UNHCR IS MY HUSBAND: INTERPRETING THE NATURE OF
MILITARIZED REFUGEE WOMEN’S LIVES IN
KANEMBWA REFUGEE CAMP
By
Barbra Lukunka
Submitted to the
School of International Service
of American University
in Partial Fulfillment of
the Requirement for the Degree of
Masters of Art
In
International Peace and C
Chair: Susan Shepler
June Mertus
Dean of School of International Service
Dai 2007
American University
Washington DC 20016
AMERICAN UNIVERSITY LIBRARY
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 1448704
Copyright 2007 by Lukunka, Barbra
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
UMI
UMI Microform 1448704 Copyright 2007 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. © COPYRIGHT
by
Barbra Lukunka
2007
All Rights Reserved
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNHCR IS MY HUSBAND: INTERPRETING THE NATURE OF
MILITARIZED REFUGEE WOMEN’S LIVES IN
KANEMBWA REFUGEE CAMP
BY
Barbra Lukunka
ABSTRACT
The nature of conflicts after the end of the Cold War leads one to question
different dimensions of the effects of war. The consequences of war transcend the
destruction of power and state structures in present conflicts however they also
destroy norms and socially woven mechanisms in place that are tied to people’s
behavior and culture. Refugee women in Kanembwa camp reveal a change in
behavior as a result of conflict in Burundi. This study analyzes the militarization
of refugee women, a phenomenon that has been made possible due to the nature
and history of conflict in Burundi, and the culture and circumstances in the
refugee camp. By analyzing the history of the Burundian conflict and the
transformation of the Burundian culture in the refugee camp, the nature of refugee
women’s militarization in Kanembwa camp can be understood.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
ABSTRACT...... ii
LIST OF ILLUSTRATIONS...... iv
LIST OF TABLES...... v
Chapter
1. INTRODUCTION...... 1
2. LONG INTERVIEWS AND OBSERVATION: COLLECTING DATA ON THE LIVES OF REFUGEE WOMEN...... 9
3. THE QUESTION OF MILITARIZATION: DIALECTIC OF UNCONVENTIONAL BEHAVIOR ...... 39
4. HISTORICAL ANALYSIS: BRIDGING POLITICS, ETHNIC ANIMOSITY, GENDER AND MILITARISM...... 85
5. EXPERIENCES OF MILITARIZED REFUGEE WOMEN: FEAR AND DEPENDENCY...... 117
6. CONCRETELY UNDERSTANDING THE CULTURE OF KANEMBWA REFUGEE CAMP...... 138
7. CONCLUSION ...... 175
APPENDICES...... 178
BIBLIOGRAPHY...... 183
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ILLUSTRATIONS
Figure
1. Causal Model Explaining the Militarization of Refugee Women......
IV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLES
Table
1. International Rescue Committee Sexual and Gender Based Violence Centre 2006 Chart......
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER ONE
INTRODUCTION
Political violence is corralled as the province of rational militaries and mostly rational soldiers controlling the dangerous elements and explosive fissures inherent in human society. A comfortable picture, but a mythical one.1
The understanding of war has undergone a metamorphosis with time. The
end of the Cold War seemed to have evoked a change in armed conflict. There
appear to be more intrastate wars and far fewer interstate wars. The end of the
Cold War put a different focus on what conflict means. This new phenomenon or
type of war has led to the understanding that perhaps the state is not the only
important actor in International Relations. There are different actors that we must
now take into account. The severities of conflicts are also pronounced considering
that “wars today are longer in duration, deadlier and kill higher percentage of
civilians than wars of preceding centuries.”2 Conflicts are no longer understood
merely as being between state armies. Armed conflicts are now being fought by
different types of actors, and against civilians. As stated by Carolyn Nordstrom,
the idea that wars equals soldier and soldier equals male is a myth.3
1 Carolyn Nordstrom,Shadows o f War: Violence, Power and International Profiteering in the twenty-first Century (Berkeley, University of California Press, 2004), 34. 2 Nordstrom,Shadows of War: Violence, Power and International Profiteering in the twenty-first Century, 43. 3 Nordstrom,Shadows o f War: Violence, Power and International Profiteering in the twenty-first Century, 33. 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Conflicts are unconventional today compared to how they have been
traditionally understood. What we see however is a new type of conflict where
civilians, children and women partake in conflict. Civilians, children and women
are three categories of people that have been viewed as non-combatants and
therefore “innocent” in conflict. Women especially are often viewed as passive,
and victims of wars. Essentially the idea that “women don’t join mobs, they are
assaulted by them”4 does not always hold weight.
Africa provides many examples of non-traditional conflicts; in addition,
the damage that the armed conflicts have caused on the continent has gone
beyond the destruction of any state or power structures but has also led to the
destruction of lives, social norms, identities, and traditions. Africa has suffered
numerous conflicts which have left the continent in a transformed state. This is
indicated not only by the number of fragile states and structures but of the
destruction of life and human well-being. In addition the remnants of conflicts
have also permeated cultural and traditional aspects of society thereby
transforming behaviors and identities.
In this study I insist on the analysis of the micro-level effects of conflict
and the changes that it brings. This study is intended to move beyond the
conventional realist study of state behavior and to look specifically at some of the
most affected individuals in society due to conflict. My study analyses the
consequences of the Burundian conflict on the lives of Burundian refugee women.
4 Nordstrom, Shadows of War: Violence, Power and International Profiteering in the twenty-first Century, 30.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I present the culture of militarized Burundian refugee women in Kanembwa
refugee camp in western Tanzania. Culture in this context is understood as “the
way of life of a particular people.”5 Therefore the culture of militarized refugee
women refers to ways in which militaristic activities and ideas are integral to the
lives of refugee women. The culture of militarized refugee women clearly reveals
the extent to which conflict and militaristic ventures and activities transform
behaviors and identities and lives. By the same token the transformation of
behaviors and identities brought about by conflict and the consequences of
conflict also explain the continuity of militaristic activities during conflict and in
post-conflict situations. The definition of militarization that drives this paper is
presented by Cynthia Enloe who describes militarization as a “step-by-step
process by which a person or a thing gradually comes to be controlled by the
military or comes to depend for its well-being on militaristic ideas.”6 Therefore
this study analyzes the ways that refugee women have come to be controlled by
the military, and how they have come to depend on the military as a result of the
conflict and the circumstances in the refugee camp. The fact that refugee women
have two contradicting connections with militarization: as victims and supporters
of militarism, can be explained by looking at the development of militarization in
history, as well as the cultural change brought by conflict, specifically in regards
to the effects and changes of patriarchal culture.
5Keith F. Otterbein, Comparative Cultural Analysis (New York: University of Kansas Press, 1966), 4. 6 Cynthia Enloe,Maneuvers: the International Politics of Militarizing Women’s Lives(London: University of California Press, 2000), 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4
The militarization of refugee women is understood in this study as an issue
that is explained by both tangible and intangible factors. Refugee women and
other refugees in a political camp can be considered militarized due to the
symbolic attributes of a refugee camp as an environment created as a result of
militarization. By the same token refugees are militarized as a result of the
influences that military activities that take place in and around the refugee camps
have on their lives. These activities include military training, political activism,
and fundraising for military purposes and much more.
Refugee women have long been portrayed as victims of violence. This is
of course because they have and continue to experience some of the most
deplorable violations of their rights before, during and after their flight. War has
created situations which permit atrocities to occur that are sometimes ethnically
focused or gender focused. What one can notice in Africa’s Great Lakes Region
in Burundi and Rwanda is that ethnicity and gender at some point or another
during the conflict become intertwined and work together in complicated ways.
The wars that have been fought in these two countries demonstrate ethnic conflict
that has a gendered focus. The tendency is to write on the violence that women
face, however this study breaks from this tendency and enters a realm of
understanding a multidimensional existence of refugee women. In this study I
often pose a liberal feminist question: “Where are the women”7 in the Burundian
history and conflict? However this study is not merely about placing women in
7 Francine D ’Amico, “Critical Feminism: Deconstructing Nationalism, Gender and war” in Making sense of international relations theory edited by Jennifer Sterling-Folker (Lynne Rienner, 2005), 269.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. history; it is also about understanding the behavioral changes that women face,
especially those that go against conventional traditional norms, especially in the
patriarchal culture of Burundi.
Broadly speaking this study is about the behavioral changes that are
brought about by conflict. More specifically, it is about the culture of militarized
refugee women and even more specifically it addresses, why the history of the
conflict, patriarchal culture and the context of the refugee camp yield two
dichotomous situations where women are controlled by the military and fear the
military, and on the other hand depend on the military. This analysis is of the two
faces that refugee women carry in Kanembwa refugee camp as victims of military
activities as well as supporters. The central question that motivates this study
deals with the puzzling elements that allow refugee women to have two faces.
How is it possible that refugee women that are deemed the most vulnerable in the
world are able to support violence and militarism? This question is posed
specifically because violence and militarism are two reasons that forced refugee
women to leave Bumndi; in addition they are two elements that exist in the
refugee camp that control the lives of refugee women. The unconventional
behavior demonstrated by refugee women is inevitably the byproduct of conflict
and the intricate aspects of conflict and refugee life. Their behavior is considered
unconventional because it goes against gender norms as well as patriarchal values.
There are different ways of tackling the questions that I pose. These are
questions that could benefit from a study that takes a normative approach. By
studying the repercussions of conflict, the social changes that take place during
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. conflict and the social issues that lead to war, one may present a study that
promotes international norms to prevent future conflicts. Especially in liberal
feminism, the normative approach can lead one to question our understanding of
certain socially constructed concepts. This can lead to an emancipatory study that
focuses on what Julie Mertus (2005) refers to as “challenging epistemological
orientations.”8 This study however is motivated by the need to establish
explanatory variables that can provide a lens through which one can understand
the unconventional behavior in conflict situations without presenting a normative
or emancipatory focus. In order to understand the militarization of refugee women
in Kanembwa camp I propose that it is important to analyze various variables that
can be categorized into two factors: history and culture. History sets the context
and it explains the important factors leading to present situations. In history we
can find the correlations that develop between militarism, politics, ethnicity and
gender. Culture presents the values and influences that lead to certain behaviors
and behavioral changes. By analyzing culture we establish the behaviors that
change over time, notably the challenge that is presented to patriarchal culture.
The emphasis in this study is how Burundian patriarchal culture is slowly losing
its grip on women in the refugee camp and therefore this explains the
unconventional behaviors demonstrated by women. Women’s unconventional
behavior goes against traditional and patriarchal norms as they support militia
groups and take up arms. By the same token, the slow loss of patriarchy is facing
8 Julie Mertus, “Liberal Feminism: Local Narratives in a Gendered Context” Makingin sense of international relations theory edited by Jennifer Sterling-Folker (Lynne Rienner, 2005), 265.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7
resistance from men. Furthermore, patriarchy and male dominance explain the
control the military has over women, while the loss of patriarchy explains the
unconventional behavior demonstrated by women as they take up militaristic
ideals. The dichotomous existence of patriarchal culture is pronounced when we
look at how many women took up arms in the Burundian conflict. Reports
indicate that only about 490 women were demobilized in the Burundian conflict.9
This number can be see as insignificant and a clear manifestation of patriarchy
and delineations of gender roles that keep women out of the military. However,
another perspective is that 490 women involved with military and militia groups
is significant in a highly patriarchal culture and is therefore evidence of the
change that the Burundian patriarchal system is undergoing.
Although history and culture are broad explanatory factors and can be
applied to any conflict, it is my premise that the specific factors that fit the
category as historical or cultural variables differ across the board. This study is
not meant to generalize about other conflicts and other behaviors of refugee
women in other countries or women in general. The data for this study was
collected in the refugee camp of Kanembwa in western Tanzania, and therefore
the findings and analysis apply to that particular camp. However, this study also
presents various constructs such as gender, ethnicity, and patriarchy which are
similar in other African settings.
9 Waldemar Vrey United Nations Operations in Burundi Official, Interviewed by Author on Email.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8
Chapter two will outline the ethnographic research methodology that was
used in this study which led me to travel to Kanembwa camp in western Tanzania
and to interview twenty-four individuals, seventeen of whom were refugee
women. Chapter three of this study is a discussion of the various literatures that
exist concerning refugee security and the phenomenon of militarization and other
concepts that help understanding its formulation and existence. Chapter four is a
historical analysis which frames the issues that have allowed for militarization by
focusing on ethnic animosity and the history of politics and militarism. Chapter
five presents the experiences of refugee women as told by refugee women that
explain how militarism is present in their lives. Chapter six will analyze the
situation and conditions in the refugee camp that explain why refugee women
have had the sort of experiences that are illustrated in chapter five.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER TWO
LONG INTERVIEWS AND OBSERVATION: COLLECTING
DATA ON THE LIVES OF REFUGEE WOMEN
My research on the militarization of refugee women in western Tanzania
aimed to uncover social issues deriving from patriarchy, and consequences of
colonialism in Burundi. An analysis of present day African conflicts reveals that
the conflicts share many similarities in that they have some roots in colonial
legacies, and they cause one to question the issues of patriarchy as women are
raped, abducted, manipulated or invited (or choose) to take up certain roles as
they participate in the conflicts. My research can be best understood in the
interpretive paradigm/ which “holds status quo assumptions about the social
world and subjectivist assumptions about epistemology.”2 Informed by this
paradigm, my research methodology was highly motivated by the objective of
understanding the “social world as it is (the status quo) from the perspective of
individual experience, hence an interest in subjective worldviews.”3 This is the
primary mode of the discipline of anthropology and I was particularly drawn to
this paradigm because of its emphasis on “thick description of the actor’s world
1A definition of critical humanism is provided by Gretchen B. Rossman and Sharon Rallis, Learning in the Field (London: Sage Publication, 2003). 2 Rossman and Rallis, 46. 3 Rossman and Rallis, 46. 9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10
views.”4 My research on Burundian refugee women militarization is not for
emancipatory purposes; instead it is for the purpose of enlightenment and adding
to the understanding of the phenomenon that I am studying. Most importantly
tapping into the experiences of those who reflect on their circumstances enables
others to be enlightened and knowledgeable of a reality that is unique and only
completely understood and experienced by those retelling their stories. Therefore
my research does not present data that is universally applicable to the situation of
all militarized refugee women. What I present is the lived experiences of
particular individuals with particular experiences in particular settings and the
interpretations that these individuals give to their world.
The methodology I used for the exploration of the experiences and culture
of militarized Burundian refugee women in western Tanzania was naturalistic and
completely interpretive. I argue that, especially in view of the questions that I
pose, this is the only reasonable methodology. By analyzing and observing
refugee settings, and by interpreting refugees’ lived experiences, I planned to
discover the various elements that lead to the culture of militarized Burundian
refugee women, and to understand the nature of the culture of the refugee
women’s militarization. By studying the “culture of militarized refugee women” I
aimed to study the aspects of refugee life that made refugee women susceptible to
being controlled and/or depending on the military, militaristic activities and ideas.
On one hand I looked for evidence of the fear that refugee women have of the
military therefore indicating a level of control by the military, and on the other
4 Rossman and Rallis, 46.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11
hand I looked for elements described by Robert Muggah (2006) that signify
militarization such as women’s voluntary participation in military activity in and
outside the camps, their political activism and perpetration of violence, their
military training and recruitment activities, and their support for combatants and
armed resistance. I looked at their surroundings and interactions and what about
their surroundings and interactions made them prone to militarization. In order to
carry out research aiming to uncover such interpretive information, I was led to a
hermeneutic process that is best achieved through qualitative research methods
because the goal for this research was not to establish an objective truth of the
overall issue of militarized refugee women. I was led to a heuristic process that
allowed for my perspectives and interpretations. In order to fully comprehend the
culture of militarized Burundian refugee women, it was necessary to not only
analyze the leading literature; but to also take into full account the lived
experiences, perceptions and meanings that the refugee women give to their world
and situation. Not only were the expressions and thoughts of the refugee women
important to understand their militarization, it was also very necessary to analyze
their refugee settings. The refugee camp setting lends itself to many perceptions,
and presents an interesting case for analysis because of its situational and
contextual aspects as a community created from a form of disaster; in this case
war. This research methodology cm be used in many different communities;
however the nature of a refugee camp made the research methodology even more
appropriate because of the experiences that refugee women have in a society that
is created unnaturally through the initiatives of actors other than the individuals
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12
that live in them. I was able to conduct in-depth interviews with refugee women
and interpret the information they provided based upon different and recurring
themes.
My goal for my research is to add to knowledge on the reality and life of
militarized Burundian refugee women, which have not been fully analyzed. I am
approaching this research as the beginning of my exploration of refugee women’s
militarization in Africa, hoping to add to the scarce literature on this subject.
Therefore, my research presents definitions and variables that can be
operationalized in further studies, even those that are quantitative. Qualitative
research presented the appropriate method for this study because it allowed me to
explore and describe the phenomenon of Burundian refugee women in a heuristic
manner, thereby leading to an inductive study.
Conducting qualitative research allowed me to describe, analyze and
interpret my findings, as I took into account my personal biography as an African
woman that inevitably shaped my interpretation and understanding of the lives of
militarized Burundian refugee women. I took the words of the refugee women as
the path to discovering their lives and their created culture as the basis for
understanding their reality, I was aware of my personal biography and the
influence my biography had on how I understood their experiences and the
influence it might have had on what they chose to tell me.
The strategy and approach that I used for this research can be best
described as participant observation. This allowed me to build rapport with the
refugee women in order to gather information about their interactions with militia
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13
groups and their ideas about militarism, two subjects that can be easily interpreted
as sensitive intelligence information. It was necessary for me to build trust with
the refugee women through my interactions with them. It was also very important
for me to observe their settings in order to properly contextualize the phenomenon
of their militarization. My interest in the culture of militarized refugee women
defines my study as an ethnographic study. Ethnographic methods however
usually entail a prolonged presence in the research site by the researcher.5 Due to
time constraints, I was not able to spend an extended amount of time in the
refugee camps. However this did not discount my intentions of using
ethnographic techniques. I intended on using ethnographic methods in order to
capture some of the cultural aspects of the refugee women’s militarization.
Liisa Malkki (1995) raises a very interesting point about ethnographic
research in refugee camps. She states that, “refugee camps are not, of course, a
routine for ethnographic field work, and displaced people are not the usual “native
informants of anthropologists.”6 She conducted her ethnographic research on
refugees in Tanzania, and analyzed the way that refugees refer to and mould their
understanding of their history and nationhood from a refugee camp setting. In my
research, I did not intend to analyze refugee women as a permanent group that
holds a specific prolonged culture. Refugee life and culture is very uncertain since
refugees can find themselves in protracted situations for extended periods of time
or their stay can be cut short when conflict ends, or when they are repatriated.
5 Rossman and Rallis, 95. 6 Liisa Malkki, Purity and Exile: Violence, Memory, and National Cosmology Among Hutu Refugees in Tanzania (Chicago: University of Chicago Press, 1995), 4.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14
My concern was the culture of militarized Burundian refugee women; this
culture, I contend, is temporary and highly conducive to change due to the
capricious nature of conflict and especially of refugee settings. The culture of
militarized refugee women, like other cultures, is not static and can be drastically
altered depending on the circumstances in the home country as well as the host
country. The purpose of this study was to analyze the type of culture that the
Burundian refugee women have that reflects the role of militarism in their lives.
As noted earlier, culture refers to the way of life of individuals. In this study I
specifically looked for ways that reveal how military ideas and activities are
integral to the lives of refugee women.
Malkki also raises the important issue that literature on refugees tends to
essentialize refugees as special persons. She states that, “as the refugee has
become idealized and generalized as a type of person, he or she has also become
an object of specialization.”7 She states that many scholars and authors have
looked at refugees as individuals tom from their culture, however her research
leads her to have the view that refugees produce their culture and maintain and
produce their own concrete identities.8 By choosing to analyze refugee women
and their culture in the militarized situations, I sought to understand the elements
that allow for a produced culture that militarized women refugees have, how
militarization is part of their lives and how they translate this culture. My findings
indicate that refugee women, like other refugees, hold on to their primary culture
7 Malkki, 9. 8 Malkki, 15.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15
but also manage to create a new culture that is easily altered and malleable based
upon circumstances and events. I will show that the culture created by militarized
Burundian refugee women is rooted in the social issues inherent in Burundian
society, the circumstances of their surrounding in the refugee camp in relation to
militarism and political activities, the influence of aid workers, and the influence
of the culture of their host community.
The culture created by militaristic ideals or control by the militia can be
seen as both an issue of essentialism and of empowerment. My research is not
necessarily intended to answer the question of which category militarization of
refugee women culture fits in but rather to portray this culture in the way that the
refugee women view it. However, it is necessary to be cognizant of the various
theories and questions that informed my motivations for this study. Although I
intended to provide an emic analysis, I am cognizant that my understanding of the
situation influenced my interpretation.
In the Field
I was in Tanzania from December 31,2006 until February 4,2007 where I
conducted field research and interviewed Burundian refugee women in western
Tanzania. In order to have access to the refugee camps, I was affiliated with the
Centre for the Study of Forced Migration in Dar es Salaam which helped facilitate
my visits to the refugee camps in Tanzania by submitting an application on my
behalf to access the camps to the Ministry of Home Affairs. In exchange I paid
the center a fee that researchers and interns pay to be affiliated with the Center.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16
In order to expedite my application process I was also assisted by the Special
Rapporteur for Refugees and Displaced Persons for the African Commission on
Human and Peoples Rights. Keeping in mind the security issues that could arise
from my study, I did not provide any of my research notes or share any of the
information that I retrieved about specific individuals in the camps to the Centre
for the Study of Forced Migration or others. I followed standard protocols for
working with human subjects in areas of political sensitivity, assuring them of
confidentiality.
I had initially intended to conduct my research in Lukole camp. This
camp is located at the border of Tanzania, Rwanda and Burundi in the Kagera
region of Tanzania. According to the United Nations High Commission for
Refugees, Lukole camp is a highly organized refugee camp with a population of
48,767.9 I was unable to secure proper assistance in Kagera from the NGOs there;
furthermore I was encouraged to go the Kigoma region because of the assistance I
could get there from the NGOs on the ground. I ended up going to Kibondo in the
Kigoma region and conducting my research in Kanembwa refugee camp.
Kibondo has four refugee camps: Kanembwa, Mkugwa, Mtendeli and Nduta
camps. I decided to conduct my research in Kanembwa camp because the camp is
older than the other camps and was thus created shortly after the massacres in
9 See UNHCR Statistics for details in “2005 Global Refugee Trends” available on line from: http://www.unhcr.org/statistics/STATISTICS/4486cebl2.pdf.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17
1993; therefore it inevitably has a longer history of militarization.10
I analyzed the lives of Hutu refugees from Burundi in Kanembwa camp. I
had initially wanted to interact with Rwandan refugees but was unable to talk to
any. Many of the Rwandan refugees have been repatriated back to Tanzania, and
some of the few that remained are in Mukugwa camp in Kibondo. The conflict in
Burundi has shaped concrete delineations of Hutu and Tutsi ethnic identity.
Therefore, even though my research was specifically on the culture of militarized
Burundian refugee women, the Hutu women were of particular interest because of
their ethnic and gender situation in the conflict. Furthermore, the categorization of
ethnicity in the conflicts has had a very vivid impact on women and women’s
position in society, as well as in conflict. Therefore, this conflict is particularly
interesting because of the particular symbolic and representative attributes11 given
to women during the conflict that has rendered women rape victims, as well as
symbols of purity and nationhood. In Malkki’s research, refugees told her stories
that ended up being narratives from which she was able to analyze the culture and
history that refugees in Mishamo camp held on to or created. One of the important
narratives recurring in her research is about Burundian men’s view of Tutsi and
10 When I arrived I was unsure of which camp would suit my research goals especially since my preparation before my visit was for Lukole camp. I visited three of the camps in Kibondo: Nduta camp, Mtendeli camp and Kanembwa camp. Nduta was the first camp I visited. It is a camp located in the forest which gives the environment very little light and a somber atmosphere. Nduta was also the farthest camp from the town of Kibondo where I was staying. The population of Nduta camp is approximately 18,000.1 later visited Mtendeli camp, which is the biggest camp in Kibondo with a population of about 20,000. Mtendeli is also far from Kibondo. I later visited Kanembwa camp which has a population of 12,500. Kanembwa is the closest camp to Kibondo and it is also the oldest camp among the three camps I visited. Due to transport issues, 11 Marcia Kovitz, “The Roots of Military Masculinity”Military in Masculinities: Identity and the State (London: Praeger, 2003), 6. Kovitz provides a discussion on the attributes given to women that are symbolic in nature.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18
Hutu women. 19 Although Malkki acknowledges that some of the narratives can be
seen as factual and others are probably not, they are strong representation of how
the Hutu refugees view themselves and therefore shaped their identity and culture.
To present a coherent study on the culture of militarization of refugee
women, I relied on observation, in-depth interviews, and one focus-group
meeting. The interviews I conducted enabled me to present a descriptive analysis
of the lived experiences of Burundian refugee women and to better understand
their culture. I initially planned on interviewing between twenty and thirty refugee
women for at least forty-five minutes to an hour each. I ended up interviewing
twenty-five individuals. I interviewed seventeen refugee women, eight other
individuals. They were a combination of male refugees, former male combatants
in the camp, and officials from the Tanzanian Ministry of Home Affairs official,
United Nations High Commission for Refugees (UNHCR) Officers, United
Nations Mission in Burundi (UNOB) Official working on DDR program in
Burundi, and International Rescue Committee (IRC) Sex and Gender Based
Violence workers. I also collected information from various NGO workers and
refugees that I had informal conversations with. My interviews in the camp lasted
between one to two and a half hours. I was not in the camps long enough to
conduct the “conventional” in-depth iterative interview that requires a researcher
to interview an informant about three times with a few days or even weeks in
between interviews.13 The conventional in-depth interviews would have been
12 Malkki, 84. 13 Iriving Steidman,Interviewing as Qualitative Research: A Guide for Researchers in Education and Social Sciences (London: Teachers College Press, 2006), 17.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19
ideal for this research because of the wealth of information presented by the
informants. After interviewing the refugees I would realize that there were other
questions I had, especially based upon the information they had given.
I also held a focus group meeting with a total of thirty-five refugee
women. This was the best way for me to gather information from a variety of
women at once. I used the focus group to gather information on topics and issues
that were priorities for most women.
Observation and Interviews
My interactions with the participants were overt. The participants and
other refugees who were curious about my activities knew that I was in the camp
as a researcher studying the lives and experiences of refugee women. I went to the
camp with the intention of making my identity clear. I was however prudent and
did not directly introduce my study as one on the culture of militarized refugee
women. Instead I chose to present the study for what it is, one on the culture and
experiences of refugee women, but also leaving out the term militarization in
order to not cause alarm. It was very important for me to inform all the refugees
that I am from Zambia and that I was conducting research. I made sure that the
refugees understood that I am not from Burundi or Rwanda. Before entering the
camp some of the aid workers explained to me that the refugees were very
suspicious of outsiders. They warned me that the distrust by many refugees of
outsiders could lead the refugees into labeling me as a Hutu or a Tutsi. This was
very alarming to me; however I did not fully understand what the aid workers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20
meant until one of my informants overtly expressed to me that based upon my
morphology I could fit into one of the categories of Hutu or Tutsi. This sparked
my reflection on the idea of an “imagined” enemy that the refugees were
suspicious about, and was wary of how my appearance could affect my
interactions with the refugee women.
Although Kanembwa camp has over 12,000 refugees, it appears as though
the news spread very quickly of what I was doing in the camp and where I had
come from. I never experienced any animosity from refugees and instead received
many refugee women and men volunteers that wanted to participate in my study.
I planned on spending as much time with the refugee women as possible
and also participating in their daily chores. When I arrived in the camp I realized
that the refugee women always seemed busy carrying firewood from the forest,
tending to their children, working in the market place. There were many women
walking in and out of the refugee hospital run by IRC. The time I spent in the
camp was not long enough for me to establish a routine to spend time with groups
of women. Instead I decided to spend time with different individuals. By taking
long walks around the camp, and going to their homes and into the market, I was
able to learn about the lives of the refugees, even though it was from the view of a
few individuals.
I went to one workshop on HIV/AIDS organized by Tanzanian Christian
Refugee Service (TCRS) and STOP SIDA, an HIV/AIDS awareness organization
initiated by a Burundian refugee where refugees spoke quite vocally about their
concerns about the virus. It was during moments like this that I realized how
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21
vocal refugee women are in the Kibondo camps. The workshops were conducted
in Kirundi, and therefore I required a translator to understand what was being
said.
I planned on using a similar interview technique as Liisa Malkki. Malkki
chose to collect various narratives from the refugees in Mishamo. By doing so,
the informants were almost delivering a monologue.141 planned to achieve a
similar process, where refugee women would speak freely with me and explain to
me in the amount of detail they chose on their experiences. Malkki also looked for
patterns and recurring stories from the narratives told to her by the refugees.15
Malkki states that, “an ethnographic representation here needs to capture not only
the content of the refugee’s conception of their history, but also a feel for the
repetition and thematic unity that characterize the way people told their stories,
and sense of the specifically narrative construction of their historicity.” 1 f The \
importance of recurring themes in my research was to establish the common
perceptions that women have of a militarized life and culture. Recurring themes
also validate the existence of some form of culture established and created
purposefully or inadvertently by individuals. When interviewing the refugee
women, I chose not to have a set of questions. Instead I asked them to explain to
me their lives before the conflict, their situation during the conflict and their lives
in the camp. I urged the women to speak to me as if they were telling me a story.
14 Malkki, 53. 15 Malkki, 54. 16 Malkki, 56.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22
There were instances where some refugees expected to be asked questions
like in conventional interviews. However, the majority of the refugees understood
my technique, and they were able to tell their stories in a monologue style. It
made me wonder if they were such good story tellers because they had told their
stories many times before in different settings. I urged them to describe their
situations in detail and would only interrupt the flow of their narratives when I
required clarification or when I asked a question triggered by something they had
said. My approach required me to be highly analytical and very sensitive to the
narratives told to me by the refugee women. The narratives helped the flow of the
interview and it also made any discussion of militarism integral to the story
instead of misplaced and risky for the refugee women to talk about. Furthermore,
I did not bring up the question of militarization because it was a “natural” part of
the story. For example, while a refugee woman would explain to me the nature of
her flight, I would ask her how long the flight took, what she ate, if she heard
gunshots while fleeing, and if she encountered any militias. Another example is
that I would ask them about their security in the camp, and they would talk about
armed bandits, a problem that everyone talked about. From there I would ask a
question about any fears they had about being attacked by armed groups and if
they had ever experienced something like that. The women seemed very
comfortable when they spoke to me, and it was often hard to turn them away
when they wanted to talk for longer periods of time. Some interviews were
conducted in French, while others were in Kirundi therefore I required a translator
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23
when the informant did not speak French. I hired two refugees that were
recommended by the camp managers to work with me as translators.
Other than conducting interviews and spending time with refugee women,
I planned on observing the refugee camps and the activities that the refugees
engage in. It was important to take note of the refugee camp settings, the
demographics of the camps, and the day to day activities. These observations
were inevitably important since they allowed me to contextualize the refugee
culture through tangible factors and to see how the everyday life in the refugee
camp did not overtly reflect militarization although it was completed tainted by it.
