring and
administering resettlement has instead spawned a sprawling edifice, some
of it out-sourced to an array of consultants, often with little experience and
few credentials, and has led to a protracted process of transfer of a much
smaller amount of land in five years than, say, Zimbabwe managed in the
same period. The reasons for this are examined.

A policy rethink during 1999 has led to changes in emphasis which,

hopefully, will speed up the redistribution of land, provide more back-up to
those resettled, and prioritise future grants for more productive agricultural
use. This latter formula, however, is constricted by old-fashioned 'modernist'
(and often implicitly colonial) orthodoxies still current in South Africa, not
least in the ANC and government. These are fixated on 'commercialisation'
- which usually translates into larger-scale and high-tech - and the
promotion of the interests of a would-be black agrarian entrepreneurial class,
rather than those of the propertyless. Some hope may derive from the
inclusion of experiments in the new programme to chart an alternative to
the 'market-led' formula which would instead allow redistribution of land
as an element within district-level planning.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
274 Review of African Political Economy

With the end of the apartheid regime in South Africa in 1994, the newly elected
government introduced a programme of land reform. This had three components:

* land redistribution,

* land restitution and

* land tenure reform.

Following the elections of 1999, a new Minister for Agriculture and Land Affairs
announced a moratorium on new land redistribution and sent the draft Land Rights
Bill, which would have enacted a comprehensive land tenure reform, back to the
drawing board, while she reassessed policy on the basis of evaluation of the past five
years' experience. Now after six months of deliberations, a new policy has been
outlined. It seeks to accelerate redistribution of white-owned commercial farm land,
setting a target of 15 per cent of it to change hands in five years, but seeking to orient
reallocation to black commercial farmers of varying scale of operation; there will also
be some pilots for a switch to a less 'market-led' form of redistribution. Restitution
will be speeded up and reoriented to incorporate a more developmental rather than
simply a rights-based approach. Land tenure in the former bantustans will be subject
to reform legislation, different in some respects from the draft Land Rights Bill of
1999, to be introduced this year, which will aim to give greater security under a
rationalised system, utilising local institutions. This article will seek to assess the first
five years of land reform, explore the characteristics of the new policy and examine
how far it meets the shortcomings of the old.

Outline of the Programme*

With the end of apartheid in the early nineties, the Mandela Government launched an
active programme of land reform, manifested in new legislation and its White Paper on
South African Land Policy, 1997. The primary objective throughout has been to redress
the gross imbalance in landholding, where more than a third of the population is
concentrated in 13 per cent of the land area, and occupy land in insecure and
secondary ways. In contrast, the minority 'white' population of South Africa hold vast
tracts of land in secure freehold type regimes. In broad terms, the reform programme
comprises three parts: initiatives to redistribute land, measures to restore property
appropriated through racially discriminatory laws and policies, and a programme to
reform the manner in which property is actually held.

Redistribution
The Land Redistribution Programme aims to reallocate land to the landless poor,
labour tenants, farm workers and emerging farmers for residential and productive
uses in order to improve their livelihoods and quality of life. Special attention is to be
given to the needs of women. The following legislation has been enacted to facilitate
the programme:

* The Provision of Certain Land for Settlement Act, 126 of 1993 (renamed in 1998):
provides for the designation of land for settlement purposes and financial
assistance to people acquiring land and for settlement support;

* This section is derived from a briefing from Robin Palmer, Oxfam/UK Land Reform
Adviser, with thanks.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
 Land Reform in South Africa 275

* The Development Facilitation Act, 67 of 1995: introduces measures to speed up

land development, especially the provision of serviced land for low-income
housing.

The programme gathered momentum so that by June 1998 almost 250,000 people, in
279 projects, had received land. More people received land during 1998 than in the
previous three years combined. In the first quarter of 1999, more than 50,000 people
obtained land through the redistribution programme. By mid-November 1999, the
Department of Land Affairs had 447 redistribution projects in the implementation
phase which involved 360,256 people (55,424 households) and the transfer of 714,407
hectares of land (see below for details).

Restitution
A Land Restitution Programme aims to restore land to those dispossessed of their
rights in land since 1913 through racially discriminatory laws and practice. The
objective is 'to promote justice and reconciliation'. The Restitution of Land Rights Act
22 of 1994 was enacted to guide implementation and give it a legal basis. As of 31
March 1999, 63,455 restitution claims had been lodged. Finalised claims totalled 241,
involving the restoration of 311,484 hectares to 13,584 households (83,378 beneficiar-
ies). A further 4,365 claims were gazetted by March 1999, with another 200 nearing
completion of the work required to certify claims. Some 284 claims were rejected.
About 80 per cent of all restitution claims relate to urban land and involve substantial
financial settlement rather than resettlement on the land confiscated originally. The
status of this programme is that once a claim is recognised by the Claims Court, the
government is required to compensate the existing 'owner' and/or pay recompense
to the claimants; the commitments under it are not limited by specific budgetary
provision or shifts in policy.

