This Content Downloaded From 102.134.149.146 On Fri, 12 Aug 2022 08:57:02 UTC
This Content Downloaded From 102.134.149.146 On Fri, 12 Aug 2022 08:57:02 UTC
This Content Downloaded From 102.134.149.146 On Fri, 12 Aug 2022 08:57:02 UTC
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
School of Oriental and African Studies is collaborating with JSTOR to digitize, preserve and
extend access to Journal of African Law
SIMON COLDHAM*
Over the last ten years land reform has been the subject of intense
Uganda. Some of the issues that have arisen (e.g. the future of custo
tenure) are familiar and have been discussed at length in the contex
reform elsewhere in sub-Saharan Africa. Others are rooted in Uganda's
historical context. The Land Act 1998 (No. 16 of 1998) is the end-
this debate. It is not a comprehensive code of land law, but it does p
legal framework governing land tenure, land administration and the
of land disputes. Issues of land policy are worked out in some detail in
reflecting the view that a law which recognizes and protects private l
needs to be specific and clear and that a law which provides for the
powers by officials needs to spell out clearly how such powers are to b
leaving as little as possible to administrative discretion.' The Land Act
the most important piece of land legislation since the Land Refo
19752 (which it repeals) and, arguably, it represents as great a revoluti
relations as the Uganda Agreement and other reforms ushered in at t
the colonial period. Inevitably, it is highly controversial.
In this article I propose first to set out the principal reforms intro
the Act under three headings (mailo land, customary land, land admin
and then to select for further discussion certain issues of general inte
by the Act.
The most intractable policy issue facing the government was, undoubtedly,
the future of mailo land. Indeed, ever since the Uganda Agreement of 1900
between the British Government and the Kingdom of Buganda granted some
8,000 square miles of land (hence known as "mailo" land) to the Kabaka of
Buganda and various other chiefs and notables, the nature of the rights of those
who had traditionally occupied and farmed those lands has always been a vexed
question. In spite of colonial legislation designed to protect customary tenants
of this kind, their position was always vulnerable. The Land Reform Decree,
1975, which declared all land in Uganda to be public land to be administered
by the Lands Commission, further weakened their position. All freeholds including
mailo ownership rights were converted into 99-year government leases and
customary tenancies on mailo land were converted into customary tenures on
65
public land with little security of tenure. In the discussions on land policy that
took place during the 1990s the former mailo owners pressed for the full
restoration of their rights while the customary tenants for their part demanded
security of tenure of the lands they occupied.
The 1995 Constitution sought to address the land question. While Article
237(1) declared that land in Uganda belonged to the citizens of Uganda, clause
(3) made it clear that land was to be held on one of the following four tenures:
customary, freehold, mailo and leasehold. Although mailo ownership is equivalent
to freehold ownership, the clause left no doubt that the 1975 conversion of mailo
rights had been reversed. On the other hand, clause (9) required Parliament
within two years to enact a law (a) regulating the relationship between the lawful
or bonafide occupants of mailo, freehold or leasehold land and the owners of the
land, and (b) providing for the acquisition of a registrable interest in the land
by the occupant. In the meantime the occupant was to enjoy security of tenure.
The determination of the precise nature of the relative rights of owners and
occupants was thus deferred. However, while it was obviously desirable to hold
consultations on the issue, it was equally obvious that it would be impossible to
secure a consensus.
must give first refusal to the owner or tenant as the case may be.'2 Although a
tenancy by occupancy is in many ways similar to a common-law tenancy, its
existence is conditional on continued occupation of the land by the tenant, his
agent or a member of his family. If the land is left unattended for three years
or more, the tenancy lapses.'3
While the Act thus strengthens the position of those in occupation of mailo
land, it does not go so far as to provide for the compulsory enfranchisement of
customary tenancies, as had been demanded in some quarters and indeed as
had been done in neighbouring Tanzania.'4 Instead, a tenant may apply to the
owner to be registered as the mailo owner in his place.15 A Mediator may be
called in to assist negotiations between the parties, but it is clear that there is
nothing to prevent the owner from simply rejecting the application. To this
extent the Act can be seen as striking a compromise between the interests of
mailo owners and occupants, albeit one that tends to favour the latter group.