When choosing participants to interview among the refugee women and
men I planned on using judgment sampling. Judgment sampling is the opposite of
random sampling and it entails selecting a sample group on the basis of the
research questions, or prior knowledge of the refugee camp.17 The criteria for my
sample were flexible and not entirely rigid because my view was that information
can present itself in different situations and from different individuals. I wanted at
least half of the women I interviewed to have had no formal education, because
the majority of the refugees do not have formal education. Furthermore, women in
Burundi generally do not have access to education. I wanted to speak with
educated women because they are theoretically the ones with the more radical
ideas compared to the other women. I wanted the majority of my informants to be
Hutu women considering the nature of my focus, and also because of the ethnic
17 For a detailed definition of judgmental sampling see Earl Babbie, Survey Research Methods (Wadsworth Publishing Company, 1990), 106.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24
composition of the camp. However I also spoke with some Tutsi women in order
to have a more holistic picture of the ethnic relations in the camp. The other
criteria that I had were generally based upon the experience the women had with
militia groups. I therefore looked to speak with women whose relatives are in
militia groups, women who were involved in political activism, women who had
experienced violence by militia groups and bandits from around the camp, women
who had been involved with militia groups, women who faced abuse and violence
in the refugee camp, women who contributed to political and militia groups while
in the camp, and women who feared the militia groups. In view of this, while in
the camp I sought out refugee women based on their experiences, education level
and ethnicity. This method was highly motivated by my time constraints as well
as my research questions. I relied on information from informants to decide which
refugees to interview. In order to not breach confidentiality, I only requested
information from refugees that I had a relationship with about other refugees. I
chose not to ask specific questions about people; however I would ask the
refugees that I had built a rapport with if they knew of other refugees in their
situation, or who had a certain type of experience or circumstance. For example,
since the majority of the camp is composed of Hutus, it is rare to find a Tutsi or a
Twa. I relied on other refugees to introduce me to Tutsi and Twa refugees. I
received information from refugees about which refugee women had been
attacked by bandits in the camp or refugees that were leaders in the community. I
also received information about refugees that were conducting secret political
meetings and that were involved with militias by using this method.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25
It must be noted that in order to use this method, one has to be extremely
vigilant, and be very intuitive of who to ask certain questions, and most
importantly at all times it is important to be careful how certain questions are
asked. I was very aware of my tone, how I phrased certain questions so as to not
alarm the informants, and to especially not be too intrusive and badgering for
certain risky information. I learned that instead of pushing for the information I
wanted, it was better to be patient and always be alert for new information to
present itself. For example, one day one of my informants that I spent a
significant amount of time with told me about a refugee woman who had been
involved with militia groups. We approached her and I asked her to talk to me
about her experiences. She seemed to be suspicious of my research from her body
language. I immediately invited her to a meeting I had planned to hold with
refugee women so that I would not seem to single her out. She never came to the
meeting, and I decided to not pursue her as an informant. Time did not permit and
her demeanor, although not overt, was tacit enough for me to understand that my
research made her uncomfortable. These are the sorts of calculations that I found
myself making. My time constraints made me more anxious for information;
however in order to conduct proper research patience and being intuitive is the
best policy. This method made it difficult and almost impossible to retrieve
information from highly militarized women that were involved in armed conflict
because of the sensitivity of the topic. The process of data collection made me
realize I would have to spend more time in the refugee camp to obtain the sort of
sensitive information that I was looking for.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26
Therefore, my research participant selection relied heavily on building
trust with the refugees and being able to initiate a snowball effect with the
women, and other refugees. My plan was to start a snowball effect where one or
more refugee women who have experienced some aspect of militarization will
lead me to others. Although some of my informants were selected through the
snowball effect, the majority of the informants were women who approached me
because they had heard about my research through announcements that I made at
various meetings or through word of mouth from other refugees that wanted to
participate. There was a certain level of curiosity that my research evoked that
ended up bringing many women to speak to me about their situation in the camp.
It must be noted that the self-selection by the refugees raises questions of how
representative my sample was. There were inevitably many refugee women that
were timid and not willing to participate. This was however a very useful
methodology for my study. I was not preoccupied with the representation of my
sample to provide any generalizable conclusions, however each woman that
participated had a story to tell that was very representative. The self-selection was
in effect necessary and exactly what I required for my study.
I asked the refugee women various questions that helped me uncover their
lives as militarized women. This is a very sensitive topic, and therefore I had to be
highly cognizant of various security issues that arise from asking such questions.
My main concern was to ensure that the refugee women accepted to participate
while knowing my research goals and purpose of interviewing them. Even though
a refugee woman would approach me and tell me that she had heard about my
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27
research and wanted to participate, I would ensure that she understood the risks of
participating in my research. Furthermore, I ensured that my questions reflected
my understanding of the security issues by not asking them information that might
identify them such as their place of birth or residence in Burundi, their names,
their exact age, names of their parents, or their family lineage. When retrieving
biographic information I only asked their age range, and how they identify
themselves ethnically; as Hutu, Tutsi, Twa or simply Burundian. I did not require
other information that could reveal their identity. If at all they chose to provide
information that I considered or knew would be dangerous to reveal, I ensured
them I would not include the information in my study.
I asked them different questions regarding the ethnic tension that ignited
the conflict. I asked them questions concerning their security in the refugee
camps and how they coped with and viewed militaristic activities in the camp. I
asked them to express their understanding of the role of women in Burundian
culture as well as their views about gender relations in Burundi, and in the refugee
camps. I asked them to describe their opinions about the conflict, and their
personal feelings concerning their ethnic identities (See appendix for sample
questions).
Research Problem: Fabricating Stories
While carrying out in-depth interviews I had to take into consideration that some
of the interviewees might fabricate their stories. As I started conducting my
interviews this became a big concern because at this time the topic of interest for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28
refugees is resettlement to various countries such as the United States, Australia
and Canada. I was immediately seen as a person who could provide the refugees
with the way to get resettlement. When conducting the interviews, I ended up
changing my introduction by adding that I was not conducting research to help
refugees with resettlement. There is also currently a strong belief among refugees
in the camp that the refugees in Tanzania will be forcibly repatriated. This fear is
a product of the peace agreement in Burundi. One of the first things I found out
while conducting interviews is that the school year in the camp which normally
ends in June is now scheduled to end in May. The refugees fear that this change is
because the UNHCR and the government of Tanzania are planning on forcibly
repatriating the refugees. With the option of resettlement being presented, and as
the fears of repatriation is growing, this poses a fundamental problem for any
researcher.
It was imperative for me to cross reference and triangulate results. It was
not unusual for people to explain things not the way that they happened but in a
way that they might think I, as a researcher, might want to hear, especially in
order to help them get resettlement. The refugees told me many stories and in
some situations a lot of them asked me if I could help them with resettlement to
the United States so that they could avoid repatriation. I responded by telling them
that I was in the refugee camp to conduct research and my role was not to assist in
repatriation. I also informed them that I was conducting research so that others
could learn about the refugee women’s life and experiences in Kanembwa.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29
I relied heavily on identifying patterns in the stories and comparing each
story with others in order to have an idea if the information I was gathering was
valid. It was also important for me to have detailed knowledge of the history of
Burundi before conducting the interviews in order to match the events the people
were telling with the time frame and events in the country.
Security Concerns
Throughout my interactions with the refugees I was cognizant of the security
issues and implications that could arise from my interactions with them. It must
be noted that the security concerns were some of the main reasons that some
refugee women decided not to be interviewed or participate in my study. I tried to
minimize any risk that my interaction could pose by ensuring the refugees were
fully aware of my purpose in the camp. I made sure that the refugees understood
that they did not have to participate in my research and could stop the interviews
at any point. I chose not to present the refugees with a written informed consent
forms (see appendix script); however I relied on verbal consent from individuals.
Some refugees would come to talk to me and immediately start talking
about their situations. In moments like this I chose not to write what the person
had told me until I had explained to them about my research and the risks of
taking part in the research. When writing down the information I did not write
down the names of the refugees or any information that could identity them. I had
initially planned on giving each participant pseudonyms, and finally decided
against that, and did not give any names to the refugees in my field notes. I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30
walked around with a note book in the camp where I would write information
from the interviews and my observation. I remained alert at all times of the risk
that someone could be interested in my notes. I kept my notes with me at all times
and kept my notes from previous days hidden and locked in my room in Kibondo,
instead of taking them with me to the refugee camp. This sometimes posed
problems for me when I wanted to reread some of my notes from previous days
while I was in the refugee camp.
Some considerations with this research include translation and
interpretation. The official language in Burundi is French, a language in which I
am fluent. The other official languages are Kirundi and Swahili, neither of which
I speak. The Centre for the Study of Forced Migration in Dar es Salaam informed
me that I would not require an interpreter because the majority of refugees speak
either English or French. However, when I arrived I found out that the only
refugees that spoke English and French were inevitably the educated refugees,
and there were very few of them that were educated despite the free education
provided in the camp. Trying to find individuals that spoke English and French
was very difficult and so I hired two interpreters who were also refugees. One
interpreter was a refugee woman, who helped me with interviews conducted with
women, and the other was a refugee man who helped me with interviews with
men. Both interpreters, as I learned from the beginning, were trusted by the other
refugees according to the NGO workers and the refugee leaders. Toward the end
of my research, some refugee women insisted that since they knew the male
interpreter more than the woman, they wanted him to be their interpreter. I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31
initially wanted an interpreter that was not from within the camp, and was from a
neighboring community or an NGO worker because of the security problems and
discomfort that might arise from hiring a person from within the camp. I later
found out that it was better for me to work with a refugee from the camp because
the refugees trusted a person that they knew more than they trusted NGO workers
and especially more than they could trust a local Tanzanian from a nearby
community. As I went along with my interviews I came to realize that my
interpreters were very instrumental because they had experienced similar
problems and were able to empathize with the participants in a way that someone
from the outside would not be able to do. It was interesting to see the interaction
between my interpreters and the participants because the interpreter understood
the refugees’ experiences, it helped to make the interview very informal, and very
much like a narrative and monologue by the informant with a few comments here
and there added by the interpreter, as opposed to a conventional interview with
questions and answers. I had to ensure that the people I hired as interpreters
understood the sensitivity of this topic and the security implications. I planned on
hiring only a female interpreter because I believed that women would feel
uncomfortable talking to a male interpreter about various issues and experiences.
As indicated earlier, I later learned that women did not mind who the
interpreter was, whether male or female. I explained to both my interpreters in
detail about my research topic, questions and goals. I chose not to present them
with a form to sign that would indicate that they understood the issue of
confidentiality because, although I had hired them in the vein of an interpreter, the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32
fact still remains that they are refugees with serious security risks. They were not
merely interpreters. Our lengthy conversations that educated me and clarified
certain events that the refugees explained to me made them informants at the
same time as being interpreters (See appendix for outline of confidentiality
explained to interpreters).
Trustworthiness of the Study: Ethical and Political Considerations
In order to have the most effective research, every researcher must be
cognizant of the need for the study being conducted to be trustworthy.18
Trustworthiness refers to whether the study was competently conducted, and if it
was ethically conducted.19 As regards the competence of the study, it was
necessary to ensure that the study was rigorous. In order to ensure rigor, I relied
on triangulation, and multiple methods of data gathering as noted above. This was
especially important because of the problem I faced with refugees approaching
me and talking to me because they wanted my assistance for resettlement. As
mentioned above, I was able to gather information about the situation of refugee
women from other individuals that I interviewed and some that I had informal
conversations with. These individuals spoke of various subjects that refugee
women spoke of which helped me verify the information I had obtained from the
refugees. It must be noted that when I interviewed the other informants I did not
ask them specific questions about specific refugee women.
18 Rossman and Rallis, 63. 19 Rossman and Rallis, 63.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33
Throughout my research, the most important ethical considerations were
those concerning how I, as a researcher, could influence those that I interviewed
and the environment that they are in. My research was designed in such a way that
would require mainly an observational role in my research and in-depth
interviews that were essentially detailed life stories. I was cognizant of the
security issues that my research could pose. I was also cognizant of the emotional
and psychological issues that some people face while recounting experiences that
have been traumatizing. Before each interview I ensured that the refugees
understood the emotional risks. I had a number of interviews where the refugee
women would be in tears as they told me their heart wrenching stories. In
moments like this I made sure to demonstrate that I was aware of the emotional
pain that was being felt from the memories while telling their stories. I made it
clear that they could stop the interview at any point, especially during those
moments. I also had to be cognizant of the emotional and psychological risks that
my interpreter faced when hearing the stories. I only became aware of the
potential risk for my interpreter when I realized that I was also affected by the
stories being told to me by the women. Listening to up to six hours of stories told
about refugee women’s experiences a day, I learned, can be emotionally
exhausting. I realized that if I felt this way, then surely my interpreter was feeling
the same way.
Many of the security risks associated with my study were inevitably linked
to political considerations. The fact that Burundi is still in the midst of a make
shift peace process, made the research risky. The Burundian refugee camps, as I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34
learned from my research, are highly political and are very connected to
militarism due to the nature of the conflict. It was absolutely crucial for me to
ensure the privacy and confidentiality of those I interview by interviewing them in
discrete and private places rather than out in the open where everyone was able to
hear what was being said. I was constantly thinking of the potential risks that a
person who is found talking to me about military and political activities might
face. Furthermore, my presence in the camp and my interactions with various
refugees were noticeable. Even though I remained in one area of the camp during
my entire research, and even though there are 12,500 refugees in Kanembwa
camp, it was evident that my presence was noticed by the refugees.
After recognizing just how obvious my presence was in the camp, I
planned on interacting with as many refugees as possible in order to minimize any
risk of overtly demonstrating through my interactions which refugees I was
seeking information from. When I was not interviewing refugees I walked around
the camp, and visited the market. I went to different meetings, observed the food
distribution by the UNHCR, and spent time at the IRC hospital talking with
various refugees and staff members from the SGBV center. In order to minimize
the risks associated with my study, I introduced myself to the camp manager and
attended a meeting held by a Ministry of Home Affairs official and the refugee
leaders. I later organized a focus group meeting where I invited refugee women to
come and discuss their lives in the camp with me. I explained to the women that
this meeting was important for my research and it helped establish what my
research was about to the refugee women. The meeting lasted about four hours. I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35
brought food for the women and we discussed many issues about life in the
camps, and the hardships the refugee women face. The meeting can be best
described as a social gathering where women came to speak freely about various
problems they faced in the camp, problems with their culture and the new culture
created in the refugee camp. When making the announcement for the meeting I
had made it clear that the meeting was for refugee women only, but there were
some men who came hoping to be part of the meeting. Unfortunately I had to turn
the men away and tell them that the meeting was for women only and if I had
time during my stay, I would try to hold one where everyone would be included.
Unfortunately I did not have enough time to hold such a meeting. The meeting
was very important not only as a data collection mechanism, but also it helped
build trust between me and the refugee women and it also enabled me to gain
acceptance and legitimacy in the eyes of the refugees. Many of the women who
came to the meeting asked me if they could be interviewed alone. I later realized
that for many women, even though resettlement was a motivating factor to talk to
me, there was also a therapeutic element to the interviews. I found out that some
women simply wanted someone to hear their story and to simply talk.
The other major concern I had while conducting research was how my
presence and my research would affect my informants. This was a concern I had
especially when I started to talk to individuals about ethnic strife. My
observations led me to believe that there were still ethnic problems in the camp.
Kanembwa like many of the other camps with Burundian refugees is composed of
a majority Hutu population. I found that when I spoke to Tutsi women, they often
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36
whispered their ethnicity. During the meeting that I held, one of my Hutu refugee
women friends came to me and told me that she noticed a couple of Tutsi women
in the meeting who seemed afraid to talk, and she asked me if she could invite
them to speak. This incident made me realize just how profound the ethnic tension
is in the camp.
As I started to interview various refugee women, I found it important to
find out their ethnicity, but I had to ask in such a way that did not evoke a
reaction. For the purpose of my research I wanted in-depth information about how
Hutu women view themselves. I managed to get this information mainly from
interviews that I conducted without an interpreter where I could phrase the
question in a way that would not instigate or re-instigate ethnic animosity. During
the interviews I was constantly aware of the body language and tone of the
refugee women in order to not offend them or create unintended emotions and
sentiments.
My research brought to light many issues concerning culture, tradition,
gender, politics and militarism. The concerns and limitations presented here may
raise questions about the trustworthiness and reliability of the information I
gathered. In this light it is important to keep in mind that this research is intended
to be an introduction to a more in-depth research that will follow. The lessons
learned during this research will serve as the foundation for further research that I
will conduct as well as for the research that others interested in this subject can
conduct.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37
There are certain aspects of conducting research like this that make it very
difficult for researchers. For example, this research was somewhat costly and it is
important to take into account that budgeting is very important. However even
after budgeting, there were many financial issues that arose that I did not
anticipate. For example to get to the refugee camp from Dar es Salaam it is
possible to take a bus, a train or a UNHCR flight. I had initially budgeted for a
bus ride until I learned that buses are usually attacked on the road by bandits in
the country side, they take a long time to get to the camp, and train rides take an
even longer time. I therefore resorted to taking a UNHCR flight which was
inevitably more costly. Furthermore, although my research yielded significant
information, having more time in the camp would have been extremely useful.
With time on my side I would have been able to take my time interviewing
individuals and interviewing individuals that were on my list to interview but I
never managed to interview. Furthermore, there is also a very significant
psychological effect that occurs when interviewing and thinking of the refugee
women on a constant basis during a short period of time. The refugee women’s
problems were saturated in my mind and for a period I was unable to distance
myself from the refugee women. I mainly was able to maintain myself from being
too involved in the lives and experiences of the refugee women because every
afternoon at 5:00pm the government of Tanzania has a stipulated that all workers,
visitors, researchers etc should leave the refugee camp. I was able to recoup in the
town of Kibondo every evening before returning to Kanembwa camp the next
day.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. My research methodology led me to capture an immense amount of
information. There were many aspects of my research that I plan on replicating
for further research. It was important to hear the women’s stories in a monologue
style because not only was the information rich, but it showed me that without
being prompted the women were telling me what they viewed was important to
share. My understanding of the conflict came in handy many times especially
when I was unsure of the information being told to me. It was important that I was
welcoming of all individuals that wanted to talk to me. Although I spent more
time with refugee women, when I had a moment with refugee men, I made sure to
listen to them. My experience enabled me to see the benefits of conducting
research in the field and by taking into account what the informants have to say.
The narratives provided by the refugees were extremely indicative of the self-
consciousness of the refugee population. Their understanding of their own
situation, their observations of what was happening around them, their opinions
and thoughts were pertinent. This methodology is imperative to leam the true
reality of a certain individuals. It would be very difficult to replicate the results
that I yielded with a quantitative method even though now that I have some of the
nuance, I would be able to write a more perceptive survey. However some of the
nuances and the sheer details of the lives can only be learned through in-depth
interviews and analysis relies on subjectivity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER THREE
THE QUESTION OF MILITARIZATION: DIALECTIC OF
UNCONVENTIONAL BEHAVIOR
The objective of this research is to explore the culture and nature of
militarized refugee women’s lives by analyzing their experiences. This study
analyzes the way that militarism is present in the lives of refugee women. The
definition of militarization that is used in this study is presented by Cynthia Enloe
as a “step-by-step process by which a person or a thing gradually comes to be
controlled by the military or comes to depend for its well-being on militaristic
ideas.”1 Therefore in this study militarization is understood as a process that
occurs as a result of a persistence of military activities throughout history that
affects the well-being, tradition and culture of individuals and that alters and/or
reinforces various aspects of culture. The aim of this study is to discover why
militarization of refugee women yields two dichotomous situations: on one hand
soldiers and militia rebels are known to intimidate refugee women during the
women’s flight and while women are in the camp; and on the other hand refugee
women sometimes partake in militaristic activities, sharing and relying on
militaristic ideologies. Understanding this dichotomous situation requires
understanding the history of Burundian conflict and the refugee camp by taking
into account ethnic, political, patriarchal, gender and military dimensions. In
1 Enloe,Maneuvers: the International Politics of Militarizing Women’s Lives, 3. 39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40
addition, analyzing Burundian culture, specifically patriarchal culture, and what
happens to patriarchal culture during war and in refugee camps is required. My
study is motivated by the thesis that refugee women’s militarization is a
byproduct of a combination of events in history and is influenced by the malleable
and non-static nature of culture. History is important because it shapes and
explains the past and current context, and culture is important because the
refugees leave almost everything behind in Burundi when they flee, and among
the things they bring with them is culture. Culture determines the behaviors and
values of individuals. Culture therefore is, in my view, part and parcel with
history and is an identifying element, which indicates who the refugees are and
who they become. The change in culture is especially significant because it
explains the adoption by refugee women of certain behaviors that can be labeled
as militarized behavior.
This research is also motivated by the premise that there is a strong
correlation between militarization and masculinity.2 Essentially literature points
out that the military is a masculine entity, and is therefore traditionally a domain
for men only. More explicitly put, being in the military is a man’s job. This
correlation inevitably affects militarized refugee women’s experiences. This leads
one to pose two main questions: how do women get involved with militia groups
and militaries, and what are women doing with militia groups and militaries?
Does this not go against their gender roles, especially in a highly patriarchal
2 Even though I bring up the issues of militarization, masculinity and violence against refugee women, the purpose of this paper is not to address the origins of male violence as done by Richard Wrangham and Dale Peterson (1996), especially since in this study I address violence by refugee women as well as violence against them.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41
society like the Burundian society? The correlation between militarization and
masculinity underlines the unconventionality of women’s involvement with
militaristic ventures. As Cynthia Enloe states militarism equals maleness.3 If
militarism equals maleness then this inevitably presents a challenge when women
are involved with militarism, especially in highly patriarchal societies, because
this goes against their gender roles and identities.
The nature of militarization associated with the refugee women in
Kanembwa camp is also highly correlated with a history of ethnic animosity and a
high degree of power politics that are manifested in militaristic patterns. History
has demonstrated that ethnicity and politics are militarized in Burundi. The
military is associated with Tutsis, and the rebel militias are Hutu. These military
entities are divided along ethnic lines. In addition, political parties are linked to
militarism considering that every political party has a military wing. The waves of
v iolence and massacres that have marred Burundian history have inevitably
affected Burundian society, and the remnants and effects of militarism as a means
to achieve ethnic hegemony and power have followed the refugees into the
camps. I contend, along with most analysts, that there is nothing inherent about
ethnic animosity and violence in the Burundian culture. Instead the situation of
Burundi can be deemed as a manufactured process by those with military and
political power to manipulate ethnic differences and spread violence at all levels
of society. What we learn is that this manufactured animosity has seeped into
3 Cynthia Enloe,Does Khaki Become You ? The Militarization of Women’s (Boston: Lives South End Press, 1983), 20.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42
Burundian culture. It is therefore necessary to have both history and culture as
explanatory variables instead of just one or the other, because they are highly
interrelated. Therefore, the combination of the two variables provides a holistic
explanation of the nature of refugee women’s militarization because history
captures the developments of political, ethnic and gendered dimensions of
militarization. On the other hand, culture captures the behavioral dimensions,
especially those relating to gender roles and patriarchy that permit militarization
to take place throughout history and in response to activities in history.
What follows is a discussion of the various literatures that informs my
study. To begin, I present a discussion on the two variables: culture and history.
This discussion is followed by a discussion on the definition of militarization used
in this study and its correlations with gender and masculinity. This discussion will
be followed by literature explaining why and how women become attached to
militarism. I then present literature on the violence and security issues faced by
refugee women and refugee populations in general.
Culture and History
This study puts forth two variables: culture and history as determinants of
the nature of refugee women’s militarization. History in the study mainly refers to
the Burundian political and ethnic conflict history. The main objective is to situate
women in the conflict history of Burundi with a focus on the connection between
gender and ethnicity. In this study, culture is used as an explanatory concept to
understand what has facilitated refugee women’s militarization, and to understand
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43
the nature of refugee women’s lives and experiences as a result of militarization.
A culture of militarized refugee women is very particular and in this study is
understood through the experiences of refugee women. What is meant by “culture
of militarized refugee women” is the way that militarism is present in the
everyday lives of refugee women. Culture explains how militarism has come into
their lives, and how it has affected their lives. The definition of culture that
informs this study is presented by Kevin Avruch who presents two forms of
culture, generic culture and local culture. Avruch’s definition of culture is
motivated by the work of Franc Boaz. Boaz takes a very liberal understanding of
culture by emphasizing the uniqueness of cultures of all societies, and also
emphasized that there should be no distinction between ‘high’ or ‘low’ culture.4
Therefore by generic culture, Avruch refers to human nature and what he terms
“universal attributes of human behavior.”5 By local culture, Avruch refers to
diversity, differences and pluralism of societies.6 In this study the latter is of
particular interest. As stated by Avruch, “culture is a derivative of individual
experiences, something learned or created by individual experience, something
learned or created by individuals themselves or passed on to them socially by
contemporaries or ancestors.”7 Most importantly, I share the view presented by
Avruch that individuals can have more than one culture, and that culture is not
timeless nor is it static. Therefore culture in this study is understood as behaviors
4 Kevin Avruch,Culture and Conflict Resolution (Washington DC: United States Institute of Peace Press, 2004), 7. 5 Avruch, 10. 6 Avruch, 10. 7Avruch, 5.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44
and values of individuals that have been created or learned through individual and
group experiences, by individuals and passed on by ancestors. Burundian
patriarchal culture will be the main element of analysis with emphasis on gender
roles and relations and familial ties.
The marriage between traditional culture and militaristic ideals leads one
to notice, as Johan Galtung (1996) terms, a culture of violence. Cultural violence
refers to:
The aspects of culture, the symbolic sphere of our existence—exemplified by religion, ideology, language and art, empirical science and formal science (logic, mathematics)—that can be used to justify or legitimize direct or structural violence.
The line between traditional cultural norms and militaristic ideals has been erased
in the Burundian conflict as a result of the persistent armed conflict, and history
of militarism propagated by the conflicting parties. The nature of the experiences
of militarized refugee women in Kanembwa reveals a “consummation” of
traditional culture and militaristic ideals. In my view, traditional cultural norms in
Burundi are not part and parcel with militarism, and therefore the marriage of
these dichotomous elements have come together through events in history which
have affected culture, through created and new cultures, which are revealed in the
lives of militarized refugee women. All individuals in the Kanembwa camp,
whether they know it or not, are militarized by virtue of being in that particular
camp because of its symbolic attributes as an environment created from military
activities, and also due to the military history of the camp, and inevitably of the
8 Johan Galtung, Peace by Peaceful Means: Peace and Conflict, Development and Civilization (London: Sage Publications, 1996), 196.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45
war. In addition the situation of refugee women is also based upon the capricious
gender dynamics in the refugee camp. Furthermore, and perhaps more importantly
in this study, the case of refugee women presents many questions on the
underlying effects of militarism on perceived cultural norms and what can be
labeled as unconventional behavior. It is not only the attributes of culture and
tradition that I have found to be a basis for militarization. It is in fact sometimes
the absence, change and loss of culture and tradition that has allowed for
militarization.
One author that has developed explanatory variables for the nature of
women’s participation in armed forces throughout history and across nations is
Mady Segal (1995). Segal explains that there are three variables that determine
women’s interaction and nature of participation with armed forces:
1. Military Factor: takes into account national security issues, combat support ration, military technology, force structure, military access; 2. Social Structure Factor: takes into account demographic patterns, labor force, economic factors, family; 3 . Culture Factor: takes in to account gender, family values, and public discourse regarding gender, values regarding equity.9
As indicated by Segal, the military has to be perceived as transformed in
order to make it more compatible with how women are or how women are
perceived.10 Segal’s analysis is mainly on national armies, unlike my research
which looks at women’s involvement in rebel militia groups and focuses on
refugee women. Therefore, although Segal aims to determine the degree and
nature of women’s participation in armed forces, her variables are much more
9 Mady Segal, “Women’s Military Roles Cross-Nationally: Past, Present, and Future”Gender and Society 9 no. 6 (1995), 758-759. 10 Segal, 758.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46
applicable to Western situations. 11 Iskra et al. (2002) revised Segal’s model by
adding one more variable: political factors in a country. These variables however
do not sufficiently explain the nature of refugee women’s participation in armed
conflict. The main difference between Iskra et al.’s variables and the two
variables that I propose is that the variables presented by the authors do not have a
dimension of time especially through history. An analysis of history helps to
identify factors that have an effect on the nature of women’s militarization and
that change over time. In the Burundian conflict these include ethnic animosity,
gender relations and political and military activities that develop or change over
time. The change in the level of ethnic animosity and political development is
significant especially in regards to women’s role and position. Furthermore, the
cultural variable that is presented in this study is different in that it too takes into
account time: notably change of culture over time from Burundi to the refugee
camp (the causal model explaining the militarization of refugee women
established in this study is illustrated in Figure 1.).
Militarization and Gender
This leads to the question: what is militarization? In this study, the
definition of militarization is derived from Cynthia Enloe. Enloe (2000) defines
militarization as a “step-by-step process by which a person or a thing gradually
comes to be controlled by the military or comes to depend for its well-being on
11 Darlene Iskra, Stephan Trainor, Marcia Leithauser and Mady Wechsler Segal, “Women’s Participation in Armed Forces Cross-Nationally: Expanding Segal’s Model”Current Sociology 50 no. 5 (2002), 772.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 47
CULTURE
Before Refugee Camp • Symbolism of Women and Patriarchal Culture Militarization of Refugee Women In Refugee Camp demonstrated by • Challenges to Patriarchal Culture militia and and Tradition military control • Effects of Loss o f Traditional Ties over women and Through Loss of Family Members thereby leading to • Adoption of New Behaviors and fear o f the Values— Women’s Emancipation military
Militarization of Refugee women demonstrated by HISTORY women’s dependency and Political and M ilitary History participation in • Political History: Militarized and support of Politics militaristic activities and Social and Gender History militaristic ideals • History of Group or Ethnic Animosity Militarized Ethnic identity • Role of Women in Social and Political Spheres
Figure 1. Causal Model Explaining the Militarization of Refugee Women
militaristic ideas.”12 Enloe is the primary theorist of militarization and gender, but
she does not address the issue of refugees. Therefore one of the contributions of
this study is to apply the theory of militarization to the situation of refugee
women. Enloe explores the concepts of militarization and its specific effects on
women by stating that, “women’s militarized lives are due not only to culture and
12 Enloe,Maneuvers: the international Politics of Militarizing Women’s Lives(London: University of California Press, 2000), 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48
tradition but to decisions as well.”13 My field research influenced my view that
refugee women’s militarization is a product of history, and various aspects and
attributes of culture. Furthermore, cognitive decisions are an inevitable element of
history and culture and should be tacitly understood as such.
Although culture is one of my explanatory variables, authors such as
Enloe have criticized the use of culture as an explanatory variable. The problem
that arises when addressing culture and tradition is mainly because of the inherent
nature of these two variables. Enloe explains that the assumption that
militarization of women is a cultural and traditional issue hinders people from
investigating how militarization occurs.14 In my view, the main reason why as
individuals we choose not to question culture and tradition is because culture and
tradition are intrinsic and fundamental to our identities. For example, Enloe states
that “the persistence of the presumption that women’s militarization is simply the
outcome of nature or custom is one of the things that grants gendered
militarization its stubborn longevity.”15 Therefore, Enloe uses a narrower
conception of culture than what I present. Decisions by masculinized and
militarized forces are the main focus for Enloe’s study; my research on the other
hand sees it crucial for an analysis of history and culture in explaining the
consequences of militarization and not as elements to provide agency,
emancipation, or change as is the goal for Enloe’s work.
13 Enloe,Maneuvers: the international Politics of Militarizing Women’s Lives, 34. 14 Enloe,Maneuvers: the international Politics of Militarizing Women’s Lives, 34. 15 EnloeManeuvers: the international Politics of Militarizing Women’s Lives, 34.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49
Some scholars in favor of using culture as an analytical tool for refugee
women’s situation argue that the situations of refugee women in camps are
strongly correlated to the breakdown of traditions and culture16 while others
indicate that it is a question of discriminatory factors in culture that determine
refugee women’s experiences.17 For example, UNHCR illustrates refugee
women’s susceptibility to abuse in refugee situations as a result of “family
protection and traditional authority structures breakdown and [also because]
economic support is less available.”18 On the other hand, “sometimes violence
against women is not a breakdown of a cultural norm but an exacerbation of some
of the most discriminatory cultural norms.”19 The view in this paper is that both
the change in cultural norms and traditions, as well as “discriminatory cultural
norms” leads to women’s susceptibility to militarization. The change in culture
allows women to partake in militarism voluntarily, but the discriminatory aspects
explain why women are controlled by militarism.