Tenure Reform
The Land Tenure Reform Programme deals directly with the means through which
land is owned. In particular, it seeks to address issues pertaining to the insecure,
overlapping and disputed land rights resulting from the previous systems of
governance, especially in the former bantustans. The following laws have been
enacted to facilitate reform: The Interim Protection of Informal Land Rights Act 31 of
1996, a mechanism to protect people with insecure tenure from losing their rights, and
interest, in land pending long-term reform measures; the Communal Property
Associations Act 28 of 1996, which enables communities or groups to acquire, hold
and manage property under a written constitution; and The Land Reform (Labour
Tenants) Act 3 of 1996, which provides for the purchase of land by labour tenants and
the provision of subsidies to this end. In addition, the Extension of Security of Tenure
Act 62 of 1997 ('ESTA') provides for tenure security in two ways: first, by helping
people living on rural or peri-urban land to obtain stronger rights to the land on
which they are living, or on land close by; and second, by laying down certain steps
that owners and persons in charge of rural or peri-urban land must follow before they
can evict people. The Act also regulates the day-to-day relationships between owners
and people living on rural or peri-urban land. A proposed Land Rights Bill was
intended to finalise the programme and, pursuant to the 1997 Land Policy, was to
significantly reform the context within which South Africans in the ex-homelands
occupy property.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
276 Review of African Political Economy

Origin & Logic of 'Market-led' Land Redistribution

The thrust behind this package of measures was a curious hybrid of pressures and
perspectives. On the one hand, policy grew out of those initiatives concerned with
land issues during the previous decade's mounting mobilisation against apartheid.
By and large these involved struggles to assert the claims of those who had lost rights
to land or who were threatened with loss of access. These included attempts to defend
the interests of those occupying 'black spots' of land within designated 'white areas',
existing and former workers and 'labour-tenants' still resident on white owned
commercial farms or squatting precariously on pockets of land, and peri-urban
dwellers threatened by urban development. They also included the assertion of the
rights of those removed from land, whether by recent removals and dispersals or by
historic land seizures. These struggles had spawned a range of different forms of
mobilisation of groups and communities, some based on traditional, even chiefly,
identities. In KwaZulu-Natal, for instance, there were reportedly over 100 nkosi
(chiefs) who, with the people who swore allegiance to them, were landless. Such
bodies were often advised or mobilised by activists and crusading lawyers. This
constituency was drawn to formulations that would have government move in to
assert, consolidate and guarantee those rights for which they had fought campaigns,
often defensive ones.

It was this thrust that was behind the inclusion of a land restitution provision in the
land reform package, distinct from the patterns of land reform in ex-settler colonies in
southern Africa where the complexity of establishing and implementing such historic
rights was put aside in favour more simply of land redistribution. But South Africa's
formula was also different from those in Zimbabwe, Namibia, even Swaziland and
before that in Kenya, in that the redistribution programme was not implemented
through a government agency acquiring large areas of land and then parcelling it up,
providing infrastructure and back-up services for individual African smallholders.
Instead redistribution was left to be initiated by 'communities', envisaged as similar
to those that would be pursuing restitution, but without qualifying legally under that
programme. Government would play a facilitating role, making grants available for
such communities to purchase land through the market. This grants-based approach
also owes much to the limited number of previous activists who had pursued land
causes in the recent past: one of the few constituencies from which policy makers,
advisers and reform implementing officials could be drawn.

On the other hand, this approach to redistribution through grants and land transfer
through the market was also very much to the economic thinking of another set of
actors: the World Bank and other international 'experts' wedded to a 'market-led'
land reform. This perspective is perhaps best represented in a collection of papers,
most of them general readings of international experience or of South African realities
and designed to yield a model for the right policy choice, authored by interconnected
'experts' from the World Bank, herded together by a senior analyst from Washington,
Hans Binswanger. They included those from the Southern Africa Development Bank
waving apartheid era reformist, verligte credentials and some agricultural economists
based in one academic institution (Van Zyl, Kirsten & Binswanger, 1996). Their
influence was fed into the policy-making process through a series of conferences and
other dialogues with the ANC, government and activists from the early 1990s. Their
thinking was based on a particular reading of the lessons of experience elsewhere in
Africa, and in other parts of the world, such as Central America, which were dealing
with different inherited structures and thus were of limited relevance. What must be
recognised about this approach is that World Bank officials and experts have been

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
 Land Reform in South Africa 277

urging some degree of land reform in South Africa since the start of the transition in
1991. Their reasons are straightforward:

* Politically it is essential so as to avoid 'decades of likely peasant insurrection,

possibly civil war, combined with capital flight and economic decline'
(Binswanger & Deininger, 1996: 94);

* In micro-economic terms, a 'small farmer' strategy will lead to inverse

economies of scale, through more intensive use of labour and relatively less
capital (hitherto available through unjustified subsidies to 'white-owned
farms);

* In macro-economic terms, transfer of land will allow a resolution of the 'debt-

crisis' affecting many white-owned commercial farms as a result of liberalisa-
tion and removal of subsidies, without the need to devote state revenues to debt
forgiveness or debt relief.