Apart from addressing the specific problems relating to mailo land, those
responsible for framing a national land policy had to consider the future of
customary land tenure. Ever since the Crown Land Ordinance, 1903, whose
effect was to make those occupying land under customary law (outside the mailo
area) tenants at will of the Crown, the legal position of such people has been
extremely precarious. The Land Reform Decree, 1975, declared the customary
occupation of public land to "be only at sufferance", though it provided for
compensation to be paid if the occupation was terminated by the Lands
Commission.l6 Instances of government alienation of such land on freehold
(before 1975) or leasehold had led to demands for some degree of security of
tenure.
The Constitution itself went some way to meeting these demands. Artic
237(2) recognized customary land tenure as being one of the four ways (tog
with freehold, mailo and leasehold) in which land could be owned, while a
237(4) provided for the possibility (a) of customary owners acquiring certif
of ownership, and (b) of customary land being converted to freehold land
registration. These provisions left open a number of obvious questions wh
were to be addressed in the course of discussions on the Land Bill, but they
indicate that it was government policy that, at least in the long term, most
in Uganda should be held on individual freehold tenure. To this extent it ado
the free market approach recommended by the 1989 study carried out by
University of Wisconsin Land Tenure Center together with the Makerere Institute
of Social Research.17
If, as the Constitution clearly envisaged, customary owners were to be able
to acquire certificates of ownership, the Land Bill had to lay down procedures for
determining who was a "customary owner" and to set out the legal consequences of
such a person acquiring a certificate of ownership. The adjudication and
registration of individual freehold titles in Kenya had threatened the customary
entitlements of members of the owner's family (particularly those of women and
children) and the question in Uganda was how to strike a balance between
promoting a free market in land on the one hand and protecting customary
rights on the other. The result was a compromise.
The definition of customary tenure contained in section 4(1) recognizes both
individual, family and communal ownership of land, and section 5 accordingly
provides that any person, family or community holding land under customary
tenure may apply for a certificate of customary ownership of that land. The
application is to be determined by the parish Land Committee, a body whose
establishment,18 composition, functions"9 and procedures20 are set out in some
detail in the Act. The Committee's recommendation has to be forwarded to the
District Land Board which has broad powers to confirm, reject or vary it.21
From the Board there is an appeal to the Land Tribunal22 and from the Land
Tribunal to the High Court.23 As in Kenya under the Land Adjudication Act,
the determination and recording of customary land rights, including the settlement
of any disputes that may arise, is entrusted largely to administrative bodies rather
than to the courts. An important difference, though, is the central role given to
"elders" in Kenya (rather appointed officers).24
Of particular interest are the provisions designed to protect those who have
customary rights in the land that do not amount to ownership. Research has
shown that the Kenyan land adjudication programme concentrated solely on
determining which individual should be recorded as the owner of a particular
plot of land (almost invariably the household-head (male)). Little attempt was
made to record and protect the customary rights of women, children and others,
and these were consequently put at risk.25 The Land Act seeks to avoid this by
requiring the Committee to record third-party rights over customary land
(including rights of occupation or use)26 and specifically to "safeguard the interests
and rights in the land ... of women, absent persons, minors and persons with
The Act recognizes that there are areas of the country where the recording
of individual ownership of land (whether customary or freehold) would not be
appropriate, that is, where land is communally owned and managed. The Act
therefore makes provision for the formation of Communal Land Associations
under the overall supervision of the District Land Registrar.35 Where not less
than 60 per cent of the members of a group wish to incorporate themselves as
27 S. 6(1)(g). The original Bill did not set out the functions of the Committee in this detail. It
made no mention of third-party rights and it did not contain the equivalent of section 6(1)(g). Neither
the Bill nor the Act require the matrimonial home to be registered in the joint names of the spouses,
as had been demanded by women's groups.
