Land Scarcity, Distribution and Conflict in Rwanda

50 Scarcity and Surfeit
Land Scarcity, Distribution and
Conflict in Rwanda
Jean Bigagaza, Carolyne Abong and Cecile Mukarubuga
Introduction
Rwanda is a small country of eight million people in central Africa, with a
long history of violent conflict dating back to 1959, and culminating in the
1994 genocide. Conflict in Rwanda has created a large refugee population in
neighbouring countries, with Uganda and Tanzania being the largest refugee
recipients before the 1994 genocide, and the Democratic Republic of Congo
(DRC) receiving the largest number of refugees after the genocide. As this
study will show, the Rwanda refugee population has had a destabilising
effect on the entire Great Lakes region, including on Rwanda itself.
This chapter examines the relationship between land scarcity and conflict
in Rwanda. Historically, land pressure has been a severe problem in Rwanda,
where over 90% of the population practises agriculture. Land pressure has
resulted in declining overall agricultural production, but increasing production
for individuals and groups with favourable land and resource access.
Cultivation is encroaching into wetlands, national parks and forest reserve
areas to satisfy unmet demands for land by some, predominately underprivileged,
groups. Large numbers of internally displaced persons have worsened
stress in some ecologically sensitive areas, such as in forests, resulting in
localised degradation of forest resources.
We will assess the power dynamics in Rwanda insofar as power through
control of the state is essential to control land. We will demonstrate that elite
power struggles for control of the state links land and conflict in Rwanda
where, historically, control of the state is the principal factor in rights to access,
use and ownership of land. While many analyses focus on linkages between
conflict and ethnicity, less attention has been focused on the role of land scarcity
in the Rwanda conflict. A thoughtful analysis of the so-called ethnic conflict
in Rwanda will show ethnicity is a cover for competition to control scarce land.
Indeed, this study argues that the causes of conflicts in Rwanda lie in competition
to access and control scarce land.1 We do not imply that land scarcity is
the ultimate or most important root cause of the Rwandan conflict. It is, however,
a critical component of the complex and intertwined causal factors.
Since 1980 powerful economic, political and social grievances in Rwanda
Chapter Two
relate to land scarcity. Overpopulation as well as inequitable distribution of
land worsened land scarcity for the rural poor. Increasingly, political power
and representation by elite groups at the national level determined control of
land. Widespread disinheritance of land rights of the rural poor coupled with
resource capture by elite groups has been closely related to deepening rural
poverty in the 1980s and 1990s.2 Deepening rural poverty, in effect, led to
violent conflict.3 Ethnic differences were less important in understanding the
dynamics of the conflict than were elite competitions to dominate critical
environmental decision-making processes through control of the state. In
turn, elite groups characterised these competitions in ethnic terms. Over
time, different groups in the conflict were polarised along ethnic lines and
were purposefully driven to conflict through ideologies propagated through
official media. Gasana observes that the rural poor (both Hutu and Tutsi)
described the ruling elite collectively as abaryi (eaters) who to them represented
a new, exploitative ethnic category.4 It can therefore be argued that
the conflict in Rwanda was ultimately a struggle against inequitable distribution
of land, tragically fought along ethnic, Tutsi versus Hutu, lines.
Recent studies of the Rwandan conflict have come to appreciate and
acknowledge the role played by ecological scarcity.5 Many studies, however,
still focus intently on the role of ethnicity in the conflict. Homer-Dixon has
described Rwanda as a country with severe demographic stress in the period
leading to the genocide.6 In 1991, Rwanda had an estimated population of 7.5
million and a growth rate of 3.3% annually. It had the highest population
density in Africa at 271 persons per square kilometre, and between 400 and
800 persons for arable land depending on the prefecture.7 Ninety-five percent
of the overall population inhabits 43% of the total cultivated land. The population
density in the rural areas is up to 843 persons per square kilometre. 8
Land contributed to conflict in the following two ways. The first is population
pressure leading to competition for scarce land; the second was the
inequitable distribution of land, most of which was controlled by elite
groups. Having established the links between land scarcity and conflict in
Rwanda, this study uncovers a number of lessons for other countries faced
with volatile land issues. These lessons may inform prevention strategies,
such as the development of early warning and early response systems.
The failure of the Arusha peace process for Rwanda and of the United
Nations intervention force in Rwanda (UNAMIR) to manage the Rwandan
conflict effectively suggests that there was a widespread failure to account
for the root causes of the conflict. This study demonstrates the importance of
including land and resource considerations in conflict prevention and management
policies and processes. The Arusha peace process failed in part
because it focused on the ethnic dimension of the conflict while ignoring
deeper and more critical issues relating to land and resource rights. The lessons
for conflict analysts and policy makers are many.
52 Scarcity and Surfeit
Background to the Conflict
Prior to the arrival of Belgian colonisers, Rwandan society had developed
over centuries into a remarkably organised state, with a high degree of centralised
authoritarian social control.9 King Rwabugiri, the former Tutsi King
of Rwanda, ruled the formerly semi-autonomous Tutsi and Hutu lineages
harshly, confiscating their lands and breaking their political power. The king
entrenched inequalities through the spread of uburetwa, a feudal system
under which poor Hutu farmers exchanged labour for access to land owned
mainly by Tutsi. As a result, polarisation and politicisation of ethnic groups
started. Land became a factor of differentiation between Hutus and Tutsis.10
There is a long history underlying the relationship between land and politics
in Rwanda. Land was used during the colonial era to divide the Rwandan
population along ethnic lines. When Belgian colonisers came to Rwanda they
favoured the Tutsi for administration, in effect establishing a governing class of
mainly Tutsi. They adopted the indirect rule system that enabled Belgium to
extract more taxes and labour from small farmers, mostly Hutu. Belgian
colonisers justified their preferential treatment of the Tutsi by relying on racist
ideologies. The Tutsi governing class, meanwhile, exploited their authority by
seizing cattle and land from other Tutsi and Hutu peasants. King Rwabugiri
also used land to increase tension between the Hutu and Tutsi. During and
after the period of colonial rule, the governing class in Rwanda once again
used land to polarise the Hutu and Tutsi ethnically. Insecure rights to land and
resources for the rural poor were mobilised for political gain.
A number of analyses have traced the historical background to the
Rwandan conflict, and in particular, the 1994 genocide which saw the death
of an estimated 800,000 Rwandans mainly Tutsi and moderate Hutus.11 Many
of these works link the civil war and the genocide to the colonial period.
There is a widespread belief that German and later Belgian colonialists
reinforced divisions between Hutu, Tutsi and Twa ethnic groups. This was in
part the result of a racialist colonial perception that viewed Tutsi as superior
to other groups, including Hutu. The Tutsi were treated preferentially by
Belgian colonial authorities. This, consequently, strengthened Tutsi hegemony
over the Hutu. Historically, Tutsi and Hutu identities were not clearly
defined. The terms Hutu and Tutsi appear to have originally been flexible in
that a man could be Tutsi in relation to his clients or inferiors, and Hutu in
relation to his patrons or superiors. It was possible for those born Hutu or
Twa to be ennobled to hold elite positions thus becoming Tutsi. Colonialists,
however, by favouring Tutsi on the basis of racialist ideology, reinforced
ethnic divisions. These differences were reinforced by the introduction of
compulsory identity cards in 1931, which indicated ethnicity.
At this point, Rwandans began to relate more to their respective ethnic
groups, which would be critical to determining access to political representation
Land Scarcity, Distribution and Conflict in Rwanda 53
and access to resources. Colonialism thus sharpened the differences between
Tutsi and Hutu. At independence the new government continued the use of
identity cards. These would be used later to identify Tutsi during the genocide.
According to Prunier12, the Republic of Rwanda, created at the end of colonial
rule in 1962, was ethnically divided. Others note the long history of political and
economic rivalries between the Hutu and Tutsi ethnic groups that predate the
contemporary conflict. This rivalry found expression in the periodic outbreaks of
violence leading to regular surges of large refugee outflow.
The first major conflict in the history of Rwanda was the 1959 so-called
‘Hutu revolution’ against Tutsi hegemony. The ethnic animosity and Hutu
discontent created in the colonial period was catalysed and Hutu chiefs
organised the killing of rich and poor Tutsi. The abrupt shift in Belgian policy,
and the role played by the Catholic Church in empowering the Hutu
against the Tutsi, paved the way for this revolution. The 1959 ‘social revolution’
marked a period during which the Tutsi were excluded from participating
in political and economic processes in Rwanda. Though known as a Hutu
‘revolution’, only a minority Hutu elite were to benefit. Poor Hutu were largely
excluded from national political and economic processes, as well. Thus,
the social revolution substituted one elite group for another.
The majority of the rural poor, both Hutu and Tutsi, remained outside the
realm of official politics and the formal economy. Education and employment
opportunities and positions in the military were reserved for a small Hutu
elite from the north. Thus in the 1990s, when power and access to resources
was concentrated in the hands of the northern elite, a pervading sense of
frustration with formal politics and economy, and its inability to ensure
livelihood security for most groups, ignited conflict. Ethnic divisions, therefore,
were not the cause of the conflict. Instead, these were the result of political
manipulation by a powerful ruling elite.
The 1959 conflict saw a mass exodus of Tutsi refugees into neighbouring
countries, especially Uganda13. Tutsi refugees in Uganda reorganised themselves
and in 1963 launched the first military invasion into Rwanda in an
attempt to capture the state. The invasion was unsuccessful and resulted in
widespread killings of Tutsi and accelerated flows of refugees into neighbouring
countries.14
The period between 1963 and the 1990 civil war was one of uncertainty,
marked by Tutsi armed incursions into Rwanda and ethnic cleansing of Tutsi
inside Rwanda. The post-independence period was thus marked by ethnic
violence between Hutu and Tutsi and Tutsi refugee outflows mainly to
Tanzania and Uganda. The subsequent Hutu governments fostered and
manipulated ethnic divisions to maintain a popular rural support base.