The validity of these assumptions continues to be part of the analytical and

ideological debate in South African circles (see Lipton, 1996; Bernstein, 1996 etc.) and
can only receive brief mention here. However, it is striking to an outside observer
how little the classic debates around such issues of agrarian change are either
understood or actively pursued by policy makers or analysts in South Africa. In both
camps it is common to argue against the first, counter-revolutionary imperative by
stating that the rural poor in South Africa don't want land but jobs, security and social
services (Longusile, 1998), a convenient device to exclude the vast majority of
residents of former bantustans from the process of land redistribution. I even heard
that line of argument from a local official dealing with land, who then blithely went
on, without any awareness of the contradiction, to defend a redistribution which
legally recognised the de facto land rights of a very large group of 'black-spot' dwellers
on the grounds that it had headed off the looming threat of a massive land invasion!

Bernstein (1998) is right to criticise the view of those like the Liptons who assert that
small always will be more efficient. However, all too often in South Africa the polar
opposite position is so deeply embedded in the subconscious that it is beyond the
powers of persuasion even of the World Bank experts to convince many practitioners
that smallholder farming might be more productive in certain circumstances.
Bernstein (1998) is also correct that one common approach to generating guidelines
for redistribution is simply to scale down, without fundamentally changing, the mix
of factors of production on large commercial farms to some minimum viable scale.
One can add that the supposed 'technical' basis for the inherited regulations which
preserve the size of farms by prohibiting sub-division beyond a certain 'floor' is
derived from assumptions of what is an appropriate target income for a white farmer
(in some cases a generation ago, and for a 'poor white' as well) rather than any logic
relating to economies of scale of production!

Whatever the status of these arguments, it is the consequences of some of the World
Bank's type of thinking for the shaping of policy that should be noted here. Convinced
of the necessity for some redistribution, authors like Binswanger (1996) have
marshalled arguments against any approach based on confiscation of land: to do so
without any compensation, they felt, was precluded because the political situation
was not revolutionary enough to overcome the inevitable opposition - probably a
correct reading of the situation even in the early 1990s - and in any case this was
precluded by South Africa's new constitution. But they also argued against

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
278 Review of African Political Economy

confiscation with compensation, especially if below the market price of land, on the
grounds that it would be costly and slow, involving contestation through the courts
or in the political arena, and would have to be administered via a state agency which
should be avoided at all costs, as it would be bureaucratic and would ultimately be
rent-seeking.

Combined with the 'rights-based' tradition, this thinking led to a policy that allowed
some restitution of judicially established rights of evicted peoples and envisaged
greater security of tenure in former bantustans, but also put the major emphasis on 'a
market-led redistribution' approach. Land would be transferred on a willing-buyer/
willing-seller basis, where the transaction would be between farm-owners and
communities, but where the latter would be entitled, on application, to a grant (as a
recognition of historic wrongs) from government. This approach would obviate the
need for a separate land reform bureaucracy (although as we shall see the market-led/
application-grant based approach (MAG) succeeded in spawning a considerable one).

Overall Performance of the Programme

The Approach to Redistribution

The Land Reform Pilot Programme sought as one of its aims to provide a range of
experimental approaches which could generate a learning experience to inform a
more widespread programme. The variety of actual 'projects' and beneficiaries was
indeed extremely wide, as we shall document. But the process of land redistribution
that emerged corresponded increasingly to a single pattern. The rules put the onus on
groups and communities of would-be beneficiaries to take the initiative: to form a
group, and eventually a legal entity, that could be assigned ownership of the land; to
indicate a targeted piece of land or type of land they were seeking; and on that basis
apply for a grant to make the purchase. The formal stages of this process and the
approvals required are listed in the next section. In practice the Department of Land
Affairs (DLA) not only registered the groups but advised on forming legal entities and
on drawing up 'business plans' for farms that were sought. Indeed, the grant had to be
approved by the Minister and was only so 'designated' on the basis of such an
approved plan. A single formula evolved as the basis of the grant - the provision of
R15,000 (later R16,000) per beneficiary household. Once 'designated', a more detailed
plan for land use had to be worked out for the new owners to implement.