28 S. 8(3).
29 S. 9(1), (2).
30 S. 69.
31 S. 9(4).
32 S. 10. Section 11 allows any person to apply to be granted land in freehold. The procedure is
the same.
33 S. 13 and 14. The procedure is simplified where the applicant already possesses a certificate
of customary ownership of the land. However, whereas an applicant for such a certificate need only
show that s/he has exercised rights over the land which "should be recognised as ownership of that
land" (s. 6(1)(d)), in an application for freehold title the Committee must consider whether the
applicable customary law "recognises or provides for individual ownership of land" (s. 12(2)). The
difference in wording is presumably intended to be significant, though the significance may well be
ignored by the Land Committees.
34 S. 14(4) and (6)(b), 15(3).
35 S. 16. The original Bill contained a single section providing for communal land-holding; it said
nothing about Communal Land Associations.
an Association, they must elect 3-9 persons to be the officers of the Association.36
The officers must then prepare a constitution for approval by the majority of
the members37 and, once this is done, they must apply for incorporation as a
"Managing Committee".38 The Committee holds the land and may deal with
the land on behalf of the Association members,39 and it may set aside areas of
land for common use to be managed in accordance with a "common land
management scheme".40 Provision is also made for an individual or family who
is occupying and using a part of the communal land in accordance with customary
law, to apply to have that part excised from the communal land and recorded/
registered in the name of the individual or family head as customary/freehold
owner.41 The carving-up of the communal land into individual farms would, of
course, eventually lead to the dissolution of the Association.
LAND ADMINISTRATION
36 S. 17.
37 S. 18.
38 S. 19.
39 S. 20.
40 S. 24(1). The ingredients of such schemes as well as the duties and rights of members in relation
to such schemes are set out in some detail in the Act. The most common type of scheme is likely
to be for the grazing and watering of livestock and such a scheme would be expected to contain
restrictions designed to maintain or increase the carrying capacity of the land. Similar schemes to
preserve the rangelands of Tanzania (under the Range Development and Management Act) and of
Kenya (under the Land (Group Representatives) Act) were not very successful.
41 S. 23.
42 S. 47 and 50. The Act makes provision for the Commission's membership, its meetings and
its powers.
43 S. 57. In addition, section 60(6) requires every District Council to have a District Land Office
containing the offices of the Physical Planner, the Land Officer, the Valuer, the Surveyor and the
Registrar of Titles.
44 S. 65(1).
45 S. 75.
46 S. 77.
CRITIQUE
The reform, indeed the upgrading, of customary land tenure lies at the heart
of government land policy. For the first time, those with customary land rights
are to be given security of tenure in the form of a documentary title, whether
customary or freehold. However, unlike what occurred in neighbouring Kenya
where the adjudication and registration of titles was conducted on a systematic
area-by-area basis, in Uganda the Act merely provides a framework within which
an individual or group may apply for a certificate of title. Such an approach
reflects the new orthodoxy which recognizes that the systematic conversion of
titles (a "replacement" strategy) is not only extremely costly but is likely to be
ineffective unless accompanied by other reforms of the rural economy. Where,
as in Uganda, the majority feel secure with customary rights of access to land,
the incremental adaptation of customary land tenure within a suitable institutional
and legal framework may be more appropriate.49 Registration of titles should be
gradual, demand-led and funded (in part, at least) by the applicant.50 The
argument is a persuasive one. Population densities and types of land use vary
considerably within the country, from the extensive rangelands of Karamoja to
the tiny fragmented smallholdings of Kigezi, and demands for enhanced security
of tenure may vary too. A Kenya-style frontal programme would be very
expensive and would require a considerable investment in technical staff such
as surveyors, registrars and other officials. On the other hand, in densely
populated areas affected by land fragmentation it would be wasteful of resources
to leave the adjudication of titles to individual initiative; it would only make
sense for a large area to be surveyed and adjudicated as a whole. Moreover,
where the grant of certificates of title is based on individual applications, there
is always a risk of land-grabbing, that is, that an applicant may lay claim to a
larger area of land than that to which he is customarily entitled. It may be that
a flexible approach, allowing for both systematic and voluntary adjudication and
making use of pilot schemes, is called for.