Not only do history and culture explain the phenomenon of militarization,
they also explain the connection between militarization and masculinity in
general. The importance of bringing up the connection between militarization and
masculinity is to underline just how unconventional refugee women’s voluntarily
interaction is, and also to explain the inevitable control that militaries have on
16 Jennifer Hyndman, “Refugee camps as conflict Zones” SitesIn of Violence: Gender and Conflict Zones edited by Wenona Giles and Jennifer Hyndman (London: University of California Press, 2004), 200. 17 Camus-Jacques, « Refugee Women : the Forgotten majority. »Refugees in and International Relations edited by Gil Loescher and Laila Monahan (New York: Oxford University Press, 1989), 146. 18 Hyndman, 201. 19 Camus-Jacques, 146.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50
refugee women, especially if we take into account Enloe’s view that militaries
control women. Enloe raises the point that militarization privileges masculinity20
by stating that, “militarization and the privileging of masculinity are both
byproducts not only of amorphous cultural beliefs but also of deliberate
decisions.”21 Furthermore Enloe adds that, “military officials have presumed that
they have needed to control not only women, but the very idea of femininity.”22
Therefore, according to Enloe, militarization not only privileges masculinity, it
also controls femininity.
Masculinity and femininity are inevitably understood in this study as
socially constructed understandings of the attributes of males and females. The
importance of bringing up the connection between militarization and masculinity
is based upon historical and cultural factors that have made the military and
military activities distinguishable activities for men. Traditionally men are the
protectors and the soldiers, and women and children are the protected. This
picture fits very well with conventional gender roles, and especially in highly
patriarchal societies. Therefore a picture of women and their interaction with
militaries can be seen as an aberration because it goes against conventional gender
roles. Burundian culture is highly patriarchal and the cultural and traditional
delineation of gender are so profound that the interaction between militia groups
and refugee women has to be questioned. How can women from a highly
20 Enloe, Maneuvers: the International Politics of Militarizing Women’s Lives, 4. 21 Enloe,Maneuvers: the International Politics of Militarizing Women’s Lives, 33. 22 Enloe,Maneuvers: the International Politics of Militarizing Women’s Lives, 34.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51
patriarchal society with strict gender roles engage in and interact with militia
groups?
The connection between masculinity and militarization affects women’s
situation in refugee camps, and this connection is not explicitly discussed by
much scholarly writing. The effects of militarization on refugee women can be
attributed to the gender dynamics of culture and tradition throughout history.
Every aspect of a Burundian refugee’s life in Kanembwa camp is a reminder of
the consequences of militarism. From the marketplace to the classrooms, each
individual in the camp assumes a symbolic identity determined by political strife,
coupled with militaristic ideals that have been created in the conflict history of the
country. The refugees in western Tanzania can be viewed, among other
descriptions, as symbols of the most pervasive militaristic ventures and political
projects.
In order to emphasize the connection between militarism and masculinity,
some authors explain that, projects of state power, control, citizenship, militarism,
revolution can be understood as masculinist projects.23 Joan Nagel outlines the
feminist notion that the national state is a masculine institution especially when
we look at the hierarchal authority structures of states where men are generally
the decision makers.24 Nagel states that:
The goal of sovereign statehood; state building; often take the form of revolutionary or anti-colonial warfare. The maintenance and exercise of statehood vis-a-vis other nation states often takes the
23 Joan Nagel, “Masculinity and Nationalism: Gender and Sexuality in the Making of the Nation” Ethics and Racial Studies (1998), 243. 24 Nagel, 251.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52
form of armed conflict. As a result nationalism and militarism seem to go in hand.25
Therefore if nationalism is attached to masculinity and nationalism is attached to
militarism, we can conclude that militarism and masculinity are connected as
well. By looking at the names by which the rebel factions call themselves in
Burundi we can see a trend of “militaristic and masculinized” motivations for
their rebellion which can be summed up as military dominance, statehood control
and their masculinized ideas of nation-building. Some examples include Parti
Pour la Liberation du Peuple Hutu and its armed wing Forces National de
Liberation, Union Pour la Liberation National, and its armed wing Forces de
Liberation National just to mention a few. These groups are highly militarized
entities because although they are political groups, they have military
components. If we take Nagel’s premise, these groups are also highly
masculinized groups based upon their projects for liberation and anti-state
projects and nationalism, especially if we understand the nation state to be a
masculine entity because it is controlled by men. Groups such as these mentioned
have been responsible for the displacement of the Burundian population, and the
way in which they lead their lives in the camps.
There is a clear understanding among certain scholars that militaries are
highly masculine institutions and they are clearly not feminine cultures (Hopton
2003, Nagel, Enloe 2000, Cockbum 2001, Kovitz 2003). Cockbum indicates that
25 Nagel, 247. 26 Erik Doxtador and Yeki Mosomothane, “Burundi: Permanent Deadlock or Tentative Peace?” In Through Fire with Water, Edited by Erik Doxtader and Charles Villa-Vicencio Trenton: Africa World Press Inc., (2003), 70.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53
women are “reminded that by biology and by tradition they are the keepers of
hearth and home, to nurture and teach children. Men, by physique and tradition,
are there to protect women, children and the nation (often also represented as the
motherland).”27 Hopton states that, “militaristic ideals represent the most extreme
expression of the values of ...hegemonic masculinity.”28 In addition to this point,
Enloe states that militaries control women, and therefore one can conclude that
the control of women leads to abuses, but also is a way to fuel “hegemonic
masculinity.” For example, guerilla fighters often attack civilians, rape women
and depend on the local populations for food, and they violently take property,
food and sex from the locals.29 Meredith Turshen (1998) states that violence is
one of the most obvious military values and that militaries intensify women’s
subordination.30 Turshen adds that:
Rape is committed to boost the soldiers’ morale, to feed soldiers hatred of the enemy, their sense of superiority, and to keep them fighting. Rape is one kind of war booty, women are raped because war intensifies men’s sense of entitlement, superiority, avidity, and social license to rape; rape is a weapon of war used to spread political terror; rape can destabilize a society and break its resistance; rape is a form of torture; gang rapes in public terrorize and humiliate women; rape is use to terrorize and silence women and to force them to flee home, families and communities...31
On the other side, as noted by various authors, we notice that the
conventional and traditional connection of militaries and masculinity is being
27 Cynthia Cockbum, “Gendered Dynamics of Armed Conflict and Political Violence”Victims, in Perpetrators or Actors? Gender, Armed Conflict and Political Violence (London: Zed Books, 2001), 19. 28 John Hopton, “The State and Military Masculinity”Military in Masculinities: Identity and the State (London: Praeger, 2003), 111. 29 Meredith Turshen, “Women’s War Stories” Whatin Women do in Wartime (London: Zed Book, 1998), 2. 30 Turshen, 5. 31 Turshen, 11.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54
challenged by the voluntary participation of refugee women. Harry West (2004)
conducted research in Mozambique that speaks to this question by pointing out
that women’s interaction in militia groups has been considered as an aberration to
traditional norms. The notion of women as pacifists who have no interaction with
militarism is challenged in political refugee camps. These interactions are part of
the elements that can clearly distinguish refugees from others that stayed behind
during war. The interaction is so pervasive that it not only alters the way that
refugee women identify themselves, however, as shown from my study, it also
influences the way others that remained behind in the country of origin view
refugee women.
However, women that interact with and are influenced by militarism do
not always take on very significant roles. The highly militarized Burundian
refugee women who chose to join the rebel militias often did not take up high
•a <2 ranking positions. Although there is the sense of female marginalization in
armed groups (Nordstrom 2005, Kovitz 2003) women still join these groups. It
has been noted by Nordstrom that:
Women and girls [are usually] not allowed military positions or equipment, but they transported messages, munitions, supplies and food. They [are] a backbone of the war: running arms, procuring survival necessities, acting as communications systems, doing reconnaissance.33
Nordstrom indicates that in Mozambique, women, whether as combatants or not,
are usually the ones who are most targeted to be tortured or killed by armed
32 Waldemar Very, interviewed by author by email. 33 Carolyn Nordstrom, “Gendered War”Studies in Conflict and Terrorism (2005), 401.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55
groups.34 Other scholars such as Marcia Kovitz (who references Deidre Meintel)
dispute and state that, “there is no universal taboo against the bearing of arms by
women.”35 However Kovitz points out that one of the main reasons for women’s
exclusion in armed conflict and marginalization is due to culture.36 She states that,
“what makes women’s presence so contentious is not what they are—their
purported essential physiology—but what they represent, their associated social
attributes.”37 Therefore I contend that the issue of women’s marginalization in
armed conflict is a highly situational questions and it is a matter of culture and
history that explains to what degree militarized women are marginalized by armed
forces or targeted by these forces. History in certain parts of the world such as
Benin has shown that women were the dominant warriors (Kovitz 2003,
Mazurana 2002), in traditions such as the Hinduism, some goddesses such as
Durga are fierce warriors.38 In addition, in present day Sri-Lanka, women are
solicited to join rebel groups for the purpose of demonstrating a need for
liberation and modernity. In other places this is clearly not the case, and
representation of women as mothers and care-givers takes precedent over women
as armed protectors.
In addition to the question of marginalization faced by female combatants,
Nordstrom also states that:
34 Nordstrom, “Gendered War,” 402. 35 Marcia Kovitz, “The Roots of Military Masculinity”Military in Masculinities: Identity and the State (London: Praeger, 2003), 4. 36 Kovitz, 5. 37 Kovitz, 6. 38 Radhika Coomaraswamy, “A question of Honor: Women, Ethnicity and Armed Conflict” in Feminists Under Fire: Exchanges Across War Zones (Toronto: Between the Lines, 2003), 93.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56
To be female today in a war zone is to be the frontlines. But women generally enter the battle armed with little more than their intelligence and values—they are often denied access to the weapons and training they require to fight back. So they create.39
Nordstrom raises a very crucial point, and it raises the question of how women
protect themselves. Women are known to commit crimes against others and
literature has shown that the words “women and pacifist” are not synonymous.40
The idea of female combatants contradicts what Mary-Jane Fox addresses as a
prevalent myth that girls are pacifists that cannot fight and that when they are in
the company of militia or military groups, they are merely camp followers (Fox
2004, Enloe 2000, McKay 2005, Manchanda 2001). Other authors, such as Susan
McKay, have addressed this myth and the violence perpetrated by girls and other
females in general. In this study, various roles of militarized women are taken into
consideration. As noted by Enloe, militarization is not only about women taking
up arms, but it is also the process that women go through from militaristic
influences.
Even though women are often marginalized in militaries and women are
targeted during conflict, on countless instances women are asked to engage in
some form of violence such as deciding the fate of prisoners.41 The nature of
militarization is such that individuals and groups of people take on different roles
from their normal roles. Women are often forced to participate in atrocities and on
other occasions they choose to engage in these activities. The fluidity of cultures
39 Nordstrom, “Gendered War,” 404. 40 Hopton, 116. 41 Kovitz, 4.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57
and traditions, and the remnants of past histories that linger in the memories of
individuals determine the roles that individuals take on during conflict.
A growing amount of literature touches upon militarization of women in
the context of women as combatants. A discussion of refugee women combatants
highlights the most pervasive elements of militarization. However it must be
noted that the lives of political refugees are militarized whether or not they are
involved in armed conflict. Wenona Giles is one of the few scholars that has
provided a conceptual framework in which to understand the concept of
militarization of refugee camp settings. She analyzes the gendered nature of
nationalism and war by focusing specifically on the violation of women, homes
and house holds in wartime.42 She states that in order to, “understand nationalism,
gendered ideas of home and households must be analytically addresses as
dynamic structures.”43 Giles defines a home as, “complex of ideas that may refer
to a country of origin and an associated national identity.”44 The concept of
nationalism, I find is highly attached to militarism, and is so demonstrated in the
case of Burundian refugees. The home is very representative in a refugee setting
for the national identity, and cultural identity. Giles states that some of the, “key
social relations of war and militarism can be considered as being immanent in
homes at three levels: in the domestic sphere, in the community, and in the
42 Wenona Giles, “ Gendered Violence in War: Reflections on Transnationalist and Comparative Frameworks in Militarized Conflict Zones” inEngendering Forced Migration (New York: Berghahn Books, 1999), 83. 43 Giles, 84. 44 Giles, 85.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58
remembrances of nations of origin.”45 Giles presents a clear conceptualization of
how the essence of refugee homes as representational of national identity yields
the militarization of all refugees. By linking the remembrance of nations of
origins into this conceptualization, an emphasis on the role of history is presented.
Giles states that, “the way that patriarchy is constructed through social and
ideological relations in homes is clearly reflected in militarization and
militarization does not only occur in so called war zones; nor is it associated with
one specific site or static phenomenon.”46 She references Enloe who states that
“militarization is informed and informs gender relations in factories, social
stations, refugee camps, and bedrooms.”47 Giles states that refugee camps become
places that are homes to refugees but also places of violence because they are the
space where there is interplay between nationalist fantasies of power and
ideologies of home.48 Giles also states that, “gendered militarized conflict zones
are, in a sense, the backdrop to how women are constrained to operate in
households and are characterized chiefly by their roles arising from the home.”49
Women Warriors
The phenomenon of refugee women combatants has led me to question
how women become attached to militia groups in the first place. The culture that
they share as refugee women, working voluntarily or involuntarily with rebel
45 Giles, 86. 46 Giles, 89. 47 Enloe cited in Giles, 89. 48 Giles, 90. 49 Giles, 92.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59
militias or for militaristic ideals, or relying on militaristic factions and ideals is of
interest especially concerning their ethnic and gender identity. The literature on
female combatants seems to be growing, although not much has been written
about the militarization of refugee women. There are three recurring themes of
why women join militia groups in literature (1) women join for liberation and
emancipation, (2) women join because they are abducted and forced to join (3)
and women join because the military or militia groups are their only means of
survival. Although many contest the idea of girls joining armed forces, girls may
volunteer to associate with militant factions. I contend that in any refugee setting,
it is inevitable that many women will associate with and join militia groups
voluntarily. The hopelessness felt in the refugee setting, or the anger and need for
revenge, or the sheer belief that being involved with these groups will bring them
power, are not ideas and feelings that are inconsistent with refugee women. Girls
join for various reasons including, the need to leave the burden of domestic work
at home, to join other family members, or even to seek adventure.50
The highest level of militarization of refugee women is produced when a
refugee women chooses to leave the “sanctuary” of the camp to join rebel militias.
Militarization of refugee women is very complex, furthermore not only are
women involved with militarized forces, however young girls are as well. Many
have written about child soldiers that are boys however the situation of girl child
soldier has only recently been explored. Mary-Jane Fox (2004) indicates that “the
50 Susan McKay, « Girls as Weapons of Terror in Northern Uganda and Sierra Leonean Rebel Fighting Forces »Studies in Conflict and Terrorism (2005), 388.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60
most extreme abuses of child soldier recruitment and use are often found within
non-state armed groups.”51 Fox also states that, “it is here that we find reports of
minors being kidnapped from their homes and schools, physically and sexually
abused, forced to choose between killing other minors who are non-compliant or
being killed themselves and a range of similar offenses.”52
There is also a delineation that has been made by some scholars between
women and girls when addressing the female combatant phenomenon (Fox 2004,
McKay 2005). Fox states that the issue of girl-soldiers is first and foremost a
child-protection issue, and should only be viewed using feminist theory on
women and conflict in limited ways.53 Fox states that the power dynamics
involved in the girl soldier phenomenon are those between adult and child.54 This
premise should be revisited because in many societies, especially in Africa, there
is a fine line between adult and child especially as regards to females, therefore
the issue of child soldiers is better understood as a child protection issue when
understanding the issue in a Western lens. However when one looks at a situation
and takes into account the cultural surroundings and context in which the
phenomenon is occurring, specifically in Africa in my study, it should be
emphasized that “the Western cultural notion and rite of passage of being a
teenager do not exist, pubescent girls are considered to be women after initiation
51 Mary Jane Fox, “ Girl Soldiers: Human Security and Gendered Insecurity”Security Dialogue (2004), 468. Fox, 468. 53 Fox, 469. 54 Fox, 470.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61
rites.”55 Therefore a fifteen year old in Western standards is still a child while in
other situations in Africa and other places, she is considered a woman. Instead of
creating delineations between girl child soldiers and women soldiers, it is
necessary for my study to acknowledge the overlap. That is why in this analysis I
will examine the experiences and militarization of female refugee combatants of
all ages.
Some of the most recurring cases of analysis as regards female combatants
are of Sri Lankan women, Northern Irish women, Sierra Leonean girls and
Ugandan girls. Even though some of these regions are culturally, historically and
ethnically different from my area of interest, the Great Lakes Region, the analyses
of all these regions are informative because due to the paucity of significant
literature, they provide an understanding of the experiences of militarized women.
Miranda Alison reiterates this point when she states that “since security has
traditionally been conceptualized in masculinized, military terms and women have
been excluded from this; the experiences and roles of women have rarely been of
interest in literature of security.”56
Women’s Militarization and Liberation
In this study I place a large emphasize on the influence of non-state armed
groups such as rebel groups and militias on the militarization of refugee women
because of what Aristide R. Zolberg et al.(1989) have termed as the phenomenon
55 McKay, 387. 56 Alison, 447.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62
of warrior-refugees who are refugees with militaristic and political agendas and
therefore associated with an armed struggle.57 In addition, the warrior-refugees
are the main militaristic actors around the refugee camps in Kanembwa. Dyan
Mazurana (2002) defines militias as “bodies of nonprofessional citizen-solders; in
the majority of cases, militias operate with some opposition groups from the
government or are condoned by the government.”58 The importance of refugee
women’s interaction with militias is based on the assumption that these groups
provide an interesting channel of “opportunity” for women who are with them
(Alison 2004, Ana Cristina Ibanez 1998). The historical events of the Burundian
conflict demonstrate that political power is associated with militaristic activities.
Therefore, when women interact with militias it is because they choose to align
themselves with potential victors. Alison states that, “theoretical analysis suggests
that anti-state, so called liberatory nationalisms often provide a greater degree of
ideological and practical space for women to participate as combatants than do
institutionalized states or pro-state nationalism.”59 This theoretical analysis also
applies to refugee women, whose lives have been altered significantly by conflict
and the uncertainties of displacement. Alison analyses why rebel and militia
groups have been inviting and seeking women’s participation. She references
Valentine Moghadam who differentiates between two models of revolution by
57 Aristide R. Zolberg et al.,Escaping from Violence: Conflict and the Refugee Crisis in the Developing World (New York: Oxford University Press, 1989), 276. 58 Dyan Mazurana et al., « Girls in Fighting Forces and Groups : Their recruitment, Participation, Demobilization, and Reintegration »Peace and Conflict: Journal of Peace Psychology (2002), 103. 59 Miranda Alison, « Women as Agents of Political Violence : Gendering SecuritySecurity » Dialogue (2004), 448.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63
rebel groups: one that indicates ideas of equality linked to progress and
modernization, and the other where women’s behaviors and appearances are
viewed as central to cultural rejuvenation.60 Regardless of the motivation
presented by the rebel militia group, the reasons why refugee women join are
highly motivated by the environment of being in a refugee camp, the experiences
that led to their displacement and to some degree the ideology of the group.
Many authors concur that one of the main reasons women join rebel
groups is for liberation (Ibanez 2001, West 2004). For example, women’s role in
El Salvador, due to patriarchy much like in many other societies in the world, was
limited to subordination and intended to be very discrete. Ibanez states that
women joined the guerillas, “because their lives and outlooks had been changed
by other experiences and because they believed they would be listened to in these
[political-military] organizations.”61 Therefore as regards to the question of why
women take up such unconventional roles, one is that they are seeking ports of
entries into a world where their voices might be heard. Liberation, however, is but
one of the main reasons that women join rebel groups. Reasons for joining these
groups, I contend depend on culture, tradition as well as history.
The main argument presented by Alison is very similar to the one made by
Rita Manchanda. Both authors take the stance that female combatants, especially
in the case of the LTTE, can be viewed as being empowered. This view is clearly
different from the prevalence of victimization of women in literature, media and
“ Alison, 452. 61 Ana Cristina Ibanez, « El Salvador: War and Untold Stories—Women Guerrillas » inVictims, Perpetrators or Actors? Gender, Armed Conflict and Political Violence (London: Zed books, 2001 ),120-121.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64
other forums that address women’s experiences in conflict (Manchanda: 2001,
D’Amico: 2005, Cockbum: 1998). In addition, Manchanda states that women
have survival strategies and have been able to mobilize themselves for resistance
and negotiations of power. She provides several examples from South Asian
conflicts by referring to the LTTE women militias and the Maoist women
guerrillas in Nepal.
Several authors thus far in my analysis agree that militancy leads to or is a
port of entry into the realm of empowerment and liberation for women. As
Manchanda states it is a matter of “empowerment through arms, namely the
woman militant.”62 Women are not only empowered however they also exhibit
what many may term as a change in behavior from their traditional gender roles
and as being more peaceful than men. Alison and Hopton challenge the notion
that women are more peaceful than men. In her research on the LTTE women,
Alison notes that the women’s military wing is well-organized and highly
disciplined.63 Her research also led her to find that Tamil women are perceived to
be more dangerous than men.64 The fact that women are perceived to be more
dangerous than men raises the question of whether they are merely being
perceived as dangerous relative to their “normal” gender role. It can be noted,
especially in highly patriarchal societies, women who take up unconventional
roles will shock others. Regardless of how dangerous a woman can truly be the
important point is that by being dangerous they generally refute the notion that the
62 Manchanda, 101. 63 Alison, 450. 64 Alison, 457.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65
words woman and pacifist are synonymous. Rita Manchanda raises a similar
remark as regards women combatants from Nepal by stating that Nepalese women
perform some of the most atrocious acts of violence.65
However the question that should be asked is: are women empowered
even after they are no longer engaged with militia groups? Manchanda, much like
Alison raise an important note that requires a closer analysis of how women
militants are treated after the conflict is over. Militancy might empower women
while conflict is still going on, however, when the conflict is over, they might
have to return to their “normail” patriarchal lives. For example, Susan McKay
notes that when the war is over in some areas (notably Sierra Leone) girls are not
viewed to have taken a significant role in combat and therefore do not even
undergo the Disarmament Demobilization and Reintegration (DDR) programs,
and community members also react with hostility to girls who come back from
their association with rebel militias.66
McKay notes that girls generally hide their engagement with rebel militias
when the conflict is over because their actions with rebel groups are seen as
breaking social and community norms.67 Twagiramiriya and Turshen also note an
interesting aspect of the ethnic conflict in Rwanda, not only of women who were
abused by militias but of women who were accused of having any sort of
connection to the militia groups. They note that, “young girls accused of being
Interahamwe were savagely violated; stripped naked, they were kicked with
65 Rita Manchanda, « Ambivalent Gains in South Asian Conflicts » inThe Aftermath: Women in Post Conflict Transformation (London: Zed Book, 2001)118. 66 McKay, 393. 67 McKay, 388.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66
boots, beaten with wooden sticks, and knives were used to cut their genitals.”68
Turshen and Twagiramariya note that when the RPF took control of Rwanda and
ended the genocide, they forced girls to offer themselves to them or be accused of
affiliating with the “disposed regime and put in jail.”69
Forced Participation or Survival Mechanisms
Although women have joined rebel groups for liberation, this study can
not ignore the fact that many women are abducted and forced to join these groups.
Furthermore, women do not always take up the romanticized role of liberated and
strong women while in the presence of these groups. Women take on other
militarized roles, and they experience things that are unfathomable. For instance
Susan Shepler (2002) addresses the various roles that girl-soldiers had while
associating with Sierra Leonean rebels. Many were abducted, some joined for
protection, and some seem to have joined at their parents urging.70 There is a clear
difference between forced and voluntary participation with armed group. Not all
situations are aligned with a romanticized vision for liberation. Girls that have
been coerced into joining have been forced to join in different scenarios.
Dyan Mazurana, Susan McKay, Kristopher C. Carlson and Janel C.
Kasper (2002) indicate that girls have been at war before from Joan D’Arc in
68 Twagiramariya and Turshen, “Favours to Give and Consenting Victims the ScAual Politics of Survival in Rwanda: in What Women do in Wartime (London: Zed Book, 1998), 106. 69 Twagiramariya and Turshen, 109. 70 Susan Shepler, “ Post-war trajectories for girls associated with the fighting in forces in Sierra Leone”Politique Africaine (2002), 4-5.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67
1492 71 to the Dahomey kingdom women warriors of Benin who were the warriors
that thrived in the 18th and 19th Century.72 Therefore the idea of female
combatants is not new, however the motivations for being militarized women
vary. The authors point out that reasons that some girls join is not always due to
coercion nor is it a question of liberation. In places like the DRC it is because if
they are not in the militia groups, they face hunger on the streets.73 Therefore, in
some instances women join out of pure coercion, others for liberation and many
others because they need to survive.
Militarized Women’s Roles
Girls and women are not only forced to participate in armed groups,
however they also take on diverse roles and activities while with the militia
groups. Many studies that have been conducted on the issue of girl soldiers in
Uganda and Sierra Leone reveal the diversity of roles and activities that girls and
women engage in while with militia groups. The magnitude of the issue can be
expressed statistically. The LRA has 70-80% child combatants, 30% of which are
girls.74 As part of rebel militias, girls partake in various acts including terrorist
mutilation, ritualistic murder, forced cannibalism, and drug use.75
The primary roles noted for girls associated with militias include providers
of sex, being wives and giving birth to children that are raised to be rebel
71 Mazurana, 101. 72 Mazurana, 102. 73 Mazurana, 106. 74 Fox, 472. 75 McKay, 386.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68
militias• K\ and in the Ugandan forces they sell goods, prepare food, stealing food
and moving weapons.77 Girls also receive military training and are forced to
perform many atrocities. In many instances they are told to kill or be killed.78
It appears that militarization sometimes provides a way for women to try and fight
the status quo of patriarchy, but others still face abuse while in these groups. For
example the lives of Sudanese female combatants do not change with their role as
combatants. Women combatants in the southern Sudanese militia groups face the
same abuse as women who are non-combatants. They are raped and forced into
marriages as a constant reminder that they are “subordinate.”79 South Sudanese
women combatants have not only been raped by Sudanese militia however they
have also been sexually abused by Uganda soldiers that sometimes train them.80
This is a clear indication that women that are involved with militia groups are
vulnerable to rape as well (Halim 1998). This is only to show that rape is not only
a weapon used to instill fear in the enemy and to reestablish the feelings of
superiority against the enemy, but it is also used to instill a feeling of superiority
and dominance, in male bonding, showing that men have control over their own
women. In addition, Turshen states that rape is pervasive in non-military armies
where there are female combatants and where the distinction between combatant
76 McKay, 388. 77 McKay, 390. 78 McKay, 392. 79 Asma Abdel Halim, “Attack with a friendly weapon” inWhat Women do in Wartime (London: Zed Book, 1998), 96. 80 Halim, 97.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69
and non-combatant is blurred,81 Such experiences of women in rebel groups raise
questions concerning the psychological well-being of women.
Dyan Mazurana, Susan McKay, Kristopher C. Carlson and Janel C.
Kasper (2002) indicate that, “very little is known regarding the distinct physical,
emotional, and spiritual long-term, even lifetime, effects of girls’ experiences
within fighting forces and groups.”82 They note that girls take on roles and chores
that they would in non-military times such as cleaning, cooking, washing and
taking care of children for the militia and military groups. These are militarized
0 1 chores because they are done to maintain and sustain a militaristic lifestyle.
Violence: against Women in War: Rape. Ethnicity and Identity
The literature above points to reasons for women’s militarization such as;
many women join voluntarily for the purpose of finding a port of entry into a
world where they are liberated and their voices are heard. However literature also
reminds us that women are not always emancipated when they are in constant
interaction with militia groups, but instead they face abuse and violence. The
recurring themes from Cockbum to Turshen are that militaries and armed groups
have a propensity for violence. The tendency for violence of armed forces is
reiterated by other authors such as Deborah Harrison (2003). In addressing gender
and war, Clotilde Twagiramariya and Meredith Turshen (1998) write about the
81 Turshen, 12. 82 Mazurana, 100. 83 Mazurana, 109, and Enloc Maneuvers: the international Politics of Militarizing Women’s Lives, 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70
politics of rape during and after the Rwandan genocide and how rape in Rwanda
was politicized for the purpose of ethnic cleansing.84 The Rwandan case is very
similar to that of Burundi. The two authors raise a clearly important point that
women are targeted because of their ethnicity and their gender.85 This is clearly
the case in Burundi as well.
Ethnicity and gender are two factors that are separate although very
intertwined. The conflict in Burundi clearly reveals delineations of ethnicity.
Furthermore, gender discrimination is partly fueled from ethnic animosity as
women from a certain ethnic group are systematically targeted and killed. Ethnic
rivalry in Rwandan and Burundian conflict have their roots in complex and often
superficial historical, social, racial and colonial factors that revolve around issues
of identity. As Cynthia Cockbum points out, identity issues stem from complex
factors that contribute and lead to inclusions or exclusions of others.86 Ethnic
identity is inevitably one of the main sources of social cohesion. Refugee women
therefore are faced with the challenge of constructing an identity that allows for
cohesion while in the camp. What is found, however, is that the structure of
certain camps, Kanembwa for example, where the refugee population is
purposefully somewhat homogeneous in terms of ethnicity, creates ethnic
tensions.87 The fact that the camp is majority Hutu reinforces and adds to the
militaristic elements that exist at all levels of the camp because Hutus in the camp
share experiences of a common enemy, therefore this can create ethnic solidarity
84 Twagiramariya and Turshen, 105. 85 Twagiramariya and Turshen, 103. 86 Cynthia Cockbum,The Space Between Us (London: Zed Books, 1998), 212. 87 The majority of refugees are Hutu, but there are a few Tutsi refugees in the camp.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71
and also increase hatred for the few Tutsis that reside in the camp. As indicated by
Georg Simmel, “conflict sets boundaries between groups within a social system of
strengthening group consciousness and awareness of separateness, thus
establishing the identity of groups within the system.”88 Therefore, it can be stated
that the Burundian conflict has strengthened ethnic solidarity, and the few Tutsis
in a predominantly Hutu refugee camp experience the effects of Hutu ethnic
solidarity. Cockbum states that, “dominant groups maintain hegemony for the
most part by discursive means rather than by direct force, mobilizing consent by
inclining us towards particular identifications.”89 The cohesion of a group and
how it identifies itself is based upon physical and tangible commonality and
shared ideology, viewed by the group members as the basis for inclusion and
exclusion of others. I contend that ethnicity provides a distinction of the enemy
even when ethnic lines are blurred and made relatively superficial through
intermarriage. Furthermore this aspect of ethnicity makes it an element that is
easily militarized, as seen in the Burundian conflict. This raises the very important
point of whether refugee women have an affinity to hold on strongly to their
ethnic identity, since the conflict and ideologies propagated by militaristic
factions around the camp are ethnically based.
88 Lewis Coser,The Functions of Social Conflict (New York: The Free Press, 1956), 34. 89 Cockburn,The Space Between Us, 213.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72
Refugee Women’s Security
By introducing the question of militarization, ethnic animosity and
violence in refugee camps this evokes a closer look at the security of refugee
women. Militia groups as noted above present many security concerns for
individuals, especially women. Refugees and other migrants are often seen as
security issues (Loescher 1994, Karen Jacobsen and Steven Wilkenson 1993).
Refugees can be seen as the referent object (the object requiring security),90 or
they can be seen as the source of security problems. In this study, the culture of
militarized refugee women raises specific security issues. For starters my study
has shown that refugee women face security issues because of their susceptibility
to abuse, especially rape. However, by the same token, an environment of fragile
security due to militarization almost automatically involves women who become
sources of security concerns. Therefore although in this study I present a human
security91 approach to refugee’s situation and therefore making refugees the
referent object, I also present the other side of the coin, which is that refugees can
be perceived as the source of insecurity to other refugees, and potentially to the
people that remained in the home country and to the host communities. By
supporting militaristic ventures, militia members, engaging in political activities,
90 Andrew Mach, “A signifier of Shared Values” What is Human Security?Security Dialogue 35 no. 3 (2004), 368. According to Mack, “for realists, the state is the referent object of security but the realist paradigm is incapable of dealing with the threat states pose to their own citizens. This is the primary reason why proponents of human security argue that the individual should be the referent object of security.” 91 The United Nations Development Program labels human security, among other things, as people-centered (UNDP 1994). Furthermore, the UNDP report states that human security is about freedom from fear and freedom from want (UNDP 1994).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73
and being part of militia forces militarized refugee women pose security threats to
other individuals and communities.