DLA sought to meet these needs by adding a lttle to the grant, but also by farming the
planning out to 'consultants'. The advice provided was often lacking or superficial,
and seldom coupled with continuing support such as credit and agricultural advice:
some consultants never consulted their clients. In those regions away from the main
metropolitan centres where there were few local consultancy firms or NGOs,
provision was particularly problematic. Without the expertise to provide the required
range of assistance, and inhibited from becoming a 'rural development agency' by the
programme's conceptual framework as 'market-led' reform, much of the pre- and
post-plan making and other advice, as well as some core functions of the DLA itself,
were 'out-sourced' in a manner in keeping with the pattern in today's climate of
economic liberalisation around the globe. While scarcely preventing the emergence of
a very considerable bureaucracy, these measures inhibited the emergence of an
agency that could institutionalise the lessons of the experience that was accruing and
the build-up of a trained and experienced cadre of professionals that could perform
the many and complex tasks required.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
 Land Reform in South Africa 279

To these design shortcomings was added a peculiar characteristic of South Africa

under transition. Over the last few years there have been successive waves of
restructuring and institutional transformation throughout the public service. One
essential element of these processes has been a concern to redress the racial imbalance
in senior staff. DLA seems to have devoted massive amounts of time, creative
thinking and finance (including aid funds) to analyses, inquests, designing proposals
documentation and discussion sessions around these structural issues. One wonders
whether these have not sometimes been at the expense of, rather than a means to,
effort and creativity being devoted to critique and refinement of policy and strategy.
These efforts have been justified on the grounds that appropriate structures are a
prerequisite to the right policies. But it is only against the yardstick of policies and
their requirements that the 'appropriateness' of structures can be assessed. Moreover,
the new initiative in 1999 to reconsider policy owed more to changes in personnel at
the level of Presidency and Cabinet than internal restructuring.

In sum, a sizeable bureaucracy has emerged, requiring complex processes of approval

built onto a notionally 'market-driven' redistribution. The consequence is that these
processes have been long and drawn out. It is not unusual for it to take two to three
years for a group registering an interest to take possession of land. And, perhaps
inevitably under such a system, many initiated projects never reach fruition. At the
same time, the consultants brought in on expensive contracts to plan with groups
often have little competence in agricultural enterprise planning, and on occasion put
together documents without on-site observation or discussion with beneficiaries - nor
does this aspect of the process yield much of a learning experience.

In spite of these cumbersome processes, efforts by DLA provincial staff have since
1997 begun to see a significant number of projects emerge at the end of a long pipeline.
But overall the process has been modest by most measurements. Against the ANC's
target of redistributing 30 per cent of white-owned land in five years, less than 1 per
cent has changed hands. Compared with Zimbabwe, for instance, where the World
Bank's nightmare scenario of a state-led redistribution nevertheless led to over two
million hectares (about 15 per cent of the large-farm sector) being redistributed in five
years, the market-led system with its much vaunted speed and efficiency, achieved
just over a third of that aggregate and the transfer of only a tiny proportion of white-
owned farms in a vastly larger country.

Review of Projects & their Implementation

The most concrete measure of the achievement of DLA is in the actual projects that
have been initiated and processed. In land redistribution, which has been the main
area for projects, the procedures laid-down envisage a project going through five-
stages, from the coming together of a group of beneficiaries around a desire to acquire
and utilise a piece of land, to the implementation of a specific plan. These phases, each
with its own sub-phases, are: Project Identification; Feasibility Investigation;
Designation & Land Transfer; Detailed Design; Development & Support.

Given the nature of the grant/application system, and the often rather vague
aspirations around which initial proposals are framed, large numbers of projects
lodged at the early, formulation stages, never come to fruition. Inevitably, therefore,
DLA offices have portfolios with large numbers of projects on their books at any one
time, and as of mid-1999 when the new Minister called a moratorium on new projects,
most were still at the pre-transfer stages. Significant catching up with this backlog has
been achieved up to March 2000, however, when the moratorium was lifted.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
280 Review of African Political Economy

The crucial two stages at which a project becomes a reality are those in Phase 3, the
moment when it is 'designated' by the Minister and announced in the Government
Gazette as approved, authorising the subsequent actual transfer of title to the legal
entity formed by the beneficiaries. Therefore, while the total number of projects on the
books, including those in the first two stages, is useful in indicating how many are in
the pipeline, the number having reached Phase 3 is more indicative of what is actually
happening on the ground.

The latest figures supplied by DLA, to mid-November 1999, and summarised in the
Table below, indicate that the number of designated projects was officially 447,
benefiting over 55,000 households on land totalling 0.7million hectares.