47 S. 89(2).
48 S. 88.
49 SeeJohn Bruce and others, "Land and natural resource tenure on the Horn of Africa: synthesis
of African trends and issues raised by land tenure country profiles of East Africa countries, 1996",
in Workshop Report on Land Tenure Issues in Natural Resources Management in the Anglophone East Africa with
afocus on the IGAD Region, Addis Ababa, 11-15 March, 1996, unpublished.
50 M. Marquardt, "Access to land and other natural resources: Research and Policy Development
Project Uganda", ibid. He argues that registration should be related to the commercialization of
agriculture and the need to secure investment.
A novel feature of the Act is that it gives an individual or group the option
of applying either for a certificate of customary ownership or for a freehold title,
and it provides a straightforward procedure for converting the former into the
latter. Given that it is government policy that, in the long term at least, all land
should be held on freehold tenure, the question arises whether it would not have
been more sensible simply to have provided for freehold applications. The
argument for a two-stage process (first register customary titles, then convert
them to freehold titles) would have been stronger if the incidents of the two
types of tenure had been markedly different, but that is not the case. It is true
that the general law applies to freehold land, whereas customary law continues
to apply to land subject to a certificate of customary ownership, yet in two
important aspects the incidents of the two types of tenure are very similar. One
is the basic requirement that dispositions of both types of land should be recorded;
in the case of freehold land they should be registered under the Registration of
Titles Act and in the case of customary land they should be registered by the
Recorder.51 The other is that both customary and freehold titles will be registered
subject to third party rights, usually customary rights of a family nature. The
mechanism for protecting such rights will depend on the form of tenure, but in
either system they will constitute a major restriction on the freedom of the title-
holder (whether customary or freehold) to deal with the land.52
Assuming that the existence of two parallel registration systems (one for
customary titles and one for freehold titles) can be justified, the question still
arises as to whether the registration programme introduced by the Land Act is
likely to achieve its objectives. It is certain that the Act will go a long way
towards clarifying land rights in Uganda and providing security of tenure for
those in occupation of land where previously they were at risk of eviction by
government or by mailo "owners" or by government lessees. However, the grant
of a documentary title to land, particularly a freehold title, is also intended to
promote development by giving the titleholder an incentive to invest in his or
her land and the opportunity to raise credit on the security of the land. In
addition, it is likely to stimulate the market in land and to ensure, the argument
goes, that land is in the hands of those best able to develop it. It was considerations
of this kind that lay behind the land adjudication and registration programme
in Kenya which resulted in the abolition of customary land tenure and the
registration of household-heads as "absolute owners" of their land.53 Moreover,
it was these considerations that informed the recommendations of the University
of Wisconsin Land Tenure Center 1989 report that Uganda should move towards
51 Note, in particular, section 9(4) which provides that no lease, mortgage or sale shall have the
effect of passing any interest in the land unless it is registered by the Recorder.
52 See ss. 9(1) and (2), 15(3), mentioned above.
53 The well-known Swynnerton Plan proposed that "the African farmer... be provided with such
security of tenure through an indefeasible title as will encourage him to invest his labour and profits
into the development of his farm and as will enable him to offer it as security against financial
credits." R. J. M. Swynnerton, A Plan to Intensif the Development of African Agriculture in Kenya, 1954, s.