Even though refugees pose security problems, refugees are individuals
characterized as having security problems. Burundian refugee women, much like
other refugees have faced, and continue to face security problems in the camps of
western Tanzania. The magnitude of the security concerns associated with
refugees can be conceptualized by the definition of a refugee. A legal definition of
a refugee is presented by the United Nations Convention Relating to the Status of
Refugees:92
Owing to well founded fear of being persecuted for reasons of race, religion, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or owing to such fear is unwilling to avail himself of the protection of that country; or who not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or owing such fear is unwilling to return to it.93
From this definition, one is quick to gather that a refugee is a person who fears
persecution and therefore is seeking sanctuary and safety. This legalistic
definition is unfortunately incomplete because it leaves out the persecution and
violence refugees face while they are in “sanctuary.” Elly-Elikunda Mtanga
(1989) expresses concerns for the need of more protection for refugee physical
security by writing that:
As armed attacks on civilian refugee population’s increase in number, the need for a new refugee instrument to deal more
92 Adopted in 1951 by the UN General Assembly. 93 United Nations Conventions and Protocol Relating to the Status of Refugees: Text of the 1951 Convention and the 1967 Protocol available from http://www.unhcr.org/cgi- bin/texis/vtx/protect/opendoc.pdf?tbl=PROTECTION&id=3b66c2aalO.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74
specifically with the problem of physical safety of refugees in particular from military or armed attacks is compelling.94
Even though the United Nations High Commission for Refugee (UNHCR)
has established and produced many documents relating to the security of refugee
women, their continues to be a need for the emphasis on refugee women’ s
security. As stated by Genevieve Camus-Jacques (1989) refugee men and women
are equally in need of protection; however, women more often than men need
protection because of their sex.95 In addition to threats such as rape, sexual
harassment, torture, forced or voluntary prostitution, refugee women also face
armed robbery, forced or voluntary recruitment into militia groups and much
more.
In addition, even though refugee women suffer different types of abuses
not all abuses are perpetrated by the military and armed factions surrounding the
camps. The vulnerability of the refugee population is such that almost anyone can
perpetrate abuses and violence against them. As Susan Forbes Martin (2004)
explains, “some of the perpetrators around these camps include male refugees
from within the camps, military personnel from the host or home country as well
as aid workers.”96
Based upon authors such as Susan Forbes Martin, one is quick to infer that
refugee camps demonstrate issues of gender and power relations with women
94 Elly Mtanga “Military and Armed Attacks on Refugee Camps” inRefugees and International Relations edited by Gil Loescher and Laila Monahan (New York: Oxford University Press, 1989), 89. 95 Camus-Jacques, 14. 96 Susan Forbes Martin,Refugee Women (New York: Lexington Books, 2004), 47. See also Radhika Coomaraswamy, “A Question of Honor: women, Ethnicity, and Armed Conflict” in Feminists Under Fire (Toronto: Between the Lines, 2003), 91-101.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75
facing many abuses from different actors. Abuse against refugee women can be
seen in the lens of what I term a dominance paradigm relating to dominance by
one group over another on all levels especially gender and ethnicity. Cockbum’s
(2001) work is very informative in analyzing some of the experiences of refugee
women. Cockbum discusses the importance of taking power relations of gender
into consideration, and how being alert to these issues “enables us to see features
of armed conflict and political violence that are otherwise overlooked.”97
Cockbum states that although gender formulations vary and are diverse in
different cultures, dominance by men is pervasive.98
There are many reasons that keep refugee women from receiving attention
especially when they face abuse, some of which is not entirely a result of
negligence of refugee women by the NGO’s or other outside community. Refugee
women sometimes find themselves in an internal tug of war, where they debate on
whether to give significance and importance to the abuses they face or simply to
accept their situations. Some authors have argued that refugee women’s abuses in
places such as Sierra Leone are often forgotten or not efficiently reported by NGO
workers (Camus-Jacques 1989, Odd Einer Olsen and Kristin S. Schaffscher
2004). Camus-Jacques refers to refugee women as the “forgotten majority.”99 As
revealed from my research in Kanembwa and as noted by numerous authors,
women sometimes fear reporting cases of abuse due to the perceptions and stigma
associated with experiences of abuse. Furthermore, they also fear the
97 Cockburn, “Gendered Dynamics of Armed Conflict and Political Violence,” 13. 98 Cockburn, “Gendered Dynamics of Armed Conflict and Political Violence,” 15. 99 Camus-Jacques, 141.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76
repercussions they might face from perpetrators if they seek assistance against the
perpetrators.
The risk of conducting a study on refugee women is that the refugee
women’s situation can be misinterpreted as one only of victims of violence,
abuse, hunger and much more. A study that presents refugee women as victims
risks essentializing women. Cynthia Cockbum state that, “essentialism is not
merely an interesting theoretical concept. It is a dangerous political force,
designed to shore up differences and inequalities, to sustain domination. It
operates through stereotypes that fix identity in eternal dualism: woman victim,
male warrior, trusty compatriot, degenerate foreigner.”100 This study therefore
presents the aspect that refugee women are sometimes involved in and with
elements of violence. Camus-Jacques, Olsen and Schaffscher and Mtanga
describe the problems refugee women face in camps and express the need to
increase protection of refugee women. Neither addresses the other side of the
spectrum which would be a more complex view of women that sees them not
merely as victims, but as survivors or even as perpetrators of violence. My
approach is to analyze refugee women in a non-categorical lens where they can be
victims, survivors and perpetrators of violence.
Militarization and Arms Proliferation
The literature presented so far demonstrates that militarization can be seen
as phenomenon hinging upon intangible issues and constructs such as those
100 Cockburn, 13.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77
pertaining to gender, patriarchy, ethnicity and dominance. Aside from these
intangible factors, there are tangible concepts that fuel militarization as well that
need to be addressed. The tangible concepts are mainly rooted in regional security
concerns in Africa, especially in regards to the proliferation of arms. As explained
by Volman (1998) the effects of militarization can be viewed in terms of
prevalence of arms in Africa. The significance of arms proliferation in Africa as a
question of refugee militarization is pronounced when one analyzes the violent
activities in and around the camp as a result of the use of firearms by refugees and
individuals in and around the refugee camps. The security complexes brought
about by colonial powers, the global superpowers, extra continental states,
international and regional organizations, and other non-state actors can not be
ignored in exploring the tangible ways in which Africa has been militarized.101
One of the main consequences of activities concerning the many actors on the
African continent is the prevalence of arms, especially light weaponry. The Great
Lakes Region is particularly an area in Africa that has experienced persistent
armed conflict. The flow of arms in the area is so pervasive that it has seeped into
the lives of local communities living in the region, and especially those around
refugee camps. Cockburn states that, “in societies that will later know open
violence there is often a prior increase in militarization and the quantity of
weapons flowing into the area.”102 Volman indicates that the prevalence of
weapons is an indication of militarization, and that the prevalence of weapons in
101 Daniel Volman, “The Militarization of Africa” inWhat Women do in Wartime (London: Zed Book, 1998), 150. 102 Cockbum, The Space Between Us, 18.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78
an area leads to the disruption of African women’s lives. He indicates this
connection without detailing the nuanced process of how the disruption of
women’s lives occurs. However, the fact that Africa is awash in arms, and that
these arms are coming from a variety of channels helps one conceptualize the
present day problems of arms proliferation, its connection to militia groups’
formation and activities in and around refugee camps, and therefore the
militarization of refugee women’s lives.
In addition, the actions of the superpowers during the Cold War and how
they created military relationships with African states to increase their economic,
political and militaristic influence are worth considering especially when one
analyzes the consequences today.103 Volman states that:
And while it has become more difficult for Africans, particularly African governments, to obtain expensive heavy weaponry, it has actually become easier for government and other armed groups to get light weapons (such as handguns, rifles, machine guns, landmines and light artillery).104
African governments have also participated in the militarization of various militia
groups. For example Congo (Kinshasa) is known to have provided arms to Hutu-
extremists in refugee camps in eastern Congo.105 The militarization and effects of
militarization on refugee women in Tanzania can be analyzed through the lenses
of culture, tradition and history. But it is in a historical analysis that one finds the
tangible ways in which militarization has been executed. The African continent
has been part of a geopolitical “game” where superpowers and colonial powers
continue to provide weapons. This leads to the understanding of the tangible
103 Volman, 151. 104 Volman, 155. 105 Volman, 157.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79
history of where militarization has some of its roots, especially when one analyzes
the prevalence of arms around refugee camps. The prevalence of arms has been a
leading cause of violence in and around refugee camps, especially in Kanembwa
camp.
The proliferation of arms in the Great Lakes Region of Africa has made
the region susceptible to armed conflict. What one notices is that as a result of
armed conflict, the nature of refugee women’s situations and experiences in
Africa is very complex because the remnants of colonialism have created regional
security complexes.106 Therefore, refugee women’s situations are connected to
different security complexes inherent in the region. Some of the reasons why
refugee women and other refugees in general face security problems can be
addressed through an understanding of how refugees get to their situations in the
first place. Patrick Matlou states that there are four reasons for migration in
Africa: (1) decolonization (2) South Africa’s project to destabilize southern Africa
(3) post independence conflict involving political, ethnicity and religious
struggles (4) economic and environmental disasters.107 African countries have
faced myriad conflicts which have externally or internally displaced their citizens.
Burundians are part of the equation of Africa’s perpetual displacement
experience.
106 Barry Buzan studied regional security complexes which refer to the interconnected security concerns of states within a region. See Miall et al. Contemporary Conflict Resolution. (Cambridge: Polity Press, 2005), 83. 07 Patrick Matlou, « Upsetting the Chart: Forced Migration and Gender Issues, the Experience of Africa » in Engendering Forced Migration (New York: Berghahn Books, 1999), 129.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80
Tanzania has the largest refugee population in Africa.108 Edward Mogire
notes that Tanzania has had a long history of accepting refugees, especially of
southern Africans who were involved in armed resistance against colonial and
racist regimes.109 However “the most consistent refugee flows into Tanzania have
been from neighboring Burundi.”110Burundian refugees fled Burundi due to
militaristic activities and fear of persecution, however activities in the refugee
camp by certain refugees has caused concern for Burundi and the region. In this
regard, Tanzanian refugee camps with Burundian refugees are militarized because
they were created in response to militaristic activities; in addition, these camps
maintain their militaristic symbolism because of the political and militaristic
activities carried out by some refugees. What one notices is that the militarization
of refugee camps in Tanzania has been partially made possible due to the
proliferation of arms in the region that are used by rebel groups.
Furthermore, politicization in the refugee camps is linked to militarism.
Many scholars have raised the view that African refugee situations pose many
security problems to the state and other refugees and civilian populations (Gil
Loescher 1989, Stephan John Stedman 2003, Mogire, and Aristide R. Zolberg et
al. 1989). However, a focus on the state as a referent object of security succumbs
to traditional realist and state-centric views. Authors like Loescher, Mogire, and
Aristide et al. analyze broadly the security issues of all refugees. Loescher states
108 Robert Mogire, « Preventing or Abetting : Refugee Militarization in Tanzania » inNo Refuge (London: Zed Books, 2006), 137. 109 Mogire, 139. 110 Mogire, 141.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81
that refugee issues are connected to their particular nation’s security problem.111
This is inevitably the truth with Burundian refugees. The main concern attached
to this concept is that nations might not be willing to address refugee issues
because “refugee camps are often times located near disputed borders and also
reside among combatants that are planning against the nation they fled.”112 The
double-identity of refugees as referent objects of security, and sources of security
problems adds to the complexity of the refugee situation. Furthermore, the
ambiguity of refugee security is at the core of the concept of militarization of
refugee women and of the experiences refugee women face as militarized
individuals. Women find themselves in a situation where on one hand they fled
militarism, however while in refuge, they take up the cause for militarism, or they
are inadvertently attached to militarism.
The argument that refugees not only face security dilemmas but that they
also pose these security threats creates uncertainty in the discourse concerning
refugee protection. This is especially the case when the issue of warrior-refugees
a term introduced by Aristide R. Zolberg et al. (1989) is raised. Refugees that take
up arms are often seen as resources of war and as being manipulated by various
political and military factions.113 Zolberg et al. explain the prevalence of warrior
refugees in camps by writing that:
Old people, women and children may be physically separated in the camps by warriors but are still connected in that refugees
111 Gil Loescher,Refugees and International Relations edited by Gil Loescher and Laila Monahan (New York: Oxford University Press, 1989), 3. 12Loescher,Refugees and International Relations, 3. 113 Stephan John Stedman,Refugee Manipulation: War, Politics and the Abuse of Human Suffering (Washington DC: Brookings Institution Press, 2003), 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82
constitute a legitimacy population for the warriors. They become physical testimony of support for the warriors in the sense that they represent a rejection of the other side in the conflict.114
The symbolism of refugees as individuals constituting legitimacy for warrior-
refugees is another abstract explanation of refugee militarization. The mere
perception that refugees are tied to warriors makes their very existence
militarized. The refugee population in political refugee camps legitimizes warriors
that use militant tactics merely by being in the camps. It can be said that refugee
populations legitimizes the militant tactics of warriors because warriors can make
a claim that militarism in the country of origin causes fear and creates refugee
camps and therefore those in power in the country of origin do not have
legitimacy. The illegitimacy of those in power is demonstrated by the number of
individuals who flee into refugee camps. The paradox however is that refugees
have faced monumental physical threats from warrior-refugees, and yet these
warriors manage to sometimes entice refugees into joining and being part of their
militaristic legions.
Therefore despite the security implications linked to warrior-refugees, a
consensual interaction with other civilian refugees sometimes exists. I contend
that the interaction between warrior-refugees and refugee women highly
reinforces the culture of refugee women’s militarization. Therefore an element
that reinforces the militarization of refugee women is partly due to women’s close
proximity and interaction with highly militarized and masculinized factions. This
of course is a simplistic explanation because militarization does not occur by
114 Zolberg et al., 277.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83
rubbing-off from one individual to the other; it is a process. There is a complex
history and culture around this phenomenon that allows it to exist. For the
Burundian case, there exists a fine line-if one at all-between politics, ethnicity and
militarization. Ethnicity in Burundi is militarized, and so are politics. The
interactions between refugee women and rebel militias and their militaristic ideals
are not merely militaristic situations involving military activity, but they are also
political situations which hinge on questions of ethnicity. The nature of a refugee
camp as an insecure, unstable, and uncertain environment allows for this
interaction. It is in this situation that refugee women are prone to becoming
militarized by the activities of masculinized and militarized forces. This
militarization, does not simply invent itself. The Burundian situation has shown
that history and culture influence this situation.
Conclusion
The literature presented above has touched upon numerous themes that
explain the development and nature of refugee women’s militarization.
Militarization is a phenomenon that is traditional not associated with women, and
its connection with masculinity points to this. However what we notice is an
increase in the number of cases where women take up arms and therefore refute
and challenge conventional gender roles and patriarchal systems. Women choose
to participate in militia groups for various reasons; however it must be noted that
in many situations their participation is not voluntary. In some cases women see
military groups as entities that can provide them with a sense of emancipation and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 84
liberation from the chains of subordination and patriarchy. On the other hand,
military groups are also known to sometimes abuse women. The contradiction in
the ambiguous relationship between militia groups and women leads one to
question the elements that create and fuel such a capricious relationship. By
looking at the events in history we find the answers of where militaristic ventures,
activities, and elements originate. In history we find elements that point to
politics, regional security complexes, and ethnic animosity. By analyzing culture
and its metamorphosis, we notice that there are certain elements of culture,
especially patriarchal elements and gender roles that change in time that explain
the militarization of refugee women and how the military is present in their every
day lives. The next chapter will analyze the history of the Burundian conflict, and
will bring together issues such as ethnicity, patriarchy, gender, politics and
migration in the Burundian history in order to establish where militarization has
its roots, and how it seeps into society and affects the culture of the Burundian
people.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER FOUR
HISTORICAL ANALYSIS: BRIDGING POLITICS, ETHNIC
ANIMOSITY, GENDER AND MILITARISM
This historical analysis brings together some elements of patriarchal
culture and traditions of the Burundian society. Although this study is about
Burundian refugee women in Kanembwa camp in the Kigoma region of Tanzania,
I present a history not only of Burundi but of Rwanda as well. The conflict
histories of Burundi and Rwanda can almost be described as mirror-like based
upon the many similar attributes of the conflicts. The two countries shared a joint
history up until post-independence where the activities in one country ignited
violent activity in the other. Taking into consideration the connection between
Rwanda and Burundi, Jean-Pierre Chretien (2003) poses the question of whether
“Burundi’s fate would have been radically different if the country had not been
falsely twinned to Rwanda under colonialism.”1 The purpose of my historical
analysis of the Burundian and Rwandan conflict is to place women within this
conflict. I contend that in order to understand the culture that militarized refugee
women create as a result of this conflict, it is important first of all to analyze
women’s position in their society. The ethnic dimensions of the conflict make this
analysis particularly interesting because of the various myths associated with
1 Jean-Pierre Chretian, The Great Lakes of Africa: Two Thousand Years of History (New York: Zone Books, 2003), 310. 85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86
Tutsi and Hutu women. I argue that myths play an important role in how societies
identify themselves.
I am particularly interested in the ways in which women have been viewed
in this society based upon the dichotomous attributes given to the two main
contesting ethnic groups “Hutu and Tutsi.” In order to present this historical
analysis I rely heavily on Liisa Malkkis’ narratives that she collected in Mishamo
refugee camp in western Tanzania. In addition I rely on the interviews that I
conducted in Tanzania. The interviews that are cited here, like in the rest of this
study, have been altered in order to ensure that the identities of the informants are
not traceable. I also rely on the scholarly writings of Mahmood Mamdani and
Rene Lemarchand and others on the issues of race and ethnicity in the conflict.
This analysis starts with an overview of the origins of the Hutu and Tutsi
distinction; this is followed by the activities during the colonial and post-colonial
period. Finally I present an overview of the history of the Burundian refugee
movement to Tanzania with an emphasis on the role of rebel groups. Throughout
each section I attempt to present where women are and how they are involved.
As I have indicated earlier, my analysis is of Burundi and Rwanda because
I am of the view that addressing one requires addressing the other. Therefore,
although I will delineate the different events in each country, I will also fuse the
events together to produce a picture of the two together.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87
Origins of Hutu and Tutsi
The main question that motivates my analysis concerns the origins of the
Hutu and Tutsi ethnic animosity and how women’s roles and place in society has
been affected by ethnic animosity. Were the divides between Hutu and Tutsi in
existence in pre-colonial periods? This question has highly contradictory answers.
Based upon the Arusha Peace Accord for Burundi signed in 2000,2 Burundi did
not know of any ethnic differences before colonial times. Article 1(1) of the
accord states that:
During the pre-colonial period, all the ethnic groups inhabiting Burundi owed allegiance to the same monarch, Umwami, believed in the same god, Imana, had the same culture and the same language, Kirundi, and lived together in the same territory. Notwithstanding the migratory movements that accompanied the settlement of the various groups in Burundi, everyone recognized themselves as Barundi.3
Authors such as Michael Banton (2000) indicate that large scale conflicts are
usually between two groups that have one or more distinct differences which help
them act collectively.4 The importance of group and collective identity is
irrefutable. Scholars such as Mamdani (2001) present an analysis of the roots of
the Rwandan and Burundian conflict. Mamdani states that the conflict can be
understood in terms of the relationship between political power and political
violence even if it is translated to Hutu and Tutsi ethnic differences.5 Therefore in
this regard we can understand that Hutu and Tutsi are merely convenient
2 Peace accord signed on August 28 2000 Full text available from http://www.usip.org/library/pa/burundi/pa_burundi_08282000_toc_html. 3 (USIP). 4 Michael Banton, “Ethnic Conflict”Sociology 34 no. 3( 2000), 485. 5 Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton: Princeton University Press, 2001), 42.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88
delineations for those who wish to gain political and hegemonic power. If the
origins are power related and ethnicity is manipulated to justify a power struggle,
how are women involved and how are they part of this equation? Mamdani and
others do not fully explore the gender dimensions of the creation of the ethnic
categories of Hutu and Tutsi.
To begin with, in order to present the origins of the differences between
Hutu and Tutsis, it is important to note the hypothesis of the elements that make
Hutus different from Tutsis. Mamdani indicates that there are three types of
hypothesis that attempt to explain the Hutu and Tutsi differences. One is
phenotype which is essentially the physical differences, the second is genotype
which in essence is based upon blood types, and the last one is memory of the
people as ways of explaining their history.6 What is meant by physical differences
is that Tutsi were seen as slender and tall while the Hutu were squat with medium
height.7 As indicated by one of Malkki’s panels “The Tutsi are taller
generally ...at the same time, thinner ...they are of a beautiful stature.”8 As regards
genotype, studies have gone as far as comparing how lactose intolerant one group
is compared to the other. Studies have shown that three out of four Tutsis can
digest lactose, while one out of three Hutus can digest lactose.9 As one informant
stated to me the reason for the conflict between the Hutus and the Tutsis seems
like a big secret that only they know. I tend to agree with such a position because
the differences between Hutus and Tutsis seem to be rooted in internalized
6 Mamdani, 43. 7 Mamdani, 44. 8 Malkki,79. 9 Mamdani, 46.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89
complexes that are fueled to some degree by a collective imagination.10 This is
why we can not rely on the simplistic physical and genotype delineations of Hutus
and Tutsi. Relying on the hypothesis of physical ethnic differences has proved to
be very unreliable especially when one attempts to understand why some Hutus
have lost their lives for looking Tutsi. It is irrefutable that the African continent is
a mosaic with different groups of people. However superficial qualifications of
each group are what lead to animosity. The qualifications have been made by
outsiders and also by Burundians and Rwandans themselves. What is noticed is
that it is partly a question of imagination that is fueled by internalized inferiority
complexes and a false sense of superiority, which leads to animosity as opposed
to “real” ethnic differences.
Furthermore, Mamdani brings up the point that the Hutu and Tutsi
“difference” were racialized. The Hutu has been considered as a Negro and a
Tutsi as having some sort of link with Caucasians based upon physical
morphology and other attributes. As pointed out by Mamdani the racialized nature
of the Tutsi and Hutu differences are products of the Hamitic myth from the
Christian and Jewish traditions.11 The myth that was inevitably introduced
through the racism that fueled slavery and colonial legacies is that when Noah
was drunk after the flood and lay naked and asleep his son Ham laughed at him
instead of covering his naked body. His other sons Shem and Japheth covered him
and therefore Noah after decided to curse Ham and his descendents for laughing
10 Jacque Semelin provides an interesting analysis on collective imagination and how it is used against an enemy and can give groups reasons for acting out their collective and imagined fear and hatred. 11 Mamdani, 80.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90
• i 1) at his naked body. Therefore the animosity between Hutus and Tutsi has its
roots in Christian and Jewish traditions, traditions that are not inherent to the Hutu
and Tutsi people. This is a demonstration of the how much of the animosity is
based upon imagination. Colonial history holds the key to the ethnic animosity
that has led to massacres.
The Peace Accord points to colonial interference in Burundi as being the
reason for the delineation of ethnicity and ethnic differences between Hutu and
Tutsi. What is of interest is how these interferences have translated into power
politics as Mamdani points out, and also how they have seeped into the identity of
the two groups. This can be seen in the way that Hutus and Tutsi view women.
Women have been viewed in a completely ethnicized manner in the conflict
which, I argue, led to the position that women had in the conflict as the ultimate
rape victims, as well as the perpetrators of conflict.
In this study I am interested in understanding how culture and peoples’
perceptions of their reality, although it appears that there is a thin line between
reality and imagination in this conflict, have influenced women’s roles, and
position in the conflict. I contend that the myths that various societies adhere to
shape the way various groups view themselves. If a woman in a given society
belongs to a group that is degraded, this surely has an effect on her self-value and
the identity that she creates for herself. In this analysis it will be important to trace
the myths and the culture that have been passed on from the past and how they
affect women during conflict.
12 Mamdani, 80.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91
The Burundians perceptions of their ethnic differences are pronounced
when the relationship between Hutu men and Tutsi women is analyzed. The Tutsi
woman is valued by the Hutu man because being with her means that he moves
up in society. This has significant implications for Hutu women, who do not seem
to have this “special” attribute. Lemarchand brings up one of Malkki’s narratives
that states:
The Tutsi came to find pasture for their herds. They asked the Hutu, who were the natives, if they could have a place for their herds and for themselves... Then the Tutsi began to give dung to the Hutu. After some time the Tutsi gave milk to the Hutu children, and then the mama found that the children became more vigorous... They started to go to the Tutsi to obtain milk...Once the Hutu receives a cow from the Tutsi, all his descendents have to work for the Tutsi... In order to get food, the Tutsi flatters a Hutu, a cultivator. He says: “I give you my daughter, even two or three cows.” Like this. Then the Hutu accepts, seeing a beautiful woman with a long nose and very tall also in stature, elegant if you wish, and who squanders smiles... This the means of flattering the Hutu... Then the Hutu has to augment his cultivating since he gets married to the beautiful Tutsi woman... Then this Hutu begins to be despised by the other Hutu because he is flattered and he boasts about his Tutsi wife. And his parents are proud. It was like this in the past.13
Another element of the dichotomous relationship between the Hutu and Tutsi is
concerning socioeconomic status. As stated by Malkki, “the Tutsi, it was said,
suffered from ‘poverty’ brought on by ‘innate laziness’ and seized any
opportunity to ‘trick’ the Hutu man into a permanent arrangement of economic
support. The beautiful Tutsi women provided the means to accomplish this.”14 In
one of her panels of narratives Malkki quotes a refugee who states that since
Hutus worked harder and since Tutsis are lazy, a Tutsi woman, who is most
13 Rene Lemarchand, Burundi: Ethnic Conflict and Genocide (Cambridge: Cambridge University Press, 1994), 21. 14 Malkki, 83.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. beautiful, will prefer to marry a Hutu, and a Hutu prefers to marry a Tutsi woman
because of her beauty.15 As regards the intermarriage between Hutu and Tutsi, the
panel states that there were more marriages between Hutu men and Tutsi women
than Tutsi men and Hutu women. The reasons were based upon the laziness of the
Tutsi which made him unattractive to a Hutu woman but also it appears as though
there was a law among the Tutsi people which forbade them to marry Hutu girls.16
This raises questions of ethnic purity and of the symbolism of women as the
bearer of children and therefore of a people. I propose that by forbidding the
marriage between Hutu women and Tutsi men, this could be translated as a
mechanism among the Tutsis to maintain the purity of the Tutsi population and to
ensure that it was not born of a Hutu woman. We can see this especially when one
analyzes the ethnicity of a child bom to a Hutu-Tutsi couple. Erin Baines (2003)
and Lemarchand explain that in pre-colonial time there was ethnic mobility where
Hutu men could become Tutsi by marrying Tutsi women. This social act is called
Kwihtura}1 Therefore children bom of a Hutu and Tutsi couple would be
considered Tutsi.18 As Lemarchand points out, in theory a child bom of a Hutu-
Tutsi couple is supposed to assume the father’s identity.19 However this was
clearly not the case as a result of Kwihtura. I propose that the paucity of marriages
between Hutu women and Tutsi men indicates that a Tutsi child bom of Hutu
women was highly undesirable. In addition, since interethnic marriage generally
15 Malkki, 82. 16 Malkki, 82. 17 Erin Baines, “Body Politics and the Rwandan Crisis”Third World Quarterly 24 no.3 (2003), 483. 18 Baines, 483. 19 Lemarchand, 9.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93
presupposes social mobility, this raises the question of what motivation a Tutsi
man had for marrying woman who was of a “lower” ethnicity.
Furthermore another panel by Malkki indicates that marriage between an
intellectual Hutu man and Tutsi peasant woman was accepted but a peasant Hutu
man could never marry a Tutsi woman.20 This particular panel emphasizes that
the beauty of a Tutsi woman is a death trap. In one of the camps in Kibondo, one
male refugee expressed to me another myth about why Tutsi women were
considered more beautiful than Hutu women. The informant stated that “the myth
about why the Tutsi women are more beautiful than Hutu women is because Tutsi
women are said to be lazy and so they are always clean and beautiful. Hutu
women are hard working and so they look dirty and not very beautiful.”21
Therefore it is of particular interest to note that the beauty of a Tutsi women
compared to a Hutu woman is linked not only to physical qualities and
socioeconomic elements but also of work ethics and culture. This informant paints
a picture of Hutu women as traditional and hardworking, as is expected in a
patriarchal society. However it is the non-traditional Tutsi woman who is the
desire of all men. In this regard the Hutu woman symbolizes traditional
patriarchal society through her hard work, while a Tutsi woman manages to
transcend some elements of patriarchal constraints and is even more desirable
than a Hutu woman simply because of her ethnicity. Liisa Malkki’s panel raises
a similar point:
20 Malkki, 85. 21 Refugee informant, Interviewed by author, January 15, 2007 Kibondo, Tanzania
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94
It seems to me that it is forbidden in their [Tutsi] law to marry with the Hutu girls but if [the Tutsi man] is very poor, he will do it to have a cultivator, a slave. But this does not usually happen. To see a Tutsi who has married a Hutu [woman], this is hardly 2 percent.22
Therefore a Hutu woman, whose lifestyle symbolizes the pervasive nature of the
patriarchal system of Burundi, is of little value. She is not only ingrained in the
patriarchal system, but also being of an ethnicity that is considered inferior,
entrenches her further into the patriarchal system.
The importance of bringing up the issues of the physical characteristics of
Hutu and Tutsi or as Lemarchand states the “nonsense that has been written about
the ‘short and stocky Hutu’ and the ‘tall, lash-thin and graceful’ Tutsi is to present
how perceived identities of the other, even though often inaccurate, play a role in
fueling historical tragedies. Hutus and Tutsi are people sharing similar histories,
language and culture therefore they are a similar people. However the ethnicized
and also racialized identities of these groups have not only remained an abstract
notion for academics to write about, but they are a complex set of issues that are
perhaps only understood by the Hutus and Tutsis. The ethnic animosity is so
ingrained in the society that parents were among those responsible for teaching
their children these ethnic issues. Therefore, the imagination of individuals
concerning the ethnic differences and issues of beauty, superiority, inferiority and
ugliness has been passed on from generation to generation. What we have is not
only a collective memory, but a collective imagination. As stated by one of my
female refugee informants:
22 Malkki, 82.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95
I am a Hutu. My father was killed by Tutsi in 1980. He was not in any political group. The man that killed my father is also dead. Life for us in Burundi was not very normal. People threatened to hurt me all the time because there were more Tutsi than Hutu in my community. It was rare to see Hutu women married to Tutsi men, and there were more Hutu men that married Tutsi women. My parents would tell me that there was a difference between Hutus and Tutsis. They told me that Tutsis had longer noses and that Hutus had flat noses. They told me that Tutsis are tall. They explained to me the ethnic differences. I learned that the Tutsi women were more beautiful. Many people believe this. Hutu women believed it but there are some Hutu women who are more beautiful. I believe it was the Tutsi men who said these things and tried to make everyone believe them.23
As written by Mamdani, Tutsi women that were married to Hutu men were killed
in the 1994 genocide in Rwanda and in some instances their Hutu husbands were
forced to kill them.24 At this point, the delineation of ethnicity took a form of
gender discrimination regardless of one’s affiliation with Hutus. It is by analyzing
such situations that one is quick to understand to what extent ethnic differences
play a role, even to cross the bonds of marriage, where a husband is made to kill
his own wife. The demonization of the other in this context does not only stem
from physical and morphological differences but it is also a matter of how these
differences regardless of how superficial they might be, shape a person’s
perception of ethnic solidarity and identity. A female refugee informant explained
to me her story of the difficulties she faced for being in an inter-ethnic marriage:
In 1993 my husband was killed by the military along with my two children. After the death of my husband and children, a year later a group of people whose faces I could not see came to beat me. I lost consciousness and awoke in the hospital. After one week I was in the
23 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 24 Mamdani, 4.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96
hospital and a group came to kill those of us in the hospital. I am a Tutsi and my husband was a Hutu.25
Despite the instances of passage from one ethnic group (mainly Hutu
becoming Tutsis) to another through Kihutura, countless people killed people of
their own ethnic group in belief that they were killing the other group.26 This adds
to the ambiguity of physical characteristics, but still does not explain why
individuals decide to internalize these superficial identities. As Malkki states,
“even while the Hutu accepted the description of the Tutsi as ‘beautiful,’ they
were busy revaluing beauty itself, casting it as a sign, not of nobility or virtue, but
of evil and danger.”27 It can be concluded that in this regard by accepting beauty
as a positive element, would mean that the Hutu were accepting the notion that
they were inferior. By labeling beauty as a negative element, they neutralized the
dominance of Hutus by Tutsi. One of my Hutu female refugee informants
explained to me the issues behind the Hutu and Tutsi differentiation as one
induced by an inferiority complex:
I noticed that more Hutu males were with Tutsi women. It was rare for Tutsi men to be with Hutu women. Tutsi are considered more beautiful and Hutus like women that are more beautiful. In Burundian culture a richer woman is better. It is very difficult for a poor Hutu man to be with a rich Tutsi woman. Hutus considered Tutsis to be better because they have a serious inferiority complex.28
25 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 22, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 26 Lemarchand, 7-8. 27 Malkki, 103. 28 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania This particular refugee woman is very educated and analytical. She used the term “inferiority complex.”