Table 1: 'Designated' Projects, Beneficiaries & Areas by Province

(mid-November 1999)

Province Number of Beneficiary Women- Area of

Projects Households headed Land
Households (hectares)

Eastern Cape 49 17,047 1,093 51,555

Free State 130 3,465 548 56,042
Gauteng 20 1,464 581 5,426
KwaZulu-Natal 52 6,024 543 156,839
Mpumalanga 42 7,642 1,109 29,427
Northern Cape 50 3,667 684 334,008
Northern Province 34 4,966 1,771 26,012
North West 30 8,700 433 13,532
Western Cape 40 2,449 498 41,566
Total 447 55,424 7,260 714,407

These totals include 46 Commonage Projects and 35 Equity schemes; the number of beneficiaries of the former
are particularly difficult to quantify as the use of these areas can be open to all (poor) municipal residents.

The great majority of these projects are of the mainstream land redistribution type,
but they do contain significant numbers of equity (joint ventures of registered groups
with commercial farmers) and commonage (transfer of mainly peri-urban state land
for use as common grazing and other uses by residents) projects. There are also a
number of projects that have emerged as a result of restitution cases and the granting
of security to labour tenants, which are not included in the Table One.

One limitation of the DLA project database from which the above figures are taken,
whether as a management or evaluative tool, stems from the paucity of data retrieved
about the type of project and the composition of beneficiaries. The data from some
provinces merely classifies projects into two categories: where land is used for
production and where it is used for settlement. This simple two-way categorisation is
used routinely by DLA in its reports and policy documents, including the influential
Transformation Workshop Document which sought to define 'The Products of the
DLA' (December 1998). However, some of the data supplied by Provincial DLA
offices incorporate further distinctions as to the type of production - irrigation,
livestock, commercial crops, etc. Such information is badly needed for planning and
review purposes, and certainly for prioritising between products.

In contrast there has been a tendency in the present policy-revision environment for
some in the DLA and many in other spheres of government to address the core of the
problem in stereotypical fashion. The picture they portray is of a long 'pipeline' of

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
 Land Reform in South Africa 281

project applications, many by sizeable groups of would-be beneficiaries, mostly

strangers to each other, who have been rather randomly brought together in order to
seek a grant and thereby obtain secure title to the land they already occupy or upon
which they hope to build a residence. These are typically regarded as 'unsuitable'
projects, now classified as 'non-products' by DLA. They are in contrast to smaller
projects where people want land to make a commercial success of farming. There is
often a corresponding mode of thinking that holds that prospective beneficiaries can
be divided into two parallel social categories: those only interested in finding a secure
place to live and those regarded as 'real farmers'.

This kind of stereotyping of projects and people is a throwback to official mentalities

found throughout colonial or apartheid southern Africa in previous generations that
led to the belief that only a minority of Africans, capable of becoming 'master
farmers', were worth resourcing. It is hardly informed by the 'smallholder' lobby of
the World Bank and the thinking of the Liptons. But it continues to be subscribed to by
a significant body of African policymakers. It has little basis in fact, as many successes
of small-scale, peasant farming in the region, often after resettlement (see Moyo in this
Issue) have indicated. Nor by any means do the actual projects that DLA has
supported fall into these two polar opposite categories. Of the 19 projects, at various
stages of pre- and post-designation, that I saw at first hand, only two could be
categorised as solely to do with secure settlement; the rest all had some productive
potential, however limited (see the 'sketches' of some of these below). Admittedly this
constituted a small and not necessarily representative sample - all the more need,
therefore, for more detail about land use and potential on projects and for empirical
testing of the categories used in planning. However, even this tiny, unrepresentative
sample contains enough exceptions to cast doubt on the simple dichotomies used in
planning documents.

Table 2: Productive Activities on Transferred Land: Numbers of Projects,

by Province 1999

Type of Land Use WC FS NW G EC KZN NC NP Mp'a Total

Cultivation 74 100 16 14 52 280 56 102 125 614

Grazing 18 4 0 1 8 32 26 2 7 98
Other agricultural 4 3 0 1 3 42 8 1 3 66
Forestry 1 0 0 0 1 0 0 0 0 2
Business 2 1 2 2 1 0 1 0 1 10
Tourism 1 0 0 1 0 7 0 0 1 10
Residential 3 0 1 6 1 29 0 2 4 46
Unutilised 6 0 2 2 7 1 1 3 0 13 44
Total 109 108 21 27 73 401 94 107 154 890

Data supplied by DLA.; Key: WC - Western Cape; FS - Free State; NW- North West; G - Gauteng; EC -
Eastem Cape; KZN - KwaZulu-Natal; NC - Northern Cape; NP - Northern Province; Mp'a - Mpumalanga