In 1990, the Rwanda Patriotic Front (RPF) organised an armed invasion
into Rwanda, from Uganda. Predominantly Tutsi, many of the RPF members
were refugees or children of refugees driven out of Rwanda in the aftermath
54 Scarcity and Surfeit
of the 1959 conflict. 1990 thus marked the beginning of a civil war that by
1992 had displaced one-tenth of the population and widely disrupted agricultural
activities.15 Even after the signing of the Arusha Peace Accords in
August 1993, hostilities continued. During this period, radical Hutu plotted
the genocide with the support of the government.
Tensions reached a climax on April 6, 1994, when the plane carrying
Rwandan President Juvenal Habyarimana and his Burundian counterpart,
President Cyprien Ntaryamira was shot down, killing them both and marking
the beginning of the genocide. Fighting between the RPF and the Rwandan
defence forces (Forces Armées Rwandaises or FAR) escalated and within a
period of 100 days an estimated 800,000 Tutsi and Hutu moderates were
killed by members of a radical Hutu militia group.16 The RPF emerged victorious
in July 1994 and leads the government in Rwanda to this day. However,
full peace and national reconciliation is still elusive. Persistent armed incursions
into Rwanda that are organised by members of FAR and the Hutu militia
group responsible for the genocide (the Interahamwe, now operating
under its French acronym PALIR, or the Peuple en Armes Pour Liberer le
Rwanda) continue.
In the post-genocide period, the precarious refugee situation continues to
be a source of insecurity and instability to Rwanda. The government of the
neighbouring Democratic Republic of Congo (DRC), where many ex-FAR
servicemen and Hutu militia sought refuge, has either been unwilling or
unable to disarm them, sometimes seeing them as allies in a common struggle
against Rwandese occupation.17 Rwanda justifies the presence of its occupying
forces in the DRC on the grounds of the continued presence of FAR and
PALIR cells inside the DRC. Rwanda feels that its national security is threatened
as long as the PALIR continues to operate from inside the DRC.
Rwandan President Paul Kagame has stressed that Rwanda’s DRC policy is to
contain PALIR military activities.
Key Factors
Although the importance of ethnicity should not be overestimated, it is still
important to trace the use of ethnicity in conflict discourse. In Rwanda, historical
narratives of social injustice were employed by Hutu extremists prior
to the genocide in 1994 to justify their exclusive claim to power.
Prunier18 analyses the role of colonialism in creating a cultural mythology
that informed the ideology and actions of Hutu and Tutsi, and the role of this
mythology in causing conflict. To Prunier, although conflict in Rwanda can be
viewed as a power struggle between Hutus and Tutsi dating to colonialism,
ethnicity is not necessarily the most important factor to understanding the
conflict. He cautions against dividing the Rwandan society into Hutu, Tutsi
Land Scarcity, Distribution and Conflict in Rwanda 55
and Twa tribes. As Prunier explains, they share the same language and culture
and, historically, coexist without separate tribal homelands.19
In Rwanda, there are no ethnic distinctions in terms of language, culture
and religion. According to Vidal, it would be simplistic to understand the
Rwandan conflict as a manifestation of ethnic differences. Like Prunier, Vidal
sees ethnicity in Rwanda as a colonial creation, but one that was exacerbated
and manipulated by extremist politicians to maintain popular support and
control of the state. Reyntjens disagrees, and claims that ethnicity always
existed in Rwanda and is to blame for the 1994 genocide.20
It is informative to note the patterns of ethnic division in Rwanda, and the
contexts in which ethnic divisions are raised in relation to specific political
and economic aims, including, importantly, control of the state. Conflict in
Rwanda to control the state illustrates the importance of state institutions and
other decision-making structures, for instance, to determine land and
resource rights. Lee observes that such conflicts have been common throughout
human history: “Communities have fought to preserve access to scarce
resources or to prevent another from gaining such access.”21
During the Habyarimana regime, power was concentrated in the hands of
the akazu22. The akazu mainly came from the northern prefecture of Gisenyi
and supplied a third of top government jobs and almost all heads of security
apparati. They also benefited disproportionately from state development projects.
23 The akazu dominated the state and maintained virtual exclusive control
over Rwanda’s land and resources using the laws and institutions of the
state. Thus in Rwanda, the question of who controlled which decision-making
structures and processes to decide ownership of what land and resources
and for which groups was the key issue underlying conflict leading up to civil
war and genocide. Conflict to capture the state was simply the means used to
gain or maintain control of scarce land and natural resources.
In viewing conflict in Rwanda through the lens of state control, it must be
critically questioned how arguments of ethnic difference were used to support
the war and genocide. Storey24 lists the following strategies adopted by
the akazu to deal with challenges to their domination of the state and control
of land and natural resources:
• mass propaganda that blamed the Tutsi minority for poverty, famine and
general economic hardship;
• violence against opposition figures; and
• genocide.
The international advocacy group Human Rights Watch has documented incidences
of ethnic incitement of rural populations by elite groups.25 In the 1980s
and 90s, the gap between the rulers and the ruled (both Hutu and Tutsi) was
clearly widening and poverty was at a record peak. But with the advent of
multiparty democracy, and the Arusha Accords, electoral competition and
56 Scarcity and Surfeit
forced government concessions to the RPF threatened to deprive the akazu of
its control of the state machinery. Therefore although both poor rural Hutu
and Tutsi were denied secure rights to land and resources, the akazu popularised
a view that predatory Tutsi were the source of deprivation among Hutu
peasants, and were accomplices of the rebel Rwanda Patriotic Front that was
seeking to overthrow the Hutu dominated government.
The akazu regime, faced with widespread internal and external opposition,
manipulated ethnic differences to incite violence and genocide in order to weaken
Tutsi-led opposition.26 In zero-sum competitions to control scarce land and
resources, where the gain of one group implies the loss of another, ethnicity is a
convenient guise for elite competition. Ultimately, grievances over access to and
control of scarce land and resources assumed an ethnic orientation in Rwanda.
The multiple effects of economic decline, population pressure, structural
adjustment policies (SAPs) and growing internal opposition weakened the
government’s legitimacy and its administrative ability, thus contributing to
conflict in Rwanda. Declining government revenues owing to the sharp fall in
world coffee prices, Rwanda’s main export earner, caused the economic
slump of the 1980s. Together with SAPs, the already exhausted economy was
weakened further. The rural poor were hardest hit.
Collins,27 Chossudovsky28 and Karnik29 attribute the weakness of the
Rwandan state to the policies of the World Bank and IMF. They observe that
SAPs imposed on Rwanda by international lenders in the early 1990s
destroyed economic activity and rural livelihoods. Ruling elites, however,
were spared devastating effects by passing the cost of structural adjustments
onto the poor. SAPs also fuelled unemployment, which was already at critically
high levels and created a situation of general social despair.
Chossudovsky claims that it was the general impoverishment of the population
that contributed to desperation, insecurity and violence.30 These factors
combined with the effects of the 1988–89 drought to induce ever higher levels
of stress among the rural poor. The rural poor, in response, devised a variety
of strategies to strengthen their livelihoods, including complicity in genocide
motivated by incitement from elite political leaders.
These crises prompted the 1990 RPF invasion. According to former
Rwandan President Pasteur Bizimingu, the Rwandan political system was on
the verge of collapse owing to this situation, and any push from outside would
only have completed the process of its collapse.31 However, with growing internal
demands for democratisation, and with the government threatened by
these crises as well as an imminent RPF invasion from Uganda, extremist political
parties, notably the Coalition Pour la Defence de la Republique (CDR) and
part of the Mouvement Republicaine pour la Democratie et le Developpment
(MRND) began systematic campaigns and attacks against the Tutsi.
Refugees have been and continue to be a source of conflict in Rwanda. The
conflicts of 1959 and 1963 forced a number of Tutsi into exile. These refugees
Land Scarcity, Distribution and Conflict in Rwanda 57
were denied the right of return by the Habyarimana government. Towards the
end of the 1980s, there were about 480 000 refugees (representing about 7% of
the population and almost half the entire Tutsi population) seeking to return to
Rwanda, but who were denied this by the government. Refugees claiming the
right of their return to Rwanda posed a major challenge to the government,
which refused, officially because of population pressure and scarce land.32 The
right of refugee return was a key objective of the RPF, comprised predominantly
of exiled Tutsi. Tutsi refugees since independence had organised armed incursions
into Rwanda. These were followed by retaliatory attacks by the Rwandan
defence forces on Tutsi populations within Rwanda. The RPF invaded Rwanda
in 1990 in order, it claims, to ensure the return of refugees to Rwanda. Thus by
failing to deal with the refugee problem prior to 1990 and by denying their
return, the government set the stage for future civil war in Rwanda.
Internally Displaced Persons (IDPs) have also contributed to the conflict
in Rwanda by worsening land and resource scarcities in areas where they settled.
Thousands of IDPs were created as a result of the civil war in Rwanda
dating from 1990. Gasana claims that Byumba, Butare and Ruhengeri prefectures
hosted up to one million IDPs, severely straining scarce resources in
these areas.33 He sees a powerful interaction between scarcity of land and
resources and IDPs, which, according to him, aggravated the conflict.