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97
From Colonial to Post Colonial Burundi and Rwanda
When searching for where the Hutu inferiority complex comes from and
the collective imagination of Hutu inferiority and Tutsi superiority, one has only
to look at the colonial history of Rwanda and Burundi. This section presents a
narrative exploring colonial to post colonial history of Burundi and Rwanda. To
begin it must be noted that missionaries were among the first ethnographers of
Ruanda-Urundi.29 In addition the colonial state relied on missionary knowledge.30
Some of the early writings of missionaries were that the Tutsi were like “a
European under a black skin.” This writing thereby gave racial qualities to the
Hutus and Tutsis.31 The Belgians used the work of the missionaries and what is
noticed is that the “Belgian power turned Hamitic racial supremacy from an
ideology into an institutional fact by making it the basis of the changes in
political, social, and cultural relations.”32 Therefore in Ruanda-Urundi some of
the ways that were used to keep the Tutsi as the superior race were through the
school system and the local administration. The point was to turn the Tutsi into
the rulers of Rwanda since they were the “bom rulers” in the eyes of the Church.33
Furthermore the 1933-34 census in Ruanda-Urundi, according to
Mamdani racialized the Hutu and Tutsi distinction. “The Information Bulletin on
Ruanda-Urundi issued by the Public Relations Office in Belgium in 1960, thus
spoke of “the inhabitants” of Rwanda as belonging “to two main racial groups:
29 Mamdani, 87. 30 Mamdani, 88. 31 Mamdani, 88. 32 Mamdani, 88. 33 Mamdani, 89.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98
the Tutsi feudal stock-breeders, comprising 14% of the population, and the Hutu
farmers amounting to 85%.”34 Leading up to independence in 1962, Hutus were
creating a Hutu consciousness; the Tutsi consciousness on the other hand had
been created long before the Hutu consciousness.35 Hutu consciousness grew with
various parties and elites that threatened the privilege the Tutsi had gained
through colonialism.36 The growth of political parties with Hutu agendas such as
PARMEHUTU in Rwanda contributed to the Hutu consciousness. As explained
by Baines the Hutu consciousness or Hutu nation was also produced in social
spaces that included private spheres as well, especially the female body.37 Hutu
women were refused to marry Tutsi men by their fathers to protect the purity of
the Hutu ethnicity.38 While for Burundian refugees interviewed by Malkki, Tutsi
men did not want to marry Hutu women, in Rwanda the Hutu consciousness
altered this rejection through patriarchy, where men controlled Hutu women’s
choice of marrying a Tutsi by ensuring that a Tutsi suitor was not accepted.
The creation of a Hutu consciousness provoked events that led to Hutus
taking over in Rwanda. After the Mwami coup of 1959 in Rwanda,
PARMEHUTU had a violent encounter with armed forces, and some 200 people
OQ were killed. The violence increased and targeted the Tutsi chiefs. The Belgians
quickly called for a state of emergency and put Colonel B.E.M. Guy Logiest in
34 Mamdani, 99. 35 Mamdani, 117. 36 Mamdani, 118. 37 Baines, 482. 38 Baines, 483. 39 Mamdani, 123.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99
charge. Logiest started replacing Tutsi chiefs with Hutus.40 Hutu chiefs and local
authorities met in Gitarama on January 28,1961, and abolished the monarchy and
created the first republic which led to Hutus taking political power in Rwanda.41
These events forced many Tutsis to flee Rwanda.42 The creation of a Hutu
consciousness in Rwanda can be analyzed as the beginning of the actualization of
a collective imagination. In addition the colonial powers fueled this collective
imagination by replacing Tutsis with Hutus. However as stated by Baines this
consciousness was not only manifest in the political arena; however it was also
manifest in the social and private spheres, therefore making it a truly deep-seated
and collective consciousness.
Events in Burundi were similar to those in Rwanda. As noted by
Lemarchand, the colonial powers played a role in the disunity of the Burundians
through their divide and rule policies 43 On November 10,1959 the Belgian
government committed itself to a two-fold-program of political reform.44 Political
parties were established notablyParti de I’Union et du Progres National
(Uprona) and the Parti Democratie Chretien (PDC).45 Unlike Rwanda, in
Burundi, the kingship Mwami was not associated with Tutsi supremacy but was
associated with Hutu identification.46 The kingship was used by princes who used
their royalty to gain power in the political parties. This is how Prince Rwagasore
40 Mamdani. 124. 41 Mamdani. 123-124. 42 Mamdani, 124. 43 Lemarchand, 24. 44 Lemarchand, 51. 45 Lemarchand, 50-52. 46 Lemarchand, 39.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100
gained power.47 However, on October 13,1961 Prince Rwagasore was shot by a
Greek gunman in a PDC plot, and as stated by Lemarchand, many believe that if
he had not been killed, “he might have spared his nation the traumas that would
soon tear it apart.”48 Rwagosore was the leader of the Uprona and the eldest son
of the Mwami. His image appealed to the Hutu masses and “the presumption that
he took a Hutu girl for a wife was enough to dissipate all suspicions of ethnic
prejudice.”49 Therefore in Burundi, much like Rwanda, the symbolic attributes of
Tutsi and Hutu women was an issue that brought the social and political realms
together. Analytically, one can surmise that a Tutsi wife symbolized ethnic
animosity because by taking a Tutsi wife a man, especially a Hutu man, is
acknowledging the ethnic superiority of the Tutsis.
The death of Prince Rwagosore was the first in the series of events that
heightened the polarization of ethnic feelings. After his death, Hutus and Tutsi
alike started to see the Mwami and kingship as a problem. What also exacerbated
the situation in Burundi was the revolution in Rwanda which led to many
Rwandan Tutsi refugees in Burundi. The Tutsi refugees added to the feelings of
ethnic animosity in Burundi as they brought with them their baggage of trauma
and horror.50
By early 1965, approximately 50,000 Rwandan Tutsi had found refuge in
Burundi.51 They planned to return to Rwanda and gain control. “Only through a
47 Lemarchand, 51. 48 Lemarchand, 54. 49 Lemarchand, 53. 50 Lemarchand, 60-61. 51 Lemarchand, 60.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101
Tutsi-dominated government in Bujumbura could Tutsi rule be restored in
Kigali.”52 The Rwandan refugees became involved in Burundi’s internal affairs.
On January 18,1965 a Rwandan Tutsi refugee employed at the US embassy in
Burundi shot to death the first appointed Prime Minister Pierre Ngendandumwe, a
Hutu.53 In Burundi interethnic tensions increased from 1968 to 1969 and a plan
was disclosed of a Hutu instigated coup.54 The disclosure of the coup led to the
arrests of thirty Hutu officials in the army and the government; twenty of the
arrested were sentenced to death.55 On April 29,1972 Hutu armed insurgents
attacked government buildings and every Tutsi in sight and Hutus that refused to
join in the areas of Nyanza Lac and Rumonge.56 These killings led to the deaths of
over 3000 individuals most of them Tutsi.57 As Lemarchand points out, the
insurgency was planned and those in Nyanza Lac used Tanzania as a base to
rearm.58 In response the government ordered the killings of all Hutus on the radio;
it was stated that all should “hunt down the python in the grass.”59
The Hutu massacres led to the deaths of educated Hutus that posed a threat
to Tutsi survival.60 Hundreds of thousands of individuals were killed.61 This wave
of violence led to the exodus of Hutu refugees from Burundi into neighboring
countries. Refugees in Tanzania launched attacks into Burundi and the
52 Lemarchand, 61. 53 Lemarchand, 69. 54 Lemarchand, 85. 55 Lemarchand, 86. 56 Lemarchand, 91. 57 Lemarchand, 92. 58 Lemarchand, 94. 59 Greendland cited in Lemarchand, 98. 60 Lemarchand, 101. 61 Lemarchand, 100.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102
government responded by attacking Tanzanian localities.62 It was in the camp of
Mishamo in Tanzania that the Burundian rebel group PALIPEHUTU was bom in
April 1980.63 This wave of events led to the militarization of refugee camps where
political and military activities were taking place. In addition, it can be said that
the events merged politics with militarism. Bumndi experienced more surges of
anger in 1980 from both Tutsi and Hutu extremists.64 The Tutsi viewed the
government to be moving to further liberalization and acceding power to Hutus,
and extreme Hutus were of the view that this was a moment to challenge even
further the Tutsi dominance.65 According to Lemarchand, the 1988 massacres
were not planned like the 1972 massacres.66 International pressure was more
present in the 1988 massacre which led to President Buyoya of Bumndi deciding
to increase the number of Hutus in his cabinet.67 What we notice therefore is that
politics in Burundi were ethnicized. They can be delineated as Hutus against
Tutsi. Furthermore, they were also militarized. Militarization of political and
ethnic rivalries is manifest not only through the means of using military means to
gain power; however after the creation of PALIPEHUTU it can be said that the
type of militarization of the opponents became a primary identity of the political
and ethnic identities. Tutsi political and ethnic identity was characterized through
the use of the army, while Hutus political and ethnic identity was characterized
through the use of unconventional rebel militia tactics.
62 Lemarchand, 104. 63 Lemarchand, 105. 64 Lemarchand, 128. 65 Lemarchand, 127. 66 Lemarchand, 128. 67 Lemarchand, 130.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103
The Burundi conflict has been complicated by the various rebel groups.
These groups include Parti pour la Liberation du Peuple Hutu (PALIPEHUTU)
which has a military wing called Forces Nationales de Liberation
(PALIPEHUTU-FNL). Linked to PALIPEHUTU are Front de Liberation
National (FROLINAT) and its armed wing Forces Armies du Peuple (FAP).
Another group is Union pour la Liberation Nationale (ULINA) and its armed
wing Forces de Liberation National (FALINA). The largest and best organized
group is the National Council for the Defense of Democracy (CNDD) and its
armed wing the Forces for the Defense of Democracy (FDD).68 CNDD is a
largely Hutu group but includes some Tutsi members as well. Each political party
is aligned with a military wing; this reflects the level of militarization in the
political and ethnic conflict of Burundi. Militarism can not be seen apart from
politics and ethnicity.
Events in Rwanda continued to be fueled by and fueled the Burundian
conflict. On October 1,1990 the Rwandan Patriotic Front (RPF) launched an
attack in Rwanda from Uganda where they were in exile.69 They suffered a lot of
casualties which influenced Paul Kagame’s decision to leave his military training
in the US in order to lead the RPF into Rwanda.70 Under Kagame’s leadership the
RPF made advances to Rwanda.71 The RPF advances and invasion sparked Hutu
Power ideology.72 This led to the massacres of many Tutsi in 1990 in Kibilira, in
68 Doxtador and Mosomothane, 69. 69 Mamdani, 186. 70 Mamdani, 186. 71 Mamdani, 187. 72 Mamdani, 190.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104
1991 in Bugogwe, the third was in 1993 in Bugesera.73 The civilians were urged
to protect themselves and to organize against the invading RPF and an estimated
3000 people were killed in massacres between 1990 and 1993.74 These events are
a clear demonstration of the mirror-like aspects of the Rwandan and Burundian
conflict as well as the militarization of politics and ethnicity. While the Tutsis
were the ones in power in Burundi and the army was controlled by the Tutsis, this
was the opposite in Rwanda. The opposition in Burundi was a group of Hutu rebel
militia who were living in refuge, while for Rwanda the rebel militia was Tutsi.
This complexity of militarism in the two countries is mirror-like however, its very
nature explains the graveness of the problem, and why the massacres in each
country reached such unprecedented proportions.
In Burundi, on June 1,1993, Front pour la Democratic Burundi
(FRODEBU) party leader Melchior Ndadaye became Burundi’s first Hutu
'JC president. He won the first free elections in the history of Burundi. Ndadaye was
assassinated in October 1993 and this sparked massacres in Burundi. This led
Burundian refugees to flee to Rwanda and where they took part in the Rwandan
genocide and were responsible for some of the most gmesome attacks.76 Many
Burundians fled to Tanzania after Ndadaye’s death and therefore leading to the
creation of some of the refugee camps in western Tanzania such as Kanembwa.
As for Rwanda, by 1992 the Arusha talks were taking place.77 The Arusha
73 Mamdani, 152. 74 Mamdani, 192. 75 Lemarchand, 178. 76 Mamdani, 205. 77 Mamdani, 210.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 105
no agreement was signed on August 3,1993. From 1991-1992 the youth militia
group Interahamwe grew into a death squad that was targeting Tutsis from house
to house.79 What sparked the conflict in Rwanda was the assassinations of
Ndadaye in Burundi and the assassination of the Rwandan President and Prime
Minister Juvenal Habyarimana and Agathe Uwilingiyimana.80 The Burundians
were angry for being forced to leave their home and therefore they become easy
recruits for the genocide in Rwanda.81 The fueling of events in one country to
another is indicative of the regional security complex that the ethnic tensions have
created.
Throughout their interaction, it is easy to see how the Hutu and Tutsi in
Rwanda and Burundi have been very distrustful of each other. The coming
together of militarism, politics and ethnicity not only has social dimensions;
however when violence is used women are targeted in very specific ways in these
conflicts. In the middle of all the massacres and genocide, women have been
targeted. Rape was used as one of the main weapons in the two conflicts. A
female refuge informant explained to me the horror she encountered during the
war:
In 1993 when the war started I was not with my parents. I left for Rwanda after hiding in various places in the hills of Burundi. In 19941 left to Rwanda, and found refuge in a convent. I was looked for. I was only 12 years old. I was looked for by armed men. The sisters at the convent did all they could to get me to return to Burundi. I was raped by someone I did not know while fleeing Rwanda. I am Tutsi. I gave birth to my child and returned to school in Burundi One day a group of armed men came to
78 Mamdani, 212. 79 Mamdani, 212. 80 Mamdani, 215. 81 Mamdani, 219.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106
attack my school. I went and hid with a priest for a year. Life was difficult with the priest and so I went to Tanzania. I found someone to take me to Tanzania where I found my family. I was mistreated by my family for being raped and having a child.82
The way women have been targeted in these two countries is very specific
and not only a matter of “acts of war” as some might see it. For instance in
Rwanda the message was that the Tutsi were to be eradicated from Rwanda and
that which symbolized the nation, women, would be ensured to go.83 Women had
a very specific and direct position in the conflict and it can be seen by the way
women are viewed in the myths. Not only do myths and stories reveal this,
however how women have been viewed during war time in Rwanda and Burundi
is revealed by the Hutu Ten Commandments from Rwanda and the Seventeen
Rules of Tutsi Conduct from Burundi.
The Tutsi Conduct state as follow:
Do not trust a Hutu or anyone supposed to be one... Do not buy anything from a Hutu...Try to locate Hutu residences so that you will know, when the times comes, whom to save and whom to liquidate... Some Hutu women look like Tutsi, and their job is to spy on us; be careful! Do not trust appearances they are as dangerous as AIDS... There are subtle ways to exterminate Hutu people: you can isolate them in the bush, and make them disappear one after the other, ni vu ni connu; you can send them pretty girls or Rwandese prostitutes... ,84
The Hutu Ten Commandments from Rwanda are much more explicit about women:
Every Hutu should know that a Tutsi woman, wherever she is, works for the interest of her Tutsi ethnic group. As a result, we shall consider a traitor any Hutu who: marries a Tutsi woman;
82 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 24, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 83 Baines, 482. 84 Lemarchand, xviii.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107
befriends a Tutsi woman; employs a Tutsi woman as a secretary or a concubine.
Every Hutu should know that our Hutu daughters are more suitable and conscientious in their role as woman, wife and mother of the family. Are they not beautiful, good secretaries and more honest?
Hutu woman, be vigilant and try to bring your husbands, brothers and sons back to reason.
The Rwandese Armed Forces should be exclusively Hutu. The experience of the October [1990] war has taught us a lesson. No member of the military shall marry a Tutsi.85
These commandments and code of conduct explicitly indicate the
importance of women in the conflicts. The conflicts are not merely about two
ethnic groups that abhor each other; it is more intrinsic than that. The nature of the
conflict includes profound dimensions that reveal just how involved women are in
the conflict, whether they took up arms to fight or not.
A question that can be raised, that is seldom addressed, is how the ethnic
animosity was manifested in the conflict in the relationships between women.86 In
Rwanda, although the main targets were Tutsi women, there were some Hutu
women that were killed and raped, especially the very few that were married to
Tutsis or those who tried to help Tutsis.87 However, there are some Hutu women
o a that participated in the genocide by urging men to rape Tutsi women. The
participation of Hutu women demonstrates to what degree the ethnic animosity
toward Tutsi women was internalized even among women. Hutu women who
85 Human Rights Watch Shattered Lives. 86 Phyllis Chesler presents theories and analysis on violence between women. See Phyllis Chesler. Women’s Inhumanity to Women. New York: Plume Printing, 2003. 87 Human Rights Watch Shattered lives. 88 Baines, 489. Elizabeth Jean Wood “Variation in Sexual violence during War”Society and Politics 34 no. 3 (2006): 325.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108
were undervalued and completely controlled by their fathers and the men in their
lives, where finally given a role in state affairs through the Ten Commandments,
to protect their homes. For some, they took it as far as participating in the conflict.
Despite how logical it seems for Hutu women to have participated in the conflict,
it seems less evident as to why many of them risked their lives to save Tutsi
women. One of my Hutu female refugee informants explained:
During the conflict in 1993 Tutsi women tried to hide Hutu women and vice versa. I never saw or heard of women being angry with each or killing each other. I did hear that Tutsi women were killing their Hutu husbands. Hutu women did not marry Tutsi men.89
What is of particular interest here is that it appears as though Hutu women and
Tutsi women are seen as not having a vendetta against each other; and any
hostility of a woman is only directed toward men. Therefore Tutsi women killed
their husbands, and Hutu women did not kill because they were not married to
Tutsi men who they would have perhaps killed had they had a chance to marry
them.
Therefore one can conclude that the Tutsi and Hutu ethnic conflict was
mainly a conflict between Hutu men and Tutsi men. However, the complexes that
Hutu men have, makes them desire and detest a Tutsi woman at the same time.
Therefore a Hutu man suffers from an internal tug-of-war, where on one hand he
believes that the Tutsi is superior and this is manifest in his preference for a Tutsi
woman. On the other hand, he rejects his so-called inferiority and demonstrates
this by raping and killing the Tutsi woman, not only because she gives birth to the
89 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109
Tutsi community, but also because his desire for her reminds him that he is
considered to be inferior. Before raping and murdering Tutsi women in Rwanda,
the Interahamwe said such things as, “You Tutsi women, you have no respect for
Hutu men. So now, choose between death and marriage to a Hutu
Interahamwe.”90 The Tutsi men on the other hand as demonstrated by the
Seventeen Rules of Tutsi Conduct, inevitably did not suffer from inferiority
complexes; instead their “complex” stemmed from their preoccupation to
maintain the purity of the Tutsi ethnicity by ensuring that Hutu women were
avoided. In addition, it was not just a matter of avoiding any type of Hutu woman;
it was especially against Hutu women that looked like Tutsi women. Therefore it
was tacitly understood that a Hutu woman that “looked” like a Hutu was
especially not a threat nor was she desirable.
In addition to the gendered dimensions of the ethnic animosity, the ethnic
animosity in the conflict was very superficial, and this can be seen during all
phases of the conflict. For the refugees the journey from Burundi to Tanzania was
very harsh, especially for women. There were concerns of encountering military
or militia groups during their flight and being raped. Many of them were saved by
their ethnicity. The question of which ethnicity one belonged to lingered on
throughout their journeys into the refugee camp. One Tutsi female refugee
informants explained:
I met men with machetes that tried to kill me at the border. I told them that I had fled because I had almost been killed so they spared me. They let me go because they thought since I was almost killed I must be Hutu. A
90 Baines, 489.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110
woman I met along the way had pity on me and took me into her house. She soon asked me to leave because she said that if people knew she was hiding a Tutsi they would kill her.91
This demonstrates the degree of superficiality of the ethnic differences. This
informant managed to survive because the militia could not tell what her ethnicity
is, and instead assumed she was Hutu and therefore sparing her life. This
informant reveals the role of imagination and superficiality of the ethnic conflict.
The Burundian Refugee Movement to Tanzania and Role of Rebel Groups
As demonstrated above, the origin of the Burundian refugee movement
into Tanzania has its roots in deep-seated ethnic and civil conflict. This section
moves from creation of ethnic and political animosity and explains the movement
of refugees to Tanzania and the role of rebel groups. This section also attempts to
demonstrate the elements that make the refugee camps militarized.
Based upon the analysis presented by Zolberg et al. (1989) the conflict in
Burundi started after independence in 1962, and after the Hutu and Tutsi conflict
in Rwanda escalated.92 The Rwandan Tutsis that had taken refuge in Zaire were
expelled from Zaire and forced to settle in Burundi. In Burundi the expelled
Rwandan Tutsis promoted anti-Hutu policies. In response to these policies, a
mutiny was staged by Hutu militias but was unsuccessful. The mutiny led to a
backlash by the Tutsis who then slaughtered many Hutu leaders. It was during this
time that the Mwami, who had been in power, left the country and was replaced
91 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 22, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 92 Zolberg, 72.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I ll
by Captain Michel Micombero, a Tutsi.93 Tutsi supremacy increased while
Micombero was in power, thus inciting Hutus to plot a coup. When the Hutu
plotting was revealed in 1967 many were executed and arrested by Tutsis. It was
also during this time that Hutus fled to Tanzania and started to use Tanzanian
refugee camps as bases from which they could operate. The Hutus and Tutsis saw
the use of military techniques as the only way to achieve their goals of political
hegemony. Militarism was seen as a solution in the refugee camps where exiled
Hutus initiated their politicization of refugee camps. As indicated by Zolberg,
refugee populations provide legitimacy for rebel militias. Therefore rebel militias
took advantage of refugees by not only establishing political ideologies but also
by completely merging politics with militarism.
In 1972 Micombero dismissed his entire government and took sole
command of the state.94 In response to Micombero’s despotic act, Hutus decided
to band together against Tutsis and Micombero. Those included were Hutu
refugees in Tanzania, and Tutsi refugees in Zaire. Together the Hutus and some
Tutsis slaughtered many Tutsis.95 Micombero retaliated by carrying out a
“selective genocide” which resulted in the deaths of approximately 200,000
people.96 An estimated 150,000 Hutus fled. In 1976 Tanzania had 110,000
Burundian refugees, Zaire had 24,000 and 15,000 were in Rwanda. The
Burundian rebels that fled to Tanzania continued to stage raids from Tanzania.97
93 Zolberg, 72. 94 Zolberg, 72. 95 Zolberg, 72. 96 Gil Loescher and James MilerProtracted Refugee Situations. (New York: Routledge, 2005), 44. 97 Zolberg, 47.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 112
Micombero was eventually overthrown by his cousin Baptiste Bagaza. By 1986
Tanzania had an estimated 160,000 refugees. In 1987 Bagaza was overthrown by
Major Pierre Buyoya.98 The turmoil in Burundi paved the way for the security
problems inherent to Burundian refugee women in Tanzanian refugee camps. The
conflict resulted in many combatants taking refuge in the same camps as ordinary
civilians, thereby creating environments and circumstances lacking in human
security. The fact that combatants were able to be in the camps demonstrates an
element of weakness in the international refugee regime in that combatants were
allowed to reside side by side with ordinary refugees. Tanzania is similar to
Congo in that it has received a large number of refugees, and among the refugees
there have been victims and perpetrators of violence. In response to the
genocidaires of Rwanda, the Tanzania government opened Mwisa camp which is
a separation facility where refugees involved in military and political activities in
the camp are taken.99 It is inevitable that the presence of combatants and their
goals to preach their political agendas, as well as their political agendas with
military tactics created a militarized environment in the refugee camps of
Tanzania. The Burundian government has often complained about the use of
refugee camps in Tanzania as bases for militaristic ventures against them by rebel
groups. In response to the complaints, UNHCR deployed a group of international
98 Zolberg, 49. 99 The Lawyers Committee for Human Rights,Refugees, Rebels and the Quest for Justice (New York: The Lawyers Committee for Human Rights, 2002), 28-29.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 113
security trainers that initiated and trained the tsungu tsungu system, a system of
local policing.100
Despite the atrocities faced in the decades after independence, Burundi
experienced a remote hope to the end of conflict in 1993 when it had its first
democratically elected president, Melchior Ndadaye. Ndadaye’s presidency ended
decades of Tutsi rule. He was the head of the Front pour la Democratic Burundi
(FRODEBU).101 The assassination of Ndadaye in October 1993 during yet
another attempted coup102 led to the escalation of more violence committed by the
Tutsi-dominated army. Approximately 150,000 people lost their lives103 and over
500,000 people fled to neighboring countries.104 In July 1996 an army-led coup
was successful in Burundi which returned Pierre Buyoya to power.105 Buyoya’s
government immediately started to discredit the various rebel groups by pointing
out their links to Rwandan Interhamwe and ex-FAR.106 By December 1997, as a
result of Burundi’s instability, Tanzania was hosting 230,000 Burundian
refugees.107 The series of coups in Burundi is indicative of the level of
militarization in the country. Militarism has seeped into all areas of life, and has
become the main mode of establishing political and ethnic power. The coups have
led to the further displacement of individuals, and Tanzania is a country that has
had to face the repercussions of the instability in Burundi.
100 The Lawyers Committee for Human Rights, 209. 101 Erik Doxtador and Yeki Mosomothane, 66. 102 Doxtador and Mosomothane, 66. 103 Doxtador and Mosomothane, 67. 104 Loescher and Milner, 44. 105 Doxtador and Mosomothane, 68. 106 Doxtador and Mosomothane, 68. 107 Loescher and Milner, 44.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114
The conflict in Burundi has led to the deaths and displacement of a
significant number of people. The situation of instability and paucity of human
security has been created by both government and rebel actions. According to
Erik Doxtador, many people have died in the crossfire between government
troops and the rebel groups; and numerous people have been killed by the
government, especially individuals suspected of collaborating with the rebel
groups.108 The rebel groups have also been involved in indiscriminate killing of
civilians.109
The rebel groups are known to receive their assistance from the
governments of Democratic Republic of Congo and Tanzania.110 For example,
Richard Giggs writes that Tanzania is known to have supported many Hutu
militias in Burundi such as the PALIPEHUTU, FDD and FROLINA. These
groups were provided with training bases near Lake Victoria and in and around
refugee camps (Richard Giggs n.d.).
Despite Tanzania’s involvement with Burundian rebel groups, Tanzania
has been known to be a very hospitable country to the Burundian refugees.
Between 1992 and 1994 the population of Burundian refugees in Tanzania tripled
from 292,100 to 882,300.111 Even though many refugees have been repatriated
back to Burundi, there still are as many as 400,000 refugees in Tanzania.112 The
108 Doxtador and Mosomothane, 70. 109 Doxtador and Mosomothane, 70. 110 Doxtador and Mosomothane, 72. 111 Loescher and Milner, 45. 112 United Nations High Commission for Refugees, “Funding Crisis Threatens return hope for thousands of Burundian Refugees” 28 October 2005 available from http://www.unhcr.org/cgi- bin/texis/vtx/news/opendoc.htm?tbl=NEWS&id=4362556c4 accessed on June 12, 2006.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115
large number of refugees is seen as problematic in Tanzania. Burundian refugees
have been deemed by the Tanzanian authorities and locals as responsible for the
insecurity, environmental degradation and the disruption of the Tanzanian
economy.113
Conclusion
The Burundian refugee movement has its roots in a history that is marred
by ethnic tensions. Taking a closer look at the history of the conflict, one notices
the intrinsic role that women have in this conflict. Women have been pivotal
throughout the conflict. They have been used as a way to demonstrate ethnic
hegemony through rape and other actions that target them as victims. Culturally
women have been used to create delineations of the differences between the
groups by attaching different attributes to Hutu and Tutsi women. By placing
women in this conflict and giving them such a position of importance, an
importance that is perhaps rarely evident, one can piece together how and why
women have been militarized in this conflict, especially the women who end up in
refugee camps. The refugee camp, although a setting far from where the original
violence occurred, is very much involved in the violent struggles of the country
and is an even more poignant reminder of the conflict. The conflict history
demonstrates a confluence of ethnicity, politics and militarism. Women are
positioned in the middle of this due to the symbolism of women in a highly
patriarchal society. Women are at the heart of the Tutsi feelings of superiority and
113 Loescher and Milner, 45.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116
the Hutus feelings of inferiority. While Tutsi women are valued, Hutu women
have insignificant attributes. The turn of events however has created different
sorts of women. It has created the sense that Tutsi women are the enemy, and
Hutu women have been given new roles as the protectors of the home and the
Hutu community. The refugee camps that are highly militarized due to the
political and militaristic activities have profound effects on the role of women. It
is from this historical background that we can understand what type of situation
refugee women are facing. However, this historical analysis only explains half of
the story. The next chapter presents the experiences of refugee women that reveal
their militarization. I discuss ways in which the military is present in the lives of
the refugee women, and how this presence labels refugee women as militarized
individuals.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER FIVE
EXPERIENCES OF MILITARIZED REFUGEE WOMEN: QUESTION
OF FEAR AND DEPENDENCY ON THE MILITARY
Kanembwa refugee camp is a militarized camp, and I contend that the
refugees in the camp are all militarized. Militarization of refugees in this context
is understood, as noted in chapter three, by the definition presented by Cynthia
Enloe that militarization is a process where someone is gradually controlled by or
depends on the military or mil itaristic ideas.1 Therefore what one learns is that
refugees in Kanembwa camp have been at one point or another controlled or/and
depended on militaristic ideas.
Muggah explains that the factors that provide evidence of refugee
militarization include: military activity in and outside camps, political activism
and violence, military training and recruitment and support for combatants and
armed resistance.2 However I present other factors indicating refugee
militarization that are not all tangible. Militarization is understood in this study
not only as participation in militaristic activities; it is also about the pervasive
nature of militaristic activities and ideals that become ingrained in an individual’s
life. Militarization in the lives of refugee women is an issue that is in some
respects very tangible and visible. For example when individuals take up arms
1 Enloe,Maneuvers: the international Politics of Militarizing Women’s Lives, 3. 2 Muggah, 8. 117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 118
and support militarism, they can be considered militarized. However, there are
other more symbolic and intangible factors that reveal militarization. For instance,
fear of militia groups in the refugee camp demonstrates control. In my view fear
of the military and militia is a sign of militarization. Kanembwa is a concrete
symbol of the consequences and continuity of militarism and refugees have been
directly affected by militarism. Some have been directly involved with militaristic
activities. In this study, individuals do not have to take up arms or be directly
involved in militarism and politicization to be militarized. However, by the same
token a refugee camp that is militarized provides a pool for recruitment and to
some degree is symbolic support for warrior-refugees’ ventures. What one finds is
that militarism is in the lives of refugees, sometimes in tangible ways and in many
instances in abstract ways.