Available figures that seek to give a more nuanced view of actual situations are
presented in Table 2. These tend to demonstrate the falsity of the view that land is
used essentially for residential purposes in a large minority of designated projects.
Only 10 per cent could be said to correspond to the literal definition of what in DLA
parlance is a 'settlement' project, that is, where the land was unutilised or used
primarily for residence. The overwhelming majority involved some agricultural
usage of the transferred land, for either cultivation or grazing - although these
categorisations do not allow distinctions to be made as to the level of production.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
282 Review of African Political Economy

However, the data do suggest that guidelines based on polarised categories of

'production' and 'settlement', for instance in the new policy's emphasis on
'commercial farm' redistribution, may well not correspond to the realities. It would
perhaps be more appropriate to deal with projects currently in the pipeline by
identifying mechanisms for utilising the productive capacity of particular land to the
full, given the nature of intended beneficiaries and the substance of their plans.

'Communities' & Organisation of Production

The application/grant-based approach grew out of the assumption that there were
'communities' in existence that were keen and able to take advantage of opportunities
to acquire farms and put them to productive use. A number of studies (see Kepe, 1998)
have cast doubt on this assumption, arguing that part of the legacy of apartheid is that
local communities have been displaced, eroded and/or atomised to the point where
many people dwell in artificial social circumstances. This view seems to be supported
by empirical findings about the nature of the groups of beneficiaries applying for
land. It is suggested that many are sets of individuals who have come together with
little if any previous mutual involvement for the sole purpose of constituting
sufficient critical mass to purchase a farm. There is, however, little hard evidence
about the origins of groups to support or refute such generalisations; detailed
sociological research into how groups come together, their basis and coherence and
the group dynamics as they make plans and acquire land should be undertaken to
inform this aspect of policy. My impressions are that there are often some connections
on which groups are formed - workers or former workers on a particular farm (even
if not the one being sought), residents on a black spot or squatter locale, those owing
allegiance to an nkosi (even if not resident in the same place). But these are often
tenuous. More importantly, they may not be an adequate basis for them to share
property, to plan its collective use or to manage the land, its use and some group-
based economic activity.

This last concern, with the type of collective activity and joint property arrangements
envisaged or implied in a project, has been even less investigated - although it goes to
the root of the appropriateness of the past policy approach. In fact, an important
assumption behind the application/grant-based approach is that the land is
transferred to a single legal entity made up of a group of households and/or
individuals, with the further implicit assumption that they will organise productive
use of that land as a collectivity of some kind. Little provision, legal or managerial,
seems to be made for land use and thus for rights to it to be sub-divided to individuals
or households. Thus within the 'market-led' approach there is a hidden presumption
for cooperative production - even within a new policy emphasis on commercial
farming ventures. Yet any kind of cooperative or collective organisation of
production is notoriously complex and difficult to introduce and maintain - desirable
though such aims might be in terms of equity and social cohesion. Such challenges
and difficulties are compounded in circumstances where the members of the group
are moving onto new land, where there is limited sense of community, and where the
only types of farms with which they are likely to be familiar are those of the individual
family plot or the authoritative management of a capitalist farm, rather than any form
of self-managed collective activity. Such problems are made more acute where the
members include those who are in fact 'workers', seeking to use the land as their main
source of livelihood, and others who are 'shareholders' with other sources of
livelihood, often at some distance away.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
 Land Reform in South Africa 283

The DLA has a Community Support Facility, funded by international donors, which is
in theory one resource to help project groups acquire the skills necessary for
cooperatively organised activities. But it seems seldom to be used for such purposes -
perhaps due to the lack of successful pilot experiences and thus of the acquired know-
how and available facilitators and trainers in South Africa. Such expertise in self-
managed activities or even in commercial small-holder production is also lacking in
the various departments of Agriculture, which come under provincial governments
in the South African constitution - and even more crucially in the assumptions which
inform the whole practice of agricultural professionals and structures in South Africa.
These assumptions are often justified as 'technical', but are more often covers for
social and economic value judgements about desirable levels of farmer incomes and
thus of the overall distribution of income in the country, and about what is considered
to be the appropriate sociological and cultural make-up of a 'real' farmer. Although
such assumptions reflect an implicit, though far from crude, form of racial
stereotyping, they are also often swallowed as part of the common sense economic
and technical wisdom by African technical personnel and policy-makers.

Some data about the actual level and type of group activity on projects was beginning
to come out of the Quality of Life Survey undertaken as part of DLA's work in late
1999. Thus, the Survey reveals that out of 87 projects for which respondents supplied
data, only 35 had any kind of 'communal farm'. However, some 45 cited other kinds of
common productive activity or shared facility.