Currently, Hutu refugees are an important source of instability in Rwanda
and the Great Lakes region in general. These are mainly ex-FAR soldiers and
Hutu militia who organise armed incursions into Rwanda from the DRC. It is
clear that as long as the refugee problem is not resolved, long term peace and
national reconciliation in Rwanda will be impossible. Thus in November
1996, the Rwandan government, frustrated by the growing number of armed
attacks from refugees based in the DRC, closed down refugee camps.
However, many of the refugees declined to return and retreated deeper into
the DRC, thus remaining a source of insecurity to Rwanda to this day.
The post-genocide massive return of Tutsi refugees (first caseload
refugees) has led to conflict over ownership of property and land. The ongoing
return of Hutu refugees (second caseload refugees) from closed refugee
camps in the DRC has placed an enormous strain on scarce resources and is
a source of conflict between first and second caseload refugees competing for
ownership of land and property. As a result Rwanda is faced with the monumental
task of resolving competing land and property claims.
Key Actors
By 1990, the Rwandan state had increasingly become authoritarian and
unpopular, especially among the rural peasantry, and was facing threats both
internally from opposition parties, and externally, from the RPF.34 It is argued
58 Scarcity and Surfeit
that extremist political parties within the Habyarimana government, struggling
for survival, saw conflict and the genocide as a last attempt to survive.35
The government and extremist political parties, notably the CDR and a section
of MRND, then became the most outspoken protagonists on the government’s
side. Senior and influential elites in the Habyarimana government
manipulated existing grievances, especially grievances concerning land
scarcity and economic hardships, to create a conflict pitting Hutus against
Tutsis to maintain political power amid growing internal opposition.
Rebel and militia groups constitute a second category of actors and are
predominantly composed of refugees, with the first caseload refugees made
up of Tutsis and the second caseload refugees made up predominantly of
Hutus. Prior to the 1994 RPF victory, Tutsi refugees forcefully attempted to
return to Rwanda, and in the process, destabilised social and political systems
in Rwanda. The RPF was composed of exiled and refugee Tutsi.
Currently, Hutu refugees, many whom are ex-FAR soldiers and members of
former paramilitary groups, have regrouped from bases in the DRC and
formed the new armed group, PALIR, whose goal is to capture the Rwandan
state by force. PALIR remains a dangerous and destabilising force to the government
of Rwanda, which is seeking reconstruction and national reconciliation
of the population. The Armée de Liberation du Rwanda (ALIR), the
armed wing of PALIR, continues to carry out armed attacks into north-west
Rwanda from bases inside the DRC. Violent conflict between the Rwandan
army, RPA, and Hutu militia groups continues today.
Local populations participated in the conflict in two ways. The first was
through direct armed support in the genocide and civil war. A second was to
assist armed opposition groups against the government, an opposition that
emanates from their grievances concerning economic marginalisation and
disinheritance of land and property. During the 1994 conflict, the rural peasantry
played a major role in terms of carrying out the genocide orders.36
Prunier observes that one of the incentives for the rural poor to be involved
in the conflict stems from land and resource scarcities. In his view, Hutu
peasants killed the Tutsi because they would inherit the land of the murdered
Tutsi.37 To the poor rural Hutu, inheritance of additional land and property
was a big incentive to participate in the genocide.
Conflict Management Strategies
The conflict in Rwanda received little international recognition for a long period.
This changed following the outbreak of civil war in 1990 and the massive
human rights violations accompanying the war, and the threat these posed to
regional peace and security in central Africa.38 Following the civil war that broke
out in 1990, the government and RPF entered into negotiations in Arusha,
Land Scarcity, Distribution and Conflict in Rwanda 59
Tanzania, with the aim of finding peaceful solution to the conflict. The negotiations,
initiated by the Organisation of African Unity (OAU) and facilitated by the
government of Tanzania, began in July 1992 and were concluded in August
1993. The Arusha peace process was the first major international response to
the war. The subsequent Arusha Accords provided for the following:
• a ceasefire;
• the formation of a broad-based transitional government with power sharing;
• incorporation of the RPA into the army; and
• return of all refugees.
In June 1993, just before the Arusha Accords were signed, the UN Security
Council established the United Nations Observer Mission for Uganda and
Rwanda (UNOMUR). The mandate of UNOMUR was to ensure that no military
assistance crossed the Uganda border into Rwanda.39 The United
Nations Assistance Mission for Rwanda (UNAMIR) was established after the
signing of the Arusha Accords and the establishment of a ceasefire between
the RPF and the Rwandan government. UNAMIR’s overall mandate was to
support implementation of the Accords as well as to protect humanitarian
organisations operating in Rwanda during the transitional period.
Extremist groups within the government, notably the CDR and a faction of
MRND, were opposed to the Arusha Accords, which they believed gave too
much power to the RPF, including key defence and health ministerial positions,
as well as large influence in the national armed forces.40 The governing
Hutu elite began to undermine the Accords with the support of extremist
Hutu elements. In particular, the extremist Hutu party (CDR) did not participate
in the peace process and was not included in the Accords, leading to
their armed opposition. At the time, the CDR was a major political force in
Rwanda. Adelman41 argues that the exclusion of the CDR from the Arusha
peace process was its major shortcoming. Only four days after the signing of
the peace agreement, ‘ethnic’ massacres began in Kibuye, for which extremist
Hutu groups were held responsible.42
The main lesson emerging from the failed Arusha peace process for Rwanda
was that zero-sum solutions that exclude some groups, no matter how strong,
are unworkable. Furthermore, the Arusha peace process focused more on the
ethnic parameters of the conflict while overlooking other important causes,
such as pervasive unequal land ownership, decreasing international value of
agricultural commodities and deepening rural poverty. However, effective
interventions in conflict are only possible when all sources of conflict are
recognised, and the dynamics between these understood.43 The Arusha peace
process was preoccupied with issues of power sharing, elections and the composition
of armed forces, all which were meant to diffuse ethnic sources of
conflict. Although these were important considerations, by not addressing
60 Scarcity and Surfeit
other sources of conflict, these considerations alone were not sufficient to manage
the conflict.
The UN assumed responsibility for overseeing the implementation of the
accords under UNAMIR, which had a broad mandate including peacekeeping,
humanitarian assistance and general support to the peace process during
the transitional period. However, the deployment of UNAMIR peacekeepers
was slow and they lacked the mandate to coerce combatants to observe the
Accords. The international community, meanwhile, was focused on the
apparent success of the peace process, and neglected to devise a contingency
plan to address the conflict should the ceasefire break (which it did).
The design of the intervention was traditional and did not consider the
nuances of the Rwandan situation, nor account for the rapidly shifting
dynamics of conflict and peace in the country. Events inside Rwanda were
widely misinterpreted by the international community and did not distinguish
between impending genocide and civil war. Hence when the genocide
began, it was considered part of the civil war. The international community
maintained its distance from events on the ground while reducing its mission
to a mere 270 observers. Apparently, the UN was basing its actions on the
Somalia experience, from which it recognised the need to maintain some
measure of neutrality. However, the conflict in Rwanda differed greatly from
Somalia.
After the genocide had begun, the UN authorised the deployment of
French troops as part of Operation Turquoise with the purpose to establish a
safe humanitarian zone for the protection of civilians. Operation Turquoise
was authorised under Chapter 7 of the UN charter to use force to achieve its
humanitarian objectives (unlike UNAMIR that was established under Chapter
6 and hence lacked enforcement powers). Even though it did manage to save
some lives in the humanitarian zones, and facilitated the flow of humanitarian
assistance to displaced populations, it did not succeed in halting further
hostilities and the massacre of civilians. It is argued that Operation Turquoise,
which was seen as more sensitive to francophone interests, protected the status
quo in Rwanda at the time, whereas the rebel RPF movement was viewed
as squarely anglophone.
In the post-genocide period, the UN explored methods of national reconciliation
and peace building with a focus on ending the culture of impunity
that was entrenched in Rwanda. The UN established the International
Criminal Tribunal for Rwanda (ICTR) in November 1994 to facilitate the
process of national reconciliation and reconstruction. The trial of suspected
perpetrators of the genocide is one of the mechanisms by which the international
community hopes to create conditions for justice and reconciliation
leading to peace building and stability in Rwanda. Also, by placing the
responsibility for human rights violations and other atrocities on specific
individuals rather than an ethnic group, it is hoped that the tribunal can help
Land Scarcity, Distribution and Conflict in Rwanda 61
to diffuse ethnic tensions, and deter such atrocities from being committed
again. Individual trials avoid laying blame for the genocide on Hutu, many of
whom were killed in the events of April and May 1994 as well.
The ICTR trials in Arusha are complemented by trials in Rwanda. Since
1999, revival of the gacaca traditional court system has been discussed, and
recently implemented, as an option to alleviate pressure on the overburdened
justice system in Rwanda. Gacaca courts will be held at the local level by
locally elected judges and have jurisdiction over the minor categories of
crimes committed during the genocide. It is too soon to judge the ability of
the gacaca courts to contribute to the process of reconciliation and justice.
However, it is clear that its success will depend on the ability of the ‘traditional’
judge to be distant enough from the genocide to pass fair judgments
that are considered legitimate and acceptable to all parties concerned.
The Rwandan justice system is overburdened with the responsibility of bringing
justice to those who survived the genocide. Many years after the genocide in
1994, several thousand suspects still languish in custody without trial or due
process of the law. In order to ease pressure on the justice system, it is proposed
that traditional gacaca courts headed by local elders be revived to deal with
minor cases. The courts will function from the cellule to the prefecture levels.