Therefore, this chapter addresses the experiences of militarized refugee
women. I explore how militarism is linked to the lives and experiences of refugee
women. By using Enloe’s definition as a guiding conceptual tool, I looked for
elements that demonstrate that refugee women were and are controlled by
militaristic elements. Control in this study also refers to elements of fear,
particularly the fear that refugee women have of militarism and militia groups.
The second element that I looked for was the dependency that refugee women
have on militarism which is demonstrated through voluntary politicization,
support of militia rebel members and ideas.
This chapter is divided into two sections based upon Cynthia Enloe’s
definition of militarization. The first part analyzes militarization as “being
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 119
controlled by military and militaristic ideals” and the second part is being
militarized “by depending on the military and militaristic ideals.”
“Control” bv Militarism
By using fear to mean control in this context I elucidate the pervasiveness
of militia and militaristic control on the lives of refugees. The control is obviously
present in that as long as conflict is going on, and militaristic activities persist, the
refugees fear returning to their homes and therefore live in the refugee camp. The
fear of returning home due to rebel militias is not a very simplistic notion but it is
also attached to other types of fears. For instance some refugees fear returning
back home to nothing and to no one. Some fear losing the assistance that the
NGOs and UNHCR provide. Many refugees would prefer to go home, however it
must be noted that there is an element of fear of militarism that exists for many
refugees. What I present in this section is a more nuanced look at fear of the
military. Fear for one’s security because of militaristic activities can be seen at all
stages of the refugee plight. Fear of militarism is not only associated with fear of
the military in Burundi, but there is also the fear of rebel militias in and around
the camp. Some of the fear that is associated with the military for some refugees
dates further back into the 1970’s even before Kanembwa camp was created. As
one informant explained to me:
My father was a technician. In 1972 my father was killed by the government military. I am a Hutu. Life became hard after my father died. My mother was a peasant but she managed to educate all of us. She would cultivate the land and do some business. My parents had seven children. My mother was a Hutu. I never understood the ethnic conflict. My father’s
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120
murder by the military has caused me to have serious fears of the military.3
The expression of fear indicated by the informant clearly delineates the pre
existence of an understanding of the horrors associated with militaristic activities.
This informant could only describe soldiers as men who do nothing good other
than kill. There are many other women who experienced traumatizing experiences
with the militias and military that influenced their fear. One female refugee was
forced to eat human flesh by militias during her flight to Tanzania from Burundi
in order to save her life. These are all experiences that refugee women had prior to
being in the refugee camp. Once in the camp however, their fears come with them
and their very lives hinge upon the fear they have of the military and militia
groups. This can be seen by the many refugee women that succumbed to the
demands of recruiters in the mid 1990’s for contributions for the cause of the
various rebel groups. One fonner rebel militia explained the fear that he and other
former combatants exerted on the other refugees. He expressed the classical
example of how warrior-refugees were viewed in the camp by stating that,
“before, they used to be afraid of us because other rebels would threaten refugees
by telling them to give up their food. Rebels would tell refugees that if they didn’t
join the CNDD-FDD and go back to Burundi the refugees would be killed.”4
Another refugee reflected on the activities of the combatants in the camp and the
sort of control that they had on the lives of other refugees:
3 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 4 Refugee informant in Kanembwa Refugee Camp, Interviewed by Author, January 17,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121
The camps started having organized military and political groups in 1995. The combatants told us to reorganize and to start working for our country. They asked for contributions and they asked especially from those who worked. I worked so they asked from me all the time. I felt that it was obligatory to give money. At first it was the FROLINAT after it was the CNDD. I did not understand them. I was afraid of asking them questions about their objectives. I felt that even if I joined these groups my family would not return. We all gave contributions because if we didn’t give them, they said they would kill us. They gave each one of us membership cards so that if we returned to Burundi we should show that we were part of them. There were women combatants during this time. Some women became members because their husbands joined the groups. And as a member you had to sensitize others. Some felt that militants protected them. Some women were just trying to take charge of their lives.5
Therefore it is clear that the refugees’ daily lives at one point were highly affected
by militaristic ideals. Refugee women were forced into militarization through fear
and coercion. By threatening their lives into supporting them, refugee women
found themselves involuntarily being members of rebel groups, and even having
membership cards. Refugees were used by rebel militias for their political and
militaristic agendas. The women that were married to men that joined rebel
groups took the role of sensitizing others about rebel militia’s agendas. This
indicates that women took an active role in politicization as militia wives. Even
though this role is significant, it presents the conventional role that women have
played as military wives whose only involvement in military activities is to
support their husbands and the militaristic institutions that their husbands are
involved with. The fear of these militias based upon experiences during the
conflict put many women under the control of militias. The feeling of obligation
to give and provide for a cause one did not even understand or did not support
5 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 20, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 122
emphasizes the magnitude of fear. It is clear that after having your entire family
killed by militarized individuals one is bound to have a level of fear. The control
aspect of the militarization of refugee women is in itself very much embedded in
the more conventional understanding of women and military. Enloe explains that
the military wants to control women and the idea of femininity.6 That was
accomplished in the refugee camp of Kanembwa. Many women still walk around
wondering if they will be killed for never having enough to contribute to the cause
of the rebels. One woman expressed, “I was asked to make contributions to
militias but had nothing to give to the recruiters. I was threatened and told that I
would be killed in Burundi for not giving contributions.”7
The fear that is associated with militia groups however does not only exist
in the refugee camp. The fear of not giving extended to those who felt that if they
returned to Burundi after not contributing to the cause, they would face death.
However there are some instances where those who are in refugee camps are
tainted for a long time if they leave the camp because they are accused of being
involved with militias. The way in which refugees that return are viewed is
affected by militarism. The identity of a returnee is under the control of militarism
because of the association that many make between the refugee camps and
military activity. One woman expressed to me the fears that she encountered
when she repatriated back to Burundi and her entire community accused her of
being a militant. The very aspect of being in a refugee camp that is militarized
6 Enloe, Maneuvers: the international Politics of Militarizing Women’s Lives, 36. 7 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 22, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123
creates a fear not only in the camp and not only of the militias and military. But
others who did not flee into the refugee camps sometimes fear those who fled
because of the clear connection that those who stayed behind in Burundi make
between militarization and refugee camps. This young woman expressed to me:
In 1996 there were problems of combatants. I was scared that they would kidnap me. I returned to Burundi in 1996 after being in the camp even though that was a dangerous time. When I returned to Burundi, people in my village thought I was a combatant because I had come from the camps. They believed this because there were many women who joined the militias. Those in the camps joined voluntarily but those who joined in Burundi were forced. My parents wanted me to leave Burundi and live elsewhere because of all the threats. I did not go to school while I was in Burundi. I faced problems because I had lived in the camps. People would ask me questions about being a combatant. I am afraid and have always been afraid of combatants. I had nothing to say to the accusations. I know women who chose to be combatants. Many of the women joined because their husbands and other family members were killed and so they joined to revenge the deaths of their lived ones. In 19971 left Burundi to come back to Kanembwa.8
This informant’s story is indicative of the perceptions that those who stayed
behind have of refugee camps. The fact that refugee women can be accused of
joining militia groups indicates the perceptions by others that refugee camps have
a changed culture, one that allows women to participate in armed conflict. The
perceptions of those who stayed behind of refugee women as militant due to
changes in culture is not surprising due to the stories that one hears about militant
women. One refugee informant indicated that militant refugee women are
“considered super patriots that are very dangerous. One woman almost killed her
8 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 19,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 124
husband when he attacked her. They apparently are not afraid of anything.”9 In
addition, the refugee woman informant indicates that there were many refugee
women that joined rebel groups to avenge the deaths of their husbands and family
members. It can be stated that after losing her family, a woman whose life has
been restricted through patriarchal and family control, might experience an
internal crisis. The loss of family, and the loss of patriarchy, could lead to socio-
psychological insecurity for women who might feel a sense of losing the norms
and community in which she belongs. I contend that by joining rebel groups,
refugee women not only avenge the deaths of their loved ones, however some
might also take it upon themselves to avenge the demise of their traditional life
and community norms.
The militia groups and military are not abstract groups that create fear
only during war and during military activities such as recruitments. Some women
face individual threats from individuals in the militia groups who use their
participation in the militia groups to threaten people they know. For example a
refugee woman in Kanembwa is threatened by her step-son who has vowed to kill
her because he is convinced that she is Tutsi even though she claims to be Hutu.
She stated that:
When the war started my step son left our home and told people that I am Tutsi and to kill me. I am a Hutu. I have had many problems with my husband’s children and family. My mother in law surprisingly saved me during this time even though I did not get along with her before the war. She told people I was her family member and that I am Hutu. The boy told everyone that I am Tutsi, even though I am Hutu. When we arrived in the
9 Refugee in Kanembwa Refugee Camp, Interviewed by Author, January 22,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125
camp we found him here as well and he told others that I am Tutsi. I was beaten in the camp because of him. My husband and I called the police and his son went to Mtendeli camp and then he went to Burundi. He wrote a letter to us from Burundi saying that he was going to kill me and his father now that he is a PALIPAHUTU combatant.1
The ethnic animosity transcends family ties. Husbands have killed their wives,
and wives have killed their husbands. At this point one should ask, why not step
sons killing their step-mothers? Therefore the fear of the militia and military at
this point is not necessarily associated with the militaristic ventures and ideals.
Militarization is very intricate in Kanembwa because of the dimensions it
manifests which seep into the private sphere and into the personal lives of refugee
women. The violence experienced by women therefore is highly associated with
militarism because it takes the form of violence meant to eliminate individuals,
Tutsis in particular, that can pose a threat to state and nation building objectives
of the rebel militias. Therefore the violence faced by women is intricately
associated with militarization.
One of the most questionable consequences of militarization is the
phenomenon of banditry. Many people in the camp and outside the camp are
convinced that the bandits that live in the forest are former combatants. The facts
that they carry ammunition that is used in combat, and they apparently use
strategies that reveal a certain level of military training, lead many to believe that
they are former combatants.11 The very fact that there is suspicion among
refugees about who the bandits are and that they could be combatants reveals a
10 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 20,2007, Kanembwa Refugee Camp Kibondo Tanzania. 11 This was expressed to me by both UNHCR staff and Ministry of Home Affairs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126
large degree of fear for personal security. A group of Twa refugees expressed to
me: “We are safe here but at night we fear the Bandits that are armed and kill.
Bandits are both Burundians and Tanzanians. They kill everyone.”12
The question that is raised is especially pertinent to women who collect
firewood and end up being raped and attacked by these bandits that live in the
forest. The bandits also attack people in their homes. One of my informants is
convinced that they are rebels and she stated that:
There are rebels who can attack the camp as well. In July 20061 was attacked for the 5th time. I don’t know why I have been attacked frequently. They hurt me in the back and until now my scar has not healed. The night they attacked me, the bandits all came with arms and they shot at whoever tried to help us. They spoke Kiswahili and so I did not understand what they were saying. I am afraid that I will be raped next time they come, and they could rape my daughters.13
Many refugee women believe that the bandits in the forest are Burundians who
disguise their identities by learning Kiswahili. The identity of the bandits is highly
contested. One UNHCR staff member stated that, “the people of Kibondo,
Rwanda and Burundi have same ancestral background so you can’t tell who is
who.”14 There is a clear debate of whether these bandits are Burundian or not.
However on some occasions they have been found to be Burundians. As
indicated by one UNHCR staff:
In Nduta two women were staying together. At night the house was attacked, the bandits raped the women. The bandit rapist was identified as
12 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 16, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 13Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 17,2007, Kanembwa Refugee Camp Kibondo Tanzania. 14 UNHCR staff, Interviewed by Author, January 24, 2007, Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 127
a Burundian who was suspected of banditry and so he fled. He was later found in the forest working with other bandits.15
Bandits present tangible evidence of the consequences of military activities and
of militarization through their use of firearms and light arms that have been
trafficked in the region. The existence of these bandits is a clear consequence of
militarization and is associated with the militaristic activities of militia groups in
the forest. There is a clear distinction between banditry and militarism; however
in this case the distinction is almost completely erased because bandits use
military tactics and use weapons that are trafficked as a result of military
activities. Therefore, the banditry and criminal behaviors are attached to the
militarism that is evident in the area. Another refugee raises the same issue by
stating that:
The forest has armed groups that attack the camp. No one knows who they are. 20011 was attacked by Bandits during the night around 9pm. They spoke Kiswahili. They could be Burundians. To not be unknown they speak Kiswahili. They came with guns, and axes. They beat us and stole what they wanted. They beat my children. When help came it was too late.16
Some officials are convinced that they are former combatants and that they
collaborate with refugees in the camp. One aid worker indicated that, “the bandits
look very trained and are well equipped. That is why they are suspected of being
ex-fighters.”17 Whether they are ex-fighters or not is highly contested, however
what is clear is that their tactics seem to indicate a level of militarization due to
the proliferation of small arms. Whether they are former combatants or not, the
15 UNHCR staff, Interviewed by Author, January 24,2007, Kibondo Tanzania. 16Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 17 NGO informant, Interviewed by Author, January 24, 2007, Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128
fear and suspicion by other refugees that they could be former combatants reveals
a high level of militarization through fear. The bandits are an indication of the
militarization that is present in the lives of refugee women. Refugee women
associate the activities of the bandits to those of rebel militias. The use of fire
arms, as indicated by Muggah, is another element that clearly signifies
militarization. Therefore in this context, refugee women are militarized in the
sense that militaristic activities or the perceptions of militaristic activities are
present in their every day lives through the activities of bandits. The distinction
between criminality and militarization is blurred and at the same time fused. The
militia and military, especially for those who fear them, are represented in all
armed activities and circumstances that bring little good. The militia groups that
have operated in and around Kanembwa camp have left their mark by instilling a
perpetual fear on refugee women. Some women overtly state their fear of being
raped by these groups. Therefore the fear has taken a gendered dimension that is
recognized by the refugee women who not only fear the bandits for the abuse they
perpetrate, but also because as women, there is a specific abuse that they can face
that is perpetrated by bandits: rape.
Militarization as Dependency
After presenting examples of militarization of refugee women through their
experiences of being controlled by and fearing militaristic ideas and activities, this
section will present the other side of the coin. What is presented here is
militarization of refugee women through the experiences of women who depend
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 129
on militaristic activities and ideas for their survival and protection. Women can be
controlled by military and that is a phenomenon that is perhaps more conventional
and understood as somewhat inevitable during military operations. The
unconventionality of refugee women’s militarization as a result of dependency is
compounded by the conventional control of women by military. I present here, a
different form of militarization. In the following chapter I will present an analysis
of how fear and dependency both create one common result of militarization.
For women in the refugee camp, whether one fears or depends on the
military, the military and its ideals become a constant shadow (or sunlight)
guiding one’s decisions, lifestyle and relationships. For example many women
who fear the military made the decision to not question when asked to make
contributions to militias and instead succumbed to the rebel militias’ demands.
Women who depend on the military made the decision to risk their lives in the
camp and work as recruiters, or even as combatants. In Burundi every political
party in one way or another is involved with militaristic ideals. For example,
FRODEBU a party that many have described to have the most non-militaristic
agenda has PALIPEHUTU members.18 In addition the CNDD was created in
1995 by one of FRODEBU’s founders Leonard Nyangoma.19 The fact that some
of the influential figures in FRODEBU have militaristic tendencies signifies that
FRODEBU has militant elements that influence its behavior.
18 Lemarchand, 186. 19 Gregory Mthembu-Salter, “Self-Determination Country Profile: Burundi”Foreign Policy Focus, 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 130
Burundi does not have a large population of women who were combatants
and took up arms. This is what makes the militarization of women in the conflict
very particular. As stated by an official from the United Nations Mission in
Burundi (UNOB), who is working on DDR process:
All rebel groups/ parties had to submit a certified name list of members before anyone was accepted [to under go DDR]. This was initially a problem because only the men were listed, but through political pressure, we were able to include the women - about 490 of them.20
The number of women as combatants in the Burundian conflict is very small
however this does not mean that militarization can only be measured by the
number of individuals that take up arms. There are other ways that women have
been militarized in the Burundian conflict. In Kanembwa camp there are women
who have been involved in recruitment of combatants, there are some who have
worked as cooks for the militia, others are married to men who work as militants,
some promote the ideologies of the military, and the list goes on. As one refugee
woman explained to me:
I lost my whole family. My husband was part of the FRODEBU. I was also part of FRODEBU. After Ndadaye was killed my husband fled to Tanzania with nothing. I lived in trees and bushes for one month, sleeping in swamps. I went to another town only to be looked for by government soldiers. They came to my house four times. My neighbors were nice and they helped save me. I fled to Tanzania with four children and a toddler on my back. I came to the camp and was active in politics. I was secretly working for FRODEBU in the camp and believed in Ndadaye.21
Another refugee woman said to me that she was a member of FRODEBU also
because it was a party with clear ideas. PALIPEHUTU on the other hand, she
20 Waldemar Very, interviewed by Author by email. 21 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 16,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 131
described as not having clear ideas.22 It is interesting to note that PALIPEHUTU
is not favored by these women because of its ideas that are not clear, instead of its
militancy. This aspect of these two women’s militarization is discussed by Enloe
who states that militarization is so pervasive that the more militarized an
individual or society is, the more the individual or society will have the view that
military needs and militaristic presumptions are valuable and normal.23 In
addition, although FRODEBU is a party that is described as not having the same
sort of military components as PALIPEHUTU or the CNDD, what is clear is that
FRODEBU has some extreme members that are members of other extreme
parties. Therefore even politicization of FRODEBU ideals is somewhat linked to
militarism. Although the women claim to support the ideas of FRODEBU, they
are indirectly supporting a militarized entity, and its ideologies. The dependency
factor as regards militarization in the two examples stated above is clearly linked
with the understanding that FRODEBU will bring a better future because of its
“clear ideas.” Therefore the women depend on FRODEBU for their future well
being to the extent that they secretly promulgate FRODEBU’s ideologies.
Burundian refugee women in Kanembwa had a vital role in the militia
activities especially those pertaining to recruitment. As one refugee informant
indicated women who were the most vocal were asked to be recruiters and to
recruit other women. Women’s dependency on the militia groups took many
22 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 17, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 23 EnloeManeuvers: the international Politics of Militarizing Women’s Lives, 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 132
different forms and many of which were characterized by strong vocal support for
militia ideologies. As stated by one refugee woman informant:
People were recruiting in the camps. They wanted young girls, boys and men. Women recruiters were in the camp. These women encouraged other women to wake up and take part in the struggle. The women who recruited were already educated. They were some women who joined but I feared the military. I see the military as a place for a man who likes to kill. Military is for men. Even my husband does not want to join.24
This informant explains her reluctance to join the rebel militia due to fear of
military. The fear of the military kept some women from joining and that is why
other refugee women encouraged other women to take part in the struggle and
“wake up.” By telling other women to wake up, this indicates that emancipation is
necessary for militarism. As one informant indicated, it was mainly the educated
refugee women who were involved in militaristic activities. Some women wanted
to join the militia groups however there were instances where women said their
husbands would not allow them to join. Therefore it is clear that there was an
inevitable clash of cultures in the refugee camp as regards to women’s
militarization; emancipation of women against patriarchal values. Some girls
wanted to join and if their parents did not allow them to join they would runaway.
Some girls were known to join because their parents wanted them to avenge the
deaths of their families and for the Hutu cause. For example one refugee
explained the reason why some refugee girls joined in Lukole camp by stating
that:
24 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 133
Some refugee women were recruited in Lukole camp and they were recruited because their parents were political and strong Hutu supporters. They joined for whatever solidarity is associated with Hutu-ness.
As one young refugee woman explained to me:
The whole idea of child soldiers started in 1994. The child soldiers were forced to fight, even girls were forced to fight but only CNDD-FDD seems to have girl soldiers that chose to fight. I know 5 girls that fought but now they are in Burundi. The girls went to fight to avenge the deaths of their family members. 26
One woman explained to me the unconventionality of women’s recruitment, and
to what extent a woman would go in order to be part of a militia group:
Some women chose to be combatants. In Burundi there are now many women in the military. People thought it was weird for women to fight. I think it’s bizarre that a woman would leave her children to fight. I am not sure why women choose to join the military. When women were being recruited they joined in the camp because of their ethnic group. Before 19931 did not notice women leaving their children. After the war, women left their children to become combatants. The life of a woman in a camp is very difficult than life in Burundi. Women can’t take care of their children in the camp, and they can’t feed or clothe their children.27
Therefore it can be established that the dependency on the military was a result of
the history that, as indicated earlier was connected to ethnic animosity. However
for some, they joined because life was hard and they could not take care of their
children. Joining the militia groups might have provided women with a better
strategy for survival. Parents caught up in the history of the conflict and the pain it
inflicted sent their children away to be part of this history and avenge the deaths
and destruction. There is also the element of cultural break down where women
25 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 16, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 26 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 17, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 27 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 134
left their children in order to fight. In a society where family ties are valued, such
an act of abandoning ones children expresses to what extent family values were
broken and how society was disrupted by war. One former combatant explained
to me that there were many reasons why refugee women fought and joined the
militia groups. He stated that many joined voluntarily for patriotic reasons, other
joined because they had family problems:
Women were used as cooks for the fighters but they were also trained to know something about the military. They were cooks because someone had to do the cooking. Women were trained as police as well as some men. The police were called military police. Some of the women were from Mtendeli, Kanembwa and Ngara refugee camps. They were all between ages 15-16 years. They did not carry weapons since this was a man’s job. The girls that joined did so voluntarily. Many girls joined because of patriotism and others joined because they had family problems. We were all Hutu. But now the girls are in the national army and have been demobilized.28
Even though women and young girls did not have very high positions in the
military and militia groups, they stayed with the militia groups and are now
undergoing the DDR process. An official from the United Nations Mission in
Burundi working on DDR stated that, “[women] were very junior graded. This is
still evident in the Army and Police as well where the first female officers were
only recently appointed at the entry levels.”29
Another form of dependency can be seen through the commitment that
women have to their spouses that are militarized. There some women who will
share the values of their husband and adopt their values. One woman in the camp
was jailed because of her husband’s banditry and militia activities. Another
28 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 17, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 29 Waldemar Vrey, interviewed by author by email.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 135
woman whose husband was accused in Burundi on many occasions of being a
dissident and potential rebel told me:
I know that he has been accused of being rebel combatant (PALIPEHUTU). In 1991 he left because he was looked for by the Tutsi military for allegedly being part of a rebel group and hiding other members. When he left he was told that he was being looked for. The military came to my door asking for him. I said he was at his restaurant. They did not find him there. They came back to look for him in the house. My husband had already left by then.30
Therefore for some women, being militarized is not revealed in what they do
consciously such as being a recruiter, but it is a matter of affiliation. The two
women described above are associated with militarism by their intimate
association with men who are in one way or another linked to militia activities.
Other women have considered depending on militia groups as a way to
avenge the death of family members but many never actually join. However the
very thought of considering the militia groups as an option to avenge the deaths of
ones family presents itself in such a way that a women can view the militia as a
potential solution for the grief that is felt from loss. One woman considered
joining but decided not to because her family would not come back even if she
had joined. For some women joining the militia groups and depending on the
militaristic ideals is more about emancipation for women. Women depend on the
military to give them an opportunity to be empowered and emancipated. Some
women expressed to me the power that they believe women would get if they
joined the militia groups. One informant indicated that it was the educated women
30 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 19,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 136
that joined because educated women understood the opportunities associated with
joining. The informant stated that, “the women that believe in joining the militia
groups are the educated women. The peasant women are the ones who think that
it is not good for women to be involved in militia groups.”31 For some women the
militia is not only about being in power and getting a place in government but it is
a source of female empowerment. As one refugee woman noted, “being in the
armed forces can be seen as a promotion of women’s rights. If you notice many of
the women who joined these groups now have power. The war provided an
opportunity for women to be militant.”32 As indicated in chapter three, women
often take up militarism as a way of emancipation and liberation.
There are other women that fear for their security because of the history of
the conflict and yet find themselves depending on the militia for protection:
My oldest son is a militant for CNDD. My son was 18 at the time he decided to leave. He talked about it to me and his father. I was not happy about this but my husband was for it. He left one day without saying goodbye, and I heard from other refugees that he had left to fight for CNDD. We spent 3 years without any news from him. He left because he did not see a future for himself. He was not happy with his father who was drinking a lot of alcohol. The witches in the camp bewitched him because they were jealous that he was a hard working boy that was helping me around the house. My husband was happy for him to leave because this meant that our son could avenge the deaths of our other children. We are scared that if our son is not with us we could be killed in Burundi. Our son can protect us.33
31 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 16,2007, Kanembwa Refugee Camp Kibondo Tanzania. 32 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 17, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 33 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 20, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137
Although some fear the military, they depend on it as well. The military is given
an ominous attribute as an entity that is both feared and to some extent revered.
Conclusion
Militarization revealed through women’s dependency on the military takes many
different shapes. The military can be seen as the institution that can bring hope for
those who grieve their loved ones, it is a place where women can find protection,
it is a place where women can assume power, and it is a place that some can go to
in order to deal with the changes that conflict brings. By the same token, the
military still remains an entity that controls women and presents a level of control.
This is seen through the fear that refugee women have of the military. In both
instances what one notices is the interplay between a history of ethnic animosity,
political and militaristic activities. In this context the experiences of refugee
women are marked by instability and influenced by a continuity of external
influences that promote militarization.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER SIX
CONCRETELY UNDERSTANDING THE CULTURE OF
KANEMBWA REFUGEE CAMP
Militarization has been demonstrated in the preceding chapter as an issue
that is integral to the lives of women. The women that I spoke with revealed ways
in which military ideas and activities affect and influence their lives. This chapter
takes a closer look at the situation and condition in the refugee camp in order to
explain what about the refugee camp makes refugee women’s militarization
possible. The historical analysis provides evidence of the events in history that
have goaded the militarization of refugee women; however it is only half of the
story. This chapter dives into the nuances of camp life. It is the context and
environment of where the militarization of refugee women takes place where we
find the other half of the story. Militarization as noted throughout this study is a
process throughout history, in addition it is a process that is affected and affects
culture and behaviors. This chapter addresses the militarization of the refugee
camp and the elements that have facilitated the militarization of the camp as well
as the militarization of refugee women. The elements discussed in this essay are
not exhaustive; however they provide a picture of the context of the camp, and
how the various elements put together explain how the camp is prone to
militarization. Refugee women’s militarization, I contend is a highly contextual
issue, and it is therefore pertinent to give significance to the refugee camp as an
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 139
environment that requires analysis. This chapter puts forth the premise that
Kanembwa camp is a militarized environment. In addition the activities,
circumstances and culture of the camp reveal why it is prone to militarization.
Therefore I contend that the reason it can be said that Kanembwa camp refugee
women are militarized is basedprimarily on the fact that the refugee camp is
militarized. Kanembwa camp is not only militarized due to its symbolic value as
an environment created from military activities, however there are concrete
activities that take place, as revealed by refugee women, in and around the camp
that demonstrate a continuity of military ventures. As stated by Robert Mogire,
the debate of whether Tanzanian camps are militarized is highly political and
highly contested.1 However when one analyzes what refugee camp militarization
refers to, it is clear that Kanembwa camp is militarized. Robert Muggah (2006)
indicates seven elements that authors have addressed as evidence of refugee camp
militarization: (1) armed violence in camps (2) political activism and violence (3)
storage and trafficking of small arms (4) inflows and outflows of weapons (5)
military training and recruitment (6) infiltration of armed elements (7) use of
relief/development resources.2 My research in Kanembwa camp indicates that the
camp has many elements of militarization such as presence of small arms,
military recruitment, military-fundraising and political activism. These are all
factors expressed to me by refugee women. Therefore this chapter will address the
1 Mogire, 142. 2 Muggah, 8.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 140
situation in and around the camp, as described by refugee women that reveals that
Kanembwa refugee camp is militarized.
In addition, in this chapter I present an analysis on the way of life and
situation of refugees that explains cultural and behavior elements that make
refugee women susceptible to militarization. I present an analysis of gender
relations, communal ties, and the situation of patriarchal culture in the camp. On
one hand what one notices are the change and loss of patriarchal culture and the
loss of family ties that can be seen as explanatory factors for women’s
unconventional behavior as combatants and supporters of militarism. On the other
hand, there is a resistance to the loss of patriarchal culture and male dominance
which is demonstrated by the abuse that women face. These factors are at the
heart of militarized refugee women’s experiences in the refugee camp.
Kanembwa Camp: Political and Social Make-up of a Militarized Camp
My emphasis on experiences is what drives this study. By analyzing the
experiences of refugee women and their social life and relations in the refugee
camp I am able to interpret refugee women’s militarized culture. The experiences
of refugee women that are detailed in chapter five do not occur in a vacuum;
instead they are experiences that occur in a specific space with specific activities
that makes these experiences possible. First and foremost, Kanembwa camp,
much like other environments has its own situational identity that contributes to
the culture of militarized refugee women. Therefore what follows is a description
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141
of four elements that distinguish the camp and make it an environment prone to
militarization and its inhabitants prone to militarization. My research on the
militarization of refugee women centers on two main variables: history, and
culture, as determinants and drivers of militarization. These variables are
conceptual tools that enable one to explain the prevalence of militarization, its
very existence and nature and the different forms it can take based upon diversity
of culture, tradition and history of communities. What falls within these two
variables is perhaps inevitable and obvious, and that is location and context where
the interaction between history and culture take place. Where these interactions
take place and the location of these places have a monumental role on the nature
of militarization. For this reason I chose to look at a political refugee setting
because of its superficial attributes as a home away from home but also because
of its particularity as an environment tacitly linked to militarism. The reason that
political camps can be classified as tacitly linked to militarism is because they
symbolize the remnants of militaristic ventures in the country of origin. Not all
refugee camps are linked to political and military conflict, and not all refugees
flee fighting, for instance some refugees flee famine or drought. In addition, not
all politically and militarily produced refugee camps have militaristic activities in
and around them. However it is irrefutable that those that are created due to armed
conflict are symbolic of the consequences of militarism. Kanembwa camp is not
only linked to militarism in the symbolic sense due to the reasons for its creation;
however it is also militarized due to the activities that have taken place in the
camp that demonstrate militarization. Therefore an analysis of refugee women in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 142
Kanembwa camp and discovering if at all they are militarized is futile because,
according to the definition I am using, by nature of being in this particular refugee
camp they are connected to militarism and are therefore militarized.
What is of particular interest in this section however is the location of
Kanembwa camp. What about the camp makes it especially prone to
militarization, and has facilitated the militarization of refugee women? My
observations of the camp have led me to analyze four variables concerning the
situation and conditions in and around the refugee camp that have made it
possible for the camp to be militarized and refugees to be susceptible to
militarization. These four factors include (1) the location of the camp, (2) the
political and sometimes militaristic activities within the camp, (3) the lifestyle and
livelihood of the refugees, (4) and the role of external actors such as NGO’s and
local Tanzanians.
The camps in Kibondo are of particular interest because they are all
located in the forest in western Tanzania. As explained by one informant from the
government of Tanzania, refugees are given designated areas by the government
of Tanzania for encampment aind are helped by international organizations
especially UNHCR.3 Refugees are encamped in these forests and are not allowed
to leave the camp without permission from the Tanzanian government.4 My
observation has led me to have the view that the location or designated areas of
the camp highly impacts the level of the camps’ militarization as well as the level
3 Liaison Office for Tanzanian Ministry of Home Affairs, Interviewed by Author, January 21, 2007, Kibondo, Tanzania. 4Khoti Kamaga, “The Tanzanian Refugee Act of 1998: Some Legal and Policy Implications” Journal of Refugee Studies no.18 1 (March 2005), 110.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 143
of militarization in the lives of refugee women. The fact that the camps are
located in the forests makes the setting very dark and easily accessible by
outsiders. The forest is unmonitored and therefore different items, notably
weapons can be trafficked within the forest from various areas into the camps.