A more immediate feel for what this programme has in fact delivered in terms of
projects may be gleaned from the following thumb-nail sketches of a selection which I
visited. While they are in no sense a representative sample, they underline the great
diversity of projects: extending from 10,000 beneficiaries in one case to a single
extended family in another and from intensive commercial crop production under
irrigation to a housing estate. They may be a useful corrective to simple
dichotomisation into 'settlement' versus 'production' or real 'commercial' farmers
versus 'subsistence', or other stereotypes.

Cornfields, KwaZulu-Natal
This is a real 'community' which formerly occupied a 'black spot' in a white farming
area, having successfully resisted removals since 1910. They were given title to the
land and a grant to buy three neighbouring white-owned farms under legislation of
the transitional government in the early 1990s. Between 8-10,000 people are
beneficiaries; many are migrant workers who have a home-base to fall back on
through resident family members. These latter have 'block farms' and are hoping to
start other group enterprises (more irrigation, poultry, handicrafts). A piped water
supply and access road are being completed, utilising a DLA redistribution grant. But
land-based economic activities are not likely to provide a sole source of adequate
livelihood for more than a proportion of beneficiaries. Some tensions still exist with
local white farmers ('we may get beaten up when we go to recover livestock that have
strayed').

Tumbleweed, KwaZulu-Natal
This is a collection of some 50 households gathered together by a 'landless' nkosi
(chief) to whom they owed allegiance. They used a grant to buy a farm in a peri-urban
area where some prospect of jobs exists and are building a small housing estate upon
it.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
284 Review of African Political Economy

Ngxito, Free State

This consists of two brothers and their families, who have years of experience as small
traders. They have acquired a peri-urban small-holding to use for vegetable growing,
as a fattening pen for sheep, and as a base for a range of trading activities in animals
and manufactured goods.

Irrigation Cooperative, Mpumalanga

An enterprising group of 20 households from former bantustan areas with
agricultural know-how occupied a set of vacant 5 hectare plots formerly designated to
'poor whites' at one end of a vast irrigation scheme. Efforts are still being made to
transfer ownership to them so they can't secure loans. They are cultivating a range of
irrigated crops - sugar, tobacco, wheat, and vegetables. They have an effective
management committee for organising the distribution of water and other common
resources to individual plots. Most households have pick-ups or other vehicles. They
are still being over-charged and in other ways hassled by super-racist members of the
local white farming community.

Perhaps only the latter of these four cases would be prioritised under the new policy
guidelines for enhanced grants.

New Policy Initiatives

The new measures to be launched later this year start from an ambitious goal of seeing
a much more significant transfer of land from the white-owned commercial sector in
the next five years than has been achieved in the last, though still only half that
promised by the ANC government. They also reflect current government thinking in
associating redistribution with facilitating the emergence of a class, or rather three
strata, of black capitalist farmers. These formulations are not completely at odds with
DLA's own thinking, which has been shaped by some systematic learning from the
pilots of the last five years.

One point of consensus that has emerged is to give sharper focus to grant/application
redistribution by prioritising such projects as have a clear potential for improving
livelihoods through meaningful land and resource-based productive activities. There
has also been a realisation that the rigid formula of a R16,000 grant per household is
unrealistic, given the present market price levels of high potential land. This emphasis
has been given a particular interpretation in the new formulations of March 2000
which announce that 'commercial farming' projects will be the main focus of the new
policy, and that they can qualify for higher levels of grant than the basic one, but at
three levels: small, medium and large-scale. This is seen as the most important short-
to-medium term measure for achieving the target of redistributing 15 per cent of the
white-owned farm sector. Certainly, the R16,000 formula has meant that a group's
total grant has often been insufficient for meeting purchase costs, let alone
development, and that more people than could obtain a reasonable livelihood from
the land have had to be incorporated to come up with enough money to get the land.
Circumstances could well remain difficult for groups looking to move into 'peasant'
farming, and to engage in commercial as well as subsistence production. Whether the
level of enhanced grants will provide working capital, or even enough to buy the high
potential (irrigated) land, is by no means certain. Nor is it clear whether this new
formula offers any mechanism for the redistribution of one huge portion of
commercial land that is marked by its owners' readiness to sell at reasonable prices:
the large maize (plus cattle) farms of the high veld (see Bernstein, 1996).

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
 Land Reform in South Africa 285

A more welcome dimension of new policy is that it incorporates some of the emerging
thinking that has questioned the whole market-led, application/grant-based ap-
proach. It will include a parallel approach to land reform programmes that is 'supply-
driven', in the common but inappropriate current terminology, whereby suitable land
coming onto the market is earmarked, purchased and then made available to
identified beneficiaries, and where such activities are more closely situated within
district-based spatial and development plans for integrated rural development that
are formulated by a cohort of district councils, provincial and national line ministries,
with participation from community-based organisations and the public at large.
However, such a shift of emphasis will require the development of new guidelines for
plan formulation and implementation, and amended methods of work and skills, not
just for DLA, but also for a range of other staff in ministries of agriculture, housing
and in local authorities. It will also require major breakthroughs in promoting 'co-
operative governance', the South African jargon for coordination. And it will
necessitate the working out of appropriate approaches to specific areas and their
physical and social environments: e.g. the maize Boer heartland, densely populated
former bantustans along with neighbouring areas, agro-pastoral areas, specialist tree
crop plantations and estates.