13.
a land tenure system based on freehold titles free from development conditions.54
Supporting this view, Marquardt claims that research points to the desirability
of moving to open land markets in Uganda and that, rather than causing
landlessness, the move may result in a more equitable distribution of land.55 In
particular, he found that the majority of women landowners had acquired their
land by purchase, leading him to conclude that a functional land market may
be beneficial rather than detrimental to women.56
The government of Uganda has adopted this policy in so far as the Land Act
sets out procedures for the move from customary to freehold tenure and imposes
no development conditions on freehold (or customary) titles.57 However, the
alleged benefits of titling are likely to be undermined by the Act's insistence on
the protection of third-party, essentially customary, land rights. It is, of course,
understandable that the government would yield to pressure to protect the rights
of spouses, children and other household members rather than putting them at
risk, as had occurred in Kenya, by failing to enter them on the register. It is
obvious, though, that the greater the number of rights or restrictions entered
against a person's title, the less negotiable that title becomes. Under the Land
Act the number could be considerable.
Some of the provisions have already been mentioned. The procedure relating
to the issue of certificates of customary ownership clearly envisages that such
certificates should be subject to such restrictions as are necessary to give effect
to customary entitlements (those of women, minors, absentees and persons under
a disability are specifically mentioned) and that an owner's freedom to deal with
the land should be circumscribed by such restrictions. Similarly, where a person
applies for the grant of a freehold title or for the conversion of a customary title
to a freehold title, third-party rights are to be protected on the land register. It
is not clear what form this protection will take. If they are to be substantively
registered, the Registrar will face the difficult task of finding in the general law
an equivalent for a customary right. Alternatively, the Registrar may simply
enter a restriction prohibiting dealings with the land without the consent of
certain named persons. In either case, given the potentially large number of
54 As mentioned above, the government of Uganda embraced the freehold option in the 1995
Constitution. So the question hardly arose in the debates on the Land Bill. It has been argued,
however, that Ugandan farmers do not need freehold titles as long as they have security of tenure.
See Nyangabyaki Bazaara, "Land reforms and agrarian structure in Uganda; retrospect and prospect",
(1995) 34/35 Nomadic Peoples 37. It has also been argued that government attempts (e.g. in Kenya
and Tanzania) to intervene in indigenous land tenure arrangements are generally misguided.
"Governments can best devote their resources to other uses, while standing ready to intervene to
the minimum extent possible ... where some tenure reform is required." T. C. Pinckney and P. K.
Kimuyu, "Land tenure reform in Africa: good, bad or unimportant?", (1994) 3(1) Journal of African
Economies 1 at 26.
55 M. Marquardt, op. cit. At the same workshop Professor Nsibambi also argued in favour of an
open land market. A. R. Nsibambi, "Land tenure relations in Uganda 1900-1995", ibid.
56 Ambreena Manji also argues that women are likely to benefit from individual titling. "In
Tanzania individualisation, registration and titling on the Kenyan model is more likely to be the
means by which the State provides women with land rights." Ambreena Manji, "Gender and the
politics of the land reform process in Tanzania", (1998) 36(4) Journal of Modern African Studies 645 at
666. Government policy in Tanzania has, of course, been based on the premise that marketability
of land increases insecurity of tenure.
57 Contrast policies adopted elsewhere in sub-Saharan Africa, e.g. in Tanzania where development
conditions are attached to granted rights of occupancy and in Zambia where they are attached to
the grant of government leaseholds. The Land Act is not concerned with land use except in relation
to common land.
58 A word like "action" might be better than "decision", though "decision" is apt to include a
determination of the Land Committee, Land Board or Land Tribunal. The expression "access to"
is unclear. The location of the section in the statute, wedged between a series of sections dealing
with Communal Land Associations and a long section dealing with the conversion of leaseholds, is
odd. In the original Land Bill the fore-runner of this section gave no protection to persons with
disability and it omitted the reference to unconstitutional conditions.