The cases to be tried would be those in Categories 2 and 3. Category 2 is for
those accused of overseeing massacres and for failing to prevent them when in
a position to do so. Category 3 is for those who killed or looted during the genocide.
Category 1 which would not be tried by the gacaca courts is for the planners
of the genocide. Those found guilty in gacaca courts will be able to appeal
to another court. Legislation to establish gacaca was adopted in October 2000.
The gacaca is controversial in Rwanda. Many, particularly Tutsi, question
whether gacaca will bring about intended justice and reconciliation in Rwanda.
Many Tutsi believe the gacaca are predisposed to leniency towards genocide
suspects. For example, the Rwandan Justice Minister, Jean de Dieu Mucyo,
claimed that community service could be an option for sentencing in gacaca.44
It is feared that revenge attacks by genocide survivors on those sentenced in
gacaca could develop, thus sabotaging the reconciliation it is aiming to achieve.
Another consideration is that the integrity and credibility of traditional
courts has rested on the people selected to fulfil the role of arbitrator or
judge. However, population displacements and resettlement after the genocide
imply that the structures of community life in most areas have been disrupted.
This would make it hard for the prospective gacaca judges to establish
credibility both with the accused and with the genocide survivors.
Credibility of the court and its decisions will be essential for real justice and
national reconciliation in Rwanda.
In contending with the destabilising refugee situation, the government,
with the help of the United Nations High Commissioner for Refugees
(UNHCR), has undertaken confidence-building measures to encourage the
62 Scarcity and Surfeit
safe return of refugees, including those engaged in the genocide. However,
the government of Rwanda has also moved to forcefully close refugee camps
within Rwanda and in the DRC. In any case, the safe return of refugees is
vital to the normalisation of political, economic and social orders in Rwanda,
and is key to the long-term stability of the country.
Prior to the genocide few non-governmental organisations and other civic
bodies operated in Rwanda due to the extensive control that the
Habyarimana government exercised over civil society. International organisations,
such as Oxfam, International Alert, and Medecins sans Frontieres,
maintained limited operations in Rwanda, primarily to observe human rights
and provide early warning information for conflict managers. Following the
1993 ceasefire brokered as part of the Arusha peace process, international
organisations monitoring the situation in Rwanda warned that extremist
forces linked to the Habyarimana government were arming and organising
themselves to derail the peace process.45 They also warned of human rights
violations, including extrajudicial executions of civilians.46 These warnings,
however, were largely ignored by the international community.
In contrast to the situation before the genocide there has been large-scale
humanitarian intervention in Rwanda by international organisations in the
post-genocide period.47 The international community, criticized for its failure
to intervene before the genocide, provided substantial emergency assistance
and reconstruction aid following the end of the civil war in 1994. It has supported
macro-economic reforms and the re-establishment of basic services in
Rwanda. Economic growth is widely seen as essential for long-term peace
and stability in Rwanda and throughout the Central Africa region. The IMF
and World Bank have supported a number of reforms in Rwanda to benefit
reconstruction, and the United Nations Development Programme (UNDP)
established a trust fund for Rwanda to channel aid for building the capacities
of institutions and the state.
There has also been a proliferation of non-governmental organisations in
Rwanda in recent years. These organisations are engaged in a wide range of
activities to support humanitarian assistance and peace building. Many local
organisations48 address issues such as social justice, human rights, and reconciliation
and rehabilitation of those traumatised by the conflict. Faith-led
initiatives are prominent among the manifold civic society initiatives started
in Rwanda.
Overview of the Environment
Land is the focus of ecological grievances in Rwanda. Rwanda is described as
a country with severe population pressure and widespread dependence on limited
resources for subsistence.49 According to Homer-Dixon, demand-induced
Land Scarcity, Distribution and Conflict in Rwanda 63
scarcity in Rwanda resulted from overpopulation and supply-induced scarcity
was caused by decline in soil fertility due to over-cultivation, degradation of
watersheds and the depletion of forests.50 The effects of ecological scarcity
were beginning to be felt in the 1980s when food production failed to keep
pace with population growth. The population was growing but there was little
or no land for agricultural expansion. Earlier expansions into forest reserves
had depleted wood fuel leading to scarcity.51
In the 1980s and the period preceding the drought, Rwanda moved from being
one of Africa’s top performers in agricultural production to one facing massive
food shortages. Homer-Dixon observes that although food output rose by 4.7%
annually between 1962–1982, a higher increase compared to population
growth, much of this was a result of expanding cropland areas and a reduction
of fallow periods rather than a result of improved agricultural methods, such
as the use of fertilizers.52 Hence by the late 1980s, most land, including steep
hillsides, was under cultivation. Soil fertility fell sharply and with a growing
population, per capita food production decreased as well. As a consequence,
the country began to face severe food shortages, more so in the southern prefectures
where internal opposition to the government grew.
The government of Rwanda received overseas development assistance to
undertake environmental and development projects to improve rural livelihoods
and alleviate poverty. International aid, however, was unevenly distributed,
with the bulk of it skewed towards President Habyarimana’s home
region in the north-west region of Rwanda. Because of this, disenchantment
with and resentment towards the government grew in southern prefectures
and encouraged the spread of internal opposition to the government. Threats
to the Habyarimana’s government emerged.
As described earlier, the rural poor bore the brunt of environmental scarcity
and were the most visibly affected. Many of these poor families, owing to
diminishing size of family land holdings, moved onto unproductive lands
that were threatened with massive soil erosion. Resource capture by elite
groups and population pressure led to unsustainable land use, such as cultivation
on steep hillsides, shortening of fallow periods and deforestation to
open additional land for farming. Food insecurity continued to grow up to the
time of the genocide. All the while, the rural poor protested against the
unequal distribution of land and resources.
Southern elites harnessed the ecological grievances of the rural poor for
their own political gains. They stressed that the government invested more in
rural development in the north while marginalising the south. When it was
clear that growing opposition to the government in the southern prefectures
was threatening the legitimacy of Habyarimana’s government, the RPF capitalised
on the opportunity and launched their invasion from Uganda to the
north. The governing Hutu elite, however, was able to cast the conflict as one
between Hutu and Tutsi, not between a marginalised majority and privileged
64 Scarcity and Surfeit
ethno-regional minority. In Gisenyi and Kibuye prefectures in the south, for
example, where violence took an ethnic form, elite Hutu exploited the grievances
of the poor landless and turned them against their Tutsi neighbours
while promising them the Tutsi lands in return.53
The civil war displaced up to one million Rwandans inside Rwanda. Many
of them settled in areas already confronted with ecological scarcity such as
the Ruhengeri and Butare prefectures, whereas others moved into sensitive
environments, such as steep hillsides. IDPs increased resource demands leading
to fierce competition between themselves and other local inhabitants.
Gasana argues that the dissatisfaction of the rural peasantry as well as the
grievances of the IDPs were channelled into anti-RPF and anti-Tutsi sentiments,
leading to conflict.
Grievances linked to soil erosion also emerged in the highlands of the
southern prefectures of Gisenyi and Gikongoro and parts of Butare, Byumba,
Cyangugu and Kibuye. These prefectures have acid soils that lack any regenerative
capacity when put under intense cropping and are highly erodible.
These prefectures all recorded negative rates of food production and hence
retained a permanent situation of food scarcity compared to the positive rates
recorded in the northern prefectures of Kibungo and Ruhengeri.
Land as a Source of Conflict
The role of land is crucial to understanding the civil war and genocide in
Rwanda. Several authors have asserted the link between land scarcity and conflict
in Rwanda. Land is the most important asset for most Rwandans and will
remain important for many years to come.54 Around 95% of the active population
derives its livelihood from the production of food crops. A common
understanding imparted in many analyses of the Rwandan conflict is that population
pressure leading to land scarcity was the ultimate cause of conflict in
Rwanda. However, it is clear that this view is limited, and that a number of
other factors interacted to cause conflict in Rwanda, as Olson55 notes. Rapid
population growth, soil degradation, low prices for agricultural produce, lack
of access to productive resources, unequal distribution of land, limited government
investment, and limited off-farm opportunities amounted to “production
pressure on a constrained resource”, in the words of Olson.56
Land Acquisition and Access
A 2001 survey carried out by the Ministry of Environmental Rehabilitation
and Protection examined several mechanisms used to acquire land in
Rwanda, including inheritance, government distribution, the market and by
donation.
Land Scarcity, Distribution and Conflict in Rwanda 65
Inheritance, historically, is the predominant mode of land acquisition for
Rwandans. Male children were entitled to inherit land and property from
their parents upon the parents’ death. However, a new law was recently
promulgated that gives the same rights of inheritance to girls.
A second mechanism was acquisition through government distribution.
Government authorities allocate land to landless peoples, including returning
refugees, in new settlement areas, including swamps and natural reserves.
Returning refugees are the primary beneficiaries of government land allocations
in the post-civil war and genocide period. However, land distributions by the
government are decreasing because there is no additional land for distribution.
Despite limitations in land laws and regulations, a small land market has
also developed in Rwanda. However, few have the ability to purchase land
on the market. There is a real risk that land distribution will become more
disparate as the rural poor are forced to sell their land due to poverty, and
other rural poor are financially unable to acquire new land.
Acquiring land through temporary lease is becoming more common owing
to the decreasing availability of land for cultivation. Leasing land is one strategy
that smallholders and landless farmers use to acquire land for cultivation.
They pay rent for each growing season to families with larger holdings
or to other poorer farmers who lack the means to farm.
There is a history of donating land to poor kin in Rwanda. However this
practice is becoming infrequent owing to growing land scarcity. Instead, relatives
of landless families may temporarily loan land for one or more growing
season.
