In 1994 the world was left astonished by the human capacity to cruelty in the Rwandan
genocide and her own incapacity to take proper action. Preceding the genocide, Rwanda
found itself in the midst of a civil war between the Hutu-led government and the
Rwandan Patriotic Front (RPF), which consisted mostly of second generation Tutsi
exiles. Whilst both parties were contributed to finding a truce at the Arusha Accords,
Hutu extremists holding offices in the top of the Rwandan government were planning a
war. On the 6th of April 1994, the Rwandan president Habyarimana’s plane was shot
down. Who was responsible for this assassination remains unknown. However, this
incident lead to the launch of the Rwandan genocide. Within a hundred days, at least half
a million Tutsis were systematically annihilated. Several governmental and military
authorities have been held responsible by the International Criminal Tribunal for
Rwanda (ICTR) for the genocide and its preliminary organization. As the authorities
were mostly responsible for the organization of the genocide, the execution largely
depended upon the participation of ordinary civilians. In order to convict these local
perpetrators, community-based Gacaca courts were created. Up to now, they been
successful in convicting a large part of the local perpetrators.
An image commonly associated with the Rwandan genocide is the machete. The machete
is an agricultural tool used by the predominantly rural Rwandan population. During the
genocide, this traditional tool fulfilled a renewed function; to kill. Besides machetes,
clubs, firearms and grenades were used as well in carrying out the genocide. These
different weapons were used by the various types of perpetrators present; from national
and local authorities to the various militia’s and ordinary civilians.
The Rwandan case has been a popular field of research within genocide studies. Scholars
have tried to explain the various aspects of the genocide, with the ultimate goal of
preventing another such an astounding process from happening. Little research has
been done on revealing the role weapons play in genocide. Logically weapons are
necessary for the execution of genocide. More importantly though, it is yet unknown
why perpetrators choose a certain type of weapon for genocide; whether they actually
have a choice, or if their decision is influenced by factors such as availability, costs, 7
capability to operate a weapon, etcetera. This research will try to shed more light on the
reason why these specific types of weapons were used in the Rwandan genocide.
An article with a similar topic has been published in 2006 by Philip Verwimp. His
research, published as Machetes and Firearms: The organization of Massacres in Rwanda
studies the variation in the weapons used in the Rwandan genocide. His research and
the consequent results will be elaborated upon in more detail in chapter 3. In brief, the
conclusions that Verwimp comes up with are only concerned with the influence of
victim identity on the weapons used. Personally, I find it hard to believe that the Page: 7
choice of weapons to use merely depends upon the victim identity. This research will
build upon the study done by Verwimp, but will focus on different variables that could
have been of influence on the weapons used.
As a consequence of limitations in the data available for use, this research will dedicate
itself to a case study of the variation in weapons used in Kibuye prefecture. The research
question is whether the variation in death by type of weapon during the Rwandan
genocide in the prefecture of Kibuye can be explained by various geographical and sociodemographic
variables and perpetrator characteristics. These variables are
subsequently translated into five hypotheses, derived from two existing theories on the
Rwandan genocide, including Verwimp’s research.
The prefecture of Kibuye is chosen as a case study for two reasons, the first being that
Verwimp’s research is also focused on Kibuye. To build on his work, it is necessary to
focus on the same subject. Secondly, the data made available by the survivors
organization IBUKA, is limited to the Kibuye prefecture. At first my research was
intended as an analysis at the Rwandan national level. However, since there was not
enough information on the use of weapons in other prefectures, I had to limit my study
to Kibuye. The prefecture of Kibuye however is the area with the most comprehensive
genocide in the whole of Rwanda.1 In 1991, a count of 71.225 Tutsis was made by the
national government. African Rights has estimated between 7.000 and 8.000 Tutsis to
have survived the genocide in Kibuye.2 Since Kibuye only consists of 9 communes, it will

1 Melvern, 2004 (p. 224)
2 African Rights, 1995 (p.394)8
not be viable to quantitative analysis. The results from this case study could lead to
interesting hypotheses for the national level, to be tested in further research.
The methodology used in this research will be both qualitative and quantitative. The
IBUKA dataset will be used to illustrate the variation in weapons used and frequency of
deaths throughout Kibuye. Subsequently, I will test the five hypotheses by making use of
quantitative analysis. In the absence of quantitative data, I will additionally make use of
qualitative analysis in the form of secondary literature to test the hypotheses.
Furthermore, qualitative analysis will be used to explain the results and correlations
from the quantitative study.
This research should be regarded as an exploration of other variables explaining the
variation in choice of weapons than those proposed by Verwimp. It is not necessarily
trying to falsify his theory. Rather the aim is to contribute to understanding the various
factors influencing the variation in weapons used during the genocide. Verwimp
assumes that perpetrators had a choice in weapon and that their decision was
determined by the victim identity. Instead, this research will mainly focus on the
question whether perpetrators actually had a choice in weapons to use, or whether
weapons were only selectively available to certain types of perpetrators. The ultimate
goal is to derive hypotheses from the case study pertained to weapon use at the national
level. Its significance lies in understanding whether the weapons used merely depend on
the identity of the victim, as claimed by Verwimp, or whether other factors are at play.
This thesis is constructed by firstly defining the concept ‘genocide’. Within the field of
genocide studies, there is an ongoing discussion on the definition this concept. The first
chapter will elaborate on this academic discussion and clarify the concept used in this
research. The third chapter will explore the existing theories on genocide that are of
relevance to my subject. From these theories, five hypotheses will be derived. Prior to
testing these hypotheses, the variation of deaths and weapons throughout Kibuye will
be researched. Consequently, I will verify the influence of Verwimp’s variables on my
sample in chapter 4. After having identified the communes with a distinct number of
deaths or use of firearms, the five hypotheses are tested separately by mostly making
use of quantitative analysis. The lack of quantitative data turned out to be a large
challenge to my research. For that reason, two hypotheses will be tested by using9
qualitative literature. Finally, the thesis will be concluded by a discussion on the
implication of the results and the construction of hypotheses on the use of weapons of
genocide at the national level.