59 For example, article 34(7) ("The law shall accord special protection to orphans and other
vulnerable children") and article 35(1) ("Persons with disabilities have a right to respect and human
dignity and the State and society shall take appropriate measures to ensure that they realise their
true potential").
60 However, there is some evidence that land control boards have been concerned to consult the
owner's family where the effect of the transaction would be to leave them with insufficient or no
land. See S. F. R. Coldham, "Land control in Kenya", [1978] J1A.L. 63.
61 During the period of British rule only a very small area of land was held on freehold or
government leasehold titles, and these were converted to statutory rights of occupancy in the 1960s.
When public land was granted, whether to Africans or non-Africans, it was granted on a statutory
right of occupancy for a term and subject to conditions. At independence around 98 per cent of
land was held according to customary tenure on a customary right of occupancy. The general
picture remains the same today.
62 See Part VI generally. The provisions relating to the need to obtain the consent of the
Commissioner for Lands for the disposition of a right of occupancy are complex. Of particular
interest in the present context is section 41(l)(e) which requires the Commissioner to have regard
for the interests of groups at risk such as displaced persons, children and low income persons.
63 Interestingly, the Draft Bill contained a whole Part (60 sections) devoted to the formalization
of land tenure on village lands. It proved too controversial, however, and that Part was omitted
from the final version.
The Ugandan land policy charts a rather curious course between those of its
East African neighbours. It espouses a move towards freehold titles (like Kenya),
it shies away from an open land market (like Tanzania),64 and (unlike both
Kenya and Tanzania) it rejects the imposition of any administrative consent
requirement.65 While its rejection of a consent requirement was justified, it has
been argued here that, in so far as the registration of titles is intended to increase
the marketability of land, that objective is likely to be undermined by restrictions
protecting customary rights.
Implementation
CONCLUSIONS
The Ugandan Land Act is important, not simply because it ushers in a new
era in land relations in Uganda, but because it embodies new ideas on land
reform in Africa, and while it seeks to provide answers to problems that are
specific to Uganda, the nature of its successes and failures will be of general
interest, just as the land reform policies of Kenya and Tanzania have spawned
a vast literature. Although the last few years have seen widespread discussions
of land matters in Uganda and although the Land Bill went through a number
of drafts,67 it would be unrealistic to suggest that the Act represents a consensus.
It is not simply that the Act may be seen as a victory for some groups and a
defeat for others, but there were a number of policy issues on which opinion
would always be divided. The Act has introduced a legal framework governing
land tenure, land administration and the settlement of land disputes. It has
64 Among the fundamental principles of land policy set out in section 3(1) of the Tanzanian Act
are the need both to facilitate the operation of a market in land and to regulate the operation of a
market in land so as to ensure that rural and urban small-holders and pastoralists are not
disadvantaged.
65 For a discussion of the free market approach versus public law regulatory approach see
P. McAuslan, op. cit., at 537ff.
66 One writer has identified corruption as a major source of conflict in land matters in Uganda.
See A. Nsibambi, "The land question and conflict", in K. Rupesinghe (ed.), Conflict Resolution in
Uganda (International Peace Research Institute), Oslo, 1989, at 239.
67 This, as well as the need to ensure enactment by 2 July, 1998, may account for some
unsatisfactory drafting, for the "scissors and paste" feel of a few provisions.
introduced certainty into an area of law for a long time beset by uncertainty
and conflict. Its detailed provisions regarding land tenure and administration
should ensure certainty in the future. It provides security of tenure for customary
occupants of land (both mailo and non-mailo land). It decentralizes land ad-
ministration and establishes specialized land courts. These are all desirable
outcomes. However, doubts have been expressed above about some of the other
reforms, about whether registration of title should be left solely to individual
initiative, about whether there is a need for both customary and freehold titles,
about whether customary restrictions on titles will not adversely affect the
marketability of land, and about whether sufficient has yet been done in terms
of public education and the training of officials for the Act to be capable of
implementation in the near future.