Jeff Crisp (2000) indicates the importance of looking at the location of refugee
camps in Kenya as well where refugees face insecurity. He quotes a UNHCR
official who states that “you cannot create an island of security in a sea of
insecurity.”5 Furthermore, there are many accounts of armed men that hide in the
forest around the camps in Kibondo that disrupt the lives of the refugees. These
bandits that live in the forest are known to be responsible for many atrocities
including rape, murder, and theft. The issue of armed bandits is not a phenomenon
that is unique to Kanembwa. According to Crisp, the refugee camps in Kenya
such as those in the Dadaab area also experience armed bandits that rape women
around the camps.6 The forest in Kibondo however also allow refugees to move
around the area of Kibondo without Tanzanian government permits because of
lack of surveillance. Therefore there is constant movement of items and of people
from the camps into the forest and into other areas of the town, and in the region,
including Burundi. In addition, the camps are located fairly close to the border as
well. Therefore it is evident as to why the location of the camp is indicative of the
level of militarization. Kanembwa camp is located in an unmonitored forest where
individuals such as rebel militias are able to move unmonitored in and out of the
5 Jeff Crisp, “A State of Insecurity: The Political Economy of Violence in Kenya’s Refugee Camps” African Affairs 99 (2000), 618 interview conducted by Crisp. 6 Jeff Crisp, 606.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 144
camps. Weapons are easily trafficked in such settings, and refugees are able to be
recruited and can easily leave the camps to fight.
The forest is also where women go in order to get firewood. With various
unmonitored activities taking place in the forest, it can be very dangerous for a
woman to get firewood. One of my informants stated that:
Being in a camp is dangerous for a woman. When you look for firewood in the forest or when you walk in the night there is danger. Other refugees rape women in the forest. There are also dangerous people who live in the forest. The refugees who live in the camps sometimes go into the forest. Through the forest they get arms from Burundi.7
This informant explicitly explains the level of arms trafficking that takes place
within the forest, from Burundi and into the camps. She indicates that refugees
from the camp are directly involved in arms trafficking, and in the midst of all this
activity, women face a degree of insecurity. In Kanembwa camp there seems to be
a mystery as to who the bandits in the forest are. Many rebels are known to have
worked in the camp to recruit fighters, and therefore some suspect the bandits to
be ex-combatants or actual combatants working in the forest. Regardless of whom
the bandits are, their presence reveals a level of militarization. It must be noted
that there is a difference between militarization and criminality, however in this
context the criminal activities of bandits are attached to militarism because of the
prevalence of arms and arms trafficking that occurs in the region. Therefore, the
bandits are not mere criminals; they are criminals that are attached to the web of
militarism that characterizes the area. Furthermore, their activities are
7 Refugee woman from Burundi, Interviewed by author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 145
representative and a reminder of the remnants of militarism and armed conflict to
everyone in the community especially refugees. The bandits are individuals that
use arms in their daily activities, and the fact that they are able to use small arms
is indicative of the level of trafficking of arms in the area. The NGO workers that
I spoke with in the area have taken the approach of stating that the armed groups
are local criminals. As indicated by Mogire, the Tanzanian government and
UNHCR have denied that the camps are militarized.8 Notwithstanding these
claims by UNHCR and the Tanzanian government, the idea and the phenomenon
of Burundian rebels working in the refugee camps and training on Tanzanian
grounds is not new. Based upon the accounts by refugees, the camps are
militarized, and the location of the camp facilitates the camp militarization.
Reminder: Political and Military Activities
The second element that makes Kanembwa camp prone to militarization is due to
the political and military activities that have taken place in and around the camp.
These activities reveal a high level of militarization of the refugee camps.
However, they also serve as elements of refugee militarization. These activities
serve as a constant reminder to the refugees as to why they are in the camp,
especially since being in the camp is attached to militarism and conflict. In
addition the events of the conflicts remain perpetually in the memories of the
refugees. Therefore, on one hand political and military activities reveal the
militarization of the camp, and on the other hand they reveal an intangible aspect
8 Mogire, 143.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 146
of refugee militarization, since they present a constant reminder of conflict and
militarization. Refugees do not have to be completely involved in the political and
military activities to be militarized, although many are involved, however, by
being affected by these activities through constant reminder of militarism, this
signifies an intangible sense of militarization. Being in the refugee camp in my
view can serve as a way to forget the past or it can be a perpetual reminder of the
events in the conflict. As Malkki states concerning the refugees in Mishamo in
Tanzania, “history had seized center stage in everyday thought and social action
in the camp.”9 As regards to the Kanembwa camp, there seems to be a mixture of
the two elements: reminder and a way to forget. The fact that the camp is a
militarized setting due to its symbolic attribute as an environment created through
militaristic activities and conflict makes it one that is a constant reminder of
conflict even though it is was set up as a place of refuge from conflict. Elements
that make the refugee camp a constant reminder of the conflict and militarism
include political activities that are carried out in the camp. Politicians and rebel
leaders are known to come into the refugee camp to “preach” their ideology. As
indicated by the Tanzanian Ministry of Home Affairs, refugees have often held
secret political meetings and some even smuggle arms into the camps. In 2006
there was an exchange of firearms in each of the refugee camps in Kibondo. The
violators were found and taken to Mwisa camp, a separation facility that has a
number of refugees that have been caught with arms.10
9 Malkki, 53. 10 Liaison Office for Tanzanian Ministry of Home Affairs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 147
In 2006 there were incidences of recruitment in Nduta refugee camp, a
camp in Kibondo. The recruiters were arrested by the Tanzanian government and
removed from the camp.11 The location of the refugee camp in the forest as
mentioned above is also another element that makes it a constant reminder for the
refugees as to why they are there and of the conflict. The problem of bandits,
although a mystery in many ways, is and has always been a reminder of the
conflict and the consequences of conflict. The forest allows for the transfer of
arms into and out of the camp and this leads to many incidents in the camp of
exchanges of firearms that can serve as reminders to the refugees of the
militarization that led to their encampment.
The problem that arises from these activities is that they not only disrupt
the lives of refugees in the camp, but they also disrupt the lives of those who are
local to the area. The activities that they carry out in the local areas have made
them very suspicious groups of people whose agendas exceed those of criminals,
but reflect those of combatants. As noted by an official from the Ministry of
Home Affairs:
Most of the bandits are not refugees even though they are Burundians. They have large fire arms with hand grenades. It is impossible for civilian bandits to possess these sorts of weapons. They must be militants and rebels. These are criminals that steal goats with grenades.12
The presence of arms in the camp is a clear reminder to the refugees as to why
they are in the camp. In addition, the Tanzanian government has played a role in
reminding the refugees that they are not home. The Tanzanian government is
11 Liaison Office for Tanzanian Ministry of Home Affairs. 12 Liaison Office for Tanzanian Ministry of Home Affairs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 148
known to tell the refugees every now and then that they should repatriate back to
Burundi. Some refugees feel that this sort of reminder that they are not “home”
reminds them that they are in a refugee camp. This has also fueled an obsession
among refugees, much like refugees in other places, for the possibility of
resettlement to Western countries especially the United States, Canada and
Australia. The desire to leave the refugee community and the fear of returning to
Burundi is front and center in the minds of refugees in Kanembwa. When the
Rwandan genocide came to an end and peace was declared the Rwandan refugees
in Tanzania were essentially forced to return.13 The fears that the Burundian
refugees have, now that there is “peace” in Burundi, are that they will go through
what the Rwandans went through. What has ended up happening in the camp is
that there are two different solutions for what to do with refugees proposed by the
host government and the international community, which are working against
each other. My observations are that the refugees are not well informed about the
procedures and the nature of resettlement. This has given many refugees false
hope because according to the UNHCR there is a list of individuals that will be
resettled and everyone is supposed to know who is on that list. Those who are not
on the list will not be resettled . If at all this is truly the case, then my observations
from the refugees I spoke with indicate that this is not clear because the refugees I
spoke with would rather be resettled than be repatriated back to Burundi. This
conflict between repatriation and resettlement is a clear manifestation of what
happens when there is a constant reminder of conflict and also it reveals to what
13 The Lawyers Committee for Human Rights, 29.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 149
extent the refugee’s connection to their home country has been disrupted. I
contend that the disruption in their connection to their homes is based upon their
symbolic attributes as militarized individuals since they live in refugee camps. I
argue that if the refugees were not constantly reminded of the conflict, and if they
were not symbolically or tangibly associated with militarism, repatriation would
be an option that would be more easily accepted. There of course is the view that
refugees merely want the opportunity to end up in the more prosperous pastures
of the Western world. It could also be a combination of both factors. The notions
of repatriation and resettlement evoke reminders for refugees as to why they left
their homes. The process of resettlement requires refugees to explain the fears
they have of returning to their home countries. Generally the fears expressed by
refugees, and the reason they seek resettlement are associated with militarism in
Kanembwa camp. The fears of repatriation are also tied to fears of violence
because some refugees realize that by returning home they might not be accepted
in their community because of their links to militarism while in the camp.
Therefore they might face rejection from their community, and potentially
violence. Therefore, what we notice is a complex web that reveals connections
between such notions as repatriation and resettlement and militarization.
Lifestyle. Livelihood and Violence
The third factor, lifestyle and livelihood of refugees, presents reasons that could
lead to frustration in the refugee community and encourage refugees to take up
arms or support militia groups. The restrictions of refugee life can lead to a sense
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 150
of hopelessness that can give individuals reasons to take up militarism. For
starters the refugees live a controlled life in a camp that is somewhat monitored
by external actors. The lack of sufficient activity for self-promotion from such
things as lucrative business ventures and job opportunities adds to the culture of
the refugee camp and fuels the experiences that refugees have in the camp. For
starters the refugees survive by receiving distributions from the United Nations
High Commission for Refugees (UNHCR) and other organizations working in the
camp such as International Rescue Committee (IRC) which provides a hospital
and a Sex Gender Based Violence Centre, and the Tanzanian Christian Refugee
Service (TCRS). The fact that the majority of refugees have virtually no control
over their lives and depend on the UNHCR and other organizations for food
rations every two weeks and soap once a month adds to the kinds of experiences
that refugees have in the camp and how they view themselves and the culture they
create. For some refugees there is a definite and inevitable sense of relative
deprivation. They analyze their lives as lacking in basic needs and in poverty,
while other refugees are content because their lives prior to the refugee camp
might have been in deeper poverty.
A combination of the history of the conflict, the location of the camp, the
constant reminders as to why the refugees are in the camp and the minimum that
they have can be a reason to fuel various acts that manifest dissatisfaction.
Violence and animosity are prevalent in the camp and are not always manifested
by acts of violence via firearms and other forms of physical violence. There is
inevitably a high level of human insecurity in the camp that is manifested in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 151
interactions that refugees have with each other. As noted by an official from the
Ministry of Home Affairs:
The refugees have a habit of trying to avenge each other in the camp. Women have witchcraft problems. Three women in Kanembwa were accused of witchcraft. Last year a woman was killed by other refugees for witchcraft. The government of Tanzania does not recognize witchcraft. When such information is neglected it might lead to the death of someone. There have been many cases of witchcraft.14
The level of violence and the level of distrust among refugees reveal the
consequences of militarization. This is because the refugee camp presents
elements of violence, is prone to interference from outsiders, and it can be argued
that this camp has not achieved a level of stability from the violence the refugees
fled. What we notice is that the violence and aggressive behaviors are a
demonstration of the instability of the camp that has been brought on by the link
the camp has to militarism and violence. It would be myopic to simply look at
overt manifestations of militarization through use of firearms and such, what we
find however is that militarization has seeped into the daily lives of individuals
and what we find is a complete culture of violence. As stated by a group of
refugee women:
Some refugees like to accuse other refugees of crimes based on hatred. During distribution one person was accused of killing someone after a skeleton was found. This accusation was not true.15
The lifestyle of refugees not only has been marked by violence, also the instability
of the camp presents serious health implications. Like any other setting the
refugee camp has been hit by HIV/AIDS. The prevalence of the disease in the
14 Liaison Officer for Tanzanian Ministry of Home Affairs. 15 Refugee women focus group meeting in Kanembwa Camp, Interviewed by Author, January 23, 2007 Kanembwa Refugee Camp, Kibondo, Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 152
camps is high and the refugees have taken the initiative to try and work to combat
the disease. HIV is seen as a result of war. A group of women informed me that
there is usually nothing to do in the camp and sometimes people resort to sexual
activity.16 The fact that people have nothing to do in the camp is a huge factor that
creates concern about the lifestyle of those in the camp. Refugees are confined in
the refugee camps where they have little or nothing to do. Their encampment is
due to Tanzanian refugee law that designates areas and disallows refugees from
moving outside the parameters of the camp without obtaining government
permission.17 With restricted movement, refugees are unable to find jobs or other
ways beyond the camp to make a living. Some refugees however have found ways
of not living only off of the UNHCR distribution and have taken it upon
themselves to find ways to make extra money in order to buy more food for the
family. One of the ways many people have been able to make some money is by
brewing local beer that is sold at the market place. What one leams is that many
men have virtually nothing to do in the camp and therefore find themselves
trapped in a habit of drinking which can often lead to domestic violence in the
home. One of my informants stated that:
The Burundians are used to drinking. Men have nothing to do so they sell the rations of food and start to drink. Women usually can’t find enough food to feed their children. Women try to tell men not to sell the food but this brings conflict in the house and leads to domestic violence. Some women do not obey their husbands. There are women who tell their husbands, “you don’t bring anything to eat, it is UNHCR that feeds us.” This also brings violence and divorce.18
16 Refugee Women Focus Group Meeting in Kanembwa Camp. 17 Khoti Kamanga, 110. 18 Refugee woman in Kanembwa Camp, Interviewed by Author, January 18, 2007 Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 153
Another informant indicated that, “there was no domestic violence in Burundi.
Men had things to do in Burundi. Men have nothing to do in the camps but drink
and so they beat their wives.”19 The women who face domestic violence often
refer to alcohol as a leading factor in their husband’s aggressive behavior. One
woman who is a victim of domestic violence expressed her story by stating that:
The biggest problem is that he beats me. In Burundi he was not a drunk. He beats when he is drunk and when he is not drunk. He wakes up in the morning and he is mean. When he beats me he says I am too old, and that I must be younger. He told me that if I don’t leave he will kill me. In November 20061 found my husband at the market with a stick getting ready to beat and kill me. I try not to go to the market very often. He has beaten me before with his head. I tried to tell other women and other women have told me to leave him and if possible to leave the camp.20
Another woman stated:
I have problems with my husband. He drinks all the money. I do everything for the children. Sometimes he beats me and when he is drunk he really beats me. I have not gone to SGBV center. I went to the block leaders but they didn’t help. My husband takes ration food and sells it for money to drink. I keep food at the neighbor’s. In the HTV/AIDS-Good Parenting Seminar, the other refugees asked my husband to come to the seminar but he refused. In this seminar parents come to help each other with problems and give each other advice. I do not want to separate from him. I can only accept him and only hide the food. I have not talked to women with the same problem. Women without this problem offer me support. Before the war he was a drunkard. In the camp he does nothing but drink. I have started out by talking to friends and maybe later on I will go to SGBV.21
The violence that is faced by women in the camp is a reflection of the instability
of the camp. Being in a camp that was created from militaristic ventures and that
19 Refugee woman in Kanembwa Camp, Interviewed by Author, January 18, 2007 Kanembwa Refugee Camp in Kibondo Tanzania. 20 Refugee woman in Kanembwa Camp, Interviewed by Author January 22, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 21 Refugee woman in Kanembwa Camp, Interviewed by Author, January 20,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 154
has militaristic activities in and around it creates a culture of violence. Violence,
especially gender based violence, becomes common in such situations, and to
some degree the level of militarization makes interpersonal violence appear
“normal.” Violence, as indicated by Muggah, is one of the elements that points to
militarization of a refugee camp. I argue that violence does not have to be
perpetrated with the use of firearms to be considered a sign of militarization.
Militarization is also an issue of interpersonal violence that occurs especially in
private spaces where the frustrations of being in the camp are manifested. The
situation of the refugee woman above, although not evident, manifests a degree of
violence in the refugee camp and instability of the camp as an environment that is
characterized by uncertainty and frustrations.
Furthermore, even though men are the ones who seem to have serious
frustrations and problems with alcoholism especially relating to their feeling of
emasculation, refugee women also have problems of fmstration. As indicated by
refugee women during the focus group meeting:
In Burundi women are not drunk; in the camp they are drunk. Women start from morning to look for local brew and go to the market to look for more. Women drink because they are traumatized.22
The trauma that is associated with being in the camp has led to a myriad of
behavioral changes. The lifestyle of the refugees is significantly influenced by
trauma and other socio-psychological problems that come with living in a
superficial environment that was created as a result of the atrociousness of
22 Refugee women focus group meeting in Kanembwa Camp.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 155
militarism, and an environment that presents a continuity of militaristic activities
and systemic violence.
Role of External Actors
The fourth factor is the role of external actors. External actors have been
responsible for presenting new behaviors and values in the camp that can easily
be translated into reasons for some women to take up militarism. As demonstrated
in the historical analysis in chapter four, the Burundian culture was very
ethnicized and militarized, and this become evident even in the social sphere.
However, women, especially Hutu women were restricted from overt
participation because of the patriarchal control. What is learned in the camp is that
women’s participation in militarization was condoned by their parents, but in
many instances it was a result of the woman’s own initiative. It is the premise of
this study that the conditions in the refugee camp altered the patriarchal culture
just enough for women to be involved with militia groups, and this alteration can
be seen partly as a result of the influence from external actors such as NGOs, and
UNHCR. Therefore when we ask the question: what leads to the unconventional
behavior of Burundian refugee women’s support for rebel militias especially
when one takes into account the patriarchal society in Burundi? The answer to
this is that there has been a slight change in Burundian patriarchal culture which
permits women to take up militarized roles.
As noted earlier, the role of external actors such as UNHCR and other
organizations also influences and adds to the culture that is created by refugees in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 156
Kanembwa camp. What is of particular interest is the sense of empowerment that
the organizations provide. From my observation, when refugee women referred to
the term empowerment, the usually meant enlightenment and the fact that they
had gained knowledge of their rights. However another aspect of their
empowerment is that they do not have to rely on the men in their community to
survive. Women now rely on UNHCR and other organizations for food and for
their well-being while in the camp. One refugee women indicated in the focus
group that since UNHCR provides food and shelter some refugee women even tell
their husband such things as, “UNHCR is my husband because UNHCR brings
rations.”23 The women in the focus group expressed a dilemma that they felt was
taking place in the camp. On one hand, the international organizations have
helped the refugees, but on the other hand they have brought behavioral changes.
A detailed discussion of the implications of the external actors’ role of
empowerment will be provided in the following section. The refugees are
educated through activities such as seminars especially HIV/AIDS seminars. This
is not to say that the refugees themselves are incapable of starting their own
initiatives for empowerment. In fact one of the leading groups STOP SIDA
(STOP AIDS) was founded by a Burundian refugee. Empowerment is also seen
through women who have been able to go to the SGBV center to get assistance
when facing domestic violence and other forms of sexual violence especially rape.
Women are educated about their rights through this center. Some informants in
the camp have gone as far as saying that women have more rights in the refugee
23 Refugee women focus group meeting in Kanembwa Camp.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 157
camps than in Burundi, and this is mainly due to the fact that the refugee camps
offer more seminars, and education for women than in Burundi.24
The IRC SGBV center tries to teach women about their rights but also to
change various mentalities and old traditions and cultural behavior that in their
view add to violence against women. For example the center tries to encourage
men to help women with various chores especially when it comes to collecting
firewood because of the abuse women face from bandits in the forest when they
collect firewood. However this solution has met some resistance because as one
informant put it “women collect firewood because men can’t do such a job.”25
There is a prevalence of rape in the camp and that is an issue at the heart
of the SGBV center. Many young girls are raped in the camp for example the case
of one informant:
One day I went to get water. I went alone. I met a boy on the way. He was with two other boys. The three of them took my water and took me to their home. The boys were about 30 years old. I didn’t know who they were. They locked me up for 2 days and they raped me. After they let me go I reported to SGBV, and SGBV put in a report to the police. The boy was arrested. The doctor helped me; I was given treatment to prevent pregnancy and HIV/AIDS infection. My mother is the one who brought me to SGBV. My father told me not to return home and he said that I should go back to my husband, the boy who had raped me. The rape made me married to the boy in the eyes of my father. My father does not beat me but he tells me to leave.26
The context of the camp manifests the various elements that characterize the camp
as one that is militarized, and also one that maintains elements of militarization.
24 Refugee woman in Kanembwa camp, interviewed by author, January 15,2007, Kanembwa Refugee Camp Kibondo Tanzania. 25 Refugee informant in Kanembwa camp, interviewed by author, January 16,2007, Kanembwa Refugee Camp Kibondo Tanzania. 26 Refugee woman in Kanembwa, interviewed by author, January 21,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 158
The location of the camp makes it susceptible to military activities, the political
and military activities that have generally been made possible due to the location
of the camp present a direct link to militarization. The lifestyle of refugees is
linked to violence as a result of militarization and finally the influences of
external actors also present changes in behavior and a challenge to patriarchal
culture. These influences are at the heart of why refugee women are militarized.
History and Culture: Conceptual Elements that Explain Kanembwa Camp’s Militarization
The history of the conflict and aspects of culture are indisputably leading
factors of militarization of refugee women in the camp of Kanembwa. As noted
earlier, the refugee camp is in itself a constant reminder of the past history. What
is noticed here is that the history of the conflict floats around in the camp and
influences the instances of animosity or trauma that one may face. As argued
earlier, the history of the conflict and the militarization in and around the camp
makes violence appear normal and almost inevitable in the refugee camp.
Therefore, although there are many motivating factors for instances such as rape,
it can be surmised that violence and rape have been facilitated in the camp
because militarization of the camp makes such situations inevitable and normal.
The paradox is that a refugee camp can help individuals feel far from the
problems they faced in the home country. However, I find that this is not always
the case in Kanembwa. The main element that influences my view that the history
of the conflict is alive in Kanembwa is the refugees’ memory. The memory of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 159
conflict is especially significant due to the many elements about the refugee camp
that serve as reminders of the conflict. In addition, the memory is hard to erase.
Being in a refugee camp and away from where massacres occurred does not mean
that you will forget what happened in the past. Another factor that is presented in
the camp due to the history of the conflict is ethnic animosity. It appears as
though ethnic animosity remains in the camp and can lead not only to a reminder
of the history of the conflict, but can also serve as a reason for participation with
rebel militias. This is especially the case since the ethnic make up of Kanembwa
camp is majority Hutu.
Although there are a few Tutsis in the camp; the camps hosting Burundian
refugees in Tanzania generally host Hutus because of the nature of the conflict:
Hutus were the ones fleeing from the Tutsi army.27 The other factor that allows
for militarization is the change in culture in the camp. In this instance I analyze
how Burundian culture and the created culture in the camp have taken a certain
metamorphosis. This transformation of culture based upon various factors can
easily be said to be part of the influencing element that mold the militarized
culture of Kanembwa camp but also it is one of the explanatory factors as to why
refugee women have participated in activities with armed forces.
When the two elements are put together one can see the nature of
militarization that refugee women in Kanembwa face. In view of history and
memory of history, many refugees are able to recall the events of the conflict with
27 The camps hosting Rwandans on the hand were hosting Tutsis because the Tutsis were the main victims in the Rwandan genocide.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 160
such precise detail. There is also an inevitable element of trauma associated with
many refugees especially when recounting their stories. The history of ethnic
animosity and ethnic delineation is at the heart of the refugees’ understanding of
the conflict. As one refugee explained her history:
When war started in 1993 it was announced on the radio that Ndadaye was murdered and all the Tutsi ran towards the army that would protect them. All the Hutu fled to the villages because there was no military or army to protect them. They started to walk in the forest on foot and hiding in villages. Tutsis were killed by Hutu in my region. They found no Tutsi that was alive.28
The fact that the camp is predominately Hutu makes it also hard to forget. There
is a sense of superficial harmony in the camp because the majority of inhabitants
are Hutu. What comes to mind however is what this somewhat “homogenous”
make up of people does to people’s understanding of ethnic animosity, the very
reason that sparked the conflict. The answer to that can be found when one talks
to any one of the few Tutsis that reside in the camp. It is interesting to note that
the reasons why there are any Tutsi at all in a predominately Hutu camp is due to
the fact that some Tutsis were threatened for being married to Hutus, in addition
some Tutsis found themselves guilty by association for merely living in the same
areas with Hutus. One Tutsi explained:
When I arrived in the camp I was threatened and the threats to kill me were so many because of my ethnicity. Tutsi women that were married to Hutus were trying to flee and were killed at the border. When the camp was created the animosity between Hutus and Tutsi was so deep that there were many killings. One day the Tanzanian authorities came to the camp and told everyone to put their ethnic differences aside because we were
28 Refugee woman in Kanembwa Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp, Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 161
now in the camp as refugees and not Hutu or Tutsi. Our new identity became refugee and not Hutu or Tutsi.29
The history of the conflict is so intertwined with the complexities of ethnicity.
Would there have been such a conflict in Burundi if ethnicity was not a factor?
The answer to that question is hard to come by because of how complex ethnicity
is as reason for conflict and a source of power in the specific history of Burundi
(and Rwanda and many other post-colonial African national contexts). Many
Hutus and Tutsis living in the camp are able to voice their thoughts about
ethnicity but the stories that are told of constant animosity confirm the premise
that this refugee camp, being quite homogenous in terms of ethnicity does not
help the situation. The memories of history live in the minds of all the refugees
despite the fact that they are in a refugee camp in Kibondo and no longer in
Burundi. One refugee explained to me:
We lived very badly with Tutsis. We fled in 1993. At school children that were Hutu were mistreated by Tutsi. My family suffered in 1972. My uncle was killed. No one in my family would ever marry a Tutsi. Many in my village did not want anything to do with Tutsis. We didn’t even keep cattle together. The Tutsis despise the Hutu and so Hutus despise the Tutsi.30
With such a history it is hard to imagine that being in a refugee camp with other
Hutus can help heal or forget the animosity of hatred. As Malkki argues, instead it
creates group identity and cohesion.31 I argue that this is why recruiting from the
29 Refugee woman in Kanembwa Camp, Interviewed by Author, January 16, 2007, Kanembwa Refugee Camp, Kibondo, Tanzania. 30 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 24, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 31 This view is similar to the one raised by Malkki who states that the Hutu refugees in Mishamo camp in Tanzania were creating a collective identity and this was done through their “construction and reconstruction of their history as a people” p. 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 162
refugee camps by the rebel groups was possible. When rebel militias entered the
camp to recruit, or when refugees within the camp were recruiting refugees to join
militia groups, many individuals joined the armed forces. My premise is that
living in close proximity to each other, even though they come from different
regions of Burundi, they learn that they share the same grievances. The Tutsi
army was the cause of their grievances and therefore anyone Tutsi is part of the
Tutsi cohort. The animosity has truly been brought in to the camp and many
women that are Tutsi face problems. Ethnicity in Burundi is not only a concept
associated with power. The conflict has also militarized the very idea of ethnicity.
To be Tutsi means to be aligned with the military and to be Hutu means to be a
rebel militia. This is obviously extended to every member of society. Tutsi
women were treated with animosity in the camp when the camp was created, and
many women remain ostracized and mistreated today in the camp because of
ethnicity. One Tutsi informant indicated her story:
I faced problems with others who were Hutu. I left my block because of these problems. One person helped me to get away from the block. I was transferred from that block. Life was so hard that I wanted to kill myself. My house was burned three times. There were some people that tried to help but they had the same problem. My children are treated badly for being Tutsi. 2
The ethnic animosity is not only directed toward women in the camp. There is
also a case of a Hutu man who is hated in the camp because he came from a Tutsi
dominated area known for some of the worst massacres of Hutus.33 The
32 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 22,2007, Kanembwa Refugee Camp Kibondo Tanzania. 33 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 24, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 163
complexity of ethnicity continues to transcend various social institutions such as
marriage. During the war many spouses were forced or voluntarily killed their
spouses that were of a different ethnicity. This phenomenon is also manifest in the
camp. In the camp however, spouses are not killing each other. There is a case of
a woman who married in the camp and is now being threatened by her husband
because her husband claims that she never told him her true ethnicity before
getting married. And since she is Tutsi he wants to have a divorce and even kill
her.
It seems that fourteen years in a refugee camp is not long enough to forget
why the Kanembwa refugees are in the camp. The fact remains that the history of
the conflict follows the refugees because it relies on their identities as a specific
ethnic group. How can that be taken away from them in the camp, especially one
that is nearly homogenous? The homogeneity of the camps is deliberate. In
Kibondo, the three big refugee camps Nduta, Mtendeli and Kibondo are majority
Hutu, then there is Mukugwa camp which is a mixed camp with people from
different parts of Central Africa but also it has the Tutsi population of Burundi.
One reason for the homogeneity is due to the nature of the conflict, but also
because it is a way for the International Community in charge of the camps to
avoid bloodshed.
There is an interaction of different cultures in the camp. As stated earlier,
my premise is that, following Avruch, individuals can have more than one culture
and that cultures are a byproduct of experiences. What is of particular interest in
Kanembwa is that there seems to be an ongoing metamorphosis of cultures. There
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 164
are three activities relating to culture that I observed are occurring in Kanembwa
camp; there is a loss of some of the original patriarchal culture and tradition due
to conflict, there is a clear manifestation of some of the questionable elements of
the original culture, and there is also a sense of challenge to the original
patriarchal culture due to an emerging new culture characterized by refugee
women empowerment. The original culture in this context is the patriarchal
Burundian culture.
The created culture in the refugee camp that fuels militarization hinges on
the metamorphosis of the original culture. The Burundian culture exists and will
always exist for the Burundian refugees. Fundamentally they adhere to the various
aspects of their culture through behaviors, and traditions that make them
Burundian like other Burundians in Burundi; however there are some elements of
their original patriarchal culture that have been transformed and this allows
existence of a new culture. This does not in any way mean that the Burundian
culture has been eliminated and that a Burundian refugee no longer holds
Burundian culture. However, as noted by a group of refugee women, this is
precisely what Burundians in Burundi believe.34 Many refugees indicated that
many Burundians in Burundi are distrustful of Burundians from the refugee camp
because they have supposedly lost their culture. However, it is not a matter of loss
of culture, as much as it is adopting a second or third culture and the loss of some
aspects, and not all aspects of the original culture. The focus in this study is the
change in patriarchal culture, amd an analysis of the way the so called
34 Refugee women focus group meeting in Kanembwa Camp.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 165
empowerment of women can explain refugee women’s militarization. There are
various behavioral changes that have been noted by refugees relating to their
change in culture. The interaction between men and women has been going
through many changes. Some women are breaking from patriarchal restrictions
and are able to disobey their husbands. Fathers do not abide to cultural norms in
place to protect their daughters, and husbands and wives are ready to divorce each
other, even though in Burundi divorce is frowned upon. One informant explained
the changes in her relationship with her father:
My father tells me to show him my boyfriends and he enters my room. In Burundian culture the father is not allowed in his daughter’s room. He has a drinking problem since Burundi. He has tried to chase me out of the house. I used to pack my luggage and try to leave but people always told me to stay. I tried to kill myself a few times.35
The change in culture in the refugee camp has led to the destruction of social and
familial ties. Some refugees noted that many young girls fled the refugee camp to
join rebel groups, and it can be seen as to why this would happen because of the
loss of familial ties and traditions that keep girls at home and under patriarchal
protection. As a former combatant explained, many girls that had joined the rebel
militias did so because they were running away from family problems.36
Polygamy and divorce are frowned upon in Bumndian culture; however it
appears that the culture in the refugee camp permits such behavior. In Burundi
polygamists are considered immoral and can not hold positions in the society as
35 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 21, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 36 Refugee informant in Kanembwa Refugee Camp, Interviewed by Author, January 17, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 166
umushingantahe or moral leaders. This was explained by refugee women in a
focus group:
Polygamy is a big problem but it is not Burundian in origin...And the other problem is that people have nothing else to do other than have sex. In Burundi a polygamist can not be considered umushingantahe,37
The remnants of conflict can be seen in the refugee camp as individuals take up
different behaviors. The change in culture, particularly patriarchal culture is at the
heart of women’s militarization. The complex web of activities in the camp
reveals that a metamorphosis of culture is taking place.