The results of a survey carried out in 1988 (Table 1) show a change in the
mode of land acquisition. Purchases of land are increasing and are now the
primary mode of land acquisition. Formerly, allocation of land by the government
was the primary means through which men inherited land on behalf
of their families. However, the government is no longer able to allocate large
areas of land given that nearly all land in Rwanda is in use.
66 Scarcity and Surfeit
Table 1: Evolution of land acquisition in time (in % of plots)57
Mode of land acquisition
Land owned for more Land owned for less
than 25 years than 10 years
Purchase 2.4 20.4
Clearing 1.6 4.7
Inheritance 72.2 67.5
Donation 8.3 4.7
State allocation 15.5 2.6
According to customary land tenure systems in Rwanda, only men had the
right of access to land. Upon marriage, young women had to leave their families
to join their husbands who inherited land from their parents. In her new
family, a woman could not inherit her husband’s property rights; only men
were entitled to inherit landed properties. A woman could inherit land only
when she had neither male children nor living male relatives of her deceased
husband. However, the widow had the right to use her late husband’s land
as long as she stayed in her husband’s house and raised their children. The
fact that women did not have the same rights to land and property as men,
however, generally did not worsen conflict.
In Rwanda, although women perform more of the agricultural labour, they
have benefited less than men from social development. Women’s rights to
property and land are limited. Moreover, women have few education or
employment opportunities often because of persistent social stereotypes that
the role of women is exclusively domestic. Women in Rwanda played a central
role in the reconstruction of the country. After the genocide they formed
an exceptionally large proportion of the population and many had to head
their households. Women also assumed new roles in Rwandan society. Some
improvements to promote gender equality were made following the war and
genocide, such as the establishment of the Ministry of Gender and Social
Affairs. The government also plans to create a legal framework that recognises
women’s rights. The revision of the matrimonial code offers couples a
choice of property regimes, including the option to own land and property
equally. The labour code and land legislation will remove restrictions on
women’s ability to work and own property. There are also efforts to mainstream
gender in all policies and programmes, empower women through education,
targeted micro credit programmes and community safety nets.
New legislation enacted in November 1999 makes it clear that men and
women are equally entitled to inherit land from their parents. In the proposed
new comprehensive land law, women and men have equal land rights. When
this law is enacted, women will be able to inherit land and property from
their parents and husbands. Empowerment of women may have indirect benefits
for food security, as strong land and resource rights should enable
women to invest in greater agricultural production.
Demographics
Population pressure is an important factor contributing to land scarcity in
Rwanda. It is well known that Rwanda is the most densely populated country
in Africa (329 per square km, against 29 in sub-Saharan Africa in 1998).
In the 1980s the population density on arable lands was estimated at 390 persons
per square kilometre. In the intensively cultivated regions in southern
Rwanda, such as Butare, the population density was an estimated 400 to 500
Land Scarcity, Distribution and Conflict in Rwanda 67
persons per square kilometre. The 1991 census estimated the Rwandan population
to be 7.15 million and overall population density to be 271 persons
per square kilometre. This was the highest population density recorded in
Africa at the time.58
Furthermore, most arable land is under cultivation. In 1998, almost 70% of
the estimated 1.3 million hectares of arable land were under cultivation.
During the period between 1980 and 1988, the population increased by 3.3 %
per annum while the overall agricultural area increased by only 0.3%.59 The
continuous decrease of agricultural land due to population pressure was worsened
by the localised degradation of lands already under cultivation. Table 2
shows that natural reserves have decreased by more than 50% as population
has increased with a concomitant increase in the demand for land. Whereas
the area of arable land increased only by 10%, agricultural households
increased by 110% during the same period. By the end of the 1980s, nearly all
the arable land for agricultural production was in use. The ultimate impact was
severe food insecurity linked to unsustainable environmental practice and
resource capture. In 1990, over two million of the total Rwandan population
was living under conditions of permanent food insecurity.
If the current trend continues, agricultural conditions in Rwanda will
worsen. Already, agriculture is encroaching into protected areas and other
ecologically sensitive areas. Based on current population growth and patterns
of land use, it is estimated that 82% of all land holdings will be less than one
hectare or smaller by 2010. An estimated 38% of all holdings will be smaller
than 0.275 hectares.
Land Distribution
Land scarcity was a perennial problem in Rwanda even before the outbreak
of civil war in 1990. Owing to a number of interrelated factors, there was
insufficient land to meet the needs of the growing population in Rwanda over
time. ‘Free’ land was exhausted and the size of family holdings was decreasing.
The size of family holdings declined on average from 3 hectares per family
in 1949 to 2 hectares in the 1960s, 1.2 hectares in the early 1980s and 0.7
hectares by the early 1990s. This average, however, conceals great disparities
in the size of land holdings, with an increasing number of landless and nearlandless
peasants at the same time that the size of the largest farms was
increasing. Stress induced by high population density chiefly affects smallholders
who have few opportunities off the land to begin with.
Unequal land distribution in Rwanda dates back to the 1940s when cattleowning
Tutsi controlled large areas of land to graze livestock. Most of the farming
population was settled in the western highlands. The eastern savannas
bordering Tanzania were reserved as pasture for grazing Tutsi livestock. The
Hutu social revolution of 1959 and national independence in 1962 resulted in
68 Scarcity and Surfeit
the opening of large rangelands to settlement by Hutu farmers. An unknown
number of Hutu migrated to these rangelands from the highlands in western
Rwanda, where the local uneven distribution of land resulted in widespread
landlessness.60 Competition for land was fierce at the time. In addition to a
cycle of conflict and population displacements prior to and following independence
in 1962, there was also a total absence of a land policy or national
land use plan.
Resource capture by the elite was evident in the 1980s when the disparity
in land ownership between poor rural peasants and the elite grew tremendously.
61 In 1984, nearly 50% of the agriculturally productive land was held
on 182 farms out of an overall total of 1 112 000 farms.62 Furthermore, whereas
43% of poor families owned only 15% of cultivated lands, 16% of rich
families owned 43% of cultivated lands.63 These figures are supported by a
survey carried out in 1988 in five prefectures of Rwanda that show 60% of
agricultural households owned only 31.4% of arable land while 20% of the
population owned 46.9% of the total cultivable land. Increase in population
led to further divisions of smallholdings through inheritance, further decreasing
the viability of subsistence farming on many plots. Land scarcity among
the rural poor forced many to cultivate steep slopes prone to erosion and that
are acidic and unproductive.
By the 1990s, Rwanda was thus facing serious land scarcities that were
worsened by unequal access to and distribution of land caused by resource
capture by the elite. Although more than 90% of the population of Rwanda
subsists on various types of farming, most lack secure rights to own land. In
1994, 57% of rural households owned less than one hectare of land, and 25%
owned less than half a hectare. In a survey carried out before the genocide in
1994, the Ministry of Agriculture found that 45% of the rural population
(estimated to constitute 92% of the total population of Rwanda) were unemployed
or landless peasants. Another survey that was carried out by a nongovernmental
organisation, the Agency for Cooperation and Research in
Development (ACORD), in 1998 among 271 village households in Rwanda
found that 26% were landless. Furthermore, according to the survey, a majority
of landholders had less than one hectare of land. Most of the land
belonged to elite groups connected to powerful government officials. Rarely
do elite groups fully utilise their land holdings.
Even though the statistics on the disparity of the distribution of land
between rich and poor are few, it is clear that the size of farms for rural poor
are decreasing in size while the land holdings of the wealthier are becoming
larger. Indeed, more privileged individuals and groups possess the financial
means to acquire additional land that in many cases they do not use. The
unequal distribution of land is presumably a larger problem than population
pressure. Widespread disinheritance and unemployment in rural areas posed
a number of alarming risks to the government. There is little question that
Land Scarcity, Distribution and Conflict in Rwanda 69
the political economy of land was a significant factor contributing to the civil
war and genocide in the 1990s. The government, unable to address the land
issue without incurring a loss of patronage, attempted to use land to maintain
control of the state and thereby dominate the institutions and other decisionmaking
structures to allocate scarce land. Meanwhile, elite groups focused
on accumulating wealth at the expense of the rural poor. By controlling the
state they dominated key decision and rule-making processes to allocate land
and resources.
As observed earlier, land scarcity is prevalent for both the Hutu and Tutsi
in Rwanda. Resource scarcity is not divided along ethnic lines. Indeed, the
great majority of both Tutsi and Hutu did not benefit from the 1959 social revolution.
The situation for the majority of the rural poor did not change.
Instead, the revolution worsened poverty and inequality and concentrated
wealth in the hands of a ruling elite. However, grievances of the poor rural
Hutu failed to materialise into protest against the control of the state by a
small ruling elite. The government became increasingly insecure as food insecurity
grew in rural areas. The government was deeply concerned that growing
food insecurity threatened the legitimacy of their rule and state control.
Ethnic differences were used to polarise Hutu and Tutsi and to shift responsibility
for social injustice onto the Tutsi, regardless of class.
Land Scarcity and Adaptation
Olson64 identifies four main strategies to mitigate land scarcities in Rwanda.
One response to land scarcity is agricultural intensification. Different methods
used by farmers included reducing fallow periods, continuous cropping,
adopting labour intensive techniques, investing in land capital and switching
to higher yielding crops. Intensification of agricultural methods predates
independence in Rwanda, when land was distributed in favour of groups
who grazed livestock, mainly Tutsi. Hence, Hutu farmers intensified agricultural
practices in the western highlands where population densities were high
and land was scarce
A second response is to generate income off the land, either through trade
and exchange, or by working as agricultural labourers on nearby farms.