Pilot district-wide programmes will be initiated in the next plan period to generate
the necessary learning experience to allow South Africa to transcend the 'market-led'
approach that their own rights-based political traditions - and their reliance on one
narrow-band of international 'expertise' - has landed them with and which, in the
eyes of this analyst, is one of the reasons for the limited extent of land reform so far.
The oft-repeated explanation of radical critics, that there is a 'lack of political will' is
not wrong but it is simplistic. A 'capitalist' reform may be the best that can be
expected in the present conjuncture, but the nature and extent of such reform should
not just be left to capitalists and, in particular, not to those whose calculations are
short-term and opportunistic. But more radical entrants to this (still ongoing) policy
debate need to generate concrete formulations. They must also challenge two
dominant paradigms in the process: first, the so-called 'technical' formulae of the old-
fashioned paternalist white agriculturalists and, equally, the black officials and
policy-makers who have bought their assumptions of World Bank-type
'smallholderism'. Only then will it be possible to go beyond aspirations of merely
Africanising prevailing (massively inegalitarian) production relations.

Another crucial area of the policy shifts is in relation to land tenure, especially in
former bantustans. After some hesitation over, and redrafting of, the 1999 draft of a
Land Rights Bill, a new law that will seek to overcome the present, often chronic,
insecurity of tenure will be presented to Parliament this year (2000). Its precise form is
thus not yet known but it aims to provide a basis for security of rights which go
beyond simply registering individual, negotiable titles. There is also talk that it will
lodge such rights with 'tribes', though what that means in practice will have to be
clarified in law. It also remains to be seen whether it will imply a democratisation of
allocation of land use and will facilitate common resource management, or be a
vehicle for reinforcing what Bernstein (1998), echoing Mamdani (1996), correctly
refers to as the system of 'decentralised despotism' set-up under apartheid.

Lionel Cliffe is in the Institute of Politics and International Studies, University of

Leeds, UK.

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms
286 Review of African Political Economy

References

Bernstein, Henry (ed.) (1996), The Agrarian Mpumalanga, South Africa', Africa World Press:
Question in South Africa, London: F Cass. Trenton, NJ.

Bernstein, Henry, (1998), 'Social Change in McIntosh Xaba Associates et al. (1999), 'Review
the South African Countryside? Land & of the Land Reform Support Programme', Report
Production, Poverty & Power',Journal of Peasant to Department of Land Affairs, Pretoria, 30
Studies, 25, 4 July: 1-32. November.

Brown, Marj et al. (1997), Land Restitution in Ministry of Agriculture & Land Affairs (1998),
South Africa: An Independent Evaluation, Institute
'Agricultural Policy in South Africa: A Dis-
for Development Policy & Management, Univer-
cussion Document', Pretoria.
sity of Manchester.
National Land Committee (NLC) (1999), 'A
Deininger, Klaus (1999), 'Making Negotiated Report on the National Land & Agrarian Reform
Land Reform Work: Initial Experience from Conference', Pretoria.
Colombia, Brazil and South Africa', World
Ntsebeza, Lungisile (1999), 'South Africa's Land
Development, 27, 4: 651-672.
Tenure Reform Programme in the former
Department of Land Affairs (DLA) (1998), Bantustans: The example of the Eastern Cape
Province', paper for NLC Conference.
White Paper on South African Land Policy, Pretoria

Kepe, Thembela (1998), 'The Problem of Van Zyl, J, J Kirsten & H P Binswanger (eds.)
Defining "Community": Challenges for the (1996), Agricultural Land Reform in South Africa:
Land Reform Programme in Rural South Policies, markets and mechanisms, Cape Town:
Africa', Land Reform and Agrarian Change in Oxford University Press.
Southern Africa, Occasional Paper no. 6, School
Wildschut, Adele & S Hulbert (1998), 'A Seed
of Government, University of Western Cape.
not Sown: Prospects for agrarian reform in South
Levin, Richard & Daniel Weiner (eds.) (1997), Africa', Deutsche Welthungerhilfe & Interfund.
'No More Tears ... Struggles for Land in

This content downloaded from

13.245.153.107 on Sat, 21 Oct 2023 08:57:10 +00:00
All use subject to https://about.jstor.org/terms