The Conflicting Cultures of Kanembwa Refugee Camp
I contend that culture is highly malleable and has attributes that reveal
fluidity. The patriarchal Burundian culture in Kanembwa camp is being
transformed. On one hand the change is facing resistance. Hutu women are slowly
moving from patriarchal control; however what one notices is that the men in
their lives are resisting this change. The slow move from the grips of patriarchy
explains the number of women who took up arms to fight with rebel militias.
However the fact that there were only 490 women38 who have been targeted as
fighters in the Burundian conflict can be explained by the resistance of patriarchy
for women to take overt militaristic actions. Notwithstanding this point, the
transformation of culture in the refugee camp has implications on the level of
militarization especially of refugee women. In Burundi Hutu women were
37 Refugee women focus group meeting in Kanembwa Camp. 38 Waldemar Very, Interviewed by author by email.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 167
completely controlled by their husbands, fathers and brothers. In the camp the
patriarchal culture is slowly losing its hold on women. Therefore the elements of
the Burundian culture that are being compromised that I observed involved those
relating to family ties and gender roles and relations. There could very well be
other aspects that are changing about the Burundian culture, but my interest in
these areas, considering the nature of my study, caught my attention. Many
refugee women that I spoke with especially during the focus group that I held
indicated that Burundians in the camp are “losing” their family ties and traditions
because many of their family members were killed in the conflict, and living in
the camp creates a sense of disconnection with their lives in Burundi. The bonds
that keep family together have been lost because many people have lost their
family members. Women indicated that when wives and husbands are having
problems they can easily cope with these problems in Burundi because of a family
support system. However, due to the conflict and deaths of many family members
and being displaced into a camp and separated from family members, the family
support system no longer exists. Many women also indicated that many husbands
and wives remarried because of losing their wives and husbands and therefore
there was also a question of bringing together two families which was always a
problem considering the context in which the two families would be brought
together. One male refugee informant expressed some of the family problems that
he faced:
When we arrived in the camp, my mother died. My father soon remarried. We lived without much food to eat. I soon decided to leave because my father mistreated me because of his new wife and I didn’t feel like I had
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 168
much of a future. I joined the CNDD-FDD. I fought for 2 years. I joined when I was 17 years old.39
Conflict, the situation in the camp, and the experiences of refugees have been a
main reason for the springing up of new cultures and habits and the abandonment
of some elements of the original culture because there are no systems that can
keep those elements alive.
Exacerbation of Culture and Tradition
Another aspect of the culture is that there seems to be an exacerbation of some of
the “questionable” elements of the original culture. By questionable elements, I
use this term cautiously to refer to the elements of culture that lead to the abuse of
certain individuals, notably women. Burundian culture like many cultures is
highly patriarchal. Patriarchal culture means that women take a submissive role in
society. The submissiveness and role of women becomes very questionable in
times of conflict and in refugee settings because many of the systems that might
have been in place in “normal” and non-conflict circumstances that protected
women from abuse are often lost. In the camp, women complain that they are
doing too much work and much more work than men. The SGBV center in
Kanembwa indicated that women were doing all the work and were often
compromising their own security. For example women are the ones who collect
firewood in the forest and in the forest they can get raped by bandits. Men tend to
refuse to go along with them because collecting firewood is net a man’s job.
39 Refugee informant in Kanembwa Refugee Camp, Interviewed by Author, January 17,2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 169
There is also a tendency for men to leave their wives at home and to take various
partners. This exacerbation of patriarchal culture is very much connected with the
changes in cultures because as women lose their family ties that are inherent in
the original Burundian culture, they become susceptible to some abuses.
The systems that keep control of patriarchy and women’s submissive role,
and ensure that they are not elements leading to complete abuse are taken away
because of conflict, because of loss of family members, and because of living in a
created environment that is restrictive and is a byproduct of animosity.
Furthermore, there is the issue that men do not have a lot of activity in the refugee
camp. Men are bored in the camp. They have little to do, little to no opportunities
in the camp and can not take care of their families and therefore they feel
emasculated. Some refugee women stated that the men try to exert their
masculinity through the abuse of women. The emasculation of refugee men is
another attributes that leads to the loss of patriarchal elements and control over
women. Therefore in order to regain their control, they abuse women. This is a
common phenomenon that has been addressed by various authors. One study by
Gary Barker and Christine Ricardo (2005) elaborates on the emasculation of men
in Africa, especially those that have experienced conflict.
Challenge of Culture and Tradition
With the absence of traditional and cultural mechanisms such as family ties to
protect women, women have to find ways to protect themselves, and sometimes
NGOs find ways for women to protect themselves. The presence of NGOs is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 170
heralded by many but taking a different lens reveals that NGOs and their various
programs present a challenge to the patriarchal Burundian culture. Activities by
NGOs have actually added to the loss of patriarchal culture and have contributed
to the creation of a new culture that has led to the education of refugee women on
their rights. As one refugee woman pointed out to me, it is the emancipated and
educated women that directly participate in militarization and the “peasant” and
uneducated women are the ones who are afraid of the militia groups. This is also
one of the main points raised in literature as presented in chapter three that
explains the type of women that engage in militarism. Some women, although not
a significant number, are known to go to the IRC SGBV center and other
organizations for assistance when raped, and when facing domestic violence.
Women have also become empowered by the various seminars that are presented
to them by the organizations that coordinate the refugee camps. The SGBV
Center has a chart every year indicating the number of sexual based violence
experienced in the camp. There are of course cases of men being abused and the
abuse of young boys, however these cases are hardly reported and therefore the
figures of abuses are overwhelmingly of women.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 171
Table 1. International Rescue Committee Sexual and Gender Based Violence Centre 2006 Chart on the number of cases per month of each form of SGBV abuse in Kanembwa camp out of the 12,500 refugees that live in Kanembwa camp.40
Month Rape Attempted Sexual Forced Early Domestic Other Total
Rape Harassment Marriage Marriage Violence
Jan 0 0 1 0 0 12 3 16
Feb 0 0 0 0 1 9 6 16
Mar 1 1 0 0 0 10 8 20
April 1 0 0 0 0 7 4 12
May 1 0 0 0 1 8 7 17
June 0 0 0 0 0 13 4 17
July 3 0 0 0 0 11 6 20
Aug 1 0 0 0 0 12 3 16
Sept 1 0 0 0 0 13 6 20
Oct 1 0 0 0 0 5 4 10
Nov 2 0 0 0 0 6 6 14
Dec
A woman who had been to the SGBV center explained to me the problems that
she faced and why she ended up going to the SGBV center:
During the 1993 war, I faced violence from my husband. My husband insisted on looking for partners. My husband is a Tutsi and I don’t really know for sure if he is really a Tutsi. I am a Hutu. He told me that he would rape me for having a different ethnicity. In the camp I am a cultivator. He tells me that I am old and that he will find a new wife. He beats me. He
40 Results for December were not yet calculated when I looked at the chart on January 17, 2007.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 172
told me that he would kill me. I came to the center in 2002. The SGBV center counseled us as a couple. I came with an injury to the center and the case was taken to the police. The MHA gave us the right for separation. My husband still threatens to kill me.41
The UNHCR has programs to protect refugee women; in addition the Tanzanian
government has laws that attempt to deter such behavior of abuse. Some refugee
men have spoken out about the protection provided by international organizations.
There have been cases of men who threaten their wives because of the protection
UNHCR provides to their wives from the abuse they commit against their wives
and some have indicated that once they return to Burundi they will regain their
control42 Tanzanian law is such that any one culpable of rape will receive up to
thirty years in prison. The international organizations are challenging what can
easily be deemed as the patriarchal culture of Burundi. By the same token some
refugees have indicated that their Burundian culture is being challenged by
Tanzanian culture as well. In this regard, many women indicated that the reason
for the abuses is not necessarily because their culture permits it; it is because
Burundian men are emulating Tanzanian men who apparently are polygamous. As
one refugee put it:
In Burundi there is no polygamy. In Tanzania polygamy is permitted and so after the conflict in Burundi men in the camps want more than one wife. In Burundi polygamy is not accepted and so we find that Burundians are imitating Tanzanian culture.43
41 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 22, 2007, Kanembwa Refugee Camp Kibondo Tanzania. 42 UNHCR staff in Kibondo, Interviewed by Author, January 24, 2007, Kanembwa Refugee Camp Kibondo, Tanzania. 43 Refugee woman in Kanembwa Refugee Camp, Interviewed by Author, January 18, 2007, Kanembwa Refugee Camp Kibondo Tanzania.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 173
There are many different activities happening around camp that are affecting the
culture of the Kanembwa refugees. It is clear that the situation of being in the
camp and the history of the conflict have contributed to the change in culture. It is
in this environment that the culture of militarization of refugee women has
emerged and been created. The formulation of this culture is based upon the
women’s experiences and activities, as well as their understanding of ethnic
animosity.
By virtue of its existence and history, Kanembwa refugee camp is a
militarized environment and therefore its inhabitants are militarized. However
there is a specific context from which a specific culture of militarization is
created. This specific culture explains women’s understanding of militaristic
activities, women’s interest in military activities, women’s fears and dependency
on militaristic activities and ideals. In this analysis, based upon Cynthia Enloe’s
definition, what is of particular interest is how women are controlled by militia
groups and militaristic ideals and how they depend on these militaristic ideals.
What I have attempted is to create an understanding of how the refugee camp is
militarized by virtue of being a political camp, and also by the activities that take
place in the camp that are politically and militaristically motivated. Furthermore, I
attempt to show how prone to militarism the refugee women are through the
intangible and more complex factors such as change in cultural norms that
challenge patriarchal structures that tend to keep women from being involved in
militaristic ventures. Therefore when one examines how women can be involved
and believe in what is arguably “men’s affairs” it is clear that for starters the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 174
refugee camp itself by virtue of existence makes certain activities inevitable and
that there is also the aspect of the very existence of the refugee camp as an
environment that is constantly formulating new cultures, and some of the systems
in the original culture that are there to protect women and to maintain normalcy
and norms are broken. There are inevitably many other variables to consider in
such a phenomenological issue; however the lived experiences of the individuals
being studied present the answers.
Conclusion
The context of where militarization of refugee women takes place is very
indicative of how militarization occurs. This chapter has demonstrated the various
factors that make Kanembwa camp a militarized camp. The location of the camp
and the militaristic/political activities that take place around the camp clearly
indicate militarization. However, these are two factors are tangible explanations.
There is also the importance of understanding the social relations and cultural
changes in the camp that explain why certain behaviors take place in the camp.
By looking at the social relations and change in behavior, we bring forth elements
that put the focus on the individual. Conflicts and militaristic ventures do not only
change power structures at the state level, however they also change behaviors
and norms at the individual level. Even though the changes in culture encounter
some resistance, what we notice is that the change in the norms in the camp
partially explains why women feel the need to depend on militia groups even
though the militia groups control refugee women.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER SEVEN
CONCLUSION
Ethnic animosity in the history of Burundi has led to many atrocious
consequences in the lives of refugee women and others. The gendered dimensions
of ethnic conflict take a clear form in the Burundian conflict even though not
many people write about it. Burundi’s problems have been shadowed by
Rwanda’s genocide. However what should be understood is that the Rwandan
conflict has been fueled by events in Burundi and vice versa. Burundian
massacres have led to the destabilization of the lives of many individuals, and
furthermore, they have altered the livelihood and norms that are integral to the
lives of Burundians. The socially woven fabric of tradition and culture is slowly
coming undone and what we have is a transformation and metamorphosis of
culture and of people. In order to understand this change in behavior I had to look
into the past, as well as the present for answers.
Refugee women have endured so much in the Burundian conflict.
Furthermore, their lives in the refugee camp are connected to militaristic ideas
and violence. There is an apparent dichotomy in the nature of militarization in the
refugee camp: some women fear the military and militia groups and other women
depend on them. Although this study delineates this dichotomy, it is irrefutable
that the two sides merge together and some women both fear and depend on the
175
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 176
military. In my view, even though refugee women’s lives are coupled with
poverty and violence, they can be deemed as the strongest woman. A refugee
woman is a woman who with very little will find a way to feed her children.
Living with a history tainted by violence where half if not all her family members
have been killed in the name of ethnic identity and power. Living in a society
where men find themselves emasculated since they have very little control over
their lives, and having to face the consequences of the emasculation of their male
counterparts. Being able to look forward to a future for their children and being
able to keep hoping for a better future.
When working with refugee women, it is easy to fall in the trap of looking
at their history and past as completely unmanageable. It is easy to fall in the trap
of looking at their present as abysmal without life or light. The life of a refugee
woman is unmanageable and abysmal with little light, however somehow the
women go on. What was revealed in Kanembwa camp is a consciousness of
women who not only understand their situation and are able to interpret it to me,
but they are also able to be daring and do what is unconventional. One might say
that they have nothing to lose. I beg to differ with such an explanation and insist
that conflict creates change, and animosity breeds within us a different being.
Conflict and war, animosity and change are at the heart of ontological questions.
A refugee woman in Kanembwa can not be easily described however her behavior
and mentality is shaped by a history of ethnic animosity, a fear or dependency or
both of military and militia activities, and a loss of culture that promotes the
norms that are intricate to her life in Burundi. A refugee woman in Kanembwa is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ill
a woman who has undergone change, metamorphosis into a woman who finds
new ways to survive in a dangerous world characterized by war.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A
SCRIPT TO MEMORIZED AND RECITED TO
REFUGEE WOMEN PARTICIPANTS
My name is Barbra N. Lukunka. I am from Lusaka, Zambia. I am a Masters student at the American University School of International Service in Washington DC. I am here in Kanembwa camp for three weeks to research the experiences of refugee women that come from Burundi. I would like to talk to you about your experiences as a refugee woman in Tanzania.
I am interested in knowing how you see yourself as a woman from Burundi especially in regards to the conflict. I am interested in how you perceive the ethnic problems in Burundi especially how they affect women. I am interested in the experiences you faced while fleeing Burundi. I am interested in knowing what experiences you have had in the camp that sometimes make you feel unsafe.
The reason for my research is to help others understand the lives of refugee women. I feel that your participation will add to this understanding. After my research in Tanzania I will return to the United States and write a report about what I learned.
There are some potential risks that can arise from your participation. There is the risk that people might find out what you have told me, especially regarding issues that you do not want people to know. There is also the psychological risk which could arise while you are telling me information. It could lead to you being sad and emotional while remembering some of the events that you might recall. There is also the security risk that people that do not want you to talk about certain events might find out what you said and this might threaten your security.
In order to minimize these risks my report will not include the identities of any person that speaks to me about their experiences. I will not include your name, place of birth or any information about your identity in my report. I want you to know that I respect your privacy and so I will keep all the information that you tell me confidential. You have the choice to not participate in this research, and if you decide to participate but change your mind in the middle of the interview, just let me know and the interview will be completely stopped. If you change your mind and decide that you do not want the information that you have told me to be
178
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 179
used please let me know and I will not use any of the information you have told me in the study. I will try my very best to make sure that what you tell me is not known by anyone and is kept confidential. I will make sure that when I write my notes that I do not write your names down. I will make up names for each participant.
Please take my card that has my email address which is [email protected] and my number here in Tanzania (0786972834) in case you have any concerns and questions. The card also has the number for the Centre for the Study of Forced Migration. The email address is [email protected] and the phone number is also on the card (0222410197). Please contact the Centre if you have any concerns and questions and they will ensure to contact me. I will also post my card on the bulletin board in the camp provided by the NGO workers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX B
SCRIPT TO INTERPRETER
My name is Barbra Lukunka, a second-year Masters student at the American University School of International Service in Washington DC. I am conducting research on the culture of refugee women in Tanzania. I would like to invite you to participate in this research as an interpreter. This research will be published online by the American University, furthermore I intend on presenting my findings in forums in North America and Africa.
If you choose to participate please note that I will not publish your name in the final project. Your identity as my assistant will not be revealed in the final project.
By taking part in this research you must adhere to certain policies that guide this research. It is required that you keep all the information that you leam during the interviews confidential. It is required that you do not share any of the information with others especially about specific individuals. It is required that you never disclose the names of individuals that are interviewed and the experiences that they share.
By participating and adhering to these policies you will be contributing to this research immensely.
Thank you,
180
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX C
INTERVIEW QUESTIONS
Refugee Women Interview in Kanembwa Camp: Biographic information limited to: ethnic identity, marital status, education level and age range
The goal was for the refugee women to present a monologue, therefore all interviews started with me saying to the refugee women: Please tell me your life story by telling me what your life was like in Burundi. Then tell me the events that took place during the conflict, the events while you fled for Tanzania, and then explain to me your life in the refugee camp.
Example of questions asked during the monologue:
1. How were the relationships between Hutus and Tutsis in your community 2. What were your specific feelings about being a Hutu Woman before the conflict in Burundi 3. What are some of the main differences that you noticed of how Hutu women are treated in Burundi compared to Tutsi women 4. What are some of the stories and myths that you have heard about the differences between Hutu and Tutsi women 5. What are some of the stereotypes of Hutu and Tutsi women 6. What were your thoughts about the animosity and conflict in Burundi when you were there 7. What year did you leave Burundi/Rwanda for Tanzania 8. Can you describe some of the circumstances that led you to leave Burundi/Rwanda 9. How long was your journey from Burundi/Tanzania to Tanzania 10. Can you describe your journey, where did you sleep, what did you eat, what were you feeling 11. Who did you travel with from Burundi 12. Where there any militia around at any point while you were fleeing Burundi 13. If yes to question 12 a. Where there any interactions between those who were fleeing and the militias that you know of b. What sort of activities were people involved with the militia performing c. Were the militia group/groups providing protection for those fleeing d. How did you feel about the militia and their ideology while fleeing
181
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 182
14. What were your thoughts and! feelings about leaving Burundi and being a refugee when you arrived in the camp 15. How did you feel about the animosity and conflict when you arrived in the camp 16. Are there any particular groups that posed security threats to the camp, or outside the camp 17. Did the camp make you feel safe 18. Did you believe the camp would be safe before arriving, what were your thoughts about that 19. Can you describe any instances when you did not feel safe when you first got to the camp 20. What are your opinions about fighting and armed conflict 21. What are your opinions about women who fight with militia 22. What are you opinions about women who try to be involved with the militia as girlfriends, fighters, messengers, workers 23. How do most women interact with militia 24. Why do you think women would want to be around militia groups 25. How has the your feeling of safety changed or has it changed over your stay in the camp 26. Who protects you and other refugees from insecurity 27. Who do you feel should protect you 28. How do you and other refugee women protect yourselves 29. Do you think you will be going home soon 30. What are your goals for the future
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BIBLIOGRAPHY
Alison, Miranda. “Women as Agents of Political Violence: Gendering Security,” Security Dialogue (2004): 447-463.
Arusha Peace and Reconciliation Agreement for Burundi 28 August 2000 available from: http://www.usip.org/library/pa/burundi/pa_ burundi_08282000_toc.html accessed December 2006.
Avruch, Kevin. Culture and Conflict Resolution. Washington DC: United States Institute of Peace Press, 2004.
Babbie, Earl.Survey Research Methods. Wadsworth Publishing Company, 1990.
Baines, Erin. “Body Politics and the Rwandan Crisis,” Third World Quarterly 24 no. 3 (2003):479-493.
Baker, Gary and Christine Ricardo. “Young men and the Construction of Masculinity in Sub-Saharan Africa: Implications for HIV/AIDS, Conflict and Violence.” World Bank Social Development Papers: Conflict Prevention and Reconstruction Paper No. 26 (June 2005).
Banton, Michael. “Ethnic Conflict,” Sociology no.34 3( 2000): 481-498.
Camus-Jacques, Genevieve. “Refugee Women : the Forgotten majority,” in Refugees and International Relations edited by Gil Loescher and Laila Monahan New York: Oxford University Press, 1989.
Chesler, Phyllis. Women’s Inhumanity to Women. New York: Plume Printing, 2003.
Chretien, Jean-Pierre. The Great Lakes ofAfrica: Two Thousand Years o f History. New York: Zone Books, 2003.
Cockbum, Cynthia. “Gendered Dynamics of Armed Conflict and Political Violence,” in Victims, Perpetrators or Actors? Gender, Armed Conflict and Political Violence London: Zed Books, 2001.
Cockbum, Cynthia. The Space Between Us. London: Zed Books, 1998.
183
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 184
Common Wealth Secretariat, Gender Mainstreaming in Conflict Transformation London: Commonwealth Secretariat, 2005.
Coomaraswamy, Radhika. “A Question of Honor: Women, Ethnicity and Armed Conflict,” in Feminists Under Fire: Exchanges Across War Zones Toronto: Between the Lines, 2003.
Coser, Lewis. The Functions o f Social Conflict. New York: The Free Press, 1956.
Crisp, Jeff. “A State of Insecurity: The Political Economy of Violence in Kenya’s Refugee Camps,” African Affairs 99 (2000): 601-632.
Crisp, Jeff. “Refugees and International Security: an Introduction to Some Key Issues and Policy Challenges,” 4th International Security Forum Geneva November 15-17 2000.
D’Amico, Francine “Critical Feminism: Deconstructing Nationalism, Gender and War,” in Making Sense o f International Relations Theory. Edited by Jennifer Sterling-Folker. Lynne Rienner, 2005.
Doxtador, Erik and Yeki Mosomothane. “Burundi: Permanent Deadlock or Tentative Peace?” in Through Fire with Water, Edited by Erik Doxtader and Charles Villa-Vicencio Trenton: Africa World Press Inc., 2003.
Enloe, Cynthia. Does Khaki Become You? The Militarization o f Women’s Lives. Boston, South End Press, 1983.
Enloe, Cynthia. Maneuvers: the International Politics of Militarizing Women’s Lives London: University of California Press, 2000.
Fox, Mary-Jane. “Girl Soldiers: Human Security and Gendered Insecurity,” Security Dialogue (2004): 465-479.
Galtung, Johan. Peace by Peaceful Means: Peace and Conflict, Development and Civilization. London: Sage Publications, 1996.
Giggs, Richard. “Geo-strategies in the Great Lakes Conflict and Spatial Designs for Peace.” Available from: http://www.cwis.org/hutu3_l.html; accessed on March 17, 2006.
Giles, Wenona. “Gendered Violence in War: Reflections on Transnationalist and Comparative Frameworks in Militarized Conflict Zones,” in Engendering Forced Migration. New York: Berghahn Books, 1999.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 185
Halim, Asma Abdel. “Attack with a Friendly Weapon,” in What Women do in Wartime London: Zed Book, 1998.
Harrison, Deborah. “Violence in the Military Community,” inMilitary Masculinities: Identity and the State. London: Praeger, 2003.
Hopton, John. “The State and Military Masculinity,” in Military Masculinities: Identity and the State. London: Praeger, 2003.
Human Rights Watch “Shattered Lives: Sexual Violence in the Rwandan Genocide and its Aftermath.” September 1996 available from http://www.hrw.org accessed March 2006.
Human Rights Watch “Burundi: Civilians Pay the Price of Faltering Peace Process” available from: http://www.hrw.org/backgrounder/africa/ burundi/ burundi0203.htm#_TOC34131812 accessed March 2007.
Human Rights Watch “Security Concerns: Accusations of Refugee Militarization and Crime.” available from: http://www.hrw.org/reports/1999/tanzania/ tanzania99_02.htm accessed March 2006.
Hyndman, Jennifer. “Refugee Camps as Conflict Zones,” in Sites o f Violence: Gender and Conflict Zones. Edited by Wenona Giles and Jennifer Hyndman. London: University of California Press, 2004.
Ibanez, Ana Cristina. “El Salvador: War and Untold Stories—Women Guerrillas,” in Victims, Perpetrators or Actors? Gender, Armed Conflict and Political Violence. London: Zed books, 2001.
Iska, Darlene, Stephen Trainor, Marchia Leithauser and Mady Wechsler Segal. “Women’s Participation in Armed Forces Cross-Nationally: Expanding Segal’s Model,” Current Sociology no. 50 5 (2002): 771-797.
Jacobsen, Karen and Steven Wilkinson. “Refugee Movements as Security Threats in Sub-Saharan Africa,” in International Migration and Security. Boulder: Westview Press, 1993.
Kamanga, Khoti. “The Tanzanian Refugee Act of 1998: Some Legal and Policy Implications,” Journal o f Refugee Studies no. 18 1 (March 2005): 100- 116.
Kovitz, Marcia. “The Roots of Military Masculinity,” in Military Masculinities: Identity and the State. London: Praeger, 2003.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 186
Lemarchand, Rene. Burundi: Ethnic Conflict and Genocide. Cambridge: Cambridge University Press, 1994.
Loescher Gilbert and James Milner. Protracted Refugee Situations. New York: Routledge, 2005.
Loescher, Gilbert. “The International Refugee Regime: Stretched to the Limit,” Journal o f International Affairs no. 47 2 (1994): 351.
Loescher, Gilbert. Refugees and International Relations. Edited by Gil Loescher and Laila Monahan New York: Oxford University Press, 1989.
Mach, Andrew “A Signifier of Shared Value,” What is Human Security? Security Dialogue 35no. 3 September 2004.
Malkki, Liisa. Purity and Exile: Violence, Memory, and National Cosmology Among Hutu Refugees in Tanzania. Chicago: University of Chicago Press, 1995.
Mamdani, Mahmood. When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda. Princeton: Princeton University Press, 2001.
Manchanda, Rita. “Ambivalent Gains in South Asian Conflicts,” in The Aftermath: Women in Post Conflict Transformation London: Zed Book, 2001.
Martin, Susan Forbes. Refugee Women. New York : Lexington Books, 2004.
Matlou, Patrick. “Upsetting the Chart: Forced Migration and Gender Issues, the Experience of Africa,” in Engendering Forced Migration New York: Berghahn Books, 1999.
Mazurana, Dyan et al. “Girls in Fighting Forces and Groups : Their Recruitment, Participation, Demobilization, and Reintegration,” Peace and Conflict: Journal o f Peace Psychology (2002): 97-123.
McKay, Susan. “Girls as Weapons of Terror in Northern Uganda and Sierra Leonean Rebel Fighting Forces,” Studies in Conflict and Terrorism (2005):385-397.
Merry, Sally Engle.Human Rights and Gender Violence: Translating International Law into Local Justice. Chicago: The University of Chicago Press, 2006.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 187
Mertus, Julie. “Liberal Feminism: Local Narratives in a Gendered Context,” in Making Sense of International Relations Theory. Edited by Jennifer Sterling-Folker Lynne Rienner, 2005.
Mertus, Julie. War’s Offensive on Women: The Humanitarian Challenge in Bosnia, Kosovo, and Afghanistan. Canada: Kumarian Press, 2000.
Miall, Hugh, Oliver Ramsbotham, Tom Woodhouse. Contemporary Conflict Resolution. Cambridge: Polity Press, 2005.
Mogire, Edward. “Preventing or Abetting : Refugee Militarization in Tanzania,” in No Refuge. London: Zed Books, 2006.
Mtanga, Elly. “Military and Armed Attacks on Refugee Camps,” in Refugees and International Relations. Edited by Gil Loescher and Laila Monahan. New York: Oxford University Press, 1989.
Muggah, Robert. No Refuge: The Crisis o f Militarization in Africa. London: Zed Books, 2006.
Mthembu-Salter Gregory. “Self-Determination Country Profile: Burundi,” Foreign Policy Focus available from: http://www.fpif.org accessed March 2006.
Nagel, Joan. “Masculinity and Nationalism: Gender and Sexuality in the Making of the Nation,” Ethics and Racial Studies (1998).
Nordstrom, Carolyn. “Gendered War,” Studies in Conflict and Terrorism (2005) 399-411.
Nordstrom, Carolyn. Shadows o f War: Violence, Power and International Profiteering in the Twenty-First Century Berkeley, University of California Press, 2004.
Olsen Odd Einer and Kristin S. Schaffscher. “Rape in Refugee Camps as Organizational Failure,” International Journal of Human Rights (1998).
Otterbein, Keith F. Comparative Culture Analysis: An Introduction to Anthropology. New York: University of Kansas Press, 1966.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 188
Peterson, Dale, and Richard Wrangham. Demonic Males. Boston: Houghton Mifflin Company, 1996.
Refugees International “Tanzania: Hope in the Flight to Reduce Gender Violence in Camps” available from: http://www.refugeesintemational.org/ content/article/ detail/73 9/7PHP SESSID=5ce00f92779cl66324eld accessed March 2007.
Rossman, Gretchen B. and Sharon F. Rallis. Learning in the Field: An Introduction to Qualitative Research. London: Sage Publication, 2003.
Segal, Mady Wechsler. “Women’s Military Roles Cross-Nationally: Past, Present, and Future,” Gender and Society no. 9 6 (1995): 757-775.
Seidman, Irving. Interviewing as Qualitative Research: A Guide for Researchers in Education and Social Sciences. London: Teacher’s College Press, 2006.
Semelin, Jacques. Purifier etDetruire. Paris: Edition du Seuil, 2005.
Shepler, Susan. “Post-war Trajectories for Girls Associated with the Fighting in Forces in Sierra Leone,” Politique Africaine (2002):49-62.
Stedman, Stephen John. Refugee Manipulation: War, Politics and the Abuse of Human Suffering. Washington DC: Brookings Institution Press, 2003.
The Lawyers Committee for Human Rights. Refugees, Rebels and the Questfor Justice. New York: The Lawyers Committee for Human Rights, 2002.
Turshen, Meredith. “Women’s War Stories,” in What Women do in Wartime. London: Zed Book, 1998.
Twagiramariya, Clotilde and Meredith Turshen. “Favours to Give and Consenting Victims the Sexual Politics of Survival in Rwanda,” in What Women do in Wartime. London: Zed Book, 1998.
United Nations Development Program. Human Development Report 1994: New Dimensions o f Human Security. New York: Oxford University Press, 1994.
United Nations Conventions and Protocol Relating to the Status of Refugees: Text of the 1951 Convention and the 1967 Protocol available from: http://www.unhcr.org/cgi-bin/texis/vtx/protect/opendoc.pdf? tbl=PROTECTION&id=3b66c2aalO accessed March 2006.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 189
United Nations High Commission for Refugees, “Funding Crisis Threatens Return Hope for Thousands of Burundian Refugees” 28 October 2005 available from: http://www.unhcr.org/cgi-bin/texis/vtx/news/opendoc. htm?tbl=NEWS&id=4362556c4 accessed March 2006.
United Nations High Commission for Refugees Statistics “2005 Global Refugee Trends” available from: http://www.unhcr.org/statistics/ STATISTICS/4486cebl2.pdf accessed February 2006.
Volman, Daniel. “The Militarization of Africa,” in What Women do in Wartime. London: Zed Book, 1998.
West, Harry. “Girls with Guns: Narrating the Experience of War of FRELIMO’s ‘Female Detachment,” In Children and Youth on the Front Line: Ethnography, Armed Conflict and Displacement. Edited by Jo Boyden and Joanna de Berry. New York: Berghahn Books, 2005.
Wood, Elizabeth Jean “Variation in Sexual Violence during War,” Society and Politics 34 no.3 (2006): 307-341.
Wrangham, Richard and Dale Peterson. Demonic Males. Boston: Mariner Book, 1996.
Yu, Lisa. “Separating Ex-Combatants and Refugees in Zongo, DRC: Peacekeeping and UNHCR Ladder of Options,” UNHCR Working Paper (2002 ).
Zolberg, Aristide et al. Escaping from Violence: Conflict and the Refugee Crisis in the Developing World. New York: Oxford University Press, 1989.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.