Opportunities off the land, however, are few. A third strategy is migration to
other rural and urban areas. The fourth strategy identified by Olson is family
planning, which includes delaying marriage and limiting the number of
births within marriage.
The former government of Rwanda, in cooperation with non-governmental
organisations and the private sector, pursued a number of other strategies to
mitigate the impact of land scarcity. Overall, government investments focused
on the agricultural sector. Agricultural policy was based on intensive farming,
protection against soil erosion, and the use of selected seeds, fertilizers and
70 Scarcity and Surfeit
other inputs. However, these were rarely available to smallholders, and the
implementation approach adopted by the government was draconian. Farmers
themselves adopted zero-grazing systems for livestock, and switched from cattle
to smaller livestock, such as goats and rabbits. However, changes to agricultural
practice had a very limited overall impact on alleviating land scarcities,
particularly for the poorest.
In light of the government’s failure to pursue a policy to promote agricultural
intensification, expansion of arable land continued throughout the
1980s ahead of the civil war. Pasture areas declined from 487 000 hectares in
1970 to 200 000 hectares in 1986 or to 19% of the overall land area devoted
to agriculture. The total cultivated land area during this time expanded from
528 000 hectares to 826 000 hectares.65 By the late 1980s, 94% of all cropland
was devoted to food crops, which took up 42% of the total land area (up
from less than 25% in 1965).
Another policy pursued by the former government was to force the Tutsi to
migrate to neighbouring countries. The departure, death or flight of more than
half of the Tutsi population in the early 1960s and in 1972 opened vast rangelands
in the east of the country where Tutsi previously grazed livestock.66 Land
formerly used by the Tutsi was redistributed to rural Hutu for farming.
Land, Politics and Power
Land forms the basis for material wealth in Rwanda. Control of land is
required to generate wealth and to sustain livelihoods. It is understandable,
therefore, that a vast majority of Rwandans in the time leading up to the
genocide were unable to meet basic livelihood needs through subsistence
production strategies given that land was scarce, and many were cultivating
unproductive marginal lands or were entirely landless. Subsistence production
was rarely adequate to sustain livelihoods in the best of times. Few
income-generating activities off the land compounded the impact of land
scarcity. According to the World Bank rural unemployment had already
reached 30% in the early 1980s. Unemployed youth migrated to urban centres
to take non-agricultural jobs, such as couriers and security guards.
It was towards the end of the 1980s that an acute famine began in the
south of the country and resulted in widespread deaths and population displacements
into less populated and more fertile areas, such as the east of the
country or to neighbouring Tanzania and Uganda. Land scarcity inevitably
led to widespread acute shortages of food. Food imports and aid became
increasingly important to cover shortfalls and weaken any potential opposition
to the government (See Table 2).
Rwanda is economically dependent on subsistence production that is highly
at risk to ecological change. Its economy, consequently, is vulnerable to
the effects of stress induced by ecological changes. By the early 1990s, when
Land Scarcity, Distribution and Conflict in Rwanda 71
Table 2: Food imports 1987–199767
Type 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
Commercial imports 25 31 48 56 56 60 42 17 21 39 52
Food aid 13 3 2 2 9 15 134 58 83 130 130
Total 38 34 50 58 65 75 175 75 204 169 182
rural livelihoods were threatened by environmental stress, dissention and
opposition grew among the rural poor. Environmental degradation and population
growth precipitated the emergence of internal opposition to the government,
which could no longer provide welfare for the rural poor. This coincided
with a weakening of the legitimacy of President Habyarimana’s government.
Prunier observes that during Habyarimana’s regime, the land question was
becoming increasingly thorny. Population pressure had reached critical levels,
and agricultural production was decreasing owing to soil erosion and
deforestation.68 This was worsened by the capture of land by powerful elite
groups. The government, however, was less concerned with redressing the
root causes of rural deprivation than with maintaining its control of the state.
Notably, the government failed to implement any substantive legal and policy
reforms to balance inequities in the distribution of land, or other reforms
to strengthen the land and resource rights of the rural poor and to prevent
further capture of land by the wealthy.
Instead, the government sought ways to maintain its complete control of
the state, and to continue its policy of exclusion of the Tutsi minority. The
most striking example is the creation of the Interahamwe militia, largely
made up of frustrated youths. Their parents were no longer able to lend support,
so they therefore sought their own means of survival. The Arusha
Accords provided for the integration of the two armies – the Armed Forces of
Rwanda (FAR) and the Rwanda Patriotic Front (RPF). The government supported
the Interahamwe as a militia force that could act with impunity to protect
the interests of the ruling Hutu elite. A confidential memo from August
1991 that was transmitted to local authorities explained: “Popular selfdefence
must be assimilated into the population up to the smallest administrative
unit (commune). The training of the volunteers shall be given by
members of the Rwandese Armed Forces.”69
For the ruling class, the popular defence force, or the Interahamwe, was a
critical part of its strategy to maintain control of the state in spite of the integration
of the Tutsi into the national armed forces. The popular defence force
itself was composed mostly of unemployed youth from rural areas that invited
the opportunity for social and economic advancement. Similarly, a news
72 Scarcity and Surfeit
commentator at the time of the genocide claimed that the ruling elite recruited
the militia (Interahamwe) from a half-schooled youth, with no sense of
direction and easily manipulated by money, beer and Indian hemp. In fact,
the Interahamwe received priority in employment, government loans, access
to land and resources, and education in return for their complete loyalty to
the interests of the ruling government.
Since control of the state enabled access to and control of scarce land, a
power struggle between competing elite groups to capture the state developed
over time. It is this competition that culminated in the genocide.
Historically, Hutu occupied land vacated by displaced Tutsi, thereby relieving
land scarcity temporarily. While conflict was used to entrench the power
of a ruling elite, it empowered poor and rich Hutu alike to claim land vacated
by the fleeing Tutsi. In the 1980s, the Habyarimana regime acknowledged
that population pressure on land was a major problem facing Rwanda.
However, the government reasoned that land scarcity prohibited the return
of refugees since there was no land for redistribution. When interviewed by
a French newspaper in 1989 on the possible return of Tutsi refugees to
Rwanda, President Habyarimana responded, “Rwanda is overpopulated,
there is no available space for Tutsi refugees.” Instead, the Rwandan government
insisted that neighbouring countries, mainly Tanzania and Uganda,
take the necessary measures to allow Tutsi refugees to remain permanently.
The government went as far as to propagate the view that the Tutsi were
colonisers from Abyssinia, to where they should return.
When Tutsi refugees demanded their right to return, the government and
Hutu extremist groups purposefully politicised the land issue to fuel resentment
among rural Hutu. Furthermore, claims to land by returning Tutsi refugees were
used to cast Tutsi intentions as an attempt to capture the state and reclaim
scarce land from Hutu. The popular understanding among Hutu peasants at the
time was that any returning Tutsi refugees would reclaim their land. The return
of Tutsi refugees, therefore, was deeply political and emotional to poor rural
Hutu who were concerned with the security of their own land rights. The government
considered that all means, including genocide, were necessary to discourage
the return of Tutsi refugees and to enlarge Hutu control of land.
Hutu extremists in support of the ruling government calculated acts of genocide
by purposefully laying the blame for land scarcity and widespread poverty
on the Tutsi minority. According to Prunier, the orders to commit the 1994
genocide (given by government authorities) were heeded by rural Hutu who
were led to believe that they would inherit the land of the killed Tutsi. As in the
past, the government assured Hutu peasants that land vacated by Tutsi who
were displaced or killed would be redistributed to landless Hutu and other
smallholder Hutu farmers. Claims to land were a strong motivation underlying
recurring pogroms in Rwanda over time. The ruling elite manipulated land
scarcity to its advantage, claiming to redress scarcity for the poor rural Hutu.
Land Scarcity, Distribution and Conflict in Rwanda 73
Thus, in the early 1990s, when the ruling regional Hutu elite from northern
Rwanda was challenged by provisions contained in the Arusha Accords,
the government increasingly emphasised Tutsi control of rural land and
resources to win the popular support of rural Hutu peasants and maintain
control of the state. Rural Hutu strongly supported the genocide of the Tutsi
and moderate Hutu in 1994. Access to and control of land was essential to
sustain rural livelihoods. Hutu are known to have participated in the genocide
in the belief that land belonging to the murdered Tutsi and moderate
Hutu would become theirs.
Responses to the Land Problem
Migration has been a common feature in the history of central Africa, including
Rwanda, since before the arrival of colonialists in the 1800s. Since independence
in 1962, migration and settlement have been important elements of
the government of Rwanda’s land policy.
One programme pursued by the former government was resettlement of
rural populations into farming villages. The expressed purpose of the villages
was to encourage rational exploitation of arable land in rural areas, mainly by
reducing population pressure on the land. Under the resettlement programme,
60 families were resettled per every 120 hectares, or on average two hectares
for every resettled family. Peasants signed a contract with the government
agreeing to reduce the number of cattle, intensify agriculture and grow cash
crops, mainly coffee, in order to increase income. Fragmenting land, as well as
inheritance, was also forbidden. The government undertook to provide adequate
training and to build basic infrastructure, such as roads, schools, and
water supplies. Lands that were redistributed under the programme included
national and communal lands, as well as lands belonging to refugees.
However, the resettlement programme failed largely because it was not
accompanied by real land reform that redistributed land to the rural poor fairly,
or the transfer of technologies to peasant farmers. Overall it was a shortterm
solution that dealt only with immediate needs. Population pressure was
reduced locally in some areas, but it remained a limited response that did not
redress the root causes of widespread land scarcity.
After the failure of the first land reform, the government proposed Rural
Development Centres (RDCs) to purportedly mitigate land scarcity. The RDCs
were to comprise existing rural villages or new centres. The RDCs were to be
nodes for rural economic development and diversification. RDCs consisted of
two zones of use. The first, the inner zone consisted of grouped settlements
around a centre where basic services were available. A second outer zone
was reserved for collective farming and livestock grazing. Examples of RDCs
include Nyacyonga and Rilima in Greater Kigali, Rutare in Byumba,
Rubengera in Kibuye and Rukumberi in Kibungo prefectures respectively.
74 Scarcity and Surfeit
Although there are few socio-economic analyses of RDCs, there was widespread
resistance to RDCs by rural populations. Complaints included that
houses were far from the farming area, and that few off-land opportunities
were established in the new RDCs.
Besides government-supported programmes for migration and resettlement,
there were large-scale spontaneous migrations of rural poor from
Rwanda to Tanzania. Although there are many reasons for this movement,
the main one must be the lack of land for agriculture and livestock breeding
in Rwanda and the availability thereof in Tanzania.
Peasants (mainly from the south), pushed by poverty and land scarcity,
sold their smallholdings to buy passage to Tanzania. However, many were
returned to Rwanda. The first group returned in 1985 and was resettled in
Kibungo region. A second group returned in the 1990s, but remained internally
displaced. Many participated in the 1994 genocide.
More recently the government has embarked upon a villigisation
(Imidugudu) resettlement programme and law reform. Each is discussed in
turn below.
The villagisation programme70 initiated by the government of Rwanda is a
reconstruction programme. The government considered Imidugudu as a pragmatic
solution to the scarcity of property and land, particularly for returning
Tutsi and Hutu refugees. It was also seen as a way to mitigate conflict emerging
over property and land by the population of returned refugees.
The main problem that the government faces in implementing Imidugudu
is high population density. The government prefers to settle populations in
previously unsettled areas such as the Akagera National Park in the northeast,
and in the former presidential hunting grounds near the border with
Tanzania in the south-east. However, there have also been cases where people
were regrouped into newly constructed villages within the periphery of
densely populated urban areas such as Kigali.
UNHCR and other non-governmental organisations initially funded
Imidugudu. However, the overall aim of the programme was disputed between
the donors and the government. The donors wanted a more limited programme
that focused on reconstructing shelters for the homeless and returning refugees,
whereas the government was interested in a much broader programme of
‘regroupment’ to promote ethnic and social integration and national reconciliation.
Villigisation embodied national reconstruction to the Rwandan government.
According to the government, services such as water provision, health
facilities and schools could be more easily built in regroupment villages.
Furthermore, regroupment villages would be built on marginal lands, opening
fertile land for agrarian reforms, including land and resource reforms.71
Imidugudu was implemented in Kibungo, Cyangugu, Butare, Byumba, and
Kigali prefectures. However, ethnic differences still trouble Imidugudu. Local
authorities are responsible for identifying beneficiaries of the programme, but
Land Scarcity, Distribution and Conflict in Rwanda 75
as Van Hoyweghen observes, apart from villages created in Kibungo and
other areas in north-western Rwanda, nearly all other settlements are inhabited
by Tutsi (either genocide survivors or first caseload refugees).72 Local
authorities, in many cases, do not recognise the Hutu as victims of the war
and genocide. Confounding the problem is that Tutsi genocide survivors
claim they are susceptible to militia attacks because they reside in clearly
identifiable ‘Tutsi’ villages. Imidugudu, therefore, has contributed to social
tensions in some areas of Rwanda.
Since the UNHCR withdrew support of Imidugudu, the programme is
threatened with a lack of sufficient financial support. It is clear that the government
cannot pursue the programme without the assistance of the UNHCR
or other international donors. It therefore remains to be seen the extent to
which the villagisation programme will be implemented.
Turning now to the issue of law reform.
The preceding pages have made the point that land is the most important
productive asset for most Rwandans. Yet Rwanda has a legacy of disputed land
rights, arising partly from the lack of legal status for land title and partly from
the return of refugees whose land has been redistributed to other occupants.
Hence the provision of security and the resolution of land disputes are important
objectives of the current government.73 In line with objectives outlined in
Rwanda’s poverty reduction strategy, the government is acting to reduce conflicts
over land, for example, through draft Land Policy and Land Law.
The Draft Land Policy includes the following provisions:
• All Rwandans enjoy the same rights of access to land, implying no ethnic
or gender discrimination.
• All land should be registered in title that can be traded, except where doing
so would fragment the land into plots that are less than one hectare in size.
• Land use should be optimal. Households will be encouraged to consolidate
plots to ensure that each holding is not less than one hectare; there
will also be a maximum size of 50 hectares for any individual landowner.
Families will be required to hold land in common should fragmenting the
land reduce the size of the plot to less than one hectare.
• Land administration will be based on a reformed cadastral system and be
subject to further consideration.
• The rights of occupants of urban land will be recognised on condition that
they conform to established rules.
The Draft Land Law specifies that:
• Persons occupying less than two hectares, and those with customary holdings
of between two and 30 hectares, where the owner has a project and
a development plan, will be recognised as the rightful owners.
76 Scarcity and Surfeit
• Transfer of title deeds requires prior consent of all family members.
• A land tax will be imposed.
• Undeveloped land reverts to the state’s private domain after a period of
three years.
• Holders of ubukonde land (originally distributed by the clan head), known
as abagererwa, will have the same rights as other customary owners.
The objective of the draft policy and law is to improve land management
while conferring security on the existing occupants of the land. It will be
important to devise cost-effective methods of resolving disputes at a community
level. It will also be essential that smallholders occupying plots that
are less than one hectare are not displaced on the basis of the Land Policy.
The process of allocating title will require inclusive local level participation.
It may be both more cost effective and transparent to conduct a survey to
determine land titles in each community on a particular occasion, rather than
to allow individuals to apply opportunistically to register particular pieces of
land. Traditionally, local land disputes were resolved by the local gacaca, or
court. Many households hold several plots at different altitudes both as an
insurance mechanism and to spread their labour inputs evenly over different
times of the agricultural calendar. As non-agricultural incomes rise, it should
be possible for households to diversify their income sources and therefore
reduce their dependence on land.
The poverty reduction strategy for Rwanda outlines a number of actions to
ensure that legal and policy reforms for land are effectively implemented,
including:74
• cabinet and parliament review of the draft land policy and land law;
• awareness raising and other civic programmes to educate the population
about their land and resource rights;
• establishing a cost-effective system to administer land and resource rights;
and
• creating mechanisms to settle disputes over land at the local level.
The measures undertaken by the Rwandan government to ensure tenure
security and ownership rights are a crucial step to redressing the role of land
in the conflict in Rwanda. Importantly, however, current reforms undertaken
by the government do not redistribute additional land to poor peasants. The
overall impact of the policy and legal reforms, therefore, may be limited as
long as redistribution is omitted.
Land Scarcity, Distribution and Conflict in Rwanda 77
Conclusion
Land is not the root cause of the Rwandan conflict. Various factors contributed
to the onset and continuation of conflict in Rwanda. The role of land,
however, is critical to understanding conflict dynamics in Rwanda. Land
scarcity in Rwanda is both a function of population pressure and the unequal
distribution of land. Control of the state by elite groups has facilitated their
domination of land ownership. This has aggravated land scarcity for the rural
poor by concentrating ownership of land with a minority.
Prior to the genocide, land was an important factor underlying the formation
of violent conflict between the ruling elite and armed opposition. The
land issue continues to complicate peace-building and national reconciliation
in the post-genocide period. The government is confronted with an enormous
responsibility to settle thousands of returning refugees and secure the land
and resource rights of the rural poor, who are the majority of the population.
At the same time the government, with the support of the international community,
is seeking ways to reduce dependence on subsistence production that
depends on access to scarce land. Dependence on access to scarce land has
reinforced the vulnerability of the rural poor, Hutu and Tutsi alike.
International policy in support of peace building in Rwanda must be based
on a more careful assessment and scrupulous investigation of the dynamics
underlying conflict in Rwanda. As this study shows, by emphasising the ethnic
parameters of the conflict over other important variables, including land,
past international interventions in Rwanda were, at least to some degree, misguided.
This was evident in the Arusha peace process, which emphasised
power sharing, the composition of the armed forces and elections in order to
diffuse ethnic tension, but did not redress other important structural factors
at play. The Arusha Accords, importantly, called for the return of all refugees.
However, the peace process neglected to understand the explanations given
by the government and did not attempt to overcome these by assisting resettlement
in terms of creating employment and helping to resolve competing
claims to land and resources. Given the highly charged environment in
Rwanda at the time, it is clear that any return of refugees required careful
planning and significant international involvement at all levels.
A further important finding is the need for inclusive peace-building
approaches. The extremist Hutu party, CDR, did not participate in the Arusha
peace process, and was later influential in undoing the Arusha Accords.
Similarly, some rebel groups fighting in the neighbouring DRC were excluded
from the Lusaka framework for peace building. It is important that all
armed factions participate in peace building. This may require ongoing dialogue
to address the hesitancy of some parties, and to reassure different factions
of the necessity and long-term dividends of peace. Furthermore, governments
must be open to negotiation with different armed groups during
78 Scarcity and Surfeit
peace dialogues in order to identify a lasting solution to the conflict for all
sides. It is also imperative that conflict prevention and management strategies
for Rwanda recognise the linkages between Rwanda and other countries
forming part of the central Africa conflict. Conflict and peace in Rwanda,
DRC, Burundi and Uganda are intimately related. Conflict in these different
countries centres on similar actors, interests and issues. Conflict managers
should devise a comprehensive framework that seeks to build peace across
borders and to minimise the impact of conflict in one country on neighbouring
countries.
Many reforms will be necessary to effectively manage the sources of conflict
in Rwanda. The government has the responsibility to strengthen the
security of rural livelihoods, and to create employment for thousands of
unemployed youth. Specifically, a new environmental policy is required to
enhance the ecological capital upon which rural livelihoods are based, as
well as to devise ways of generating greater goods and services from these.
Improving overall land use should be a key priority of the Rwandan government
and international development agencies operating in the country.
This will imply a scrupulous assessment of existing uses and devising a new
national land use plan to increase the provision of ecological goods and services
for the poorest. The introduction of a broader range of agricultural methods
through the transfer of new technologies that are suited to Rwanda’s ecological
vagaries may be a necessary element of improving overall land use.
It highly probable that Rwanda’s agriculture will reach its natural limit in
the near future, making it necessary to diversify the country’s overall production
methods and income-generating opportunities. Government initiatives
should reduce dependence on the land, while offering real opportunities
for generating income in other sectors.
This study points strongly to the unequal distribution of scarce land as a
significant factor contributing to the civil war and genocide in Rwanda.
Substantive agrarian reforms that would lead to a more equitable distribution
of land have been delayed for more than 30 years by the government. The
government, however, has a moral duty and responsibility to redress gross
inequalities in land ownership, and to improve livelihoods for the rural poor.
Land redistribution to benefit the poorest will be a necessary part of any
strategy for meeting these responsibilities. Doing so will reduce powerful tensions
related to access to and control of land, and contribute to the process
of national reconciliation and peace building.
Finally, land issues must be integrated into the training of conflict prevention
and management specialists in Rwanda. Non-governmental organisations
need to refocus their emphasis on environmental or ecological issues to
address issues of land and resource rights. This may help to shift the emphasis
from preventing perceived degradation of the environment to advocating
for and defending the rights of the landless and the rural poor. There is no
Land Scarcity, Distribution and Conflict in Rwanda 79
doubt that dealing with the issue of land and resource rights in a considered
and open way at the level of policy making to the level of local level dispute
resolution will have enduring benefits for peace building in Rwanda.
Endnotes
1 See also M A M Salih, Political narratives and identity formation in post-1989
Sudan, M A M Salih & J Markakis (eds) Ethnicity and the state in Africa,
Nordisksa Afrikaininstitutet, Uppsala, 1998, for further analysis of land conflict
linkages in Rwanda.
2 See T Homer-Dixon & V Percival, Environmental scarcity and violent conflict:
The case of Rwanda, Journal of Environment and Development, no 5, 1996 for
greater explanation of the role of resource capture in violent conflict.
3 See J K Gasana, Natural resource scarcity and violence in Rwanda, Paper presented
to the IUCN World Conservation Congress, Jordan 4 November, 2000, p 5.
4 Ibid, p 5.
5 See Homer-Dixon et al, op cit; H Hintjens, Explaining the genocide in Rwanda,
Journal of Modern African Studies, no 34, 1999 and S Van Hoyweghen, From
humanitarian disaster to development success?, Centre for Development Studies,
University of Leeds, working paper no 18, 2000.
6 Homer-Dixon et al, op cit, p 270.
7 See United Nations, The United Nations Environmental Programme, Environmental
Data Report 1993–1994, Blackwell, Oxford, 1993, p 217, and The
Economist Pocket Africa, The Economist, London ,1995, p 20.
8 See Gasana, op cit, p 6.
9 Ibid.
10 Danish International Development Agency (DANIDA), Special Report for Rwanda,
1997. <www.um/dk/danida/>
11 See P Chretien La Defi de l’Ethnisme: Rwanda et Burundi, 1990–1996, Karthala,
Paris, 1997: G Prunier, The Rwanda crisis: The history of a genocide, Fountain
Publishers, Kampala, 1995 and F Reyntjens, Afrique des Grandes Lacs en Crise:
Rwanda, Burundi 1988–1994, Karthala, Paris, 1994.
12 Prunier, op cit.
13 These refugees were later to become a major destabilising force in Rwandan politics.
14 See Prunier, op cit, 1995, p 52.
15 T Homer-Dixon, Strategies for studying causation in complex ecological political
systems, Occasional Paper, Project on Environment, Population and Security,
American Association for the Advancement of Science, Washington DC, 1995, p 12.
16 See Prunier, op cit, 1995, p 261.
17 This is the view of the Rwandan government.
18 Prunier, op cit.
19 See also C Vidal, Les Genocide des Rwandais Tutsis:Trois Questions d’Histoire,
Afrique Contemporaine, vol 17, 2e trimester, 1995.
20 Reyntjens, op cit, p 326.
21 S W Lee, Not a one time event: Environmental change, ethnic rivalry and violent
conflict in the 3rd world, Journal of Environmental Development, no 6, 1997, p 369.
80 Scarcity and Surfeit
22 The akazu in Kinyarwanda means a ‘little house’. The term applied to the ruling
clique in Rwanda then, centred around the family of Habyarimana’s wife. See
Reyntjens, op cit, 1994.
23 See Human Rights Watch, Leave none to tell the story: Genocide in Rwanda, New
York, 1999.
24 M A Storey, Structural adjustment, state power and genocide: The World Bank
and Rwanda, African Review of Political Economy, vol 28, no 89, 2001.
25 See various Human Rights Watch reports.
26 Hintjens, op cit, p 248.
27 B Collins, Obedience in Rwanda: A critical question, Hallam University School of
Cultural Studies, Sheffield, 1998.
28 M Chossudovsky, The globalization of poverty: Impacts of IMF and the World
Bank reforms, Zed Books with the Third World Network, London, 1997.
29 N S Karnik, Rwanda and the Media: Imagery, war and refuge, Review of African
Political Economy, no 78, 1998.
30 M Chossudovsky, op cit, pp 111–112.
31 Prunier, op cit, p 90.
32 See United Nations, The UN and Rwanda, 1993–1996, New York, 1996, p 11;
H Adelman & A Suhrke, The international response to genocide: Lessons from the
Rwanda experience, Joint Evaluation of Emergency assistance to Rwanda,
Copenhagen, 1997, p 20.
33 Gasana, op cit, p 12.
34 See Hintjens, op cit, p 241.
35 Ibid, p 242.
36 They were manipulated and used by extremist Hutu politicians to fulfil their
desire of creating an exclusive Hutu state, which would ensure their political survival
and holding of power.
37 Prunier, op cit, p 142.
38 United Nations, op cit, p 14.
39 Ibid.
40 See Homer-Dixon et al, op cit, p 2.
41 Adelman et al, op cit.
42 Prunier, op cit, p 62.
43 See F Cliffe & R Luckham, Complex political emergencies and the state: Failure
and the fate of the state, Third World Quarterly, no 20, 1999.
44 Gacaca would use a mixture of prison terms and community service in its sentencing.
See The Economist Intelligence Unit, Rwanda Country Report, 1st quarter,
2000, London, p 9.
45 See Adelman et al, op cit, p 7.
46 See United Nations, op cit, p 20.
47 The USA, for example, which had opposed UNAMIR has been one of the biggest
donors to the relief operations. See, e.g, Van Hoyweghen, op cit, p 18.
48 These local NGOs include ACORD (Agency for Cooperation and Development)
Ligue des Droits de la Resource dans les Region des Grands Lacs (LDGL), Conseil
de Concentration des Organisations d’Appui Aux Initiatives de Base (CCOAIB),
Reseau des Femmes, and Haguruka.
Land Scarcity, Distribution and Conflict in Rwanda 81
49 Homer-Dixon et al, op cit, p 270.
50 Demand-induced scarcity results from population pressure whereas supplyinduced
scarcity results from the degradation of resources.
51 91% of total wood consumption in Rwanda was for fuel wood. See Homer-Dixon,
op cit.
52 Ibid, p 3.
53 Gasana, op cit, p 12.
54 See International Monetary Fund (IMF), An approach to the poverty reduction
action plan for Rwanda, The Interim Poverty Reduction Strategy Paper, 2000.
55 J Olson, Demographic Responses to Resource Constraints in Rwanda, Rwanda
Society-Environment Project, working paper no 7, 1994.
56 Ibid.
57 B Blarel, Banque Mondiale with SESA, 1988.
58 See Republique Rwandaise, Recement Generale de la Population et de l’Habitat au
15 Aout 1991, Kigali, 1993.
59 G Baechler, Violence through environmental discrimination, Academic, Kluwer,
1999.
60 Olson, op cit.
61 Resource capture applies to both control of resources and control of the state for
political and economic gains. See Gasana, op cit. Gasana views both access to
power and access to land as intimately related as control of the state enabled elite
groups to capture land among other property.
62 Baechler, op cit, p 139.
63 See Gasana, op cit. See also Adelman et al, op cit, p 18.
64 J Olson, Behind the Recent Tragedy, Rwanda Geo Journal, vol 35, no 2, 1995.
65 See Olson 1995, p 326.
66 P Uvin, Aiding violence: The development enterprise in Rwanda, Kumarian Press,
West Harford, CT, 1998, p 187.
67 World Bank, Rwanda Poverty Survey, 1998.
68 Prunier, op cit, p 16.
69 See French Policy in Rwanda 1973–1994.
70 Known as Imidugudu in Rwanda.
71 See Van Hoyweghen, op cit, p 28.
72 Ibid, p 29.
73 See International Monetary Fund, op cit.
74 Ibid.
82 Scarcity and Surfeit